-, P. N. SALKELD, B. L. BAYNE, E. GNAIGER, AND D. TANDE, K. S., AND D. SLAGSTAD. 1985. Assimilation
M. LOWE. 1985. Feeding and resource allocation in efficiency in herbivorous aquatic organisms-The po-
the mussel Myths edulis: Evidence for time-averaged tential of the ratio method using 14C and biogenic
optimization. Mar. Ecol. Prog. Ser. 20: 273-287. silica as markers. Limnol. Oceanogr. 30: 1093-1099.
LAIWRY, M. R., R. P. HASSE-I-I-,V. FAGERNESS,J. DOWNS, YENTSCH, C. S., AND D. W. MENZEL. 1963. A method
AND C. J. LORENZJZN. 1984. Effect of food acclima- for the determination of phytoplankton chlorophyll
tion on assimilation efficiency of Calanus pacificus. and pheophytin by fluorescence. Deep-Sea Res. 10:
Limnol. Oceanogr. 29: 36 l-364. 221-231.
PARSONS, T. R., Y. MAITA, AND C. M. LALLI. 1984. A
manual of chemical and biological methods for sea-
water analysis. Pergamon.
SHUMAN, F. R., AND C. J. LORENZEN. 1975. Quantitative
Submitted: 15 October 1992
degradation of chlorophyll by a marine herbivore. Accepted: 25 May 1993
Limnol. Oceanogr. 20: 580-586. Amended: 7 July 1993
Limnol. Oceanogr., 39( 1). 1994, 164-169
0 1994, by the American Society of Limnology and Oceanography, Inc.
Hydrodynamic impediments to settlement of marine propagules, and
Abstract- “Protruding bodies,” such as kelp stems, site selection, passive transport of propagules
seagrasses, filiform algae, artificial reefs, and engi- to the substratum, or both (Butman 1987; Eck-
neered structures, constitute substrata for prolifer-
ation of benthic communities of great ecological and man 1990; Mullineaux and Butman 1990).
economical importance. Unfortunately, very little is Many studies dealing with settlement have
known of hydrodynamic aspects of settlement in such shown the significance of various environ-
habitats. Based on flow-tank experiments and the- mental factors, such as chemical cues, sub-
oretical considerations, we discuss hydrodynamic
interference with settlement of larvae on protruding-
stratum heterogeneity, and flow pattern, on dif-
body habitats. We suggest that larvae may overcome ferential settlement (Jumars and Nowell 1984;
these interferences by producing mucous threads up Chabot and Bourget 1988; Morse 1990). Stud-
to 100 body lengths in size. These adhesive threads ies that deal with flow pattern effects on set-
enable propagules of suspension feeders to settle in tlement consider “planar” substrata, but none
environments of high food-particle flux and low sed-
imentation rate. The results suggest that hydrody- of them approaches the problems of settlement
namic impediments to encounter play a major role on “protruding bodies” (for definitions, see Fig.
in determining the spatial distribution of benthic I). Coral reefs, like other benthic environ-
species on protruding bodies by favoring propagules ments, consist of various protruding-body
of species with such adhesive devices.
habitats dominated by species that are ex-
tremely rare in nearby planar habitats and vice
versa (Ableson and Loya unpubl.).
Differential settlement by propagules of ben-
Protruding-body substrata differ dramati-
thic organisms may play an important role in
cally from planar substrata in their flow re-
determining the survi vorship of settled indi-
gimes. Planar substrata (characterized by high
viduals, adult spatial distribution, and com-
boundary-layer Reynolds numbers), due to the
munity structure (e.g. Gaines and Roughgar-
adverse pressure gradient that is developed on
den 1985; Mullineaux 1988). Differential
their surface, induce fully turbulent, thick
settlement is thought to be a result of active
boundary layers characterized by eddies and
sweeps (Fig. 1, A). Protruding bodies, because
Acknowledgments of the reduction of the cross-sectional area for
We thank P. Jumars, J. Eckman, M. Patterson, M. Ilan, the flow to pass through and the resultant, fa-
and two anonymous reviewers for comments on the manu- vorable pressure gradient, induce accelerating,
script. We are indebted to Mark Patterson who provided laminar, thin boundary-layer flows (Fig. 1,
us the scheme of the flow tank. Special thanks to Y. Be-
nayahu, Y. Shlesinger, D. Veil, and M. Dahan for help in B,C). The hypothesis proposed here is that the
obtaining the coral larvae. The MBL at Eilat provided flow regime characterizing protruding-body
hospitality and lab facilities. substrata exerts hydrodynamic impediments
Fig. 2. Plan view of the nozzle-diffuser flow tank,
showing the two acrylic cones (expansion angle of 11”; 2.5
x 10 x 34 cm high) functioning as a nozzle (n) and a
diffuser (d), the closed flume (f) bounded by acrylic walls
(76.5 x 10.6 x 10.8 cm deep), returning reservoir (r; 20
x 20 cm in diam), and acrylic tubing (1.6-cm i.d. x 220
cm long) connecting the returning reservoir to the nozzle
and diffuser. Water is recirculated by an electric bilge pump
(p; 50 liters min-’ capacity) located in the reservoir, and
Fig. 1. Schematic representation of a planar substra- DC power supply (s) used to set the flow velocity. Arrows
tum (A) and a protruding body (B) and its cross-section indicate flow direction.
(C) and the nearby flow regimes. Planar substrata are de-
fined as hard- or soft-bottom areas of spacious and flat-
tened or depressed surfaces, generally horizontal; protrud-
ing bodies are defined as substrata of high slenderness ratio of the cones. Flow velocities were determined
(i.e. height to width of the body plane normal to the flow) from analysis of the motions of fine Pliolite
extending above the seabed with most of their surfaces VT-particles (Goodyear Co.) recorded on vid-
vertical. Planar substrata induce fully turbulent, thick eotape. Larvae were added to the tank through
boundary layers (6) characterized by eddies and sweeps.
The flow pattern induced by protruding bodies (in high the return reservoir (Fig. 2) and had a choice
Reynolds numbers) is of two zone types: accelerated lam- of various settling substrata. In addition to the
inar, thin boundary-layer flow (L) in upstream turns and tank walls which were available as planar sub-
retarded turbulent flow in downstream wakes (W). U,,,- strata, we placed flattened pebbles and coral
free-stream flow; broken lines-the outer edge of the
boundary layers; broken arrows-streamlines; solid ar-
skeletons in the trough, as well as upright acryl-
rows- flow velocity. ic cylinders and skeletons of coral branches
which served as protruding bodies. During the
experiments, larval behavior was detected by
to settlement of larvae on their surfaces. These direct observations and by videotaping. Each
impediments prohibit settlement of larvae that experiment lasted 3-10 d, at the end of which
do not have a mechanism to overcome them. the tank walls and the various bodies were
Flow-tank experiments were conducted to surveyed by dissecting microscope to count
determine the role of hydrodynamic processes settled larvae.
and larval behavior in the settlement of four The larvae of S. caliendrum and A. dae-
species of coral larvae in different flow regimes. dalea attached and metamorphosed exclusive-
Seriatopora caliendrum and Alveopora dae- ly on substrata characterized by decelerating
dalea inhabit planar substrata, and Litophyton (dilhtser) and nonaccelerating flow (trough walls
arboreum and Dendronephthya hemprichi and separation zones that are sites on the
dominate protruding-body substrata (Abelson trough’s bottom either where the boundary
and Loya unpubl.). The experiments were con- layer separates from the wall or in the down-
ducted in a “nozzle-diffuser” recirculating flow stream wake of bodies; Table 1, Fig. 3A,B).
tank designed to create steady, non-uniform Larvae of L. arboreum and D. hemprichi set-
flows in a wide velocity range (Fig. 2). Two tled preferentially in accelerating flows (nozzle
velocity ranges were established: 0.5-l 2 cm and protruding bodies); no settlement was ob-
s-l (Exp. 1 and 3) and 2-50 cm s-l (Exp. 2) served in decelerating flows (Table 1; Fig.
where 0.5 and 2 cm s-l are the average free- 3C,D). Direct observations of larval behavior
stream velocities in the closed flume and 12 during the experiments clarified the causes for
and 50 cm s-l the velocities in the narrow base the significant differences in settlement pat-
Table 1. Distribution of larval settlement of four coral species among different substratum categories. Upper numbers
of each substratum-category row indicate the observed frequencies of settled larvae; lower numbers indicate the expected
frequencies based on the relative surface area of each category. G critical value at P = 0.00 1 is 18.467. Alveo. -Alveoporu
daedalea; Seria. -Seriatopora caliendrum; Dendro. - Dendronephthya hemprichi; Litoph. - Litophyton arboreum.
Alveo. Seria. Dendro. Litoph.
cw Exp. 1 Exp. 2 Exp. 1 Exp. 2 Exp. 3 Exp. 1 Exp. 2 Exp. 1 Exp. 2
Protruding bodies 12 0 0 0 0 0 51 18 287 42
4.8 12.6 1.9 21 3.4 11.6 2.16 56.76 12.6
Nozzle 21 0 2 0 0 0 23 27 92 39
8.4 22 3.4 36.8 5.9 19.53 12.18 99.33 22
Trough 43 5 20 1 84 10 17 4 0 4
17.2 45.1 6.9 75.2 5.2 40.92 25.52 203.4 45.1
Separation zone 3 34 74 2 82 0 2 9 94 20
1.2 3.2 0.5 5.3 0.8 2.8 1.7 14.2 3.2
Diffuser 21 1 9 13 9 18 0 0 0 0
8.4 22.1 3.4 36.8 5.9 19.5 12.2 99.33 22.1
Total of settled larvae 40 105 16 175 28 93 58 473 105
G-value 211 406 37 611.5 53 131 135 1,271 218
terns. The settlement pattern of L. arboreum the body is reduced; second, after encounter,
and D. hemprichi is a consequence of larval the accelerated flow interferes with larval at-
rejection of decelerating flow zones and their tachment and establishment. In the encounter
high ability to encounter and attach to sub- phase, the quantities of larvae that are trans-
strata in accelerating flows. The larvae of these ported to the protruding body surface are dras-
species were observed to deposit in deceler- tically reduced due to decline in the effective-
ating and nonaccelerating flow zones, as did ness of two important transport mechanisms -
larvae of other species, but most of them, after turbulent transport and gravitational deposi-
brief exploration, actively detached from the tion. The first transport mechanism operates
substratum and swam back into the flow. Like- via eddies or sweeps to enhance particle de-
wise, they were observed being caught by the position (Sumer and Oguz 1978; Abelson et
protruding bodies, at distances from the body al. 199 1) and propagular settlement (Charack-
where larvae of the other two species were un- lis 198 1; Mullineaux and Butman 1990).
able to encounter it. For larvae of A. daedaZea Sweeps characterize planar substrata and do
and S. caliendrum, in contrast, the protruding not exist in the vicinity of protruding bodies.
bodies and nozzle greatly hampered settle- Turbulent eddies, however, which might exist
ment; hence, their settlement pattern is a result in the vicinity of protruding bodies, are less
of their inability to obtain access to these zones. effective as transport mechanisms due to the
Theoretically, larvae gain access to protruding laminar boundary layer of the body which in-
bodies through the wake zone of the body by hibits their penetration to the body’s surface.
active swimming. In our experiments, how- The efficiency of the second mechanism-
ever, no case of such active settlement was gravitational deposition-is dependent on
observed. Our observations have shown that substratum orientation relative to the gravi-
encounter events of S. caliendrum and A. dae- tational vector (Rubenstein and Koehl 1977;
dalea larvae with the protruding bodies are LaBarbera 1984; Shimeta and Jumars 199 1).
extremely infrequent. Moreover, in those cases Thus, in cases of many vertically oriented sur-
where physical contact took place, larvae were faces of the protruding bodies, gravitational
instantaneously swept off the body surface by deposition is negligible. Due to the size and
the currents. density of the larvae, the main mechanism
Obstructions of the flow regime act on set- transporting larvae to protruding bodies is di-
tling larvae during two stages of settlement. rect interception, for which the encounter ef-
First, the probability of larval encounter with ficiencies of larvae are extremely low due to
A Seriatopora caliendrum n rxp.1
the small size of the larvae and the large size
2 la exp.2
of the bodies (Rubenstein and Koehl 1977;
LaBarbera 1984; Shimeta and Jumars 199 1).
The establishment phase, which is crucial
for settlement, requires a finite time to achieve
firm attachment (e.g. Jumars and Nowell 1984;
Denny 1988). Such a residence time is much
harder to attain in the case of protruding bod-
no*z,s protruding bodieo reparation trough walls diffuser
ies, as compared with the seabed, due to a
accelerating nonaccelerating decelerating combination of high shear stresses (stemming
flow flow flow
from the relatively high velocity gradients in-
duced by the protruding bodies) and two forces
0 exp.2 that sweep the larvae away.
The first is drag forces that are much higher
than those on the seabed due to the higher flow
velocities. Protruding bodies experience flow
velocities higher than those of seabed envi-
ronments because of their positions higher in
the benthic boundary layer. Also, flow in the
vicinity of the protruding body may be several
nozzle protruding bodiaa rqmrrtion trough V&II) 1 difturer , times the ambient velocity, depending on the
accelerating nonaccelerating deceleratina
cross-section aspect ratio of the body (i.e. the
ratio between the two diameters of the body’s
cross-section; for circular cylinders which have
a ratio of 1, the velocity achieves about twice
the free-stream velocity).
The second is acceleration reaction forces,
d O” which are unique for accelerating flows and are
B 40 accompanied by drag forces. Protruding bod-
z ies induce non-uniform flow, which under
cn 20 steady flow conditions has an acceleration
nollfs protruding bodie reparation trough wat,s 1 ~,,“,i;U~~e;~,;
accelerating nonaccelerating &A
flow flow flow
‘E D Dendronephthya hemprichi
‘;; 80 n exp.1 u is the velocity and s is distance. The accel-
eration reaction force acting on a propagule in
: 60 such a flow becomes
30 I;, = V(P~ + kapwb $.
v/is the propagule volume, &, is the propagule
“OLZIe protruding bodierll reparation trough waIIs 11 diffuser density, pwthe water density, and k, the added
mass coefficient which is the ratio of the added
Flow regime volume of fluid to the volume of the propagule.
In cases of curvilinear protruding bodies, the
Fig. 3. Larval settlement of four coral species in dif-
ferent flow pattern conditions (n-number of larvae in a flow may accelerate up to several times the
given experiment). A. n, = 16, n, = 175, n, = 28. B. n, free-stream velocity along the sides of the body.
= 40, n, = 105. C. n, = 473, n, = 105. D. n, = 93, n, = The vector direction of this acceleration is the
58. The experiments were conducted in a nozzle-diffuser same as for the drag force. The forces resulting
flow tank in two flow-velocity ranges: slow [Exp. 1 (and
3 in panel A)] and fast (Exp. 2). G-tests for goodness-of-
from acceleration might be significant in cases
fit were run for each of the nine experiments and in all of large propagules settling at high accelera-
cases P KO.001. tion.
and other adhesive devices produced by prop-
agules of many taxa are used for feeding (Emlet
and Strathmann 1985), dispersion (Lane et al.
1985), and for high aggregation efficiency
among bacteria (Cowen 1992). In our study,
however, the common denominator of larvae
that produce threads is their ability to settle in
protruding habitats, resulting in high abun-
dance of their adults in such habitats (Abelson
and Loya unpubl.). A related issue for these
species is their potential food, which consists
of fine particles or organic molecules. Such food
particles are distributed nearly uniformly
throughout the benthic boundary layer. Their
highest fluxes are obtained in high flow-veloc-
ity environments such as those surrounding
We suggest that the mucous threads and oth-
er adhesive devices produced by propagules of
various taxa are used as mechanisms to over-
come hydrodynamic interference exerted by
Fig. 4. Schematic representation of two main catego- protruding-body substrata, enabling the prop-
ries of orientation, trajectory, and subsequent retainment agules to settle on such substrata. We conclude
of “larvae-bearing threads” by protruding bodies. A. Ori- that the described hydrodynamic impedi-
entation of larvae and threads, in which the thread is ex- ments to settlement may play a major role in
tended into the trapping zone of the body (2rJ determined
by larval radius (r,,). B. Orientation of larvae and threads, determining the spatial distribution of benthic
in which the local shear flow of the body rotates the larva species by favoring propagules of species with
and thread so that the thread comes into contact with the such adhesive devices, while prohibiting the
body. U,,, - mainstream flow; L-larva; T-thread; broken settlement of larvae that lack them.
arrows-larval trajectory; solid arrows-flow direction.
Numbers l-3 describe the sequence of arrival, rotation, Avigdor Abelson’
and attachment of the larva.
Department of Zoology
Tel Aviv University
Our observations on larval behavior in the Tel Aviv 69978, Israel
flow tank indicate that the capacity of larvae
to overcome the impedence and settle onto
protruding habitats is directly attributed to Faculty of Aerospace Engineering
mucous threads that are secreted by the aboral Technion- Israel Institute of Technology
part of the larvae. The mucous threads are used Haifa, Israel
to overcome impedance through the two phases
of settlement. In the first phase, the threads Yossi Loya
greatly increase the efficiency of larval en- Department of Zoology
counter with the substratum over that for lar- Tel Aviv University
vae without threads (Fig. 4). In the second
phase, the mucous threads enable an instan-
taneous attachment to the body. The adhesive References
mucous threads have been observed to adhere ABELSON, A., B. S. GALIL, AND Y. LDYA. 199 1. Skeletal
by transient contact in continuous flow and to modifications in stony corals caused by indwelling
crabs: Hydrodynamical advantages for crab feeding.
retain the tethered larvae on the substratum. Symbiosis 10: 233-248.
Hence the larva does not require a finite res-
idence time of locally calm flow conditions to
attach. ’ Present address: Hopkins Marine Station, Stanford
Previous studies have suggested that mucus University, Pacific Grove, California 93950-3094.
BUTMAN, C. A. 1987. Larval settlement of soft-sediment LANE, D. J. W., A. R. BEAUMONT, AND J. R. HUNTER.
invertebrates: The spatial scales of pattern explained 1985. Byssus drifting and the threads of the young
by active habitat selection and the emerging role of post-larval mussel Myths edulis. Mar. Biol. 84: 30 l-
hydrodynamical processes. Oceanogr. Mar. Biol. 308.
Annu. Rev. 25: 113-165. MORSE, D. E. 1990. Recent progress in larval settlement
CHABOT, R., AND E. BOURGET. 1988. Influence of sub- and metamorphosis: Closing the gaps between mo-
stratum heterogeneity and settled barnacle density on lecular biology and ecology. Bull. Mar. Sci. 46: 465-
the settlement of cypris larvae. Mar. Biol. 97: 45-56. 483.
CHARACKLIS, W. G. 198 1. Fouling biofilm development: MULLINEAUX, L. S. 1988. The role of settlement in struc-
A process analysis. Biotechnol. Bioeng. 23: 1923-l 960. turing a hard-substratum community in the deep sea.
COWEN, J. P. 1992. Morphological study of marine bac- J. Exp. Mar. Biol. Ecol. 120: 247-26 1.
terial capsules: Implications for marine aggregates. -, AND C. A. BUTMAN. 1990. Recruitment of en-
Mar. Biol. 114: 85-95. crusting benthic invertebrates in boundary-layer flows:
DENNY, M. W. 1988. Biology and the mechanics of the A deep-water experiment on Cross Seamount. Lim-
wave-swept environment. Princeton. nol. Oceanogr. 35: 409-423.
ECKMAN, J. E. 1990. A model of passive settlement by RUBENSTEIN, D. I., AND M. A. R. KOEHL. 1977. The
planktonic larvae onto bottoms of differing roughness. mechanisms of filter feeding: Some theoretical con-
Limnol. Oceanogr. 35: 887-90 1. siderations. Am. Nat. 111: 981-994.
EMLET, R. B., AND R. R. STRATHMANN. 1985. Gravity, SHIMETA, J., AND P. A. JUMARS. 199 1. Physical mech-
drag, and the feeding currents of small zooplankton. anisms and rates of particle capture by suspension-
Science 228: 1016-1017. feeders. Oceanogr. Mar. Biol. Annu. Rev. 29: 19 l-
GAINES, S., AND J. ROUGHGARDEN. 1985. Larval settle- 257.
ment rate: A leading determinant of structure in an SUMER, B. M., AND B. OGUZ. 1978. Particle motions near
ecological community of the marine intertidal zone. the bottom in turbulent flow in an open channel. J.
Proc. Natl. Acad. Sci. 82: 3707-37 11. Fluid Mech. 86: 109-127.
JUMARS, P. A., AND A. R. M. NOWELL. 1984. Fluid and
sediment dynamic effects on marine benthic com-
munity structure. Am. Zool. 24: 45-55.
LABARBERA, M. 1984. Feeding currents and particle cap-
Submitted: 24 May 1993
ture mechanisms in suspension-feeding animals. Am. Accepted: 31 July 1993
Zool. 24: 71-84. Amended: 1 September 1993
Limnol. Oceanogr., 39(l), 1994, 169-175
0 1994, by the American Society of Limnology and Oceanography, Inc.
Primary production of prochlorophytes, cyanobacteria, and
eucaryotic ultraphytoplankton: Measurements from
flow cytometric sorting
Abstract-A partitioning of ultraphytoplankton for prochlorophytes and 0.2 to 10 fg C celll’ hm ’ for
primary production among prochlorophytes, cyano- cyanobacteria. Results indicated that the dominant
bacteria, and eucaryotic algae was made by ship- primary producer was not necessarily the numerical
board flow cytometric sorting of 14C-labeled cells. dominant nor necessarily the group with the highest
Aggregate primary production was derived from the cell-specific rate of 14Cuptake. Generally, eucaryotic
sum, over all three ultraplankton groups, of the prod- ultraphytoplankton are dominant because of their
uct of cell abundance and cell-specific rate of 14C high cell-specific rate of 14C uptake and in spite of
uptake which ranged from 0.03 to 4 fg C celll’ h-l their relatively low abundance. Less often, it seems,
procaryotic picoplankton may dominate in spite of
their low cell-specific rate of 14C uptake because of
their high abundance.
I am grateful to W. G. Harrison for unpublished data
shown in Fig. 3C. Primary production, meaning the rate at which
This work was supported by the following Canadian carbon is converted from inorganic to organic
government organizations: Department of Fisheries & form by photosynthesis, is often measured by
Oceans, Department of National Defense, Canadian Panel
on Energy R&D (PERD), and the interdepartmental “Green the rate at which phytoplankton become radio-
Plan.” Other support was provided by the Joint Research labeled when supplied with NaH14C0,. Various
Center, Commission of the European Communities. methods exist which allow this production to