Document Sample
_Fish_Physiology_Volume_26__David_J._McKenzie__Anthony_P._Farrell__Colin_J._Brauner-Primitive_Fishes__Fish_Physiology__Vol._26_-Academic_Press_2007_ Powered By Docstoc
                    This is Volume 26 in the
             FISH PHYSIOLOGY series
  Edited by Anthony P. Farrell and Colin J. Brauner
Honorary Editor: William S. Hoar and David J. Randall

 A complete list of books in this series appears at the end of the volume

                      Edited by

        Institut des Sciences de l’ Evolution
   UMR 5554 CMRS‐Universite de Montpellier II
          ´        ´
Station Mediterraneenne de l’ Environnement Littoral
                     Sete, France

             Department of Zoology
          University of British Columbia
       Vancouver, British Columbia, Canada

             Department of Zoology
          University of British Columbia
       Vancouver, British Columbia, Canada

             Academic Press is an imprint of Elsevier
Front Cover Photograph: The cover shows various present‐day reputedly primitive fishes
pointing downwards and some of their earliest fossil relatives pointing upwards.
The tails of two examples of extinct fish taxa, Palaeozoic in age, are also just visible
at the bottom. Drawing by P. Janvier

Academic Press is an imprint of Elsevier
525 B Street, Suite 1900, San Diego, California 92101-4495, USA
84 Theobald’s Road, London WC1X 8RR, UK

This book is printed on acid-free paper.

Copyright # 2007, Elsevier Inc. All Rights Reserved.

No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing from the Publisher.

The appearance of the code at the bottom of the first page of a chapter in this book
indicates the Publisher’s consent that copies of the chapter may be made for
personal or internal use of specific clients. This consent is given on the condition, however, that
the copier pay the stated per copy fee through the Copyright Clearance Center, Inc.
(, for copying beyond that permitted by Sections 107 or 108 of the U.S.
Copyright Law. This consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new collective
works, or for resale. Copy fees for pre-2007 chapters are as shown on the title pages.
If no fee code appears on the title page, the copy fee is the same as for current chapters.
1546-5098/2007 $35.00

Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone: (þ44) 1865 843830, fax: (þ44) 1865 853333,
E-mail: You may also complete your request on-line
via the Elsevier homepage (, by selecting “Support & Contact”
then “Copyright and Permission” and then “Obtaining Permissions.”

For information on all Elsevier Academic Press publications
visit our Web site at

ISBN-13: 978-0-12-373671-0

07 08 09 10    9 8 7 6 5 4 3 2 1

CONTRIBUTORS                                                                             ix
PREFACE                                                                                  xi

 1.        Living Primitive Fishes and Fishes From Deep Time
           Philippe Janvier
   1.      Introduction                                                                  2
   2.      Primitive Characters, Primitive Taxa, and Ancient Taxa                        4
   3.      Living Fossils                                                                6
   4.      Living Primitive Fishes in Vertebrate Phylogeny                               9
   5.      Living Primitive Fishes and Their Fossil Relatives: Naming and Dating Taxa   16
   6.      Extinct Major Fish Taxa and Their Position in Vertebrate Phylogeny           28
   7.      How Stable is Vertebrate Phylogeny?                                          38
   8.      Fossils and Physiology                                                       39
   9.      The Environment of Early Fishes: Marine Versus Freshwater Vertebrates        41
  10.      Conclusions                                                                  45
           References                                                                   45

 2.        Cardiovascular Systems in Primitive Fishes
           Anthony P. Farrell
      1.   Introduction                                                                  54
      2.   An Overview of Evolutionary Progressions                                      57
      3.   Details of the Cyclostome Circulatory Systems                                 64
      4.   Details of the Sarcopterygii (Lobe‐Finned Fishes) Circulatory Systems         86
      5.   Details of the Circulatory Systems in Polypterids, Gars, and Bowfins          105
      6.   Details of the Sturgeon Circulatory Systems                                  109
      7.   Conclusions                                                                  111
           References                                                                   112

vi                                                                               CONTENTS

 3.        Nervous and Sensory Systems
           Shaun P. Collin
      1.   Introduction                                                               122
      2.   Development of the CNS                                                     123
      3.   The Brains of Primitive Fishes                                             124
      4.   Functional Classification of Cranial Nerves in Fishes                       129
      5.   The Visual System                                                          132
      6.   Chemoreceptive Systems                                                     144
      7.   Octavolateralis System                                                     152
      8.   Electroreception                                                           160
      9.   Concluding Remarks                                                         165
           References                                                                 166

 4.        Ventilatory Systems
           Emily Coolidge, Michael S. Hedrick, and William K. Milsom
      1.   Introduction                                                               182
      2.   Respiratory Strategies                                                     183
      3.   Respiratory Organs                                                         184
      4.   Ventilatory Mechanisms                                                     189
      5.   Respiratory Control                                                        196
      6.   Conclusions                                                                206
           References                                                                 206

 5.        Gas Transport and Exchange
           C. J. Brauner and M. Berenbrink
      1.   Introduction                                                               214
      2.   Partitioning of O2 and CO2 Exchange Across the Respiratory Surfaces        214
      3.   Blood O2 Transport                                                         230
      4.   Transport and Elimination of CO2                                           253
      5.   Synthesis                                                                  262
           References                                                                 270

 6.        Ionic, Osmotic, and Nitrogenous Waste Regulation
           Patricia A. Wright
      1.   Introduction                                                               284
      2.   Ionic and Osmotic Regulation                                               285
      3.   Nitrogen Excretion                                                         291
      4.   Concluding Remarks                                                         309
           References                                                                 310
CONTENTS                                                                      vii

 7.        Locomotion in Primitive Fishes
           D. J. McKenzie, M. E. Hale, and P. Domenici
      1.   Introduction                                                       320
      2.   Swimming Modes and Associated Morphological Adaptations            321
      3.   Locomotor Muscles                                                  328
      4.   Neuromotor Coordination                                            331
      5.   Locomotor Performance and Physiology                               338
      6.   Conclusions                                                        368
           References                                                         370

 8.        Peripheral Endocrine Glands. I. The Gastroenteropancreatic
           Endocrine System and the Thyroid Gland
           John H. Youson
      1.   Introduction                                                       382
      2.   Endocrine Pancreas and Related Gastrointestinal Endocrine System   383
      3.   Thyroid Gland                                                      405
      4.   Summary and Conclusions                                            440
           References                                                         442

 9.        Peripheral Endocrine Glands. II. The Adrenal Glands and the
           Corpuscles of Stannius
           John H. Youson
      1.   Introduction                                                       458
      2.   Adrenal Glands                                                     459
      3.   Corpuscles of Stannius                                             487
      4.   Summary and Conclusions                                            500
           References                                                         502

10.        Why Have Primitive Fishes Survived?
           K. L. Ilves and D. J. Randall
      1.   Introduction                                                       516
      2.   Life During the Early Phanerozoic                                  516
      3.   The Teleosts                                                       518
      4.   Primitive Fishes: Relationships Between Groups                     520
      5.   Why Have These Primitive Fishes Survived?                          530
      6.   Conclusions                                                        532
           References                                                         533

INDEX                                                                         537
OTHER VOLUMES        IN THE   SERIES                                          561

The numbers in parentheses indicate the chapter(s) that the authors have written.

MICHAEL BERENBRINK (213), Integrative Biology Resarch Division, School of
  Biological Sciences, The University of Liverpool, Liverpool, United
COLIN J. BRAUNER (213), Department of Zoology, University of British
   Columbia, Vancouver, Canada
SHAUN P. COLLIN (121), Vision, Touch and Hearing Research Centre, School
   of Biomedical Sciences, The University of Queensland, Brisbane, Queens-
   land, Australia
EMILY COOLIDGE (181), Department of Zoology, University of British
   Columbia, Vancouver, Canada
PAOLO DOMENICI (319), IAMC-CNR, Localita’ Sa Mardini, Torregrande
   (Oristano), Italy
ANTHONY P. FARRELL (53), Department of Zoology, University of British
  Columbia, Vancouver, Canada
MELINA HALE (319), Department of Organismal Biology & Anatomy,
  University of Chicago, Chicago, IL, USA
MICHAEL S. HEDRICK (181), Department of Biological Sciences, California
  State University, Hayward, CA, USA
K. L. ILVES (515), Department of Zoology, University of British Columbia,
   Vancouver, BC, Canada
PHILIPPE JANVIER (1), UMR 5143 du CNRS, Museum National d’Histoire
   Naturelle, Departement Histoire de la Terre, Paris, France
D. J. MCKENZIE (319), UMR 5554 du CNRS, Institut des Sciences de
                         ´        ´                                   `
   l’Evolution, Station Mediterraneenne de L’Environnement Littoral, Sete,

x                                                          CONTRIBUTORS

WILLIAM K. MILSOM (181), Department of Zoology, University of British
   Columbia, Vancouver, Canada
D. J. RANDALL (515), Department of Zoology, University of British
   Columbia, Vancouver, BC, Canada
PATRICIA A. WRIGHT (283), Department of Integrative Biology, University of
   Guelph, Guelph, ON, Canada
JOHN H. YOUSON (381, 457), Department of Life Sciences, University of
   Toronto, Scarborough, Toronto, Ontario, Canada

     We had many discussions about the most appropriate title for this volume
of the Fish Physiology series. “Primitive fishes” is a loose denomination that
is typically used to describe species from taxonomic groups which appeared
in vertebrate evolution earlier than the modern elasmobranchs and the tele-
osts. In this context, the term “primitive” is synonymous with the more
scientifically correct “plesiomorphic,” which indicates the possession of prim-
itive morphological characters, hence characters that occurred earlier in the
fossil record than those by which dominant modern groups are defined. In
most cases, primitive fishes are the extant remnants of taxa that dominated
periods of the fossil record but comprise a limited number of species today.
This has led them also to be described as “living fossils,” “evolutionary relics,”
or “ancient fishes.” Therefore, by selecting primitive fishes, we elected for a
simpler descriptor, rejecting the more scientifically robust or more emotive
     The primitive fishes that this book focuses on include the jawless ag-
nathans (hagfishes and lampreys), the lobe-finned sarcopterygians (coela-
canth and lungfishes), and the primitive ray-finned actinopterygian fishes
(the sturgeons, the bichirs and the ropefish, the gars, and the bowfin). This
is, therefore, a rather diverse collection of taxa whose only universal feature is
being of ancient lineage.
     The primitive fishes, so defined, are all interesting because they are repre-
sentative of stages in the evolution of physiological systems in fishes, and in
some cases also of the tetrapods. This book reviews what is known about the
physiology of these unusual animals, by comparison with the two fish groups
that dominate today, the modern elasmobranchs and the teleosts. The book
takes a systems approach, with chapters that review and summarize what is
known about cardiovascular, nervous, and ventilatory systems, gas exchange,
ion and nitrogenous waste regulation, locomotion, and the endocrine systems.
     The cardiovascular system is crucial by virtue of its role in transporting
nutrients, respiratory gases, hormones, and waste products. A chapter focuses
on circulatory form and function: cardiovascular anatomy, cardiac dynamics,
and cardiovascular control. Unusual adaptations of primitive fishes that
xii                                                                   PREFACE

deviate from features common to elasmobranchs and teleosts are highlighted,
and the chapter examines the evolutionary roots and evolutionary divergence
of the piscine cardiovascular system.
    The nervous and sensory systems of the primitive fishes are reviewed and
compared, with anatomical, physiological, molecular, and behavioral data
discussed in relation to both ecological and phylogenetic relationships. The
peripheral and central components of the sensory systems are examined
in some detail highlighting the physiological basis for behavior wherever
    Primitive fishes exhibit a tremendous adaptive radiation in their respira-
tory physiologies. A chapter reviews respiratory strategies, respiratory organs,
ventilatory mechanisms, and the control systems that integrate multiple
exchange sites and receptors into the overall ventilatory response to environ-
mental perturbations such as hypoxia and hypercarbia.
    The following chapter reviews the physiology of gas transport and ex-
change, in particular, those adaptations for gas exchange, such as air breath-
ing, which may have contributed to the survival of the primitive fishes. It also
considers the evolution of the Bohr/Haldane eVect, Root eVect, hemoglobin
buVer value, and the appearance of the choroid rete mirabile, in the primitive
ray-finned fishes.
    Among the primitive fishes, there is a diversity of strategies that have
evolved to cope with ion, water, and nitrogen balance. A chapter reviews
the regulation of salt and water balance, from the ionic and osmotic confor-
mation seen in hagfish to the strategies for regulation of body fluids distinct
from the environment that characterizes most other fish groups. Strategies for
nitrogen balance are also reviewed, including urea synthesis via the urea cycle
in the coelacanth and estivating lungfish.
    Swimming is critical to the ecology of many fishes as it determines, for
example, their ability to forage, to escape predators, and to migrate. A chapter
reviews the information about swimming modes, locomotor muscles, and
neuromotor coordination in primitive fishes. It then compares swimming
endurance, prolonged exercise performance, fast start escape responses,
and recovery from anaerobic burst exercise between these fish and the modern
elasmobranchs and teleosts.
    The peripheral endocrine systems of plesiomorphic fishes have been the
focus of quite significant research. A chapter focuses on the phylogenetic
development of two elements of the peripheral endocrine system, the gastro-
enteropancreatic (GEP) system, and the thyroid gland. In particular, evidence
is provided for clear phylogenetic patterns in distribution and structure of the
GEP system from protochordates to the ancient agnathans through more
generalized teleosts.
PREFACE                                                                    xiii

    A further chapter reviews morphological and molecular data in a phylo-
genetic analysis of two other elements of the peripheral endocrine system in
primitive fishes. A scheme is provided of the phylogeny of the steroid-
synthesizing, adrenocortical homologue, and of the catecholamine-secreting,
chromaYn tissue in agnathans (hagfish and lamprey) and in bony fishes of
ancient lineage. The corpuscles of Stannius first appear in the primitive acti-
nopterygians, and the subsequent phylogenetic trends of these glycoprotein-
secreting glands are reviewed.
    These chapters are preceded by a chapter that places the primitive fish
groups within their evolutionary context relative to other vertebrates, and the
volume concludes with a chapter that ponders on how each primitive fish (or
fish group) might have endured while their evolutionary contemporaries have
gone extinct.
    We wish to thank the reviewers for their suggestions and criticisms and
Kirsten Funk at Elsevier for publication advice and support.
                                                          David J. McKenzie
                                                          Anthony P. Farrell
                                                            Colin J. Brauner


 1.   Introduction
 2.   Primitive Characters, Primitive Taxa, and Ancient Taxa
 3.   Living Fossils
 4.   Living Primitive Fishes in Vertebrate Phylogeny
      4.1. The Hagfish‐Lamprey‐Gnathostome Node
      4.2. The Gar‐Bowfin‐Teleosts Node
      4.3. The Coelacanth‐Lungfish‐Tetrapod Node
      4.4. Other Problematic Nodes
 5.   Living Primitive Fishes and Their Fossil Relatives: Naming and Dating Taxa
      5.1. Hagfishes and Lampreys
      5.2. Chondrichthyans
      5.3. Actinopterygians
      5.4. Sarcopterygians
 6.   Extinct Major Fish Taxa and Their Position in Vertebrate Phylogeny
      6.1. Yunnanozoans and Myllokunmingiids
      6.2. ‘‘Ostracoderms’’
      6.3. Placoderms
      6.4. Acanthodians
      6.5. ‘‘Paleoniscoids’’ and Basal Neopterygians
      6.6. Extinct Sarcopterygian Taxa
 7.   How Stable is Vertebrate Phylogeny?
 8.   Fossils and Physiology
 9.   The Environment of Early Fishes: Marine Versus Freshwater Vertebrates
10.   Conclusions

   The notion of ‘‘primitive living taxon,’’ or ‘‘living fossil,’’ largely stems
from the evolutionary concepts that have pervaded systematics for nearly a
century, notably the view that paraphyletic taxa are real and ancestral.
Certain living taxa are regarded as primitive because some of their characters
remain in a plesiomorphic state relative to their homologues in other, closely

Primitive Fishes: Volume 26                            Copyright # 2007 Elsevier Inc. All rights reserved
FISH PHYSIOLOGY                                                   DOI: 10.1016/S1546-5098(07)26001-7
2                                                              PHILIPPE JANVIER

related living taxa, and this assessment rests on both out‐group comparison
and fossil data. The biology of ‘‘primitive living taxa’’ is thus supposed
to mirror that of the related fossil taxa they resemble. Physiologists, there-
fore, bet that the physiological functions of a reputedly ‘‘primitive living fish’’
are the same as those of its fossil anatomical proxies, but paleontologists
often infer those of the latter on the basis of ‘‘primitive’’ living models.
In some cases, such circular reasoning can be avoided by considering
paleoenvironmental data that are inferred preferably from geochemical
parameters. An overview of living and fossil vertebrate phylogeny, however
stable it may seem, shows that there are several ways of defining and naming
taxa, and that shared physiological characters of a crown group may not
be extrapolated to its stem group, the divergence of which may be much
earlier. Physiological characters are probably no more and no less homoplas-
tic than morphological characters. Like the latter, they can be decomposed
into series of states that can be included in fractioned and combined parsi-
mony analyses, and can contribute to patterning the trees, instead of being
interpreted a priori as adaptive and mapped as attributes on trees based on
other kinds of characters.


    By comparison to that of the morphological or molecular characters, the
question of the homology of physiological characters has been little debated
during the last three decades of the twentieth century, which roughly cor-
respond to the time of the ‘‘cladistic revolution’’ in comparative biology. The
reason for this neglect is that physiology was long regarded as a discipline
of ‘‘general biology’’: the biology of processes, as opposed to ‘‘comparative
biology,’’ that is, the biology of patterns, as outlined by Nelson (1970, 1994).
There is indeed an old tradition of considering physiological characters as
highly ‘‘adaptive’’; that is, they are assumed to be commonly subject to
homoplasy and thus their distribution, however hierarchical it may some-
times look, tells us little about their evolutionary history. There are multiple
historical reasons for this deep‐rooted belief, some of which date back to the
nineteenth century, possibly with a scent of Lamarckism, but there is no
clear evidence that physiological characters are more ‘‘adaptive’’ than such
anatomical structures, as the pattern of the skull bones or tail skeleton
morphology. Nevertheless, already in the early twentieth century, some
physiologists pointed out congruences between the ‘‘laws’’ based on morpho-
logical character distributions (or evolution) and the presumed history
of physiological characters. For example, Needham (1938) came to the
conclusion that, following the morphology‐based ‘‘Dollo’s Law,’’ losses of
physiological functions are irreversible.
1.   PRIMITIVE FISHES THROUGH TIME                                               3

    During the past two or three decades, there have been some attempts
at exploiting the phylogenetic message of physiological characters, notably
by Løvtrup (1977), who was the first to suggest cyclostome paraphyly on
this ground (see below). More recently, Cunchillos and Lecointre (2005)
demonstrated that metabolic pathways could be described as nested sets of
character states and coded like any other kind of characters in a data matrix
aimed at reconstructing phylogenetic relationships between taxa. Thus, there
are multiple ways of including physiological traits in phylogenetic analyses,
by considering either the distribution of a particular function (coded
as absent/present) or that of diVerent states of a function (as a hierarchy of
more and more complex pathways). A classical bias in physiology‐based
phylogenies is perhaps that physiologists readily know (or think they can
readily know) the selective advantage of a physiological character, notably
by means of experiments. Therefore, they are tempted to first make infer-
ences about the evolution of physiological characters on the basis of the
supposedly known history of the environment or behavior of an organism. In
contrast, morphologists generally can make only vague inferences about the
selective advantage of morphological characters, not to speak of molecular
phylogeneticists, whose nucleotide sequences tell little about their impact on
the phenotype. Yet comparative functional genomics may soon provide
information in this field.
    It is thus time to restore the consideration of physiological characters as a
source of potential shared homologies, irrespective of the morphological
characters they are inferred from, and stop considering that their interest
essentially lies in their adaptive plasticity, that is homoplasy. Physiological
characters are no worse, no better than any other legacy of evolution: they
provide examples of both phylogenetic messages (synapomorphies) and
adaptive convergences (homoplasies), but their assessment is always relative,
in the light of parsimony. There are nevertheless certainly some very robust
physiological ‘‘signatures’’ in phylogeny (e.g., uric acid excretion in sauropsid
amniotes), which can be regarded as being just as good node supports as, for
example, gnathostome jaws or tetrapod limbs.
    In this introduction, I should like first to make clear that the relationships
between organisms (and thus the criteria on the basis of which we decide
whether the latter are ‘‘primitive’’) are exclusively based on assumptions
about homology relationships between parts of these organisms, be they ana-
tomical or physiological characters, or even nucleotide sequences. Therefore,
the phylogenetic trees from which physiologists may infer evolutionary pat-
terns are mere theories based on most parsimonious character distributions,
and in which fossils provide additional character combinations, as well as
information about the minimum age of characters and taxa. Then, I shall
briefly depict the lost world of the ancient fossil fishes, on which rests the notion
of ‘‘living primitive fishes,’’ or, more generally, ‘‘living fossils.’’
4                                                             PHILIPPE JANVIER


    Taxa reflect relationships between organisms, which are inferred from
homology relationships between parts of these organisms (Nelson, 1994).
Consequently, a theory of phylogenetic relationships (and thus a phylogenetic
classification) reflects the most parsimonious distribution of congruent
homology relationships at one particular time and for one particular sample
of characters and terminal taxa. The structure of the vertebrate tree at the
level of the higher terminal taxa has been relatively stable during the past two
decades, despite some divergent theories that essentially arose from conflicts
between morphological and molecular sequence data (see below). However,
in detail, the numerous trees of each of the higher terminal taxa, expressed
in the same graphic way (a branching diagram), are derived from several,
entirely diVerent, conceptual backgrounds, such as genetic distance (phenet-
ics), parsimony, or model‐based approaches such as maximum likelihood or
Bayesian approaches. Although phylogeneticists now tend to provide, for
the same data set, the trees that are yielded by these respective methods and
generally consider that the diVerences are minimal, this often generates con-
fusion because subsequent authors often compare trees that are not, in fact,
comparable. Phylogenetic trees are doomed to remain theories forever, some
of which will be less and less frequently refuted. Therefore, the unending quest
of phylogenetics (i.e., to progressively tend toward a more and more stable
pattern of relationship) requires ever more data, and, above all, diVerent
kinds of data of approximately equal quality. Physiological characters are
certainly underexploited to this end. Perhaps genomics (i.e., parsimony‐based
analysis of the organization of the genome as a whole, rather than mere
sequences of particular genes) may also provide one of these new sources of
    Vertebrates are currently regarded as a taxon because they are the only
living beings that share such diverse characters as migrating neural crest cells,
massive gene duplication, or a labyrinth with at least one semicircular canal
with ampullae. The congruence of these character distributions is, to date,
not contradicted by conflicting distributions of other characters, apart from
minor examples currently regarded as homoplasies. Although phylogeneti-
cists focus their interest on the search for shared derived characters (synapo-
morphies) and thus regard shared primitive characters (symplesiomorphies)
as uninformative, other biologists whose main interest lies in biological or
evolutionary processes consider living organisms as a functional assemblage
of characters that can tell us something about the biology of a hypothetical
ancestor at one particular node of the tree of life. Thus, the closer to that
1.   PRIMITIVE FISHES THROUGH TIME                                              5

node, the more interesting the real organisms! This is particularly true for
physiologists, who pay much attention to the so‐called primitive living taxa,
that is, extant taxa that retain a large number of symplesiomorphies. For
example, hagfishes have a number of characters, notably physiological ones,
that are lacking in all other vertebrates (lampreys and gnathostomes, or
jawed vertebrates), but shared with other chordates (cephalochordates, tuni-
cates), and other deuterostomes (Jørgensen et al., 1998). These plesiomor-
phous characters are thus regarded as ‘‘primitive’’ and inherited from
‘‘invertebrates,’’ and tell us nothing about the particular relationships of
hagfishes but their absence (or presumed modification) in lampreys and
gnathostomes does as they exclude hagfishes. However, hagfishes also
share some unique anatomical characters with lampreys (Yalden, 1985),
and these homology relationships might also suggest that hagfishes and
lampreys are a taxon, the cyclostomes, unless these characters are lost in
jawed vertebrates. Relationships between parts of organisms almost inevita-
bly conflict, and relationships between organisms are, in principle, never
stable. They depend on the number and quality of characters that biological
research progressively and endlessly pours into the bag of comparative
biology. The principle of parsimony is currently considered the best way to
choose one theory of relationships between organisms: the one that is sup-
ported by the largest number of congruent derived character distributions
beyond the bounds of chance.
     However, this way of considering homology relationships (and thus
relationships between taxa) developed only in the 1960s, with the rise of
Hennig’s (1950) phylogenetic systematics. Before that time (and occasionally
still now), evolutionary systematics defined taxa on the basis of the overall
resemblance of the organisms they include, and a demarcation was arbitrarily
drawn between groups of organism that possessed a character and others that
did not, but both groups were named and thus were equally regarded as taxa.
Therefore, we still find such names as protozoans (versus metazoans), inverte-
brates (versus vertebrates), fishes (versus tetrapods), anamniotes (versus
amniotes), and so on, used by some biologists, not in a colloquial context
but as real groups of organisms whose history has a beginning and an end. The
beginning of fishes is that of vertebrates, but do fishes end with the first
tetrapod? In all these cases, the group that lacks the character that defines
the other group is generally said to be ‘‘primitive’’ or ‘‘ancestral.’’ The former
is what Hennig (1950) called a paraphyletic group (or a grade), the latter
is what he called a monophyletic group (or a clade). In a historical context,
these notions are admittedly relative, and fishes were monophyletic before the
rise of limbs (and thus tetrapods).
     Grades were very convenient to paleontologists for accommodating
fossils whose relationships were unclear and whose anatomy displayed
6                                                              PHILIPPE JANVIER

an overwhelming number of general (i.e., plesiomorphous) characters.
Among vertebrates, agnathans (jawless fishes), crossopterygians (‘‘lobe‐
finned’’ fishes), or paleoniscoids (primitive ray‐finned fishes) were such groups
that are only ‘‘defined’’ by the lack of the characters of their presumed
descendants. ‘‘Primitive living fishes’’ are thus such extant taxa that have
once been classified in a grade along with many extinct taxa.


     The term ‘‘living fossil,’’ coined by Darwin in his Origin of Species,
became widespread in the literature by the end of the nineteenth century.
It has been applied to a wide range of modern taxa for a variety of reasons.
The concept of ‘‘living fossil’’ is in general linked to evolutionary classifica-
tions and thus to the notion that grades are taxa. The case of the living
actinistian (coelacanth) Latimeria is a good example of how a taxon became
regarded as a ‘‘living fossil.’’ The status of ‘‘living fossil’’ assigned to Lati-
meria is essentially due to the fact that actinistians (which were known only
as fossils for almost a century) have long been classified in a taxon Cross-
opterygii (crossopterygians), which also included a number of Paleozoic
taxa and is now considered to be paraphyletic because it includes
stem tetrapods (rhizodontids, ‘‘osteolepiforms,’’ elpistostegalians) and stem
dipnoans (porolepiforms, youngolepidids; Figure 1.1). All the ‘‘crossoptery-
gian’’ characters that Latimeria shares with these taxa are merely general
sarcopterygian or even osteichthyan characters (e.g., monobasal‐paired
fin skeleton, intracranial articulation, large notochord, Figure 1.1A), which
have been modified or lost in most of the living descendants of ‘‘crossoptery-
gians’’ (Figure 1.1C and D). Although actinistians in general show a relatively
stable morphology during the past 380 million years (Myr), Latimeria is no
more a ‘‘living fossil’’ than many other fish taxa, such as lungfishes, and
probably less so, unless by the virtue of its retaining a few plesiomorphous
characters, such as the intracranial joint (Figure 1.1B). The survival of
‘‘living fossils’’ is thus rather a survival of some characters that were once
general, but have largely been lost in related extant taxa (Janvier, 1984).
     The concept of ‘‘living fossil’’ can also be seen in a diVerent way, that
is when conspicuous specializations appear very early in phylogeny, and are
retained during a long period of morphological stability. This refers to
the ‘‘panchronic’’ taxa which, among fishes, are relatively rare. Example of
such ‘‘panchronic taxa’’ are hagfishes and lampreys (Figure 1.2). The earliest
fossil hagfishes are 305 Myr old, and the earliest lamprey 360 Myr, and, as far
as it can be seen from fossils preserved as soft‐tissue imprints, looked practi-
cally identical to the living hagfishes and lampreys (Figure 1.2A, C, E, and G).
1.   PRIMITIVE FISHES THROUGH TIME                                                                  7


         B                           C                                 D

                                          Lungfishes                          Tetrapods
        0 Myr

       299 Myr


        443 Myr

Fig. 1.1. Latimeria as a ‘‘living fossil.’’ In typical evolutionary classifications, Latimeria is often
referred to as the only living ‘‘crossopterygian,’’ an essentially Paleozoic, paraphyletic group that
includes stem lungfishes and stem tetrapods (in gray). ‘‘Crossopterygians’’ are in fact diagnosed
on the basis of characters that have been lost or modified in their descendants, lungfishes and
tetrapods. One of these characters is the intracranial joint (arrowheads), a general (plesiomor-
phous) character of the sarcopterygian braincase (A), conserved in Latimeria (B), but lost in
lungfishes (C) and tetrapods (D). (A) Braincases of Nesides, (B) a late Devonian actinistian,
Latimeria, (C) Neoceratodus, and (D) a primitive Carboniferous tetrapod. [Based on Janvier
(1984, 1996a) by permission of Oxford University Press.]

Hagfishes already possessed the characteristic tentacles (Figure 1.2B
and D) and lampreys already possessed a piston cartilage (hence a protrac-
tible and retractable ‘‘tongue’’), an annular cartilage surrounding the mouth
(Figure 1.2F and H), and a sucker (Figure 1.2I and J; see also Gess et al., 2006
for new data). Such cases of panchronic taxa are perhaps the only ones that
can actually be referred to as ‘‘living fossils’’ since it does not rest on the
retention of a few plesiomorphous characters, but the conservation (or sur-
vival) of a large number of apomorphous characters that appeared early in
time and have remained unmodified (Janvier, 1984). Large gaps in the fossil
record, like in the case of Latimeria (for which there is no Tertiary record), are
also regarded as a criterion to assess a living taxon as a ‘‘living fossil.’’
8                                                                            PHILIPPE JANVIER


                             Prenasal duct

                                                         Prenasal duct



                                                  Piston cartilage



                                        Piston cartilage
                                              J                          K

Fig. 1.2. Hagfishes and lampreys are regarded as ‘‘living fossils,’’ not because they retain
characters that have once been general to a larger group but because they have gained some
highly derived characters that remained extremely stable throughout time. The overall morphol-
ogy (A) and skull structure (B) of living hagfishes strikingly resemble those of the late Carboni-
ferous hagfish Myxinikela (C and D). Although Myxinikela is somewhat stouter in body shape
than the living forms (C), it already displays the main unique characters of the living hagfish head
skeleton (D), such as the tentacles and long prenasal duct. Similarly, the overall morphology
(E) and head skeleton of the living lampreys (F) are almost identical to those of the late
1.   PRIMITIVE FISHES THROUGH TIME                                                                 9

    The notion of ‘‘living fossil,’’ in common with evolutionary systematics,
remains rather vague and subjective. This is reflected in Eldredge and
Stanley’s (1984) review of the diVerent taxa that may be regarded as such
and which includes elopomorph teleosts but, strangely, not lampreys. What-
ever the ground for regarding a taxon as a ‘‘living fossil,’’ it always rests on
morphology since this is essentially what is known in fossils. It is, of course,
tempting to consider that if the morphology of an extant taxon has under-
gone little change through time, the same may apply to its physiological
characters that leave no fossil record. This postulate, which cannot be
tested by direct evidence, is nevertheless often regarded as the only means
for attempting a reconstruction of the history of physiological characters.
Thus, the only possible test for theories about physiological character phy-
logeny is their congruence with molecular sequence‐based phylogenies
of living taxa and/or morphological character distributions in living and
fossil taxa.


    Living fish taxa that are traditionally regarded as ‘‘primitive’’ are essen-
tially the jawless vertebrates or agnathans [hagfishes (Hyperotreti) and
lampreys (Hyperoartia)], and, among the gnathostomes, some chondri-
chthyans [batomorphs (sawfishes, torpedoes, skate, and rays), hexanchiform
sharks, Chlamydoselachus, and chimaeriforms], cladistians (i.e., polypteri-
formes or bichirs), acipenceriforms (sturgeons and paddlefishes), ginglymods
(or lepisosteiforms, i.e., gars), Amia calva (bowfin), actinistians (Latimeria),
and lungfishes (or dipnoans). In addition, the osteoglossomorphs (e.g., bony‐
tongues) and elopomorphs (e.g., tarpons, eels) are sometimes regarded
as examples of primitive teleosts. The trees in Figure 1.3 show the relation-
ships of the major living vertebrate taxa, with special reference to the
so‐called ‘‘primitive fishes.’’ The fully resolved tree on the left‐hand side
shows the phylogeny that is most widely accepted by morphologists and
paleontologists. The tree on the right‐hand side shows a consensus of
the various trees based on either morphological or molecular sequence
data, which may yield slightly diVerent topologies (hence, the debated

Carboniferous Mayomyzon (G and H). Both share the characteristic annular and piston cartilages.
The late Carboniferous Pipiscius (I), regarded as a fossil lamprey, also possessed an oral funnel (J)
armed with horny plates, which strikingly resembles that of living lampreys (K). Scale
bars ¼ 10 mm. [A–C, E–G, J, and K, from Janvier (1984, 1996a by permission of Oxford
University Press); H, redrawn and modified from Bardack and Zangerl (1968); D, redrawn and
modified from Bardack (1991); and I, redrawn and modified from Bardack and Richardson (1977).]
10                                                                                      PHILIPPE JANVIER



              1                                          Sharks

                                       4               Batomorphs


                                                Cladistians (Polypteriformes)

                              7            9          Acipenserids
                                                    Ginglymods (lepisosteids)
                          6                          Osteoglossomorphs
                                                       Elopomorphs              D
                                               13     Other teleosts


                                  14 15              Neoceratodus
                                      17               Lepidosiren


Fig. 1.3. Interrelationships of the living vertebrates, with particular reference to the reputedly
primitive fishes. The tree on the right‐hand side shows the four major polytomies (A–D), which
are essentially due to conflicts between morphological and molecular sequence data. The ques-
tion of the relationships of sharks and batomorphs is regarded here as resolved as there is
increasingly strong support for their sister‐group relationships (thus shark monophyly). The
tree on the left‐hand side depicts the topology that is most widely accepted by morphologists
1.   PRIMITIVE FISHES THROUGH TIME                                                               11

nodes A–D for which diVerent data sets provide diVerent topologies for the
more crownward taxa).
    There are, however, some widely diVerent, molecular sequence‐based
phylogenies, notably that proposed by Arnason et al. (2001, 2004) and
Rasmussen and Arnason (1999a,b), one of which is shown here in
Figure 1.4. To date, these phylogenies are diYcult to reconcile with most
of the currently available morphological data as they support fish mono-
phyly but not osteichthyan, sarcopterygian, and actinopterygian
monophyly. However, provided that they are merely noise due to inappro-
priate gene sequences, these ‘‘odd phylogenies’’ (Janvier, 1998) are not to be
ignored and should stimulate a new look at certain conflicting character
combinations displayed by some early fossil chondrichthyans and
osteichthyans (Janvier, 1998; Zhu et al., 1999, 2006; Maisey, 2001),
which, at any rate, suggest that some of the classical osteichthyan and
sarcopterygian characters may be more general than currently believed.
    Current morphology‐based gnathostome phylogenies (basically that in
Figure 1.3, left‐hand tree), notably the assumption that chondrichthyans are
the plesiomorphous sister group of osteichthyans, stem from (or is consistent
with) Huxley’s (1880) conception that chondrichthyans were ancestral to
osteichthyans. Most molecular sequence‐based phylogenies have proven
to be consistent with the monophyly of these two respective groups (Hedges,
2001; Zardoya and Meyer, 2001). Arnason et al. (2004) pointed out, however,
that molecular (mitogenomic) sequence‐based osteichthyan phylogenies may
be biased by the fact that they are generally rooted with chondrichthyans,
rather than being rooted with either lampreys or hagfishes.
    When considering the current (or ‘‘conventional’’) consensus tree for higher
vertebrate taxa, only three major ‘‘piscine’’ nodes remain controversial:

and paleontologists. Main clades and selected characters applying to living taxa: 1, craniates
(migrating neural crest cells, epidermal placodes, skull, olfactory, optic, and otic capsules);
2, vertebrates (arcualia, extrinsic eye muscles, radial muscles in unpaired fins); 3, gnathostomes
(jaws, horizontal semicircular canal, paired fins, calcified or ossified endo‐ and exoskeleton,
epicercal tail); 4, chondrichthyans (prismatic calcified cartilage, pelvic claspers); 5, elasmobranchs
(posteriorly directed basibranchials, paired occipital condyles); 6, osteichthyans (endochondral
bone, large dermal bones covering the head and shoulder girdle, lepidotrichs); 7, actinopterygians
(only one dorsal fin, primitively ganoid scales, acrodin cap on teeth, everted telencephalon); 8,
actinopterans (fringing fulcla on the leading edge of fins); 9, acipenseriforms (anterior symphysis
of palatoquadrate); 10, neopterygians (unpaired fin lepidotrichs equal in number to their sup-
ports); 11, halecostomes (mobile maxillae); 12, teleosts (median tooth plate on basihyal, mobile
premaxillae); 13, elopocephalans (two uroneurals); 14, sarcopterygians (monobasal paired fins,
pulmonary vein, veina cava); 15, lungfishes (massive entopterygoid and prearticular tooth
plates); 16, lepidosirenidae (reduced paired fins); and 17, rhipidistians, or choanates (alveolae
in lungs, partially divided and sigmoid arterial cone in heart, incipient atrial septum). [Illustra-
tions for terminal taxa from Janvier (1996a) by permission of Oxford University Press.]
12                                                                          PHILIPPE JANVIER



                                    Cladistians (polypteriformes)







                                            Acipenserids              Acipenseriforms


                                          Ginglymods (lepisosteids)



Fig. 1.4. Mitochondrial DNA‐based vertebrate tree (hagfishes not considered) proposed
by Arnason et al. (2004). This tree is strongly at odds with the current consensus (compare to
Figure 1.3), notably by the breakdown of the osteichthyans, sarcopterygians, actinopterygians, and
neopterygians. [Illustrations for terminal taxa from Janvier (1996a) by permission of Oxford
University Press.]

the hagfish‐lamprey‐gnathostome node, the gar‐bowfin‐teleost node, and the
coelacanth‐lungfish‐tetrapod node. However, one must concede, in agreement
with Arnason et al. (2004), that the morphological support to the osteichthyan
clade is relatively low and even more so when early fossil taxa are considered.
1.   PRIMITIVE FISHES THROUGH TIME                                            13

4.1. The Hagfish‐Lamprey‐Gnathostome Node

    Since the early nineteenth century, lampreys and hagfishes have long been
gathered in a group, the Cyclostomi (cyclostomes) characterized by horny
‘‘teeth’’ covering a protractible and retractable ‘‘tongue,’’ and pouch‐shaped
gills. All other cyclostome features are either absence of gnathostome char-
acters or characters whose state cannot be assessed on the basis of out‐group
comparison (i.e., not applicable to nonvertebrate chordate taxa). Since the
mid‐twentieth century, morphologists and physiologists had long been aware
that lampreys were in many respects more similar to gnathostomes than
to hagfish, but the apparently more ‘‘simple’’ or ‘‘invertebrate‐like’’ characters
of hagfishes were regarded as a consequence of ‘‘degeneracy,’’ due to their
legendary endoparasitic habits (already alluded to by Linnaeus; Jørgensen
et al., 1998). Løvtrup (1977) was the first to suggest clearly that this appeal to
‘‘degeneracy’’ was groundless and that lampreys display an overwhelming
number of morphological and physiological characters shared only with
gnathostomes, which suggest sister‐group relationships between these two
taxa (2, Figure 1.3). Then, cyclostome paraphyly became progressively
accepted by many comparative biologists (Hardisty, 1982; Maisey, 1986;
Janvier, 1996a,b). In contrast, molecular sequence‐based phylogenies tended
to support cyclostome monophyly (Stock and Whitt, 1992; Delarbre
et al., 2000; Hedges, 2001; Mallatt et al., 2001), although analysis of data
sets partitioned into small and large subunit components provided
conflicting support for both monophyly and paraphyly (Zrvav’y et al.,
1998). Cyclostome monophyly in molecular sequence‐based phylogenies
was first regarded as an artifact of long‐branch attraction, but it is still
strongly supported by Bayesian methods, which are supposed to minimize
this bias (Furlong and Holland, 2002). Some molecular sequence‐based data,
however, rather support the position of lampreys as sister group to gnathos-
tomes (cyclostome paraphyly) such as those based on small RNA units
(Gursoy et al., 2000). Currently, the problem is still unresolved. To morpho-
logists and physiologists, cyclostome monophyly would imply an impressive
number of either reversions in hagfishes or convergences in lampreys and

4.2. The Gar‐Bowfin‐Teleosts Node

    Gars and bowfins have long been grouped in a group called Holostei
(holosteans), along with a number of fossil neopterygian actinopterygian taxa
(e.g., semionotids, parasemionotids, macrosemiiforms). However, Patterson
(1973) regarded bowfins as more closely related to teleosts than gars. Bowfins
and teleosts were thus gathered in the clade Halecostomi (halecostomes),
characterized notably by mobile maxillae (11, Figure 1.3). Apart from a
14                                                             PHILIPPE JANVIER

paleontology‐based phylogeny proposed by Olsen and McCune (1991),
very few anatomical data support a sister‐group relationship between gars
and teleosts. Similarly, no anatomical data supports the topology proposed
by Inoue et al. (2003) and Arnason et al. (2004), in which gars, bowfins, and
acipenseriforms form an unresolved clade, sister to the teleosts (Figure 1.4).
Conversely, a reconsideration of fossil neopterygians anatomy (notably
for early gars, bowfins, semionotids, and parasemionotids) as well as molecular
sequence‐based neopterygian phylogenies (Kikugawa et al., 2004) now strongly
supports holostean monophyly again; that is, gars and bowfins would be
sister groups.

4.3. The Coelacanth‐Lungfish‐Tetrapod Node

    Piscine sarcopterygians are only represented by two strongly depau-
perized living taxa: actinistians (coelacanths) and dipnoans (lungfishes).
The living coelacanths are represented by Latimeria chalumnae from the
Strait of Mozambique, and possibly a second species Latimerai menadoensis
from Sulawezi (Indonesia). Lungfishes fall into three genera, the Australian
Neoceratodus (one species), the African Protopterus (four species), and
the South American Lepidosiren (one species). Neoceratodus is the sister
group of the clade Lepidosirenidae, which includes the other two genera
(15, Figure 1.3).
    Although long debated, the question of actinistian‐lungfish‐tetrapod
relationships is now generally regarded as settled by morphologists and
paleontologists (Rosen et al., 1981; Cloutier and Ahlberg, 1996; Zhu et al.,
2006), with coelacanths being sister to lungfishes and tetrapods, despite a few
anatomical data that may support an actinistian–tetrapod sister‐group rela-
tionship. In contrast, molecular data remain ambiguous on this issue
(Zardoya and Meyer, 1997, 2001; Zardoya et al., 1998; Brinkmann et al.,
2004; Takezaki et al., 2004). Molecular sequence‐based trees sometimes
show coelacanths as sister to tetrapods and sometimes as sister to lungfishes.
Only the molecular sequence‐based phylogeny proposed by Arnason et al.
(2004) is strongly at odds with the current consensus, as it shows monophy-
letic living fishes, with coelacanths, chondrichthyans, lungfishes, and actino-
pterans (i.e., actinopterygians minus cladistians) as forming an unresolved
clade, sister to cladistians (Figure 1.4).

4.4. Other Problematic Nodes

   The in‐group relationships of some other higher piscine gnathostome
taxa are also the subject of controversies, yet to a lesser degree. Within elasmo-
branch chondrichthyans, the relationships of batomorphs (pristiophoriforms,
1.   PRIMITIVE FISHES THROUGH TIME                                            15

sawfish, torpedoes, skates, and rays) have long been debated, but, during the
past 20 years, most morphologists agreed that batomorphs were nested within
the squalomorph sharks as sister group to pristiophoroids (Maisey, 1984;
Shirai, 1996). Molecular sequence‐based phylogenies do not clearly support
this relationship and suggest, rather, that batomorphs are the sister group
of all living sharks (Arnason et al., 2001). Recent consideration of this
question, involving collaboration between morphologists, paleontologists,
and molecular phylogeneticists, has provided much stronger support for
this theory, which entails living shark monophyly (Maisey et al., 2004;
Figures 1.3 and 1.4). It now appears that living shark monophyly is actually
supported by a number of morphological characters, but previous parsimony
analyses considered too many doubtful characters, which now turn out to be
    Cladistians (bichirs) are yet another taxon whose phylogenetic position
has been much debated since the nineteenth century. Cladistians share some
unique skeletal characters with actinopterans (i.e., acipenseriformes and neop-
terygians), such as the ganoid scale structure or the acrodin cap on teeth, and
numerous soft‐tissue characters, notably in brain development (everted telen-
cephalon) and muscles of the jaw and gill arches. Most other characters seen in
cladistians are unique to them (pectoral fin and dorsal fin structures), but a
few characters were once regarded as either shared only with sarcoptery-
gians, notably actinistians, or supposedly primitive for osteichthyans in general
(Jarvik, 1980). Cladistians are almost unanimously regarded as the sister group
to actinopterans (7, Figure 1.3), but molecular sequence data remain ambi-
guous in this respect. Early sequence‐based gnathostome phylogenies generally
ignored cladistians because they made trees either collapse or display very odd
topologies, as in Arnason et al.’s (2004) tree (Figure 1.4), where they appear as
the sister group of all other living fishes. Other molecular trees show a more
conventional position for cladistians as sister to actinopterans (Noack et al.,
1996). The question may not be entirely settled as long as the early phase
of cladistian history, possibly in early Devonian times, remains undocumented
by fossils.
    Within teleosts, osteoglossomorphs (osteoglossids, arapaimids, mor-
myrids, and hiodontids) and elopomorphs (elopids, megalopids, albuloids,
nothacanthids, anguilloids, and saccopharyngids) are currently considered
as the most inclusive living teleost taxa. However, the relationships of these
two taxa to other teleosts (i.e., clupeocephalans) remain debated. A classical
theory is that osteoglossomorphs are the sister group of elopomorphs and all
other living teleosts (Patterson, 1977; Patterson and Rosen, 1977; 12 and 13,
Figure 1.3), but this has been challenged over the last decade, notably by
Arratia (2004), who favors the theory that elopomorphs are the sister group
of osteoglossomorphs and all other living teleosts. The latter result is a
16                                                             PHILIPPE JANVIER

consequence of the new inclusion of a large number of extinct Cretaceous
and Jurassic teleost taxa in data matrices. Other phylogenies, either molecu-
lar and morphological, suggest that either the two groups form a clade, sister
to all other living teleosts (Le et al., 1993), or the relationships of osteoglos-
somorphs, elopomorphs, and the ensemble of all other living teleosts are
unresolved (Li and Wilson, 1996) (D, Figure 1.3). Elopomorph monophyly
has long been regarded as strongly supported by at least the leptocephalous
larva, supposed to be a uniquely derived condition. However, by combining
sequence‐based and morphology‐based analyses, Filleul and Lavoue (2001)   ´
have challenged elopomorph monophyly. Surprisingly, they suggest that the
leptocephalous larva may be a primitive condition for teleosts and that eels
are the sister group of osteoglossomorphs.


    Most of the reputedly primitive fish taxa referred to above have a large
number of fossil relatives. However, the fossil record for some of these
taxa remains desperately poor, as is notably the case for hagfishes and
lampreys because they lack an extensively mineralized skeleton and can
only be fossilized under particular conditions.
    The principle of a molecular clock (which followed in the wake of phenet-
ics) was based on the assumption of a constant mutation rate (now regarded
as unfounded; Maisey et al., 2004) and required accurate, paleontology‐based
calibrations of divergence times for taxa that have extant representatives.
Bracketing divergence times thus became a raison d’etre for paleontologists
working on early vertebrates (see Donoghue et al., 2003 for a review of the
question). The phylogenetic tree of any taxon that includes living and fossil
representatives comprises a ‘‘crown group’’ and a ‘‘stem group’’ (Figure 1.5).
The crown group includes the youngest common ancestor to all the living
representatives of the taxon under consideration and their respective
fossil relatives (A1, Figure 1.5). The stem group includes all the taxa that
have diverged before the common ancestor of the crown group and after
the youngest common ancestor it shares with its living sister group
(A2, Figure 1.5). For example, assuming that lampreys are the sister group
of gnathostomes, crown‐group gnathostomes include the youngest common
ancestor to a shark and a tetrapod and all its other living and fossil descen-
dants such as actinopterygians, piscine tetrapodomorphs, or extinct chondri-
chthyan taxa (A1, Figure 1.5). Stem gnathostomes include all the extinct taxa
1.   PRIMITIVE FISHES THROUGH TIME                                                                                   17


                        ey                                                us

                      pr                                               al
































                                   A2                          Pectoral fins

Fig. 1.5. The structure of taxa. When considering the living (above t0) and fossil (below t0)
representatives of a taxon, there are various ways of considering its composition and definition,
as exemplified here by some fossil and living gnathostomes. The crown‐group gnathostomes
(gray) include the most recent common ancestor of the three major living gnathostome groups,
namely chondrichthyans (e. g., Squalus), actinopterygians (e.g., Elops), and sarcopterygians
(e.g., Salamandra), as well as all their respective extinct fossil relatives (e. g., Xenacanthus,
Cheirolepis, Eusthenopteron) (A1). The stem‐group gnathostomes (light gray) include all the
extinct taxa (e.g., heterostracans, osteostracans, and placoderms) that have diverged before the
common ancestor to crown‐group vertebrates (A1) and after the most recent common to crown‐
group gnathostomes and lampreys (A2; assuming that lampreys are the sister group of gnathos-
tomes). Some stem‐group gnathostomes are jawless (heterostracans, osteostracans), whereas
others are jawed (placoderms). The total‐group gnathostomes (dark gray) include both the stem‐
and crown‐group gnathostomes, that is, all fossil or living vertebrates that are more closely
related to any member of the crown‐group than to lampreys. An apomorphy‐based taxon is
defined on the basis of at least one particular derived character that supports a node of the tree.
Apomorphy‐based gnathostomes defined on the basis of the presence of jaws would include both
the crown‐group gnathostomes and some stem gnathostomes (placoderms). [Illustrations for
terminal taxa from Janvier (1996a), by permission of Oxford University Press.]

that diverged before the divergence between any crown‐group gnathostome
but after the divergence between lampreys and the latter (A2, Figure 1.5).
Some of these stem gnathostomes have jaws (e.g., placoderms), whereas
others do not (e.g., heterostracans and osteostracans), but they all share at
least one uniquely derived character with crown‐group gnathostomes (the
most general one being the ability to produce bone). The ‘‘total group’’
includes the crown group and the stem group. The crown group and the
total group are clades (monophyletic taxa), the stem group is a grade (para-
phyletic taxon). The next question is how to name these diVerent segments of
18                                                            PHILIPPE JANVIER

a tree. This is currently a much debated matter among certain systematicists,
within the framework of the Phylocode debate (for information about the
Phylocode, see In brief, there are three
ways of considering a clade: node‐based (e.g., a crown group), stem‐based
(i.e., a total group), or apomorphy‐based (i.e., all organisms that display a
particular, uniquely derived character, e.g. pectoral fins or jaws; Figure 1.5).
Consequently and depending on which type of clade one considers for a taxon
name, the range of this taxon in time (thus its earliest occurrence) may vary
considerably. Biologists who are interested in the minimum age of characters
that cannot be directly documented by fossils (e.g., gene sequences or physio-
logical functions) generally prefer to use node‐based crown groups. However,
one must keep in mind that a node is a matter of data set, methodological
procedure, and options in phylogenetic softwares. Therefore, it may not
always refer to any conspicuous character and is often labile. Examples are
the node‐based definition of tetrapods (Laurin and Anderson, 2004), which
does not refer to the presence of digits and limbs, or the node‐based definition
of gnathostomes, which does not refer to jaws (Figure 1.5). Apomorphy‐
based clades may be of interest to ecophysiologists, as certain fossil taxa
that are members of a stem group may provide indirect evidence for a
particular function in the form of a unique combination of anatomical struc-
tures. An apomorphy‐based definition of gnathostomes would include the
presence of jaws (Figure 1.5), which actually are ecologically important
structures. In contrast, an apomorphy‐based definition of tetrapods would
include limbs with digits, supposedly important for the conquest of land, but
the taxon would also comprise such stem tetrapods as Acanthostega (see
tree in Figure 1.12) in which limbs and digits had no role in terrestrial
locomotion. The interest of stem‐based taxa is that they may provide infor-
mation about a ‘‘ghost range,’’ that is a segment of the phylogeny of a taxon
that is not documented by fossils but which must have existed because
its sister group has earlier representatives (Figure 1.6). For example, assuming
that the most general character of the total‐group gnathostomes is the
presence of bone, which occurs first in the early Ordovician, the minimum
age of this total group is about 475 Myr (Figure 1.6A). In contrast, the earliest
record of lampreys is only 360 Myr (Figure 1.6). Now, assuming that
lampreys alone are the sister group of the gnathostomes, one may infer
that lampreys and/or hagfishes have a ghost range of at least 115 Myr. This
ghost range may even be increased by 25 Myr if the character ‘‘bone’’ is
extended to ‘‘mineralized dermal skeleton’’ in general, and if one considers
that the denticles of euconodonts (‘‘conodonts’’; see Figure 1.15) actually are
evidence for dermal skeleton and thus that euconodonts are stem gnathostomes
(Figure 1.6B) (Donoghue and Sansom, 2002).
1.   PRIMITIVE FISHES THROUGH TIME                                                          19

                                  A                  B




                                   na s












Fig. 1.6. The length of the ‘‘ghost ranges’’ that are inferred from phylogenetic trees may vary
considerably, depending on the topology of the latter. Assuming that lampreys are the sister
group of the total‐group gnathostomes and that the earliest evidence for the latter are early
Ordovician arandaspid remains, the minimum age for the divergence between the two taxa is
about 475 Myr (A). The ghost range of the lampreys (in gray) is thus about 115 Myr. However, if
euconodonts are regarded as stem gnathostomes, the minimum age for the lamprey‐gnathostome
divergence would be about 500 Myr (B), and the ghost range of lampreys about 140 Myr.

5.1. Hagfishes and Lampreys

    Only two fossils, Myxinikela and Myxineides, both late Carboniferous
(305 Myr) in age, are referred to hagfishes. Myxinikela (Figure 1.2C and D)
is perhaps the most convincing fossil hagfish as it shows traces of the typical
nasal basket, prenasal sinus, and tentacles, but its body is stouter than that of
living hagfishes (Bardack, 1991). Myxineides shows no clear evidence for
tentacles, but the internal cast of its oral cavity clearly shows the imprint of
the two V‐shaped rows of horny teeth. Its body is eel‐shaped, like in modern
hagfishes (Poplin et al., 2001). Myxinikela occurs in marine sediments, but
Myxineides poses a problem, as it occurs in reputedly lacustrine sediments,
whereas hagfish physiology supposedly precludes freshwater habits.
    Two fossil lampreys, Mayomyzon and Hardistiella, are known from the
late Carboniferous (Bardack and Zangerl, 1968; Janvier and Lund, 1983).
Mayomyzon (Figure 1.2G and H) is from the same locality as the hagfish
Myxinikela, and its excellent preservation leaves little doubt about its assign-
ment. Certain specimens, preserved in lateral view, display imprints of exactly
20                                                            PHILIPPE JANVIER

the same cartilages as in modern lampreys, notably a piston cartilage, large
tectal cartilages, and an annular cartilage (Figure 1.2F and H). Another form
from the same locality, Pipiscius (Figure 1.2I and J), may also be a lamprey
and possesses an oral funnel armed with horny plates that recall those of
modern lampreys (Figure 1.2K; Bardack and Richardson, 1977). Hardistiella
is about 20‐Myr older than Mayomyzon. It is quite similar to the latter but
displays less distinct cartilage imprints. The recent discovery of the first
Devonian (360 Myr) lamprey, Priscomyzon (Gess et al., 2006), also confirmed
the remarkable conservatism of lamprey morphology through time. These
fossil lampreys only diVer from the living ones by their lack of separate dorsal
fins and their shorter branchial basket.
    It is diYcult to assign these fossil hagfishes and lampreys a particular
position relative to the members of their respective crown groups, all the
more so because there are no reliable phylogenies for the living representa-
tives of these two taxa. Nevertheless, it is likely that the Carboniferous forms
are stem hagfishes and stem lampreys, respectively. The apomorphy‐based
minimum age of these two taxa is thus 300 Myr for hagfishes and 360 Myr
for lampreys, and the stem‐based minimum age for both is from 475 to about
500 Myr (see above).

5.2. Chondrichthyans

    The morphological disparity of the living chondrichthyans is significantly
lower than the late Paleozoic ones. The trees in Figures 1.7 and 1.8 show
reconstructions of some representatives of the major Paleozoic elasmo-
branch‐ and chimaeriform‐related extinct chondrichthyan taxa. The overall
morphology of some of them is grossly sharklike (e.g., ctenacanthiforms,
eugeneodontids), but others, in particular stem chimaeriformes, display a
most unusual aspect (e.g., petalodontids, iniopterygians, chondrencheliids).
    All living elasmobranchs (sharks and rays) are neoselachians, whose ear-
liest undisputable representatives are only early Jurassic (199 Myr) in age
(3, Figure 1.7). However, on the basis of their structure, some isolated scales
and teeth suggest, albeit with great reservations, a Carboniferous and even
Devonian neoselachian record. By the early Jurassic, several major living
neoselachian higher taxa were already represented, notably batomorphs,
hexanchoids, heterodontoids, and orectoloboids, and in the late Jurassic
appear the earliest carcharinoids, lamnoids, and squatinoids (Maisey et al.,
2004). Thus, the minimum age for crown‐group neoselachians is about 200
Myr, but a number of Triassic and even Permian taxa are regarded as stem
neoselachians on the basis of various anatomical characters. Major extinct
elasmobranch clades are the Carboniferous to Cretaceous hybodontiforms, the
sister group of neoselachians, and the essentially Paleozoic ctenacanthiforms
and xenacanthiforms (Figure 1.7). The earliest evidence for elasmobranch
1.   PRIMITIVE FISHES THROUGH TIME                                                             21

            Cam 488
            Ord 443
            Dev 359
            Perm 251
            Jur 145




                                     Symmoriids and stethacanthids












                                     Batomorphs (rays, skates,
                                     torpedoes, and sawfishes)

Fig. 1.7. Elasmobranch phylogeny (the out‐group being euchondrocephalans). Geologic time-
scale in Myr, with dates referring to the limits between periods (Cam, Cambrian; Carb, Carboni-
fere; Cen, Cenozoic; Cret, Cretaceous; Dev, Devonian; Jur, Jurassic; Ord, Ordovician; Perm,
Permian; Sil, Silurian; Tr, Triassic). Paleozoic in white, Mesozoic in light gray, Cenozoic in dark
gray. Horizontal bars in the timescale indicate the distributions of the taxa through time. Inter-
relationships of batomorphs not detailed. Carcharhinoids include scylliorhinids. Symmoriids and
22                                                                          PHILIPPE JANVIER

based on articulated material is from the early Devonian (400–410 Myr)
(Figures 1.7 and 1.15).
    Living holocephalans, or chimaeriforms, are the relicts of a very diverse
ensemble of chondrichthyans, referred to as ‘‘euchondrocephalans’’ (Grogan
and Lund, 2004) (1, Figure 1.8). Living chimaeriform overall morphology is
certainly much derived, relative to the general chondrichthyan morphology,
but it appeared early in time, probably in the late Paleozoic (e.g., Echinochi-
maera, Figure 1.8). The living chimaeriforms only consists of six genera
distributed in three families: the Callorhinchidae, Rhinochimaeridae, and
Chimaeridae, the former being sister group to the latter two (Didier, 2004)
(3, Figure 1.8). The phylogeny of the crown‐group chimaeriforms is thus
quite clear, but informative fossils referred to these three families are rela-
tively rare, hence the diYculty to infer their minimum age. Nevertheless the
earliest representatives of the crown‐group chimaeriforms are late Jurassic
callorhinchids, and egg capsules referred to rhinochimaerids are known since
the early Triassic (245 Myr) (Stahl, 1999). Stem chimaeriforms are known
since the late Devonian (370 Myr) and include notably the ‘‘bradyodonts’’
(e.g., chondrenchelyids), a paraphyletic ensemble of Paleozoic taxa which
display chimaeriform‐like features such as holostyly and tubular tooth
    Apart from the in‐group phylogeny of elasmobranchs and chimaeriforms,
chondrichthyan phylogeny remains tenuously supported, and the phylogenetic
position of a number of fossil higher taxa, such as cladoselachids, ede-
stids, petalodontids, symmoriids (including stethacanthids), inopterygians, or
Pucapampella, remain extremely labile (Figures 1.7, 1.8, and 1.15).

5.3. Actinopterygians

    The fossil record of cladistians is very poor and the earliest evidence
for this taxon are isolated scales, dorsal fin pinnules, and vertebrae from
the early Cretaceous of Africa. Serenoichthys, from the late Cretaceous
(98 Myr) freshwater deposits of Morocco (Dutheil, 1999), is the only fossil
cladistian known from articulated specimens (Figure 1.9). It is remarkably
similar to modern cladistians, except for its stouter body shape, but certainly
represents a stem cladistian as it is not the sister group of one or the
other living cladistian genera, Polypterus and Erpetoichthys. Considering

stethacanthids are regarded as single clade, but their relationships remain debated, and it is not
inconceivable that they are in fact basal euchondrocephalans (Janvier, 1996a). The clade that
includes xenacanthiforms and all the more crownward taxa is better supported, and is therefore
regarded as the most reliable total‐group elasmobranchs. 1, elasmobranchs; 2, euselachians; and 3,
neoselachians. [Illustrations for terminal taxa from Janvier (1996a), by permission of Oxford
University Press.]
1.   PRIMITIVE FISHES THROUGH TIME                                                           23

             Ord 443
             Dev 359
             Perm 251
             Jur 145











Fig. 1.8. Euchondrocephalan phylogeny (the out‐group being elasmobranchs). For abbrevia-
tions to the geologic timescale, see Figure 1.7. 1, euchondrocephalans; 2, holocephalans; and 3,
chimaeriforms. [Illustrations for terminal taxa from Janvier (1996a), by permission of Oxford
University Press.]

that (according to current morphology‐based actinopterygian phylogenies)
cladistians are supposed to have diverged from actinopterans before such
late Devonian taxa as Moythomasia (Coates, 1999; Cloutier and Arratia,
2004) and, at any rate, before the earliest crown‐group actinopterans
(2 and 4, Figure 1.9), the rather late occurrence of the earliest cladistians
remains a riddle. It is possible that earlier (Paleozoic) stem cladistians are
in fact known, but we do not recognize them because they lack conspic-
uous cladistian characters. Lund (2000) suggested that the Carboniferous
guildayichthyiform actinopterygians are the fossil sister group of cladistians,
24                                                                                  PHILIPPE JANVIER

         Cam 488
         Ord 433
         Dev 359
         Perm 251




                                                         3                               1

                                 Moythomasia                                         2

                                 Chondrosteus                                   4



                                 Ginglymods                             8

                                                             11 9

                                 Caturids                 12



                                 Crown-group teleosts

Fig. 1.9. Actinopterygian phylogeny. For abbreviations to the geologic timescale, see Figure 1.7.
The assignment of Dialipina to the actinopterygians still rests on tenuous characters, notably
scale histology. In this tree, we regard as undoubted actinopterygians Cheirolepis and all
more crownward taxa. 1, actinopterygians; 2, crown‐group actinopterygians; 3, cladistians
1.   PRIMITIVE FISHES THROUGH TIME                                                           25

but this rests on a very small number of characters, which are frequently
homoplastic among early actinopterygians (Cloutier and Arratia, 2004).
    The earliest crown‐group acipenseriforms are late Jurassic polyodontids
and late Cretaceous acipenserids (Figure 1.9). The minimum age for the
crown‐group acipenseriforms is thus about 150 Myr (Bemis et al., 1997).
However, stem acipenseriformes, such as Chondrosteus (Figure 1.9), are
known since the early Jurassic, and certain extinct taxa, such as the early
Triassic (250 Myr) Birgeria (Figure 1.9) and saurichthyids, are possibly more
closely related to the acipenseriforms than to any other actinopterygian
group (6 and 7, Figure 1.9).
    Living gars (ginglymods) fall into two genera, Lepisosteus (four species)
and Atractosteus (three species), both of which are known since the
late Cretaceous (65–100 Myr) (Figure 1.9). The stem lepisosteid Obaichthys
is early Cretaceous in age, and the minimum age of the crown‐group gingly-
mods is thus about 112 Myr. The earliest known gars display much the same
morphology as the extant ones. Stem gars are represented by the early
Triassic to late Cretaceous (250–65 Myr) semionotids, a probably para-
phyletic taxon that includes large, ubiquitous fishes with thick ganoid scales
(10, Figure 1.9).
    The bowfin (A. calva) is represented by a single living species, which is
thus the crown group, and is recorded since the early Pleistocene (about
1 Myr). However, the more and more inclusive taxa Amiinae, Amiidae,
Amiiformes, and Halecomorphi comprise a large number of fossil marine
and freshwater species. The earliest Amiinae are known since the early
Cretaceous and were already exclusively freshwater, but the earliest haleco-
morphs, the middle Triassic (230 Myr) paraseminotids, were exclusively
marine (Grande and Bemis, 1999) (11 and 12, Figure 1.9).
    Osteoglossomorphs and elopomorphs are known since the late Jurassic
(elopomorphs having a slightly older first occurrence than osteoglosso-
morphs) and include a large number of fossil species. The debate about
their relationships to other teleosts arose with the interpretation of certain
fossil representatives of those respective taxa, as well as that of some stem
teleosts (Arratia, 2004). The minimum age for crown‐group teleosts is thus
indicated by the earliest elopomorphs, that is, about 150 Myr. However,
there is a large number of stem teleost taxa the earliest of which, such as
Pholidophorus (Figure 1.9), are late Triassic (215 Myr) in age.

(Polypteriformes); 4, actinopterans; 5, crown‐group actinopterans; 6, chondrosteans; 7, Acipen-
seriformes; 8, neopterygians; 9, crown‐group neopterygians; 10, Semionotiformes; 11, haleco-
morphs; 12, Amiiformes; and 13, teleosts. [Illustrations for terminal taxa from Janvier (1996a),
by permission of Oxford University Press, except for the reconstruction of Dialipina, which is
based on photographs in Schultze and Cumbaa (2001).]
26                                                                        PHILIPPE JANVIER

5.4. Sarcopterygians

    Apart from the two living species of Latimeria, all other actinistians
are fossils and thus stem actinistians (Figure 1.10). The earliest known
actinistians were long believed to date from the middle Devonian only, but
their position within sarcopterygian phylogeny entailed a rather long ‘‘ghost
range’’ because their living sister group, the rhipidistians (i.e., dipno-
morphs and tetrapodomorphs), are known since the base of the Devonian
          Cam 488
          Ord 443
          Dev 359
          Perm 251












               No fossil known


Fig. 1.10. Actinistian (coelacanth) phylogeny. For abbreviations to the geologic timescale, see
Figure 1.7. 1, actinistians; 2, coelacanthiformes. [Illustrations for terminal taxa from Janvier
(1996a) by permission of Oxford University Press, except for that of Holopterygius, redrawn and
modified after Friedman and Coates (2005).]
1.   PRIMITIVE FISHES THROUGH TIME                                                          27

(Figure 1.15). However, there is now some evidence for actinistian remains in
the early Devonian (400 Myr) (Johanson et al., 2006). Although the group
superficially shows relatively little morphological disparity (in particular since
the beginning of the Mesozoic), some Paleozoic coelacanths display strongly
divergent morphologies such as the eel‐shaped middle Devonian Holoptery-
gius or the deep‐bodied Carboniferous Allenypterus (Friedman and Coates,
2005) (Figure 1.10). There is no fossil actinistian record between the late
Cretaceous (70 Myr) and present.
    Lungfishes are among the earliest known osteichthyans (2, Figure 1.11),
as they are recorded since the beginning of the Devonian (415 Myr) and show
a spectacular radiation during the Devonian, with a majority of marine
         Cam 488
         Ord 443
         Dev 359
         Perm 251












Fig. 1.11. Dipnomorph phylogeny. For abbreviations to the geologic timescale,
see Figure 1.7.1, Dipnomorphs; 2, dipnoiforms; 3, dipnoans; and 4, crown‐group dipnoans.
[Illustrations for terminal taxa from Janvier (1996a, by permission of Oxford University Press,
28                                                            PHILIPPE JANVIER

forms, often associated with coral reef environments. Lungfish diversity
declines progressively during the Carboniferous, when most species occur
in reputedly freshwater environments. After the Paleozoic, lungfishes under-
go a considerable reduction of their dermal skeleton and are essentially
known by poorly informative isolated tooth plates. Therefore, the relation-
ships of the Mesozoic taxa to the living Neoceratodus and lepidosirenids
remain unclear, except for some species. It is generally assumed that the
earliest tooth plates of Neoceratodus type occur in the early Cretaceous, and
tooth plates of lepidosirenid type are known in the late Cretaceous. The
minimum age for crown‐group lungfishes is thus about 140 Myr. The earliest
lungfish estivation burrows are from the late Permian.
    Although this book is about fishes, it must be kept in mind that tetrapods
are only part of a large total group called tetrapodomorphs and which
includes a number of piscine taxa (Figure 1.12). The paleobiology of
the piscine tetrapodomorphs has little bearing on the biology of the living
tetrapods, but these extinct taxa will be dealt with briefly in the next section.


    Besides these living primitive taxa and their extinct closest relatives,
vertebrates include a number of major extinct clades, which are shown in
the trees in (Figures 1.13 and 1.15). Although these will not be discussed in
detail here, they deserve some comment because they illustrate the most
interesting property of fossils, that is they provide character combinations
that no longer exist in present‐day nature and thus often help in resolving
conflicting relationships between the major living taxa.

6.1. Yunnanozoans and Myllokunmingiids

    Strangely, there are very few stem vertebrates (i.e., fossil vertebrates that
have diverged earlier than the most recent common ancestor to all living
vertebrates, and after the divergence between vertebrates and either cepha-
lochordates or tunicates). There is no fossil cephalochordate, despite certain
claims (Blieck, 1992; Janvier, 1997), but there are undoubted tunicates
in the lower Cambrian Chengjiang marine Lagerstatte of Yunnan, China
(535 Myr) (Shu et al., 2001). Two taxa from the same fossil locality, Yunna-
nozoa and Myllokunmingiida, have been regarded as possible stem verte-
brates that ‘‘filled’’ the morphological gap between cephalochordates and
vertebrates. Yunnanozoans possess a distinct branchial apparatus with six
pairs of filamentous gills, followed posteriorly by a series of vertical body
1.   PRIMITIVE FISHES THROUGH TIME                                                          29

         Ord 443
         Dev 359
         Perm 251










                                  lchthyostega                     2

                                  Crown-group tetrapods

Fig. 1.12. Tetrapodomorph phylogeny. For geologic timescale, see Figure 1.7. 1, Tetrapodo-
morphs; and 2, apomorphy‐based tetrapods. [Illustrations for terminal taxa from Janvier (1996a,
by permission of Oxford University Press, and 2004b), except for Tiktaalik, redrawn and
modified from Daeschler et al. (2006).]

muscle blocks. Despite Mallatt and Chen’s (2003) attempts to interpret other
structures of the yunnanozoan’s head with regard to larval lamprey anatomy,
there is no clear evidence that this taxon is more closely related to the verte-
brates than to any other chordate or deuterostome taxon. Yunnanozoans have
been assigned to a wide range of phylogenetic positions, notably stem verte-
brates, stem cephalochordates, stem hemichordates, and finally stem deuter-
ostostomes (see review in Janvier, 2003). It is possible that Pikaia, from the late
Cambrian Burgess Shale Lagertstatte, long popularized as a the earliest chor-
date (Conway Morris, 1998), is in fact a close relative of yunnanozoans
(Janvier, 2003). In contrast, myllokunmingiids (Figure 1.13) are more likely
to be stem vertebrates. Although preserved as imprints, they display most of the
30                                                                        PHILIPPE JANVIER

      Cam 488
      Ord 443
      Dev 359
      Perm 251




                               Lampreys                                       1

                               Euphaneropids                              2



                               Astraspids                  3





                               Jawed vertebrates

Fig. 1.13. Vertebrate phylogeny. For abbreviations to the geologic timescale, see Figure 7. 1,
vertebrates/craniates; 2, crown‐group vertebrates/craniates; and 3, pteraspidomorphs. The name
‘‘ostracoderms,’’ a grade of basal, jawless gnathostome‐related vertebrates, generally refers to
anaspids, pteraspidomorphs, thelodonts, galeaspids, pituriaspids, and osteostracans. The mono-
phyly of the thelodonts remains poorly supported. [Illustrations for terminal taxa from Janvier
(1996a by permission of Oxford University press, and in press).]

vertebrate characters such as olfactory organs, eyes,and possibly otic capsules,
which are acceptable evidence for neurogenic placodes, as well as six gill arches,
which may be evidence for migrating neural crest cells. They also display
W‐shaped myomeres that resemble more those of vertebrates than those of
1.   PRIMITIVE FISHES THROUGH TIME                                         31

cephalochordates (Shu et al., 2003). Although myllokunmingiids were initially
regarded as a paraphyletic group (Shu et al., 1999), including stem lampreys
(Myllokunmingia) and the sister group of lampreys and all other vertebrates
except for hagfishes (Haikouichthys), they are now regarded as probably a
monophyletic group; they lack endoskeletal fin radials, which are a character
of crown‐group vertebrates (Janvier, 2003, in press; Zhang and Hou, 2004).

6.2. ‘‘Ostracoderms’’
    The name Ostracodermi was coined by Cope (1889) for lower Paleozoic
vertebrates (essentially heterostracans and osteostracans) that lack jaws but
possess a heavily ossified dermal skeleton, like gnathostomes. These armored
jawless vertebrates are regarded as stem gnathostomes (Figures 1.5 and 1.13),
but the name ‘‘ostracoderm’’ is still used informally by some authors. Eight
major taxa are commonly referred to as ‘‘ostracoderms.’’ In addition to
heterostracans and osteostracans, ‘‘ostracoderms’’ include anaspids, galeas-
pids, arandaspids, eriptychiids, pituriaspids, and thelodonts. Even some
partly or entirely soft‐bodied jawless vertebrate taxa (such as euconodonts
and euphaneropids; Figure 1.13) are now sometimes referred to as ‘‘ostraco-
derms’’ as they are regarded as stem gnathostomes. At the beginning of the
twentieth century, ‘‘ostracoderms’’ were thought to be ancestral to hagfishes
and lampreys essentially because they were jawless and ancient. Stensio      ¨
(1927) was the first to unravel the details of the internal anatomy of osteos-
tracans and pointed out some unique characters they (and also anaspids)
shared with lampreys, notably the presence of a median, dorsal nasohypo-
physial opening. Stensio also proposed that hagfishes were derived from yet
another ‘‘ostracoderm’’ group, heterostracans, although this idea is now
discarded. Stensio’s conception of vertebrate phylogeny was that agnathans
are monophyletic and that ‘‘ostracoderms’’ are ancestral to the modern
cyclostomes as a whole. This theory was also coherent with the old idea
that hagfishes and lampreys were ‘‘degenerate’’ and had lost numerous
characters such as the paired fins and mineralized skeleton. Some authors,
however, suggested that certain ‘‘ostracoderms,’’ notably heterostracans,
could be ancestral to the gnathostomes because they possessed paired olfac-
tory organs (Halstead, 1973). During the last two decades, evidence has
accumulated to indicate that ‘‘ostracoderms,’’ though undoubtedly lacking
jaws, are in fact more closely related to the jawed vertebrates than to either
hagfishes or lampreys (Gagnier, 1991; Forey and Janvier, 1993; Forey, 1995;
Janvier, 1996b; Donoghue et al., 2000; Donoghue and Smith, 2001). As a
consequence, the classical assumption that hagfishes and lampreys had lost
many characters became pointless, as the presence of bone, along with other
characters uniquely shared by ‘‘ostracoderms’’ and gnathostomes, came to
32                                                            PHILIPPE JANVIER

be considered as synapomorphies. Although some characters, such as the
strikingly similar condition of the nasohypophysial complex in lampreys and
osteostracans (and probably anaspids), have to be regarded as homoplastic,
the current tree of fossil and living vertebrates is, if not robust, at any rate
clearly more parsimonious than any tree assuming agnathan monophyly
(Donoghue et al., 2000). It shows ‘‘ostracoderms’’ as a grade of stem
gnathostomes that fills the morphological gap between the living cyclostomes
(or the living lampreys, if cyclostomes are not a clade) and the living
gnathostomes (Figures 1.5 and 1.13).
    The most interesting implication of the current vertebrate tree is that
what is often referred to as the so‐called ‘‘gnathostome body plan’’ is not the
result of a ‘‘burst of anatomical innovations,’’ as frequently alleged, but a
progressive accumulation of new characters that can be inferred at several
nodes of the tree (Mazan et al., 2000; Janvier, 2001; Donoghue and Sansom,
2002). For example, this tree tells us that the dermal skeleton became ossified
before the endoskeleton or that paired fins appeared before jaws. Osteostra-
cans, which now appear as the closest jawless fossil relatives of the jawed
vertebrates, show that such characters as cellular bone, perichondral bone,
sclerotic ring, pectoral fins, shoulder girdle, and epicercal tail have also
preceded jaws in vertebrate evolution (Figure 1.13). Of course, we know
nothing of the evolution of physiology along this long segment of vertebrate
phylogeny that extends between the divergence of lampreys (or the cyclos-
tomes as whole) and crown‐group gnathostomes, and that is only documen-
ted by fossil taxa (apart from functions linked to calcified tissue
development), but one might expect that some anatomical or histological
data might inform insights in this field. The input is now expected from
physiologists, rather than from paleontologists.
    During the last decade, considerable attention has been paid to ‘‘cono-
donts’’ (or, more restrictively, euconodonts; Figure 1.13), which are currently
regarded by many paleontologists as the basalmost stem gnathostomes.
From the mid‐nineteenth century, conodonts (including at least paracono-
donts and euconodonts) were only known by minute comb‐ or tooth‐shaped
denticles of unknown derivation. They always occur in marine sediments
from the late Cambrian to the Triassic (500–220 Myr) and were widely
used by stratigraphers for dating the rocks. Since the first discovery of the
soft‐tissue imprint of articulated euconodonts from the Carboniferous and
Ordovician (Briggs et al., 1983; Gabbott et al., 1995), it became widely
accepted that euconodonts are vertebrates as they display a caudal fin
supported by radials, a body musculature made up by chevron‐shaped
myomeres, and a pair of large eyes. Their mouth is armed with denticles
made of calcium phosphate, whose structure somewhat recalls the verteb-
rate bone, dentine, and enamel (though this homology is far from widely
1.   PRIMITIVE FISHES THROUGH TIME                                          33

accepted; Schultze, 1996). The phylogenetic position of euconodonts as stem
gnathostomes remains tenuously supported, and they may turn out to be
either more closely related to hagfishes or lampreys (or cyclostomes as a
whole), or even stem vertebrates. Euconodonts are, however, interesting
because of their extraordinary abundance and specific diversity during the
Paleozoic. While most of the classical ‘‘ostracoderm’’ groups are relatively
rare and have a relatively low specific diversity, the contemporary eucono-
donts comprise hundreds of species, despite their low morphological dispari-
ty. In a sense, they could compare to rodents among the living mammals.

6.3. Placoderms

    Placoderms, or armored jawed vertebrates (Figures 1.14 and 1.15), lived
from the early Silurian to the very end of the Devonian (435–360 Myr). They
are characterized by a massive armor made up by bony plates that cover the
anterior part of the trunk and the head. The trunk armor articulates with the
head armor at the level of the neck. The jaws are armed with bony, denticle‐
bearing, dermal plates. Placoderms were the most diverse and abundant fish
group during the Devonian, and their extinction at the end of this period is as
sudden and enigmatic as that of the nonavian dinosaurs at the end of the
Cretaceous. Geologic events, such as an extensive marine regression at the
end of the Devonian, may have been the coup de grace that caused their
disappearance, but they may also have been outcompeted by the sudden
burst of the chondrichthyans, notably the durophagous, chimaeriform‐
related ‘‘bradyodonts.’’
    The diversity of placoderms during the Devonian may compare to that of
living teleosts, with about 400 known species ranging from 10 mm to 6–7 m
in total length (adult size), and morphologies that vaguely parallel those of
rays, catfishes, boxfishes, tunas, or swordfishes (Figure 1.14). Placoderms
were once regarded as either chondrichthyan relatives because of their neu-
rocranial anatomy, or osteichthyan relatives because of their dermal bone
pattern, but they are currently regarded as the sister group to all other jawed
vertebrates, based on various morphological characters (Figure 1.15).

6.4. Acanthodians

    Acanthodians are a poorly understood group of late Ordovician to early
Permian (445–295 Myr) gnathostomes (Figure 1.15), whose exoskeleton
generally consists of minute, square‐shaped scales covered with overgrowing
layers of dentine. They display significant specific diversity, but comparatively
low morphological disparity. They have long been characterized by the
presence of large fin spines in front of the paired and unpaired fins until the
34                                                                          PHILIPPE JANVIER


                           Antiarchs                            Rhenanids


Fig. 1.14. Placoderm diversity. Placoderms are one of the two major extinct jawed vertebrate clades
and were among the most abundant fishes during the Devonian. These reconstructions of repre-
sentatives of four main placoderm groups illustrate their morphological diversity. The antiarchs
(e.g. Asterolepis) possessed extensive trunk and head armors, and their pectoral fins were covered
with bony plates and articulated in a crab leg‐like manner. The rhenanids (e.g., Gemuendina)
displayed a ray‐ or squatinoid‐like overall morphology. Arthrodires (e.g., Coccosteus), the most
diversified placoderm group, possessed a massive dermal armor and could grow up to 7 m in total
length. In contrast, ptyctodonts (e.g., Ctenurella) had a much reduced dermal armor, and their
overall shape recalls that of living chimaeriformes. Scale bar ¼ 10 mm. [From Janvier (1996a) by
permission of Oxford University Press, Janvier and Racheboeuf (2003), and Janvier et al. (2003).]

discovery of paired fin spines in early Devonian primitive chondrichthyans
(Miller et al., 2003). Unfortunately, very little is known of acanthodian
anatomy, in particular for the endoskeleton, except in the Permian genus
Acanthodes, which is the last surviving taxon and shares some unique derived
characters with osteichthyans, notably three otoliths. The status of acantho-
dians is debated, and they are increasingly suspected to be nonmonophyletic.
If Acanthodes (and acanthodids as a whole) are still regarded as a potential
1.   PRIMITIVE FISHES THROUGH TIME                                                            35

         Ord 443
         Dev 359
         Perm 251



                                                                  3                       1


                                 Euchondrocephalans (incl. chimaeriforms)


                                 Dialipina                                        5

                                 All other actinopterygians
                                 Meemannia                                    6






Fig. 1.15. Jawed vertebrate phylogeny. For abbreviations to the geologic timescale, see
Figure 1.7. The monophyly of acanthodians is currently debated and the position of Dialipina
is tenuously supported. Meemannia, Psarolepis, and Styloichthys are known from fragmen-
tary material and are not reconstructed here. 1, Jawed vertebrates (apomorphy‐based gnathos-
tomes); 2, crown‐group gnathostomes; 3, chondrichthyans; 4, crown‐group chondrichthyans;
36                                                                          PHILIPPE JANVIER

sister group to osteichthyans, there are increasingly numerous data, notably
histological ones, suggesting that certain Silurian and early Devonian
acanthodian‐like taxa are in fact stem chondrichthyans. Thus, acanthodians
are likely to turn out to be a paraphyletic group, comprising both stem
osteichthyans, stem chondrichthyans, and possibly stem gnathostomes
(Janvier, 1996a).

6.5. ‘‘Paleoniscoids’’ and Basal Neopterygians

    The earliest known undoubted actinopterygians are early Devonian (410
Myr) in age, although certain late Silurian (420 Myr) remains (e.g., Andreo-
lepis, Lophosteus) have been tentatively referred to actinopterygians, but may
in fact be derived from stem osteichthyans. Basal actinopterygian and basal
actinopteran relationships remain heatedly debated and extremely labile
(see for review Cloutier and Arratia, 2004). During the past three decades,
a number of trees have been proposed for the total‐group actinopterygians
(Patterson, 1982; Gardiner, 1984; Long, 1988; Gardiner and SchaeVer, 1989;
Gardiner et al., 1996; Coates, 1999; Cloutier and Arratia, 2004). These trees
diVer in many respects, although the relationships between the extant taxa
generally remain the same. A large number of taxa that have long been
referred to as ‘‘paleonisciforms’’ or ‘‘paleoniscoids,’’ an obviously paraphy-
letic taxon composed of Devonian to Jurassic taxa, are in fact a mix of stem
actinopterygians (e.g., Cheirolepis), stem actinopterans (e.g., Moythomasia),
and stem neopterygians (e.g., Australosomus) (Figure 1.9). That is, they
diverged before either cladistians, acipenseriforms, or crown‐group neopter-
ygians. It is also probable that some of the ‘‘paleonisciforms,’’ such as the
platysomids, may be gathered into large extinct actinopterygian clades.
The diVerences between the various tree topologies published during the
past three decades rest mainly on the minimum age of total groups. For
example, depending on the earliest fossil sister group of the total‐group
acipenseriforms, the minimum age for the divergence between acipenseri-
forms and neopterygians may be dated as either early Triassic or early
Carboniferous (Coates, 1999).
    One of the most intriguing early actinopterygians is certainly the early
Devonian genus Dialipina (Figures 1.9 and 1.15), which possesses typical
actinopterygian ‘‘ganoid’’ scales but retains two dorsal fins (all other
actinopterygians have a single dorsal fin—a derived character), possesses a
diphycercal caudal fin that resembles that of coelacanths, and shows a dermal

5, osteichthyans (or teleostomes); 6, crown‐group osteichthyans; 7, actinopterygians; 8, sarcopter-
ygians; 9, crown‐group sarcopterygians; 10, rhipidistians; and 11, crown‐group rhipidistians (or
‘‘choanates’’) [Illustrations for terminal taxa from Janvier (1996a) by permission of Oxford
University Press, Janvier and Maisey (in press), and the reconstruction of Dialipina is based on
photographs in Schultze and Cumbaa (2001).]
1.   PRIMITIVE FISHES THROUGH TIME                                            37

bone pattern that is strongly at odds with that of all other osteichthyans,
notably because it lacks the classical osteichthyan cheek bones (opercular,
preopercular jugal, and so on; Schultze and Cumbaa, 2001). The peculiar
anatomy of Dialipina is interpreted as a mix of general gnathostome characters
(e.g., two dorsal fins) and general osteichthyan or actinopterygian char-
acters, with some unique highly derived features. However, it is possible that
Dialipina tells us that some of the characters long used to define actinopter-
ygians (e.g.,the ganoid scales) are in fact more general than currently believed.

6.6. Extinct Sarcopterygian Taxa
    Extinct nontetrapod sarcopterygian taxa are relatively numerous, notably
in the Devonian (415–360 Myr). However, their relationships are comparably
better elucidated than for extinct actinopterygian taxa. Most of these piscine
sarcopterygians fall into two major clades of the crown‐group sarcoptery-
gians, the dipnomorphs and tetrapodomorphs (Figure 1.15), which are cur-
rently regarded as sister groups (10, Figure 1.15).
    Dipnomorphs include dipnoans (lungfishes) and porolepiforms
(Figure 1.11), as well as some monogeneric taxa (e.g., Powichthys, Youngolepis)
that are variously regarded as either basal lungfishes or porolepiforms. The
most generalized lungfish (or dipnoiform), Diabolepis, from the early Devonian
of China already possesses the characteristic entopterygoid and prearticular
tooth plates, but retains (at any rate ventrally) the characteristic intracranial
joint of basal sarcopterygians and the living Latimeria (Figure 1.1), and two
external nostrils.
    Tetrapodomorphs are notably characterized by the choana, which is as-
sumed to be the posterior nostril that has migrated into the palate, though
independently from the intrabuccal posterior nostril of lungfishes (Zhu and
Ahlberg, 2004; Janvier, 2004b). They include Kenichthys, the rhizodontids,
osteolepids, megalichthyids, tristichopterids, elpistostegalians, and tetrapods
(Figure 1.12). Osteolepids, megalichthyids, and tristichopterids were formerly
included in a taxon ‘‘Osteolepiformes,’’ now proved to be paraphyletic
(Ahlberg and Johanson, 1998), and so are also probably elpistostegalians,
that include Panderichthys and Tiktaalik (Figure 1.12) (Daeschler et al., 2006).
    Onychodontiforms are only known from marine Devonian sediments and
display a curious assemblage of sarcopterygian characters (intracranial joint),
actinistian characters (unlobed anterior dorsal fin and diphycercal tail, also
shared with Dialipina), and actinopterygian‐like characters (large and poste-
riorly expanded maxilla). They are currently regarded as the sister group of
either actinistians or crown‐group sarcopterygians (i.e., actinistians and rhi-
pidistians) (Figure 1.15).
    In addition to these taxa, the early Devonian of China has yielded a
number of sarcopterygians that are referred to monotypic genera, such as
38                                                           PHILIPPE JANVIER

Meemannia, Psarolepis, Achoania, Styloichthys, and Kenichthys, and current-
ly regarded as either stem sarcopterygians, stem rhipidistians, or stem tetra-
podomorphs (Figure 1.15) (Zhu et al., 1999, 2001, 2006; Zhu and Yu, 2002;
Zhu and Ahlberg, 2004). Psarolepis, in particular, is a puzzling form that
allies classical sarcopterygian characters (e.g., folded dentine in teeth, cos-
mine, intracranial articulation), some reputedly actinopterygian characters
(e.g., structure of the preopercular, median rostral bone), and characters that
occur outside osteichthyans, in placoderms, chondrichthyans, and acantho-
dians (e.g., median fin spines, placoderm‐like shoulder girdle), the latter
being probably general gnathostome characters that have been lost in all
other osteichthyans. Therefore, it has been suggested that Psarolepis is a stem
osteichthyan (Zhu et al., 1999). Phylogenetic analyses would, however, rath-
er place it as a stem sarcopterygian (Zhu and Yu, 2002; Zhu et al., 2006).
Meemannia is yet another lower Devonian stem sarcopterygian that is mor-
phologically very close to the currently accepted (or imagined) ancestral
morphotype of actinopterygians and sarcopterygians (Zhu et al., 2006).
Interestingly, these five taxa all occur in the same locality of South China
and range in age from 416 to 400 Myr. This suggests that the minimum
age for the divergence between actinopterygians and osteichthyans is about
416 Myr.


    There is a widespread belief that, besides some problematical nodes
(see above), the vertebrate tree is relatively stable at the level of the major
extant terminal taxa. The current consensus about vertebrate phylogeny is a
consequence of an overall agreement between morphology‐based and most
molecular sequence‐based phylogenies. There are, however, some discordant
views about vertebrate phylogeny, the most striking incongruence with
morphology‐based trees being Arnason et al.’s (2001, 2004) consideration
of mitochondrial DNA sequence data (Figure 1.4). This gnathostome tree
has been the subject of little debate because it was regarded as simply the
result of methodological biases, notably the inadequacy of mitochondrial
DNA for resolving very deep divergences (Zardoya and Meyer, 2001).
    Could some paleontological data accommodate Arnason et al.’s (2004)
tree? The latter raises two major questions: (1) the monophyly of actinopter-
ygians (cladistians þ actinopterans) and (2) the monophyly of sarcoptery-
gians. Cladistians still trouble molecular phylogenies but are grouped with
actinopterans into the actinopterygians on the basis of a few morphological
characters. Morphologists classically regard sarcopterygians as a well‐
supported clade. During the last decade, new early Devonian sarcopterygians
1.   PRIMITIVE FISHES THROUGH TIME                                           39

have raised questions about sarcopterygian monophyly, as currently defined.
Notably, Psarolepis (see above) displays some classical sarcopterygian char-
acters, along with characters that are only known in non‐osteichthyan
gnathostomes. The early Devonian chondrichthyan Pucapampella, one
of the earliest known chondrichthyans, also raises the question of the distri-
bution of certain osteichthyan morphological characters. Although it pos-
sesses the characteristic prismatic calcified cartilage of chondrichthyans, the
braincase of Pucapampella also displays a complete ventral fissure (once
regarded as the homologue of the sarcopterygian intracranial joint).
In addition, the overall braincase morphology of Pucapampella is strikingly
similar to that of the earliest known actinopterygian and sarcopterygian
braincases. In a diVerent (nonphylogenetic) context, Jarvik (1981) considered
that sarcopterygian skull morphology was an image of the ideal gnathostome
skull morphology. At any rate, it now seems that we are coming closer and
closer to the ancestral morphotype of the crown‐group gnathostomes, and
that endoskeletal cranial characters suggest that it was more similar to the
osteichthyan condition than to the modern chondrichthyan one. Although
this is not enough to refute the current gnathostome phylogeny, it does show,
at least, that its morphological bases are not as stable as once believed.
Notably, it addresses the question of the reputed primitiveness of the chon-
drichthyans, as did Arnason et al.’s (2001) results. The paleontological
‘‘black box’’ of crown‐group gnathostome divergences seems thus situated
in time somewhere in the Silurian period between 440 and 415 Myr ago.


    Fossils have only three unique properties that no other source of
biological data can replace: (1) they provide morphological character com-
binations which no longer exist in living organisms; (2) they provide a
minimal age for characters, and thus the taxa that they define; and (3) they
may show geographical (and ecological) distributions that are diVerent from
the present day. Property (1) is essentially used by phylogeneticists as it may
help in resolving (or, at any rate, better support) relationships between extant
taxa by answering questions about homologies or homoplasies. To evolu-
tionary morphologists, it provides material evidence for ‘‘transitional forms’’
in such scenarios of evolutionary transition, as the ‘‘agnathan–gnathostome’’
or the ‘‘fish–tetrapod’’ transitions. In rare cases, this property may allow
inferences about the distribution of physiological characters, when reflected
in hard tissues, such as the presence of bone in a jawless vertebrate, which
suggests that the physiological functions involved in bone production did
exist before the rise of jaws. Property (2) is now much used by molecular
40                                                            PHILIPPE JANVIER

phylogeneticists for calibrating molecular clocks (however reliable they may
be), but may be of interest to physiologists to explore the stability of a
physiological character through time. Finally, property (3), which is mainly
useful to historical biogeographers, may have bearings on inferences about
the past physiological adaptations of living taxa. For example, physiologists
are sometimes surprised to learn that, until about 370 Myr, all lungfishes
were marine.
    In sum, very few fossil data are directly informative to primitive verte-
brate physiologists. Generally, they concern the physiology of hard tissues
(bone, teeth, calcified cartilage) and are relevant to calcium regulation. Some
anatomical data that can be linked to particular functions (e.g., evidence
for a pineal foramen that suggests the presence of a photosensory pineal
organ, or pelvic claspers that suggest internal fecundation) can provide a
minimum age for the latter by inference from one particular phylogenetic
tree. In fact, questions about the evolution of physiological functions arise
essentially from their distribution in living taxa, and the role of fossils is
to provide a basis for the calibration of the divergence times of these taxa.
In a sense, evolutionary physiology is in much the same situation as mole-
cular phylogenetics, relative to paleontological data. The literature about
fossil fishes probably contains many morphological data on skeletal struc-
tures or exceptional soft‐tissue preservations that may be relevant to physio-
logy, but paleontologists are rarely able to properly assess their bearings on
this field.
    Traces of activity left by fossil fishes are very rare and poorly informative
as to the mode of living. Fish trails in the sediment, referred to as Undichna,
that were left by the fins or fin spines of fishes trapped in shallow pools,
can hardly be assigned to a particular taxon. The most interesting of these
traces are those clearly left by early Devonian osteostracans, as they repre-
sent a unique source of information about the locomotion of an ‘‘ostraco-
derm’’ and show that the fish moved by short successive ‘‘jumps’’ (Morrissey
et al., 2006). Lungfish estivation burrows are known from the Permian and
can be readily identified by the articulated lungfish skeletons found inside
them (McAllister, 1992). Stomach contents are frequently found in fossil
fishes and provide some information about their diet. They also allow to
reconstruct the original trophic network (Maisey, 1994). The stomach
contents of ‘‘ostracoderms’’ (known in anaspids, euphaneropids, and thelo-
donts) consist of fine‐grained sediment, show no large food particle,
and suggest microphagous particulate feeding. In contrast, all major Devo-
nian gnathostome groups in which the stomach contents are known show
evidence for predatory habits.
    Fossils provide some information about reproduction strategies and
sequences of skeletal development in early fishes, for example, fish eggs, egg
1.   PRIMITIVE FISHES THROUGH TIME                                           41

capsules, and growth series from juvenile to adults are known in heterostra-
cans, placoderms, acanthodians, many actinopterygians, and some piscine
sarcopterygians (coelacanths, lungfishes, and tristichopterid tetrapodo-
morphs). A Carboniferous coelacanth (Rhabdoderma) shows that the young
had a yolk sac, like Latimeria. Among early fishes, evidence for larval devel-
opment with metamorphosis is only known in the Devonian lungfish Dipterus
(Janvier, 1996a).
    Finally, by unraveling characters or character combinations that no
longer exist in the present day, fossils may challenge functional models
based exclusively on living taxa. An interesting example is provided by the
‘‘polybranchic’’ condition in extinct jawless vertebrate taxa, which display an
extremely large number of gills, such as some early Devonian galeaspids and
the late Devonian euphaneropids. While most studies on fish respiration are
based on the living hagfishes, lampreys, and gnathostomes, which have from
15 to 5 pairs of gill pouches or gill arches, the 33 gill pairs of euphaneropids
or the 45 gill pairs of certain galeaspids (Figure 1.16) may raise questions
relating to gill ventilation (Janvier, 2004a; Janvier et al., 2006). Notably, it
suggests that for jawless fishes with lamprey‐like gill pouches and passive
inspiration, the ventilatory problems posed by life in dysoxic environments
may have been solved by increasing the number of gills, whereas the active
inspiration of jawed fishes would not require this specialization (Mallatt,


    Most of the works that allude to the physiology of early fossil fish taxa
focus on the question of their freshwater or marine habitat, and sometimes
the possible evidence for anadromy (Denison, 1956; White, 1958; GriYth,
1987; Hardisty et al., 1989; Janvier, 1996a). The environment of the earliest
vertebrates remains vividly debated and rooted in certain geologic traditions.
The first Silurian and Devonian fish remains described in the beginning of the
nineteenth century were preserved in sandstones (e.g., the ‘‘Old Red Sand-
stone’’ of Britain and the Baltic States) and generally associated with plant
remains, but rarely with marine invertebrates. In addition, these heavily
armored fishes were regarded as ‘‘ganoids,’’ a group which classically includ-
ed living bichirs, gars, and catfishes, all reputedly freshwater. Progressively,
the received wisdom became that all these early fishes lived in freshwater and
occasionally passed into the sea, when found in marine sediments. In the mid‐
twentieth century, paleontologists began to raise the question of the evidence
for freshwater environment when dealing with sediments and fossils. Marine
42                                                                        PHILIPPE JANVIER

          A                                        B



                        Branchial basket

Fig. 1.16. Polybranchic vertebrates. (A and B) Endoskeletal head shield (braincase) of two early
Devonian galeaspids showing the roof of the oralobranchial chamber and the numerous bran-
chial fossae that housed the gills or gill pouches. (A) Duyunolepis, (B) Zhaotongaspis, and
(C) specimen (C1) and reconstruction (C2) of the late Devonian euphaneropid Euphanerops,
showing the very elongated branchial basket. Scale bar ¼ 10 mm. [Redrawn and modified from
Janvier (1996a by permission of Oxford University Press, and 2004a).]

sediments can be readily characterized by the presence of such invertebrates
as brachiopods or echinoderms, for which no freshwater representative is
known (Figure 1.17), but freshwater has no obvious paleontological ‘‘signa-
ture,’’ at any rate for such ancient periods as the Paleozoic. To make a long
story short, there are currently two competing trends in the interpretation of
these environments: some paleontologists consider the ‘‘Old Red Sandstone’’
as having been deposited in marginal marine environments, whereas others
consider that most of these sediments were deposited in either fluvial or
1.   PRIMITIVE FISHES THROUGH TIME                                                          43







Fig. 1.17. Reconstruction of a typical marine environment at the end of the middle Devonian
(385 Myr), in what is now northern France. The fish fauna is dominated by placoderms
(a, arthrodires; b, antiarchs; and c, ptyctodonts) and large sarcopterygians (d, lungfishes; and
e, onychodonts), but also includes small primitive actinopterygians (f, ‘‘paleonisciforms’’)
and acanthodians (g). Many of these fish taxa, notably arthrodires, antiarchs and acanthodians
also occur further north in the contemporary, but reputedly non‐arine, ‘‘Old Red Sandstone’’ of
Scotland. [Drawing by P. Janvier, in Lelievre et al. (1986).]

lacustrine environments (see review in Janvier, 1996a; Lelievre, 2001).
The arguments of the former are essentially based on the geologic context
(sedimentologic characteristics, geographical position relative to the closest
marine deposits), and those of the latter are that the same Silurian and
Devonian fishes, sometimes at the species level, may occur in both reputedly
freshwater and marine environments. Moreover, many of these early fish
taxa found in reputedly freshwater deposits generally display such a broad
global distribution that, considering current Devonian paleogeographic
reconstructions, they must have dispersed via marine environments.
    Geochemistry could theoretically settle the question. Notably, stable
isotope ratios of certain elements, such as 86Sr/87Sr, significantly diVer
between fresh and marine waters, but mainly at higher latitudes. However,
attempts at using this criterion on Paleozoic fishes have been scarce (Schmitz
et al., 1991), and the results must be taken with great reservation because of
the numerous biases induced by diagenesis or percolation, especially for such
ancient periods. In addition, the geochemical signals for marginal marine
environments (and organisms) situated at low latitudes (tropical waters) or
44                                                           PHILIPPE JANVIER

submitted to frequent variations of salinity, as was probably the environment
of the ‘‘Old Red Sandstone’’ deposits, are often unclear. It is probably
the reasons why the signal found for ‘‘Old Red Sandstone ostracoderms’’
(mainly heterostracans) by Schmitz et al. (1991) is ambiguous: freshwater for
some samples, marine for others, or undiagnostic. The question is thus not
yet settled, but one must keep in mind the fact that the Devonian was a
period of extensive peneplanization, during which the continental margins
were unusually flat and occupied by vast deltas and tidal flats. The environ-
ment of most Silurian and Devonian (and probably earlier) fishes was thus
comparable to the present‐day major tropical deltas and mangroves.
    This question of the marine versus freshwater environment is not restricted
to the early Paleozoic. Many late Paleozoic and Mesozoic fish sites also pose
the same problem, in connection with the presence of fish taxa whose extant
representatives are either essentially or exclusively marine. For example, the
Carboniferous and Permian xenacanthiform sharks of the intramontane
basins of Europe were almost certainly freshwater, although the same taxa
are also known elsewhere from undoubtedly marine sites. Similarly, the
earliest known Devonian actinistians were all marine, as is also Latimeria,
but several Carboniferous and Mesozoic actinistian taxa are regarded as
freshwater, essentially on the basis of the sedimentologic context. A single
stable isotope analysis has been performed on a coelacanth from a Creta-
ceous lacustrine deposit, and eVectively provided a clear freshwater signal,
for both the fish and the sediment (Poyato‐Ariza et al., 1998).
    Anadromy has been frequently invoked to explain the occurrence of
the same Paleozoic fish species or genera in marine and reputedly freshwater
localities (Halstead, 1973; Trewin and Davidson, 1999). Another argument,
albeit weak, in favor of anadromy is the homogeneity in size of the populations
of certain fossil fish species and the absence of populations of intermediate
sizes (Nielsen, 1949).
    Only two of the major living fish groups, hagfishes and chimaeriforms,
are exclusively marine. Hagfish physiology is supposed to preclude any
possibility for the group to have once been freshwater, but the presumed
freshwater environment of the Carboniferous hagfish Myxineides is ambigu-
ous and only based on a supposedly intramontane geologic context. Neither
the fossil chimaeriforms nor any stem chimaeriform, or fossil euchondroce-
phalan, has ever been found in a reputedly nonmarine geologic formation.
The earliest (Cambrian–Ordovician) vertebrates are all marine, and the
freshwater versus marine debate principally begins with the late Silurian
and Devonian fishes. However, indisputable freshwater fishes occur first in
the latest Devonian or the early Carboniferous. The early phase of vertebrate
and even gnathostome evolution thus took place almost certainly in marine
waters, though in coastal environments, but the conquest of freshwater
1.   PRIMITIVE FISHES THROUGH TIME                                                          45

environments occurred many times independently. Among the major extinct
taxa, it may have occurred in placoderms, but certainly in acanthodians. As
far as living taxa and their closest fossil relatives are concerned, it occurred
once in lungfishes and lampreys, respectively, possibly twice in actinistians, at
least four times in elasmobranchs, and many times in actinopterygians.


    The character combinations that are displayed by the fossil piscine verte-
brates rarely overturn phylogenies based on living taxa. However, they
provide key information for dating the earliest occurrence of taxa and
estimating ‘‘ghost ranges.’’ Therefore, the assessment of a living fish as
‘‘primitive’’ (or ‘‘ancient’’) does not only rest on its vague resemblance to
an early fossil form, as in the case of Latimeria, but also on the minimum age
of the taxon it belongs to, including when this is based on inferences from the
range of its sister group, as in the case of Polypterus. Moreover, regarding the
biology of the living primitive fishes, one must keep in mind that their stem‐
group relatives may have lived in environments that were radically diVerent,
as exemplified by lungfishes. The geologic context of the early fossil fishes can
indeed provide information about their environment, but in such cases as the
marine versus freshwater habits the data may be ambiguous, and any large‐
scale conclusions about evolutionary fish physiology should consider them
with great reservations. Attempts at inferring the physiology of ancient fishes
on the basis of their presumed descendants have generally been made by
paleontologists with little factual background. It may be timely for physiol-
ogists to consider shared physiological characters as an additional source of
data for phylogeny reconstructions and also to take a new look at certain
fossil data for hints into the physiology of the past.


Ahlberg, P. E., and Johanson, Z. (1998). Osteolepiforms and the ancestry of tetrapods. Nature
    395, 792–794.
Arnason, U., Gullberg, A., and Janke, A. (2001). Molecular phylogenetics and Gnathostomous
    (jawed) fishes: Old bones, new cartilage. Zool. Scr. 30, 249–255.
Arnason, U., Gullberg, A., Janke, A., Joss, J., and Elmrot, C. (2004). Mitogenomic analyses of
    deep gnathostome divergences: A fish is a fish. Gene 333, 61–70.
Arratia, G. (2004). Mesozoic halecostomes and the early radiation of teleosts. In ‘‘Mesozoic
    Fishes 3‐ Systematics, Palaeoenvironments and Biodiversity’’ (Arratia, G., and Tintori, A.,
    Eds.), pp. 279–316. Pfeil, Munich.
Bardack, D. (1991). First fossil hagfish (Myxinoidea): A record from the Pennsylvanian of
    Illinois. Science 254, 701–703.
46                                                                            PHILIPPE JANVIER

Bardack, D., and Richardson, E. S. (1977). New agnathous fishes from the pennsylvanian of
     Illinois. Fieldiana Geol. 33, 489–510.
Bardack, D., and Zangerl, R. (1968). First fossil lamprey: A record from the Pennsylvanian of
     Illinois. Science 1962, 1265–1267.
Bemis, W. E., Findeis, E. K., and Grande, L. (1997). An overview of Acipenseriformes. Environ.
     Biol. Fish 48, 25–71.
Blieck, A. (1992). At the origin of chordates. Geobios 25, 101–113.
Briggs, D. E. G., Clarkson, E. N. K., and Aldridge, R. J. (1983). The conodont animal. Lethaia
     20, 1–14.
Brinkmann, H., Venkatesh, B., Brenner, S., and Meyer, A. (2004). Nuclear protein‐coding genes
     support lungfish and not coelacanth as the closest relatives of land vertebrates. Proc. Natl.
     Acad. Sci. USA 101, 4900–4905.
Cloutier, R., and Ahlberg, P. E. (1996). Morphology, characters and the interrelationships of
     basal sarcopterygians. In ‘‘Interrelationships of Fishes’’ (Stiassny, M. L., Parenti, L. R., and
     Johnson, G. D., Eds.), pp. 445–480. Academic Press, San Diego.
Cloutier, R., and Arratia, G. (2004). Early diversification of actinopterygians. In ‘‘Recent Advances
     in the Origin and Early Radiation of Vertebrates’’ (Arratia, G., Wilson, M. V. H., and
     Cloutier, R., Eds.), pp. 217–270. Pfeil, Munich.
Coates, M. I. (1999). Endocranial preservation of a Carboniferous actinopterygian from Lanca-
     shire, UK, and the interrelationships of primitive actinopterygians. Philos. Trans. R. Soc.
     Lond. Ser.B 354, 435–462.
Conway Morris, S. (1998). ‘‘The Crucible of Creation.’’ Oxford University Press, Oxford.
Cope, E. D. (1889). Synopsis of the families of Vertebrata. Am. Nat. 23, 1–29.
Cunchillos, C., and Lecointre, G. (2005). Integrating the universal metabolism into a phyloge-
     netic analysis. Mol. Biol. Evol. 22, 1–11.
Daeschler, E., Snubin, N., and Jenkins, F., Jr. (2006). A Devonian tetrapod‐like fish and the
     evolution of the tetrapod body plan. Nature 440, 757–763.
Delarbre, C., Escriva, H., Gallut, C., Barriel, V., Kourilsky, P., Janvier, P., Laudet, V., and
     Gachelin, G. (2000). The complete nucleotide sequence of the Mitochondrial DNA of the
     Agnathan Lampetra fluviatilis: Bearings on the phylogeny of cyclostomes. Mol. Biol. Evol.
     17, 519–529.
Denison, R. H. (1956). A review of the habitat of the earliest vertebrates. Fieldiana Geol. 11,
Didier, D. A. (2004). Phylogeny and classification of extant holocephali. In ‘‘Biology of Sharks
     and Their Relatives’’ (Carrier, J. C., Musick, J. A., and Heithaus, M. R., Eds.), pp. 115–135.
     CRC Press, Boca Raton.
Donoghue, P. C. J., and Sansom, I. J. (2002). Origin and evolution of vertebrate skeletonization.
     Microsc. Res. Tech. 59, 352–372.
Donoghue, P. C. J., and Smith, M. P. (2001). The anatomy of Turinia pagei (Powrie) and the
     phylogenetic status of the Thelodonti. Trans. R. Soc. Edinb. Earth Sci. 92, 15–37.
Donoghue, P. C. J., Forey, P. L., and Aldridge, R. J. (2000). Conodont aYnity and chordate
     phylogneny. Biol. Rev. Camb. Philos. Soc. 75, 191–251.
Donoghue, P. C. J., Smith, M. P., and Sansom, I. J. (2003). The origin and early evolution of
     chordates: Molecular clocks and the fossil record. In ‘‘Telling the Evolutionary Time:
     Molecular Clocks and the Fossil Record’’ (Donoghue, P. C. J., and Smith, M. P., Eds.),
     pp. 190–223. Taylor and Francis, London.
Dutheil, D. B. (1999). The first articulated fossil cladistian: Serenoichthys kemkemensis, gen. et
     sp. nov., from the Cretaceous of Morocco. J. Vertebr. Paleontol. 19, 243–246.
Eldredge, N., and Stanley, S. M., (Eds.) (1984). ‘‘Living Fossils.’’ Springer, New York.
Filleul, A., and Lavoue, S. (2001). Basal teleosts and the question of elopomorph monophyly.
     C. R. Acad. Sci. III 324, 393–399.
1.   PRIMITIVE FISHES THROUGH TIME                                                             47

Forey, P. L. (1995). Agnathans recent and fossil, and the origin of jawed vertebrates. Rev. Fish
    Biol. Fisheries 5, 267–303.
Forey, P. L., and Janvier, P. (1993). Agnathans and the origin of jawed vertebrates. Nature 361,
Friedman, M., and Coates, M. I. (2005). A newly recognized fossil coelacanth highlights the
    early morphological diversification of the clade. Proc. R. Soc. Lond. Ser. B 273, 245–250.
Furlong, R. F., and Holland, P. W. H. (2002). Bayesian phylogenetic analysis supports mono-
    phyly of Ambulacria and of cyclostomes. Zool. Sci. 19, 593–599.
Gabbott, S., Aldridge, R. J., and Theron, J. (1995). A giant conodont with preserved muscle
    tissue from the Upper Ordovician of South Africa. Nature 374, 800–803.
                                                      ´ ´
Gagnier, P.‐Y. (1991). Sacabambaspis janvieri, Vertebre ordovicien de Bolivie. 2: Analyse phy-
    logenetique. Annales de Paleontologie 79, 119–166.
        ´ ´                      ´
Gardiner, B. G. (1984). The relationships of the palaeoniscid fishes, a review based on new
    specimens of Mimia and Moythomasia, from the Upper Devonian of Western Australia.
    Bull. Br. Mus. (Nat. Hist.) 37, 173–428.
Gardiner, B. G., and SchaeVer, B. (1989). Interrelationships of lower actinopterygian fishes.
    Zool. J. Linnean Soc. Lond. 97, 135–187.
Gardiner, B. G., Maisey, J. G., and Littlewood, D. T. J. (1996). Interrelationships of basal
    neopterygian fishes. In ‘‘Interrelationships of Fishes’’ (Stiassny, M., Parenti, L., and
    Johnson, G. D., Eds.), pp. 117–146. Academic Press, San Diego.
Gess, R. W., Coates, M. I., and Rubidge, B. S. (2006). A lamprey from the Devonian of South
    Africa. Nature 443, 981–984.
Grande, L., and Bemis, W. (1999). Historical biogeography and historical paleoecology of
    amiidae and other halecomorph fishes. In ‘‘Mesozoic Fishes 2 ‐Systematics and the Fossil
    Record’’ (Arratia, G., and Schultze, H. P., Eds.), pp. 413–424. Pfeil, Munich.
GriYth, R. W. (1987). Freshwater or marine origin of the vertebrates? Comp. Biochem. Physiol. A
    87, 523–531.
Grogan, E. D., and Lund, R. (2004). The origin and relationships of early chondrichthyes.
    In ‘‘The Biology of Sharks and Their Relatives’’ (Carrier, J. C., Musick, J. A., and
    Heithaus, M. R., Eds.), pp. 3–31. CRC Press, Boca Raton.
Gursoy, H. C., Koper, D., and Benecke, B. J. (2000). The vertebrate 7S K RNA separates hagfish
    (Myxine glutinosa) and lamprey (Lampetra fluviatilis). J. Mol. Evol. 50, 456–464.
Halstead, L. B. (1973). The heterostracan fishes. Biol. Rev. 48, 279–332.
Hardisty, M. W. (1982). Lampreys and hagfishes: Analysis of cyclostome relationships. In ‘‘The
    Biology of Lampreys’’ (Hardisty, M. W., and Potter, I. C., Eds.), Vol. 4B, pp. 165–259.
    Academic Press, London.
Hardisty, M. W., Potter, I. C., and Hilliard, R. W. (1989). Physiological adaptations of the living
    agnathans. Trans. R. Soc. Edinb. Earth Sci. 80, 241–254.
Hedges, S. B. (2001). Molecular evidence for the early history of living vertebrates. In ‘‘Major
    Events in Early Vertebrate Evolution’’ (Ahlberg, P. E., Ed.), pp. 118–134. Taylor and
    Francis, London.
Hennig, W. (1950). ‘‘Grundzuge Einer Theorie Der Phylogenetischen Systematik.’’ Deutscher
    Zentralverlag, Berlin.
Huxley, T. H. (1880). On the application of the laws of evolution to the arrangement of the
    vertebrata, and more particularly of the mammalia. Proc. Zool. Soc. Lond. 1880, 640–662.
Inoue, J. G., Miya, M., Tsukamoto, K., and Nishida, Y. (2003). Basal actinopterygian relation-
    ships: A mitogenomic perspective on the phylogeny of the ‘‘ancient fish.’’ Mol. Biol. Evol. 26,
Janvier, P. (1984). Groupes panchroniques, ‘‘fossiles vivants’’ et systematique: l’exemple des
    ‘‘Crossopterygiens’’ et des Petromyzontida. Bull. Soc. Zool. Fr. 109, 609–616.
Janvier, P. (1996a). ‘‘Early Vertebrates.’’ Oxford University Press, Oxford.
48                                                                          PHILIPPE JANVIER

Janvier, P. (1996b). The dawn of the vertebrates: Characters versus common ascent in phyloge-
     netic theories. Palaeontology 39, 259–287.
Janvier, P. (1997). Les vertebres avant le silurien. Geobios 30, 931–951.
                             ´ ´
Janvier, P. (1998). A cold look at odd vertebrate phylogenies. J. Mol. Evol. 46, 375–377.
Janvier, P. (2001). Ostracoderms and the shaping of the gnathostome characters. In ‘‘Major
     Events in Early Vertebrate Evolution’’ (Ahlberg, P. E., Ed.), pp. 172–186. Taylor and
     Francis, London.
Janvier, P. (2003). Vertebrate characters and Cambrian vertebrates. Comptes Rendus Palevol 2,
Janvier, P. (2004a). Early specializations of the branchial apparatus in jawless vertebrates:
     A consideration of gill number and size. In ‘‘Recent Advances in the Origin and Early
     Radiation of Vertebrates’’ (Arratia, G., Wilson, M. V. H., and Cloutier, R., Eds.),
     pp. 29–52. Pfeil, Munich.
Janvier, P. (2004b). Wandering nostrils. Nature 432, 23–24.
Janvier, P. (in press). Homologies and evolutionary transitions in early vertebrate history.
     In ‘‘Major Transitions in Vertebrate Evolution’’ (Anderson, J. S., and Sues, H.‐D., Eds.).
     Indiana University Press, Bloomington.
Janvier, P., and Lund, R. (1983). Hardistiella montanensis n.gen., n.sp. (Petromyzontida) from
     the Lower Carboniferous of Montana, with remarks on the aYnities of lampreys. J. Vertebr.
     Paleontol 2, 407–413.
Janvier, P., and Maisey, J. G. (in press). The Devonian vertebrates of South America and their
     biogeographical relationships. In ‘‘Morphology, Phylogeny and Biogeography of Fossil
     Fishes’’ (Yu, X., Maisey, J. G., and Miao, D., Eds.). Pfeil, Munich.
Janvier, P., and Racheboeuf, P. (2003). A ptyctodontid fish (Placodermi) from the Kersadiou
     Formation (Givetian) of western Brittany, France. Neues Jahrb. Geol. Palaeontol. Monatsh.
     10, 630–640.
Janvier, P., Racheboeuf, P., Nguyen Huu, H., and Doan Nhat, T. (2003). Devonian fish
     (Placodermi, Antiarcha) from the Tra Ban Island (Bai Tu Long Bay, Quang Ninh, Vietnam)
     and the question of the age of the Do Son Formation. J. Asian Earth Sci. 21, 795–801.
Janvier, P., Desbiens, S., Willett, J. A., and Arsenault, M. (2006). Lamprey‐like gills in a
     gnathostome‐related Devonian jawless vertebrate. Nature 440, 1183–1185.
Jarvik, E. (1980).‘‘Basic Structure and Evolution of Vertebrates,’’ Vol. 1. Academic Press, London.
Jarvik, E. (1981).‘‘Basic Structure and Evolution of Vertebrates,’’ Vol. 2. Academic Press,
Johanson, Z., Long, J. A., Talent, J. A., Janvier, P., and Warren, J. A. (2006). Oldest coelacanth
     from the Early Devonian of Australia. Biol. Lett. 2, 443–446. doi: 10.1098/rsbl.2006.0470.
Jørgensen, J. M., Lomholt, J. P., Weber, R. E., and Malte, H. (1998). ‘‘The Biology of Hag-
     fishes.’’ Chapman and Hall, London.
Kikugawa, K., Katoh, K., Kuraku, S., Sakurai, H., Ishida, O., Iwabe, N., and Miyata, T. (2004).
     Basal vertebrate phylogeny inferred from multiple nuclear DNA‐coded genes. BMC Biol. 2,
Laurin, M., and Anderson, J. S. (2004). Meaning of the name Tetrapoda in the scientific
     literature: An exchange. Syst. Biol. 53, 68–80.
Le, H. L. V., Lecointre, G., and Perasso, R. (1993). A 28S rRNA‐based phylogeny of the
     gnathostomes: First steps in the analysis of conflict and congruence with morphologically
     based cladograms. Mol. Phylogenet. Evol. 2, 31–51.
    `                          ´
Lelievre, H. (2001). ‘‘Phylogenie des brachythoraci (Vertebrata, Placodermi) et ajustement de la
             ´    `                                   ´                           ´
     phylogenie a la stratigraphie. Les sites du Devonien terminal, la caracterisation de leur
1.   PRIMITIVE FISHES THROUGH TIME                                                              49

                 ´ ˆ
     milieu de depot par analyse de similitude de leur ichthyofaune.’’ Unpublished Thesis, H 363.
     Universite des Sciences et Techniques de Lille, Villeneuve d’Ascq.
    `                                                                       ´
Lelievre, H., Goujet, D., Blieck, A., and Janvier, P. (1986). Poissons devoniens du Boulonnais
     France. In ‘‘Le Devonien de Ferques, Bas‐Boulonnais (N. France)’’ (Brice, D., Ed.), Vol. 7,
                                             ´    ¨
     pp. 504–519. Biostratigraphie du Paleozoıque, Brest France.
Li, G.‐Q., and Wilson, M. V. H. (1996). Phylogeny of Osteoglossomorpha. In ‘‘Interrelationships
     of Fishes’’ (Stiassny, M. L. J., Parenti, L. R., and Johnson, G. D., Eds.), pp. 163–174.
     Academic Press, San Diego.
Long, J. A. (1988). New palaeoniscid fishes from the late Devonian and Early Caboniferous of
     Victoria. Mem. Assoc. Aust. Palaeontol. 7, 1–63.
Løvtrup, S. (1977). ‘‘The Phylogeny of the Vertebrata.’’ John Wiley, London.
Lund, R. (2000). A new actinopterygian order guildayichthyiformes from the lower carbonifer-
     ous of Montana (USA). Geodiversitas 22, 171–206.
Maisey, J. G. (1984). Higher elasmobranch phylogeny and biostratigraphy. Zool. J. Linnean Soc.
     Lond. 82, 33–54.
Maisey, J. G. (1986). Heads and tails: A chordate phylogeny. Cladistics 2, 201–256.
Maisey, J. G. (1994). Predator‐prey relationships and trophic level reconstruction in a fossil fish
     community. Environ. Biol. Fish 40, 1–22.
Maisey, J. G. (2001). A primitive chondrichthyan braincase from the Middle Devonian of
     Bolivia. In ‘‘Major Events in Early Vertebrate Evolution’’ (Ahlberg, P. E., Ed.),
     pp. 263–288. Taylor and Francis, London.
Maisey, J. G., Naylor, G. J. P., and Ward, D. J. (2004). Mesozoic elasmobranchs, neoselachian
     phylogeny and the rise of modern neoselachian diversity. In ‘‘Mesozoic Fishes 3 – System-
     atics, Paleoenvironments and Biodiversity’’ (Arratia, G., and Tintori, A., Eds.), pp. 17–56.
     Pfeil, Munich.
Mallatt, J. (1996). Ventilation and the origin of jawed vertebrates: A new mouth. Zool.
     J. Linnean Soc. Lond. 117, 329–404.
Mallatt, J., and Chen, J. (2003). Fossil sister group of craniates: Predicted and found. J. Morphol.
     258, 1–31.
Mallatt, J., Sullivan, J., and Winchell, C. J. (2001). The relationships of lampreys to hagfishes:
     A spectral analysis of ribosomal DNA sequences. In ‘‘Major Events in Early Vertebrate
     Evolution’’ (Ahlberg, P. E., Ed.), pp. 106–118. Taylor and Francis, London.
Mazan, S., Jaillard, D., Baratte, B., and Janvier, P. (2000). Otx1 gene‐controlled morphogenesis
     of the horizontal semicircular canal and the origin of the gnathostome characteristics. Evol.
     Dev. 2, 186–193.
McAllister, J. A. (1992). Gnathorhiza (Dipnoi): Life aspects, and lungfish burrows. In ‘‘Fossil
     Fishes as Living Animals’’ (Mark‐Kurik, E., Ed.), pp. 91–105. Akademia 1, Tallinn.
Miller, R. F., Cloutier, R., and Turner, S. (2003). The oldest articulated chondrichthyan from the
     Early Devonian period. Nature 425, 501–504.
Morrissey, L. B., Janvier, P., Braddy, S. J., Bennett, J. P., Marriott, S. B., and Tarrant, P. R.
     (2006). Swimming with devonian fish. Geology Today 22, 66–67.
Needham, J. (1938). Contributions of chemical physiology to the problem of the reversibility in
     evolution. Biol. Rev. 13, 225–251.
Nelson, G. J. (1970). Outline of a theory of comparative biology. Syst. Zool. 19, 373–384.
Nelson, G. J. (1994). Homology and systematics. In ‘‘Homology: The Hierarchical Basis of
     Comparative Biology’’ (Hall, B. K., Ed.), pp. 101–149. Academic Press, San Diego.
Nielsen, E. (1949). Studies on Triassic fishes from East Greenland. 2. Australosomus and
     Birgeria. Meddr. Grnl. 146, 1–309.
50                                                                              PHILIPPE JANVIER

Noack, K., Zardoya, R., and Meyer, A. (1996). The complete mitochondrial DNA sequence of
    the bichir (Polypterus ornatipinnis), a basal ray‐finned fish: Ancient establishment of the
    consesus vertebrate gene order. Genetics 144, 1165–1180.
Olsen, P. E., and McCune, A. R. (1991). Morphology of the Semionotus elegans species group
    from the Early Jurassic part of the Newark Supergroup of Eastern North America with
    comments on the family semionotidae (Neopterygii). J. Vertebr. Paleontol. 11, 269–292.
Patterson, C. (1973). Interrelationships of holosteans. In ‘‘Interrelationships of Fishes’’ (Greenwood,
    P. H., Miles, R. S., and Patterson, C., Eds.). Zool. J. Linnean Soc. Lond. 53(Suppl. 1), 233–305.
Patterson, C. (1977). The contribution of paleontology to teleostean phylogeny. In ‘‘Major
    Patterns in Vertebrate Evolution’’ (Hecht, P. C., Goody, P. C., and Hecht, B. M., Eds.),
    pp. 579–643. Plenum, New York.
Patterson, C. (1982). Morphology and interrelationships of primitive actinopterygian fishes.
    Am. Zool. 22, 241–259.
Patterson, C., and Rosen, D. E. (1977). Review of ichthyodectiform and other Mesozoic teleost
    fishes, and the theory and practice of classifying fossils. Bull. Am. Mus. Nat. Hist. 159,
Poplin, C., Sotty, D., and Janvier, P. (2001). Un Myxinoıde (Craniata, Hyperotreti) dans le
                         ¨              `      ´
    Konservat‐Lagerstatte Carbonifere superieur de Montceau‐les‐Mines (Allier, France). C. R,
    Acad. Sci. Terre Planete, 332, 345–350.
Poyato‐Ariza, F. J., Talbot, M. R., Fregenal‐Martinez, M. A., Melendez, N., and Wenz, S.
    (1998). First isotopic and multidisciplinary evidence for nonmarine coelacanths and pycno-
    dontiform fishes: Palaeoenvironmental implications. Palaeogeogr. Palaeoclimatol. Palaeoe-
    col. 144, 65–84.
Rasmussen, A.‐S, and Arnason, U. (1999a). Phylogenetic studies of complete mitochondrial
    DNA molecules place cartilaginous fishes within the tree of bony fishes? J. Mol. Evol. 48,
Rasmussen, A.‐S, and Arnason, U. (1999b). Molecular studies suggest that cartilaginous fishes
    have a terminal position in the piscine tree. Proc. Natl. Acad. Sci. USA 96, 2177–2182.
Rosen, D. E., Forey, P. L., Gardiner, B. G., and Patterson, C. (1981). Lungfishes, tetrapods,
    paleontology, and plesiomorphy. Bull. Am. Mus. Nat. Hist. 167, 157–276.
Schmitz, B., Aberg, G., Werdelin, L., Forey, P. L., and Bendix‐Almgren, S. E. (1991). 87Sr/86Sr,
    Na, F, Sr and La in skeletal fish debris as a measure of the paleosalinity of fossil‐fish
    habitats. Geol. Soc. Am. Bull. 103, 786–794.
Schultze, H.‐P. (1996). Conodont histology: An indicator of vertebrate relationship? Mod. Geol.
    20, 275–285.
Schultze, H.‐P., and Cumbaa, S. L. (2001). Dialipina and the characters of basal actinopteryp-
    terygians. In ‘‘Major Events in Early Vertebrate Evolution’’ (Ahlberg, P. E., Ed.),
    pp. 315–332. Taylor and Francis, London.
Shirai, S. (1996). Phylogenetic interrelationships of neoselachians (Chondrichthyes: Euselachii).
    In ‘‘Interrelationships of Fishes’’ (Stiassny, M. L. J., Parenti, L. R., and Johnson, G. D.,
    Eds.), pp. 9–34. Academic Press, London.
Shu, D., Luo, H.‐L., Conway Morris, S., Zhang, X.‐L., Chen, S.‐X., Han, J., Zhu, M., Li, Y.,
    and Chen, L.‐Z. (1999). Lower Cambrian vertebrates from South China. Nature 402, 42–46.
Shu, D., Chen, L., Han, J., and Zhang, X.‐L. (2001). An early Cambrian tunicate from China.
    Nature 411, 472–473.
Shu, D., Conway Morris, S., Han, J., Zhang, Z.‐F., Yasui, K., Janvier, P., Chen, L.,
    Zhang, X.‐L., Liu, J.‐N., Li, Y., and Liu, H. K. (2003). Head and backbone of the early
    Cambrian vertebrate Haikouichthys. Nature 421, 526–529.
Stahl, B. (1999). Chondrichthyes III, Holocephali. In ‘‘Handbook of Palaeoichthyology’’
    (Schultze, H.‐P., Ed.). Pfeil, Munich. 1–164.
1.   PRIMITIVE FISHES THROUGH TIME                                                           51

Stensio, E. A. (1927). The Devonian and Dowtonian vertebrates of Spitsbergen. I. Family
    Cephalaspidae. Skrifter om Svalbard og Ishavet 12, 1–391.
Stock, D. W., and Whitt, G. S. (1992). Evidence from 18S ribosomal RNA sequences that
    lampreys and hagfishes form a natural group. Science 257, 787–789.
Takezaki, N., Figueroa, F., Zaleska‐Rutccynska, Z., Takahata, N., and Klein, J. (2004). The
    phylogenetic relationship of tetrapod, coelacanth, and lungfish revealed by the sequences of
    44 nuclear genes. Mol. Biol. Evol. 21, 1512–1524.
Trewin, N. H., and Davidson, R. G. (1999). Lake level changes, sedimentation and faunas in a
    Middle Devoniuan basin‐margin fish bed. J. Geol. Soc. Lond. 156, 535–548.
White, E. I. (1958). Original environment of the craniates. In ‘‘Studies on Fossil Vertebrates’’
    (Westoll, T. S., Ed.), pp. 213–233. Athlone Press, London.
Yalden, D. W. (1985). Feeding mechanisms as evidence for cyclostome monophyly. Zool. J.
    Linnean Soc. Lond. 84, 291–300.
Zardoya, R., and Meyer, A. (1997). Molecular phylogenetic information on the identity of the
    closest living relative(s) of land vertebrates. Naturwissenschaften 84, 389–397.
Zardoya, R., and Meyer, A. (2001). Vertebrate phylogeny: Limits of inference of mitochondrial
    genome and nuclear DNA sequence data due to an adverse phylogenetic signal/noise ratio.
    In ‘‘Major Events in Early Vertebrate Evolution’’ (Ahlberg, P. E., Ed.), pp. 135–155. Taylor
    and Francis, London.
Zardoya, R., Cao, Y., Hasegawa, M., and Meyer, A. (1998). Searching for the closest living
    relative(s) to tetrapods through evolutionary analyses of mitochondrial DNA and nuclear
    data. Mol. Biol. Evol. 15, 506–517.
Zhang, Y.‐G., and Hou, X.‐G. (2004). Evidence for a single median fin‐fold in tail in the Lower
    cambrian vertebrate, Haikouichthys ercaicunensis. J. Evol. Biol. 17, 1162–1166.
Zhu, M., and Ahlberg, P. E. (2004). The origin of the internal nostril of tetrapods. Nature 432,
Zhu, M., and Yu, X. (2002). A primitive fish close to the common ancestor of tetrapods and
    lungfish. Nature 418, 767–770.
Zhu, M., Yu, X., and Janvier, P. (1999). A primitive fossil fish sheds light on the origin of bony
    fishes. Nature 397, 607–610.
Zhu, M., Yu, X., and Ahlberg, P. E. (2001). A primitive sarcopterygian fish with an eyestalk.
    Nature 410, 81–84.
Zhu, M., Yu, X., Wang, W., Zhao, W., and Jia, L. (2006). A primitive fish provides key
    characters bearing on deep osteichthyan phylogeny. Nature 44, 77–80.
Zrvav’y, J., Lilhulka, S., Kepka, P., Bezdek, A., and Tietz, D. (1998). Phylogeny of the Metazoa
    based on morphological and 18S ribosomal DNA evidence. Cladistics 14, 249–285.
This page intentionally left blank


1. Introduction
   1.1. Scope of the Chapter
   1.2. Measurement Systems: Their Benefits and Limitations
2. An Overview of Evolutionary Progressions
   2.1. Anatomical Patterns
   2.2. Physiological Patterns
3. Details of the Cyclostome Circulatory Systems
   3.1. Hagfishes
   3.2. Lampreys
4. Details of the Sarcopterygii (Lobe‐Finned Fishes) Circulatory Systems
   4.1. Coelacanth
   4.2. Dipnoi (Lungfishes)
5. Details of the Circulatory Systems in Polypterids, Gars, and Bowfins
   5.1. Polypterids (Bichirs and Reedfish)
   5.2. Garfishes
   5.3. Amia (Bowfins)
6. Details of the Sturgeon Circulatory Systems
   6.1. Cardiac Anatomy
   6.2. Circulatory Patterns
   6.3. Cardiac Dynamics
   6.4. Circulatory Control
7. Conclusions

    The cardiovascular system is crucial by virtue of its role in transporting
nutrients, respiratory gases, hormones, and waste products. This chapter
focuses on circulatory form and function: the anatomy of the cardiovascular
system, cardiac dynamics, and cardiovascular control. Studying circulatory
control in any fish is particularly diYcult because discrete circulations of
specific organs are not easily accessible. Therefore, by necessity, most infor-
mation on cardiovascular control in primitive fishes is limited largely to the
control of cardiac output (Q), as well as control of blood flow through
the gills, to air‐breathing organs, and the gastrointestinal tract. Unusual
Primitive Fishes: Volume 26                            Copyright # 2007 Elsevier Inc. All rights reserved
FISH PHYSIOLOGY                                                   DOI: 10.1016/S1546-5098(07)26002-9
54                                                         ANTHONY P. FARRELL

adaptations of primitive fishes that deviate from those piscine features com-
mon to elasmobranchs and teleosts are highlighted. The chapter starts with
the most primitive fishes, the cyclostomes, and moves through the cardiovas-
cular anatomy of the coelacanth to the cardiovascular anatomy and physiol-
ogy of dipnoans, the forerunners to tetrapods. It then closes by covering the
limited physiological information for Polypterids, gars, bowfins, and stur-
geons. By comparing cardiovascular adaptations among these primitive
fishes, this chapter examines the evolutionary roots and the evolutionary
divergence of the piscine cardiovascular system.


1.1. Scope of the Chapter

    This chapter assumes that the reader has a general knowledge of the
circulatory form and function in elasmobranchs and teleosts, as described
elsewhere (Randall, 1968; Olson and Farrell, 2006). Additional details can
be found in books and reviews, such as Randall (1970), Satchell (1971, 1991),
Johansen (1971), Farrell (1984, 1991), Santer (1985), and Butler and Metcalfe
(1988), as well as two previous volumes of Fish Physiology (12A and 12B;
1992) that were dedicated to the cardiovascular system. At a finer scale, the
cellular structure of fish hearts has been reviewed by Yamauchi (1980). The
physiological processes that govern transsarcolemmal and intracellular ion
movements associated with cardiac excitation and contraction are being
slowly unraveled for teleosts (see reviews by Tibbits et al., 1992; Shiels et al.,
2002; Vornanen et al., 2002), but not yet for primitive fishes.
    Goodrich (1930) is an excellent starting place for readers interested in
a description of the fish circulatory system in a broader evolutionary
context, with detailed physiological perspectives provided by Johansen
(1965), Johansen and Hanson (1968), Johansen and Burggren (1980), and
Burggren et al. (1997). The neural and humoral controls of the circulatory
system have been reviewed from several evolutionary perspectives (Laurent
et al., 1983; Nilsson, 1983; Nilsson and Axelsson, 1987). Linkages can be
made between the present chapter and others in this volume, for example,
Chapter 3 (nervous controls) and Chapter 5 (respiratory functions), and with
the description of cardiorespiratory control in tropical fishes found in an
early volume of this series (see Reid et al., 2006 in volume 21).
    An impressive fact is that many of the older observations of circulatory
anatomy, some approaching 200 years old, were remarkably accurate in many
respects and thus continue to provide a foundation for our understanding of
the piscine circulatory system. Newer techniques, such as vascular casting
with synthetic resins and electron microscopy, have added important details
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                55

of the microcirculation, as well as correcting older literature where necessary.
This chapter, rather than fully citing the older literature, favors works that are
generously referenced and can take the reader back to the roots of important
anatomical and physiological discovery in the nineteenth and early twentieth
centuries. Among these works are Biology of Myxine (Brodal and Fange,       ¨
1963), Biology of the Lampreys (Hardisty and Potter, 1972), and Biology of the
Cyclostomes (Hardisty, 1979). Satchell’s chapter in his book (Satchell, 1991),
‘‘Myxine, a Speculative Conclusion,’’ is fascinating reading. The anatomy of
the coelacanth circulatory system is meticulously detailed in the monograph
by Millot et al. (1978). Satchell (1976) has provided an extremely insightful
chapter on the circulatory system of air‐breathing fish, which has been gener-
ously extended by Graham’s (1997) masterful monograph on air‐breathing
fishes. Other relevant works appear at the outset of each section.
    Compared with anatomical studies, physiological studies on primitive
fishes are spartan in number. Furthermore, techniques for studying circula-
tory function are continually emerging and improving. Therefore, older
physiological literature may need verification. Concerns include measure-
ment precision, data replication, and animal welfare (the later because of
the potential impact of surgical stress and animal restraint on routine physi-
ological variables). Two of these concerns are considered briefly so that the
reader can make a more informed decision about the quality of the detailed
information presented in this chapter.

1.2. Measurement Systems: Their Benefits and Limitations
    Early in vivo estimates of blood flow and Q were obtained by either
direct measurement with electromagnetic and Doppler flow probes, or
indirect measurement using the Fick Principle. Electromagnetic probes
report blood flow but require a reliable means to regularly check zero flow.
Doppler flow probes have the advantage of being easy to zero in vivo. If either
electromagnetic or Doppler flow probes do not fit snuggly to the vessel wall,
they become excessively noisy, and if the orientation on the vessel is incor-
rect, they can underestimate flow. Attention is needed to ensure that the
Doppler signal is focused on the center of the vessel. Doppler probes are
more diYcult to calibrate and this must be done in situ (with relevant
blood flows and blood pressures) after the experiment. Doppler probes
measure blood velocity, and a loose fit on a highly elastic cardiac outflow
vessel in a fish could underestimate flow if the vessel expands when blood
pressure increases. A full strength signal from a Doppler flow probe
requires a hematocrit of the order of 10%, which should not be a problem
for normocythemic primitive fishes, but this could be an issue during
56                                                        ANTHONY P. FARRELL

    Today’s preferred blood flow measurement technique utilizes a transit
time flow probe, which is self‐zeroing, precalibrated, and has a lower toler-
ance for a snug fit on the vessel. However, transit time flow probes (e.g.,
Transonic Instruments) are expensive and their temperature sensitivity
requires precalibration to the experimental temperature. In addition, probes
for small diameter vessels have a limited lead length, which can be a consid-
erable challenge with aquatic animals. Modern Doppler flow probes with
their small probe head and fine lead remain the technique of choice for small
fish, especially if vessel access is limited.
    Indirect measurement of flow using the Fick Principle has the advantage
of little to no surgical intervention. Nevertheless, it has two shortcomings:
(1) it is possible to overestimate Q in fish because oxygen consumed directly
by gills and skin is not properly taken into account (Randall et al., 1981) and
(2) the possibility of making continuous measurements of gas tensions is
limited, which is a problem when fish perform arrhythmic and dynamic
behaviors such as air breathing, blood shunting, exercise, and responding
to gas tension changes. Hopefully, this latter problem can be resolved using
continuous measurements of oxygen tension with fiber optic technologies.
    Placement of a flow probe around a blood vessel requires surgical inter-
vention. Early studies rarely described how invasive the surgery was and
this is a special concern given that early electromagnetic and Doppler flow
probes were bulky. Surgical techniques and the size of flow probes have
improved considerably over the past 40 years. For example, Q used to be
measured by accessing the ventral aorta or bulbus arteriosus via ventral
dissection through the pectoral musculature. Beyond tissue and vascular
trauma, the pericardium was often cut which alters cardiac pumping (reduces
maximum performance—Farrell et al., 1988a; and changes venous blood
pressures—Johansen, 1965; Shabetai et al., 1985; Sandblom et al., 2006). In
rainbow trout, access to the ventral aorta is now gained through a minor
incision in the lateral wall of the isthmus, anterior of the pericardial cavity,
and the union of the coronary artery (Axelsson and Farrell, 1993). This
approach avoids excessive tissue trauma, opening of the pericardium, and
possible occlusion of the coronary artery from a tightly fitting flow probe on
the bulbus arteriosus.
    Routine cardiovascular performance is best assessed after full recovery
from anesthesia and surgery. Therefore, measurements made either during
anesthesia or shortly after recovery must be treated with caution. Overnight
recovery is a generally accepted compromise between recovery and the risk
of damage to expensive equipment, even though rainbow trout swim well
after just a 2‐h recovery from invasive surgery (Farrell and Clutterham,
2003). Nevertheless, since heart rate typically decreases with recovery time,
a protracted recovery may be essential to properly assess certain aspects
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                              57

of cardiovascular control. However, even fully recovered fish must be
restrained to reduce the risk to expensive recording equipment. Hagfishes,
for example, have a remarkable ability to tie multiple knots in any leads
attached to them. Therefore, future work should attempt to allow fish to
behave more normally. This goes beyond obvious issues of correct lighting,
appropriate temperatures, and diurnal rhythms. The use of remote telemetry
and video observation will reduce the influence of an investigator’s presence
during cardiovascular recordings and allow the animals to behave more
normally and with less physical restriction compared to animals physically
connected to a recording system.
    Even simple measurements such as cardiac (cardiosomatic) index need
careful inspection for what cardiac chambers (sinus venosus, atrium, ventri-
cle, and conus/bulbus arteriosus) are included in the measurement. While the
ventricle is usually the largest of the cardiac chambers contained within
the pericardium and atrial mass is about 25% of ventricular mass, this is
clearly not the case for either Latimeria or lungfishes (see below). Care is also
needed to remove excess blood from the extensive trabecular spaces of the
fish heart before weighing (a task that is virtually impossible if blood has
coagulated) and to prevent changes in tissue water content through either
dehydration in small hearts or fixation techniques. Physiological studies tend
to report only relative ventricular mass as part of the calculation of myocar-
dial power output. However, sexual maturation and temperature acclimation
influence ventricular mass (Farrell et al., 1988b; Thorarensen et al., 1996).
    Measurement of blood volume involves marker techniques. Red blood
cell markers apparently provide a better estimate of blood volume than
plasma markers, which overestimate blood volume to varying degrees and
in relation to the equilibration time (Bushnell et al., 1992). In fact, careful
consideration of the secondary circulations and blood sinuses is needed to
interpret the blood volume estimates because red blood cells do not necessarily
move quickly or freely into these fluid compartments.
    Despite these overarching concerns, the following descriptions take data
at face value unless a clear problem is noted. In this regard, it is encouraging
to reevaluate Greene’s (1926) arterial blood pressures for chinook salmon
and to see that they span the range we now know to be between routine and
maximum blood pressure.


   The purpose of this section is to provide the reader with a general
overview of the various patterns of change evident among primitive fishes.
This section could have appeared at the end of the chapter as a synthesis
58                                                         ANTHONY P. FARRELL

(i.e., after the detailed information that follows), but I felt it was more useful
at the outset to provide a broad framework. For this reason, few citations are
found in this section because they appear subsequently.

2.1. Anatomical Patterns

2.1.1. The Heart
    The number of cardiac chambers contained in the pericardial cavity is
three in cyclostomes and four in all other fishes. All fish have a sinus venosus,
an atrium, and a ventricle. Cyclostomes lack a ventricular outflow tract
within the pericardium.
    There has been extensive debate on the origins of and the terminology for
the ventricular outflow tract (the conus arteriosus, bulbus arteriosus, and
truncus cordis: see Smith, 1918; Wright, 1984; Farrell and Jones, 1992;
Icardo et al., 2005b). Long ago, Wilder (1876) distinguished a rhythmically
contractile ‘‘bulbus arteriosus’’ with several rows of valves in elasmobranchs
and ganoids (i.e., Polypterids, gars, and bowfins) from a noncontracting
bulbus arteriosus in teleosts. Here, I use the term bulbus arteriosus if cardiac
muscle is lacking and conus arteriosus if cardiac muscle is present.
    All primitive fishes except cyclostomes possess a conus arteriosus proxi-
mal to the ventricle, just like elasmobranchs. However, unlike elasmo-
branchs, the conus in most primitive fish is reduced in size and lacks true
valves, and a bulbus arteriosus lies distal to the conus. Furthermore, not all
primitive fishes have conal muscle that is rhythmically contractile, at least to
the point of altering ventral aortic pressure recordings.
    The atrium and ventricle of primitive fishes do not diVer from the general
piscine arrangement of trabecular muscle (spongy myocardium) with deep
lacunae that allow venous blood to almost reach the epicardial surface of the
heart. The spongy myocardium derives its nutrition from, and excretes its
wastes into, these sinusoids (lacunae, cardiac circulation). The ventricle can
have an additional muscle type (compact myocardium) with an arterial
(coronary) circulation that surrounds the spongy myocardium. Thus, a
variable portion of the ventricle can have a secondary, fully oxygenated
blood supply directly from the gills via the hypobranchial artery or dorsal
    The most primitive form for the ventricle is an entirely trabecular arrange-
ment found in cyclostomes. Even so, the coronary circulation appeared early
in the evolution of the vertebrate heart, being present in elasmobranchs and
apparently all other primitive fishes. The need for a more secure, arterial
oxygen supply to the heart may reflect both the cardiac evolution toward a
higher workload capability, as well as exposure to environmental conditions
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                  59

that rendered a venous oxygen supply unreliable (e.g., aquatic hypoxia and
exercise). The coronary circulation has appeared, been lost, and reappeared
during the evolution of fishes. It has cephalad (hypobranchial artery) and
caudal (subclavian and coracoid arteries) origins (Parker and Davis, 1899;
Grant and Regnier, 1926; Foxon, 1950; Davie and Farrell, 1991). Rays
possess both caudal and cephalad coronaries, and while Latimeria and some
Chondrosteans have a caudal supply, other primitive fishes and sharks have a
cephalad supply. Most teleosts lack a coronary circulation, but when present
it is usually a cephalad supply (a few have both cephalad and caudal supplies,
e.g., eels and marlin). While the evolution of the cephalad coronary circula-
tion appears to be tied to evolution of the compact myocardium on both the
conus arteriosus (whose cardiac muscle cannot be easily supplied with a vasa
vasorum) and the ventricle, the caudal origin of the coronary system in
Latimeria is diYcult to reconcile, especially since there is little, if any, compact
myocardium on the ventricle (see below). There are indications for some
primitive fishes (e.g., Acipenser) that at least a portion of the coronary
circulation is not restricted to outer compact myocardium as in most teleosts,
but reaches the ventricular trabeculae as in all elasmobranchs.

2.1.2. The Branchial Circulation
     The number of gill pouches and gill arches is variable among primitive
fishes and this results in diVerent numbers of paired branchial arteries. Hag-
fishes have 5–13 pairs and lampreys have 6. Teleosts and most primitive fish
have 4 pairs of branchial arteries, but dipnoans and Amia have 5, although
the most anterior branchial artery does not serve a respiratory function. A
hyoid arch is present in Acipenser, Lepisosteus (¼Lepidosteus), Lepidosiren,
and Protopterus, but absent in Neoceratodus and Polypterus. A pseudo-
branch is present in Acipenser, Lepisosteus, and Neoceratodus, but absent in
Lepidosiren, Protopterus, and Polypterus.
    A high capillary density is a prerequisite for an eYcient gas exchange
surface. The capillaries of mammalian and reptilian lungs, as well as teleost
gills, have been described as a vascular sheet, in part because almost 90% of
the surface area is occupied by blood vessels (Farrell et al., 1980). The
vascular density in the gills of primitive fishes has not been well documented,
but appears to be high even in cyclostomes. Thus, the gill surface area of
primitive fishes probably has not been overdesigned at the expense of vascu-
lar density, except in the gills of air‐breathing fish where the lamellar vessels
are ‘‘channelized’’ at the expense of vascular density to provide a more direct
routing between the aVerent and eVerent arteries. Protopterus show the most
extreme form of these lamellar vascular channels, which even possess an
endothelial lining. The lamellar channels in Amia are much less extreme
and lack an endothelium, as do teleost secondary lamellar vessels.
60                                                        ANTHONY P. FARRELL

2.1.3. The Systemic Circulation
     All fish have an anatomically undivided circulation. The air‐breathing
lungfishes come closest to having a functionally divided circulation. Also,
there appears to be an early evolutionary polarization among vertebrates for
the left side of the heart to receive oxygenated blood. Protopterus certainly
return blood from the lung to the left side of the atrium and ventricle and
keep it largely separated from the systemic venous return. However, recent
work (Icardo et al., 2005b) has challenged the idea that the pulmonary vein of
Protopterus physically terminates in the atrium proper (see below), although
its terminus is such that oxygenated blood is directed to the left side of the
atrium. Thus, the veins of all primitive fish anatomically terminate in the
sinus venosus.

2.2. Physiological Patterns

    All primitive fishes have a myogenic heart that often operates at a low
heart rate when compared with teleosts at similar temperatures. The poten-
tial advantage of these slower heart rates is unclear (possibilities include a
lower myocardial oxygen demand or a laminar flow pattern through the
heart to prevent mixing of oxygenated and deoxygenated bloodstreams in
air‐breathing fish). The pacemaker rate in rainbow trout at 20  C is over
100 minÀ1 and such high heart rates are rarely observed at these tempera-
tures in primitive fishes. Therefore, the low heart rate in primitive fishes is
likely a result of a slow pacemaker rate, even though many primitive fishes
lack a neural cardiac excitatory mechanism.
    Ventricular myocytes also show a clear evolutionary progression
toward being intrinsically capable of more forceful contractions. The hagfish
ventricle has the poorest pressure generating ability and yet it has a mass
comparable to that found in teleosts. The lamprey ventricle generates a
higher central arterial blood pressure than hagfish but apparently uses
a larger ventricle. Ventral aortic blood pressure in other primitive fishes is
even higher, often more similar to elasmobranchs and lower than in most
teleosts. Thus, although cardiac myocytes superficially share a similar struc-
ture among all fishes, the integration of the cellular structures must be
diVerent for them to beat with diVerent tensions and at diVerent rates. The
cellular basis for these evolutionary shifts in rates and quantities of ion fluxes
associated with excitation–contraction coupling needs to be investigated.
    The progression toward higher arterial blood pressure among primitive
fishes has rendered redundant the accessory hearts that assist venous return
in hagfishes. Higher blood pressures likely allow regional blood flow to be
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                61

regulated in a more local and reliable manner with arterial, and perhaps
venous, vasoactivity.
    The evolutionary shift from the 15% blood volume for hagfish to the 3%
blood volume for rainbow trout has resulted in circulation time becoming
about fivefold shorter because routine Q does not vary greatly among fishes.
A faster circulation allows fish to detect variations in blood composition more
rapidly. A more rapid detection system is consistent with the other progressive
adaptations among primitive fishes of additional sensory systems (e.g., eyes),
eVector systems (e.g., neural control of the heart), more complex breathing
(e.g., bimodal water and air breathing), and more powerful locomotion.
    The evolution toward a higher central arterial blood pressure necessitated
mechanisms to protect the delicate blood vessels of the branchial circulation.
The evolution of a fourth cardiac chamber (the conus or bulbus arteriosus) is
likely related to the need for damping the systolic oscillations in blood
pressure and flow before they reach the lamellar capillaries. Why the conus
has been superseded by the bulbus is unclear, but it corresponds with higher
ventral aortic blood pressures. Certainly, the design characteristics of the
bulbus arteriosus are such that they allow accurate predictions of ventral
aortic systolic and diastolic blood pressures (David Jones, personal commu-
nication), but nothing is known about the conus in this regard. Despite this,
blood pressures in the secondary lamellae can be higher and more pulsatile
than those in mammalian capillaries.
    Cardiac filling requires energy and this can come in one of the several
forms, or in some combination. Vis‐a‐tergo cardiac filling (force from behind)
refers to kinetic and potential energy stored in the venous return as well as
energy that is generated by a preceding cardiac chamber. Vis‐a‐fronte cardiac
filling (force from front) refers to energy from the contraction of the ventricle
that is hydraulically coupled to expansion of the atrium and sinus venosus by
means of the pericardial cavity acting like a rigid box. Similar to elasmo-
branchs, a number of primitive fish orders have a stiV pericardium and can
utilize vis‐a‐fronte cardiac filling. A complete pericardium is absent in hag-
fishes, but in lampreys a closed and rigid pericardium assists venous return to
the heart. Thus, the evolution of a more powerful ventricle may have been in
part to assist venous return and replace the accessory hearts that are not
necessarily tightly coupled with the activity of the branchial heart. Alterna-
tively, vis‐a‐fronte filling may have been simply a consequence of the need for a
protective structure around the heart. The relationship between the evolution
of vis‐a‐fronte filling and vasoactivity in venous vessels is unclear, but both are
present in dogfish (Sandblom et al., 2006).
    Intrinsic modulation of cardiac stroke volume is possible by stretch (the
Frank‐Starling mechanism) in hagfishes. Thus, the underlying cellular
62                                                      ANTHONY P. FARRELL

mechanisms, which may be universal among vertebrate hearts, clearly
evolved before any neural mechanism to modulate pacemaker rate. Similar-
ly, paracrine excitatory control of cardiac activity (catecholamines stored in
cardiac chromaYn tissue) evolved before any neural mechanism to modulate
pacemaker rate being present in the cyclostome heart and several other
primitive fishes (dipnoans and Lepisosteus). These catecholamine stores pro-
vide tonic cardiac stimulation but the stimuli causing their release need
further study. The most primitive eVector system is therefore a paracrine
adrenergic stimulation of b‐adrenoceptors. This adrenergic transduction
mechanism allowed the evolution of humoral control of cardiac tissues by
plasma catecholamines well before sympathetic autonomic innervation of the
heart first appeared (e.g., in Amia). Therefore, although appropriate cellular
transduction mechanisms were present in primitive fish heart, eVerent sym-
pathetic nerve fibers were lacking. Consequently, a clear evolutionary pro-
gression exists toward (1) a more rapid eVector system to control cardiac
activity and, then, (2) a dual, push–pull (excitatory/inhibitory) control.
    The neural eVector system to rapidly slow heart rate appeared early in
vertebrate evolution. All primitive fishes, with the exception of cyclostomes,
have vagal cholinergic (muscarinic) inhibitory cardiac control. A cardiac
branch of the vagus nerve is absent in hagfish and first appears in lampreys.
However, vagal cholinergic control in lampreys is an obscure nicotinic car-
diostimulatory mechanism that predates the muscarinic inhibitory control
and, instead, resembles the control of the catecholamine release from stores
in the head kidney.
    Water‐breathing fish respond to environmental hypoxia with bradycar-
dia, but air‐breathing mammals do not, although both possess the same
vagal inhibitory control mechanism. Hypoxic bradycardia can reduce Q, if
there is no compensatory increase in cardiac stroke volume, which would
reduce myocardial oxygen demand. In addition, bradycardia can promote
oxygen transfer to the cardiac myocardium by extending diastole which then
prolongs blood residence time in the lumen and diastolic blood flow in the
coronary arteries. However, in most cases, Q is maintained by an increase in
stroke volume, which increases stroke work depending on how Laplace’s law
comes into play for trabeculated myocardium with a larger end‐diastolic
diameter. Randall (1982) suggested that the bradycardia and increased
stroke volume allow stroke volume to be matched with secondary lamellar
blood volume, thereby providing for a more even flow of blood through the
respiratory lamellae and more eYcient gas exchange. Another advantage
might be to prevent an excessive back pressure building up in sensitive gill
secondary lamellae if the heart continued to pump blood against a large
increase in systemic vascular resistance when strong skeletal muscle contrac-
tions compress skeletal muscle capillaries and stop blood flow. Lungfish lack
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                              63

this hypoxic bradycardia (Fritsche et al., 1993; Sanchez et al., 2001) and so
the evolution of lungs may have been associated with the loss of this piscine
    Primitive fishes tolerate severely hypoxic environments, which is perhaps
not surprising considering that their lineages have survived many environ-
mental changes. This tolerance is expressed in the circulation in at least three
important ways: a low cardiac ATP demand, a coronary circulation, and an
ability to breathe air. Farrell and Stecyk (2007) have argued that the low
myocardial power output of hagfishes has led to a routine cardiac ATP
demand that can be completely supplied through their glycolytic ATP capac-
ity during anoxia (rather than by a greatly elevated cardiac glycolytic capac-
ity). How widely this strategy extends among primitive fishes is diYcult to
determine given the relatively few measurements of myocardial power output
and maximum glycolytic capacity. Nevertheless, given that hagfishes have
the lowest ventral aortic blood pressure (one of the main determinants of
cardiac power output), the strategy may not be widespread. Instead, it
appears that improving oxygen supply to the myocardium through a coro-
nary circulation and air breathing has been a solution to inhabiting hypoxic
    While the cyclostome heart relies on venous blood for its oxygen supply,
all other primitive fishes have a coronary circulation in one form or another.
The compact myocardium benefits from the higher arterial oxygen tension
during hypoxia and especially during exercise. Farmer (1997) has argued that
the evolution of air breathing among fishes was to provide a myocardial
oxygen supply during exercise. This has to be subsequent to the evolution of
the coronary circulation, since a coronary circulation appeared in elasmo-
branchs apparently well before air breathing evolved in fishes. Although the
apparent loss of coronary support for cardiac muscle for the ventricle (but
not the conus) in Lepidosiren (Foxon, 1950) is consistent with Farmer’s
suggestion, problematic is the finding that the compact myocardium is
more extensive in the facultative air‐breathing Pacific tarpon (Megalops)
than in the water‐breathing rainbow trout (Farrell et al., 2007). Air breathing
is certainly beneficial to cardiac oxygen supply, but not necessarily at the
exclusion of a coronary circulation and probably more so during aquatic
hypoxia rather than exercise.
    A high proportion of primitive fishes are air‐breathers, and this has
resulted in numerous cardiovascular modifications that allow varying degrees
of functional separation between oxygenated venous return from the air‐
breathing organ and deoxygenated venous return from the rest of the body.
(Some of these features are shared by amphibians and reptiles.) Among the
three extant genera of lungfishes, there is a clear progression toward a more
divided circulation that parallels a greater dependence on air breathing
64                                                       ANTHONY P. FARRELL

(the Australian Neoceratodus only breathes air under hypoxic conditions,
while the South American Lepidosiren and the African Protopterus species
are obligate air‐breathers). In fact, deoxygenated blood from the body can
remain largely separated from the oxygenated blood returning from the lungs
during its passage through the lungfish heart and gills. However, no fish
possesses an anatomically divided circulation, as is the case in crocodiles,
mammals, and birds. Lungfishes show four circulatory changes apparently
critical to their successful transition to obligate lung breather:
     1. They send pulmonary venous return directly to the atrium and this
        prevents mixing of pulmonary and systemic venous return in the sinus
     2. They can keep pulmonary and systemic venous return largely sepa-
        rated within the atrium and ventricle (a prelude to a proper anatomi-
        cally divided, four‐chambered heart).
     3. They can direct deoxygenated arterial blood flow from the heart to the
     4. Cardiac output can bypass the respiratory surfaces of the gill.


    The superclass Agnatha contains the two orders Myxiniformes, the hag-
fishes, and Petromyzontiformes, the lampreys, also known as the cyclos-
tomes. The circulatory system of cyclostomes follows a basic craniate
pattern, but commands great interest by informing us on the cardiovascular
features likely present in the most primitive fish.

3.1. Hagfishes
    Excellent summaries of hagfish circulatory systems are found in Cole
(1926), Johansen (1963), Hardisty (1979), Forster et al. (1991), and Satchell
(1991). A comprehensive historical review of anatomical discoveries (Chapman
et al., 1963) suggests that these discoveries date back to Home (1815), Retzius
(1824), and Muller (1841). Important histological and cellular studies include
those by Augustinsson et al. (1956), Bloom et al. (1963), Leak (1969), Wright
(1984), and Fock and Hinssen (2002). Bloom et al. (1961) have summarized the
studies on the catecholamine‐containing granules found in cyclostome hearts.

3.1.1. Cardiac Anatomy
   The hagfishes have a main branchial (systemic) heart and three sets of
accessory hearts (see Section 3.1.2). The branchial heart has a sinus venosus,
an atrium, and a ventricle (Figure 2.1A), but lacks either a bulbus or a conus.
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                            65

                    RACV                                                LACV






 Internal     Hyoid efferent Efferent branchial
 carotids                                               Dorsal aorta

     Velar                                                                       Gill
     artery                                                                      pouches
 External                                   4        5
 carotid             1     2       3 Afferent branchials        6            7

                                                        Ventral aorta

Fig. 2.1. Arrangement of the hagfish circulation. (A) The major inflow and outflow vessels
for the branchial heart (Atr, atrium; Ven, ventricle; and SV, sinus venosus) and the portal
vein heart (PVH) (arrows indicate direction of flow; VA, ventral aorta; RACV and LACV, right
and left anterior cardinal veins; IJV, internal jugular vein; PCV, posterior caudal vein; CPV,
common portal vein; and SupV, intestinal portal vein). [Taken from Forster et al. (1991).]
(B) A schematic diagram of the major arteries to and from the gill pouches. [Taken from
Hardisty (1979).]
66                                                        ANTHONY P. FARRELL

The long ventral aorta (that traverses the elongate pharynx to reach the gills)
has a swelling just outside of the pericardium that resembles a bulbus
arteriosus and has a pair of semilunar valves. The posterior pericardium is
connected with the perivisceral coelom and therefore is not completely
    The hagfish heart is relatively large given its low power‐generating ability
(see below). Cardiac index is reported as 0.18% in the Atlantic hagfish
Myxine glutinosa (Satchell, 1986). Relative ventricular mass is reported as
0.1% for both Myxine (Johnsson and Axelsson, 1996) and the more active
Pacific hagfish Eptatretus cirrhatus, where atrial mass is one‐third of ventric-
ular mass (Forster, 1991). The ventricle is composed of spongy myocardium
that lacks a coronary circulation. Hagfish myocytes have a low myofibrillar
volume (consistent with a poor pressure‐generating ability), a large sarco-
plasmic reticulum, and lack a t‐tubule system, while possessing glycogen and
atrionaturetic peptide (ANP) granules in the sarcoplasm, and intercellular
desmosome connections (Augustinsson et al., 1956; Bloom et al., 1963; Leak,
1969; Helle and Lonning, 1973; Reinecke et al., 1987).
    The other main cardiac cell type, found next to the endothelium, is devoid
of myofibrils and is filled with densely staining granules associated with Golgi
bodies. These granules contain catecholamines (Bloom et al., 1961; Euler and
Fange, 1961; Nilsson, 1983), whose concentration varies considerably among
cardiac chambers (Table 2.1). Adrenaline is dominant in the ventricle, while
noradrenaline is dominant in the atrium and portal heart. The concentration
of adrenaline in the ventricle of the branchial heart of Myxine was approxi-
mately the same as in the head kidney of Atlantic cod (Gadus morhua) and
1/100th of that in the chromaYn cells of dogfish (Squalus acanthias).
    Greene (1902) was the first to suggest that cardiac innervation is lacking
in adult hagfish hearts, making them unique among vertebrates.

3.1.2. Circulatory Patterns
    The circulatory pattern of hagfish shows some important diVerences to
the generalized single circulation of fishes. Foremost, the circulatory pattern
to the 5–13 pairs of gill pouches is unusual among fishes (Figure 2.1B). Each
aVerent branchial artery divides and encircles a water duct, thereby supply-
ing both hemibranchs of a single gill pouch as well as the lamellae that project
radially into that gill pouch. Each pair of eVerent branchial vessels unites
onto two lateral aortae (an unusual duplication compared with other fishes),
which then connect to a single median dorsal aorta. Another unusual feature
is that the blood supply to the kidney is entirely arterial, that is the renal
portal circulation of other fishes is lacking. In addition, the venous system
has unusual asymmetries: the right ductus Cuvier is absent, and the left
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                               67

                                         Table 2.1
     Catecholamine (AD, adrenaline; NA, noradrenaline) Concentrations in Tissues (mg gÀ1;
      Cardiac Chamber and Blood Vessels) and Plasma (nmol literÀ1) of Primitive Fishes

      Fish               Tissue               AD        NA     AD (%)         References

M. glutinosa      Ventricle               59        6.5        92                   ¨
                                                                         Euler and Fange,
                  Atrium                  8.1       18         31          1961
                  Portal heart            3.1       58         5
                  Kidney                  –         16         –
M. glutinosa      Ventricle               49        6.2        89        Reported in Bloom
                  Atrium                  13        47         22          et al., 1963
                  Portal heart            3.4       53         64
M. glutinosa      Ventricle and           21        20         53        Perry et al., 1993
                  Posterior               1.9       49         4
                    cardinal vein
L. fluviatilis     Ventricle               81        12         87        Reported in Nilsson,
                  Atrium                  127       16         89          1983
L. fluviatilis     Ventricle               28        0          100       Reported in Bloom
                  Atrium                  130       6.3        95          et al., 1963
P. marinus        Ventricle               9.8       1.3        88        Reported in Bloom
                  Atrium                  51        2.0        96          et al., 1963
Protopterus       Atrium                  4.2       71         6         Abrahamsson et al.,
                  Prox. intercostal       216       94         70          1979a
                  Left cardinal vein      0.55      0.03       95
                  Plasma mg/100 ml        11 (27)   11 (160)   50 (17)
                    (stressed fish)
Protopterus       Plasma—normoxia         5         5          50        Perry et al., 2005
                  Plasma—aquatic          8         8          50
                  Plasma—aerial           18        20         47
L. platyrhincus   Ventricle               0.27      0.03       90        Nilsson, 1981
                  Atrium                  1.8       0.23       89
                  Cardinal vein           47.5      21.5       69
                  Plasma                  54        4.4        93
Amia calva        Plasma                  13.3      9.0        40        McKenzie et al.,
                  Plasma: HCl infusion    730       703        49         1991a
                  Plasma: hypoxia         42.6      7.8        15        McKenzie et al.,
A. naccarii       Plasma: normoxia        4.3       5.2        55        Randall et al., 1992
                  Plasma: hypoxia         29.9      45.1       60
68                                                                                                          ANTHONY P. FARRELL

             A                          cbv                               B

                    rsc               rdc


                          roc                                lac
                                              ijv                             Position of                         Heart

                                  pbd                                                                               is


                 ph                                                 saa
                                       gut                   sv
                                       lac                                                                                        ilt

                 pv                    ah                                                             lcs
                                                             siv                                                            ilt


                                rpc                    lpc

                                                                                            Right                   Left
                                                    cv                                                                cv
                                                                                       scs                               cdh
                 Right                                             Left
                                                  cv                                                              cv
                                        mvb                                                                 mvb

Fig. 2.2. Arrangement of major (A) veins and (B) sinuses in hagfish. [ah, anterior hepatic vein;
ca, anterior cardinal anastomosis; cbv, branchial constrictor; cdh, caudal heart; clh, cardinal
heart; cysv, cystic vein; cpv, common portal vein; cv, caudal vein; hvs, hypophysiovelar sinus; ijv,
inferior jugular vein; ilt, intestinal lymphatic trunk; is, intestinal sinus; lac, left anterior cardinal
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                    69

anterior cardinal vein is larger than the right one and drains into both the
branchial and portal hearts (Figure 2.2A).
     The chronology of the discovery of the accessory portal heart has been
described by Fange et al. (1963) and Johansen (1963). The portal heart has a
single chamber that consists of cardiac muscle arranged as thin trabeculae
(each containing four to six fibers). It has its own pacemaker activity and
looks much like the atrium of the branchial heart. Also like the atrium, the
portal heart stores noradrenaline at higher (20 times) concentrations than
adrenaline (Table 2.1). The cellular structure of the portal heart has been
described for Paramyxine atarii (Endo et al., 1997) and Myxine (Helle and
Lonning, 1973) and, much like the branchial heart, its myocytes are char-
acterized by the presence of myofibrils, mitochondria, an extensive sarco-
plasmic reticulum, glycogen granules, ANP immunoreactivity, and the
absence of a t‐tubular system. The portal heart has two input vessels and
one output vessel, each protected by a valve. Its main input comes from the
supraintestinal/portal veins that drain the gut circulation. Thus, its main role
is to overcome the vascular resistance of the hepatic circulation (Fange et al.,
1963), moving blood through the liver to the sinus venosus via hepatic veins.
The portal heart has additional input from the left anterior cardinal vein, but
why a portion of the cephalic venous drainage can go directly to the liver via
the portal heart is a mystery.
     Blood volume of hagfishes is around 18% and is the largest of all fishes
(Hardisty, 1979). This large volume reflects extensive subcutaneous blood
sinuses (Figure 2.2B), whose structure, function, control, and associated
accessory hearts have received much attention and debate (Johansen et al.,
1962; Satchell, 1984, 1991; Forster, 1997). The original idea that the sinuses
were a primitive lymphatic system has been abandoned based on their high
plasma protein content (Johansen et al., 1962) and the presence of red blood
cells in cannulated hagfish (Forster et al., 1988), albeit at about one‐third of
the hematocrit of the primary circulation.
     There are three large blood sinuses, each connected to arterial vessels via
vascular papillae, about the caliber of one red blood cell (Cole, 1926). The
secondary circulatory system of teleosts has similar arterial connections that
filter red blood cells (SteVensen and Lomholt, 1992), and this has led to the
suggestion that hagfish sinuses are the forerunner of the less capacious

vein; lcs, lateral chordal sinus; lpc, left posterior cardinal vein; ls, lingual sinus; mvb, median
ventral bar of caudal fin; pba, peribranchial anastomosis; ph, portal heart; pv, portal (supra
intestinal) vein; rac, right anterior cardinal vein; rdc, right deep anterior cardinal vein; rpc, right
posterior cardinal vein; rs, rectal sinus; rsc, right superficial cardinal vein; saa, sinoatrial aper-
ture; sca, subcutaneous anastomosis; scs, subcutaneous sinus; siv, subintestinal (posterior)
hepatic vein; sv, sinus venosus]. [Taken from Cole (1926).]
70                                                        ANTHONY P. FARRELL

secondary circulatory system (Satchell, 1992). The caudal subcutaneous
sinus is the largest sinus, and it drains into an accessory caudal heart. The
caudal heart of the hagfish provides venous return from the tail, as do
analogous caudal heart structures in sharks and teleosts (see Satchell, 1992
for discussion).
    The caudal heart arises, just posterior to the last of the mucus glands, as
two swellings of the paired caudal veins separated by a medial cartilaginous
fin plate (Figure 2.3A and B). The swellings are encased on either side by a
fan‐shaped skeletal muscle, each of which is innervated by single motor
nerves arising from three spinal roots (Satchell, 1984). In addition to two
veins from the subcutaneous sinus and minor inputs from the caudal veins,
there is input from the marginal veins that collect blood from radial fin veins.
The caudal heart pumps blood into the posterior cardinal vein and contrac-
tion of each side of the caudal heart generates a venous pressure pulse
(Figure 2.3B).
    The peribranchial sinus has anastomoses with the anterior cardinal veins
(Figure 2.2A). The hypophysiovelar sinus is located in the head region and
drains into the inferior jugular vein via the cardinal ‘‘hearts,’’ which Satchell
(1991) has considered ‘‘propulsors’’ rather than ‘‘hearts’’ since they are
driven by the extrinsic muscles of the velar.

3.1.3. Circulatory Dynamics
    Various aspects of circulatory physiology in hagfish have been reviewed
in Satchell (1984), Forster et al. (1991), and Forster (1997). The electrical
properties of hearts from M. glutinosa, Eptatretus stoutii, and E. cirrhatus
have been described by Bloom et al. (1963), Chapman et al. (1963), Arlock
(1975), Satchell (1986), and Davie et al. (1987).
    The branchial heart is myogenic, a characteristic of all vertebrate hearts.
Desmosomes provide intercellular connections between cardiomyocytes (Leak,
1969). All three cardiac chambers are capable of independent pacemaker
activity, although this capability is weakest for the ventricle (Bloom et al.,
1963). Pacemaker potentials have been reported in atrial cells from Myxine
(Arlock, 1975). Jensen (1965) reported unusually low resting membrane
potentialsfor E. stoutii (atrium ¼ –41 mV and ventricle ¼ –48 mV).
    The branchial heart of both Myxine and E. cirrhatus has a prolonged
electrocardiogram (ECG) (Figure 2.4A and B). The electrical conduction
times between the cardiac chambers are two to three times longer than other
fishes. The prolonged ECG results in atrial depolarization being visible in the
trace (a Pr wave; Figure 2.4A). A contributing factor in this electrical delay is
likely the long, funnel‐like atrioventricular connection (Figure 2.1). Never-
theless, ventricular dP/dt is about 10 times slower that that found in teleosts
2.    CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                               71

 A                                  B



                                   LMG                                                        DCV

                                            VScS                          MV

 Caudal vein pressure


     Mechanogram                                                Down
                                                     (2.5 s)

Fig. 2.3. Caudal hearts of hagfish. (A) A dorsal view to show the two valved swellings either side
of a median plate (hatched, MP) that formed the caudal heart and the direction of blood flow
(arrows) in the major inflow and outflow vessels (CV, caudal vein; VScS, vein from subcutaneous
sinus; and MV, marginal vein). (B) A lateral view of one side of the caudal heart (CH) illustrating
the motor nerves (MN) from the spinal column (SpC) that innervate the caudal heart muscle
(CHM) (Nch, notochord; KMP, knob of median plate; LMG, last of the series of mucus glands;
RV, radial vein; and DCV, distal caudal vein). (C) A tracing of blood pressure in the caudal vein
and a mechanogram from the surface of a caudal heart. Note that the venous pulse pressure is
caused by the alternate activity of each side of the caudal heart. [Taken from Satchell (1992).]

and this is probably due to slow muscle activation kinetics (Satchell, 1986;
Davie et al., 1987).
     Routine heart rates in hagfishes are typically reported as 20–30 minÀ1 at
10  C, and an unimpressive maximum of 35 minÀ1 in vivo (but 15–42 minÀ1

at 8–10  C in anesthetized E. stoutii; Chapman et al., 1963). Greene (1900,
1902) and Augustinsson et al. (1956) conclusively showed that neither vagal
stimulation nor applied acetylcholine altered branchial heart rate. Therefore,
any control of heart rate in hagfish must be aneural. (Curiously acetylcholine
has been isolated from both branchial and portal cardiac tissues.)
     There is very strong evidence that the hagfish heart is under a tonic,
paracrine b‐adrenergic stimulation, likely from cardiac catecholamine stores.
72                                                                                       ANTHONY P. FARRELL


                                                  V      P         Q S

                                              Ventral aorta
                   Pressure cm H2O





                                              0                1            2      3    s
                                                       A       B        C   D EF
                   cm H2O

                                                        Q          S


                                      8                                                  aorta
                   Pressure cm H2O


                                      4                                                Ventricle


                                          0                        1            2 s
                                                   A           B    C       D   E F

Fig. 2.4. ECGs and superimposed traces from cardiac chambers and major arteries of lightly
anesthetized hagfishes. (A) Pacific hagfish E. cirrhatus [taken from Davie et al. (1987)] and
(B) Atlantic hagfish M. glutinosa [taken from Satchell (1986)]. Note the higher pressure develop-
ment in the Pacific hagfish.
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                            73

Depletion of catecholamine stores revealed positive inotropic and chrono-
tropic eVects of adrenaline and noradrenaline (Fange and Ostlund, 1954).
Likewise, Chapman et al. (1963) found a modest inhibitory inotropic eVect of
reserpine, which then allowed a restorative, stimulatory eVect of adrenaline
on isolated E. stoutii hearts. Sotalol reduced intrinsic heart rate of the
working, perfused Myxine hearts (Figure 2.5A) but did not prevent a modest
3–4 minÀ1 stretch‐induced increase in heart rate (Johnsson and Axelsson,
1996). Indeed, in vivo injections of sotalol in Myxine (Figure 2.5B) and
propranolol in E. cirrhatus produced a large reduction in heart rate to around
15 minÀ1 (Axelsson et al., 1990; Forster et al., 1992). Thus, temperature and
the slow release of endogenous catecholamines are the two main modulators
of heart rate in hagfishes.
    An unresolved question regarding the control of heart rate is to what
degree and under what conditions there is a paracrine regulation of cardiac
contractility and heart rate (see Reid et al., 1998 for general discussion of
catecholamine release in fishes). Perfused Myxine hearts normally release
endogenous catecholamines (Bloom et al., 1963) but at low levels (1 nmol
literÀ1; Perry et al., 1993). In perfused hearts, this release is unaVected by
either acidosis or high filling pressure, but is stimulated tenfold by nonphy-
siological conditions of 60‐mM potassium and carbachol (a cholinergic
agonist), and 100‐fold with pituitary extracts. Thus, it is possible that
the cardiac tissues have a very high aYnity for catecholamines and paracrine
release results in very little overflow into the circulation. Alternatively,
there may be a central humoral control (Perry et al., 1993). Potentially,
release could even be triggered by a very low threshold to stretch, which
could then kick‐start a quiescent heart (Axelsson et al., 1990). But whether
heart rate declines under long‐term adverse conditions, thereby slowing the
hagfish circulation, is still not known.
    Another unresolved question regarding the control of heart rate is to
what degree increased venous return or cardiac filling can increase heart rate.
Results are highly variable among studies, with a quadrupling of heart rate
in E. stoutii (up to 30 minÀ1 at 15  C; Jensen, 1961) versus a modest 20%
increase (again up to 30 minÀ1 at 10  C; Chapman et al., 1963). These
responses could reflect either a stretch‐induced pacemaker eVect (as sug-
gested by Jensen, 1961) or a release of catecholamine stores (as suggested
by Johansen, 1963). More recent studies of working, perfused hearts from
E. cirrhosus and Myxine point to a more modest stretch‐induced eVect
because intrinsic heart rates in hagfish are similar to those recorded in vivo
(Table 2.2) and only a modest (<15%), if any, increase in heart rate occurs
with cardiac stretch (Forster et al., 1992; Johnsson and Axelsson, 1996).
    On the basis of in vitro studies, it appears that humoral eVects of cate-
cholamines on heart rate are likely very modest. Heart rate in working,
74                                                                                                                                ANTHONY P. FARRELL

                                                                                                   Sotalol treated

                                                (ml min−1 kg−1 BM) (ml beat−1 kg−1 BM)
                                            A                                            2.0
                                                                                         1.5                         *

                                                                                         0.5                         *

                                                                                         40 B


                                                                                         0.8 C
                                                Power output
                                                (mW g−1 VM)

                                                                                         0.6                         *
                                                                                         0.2                         *

                                                                                               0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16
                           B                                                                                  Pin (kPa)
      P VA (kPa)

      P DA (kPa)

      f H (beats min−1) Q (ml min−1)


                                                                                                    Sotalol                               1 min

Fig. 2.5. Cardiac performance in Myxine and the inhibitory eVect of b‐adrenergic blockade.
(A) Performance of the working, perfused heart illustrating (1) the Frank‐Starling eVect of
cardiac filling pressure (Pi) on cardiac stroke volume (Vs), cardiac output (Q), and myocardial
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                               75

                                        Table 2.2
                 A Comparison of Cardiovascular Performance for Hagfishes

                                           Myxine 10  C                 Eptatretus 17  C

                                                     In vitro—                       In vitro—
                                       In vivo          max            In vivo          max

Power output (mWgÀ1)                 0.15 (0.62)       0.50          0.42 (0.88)       0.37
Cardiac output (ml minÀ1 kgÀ1)       8.7 (25)          29.4          15.8 (23)         22
Heart rate (minÀ1)                   22 (25)           23            25 (29)           32
Stroke volume (ml kgÀ1)              0.41 (1.0)        1.3           0.67 (1.0)        0.71
Ventral aortic pressure (kPa)        1.0 (1.6)         1.8           1.6 (2.3)         1.4
Dorsal aortic pressure (kPa)         0.77 (1.0)        NA            1.3 (1.9)         NA
Branchial resistance                 31 (24)           NA            20 (17)           NA
  (Pa min kg mlÀ1)
Systemic resistance                  89 (40)           NA            84 (82)           NA
  (Pa min kg mlÀ1)

    Values in parentheses are the peak in vivo response to an injection of 10‐nmol kgÀ1
    Myxine data taken from Axelsson et al. (1990) and Johnsson and Axelsson (1996).
    Eptatretus data taken from Forster (1989) and Forster et al. (1992).

perfused Myxine heart was largely insensitive to adrenergic agonists
(increases of just 3–5 minÀ1), cholinergic agonists, and atropine (Johnsson
and Axelsson, 1996), confirming earlier findings with adrenaline for Myxine
and E. stoutii hearts (Fange and Ostlund, 1954; Chapman et al., 1963).
Furthermore, plasma catecholamines are normally in the nanomolar range
(Table 2.1) and even severe, acute hypoxia that reduced blood oxygen tension
below the P50 for hemoglobin did not significantly increase these plasma
catecholamine levels (Perry et al., 1993).
    Cardiac performance and contractile properties have been studied
in unanesthetized hagfish (Forster et al., 1988, 1992; Axelsson et al., 1990),
lightly anesthetized hagfish (Johansen, 1960; Chapman et al., 1963; Satchell,
1986; Davie et al., 1987), and working, perfused heart preparations (Forster,
1989, 1991; Johnsson and Axelsson, 1996). Like all vertebrate hearts, the
branchial and portal hearts both follow the Frank‐Starling law of the heart:
an increase in cardiac filling pressure can increase stroke volume severalfold
(Figure 2.5A). Stroke volume can reach 1.3 ml kgÀ1 in Myxine and 0.71 ml kgÀ1

power output and (2) the inhibitory eVect on this response of the b‐adrenergic blocking
agent sotalol. [Taken from Johnsson and Axelsson (1996).] (B) In vivo cardiovascular recordings
(PVA and PDA, ventral and dorsal aortic blood pressures) to illustrate the marked slowing of
heart rate ( fH) after sotalol injection. [Taken from Axelsson et al. (1990).]
76                                                         ANTHONY P. FARRELL

in E. cirrhatus (Table 2.2). Venous blood pressures and cardiac filling pres-
sures are above ambient (Figure 2.4A and B), which is consistent with vis‐a‐
tergo cardiac filling (Johansen, 1960; Satchell, 1986).
    Cardiac output is routinely 9–16 ml minÀ1 kgÀ1 (Table 2.2), a range that
is comparable with benthic teleosts at similar temperatures (Forster et al.,
1991; Farrell and Jones, 1992). Routine circulation time is estimated as a
sluggish 12 min for Myxine and 6 min for E. cirrhatus (based on a blood
volume of 150 ml kgÀ1 and ignoring the 30% volume found in sinuses, which
has a turnover time of many hours; Forster, 1997). Maximum Q is unimpres-
sive in hagfish (22–30 ml minÀ1 kgÀ1; Table 2.2) but decreases circulation
time to 3–5 min. Consistent with the absence of a bulbus arteriosus, flow
traces from the ventral aorta show little diastolic run oV and a zero flow
(Figure 2.6A).
    Hagfish hearts stand out from all other vertebrates by their low and
slow pressure‐generating ability. Intracardiac pressure recordings in anesthe-
tized animals (Figure 2.4A and B) have been confirmed in vivo and with
working, perfused heart preparations (Table 2.2). A peak ventricular pres-
sure of just 1.04 kPa was reported for Myxine (Johansen, 1960) and 2.6 kPa
for E. stoutii (Chapman et al., 1963). End‐systolic volume is low for the
hagfish ventricle (Hol and Johansen, 1960), as in rainbow trout (Franklin
and Davie, 1992). Chapman et al. (1963) found little eVect of temperature on
peak ventricular performance of E. stoutii between 11 and 21  C; peak
inotropic responsiveness occurred around 14  C. The hagfish heart uses
oxygen eYciently, with a myocardial oxygen consumption rate that is not
vastly diVerent to teleost values (around 0.4‐ml O2 sÀ1 mWÀ1; Forster et al.,
1991). Furthermore, the glycolytic capacity of the heart can support most of
the routine myocardial power (Forster, 1991; Farrell and Stecyk, 2007),
suggesting that cardiac contractility is unlikely to be severely compromised
by a poor oxygen supply. The poor cardiac contractility of the hagfish
heart reflects a low myofibril content of cardiomyocytes, but other cellular
processes related to excitation–contraction coupling need to be studied in
this regard.
    Hagfishes have a low central arterial blood pressure (1.0–1.6 kPa;
Table 2.2). This results in routine and maximum myocardial power output
values that are lower than all other vertebrates. E. cirrhatus has a slightly more
powerful heart than Myxine (Table 2.2). Maximum arterial pressure is an
unremarkable 1.8–2.3 kPa, about one‐fifth of the maximum pressure attained
by the rainbow trout heart. Approximately one‐third of arterial blood pres-
sure is lost across the gill circulation (Table 2.2) (Bushnell et al., 1992).
Johansen (1960) reported mean dorsal aortic blood pressure as 0.52–
0.78 kPa for Myxine, a range similar to subsequent reports (Table 2.2), but
higher than the 0.4 kPa reported for E. stoutii (Chapman et al., 1963).
2.                 CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                                  77



                                                            P (kPa)
                                                                                               Ventral aortic pressure
                   2                                                                           Dorsal aortic pressure
PVA (kPa)


                                                          fH (beats min−1)
                   0                                                               30
PDA (kPa)


                   0                                                               80

                                                             Q and VS (% change)
                    3                                                                        =Q
Q (ml min−1)

                                                                                             = Vs
                    0                                                               0
fH (beats min−1)

                   30                                                              −20
                                                                                              = Systemic
                                                             R (% change)

                         2s           2 min                                        25         = Branchial


                                                                                         0    1     2   3 4 5 6       7   8   9 10
                                                                                                         Time (min)

Fig. 2.6. Cardiovascular performance in Myxine. (A) In vivo cardiovascular recordings to illus-
trate the zero diastolic flow in the ventral aorta (PVA and PDA, ventral and dorsal aortic blood
pressures; Q, cardiac output; and fH, heart rate). [Taken from Axelsson et al. (1990).] (B) The
stimulatory eVect of an intraarterial injection of adrenaline (P, blood pressure; Vs, cardiac stroke
volume; Q, cardiac output; and R, vascular resistance). [Taken from Forster et al. (1992).]

     The accessory portal heart has weaker contractions, as well as a lower
output, than the branchial heart. Blood pressure was reported as 0.25 kPa
(pulsing between 0.39 and 0.19 kPa) for Myxine (Johansen, 1960) and similarly
0.03–0.28 kPa for E. cirrhatus (Davie et al., 1987), that is, mean pressures that
are about one‐fifth of those produced by the systemic heart. Flow was 0.1–
0.3 ml minÀ1kgÀ1 in Myxine (Johansen, 1960). Whether performance of the
portal heart increases appreciably, say during digestion, has not been tested.
Fange et al. (1963) noted that the portal heart could be distended to the size of
the ventricle. Chapman et al. (1963) reported portal heart rates of 43–60 minÀ1
at 8–10  C in anesthetized E. stoutii. For Myxine, beating rates of 47 minÀ1 at
20  C and 24 minÀ1 at 10  C (Fange et al., 1963) suggest a Q10 around 2. The
myogenic portal heart beat is uncoupled from that of the branchial heart beat
and the two hearts do not necessarily beat synchronously. Pacemaker activity
78                                                         ANTHONY P. FARRELL

of the portal heart originates upstream in the supraintestinal vein, as demon-
strated by the vein continuing to pulse and responded with an increase in pulse
frequency with electrical and mechanical stimulation when severed from the
heart (Fange et al., 1963), which is similar to the behavior of the portal heart
when filled (Johansen, 1960). The modest inhibitory eVect of sotalol on portal
heart rate in vitro and the modest stimulatory eVect of adrenaline following
sotalol application suggest a tonic b‐adrenergic control similar to the branchial
heart (Johnsson and Axelsson, 1996).
    The caudal accessory heart regulates venous return blood from the tail and
the posterior subcutaneous sinuses. Excellent descriptions of hagfish caudal
heart function and comparisons among fishes are provided by Satchell (1984,
1991, 1992). The caudal heart in hagfish, like other fishes, is not myogenic, but is
reflexly driven by spinal nerves (Figure 2.3B). It has a rate faster than the
branchial and portal hearts, can pause for many minutes, and becomes active
shortly after swimming. Greene (1900) provided the first mechanograms of the
beating caudal heart, demonstrating an arrhythmic beat in vivo, a steady beat
in vitro, and inhibition by touch to the skin (suggesting aVerent neural control).
Chapman et al. (1963) reported pulse rates of 60–72 minÀ1 at 8–10  C in
anesthetized E. stoutii. The alternate contractions of the paired skeletal mus-
cles, with a slight pause between them, create an intriguing double pressure
pulse of about 0.002 kPa superimposed on a diastolic venous blood pressure of
about 0.162–0.165 kPa (Figure 2.3C).
3.1.4. Circulatory Control
    In vivo studies of circulatory control in nonanesthetized hagfish are quite
limited (Forster et al., 1988, 1992; Axelsson et al., 1990). In addition, drug
injection studies must use small volumes and be interpreted with care because
hagfish have a slow circulation and a sluggish regulation of blood pressure
(barostatic reflexes are impossible with an aneural heart) (Forster, 1997;
see also Chapman et al., 1963). Consequently, increases in blood volume
through injections can produce slowly compensated hypertension. A further
complication is that blood pressure is also vulnerable to muscular activity.
    Adrenaline injection in Myxine increases Q and heart rate (Figure 2.6B)
to levels close to maximum cardiac performance (see Table 2.2). This result
could be interpreted as a consequence of either direct adrenergic stimulation
of the heart or venous vasoconstriction mobilizing venous return to trigger
Frank‐Starling and pacemaker stretch responses.
    Gill and systemic vascular resistances in hagfishes are at the low end of
the range for fishes (Table 2.2) (Bushnell et al., 1992). In vivo injection of
adrenaline decreases systemic vascular resistance without altering branchial
vascular resistance (Axelsson et al., 1990). However, adrenaline and acetyl-
choline constrict perfused gill preparations from Myxine, while isoprenaline,
noradrenaline, and adenosine produce vasodilatations (Axelsson et al., 1990;
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                             79

Sundin et al., 1994). Acetylcholine and adrenaline constrict the perfused
systemic vasculature of Myxine (Reite, 1969).
    Spontaneous swimming activity modestly increases ventral aortic blood
pressure and heart rate without changing dorsal aorta blood pressure
(Forster et al., 1988). Postactivity, heart rate increases further, blood pres-
sure is restored, and hematocrit in the sinuses increases (Forster, 1997).
Blood flow into the sinus is regulated by arterial blood pressure (Forster,
1997) and gill vasoactivity (Sundin et al., 1994). Blood flow distribution
between the primary circulation and sinuses of hagfish gills is likely regulated
by tonic adrenergic mechanisms since adrenaline (more so than isoprote-
renol) decreases sinus blood flow in perfused gills from 50% to about 10%.
Thus, the common neural and humoral vascular smooth muscle control
mechanisms (cholinergic and a‐adrenergic vasoconstrictions and b‐adrener-
gic vasodilation) are found in hagfishes. Likewise, Myxine possess many
paracrine vascular smooth muscle control mechanisms. Using an isolated
ventral aorta preparation, Evans and Harrie (2001) showed that endothelin‐1
elicited a strong contraction and NO a modest contraction, whereas a strong
relaxation was elicited by ANP, a prostaglandin I2 agonist, and prostaglan-
din E2. Prostacyclin was without eVect.
    Hagfishes inhabit hypoxic sediments, burrow into prey during feeding,
and tolerate extreme hypoxia (Hansen and Sidell, 1983). Hypoxia tolerance is
aided by a high blood volume (to buVer anaerobic wastes), high cardiac
glycogen stores, a peculiarly thick cardiac glycocalyx (which may be impor-
tant in protecting the extracellular calcium supply to cardiac myocytes from
the eVects of extracellular acidosis; Poupa et al., 1985), and an extremely low
myocardial power output. On the basis of Forster’s (1991) estimates of
the glycolytic ATP‐generating capacity of the E. cirrhatus heart, maximum
glycolytic ATP turnover rate (41 nmol ATPsÀ1 gÀ1) actually lies below
that for the trout, but ATP turnover rate normalized to work output
(146 nmol JÀ1) is similar to other vertebrates (Farrell and Stecyk, 2007). As
a result of this cardiac glycolytic capacity closely matching routine needs,
hagfish can likely circumvent the need for a myocardial oxygen supply
(Farrell, 1991). Therefore, hagfish may not need to downregulate cardiac
activity during anoxia as a protective mechanism, a feature that could be
important for the assimilation of food and the distribution of glucose during
feeding. Acute (<20 min) hypoxia actually increased rather than decreased
myocardial power output (up to 0.4 and 0.7 mW gÀ1 in Myxine and Eptatre-
tus, respectively) and Q (Figure 2.7) possibly because the fish became agitated.
However, the eVects of chronic hypoxia in hagfishes are unknown, other than
Myxine can maintain the same ‘‘relative cardiac performance’’ (calculated as
the product of heart rate and displacement of the pericardium) for 3 h of
anoxia at 5  C and, over a 20‐h anoxic period, during which cardiac glycogen
stores are depleted from 22 to 0.9 mmol gÀ1 (Hansen and Sidell, 1983).
80                                                                                                            ANTHONY P. FARRELL

A                      1.0                                                     B                25
                                                          Severe hypoxia
                       0.8                                                                      20
Cardiac powe routput

                                                                            (ml min −1 kg −1)
                                                                            Cardiac output
      (mW g −1)

                       0.6                                                                      15

                       0.4                                                                      10

                       0.2                                                                       5

                       0.0                                                                       0
                             Myxine                        Eptatretus                                  Myxine        Eptatretus

                                      C                   2.5

                                      Arterial pressure




                                                                   Myxine                        Eptatretus

Fig. 2.7. EVect of an acute, severe hypoxia exposure on cardiac performance of M. glutinosa and
E. cirrhatus. Note that performance is not depressed with hypoxia. [Taken from Farrell and
Stecyk (2007).]

3.2. Lampreys
   Fange (1972) has provided an excellent summary of the lamprey circula-
tory system, which has been studied to a lesser extent than the hagfish.
Hardisty (1979) has compared the two cyclostome circulatory systems.

3.2.1. Cardiac Anatomy
    Descriptions and historical citations on various aspects of cardiac anato-
my are found in Augustinsson et al. (1956), Bloom et al. (1961), Bloom
(1962), Johansen (1963), Kilarski (1964), and Wright (1984).
    The branchial heart of the lamprey is unusually large by vertebrate
standards. Remarkably, the adult cardiac index [reported as 0.59% in
Fange (1972) and 0.25–0.41% (but only 0.1% in ammocoetes) in Hardisty
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                  81

(1979)] exceeds the ventricular index of tunas (but note that relative
ventricular mass was only 0.07% for Lampetra planeri; Ostadal and     ¨
Schiebler, 1971). The ventricle is composed of spongy myocardium and
lacks a coronary circulation. The blood volume of lampreys is large (8%;
Fange, 1972) by vertebrate standards, and about half that of hagfishes. The
cardiomyocytes contain well‐defined myofibrils, sarcoplasmic reticulum, and
desmosomes (Bloom et al., 1961; Kilarski, 1964), similar to the situation in
    Lampreys are extant representatives of the first vertebrate with cardiac
innervation. The vagus nerve travels along the median jugular vein and
innervates the sinus venosus. Vagal stimulation accelerates the heart via
nicotinic receptors (Augustinsson et al., 1956), a unique feature among
vertebrate hearts because cardiac vagal stimulation typically causes brady-
cardia via muscarinic receptors (Nilsson, 1983).
    Lampreys, like hagfishes, have specialized cardiac cells with catecholamine‐
containing granules located in a loose subendothelial layer (Augustinsson et al.,
1956; Bloom et al., 1961, 1963; Lignon, 1979). The ratio of adrenaline to
noradrenaline varies among tissues, with adrenaline being dominant in the
ventricle and atrium, and noradrenaline being dominant in the sinus venosus,
similar to the portal heart of hagfishes (Table 2.1). The granules are arranged
like chains of ganglionic cells throughout the three cardiac chambers (and to a
lesser extent in the jugular vein), giving them a resemblance to the varicosities of
adrenergic nerve terminals, but without any synaptic structures (Augustinsson
et al., 1956; Falck et al., 1966).
    The sinus, atrium, and ventricle are fully enclosed in a pericardium that
is reinforced by cartilage. There is no bulbus or conus arteriosus and a
ventral aorta (truncus arteriosus) lies outside the pericardium. Two semilu-
nar valves guard the outflow of the ventricle. The wall of the ventral aorta
contains vascular smooth muscle and collagen, but apparently lacks elastin
(Wright, 1984, but see Fange, 1972) and is thicker than that of hagfish,
presumably reflecting a higher blood pressure (see below). Near to the
heart, and unique among vertebrates, is an ‘‘intra‐arterial cushion,’’ which
is a loose arrangement of connective tissue that greatly restricts the lumen of
the ventral aorta (Figure 2.8A). The function of this cushion is unknown, but
could either help create turbulent blood flow (to limit settling of red blood
cells; see Wright, 1984 for discussion) or aid in vasoconstriction (see below
for vasoactivity).
3.2.2. Circulatory Patterns
    The gill vascular pattern in Lampetra japonica has been superbly detailed
(Nakao and Uchinomiya, 1978). Seven pairs of aVerent branchial arteries
serve the anterior and posterior hemibranchs of adjacent gill pouches
82                                                                        ANTHONY P. FARRELL

                     A                                          VA

                         CA          D




              Hyoid efferent Efferent branchial
 carotids                                                 Dorsal aorta

     Velar                                                                          Gill
     artery                                                                         pouches
 External                                     4         5
 carotid            1       2        3 Afferent branchials           6        7

                                                          Ventral aorta

Fig. 2.8. Major vessels associated with the gills of the lamprey. (A) A histological section of the
ventral aorta (VA and CA) just outside of the pericardium and ventricle (V) to illustrate
the vascular cushion that protrudes into the lumen (L) beyond the ostial valves (arrows).
[Taken from Wright (1984).] (B) A schematic diagram of the arterial arrangement. [Taken from
Hardisty (1979).]

(Figure 2.8B) and, in doing so, they resemble other fish rather than hagfish,
where an aVerent branchial artery supplies just one gill pouch. The eVerent
branchial arteries mimic this pattern and join the dorsal aorta. AVerent and
eVerent lamellar arteries are lacking in lampreys (Figure 2.9A). Blood vessels
in the secondary lamellae have a sheetlike arrangement that is bounded by a
large diameter marginal vessel (about twice that of the lamellar sheet thick-
ness) and an axial plate that is shared by opposing secondary lamellae
(Figure 2.9B). Arteriovenous anastomoses that are protected by microvilli,
as in hagfishes, provide extensive connections between the aVerent filament
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                             83


                       Anterior                       re


          efa                                                              pca




                                                             Inner ridge
                     Efferent filament a.

                                                             Smooth muscle
                      Marginal channel

                                                              Respiratory region
                   Axial blood lacunae                        (Secondary lamella)

                Lamellar blood lacunae

                      Canvernous body
                                                             Osmoregulatory region
                                                               l... Chloride cells:

                    Afferent filament a.
                     Filament vein
                                                                 Adipose tissue

                                            Peribranchial      Smooth muscle
                                              venous sinus

Fig. 2.9. Circulatory pattern for the lamprey gill. (A) Major vessels to and from the gill (Av,
ventral aorta; aba and eba, aVerent and eVerent branchial arteries; afa and efa, aVerent and
eVerent filament arteries; aca and pca, anterior and posterior collecting artery; ibp, internal
branchiopore; re, oesophageal branch; Ad, dorsal aorta; and arrows, direction of blood flow)
and (B) major vessels to and from a secondary lamella. [Taken from Nakao and Uchinomiya
84                                                        ANTHONY P. FARRELL

artery and the gill sinuses. Sinus blood of lampreys has a low hematocrit.
There is rich innervation along the eVerent branchial artery (Figure 2.9B).
    Lampreys have four unusual circulatory features. One feature is the lack
of a renal portal system, already noted for hagfishes, such that the kidney
only has an arterial supply. Intestinal ‘‘vascular couples’’ are another unusu-
al circulatory arrangement (detailed by Baxter, 1958). These are composite
vessels made up of an artery contained within a vein. In this case, each couple
involves an intestinal artery vessel from the dorsal aorta and intestinal vein
returning to the posterior cardinal vein. The arrangement arises developmen-
tally by an invasion of a vein into the outer connective tissue of the artery.
Their functional significance is unknown (Baxter, 1958). The large (600 mm)
vascular cushions (sphincters) in the dorsal aorta associated with the origins
of the segmental parietal arteries are a third unusual feature. Again, they may
induce turbulent blood flow to ensure that red blood cells leave the dorsal
aorta during sluggish aortic blood flow. Baxter (1958) emphatically rejected
the presence in lampreys of either valves in the wall of the dorsal aorta or
lymphatics. The fourth unusual feature is the developmental loss of the left
ductus Cuvier in adult lampreys (Fange, 1972).

3.2.3. Circulatory Dynamics
    In vivo heart rate is 33–50 minÀ1 at 16  C in adult Lampetra fluviatilis
(Hardisty, 1979) and 23–37 minÀ1 at 15  C in Geotria australis (Macey et al.,
1991). Similar heart rates at 18–20  C are apparent from traces presented for
isolated L. fluviatilis hearts (Augustinsson et al., 1956; Falck et al., 1966). A
heart rate of 8 minÀ1 at 11  C for L. planeri ammocoetes (Lignon, 1979)
seems unusually low. For Geotria, the Q10 is 1.2 between 5 and 15  C and
2.2 between 15 and 25  C (Macey et al., 1991). For Entosphenus tridentatus,
heart rate increased from 25 to 45 minÀ1 between 4 and 20  C, and then
jumped to between 80 and 130 minÀ1 from 20 to 25  C (Johansen et al.,
1973), perhaps signifying arrhythmias at these unusually high temperatures
for lampreys.
    Lampreys possess two cardioaccelerator mechanisms and therefore are
unusual among vertebrates. Neural regulation of the myogenic lamprey heart
has been detailed by Augustinsson et al. (1956) and Falck et al. (1966), who
conclusively confirmed Carlson’s (1906) unusual observation that stimula-
tion of the vagal nerve accelerated heart rate. The findings that acetylcholine,
nicotine, and even cigarette smoke cause cardioacceleration led to the con-
clusion that the cardiac receptors were nicotinic. These researchers (see also
Lignon, 1979) also characterized a positive inotropic and chronotropic eVect
through the stimulation of b‐adrenergic receptors.
    Adrenergic stimulation likely involves the paracrine action of the cardiac
chromaYn tissue. In isolated lamprey hearts, b‐adrenergic antagonists
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                 85

and reserpine treatment (to deplete catecholamine stores) have negative
inotropic and chronotropic eVects and subsequently enhance the stimul-
atory eVects of b‐adrenergic agonists. Extracellular calcium, especially
with acidosis, also enhances the positive inotropic eVect of catechola-
mines, suggesting b‐adrenergic modulation of intracellular calcium availa-
bility, similar to the situation in teleosts (Shiels et al., 2002; Vornanen et al.,
2002). Perfused ammocoete hearts show an overflow of adrenaline into the
perfusate (Lignon, 1979). While the mechanism behind catecholamine
release from the lamprey heart has not been adequately studied, it is not a
result of vagal stimulation (Augustinsson et al., 1956). Humoral chrono-
tropic responses are also possible in lampreys because circulating catechol-
amine levels rise during stress (Dashow et al., 1982), and both adrenaline
and noradrenaline increase in vivo heart rate to 47 minÀ1 in Geotria
(Macey et al., 1984). In the absence of cardiac vagal inhibitory innerv-
ation, slowing of the lamprey heart must come about by the withdrawal
of either the excitatory vagal tone or the excitatory paracrine/humoral
adrenergic tone.
    The physiological importance of stretch‐induced tachycardia in lampreys
needs to be revisited. Although Jensen (1969) reported tachycardia
(22–44 minÀ1) in Petromyzon marinus, an excessive increase in ventricular
filling pressure (up to 2.1 kPa) makes the physiological significance of this
finding questionable. Johansen et al. (1973) reported subambient central
venous pressures (À0.4 to À0.1 kPa) for E. tridentatus. These data indicate
vis‐a‐fronte suction filling is important as a result of the rigid pericardium in
lampreys (Satchell, 1991).
    Information on cardiac performance and blood pressures in lampreys is
conspicuously absent from general reviews (Bushnell et al., 1992; Farrell
and Jones, 1992). Instead, indirect information points to a more powerful
heart than hagfishes. The Fick estimate of routine Q was 32 ml minÀ1 kgÀ1
for E. tridentatus (Johansen et al., 1973). Geotria has a similar routine oxygen
uptake as E. tridentatus (Macey et al., 1991) and so Q may be similar.
Routine Q in lampreys is potentially about the same as maximum Q in
    Mean dorsal aortic blood pressure is between 2.5 and 4.4 kPa in
E. tridentatus, with pulse pressure usually less than 1.2 kPa (Johansen
et al., 1973). This means that the lamprey heart can generate a central arterial
blood pressure two to four times greater than the Myxine heart, but it uses
apparently two to six times more cardiac mass. As a result, performance of
the lamprey heart is superior to hagfishes likely because of a larger ventricu-
lar mass. Potentially, the mass‐specific myocardial power output is much
lower in lampreys than in teleosts, but accurate measurements are needed to
test this prediction.
86                                                        ANTHONY P. FARRELL

3.2.4. Circulatory Control
    Recovery from stress in Geotria decreases heart rate from 64 to 20 minÀ1
over 4 h. Since this decline in heart rate is slower and smaller than the fivefold
decrease in oxygen uptake (145‐ to 25‐ml O2 hÀ1 kgÀ1; Macey et al., 1991),
cardiac stroke volume and venous oxygen content are restored faster than
heart rate. Seasonal variation in heart rate in Geotria is in proportion with
oxygen uptake, but independent of temperature (Macey et al., 1991). The
basis for this regulation is unknown.
    Using isolated ventral aortic tissue, Evans and Harrie (2001) showed that
P. marinus has many of the paracrine vascular control mechanisms found in
Myxine. DiVerences for the lamprey included very strong contractions with
endothelin, strong relaxations with NO and prostacyclin, and acetylcholine
having no eVect.
    Studies with isolated dorsal aortic tissue have led to the suggestion that a
primordial feature of the earliest vertebrates is the hypoxic response of large
vessels being mediated by the gas H2S (Olson et al., 2006). Both H2S and
hypoxia produce similar responses (these are vessel‐specific vasodilations and
vasoconstrictions) in at least one representative species of each vertebrate
class and show essentially identical depolarizing eVects on the transmem-
brane potential. Other criteria used in reaching this conclusion include:
H2S being vasoactive at relevant physiological concentrations; blood vessels
producing H2S enzymatically; the competitive eVects of H2S and hypoxia;
inhibitors of H2S synthesis partially or completely blocking hypoxic
responses; and the addition of cysteine, the precursor for H2S production,
enhancing the hypoxic response (Olson et al., 2006).


4.1. Coelacanth

    Nothing is known of the circulatory physiology of coelocanths except
what can be deduced from anatomical studies. Circulatory anatomy for
Latimeria chalumnae is extensively detailed by Millot et al. (1978) and the
following account is taken from this treatise.
    The ventricle is composed largely of muscular trabeculae (Figure 2.10A)
without any remarkable features. The atrium is larger than the ventricle and
has ostial valves protecting the inflow and outflow. The sinus venous also has
trabecular structures (Figure 2.10A), which are unusual among fishes. The
heart is contained in a tough, fibrous pericardium and there are vestiges of
2.    CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                       87

                   A                                                B 1 2 3 4

                                                     2 cm
      per.                                                                                  valv.c.a

                                   2 cm


                                              can.C.g.                                       can.C.dr.
                                  v.pulm.                                                    v.pulm.

Fig. 2.10. The coelacanth heart and its major inflow and outflow vessels. [, aVerent
branchial arteries; at, atrium; c.a., conus arteriosus and its valves (valv.c.a.); can.C.g., left ductus
Cuvier; can.C.dr., right ductus Cuvier; cav.sin.v., sinus venosus lumen; cav.vt, ventricular
lumen;, atrioventricular ostium;, sinoatrial ostium; par.vt., retracted ventricle;
per., pericardium; sin.v., sinus venosus; trc.a., truncus arteriosus;, sinoatrial valve;
V.C., vena cava; v.hep.g., left hepatic vein; v.jug.i., internal jugular vein; v.pulm., pulmonary
vein;, subintestinal vein; v.s.oes., oesophageal vein; and vt., ventricle]. [Taken from Millot
et al. (1978).]

the pericardioperitoneal canal found in elasmobranchs. Both features
raise the possibility of vis‐a‐fronte cardiac filling. A 35‐kg coelacanth is
reported to have a 20‐g heart, which converts to a cardiac index of 0.57%.
What proportion of this large relative cardiac mass is ventricle is unclear.
    The cardiac outflow tract is unusually long (Figure 2.10B), consisting of a
posterior conus arteriosus and an anterior bulbus arteriosus, each of similar
length. Two sagittal valves protect the opening of the ventricle and the conus
contains four longitudinal rows of five to six valves that increase in size
88                                                        ANTHONY P. FARRELL

anteriorly. The short ventral aorta fans out into three pairs of aVerent
branchial arteries, with the most posterior pair branching further to supply
the third and fourth branchial arches. In addition, there is a vestige of an
aVerent hyoidean artery. The eVerent branchial arteries unite onto the ante-
rior dorsal aorta and its first major posterior branch is the coeliomesenteric
    Two lateral coronary arteries supply the ventricle and conus arteriosus.
They have a caudal origin from the subclavian arteries and reach the heart
laterally in association with the ductus Cuvier (Figure 2.11A). This caudal
coronary arrangement is similar to that found in skates, rays, and some
Chondrosteans. Large coronary veins drain into the sinus venosus. Despite
a well‐developed coronary circulation, a thick compact myocardium is not
evident in any cardiac illustration, which raises the possibility that coronary
vessels of Latimeria largely invest ventricular trabeculae.
    The venous arrangement in Latimeria is illustrated in Figure 2.11B.
Notably, a vestigial pulmonary vein can be traced from the fat‐filled lung,
but there is no vestigial pulmonary artery from the fourth eVerent branchial
artery, as in the dipnoans. The vestigial lung (air bladder) is instead supplied
by several small arterial branches from the dorsal aorta.

4.2. Dipnoi (Lungfishes)

    The lungfish circulatory system has commanded the most attention of all
primitive fishes because lungfish are the oldest of extant air‐breathing fish
and are at the root of tetrapod evolution. In fact, many evolutionary parallels
have been drawn between the lungfish and urodele circulatory systems.
However, despite a steady interest in the macrocirculation after the early
anatomical studies (Muller, 1842; Gunther, 1871) and the pioneer physiolog-
ical studies (summarized in Johansen, 1970), few physiological studies have
followed the excellent review of the dipnoan circulatory system by Burggren
and Johansen (1986). The circulatory adaptations of dipnoans are placed in
the broader context of all air‐breathing fishes by Graham (1997) and Olson
    The hearts and circulatory patterns of lungfishes diVer from the typical
piscine arrangements in a number of important respects. Here, the emphasis
is on how the basic fish circulatory system was modified for air breathing. As
noted already, Neoceradatus breathes air only under hypoxic conditions,
while Lepidosiren and Protopterus species are obligate air‐breathers. Also,
Lepidosiren and Protopterus can live for long periods out of water (estiva-
tion). Along with the greater dependence on air breathing, Lepidosiren and
Protopterus show greater deviation from the normal piscine circulatory
design and a progression toward a more divided circulatory system than
      A          2 cm

            sin.v.            can.C.dr. v.jug.i.dr. at.                 c.a.


                        a.cor.dr. v.cor.dr.

      v.caud.                                                v.gén.dr.            



                                                                                                                                                       v.hep.g. sin.v. v.jug.i.
                                                   V.C.                                                             v.hep.dr.
                                                                           v.gastr. trc.p.hep.

Fig. 2.11. (A) Lateral view of the right caudal coronary supply to the coelacanth heart. (B) The major posterior veins of the coelacanth. [,
aVerent branchial arteries; a.cor.dr., right coronary artery; at., atrium; c.a., conus arteriosus and its valves (valv.c.a.); can.C.g., left ductus Cuvier; can.
C.dr., right ductus Cuvier; cav.sin.v., sinus venosus lumen; cav.vt., ventricular lumen;, atrioventricular ostium;, sinoatrial ostium; par.
vt., retracted ventricle; per., pericardium; sin.v., sinus venosus; trc.a., truncus arteriosus;, sinoatrial valve; v.caud., caudal vein; v. interren.,
interregnal vein; V.C., vena cava; v.cor.dr., right coronary vein; v.gast., gastric vein; v.hep.g., left hepatic vein; v.jug.i., internal jugular vein; v.pulm.,
pulmonary vein;, subintestinal vein; v.s.oes., oesophageal vein; and vt., ventricle]. [Taken from Millot et al. (1978).]
90                                                                            ANTHONY P. FARRELL





       A                                                                                  A


                 L                            V

                                                      1 23 4 5




Fig. 2.12. The lungfish heart. (A) A scanning electron microscopic image of the inside of the
sinus venosus (SV) of the Protopterus heart to illustrate the pulmonary vessel is embedded in its
wall (arrow) and the left (L) atrium is larger than right (R) atrium (A). [Taken from Icardo et al.
(2005a).] (B) A scanning electron microscopic image of a sagittal section through the ventricle
(V) of the Protopterus heart to illustrate the left (L) and right (R) sides to the ventricle that are
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                  91

Neoceratodus. To reiterate and as detailed below, lungfishes show four
circulatory changes critical for the transition to obligate air breathing with
a lung: They do not mix pulmonary and systemic venous return in the sinus
venosus because of a separate return of pulmonary blood to the atrium;
pulmonary and systemic flows remain largely separated during their passage
through the atrium and ventricle; deoxygenated arterial blood flow can flow
directly from the heart to the lung; and Q can bypass the gill respiratory
surface. Radiographic imaging of blood flow streams (Johansen and Hol,
1968; Szidon et al., 1969) and measurements of gas tensions (Johansen et al.,
1968) have provided the key evidence for this functional separation between
the pulmonary and systemic circulations.

4.2.1. Cardiac Anatomy
    In Neoceratodus, the single pulmonary vein joins the sinus venosus
(Foxon, 1950). This union is unlike other air‐breathing fishes, where the
venous drainage from the air‐breathing organ is less central. Protopterus
and Lepidosiren have an even more unusual cardiac modification since they
are the only fishes known to separate the pulmonary and systemic venous
return to the heart. After the paired pulmonary veins unite, the single
pulmonary vein becomes embedded in the wall of the sinus venosus of
Protopterus and Lepidosiren (Figure 2.12A). The vein then runs longitudi-
nally toward the atrioventricular plug and fuses with the membranous pul-
monalis fold of the atrium. Meticulous examination of the inflow tract of
Protopterus led Icardo et al. (2005a) to conclude that the pulmonary vein
terminates in the sinus venosus so close to the sinoatrial junction that
oxygenated blood from the pulmonary vein empties directly into the left
lobe of the atrium. Thus, they rejected the previously held idea that the
pulmonary vein actually terminated in the atrium (unlike all other fishes
where veins do not terminate beyond the sinus venosus). Instead, the pulmo-
nary vein in Protopterus has a functional termination in the atrium and an
anatomical termination in the sinus venosus.
    All three genera have a single atrium that is partially divided externally
and internally into a larger right and smaller left side, and that is heavily
trabeculated. This appearance caused Gunther (1871) to incorrectly refer to
the heart as trilocular (Figure 2.12B). Lepidosiren shows the greatest degree

separated by the vertical septum (VS) and the atrioventricular plug (P), and through the outflow
tract of the heart (C) with its intralumenal spiral and bulbar folds (1), which fuse at the asterisk.
[Taken from Icardo et al. (2005b).] (C) A schematic diagram of the outflow tract in Lepidosiren,
which has a reduced spiral valve (sp.v) compared with Protopterus. (The aVerent brachial arteries
are numbered;, dorsal channel;, ventral channel; sw, swelling in the spiral valve; pr.v,
proximal semilunar valves; Vn, ventral; and Dr, dorsal). [Taken from Satchell (1976).]
92                                                          ANTHONY P. FARRELL

of atrial subdivision and Neoceratodus the least. The left lobe of the atrium
represents the pulmonary channel into which the pulmonary vein empties
and is formed in part by the pulmonary fold in the wall of the atrium and the
atrioventricular plug. A sinoatrial valve guards the entrance into the right
atrium, but the pulmonary vein is apparently not valved at its termination.
    The atrioventricular channel has an atrioventricular plug (cushion),
formed from a cartilaginous core surrounded by connective tissue (see
Icardo et al., 2005a for details). The atrioventricular plug moves into and
out of the atrioventricular orifice during cardiac contraction and relaxation,
and while serving as a one‐way valve during ventricular systole, it allows the
right and left bloodstreams to remain separated as they leave the atrium. This
type of separation would be diYcult with the typical ostial valve found in
other fishes.
    The rounded lungfish ventricle is highly trabeculated and compact myo-
cardium is not evident (Figure 2.12B). Again, in all three genera, the ventricle
is divided into right and left sides by a large muscular ridge, the vertical
septum, which extends from the apex between the dorsal and ventral walls
(Figure 2.12B). The vertical septum is developed to the greatest extent in
Lepidosiren and least in Neoceratodus. Collectively, the arrangement of the
pulmonary vein, the pulmonary fold, the atrioventricular plug, and the
vertical septum results in oxygenated blood returning from the lung being
preferentially directed through the left side of the atrium to the left side of the
ventricle. Deoxygenated systemic venous return is preferentially directed
through the larger right side of the atrium to the right side of the ventricle.
To what degree the highly trabeculated structure of the cardiac chambers
restricts intracardiac mixing by immobilizing blood during diastole [as sug-
gested by Shelton (1976) for amphibian hearts] is unclear.
    The ability of lungfish to direct deoxygenated blood flow from the heart
to the lung depends on specializations in the cardiac outflow tract and in the
circulation (see Section 4.2.2). The long outflow tract from the lungfish heart,
described by Icardo et al. (2005b), is perhaps the most complex of all fishes
(Figure 2.12B and C). In Protopterus and Lepidosiren, it has three sections, a
270 rotation, conal valves, and incomplete partitioning of its lumen with a
spiral valve. The net eVect of these anatomical adaptations is that the
separation of oxygenated and deoxygenated blood is maintained to a large
degree within the outflow tract. As a result, deoxygenated systemic blood is
preferentially directed to the two posterior gill arches, while the oxygenated
pulmonary blood is preferentially directed to the two anterior gill arches
(Johansen and Hol, 1968; Johansen et al., 1968; Szidon et al., 1969).
    The proximal portion of the outflow tract displays several rows of valves
on the dorsal and ventral walls. It possesses a thick layer of well‐vascularized,
circumferentially arranged, and compact myocardium and should be termed
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                93

conus arteriosus (Icardo et al., 2005b). A large fold of loose connective tissue,
the spiral fold, arises on the inside left wall and runs most of the length of the
outflow tract in Protopterus and Lepidosiren. The spiral fold is greatly
reduced in Neoceratodus, especially in the distal region (Figure 2.12C). In
Protopterus and Lepidosiren, a smaller bulbar fold arises in the middle
portion of the outflow tract, but it does not fuse with the spiral fold until
the distal portion of the outflow tract. The medial and distal portions of the
outflow vessel contain predominantly elastic and connective tissue and
should be termed bulbus arteriosus (Icardo et al., 2005b). A cephalad coro-
nary artery supply reaches to the compact cardiac muscle of the conus
arteriosus and, to a limited extent, the ventricle. The possibility of coronary
vessels reaching the large ventricular trabeculae has not been explored.
    Innervation of the lungfish heart has been described in Abrahamsson
et al. (1979a,b), Nilsson (1983), and Axelsson et al. (1989). Vagal innerv-
ation reaches the sinus venosus and probably the atrium (see Section 4.2.3).
Spinal autonomic innervation is absent. Instead, a putative adrenergic con-
trol is provided by the large masses of chromaYn cells that line the inner
wall of the atrium (but not the ventricle), the left cardinal vein, and curiously
the walls of the intercostal arteries near the dorsal aorta (Abrahamsson
et al., 1979a). Noradrenaline predominates in the atrium, while adrenaline
predominates in the other locations, with the exception of the plasma
(Table 2.1).
4.2.2. Circulatory Patterns
    The circulatory patterns of the gills and lungs are summarized in
Burggren and Johansen (1986). Early macroscopic studies include Boas
(1880), Spencer (1892), Lankester (1878), Robertson (1913), and Bugge
(1961). The functional arrangement of the gill arteries is summarized in
Johansen et al. (1968), Szidon et al. (1969), Gannon et al., (1983), and
Fishman et al. (1985). The circulatory pattern of the gill diVers considerably
among the three genera, and Protopterus shows the greatest divergence from
the basic piscine pattern (Burggren and Johansen, 1986). The gill
microcirculation of Protopterus is detailed in Laurent et al. (1978).
    The ventral aorta is almost nonexistent because four aVerent branchial
arteries originate immediately outside the pericardium, similar to elasmo-
branchs. In Neoceratodus, the first and second (anterior) gill arches are
filament‐bearing holobranchs and their eVerent branchial arteries directly
join the dorsal aorta. However, the two anterior gill arches in Protopterus
and Lepidosiren are not true holobranchs because the arches lack gill
filaments and the aVerent branchial artery passes directly to the dorsal
aorta (Figure 2.13A). Thus, oxygenated blood from the ventral root of the
bulbus arteriosus reaches to the dorsal aorta and the systemic circulation
94                                                                                ANTHONY P. FARRELL



                                                                         Ductus    Aorta

B                                       da          p.a



     3a   4a                                       c


Fig. 2.13. The circulation pattern in the obligate air‐breathing lungfish Protopterus. (A) A
vascular cast of the gills showing that the anterior two gills arches are devoid of gill arches and
each contains a single vessel that connects directly to the dorsal aorta (A), while the posterior two
gill arches possess a coarse arrangement of gill filaments and their eVerent flow is directed more
to the pulmonary artery (P). [Taken from Jesse et al. (1967).] (B) A schematic diagram of the
vascular arrangement of the gills and the lung. (Aortic arches are numbered; c, conus; da, dorsal
aorta; and p.a and p.v, pulmonary artery and vein.) [Taken from Satchell (1976).] (C) A close of
the connecting vessel between the third and fourth gill arch and the origin of the pulmonary
artery. [Taken from Szidon et al. (1969).]

without going through a gill respiratory surface, which is not the case for
   The gill circulation of Neoceratodus is distinguished by the absence of a
hemibranch on the hyoid arch. However, in Protopterus and Lepidosiren, the
aVerent branch from the first gill arch supplies an additional hemibranch
located on the hyoid arch (Figure 2.13B). This is analogous to the pseudo-
branch because the eVerent vessels of the hyoid arch go on to form the main
vessel supplying the brain, the internal carotid artery.
   All three dipnoan genera have primary gill filaments and secondary
lamellae on their third and fourth (posterior) gill arches. The gill filaments
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                               95

in Lepidosiren are sparse and coarse (Laurent et al., 1978). The aVerent
branchial artery of the fourth gill arch also supplies blood to a hemibranch
located on a fifth gill arch (Figure 2.13B).
    The posterior gill arches are supplied predominantly with deoxygenated
systemic venous blood from the dorsal channel of the bulbus arteriosus.
Thus, combined with a complex arrangement of the posterior eVerent bran-
chial arteries (Figure 2.13B), dipnoans can preferentially direct deoxygenated
blood to the lung and be highly successful air‐breathers. The fourth eVerent
branchial artery is contiguous with the pulmonary artery (Szidon et al.,
1969), while the third eVerent branchial artery goes more directly to the
dorsal aorta. However, the third and fourth eVerent branchial arteries have
a short connecting vessel, the ductus (ductus arteriosus; Figure 2.13C). The
ductus is highly invested with vascular smooth muscle (up to 30 layers as
compared with just 5–7 in the dorsal aorta) and clearly plays a regulatory role
in blood flow distribution between the systemic and pulmonary circuits (see
below). The lung is perfused by two pulmonary arteries that originate at the
base of the fourth eVerent branchial arteries close to the dorsal aorta. The left
pulmonary artery perfuses the lung from a ventral aspect and becomes
branched partway down the lung, while the right pulmonary artery perfuses
the lungs from the dorsal side. Neoceratodus has one lung, and Protopterus
and Lepidosiren have two lungs.
    The vascular pathways in the posterior gill filaments of Protopterus are
highly modified. Three arterio‐arterial vascular shunt pathways allow
(Laurent et al., 1978) blood to bypass the normal lamellar respiratory area.
These vascular adaptations would be important either in severely hypoxic
water, when oxygen could be lost across to the water from blood across the
gill secondary lamellae, or in estivating lungfish when secondary lamellae
might collapse during air exposure. A short shunt vessel connecting the base
of the aVerent filament artery to the base of the eVerent filament artery
provides one bypass route (Figure 2.14A). This shunt has not been reported
for other fishes. The aVerent filament artery that continues around the tip of
the filament provides a second bypass route, although it does taper and lose
vascular smooth muscle, (Figure 2.14A). The vessels of the secondary lamel-
lae resemble second order arterioles, designed more for throughput rather
than for exchange, as is the high‐density vascular sheet typically found in
teleost secondary lamellae (Farrell et al., 1980). They also have distinct
origins and terminations on the aVerent and eVerent filament arteries, pos-
sess vascular smooth muscle, are lined with endothelium, and are covered by
multiple layers of epithelium (Figure 2.14B and C) (Laurent et al., 1978). All
these features are normally absent in secondary lamellae. Angiogenesis in the
gill lamellae of Protopterus must follow a unique trajectory among fishes
and further work is needed to ascertain the role that pillar cells might play
96                                                                           ANTHONY P. FARRELL





                                   na                                   n

 B                             C



              IV V

  2 mm                                                                                     100

Fig. 2.14. Details of the vascular arrangement in the posterior gill arches of Protopterus. (A) A
schematic of vascular arrangement of a gill filament illustrating the shunt vessels (S) between the
aVerent (aba) and eVerent (eba) branchial arteries, as well as the primary aVerent and eVerent
primary (aa1, ea1) and the secondary (aa2, ea2) aVerent and eVerent arteries (bv, branchial vein;
cg, cartilage; m, muscle; na, nutritive artery; v, vein; and arrows, direction of blood flow) [inset
shows detail of the tip of the filament with the connection between the aVerent (aa) and
eVerent (ea) filament arteries (a‐a, arterioarterial capillaries and a‐v, arteriovenous system).
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                              97

in lamellar vessel development. Like other fishes, Protopterus gills possess
a well‐developed arteriovenous system, with vessels originating from the
eVerent filament artery (Laurent et al., 1978). Gannon et al., (1983)
concluded that the arterial system of the gift filaments of Neoceratodus was
similar to that of the elasmobranch pattern and lacked the specializations and
shunt vessels evident in Protopterus. In addition they observed that both the
lung and the secondary lamellar contained highly vascularized capillary
4.2.3. Circulatory Dynamics
    Dipnoans are prime candidates for high‐quality studies of cardiovascular
physiology. Studies of circulatory physiology in lungfish are very limited,
unlike studies of the anatomy and respiratory properties of the circulatory
system. Also, the classical studies by Johansen et al. (1968) on unanesthetized
lungfishes and by Szidon et al. (1969) on anesthetized Protopterus have
    For Protopterus at 18  C, heart rate, stroke volume, and Q are reported as
15 minÀ1, 1.3 ml kgÀ1, and 20 ml minÀ1 kgÀ1, respectively (Johansen et al.,
1968). For Protopterus at 25  C, dorsal aortic flow is reported as 34 ml minÀ1
(range 20–48 ml minÀ1; body mass was not given and varied from 0.3 to 4 kg)
(Szidon et al., 1969). Cardiac output increased with air breathing but values
were not given. Using the Fick Principle, Q in Neoceratodus apparently
increased sevenfold following 45 min of hypoxia (Johansen et al., 1967),
which suggests Q had been extremely low during hypoxia. In contrast, direct
measurement of Q with a Doppler flow probe failed to find such large
changes in Q during hypoxia with Neoceratodus (Fritsche et al., 1993).
    Compared with teleosts, mean arterial blood pressures seem to be lower
for all three genera of dipnoans. Johansen et al. (1968) suggested, without
convincing data and great variability in the individual blood pressure tra-
cings (1.9–2.3 kPa), that arterial blood pressures in Neoceratodus were higher
than in the other two genera. While diastolic and systolic blood pressures
in the ventral aorta were reported as 2.6 and 4.2 kPa, respectively (Fig. 2 in
Johansen et al., 1968), peak ventricular pressure was only 2.3 kPa in another
figure (their Fig. 3). Also, in a later figure (their Fig. 19), the highest ventral
aortic blood pressure was clearly associated with the heart being unable
to pump blood through the gills. A mean ventral aortic blood pressure of
2.7 kPa (pulse pressure ¼ 1.1 kPa) may be normal for Neoceratodus.

(B) A vascular cast of the coarse arrangement of the filaments and the eVerent branchial arteries
(eba) on the IV, V, and VI aortic arches. A close‐up of a vascular cast near the tip of a gill
filament illustrating the large diameter and direct vascular connections between the aVerent (aa1)
and eVerent (ea1) filament arteries. [Taken from Laurent et al. (1978).]
98                                                          ANTHONY P. FARRELL

For anesthetized Protopterus, Szidon et al. (1969) showed a representative
peak ventricular pressure of 2.0 kPa (Figure 2.15) and suggested that peak
ventricular pressure can reach 2.5 kPa, but then stated that the postbranchial
blood pressure was 3.0 kPa in an unanesthetized specimen. Assuming Q is
20–35 ml minÀ1 kgÀ1, mean ventral aortic pressure is 3.0 kPa and relative
ventricular mass is 1g kgÀ1 body mass, routine myocardial power output of
lungfish can be estimated as 1.0 to 1.8 mW gÀ1 ventricular mass.
     Heart rates are comparatively low in dipnoans: 29 minÀ1 at 27  C in
Lepidosiren (Axelsson et al., 1989), 38–41 minÀ1 at 25  C in Lepidosiren
(Sanchez et al., 2001), 47 minÀ1 in Protopterus at 25  C (Perry et al., 2005),
32 minÀ1 in Protopterus (from Fig. 15 in Johansen et al., 1968), 20–30 minÀ1
and 30 minÀ1, respectively, in anesthetized and unanesthetized Protopterus at
25  C (Szidon et al., 1969), 22 minÀ1 at 20  C in Neoceratodus (Fritsche et al.,
1993), and 24 minÀ1 in Neoceratodus (from Fig. 14 in Johansen et al., 1968).
It may be that a low heart rate and a gentle cardiac contraction are needed to
ensure laminar, rather than turbulent, flow in the cardiac chambers, and
thereby aid the separation of oxygenated and deoxygenated bloodstreams.
     The thick pericardium suggests that vis‐a‐fronte cardiac filling is important
in dipnoans. In fact, subambient cardiac filling pressures were reported for
all three genera by Johansen et al. (1968), who concluded that suctional
attraction by the heart was important for the systemic veins; however, the
pulmonary veins had above ambient blood pressures. Szidon et al. (1969)
similarly concluded that vis‐a‐fronte filling was important based primarily on
subambient pressure measurements within the pericardial space of Protopterus,
but in their figures, venous pressures were above ambient (Figure 2.15; these
experiments referenced the pressure transducer to the center of the lungfish out
of water, leaving a large margin for a referencing error).
     Beyond a passive capacity to separate bloodstreams, the conus arteriosus
may play an active role in the circulation of dipnoans (Satchell, 1991). Distinct
rhythmic conal contractions were present in blood pressure traces for Neocer-
atodus (Johansen et al., 1968), but not for either anesthetized or unanesthetized
Protopterus (Johansen et al., 1968; Szidon et al., 1969). Perhaps the contractile
function was lost as dipnoans evolved a greater dependency on air breathing.
Even so, secondary pressure oscillations became apparent when blood pressure
in the aVerent branchial artery of Protopterus was excessively elevated
following acetylcholine injection (Johansen et al., 1968).
     Measurements of dorsal aortic blood pressures are variable. EVerent bran-
chial blood pressures were reported as 1.7–2.0 kPa in Protopterus (Perry et al.,
2005), 2.6 kPa in Lepidosiren (Axelsson et al., 1989), and 2.9 (Szidon
et al., 1969) and 2.1–2.3 kPa in Neoceratodus (Johansen et al., 1968; Fritsche
et al., 1993). Collectively, these data point to the latter authors’ suggestion that
about one‐third of the arterial blood pressure can be lost across the dipnoan
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                  99

                                                 O           P      QRS           O          P



                                     Ventricular ejection   Isovolumic ventricular contraction




                5 Vena cava
              B   Sinus venosus

                                                            Atrial systole

                                                     Sinus venosus




Fig. 2.15. ECG and superimposed traces from the cardiac chambers and the pericardial cavity of
lightly anesthetized Protopterus. [Taken from Szidon et al. (1969).]
100                                                         ANTHONY P. FARRELL

gills, as in other fishes. However, when nearly simultaneous measurements of
blood pressure were made in the aVerent branchial artery and the coeliac artery
of Protopterus, there was a very small transbranchial vascular resistance (i.e., a
similar diastolic blood pressure and a systolic diVerence of just 0.23 kPa), which
then disappeared when Protopterus was exposed to air (Johansen et al., 1968).
This result was taken to indicate that the vascular resistance of the proximal gill
arches is very small, as expected given the lack of primary filaments. The
implication of this finding is profound because the eVerent branchial blood
pressure of the anterior gill arches must be greater than that in the posterior gill
arches, and yet both are connected to the dorsal aorta. Thus, blood cannot flow
from the posterior gill arches into the dorsal aorta unless there is a means to
either increase vascular resistance in the anterior arches (vasoconstriction) or
decrease vascular resistance in the posterior arches (perhaps opening of gill
shunt vessels). Conversely, blood from the posterior arches will tend to flow
into the pulmonary circulation rather than the dorsal aorta. In addition, unless
the ductus is closed, blood will easily flow from the dorsal aorta into the
pulmonary arteries retrograde via the posterior eVerent branchial arteries.
Clearly, vasoactive control systems are important for eVective perfusion of
the lung since the anatomical arrangement in the gills is not entirely adequate
by itself. As shown in the next section, some aspects of this control have been
4.2.4. Circulatory Control
    The Lepidosiren heart has a routine cholinergic inhibitory tone (Axelsson
et al., 1989). Also, resting heart rate decreased from 32 to 25 minÀ1 when
propranolol was injected, suggesting a b‐adrenergic tonus (Axelsson et al.,
1989). In contrast, Fritsche et al. (1993) found neither vagal nor adrenergic
tone on resting heart rate in Neoceratodus. An abrupt, fright‐induced brady-
cardia was abolished with atropine treatment in both Lepidosiren and Neo-
ceratodus, suggesting a common vagal cardioinhibitory mechanism in
dipnoans (Axelsson et al., 1989; Fritsche et al., 1993).
    Adrenergic cardiac control is possible through humoral and paracrine
mechanisms, but adrenergic innervation is lacking (Abrahamsson et al.,
1979a). Paced ventricular strips from Protopterus show a positive inotropic
response to adrenaline but no response to cholinergic stimulation with carba-
chol (Abrahamsson et al., 1979b). Conversely, paced atrial strips show a
strong negative inotropic eVect with cholinergic stimulation and no eVect
with adrenaline. However, injections of adrenaline into Protopterus and
Neoceratodus have failed to change heart rate (Johansen and Reite, 1967;
Fritsche et al., 1993), and so experiments with adrenergic antagonists are
needed to properly describe catecholamine control of cardiac function in vivo.
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                              101

    Despite their functional vagal cardioinhibitory control system, Protopterus
lack the typical piscine bradycardia response to aquatic hypoxia; heart rate
remains at 46–48 minÀ1 at a water oxygen tension of 2.6 kPa (Perry et al., 2005).
Absence of bradycardia in response to aquatic hypoxia (3–4 kPa) was noted
earlier for Lepidosiren (Sanchez et al., 2001) and Neoceratodus (Fritsche et al.,
1993). Also, aquatic hypoxia did not alter oxygen uptake (Perry et al., 2005)
and plasma catecholamine levels (Table 2.1). Likewise, aerial hypoxia (4.6 kPa)
had no eVect on heart and ventilation rates, but depressed oxygen uptake by
fivefold (Perry et al., 2005), suggesting some combination of a decrease in
cardiac stroke volume and tissue oxygen extraction.
    Lungfish increase pulmonary blood flow after an air breath to ensure
that lung ventilation and perfusion are closely matched. However, quantify-
ing such changes in vivo has proven diYcult because of limited vascular
access to implant flow probes. Johansen et al. (1968) reported that spontane-
ous air breathing was consistently associated with increased pulmonary
flow and, sometimes, increased heart rate and Q. Unfortunately, no numeri-
cal information was reported for this response, or for the stated tachycardia
and pressor eVects associated with artificial lung inflation, or for the stated
bradycardia and depressor eVects associated with lung deflation. Instead,
pulmonary flow was estimated to vary between 20% and 70% of Q. Other
reports suggest that the change in pulmonary blood flow with air breathing
ranges from as little as 50% to over fourfold. Szidon et al. (1969) reported a
mean flow in the left pulmonary artery of Protopterus of 18 ml minÀ1 that
varied more than fourfold from 7 to 33 ml minÀ1 (body mass was not given).
However, if we assume similar blood flow rates in the right and left pulmo-
nary arteries, such a response would mean that total pulmonary blood flow
could exceed their measurement of Q. When Lepidosiren were taking air
breaths every 4–12 min, pulmonary blood flow increased 50% (from 13 to
20 ml minÀ1 kgÀ1) and heart rate increased 10% (Axelsson et al., 1989), with
some of the tachycardia being preemptive. This increase in pulmonary flow
was blocked by atropine pretreatment, suggesting that a cholinergically
mediated vasodilation (possibly of the ductus) can regulate pulmonary
flow. In Neoceratodus, pulmonary flow increased with air breathing (without
aVecting heart rate) through a decrease in pulmonary vascular resistance and
a slight increase in resistance of, and decrease in flow in, the coeliacomesen-
teric circulation (Fritsche et al., 1993). Also, pulmonary blood flow was
estimated to increase 25% after 25 min of progressive hypoxia in Neocerato-
dus (Johansen et al., 1967). In Protopterus, pulmonary blood flow decreased
with swimming and aerial exposure to 5% carbon dioxide (again numerical
values were not presented; Johansen et al., 1968).
    Pulmonary blood flow must be under active control because the ratio of
systemic and pulmonary blood flow can change, that is a change in
102                                                       ANTHONY P. FARRELL

pulmonary blood flow goes beyond that associated with the change in Q. In
this regard, vasoactivity in the pulmonary artery and the well‐innervated
ductus becomes important. Fishman et al. (1985) proposed the following
general mechanism for cyclic control of pulmonary flow in Protopterus.
During submersion when the posterior gills are engaged in gas exchange
with the water, the gill shunt vessels are closed, the pulmonary artery is
constricted, and the ductus is relaxed. When Protopterus breathe air, the
pulmonary artery opens, the gill shunts open, and the ductus closes. This
hypothesis is supported by vascular casts from estivating Protopterus
showing a clear constriction site at the base of pulmonary artery and dilated
gill aVerent to eVerent shunt vessels (Laurent et al., 1978), as well as by
studies of putative vasoactivity control mechanisms with isolated and per-
fused vessels of Protopterus (Table 2.3). Even so some of these in vitro results
are contradictory. For example, while Johansen and Reite (1967) found that
both the ductus and pulmonary artery constricted with acetylcholine,
Fishman et al. (1985) found opposite responses of the two vessels to
    In vivo studies in Neoceratodus revealed cholinergic vasoconstriction of
the pulmonary artery and cholinergic relaxation of the ductus (perhaps
through NO release) (Fritsche et al., 1993). Acetylcholine increased pulmo-
nary blood flow, while atropine injection decreased pulmonary vascular
resistance. In the same study with Neoceratodus, adrenaline injection
increased pulmonary blood flow, dorsal aortic pressure, and coeliacomesen-
teric vascular resistance, and decreased coeliacomesenteric blood flow. Be-
cause sotalol injection increased vascular resistance in both the pulmonary
and coeliacomesenteric circulations, and phentolamine decreased dorsal aor-
tic blood pressure, at least three tonic mechanisms appear to influence
pulmonary blood flow in Neoceratodus: cholinergic vasodilation, a‐adrener-
gic vasoconstriction, and b‐adrenergic vasodilation.
    Despite these studies, a complete understanding of vasoactive control for
the ductus and pulmonary artery in dipnoans is lacking. Even less is known
concerning the control of the gill shunt vessels (e.g., how is blood flow
distributed among gill arches? What determines inter‐ and intrabranchial
shunt patterns?). A cholinergic control mechanism may be important be-
cause injection of acetylcholine in Protopterus increased branchial vascular
resistance to such a degree that Q and pulmonary blood flow stopped
(Johansen et al., 1968). The decrease in gill resistance with air exposure in
Protopterus (Johansen et al., 1968) contrasts with the large increases in gill
vascular resistance seen for water‐breathing teleosts during air exposure.
    Holmgren et al. (1994) examined the in vivo responses of Neoceratodus
to several neuropeptides (neurotensin, galanin, bombesin, substance P,
and cholecystokinin 8) and concluded that they followed the general
                                                                 Table 2.3
                Vasoactivity of the Isolated Ductus and the Proximal Pulmonary Artery of the African Lungfish Protopterus

                                             Ductus                                                    Pulmonary artery

                    Fishman et al., 1985          Johansen and Reite, 1967           Fishman et al., 1985         Johansen and Reite, 1967

Hypoxia                 Relaxes                       No response                       No response                  No response
Hyperoxia               Contracts                     NA                                NA                           NA
Hypercapnia             NA                            No response                       NA                           No response
Dopamine                Contracts (strong)            NA                                No response                  NA
Noradrenaline           Contracts (strong)            Relaxes (slow)                    No response                  Relaxes (slowly)
Adrenaline              NA                            Relaxes (slowly)                  NA                           Relaxes (slowly)
Isoproterenol           Relaxes (weak)                Relaxes (weak)                    No response                  Relaxes (weak)
Acetylcholine           Relaxes (weak)                Contracts (strong)                Contracts (strong)           Contracts (strong)
                          atropine blocks               atropine blocks                   atropine blocks              atropine blocks
PGE2                    Relaxes (strong)              NA                                NA                           NA
Histamine               NA                            Contracts (strong)                NA                           Contracts (strong)
Serotonin               NA                            Contracts (strong)                NA                           Contracts (strong)
                                                              p.a                                 B



                                      c                  6a

Fig. 2.16. Unusual features in ganoid circulatory systems. (A) A schematic diagram of the circulatory arrangement of the gills and lung (abl) of the air‐
breathing Polypterus. (Aortic arches are numbered; c, conus; p.a and p.v, pulmonary artery and vein). [Taken from Satchell (1976).] (B) A vascular cast
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                             105

vertebrate pattern of cardiovascular control by these peptides. In addition,
Abrahamsson et al. (1979b) found muscarinic and adrenergic constrictions
of the spleen in Protopterus.


    The fish orders comprising the Polypterids, gars, and bowfins have been
grouped together in early literature as the ganoids. Collectively, their primi-
tive status is reflected in the possession of a conus arteriosus [termed con-
tractile bulbus arteriosus by Wilder (1876)] that contains serial rows of
valves. Here, they are grouped because so little is known of their individual
circulatory physiology and because all three orders are facultative air‐
breathers (Smatresk and Cameron, 1982; Graham, 1997). Amia rarely
breathe air at 24  C, but at 30  C, and when exposed to aquatic hypoxia,
they take regular air breaths every 10 min or so (Randall et al., 1981). While
the gill structure of garfish is reduced (Potter, 1927), Amia have well‐
developed gills (Daxboeck et al., 1981; Olson, 1981).

5.1. Polypterids (Bichirs and Reedfish)
    Polypterus possess a number of anatomical features that have drawn atten-
tion. They have four gill arches, and the most posterior arch bears a single
hemibranch without a gill slit behind it (Figure 2.16A). While much debate has
been given to whether this last gill arch is a fifth gill arch with an accompanying
loss of the fourth gill arch or a reduced fourth gill arch, or (see Britz and
Johnson, 2003), recent literature favors the latter view, in part because the
fourth gill arch possesses its own aVerent and eVerent branchial arteries.
    A cephalad coronary artery from the subclavian artery supplies the
compact myocardium of the conus arteriosus and the ventricle. A thick‐
walled coronary vein drains into the ductus Cuvier (Budgett, 1901).
    Budgett (1901) has shown Polypterus with a similar eVerent branchial
artery arrangement as Amia (and lungfish). The fourth eVerent branchial artery

of a gill filament from the air‐breathing garfish Lepisosteus with most lamellae removed to
reveal channelized vessels in the lamella vessels. [Taken from Smatresk and Cameron (1982).]
(C) A vascular cast of a secondary lamella from the air‐breathing Amia to show the limited
degree of channelized lamellar vessels with the exception of the larger inner (IM) and outer (OM)
marginal vessels. [Taken from Olson (1981).] (D) A histological cross‐section through the
secondary lamellae of Amia to show the fusion of adjacent lamellae (S), which results in lamellar
vessels being buried in epithelium, as they are in the basal region of the lamellae near to the
filament (F). [Taken from Olson (1981).]
106                                                      ANTHONY P. FARRELL

passes directly to the pulmonary artery and is connected via a ductus to
the third eVerent branchial artery. However, the third eVerent branchial
artery joins the coeliac artery directly, rather than joining the dorsal aorta
(Figure 2.16A).
    The lungs are supplied by a pair of pulmonary arteries, with the left
pulmonary artery passing over the left ductus Cuvier en route to the lung.
Left and right posterior pulmonary veins drain the posterior two‐third of the
lungs. These vessels unite to form a common pulmonary vein, which then
empties into the large posterior vena cava (Kerr, 1910) that runs ventral to
the right lung between the posterior caudal vein and ductus Cuvier. The two
anterior pulmonary veins unite before entering the vena cava at the same
point (Magid, 1967). While beautifully describing the pattern of venous
system, Magid (1967) termed this large collecting vessel the hepatopulmon-
ary vein because it received numerous collateral venous branches from the
lung before receiving the hepatic veins caudal to the union of the common
pulmonary vein. The caudal vein also connects to an unusual pairing of
posterior cardinal veins, which have lateral connections, a dominant right
side, and empty into the ductus Cuvier. Additional examples of venous
asymmetry in Polypterus, first noted by Muller (1845; as quoted in Magid,
1967), are the relatively small and short left ductus Cuvier and small left
jugular vein.
    Surprisingly, physiological studies are lacking for the circulatory system
of Polypterids, despite the fact that they are extant fish with paired lungs.

5.2. Garfishes
    Although the gill structure of garfish is reduced, remarkable modifica-
tions are lacking (Landolt and Hill, 1975). Smatresk and Cameron (1982)
examined the gill and pulmonary circulations in Lepisosteus oculatus using
corrosion vascular casts, radiolabeled microspheres, and gas tensions. They
concluded that the arterial system of the gill filaments was similar to the
typical teleost pattern. Three paired aVerent branchial arteries originated on
the ventral aorta, with the third and fourth gill arches having a common
aVerent branchial artery. Similarly, the eVerent branchial arteries united into
a common vessel before directly entering the dorsal aorta. No major non-
respiratory shunt vessels were evident.
    The vascular pathways in the secondary lamellae of garfish are unusual in
that the pathways are clearly ‘‘channelized’’ (Figure 2.16B), but not to the
extremes seen in protopterus. Lamellar vascular channels provide a more direct
flow of blood across the respiratory surface and reduce the vascular density of
the exchange area. Nevertheless, oxygen is lost as blood passes through the
lamellae, as indicated by oxygen tensions in the ventral aortic blood exceed-
ing those in the dorsal aortic blood when the garfish are breathing air
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                             107

(Smatresk and Cameron, 1982). A few vascular anastomoses between the
aVerent filament artery and the central venous sinus have been observed.
    Blood flow to the lung is derived caudally from the dorsal aorta as
numerous paired branches (Potter, 1927). Venous drainage from the lung is
also by means of numerous paired vessels that join only to the right side of
the posterior cardinal vein (Potter, 1927).
    L. oculatus had a routine heart rate of 33 minÀ1 at 20  C, a ventral aortic
blood pressure of 2.5 kPa, and a dorsal aortic blood pressure of 1.9 kPa
(Smatresk and Cameron, 1982). These arterial blood pressures are somewhat
lower than those of teleosts. However, similar to teleost fish, gill vascular
resistance was about 25% of the total vascular resistance. Air breathing was
found to have no eVect on heart rate and arterial blood pressures. In the
absence of direct measurements, Smatresk and Cameron (1982) used Fick
calculations for carbon dioxide to estimate Q. During hypoxia, Q increased
from 31 to 40 ml minÀ1 kgÀ1, with a modest hypertension (ventral aortic
blood pressure ¼ 3.0 kPa and dorsal aortic blood pressure ¼ 1.9 kPa), tachy-
cardia (up to 36 minÀ1), and increased air‐breathing frequency (from 1 hÀ1 to
7 hÀ1). Using these measurements and assuming a ventricular mass of 1 g kgÀ1
body mass, routine myocardial power output can be estimated as 0.8 mW gÀ1,
increasing to 2.0 mW gÀ1 during hypoxia.
    Blood flow to the lung was estimated as 19% of Q during normoxia.
Pulmonary flow increased from 5.9 to 12.1 ml minÀ1 kgÀ1 during hypoxia.
Consequently, most of the increase in Q during hypoxia was diverted directly
to the lung (Smatresk and Cameron, 1982). During normoxia, the proportion
of blood flow was similar for each gill arch and remained unchanged during
    There is no evidence for sympathetic innervation of the garfish heart since
postganglionic fibers are absent in the atrium and ventricle (Nilsson, 1981).
The atrium of Lepisosteus platyrhinchus receives vagal innervation. Catechol-
amine stores are found in the sinus venosus, atrium, and the small coronary
vessels in the conus arteriosus, but not in the ventricle (Table 2.1). Both the
atrium and ventricle may have adrenergic control via either humoral or
paracrine mechanisms because adrenaline had a positive inotropic eVect on
paced atrial and ventricular muscle strips (Nilsson (1981). Carbachol had a
negative inotropic eVect on the atrium, but not the ventricle (Nilsson, 1981).
The garfish spleen contracted with both adrenaline and noradrenaline, as in
teleosts (Nilsson, 1981).

5.3. Amia (Bowfins)
   The heart of Amia has no unusual features. A cephalad coronary artery,
which originates in the first and second gill arches (Randall et al., 1981),
supplies the compact myocardium of the ventricle. The conus arteriosus is
108                                                       ANTHONY P. FARRELL

greatly reduced and gives way to a longer bulbus arteriosus. Three aVerent
branchial arteries arise from the ventral aorta, with the posterior one dividing
to supply the third and fourth gill arches, as in garfish.
     The gill respiratory dimensions in Amia are similar for all four gill arches
(Olson, 1981). The central vessels of the secondary lamellae lack an endothe-
lial lining (Olson, 1981) and are not channelized to the extent seen in garfish
(Figure 2.16D). The secondary lamellae on apposing gill filaments are fused
by epithelium (Figure2.16C) (Bevelander, 1934), presumably to prevent la-
mellar collapse in air, and this means that the most marginal lamellar blood
channels ( just like basal lamellar channels in most fish) are buried in epithe-
lium which restricts gas transfer (Olson, 1981). However, these potential
intralamellar nonrespiratory shunts have a limited benefit when Amia
breathe air during aquatic hypoxia because oxygen can be lost from the
gills and the transfer factor for this oxygen loss was calculated to be the
same as that for oxygen uptake during water breathing (Randall et al., 1981).
The few connections between the aVerent filament artery and the secondary
circulation of the gills are outnumbered by those from the eVerent filament
artery (Olson, 1981; but see Daxboeck et al., 1981). Thus, no major non-
respiratory shunt vessels have been identified from vascular corrosion casts
of Amia gills (Olson, 1981; Randall et al., 1981).
     The first and second eVerent branchial arteries supply the pseudobranch,
the cranial vessels, and the dorsal aorta. The third and fourth eVerent
branchial arteries are modified to supply the air bladder (Olson, 1981;
Randall et al., 1981). The fourth eVerent branchial artery is directly confluent
with the air bladder artery. The third and fourth eVerent branchial arteries
are connected by a short ductus near their dorsal emergence from the gill
arches. The ductus is orientated to favor flow from the third into the fourth
eVerent branchial artery, and thus blood flow from the fourth gill arch may
reach the dorsal aorta only via the ductus and the third eVerent branchial
     A single coeliacomesenteric artery arises from the dorsal aorta just caudal
to the junction of the third eVerent branchial artery (Olson, 1981). A more
direct connection between the coeliacomesenteric artery and the third eVer-
ent branchial artery has been suggested in the schematic diagram of Randall
et al. (1981) and in some of the specimens examined by Olson (1981).
     Studies of cardiovascular physiology are limited. The Fick estimate for Q
was 66–69 ml minÀ1 kgÀ1 during regular air breathing at 30  C. The air
bladder received 39% (25.9 ml minÀ1 kgÀ1) of Q (Randall et al., 1981), a
percentage that would represent almost the entire flow from the third and
fourth gill arches. Wilder (1876) had earlier suggested that blood flow in the
posterior dorsal aorta and the air bladder should be similar given their
similar vessel diameters. McKenzie et al. (1991b) measured routine heart
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                            109

rate (30 minÀ1) and dorsal aortic blood pressure (2.9 kPa) in Amia at 20  C. A
single tracing in Amia at 27  C of ventral aortic flow after an air breath
showed Q increasing and heart rate remaining unchanged (32 minÀ1;
Johansen et al., 1970). By combining these measurements routine Q in
Amia has an impressive cardiac stroke volume of 2 ml kgÀ1.
    During hypoxia at 24  C, Amia increased Q to 96.3 ml minÀ1 kgÀ1 (stroke
volume ¼ 3 ml kgÀ1; but note these estimates were based on some 30  C
data), which is an impressive Q among fishes. At the same time, blood flow to
the air bladder increased to 25.9 ml minÀ1 kgÀ1, but not at the expense of the
systemic circulation, which remained unchanged at 44 ml minÀ1 kgÀ1, that is,
all of the increase in Q was directed to the air bladder.
    Aquatic hypoxia (6.3 kPa) and NaCN injection into the buccal cavity
both decreased heart rate by about 15%. Aquatic hypoxia increased and
NaCN decreased dorsal aortic blood pressure. Adrenaline and noradrenaline
infusions both increased heart rate by 30–40% and dorsal aortic blood
pressure by 60–70% (McKenzie et al., 1991b).


6.1. Cardiac Anatomy
     The anatomy of the sturgeon heart has a few notable peculiarities. The
ventricle has nodular tissue located in the subendocardium (Hertwig, 1873;
Khloponin, 1979; Fange, 1986). This nodular tissue can be several milli-
meters thick in a 30‐kg fish and has outer and inner layers resembling,
respectively, the cortical and medullar regions of the thymus. It apparently
serves as a lymphohemopoietic organ, is well vascularized, and contains
lymphocytes and young forms of white blood cells (Fange, 1986; Icardo
et al., 2002b). Such a function for the ‘‘heart’’ may be unique among fishes.
     Myklebust and Kryvi (1979), who examined the cardiac ultrastructure of
Acipenser stellatus, showed that the myofibrils and much of the sarcoplasmic
reticulum have peripheral cellular locations, while the mitochondria and
the nucleus have central locations. T‐tubules were not observed and large
(þ300 nm) glycogen granules were observed only in the atrium. Nerves and
nerve endings were apparent among the myocytes but their locations were
not given.
     The outflow tract of the heart comprises a proximal conus arteriosus and
distal bulbus arteriosus within the pericardium [see descriptions for Acipen-
ser naccarii in Icardo et al. (2002a,b) and Guerroero et al. (2004). The conus
is rich in collagen but not elastin, is surrounded by well‐vascularized, com-
pact myocardial muscle, and contains rows of conal valves (cushions).
110                                                       ANTHONY P. FARRELL

In contrast, the bulbus is rich in elastin, lacks cardiac muscle, and carries the
coronary artery on its surface. Embryonic development of the heart has been
carefully documented for A. naccarii (Icardo et al., 2004). Notably, the conus
and its endocardial cushions appear 4 days posthatch and the coronary
artery appears 10 days posthatch.
    The coronary circulation has diVerent origins among sturgeons: a cepha-
lad origin in the paddlefish Polyodon spathula (from the most posterior
eVerent filament artery) and a caudal origin in the shovelnose sturgeon
Scaphirhynchus platorynchus (from the subclavian artery) (Danforth, 1916;
Foxon, 1950). In addition to supplying the compact myocardium of the
conus and ventricle, the coronary circulation apparently serves the inner
spongy myocardium of the ventricle in the beluga sturgeon (Huso huso),
                                                      ¨     ¨
stellate sturgeon (A. stellatus), and Acipenser guldenstadti, much like the
situation in skates and sharks (Romensky, 1978).

6.2. Circulatory Patterns
    For A. stellatus, the circulatory pattern in the gills (Byczkowska‐Smyk,
1962) and the swim bladder (supplied by a branch of the coeliacomesenteric
artery and drained to the hepatic portal system; Jasinski, 1965) are not

6.3. Cardiac Dynamics
     For A. naccarii at 23  C, heart rate and dorsal aortic blood pressure are
reported as 62 minÀ1 and 2.8 kPa (McKenzie et al., 1995), and 60 minÀ1 and
2.5–3.0 kPa (Agnisola et al., 1996). For Acipenser transmontanus at 19  C,
heart rate and dorsal aortic blood pressure are reported as 48 minÀ1 and
2.9 kPa (Crocker et al., 2000). Routine heart rate is under modest b‐adrener-
gic and cholinergic tonus, as suggested by the findings of tachycardia (up to
67 minÀ1) after atropine injection, bradycardia (down to 52 minÀ1) after
injection of propranolol (McKenzie et al., 1995), and a 9% increase in heart
rate after isoproterenol injection (Crocker et al., 2000). Kisch (1950) reported
heart rate as 70 minÀ1 (temperature was not given) and a fright‐induced
bradycardia reflex for Acipenser sturio.
     In A. transmontanus at 19  C, Q was 36 ml minÀ1 kgÀ1 and stroke volume
was 0.83 ml kgÀ1 (Crocker et al., 2000). In an anesthetized A. naccarii at
23  C, Q was 13 ml minÀ1 kgÀ1 and stroke volume 0.2 ml kgÀ1 (Agnisola
et al., 1996). Systemic vascular resistance was 88 Pa min kg mlÀ1. Following a
struggle, Q increased by 29%. Hypercarbia (2 h at 2.6 kPa) decreased system-
ic resistance, which produced systemic hypotension despite an increase in Q
to 68 ml minÀ1 kgÀ1 (occasionally up to 82 ml minÀ1 kgÀ1). Combining these
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                             111

in vivo data, and assuming a blood pressure loss of one‐third across the
branchial circulation and a relative ventricular mass of 1 g kgÀ1 body mass,
routine myocardial power output is estimated at 2.3–5.3 mW gÀ1.
    By comparison, isolated, perfused ventricles from A. naccarii with a low
spontaneous heart rate (20 minÀ1 at 20  C) had a very low basal power
output (0.3 mW gÀ1; Agnisola et al., 1996). Heart rate was independent of
filling pressure and output pressure. However, power output was doubled at
maximum stroke volume and increased a further 50% when diastolic output
pressure was increased from 2.0 kPa to a maximum level, such that the
maximum power output was 1.5 mW gÀ1. Subambient filling pressures
were reported for the perfused heart (Agnisola et al., 1996). Thus, vis‐a‐fronte
cardiac filling seems likely especially since the heart is surrounded by a
fibrous pericardium and there is pericardioperitoneal connection, as in

6.4. Circulatory Control
    Hypoxia causes bradycardia. In A. naccarii at 23  C, acute hypoxia
(2.5 kPa for 20 min) and externally applied NaCN slowed heart rate (cholin-
ergically mediated) without a pressor eVect on dorsal aortic blood pressure
(McKenzie et al., 1995). Agnisola et al. (1996) found that moderate hyp-
oxia (6.6 kPa for 30 minÀ1) had no cardiovascular eVects, but deep hypoxia
(4.6 kPa for 15 min) produced bradycardia (4–50) and, in contrast, systemic
hypotension (dorsal aortic blood pressure ¼ 2.0–2.5 kPa). Adrenaline injec-
tions cause tachycardia and systemic hypertension, while propranolol causes
bradycardia and systemic hypotension in A. naccarii (McKenzie et al., 1995).
    Blood flow in the coeliacomesenteric artery was 20% of routine Q
(Crocker et al., 2000). Struggling increased splanchnic vascular resistance,
which rapidly reduced splanchnic blood flow by 72%. An a‐adrenergic
vasoconstriction of the systemic and splanchnic circulations was demon-
strated by increases in vascular resistance after phenylepherine injection,
but splanchnic vascular resistance was unaVected by isoproterenol injection,
even though it increased Q and decreased systemic vascular resistance.


    There is certainly still much to be discovered about the circulatory
physiology of primitive fishes. For many orders, other than cyclostomes
and lungfishes, there are just a few physiological studies of cardiovascular
control, and none for Polypterids. More studies are certainly needed to either
validate or refute the speculation presented here in the synthesis and the
112                                                                 ANTHONY P. FARRELL

detailed descriptions. In a much earlier evolutionary synthesis based on
brain, lobe fin, and cardiac anatomy, Wilder (1876) pleaded ‘‘guilty’’ to not
including any ‘‘osteological characters by which fossil forms may be collo-
cated with the living.’’ Today we can be more confident that the study of
extant species can reveal important information about the evolution of the
vertebrate circulatory system. Encouragement is provided by the derived
circulatory diversity among lungfishes (Burggren and Johansen, 1986), as
well as the relics of circulatory design (e.g., the vestigial pulmonary vessel in
Latimeria). Likewise, the ontogenetic loss of a ductus Cuvier in lampreys and
the developmental studies of the sturgeon cardiac outflow tract indicate that
much remains to be discovered in terms of evolutionary trends from studies
of cardiovascular development. Therefore, this chapter closes with two
remarks and no apology.
     The evidence presented here for very diVerent circulatory anatomy and
physiology in hagfishes and lampreys must surely place them further apart
from each other than their current phylogeny indicates. Many more circula-
tory similarities can be described for other, more phylogenetically divergent
primitive fishes. Finally, primitive fishes are clearly survivors and this is
reflected well in their cardiac hypoxia tolerance. Hagfishes and perhaps
lampreys apparently took the path of ‘‘low cardiac ATP demand.’’ Other
primitive fishes have a higher cardiac ATP demand, but their hearts benefit
from an additional oxygen supply in their coronary circulation and from air
breathing. A notable exception is Latimeria, which appears to have adopted
a sedentary lifestyle in stable and remote marine environments. Added to this
is the remarkable tolerance of hypercarbia by sturgeon, that is, levels that will
anesthetize salmon. Therefore, exciting discoveries remain ahead for
researchers who tackle why the hearts of primitive fishes are so tolerant.


    The assistance of Christopher Wilson, Joanna Bernhardt, and Linda Hansen in locating and
organizing reference material and figures is gratefully appreciated.


Abrahamsson, T., Holmgren, S., Nilsson, S., and Pettersson, K. (1979a). On the chromaYn
   system of the African lungfish, Protopterus aethiopicus. Acta Physiol. Scand. 107, 135–139.
Abrahamsson, T., Holmgren, S., Nilsson, S., and Pettersson, K. (1979b). Adrenergic and
   cholinergic eVects on the heart, the lung and the spleen of the African lungfish, Protopterus
   aethiopicus. Acta Physiol. Scand. 107, 141–147.
Agnisola, C., McKenzie, D. J., Taylor, E. W., Bolis, C. L., and Tota, B. (1996). Cardiac
   performance in relation to oxygen supply varies with dietary lipid composition in sturgeon.
   Am. J. Physiol. 271, R417–R425.
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                 113

Arlock, P. (1975). Electrical activity and mechanical response in the systemic heart and the portal
     vein heart of Myxine glutinosa. Comp. Biochem. Physiol. 51A, 521–522.
Augustinsson, K. B., Fange, R., Johnels, A., and Ostlund, E. (1956). Histological, physiological
     and biochemical studies on the heart of two cyclostomes, hagfish (Myxine) and lamprey
     (Lampetra). J. Physiol. 131, 257–276.
Axelsson, M., and Farrell, A. P. (1993). Coronary blood flow in vivo in the coho salmon. Am.
     J. Physiol. 264, R963–R971.
Axelsson, M., Abe, A. S., Bicudo, J. E. P. W., and Nilsson, S. (1989). On the cardiac control in
     the South American lungfish, Lepidosiren paradoxa. Comp. Biochem. Physiol. 93A, 561–565.
Axelsson, M., Farrell, A. P., and Nilsson, S. (1990). EVects of hypoxia and drugs on the
     cardiovascular dynamics of the Atlantic hagfish Myxine glutinosa. J. Exp. Biol. 151, 297–316.
Baxter, E. W. (1958). The development of the intestinal blood‐system of lampreys (Petromyxoni-
     dae), with an account of the origin of vascular couples. Proc. Zool. Soc. Lond. 129, 371–396.
Bevelander, G. (1934). The gills of Amia calva specialized for respiration in an oxygen deficient
     habitat. Copeia 1934, 123–127.
              ¨                                                 ´
Bloom, G., Ostlund, E., von Euler, U. S., Lishajko, F., Ritzen, M., and Adams‐Ray, J. (1961).
     Studies on catecholamine‐containing granules of specific cells in cyclostome hearts. Acta
     Physiol. Scand. 53, 1–34.
               ¨                    ¨
Bloom, G., Ostlund, W., and Fange, R. (1963). Functional aspects of cyclostome hearts in
     relation to recent structural findings. In ‘‘The Biology of Myxine’’ (Brodal, A., and
     Fange, R., Eds.), pp. 317–339. Grondahl and Son, Oslo.
Bloom, G. D. (1962). The fine structure of the cyclostome cardiac muscle cells. Z. Zellforsch.
     Mikrosk. Anat. 57, 213–239.
Boas, J. E. V. (1880). Ueber Herz und Arterienbogen bei Ceratodus und Protopterus. Morph.
     Jahrb. 6, 321–354.
Britz, R., and Johnson, G. D. (2003). On the homology of the posterior most gill arch in
     Polypterids (Cladistia, Actinopterygii). Zoolog. J. Linnean Soc. 138, 495–503.
Brodal, A., and Fange, R. (1963). ‘‘The Biology of Myxine.’’ Grondahl and Son, Oslo.
Budgett, J. S. (1901). On some points in the anatomy of Polypterus. Trans. Zool. Soc. Lond. 15,
Bugge, J. (1961). The heart of the African lungfish, Protopterus. Vidensk. Meddr. Dansk. Natuhr.
     Foren. 123, 193–210.
Burggren, W. W., and Johansen, K. (1986). Circulation and respiration in lungfishes (Dipnoi).
     J. Morphol. Suppl. 1, 217–236.
Burggren, W. W., Farrell, A. P., and Lillywhite, H. (1997). Vertebrate cardiovascular systems.
     In ‘‘The Handbook of Physiology. Section 13: Comparative Physiology’’ (Dantzler, W. H.,
     Ed.), Vol 1, pp. 215–308. Oxford University Press, New York, Oxford.
Bushnell, P. G., Jones, D. R., and Farrell, A. P. (1992). The arterial system. In ‘‘Fish Physiology’’
     (Hoar, W. S., Randall, D. J., and Farrell, A. P., Eds.), Vol. 12A, pp. 89–139. Academic
     Press, New York.
Butler, P. J., and Metcalfe, J. D. (1988). Cardiovascular and respiratory systems. In ‘‘Physiology
     of Elasmobranch Fishes’’ (Shuttleworth, T. J., Ed.), pp. 1–47. Springer‐Verlag, Berlin.
Byczkowska‐Smyk, W. (1962). Vascularization and size of the respiratory surface in. Acipenser
     stellatus Pall. Acta Biol. Cracov. 5, 303–316.
Carlson, A. J. (1906). The presence of cardioregulative nerves in lampreys. Am. J. Physiol. 16,
Chapman, C. B., Jensen, D., and Wildenthal, K. (1963). On circulatory control mechanisms in
     the Pacific hagfish. Circ. Res. 12, 427–440.
Cole, F. J. (1926). A monograph on the general morphology of the myxinoid fishes, based on a
     study of Myxine. Part IV. The morphology of the vascular system. Trans. Roy. Soc. Edinb.
     54, 309–342.
114                                                                    ANTHONY P. FARRELL

Crocker, C. E., Farrell, A. P., Gamperl, A. K., and Cech, J. J., Jr. (2000). Cardiorespiratory
    responses of white sturgeon to environmental hypercapnia. Am. J. Physiol. Regul. Integr.
    Comp. Physiol. 279, R617–R628.
Danforth, C. H. (1916). The relation of the coronary and hepatic arteries in the common
    ganoids. Am. J. Anat. 19, 391–400.
Dashow, L., Epple, A., and Nibbio, B. (1982). Catecholamines in adult lampreys: Baseline levels
    and stress induced changes, with a note on cardiac cannulation. Gen. Comp. Endocrinol. 46,
Davie, P. S., and Farrell, A. P. (1991). The coronary and luminal circulations of the myocardium
    of fishes. Can. J. Zool. 69, 1993–2001.
Davie, P. S., Forster, M. E., Davison, B., and Satchell, G. H. (1987). Cardiac function in the New
    Zealand hagfish. Eptatretus cirrhatus. Physiol. Zool. 60, 233–240.
Daxboeck, C., Barnard, D. K., and Randall, D. J. (1981). Functional morphology of the gills of
    the bowfin, Amia calva, with special reference to their significance during air exposure.
    Respir. Physiol. 43, 349–364.
Endo, H., Yamada, T. K., Kurohmaru, M., and Hayashi, Y. (1997). Structure of portal vein
    heart wall in the brown hagfish (Paramyxine atami). J. Morphol. 231, 225–230.
Euler, U. S. v., and Fange, R. (1961). Catecholamines in nerves and organs of Myxine glutinosa,
    Squalus acanthias, and Gadus callarias. Gen. Comp. Endocrinol. 1, 191–194.
Evans, D. H., and Harrie, A. C. (2001). Vasoactivity of the ventral aorta of the American eel
    (Anguilla rostrata), Atlantic hagfish (Myxine glutinosa), and sea lamprey (Petromyzon
    marinus). J. Exp. Zool. 289, 273–284.
Falck, B. V., Mecklenburg, C., Myhrberg, H., and Persson, H. (1966). Studies on adrenergic and
    cholinergic receptors in the isolated hearts of Lampetra fluviatilis (Cyclostomata) and
    Pleuronectes platessa (Teleostei). Acta Physiol. Scand. 68, 64–71.
Fange, R. (1972). The circulatory system. In ‘‘The Biology of Lampreys’’ (Hardisty, M. W., and
    Potter, I. C., Eds.), Vol. 2, pp. 241–259. Academic Press, London, New York.
Fange, R. (1986). Lymphoid organs in sturgeons (Acipenseridae). Vet. Immunol. Immunopathol.
    12, 153–161.
Fange, R., and Ostlund, E. (1954). The eVects of adrenaline, noradrenaline, tyramine and other
    drugs on the isolated heart from marine vertebrates and a cephalopod (Eledone cirrosa).
    Acta Zool. 35, 289–305.
 ¨                             ¨
Fange, R., Bloom, G., and Ostlund, E. (1963). The portal vein heart of myxinoids. In ‘‘The
    Biology of Myxine’’ (Brodal, A., and Fange, R., Eds.), pp. 340–351. Grondahl and Son, Oslo.
Farmer, C. (1997). Did lungs and the intracardiac shunt evolve to oxygenate the heart in
    vertebrates? Paleobiology 23, 358–372.
Farrell, A. P. (1984). A review of cardiac performance in the teleost heart: Intrinsic and humoral
    regulation. Can. J. Zool. 62, 523–536.
Farrell, A. P. (1991). From hagfish to tuna: A perspective on cardiac function in fish. Physiol.
    Zool. 64, 1137–1164.
Farrell, A. P., and Clutterham, S. M. (2003). On‐line venous oxygen tensions in rainbow trout
    during graded exercise at two acclimation temperatures. J. Exp. Biol. 206, 487–496.
Farrell, A. P., and Jones, D. R. (1992). The heart. In ‘‘Fish Physiology’’ (Hoar, W. S.,
    Randall, D. J., and Farrell, A. P., Eds.), Vol. 12A, pp. 1–88. Academic Press, San Diego.
Farrell, A. P., and Stecyk, J. A. W. (2007). The heart as a working model to explore themes and
    strategies for anoxic survival in ectothermic vertebrates. Comp. Biochem.Physiol. 147A,
Farrell, A. P., Sobin, S. S., Randall, D. J., and Crosby, S. (1980). Intralamellar blood flow
    patterns in fish gills. Am. J. Physiol. 239, R428–R436.
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                115

Farrell, A. P., Johansen, J. A., and Graham, M. S. (1988a). The role of the pericardium in cardiac
     performance of the trout (Salmo gairdneri). Physiol. Zool. 61, 213–221.
Farrell, A. P., Hammons, A. M., Graham, M. S., and Tibbits, G. F. (1988b). Cardiac growth in
     rainbow trout, Salmo gairdneri. Can. J. Zool. 66, 2368–2373.
Farrell, A. P., Simonot, D. L., Seymour, R. S., and Clark, T. D. (2007). A novel technique for
     estimating the compact myocardium in fish reveals surprising results for an athletic air
     breathing fish, the pacific tarpon. J. Fish. Biol. in press .
Fishman, A. P., DeLaney, R. G., and Laurent, P. (1985). Circulatory adaptation to bimodal
     respiration in the dipnoan lungfish. J. Appl. Physiol. 59, 284–294.
Fock, U., and Hinssen, H. (2002). Nebulin is a thin filament protein of the cardiac muscle of the
     Agnathans. J. Muscle Res. Cell Motil. 23, 205–213.
Forster, M. E. (1989). Performance of the heart of the hagfish. Eptatretus cirrhatus. Fish Physiol.
     Biochem. 6, 327–331.
Forster, M. E. (1991). Myocardial oxygen consumption and lactate release by the hypoxic
     hagfish heart. J. Exp. Biol. 156, 583–590.
Forster, M. E. (1997). The blood sinus system of hagfish: Its significance in a low‐pressure
     circulation. Comp. Biochem. Physiol. 116A, 239–244.
Forster, M. E., Davie, P. S., Davison, W., Satchell, G. H., and Wells, R. M. G. (1988). Blood
     pressures in swimming hagfish. Comp. Biochem. Physiol. 89A, 247–250.
Forster, M. E., Axelsson, M., Farrell, A. P., and Nilsson, S. (1991). Cardiac function and
     circulation in hagfishes. Can. J. Zool. 69, 1985–1992.
Forster, M. E., Davison, W., Axelsson, M., and Farrell, A. P. (1992). Cardiovascular responses
     to hypoxia in the hagfish, Eptatretus cirrhatus. Respir. Physiol. 88, 373–386.
Foxon, G. E. H. (1950). A description of the coronary arteries in dipnoan fishes and some
     remarks on their importance from the evolutionary standpoint. J. Anat. 84, 121–131.
Franklin, C. E., and Davie, P. S. (1992). Sexual maturity can double heart mass and cardiac
     power output in male rainbow trout. J. Exp. Biol. 171, 139–148.
Fritsche, R., Axelsson, M., Franklin, C. E., Grigg, G. C., Holmgren, S., and Nilsson, S. (1993).
     Respiratory and cardiovascular responses to hypoxia in the Australian lungfish. Respir.
     Physiol. 94, 173–187.
Gannon, B. J., Randall, D. J., Browning, J., Lester, R. J. G., and Rogers, L. J. (1983). The
     microvascular organization of the gas exchange organs of the Australian lungfish. Neocer-
     atodus foresteri (Krefft). Aust. J. Zool. 31, 651–676.
Goodrich, E. S. (1930). ‘‘Studies on the Structure and Development of Vertebrates,’’ p. 837.
     Dover, New York.
Graham, J. B. (1997). ‘‘Air‐Breathing Fishes: Evolution, Diversity, and Adaptation,’’ p. 299.
     Academic Press, New York.
Grant, R. T., and Regnier, M. (1926). The comparative anatomy of the cardiac coronary vessels.
     Heart J. 13, 285–316.
Greene, C. W. (1900). Contributions to the physiology of the California hagfish, Polistotrema
     stouti—I. The anatomy and physiology of the caudal heart. Am. J. Physiol. 3, 366–382.
Greene, C. W. (1902). Contributions to the physiology of the California hagfish, Polistotrema
     stouti—II. The absence of regulative nerves for the systemic heart. Am. J. Physiol. 6, 318–324.
Greene, C. W. (1926). The physiology of the spawning migration. Physiol. Rev. 6, 201–241.
Guerrero, A., Icardo, J. M., Duran, A. C., Gallego, A., Domezain, A., Colvee, E., and Sans‐
     Coma, V. (2004). DiVerentiation of the cardiac outflow tract components in alevins of the
     sturgeon Acipenser naccarii (Osteichthyes, Acipenseriformes): Implications for heart evolu-
     tion. J. Morphol. 260, 172–183.
Gunther, A. (1871). Description of Ceratodus, a genus of ganoid fishes, recently discovered in
     rivers of Queensland, Australia. Philos. Trans. R. Soc. Lond. 161, 511–792.
116                                                                     ANTHONY P. FARRELL

Hansen, C. A., and Sidell, B. D. (1983). Atlantic hagfish cardiac muscle: Metabolic basis of
     tolerance to anoxia. Am. J. Physiol. 244, R356–R362.
Hardisty, M. W. (1979). ‘‘Biology of the Cyclostomes.’’ Cambridge University Press,
Hardisty, M. W., and Potter, I. C. (1972).‘‘The Biology of Lampreys,’’ Vol. 2. Academic Press,
Helle, K. B., and Lonning, S. (1973). The sarcoplasmic reticulum in the portal vein heart and
     ventricle of the cyclostome Myxine glutinosa (L). J. Mol. Cell. Cardiol. 5, 433–439.
Hertwig, R. (1873). Die Lymphoiden Drusen auf der Oberflache des Storherzens. Arch. F. Mikr.
     Anatomie 9, 62–79.
Hol, R., and Johansen, K. (1960). A cineradiographic study of the central circulation in the
     hagfish, Myxine glutinosa L. J. Exp. Biol. 37, 469–474.
Holmgren, S., Fritsche, R., Karila, P., Gibbins, I., Axelsson, M., Franklin, C., Grigg, G. C., and
     Nilsson, S. (1994). Neuropeptides in the Australian lungfish Neoceratodus fosteri: EVects
     in vivo and presence in autonomic nerves. Am. J. Physiol. 266, R1568–R1577.
Home, E. (1815). On the structure of the organs of respiration in animals which appear to hold
     an intermediate place between those of the class Pisces and the class Vermes, and in two
     genera of the last mentioned class. Philos. Trans. R. Soc. 105, 256.
Icardo, J. M., Colvee, E., Cerra, M. C., and Tota, B. (2002a). Structure of the conus arteriorus of
     the sturgeon (Acipenser naccarii) heart. I: The conus valves and the subendocardium. Anat.
     Rec. 267, 17–27.
Icardo, J. M., Colvee, E., Cerra, M. C., and Tota, B. (2002b). Structure of the conus arteriosus of
     the sturgeon (Acipenser naccarii) heart: II. The myocardium, the subepicardium and the
     conus‐aorta transition. Anat. Rec. 268, 388–398.
Icardo, J. M., Guerrero, A., Duran, A. C., Domezain, A., Colvee, E., and Sans‐Coma, V. (2004).
     The development of the sturgeon heart. Anat. Embryol. 208, 439–449.
Icardo, J. M., Ojeda, J. L., Colvee, E., Tota, B., Wong, W. P., and Ip, Y. K. (2005a). Heart inflow
     tract of the African lungfish Protopterus dolloi. J. Morphol. 263, 30–38.
Icardo, J. M., Brunelli, E., Perrotta, I., Colvee, E., Wong, W. P., and Ip, Y. K. (2005b). Ventricle
     and outflow tract of the African lungfish. Protopterus dolloi. J. Morphol. 265, 43–51.
Jasinski, A. (1965). The vascularization of the air bladder in fishes. II. Sevruga (Acipenser
     stellatus Pallas) grayling (Thymallus thymallus L.), pike (Esox lucius L.) and umbra
     (Umbra krameri Walbaum). Acta Biol. Cracoviensia Zool. 8, 199–210.
Jensen, D. (1961). Cardioregulation in an aneural heart. Comp. Biochem. Physiol. 2, 181–201.
Jensen, D. (1965). The aneural heart of the hagfish. Ann. N. Y. Acad. Sci. 127, 443–458.
Jensen, D. (1969). Intrinsic cardiac rate regulation in the sea lamprey, Petromyzon marinus and
     rainbow trout, Salmo gairdineri. Comp. Biochem. Physiol. 30, 685–690.
Jesse, M. J., Shub, C., and Fishman, A. P. (1967). Lung and gill ventilation of the African
     lungfish. Respir. Physiol. 3, 267–287.
Johansen, K. (1960). Circulation in the hagfish, Myxine glutinosa L. Biol. Bull. 118, 289–295.
Johansen, K. (1963). Circulation in the hagfish, Myxine glutinosa L. In ‘‘The Biology of Myxine’’
     (Brodal, A., and Fange, R., Eds.), pp. 289–316. Grondahl and Son, Oslo.
Johansen, K. (1965). Cardiovascular dynamics in fishes, amphibians and reptiles. Ann. N. Y.
     Acad. Sci. 127, 414–442.
Johansen, K. (1970). Air breathing in fishes. In ‘‘Fish Physiology’’ (Hoar, W. S., and
     Randall, D. J., Eds.), Vol. 4, pp. 361–411. Academic Press, New York.
Johansen, K. (1971). Comparative Physiology: Gas Exchange and Circulation in Fishes. Ann.
     Rev. Physiol. 33, 569–612.
Johansen, K., and Burggren, W. W. (1980). Cardiovascular function in the lower vertebrates.
     In ‘‘Hearts and Hearts‐Like Organs’’ (Bourne, G. H., Ed.), Vol. 1, pp. 61–117. Academic
     Press, New York.
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                              117

Johansen, K., and Hanson, D. (1968). Functional anatomy of the hearts of lungfishes and
    amphibians. Am. Zool. 8, 191–210.
Johansen, K., and Hol, R. (1968). A radiological study of the central circulation in the lungfish,
    Protopterus aethiopicus. J. Morphol. 126, 333–348.
Johansen, K., and Reite, O. B. (1967). EVects of acetylcholine and biogenic amines on pulmo-
    nary smooth muscle in the African lungfish, Protopterus aethiopicus. Acta Physiol. Scand.
    71, 248–252.
Johansen, K., Fange, R., and Johannessen, M. W. (1962). Relations between blood, sinus fluid
    and lymph in Myxine glutinosa L. Comp. Biochem. Physiol. 7, 23–28.
Johansen, K., Lenfant, C., and Grigg, G. C. (1967). Respiratory control in the lungfish,
    Neoceratodus forsteri (KreVt). Comp. Biochem. Physiol. 20, 835–854.
Johansen, K., Lenfant, C., and Hanson, D. (1968). Cardiovascular dynamics in lungfishes.
    Zeitschrift fur vergleichende Physiologie 59, 157–186.
Johansen, K., Hanson, D., and Lenfant, C. (1970). Respiration in the primitive air breather,
    Amia calva. Respir. Physiol. 9, 162–174.
Johansen, K., Lenfant, C., and Hanson, D. (1973). Gas exchange in the lamprey, Entosphenus
    tridentatus. Comp. Biochem. Physiol. 44A, 107–119.
Johnsson, M., and Axelsson, M. (1996). Control of the systemic heart and the portal heart of
    Myxine glutinosa. J. Exp. Biol. 199, 1429–1434.
Kerr, J. G. (1910). Note on the posterior vena cava in Polypterus. Proc. R. Phys. Soc. Edinb. 18,
Khloponin, P. A. (1979). Morphological aspects on diVerentiation of cardiac myocytes in the
    Russian sturgeon. Arkhiv Anatomii Gistoligii I Embriologii 77, 44–51.
Kilarski, W. (1964). The organization of the cardiac muscle cell of the lamprey (Petromyzon
    marinus L). Acta Biol. Cracov. (Zool.) 7, 75–87.
Kisch, B. (1950). Reflex cardiac inhibition in the Ganoid Acipenser sturio. Am. J. Physiol. 160,
Landolt, J. C., and Hill, L. G. (1975). Observations of the gross structure and dimensions of the
    gills of three species of gars (Lepisosteidae). Copeia 1975, 470–475.
Lankester, E. R. (1878). On hearts of Ceratodus, Protopterus and Chimera with an account of
    undescribed pocket valves in the conus arteriosus of Ceratodus and Protopterus. Trans. Zool.
    Soc. Lond. 10, 493–505.
Laurent, P., DeLaney, R. G., and Fishman, A. P. (1978). The vasculature of the gills in the
    aquatic and aestivating lungfish (Protopterus aethiopicus). J. Morphol. 156, 173–208.
Laurent, P., Holmgren, S., and Nilsson, S. (1983). Nervous and humoral control of the fish heart:
    Structure and function. Comp. Biochem. Physiol. 76A, 525–542.
Leak, L. V. (1969). Electron microscopy of cardiac tissue in a primitive vertebrate. Myxine
    glutinosa. J. Morphol. 128, 131–157.
Lignon, J. M. (1979). Responses to sympathetic drugs in the ammocoete heart: Probable
    influence of the small intensely fluorescent (SIF) cells. J. Mol. Cell. Cardiol. 11, 447–465.
Macey, D. J., Epple, A., Potter, I. C., and Hilliard, R. W. (1984). The eVect of catecholamines on
    branchial and cardiac electrical recordings in adult lampreys (Geotria australis). Comp.
    Biochem. Physiol. 79C, 295–300.
Macey, D. J., Clarke, L. M., and Potter, I. C. (1991). Basal oxygen consumption, ventilatory
    frequency, and heart rate during the protracted spawning run of the Southern Hemisphere
    lamprey Geotria australis. J. Comp. Physiol. B. 161, 525–531.
Magid, A. M. A. (1967). Observations on the venous system of three species of Polypterus
    (Pisces). J. Zool. Lond. 152, 19–30.
McKenzie, D. J., Aota, S., and Randall, D. J. (1991a). Ventilatory and cardiovascular responses
    to blood pH, plasma PCO2 blood O2 content and catecholamines in an air‐breathing fish,
    Amia calva. Physiol. Zool. 64, 432–450.
118                                                                    ANTHONY P. FARRELL

McKenzie, D. J., Burleson, M. L., and Randall, D. J. (1991b). The eVects of branchial denerva-
    tion and pseudobranch ablation on cardioventilatory control in an air‐breathing fish.
    J. Exp. Biol. 161, 347–365.
McKenzie, D. J., Taylor, E. W., Bronzi, P., and Bolis, C. L. (1995). Aspects of cardioventilatory
    control in the Adriatic sturgeon (Acipenser naccarii). Respir. Physiol. 100, 45–53.
Millot, J., Anthony, J., and Robineau, D. (1978). Anatomie de Latimeria chalumnae. Tome 3,
    Appareils digestifs, respiratoire, urogenetal, circulatoire, glandes endocrine, teguments,
    ecailles, conclusions generales. Paris, Editions de Centre National de Rech. Scient p. 198.
Muller, J. (1841). Vergleichende Anatomie der Myxinoiden, der Cyclostomen mit durchbohrten
    Gaumen. IV: Uber das Gefassystem. Abhandl. D. Konigl. Akad. D. Wissensch. zu Berlin
    1839, 175–303.
Muller, J. (1842). Beobachtungen uber die Schwimblase der Fishe, mit Bezug auf einige neue
    Fischegattungen. Archiv fur Anatomie Physiol. Wiss. Medicin. 307–329.
Muller, J. (1845). Uber den Bau und die Grenzen der Ganoiden und uber das naturliche System
    der Fishe. Abh. Preuss. Akad. Wiss. 1845, 85–93.
Myklebust, R., and Kryvi, H. (1979). Ultrastructure of the heart of the sturgeon Acipenser
    stallatus (Chondrostei). Cell Tissue Res. 202, 431–438.
Nakao, T., and Uchinomiya, K. (1978). A study on the blood vascular system of the lamprey gill
    filament. Am. J. Anat. 151, 239–264.
Nilsson, S. (1981). On the adrenergic system of Ganoid fish: The Florida spotted gar, Lepisosteus
    platyrhincus (Holostei). Acta Physiol. Scand. 111, 447–454.
Nilsson, S. (1983). ‘‘Autonomic Nerve Function in the Vertebrates.’’ Springer‐Verlag, Berlin.
Nilsson, S., and Axelsson, M. (1987). Cardiovascular control system in fish. In ‘‘Neurobiology of
    the Cardiorespiratory System’’ (Taylor, E. W., Ed.), pp. 115–133. Manchester University
    Press, UK.
Olson, K. R. (1981). Morphology and vascular anatomy of the gills of a primitive air‐breathing
    fish, the bowfin (Amia calva). Cell Tissue Res. 218, 499–517.
Olson, K. R. (1994). Circulatory anatomy in bimodally breathing fish. Am. Zool. 34, 280–288.
Olson, K. R., and Farrell, A. P. (2006). The Cardiovascular System. In ‘‘The Physiology of Fishes’’
    (Evans, D. H., and Claiborne, J. B., Eds.), 3rd edn., pp. 119–152. CRC Press, Boca Raton.
Olson, K. R., Dombkowski, R. A., Russell, M. J., Doellman, M. M., Head, S. K., Whitfield, N. L.,
    and Madden, J. A. (2006). Hydrogen sulfide as an oxygen sensor/transducer in vertebrate
    hypoxic vasoconstriction and hypoxic vasodilation. J. Exp. Biol. 209, 4011–4023.
Ostadal, B., and Schiebler, T. H. (1971). Uber die terminale Strombahn in Fischherzen. Z. Anat.
    Entwickl. Gesch. 134, 101–110.
Parker, G. H., and Davis, F. K. (1899). The blood vessels of the heart of Carachias, Raja, and
    Amia. Boston Soc. Nat. Hist. Proc. 29, 163–178.
Perry, S. F., Fritsche, R., and Thomas, S. (1993). Storage and release of catecholamines from the
    chromaYn tissue of the Atlantic Myxine glutinosa hagfish. J. Exp. Biol. 183, 165–184.
Perry, S. F., Gilmour, K. M., Vulesevic, B., McNeill, B., Chew, S. F., and Ip, Y. K. (2005).
    Circulating catecholamines and cardiorespiratory responses in hypoxic lungfish (Proto-
    pterus dolloi): A comparison of aquatic and aerial hypoxia. Physiol. Biochem. Zool. 78,
Potter, G. E. (1927). Respiratory function of the swim bladder in Lepidosteus. J. Exp. Zool. 49,
Poupa, D., Helle, K., and Lomsky, M. (1985). Calcium paradox from cyclostome to man—a
    comparative study. Comp. Biochem. Physiol. 81A, 801–806.
Randall, D. J. (1968). Functional morphology of the heart in fishes. Am. Zool. 8, 179–189.
Randall, D. J. (1970). The circulatory system in fish physiology. In ‘‘Fish Physiology’’
    (Hoar, W. S., and Randall, D. J., Eds.), pp. 132–172. Academic Press, New York.
2.   CARDIOVASCULAR SYSTEMS IN PRIMITIVE FISHES                                                119

Randall, D. J. (1982). The control of respiration and circulation in fish during exercise and
    hypoxia. J. Exp. Biol. 100, 275–288.
Randall, D. J., Cameron, J. N., Daxboek, C., and Smatresk, N. (1981). Aspects of bimodal gas
    exchange in the bowfin, Amia calva L. (Actinopterygii: Amiiformes). Respir. Physiol. 43,
Randall, D. J., McKenzie, D. J., Abrami, G., Bondiolotti, G. P., Natiello, F., Bronzi, P.,
    Bolis, L., and Agradi, E. (1992). EVects of diet on responses to hypoxia in the sturgeon
    (Acipenser naccarii). J. Exp. Biol. 170, 113–125.
Reid, S. G., Bernier, N. J., and Perry, S. F. (1998). The adrenergic stress response in fish: Control
    of catecholamine storage and release. Comp. Biochem. Physiol. 120C, 1–27.
Reid, S. G., Sundin, L., and Milson, W. K. (2006). The cardiorespiratory system in tropical
    fishes: Structure, function, and control. In ‘‘The Physiology of Tropical Fishes’’ (Val, A. L.,
    Almeida‐Val, V. M. F., and Randall, D. J., Eds.), pp. 225–275. Academic Press, San Diego.
Reinecke, M., Betzler, D., and Forssmann, W. G. (1987). Immunohistochemistry of cardiac
    polypeptide hormones (cardiodilatin/atrial natriuretic polypeptide) in brain and hearts of
    Myxine glutinosa. Histochemistry 86, 233–239.
Reite, O. B. (1969). The evolution of vascular smooth muscle responses to histamine and
    5‐hydroxytryptamine. I. Occurrence of stimulatory actions in fish. Acta Physiol. Scand.
    75, 221–239.
Retzius, A. (1824). Yeterligare Bigrag till Anatomien ab Myxine glutinosa. Kongl. Vetenskapsa-
    cademiens Handligar (Stockholm) 408–431.
Robertson, J. I. (1913). The development of the heart and vascular system of Lepidosiren
    paradoxa. Q. J. Micro. Sci. 59, 53–132.
Romensky, O. Y. (1978). Blood supply for compact and spongy myocardium in fish, amphibians
    and reptiles. Arkh. Anat. Gistol. Embriol. 75, 91–95.
Sanchez, A., Soncini, R., Wang, T., Koldkjaer, P., Taylor, E. W., and Glass, M. L. (2001). The
    diVerential cardio‐respiratory responces to ambient hypoxia and systemic hypoxaemia in
    the South American lungfish, Lepidosiren paradoxa. Comp. Biochem. Physiol. 130A,
Sandblom, E., Axelsson, M., and Farrell, A. P. (2006). Central venous pressure and mean
    circulatory filling pressure in the dogfish Squalus acanthias: The roles of adrenergic controls
    and the pericardium. Am. J. Physiol. 291, R1465–R1473.
Santer, R. (1985). Morphology and innervation of the fish heart. Adv. Anat. Embryol. Cell Biol.
    89, 1–102.
Satchell, G. H. (1971).‘‘Circulation in Fishes,’’ 131 London, Cambridge University Press.
Satchell, G. H. (1976). The circulatory system of air breathing fish. In ‘‘Respiration of Amphibi-
    ous Vertebrates’’ (Hughes, G. M., Ed.), pp. 105–123. Academic Press, London.
Satchell, G. H. (1984). On the caudal heart of Myxine (Myxinoidea: Cyclostomata). Acta Zool.
    65, 125–133.
Satchell, G. H. (1986). Cardiac function in the hagfish, Myxine (Myxinoidea: Cyclostomata).
    Acta Zool. 67, 115–122.
Satchell, G. H. (1991). ‘‘Physiology and Form of Fish Circulation,’’ p. 235. Cambridge
    University Press, Cambridge.
Satchell, G. H. (1992). The venous system. In ‘‘Fish Physiology’’ (Hoar, W. S., Randall, D. J.,
    and Farrell, A. P., Eds.), Vol. 12A, pp. 141–183. Academic Press, New York.
Shabetai, R. D., Abel, C., Graham, J. B., Bhargave, V., Keyes, R. S., and Witzum, K. (1985).
    Function of the pericardium and pericarioperitoneal canal in elasmobranch fishes. Am. J.
    Physiol. 248, H198–H207.
Shelton, G. (1976). Gas exchange, pulmonary blood supply, and the partially divided amphibian
    heart. Perspect. Exp. Biol. 1, 247–259.
120                                                                      ANTHONY P. FARRELL

Shiels, H. A., Vornanen, M., and Farrell, A. P. (2002). Force‐frequency relationship in fish
    hearts—a review. Comp. Biochem. Physiol. 132A, 811–826.
Smatresk, N., and Cameron, J. (1982). Respiration and acid‐base physiology of the spotted gar,
    a bimodal breather. I. Normal values and the response to severe hypoxia. J. Exp. Biol. 96,
Smith, W. C. (1918). On the process of the disappearance of the conus arteriosus in teleosts.
    Anat. Rec. 15, 65–72.
Spencer, W. B. (1892). Note on the habits of Ceratodus forsteri. Proc. R. Soc. Vict. 4, 81–84.
SteVensen, J. F., and Lomholt, J. P. (1992). The secondary vascular system. In ‘‘Fish Physiology’’
    (Hoar, W. S., Randall, D. J., and Farrell, A. P., Eds.), Vol. 12A, pp. 185–217. Academic
    Press, New York.
Sundin, L., Axelsson, M., Nilsson, S., Davison, W., and Forster, M. (1994). Evidence of
    regulatory mechanisms for the distribution of blood between the arterial and the venous
    compartments in the hagfish fill pouch.. J. Exp. Biol. 190, 281–286.
Szidon, J. P., Lahiri, S., Lev, M., and Fishman, A. P. (1969). Heart and circulation of the African
    lungfish. Circ. Res. 25, 23–38.
Thorarensen, H., Young, G., and Davie, P. (1996). 11‐Ketoestosterone stimulates growth of
    heart and red muscle in rainbow trout. Can. J. Zool. 74, 912–917.
Tibbits, G. F., Moyes, C. D., and Hove‐Madsen, L. (1992). Excitation‐contraction coupling in
    the teleost heart. In ‘‘Fish Physiology’’ (Hoar, W. S., Randall, D. J., and Farrell, A. P., Eds.),
    Vol. 12A, pp. 267–304. New York, Academic Press.
Vornanen, M., Shiels, H. E., and Farrell, A. P. (2002). Plasticity of excitation‐contraction
    coupling in fish cardiac myocytes. Comp. Biochem. Physiol. 132A, 827–846.
Wilder, B. G. (1876). Notes on the North American Ganoids, Amia, Lepidosteus, Acipenser
    and Polyodon. (Taken from: 1875, Proc. Am. Assoc. Adv. Sci. 24 ). Salem Press, Salem,
    Massachussetts. 149–197.
Wright, G. M. (1984). Structure of the conus arteriosus and ventral aorta in the Sea Lamprey,
    Petromyzon marinus, and the Atlantic hagfish, Myxine glutinosa: Microfibrils, a major
    component. Can. J. Zool. 62, 2445–2456.
Yamauchi, A. (1980). Fine structure of the fish heart. In ‘‘Hearts and Heart‐like Organs’’
    (Bourne, G. H., Ed.), pp. 119–148. Academic Press, New York.


1. Introduction
2. Development of the CNS
3. The Brains of Primitive Fishes
   3.1. Agnathans (Hagfishes and Lampreys)
   3.2. Sarcopterygians (Lobe‐Finned Fishes)
   3.3. Actinopterygians (Early Ray‐Finned Fishes)
4. Functional Classification of Cranial Nerves in Fishes
5. The Visual System
   5.1. The Optical Apparatus
   5.2. Retina and Visual Function
   5.3. Spectral Filters
   5.4. Visual Sensitivity
   5.5. Visual Resolution
   5.6. Visual Input to the CNS
   5.7. Nonvisual Photoreception
6. Chemoreceptive Systems
   6.1. Olfaction
   6.2. Gustation
   6.3. Solitary Chemoreceptor Systems
   6.4. Common Chemical Sense
7. Octavolateralis System
   7.1. Audition
   7.2. Vestibular Control
   7.3. Lateral Line
8. Electroreception
   8.1. Structure, Function, and Evolution of Ampullary Receptors
   8.2. Role in Passive Electrolocation
9. Concluding Remarks

   The nervous and sensory systems of lampreys, hagfishes, the coelacanth,
lungfishes, and the basal ray‐finned fishes, such as bichirs and reedfishes,
paddlefishes and sturgeons, garfishes and bowfins, are reviewed and compared

Primitive Fishes: Volume 26                          Copyright # 2007 Elsevier Inc. All rights reserved
FISH PHYSIOLOGY                                                 DOI: 10.1016/S1546-5098(07)26003-0
122                                                           SHAUN P. COLLIN

with closely related groups of ‘‘primitive’’ fishes. Anatomical, physiological,
molecular, and behavioral data are discussed in relation to both ecological and
phylogenetic relationships. In addition to overviews of their embryology,
development, and gross anatomy, a functional classification of the cranial
nerves is also presented. The peripheral and central components of the visual
(including nonvisual photoreception), chemoreceptive (olfaction and gusta-
tion), octavolateralis (audition and lateral line), and electroreceptive systems
are examined in some detail, highlighting the physiological basis for behavior
wherever possible. Although a number of neuroanatomical, neurochemical,
physiological, and molecular studies have greatly enhanced our understanding
of brain evolution in these important groups, physiological studies are remark-
ably still scarce and need to be undertaken in order to trace the origins of
craniate brains and the evolutionary constraints placed on neural plasticity and


    The central nervous system (CNS) and sensory systems of ‘‘primitive’’
fishes have received considerable attention, and these studies provide crucial
insights into the evolution and origins of the structure and function of the
nervous systems of vertebrates. As in more advanced fishes, the peripheral
sense organs and their central input are under intense selection pressure
according to their ecological niche and the evolutionary constraints on their
sensory and motor lifestyles. The aquatic environment is diVerent to the
terrestrial environment in its ability to convey sensory information, and the
physical environment has been shown to play a major role in the propagation
and reception of signals.
    This chapter will concentrate on the nervous and sensory systems of the
Agnatha (Myxiniformes, hagfishes and Petromyzontiformes, lampreys),
the Sarcopterygii or lobe‐finned fishes (Coelacanthiformes, coelacanth
and the Dipnoi, lungfishes), and the basal Actinopterygii (Polypteriformes,
bichirs and reedfishes; Acipensiformes, paddlefishes and sturgeons; Semiono-
tiformes, garfishes; and Amiiformes, bowfins), although reference will be made
to other closely related groups in order to reveal phylogenetic trends in the
evolution of specific neural characters. Knowledge of these ‘‘living fossils’’ is
really our only window into the physiology and behavior of the ancestral
vertebrates and the processes of brain evolution that have shaped the diversity
of the nervous systems in extant species.
    In reviewing the research done on the nervous and sensory systems in
primitive fishes, it became apparent that an appreciable wealth of anatomical
knowledge exists, with little or no physiological studies for many species.
3.   NERVOUS AND SENSORY SYSTEMS                                              123

Therefore, anatomy forms the basis of some of the sections, although
structure–function relationships are discussed wherever possible. Special
emphasis is placed on the few physiological studies and the physiological
mechanisms influencing behavior.


    In anamniotes, the embryo is dependent on nutrition from either a large
yolk sac within the egg (oviparous fishes) or via a direct placenta‐like con-
nection to the mother (viviparous fishes). Cleavage divisions in the animal
pole exceed those in the vegetal pole but eventually the blastomeres migrate
and surround the yolk. The dorsal margin of the embryo thickens and
gastrulation takes place, where mesodermal cells move inside the ectodermal
layer and migrate back toward the animal pole. At this stage, the ectoderm
begins to fold onto itself to form the neural tube, which gives rise to nearly all
the neurons and glia. Invading ectoderm then forms a continuous layer over
the neural tube, which is sealed oV at both ends. The lumen forms the
ventricles of the brain (rostrally) and the central canal of the spinal cord
(caudally). The undiVerentiated nerve cells within the neural crest grow
processes, which either grow out to the periphery (aVerents) or invade
the neural tube (eVerents). The evolution of these neural crest cells and the
subsequent development of complex head structures are thought to be a
major reason for the success of vertebrates (Gans and Northcutt, 1983).
The development of oligodendrocytes (centrally) and Schwann cells (periph-
erally) at this time is also thought to underlie the success of jawed vertebrates.
These cells produce myelin to ensheath the neural processes (axons) and
increase the conduction velocity of neural impulses, a distinct advantage
over the jawless protochordates (lancelets) and agnathans (hagfishes and
lampreys) (Zalc and Colman, 2000).
    The rostral portion of the neural tube enlarges and diVerentiates to form
three primary vesicles; the prosencephalon, the mesencephalon, and the
rhombencephalon. The prosencephalon subdivides to form the telencephalon
and the diencephalon (thalamus and hypothalamus), and after flexion, the
rhombencephalon or hindbrain further subdivides to form the metence-
phalon (pons and cerebellum) and the myelencephalon (medulla oblongata).
The mesencephalon becomes the optic tectum and tegmentum. The identity
and boundaries of segmental units or compartments of the vertebrate brain
(neuromeres) depend on cell surface molecules such as Eph and a number of
transcription factor‐encoding genes such as Hox and Krox20 (see review
by Murakami et al., 2005). A comparison of the expression patterns of the
Pax6 gene in the dogfish, Scyliorhinus canicula, and the lamprey, Lampetra
124                                                                          SHAUN P. COLLIN

fluviatilis during neurulation, reveals that this gene plays an important role in
brain regionalization in primitive fishes and that this has been conserved
during vertebrate evolution (Derobert et al., 2002).


3.1. Agnathans (Hagfishes and Lampreys)

    A number of studies have concentrated on establishing the morphotype of
the craniate brain based on the brains of both hagfishes and lampreys (Wicht
and Northcutt, 1992, 1994; Fritzsch and Northcutt, 1993; Northcutt, 1996;
Wicht, 1996). The neuroanatomy of these jawless fishes has also received
much attention (Bone, 1963; Braun, 1996; Northcutt, 1996; Niewenhuys and
Nicholson, 1998; Wicht and Niewenhuys, 1998). The hagfish brain comprises
telencephalic, diencephalic, mesencephalic, and rhombencephalic divisions,
although the presence of a metencephalic/cerebellar division is questionable
in lampreys (Wicht, 1996) (Figure 3.1A–D). There is no cerebellum in hagfishes

Fig. 3.1. (A and B) The brain of the Atlantic hagfish, M. glutinosa in dorsal (A) and lateral (B)
views. (C and D) The brain of the river lamprey, L. fluviatilis in dorsal (C) and lateral (D) views.
The tela choroidea are removed on the left‐hand side. [(A) and (B) adapted from Wicht and
Niewenhuys (1998).] [(C) and (D) adapted from Niewenhuys and Nicholson (1998).] Ac, auricula
cerebelli; Cc, corpus cerebelli; D, diencephalon; Ht, hypothalamus; M, mesencephalon; OB,
olfactory bulbs; OT, optic tectum; Po, pedunculus olfactorius; R, rhombencephalon; S, spinal
cord; and Tel, telencephalon. Abbreviations apply to Figures 3.1–3.3. Reproduced with kind
permission of Springer Science and Business Media.
3.   NERVOUS AND SENSORY SYSTEMS                                             125

(Figure 3.1A and B). The rhombencephalon is divided into dorsal (visceral and
somatosensory) and ventral (branchiomotor and somatomotor) zones. The
spinal cord has both ascending and descending projections between the mesen-
cephalic and rhombencephalic areas. In myxinoids, the brain is large and
approaches the high encephalization quotients of teleosts and amphibians
(Platel and Delfini, 1981). The cerebellum and choroid plexus are lacking,
and in hagfishes, sulci do not always demarcate the boundaries of brain divi-
sions. The narrow optic nerves form a chiasm and decussate within the brain
rather than occur externally. The ventricular system is almost obliterated by the
compression of the brain in the rostrocaudal axis (Wicht and Niewenhuys,
    The brains of lampreys are slender and possess a well‐developed ventri-
cular system and a choroid plexus in the roof of the mesencephalon, a unique
feature not found in other craniates. The ratio of brain to body weight is also
the lowest among craniates (Platel and Delfini, 1986). The spinal cord is
hypertrophied in lampreys, is ribbonlike, and extends throughout the length
of the vertebral canal (Figure 3.1C and D). The forebrain of hagfishes and
lampreys diVers appreciably. In lampreys, the telencephalic hemispheres are
formed from the paired lateral evaginations of the rostral neural tube, while
in hagfishes, the forebrain is composed of paired anterior and posterior
evaginations from the prosencephalic vesicle, which include the olfactory
bulb and the pallial and subpallial grisea (Wicht, 1996).

3.2. Sarcopterygians (Lobe‐Finned Fishes)
    The lobe‐finned fishes comprise the lungfishes and the coelacanth. The
three main groups of lungfishes are the African, for example, Protopterus, the
South American Lepidosiren, and the Australian Neoceratodus. All lungfish
brains show little histological diVerentiation and are considered ‘‘simple’’
compared to other primitive fishes and teleosts (Striedter, 2005). The spinal
cord is cylindrical and the rhombencephalon resembles that of urodele
amphibians. The cranial nerves reveal a pattern similar to that of other
gnathostomatous fishes, although the anterior lateral line nerve is divided
into two branches; pars dorsalis and pars ventralis (Niewenhuys, 1998a). The
mesencephalic optic tectum has retained its embryonic, tubelike appearance.
The telencephalon consists of three parts: the telencephalon impar, the
cerebral hemispheres, and the olfactory bulbs. The brains of Protopterus
and Lepidosiren (Figure 3.2A and B) are very similar and diVer to that of
Neoceratodus. In Neoceratodus forsteri, the olfactory bulbs are connected to
the cerebral hemispheres by short hollow peduncles instead of being sessile as
in the other two groups (Niewenhuys, 1998a). Neural development and a
range of neural characters have been useful in clarifying phylogenetic
126                                                                        SHAUN P. COLLIN

Fig. 3.2. A, B. The brain of the South American lungfish, Lepidosiren paradoxa in dorsal (A) and
lateral (B) views. The tela choroideae are removed on the left‐hand side. (C and D) The brain of
the coelacanth, L. chalumnae in dorsal (C) and lateral (D) views. [(A) and (B) are adapted from
Niewenhuys (1998a) and (C) and (D) are adapted from Niewenhuys (1998b).] For abbreviations
see Figure 3.1. Reproduced with kind permission of Springer Science and Business Media.

relationships between this group and tetrapods (Bemis and Burggren, 1986;
Northcutt, 1986b; Kemp, 2000).
     The coelacanth, Latimeria chalumnae, has a small brain occupying
less than 1% of the total endocranial volume and appears superficially to
comprise a mixture of neural features from both teleosts and cartilaginous
fishes (Millot and Anthony, 1965; Striedter, 2005). Sitting within the otico‐
occipital division of the braincase and covered with adipose tissue, the brain
is long and slender and gives rise to very elongate cranial nerves and
‘‘stretched’’ olfactory peduncles (Niewenhuys, 1977, 1998b; Northcutt
et al., 1978) (Figure 3.2C and D). The spinal cord has slight pelvic and
3.   NERVOUS AND SENSORY SYSTEMS                                               127

pectoral thickenings associated with the species’ extraordinarily mobile,
paired fins. All the major brain divisions can be identified, especially the
cerebellum, which is well developed (diVerentiated into a corpus cerebelli and
a lobus vestibulolateralis). The mesencephalon is small and not well diVer-
entiated. The olfactory bulbs are greatly separated, apposed to the olfactory
mucosa, and connected to the telencephalon by slender stalks (Niewenhuys,
1998b) (Figure 3.2C and D). An elegant reconstruction of the cranial nerves
of the prenatal coelacanth reveals that (1) the optic nerves are interdigitated
and therefore provide a partial decussation of retinal input to the brain; (2)
the profundal ganglion and ramus are separate from the trigeminal system in
contrast to hagfishes, lampreys, lungfishes, and tetrapods; and (3) there are
three postotic lateral line nerves (Northcutt and Bemis, 1993).

3.3. Actinopterygians (Early Ray‐Finned Fishes)

     The early ray‐finned fishes are a collection of bony fishes: the Polypter-
iformes or bichirs and reedfishes (Polypterus and Erpetoichthys formerly
Calamoichthys), the Acipenseriformes or sturgeons and paddlefishes (Acipen-
ser, Polyodon, and Psephurus), the Semionotiformes or garfishes (Lepisos-
teus), and Amiiformes or bowfins (Amia). According to Northcutt and
Braford (1980), Polypterus possesses a higher brain weight to body weight
ratio than the other basal actinopterygians. In this genus, the rhombenceph-
alon is well developed and the cerebellum is not fused as in most other
vertebrates but is paired, connected only by a thin lamella (Niewenhuys,
1998c). The habenular ganglia are asymmetrical with the right ganglion
larger than the left. The telencephalon is large and the two elongated hemi-
spheres are connected by the anterior commissure. Bichirs appear to retain
all six pairs of lateral line nerves that characterize the earliest gnathostomes
and, unlike most non‐teleost bony fishes, have lost the spiracular organ
(Piotrowski and Northcutt, 1996). The hypobranchial nerve of bichirs is
formed by only two spinal nerves, a pattern seen also in gars.
     The rhombencephalon, hypothalamus, and cerebellum are all well devel-
oped in the Acipenseriformes. The rhombencephalon contains the termi-
nation of fibers from the viscerosensory, somatosensory, electroreceptive,
lateral line, auditory, and vestibular systems. The telencephalon is large
and of the everted type, that is the walls have recurved laterally during
ontogenesis (Niewenhuys, 1998d; Striedter and Northcutt, 2006). The rela-
tively minimal development of the mesencephalic tectum in Acipenser nac-
carii (Figure 3.3) is a derived characteristic that is shared with other species of
sturgeon, suggesting that the visual system is not well developed in this group
(Vazquez et al., 2002). In the developing paddlefish, Polyodon spathula,
the brain fills the braincase in larval forms but not in larger fish, and as
128                                                                         SHAUN P. COLLIN

Fig. 3.3. (A and B) The brain of the reedfish, E. calabaricus in dorsal (A) and lateral (B) views.
The tela choroideae are removed on the left‐hand side. (C and D) The brain of the shovelnose
sturgeon, Scaphirhynchus in dorsal (C) and lateral (D) views. The tela choroideae are removed on
the left‐hand side. [(A) and (B) adapted from Niewenhuys (1998c) and (C) and (D) adapted from
Niewenhuys (1998d).] For abbreviations see Figure 3.1. Reproduced with kind permission of
Springer Science and Business Media.

the rostrum protrudes forward, the position of the brain, barbels, and eyes all
change (Larimore, 1949).
    The garfishes and bowfins possess similar brains with relatively small
cerebella and optic tecta, when compared to the brain of teleosts, that is
Salmo, although gars have a higher brain weight to body weight ratio than
most other primitive fishes (Striedter, 2005). A number of studies tracing the
3.   NERVOUS AND SENSORY SYSTEMS                                              129

innervation of the brain stem (Niewenhuys and Pouwels, 1983), lateral line
(Song and Northcutt, 1991a), retina (Northcutt and Butler, 1976; Butler and
Northcutt, 1992; Collin and Northcutt, 1995), and tectum (Northcutt, 1982)
have previously been published and are reviewed in Meek and Niewenhuys


     Cranial nerves are components of the peripheral nervous system
that connect directly to the brain rather than the spinal cord (Figure 3.4A
and B). The numbers assigned to a cranial nerve normally indicate both
its location on the brain stem in the rostro‐caudal direction and its function.
The cranial nerves of fishes are classified as primarily sensory (carrying
somatic sensory information including touch, pressure, vibration, tempera-
ture, or pain), special sensory (carrying the sensations of smell, sight, hearing,
or balance), motor (carrying information to somatic muscle), or mixed
(comprising sensory and motor axons). The size and development of the
cranial nerves are directly related to their importance and organizational
     A total of 20 cranial nerves have now been identified in vertebrates,
although some of these are not found in fishes, i.e. vomeronasal (Butler
and Hodos, 1996) (Figure 3.4A and B). The olfactory nerve (nI) carries
sensory information from the olfactory epithelium situated at the base of
the nares into the olfactory bulbs where they aggregate with axons sensitive
to the same water‐soluble stimuli within glomeruli. Olfactory information is
then carried to the telencephalon via the olfactory tract. The optic nerve (nII)
contains the axons of ganglion cells located within the inner retina, which
terminate in the hypothalamus, thalamus, pretectum, and optic tectum. The
optic nerve may be pleated in some species (Collin, 1989). The epiphyseal
nerve carries aVerent and eVerent axons from the epithalamus (epiphysis and
habenular nucleus) to the diencephalon and midbrain. The terminal nerve is
closely associated, but separate to, the olfactory nerve input and projects
from a ganglia, situated generally between the olfactory bulbs and the
telencephalon, to a large range of sites throughout the CNS including the
retina. Terminal nerve fibers may be luteinizing hormone‐releasing hormone
(LHRH) immunoreactive and may be involved in reproductive behaviors.
Six extraocular muscles are innervated by the oculomotor (nIII), the trochlear
(nIV), and the abducens (nVI) motor nerves, which appear to be found in all
primitive fishes except hagfishes (with no extraocular eye muscles). Interest-
ingly, lampreys possess only five extraocular eye muscles. The complement of
130                                                                                 SHAUN P. COLLIN


                                                                                        1 mm

          B                                                                                    d

                                                                MLLN   STL
                                                        vr                   PLLN             pd
                                                                             vX               pv

                                     buc max

                                                                                           2 mm

Fig. 3.4. (A) Photograph of the cleared head of a juvenile Senegal bichir, P. senegalus stained
with Sudan black to reveal the peripheral course of the cranial nerves. (B) Camera lucida drawing
of the course of the cranial nerves seen in (A). buc, buccal ramus of anterodorsal lateral line
nerve; d, dorsal ramus of posterior lateral line nerve; dr, dorsal root of anterodorsal and
anteroventral lateral line nerves; gAD, sensory ganglion of anterodorsal lateral line nerve; gV,
sensory ganglion of trigeminal nerve; hy‐op, hyo‐opercular ramus of anteroventral lateral line
and facial nerves; hym, hyomandibular trunk of anteroventral lateral line and facial nerves;
mAV, mandibular ramus of anteroventral lateral line nerve; max, maxillary ramus of trigeminal
nerve; MLLN, middle lateral line nerve; mV, mandibular ramus of trigeminal nerve; mVII,
mandibular ramus of facial nerve; pd, pars dorsalis of lateral ramus of posterior lateral line
nerve; PLLN, posterior lateral line nerve; Pr, profundal nerve; pv, pars ventralis of lateral ramus
of posterior lateral line nerve; so, superficial ophthalmic ramus of anterodorsal lateral line
nerve; ST, supratemporal sensory canal; STL, supratemporal lateral line nerve; VIII, octaval
nerve; vr, ventral root of preoptic lateral line nerve; vx, visceral trunk of vagal nerve; X, vagal nerve.
[Adapted from Piotrowski and Northcutt (1996) with kind permission of S. Karger AG, Basel.]
3.   NERVOUS AND SENSORY SYSTEMS                                            131

extraocular eye muscles in gnathostomes is four recti (inferior, superior,
lateral, and medial) and two oblique (inferior and superior) muscles, which
control eye movements and the retraction of the nictitating membrane where
present, that is, in cartilaginous fishes.
    The sensations of pain, temperature, touch, and proprioception from the
skin and muscles of the head and jaws are conveyed to the CNS via the
trigeminal (nV) nerve. Three major branches (ophthalmic, maxillary, and
mandibular) of the trigeminal nerve terminate within either the descending or
principal nuclei. In the coelacanth, the mucosal walls are innervated by the
profundus nerve (a homologue of the ophthalmic branch of the trigeminal).
Taste buds are often distributed within the oral cavity, pharynx, gills, and
over the barbels and skin of the body. The responses of gustatory fibers
conveyed from taste cells are carried by three diVerent nerves (the ventrolat-
eral regions of the facial nVIIVL, glossopharyngeal nIXVL, and vagus nXVL)
and terminate within the two divisions of the nucleus solitarius, that is,
rostral (also called the gustatory nucleus) and caudal regions. The input to
these subdivisions is topographic. The dorsal component of the facial nerve
(nVD) contains motor axons and innervates the superficial muscles of the
head, including muscles of the cheeks, lips, and nares. Ingestion of food
into the mouth is controlled by the motor component of nV and the dorsal
component of nVII. Swallowing is controlled by inputs from the dorsal glosso-
pharyngeal (NIXD), dorsal vagus (nXD), and accessory nerves (nXI), which
convey eVerent axons to the pharynx and palate from the nucleus ambiguous in
the medulla.
    The octaval or vestibulocochlear nerve (nVIII) comprises two compo-
nents (auditory and vestibular), which are found in all primitive fishes and
terminate primarily in the octaval column but also the cerebellum and
the reticular formation (but not in lampreys). In fishes, the octaval column
comprises at least four nuclei: anterior, magnocellular, descending, and
posterior, and receives information from both the auditory and vestibular
systems. In hagfishes, the eighth cranial nerve projects primarily to the ventral
nucleus of the area acousticolateralis. Unique to aquatic vertebrates and
all primitive fishes is the lateral line, which detects either water movements
(mechanosensory) or electrical fields (electrosensory) using neuromasts
and/or pit organs and either ampullary and/or tuberous organs, respectively.
Input from these two systems is mediated by up to six nerves: antero-
dorsal, anteroventral, otic, middle, supratemporal, and posterior lateral line
nerves. The mechanosensory fibers of the lateral line terminate within two
regions (nucleus medialis and nucleus caudalis), although in electroreceptive
non‐teleost gnathostomes, a third region also receives input (nucleus dorsalis).
In electroreceptive teleost fishes, the input terminates in a long electrosensory
lateral line lobe within the dorsal zone of the medulla.
132                                                           SHAUN P. COLLIN


5.1. The Optical Apparatus

     All primitive fishes possess single‐chambered eyes, which scale positively
with body size, although primitive fishes with long bodies, that is lampreys
(Petromyzon marinus) and reedfishes (Erpetoichthys calabaricus), do not obey
the same allometric relationships as other fishes (Howland et al., 2004). With
the exception of some species of hagfishes (i.e., Paramyxine and Myxine sp.),
which possess degenerate eyes and lack a cornea, iris, and lens (Holmberg,
1977), the eyes of primitive fishes all possess a cornea and lens and focus light
onto a diVerentiated retina. The cornea is the first optical interface and acts
as a protective goggle and is composed predominantly of collagen fibers
(stroma) interposed between an epithelium and an endothelium. The fish
cornea confers little, if any, refractive power due to the comparable refractive
indices of the cornea and the surrounding aquatic media. In some primitive
fishes, that is lampreys (van Horn et al., 1969), the cornea is split into dermal
(continuous with the skin) and scleral (continuous with the eyecup) compo-
nents, which allows the underlying globe to move freely. The evolution of
the dermal cornea or secondary spectacle is thought to streamline the head
and, in benthic species, protects the eye from abrasion. Microprojections
(microridges, microplicae, and microvilli) extend from the surface of the
superficial corneal epithelial cells and stabilize the tear film to provide an
optically smooth interface (Collin and Collin, 2006). In lampreys and lung-
fishes, microholes in the cornea constitute the surface openings of large
mucus‐secreting cells. The mucus provides a protective coating during
burrowing and estivation (Collin and Collin, 2001).
     Colored pigments within the cornea and lens act as short wavelength‐
absorbing filters in some primitive fishes (the sea lamprey, P. marinus; the
Australian lungfish, N. forsteri; and the bowfin, Amia calva) to minimize
chromatic aberration and to tune the light reaching the retinal photoreceptors.
In agnathans and elasmobranchs, sutural fibers inhibit the corneal stroma from
swelling in response to low temperatures and changes in osmotic pressure
associated with moving between saltwater and freshwater. Multilayered stacks
of materials (such as connective tissue, modified rough endoplasmic reticulum,
collagen fibrils, and cytoplasmic plates) with diVerent refractive indices
are common inclusions in some teleost and non‐teleost fishes, that is the bowfin
A. calva (Munk, 1968), and produce iridescence (Collin and Collin, 2001). In
the sturgeon Acipenser sp. and the garfish Lepisosteus platyrhinchus, the cornea
is supported by a loose meshwork of cellulofibrous tissue (or annular ligament)
to support the iridocorneal angle (Collin and Collin, 1993).
     In water, and in the absence of any corneal refraction, the refractive
power of the fish eye lies solely with the spherical lens. Although it is assumed
3.   NERVOUS AND SENSORY SYSTEMS                                              133

that all primitive fishes adhere to this general mechanism, the lenses of
downstream transformers of the sea lamprey, P. marinus, exhibit positive
spherical aberration, while adults are corrected for spherical aberration
(Bantseev et al., 2005). In biological lenses, short wavelengths of light are
refracted more than long wavelengths of light (chromatic aberration). This
results in blue light focusing closer to the lens than red light. However,
interruptions to the otherwise smooth refractive index gradients of the lens
in lampreys, lungfishes, and teleosts bring the focal length of the light
spectrum back to a single focus on the retina. These multifocal lenses allow
the incident light to be focused on a single layer of retinal photoreceptors,
thereby optimizing the capture of light by multiple cone types involved in
mediating color vision (Kroger et al., 1999). The size of the pupillary aperture
can improve the level of spherical aberration by reducing the number of rays
passing through the lens periphery. The shape and size of the pupil varies
across fishes from a circular to a U‐ or W‐shaped with multiple apertures to a
slit. Lampreys possess little, if any, active movement of the iris, while most
elasmobranchs elicit rapid changes in pupil shape in response to ambient
light levels.
     Accommodatory mechanisms have been subject to appreciable selection
pressure in primitive fishes and vary across many of the groups. Many hag-
fishes do not possess a lens and in both hagfishes and lampreys, the eyes do not
possess intraocular eye muscles, a ciliary ganglion, and an Edinger‐Westphal
nucleus. However, in other agnathans (lampreys), a cornealis muscle lying in
the head next to the eye is thought to retract the cornea, and thereby the
closely apposed lens, toward the retina during accommodation. However,
stimulation of the cornealis muscle does not elicit lens movement (Sivak and
Woo, 1975), and other static forms of accommodation may provide a focused
image on the retina. In lampreys, some batoid elasmobranchs, and developing
teleosts, the eyes are not symmetrical, where the dorsal retina sits closer to the
lens than the ventral retina. This ‘‘ramped retina’’ allows both near and far
objects to be focused on the retina simultaneously. Both the garfish, Lepisos-
teus osseus oxyurus, and the bowfin, A. calva, possess a retractor lentis muscle,
which is responsible for accommodatory lens movement in the same direction
(toward the retina) as in teleosts (Sivak and Woo, 1975). Cartilaginous fishes
possess a protractor lentis muscle within the ventral papillae of the ciliary
body, which moves the spherical lens anteriorly (toward the cornea) accom-
modating for near objects.

5.2. Retina and Visual Function
    With the exception of the hagfish retina, which is undiVerentiated, the
retina of both primitive and advanced fishes possesses all of the major retinal
neurons found in other vertebrates. These include photoreceptors, horizontal,
134                                                              SHAUN P. COLLIN

bipolar, amacrine, and ganglion cells, which are arranged into three nuclear
layers and two plexiform (synaptic) layers. In lampreys, most (75%) of
the ganglion cells lie within the inner nuclear layer. Axons of these ‘‘ectopic’’
ganglion cells join those lying within the ganglion cell layer to exit the retina at
the boundary of the inner nuclear and inner plexiform layers at the optic nerve
head, eVectively negating the blind spot (Fritzsch and Collin, 1990; Fritzsch,
1991). In all other primitive fishes, most of the ganglion cells lie within the
ganglion cell layer, with a small proportion ‘‘displaced’’ to the inner nuclear
layer, and their axons traverse the retina within the nerve fiber layer abutting
the inner limiting membrane. A great deal of work still remains to be done on
the evolution of the inner retina in primitive fishes, but it appears that
lampreys possess a population of biplexiform ganglion cells that make direct
connections with the photoreceptors and inner nuclear layer cells (Fritzsch
and Collin, 1990).
    The photoreceptors within the outer retina of a range of extant primitive
fishes have received more attention. Their visual pigments phototransduce
light energy into electrical impulses that trigger a cascade of enzymatic
reactions that amplify the signal and ultimately change the rate of neuro-
transmitter release from their synaptic terminals. The signals are conveyed to
the ganglion cells via bipolar interneurons. The visual pigments comprise a
chromophore based on either vitamin A1 (rhodopsin) or A2 (porphyropsin)
covalently bonded to an opsin protein, composed of seven transmembrane
a‐helices embedded within the outer segment discs. The amino acid sequence
of the opsin protein and the type of chromophore used determines the
spectral sensitivity/tuning of the visual pigment and therefore the range of
wavelengths to which an animal is sensitive. The possession of an A2‐based
visual pigment by the pouched lamprey, Geotria australis, at the time this
species enters the sea, contrasts with the situation in the comparable stage of
P. marinus and in marine teleosts and elasmobranchs, which generally have
vitamin A1‐based visual pigments (rhodopsins). Interestingly, during the
upstream migration of P. marinus, the chromophore becomes A2‐based
(Harosi and Kleinschmidt, 1993), as is typically the case in freshwater tele-
osts. The photoreceptors of both downstream and upstream migrants of the
Pacific lamprey, Entosphenous tridentatus, possess a vitamin A1‐based photo-
pigment, typical of rhodopsin (Crescitelli, 1972). When considered together
with the finding of the entire complement of visual pigments incorporating a
chromophore based on vitamin A2 in the anadromous white sturgeon
Acipenser transmontanus, it appears that migration between freshwater and
saltwater may not be suYcient to induce a paired A1/A2 visual pigment
system (Whitmore and Bowmaker, 1989; Sillman et al., 1995). Given the
interspecific variability, it is premature to predict whether rhodopsin or
3.   NERVOUS AND SENSORY SYSTEMS                                          135

porphyropsin is the ancestral photopigment. However, it is interesting to
note that only vitamin A1, and not A2, could be isolated from the liver in
Myxine glutinosa (Vigh‐Teichmann et al., 1984).
     Rods mediate dim light (scotopic) vision and cones mediate bright
light (photopic) vision. In jawed vertebrates, rod visual pigments are
classified as Rh1, while cone visual pigments fall into four classes (long
wavelength‐sensitive or LWS, ultraviolet (UV)‐sensitive or SWS1, blue wave-
length‐sensitive or SWS2, and medium wavelength sensitive or Rh2). The
presence of more than one cone type each with a diVerent spectral sensitivity
provides the basis for color vision. The eyes of present day hagfishes, such as
M. glutinosa and Eptatretus stoutii, are poorly developed and lie beneath
muscle (M. glutinosa) or unpigmented skin (E. stoutii) (Locket and
Jorgensen, 1998). On morphological criteria, only a single rodlike photore-
ceptor has been identified in M. glutinosa (Locket and Jorgensen, 1998),
although Vigh‐Teichmann et al. (1984) found two unidentified classes of
outer segments that could be distinguished on immunocytochemical, but
not ultrastructural, criteria. Despite the degenerative state of the retina in
hagfishes, M. glutinosa and Eptatretus burgeri both respond to changes
in illumination by active locomotory movements, possibly ensuring that
it remains buried during the day or maintains its circadian rhythm of activity
(Kabasawa and Ooka‐Souda, 1989). However, the relative contributions of
the retinal photoreceptors in the eye and those situated in the cloacal region
(Myxine sp.) or the tail (Lampetra sp., Young, 1935; Ronan and Bodznick,
1991) still need to be assessed.
     The retinae of a number of northern hemisphere (or holarctic) lampreys
have been examined morphologically and the consensus is that two photore-
ceptor types exist, a short and a long, and are putatively a rod and a cone
receptor, although the classification of these receptors has been a subject
of contention for many years (reviewed in Crescitelli, 1972; Collin et al.,
1999) (Figure 3.5A). On the basis of the continuity of the outer segment discs
with the extracellular matrix and a pattern of protein labeling throughout the
outer segment (following incorporation of radioactively labeled amino
acids to indicate the process of outer segment disc renewal), both receptor
types were considered cones in the sea lamprey, P. marinus, by Dickson and
Graves (1979). In contrast, the southern hemisphere lamprey, G. australis,
possesses five morphologically distinct photoreceptor types, all of which
possess cone‐like characteristics and contain a diVerent visual pigment
(LWS, SWS1, SWS2, RhA, and RhB), providing the basis for pentachro-
matic vision (Figure 3.5B). Three of these opsin genes are orthologous to the
visual pigment classes of gnathostomatous (jawed) vertebrates, but the other
two (RhA and RhB) have evolved by an independent gene duplication event,
136                                                                                                                                                                        SHAUN P. COLLIN

                                                 1.2                                                                                           1.2

                 Normalized quantal

                                                                                         Normalized quantal
                                                                              — 507 nm                                                                                 — 503 nm
             A                                   1.0                          — 552 nm                                                         1.0                     — 517 nm


                                                 0.8                          — 614 nm                                                         0.8                     — 618 nm
                                                 0.6                                                                                           0.6
                                                 0.4                                                                                           0.4
                                                 0.2                                                                                           0.2
                                                   0                                                                                             0
                                                   350      450  550     650     750                                                             350   450  550     650    750
                                                              Wavelength (nm)                                                                            Wavelength (nm)
                           Relative absorbance

                                                                                                   Normalized absorbance Relative absorbance

                                                                  PB                                                                                         PB

                                                             BL                                                                                             BL
                                                    400        500        600                                                                    400      500        600
                                                 1.2          Wavelength (nm)                                                                            Wavelength (nm)
                           Relative absorbance

                                                 1.0                                                                                1.2
                                                            SR           LR
                                                 0.8                                                                                1.0
                                                 0.6                                                                                0.8
                                                 0.4                                                                                0.6
                                                 0.2                                                                                0.4
                                                   0                                                                                0.2
                                                      400     500   600    700                                                        350              450   550     650     750
                                                              Wavelength (nm)                                                      −0.2                 Wavelength (nm)

             B                                                                                         Catfish Rh1
                                                                                                         Dogfish Rh1           99 100
                                                                                        36            Skate Rh1
                                                                                                        Chameleon Rh1
                                                                                       99              Coelacanth Rh1
                                                                                                        Goldfish Rh1
                                                                                                      Dunnart Rh1
                                                                                   92               Mouse Rh1
                                                                                            100     Human Rh1
                                                                                                       Coelacanth Rh2
                                                                                 100                      Goldfish Rh1-1
                                                                                      100          100     Goldfish Rh1-2
                                                                                         79             Chameleon Rh2
                                                                                                     Geotria RhB
                                                                           96                   Geotria RhA
                                                                                      100       Lamprey Petromyzon RhA
                                                                                        100     Lamprey Lampetra RhA
                                                                                                   Geotria SWS2
                                                                              100                         Goldfish SWS2
                                                                                  93                  Newt SWS2
                                                                                    81                Chameleon SWS2
                                                                                                 Geotria SWS1
                                                                                                    Frog SWS1
                                                                                               Chameleon SWS1
                                                                                   96               Wallaby SWS1
                                                                                       100          Mouse SWS1
                                                                                           97      Human SWS1
                                                                                                     Geotria LWS
                                                                                          84            Goldfish LWS
                                                                                   100                  Chameleon LWS
                                                                                        97           Wallaby LWS
                                                                                          65           Mouse LWS
                                                                                            100      Human LWS-R
                                                                                                100 Human LWS-G
                                                                                                          Fruitfly Rh4

Fig. 3.5. Spectral sensitivity of lamprey photoreceptors. (A) Upper two panels: summary of the
calculated photoreceptor quantal spectral sensitivity curves for three of the five types of photo-
receptors in downstream (left) and upstream (right) migrants of G. australis. Middle two panels:
3.   NERVOUS AND SENSORY SYSTEMS                                                               137

which occurred within the agnathan lineage (Collin et al., 2003b)
(Figure 3.5B). Given the absence of the Rh1 opsin gene within the agnathan
lineage, scotopic vision may have evolved exclusively within the gnathos-
tomatous lineage prior to the evolution of the cartilaginous fishes.
The functional characterization of rods and cones requires further examina-
tion in the early vertebrates at both the physiological (photokinetics) and
the biochemical (phototransduction genes) levels. Interestingly, the lamprey
visual pigments react with hydroxylamine in a similar manner to cone
pigments (Hisatomi et al., 1988), they do not saturate at high light inten-
sities (Govardovskii and Lychakov, 1984) and appear to be designed for
photopic vision.
     Multiple cone types have been retained within the cartilaginous (three
cone types in the shovelnose rays, Hart et al., 2004), dipnoan (four cone
types in the Australian lungfish, Bailes et al., 2006), early ray‐finned (up to
three cone types, that is, in the sturgeon and paddlefish, Munk, 1964, 1968,
1969; Sillman et al., 1999), and teleostean (up to seven cone types in cichlids,
Parry et al., 2005) fishes. Color‐coding mechanisms and various subtypes of
color opponent horizontal cells are also present in primitive fishes, that is in
the bowfin, A. calva, the shortnose garfish, Lepisosteus platostomus, and
the Siberian sturgeon, Acipenser baeri (Burkhardt et al., 1983; Gottesman
and Burkhardt, 1987). Therefore, it appears that all vertebrate classes possess
the potential for color vision, but this has not yet been confirmed behavior-
ally in the non‐actinopterygian fishes. The comoran coelacanth appears to
have lost the SWS1, SWS2, and LWS opsin genes and retained the Rh1 and
Rh2 opsin genes that are tuned to detect the full spectrum of light available at
the depth it inhabits (200 m) (Dartnall, 1972; Yokoyama, 2000).

spectral sensitivity curves for the visual pigments in the short (SR) and long (LR) receptor types
in the river lamprey, L. fluviatilis, giving lmax values of 517 and 555 nm, respectively. Lower two
panels: normalized absorbance spectra from the outer segments of the short (SR) and long (LR)
receptors in the sea lamprey, P. marinus, giving lmax values of 525 and 600 nm, respectively (left),
and the relative quantal spectral sensitivity of the whole photoreceptor in M. mordax (lmax
514 nm), based on both the visual pigment and ellipsosome spectra and the dimensions of the
inner and outer segment (right). PB, postbleaching and BL, baseline. [Figure from Collin and
Trezise (2006).] (B) Phylogenetic tree (based on codon‐matched nucleotide sequences compar-
isons) showing the relationships between the opsin genes of G. australis, the northern hemi-
sphere lampreys, Lampetra japonica and P. marinus, representative jawed vertebrates, and an
invertebrate out‐group. See text for explanations of the diVerent opsin groups. The dashed line
indicates the predicted genetic complement of opsins present in the most recent common
ancestor of the jawed and jawless vertebrates, $540 mya. The number at each branch point
reflects its robustness (maximum 100). The scale bar is calibrated in nucleotide substitutions per
site. [Reproduced from Collin et al. (2003b) with kind permission from Science Publishers Inc.]
138                                                            SHAUN P. COLLIN

5.3. Spectral Filters

    Filtering mechanisms in fishes are common and typically comprise the
accumulation of short wavelength‐absorbing pigment within the ocular
media (cornea, lens, and vitreous humor) (Siebeck and Marshall, 2001) and
within the inner regions of some photoreceptor types (myoidal pigment
and oil droplets) (Collin et al., 1999; Vorobyev, 2003). These spectral filters
narrow the absorption spectrum of the visual pigment housed in the outer
segment of the photoreceptors, shift the peak absorption of the visual pig-
ment toward longer wavelengths, and decrease the absorption eYciency.
Accumulations of yellow myoidal pigment exist in lampreys (Collin et al.,
1999), Australian lungfishes (Bailes et al., 2006), and reptiles (Barbour et al.,
2002). Oil droplets exist in the lobe‐finned (lungfish, Robinson, 1994; Bailes
et al., 2006) and early ray‐finned (sturgeon and paddlefish, Sillman, et al.,
1999) fishes and may be red or colorless. Oil droplets allow the discrimination
of more colors under bright light conditions. Intracellular structures resem-
bling oil droplets lie within the photoreceptor inner segments in two species
of lamprey (Collin and Potter, 2000; Collin and Tresize, 2006) and some
teleostean cyprinids (Nag and Bhattacharjee, 1995). Termed ellipsosomes, on
the basis of their elliptical shape, these structures are of mitochondrial origin
and may either contain a heme pigment, thereby acting as a spectral filter
(cyprinids), or lack any light‐absorbing pigment and may act as an
intracellular focusing device (lampreys, Collin and Potter, 2000).

5.4. Visual Sensitivity

    Increased visual sensitivity is often mediated by the adaptive advantage of a
mirror or tapetum located behind the retina. Sensitivity is increased by reflect-
ing light back onto the photoreceptors for enhanced photon capture, an early
invention in vertebrate evolution. Of the 33 species of lampreys described, a
single species (Mordacia mordax) possesses a mixture of reflective needles and
pigment granules within the retinal pigment epithelium, which elicits a yellow
eyeshine (Collin and Potter, 2000). Retinal tapeta are present in the garfishes
and several species of teleosts. These comprise spheres containing astaxanthin,
phenolic compounds, or lipids packed into a hexagonal array within the con-
fines of the retinal pigment epithelial cell membrane. All these spheres are
reflective and elicit a colored reflex produced by diVuse scattering. Choroidal
tapeta typically with guanine as the reflector are found in elasmobranchs
(sharks, skates, rays, and ratfishes), Polypteriformes (bichirs), Semionoti-
formes (gars), Acipenseriformes (sturgeons), Dipnoi (lungfishes), the coela-
canth, and a few nocturnal ray‐finned fishes (Nicol and Arnott, 1973).
In cartilaginous fishes, the choroidal tapetum is occlusable, masking the
mirrored surface with pigment granules in bright light.
3.   NERVOUS AND SENSORY SYSTEMS                                              139

    Another way of increasing sensitivity to a range of ambient light condi-
tions is to diVerentially place the rod photoreceptors adapted for dim light
vision closer to the incident light (at the outer limiting membrane) during
times of low light. Under these conditions, the cone photoreceptors migrate
toward the back of the retina. The opposite occurs in bright light, where the
cones adopt a position at the outer limiting membrane and the rods are
masked by the migration of the melanosomes within the retinal pigment
epithelium. Retinomotor movements are not present in lampreys (Walls,
1942), reedfishes, Calamoichthys calabaricus and Polypterus delhezi, and in
the African lungfishes Protopterus dolloi (PfeiVer, 1968a). In teleosts, not all
photoreceptor types undergo photomechanical movements. There appears to
be a trade‐oV between the level of photomechanical retinomotor movements
(inherently slow) and the evolution of more rapid pupillary movements.
In evolutionary terms, pupillary and retinomotor movements appear to be
most developed within the elasmobranchs and teleosts, respectively.

5.5. Visual Resolution
     In contrast to optimizing sensitivity with either retinomotor movements or
a tapetum, many fishes are specialized for acute vision, sampling a particular
part of their visual field with high spatial resolving power. The lamprey retina
is specialized for acute vision, where ganglion cell densities increase to form an
area centralis in central retina (Fritzsch and Collin, 1990). All of the fishes
examined thus far, irrespective of their phylogenetic origins, possess some
form of retinal specialization, which may be in the form of a concentric
increase in cell density (area centralis) or an elongated increase in cell density
across the retinal meridian (horizontal streak). The silver lamprey, Ichthyo-
myzon unicuspis, possesses an area centralis located in central retina (peak of
over 4000 ganglion cells per mm2), which relocates to the dorsal peripheral
retina (peak of over 3600 ganglion cells per mm2) during growth. In the
Florida garfish, L. platyrhinchus, a pronounced horizontal streak lies across
the ventral meridian of the eye. This specialization subtends the surface of the
water, where this predator preys on live fish with its long snout armed with
needlelike teeth (Collin and Northcutt, 1993). Together with a temporal area
centralis, which receives input from the frontal visual field, much of its visual
field is monitored with increased spatial resolving power without the need for
scanning eye movements. Environmental cues and the symmetry of each
species’ perceived world play a large role in the topography of retinal cells
rather than any phylogenetic relationships.
     The first appearance of a retinal invagination or foveal pit in predatory
fishes appears to be in the euteleosts, that is in the seahorse and the sandlance
(see reviews by Collin, 1997; Collin, 1999; Collin and Shand, 2003). Although
there is little known about the resolution of primitive vertebrate eyes, there
140                                                           SHAUN P. COLLIN

does appear to be a large variation in the type and total number of retinal
ganglion cells in lampreys (12,000–35,000, Vesselkin et al., 1989; Fritzsch and
Collin, 1990), elasmobranchs (100,000, Collin, 1988), and garfishes (80,000
cells, Collin and Northcutt, 1993), reflecting functional diVerences in acuity
and function. The homogeneous distribution of large a‐like ganglion cells
across the retina in L. platyrhinchus suggests that they possess large receptive
fields and are motion sensitive as found in various other species of vertebrates
(Cook et al., 1992; Collin and Northcutt, 1993). The topographic distribution
of rods and cones in the paddlefish, P. spathula, and the Australian lungfish,
N. forsteri (Bailes et al., 2006), reveals that these primitive species lack a
retinal specialization for acute vision, but have increased receptor size to
increase sensitivity, while maintaining chromatic sampling of their visual

5.6. Visual Input to the CNS
    The optical image formed by the visual apparatus is transformed into an
electrical image by the retina, which is conveyed to the visual centers of the
brain via the optic nerve. The optic nerve comprises the axons of the retinal
ganglion cells and eVerent fibers. In lampreys, the optic nerve is avascular
and contains an ependymal core and unmyelinated axons. However, typical-
ly the axons of gnathostomatous (jawed) fishes are myelinated and form
fascicles or bundles subdivided by astroglia. The optic nerves from the left
and right eyes cross (decussate) at the optic chiasm crossing as separate
nerves (most teleosts) or interlacing with each other (cartilaginous fishes
and a few teleosts). Three major fiber tracts are present in virtually all
vertebrates, that is the basal optic tract, the axial or medial optic tract, and
the marginal optic tract (Fritzsch, 1991). In lampreys, 75% of the ganglion
cells are ‘‘displaced’’ to the inner nuclear layer, while less than 1% of these
ganglion cells exist in jawed vertebrates. In gnathostomatous vertebrates, the
displaced ganglion cells project to the basal optic root and the basal optic
nucleus, but this pathway does not exist in hagfishes (Wicht and Northcutt,
1990) and receives little input in lampreys (Fritzsch and Collin, 1990;
Fritzsch, 1991). The axial optic tract occurs in lampreys and conveys retino-
petal (eVerent) fibers rather than retinofugal (aVerent) fibers (Rubinson,
    In most fishes, contralateral and ipsilateral projections from the eyes
terminate in the suprachiasmatic nucleus, the posterior parvocellular preop-
tic nucleus, the lateral geniculate nucleus, the dorsolateral thalamic nucleus,
the pretectal nuclei, and the optic tectum. In all species examined, retinal
input to the optic tectum is retinotopic and is restricted to the stratum
opticum (SO), the stratum fibrosum et griseum superficiale (SFGS), the
3.   NERVOUS AND SENSORY SYSTEMS                                            141

stratum griseum centrale (SGC), and the junction between the stratum
album centrale (SAC) and the stratum periventriculare (SPV), although
some phylogenetic variation exists. Ipsilateral (non‐decussating) input to
the visual centers of the brain is thought to be an inherent component
of the visual pathway in all vertebrates, including hagfishes (15%) and
lampreys. However, the relationship between ipsilaterally projecting gangli-
on cells and the extent of the binocular visual field is not well understood in
early fishes. The optic tracts in both hagfishes and lampreys project bilater-
ally to the preoptic, thalamic, and pretectal nuclei and terminate in the
mesencephalic optic tectum (Wicht and Northcutt, 1990) and may represent
the ancestral condition, which has been retained in hagfish (Fritzsch, 1991).
In garfishes, the mediorostral and ventrolateral regions of the optic tectum
receive ipsilateral input from the retina and subtend the dorsal and ventral
binocular fields, respectively (Collin and Northcutt, 1995). Discrete ipsilater-
al input to the entire optic tectum is found in Australian lungfishes
(Northcutt, 1980) and juvenile teleosts (Collin et al., 2001), suggesting that
there may be phylogenetic diVerences in binocular partitioning (when com-
pared to most modern teleosts). Ipsilateral input via the intertectal and
posterior commissures does not appear to occur in all non‐actinopterygian
fishes and appears to have evolved independently many times (Northcutt and
Butler, 1992). Retinal projections to non‐teleost actinopterygians, that is
sturgeon, and to chondrichthyans (sharks) terminate bilaterally in the pre-
optic area, thalamus, area pretectalis, nucleus of the posterior commissure,
optic tectum, and the nuclei of the accessory optic tract (Ito et al., 1999).
    In both lampreys and hagfishes, two tegmental cell groups (the reticular
mesencephalic area and the nucleus M5 of Schober) give rise to centrifugal
fibers and, as in teleosts, also make contact with bipolar, horizontal, and
ganglion cells. In the river lamprey, L. fluviatilis, it is thought that postsyn-
aptic ganglion cell responses are the synaptic potentials from amacrine cell
contacts and that the amacrine cells are also directly stimulated by retinope-
tal fibers (Vesselkin et al., 1996). The centrifugal system responds to chemical
cues such as sex pheromones and regulates visually mediated sexual and
reproductive behavior in addition to altering ganglion cell responses to
color contrast. The increased proportion of eVerents in agnathans (2%
in lampreys and 5% in hagfishes) in contrast to jawed vertebrates (0.5%)
suggests that this is the ancestral condition.

5.7. Nonvisual Photoreception

    Almost all organisms have evolved some form of endogenous time‐
keeping mechanism or biological clock to respond to changes in environmen-
tal conditions, for example seasons, tides, light cycles, and temperature. The
142                                                           SHAUN P. COLLIN

receptors that sense these regular, environmental perturbations are diverse in
their structure, function, location, and role. The synchronization or entrain-
ment between the external environment and internal rhythms regulates the
period and phase of the rhythm, and, for the majority of animals, this
biological clock is entrained by changes in both the intensity and the spectral
composition of light. Photoentrainment, or using the transition between light
and dark (dawn and dusk) to adjust circadian phase to local time, is consid-
ered primarily to be mediated by nonvisual photoreceptors (Foster and
Provencio, 1999). These photoreceptors do not contribute to image‐forming
vision or object detection (rods and cones in the eye) but respond only to
brightness information. However, as both detector systems occur in all
vertebrate classes, the selection pressures underlying their coevolution and
maintenance are crucial to survival.
5.7.1. Pineal and Parapineal Organs
     The pineal makes the hormone melatonin, which is produced only in the
dark portion of the light/dark cycle and provides a slow signal that is impor-
tant in regulating circadian and/or photoperiodic behaviors. The pineal arises
from an evagination from the roof of the diencephalon and sits beneath a
translucent area or ‘‘window’’ in the skull. Pineal photoreceptors appear to be
highly specialized for detecting gradual changes in environmental light rather
than transient light stimuli (Shand and Foster, 1999). Light responses from
the pineal may be color‐coded with sensitivity extending into the UV range
(Koyanagi et al., 2004). There is no pineal body described in hagfishes, but in
holarctic lampreys and G. australis, pinealocyte outer segments are immuno-
labeled with both anticone and antirod opsin antisera (Garcia‐Fernandez and
                    ´      ´
Foster, 1994; Garcıa‐Fernandez et al., 1997). For G. australis and M. mordax,
metamorphosis and reproduction are both accompanied by migration be-
tween freshwater rivers and the sea, occurring at very precise times of the year
and are most likely controlled by seasonal changes in photoperiod. Pinealec-
tomy has been shown to aVect all the phases of the life cycle (Joss, 1973).
In M. mordax, the parapineal is absent (Eddy and Strahan, 1968). Elasmo-
branchs possess UV or blue sensitive photoreceptors, which serve to monitor
circadian and circannual variations in light intensities (Vigh‐Teichmann et al.,
1990). The outer segments of photoreceptors in the pineal have also been
discovered in the Dipnoi (lungfishes, P. dolloi, Ueck, 1969) and the coelacanth
L. chalumnae (Hafeez and Merhige, 1977).
     The parapineal arises from a dorsal evagination from the diencephalon
(Vollrath, 1981). In lampreys, the parapineal organ consists of a vesicle that
sits beneath the pineal organ and possesses a few photoreceptors. Parapinea-
locytes are not labeled by anticone opsin antisera in the sea lamprey
P. marinus (Garcia‐Fernandez and Foster, 1994), but are densely labeled in
3.   NERVOUS AND SENSORY SYSTEMS                                            143

                                     ´      ´
the river lamprey L. fluviatilis (Garcıa‐Fernandez et al., 1997) with potentially
two types of parapinealocytes based on opsin immunoreactivity (Vigh‐
Teichmann et al., 1984). The parapineal has not been found in adult lung-
fishes (African lungfish, P. dolloi, Ueck, 1969) or elasmobranchs (Bertolucci
and Foa, 2004).
5.7.2. Deep‐Brain Photoreceptors
    As early as 1935, J. Z. Young revealed that deep‐brain photoreceptors
were responsible for body movements in blinded and pinealectomized lamp-
reys following illumination of the head (Young, 1935) and since then all
nonmammalian vertebrates have been found to possess deep‐brain photore-
ceptors (Foster et al., 1994). Although photons are scattered and selectively
filtered by neural tissue, large amounts of light penetrate deep into the brain,
thereby producing a measure of environmental irradiance and hence time of
day. There are two classes of encephalic photoreceptors: cerebrospinal fluid
(CSF)‐contacting neurons in the hypothalamus of lampreys, reptiles, and birds
(Garcia‐Fernandez and Foster, 1994) and cells within the nucleus magnocellu-
laris preopticus in fish and amphibians (Alvarez‐Viejo et al., 2004). The hypo-
thalamic and septal nuclei of the periventricular CSF‐contacting neuronal
system are already present in the lancelets and hagfishes (Vigh et al., 2002;
David et al., 2003).
5.7.3. Dermal Photoreceptors
    Dermal cells are regulated by photoreceptor cells in the CNS via pituitary
melanocyte‐stimulating hormone, pineal melatonin, or directly by photopig-
ments localized within the dermal cells themselves. Dermal chromatophores
or iridophores of many fishes and amphibians are directly sensitive to light
and will aggregate or disperse pigment granules on light exposure. Dermal
photoreceptors are able to initiate behavioral responses in lampreys and
teleosts, where, following illumination, an ‘‘eyespot’’ on the tail of both
ammocoete (Ronan and Bodznick, 1991) and adult (Ullen et al., 1993)
lampreys mediates tail movement and avoidance behavior, respectively. Hag-
fishes have also been reported to possess light‐sensitive pigment in both the
tail and a line of unpigmented skin running down the back in E. burgeri
(Patzner, 1978). Iridophores and chromatophores also occur in the cornea
and act either as a yellow filter, to reduce chromatic aberration and/or
enhance contrast perception (chromatophores), or reflect bright downwelling
light (iridophores, Muntz, 1976). Yellow/orange pigment granules within the
corneal stroma can migrate across the teleost fish cornea in response to
environmental lighting conditions (Siebeck et al., 2003). Recent ultrastruc-
tural studies of the eyes of the river lamprey L. fluviatilis reveal corneal
pigment in the primary spectacle (S.P.C., unpublished data).
144                                                             SHAUN P. COLLIN


    Anamniotic gnathostomes possess five types of chemosensory systems:
(1) olfaction or smell for social interactions, homing, alerting to the presence
of food and avoiding predators; (2) gustation or taste used in feeding and the
localization, acceptance, or rejection of food; (3) isolated chemosensory
receptors for feeding and predator avoidance; (4) a common chemical sense
(free nerve endings) for irritant detection (Finger, 1997); and (5) chemore-
ceptors involved in cardiovascular responses to changes in oxygen and
carbon dioxide levels. The olfactory epithelium lies within a water‐filled
chamber at the base of the nares and comprises a series of folded lamellae
adorned with sensory receptors to detect water‐soluble odors. The gustatory
system comprises taste buds or aggregations of sensory receptors situated
often on raised papillae within the oral cavity, pharynx, or over the head and
body. Solitary chemoreceptors exist as small, encapsulated nerve endings
over the epidermis or, more commonly, on the pectoral fin rays. The common
chemical sense is closely associated with the somatosensory system. Branchial
and extrabranchial chemoreceptors play a role in controlling cardiorespira-
tory responses to changes in oxygen and carbon dioxide levels in addition to
pH. These receptors, sensitive to dissolved gases and innervated by cranial
nerves V, VII, IX, and X, will not be discussed further in this chapter, but see
Chapters 4 and 5 for a comprehensive review.

6.1. Olfaction
6.1.1. Olfactory Epithelium
    Olfaction constitutes a vital role in the lives of primitive fishes, especially
hagfishes and lampreys. In hagfishes, water enters a single median nasohy-
pophysial duct and then passes ventrally to pass over the olfactory organ
(Figure 3.6A). The olfactory organ is housed within a single cartilaginous
capsule but comprises left and right convoluted epithelial surfaces divided by
a medial septum (Northcutt, 1989a). Caudal to the brain, water is then
pumped from another chamber with the help of a velum, which forces
water backward to the gills and out through the branchiopores (Zeiske
et al., 1992; Braun, 1996) (Figure 3.6A). A similar arrangement is found in
lampreys, where water passes over an olfactory median organ before being
pumped toward the gills. The pumping mechanism is an enlarged sac dorsal
to the branchial pharynx, which contracts and subsequently recoils to force
water in and out of the nasal cavity (Kleerekoper and van Erkel, 1960;
Gradwell, 1972). Agnathan water sampling may be described as cyclosmate
(reliant on pumping). This is in contrast to the isosmate (reliant on cilia)
3.   NERVOUS AND SENSORY SYSTEMS                                                                   145

                            nd                v





                                                                                           2 mm

       B                                               C


           Nonsensory                          10 µm                                        2 µm

Fig. 3.6. (A) The olfactory organ of the hagfish, M. glutinosa, showing the head in sagittal
section and in transverse section (inset). The broken line indicates the level of section of the inset.
The arrows show the direction of the water current. n, nostril; nd, nasal duct; np, nasopharyngeal
duct; ol, olfactory lamellae; v, valve; and ve, velum. (B and C) Scanning electron micrographs of
the olfactory epithelium of the gray reef shark, Carcharhinus amblyrhincos, in low (B) and high
(C) power. Note the clear diVerentiation between the sensory and nonsensory epithelium. The
sensory epithelium is characterized by both microvillous (m) and ciliated epithelial cells. [(A) is
adapted from Zeiske et al. (1992). (B and C) are courtesy of Vera Schluessel.]

arrangement in the garfish, L. osseus, where ciliary action moves an odorant
wave front over the olfactory lamellae only once (Bashor et al., 1974).
    In both hagfishes and lampreys, the presence of a prenasal sinus and an
adenohypophysis, respectively, has received much attention along with the
relationship between the olfactory and endocrine systems (Gorbman, 1995).
In bony fishes, the olfactory epithelium is located within specialized nares,
one on each side of the dorsal surface of the head, often with an inhalent
and an exhalent opening. The cartilaginous ratfish, Chimaera monstrosa,
possesses internal nares. Bipolar neurons of the olfactory epithelium are
146                                                           SHAUN P. COLLIN

generally characterized by ciliated apical surfaces and a basal axon that
forms the first cranial nerve (olfactory nerve, nI, Figure 3.6B and C).
     The olfactory epthelium of lampreys is composed of only ciliated recep-
tors, whereas hagfish olfactory epithelial cells possess both cilia and micro-
villi (Thornhill, 1967; Theisen, 1976). A similar dichotomy is revealed in
lungfishes, where the African lungfish (Protopterus annectens) possesses
both microvillous and ciliated receptor cells, but the Australian lungfish
(N. forsteri) has only microvillous receptor cells as adults but both as juve-
niles (Hansen et al., 1994). The olfactory organs of dipnoans (Derivot et al.,
1979) and the coelacanth (Millot and Anthony, 1965; PfeiVer, 1969) have
also been described.

6.1.2. Sensitivity and Odor Discrimination
     The olfactory epithelium is sensitive to a range of odors, including amino
acids, bile acids, steroid pheromones, and aromatic compounds. For teleosts,
sensitivity thresholds vary for amino acids (10–9 M in Ictalurus catus, Caprio,
1980), steroids (10–13 in Carassius auratus, Sorensen et al., 1987), and pros-
taglandins (10–13 M in Misggurnus anguillicaudatus, Kitamura and Ogata,
1980). In the hagfish M. glutinosa, the olfactory system reveals a sensitivity
threshold of 10–6 for both amino acids and steroids (Doving and Holmberg,
1974). Acids (such as taurocholic acid) are produced by the bile to solubilize
ingested fat and are then released into the environment, where they can be
used as a signal by conspecifics, especially in the context of homing and sex. In
anadromous lampreys, individuals return to their natal stream to spawn and
eventually die based on an acute sensitivity to specific amino acids and bile
acids produced by ammocoetes and other fishes (Teeter, 1980; Li et al., 1995).
Similarly, hagfish are thought to rely heavily on olfaction to find prey and,
with their limited visual and hydrodynamic abilities, chemosensory sensitivity
may be their most important sensory modality. DC recordings [analogous to
an electro‐olfactogram (EOG)] from the olfactory epithelium of hagfishes
reveal that L‐alanine is detected at lower concentrations than L‐glutamine
and that g‐aminobutyric acid, glutathione, and 4‐hydroxy‐l‐proline are also
potent stimuli (Doving and Holmberg, 1974).
     Sea lampreys (P. marinus) undergo three major phases during their
life cycle (a larval phase, where they travel down freshwater streams after
metamorphosis; a marine phase, where they become parasitic on large
teleosts; and an upstream phase, returning to their natal stream to spawn
and die) (Hardisty and Potter, 1971). In the upper regions of the streams,
spermiating male bile acids are thought to function as a mating pheromone,
while larval bile acids act as a migratory pheromone. The bile acid 3kPZS
acts as a potent male pheromone to adult female lampreys and allows
discrimination between ovulating females and ever present larvae in the
rapids, where the released bile acids would be rapidly diluted (Siefkes and
3.   NERVOUS AND SENSORY SYSTEMS                                              147

Li, 2004). Suzuki (1978) suggests that calcium ions play an important role in
the generation of olfactory receptor potentials (resting membrane potentials
of the olfactory neurons in the lamprey Entosphenus japonicus range from
À36.7 to À60.0 mV, Suzuki, 1977). In general, olfactory acuity varies among
species. In studies utilizing a ‘‘natural’’ stimulus of beef heart extract, rather
than pure compounds, PfeiVer (1969) revealed that sensitivity thres-
holds varied between 10–11 g literÀ1 in C. calabaricus and 10–14 g literÀ1 in
the bichir Polypterus bichir, which was about 10 times more sensitive
(macrosmatic) than the teleost Phoxinus.
    Olfactory receptor proteins are encoded by a large multigene family,
which may consist of up to 1000 genes in mammals (Buck and Axel, 1991)
but in the order of 50–100 genes in teleost fishes (Weth et al., 1996). Individual
olfactory receptor genes are selectively expressed in a small subpopulation
of olfactory neurons, each of which expresses only one or a few receptor types
(Freitag et al., 1998). Comparison of the deduced amino acid sequences of the
olfactory receptor gene family reveals that there are two basic types (Types I
and II) that are thought to be specialized for detecting water‐soluble odorants
and volatile odorants, respectively. Analysis of the receptors in L. chalumnae
shows that although both Types I and II are both present, the Type II receptor
genes represent nonfunctional pseudogenes (Freitag et al., 1998). It is inter-
esting to speculate whether the diminishing importance of the Type II recep-
tors in these lobe‐finned ‘‘living fossils,’’ which hold such vital clues to the
evolution of tetrapods, has resulted either from selection pressures associated
with its origin or a new function for the recognition of volatile odorants.
6.1.3. Primary and Secondary Olfactory Input to the CNS
    Hagfishes and lampreys both send large numbers of olfactory axons into
the olfactory bulbs (primary input) and telencephalon (secondary input). The
secondary projections in hagfishes are so pronounced that a large, laminated
olfactory cortex has been identified within the telencephalon (Wicht and
Northcutt, 1992). In lampreys, the receptor cells within the left and right
epithelia project to the ipsilateral olfactory bulbs and form spatially and
functionally distinct arrangements. The number of types of olfactory recep-
tors in lampreys is relatively small and only a few odorants stimulate olfac-
tory activity (Li et al., 1995; Frontini et al., 2003). From the olfactory bulbs,
axons of the mitral cells leave the glomerular zones and form fiber bundles
(medial and lateral) collectively called the olfactory tract, which terminate
within various regions of the telencephalon. In contrast to the olfactory nerve
axons, which are all unmyelinated, the olfactory tract is composed of a
mixture of myelinated and unmyelinated axons.
    In all primitive fishes studied thus far (Lampetra planeri, I. unicuspis,
Polypterus palmas, C. calabaricus, A. calva, and P. dolloi), a subset of primary
olfactory projections bypass the olfactory bulbs and terminate in both the
148                                                           SHAUN P. COLLIN

telencephalon and the diencephalon, with the exception of Acipenser ruthe-
nus, that does not possess extrabulbar projections to the diencephalon (see
review by Hofmann and Meyer, 1995). The function of these extrabulbar
olfactory projections is speculative but is thought to be involved in the
detection of waterborne substances that are either in high enough concentra-
tions or are suYciently distinct to elicit a response without the processing
necessary for bulbar projections. Despite their putative function, there is
little evidence to suggest that extrabulbar projections are of olfactory origin,
indicating that their inclusion within the terminal nerve complex may be
more appropriate (Demski, 1993; von Bartheld, 2004).

6.1.4. The Terminal Nerve
    Originally discovered in the African lungfish, P. annectens, by Pinkus
(1894), the terminal nerve (or nervus terminalis) is considered to be a group
of ganglion cells, which possess fibers projecting from the nasal epithelium,
that bypass the olfactory bulb and terminate in various regions of the
CNS. The nervus terminalis is considered to be a separate cranial nerve
(n0), but its projection patterns and characterization are variable in the
range of species examined, making it diYcult to trace its origins or function.
In lungfishes, an anterior root enters the brain among the olfactory nerve and
bulb, while the posterior root enters the CNS at the level of the optic nerve
(as the nervus praeopticus, SewertzoV, 1902). The two components of the
terminal nerve in dipnoans possess diVerent neurochemical signatures (ante-
rior root is GnRH immunoreactive and the posterior root is acetyl-
cholinesterase immunoreactive, von Bartheld et al., 1988, 1990). The
situation appears to be diVerent for lampreys that have lost the GnRH or
FMRF‐amide terminal nerve components but retained other features
thought to be characteristic of this system (Eisthen and Northcutt, 1996).
Bichirs (Polypterus) possess similar projection patterns to lampreys and
lungfishes (von Bartheld and Meyer, 1986). The terminal nerve within the
brain of both the bichir, Polypterus senegalus (Chiba, 1997), and the gar,
Lepisosteus oculatus (Chiba, 2005), is immunoreactive to neuropeptide Y and
GnRH, suggesting that it acts as a neuromodulator, especially in reproduc-
tive behaviors via the hypothalamic pituitary system. Von Bartheld (2004)
has recently suggested that the terminal nerve in primitive fishes is extremely
variable and is probably a component of all primitive species, albeit defined
using diVerent anatomical, neurochemical, and developmental criteria.

6.2. Gustation
    Taste or the central perception of both mechanical and chemical stimula-
tion by food is mediated by taste buds. Taste buds are pear‐shaped organs,
which consist of up to 100 specialized epithelial cells concentrated into
3.   NERVOUS AND SENSORY SYSTEMS                                            149

aggregations over the oropharynx and often also covering the body surface
or branchial cavity. Unlike olfactory neurons, taste receptors lack an axon,
where their bases are synaptically connected to aVerent nerve fibers via
chemical synapses. From developmental studies, taste buds are considered
secondary sensory cells (Reutter and Witt, 1993) and are always stimulated
by nutritionally important substances. In most anamniotes, the diVerent
sized villi at the apical ending protrude above the surface of the epithelium.
In teleost fish, taste buds are able to discriminate between palatable
substances before ingestion and thereby act over very small distances
(Kasumyan, 1997). The gustatory system comprises peripheral receptors or
taste buds, which are innervated by the facial (nVII), glossopharyngeal
(nIX), and vagal (nX) cranial nerves (Finger, 1983). Oropharangeal taste
buds appear to be the primitive condition with external taste buds over the
head and even trunk regions evolving independently a number of times
(Northcutt, 2004).
    Lampreys (P. marinus and I. unicuspis) do not appear to possess external
taste buds (Baatrup, 1985), but have terminal (ciliated) buds over the internal
surfaces of the pharyngobranchial ducts (Mallat and Ridgeway, 1984), which
respond preferentially to amino acids. Given the similar distribution, inner-
vation, and central projections of the internal buds in lampreys and the taste
buds of teleosts, Braun and Northcutt (1997) have considered them to be
    External taste buds are frequently observed over the head of gars, that is
L. platyrhinchus (Norris, 1925; Song and Northcutt, 1991b), the bowfin,
A. calva (Allis, 1889), and sturgeons (Norris, 1925; Figure 3.7A and B), but
not in Polypterus (PfeiVer, 1968b) and Polyodon (Norris, 1925). This distri-
bution suggests that the earliest ray‐finned fishes possessed only internal taste
buds and that external taste buds evolved independently in chondrosteans
and neopterygians. Although there is considerable variation in the morphol-
ogy of taste buds, four types have been characterized: dark cells (Type I),
light cells (Type II), basal cells, and stem cells (Reutter and Witt, 1993;
Finger, 1997). In Lepisosteus, taste buds are composed of two types of
elongated light cells and one type of dark cell (Reutter and Hansen, 2005)
(Figure 3.7C and D). In contrast, Amia taste buds contain two types of light
cells and two types of dark elongated cells. On the basis of the height of the
microvillar projections above the epidermis, both the Australian lungfish,
N. forsteri, and the sturgeon, Scaphirhynchus platorynchus, possess three
types of light cells and one type of dark cell (Reutter and Hansen, 2005).
AVerent synapses are common in the buds of both species, while eVerent
synapses occur only in Lepisosteus taste buds (Reutter et al., 2000). The inter-
nalized taste buds within the oropharynx in the plesiomorphic representatives
of the sarcopterygian group of lobe‐finned fishes (Latimeria and Neoceratodus)
appear to confirm that this is the primitive pattern and that protopterid
150                                                                         SHAUN P. COLLIN

      A                                          B


                                       200 µm                                     20 µm

      C                                                    D


                            CI1               Cd             Cd           CI2         Cd

                                             1 µm                            1 µm

Fig. 3.7. (A) Scanning electron micrograph of the tip of the barbel of the sturgeon, A. baeri,
showing a series of epidermal hillocks each containing several taste buds. (B) High power of a
hillock in A. baeri showing three taste bud (TB) receptor regions. (C and D) Transmission
electron micrographs of the apical region of two types of light (Cl1 and Cl2) and dark (Cd) taste
receptors in the garfish, L. oculatus. Note one type of light cell has a single large villus.
[Reproduced with permission from Reutter and Hansen (2005).]

lungfishes (that possess external taste buds, PfeiVer, 1968b) have independently
evolved external taste buds as has occurred in teleosts (especially over barbels,
fin rays, and around the mouth, Hansen and Reutter, 2004).
    Taste buds are used to assess palatability of food and may be ‘‘tuned’’ to
specific prey items or nutritional needs or both. At least teleost fish taste buds
are sensitive to diVerent amino acids such as arginine, proline, and alanine
3.   NERVOUS AND SENSORY SYSTEMS                                              151

(Caprio et al., 1993). Presumably, taste buds in lampreys may be used to
monitor respiratory currents for the presence of potential food particles
(Finger, 1997). Little is known about the taste preferences of other primitive
fishes. However, using the teleost systems as a guide, one may expect that
taste preferences are highly species specific, that there is a strong similarity in
the taste preferences between geographically isolated fish populations, and
that environmental factors, such as water temperature and pollutants (heavy
metals and low pH water), may shift taste preferences and fish‐feeding
motivation (Kasumyan and Doving, 2003).

6.3. Solitary Chemoreceptor Systems

    Solitary chemosensory cells are present in lampreys, elasmobranchs,
teleost fishes, and some amphibians (Whitear, 1992). The hagfishes, Epta-
tretus stouti and Myxine sp., do not possess taste buds but possess solitary
chemosensory receptors over the oral tentacles, at the opening of the prenasal
duct and in the epidermis surrounding the mouth and oropharynx (Retzius,
1892; Patzner et al., 1977; Georgieva et al., 1979). These solitary receptors are
innervated by either cranial (trigeminal) or spinal nerves, whose central
projections form a continuous dorsolateral tract, which runs the length of
the neuraxis (Whitear, 1992; Braun, 1996; Finger, 1997). Hagfishes also
possess 180,000 Schreiner organs (Schreiner, 1919), which are similar to the
solitary receptor organs, within the oropharynx and over the entire external
surface of the body (Northcutt, 2004). Fibers from the Schreiner organs over
the barbels, snout, oral epidermis, and velum terminate within the medullary
trigeminal complex, while those on the body project to the dorsal spinal cord.
Those organs over the pharynx terminate within the visceral column (vagal
lobe and fasciculus communis) (Matsuda et al., 1991; Finger, 1997). There is
no information pertaining to the stimuli that elicit responses in the sensory
bud system of hagfishes. However, the large representation of the terminal
field in the medulla of the CNS (occupying 10%) suggests an important role.
    In both ammocoetes and adult Lampetra planeri, solitary chemosensory
receptors occur over the head and trunk that are characterized by apical
microvilli and a basal synapse (Northcutt, 1989a). The oligovillous solitary
chemosensory cells of lampreys physiologically respond to epithelial secre-
tions (sialic acid), fish mucous, and water tainted by the odor of other fish.
Sialic acid is stimulatory in concentrations as low as 10–9 and could be a
meaningful stimulus to parasitic lampreys in addition to being important in
social behavior (Baatrup and Doving, 1985). The identification of sialic acid
in the epithelial cells of the pouched lamprey, G. australis (Lethbridge and
Potter, 1979), could be also detected by the numerous solitary oligovillous
cells on the male genital papilla (Whitear and Lane, 1983).
152                                                           SHAUN P. COLLIN

6.4. Common Chemical Sense

     The common chemical sense is mediated by epithelial free nerve endings
that are innervated by cranial and spinal nerves, many associated with the
somatosensory system. Stimulation of these nerve endings (nociceptors
and thermoreceptors) by various chemicals produces sensations including
‘‘pain,’’ warmth, and cold; however, others may elicit irritation and subse-
quently aversive behavior (Rovainen and Yan, 1985; Andres and von
During, 1993).


    The octavolateralis system comprises end organs responsible for audi-
tion, vestibular control, and the lateral line, which all utilize hair cells
as receptors and branches of the eighth (octaval) and lateral line nerves
(Northcutt, 1981; McCormick, 1982; Platt et al., 1989). In contrast to earlier
theories that suggested that the ear was derived from the lateral line, the term
‘‘octavolateralis system’’ should not inherently suggest acoustic function, nor
imply evolutionary or ontogenetic origins (see review by Popper et al., 1992).
It was Wever (1974) who first suggested that the auditory system
and the lateral line may have evolved concurrently, but there is still much
to learn about the role each system plays in the detection of propagated
(sound) energy at low frequencies. The lateral line appears to respond to
particle motion below 1 Hz, whereas the low limit of the inner ear in most
fishes is about 35–50 Hz (Popper, 1983). Therefore, the two systems are
complementary with the auditory system operating as a high‐frequency
sound detector and the lateral line acting as a low frequency detector. The
lateral line and auditory innervation also maintain their separate integrity to
at least the level of the midbrain or the torus semicircularis in A. calva
(McCormick, 1981; Northcutt, 1981).

7.1. Audition

    Three major evolutionary patterns exist with respect to the vertebrate ear.
In hagfishes, a single torus with two rings of sensory epithelia and a single
macula comprise the ear. The lamprey ear comprises two semicircular canals
with trifid cristae and a partially divided macula, while the gnathostomatous
ear has three distinct semicircular canals and at least two or more maculae
with otoconia (Fritzsch, 1998). In general, the angular acceleration‐sensing
semicircular canals and the linear accelerating‐sensing systems of the otoconia‐
bearing maculae have remained relatively constant (Lewis et al., 1985).
3.   NERVOUS AND SENSORY SYSTEMS                                            153

Classically, the utricle has been considered as the primary vestibular organ for
postural control and the sacculus as the primary auditory end organ (with the
lagena used for both). However, the utricle has undergone large evolutionary
change where it lies either in a recess of its own (elasmobranchs and lungfishes)
or in the common crus of the anterior and horizontal canal (bony fish
and tetrapods, Fritzsch, 1998), and all three maculae may be involved in
both vestibular and auditory functions (Popper and Fay, 1993; Popper and
Platt, 1993). The eighth cranial nerve is divided into (1) the anterior ramus,
which innervates the anterior horizontal semicircular cristae, the utricular
macula, and part of the saccular macula and (2) the posterior ramus, which
innervates most of the saccular macula, the lagena macula, the macula
neglecta, and the posterior semicircular crista (Meredith and Butler, 1983;
Popper and Northcutt, 1983). A single macula communis is present in the
hagfish, Myxine, and two ampullae appear as dilations on either end. Subdivi-
sions of this torus are thought to be homologous to the utricular and lagenar
maculae of gnathostomes (Lowenstein and Thornhill, 1970). The macula
communis in lampreys can be divided into anterior, vertical, and posterior
subdivisions, which are thought to be homologous to the utricular, saccular,
and lagenar maculae in gnathostomes (Lowenstein et al., 1968; Popper and
Hoxter, 1987).

7.1.1. The Inner Ear and Hair Cells
    The mechanism of transduction from a mechanical signal to neurotrans-
mitter release is still not well understood but is mediated by bundles of hair
cells, composed of a single long kinocilium and a number of stereocilia. On
movement of the surrounding endolymph, and thereby the cupula, the hair
cells are deflected possibly changing the ionic permeability and/or resistance
of their membranes to produce a voltage diVerential. The inner ear in six
species of hagfishes, all possess a labyrinth containing a single macula and
two cristae in a single semicircular canal. The macula consists of a horizontal,
a middle vertical, and a posterior horizontal component, each of which is
covered by numerous round statoconia (otoconia) (Retzius, 1881; Jorgensen
et al., 1998). The hair cell bundles contain long kinocilia (up to 35 mm in
length) and lack a cupula.
    The ear of the coelacanth, L. chalumnae, possesses fewer otoliths than the
three found in bony fishes, where the upper part of the inner ear adheres to
the gnathostomatous pattern of a moderately sized utricle and three perpen-
dicularly oriented semicircular canals (Millot and Anthony, 1965; Bernstein,
2003). The lower part comprises the sacculus with a large otolith and sepa-
rated lagenar recess. Analyses of larval and adult specimens of Latimeria
have revealed that both ears are linked by a canalis communicans,
which is enclosed by cartilage (Bernstein, 2003; Fritzsch, 2003). Where the
154                                                           SHAUN P. COLLIN

communicans enters the saccular/lagena region, an end organ lies innervated
by the lagenar branch of the statoacoustic nerve VIII. The end organ is
proported to represent a basilar papilla, a potential precursor to the cochlea
of tetrapods (Fritzsch, 1987, 2003), although Bernstein (2003) considers
this structure more suited to sensing pressure changes during movements
involving the intracranial joint rather than for audition.
    Many species of elasmobranchs also have one or two accessory maculae,
that is the macula neglecta, which contain patches of sensory epithelium
covered by a gelatinous membrane. In sharks and rays, the macula neglecta
possesses high numbers of hair cells (267,000 in the gray reef shark, Carch-
arhinus menisorrah, and 6,000 in the thornback ray Raja clavata), which are
reported to detect forces directed posteroventrolaterally in the posterior
canal duct, suggestive of an auditory function (Corwin, 1977, 1983). Physio-
logical recordings confirm auditory sensitivities of between 40 and 200 Hz
for the ray, R. clavata (Corwin, 1983), and between 31 and 375 Hz for the
lemon shark, Negaprion brevirostris (Corwin, 1981), where both nonotolithic
(macula neglecta) and otolithic (sacculus) organs may mediate sound source
localization (Corwin, 1981). Popper (1977) has reported that sharks can
behaviorally detect sound at frequencies of up to 1000 Hz and that certain
low frequency signals attract sharks from large distances.
    The sensory cells in the sacculus are innervated by both aVerent and
eVerent fibers, and although the precise number of neurons innervating the
otolithic organs is unknown for most primitive fishes, the number of sensory
cells varies (216,000 cells in the lagena macula; 56,400 cells in the utricular
macula; and 8,600 cells in the saccular macula) in the bowfin, A. calva. These
high receptor numbers indicate a high convergence of information in the
order of 90 sensory cells to ganglion cells (Popper and Northcutt, 1983),
which is appreciably higher than the hair cell to myelinated axon ratio of 57:1
found in the macula neglecta in the shark C. menisorrah (Corwin, 1977).
In the sturgeon and the bichir, the saccular sensory hair cells are divided into
two groups as opposed to four in non‐osteriophysans (Popper, 1983). The
sacculus and lagena comprise one continuous chamber in the sturgeon,
S. platorynchus, and are connected through a wide canal in the bowfin,
A. calva (Figure 3.8A and B). Unlike the condition in teleosts, early ray‐
finned fishes have horizontally oriented cells rostrally, and vertically oriented
cells caudally in the saccular macula. The horizontally oriented cells also
appear to be derived from vertically oriented cells through ontogenetic shifts
in macula curvature (Popper, 1978). Therefore, the most primitive ray‐finned
fishes may have originally been equipped with vertically oriented hair cells,
where the subsequent evolution of horizontally oriented hair cells enabled
them to do a vectorial analysis of acoustic signals associated with sound
localization (Schuijf and Buwalds, 1980; Platt and Popper, 1981). The
3.   NERVOUS AND SENSORY SYSTEMS                                                              155

 A                                                            B

                 hc        u
       ac                                  cn

                 cn                                      Im
                                      In          I

Fig. 3.8. (A) Illustration of the medial side of the left ear of the bowfin, A. calva, based on an
early depiction by Retzius (1881). The macula neglecta is not shown. ac, crista of the anterior
semicircular canal; ar, anterior ramus of the otic nerve; cn, branches of otic nerve to the various
cristae; de, ductus endolymphaticus; h, crista of the horizontal semicircular canal; l, lagena; lm,
lagenar macula; ln, lagenar branch of the eighth nerve; pb, posterior branch of the otic nerve; pc,
crista of the posterior semicircular canal; s, sacculus; sm, saccular macula; u, utricle; and um,
utricular macula. (B) Ciliary bundles from the saccular macula. Note the long kinocilia (K) and
shorter stereocilia (S). [Reproduced from Popper and Northcutt (1983) with kind permission of
John Wiley & Sons, Inc.]

saccular macula in lungfishes is diVerent to that of teleosts, and there is no
discrete boundary between vertical and horizontal groups of hair cells, which
comprise up to 80 cilia each (Platt et al., 2004). DiVerences in the size of the
ciliary bundles located in the striolar and extrastriolar regions of the lagenar
and saccular maculae may indicate that they respond to diVerent types of
stimuli and have diVerent frequency response characteristics.

7.1.2. Sound Source Localization, Sensitivity, and
       Frequency Tuning
    The perpendicular orientation of the hair cells in the saccular and lagena
in both the bichir, P. bichir, and the sturgeon, S. platorynchus, suggests
that these primitive species, like teleosts, are able to detect signals at right
angles from one another and therefore localize a sound source (Popper,
1978). Whether the direct displacement of the otolith is mediated by bone
conduction (appreciable in these two species) or pressure displacement is
unknown. The hearing abilities of the paddlefish, P. spathula, and the lake
sturgeon, Acipenser fulvescens, have revealed that both fish are responsive to
sounds ranging in frequency from 100 to 500 Hz and their lowest hearing
thresholds are acquired from frequencies in a bandwidth of between 200 and
156                                                           SHAUN P. COLLIN

300 Hz (Lovell et al., 2005). Both species are not sensitive to sound pressure
but to particle motion and will be able to detect the knocks and moans
produced during their breeding season, thereby serving as some form of
intraspecific communication. Both P. spathula and A. fulvescens do not
possess an internal division of their saccule and lagena, a feature shared
with African lungfishes, Protopterus (Platt et al., 2004). The saccule bears
hair cells divided into two oppositely oriented groups allowing these species
to locate the source of a sound in both the horizontal and vertical planes,
relying on the stimulation of ciliary bundles oriented specifically along the
sound propagation axis (Lu and Popper, 1998).

7.2. Vestibular Control

    In vertebrates, a membranous labyrinth contains specialized organs
that mediate the reception of gravity, linear, and angular accelerations.
This mechanosensory complex is innervated by the eighth cranial nerve.
Fishes lack a cochlea but otherwise the semicircular canal system has not
changed much during vertebrate evolution (Platt, 1983a,b). The semicircular
canal system monitors rapid head movements and provides spatial orienta-
tion. This is achieved by the displacement of endolymph within the canals,
which stimulate hair cells located in a dome‐like enlargement (ampulla) and
which are covered by a gelatinous cupula. Most fishes are considered to
possess Type II vestibular hair cells but diVerent hair cell bundle forms
have been diVerentiated on the diameter of the stereocilia, bundle size, and
the length of the kinocilium.
    Three otolithic organs (utriculus, sacculus, and lagena) extend from
the bases of the semicircular canals and are covered by either crystalline
structures of aragonite (bony fishes) or aggregations of calcite crystals (carti-
laginous fishes). The utriculus is thought to be the superior part of the
vestibular system responding primarily to movement and postural changes.
The inferior part of the ear (encompassing the sacculus and lagena) responds
primarily to vibration and acoustic stimuli. The hair cells of non‐teleosts
such as the bichir (P. bichir) and the sturgeon (S. platorynchus) have
been characterized (Popper, 1978), and hair cell bundles with long kinocilia
located near the margins of the otolithic organs have been described in
chondrosteans (Popper, 1978), elasmobranchs (Lowenstein et al., 1964),
and agnathans (Lowenstein et al., 1968). The vestibular maculae in the
lampreys, L. fluviatilis and E. japonicus, possess only two ciliary bundle
forms, where the forms with the shortest stereocilia are localized in areas
with vibration sensitivity (Lowenstein et al., 1968; Lowenstein, 1970). The
large size of the kinocilium dictates that the hair cell bundle has an orienta-
tion (Figure 3.8B), where displacement toward the kinocilium produces
3.   NERVOUS AND SENSORY SYSTEMS                                            157

excitation and displacement away from the kinocilium produces inhibition.
Although the parts of the labyrinth concerned with balance and postural
control seem to show less variability than the parts related to acoustic
reception, the demarcation between hair cells of diVerent polarities lies near
the curving anterior edge of the utricular macula in both the shovelnose
sturgeon, S. platorynchus, and the bichir, P. bichir, which is wide and disc
shaped, respectively (Popper, 1978).
    The otoconia‐bearing organs can also detect linear accelerations. Surgical
lesions and examination of the resultant behavioral deficits have established
that the utricle has a gravistatic function and the saccule has a specialized
acoustic function. The lagena is thought to participate in both (Schoen and
von Holst, 1950).

7.2.1. Semicircular Canals and Balance
    Hagfishes possess a single canal (simple torus), which is oriented about
55 to the vertical plane (Lowenstein and Thornhill, 1970). In the hagfish,
Myxine a large internal radius of the single semicircular canal is the result of
the need to increase the sensitivity within a canal that must signal rotation in
three planes, while the two cristae within the canal are known to respond to
angular accelerations despite the absence of a cupula (McVean, 1991).
Lampreys possess two separated semicircular canals. An additional horizon-
tal canal evolved with the jawed vertebrates. Lampreys and hagfishes also
appear to possess a more primitive pattern of oculomotor innervation
(Fritzsch, 1991) than in jawed vertebrates. This reorganization may be
related to the evolution of the inner ear. The innervation pattern of four
extraocular muscles supplied by the oculomotor nucleus (two contralateral
and two ipsilateral), one muscle by the trochlear, and one muscle by the
abducens nucleus in elasmobranchs and lungfishes appears to have evolved
into the tetrapod pattern of four extraocular muscles innervated by the
oculomotor nucleus (one contralateral and three ipsilateral) and one muscle
by the trochlear, and one muscle by the abducens nucleus after the acquisi-
tion of the horizontal canal (Fritzsch, 1998). The large Muller and Mauthner
cells in the lampreys, I. unicuspis and P. marinus, also respond to vestibular
stimulation as part of a general arousal response (Rovainen, 1980), where in
addition to other motor functions, reticulospinal neurons participate in
polysynaptic vestibular reflexes (Rovainen, 1979).

7.2.2. Vestibulo‐Ocular Control
    Visual input is important for controlling body orientation in the three
dimensionalities of the water column. Hagfish canal aVerents are about half
as sensitive as those of other vertebrates due to their lack of image formation
and not having to stabilize their eyes (McVean, 1991). However, the neural
158                                                             SHAUN P. COLLIN

mechanisms controlling spatial orientation (roll and pitch angles in the
vertical plane and the yaw angle in the horizontal plane) with respect to the
eyes have received a great deal of attention in lampreys. Vestibular informa-
tion is crucial for the roll control and vestibular reflexes are tuned to stabilize
a dorsal‐side‐up orientation (Ullen et al., 1995a,b). Lampreys rely on extra-
tectal inputs to evoke the dorsal light response (a roll tilt toward the light)
and negative phototaxis (lateral turn away from the light) (Ullen et al.,
1995b) and use the eyes to compensate for any vestibular impairment
(Degliagina, 1997). Lampreys have also been used to highlight the striking
capacity of the CNS to compensate for unilateral loss of the vestibular system
by restoring the symmetry of reticulospinal commands that control rotation
around the horizontal axis (Degliagina et al., 1993; Pavlova et al., 2004).
Behavioral responses to vestibular stimuli depend on modulation by the
CNS, where the state of arousal, adaptive compensation, and eVerent inner-
vation can play a large role (Platt, 1983a). Responses to the dorsal light reflex
depend on the excitability of the fish, and restoration of normal responses
following vestibular lesions may take over a month in teleosts. EVerent
neural pathways can modulate postural behavior and ‘‘reaVerence’’ would
allow receptors to retain sensitivity to other external stimuli.

7.3. Lateral Line
     Unlike the auditory portion of the ear and the closely related electro-
sensory system, the mechanosensory lateral line is thought to have arisen
from a single common ancestor. Lateral line organs are mechanoreceptors,
which are direction selective neuromasts containing hair cells (with a rela-
tively immobile kinocilium and up to 100 stereocilia) over the surface of a
sensory macula. Embedded within a gelatinous cupula, the hair cells are
connected by fine extracellular threads, where the displacement of the stereo-
cilia is proportional to the water flow (Dijkgraaf, 1963). Movement of the
sterocilia relative to the kinocilium provides a directional indicator (deflec-
tion toward the kinocilium causes a depolarization and an increase in the
aVerent firing rate; deflection away from the kinocilium causes a hyperpolar-
ization and a decrease in aVerent nerve firing) (Blaxter, 1987). The ability to
detect both particle motion and pressure may have evolved independently for
both the ear and the lateral line. Significant changes in the relative excitation
of diVerent lateral line canal neuromasts can result from mixing the pressure‐
induced and motion‐induced signals (Coombs et al., 1988). These changes
could encode information about the distance of a source, about sudden
changes in acceleration of the source, and about small changes in its relative
position (Gray, 1984).
3.   NERVOUS AND SENSORY SYSTEMS                                               159

7.3.1. Structure and Function of Neuromasts
    Neuromasts may be located superficially or in canals. Superficial neuro-
masts lie in shallow dermal pits or grooves and function by projecting their
cupula into open water with slow‐moving fishes possessing more superficial
neuromasts than fast moving ones (Dijkgraaf, 1962). On the other hand,
canal neuromasts are stimulated by local displacements of the canal fluid
perpendicular to the surface of the skin. Pores to the external environment
allow water to pass into the semicompartmentalized canals, where the hair
cells act as displacement sensors allowing the animal to detect something
moving toward it. Occluded canals produce an acceleration of the water past
the neuromast, thereby amplifying the signal (Coombs et al., 1988). Seven
lateral line canals represent the basic pattern in fishes: supraorbital, infra-
orbital, mandibular, otic, temporal, supratemporal, and trunk canals (Song
and Northcutt, 1991b).
    The hagfish lateral line is composed of rudimentary clusters of a single
type of ciliated cell sitting in a series of two or three shallow epidermal grooves
(Braun and Northcutt, 1997). Although there is a central kinocilium
surrounded by stereocilia, there is no apparent orientation and no cupula.
The development (Wicht and Northcutt, 1995), innervation (Braun and
Northcutt, 1997), and central projections (Kishida et al., 1987) indicate that
the lateral line of hagfishes is homologous to other craniates but its absence in
the genus Myxine (Braun and Northcutt, 1997), the lack of eVerent inner-
vation, and large interindividual variation suggest that the role and impor-
tance of this sense is still to be elucidated. It is interesting to note that
instead of lateral line receptors, mechanoreceptive‐lamellated receptors are
present in the hyperdermal cutaneous layer (Andres and von During, 1993).
These spindle‐shaped or cylindrical corpuscles are innervated by unmyelinat-
ed spinal nerve axons and are thought to be encapsulated stretch receptors,
which appear to have evolved in this group independently in addition to
species of anurans and reptiles (von Du                                        ¨
                                               ¨ring and Seiler, 1974; von During
and Miller, 1978).
    Lampreys possess three classes of lateral line receptors, and the appear-
ance and polarized orientation of their hair cells are similar to those in
gnathostomes, except that they do not receive eVerent innervation and lack
a cupula (Lane and Whitear, 1982; Fritzsch et al., 1989). The lateral line in
hagfishes comprises only a few shallow epidermal grooves containing a single
class of sensory hair cell (Braun, 1996), which could respond to spatial
disturbances in the hydrodynamic field surrounding the animal. All the
electroreceptors are innervated by the anterior lateral line nerve (Ronan
and Northcutt, 1983). While cartilaginous fishes possess canal neuromasts,
in addition to pit lines, lateral line receptors in lungfishes either lie in grooves
160                                                          SHAUN P. COLLIN

or are superficial. It appears that neuromasts in lampreys, lepidosirenid
lungfishes, and some teleosts have independently evolved from being housed
in grooves and/or canals to occurring as lines of superficial neuromasts
(Northcutt, 1989b). This tendency toward more superficially placed neuro-
masts may be correlated with a progression into more still water environ-
ments and/or sedentary behavior, where these types of neuromasts may
function as proprioceptors of swimming velocity (Blaxter, 1987). The reduc-
tion of the lateral line canals in the bichir, Polypterus, and the lungfishes,
Protopterus, Neoceratodus, and Lepidosiren, could also be the result of selec-
tion for changes in neuromast function, the nonadaptive reduction of dermal
bone around the lateral line canals and their neuromasts (Webb and
Northcutt, 1997), or a by‐product of the need to truncate their development
in an environment prone to desiccation (Northcutt, 1989b).
7.3.2. Frequency Sensitivity and Object Localization
    The apical surfaces of canal neuromasts are 2–4 times larger and more
oval shaped than those of superficial neuromasts, and canal neuromasts
possess 4–15 times more hair cells than do superficial neuromasts in the
Florida garfish, L. platyrhinchus (Song and Northcutt, 1991b) (Figure 3.9).
Similar diVerences have been noted in the cladistian, Polypterus, by Webb
and Northcutt (1988). These morphological diVerences may reflect physio-
logical diVerences in the transduction properties and the sensitivity of the
neuromasts (van Netten and Kroese, 1989). It appears that canal neuromasts
respond better to high frequencies than do superficial neuromasts and
that superficial neuromasts may respond preferentially to water velocity in
contrast to acceleration (Kroese and Schellart, 1987).


    Electroreceptors are found in most primitive fishes, including lampreys,
elasmobranchs, non‐teleost ray‐finned fishes (such as polypterids and
chondrosteans), some teleosts (siluriforms, gymnotids, mormyrids, and gym-
narchids), dipnoans, crossopterygians, and aquatic amphibians (urodeles
and apodans) (Bodznick and Northcutt, 1981; Bullock et al., 1982;
Jorgensen, 1982; Bullock and Heiligenberg, 1986; Northcutt, 1986a;
Blaxter, 1987; Zakon, 1988; Jorgensen, 2005). Although both hagfishes and
lampreys possess a lateral line, only the lampreys possess an electrosense
(Ronan and Bodznick, 1986).
    All bony fish with electroreception possess cathodally stimulated recep-
tors with the exception of some members of the neopterygians (four groups
of teleosts mentioned above), which are stimulated by anodal signals
3.   NERVOUS AND SENSORY SYSTEMS                                                                        161

                                                                         al       ml ST PT
                                                             SO                    T
                                                                        OT                   ll
                                          IO                                sp
                                     MD                                hl


                                                                                                  10 mm

Fig. 3.9. Drawing of the lateral view of the head of the Florida gar, L. platyrhinchus, illustrating
the relative positions and extent of the cephalic superficial (pit) lines and canals of the lateral line.
al, anterior pit line; dl, dorsal trunk pit line; gl, gular pit line; hl, horizontal pit line; IO,
infraorbital canal; ll, lateral trunk pit line; MD, mandibular canal; mdl, first mandibular motor
ramus of trigeminal nerve; ml, middle pit line; OT, otic canal; PO, preopercular canal; PT,
posttemporal canal; SO, supraorbital canal; sp, spiracular diverticulum; ST, supratemporal
commissure (or canal); T, temporal canal; TL, trunk canal; and vl, vertical pit line. [Reproduced
from Song and Northcutt (1991b) with kind permission of S. Karger AG, Basel.]

(Northcutt, 1986a). The electroreceptors occur in two types (ampullary and
tuberous organs) but only the ampullary receptors are present in primitive
fishes. Evolved for the detection of weak electromagnetic fields produced by
living organisms (and inanimate sources), electroreception was thought to be
present in the ancestor of the earliest vertebrates and was reinvented after
the ancestors of the teleost fishes lost the sense (New, 1997). Electrical
fields provide a rich source of information about prey and mate locations
(Tricas et al., 1995), local electrogenic landmarks (Peters and Bretschneider,
1972; Pals et al., 1982), and the animal’s orientation with respect to currents
induced by the earth’s magnetic field (Kalmijn, 1982; Paulin, 1995). Some
teleosts are able to generate electrogenic signals or electric organ discharges
for interspecific communication (mediated by tuberous organs) but this will
not be discussed further.

8.1. Structure, Function, and Evolution of Ampullary Receptors

    Ampullary receptor organs are superficial structures embedded within the
epidermis and connected to the surface by a canal or ampullary pore filled
with a mucopolysaccharide gel. At the base of the canal is an ovoid capsule
containing aggregations of receptor cells, each of which supplies an aVerent
axon that projects to the medulla. The length of the canal changes according
to the position of the receptors over the head and, at least in elasmobranchs,
to environmental factors such as osmoregulatory constraints and concomi-
tant changes in skin resistance (Kalmijn, 1982; Raschi and Mackanos, 1989).
162                                                           SHAUN P. COLLIN

     Lamprey electroreceptors were once considered to be taste buds
(Johnston, 1902) and were subsequently termed end buds. However, they
are considered to be homologous to electroreceptors (Bodznick and
Northcutt, 1981; Bullock et al., 1982, 1983; Fritzsch et al., 1984; Ronan
and Bodznick, 1986). Scattered over the head and body trunk, each organ
comprises between 3 and 25 sensory cells surrounded by support cells.
Each ampullary organ usually comprises a single kinocilium surrounded by
numerous stereocilia/microvilli (as found in cladistians and dipnoans,
Jorgensen, 1984), although in lampreys, elasmobranchs, and chondrosteans,
a kinocilium is lacking (Waltman, 1966; Ronan and Bodznick, 1986). There
is a wide variation in the number and distribution of electroreceptor organs
(see review by Collin and Whitehead, 2004 for elasmobranchs), ranging from
6 tubules in the rostral organ of the coelacanth (Millot and Anthony, 1956)
to 75,000 in the paddlefish, Polyodon (Jorgensen et al., 1972; Pettigrew
and Wilkens, 2003) (Figure 3.10). The ciliated receptors in the coelacanth
lie within a chamber at the base of three tubules within the ethmoid chon-
drocranium on each side of the head (Bemis and Hetherington, 1982;
Jorgensen, 1991). The octavolateralis nucleus in the brain of L. chalumnae
is also hypertrophied, indicative of a well‐developed electroreceptive capacity
(Northcutt, 1980).
     Ampullary electroreceptors are broadly tuned to low‐frequency electric
fields from less than 0.1 to 25 Hz. This low‐frequency range of sensitivities
corresponds well to the frequency ranges of standing or modulated fields
produced by aquatic environments (New, 1997). Ampullary receptors are
tonic receptors with a long‐lasting response to low‐frequency stimulation
and possess threshold sensitivities of less than 20 nV cmÀ1 in elasmobranchs
to 100 mV cmÀ1 in other taxa (New, 1997; Tricas and New, 1998). At least in
elasmobranchs, the ampullae of Lorenzini are also thought to detect changes
in ambient water temperature using the glycoprotein‐based (semiconductor‐
like) gel within the epidermal canal to elicit a thermoelectric signal (Brown,
2003). Ampullary organs appear to be the ancestral condition and are restrict-
ed to the head, except in lungfishes, which also possess trunk and tail ampul-
lary organs (Gibbs, 2004). The loss of electroreception in the neopterygians is
thought to be pleiotropically linked to the reduction of the cranial dermal
armor (Moy‐Thomas and Miles, 1971; Lauder and Liem, 1983), a significant
reorganization of the bones of the cranium and a reduction in size and
thickness of the scales (Northcutt and Gans, 1983; Northcutt, 1986a). The
evolutionary mechanism underlying the reinvention of electroreception
(based on the ampullary system of receptors) in the teleosts, that is in the
mormyrid, gymnotiform, gymnarchid, and siluriform lineages, is not well
understood. Specialization of the superficial neuromasts of the lateral line,
which underwent subsequent changes to give rise to tuberous organs, has
been hypothesized (New, 1997), although electroreceptors do not exhibit a
3.   NERVOUS AND SENSORY SYSTEMS                                                                    163

 A                                                                            B   c            c


 C                                                             D                                   Hz 80
                  Pore                                                                                60
                           Skin                                                                       40
                           Canal                                                                      20
                           Hair cells in epithelium
                          synapses                                 Afferent
                           SIZ ?

                               Afferent                            Canal 1                           200
                               axon                   Spikes                                         µV
         e                             Micro
                                   electrode    ALLn
     e                                                 Brain       Canal 2                          0.2 s
                   of canals
  0.5 mm                                                                                            0.1 s
             Receptive Head


Fig. 3.10. (A) A paddlefish, P. spathula, attacking dipole wires (asterisk) that are mimicking the
electrical discharges of live prey. (B) A schematic diagram of the ampullary electroreceptors found
in the rostrum of the paddlefish. Note that primary aVerent nerve endings PA1 and PA2 innervate
ciliated sensory epithelial cells, which mediate signal transduction. C, cilia; M, microvilli; NE, nerve
endings; RC, receptor cell; and SC, supporting cell. (C) Diagram of an electroreceptor in the
paddlefish rostrum. e, photograph of pipette electrodes in two canals of one large cluster, which is
the receptive field of the aVerent in (D) and (E). SIZ, presumed spike initiating zone. (D) Raw
recording of the spontaneous firing of an electroreceptor aVerent with its spontaneous firing
frequency (top trace) and simultaneous pipette recordings from two canals. (E) Expanded segment
showing canal oscillations at 26.7 Hz (middle trace) and 53 Hz (bottom trace, left). [(A) and (B) are
from Pettigrew and Wilkins (2003) and are reproduced with kind permission of Springer Science
and Business Media. (C), (D), and (E) are reproduced from Neiman and Russell (2004).]

neuromast‐like developmental stage and it may be that developing ampullary
electroreceptors have arisen from epithelial cells overlying a new class of
aVerent nerve fiber.

8.2. Role in Passive Electrolocation

    The ampullary electroreceptors are used to detect animate and inanimate
electric fields, by measuring minute changes in potential between the water
at the skin surface and the basal surface of the receptor cells. Epidermal pores
164                                                             SHAUN P. COLLIN

and the jelly‐filled canals comprising each electroreceptor organ ensure
that the potential within the ampullary lumen is the same as that at the
surface. The hair cells of each receptor act as voltage detectors and release
neurotransmitter onto the primary aVerent neurons according to the diVer-
ence between the basal and apical potentials (Tricas, 2001). The primary
aVerent neurons encode stimulus amplitude and frequency data that is sent
to the brain (Montgomery, 1984; Tricas and New, 1998), where a sophisti-
cated set of filter mechanisms are used for extracting the weak electrosensory
signals from a much stronger background noise, predominantly created by
the animal’s own movements (see review by Bodznick et al., 2003). Therefore,
the distribution of the ampullary organs may provide information about the
electric field’s intensity, its spatial configuration, and possibly the direction of
its source (Tricas, 2001). The behavioral relevance of this level of sensitivity
was uncovered in a series of experiments involving both experimental and
free‐living elasmobranchs, which induced feeding responses toward either
buried fish or a pair of buried electrodes, that could not otherwise be detected
using other sensory modalities. Feeding was subsequently terminated when
the bioelectric field of either source was masked by thin plastic film (Kalmijn,
     Therefore, the high sensitivity of electroreceptors enables elasmobranchs
to localize prey by detecting the very faint potentials associated with the ionic
leakage of the gills (modulated by ventilatory movements) of buried teleosts.
Wounded crustaceans produce higher bioelectric fields (1000 mV cmÀ1).
Bodznick and Northcutt (1981) revealed that lampreys (Lampetra tridentata)
possess sensitivity thresholds (0.1 mV cmÀ1) comparable to electrosensory
freshwater teleosts. Watt et al. (1999) have also shown that the ampullary
organs in the Australian lungfish, N. forsteri, use passive electroreception to
perceive weak electric fields emanating from hidden prey in much the same
way as elasmobranchs.
     The rostrum of the paddlefish, P. spathula, possesses a rostrum (not
unlike that of the platypus, Pettigrew and Wilkens, 2003) that is adorned
with ampullary electroreceptors, which act as an antenna (Figure 3.10B). The
rostrum is considered a sensory device with suYcient sensitivity to detect the
electric fields of planktonic prey with a sensitivity threshold of 10 mV cmÀ1, a
considerably higher sensitivity than the sensitivity of individual electrorecep-
tors (Wilkens et al., 1997; Russell et al., 1999). Paddlefishes use this rostrum
to laterally strike at planktonic prey using its electric sense passively without
the use of visual, chemical, and hydrodynamic senses at distances of 8–9 cm
(Wilkens et al., 2001) (Figure 3.10A). Higher concentrations of receptors
along the edges of the rostrum and its saccade‐like motion through the water
may serve to enhance prey detection by increasing the width of the electrical
scan field (Pettigrew and Wilkens, 2003). It appears that a single receptor
traversing an electrical field receives an electrical signal over time that
3.   NERVOUS AND SENSORY SYSTEMS                                            165

contains suYcient information necessary for the dorsal octavolateral nucleus
to extract the location, size, and orientation of a source, rather than relying
on topographic input (Hofmann et al., 2005). Recent findings of two noise
oscillators in the electroreceptors in paddlefish, one in the sensory epithelia
and the other in the aVerent terminals, reveal that there are mechanisms for
(1) driving spontaneous firing, conferring the advantage that both inhibitory
and excitatory stimuli can be detected (aVerent oscillator); (2) dealing with
high levels of synaptic convergence ($45,000 hair cells onto 1 aVerent axon)
(aVerent oscillator); (3) increasing receptor sensitivity near threshold by
mediating stochastic resonance (epithelial oscillator); and (4) encoding
water temperature by frequency modulation of aVerent firing (epithelial
oscillator) (Neiman and Russell, 2004, 2005) (Figure 3.10C–E).
    Behavioral response to weak cathodal stimulation has been confirmed
in the sea lamprey, P. marinus, with a threshold sensitivity of 30 mV cmÀ1
(Chung‐Davidson et al., 2004). Using neuronal activity markers such as Fos,
FosB, and Jun, it has also been established that, in addition to the involve-
ment of the octavolateralis region of the medulla and the torus semicircularis
(Gonzalez et al., 1999), the habenula‐fasciculus retroflexus‐interpeduncular
nucleus system also receives input in this species.


     In this chapter, an attempt has been made to reveal the evolution and
complexity of the nervous and sensory systems in primitive fishes. Drawing
on detailed information available on the neuroanatomy of specific represen-
tatives, structure–function relationships have been developed wherever pos-
sible in order to gain insights into the physiology. Unfortunately, this has not
always been possible for each of the major groups of fishes. Inevitably, and
due in part to the accessibility of both the animals and the neural system(s)
under consideration, detailed analyses of some species have not been under-
taken. However, in some neural systems, a wealth of information at the
neuroanatomical, neurochemical, physiological, and molecular levels has
greatly enhanced our understanding of brain evolution and the mechanisms
regulating the genesis of new neural tissues as opposed to the modification of
existing neural organization. However, physiological studies are remarkably
still scarce and need to be undertaken in order to trace the origins of craniate
brains and the evolutionary constraints placed on neural plasticity. In com-
bination with molecular and embryological investigation, a multidisciplinary
approach will allow a more comprehensive understanding of brain complex-
ity, function, and therefore behavior. A clear understanding of their behavior
in the natural environment can also help protect these unique animals that
hold so many keys to brain evolution.
166                                                                           SHAUN P. COLLIN


Allis, E. P. (1889). The anatomy and development of the lateral line system in Amia calva.
     J. Morphol. 2, 463–566.
Alvarez‐Viejo, M., Cemuda‐Cermuda, R., Alverez‐Lopez, C., and Garcia‐Fernandez, J. M.
     (2004). Identification of extraretinal photoreceptors in the teleost fish Phoxinus phoxinus.
     Histol. Histopathol. 19, 487–494.
Andres, K. H., and von During, M. (1993). Lamellated receptors in the skin of the hagfish,
     Myxine glutinosa. Neurosci. Lett. 151, 74–76.
Baatrup, E. W. (1985). Physiological studies on the pharangeal terminal buds in the larval brook
     lamprey, Lampetra planeri (Bloch). Chem. Senses 10, 549–558.
Baatrup, E. W., and Doving, K. B. (1985). Physiological studies on solitary receptors of the
     oral disc papillae in the adult brook lamprey, Lampetra planeri (Bloch). Chem. Senses 10,
Bailes, H. J., Robinson, S. R., Trezise, A. E. O., and Collin, S. P. (2006). Morphology,
     characterization and distribution of retinal photoreceptors in the Australian lungfish Neo-
     ceratodus forsteri (KreVt, 1870). J. Comp. Neurol. 494, 381–397.
Bantseev, V., Auclair, F., Dubuc, R., and Sivak, J. G. (2005). Optical quality of the ocular lens of
     the sea lamprey (Petromyzon marinus) during the mature and transformer periods of life.
     J. Comp. Physiol. A 191, 505–509.
Barbour, H. R., Archer, M. A., Hart, N. S., Thomas, N., Dunlop, S. A., Beazley, L. D., and
     Shand, J. (2002). Retinal characteristics of the ornate dragon lizard, Ctenophorous ornatus.
     J. Comp. Neurol. 450, 334–344.
Bashor, D. P., Beuerman, R. W., and Easton, D. M. (1974). Ciliary action normal movement
     of odorant wavefronts in garfish nasal capsule of hepisosteus osseus. Experientia, 30,
Bemis, W. E., and Burggren, W. W. (1986). The biology and evolution of lungfishes. J. Morphol.
     Suppl. 1, 1–83.
Bemis, W. E., and Hetherington, T. E. (1982). The rostral organ of Latimeria chalumnae:
     Morphological evidence of an electroreceptive function. Copeia 1982, 467–471.
Bernstein, P. (2003). The ear region of Latimeria chalumnae: Functional and evolutionary
     implications. Zoology (Jena) 106, 233–242.
Bertolucci, C., and Foa, A. (2004). Extraocular photoreception and circadian entrainment in
     non‐mammalian vertebrates. Chronobiol. Int. 21, 501–519.
Blaxter, J. H. S. (1987). Structure and development of the lateral line. Biol. Rev. 62, 471–514.
Bodznick, D., and Northcutt, R. G. (1981). Electroreception in lampreys: Evidence that the
     earliest vertebrates were electroreceptive. Science 212, 465–467.
Bodznick, D., Montgomery, J. C., and Tricas, T. C. (2003). Electroreception: Extracting beha-
     viourally important signals from noise. In ‘‘Sensory Processing in Aquatic Environments’’
     (Collin, S. P., and Marshall, N. J., Eds.), pp. 389–403. Springer, New York.
Bone, Q. (1963). Some observations upon the peripheral nervous system of the hagfish, Myxine
     glutinosa. J. Mar. Biol. Assoc. UK 43, 31–47.
Braun, C. B. (1996). The sensory biology of the living jawless fishes: A phylogenetic assessment.
     Brain Behav. Evol. 48, 262–276.
Braun, C. B., and Northcutt, R. G. (1997). The lateral line system of hagfishes (craniata:
     Myxinoidea). Acta Zool. 78, 247–268.
Brown, B. R. (2003). Sensing temperature without ion channels. Nature 421, 495.
Buck, L., and Axel, R. (1991). A novel multigene family may encode odorant recepors –
     A molecular basis for odor recognition. Cell 65, 175–187.
3.   NERVOUS AND SENSORY SYSTEMS                                                               167

Bullock, T. B., Bodznick, D., and Northcutt, R. G. (1983). The phylogenetic distribution of
    electroreception: Evidence for convergent evolution of a primitive vertebrate sense modality.
    Brain Res. 6, 25–46.
Bullock, T. H., and Heiligenberg, W. (Eds.) (1986). ‘‘Electroreception.’’ John Wiley and Sons,
    New York.
Bullock, T. H., Northcutt, R. G., and Bodznick, D. A. (1982). Evolution of electroreception.
    Trends Neurosci. 5, 50–53.
Burkhardt, D. T., Gottesman, J., Levine, J. S., and MacNichol, E. F., Jr. (1983). Cellular
    mechasnisms for color‐coding in holostean retinas and the evolution of color vision. Vision
    Res. 23, 1031–1041.
Butler, A. B., and Hodos, W. (1996). ‘‘Comparative Vertebrate Neuroanatomy. Evolution and
    Adaptation.’’ Wiley‐Liss, New York.
Butler, A. B., and Northcutt, R. G. (1992). Retinal projections in the bowfin, Amia calva:
    Cytoarchitectonic and experimental analysis. Brain Behav. Evol. 39, 169–194.
Caprio, J. (1980). Similarity of olfactory receptor responses (EOG) of freshwater and marine
    catfish to amino acids. Can. J. Zool. 58, 1778–1784.
Caprio, J., Brand, J. G., Teeter, J. H., Valentincic, T., Kalinoski, D. L., Kohbara, J.,
    Kumazawa, T., and Wegert, S. (1993). The taste system of the channel catfish: From
    biophysics to behavior. Trends Neurosci. 16, 192–197.
Chiba, A. (1997). Distribution of neuropeptide Y‐like immunoreactivity in the brain of the bichir,
    Polypterus senegalus, with special regard to the terminal nerve. Cell Tissue Res. 289, 275–284.
Chiba, A. (2005). Neuropeptide Y‐immunoreactive (NPY‐ir) structures in the brain of the gar
    Lepisosteus oculatus (Lepisosteiformes, Osteichthyes) with special regard to their anatomical
    relations to gonadotropin‐releasing hormone (GnRH)‐ir structures in the hypothalamus
    and the terminal nerve. Gen. Comp. Endocrinol. 142, 336–346.
Chung‐Davidson, Y.‐W., Yun, S.‐S., Teeter, J., and Li, W. (2004). Brain pathways and beha-
    vioiural responses to weak electric fields in parasitic sea lampreys (Petromyzon marinus).
    Behav. Neurosci. 118, 611–619.
Collin, S. P. (1988). The retinal structure of the shovel‐nosed ray, Rhinobatos batillum (Rhino-
    batidae). Morphology and quantitative analysis of ganglion, amacrine and bipolar cell
    populations. Exp. Biol. 47, 195–207.
Collin, S. P. (1989). Anterograde labelling from the optic nerve reveals multiple central targets in
    the teleost, Lethrinus chrysostomus (Perciformes). Cell Tissue Res. 256, 327–335.
Collin, S. P. (1997). Specializations of the teleost visual system: Adaptive diversity from shallow‐
    water to deep‐sea. Acta Physiol. Scand. 161(Suppl. 638), 5–24.
Collin, S. P. (1999). Behavioural ecology and retinal cell topography. In ‘‘Adaptive Mechanisms
    in the Ecology of Vision’’ (Archer, S. N., Djamgoz, M. B. A., Loew, E., Partridge, J. C., and
    Vallerga, S., Eds.), pp. 509–535. Kluwer Academic Publishers, Great Britain.
Collin, S. P., and Collin, H. B. (1993). The visual system of the Florida garfish, Lepisosteus
    platyrhincus. (Ginglymodi): II. Cornea and lens. Brain Behav. Evol. 42, 98–115.
Collin, S. P., and Collin, H. B. (2001). The fish cornea: Adaptation for diVerent aquatic
    environments. In ‘‘Sensory Biology of Jawed Fishes. New Insights’’ (Kapoor, B. G., and
    Hara, T. J., Eds.), pp. 57–96. Science Publishers Inc., Enfield (NH), USA.
Collin, S. P., and Collin, H. B. (2006). The corneal epithelial surface in the eyes of vertebrates:
    Environmental and evolutionary influences. J. Morphol. 267, 273–291.
Collin, S. P., and Northcutt, R. G. (1993). The visual system of the Florida garfish, Lepisosteus
    platyrhincus (Ginglymodi): III. Retinal ganglion cells. Brain Behav. Evol. 42, 295–320.
Collin, S. P., and Northcutt, R. G. (1995). The visual system of the Florida garfish, Lepisosteus
    platyrhincus (Ginglymodi): IV. Bilateral projections and the binocular visual field. Brain
    Behav. Evol. 45, 34–53.
168                                                                            SHAUN P. COLLIN

Collin, S. P., and Potter, I. C. (2000). The ocular morphology of the southern hemisphere
    lamprey Mordacia mordax Richardson with special reference to a single class of photore-
    ceptor and a retinal tapetum. Brain Behav. Evol. 55, 120–138.
Collin, S. P., and Shand, J. (2003). Retinal sampling and the visual field in fishes. In ‘‘Sensory
    Processing in Aquatic Environments’’ (Collin, S. P., and Marshall, N. J., Eds.), pp. 139–169.
    Springer, New York.
Collin, S. P., and Trezise, A. E. O. (2006). Evolution of colour discrimination in vertebrates and
    its implications for visual communication. In ‘‘Communication in Fishes’’ (Ladich, F,
    Collin, S. P., Moller, P., and Kapoor, B. G., Eds.), Vol. 2, pp. 303–335. Science Publishers
    Inc., Plymouth.
Collin, S. P., and Whitehead, D. (2004). The functional roles of electroreception in non‐electric
    fishes. Animal Biol. 54, 1–25.
Collin, S. P., Potter, I. C., and Braekevelt, C. R. (1999). The ocular morphology of the
    southern hemisphere lamprey Geotria australis Gray, with special reference to optical specia-
    lizations and the characterization and phylogeny of photoreceptor types. Brain Behav. Evol. 54,
Collin, S. P., Shand, J., Tee, L. B. G., and Beazley, L. D. (2001). Transient ipsilateral projections
    in a developing teleost is related to the relocation of the area centralis. Proc. Aust. Neurosci-
    ence Soc. 12, 179.
Collin, S. P., Hart, N. S., Shand, J., and Potter, I. C. (2003a). Morphology and spectral
    absorption characteristics of retinal photoreceptors in the southern hemisphere lamprey
    (Geotria australis). Vis. Neurosci. 20, 119–130.
Collin, S. P., Knight, M. A., Davies, W. L., Potter, I. C., Hunt, D. M., and Trezise, A. E. O.
    (2003b). Ancient colour vision: Multiple opsin genes in the ancestral vertebrates. Curr. Biol.
    13, R864–R865.
Cook, J. E., Becker, D. L., and Kapila, R (1992). Independent mosaics of large inner‐ and outer‐
    stratified ganglion cells in the goldfish retina. J. Comp. Neurol. 318, 355–366.
Coombs, S., Janssen, J., and Webb, J. F. (1988). Diversity of lateral line systems: Evolu-
    tionary and functional considerations. In ‘‘Sensory Biology of Aquatic Animals’’
    (Atema, J., Fay, R. R., Popper, A. N., and Tavolga, W. N., Eds.), pp. 553–593.
    Springer, New York.
Corwin, J. T. (1977). Morphology of the macula neglecta in sharks of the genus Carcharhinus. J.
    Morphol. 152, 341–362.
Corwin, J. T. (1981). Peripheral auditory physiology in the lemon shark: Evidence of parallel
    otolithic and non‐otolithic sound detection. J. Comp. Physiol. 142, 379–390.
Corwin, J. T. (1983). Postembryonic growth of the macula neglecta auditory detector in the ray,
    Raja clavata: Continual increases in hair cell number, neural convergence, and physiological
    sensitivity. J. Comp. Neurol. 217, 345–356.
Crescitelli, F. (1972). The visual cells and visual pigments of the vertebrate eye. In ‘‘Handbook of
    Sensory Physiology, VII/1, Photochemistry of Vision’’ (Dartnall, H. J. A., Ed.),
    pp. 245–363. Springer‐Verlag, Berlin.
Dartnall, H. J. A. (1972). Visual pigment of the coelacanth. Nature 239, 341–342.
David, C., Frank, C. L., Lukats, A., Szel, A., and Vigh, B. (2003). Cerebrospinal fluid contacting
    neurons in the reduced brain ventricular system of the Atlantic hagfish, Myxine glutinosa.
    Acta Biol. Hung. 54, 35–44.
Degliagina, T. G. (1997). Vestibular compensation in lampreys: Impairment and recovery of
    equilibrium control during locomotion. J. Exp. Biol. 200, 1459–1471.
Degliagina, T. G., Grillner, S., Orlovsky, G. N., and Ullen, F. (1993). Visual input aVects the
    response to roll in reticulospinal neurons of the lamprey. Exp. Brain Res. 95, 421–428.
Demski, L. S. (1993). Terminal nerve complex. Acta Anat. (Basel.) 148, 81–95.
3.   NERVOUS AND SENSORY SYSTEMS                                                              169

Derivot, J. H., Mattei, X., Godet, R., and Dupe, M. (1979). Ultrastructural study of apical
     region of olfactory epithelial cells of Protopterus annectens Owen (Dipneustes). J. Ultra-
     struct. Res. 66, 22–31.
Derobert, Y., Baratte, B., Lepage, M., and Mazan, S. (2002). Pax6 expression patterns in
     Lampetra fluviatilis and Scyliorhinus canicula embryos suggest highly conserved roles in
     the early regionalization of the vertebrate brain. Brain Res. Bull. 57, 277–280.
Dickson, D. H., and Graves, D. A. (1979). Fine structure of the lamprey photoreceptors and
     retinal pigment epithelium (Petromyzon marinus L.). Exp. Eye Res. 29, 45–60.
Dijkgraaf, S. (1963). The functioning and significance of the lateral line organs. Biol. Rev. 38,
Doving, K. B., and Holmberg, K. (1974). A note on the function of the olfactory organ of the
     hagfish Myxine glutinosa. Acta Physiol. Scand. 91, 430–432.
Eddy, J. M. P., and Strahan, R. (1968). The role of pineal complex in the pigmentary effector
     system of the lampreys, Mordacia mordax (Richardson) and Geotria australis. Grey. Gen.
     Comp. Endocrin. 11, 528–534.
Eisthen, H. L., and Northcutt, R. G. (1996). Silver lampreys (Ichthyomyzon unicuspis) lack a
     gonadotropin‐releasing hormone‐ and FMRF‐amide‐immunoreactive terminal nerve.
     J. Comp. Neurol. 370, 159–172.
Finger, T. E. (1983). The gustatory system in teleost fish. In ‘‘Fish Neurobiology’’
     (Northcutt, R. G., and Davis, R. E., Eds.), Vol. 1, pp. 285–310. University of Michigan
     Press, Ann Arbor.
Finger, T. E. (1997). Evolution of taste and solitary chemoreceptor cell systems. Brain Behav.
     Evol. 50, 234–243.
Foster, R. G., and Provencio, I. (1999). The regulation of vertebrate biological clocks by light.
     In ‘‘Adaptive Mechanisms in the Ecology of Vision’’ (Archer, S. N., Djamgoz, M. B. A.,
     Loew, E., Partridge, J. C., and Vallerga, S., Eds.), pp. 223–243. Kluwer Academic Publish-
     ers, Great Britain.
                                                                        ´       ´
Foster, R. G., Grace, M. S., Provincio, I., DeGrip, W. J., and Garcıa‐Fernandez, J. M. (1994).
     Identification of vertebrate deep brain photoreceptors. Neurosci. Biobehav. Rev. 18,
Freitag, J., Ludwig, G., Andreini, I., Rossler, P., and Breer, H. (1998). Olfactory receptors in
     aquatic and terrestrial vertebrates. J. Comp. Physiol. A 183, 635–650.
Fritzsch, B. (1987). Inner ear of the coelacanth fish Latimeria has tetrapod aYnities. Nature 327,
Fritzsch, B. (1991). Ontogenetic clues to the phylogeny of the visual system. In ‘‘The Changing
     Visual System’’ (Bagnoli, P., and Hodos, W., Eds.), pp. 33–49. Plenum Press, New York.
Fritzsch, B. (1998). Evolution of the vestibulo‐ocular system. Arch. Otolaryngol. Head Neck
     Surg. 119, 182–192.
Fritzsch, B. (2003). Commentary: The ear of Latimeria chalumnae revisited. Zoology 106,
Fritzsch, B., and Collin, S. P. (1990). The distribution and dendritic fields of two populations of
     ganglion cells and the retinopetal fibers in the retina of the lamprey, Ichthyomyzon unicuspis.
     Vis. Neurosci. 4, 533–545.
Fritzsch, B., and Northcutt, R. G. (1993). Origin and migration of trochlear, oculomotor and
     abducent motoneurons in Petromyzon marinus L. Brain Res. Dev. Brain Res. 74, 122–126.
                                                ¨                    ¨
Fritzsch, B., Crapon de Caprona, M. D., Wachtler, K., and Kortje, K. H. (1984). Neuroana-
     tomical evidence for electroreception in lampreys. Z. Naturhforsch. 39, 856–858.
Fritzsch, B., Dubuc, R., Ohta, Y., and Grillner, S. (1989). EVerents to the labyrinth of the river
     lamprey (Lampetra fluviatilis) as revealed with retrograde tracing techniques. Neurosci. Lett.
     96, 241–246.
170                                                                           SHAUN P. COLLIN

Frontini, A., Zaidi, A. U., Hua, H., Wolak, T. P., Greer, C. A., Kafitz, K. W., Li, W., and
    Zielinski, B. S. (2003). Glomerular territories in the olfactory bulb from the larval stage of
    the sea lamprey Petromyzon marinus. J. Comp. Neurol. 465, 27–37.
Gans, C., and Northcutt, R. G. (1983). Neural crest and the origin of vertebrates: A new head.
    Science 220, 268–274.
Garcia‐Fernandez, J. M., and Foster, R. G. (1994). Immunocytochemical identification of
    photoreceptor proteins in hypothalamic cerebrospinal fluid‐contacting neurons of the larval
    lamprey (Petromyzon marinus). Cell Tissue Res. 275, 319–326.
     ´        ´                  ´               ´
Garcıa‐Fernandez, J. M., Jiminez, A. J., Gonzalez, B., Pombal, M. A., and Foster, R. G. (1997).
    An immunocytochemical study of encephalic photoreceptors in three species of lamprey.
    Cell Tissue. Res. 288, 267–278.
Georgieva, V Patzner, R. A., and Adam, H. (1979). Trasnmissions‐ und raster‐
    elektronenmikroskopische Untersuchung an den Sinnesknospen der Tentaklen von Myxine
    glutinosa L. (Cyclostomata). Zool. Scripta 8, 61–67.
Gibbs, M. A. (2004). Lateral line receptors: Where do they come from developmentally and
    where is our research going? Brain Behav. Evol. 64, 163–181.
       ´            ´                  ´
Gonzalez, M. J., Yanez, J., and Anadon, R. (1999). AVerent and eVerent connections of the torus
    semicircularis in the sea lamprey: An experimental study. Brain Res. 826, 83–94.
Gorbman, A. (1995). Olfactory origins and evolution of the brain‐pituitary endocrine system:
    Facts and speculation. Gen. Comp. Endocrinol. 97, 171–178.
Gottesman, J., and Burkhardt, D. A. (1987). Response properties of C‐type horizontal cells in
    the retina of the bowfin. Vision Res. 27, 179–189.
Govardovskii, V. I., and Lychakov, D. V. (1984). Visual cells and visual pigments of the lamprey,
    Lampetra fluviatilis L. J. Comp. Physiol. A 154, 279–286.
Gradwell, N. (1972). Hydrostatic pressures and movements of the lamprey, Petromyzon during
    suction, olfaction, and gill ventilation. Can. J. Zool. 50, 1215–1223.
Gray, J. A. B. (1984). Interaction of sound pressure and particle acceleration in the excitation of
    lateral line neuromasts of sprats. Proc. R. Soc. Lond. B 220, 299–325.
Hafeez, M. A., and Merhige, M. E. (1977). Light and electron microscopic study on the pineal
    complex of the coelacanth, Latimeria chalumnae Smith. Cell Tissue Res. 178, 249–265.
Hansen, A., and Reutter, K. (2004). Chemosensory systems in fish: Structural, functional and
    ecological aspects. In ‘‘The Senses of Fishes: Adaptations for the Reception of Natural
    Stimuli’’ (von der Emde, G., Mogdans, J., and Kapoor, B. G., Eds.), pp. 55–89. Narosa
    Publishing House, New Delhi and Kluwer Academic Publishers, Dordrecht.
Hansen, A., Zeiske, E., and Reutter, K. (1994). Microvillous and ciliated receptor cells in the
    olfactory epithelium of the Australian lungfish, Neoceratodus forsteri (KreVt 1870). Adv.
    Biosci. 93, 43–51.
Hardisty, M. W., and Potter, I.C (1971). The behaviour, ecology and growth of larval lampreys.
    In ‘‘The Biology of Lampreys’’ (Hardisty, M. W., and Potter, I. C., Eds.), Vol. 1,
    pp. 127–206. Academic Press, London.
Harosi, F. I., and Kleinschmidt, J. (1993). Visual pigments in the sea lamprey, Petromyzon
    marinus. Vis. Neurosci. 10, 711–715.
Hart, N. S., Lisney, T. J., Marshall, N. J., and Collin, S. P. (2004). Multiple cone visual pigments
    and the potential for trichromatic colour vision in two species of elasmobranch. J. Exp. Biol.
    207, 4587–4594.
Hisatomi, O., Iwasa, I., and Tokunaga, F. (1988). The visual pigment of lamprey. In Molecular
    Physiology of Retinal Proteins, (Hara, T., Ed.), pp. 371–372. Yamada Science Foundation,
Hofmann, M. H., and Meyer, D. L. (1995). The extrabulbar olfactory pathway: Primary
    olfactory fibres bypassing the olfactory bulb in bony fishes? Brain Behav. Evol. 46, 378–388.
3.   NERVOUS AND SENSORY SYSTEMS                                                              171

Hofmann, M. H., Chagnaud, B., and Wilkens, L. A. (2005). Response properties of electro-
     sensory aVerent fibers and secondary brain stem neurons in paddlefish. J. Exp. Biol. 208,
Holmberg, K. (1977). The cyclostome retina. In ‘‘Handbook of Sensory Physiology: The Visual
     System in Vertebrates’’ (Mackay, D. M., and Teuber, H. L., Eds.), pp. 47–66. Springer‐
     Verlag, New York.
Howland, H. C., Merola, S., and Basarab, J. R. (2004). The allometry and scaling of the size of
     vertebrate eyes. Vision Res. 44, 2043–2065.
Ito, H., Yoshimoto, M., Albert, J. S., Yamamoto, N., and Sawai, N. (1999). Retinal projections
     and retinal ganglion cell distribution patterns in a sturgeon (Acipenser transmontanus), a
     non‐teleost actinopterygian fish. Brain Behav. Evol. 53, 127–141.
Johnston, J. B. (1902). The brain of Petromyzon. J. Comp. Neurol. 12, 1–86.
Jorgensen, J. M. (1982). Fine structure of the ampullary organs of bichir (Polypterus senegalus)
     with some notes on the phylogenetic development of electroreceptors. Acta Zool. 63, 211–217.
Jorgensen, J. M. (1984). On the morphology of the electroreceptors of the two lungfish: Neocer-
     atodus forsteri KreVt and Protopterus annectens Owen. Vidensk. Meddr dansk naturh. Foren
     145, 77–85.
Jorgensen, J. M. (1991). Regeneration of lateral line and inner ear vestibular cells. In ‘‘Regener-
     ation of Vertebrate Sensory Receptor Cells. Ciba Foundation Symposium,’’ pp. 151–170.
     Wiley, Chichester.
Jorgensen, J. M. (2005). Morphology of electroreceptive sensory organs. In ‘‘Electroreception’’
     (Bullock, T. H., Hopkins, C. D., Popper, A. N., and Fay, R. R., Eds.), pp. 47–67. Springer,
     New York.
Jorgensen, J. M., Flock, A., and Wersall, J. (1972). The Lorenzinian ampullae of Polyodon
     spathula. Z. Zellforsch. 130, 362–377.
Jorgensen, J. M., Shichiri, M., and Geneser, F. A. (1998). Morphology of the hagfish inner ear.
     Acta Zool. 79, 251–256.
Joss, J. M. P. (1973). Pineal complex, melatonin, and color change in lamprey Lampetra. Gen.
     Comp. Endocrinol. 21, 118–122.
Kabasawa, H., and Ooka‐Souda, S. (1989). Circadian rhythms in locomotor activity of the
     hagfish, Eptatretus burgeri (IV). The eVect of eye ablation. Zool. Sci. 6, 135–139.
Kalmijn, A. J. (1982). Electric and magnetic field detection in elasmobranch fishes. Science 218,
Kasumyan, A. O. (1997). Gustatory reception and feeding behaviour in fish. J. Ichthyol. 37,
Kasumyan, A. O., and Doving, K. B. (2003). Taste preferences in fish. Fish Fisheries 4, 289–347.
Kemp, A. (2000). Early develoment of neural tissues and mesenchyme in the Australian lungfish
     Neoceratodus forsteri (Osteichthyes: Dipnoi). J. Zool. Lond. 250, 347–372.
Kishida, R., Goris, R. C., Nishizawa, H., Koyama, H., Kadota, T., and Amemiya, F. (1987).
     Primary neurons of the lateral line nerves and their central projections in hagfishes. J. Comp.
     Neurol. 264, 303–310.
Kitamura, S., and Ogata, H. (1980). Olfactory responses of male loach, Misgurnus anguillicada-
     tus, to F‐type prostaglandins. Taste Smell 24, 163–166.
Kleerekoper, H., and van Erkel, G. A. (1960). The olfactory apparatus of Petromyzon marinus L.
     Can. J. Zool. 38, 209–223.
Koyanagi, M., Kawano, E., Kinugawa, Y., Oishi, T., Shichida, Y., and Tamotsu, S. (2004).
     Bistable UV pigment in the lamprey pineal. Proc. Natl. Acad. Sci. USA 101, 6687–6691.
Kroese, A. B. A., and Schellart, N. A. M. (1987). Evidence for velocity and acceleration sensitive
     unites in the trunk lateral line of the trout. J. Physiol. 393, 29P.
172                                                                            SHAUN P. COLLIN

Kro¨ger, R. H. H., Campbell, M. C. W., Fernald, R. D., and Wagner, H‐J. (1999). Multifocal lenses
    compensate for chromatic defocus in vertebrate eyes. J. Comp. Physiol. A 184, 361–369.
Lane, E. B., and Whitear, M. (1982). Sensory structures at the surface of fish skin. II. Lateralis
    system. Zool. J. Linn. Soc. 76, 19–28.
Larimore, R. W. (1949). Changes in the cranial nerves of the paddlefish, Polyodon spathula,
    accompanying development of the rostrum. Copeia 1949, 204–212.
Lauder, G. V., and Liem, K. F. (1983). The evolution and interrelationships of the actinopter-
    ygian fishes. Bull. Mus. Comp. Zool. 150, 95–197.
Lethbridge, R. C., and Potter, I. C. (1979). The oral fibriae of the lamprey, Geotria australis.
    J. Zool. 188, 267–277.
Lewis, E. R., Leveren, E. L., and Bialek, W. (1985). ‘‘The Vertebrate Inner Ear.’’ CRC Press,
    Boca Raton.
Li, W., Sorensen, P. W., and Gallaher, D. D. (1995). The olfactory system of migratory adult sea
    lamprey (Petromyzon marinus) is specifically and acutely sensitive to unique bile acids
    relaeased by conspecific larvae. J. Gen. Physiol. 105, 569–587.
Locket, N. A., and Jorgensen, J. M. (1998). The eyes of hagfishes. In ‘‘The Biology of Hagfishes’’
    (Jorgensen, J. M., Lomholt, J. P., Weber, R. E., and Malte, H., Eds.), pp. 541–556.
    Chapman and Hall, London.
Lovell, J. M., Findlay, M. M., Moate, R. M., Nedwell, J. R., and Pegg, M. A. (2005). The inner ear
    morphology and hearing abilities of the paddlefish (Polyodon spathula) and the lake sturgeon
    (Acipenser fulvescens). Comp. Biochem. Physiol. A Mol. Integr. Physiol. 142, 286–296.
Lowenstein, O. (1970). The electrophysiological study of the responses of the isolated labyrinth
    of the lamprey (Lampetra fluviatilis) to angular acceleration, tilting and mechanical vibra-
    tion. Proc. R. Soc. Lond. B Biol. Sci. 174, 419–434.
Lowenstein, O., and Thornhill, R. A. (1970). The labyrinth of Myxine: Anatomy, ultrastructure
    and electrophysiology. Proc. R. Soc. Lond. B. Biol. Sci. 176, 21–42.
Lowenstein, O., Osborne, M. P., and Wersall, J. (1964). Structure and innervation of the sensory
    epithelia of the labyrinth in the thornback ray (Raja clavata). Proc. R. Soc. Lond. B Biol. Sci.
    160, 1–12.
Lowenstein, O., Osborne, M. P., and Thornhill, R. A. (1968). The anatomy and ultrastructure of the
    labyrinth of the lamprey (Lampetra fluviatilis L.). Proc. R. Soc. Lond. B Biol. Sci. 170, 113–134.
Lu, Z., and Popper, A. N. (1998). Morphological polarization of sensory hair cells in the three
    otolithic organs of a teleost fish: Fluorescent imaging of ciliary bundles. Hear. Res. 126,
Mallat, J, and Ridgeway, R. L. (1984). Ultrastructure of a complex epithelial system: The
    Pharangeal lining of the larval lamprey, Petromyzon marinus. J. Morphol. 180, 271–296.
Matsuda, H., Goris, R. C., and Kishida, R. (1991). AVerent and eVerent projections of the
    glossopharyngeal‐vagal nerve in the hagfish. J. Comp. Neurol. 311, 520–530.
McCormick, C. A. (1981). Central projections of the lateral line and eighth nerves in the bowfin,
    Amia calva. J. Comp. Neurol. 197, 1–15.
McCormick, C. A. (1982). The organisation of the octavolateralis area in actinopterygian fishes:
    A new interpretation. J. Morphol. 171, 159–181.
McVean, A. (1991). The semicircular canals of the hagfish Myxine glutinosa. J. Zool. Lond. 224,
Meek, J., and Niewenhuys, R. (1998). Holosteans and teleosts. In ‘‘The Central Nervous System
    of Vertebrates’’ (Niewenhuys, H., Ten Donkelaar, J., and Nicholson, C., Eds.), Vol. 2,
    pp. 759–938. Springer‐Verlag, Berlin.
Meredith, G. E., and Butler, A. B. (1983). Organization of eighth nerve aVerent projections from
    individual endorgans of the inner ear of the teleost, Astronotus ocellatus. J. Comp. Neurol.
    220, 44–62.
3.   NERVOUS AND SENSORY SYSTEMS                                                              173

Millot, J, and Anthony, J. (1956). Note preliminaire sur le thymus et la Glande Thyroide de
    Latimeria chalumnae (crossopterygien coelacanthide). Comptes Rendus Hebdomadaires
    des seances de l’academie des sciences. 242, 560–561.
Millot, J., and Anthony, J. (1965). ‘‘Anatomie de Latimeria chalumnae. II. Systeme nerveux et
    organs de sense.’’ Centre Nationale de la Recherche Scientifique, Paris.
Montgomery, J. C. (1984). Frequency response characteristics of primary and seconday neurons
    in the electrosensory system of the thornback ray. Comp. Biochem. Physiol. A Physiol. 79,
Moy‐Thomas, J. A., and Miles, R. S. (1971). ‘‘Paleozoic Fishes.’’ W. B. Saunders, Philadelphia.
Munk, O. (1964). The eye of Calamoichthys calabaricus Smith, 1865 (Polypteridae, Pisces)
    compared with the eye of other fishes. Vidensk. Meddr. dansk naturh. Foren 127, 113–126.
Munk, O. (1968). The eyes of Amia and Lepisosteus (Pisces, Holostei) compared with the
    brachiopterygian and teleostean eyes. Vidensk. Meddr. dansk naturh. Foren. 131, 109–127.
Munk, O. (1969). On the visual cells of some primitive fishes with particular regard to the
    classification of rods and cones. Vidensk. Meddr. dansk naturh. Foren. 132, 25–30.
Muntz, W. R. A. (1976). The visual consequences of yellow filtering pigments in the eyes of fishes
    occupying diVerent habitats. In ‘‘Light as an Ecological Factor II’’ (Evans, G. C.,
    Bainbridge, R., and Rackman, O., Eds.), pp. 271–287. Blackwell, Oxford.
Murakami, Y., Uchida, K., Rijli, F. M., and Kuratani, S. (2005). Evolution of the brain
    developmental plan: Insights from agnathans. Dev. Biol. 280, 249–259.
Nag, T. C., and Bhattacharjee, J. (1995). Retinal ellipsosomes: Morphology, development,
    identification, and comparison with oil droplets. Cell. Tissue Res. 279, 633–637.
Neiman, A. B., and Russell, D. F. (2004). Two distinct types of noisy oscillators in electro-
    receptors of paddlefish. J. Neurophysiol. 92, 492–509.
Neiman, A. B., and Russell, D. F. (2005). Models of stochastic biperiodic oscillations and extended
    serial correlations in electroreceptors of paddlefish. Phys. Rev. E, 061915‐1‐061915‐10.
New, J. G. (1997). The evolution of vertebrate electrosensoy systems. Brain Behav. Evol. 50,
Nicol, J. A. C., and Arnott, H. J. (1973). Studies on the eyes of gars (Lepisosteidae) with special
    reference to the tapetum lucidum. Can. J. Zool. 51, 501–508.
Niewenhuys, R. (1977). The brain of the lamprey in a comparative perspective. Ann. N. Y. Acad.
    Sci. 299, 97–145.
Niewenhuys, R. (1998a). Lungfishes. In ‘‘The Central Nervous System of Vertebrates’’
    (Niewenhuys, H., Ten Donkelaar, J., and Nicholson, C., Eds.), Vol. 2, pp. 939–1006.
    Springer‐Verlag, Berlin.
Niewenhuys, R. (1998b). The coelacanth, Latimeria chalumnae. In ‘‘The Central Nervous System
    of Vertebrates’’ (Niewenhuys, H., Ten Donkelaar, J., and Nicholson, C., Eds.), Vol. 2,
    pp. 1007–1044. Springer‐Verlag, Berlin.
Niewenhuys, R. (1998c). Brachiopterygian fishes. In ‘‘The Central Nervous System of Verte-
    brates’’ (Niewenhuys, H., Ten Donkelaar, J., and Nicholson, C., Eds.), Vol. 1, pp. 655–700.
    Springer‐Verlag, Berlin.
Niewenhuys, R. (1998d). Chondrostean fishes. In ‘‘The Central Nervous System of Vertebrates’’
    (Niewenhuys, H., Ten Donkelaar, J., and Nicholson, C., Eds.), Vol. 1, pp. 701–757. Springer‐
    Verlag, Berlin.
Niewenhuys, R., and Nicholson, C. (1998). Lampreys, Pertomyzontoidea. In ‘‘The Central
    Nervous System of Vertebrates’’ (Niewenhuys, H., Ten Donkelaar, J., and Nicholson, C.,
    Eds.), Vol. 1, pp. 397–496. Springer‐Verlag, Berlin.
Niewenhuys, R., and Pouwels, E. (1983). The brain stem of actinopterygian fishes. In ‘‘Fish
    Neurobiology. 1. Brain Stem and Sense Organs’’ (Northcutt, R. G., and David, R. E., Eds.),
    pp. 25–87. University of Michigan Press, Ann Arbor.
174                                                                          SHAUN P. COLLIN

Norris, H. W. (1925). Observations on the peripheral distribution of the cranial nerves of certain
    ganoic fishes (Amia, Lepisosteus, Polyodon, Scaphirhynchus, and Acipenser). J. Comp.
    Neurol. 39, 345–432.
Northcutt, R. G. (1980). Anatomical evidence of electroreception in the coelacanth (Latimeria
    chalumnae). Zbl. Vet. Med. C. Anat. Histol. Embryol. 9, 289–295.
Northcutt, R. G. (1981). Audition in the central nervous systemof fishes. In ‘‘Hearing and Sound
    Communication in Fishes’’ (Tavolga, W. N., Popper, A. N., and Fay, R. R., Eds.),
    pp. 331–355. Springer‐Verlag, New York.
Northcutt, R. G. (1982). Localization of neurons afferent to the optic tectum in longnose gars.
    J. Comp. Neurol. 204, 325–335.
Northcutt, R. G. (1986a). Electroreception in non‐teleost bony fishes. In ‘‘Electroreception’’
    (Bullock, T. H., and Heiligenberg, W., Eds.), pp. 257–285. Wiley, New York.
Northcutt, R. G. (1986b). Lungfish neura characters and their bearing on sarcopterygian
    phylogeny. J. Morphol. Suppl. 1, 277–297.
Northcutt, R. G. (1989a). The brain and sense organs of the earliest vertebrates: Reconstruction
    of a morphotype. In ‘‘Evolutionary Biology of Primitive Fishes’’ (Foreman, R. E.,
    Gorbman, A., Dodd, J. M., and Olsson, R., Eds.), pp. 81–112. Plenum Press, New York.
Northcutt, R. G. (1989b). The phylogenetic distribution and innervation of craniate mechanor-
    eceptive lateral lines. In ‘‘The Mechanosensory Lateral Line: Neurobiology and Evolution’’
                     ¨                 ¨
    (Coombs, S., Gorner, P., and Munz, H., Eds.), pp. 17–78. Springer‐Verlag, New York.
Northcutt, R. G. (1996). The agnathan ark: Origin of cranaite brains. Brain Behav. Evol. 48,
Northcutt, R. G., and Grans, C. (1983). The genesis of neural crest and epidermal placodes‐A
    reinterpretation of vertebrate origins. Quart. Rev. Biol. 58, 1–28.
Northcutt, R. G. (2004). Taste buds: Development and evolution. Brain Behav. Evol. 64,
Northcutt, R. G., and Bemis, W. E. (1993). Cranial nerves of the coelacanth, Latimeria chalum-
    nae (Osteichthyes: Sarcopterygii: Actinistia), and comparions with other craniata. Brain
    Behav. Evol. 42(Suppl. 1), 1–76.
Northcutt, R. G., and Braford, M. R., Jr. (1980). New observations on the organization and
    evolution of the telencephalon of actinopterygian fishes. In ‘‘Comparative Neurology of the
    Telencephalon’’ (Ebbesson, S. O. E., Ed.), pp. 41–98. Plenum Press, New York.
Northcutt, R. G., and Butler, A. B. (1976). Retinofugal pathways in the longnose gar Lepisosteus
    osseus (Linnaeus). J. Comp. Neurol. 166, 1–16.
Northcutt, R. G., and Butler, A. B. (1992). Retinofugal and retinopetal projections in the green
    sunfish, Lepomis cyanellus. Brain Behav. Evol. 37, 333–354.
Northcutt, R. G., Neary, T. J., and Senn, D. G. (1978). Observations on the brain of the coelacanth
    Latimeria chalumnae: External anatomy and quantitative analysis. J. Morphol. 155, 181–192.
Pals, N., Valentijn, P., and Verwey, D. (1982). Orientation reactions of the dogfish, Scyliorhinus
    canicula to local electric fields. Neth. J. Zool. 32, 495–512.
Patzner, R. A. (1978). Experimental studies on the light sense in the hagfish, Eptatretus burgeri
    and Paramyxine atami (Cyclostomata). Helgl. wiss. Meeres. 31, 180–190.
Patzner, R. A., Georgieva, V., and Adam, H. (1977). Sinneszellen an den Tentakeln der
    Schleimaale Myxine glutinosa und Eptatretus burgeri (Cyclostomata), Sitzungbericht Osterr.
    Akad. Wiss. 5, 77–79.
Paulin, M. G. (1995). Electroreception and the compass sense of sharks. J. Theor. Biol. 174,
Parry, J. W. L., Carleton, K. L., Spady, T., Carboo, A., Hunt, D. M., and Bowmakor, J. K.
    (2005). Mix and match color vision: Tuning spectral sensitivity by differential opsin gene
    expression in Lake Malauri cichlids. Current Biol. 15, 1734–1739.
3.   NERVOUS AND SENSORY SYSTEMS                                                              175

Pavlova, E. L., Popova, L. B., Orlovsky, G. N., and Deliagina, T. G. (2004). Vestibular
     compensation in lampreys: Restoration of symmetry in reticulospinal commands. J. Exp.
     Biol. 207, 4595–4603.
Peters, R. C., and Bretschneider, F. (1972). Electric phenomena in the habitat of the catfish,
     Ictalurus nebulosus LeS. J. Comp. Physiol. 81, 345–362.
Pettigrew, J. D., and Wilkens, L. (2003). Paddlefish and platypus: Parallel evolution of passive
     electroreception in a rostral bill organ. In ‘‘Sensory Processing in Aquatic Environments’’
     (Collin, S. P., and Marshall, N. J., Eds.), pp. 420–433. Springer, New York.
PfeiVer, W. (1968a). Retina und retinomotorik der Dipnoi und Brachiopterygii. Zeitschrift
     Zellforsch 89, 62–72.
PfeiVer, W. (1968b). Die Fahrenholzschen Organe der Dipnoi und Brachiopterygii. Z. Zell-
     forsch. Mikrosk. Anat. 90, 127–147.
PfeiVer, W. (1969). Der Geruchssinn der Polypteridae (Pisces, Brachioptergii). Zeit. Vergl.
     Physiol. 63, 151–164.
Pinkus, F. (1894). Uber einen noch nicht beschreibenen Hirnnerv des Protopterus annectens.
     Anat. Anz. 9, 562–566.
Piotrowski, T., and Northcutt, R. G. (1996). The cranial nerves of the senegal bichi, Polypterus
     senegalus (Osteichthyes: Actinoptergii: Cladistia). Brain Behav. Evol. 47, 55–102.
Platel, R., and Delfini, C. (1981). L’encephalization chez la Myxine (Myxine glutinosa L.).
     Analyse quantifieedes principales subdivisions encephaliques. Cah. Biol. Mar. 22, 407–430.
                        ´                                ´
Platel, R., and Delfini, C. (1986). L’encephalisation chez la lamproie marine, Petromyzon marinus
     (L.). Analyse quantifiee des principales subdisions encephaliques. J. Hirnforsch. 27, 279–293.
                            ´                               ´
Platt, C. (1983a). Equilibrium in the vertebrates: Signals, senses and steering underwater.
     In ‘‘Sensory Biology of Aquatic Animals’’ (Atema, J., Fay, R. R., Popper, A. N., and
     Tavolga, W. N., Eds.), pp. 783–809. Springer, New York.
Platt, C. (1983b). The peripheral vestibular system of fishes. In ‘‘Fish Neurobiology, Vol. 1, Brain
     Stem and Sense Organs’’ (Northcutt, R. G., and Davis, R. E., Eds.), pp. 89–123. The
     University of Michigan Press, Ann Arbor.
Platt, C., and Popper, A. N. (1981). Structure and function in the ear. In ‘‘Hearing and Sound
     Communication in Fishes’’ (Tavolga, W. N., Popper, A. N., and Fay, R. R., Eds.), pp. 3–38.
     Springer‐Verlag, New York.
Platt, C., Popper, A. N., and Fay, R. R. (1989). The ear as part of the octavolateralis system.
     In ‘‘The Mechanosensory Lateral Line: Neurobiology and Evolution’’ (Coombs, S.,
        ¨                 ¨
     Gorner, P., and Munz, H., Eds.), pp. 633–651. Springer‐Verlag, New York.
Platt, C., Jorgensen, J. M., and Popper, A. N. (2004). The inner ear of the lungfish Protopterus.
     J. Comp. Neurol. 471, 277–288.
Popper, A. N. (1977). Structure and function of the elasmobranch auditory system. Am. Zool.
     17, 443–452.
Popper, A. N. (1978). Scanning electron microscopic study of the otolithic organs in the bichir
     (Polypterus bichir) and shovel‐nose sturgeon (Scaphirhynchus platorynchus). J. Comp.
     Neurol. 181, 117–128.
Popper, A. N. (1983). Organisation of the inner ear and auditory processing. In ‘‘Fish Neurobi-
     ology and Behavior’’ (Northcutt, R. G., and Davis, R. E., Eds.), pp. 125–178. University of
     Michigan Press, Ann Arbor.
Popper, A. N., and Fay, R. R. (1993). Sound detection and processing by fish: Critical review and
     major research questions. Brain Behav. Evol. 41, 14–38.
Popper, A. N., and Hoxter, B. (1987). Sensory and non‐sensory ciliated cells in the ear of the sea
     lamprey, Petromyzon marinus. Brain Behav. Evol. 30, 43–61.
Popper, A. N., and Northcutt, R. G. (1983). Structure and innervation of the inner ear of the
     bowfin, Amia calva. J. Comp. Neurol. 213, 279–286.
176                                                                         SHAUN P. COLLIN

Popper, A. N., and Platt, C. (1993). Inner ear and lateral line of bony fishes. In ‘‘The Physiology
    of Fishes’’ (Evans, D. H., Ed.), pp. 99–136. CRC Press, Boca Raton, Florida.
Popper, A. N., Platt, C., and Edds, P. L. (1992). Evolution of the vertebrate inner ear: An
    overview of ideas. In ‘‘The Evolutionary Biology of Hearing’’ (Webster, D. B., Fay, R. R.,
    and Popper, A. N., Eds.), pp. 49–57. Springer‐Verlag, New York.
Raschi, W., and Mackanos, L. A. (1989). The structure of the ampullae of Lorenzini in Dasyatis
    garouaensis and its implications on the evolution of freshwater electroreceptive systems.
    J. Exp. Zool. 2, 101–111.
Retzius, G. (1881). ‘‘Das Gehororgan der Wirbeltiere.’’ Samson and Wallin, Stockholm.
Retzius, G. (1892). Zur Kenntnis der motorischen Nervendigungungen. Biol. Untersuch. N. F.
    Bd. 3, 41–52.
Reutter, K., and Hansen, A. (2005). Subtypes of light and dark elongated taste bud cells in fish.
    In ‘‘Fish Chemosenses’’ (Reutter, K., and Kapoor, B. G., Eds.), pp. 212–230. Science
    Publishers,Inc., Plymouth, UK.
Reutter, K., and Witt, M. (1993). Morphology of vertebrate taste organs and their nerve supply.
    In ‘‘Mechanisms of Taste Transduction’’ (Simon, S. A., and Roper, S. D., Eds.), pp. 30–82.
    CRC Press, Boca Raton, Florida.
Reutter, K., Boudriot, F., and Witt, M. (2000). Heterogeneity of fish taste bud ultrastructure as
    demonstrated in the holosteans Amia calva and Lepisosteus oculatus. Philos. Trans. R. Soc.
    Lond. B 355, 1225–1228.
Robinson, S. R. (1994). Early vertebrate colour vision. Nature 367, 121.
Ronan, M. C., and Bodznick, D. (1986). Endbuds: Non‐ampullary electroreceptors in adult
    lampreys. J. Comp. Physiol. 158, 9–16.
Ronan, M. C., and Bodznick, D. (1991). Behaviural and neurophysiological demonstration of a
    lateralis skin photosensitivity in larval sea lampreys. J. Exp. Biol. 161, 97–117.
Ronan, M. C., and Northcutt, R. G. (1983). Primary projections of the lateral line nerves in the
    northern silver lamprey. Soc. Neurosci. 8, 764.
Rovainen, C. M. (1979). Electrophysiology of vestibulospinal and vestibuloreticulospinal system
    in lampreys. J. Neurophysiol. 42, 745–766.
Rovainen, C. M. (1980). Neurophysiology of lampreys. Can. J. Fish. Aquat. Sci. 37, 1723–1738.
Rovainen, C. M., and Yan, QW. (1985). Sensory responses of dorsal cells in the lamprey brain.
    J. Comp. Physiol. A 156, 181–183.
Rubinson, K. (1990). The developing visual system and metamorphosis in the lamprey.
    J. Neurobiol. 21, 1123–1135.
Russell, D. F., Wilkens, L. A., and Moss, F. (1999). Use of behavioural stochastic resonance by
    paddlefish for feeding. Nature 402, 291–294.
Schoen, L., and von Holst, E. (1950). Das Zusammenspiel von Lagena und Utriculus bei der
    Lageorientierung der Knochenfische. Z. Vergl. Physiol. 32, 552–571.
Schreiner, K. E. (1919). Zur Kenntis der Zellgranula. Untersuchungen uber den feineren Bau der
    Haut von Myxine glutinosa. Archiv. Mikr. Anat. 92, 1–63.
Schuijf, A., and Buwalds, R. J. A. (1980). Underwater localisation: A major problem in fish
    acoustics. In ‘‘Comparative Studies of Hearing in Vertebrates’’ (Popper, A. N., and
    Fay, R. R., Eds.), pp. 43–77. Springer‐Verlag, New York.
SewertzoV, A. N. (1902). Zur Entwicklungsgeschichte des Ceratodus forsteri. Anat. Anz. 21,
Shand, J., and Foster, R. G. (1999). The extraretinal photoreceptors of non‐mammalian verte-
    brates. In ‘‘Adaptive Mechanisms in the Ecology of Vision’’ (Archer, S. N.,
    Djamgoz, M. B. A., Loew, E., Partridge, J. C., and Vallerga, S., Eds.), pp. 197–222. Kluwer
    Academic Publishers, Great Britain.
Siebeck, U. E., and Marshall, N. J. (2001). Ocular media transmission of coral reef fish—can
    coral reef fish see ultraviolet? Vision Res. 41, 133–149.
3.   NERVOUS AND SENSORY SYSTEMS                                                              177

Siebeck, U. E., Collin, S. P., Ghoddussi, M., and Marshall, N. J. (2003). Occlusable corneas in
      toadfishes: Light transmission, movement and ultrastructure of pigment during light and
      dark adaptation. J. Exp. Biol. 206, 2177–2190.
Siefkes, M. J., and Li, W. (2004). Electrophysiological evidence for detection and discrimination
      of pheromonal bile acids by the olfactory epithelium of female sea lampreys (Petromyzon
      marinus). J. Comp. Physiol. A 190, 193–199.
Sillman, A. J., Sorsky, M. E., and Loew, E. R. (1995). The visual pigments of wild white sturgeon
      (Acipenser transmontanus). Can. J. Zool. 73, 805–809.
Sillman, A. J., O’Leary, C. J., Tarantino, C. D., and Loew, E. R. (1999). The photoreceptors and
      visual pigments of two species of Acipenseriformes, the shovelnose sturgeon (Scaphirhynchus
      platorynchus) and the paddlefish (Polyodon spathula). J. Comp. Physiol. A 184, 37–47.
Sivak, J. G., and Woo, G. C. S. (1975). Accommodative lens movement in holosteans (Amia
      calva and Lepisosteus osseus oxyurus) and in the sea lamprey (Petromyzon marinus). Can.
      J. Zool. 53, 516–520.
Song, J., and Northcutt, R. G. (1991a). The primary projections of the lateral line nerves of the
      Florida gar, Lepisosteus platyrhincus. Brain Behav. Evol. 37, 38–63.
Song, J., and Northcutt, R. G. (1991b). Morphology, distribution and innervation of the lateral
      line receptors of the Florida gar, Lepisosteus platyrhincus. Brain Behav. Evol. 37, 10–37.
Sorensen, P. W., Hara, T. J., and Stacy, N. E. (1987). Peripheral olfactory responses of mature
      male, regressed, and hypophysectomized goldfish to a steroidal sex pheromone, L‐serine,
      and taurocholic acid. Ann. N. Y. Acad. Sci. 510, 635–637.
Striedter, G. F. (2005). ‘‘Principles of Brain Evolution.’’ Sinauer Associates Inc, Massuchesetts,
Striedter, G. F., and Northcutt, R. G. (2006). Head size constrains forebrain development and
      evolution in ray‐finned fishes. Evol. Dev. 8, 215–222.
Suzuki, N. (1977). Intracellular responses of lamprey olfactory receptors to current and chemical
      stimulation. In ‘‘Food Intake and Chemical Senses’’ (Katsuki, Y., Sato, M., Takagi, S. F.,
      and Oomura, Y., Eds.), pp. 13–22. University of Tokyo Press, Tokyo.
Suzuki, N. (1978). EVects of diVerent ionic environments on the responses of single
      olfactory receptors in the lamprey. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 61,
Teeter, J. (1980). Pheromone communication in sea lampreys (Petromyzon marinus): Implica-
      tions for population management. Can. J. Zool. 37, 2123–2132.
Theisen, B. (1976). The olfactory system in the pacific hagfish Eptatretus stouti, Eptatretus deani,
      and Myxine circifrons. Acta Zool. 57, 167–173.
Thornhill, R. A. (1967). Ultrastructure of olfactory epithelium of lamprey hampetra fluriatilis. J.
      Cell. Science 2, 591–602.
Tricas, T. C. (2001). The neuroecology of the elasmobranch electrosensory world: Why periph-
      eral morphology shapes behavior. Environ. Biol. Fishes 60, 77–92.
Tricas, T. C., and New, J. G. (1998). Sensitivity and response dynamics of electrosensory primary
      aVerent neurons to near threshold fields in the round stingray. J. Comp. Physiol. 182, 89–101.
Tricas, T. C., Michael, S. W., and Sisneros, J. A. (1995). Electrosensory optimization to
      conspecific phasic signals for mating. Neurosci. Lett. 202, 129–338.
Ueck, M. (1969). Ultrastrukturbesonderheiten der pinealen Sinneszellen von. Protopterus dolloi.
      Z. Zellforsch. 100, 560–580.
Ullen, F., Orlovsky, G. N., Degliagina, T. G., and Grillner, S. (1993). Role of dermal photo-
      receptors and lateral eyes in initiation and orientation of locomotion in lamprey. Behav.
      Brain Res. 54, 107–110.
Ullen, F., Degliagina, T. G., Orlovsky, G. N., and Grillner, S. (1995a). Spatial orientation of
      lamprey. 1. Control of pitch and roll. J. Exp. Biol. 198, 665–673.
178                                                                         SHAUN P. COLLIN

Ullen, F., Degliagina, T. G., Orlovsky, G. N., and Grillner, S. (1995b). Spatial orientation of
      lamprey. 2. Visual influence on orientation during locomotion and in the attached state.
      J. Exp. Biol. 198, 675–681.
van Horn, D. L., Edelhauser, H. F., and Schultz, R. O. (1969). A comparative study of stromal
      swelling in sea lamprey spectacle and brook trout cornea. J. Ultrastruct. Res. 28, 452–461.
van Netten, S. M., and Kroese, A. B. A. (1989). Dynamic behavior and micromechanical
      properties of the cupula. In ‘‘The Mechanosensory Lateral Line’’ (Coombs, S., Gorner, P.,
      and Munz, H., Eds.), pp. 247–263. Springer, New York.
Vazquez, M., Rodriguez, F., Domezain, A., and Salas, C. (2002). Development of the brain of
      the sturgeon Acipenser nccarii. J. Appl. Ichthyol. 18, 275–279.
Vesselkin, N. P., Reperant, J., Keningfest, N. B., Rio, J. P., Micali, D., and Shupliakov, O. V.
      (1989). Centrifugal innervation of the lamprey retina: Light microscopic and electron
      microscopic and electrophysiological investigations. Brain Res. 493, 51–65.
Vesselkin, N. P., Rio, J. P., Reperant, J., Kenigfest, N. B., and Adanina, V. O. (1996). Retino-
      petal projections in lampreys. Brain Behav. Evol. 48, 277–286.
Vigh, B., Manzano, M. J., Zadori, A., Frank, C. L., Lukats, A., Rohlich, P., Szel, A., and
      David, C. (2002). Non‐visual photoreceptors of the deep brain, pineal organs and retina.
      Histol. Histopathol. 17, 555–590.
Vigh‐Teichmann, I., Vigh, B., Olsson, R., and van Veen, T. (1984). Opsin‐immunoreactive outer
      segments of photoreceptors in the eye and in the lumen of the optic nerve of the hagfish,
      Myxine glutinosa. Cell Tissue Res. 238, 515–522.
Vigh‐Teichmann, I., Szel, A., Rohlich, P., and Vigh, B. (1990). A comparison of the ultrastruc-
      ture and opsin immunocytochemistry of the pineal organ and retina of the deep‐sea fish
      Chimaera monstrosa. Exp. Biol. 48, 361–371.
Vollrath, L. (1981). ‘‘The Pineal Organ.’’ Springer‐Verlag, New York.
von Bartheld, C. S. (2004). The terminal nerve and its relation with extrabulbar ‘‘olfactory’’
      projections: Lessons from lampreys and lungfishes. Microsc. Res. Tech. 65, 13–24.
von Bartheld, C. S., and Meyer, D. L. (1986). Central connections of the olfactory bulb in the
      bichir, Polypterus palmas, re‐examined. Cell Tissue Res. 244, 527–535.
von Bartheld, C. S., Claas, B., Munz, H., and Meyer, D. L. (1988). Primary olfactory projections
      and the nervus terminalis in the African lungfish: Implications for the phylogeny of cranial
      nerves. Am. J. Anat. 182, 325–334.
von Bartheld, C. S., Collin, S. P., and Meyer, D. L. (1990). Dorsomedial telencephalon of
      lungfishes: A pallial or subpallial structure? Criteria based on histology, connectivity, and
      histochemistry. J. Comp. Neurol. 294, 14–29.
von During, M., and Miller, M. R. (1978). Sensory nerve endings of the skin and deeper
      structures. In ‘‘Biology of the Reptilia, Vol. 1, Neurology’’ (Gans, C., Ed.), pp. 407–441.
      Academic Press, London.
von During, M., and Seiler, W. (1974). The fine structure of lamellated receptors in the skin of
      Rana esculenta. Z Anat Entwicklungsgesch 144, 165–172.
Vorobyev, M. (2003). Coloured oil droplets enhance colour discrimination. Proc. R. Soc. Lond.
      B. 270, 1255–1261.
Walls, G. L. (1942). ‘‘The Vertebrate Eye and its Adaptive Radiation.’’ Cranbrook Press,
      Bloomfield Hills.
Waltman, B. (1966). Electrical properties and fine structure of the ampullary canals of Lorenzini.
      Acta Physiol. Scand. 66(Suppl. 264), 1–60.
Watt, M., Evans, C. S., and Joss, J,M. P. (1999). Use of electroreception during foraging by the
      Australian lungfish. Anim. Behav. 58, 1039–1045.
Webb, J. F., and Northcutt, R. G. (1988). Bony fish pit line neuromasts are homologous to canal
      neuromasts. Soc. Neurosci. Abstr. 14, 54.
3.   NERVOUS AND SENSORY SYSTEMS                                                              179

Webb, J. F., and Northcutt, R. G. (1997). Morphology and distribution of pit organs and canal
    neuromasts in non‐teleost bony fishes. Brain Behav. Evol. 50, 139–151.
Weth, F., Nadler, W., and Korsching, S. (1996). Nested expression domains for odorant
    receptors in zebrafish olfactory epithelium. Proc. Natl. Acad. Sci. USA 93, 13321–13326.
Wever, E. G. (1974). The evolution of vertebrate hearing. In ‘‘Handbook of Sensory Physiology,
    Vol. 1, Auditory System’’ (Keidel, W. D., and NeV, W. D., Eds.), pp. 423–454. Springer, Berlin.
Whitear, M. (1992). Solitary chemosensory cells. In ‘‘Fish Chemoreception’’ (Hara, T. J., Ed.),
    pp. 103–125. Chapman and Hall, London.
Whitear, M., and Lane, E. B. (1983). Multivillous cells: Epidermal sensory cells of unknown
    function in lamprey, Entosphenous japonicus. J. Ultrastruct. Res. 43, 1–17.
Whitmore, A. V., and Bowmaker, J. K. (1989). Seasonal variation in cone sensitivity and
    shortwave absorbing visual pigments in the rudd, Scardinius erythophthalmus. J. Comp.
    Physiol. A 166, 103–115.
Wicht, H. (1996). The brains of lampreys and hagfishes: Characteristics, characters, and com-
    parisons. Brain Behav. Evol. 48, 248–261.
Wicht, H., and Niewenhuys, R. (1998). Hagfishes, Myxinoidea. In ‘‘The Central Nervous System
    of Vertebrates’’ (Niewenhuys, H., Ten Donkelaar, J., and Nicholson, C., Eds.), Vol. 1,
    pp. 497–550. Springer‐Verlag, Berlin.
Wicht, H., and Northcutt, R. G. (1990). Retinofugal and retinopetal projections in the Pacific
    hagfish, Eptatretus stouti. Brain Behav. Evol. 36, 315–328.
Wicht, H., and Northcutt, R. G. (1992). The forebrain of the pacific hagfish: A cladistic
    reconstruction of the ancestral craniate forebrain. Brain Behav. Evol. 40, 25–64.
Wicht, H., and Northcutt, R. G. (1994). An immunohistochemical study of the telecephalon and
    the diencephalon in a myxinoid jawless fish, the Pacific hagfish Eptatretus stouti. Brain
    Behav. Evol. 43, 140–161.
Wicht, H., and Northcutt, R. G. (1995). Ontogeny of the head of the Pacific hagfish (Eptatretus
    stouti, Myxinoidea): Development of the lateral line system. Philos. Trans. R. Soc. Lond. B.
    349, 119–134.
Wilkens, L. A., Russell, D. F., Pei, X., and Gurgens, C. (1997). The paddlefish rostrum functions
    as an electrosensory antenna in plankton feeding. Proc. R. Soc. Lond. B 264, 1723–1729.
Wilkens, L. A., Wettring, B., Wagner, E., Wojtenek, W., and Russell, D. (2001). Prey detection
    in selective plankton feeding by the paddlefish: Is the electric sense sufficient? J. Exp. Biol.
    204, 1381–1389.
Yokoyama, S. (2000). Color vision of the coelacanth (Latimeria chalumnae) and adaptive
    evolution of rhodopsin (RH1) and rhodopsin‐like (RH2) pigments. J. Hered. 91, 215–220.
Young, J. Z. (1935). The photoreceptors of lampreys. I. Light sensitive fibres in the lateral line
    nerves. J. Exp. Biol. 12, 229–238.
Zakon, H. H. (1988). The electroreceptors: Diversity in structure and function. In ‘‘Sensory
    Biology of Aquatic Animals’’ (Atema, J., Fay, R. R., Popper, A. N., and Tavolga, W. N.,
    Eds.), pp. 813–850.
Zalc, B., and Colman, D. R. (2000). Origins of vertebrate success. Science 288, 271–272.
Zeiske, E., Theissen, B., and Breucker, H. (1992). Structure, development, and evolutionary
    aspects of the peripheral olfactory system. In ‘‘Fish Cemoreception’’ (Hara, T. J., Ed.),
    pp. 13–39. Chapman and Hall, London.
This page intentionally left blank


1. Introduction
2. Respiratory Strategies
3. Respiratory Organs
   3.1. Water Breathing
   3.2. Air Breathing
4. Ventilatory Mechanisms
   4.1. Cutaneous Gas Exchange
   4.2. Ventilation of External Gills
   4.3. Ventilation of Internal Gills
   4.4. Ventilation of ABOs
5. Respiratory Control
   5.1. Hypoxic and Hypercarbic Ventilatory Responses and Reflex Pathways
   5.2. Receptors Involved in Reflex Ventilatory Control
6. Conclusions

    Primitive fishes are widespread in geographical distribution, and the
diverse nature of their aquatic environments has given rise to a tremendous
adaptive radiation in respiratory physiologies. The group of primitive fishes
comprises water‐breathing, bimodal breathing, and obligate air‐breathing
species, which possess a variety of respiratory strategies, respiratory organs,
pumping mechanisms, and control systems that integrate multiple exchange
sites and receptors into their overall ventilatory response. Many studies have
described the reflex responses of this group to environmental perturbations
such as hypoxia and hypercarbia, but one area that clearly needs more study
is our understanding of the mechanisms by which peripheral receptors and
neural pathways are centrally integrated to produce the complex ventilatory
behaviors seen in these fishes. Although these species are highly evolved
and adapted to their particular ecological niches, and represent significant

Primitive Fishes: Volume 26                        Copyright # 2007 Elsevier Inc. All rights reserved
FISH PHYSIOLOGY                                               DOI: 10.1016/S1546-5098(07)26004-2
182                                                    EMILY COOLIDGE ET AL.

departures in morphology, ecology, and behavior from the stem groups that
gave rise to them, further examination of the mechanisms underlying their
respiratory behavior may lead to greater insight into the evolutionary forces
that have shaped the transition from living in water to living on land among


     The first chapter in this volume has clearly defined those groups of living
fishes that are considered ‘‘primitive fishes.’’ In this chapter, we will examine
what is known of the respiratory systems of these species. While there is
an extensive literature describing respiratory processes in fish, much of this
literature comes from studies of elasmobranchs and teleosts. These are con-
sidered ‘‘modern fishes’’ and previous volumes in this series have focused
exclusively on these groups (Volumes 5, 10, 12, and 17; Randall et al., 1981a;
Graham, 1997; Maina, 2003). The challenge of this chapter is to summarize
data specific to the primitive fishes and build on previous reviews on this topic
(Randall et al., 1981a, Burggren et al., 1985; Shelton et al., 1986). Several
things are of note in this regard.
     Ventilatory systems regulate the diVusion of respiratory gases in (O2) or
out (CO2 and NH4) of the body, and hence are strictly linked to the regulation
of all metabolic processes. They are also highly modified as a function of
the environments within which these fish live. The various groups of primitive
fishes have a widespread geographical range of distribution, and the diverse
nature of their aquatic environments has given rise to a tremendous adap-
tive radiation in respiratory strategies. Thus, the group of primitive fish
defined in this volume consists of water‐breathing, bimodal breathing, and
obligate air‐breathing fish, possessing a variety of respiratory strategies,
respiratory organs, pumping mechanisms, and control systems. Furthermore,
all extant species of fish (primitive fish included) are highly derived and
represent significant departures in morphology, ecology, and behavior from
the stem groups that gave rise to them. Given this background, it is extremely
difficult to distinguish ‘‘primitive’’ from ‘‘derived’’ physiological character-
istics, particularly given how plastic physiological processes can be at the
organismal level. In summarizing studies of these processes, trends will be
shown that are suggestive of evolutionary progression. All attempts to de-
scribe evolutionary trends based on the physiology of present‐day fishes,
however, must be viewed within this context.
4.   VENTILATORY SYSTEMS                                                      183


    Because environmental hypoxia is such a pervasive problem for aquatic
organisms, the ventilatory adaptations to hypoxia have been investigated
in fishes for many years. As with all fish, primitive fish have developed
strategies (behavioral, morphological, anatomical, physiological, and bio-
chemical) either to avoid low O2 conditions, increase O2 transfer from the
environment to the tissues, reduce O2 demands, or some combination of
these. Many species constantly sense and monitor environmental conditions
and migrate to better areas (Junk et al., 1983; Wootton, 1990). Other species
increase O2 extraction and/or reduce O2 demands through a host of physio-
logical and biochemical adjustments including regulation of diVerent Hb
fractions, adjustment of intraerythrocytic levels of organophosphates,
changes in hematocrit and Hb concentration, and metabolic suppression—
almost all under catecholaminergic control. These are described in detail in
Chapter 5 of this volume (Milligan and Wood, 1987; Perry and Kinkead,
1989; Nikinmaa, 1990; Randall, 1990; Val et al., 1992; Almeida‐Val and Val,
1993). They also include increased O2 uptake via the respiratory system.
    In general terms, nearly all aquatic organisms exhibit increases in ventila-
tion either as increases in ventilatory frequency, tidal volume, or both, in order
to improve O2 extraction at the gills. A surprisingly large number of the
‘‘primitive fishes’’ are also air‐breathing fishes. Indeed, of the bony fishes,
only two Chondrosteans, the sturgeon and the paddlefish, have retained total
dependency on aquatic gas exchange. For some species, air breathing is a
facultative event that occurs only when water O2 levels are low. These species
tend to have functional gills that are used in conjunction with air breathing
(e.g., gar, bowfin). For other species, air breathing is an obligatory behavior
and these species rely primarily, if not exclusively, on O2 taken from the air.
These species tend to have greatly reduced gill structure (e.g., lungfish). There
are many species that utilize both strategies either as a function of develop-
mental age or environmental conditions. Thus, the gar, Lepisosteus, is a facult-
ative air‐breather at low temperatures but becomes an obligate air‐breather
when O2 uptake increases at higher temperatures (Rahn et al., 1971). Obligate
air‐breathing fish, such as Protopterus, do not respond to aquatic hypoxia
but, instead, respond to aerial hypoxia. In facultative air‐breathers such as
Lepisosteus or Amia, there is a reduction in gill ventilation and a behavioral
switch to air breathing as the major source of O2 acquisition under hypoxic
conditions. These are generalizations, however, and more specific examples of
the ventilatory response to hypoxia are detailed below.
184                                                                  EMILY COOLIDGE ET AL.

    Environmental hypercarbia, although a prevalent and important ventilatory
stimulus in freshwater, has not been studied to the same extent as environ-
mental hypoxia. Because increased ventilation cannot ameliorate the eVects
of environmental hypercarbia, alternative respiratory strategies must be
employed when fish encounter such a stimulus. Moreover, owing to the
Root eVect in fish Hb, environmental hypercarbia may produce secondary
eVects on ventilation through reductions in O2 content.


3.1. Water Breathing
3.1.1. Gills
   From the Agnatha to the elasmobranchs to the bony fish, there is a shift
from pouched gills to septal gills to opercular gills (Figure 4.1). The funda-
mental structure of the gill remains the same but the eYciency with which the
actual gas exchange surface, the secondary lamellae, is exposed to the water
flowing through the gills increases progressively (see below). While most of
these gills are internal and situated within the respiratory cavity, external gills

                       A                  B                  C

                     Juvenile    Adult
                           Lamprey              Shark            Teleost fish

Fig. 4.1. Schematic diagrams of (above) the arrangement and coverings of the pharyngeal slits
and (below) the individual pharyngeal arches of a (A) lamprey (ammocoetes larva on left and
adult on right), (B) shark, and (C) teleost fish. Black arrows indicate the direction of water flow.
[Modified from Kardong (2002).]
4.   VENTILATORY SYSTEMS                                                     185

are common in the larval stages of primitive jawed fishes and are also found
in the adults of some Polypterids and Dipnoans (Burggren et al., 1985).
     The diVusing capacity of any species can be increased by alterations in the
number of gill arches, the length and number of gill filaments on each arch,
the spacing of the lamellae along the filament, the surface areas of individual
gill lamellae, the thickness of the water–blood interface, and the resistance
to water flow through the gill sieve (Hughes, 1984). Changes in any or all of
these variables occur as a function of lifestyle and habitat throughout all
taxonomic groups of fishes.
     The diVusing capacity of any individual fish can also be changed in a
number of ways. These include increasing the number of lamellae perfused
at any one time (and hence the functional area available for gas transfer)
(Booth, 1978), redirecting blood through sections of lamellae exposed to gill
water flow, and reducing lymphatic space (Randall et al., 1981a). All result in
a reduction in diVusion distance between blood and water and an increase
in the surface area across which gas exchange occurs.
     These are the general trends. Specific details are available for many
species. Thus, as mentioned above, the gills of agnathans are located
in saclike pouches. Lampreys have seven pairs of gill pouches while hag-
fish have from 5 to 14 pairs, depending on the species. The anterior and
posterior walls of each pouch bear a hemibranch on which the gill filaments
are diVerentiated into secondary lamellae. The anatomy of the secondary
lamellae, their epithelial thickness, and their total surface area in lampreys
are similar to those in teleost fish (Hughes and Morgan, 1973; Lewis, 1976;
Lewis and Potter, 1976; Hughes, 1984).
     In elasmobranchs each gill possesses a central partition, the interbran-
chial septum, which is covered on both sides by primary lamellae (or gill
filaments). In the bony fishes, the interbranchial septum is reduced or absent
and the gill filaments often arise directly from the base of the branchial
arch. This frees the individual filaments in such a way that water flow over
the filaments is enhanced (Figure 4.1). In the Chondrosteans (sturgeon
and paddlefish), the gill septum is reduced and only about 50% of the gill
filaments attach to the septum. The general organization of the secondary
lamellae is similar to that of other bony fishes although the total surface area
is greatly reduced (Byczkowska‐Smyk, 1962; Hughes, 1984).
     In Neopterygii (bowfin and gar), the interbranchial septum is well devel-
oped proximally but the distal one‐third of the gill filaments are unsupported
(Randall et al., 1981b; Smatresk and Cameron, 1982). While the gills are
well developed in general, in Amia, the first gill arch bears no hemibranch,
whereas in Lepisosteus, it possesses only a posterior hemibranch. Despite
this, the total surface area of the gills of Amia is similar to that of a teleost
with similar activity levels (Daxboeck et al., 1981). Furthermore, in Amia,
the secondary lamellae are fused into a lattice of rectangular pores, which
186                                                                   EMILY COOLIDGE ET AL.

is believed to prevent their collapse in air (Daxboeck et al., 1981). The
reduced first gill arch and strengthened secondary lamellae are believed to
be adaptations associated with air breathing.
    In the Dipnoi, there are considerable diVerences in gill structure between
the three extant species. The Australian lungfish, Neoceratodus forsteri, is
an obligate water‐breather with well‐developed gills. The African lungfish,
Protopterus sp., and the South American lungfish, Lepidosiren paradoxa, on
the other hand are all obligate air‐breathers with diVerent degrees of reduc-
tion in their gill development and surface area (Johansen and Lenfant, 1967;
Brauner et al., 2004) (Figure 4.2).
    For species possessing well‐developed air‐breathing organs (ABOs), there
are conflicting functional requirements placed on the design of their gills.
This arises because the more O2‐rich blood draining the ABO returns to the
heart and must then pass through the gills before entering the systemic
circulation. In the process, the potential exists for significant loss of O2 to
the hypoxic water during transit through the gills (Randall et al., 1981a). As a
result, many of these fish exhibit reductions in functional gill surface area.
This may be in the form of reductions in the number of gill arches, the
number of secondary lamellae, secondary lamellar thickening, or presence
of gill vascular shunts. The extent to which any or all of these occur is
generally a function of the dependence of the species on air breathing.
3.1.2. Skin
    Although the gills are the primary gas exchange organs in water, the
skin also plays a significant role in many species of fish. Agnathans lack
dermal bone and the skin is smooth and without scales. At least 80% of O2
uptake in the hagfish, Myxine glutinosa, can occur across the skin (SteVensen
et al., 1984) while the respiratory role of the skin of adult lampreys has

  A                                            B

Fig. 4.2. Scanning electron micrographs (SEM) of the gills from (A) a typical bony fish and (B) the
obligate air‐breathing teleost Arapaima gigas. Scale bar, 500 mm. [Figure courtesy of Colin Brauner.]
4.   VENTILATORY SYSTEMS                                                        187

been estimated to be only 20% of total O2 uptake (Korolewa, 1964). The
remaining primitive fishes, including both the actinopterygians and sarcop-
terygians, are bony fishes. While bone is pervasive throughout the endoskel-
eton in this class, the trend is reversed in groups such as the sturgeons,
paddlefishes, and some lungfishes in which the endoskeleton is primarily
cartilaginous. Despite this, all have scales but cutaneous gas exchange is
still very significant in many, particularly in their larval stages. In most air‐
breathing fish, while other specialized exchange surfaces become the major
site of O2 uptake from air, the gills and/or skin remain one of the major sites
of CO2 excretion (into water). In Amia, the gas exchange ratio (CO2 elimina-
tion/O2 uptake) is 0.21 for the ABO and 1.61 for the gills (Randall et al.,
1981b). CO2 excretion is exclusively aquatic in Neoceratodus, whereas in
Protopterus and Lepidosiren air breathing can account for up to 30% of
CO2 elimination (Johansen et al., 1970).

3.2. Air Breathing
3.2.1. Lungs and Respiratory Gas Bladders
    An accessory or alternative strategy employed by a large number of
phyletically primitive fishes is air breathing. The diversity of sites and sur-
faces that are utilized for gas transfer from air to blood in fish is remarkable.
While a few species do utilize their gills for gas exchange in air, this is a rare
occurrence and most air‐breathing fish utilize other surfaces. Graham (1997)
has put forward a simplified classification scheme for structures utilized
by fish for aerial gas exchange (ABO). He suggests that ‘‘even though air
breathing has evolved numerous times and independently, the location of aerial
exchange sites has remained largely under the conservative influence
of structures ‘‘predisposed’’ for air gulping and sites in the body where gas
storage and the requisite vascularization could be developed.’’ This scheme
divides structures into three groups: (1) those associated with the skin; (2) struc-
tures associated with organs in the head region (buccal and opercular cavities,
pharynx) or along the digestive tract; and (3) the lungs and respiratory gas
bladders (Carter and Beadle, 1931; Johansen, 1970; Randall et al., 1981a).
    In the primitive fishes, however, structures associated with air breathing
fall into only the last category (lungs or respiratory gas bladders). At least
47 species from 24 genera of bony fish are known to breathe air using a lung
or a respiratory gas bladder. By the scheme put forward by Graham (1997),
gas bladders have an embryonic origin from the side or dorsal aspect of the
alimentary canal, are not paired, do not always have a glottis (and may or
may not retain an open pneumatic duct), and in most cases receive blood in
parallel with the systemic circulation and lack a specialized pulmonary
circulation. Lungs, on the other hand, have an embryonic origin from the
ventral wall of the alimentary canal, are paired, possess a valvular glottis in
188                                                                           EMILY COOLIDGE ET AL.

the floor of the alimentary canal, and have a proper pulmonary circulation
in which eVerent vessels return blood directly to the heart (not the vena cava)
(Figure 4.3). By this scheme, lungs are possessed only by the lungfishes
(Neoceratodus, Lepidosiren, and Protopterus) and the polypterids (Polypterus
and Erpetoichthys). Gas bladders are found in both Amia and the garfishes
and are scattered throughout the teleosts. Among these fishes, respiratory gas
bladders diVer greatly in complexity (Jarvik, 1980; Graham, 1997). Note that
this scheme implies that both lungs and gas bladders must have evolved
independently multiple times.
    The relative roles of the various gas exchange surfaces vary tremendously
between species. In the reedfish, Calamoichthys calabaricus, when in oxyge-
nated water, total O2 uptake is 28% from the gills, 32% across the skin, and
40% across the lungs (Sacca and Burggren, 1982). In Amia and Lepisosteus,
while the gills are the major site of gas exchange at low temperatures (10  C),
at warmer temperatures (20–25  C), the ABO accounts for 40–75% of O2
uptake (Johansen et al., 1970; Rahn et al., 1971; Randall et al., 1981a;
Smatresk and Cameron, 1982b). The relative contributions of aquatic and
aerial O2 uptake in the three genera of lungfish range from nearly 100% from

 A                                                       B
                                   Swim bladder artery                                      Pulmonary artery
                Dorsal aorta                                         Dorsal aorta

 Swim bladder                                                 Lung
                                             Gill                                                  Gill

 Postcava                                                Postcava
  Swim bladder vein                                          Pulmonary vein
                               Heart                                                Heart

 C                                                       D

Fig. 4.3. Schematic diagrams illustrating the generalized circulation to (A) the ABO of a typical
bony fish and (B) a lungfish (see text for details). Panels (C) and (D) illustrate the relation of the
ABO and lung to the oesophagus as seen from the side and in cross section. [Modified from
Kardong (2002).]
4.   VENTILATORY SYSTEMS                                                    189

water in Neoceratodus to 80–90% from air in Protopterus and Lepidosiren
(Johansen et al., 1970).
3.2.2. Skin
    Many fish that spend time out of water (amphibious fish) do use their skin
for aerial gas transfer and, although subject to uncontrolled water loss and
limited as an organ for O2 uptake, the surface is adequate for CO2 excretion
(Graham, 1997).


    No matter what the exchange organ, water or air must move actively
across the respiratory surfaces to increase the rate of diVusion. Invariably,
this requires muscular action.

4.1. Cutaneous Gas Exchange

    While cutaneous gas exchange in aquatic organisms is often viewed as
purely a passive phenomenon, active movement of water over the skin surface
is common (Feder and Burggren, 1985). This may involve general body move-
ments or more specialized movements. Larvae of the lungfish, Neoceratodus,
possess cilia which develop respiratory currents across the general body surface
and inside the opercula (Whiting and Bone, 1980). Larvae of Monopterus use
movements of the pectoral fins to produce water currents that run over the
body surface counter to the direction of skin blood flow (Liem, 1982).

4.2. Ventilation of External Gills

    While external gills are rare in general, they are common in the larval
stages of primitive jawed fishes (Burggren et al., 1985). The external gills
of larval Protopterus are ciliated and produce convective currents prior
to the development of a muscular apparatus for active movement of the
gills. Intrinsic musculature does develop and the external gills of lungfishes
are capable of sweeping movements that irrigate the gills and break up
stagnant boundary layers surrounding the gills (Greenwood, 1958).

4.3. Ventilation of Internal Gills
4.3.1. Phylogenetic Perspectives
    One of the characteristics that define all chordates is the possession of
pharyngeal slits, at least at some point during their development (Cameron
et al., 2000). It is believed that these pharyngeal slits likely first evolved
190                                                       EMILY COOLIDGE ET AL.

from the corners of the mouth to aid in suspension feeding in primitive
chordates and protochordates (Gutmann, 1981). This allowed a one‐way
flow of water—in at the mouth and out through the pharyngeal slits.
Initially, this form of suspension feeding depended solely on ciliary pumps
to create the flow of water (Gilmour, 1979). As the primitive chordates
enlarged, the flanks of the body weakened favoring the evolution of support-
ing structures between successive slits. These ultimately gave rise to cartila-
ginous pharyngeal arches. Initially, the walls of the slits were associated
with mucus‐bearing cilia that served to trap suspended particles; respir-
ation was primarily cutaneous. Only secondarily did the walls defining the
slits become associated with gills and begin to participate in respiratory
gas exchange. At this point, water entering the mouth could bring sus-
pended food and O2 to the animal. This increase in feeding eYciency gave
rise to more active lifestyles. With this came the evolution of a muscular
buccal pump that helped to produce the food bearing current. This allowed
animals to attain larger mass and led to the loss of ciliary mechanisms
for moving water (Sanderson and Wassersug, 1990) (Figure 4.1A). Transi-
tional species probably became raptorial feeders, plucking individual parti-
cles selectively from suspensions or oV surfaces. The supporting structures
of the first pharyngeal slit moved forward and evolved into jaws, further
increasing feeding eYciency and giving rise to the origins of active predation
and a shift away from sessile suspension feeding (Mallatt, 1996). With
this, the pharyngeal slits were no longer required for feeding but the active
lifestyle demanded greater gas exchange than was provided by cutaneous
exchange alone. With removal of the constraints placed on the pharyn-
geal slits for feeding, true gills evolved, and the pharyngeal slits and buccal
pump that originally evolved for feeding gave rise to gills for breathing
with water flow being driven by a buccal pump involving muscles primarily
innervated by the trigeminal and facial nerves. Muscles in the walls of
the pharyngeal arches, innervated by the glossopharyngeal and vagus nerves
acted as accessory muscles to stabilize and maintain the gill curtain (Mallatt,
1996; Kardong, 2002). This is the situation found today in the cartila-
ginous and bony fishes, and to some extent in the agnathans (hagfish and

4.3.2. Agnatha
    Ventilation in Agnatha is powered by muscular velar folds and by
compression and expansion of the branchial apparatus. In the larvae of the
lamprey as well as in adult hagfish, the primary respiratory pump is
the velum. This is a muscular structure attached to the roof of the pharynx
in the midline. It consists of two leaves that are tightly furled at rest. Scrolling
and unscrolling of the velum on each side of the buccal cavity, together with
4.   VENTILATORY SYSTEMS                                                                         191

synchronized contraction and relaxation of the branchial pouches, produce a
flow of water in at the nostril (singular for hagfish) and out across the gills
via the branchial pouches. This is an amazingly eVective mechanism that can
generate water flows that are not dissimilar from those produced by many
teleost fishes (Figure 4.4) (Shelton, 1970; Rovainen and Schieber, 1975).

     A                       Vel.            Ph.






Fig. 4.4. (A) Schematic sagittal section through the anterior trunk of a hagfish (from Johansen
and Strahan, 1963). (B) Schematic diagram of the velar chamber with the left pharyngeal wall
removed. 1, resting; 2, velar scroll beginning to unroll; 3, velar scroll unrolled to the full extent;
and 4, velar scroll beginning to roll again. Arrow shows direction of water flow. [From Johansen
and Strahan (1963).] (C) Schematic of a lateral view of the velum scrolling and unscrolling to
move water through the pharynx. [From Kardong (2002).] Bucc., buccal cavity; Ph., pharynx;
Vel., velum and Vel. ch., velar chamber.
192                                                     EMILY COOLIDGE ET AL.

    In adult lampreys, the velum is reduced and ventilation is primarily
generated by compression and expansion of the branchial apparatus alone.
Closure of the velum and active compression of the branchial apparatus
drive water out through the branchial pouches. Relaxation of these muscles
allows the elastic branchial apparatus to recoil into its expanded position
passively drawing water back in. Although unidirectional water flow (in at
the mouth and out over the gills) is possible, in most adult lampreys the
pattern of water flow is tidal with water flowing out and back in through
the openings of the branchial pouches. A partition dividing the pharynx
thus allows the lamprey to attach to its prey by suction and use its tongue to
scrape flesh into the esophagus while continuing to ventilate tidally through
the branchial pouches (Johansen and Strahan, 1963; Jensen, 1966; Randall,

4.3.3. Bony Fishes
    In most fishes, the buccal and opercular cavities form dual pumps on
either side of the gill curtain. Both cavities are expanded simultaneously by
muscular action creating a suction that closes the operculae and draws water
in through the mouth. Both cavities are then compressed by muscular action
while the mouth closes, forcing water over the gill curtain and out through
the operculae. Because of a slight diVerence in pressure between buccal and
opercular cavities, water flows almost continuously across the gills in one
direction (Figure 4.5A) (Hughes, 1984). This basic mechanism powers gill
ventilation in all primitive and modern jawed fishes (McMahon, 1969;
Shelton, 1970; Hughes, 1984).

4.3.4. Acipenseriformes (Sturgeon) and Polypteriformes (Birchirs)
     Many species of these two groups are bottom feeders eating buried
invertebrates and carrion. Their mouths are modified for sucking mud.
In most, the first gill arch is reduced to a spiracle. When these fish are in
open water they ventilate the gills just as other fish do. In sturgeon (Acipenser
transmontanus), when the mouth is buried, the fish ventilate only via the
opercular opening (the spiracle plays little role) but in a unique way that
retains a unidirectional flow of water over the gills. During buccal and
opercular expansion, water enters the opercular cavity through a permanent
aperture in the dorsal margin of the operculum. This water then continues
into the buccal cavity flowing dorsally over the gill sieve, not passing over the
gill lamellae. During buccal compression, the water then passes over the gill
sieve in a normal fashion and exits via the operculum (Figure 4.5B)
(Burggren, 1978). It has been postulated that a similar pattern is produced
using the spiracle in Polypterus (Magid, 1966).
4.   VENTILATORY SYSTEMS                                                                193

                     Dual pump
                    Buccal     Gill
                    cavity     curtain     Opercular

                   valve                      Opercular

                                                            Force phase
                           Suction phase


          Buccal               Dorsal opercular   Buccal             Dorsal opercular
          cavity    Spiracle   channel            cavity Spiracle    channel

Fig. 4.5. Schematic diagrams illustrating the dual pump found in most bony fishes (A), and the
modified pumping mechanism found in sturgeon (B). See text for details on all pumping
mechanisms. [From Burggren (1978) and Kardong (2002).]

4.4. Ventilation of ABOs
4.4.1. Sarcopterygian Fishes
    In air‐breathing fish, the buccal pump usually exclusively produces venti-
lation. In sarcopterygian fishes, an initial buccal expansion phase draws both
air from the ABO and fresh air from the environment into the buccal cavity
194                                                                EMILY COOLIDGE ET AL.

              A            Exhalation       Water        Inhalation
                             Compression (2)      Expansion (3)
                     Expansion (1)                         Compression (4)


                        Expansion                      Compression

Fig. 4.6. Schematic diagrams illustrating the two‐stroke buccal pump found in sarcopterygian
fishes (A), the four‐stroke pump found in most actinopterygian fishes (B). See text for details on
all pumping mechanisms. [From Randall et al. (1981a) and Kardong (2002).]

simultaneously. Lung emptying is due to a combination of elastic recoil,
contraction of muscles within the lung wall, hydrostatic forces, and the
negative pressure created by buccal expansion. In the next step, buccal
compression in series with jaw closure and sealing of the opercula forces
mixed air into the lungs with any excess being expelled through the mouth,
operculae, or nares (Figure 4.6B) (McMahon, 1969; Brainerd, 1994).

4.4.2. Actinopterygian Fishes
    While in sarcopterygian fishes, air breathing occurs in two phases, in
actinopterygian fishes this occurs in four phases. In the former case, initial
buccal expansion occurs with the mouth closed and draws air from the ABO
into the buccal cavity. This may be assisted by elastic recoil of the ABO as
well as compression of muscles in the wall of the ABO. Hydrostatic pressure
gradients in submerged fish may also assist in this air movement. This air is
then expelled during buccal compression through the mouth or operculae.
A second buccal expansion now draws in fresh air through the open mouth
and the subsequent buccal compression, which takes place with the operculae
and mouth closed, forces this air into the ABO (Figure 4.6A) (Liem, 1988;
Brainerd, 1994).
4.   VENTILATORY SYSTEMS                                                                 195

4.4.3. Polypterids
    A notable exception to this general trend is found in the polypterids (Poly-
pterus and Erpetoichthys) in which elastic recoil from emptying of the
lungs leads to aspiration breathing. Exhalation in these fishes is driven by
contraction of the lung wall, which also deforms the body wall. When the
muscles subsequently relax, a negative, recoil pressure is created within the
lungs, enhanced by the ganoid scale‐reinforced skin and body wall, which
serves to reinflate the lungs (Figure 4.7) (Purser, 1926; Brainerd et al., 1989).

                                     Stiff dermal armor

                                                       Contraction of lung

                              Escaping spent air

                                                             Inward buckling
     Fresh air sucked                                        of ventral wall
     in by aspiration

                                                               Elastic recoil

Fig. 4.7. Schematic diagrams illustrating the modified dual pump found in polypterids. See text
for details on all pumping mechanisms. [From Liem et al. (2001).]
196                                                     EMILY COOLIDGE ET AL.

Claims of suction filling of lungs by estivating Protopterus (Lomholt et al.,
1975) and of ABOs by Arapaima (Farrell and Randall, 1978) have not been
substantiated (DeLaney and Fishman, 1977; Greenwood and Liem, 1984).


    The ability to assess respiratory control in fishes requires experiments that
distinguish between the diVerent sites where respiratory stimulation might
occur. For example, hypoxia and/or hypercapnia may stimulate externally
oriented (water) or internally oriented (blood or brain tissue) chemoreceptors.
Respiratory control in purely water‐breathing fishes supports the hypothesis
that there are two separate populations of O2‐sensitive chemoreceptors: in
general, one population monitors the internal environment (blood) and elicits
ventilatory reflexes while those that monitor the external environment (water)
elicit both ventilatory and cardiovascular reflexes. Mechanical deformation
of the gills or ABO is sensed by mechanoreceptors that may be integrated into
the overall ventilatory response. Thus, a full understanding of respiratory
control, and the mechanisms involved in these responses, needs to distinguish
between stimuli arising from water, blood, and, if present, the ABO.

5.1. Hypoxic and Hypercarbic Ventilatory Responses and Reflex Pathways
5.1.1. Agnathans
    There has been little work on the ventilatory responses to hypoxia and/or
hypercarbia in hagfishes or lampreys. Hagfishes are known to be very hyp-
oxia tolerant with a low metabolic rate (Munz and Morris, 1965) and high
anaerobic potential (Sidell et al., 1984). Measurements of ventilation of
hagfishes are rare, probably owing to the diYculty of making direct measure-
ments of ventilation on an animal that produces copious amounts of slime
when disturbed (Lim et al., 2006). Branchial ventilation has been measured in
animals at rest in 7  C water and the frequency was 18 beats minÀ1 generating
a water flow of 0.019 ml minÀ1 gÀ1 (SteVensen et al., 1984). Given that
hagfish are hypoxia, and perhaps, anoxia tolerant, it is possible that ventila-
tion in these animals is oxyconforming; that is, ventilation may be depressed
on hypoxic exposure and hagfish rely primarily on anaerobic metabolism.
On the other hand, hypoxic exposure results in the release of catecholamines
in hagfish (Bernier et al., 1996), and catecholamines have been shown
to increase ventilation in fish. The ventilatory response to hypoxia in hag-
fish awaits experimentation in these animals and a solution to the ‘‘slime
4.   VENTILATORY SYSTEMS                                                     197

    Compared with hagfishes, there has been considerably more work on the
ventilatory responses to hypoxia/hypercarbia in lampreys. Larval lampreys
(ammocoetes) are not particularly hypoxia tolerant, although hypoxia toler-
ance varies inversely with temperature (Potter et al., 1970). Ventilation
frequency in Ichthyomyzon hubbsi ammocoetes increases with hypoxia, but
eventually decreases over time when PO2 < 10 mmHg at 15.5  C (Potter et al.,
1970). Direct measurements of ventilation in larval sea lampreys (Petromy-
zon marinus) indicate that severe hypoxia (4% O2) increases ventilation by
nearly tenfold with increases in both respiratory frequency and stroke vol-
ume that account for the large increase in minute ventilation (Rovainen and
Schieber, 1975). In the same study, aquatic hypercarbia (3% CO2) also
significantly increased minute ventilation by about fivefold with increases
in both ventilatory frequency and stroke volume (Rovainen and Schieber,
1975); however, the authors were unsure whether the increased ventilation in
response to hypercarbia resulted from the reduction in aquatic pH or a direct
eVect of CO2 as a ventilatory stimulus.
    Adult lampreys (Lampetra fluviatilis) increase ventilation frequency
in response to reductions in O2 saturation (Claridge and Potter, 1975).
At a temperature of 9.5  C, frequency increased from 60 breaths minÀ1
with 100% air saturation to $180 breaths minÀ1 at 15% air saturation. As
a consequence, they can maintain or even increase their O2 uptake down to a
PO2 as low as 10 mmHg even in comparatively warm water (Claridge and
Potter, 1975). Further reductions to 7.5% saturation, however, were lethal.
Although ventilation volume was not measured, it is likely that volume
increases also contributed to the overall increase in ventilation.

5.1.2. Dipnoans
    Members of the three extant lungfish genera, Protopterus, Lepidosiren,
and Neoceratodus, occupy an important position in the evolution of tetra-
pods. In addition, these groups have provided a substantial amount of
information about the transition from aquatic to aerial ventilation. The
African and South American lungfishes, Protopterus and Lepidosiren, respec-
tively, are primarily obligate air‐breathers and exhibit little or no response to
aquatic hypoxia (Johansen and Lenfant, 1967, 1968), whereas the Australian
lungfish, Neoceratodus, is a facultative air‐breather that uses its gills as the
primary gas exchange organ (Johansen et al., 1967a; Fritsche et al., 1993).
African lungfish ventilate the gills, but gill ventilation has been found both to
be unresponsive to aquatic hypoxia (Johansen and Lenfant, 1968) and
to exhibit increases in response to hypoxia (Jesse et al., 1968). The lungfish
in the latter study were smaller than in the study by Johansen and Lenfant
(1968), thus developmental stage might be responsible for the diVerences
between the two studies. Branchial denervation in Protopterus abolished the
198                                                     EMILY COOLIDGE ET AL.

increase in gill ventilation and enhanced air‐breathing behavior during
hypoxia (Lahiri et al., 1970), suggesting that branchial O2 chemoreceptors
are a vital component of aerial respiratory control in these animals. Experi-
ments with the African lungfish, Protopterus amphibius, indicate that juvenile
members of this species use the gills to obtain about 70% of total O2 uptake,
whereas adults use aquatic respiration to obtain about 10–15% of the total
O2 uptake (Johansen et al., 1967b). However, a study with Protopterus
aethiopicus indicates that in hypoxic water, both juvenile and adult animals
obtain more than 90% of O2 uptake via aerial respiration (Seifert and
Chapman, 2006). Although African lungfish are largely unresponsive to
hypoxic water, lungfish forced to breathe hypoxic gas from the aerial
environment or when N2 is injected into the lung significantly increase air‐
breathing frequency (Burggren and Johansen, 1968; Johansen and Lenfant,
1968; Perry et al., 2005), suggesting that lungfish have internally oriented
O2‐sensitive chemoreceptors. The Australian lungfish, as a facultative air‐
breather, inhibits gill ventilation and increases air‐breathing frequency in
response to aquatic hypoxia (Johansen et al., 1967a; Fritsche et al., 1993),
similar to the responses of other facultative air‐breathing fish such as gar
(Lepisosteus), bowfin (Amia) and reedfish (Erpetoichthys) (Pettit and
Beitinger, 1981; Smatresk et al., 1986; McKenzie et al., 1991a).
    Hypercarbia in the African lungfish reduces gill ventilation rate while
increasing air‐breathing frequency (Johansen and Lenfant, 1968). In the
South American lungfish, prolonged aquatic or combined aquatic/aerial
hypercarbia elicits an initial large increase in air‐breathing frequency that
steadily declines to preexposure levels after 8 h (Sanchez et al., 2005).

5.1.3. Polypterids
    The African reedfish (Erpetoichthys calabaricus), a bimodally breathing
fish, is capable of supporting metabolism exclusively by air breathing in
aquatic hypoxia (Pettit and Beitinger, 1981), although in normoxic condi-
tions it obtains most of its O2 requirements through gill ventilation (Pettit
and Beitinger, 1985). Reedfish are also amphibious, making terrestrial excur-
sions that are supported entirely by air breathing (Sacca and Burggren,
1982). This fish also does not avoid low dissolved O2 concentrations
(0.5 mg literÀ1) in comparison to an obligate water‐breathing fish (Percina
caprodes), suggesting that habitat selection and distribution of E. calabaricus
is not limited by aquatic hypoxia (Beitinger and Pettit, 1984). Gill ventilation
in E. calabaricus is reduced in hyperoxia (100% O2), and is completely
inhibited in aquatic hypoxia when the fish have access to aerial normoxia
(Pettit and Beitinger, 1981); however, exposure to combined aquatic and
aerial hypoxia does not alter gill ventilation from control levels (Pettit
and Beitinger, 1985), suggesting a complex interaction of O2 partitioning
between the aquatic and aerial environments. In the same study, gill ventilation
4.   VENTILATORY SYSTEMS                                                  199

increased significantly in response to low (0.5%) or high (5.0%) levels of CO2
(Pettit and Beitinger, 1985). Air breathing in E. calabaricus is strongly sti-
mulated by aquatic hypoxia (3% O2), but also increases in response to
combined hypoxia/hypercarbia (8% O2/5% CO2), or by hypercarbia alone
(5% CO2) (Pettit and Beitinger, 1985). The birchir (Polypterus senegalus), the
only other extant polypterid, also increases air breathing in response to
aquatic hypoxia (Magid, 1966) and is an obligate air‐breather even under
normoxic conditions (Magid and Babiker, 1975). Overall, air breathing is an
important feature of the respiratory physiology of polypterid fishes and
allows them to invade hypoxic/hypercarbic habitats.

5.1.4. Chondrostei
    The ventilatory responses of chondrosteans to aquatic hypoxia have been
the subject of several studies, but have been limited to sturgeons with no
apparent examination of the ventilatory responses to hypoxia in paddle-
fishes. The majority of the studies with sturgeon report typical ventilatory
responses to hypoxia including increases in ventilatory frequency and vol-
ume. One study, however, reported that the sturgeon A. transmontanus was
an oxyconformer, lowering its metabolic rate and ventilation in response to
progressive hypoxia (Burggren and Randall, 1978). This view was challenged
by a study in a congener (Acipenser baeri) that demonstrated oxyregulation
and increases in respiratory frequency and gill amplitude in response to
progressive hypoxia (Nonnotte et al., 1993). The discrepancies between the
two studies appear to result from technical diVerences in the measurements
of ventilation: the study by Burggren and Randall (1978) used a flexible
tube sewn directly to the mouth of the sturgeon which may have imposed a
resistance to gill water flow (see Nonnotte et al., 1993 for discussion). Other
studies support the view that sturgeon respond to hypoxia with increased
ventilatory frequency and increased opercular pressure amplitude (Maxime
et al., 1995; McKenzie et al., 1995) and are, until a critical PO2 is reached,
oxyregulators. Stimulation of external chemoreceptors in the sturgeon by
addition of NaCN to the inspired water elicited a transient bradycardia and
stimulated ventilation, whereas intra‐arterial injections of NaCN stimulated
ventilation and had no eVect on heart rate (McKenzie et al., 1995), support-
ing the hypothesis that two separate populations of chemoreceptors monitor
the external and internal environments and are responsible for the cardio-
ventilatory responses to hypoxia (Randall, 1982).
    Garfishes (Lepisosteus spp.) and bowfin (Amia calva) are bimodally
breathing fish with functional gills and a well‐vascularized gas bladder that
is used for air breathing. Several studies have examined gill ventilation
and air breathing in these species in response to hypoxic and hypercarbic
200                                                    EMILY COOLIDGE ET AL.

challenges. These fishes shift the emphasis of ventilation from gill ventilation
to air breathing as temperature increases or in response to aquatic hypoxia.
In the spotted gar, Lepisosteus oculatus, severe hypoxia ($12 mmHg) results
in a significant increase in air‐breathing frequency from about 1 breath hÀ1 to
8–9 breaths hÀ1, which supports 100% of the metabolic requirements of the
animal (Smatresk and Cameron, 1982a). Gill ventilation was significantly
depressed in hypoxia, decreasing from 35 breaths minÀ1 to 21 breaths minÀ1
(Smatresk and Cameron, 1982a). The depression of gill ventilation is thought
to limit the diVusional loss of O2 from blood to water during severe hypoxia.
A similar study in conscious longnose gar (Lepisosteus osseus) revealed that
progressive hypoxia significantly stimulated air breathing, but had little eVect
on gill ventilation frequency (Smatresk, 1986). Opercular pressure amplitude,
used as an index of gill ventilation volume, initially increased with moderate
hypoxia, but was inhibited with more severe hypoxia, as shown previously.
Smatresk (1986) also used NaCN applied to the inflow water and injected
intra‐arterially in an attempt to distinguish between externally oriented and
internally oriented O2 chemoreceptors that mediate the ventilatory responses
to hypoxia. NaCN applied externally or internally stimulated air‐breathing
frequency, indicating that O2‐sensitive chemoreceptors monitor both the
external and internal environments (Smatresk, 1986). Gill ventilation, how-
ever, was inhibited by NaCN applied externally, similar to the hypoxic
response, but stimulated gill ventilation when injected intra‐arterially, sug-
gesting that internal hypoxia exerts the dominant control over gill breathing
in gar (Smatresk, 1986). Studies with anesthetized, spontaneously breath-
ing gar where internal and external O2 levels were precisely regulated show
that hypoxia, whether internal or external, consistently stimulated air breath-
ing whereas the responses to hypoxia by gill breathing were more complex
(Smatresk et al., 1986). Hypoxia appeared to consistently inhibit gill ventila-
tion, regardless of internal O2 levels, but there was also an interaction
between internal and external O2 chemoreceptors for establishing gill venti-
lation levels. Hypercarbia significantly stimulated gill ventilation in spotted
gar, but had no consistent eVect on air breathing (Smatresk and Cameron,
    Bowfin (A. calva) is also a facultative air‐breather that responds to
hypoxia by increasing air‐breathing frequency (Johansen et al., 1970;
Randall et al., 1981a; McKenzie et al., 1991a). There are conflicting reports
on the response of gill ventilation to aquatic hypoxia. Some studies have
shown that severe hypoxia inhibits gill ventilation (Johansen et al., 1970), as
seen in gar, but studies using both hypoxia and NaCN show that bowfin does
not inhibit gill ventilation (McKenzie et al., 1991a). In Amia, denervation of
the cranial nerves serving the gill arches (glossopharyngeal and vagus) com-
bined with pseudobranch ablation (innervated by the facial nerve) eliminated
4.   VENTILATORY SYSTEMS                                                                          201

the air‐breathing responses to brief (15 min) aquatic hypoxia (McKenzie
et al., 1991a). Although denervation experiments can produce side eVects
such as altered behavior (see below), metabolic depression, and stress, cranial
nerve denervation has been used successfully to elucidate chemoreceptor
pathways in a number of fish species (Burleson et al., 1992).
    Additionally, Amia uses two types of air breaths that appear to be linked
to gas exchange and buoyancy functions (Figure 4.8) (Hedrick and Jones,
1993; Hedrick and Jones, 1999). Type I breaths (exhalation followed by
inhalation) are gas exchange breaths and are predominantly stimulated by
aquatic hypoxia, or by internal hypoxia created by forcing bowfin to breathe
hypoxic aerial gas mixtures (Figure 4.8A). Type I breaths also exhibit a
periodicity with a mean interbreath interval of about 30 min in normoxia
and 16 min in hypoxia (Hedrick et al., 1994). The periodic air‐breathing
pattern is thought to arise from periodic fluctuations in blood PO2 that
stimulate Type I (gas exchange) breaths. Type II breaths (inhalation only)
are stimulated by aerial hyperoxia (Hedrick and Jones, 1993) or by gas
bladder deflation (Hedrick and Jones, 1999), supporting the hypothesis
that Type II breaths regulate gas bladder volume and have a buoyancy‐
related function (Figure 4.8B). These studies point to the importance of
buoyancy as a proximate factor in the air‐breathing behavior of primitive
air‐breathing fishes. It is unknown if gar, or other air‐breathing fishes that

                          A                                       B
                                                         Inhale                          Inhale
                               B       Type I                                  Type II


                          30                                                   0
                           0                                               −10
      Air flow (ml s−1)



                                                                      100 ms

                                                           Inhale                        Inhale

                           0                     TE


                                   T            Exhale

Fig. 4.8. Records of air flow (ml sÀ1) for two A. calva in control conditions illustrating (A) Type I
and (B) Type II air‐breaths. For Type I breaths, the transfer phase (T) and expiratory time
interval (TE) is shown. [From Hedrick and Jones (1993).]
202                                                    EMILY COOLIDGE ET AL.

use the gas bladder as a buoyancy organ, also exhibits air‐breathing mechan-
isms similar to the Type II breaths described in bowfin. Complete branchial
denervation in Amia, without pseudobranch ablation, did not abolish air‐
breathing responses to aquatic hypoxia, but did interfere with the ability
to capture inhaled air by Type I air‐breaths (Hedrick and Jones, 1999).
Air‐breathing frequency actually increased due to the inability to transfer
inhaled gas to the gas bladder; the increase in frequency was due to a large
increase in the number of Type II air‐breaths further supporting the hypoth-
esis that Type II breaths monitor gas bladder volume and have a buoyancy‐
related function (Hedrick and Jones, 1999). These experiments also suggest
that some components of branchial innervation, perhaps arising from gill
mechanoreceptors, play an important role in the coordination of ventilatory
muscles important for air breathing.

5.2. Receptors Involved in Reflex Ventilatory Control
5.2.1. O 2‐Sensitive Chemoreceptors
    Physiological data provides indirect evidence for the location of
O2‐sensitive chemoreceptors. However, in combination with whole‐animal
responses, putative locations for chemoreceptors may be distinguished using
electrophysiology and histology. Chemoreceptive cells are characterized by
neurotransmitter storage and possess irregular discharge at rest, which
increases frequency and irregularity exponentially with decreasing levels of
O2 (PO2 or CaO2).
    In all classes of vertebrates, O2‐sensitive chemoreceptors are composed of
glomus cells (Type I), companion cells (Type II), and nerve tissue. Emphasis
has been placed on the role of the glomus cells in aVerent sensing, and these
cells are characterized by cytoplasmic vesicles, ribosomes, endoplasmic retic-
ulum, and mitochondria, similar to endocrine tissue. Putative chemoreceptive
cells, neuroepithelial cells (NEC), in the gills of water‐breathing fish share
many similar characteristics with the glomus cells of the mammalian carotid
body. Examination of the NECs have shown serotonin (5‐HT) to be the major
monoamine present in the dense‐cored vesicles and 5‐HT has been used
as a marker for these cells (Dunel‐Erb et al., 1982; Zaccone et al., 1992;
Goniakowska‐Witalinska et al., 1995; Jonz and Nurse, 2003). Neuroepithelial
cells have been identified in a few primitive fish, including the gills and lungs
of Protopterus (Figure 4.9), the swimbladder of Polypterus, and the gills
of bowfin (Zaccone et al., 1989; Adriaensen et al., 1990; Goniakowska‐
Witalinska et al., 1995). Therefore, it appears that putative O2 chemorecep-
tors carrying aVerent information to a respiratory center are present in both
the respiratory structures of the gills and ABOs.
4.   VENTILATORY SYSTEMS                                                                    203


                     B                                              A



Fig. 4.9. (A) Serotonin‐like immunoreactive cells (brown) within the epithelium of the alveoli of
Protopterus annectens (100Â) (PAP method) [(from Zaccone et al. (1989)] and (B) low‐power
electron micrograph of a solitary neuroendocrine cell in the pneumatic duct region of the lung
of P. aethiopicus. Dense cored vesicles are observed in the basal cytoplasm (6800Â). [From
Adriaensen et al. (1990).] NE, neuroendocrine cell.

5.2.2. Peripheral CO 2‐Sensitive Chemoreceptors
    Ventilation, aquatic, aerial or both, increased in fish exposed to environ-
mental acidosis, hypercapnia, and arterial acidosis. Acid–base disturbances
come hand‐in‐hand with changes in CO2 dissociation, so it is diYcult to
isolate the stimulus for potential chemoreceptors as solely CO2 or concomi-
tant fluctuations in pH. CO2 sensitivity may have evolved with primitive
bimodal breathing fish, as these animals continually face the challenge of
both O2 and CO2 regulation. Although the O2 ventilatory drive is supported
by the lower content of O2 than CO2 in water, obligate air‐breathing fish,
such as the African lungfish, increased both air‐breathing frequency and
branchial ventilation in mildly hypercarbic water (Jesse et al., 1968;
Johansen and Lenfant, 1967; Johansen et al., 1967a). However, higher levels
204                                                    EMILY COOLIDGE ET AL.

of CO2 in the water decreased aquatic ventilation. The isolated brainstem of
L. osseus is responsive to changes in superfused CO2 as well as changes in O2
(Wilson et al., 2000); however, Amia is not responsive to any changes in
arterial CO2 levels (McKenzie et al., 1991b).
    There is no histological identification of specific CO2‐sensitive chemor-
eceptors in primitive fish; however, the proposed location is again the
gills. A separate population of CO2/Hþ chemoreceptors, separate from the
O2‐sensitive NECs, projecting aVerent sensory input may exist. Additionally,
CO2 sensitivity may be associated with central chemosensitivity, and research
using hypercarbic and acidotic superfusates on isolated brains is discussed
briefly in the next section.

5.2.3. Central CO 2 Chemosensitivity
    The central site for respiratory integration remains unknown for both
water‐breathing and air‐breathing fish. Numerous studies have suggested
two separate central pattern generators (CPGs) for gill ventilation and air
breathing, presumably located in the medulla (Rovainen, 1977; Ballintijn,
1987; Taylor et al., 1999). [For more detailed reviews on the location and
function of CPG in fish, see Ballintijn (1987) and Taylor et al. (1999).]
    The isolated brainstem prep of longnose gar (L. osseus) showed a
motor pattern for lung ventilation that was responsive to CO2, and the
authors correlated this proposed central chemoreceptor to a CPG
(Remmers et al., 2001). However, this is not true of all air‐breathing fish.
Amia do not appear to have central chemoreceptors mediating cardioventi-
latory responses, as altering gas tensions and pH on the extradural fluid has
no eVect on any cardioventilatory variables in normoxic water (Hedrick
et al., 1991).

5.2.4. Mechanoreceptors
    Physical displacement of respiratory passages and gas exchange surfaces is
detected by mechanoreceptors, which are a broad group of receptors sensitive
to changes in flow, volume, transpulmonary pressure, and tension. Mechan-
oreceptors are characterized by simple free nerve endings located in connec-
tive tissue or muscle and it is the deformation of these nerve endings that
results in changes in membrane and channel geometry, ion flux, and mem-
brane potential, and therefore a change in discharge frequency. Typically, the
sensory modalities and response characteristics of these mechanoreceptors
are determined by the location of receptors. In fish, mechanoreceptors have
been identified in the buccal and opercular cavities, pharynx, gill arches, gill
rakers and filaments, and ABOs.
    In general, there are two divisions of mechanoreceptors based on their
response to the degree or rate of change during inspiration and expiration.
Slowly adapting receptors (SAR) demonstrate a static change in discharge
4.   VENTILATORY SYSTEMS                                                    205

that is proportional to the degree of change in flow, volume, pressure, or
tension. The second group, rapidly adapting receptors (RAR), responds to
the rate of change, showing a dynamic change in discharge with an initial
burst on inspiration and a second burst corresponding to deflation of the
respiratory surfaces. Mechanoreceptors that are sensitive to both the degree
and rate of change are characterized as SAR with a dynamic aspect that
gradually levels oV to a steady state in proportion to the change in volume,
pressure, and tension.

    a. Gill Mechanoreceptors. Although proprioceptors do exist in the buccal
and opercular cavities of fish, the gills are the primary site of mechanorecep-
tors aVecting ventilatory control in water‐breathing fish. Within the gills,
mechanoreceptors have been identified along the gill filaments, gill rakers,
and in the cartilaginous strip between the gill arches (De Graaf and
Ballintijn, 1987). The receptors of the gill filaments are of the slowly adapting
variety with a dynamic response to displacement. Gill raker mechanorecep-
tors are phasic and predominantly detect damaging material and work to
maintain the gill sieve by producing the ‘‘cough’’ reflex of fish and a reflect
bradycardia (De Graaf and Ballintijn, 1987). However, there is no evidence
that they contribute to respiratory control (Burleson et al., 1992).

    b. ABO Mechanoreceptors. With the addition of air breathing in fish,
mechanoreceptors are found in both the gills and ABOs. Studies documented
both SAR and RAR of their ABOs in lungfish and gar (DeLaney et al., 1983;
Smatresk and Azizi, 1987), while bowfin only appeared to have SAR of its
swimbladder (Milsom and Jones, 1985). The SAR of the lungfish increased
firing with progressive inflation, but also displayed a dynamic component as
a function of the inflation rate (Fishman et al., 1989). Receptors sensitive to
the degree of physical change in the lung predominated, while RAR fired
briefly only during inflation and deflation. AVerent information from both
of the degree and rate of inflation in the lungfish was transmitted to the
brainstem where it likely plays a role in ventilatory control (Fishman et al.,
1989). Lungfish stretch receptors responded strongest to transpulmonary
pressure yet they also had a unique inhibition of receptor discharge in
intrapulmonary hypercapnia (DeLaney et al., 1983). Receptor inhibition by
intrapulmonary CO2 concentrations also occurred in the spotted gar,
L. oculatus (Smatresk and Azizi, 1987); however in A. calva, the receptors
were insensitive to CO2 concentrations (Milsom and Jones, 1985). The ABOs
of these primitive fish appeared to function both in buoyancy control as
well as gas exchange, thus adding to the importance of aVerent information
of stretch receptors (Johansen et al., 1967b; Smatresk and Azizi, 1987).
In support of this hypothesis, rapid deflation of the ABO in Amia elicited
a Type II breath while inflation causes fish to stop an air‐breath attempt
(Hedrick and Jones, 1999).
206                                                                  EMILY COOLIDGE ET AL.


    The ‘‘primitive fishes’’ described in this volume display a wide variety of
ventilatory patterns and control mechanisms. As this chapter has illustrated,
many of these fishes are facultative or obligate air‐breathers, thus increasing
the complexity of control mechanisms when several sites and receptors are
integrated into the overall ventilatory response. Because many of these fishes
use both water and air breathing in their respiratory behavior, they are
important for understanding the physiological mechanisms in the transition
from water to land. Many studies have described the reflex responses to
environmental perturbations such as hypoxia and hypercarbia, but one
area that clearly needs more study is how peripheral receptors and neural
pathways are centrally integrated to produce the complex ventilatory beha-
viors seen in these fishes. Although these fishes are highly evolved and
adapted to their particular ecological niches, further examination of mechan-
isms underlying respiratory behavior may lead to greater insight into the
evolutionary forces that have shaped the transition from water to land
among vertebrates.


Adriaensen, D., Scheuermann, D. W., Timmermans, J.‐P., and De Groodt‐Lassel, M. H. A.
     (1990). Neuroepithelial endocrine cells in the lung of the lungfish Protopterus aethiopicus. An
     electron‐ and fluorescence‐microscopial investigation. Acta Anat. 139, 70–77.
Almeida‐Val, V. H. M., and Val, A. L. (1993). Evolutionary trends of LDH isozymes in fishes.
     Comp. Biochem. Physiol. 105B, 21–28.
Ballintijn, C. M. (1987). Evaluation of central nervous control of ventilation in vertebrates.
     In ‘‘The Neurobiology of the Cardiovascular System’’ (Taylor, E. W., Ed.), pp. 3–27.
     Manchester University, Manchester, UK.
Beitinger, T. L., and Pettit, M. J. (1984). Comparison of low oxygen avoidance in a bimodal
     breather, Erpetoichthys calabaricus and an obligate water breather, Percina caprodes.
     Environ. Biol. Fish 11, 235–240.
Bernier, N. J., Harris, J., Lessard, J., and Randall, D. J. (1996). Adenosine receptor blockade and
     hypoxia‐tolerance in rainbow trout and Pacific hagfish. I. EVects on anaerobic metabolism.
     J. Exp. Biol. 199, 485–495.
Booth, J. H. (1978). The distribution of blood flow in the gills of fish: Application of a new
     technique to rainbow trout (Salmo gairdneri). J. Exp. Biol. 73, 119–129.
Brainerd, E. L. (1994). The evolution of lung‐gill bimodal breathing and the homology of
     vertebrate respiratory pumps. Am. Zool. 34, 289–299.
Brainerd, E. L., Liem, K. F., and Samper, C. T. (1989). Air ventilation by recoil aspiration in
     polypterid fishes. Science 246, 1593–1595.
Brauner, C. J., Matey, V., Wilson, J. M., Bernier, N. J., and Val, A. L. (2004). Transition in
     organ function during the evolution of air‐breathing; Insights from Arapaima gigas, an
     obligate air‐breathing teleost from the Amazon. J. Exp. Biol. 207, 1433–1438.
4.   VENTILATORY SYSTEMS                                                                       207

Burggren, W. W. (1978). Gill ventilation in the sturgeon, Acipenser transmontanus: Unusual
     adaptations for bottom dwelling. Respir. Physiol. 34, 153–170.
Burggren, W. W., and Johansen, K. (1968). Circulation and respiration in lungfishes (Dipnoi).
     J. Morphol. (Suppl. 1), 217–236.
Burggren, W. W., and Randall, D. J. (1978). Oxygen uptake and transport during hypoxic
     exposure in the sturgeon Acipenser transmontanus. Respir. Physiol. 34(2), 171–183.
Burggren, W. W., Johansen, K., and McMahon, B. (1985). Respiration in phyletically ancient
     fishes. In ‘‘Evolutionary Biology of Primitive Fishes’’ (Forman, R. E., Gorbman, A.,
     Dodd, J. M., and Olsson, R., Eds.), pp. 217–252. Plenum Press, New York.
Burleson, M. L., Smatresk, N. J., and Milsom, W. K. (1992). AVerent inputs associated with
     cardioventilatory control in fish. In ‘‘Fish Physiology’’ (Hoar, W. S., and Randall, D. J.,
     Eds.), pp. 389–426. Academic Press, New York.
Byczkowska‐Smyk, W. (1962). Vascularization and size of the respiratory surface in Acipenser
     stellatus. Respir. Physiol. 34, 171–183.
Cameron, C., Garey, J., and Swalla, B. J. (2000). Evolution of the chordate body plan:
     New insights from phylogenetic analysis of deuterostome phyla.. Proc. Natl. Acad. Sci.
     USA 97, 4469–4474.
Carter, G. S., and Beadle, L. C. (1931). Notes on the habits and development of Lepidosiren
     paradoxa. J. Linn. Soc. Lond. Zool. 37, 197–203.
Claridge, P. N., and Potter, I. C. (1975). Oxygen consumption, ventilatory frequency and heart
     rate of lampreys (Lampetra fluviatilis) during their spawning run. J. Exp. Biol. 63, 193–206.
Daxboeck, C., Barnard, D. K., and Randall, D. J. (1981). Functional morphology of the gills of
     the bowfin, Amia calva, with special reference to their significance during air exposure.
     Respir. Physiol. 43, 349–364.
De Graaf, P. J. F., and Ballintijn, C. M. (1987). Mechanoreceptor activity in the gills of carp. II.
     Gill arch proprioceptors. Respir. Physiol. 233, 183–194.
DeLaney, R. G., and Fishman, A. P. (1977). Analysis of lung ventilation in the aestivating
     lungfish Protopterus aethiopicus. Am. J. Physiol. 233(5), R181–R187.
DeLaney, R. G., Laurent, P., Galante, R. J., Pack, A. I., and Fishman, A. P. (1983). Pulmonary
     mechanoreceptors in the dipnoi lungfish Protopterus and Lepidosiren. Am. J. Physiol. 232,
Dunel‐Erb, S., Bailly, Y., and Laurent, P. (1982). Neuroepithelial cells in fish gill primary
     lamellae. J. Appl. Physiol. 53, 1342–1353.
Farrell, A. P., and Randall, D. J. (1978). Air‐breathing mechanics in two Amazonian teleosts,
     Arapaima gigas and Hoplyerythrinus unitaeniatus. Can. J. Zool. 56, 939–945.
Feder, M. E., and Burggren, W. (1985). Cutaneous gas exchange in vertebrates: Design, patterns,
     control and implications. Biol. Rev. (Camb. Philos. Soc.) 60(1), 1–45.
Fishman, A. P., Galante, R. J., and Pack, A. I. (1989). Diving physiology in lungfish.
     In ‘‘Comparative Pulmonary Physiology: Current Concepts’’ (Wood, S. C., Ed.),
     pp. 645–676. Dekker, New York.
Fritsche, R., Axelsson, M., Franklin, C. E., Grigg, G. G., Holmgren, S., and Nilsson, S. (1993).
     Respiratory and cardiovascular responses to hypoxia in the Australian lungfish. Respir.
     Physiol. 94, 173–187.
Gilmour, T. H. J. (1979). Feeding in pterobranch hemichordates and the evolution of gill slits.
     Can. J. Zool. 57, 1136–1142.
Goniakowska‐Witalinska, L., Zaccone, G., Fasulo, S., Mauceri, A., Licata, A., and Youson, J.
     (1995). Neuroendocrine cells in the gills of the bowfin Amia calva. An ultrastructural and
     immunocytochemical study. Folia Histochem. Cytobiol. 33, 171–177.
Graham, J. B. (1997).‘‘Air‐Breathing Fishes: Evolution, Diversity and Adaptation,’’ 299 p.
     Academic Press, New York.
208                                                                   EMILY COOLIDGE ET AL.

Greenwood, P. H. (1958). Reproduction in the East African lungfish Protopterus aethiopicus.
     Proc. Zool. Soc. (Lond.) 130, 547–567.
Greenwood, P. H., and Liem, K. F. (1984). Aspiratory respiration in Arapaima gigas (Teleostei,
     Osteoglossomorpha): A reappraisal. J. Zool. (Lond.) 203, 411–425.
Gutmann, W. F. (1981). Relationships between invertebrate phyla based on functional mechani-
     cal analysis of the hydrostatic skeleton. Am. Zool. 21, 63–81.
Hedrick, M. S., and Jones, D. R. (1993). The eVects of altered aquatic and aerial respiratory gas
     concentrations on air‐breathing patterns in a primitive fish (Amia calva L.). J. Exp. Biol. 181,
Hedrick, M. S., and Jones, D. R. (1999). Control of gill ventilation and air‐breathing in the
     bowfin Amia calva. J. Exp. Biol. 202, 87–94.
Hedrick, M. S., Burleson, M. L., Jones, D. R., and Milsom, W. K. (1991). An examination of
     central chemosensitivity in an air‐breathing fish (Amia calva). J. Exp. Biol. 155, 165–174.
Hedrick, M. S., Katz, S. L., and Jones, D. R. (1994). Periodic air‐breathing behaviour in a
     primitive fish revealed by spectral analysis. J. Exp. Biol. 197, 429–436.
Hughes, G. M. (1984). Measurement of gill area in fishes: Practices and problems. J. Mar. Biol.
     Assoc. UK 64, 637–655.
Hughes, G. M., and Morgan, M. (1973). The structure of fish gills in relation to their respiratory
     function. Biol. Rev. 48, 419–475.
Jarvik, E. (1980). ‘‘Basic Structure and Evolution of Vertebrates.’’ Academic Press, London.
Jensen, D. (1966). The hagfish. Sci. Am. 214, 82–90.
Jesse, J., Shub, C., and Fishman, A. P. (1968). Lung and gill ventilation of the African lungfish.
     Respir. Physiol. 3, 267–287.
Johansen, K. (1970). Air‐breathing in fishes. In ‘‘Fish Physiology’’ (Hoar, W. S., and
     Randall, D. J., Eds.), pp. 361–411. Academic Press, New York.
Johansen, K., and Lenfant, C. (1967). Respiratory function in the South American lungfish,
     Lepidosiren paradoxa (Fitz.). J. Exp. Biol. 46, 205–218.
Johansen, K., and Lenfant, C. (1968). Respiration in the African lungfish, Protopterus aethiopi-
     cus. II. Control of breathing. J. Exp. Biol. 49, 453–468.
Johansen, K., and Strahan, R. (1963). The respiratory system of Myxine glutinosa. In ‘‘The
     Biology of Myxine’’ (Brodal, A., and Fange, R., Eds.), pp. 352–371. Universitetsforlaget,
     Oslo, Norway.
Johansen, K., Lenfant, C., and Grigg, G. G. (1967a). Respiratory control in the lungfish,
     Neoceratodus forteri (KreVt). Comp. Biochem. Physiol. 20, 835–854.
Johansen, K., Lomholt, J.P, and Maloiy, G. M. O. (1967b). Importance of air and water
     breathing in relation to size of the African lungfish Protopterus amphibius Peters. J. Exp.
     Biol. 65, 395–399.
Johansen, K., Hansen, D., and Lenfant, C. (1970). Respiration in a primitive air breather, Amia
     calva. Respir. Physiol. 9, 162–174.
Jonz, M. G., and Nurse, C. A. (2003). Neuroepithelial cells and associated innervation of the
     zebrafish gill. A confocal immunofluorescence study. J. Comp. Neurol. 461, 1–17.
Junk, W. J., Soares, G. M., and Carvalho, F. M. (1983). Distribution of fish species in a lake of
     the Amazon river floodplain near Manaus (Lago Camaleao), with special reference to
     extreme oxygen conditions. Amazoniana 7(4), 397–431.
Kardong, K. V. (2002). ‘‘Vertebrates: Comparative Anatomy, Function, Evolution.’’ McGraw‐
     Hill, New York.
Korolewa, N. W. (1964). Water respiration of lamprey and survival in a moist atmosphere.
     Isv. vses. nauchno‐issled. Inst. ozern. rechn. ryb. Khoz. 58, 186–190 (in Russian).
Lahiri, S., Szidon, J. P., and Fishman, A. P. (1970). Potential respiratory and circulatory
     adjustments to hypoxia in the African lungfish. Fed. Proc. 29, 1141–1148.
4.   VENTILATORY SYSTEMS                                                                    209

Lewis, S. V. (1976). Respiration and gill morphology of the paired species of lampreys. Lampetra
    fluviatilis (L.) and Lampetra planeri (Bloch). Ph.D. thesis, University of Bath, UK.
Lewis, S. V., and Potter, I. C. (1976). Gill morphometrics of the lampreys, Lampetra fluviatilis
    (L.) and Lampetra planeri (Bloch). Acta Zool. 57, 103–112.
Liem, K. (1982). Larvae of air‐breathing fishes as countercurrent flow devices in hypoxic
    environments. Science 211, 1177–1179.
Liem, K. F. (1988). Form and function of lungs: The evolution of air breathing mechanisms.
    Am. Zool. 28, 739–759.
Liem, K. F., Bemis, W. E., Walker, W. F., and Grande, L. (2001). ‘‘Functional Anatomy of
    the Vertebrates: An evolutionary perspective,’’ 586 p. Brooks/Cole‐Thompson Learning,
    Belmont, CA.
Lim, J., Fudge, D. S., Levy, N., and Gosline, J. M. (2006). Hagfish slime ecomechanics: Testing
    the gill‐clogging hypothesis. J. Exp. Biol. 209, 702–710.
Lomholt, J. P., Johansen, K., and Maloiy, G. M. O. (1975). Is the aestivating lungfish the first
    vertebrate with sectional breathing? Nature 257, 787–788.
Magid, A. M. (1966). Breathing and function of the spiracles in Polypterus senegalus. Anim.
    Behav. 14, 522–530.
Magid, A. M., and Babiker, M. M. (1975). Oxygen consumption and respiratory behavior of
    three Nile fishes. Hydrobiologia 46, 359–367.
Maina, J. N. (2003). Functional morphology of the vertebrate respiratory systems. In ‘‘Biological
    Systems in Vertebrates’’ (Dutta, H. M., and Kline, D. W., Eds.). 192 p. Science Publishers,
Mallatt, J. (1996). Ventilation and the origin of jawed vertebrates: A new mouth. Zool. J. Linn.
    Soc. 117, 329–404.
Maxime, V., Nonnotte, G., Peyraud, C., Williot, P., and Truchot, J. P. (1995). Circulatory and
    respiratory eVects of an hypoxic stress in the Siberian sturgeon. Respir. Physiol. 100,
McKenzie, D. J., Burleson, M. L., and Randall, D. J. (1991a). The eVects of branchial denerva-
    tion and pseudobranch ablation on cardioventilatory control in an air‐breathing fish.
    J. Exp. Biol. 161, 347–365.
McKenzie, D. J., Aota, S., and Randall, D. J. (1991b). Ventilatory and cardiovascular responses
    to blood pH, plasma PCO2, blood O2 content and catecholamines in an air‐breathing fish,
    Amia calva. Physiol. Zool. 64, 432–450.
McKenzie, D. J., Taylor, E. W., Bronzi, P., and Bolis, C. L. (1995). Aspects of cardioventilatory
    control in the Adriatic sturgeon (Acipenser naccarii). Respir. Physiol. 100, 45–53.
McMahon, B. R. (1969). A functional analysis of the aquatic and aerial respiratory movements
    of an African lungfish, Protopterus aethiopicus, with reference to the evolution of the lung
    ventilation mechanism in vertebrates. J. Exp. Biol. 51, 407–430.
Milligan, C. L., and Wood, C. M. (1987). Regulation of blood oxygen transport and red cell pHi
    after exhaustive activity in rainbow trout (Salmo gairdneri) and starry flounder (Platichthys
    stellatus). J. Exp. Biol. 133, 263–282.
Milsom, W. K., and Jones, D. R. (1985). Characteristics of mechanoreceptors in the air‐
    breathing organ of the holostean fish, Amia calva. J. Exp. Biol. 117, 389–399.
Munz, F. W., and Morris, R. W. (1965). Metabolic rate of the hagfish, Eptatretus stouti (Lock-
    ington). Comp. Biochem. Physiol. 16, 1–6.
Nikinmaa, M. (1990). ‘‘Vertebrate Red Blood Cells. Adaptations of Function to Respiratory
    Requirements,’’ 262 p. Springer, Berlin, Heidelberg, New York.
Nonnotte, G., Maxime, V., Truchot, J. P., Williot, P., and Peyraud, C. (1993). Respiratory
    responses to progressive ambient hypoxia in the sturgeon, Acipenser baeri. Respir. Physiol.
    91, 71–82.
210                                                                EMILY COOLIDGE ET AL.

Perry, S. F., and Kinkead, R. (1989). The role of catecholamines in regulating arterial oxygen
     content during acute hypercapnic acidosis in rainbow trout (Salmo gairdneri). Respir.
     Physiol. 77, 365–377.
Perry, S. F., Gilmour, K. M., Vulesevic, B., McNeill, B., Chew, S. F., and Ip, Y. K. (2005).
     Circulating catecholamines and cardiorespiratory responses in hypoxic lungfish (Proto-
     pterus dolloi): A comparison of aquatic and aerial hypoxia. Physiol. Biochem. Zool. 78,
Pettit, M. J., and Beitinger, T. L. (1981). Aerial respiration of the brachiopterygian fish,
     Calamoichthys calibaricus. Comp. Biochem. Physiol. A 68, 507–509.
Pettit, M. J., and Beitinger, T. L. (1985). Oxygen acquisition of the reedfish, Erpetoichthys
     calabaricus. J. Exp. Biol. 114, 289–306.
Potter, I. C., Hill, B. J., and Gentleman, S. (1970). Survival and behaviour of ammocoetes at
     low oxygen tensions. J. Exp. Biol. 53, 59–73.
Purser, G. L. (1926). Calamoichthys calabaricus (Smith, J. A.). Part I. The alimentary and
     respiratory systems. Trans. R. Soc. Edinburgh 54, 767–784.
Rahn, H., Rahn, K. B., Howell, B. J., Gans, C., and Tenney, S. M. (1971). Air breathing of the
     garfish (Lepisosteus osseus). Respir. Physiol. 11, 285–307.
Randall, D. J. (1972). Respiration. In ‘‘The Biology of Lampreys’’ (Haridisty, N. W., and
     Potter, I. C., Eds.), Vol. 2, pp. 287–306. Academic Press, London.
Randall, D. J. (1982). The control of respiration and circulation in fish during hypoxia and
     exercise. J. Exp. Biol. 100, 275–288.
Randall, D. J. (1990). Control and co‐ordination of gas exchange in water breathers.
     In ‘‘Advances in Comparative and Environmental Physiology. Vertebrate Gas Exchange:
     From Environmental to Cell’’ (Boutilier, R. G., Ed.), pp. 253–278. Springer, Berlin,
Randall, D. J., Burggren, W. W., Farrell, A. P., and Haswell, M. S. (1981a). ‘‘The Evolution of
     Air‐Breathing Vertebrates,’’ 133 p. Cambridge University Press, Cambridge.
Randall, D. J., Cameron, J. N., Daxboeck, C., and Smatresk, N. (1981b). Aspects of bimodal gas
     exchange in the bowfin, Amia calva L. (Actinopterygii: Amigormes). Respir. Physiol. 43,
Remmers, J. E., Thorgerson, C., Harris, M., Perry, S. F., Vasilakos, K., and Wilson, R. J. A.
     (2001). Evolution of central respiratory chemoreception: A new twist on an old story.
     Respir. Physiol. 129, 211–217.
Rovainen, C. M. (1977). Neural control of ventilation in the lamprey. Fed. Proc. 36, 2386–2389.
Rovainen, C. M., and Schieber, M. H. (1975). Ventilation of larval lampreys. J. Comp. Physiol.
     104, 185–203.
Sacca, R., and Burggren, W. W. (1982). Oxygen uptake in air and water in the air‐breathing
     reedfish, Calamoichthys calabaricus: Role of skin, gills and lungs. J. Exp. Biol. 97,
Sanchez, A. P., Giusti, H., Bassi, M., and Glass, M. L. (2005). Acid‐base regulation in the South
     American lungfish Lepidosiren paradox: Effects of prolonged hypercarbia and blood gases
     and pulmonary ventilation. Physiol. Biochem. Zool. 78, 908–915.
Sanderson, S. L., and Wassersug, R. (1990). Suspension‐feeding vertebrates. Sci. Am. 262(3),
Seifert, A. W., and Chapman, L. J. (2006). Respiratory allocation and standard rate of metabo-
     lism in the African lungfish, Protopterus aethiopicus. Comp. Biochem. Physiol. 143A,
Shelton, G. (1970). The regulation of breathing. In ‘‘Fish Physiology’’ (Hoar, W. S., and
     Randall, D. J., Eds.), Vol. IV, pp. 293–359. Academic Press, New York.
4.   VENTILATORY SYSTEMS                                                                     211

Shelton, G., Jones, D. R., and Milsom, W. K. (1986). Control of breathing in ectothermic
    vertebrates. In ‘‘Handbook of Physiology’’ (Cherniack, N. S., and Widdicombe, J. G.,
    Eds.), Vol. 2, pp. 857–909. American Physiological Society, Bethesda, MD.
Sidell, B. D., Stowe, D. B., and Hansen, C. A. (1984). Carbohydrate is the preferred metabolic
    fuel of the hagfish (Myxine glutinosa) heart. Physiol. Zool. 57, 266–273.
Smatresk, N. J. (1986). Ventilatory and cardiac reflex responses to hypoxia and NaCN in
    Lepisosteus osseus, an air‐breathing fish. Physiol. Zool. 59, 385–397.
Smatresk, N. J., and Azizi, S. Q. (1987). Characteristics of lung mechanoreceptors in spotted gar,
    Lepisosteus oculatus. Am. J. Physiol. 252, R1066–R1072.
Smatresk, N. J., and Cameron, J. N. (1982a). Respiration and acid‐base physiology of the
    spotted gar, a bimodal breather. I. Normal values, and the response to severe hypoxia.
    J. Exp. Biol. 96, 263–280.
Smatresk, N. J., and Cameron, J. N. (1982b). Respiration and acid‐base physiology of the
    spotted gar, a bimodal breather. II. Responses to temperature change and hypercapnia.
    J. Exp. Biol. 96, 281–293.
Smatresk, N. J., Burleson, M. L., and Azizi, S. Q. (1986). Chemoreflexive responses to hypoxia
    and NaCN in longnose gar: Evidence for two chemoreceptor loci. Am. J. Physiol. 251,
SteVensen, J. F., Johansen, K., Sindberg, C. D., Sorensen, J. H., and Mooler, J. L. (1984).
    Ventilation and oxygen consumption in the hagfish, Myxine glutinosa L. J. Exp. Mar. Biol.
    Ecol. 84, 173–178.
Taylor, E. W., Jordan, D., and Coote, J. H. (1999). Central control of the cardiovascular and
    respiratory systems and their interactions in vertebrates. Physiol. Rev. 79, 855–916.
Val, A. L., AVonso, E. G., and Almeida‐Val, V. M. F. (1992). Adaptive features of Amazon
    fishes: Blood characteristics of Curimata (Prochilodus cf nigricans, Osteichthyes). Physiol.
    Zool. 65(4), 832–843.
Whiting, H. P., and Bone, Q. (1980). Ciliary cells in the epidermis of the larval Australian
    dipnoan, Neoceratodus. Zool. J. Linn. Soc. 68, 125–137.
Wilson, R. J. A., Harris, M. B., Remmers, J. E., and Perry, S. F. (2000). Evolution of air‐
    breathing and central CO2/Hþ respiratory chemosensitivity: New insights from an old fish?
    J. Exp. Biol. 203, 3505–3512.
Wootton, R. J. (1990). ‘‘Ecology of Teleost Fishes,’’ 404 p. Chapman and Hall, New York.
Zaccone, G., Tagliaferro, G., Goniakowska‐Witalinska, L., Fasulo, S., Ainis, L., and
    Mauceri, A. (1989). Serotonin‐like immunoreactive cells in the pulmonary epithelium of
    ancient fish species. Histochemistry 92, 61–63.
Zaccone, G., Lauweryns, J. M., Fasulo, S., Tagliafierro, G., Ainis, L., and Licata, A. (1992).
    Immunocytochemical localization of serotonin and neuropeptides in the neuroendocrine
    paraneurons of teleost and lungfish gills. Acta Zool. 73, 177–183.
This page intentionally left blank


1. Introduction
2. Partitioning of O2 and CO2 Exchange Across the Respiratory Surfaces
   2.1. Primitive Ray‐Finned Fishes (Actinopterygii)
   2.2. Lobe‐Finned Fishes (Sarcopterygii)
   2.3. Jawless Fishes (Agnatha)
3. Blood O2 Transport
   3.1. General Principles of Hb Function
   3.2. Factors AVecting the ArterioVenous O2 DiVerence
   3.3. Survey of Extant Primitive Fishes
4. Transport and Elimination of CO2
   4.1. General Model of CO2 Transport and Excretion
5. Synthesis
   5.1. How Do Primitive Fishes Compete with Other Fishes?
   5.2. Primitive Fishes and the Evolution of Vertebrate Blood
        O2 and CO2 Transport Characteristics

    Gas exchange is a prerequisite of vertebrate life. In terms of structural and
functional diversity and habitats occupied, extant teleosts clearly outcompete
extant primitive fishes; however, there are a few aspects related to gas
exchange that may have contributed to the survival of these primitive fishes.
Most of the primitive fishes either have the ability to breath air, have the
ability to tolerate aerial exposure (and in some cases estivate), or are tolerant
to aquatic hypoxia. Many of the bimodal breathers retain fully functional
gills, which at times allow strictly aquatic breathing over prolonged periods
which may be important for aerial predator avoidance or surviving ice cover
in temperate climates. While air breathing is important for surviving aquatic
hypoxia, it is also important in enhancing O2 uptake during exercise. Living
primitive fishes occupy strategic positions in the evolutionary tree of

Primitive Fishes: Volume 26                          Copyright # 2007 Elsevier Inc. All rights reserved
FISH PHYSIOLOGY                                                 DOI: 10.1016/S1546-5098(07)26005-4
214                                           C. J. BRAUNER AND M. BERENBRINK

vertebrates and may shed light on the evolution of blood O2 and CO2 trans-
port characteristics. Evolutionary reconstruction indicates that the increase in
the Bohr–Haldane eVect in primitive ray‐finned fishes was followed first by a
gradual increase in the magnitude of the Root eVect and then a gradual
reduction in specific Hb buVer value. This was followed by the evolution of
a choroid rete mirabile and ocular O2 secretion in the last common ancestor of
Amia calva and teleosts. Finally, the adrenergic red blood cell Naþ/Hþ
exchanger was never present in primitive ray‐finned fishes or primitive teleosts
and only evolved in advanced teleosts. No such evolutionary trends are
observed in primitive lobe‐finned fishes.


    The uptake of O2 from the environment and elimination of metabolically
produced CO2 are prerequisites of vertebrate life. A great deal is known about
the diVerences in O2 and CO2 transport and exchange between water and air‐
breathing vertebrates; however, this stems largely from studies on teleost fishes
in the former, and mammals in particular in the latter. Fishes possess great
diversity in gas exchange strategy, ranging from completely water breathing to
obligate air breathing, and thus occupy a crucial phylogenetic position in the
transition of life from water to land which has large implications for gas
exchange (Dejours, 1988; Graham, 1997). Relatively little in relation to O2
and CO2 transport and exchange is preserved within the fossil record, and
consequently, reconstruction of the evolution of gas exchange is limited largely
to studies on extant species. In the following sections, primitive fishes will be
discussed going backward in time from the closest living relatives of teleosts to
successively more distantly related groups of primitive fishes. We first discuss
the relative roles of the respective gas‐exchange surfaces [gills, skin, and air‐
breathing organs (ABOs)] to O2 and CO2 exchange in each primitive fish group.
We then discuss general aspects of O2 and CO2 exchange, largely on the basis of
what is known in teleosts and then what is known for primitive fishes. Finally,
we discuss how this information on primitive fishes helps to identify some
general trends in the evolution of vertebrate blood gas transport characteristics.


     In typical water‐breathing teleosts, the gills are the predominant surface for
both O2 and CO2 exchange; but in some cases, there can be appreciable O2
uptake across the skin. Many of the primitive fish groups discussed in this
chapter contain species that are facultative or obligate air‐breathers. Thus, the
gills, skin, and ABOs are all potential sites for gas exchange in many primitive
5.   GAS TRANSPORT AND EXCHANGE                                               215

fishes. There has been considerable interest and research conducted on the
morphologies of the respective gas exchange structures (see Chapter 4, this
volume) and a great number of direct measurements of the relative role and
eYciency of each structure to both O2 and CO2 exchange, which are briefly
summarized below. In most air‐breathing fishes studied to date, there appears
to be a spatial separation of O2 and CO2 exchange. That is, the majority of O2
uptake may occur across the ABO, and the majority of CO2 excreted across the
gills and/or skin. This is largely related to the fact that the capacitance coeY-
cient for CO2 does not change much between water and air, while that for O2 is
20‐ to 30‐fold higher (depending on the temperature) in air than water (Dejours,
1988). Because ventilation‐rate volume (ventilation frequency  volume) of gas
exchangers in fish is largely regulated to secure adequate O2 uptake, ventilation‐
rate volume of the ABO in an air‐breather is greatly reduced relative to that of
the gills in a water‐breather. The reduced ventilation‐rate volume is suYcient
for O2 uptake, but insuYcient for CO2 elimination across the ABO, and
consequently CO2 diVuses out across the gills and/or skin (Dejours, 1988;
Graham, 1997). The spatial uncoupling of O2 and CO2 transport has interesting
implications for gas exchange in fish, given that at least in most teleost fishes
there is a tight interaction between O2 and CO2 exchange that resides at the level
of Hb in the red blood cell (RBC) (Jensen, 1989; Brauner and Randall, 1996,
1998; Brauner and Val, 1996; Nikinmaa, 2001).
     Air breathing not only permits fishes to survive exposure to aquatic
hypoxia but also allows them to maintain normal levels of metabolism and
activity in aquatic hypoxia, provided O2 taken up in the ABO is not subse-
quently lost across the gills. Consequently, many air‐breathing fishes possess
circulatory adaptations in the gills. Of the four gill arches, eVerent vessels
from the third and fourth arches give rise to the pulmonary artery leading to
the ABO, and venous return from the ABO is direct to the heart. The first and
second gill arches lead exclusively to the dorsal aorta. The creation of a
double circulatory loop is most developed in the obligate air‐breathing lung-
fishes but present to some degree in some of the other groups described below
(reviewed in Graham, 1997). Those primitive fishes that are not air‐breathers
tend to be tolerant of aerial exposure and/or aquatic hypoxia and thus are
very tolerant of adverse environments. The following sections review the
limited information that exists for primitive fishes.

2.1. Primitive Ray‐Finned Fishes (Actinopterygii)
2.1.1. Bowfin (Amiiformes)
   Like many basal teleosts, the bowfin Amia calva uses a swim bladder for
accessory air breathing, with air‐breathing frequency and the fraction of aerial
M O2 increasing with rising temperatures, hypoxia, and exercise (Johansen et al.,
1970; Farmer and Jackson, 1998).
216                                         C. J. BRAUNER AND M. BERENBRINK

    The gills in A. calva are well developed and similar to those of teleost
fishes with a few minor modifications (Daxboeck et al., 1981; Olson, 1981)
(Figure 5.1). At temperatures below 10–15  C in normoxia, A. calva is an
exclusive water‐breather (Johansen et al., 1970; Horn and Riggs, 1973).
Under these conditions, a high arterial PO2 (PaO2) of about 110 mmHg
(measured in a branch of the celiac artery) approaches the PO2 of inspired
water (Table 5.1). This was taken to indicate countercurrent O2 exchange
between water and blood in the gills (Johansen et al., 1970).
    Air‐breathing frequency increases almost linearly with an increase in
temperature above 10  C, to an upper thermal tolerance of 35  C (Horn and
Riggs, 1973). At 20  C, the ABO accounts for about 35% of the total O2
consumed in normoxia, and at 30  C contribution by ABO increases to 75%
at which time A. calva is an obligate air‐breather (Johansen et al., 1970). The
role of the ABO in CO2 excretion is not proportional to the O2 uptake,
accounting for 25% and 40% of total CO2 removal at 20 and 30  C, respec-
tively. Thus, at 30  C, the majority of CO2 is excreted aquatically, presumably
across the gills, although transcutaneous excretion has not been measured
(Johansen et al., 1970).
    During exposure to hypoxia, the dependence on the ABO remains tem-
perature dependent. At 10  C, there is no change in air‐breathing frequency
down to a water PO2 (PwO2) of 70–80 mmHg, and in many specimens, no
changes down to 40 mmHg (Johansen et al., 1970). There is a large increase
in branchial ventilation over this PwO2 range. At a PwO2 of 70–80 mmHg,
the ABO is responsible for about 50% of the total O2 uptake at 20  C, and
almost 100% at 30  C. Johansen et al. (1970) postulated that A. calva possess
shunts in the gills to bypass the lamellae and prevent O2 loss from the
blood to hypoxic water during air breathing. However, after this study,
there has neither been physiological (Randall et al., 1981) nor anatomical
evidence (Daxboeck et al., 1981; Olson, 1981) for shunts of this nature in
the gill. The creation of a double circulatory loop is present to some degree
based on anatomy, but is not as developed as in the obligate air‐breathing
    Activity level also influences the role of the ABO in gas exchange
(Johansen et al., 1970). At a temperature of 24  C in fish at rest, O2 uptake
from the air was 10% of whole animal metabolic rate, the remainder secured
from the water across the gills and/or skin. When fishes were forced to swim
at a low but sustained swimming velocity, there was an increase in air‐
breathing frequency and O2 uptake across the ABO was elevated to 66% of
whole‐animal metabolic rate. Despite the large increase in aerial O2 uptake,
there was also a ca 50% increase in O2 uptake from the water (Farmer and
Jackson, 1998), indicating that there is still some capacity to elevate aquatic
gas exchange in resting fish at this temperature.
5.   GAS TRANSPORT AND EXCHANGE                                                                                                      217


                                      105                                                        lam
                                                                                          s   pe                      os
                                                                                       nu                        Te
                                                               Amia                  wo
                                                                                 tsu   Acipenser
                                                               calva          Ka
            Gill surface area (cm2)

                                            fluviatilis                                                       Latimeria
                                      103                                                                    chalumnae

                                                               u st                       Neoceratodus
                                                            an                               forsteri
                                      102            O    ps



                                      100                            Lepidosiren

                                 10 −1
                                     0.01                      0.1              1                           10                 100
                                                                          Body mass (kg)

Fig. 5.1. Comparison of gill surface area in relation to body weight in primitive fishes. Biloga-
rithmic plot. The shaded area indicates the typical range for teleosts as obtained by extrapolation
of the regression lines for the sluggish toadfish Opsanus tau and the highly active skipjack tuna
Katsuwonus pelamis (Muir and Hughes, 1969; Hughes and Gray, 1972). For primitive fishes,
actual data points and, if applicable, regression lines are given, except for the river lamprey
L. fluviatilis, where minimal and maximal values from Lewis (1980) are shown. The single value
for the Australian lungfish N. forsteri was estimated from Figure 6 in Hughes (1976). Further
references: A. transmontanus (white sturgeon; Burggren et al., 1979), A. calva (bowfin; Daxboeck
et al., 1981), Latimeria chalumnae (coelacanth; Hughes, 1995), Lepidosiren paradoxa (South
American lungfish; de Moraes et al., 2005), and Lepisosteus oculatus (spotted gar; Landolt and
Hill, 1975). Values for the elasmobranch nursehound Scyliorhinus stellaris from Hughes et al.
(1986) are included for comparison. Note the teleost‐like gill surface area of bowfin and river
lamprey, the reduced gill surface area in the coelacanth, and the diVerence between the obligate
and facultative air‐breathing South American and Australian lungfishes, respectively.
                                                                     Table 5.1
                                              Whole‐Blood and Hemoglobin Characteristics of Primitive Fishes

                                               [Hb]     Hct    MCHC
            Species           PaO2    PvO2    (g/dl)    (%)     (g/dl)   P50    nH            F      bHb           Comments                References

      Amia calva                                                         4                                     Whole blood,          Black, 1940
                                                                                                                  15  C, PCO2 0–1
                                                                         11                                    Whole blood,          Black, 1940
                                                                                                                  15  C, PCO2 12
                                                                         15                                    Whole blood,          Black, 1940
                                                                                                                  15  C, PCO2 15
                             110             5.8       22.8   254        9     1.3    À0.43                    Whole blood,          Johansen et al., 1970
                                                                                                                  pH 7.6, 15  C
                                                                         24    2.6    À0.51                    Whole blood,          Johansen et al., 1970
                                                                                                                  pH 7.6, 27  C

                             43.7    31.1*   4.5                                                               Whole blood, pH       Randall et al., 1981
                                                                                                                  7.5, 30  C,
                                                                                                                  *ventral aortic
                             35–59                                                                             Whole blood,          McKenzie et al., 1991a
                                                                                                                  pH 7.60–7.67,
                                                                                                                  20  C
                             47                                                                                Whole blood,          McKenzie et al., 1991b
                                                                                                                  pH 7.72, 20  C
                                             7.5       26.4   284                     À1.0                     Washed RBCs 15       Weber et al., 1976b
                                                                                                                  C, no CO2
                                                                                                   À6.4        25  C                Berenbrink et al., 2005
      Atractosteus                           9.1       29     314                                                                    Siret et al., 1976
      Lepisosteus oculatus   24      20      7.0–8.0                     24           À0. 5                    Whole blood           Smatresk and
                                                                                                                in vitro, 1%           Cameron, 1982a
                                                                                                                CO2, pH 7.8,
                                                                                                                20  C
                                                   30     À0. 5               Whole blood             Smatresk and
                                                                                in vitro, 2%            Cameron, 1982a
                                                                                CO2, pH 7.4,
                                                                                20  C
      Lepisosteus osseus                                               À9*    *Whole blood            Rahn et al., 1971
                                                                                buVer value in
                                                                                mmol HCOÀ    3
                                                                                per liter and pH
                                                                                unit, winter fish,
                                                                                10  C
                                                                       À16*   *Whole blood            Rahn et al., 1971
                                                                                buVer value in
                                                                                mmol HCOÀ    3
                                                                                per liter and pH
                                                                                unit, summer
                                                                                fish, 15  C
                                 9.0                                                                  Lenfant and Johansen,

      Lepisosteus                                         À0.525   *
                                                                       À8.6   25  C, *Haldane        Berenbrink et al., 2005
        platyrhincus                                                             coeYcient
      Acipenser            90    5.0þ              21.5   À0.55               Whole blood             Burggren and Randall,
        transmontanus                                                            in vitro, PO2 3.5,     1978
                                                                                 pH 7.81, 15  C
                           115   5.5    21   262          À0.40               Whole blood             Crocker and Cech,
                                                                                 in vitro,              1998
                                                                                 pH 8.23–7.27,
                                                                                 15  C
                                                          À0.50               Whole blood             Crocker and Cech,
                                                                                 in vitro, pH           1998
                                                                                 8.14–7.20, 20  C
      Acipenser baeri      80*                                                *pH 7.85, 18  C        Maxime et al., 1995
                           70*                                                *pH 7.85, 15  C        Nonnotte et al., 1993
                                 6.9                                                                  O. Kepp and M.
                                                                                                        unpublished data

                                                                 Table 5.1 (continued )

                                           [Hb]     Hct    MCHC
            Species        PaO2   PvO2    (g/dl)    (%)    (g/dl)      P50        nH       F     bHb         Comments                 References

      Acipenser naccarii   72            7.4þ      23.4    316                                           pH 7.91, 23  C        McKenzie et al., 1997
                                                   30                                                                           Clementi et al., 1999
      Acipenser ruthenus                                                                        À9.4     15  C                 Berenbrink et al., 2005
      Polypterus                         4.3–14    17–43   253–334     23.5   2.68     À0.43    À15.4*   Whole blood            Vokac et al., 1972
        senegalus                                                                                           in vitro, PCO2 6,
                                                                                                            pH 7.7, 30  C, F
                                                                                                            calculated from
                                                                                                            pH 7.6–7.3,
                                                                                                            *whole blood
                                                                                                            buVer value in
                                                                                                            mmol HCOÀ    3
                                                                                                            per liter and pH

      Erpetoichthys                      7.5       22      341         17.9   2.0      À0.247            Whole blood            Beitinger et al., 1985
        calabaricus                                                                                         in vitro, PCO2 7,
                                                                                                            pH 7.56, 25  C,
                                                                                                            F calculated
                                                                                                            from pH 7.56–
                                                                                                À11.6    25 C                   Berenbrink et al., 2005
                                         11.9                                                                                   Kepp and Berenbrink,
                                                                                                                                  unpublished data
      Latimeria                          3.4       20      170         3.3    1        À0.51    À9.0*    Whole blood, pH        Wood et al., 1972
        chalumnae                                                                                         7.8, 20 ºC, F
                                                                                                          calculated from
                                                                                                          pH 8.8–6.6,
                                                                                                          *whole blood
                                                                                                          buVer value in
                                                                                                          mmol HCOÀ    3
                                                                                                          per liter and pH
                                         3.7þ                2.1    1                       Whole blood           Hughes and Itazawa,
                                                                                              in vitro, pH 7.3,     1972
                                                                                              15  C
                                                                                    À11.9   Calculated based      Berenbrink, 2006
                                                                                              on number and
                                                                                              location of
      Neoceratodus fosteri   38.9   20   5.5þ   31     177   11            À0.62    À13.3   Whole blood, pH       Lenfant et al., 1966
                                                                                              7.5, PCO2 3.5,
                                                                                              18  C, *whole
                                                                                              blood buVer
                                                                                              value in mmol
                                                                                              HCOÀ per liter
                                                                                              and pH unit
                                         6.0    30     200   22     2.27   À0.48            Whole blood,          Kind et al., 2002
                                                                                              pH 7.5, PCO2 16,
                                                                                              20  C

                                         7.0    35     200                                                        Johansen, 1970
      Lepidosiren            30          4.1þ   15.7   261   10.5          À0.234           Whole blood,          Johansen and Lenfant,
        paradoxa                                                                             PCO2 6, 23  C         1967
                                         6.5    28     232                 À0.295                                 Johansen, 1970
                                         7.0    39.8   176                 À0.31                                  Johansen et al., 1978
                             81     43          21           18.6   2.00   À0.66            Whole blood, 25      Bassi et al., 2005
                                                                                             C, F calculated
                                                                                             from pH 7.62–
                             76     49                       29.8   1.97   À0.44            Whole blood, 35      Bassi et al., 2005
                                                                                             C, F calculated
                                                                                             from pH 7.71–
                                         6.2    33.5   185                                                        Isaacks et al., 1978
                                                                           À0.33            Whole blood           Powers et al., 1979
                                                20                                                                Amin‐Naves et al.,
                                                35                                                                Sanchez et al., 2001
                                                                                    À16.5   25  C                Berenbrink et al., 2005

                                                                   Table 5.1 (continued )

                                            [Hb]     Hct     MCHC
            Species         PaO2   PvO2    (g/dl)    (%)      (g/dl)     P50     nH         F          bHb       Comments             References

      Protopterus           36                                                          À0.28                Whole blood,       Lahiri et al., 1968
        aethiopicus                                                                                           25  C
                                          7.0       25.3     277                        À0.35                Whole blood,       Swan and Hall, 1966
                                                                                                              23  C
                                          6.2       25       248                        À0.47                Whole blood,       Lenfant and Johansen,
                                                                                                              25  C              1968
                                          7.4       27.4     270                                                                DeLaney et al., 1977
                                          7.1       29       245                                                                Jensen et al., 2003
                                                    32.7                                                                        Bartlett, 1978b
      Protopterus                         5.8þ      26.5     219                        À0.20 À0.29          Whole blood from   Johansen et al., 1976b
        amphibius                                                                                             nonestivating
                                                                                                              fish > 2 years
                                          8.7þ      38.3     227                        À0.41 À0.68          Whole blood from   Johansen et al., 1976b

                                                                                                              estivating fish
      Protopterus                         6.9       32.1     215                                                                Babiker, 1979
      Protopterus dolloi                            15.3                                                                        Perry et al., 2005
      Protopterus sp.                     7.8       30.1     259                                                                Johansen, 1970
      Lampetra fluviatilis                                                11.8   1.21                         Adult fish whole    Bird et al., 1976
                                                                                                               blood, 10  C,
                                                                                                               pH 7.75
                                                                         1.8                                 Larvae whole       Bird et al., 1976
                                                                                                               blood, 10  C,
                                                                                                               pH 7.75
                                          6.1       31.4þþ   194                                             2C                Nikinmaa and Weber,
                                          9.5       33.5þþ   284                                             15  C
                                                                                        À1.03                Intracellular F    Nikinmaa et al., 1995
                                                                                        À0.9*         À3.0   *Haldane           Jensen, 1999
      Lampetra               58 À77   24 À40   6.7      21.3     315        $18    1.88   À0.41          À3.5*     Whole blood           Johansen and Lenfant,
        (Entosphenus)                                                                                                 14  C, pH 7.6       1972; Johansen
        tridentata                                                                                                    *Whole blood         et al., 1973
                                                                                                                      buVer value in
                                                                                                                      mmol HCOÀ     3
                                                                                                                      per liter and pH
      Petromyzon marinus     120               6.5þþ    24.6     263                                               Whole blood           Tufts, 1991
                                                                                                                      in vivo, 10  C
                                                                            23.6   1.89                            Whole blood, 10      Ferguson et al., 1992
                                                                                                                      C, PCO2 1.5
                                                                            40.8   1.52                            Whole blood, 10      Ferguson et al., 1992
                                                                                                                      C, PCO2 35
                                                                                          À0.63                    Intracellular F       Ferguson et al., 1992
                                                                                                         À4.78*    *Whole blood          Ferguson et al., 1992
                                                                                                                      buVer value in
                                                                                                                      mmol HCOÀ     3
                                                                                                                      per liter and pH

      Eptatretus cirrhatus   90–110   17.2                                  12.3   1.38   À0.43                    Whole blood           Wells et al., 1986
                                                                                                                    in vitro
                                               3.0      12.6     242                                                                     Wells and Forster, 1989
      Eptatretus stoutii                       2.4þ                         2–4    1.0    0.0                      Whole blood,          Manwell, 1958;
                                                                                                                    in vitro               Johansen and
                                                                                                                                           Lenfant, 1972
      Myxine glutinosa                                                                    À0.35*         À8.2      *Haldane              Jensen, 1999
                                               4.1      19.1     215                                                                     Larsson et al., 1976

          PaO2 and PvO2 refer to the in vivo partial pressure (mmHg) of O2 in arterial and venous blood, respectively, P50 refers to the PO2 (mmHg) at which
      whole blood is 50% saturated, nH refers to the Hill number at 50% hemoglobin saturation, F refers to the Bohr coeYcient, bHb refers to the hemoglobin
      buVer value in organic phosphate‐free, deoxygenated hemolysates at physiological pH and ClÀ and is given in mmol per mmol Hb4 and pH, unless
      otherwise indicated. Comments refer to conditions under which P50, nH, F, and bHb were made unless otherwise indicated by * in that row. þHb
      concentration ([Hb4]) estimated from O2 capacity, þþhematocrit (Hct) or [Hb4] calculated from mean cellular Hb concentration (MCHC) and [Hb4] or
      Hct, respectively.
224                                          C. J. BRAUNER AND M. BERENBRINK

    Immediately after exhaustive exercise, fish that had access to air experi-
enced less of a metabolic acidosis than those denied air access, presumably
due to the increased ability to secure O2 aerially (Gonzalez et al., 2001).
Interestingly, the rate of recovery from acidosis following exhaustive exercise
was slower in fish with than without access to air, which the authors attribute
to a reduced ability to excrete CO2 and Hþ during air breathing due to
inadequate gill ventilation (Gonzalez et al., 2001).
    In general, A. calva utilizes aerial respiration to deal with aquatic hypoxia
at higher temperatures (20–30  C) and to augment O2 uptake during periods
of temperature‐ or activity‐induced elevations in metabolism. The fraction of
total CO2 excreted across the ABO is less than the fraction of total O2 uptake
across this organ, consistent with many other air‐breathing fishes.

2.1.2. Gars (Lepisosteiformes)
    Of the seven species of extant gars found in two genera, most of the
research conducted on gas exchange to date is on Lepisosteus. Like A. calva
and many basal teleosts, gars use a swim bladder for accessory air breathing
(Graham, 1997). Under resting conditions, the spotted gar Lepisosteus
oculatus is generally considered to be a facultative air‐breather. It possesses
reasonably well‐developed gills, although with a slightly reduced mass specific
total gill surface area relative to that observed in studied teleosts (Landolt and
Hill, 1975) (Figure 5.1). An elevation in temperature (Smatresk and Cameron,
1982b), or a reduction in aquatic O2 levels (Smatresk and Cameron, 1982a),
results in an elevation in air‐breathing frequency. At 20–25  C, 70–80% of
total M O2 was obtained from the air and air breathing became obligatory
(Rahn et al., 1971). The gills remained the principal site for CO2 excretion at
all temperatures, with the ABO contributing 0% and 8% to the total CO2
excretion at 13 and 25  C, respectively (Rahn et al., 1971).
    During exposure to severe aquatic hypoxia (PwO2 ¼ 12 mmHg), 100% of
the O2 uptake was across the ABO; however, the majority of CO2 was excreted
to the water (Smatresk and Cameron, 1982a). L. oculatus does not appear to
actively avoid O2‐depleted water, provided it has access to air (Graham, 1997).
    An exercise‐induced elevation in metabolic rate also stimulates air breath-
ing in L. oculatus. During a low‐level exercise‐induced elevation in metabolic
rate, aerial O2 uptake accounted for 53% of total uptake (Farmer and Jackson,
1998). During forced exhaustive exercise, the animals continuously breathed
air during the exercise bout (30–40 min) and during recovery (Burleson
et al., 1998). Consequently, exposure to water with PwO2 < 19 mmHg had a
minimal eVect relative to aquatic normoxia on the duration over which the fish
could sustain the swimming bout, or the pattern of recovery following
exhaustive exercise, provided the animals had unlimited access to air.
5.   GAS TRANSPORT AND EXCHANGE                                               225

Recovery from exercise was associated with relatively minor changes in gill
ventilation, but a large elevation in air‐breathing frequency relative to preswim
levels further emphasizing the reliance on air‐breathing in this species during
periods of elevated metabolism.

2.1.3. Sturgeons and Paddlefishes (Acipenseriformes)
     Sturgeons and paddlefishes are strictly water‐breathers, unlike many of
the other primitive fishes, and the gills possess filaments with relatively well‐
developed lamellae that are similar to those of teleost fishes. Paddlefishes are
filter feeders, and juvenile Polyodon spathula are obligate ram‐ventilators in
normoxia and swim continuously shortly following hatch and throughout
life (Burggren and Bemis, 1992). While they have the ability to ventilate their
gills at low swimming speeds (<0.6–0.8 body lengths sÀ1), juveniles are not
tolerant of hypoxia and die at a PwO2 < 90 mmHg (Burggren and Bemis,
1992). The paddlefishes have reduced scales, and thus transcutaneous gas
transfer may occur; however, this has not been measured directly.
     The open swim bladder in Acipenseriformes is not used as an accessory
ABO (Graham, 1997) and probably only participates in buoyancy regulation.
In the white sturgeon, Acipenser transmontanus, the total gill surface area
relative to body weight is at the lower margin of the values in typical teleosts
(Figure 5.1), which is thought to be related to a low activity level and meta-
bolic rate (Burggren et al., 1979). The eYciency of the gills for gas exchange
in the genus Acipenser is demonstrated by high‐average normoxic PaO2 values
ranging between 70 and 110 mmHg at temperatures between 23 and 15  C,
depending on species (Table 5.1) (Burggren, 1978; Nonnotte et al., 1993;
Maxime et al., 1995; McKenzie et al., 1997; Crocker and Cech, 1998). Average
dorsal aortic blood pH was between 7.73 and 7.91 in these studies. At these
PO2 and pH levels, dorsal aortic hemoglobin‐O2 (HbO2) saturations above
90% can be calculated from in vitro O2 equilibrium curves of whole blood from
A. transmontanus (Burggren and Randall, 1978; Crocker and Cech, 1998).
A unique feature in this species, and probably also in other members of this
genus, is the ability to eYciently ventilate the gills with water drawn in through
the upper part of the opercular slits when the retractable, tubelike mouth is
blocked, as may occur during feeding (Burggren, 1978).
     In sturgeons, up to 10% of O2 uptake may occur across the skin (Burggren
et al., 1979). Although the sturgeons are water‐breathers, anecdotally they
have been reported to be extremely tolerant of aerial exposure. Whether this is
associated with transcutaneous gas exchange or the possession of lamellae
that do not collapse and are still functional in air is not known.
     In addition to aerial exposure, white sturgeons are very tolerant of high
aquatic CO2 levels (Crocker and Cech, 1998; D. Baker and C. J. Brauner,
226                                          C. J. BRAUNER AND M. BERENBRINK

unpublished data), and sturgeons in general are very tolerant of aquatic
hypoxia [Acipenser baeri (Nonnotte et al., 1993; Maxime et al., 1995, 1998),
Acipenser oxyrinchus and Acipenser brevirostrum (Baker et al., 2005), and
A. transmontanus (Burggren and Randall, 1978)]. Burggren and Randall
(1978) reported that white sturgeons were oxyconformers during exposure
to short‐term hypoxia, exhibiting 50%, 15%, and 5% of control M O2 during
exposure to a PwO2 of 100, 60, and 30 mmHg, respectively. However, in
studies on the Siberian (Acipenser baeri) and the Adriatic sturgeon (Acipenser
naccarii), no oxyconforming behavior was observed, indicating that these
species were oxyregulators (Randall et al., 1992; Nonnotte et al., 1993;
McKenzie et al., 1997), like most other fishes studied to date.

2.1.4. Reedfish and Bichirs (Polypteriformes)
    In the gray bichir, Polypterus senegalus, and the reedfish, Erpetoichthys
calabaricus, the ABO is quite well developed (Magid, 1967; Lechleuthner
et al., 1989). However, under normoxia all O2 uptake occurs from the water
(Babiker, 1984b; Pettit and Beitinger, 1985). In E. calabaricus, transcutane-
ous O2 uptake accounts for up to 32% of whole‐animal metabolic rate (Sacca
and Burggren, 1982). During elevated activity levels or exposure to hypoxia,
aerial respiration increases in both species indicating that they are facultative
air‐breathers (Babiker, 1984b; Pettit and Beitinger, 1985). In P. senegalus
smaller than 25 g, the lungs are not well developed and aquatic hypoxia does
not result in air breathing. At this stage of development, a water O2 level of
1.8 mg literÀ1 results in 100% mortality (Babiker, 1984b). With an increase in
body mass up to about 100 g, there is a progressive, almost linear increase in
percentage of total O2 consumed that is secured from the ABO (up to 80%)
during exposure to aquatic hypoxia (2.8 mg literÀ1 O2; calculated PwO2 of
57 mmHg). At 100 g, survival was 100% in animals with free access to air
during exposure to low aquatic O2 (1.8 mg literÀ1; calculated PwO2
of 37 mmHg). In adults, air‐breathing frequency increases linearly with a
reduction in water O2 levels down to 2.4 mg literÀ1 (calculated PwO2 of
49 mmHg) or less, at which time 95% of total M O2 is secured across the
ABO (Babiker, 1984b). Both P. senegalus and E. calabaricus survive low
aquatic PO2 values through air breathing, and do not strongly avoid O2
poor waters. E. calabaricus makes voluntary excursions onto land and is
tolerant of aerial exposure for several hours without negative eVects (Sacca
and Burggren, 1982; Pettit and Beitinger, 1985). However, there are contra-
dictory reports on whether this is associated with an increase or decrease
in M O2 (Sacca and Burggren, 1982; Pettit and Beitinger, 1985).
    The creation of a double circulatory loop is present to some degree
based on anatomy, but is not as developed as in the obligate air‐breathing
5.   GAS TRANSPORT AND EXCHANGE                                                227

2.2. Lobe‐Finned Fishes (Sarcopterygii)

2.2.1. Coelacanths (Coelacanthiformes)
    Superficially, the general gill structure in Latimeria chalumnae is similar
to that of teleosts. However, it has been hypothesized that their O2 extraction
eYciency may be quite low (Hughes, 1976). The lamellae are less well
developed, and mass specific total gill surface area is reduced relative to
other strictly water‐breathing fishes (Figure 5.1). However, this is a general
characteristic of deep dwelling fishes and may not reflect the ancestral condi-
tion of lobe‐finned fishes. Metabolic rate has never been measured directly;
but on the basis of the low mass specific total gill surface area, metabolic rate
has been calculated to be relatively low (i.e., 10 ml kgÀ1 hÀ1) (Hughes, 1976)
in comparison with teleost fishes. It is likely that the gills are the predominant
sites for O2 and CO2 exchange. However, nothing is known of the potential
for transcutaneous gas exchange.
2.2.2. Lungfishes (Dipnoi)
    The role of the lungs, gills, and skin in O2 and CO2 exchange in lungfishes
has been studied extensively in the past century, largely with the goal to gain
insight into how aerial respiration may have arisen in ancestral tetrapods. There
are a number of excellent reviews on the topic to which the reader is referred (see
Johansen, 1970; Burggren and Johansen, 1986; Graham, 1997). Of the three
extant lungfish genera, the Australian lungfish (Neoceratodus forsteri) is a
facultative air‐breather and its physiology diVers markedly from the obligate
air‐breathing lungfishes of South America (Lepidosiren paradoxa) and Africa
(Protopterus species). In the two most extensively studied species of obligate air‐
breathing lungfishes, L. paradoxa and P. aethiopicus, the lung has long been
thought to be the dominant site for O2 uptake and the gills and/or skin has
been the dominant site for CO2 excretion (Johansen and Lenfant, 1967; Lenfant
and Johansen, 1967, 1968; McMahon, 1970; Burggren and Johansen, 1986).
In adults of these species, aerial O2 uptake satisfies 90–97% of the animal’s
O2 requirements, and CO2 excretion to the aquatic environment comprises
40–77% of total CO2 excretion.
    The circulatory system of these obligate air‐breathing lungfishes has
received a great deal of attention due to the degree of separation between
pulmonary and systemic blood flow. The eVerent blood vessels from gill
arches 3 and 4 give rise to the pulmonary artery leading to the lung, and
the pulmonary venous return is delivered directly to a functionally divided
heart (see Chapter 2, this volume). Oxygenated blood returning from the
lung is then directed toward gill arches 1 and 2, which are largely devoid of
respiratory lamellae, and this prevents O2 loss to the water (see Graham,
1997 for a review). While it has long been assumed that the lamellae on
228                                          C. J. BRAUNER AND M. BERENBRINK

arches 3 and 4 are retained for CO2 excretion, there have been few studies
that have tried to delineate the relative role of the gills and skin in CO2
excretion. In L. paradoxa, a detailed morphological analysis revealed that
99.15% of the physiological diVusing capacity lies in the lungs, 0.85% in the
skin, and only 0.0013% in the gills (de Moraes et al., 2005). This is reflected in
the extremely reduced gill surface area shown in Figure 5.1. From these data,
the authors conclude that gills are virtually without importance as a gas
exchange organ. Despite the relatively small diVusing capacity of the skin,
this route could still account for a large proportion of CO2 excretion (de
Moraes et al., 2005). Clearly, physiological studies are required to delineate
the relative roles of the gills and skin to CO2 elimination in these obligate
air‐breathing lungfishes.
    In L. paradoxa, temperature has a large eVect on the role of the lung
in CO2 excretion. A 20  C increase in temperature (from 15 to 35  C),
resulted in a 15‐fold elevation in metabolic rate, and CO2 excretion across
the lung increased from less than 15% to about 70% of total CO2 excreted
(Amin‐Naves et al., 2004). Thus, with an elevation in metabolic rate, the lung
may take on a greater role in CO2 excretion; however, this was associated
with a respiratory acidosis where PaCO2 increased from 12 to 35 mmHg, and
pHa decreased from 7.6 to 7.35 (Amin‐Naves et al., 2004).
    In a study on an obligate air‐breathing species of African lungfish,
P. dolloi, the lung plays the dominant role in both O2 and CO2 exchange
(91% and 76%, respectively; Perry et al., 2005). During exposure to aerial and
aquatic hypercapnia, only the former induced a respiratory acidosis. The
apparent lack of CO2 uptake across the gills during hypercapnia could be
due to a limited role of the gills for CO2 exchange, or, alternatively, CO2
excretion across the lung could be suYcient to excrete the CO2 taken up
across the gills in this species. Thus, the textbook description of the relative
role of the lungs and gills to obligate air‐breathing lungfishes in terms of O2
and CO2 exchange is not as clear as once thought, and is clearly in need of
further detailed investigation.
    The Australian lungfish, N. forsteri, is a facultative air‐breather and
obligate water‐breather incapable of surviving prolonged aerial exposure,
unlike P. aethiopicus and L. paradoxa. Gill surface area is slightly reduced
compared to teleosts (Figure 5.1), but aerial respiration can augment O2
uptake during elevated metabolic rate associated with activity or during
exposure to low water PO2 . Interestingly, during exposure to aquatic hypoxia,
N. forsteri responds with an immediate, large, and sustained increase in gill
ventilation, which diVers markedly from the obligate air‐breathing lungfishes
P. aethiopicus and L. paradoxa, which exhibit no change in branchial venti-
lation, consistent with the limited role of gills for O2 uptake in the latter
(Burggren and Johansen, 1986; Sanchez et al., 2001). The increase in gill
ventilation during hypoxia in N. forsteri is maintained during hypoxia even
5.   GAS TRANSPORT AND EXCHANGE                                                229

following the onset of air breathing (Fritsche et al., 1993; Kind et al., 2002).
In P. aethiopicus, P. amphibius, and P. annectens, juvenile stages are more
dependent on aquatic O2 uptake than adults (Johansen et al., 1976a; Babiker,
1979; Burggren and Johansen, 1986; Seifert and Chapman, 2006), and
juveniles respond to hypoxic water by elevating both gill ventilation and
air‐breathing frequency. The greater dependence on aquatic O2 uptake in
juvenile animals is likely a reflection of a less‐developed lung (Babiker, 1979).
Clearly, there are interesting developmental changes that occur within these
obligate air‐breathing lungfishes that remain to be investigated.
    All lungfishes are able to tolerate aquatic hypoxia through the use of air
breathing, as is the case for many other primitive fishes. However, both
L. paradoxa and the species of Protopterus possess yet another line of defense
against unfavorable environments, in that they can estivate. In estivating
animals, the metabolic rate is greatly reduced (to 1–20% of resting rate in
water) and the animals become ureotelic (Fishman et al., 1986; reviewed by
Graham, 1997). Thus, in addition to being virtually unaVected by poorly
oxygenated waters through breathing air, they tolerate desiccation of the
environment for long periods of time.

2.3. Jawless Fishes (Agnatha)

2.3.1. Lampreys (Petromyzontiformes)
    Both larval and adult stages of the lamprey Geotria australis are capable of
surviving out of water in humidified air for up to at least 4 days (Potter et al.,
1996b) without apparent negative eVects. During this time, metabolic rate is
similar to that measured in G. australis held in water, and most of the O2
uptake during air exposure is cutaneous (Potter et al., 1996b). However, in
air‐exposed adult G. australis, the gills are responsible for 87% of O2 uptake
and 80% of CO2 excretion, indicating that the gills in adults retain their
integrity in air (Potter et al., 1997). Lampreys, such as G. australis, are exposed
to air when they leave the water to overcome barriers during their upstream
migration (Potter et al., 1996a), and thus aerial exposure may be common.
Total gill surface area to body weight in adult river lamprey, Lampetra
fluviatilis, is similar to that of active teleosts (Figure 5.1) (Lewis and Potter,
1976), and metabolic rate in resting and exercising Petromyzon marinus
appear to be in the order of that of teleosts (Beamish, 1973). The gills are
the predominant site of O2 uptake, and presumably CO2 excretion, in water.
2.3.2. Hagfishes (Myxiniformes)
    Hagfishes (Eptatretus cirrhatus and E. stoutii) have a metabolic rate that is
considerably lower than in teleost fishes of similar mass and temperature, and
is among the lowest recorded among vertebrates (Munz and Morris, 1965;
Forster, 1990). The skin in hagfishes may play a significant role in O2 uptake,
230                                         C. J. BRAUNER AND M. BERENBRINK

and, on the basis of calculations by SteVensen et al. (1984) for the Atlantic
hagfish (Myxine glutinosa), may be responsible for up to 80% of whole‐animal
metabolic rate! Dermal capillaries have been identified in several species of
hagfishes (Potter et al., 1995). During feeding, this may be important when
the head is buried in the prey, preventing ventilation through the nostril.


    The general blood O2 transport characteristics of fishes have been
reviewed in detail in a previous volume of the Fish Physiology series, includ-
ing comprehensive chapters on Hb structure and function (Jensen et al.,
1998a), general O2 transport (Nikinmaa and Salama, 1998), hematocrit and
blood O2‐carrying capacity (Gallaugher and Farrell, 1998), the physiology of
the Root eVect (Pelster and Randall, 1998), and the linkage between O2 and
CO2 transport (Brauner and Randall, 1998). These chapters extensively dealt
with teleost fishes and also covered elasmobranchs and agnathans. However,
the basal members of the ray‐finned fishes and lobe‐finned fishes were usually
only treated in passing, if at all. Here we first recapitulate general aspects of
vertebrate blood O2 transport and then review the fragmentary information
on primitive fishes.

3.1. General Principles of Hb Function
    Reversible binding of O2 to Hb inside RBCs is crucial to the uptake,
transport, and delivery of O2 in whole blood of nearly all vertebrates. Briefly,
O2 reversibly binds to the Fe atom in a heme group, which itself is attached
to a globin polypeptide chain. Hbs of jawed vertebrates consist of a‐type and
b‐type globins, which align to form tetramers consisting of two ab‐dimers
and thus contain four O2‐binding sites.
    In most cases, O2 binding by the tetrameric Hb of jawed vertebrates can
be satisfactorily explained by the two‐state allosteric model, whereby Hb is
in equilibrium between a low‐O2 aYnity, T(ense)‐state conformation, and a
high‐O2 aYnity, R(elaxed)‐state conformation. Changes in the O2‐binding
properties of Hb can be explained by a number of factors interacting diVer-
entially with the two conformations, thereby stabilizing one form over the
other. Hb in the jawless Agnatha contains only one general type of globin
chain, which is in monomer–oligomer equilibrium and has a higher O2 aYnity
in the monomeric than in the dimeric or tetrameric aggregation state. Again,
factors aVecting the equilibrium between aggregation states determine the
whole‐blood O2‐binding properties.
5.   GAS TRANSPORT AND EXCHANGE                                             231

    The Hb‐O2‐binding characteristics are generally believed to be fine‐tuned
to meet the variable O2 requirements of an organism. Increased metabolic
demand for O2 due to, for example, exercise, digestion, or an increased body
temperature, can principally be met by an increased cardiac output and/or an
increased arteriovenous O2 diVerence. In the following sections, the latter will
be discussed in this respect. With the exception of Hb‐less icefishes, changes
in the amount of physically dissolved O2 between arterial and venous blood
of vertebrates are usually negligible when compared to the changes in
Hb‐bound O2. Therefore, the arteriovenous O2 diVerence is mainly a func-
tion of the diVerence in Hb‐O2 saturation between arterial and venous
blood and blood Hb concentration.

3.2. Factors AVecting the ArterioVenous O2 DiVerence
    The amount of O2 that can be maximally taken up by the blood in
addition to physically dissolved O2 is a direct function of the functional
blood Hb concentration and is termed the O2 capacity. Several early studies
did not correct their O2 capacity measurement for physically dissolved O2,
which is insignificant at high Hb concentrations, but can make up a large
fraction of total O2 at low Hb concentration, high PO2 , or low temperature.
There is generally a strong positive correlation between blood Hb concentra-
tion and hematocrit, both within and across species (Graham, 1997;
Gallaugher and Farrell, 1998). This supports the early finding of a relatively
constant Hb concentration inside RBCs across vertebrates (Wintrobe, 1934).
    Table 5.1 gives values for blood Hb concentration, hematocrit, and mean
cellular Hb concentration in a number of primitive fishes. As can be seen,
blood Hb concentration can be extremely variable even within the same
species, making it diYcult to identify adaptive trends in this parameter
when comparing species in diVerent environments or with diVerent lifestyles
(Gallaugher and Farrell, 1998). During exercise or hypoxia, a short‐term
response of several teleosts is to release additional RBCs form their spleen
and thereby increase blood Hb concentration; a more long‐term response
involves increased production of new RBCs. With few exceptions, hypoxia
induces erythropoiesis also in teleost fishes as in other vertebrates
(Gallaugher and Farrell, 1998).
3.2.2. The O 2 Equilibrium Curve and Its Modulators
   Outside extremely low and high O2 saturations, the shape of the O2
equilibrium curve can be satisfactorily characterized by the overall aYnity
and cooperativity of Hb‐O2 binding. A useful, inverse measure for Hb‐O2
232                                         C. J. BRAUNER AND M. BERENBRINK

aYnity is the P50, the PO2 required for half‐maximal saturation of Hb
with O2. The cooperativity of Hb‐O2 binding determines the steepness of
the O2 equilibrium curve in the vicinity of the P50. Cooperativity is expressed
as the Hill‐coeYcient, nH, which is usually determined at 50% O2 saturation
and is obtained as the slope of the line that results when the O2 equilibrium
curve data are transformed such that the log of the oxyHb to deoxyHb ratio
is plotted against log PO2 . Values for nH around unity signify noncooperative,
hyperbolic O2 equilibrium curves, whereas nH values increasing above unity
indicate increasing degrees of cooperativity. In fish Hbs that display a Root
eVect, nH values below unity are frequently observed at low pH and are
thought to reflect heterogeneity in subunit O2 aYnity.

    a. Hb Multiplicity. In the simplest case, Hb‐O2 saturation at any one PO2
can be calculated from the prevailing P50 and nH value. However, many fish
species, primitive and modern, contain multiple Hbs in their blood, which
often diVer significantly in aYnity and cooperativity. This can result in
undulatory O2 equilibrium curves because at increasing PO2 values, the Hb
fraction with the highest O2 aYnity is saturated significantly before the others,
resulting in a flattening of the overall curve at intermediate PO2 values
(Maginniss et al., 1980). Whenever O2‐binding sites diVer in O2 aYnity, either
as a result of diVerent Hb isoforms or because of subunit heterogeneity, the
apparent nH values are reduced as a mathematical consequence, sometimes
below unity.

    b. Evolutionary and Ontogenetic Changes. Even with only one major
Hb component, the shape of the O2 equilibrium curve is quite plastic and
influenced by several factors. Over evolutionary timescales, genetic changes in
the amino acid sequence of the globin chains can aVect the intrinsic O2 aYnity
and cooperativity. Further, over the life span of individuals, ontogenetic
changes in the expression of genetically diVerent Hbs have been described in
virtually all major vertebrate groups such that embryos, juveniles, and adults
often express diVerent, life stage‐specific Hbs (Nikinmaa, 1990).

   c. Organic Phosphates. Over even shorter timescales, phenotypic
changes in the shape of the O2 equilibrium curve of jawed vertebrates may
be brought about by species‐specific organic phosphate molecules, which
bind allosterically to the b‐ chains of deoxygenated Hb and thereby reduce
Hb‐O2 aYnity. In contrast to mammalian and avian RBCs, which contain
high concentrations of 2,3‐diphosphoglycerate (2,3‐DPG or 2,3‐BPG)
and inositol‐pentaphosphate (IP5), respectively, the principal organic phos-
phates of most fishes are ATP and GTP, in varying amounts. The RBC
concentrations of ATP and GTP are reduced during hypoxic acclimation or
5.   GAS TRANSPORT AND EXCHANGE                                           233

on RBC swelling, leading to a higher Hb‐O2 aYnity under these conditions
(Val, 2000).

    d. Temperature. Due to the exothermic nature of Hb‐O2 binding,
increased temperatures usually decrease Hb‐O2 aYnity. This physicochemi-
cal property of Hb will necessarily cause a shift in O2 aYnity of blood
between metabolically active, heat‐producing tissues and relatively cooler,
water‐exposed respiratory organs. O2 aYnity will also vary with daily and
seasonal temperature changes and with the evolution of regional or complete
endothermy, unless compensated by phenotypic or genetic mechanisms.

   e. pH. Hb‐O2 aYnity is also profoundly aVected by pH and this eVect
dynamically changes O2‐binding properties in the time frame of seconds when
blood circulates between arterial and venous capillaries because of the meta-
bolic acids produced and released by the tissues. In the physiological pH
range, a decrease in pH lowers O2 aYnity and consequently increases P50,
causing a right‐shift of the O2 equilibrium curve. This is the so‐called Bohr
eVect, various aspects of which have been reviewed, including a series of
papers celebrating the centennial of the first description of the phenomenon
by Bohr et al. (1904; Giardina et al., 2004; Jensen, 2004; Berenbrink, 2006).
   The magnitude of the Bohr eVect is often quantified by the Bohr coeYcient
F, which gives the change in log P50 on a unit change in pH:

                              D log P50    1
                         f¼             ¼ À DzHþ                          ð1Þ
                              D log pH     4

    Alternatively, F can be estimated from acid–base titrations of Hb as one‐
fourth of the number of protons that are released from tetrameric Hb when it
changes from being completely deoxygenated to being fully oxygenated at
constant pH (Haldane eVect), as also shown in Eq. (1). Strictly, the two
expressions for the Bohr eVect are only equivalent when the shape of the O2
equilibrium curve is symmetrical (otherwise the log of the median PO2 has to
be used), and in the absence of other eVectors that may interact diVerentially
with oxygenated and deoxygenated Hb (including CO2, and the anion AÀ of
the acid AH that is used to change pH) (Wyman, 1964). In whole blood
in vivo, in the presence of multiple Hbs or organic phosphate modulators,
these conditions are rarely achieved.
    The source of the so‐called Bohr protons [DzHþ in equation (1)], and
thereby also the molecular mechanism of the Bohr eVect, is quite diVerent in
Hbs of jawed vertebrates and agnathans, and increasing evidence suggests
that the mechanism also diVers within the jawed vertebrates (Berenbrink,
234                                         C. J. BRAUNER AND M. BERENBRINK

     The Bohr eVect is advantageous in increasing O2 unloading by a right‐
shift of the O2 equilibrium curve in acidic tissues and enhancing O2
loading by a left‐shift at the respiratory organs, where CO2 is released
and pH increases. Its ultimate benefit is more accurately described by the
extent to which it increases the arteriovenous O2 diVerence, which does
not only depend on the shift of logP50 with pH (i.e., F) but also on the
respective arterial and venous PO2 values and nH (Bartels, 1972). But even
when these parameters are identical, the extent to which a given acid load
in the tissues aVects the arteriovenous O2 diVerence in two species may
be quite diVerent. This is due to species‐specific diVerences in (1) the transfer
mechanisms of extracellular acid–base loads to the interior of the RBCs,
where protons ultimately act on Hb‐O2 aYnity; (2) the net charge of Hb,
which aVects RBC pH at a given extracellular pH; and (3) the degree
of intracellular proton buVering, mainly by Hb itself (Sections 3.2.2.g and
     Many fishes show an extreme reduction of Hb‐O2 aYnity at low‐pH
values, which results in whole blood not being fully saturated with O2 even
when equilibrated with air (Root, 1931). In some cases, the blood at low pH
cannot even be fully saturated when equilibrated with pure O2 at pressures of
several atmospheres. This has been ascribed to a decrease in Hb‐O2 capacity
and is known as the Root eVect (Scholander and Van Dam, 1954). However,
it is somewhat arbitrary to define a fixed PO2 above which failure to become
fully saturated constitutes a Root eVect and below which that failure
constitutes a Bohr eVect. Clearly, on the basis of the saturation criterion
at fixed PO2 alone, the distinction between a strong Bohr eVect and a small
Root eVect is diYcult (Farmer et al., 1979), as shown by the debate about
whether the amphibian Xenopus laevis possesses a Root eVect (Perutz and
Brunori, 1982; Bridges et al., 1985; Kister et al., 1989; Berenbrink et al.,
2005). Dilute hemolysates usually have a much higher O2 aYnity than whole
blood because of the diminished interaction between organic phosphates and
Hb. Thus, the failure of air‐equilibrated hemolysates to become completely
saturated at low pH and physiological temperature is often accepted as
indicating the presence of a Root eVect, and the percentage of Hb remaining
deoxygenated under these conditions is used to quantify the eVect (Farmer
et al., 1979; Berenbrink et al., 2005). In contrast to the Bohr eVect, many
Root eVect Hbs show a conspicuous decrease in nH at low pH, and this
property has been proposed to diVerentiate between the two eVects (Perutz
and Brunori, 1982).
     In teleosts and the bowfin, the presence of a Root eVect is associated
with the presence of specialized vascular countercurrent exchangers in the eye
or swim bladder (choroid and swim bladder retia mirabilia, singular: rete
mirabile; Farmer et al., 1979; Berenbrink et al., 2005). The Root eVect allows
5.   GAS TRANSPORT AND EXCHANGE                                               235

the acid‐induced oZoading of Hb‐bound O2 even at high PO2 values.
Together with countercurrent multiplication by the retia mirabilia, this is
used to fill the swim bladder with O2 against high hydrostatic pressures for
buoyancy regulation or to sustain the high metabolic demand of the poorly
vascularized retina of many fishes (Pelster and Randall, 1998; Waser and
Heisler, 2005). Theoretically, a Root eVect may be used to increase the PO2 in
other tissues. However, as there are no reports of gas‐exchanging vascular
retia mirabilia outside the eye and swim bladder, which could act as counter-
current multipliers, the degree of PO2 elevation is necessarily limited. With a
blood O2 solubility coeYcient of 2 mmol literÀ1 mmHgÀ1, a typical tetrameric
blood Hb concentration of 1 mmol literÀ1, and a maximal Root eVect of
50%, fully oxygenated blood that is acidified in a closed system will show a
theoretical PO2 increase of 1000 mmHg. Although in vivo the assumption of
a closed system is practically never fulfilled, especially so in the absence of the
barrier function of a rete mirabile, a Root eVect may be the cause behind PO2
values slightly above air saturation in postbranchial blood of exercised
striped bass (Nikinmaa et al., 1984) and higher than expected PO2 values in
red muscle tissues of rainbow trout (McKenzie et al., 2004). However, it is
currently unclear whether blood in red muscle is suYciently acidotic, even
transiently, for a significant Root eVect to occur. Clearly more research is
needed before a physiological function of the Root eVect outside the eye and
swim bladder can be ascertained.

    f. CO2. A large part of the eVect of elevated blood PCO2 on Hb‐O2
transport is caused by the pH change generated by the hydration of CO2 to
carbonic acid and its subsequent dissociation to bicarbonate and protons.
However, as in mammals, a specific eVect of CO2 has also been reported for
the Hb‐O2 aYnity of some fishes (Farmer, 1979). Again as in mammals, CO2
is thought to bind preferentially to the N‐terminal amino groups of deoxy-
genated Hb as a carbamino compound, thereby decreasing Hb‐O2 aYnity.
The eVect is stronger at high pH values because CO2 reacts with the unpro-
tonated N‐terminal amino groups, whereby the Bohr eVect may be partially
oVset in the presence of CO2. However, the eVect of CO2 is not as strong in
teleosts as in mammals, most likely because of the blockage of some of the
N‐terminal amino groups in teleosts (Jensen et al., 1998a). It is unclear
whether CO2 competes with organic phosphates in binding to deoxygenated
fish Hbs, as shown for human Hb (Kilmartin and Rossi‐Bernardi, 1973), and
whether CO2 will have an appreciable eVect in the presence of physiological
ATP and GTP concentrations. Purely water‐breathing fishes are generally
characterized by lower blood PCO2 values than air‐breathers (Rahn, 1966),
questioning the physiological role of CO2 in modulating Hb‐O2 aYnity in
these fishes (but see hagfishes below).
236                                        C. J. BRAUNER AND M. BERENBRINK

    g. RBC pH Regulation. Possession of even the strongest Bohr eVect Hb
would be physiologically meaningless if the changes in plasma acidity that
occur between tissues and respiratory surfaces were not transferred from the
plasma to the RBC interior. In all jawed vertebrates studied so far, a power-
ful anion exchange protein (AE1, band 3) rapidly equilibrates acid–base
equivalents in form of the bicarbonate anion across the RBC membrane
(Nikinmaa and Salama, 1998; Jensen, 2004). Whether blood CO2 increases
by respiratory acidosis or whether it is generated during a metabolic acidosis
when increased proton concentrations in the plasma lead to the formation of
CO2 from plasma bicarbonate, CO2 readily diVuses across the RBC mem-
brane into the RBC. Here the equilibrium of the CO2 hydration reaction is
shifted toward the formation of bicarbonate and protons. RBC bicarbonate
is exchanged for plasma chloride and leaves the cell, whereas the proton
acidifies the RBC interior. The net eVect is an entry of Hþ and ClÀ (the
dissociation products of HCl) by which extracellular pH changes are directly
transmitted to the RBC cytosol.
    Sodium and potassium, the main intra‐ and extracellular cations, are
regulated at a rather constant level, and in the absence of acid–base relevant
ion transporters other than AE1, the passive equilibrium distribution of
protons across the RBC membrane largely depends on the net charge of
membrane impermeable anions, mainly Hb and plasma proteins (Nikinmaa,
1990; Nikinmaa and Salama, 1998). The much higher concentration of
intracellular Hb compared to plasma proteins together with its negative
charge causes a higher intracellular than extracellular proton concentration.
Across the physiological plasma pH range, intracellular pH is therefore
distinctly lower than plasma pH. The transmembrane pH diVerence is directly
influenced by changes in Hb charge, such that it decreases on deoxygenation
when the binding of Bohr protons neutralizes negative charges on Hb.
Titration of negative charges on intracellular Hb at low pH also causes the
transmembrane pH diVerence to diminish. Therefore, a unit decrease in
plasma pH causes a somewhat smaller decrease in RBC pH so that the
Bohr coeYcient F in studies on whole blood or isolated RBCs needs to be
related to the intracellular pH when it is compared with studies on Hb in
    Due to their negatively charged phosphate groups, organic phosphates
contribute to the number of impermeable intracellular negative charges, and
their concentration therefore also aVects the transmembrane Hþ equilibrium
of RBCs. Thus, a decrease in intracellular organic phosphates will decrease
the transmembrane pH diVerence (Nikinmaa and Salama, 1998).
    In many teleost species and lampreys, secondarily active transport of
protons elevates intracellular pH above its equilibrium value (Nikinmaa
and Huestis, 1984; Nikinmaa, 1986; Berenbrink and Bridges, 1994a,b).
5.   GAS TRANSPORT AND EXCHANGE                                            237

In lampreys, RBC pH is continually elevated above the Hþ equilibrium
distribution by a Naþ/Hþ exchanger (NHE), increasing Hb‐O2 aYnity
because of the strong Bohr eVect of lamprey Hb (described below).
In teleosts, a similar mechanism is activated by catecholamines, whose plasma
levels increase during exercise stress or hypoxia. As in lampreys, the ensuing
increase of RBC pH via the adrenergic NHE increases Hb‐O2 aYnity in
teleosts, which often have a very large Bohr eVect in addition to a Root eVect.
    The method of blood sampling in many early studies very likely caused
elevated plasma catecholamine levels. If the species under investigation pos-
sessed an adrenergic NHE, the reported whole‐blood O2‐binding properties
clearly would not represent resting conditions.
    In Atlantic cod RBCs, hypercapnic acidosis activates another secondarily
active, pH regulatory transporter, which has been characterized as sodium‐
dependent chloride/bicarbonate exchange. This transport mechanism elevates
intracellular pH and will thereby also aVect Hb‐O2 binding (Berenbrink and
Bridges, 1994a).

    h. Specific Hb BuVer Capacity. The extent to which a given acid load
entering the RBC aVects intracellular pH and thereby the Bohr and Root
eVect critically depends on the specific buVering capacity of Hb (bHb). Hb is
the main nonbicarbonate blood buVer and bHb is the molar concentration of
acid–base equivalents that is required to change the pH of an Hb solution
of a given molar concentration (in the absence of other buVers) by one unit.
It is determined by acid–base titration at constant oxygenation status and
yields negative values because addition of acid decreases the pH. In the
following sections, values for bHb refer to measurements on deoxygenated,
tetrameric Hb at physiological temperature, chloride concentration, and

3.2.3. Arterial andVenous PO2
    Hb‐O2 saturation is principally determined by the shape of the prevailing
Hb‐O2 equilibrium curve and PO2 . High PaO2 values ensure full blood O2
saturation in respiratory organs. Generally, PaO2 in gills and lungs is a
function of the PO2 in inspired water or air and therefore environmental
PO2 . It is aVected by the extent and type of ventilatory movements and the
general anatomy of the gas exchange organs, such as countercurrent arrange-
ment of gill blood and water flow (Chapter 4, this volume). As O2 transfer
from water or air to blood ultimately occurs by diVusion, maximal M O2 is
limited by the respiratory surface area of gills and lungs and the length of
the water to blood diVusion path.
    Strictly water‐breathing teleosts fall into two broad groups regarding the
level of resting PaO2. Contrary to the suggestion of generally low PaO2
238                                           C. J. BRAUNER AND M. BERENBRINK

values in water‐breathing fishes (Lenfant et al., 1970), available data suggest
that salmonids and other teleosts that require well‐oxygenated water, such as
Atlantic cod or striped bass, have a high normoxic resting PaO2 around
100 mmHg (Nikinmaa and Salama, 1998), similar to or sometimes even
exceeding values in the air‐breathing birds and mammals. On the other
hand, cyprinids show normoxic resting PaO2 values of 20–40 mmHg.
This is not due to a low diVusion capacity of the gills in these animals but
is regulated by intermittent ventilation as shown by the increased PaO2
values under exercise‐induced hyperventilation (Jensen et al., 1983;
Koldkjær and Jensen, 1998; see Nikinmaa and Salama, 1998). Species with
low resting PaO2 values commonly have a high blood O2 aYnity, such that
nearly complete Hb‐O2 saturation can nevertheless be achieved. These spe-
cies are often also hypoxia tolerant, like the common carp, but so far the
diVerences in the regulation of PaO2 values in teleosts have not been system-
atically addressed and it is unclear to what extent phylogenetic position plays
a role.
    PO2 at the principle site of O2 consumption in the mitochondria can be
considered to be close to zero, and therefore the diVusive O2 transfer from
venous blood to tissue mitochondria is directly related to venous PO2 (PvO2).
Tissue O2 supply is therefore enhanced by Hbs, which oZoad O2 at relatively
high venous PvO2 values. The other major factor determining the rate
of diVusive O2 transfer from blood to mitochondria in a given tissue is the
diVusive O2 conductance. The latter depends on factors such as the geometry
of the structures through which diVusion occurs (e.g., capillary density and
length), the O2 diVusion coeYcients of the various tissue components, their
O2 capacitance coeYcients, and whether myoglobin‐facilitated diVusion
occurs (Dejours, 1975). Except under special circumstances like in O2 secre-
tory structures, PvO2 must be lower than PaO2 for tissue oxygenation, and
Hbs requiring only a small decrement from PaO2 for oZoading significant
amounts of O2 (i.e., Hbs with PaO2 and PvO2 in the steepest part of the O2
equilibrium curve) are most eYcient for tissue O2 supply. On the other hand,
to ensure near complete saturation of Hb in arterial blood, PaO2 should
ideally be regulated in the flat upper part of the O2 equilibrium curve, that is,
away from the steep portion. Perhaps it is for this reason that PaO2 usually
exceeds PvO2 by a safety margin of several mmHg.
    In primitive air‐breathing fishes, the accessory ABOs are arranged in
parallel to the systemic tissues and even if PO2 in blood leaving these organs
is high, the admixture of systemic mixed venous blood in an undivided heart
will subsequently reduce it. If the gills are reduced, collapsed, or contain shunt
pathways, the PO2 of mixed blood will be little changed and constitute the
systemic arterial PO2 . If the gills are well developed, not collapsed, and contain
no shunts, PO2 of mixed blood can be increased or decreased, depending on
whether the PO2 gradient to inspired water is positive or negative.
5.   GAS TRANSPORT AND EXCHANGE                                            239

3.3. Survey of Extant Primitive Fishes

3.3.1. Ray‐Finned Fishes (Actinopterygii)
    a. Bowfin (Amiiformes). Relative to many teleost fishes, the P50 in whole
blood of A. calva is quite low, ranging from 4 to 24 mmHg depending on the
temperature, pH, and CO2 level (Table 5.1). The Bohr coeYcient (F) in whole
blood is largely temperature independent and ranges from –0.43 to À0.51
(Table 5.1). The Bohr eVect in washed RBCs at 15  C in the absence of CO2 is
large, with F ¼ À1.0 (Table 5.1) (Weber et al., 1976b). It is unclear whether
the somewhat reduced Bohr eVect in whole blood is due to a specific CO2
eVect since arterial PCO2 is low when the fish is water breathing (>4 mmHg,
Randall et al., 1981). A. calva blood contains at least five diVerent Hb
components with broadly similar functional properties and amino acid
compositions (Weber et al., 1976b).
    At a PO2 of 150 mmHg, a PCO2 of 26 mmHg depressed Hb‐O2 saturation
in A. calva whole blood by only 8% (Black and Irving, 1938). Johansen et al.
(1970) similarly found that a pH change from 8.0 to 7.2 only caused a 10%
reduction in O2 capacity and judged the Root eVect in A. calva whole blood as
insignificant. The apparent lack of a Root eVect in whole blood is surprising
given the importance of the Root eVect for O2 secretion (Waser and Heisler,
2005), the presence of a well‐developed choroid rete mirabile in the eye of
A. calva and PO2 values considerably above air saturation close to the retina
(Wittenberg and Wittenberg, 1974). In contrast, the hemolysate shows a
strong Root eVect of about 60% (Weber et al., 1976b; Berenbrink et al.,
2005). Plasma pH changes during and after exhaustive exercises are trans-
ferred to the RBC interior (Gonzalez et al., 2001). The changes in RBC
chloride concentration and water content with changes in extra‐ and intracel-
lular pH in vitro and the inhibition of proton equilibration across the RBC
membrane by the anion exchange inhibitor DIDS further all point to a passive
equilibration of protons across the RBC membrane (Tufts et al., 1994). Using
the formula obtained for washed A. calva RBCs by the latter authors, the
extracellular pH of 7.2 in Johansen et al. (1970) corresponds to an RBC pH of
about 6.8. This value is suYcient for 50% Hb deoxygenation in the hemolysate
at saturating ATP concentrations (Figure 4 in Weber et al., 1976b). A. calva
RBCs do not possess an adrenergic NHE (Tufts et al., 1994; Berenbrink et al.,
2005), which could otherwise have elevated RBC pH and thereby shifted the
onset of the Root eVect to lower plasma pH values in the study by Johansen
et al. (1970). Perhaps reduction of O2 capacity by only 10% at pH 7.2 in the
latter study was due to less than saturating organic phosphate concentrations
in the RBCs. Saturating levels of organic phosphates are known to shift the
onset of the Root eVect to higher pH values (Weber et al., 1976b). From this it
follows that blood pH in the choroid rete mirabile of A. calva must be below
7.2 for maximal exploitation of the Root eVect.
240                                         C. J. BRAUNER AND M. BERENBRINK

     b. Gars (Lepisosteiformes). At 20  C, whole blood of the spotted gar,
L. oculatus, had a P50 of 24.1 mmHg at 1% CO2 in vitro, which increased to
30.4 at 2% CO2. The shape of the O2 equilibrium curves was distinctly
sigmoidal, and possessed F ¼ À0.5 (Table 5.1) (Smatresk and Cameron,
1982a), which was similar to the value measured above in whole blood from
A. calva. RBCs of L. oculatus and the Florida gar Lepisosteus platyrhincus
lack significant adrenergic NHE activity (Berenbrink et al., 2005). In L.
platyrhincus, the Bohr eVect in organic phosphate‐free hemolysates at physi-
ological temperature, pH, and chloride concentrations was F ¼ À0.525 and
bHb was À8.6 per mmol Hb4 and pH. This magnitude of the Bohr eVect is
distinctly lower and the value for bHb distinctly higher than in isolated Hbs of
A. calva and most teleosts under the same conditions (Table 5.1) (Berenbrink
et al., 2005).
     Changing the PCO2 of air‐equilibrated whole blood of L. oculatus from
1% to 2% CO2 caused pH to drop from 7.8 to 7.4 and was associated with a
statistically significant decrease in O2 capacity by about 13%, which was
taken to indicate a Root eVect (Smatresk and Cameron, 1982a). However,
it is not clear to what extent the decrease in Hb‐O2 binding was caused by
oxidation of Hb, which readily occurs at low pH values. Hemolysates of
L. oculatus and L. platyrhincus show a Root eVect of about 43% and 40%,
respectively (Berenbrink et al., 2005). The presence of strong Root eVect
Hbs in Lepisosteiformes may be surprising in view of the general absence
of a choroid or swim bladder rete mirabile outside teleosts and A. calva
(Berenbrink et al., 2005). If the reduction of whole‐blood O2 capacity at
pH 7.4 was indeed caused by a Root eVect (Smatresk and Cameron, 1982a),
the low blood pH values under severe exercise (Burleson et al., 1998)
indicate that the Root eVect may be quite important at physiological blood
pH values. However, in the absence of gas exchanging retia mirabilia in
gars, which will concentrate CO2 as well as O2, it is unclear whether low
enough RBC pH values can be achieved to elicit the full magnitude of the
Root eVect in vivo. Despite the absence of a choroid rete mirabile, maximal
ocular PO2 values ranged between 72 and 145 mmHg in L. oculatus and
alligator gar [Atractosteus (formerly Lepisosteus) spatula], with an average
of 90 mmHg (Wittenberg and Wittenberg, 1974). Even the lowest recorded in-
dividual ocular PO2 value is about twice as high as the highest recorded
average dorsal aortic PO2 value of 37 , which was observed after recovery
from exhaustive exercise in normoxic water at 23–24  C (Burleson et al.,
     Ocular PO2 measurements were performed on restrained animals
submerged in surface water between 22 and 33  C immediately after they
had been caught from bottom temperatures of 23  C. However, even when it
is assumed that the eye is a closed system and does not consume O2 (neither
5.   GAS TRANSPORT AND EXCHANGE                                             241

of which is true), the calculated temperature‐induced increase in PO2 by the
decreased O2 solubility is only 20% (Wittenberg and Wittenberg, 1974). This
indicates that ocular PO2 is elevated above PaO2 in the absence of a choroid
rete mirabile and suggests that the strong Root eVect of Lepisosteus Hbs may
be involved in O2 delivery to the eye. Even in the absence of a distinct choroid
rete mirabile, a certain degree of countercurrent gas exchange may contribute
to the elevated ocular PO2 given the general blood supply characteristics of
vertebrate eyes. The ophthalmic artery and vein often share a single entry/
exit point in the sclera. From here, arterial and venous networks fan out in
close vicinity to supply and drain the choroid capillaries. This arrangement
by necessity already creates favorable conditions for countercurrent gas
exchange. Clearly more studies are necessary to establish whether the eyes
of gars resemble an intermediate stage in the evolution of O2 secretion, which
is so powerfully developed in A. calva and teleosts (Wittenberg and
Wittenberg, 1974; Berenbrink et al., 2005).

    c. Sturgeons and Paddlefishes (Acipenseriformes). No published infor-
mation appears to exist on the blood O2 transport characteristics of the two
paddlefish genera and all information for Acipenseriformes stems from
studies on just one out of four sturgeon genera, namely Acipenser. Whole
blood P50 is 21.5 mmHg in A. transmontanus (Table 5.1). The Bohr coeY-
cient in whole‐blood ranges from À0.4 to À0.55 (Table 5.1).
    Intracellular RBC pH in A. naccarii is much lower than plasma pH
(McKenzie et al., 1997). Studies on washed RBCs of A. baeri suspended in
CO2/bicarbonate buVered salines show that extracellular pH changes are
transferred to the RBC interior to a similar extent as in rainbow trout
RBCs under the same conditions (A. J. da Silva and M. Berenbrink, unpub-
lished data). The latter study provided evidence for a small Root eVect as
RBC acidification by CO2 at a PO2 of 150 mmHg caused 15% and 20%
deoxygenation at pH 7.1 and 6.6, respectively.
    Plasma catecholamine levels sharply increase during mild and severe
acute hypoxia in A. naccarii and A. transmontanus (Randall et al., 1992;
Maxime et al., 1995), an adrenergic NHE, however, appears to be missing
in washed RBCs of A. baeri and sterlet (Acipenser ruthenus; A. J. da Silva and
M. Berenbrink, unpublished data; Berenbrink et al., 2005), as well as in
A. transmontanus (Baker, Rummer, and Brauner, unpublished data).
    The RBC GTP concentration in A. naccarii is equal to, and in A. ruthenus
is twice as high as the RBC ATP concentration (Clementi et al., 1999;
O. Kepp and M. Berenbrink, unpublished data). The Bohr eVect and P50
of purified hemolysates of A. naccarii in the presence of physiological chlo-
ride concentrations and excess GTP are similar to the values in washed RBCs
in phosphate buVer (Clementi et al., 1999).
242                                         C. J. BRAUNER AND M. BERENBRINK

     The Hb system of A. naccarii consists of two Hb components in equal
amounts composed of three a‐ and two b‐chain types, with the individual
Hb components apparently containing diVerent a‐ and b‐chains in each
component (Clementi et al., 2001a,b).
     The small Root eVect in RBCs of A. baeri (see above) is also seen in
hemolysates, which show Root eVect magnitudes of 15–26% in A. ruthenus,
A. naccarii, and A. baeri (Clementi et al., 1999; Berenbrink et al., 2005).
     No swim bladder rete mirabile is present in the genus Acipenser (Muller,
1839; Bridge, 1904), and swim bladder O2 secretion also appears to be absent
(Fange, 1966). Despite the generally more benthic lifestyle, the swim bladder
is regressed only in a few sturgeon species (Rauther, 1922). Presumably a
normal‐sized swim bladder is useful during the long migrations undertaken
by some sturgeon species. No choroid rete mirabile has been found in the
genera Acipenser, Scaphirhynchus, and Polyodon (Muller, 1839; Rauther,
1937; Wittenberg and Haedrich, 1974). However, Rodrıguez and Gisbert
(2001) have mentioned a choroid gland, an old name for the choroid rete
mirabile, in 5‐ to 6‐day‐old larvae of A. baeri. The structure is described as
conjunctive tissue with two pigmented layers and hence may not constitute
the vascular countercurrent exchange system usually described by this name.
TretjakoV (1926) has described a suprachorioidea in the starry sturgeon,
A. stellatus, consisting of closely arranged sinusoid veins, of which the larger
veins regularly alternate with larger arteries. He suggested that the choroid
rete mirabile of teleosts may have evolved from a similar arrangement
(TretjakoV, 1926). PO2 and pH in choroid capillaries in the eyes of sturgeons
are not known, so it is unclear whether the small Root eVect is used in ocular
O2 secretion. Arterial pH values of 7.1 or below have been measured during
external hypercapnia or deep hypoxia (Burggren and Randall, 1978; Crocker
and Cech, 1998), and the very much lower RBC pH compared to plasma pH
(McKenzie et al., 1997) all suggest that the Root eVect may be occurring
in vivo in the general circulation, let alone under the more acidic conditions
in tissues.

    d. Reedfish and Bichirs (Polypteriformes). The P50 of in vitro whole
blood of the gray bichir (P. senegalus) at 30  C is 23.5 mmHg (Table 5.1)
(Vokac et al., 1972). Hb‐O2 binding is sigmoid with nH ¼ 2.68 (calculated
from Fig. 2B in Vokac et al., 1972) and F ¼ À0.43 (Table 5.1). Using this
value and assuming that the normoxic, resting arterial blood pH at 30  C is
about 7.55, like in other facultative air‐breathing ray‐finned fishes such as
A. calva and Lepisosteus species (Ultsch and Jackson, 1996), the P50 in
arterial blood is calculated to be 27.3 mmHg.
    RBCs of P. senegalus contain high levels of ATP and GTP, which change
during ontogeny as well as hypoxia acclimation (Babiker, 1984a). The sum of
5.   GAS TRANSPORT AND EXCHANGE                                              243

these nucleotides changes little during hypoxia acclimation of juveniles that
have not yet developed the ability to breathe air, or during maturation from
juveniles to bimodally breathing adults. However, in both cases the relative
contribution of GTP increases, from 20% to 35% during hypoxia acclimation
and 5–30% during maturation (Babiker, 1984a). Unfortunately, whole blood
Hb‐O2 aYnity under these conditions is not known. Because GTP is a more
eVective organic phosphate than ATP, a decrease in Hb‐O2 aYnity may be
expected (Val, 2000).
    In E. calabaricus, whole blood in vitro has a P50 of 17.9 mmHg and an nH
value of 2.0 at 25  C, and F is À0.247 (Table 5.1). Air exposure for 4 h does
not change P50 or the Bohr eVect but significantly increases nH (Beitinger
et al., 1985). Assuming an in vivo normoxic resting arterial pH at 25  C of 7.65
(close to values in A. calva and Lepisosteus species), a P50 of 17.0 mmHg can
be calculated.
    E. calabaricus RBCs contain ATP and GTP in equal amounts and
relatively high 2,3‐DPG levels of about 40% of the sum of ATP and GTP.
These levels show no major changes after 4‐h air exposure (Beitinger et al.,
1985). Referring to the latter work, Graham (1997) has erroneously stated
that reedfish RBCs also contain IP5, which is the major organic phosphate in
birds but also occurs in lungfishes (see below).
    Whole blood of P. senegalus shows a small Root eVect. Hb‐O2 saturation
at 30  C and a PO2 of 200–220 mmHg is close to 100% between pH 7.74 and
7.47 and decreases to about 93% between pH 7.16 and 7.00 (Vokac et al.,
1972). However, the maximal Root eVect in hemolysates of the ornate bichir
(Polypterus ornatipinnis) and E. calabaricus are somewhat higher with 14%
and 15%, respectively, at 25  C (Berenbrink et al., 2005).
    Retinal PO2 in Polypteriformes is not known, but choroid and swim
bladder retia mirabilia are missing in all members of this group that have
been studied (Bridge, 1904; Wittenberg and Haedrich, 1974).
    Berenbrink et al. (2005) found no evidence of an adrenergic NHE in
washed RBCs of P. ornatipinnis.
3.3.2. Lobe‐Finned Fishes (Sarcopterygii)
    a. Coelacanths. At 20  C and a pH of 7.8, Wood et al. (1972) determined
the rather low P50 of 3.3 mmHg in the blood of L. chalumnae. In the same
study, whole blood F was À0.51 (Table 5.1). A low P50 and the presence of a
Bohr eVect were confirmed by Hughes and Itazawa (1972), who measured a
P50 value of 2.1 mmHg and showed that P50 in whole blood increases with
increasing temperature, as in most other vertebrates. Both groups found
that blood‐O2 binding was essentially noncooperative at high pH values
(nH ca 1.0), although Wood et al. (1972) showed an unusual increase in
cooperativity at low pH (nH ca 1.6 at pH 6.7).
244                                         C. J. BRAUNER AND M. BERENBRINK

    Studies on purified Hb at 20  C revealed even lower P50 values than
in whole blood but similar values for the Bohr coeYcient. A change from
hyperbolic to cooperative Hb‐O2 binding below pH 7.5 was observed
and an increase in P50 and nH in the presence of phosphate buVer was
demonstrated (Wood et al., 1972; Bonaventura et al., 1974). Increasing
temperature decreased Hb‐O2 aYnity to a similar extent as in other fish
Hbs (DH ¼ À10.4 kcal per mol Hb; Wood et al., 1972). Unlike mammalian
Hbs, but similar to elasmobranchs, coelacanth Hb is insensitive to high urea
concentrations, which correlates with high urea concentrations in the blood
of elasmobranchs and the coelacanth (Mangum, 1991).
    Wood et al. (1972) also found a 12% reduction in O2 capacity when blood
was acidified with 3% CO2 in air and attributed this to a moderate Root
eVect. In contrast, Hughes and Itazawa (1972) did not find any evidence of
a Root eVect because blood O2 capacity at pH 7.30 and 6.88 at 15  C and
pH 6.55 at 28  C was the same. However, pH 7.3 is 0.5 pH units below
the estimated normal arterial pH value in L. chalumnae (about 7.8, see
below), and it is possible that the maximum decrease in O2 capacity via the
Root eVect already occurs at higher pH values than this. Unfortunately,
Hb concentration and methemoglobin (metHb) fraction were not reported,
precluding the calculation of functional Hb‐O2 saturation under these exper-
imental conditions. Similarly, the pH values before and after equilibration to
3% CO2 in air in the study by Wood et al. (1972) are not known, and it is
unclear whether the 10% decrease in O2 capacity represents a maximal
reduction or whether further acidification would have resulted in even higher
decreases. In addition, since 10% metHb had already been formed in their
sample and lower pH values generally increase the rate of metHb formation
(Wallace et al., 1982), it is possible that this was the cause of reduced Hb‐O2
binding at low pH and that L. chalumnae blood does not show a Root eVect.
    Gorr et al. (1991a) have determined the amino acid sequence of
L. chalumnae Hb, and on the basis of the molecular mechanism for the
Root eVect proposed by Perutz and Brunori (1982), they predicted in a
subsequent paper the absence of a Root eVect in L. chalumnae Hb (Gorr
et al., 1991b). However, the model of Perutz and Brunori for the Root eVect
mechanism has been subsequently challenged (Ito et al., 1995; Yokoyama
et al., 2004; see Berenbrink, 2006). Thus, the prediction by Gorr et al. (1991b)
about the absence of a Root eVect may be invalid. The question whether
L. chalumnae possesses a large Root eVect Hb is still open.
    The high‐O2 aYnity of L. chalumnae blood deserves special comment.
However, before discussing its physiological relevance, it is worthwhile to
briefly mention three possible reasons for measuring erroneously high blood
O2 aYnities. First, long storage times until analysis may reduce RBC organic
phosphates and thereby increase Hb‐O2 aYnity. The natural organic
5.   GAS TRANSPORT AND EXCHANGE                                           245

phosphate, if any, that modulates Hb‐O2 aYnity in L. chalumnae RBCs is not
known. At pH 7.47, which may be close to physiological RBC pH at a plasma
pH of 7.8, ATP has only a small eVect on P50 of isolated Hb in Tris buVer
(Wood et al., 1972), but the eVects of other organic phosphates, such as GTP,
2,3‐DPG, IP5, or IP2, have not been tested. Another reason for the measure-
ment of high‐O2 aYnities may be metHb formation. 10% of the sample was
oxidized to methemoglobin in the study by Wood et al. (1972). Although
metHb itself is unable to bind O2, it is known to increase the average O2
aYnity of the remaining functional heme groups (Falcioni et al., 1977;
Kwiatkowski et al., 1994). Mangum (1991) showed that procedures normally
removing metHb did not reduce the high‐O2 aYnity in L. chalumnae Hb
preparations, although the initial metHb concentration was not reported.
Finally, the animals live at much higher hydrostatic pressures than those
under which O2 equilibrium curves are normally established, and it is not
known how this diVerence aVects blood O2 aYnity.
     Bearing the above caveats in mind, a high blood O2 aYnity in L. chalumnae
may signify tolerance to a low‐O2 environment (Wood et al., 1972). Ever since
the pioneering work of Krogh and Leitch (1919), a correlation between high
blood O2 aYnities and hypoxic habitats has been noted in water‐breathing
vertebrates. Thus, common carp Cyprinus carpio, which frequently encoun-
ters hypoxic environments, has a whole blood P50 of ca 7 mmHg at pH 7.9 and
20  C (Weber and Lykkeboe, 1978). This contrasts with a P50 of ca 23 mmHg
at pH 7.93 and 15  C in rainbow trout Oncorhynchus mykiss (Soivio et al.,
1980). Thus, rainbow trout, which live in well‐oxygenated waters and are
relatively hypoxia intolerant, have more than a threefold higher P50 value,
despite the lower measurement temperature that tends to decrease P50.
     Using their submersibles GEO and JAGO, Fricke and Hissmann (2000)
established the depth profile of physically dissolved O2 in the habitat of
L. chalumnae. They found that the caves, which are used as resting places
during the day, were located at depths corresponding to an O2 minimum
layer at about 200 m. Dissolved O2 concentration decreased from about
7.6 mg literÀ1 in 27  C surface water to about 5.1 mg literÀ1 in 20  C water
at the level of the caves, as estimated from Fig. 1 in Fricke and Hissmann
(2000). However, to assess the gradient for diVusive O2 uptake under these
conditions, the actual PO2 of the water, which was not given, is more
important than the absolute O2 concentration. Using the O2 solubility coeY-
cient for seawater at 20  C (tabulated in Dejours, 1975), a PO2 of about
100 mmHg or 2/3 of air‐equilibrated seawater can be calculated. During
the night, L. chalumnae uses energy‐saving drift‐hunting at depths below
the level of the caves, where dissolved O2 concentrations gradually increase
again (ca 6.0 mg literÀ1 at 400 m and 12.5  C; Fricke and Hissmann, 2000).
However, the increased concentration of dissolved O2 is accompanied by an
246                                         C. J. BRAUNER AND M. BERENBRINK

increased solubility at the lower temperature prevailing at 400 m such that
the calculated PO2 remains unchanged at about 100 mmHg. Thus, a high
blood O2 aYnity in L. chalumnae may indeed be related to the low‐O2
availability in its habitat and ensures full Hb‐O2 saturation in the gills.
    Another reason for a high blood O2 aYnity may be the low gill surface
area and relatively thick water–blood diVusion barrier of L. chalumnae
gills measured by Hughes (1972, 1995), which results in a very low diVusing
capacity of the water–blood barrier and may result in a large PO2 drop
between inspired water and arterial blood. As the PO2 of inspired water
is already low due to environmental hypoxia, a high‐O2 aYnity may be
necessary for adequate oxygenation of Hb at low arterial blood PO2 .
    On the basis of the low body mass specific gill area (Figure 5.1), Hughes
(1972, 1976, 1995) has predicted a low resting M O2 and a sluggish lifestyle
for L. chalumnae, similar to some deepwater teleosts. The latter has been
confirmed by observations from submersibles (Fricke and Hissmann, 2000).
    From a high blood O2 aYnity it follows that PvO2 must be comparatively
low in order to allow significant O2 unloading in tissues. The P50 values
measured by Wood et al. (1972) and Hughes and Itazawa (1972) suggest that
PvO2 values as low as 2–3 mmHg are required for 50% Hb‐O2 unloading.
Unloading may be facilitated by slightly lower venous than arterial pH
values and the Bohr eVect of L. chalumnae blood, but the noncooperative,
hyperbolic shape of the blood O2 equilibrium curve appears disadvanta-
geous, as a given arteriovenous PO2 diVerence at a given P50 causes smaller
changes in Hb saturation when compared to the cooperative, sigmoid O2
equilibrium curves found in many other vertebrate groups.
    L. chalumnae is live bearing (Smith et al., 1975) with a 98‐kg female
containing as many as 26 fetuses with a near‐term body mass up to about
500 g (Heemstra and Greenwood, 1992). The extent of maternal trophic input
in L. chalumnae has been a matter of intense debate, and suggestions of
prenatal oophagy, uterine cannibalism, and placentotrophy in an earlier
volume of the Fish Physiology series (Wourms et al., 1988) have been strongly
criticized (Heemstra and Greenwood, 1992). Nevertheless, Wourms et al.
(1991) demonstrated an extensive plexus of blood vessels in a specialized
area in each of the uterine compartments closely surrounding individual
fetuses. These blood plexuses lie in close apposition to the heavily vascular-
ized yolk sac and are discussed as yolk sac placenta specialized for molecular
transport and gas exchange by Wourms et al. (1991). Live bearing has
independently evolved in many elasmobranch and teleostean lineages
(Reynolds et al., 2002). In several cases, maternal‐fetal O2 transfer appears
to be facilitated by a shift toward a distinctly higher blood O2 aYnity in fetal
blood compared to maternal blood (Ingermann et al., 1984; Weber and
Hartvig, 1984; King, 1994). It will be most interesting to see whether this
5.   GAS TRANSPORT AND EXCHANGE                                             247

is also true for fetal L. chalumnae blood because of the already very high
blood O2 aYnity of the adult described above.
    Early studies showed that blood of adult L. chalumnae contains a single
tetrameric Hb composed of two diVerent subunit types conforming to the
general a2b2 subunit structure of other jawed vertebrates (Weber et al., 1973;
Bonaventura et al., 1974). The amino acid sequences of the a‐ and b‐chains
were determined by Gorr et al. (1991a), who found that these chains share
51 and 48 identical residues with the respective chains in human HbA.
The molecular mechanism behind the low cooperativity and low intrinsic
O2 aYnity of L. chalumnae Hb was explored by molecular modeling and
comparison with human HbA. These studies are exceedingly diYcult when
dealing with a large number of amino acid diVerences. Nonetheless, Gorr
et al. (1991b) were able to predict a distinct diVerence in the mechanism of the
Bohr eVect between Latimeria Hb and human HbA, despite the similar Bohr
coeYcient displayed by these two Hbs. They also predicted the absence of
a Root eVect in L. chalumnae Hb (see above). In contrast to the a‐chains
of most teleost Hbs, the N‐terminal amino groups of the globin chains in
L. chalumnae are not acetylated (Gorr et al., 1991b).

    b. Lungfishes (Dipnoi). Blood respiratory characteristics in the faculta-
tive air‐breathing, nonestivating N. forsteri were investigated by Lenfant
et al. (1966). At 18  C during normoxia at rest, this species is predominantly
water breathing and it is assumed that blood in the pulmonary artery
represents systemic arterial blood, that is that no central blood shunting
occurs. Under these conditions, a PaCO2 of 3.5 mmHg is as low as in typical
water‐breathing fishes (Table 5.1). The PaO2 of 40 mmHg is rather low.
Nevertheless an Hb‐O2 saturation of 95% can be calculated for these condi-
tions from in vitro whole‐blood O2 equilibrium data. This is due to a low P50
of 11 mmHg and sigmoid Hb‐O2 binding (Lenfant et al., 1966). The Bohr
eVect was rather high with F ¼ À0.620.
    In a more recent study on N. forsteri at 20  C, P50 was 22.0 mmHg and nH
was 2.27 at pH 7.5 and PCO2 of 16 mmHg. The Bohr eVect was considerably
lower with F ¼ À0.48 (Kind et al., 2002). These authors found a significant
left‐shift of the O2 equilibrium curve with hypoxia acclimation, although
ATP levels did not change. However, GTP, which makes up 30% of the NTPs
in N. forsteri (Isaacks and Kim, 1984), was not measured.
    In contrast to N. forsteri, the other two genera of lungfishes are obligato-
ry air‐breathers grouped in their own separate family. Working on juvenile
L. paradoxa, Johansen and Lenfant (1967) clearly demonstrated a selective
passage of oxygenated pulmonary blood and systemic venous blood through
the only partially divided heart because PO2 in dorsal aortic blood exceeded
pulmonary arterial PO2 (10–43 mmHg versus 5–16 mmHg, respectively).
248                                        C. J. BRAUNER AND M. BERENBRINK

Higher dorsal aortic PO2 values of 80–100 mmHg have been reported from
somewhat larger specimens (Amin‐Naves et al., 2004).
     In vivo whole blood P50 at 23  C and PCO2 of 6 mmHg was 10.5 mmHg in
the study by Johansen and Lenfant (1967), showing a rather high Hb‐O2
aYnity. The Bohr coeYcient was F ¼ À0.234. In later studies on adult
Lepidosiren, slightly higher values of F ¼ À0.293 and À0.31 have been
measured (Johansen, 1970; Johansen et al., 1978). All three foregoing values
are less than half the value of F ¼ À0.66 determined at 25  C by Bassi
et al. (2005). The latter study found a smaller Bohr eVect at 35  C, where F
was À0.44. In contrast, Powers et al. (1979) found F ¼ À0.33 at 20  C and
30  C (Table 5.1).
     The values for the Bohr coeYcient reported in P. aethiopicus are equally
variable, ranging from F ¼ À0.28 to À0.35 and À0.47 at 23–25  C (Lahiri
et al., 1968; Swan and Hall, 1966; Lenfant and Johansen, 1968) (Table 5.1).
     Some of the above authors have oVered an adaptive explanation for the
particular size of the Bohr eVect measured in their study. These interpreta-
tions are tenuous because it is not known to what extent the variations are
due to experimental technique, phenotypic plasticity, or inheritable, genetic
diVerences between individuals, populations, and species.
     Several of the whole‐blood O2 equilibrium curves in earlier studies by
Johansen and coworkers were obtained by the mixing method, which was
previously successfully used on invertebrate blood (Lenfant and Johansen,
1965) but, as later admitted by the authors, may not work well with fish
blood (Johansen et al., 1976b). Problems include accurate PO2 measurements
in bloods with high rates of O2 consumption and depletion of organic
phosphates in anoxically incubated subsamples (Scheid and Meyer, 1978;
Tetens and Lykkeboe, 1981). The latter may also be a problem in the study
by Johansen and Lenfant (1967), where whole blood was stored for more
than 48 h before O2 equilibrium curves were determined.
     Evidence for some individual and phenotypic plasticity of the Bohr
eVect comes from studies on estivating African lungfishes (Johansen et al.,
1976b). At 26  C, in vitro whole blood F in two active P. amphibius was À0.29
and À0.20 relative to À0.41 and À0.68 in two individuals that had been
estivating for 28–30 months. The elevated whole blood F in estivating
animals was associated with a large reduction in RBC GTP levels, whereas
ATP levels remained unchanged. In a companion study, it was shown that
estivation did not change the functional properties or relative proportion of
the three Hb isoforms in organic phosphate‐free solutions. Addition of ATP
increased the Bohr eVect, but the eVects of GTP on the Bohr eVect have not
been reported. At constant pH, GTP had a stronger eVect on P50 than ATP
(Weber et al., 1976a). The dynamics of the Bohr eVect changes between
estivating and active animals in vivo are not known, but it is possible that
5.   GAS TRANSPORT AND EXCHANGE                                                249

part of the variable Bohr eVect in active animals can be explained by the time
that has elapsed since the last estivation. Although F in whole blood is
correlated with RBC GTP levels, the mechanism for the increased Bohr
eVect during estivation remains to be investigated.
    Babiker (1984a) confirmed decreased GTP levels in RBCs of estivating
P. annectens and further showed an increase in relative and absolute GTP
levels during the ontogenetic change from almost completely water‐breathing
juveniles to obligatory air‐breathing adults. Babiker (1984a) further showed
that acclimation to aquatic hypoxia in maturating juveniles led to an increase
in GTP levels while ATP levels decreased slightly. Unfortunately, there are no
data on P50 and the Bohr eVect under these conditions, but it is possible that
ontogenetic changes and levels of previous hypoxic exposure contribute to the
variation of the Bohr eVect within species.
    In contrast to P. amphibius, P. annectens and L. paradoxa contain only
one major Hb isoform, and studies on their functional properties in solution
suggest distinct interspecific diVerences in their intrinsic O2 aYnity and
interaction with allosteric modifiers (Weber et al., 1976a; Phelps et al.,
1979). Unlike N. forsteri, Protopterus species and L. paradoxa are unusual
in containing significant amounts of inositol‐diphosphate (IP2) and uridine
phosphates (UTP and UDP) in their RBCs (Bartlett, 1978b; Isaacks
et al., 1978). The eVects of these organic phosphates on Hb‐O2 binding are
    High urea concentrations accumulate in the blood of African lungfishes
during estivation (Smith, 1930). A plasma osmolality increase from 235 to
650 mOsm over 13 months of estivation has been reported (DeLaney et al.,
1977). Urea concentrations in estivating L. paradoxa are unknown, but the
RBCs of the species have unusually high urea permeability (Kim and Isaacks,
1978). In contrast to humans, high urea does not decrease P50 in P. amphibius
hemolysates (Weber et al., 1976a).
    RBCs of P. aethiopicus possess an anion exchanger and extracellular pH
changes are transferred to the interior, suggesting passive proton equilibra-
tion across the RBC membrane (Jensen et al., 2003). RBCs of this species
express b‐adrenergic receptors, and aerial (but not aquatic) hypoxia signifi-
cantly increases plasma catecholamine levels in the related P. dolloi (Koldkjær
et al., 2002; Perry et al., 2005). However, in neither of these two species, nor in
L. paradoxa, could an adrenergic NHE be demonstrated using diVerent
methods (Koldkjær et al., 2002; Berenbrink et al., 2005; Perry et al., 2005).
    A Root eVect is absent in whole blood of all three lungfish genera
(Johansen, 1970), which has been confirmed for the hemolysates of an
unidentified Protopterus species and L. paradoxa (Farmer et al., 1979;
Berenbrink et al., 2005). A choroid rete mirabile is also absent (Wittenberg
and Haedrich, 1974).
250                                         C. J. BRAUNER AND M. BERENBRINK

3.3.3. Jawless Fishes (Agnatha)
    The biology of hagfishes has been reviewed in two compilations of articles
by Brodal and Fange (1963) and, more recently, by Jørgensen et al. (1998).
A similar compilation about the biology of lampreys has been edited by
Hardisty and Potter (1971, 1972). All three compilations contain chapters
about the blood O2 transport characteristics (Manwell, 1963; Riggs, 1972;
Fago and Weber, 1998). Comparative aspects of blood O2 transport in
agnathans relative to teleosts, elasmobranchs, and terrestrial vertebrates
are discussed by Nikinmaa (1990, 1992, 1997, 2001). As mentioned earlier,
agnathan Hbs consist of high O2 aYnity monomers, which aggregate to low
O2 aYnity dimers and tetramers at low pH and PO2 .

     a. Lampreys (Petromyzontiformes). Despite the general anatomical
diVerences between the gills in agnathans and jawed fishes, gill O2 diVusion
capacity is high in the Pacific lamprey Lampetra tridentata and the sea lamprey
P. marinus as indicated by high normoxic PaO2 values of up to 77 and
120 mmHg at 14 and 10  C, respectively, which is associated with an Hb‐O2
saturation of about 95% (Table 5.1) (Johansen et al., 1973; Tufts, 1991). At
10  C and pH of 7.75, whole blood P50 in L. fluviatilis is 11.8 mmHg in the
adults but only 1.8 mmHg in the larvae (Bird et al., 1976). The higher O2
aYnity of the larvae has been attributed to their buried lifestyle in presumably
hypoxic sediments, whereas the parasitic adults have an active lifestyle in more
oxygenated waters, actively attaching themselves to teleost prey, and under-
taking long spawning migrations (Bird et al., 1976). When the Bohr coeYcient
is related to extracellular pH values, lamprey blood shows low F values
between À0.1 and À0.3. However, since a functional anion exchanger is
lacking and extracellular pH changes are not readily transferred to the RBC
interior in lampreys, this underestimates the pH sensitivity of Hb‐O2 binding
and values of F ¼ À0.63 to À1.03 have been measured for the intracellular
Bohr coeYcient (Table 5.1) (Ferguson et al., 1992; Nikinmaa et al., 1995).
     At high pH, cooperativity of Hb‐O2 binding is saturation dependent, nH
at 35%, 50%, and 80% saturation increases from 1.00 to 1.21 and 2.32,
respectively (Bird et al., 1976). At lower pH values, Hb‐O2 binding becomes
hyperbolic and nH is close to 1.00 regardless of saturation (Nikinmaa, 1993).
     After exhaustive exercise, normoxic dorsal aortic pH in P. marinus at
10  C fell from 7.9 to ca 7.55. However, as this change was not transferred to
the intracellular RBC compartment, pHi stayed constant (Tufts, 1991).
Nevertheless, arterial Hb‐O2 saturation fell transiently to about 80% because
PaO2 decreased from 120 to about 80 mmHg.
     Acute hypoxia (PwO2 ¼ 40–50 mmHg) in L. fluviatilis resulted in a large
increase in ventilation frequency, accompanied by a 50% increase in M O2 ,  _
5.   GAS TRANSPORT AND EXCHANGE                                             251

which after 1 week was reduced to 125% of normoxic control values
(Nikinmaa and Weber, 1984). This was paralleled by an acclimatory increase
in Hb‐O2 aYnity, brought about by an increase in RBC pH and the Bohr
eVect, and also by a decrease in intracellular Hb concentration, which is
known to shift the monomer–oligomer equilibrium toward the higher aYnity
monomeric Hb form (Nikinmaa and Weber, 1984).
    RBC pH in lampreys is elevated above the proton equilibrium distribution
by the activity of an NHE mechanism as first shown by Nikinmaa (1986).
The high RBC pH together with the strong Bohr eVect shifts the P50 of Hb to
the region of active teleosts. If RBC pH was as low as in a rainbow trout at the
same temperature, P50 in L. fluviatilis would be as high as 73 mmHg
(Nikinmaa, 1993). The same study showed that at low RBC pH (6.8), brought
about by very high PCO2 values (152 mmHg), L. fluviatilis Hb failed to become
saturated even at PO2 values of 530 mmHg and that Hb‐O2 saturation at PO2
152 mmHg was below 40%, indicating an eVect analogous to the Root eVect
in ray‐finned fishes. Swim bladders are absent in lampreys and no choroid
rete mirabile has been found (Wittenberg and Haedrich, 1974).
    Under physiological conditions, catecholamines do not significantly
activate the NHE in RBCs of P. marinus and L. fluviatilis (Tufts, 1991; Virkki
and Nikinmaa, 1994).
    RBCs of the Pacific lamprey, L. (formerly Entosphenus) tridentatus,
contain high levels of ATP and 2,3‐DPG and a small amount of GTP
(Johansen et al., 1973; Bartlett, 1982). However, Hb‐O2 binding in this
species is not aVected by 2,3‐DPG (Johansen et al., 1973), which is in line
with the absence of the typical binding site in the crevice of deoxygenated
tetramers, where organic phosphates bind in jawed vertebrate Hbs.
    Lamprey hemolysates often contain multiple Hbs, several of which have
been sequenced. The mechanism of the strong Bohr eVect has been elucidated
in a major Hb component of P. marinus. Structural X‐ray crystallography and
functional studies on mutant Hbs obtained by site‐directed mutagenesis have
revealed that, unlike in Hbs of all other vertebrate classes, the Bohr eVect is
brought about by preferential proton binding in the deoxygenated Hb to a
cluster of glutamic acid residues in the interface of the dimeric Hb and to the
distal histidine of the heme group (Heaslet and Royer, 1999; Qiu et al., 2000).

   b. Hagfishes (Myxiniformes). Blood O2 transport in the hagfish
E. cirrhatus was studied by Wells et al. (1986; Table 5.1). At 16  C, they
found a high normoxic resting PaO2 of about 90–110 mmHg, which was
not reduced during 10–15 min of swimming at 0.4 body lengths per second.
PvO2 was rather high at 17.2 mmHg, falling to 3.5 mmHg during swimming.
PCO2 was low in arterial and venous blood, as typical for water‐breathers, and
pH was 7.92 in arterial and 7.77 in venous blood. Arteriovenous pH diVerence
252                                           C. J. BRAUNER AND M. BERENBRINK

slightly increased to 0.25 pH units during swimming. Hematocrit value and
Hb concentration were comparatively low relative to lampreys and jawed
vertebrate fishes (Table 5.1), and physically dissolved O2 composed 15% of
normoxic arterial O2 content. In vitro, whole‐blood O2 equilibrium curves
were slightly sigmoid with nH ¼ 1.38. The latter did not change with satura-
tion or pH. At pH 7.8, P50 was 12.3 mmHg and F ¼ À0.43 (Wells et al., 1986)
(Table 5.1).
    These data are quite diVerent from observations on the congeneric Pacific
hagfish Eptatretus (formerly Polistotrema or Bdellostoma) stoutii (Manwell,
1958, 1963; Johansen and Lenfant, 1972). Whole blood of the latter species
did not show a Bohr eVect at physiological blood pH. Above pH 7.0, O2
binding was strictly hyperbolic (nH ¼ 1.0) and P50 was 2–4 mmHg between
5 and 18  C and was unaVected by CO2 at constant pH (Manwell, 1958;
Johansen and Lenfant, 1972) (Table 5.1). Similarly, Manwell (1963) reported
a high blood O2 aYnity in Atlantic hagfish (M. glutinosa) RBCs, very low
cooperativity, and no Bohr eVect. From a crude visual estimation of arterial
and venous Hb‐O2 saturation in exposed dorsal aortic and liver vessels in
restrained animals, Manwell (1958, 1963) concluded that hagfish Hb essen-
tially functions over a restricted PO2 range of maximally 10 mmHg, with very
low venous PO2 values.
    It is possible that these diVerences are related to diVerences in the lifestyle
and environmental PO2 of the species, with some preferring rocky substrates
and others living in muddy burrows (Wells et al., 1986). However, methodo-
logical diVerences also need to be taken into account since Manwell (1958,
1963) used RBCs in phosphate buVers, whereas Wells et al. (1986) used whole
blood, pH‐adjusted with CO2. The above diVerence is important since
E. stoutii RBCs appear to lack the anion exchanger (Ellory et al., 1987).
Thus, extracellular pH changes with the membrane permeable CO2 are more
likely to change RBC pH than those with phosphate buVers. In M. glutinosa
whole blood in vitro, extracellular pH changes caused by changes in PCO2 are
transferred to the RBC interior, but intracellular pH is much lower than
in lampreys (Tufts and Boutilier, 1990). In further contrast to the river
lamprey, no RBC pH regulation could be observed after intracellular acidifi-
cation (Nikinmaa et al., 1993). Studies on the isolated multiple Hbs of
M. glutinosa have further shown that the very small Bohr eVect of F ¼ À0.07
can be increased to F ¼ À0.17 in the presence of CO2 (Bauer et al., 1975). These
authors showed a specific CO2 eVect, which decreased Hb‐O2 aYnity
at constant pH. However, these experiments were performed at low
chloride concentrations, which have been shown to facilitate the Bohr eVect
of M. glutinosa Hb (Fago and Weber, 1998). The CO2 eVect described by Bauer
et al. (1975) has later been ascribed to an interaction between bicarbonate ions
and Hb (Fago et al., 1998).
5.   GAS TRANSPORT AND EXCHANGE                                               253

    RBCs of E. stoutii contain high ATP and ADP levels, together with small
amounts of GTP (Bartlett, 1982). However, in the presence of chloride at
least ATP, 2,3‐DPG, and inositol hexaphosphate do not aVect O2 aYnity or
cooperativity in isolated M. glutinosa Hb (Bauer et al., 1975; Fago and
Weber, 1995).


4.1. General Model of CO2 Transport and Excretion

    There have been a number of reviews on CO2 excretion in vertebrates in
general (Swenson and Maren, 1978; Klocke, 1987, 1988; Swenson, 1990) and
fish in particular (Perry, 1986; Tufts and Perry, 1998) to which the reader is
referred; however, the basic teleost model is briefly discussed here. As in all
vertebrates, the majority of total CO2 is transported in the blood as HCOÀ in3
fish. At the tissues, CO2 diVuses down its partial pressure gradient into the
blood, and once within the RBC is converted to HCOÀ and Hþ in the presence
of high levels of carbonic anhydrase (CA). There is no plasma accessible CA in
the gill of teleosts, and thus CO2 hydration/dehydration in the blood is restrict-
ed to within the RBC. In general, HCOÀ is transported across the RBC anion
exchanger (AE1) in exchange for ClÀ (with the exception of hagfishes and
lampreys discussed below), and the majority of HCOÀ is transported in the
plasma compartment. A varying proportion of the protons resulting from CO2
dehydration are then bound to Hb, either through the ability of Hb to act as a
buVer or through the binding of protons associated with deoxygenation (Hal-
dane eVect), minimizing the pH changes that occur within the RBC during CO2
loading. In general, nonoxygenation‐dependent Hþ buVer sites in Hb under
physiological conditions consist predominantly of histidine residues (Jensen,
1989), and in some cases the terminal amines of the a‐ and b‐chains. In general,
animals appear to have Hbs with relatively high buVer values at fixed oxygena-
tion status and small Haldane coeYcients, as in mammals and elasmobranchs,
or low buVer values and large Haldane coeYcients, as in teleost fishes (Jensen,
1989). These two patterns represent two diVerent, but equally eVective strate-
gies for CO2 transport and excretion (Jensen, 1989). Despite the binding
potential of Hb for Hþ, there is often a reduction in pH during blood tissue
transit due to metabolic CO2 production, that in conjunction with the Bohr
eVect (see above) facilitates O2 delivery to the tissues. This has been well
described in the blood of vertebrates (Bartels, 1972; Lapennas, 1983;
Riggs, 1988). Lapennas (1983) has conducted a theoretical analysis which
concludes that the optimal Bohr coeYcient for O2 delivery at the tissues is
one‐half of the respiratory quotient (rate of CO2 production/rate of O2
254                                        C. J. BRAUNER AND M. BERENBRINK

consumption by the tissues) which in most animals is between À0.35 and À0.5.
This value represents a compromise between the sensitivity of the Hb to
changes in pH (Bohr eVect) and the net acidification in the blood associated
with metabolic CO2 production and the binding of protons to Hb during
deoxygenation (Haldane eVect).
    The reverse process occurs at the gas exchange site responsible for CO2
elimination (gills or ABO), where the majority of CO2 excreted consists
of HCOÀ transported from the plasma into the RBC (Perry et al., 1982;
Brauner et al., 2000) (via AE1) where it combines with Hþ to form CO2 and
diVuses down its concentration gradient to the environment. Again, protons
are generally either titrated from Hb or released from Hb during oxygenation
(Haldane eVect), and HCOÀ dehydration occurs at a rapid rate in the presence
of high RBC levels of CA. In fish which possess a low‐Hb buVer value and
large Haldane eVect (i.e., teleosts), there is a tight interaction between O2
uptake and CO2 removal at the gills (Jensen, 1986, 1989; Brauner et al., 1996;
Brauner and Jensen, 1998; Brauner and Randall, 1998). The rate‐limiting
step in CO2 excretion is thought to exist at the RBC AE1 (Perry, 1986), and
when all steps associated with CO2 excretion are considered, the gills of fish
are thought to be diVusion limited for CO2 (Perry and Gilmour, 2002).
    A number of aspects that influence the pattern of CO2 transport and
elimination in primitive fish groups diVer from teleosts and air‐breathing
vertebrates. These include aspects related to (1) Hb and RBC function, in
particular Hb buVer values and the magnitude of the Haldane eVect which
influence the degree of interaction between O2 and CO2 exchange (Brauner
and Randall, 1996; Brauner and Randall, 1998), as well as RBC permeability
to HCOÀ via AE1, and (2) the catalytic activity and location of CA. The
limited research conducted to date on these two aspects in primitive fishes is
discussed below.

4.1.1. Hemoglobin and RBC Function
    a. Ray‐finned fishes (Actinopterygii)
    i. Bowfin (Amiiformes). Blood bicarbonate concentration at physio-
logical pH and PCO2 is 6–7 mM at 20  C and corresponds to the typical values
for water‐breathers (Johansen et al., 1970). The latter authors noted a non-
bicarbonate buVering capacity similar to the blood of other primarily water‐
breathing fishes. The specific buVer capacity of the composite Hb system bHb
is À6.4 mol Hþ per mol Hb4 and pH, a value distinctly lower than in other
primitive ray‐finned fishes such as L. platyrhincus, A. ruthenus, and E. cala-
baricus, but very similar to many teleosts such as dolphin fish, rainbow trout,
and European eel (Coryphaena hippurus, O. mykiss, and Anguilla anguilla;
Berenbrink et al., 2005) (Table 5.1; Figure 5.2F). A. calva RBCs appear to
possess a high rate of ClÀ/HCOÀ exchange, and in conjunction with the
5.   GAS TRANSPORT AND EXCHANGE                                             255

presence of intermediate levels of RBC CA activity (relative to agnathans
and teleosts), and the presence of a plasma CA inhibitor, it is thought that
blood CO2 transport is similar to that of teleost fishes (Gervais and Tufts,
     ii. Gars (Lepisosteiformes). In longnose gar (L. osseus) exposed to 10  C,
PCO2 and bicarbonate concentration of ventricular blood were low as
expected for strict water‐breathers and average pH was 7.831 (Rahn et al.,
1971). In animals at 25  C, pH was 7.440, which is lower than predicted
by the general temperature–pH relationship of ectotherms (À0.016/ C,
Ultsch and Jackson, 1996). The low pH at 25  C was primarily due to an
elevation in PCO2 associated with air breathing (Rahn et al., 1971). In L.
platyrhincus, bHb was À8.6 mmol Hþ/mmol Hb4 and pH. While the Bohr
eVect in these Hbs has not been measured directly, the Haldane coeYcient in
organic phosphate‐free hemolysates at physiological temperature, pH, and
chloride concentrations was F ¼ À0.525. Assuming numerical identity be-
tween the Bohr and Haldane coeYcients, the magnitude of the Haldane eVect
is lower, and bHb higher than in Hbs of A. calva and most teleosts under the
same conditions (Table 5.1; Figure 5.2) (Berenbrink et al., 2005).
     iii. Sturgeons and Paddlefishes (Acipenseriformes). The Adriatic stur-
geon A. naccarii is the only primitive ray‐finned fish whose globins have at
least been partially sequenced (Clementi et al., 1999). These partial sequences
are maximally 30 amino acids long but already contain some very interesting
information that sets them apart from teleosts. In contrast to virtually all
teleosts, the N‐terminal amino groups of the a‐globins are not acetylated and
hence are available for binding of CO2 as carbamino compound or for
buVering of protons. The number of physiological buVer groups is further
increased by the presence of histidines in position 9 of both the a‐ and
b‐chains and in position 2 of the b‐chains. In teleosts, these positions are
usually occupied by amino acids that do not buVer at physiological pH. This
is consistent with a distinctly higher bHb of À9.4 mol Hþ per mol Hb4 and pH
in A. ruthenus as compared to L. platyrhincus, A. calva, and most teleosts
(Table 5.1; Figure 5.2F) (Berenbrink et al., 2005).
     iv. Reedfish and Bichirs (Polypteriformes). No measurements of in vivo
blood CO2 transport characteristics exist for this group. However, in vitro
characterization of the blood of P. senegalus reveals a typical CO2‐combining
curve, where the majority of HCOÀ is held within the plasma compartment
(Vokac et al., 1972). The whole blood buVering capacity was calculated to be
À15.4 mmol HCOÀ per liter and per pH unit (Vokac et al., 1972). Hemoly-
sates of the reedfish E. calabaricus show a bHb value of À11.6 mmol/mmol
Hb4 and pH, which is similar to values in many tetrapods and elasmo-
branchs, but higher than the values in more advanced ray‐finned fishes,
especially those in many teleosts (Table 5.1; Figure 5.2F) (Berenbrink




                                                                                                                                                                               Primitive fishes
                                                                                                                                                              regulatory NHE
                                                                                                                                              rete mirabile

                                                                                                                                                                 RBC pH

                                                       (mol H+ per mol Hb4 per pH)   Fixed acid Haldane effect         Root effect (%)
                                                                                       (mol H+ per mol Hb4)











                                  Myxine glutinosa


                                Lampetra fluviatilis

                              Scyliorhinus stellaris
Fig. 5.2. (continued)

                                  Mustelus asterias
                                Squalus acanthias
                             Coryphaena hippurus

                             Oncorhynchus mykiss

                                   Cyprinus carpio

                               Scleropages jardinii


                                                                                                                                                                                                                                                                 C. J. BRAUNER AND M. BERENBRINK
                                        Amia calva

                         Lepisosteus platyrhincus


                               Acipenser ruthenus

                        Erpetoichthys calabaricus
                              Latimeria chalumnae


                             Lepidosiren paradoxa

                                   Xenopus laevis

                         Alligator mississippiensis


                                      Gallus gallus
                                        Sus scrofa
                                    Homo sapiens
5.   GAS TRANSPORT AND EXCHANGE                                                                                                                                                                                                                       257











                                   b c
                                       e                                               e
                                                                                           c d

                                                                                                                                                                                                   a         - Fixed-acid Haldane
                                                                                                                                                                                                               effect increases
                                                                                                                                                                                                   b         - Root effect increases
                                                                                                                                                                                                   c         - Hb buffer value
                                                                                                                                                                                                   d         - Choroid rete mirabile
                                                                                                                                                                                                   e         - RBC pH regulation

Fig. 5.2. Selected blood O2 and CO2 transport characteristics and other relevant factors in
representative primitive fishes and other vertebrates. (A) Number of primitive fish species com-
pared to other vertebrate groups. (B) Presence (solid circles) of RBC pH regulatory Naþ/Hþ
exchange. (C) Presence (solid circles) of a choroid rete mirabile. (D) Root eVect in RBCs or Hb
solutions, expressed as maximal percentage deoxygenation induced by decreased pH at PO2 ¼
150 mmHg compared to pH 8 or higher. (E) Fixed‐acid Haldane eVect, expressed as maximal
number of protons taken up per Hb tetramer on changing the oxygenation status of hemolysates
at constant pH from fully oxygenated to fully deoxygenated in the absence of organic phosphates
and at physiological chloride concentrations. (F) Specific Hb buVer value, bHb, expressed as mol
protons released per mol Hb tetramer on a unit increase in pH, obtained in deoxygenated
hemolysates in the absence of organic phosphates and at physiological chloride concentrations
and pH. (G) Reconstruction of the evolution of blood respiratory gas transport characteristics
on a vertebrate phylogenetic tree (for detailed methods see Berenbrink et al., 2005). Dashes
through tree branches indicate the level at which certain characteristics (a–e) were gained.
References: (A) Nelson (1994); (B) Berenbrink et al. (2005) and Wittenberg and Haedrich
(1974); (C) Berenbrink et al. (2005), Nikinmaa (1986), and Nikinmaa et al. (1993);
(D) Berenbrink et al. (2005) and Nikinmaa (1993); (E and F) Berenbrink et al. (2005), Jensen
(1999), Jensen et al. (1998b), and Siggaard‐Andersen (1975); and (G) phylogenetic tree assembled
from information given in Berenbrink et al. (2005) and Takezaki et al. (2003). ‘‘?’’ indicates no
data. See text for further details.

et al., 2005). The Haldane coeYcient in P. senegalus calculated from in vitro
CO2 dissociation curves was determined to be À0.10 vol% CO2 per vol% O2
capacity (Vokac et al., 1972) and the Bohr coeYcient measured in the same
study was F ¼ À0.43. The latter value is low compared with values observed
in most teleosts, and this occurs in conjunction with relatively high buVer
258                                        C. J. BRAUNER AND M. BERENBRINK

    b. Lobe‐Finned Fishes (Sarcopterygii)
    i. Coelacanths (Coelacanthiformes). The physiological pH range of
L. chalumnae blood is not known. Assuming that the general relationship
between resting arterial pH and body temperature in elasmobranchs and
marine teleosts (Ultsch and Jackson, 1996) holds, values of 7.85 to 7.75 can
be expected at the habitat temperatures of 13–25  C, respectively, that were
measured by Fricke and Hissmann (2000). The limited work conducted on
aspects related to CO2 transport in L. chalumnae indicates that it has a small
to moderate Haldane eVect (F ¼ À0.5; Bonaventura et al., 1974) and a fairly
high‐Hb buVer value based on the total number of histidine residues (36 per
tetramer) (Gorr et al., 1991b). Wood et al. (1972) determined a whole blood
nonbicarbonate buVer value of À9.0 mmol HCOÀ per liter and pH unit. The
primary sequence of Latimeria globin chains allows an estimation of the
number of physiological buVer groups per tetrameric Hb and this can be
used to predict bHb (Berenbrink et al., 2005). Using the number of predicted
physiological buVer groups per Hb tetramer and the regression line obtained
by phylogenetically independent contrast from Berenbrink (2006), bHb is
calculated to be À11.9 mmol Hþ per mmol Hb4 and pH. This value is twice
as high as in the bowfin A. calva and typical teleosts, somewhat lower than
the value measured for the South American lungfish, but in the same range as
the values measured for various tetrapods, elasmobranchs, and E. calabar-
icus as a member of the most basal living actinopterygian lineage (Table 5.1;
Figure 5.2F) (Berenbrink et al., 2005). Thus, in terms of CO2 transport and
excretion, L. chalumnae possesses characteristics that are more similar to that
of elasmobranchs than teleosts (Jensen, 1989), namely a high Hb buVer value
and moderate Haldane eVect.
    ii. Lungfishes (Dipnoi). In L. paradoxa, the Haldane eVect is small and
the buVer value of oxygenated blood was high in relation to typical water‐
breathers (Johansen and Lenfant, 1967). This has also been found in
P. aethiopicus (Jensen et al., 2003). Rodewald et al. (1984) sequenced the
globin chains of the single major Hb of L. paradoxa. The N‐terminal amino
groups are free and thus available for proton buVering and CO2 binding,
unlike in most fish investigated to date. The Hb contains 50 histidines per
tetramer (Rodewald et al., 1984) and the total number of physiological buVer
groups has been estimated to be 40 per tetramer, which corresponds to one of
the highest bHb values measured so far of À16.5 mmol Hþ per mmol Hb4 and
pH (Berenbrink et al., 2005).
    In the African lungfish P. aethiopicus, the RBC volume approaches
7000 mm3 (Koldkjær et al., 2002; Jensen et al., 2003), more than 30‐fold
larger than that commonly observed in teleost fishes (200 mm3; Jensen and
Brahm, 1995). The plasma compartment in the blood of P. aethiopicus
contains the highest concentration of total CO2 in the blood. Thus, CO2
5.   GAS TRANSPORT AND EXCHANGE                                              259

excretion is likely to be very dependent on RBC ClÀ/HCOÀ exchange, the
rate of which may be limited by the large size of lungfish RBCs. The rate
coeYcient for unidirectional RBC ClÀ eZux is much slower than that
measured in rainbow trout (O. mykiss) (Jensen et al., 2003). No studies
have been conducted to determine the degree to which the large cells may
aVect CO2 excretion in vivo.
    c. Jawless Fishes (Agnatha). Both lampreys and hagfishes appear to
virtually lack RBC anion exchange (Ellory et al., 1987; Nikinmaa and
Railo, 1987), which greatly alters the general pattern of CO2 excretion from
most other vertebrates. River lampreys (L. fluviatilis) have Hbs with relative-
ly low buVer values (1–1.5 titratable groups per monomer due to a low
number of histidines in the Hb) but a relatively large Haldane eVect
(0.9 Hþ per monomer on deoxygenation). In contrast, the Atlantic hagfish
(M. glutinosa) exhibits the reverse pattern, with relatively high‐Hb buVer
values (four to five titratable groups per monomer) and a relatively small
Haldane eVect (0.35 Hþ per monomer on deoxygenation) (Jensen, 1999)
(Table 5.1; Figure 5.1E).
    In the river lamprey, these Hb–Hþ equilibrium characteristics result in a
very interesting system that helps to partially compensate for the lack of an
RBC AE1 that would otherwise be assumed to limit the rate of CO2 excretion
in vivo. It must be remembered that the amount of HCOÀ formed for a given
PCO2 increases with increasing intracellular pH. Because of the large Hþ
binding associated with deoxygenation in conjunction with a low Hb buVer
value, deoxygenation results in an alkalinization of the RBC by 0.3–0.4 pH
units, and venous pHi is greater than arterial pHi in vivo (Tufts et al., 1992).
Furthermore, RBC membrane Naþ/Hþ exchange maintains the RBC pHi for
a given pHe at a more alkaline level than in most other fishes (Tufts et al.,
1992; Nikinmaa, 1997). Thus, the relatively alkaline RBC results in a greatly
elevated HCOÀ concentration within the erythrocyte (Tufts and Boutilier,
1989), and the majority of CO2 excreted at the gills will consist of HCOÀ       3
carried within the RBC that combines with Hþ released from Hb during
oxygenation. Thus, in terms of CO2 transport and excretion, there is a tight
coupling of CO2 excretion with O2 uptake in the river lamprey, that is
accomplished in the absence of AE1.
    The pHi in the Atlantic hagfish M. glutinosa is not elevated as it is in river
lampreys, and the Hb–Hþ characteristics appear less eVective in facilitating
CO2 removal in the absence of AE1. However, the Hb of the Atlantic hagfish
exhibits oxygenation‐linked binding of HCOÀ (Fago and Weber, 1998; Fago
et al., 1998) analogous to that observed in crocodiles, where deoxygenated
Hb binds HCOÀ . Thus, during tissue capillary transit, HCOÀ could be bound
                 3                                           3
to Hb as O2 is delivered, and transported within the RBC until blood reaches
the gills. Oxygenation‐dependent HCOÀ binding does not exist in lamprey
260                                       C. J. BRAUNER AND M. BERENBRINK

Hb, and thus there appear to be two very diVerent strategies to compensate
for the lack of RBC AE1 in Atlantic hagfish and river lamprey, which likely
reflects the early evolutionary divergence of the two groups.
    The lack of RBC AE1 in both lampreys and hagfishes indicates that the
RBC may have originally evolved to transport both O2 and CO2 (Tufts and
Boutilier, 1989). In the absence of appreciable RBC HCOÀ permeability,
extracellular acid loads cannot be buVered by Hb within the RBC
(Nikinmaa, 1997), and the intra‐ and extracellular compartments within the
blood functionally operate as two separate compartments. With the incor-
poration of AE1 into vertebrate RBCs, extracellular protons can be buVered
by Hb with the export of HCOÀ into the plasma, and the two‐compartmental
system results in the majority of HCOÀ being transported in the plasma, and
the total CO2 content of blood for a given PCO2 , to be elevated.
4.1.2. CA Activity and Location
    CA is a ubiquitous enzyme that catalyzes the hydration/dehydration of
CO2 and is involved in processes related to CO2 transport and excretion
(as described above), but also ion regulation, acid‐base balance, and fluid
secretion among others. CA is found in very high concentrations within the
erythrocyte of all vertebrates, second only in concentration to Hb, and plays
a crucial role in CO2 transport and excretion. In teleost fishes, there is no
plasma accessible CA in the gill, and thus HCOÀ dehydration at the gills is
restricted to within the RBC. In the elasmobranchs (Gilmour et al., 2001),
CA accessible to plasma is present in the gills, and there is significant non‐
RBC HCOÀ dehydration associated with CO2 excretion. CA is also found in
the endothelium of the ABO of many fishes, and this is also thought to play
an important role in CO2 excretion from the ABO. The following sections
describe what is known about CA characteristics in the RBCs, gills, and
ABOs of primitive fishes.

    a. RBC CA. It has been proposed that there may be a trend from
agnathans to teleosts toward a faster RBC CA isozyme (Tufts et al., 2003).
However, it was cautioned that this trend was based on comparison of a few
broad phylogenetic groups from diVerent studies and using diVerent meth-
odologies. In a later study, the single cytoplasmic CA isozyme expressed in
numerous tissues in the sea lamprey (P. marinus) exhibited a turn‐over rate
that was similar to that of teleosts such as the rainbow trout (Esbaugh and
Tufts, 2006). Furthermore, the amino acid sequence of the active site pocket
of the lamprey RBC CA isozyme is very similar to that of rainbow trout,
which is known to be a high‐turnover isozyme. These data indicate that the
lamprey RBC CA isozyme is in fact a high‐turnover isozyme, and thus a
high‐activity CA isozyme was present in RBCs early in vertebrate evolution.
5.   GAS TRANSPORT AND EXCHANGE                                              261

On the basis of the amino acid sequences of the putative active site, as well as
the hydrophobic core of RBC CA, the emerging consensus is that RBC CA is
highly conserved during vertebrate evolution. For example, relative to mam-
malian CA VII, the active site of lamprey (Esbaugh and Tufts, 2006), gar
(Lund et al., 2002), and rainbow trout (see Esbaugh and Tufts, 2006) all only
diVered by 3 of 36 amino acids. Biochemical measurements on RBC CA
activity in lamprey yield a value, an order of magnitude lower than that of
rainbow trout; however, this could be accounted for solely by the lower CA
concentration within lamprey RBCs (Esbaugh et al., 2004). The fact that
large diVerences in RBC CA concentration exists among the few primitive
fish groups investigated to date [i.e., sea lamprey (P. marinus) < A. calva <
O. mykiss (Gervais and Tufts, 1999; Esbaugh et al., 2004)] is interesting given
that CA levels appear to be far in excess of that thought to be required in vivo.

    b. Gill CA. High levels of CA are found in the gills of most fishes studied
to date. However, for the most part, it appears to be restricted to the
cytoplasm where it plays an important role in ion and acid–base regulation
by catalyzing the conversion of CO2 to HCOÀ and Hþ; crucial counterions
for ClÀ and Naþ exchange in particular (Perry and Laurent, 1990; Randall
and Brauner, 1998). In elasmobranchs, membrane‐bound CA that is plasma
accessible in the gills has been identified, and there is significant non‐RBC
HCOÀ dehydration associated with CO2 excretion (Gilmour et al., 2001).
Plasma accessible CA in the gills of the dogfish Squalus acanthias is also
thought to be important in equilibrating postbranchial CO2 and Hþ, for
chemoreceptors regulating ventilation (Gilmour et al., 1997; Henry et al.,
1997). In general, very little is known about the subcellular localization of
CA within the gills of primitive fishes. In the sea lamprey, P. marinus, there
does not appear to be any plasma accessible membrane‐bound CA in the gill
(Henry et al., 1993). In A. calva, subcellular fractionation indicates that the
majority (>97%) of CA exists within the cytosol, and the small amount
associated with membranes was not likely membrane bound (Gervais
and Tufts, 1998). Clearly a great deal of more research is required to deter-
mine whether elasmobranchs represent the only fish group to have plasma
accessible CA in the branchial circulation.

    c. CA in the ABO. There have been a number of studies investigating the
possible role of CA in aerial CO2 excretion in the ABO of fishes. In the lungs
of air‐breathing vertebrates, plasma accessible CA is available through a
membrane‐bound CA (CA IV), which is anchored to the extracellular lumi-
nal surface of capillary lung endothelium (Heming et al., 1993), and this is
thought to be the case in all air‐breathing animals (Stabenau and Heming,
2003). While extracellular HCOÀ dehydration is thought to play a role in
262                                         C. J. BRAUNER AND M. BERENBRINK

CO2 excretion during lung capillary transport, the low plasma buVer value,
and therefore lack of Hþ availability for HCOÀ dehydration, limits the
degree to which this facilitates CO2 excretion (Bidani and Heming, 1991).
Current estimates in the lungs of mammals indicate that less than 10% of
total CO2 excretion may be associated with plasma accessible CA.
     Membrane‐bound CA has been identified in the ABO of bowfin. Inter-
estingly, the membrane‐bound CA in the bowfin ABO was about three times
less sensitive to the plasma CA inhibitor than RBC CA (Gervais and Tufts,
1998), and thus plasma CA inhibitors may be more eVective at scavenging
and inhibiting CA released from lysed RBCs while minimally aVecting
membrane‐bound CA as has been observed in mammals (Heming et al.,
1993). In the lungfish P. dolloi, injection of an impermeant CA inhibitor
that was hypothesized to inhibit any plasma accessible CA did not signifi-
cantly aVect CO2 excretion rate into water or air, or alter arterial PCO2 or pH.
Although it is not known whether plasma accessible CA exists in this species,
if it does, it does not appear to contribute to CO2 excretion (Perry et al.,
2005), and in general, the degree to which ABO CA influences aerial CO2
excretion remains controversial (Graham, 1997). Membrane‐bound CA in
the lungs of vertebrates has also been proposed to ensure complete pH/PCO2
equilibration during blood capillary transit (Henry and Swenson, 2000),
which may be important for the control of ventilation; however, this remains
to be investigated in primitive fishes.


5.1. How Do Primitive Fishes Compete with Other Fishes?

   In terms of species numbers, living primitive fishes are dwarfed by the
enormous species richness of their extant relatives, which had the same or
even less time available for speciation. Thus, the 84 agnathan, 8 primitive
lobed‐finned, and 44 primitive ray‐finned species are opposed by the at least
500‐fold more speciose gnathostomes, tetrapods, and teleosts, respectively
(Figure 5.2A).
   A priori species number is not necessarily a measure of the superiority of
some structure or function in one group over that in another group since
highly asymmetric species numbers in two sister groups can also evolve in
simulations under a model of random speciation and extinction (Slowinski
and Guyer, 1993).
   However, in terms of structural and functional diversity and habitats
occupied, extant teleosts and tetrapods clearly outcompete living primitive
ray‐finned and lobe‐finned fishes, respectively, and the same is true comparing
5.   GAS TRANSPORT AND EXCHANGE                                                263

all these jawed vertebrates with the living agnathans. After emerging onto
land, the new environment presumably allowed adaptive radiation and the
explosion of species numbers in tetrapods, compared to aquatic lobefins.
Another adaptive radiation (or several successive ones) occurred in the aquat-
ic habitat, giving rise to the current diversity of teleosts fishes, perhaps related
to an additional whole‐genome duplication event in teleosts as compared
to all other vertebrates (see below). The question arises: to what extent
does the respiratory physiology of living primitive fishes contribute to
their survival, given the competition by teleosts and by terrestrial, aerial,
and secondarily aquatic lobe‐fins, such as man, birds, and seals, which prey
on them?

5.1.1. Primitive Ray‐Finned Fishes (Actinopterygii)
     In contrast to many bimodally breathing members of teleosts, air‐
breathing primitive fishes such as the Polypteriformes, A. calva, and the
genus Lepisosteus retain fully functional gills, which at times allow strictly
aquatic breathing over prolonged periods. This may be important to avoid
aerial or terrestrial predators lurking at the water surface (Graham, 1997;
Farmer, 1999). It is of additional importance in temperate zone fishes like
A. calva and L. osseus, where ice cover in winter may prevent air breathing
for months (Rahn et al., 1971). Alternatively, these two genera are also able
to support their M O2 almost completely aerially even under exercise in hot,
severely O2‐depleted water. Thus, Burleson et al. (1998) concluded that using
their ABO, ‘‘gar can maintain activity under . . . conditions that incapacitate
virtually every other fish in their environment.’’
     Sturgeons and paddlefishes are unusual among primitive ray‐finned fishes
in that they do not breathe air. Both groups have evolved a highly specialized
feeding mode, together with unusual ventilatory adjustments. Sturgeons are
able to ventilate their gills via water intake through the dorsal part of the gill
slit while their protrusible mouth is involved in suctional feeding, whereas
paddlefishes are continually moving filter feeders, which use ram ventilation
(see above). These specializations may contribute to their success despite
competition by teleosts.

5.1.2. Primitive Lobe‐Finned Fishes (Sarcopterygii)
    The extraordinary capacity of African lungfishes to estivate in cocoons
for months or even years in hard‐baked mud until favorable conditions are
encountered again allows them to inhabit desiccation‐prone swamp areas
and is unparalleled by teleosts. The South American lungfish L. paradoxa is
not known to produce a cocoon but also estivates in damp burrows when
water levels fall. Functional separation of deoxygenated systemic venous
blood and oxygenated pulmonary blood in the heart of lungfishes provides
264                                         C. J. BRAUNER AND M. BERENBRINK

their blood–gas transport system with an eYciency that is usually only found
in the single‐loop arrangement of pure water‐breathers or the double‐loop
arrangement found in birds and mammals. Interestingly, the swamp eel
Synbranchus marmoratus, perhaps the only teleost able to compete with
Lepidosiren in its natural habitat, has also evolved a way to avoid central
mixing of oxygenated ABO blood and deoxygenated systemic blood, namely
by using its gills as an ABO (Graham, 1997).
    For the coelacanth, Fricke and Hissmann (2000) have suggested that a
low metabolic rate, together with a low‐energy drift‐hunting strategy and the
peculiar electroceptive rostral organ for prey detection, has enabled
L. chalumnae to successfully compete with modern ray‐finned fishes in a
hypoxic and low biomass deepwater environment. The extremely low P50
would also be beneficial during exposure to hypoxia, but these low values
need to be reconfirmed.
5.1.3. Jawless Fishes (Agnatha)
    Surprisingly, at least in one way both groups of living agnathans have
turned the emergence of teleosts and tetrapods to their advantage. Thus,
adult pacific lamprey L. tridentatus parasitize on teleosts and sperm whales
(Hart, 1973), and, when given the opportunity, the Atlantic hagfish
M. glutinosa feeds on teleostean, avian, and mammalian carcasses, often
penetrating with head and anterior body into the carcass (Strahan, 1963).
In both cases, respiratory water intake via the mouth or nostril is impeded.
In lampreys, this can be compensated by unique tidal ventilation of gill
pouches via their external openings (Randall, 1972). Hagfishes with their
heads immersed in a carcass may rely on O2 uptake via the skin of their
elongated bodies (Strahan, 1958; SteVensen et al., 1984).
    Feeding on the evolutionary younger groups of teleosts and tetrapods is
clearly a secondary specialization, and fossils demonstrate that today’s jaw-
less fishes are at least structurally quite diVerent from their early agnathan
ancestors (Benton, 2000; Chapter 1, this volume) and thereby probably also
from the last common ancestor with jawed vertebrates.

5.2. Primitive Fishes and the Evolution of Vertebrate Blood
     O2 and CO2 Transport Characteristics

    Living primitive fishes occupy strategic positions in the evolutionary tree
of vertebrates and may shed light on the respiratory physiology of the first
vertebrates and the first Osteichthyes (Figure 5.2A). Reconstructing the
evolution of blood O2 and CO2 transport characteristics and their condition
in ancestral species from the condition in living species is a fascinating task
that may unravel general trends in physiological evolution but has its pitfalls.
5.   GAS TRANSPORT AND EXCHANGE                                          265

5.2.1. A Note of Caution
    In most cases, it is not clear to what extent phenotypic similarities and
diVerences between species are genetically based and therefore subject to
Darwinian evolution. Genetic components are diYcult to clarify because of
the enormous phenotypic plasticity of blood O2 and CO2 transport charac-
teristics and because diVerent experimental methods and approaches some-
times give significantly diVerent values for the same variable (see above
discussion of the Bohr eVect in lungfishes).
    Because the primary structure of Hb is undoubtedly genetically deter-
mined, one useful approach to unravel evolutionary trends in blood O2 and
CO2 transport is to relate Hb‐O2‐binding characteristics to its molecular
structure. Although the already large number of nucleotide and amino acid
sequences for fish globins seems to be ever increasing, there is a conspicuous
lack in globin sequences and associated structure–function analyses in primi-
tive fishes. No Hb sequences are available for primitive ray‐finned fishes such
as Polypteriformes, gars, bowfin, and even basal teleosts such as the Osteo-
glossiformes. Even the Elopomorpha are only covered by globins from the
highly modified Anguilliformes. Similarly among primitive lobe‐finned fishes,
no sequences are available for the Australian and African lungfish species.
    Another diYculty is that, whether due to insuYcient data or an attempt
to simplify, discussions of evolutionary trends in terms of vertebrate blood
O2 and CO2 transport characteristics are often limited to the larger sub-
groups, which are treated sequentially, such as from agnathans, elasmo-
branchs, teleosts to one or more groups of tetrapods (Lenfant et al., 1970;
Wood and Lenfant, 1987; Nikinmaa, 1990, 1997, 2001). A hidden danger of
such studies is that, even if unintended by the authors, they can easily be
interpreted in a teleological and anthropocentric way such that living mem-
bers of increasingly closer relatives to tetrapods (or mammals or man) are
interpreted to have changed little since they split from the last common
ancestor with tetrapods and are taken to represent a linear succession of
stages that the ancestors of tetrapods underwent before reaching their
present form.
    In strong contrast to such a view, increasing evidence points to an
additional, whole‐genome duplication event in the ray‐finned fish lineage as
compared to lobe‐finned fishes (including tetrapods) that may have provided
the genetic basis for the adaptive radiation and evolutionary novelties in
modern teleosts (Meyer and Schartl, 1999; Hoegg et al., 2004; Crow et al.,
2006; but see Donoghue and Purnell, 2005).
    Yet, although not explicitly stated, an inherent assumption in the older
literature appears to be that the fishlike ancestors of tetrapods went through
a teleost‐like state before they acquired the current blood O2 and CO2
266                                          C. J. BRAUNER AND M. BERENBRINK

transport characteristics. For example, it was held that to venture onto land,
the tetrapod ancestors had to reduce the strong CO2 sensitivity of blood O2
binding (i.e., the Bohr and Root eVects) typical of modern teleosts and had to
acquire a higher blood‐buVering capacity than found in teleosts (Carter,
1931; Root, 1931). Studies of primitive sarcopterygians (lungfishes and coe-
lacanth) and actinopterygians (Polypteriformes, sturgeons, gars, and bowfin)
now suggest that the last common ancestor of these groups had a high blood
buVer value and that the low‐Hb buVer capacity, the Root eVect, and the
exceptionally large Bohr eVect evolved only within the actinopterygians after
they split form the ancestor of tetrapods (see below; Berenbrink et al., 2005;
Berenbrink, 2006). Thus, the jawed vertebrate ancestors of air‐breathing
tetrapods never went through a teleost‐like state and were more likely to
show elasmobranch‐like blood O2 transport properties.
     Similarly, modern lungfishes are often seen as resembling the ancestors of
today’s tetrapods at a time when they first emerged onto land. Most scientists
now agree that lungfishes are the closest living relatives of tetrapods; how-
ever, that does not mean that lungfishes have survived unchanged since their
split from the lineage that gave rise to tetrapods. Thus, the first lungfishes
may have been non‐air‐breathing fishes of open marine habitats (Marshall
and Schultze, 1992; Cloutier and Ahlberg, 1996; Chapter 1, this volume).
Among living lungfishes, the Australian N. forsteri is the least dependent on
air breathing and does not estivate. In this regard, it may more closely
resemble ancestral forms than the obligate air‐breathing and estivating
African and South American lungfish genera Protopterus and Lepidosiren,
respectively. However, fossils of estivating lungfishes in cocoons can be dated
back to 286 mya (Graham, 1997), whereas the fossil record of the group
containing N. forsteri is dated back to only 245 mya (Schultze, 1993). Thus,
given the current uncertainty regarding the relationships of fossil lungfishes,
it is also possible that N. forsteri secondarily lost the ability to estivate and
changed from an obligate to a facultative air‐breather. This is supported
by their lung anatomy, whereby the dorsal unpaired lung in N. forsteri may
be derived from the more primitive arrangement of paired ventral lungs in
African and South American lungfishes (Schultze, 2003). Moreover, despite
the ventral location of the primitive lungfish lung and amphibian lungs,
anatomical diVerences suggest that the amphibian lung is not directly derived
from the organ in lungfishes (Perry and Sander, 2004).
     Thus, it appears possible that air breathing evolved independently in
modern lungfishes and tetrapods and that consequently the partially divided
heart with functional separation of pulmonary and systemic blood pathways
was also acquired independently.
     Another pitfall of this kind in the reconstruction of ancestral respiratory
physiology is illustrated by the use of modern amphibians to infer the mode
5.   GAS TRANSPORT AND EXCHANGE                                             267

of CO2 elimination in the first vertebrates emerging onto land in the Devonian.
It was proposed that much like present‐day amphibians these protoamphi-
bians eliminated CO2 via their moist skin and thereby avoided the general
respiratory acidosis otherwise occurring when due to a presumed general
decrease in ventilation rates the first terrestrial air‐breathers could no longer
readily eliminate CO2 (Howell, 1970; see also Ultsch, 1996). However, fossils
show that in contrast to modern amphibians, several protoamphibians were
heavily scaled and rather large, with a low surface area to volume ratio,
creating unfavorable conditions for CO2 elimination via the skin. Thus,
Ultsch (1996) proposed that protoamphibians living in hypercapnic waters
had already coped with elevated blood CO2 levels before they ventured onto
land by evolving ion‐regulatory mechanisms for an equivalent rise in plasma
    In the following discussion, it may be helpful to remind oneself of the
provocative but well‐supported statement that lungfishes share more derived
characters with—and therefore are more closely related to—a cow than to a
salmon (Gardiner et al., 1979).

5.2.2. Trends in the Evolution of Vertebrate Blood O 2 and CO 2
       Transport Characteristics
    Bearing these points in mind, we tentatively identify some general trends
in the evolution of vertebrate blood respiratory properties. It is generally
accepted that vertebrate life began in water and the blood respiratory pro-
perties of the earliest vertebrates were presumably shaped by the physico-
chemical consequences of water breathing such as the relative ease of
CO2 release into the water and the relatively low availability of O2. This
is reflected in the unanimously low CO2/bicarbonate levels and the high
intrinsic Hb‐O2 aYnity in virtually all present‐day water‐breathers, primi-
tive or modern. Lampreys are a curious exception because of their low
intrinsic Hb‐O2 aYnity. This may therefore be a secondary specialization.
As hypothesized by Wood and Lenfant (1987), the generally high intrinsic
O2 aYnity of vertebrate Hbs may be related to the lower atmospheric PO2
values prevalent in the Devonian, where several of the main vertebrate
lineages evolved.
    It is often presumed that the lack of stable tetrameric Hbs consisting of
two ab‐heterodimers in living agnathans represents the ancestral condition
of vertebrates (Wood and Lenfant, 1987). This seems plausible since a‐ and
b‐globins appear to have arisen by duplication of an ancient globin gene
(Goodman et al., 1975). However, if molecular evidence, which suggests that
living lampreys and hagfishes are not consecutive sister groups of extant
jawed vertebrates but rather together form the sister group of gnathostomes
268                                          C. J. BRAUNER AND M. BERENBRINK

(Takezaki et al., 2003; Chapter 1, this volume), is accepted, this is not the
single most parsimonious explanation. Under this scenario, a last common
ancestor of agnathans and gnathostomes with a2b2 tetramers is equally
    Similarly, lack of significant RBC AE in agnathans is considered the
primitive, ancestral state in the ancestor of agnathans and jawed vertebrates.
However, it should be remembered that, again accepting that lampreys
and hagfishes together are the sister group to gnathostomes, it is conceivable
that RBC AE was present in all early vertebrates and only secondarily lost in
an ancestor of living agnathans.
    In previous reviews, the first organophosphate that was used to modulate
vertebrate Hb function was considered to be ATP, which was found to be the
predominant organophosphate in some shark RBCs (Wood and Lenfant,
1987). However, GTP, which is a stronger eVector than ATP, is found in
higher concentrations than ATP in some other sharks and in a member of the
Holocephali, the sister group of elasmobranchs (Bartlett, 1978a). Evolu-
tionary analysis of RBC organic phosphate concentrations in a large number
of vertebrates confirms that both ATP and GTP were likely present in signi-
ficant amounts in the first jawed vertebrates (Val, 2000). Even some agnathans
contain traces of GTP next to large amounts of ATP, ADP, and, surprisingly,
2,3‐DPG (Johansen et al., 1973; Bartlett, 1982). Although we do not know
about ancestral vertebrates, organic phosphates in living agnathans have
only an unspecific anion eVect on the monomer–oligomer equilibrium in
hagfish Hb, and no eVect at all on lamprey Hb (Nikinmaa, 2001).
    Leaving speculations about the presence or absence of stable tetrameric
Hbs, AE1, or organophosphate eVects in RBCs of ancestral vertebrates
aside, Figure 5.2G illustrates the most parsimonious reconstructions of
some other blood gas transport properties, which are better substantiated.
These reconstructions are not aVected by the alternative positionings of
lampreys as the sister group to hagfishes or jawed vertebrates. They use
linear parsimony as the optimality criterion for reconstruction (Berenbrink
et al., 2005; Berenbrink, 2006).
    The fixed‐acid (i.e., CO2‐independent) Haldane eVect, DzHþ, is mechanis-
tically linked to the fixed‐acid Bohr eVect and numerical similar to À4F in the
absence of other interacting allosteric eVectors (Eq. 1). Its value is recon-
structed to be below 1.5 mol Hþ per mol Hb tetramer in the last common
ancestor of vertebrates (Figure 5.2G). This is still seen in M. glutinosa, living
sharks, and the non‐amniotic sarcopterygians L. paradoxa and X. laevis
(Figure 5.2E). Increased fixed‐acid Haldane eVects above 1.5 mol Hþ per
mol Hb tetramer evolved three times independently, namely in L. fluviatilis,
ray‐finned fishes, and to a lesser extent in amniotes (Figure 5.2E,G).
In primitive ray‐finned fishes, the evolutionary increase in the fixed‐acid
5.   GAS TRANSPORT AND EXCHANGE                                                   269

Haldane eVect is followed first by a gradual increase in the Root eVect and
then a gradual reduction in specific Hb buVer value. This was followed by the
evolution of a choroid rete mirabile and ocular O2 secretion in the last
common ancestor of A. calva and teleosts. Finally, the adrenergic NHE was
never present in primitive ray‐finned fishes or primitive teleosts and only
evolved in RBCs of advanced teleosts (Figure 5.2G) (Berenbrink et al.,
2005). No such evolutionary trends are observed in primitive lobe‐finned
fishes, which only very rarely possess Hbs that are more than 10% deoxygen-
ated at low pH and air equilibration as seen in X. laevis, have usually high Hb
buVer values, no choroid rete mirabile, and no pH regulatory NHE in their
RBCs (Figure 5.2G) (Berenbrink et al., 2005). However, in some advanced
tetrapods Hb amino acid sequence data suggest similar reductions in Hb
buVer values as seen in teleosts (Berenbrink, 2006).
     Interestingly, the elevated fixed‐acid Haldane eVect in lampreys is asso-
ciated with exactly the same changes in RBC physiology as seen in primitive
ray‐finned fishes. Thus, lamprey RBCs show a strong Root eVect, the lowest
ever measured specific Hb buVer value, and RBC pH regulation by an NHE,
although a choroid rete mirabile never evolved (Figure 5.2B–G) (Nikinmaa,
1986, 1993; Jensen, 1999). The sequence in which these diVerences evolved
from the condition in the ancestral vertebrates is not resolved.
     It is not known why this trend is not seen in advanced lobe‐fins (amniotes,
i.e., reptiles, birds, and mammals), which present the third case that shows an
(albeit modest) evolutionary increase in the fixed‐acid Haldane eVect.
Berenbrink et al. (2005) and Berenbrink (2006) have discussed diVerences
in the mechanism of the fixed‐acid Bohr eVect for the evolution of a Root
eVect in ray‐finned and lobe‐finned fishes and the implications of terrestrial
air breathing and the associated elevation in blood CO2/bicarbonate buVer
capacity for the evolution of specific Hb buVer values and RBC pH
     Lampreys and teleosts are among the most active aquatic vertebrates,
with lampreys in some rivers undertaking the same exhausting spawning
migrations, overcoming the same rapids and other obstacles as salmonids.
Little is known about the exercise physiology of sharks, but at least in
comparison with hagfishes and primitive lobe‐finned fishes, it appears that
the suite of respiratory blood–gas transport characteristics, which indepen-
dently evolved in lampreys and ray‐finned fishes, ideally poised them for
exercising in water.


    CJB was supported by an NSERC Discovery grant and MB’s research was supported by
funds from BBSRC, United Kingdom. We would like to thank Pia Koldkjær and Kim Suvajdzic
270                                                    C. J. BRAUNER AND M. BERENBRINK

for editorial assistance, Pia Koldkjær and Mathew Regan for commenting on an earlier version
of this chapter, and the reviewers for valuable criticisms and suggestions.


Amin‐Naves, J., Giusti, H., and Glass, M. L. (2004). EVects of acute temperature changes on
     aerial and aquatic gas exchange, pulmonary ventilation and blood gas status in the South
     American lungfish, Lepidosiren paradoxa. Comp. Biochem. Physiol. A Mol. Integr. Physiol.
     138, 133–139.
Babiker, M. M. (1979). Respiratory behaviour, oxygen consumption and relative dependence on
     aerial respiration in the African lungfish (Protopterus annectens, Owen) and an air‐breathing
     teleost (Clarias lazera, C.). Hydrobiologia 65, 177–187.
Babiker, M. M. (1984a). Adaptive respiratory significance of organophosphates (ATP & GTP)
     in air‐breathing fishes. Hydrobiologia 110, 339–349.
Babiker, M. M. (1984b). Development of dependence on aerial respiration in Polypterus‐
     Senegalus (Cuvier). Hydrobiologia 110, 351–363.
Baker, D. W., Wood, A. M., and KieVer, J. D. (2005). Juvenile atlantic and shortnose sturgeons
     (Family: Acipenseridae) have diVerent hematological responses to acute environmental
     hypoxia. Physiol. Biochem. Zool. 78, 916–925.
Bartels, H. (1972). The biological significance of the Bohr eVect. In ‘‘Oxygen AYnity of Hemo-
     globin and Red Cell Acid Base Status’’ (Rørth, M., and Astrup, P., Eds.), pp. 717–734.
     Munksgaard, Copenhagen.
Bartlett, G. R. (1978a). Water‐soluble phosphates of fish red cells. Can. J. Zool. 56, 870–877.
Bartlett, G. R. (1978b). Phosphates in red cells of two lungfish: The South American, Lepidosiren
     paradoxa, and the African, Protopterus aethiopicus. Can. J. Zool. 56, 882–886.
Bartlett, G. R. (1982). Phosphates in red cells of a hagfish and a lamprey. Comp. Biochem.
     Physiol. A Mol. Integr. Physiol. 73, 141–145.
Bassi, M., Klein, W., Fernandes, M. N., Perry, S. F., and Glass, M. L. (2005). Pulmonary oxygen
     diVusing capacity of the South American lungfish Lepidosiren paradoxa: Physiological
     values by the Bohr method. Physiol. Biochem. Zool. 78, 560–569.
Bauer, C., Engels, U., and Paleus, S. (1975). Oxygen binding to haemoglobins of the primitive
     vertebrate, Myxine glutinosa L. Nature 256, 66–68.
Beamish, F. W. H. (1973). Oxygen‐consumption of adult Petromyson marinus in relation to
     body‐weight and temperature. J. Fish. Res. Bd. Can. 30, 1367–1370.
Beitinger, T. L., Pettit, M. J., and Hutchison, V. H. (1985). Oxygen transfer characteristics of the
     blood of reedfish. Erpetoichthys calabaricus. Comp. Biochem. Physiol. A Mol. Integr.
     Physiol. 82, 553–558.
Benton, M. J. (2000).‘‘Vertebrate Palaeontology,’’ 2nd edn. Blackwell, Oxford.
Berenbrink, M. (2006). Evolution of vertebrate haemoglobins: Histidine side chains, specific
     buVer value and Bohr eVect. Respir. Physiol. Neurobiol. 154, 165–184.
Berenbrink, M., and Bridges, C. R. (1994a). Active Naþ, Cl–, and HCO3 ‐dependent acid

     extrusion in Atlantic cod red blood cells in winter activated by hypercapnia. J. Exp. Biol.
     192, 239–252.
Berenbrink, M., and Bridges, C. R. (1994b). Catecholamine‐activated sodium/proton exchange
     in red blood cells of the marine teleost. Gadus morhua. J. Exp. Biol. 192, 253–267.
Berenbrink, M., Koldkjær, P., Kepp, O., and Cossins, A. R. (2005). Evolution of oxygen secretion
     in fishes and the emergence of a complex physiological system. Science 307, 1752–1757.
Bidani, A., and Heming, T. A. (1991). EVects of perfusate buVer capacity on capillary CO2–
     HCO3–‐Hþ reactions: Theory. J. Appl. Physiol. 71, 1460–1468.
5.   GAS TRANSPORT AND EXCHANGE                                                                 271

Bird, D. J., Lutz, P. L., and Potter, I. C. (1976). Oxygen dissociation curves of the blood of larval
    and adult lampreys (Lampetra fluviatilis). J. Exp. Biol. 65, 449–458.
Black, E. C. (1940). The transport of oxygen by the blood of freshwater fish. Biol. Bull. 79,
Black, E. C., and Irving, L. (1938). The eVect of hemolysis upon the aYnity of fish blood for
    oxygen. J. Cell. Comp. Physiol. 42, 255–262.
Bohr, C., Hasselbalch, K. A., and Krogh, A. (1904). Uber einen in biologischer Beziehung
    wichtigen Einfluss, den die Kohlensaurespannung des Blutes auf dessen SauerstoVbindung
    ubt. Skand. Arch. Physiol. 16, 402–412.
Bonaventura, J., Gillen, R. G., and Riggs, A. (1974). The hemoglobin of the crossopterygian fish,
    Latimeria chalumnae (Smith). Arch. Biochem. Biophys. 163, 728–734.
Brauner, C. J., and Jensen, F. B. (1998). O2 and CO2 exchange in fish: The non‐linear release of
    Bohr/Haldane protons with oxygenation. In ‘‘Physiology and Biochemistry of the Fishes of
    the Amazon’’ (Val, A. L., and Almeida‐Val, V. M. F., Eds.). Vol. 2. INPA, Manaus.
Brauner, C. J., and Randall, D. J. (1996). The interaction between oxygen and carbon dioxide
    movements in fishes. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 113, 83–90.
Brauner, C. J., and Randall, D. J. (1998). The linkage between oxygen and carbon dioxide
    transport. In ‘‘Fish Physiology: Fish Respiration’’ (Perry, S. F., and Tufts, B., Eds.), Vol. 17,
    pp. 283–319. Academic Press, New York.
Brauner, C. J., and Val, A. L. (1996). The interaction between O2 and CO2 exchange in the
    obligate air breather, Arapaima gigas, and the facultative air breather, Lipossarcus pardalis.
    In ‘‘Physiology and Biochemistry of the Fishes of the Amazon’’ (Val, A. L., Almeida‐
    Val, V. M. F., and Randall, D. J., Eds.), pp. 101–110. INPA, Manaus.
Brauner, C. J., Gilmour, K. M., and Perry, S. F. (1996). EVect of haemoglobin oxygenation
    on Bohr proton release and CO2 excretion in the rainbow trout. Respir. Physiol. 106, 65–70.
Brauner, C. J., Thorarensen, H., Gallaugher, P., Farrell, A. P., and Randall, D. J. (2000). CO2
    transport and excretion in rainbow trout (Oncorhynchus mykiss) during graded sustained
    exercise. Respir. Physiol. 119, 69–82.
Bridge, T. W. (1904). The air‐bladder. ‘‘Cambridge Natural History,’’pp. 297–312. MacMillan
    and Co., Ltd., London.
Bridges, C. R., Pelster, B., and Scheid, P. (1985). Oxygen binding in blood of Xenopus leavis
    (Amphibia) and evidence against Root eVect. Respir. Physiol. 61, 125–136.
Brodal, A., and Fange, R. (1963). ‘‘The Biology of Myxine.’’ Scandinavian University Books, Oslo.
Burggren, W. W. (1978). Gill ventilation in the sturgeon, Acipenser transmontanus: Unusual
    adaptations for bottom dwelling. Respir. Physiol. 34, 153–170.
Burggren, W. W., and Bemis, W. E. (1992). Metabolism and ram gill ventilation in juvenile
    paddlefish, Polyodon spathula (Chondrostei, Polyodontidae). Physiol. Zool. 65, 515–539.
Burggren, W. W., and Johansen, K. (1986). Circulation and respiration in lungfishes (Dipnoi).
    J. Morphol. Supplement 1, 217–236.
Burggren, W. W., and Randall, D. J. (1978). Oxygen uptake and transport during hypoxic
    exposure in the sturgeon Acipenser transmontanus. Respir. Physiol. 34, 171–183.
Burggren, W. W., Dunn, J., and Barnard, K. (1979). Branchial circulation and gill morpho-
    metrics in the sturgeon Acipenser transmontanus, an ancient chondrosteian fish. Can.
    J. Zool. 57, 2160–2170.
Burleson, M., Shipman, B., and Smatresk, N. (1998). Ventilation and acid‐base recovery
    following exhausting activity in an air‐breathing fish. J. Exp. Biol. 201, 1359–1368.
Carter, G. S. (1931). Aquatic and aerial respiration in animals. Biol. Rev. 6, 1–35.
Clementi, M. E., Cataldi, E., Capo, C., Petruzzelli, R., Tellone, E., and Giardina, B. (1999).
    Purification and characterization of the hemoglobin components of Adriatic sturgeon
    (Acipenser naccarii) blood. J. Appl. Ichthyol. 15, 78–80.
272                                                    C. J. BRAUNER AND M. BERENBRINK

Clementi, M. E., De Rosa, M. C., Bertonati, C., Capo, C., Cataldi, E., Petruzzelli, R., and
    Giardina, B. (2001a). Functional and structural properties of the hemoglobin components
    from Italian sturgeon (Acipenser naccarii). Fish Physiol. Biochem. 24, 191–200.
Clementi, M. E., De Rosa, M. C., Bertonati, C., Capo, C., Cataldi, E., Petruzzelli, R., and
    Giardina, B. (2001b). The hemoglobins of the ‘‘fossil fish’’ Acipenser naccarii: Functional
    properties and their structural basis. Hemoglobin 25, 447–451.
Cloutier, R., and Ahlberg, P. E. (1996). Morphology, characters, and the interrelationships of
    basal sarcopterygians. In ‘‘Interrelationships of fishes’’ (Stiassny, M. L. J., Parenti, L. R.,
    and Johnson, G. D., Eds.), pp. 445–479. Academic Press, San Diego.
Crocker, C. E., and Cech, J. J., Jr. (1998). EVects of hypercapnia on blood‐gas and acid‐base
    status in the white sturgeon, Acipenser transmontanus. J. Comp. Physiol. 168, 50–60.
Crow, K. D., Stadler, P. F., Lynch, V. J., Amemiya, C., and Wagner, G. P. (2006). The ‘‘fish‐
    specific’’ Hox cluster duplication is coincident with the origin of teleosts. J. Mol. Evol. 23,
Daxboeck, C., Barnard, D. K., and Randall, D. J. (1981). Functional morphology of the gills of
    the bowfin, Amia calva L., with special reference to their significance during air exposure.
    Respir. Physiol. 43, 349–364.
de Moraes, M. F. P. G., Holler, S., Da Costa, O. T. F., Glass, M. L., Fernandes, M. N., and
    Perry, S. F. (2005). Morphometric comparison of the respiratory organs in the South
    American lungfish Lepidosiren paradoxa (Dipnoi). Physiol. Biochem. Zool. 78, 546–559.
Dejours, P. (1975). ‘‘Principles of Comparative Respiratory Physiology.’’ North‐Holland
    Publishing Company, Amsterdam.
Dejours, P. (1988). ‘‘Respiration in Water and Air: Adaptations, Regulation, Evolution.’’ Elsevier,
DeLaney, R. G., Lahiri, S., Hamilton, R., and Fishman, A. P. (1977). Acid‐base balance
    and plasma composition in the aestivating lungfish (Protopterus). Am. J. Physiol. 232,
Donoghue, P. C. J., and Purnell, M. A. (2005). Genome duplication, extinction and vertebrate
    evolution. Trends Ecol. Evol. 20, 312–319.
Ellory, J. C., Wolowyk, M. W., and Young, J. D. (1987). Hagfish (Eptatretus stoutii) erythro-
    cytes show minimal chloride transport activity. J. Exp. Biol. 129, 377–383.
Esbaugh, A., Lund, S., and Tufts, B. L. (2004). Comparative physiology and molecular analysis
    of carbonic anhydrase from the red blood cells of teleost fish. J. Comp. Physiol. [B] 174,
Esbaugh, A. J., and Tufts, B. L. (2006). Tribute to R. G. Boutilier: Evidence of a high activity
    carbonic anhydrase isozyme in the red blood cells of an ancient vertebrate, the sea lamprey
    Petromyzon marinus. J. Exp. Biol. 209, 1169–1178.
Fago, A., and Weber, R. E. (1995). The hemoglobin system of the hagfish Myxine glutinosa:
    Aggregation state and functional properties. Biochim. Biophys. Acta 1249, 109–115.
Fago, A., and Weber, R. E. (1998). Hagfish haemoglobins. In ‘‘The Biology of Hagfishes’’
    (Jørgensen, J. M., Lomholt, J. P., Weber, R. E., and Malte, H., Eds.), pp. 321–333.
    Chapman and Hall, London.
Fago, A., Malte, H., and Dohn, N. (1998). Bicarbonate binding to hemoglobin links oxygen and
    carbon dioxide transport in hagfish. Respir. Physiol. 115, 309–315.
Falcioni, G., Fortuna, G., Giardiba, B., Brunori, M., and Wyman, J. (1977). Functional
    properties of partially oxidized trout hemoglobins. Biochim. Biophys. Acta 490, 171–177.
Fange, R. (1966). Physiology of the swimbladder. Physiol. Rev. 46, 299–322.
Farmer, C. G. (1999). Evolution of the vertebrate cardio‐pulmonary system. Annu. Rev. Physiol.
    61, 573–592.
5.   GAS TRANSPORT AND EXCHANGE                                                                273

Farmer, C. G., and Jackson, D. C. (1998). Air‐breathing during activity in the fishes Amia calva
     and Lepisosteus oculatus. J. Exp. Biol. 201, 943–948.
Farmer, M. (1979). The transition from water breathing to air breathing: EVects of CO2 on
     hemoglobin function. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 62, 109–114.
Farmer, M., Fyhn, H. J., Fyhn, U. E. H., and Noble, R. W. (1979). Occurrence of Root eVect
     hemoglobins in Amazonian fishes. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 62,
Ferguson, R. A., Sehdev, N., Bagatto, B., and Tufts, B. L. (1992). In vitro interactions between
     oxygen and carbon dioxide transport in the blood of the sea lamprey (Petromyzon marinus).
     J. Exp. Biol. 173, 25–41.
Fishman, A. P., Pack, A. I., DeLaney, R. G., and Galante, R. J. (1986). Estivation in
     Protopterus. J. Morphol. 1(Suppl.), 237–248.
Forster, M. E. (1990). Confirmation of the low metabolic rate of hagfish. Comp. Biochem.
     Physiol. A Mol. Integr. Physiol. 96, 113–116.
Fricke, H., and Hissmann, K. (2000). Feeding ecology and evolutionary survival of the living
     coelacanth Latimeria chalumnae. Mar. Biol. 136, 379–386.
Fritsche, R., Axelsson, M., Franklin, C. E., Grigg, G. G., Holmgren, S., and Nilsson, S. (1993).
     Respiratory and cardiovascular‐responses to hypoxia in the Australian lungfish. Respir.
     Physiol. 94, 173–187.
Gallaugher, P., and Farrell, A. P. (1998). Hematocrit and blood oxygen‐carrying capacity.
     In ‘‘Fish Physiology: Fish Respiration’’ (Perry, S. F., and Tufts, B. L., Eds.), pp. 185–227.
     Academic Press, New York.
Gardiner, B. G., Janvier, P., Patterson, C., Forey, P. L., Greenwood, P. H., Miles, R. S., and
     JeVeries, R. P. S. (1979). The salmon, the lungfish and the cow: A reply. Nature (Lond.) 277,
Gervais, M. R., and Tufts, B. L. (1998). Evidence for membrane‐bound carbonic anhydrase in the
     air bladder of bowfin (Amia calva), a primitive air‐breathing fish. J. Exp. Biol. 201, 2205–2212.
Gervais, M. R., and Tufts, B. L. (1999). Characterization of carbonic anhydrase and anion
     exchange in the erythrocytes of bowfin (Amia calva), a primitive air‐breathing fish. Comp.
     Biochem. Physiol. A Mol. Integr. Physiol. 123, 343–350.
Giardina, B., Mosca, D., and De Rosa, M. C. (2004). The Bohr eVect of haemoglobin in
     vertebrates: An example of molecular adaptation to diVerent physiological requirements.
     Acta Physiol. Scand. 182, 165–244.
Gilmour, K. M., Henry, R. P., Wood, C. M., and Perry, S. F. (1997). Extracellular carbonic
     anhydrase and an acid‐base disequilibrium in the blood of the dogfish Squalus acanthias.
     J. Exp. Biol. 200, 173–183.
Gilmour, K. M., Perry, S. F., Bernier, N. J., Henry, R. P., and Wood, C. M. (2001). Extracellular
     carbonic anhydrase in the dogfish, Squalus acanthias: A role in CO2 excretion. Physiol.
     Biochem. Zool. 74, 477–492.
Gonzalez, R. J., Milligan, L., Pagnotta, A., and McDonald, D. G. (2001). EVect of air breathing
     on acid‐base and ion regulation after exhaustive exercise and during low pH exposure in the
     bowfin, Amia calva. Physiol. Biochem. Zool. 74, 502–509.
Goodman, M., Moore, G. W., and Matsuda, G. (1975). Darwinian evolution in the genealogy of
     haemoglobin. Nature 253, 603–608.
Gorr, T., Kleinschmidt, T., and Fricke, H. (1991a). Close tetrapod relationship of the coelacanth
     Latimeria indicated by haemoglobin sequences. Nature 351, 394–397.
Gorr, T., Kleinschmidt, T., Sgouros, J. G., and Kasang, L. (1991b). A ‘‘living fossil’’ sequence:
     Primary structure of the coelacanth (Latimeria chalumnae) hemoglobin – evolutionary and
     functional aspects. Biol. Chem. Hoppe‐Seyler 372, 599–612.
274                                                   C. J. BRAUNER AND M. BERENBRINK

Graham, J. B. (1997). ‘‘Air‐Breathing Fishes. Evolution, Diversity, and Adaptation.’’ Academic
    Press, San Diego.
Hardisty, M. W., and Potter, I. C. (1971). ‘‘The Biology of Lampreys.’’ Academic Press, London.
Hart, J. L. (1973). Pacific fishes of Canada. Fisheries Research Board of Canada (Bulletin 180),
    Ottawa, Canada. p.749.
Heaslet, H. A., and Royer, W. E. (1999). The 2.7 A crystal structure of deoxygenated hemo-
    globin from the sea lamprey (Petromyzon marinus): Structural basis for a lowered oxygen
    aYnity and Bohr eVect. Structure 7, 517–526.
Heemstra, P. C., and Greenwood, P. H. (1992). New observations on the visceral anatomy of the
    late‐term fetuses of the living coelacanth fish and the oophagy controversy. Proc. R. Soc.
    Lond. 249B, 49–55.
Heming, T. A., Vanoye, C. G., Stabenau, E. K., Roush, E. D., Fierke, C. A., and Bidani, A.
    (1993). Inhibitor sensitivity of pulmonary vascular carbonic‐anhydrase. J. Appl. Physiol. 75,
Henry, R. P., and Swenson, E. R. (2000). The distribution and physiological significance of
    carbonic anhydrase in vertebrate gas exchange organs. Respir. Physiol. 121(1), 1–12.
Henry, R. P., Tufts, B. L., and Boutilier, R. G. (1993). The distribution of carbonic anhydrase
    type I and II isozymes in lamprey and trout: Possible co‐evolution with erythrocyte chloride/
    bicarbonate exchange. J. Comp. Physiol. [B] 163, 380–388.
Henry, R. P., Gilmour, K. M., Wood, C. M., and Perry, S. F. (1997). Extracellular carbonic
    anhydrase activity and carbonic anhydrase inhibitors in the circulatory system of fish.
    Physiol. Zool. 70, 650–659.
Hoegg, S., Brinkmann, H., Taylor, J. S., and Meyer, A. (2004). Phylogenetic timing of the fish‐
    specific genome duplication correlates with the diversification of teleost fish. J. Mol. Evol.
    59, 190–203.
Horn, M., and Riggs, C. (1973). EVects of temperature and light on the rate of air breathing of
    the bowfin, Amia calva. Copeia 4, 653–657.
Howell, B. J. (1970). Acid‐base balance in transition from water breathing to air breathing. Fed.
    Proc. 29, 1130–1134.
Hughes, G. M. (1972). Gills of a living coelacanth, Latimeria chalumnae. Experientia 28,
Hughes, G. M. (1976). On the respiration of Latimeria chalumnae. Zool. J. Linn. Soc. 59,
Hughes, G. M. (1995). The gills of the coelacanth, Latimeria chalumnae, a study in relation to
    body size. Philos. Trans. R. Soc. Lond. 347B, 427–438.
Hughes, G. M., and Gray, I. E. (1972). Dimensions and ultrastructure of toadfish gills. Biol. Bull.
    143, 150–161.
Hughes, G. M., and Itazawa, Y. (1972). The eVect of temperature on the respiratory function of
    Coelacanth blood. Experientia 28, 1247.
Hughes, G. M., Perry, S. F., and Piiper, J. (1986). Morphometry of the gills of the elasmobranch
    Scyliorhinus stellaris in relation to body size. J. Exp. Biol. 121, 27–42.
Ingermann, R. L., Terwilliger, R. C., and Roberts, M. S. (1984). Foetal and adult blood oxygen
    aYnities of the viviparous seaperch, Embiotica lateralis. J. Exp. Biol. 108, 453–457.
Isaacks, R. E., and Kim, H. D. (1984). Erythrocyte phosphate composition and osmotic fragility
    in the Australian lungfish, Neoceratodus fosteri, and osteoglossid, Scleropages schneichardti.
    Comp. Biochem. Physiol. A Mol. Integr. Physiol. 79, 667–671.
Isaacks, R. E., Kim, H. D., and Harkness, D. R. (1978). Inositol diphosphate in erythrocytes of
    the lungfish, Lepidosiren paradoxa, and 2,3‐diphosphoglycerate in erythrocytes of the
    armored catfish, Pterygoplichtys sp. Can. J. Zool. 56, 1014–1016.
5.   GAS TRANSPORT AND EXCHANGE                                                               275

Ito, N., Komiyama, N. H., and Fermi, G. (1995). Structure of deoxy‐haemoglobin of
     the Antarctic fish Pagothenia bernacchii with an analysis of the structural basis of the
     Root eVect by comparison of the liganded and unliganded structures. J. Mol. Biol. 250,
Jensen, F. B. (1986). Pronounced influence of Hb‐O2 saturation on red cell pH in tench blood
     in vivo and in vitro. J. Exp. Zool. 238, 119–124.
Jensen, F. B. (1989). Hydrogen ion equilibria in fish haemoglobins. J. Exp. Biol. 143, 225–234.
Jensen, F. B. (1999). Haemoglobin H þequilibria in lamprey (Lampetra fluviatilis) and hagfish
     (Myxine glutinosa). J. Exp. Biol. 202, 1963–1968.
Jensen, F. B. (2004). Red blood cell pH, the Bohr eVect, and other oxygenation‐linked phenom-
     ena in blood O2 and CO2 transport. Acta Physiol. Scand. 182, 215–227.
Jensen, F. B., and Brahm, J. (1995). Kinetics of chloride transport across fish red blood cell
     membranes. J. Exp. Biol. 198, 2237–2244.
Jensen, F. B., Nikinmaa, M., and Weber, R. E. (1983). EVects of exercise stress on acid–base
     balance and respiratory function in blood of the teleost Tinca tinca. Respir. Physiol. 62, 51,
Jensen, F. B., Fago, A., and Weber, R. E. (1998a). Hemoglobin structure and function. In ‘‘Fish
     Physiology: Fish Respiration’’ (Perry, S. F., and Tufts, B. L., Eds.), Vol. 17, pp. 1–40.
     Academic Press, New York.
Jensen, F. B., Wang, T., Jones, D. R., and Brahm, J. (1998b). Carbon dioxide transport in
     alligator blood and its erythrocyte permeability to anions and water. Am. J. Physiol. 274,
Jensen, F. B., Brahm, J., Koldkjær, P., Wang, T., McKenzie, D. J., and Taylor, E. W. (2003).
     Anion exchange in the giant erythrocytes of African lungfish. J. Fish Biol. 62, 1044–1052.
Johansen, K. (1970). Air breathing in fishes. In ‘‘Fish Physiology’’ (Hoar, W. S., and
     Randall, W. S., Eds.), Vol. 4, pp. 361–411. Academic Press, New York.
Johansen, K., and Lenfant, C. (1967). Respiratory function in the South American lungfish,
     Lepidosiren paradoxa (Fitz). J. Exp. Biol. 46, 205–218.
Johansen, K., and Lenfant, C. (1972). A comparative approach to the adaptability of O2‐Hb
     aYnity. In ‘‘Oxygen AYnity of Hemoglobin and Red Cell Acid Base Status’’ (Rørth, M.,
     and Astrup, P., Eds.), pp. 750–780. Munksgaard, Copenhagen.
Johansen, K., Hanson, D., and Lenfant, C. (1970). Respiration in a primitive air breather, Amia
     calva. Respir. Physiol. 9, 162–174.
Johansen, K., Lenfant, C., and Hanson, D. (1973). Gas exchange in the lamprey, Entosphenus
     tridentatus. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 44, 107–119.
Johansen, K., Lomholt, J. P., and Maloiy, G. M. O. (1976a). Importance of air and water
     breathing in relation to size of the African lungfish Protopterus amphibius Peters. J. Exp.
     Biol. 65, 395–399.
Johansen, K., Lykkeboe, G., Weber, R. E., and Maloiy, G. M. O. (1976b). Respiratory proper-
     ties of blood in awake and estivating lungfish, Protopterus amphibius. Respir. Physiol. 27,
Johansen, K., Mangum, C. P., and Lykkeboe, G. (1978). Respiratory properties of the blood of
     Amazon fishes. Can. J. Zool. 56, 896–906.
Jørgensen, J. M., Lomholt, J. P., Weber, R. E., and Malte, H. (1998). ‘‘The Biology of
     Hagfishes.’’ Chapman and Hall, London.
Kilmartin, J. V., and Rossi‐Bernardi, L. (1973). Interaction of hemoglobin with hydrogen ions,
     carbon dioxide, and organic phosphates. Phys. Rev. 53, 836–890.
Kim, H. D., and Isaacks, R. E. (1978). The membrane permeability of nonelectrolytes and
     carbohydrate metabolism of Amazon fish red cells. Can. J. Zool. 56, 863–869.
276                                                    C. J. BRAUNER AND M. BERENBRINK

Kind, P. K., Grigg, G. C., and Booth, D. T. (2002). Physiological responses to prolonged aquatic
    hypoxia in the Queensland lungfish Neoceratodus forsteri. Respir. Physiol. Neurobiol. 132,
King, L. A. (1994). Adult and fetal hemoglobins in the oviparous swell shark, Cephaloscyllium
    ventriosum. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 109, 237–243.
Kister, J., Bohn, B., Marden, M. C., and Poyart, C. (1989). Analysis of oxygen binding by
    Xenopus laevis hemoglobin: Implications for the Root eVect. Respir. Physiol. 76, 191–204.
Klocke, R. A. (1987). Carbon dioxide transport. In ‘‘Handbook of Physiology: The Respiratory
    System’’ (Fishman, A. P., Farhi, L. E., Tenney, S. M., and Geiger, S. R., Eds.), Vol. 4,
    pp. 173–197. American Physiological Society, Maryland.
Klocke, R. A. (1988). Velocity of CO2 exchange in blood. Annu. Rev. Physiol. 50, 625–637.
Koldkjær, P., and Jensen, F. B. (1998). EVects of exhausting exercise and catecholamines on Kþ
    blance, acid‐base status and blood respiratory properties in carp. Comp. Biochem. Physiol.
    A Mol. Integr. Physiol. 119, 301–307.
Koldkjær, P., Taylor, E. W., Glass, M. L., Wang, T., Brahm, J., McKenzie, D. J., and
    Jensen, F. B. (2002). Adrenergic receptors, Naþ/Hþ exchange and volume regulation in
    lungfish erythrocytes. J. Comp. Physiol. 172B, 87–93.
Krogh, A., and Leitch, I. (1919). The respiratory function of the blood in fishes. J. Physiol.
    (Lond.) 52, 288–300.
Kwiatkowski, L. D., DeYoung, A., and Noble, R. W. (1994). Isolation and stability of partially
    oxidized intermediates of carp haemoglobin: Kinetics of CO binding to the mono‐ and
    triferric species. Biochemistry 33, 5884–5893.
Lahiri, S., Shub, C., and Fishman, A. P. (1968). Respiratory properties of blood of the African
    lungfish (Protopterus). Proc. 24th Intern. Union Physiol. Sci. Washington DC 7, p. 251.
Landolt, J. C., and Hill, L. G. (1975). Observations of gross structure and dimensions of gills of 3
    species of gars (Lepisosteidae). Copeia 3, 470–475.
Lapennas, G. N. (1983). The magnitude of the Bohr coeYcient: Optimal for oxygen delivery.
    Respir. Physiol. 54, 161–172.
                             ¨                     ¨
Larsson, A., Johansson‐Sjobeck, M.‐L., and Fange, R. (1976). Comparative study of som
    haematological and biochemical blood parameters in fishes from the Skagerrak. J. Fish
    Biol. 9, 425–440.
Lechleuthner, A., Schumacher, U., Negele, R. D., and Welsch, U. (1989). Lungs of Polypterus
    and Erpetoichthys. J. Morphol. 201, 161–178.
Lenfant, C., and Johansen, K. (1965). Gas transport by hemocyanin‐containing blood of the
    cephalopod Octopus dofleini. Am. J. Physiol. 209, 991–998.
Lenfant, C., and Johansen, K. (1967). Gas exchange and response to environment in lungfish
    Protopterus aethiopicus (Heck). Fed. Proc. 26, 441.
Lenfant, C., and Johansen, K. (1968). Respiration in African lungfish Protopterus aethiopicus.
    Respiratory properties of blood and normal patterns of breathing and gas exchange. J. Exp.
    Biol. 49, 437–452.
Lenfant, C., and Johansen, K. (1972). Gas exchange in gill, skin, and lung breathing. Respir.
    Physiol. 14, 211–218.
Lenfant, C., Johansen, K., and Grigg, G. C. (1966). Respiratory properties of blood and
    pattern of gas exchange in the lungfish Neoceratodus forsteri (KreVt). Respir. Physiol. 2, 1–21.
Lenfant, C., Johansen, K., and Hanson, D. (1970). Bimodal gas exchange and ventilation‐
    perfusion relationship in lower vertebrates. Fed. Proc. 29, 1124–1129.
Lewis, S. V. (1980). Respiration of lampreys. Can. J. Fish. Aquat. Sci. 37, 1711–1722.
Lewis, S. V., and Potter, I. C. (1976). Gill morphometrics of the lampreys Lampetra fluviatilis
    (L.) and Lampetra planeri (Bloch). Acta Zool. 57, 103–112.
5.   GAS TRANSPORT AND EXCHANGE                                                              277

Lund, S. G., Dyment, P., Gervais, M. R., Moyes, C. D., and Tufts, B. L. (2002). Characteriza-
    tion of erythrocyte carbonic anhydrase in an ancient fish, the longnose gar (Lepisosteus
    osseus). J. Comp. Physiol. 172B, 467–476.
Magid, A. M. A. (1967). Respiration of air by the primitive fish Polypterus senegalus. Nature 215,
Maginniss, L. A., Song, Y. K., and Reeves, R. B. (1980). Oxygen equilibria of ectotherm blood
    containing multiple hemoglobins. Respir. Physiol. 42, 329–343.
Mangum, C. P. (1991). Urea and chloride sensitivities of coelacanth hemoglobin. Environ. Biol.
    Fishes 32, 219–222.
Manwell, C. (1958). On the evolution of hemoglobin: Respiratory properties of the hemoglobin
    of the California hagfish, Polistotrema stouti. Biol. Bull. 115, 227–238.
Manwell, C. (1963). The blood proteins of cyclostomes. A study in phylogenetic and ontogenetic
    biochemistry. In ‘‘The Biology of Myxine’’ (Brodal, A., and Fange, R., Eds.), Vol. 17,
    pp. 372–455. Scandinavian University Books, Oslo.
Marshall, C. R., and Schultze, H. P. (1992). Relative importance of molecular, neontological,
    and paleontological data in understanding the biology of the vertebrate invasion of land.
    J. Mol. Evol. 35, 93–101.
Maxime, V., Nonnotte, G., Peyraud, C., Williot, P., and Truchot, J. P. (1995). Circulatory and
    respiratory eVects of an hypoxic stress in the Siberian sturgeon. Respir. Physiol. 100,
Maxime, V., Nonnotte, G., and Williot, P. (1998). Respiratory and circulatory adaptations of
    Siberian sturgeon to environmental hypoxia. BFPP Bull. Fr. Peche Prot. Milieux Aquat.
    350–351, 377–391.
McKenzie, D. J., Aota, S., and Randall, D. J. (1991a). Ventilatory and cardiovascular responses
    to blood pH, plasma PCO2, blood O2 content, and catecholamines in an air‐breathing fish,
    the bowfin (Amia clava). Physiol. Zool. 64, 432–450.
McKenzie, D. J., Burleson, M. L., and Randall, D. J. (1991b). The eVects of branchial denerva-
    tion and pseudobranch ablation on cardioventilatory control in an air‐breathing fish.
    J. Exp. Biol. 161, 347–365.
McKenzie, D. J., Piraccini, G., Papini, N., Galli, C., Bronzi, P., Bolis, C. G., and Taylor, E. W.
    (1997). Oxygen consumption and ventilatory reflex responses are influenced by dietary lipids
    in sturgeon. Fish Physiol. Biochem. 16, 165–179.
McKenzie, D. J., Wong, S., Randall, D. J., Egginton, S., Taylor, E. W., and Farrell, A. P. (2004).
    The eVects of sustained exercise and hypoxia upon oxygen tensions in the red muscle of
    rainbow trout. J. Exp. Biol. 207, 3629–3637.
McMahon, B. R. (1970). The relative eYciency of gaseous exchange across the lungs and gills of
    an African lungfish Protopterus aethiopicus. J. Exp. Biol. 52, 1–16.
Meyer, A., and Schartl, M. (1999). Gene and genome duplications in vertebrates: The one‐to‐
    four (‐to‐eight in fish) rule and the evolution of novel gene functions. Curr. Opin. Cell Biol.
    11, 699–704.
Muir, B. S., and Hughes, G. M. (1969). Gill dimensions for three species of tunny. J. Exp. Biol.
    51, 271–285.
  ¨                                                                  ¨              ¨
Muller, J. (1839). Vergleichende Anatomie der Myxinoiden. III. Uber das Gefassystem. Abh.
    Dtsch. Akad. Wiss. Berlin 1839, 175–303.
Munz, F. W., and Morris, R. W. (1965). Metabolic rate of the hagfish, Eptatretus stoutii (Lock-
    ington) 1878. Comp. Biochem. Physiol. 16, 1–6.
Nelson, J. S. (1994). ‘‘Fishes of the World,’’ 3rd edn. Wiley, New York.
Nikinmaa, M. (1986). Red cell pH of lamprey (Lampetra fluviatilis) is actively regulated.
    J. Comp. Physiol. [B] 156, 747–750.
278                                                    C. J. BRAUNER AND M. BERENBRINK

Nikinmaa, M. (1990). ‘‘Vertebrate Red Blood Cells. Adaptations of Function to Respiratory
     Requirements.’’ Springer, Berlin.
Nikinmaa, M. (1992). Membrane transport and control of hemoglobin‐oxygen aYnity in nucle-
     ated erythrocytes. Physiol. Rev. 72, 301–321.
Nikinmaa, M. (1993). Haemoglobin function in intact Lampetra fluviatilis erythrocytes. Respir.
     Physiol. 91, 283–293.
Nikinmaa, M. (1997). Oxygen and carbon dioxide transport in vertebrate erythrocytes:
     An evolutionary change in the role of membrane transport. J. Exp. Biol. 200, 369–380.
Nikinmaa, M. (2001). Haemoglobin function in vertebrates: Evolutionary changes in cellular
     regulation in hypoxia. Respir. Physiol. 128, 317–329.
Nikinmaa, M., and Huestis, W. H. (1984). Adrenergic swelling in nucleated erythrocytes:
     Cellular mechanisms in a bird, domestic goose, and two teleosts, striped bass and rainbow
     trout. J. Exp. Biol. 113, 215–224.
Nikinmaa, M., and Railo, E. (1987). Anion movements across lamprey (Lampetra fluviatilis) red
     cell membrane. Biochim. Biophys. Acta 899, 134–136.
Nikinmaa, M., and Salama, A. (1998). Oxygen transport in fish. In ‘‘Fish Physiology: Fish
     Respiration’’ (Perry, S. F., and Tufts, B. L., Eds.), Vol. 17, pp. 141–184. Academic Press,
     New York.
Nikinmaa, M., and Weber, R. E. (1984). Hypoxic acclimation in the lamprey Lampetra fluvia-
     tilis: Organismic and erythrocytic responses. J. Exp. Biol. 109, 109–119.
Nikinmaa, M., Cech, J. J., Jr., and McEnroe, M. (1984). Blood oxygen transport in stressed
     striped bass (Morone saxatilis): Role of beta‐adrenergic responses. J. Comp. Physiol. 154B,
Nikinmaa, M., Tufts, B. L., and Boutilier, R. G. (1993). Volume and pH regulation in agnathan
     erythrocytes: Comparisons between the hagfish, Myxine glutinosa and the lampreys, Petry-
     myson marinus and. Lampetra fluvitilis. J. Comp. Physiol. 163B, 608–613.
Nikinmaa, M., Airaksinen, S., and Virkki, L. V. (1995). Haemoglobin function in intact lamprey
     erythrocytes: Interactions with membrane function in the regulation of gas transport and
     acid‐base balance. J. Exp. Biol. 198, 2423–2430.
Nonnotte, G., Maxime, V., Truchot, J. P., Williot, P., and Peyraud, C. (1993). Respiratory responses
     to progressive ambient hypoxia in the sturgeon, Acipenser‐Baeri. Respir. Physiol. 91, 71–82.
Olson, K. R. (1981). Morphology and vascular anatomy of the gills of a primitive air‐breathing
     fish, the bowfin (Amia calva). Cell Tissue Res. 218, 499–517.
Pelster, B., and Randall, D. J. (1998). The physiology of the Root eVect. In ‘‘Fish Physiology:
     Fish Respiration’’ (Perry, S. F., and Tufts, B. L., Eds.), Vol. 17, pp. 113–140. Academic
     Press, New York.
Perry, S. F. (1986). Carbon dioxide excretion in fishes. Can. J. Zool. 64, 565–572.
Perry, S. F., and Gilmour, K. M. (2002). Sensing and transfer of respiratory gases at the fish gill.
     J. Exp. Zool. 293, 249–263.
Perry, S. F., and Laurent, P. (1990). The role of carbonic anhydrase in carbon dioxide excretion,
     acid‐base balance and ionic regulation in aquatic gill breathers. In ‘‘Animal Nutrition and
     Transport Processes 2. Transport, Respiration and Excretion: Comparative and Environ-
     mental Aspects’’ (Truchot, J. P., and Lahlou, B., Eds.), pp. 39–57. Karger, Basel.
Perry, S. F., and Sander, M. (2004). Reconstructing the evolution of the respiratory apparatus in
     tetrapods. Respir. Physiol. Neurobiol. 144, 125–139.
Perry, S. F., Davie, P. S., Daxboeck, C., and Randall, D. J. (1982). A comparison of CO2
     excretion in a spontaneously ventilating blood‐perfused trout preparation and saline‐
     perfused gill preparations: Contribution of the branchial epithelium and red blood cell.
     J. Exp. Biol. 101, 47–60.
5.   GAS TRANSPORT AND EXCHANGE                                                               279

Perry, S. F., Gilmour, K. M., Swenson, E. R., Vulesevic, B., Chew, S. F., and Ip, Y. K. (2005).
     An investigation of the role of carbonic anhydrase in aquatic and aerial gas transfer in the
     African lungfish Protopterus dolloi. J. Exp. Biol. 208, 3805–3815.
Perutz, M. F., and Brunori, M. (1982). Stereochemistry of cooperative eVects in fish and
     amphibian hemoglobins. Nature 299, 421–426.
Pettit, M. J., and Beitinger, T. L. (1985). Oxygen acquisition of the reedfish, Erpetoichthys
     calabaricus. J. Exp. Biol. 114, 289–306.
Phelps, C., Farmer, M., Fyhn, H. J., Fyhn, U. E. H., Garlick, R. L., Noble, R. W., and
     Powers, D. A. (1979). Equilibria and kinetics of oxygen and carbon monoxide binding to
     the haemoglobin of the South American lungfish, Lepidosiren paradoxa. Comp. Biochem.
     Physiol. A Mol. Integr. Physiol. 62, 139–143.
Potter, I. C., Welsch, U., Wright, G. M., Honma, Y., and Chiba, A. (1995). Light and electron‐
     microscope studies of the dermal capillaries in 3 species of hagfishes and 3 species of
     lampreys. J. Zool. 235, 677–688.
Potter, I. C., Macey, D. J., and Roberts, A. R. (1996a). Oxygen consumption by adults of the
     southern hemisphere lamprey Geotria australis in air. J. Exp. Zool. 276, 254–261.
Potter, I. C., Macey, D. J., Roberts, A. R., and Withers, P. C. (1996b). Oxygen consumption by
     ammocoetes of the lamprey Geotria australis in air. J. Comp. Physiol. 166B, 331–336.
Potter, I. C., Macey, D. J., and Roberts, A. R. (1997). Oxygen uptake and carbon dioxide
     excretion by the branchial and postbranchial regions of adults of the lamprey Geotria
     australis in air. J. Exp. Zool. 278, 290–298.
Powers, D. A., Fyhn, H. J., Fyhn, U. E. H., Martin, J. P., Garlick, R. L., and Wood, S. C. (1979).
     A comparative study of the oxygen equilibria of blood from 40 genera of Amazonian fishes.
     Comp. Biochem. Physiol. A Mol. Integr. Physiol. 62, 67–86.
Qiu, Y., Maillett, D. H., Knapp, J., Olson, J. S., and Riggs, A. F. (2000). Lamprey hemoglobin.
     Structural basis of the Bohr effect. J. Biol. Chem. 275, 13517–13528.
Rahn, H. (1966). Aquatic gas exchange: Theory. Respir. Physiol. 1, 1–12.
Rahn, H., Rahn, K. B., Howell, B. J., Gans, G., and Tenney, S. M. (1971). Air breathing of the
     garfish (Lepisosteus osseus). Respir. Physiol. 11, 285–307.
Randall, D. J. (1972). Respiration. In ‘‘The Biology of Lampreys’’ (Hardisty, M. W., and
     Potter, I. C., Eds.), pp. 287–306. Academic Press, London.
Randall, D. J., and Brauner, C. J. (1998). Interactions between ion and gas transfer in freshwater
     teleost fish. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 119A, 3–8.
Randall, D. J., Cameron, J. N., Daxboeck, C., and Smatresk, N. (1981). Aspects of bimodal
     gas exchange in the bowfin, Amia calva (Actinopterygii: Amiiformes). Respir. Physiol. 43,
Randall, D. J., McKenzie, D. J., Abrami, G., Bondiolotti, G. P., Natiello, F., Bronzi, P.,
     Bolis, L., and Agradi, E. (1992). EVects of diet on responses to hypoxia in sturgeon
     (Accipenser naccarii). J. Exp. Biol. 170, 113–125.
Rauther, M. (1922). Zur vergleichenden Anatomie der Schwimmblase der Fische. Ergebnisse und
     Fortschritte der Zoologie 5, 1–66.
Rauther, M. (1937). Die Schwimmblase. In ‘‘Handbuch der vergleichenden Anatomie der
     Wirbeltiere’’ (Bolk, L., Goppert, E., Kallus, E., and Lubosch, W., Eds.), Vol. 3,
     pp. 883–908. Reprinted 1967, Asher and Co., Amsterdam.
Reynolds, J. D., Goodwin, N. B., and Freckleton, R. P. (2002). Evolutionary transitions in parental
     care and live bearing in vertebrates. Philos. Trans. R. Soc. Lond. B Biol. Sci. 357, 269–281.
Riggs, A. (1972). The haemoglobins. In ‘‘The Biology of Lampreys’’ (Hardisty, M. W., and
     Potter, I. C., Eds.), pp. 261–286. Academic Press, London.
Riggs, A. F. (1988). The Bohr eVect. Annu. Rev. Physiol. 50, 181–204.
280                                                    C. J. BRAUNER AND M. BERENBRINK

Rodewald, K., Stangl, A., and Braunitzer, G. (1984). Primary structure, biochemical and
     physiological aspects of hemoglobin from South American lungfish (Lepidosiren paradoxus,
     Dipnoi). Hoppe‐Seyler’s Z. Physiol. Chem. 365, 639–649.
Rodrıguez, A, and Gisbert, E. (2001). Morphogenesis of the eye of Siberian sturgeon. J. Fish
     Biol. 59, 1427–1429.
Root, R. W. (1931). The respiratory function of the blood of marine fishes. Biol. Bull. 61, 427–456.
Sacca, R., and Burggren, W. (1982). Oxygen uptake in air and water in the air‐breathing reedfish
     Calamoichthys calabaricus: Role of skin, gills and lungs. J. Exp. Biol. 97, 179–186.
Sanchez, A., Soncini, R., Wang, T., Koldkjær, P., Taylor, E. W., and Glass, M. L. (2001). The
     diVerential cardio‐respiratory responses to ambient hypoxia and systemic hypoxaemia in the
     South‐American lungfish, Lepidosiren paradoxa. Comp. Biochem. Physiol. A Mol. Integr.
     Physiol. 130, 677–687.
Scheid, P., and Meyer, M. (1978). Mixing technique for study of oxygen‐hemoglobin equilibri-
     um: A critical evaluation. J. Appl. Physiol. 45, 818–822.
Scholander, P. F., and Van Dam, L. (1954). Secretion of gases against high pressures in the
     swimbladder of deep sea fishes. I. Oxygen dissociation in blood. Biol. Bull. 107,
Schultze, H.‐P. (1993). Osteichthyes: Sarcopterygii. In ‘‘The Fossil Record 2’’ (Benton, M. J.,
     Ed.), pp. 657–663. Chapman and Hall, London.
Schultze, H.‐P. (2003). Dipnoi, Lungenfische. In ‘‘Spezielle Zoologie; Teil 2: Wirbel‐ oder
     Schadeltiere’’ (Westheide, W., and Rieger, R., Eds.), p. 290. Spectrum Fischer, Heidelberg.
Seifert, A. W., and Chapman, L. J. (2006). Respiratory allocation and standard rate of metabo-
     lism in the African lungfish, Protopterus aethiopicus. Comp. Biochem. Physiol. A Mol. Integr.
     Physiol. 143, 142–148.
Siggaard‐Andersen, O. (1975). ‘‘The Acid‐Base Status of the Blood,’’ 4th edn. Munksgaard,
Siret, J. R., Carmena, A. O., and Callejas, J. (1976). Erythrokinetic study in the fish manjuari
     (Atracosteus tristoechus). Comp. Biochem. Physiol. A Mol. Integr. Physiol. 55, 127–128.
Slowinski, J. B., and Guyer, C. (1993). Testing whether certain traits have caused amplified
     diversification: An improved method based on a model of random speciation and extinction.
     Am. Nat. 142, 1019–1024.
Smatresk, N. J., and Cameron, J. N. (1982a). Respiration and acid‐base physiology of the
     spotted gar, a bimodal breather: I. Normal values, and the response to severe hypoxia.
     J. Exp. Biol. 96, 263–280.
Smatresk, N. J., and Cameron, J. N. (1982b). Respiration and acid‐base physiology of the
     spotted gar, a bimodal breather: II. Responses to temperature change and hypercapnia.
     J. Exp. Biol. 96, 281–293.
Smith, C. L., Rand, C. S., SchaeVer, B., and Atz, J. W. (1975). Latimeria, the living coelacanth, is
     ovoviviparous. Science 190, 1105–1106.
Smith, H. W. (1930). Metabolism of the lungfish Protopterus aethiopicus. J. Biol. Chem. 88,
Soivio, A., Nikinmaa, M., and Westman, K. (1980). The blood oxygen binding properties of
     hypoxic Salmo gairdneri. J. Comp. Physiol. [B] 136, 83–87.
Stabenau, E. K., and Heming, T. (2003). Pulmonary carbonic anhydrase in vertebrate gas
     exchange organs. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 136, 271–279.
SteVensen, J. F., Johansen, K., Sindberg, C. D., Sørensen, J. H., and Møller, J. L. (1984).
     Ventilation and oxygen‐consumption in the hagfish, Myxine glutinosa L. J. Exp. Mar.
     Biol. Ecol. 84, 173–178.
Strahan, R. (1958). The velum and respiratory current of Myxine. Acta Zool. 39, 227–240.
5.   GAS TRANSPORT AND EXCHANGE                                                                281

Strahan, R. (1963). Behaviour of Myxine and other myxinoids. In ‘‘The Biology of Myxine’’
    (Brodal, A., and Fange, R, Eds.), pp. 22–32. Scandinavian University Books, Oslo.
Swan, H., and Hall, F. G. (1966). Oxygen‐hemoglobin dissociation in. Protopterus aethiopicus.
    Am. J. Physiol. 210, 487–489.
Swenson, E. R. (1990). Kinetics of oxygen and carbon dioxide exchange. In ‘‘Advances in
    Comparative and Environmental Physiology’’ (Boutilier, R. G., Ed.), Vol. 6, pp. 163–209.
    Springer‐Verlag, Berlin.
Swenson, E. R., and Maren, T. (1978). A quantitative analysis of CO2 transport at rest and
    during maximal exercise. Respir. Physiol. 35, 129–159.
Takezaki, N., Figueroa, F., Zaleska‐Rutczynska, Z., and Klein, J. (2003). Molecular phylogeny
    of early vertebrates: Monophyly of the agnathans as revealed by sequences of 35 genes. Mol.
    Biol. Evol. 20, 287–292.
Tetens, V., and Lykkeboe, G. (1981). Blood respiratory properties of rainbow trout, Salmo
    gairdneri: Responses to hypoxia acclimation and anoxic incubation of blood in vitro.
    J. Comp. Physiol. 145, 117–125.
TretjakoV, D. (1926). Die Orbitalen Venensinusse der niederen Wirbeltiere. Gegenbaurs
    Morphol. Jahrb. 56, 402–445.
Tufts, B. L. (1991). Acid‐base regulation and blood gas transport following exhaustive exercise in
    an agnathan, the sea lamprey Petromyzon marinus. J. Exp. Biol. 159, 371–385.
Tufts, B. L., and Boutilier, R. G. (1989). The absence of rapid chloride/bicarbonate exchange in
    lamprey erythrocytes: Implications for CO2 transport and ion distributions between plasma
    and erythrocytes in the blood of Petromyzon marinus. J. Exp. Biol. 144, 565–576.
Tufts, B. L., and Boutilier, R. G. (1990). CO2 transport properties of the blood of a primitive
    vertebrate, Myxine glutinosa (L.). Exp. Biol. 48, 341–347.
Tufts, B. L., and Perry, S. F. (1998). Carbon dioxide transport and excretion. In ‘‘Fish Physiology:
    Fish Respiration’’ (Perry, II, S. F., and Tufts, B. L., Eds.), Vol. 17, pp. 229–282. Academic
    Press, San Diego.
Tufts, B. L., Bagatto, B., and Cameron, B. (1992). In vivo analysis of gas transport in arterial and
    venous blood of the sea lamprey Petromyzon marinus. J. Exp. Biol. 169, 105–119.
Tufts, B. L., Drever, R. C., Bagatto, B., and Cameron, B. A. (1994). In vitro analysis of volume
    and pH regulation in the red blood cells of a primitive air‐breathing fish, the bowfin, Amia
    calva. Can. J. Zool. 72, 280–286.
Tufts, B. L., Esbaugh, A., and Lund, S. G. (2003). Comparative physiology and molecular
    evolution of carbonic anhydrase in the erythrocytes of early vertebrates. Comp. Biochem.
    Physiol. A Mol. Integr. Physiol. 136, 259–269.
Ultsch, G. R. (1996). Gas exchange, hypercarbia and acid‐base balance, paleoecology, and the
    evolutionary transition from water breathing to air‐breathing among vertebrates. Palaeo-
    geogr. Palaeoclimatol. Palaeoecol. 123, 1–27.
Ultsch, G. R., and Jackson, D. C. (1996). pH and temperature in ectothermic vertebrates. Bull.
    Alabama Mus. Nat. Hist. 18, 1–41.
Val, A. L. (2000). Organic phosphates in the red blood cells of fish. Comp. Biochem. Physiol.
    A Mol. Integr. Physiol. 125, 417–435.
Virkki, L. V., and Nikinmaa, M. (1994). Activation and physiological role of Naþ/Hþ exchange
    in lamprey (Lampetra fluviatilis) erythrocytes. J. Exp. Biol. 191, 89–105.
Vokac, Z., Ahmed, N. E. D., and Abdel Magid, A. M. (1972). The in vitro respiratory properties
    of blood of the primitive fish Polypterus senegalus. J. Exp. Biol. 57, 461–469.
Wallace, W. J., Houtchens, R. A., Maxwell, J. C., and Caughey, W. S. (1982). Mechanism of
    autooxidation for haemoglobins and myoglobins: Promotion of superoxide production by
    protons and anions. J. Biol. Chem. 257, 4966–4977.
282                                                  C. J. BRAUNER AND M. BERENBRINK

Waser, W., and Heisler, N. (2005). Oxygen delivery to the fish eye: Root eVect as a crucial factor
    for elevated retinal PO2. J. Exp. Biol. 208, 4035–4047.
Weber, R. E., and Hartvig, M. (1984). Specific fetal hemoglobin underlies the fetal‐maternal shift
    in blood‐oxygen aYnity in a viviparous teleost. Mol. Physiol. 6, 27–32.
Weber, R. E., and Lykkeboe, G. (1978). Respiratory adaptations in carp blood. Influences of
    hypoxia, red cell organic phosphates, divalent cations and CO2 on hemoglobin‐oxygen
    aYnity. J. Comp. Physiol. [B] 128, 127–137.
Weber, R. E., Bol, J. E., Johansen, K., and Wood, S. C. (1973). Physicochemical properties of
    hemoglobin of coelacanth Latimeria chaluumnae. Arch. Biochem. Biophys. 154, 96–105.
Weber, R. E., Johansen, K., Lykkeboe, G., and Malooiy, G. M. O. (1976a). Oxygen‐binding
    properties of hemoglobins from estivating and active African lungfish. J. Exp. Zool. 199,
Weber, R. E., Sullivan, B., Bonaventura, J., and Bonventura, C. (1976b). The hemoglobin
    system of the primitive fish, Amia calva: Isolation and functional characterization of the
    individual hemoglobin components. Biochim. Biophys. Acta 434, 18–31.
Wells, R. M. G., and Forster, M. E. (1989). Dependence of blood viscosity on haematocrit and
    shear rate in a primitive vertebrate. J. Exp. Biol. 145, 483–487.
Wells, R. M. G., Davie, P. S., and Satchell, G. H. (1986). Blood oxygen transport in the
    free‐swimming hagfish, Eptatretus cirrhatus. J. Exp. Biol. 123, 43–53.
Wintrobe, M. M. (1934). Variations in the size and hemoglobin content of erythrocytes in the
    blood of various vertebrates. Folia Haematol. 51, 32–49.
Wittenberg, J. B., and Haedrich, R. L. (1974). The choroid rete mirabile of the fish eye. II.
    Distribution and relation to the pseudobranch and to the swimbladder rete mirabile. Biol.
    Bull. 146, 137–156.
Wittenberg, J. B., and Wittenberg, B. A. (1974). The choroid rete mirabile of the fish eye. I.
    Oxygen secretion and structure: Comparison with the swimbladder rete mirabile. Biol. Bull.
    146, 116–136.
Wood, S. C., and Lenfant, C. (1987). Phylogeny of the gas exchange system: Red cell function.
    In ‘‘Handbook of Physiology, Section 3: The Respiratory System’’ (Fahri, L. E., and
    Tenney, S. M., Eds.), Vol. 4, pp. 131–146. American Physiological Society, Bethesda.
Wood, S. C., Johansen, K., and Weber, R. E. (1972). Haemoglobin of the coelacanth. Nature
    239, 283–285.
Wourms, J. P., Grove, B. D., and Lombardi, J. (1988). The maternal‐embryonic relationship in
    viviparous fishes. In ‘‘Fish Physiology’’ (Hoar, W. S., and Randall, D. J., Eds.), Vol. 11B,
    pp. 1–134. Academic Press, San Diego.
Wourms, J. P., Atz, J. W., and Stribling, M. D. (1991). Viviparity and the maternal‐embryonic
    relationship in the coelacanth Latimeria chalumnae. Environ. Biol. Fishes 32, 225–248.
Wyman, J. (1964). Linked functions and reciprocal eVects in haemoglobin: A second look. Adv.
    Protein Chem. 19, 223–286.
Yokoyama, T., Chong, K. T., Miyazaki, G., Morimoto, H., Shih, D. T.‐B., Unzai, S.,
    Tame, J. R. H., and Park, S.‐Y (2004). Novel mechanism of pH sensitivity in tuna hemoglo-
    bin. A structural explanation of the root eVect. J. Biol. Chem. 279, 28632–28640.


1. Introduction
   1.1. Origins in Seawater
   1.2. ‘‘Parting of the Ways’’: The Move to Freshwater
   1.3. Key Sites of Osmoregulation and Nitrogen Excretion in Fishes
2. Ionic and Osmotic Regulation
   2.1. In Seawater
   2.2. In Freshwater
   2.3. Moving Between the River and Sea
3. Nitrogen Excretion
   3.1. Toxic Ammonia
   3.2. Synthesis of Nitrogen End‐Products
   3.3. Excretion
   3.4. The Challenges of Estivation
4. Concluding Remarks

     Among primitive fishes, there is a diversity of strategies that have evolved to
cope with ion, water, and nitrogen balance. The whole physiological spectrum
is found from ionic and osmotic conformation to the regulation of body fluids
distinct from the environment. The most primitive of vertebrates, the marine
hagfish iono‐ and osmoconforms to its seawater environment, whereas their
euryhaline relatives, the lampreys, iono‐ and osmoregulate. The gills of
Agnathans contain both pavement and mitochondrial rich cells, but the ar-
rangement of cells and structural features are unique relative to euryhaline
teleosts. Coelacanths are osmoconformers but ionoregulators, maintaining
high internal urea levels like the elasmobranchs. In many primitive species,
ammonia is the dominant excretory product as it is in most teleost fishes.
The exception is the coelacanth and estivating lungfish that synthesize urea
via the urea cycle and excrete urea. Membrane transporters have been isolated
in fish that regulate urea and possibly ammonia movements between tissue

Primitive Fishes: Volume 26                          Copyright # 2007 Elsevier Inc. All rights reserved
FISH PHYSIOLOGY                                                 DOI: 10.1016/S1546-5098(07)26006-6
284                                                         PATRICIA A. WRIGHT

compartments and to the environment. Nitrogen excretion during early life
stages presents a particular challenge in encapsulated embryos dependent on
yolk protein catabolism. As yet, little is known about how primitive fish
embryos face these challenges. Research on primitive fish species will broaden
our knowledge of the evolution of osmoregulation and excretion in fish and
terrestrial vertebrates.


    The focus of this chapter is on the Agnathans (lampreys and hag-
fishes), the Sarcopterygians (the coelacanth and the lungfishes), and the
primitive Actinopterygians such as the Polypteriformes (bichirs and reed-
fish), the Chondrostean Acipenseriformes (paddlefishes and sturgeons),
and the Neopterygians (gars and the bowfin). The systematics and phylogeny
of these fishes is outlined by P. Janvier in Chapter 1. This chapter con-
cerns the regulation of ions, water, and nitrogen end‐products. Although
the endocrine control of these processes is important in their regula-
tion, the current chapter focuses on the sites, structures, and mechanisms
involved in iono‐ and osmoregulation. Readers are referred to two compre-
hensive chapters on ‘‘Peripheral Endocrine Glands’’ by J. Youson in this
volume (Chapters 8 and 9) for a more detailed review of endocrine control.

1.1. Origins in Seawater
    Over the last 100 years or so, scientists have debated whether the first fishes
evolved in freshwater or seawater (Smith, 1932, 1961; Munz and McFarland,
1964; Fange, 1998). These arguments are based on osmoregulatory structures
present in extant fishes, including the most primitive jawless fishes, the
Agnathans. Collective opinion weighs in on the seawater side (GriYth,
1994; Holland and Chen, 2001; Chang et al., 2006); early Agnathans are
thought to have inhabited shallow seas or estuaries (Helfman et al., 1997).
The living jawless hagfishes are entirely marine. They are unique among
vertebrates in having plasma ion concentrations and osmolarity roughly the
same as seawater, similar to marine invertebrates.

1.2. ‘‘Parting of the Ways’’: The Move to Freshwater

    Smith (1932) first noticed the interesting contrast between the blood
osmotic concentration of the hagfish Myxine glutinosa with that of the
lamprey Petromyzon fluviatilis. Hagfish iono‐ and osmoconform to their
seawater habitat, whereas lamprey iono‐ and osmoregulate in either seawater
or freshwater. It has taken decades to form a more complete understanding of
vertebrate osmoregulation, but the following prescient statement by Smith
6.   OSMOREGULATION                                                          285

(1932) captures so much. It is thus possible that these two groups lead back to a
‘‘parting of the ways’’ in the evolution of body fluid regulation. (The two groups
he refers to are the hagfish and the lamprey.) After early beginnings in sea-
water, lamprey ancestors moved back to freshwater, no later than the early
Cretaceous (Chang et al., 2006). Osmotic control that evolved first in lamprey
and teleost ancestors has been an adaptive trait with a selective advantage, in
both freshwater and marine environments (Robertson, 1963). Although the
hagfish lineage has survived for millions of years, their unique form of iono‐ and
osmoconformation has not appeared in any other group of aquatic vertebrate.

1.3. Key Sites of Osmoregulation and Nitrogen Excretion in Fishes
    The gills, kidney, intestine, urinary bladder, and integument are the
key sites of ion exchange, nitrogen elimination, and osmoregulation in most
fishes (for a review see Marshall and Grosell, 2006). In many of the primitive
fish groups, there is limited information to verify the involvement of these sites
or the importance of one tissue over another. However, it is probably safe to
say that the gill is the dominant site of exchange of ions, water, and nitroge-
nous waste products in most of the primitive fishes discussed below.
    Among primitive fishes, with the exception of lungfish, the gill has a large
surface area in contact with flowing water and with the aid of specialized
branchial epithelium, materials are transported between the blood and water.
The kidney is also an important structure, particularly in freshwater fishes
where passive water gain is countered by a high urine output with reabsorption
of key monovalent ions (see Section 2.2); however, renal nitrogen excretion is
typically low. The intestine and urinary bladder are important for absorption of
ions and water, but will not be considered further due to the lack of research on
primitive fishes. The integument in most fish presents a barrier to the exchange
between the internal and external environments, and may only play a critical
role in a few unusual species (Wood, 1993; see Section 2.2). It should also be
noted that a postanal gland, similar in structure to the elasmobranch rectal
gland, has been described in the coelacanth Latimeria chalumnae (reviewed by
Locket, 1980). This gland probably represents an additional site of NaCl


2.1. In Seawater

2.1.1. Ionoconform, Osmoconform: The Hagfish Strategy
   Hagfish are the only aquatic vertebrate known that ionoconform as well as
osmoconform to their seawater environment (Figure 6.1A) (Robertson, 1963;
Evans, 1993; Karnaky, 1998). They are not found in water of low salinity,
286                                                                     PATRICIA A. WRIGHT

Fig. 6.1. Schematic representation of ionic and osmotic balance in (A) hagfish, (B) coelacanth, (C)
seawater lamprey, and (D) freshwater lamprey. Total plasma osmolality (mosmol kgÀ1), plasma
osmolality attributed to NaCl (mosmol kgÀ1), and plasma urea concentrations (mmol kgÀ1) are
given inside each diagram, were appropriate. Passive fluxes are represented by a dashed line,
whereas active mechanisms are shown as a solid line. [A, values from McDonald and Milligan
(1992), B, GriYth (1991), and C and D modified from Bartels and Potter (2004).]
6.   OSMOREGULATION                                                         287

contrasting sharply with their Agnathan relatives, the lampreys (see Section
2.1.3, below). They gain weight in hyposmotic water, slowly recovering after
7 days, whereas in hyperosmotic water they lose weight and fail to recover
(McFarland and Munz, 1965). Smith (1932) established that the ionic com-
position of hagfish blood approximated the inorganic ion concentrations of
seawater (Figure 6.1A), a very diVerent osmoregulatory strategy compared
to osmoconforming coelacanth and elasmobranch fishes (Figure 6.1B; see
Section 3.3.3). Similar to elasmobranchs, the intracellular osmotic compo-
sition of marine hagfish has a large organic component, with trimethyla-
mine oxide (TMAO) concentrations exceeding 200 mM and one‐third lower
inorganic ion levels relative to serum levels (Bellamy and Jones, 1961).
     There has been very little attention to the mechanisms of iono‐ and
osmoregulation in hagfish, possibly because it was assumed that iono‐
and osmoconforming with the environment requires minimal eVort. Large
mitochondrial rich cells (MRCs) are numerous on hagfish gill lamellae
(Mallat and Paulsen, 1986; for a review see Bartels, 1998; Choe et al.,
1999), but unlike seawater‐acclimated lampreys (see Section 2.1.3) the
MRCs appear singly, sandwiched between gill pavement cells. A leaky para-
cellular junction that is so clearly observed between MRCs in marine teleosts
and lampreys and allows for the passive leak of Naþ is not apparent in
hagfish gills. Hagfish gill MRCs stain for Naþ/Kþ ATPase (Mallat et al.,
1987; Choe et al., 1999) and Naþ/Hþ exchanger isoforms (NHE) are
also expressed in the M. glutinosa gill but cell localization has yet to be
determined (Edwards et al., 2001; Choe et al., 2002). Evans (1984) proposed
that NHE and ClÀ/HCOÀ exchangers were operating in parallel in hagfish
gill epithelium for acid or base excretion, and provided a ‘‘preadaptation’’ for
ion regulation in species that later inhabited freshwaters. This proposition
was later supported by McDonald et al. (1991) who showed that acid–base
disturbances in M. glutinosa were fully corrected by gill mechanisms, proba-
bly involving NHEs. Indeed, gill NHE mRNA is upregulated in hagfish gill
tissue following an acid infusion (Edwards et al., 2001).
     It might be expected of the iono‐ and osmoconforming hagfish that passive
water influx (found in freshwater teleosts) or passive water eZux (found in
seawater teleosts) would be minimal, raising the question of kidney structure
and function (GriYth, 1994; Fels et al., 1998). The hagfish kidney is unusual in
having large glomeruli (500–1500 mm) and two archinephric ducts or ureters
(Riegel, 1998). There have been only a few studies on kidney function
(Karnaky, 1998). Overall, the hagfish kidney functions in the reabsorption of
glucose and amino acids and secretion of some ions (Munz and McFarland,
1964; Riegel, 1998); however, there are some discrepancies in the literature
whether reabsorption of Naþ or ClÀ occurs (McInerney, 1974; Alt et al., 1981).
     Little progress has been made over the last 40 years or so on understand-
ing of iono‐ and osmoregulation in hagfish. More information is required on
288                                                      PATRICIA A. WRIGHT

the stability of plasma NaCl concentrations under diVerent physiological
conditions (e.g., feeding, exercise, and acid–base disturbances) to determine
if NaCl in the blood is regulated under some circumstances. This would go
a long way in understanding the potential roles of gill MRCs and specific
ion transporters. Osmotic disturbances may be primarily due to the compo-
sition of the diet in this stenohaline fish. Using traditional methods to study
ion fluxes and water balance, it would be valuable to know how hagfish
compensate for a meal with a high water content (e.g., teleost tissues) versus
an isosmotic meal (e.g., invertebrate, elasmobranch tissues).

2.1.2. Ionoregulate, Osmoconform: The Coelacanth Strategy
    An overall understanding of ionic and osmotic regulation is lacking in
Latimeria, with only two surviving species of Coelacanths, L. chalumnae
(Comoro Islands and vicinity) and L. menadoensis (discovered in 1999 oV
the coast of Indonesia) (Holder et al., 1999). It is well accepted that L.
chalumnae plasma and tissues contain elevated concentrations of urea similar
to marine elasmobranchs and they osmoconform to their seawater environ-
ment (Figure 6.1B). What is less clear is whether plasma osmolality is some-
what greater (as in elasmobranchs); equal to or slightly less than (GriYth,
1991) the local marine environment. The confusion over this issue no doubt
relates to a limited number of samples that have been collected from either
frozen (Pickford and Grant, 1967) or moribund specimens (GriYth et al.,
1976). On the basis of available evidence, it does appear that plasma NaCl
concentrations are $25% lower than values reported in marine elasmo-
branchs (McDonald and Milligan, 1992). If this is indeed the case, then
there are several implications. First, other organic osmolytes besides urea
and TMAO (GriYth et al., 1974) must be present in the plasma to add up to
$1000 mosmol kgÀ1. Second, to maintain a combined osmolality of plasma
NaCl at $350 mosmol kgÀ1 (Figure 6.1B), a value almost identical to steno-
haline marine teleosts (346 mosmol kgÀ1; McDonald and Milligan, 1992),
L. chalumnae, must have powerful mechanisms to secrete NaCl. Are these
mechanisms partly in the gill (i.e., chloride‐type cells) or solely in the
postanal gland? The structure of the kidney resembles other osteichthyes
(Locket, 1980), but measurements of a single urine sample are insuYcient to
understand renal function (GriYth et al., 1976). There is much to learn and
physiologists await the opportunity to study multiple live Latimeria.

2.1.3. Ionoregulate, Osmoregulate: The Alternative Strategy
    In seawater, marine teleosts as well as lampreys and sturgeons maintain
body fluid osmolality and NaCl concentrations at about one‐third of
their environment (Morris, 1972; Potts and Rudy, 1972; Beamish, 1980a).
The seawater origins of Agnathans [with lampreys later entering freshwater
6.   OSMOREGULATION                                                              289

(Chang et al., 2006)], but freshwater origins of Teleostei (with subsequent
forays to seawater) suggest that convergent evolution may explain the
remarkable similarity of gill and renal osmoregulatory mechanisms in these
two groups of fish (Bartels and Potter, 2004). Fish ion and osmoregulation
have been well reviewed (Marshall, 2002; Bartels and Potter, 2004; Marshall
and Grosell, 2006).
    Ionic and osmotic gradients result in the constant influx of NaCl and loss
of body water that are counterbalanced by active excretion of NaCl across
the gills and replenishment of water by drinking (Figure 6.1C). The branchial
epithelium, therefore, plays an important role in ionoregulation. The struc-
ture and cellular composition of the lamprey gill have been extensively
described (reviewed by Bartels and Potter, 2004). Chloride cells in lamprey
gills in seawater form long rows and lack accessory cells that are associated
with chloride cells in teleosts. The apical crypts so distinctive in teleost gills in
seawater are absent in lampreys. Despite these small structural diVerences,
the lamprey chloride cells share most of the other characteristics typical of
teleost fishes and other salt‐secreting epithelia, such as a high density of
mitochondria, basolateral membrane elaboration, and a leaky paracellular
pathway (Laurent, 1984; Karnaky, 1986; Bartels and Potter, 2004). Further-
more, the mechanism of active NaCl secretion, involving Naþ/Kþ ATPase,
Naþ/Kþ/2ClÀ cotransporter, and a chloride channel (Marshall and Grosell,
2006), is assumed to be present in seawater lamprey gills (Bartels and Potter,
2004), but this has not been verified.
    The kidney plays a small role in lamprey osmoregulation in seawater,
producing low volumes of urine as would be expected of marine osmoregu-
lators (Logan et al., 1980). Similar to teleosts, lamprey and sturgeon kidneys
preferentially secrete divalent ions in the urine (Pickering and Morris, 1970;
Logan et al., 1980; Krayushkina et al., 1996). Following transfer from
freshwater to brackish water, glomeruli size declines and the tubule
cells and brush border were reduced in two sturgeon species, Acipenser
naccarii (Cataldi et al., 1995) and Huso huso (Krayushkina et al., 1996),
implying reduced function in hypersaline water.

2.2. In Freshwater
    In freshwater, iono‐ and osmoregulation in primitive fishes is accom-
plished as outlined in Figure 6.1D for freshwater lamprey, similar to fresh-
water teleosts. Plasma is hyperosmotic to the surrounding media and
therefore passive water gain and NaCl loss must be compensated by the
elimination of copious volumes of urine and the active uptake of ions via
the branchial epithelium. The majority of studies on primitive fish osmoreg-
ulation have focused on lampreys and sturgeon, with far fewer on bowfin,
290                                                       PATRICIA A. WRIGHT

gar, paddlefishes, and birchirs. Elegant studies by Bull and Morris (1967) and
Morris and Bull (1968, 1970) established that freshwater ammocoete larva of
Lampetra planeri carefully regulate serum and tissue water and ion content,
that external calcium aVects gill permeability to ions, and that sodium influx
is dependent on both internal and external sodium concentrations. The life
cycle of all species of lamprey consists of a larval phase (ammocoetes) in
freshwater, metamorphosis into young adults that may migrate downstream
to the ocean, if anadromous, a marine trophic phase, followed by a return
migration upstream to freshwater streams where they spawn and die
(Beamish, 1980b).
    The gill cell composition of lampreys changes with life cycle and external
salinity. Larval gills have both ammocoete MRCs and intercalated MRCs, as
well as pavement cells. Downstream migrants (freshwater) retain the inter-
calated MRCs and pavement cells, and new chloride cells develop (Youson
and Freeman, 1976; Bartels and Potter, 2004). Choe et al. (2004) further
identified two subtypes of MRCs based on immunohistochemical staining of
the gills of freshwater adult lampreys. They proposed a model of ion trans-
port where MRC‐A that express Naþ/Kþ ATPase are responsible for Naþ
uptake, whereas MRC‐B that stain for carbonic anhydrase and V type
Hþ ATPase transport ClÀ. Verification of this model will require sophisti-
cated separation of MRC‐A and MRC‐B type cells, similar to isolated cell
studies in freshwater teleosts (see review by Marshall and Grosell, 2006).
    Kidney function in freshwater primitive fishes is probably comparable to
teleosts. In bowfin and lampreys, the kidney reabsorbs Naþ and ClÀ, and has
a relatively high glomerular filtration rate that is correlated with urine flow
rate (Logan et al., 1980; Butler and Youson, 1988). The renin‐angiotensin
system (RAS) has been identified in the river lamprey Lampetra fluviatilis
(Cobb et al., 2002; Brown et al., 2005). In vertebrates, the RAS plays an
important role in blood volume and pressure regulation, and studies in
L. fluviatilis indicate that the RAS responds to external water salinity
changes (Rankin et al., 2001; Brown et al., 2005).
    Ionic and osmotic balance in submerged African lungfish has not been
well studied, but presents an interesting challenge given the reduction of gill
surface area (Laurent et al., 1978) and reliance on lung respiration (Perry
et al., 2005a,b). Moreover, the absence of gill convection during terrestrial
episodes may further exacerbate osmoregulatory control. Wilkie et al. (2007)
have discovered that Protopterus dolloi remains in ionic and osmotic balance
after a 6‐month episode on moist land, partly because they exchange water
and ions across their ventral body surface. In fact, their data indicates that
the majority of water exchange in submerged lungfish also occurs across the
ventral skin, which may play a similar role as the pelvic region in amphibians.
This is a fascinating avenue for future study.
6.   OSMOREGULATION                                                         291

2.3. Moving Between the River and Sea

    The life history of anadromous lampreys and sturgeon involves an initial
freshwater phase, followed by migration to the sea and a later return
to freshwater streams to spawn. In lamprey (Petromyzon marinus), anadro-
mous populations are far better at maintaining plasma osmolality with rising
external salinity relative to landlocked populations (Beamish et al., 1978).
Plasma osmolality increases on exposure to saline water in sturgeon
(McEnroe and Cech, 1985; Cataldi et al., 1995; Krayushkina et al., 1996;
                                                      ´      ´
Altinok et al., 1998; McKenzie et al., 2001; Martınez‐Alvarez et al., 2002;
Rodrıguez et al., 2002) and lamprey (Beamish, 1980a). Gill chloride cell size
and number increase when sturgeon are transferred to hypersaline waters
(Altinok et al., 1998) accompanied by an upregulation of gill Naþ/Kþ ATPase
activity in some species (McKenzie et al., 1999; Rodrıguez et al., 2002), but
not all (Jarvis and Ballantyne, 2003). These responses are comparable to
euryhaline teleosts moving from fresh to seawater environments, although
far more details of molecular and cellular changes have been uncovered in
teleosts (for a review see Marshall and Grosell, 2006).
    Gill cell composition in euryhaline lamprey is distinct from teleosts. Prior
to migration down river to the sea, the surface of the gill MRC of young adult
lampreys is covered by the flanges of adjacent pavement cells, with only a
relatively small circular area exposed covered with microvilli (Peek and
Youson, 1979; Mallat et al., 1995; Bartels and Potter, 2004). After young
lampreys enter seawater, the pavement cells retract revealing a larger rectan-
gular microvilli‐free chloride cell surface area, as well the paracellular chan-
nel between adjacent cells widens. These excellent structural studies now need
to be linked to functional investigations to understand the role of ion
transport proteins in specific cell types.


3.1. Toxic Ammonia
    In aqueous solution, ammonia exists both as NH3 and NHþ, according to
the equation:

                        NH3 þ H3 Oþ $ NHþ þ H2 O

The term ammonia represents the sum of the NH3 and NHþ concentrations.
The pK of the reaction is about 9.5 so that at fish blood pH values (pH $ 8)
about 96% of ammonia will be in the NHþ form at 25  C. NHþ is charged and
                                         4                   4
larger than NH3 and therefore has a lower diVusivity compared to NH3.
292                                                        PATRICIA A. WRIGHT

     Ammonia may accumulate in fish under a variety of conditions and, if
severe, can result in convulsions, coma, and eventually death (for reviews see
Ip et al., 2001, 2004; Randall and Tsui, 2002). Most of the fish discussed
in this chapter are primarily ammoniotelic, that is, they excrete primarily
ammonia. If environmental conditions preclude normal rates of ammonia
elimination (e.g., elevated water pH, air exposure, or limited access to water),
then endogenously produced ammonia may accumulate in the fish. Elevated
environmental ammonia occurs in natural and hatchery freshwaters. The
reversal of the branchial blood‐to‐water ammonia gradient results in ammo-
nia uptake and elevated plasma and tissue ammonia concentrations in a
variety of fish species, including lungfish (Chew et al., 2005).
     The toxicity of elevated environmental ammonia varies with water pH,
temperature, salinity, and oxygen levels (for review see Ip et al., 2001). As
well, intraspecific variation, developmental eVects, nutritional status, and
acute versus chronic exposure all impact ammonia toxicity. There is a paucity
of data on primitive fishes. In one study by Fontenot et al. (1998), the 96‐h
median‐lethal concentration (96‐h LC50) for NH3 for fingerling shortnose
sturgeon Acipenser brevirostrum was 0.58 Æ 0.21 mg literÀ1 (mean Æ SD,
18  C). As a comparison, the 96‐h LC50 value for NH3 in both rainbow trout
(Oncorhynchus mykiss) and fathead minnow (Pimephales promelas) was
0.37 mg literÀ1 (14  C) (Thurston et al., 1981). With very little solid data on
ammonia tolerance between diVerent orders of fishes, it is not easy predicting
which fish might demonstrate a higher tolerance to environmental ammonia.
The strongest guidelines may be environmental or ecological considerations.
For example, there is evidence that freshwater fishes are less susceptible to
ammonia toxicity compared to their seawater counterparts (Ip et al., 2001).
The tolerance to elevated water ammonia levels may be high in fish that form
aggregations, burrow into confined spaces, or encounter low water volumes.
Thus, there is an apparent correlation between ammonia and hypoxia toler-
ance in fish (for review see Walsh et al., 2006). For example, high densities of
hagfish have been observed feeding on whale carcasses in deep ocean envir-
onments (Martini, 1998). Rotting flesh combined with hundreds of relatively
large ($0.5 m) ammonia‐excreting hagfish may create a local environment
high in ammonia. As well, hagfish normally reside in mud burrows on the
ocean floor, such a confined space may also result in elevated external
ammonia. Due to these circumstances, it is possible that hagfish may have
evolved a high tolerance to ammonia, but this has not been tested.
     Another Agnathan, the lamprey burrows into soft mud for several years
in the larval stage (ammocoete) and feeds on detritus (Moore and Mallatt,
1980). Depending on the rate of water exchange near the ammocoete sur-
face, endogenous ammonia may accumulate in the local environment. In
fact, the 96‐h LC50 value for NH3 was 1.7 mg literÀ1 in P. marinus
6.   OSMOREGULATION                                                          293

ammocoetes (Wilkie et al., 1999), fivefold higher relative to the values
reported for teleosts (see above).
    Lungfish encounter limited water availability or a complete absence of
water during estivation (see Section 3.4), and therefore may frequently
encounter ammonia loads and have a correspondingly higher tolerance for
ammonia. LC50 values have not been reported in the African lungfish, but
this group of fish certainly appears to be ammonia tolerant. P. dolloi survives
6 days in water containing 100 mmol literÀ1 NH4Cl and barely accumulates
ammonia in the extracellular compartment (Chew et al., 2005). It has recently
been discovered that their remarkable insensitivity to elevated external
ammonia may be partly linked to the excretion of acid. Wood et al. (2005a)
found that when P. dolloi are exposed to 309 mmol literÀ1 NH4Cl for 7 days
they excrete both CO2 and titratable acid (e.g., Hþ) into their external
environment, lowering water pH to as low as pH 3.7 in one case (even with
aeration). Environmental acidification ensures that the highly diVusible NH3
remains low in the external water, thereby lowering the overall uptake of
ammonia by the lungfish. P. dolloi also detoxifies excessive ammonia by
conversion to urea via the urea cycle (see Section 3.2.2) when confronted
with exceptionally high external ammonia concentrations (Chew et al., 2005).
Hence, the African lungfish may have multiple strategies of coping with this
toxic compound. It is likely that African lungfish rank up there with other
highly ammonia‐tolerant species, such as the mudskipper (Periophthalmodon
schlosseri), that can survive also in 100‐mmol literÀ1 NH4Cl and has a 96‐h
LC50 for NH3 of 7.6 mg literÀ1 (Ip et al., 2004)!
    The brain of fish, as well as all vertebrates, is the most vulnerable organ to
elevated plasma ammonia levels (Felipo and Butterworth, 2002; Walsh et al.,
2006). This topic has been extensively reviewed elsewhere and will only be
briefly described here. In the case of high extracellular ammonia, if NH3
is the primary permeant species, cytosol pH will increase. If mostly NHþ        4
enters the cell, intracellular pH will decrease. Any pH change will influence
the function of intracellular processes. Ammonia has numerous other eVects
that appear to be due to the unique properties of the NH3 or NHþ molecules
themselves. NHþ can directly substitute for Kþ or Hþ in ion exchangers,
disrupting ion balance and nerve propagation (Cooper and Plum, 1987).
Elevated brain ammonia in fish interferes with normal cell metabolism and
the synthesis of the neurotransmitter, glutamate (Wicks and Randall, 2002).
The most important detoxification enzyme, glutamine synthetase (GSase),
catalyzes the conversion of glutamate and NHþ to glutamine and is induced
in the brain of some ammonia‐exposed teleost fishes (Wicks and Randall,
2002; Wright et al., submitted for publication). As well, tolerance to elevated
environmental ammonia levels is correlated with constitutive activities of
GSase in the brain of closely related Batrachoididae fishes (Wang and
294                                                        PATRICIA A. WRIGHT

Walsh, 2000). Comparable data on brain metabolism in ammonia‐exposed
primitive fishes would be a step toward understanding the evolution of
ammonia detoxification mechanisms.

3.2. Synthesis of Nitrogen End‐Products

    Nitrogen end‐products are the result of protein catabolism. Ingested or
cellular proteins are hydrolyzed to the component amino acids by proteolytic
enzymes. Excess amino acids are catabolyzed forming ammonia, the primarily
nitrogen end‐product in fishes. Ammonia may be further ‘‘repackaged’’ as urea,
glutamate, or glutamine. Several reviews have been written on nitrogen metab-
olism and excretion in fishes (Wood, 1993, 2001; Walsh, 1998; Anderson, 2001;
Ip et al., 2001; Walsh and Mommsen, 2001; Wilkie, 2002).
3.2.1. Ammonia
    Figure 6.2 describes the three pathways for ammonia synthesis. Amino
acid transferase enzymes transfer an amino group from the L‐amino acid to
a‐ketoglutarate (a‐KG) forming glutamate and a‐keto acid (Figure 6.2A).
Mitochondrial glutamate dehydrogenase (GDH) catalyzes the conversion of
glutamate to a‐KG and NHþ . Hepatic and red muscle mitochondrial
GDH activities are comparable between bowfin (Amia calva), a Holostean
fish and various teleost species (Chamberlin et al., 1991; Felskie et al., 1998).
In lamprey (P. marinus) GDH activities in liver, intestine, and muscle vary
with developmental stage. In the parasitic phase, liver GDH activity and
protein abundance were six times higher relative to the lamprey ammocoete
or upstream migrant (Wilkie et al., 2006). The authors proposed that high
levels of GDH in the liver of parasitic lamprey allow for rapid catabolism of
amino acids when feeding opportunities arise. This supposition is further
strengthened by parallel high activities of two liver transferase enzymes,
alanine, and aspartate aminotransferase in the parasitic lampreys (Wilkie
et al., 2006).
    Ammonia is also created when AMP is degraded to IMP, catalyzed by
AMP deaminase (Figure 6.2B). Although present in fish liver (Casey
and Anderson, 1983), it probably only makes a significant contribution
to ammonia synthesis in skeletal muscle tissue after exhaustive exercise
(Mommsen and Hochachka, 1988; Wright et al., 1988). Finally, the break-
down of glutamine also results in the generation of ammonia and glutamate,
catalyzed by glutaminase (GLN) (Figure 6.2C). The reverse reaction, cata-
lyzed by GSase, consumes ammonia (Figure 6.2D). Thus, the balance
between the two enzymes will determine the net ammonia synthesis in fish
tissues (Chamberlin et al., 1991). In bowfin, liver GLN activity is higher
than GSase activities (Chamberlin et al., 1991), suggesting a net production
6.   OSMOREGULATION                                                                    295

Fig. 6.2. Pathways for ammonia synthesis in and out of the mitochondrion. (A) L‐Amino acids
are transaminated forming glutamate and an a‐Keto acid. Glutamate enters the mitochondrion
where the enzyme GDH deaminates glutamate forming a‐ketoglutarate (a‐KG) and NHþ . (B) In
the purine nucleotide cycle, the adenylate AMP is degraded to IMP and NH3, catalyzed by AMP
deaminase. (C) Glutamine may enter the mitochondrion where the enzyme GLN catalyzes the
reaction forming glutamate and NH3. (D) Glutamate and NHþ are combined by the enzyme
GSase to form glutamine. GS is typically cytosolic in ammoniotelic fish, but mitochondrial in
ureotelic fish.

of ammonia by the liver typical of other ammonotelic fishes. Ureotelic species
such as elasmobranchs and most likely the coelacanth (not measured) have
much higher liver GSase activities because available ammonia is scavenged to
form glutamine, the nitrogen donating substrate for the urea cycle (see below).
3.2.2. Urea
    Urea is formed by three known pathways in fish: uricolysis, arginolysis,
and the urea cycle (Figure 6.3). Uric acid arises from a purine ring that is
formed by a series of complex reactions involving glutamine, aspartate,
glycine, HCOÀ , and phosphoribosyl pyrophosphate (PRPP) (Figure 6.3A).
Uric acid degradation to urea occurs in the peroxisomes in fish (Noguchi
296                                                                  PATRICIA A. WRIGHT

Fig. 6.3. Pathways for urea synthesis in the peroxisome (A) and mitochondrion (B), (C).
Uricolysis is depicted in (A), where a purine ring is formed by a series of complex reactions
involving 2 glutamine, aspartate, glycine, HCOÀ , and PRPP. Uric acid is degraded by uricase
(URC) to allantoin which is further degraded by allantoinase (ALN) and allantoicase (ALC) to
urea. Ariginolysis (B) is the simple conversion of arginine to ornithine and urea catalyzed by
arginase. The urea cycle in fish is initiated by the enzyme carbamoyl phosphate synthetase III
(CPS III) that combines glutamine and HCOÀ to form carbamoyl phosphate, which in turn is
converted to citrulline catalyzed by ornithine carbamoyl transferase (OTC). Citrulline is con-
verted to arginine by two enzymes, argininosuccinate synthetase (ASS) and argininosuccinate
lyase (ASL), the final step is arginolysis (B) forming urea.
6.   OSMOREGULATION                                                         297

et al., 1979) and involves three enzymes, uricase (URC), allatoinase (ALN),
and allatoicase (ALC).
    Activities for all three uricolytic enzymes were first reported in lungfish in
an aquatic habitat (Brown et al., 1966). Later studies revealed that urea
synthesis via the urea cycle in estivating lungfish far surpassed the capacity
of uricolysis (Forster and Goldstein, 1966). It is likely that all fish have the
capacity to degrade uric acid because the pathway plays an important role in
nucleic acid metabolism. It is surprising therefore that in hagfish (Bdellos-
toma cirrhatum renamed Eptatretus cirrhatus), no uricolytic enzymes were
detected (Read, 1975), a finding that has not been confirmed in this or other
hagfish species. In contrast, the full suite of uricolytic enzymes were found in
liver of lamprey ammocoetes (P. marinus), and the level of activities were
similar to teleost values (Wilkie et al., 1999).
    The first uricolytic gene, uricase (or urate oxidase), has been cloned in
African lungfish (P. annectens), as well as in several teleost species (Andersen
et al., 2006). Although the allantoicase gene has been identified in the puVer
fish (Fugu rubripes) genome database (Vigetti et al., 2003), cloning of allan-
toicase and allantoinase genes in fish has not been successful to my knowl-
edge. Given that several enzymes in the uricolytic pathway have been lost
during the evolution of higher vertebrates, much would be gained from
understanding the evolution of the genes coding for the uricolytic enzymes
in primitive as well as other fishes.
    Urea can also be formed from arginine degradation in fish mitochondria
catalyzed by arginase (ARG) (Figure 6.3B), independent of a complete urea
cycle. ARG is widespread in fish tissues, including hagfish (Read, 1975),
lamprey (Read, 1968; Wilkie et al., 1999, 2004, 2006), lungfish (Janssens
and Cohen, 1966), coelacanth (Brown and Brown, 1967), sturgeon and gar
(Cvancara, 1969), as well as bowfin (Felskie et al., 1998). Two mitochondrial
isoforms of ARG, ARG type I and II, are coded by two genes in puVer fish
(Takifugu rubripes) and four genes in rainbow trout (O. mykiss) (Wright
et al., 2004), whereas ARG type I in terrestrial ureotelic vertebrates is
cytosolic (Ikemoto et al., 1990). At what stage during the transition to a
terrestrial habitat did ARG type I lose its mitochondrial leader sequence?
Mommsen and Walsh (1989) proposed that this shift in the intracellular
location of liver ARG first appeared in the lungfish. Although enzyme
activity data support this notion (Mommsen and Walsh, 1989), the sequenc-
ing of lungfish ARG genes (type I and II) would provide valuable informa-
tion for a more complete phylogenetic analysis and lead to a better view of
the evolution of the urea cycle in vertebrates.
    The urea cycle is the main pathway for urea synthesis in terrestrial
vertebrates (amphibians, mammals), as well as elasmobranchs (Anderson,
2001), coelacanths, lungfish (GriYth, 1991), and a few teleosts (Walsh and
298                                                        PATRICIA A. WRIGHT

Mommsen, 2001). Five enzymes form the backbone of the pathway: carba-
moyl phosphate synthetase III (CPS III), ornithine carbamoyl transferase
(OTC), argininosuccinate synthetase (ASS) and lyase (ASL), and ARG
(Figure 6.3B and C). CPS III requires glutamine as the nitrogen donating
substrate and, therefore, mitochondrial GS is considered an important
accessory enzyme. The properties of CPS III and the subcellular location of
ARG distinguish the fish urea cycle from that of amphibians and mammals
(Anderson, 2001). In terrestrial vertebrates, CPS I utilizes ammonia as the
N‐donating substrate and ARG is cytosolic. A third CPS cousin, CPS II, is
prevalent in all vertebrate tissues, is part of the pyrimidine pathway, requires
glutamine as a substrate, but does not require the eVector N‐acetyl glutamate
(required by both CPS I and III).
     Confusion in the literature arose a few decades ago when it was first
reported that CPS activity was fairly common in fish (Cvancara, 1974), but
subsequent work clarified the importance of careful assay technique to pro-
vide the correct conditions for the urea cycle‐related CPS III and separate it
from CPS II (Felskie et al., 1998). Thus, it is now clear that hagfish (Read,
1975; Mommsen and Walsh, 1989), lamprey (Wilkie et al., 1999, 2004),
and bowfin (Felskie et al., 1998) have extremely low or nondetectable levels
of CPS III and lack a functional hepatic urea cycle. There is limited informa-
tion on urea cycle enzymes in sturgeon, paddlefish, and bichir; however, the
absence of the pathway is reported in Mommsen and Walsh (1989). Despite
this, the bichir was assigned a CPS type III enzyme by Mommsen and
Walsh (1989) and would be interesting to investigate further given this fish’s
predilection for river margins, flood plains, and swamps, similar to lungfishes.
     The coelacanth, L. chalumnae, first came to the attention of urea cycle
enthusiasts with the reports of Pickford and Grant (1967) and Brown and
Brown (1967) that blood and tissue urea levels as well as OTC and ARG
activities were similar to the ureosomotic elasmobranchs (GriYth, 1991).
A number of follow‐up studies expanded on these initial findings but
fascinating questions remain. For example, during early development, coela-
canth pups are carried within the oviducts of the mother (Wourms et al.,
1991) and obtain nutrition from large attached yolk sacs (Locket, 1980). Are
these developing young ureosmotically independent as have been found in
little skate (Raja erinacea) embryos (Steele et al., 2004) or is urea cycle
capacity only apparent after release from internal maternal support?
     African lungfish express all urea cycle enzymes (Janssens and Cohen,
1966) and detoxify ammonia to urea when water supplies become limited
(Janssens and Cohen, 1968; see also Section 3.4). Mommsen and Walsh
(1989) classified CPS in the lungfish P. aethiopicus as type I, with a preference
for ammonia over glutamine as the N‐donating substrate. This designation
was later challenged by Loong et al. (2005) who claimed that hepatic CPS
activities in P. aethiopicus and P. annectens were 30‐ to 60‐fold higher if
6.   OSMOREGULATION                                                        299

glutamine was the available substrate rather than ammonia. It is possible,
however, that both subtypes of CPS exist, but that the developmental stage
or previous environmental conditions of the lungfish play a role in deter-
mining whether CPS I or III dominates (P. Walsh, personal communication).
Clearly, sequence information on lungfish CPS genes would help to illuminate
the controversy.

3.3. Excretion
    Aquatic animals tend to be ammoniotelic, whereas terrestrial animals
excrete mostly urea (ureotelic) or uric acid (uricotelic) (reviewed by Wright,
1995). The key nitrogen excretory product in fish is ammonia, but elasmo-
branchs (and probably the coelacanth) release primarily urea (ureotelic) as a
by‐product of their ureosmotic strategy (Table 6.1). Other nitrogenous end‐
products including amino acids, proteins, creatine, creatinine, and unknown
substances, which may account for 20–40% of the total nitrogen excreted in
teleost fishes, are less well understood (Kajimura et al., 2004).

3.3.1. Ammonia
    Fish excrete ammonia mostly as NH3 down the blood‐to‐water partial
pressure gradient (Wilkie, 2002; Evans et al., 2005). To determine if
NH3 diVusion dominates at the gill, researchers have manipulated external
water pH (Wright and Wood, 1985). The diVusion trapping model predicts
that NH3 excretion will increase in acid environments and decrease in
alkaline environments (Wright and Wood, 1985; Wilkie and Wood, 1991).
This phenomena is nicely illustrated in sturgeon (Acipenser ruthenus) finger-
lings acclimated gradually for 1 week to water pH values ranging from pH 4.0‐
to 9.4 (fish died at pH 9.6) (Figure 6.4). If one takes $500 mmol kgÀ1ÁhÀ1
as the control excretion rate (pH 7–8.4), then there is an 80% increase in Jamm
at pH 4.0 and a 34% decrease in Jamm at pH 9.4, with virtually no change
in Jurea. The very high rate of Jamm under acid conditions would be hard
to maintain and it is not surprising that some mortalities were noted after
2 days at pH 4.0.
    There is evidence for some branchial NHþ eZux possibly linked to Naþ
influx in fish (Wilkie, 2002; Evans et al., 2005). Interestingly, in the hagfish
M. glutinosa, Evans (1984) proposed that branchial ammonia excretion was
dominated by NH3 diVusion, with no evidence for Naþ‐dependent NHþ             4
transport. In general, gill NHþ diVusion may be more important in seawater
compared to freshwater fish. In seawater lamprey and sturgeon, leaky para-
cellular junctions adjacent to gill MRCs (see Section 2.1.3) may provide a
passageway for NHþ . In contrast, in the hagfish gill MRCs are tightly
associated with pavement cells and this arrangement probably prevents
NHþ diVusion between cells.
                                                                    Table 6.1
                                      Ammonia (Jamm) and Urea (Jurea) Excretion Rates and the Percentage of
                                  Nitrogen Wastes Excreted as Urea (% urea) in Hagfish, Lamprey, Elasmobranch,
                                                     Lungfish, Sturgeon, and Bowfin Species

                                          Jamm (mmol N         Jurea (mmol N
                    Species                 kgÀ1ÁhÀ1)            kgÀ1ÁhÀ1)        Urea (%)             Comments        References

       Eptatretus stoutii                          73             5                   6           SW                Walsh et al., 2001

       Myxine glutinosa                           218             –                   –           SW                Evans, 1984
                                                  200             –                   –           SW                McDonald et al.,
        Petromyzon marinus                         50             9                   18          FW (ammocoetes)   Wilkie et al., 1999
                                                  100             15                  13          FW (adults)       Wilkie et al., 2004
                                                 2500             200                 7           FW (adults, fed   Wilkie et al., 2004
                                                                                                   trout blood)
        Entosphenus tridentatus                   119             0                   0           FW                Read, 1968
        Squalus acanthias                          11             547                 98          SW                Wood et al., 1995
        Raja erinacea                             150             411                 73          SW (adult)        Steele et al., 2005
                                                  124             111                 47          SW (embryo)       Steele et al., 2004
        Protopterus dolloi                        170             21                  11          2‰                Wood et al., 2005
        Protopterus aethiopicus         63   66    51   FW                  Loong et al., 2005
        Protopterus annectens           96   50    34   FW                  Loong et al., 2005
        Protopterus dolloi             133   42    24   FW (fasted)         Lim et al., 2004
                                       265   221   46   FW (fed)            Lim et al., 2004
        Acipenser baeri                469   –     –    FW (juvenile)       Dabrowski et al.,
        Acipenser oxyrinchus desotoi   208   16    7    FW                  Altinok and
                                                                              Grizzle, 2004
                                       142   15    10   9‰ SW               Altinok and
                                                                              Grizzle, 2004
        Acipenser oxyrhynchus          438   –     –    FW (juvenile)       KieVer et al.,
        Acipenser brevirostrum         381   –     –    FW (juvenile)       KieVer et al.,
        Acipenser gueldenstaedti       724   95    12   FW (juvenile)       Gershanovich

                                                                              and Pototskij,
                                       594   130   18   10‰ SW (juvenile)   Gershanovich
                                                                              and Pototskij,
        Amia calva                     607   60    9    FW                  McKenzie and
                                                                             Randall, 1990
302                                                                 PATRICIA A. WRIGHT

Fig. 6.4. Ammonia (Jamm) and urea (Jurea) excretion rates (mmol kgÀ1 hÀ1) in Sturgeon finger-
lings A. ruthenus in response to 1 week of exposure to water of varying pH. [Reprinted from
Gershanovich and Pototskij (1995) with permission from Elsevier.]

    Gill ammonia transporters have been isolated in two teleost species,
Rivulus marmoratus (Hung et al., 2007) and O. mykiss (Nawata, et al., submit-
ted manuscript), which share sequence similarities to the Rhesus‐associated
glycoproteins (RhG) (Huang and Peng, 2005). There is an ongoing controversy
whether RhG proteins transport NH3 (Ripoche et al., 2004), NHþ (Nakhoul
et al., 2005), and/or mediate NHþ /Hþ exchange (Handlogten et al., 2004) in
mammals. In crab gills, an Rh‐like protein (RhCM) is thought to actively
transport NHþ across the cuticle (Weihrauch et al., 2004). Huang and Peng
(2005) have reported the presence of ‘‘genuine Rh genes in hagfish’’ and suggest
further characterization of these genes may shed light on CO2 conductance
across red blood cell membranes. Thus, across the spectrum of organisms and
tissues many putative roles have been associated with RhG genes and the path
forward in fish will be clearly very interesting.
    The major site of Jamm is typically the gills in fish (Smith, 1929). Divided
chamber experiments performed by Read (1968) on the lamprey Entosphenus
tridentatus demonstrated that 87% of the ammonia was eliminated across the
gills, 8% across the skin, and $4% via the kidneys. Little attention has been
paid to cutaneous ammonia excretion, but premetamorphic lamprey larvae
have a thinner integument (Youson, 1980) and possibly rely less on branchial
excretion while in burrows compared to adults; however, this has not been
tested. In the submerged lungfish P. dolloi, ammonia and urea excretion are
almost equally partitioned between the anterior (internal and external gills)
and posterior end (most of the skin and urinary opening) (Wood et al., 2005).
6.   OSMOREGULATION                                                          303

What percentage of the posterior excretion is attributable to the skin versus
urine is unknown; however, the skin may be the key site of urea elimination
following prolonged terrestrialization in lungfish (see Section 3.4).
    Feeding has a profound eVect on nitrogen excretion rates in fish, but most
Jamm values in the literature are for starved fish (Wood, 2001). One of the
most dramatic examples of the influence of nutrition is found in the parasitic
lamprey (P. marinus) (Wilkie et al., 2004), where postprandial Jamm was
25 times higher after a blood meal (Table 6.1). Parasitic lampreys may
consume up to 30% of their body weight/day and are very eYcient at
assimilating the energy in a blood meal (Farmer et al., 1975). The African
lungfish (P. dolloi) appears to take a diVerent approach to a surfeit of amino
acids following feeding. Although Jamm does increase significantly after a
meal, urea synthesis is stimulated to a greater extent resulting in higher tissue
urea contents and a greater proportion of nitrogen excreted as urea (Lim
et al., 2004).
3.3.2. Urea
    In jawless and ray‐finned primitive fishes, Jurea constitutes 6–18% of total
ammonia and urea excretion (Table 6.1). Jurea depends on both the simple
diVusion of urea as well as facilitated transport (Walsh and Smith, 2001). The
first fish urea transporter (UT) was isolated and characterized in the dogfish
shark (Squalus acanthias) kidney (ShUT; Smith and Wright, 1999). ShUT,
homologous to the mammalian facilitated transporter UT‐A2 family, shares
sequence similarity with UTs cloned from the kidneys of other elasmobranch
species (Janech et al., 2003, 2006; Morgan et al., 2003; Hyodo et al., 2004).
A survey of several marine fish indicates that UT gill expression may be fairly
widespread, although no signal was detected in hagfish (Eptatretus stoutii)
(Walsh et al., 2001). This negative result requires further investigation
because a teleostean UT probe was used. Other UT isoforms may be present
in hagfish or branchial urea transport may be solely dependent on simple
diVusion. Due to the unique position of hagfish in the fish evolutionary tree,
a more complete picture of Agnathan urea transport would be valuable.
    There have been only a few studies in the primitive fishes that report
fluctuations in Jurea with changing physiological conditions. Jurea is not
particularly sensitive to changes in the external water pH (Figure 6.4) or
salinity (Altinok and Grizzle, 2004). Feeding enhances Jurea, just as it does
Jamm (see above). In parasitic lamprey feeding on rainbow trout, Jurea was
elevated by 15‐fold initially and remained elevated for 8 h after the
meal (Wilkie et al., 2004). More remarkably, two lampreys (P. marinus)
were caught parasitizing basking sharks and Jurea was as high as
$9000 mmolÁNÁkgÀ1ÁhÀ1 after the meal (Wilkie et al., 2004). This surge in
Jurea is presumably necessary to clear the lamprey body fluids of the excessive
304                                                      PATRICIA A. WRIGHT

urea concentrations taken in with the shark blood. The rapid eZux of urea
under these conditions may be dependent, in part, on the upregulation of gill
UTs (Wilkie et al., 2004).

3.3.3. Retention of Urea in Coelacanths and Elasmobranchs
    The coelacanth, holocephalans (chimaeras), and marine elasmobranchs
evolved a strategy of osmoregulation that sets them apart from all other
known fish species, namely the retention of urea to counterbalance the
osmotic strength of seawater. Like many other evolutionary experiments, it
is not apparent why an excretory waste product (urea) would be retained at
relatively high concentrations as an osmolyte (for a discussion see Walsh and
Mommsen, 2001). Kirschner (1993) compared the energetic costs of hypos-
motic regulation by marine teleosts with ureosmotic regulation by marine
elasmobranchs and concluded that they were similar.
    Urea is less toxic than ammonia at the same concentration, but urea at
high concentrations denatures proteins (Yancey, 2001). In elasmobranchs,
the destabilizing eVects of urea are counterbalanced by methylamines or
other organic osmolytes (Yancey, 2001). In some marine elasmobranchs,
the ratio of urea to TMAO concentrations is $2:1, whereas in other species
a variety of counteracting osmolytes along with TMAO have an additive
eVect (Steele et al., 2004, 2005). Embryos of the oviparous little skate,
R. erinacea, have a urea: TMAOþ other stabilizing osmolytes ratio of 2.3:1
(4 months) and 2.7:1 (8 months) (Steele et al., 2004), whereas the ratio was
1.68:1 in the skeletal muscle of adult skates (Steele et al., 2005). These
findings imply developmental changes in osmolyte regulation.
    Coelacanth hemoglobin is unaVected by urea concentrations >3 M,
similar to elasmobranchs (Mangum, 1991). Are other coelacanth proteins
sensitive, insensitive, or urea‐requiring, as has been found in some elasmo-
branchs? In the coelacanth, skeletal muscle TMAO was $300 mmol literÀ1
(Lutz and Robertson, 1971) considerably higher than the little skate value of
$50 mmol literÀ1 (Steele et al., 2005) or $180 mM in the shark Scyliorhinus
canicula muscle (Robertson, 1989). These diVerences raise many interesting
questions about the coelacanth counterbalancing osmolyte strategy. For
example, do coelacanth pups within the mother (Wourms et al., 1991) retain
the same organic osmolyte ratios as adults? Are these osmolyte concentra-
tions sensitive to external salinity changes, as reported in elasmobranch
embryos (Steele et al., 2004) and adults (Steele et al., 2005)?
    Maintaining elevated urea concentrations in body fluids is a considerable
challenge for coelacanths and elasmobranchs, given the large blood
($400 mmol literÀ1)‐to‐water ($0 mmol literÀ1) gradient across the gills
and the obligatory release of more or less isosmotic urine. In marine elasmo-
branchs, renal reabsorption of urea has been long established, but the
6.   OSMOREGULATION                                                         305

mechanisms involved have not been resolved. The passive countercurrent
reabsorption of urea by the kidney (Boylan, 1972; Lacy et al., 1985) is
probably dependent, in part, on UT proteins (Smith and Wright, 1999)
with diVerent functional characteristics and regional heterogeneity (Walsh
and Smith, 2001; Janech et al., 2003, 2006; Morgan et al., 2003; Hyodo et al.,
2004). Although sequence and tissue distribution information on UTs in
elasmobranchs is valuable, what is now needed is a more complete under-
standing of the functional role of various tubule segments. This work has
been impeded by the complexity of the elasmobranch nephron structure. On
the other hand, the coelacanth kidney is thought to be more similar to the
Osteichthyes (see Section 2.1.2); a more detailed structural analysis needs to
be performed for comparisons with the elasmobranch nephron. GriYth
(1991) postulated that coelacanths do not reabsorb urea from the renal
filtrate based on samples collected from the urinary bladder of a moribund
specimen (GriYth et al., 1976). Urea is energetically expensive to produce
and, therefore, it is hard to imagine how the coelacanth could manage such a
high rate of urea loss via the urine unless urine flow is remarkably low.
    A secondary active Naþ/urea antiporter in the gills of the dogfish shark is
thought to transport urea out of the gill epithelial cells back into the blood
against the urea concentration gradient (Fines et al., 2001). The gene coding
for this Naþ/urea antiporter has not been isolated in elasmobranchs, nor in
the mammalian kidney where active Naþ‐coupled urea transport is thought
to play an important role (Kato and Sands, 1998). Dogfish gill basolateral
membranes have a very high cholesterol content (Fines et al., 2001) that
decreases the permeability to urea (Pugh et al., 1989). Hill et al. (2004)
reported that apical and basolateral gill membrane vesicles in dogfish and
marine flounder share relatively low permeabilities to water and urea.
A picture of branchial urea retention in elasmobranchs is starting to emerge,
but no comparable data is available for coelacanth. Even though it is unlikely
that physiologists will obtain live coelacanth for whole‐animal experi-
ments, fresh gill and kidney tissue could provide a wealth of information
on ultrastructure and molecular composition.
3.3.4. The Developmental Perspective
    An understanding of nitrogen excretion during early development in
teleost fishes (Korsgaard et al., 1995; Wright and Fyhn, 2001) may shed
some light on unstudied ammoniotelic primitive fishes (hagfish, lungfish,
and Actinopterygii fishes) or provide comparisons to the one group that
has been examined, the lampreys. Ammonia excretion (Jamm) can be detected
in many teleost embryos very early after fertilization and depending on the
species, Jurea can account for a significant fraction of total nitrogen excretion
(Wright and Fyhn, 2001). GriYth (1991) proposed that urea synthesis via
306                                                         PATRICIA A. WRIGHT

the urea cycle arose in early gnathostome fishes as a protective mechanism to
ensure low tissue ammonia levels during a long embryonic development
phase solely dependent on yolk proteins and amino acids for energy. Indeed,
urea cycle enzymes, including the key enzyme CPS III, are expressed in
freshwater and marine teleosts embryos encompassing a variety of early
life histories (Wright et al., 1995; Chadwick and Wright, 1999; Terjesen
et al., 2001; Barimo et al., 2004). Interestingly, in most of these species
the urea cycle is not operational in the adult stage. The exception is the
embryos of the gulf toadfish, Opsanus beta, which develop into ureagenic
adults, turning on urea synthesis under stressful conditions (for a review see
Walsh, 1997).
    An alternative route for ammonia detoxification during early develop-
ment is glutamine synthesis, which may or may not feed into the urea cycle.
Early induction of GSase genes in rainbow trout embryos and subsequent
formation of the active enzyme before hatching may be necessary to prevent
excessive accumulation of ammonia (Essex‐Fraser et al., 2005).
    Given this background on teleost early development, it would be
fascinating to know more about the ontogeny of nitrogen excretion in
the oldest living vertebrate, the hagfish. Female hagfish are thought to
produce 20–30 yolky encapsulated embryos varying in size between 20 and
70 mm (Martini, 1998). In large embryos such as these, ammonia diVusion
to the surrounding seawater would be comparatively slow, and therefore
detoxification pathways such as the urea cycle and/or glutamine synthesis
might be imperative. Embryonic hagfish have rarely been found over the last
100 years (Martini, 1998).
    The picture is brighter for the lampreys. Larval ammocoetes synthesize
low levels of urea via uricolysis, not the urea cycle (Wilkie et al., 1999). When
exposed to elevated external ammonia, plasma and tissue ammonia levels
increased in ammocoetes without changes in urea or glutamine (Wilkie et al.,
1999). Premetamorphic lampreys have a depressed metabolic rate compared
to postmetamorphic stages (Wilkie et al., 2001), partly explaining the low
rates of nitrogen excretion relative to postmetamorphic stages (Table 6.1).
After metamorphosis, parasitic and upstream‐migrant lampreys express only
very low or nondetectable levels of the urea cycle enzymes, CPS III and OTC
(Wilkie et al., 2006). Is the urea cycle expressed in embryonic lamprey?
Embryos of P. marinus are about 1–2 mm in diameter (Richardson and
Wright, 2003), have a relatively large yolk sac, and hatch after $20 days
(22  C) in the nest (Applegate, 1950). One might predict that lamprey
embryos consuming yolk proteins may induce urea cycle enzymes prior to
hatching, but later repress the activity of urea cycle enzymes when metabolic
rate and protein intake is low when larvae inhabit mud burrows and feed on
6.   OSMOREGULATION                                                        307

3.4. The Challenges of Estivation

     The three families of lungfish tolerate terrestrial conditions to varying
degrees. Some species of African lungfish Protopterus estivate (i.e., survive
the dry season by forming a cocoon or a protective layer of dried mucus and
reduce metabolic rate), whereas the South American Lepidosiren will partially
estivate in a moist environment, and the Australian Neoceratodus is
completely aquatic and does not estivate. GriYth (1991) and Graham
(1997) presented excellent reviews of the older literature on nitrogen metab-
olism and excretion in the three families of lungfish. There have been several
new studies on the African lungfish Protopterus, mostly due to the ease of
shipping these animals to laboratories far away from their native habitat.
P. dolloi form dry, brown mucus cocoons in the laboratory when water
is removed and animals are kept slightly moist for 30 or 40 days (Chew
et al., 2004; Wood et al., 2005).
     During estivation, Protopterus and Lepidosiren maintain low tissue
ammonia levels, but accumulate urea in order to avoid ammonia toxicity
(Janssens, 1964; Carlisky and Barrio, 1972; Chew et al., 2004; Wood et al.,
2005). Urea synthesis in these air‐exposed lungfish occurs via the hepatic urea
cycle (see Section, 3.2.2). The activities of urea cycle enzymes were enhanced
by approximately twofold in P. dolloi, although the increase is remark-
ably modest given that these fish were without water for 40 days (Chew
et al., 2004). Moreover, the data suggest that P. dolloi maintains a high
reserve capacity for urea synthesis under control or immersed conditions,
even when they are ammoniotelic (Chew et al., 2003). This may be beneficial
if low water or impending drought is not readily anticipated by the fish. In the
nonestivating Neoceratodus forsteri, the rate of urea synthesis is 100 times
lower than that in Protopterus (Goldstein et al., 1967).
     A ‘‘washout’’ of urea was observed in Protopterus when estivating lung-
fish were returned to water (Smith, 1930; Janssens, 1964). An initial pulse of
Jurea (peak 0–1 h) was followed by a second pulse (peak $12 h) of greater
magnitude (Wood et al., 2005). When lungfish were placed in divided cham-
bers to separate the anterior (gills) from the posterior (most of the skin and
urine/feces) end of the fish, more Jurea occurred from the posterior end in the
second phase of the urea ‘‘washout’’ (Figure 6.5). Further characterization of
the second pulse of Jurea suggested that facilitated type UTs may have been
mobilized to accommodate the enormous flux at this time (Wood et al.,
2005). Isolation of lungfish UTs and their tissue distribution will be an
important next step.
     The environmental or endogenous signal(s) that stimulates estivation has
not been identified in lungfish. Ip et al. (2005) hypothesized that a subtle
increase in external salinity as the river water evaporates prior to a drought
308                                                                    PATRICIA A. WRIGHT

Fig. 6.5. Nitrogen excretion in the African lungfish (Protopterus dolloi) submerged (A) and after
21 days of terrestrial conditions (skin remained moist), (B) 0–3 h and (C) 12–13 h. Nitrogen
6.   OSMOREGULATION                                                                           309

may be one instigator of estivation in P. dolloi. Multiple factors are probably
involved and identification of changes in hormone‐signaling pathways would
be valuable information toward a more thorough understanding of the
control of estivation.
    Bowfin is tolerant of air‐exposure and there were suggestions in the
literature that Amia estivates. McKenzie and Randall (1990) attempted to
induce estivation in A. calva in the laboratory by gradually air‐exposing fish
over a 10‐day period, or elevating external water ammonia or decreasing
water oxygen levels. None of these treatments induced estivation, and
A. calva died following 3–5 days in air.


    In the area of ionic, osmotic, and nitrogenous waste regulation in
primitive fishes, there are as many gaps in our knowledge as there is detailed
information. For example, considerable research has focused on gill mor-
phology, cell type, and ultrastructure of the Agnathans, but limited
work has been directed toward the functional role of gill subtypes and the
expression of ion transporters. Likewise, an explosion of papers on lungfish
nitrogen excretion has unveiled fascinating responses to environmental per-
turbations but much less is known of ionic and osmotic regulation. As out-
lined in specific sections above, molecular approaches may provide the
missing links in a number of cases (e.g., the urea cycle enzyme CPS in
lungfish, expression of UTs and Rh‐factor ammonia transporters in gills).
Finally, we have negligible data on any aspect of osmoregulation and nitro-
gen excretion in birchir, gar, bowfin, and paddlefish (not to mention coela-
canth!). Studies of extant primitive fishes may provide more than just data on
another fish species, but lead to a broader understanding of how early
vertebrates evolved under changing external conditions (e.g., ions, salinity,
water availability).

excretion was partitioned between the anterior (i.e., head) and posterior (i.e., body) compart-
ments by placing lungfish in divided chambers under (A) control aquatic conditions (N ¼ 10),
(B) after return (0–3 h) to aquatic conditions following 21 days of terrestrial conditions (N ¼ 5),
and (C) 12–13 h after return to aquatic conditions following 21 days of terrestrial conditions
(N ¼ 9). Asterisks indicate significant diVerence ( p       0.05) from the aquatic rates in panel
A, whereas crosses indicate significant diVerence ( p 0.05) from the corresponding rates in the
anterior compartment. [After Wood et al. (2005b) with permission from University of Chicago
310                                                                       PATRICIA A. WRIGHT


    The author wishes to thank Drs. Wilkie, Wood and Terjesen for access to unpublished
material, Tom Binder for helpful discussions on lamprey development, Ian Smith for graphics,
Kim Ong for digging up references and Lori Ferguson for skilled clerical help. The comments of
two anonymous reviewers are very much appreciated. Financial support was provided by the
Natural Sciences and Engineering Research Council.


Alt, J. M., Stolte, H., Eisenbach, G. M., and Walvig, F. (1981). Renal electrolyte and fluid
     excretion in the Atlantic hagfish Myxine glutinosa. J. Exp. Biol. 91, 323–330.
Altinok, I., and Grizzle, J. M. (2004). Excretion of ammonia and urea by phylogenetically
     diverse fish species in low salinities. Aquaculture 238, 499–507.
Altinok, I., Galli, S. M., and Chapman, F. A. (1998). Ionic and osmotic regulation capabilities of
     juvenile Gulf of Mexico sturgeon, Acipenser oxyrinchus de sotoi. Comp. Biochem. Physiol.
     120, 609–616.
Andersen, Ø., Aas, T. S., Skugor, S., Takle, H., van Nes, S., Grisdale‐Helland, B., Helland, S. J.,
     and Terjesen, B. F. (2006). Purine‐induced expression of urate oxidase and enzyme activity
     in Atlantic salmon (Salmo salar). Cloning of urate oxidase liver cDNA from three teleost
     species and the African lungfish Protopterus annectens. FEBS J. 273, 2839–2850.
Anderson, P. M. (2001). Urea and glutamine synthesis: Environmental influences on nitrogen
     excretion. In ‘‘Fish Physiology Vol. 20, Nitrogen Excretion’’ (Wright, P., and Anderson, P.,
     Eds.), pp. 239–278. Academic Press, San Diego.
Applegate, V. C. (1950). Natural history of the sea lamprey Petromyzon marinus, in Michigan.
     US Department of Interior, Fish and Wildlife Service, Special Scientific Report: Fisheries.
     no. 55.
Barimo, J. F., Steele, S. L., Wright, P. A., and Walsh, P. J. (2004). Ureotely and ammonia
     tolerance in early life stages of the gulf toadfish. Opsanus beta. J. Exp. Biol. 207, 2011–2020.
Bartels, H. (1998). The gills of hagfishes. In ‘‘The Biology of Hagfishes’’ (Jørgensen, J. M.,
     Lomholt, J. P., Weber, R. E., and Malte, H., Eds.), pp. 205–222. Chapman and Hall,
Bartels, H., and Potter, I. C. (2004). Cellular composition and ultrastructure of the gill epitheli-
     um of larval adult lampreys. J. Exp. Biol. 207, 3447–3462.
Beamish, F. W. H. (1980a). Osmoregulation in juvenile and adult lampreys. Can. J. Fish. Aquat.
     Sci. 37, 1739–1750.
Beamish, F. W. H. (1980b). Biology of the North American anadromous sea lamprey, Petromy-
     zon marinus. Can. J. Fish. Aquat. Sci. 37, 1924–1943.
Beamish, F. W. H., Strachan, P. D., and Thomas, E. (1978). Osmotic and ionic performance of
     the anadromous sea lamprey, Petromyzon marinus. Comp. Biochem. Physiol. 60A, 435–443.
Bellamy, D., and Jones, I. C. (1961). Studies on Myxine glutinosa. I. The chemical composition of
     the tissues. Comp. Biochem. Physiol. 3, 175–183.
Brown, G. W., Jr., and Brown, S. G. (1967). Urea and its formation in coelacanth liver. Science
     155, 570–573.
Brown, G. W., Jr., James, J., Henderson, R. J., Thomas, W. N., Robinson, R. O.,
     Thompson, A. L., Brown, E., and Brown, S. G. (1966). Uricolytic enzymes in liver of the
     Dipnoan Propterus aethiopicus. Science 153, 1653–1654.
Brown, J. A., Cobb, C. S., Frankling, S. C., and Rankin, J. C. (2005). Activation of the newly
     discovered cyclostome rennin‐angiotensis system in the river lamprey Lamprey fluviatilis.
     J. Exp. Biol. 208, 223–232.
6.   OSMOREGULATION                                                                         311

Boylan, J. W. (1972). A model for passive urea reabsorption in the elasmobranch kidney. Comp.
    Biochem. Physiol. 42A, 27–30.
Bull, J. M., and Morris, R. (1967). Studies on freshwater osmoregulation in the ammocoete larva
    of Lampetra planeri (Bloch). J. Exp. Biol. 47, 485–494.
Butler, D. G., and Youson, J. H. (1988). Kidney function in the bowfin (Amia calva L.). Comp.
    Biochem. Physiol. 89A, 343–345.
Carlisky, N. J., and Barrio, A. (1972). Nitrogen metabolism of the South American Lungfish
    Lepidosiren paradoxa. Comp. Biochem. Physiol. 41B, 857–873.
Casey, C. A., and Anderson, P. M. (1983). Glutamine‐ and N‐acetyl‐glutamate‐dependent
    carbamoyl phosphate synthetase from Micropterus salmoides. Purification, properties and
    inhibition by glutamine analogs. J. Biol. Chem. 258, 8723–8732.
Cataldi, E., Ciccotti, E., di Marco, P., di Santo, O., Bronzi, P., and Cataudella, S. (1995).
    Acclimation trials of juvenile Italian sturgeon to diVerent salinities: Morpho‐physiological
    descriptors. J. Fish Biol. 47, 609–618.
Chadwick, T. D., and Wright, P. A. (1999). Nitrogen excretion and expression of urea cycle
    enzymes in the Atlantic cod (Gadus morhua L.): A comparison of early life stages with adults.
    J. Exp. Biol. 202, 2653–2662.
Chamberlin, M. E., Glemet, H. C., and Ballantyne, J. S. (1991). Glutamine metabolism in a
    holostean (Amia calva) and teleost fish (Salvelinus namaycush). Am. J. Physiol. 260,
Chang, M., Zhang, J., and Desui, M. (2006). A lamprey from the Cretaceous Jehol biota of
    China. Nature 44, 972–974.
Chew, S. F., Ong, T. F., Lo, L., Tam, W. L., Loong, A. M., Hiong, K. C., Wong, W. P., and
    Ip, Y. K. (2003). Urea synthesis in the African lungfish Propterus dolloi—hepatic carbamoyl
    phosphate synthetase III and glutamine synthetase are upregulated by 6 days of aerial
    exposure. J. Exp. Biol. 206, 3615–3624.
Chew, S. F., Chan, N. K. Y., Loong, A. M., Hiong, K. C., Tam, W. L., and Ip, Y. K. (2004).
    Nitrogen metabolism in the African lungfish (Protopterus dolloi) aestivating in a mucus
    cocoon on land. J. Exp. Biol. 207, 777–786.
Chew, S. F., Ho, L., Ong, T. F., Wong, W. P., and Ip, Y. K. (2005). The African lungfish,
    Protopterus dolloi, detoxifies ammonia to urea during environmental ammonia exposure.
    Physiol. Biochem. Zool. 78, 31–39.
Choe, K. P., Edwards, S., Morrison‐Shetlar, A. I., Toop, Tes, and Claiborne, J. B. (1999).
    Immunolocalization of Naþ/Kþ‐ATPase in mitochondrion‐rich cells of the atlantic hagfish
    (Myxine glutinosa) gill. Comp. Biochem. Physiol. 124, 161–168.
Choe, K. P., Morrison‐Shetlar, A. I., Wall, B. P., and Claiborne, J. B. (2002). Immunologi-
    cal detection of Naþ/Hþ exchangers in the gills of a hagfish, Myxine glutinosa, an elasmo-
    branch, Raja erinacea, and a teleost, Fundulus heteroclitus. Comp. Biochem. Physiol. 131A,
Choe, K. P., O’Brien, S., Evans, D. H., Toop, T., and Edwards, S. L. (2004). Immunolocalization
    of Naþ/Kþ‐ATPase, carbonic anhydrase II, and vacuolar Hþ‐ATPase in the gills of fresh-
    water adult lampreys,. Geotria australis. J. Exp. Zool. 301A, 654–665.
Cobb, C. S., Frankling, S. C., Rankin, J. C., and Brown, J. A. (2002). Angiotensin converting
    enzyme‐like activity in tissues from the river lamprey or lampern, Lampetra fluviatilis,
    acclimated to freshwater and seawater. Gen. Comp. Endocrinol. 127, 8–15.
Cooper, A. J. L., and Plum, F. (1987). Biochemistry and physiology of brain ammonia. Physiol.
    Rev. 67, 440–519.
Cvancara, V. A. (1969). Studies on tissue arginase and ureogenesis in freshwater teleosts. Comp.
    Biochem. Physiol. 30, 489–496.
Cvancara, V. A. (1974). Liver carbamyl phosphate synthetase in the primitive freshwater bony
    fishes (Chondrostei, Holostei). Comp. Biochem. Physiol. 49B, 785–787.
312                                                                       PATRICIA A. WRIGHT

Dabrowski, K., Kaushik, S. J., and Fauconneau, B. (1987). Rearing of sturgeon (Acipenser baeri
     Brandt) larvae. III. Nitrogen and energy metabolism and amino acid absorption. Aquacul-
     ture 65, 31–41.
Edwards, S. L., Claiborne, J. B., Morrison‐Sheltar, A. I., and Toop, T. (2001). Expression of
     Naþ/Kþ exchanger mRNA in the gills of the Atlantic hagfish (Myxine glutinosa) in response
     to metabolic acidosis. Comp. Biochem. Physiol. 130A, 81–91.
Essex‐Fraser, P. A., Steele, S. L., Bernier, N. J., Murray, B. W., Stevens, E. D., and Wright, P. A.
     (2005). Expression of four glutamine synthetase genes in the early stages of development of
     rainbow trout (Oncorhynchus mykiss) in relationship to nitrogen excretion. J. Biol. Chem.
     280, 20268–20273.
Evans, D. H. (1984). Gill Naþ/Kþ and ClÀ/HCOÀ exchange systems evolved before the verte-
     brates entered fresh water. J. Exp. Biol. 113, 465–469.
Evans, D. H. (1993). Osmotic and ionic regulation. In ‘‘The Physiology of Fishes’’ (Evans, D. H.,
     Ed.), pp. 315–341. CRC Press, Boca Raton.
Evans, D. H., Piermarini, P. M., and Choe, K. P. (2005). The multifunctional fish gill: Dominant
     site of gas exchange, osmoregulation, acid‐base regulation, and excretion of nitrogenous
     waste. Physiol. Rev. 85, 97–177.
Fange, R. (1998). Introduction: Early hagfish research. In ‘‘The Biology of Hagfishes’’
     (Jørgensen, J. M., Lomholt, J. P., Weber, R. E., and Malte, H., Eds.), pp. 13–19. Chapman
     and Hall, London.
Farmer, G. J., Beamish, F. W. H., and Robinson, G. A. (1975). Food consumption of the adult
     landlocked sea lamprey, Petromyzon marinus L. Comp. Biochem. Physiol. 50A, 753–757.
Felipo, V., and Butterworth, R. F. (2002). Neurobiology of ammonia. Prog. Neurobiol. 67, 259–279.
Fels, L. M., Kastner, S., and Stolte, H. (1998). The hagfish kidney as a model to study renal
     physiology and toxicology. In ‘‘The Biology of Hagfishes’’ (Jørgensen, J. M., Lomholt, J. P.,
     Weber, R. E., and Malte, H., Eds.), pp. 347–363. Chapman and Hall, London.
Felskie, A. K., Anderson, P. M., and Wright, P. A. (1998). Expression and activity of carbamoyl
     phosphate synthetase III and ornithine urea cycle enzymes in various tissues of four fish
     species. Comp. Biochem. Physiol. 119B, 355–364.
Fines, G. A., Ballantyne, J. S., and Wright, P. A. (2001). Active urea transport and an unusual
     basolateral membrane composition in the gills of a marine elasmobranch. Am. J. Physiol.
     280, R16–R24.
Fontenot, Q. C., Isely, J. J., and Tomasso, J. R. (1998). Acute toxicity of ammonia and nitrite to
     shortnose sturgeon fingerlings. Prog. Fish Cult. 60, 315–318.
Forster, R. P., and Goldstein, L. (1966). Urea synthesis in the lungfish: Relative importance of
     purine and ornithine cycle pathways. Science 153, 1650–1652.
Gershanovich, A. D., and Pototskij, I. V. (1995). The peculiarities of non‐faecal nitrogen
     excretion in sturgeons (Pisces: Acipenseridae). 2. EVects of water temperature, salinity and
     pH. Comp. Biochem. Physiol. 111A, 313–317.
Goldstein, L., Janssens, P. A., and Forster, R. P. (1967). Lungfish Neoceratodus forsteri:
     Activities of ornithine–urea cycle and enzymes. Science 157, 316–317.
Graham, J. B. (1997). ‘‘Air‐Breathing Fishes.’’ Academic Press, San Diego.
GriYth, R. W. (1991). Guppies, toadfish, lungfish, coelacanths and frogs: A scenario for the
     evolution of urea retention in fishes. Environ. Biol. Fish. 32, 199–218.
GriYth, R. W. (1994). The life of the first vertebrates. Bioscience 44, 408–418.
GriYth, R. W., Umminger, B. L., Grant, B. F., Pang, P. K. T., and Pickford, G. E. (1974). Serum
     composition of the coelacanth, Latimeria chalumnae Smith. J. Exp. Biol. 187, 87–102.
GriYth, R. W., Umminger, B. L., Grant, B. F., Pang, P. K. T., Goldstein, L., and Pickford, G.
     (1976). Composition of bladder urine of the coelacanth,. Latimeria chalumnae. J. Exp. Zool.
     196, 371–380.
6.   OSMOREGULATION                                                                            313

Handlogten, M. E., Hong, S.‐P., WesthoV, C. M., and Weiner, I. D. (2004). Basolateral ammonium
     transport by the mouse inner medullary collecting duct cell (mIMCD‐3). Am. J. Physiol. 287,
Helfman, G. S., Collette, B. B., and Facey, D. E. (1997). ‘‘The Diversity of Fishes.’’ Blackwell
     Science, Malden, Massachusetts.
Hill, W. G., Mathai, J. C., Gensure, R. H., Zeidel, J. D., Apodaca, G., Saenz, J. P., Kinne‐
     SaVran, E., Kinne, R., and Zeidel, M. L. (2004). Permeabilities of teleost and elasmobranch
     gill apical membranes: Evidence that lipid bilayers alone do not account for barrier function.
     Am. J. Physiol. 287, C235–C242.
Holder, M. T., Erdmann, M. V., Wilcox, T. P., Caldwell, R. L., and Hillis, D. M. (1999). Two living
     species of coelacanths? Proc. Natl. Acad. Sci. USA 96, 12616–12620.
Holland, N. D., and Chen, J. (2001). Origin and early evolution of the vertebrates: New insights
     from advances in molecular biology, anatomy, and palaeontology. Bioessays 23, 142–151.
Huang, C.‐H., and Peng, J. (2005). Evolutionary conservation and diversification of Rh
     family genes and proteins. Proc. Natl. Acad. Sci. USA 102, 15512–15517.
Hung, C. Y. C., Tsui, K. N. T., Wilson, J. M., Nawata, C. M., Wood, C. P., and Wright, P. A.
     (2007). Rhesus glycoprotein gene expression in the mangrove killifish kryptolebias marmor-
     atus exposed to elevated environmental ammonia levels and air. J.Exp.Biol. (in press).
Hyodo, S., Katoh, F., Kaneko, T., and Takei, Y. (2004). A facilitative urea transporter is localized
     in the renal collecting tubule of the dogfish Triakis scyllia. J. Exp. Biol. 207, 347–356.
Ikemoto, M., Tabata, M., Miyake, T., Kono, T., Mori, M., Totani, M., and Murachi, T. (1990).
     Expression of human liver arginase in Escherichia coli. Purification and properties of the
     product. Biochem. J. 270, 697–703.
Ip, Y. K., Chew, S. F., and Randall, D. J. (2001). Ammonia toxicity, tolerance, and excretion.
     In ‘‘Fish Physiology Vol. 20, Nitrogen Excretion’’ (Wright, P., and Anderson, P., Eds.),
     pp. 109–148. Academic Press, San Diego.
Ip, Y. K., Chew, S. F., and Randall, D. J. (2004). Five tropical air‐breathing fishes, six
     diVerent strategies to defend against ammonia toxicity on land. Physiol. Biochem. Zool.
     77, 768–782.
Ip, Y. K., Peh, B. K., Tam, W. L., Lee, S. L. M., and Chew, S. F. (2005). Changes in salinity
     and ionic compositions can act as environmental signals to induce a reduction in
     ammonia production in the African lungfish Protopterus dolloi. J. Exp. Zool. 303A, 456–463.
Janech, M. G., Fitzgibbon, W. R., Chen, R., Nowak, M. W., Miller, D. H., Paul, R. V., and
     Ploth, D. W. (2003). Molecular and functional characterization of a urea transporter
     from the kidney of the Atlantic stingray. Am. J. Physiol. 284, F996–F1005.
Janech, M. G., Fitzgibbon, W. R., Nowak, M. W., Miller, D. H., Paul, R. V., and Ploth, D. W.
     (2006). Cloning and functional characterization of a second urea transporter (strUT‐2)
     from the kidney of the Atlantic stingray, Dasyatis sabina. Am. J. Physiol. Regul. Integr.
     Comp. Physiol. 291, R844–R853.
Janssens, P. A. (1964). The metabolism of the aestivating African lungfish. Comp. Biochem. Physiol.
     11, 105–117.
Janssens, P. A., and Cohen, P. P. (1966). Ornithine‐urea cycle enzymes in the African lungfish,
     Protopterus aethiopicus. Science 152, 358–359.
Janssens, P. A., and Cohen, P. P. (1968). Nitrogen metabolism in the African lungfish. Comp.
     Biochem. Physiol. 24, 879–886.
Jarvis, P. L., and Ballantyne, J. S. (2003). Metabolic responses to salinity acclimation in
     juvenile shortnose sturgeon Acipenser brevirostrum. Aquaculture 219, 891–909.
Kajimura, M., Croke, S. J., Glover, C. N., and Wood, C. M. (2004). The eVect of feeding and
     fasting on the excretion of ammonia, urea and other nitrogenous waste products in rainbow
     trout. J. Exp. Biol. 207, 1993–2002.
314                                                                       PATRICIA A. WRIGHT

Karnaky, K. J., Jr. (1986). Structure and function of the chloride cell of Fundulus heteroclitus and
    other teleosts. Am. Zool. 26, 209–224.
Karnaky, K. J., Jr. (1998). Osmotic and ionic regulation. In ‘‘The Physiology of Fishes’’
    (Evans, D. H., Ed.), 2nd edn., pp. 157–176. CRC Press, Boca Raton.
Kato, A., and Sands, J. M. (1998). Active sodium-urea counter-transport is inducible in the
    basolateral membrance of rat renal initial inner medullary collecting ducts. J. Clin. Invest.
    102, 1008–1015.
KieVer, J. D., Wakefield, A. M., and Litvak, M. K. (2001). Juvenile sturgeon exhibit reduced
    physiological responses to exercise. J. Exp. Biol. 204, 4281–4289.
Kirschner, L. B. (1993). The energetics of osmotic regulation in ureotelic and hypoosmotic fishes.
    J. Exp. Zool. 267, 19–26.
Korsgaard, B., Mommsen, T. P., and Wright, P. A. (1995). Adaptive relationships to environ-
    ment ontogenesis and viviparity. In ‘‘Nitrogen Metabolism and Excretion’’ (Walsh, P. J.,
    and Wright, P., Eds.), pp. 259–287. CRC Press, Boca Raton.
Krayushkina, L. S., Panov, A. A., Gerasimov, A. A., and Potts, W. T. W. (1996). Changes in
    sodium, calcium and magnesium ion concentrations in sturgeon (Huso huso) urine and in
    kidney morphology. J. Comp. Physiol. 165B, 527–533.
Lacy, E. R., Reale, E., Schusselburg, D. S., Smith, W. K., and Woodward, D. J. (1985). A renal
    counter‐current system in marine elasmobranch fish: A computer aided reconstruction.
    Science 227, 1351–1354.
Laurent, P. (1984). Gill internal morphology. In ‘‘Fish Physiology’’ (Hoar, W. S., and
    Randall, D. J., Eds.), Vol. 10A, pp. 73–183. Academic Press, Orlando.
Laurent, P., Delaney, R. G., and Fishman, A. P. (1978). The vasculature of the gills in the
    aquatic and aestivating lungfish (Protopterus aethiopicus). J. Morphol. 56, 173–208.
Lim, C. K., Wong, W. P., Lee, S. M. L., Chew, S. F., and Ip, Y. K. (2004). The ammonotelic
    African lungfish, Protopterus dolloi, increases the rate of urea synthesis and becomes
    ureotelic after feeding. J. Comp. Physiol. B 174, 555–564.
Locket, N. A. (1980). Some advances in coelacanth biology. Proc. R. Soc. Lond. B 208, 265–307.
Logan, A. G., Morris, R., and Rankin, J. C. (1980). A micro‐puncture study of kidney function
    in the river lamprey Lampetra fluviatilis adapted to sea water. J. Exp. Biol. 88, 239–247.
Loong, A. M., Hiong, K. C., Lee, S. M. L., Wong, W. P., Chew, S. F., and Ip, Y. K. (2005).
    Ornithine‐urea cycle and urea synthesis in African lungfishes, Protopterus aethiopicus and
    Protopterus annectens, exposed to terrestrial conditions for six days. J. Exp. Zool. 303A,
Lutz, P. L., and Robertson, J. D. (1971). Osmotic constituents of the coelacanth Latimeria
    chalumnae Smith. Biol. Bull. 141, 553–560.
Mangum, C. P. (1991). Urea and chloride sensitivities of coelacanth hemoglobin. Environ. Biol.
    Fishes 32, 219–222.
Mallat, J., and Paulsen, C. (1986). Gill ultrastructure of the Pacific hagfish Eptatretus stouti.
    Am. J. Anat. 177, 243–269.
Mallat, J., Bailey, J. F., Lampa, S. J., Evans, M. A., and Tate, W. (1995). Quantitive ultrastruc-
    ture of gill epithelial cells in the larval lamprey Petromyzon marinus. Can. J. Fish Aquat. Sci.
    52, 1150–1164.
Mallat, J., Conley, D. M., and Ridgway, R. L. (1987). Why do hagfish have gill ‘‘chloride cells’’
    when they need not regulate plasma NaCl concentration? Can. J. Zool. 65, 1956–1965.
Marshall, W. S. (2002). Naþ, ClÀ, Ca2þ and Zn2þ transport by fish gills: Retrospective review
    and prospective synthesis. J. Exp. Zool. 293, 264–283.
Marshall, W. S., and Grosell, M. (2006). Ion transport, osmoregulation, and acid‐base balance.
    In ‘‘The Physiology of Fishes’’ (Evans, D. H., and Claiborne, J. B., Eds.), 3rd edn.,
    pp. 177–230. CRC Press, Boca Raton.
6.   OSMOREGULATION                                                                          315

     ´      ´                                                                      ´
Martınez‐Alvarez, R. M., Hidalgo, M. C., Domezain, H., Morales, A. E., Garcıa‐Gallego, M.,
    and Sanz, A. (2002). Physiological changes in sturgeon Acipenser naccarii caused by increas-
    ing environmental salinity. J. Exp. Biol. 205, 3699–3706.
Martini, F. H. (1998). Secrets of the slime hag. Sci. Am. 279, 70–75.
McDonald, D. G., and Milligan, C. L. (1992). Chemical properties of the blood. In ‘‘Fish
    Physiology’’ (Hoar, W. S., Randall, D. J., and Farrell, A. P., Eds.), Vol. XIIB,
    pp. 55–133. Academic Press, San Diego.
McDonald, D. G., Cavdek, V., Calvert, L., and Milligan, C. L. (1991). Acid‐base regulation in
    the Atlantic hagfish. Myxine glutinosa. J. Exp. Biol. 161, 201–215.
McEnroe, M., and Cech, J. J., Jr. (1985). Osmoregulation in juvenile and adult white sturgeon,
    Acipenser transmontanus. Environ. Biol. Fishes 14, 23–30.
McFarland, W. N., and Munz, F. W. (1965). Regulation of body weight and serum composition
    by hagfish in various media. Comp. Biochem. Physiol. 14, 383–398.
McInerney, J. E. (1974). Renal sodium reabsorption in the hagfish, Eptatretus stouti. Comp.
    Biochem. Physiol. 49A, 273–280.
McKenzie, D. J., and Randall, D. J. (1990). Does Amia calva aestivate? Fish Physiol. Biochem. 8,
McKenzie, D. J., Cataldi, E., Di Marco, P., Mandich, A., Romano, P., Ansferri, S., Bronzi, P.,
    and Cataudella, S. (1999). Some aspects of osmotic and ionic regulation in Adriatic sturgeon
    (Acipenser naccarii). II. Morpho‐physiological adjustments to hyperosmotic environments.
    J. Appl. Ichthyol. 15, 61–66.
McKenzie, D. J., Cataldi, E., Romano, P., Owen, S. F., Taylor, E. W., and Bronzi, P. (2001).
    EVects of acclimation to brackish water on the growth, respiratory metabolism, and swim-
    ming performance of young‐of‐the‐year Adriatic sturgeon (Acipenser naccarii). Can. J. Fish
    Aquat. Sci. 58, 1004–1112.
Mommsen, T. P., and Hochachka, P. W. (1988). The purine nucleotide cycle as two temporally
    separated metabolic units—a study on trout muscle. Metabolism 36, 552–556.
Mommsen, T. P., and Walsh, P. J. (1989). Evolution of urea synthesis in vertebrates: The piscine
    connection. Science 243, 72–74.
Moore, J. W., and Mallatt, J. M. (1980). Feeding of larval lamprey. Can. J. Fish Aquat. Sci. 37,
Morgan, R. L., Ballantyne, J. S., and Wright, P. A. (2003). Regulation of a renal urea transporter
    with reduced salinity in a marine elasmobranch, Raja erinacea. J. Exp. Biol. 206, 3285–3292.
Morris, R. (1972). Osmoregulation. In ‘‘The Biology of Lampreys ’’ (Hardisty, M. W., and
    Potter, I. C., Eds.), Vol. 2, pp. 193–239. Academic Press, London.
Morris, R., and Bull, J. M. (1968). Studies on freshwater osmoregulation in the ammocoete larva
    of Lampetra planeri (Bloch). II. The eVect of de‐ionized water and temperature on sodium
    balance. J. Exp. Biol. 48, 597–609.
Morris, R., and Bull, J. M. (1970). Studies on freshwater osmoregulation in the ammocoete larva
    of Lampetra planeri (Bloch). III. The eVect of external and internal sodium concentration on
    sodium transport. J. Exp. Biol. 52, 275–290.
Munz, F. W., and McFarland, W. N. (1964). Regulatory function of a primitive vertebrate
    kidney. Comp. Biochem. Physiol. 13, 381–400.
Nakhoul, N. L., DeJong, H., Abdulnour‐Nakhoul, S. M., Boulpaep, E. L., Hering‐Smith, K.,
    and Hamm, L. L. (2005). Characteristics of renal Rhbg as an NHþ transporter.    4
    Am. J. Physiol. 288, F170–F181.
Noguchi, T., Takada, Y., and Fujiwara, S. (1979). Degradation of uric acid to urea and
    glyoxylate in peroxisomes. J. Biol. Chem. 254, 5272–5275.
Peek, W. D., and Youson, J. H. (1979). Ultrastructure of chloride cells in young adults of the
    anadromous sea lamprey, Petromyzon marinus L., in freshwater and during adaptation to
    sea water. J. Morphol. 160, 143–164.
316                                                                     PATRICIA A. WRIGHT

Perry, S. F., Gilmour, K. M., Swenson, E. R., Vulesevic, B., Chew, S. F., and Ip, Y. K. (2005a).
    An investigation of the role of carbonic anhydrase in aquatic and aerial gas transfer in the
    African lungfish Protopterus dolloi. J. Exp. Biol. 208, 3805–3815.
Perry, S. F., Gilmour, K. M., Valesevic, B., McNeill, B., Chew, S. F., and Ip, Y. K. (2005b).
    Circulating catecholamines and cardiorespiratory responses in hypoxic lungfish (Proto-
    pterus dolloi): A comparison of aquatic and aerial hypoxia. Physiol. Biochem. Zool. 78,
Pickering, A. D., and Morris, R. (1970). Osmoregulation of Lampetra fluviatilis L. and Petro-
    myzon marinue (Cyclostomata) in hyperosmotic solutions. J. Exp. Biol. 53, 231–243.
Pickford, G. E., and Grant, F. B. (1967). Serum osmolality in the coelacanth, Latimeria
    chalumnae: Urea retention and ion regulation. Science 155, 568–570.
Potts, W. T. W., and Rudy, P. P. (1972). Aspects of osmotic and ionic regulation in the sturgeon.
    J. Exp. Biol. 56, 703–715.
Pugh, E. L., Bittman, R., Fugler, L., and Kates, M. (1989). Comparison of steady‐state fluores-
    cence polarization and urea permeability of phosphatidylcholine and phosphatidylsulfocho-
    line liposomes as a function of sterol structure. Chem. Phys. Lipids 50, 43–50.
Randall, D. J., and Tsui, T. K. N. (2002). Ammonia toxicity in fish. Mar. Pollut. Bull. 45, 17–23.
Rankin, J. C., Cobb, C. S., Frankling, S. C., and Brown, J. A. (2001). Circulating angiotensins in
    the river lamprey, Lampetra fluviatilis, acclimated to freshwater and seawater: Possible
    involvement in the regulation of drinking. Comp. Biochem. Physiol. 129B, 311–318.
Read, L. J. (1968). A study of ammonia and urea production and excretion in the fresh‐water‐
    adapted form of the Pacific lamprey, Entosphenus tridentatus. Comp. Biochem. Physiol. 26,
Read, L. J. (1975). Absence of ureogenic pathways in liver of the hagfish Bdellostoma cirrhatum.
    Comp. Biochem. Physiol. 51B, 139–141.
Richardson, M. K., and Wright, G. M. (2003). Developmental transformations in a normal
    series of embryos of the sea lamprey Petromyzon marinus (Linnaeus). J. Morphol. 257,
Riegel, J. A. (1998). An analysis of the function of the glomeruli of the hagfish mesonephric
    kidney. In ‘‘The Biology of Hagfishes’’ (Jørgensen, J. M., Lomholt, J. P., Weber, R. E., and
    Malte, H., Eds.), pp. 364–377. Chapman and Hall, London.
Ripoche, P., Bertrand, O., Gane, P., Birkenmeier, C., Colin, Y., and Cartron, J.‐P. (2004).
    Human Rhesus‐associated glycoprotein mediates facilitated transport of NH3 into red
    blood cells. Proc. Natl. Acad. Sci. USA 101, 17222–17227.
Robertson, J. D. (1963). Osmoregulation and ionic composition of cells and tissues. In ‘‘The
    Biology of Myxine’’ (Brodal, A., and Fange, R., Eds.), pp. 503–515. Scandinavian Univer-
    sity Books, Oslo.
Robertson, J. D. (1989). Osmotic constituents of the blood plasma and parietal muscle of
    Scyliorhinus canicula (L.). Comp. Biochem. Physiol. 93A, 799–805.
Rodrıguez, A., Gallardo, M. A., Gisbert, E., Santilari, S., Ibarz, A., Sanchez, J., and Castello‐
     ´                                                                   ´                      ´
    Orvay, F. (2002). Osmoregulation in juvenile Siberian sturgeon (Acipenser baerii). Fish
    Physiol. Biochem. 26, 345–354.
Smith, C. P., and Wright, P. A. (1999). Molecular characterization of an elasmobranch urea
    transporter. Am. J. Physiol. 276, R622–R626.
Smith, H. W. (1929). The excretion of ammonia and urea by the gills of the fish. J. Biol. Chem. 81,
Smith, H. W. (1930). Metabolism of the lung‐fish, Protopterus æthiopicus. J. Biol. Chem. 88,
Smith, H. W. (1932). Water regulation and its evolution in the fishes. Q. Rev. Biol. 7, 1–26.
6.   OSMOREGULATION                                                                           317

Smith, H. W. (1961). ‘‘From Fish to Philosopher.’’ American Museum of Natural History,
    New York.
Steele, S. L., Yancey, P. H., and Wright, P. A. (2004). Osmoregulation during early embryonic
    development in the marine little skate Raja erinacea; response to changes in external salinity.
    J. Exp. Biol. 207, 2021–2031.
Steele, S. L., Yancey, P. H., and Wright, P. A. (2005). The little skate Raja erinacea exhibits an
    extrahepatic ornithine urea cycle in the muscle and modulates nitrogen metabolism during
    low‐salinity challenge. Physiol. Biochem. Zool. 78, 216–226.
Terjesen, B. F., Chadwick, T. D., Verreth, J. A. J., Rønnestad, I., and Wright, P. A. (2001).
    Pathways for urea production during early life of an air‐breathing teleost, the African catfish
    Clarias gariepinus Burchell. J. Exp. Biol. 204, 2155–2165.
Thurston, R. V., Russo, R. C., and Vinogradov, G. A. (1981). Ammonia toxicity to fishes.
    EVects of pH on the toxicity of the un‐ionized ammonia species. Environ. Sci. Technol. 15,
Vigetti, D., Binelli, G., Monetti, C., Prati, M., Bernardini, G., and Gornati, R. (2003). Selective
    pressure on the allantoicase gene during vertebrate evolution. J. Mol. Evol. 57, 650–658.
Walsh, P. J. (1997). Evolution and regulation of urea synthesis and ureotely in (Batrachoidid)
    fishes. Annu. Rev. Physiol. 59, 299–323.
Walsh, P. J. (1998). Nitrogen excretion and metabolism. In ‘‘The Physiology of Fishes’’
    (Evans, D. H., Ed.), 2nd edn., pp. 199–214. CRC Press, Boca Raton.
Walsh, P. J., and Mommsen, T. J. (2001). Evolutionary considerations of nitrogen metabolism
    and excretion. In ‘‘Fish Physiology Vol. 20, Nitrogen Excretion’’ (Wright, P., and
    Anderson, P., Eds.), pp. 1–30. Academic Press, San Diego.
Walsh, P. J., and Smith, C. P. (2001). Urea transport. In ‘‘Fish Physiology, Vol. 20, Nitrogen
    Excretion’’ (Wright, P. A., and Anderson, P. M., Eds.), pp. 279–307. Academic Press, San
Walsh, P. J., Wang, Y., Campbell, C. E., DeBoeck, G., and Wood, C. M. (2001). Patterns of
    nitrogenous waste excretion and gill urea transporter mRNA expression in several species of
    marine fish. Mar. Biol. 139, 839–844.
Walsh, P. J., Veauvy, C. M., McDonald, M. D., Pamenter, M. E., Buck, L. T., and Wilkie, M. P.
    (2006). Piscine insights into comparisons of anoxia tolerance, ammonia toxicity, stroke and
    hepatic encephalopathy. Comp. Biochem. Physiol. Part A (in press).
Wang, Y., and Walsh, P. J. (2000). High ammonia tolerance in fishes of the family Batrachoi-
    didae (toadfish and midshipman). Aquat. Toxicol. 50, 205–219.
Weihrauch, D., Morris, S., and Towle, D. W. (2004). Ammonia excretion in aquatic and
    terrestrial crabs. J. Exp. Biol. 207, 4491–4504.
Wicks, B. J., and Randall, D. J. (2002). The eVect of sub‐lethal ammonia exposure on fed and
    unfed rainbow trout: The role of glutamine in regulation of ammonia. Comp. Biochem.
    Physiol. 132A, 275–285.
Wilkie, M. P. (2002). Ammonia excretion and urea handling by fish gills: Present understanding
    and future research challenges. J. Exp. Zool. 293, 284–301.
Wilkie, M. P., and Wood, C. M. (1991). Nitrogenous waste excretion acid‐base regulation, and
    ionoregulation in rainbow trout (Oncorhynchus mykiss) exposed to extremely alkaline water.
    Physiol. Zool. 64, 1069–1086.
Wilkie, M. P., Wang, Y., Walsh, P. J., and Youson, J. H. (1999). Nitrogenous waste excretion by
    the larvae of a phylogenetically ancient vertebrate: The sea lamprey (Petromyzon marinus).
    Can. J. Zool. 77, 707–715.
Wilkie, M. P., Bradshaw, P. G., Joanis, V., Claude, J. F., and Swindell, S. L. (2001). Rapid
    metabolic recovery following vigorous exercise in burrow‐dwelling larval sea lampreys
    (Petromyzon marinus). Physiol. Biochem. Zool. 74, 261–272