ABSTRACTS - The American Ornithologists Union

Document Sample
ABSTRACTS - The American Ornithologists Union Powered By Docstoc
					                                                        One Hundred and Nineteenth Stated Meeting of the
                                                                          American Ornithologists' Union
                                                                                                   in conjunction w ith t he
                                                                                             20 th Annual Meeting of t he
                                                                                      Society of Canadian Ornithologists
                                                                                    Socié des Ornithologistes du Canada
                                                                                                     16 - 18 August 2001
                                                                                Burke Museum, University of Washington
                                                                                                      Seatt le, Washington

version 13 Aug 2001

This Abstract Book includes the following disclaimer:
"The abstracts in this work are not issued for the permanent scientific record."

Plenary I        Ricklefs
The physiology - life history nexus. ROBERT RICKLEFS, Dept. Biol., Univ. Missouri-St. Louis, St. Louis, MO.
          Life-hist ory t heory is based on t he opti mization of c onst rained responses to selection. Whereas most
constraints are seen as resulting from allocation of time and externally acquired resources or from t he
acceptance of risk, many ot her trade-off s influence an individual's lif e-history strategy. A mong these are the
organism's development, struct ure, physiological control mechanisms, immune responses, and parent-off spring
conflict . By adopting a broader concept of the organism, evolutionary ecologists interested in the diversification
of lif e histories may discover new w ays to interpret natural variation and devise new comparative and
experimental approaches to understanding the organism-environment relationship.

Plenary II       Boersma
Seabird conservation: What does the future hold? DEE BOERSMA, Dept . Zo ol. , Univ . W ashingt on, Seattle,
         Like all conservation problems, conservation of seabirds is directly related to how humans use the
environment. Harvest, habitat modification, introduced predators, invasive species, pollution, f ishery
interactions, and climate change are the anthropogenic factors that threaten many of t he worlds seabird
spec ies. These threat s are driv en by 2 f und amental t rend s -- i ncreased hum an po pul ation and i ncreased per
capita consump tion. Even the most isolated and remote places have been stamped by our human f ootp rint . In
the past , harv est f or f ood and f eathers to ok many seabird species to lo w numbers, as evidenced by t he demise
of Great Auks and Shor t-t ailed Albat ros s. Because kil lin g of bir ds is relatively easy to co nt rol , m ost cou nt ries
have stopped large scale harvest through regulation, and social pressure. The current threats to seabirds are
more in dir ect and c omplex . Sp ecies of ten f ace mult ipl e threat s, and t he spatial scale o f int eraction has w iden ed
from poi nt sources of effect (such as t he co lon y) to basin an d even gl obal scal es (in the cas e of glo bal
w arming). For example, many species of penguins are rapidly declining and even with t he best management,
cli mat e chan ge may be too r apid for m any pop ulat ion s t o surv ive. The ch allen ge bef ore u s is how to mit igat e
human impacts in the face of unprecedented increases in consumption and population. Management is one of
the few tools that c an help. Because of the scale and rapidity of effect, it is increasingly important to
concent rate on im prov ing colo ny habit ats f or seabirds as w ell as zoning large areas of th e ocean so as to
reduce conf lict s w ith humans.

Plenary III       Petrie
Current issues in mate choice and sexual selection. MARION PETRIE, Evol.& Behav. Res. Group, Dept.
Psychology, Univ. Newcastle, UK
          The aim of t his talk is to introduce some of the current areas of debate in the field of sexual selection.
In particular, I will concentrate on t w o questions: How is genetic variation maintained in characters undergoing
st ron g di rection al select ion ? and, What is t he rel ation ship bet w een sex ual select ion and speci ation ?

Plenary IV        Bensch
Mol ecular meth ods t o stu dy pop ulatio n divergenc e - the f irst steps of speciation . STA FFAN BENSCH,
Molecular Pop. Biol. Lab., Dept. A nimal Ecol., Lund Univ., Sw eden.
          Analysis of mit ochondrial DNA of ten fails to distinguish betw een populations that otherw ise differ in
morphology or behaviour, traits that presumably have a genetic basis. I w ill exemplify how PCR based methods
to st udy mult ipl e genet ic r egio ns c an be u sed t o id ent if y subt le gen etic d if ferenc es bet w een rec ent ly div erged
pop ulat ion s. Thi s approac h w ill be par ticu larl y usef ul i n 1 ) evolu tion ary st udi es of seco ndar y c ont act zones
bet w een rec ent ly dif ferent iat ed t axa and 2 ) in ecol ogi cal s tudi es at migrat ory st opo ver sit es or in w int er
quarters where unambiguous identifications of indiv iduals to populations are crucial.

Plenary V         Schwabl
Maternal steroid hormones in the egg: functions, mechanisms, and implications. HUBERT SCHWABL, Cent er
of Reproduct. Biol. and Graduate Prog. Neurosci., Washington State Univ., Pullman, WA.
          The aim of t his talk is to introduce the current state of research on maternal steroid hormones in the
avian egg. I w ill focus on 2 questions. What is t he role of w ithin-clutch v ariation in yolk androgens in sibling
rivalry ? Do maternal st eroids prov ide a non-genetic l ink bet w een the mot her and the of fspr ing t hat f unct ions t o
adju st of fspr ing phen ot yp es t o env iro nment al co ndi tion s?

Plenary VI        Wilcove
Putting w ood peck ers in t he ban k – and o ther s trang e, n ew appr oach es t o sav ing North A merica’ s endanger ed
birds. DAVID S. WILCOVE, Ecosyst em Restorat ion Progr am, Envir onm ent al Def ense, Washington, DC.
        The long-term survival of North America’s endangered birds will require the restoration of their
degraded, fragmented habitats and the proper management of those areas follow ing restoration. But if much of
that habitat is on priv ately ow ned land, how can w e elicit t he cooperati on of landow ners? I w ill discu ss the use
of safe harbor agreements, habitat conservation banks, direct payments and other incentives to promote bird
conservation, and I will highlight t he role that scientists can play in making those programs succeed.


S-01 Burger & Gochfeld
Effects of pollutants on seabirds. JOANNA BURGER, Div. Life Sci., Rutgers Univ. , Piscataw ay, NJ , and
MICHA EL GOCHFELD, Environ. & Community M edicine, UMDNH-Robert Wood Johnson Med. School,
Piscataway, NY.
        A range of inorganic and organic chemicals, plastics, and other floatables can result in direct mortality
of seabirds, as w ell as causing sublethal behavioral, reproductive, and physiological effects. It is more difficult
to show how direct mortality and sublethal effects inf luence population dynamics, largely because seabirds
have evolved mechan ism s f or c opi ng w it h nat ural adversities (s torm tides, f ood short ages, chan ges in hab it at)
that cause mortality and sublethal effects, and these same mechanisms confer resiliency to pollution events.
How ever , t here ar e exam ples w here c hemical s hav e caus ed t emporar y popu lat ion decl ines, i nclud ing several
species follow ing the use of DDT (pelicans and cormorants), PCBs and other contaminants (gulls and terns in
the Great Lakes), and oil spills (seabirds in the Gulf of Alaska). These cases illustrate the potential for
w ide-spatial-scale effects on seabirds under several conditions: 1 ) multiple point-source events, 2 ) overlapping
events in the sam e areas, 3) inc reased atmospheric trans por t and subsequen t depo sit ion , an d 4 ) lack of hum an
intervention. There is no cause for complacency, and every reason to continue wit h vigorous research
pro gram s aimed at und erst andi ng t he f ate and ef fect s of chem ical s on seabirds at a loc al, regi onal and g lob al

S-02 Roby, Lyons, Anderson, Antolos, Anderson, Suryan, Wright, Collis & Craig
Seabirds vs. salmon : M anaging Caspian Terns in t he Columbia River. DANIEL D. ROBY* , DONALD E. LYONS,
WRIGHT, USGS-Oregon Coo p. Fish and Wi ldl. Res. Unit , Oreg on St ate Uni v. , Cor valli s, OR; KEN COLLIS, RTR
Consult. , Bend, OR; and DAVID P. CRAIG, Will amett e Univ. , Salem, OR.
          Seabir d/ fisheries con flic ts hav e gener ally inv olv ed eit her ( 1) bycat ch of seabirds , (2) fisheries
ov erex plo it ation of seabird f orag e base, or ( 3) loc alized pred ation pro blem s at maricu lt ure s it es. The Caspian
Tern /Sal monid issue in the Col umbia Riv er est uary is dif ferent , i n t hat most Colu mbia Basin salm oni ds ar e listed
und er t he Endangered Species A ct , b ill ion s of dol lars have recent ly been spent attempting to res tore l ist ed
salmonids, and salmon and steelhead are an important cult ural icon in the Pacific Northw est. Caspian Terns
nesting in the Columbia River estuary have recently been estimated to consume 10 - 15 % of all sea-going
juvenile salmonids from the entire Columbia Basin. This Caspian Tern population nests on artificial dredge spoil
islands, w as first recorded breeding in the estuary in 1984 , and has rapidly become the largest Caspian Tern
colony ever recorded in the w orld. Yet Caspian Terns are a protected species of conservation concern, and
worldwide populations are generally not faring well. The Columbia River estuary population represents ca. 10%
of w orldw ide numbers, 30 % o f Nor th A merican num bers, and 7 5% of Pacif ic Coast nu mbers of thi s species.
Resource managers are att empting to resolve this conflict by restoring Caspian Tern colonies at sites where
salmonids will not comprise the bulk of t he diet, and discouraging nesting at sites w here they will. This
approach has the potential to benefit both salmonids and Caspian Terns as part of a larger effort to restore the
ecosystem of the Columbia River estuary.

S-03 Melvin & Parrish
Resolving seabird bycatch. ED MELVIN* and JULIA K. PARRISH, Washington Sea Grant Prog., School Aquatic
& Fish. Sci., Univ. Washington, Seattle, WA.
        Seabirds and fisheries are inextricably linked by tw o factors: they share fishing grounds, and fishery
w aste is a magnet for foraging seabirds. Given this strong association, seabirds are incidentally killed in a wide
range of f ishing gears throughout t he world' s oceans. Seabird mortality is of ten rare relative to target catch and
the number of seabirds attending fishing operations. However, by virtue of t heir life history, seabirds are
vulnerable to population decline from subtle and chronic mortality. Seabird bycatch as a conservation issue is
characterized in general terms by several themes: population level effects are often unclear; ecological and
emotional concerns are often muddled; and endangered and healthy seabird species frequently overlap and are
both t aken by fishing gear. Solutions to seabird bycatch are hindered by: t he rarity of mort ality events; the
absence and/or inconsistency of conservation law protecting seabirds; institutional inertia/lack of seabird
expertise in fishery agencies, impatience to create bycatch-solving regulations, and - most importantly - t he
absence of trust, resources, and process to find solutions. The Puget Sound drift gillnet fishery for sockeye
salmon and t he Alaska longl ine fish ery f or grou ndf ish are cases that share t hese characterist ics and challenges.
Technological solutions w ere developed for both fisheries using fisher know-how and tested on active fishing
vessels using strict scientific prot ocols. These case studies illustrate that development of eff ective seabird
by cat ch solut ion s req uir es: cross cul tural teamw ork , sensitivit y t o ef fect s of techn olo gic al solut ion s on target
catch , pro fit abilit y and t he bycat ch of non-bird species, and pract ical and mult ifacet ed regulation s. Wh ile these
are su ccessf ul c ase st udi es, they are no t nearl y enoug h. Ult imately, seabir d co nser vat ion in f isheries requ ires
continued improvements to t echnologies and practices, education (of fishers, managers and conservation
organizations), and above all, international action.

S-04 Walker & Boersma
Seabirds and ecotourism: Evaluation, concerns, and a case study from t he breeding Magellanic Penguins of
Patagonia. BRIAN G. WALKER* and P. DEE BOERSMA, Dept. Zool., Univ. Washington, Seattle, WA.
          Ecotourism is one of the fastest grow ing industries in the world, w ith billions of dollars spent annually
by people v isit ing w ildlif e and w ild places. Seabird br eeding colonies are part icularly att racti ve ecot ourism
destinations, as seabirds are typically large and showy , nest in dense aggregations, and breed in beautif ul
coast al set ting s. The i mpact of ecot our ism on seabir ds is a co ncern, and n umerou s researchers have focused
att ention on how colonies react to t ourist s. Certain seabird g roups - e.g ., Pelecanifor mes and Charadriifor mes -
appear more sensitiv e to hum an distu rbance, w hile ot hers - e.g., Procellariifor mes and Spheniscif ormes -
oft en, but not alw ays, have been shown t o habituate to human disturbance. Here we discuss an example of
apparent successful interactions between ecotourists and the breeding Magellanic Penguins of Punta Tombo,
Argentina. Wit h over 50, 000 tourists v isiting the colony annually, Punta Tombo is the most visit ed penguin
colon y in t he West ern Hemis pher e. W e have monit ored mult ipl e levels of resp onse: b ehavior al and hor monal
changes of individuals, as well as measures of chick grow th and reproductive success of groups, and found no
signif ican t dif ferenc es bet w een visited and non -visited bi rds . W hil e cur rent ecot our ism levels appear
sustainable, continued increases in tourist numbers, wit hout concurrent changes in the infrastructure of support
(sanitary facilities, refuse disposal, increased supervision, etc.) may negatively impact the Magellanic penguin
breeding colony.

S-05 Tershy, Croll, Sanchez, Keitt, Donlan, Wood, Hermosillo & How ald
Protecting seabirds by conserving islands: an integrated regional approach. BERNIE TERSHY* , Island Conserv.,
Univ . Calif orn ia, Sant a Cruz, CA; DONA LD CROLL, Island Con serv . an d Ecol . & Evol. Biol. , Univ . Cal if orn ia,
Sant a Cruz, CA; JOSE ANGEL SANCHEZ, Grupo Ecol. & Conserv. de Islas, Mexico; BRAD KEITT, C. JOSH
DONLAN, Island Conserv.; BILL WOOD, MIGUEL ANGEL HERMOSILLO, Grupo Ecol. Conserv.; and GREGG
HOWALD, Island Conservation, Canada.
          Intr oduced mamm als are 1 of t he 2 largest causes of recent seabird exti nct ions. T his is because most
seabirds breed on islands and most island ecosystems are threatened by the introduction of non-native
mammals. Consequently, organizations that int egrate key aspects of island conservation (applied research,
priori ty sett ing, p lanning, public educ ation, fun d raising, non-nat ive mammal remov al, and prot ection against
new int rod uction s) can be p ow erf ul t ool s f or s eabir d co nser vat ion . In Nort hw est Mexico w e develop ed such an
org anizat ion to pr ot ect the reg ion s 230 + islands. These islands have 2 6 speci es of breed ing seabirds and o ver
21 0 species and subsp ecies of endemic v ertebrat es. Non-nat ive mammals have been int roduced t o at least 4 4
islands and are responsible f or t he probable ext inct ion of 21 endemic vert ebrate species and subspecies,
including the w orlds largest storm pet rel Oceanodroma macrodactyla. Int rod uced mammals have also c aused
the extirpation of 1 or more seabird species from at least 8 islands in Northwest M exico. Island Conservation,
the Universidad Nacional Aut onoma de Mexico, Centro de Investigaciones. Biologicas del Noroeste, and the
Mexican National Protected Areas Department collaborated w ith local people and NGOs to remove 1 or more
int rod uced mammals f rom each o f 22 islands and w ill soon co mplet e eradi cat ion on 1 more. Thi s w ork w as
initiated in 1994 and has protected habitat for 22 seabirds and 51 endemic vertebrates for about US$700,000.
Integr ated island con servatio n progr ams in ot her parts o f t he w orld w ould pro tect more seabirds.

S-06 Hass & Parrish
Saving seabirds with soapboxes. TODD HASS and JULIA K. PARRISH, School Aquatic & Fish. Sci., Univ.
Washington, Seattle, WA.
          Scientists are oft en criticized for failing to communicate the relevance of their studies outside the
classroom or even the laboratory. Such exclusivity may be understandable w hen results are highly esoteric or
techn ical, bu t il l-advised w hen the result s could in for m or int egrate crit ical stakehol ders. Wh ile many West
Coast seabirds appear to be in decline, the causes and magnitude of population change remain largely
unknow n. At the same time, recognition of seabird vulnerability to anthropogenic and natural forces is
mou nt ing ( e.g. , Com mon Mu rre su scept ibil it y t o oil spil ls and El Niño condit ions). T o addr ess t his, COASST
(Coastal Observ ation and Seab ird Survey Team) is a W ashingt on-based beach ed bi rd s urv ey prog ram designed
to couple 'hard science' w ith 'public outreach.’ As no public agency has the necessary personnel or money to
conduc t sur veys along W ashingt on' s entire co astline (t oo expansiv e) on a bi-mont hly basis (t oo f requent),
COASST vol unt eers provide an invaluable servi ce. W it h a roster exceedi ng 5 0 p eople and 25 beaches, COASST
has identif ied more than 30 species and 250 carcasses in its first 18 m on of existence. By monitoring the
deposition and persistence of individually identified and marked beachcast carcasses, COASST volunteers
pro vide hi gh q uali ty data on t he st atus and t rend s of coas tal seabird s f or t he pu rpo ses of sci ence, inf orm ed
management and conservation, and proactive citizen involvement and action. M ost COASST volunteers are
beachcombers, drawn from coastal communities and lack field experience with birds – yet they become
eff ectiv e citizen scient ists and l ocal stew ards.

S-07 Naughton
Regional Seabird Conservation Plan. MAURA B. NAUGHTON, U.S. Fish & Wildl. Ser., M igrato ry Birds &
Habit at Programs, Portl and, OR.
         U.S. Fish and Wildlife Service, Region One, is developing a Regional Seabird Conservation Plan. As t he
Federal agency w ith t he primary responsibility for t he management of migratory birds, and managers of a
significant portion of West Coast and Pacific Island seabird nesting habitat, this plan w ill serve to guide and
coordinate Service activities to conserve seabird populations and habitats at t he Regional scale. The plan will
encompass seabirds in Calif ornia, Oregon , Washin gto n, Haw aii and the U. S. Pacific Island Possessions,
Territories and Commonwealths. M ore than 15 million seabirds representing 59 species breed in this region.
Status and trends of each species w ill be discussed and all species will be prioritized by conservation need.
Colony and at-sea habitats used by seabirds will be described. The Regional Seabird Conservation Plan w ill
provide a comprehensive review of existing research and monitoring. Coordinated management and monitoring
strat egies w ill be developed in st ep-dow n plans. Th reats such as pol lut ion, i nteract ions w ith fish eries,
introduced species, disease, loss of habitat, disturbance, and global w arming will be discussed as will
recommendations for conservation actions. Opportunities to further seabird conservation through cooperative
and p art nership effort s w it h agen cies , ac ademia, and o thers w ill be explo red. Thi s plan w ill pro vide an
overarching review and discussion of seabird conservation in the Pacific Region and will identify priorities for
research, management, monit oring and f utu re direct ions f or conserv ation act ivit ies.

S-08 Parrish
Summary. JULIA PARRISH, School Aquatic & Fish. Sci., Univ. Washington, Seattle, WA.
     No abstract.

S-09 Hill & Badyaev
New perspectives in the evolution of sexual traits. GEOFFREY HILL and ALEXANDER BADYAEV, Dept. Biol.
Sci., Aubu rn Univ ., Aubu rn, AL.
        The evolution of condition-dependent t raits is a topic of central interest to evolutionary biologists, but
suc h w ork has been ham pered by an ov erly sim pli st ic ap pro ach i n w hic h t he beh avioral resp onses of females
are typically viewed as static. In this symposium, w e emphasize a new conceptual framew ork that explicit ly
recognizes that selection for higher mating success is more than just selection f or a bigger or more colorful
trait. Breeding success results from a complementary interaction t hat encompasses the preference and
performance of both sexes. The new perspective explicitly incorporat es life history theory into sexual selection
studies and includes the concepts of residual reproductive value and reproductive allocation into discussions of
sexual selection. Ult imately individual variation in allocation and choice in both males and females should favor
mult ipl e solut ion s and the evolu tion of con dit ion -dependen t repr odu ct ive tact ics . The im pli cat ion is t hat to
und erst and t he evolu tion of sexual t rait s, st udi es of sexual select ion should int egrat e the st udy of ind ividu al
hist ory, condit ion, m orphol ogy, and behavior f or bot h sexes.

S-10 Qvarnström
Adaptive plasticity in mate preferences. ANNA QVARNSTRÖM, Dept. Animal Ecol. Evol. Biol.Centre, Uppsala
Univ., Uppsala, Sweden.
         There is abundant evidence for the existence of marked mate preferences in natural populations, but t he
occurrence o f w it hin -popul ation var iat ion in m ate pref erenc es has received li ttle at tent ion and i s of ten reg arded
as non-adapt ive deviat ion s f rom the op timal no rm. How ever , i t has r ecent ly been show n t hat mat e pref erenc es
vary w ith context, as expected benefits from mate choice also vary. In the Collared Flycatcher (Ficedula
albicollis) males w ith large forehead patches (a sexually selected trait) trade-off their investment in mating
compet iti on and parental c are diff erently , depending o n th eir arrival dat e. Males arriv ing lat e, w hen most
females are f ert ili sed already , spend relat ively more ef fort on p arent al car e. A s a co nsequenc e, f emale c oll ared
flycat chers o nly display a pref erenc e for m ales w it h lar ge f oreh ead pat ches lat e in t he seaso n, w hen t hey
benefit from breeding wit h such males. In addition, female reproductive eff ort appears adjusted relative to both
timing of breeding and male phenotype. In conclusion, mate preferences not only display adaptive plasticity
w ithin populations, but this plasticity can also be linked to dif ferences in reproductive investment.

S-11 Price
Genetics of hybrid crosses. TREVOR D. PRICE, Dept . Bio l., Un iv. Calif orn ia at San Dieg o, San Dieg o, CA.
           Because many bird spec ies are kno w n t o hy bri dize i n nat ure, spec ies int egri ty is likely to be m aint ained
by various forms of post-mating isolation. I distinguish three kinds of post-mating isolation. First, there are
genetic incompatibilit ies that result in infertility or inviability of hy brids. Second, there are ecological causes of
isolat ion that arise becau se t he hy bri ds, alt hou gh v iabl e and f ert ile, are un able t o surv ive and r epro duc e as
effici ent ly in t he env iro nment . Fin ally , t here m ay be sexual select ion again st hybri ds. In t his talk I evaluat e
these causes. Genetic incompatibilit ies that result in infertility or inviability of hy brids accumulate over long
timescales, of ten lo nger than t he t imescale to speciat ion . The gen etics of hybri d cr osses in sexuall y select ed
trait s (i.e. , song and p lum age pat terns ) indic ate that hybri ds ar e of ten in termediat e or h ave unusual
combinations of color patt erns, which may place them at a moderate mating disadvantage, but t his is unlikely
to prevent females from mating. I conclude that ecological causes of post-mating isolation may often play a
crucial role in bird speciation.

S-12 Parker
Good genes separated from the possi bilit y of dif ferent ial maternal inv estment in Red Junglefo w l. TIM OTHY H.
PARKER, Dept. Biol., Univ. New Mexi co, A lbuquerqu e, NM.
         Accord ing to t he go od g enes hyp ot hesis, male at tractiveness signals male g enet ic q uali ty, and f emales
mate with more attractive males to obtain good viability genes for their offspring. A number of studies have
demonstrated a positive relationship betw een paternal att ractiveness and off spring quality. This pattern could
be due to a good genes eff ect, or it could be due to increased maternal investment in the offspring of more
attractiv e males. To isolate these tw o possibilities, I housed females w ith vasectomized (sterile) males and
inseminated them artificially. A ll sterile and sperm donor males were in the upper or low er third of t he
comb-size (attractiveness) distributions. Females did not differentially invest in reproduction in response to t heir
social mate' s attractiveness. However, condit ion and comb size were higher for the male off spring of
large-combed sperm donor males. This is strong support f or the action of a good genes mechanism in
maintaining female preference for large-combed males in this population. How ever, the lack of such a good
genes affect for female offspring suggests that genes affecting condition differ between males and females.
S-13 Lyon & Greene
Habitat-dependent variation in complex mating tactics in Lazuli Buntings. BRUCE E. LYON, Dept Ecol. Evol.
Biol. , Un iv. Calif orn ia, Sant a Cruz CA ; and ERICK GREENE, Div. Biol. Sci. Univ. M ontana, Missoula M T.
          The socially m onogamous l azuli bunt ing show s extr eme plumage color v ariation amo ng yearling m ales.
We examined the behavioral and fitness consequences of t his variation at 2 sites that diff er dramatically in the
availability of high quality habitat preferred by females, and consequently, in t he intensity of male-male
competition over t erritories. At M issoula, Montana, w here male competition for limit ed high quality t erritories is
intense, we found st riking disruptive sexual selection f or yearling male plumage coloration. The brightest and
dullest yearlings w ere able to obtain high quality territories, and hence females, w hile yearlings wit h
intermediate plumage obtained low quality territories and were unsuccessful at pairing. This pattern reflects, in
part, tolerance of dull yearlings by adults, who gained genetic benefits from having dull yearlings as neighbors.
In Alberta, w here buntings nest in aspen forest, t here is little competit ion among males for t he abundant high
quality territories. In cont rast to M ontana, plumage color had no influence on yearling male territory quality or
pair ing suc cess. The di fferenc e bet w een t he t w o st udy sit es co nf irm s a direct rol e for v arian ce in habi tat
territ ory qual it y as a prer equi sit e for t he co mplex mat ing tact ics observ ed at Missoul a.

S-14 Weatherhead
Sexual selec tion in Red-w inged Black bir ds: Mat chi ng m ult iple tr ait s t o mu lt iple fu nct ions. PATRICK
WEATHERHEAD, Program in Ecolog y and Evolut ion ary Biology , Univ . Illi noi s, Urbana, IL.
         To achieve reproductive success, male Red-winged Blackbirds must acquire a territory and then sire
young w ith f emales that nest on their territ ory or on the territ ories of nearby males. Conventional w isdom
suggests t hat, of t he males' several secondary sexual t raits, large size should help in co mpeti tio n fo r terr ito ries,
w hile song and plumage displays should help males acquire mates. Contrary to that w isdom, however,
acquisition of t erritories is unrelated to male size, or indeed to any other intrinsic male trait. Large males are
more successful att racting social mates independent of t erritory quality, but no male secondary sexual traits
appear relevant to siring young on or of f a male's territory . Particularly curious is the absence of an obvious
funct ion to t he males' signature feature, t heir brilliant red epaulets. The goal of this t alk will be to review the
evidence leading to these conclusions, to consider the insights to be drawn from experiments that have
manipu lat ed males' epaul ets, and t o pr opo se t he hy pot hesis t hat reso lut ion of these parado xes of Red-w ing ed
Blackbird reproduction might be found by examining aspects of the species' non-breeding ecology.

S-15 Badyaev
Selection on int egration of sexual traits and reproductive behaviors: Paternal care in relation to ornamentation
in the House Finch. ALEXANDER BADYAEV, Dept . Biol. Sci., Aubu rn Univ ., Aubu rn, AL.
           When in div idu als in a po pul ation dif fer in phy siolog ical con dit ion and r esidual r epro duc tive value,
selection should favor ph enot yp ic p last ici ty in cu rrent repr odu ct ive inv est ment suc h t hat ind ividu als are able t o
adopt the reproductive tactic t hat allows achievement of the highest fitness under given conditions. In a
Mont ana population of t he House Finches males used distinct reproductive tactics depending on elaboration of
their sexual ornamentation. Males w ith greater ornament elaboration paired w ith females that nested earlier,
but t hese males provisioned incubating f emales and nestlings litt le. In contrast, males wit h lesser ornament
elaboration p aired w ith females th at nest ed later, b ut t hese males fed female and nestlin gs more. The fit ness
consequences of alternative reproductive tactics of males w ere strongly aff ected by breeding experience and
fidel it y of their mat es. High variat ion in sexual or nament ation in a M ont ana po pul ation of the House Fin ch may
favor distinct associations of sexual displays with a particular set of reproductive behaviors. An appreciation of
individual variation in reproductive behaviors is an important st ep in establishing the selective pressures and
mechanisms underlying the operation of sexual selection.

S-16 Duckw orth & Stoehr
Proximate regulation of male mating phenotype in the House Finch: A role for condit ion-dependent hormone
int eraction s? RENÉ A. DUCKWORTH* Biol. Dept ., Duke Univ. , Durh am, NC; and ANDREW STOEHR, Dept.
Biol. , Un iv. Calif orn ia, Riverside, CA.
          The p rod uction of test ost eron e may serv e as a physiol ogi cal l ink behi nd t he co ord inat ion of male s exual
trait s and behav ior s and ind ividu al co ndi tion . A lt hou gh ab und ant evidenc e exist s t hat the pr odu ct ion of sexual
traits and behaviors is stimulated by a rise in testosterone levels, the link betw een testosterone production and
male condition is not w ell understood. We examined interactions betw een testosterone, individual condition,
sexual behaviors, and disease in both f ree-living and captive male House Finches. In free-living males, males in
high physiological condition and healthy males had higher levels of testost erone than males infected wit h
coccidi a. In t he cap tive st udy , ar tif ici al elev ation of test ost eron e levels resulted in more rapid onset of coc cid ial
infection w hile gonadal removal resulted in delayed onset of coccidial infection. In addition to eff ects on
disease and condition, w e found that t estosterone treatment had a positive effect on dominance rank and song
rate and a negative eff ect on parental eff ort. Our f indings suggest that testosterone is a physiological link
betw een male mating behaviors, sexual traits and individual condition. Because the elaboration of sexual traits
can inform females about a prospective mate's overall condition and health, f emales could benefit by assessing
male q uali ty based on t est ost eron e-dependen t sexual t rait s. We w ill also discuss int eraction s bet w een
pro lact in, test ost eron e and i ndi vidual con dit ion and t heir joi nt effect on p arent al car e.

S-17 Winker
Bird collections: Development and use of a scientific resource. KEVIN WINKER, Univ. Alaska M us., Fairbanks,
          Bird col lect ion s hav e und ergo ne pr of oun d ch anges since t heir inc ept ion . Today they serv e a bro ader
variety of scientist s asking a w ider array of questions that at any t ime in history. It is useful t o extend the
classic analogy that collections are like libraries to consider parallels betw een specimens and books as objects
st andi ng c onstant in a stream of cul tural and en viron mental ch ange. From this pers pect ive it is c lear t hat
collections suffer from temporal inadequacy. I w ill also consider science half-lives (our publications are long
forgot ten w hile specimens are still making contributions); specimens as biological filter paper (they represent
samples of t he environm ents in w hich t hey liv ed); and t he util ity of sp ecimens fo r monit oring (po pulati ons,
species, and environmental changes). Future collection growt h is urgently necessary. It must be driven more
by the gen eral p atterns of acqu isition and u se t hat giv e these r esources t heir value t oday than b y w hat ever
questions are presently f ashionable. We must become bett er at coupling short-term scientific gains wit h
lon g-t erm out loo ks an d act ivit ies, and i n t his arena i t seems that ind ividu als - n ot instit ut ion s - are the real
drivers o f pr ogress.

S-18 Causey & Trimble
Old bones in new boxes. DOUGLAS CAUSEY and JEREMIAH TRIMBLE, Mus. Comp. Zool., Harvard Univ.,
Cambridge, MA .
         Bird skeletons are a small part of most museum collections, but their use in scientif ic research is
increasing. Traditional uses include morphology, morphometrics, and comparative and functional anatomy.
Current research includes biogeochemical analyses, molecular studies of protein and DNA composition, and
other techniques. I will discuss the status of the w orld collection of avian skeletal material, the problems and
some solutions associated with curation and collection, and some examples of current research on osteology. I
will also address the issues of specimen collecting in modern ornithology.

S-19 Edwards, Birks & Brumfield
Geneti c reso urc es collect ions : A rch ives of evol ut ionar y hi st ory . SCOTT V. EDWARDS* , SHA RON BIRKS,
Burke Mus., Univ. Washington, Seattle, WA; and ROBB BRUMFIELD, Dept. Zool., Univ. W ashington.
          Genetic resources collections (collections of avian tissue and blood) drive much of t he research in the
systematics and molecular evolution of birds t oday. Because these collections usually consist of a variety of
tissues frozen in ultracold freezers or liquid nitrogen, are consumable, and are often linked to specimens in
tradi tion al co llec tion s, their lon g-t erm main tenanc e, d atabasing an d lo an po lic ies raise a number o f issues t hat
are diff erent from traditional collections. The various methods of field collection and long-term storage of blood
and tissues w ill determine the types of genetic analyses that can be accomplished; as molecular methods in
syst emat ics div ersify, inc reased attent ion to method s of field st orag e is w arrant ed. Genet ic r esources
collections are typically much younger t han traditional collections, and so their size, composition and diversity
oft en differ from t raditional collections. In addition, t he number of species and the number of
ind ividu als/ spec ies c ompri sing a t yp ical spec imen lo an di ffer bet w een t issue co llec tion s and tradi tion al
collections. These points are illustrated w ith data from a few of t he major genetic resources collections in the
U.S., relying principally on data from the Burke Museum collections. Genetic resources collections also provide
crit ical dat a for a v ariet y of resear ch pur suits in con serv ation genet ics , ec olo gy , b iom edic ine, enviro nment al
change, anthropology and forensics, and are likely to be key players in the increased use of museum collections
in the 21st Century.

S-20 Gaunt, Nelson, Dantzeker, Budney & Bradbury
New directions for bioacoustics collections. SANDRA L. L. GAUNT* , DOUGLAS A. NELSON, Borror Lab.
Bioacousti cs, Mu s. Biol . Div ersit y, Ohio St ate Uni v. , Columbu s, OH, MARC S. DA NTZEKER, GREGORY F.
BUDNEY, and JACK W. BRADBURY, Macaulay Lib. Nat. Sounds, Cornell Lab. Ornith ology, Ithaca, NY.
           The collection, st orage, retrieval, and distribution of specimens in the bioacoustics collections of t he
BLB and M LNS are u nder goi ng r evolut ion ary trans form ation s as t hese ex tensive c oll ect ion s are being dig it ized
and are opening portions of t heir databases to the Internet. These changes w ill markedly increase the
pres erv ation of , an d acc ess t o, spec imens and t heir associat ed metadat a. W hil e the use of dig it al sound has
been on the rise since its advent to bioacoustics in the 19 80s, digitization of sound collections is only now
becoming practical as the costs of reliable memory, especially optical media (CD and DVD), have fallen to
accessible l evels. Digit izat ion to op tical media no t onl y increases the arc hiv es’ lon gevit y, but also facil it ates
more auto mated access t o th e specimens. Wh en coupled w ith w eb-accessible databases, public access to
these archives will rise to new heights thereby broadening the base of users and research topics t hat can now
be accomplished in these collections. At the MLNS, the digital revolution also includes the move to multiple
media, adding a sister collection of visual representations of animal behavior (still and moving images) to the
grow ing digital library. Throw ing open the digital doors to the BLB and MLNS collections w ill increase their
uti lit y and relevance f or t omorr ow ’ s pressing research, c onservat ion, and educ ational needs.

S-21 Peterson & Navarro
The Species Analyst: Making the most of natural history museum specimen information for natural history
museums and for science, policy, and conservation. A. T. PETERSON, Nat. Hi st. Mus. , Univ . Kansas,
Lawrence, KS, and ADOLFO NA VA RRO, Facultad de Ciencias, UNAM, M exico, Mexico.
         Natural hist ory m useums hold v ast st ores of i nfo rmati on about biodiv ersity : dist ribut ion of species,
seasonality, breeding biology, abundance, historical changes, and conservation can benefit enormously from
this information. Access to this information, how ever, is woefully inefficient, reducing the uses to which
specimen data are actually put. The Species Analyst (htt p:// speciesanalyst. net) represents a new solution to
this problem: data resources w orldw ide are united in a single virtual database accessible to all. By maintaining
dat a reso urc es at instit ut ion s w here v ouc her s peci mens are housed, TSA emphasizes bo th ow ners hip of dat a
by the museum and the critical need for specimen and data to be intimately associated. By providing eff icient
access to dat a, specimen inf ormat ion becomes cent ral in div erse applications: conservat ion, p redict ing species'
invasions, and others. By emphasizing community c ooperation, TSA of fers (1) a means of cooperation w ith
countries worldw ide that are rich in biodiversity; (2) an avenue to increased funding for natural history
museum s; and (3) a more cent ral r ole f or n atural history museum s in bio div ersity issues. TSA pres ent ly serv es
about 15 m illion specimen records from 11 institut ions, including 4 ornithological collections (several more soon
to be connected). Approximately 4 5 million specimen records are presently comm itt ed to participation in the
netw ork, making it an important source of specimen information.

S-22 Remsen
New uses of specimen data for distributional and ecological analyses. J. V. REMSEN, Jr., Louisiana State Univ.
Mus. Nat . Sc i. , Bat on Rou ge, LA.
         Although specimen localities have provided the backbone for biogeographic analyses in ornithology,
their full pot ent ial i s seldom reali zed. For ex ample, inc orp orat ing elevation al dat a int o such analy ses has
provided new insights into hist orical and ecological distribution patt erns. Also, incorporating seasonal data into
suc h anal yses has rev ealed t hat our kno w ledg e of w int er di st rib ut ion s in the t rop ics is dominat ed by cop y-cat
error perpetuation. A dditional data fields on modern specimen labels allow analyses that are also currently
under-used. For example, label data on gonad size, subcutaneous fat levels, and stomach contents can provide
rapid and reasonable approximations of basic biological information on seasonal life-history patt erns and diets of
birds that w ould take decades of fieldwork to ot herwise study. Sample sizes of specimens needed for such
analyses are smaller than an average day' s accumulation of bird deaths from natural causes for most bird

S-23 Rocque
The use of avian collections in contaminant and stable isotope studies. DEBORAH A. ROCQUE, Univ. Alaska
Mus. , Fairbanks, AK.
          As the scale and diversity of environmental changes increase, museum specimens can be invaluable
tools for assessing ecosystem status and trends. Technological advances have helped release new information
from preserved specimens and enable researchers to conduct long-term monit oring studies using data from
histori c material . For exam ple, enviro nment al t race met als and st able i sot opes are in corpo rat ed in to f eather
keratin during growt h and remain inert, m aking specimens a valuable resource for long-term biomonit oring
studies. Elements bound in feathers can be used in retrospective studies to document changes in contaminant
concent rat ion s and, i n t he cas e of st able i sot opes, c hang es in diet or f ood w eb st ruc ture. Stable i sot opes can
also be used t o rec onstruc t diet s in ver y r are or ext inc t spec ies. Used i n co ncert , an alyses of st able i sot opes
and contaminants from specimens can be used to assess the health of specific systems or food w ebs and
provide insight into cont aminant pathways. A s climate change, loss of biodiversity, and pollution impact t he
environment on a global scale, baseline data for assessing long-term trends becomes increasingly important.
Given the accelerated rate of global change, the need to collect and archive specimens for biosphere monitoring
has never been greater.

S-24 Kiff
The history, present status, and future prospects of avian egg collections. LLOYD F. KIFF, The Peregrine Fund,
Boise, ID.
          Egg collecting, i.e., t he preservation of empty eggshells, w as a popular hobby in the British Isles, North
America, and a f ew ot her c oun tries dur ing the last half of the 1 9t h Cent ury and t he f irs t half of the 2 0t h
Century. Alt hough most " oologists" w ere amateurs, they recorded reliable data wit h their specimens, and
nearly all of the major collections found their w ay into museums. About 80 egg collections presently survive in
North Am erican institut ions, and they include about 50 0,0 00 egg sets. On a global basis, there are probably
no more than 300 major collections, and they contain a disproportionate representation of taxa from w estern
Europ e, Nort h A merica, Sout h A fric a, an d A ust rali a. Eggshells and t heir associat ed dat a have pro babl y been
used in a g reat er v ariet y of bio log ical st udi es t han an y other t yp e of avian specim ens. Egg co llec tion s are a
part icu larl y v aluab le t ool for c onserv ation bio log ist s and have been c rit ical ly impor tant , f or ex ample, in studi es
of the effects of modern environmental contaminants and documenting historical ranges of vulnerable species.
However, f ew egg specimens are being preserved by contemporary ornithologists, ow ing to t he loss of
knowledge of proper preparation techniques, lack of institutional support, and unfavorable funding fads. As
w it h ot her t yp es of syst emat ics col lect ion s, the co nsolid ation of egg c oll ect ion s in regi onal cent ers d evoted
principally to research may be the eventual solution to perennial funding, permitting, and conservation

1        Beresford
Hints and guesses about hig her- level py cnono tid systemat ics from mt & nu sequen ce dat a. PA MELA
BERESFORD, Am . M us. Nat . Hist . and Cit y Univ . New York, New York, NY.
         While searching for the sister group(s) to Bleda and Criniger (Beresford 199 9, A OU), more than 30
exem plar s in 14 gener a of Pycn ono tidae w ere sam pled for m it ochon dri al cy t-b (11 43 bps) and nuc lear b eta
fibinogen intron 5 (~5 40 bp) sequence data. The phylogenetic relationships among members of bulbuls should
be of particular interest t o those wishing to understand the evolution of Old World forest avifaunas: in Africa
the family contributes as much as 10% of the species (Moreau 1966, Birds of Af rica), and bulbuls have the
highest biomass in Gabon (Brosset 199 0: 2 63 in Biogeography and ecology of forest bird communities). Here
results f rom mi toc hondri al and nuclear data are presented prim arily t o clarif y f utu re research needs.
Characteristics of t he 2 loci are compared: cyt b w as variably informativ e for recovering intergeneric
relationships, and several large gap regions were apparent in t he nuclear intron for species of Pycnonotus. The
combined dat a ind icat es an A fro- Asian clad e distinc t from an end emic Afric an cl ade; ref ut es t he sist er
relationship betw een Nicator and Bleda (contra classifications from Delacour (1943, Zoolgica NY 28:17-28)
thr ough M onroe and Sibley' s Checklist (199 3), supports t he lack of close affinit y betw een Nicator and o ther
pycnonotids found by Dowsett et al. (1999, Ibis 141:22-28), corroborates the finding of Roy et al. (1998,
Steenstrupia 24: 51-66) that " Andropadus" is not monoph ylet ic and suppo rts Hall & Mor eau's (At las of
speciation in Af rica Passerine Birds, 19 70: 68) removal of Asiatic " Crin iger " to Alophoixus. Si nce a
phylo genetic f ramew ork is in valuable fo r stu dies foc used on t he evolut ion of behavioral and ecolog ical tr aits,
strategies for increasing accuracy in future efforts to resolve phylogenetic relationships among bulbuls are

2        Cibois
Phylogenetic relationships of babblers based on mitochondrial DNA sequence. ALICE CIBOIS, Am. Mus. Nat.
Hist. , New York, NY.
           The babblers, family Timaliidae, are an important group of Old World insectivorous passerines. The
family comprises > 200 species, primarily forest birds, distribut ed for the most part in Southeast Asia and
Afric a. The bab bler s display great morph olo gic al and ecol ogi cal d iversity, and t he sy st emat ics of the Ti mali idae
and related members of the Old World insectivorous passerines have been particularly difficult . In order to
clar if y our u nder st andi ng o f the bab bler gro up, phy log enet ic r elat ion ships w ere recon st ruc ted using sequences
of 3 m it ochon dri al gen es (cyt ochro me b, r RNA 1 2S and 1 6S). The r esults indic ate that several speci es
tradit ionally p laced among babblers are not related t o th e Timaliidae, but belong t o ot her passerine groups.
Furt herm ore, the ph ylogen etic h ypot heses inf erred f rom molecu lar d ata show that the bab bler s can be di vided
in 2 m ajor group s. The f irst one includes Sylvia and 5 Asian and African babbler genera, the second group
includes all the other babblers studied (> 50 species). Several babbler genera are polyphyletic, and results
from the molec ular phy log eny chal leng e the t radi tion al cl assificat ion of the Ti mali idae.

3        Cracraf t & Feinstein
Species and speciation in the birds of paradise. JOEL CRACRAFT and JULIE FEINSTEIN, Am. Mus. Nat. Hist.,
New Yor k, NY.
          Under th e biological species co ncept (BSC) the " standard m odel" of allo patric speciation envisions a
time-line of allopatry, divergence, barrier breakdown, sympatry, f ollow ed by either coexistence (if reproductively
isolat ed and /o r ecolo gic ally div ergen t) or hybri dizat ion . This cano nic al view of spec iat ion w as long ag o co dif ied
in t he minds of many bio logist s by t he classific atory view s of ev olut ionary sy stemat ists w ho applied ranks t o
pop ulat ion s/taxa t hou ght to f oll ow anot her t ime-line: subspecies t o species to gen us. Because the BSC co upl ed
not ion s of how taxic di fferent iat ion unf old s w it h t axonom ic r anki ng, the st andar d model has c onf oun ded
speciation analysis in three important w ays: (a) ranks have been explicitly linked to perceptions of the amount
of change, (b) there has been a frequent and tacit assumption of constant rates of dif ferentiation (i.e., using
taxonomic rank as an index of amount of change and age), and most importantly (c) it has hindered the search
for b asal diagn osably distinc t pop ulat ion s (basal t axa) and t heir history . Few gro ups of bir ds have at tracted
mor e int erest fr om advocat es of th e BSC than b irds of paradise, and no group bet ter exemp lif ies how th e BSC
and evolutionary classification have led to misinterpretations of t heir history. Our ongoing phylogenetic
analy ses of basal taxonom ic u nit s (phy log enet ic s peci es) in dic ate that div ersificat ion in t his gro up h as been
anything but st andard: (a) little sympatry exists among close relatives, (b) hybridization is routine among
phenotypes (ranked at genera) that have been stable over millions of years, and (c) morphological and genetic
div ergen ce has appar ent ly been u ncoup led an d epi sodic. The BSC, w it h it s ahist ori cal m ethod olo gy , h as
obscured understanding of BOP diversification.

4       Klicka
A m olecular phy logeny f or t he avian genus Geothlypis and some allies. JOHN KLICKA, Barrick Mus., Univ.
Nevada Las Vegas, Las Vegas, NV.
         Mor phologi cal similarit ies among several members of thi s genus (G. t ric has, G. beldingi, G. r ost rat a, and
G. f lavov elat a) have led to t axonomic uncertainty. I use molecular methods to investigate generic limits among
members of the ent ire Geothlypis-Oporornis complex and use this informat ion to clarify relationships among the
taxa co mprisin g th is Geothlypis " superspecies" group. Representatives of all species of Geothlypis and
Oporornis w ere sampled. Sequenced regions of the mt DNA genome were chosen to allow for 2 distinct levels
of reso lut ion . These includ e an 1 03 2 b p segment of the mtDNA cy tochro me b gene and 1243 bp from t he
(pr esum ably ) mo re v ariab le co nt rol regi on. Prelim inar y analyses indi cat e that G. speciosa, G. semiflava, and G.
nelsoni, t oget her w it h t he members o f the " super sp ecies" gro up d efined abov e, f orm a hig hly suppor ted
(100 % MP boot st rap) clad e. The Com mon Yellow thro at (G. t ric has) as presently recognized is polyphyletic.
Tw o (eastern and w estern) Geothlypis clades are recovered, diff ering by approximately 1.6% sequence
div ergen ce. Geot hly pis tric has (w est) and G. beldingi are closely related sister taxa. Geot hly pis tric has (east)
are closely related to G. f lavovelat a and G. nelsoni. These results are examined from a biogeographic

5       Outlaw, V oelker, Mila & Girman
Molec ular syst emat ics of the Catharus thrush complex: Implications for the evolution of avian migratory
behavior. DIANA C. OUTLAW* , GARY VOELKER, Barrick M us., Univ. Nevada, Las Vegas, NV; BORJA MILA,
Dept . Or g. Biol. , Un iv. Calif orn ia, Los A ngeles, CA ; and DEREK GIRMAN, Dept. Biol., Sonoma State Univ.,
Sono ma, CA.
          The Catharus thru shes are a gr oup of 7 r esident and 6 migran t spec ies. They pro vide a po tent ial m odel
system wit h w hich to examine phylogenetic and morphological patterns in the evolution of migratory behavior,
and to assess potential Pleistocene glacial effects on speciation timing of m igrant species. We reconstructed a
molecu lar p hy log eny (976 base pairs of the cy tochro me b gene) of all Catharus spec ies and w ere abl e to
determine several important aspects in the evolution of migratory behavior in this lineage. Migrant and resident
Catharus species do not form reciprocally monophyletic groups; phylogenetic evidence suggests that migratory
behavior has evolved at least tw ice from a resident ancestral character state. One radiation of migrant
Catharus species occurred w ithin the Pleistocene; a species-level radiation during this period is a rare
occurrence, although Pleistocene glacial activity is an historically-invoked avian speciation mechanism.
Morphological diff erences betw een migratory and resident Catharus species suggest t hat the longer wings of
migran t bir ds ar e an adap tive char act er st ate for m igr ant spec ies. The Catharus lineage fits Cox’s (1985, Am.
Nat. 1 26 : 4 51 -474 ) st eppi ng-st one m odel for t he evolu tion of migrat ory behav ior , w hic h pr edic ts t hat seasonal
enviro nment s nort h of the M exican Plateau may have driv en t he evolu tion of migrat ion in some Neot rop ical
avian li neages
6        Burns, Hackett & Klein
Phylo genet ic r elati onsh ips and mo rpholog ic diver sit y in Darw in' s f inches and t heir r elati ves. KEVIN J. BURNS,
Dept . Bio l., San Diego State Univ ., San Dieg o, CA; SHANNON J. HACKETT, Field Mus, Chicago, IL; and
NEDRA K. KLEIN, Div. Sci., Truman State Univ., Kirksville, MO.
          Despite the importance of Darw in' s finches to the development of evolutionary theory, the origin of t he
gro up h as only recent ly been ex amin ed using rig oro us, phy log enet ic m ethod s t hat inc lud e many pot ent ial
outgroups. Know ing the evolutionary relationships of Darw in' s finches to other birds is important f or
understanding the context from w hich this adaptive radiation arose. DNA sequence data from the cytochrome
b gene indicate that Darwin' s finches are derived from a group of species whose geographic distributions are
cent ered i n t he Cari bbean . These c lose relat ives of Darw in' s f inc hes show a div ersity of bil l t yp es as great as
that observed among t he Darw in' s fin ches th emselves. Previous t axa that have been proposed as the cl osest
livin g relativ e of Darw in' s fin ches (Volatinia, Coereba, Sporophila, Melanospiza, and Tiaris obscura) can be
excluded using likelihood-based statistical tests. Recent studies have show n that adaptive evolution in
Darwin' s finches occurred relatively quickly. Our data show t hat among the relatives of Darw in' s finches, the
evolution of bill diversity w as also rapid and extensive. Thus, the adaptive evolution of bill size and shape
among Darw in' s f inc hes can no w be seen as an ex tension o f the evolu tion of div erse f eedin g morp hol ogi es
among t heir close relati ves.

7        Joseph, W Ilke & Alpers
A m olecular approach t o th e evolut ion of host specifi cit y in som e Aust ralian bronze-cuck oos Chalc it es spp.
LEO JOSEPH* , THOMAS WILKE and DERYN ALPERS, Acad. Nat . Sci. , Philadelp hi a PA .
         We used mtDNA t o test a hypothesis derived from theoretical and field studies that host races exist in
Horsfield's Bronze-Cuckoo Chalcites basalis. We also studied several other Australian bronze-cuckoos in which
evidence of host races is absent or equivocal. We extend molecular study of t he coevolutionary history of
brood parasitic cuckoos and their hosts to the evolutionary "theatre" provided by the Australian environment.
Our expectation w as that if int rinsically controlled host races in C. basalis have had long, stable co-evolutionary
histories wit h their present Malurus and Ac ant hiza hosts as argued by others, then sequence divergences and
nucleotide diversities in C. basalis should be high and phylogenetic structure in mtDNA should be deep.
Analyses to date show t hat these expectations are not met. If anyt hing, our data have brought to light a
paradox: unexpectedly low nucleotide diversities and lack of phylogenetic struct ure in mtDNA of C. basalis
despite evidence from ecology and behaviour for host races in it. The data provide an opportunity to apply a
range of stat istical tests and analyses that test f or neutrality of m tDNA evolution and/or w hether a population
has been stable, expanding or declining. These include mismatch analyses and Tajima's D, Fu & Li' s F* and
D* , an d Fu' s Fs st atist ics . A naly ses t o dat e suggest that the 3 Chalc it es species studied have had differing
evolutionary histories, C. basalis in p art icu lar l ikel y having u nder gon e a rang e expansion t hat has over w rit ten
any mt DNA sign al arising f rom it s undoubt ed host sp ecific ity and host races, if they exist .

8        Marshall
* Nest predation and microclimate as factors in nest site selection. JAMES S. MARSHALL, Dept. Environ. &
Forest Biol., SUNY-ESF, Syracuse, NY.
         Nest site selection has most often been att ributed to avoidance of nest predation, f avorable
microclimate, or proximit y to f ood resources. Of t hese, nest predation and microclimate have received the
most suppor t. I sought to add ress bot h of these f act ors in t he nest sit e selection of the A lder Flyc atcher, Gray
Catbird, and Song Sparrow in a power line right-of-w ay. All 3 species exhibited preferences for certain
vegetation . The Son g Spar row w as t he on ly spec ies, how ever , w hose vegetation pref erenc e cor resp ond ed
w ith a factor t hat significantly inf luenced nest predation. Temperatures at nest sites did not dif fer significantly
from temperat ures at rand om sit es. Result s f rom the Son g Spar row suggest that nest pred ation may be an
impor tant fact or i n nest sit e selection , an d t hat I did no t measu re t he rel evant vegetation var iabl es f or t he ot her
species. Birds do not appear to select nest sites based on microclimate in this habitat. Food may also play a
larger role than suggested by some earlier research.
9       Nagy
* The role of territory-based food availability on female variation in double brooding. LAURA R. NAGY, Dept.
Biol. Sci. , Dart mout h Coll., Hanover, NH.
           An im por tant det erminan t of annu al br eedin g pr odu ct ivit y is t he nu mber of nest ing attempts/season, yet
lit tle is kno w n abo ut w hat fact ors det ermine i ndi vidual var iat ion in t he nu mber of nest ing attempts. I pr ov ided
supplem ent al f ood to 1 4 f emale Black-t hro ated Blue Warbler s t o experi mentally test the ro le of territ ory -based
food availability on the number of broods produced. Females provided with f ood initiated second broods more
of ten t han c ont rol females (7 of 7 v s. 5 o f 10 , P = 0. 04 ). In c ont rast , t here w as no dif ferenc e bet w een f ed
and control nests in the number of young fledged or the weight of the young in the first nest (3.25 ± 1. 03 and
3. 71 ± 0. 49 fledg lin gs, 7. 92 ± 0. 51 and 8 .0 3 ± 0. 76 g, respect ively). Thi s co mbinat ion of resu lt s suggest that
adults (probably through energetic modifications) w ere able to produce a first nest of comparable young, but, in
doing so, sacrificed the ability t o produce a second nest. These results demonstrate that dif ferences in food
availability at t he territory level influences the frequency at w hich broods are produced and therefore ultimately
annual recruitment and population size. In addition, the increase in double brooding w ith t he presence of
supplement foo d suggest s that a cost of reproduc tio n exist s and the abilit y of a female to m eet th ese costs
const rains th e number of breeding att empts.

10      Quader
* Mat e choice and nesti ng success in r elation t o nest st ruct ure in a w eaver-finc h (Ploceidae). SUHEL QUADER,
Dep t. Zool., Univ . Flori da, Gainesv ille, FL.
          Poly gy nou s w eaver s are am ong the f ew bir d species in w hic h nests are buil t solely by males . In sev eral
of t hese species, males display partially completed nests to visiting f emales. If a female accepts a nest, t he
male completes it, and eggs are laid. Do females use nest struct ure as a mate-choice cue? I tested this
hypothesis in an Indian species, the Baya Weaver (Ploceus philippinus). I measured female visits t o nests, and
then exchanged nests between males w ho received frequent visits (popular males) and those w ho did not
(unpopular males). If nest structure is the primary determinant of female choice, post-manipulation visit ation
should change such that f ormerly unpopular males are now popular. Contrary t o this prediction, f emale
visit ation f requency r emained higher at f ormerly popular males t han at f ormerly unpopul ar males. Because
visitation f requency is a good predictor of eventual nest acceptance, I conclude that nest struc ture is not a
primary mate-choice cue. Additional support f or this conclusion comes from the absence of measurable
differences in structure betw een unmanipulated nests that were eventually chosen and those that were not. A
multivariate analysis revealed that nest location is a better predictor of nesting success than is nest structure.
This is a possible reason w hy nest stru ct ure is not an import ant mat e choice cue in t his species.

11      Gjerdrum
* Regulation of parental provision ing and nestling depart ure decisions: A supplementary f eeding experiment In
Tufted Puffins. CARINA GJERDRUM, Dept. Bi ol. Sci., Sim on Fraser Univ. , Bu rnaby , BC.
           In 199 9 and 200 0, a supplementary feeding experiment w as used to examine the regulation of
provisioning effort and nestling departure behaviour in Tufted Puffins on Triangle Island, British Columbia. From
the parents' perspective, life history t heory predicts regulation of parental eff ort in order to balance current
investment w ith f uture expectation of survival and reproduction. I tested the hypot hesis that nestling
nut rit ional demand regulat es provisio ning and predic ted t hat parent s of su pplement ed nestling s w ould decrease
pro visionin g ef fort . The bi ll l oads deli ver ed by paren ts of supplem ent ed nestlin gs did n ot dif fer bet w een
treat ment gro ups. How ever , p arent s of supplem ent ed nestlin gs prov isioned signif ican tly less frequ ent ly than
did parent s of c ont rols. Result s fro m t he years of t his st udy sup port the hy pot hesis that parents adjust
provisioning effort in accordance w ith nestling demand. The second objective of t he study w as to test the
hypothesis that nestlings consider the provisioning behaviour of their parents when t iming their departure from
the nest (Morbey et al. 1999, Anim. Behav. 57 : 87 3). By supplementing nestlings late in the nestling period
w hen p arent s t yp ical ly decr ease t heir rat es of pro visionin g and nest lin gs lose m ass, I experi mentally inc reased
the nestling' s ' expectation' of another feed and therefore the benefits of remaining in the nest. A s predicted,
supplement ary f ed nestling s remained in t he nest signi fic antly longer t han did con tro l nestlin gs.

12      Coe
* How does w ater availability affect reproduct ive success in desert birds? A t est using water supplementation
in the Black-throated Sparrow. SHARON J. COE, Dept . Bio l., Un iv. Calif orn ia, Riverside, CA.
         The w ays in w hic h desert bir ds m aint ain suf fici ent bod y w ater t o meet their day-t o-day phy siolog ical
needs has been the focus of a large body of scientif ic research. However, t he relationship betw een aridity and
reproduc tiv e success has not b een well st udied. Dur ing 2 consecut ive breeding seasons (20 00 and 20 01 ),
water was provided to Black-throated Sparrows on breeding territories in the e. Mojave Desert of s. California.
Nests of pairs on both a treatment plot (surface water provided) and a control plot (no water provided) w ere
monitored to determine clutch size, the number of young that hatched, and fledging. During the 20 00 breeding
season, 19 treatment nests and 16 cont rol nests were located. The results of a Mann-Whitney t est showed a
significant difference in clutch size between treatment and control nests (U = 162; df = 16,14; P< 0.05).
There w as no significant dif ference in daily nest survival probability (1- [number of f ailed nests/number of
exposure days]) for treatment and control nests observed during incubation and nestling stages ( P> 0.05;
significance test based on Hensler & Nichols 1981, Wilson Bull. 93: 42-5 3). Results from the 200 1 breeding
season will be presented. This research has implications for estimating the proximate eff ects on bird
reproduction should global climate changes result in deserts becoming hott er and dryer, as well as for assessing
the effects from artificial water sources placed for wildlife in arid regions of the U.S.

13      Wilhelm
* Var iat ion in p arent al ro les am ong Common M urr es: Eff ect of food availabi lit y. SABINA I. WILHELM,
Bio psych . Prog. , Memor ial Univ. New foundland, St . John' s, NF.
           Coop erat ion bet w een mates durin g ch ick rearing inc reases paren tal ef fici ency . W it hin the A lci dae
family, most species studied to date show m ales as primary "brooder" and females as primary "provisioner",
including Common Murres. The present study investigated how parental roles w ithin pairs change depending
on food availability in different years. Chick feeding rates and time spent brooding chick w ere obtained from
behavioural observations of breeding Common Murres during chick rearing periods wit h low , high, or average
food availability, on Great Island, Newf oundland. Role partit ioning w as examined for the same pairs (n = 15)
across all 3 con dit ion s. During aver age and low food year s, males spent more t ime bro odi ng an d f emales
provisioned more only when food availability w as average. No sex differences were found during high food
availability as a result of males' increased provisioning rates and females' increased brooding time. However,
role partitioning at the individual level revealed that over 60% of individuals brooded and provisioned as much
as their mates, independent of condition; only w hen individuals assumed one role during average and low f ood
year s did males ass ume the ro le of bro oder and f emales of pro visioner. Assuming a p art icu lar r ole u nder
average or poorer chick-rearing conditions may reflect a selective advantage of role partit ioning under certain
ecological constraints.

14       Keedwell
* Post-fledging mortality in Black-fronted Terns: Does fledging equal success? RACHEL KEEDWELL, Ecology
Group, Massey Univ., New Zealand.
         Black-fron ted Terns (Sterna al bostriat a), an end emic New Zealand spec ies, are li st ed gl obal ly as
endan gered and ar e in d ecli ne pr imaril y due t o t he im pact s of int rod uced mammalian pred ators . M ort alit y r ates
are high at for black-front ed tern eggs (60% ) and chicks (75% ), but lit tle is know n about the impact of
predation at other lifestages. In this study, I attached radio transmitters to black-fronted tern chicks prior to
fledging to monit or rates and causes of mort ality in the immediate post-fledging period. Minimum m ortality
rat es f or t he f irs t mont h af ter f ledg ing w ere 3 1% (n = 13 ), 22 % (n = 18 ) and 13 % (n = 32 ) in the 1 99 8,
19 99 and 20 00 breeding seasons, respect ively. Intr oduced predat ors such as f eral cats (Felis catus) and
Norw ay rats (Ratt us norvegicus) w ere responsible for a minimum of 7 5% of all deaths. A dditional causes of
mortality included starvation and necrotising enteritis. A ll except 1 of t he juveniles that died did so wit hin the
first w eek after fledging, all other juveniles survived until they dropped their transmitters or left t he area. Some
new ly fledg ed ju veniles cou ld b e easily r ecapt ured by hand after t hey w ere seen flying , sugg est ing that
although they w ere capable of f lying, t hey took several days to rely on flight as their primary mechanism of
escap e from pred ators . These r esults indic ate that mort alit y init iall y r emain s high p ost -f ledg ing and t hat
fledgi ng rate alone is not an accurate measure of produc tiv ity in t his species.

15      Sanders, Otis, Costa & Gauthreaux
Brood reduc ti on in Red-cock aded Wo odpeckers . FELICIA SA NDERS* , South Carolina Dept.Nat. Res.,
Georget ow n, SC; DAVE OTIS, Sout h Carol ina Coop. Fish & Wil dl. Res. Unit , Clem son Un iv. , Clem son, SC;
RALPH COSTA, U.S. Fish & Wildl. Ser., Clemson, SC; and SIDNEY A. GAUTHREAUX, Dept. Biol. Sci.,
Clemson Univ.
          Know ledge of Red-cockaded Woodpecker (RCW) reproductive biology has been constrained because of
the diff iculty of repeated observation of the nest cavity w ithout potentially impacting reproductive success of
this federal ly endan gered spec ies. We receiv ed special perm ission from the U. S. Fish and Wild lif e Serv ice t o
use 3 new field t echniques t o examine RCW brood rearing. First, modif ied artif icial cavit y insert s allow ed easy
remo val of eggs an d nest lings. Secon d, camer as installed i n insert s record ed t he f irst videot apes of RCW
nesting activity in the nest chamber. Lastly, 40 groups of RCWs w ere examined daily from t he time of hatching
until nestlings were 10 d old. Data collected was used to investigate the brood reduction and insurance
hypot heses as explanat ions f or asynch ronous h atchin g in a South Carolina RCW popu lation . Partial broo d loss
of eggs and n est lin gs w as 41% for 1 99 8 an d 1 99 9 c ombin ed. Daily m ort alit y r ates of nest lin gs f rom part ial
brood loss were higher for the period of 0-8 d (0. 029 ) than from 9 -fledging (0.00 9) (P = 0.0 001 ). The
percentage of unhatched eggs was variable among most clutch sizes and average 20% for 1 998 and 1999
combined. Brood reduction increased w ith greater brood size (number of eggs that hatched; P = 0.0 07) but
did not vary sig nif icantl y w ith the num ber of eggs t hat w ere laid (P = 0. 52 ). These result s suggest
asynchronous hatching may provide RCWs a means to eliminate extra young w ith minimal investment if last laid
eggs are not needed as replacements, which is consistent with the insurance hypothesis.

16      Conner, Saenz, Rudolph & Schaefer
Does t he availab ili ty of art if ici al cav it ies af fect cav it y ex cav ation rat es in Red-co ckaded W ood peck ers?
Lab., Southern Res. Sta., Nacogdoches, TX.
          Rates of cavity excavation by Red-cockaded Woodpeckers w ere examined from 198 3 t o 199 9 on t he
Angelina National Forest in eastern Texas. We compared the rate of natural cavity excavation betw een 198 3
and 1990 (before artif icial cavities w ere available) w ith t he rate of cavity excavation betw een 199 2 and 199 9;
a period when artificial cavities w ere available. Our comparison was restricted to cavity-tree clusters w here
w ood peck er gr oup s w ere pr esent for t he ent ire 1 7-yr peri od. The ex cav ation rat e of new cav it ies w as
signif ican tly hig her i n lo ngl eaf pin e habi tat (n = 9) dur ing the f irs t 8 y r (0 .3 06 new cav it ies
excavated/group/year) when artificial cavities were not available than it was during the subsequent 8 yr (0.125
new cavit ies excavated/group/year). In loblolly-shortleaf pine habitat (n = 5), w e failed to detect a diff erence in
the rat e of new cav it y ex cav ation bef ore ( 0. 55 0 c avities ex cav ated/ gro up/ year ) and after (0 .3 75 cav it ies
excavated/group/year) the use of artificial cavities. We attribute this difference betw een habitats to a relative
scarcity of cavities in loblolly-shortleaf pine sites due to much higher bark beetle-induced cavity-tree mortality.

17      Saenz, Conner & McCormick
Do Red-cockaded Woodpecker c avit y t rees attr act Pileated Wood peckers? DANIEL SAENZ* , RICHARD N.
CONNER, Wildlife Habitat Lab., Sout hern Res. Sta., Nacogdoches, TX; and JAMES R. McCORMICK, Dept.
Biol., Stephen F. Austin State Univ., Nacogdoches TX.
          Red-co ckaded W ood peck ers ex cav ate their roo st and n est cav it ies ex clu siv ely in l iving pin es. Pileat ed
Woodpeckers are known t o cause damage to Red-cockaded Woodpecker cavity trees in the form of cavity
enlar gement or o ther ex cav ation s on the surf ace of the pi ne t ree. How ever , i t is not kno w n w het her Pil eated
Woodpeckers excavate more frequently on Red-cockaded Woodpecker cavity t rees than on non-cavity t rees or
how stand structure aff ects the frequency of Pileated Woodpecker excavation. It is also unclear w hether the
cavity it self is providing the stimulus to t he Pileated Woodpecker to excavate or whether the presence of
Red-cockaded Woodpeckers and their activities are att racting them. We surveyed all of the cavity t rees (n =
827 ) on the Angelina National Forest and 110 cont rol trees for recent Pileated Woodpecker excavation and we
found t hat approximately 5% of all cavity trees were damaged while no control t rees showed any evidence of
Pileated Woodpecker damage. The rate of Pileated Woodpecker excavation w as negatively correlated to
hardwood midst ory height and density. Pileated Woodpeckers appeared to focus most of t heir excavations on
Red-cockaded Woodpeckers cavity entrances. We suggest that Pileated Woodpeckers are attracted to
Red-cockaded Woodpecker cavity t rees, especially the cavity, and current management practices (midstory
removal) used to improve Red-cockaded Woodpecker habitat may increase the incidence of damage to t he
cavit y t rees by Pileated Woodpeckers.

18      Koenig & Haydock
Reproductive skew in the polygynandous Acorn Woodpecker. WALT KOENIG*, Hastings Reservation, Univ.
Calif orn ia, Carmel Valley, CA ; and JOEY HAYDOCK, Biol. Dept., Gonzaga Univ., Spokane, WA.
          Reproductive skew m odels, which f ocus on the degree to w hich reproduction is shared equally or
monopolized by a single individual wit hin groups, have been heralded as providing a unifying f ramework f or
und erst andi ng social ev olu tion . Based on o ur d emograp hic and p arent age dat a, w e test the abi lit y of opt imal
skew models to pr edic t repr odu ct ive part it ion ing in t he A corn W ood peck er w hic h, as one of the f ew ver tebrat es
exhibiting polygynandry, prov ides a key test case of this approach. Joint-nesting females share reproduction
more equitably than expected by chance, apparently due to egg destruction combined w ith t he inability of
females to recognize or defend their eggs. In contrast, male skew is considerable, and is consistent w ith some
predictions of skew t heory. However, high male skew appears not to be consistent betw een nests and we have
thus far been unable to determine any cue that predicts w hich of a set of cobreeder males is likely to sire the
majority of y oung. These results suggest an alternative model by w hich paternity is not determined primarily by
male-male interactions, but is either random or controlled by females. We are currently testing this hy pothesis
by forcing renests of groups containing cobreeder males. Thus, it is possible that, despite superficial support,
patterns of reproductive bias in neither sex of Acorn Woodpeckers is predictable by optimal skew theory.

19       Dodenhoff
* Individual ident if ic at ion and variation w it hin Dow ny Wo odpeck er acoust ic signals. DA NIELLE J. DODENHOFF,
Dept. Evol. Ecol. & Organ. Biol. , Ohio St ate Univ. , Columb us, OH.
         Downy Woodpeckers use 2 vocal (pik and ratt le call) and 1 nonvocal (drum) signals for long distance
communication during the breeding season. I recorded and analyzed these signals from color banded Downy
Woodp eckers d uri ng t he 1 99 8 - 20 00 breed ing seasons t o t est w het her al l 3 signals enco ded i ndi vidual
recognition cues, and which variable(s) were responsible for encoding identit y. I also compare signals to t est if
the signal s ch anged w it hin or b etw een br eedin g seasons. Result s indic ated signi fican t dif ferenc es bet w een
individuals, breeding status, and year. There w as insufficient variability betw een individuals to correctly
classify individuals by drum or pik calls, but rat tle calls were individually identif iable. Results from t he analysis
of drums cont radicted the prediction that long distance territorial signals encode individual identit y. Results
indicated variation between individuals was due to differences in average frequency properties of the rattle call.
There w ere significant dif ferences betw een breeding stages in the average call and note duration, and average
not e dur ation pat tern w it hin 1 y ear. Dif ferenc es w ere f oun d in the average f requ ency and p attern b etw een
consec ut ive year s f or t he rat tle cal l. I t est ed w het her o bser ved chan ges in t his call w oul d be p erceived as
dif ferences in in divid ual identi ty . Playback result s indicat ed a signific ant dif ference bet w een responses to
betw een year rattle calls vs. intersexual neighbors. Thus, observed changes within rattle calls betw een years
represented a change of the characteristics that encode individual identity.

20      Kellam
Effect of pair bond maintenance on mixed-species flocking benefits in Downy Woodpeckers during winter.
JAMES S. KELLAM, Dept. Biol. Sci. , Purdue Univ. , West Lafayett e, IN.
         A variety of bird species show v ariation in t he extent t o w hich individuals maintain pair bonds during
the w int er. It is hypo thesized t hat t his variat ion is a result of i ndivi duals facin g tr adeoff s betw een fit ness
benefit s and costs associated wit h pair bond maintenance (PBM). In Dow ny Woodpeckers, PBM may be costly
due to heightened intra-pair dominance interactions and reduced foraging eff iciency. How ever, since Downy
Woodpeckers occasionally join mixed-species flocks during w inter, any PBM-mediated decline in foraging
efficiency may be off set by the benefits of f locking. Using radio telemetry in 19 99 and 20 00, I calculated the
frequency at w hich each of 26 individual w oodpeckers was seen w ithin 40 m of their mate and other flock
members (if any). These association data were compared with time-budget analyses conducted on each bird. I
foun d t hat male w ood peck ers m ost frequ ent ly associat ing w it h mixed-speci es f loc ks c ont aini ng t heir mat e
spent the least amount of time in foraging activities, and had the highest fly ing rates. In contrast, male
w ood peck ers assoc iat ing more w it h similarl y sized m ixed-speci es f loc ks w it hou t the mate show ed hi gher rat es
of f oraging and low er flying rates. Combined, these results show t hat although male downy w oodpeckers do
appear to hav e inc reased forag ing effici ency w hil e associat ing w it h a mixed f loc k, the ben efit s disappear w hen
the female mate is also a flock member.

21      Stark
* Are eastern nearctic woodpecker drums species-specific? ROBERT D. STARK, Dept. Evol., Ecol. & Organ.
Biol., Ohio Stat e Univ., Columbus, OH.
  Drumming in w oodpeckers is a rapid repetit ive series of strikes w ith t heir bill on a resonant substrate, not
associated w ith f oraging or cavity excavation. Drumming has been implicated in territoriality, mate attraction,
loc alizat ion of ind ividu als and even as a si gnal of territ ory qual it y. Previ ous resear ch on d rum ming h as indic ated
that w estern nearct ic w oodpecker dr ums are synt opically species-specific , w ith the dru ms cadence (str ikes/s)
as the primary variable for species discrimination and recognition. How ever, current research comparing
east ern n earctic w ood peck ers indi cat es t hat spec ies recog nit ion by dru m t o be m ore c ont ent iou s. I rec ord ed
the responses of 9 eastern woodpecker species to playbacks of syntopic and allotopic woodpecker drums.
Response variab les w ere reduc ed using a pri ncipal com pon ent s analysis and co mpared u sing a Wilc ox on sign ed
rank test . Result s indic ated t hat spec ies w it h di ver gent dru m cadenc es w ere abl e to acc urat ely discriminat e
heterospecif ics versus c onspecif ics, as predic ted. How ever, eastern species w ith similar dru m variables,
notably Downy and Red-bellied woodpeckers, w ere equally responsive to heterospecifics versus conspecifics
w hile in syn top y. Red-naped and Yellow -bellied sapsuckers w ere found t o have similar behavi oral responses t o
allotopic heterospecific versus conspecific drums in reciprocal playbacks. These results demonstrate that drums
of w oodpeckers are not species-specific. The implications of these results are discussed.

22      Hosoi & Rothstein
The eff ect of repertoire size and song series length on f emale responsiveness to copulatory vocalizations by
male Brown-headed Cowbirds. S. AKI HOSOI* and STEPHEN I. ROTHSTEIN, Dept. Ecol., Evol. & Marine Biol.,
Univ . Calif orn ia, Sant a Barbara, CA.
         The importance of repertoire size and content to mate choice by female songbirds has been widely
debated in the literature. Brown-headed Cowbirds have a repertoire of 2 - 7 perched songs (PSs) used in
close-range courtship of females and in dominance interactions among males, and a single flight whistle (FW)
used primarily in long-range communication. M ales also perform a mean of 3.0 PSs and 1.9 FWs in rapid
sequence i mmediat ely pri or t o and dur ing cop ulat ion . V ariat ion in FWs and PS repertoir e size c oul d be u sed as
ind icat ors of male q uali ty, bot h du rin g co urt ship and cop ulat ion . Prev iou s playback st udi es hav e exam ined
female p ref erenc es by pres ent ing them w it h a singl e PS or FW, but w e also pres ent ed est radi ol i mplant ed
females w ith playbacks of a rapid series of 3 - 5 vocalizations, such as w ould be heard in a natural copulation
sequence. We assessed female responses to diff erent playback ty pes by measuring the durations of their
pre-copulatory lordosis displays. Females responded similarly to a series of three PSs of diff erent types and a
series of 3 PSs of the same type. We also found no consistent preference for local PS types over PSs from a
neighboring population. How ever, females gave longer responses to a series of 3 PSs and/or FWs than to
playback of a single PS or FW. Females appear to place greater importance on the duration of a song sequence
than o n t he size of the PS repert oir e pres ent ed in the sequenc e, t hou gh song typ e or p erf orm ance q uali ty m ay
also affect female c hoi ce.

23      O' Loghlen & Rothstein
Ecological eff ects on song learning in w ild Brow n-headed Cowb irds. ADRIAN L. O' LOGHLEN* , Dept. Psychol.,
Univ. Washington, Seattle, WA, and STEPHEN I. ROTHSTEIN, Dept. Ecol., Evol. & Marine Biol., Univ.
Calif orn ia, Sant a Barbara, CA.
          The timing and extent of early exposure to conspecific song can have critical influences on subsequent
vocal development in songbirds. In populations wit h comparatively short breeding seasons, such as northern or
high elevation ones, restricted access to local songs may delay development in a large proportion of yearlings.
Our previous studies described extreme examples of delayed development in high elevation populations of
Brow n-headed Co w bir ds in t he Sier ra Nev ada of Calif orn ia. In t he cu rrent st udy , w e assessed vocal
development in a low elevation population at a more northerly latitude than those in the Sierra. Because birds
w it h def ici ent develop ment, such as small reper toir es or non -local s ong typ es, may not sing read ily in n ature w e
recorded songs from captive yearling and adult males shortly after they w ere trapped at 2 sit es in New York
state. Recordings w ere obtained from all birds captured and thus represent an unbiased sampling of both age
classes. Yearling repertoires had significant deficits compared to adults. Furthermore, there w as a significant
correlation between the extent of t hese deficiencies and relative dates when yearlings hatched the previous
season. These results strongly suggest that juvenile access to local songs is also restrict ed in this population
and that delayed vocal development is w idespread in cowbirds. These findings indicate that reliance on field
recordings alone may underestimate adult-yearling diff erences in vocal competency because yearlings that sing
readi ly in n ature m ay not be rep resen tative of year lin gs in gen eral, a result that may also appl y t o ot her

24      Rothstein & O' Loghlen
The developments of male song production and of female recognition of songs follow parallel ontogenies in
Brown-headed Cowbirds. STEPHEN I. ROTHSTEIN,* Dept. Ecol. Evol. & Marine Biol., Univ. California, Santa
Barbara, CA, and ADRIAN O'LOGHLEN, Dept. Psychol., Univ. Washington, WA.
          Brow n-headed Co w bir d f lig ht w histle (FW) dialect s are often ch aracterized by vocal el ements and even
entir e voc alizat ions that have no c om parable elements in other d ialect s. The ex traor dinary diversity of FW
structures matches the range of sounds produced by songbirds in general, because of the large extent to w hich
FWs are learned. In contrast, the cow bird' s other song category, perched song (PS), has a unique basic
st ruc ture t hat is c onstant across the species' rang e. M ales reared i n aco ust ic i solation develop nor mal o r near
normal PSs so the basic template for this song category is innate, although males learn to make minor
modifications t o the detailed struct ure of PSs. These learned modifications result in local PS sharing but do not
alter the unique struct ure that characterizes all PSs. We tested w hether female development of recognition of
these song categories parallels the development of male song production. Females reared in acoustic isolation
respond to PSs but not t o FWs. In addition, w e tested the lordosis responses of f emales from New York w ith
foreign FWs and PSs from t he western half of Nort h America and with local FWs and PSs. Females w ere highly
resp onsive to lo cal FW s and to f orei gn an d lo cal PSs. But they w ere as unres pon siv e to t he f orei gn FW s as
they w ere to heterospecific cont rol songs from a range of ot her species, indicating that females recognize any
PS as a conspecific vocalization but must have direct experience with a FW type to respond to it.

25      Chilton
Songs of Townsend's and Hermit warblers in allopatric populations. GLEN CHILTON, Dept. Biol., St. Mary's
Coll. , Calgary , A B.
         I recorded the songs of male Townsend's and Hermit w arblers in allopatric populations betw een 199 9
and 2001 . Tow nsend's Warblers were recorded in Washington, Idaho, British Columbia, and Alberta, and
Hermit Warblers w ere recorded in Washington and Oregon. As wit h other w ood w arblers, individuals of both
species sang 2 song types. Songs were composed of buzzes, w histled notes, and more complex syllables. For
both species and both song ty pes, there was some variation among males w ithin t he population, but
population-specific f eatures are recognizable. In general, the diff erences in songs betw een populations
inc reased w it h t he di st ance s eparat ing them. The res ult s of this st udy can s erv e as a f ramew ork to an
examination of songs in zones w here these species hybridize. Wit h a sufficiently large lexicon of songs, it
should be possible to differentiate betw een Tow nsend's song types and Hermit song t ypes in hybrid zones in

26      Bard
Control of song learning and aggression to playback along a phenotypic t ransition zone betw een Tow nsend's
and Hermit warblers. SUSANNE C. BARD* , Dept. Psychol., Univ. Washington, Seattle, WA.
        Tow nsend' s and Herm it w arbl ers int erbr eed in 3 h yb rid zones in W ashingt on an d Oregon . It has been
found that Tow nsend' s warblers are aggressively dominant to Hermit w arblers, and their off spring are
intermediate. Songs w ere recorded and analyzed from a north-south t ransect along the w estern side of t he
Cascade Mountain Range, representing a plumage transition from Townsend's to Hermit phenotypes.
Individuals sang local songs, regardless of plumage. This suggests that songs in these species are learned from
neighbors, and are not constrained by genotype. Playbacks were conducted along the transect, using
Tow nsend' s st imuli record ed no rt h of the in tergr ade zon e, Hermit st imuli record ed sout h of the in tergr ade zon e,
and local song from t he neighborhood the subject occupied. I predicted diff erential aggression along the
transect to t he 3 diff erent playback conditions, w ith response being greatest t o local playback. The results of
this study have important implications for our understanding of species recognition and aggressive interactions
betw een species.

27      Price & Lanyon
The ev olution of st ereot ypy in birdsong : Pat terns of song evolution in t he or opendolas. J . J ORDAN PRICE*
and SCOTT M. LANYON, Bell Mus., Univ. Min nesota, St . Paul, M N.
          Sexual selection is presumed to cause male traits, including oscine song diversity, t o become more
extravagant and complex. Signal detection theory, how ever, predicts that female choosiness favors the
evolut ion of male s ign als t hat are more s tereot yp ed and temporal ly redu ndan t. W e explor ed t hese alternat ives
by recon stru ct ing song ev olut ion in t he oropendolas (Psarocolius), a group wit h extreme examples of
pol yg yn ous breed ing and sexual size dim orp hism. Male or open dol as per form elabo rat e cou rt ship di splays that
include species-specif ic songs. We measured a variety of aco usti c and behavioral f eatures of these display s,
including song stereotypy, and mapped these characters onto a robust phylogeny w e constructed using
mitochondrial DNA sequence data. Reconstructions of song evolution on this tree reveal that many features are
surprisingly evolutionarily conservative. Although some acoustic features appear to have become more
extravagant wit h increased intensity of sexual selection, song diversity appears to have decreased. The high
stereotypy and redundancy of oropendola songs apparently evolved from much more versatile modes of singing
in less polygynous ancestors.
28      Gammon, Baker & Betancourt
How do juvenile Black-capped Chickadees behave in response to playback of conspecif ic vocalization s? DAVID
E. GAMMON* , MYRON C. BAKER, and LISETTE BETANCOURT, Dept. Biol . Colo. St . Uni v. , Ft. Collins, CO.
         It has long been know n that songbirds must learn their songs, but scientists are only beginning to
und erst and t he mechan ism s of song lear nin g. The m ost com preh ensive m odel of song lear nin g pr opo ses t hat
developing songbirds possess a gen eti cally inher it ed " auditory templ ate" w hic h allow s yo ung bir ds t o rec ogn ize
and preferentially learn conspecific song. Employing methods that had been successfully used on juvenile
White-crow ned Sparrow s, w e played conspecific and heterospecific (House Finch) vocalizations to juv enile
Black-capped Chickadees in the lab and then monitored their behavior to see if they w ould respond
preferentially to conspecific vocalizations. Contrary t o the case of t he White-crowned Sparrow, birds did not
demonstrate a predictable change in behavior in response to either stimulus. In addition, birds did not give
more vocalizations or show heightened activity levels in response to conspecific playback as compared to
play back of het eros peci fic v ocalizat ion s. Thi s lac k of behav ior al respon se raises t he qu est ion of w het her
juvenile Black-capped Chickadees actually recognize conspecific vocalizations as required by the prevailing
model of song-learning. Alt hough our study cannot answ er this question, w e conclude that detecting song
recognition in juvenile Black-capped Chickadees w ould require diff erent methods (such as measuring the firing
of neurons in response to song playback) than the methods w e used.

29      Tello
* Mol ecular phyl ogenetics o f t he genus Cercomacra (Aves: Thamnophilidae). JOSE G. TELLO, Dept. Biol. Sci.,
Univ. Illino is at Chic ago an d Div. Bird s, Field M us. , Ch icag o, IL.
         The Neotrop ical genus Cercomacra con tains 12 spec ies and 2 0 r ecog nized subspec ies of mid-sized
insectivorous antbirds found mainly in forest understory and w oodlands in tropical lowlands, especially in
Amazonia b ut w it h a series of spec ies (t he C. nigricans group) found only around its periphery follow ing a
circum-Amazonian distribution. Cercomacra nigricans reaches Central America, and C. tyrannina reaches s.
Mexi co. T his dist ribut ion pat tern m akes Cercomacra a model f or s tudy of the pat terns and p roc esses
underlying diversification in Neotropical forests. To date, I have sequenced 32 indiv iduals of Cercomacra
(representing 9 species and 15 taxa) and appropriate outgroups. I sequenced 713 bps of mtDNA, containing
fragments of the cyt b and ND2 genes. Levels of phylogenetic informativeness for the mitochondrial data are
hig h; 35 .3 % var iabl e sit es, w it h 7 9. 4% of those var iabl e sit es being p arsimony inf orm ative. W eigh ted
parsimony analysis resulted in 2 most-parsimonious trees. Cercomacra appears to be monophyletics and the
boot strap c onsensus t ree clustered species in 2 major clades: (1 ) tyrannina, laet a, serva, nigrescens; and (2)
cinerascens, melanaria, carbonaria, nigricans, manu. The positions of C. brasiliana, C. ferdinandi, and C.
parkeri in relation to the other major groups require tissue samples that are not currently available.

30      Paton
* Resolving the systematics of Charadriiformes using complete mitochondrial genomes. TARA A. PATON, Royal
Ontario Mus. and Dept. Zool., Univ. Toront o, Tor ont o, ON.
          Members of the avian order Charadriiformes exhibit a wide variety of morphological forms that have
been the source of much debate in systematics. This problem has been exacerbated by many enigmatic taxa
w ith uncertain phylogenetic placement. M oreover, studies using morphological and biochemical evidence show
litt le congruence concerning the deep-branch phylogenetic relationships w ithin t his clade of birds. To clarify t he
famil y level relat ion ships w it hin the Char adri if orm es w e have been u sing app rox imately 29 00 bp o f the nu clear
RAG-1 gene (to be presented at t his meeting). How ever, to provide an independent estimate and bett er support
for recently diverged cl ades, 12 prot ein-coding and 2 ribosom al RNA genes f rom mi toc hondri al DNA
(approxim ately 1 1, 00 0 bp) w ere sequenced for represent ativ es from 11 charadriif orm f amilies plus out groups.
Phyl ogen etic t rees w ere construc ted using neig hbo ur- joi nin g w it h lo g-determinan t distances as w ell as
maximum likelihood. The relationships among the shorebird families examined here agree w ith t hose found
using nuc lear RAG-1 sequenc es. Cont rary t o previou s stud ies, t he clade consist ing of the sheath bills,
oystercatchers and allies are the basal lineage of Charadriiformes. The placements of some enigmatic taxa are
reso lved as w ell. As t he rem aini ng Ch aradr iif orm famil ies are samp led, these r elat ion ships c an be c onf irm ed
and c ompared to hy pot heses gener ated f rom nuc lear sequen ces. Combin ed sequenc e dat a from bot h gen omes
will undoubtedly generate a strongly supported and comprehensive phylogeny for shorebird families.

31       Witt
* A molecular phylogeny of the jacamars and puffbirds: Insights into the origins of Neotropical bird diversity.
CHRISTOPHER C. WITT, Dept . Biol. Sci. an d M us. Nat . Sc i. , Lo uisiana St . Univ ., Bato n Roug e, LA .
          The jacamars (Galubulidae) and puffbirds (Bucconidae) are moderately diverse families that are entirely
restricted to Neotropical forests. M itochondrial and nuclear DNA sequences w ere used to estimate a phylogeny
of nearl y all species of jacam ars an d pu ffbir ds using maximum likel iho od. Thi s phylogen y includ es sev eral
w ell-supported results that are unexpected on the basis of current classification, including non-monophyly f or
the gen era Galbula, Bucco, and Notharchus. A new generic classification is proposed that is concordant w ith
the mo lecular phy logeny. The phylo geny suggest s that the Galbulif ormes are an old radiatio n, bu t t hat bur sts
of speciation have occurred as recently as the Pleistocene. The timing of div ersification in these 2 families is
examined in detail and compared to previously studied groups and to a model of random lineage formation and
ext inc tion . Rat es of spec iat ion , ec olo gic al ch ange, and d ist rib ut ion chan ge are c ompared among t emperat e,
tropical, and island groups, using molecular divergence as an estimate of t ime. The results are discussed in
lig ht of the hy pot heses that age and st abil it y ac count for h igh spec ies diversity in t rop ical regi ons, an d t hat
diversification occurs more rapidly in the tropics.

32       Moyle
* Phylogenetics of barbet s (A ves: Pici formes): Pant ropical biogeography and life hist ory c onvergence. ROBERT
G. M OYLE, Mus. Nat . Sc i. , Lo uisiana St ate Univ. , Bat on Rug e, LA .
           I reconstructed phylogenetic relationships of all currently recognized genera of barbets and toucans.
Analysis of mitochondrial (cyt-b) and nuclear (" -f ibr ino gen i nt ron 7) dat a pro vide st ron g suppo rt for d eep no des
in t he ph ylogen y. Cont rary t o earl ier m orp hol ogi cal an d molec ular evidenc e, t hese data ind icat e that the 3
continental radiations (Africa, Asia, and New World) are reciprocally monophyletic and the Asian clade is basal.
Biogeographically, this pattern is congruent w ith that f ound in other pantropical groups of birds. However,
vicariance or dispersal explanations are not readily apparent. M onophyly of each continental assemblage
ind icat es mu lt ipl e ind epend ent ori gin s of lif e history trait s. Of part icu lar i nt erest are t he similari ties betw een
unrelated Asian (Calorhamphus) and African (Gymnobucco) genera.

33       Yuri
* Evolution of plumage color change and migration in Passerina buntings. TAMAKI YURI, Mus. Zool. and Dept.
Biol., Univ . M ichigan, A nn Ar bor, M I.
           Plumage maturation patterns of Passerina bun ting s hav e draw n much at tent ion becau se second -year
males of Passerina exhibit " delayed plumage maturation," and many adaptive hypotheses for these patt erns
have been suggested. The extent of f irst-year molts that generate their second-year plumage color variations
varies by environmental factors such as age (i.e. hatch date), physical conditions, and time constraint during
the mo lt, and genetic c omponent s underlyi ng t hese variations appear t o be small. For f irst -year migrant s,
set tlin g in the w int erin g gr oun ds shou ld b e a dif ficu lt task, and t hey may experien ce gr eater constrain ts t hat
limit t he extent of molt t han young birds of sedentary species. In this study, I described patt erns of molt and
plumage color change in Passerina bunt ings using captiv e birds and museum sp ecimens, and co mpared th ese
patterns w ith t he occurrence of migratory behavior based on a phylogenetic tree reconstruct ed using both
mitochon dri al and nuc lear DNA sequences . Phy log enet ic an alyses indi cat ed t hat Passerina, Guiraca,
Cyanocomp sa, and Spiza form a w ell-supported monophyletic group, and the possession of migratory habit is
the ancestral state of t he clade. Second-year plumage color was more variable in migratory species than in
non-migratory species, w hich is consistent w ith t he molt constraint hypot hesis. Close examinations of molt
patterns, however, indicated that the forms of delayed plumage maturation are different betw een migratory
Passerina species, w hich may suggest independent evolutions of migration, and associated modifications of
molt and plumage color patterns.

34      Burbidge
* Genetic analyses contribute t o the management and recovery of Brown Kiw i (Apteryx spp.). M ARYANN L.
BURBIDGE, Univ. Toront o and Royal Ontario M us, Tor ont o, ON.
         The kiwi f amily (Apterygidae) is composed of at least 5 species, all of w hich are threatened according
to IUCN criteria. All species require immediate conservation management effort s. In response to t his need, the
Kiwi Recovery Programme (KRP) w as developed to integrate education, advocacy, research and management
programmes to save New Zealand's unoff icial national bird from extinct ion. Adv ances have been made in all
areas of the KRP, and in particular, field research eff orts by t he Department of Conservation and other groups
have lead to major contributions regarding kiwi biology, ecology, behaviour and predator cont rol. Utilization of
this informat ion for t he most effectiv e management of kiw i requires a greater understanding of population
dynamics and struct ure. The most efficient method t o acquire this know ledge is via genetic analyses. Genetic
studies (Baker et al. 1995, Proc. Natl. Acad. Sci. 92: 8254-8258; Burbidge 2000, AOU lect ure) hav e iden tif ied
3 mo rpholo gically c rypt ic species w ith in t he Brow n Kiw i, f ormally classifi ed as Apt eryx australis. Other prior
genetic w ork (Burbidge 200 0) demonstrated that conservation management units should be defined minimally
at the regional population level. Mitochondrial DNA control region sequence data are being used to determine
w hether management is warranted at the local population level as well. The preliminary results of this study
emphasize the distinct ness of the regional populations on the North Island and indicate that these populations
are nearly as structured as those on the South Island (Baker et al. 1995; Burbidge et al. 1999, AOU poster).
The r esults also suggest that management may be req uir ed f or s ome loc al po pul ation s. The g enet ic s tudy is an
integral part of the KRP and is imperative for designing the most appropriate management and recovery plan for
Brow n Kiw i species.

35      Aleixo
* Molecular systematics of Xiphorhynchus w oodcreepers (Dendrocolaptidae): Implications for species limits and
Neotropical historical ecology. ALEXANDRE ALEIXO, Dept . Biol. Sci., and M us. Nat . Sc i. , Lo uisiana St ate
Univ. , Bat on Rou ge, LA.
          I reconst ruct ed phylog enies of t he genus Xiphorhynchus (Dend roc olap tidae) that inc lud ed all it s species
and m any subspec ies, to evaluat e spec ies limit s and to in vestigat e the ro le pl ayed by the A mazonian vár zea
(floodplain forest) - terra-firme (upland forest) ecotone in its diversification. Phylogenies were based on 2,430
bp of t he mtDNA genes ND2, ND3, and cyt ochrome-b. All phylogenies supported the monophyly of
Xiphorhynchus to t he exclusion of t he sibling species pair X. picus (Straight-billed Woodcreeper) and X.
necopinus (Zim mer’ s Wo odc reeper ). Conf irm ing find ing s of a prev iou s mo lecu lar study , stron g suppo rt w as
foun d t o in clu de Lepidocolaptes fuscus (Lesser Woodcreeper) in Xiphorhynchus. Uncorrected sequence
divergence among some subspecies of X. gut tatus (Buff-throated Woodcreeper), X. ocellatus (Ocel lat ed
Woodcreeper), and X. spixii (Spix’s Woodcreeper) reached or exceeded those found betw een closely related,
undisputed biological species of Xiphorhynchus. High levels of sequence d if ferent iat ion and p hy log enies
suppor ting the par aphy ly of som e Xiphorhynchus spec ies indi cat e that the f oll ow ing taxa shou ld b e rank ed as
species: X. gut tatoid es (Lafresnaye’s Woodcreeper), X. chunchot ambo (Tshudi’s Woodcreeper), X. napensis
(Napo Woodcreeper), and X. elegans (Elegant Woodcreeper). All Xiphorhynchus species restricted to
terra-f irme f orest f ormed a w ell-support ed monophy letic g roup, w hereas species restrict ed to v árzea forest
w ere either basal to a clade containing species found in a w ide variety of habitats, or belonged to a distinct
lineage likely to be regarded as a separate genus. These findings falsify an hypothetical sister relationship
betw een várzea and terra-firme species, as expected if the várzea - terra-firme ecotone played a decisive role in
population differentiation and subsequent speciation w ithin Xiphorhynchus.
36      Mobley
* Resolv ing deep d ivergen ces am ong the gen era of Kingbi rds and t heir alli es: New find ing s f rom com bin ed
analyses of CytB and ND2. JASON A. MOBLEY, Mus. Vert. Zo ol. , Un iv. Calif orn ia, Berkeley , CA .
         Previous att empts t o reconst ruct phylo genetic r elationshi ps among t he Kingbirds ( Tyrannus) and their
allies (Griseotyrannus, Empidonomus, Megarynchus, Pitangus, Philohydor, Myiody nast es, Myiozet etes,
Phelpsia, Cono pias, Legatus, and Tyrannopsis) based on an 8 60 bp p ort ion of the mit ochon dri al cy tochro me b
gene ac hiev ed on ly lim it ed success. Mono phy ly of the rec ogn ized g enera an d rel ation ships am ong spec ies
w it hin most gener a w ere w ell supp ort ed, w hil e basal relat ion ships am ong gener a w ere eit her u nres olv ed or had
low boo tst rap s upp ort . New phy log enet ic h yp ot heses based on an 80 0 b p po rt ion of the mit ochon dri al
protein-coding NADH dehydrogenase 2 gene (ND2) and trees derived from a combined analysis of cyt B and
ND2, using 2 species of Myiarchus to root the tree, w ill be presented. Phylogenetic hypotheses based on the
combined dat a set show bet ter resolut ion and m uch hi gher boo tst rap s upp ort for d eeper d ivergen ces t han
ind epend ent analy ses of the t w o markers. The i ngr oup topo log y is not rob ust to vario us w eigh ting sch emes
and tree building methods suggesting t he need for additional data to achieve a w ell supported phylogeny for t he
ty rannine fl ycat chers. Comp eting h ypot heses wi ll be presented and disc ussed in relatio n to Lanyon' s (19 84 ,
Am. Mus. Novit ates, 27 97) phylogenetic hypot hesis based on morphology of the syrinx, skull, and nesting
habits. Implications for t he major transitions in nesting behavior w ithin t his group from character optimization
on the combined molecular phylogeny will also be discussed.

37      Given
* Popul ation st ruc ture an d speciat ion in t he Larus cirrocephalus species complex. ANDREW D. GIVEN, Dept.
Zool., Univ. Toront o, and Cent re Biodiv. & Conserv . Biol. , Royal Ont ario Mu s., Toront o ON.
         The gulls (Tribe Larini) are among the most widespread and w ell-studied bird groups. The
approximately 50 species occur in a w ide range of habitat ty pes from tropical regions to arctic and subantarctic
regions. However, attempt s to reconstruct phylogenetic relationships w ithin t he gulls have proved difficult
because of their conservative evolution. In this study w e investigate the relationships of gulls belonging to the
southern hemisphere Larus cirrocephalus com plex . This monop hly let ic c lade c onsist s of 4 r ecog nized spec ies
and 7 subspec ies f rom Sout h A merica, southern Afric a, A ust rali a, New Caledo nia an d New Zealand . Ten
polymorphic microsatellite loci were used for the 600 gulls sampled from 26 populations. The main aims of this
stud y w ere to (1 ) determin e w hether t he current classifi cation of t his group of g ulls is approp riate and (2 ) to
investigate population divergence and levels of gene flow among the Australian and New Zealand populations.
Results show a gradual westw ard cline from New Zealand through eastern and western Austr alia to South
Af rica, resulting in fixed allele differences in some loci betw een the Silver Gull populations and Hartlaub's gull
(Larus hartlaubi). The classification of Hartlaubi' s Gull as a separate species is therefore supported. The
Australian and New Zealand populations of Silver Gulls are well diff erentiated genetically, consistent w ith
observed morphometric dif ferences. The Black-billed Gull (Larus bulleri) is the sister group to the New Zealand
Silv er Gul ls w hic h means t hat Larus nov aehol land iae as currently defined is paraphyletic. The Aust ralian and
New Zealand Silver Gulls should t herefore be classif ied as separate phylogenet ic species (Larus nov aehol land iae
and Larus scopulinus, respectively).

38      Paton, Barrowclough, Groth & Baker
Phylogenetic inf erence of Charadriif ormes relatio nships usin g nuclear RAG-1 sequence. T ARA A . PATON,
Royal Ont ario M us., To ront o, ON; GEORGE F. BA RROWCLOUGH, JEFF G. GROT H, Am. Mus. Nat . Hist ., New
York, NY; and ALLAN J. BAKER* , Royal Ontario Mus.
         The use of nuclear genes to determine higher-level phylogenetic relationships in birds has been gaining
popularity in recent years, due to t he relatively slow er rate of evolution and more homogeneous distribution of
substitut ions compared to mit ochondrial genes. Additionally, nuclear sequences collected in combination wit h
mitochon dri al sequenc es pr ov ide i ndep enden t est imates of phy log enet ic r elat ion ships as they evolve in separat e
cellular environments. Recent studies have demonstrated the nuclear RAG-1 gene to be phylogenetically useful
at the ordinal level in birds and more recently, betw een and wit hin avian families. Therefore, we tested the
ut ili ty of RAG-1 to det ermine p hy log enet ic r elat ion ships am ong the f amil ies and gen era of the avian o rder
Charadriiformes, as morphological and other biochemical evidence generate conflicting hypotheses. To resolve
these systemat ic u ncert aint ies and t o pr ov ide an ind epend ent chec k of trees g enerat ed f rom mitochon dri al dat a,
41 t axa were sequenced for approximately 29 00 bp of the RAG-1 gene. The taxa sampled represent 1 6
charadriiform families as w ell as outgroups. The sequences were analyzed under maximum likelihood,
max imum pars imony and n eigh bou r-join ing w it h lo g-determinan t distances . The t opo log ies of the co nst ruc ted
trees were very similar among these tree-building methods, and deep-branch nodes were well supported in
boot strap t ests. Furt her, t he RAG-1 top ology is congr uent w ith the t opology of a mt DNA t ree. Because
mitochondrial sequences provide improved support for t he positions of recently diverged clades of this group,
this suggest s t hat a full y r esolved tree can soon b e con st ruc ted by a joi nt use of bot h nu clear and m it ochon dri al
data, which will be valuable to comparative studies of shorebird biology.

39      Driskell
A consideration of cooperative breeding in the honeyeaters. AMY C. DRISKELL, Field Mus. and Univ. Chicago,
Chicago, IL.
          A robust phylogeny of the honeyeaters (Meliphagidae) would seem to provide an ideal opportunity for a
rig oro us c omparat ive st udy of the co rrelat es of coo perat ive breed ing (CB). Melip hagi ds ar e gener ally cit ed as
one o f the pr ime exam ples of a coo perat ively breed ing passerin e famil y. Ther e are a nu mber of det ailed st udi es
on a few honeyeaters, and cooperative behavior has been observed in a number of other meliphagids.
However, it is unclear just how many meliphagids breed cooperatively. It is equally unclear how many
honeyeater species do not breed cooperatively as a large proportion have not been carefully studied nor live in
densely populated areas. For these reasons and a number of others, a comparative analysis of the evolution of
CB in t he f amil y M elip hagi dae w oul d be p remature. How ever , a n umber o f con clu sions c an be d raw n. First , an
examination of this trait on the phylogeny reveals that CB is not as widespread in the family as widely thought.
Second, CB is not the basal condition for the Meliphagidae and this behavior has evolved independently a
number of tim es. In addit ion, t he miners (genus Manorina), oft en called obligate cooperative breeders, are
sister t o th e New Guinean genus Melid ect es. No breeding studies of Melid ect es have been published, b ut t hese
birds would appear to be excellent candidates for study. Also, the CB exhibited by Melithreptus and
Lichenostomus appears to be an instance of convergent evolution. A search for ecological or demographic
fact ors in c ommon among the species in t hese 2 gener a w oul d be v aluab le.

40      McDonald
Cow birds and Kent ucky Warblers: Fourteen years of no signif icant ef fect . M . V ICTORIA M cDONALD,
Smithsonian Conserv. & Res. Center, Front Royal, VA.
           Kentucky Warblers (KEWA) have been studied intensively for 23 yr at a relatively undisturbed second
grow th hardw ood research site in n.-central Virginia (the Conservation and Research Center [CRC]). Concurrent
w it h t he KEWA st udy , d ata on p resen ce abs ence an d rel ative abun danc e of ot her av ian speci es hav e also been
taken since 1988. Noted especially have been Brown-headed Cowbirds (BHCO), whose occurrence, relative
abundance, dist ribut ion, and p arasitism o n KEWAs and ot her species has been observed thr oughout the past 14
breeding seasons. KEWA numbers at CRC have been declining slowly since 197 9; also, their population
density centers have shifted gradually during these 23 y r. To analyze reasons for the population decline and
shift, I examined several hypotheses that have been previously tested and found significant in low ering
reproductive output of some migratory passerine species, of w hich the KEWA is ty pical. Predation, severe
w eather, and nest desertion (potentially caused by an increase in human observers) do not correlate with t he
overall long term decline. The habitats of formerly occupied KEWA t erritories wit hin the confines of the study
areas have c hang ed sligh tly, b ut measu rabl y; thus, succession al and /o r deer ind uced un ders tory depl etion may
be a factor. Most not able, are that w hile BHCO abundance and dispersion throughout CRC have apparently
increased since 1988 (when quantification began), their rate of parasitism on KEWAs has decreased.
Parasitism on other species, however, has apparently increased during the same time.
41      Hau, W ikelski, Gw inner & Gwinner
Are Darw in' s finch es opportun istic b reeders? MICHAELA HAU* , M ARTIN WIKELSKI, Dept. Ecol. & Evol. Biol.
Princeton Uni v. , Princet on, NJ , HELGA GWINNER and EBERHARD GWINNER, Research Centre for Ornithology,
Andechs, Germany.
         Reproduction of Darwin' s finches on their equatorial islands is closely associated with rainfall. Rainfall
on Gal ápago s is ver y v ariab le and bir ds have been f oun d t o br eed at ver y dif ferent times of the year. We t est ed
w hether Darw in’s f inches are strict opport unistic breeders even in the physiological sense by examining
seasonal changes in gonad size and reproduct ive horm ones of Small Grou nd Finches (Geospiza fuligino sa).
Small Gro und Finches c omplet ely regr essed their gon ads in t he no n-breedi ng season. In t hat resp ect they dif fer
from st ric t opp ort uni st ic b reeder s w hic h maint ain en larg ed go nads year -ro und . How ever , Sm all Gr oun d Finches
started to grow their gonads at dif ferent times in 2 years wit h diff ering rainfall patterns. Moreover, t iming of
gonadal growt h in 3 different populations of Small Ground Finches matched the rainfall pattern at each site.
The lack of a physiological preparation for t he breeding season and the adjustment in gonadal development to
rainfall patt erns suggest a striking flexibility in the regulation of breeding in these birds. Other species of
Darw in’ s f inc hes show ed similar p atterns . W e experim ent ally test ed in capt ive Small Gro und Finches w het her
an in crease in food abun danc e trig gers gon adal g row th. Cont rary t o ou r ex pect ation s, bot h suppl ementally fed
and control birds grew t heir gonads. In both groups, gonadal growt h coincided wit h a precocious start of t he
rainy season in that year, which may also have stimulated the birds in their outdoor aviaries. However, a direct
stimulation of r eproductive activity by rain appears unlikely because in free-living birds levels of Luteinizing
hormone were elevated 2 wk before the start of the rainy season, before any gonad development had occurred.
Future ex peri ments w ill aim at eluc idat ing the t yp es of enviro nment al cu es t hat Small Gro und Finches use t o
time breeding and how much their physiological organization differs from strict opportunistic breeders.

42       Wilson & Holberton
The endocrine basis for trade-offs betw een immediate survival and reproductive success in arctic- and
temperat e-breeding Yellow Warblers. C. M. WILSON, Dept. Biol., Univ. Mississippi, University, MS; and R. L.
HOLBERTON,* Dept. Biol., Univ. Maine, Orono, ME.
          While temperate breeding birds can renest or produce more than one successful clutch each year,
arctic-breeders are unlikely to raise more than one clutch annually due to shorter breeding seasons and
unp redi ct able w eather. Despi te these c onstrain ts, many bird s successf ull y breed i n t he A rctic, suggest ing that
trade-offs betw een individual risk and reproductive success diff er between arctic- and temperate breeders. The
mobilizat ion of higher plasma levels of cort icost erone helps indiv iduals surv ive enviro nment al pertur bation s,
particularly for t hose living in extreme and stochastic environments. However, high concentrations of
cort ico st eron e are in compatibl e w it h rep rod uction . W e have asked how hig h lat it ude-breed ing bir ds w it h li mited
opport unit ies to br eed cope w ith the t rade-off betw een the personal benefit s of t he cort icost erone stress
response needed in such an environment, and the possible cost in reproduction that such a response may incur.
In 1999 and 2000, w e found that arctic-breeding male Yellow Warblers, who must contribute significantly to
parental care, reduce acute corticosterone secretion during times of highest parental expenditure.
Temperate-breeding males, w ho also contribute significantly t o parental care, show no such reduction. The
cost of this reduced response to the individual, and the mechanism by which it occurs, are under investigation.
This phenomenon may represent a mechanism by which tw o different strategies expressed wit hin the same
species might maximize reprod uct ive success und er disparate ecologi cal const raints.

43      Kosciuch, Kasner & Arnold
Reproduc tiv e success and colony grow th o f Clif f Sw allow s in east-cent ral Texas. KA RL L. KOSCIUCH,
ANDREW C. KASNER and KEITH A. ARNOLD, Dept. Wildl. & Fish. Sci., Texas A&M Univ., College Station, TX.
       Most st udi es describi ng r epro duc tive bio log y of an avian speci es pr ov ide some measu re of annu al
reproductive success (ARS) most frequently reported in terms of nests or eggs. These measurements do not
indicate the success of the average female in the population. We report conventional measures of reproductive
success along with ARS (b), broods successfully reared/female and ARS (k), young successfully reared/female.
We calculated ARS for 4 culvert-dw elling Cliff Swallow colonies in e.-central Texas from 19 82 t o 198 5 (n =
180 5 nests) and analyzed differences in number of f ledglings/nest to estimate variation in ARS among years
and among culverts. We attr ibuted variation in ARS to predation and abiotic f actors. We also tested the
conspecific performance-based attraction hypothesis (Danchin et al. 1999, Ecology 79: 2415-2428) to
determine if colonies w ith t he highest ARS increased in size. Results from a simulation model indicate the
importance of predation and mortality in colony grow th and reveal the sensitivity of colony size to minimum
requ ired ARS for return . The use of ARS (b) and A RS (k) along w it h an i nv est igat ion of the A RS thresh old may
lend new inf ormat ion int o colon y cho ice by Clif f Sw allow s.

44      Stutchbury
Is reproduction as "costly" in tropical birds as compared with temperate zone birds? BRIDGET J.M.
STUTCHBURY, Dept. Biol., York Univ ., Toront o, ON.
        Why do birds lay so few eggs? Because there is a high cost of reproduction; large reproductive effort
now redu ces f ut ure s urv ival and /o r repro duc tion . This con vention al w isdom is based entirel y on t emperat e
zone birds that t ypically lay large clutches. Do tropical birds, w hich oft en lay only 2 - 3 eggs, also face a high
cost of reproduction? Brood manipulation studies, the main tool for studying the cost of reproduction, have
been p erf orm ed f or o nly a hand ful o f trop ical bir ds. In m ost cases pair s can not raise enlar ged b roo ds. In t he 2
species where parental effort w as experimentally increased, no eff ect on future survival was found. Together,
these few st udies suggest that clut ch size in tropical birds may by limited by immediate costs (lack of f ood)
rather than high future costs t o survival.

45      Lyons, W alters, Collazo & Saracco
Nest succ ess of fo rest birds in bot to mland hardw ood fo rest : ef fec ts of levee w idt h. JA MES E. LYONS* ,
JEFFREY R. WALTERS, Dept. Biol., Virginia Tech, Blacksburg, VA, JAIME A. COLLAZO, NCCFWRU, North
Carolina State Univ ., Raleigh, NC, and JAMES F. SARACCO, Dept. Zool., North Carolina State Univ.
          Bot tomland hard w ood s are natural ly pat chy landscapes that con tain m any dif ferent typ es of edges (i. e.
abrupt-gradual). We studied nest survival near habitat edges and w ithin core areas in bottomland forests of the
Roanok e River to det ermine i f nest pred ation and p arasitism w ere gr eater near edges. Nest ing suc cess w as
monitored in 2 f orest w idth classes: narrow bands of levee forest t hat w ere dominated by 2 edge types, and
w ide, continuous levee forest stands that have edges but most forest is relatively far from edge. Vegetation
stru ct ure and compo siti on w ere measured at nest sit es and random sit es to examin e habitat selection. Nest
success of Acadian Flyc atchers and Prot honot ary Warblers w as similar in narrow and w ide levees; nest suc cess
of Northern Cardinals was greater in narrow levees. Wide, continuous levees had a greater canopy height,
denser shrub/midstory layer, and few er large trees than narrow levees. Significant differences in vegetation
st ruc ture an d co mposition w ere ob serv ed at narr ow vs. w ide l evee nest sit es. These dif ferenc es w ere similar
to pat terns foun d at rand om sit es, how ever , an d w e con clu de t hat bir ds w ere no t selecting dif ferent habi tat
features depending on levee widt h. When nest sites were compared to random sites, variation in vegetation
struct ure and composition indicated that nest placement w as not random; each species selected diff erent
microhab it at features . Ou r results indic ate that edge ef fect s are not con sistent among ed ge t yp es, and t hat
amount of c ont rast at edges may int eract w ith landscape cont ext t o alter ecolo gical proc esses such as nest

46      Blums & Clark
Trade-o ff s ov er t iming of b reedin g and r eprod uct ive ef fo rt in bi rds. PETER BLUMS, Gaylord Mem. Lab., Univ.
Missouri, Puxico, MO; and ROBERT G. CLARK* , Can. Wildl. Serv. , Saskatoon, SK.
        We tested hypotheses about timing of breeding and reproductive effort in birds, by analyzing long-term
data for individually-marked females in 3 species of Latvian ducks. The number of recruits (per brood)
produced by breeding females declined with advancing hatch date in all species aft er controlling other factors
w ith path analysi s. Duck ling surv ival and recrui tm ent declin ed w ith advancing h atch d ate in bot h 2 speci es,
ind icat ing that advant ages ass ociat ed w it h earl y hatching ext ended beyond the pr e-f ledg ing peri od. Ind ividu al
duckling survival was moderate early in the breeding season, greatest in mid-season and low est thereafter; this
stabilizing selection was countered by directional selection favoring recruitment of early-hatching females.
Dependin g on spec ies, fledg ing or r ecruit ment ind ices w ere neg atively relat ed t o ann ual estimates of tot al
duckling production during brood-rearing, consistent wit h an hypothesis that duckling survival is inversely
density-dependent. Fecundity-independent variables (e.g., hatching date, w eather, indices of duckling
production and habitat quality) generally had 2 to 1 0 t imes greater influence on recruitment t han did
fecun dit y-depen dent var iabl es (duckli ng m ass and br ood size), sugg est ing a crit ical rol e for ex ternal
environmental factors versus individual female-specific traits in the recruitment process. Recruitment patt erns
revealed by fitness surface response analysis were similar to those obtained for several North American duck

47       Schoech & Bowman
Nutrition and the timing of reproduction in Florida Scrub-Jays. STEPHAN J. SCHOECH* , Dept. Biol., Univ.
Memphis, M emphis, TN; and REED BOWMAN, Archbo ld Biol. Stn. , Lak e Placid , FL.
          Florida Scrub-Jays in their natural habitat at Archbold Biological Station show considerable
bet w een-y ear v ariat ion in t he t iming of repr odu ct ion . In co nt rast , j ays in a n earby suburb an env iro nment ini tiat e
clut ches earlier and w ith less betw een-year variation t han jays at A rchbol d. If , as w e hypot hesize, these
dif ferences are att ribut able to t he availability of ant hropog enic fo ods in t he suburbs and result in access to
limiting nut rients or an overall increase in body condition, then jays should exhibit betw een-site differences in
certain physiological variables. We tested this by measuring plasma levels of protein, calcium, luteinizing
hormone, and estradiol, as well as the total body fat of female breeders. Between-site comparisons were made
relative to calendar and clutch initiation dates. Relative to calender date, suburban females had higher plasma
levels of pro tein, lut eini zing hor mone, and estradi ol t han f emales at Archbo ld. How ever , w hen w e con trol led
for clut ch initiation date, no diff erences existed. Most phy siologic measures increased as clutch initiation
neared but body lipids decreased. Plasma protein levels during the earliest sampling period relative to laying,
w ell before bir ds should b e mobilizing p rot eins for vit ellogenesis, w ere signifi cantly higher f or subur ban birds,
suggesting an association betw een anthropogenic foods, protein, and time of laying. In a subsequent year we
test ed if the t iming of breed ing of jays in nat ural habi tats w as limited by pro tein av ailab ili ty by prov idi ng t hem
w ith supplements through laying that w ere either high in fat or high in fat and protein (high in just prot ein
proved unpalatable). All supplemented jays advanced the date of clutch initiation relative to cont rols. Alt hough
females fed fat plus protein tended to lay earlier than those fed fat alone, there was no statistical difference.
Supplemented f emales laid larger clutc hes and had more youn g surv ive t o independence, sug gesting a fit ness
benefit to early breeding. These findings fail to support the hypothesis that jays in natural habitats are protein

48       Conw ay & Martin
Effects of clutch size on nest attentiveness and incubation efficiency. COURTNEY J. CONWAY* , Arizona
Coop. Fish & Wildl. Res. Unit, Univ. Arizona, Tucson, AZ; and THOMAS E. MARTIN, Mont ana Coop. Wildl.
Res. Unit, Univ. M ontana, Missoula, M T.
         Experimental tests of Lack's clut ch size hypothesis and eff orts t o quantify t he cost of reproduction in
birds have pr imarily foc used on broo d size manipulati ons. In stead, w e examined pot ential f itn ess costs t o
embryos assoc iated w ith incubat ing a larger clut ch in 3 ground -nesting passerine birds: Dark-ey ed Junco s,
Orange-c row ned W arbl ers, and V irg ini a' s War bler s. We t est ed w het her c lut ch size af fect s 1) egg temperat ure,
and 2) number and duration of daily foraging bouts of f t he nest. We alternately increased and decreased the
original clutch size by 1-2 eggs at each of 8 nests daily. Incubating females did not alter the number or duration
of f oraging bouts off t he nest in response to experimental changes in clutch size. How ever, mean egg
temperat ure w as inver sely relat ed t o cl ut ch size. Relati ve t o egg s in larg er clut ches, eggs in small er clut ches
cooled more quickly and cooled to lower t emperatures during off -bouts, but also were re-w armed more quickly
and heated to higher temperatures during on-bouts. Mean daily egg temperature averaged 0.7 4°C lower on
days w hen c lut ch size w as enlarged com pared to day s w hen c lut ch size w as red uced. Our resu lt s suggest that
avian cl ut ch size c an be l imit ed by cos ts asso ciat ed w it h in cubat ing more egg s in addi tion to co st s asso ciat ed
w ith f eeding more nestlings. These costs have been ignored in previous tests of Lack' s hypothesis and
previou s att empts t o quant ify the co sts associat ed w ith laying m ore eggs.

49      Fowler & McGill
Reproductive and stress endocrinology of Micronesian Kingfishers. GENE S. FOWLER* , Dept. Biol., Pomona
Coll., Claremont , CA and PATTY McGILL, Chic ago Zool . So c. , Brook field Zoo, Brook field , IL.
         Guam' s subspecies of Micronesian Kingfisher (Halcyon c. cinnamomina) w as driven to extinction in t he
w ild by the in trod uced br ow n t ree sn ake (Boiga irregularis) and this kingfisher now persists only as a population
of about 6 0 birds in U.S. zoos. The population w as founded wit h the intention of eventual reintroduct ion.
However, problems w ith reproduction have been identified, namely many birds do not breed or lay eggs at all,
egg fertility rates are low and fertile pairs have high rates of chick mortality. Because each of these problems
has been associated with horm one excesses or deficiencies in other species, this study examined patterns of
repr odu ct ive and stress hor mones in successf ul p airs and n on-breed ing pair s. Paired b loo d and fecal samples
w ere coll ect ed f rom 5 successful p airs dur ing 3 k ey m ilestones in t he br eedin g season – cou rt ship, just after
laying, and just before hatching. Fecal samples were collected weekly for 15 wk from 7 non-breeding pairs.
Samples were analyzed using radioimmunoassays for estradiol, testosterone, progesterone, and corticosterone.
Preliminary analyses show that there is considerable variation in steroid levels. Some birds appear to be
stressed, but the effect is not systematic and does not appear to account for poor reproduction. Continuing
analyses of sex st eroids show less variatio n among ind ividu als that may accoun t f or reprodu ct ive st atus.

50      Kesler & Haig
Nest sit e selection in c oop erat ive breed ing Micr onesian King fishers: An in form ation -t heor etic an alysis. DYLA N
C. KESLER* and SUSAN M. HAIG, USGS For. & Range. Ecosyst. Sci. Ctr ., and Dept. Fish & Wildl., Oregon
Stat e Univ, Corvallis, OR.
         Delayed dispersal and cooperative breeding have been theorized to be adaptive strategies for
ov ercoming lim it ed br eedin g res our ces. Our recent discover y of coo perat ive breed ing in M icr onesian
Kingf ishers (Halcyon cinnamomina) on Pohnpei Island, Federated States of M icronesia, led to an investigation
into t he selection and use of a potentially limited nesting resource, arboreal termitaria. We used logistic
regression and an inf ormat ion-t heoretic analysis t o assess the relativ e influ ence of t ermit aria characterist ics,
vegetation, and distances to f oraging areas on probability of selection for nesting. The top t w o competing
models (Delta AICc < 2) suggest that increased termitaria height from ground and volume have a strong
posit ive eff ect on t he pr obability of select ion as a nest sit e. These models st ron gly out rank ed (cumu lat ive AI Cc
w eight = 91% ) models wit h factors for vegetation and distance to foraging areas. Results suggest, however,
that Micr onesian King fishers d o no t use all t ermit aria t hat have the volu me and hei ght char act eristics associat ed
w ith nesting , and t hat ot her fact ors may li mit breeding oppor tun iti es in thi s species.

51      Renfrew & Ribic
Edge eff ects and gr assland passerines: Ar e larger pastures greener? ROSALIND B. RENFREW* and CHRISTINE
A. RIBIC, Dept. W ildl. Ecol., Univ. W isconsin, M adison, WI.
        Studies evaluating edge effects on nesting grassland passerines have focused primarily on Conservation
Reserve Program (CRP) lands and native prairie fragments. How ever, managers have recognized the need to
include pr ivat e, agricult urally pr oduct ive lands in gr assland bird con servatio n plans. W e studied nest success
and predation in grazed pastures in the Drift less Area of sw . Wisconsin to determine factors influencing
productivit y. From 199 8 - 20 00, w e monitored 820 nests of Savannah Sparrow , Grasshopper Sparrow ,
meadowlark, and Bobolink on 26 pastures. We modeled nest success and nest predation in relation t o edge
and oth er variables. Independent variables includ ed distance f rom edge, edge t ype (w ooded vs. crop/ CRP),
nest concealment, year, site, and initiation date. We developed models a priori and used Akaike Information
Criteria to evaluate models. To determine whether edges influence nest site selection, w e modeled nest density
as a funct ion of distance from edge. Distance from edge w as not an important predictor of nest success or
predation regardless of edge t ype; n est con cealment w as most f requently the best predict or. In cont rast, nest
density w as significantly low er near wooded edges and all edges combined compared to pasture interiors. Lack
of a detectable edge eff ect on productivit y may be due to ease of travel for edge-associated predators in
vegetation that is less dense t han i n CRP or prair ie. Avo idan ce of edges may be an ev olv ed ch aracteristic t hat
does not substantially enhance grassland bird productivity in pastures.

52      Schoeberl & Knick
Inf luence of shrubsteppe f ragm ent ati on o n pr odu ct ivit y and po st -f ledging dispersal of Sage Thr ashers . BRUCE
C. SCHOEBERL* , Dept. Biol., Boise Stat e Univ., Boise, ID; and STEVEN T. KNICK, USGS For. & Rangeland
Ecosyst.Sci. Cntr., Boise, ID.
         We examined mechanisms that may be responsible for the abundance and distribution of
shr ubsteppe-obl igat e passerin es in fragm ent ed lan dsc apes. In 1 99 9 an d 2 00 0, w e monit ored 45 Sage Th rash er
nests to ascertain nest success and post-fledging dispersal patterns in fragmented and contiguous shrubsteppe
habitats in sw. Idaho. Mayfield nest success (P = 0.0 86) and fledging rate (P = 0.0 06) estimates of Sage
Thr asher s w ere hi gher in c ont igu ous than f ragm ent ed hab it ats durin g bo th years. How ever , Sag e Thr asher
abundance between treatments and years did not diff er. In addition, nest success w as positively related to
cheatgras s co ver and n egat ively relat ed t o slope an d average shru b hei ght near n est s. We at tached
trans mitters t o Sage T hras her f ledg lin gs f rom 10 nest s in con tigu ous habi tats durin g bo th years. Disp ersal
pro gres sed aw ay f rom nest s in a lin ear manner thro ugh day 36 foll ow ing fledg ing , af ter w hic h di st ances
deviated and we lost contact w ith all birds by day 45. The furt hest distance that w e observed was 11. 8 km
from the nest at day 40 . No Sage T hras hers fledg ed f rom fragm ent ed hab it ats. Our dat a suggest that
contiguous habitats may funct ion as sources and fragmented habitats as sinks because of diff erences in
productivit y. Furthermore, post-fledging dispersal may facilitate habitat sampling that leads to redistribution of
birds f ollow ing sprin g migrat ion and simil ar abundances betw een treatment s.

53      Hartman & Oring
Breeding bird community composition and nesting density on improved pastures: Eff ects of delayed catt le
grazing. C. ALEX HARTMAN* and LEWIS W. ORING, Dept. Environ. Res. Sci., Univ. Nevada-Reno, Reno, NV.
         The irrigation of pastureland for spring and summer livestock grazing is a common agricultural practice
throughout much of w . North A merica. These pastures often harbor breeding w aterfow l, shorebirds and
ground-nesting passerines. As these birds typically nest between Apr and Jul, postponement of livestock
grazing until Jul may increase nest site availability and nest success through increased vegetation cover and
decreased disturbance. We are in the third year of a long-term study examining the compatibility of livestock
grazing regimes wit h breeding bird populations wit hin a wetland ecosystem in Honey Lake, CA. During t he
spring and summer of 1 999 - 200 1, w e conducted biweekly point count s to monitor bird use of 12 4 .04 -ha
pastures divided into 4 t reatments; (1) irrigated, grazed May - Nov, (2) irrigated, grazed Jul - Nov, (3) irrigated,
ungrazed, (4) non-irrigated, ungrazed. These data were compared to point counts collected from the same
past ures , p rio r t o any impro vem ent (i. e., no i rrigat ion ), in 1 99 4 - 19 96 . In 2 00 0 - 20 01 , w e con duc ted
extensive nest searches to determine nesting species, nest density and nest success. Preliminary results
demonstrate that the development of irrigated pastures radically alters bird community composition and density.
Within -year anal yses show the gr eatest bir d use of past ures w it h M ay-ini tiat ed cat tle gr azing . Nest densit y w as
highest in pastures w ith Jul-init iated cattle grazing and is likely the result of many birds preferentially seeking
increased vegetation cover. Our results suggest that the presence of irrigated pastures is beneficial to some
birds by providing increased food, nesting habitat and cover. The effects of catt le grazing, however, vary
depending on c att le intr oduct ion dat e, bird sp ecies, f oraging behavio r and nestin g requirement s.

54       Mazerolle & Hobson
Physiological cost s of habit at selection in m ale Ovenbirds: Consequences of landscape fragmentatio n. DA NIEL
F. MAZEROLLE* , Dept. Biol, Uni v. Saskatchew an, Saskatoon, SK. and KEITH A. HOBSON, Canadian Wildl.
Serv., Prairie & Nort hern Wildl. Res. Centre, Saskatoon, SK.
          Sinc e bor eal f ores t fragm ent s are of low er qu alit y t han c ont igu ous fores t for Ovenbir ds, w e pred ict ed
that competit ion for breeding sites in contiguous forest should lead to a greater prevalence of high-quality
individuals in these habitats. We quantified male quality using morphological and haematological indices.
Males in contiguous forest w ere larger than males in forest fragments, and they had higher packed cell
volumes, mean corpuscular volumes and a greater prevalence of polychromat ic cells, indices that are positively
associated w ith energetic demands. Furthermore, differential ratios, an immunological index positively
associated with numerous stressors, decreased through the breeding season only for males in forest fragments.
Tot al pl asma p rot ein an d mass corr ect ed f or s truc tural size, ind ices that ref lect nut rit ion al st atus, did not dif fer
betw een landscapes. All of t hese trends w ere independent of age. Overall, these data indicate that size of
male Ovenbirds could play a role in habitat selection, but that defending territories in contiguous forest, w here
breed ing suc cess is high er and pop ulat ion s are denser , r esults in great er ener get ic d emands and red uced
immunological condition. These results also demonstrate a physiological component to contrasting cost s and
benefit s associated w ith terri tor y acquisit ion in bi rds.

55       Norris & Stutchbury
Sexual diff erences in the gap-crossing ability of a f orest songbir d in a fragment ed landscape. D. RYAN NORRIS
and BRIDGET J. M. STUTCHBURY, Dept. Biol., York Univ ., Toront o, ON.
        Determining the frequency and purpose of movements by forest songbirds in fragmented landscapes is
essential t o un ders tand p opu lat ion distrib ut ion and p ersist ence. Recent ly, w e repo rt ed t hat male Hood ed
Warblers left their resident forest fragments and crossed gaps frequently to visit adjacent forests, primarily to
seek extra-pair copulation’s (EPC’s; Norris & Stutchbury 2001, Conserv. Biol. i n pr ess). We radio tracked
females breeding in similar size fragments (< 2 ha) to determine if females possessed a similar ability to move
betw een forest s. A lth ough 4 4% of f emales left t heir fr agment at least once, the f requency of for ays w as 6
times low er than males and significantly low er than the rate at w hich females left t heir territories in continuous
forest. Alt hough males crossed gaps as large as 465 m, f emales did not travel between forests that w ere
separ ated by > 50 m. Unlike m ales, w e foun d no evidenc e that females w ere actively seeking EPC’ s w hen t hey
made extra-territorial forays. This st udy emphasizes the need to address effects of fragmentation on both
sexes. The reluctance of f emales to cross gaps may result in females actively avoiding isolated woodlot s during
the breeding season, explaining why a high percentage of males are unmated in fragmented landscapes.

56       Rideout & Belthoff
Eff ects of habitat fragmentation on abundance and reproductive success of shrub-steppe birds in southeastern
Idaho. CATHERINE W. RIDEOUT* and JAMES R. BELTHOFF, Dept. Biol., Boise State Univ. , Boise, ID.
         During 19 99 - 200 0, our st udy exam ined pot ent ial ef fect s of habi tat fragm ent ation on t hree
sagebrus h-obli gat e bir d species, Brew er' s Spar row s, Sage Sparrow s, and Sag e Thr asher s, at the Id aho Nation al
Engineering and Environmental Laboratory (INEEL) in se. Idaho. Road construct ion and invasion by exotic
gras slands f oll ow ing fire ar e rapi dly fragm ent ing this onc e more cont inu ous land scap e. W e com pared spec ies
richness, bird abundance, nest success, productivit y, and incidence of brood parasitism betw een study plot s in
continuous sagebrush and fragmented patches. Alt hough species richness did not diff er between continuous
and fragmented habitat, Sage Sparrows and Sage Thrashers were significantly more abundant in the latter.
Despite greater abundance in fragmented habitat, likelihood of nest success was low er in fragmented habitat f or
Sage Sparrows. Productivit y (fledglings/successful nest) was significantly low er in fragmented habitat f or
Brew er's Sparrows. Brown-headed Cowbirds parasitized both Brewer's and Sage Sparrows but no more
frequ ently in f ragmented habit at. T hus, ef fect s of h abitat fragm entat ion w ith in shrub st eppe at th e INEEL
appear to be related to (1) increased abundance, (2) low er nest success, and (3) lower productivity f or some
sagebrush-obligate species.

57      Navarro & Calmé
Exotic tree plant at ion as a subst it ut e f or nativ e f or est in a highly defor est ed tro pical landscape. EFREN
HERNÁNDEZ NAVARRO and SOPHIE CALMÉ , Dept. Terrestrial Ecol., El Colegio de la Frontera Sur, Chetumal,
Quintana Roo, Mé     xico.
          We evaluated the use birds make of Gmelina tree plantations in a highly deforested tropical landscape in
southern Campeche, Mexico. Bird relative abundance and activit y w ere recorded using 25-m radius point
counts of 10 -min duration in three adjacent habitats (induced pasture, young and medium-aged secondary
forests) and wit hin the plantation at 3 distances from t he edge (25 m, 7 5 m, 1 25 m ). Bird movements to, from
or w it hin the pl ant ation w ere det ect ed using mist net s t hat w ere lo cat ed in the pl ant ation edge an d pl aced
parallel and perpendicular to the edge. Bird abundance wit hin the plantation decreased wit h the distance from
the edge. There were significant diff erences in bird abundance in the plantation relative to t he type of adjacent
habitat, but only near the edge. Highest abundance w as found near the pasture, then near the young
seco ndar y f ores t, an d f inal ly near t he mediu m-ag ed sec ond ary fores t. Si milari ty in bi rd c omposit ion bet w een
the plantation and adjacent habitats was about 0.5 for all 3 combinations. Bird movements in the plantation
edge w ere mostly f rom the plantation to adjacent habitats, in a lesser extent w ithin t he plantation edge, and
finally from adjacent habitats to t he plantation. Birds used the plantation mainly to song and feed (both
insectivorous and omnivorous), but also to move and rest. M ost species that used the plantation are common
in secondary forest, but a few mature forest specialists that are confined to t he tiny mature forest remnants
present in the landscape also used it. Gmelina plantat ion cou ld serve as an alternati ve habit at f or f orest
dwelling birds or at least help connect forest remnants to allow bird movements betw een them.

58       lisle & St. Clair
Cumulative effects of barriers on the movement of forest birds. MARC BÉ               LISLE, Metapop. Res. Group, Dept.
Ecol. & Syst., Univ. Helsinki, Helsinki, Finland; and COLLEEN CASSADY ST. CLAIR,* Dept. Biol. Sci., Univ.
Albert a, Edmont on, A B.
         Alt hough there is agreement t hat habitat fragmentation has deleterious effects on animal populations
primarily by inhibiting dispersal, there have been few explicit demonstrations of the w ays individual movement
is impeded. Tw o diff iculties are primarily responsible for this paucity: it is diff icult t o separate the effects of
habi tat fragm ent ation (co nf igu rat ion ) f rom habi tat loss (composit ion ) and con vention al meas ures of fragm ent ed
habitats are assumed to be, but probably are not, isotropic. We addressed these limitations w hile standardizing
diff erences in forest cover in a clearly anisotropic configuration of habitat fragmentation by conduct ing a
homing experiment wit h three species of forest birds in the Bow Valley of Banff National Park. Birds were
translocated (1.5 - 2. 5 km) either parallel or perpendicular to 5 parallel barriers that are assumed to impede the
cross-valley trav el of fores t-dependen t anim als. Taken toget her, ind ividu als ex hib it ed lo nger ret urn times w hen
they w ere t rans loc ated acr oss these barriers, b ut dif ferenc es amo ng speci es suggest a more complex
interpretation. A long distant migrant (Yellow-rumped Warbler) behaved as predicted, but a short distant
migran t (Golden-crow ned King let ) w as indif ferent to bar rier con figu rat ion , an d a residen t (Red-br east ed
Nuthatch) exhibited longer return t imes parallel to t he barriers. Our results suggest that an anisotropic
arrangement of small open areas in fragmented landscapes can have a cumulative barrier effect on the
movem ent of fores t anim als, but that spec ies dif ferenc es w ill have to be ac know ledg ed in pot ent ial
management imp licati ons.

59      Brawn, Robinson, Herkert & Heske
Are savanna and shrubland birds area sensitive? JEFFREY D. BRAWN* , Illinois Nat. Hist. Surv. and Univ.
Illinois, Champaign, IL; SCOTT K. ROBINSON, Dept. A nim. Biol. , Univ. Illinois, Champaign, IL; JAMES
HERKERT, The Nature Conservancy, Peoria, IL; and EDWARD J. HESKE, Illinois Nat. Hist. Survey.
          The ad ver se ef fect s of fragm ent ation on f ores t bir ds and, to a lesser ex tent , g rassland bir ds have been
observed frequently. Few er species occur in small fragments and the nesting success of birds that do breed in
these fragments is usually less than that in larger patches. Species characteristic of savanna and shrub
habitats, how ever, may be less area sensitive or prone to negative edge eff ects. We summarize results of
several studies conducted in savannas and shrublands in Illinois that reveal overall weak associations wit h
fragmentation and tract size in terms of presence/absence and breeding success. These habitats are
transitional (spatially and temporally) and, historically, may not have been all that expansive in the midwestern
U.S. o r elsew here. Area sensit ivit y of fores t bir ds has led to t he per vasive view that " big ger i s bet ter" w hen
restoring or conserving avian habitat. Therefore, the issue of dif ferential area sensitivit y among groups of birds
is important for conservation policy. Small patches (e.g., < 50), when managed appropriately, may have
generally unrecog nized value for certain t ypes of birds.

60      Studds
* Landscape composition modulates the effects of forest f ragmentation: Variable consequences for breeding
Ovenbirds. COLIN E. STUDDS, Dept. Nat . Res. Sci., Univ . Rhode Island, Kingston, RI.
         The negative impacts of forest f ragmentation on migrant songbirds are generally attributed t o patterns
of f orest loss. However, landscapes (defined here as the land area within 2 km) w ith similar amounts of
fores t cov er may be fragm ent ed by a var iet y of land uses. Su ch dif ferenc es in land scap e com posit ion may
have uni que c onsequen ces f or b reedi ng b ird s. I t est ed t he hy pot hesis t hat land scap e com posit ion exer ts an
additional impact on breeding Ovenbirds beyond the eff ect of reducing forest cover. I documented Ovenbird
density and reproductive success in 12 randomly selected 12 ha study plots representing 3 dif ferent patterns of
landscape composi tio n: (1 ) unf ragmented landsc apes (> 90 % f orest), (2) agric ult ural landscapes (> 50 %
forest and > 15% agriculture), (3) residential landscapes (> 50% forest and > 15% residential development).
Ovenbird d ensity dif fered signi fic antly w ith landscape composi tio n (F2,9 = 29. 35, P < 0.0 01). Density/12 -ha
was highest in unfragmented landscapes (mean 13.97 ± 0.85 SE), intermediate in agricultural landscapes (8.37
± 0. 94) and low est in residential landscapes (3.02 ± 1.2 2). Nest predation and brood parasitism were not
explained by patterns of f orest cover and forest patch size. Instead, increases in predation w ere linked to
resident ial d evelopm ent and i ncreases in p arasitism w ere relat ed t o agr icu lt ure. These pat terns w ere par allel ed
by inc reases in p redat or ab und ance i n res iden tial l andscapes and i ncreases in c ow bir ds in res pon se t o
agriculture and cowbird host abundance. Annual fecundity w as higher in unfragmented landscapes than in both
the fragmented landscapes, but the mechanism depressing fecundity diff ered and reflected differences in
predation and parasitism pressure among landscapes. These results highlight t he importance of modifying
conservation strategies and management protocols to include consideration of emerging properties of landscape

61      Ogden
* Evidence for age-bias in agricultural habitat use by wintering Dunlin. LESLEY J. EVANS OGDEN, Centre Wildl.
Ecol. , Dep t. Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
         The mudflats of t he Fraser River Delta, Canada' s largest West Coast estuary, are internationally
important habitat f or migrating and over-wintering shorebirds. Adjacent agricultural land is used by shorebirds
for roost ing and feeding, but t he dietary import ance of farmland has not been previously quantified. Using a
multiple source mixing model (e.g. Ben-David et al 1997, Oecologia, 11 1: 2 80-2 91), the proportional use of
agricult ural land versus est uarine mudf lat w as quantif ied via st able isotop e analysis (* 13 C, * 15 N) of Dunlin
(Calidris alpina pacifica) blood samp led o ver three w int ers (19 97 - 200 0). W hil e the 3 -yr mean in dic ated t hat
fields contribute approximately 30 % of Dunlin diet, there was extreme variability in t he extent of field feeding
(range = 0 - 87 % of diet derived from fields). No evidence for diff erential habitat use by sex was found, but
significant age and seasonal differences w ere evident. Younger birds had significantly more terrestrial isotopic
values than adults, indicating t hat fields are particularly important f or juveniles. A juv enile bias in shorebird field
feeding has also been found in Europe (Goss-Custard & Durell 1983, Ibis 12 5: 15 5-17 1). Use of field s dec lin ed
throughout t he winter for adults, but not for juveniles. While the estuarine mudflats provide the main dietary
source for Dunl in, con trib ut ing 70 % of their diet , t he f act that 30 % of Dunlin diet is deriv ed f rom terrest rial
prey suggests that farmland is important to t his wint ering population, and an essential habitat component in
conservation planning for this species.

62       Bourguelat, Gauthier & Pradel
New analyti cal too ls to stud y st opover lengt h in bird s: Wh at can w e learn from t he Greater Snow Goose
example? GRÉ                                                           pt
                  GORY BOURGUELAT* , GILLES GAUTHIER, Dé . Biol ., CEN, Un iv. Laval, Qu é                  bec, QC, and
ROGER PRADEL, CEFE, CNRS, Montpellier, France.
         Migratory birds are known t o be highly mobile and to disperse long distances to find f ood. Under such
condit ions, i t is almo st im possible t o get a com plete f ollow -up of marked indiv iduals on st aging areas.
Ther efore, ver y f ew st udi es hav e been ab le t o qu ant if y stopo ver leng th of bir ds in a st agin g area i n an u nbi ased
w ay. Our objective was to apply and expend recent developments in capture-recapture methods to obt ain
unb iased est imates of st opo ver leng ths in Great er Snow Geese (Chen caerulescens atlantica) staging along the
St. Lawrence estuary in fall. We also examined annual variations in length of stay over the short t erm (recent
years) and the long term (since mid-1980s) to test the hypothesis of a decrease in stopover duration. Intensive
observ ation s of neck -banded i ndi vidual s w ere cond ucted du rin g f alls 19 85 - 198 7 an d 1 99 4 - 20 00 . Su rv ival
analyses using classical encounter histories were performed to estimate the time spent by the geese in the
est uary after an o bser vat ion based on em igr ation pro babi lit ies. Recruit ment analy ses using rev erse encount er
histories were performed to estimate the time spent by the geese in the estuary before an observation based on
immigration probabilities. There w as no significant dif ference betw een the tw o periods, consequently the
hy pot hesis of a decr ease in st opo ver leng th w as not suppor ted. As ant ici pat ed, these est imates are larger than
the on es calculat ed using the mini mum st opo ver leng th method . W e foun d t hat st opo ver leng th w as co rrelat ed
to the proportion of juvenile and transient birds in the fall flock.

63       Carlisle, Rocklage & Kaltenecker
Post-breeding habitat use and stopover biology of Brewer's Sparrows in Idaho. JAY D. CARLISLE* , Idaho Bird
Observ ato ry, Boise, ID and Dept . Biol., Un iv. Sout h Dakota, Verm illi on, SD; A NN ROCKLA GE and GREG
KALTENECKER, Idaho Bird Observatory.
         The Brewer' s Sparrow has suffered significant population declines in recent years and is considered a
shrub-steppe obligate species. Habitat loss may be the most significant f actor affect ing Brew er's Sparrow
populations. Because Brewer' s Sparrow s are migratory, conservation effort s must be informed by life history
inf ormat ion f rom br eeding, w int ering, and m igrati on periods. Here, w e present evidence f rom w ork in t he Boise
Foothills suggesting that Brew er's Sparrows rely t o a large degree on deciduous habitats during post-breeding
and autumn migration periods. Post-breeding and migratory m ovements occurred from mid-Jul through
mid-to-late Sep. Aut umn mist-netting results show ed relatively high capture numbers in both mount ain shrub
and willow riparian habitats and Brewer's Sparrows have been observed foraging in these deciduous habitats.
Brew er's Sparrows captured during autumn generally carried small to moderate fat stores and appeared to
maintain a neutral to positiv e energy balance through morning hours. Addit ionally, a summer and fall study of
bird use of burned and unburned habitats show ed a significant habitat shift from summer to fall by Brew er's
Sparrow s. During the br eedin g season, the sparrow s w ere most com mon in unb urn ed shrub -st eppe f oll ow ed
by burned mountain shrub. During autumn, Brewer' s Sparrow s w ere most common in unburned mountain
shrub and riparian (both burned and unburned) habitats. Considered together, these data suggest that w hile
shr ub-st eppe h abit ats are c ert ainl y of param oun t impor tance t o br eedin g Brew er' s Spar row s, deci duo us habit ats
may be used subst ant iall y duri ng o ther s tages of their lif e cy cle.

64       Webb & Boarman
Post -f ledging movemen ts of juvenile Comm on Ravens in the w est ern Mojave Desert . W ILLIAM C. WEBB* ,
Dept . Bio l., Un iv. Calif orn ia., Riv ersid e, CA ; and WILLIAM I. BOARMAN, U.S.G.S., Western Ecol. Res. Center,
Rivers ide, CA.
          Common m odels fo r explaining vert ebrate dispersal inv olve popu lation dynamic s, pop ulatio n genetics,
and social behavior. We evaluated the post-fledging movements of juvenile Common Ravens in the context of
each of these categories of dispersal models. We marked and follow ed 2 cohorts (n = 240 ) of juvenile ravens
in the w estern Mojave Desert. We evaluated the dispersal models by investigating habitat selection, looking for
evidence of sibling recognition, and testing for sexual dimorphism in movements. Raven juvenile movements
display ed li ttle sex ual d imorp hism. No sex -related di fferenc es w ere ob serv ed in a suite of dispers al variab les
that inc lud ed t he maximum dispersal distance an d t he rat e of movem ent . On ly hom e rang e size show ed
sex-related dif ferences, w ith male home ranges signif icantl y larger. Af ter disp ersing f rom t heir natal t errit ories,
w e frequ ent ly observ ed Raven siblin gs t o be i n cl ose phy sic al pr ox imit y t o each ot her. The d ist ances observ ed
betw een sibling pairs w ere significantly less than non-sibling pairs observed w ithin t he same day, suggesting
some level of sibling recognition. We investigated Raven habitat selection by comparing the relative proportions
of h abitat use and availability . Raven juveniles ut ilized urban and agricul tur al habitat s more t han expect ed (97 %
of t he total), w hile under-utilizing natural habitats. A lack of female-biased movements cont radicted predictions
based upon the mating system models. Evidence for sibling recognition w as consistent w ith population genetic
models. The patterns of habitat selection related to resource abundance supported population dynamic models.
Numerous factors affect avian movements and a comprehensive understanding of dispersal may only come
fro m consid ering a combi nation of m odels.

65       Morrissey
Population structure and movement patterns in American Dippers and its implications for contaminant exposure.
CHRISTY A. MORRISSEY, Dept . Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
          A considerable lack of understanding about t he population dynamics and movement patterns of the
American Dipper exists in British Columbia. The objective of this study w as to determine the patt erns of
movement in addition to t he overall population structure. In t he south coastal region of t he province, rivers and
streams w ere found to host relatively high densities of dippers in the fall and wint er months. These numbers
greatly exceeded the resident breeding population indicating significant seasonal movement of a large portion
of the po pul ation . Cu rrent lit erat ure s ugg est ed t hat the A merican Dipper i s non-m igr atory , b ut possibly makes
some seasonal altitudinal movements. Conversely, ot her researchers suggested that t he large numbers of
bir ds w int erin g in this regi on w ere f rom more no rt hern lat it udes subject to f reezin g. We determined that < 20 %
of the winter population was resident year round while the remaining majority w ere considerably more mobile.
Individually color-marked birds and radio-tagged birds were successfully used to identify seasonal movement t o
hig her el evat ion s in spr ing and l ow elevation s in fall (alt it udi nal m igr ation ). Many w int er v isitors w ere ob serv ed
in t he same location f or 2 or mor e w int ers. Since t here is a considerable int erest in usin g th is species as a
biological indicator of environmental change, w e have applied this know ledge to determine differences in
exposure t o cont aminants f rom t he diet in r esident and non -resident di ppers.

66       Johnson, Crook, Burt, Saenz & Conner
An aut omated reco rding sy stem f or detect ing noc tur nal birds. JA MES B. JOHNSON* , SHAUN L. CROOK, D.
BRENT BURT, Stephen F. Austin State Univ. Dept. Biol. Nacogdoches, TX, DANIEL SAENZ, and RICHARD N.
CONNER, Southern Res. Station, U.S. For. Serv., Nacogdoches, TX.
          Nocturnally vocalizing birds are typically diff icult t o sample because of diff iculty in navigating w ild land
habitat after dark. Automated recording devices known as Frogloggers were used to conduct nocturnal surveys
of birds from May 20 00 t o Jul 20 01. We programmed the Frogloggers to record for one-minute intervals at the
begi nni ng o f each h our st art ing at 21 :0 0 an d end ing at 02 :0 0, for a t ot al of 6 m in at each s it e/n igh t. Data
w ere recorded at 8 diff erent sites wit hin the Davy Crockett National Forest and the Stephen F. Austin
Experimental Forest in e. Texas. Av ian vocalizations at each site w ere sampled simultaneously every night,
w hich w ould not have been possible using traditional bird surveying techniques. We detected a variety of
nocturnal and diurnal species. Our technique allow ed us to accurately determine the seasonal occurrence of
species because we w ere able to survey every night of t he year. We w ere also able to accurately determine the
number of times each individual bird called because we were able to review t he recordings repeatedly.
Technology that w as originally developed for anuran surveys is proving quite useful in its application to

67      Titus, Flat ten & Low ell
Breeding d ispersal of adul t Northern Goshaw ks i n sout heast Alask a: Implic ation s f or c onserv ation . KIM BERLY
TITUS* , Alaska Dept. Fish & Game, Juneau, AK, CRAIG FLATTEN, Alaska Dept. Fish & Game, Ketchikan, AK;
and RICHARD LOWELL, Petersburg, AK.
         Northern Goshaw ks nest in the old-growt h temperate rainforests of se. A laska and are a conservation
concern f or f orest m anagement acti vit ies. We evaluated br eeding dispersal (mov ement f rom one nest to
anot her i n co nsec ut ive year s), nest ing st atus, and f ate of adul t goshaw ks d uri ng 1 99 2 - 19 99 by trackin g t hem
w ith radiotelemetry. We defined a nest area as a 3.2 km diameter area because this w as the maximum
distance a pair moved in consecutive years while maintaining the same home range. Multi-year movements
w ere determined for 23 females and 21 males at 2 7 nest areas. Breeding dispersal was observed only for adult
females and no adult male moved to a new home range or nest area. For 13 nests located by tracking adult
females to a diff erent nest area in consecutive years, the median distance moved w as 18.5 km; maximum
distance moved was 152 km. In 55% of our consecutive year outcomes the goshawk pair nested in the same
nest area as the previous year. Overall, 31% of adult females dispersed to a new home range in consecutive
years and either nested w ith a diff erent mate or did not nest. These complex dispersal patt erns by adult female
goshawks present challenges to those charged with monitoring goshawks and/or their nests.

68      Winkler, W rege, Allen, Wasson, Kast & Senesac
Natal dispersal of Tree Swallow s in a continuous mainland environment: Results from an intensive local study
and t he Corn ell Nest Box Net w or k. DA VID W. WI NKLER* , PETER H. WREGE, PAUL E. ALLEN, MATTHEW F.
WASSON, TRACEY L. KAST and PIXIE SENESAC, Dept. Ecol. &amp; Evol. Biol., Lab. Ornithol., Cornell Univ.,
Ithaca, NY.
         Tree Swallow s have been studied near Ithaca, NY, since 198 5, w ith t he numbers of nests and
geographic scale of sampling being increased substantially to include much of surrounding Tompkins Co.
starting in 1991 and the addition of a network (“ CNBN” ) of collaborating lay banders throughout a 400 km
circle around Ithaca starting in 1995 . Working w ith a combined network of almost 100 banders in most years
since 1997 , the t otal number of natal dispersal events observed since 198 6 now tot als 791. On average, in
the years since the start of CNBN, about 40 00 nestlings have been banded/year, 45% of w hich have come
from Tompkins Co. Overall, the mean natal dispersal distance for males was 2.5 km and for females was 6.9
km, an d t he mean and rang e of recapt ure d ist ance w as subst ant iall y shor ter t han t he di st ances over w hic h t hey
could have been detected. M ethods to statist ically correct for t he biases imparted by the geography of trapping
effort depen d up on assum pt ion s about the di sper sal proc ess. Juveniles dispers ing from areas of hig her
nest-site density disperse shorter distances than those from lower nest-density, and dispersal vectors appear to
be no more conc ent rat ed in any giv en di rection s t han ex pect ed by chan ce.

69      Wiedenfeld, W olfe & Sherrod
Home ranges and movement s of r adio-tagged Greater Prairie-Chickens in a homo geneous, unb ounded t allgrass
prairie in northeastern Oklahoma. DAVID A. WIEDENFELD* , DON H. WOLFE, and STEVE K. SHERROD, Sutton
Av ian Research Center, Bartlesvill e, OK.
         Alt hough Greater Prairie-Chickens often live in fragmented habitats divided by cultivated fields, roads, or
woodlands, it is useful to understand their range requirements in a homogenous, unbounded habitat. We
measu red t he ho me ranges and m ov ements of 45 radi o-t agged Great er Prairie-Chic kens in t allg rass prai rie o f ne.
Oklahoma. The habitat is homogenous grassland, w ith f ew roads or other struct ures and no cultivation. All
birds were tracked and located at least 50 times. Tracking began on all birds in the spring of the year (Mar -
May) of 1997, 1998 or 1999, and continued for varying amounts of time until the bird was killed or lost.
Home ranges were defined using a median-centered minimum convex polygon of 98 % of t he points. Home
range size averaged 6.61 km 2 . Home ranges of males (5.9 4 km 2 ) averaged about b the size of t hose of
females (8.64 km 2 ), but the dif ference w as not sig nif icant (M ann-Whit ney U = 25 4, P = 0. 15 ). The largest
home range was of a hen (21.6 7 km 2 ). Females are more likely to have moved from one activity center to
another t han males (P 2 = 14 .7 8, df = 1, P < 0. 00 0). The maximum dist ance m ov ed by ind ividu al bi rds w as
also signif ican tly larg er in females (7.9 6 k m) t han m ales (3 .9 0 k m; U = 31 3, P = 0. 00 3). The shape (roun ded
vs. linear) of home ranges, as measured by the index (area / maximum dimension), w as not significantly
different betw een the sexes. In this unbounded habitat the prairie-chickens generally remained w ithin a few km
of their first location, although some individuals, especially hens, did relocate their activity center after some
tim e. Some long-range mo vement s, how ever, may have gone undet ected by our met hods.

70      Plissner, W arnock, Carlisle, Blackford, Garcelon, Brubaker, Opdycke & Brock
Juv enile surviv orship and di spersal of w ild and reint roduced San Clemente Loggerhead Shrikes. J ONATHA N
Stinson Beach , CA ; DAVID GARCELON, DON BRUBAKER, Inst . W ildl. Stud. , A rcat a, CA , JEFFREY OPDYCKE,
Zool . So c. San Dieg o, San Dieg o, CA; and KELLY BROCK, U.S. Navy , Commander Navy Region Sout hw est,
San Dieg o, CA.
          In 199 9 and 200 0, as part of recovery eff orts f or the subspecies, captive-reared San Clemente
Loggerhead Shrikes (SCLS) w ere released on San Clemente Island, Califo rnia, usi ng several " soft " release
techn iqu es. Compari sons of the beh avior an d po pul ation dynamics of w ild and c apt ive-released bird s pr ov ide an
assessment of the rel ative suc cess of these v ario us t echn iqu es. In t he t w o years, 18 juv enil es w ere released
w ith t heir parents from f amily cages in suitable breeding habitats. In addition, 29 juveniles w ere released,
following independence, in groups of 4 - 7 birds. During the same period, 22 individuals were known to have
fl edged in t he w ild popul atio n. 6 4% of w ild and 89 % of fam ily release fledglin gs sur viv ed to independ ence. Of
individuals att aining independence, overwint er survivorship and recruitment rates into t he breeding population
of released ju veniles w ere at least as high as those of w ild bir ds. The m ean age at dispers al f rom nat al sites
w as equal among w ild and family release juveniles, although birds reintroduced in juvenile flocks remained in
the vici nit y of the rel ease cages 9 d lo nger bef ore d ispersing f rom the site. W ild -rear ed ju veniles dispers ed
shorter distances to wint ering areas (mean 3.2 km) than did first-year birds from f amily (4.2 km ) or juvenile
(5.0 km) releases. These data indicate the potential for captiv e-reared juveniles to be successfully released into
and supplement the wild population of SCLS on San Clemente Island.

71      Cooper, Daniels & Walters
Improving estimates of juvenile dispersal and recruitment. CAREN B. COOPER* , Lab. Ornithol., Cornell Univ.,
Ithaca, NY, SUSAN DANIELS and JEFFERY R. WALTERS, Dept. Biol., Virginia Tech, VA.
           Estimates of distribut ions of natal dispersal distances and juvenile recruitment rates can be strongly
inf luen ced b y size and shape of st udy areas. A met hod to co rrect for t hese inf luen ces w as pr opo sed by Baker
et al. (1995, Condor 97: 663 -674 ). We examined the effectiv eness of the Baker correction method by creating
' sample' study areas of different sizes and shapes nested w ithin a large, individually marked population of
Red-co ckaded W ood peck ers (n = 22 0 b reedi ng g rou ps) in Nort h Caro lin a. Fir st , w e est imated di sper sal
distributions and recruitment f or the small sample study areas. Then we applied Baker's correction and
compared corrected estimates to our most accurate estimate, that observed by including all birds dispersing
outside the small sample study areas but w ithin t he larger population. Sample study areas composed of
terri tor ies selected at random fro m t he entire popu lation produc ed unbiased dispersal estimat es and most
clo sely mat ched t hose of the t rue p opu lat ion . Ci rcular and l inear samp le st udy areas prod uced bi ased dispersal
est imates t hat w ere gen erall y improv ed by the Baker cor rection met hod . Su rv ival t o rec rui tment w as
underestimated in each sample study area, and oft en overestimated by the Baker method.
72       Cooper, Walters & Priddy
Effects of landscape patterns on dispersal success: Field data and a simulation. CAREN B. COOPER,*
JEFFREY R. WALTERS, Dept. Biol., Virginia Tech, Blacksburg, VA, and JEFFERY PRIDDY, Duke Marine Lab,
Beaufort , NC.
          We used a simulat ion model t o explai n t he po ssible r ole o f pat ch isolat ion in c ausing o bser ved decl ines
of Brown Treecreeper (Climacteris picumnus) in n. New South Wales, Australia. We created a spatially realistic
land scap e in w hic h t errit ori es w ere constrain ed t o w ood land s and the matrix com posit ion mimicked t he act ual
land scap e in t he st udy area. We compared observ ed po pul ation behav ior to t he ou tcome of sim ulat ion s based
on 2 dispersal movement rules and 3 dispersal mortality rules. When initial dispersal direction w as random,
populations in contiguous habitat w ere relatively stable while populations in fragmented habitat st eadily
declined due to low female recruitment, w hich is the same pattern observed in the real population.
Populations in both contiguous and fragmented habitat increased when initial dispersal direction was tow ard
the nearest neighbor, suggesting that population dynamics were sensitive to dispersal search patterns.
Varying disperser mortality w ith habitat type strongly affected population behavior even though long-distance
dispersal was infrequent and short-distance dispersal exceedingly common. Thus, for this cooperative breeder,
w here territories become clumped due to a high budding rate, matrix habitats continue to inf luence overall
pop ulat ion perf orm ance.

73       Sgariglia & Burns
Phylogeography of the California Thrasher. ERIK SGARIGLIA* and KEVIN J. BURNS, Dept. Biol., San Diego
State Univ ., San Dieg o, CA.
         Phylogeography utilizes geographic distributions and evolutionary relationships to inf er the
bio geog raph ic h ist ory of an in div idu al species. A lt hou gh a n umber o f phy log eogr aphi c st udi es investigat e
California vertebrate taxa, very few have focused on birds of t his region. Significant geographic structure in
genet ic v ariat ion w as det ect ed in the Cali forn ia Th rash er us ing the cy tochro me b gene. Over the entire range,
isolation-by-distance (IBD) w ith restrict ed gene flow is identified by nested cladistic analysis and supported by
IBD st atist ical analy ses. In addit ion , g enet ic d ata cor resp ond w it h geo log ic ev ent s t o explai n t he in ferred
areas of ancestry, f ragmentation, and expansion wit hin this species. Comparative phylogeographic analysis
recognizes concordances betw een the California Thrasher and other co-distributed vertebrate taxa.

74       Stenzler & Fitzpatrick
Micr osat elli te markers rev eal gen etic d if ferent iat ion in Flo rid a Scrub-Jay pop ulat ion s along Flori da' s Lake W ales
Ridge. LAURA STENZLER* and JOHN W. FITZPATRICK, Cornell Lab. Ornithol. , It haca, NY.
         Genetic variation and its geographic distribution of ten correlate with dispersal ability. Especially in
animals having limited dispersal, genetic dif ferentiation should increase wit h geographic separation even wit hin
continuous habitat. We sought evidence for this relationship among subpopulations of the Florida Scrub-Jay,
FSJ), a non-migratory cooperative breeder restrict ed to patchily distribut ed oak scrub of peninsular Florida. The
FSJ is among the most sedentary birds know n (most recruit into breeding ranks within 2 - 3 territories of t heir
natal site). We sampled jays along the north/south-oriented Lake Wales Ridge of central Florida, a stretch of
historically continuous scrub habitat. We examined geographic distribution of 1 8 microsatellite markers in 9
subpopulations to determine 1) if genetic dif ferentiation exists, 2) t he minimum distance at w hich genetic
struct ure could be detected, and 3) whet her a pattern of genetic isolation by distance exists. Significant
genetic dif ferentiation w as found at some loci, but not among all subpopulations; genetic struct ure was evident
at distances of 11 km and ab ov e; g enet ic i solation by distance w as eviden t. Ou r results co nf irm that lim it ed
dispersal in this species leads to genetic diff erentiation over remarkably short distances, and could provide
baseline calibration for measuring gene flow betw een isolated populations in this Threatened and declining
75      Hendrickson, Bleiweiss, M atheus, Silva de Matheus, Jácome & Pavez
Low genetic var iabilit y i n t he geographically w idespread Andean Co ndor. SHER L. HENDRICKSON* , ROBERT
BLEIWEISS, Dept. Zool., Univ. Wisconsin, Madison, WI, J UAN CA RLOS MA THEUS, LILLY SILVA DE
MA THEUS, Fundación para el Estudio e Investigación de los Colibrí es Ecuatorianos (F.E.I.C.E.), Quito,
Ecuador, NORBERTO LUIS JÁCOME, Zoo de Buenos Aires, Buenos Aires Argentina, and EDUARDO PAV EZ,
Unión d e Ornit ólo gos de Chi le, Sant iago , Ch ile.
          The Andean Condor (Vultur gry phus) is a geographically widespread and exceptionally large scavenger.
We characterized DNA sequence variation for 13 46 base pairs comprising the mitochondrial control-region and
adjacent 1 2S ribo somal subuni t f rom ind ividu als representing po pulati ons dist ribut ed thro ughout the species'
rang e (Colombia t o cen tral A rgen tina an d Chi le). We ident if ied a t ot al of 5 h aplo typ es based on 4 v ariab le sites
in domains II and III of the control-region and the 12S gene. An additional variable site was identified in domain
I in a sample of 5 birds w ith complete sequence. All changes w ere transitions and no more than 3 basepairs
dif fered b etw een in div idu als. Across the co ndo rs' rang e, w e det ect ed lo w var iabi lit y w it hin samp les and w eak
distinc tion among samples. Var iat ion in t he co nt rol -reg ion of con dor s w as low er t han f or o ther b ird s analyzed
for t hese same loci. Alt hough low genetic variability is oft en associated with endangered megafauna, the
condor example is notable because the samples were obtained from conspecific populations distributed over a
large and physically heterogeneous area. It is not know n w hether the low v ariability in condors is due to a
historical bottleneck, or to some intrinsic f eature of megafauna. In either case, the Andean Condor example
pro vides an impor tant con text for u nder st andi ng p atterns of genet ic v ariab ili ty in rel atives and in o ther
endangered species.

76      Bates, Zimmer, Cardoso da Silva & Hunt
On the discovery of the Chapada Flycatcher (" Suiriri" islerorum) and molecular evidence for its surprising
systemtatic position. JOHN M. BATES, Field Mus., Chicago, IL; KEVIN J. ZIMMER, Los Angeles Co. Mus., Los
An geles, CA; JOSE MARIA CARDOSO DA SILVA, Universidade Federal do Pernambuco, Recife, Brazil; and
          When the Chapada Flycatcher (" Suiriri" islerorum) w as described earlier this year, it w as called a cryptic
species because it was so similar to the Cerrado subspecies of Suiriri Flycatcher (Suiriri suiriri aff inis) that all
older specimens were in museum trays as this taxon. The 2 birds co-occur in and are restricted to t he Cerrado
regi on o f Sout h A merica. Because of the morp hol ogi cal s imilar it ies, it seemed appr opr iat e to pl ace t he new
species in Suiriri even though it possesses distinctive vocalizations and behaviors. We analyzed tissues from
the type series of Chapada Flycatcher w ith an extensive sampling of Suiriri and o ther o pen c oun try flycat cher
species. Both mitochondrial and nuclear sequences provide strong support f or a surprising systematic position
for t his new species, which t urns out to be not only cryptic but also not closely related to Suiriri. The
discovery of this species illustrates well the value of multiple data sets, general collecting, and the importance
of behav ior al dat a.

77      Chesser & ten Have
Continent-w ide phylogeography of a morphologically variable Australian scrubwr en. R. TERRY CHESSER and
JOSE TEN HAVE, Aust rali an Nat l. Wildl . Co llec tion , Can berr a, A ust rali a.
          The White-browed Scrubwr en (Sericornis front alis) is among the most morphologically variable of
Australian birds, having at times been considered a single polytypic species or as many as 4 distinct species.
We studied variation in mit ochondrial DNA of 100 + individuals sampled from each of the 12 current ly
recognized subspecies and from zones of morphological intermediacy. Our results reveal several divergent
w ell-supported groupings of haplotypes, groupings that dif fer from previous assessments based on morphology
and protein electrophoresis. Implications of t hese findings w ill be discussed and comparisons made wit h
st udi es of co- distrib ut ed t axa.
78      Pavlova, Zink, Drovetski, Rohwer & Red' kin
Phylogeographic patterns in Yellow Wagtail and Citrine Wagtail: Species limits and population history.
ALEXANDRA PAVLOVA* , ROBERT M. ZINK, Bell M us. and Dept . Ecol. Evol . & Behav. , Un iv. Mi nnesota, St.
Paul, MN; SERGEY DROVETSKI, SIEVERT A. ROHWER, Burk e Mus. and Dept Zool. , Univ . W ashingt on, Seattle,
WA; and YAROSLAV RED'KIN, Zool. Mus. , M osc ow State Univ. , M osc ow , Russia.
          Phylogenetic analysis of mitochondrial DNA sequences from 162 Motacilla flava (Yellow Wagtail) and
38 Mot acilla citreola (Cit rin e Wagtail) rev ealed 3 clad es w it hin tradi tion ally recogn ized M. flava, distribut ed in
w est ern, nor theast ern an d sout heast ern Eur asia. These 3 c lades are no t monop hy let ic, and ar e int ersp ersed
w ith M. cit reola, M. c iner ea and M. alba. Mot acilla citreola is also paraphyletic. Northeastern and southeastern
groups of M. flava each appear to be sister-taxa to eastern and western groups of M. cit reola respectively.
Within each of the 3 groups of M. flava and the 2 groups of M. cit reola, l it tle geo grap hic st ruc ture w as
detected. Population expansion and gene flow taken together likely explain the lack of phylogeographic
st ruc ture.

79      Voelker
Syst ematics and bi ogeography of w agtails (Mot acilla: Motacillidae). GARY VOELKER, Barrick Mus., Univ.
Nevada Las Vegas, Las Vegas, NV.
         Usin g 1 0 o f 11 cur rent ly recogn ized speci es, 20 00 base pairs of the cy tochro me b and ND2 genes, and
various w eighti ng schemes and algorit hms, I reconst ruct ed molecular phy logenies of the genus Mot acilla. All
metho ds of analysis result ed in th e same basic phylogenet ic hy pot hesis w ith respect t o species relationsh ips;
alternate topologies consisted only of diff erent rearrangements of indiv iduals within 2 species. I compared the
best estimate of relationships to alternate topologies w hich forced currently recognized species and
superspecies groups to be monophyletic; all but one of t hese alternate topologies represents a significantly
w orse estimate of the data. The three traditional superspecies groups of " pied" , " yellow" and "stream"
w agtails are not monophyletic. Both Motacilla flava and M. cit reola are paraphyletic assemblages: a southeast
" flava" group is sister to w estern "citreola" , eastern Palearctic " flava" is sister to eastern " citreola" , w ith a third
central and western Eurasian " flava" group representing a more basal radiation. M. lugens and M. alba w ere
shown t o be paraphyletic w ith respect to one another in all analyses, although a tree forcing t hese taxa to be
monophyletic was not rejected as a worse estimate of relationships. M. m adaraspatensis is likely a valid
species, and not a subspecies of alba. Biogeo grap hic reconstruc tion s suggest that Mot acilla evolved during the
late Pliocene in the eastern Palearctic and colonized Africa tw ice, and Madagascar once during that same
period. Seven other intercontinental movements are hypothesized to be recent occurrences.

80      Cicero & Johnson
Mosaic patterns of evolution in t he Sage Sparrow (Amphispiza bellii) complex. CARLA CICERO* , Mus. Vert.
Zool ., Univ . Calif ., Berkeley , CA and NED K. JOHNSON, Mus. Vert. Zool. and Dept. Integrative Biol., Univ.
Calif ., Berkeley , CA .
          The Sage Sparrow complex contains three strongly-characterized subspecies that diff er in size, plumage
color, habitat, migratory tendency, and continuity of nesting distribut ion. In a previous study, Johnson &
Marten (1992, Condor 94: 1-19 ) defined 2 major groups of populations based on morphometrics and
allozymes: (1 ) A. b. belli (Coast ranges of Californi a, w est slope of Sierra Nevada) plus A. b. canescens
(sout hern San Joaquin V alley, n. Moj ave Desert); and (2 ) A. b. nevadensis (Great Basin, Colorado Plateau). The
alliance of canescens and belli based on size and allozymes conflicts w ith t he patt ern shown by plumage color
and migrat ory t endency, w hich places canescens and nevadensis as most similar. To further explore
relationsh ips among t hese groups of populat ions, w e sequenced 218 4 bp o f m tDNA (complet e cyt b and ND2
genes) f rom 15 sit es t hro ugh out the ran ge of the species. Divergen ce among pop ulat ion s w it hin each o f the 3
subspecies w as low compared to divergence betw een subspecies. The mtDNA data agreed wit h allozymes in
showing genetic intr ogression betw een phenotypically divergent populations of belli and canescens, although
these taxa are reported not to intergrade. In contrast to all other characters, however, mt DNA placed belli and
nevadensis as sisters relative to canescens. This mosaic patt ern of evolution in diff erent characters w ill be
disc ussed i n lig ht of th e biog eograp hy of Sage Sparro w s in t he w . U. S.

81      Dietsch
* Assessing con serv ation suc cess for Neot rop ical bir ds ac ros s an int ensity gradi ent of cof fee agro ecos ystem
management in Chiapas, Mexico. THOMAS V. DIETSCH, School Nat. Res. & Environ., Univ. M ichigan, Ann
Arbor, MI.
          Certif icatio n of agricult ure as biodiversit y-f riendly o r ecologic ally-sust ainable has been put for w ard as a
method of improving conservation in established agricultural areas by linking economic returns to more
desirable growing methods. Success of this approach requires testing w hether certification criteria, specific
vegetative measures, can meet conservation objectives. This study examined Neotropical bird diversity and
abundance across a gradient of coffee growing practices to relate changes in the avifauna to vegetative
struct ure resulting f rom coff ee management. Point counts w ere used to sample the avifauna during wint er and
summer in 5 coff ee management systems for 2 yr. Points in associated forest reserves provided a conservation
baseline. Vegetative st ruc ture w as ch aracterized at each p oin t cou nt loc ation for c ompari son to co ffee
certification crit eria. A Bray-Curtis dissimilarity index w as used to compare each point location t o the avifauna
found in the associated forest reserve. The birds found in rustic coff ee management w ere significantly more
sim ilar to t he f ores t avifauna t han o ther m anagem ent typ es. Usin g li near r egres sion, the co ffee avifauna w as
significantly less similar to t he forest avifauna wit h more intensive management. In particular, this st udy
suggests that forest-associated birds are sensitive to changes in vegetation structure resulting from
management intensity. Additionally, resident birds demonstrated a stronger forest association than migrants.
These results sugg est t hat t he stron gest co nservat ion benefi t f or f orest-associ ated birds i s likely t o be fro m less
intensive coffee agroecosystems, such as those using rustic management practices.

82      Evans
* Brood rearing habitat selection by Barrow' s Goldeneye and its effect on duckling growt h, survival, and return
rates. MATTHEW R. EVANS, Cent re Wi ldl. Ecol ., Simo n Fraser Univ ., Burnaby, BC.
         Obtaining a suitable nest site is generally of paramount importance to breeding success in secondary
cavity nesting birds. Barrow' s Goldeneye is a cavity nesting duck t hat defends exclusive brood rearing areas on
ponds, and this defense has been interpreted as defense of f ood, suggesting that variation in pond quality is
also quite important. I indexed the aquatic macroinvertebrate prey abundance available to goldeneye ducklings
on 3 0 p ond s f or 3 yr (199 7 - 19 99 ). Var iat ion in i nv ert ebrat e abun danc e w as co nsist ent among p ond s bet w een
years, thus indicating that ponds vary predictably in their probable annual productivity . Neither initial clutch
size nor pre-fledging duckling survivorship correlated with a pond's invertebrate abundance. How ever,
ducklings raised on ponds with higher productivity had substantially higher body weights at 45 d old, and w ere
more likely to return to t he breeding grounds the follow ing year. I conclude that f ood abundance in brood
rearing ponds has an immediate influence on duckling grow th, w hich appears to affect post -fledging survival,
making a substantial contribut ion to f itness differences among females. Predictable variation in pond
productivit y w ill strongly select for annual philopatry and territ orial defense of more productive ponds by
breeders, while birds breeding on less productive ponds would be expected to attempt to mov e to more
productive sites from year to year. In this species, variation in duckling food, as well as nest site availability,
appear to be strong components of f itness variation.

83      Battin
* Source, sin k, or trap? Populatio n dynami cs of breeding Plumbeous Vi reos under pond erosa pine for est
restoration. JAMES BATTIN, Dept. Biol . Sci ., Nort hern A rizona Univ. , Flagst aff , A Z.
        Sever al sourc e-sin k model s hav e been p rop osed to describe h ow bir d po pul ation s behave in lan dsc apes
in w hich habitat quality varies widely. These models share a number of assumptions, including the assumption
of an id eal pr eemptive or i deal d espotic m echan ism of habi tat selection (i. e., a behav ior al co nst rain t on h abit at
selection t hat forces late arrivers or poor competitors into poorer habitat patches). An alternative to t he
source-sink concept is the ecological trap, in w hich animals are unable to assess habitat quality correctly and
end up choosing to breed in low er-quality habitats. Source-sink and trap dynamics can be difficult to discern in
the field. Both models predict large differences in reproductive success between habitats, but t he long-term
implications of the t w o models differ greatly, w ith source-sink theory generally predicting population stability
and ec olo gic al t rap t heor y general ly pred ict ing pop ulat ion ext inc tion . To t est the assumpt ion s of these t w o
models, I examine habitat selection and its fitness consequences in Plumbeous Vireos in a landscape that is
und ergo ing pon dero sa pine f ores t rest orat ion . The res torat ion pro cess creates a landsc ape co nt aini ng p atches
of open, savanna-like restored forest, and closed-canopy, unrestored forest. I report data on early-season
habi tat selection by male an d f emale v ireo s, vireo d ensities in t he 2 habi tat typ es, evidenc e of habi tat
saturat ion, and b etw een-habitat c omparisons o f repr oduct ive success, nestlin g grow th r ates, and parasit ism
rates by dipterans. I compare the predictions and assumptions of various theoretical models to the data and
suggest modifications that might make current models more effective for this system.

84       Niven & W ells
The Important Bird Area Program: A site-based approach for identifying and conserving critical bird habitat.
DANIEL K. NIVEN* , Nat io nal A ud ub on Soc. , Aud ub on Scienc e Cent er, Lahaska, PA; and JEFF WELLS, Nat ion al
Audub on Soc., Cornell Lab. Ornithol. , It haca, NY.
           Important Bird Areas (IBAs) are sites that provide critical habitat f or one or more bird species. Sites are
identified based on a few general criteria: sites w ith t hreatened or endangered species, species of concern, or
Watc hlist species; sit es representativ e of rare or un ique habit at t ypes t hat supp ort unique bird commu nit ies;
crit ical research sites; and sites t hat have large con gregation s of sp ecies groups such as w aterf ow l, w aterbird s,
shorebirds, raptors, or migratory landbirds. Thresholds are established to identif y sites of significance at the
state, national, continental, or global level. The IBA program was initiated in Europe in the mid-1980 s by
Bird Lif e Int ernat ion al. Sinc e then o ver 50 00 IBAs hav e been i dent if ied i n Euro pe and pro gram s hav e been
est abli shed in t he M idd le East, A sia, Afric a, an d t he A mericas . IBAs have been l abeled by som e as a global
currency for site protection. In the U.S., Audubon initiated an IBA program in 1995 on a state-by-state basis.
Currently, over 30 states have established programs and many states have begun the process of implementing
conservation actions at these sites. With t he advent of NA BCI, IBAs are playing a role in identifying
on-the-grou nd sit es f or i mplemen ting all b ird , all hab it at con serv ation in p art ners hip w it h t he US/ North A merican
conservation plans for w aterfow l, w aterbirds, shorebirds, and landbirds. Through the implementation of BirdLife
International's World Bird Database and effort s of Audubon/ Cornell' s BirdSource staff, IBA data will soon be
available to the public over the Internet w ith a suite of user-defined query tools.

85       Marzluff
Understanding why urbanization affects birds. JOHN M. MARZLUFF, Coll. For. Res., Univ. W ashington,
Seattle, WA.
        Bird communities change in response to urbanization because some species are favored and others are
reduced by sett lement. However, t o better understand this process we need to identif y the mechanisms by
w hich species' reproduction, survivorship, and dispersal are affected by human settlement. I report on our
state of understanding by reviewing t he scientif ic literature and sharing new information f rom a study of birds
in urbanizing w . Washington. Primary among factors identified to favor species is increasing availability of f ood
w hich increases fecundity and occasionally overw inter survival. Predator reduction, and to a lesser extent,
redu ced h uman per secu tion also have benef it ted species in u rban areas, most not ably smal l rap tors . Urban
enhancements to the natural habitat also benefit species in certain settings. A complex w eb of factors, bot h
direct and indirect reduce the suitability of urban areas to birds. Important direct effects on fecundity are
decr eased habit at availabi lit y, redu ced p atch size, inc reased edge, inc reased exotic v eget ation , an d dec reased
vegetative com plex it y. Ind irec tly, n est pred ation inc reases in u rban areas becaus e many of the species favored
by increased food and reduced predation are eff ective nest predators. Competitors, brood parasites, and direct
human disturbance also reduce fecundity of native birds. Survivorship of birds in urban areas may also be
reduced, principally by indirectly facilitated increased in small raptors and direct introductions of domestic cats.
Reduction of insects, increased toxins, and exotic diseases are reported less frequently to limit birds.

86      Donnelly & Marzluff
Combining bird community models for more effectiv e conservation in urban areas. ROARKE DONNELLY* and
JOHN M ARZLUFF, Ecosys. Sci. & Conserv., Univ. Washington, Seattle, WA.
         Bird conservation in urban areas has relied heavily on t he competition-based Theory of Island
Biogeography because urbanization produces "islands" of native habitat w ithin a "sea" of other habitat types.
Yet, wit h the increasing importance placed on edge effects created by nest depredation and parasitism,
managers and researchers have begun to mingle models of communit y struct ure that have different struct uring
mechanisms. In order to reconcile these models with each other and adapt them to the urban ecosystem, we
test a few assumptions and predictions of these models: (1) landscape context does not affect an island's
conservation value, (2) native bird species richness is greater in larger islands, and (3) native bird productivit y
w ithin habitat islands decreases w ith increasing abundance of urban-adapted predators. We quantif ied the
urb an spatial p attern s urr oun din g f ores ts in the Seat tle, WA , m etrop oli tan area u sing a lan d co ver GIS lay er
classified from a satellite image. Within the same forests, w e surveyed bird communities using point counts and
estimated the productivit y of ground and shrub nesting species by monitoring real and artif icial nests and
trapping juveniles. Bird species richness varied only w ithin small islands and depended on the interaction of
habitat island size and the percent of the landscape that w as urbanized. The abundance of American Crows
and rats was unrelated to mult iple measures of nest success. Instead, nest depredation w as more intense in
less urban landscapes with int act native predator communities. We present an improved model of urban bird
community composition based on this information.

87      Francis & Bart
Quantitative goals for avian monitoring programs. CHARLES M. FRANCIS, Bird Studies Canada, Port Rowan,
ON, and JONATHAN BART, Snake River Field Station, USGS, Boise, ID.
         Avian monitoring programs can be used for evaluating management actions, testing hypot heses about
ecosystem responses, and identifying species in need of management action or for w hich current management
is inadeq uat e. W e suggest that , f or t he last of these purp oses, ef fort s shoul d be m ade t o moni tor al l avian
species to a minimum standard to ensure that no species "f all through the cracks." We suggest that a suitable
minimum standard is that an annual 3% decline (equivalent to a 50% decline over 20 y r) can be detected wit h
an 80% probability at a significance level (" ) of 0. 15 on a 2-t ailed test, aft er incorporating potential effects of
bias. For popul ations b reeding w ith in Canada and the U.S., the mo nit oring pr ogram shoul d sample at least b of
the species range. The effects of bias need to be included in setting an accuracy target, because many
widespread surveys in North America are potentially biased due to limitations in the sampling methods (e.g.,
roadside sampling, observer effects, etc .). For the Breeding Bird Survey (BBS), an upper limit for bias of
1% /year seems reasonable at present. Using these standards, species are adequately monitored, at the
rang ew ide l evel, b y t he BBS w hen t he st andar d err ors of the t rend est imates are < 0. 9% and b y surv eys that
yield unbiased estimates of t rend when the SE of t he trend is < 1.3 %. At present, t his accuracy target is
achieved for about a third of landbirds, most waterfowl, but < 10% of other waterbirds and shorebirds.
Implement ing t he monit oring pr ograms pro posed by t he 4 avian ini tiat ives under NA BCI (wat erfow l, w aterbird s,
shorebirds, landbirds) should lead to meeting the accuracy standard for most species breeding wit hin Canada
and the U.S.

88      Wells, Robertson, Rosenberg & Mehlman
Looking in or looking out: A comparison of state and national bird conservation priorities. JEFFREY V.
WELLS, National A udubon Soc ., Ithaca, NY, BRUCE ROBERTSON, KENNETH V. ROSENBERG, Cornell Lab.
Ornit hol., Ithaca NY, and DAVID W. MEHLMAN, The Nature Conservancy , A lbuquerqu e, NM.
          Endangered/threatened (E/T) species listings at both national and state levels have been the standard
paradigm used for the prioritization of species for conservation action in the U.S. At t he state level, lack of
formal among-state standardization of criteria for developing such lists may have important implications for
develop ment of con serv ation pri ori ties. To i nv est igat e this question , w e com pared E/T li st s dev elop ed by st ate
agencies wit h lists derived from a standardized national prioritization system developed by Partners in Flight
(PIF). Most states (90% ) had an off icial E/T species list; 7 8% of stat es also had a list of additional species of
spec ial c onc ern. 51 % of st ates w it h E/T lists of fer sp ecif ic l egal p rot ect ion s f or en dang ered s peci es t hat
exceeded those provided through the federal Migratory Bird Treaty Act . Of t he 933 species that occur w ithin
the continental U.S., 35% are listed as endangered, threatened, or special concern in 1 or more U.S. states.
State list s w ere dominated by species that w ere typically uncommo n or peripheral in a state and that the PIF
system ranked as moderate or low continental concern. In contrast, species ranking highest in a state by t he
PIF system exhibited rangewide conservation vulnerability and w ere typically those for w hich a state could
cont rib ut e great ly to t heir con serv ation . Ov erall , t he t w o sy st ems o ver lapp ed li ttle in the species they
identified, and both approaches may be necessary to conserve the full range of species and habitat diversity
w ith in st ates.

89      Van Riper, Van Riper & Hansen
The impact of avian pox on the conservation of native Hawaiian birds. CHARLES VAN RIPER III* , USGS,
Flagstaff, AZ, SANDRA G. VAN RIPER, Northern Arizona Univ., Flagstaff, AZ, and WA LLACE R. HANSEN,
USGS, M adison, WI.
          At 16 st udy sites, from sea level to tree line in mesic and xeric habitats on the island of Haw aii, birds
w ere captured from 1 977 - 198 0 and examined to determine the prevalence and altitudinal distribution of
individuals infected with avian pox. Infected birds from the wild w ere brought into the laboratory in order to
determine the course of infection. Isolates from lesions were cultured in the laboratory for positive
identification of Poxvirus avium. We also documented distributions and activit y cycles of pot ential avian pox
vectors. We found avian pox from sea level to tree line, concentrated in the mid elevational ranges in the
ecotonal area w here vectors and native birds had the greatest overlap. Native forest birds were: a) more
susceptible to infection than w ere introduced species; b) most likely to be infected during the w et season; and,
c) f oun d t o hav e a hig her p rev alenc e of avian po x inf ect ion s in mesi c w hen c ompared to xeric fores ts. Avian
pox probably did not reach epizootic proportions on Hawaii until aft er introduction of the mosquito and
domestic birds in the early 18 00s, and since then has had a negative impact on the population dynamics of
native forest birds. Today, t his introduced disease is an important f actor that should be considered in
conservat ion eff ort s that are directed at the recov ery of nativ e Hawaiian bird s.

90      Moyer, Gardiner & Clayton
Impact of f eather molt on ectoparasites: Looks can be deceiving. BRETT R. MOYER, DAVID W. GARDINER and
DALE H. CLAYTON, Dept. Biol., Univ. Utah, Salt Lake City, UT.
        Conventional wisdom has it that feather molt reduces the abundance of harmful ectoparasites on birds.
We tested this hypothesis by experimentally manipulating molt in Rock doves infested with f eather lice
(Phthiraptera: Ischnocera). We used 2 standard methods, v isual examination and body washing, to quantif y the
abundance of lice on the birds. The visual data indicated a significant effect of molt on lice. How ever, the
more robu st bod y w ashing m ethod show ed t hat molt had n o ef fect on l ouse abun danc e. Tw o f act ors caus ed
visual examination to underestimate the number of lice on molting birds. First, molt replaces worn f eathers
w ith new , lush plumage that obscures lice during visual examination. Second, w e discovered that lice actively
seek refuge inside the sheath t hat encases developing feathers, where they cannot be seen. The apparent
reduction in louse abundance caused by these factors may, in part, explain the conventional w isdom that molt
redu ces ec topar asit e abun danc e.
91      Altizer, Hochachka, Hartup, Dobson & Dhondt
Being yo ur ow n w orst enemy: House Finch natu ral histo ry and t he persist ence of a disease. SONIA ALTIZER,
WESLEY M. HOCHACHKA* , Cornell Lab. Ornit hol., Ithaca, NY; BARRY HARTUP, School Vet. Med., Univ.
Wisc., Madison, WI; ANDREW DOBSON, Ecol. Evol. Biol., Princeton Univ., Princeton, NJ; and ANDRÉA.
DHONDT, Cornell Lab. Ornithol.
          A disease, my coplasmal c onj unc tivit is, larg ely dri ves the cu rrent pop ulat ion dynamics of House Finc hes
in e. Nort h Am erica. Not only h ad the disease decreased House Finch po pulati ons by u p to 60 % i n some areas,
but House Finch populations have not recovered and the disease has persisted at relatively high prevalence.
There are consistent peaks in prevalence of the disease during the fall and late winter months; fall epidemics
are earl ier an d mov e ext reme in sout hern loc ation s. Analy sis of a mat hematical model sugg est s t hat seasonal
variati on in t he natural hist ory o f Hou se Finches is a major cont ribut or t o th e observed host -pathog en dynamics,
bot h on con tin ent-w ide and local scales.

92      Raffel, Ichida, Head & Burtt
UV inac tivation of feather- degr adin g bac teria as one ex plan ation for av ian sunn ing . THOM AS R. RAFFEL* ,
Dept. Zoo l., Ohio W esleyan Uni v. , Delaw are, OH; J ANN M . ICHIDA , Dept. Bot./ Micro., Ohio Wesleyan Univ.;
SARA K. HEAD and EDWARD H. BURTT, Jr., Dept. Zool., Ohio Wesleyan Univ.
         Survival was measured as a funct ion of t he dose of UV-C for spores and vegetative cells of Bacillus
licheniformis strain 1 38 B, a bacterium isolated f rom bir ds and know n to degrade feathers. These
measurements were compared to spores and vegetative cells of B. subtilis strain OW U11 5. B. licheniformis
spores are 3 t o 4 t imes more resist ant t o UV-C t han vegetat ive cells w hereas B. subtilis spores are 10 t o 20
times more resistant t han B. subtilis vegetative cells (Slieman & Nicholson 2000, Appl. Environ. Micro.
66:199-205). B. subtilis spores were 4 to 6 t imes more resistant to UV-C than B. licheniformis spores.
Furthermore, B. licheniformis w as incapable of phot oreactivation. The sensitivit y of B. licheniformis to UV light
and its inability to phot oreactivate suggests that t his bacterium would also be susceptible to solar UV, and
supports the hypot hesis that avian sunning behavior funct ions to reduce the load of pot entially harmful,
feather- degr adin g bac teria i n t he pl umage.

93       Farmer, Hill & Roberts
Host range of the House Finch strain of Mycoplasma gallisepticum. KRISTY L. FARM ER* , GEOFFERY E. HILL
and SHARON R. ROBERTS, Dept . Biol. , A ubu rn Univ ., Aubu rn, AL.
        Since 1994, an epidemic of mycoplasmal conjunctivitis has spread through t he eastern population of
the House Fin ch, l eadin g t o a signi fican t redu ct ion in t his pop ulat ion . Bacteria i solated f rom lesions of inf ect ed
finches w ere determined to be a unique strain of Mycoplasma gallisepticum (MG). MG, a longstanding and
important pathogen in the poultry industry, had not been previously known t o infect passerine birds. Since
199 4, my coplasmal conjunctivit is has been seen in several other passerine species, most commonly in
American Goldfinches. We tested the host range of the House Finch MG strain by experimentally infecting
House Finches, American Goldfinches, House Sparrows, Pine Siskins, Chipping Sparrow s, Budgerigars, and
Zebra Finches. Follow ing ocu lar inocul ation w ith an MG cult ure the bir ds w ere monit ored fo r clini cal disease
and MG inf ection . M G w as detected by Polymerase Chain Reaction in trach eal swabs f rom House Finches,
American Goldfinches, Pine Siskins, Zebra Finches, and House Sparrows. How ever, clinical disease was only
observ ed in the House Fin ch, A merican Gold finc h, and Pin e Sisk in. Our resu lt s indic ate that carduel ine f inc hes
are more suscepti ble to i nfect ion by the House Finch M G strain t han oth er passerine species. Our result s also
indicate that House Sparrows may serve as an asymptomatic carrier.

94       Johnson, Adams & Clayton
When do parasites fail to speciate in response to host speciation? - dove lice. KEVIN P. JOHNSON* , Illinois
Nat. Hist . Surv. , Champaign, IL; RICHARD J. ADAMS and DALE H. CLAYTON, Dept. Biol., Univ. Utah, Salt
Lake City, UT.
        Several evolutionary processes can influence the degree to which phy logenies of hosts and their
parasites are congru ent. Cospeciation i s likely t o increase the simi larity betw een host and parasit e phylogenies,
while extinction, parasite duplication, and host switching tend to reduce the similarity betw een phylogenies.
Anot her p roc ess, not as w idel y ex plo red i n ho st -par asit e cop hy log enet ic s tudi es, is f ailu re of the par asit e to
speciate w hen the host speciates. Failure to speciate is also likely to reduce congruence between host and
parasite phylogenies. We explored the relative frequency of cophylogenetic processes in avian lice by studying
phylogenies of doves and their parasitic lice. Using comparisons of t erminal taxa, w e found evidence for all of
the processes. Failure of the parasite to speciate when the host speciates appears to be common. The ability
of parasites to disperse and establish within and among host species has an important inf luence on the various
cophy logenetic processes.

95       Al-Tamimi
* Hangin g on: Is w ing l ouse insert ion b ehavior an ad apt ati on t o remain at tac hed t o a f lyi ng host ? SARAH
AL-TAM IMI, Dept. Biol., Univ. Utah, Salt Lake City, UT.
          Wing lice, as their name suggests, spend a considerable amount of t ime on the flight f eathers of t heir
host w here they are exposed to direct airflow and sharp w ing movements during flight. Wing lice on pigeons
and doves (Columbiformes) have a unique adaptation t hat may allow t hem to remain att ached during flight;
they can i nser t their ent ire b odi es bet w een adj acent barb s of flig ht feathers . I experim ent ally manipu lat ed
insert ion abil it y of w ing lic e (Colu mbico la co lum bae) from Rock Doves, and found t hat insertion significantly
improves the ability of C. c olu mbae to remain attached to a flying bird. Insertion requires louse morphology
mat ch the size of the ho st ' s interbar b space. Lice l arger than t he in terbar b space are unab le t o in sert and m ay
be dislodged during flight. This strong selection on louse size may also maintain host specificity by preventing
lice from sw itching t o foreign host species of dif ferent sizes. I measured the survival of C. c olu mbae on
feathers from 6 dif ferent host species ranging from t he Common Ground-Dove (30 g) t o the Victoria
Crowned-Pigeon (Goura victoria; 2 00 0 g).

96       Weckstein
* Cophylogenetic analysis of toucans (Aves: Piciformes) and their ischnoceran chewing lice: Ar e host and
parasite phylogenies congruent? JASON D. WECKSTEIN, Dept . Biol. Sci. an d M us. Nat . Sc i. , Lo uisiana St ate
Univ. , Bat on Rou ge, LA.
         Historically, biologists assumed that cospeciation, rather than ecological interactions and adaptation,
w as the most important f actor struct uring host-parasite assemblages. Comparisons of host and parasite
phylo genies have also concent rated on c ospeciatio n. How ever, many of t hese comparisons hav e also
demonstrat ed t hat the ph ylogen ies of hosts and paras it es are seldo m complet ely con gru ent , sugg est ing that
phen omena ot her t han c ospeciat ion play an im por tant rol e in t he evolu tion of host-par asit e assemblag es. Other
coevolutionary phenomena, such as host swit ching, parasite duplication (speciation on the host), sorting
(extinction), and failure to speciate can also structure host-parasite assemblages. Using mtDNA sequences, I
reconstructed phylogenies of Ramphastos toucans and their ectoparasitic chewing lice in the genus
Austrophilopt erus. I then compared these phylogenies to reconstruct the history of coevolutionary events in
this host-parasite assemblage. Two salient findings emerge. First, t he phylogeny of Ramphastos toucans is not
concordant w ith t he phylogeny of Austrophilopt erus lice. Second, geographic distribution of the hosts seems
to pl ay a r ole i n t he speciat ion of these lice. For ex ample, lic e from distant ly relat ed sy mpatric touc an species
form a monophyletic group. I analyze and discuss the possible causes and implications of the lack of
concor dance betw een these host and p arasite phyl ogenies.

97       Fernandez & Lank
Non-breeding territoriality of Western Sandpipers. GUILLERMO FERNANDEZ and DAVID B. LANK, Centre Wild.
Ecolog y, Dept . Bio l. Scien ces, Simo n Fraser Univ ., Burnaby, BC.
           Bahia Santa Maria, on the coast of w . Mexico, supports over 400 ,00 0 w intering Western Sandpipers in
a diverse mosaic of fresh and saltw ater habitats. We found diff erent bird densities, adult and male ratios, and
territ ori al beh avior am ong habi tats. Start ing at the shorel ine, beach es at the edg e of cat tail m arsh es suppo rt ed
a density of 50 birds/ ha, 64% adults and 57% m ales, and no territorial birds. Moving t ow ards the ocean,
lagoon and mudflat int erspersed wit h mangrove and brackish vegetation supported 30 3 birds/ha, 84 % adults
and 71% males, and 2% territorial birds. Further tow ards the ocean, open brackish lagoon flats, w ith a density
of 57 5 b ird s/ha, suppor t 90 % adul ts, 80 % males , an d 6 % territ ori al bi rds . Birds on t he shorel ine h ad bi gger
flo ck size, hig her feeding, and smaller w alking rat es than bird s on ot hers habitat s. Territ orial birds sp ent less
time foraging, had smaller flock sizes, higher walking rates, spent more time in vigilance and aggression, and
fed more visually than non-territorial birds. 80% of t erritorial birds were male, compared wit h 73% overall, but
territ ori al bi rds ' behav ior did not dif fer bet w een t he sex es or w it h hab it at typ e. The po pul ation st ruc ture w e
have doc ument ed may r elat ed t o in terf erenc e behav ior or c ond it ion depen dent habi tat use, and sugg est s t hat
individuals pursue a facultative strategy w ith respect t o non-breeding space use and territorial behavior, which
may provide benefits in coping with habitat attributes and bird density.

98       Morton
Temperate zone bias and territorial systems in birds. EUGENE S. MORTON, Smithsonian Inst., Washington,
         Over 90% of North A merican passerines have a similar territorial system, they defend breeding
territories for only a few months each summer. In the tropics, only 13 % of passerines defend territories only
for b reedi ng. Instead, the pr edom inan t territ ori al sy st em is year -ro und def ense of feedin g and nest ing territ ori es
coupled wit h perennial pair bonding (Stutchbury & Mort on 200 1, Behavioral ecology of tropical birds). I
discuss important variations in the permanent territorial systems in tropical birds, the need for more study, and
ramifications of t he unjustifiable emphasis on breeding territoriality for our understanding of avian evolution.

99       Fedy & Stutchbury
The variable terri tor ial syst em of a resident tro pical passerine, t he Whit e-bellied Ant bird. BRADLEY C. FEDY
and BRIDGET J. M. STUTCHBURY, Dept. Biol., York Univ ., Nort h York, ON.
          We stud ied the t errit orial behaviour of a resident populat ion of Whit e-bellied Ant birds (Myr meci za
lon gip es) in low land seco ndar y f ores ts in cent ral Panam a. Su perf ici ally this spec ies seems t o hav e a territ ori al
system similar to many other resident t ropical insectivores characterized by year-round territory defence and
lon g-t erm pair bon ds. How ever , o ur s tudy has r evealed a mu ch more complex and v ariab le sy st em t han
previou sly suspect ed. One radio-t racked male w ho w as mated and defended t he same territ ory f or at least a
year nevertheless made long and silent forays off territory during the non-breeding season. A t otal of 3 6
territories were monitored over 2 yr. Removal experiments w ere conducted on 17 of t hese territories over 2
non-breeding periods and included both male and female removals. In 7 cases, radio transmitters w ere
attached to bo th members o f a pair to pr ov ide d etailed dat a on t errit ory use before, dur ing and af ter removal
experiments. Data w ere collected on song rates, individual movements in relation t o territory boundaries and
mates, replacement rates, conspecific t erritory disputes and behaviour after release. We also radio-tagged and
monit ored th e behaviour of 6 ' flo ater' indiv iduals.

100      Leu
* Behavioral and physiological factors influencing breeding territory sett lement patterns in male Pacific-slope
Flycatchers. MATTHIAS LEU, Coll. For. Res., Univ. Washington, WA.
         Pacific-slope Flycatchers follow a despotic distribut ion on the breeding grounds. Males arrive and pair
earliest in red alder (Alnus ruba) habit at w here r epro duc tive out put is high est . In t his st udy I investigat ed
w hether intense competition occurs among males establishing territories in this high-quality habitat. During the
territorial phase (i.e., before pairing occurred), I assessed territorial aggression both at the behavioral and
physiological level. At the behavioral level I found that males exhibit higher territorial defense song rates after a
sim ulat ed t errit ori al ch allen ge co mpared t o t he pr e-ch allen ge ph ase. How ever , t he spont aneou s t errit ori al
def ense song rat e did not dif fer among habi tat typ es. Furt herm ore, resp onse time to simulat ed t errit ori al
intrusions did not dif fer among habitat types. At the physiological level I found that plasma testosterone levels
did not differ among habitat types. These data suggest that high-quality habitat is settled preemptively.
Over all, arrival t ime on t he br eedin g gr oun ds determines territ ory qual it y. Males in t op c ond it ion arrive earli er
on the breeding grounds; body condition indexes (body w eight adjusted for body size) correlated negatively
w ith arrival time. As show n elsew here, these data suggest that conditions on the w intering grounds or during
migration may be important in determining male arrival time and therefore settling patterns on the breeding
ground s.

101     Janota
* Experimental manipu lation of m aternal cond iti on aff ects of fspr ing sex rat io in t he House Wren. SHA NNON M.
JANOTA, Dept. Biol. Sci. , Illinois St ate Univ., Normal, IL.
          Nat ural selection should favor f emales that adju st of fspr ing sex rat ios accord ing to t he po tent ial
reproductive success of male and female offspring under existing ecological conditions. In socially and
genetically polygynous mating syst ems, male off spring in good condition should have higher reproductive value
than females, which should have higher reproductive value than males in poor condition. Condition should
affect a female's ability t o produce offspring in good condition, so females should bias offspring sex ratios
according to their ow n condition. This hypot hesis can be tested unambiguously in polygynous, monomorphic
species, such as the House Wren. If females in good condition are more likely to produce nestlings in good
condi tion , t hen t he sex rat io o f bro ods pro duc ed by females in g ood con dit ion should be male-biased, w hereas
those pro duc ed by females in p oor con dit ion should be f emale- biased. I experi mentally manipu lat ed maternal
condition by clipping flight feathers to increase energetic costs of flight. Experimental females (clipped) were
signif icantl y ligh ter t han cont rol f emales, and experiment al females produc ed light er eggs than co ntr ol f emales.
Control females produced more male offspring than did experimental females. Thus, female House Wrens
apparently adjust off spring sex ratio according to t heir condition.

102     Eckerle
* Female House Wrens may not be very choosy when selecting a mate. KEVIN P. ECKERLE, Dept. Biol.
Sciences , Illi noi s St ate Univ. , Norm al, IL.
         In birds su ch as house w rens in w hich males occ upy and def end all-purpose breeding t errit ories,
females are predicted to select m ates based on traits indi cating m ale quality or t erritory quality , or bot h. In
house wrens, as in many migratory songbirds, measures of male and territory quality are correlated; the
earliest-arriving males are older and larger than later-arriving males, and early-arriving males establish breeding
territories in preferred vegetation. The earliest-arriving males are also the first t o acquire a mate, as the time of
male ar riv al and female s ettlemen t are po sit ively cor relat ed. Alt hou gh t his cor relat ion suggest s t hat females
prefer early-arriving males, covariation betw een measures of male and territory quality make it diff icult t o
det ermine t he t rait s used by females to select mat es. Ther efore, I design ed a series of experim ent s t hat
disrupted the relationship betw een male and territory quality to t est the hypotheses that female house wrens
select mates on the basis of male body condition, song rate, use of arthropod egg sacs during nest building,
territory vegetation, or quantity of nest cavities. The results of these studies demonstrate that f emales do not
prefer males that are in best condition, sing at the highest rate, use arthropod egg sacs during nest building, or
occupy territ ories that are in preferred vegetation or have the most nest sites. Although f emale house wrens
may select m ates on t raits t hat w ere either unmeasured or un manipulat ed (e.g., male song reperto ire size, nest
cavity quality), I propose that females are not very choosy w hen selecting a mate and maximize their
reproductive success by mating quickly because the costs of mat e searching and delayed breeding outw eigh
the benefits of being choosy.
103     Yaber & Rabenold
Sociality and dispersal in tropical wrens. M. CAROLINA YABER, Dept. Biol., Hampden-Sydney Coll.,
Hampden-Sydney, VA; and KERRY N. RABENOLD, Purdue Univ., West. Lafayett e, IN.
          Strip e-backed Wrens (Campylorhynchus nuchalis) are cooperatively breeding birds, in w hich young
individuals delay breeding and natal dispersal. Five marked populations w ith study histories of 6 - 21 yr make it
possible t o det ermine h ow soc iali ty inf luen ces t he pat tern o f int ra- and in ter-p opu lat ion dispers al f or f emales
and m ales; and ev aluat e the in fluen ce of dispers al in the viabi lit y of pop ulat ion s. Disp ersal in St rip e-bac ked
Wrens is female-biased; reproductive success of female dispersers is higher than for male dispersers, and
short-distance dispersal is favored by an advantage of proximit y in competition, particularly in productive large
gro ups. M ales have similar fit ness (measured as t he mean nu mber of juv enil e firs t-or der r elat ives pr odu ced
annu ally ) w het her t hey breed in an ot her g rou p or help in t heir nat al t errit ory . M ales t hat dispers e have low er
reproduc tiv e success than f emales because they are mainly lim ited t o breeding in pairs. Females, in c ont rast,
gain breeding positions almost exclusively through dispersal. Male and female betw een-population dispersers
had low reproductive success, and such long-distance dispersal results in a very low level of effective
inter-population dispersal. Population dynamics are variable and likely independent among populations.
Survival, reproduction and dispersal appear not t o be density-dependent, although breeding females in 2
pop ulat ion s at low densit ies c onsist ed mainl y of immigran ts f rom ot her p opu lat ion s. In spit e of soc ial
constraints on dispersal, and the handicap of small effective population size imposed by sociality, sets of
int eracting populat ions in f ragmented landsc apes may be less extinc tio n-prone t han similar isolat ed populat ions.

104     Robinson
What w omen want: Song Wren females benefit from polyandrous matings. TARA R. ROBINSON, Dept. Biol.
Sci., Aubu rn Univ ., Aubu rn, AL.
          I stud ied Song Wrens (Cyphorhinus phaeocephalus), year-round residents of humid low land forests of
Panama, from 19 94 to 2 00 0. Breeding p airs repr odu ce on ce or tw ice/ yr raising 1 - 3 y oun g/ attempt dur ing a 7
mon breeding season, w hich overlaps the period of highest annual rainfall. Song Wrens live in groups
throughout the year, but cooperative breeding is rare. Breeding females have a significantly higher rate of
breeding dispersal than do males (60% vs. 28 %, respectively). Females frequently abandoned males w ith
w hom they had b red s uccessfull y in pr evious year s. Due t o t he rel atively hig h rat e of female- ini tiat ed di vo rce,
step families consisting of a male, his son, and an unrelated female formed during every year of the study.
Both males in such trios gave parental care to of fspring produced. For all years together, trios had significantly
hig her r epro duc tive suc cess than d id u naid ed pai rs (2. 0 v s. 1. 2; mean fledg lin gs/ pair ). During the El
Niño-Southern Oscillation (ENSO) event of 1997 - 1998, this difference was particularly striking (2.0 vs. 0.85).
I hypothesize that females benefit from polyandrous matings in tw o w ays. First, she gains additional help in
raising her offspring. This help is particularly valuable during unusually dry years such as the recent ENSO.
Second, a female likely gains increased genetic diversity f or her clutch by mating with 2 males. Preliminary
results f rom genot yping indicat e that fat hers and sons share paternit y w ith unrelated f emale mates.

105      Hackett & Bates
Exploring genetic diversification in birds from two tropical systems, Amazonia and Madagascar. SHANNON J.
HACKETT* and JOHN M. BATES, Field Mus., Chicago, IL.
          Patt erns and levels of genetic diversity w ithin and among populations are important indicators of
historical connections among populations, current gene flow , historical population sizes, and changes in historic
population size. Previous studies have demonstrated high levels of genetic diff erentiation among populations of
Am azonian forest birds. We use sequence data fro m mit ochond rial genes and modern analyt ical approaches t o
compare pat terns of diversif icatio n in A mazonian and Malagasy f orest bi rds. In cont rast t o Am azonian birds,
w e present data that indicate little genetic differentiation among populations of forest birds on M adagascar and
implic ate relativ ely recent populat ion incr eases and range expansions f or some species. Our results su ggest
that dif ferent time peri ods and p ossibl y dif ferent pro cesses hav e shaped gen etic d iversity in t hese t w o t rop ical

106        Reddy, Cibois, Feinstein & Cracraft
Shrike-babblers (Pteruthius: Ti malii dae) are corvidans, not passeri dans. SUSHMA REDDY* , A LICE CIBOIS,
JULIE FEINSTEIN and JOEL CRACRAFT, Dept. Ornit h., Am . M us. Nat . Hist ., New York, NY.
          The avian famil y Timalii dae, an old-w orld gr oup of babblers, has hist orically been used as a
“ scrap-basket” to place species that w ere hard to classify (Mayr & Amadon 19 51, Am. Novit . 14 96). The
shrike-babblers, genus Pteruthius, consist of a small group of 5 species and about 21 subspecies that are
endemic to Southeast Asia. This genus has traditionally been placed within t he Timaliidae, how ever a recent
int ergen eric st udy of the f amil y indi cat ed t hat this might not be co rrect . A syst emat ic s tudy of this genu s w as
undertaken in order to t est its monophyly, find it s closest relatives, and analyze the biogeographic implications
of t heir relationships. All the proposed basal taxonomic units w ere investigated by examining specimens for
diag nosable ex ternal char act ers as w ell as sequenci ng t he co mplet e cy tochro me b gene of mi toc hondri al DNA
from availabl e tissue and ski n sam ples. Phy log enet ic an alyses based on cy tochro me b of all subspecies of
Pteruthius and o ther t imalii ds, as w ell as repr esent atives of many other p asserine f amil ies, ind icat e that
Pteruthius is monophyletic and does not group w ith members of the Timaliidae. Instead, they are deeply
imbedd ed w it hin the co rv ines. How ever , c yt ochro me b is not capable of identif ying the specific sister group of
Pteruthius, t heref ore t heir phy log enet ic p lacem ent w it hin cor vidans w ill be f urt her i nv est igat ed using nuc lear
(RAG2) sequences.

107        Eaton & Lanyon
Ultr aviolet plumage refl ectance: A survey of b irds of the w orld. MUIR D. EATON* and SCOTT M. LA NYON,
Dept. Ecol. Evol. & Behav., Univ. Minnesota, St. Paul, MN.
           Alt hough ul trav iolet p lumage reflect ance has received considerable att ention in t he literat ure of t he last
10 yr , l it tle is kno w n abo ut it s per vasiveness w it hin the cl ass A ves. A bro ad t axonom ic s urv ey (32 5 speci es
representing all avian families) w as conducted to assess the overall distribution of UV reflectance, the
frequ ency of sexual dimor phism in UV ref lectance, as w ell as correlatio ns w ith body r egions and visu al colors.

108        Porzecanski
Historical biogeography of the South A merican aridlands: Preliminary hypot heses based on phylogenies for f our
avian groups. ANA LUZ PORZECANSKI, Am . M us. Nat . Hist . and Columb ia Univ. , New York, NY.
          A nu mber of biogeographic hypot heses have been proposed to exp lain diversif icatio n in t he w et f orests
of South America, but the endemic aridland biota has seldom been studied. Although endemic bird taxa have
been t abul ated and areas of endem ism postulat ed, histori cal an alysis of the ari dlan d end emic avifauna h as been
hindered by the lack of phylogenetic hypot heses for t he taxa involved. In this study, species limits and
phy log enet ic r elat ion ships w it hin 4 av ian g rou ps w it h rep resen tatives endemic to t he Neotrop ical arid areas
w ere in vestigat ed using ext ernal morph olo gy and m olec ular char act ers (cy tochro me b gene). In the tinamou
gener a Nothura and Nothoprocta, m ost desc rib ed subspecies w ere no t foun d t o be d iagn osably distinc t based
on ex ternal morph olo gy . The ph ylogen etic r elat ion ships of their con st it uent spec ies sugg est con tradi ct ory set s
of area-relationships, indicating complex biogeographic histories for these taxa. In the passerine genus of
cardinals Paroaria and the woodcreeper species complex Lepidocolaptes angustirostris, at least half of the
desc rib ed subspecies w ere f oun d t o be d iagn osable. Thei r ph ylogen etic r elat ion ships also indi cat e com plex
biogeographic histories for these taxa. These results suggest that it is unlikely that a single set of relationships
exist s f or t he ari d areas of endem ism , an d t hat div ersificat ion of the end emic avifauna h as a mo re complex
history than what is suggested by their congruent distribution patterns.

109        Braun & Huddleston
Molecular phylogenetics of caprimulgiform nightbirds. MICHAEL J. BRAUN* and CHRISTOPHER J.
HUDDLESTON, Smith sonian Inst ., Washingt on, DC.
         The Caprimulgiformes comprises 5 bird families adapted to noct urnal activit y. They are morphologically
similar in their cryptic plumage colors and patterns and have generally been regarded as monophyletic, w ith
ow ls, sw if ts and hum mingb ird s as t heir clo sest relat ives. How ever , DNA hybri dizat ion and i sozy me data
indicat e that there are many ancient lineages in the gro up, and c ertain mo rpholo gical charact ers suggest
affinities betw een some of the families and other groups of birds. To estimate the phylogeny of the group, we
obt ained more t han 2 00 0 b p of DNA sequence f rom tw o un lin ked g enes (cy tochro me b, 11 43 bp; c-myc,
~1 06 0 b p) f rom 19 ing rou p and 5 o ut gro up t axa, inc lud ing all speci es of pot oos (Ny ct ibi idae). Deep
divergences exist both betw een and wit hin families. A variety of analyses all strongly conf irm monophyly of
each of the 5 traditional families. However, relationships among families are poorly resolved and a 4 codon
insertion in c-myc linking ow let-nightjars wit h sw ift s and hummingbirds suggests the possibility that t he
caprimulgiforms are non-monophyletic. A ll nodes within t he Nyctibiidae are reasonably well resolved, wit h
cy tochro me b contribut ing more information at the tips and c-myc more at the basal nodes. The
morphologically divergent Rufous Potoo (Nyctibius bracteatus) branches first, followed by N. grandis, then N.
aethereus. The remaining 4 species form a terminal clade with leucopterus sister to maculosus and griseus
sister to jamaicensis.

110     Barker, Barrowclough & Groth
Passerine phylogeny from nuclear DNA sequences: Taxonomic and biogeographic implications. F. KEITH
BARKER, GEORGE F. BARROWCLOUGH and JEFF G. GROTH, Dept. Ornithol., Am. M us. Nat. Hist., New York,
        Passerine bi rds com pri se ov er hal f of avian di versit y, but have pro ven dif ficu lt to cl assify. Despi te a
long history of w ork on this group, no comprehensive hypothesis of passerine relationships w as available until
recent analyses of DNA-DNA hybridization data. Unfortunately given the value of such a hypothesis in
comparat ive st udi es of passerin e ecol ogy and b ehavior , f ew st atements of relat ion ship pr opo sed have been
tested w ith independent data and analytical approaches. We analyzed nucleotide sequence variation at t he
nuclear RAG-1 and c-mo s genes, fr om 6 9 passerine t axa and 3 out groups. The sampled taxa represent most
currently recognized families of passerine birds. Our analyses suggest paraphyly of the suboscines, w ith t he
New Zealand w ren Acanthisitta sister to all other passerines. We also found support for paraphyly of t he
parvorder Corvida, as well as a number of other relationships which conflict w ith previous analyses. We
reconstruct the Corvida as a basal grade within the oscines, the earliest divergences within which involve
endemic Aust ralo-Papuan lineages. This arrangement suggests an origin of the suborder Passeri on the
Australian continental plate during its long period of isolation, w ith a major radiation of oscines (e.g., t he
parvorder Passerida) subsequent t o the reduction of oceanic wat er barriers due to continental drift .

111     Bassett
* Long distance migrants breeding in hardwood and pine plantations in southeastern Louisiana: Emphasis on
species of management concern. C. AUDRA BASSETT, School For. & Wood Products, Michigan Tech. Univ.,
Hought on, M I.
        I sampled the breeding birds of hardw oods and 4 different age classes and management histories of
even-age loblolly p ine plantat ions (Pinus taeda) in Tangipahoa Parish, Louisiana. Birds were sampled w ith 10
min, 25 m fixed radius point counts in 5 habitat s, and mist netting w as conducted in 4 habitats to provide
physical evidence of breeding. 26 species of long distance migrants were detected over the 2-yr study.
Breeding w as co nf irm ed or pro babl e for 1 5 o f these speci es. Prairie Warbl ers an d Yellow -br east ed Chat s br ed
in young pine plantations. Worm -eating Warblers bred only in unthinned mid-age pine plantations, w hile
Swainson's Warblers and Wood Thrushes bred in unthinned mid-age pine and hardwood habitat. Kentucky
Warblers and Hooded Warblers bred within hardwoods and pine plantations. Acadian Flycatchers and Northern
Parulas bred only in hardwoods. Hardwoods provided breeding habitat for more species of long distance
migran ts t han p ine p lant ation s. How ever , t he availab ili ty of com mercial ly managed p ine p lant ation s in the se.
U.S. provides breeding habitat for several long distance migrants including some species of management
concern. To m anage f or s ong bir d di ver sit y large s tands of dif ferent ages and successional st ages are needed
w it hin a land scap e.

112     Stouffer
Geographical v ariat ion in t errestrial insect ivoro us bird com mun it ies in Neot rop ical rainforests. PHILIP C.
STOUFFER, Dept . Biol. Sci., Sout heast ern Lo uisiana Univ ., Hamm ond , LA .
         Our earlier analysis of the avifauna of lowland terra firme rainforest north of Manaus, Brazil, showed the
species richness of for est bird s to b e comparable to sites in t he Peruvian A mazon. This i s int eresting because
the central Amazon is generally considered to be less productive than the w. Amazon, due mostly to
diff erences in soil fertility. Here I compare species richness, density, and biomass of ground foraging birds
among t he few Neotrop ical sit es w ith density data, in cluding Manaus, Peru (M anu), French Guiana (Nouragues),
Ecuador (Yasuni), and Panama (Soberania). Amazonian sites all had 11 - 1 3 species. Panama had 7 permanent
resident species and 4 migrants. Density varied considerably among sites. In Panama, there were nearly 100
resident pairs/100 ha. The Amazonian sites were all lower, with a low of about 50 pairs/100 ha in Manaus.
Biomass follow ed the same trend, w ith t he Amazonian sites all low er than Panama, and Manaus at about 40%
of the biomass of Panama. Considering species common to at least tw o sites, biomass varied considerably.
Each sit e, even Manaus, had species that w ere at least t w ice as common there t han at ot her sites. These
results support the idea that species richness in Neotropical rainforests may be decoupled from productivity.
Further, regional patterns of abundance vary among species but are unrelated to overall productivity.

113     Leberg, Ferrence, Leonard, V ermillion, Carloss & Linscombe
Use of Indicator Value Analysis for interpreting patterns of species associations in colonies of waterbirds.
PAUL LEBERG, JEANINE FERRENCE, BILLY LEONARD, Dept . Biol. , Univ . Lo uisiana, Lafayet te, LA ; WILLIAM
VERMILLION, MIKE CARLOSS and GREG LINSCOMBE, Loui siana Dept . W ild l. & Fish. , Laf ayet te, LA .
           Alt hough strong associations among species are sometimes evident, variation in the composition of
w aterbirds nesting in mixed-species colonies often defies characterization w ith t raditional approaches. Since its
int roduct ion in 1 99 7, Indicat or Value A nalysis (IV A) has been used not only t o detect indicat or species, bu t also
to characterize the composition of species assemblages. Our objective was to assess the utility of t his
appr oach for i nt erpr eting var iat ion in speci es co mposition associat ed w it h co lon ies of w aterbi rds . W e appl ied
IVA to tw o data sets obtained through helicopter surveys of coastal Louisiana: 263 colonies occupied by 13
spec ies of w adin g bi rds and 1 53 col oni es used by 9 speci es of seabirds . Both no nhi erarchical and h ierar chical
clu st erin g w ere us ed t o cl assify c olo nies. Utili zing inf orm ation on t he rel ative abun danc e of each s peci es, as
w ell as the absence of species from subsets of colonies, IVA indicated that the optimal level of partitioning of
each data set w as obtained by dividing the colonies into 5 assemblages. Some of these assemblages appear to
resu lt from doc ument ed beh avioral int eraction s amo ng speci es; ot hers are associat ed w it h shared h abit at
pref erenc es f or n est sit es. IVA w as also usef ul f or d ist ing uishin g bet w een species that occ urr ed acr oss a bro ad
rang e of col ony typ es and those that w ere charac teristic o f col oni es w it h a nar row rang e of spec ies

114     Haskell, Evans & Pelkey
Bird communities in native forests and industrial pine plantations on the Southern Cumberland Plateau,
Tennessee. DAVID G. HASKELL* , JONATHAN P. EVANS and NEIL W. PELKEY, Dept. Biol., Univ. South,
Sewanee, TN.
        We examined the effects of conv ersion of oak-hickory forests to loblolly pine plantations on the
abundance and diversity of birds on the Southern Cumberland Plateau in Tennessee. We conducted distance
count s of breed ing bir ds in t he sprin g and sum mer of 20 00 and 2 00 1 t o est abli sh t he associat ion bet w een
habitat type and bird populations. Counts were made in native forests and in plantations of different ages.
Data from 200 0 indicate that bird diversity is lower at all spatial scales in pine plantations. For example, pine
plantations eliminate approximately a of the nat ive bir d species and m ost that remain ar e pres ent a low er
densities. A few " early successional" species, how ever, do benefit from t he presence of scrubby vegetation in
young pine stands. Bird abundance was also lower in plantations for most species, except for the " early
successional" species. We combined these data w ith an eighteen-year GIS record of land use changes in the
region t o assess the landscape-level signif icance of smaller-scale changes in comm unit ies.

115     DeSante, Pyle & O' Grady
Population dy namics of landbirds in Denali National Park, AK, and on Fort Hood, TX . DAV ID F. DeSANTE* ,
PETER PYLE, and DANIELLE R. O'GRADY, Inst . Bir d Populat ions, Point Reyes St ati on, CA.
         We monitored the vital rates and population trends of landbirds during 1992 - 200 0 using 4 M onitoring
Avian Productivity and Survivorship (MAPS) stations in Denali National Park, AK, and during 199 4 - 20 00 using
6 M APS stations on Fort Hood, TX. Overall, population sizes w ere fairly stable at both locations, w ith
non-significant annual decreases for all species pooled of 1.7% and 1.4% , respectively, and slightly more
decr easing t han i ncreasing t arget spec ies. Nevert heless, subst ant ial c hang es in pop ulat ion size w ere recor ded
for 3 t arget species at Denali and 5 at Fort Hood. At each location, annual changes in population size for each
target spec ies and all spec ies pool ed w ere po sit ively cor relat ed w it h ann ual c hang es in pro duc tivit y t he year
before. The strength of this " productivit y-population" correlation varied among target species at both locations
in such a manner that spec ies w it h more i nt erann ual v ariat ion in ad ult sur vival rat es had w eaker
productivity-population correlations. Productivity at both locations (and throughout w . and s.-central North
America) w as negatively cor relat ed w it h t he El Niñ o/ Southern Oscill ation Ind ex (SOI), tendi ng t o be h igh er
dur ing El Niño year s (w hic h pr edom inat ed du rin g t he earl y 1 99 0s ) t han La Niña y ears (w hic h pr edom inat ed
during the late 1990 s). Thus, w e predict t hat productivit y should have decreased during the 1990 s throughout
w estern and south-central North America. Productivity did decrease at Denali and virtually all other locations
except Fort Hood, where active management to lessen cowbird parasitism caused increases in productivity.
Despite the fact that annual changes in population size were driven primarily by annual changes in productivity,
overall population t rends for both decreasing and increasing species at bot h locations were driven by low or
high adult survivorship nearly as often as by low or high productivity.

116     Mabb
Seasonal patt erns in parrot roost flocks in the San Gabriel Valley, California: Size, composition, tree usage, and
site location. KAREN MABB, Dept . Bio l., Calif orn ia St ate Polyt ech. Univ ., Pomon a, CA .
         Naturalized parrot s in t he San Gabriel Valley, Californi a, w ere observed roost ing in st reet t rees.
Population numbers, species present, t ree usage, and roost site location w ere recorded on 181 days during a
period from 25 May 19 95 t o 28 J an 2001 . Parrot population numbers w ere highest in the late wint er (the
highest number of 144 7 parrots w as observed on 8 Feb 200 0). The tot al number of dif ferent species using the
roost w as highest in the late w inter (shortly before the roost f lock breaks up and begins breeding activity ) and
in the late summer (shortly aft er the end of breeding activity ). Red-crow ned (Amazona viridigenalis) and
Lilac-crow ned (A. f inschi) parrots comprised the majority of t he roost f lock; 4 other Amazona species were
present at t he roost. Mit red (Aratinga mitrata) and Red-masked (Aratinga erythrogenys) parakeets were the
most f requently occurr ing Aratinga species. Tw o other Aratinga species were also recorded at roost sites. A
flo ck of < 25 Rose-ringed Parakeets (Psittacula krameri) w ere present w ith t he parrot roost flock year-round for
the ent ire study peri od. Dependent juv enil e Red-cr ow ned Parr ot s w ere most num erou s in Sep, and ac count ed
for approximately 20 % of the roost f lock. Parrots preferred deciduous trees for roosts and only roosted in
evergreen trees w hen no ot her suit able alternativ es w ere available. Roosting oc curred in 6 tree species almost
excl usiv ely (3 decid uous, 3 everg reen). An alysis of roost sit es usin g Geographic Inf orm ati on Sy st ems (GIS)
showed t hat roost site location changed seasonally, presumably in response to the tree type or availability of
for aging resources.
117     Hodgens, Blumenshine & Bednarz
Evaluat ion o f p redat ion b y Great Blue Heron s as a sour ce of tr out mor tal it y on th e upper Whi te Riv er. LYNN S.
HODGENS* , STEVE C. BLUMENSHINE and JAMES C. BEDNARZ, Dept . o f Biol., Arkansas State Univ. , St ate
University, AR.
          A pr imary qu estion f or f ishery managers is w hether avian pi scivo ry represent s a large proport ion of fish
mortality relative to angling. M ore than 1.5 million catchable trout (23 - 30 cm) are stocked annually in the Bull
Shoals and Norfork tailw aters on the upper White River in Arkansas. Recent creel surveys conducted by the
Arkansas Game and Fish Commission suggest that angler harvesting accounts for only a small percentage of
trout mortality . Our primary objective is to provide a realistic assessment of the magnitude of Great Blue Heron
predation on the trout population of these tailwaters. Betw een Nov 200 0 and Mar 200 1, w e conducted 20
censuses of Great Blue Herons along 311 t ailwater river km on the White River below Bull Shoals Dam. Heron
densities ranged from 0.2 to 3.9 individuals / river km, but w ere significantly higher near the Dam (2.8
ind iv/k m) v ersu s low er t ailw ater reaches (1.0 ind iv/k m). Taxono mic and size dist rib ut ion dat a on t he 2 67 prey
captures by herons over 96 observation hours demonstrated that sculpin are the most common prey (26.9%).
Live trout represented just 16 captures (6%). The length of captured live trout ranged from 7 to 2 8 cm, w ith a
modal r ange f rom 10 .5 - 14 c m. Of these live trou t capt ures , o nly 6 (37 .5 %) w ere of st andar d " st ockin g size"
(23 - 30 cm) or larger. To date, our data suggest that predation by Great Blue Herons is likely a minor source
of trout mortality.

118      Homberger
Biom echanics o f a predat or-pr ey r elat ionsh ip: A ust ralian cockatoos and eucaly pt fruit s. DOM INIQUE G.
HOMBERGER, Biol. Sc i. , Lo uisiana St ate Univ. , Bat on Rou ge, LA.
         A f resh analysis of the unique structure of t he psittaciform quadrato-mandibular joint show s that t his
joint prevents a protraction of the mandible (in contrast to earlier interpretations), but allows lateral movements.
Paradoxically, most parrots and cockatoos do not move their mandibles laterally, except very slightly during
bill-honing. They extract kernels from f ruits and seeds w ith t he transverse cutt ing edge of t heir mandible by
exerting pressure on weak points of t he fruit w alls. However, certain red-tailed Black-Cockatoos
(Calyptorhynchus lathami, Callocephalon fimbriatum, an d some C. banksii subspecies) regularly move their
mandibles sidew ays to place one of their mandibular corners directly opposite their upper bill tip. They extract
seeds with a pincer-like action, ripping through large woody Eucalyptus and Casuarina fruit s, w hose walls
exhibit a "ply-wood" construction that is part of a complex hydraulic mechanism for releasing mature seeds.
Parrots and other cockatoos (including some other C. banksii subspecies) do not crack open the walls of such
fruit s. The uniform quadrato-mandibular joint and the diversified bills of Psittaciformes, if t ested against the
biomechanical properties and functional morphology of t he food sources and used with observations in the field
und er nat ural con dit ion s, pro vide co rrobo rat ed eviden ce f or t he hy pot hesis t hat the Psit tacif orm es or igi nat ed
from an ancestral form that had evolved the capacity f or moving its mandible sidew ays to rip open woody
materials, pr obably or iginally to extract w ood-bori ng or gallic ole insect larv ae. (Su ppor ted by NSF)

119       ty,
         Bê Gauthier, Korpimäki & Giroux
Shared predators and indi rect t rophic int eractions: l emming cyc les and arctic-nesting g eese. JOË BÊ ,    L TY*
GILLES GAUTHIER, Dé . Bio l. and C.E.N., Univ . Lav al, Qué         bec, PQ, ERKKI KORPIMÄKI, Dept. Biol, Univ. Turku,
Turku, Finland, and JEAN-FRANÇOIS GIROUX, Dé . Sc. Biol., UQA M, Mo nt ré PQ.            al,
         We test ed the hy pot hesis that cycl ic lemmin g populat ions indi rectl y aff ect br eeding Greater Snow Geese
in the High Arctic t hrough the behavioral and numerical responses of shared predators. Behavioral observations
indicate that predators were selective in their prey and increased their hunting effort on goose nests at low
lemming densit y. Arc tic fox es (Alopex lagopus), main nest predators, show ed a numerical response to rodent
densit ies, as t heir breed ing out put w as low est dur ing the lo w phase of the ro dent cy cle. Act ivit y r ates of fox es
and Paras it ic J aegers in t he go ose c olo ny show ed st ron g in ter-annual fluc tuat ion s and w ere gen erall y r educ ed
at high lemming density. The total response of predators varied from 19% to 8 8% of t he total number of eggs
depredated and was cyclic, being lowest during peak lemming years. Peak of egg predation by f oxes and
jaegers occurred w ith a time lag of 2 and 3 yr w ith respect to t he lemming cycle, respectively. Results from
artificial nest experiments w ere consistent w ith t hose obtained on real goose nests and confirmed the
int er-annual var iat ion in n est pred ation pres sur e by dif ferent pred ator s peci es. Thu s, w e con clu de t hat
responses of predators to cyclic lemmings cause strong indirect eff ects on goose nesting productiv ity and
generate both short-term apparent mutualism and long-term apparent competition among prey.

120      Heath
* Predation risk from nesting raptors influence the metapopulation dynamics of breeding Harlequin Ducks in
northern Labrador. JOEL P. HEATH, Biopsychol. Programme, Dept. Biol., and Dept. Psychol., Memorial Univ.
New foundland, St . Johns, NF.
          Landscape considerations are essential for understanding the distribution of many species, especially
highly mobile ones. The structure and dynamics of populations, measured at small spatial scales, may not
reflect the characteristics of the overall population, across the landscape. The application of metapopulation
theor y t o migr atory bir ds has pr ov en in form ative in t his regar d. In e. North A merica, Harlequi n Ducks aggregat e
at coastal sites during winter, but breed on inland rivers that are patchily distributed, w ith large areas of
unsuit able h abit at bet w een w atersh eds. High ret urn rat es t o bo th w int er and sum mer sit es pr ov ide a p ot ent ial
mechani sm f or m etapop ulat ion st ruc ture. We st udi ed po pul ation s of breed ing Harlequi n Ducks on 9 r iver
systems in n. Labrador. Site persistence (reuse of sites w ithin river systems, among years) increased w ith t he
density and stability of local populations, and considering river systems as demographically independent
sub-popul ations st rongly implied sour ce-sink met apopulati on st ruct ure. Wi thi n th is eco-region, predation risk
appears to be the primary determinant that produces diff erences in local populations. Density, st ability and site
persistence all decreased wit h increasing densities of birds of prey. A comparison betw een putative source and
sink river sy st ems i ndi cat ed no dif ferenc es in loc al hab it at features relev ant to t he du cks, suggest ing that avian
predators may exclude Harlequin Ducks from potential habitat, in areas where cliff nesting habitat for rapt ors is
available. These findings hold critical implications for the conservation of Harlequin Ducks and illustrate the
import ance of a landscape approach t o research and management of migrat ory b irds.

121      Callo
* To defend or not to defend? Cost conscious decisions by breeding songbirds. PAUL A. CALLO, Dept. Biol.,
Univ. Mary land, College Park, M D.
           The relationship between predator and prey is a complex behavioral interaction. Defense of offspring in
additi on t o one' s self against predator s adds yet anot her level of complex ity to t his int eraction . Parents must
make deci sions about how much effort to dev ot e to det erring pred ators that may also dir ect ly threat en
themselv es. While i t appear s init iall y t hat level of paren tal def ense w ill inc rease in di rect pro por tion to t he t ot al
fit ness at ris k, there s hou ld b e an up per l imit of ris k t o eit her o neself or o ffspr ing at w hic h def ense becom es
unp rof it able. To t est the ef fect of pred atory st imuli on p arent al respon ses, I co ndu ct ed pr edat or m odel
presentations that manipulated the threat of predation to bot h the nest and parents w ithin a class of predators
instead of across classes as has been done in the past. Taxidermic models of a Blue Jay, American Crow, and
a Veer y w ere pr esent ed at the nests of Blue-h eaded V ireo s and Red-eyed Vireo s at the Hem loc k Hill Biolo gic al
Research Area in nw. Pennsylvania. Parent birds responded with signif icantly few er vocalizations, attacks and
approaches to the crow model than to the jay model. Thus parents do modify t heir responses with v ariation in
w hen the costs of defense to the parents in either energy expended or threat-to-self have exceeded the benefits
of of fspring survival.

122      Bradley & Marzluff
Changes in know ledge of nest predat or diversit y and consequences for co nservation. JEFFREY E. BRADLEY*
and JOHN M. MARZLUFF, Coll. For. Res., Univ. Washington, Seattle, WA.
      Predator diversity can have important conservation and management implications. We reviewed the
artificial nest literature to determine how t he aw areness of nest predator diversity has changed, and tested the
influence of predator diversity on the effects that habitat or nest site attributes have on nest predation rates.
Changes in techniques used in artif icial nests have resulted in a more accurate identification of predators, and
have show n t hat bir ds ar e und er pr essure f rom a hig h di versit y of nest pred ators . Co mparisons of st udi es t hat
dif fered i n t heir abil it y t o det ect pred ation attempts (i.e. , d if fered i n t he di ver sit y of pred ators samp led) show ed
that nests under pressure from a higher diversity of predators are less likely to show t he expected effects of
habitat and nest site att ributes know n to inf luence nest predation rates. Know ledge of how nest predation and
the factors t hat influence predation rates are affected by predator diversity can suggest effective conservation
st rat egies and al low us t o successf ull y int egrat e the need s of w ild lif e w it h t he dem ands of a gro w ing hum an

123      Sanzenbacher & Haig
Winter sit e fidelity and mov ement patterns of Killdeer in the Willamette Valley of Oregon. PETER M.
SANZENBACHER* , Dept. Fish. & Wild l., Oregon St ate Uni v. , Cor valli s, OR; and SUSAN M. HAIG, USGS Forest
& Rangeland Ecosyst . Sci . Ct r., Corvallis, OR.
        Killdeer are considered a common species wit h a range that extends throughout much of North
America. How ever , d ata from larg e-scale surv eys ind icat e lon g-t erm decl ines in m any regi ons. There ar e few
studies of this species and most eff orts have focused on breeding behavior. We quantified fidelity and
movement patterns of 2 4 radio-tagged Killdeer in the Willamette Valley of Oregon during the winter of 1 999 -
200 0. Results revealed that t he wint er population w as composed primarily of migrants (86 .96 %), w ith a
smaller proportion of resident breeders (13. 04% ). Mov ements were localized wit h birds detected at an average
distance of 5. 15 km ± 0.9 1 SE from site of capture. Mean home range size (95% kernel) was 7.7 3 km 2 ± 3.19
SE. How ever, results also indicate periodic exploratory movements w ith some birds detected up to 30 km f rom
band sit es. Over all, ind ividu als ex hib it ed lo w degr ees of fidel it y t o specif ic s it es and w ere det ect ed at an
aver age of 11 .9 1 sit es ± 1. 14 SE. A naly ses f oun d no dif ferenc es in mont hly movem ent pat terns , h ow ever
there w as a t enden cy for b ird s t o ran ge f art her as the no nbr eedin g per iod pro gres sed. In m ost cases, estimates
of movements f or resident breeders w ere less than those of migrants. Findings and management implications
are contrasted with conservation strat egies for Dunlin, a more mobile and wide ranging wint er resident of t he
Willamette Valley.

124      Martinson
* Effect s of su pplement al w int er bird f eeding on experiment ally placed art hropod s. TA MM IE J. MA RTINSON,
School For. & Wood Prod., Mich igan Tech. Univ. , Hought on, M I.
          Observations suggest that birds at feeders spend time foraging on trees in the vicinity of the feeder.
Alt hou gh sever al species of bir ds t hat use w int er f eeders are pr imaril y insec tivoro us, lit tle research has
investigated the influence of wint er bird feeding on the surrounding non-avian community. I examined the
effects of supplemental w inter bird feeding on populations of experimentally placed bark-inhabiting mealworms
in northern hardwoods and mixed northern hardwood f orests in the Upper Peninsula of Michigan using a
treat ment-co nt rol design w it h 1 5 r epli cat ed pl ot s in 2 study areas. Treat ment plo ts had a blac k su nf low er seed
tube f eeder an d suet feeder at the pl ot cent er w hil e con trol plo ts lac ked a f eedin g st ation . Each plo t con tained
64 m ealw orms placed in the bark of 16 trees at t he onset of the study. Numbers of remaining mealw orms in
treat ment pl ot s w ere com pared t o n um bers remain in g i n c on tro l p lo ts af ter 6 w eeks. I f ou nd sign if ic antly (P <
0.0 2) more mealw orms missing in treatment plots (60. 3 ± 2 0.3 ) compared to cont rol plots (50 .3 ± 9.8 ). This
evidence supports the hypot hesis that supplemental wint er bird feeding increases predation pressure on
experiment ally placed art hropod s.

125      Shepherd
* The influence of predators and prey on space use patt erns and time-activity budgets of individual Dunlin
wintering at the edge of their range. PHILIPPA C. F. SHEPHERD, Dept. Biol. Sci., Simon Fraser Univ., Burnaby,
          I used rad io t elemet ry and GIS to qu ant if y space u se pat terns and t ime-act ivit y budg ets of ind ividu al
Dunlin of know n sex and age from 3 dif ferent sites w ith in t he Fraser River Delta, BC. I int egrated t hese
behavioral data w ith data on biotic factors such as the distribution of food and predators, and abiotic factors
suc h as landscape f eatures and t ime of day. Both among and w it hin sit es, marine h ome rang e size decreased
as invertebrate prey density w ithin t he home range increased, w ith prey density accounting f or 63% of t he
var ianc e in h ome rang e size ac ros s sites and 89 % w it hin -site. W it hin ind ividu al Dunli n, their cor e areas
contained higher prey densities than did their home ranges, indicating that individuals focused their use of
space on better feeding areas. In spite of diff erences in prey density and the amount of space used among
sites, I found no diff erences in Dunlin body condition or in the percent time spent foraging (as indices of
perfo rmance). Th is may have been because Dunlin densit y, and theref ore int erference com petit ion, w as also
low est w here prey density w as lowest. Dunlin spent on average 16.4 h/24 -h day foraging, and 3 h/d flying
(likely as a mean s t o dec rease predat ion ris k). The p ercent time Dunlin spent forag ing did not dif fer bet w een
day and n igh t, n or b etw een marin e and t errest rial habi tats. Dunlin made lit tle use of the t errest rial habi tat
dur ing the day (1.6 % of loc ation s) w hen f alco n species, t heir main pred ators , w ere active, so t he t errest rial
habitat did not cont ribute as much as the marine habitat to t heir overall energy budget. How ever, Dunlin did
use the terrestrial habitat at night (46.9 % of locations), primarily for foraging, so access to this alternative
habitat may be impo rtant for Dunlin t o be able to balance t heir energy bud gets.

126     Barton, Ballard, Gardali & Geupel
Demography of w intering and migrant Townsend's Warblers in coastal California: The importance of weather.
DANIEL C. BARTON* , Evergreen State Coll., Olympia, WA, GRANT BALLARD, THOMAS GARDALI and
GEOFFREY R. GEUPEL, Point Reyes Bird Observ. , St inson Beach , CA .
         We banded 112 5 migrating and w intering Tow nsend's Warblers from 197 9 - 19 99 (2 1 yr) as part of a
long-term monitoring program in central coastal California. We investigated the relationships betw een w eather,
the El Niño Southern Oscillation (ENSO), w inter survival, and capture rates during the fall and wint er. Using
Program MARK, w e est imated surv ival of w int erin g in div idu als. We t est ed f or d if ferenc es in appar ent sur vival
and recapture probability between age, sex, and transient/resident groups. There w as no significant difference
in survival betw een sexes and age classes, and data was pooled in subsequent analyses. We controlled for
diff erences in recapture probability and apparent survival of t ransient vs. resident groups in all subsequent
models. Winter survival was significantly posit ively related to ENSO, indicating increased survival in La Niña
year s and depr essed sur vival dur ing El Niño year s. During unu sually col d w int ers at the st udy sit e, surv ival w as
significantly low er than average. Cold w inters may limit food availability and stress the energy budget of
w intering individuals. Overall annual survival probability, averaging the best model, w as 0.52 (± 0. 05 SE). The
number of hatching-year birds captured during the fall w as significantly posit ively related to ENSO in the
previous year. The El Niño-La Niña phenomenon and weather may play a significant role in the population
regulatio n of Tow nsend' s Warblers.

127     Iverson
Finding the secret garden of f emales: The mystery of Surf Scot er demography and wint er distribution patterns
revealed. SAMUEL A. IVERSON, Dept . Bio l., Simon Fraser Un iv., Burnaby, BC.
             Nearly all North American seaduck species are exhibiting continental population declines. Surf Scoters
have been designated a " Blue List" species by the British Columbia Ministry of Environment due to the diff iculty
of assessing non -br eedin g po pul ation s. Blue-l ist ed t axa ar e con sidered at ris k, but not ext irp ated, endan gered
or threatened. During the winters of 19 99 - 20 00 and 200 0 - 200 1, I investigated the demographic structure
and distribution patterns of Surf Scoters at a series of geographic levels, ranging from foraging flocks t o the
entire Pacific Coast. Distinctive gender and age specific plumages allow ed direct counts of w inter flock
demography. However, a ratio-based approach is valid only if the classes are sampled equally. I discuss the
reli abil it y of dir ect w int er coun ts mad e on t he basis of plu mage discriminat ion , t he op timal t iming of w int er
counts considering seasonal movements, and the degree to w hich Surf Scoters segregate according to
age/gender class. Results suggest consistently low recruitment rates and a strongly male-biased adult sex ratio
in al l po pul ation s. Surf Scoters d o no t segr egat e by age or gend er in forag ing floc ks o r ex hib it dif ferent ial
migration according to latitude. How ever, female Surf Scoters are under represented relative to males in
habitats w here prey is dominated by clams rather than bay mussels. If t he spatial extent and timing of counts
are considered, w inter ratios can be a useful conservation tool f or indexing annual recruitment and give valuable
insight int o populat ion dy namic pro cesses in seaducks.

128     James
Winter season bird populations in central Belize. DOUGLAS A. JAMES, Dept. Biol. Sci. , Univ . A rkansas,
Fayetteville, AR.
        Bird populations w ere measured using nets and census plots in 7 habitats in central Belize. The study
lasted several years in Dec through Mar when resident and wintering migrant birds w ere present. Naturally
occurring gall ery fores t habi tat topp ed t he res t in speci es number and pop ulat ion level, foll ow ed by 2 m odi fied
habitats: overgrow n milpa (farm plot) and overgrown pasture. Anot her modified habitat (village grounds) and 2
natural ones (pine ridge savanna and broken ridge) trailed these. Managed pasture was far behind. Findings
show dist urbed habitats in the region have high avian diversity. Netting results did not correspond well w ith
census results and put overgrow n pasture at t he top.

129     Pruett & W inker
Landbirds of the Aleutian Islands: Assembly of a simple high-latitude community. CHRISTIN L. PRUETT and
KEVIN WINKER, Univ. A laska Mus. and Dept. Biol. Wildl. , Univ. Alaska Fairbanks, Fairbanks, AK.
         The study of communit y assembly has primarily focused on contemporary forces such as competition
and habitat availability. A basic assumption of t his approach is that historical processes play a minimal role in
how communities are assembled. We evaluated this assumption by examining a simple 8 member community
w ith a presumably recent, common assembly history. Geological evidence suggests that the Aleutian Islands
w ere covered by ice during the last glacial maximum. Therefore all 8 co-distributed species should share the
same colonization history . Thus, communit y assembly hypot heses for Aleut ian landbirds should discount
hist orical pro cesses because they should have had lit tle ef fect on curr ent dist ribut ions. T o test thi s hypot hesis,
w e sequenced the mit ochon dri al cy tochro me b gene for individuals from Att u and Adak Islands and the Alaska
mainland for each species. We assumed that each species should have the same polyphyletic patt ern if
community struct ure was driven by recent colonization. Inst ead of 1 overriding pattern, how ever, we found 5
diff erent phylogenetic patt erns. Testing t he appearance of 5 patterns among 8 species against a random
distribution show ed no significant difference (p = 0.3 2). Therefore, the assumption that t his simple
community is the product of a single colonization event is incorrect. Instead stochastic historic factors seem to
be driving community st ructure across these replicated communities. These results indicate that ignoring
history, even in such a simple community , could lead to errors in identif ying the processes of assembly. The
role of history in t his simple community seems to be far more profound than current communit y assembly
theor y w ould sugg est.

130     Williams & Steadman
Historic and prehistoric records of parrots from the West Indies, including descriptions of three new species.
MATTHEW I. WILLIAMS*, Dept. Biol. Sci., Auburn Univ., Auburn, AL; and DAVID W. STEADMAN, Florida
Mus. Nat . Hist ., Univ . Flori da, Gainesv ille, FL.
        Newly discovered bones from prehistoric archaeological and paleontological sites represent 3
undescribed species of West Indian parrots, an Ara and Amazona from M ontserrat, and an Aratinga from
Barbuda. We also report prehistoric material of Amazona vittata (the currently endangered Puerto Rican Parrot)
from Barbuda. Traditionally thought to be endemic to Puerto Rico, A. vittata had been reported recently from
an arc haeol ogi cal c ont ext on A nt igu a, w hic h li es on the sam e bank as Barbuda. From Marie Gal ant e, w e ref er
skeletal material tentatively to Ara cf. guadeloupensis and Amazona cf. violac ea, 2 exti nct species otherw ise
know n on ly from histori c account s on the near by island o f Guadel oup e. Fin ally , w e evaluat e selected hi st ori cal
record s of many named ext inc t spec ies of West Ind ian p sit tacid s. The c ombin ed pr ehistori c and histori c dat a
suggest that at least one speci es in each o f the 3 West Ind ian p sit tacid gener a (Ara, Aratinga, Amazona) used
to occur on each major island in the Greater and Lesser Antilles. This is yet another indication of how strongly
avian dist ribut ions in t he West Ind ies have been influenced by human-caused ext irpati ons and ext inct ions.

131     Flatten, Titus & Lowell
Color and size of the Northern Goshawk in southeast Alaska. CRAIG FLATTEN* , Alask a Dept . Fish & Game,
Ketchikan, AK; KIM TITUS, Alaska Dept. Fish & Game, Douglas, AK; and RICHARD LOWELL, Alaska Dept . Fish
& Game, Petersburg, A K.
          The taxonomy of Northern Goshawk (Accipit er gentilis) subspecies in North A merica is currently a topic
of int erest and d ebat e. Resourc e managers are req uir ed t o maint ain w ell-distrib ut ed, viable g oshaw k
popu lati ons and to prot ect dist inct popu lati on segm ents th at m ay be t hreat ened or end angered. The A OU
recognizes 2 Northern Goshawk subspecies in North America: A. g. atricapillus and A. g. laingi. Some
quest ion the validit y of Northern Goshaw k subspecies t hat are based pri mari ly on subt le color and si ze
distinc tion s. The laingi subspec ies has been desc rib ed as a sm aller and d arker race in habi ting the co ast al
temperat e rain fores ts of Brit ish Colu mbia and se. Alask a. Inf orm ation on p lum age co lor ation and b ody size w as
collected from 68 adult and 70 juvenile goshawks captured at nest sites in se. Alaska betw een 1992 and 2000.
Phenotypes ranged f rom dark f orms ident ifi ed as laingi to l ight er for ms ident ifi ed as atricapillus. Mean w ing
chord s w ere sm aller than t hose repo rt ed f or Nort hern Goshaw ks f rom ot her r egio ns of Alask a, b ut larg er t han
those reported for laingi spec imens f rom coas tal British Colu mbia. Slig ht cli nal v ariat ion in size w it hin se.
Alaska was detected in some age-sex classes w ith smaller birds occurring in the south. Results generally
support the original description of laingi, noting t he occurrence of some clinal variation and probable
int ergr adat ion of subspec ies w it hin se. Alask a.

132     Prum
The theropod origin of feathers. RICHARD O. PRUM, Dept. Ecol & Evol. Biol. , Nat. Hist. Mus. , Univ . Kansas,
Lawrence, KS.
         The earliest know n bird feathers are the fully modern feathers of Archaeopteryx. In absence of any
data about the morphology of primit ive feather morphology, funct ional theories w ere used to speculate on the
form of ancestral feathers. Recently, Prum (1999, J. Exp. Zool. 2 85 : 2 91 -306 ) pr opo sed a non-func tion al
developmental theory of t he origin of feathers. Recent analyses of the integumental appendages of theropod
din osaurs o f the earl y Cr etaceou s Liaonin g f orm ation of Chin a (~1 24 my a) hav e con firm ed t hat these struc tures
share several unique features w ith avian feathers: multiple filament structure, branched at the base, and
serially branch ed. In com binati on w ith the rob ust p hylog enetic evid ence support ing t he theropo d origin of b irds,
these details confirm t hat these structures are homologous wit h avian feathers. Further, the diversity of
primitive morphologies found in dinosaurs are exactly congruent w ith t he developmental model of t he origin of
feathers. Feathers evolved in theropods long before the origin of flight and before the origin of birds.

133     Pearce, Pierson, Talbot, Derksen, Kraege & Scribner
A g enet ic ev aluat ion of mor pho logy used t o ident if y Pacific Fly w ay Canad a Geese. J OHN M . PEARCE* ,
BARBARA J. PIERSON, SANDRA L. TA LBOT, DIRK V. DERKSEN, U.S Geol. Surv., A laska Biol. Sci. Center,
Anchorage, AK; DON KRAEGE, Washington Dept. Fish & Wildl., Olympia, WA; and KIM T. SCRIBNER, Dept.
Fish. & W ildl., Mich igan State Unive., East Lansing, MI.
         We used nuclear an d mit ochon dri al (m t) DNA markers and a mix ed st ock anal ysis appro ach t o evaluat e
the abi lit y of cul men leng th and breas t plu mage col or m easur es t o co rrect ly iden tif y grou ps of hun ter-h arv est ed
Dusky (Branta canadensis occidentalis) and Dusky -like Canad a Geese on the w int erin g gr oun ds. We est imated
contribut ions from t he Copper River Delta, the primary breeding area of Dusky Canada Geese, to groups of
harvested geese classified as Dusky Canada Geese on the basis of culmen and plumage as 50.6 % ± 0.1 SE for
females and 50.3 % ± 13. 0 SE for males. Estimates of cont ribution f rom the Copper River Delta to the
dusky-like category (birds of lighter plumage color) were < 5% . The use of culmen length and plumage color
to identify breeding origins of harvested geese provides a conservative criteria for management of Dusky
Canada Geese. Multiv ariate analysis revealed that additional discriminatory m orphological characters exist and
their u se may impro ve ident ifi cation of Du sky Canada Geese on the w int ering grou nds.

134      Kessen & Zink
Microsatellite analysis of Fox Sparrow population structure. ANN E. KESSEN* and ROBERT M. ZINK, Bell
Mus. , Univ . M innesot a, St. Paul, M N.
        The Fox Sparrow (Passerella iliaca) is a morphologically diverse taxon with 18 named subspecies (Am.
Ornithol. Union 1957, Check-list ) that assort in to 4 groups b ased on morphol ogy and on mit ochond rial DNA.
Hybrids occur in some of the contact zones between the groups, but t he full extent and significance of this
hybridization has been unknown. We analyzed allelic variation at 4 microsatellite loci, developed using
heterospecific primers, in 15 1 individuals representing 16 of the 18 subspecies and all 4 of the groups. Results
indicate population dif ferentiation among the groups and, unexpectedly, demonstrate a higher degree of
microsatellite polymorphism in the Fox Sparrow t han in the taxon for w hich the primers were developed.

135      Warheit & Schroeder
Columbian Sharp-tailed Grouse population genetics -- preliminary analysis. KENNETH I. WA RHEIT, Dept. Fish &
Wildl., Olympia, WA; and MICHAEL A. SCHROEDER, Dept. Fish & Wildl., Bridgeport, WA.
          The h ist ori cal r ange o f the Col umbian Sharp-t ailed Grou se (Tympanuchus phasianellus columbianus)
extended from the steppe- and shrub-dominated habitats in the inter-mountain regions from British Columbia
south t o Cali forn ia, Nevada, and Utah, and east to w . M ont ana, Wy oming and Co lor ado. The subspeci es has
been extirpated from most of it s range and exists now as remnant and isolated populations. We conducted a
preliminary analysis of the genetic st ructure of 3 of t hese populations (Washington, n = 19; se. Idaho, n = 31;
Montana, w est of the divide [wM T], n = 3). Also included in the analysis w ere samples from the Alaska
subspecies (T. p. caurus, n = 3) and from M ontana, east of t he divide [eMT] (n = 5). The taxonomic status of
the eMT samples is uncertain (either Columbian or Plains, T. p. jamesi). Preliminary analysis of 7 v ariable
micro satellit e loci suggest s signif icant g enetic st ruct uring amon g th e 5 geographic areas (Fst = 0.09), with
pair w ise Fst ranging f rom 0 .0 8 t o 0. 21 . Of the 3 Columbian sit es (Fst = 0.0 8), Washington and wMT w ere
most sim ilar geno typ ical ly. How ever , t he Id aho samples appeared more similar t o t he eM T and Alask a samp les
than to t he Washington or w MT samples. The w MT and eMT samples showed highly significant genoty pic
diff erentiation, although they are located w ithin approximately 16 0 km of each other. A lthough t hese results
are preliminary, the data suggest t hat the genetic relationships w ithin t his species may not now correlate well
w ith subspecific boundaries, and that t here may be significant genetic structure among the remnant populations
of the Columbian Sharp-tailed Grouse.

136      Hess & Edwards
MHC evolution in House Finches. CHRISTOPHER M. HESS and SCOTT V. EDWARDS, Dept. Zool., Univ.
Washington, Seattle, WA.
        Underst andi ng t he in teraction bet w een select ion act ing at the lev el of the gen e and t he or gani sm h as
long been a goal of evolutionary biologists. W e discuss our ongoing research concerning the evolution of the
Major Histocompatability Complex (MHC) in House Finches. MHC genes are involved in mounting immune
resp onses t o pat hog ens and t heref ore ar e an ob vious lin k bet w een select ion at the gen etic an d or gani smal
levels. We discuss how t he mechanisms of evolution w orking on the avian Mhc differ f rom those of mammals
and discuss pr eliminary d ata on geographic and tempo ral variabilit y of these genes.

137     Sharp
* Factors influencing the incidence of brood parasitism by Brown-headed Cowbirds on Least Bell's Vireos.
BRYA N L. SHARP, Dept . Bio l., San Diego State Univ ., San Dieg o, CA.
          I examined nest concealment and host behavior of Least Bell's Vireos (LBVI' s) wit h respect t o
pro babi lit y of bro od p arasitism . I samp led v eget ative cov er su rroun din g LBVI nests (n = 11 3; 42 paras it ized
[P], 7 1 non -parasitized [NP]; 19 99 -20 00 ). Numb er of v egetatio n hit s along t he upper diagonal t ransect (4 5°
angle) above nests, and percent vegetative cover lateral to nests (both w ithin 1 m of nests) diff ered betw een P
and NP nests (2-tailed 2 sample t-tests; mean NP = 3.145, mean P = 2.324, p = 0.053; mean NP = 18.5,
mean P = 13. 7, p = 0.0 56, respectively). P and NP nests also diff ered in percent canopy cover at t he
mesohabit at scale (1 - 1 1. 3 m f rom nest s, (mean NP = 53 .7 , mean P = 59 .8 , p = 0. 00 2). I examined host
vocalization rates near nests (n = 21; 8 P, 13 NP; 200 0 only) under the assumption that cow birds cue on this
behav ior w hen searchin g f or n est s. I f oun d no signif ican t dif ferenc e in v ocalizat ion behav ior of adul ts bet w een
PAR and NPAR nests. In short, greater cover within 1 m of a nest reduces the chance it will be parasitized.
Conversely, greater canopy cover at t he mesohabitat scale resulted in a higher likelihood of parasitism. LBVI is
a federal ly endan gered subspec ies of Bell' s Vireo . I suggest that rip arian habi tat of LBVI' s shoul d be m anaged
to increase understory density, and to reduce numbers of large, mature trees. I offer t hese suggestions
because LBVI's prefer to nest in dense understory, and Brow n-headed Cow birds use perches available in large
trees t o surv ey f or n est s. One pot ent ial m eans t ow ard t his management st rat egy w oul d be a r eturn to nat ural
flood regimes in riparian woodlands.

138      Hauber
* Host-parasite hatching asynchrony and the cost of interspecific brood parasitism. MARK E. HAUBER, Dept.
Neurobiol. & Behav. , Cornell Univ ., Ithaca, NY.
         All p arent al ho st s of int ersp ecif ic b roo d par asit es mu st pay the co st of rearing non -ki n. For ex ample,
the presence of Brown-headed Cowbird chicks is frequently accompanied by the reduction of host f ledging
success. Previous studies have indicated that the relative size advantage of cow birds is associated wit h
preferential feeding by fost er parents and the extent of loss of host nestmates. Size advantage of nestling
cow birds can be the result of at least 2 diff erent causes: i) cowbirds preferentially parasitize species wit h
smaller nestlings and low er growt h rates and/or ii) cow birds hatch earlier than hosts. In a comparative analysis
I found that both host incubation duration and adult host mass, surrogate measures for host-parasite hatching
asynchrony and nestling grow th rat e, w ere significant predictors of the cost of rearing parasitic y oung. To
determine whether hatching asynchrony per se is a causal fact or i n dec reasing h ost bro od sizes, I introd uced
single, early-hatching Eastern Phoebe eggs or nestlings into non-parasitized clutches of this common cowbird
host. A s predicted, induced hatching asynchrony had a negative effect on egg-to-nestling survival rates of the
original brood. However, the extent of brood reduction caused by early hatching conspecifics consistently
und erest imated ob serv ed lev els of bro od r educ tion in c ow bir d-parasitized p hoeb e nest s. Thi s suggests t hat
features of cow bird nestlings other than early hatching also contribute to t he reproductive costs of brood

139      Kilpatrick
* Variation in growth of Brown-headed Cowbird nestlings and energetic impacts on their host parents. A.
MA RM KILPATRICK, Zool. Dept., Univ. Wisconsin-Madison, Madison, WI.
         I t est ed sev eral h yp ot heses abou t the pl ast ici ty of avian gr ow th by com pari ng g row th of Brow n-Head ed
Cowbird nestlings in tw enty diff erent host species. Cowbird grow th w as unrelated to host adult mass and the
data suggest that nearly all host s are able to prov ision cow birds w ith enough f ood t o grow near the highest
observed rate. Cow bird grow th w as positively related to site latit ude (which negatively covaried wit h maximum
temperature) and negatively related to the nestling period of the host species which combined to explain nearly
fif ty percent o f t he variatio n in cow bird gr ow th. The metabolizable energy exp enditu re of c ow bird and host
nestlings was also estimated as an approximate measure of the food provisioned by parents. One cow bird
nestling was equivalent t o 0.6 to 3 .9 host nestlings in terms of peak energy intake, depending on the size of
the ho st spec ies. Cow bir d nestlin gs impose a substant ial en erget ic d emand on smal ler h ost paren ts w hic h may
redu ce t heir fut ure s urv ival or fecun dit y. The q uant it y of energ y prov isioned to par asit ized n est s dem onstrat es
that parents are often willing t o provision nests at a much higher rate than they do for an average clutch of
their ow n young.

140     Chace
Host selection by sympatric brood parasites in southeastern Arizona. JAMESON F. CHACE, Biol. Dept.,
Villano va Un iv ., Villano va, PA.
         Avian brood parasites w ith broad host-resource base potentially compete for nests to parasitize. Old
World cuckoos exhibit host-resource partit ioning, or “ alloxenia,” w here they are sympatric in Afric a, Japan and
Australia. However, alloxenia is relatively unstudied among New World brood parasites. In 1997 - 1999, I
examined host use by sympatric Brown-headed and Bronzed Cow birds in pine-oak and riparian deciduous
for ests in se. A rizona. Bronzed and Brow n-headed Cowb irds select ed nests nonr andomly , and part iti oned host s
by host body size. Brown-headed Cowbirds did not parasitize the large tanagers, w hile Bronzed Cow birds did
not parasitize the small Painted Redstart and Bell’s Vireo. Cowbird egg-laying periods overlapped broadly,
how ever , Bronzed Cow bir d egg -lay ing occ urr ed lat er in the seaso n and coi ncided w it h t he peak of tanager
clu tch ini tiat ion . A lt hou gh t here w as some over lap o f host use, there w ere on ly 3 simult aneou sly paras it ized
nests of the Plumbeous Vireo. These results differ significantly from tw o earlier studies in s. Texas where
Bronzed and Brown -headed Cowbi rds ov erlapped extensiv ely in host use (i.e., homox enia).

141     Hannon
Cowbird parasitism increases predation risk to American Redstart nests. SUSAN J. HANNON, Dept. Biol. Sci.,
Uni v. Albert a, Edmont on, A B.
          Cowbirds may perceive a landscape as a mosaic of suitable patches of hosts. The profitability of
searching patches for host nests likely depends on the species of hosts present. Some host species may have
adaptations to deter or reduce cowbird success, making them suboptimal "habitat" . American Redstarts are
not common hosts for Brown-headed Cowbirds in our study area in an agricultural landscape in central Alberta.
Only 1 0% of 3 03 nests f ound ov er 5 y r w ere parasitized. In addit ion most cow bird eggs laid in redstart nests
do not hatch as 44% of n ests are abandoned aft er the parasit ism event . Redstart s that abandon fi rst n ests
rebuild and relay. A n additional 37% of parasitized nests are depredated before fledging any young. The
presence of a cowbird egg increases the risk of predation of the redstart eggs and increases the probability of
nest aband onm ent com pared to un paras it ized n est s. Only 1 8% of cow bir d egg s laid in reds tart nest s pr odu ce a
fledged cowbird chick. Hence, redstarts appear to be unprofitable as hosts for cowbirds in this system. I
discuss mech anisms f or how cow birds inc rease nest predation risk and relat e spatial patt erns of p arasitism
relativ e to pat ch size, isolat ion and relati ve abundance of alternat ive host s.

142      Winter & Johnson
What f actors inf luence cowbird parasitism in grassland birds? MAIKEN WINTER* , State Univ. New York,
Syracuse, NY; and DOUGLAS H. JOHNSON, U.S. Geol. Surv., Nort hern Prairie Wildl. Res. Ctr., Jamestow n,
        One reason for the apparent decline in many populations of grassland-nesting birds is decreased nesting
success. Nesting success can be reduced due to either heightened nest depredation by a variety of predators
or increased brood parasitism by Brown-headed Cowbirds. Grassland passerines seem to suff er from
notoriously high rates of nest predation. Cowbird parasitism, in cont rast, is generally low but can be serious in
certain situations. Such variation might be caused by differences in characteristics of nest sites, or by
larger-scale habitat features such as distance to edge, patch size, landscape struct ure, and the presence of
catt le or bison. We invest igated how rates of cow bird parasit ism varied amon g 3 gr assland passerine species,
and i f this var iat ion ref lect ed di fferenc es in nest sit e char act eristics or h abit at features . To add ress this issue,
w e examined parasitism rates of 145 0 nests in 29 native prairies in 3 regions in Minnesota and North Dakota
during 199 8 - 20 00. Nest site characteristics did not seem to influence brood parasitism, but certain
larger-scale habitat featu res w ere associated w ith brood p arasitism r ates.

143      Curson & Mathew s
Reproductive costs of commuting by Brown-headed Cowbirds in New M exico. DAVID R. CURSON* and
NANCY E. MATHEWS, Dept. W ildl. Ecol., Univ. W isconsin, M adison, WI.
         Broo d par asit ic Br ow n-headed Co w bir ds f requ ent ly exhibi t a pat tern o f dail y c ommuting bet w een
forested breeding areas and open habitats, upon w hich they depend for feeding. In certain landscape
condi tion s t hey are know n t o co mmut e up t o 1 3 k m each w ay on a dai ly basis. Because the est imated
energetic cost of a 20 km f light f or a cowbird is 60 % of the energetic cost of producing an egg, a trade-off
betw een these tw o energy demands is expected. We measured the rate of cow bird egg production in
long-distance commuters (LDCs; > 10 km each way) and short distance commuters (SDCs; < 2 km each way)
at a majo r pr airi e-f ores t int erf ace in ne. New Mexico, f rom cou nt s of post-ov ulat ory foll icl es in dissect ed
ov aries. LDC' s laid 5 0% few er egg s ov er a 5 day peri od t han d id SDCs. Thi s result suggest s a pr ox imate
energetic trade-off as predicted, but could alternatively be due spatial structuring by age or dominance in the
local cowbird population, which could not be measured directly.

144      Kus
Just desert: Is desertion o f parasit ized nests an eff ectiv e strat egy f or maxim izing prod uct ivit y in t he Least
Bell's Vireo? BARBARA E. KUS, USGS West ern Ecol ogical Researc h Cent er, San Dieg o, CA.
         Recent analyses of the im pact o f c ow bird parasit ism on ho st pr oduct ivit y suggest that w hile parasitism
redu ces prod uctivit y on a per nest basis, the abi lit y of pair s t o desert paras it ized n est s and re-n est allo w s t hem
to achieve productivity comparable to t hat of unparasitized pairs. This has potential implications for the
management of several endangered species highly vulnerable to parasitism and consequently t he target of
w idespread cowbird cont rol programs. I calculated seasonal nesting eff ort (number of nests/pair) and
productivit y of 5 73 pairs of Least Bell' s Vireos monitored over 11 yr at t he San Luis Rey River in San Diego Co.
(w here cow bird t rapping has reduced, but not eliminated, parasitism ), assigning pairs to 1 of 4 gr oups: (1)
deser ters, (2) rescued (parasitized p airs w it h nests " resc ued" from deser tion or p rob able f ailu re t o f ledg e vireo
young by the removal of cow bird eggs), (3) desert and rescued, and (4) unparasitized. Parasitized pairs
att empted sig nif icantl y mor e nests/season t han did unparasit ized pairs, w ith deserters prod ucing m ore nests
than r escu ed pai rs. How ever , p rod uctivit y of deser ting pair s w as signi fican tly low er t han t hat of bot h res cued
and unparasitized pairs, largely the result of parasitism of subsequent re-nesting attempts by deserting pairs.
Seasonal productivity did not diff er significantly betw een rescued and unparasitized pairs, indicating that in this
species, reduction of cowbird impacts through nest manipulation is an effective practice.

145      Becker & Beissinger
Seasonal, annual and centurial variation in Marbled Murrelet t rophic level. BENJAM IN H. BECKER* and
STEVEN R. BEISSINGER, Div . Ecos ys. Sci. , Un iv. Calif orn ia Berkel ey, Berkeley , CA .
         We used stable n it rog en and carbon isotope an alysis of Marbl ed M urr elet feathers and p ot ent ial p rey
species from 19 98 - 2 000 to inf er the seasonal and annual relationships betw een murrelet diet, indicators of
repr odu ct ive suc cess and c ond it ion , o ceano grap hy , an d pr ey av ailab ili ty. Murr elet s f ed on a low er t rop hic level
dur ing the pr e-br eedin g molt of 19 99 and 2 00 0, w hic h w ere no n-El Niño y ears. These year s also show ed
higher murrelet fecundity, and increased lower trophic level prey availability. Murrelets with indicators of active
reproduc tio n (presence of b rood pat ches and inland f lying behaviors) f ed on a low er tro phic level t han tho se
w it hou t bro od p atches or inl and f lying behav ior s. Murr elet s may feed on a low er, less div erse t rop hic level
because such prey may be more common near the nesting habitat, and nesting birds merely happen to prey on
these resources. Alternatively, early spring food w eb development in oceanographically favorable years begins
w ith a pulse of lower t rophic level prey and birds that detect this pulse prior to breeding may use it as a cue
that the year will be biologically productive and initiate nesting activities. We also sampled murrelet feathers
from museum specimens collected in central California from 18 95 - 1 911 . These birds did not undergo a
significant pre-breeding shift t o a lower trophic level diet, suggesting that changes in prey composition or
availability may have occurred in the California current during the past century.

146     Sherry, Chavez-Ramirez & Slack
Factor s affect ing patt erns of f oraging habitat use by w intering w ading birds in Texas. DA WN A. SHERRY* ,
Caesar Kleberg Wildl. Res. Inst., Texas A&M Univ.-Kingsville, Texas A&M Univ., College Station, TX; FELIPE
CHAVEZ-RAMIREZ, World Wildlife Fund, Chihuahuan Desert Ecoreg. Dir., Mont errey, M exico; an d R. DOUGLA S
SLACK, Texas A&M Univ., College Station, TX.
          We examined interspecific habitat-use patterns of w intering w ading birds and factors t hat might
influence these patterns. Surveys w ere conducted on Matagorda Island Natl. Wildl. Refuge, Texas, from Nov
thro ugh Mar of 19 97 , 1 99 8 an d 1 99 9. We classif ied w adin g bi rd f orag ing habi tat int o 3 hier archic al
categories: macrohabitat (size of w ater body based on aerial extent), mesohabitat (distance of individual to
edge) and microhabitat (w ater depth). Patterns of habitat use by w ading birds could be influenced by the
follow ing factors: morphology , foraging behavior, vegetation and prey density. Surveys w ere conducted for the
fol low ing f oraging w ading bird sp ecies: Great Blue Herons (n = 10 ), Great Egrets (n = 13 1), Snow y Egrets (n
= 142 ), Reddish Egrets (n = 22) and Little Blue Herons (n= 16). When foraging birds were located,
macrohabitat was classified, mesohabitat w as estimated and microhabitat (water depth) w as measured using a
meter stick. A t t he macrohabitat level, Great Egrets, Reddish Egrets, Little Blue Herons and Tricolored Herons
occurred more often in narrow channels, whereas Snow y Egrets preferred open water habitats. At the
mesohabit at level all species occu rred most frequent ly near t he land-w ater int erface (edge). How ever, Reddish
Egrets occurred more frequently aw ay from t he edge than other species, probably due to t heir active feeding
st yle.

147     Krapu & Brandt
Eff ect of decl ini ng f ood abun danc e and b ody size on f at st orag e in spri ng-st agin g Sand hil l Cran es. GARY L.
KRAPU and DAVID A. BRANDT, Nort hern Prairie Wildl. Res. Center, Jamesto w n, ND.
         The midcontinental population of Sandhill Cranes for the past several decades has relied on waste corn
in the Central Platte River Valley (CPRV) of Nebraska, the principal spring staging area, to supply a significant
portion of energy requirements for migration and reproduction. We compared w aste corn availability, crane
food habits, and magnitude of fat st orage by cranes in the CPRV in the springs of 19 98 - 1 999 and 1978 -
1979. Waste corn post harvest declined from 5% of the standing crop in fall 1978 to 2% in fall 1998.
Despite this change, corn continued to dominate the diet in 199 8 - 19 99 as in 19 78 - 1 979 and remains the
source of most fat stor ed by cranes w hile in t he CPRV. Wit h w aste corn d eclining f rom being p resent in exc ess
of needs in the 1970 s to being relatively scarce in 1999 , fat (as a percent of body mass) went f rom not varying
w ith body size to varying inversely. Implications of w aste corn declining in the CPRV and fat st orage being
related to body size and latit ude of breeding w ill be discussed.

148     Silverman, Nevitt & V eit
Intr a- and interspecif ic associat ions among seabirds f oraging in t he south ern ocean. EMILY SILVERMA N* ,
School Nat. Res. & Environ., Univ. Michigan, Ann Arbor, MI; GABRIELLE NEVITT, Bodega Marine Lab., Univ.
Calif orn ia Davis, Bodega Bay , CA ; and RICHARD VEIT, Dept. Biol., Coll. Staten Island, CUNY, Staten Island,
          We analyze patt erns of species occurrence along strip transects conducted near South Georgia during
the austral summer. This analysis explores the foraging behavior of individual seabirds and quantifies intra- and
interspecific association. Using data collected over 9 d and hundreds of kilometers of ocean, we f ocus on the
behavior of 4 div erse species that are abundant in A ntarc tic w aters: Black-br ow ed Albat ross Diomedea
melanophris, Whit e-chinned Petrel Procellaria aequinoctialis, prion Pachyptila spp. and diving petrel Pelecano ides
spp. We analyze the identit y and behavior of t he nearest neighbors of focal individuals and demonstrate strong
intraspecific associations for all 4 species. These associations suggest that the species may forage
co-operat ively w ith conspecif ics; o f t he 4 species consi dered, prio ns fo rmed t he largest singl e species groups,
follow ed by diving petrels. Patterns of int erspecific association are also revealed. Specifically, individual birds
are most likely to be found near heterospecifics that are engaged in feeding, and not foraging, behavior.
Individual White-chinned Petrels, prions, and diving petrels with heterospecific neighbors demonstrate positive
associations w ith Black-browed Albatross. None of the 4 species, however, show association w ith penguins or
marine mammals.

149      Styring
* Resource partitioning in Malaysian woodpeckers. ALISON R. STYRING, Mus. Nat. Sci. & Dept. Biol. Sci.,
Loui siana State, Univ ., Bato n Roug e, LA .
         Woodpeckers are particularly well suited to studies involving niche theory because their obvious
morphological adaptations allow them to exploit resources inaccessible to other species. Woodpeckers are
extremely diverse in the forests of Sout heast Asia; w here as many as 15 species can occur in a single patch of
low land rainforest. This study explored niche diversification in a guild of 13 w oodpecker species found in the
low land rain fores ts of peni nsular M alaysia. The f ocus of the res earch w as f orag ing ecol ogy , an d t he go al w as
to det ermine t he deg ree of overlap these speci es exhib it ed in their forag ing pref erenc es. A majo rit y of spec ies
segr egat ed res our ces spat iall y by either u sing subst rat es of dif ferent sizes or b y f orag ing in d if ferent ver tical
levels of t he forest. Six species overlapped spatially, but they div ersified niche space by either using different
types of substrates, or by using different foraging maneuvers. Several species used resources not commonly !
exploi ted by w ood peck ers such as arbor eal ant and t ermit e nest s and leaves. The ab ili ty t o use of these nov el
resources may explain, in part, why this guild is so diverse in Southeast Asia. Two congeners, Picus mentalis
and P. puniceus, o ver lapp ed in all v ariab les analyzed ex cept forag ing maneuver. These clo sely relat ed species
often foraged together in flocks, and their ecological dynamics warrant further study.

150      Lyons, Roby, Collis & Craig
Population bi oenergetics m odeling of Double-crest ed Cormorant s in t he Columbia River estuary. DONALD E.
LYONS* , DANIEL D. ROBY, USGS-Oregon Coop. Fish & Wildl. Res. Unit, USGS; Dept. Fish. & Wildl., Oregon
Stat e Univ. , Cor valli s, OR; KEN COLLIS, Real Tim e Research, Bend, OR; and DAVID P. CRAIG, Dept. Biol.,
Will amett e Univ. , Salem, OR.
        We used a bioen erget ic m odel of prey con sum pt ion to assess int eraction s bet w een do ubl e-cr est ed
cormoran ts and juv enil e salm oni ds in t he Col umbia Riv er est uary , d uri ng t he years 1 99 8 - 20 00 . A vian
predation has been proposed as a possible mechanism limiting recovery of Columbia Basin salmonid
populations, a majority of w hich are listed under the federal Endangered Species Act. The breeding colony of
double-crested cormorants on East Sand Island in the Columbia River estuary has grow n from 9 1 pairs in 198 9
to over 70 00 pairs in 19 99, making it the largest Double-crested Cormorant colony on the Pacific Coast of
North America. The diet of cormorants during the breeding season, collected from stomach samples and chick
regurgitations, has consisted of 15 - 55% juvenile salmonids by mass, depending on colony location and year.
Seasonal use of juvenile salmonids by cormorants has peaked at t he time of highest out-migration of salmonid
smolts, generally during May of each year. Bioenergetics modeling estimates of t he yearly consumption of
juvenile salmonids by cormorants in the estuary are several million fish/yr, ranging from 2 - 9% of the smolts
that pass through the estuary.
151      Antolos
* Caspian Ter ns in t he mid-Colu mbia River: A loo k at nest densit y, diet , an d f orag ing rang e. M ICHELLE
ANTOLOS, USGS - Oregon Coop. Fish & Wi ldl. Res. Unit , Dept . Fish. & Wildl. , Oregon St ate Univ. , Corv allis,
           I examin ed t he nesting and f orag ing ecol ogy of Caspian Ter ns at 2 c olo nies in t he mid-Colu mbia River
(Three Mile Canyon Isl and [2 60 breeding pairs] and Crescent Island [ 57 1 pairs] ). I com pared these colon ies to
2 c olo nies in t he Col umbia Riv er est uary : East Sand I sland (9 09 7 p airs) and Rice Island (580 pair s). The Three
Mile Can yo n Island col ony failed due t o mink pred ation . Nest ing suc cess at Cresc ent Island, how ever , w as
comparabl e to t hat at East Sand Island , t he lar gest kno w n Caspian T ern c olo ny in t he w orl d. Nest ing densit y at
bot h mid-Colu mbia co lon ies w as higher than at the est uary col oni es. Regardl ess, nest densit y did n ot have a
signif ican t effect on p rod uctivit y at Cresc ent Island, after account ing for p osition w it hin the co lon y, hat ch dat e,
and adult att endance. Higher densities of foraging terns in the mid-Columbia River w ere observed at small
trib ut aries, n ear-shor e shallow s on the main r iver, encl osed pon ds, and d ams. Diet com posit ion w as
determined at mid-Columbia colonies by visual identif ication of adult bill loads on-colony. The proportion of
juvenile salmoni ds in t he diet of Crescent Island t erns w as 61% , less t han at Rice Island, but more t han at East
Sand I sland. In addit ion , I discover ed a Caspian T ern c olo ny (ca. 15 0 p airs) on Sols tice I sland in the Pot hol es
Reservoir, WA, w here recovered PIT tags demonstrated that terns w ere commuting over 50 km t o the
Columbia River to for age on salmonids. These results indicate that dietary specialization on salmonids is not
solely an attribute of tern colonies in the Columbia River estuary.

152      Borgmann
* Evaluating the im por tance o f w ax m yr tle (Myrica cerifera) t o ov er-w int erin g bi rds in So ut h Caro lin a. KATHI L.
BORGMA NN, Dep t. Zool., Univ . Flori da, Gainesv ille, FL
         Lit tle is kno w n abo ut w int er bi rd ec olo gy in t he sout heast , an d in part icu lar t he im por tance o f w int er
fruit as a food resource for birds. I experimentally examined the importance of Myrica cerifera fruit to w intering
birds in managed longleaf and loblolly pine stands in the upper coastal plain of South Carolina by manipulating
fruit availability. In early w inter, I covered fruiting M. cerifera shrubs wit h protective nett ing to prevent birds
from eating fruit s on treatment sites. I uncovered these shrubs in late w inter w hen M. cerifera fru it w as less
common on the landscape, allow ing me to document the response of the bird community t o these resource
patches. Total bird abundance and species richness did not change aft er the fruit w as exposed. However,
more Yellow -rumped Warblers w ere detected on treatment plot s relative to cont rols after net removal,
suggesting that Yellow-rumped Warblers were keying in on M. cerifera frui ts and that M. cerifera fruit may be
an im por tant w int er f ood reso urc e. Because M. cerifera fruit appears to be an important resource and
prescribed fire regimes tend to reduce the amount of available fruit , I examined the eff ect of f ire frequency on
M. cerifera frui t pro duc tion by com pari ng t he amoun t of frui t pro duc ed and the nu mber of shr ubs on sit es
differing in time since burning. The amount of fruit and the number of shrubs increased with t ime since
burning. M. cerifera did not pro duc e signif ican t amount s of frui t unt il 4 to 6 yr after t he pr escr ibed fire,
consequently short intervals between prescribed burns appears to reduce the amount of M. cerifera fruit
available to bir ds.

153      Cohen
Sonagram comparisons of Tree Swallow and Violet-green Swallow calls. ROBERT R. COHEN, Metrop oli tan
Stat e Coll. Denver, Denver, CO.
         I recorded Tree Swallows (TS) and Violet-green Sw allow s (VS) in Colorado and developed sonagrams
w ith A visoft -Sonagraph Pro soft w are. Only the distress and young-nestling begging calls are very similar in
both species. The VS alarm call, a pure note of f alling pitch, is similar to only the brief last component of t he
TS alarm call and is much more similar to the TS anxiety call, w hich typically has both falling and rising pitch
components but sometimes lacks the rising end; some VS alarm calls have a very brief rising-pit ch end. The TS
raspy aggression call, given in close agonistic interactions, resembles the VS raspy call, given to a distant
intruding nest-site competitor, but only t he latt er has rising pitch; bot h resemble the distress call. Compared to
the TS pleasure call, the VS pleasure call has fewer notes in its first part, its last part is much simpler in
struct ure, and its notes are given much more slowly. The TS contact call, t he VS distant-contact call, and the
nest-site advertising call of both species are similar to the begging calls of their fledglings, but the VS
close-contact call instead resembles the contact call of the Cliff and Northern Rough-w inged Sw allows. A rapid
zig-zag pitch pattern occurs in the distress calls of both TS and VS and also in the TS alarm and tee-tee calls.

154      Templeton, Davis & Greene
Variation in the " chick-a-dee" calls and behavior of Black-capped Chickadees in response to diff erent species of
predators. CHRISTOPHER N. TEMPLETON* , Div. Biol. Sci., Univ. Montana, Missoula, MT; KA TE DAV IS,
Raptors of the Rockies, Clinton, MT; and ERICK GREENE, Div. Biol. Sci. Univ. Mont ana.
           Many anim als use vocalizat ion s t o aler t con spec if ics to t he t hreat of pot ent ial p redat ors . If a prey
spec ies m ust use dif ferent behav ior s t o avoid pred ators w it h ch aracteristical ly dif ferent hun ting st rat egies, t hen
more complex signals pro vidin g in form ation abou t the t yp e of pred ator m ay ev olv e. For instance, a few spec ies
of birds and mammals have been shown to have different anti-predator calls for terrestrial and aerial predators.
How ever , l it tle is kno w n abo ut how finel y animals c an di scr iminat e among c lasses of pred ators and h ow they
trans mit this inf orm ation . W e inv est igat ed t hese issues w it h co lor -mar ked Bl ack-cap ped Ch ick adees in lar ge,
semi -nat ural aviaries. W e pres ent ed ch ick adees w it h 2 spec ies of terrest rial pred ators (domest ic c at and f erret)
and 8 dif ferent spec ies of liv e, t ethered , r apt ors (Pygmy -ow l, Saw -w het Ow l, Great -horned Ow l, American
Kest rel, Merli n, Prairie Falcon, Coop er' s Haw k, and Rou gh-legg ed Haw k). Chic kadee b ehavior al respon ses
varied dramatically for dif ferent predators and included very close mobbing approaches, freezing far away from
the predator, and normal foraging behavior. Analysis of t he acoustic struct ure of the chickadee' s vocalizations
showed substantial variation in all elements of the " chick-a-dee" call. There was also variation in t he
call -st ruc tures of dif ferent ind ividu als. Despi te this var iat ion , t here ap pear t o be systemat ic s truc tural
diff erences in the ' chick-a-dee" call that relate to t he type of predator encountered.

155      Barlow
The Cuban Vireo - What are it's " vocal" affinities? JON C. BARLOW, CBCB, Royal Ontario M us., Toront o,ON.
        The Cuban Vireo (Vireo gundlachii) is 1 of 2 Greater A nti llean vireos (subgenus Vireo) -- the other being
the Puerto Rican Vireo (V. latimeri) that has songs of diverse syllables all of w hich are pure tone (no frequency
modulated syllables). Why is this so? Two com peting hypotheses may be proposed: the Cuban Vireo is
" ancient" and long isolated and has few competitors f or it' s airwave space so that f m is not needed. Or, it
reached Cuba early as part of a lineage in which f m had never developed as proved unnecessary in the vocally
low -key relatively sound poor Cuban environment developed. Evidence is presented in support of both

156      van Gelder
* A t est of the acoust ic adaptat ion hy pot hesis: Spot ted Ow ls in f orested habit at. J ENNIFER J. VA N GELDER,
Wil dlife Dept ., Humboldt State Univ ., Ar cat a, CA .
          The Acoustic Adaptat ion Hypothesis (AAH) predicts diff erences in avian vocalizations as a function of
the environment w ithin w hich a species evolved. Direct predictions include variation in call frequency and call
modul ation s (ampli tude, frequ ency , an d rat e) bec ause soun d deg rades as a f unc tion of habi tat typ e. St udi es
supporting the AA H typically base comparisons between open and closed habitats, how ever variation w ithin
forest stands may also impose selective pressures that result in subspecific vocal variation. I examined the
acoustic properties of Spotted Ow l (Strix occidentalis) calls transmitted in "nativ e" v ersus "f oreign" habitat by
broadcasting calls of all 3 Spotted Ow l subspecies in Northern Spotted Owl (S. o. caurina) habitat . Forest
stand characteristics at ow l nest sites indicated that habitat of the Mexican Spotted Ow l (S. o. lucida) w as
more op en t han ei ther t he Cali forn ia (S. o. occidentalis) or Northern subspecies. Calls from 3 r ando mly select ed
ow ls of each s ubspeci es w ere br oadc ast and t hen r e-rec ord ed at distances 5 an d 3 0 m in 1 0 r ando mly select ed
Northern Spot ted Ow l st ands. From s pect rog rams of the br oadc ast calls, I measu red p eak intensit y per hertz,
peak frequency, and call length. I conducted one-w ay repeated measures ANOVAs using the change in each
spec trog ram var iabl e from 5 t o 3 0 m as t he res pon se variab le. I f oun d a signi fican t dif ferenc e in t he peak
intensity per hertz in Mexican Spotted Owl calls in comparison to Northern and California subspecies. I found
no significant subspecies difference in peak frequencies of note 2 or 4 , or change in overall call length, although
there w as a sig nif ican t st and ef fect in c all l engt h in dic ating inf luen ces i mposed by habi tat. The in creased
degradation incurred by Mexican Spotted Owl calls provides limited support of the AAH.

157     Waldo
* Discrimination between neighbors and strangers in Spotted Owls: Application of a crossover design.
STEPHANIE L. WALDO, Humboldt State Univ ., Ar cat a, CA .
         I co ndu ct ed a pl ayback experim ent to det ermine i f territ ori al mal e Northern Spot ted Ow ls dist ing uished
betw een calls of familiar and unfamiliar owls. I used a crossover design where territories were randomly
assigned w ithin 3 m onthly blocks. Three playback treatments w ere used: neighbor, stranger, and opposite
neighbor. The neighbor treatment consisted of playing back calls of an individual occupying an adjacent territory
for > 1 yr and w hom also shared a common boundary w ith t he individual of int erest. The stranger treatment
w as the playback of an individual who w as completely unfamiliar to t he bird being tested. The opposite
neighbor treatment w as the playback of t he familiar individual given from the opposite boundary from w hich it
w as normally heard. Thirteen male Northern Spott ed Owls w ere tested w ith 3 r esponse variables measured for
each owl w ithin each treatment. The response variables tested were latency t o first r esponse, number of calls
giv en, and av erage t ime elapsed bet w een each call giv en. Ut ili zing a crossov er des ign I blocked f or o ther
possible sources of variation including month and order in which the treatments w ere administered. Thus my
results focused on the treatments themselves; how ever, a significant t reatment effect w as only detected in the
latency variable. What can be seen from the blocked variables is that both the time of year and the order in
w hich t reatment s are administered can play a large role in a bird’ s response to c alls of c onspecif ics.

158     Romo de Vivar-Alvarez & Urbina-Torres
Songs of birds f rom San Blas, Nayarit Mexi co. CLA UDIA A . ROMO DE VIVA R - ALVAREZ* and FERNANDO
URBINA-TORRES, Lab. Ornit hol. Cent er Biol. Res., M orelos Stat e Aut onomou s Univ. , Cuernavaca, M orelos,
         This work w as made in the municipality of San Blas situated in the Mexican State of Nayarit, located in
the w estern portion of t he country, on t he Pacific Coast. The main objective of this w ork w as to design,
produce and publish an audiocassett e of bird songs of San Blas, Nayarit, in order to promot e to see and identif y
bir ds of this plac e. The pr ojec t w as done at tendi ng t he req uest of the in habi tant s of this tow n and w as
coordinated by the Nayarit State government as a collaboration agreement. A s a result w e obtained 7.5 hr of
songs r ecor din gs of 55 spec ies of bir ds, among t hem I got the song o f Whit e-eyed Vireo w hic h is a new
register for the region. We produced a tape of 40 min w ith songs and calls of 46 species of birds from w hich
41 are res iden t spec ies, 10 of them are end emic to M exico and f rom these speci es Lilac-c row ned Parr ot is Near
threatened (Collar et al. 1994, Birds to Watch 2(4):229); and the last 5 species are winter visitors.
Beneficiaries took part in t he development of t his material because they contributed w ith t heir knowledge about
species. The inhabitants of San Blas w ill be benefited w ith t his project trough the t ourist promot ion, the sale of
souvenirs and all they can learn about birds, allowing them to make use of flora and fauna in a new w ay which
means a double fun ct ion, c onservat ion and econom ical grow th. But t his w ork is st ill unf inished, n ow adays w e
are w ork ing to ob tain ec ono mical r esources in ord er t o pr odu ce copi es of the aud iov isual mat erial s t hat w e

159     Clark
On the natural history of Magellanic Penguin vocalizations. J. ALAN CLARK, Dept. Zool., Univ. W ashington,
Seattle, WA.
          The basic calls of wild adult M agellanic Penguins at Punta Tombo, Argentina, are described. These calls
inc lud e display call s, figh t call s, pair duet s, and c ont act call s. Some call s are used f or m ore t han o ne pu rpo se,
and b ehavior al co rrelat es of these c alls are also no ted. In addit ion , r ecor din gs of the di splay c alls of ind ividu al
males of know n age or reproductive success are analyzed. Elements of t hese calls are then correlated w ith
select male penguin physical characters. This analysis will likely suggest that display calls can provide
information on male quality and could thus be used by females during mate selection.

160     Lank & Dale
Visual v ersus vocal sig nals for indiv idual ident ifi cation : The silent song of Ruffs (Philomachus pugnax). DAV ID
B. LANK* , Dept. Bi ol. Sci., Sim on Fraser Univ. , Bu rnaby , BC; and JAMES DALE, Dept. Neurobiol. & Behav.,
Cornell Univ., Ithaca, NY.
         Numerous species of birds co mmunicat e individual identi ty thro ugh variatio n in vocal signals. In
contrast, breeding male Ruffs appear to do so exclusively through variation in visual signals. Male Ruff
courtship and aggressive displays are silent, but indiv iduals vary extremely in breeding plumage coloration and
patt erning, suggest ing an evolut ionary shif t in sensory modality for sig naling individu al identit y has occurred. If
plumage variation evolved to provide to signal individual identity, components of plumages should: 1) vary
ind epend ent ly of each o ther, 2) be a lo w cos t, select ively neut ral t rait ; an d 3 ) per haps have a genet ic, rat her
than environmental developmental basis (Dale et al. 2001, Am. Nat. in press). Our quantitative analysis show s
that variation in plumage characteristics provides a surfeit of information needed to identify individuals.
Plumage coloration appears to be neutral wit h respect t o male mating success, and the variation has a genetic
basis. We propose that the use of plumage variation to signal individual identity in Ruffs (rather then auditory
cues) w as favored pr imarily by lengt hy dayt ime male display occ urring in open habit ats in un usually cl ose
proximity to mult iple receivers. In addition signaling individual identity, plumage diff erences also signal how a
male w ill behav e, in t his spec ies w it h a un iqu e dim orp hism in male mat ing behav ior . This may have been an
additi onal, uni que fact or leading t o th e extraord inary plu mage diversit y in t his species.

161     Zink, Rohwer & Drovetski
Phylogeography of Eurasian Woodpeckers. ROBERT M. ZINK, Bell Mus., Univ. Minnesota, St. Paul, MN;
SIEVERT A. ROHWER and SERGEI DROVETSKI, Burke Mus., Univ. Washington, Seattle, WA.
         We sequenced parts of 3 mi toc hondri al genes in 67 Great Spott ed Woodpeck ers (Dendrocopos major)
taken from 17 sites across Eurasia and Japan. Two groups, differing by ca. 3% sequence divergence, have
had independent evolutionary histories: samples from the extreme southeast (Sakhalin, Primor’e, Hokkaido), and
the rest of Eurasia. From the United Kingdom to the Russian North Pacific, no phylogeographic divisions w ere
found. The reciprocal monophyly of haplotypes from t he 2 main groups suggests that t he southeastern group
should be recog nized as Dendrocopos japonicus. W e also com pared mt DNA sequences of Thr ee-t oed
Woodp eckers (Picoides tridactylus) from Eurasia and North America. Samples from the 2 continents are each
recipr ocally monop hy let ic, and d if fer by appr ox imately 3% sequence d ivergen ce, suggest ing that a separate
species exists on each continent. Within cont inents, no phylogeographic structure exists, suggesting that bot h
species have recently expanded their ranges following the end of the last ice age.

162      Harshman, Huddleston & Braun
C-myc and the high-level phylogeny of birds. J OHN HARSHMA N* , CHRISTOPHER J. HUDDLESTON, and
MICHAEL J. BRAUN, Lab. Mo lec. Syst ., Natl. Mu s. Nat. Hi st ., Smit hsonian Inst ., Washington, DC.
          We investigated the phylogeny of birds using sequences from part of t he nuclear oncogene c-myc,
inc lud ing all o f exon 3 and p art s of int ron B and t he 3 ' UTR. W e find st ron g suppo rt for Gal loan serae,
Paleognathae, Neognathae, and many clades wit hin Neoaves. Support for Neoaves itself is present but w eak.
We also discuss phylogeny w ithin Galloanserae and Paleognathae. The intron and the UTR are a richer source
of c haracters t han the exon , and inf ormat ive indels are present in all thr ee regions.

163     Bowie, Hacket t & Crowe
Av es largest genus (Nectarinia) takes a tum ble: Syst ematics and bi ogeography of t he w orld' s sunbirds. RAURI
C. K. BOWIE* , Percy FitzPatrick Inst., Univ. Cape Town, South Africa, SHANNON J. HACKETT, Field Mus.,
Chicago, IL; and TIMOTHY M. CROWE, Percy Fit zPatri ck Inst .
           Sunbirds are members of an Old World family (Nectarinii dae) of small nect arivoro us songbir ds. M ost
species exhibit extreme plumage dimorphism with male sunbirds typically having bright metallic plumages.
However, all members of the family have very similar skeletal structure and body plans. These similarities have
made it difficult for any consensus to be established with regard to 1) t he monophyly of t he family, 2) t he
number of genera and 3) the taxonomic association among genera. Recent classification approaches have
focused on bill variation as a key character and have considered plumage to be secondary w hen assigning
spec ies t o gen era. Usin g maximum likel iho od an d maximum parsimony algo rit hms w e have con st ruc ted
phy log enies of over 7 0% of the w orl d' s sunbi rds using b ot h mit ochon dri al (ND3 & ATPase6) and nuc lear
(Beta-fibrinogen Intron 5) DNA markers. Our results suggest t hat the nectarivorous Passeroidea are
polyphyletic, and that t he sugarbirds of southern Af rica are not members of the Nectariniidae as they are now
classified. The African sunbirds represent an old explosive radiation, derived from an Asian ancestral species.
How ever , t here i s some ev iden ce t o suggest a more recent back col oni zation from Afric a to A sia
Taxono micall y, our resu lt s do not suppor t recent attempts t o rec lassif y t he f amil y based on ph enet ic an alyses
of morphological characters, as none of the proposed polytypic genera are monophyletic in our analyses. We
present a novel classification scheme for the Nectariniidae, w hich is in general agreement w ith plumage
var iat ion among t axa.

164     Crowe, Bloomer, Randi, Lucchesi, Kimball & Groth
Phylogenetics of t he Galliformes: Sex, show y plumage, weapons and size are not important! or " What kind of
fowl am I? TIMOTHY M. CROWE, Percy FitzPatrick Inst., Univ. Cape Town, South Africa and Dept. Ornithol.,
Am . M us. Nat . Hist .; PAULETTE BLOOM ER, Dept. Genetics, Univ. Pretoria, Pretoria, South Af rica; ETTORE
RANDI, VITTORIO LUCCHESI, Istitut o Nazionale per la Fauna Selvatica, Ozzano dell'Emilia, Italy; REBECCA T .
KIM BALL, Dept . Evol. , Ecol. & Organ. Biol. , Ohi o St ate Uni v. , Columbu s, OH; and JEFFREY G. GROTH, Dept.
Ornitho l., Am. Mus. Nat. Hist ., New York, NY.
         Molecular and organismal evidence supports the monophyly of most traditional supra-generic galliform
taxa, but t he Phasianini (pheasants) and Perdicini (partridges/quails) are polyphyletic. Contrary t o results
suggested by DNA-DNA hybridization, New World quails (Odontophoridae) are not basal to guineafow l
(Numididae) and cracids (Cracidae) and megapodes (Megapodiidae) are not sister groups. A very early
divergence among taxa confined to the New World, Af rica and southeast Asia is hypothesized. Data from
mt DNA sequences (cy tochro me b and D-loop), proteins (ovomucoids), life history, behaviour, bones and
feath ers are analyzed separately, combin ed and diff erentially w eighted. Separate analysis produc es less
w ell-resolved trees. Deweighting ' less informative' evidence undermines phylogenetic signal. Analysis of
combined data produces a novel, biogeographically meaningful, t ree w ith organismal information being more
useful at basal (and molecular information at relatively terminal) branches. When 'adaptive' characters (e.g.
spurs, plu mage dic hro mat ism , m ating syst em, num ber o f tail f eathers , an d br igh t/f leshy ornam ent s) are mapped
on to the phylogeny, they are shown to have evolved multiply.

165     Lovette
What is a w ood-w arbler?: The molecular characterization of a monophyletic Parulidae. IRBY J. LOVETTE,
Cornell Lab. Ornithol. , It haca, NY.
        The wood-w arblers (family Parulidae) fall wit hin a radiation of passerine birds commonly know n as the
New World nin e-pr imaried osc ines The r econ st ruc tion of phy log enet ic r elat ion ships w it hin this assemblag e has
proven challlenging owing to its extremely high diversity, a paucity of phylogenetically informative
morphological characters, and an apparent high rate of cladogenesis early in the history of the radiation. Here, I
use both mitochondrial and nuclear DNA sequences to explore systematic relationships among the Parulidae and
other nine-primaried oscines w ith t he goal of determining w hether the 24 extant genera traditionally placed in
the Parulidae-or some subset of t hose genera-form a w ell-supported monophyletic group. Phylogenetic
reconstruc tion s based on a rang e of nuc leot ide-int ensive and t axon-int ensive analyses of mt DNA sequences
resu lt ed in many poor ly reso lved in terno des, but all r econ st ruc tion s co nt ained a w ell-suppor ted cl ade co mpris ed
of 18 genera, all of which are morphologically "typical" w ood-warblers traditionally placed in the Parulidae.
This same clade w as also highly supported in separate reconstruct ions based on nuclear-encoded loci. The
lon g, w ell-suppor ted, and i ndep enden tly con firm ed in terno de at the base of this w ood -w arbl er clade p rov ides
the op por tuni ty t o def ine a m ono phy let ic Paruli dae on several c omplem ent ary molecu lar p hy log enet ic c rit eria.

166     Fleischer
Phylogeny of the extinct Moho of Hawaii. ROBERT C. FLEISCHER, Smit hsonian Inst ., Washington, DC.
         The o` os (Melip hagidae; genus Moho) co nt ained 4 d escr ibed spec ies dist rib ut ed among the Haw aiian
Islands: M. nobilis on Hawaii, M. bishopi on Molokai, M. apicalis on Oahu, and M. braccatus on Kauai. I
conducted mtDNA sequence analyses on the group using DNA isolated from museum specimens in order to
det ermine t he ph ylogen etic r elat ion ships w it hin the gen us, and t o see if the t ree t opo log y supp ort ed
assumption s f or u se in a rat e cali brat ion based on i sland age. Cyt ochro me b and ATP sequences were
obtained, and preliminary trees rooted wit h Chaetopt ila and a few meliphagid outgroups indicate monophyly of
Moho and a t ree t opo log y t hat does cor relat e w it h island age (i.e. , Kauai i s basal, Molo kai an d Haw aii ar e a
deri ved clad e). A rat e cali brat ion using t he co rrect ed in terisland distances and K-A r is land ages is similar t o t hat
obt ained for Haw aiian = drepanidines and ot her rates estim ated f or bird s.

167     Slikas, James, Fleischer & Olson
Genetic relationships and evolution of f lightless morphology in the turt le-jaw ed moa-nalo of Kauai (Anatini:
Chelychelynechen quassus). BETH SLIKA S* , HELEN F. J AMES, ROBERT C. FLEISCHER and ST ORRS L.
OLSON, Smit hsonian Inst ., Washington, DC.
          The turtle-jawed moa-nalo, Chelychelynechen quassus, is an ext inct species of large, fli ght less
w aterfow l, endemic to the Hawaiian island of Kauai. Unlike the better-know n flight less waterfow l from t he
younger Hawaiian islands, Chely chelynech en was previously known from only 3 incomplete specimens. We
obt ained mitochon dri al gen e sequence ( cy tochro me b, ATPase8 cont rol region) of Chely cheynechen from
Holocene lake deposits in Mahaí uulepu Cave, and also studied the comparative osteology of new material from
consolidated dunes on Kauai, including the previously unknown cranium, sternum distal wing, and hindlimb.
Phylogenetic analysis of the sequence data co nfi rms t hat, like th e moa-nalos of t he younger isl ands (Ptaiochen
and Tham bet ochen), Chely chelynech en is most closely related to dabbling ducks (Anatini). Chely chelynech en
resembled the other moa-nalos in the extreme reduction of the w ing bones and hypertrophy of the axial and
hindlimb skeleton, but diff ered in such features as much larger size, turt le-like bill morphology, and sternum
shape. We evaluate our genetic and morphological evidence in light of t he hypothesis that the similarity in
flight less morphology betw een Chely chelynech en and the other moa-nalos can be att ributed to parallel evolution
from flying ancestors.

168     Omland, Marzluff, Boarman & Fleischer
Chihuahuan Raven phylogenetic relationships based on mitochondrial sequences and nuclear microsatellites.
KEVIN E. OMLAND, Dept. Biol., Univ. M aryland, Baltimore County; JOHN M. MARZLUFF, Col. Forest
Resources, Univ. Washington, Seattle, WA; WILLIA M I. BOARMAN, U.S. Geo log ical Survey , Univ . Cal if orn ia,
Rivers ide, CA; and ROBERT C. FLEISCHER, Smit hsonian Inst it ut ion, W ashin gt on, DC.
         Previous mitochondrial sequence data suggested that the Chihuahuan Raven is genetically nested within
the paraphyletic Common Raven (Omland et al., 2000, Proc. Roy. Soc. Lond. 2 67 B: 2 47 5). W e have added
addi tion al sequenc e dat a, an d add it ion al t axa t o f urt her t est this possibili ty. 92 5 b ase pairs o f cy tochro me b
sequence f rom mult ipl e repr esent atives of the " Calif orn ia Clad e" , " Holarctic Cl ade" and Ch ihu ahuan Raven
confirm t he paraphyly of t he Common Raven. The Chihuahuan Raven branches w ith t he California Clade of the
Common Raven in 100% of boot strap replicates. The Pied Crow (C. albus) f rom Afric a has prov en t o be a
good close outgroup for this raven study as it is sister to t he whole raven group and only 5% divergent in
coding sequence from each of the raven clades. We will also present microsatellite data from at least 4 loci t o
test t he paraphyly of t he Common Raven using nuclear markers. Finally, w e will present our plans for studying
the evolu tion of the 3 distinc t clad es of rav ens in w est ern Nort h A merica.

169      Safran
Female reproductive strategies and variation in group size of Barn Swallows: Sit e familiarity or the benefit s of
old nests? REBECCA J. SAFRAN, Dept. Ecol. and Evol. Biol., Cornell Univ., Ithaca, NY.
         It is w idely known t hat individuals employ diff erent strategies to solve the same problems, including
making decisions about habitat and mate selection. One research question of my study on the relationship
betw een individual reproductive strategies and semi-colonial breeding in Barn Swallows is: do f emales breed in
groups to increase their chances of using an old nest? In Tompkins Co., NY, there is a significant relationship
betw een the age of nest used during the first breeding attempt and clutch init iation date. Pairs that use old
nest s (construc ted in prev iou s years) lay earli er t han p airs that con st ruc t new nest s. Breeding ear lier has
fit ness consequences because birds that lay earlier have higher reproductive success. Interestingly, despite the
availability of old nests, some pairs construct new nests. Moreover, preliminary results suggest that birds w ill
return to breed at sites where old nests are experimentally removed during the w inter, suggesting a benefit of
sit e famil iari ty. More o ld n est s are av ailab le at larg e gro up sit es w here, on av erage, Barn Sw allo w s br eed
earlier than pairs that breed at solitary sites or in smaller groups. However, on average, the reproductive
suc cess of ind ividu als at larg e sit es is not hig her t han t he average r epro duc tive suc cess of ind ividu als t hat
breed at solitary sit es and in smaller groups, suggesting a cost of group breeding. The relative costs and
benefits of the presence of old nests and site familiarity in relation to breeding group size will be explored.

170      Schamel, Tracy & Lank
Extra-p air f ert ili zation in t he social ly pol yandro us, non territ ori al Red-necked Phalar ope. DOUGLA S SCHA MEL,
Dept. Biol. Sci. , Simon Fraser Univ. , Burnaby, BC and Dept. Biol. & Wildl. , Univ . A laska Fairbanks, Fairbanks,
AK, DIANE M. TRACY, Fairbanks, AK, and DAVID B. LANK, Dept . Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
         Males in polyandrous species have both extra-pair fertilization (EPF) opportunities and risks, and should
attempt to have other males care for eggs they fertilize, and guard against caring for eggs fertilized by others.
We predi ct ed lo w er lev els of ext rapai r y oun g f or l ater males in Red-necked Phalar opes, c ompared to t errit ori al
species, because male mate choice is not constrained by territorial considerations. EPFs were low overall:
1. 7% of 22 6 c hic ks, w it h 1 chi ck in eac h of 4 o ut of 63 clu tches. Second m ales of pol yandro us f emales faced
substantial risk, how ever, wit h 3 of 6 such clutches containing extra-pair young, in contrast t o none in a
female's first clutch (n = 25) and only 1 in 1 6 monogamous replacement clutches. First males w ere likely
fathers of t he EPF young of polyandrous females. Males mate guarded strongly. M ales increased wit hin-pair
copulation attempts coincident w ith, and beyond, the f emale' s fertile period. Paired males also sought extr apair
copulations wit h females re-entering the breeding pool. The risk of raising EPF young faced by later-nesting
males may limit polyandry (8 - 10 % of females) in this species through male discrimination against previously
mated f emales.

171      Etterson
An analytical method for estimating power for paternity exclusion using codominant genetic markers.
MATTHEW A. ETTERSON, Smit hsonian Mi grat ory Bird Cent er, Washington, DC.
          Paternit y analysis usi ng molecul ar markers is done eit her by searching a set of candidates f or t he most
likely parent (paternity inclusion), or by logical exclusion based on Mendelian Inheritance patterns. Inclusion
requires much more discriminatory pow er, and usually multiple potential fathers cannot be excluded. However,
in a sociall y m ono gamous spec ies t he nu ll h yp ot hesis is that the at tendan t male c oul d hav e sired a gi ven
of fspr ing . I assessed st atist ical pow er of pat erni ty ex clu sion (PE) and i nclusion (PI) us ing 6 n uclear
microsatellite loci for 174 offspring from 35 broods of Loggerhead Shrikes, a socially monogamous songbird. I
demo nstrate t hat PE is an up per b ou nd to PI. I also d evelop ed an analytic al metho d f or pr edic tin g average PE
based on allele frequencies and compare the predicted average PE to t he observed. Extra-pair fertilizations
accounted for 10 .6% of Loggerhead Shrike offspring. Av erage PE w as high (0.94), w hereas PI w as low
(0.4 3). Exp ect ed average PE w as an unbi ased and acc urat e pred ict or o f observ ed average PE. These analytical
methods could help researchers to precisely determine their needs prior to the expensive and time-consuming
process of developing microsatellite loci. This is also the first genetic assessment of paternity in a w ild
populat ion of Loggerhead Shrikes.

172      Dunn, Thusius, Peterson & Whittingham
Bigger is better: Mask size and mating success in the Common Yellowt hroat. PETER O. DUNN* , KEVIN J.
THUSIUS, KARA A. PETERSON and LINDA A. WHITTINGHAM, Dept . Biol. Sci., Univ. Wisco nsin-M ilw aukee,
Milw aukee, WI.
        The ev olu tion of male s exual or nament s in monog amous speci es is puzzlin g as sexual select ion is often
thought to be relatively weak in these species. How ever, sexual selection could be stronger than generally
believed if the males possessing larger ornamental traits are more likely to: 1) gain a mate, 2) mate with a more
fecund female or 3) gain extrapair fertilizations. We studied these measures of mating success in relation t o
mask size of male Co mmon Yello w thro ats. On our s tudy area in se. Wisco nsin t he bl ack facial mask var ied
almost 2 tim es in size among males. We found t hat males w ith larger masks gained both a social and extrapair
mating advantage. In terms of social mating success, males with larger masks were more likely to gain a mate.
There w as no clear evidence th at t hese males also gained more f ecund mat es. In a populat ion-w ide analysis,
males w ith larger masks w ere also more likely t o gain ext rapair fert ilization s. Similarly , in a pairw ise
comparison, ext rapai r si res h ad lar ger m asks than t he males t hat they cuc kolded . These r esults suggest that
sexual selection has the potential to inf luence mask size as a consequence of variation in both social and
genetic c omponent s of m ating su ccess.

173      Ramsay & Tuttle
Mat e choice and genetic relatedness in Whi te-t hroat ed Sparrow s. SCOTT M. RAMSA Y* and ELAINA M.
TUTTLE, Life Sci. Dept ., Indiana Stat e Univ., Terre Haute, IN.
          White-throated Sparrows are one of the few species in nature to exhibit strong negative assortative
mating. The basis of t he mating pattern in this species is an inversion on the second chromosome which is
ref lect ed in adul t crow n pl umage; ind ividu als w it h t an cr ow n st rip es hav e 2 n orm al co pies of the ch rom osome,
w hile white-striped individuals have 1 normal and 1 inverted chromosome. Over 95% of pairs are
heteromorphic for crow n plumage and the chromosomal inversion. This species has been suggested as a good
case for mate choice based on heterozygous advantage (Tregenza & Wedell 2000, Mol. Ecol. 9 :1 01 3-1 02 7):
w hite-w hite pairings would suff er from deleterious alleles accumulated on the inverted chromosome while
tan-tan pairs would suff er from competitive inferiority. The genetic compatibility hypothesis suggests that t here
is no single male who is the best partner for all females, rather categories of males who are best suited to
certain females. In this paper we examine genetic similarity using polymorphic A FLP markers in social pairs of
White-throated Sparrows from a population breeding at Cranberry Lake, NY. We compare the relatedness of
pairs to the background relatedness of the population as a w hole. Just as females may choose social partners
for genetic compatibility, because of constraints imposed by other females some individuals may resort to
ext rapai r mating ; w e com pare s ocial v ersu s extrapai r par tners to see w het her f emales cho osing ex trapai r mates
do so for genetic benefit s. We also compare simulated heteromorphic and homomorphic pairings. If t he
hom ozy got e disadvant age f or a t an-t an pai r li es solely in t heir com pet it ive abil it y, then w e pred ict that sim ulat ed
pairs will be at least as heterozygous as the population average. Because of the accumulation of deleterious
alleles, w e predict low er heterozygo sity in simulat ed w hit e-w hit e pairs.

174     Woolfenden, Stutchbury & Morton
The extra-pair mating behaviour of male Acadian Flycatchers. BONNIE E. WOOLFENDEN* , BRIDGET J. M.
STUTCHBURY, Dept. Biol ., York Univ ., Toro nt o, ON and EUGENE S. MORTON, Conserv. & Res. Center,
Smithsonian Inst., Front Royal, VA.
          The Acadian Flycatcher is a small, sexually monomorphic, neotropical migrant songbird. Despite the
fact t hat this species is relatively w idespread and abundant over much of its range, little is know n about the
mating sy stem, v ariance in reproductiv e success or the fact ors th at inf luence mate choice in either sex. In
ord er t o bet ter un ders tand t he rep rod uctive behav iou r of this spec ies w e have been m oni tori ng a b anded
population in Cambridge Springs, PA, since 199 8. In order to document t actics used by males to increase their
reproductive success w e are using a combination of behavioural observations, radio-telemetry and genetic
paren tage anal ysis. Prelim inar y r esults of our microsat elli te paren tage anal ysis sugg est that there ar e moderat e
levels of extr a-pair mating in this population. We w ill present t hese data as w ell as information on breeding
territory sizes, locations and the movement patt erns of males. Our results will be interpreted in the context of
the opportunit ies for sexual selection t o act on males. The possible effects of male song, morphology and
parental care on female mate choice are topics for future analysis.

175     Ibarguchi, Gaston, Boag, Gissing & Friesen
Over est imating ext ra-p air p aterni ty and un deres timating mutation : Lessons f rom Thi ck-billed Murr es. GABRIELA
IBARGUCHI* , Dept. Biol ., Queen' s Univ ., Kings to n, ON, ANTHONY J. GASTON, Canadi an Wild lif e Serv ice,
Hull, QC, PETER T. BOAG, GRAEME J. GISSING and VICKI L. FRIESEN, Dept. Biol., Queen's Univ.
         Molec ular markers hav e been i ndi spensable tool s in st udi es of mat ing syst ems w hic h t radi tion ally used
long-t erm behaviour al observati ons. Th e use of high ly v ariable molecular markers (e.g . mic rosatellit es and DNA
fingerprinting) has revealed extra-pair paternity (surprisingly high in some cases) in species previously
considered to be monogamous. Based on genetic paternity data obtained from Thick-billed Murres (fam.
Alcidae), we suggest that mut ation rates of some highly variable markers may be greater than is normally
acknowledged, and that there is a male-biased mutation rate. The results from t his study suggest that ext ra-pair
paternity may be overestimated in many studies that employ highly variable molecular markers. We show a
simple but pow erful statistical method f or discriminating betw een possible cases of mut ation versus extra-pair
paternity, based on the probability that t w o individuals, sampled simultaneously (e.g. at the same nest), share
half of the alleles at t he screened loci, b y chance (i. e. appear to be ' related' ).

176     Webster & Holmes
The Bateman gradient in a socially monogamous bird: Effects of EPF on male and female reproductive success.
MICHA EL S. WEBSTER* , School Biol. Sci., Washington State Univ., Pullman, WA; and RICHARD T. HOLMES,
Dept. Biol. Sci. , Dart mout h Coll., Hanover, NH.
           Sexual selection is driven by variation in mating success, and extra-pair fertilizations (EPF) may lead to
variation in mating success even in socially monogamous mating syst ems. For this reason EPF are thought to
create strong sexual selection, possibly even producing "hidden leks". W e examined the eff ects of EPF on
var iat ion in m ale and female r epro duc tive suc cess (RS) in a pop ulat ion of soc iall y m ono gamous Black -t hro ated
Blue Warblers. Variation in male mating success was higher when the eff ects of EPF were included, but t his
variation was still considerably less than that seen in most lekking species. Nevertheless, male reproductive
success w as strong ly associat ed w ith number of reproduc tiv e partners (a posit ive " Bateman gradient" ),
suggesting that any traits associated with ability to obtain EPF should be strongly favored by sexual selection.
Int erest ing ly, an d co nt rary t o no rmal expect ation , t he Bat eman grad ient w as also posit ive for f emales : f emales
w ith several reproductive partners had higher RS than females with f ew reproductiv e partners. Implications of
this result for t he evolution of female promiscuity w ill be discussed.

177     Hinam
* Habitat associations of five forest owl species in the Manitoba Escarpment. HEATHER L. HINAM, Dept. Zool.,
Uni v. Manitoba, Winnipeg, MB.
          The preliminary results of a habitat use versus availability analysis for 5 species of forest ow ls in the
Manit oba Escarpm ent are pr esent ed in this paper . Lo cat ion s f or Gr eat Horned Ow ls, Great Gray Ow ls, Barred
Owls, Northern Saw -w het Owls and Boreal Owls w ere obtained through noct urnal surveys run from Mar to Jun
199 9 - 20 00. Species' locations w ere overlaid on digital forest resource inventory maps and examined for
stand type, age and degree of canopy closure, which were then compared with similar data from random sites.
Barred , Gr eat Gray and Nort hern Saw -w het Ow l pl ot s co nt ained signif ican tly less unn atural open ing /b urn area
and the associated young open forest t han did random plots. In cont rast, Great Horned and Boreal Owl plot s
did not dif fer signi fican tly from rand om in t his resp ect . Boreal Ow l pl ot s co nt ained signif ican tly great er
pro por tion s of treed m usk eg and less nat ural open ing s and deci duo us f ores t. Barred Ow ls w ere associat ed
w ith f orests wit h high crow n closure values, w hereas Northern Saw -w het Owl plot s contained significantly
greater proportions of stands wit h moderate crown closure. The relationship betw een ow l habitat associations
and life history traits, as w ell as the application of these results to owl and forest management in Manitoba are

178     Hazler
* Habitat selection in a dynamic mosaic landscape: Is bird territorial behavior ideal? KIRSTEN R. HAZLER, Dept.
Zool . No rt h Carolina Stat e Univ ., Raleigh, NC.
         It has been suggested t hat territ ori al ani mals , such as bird s, distrib ut e themselv es amo ng h abit at
patches in an ideal despotic manner (= ideal dominance, Fretw ell & Lucas 1970, Acta Biotheoretica 19:
16-3 6). In ot her words, t he highest quality habitats w ill be preferentially selected, but late-arriving or
subord inat e ind ividu als m ay be forc ed t o set tle in low er qu alit y habit ats. I looked f or ev iden ce of an id eal
despotic distribution in 2 bird species, Acadian Flycatcher and Hooded Warbler, breeding in a dynamic mosaic
landscape. The study w as conducted in an intensively managed pine forest in w hich relatively mature stands
are juxtaposed wit h clearcuts and regenerating patches of int ermediate ages. Territories were mapped and
breed ing suc cess w as documented in 7 stands rang ing in ag e from 18 - 27 y r. The d ensity of Acadi an
Flycatch ers w ith in habit at pat ches w as positi vely relat ed to pr oduct ivit y (R2 = 0.92, F1,5 = 60 .6 , p = 0. 00 06 ),
strongly support ing the ideal despotic model. Habitat quality, measured in terms of the number of broods
fledged/ ha habitat , w as positi vely relat ed to c anopy height (R2 = 0.96, F1,5 = 117 .5, p = 0.0 001 ), and I
spec ulat e that cano py heig ht may serv e as a prox imate cue f or h abit at qual it y f or A cadi an Fly cat chers an d ot her
forest birds. Density of Hooded Warblers was relatively low so that statistical analyses were not appropriate.
However, Hooded Warblers did not appear to be distributed in an ideal despotic manner, and density may not
be a good indicator of habitat quality f or this species. I discuss how t he achievement of an ideal population
distribution may be limited by a species' ability to eff ectively sample potential habitats (Pulliam & Danielson
1991, Am. Nat. 1 37 : 5 50 -56 6), and the con servatio n implic ations.

179     Komar
* Effects on resident birds of shade trees in coffee plantations, and applications to coffee certification. OLIVER
KOMAR, Nat. Hist . M us. & Biodiv. Res. Cent er and Dep t. Ecol. & Evol. Biol. , Un iv. Kansas, Law rence, KS.
         The coffee certification criteria recommended by the Smithsonian Migratory Bird Center, and used by
the ECO-OK program of the Rainforest A lliance, include: > 40% canopy cover, > 10 t ree species/ha, and
> 20% of shade provided by emergent trees. I tested these criteria on resident birds in El Salvador, on 5-ha
plot s at 2 4 co ff ee and 5 natural f orest (c ont rol) sit es. Coff ee sites spanned gradients o f shade cov er (18 -
70% ) and tree diversity (4 - 39 species on 0.5 ha sampled wit hin the larger study plot s). Elevation, use of
agrochemicals, and distance from natural forest w ere kept relatively constant. I used a principal components
axis of 13 habitat variables as an index of disturbance. Many bird species in coff ee are generalists of no
conservation concern. To evaluate effects on resident birds, I considered 16 species negatively sensitive to t he
disturb ance i ndex -t hese speci es rep resen t pot ent ial b enef ici aries of ecol ogi cal c ert if icat ion . The gr oup inc lud ed
tw o gl obal -co ncern s peci es (Fan-t ailed Warbler and W hit e-eared Grou nd-Sparrow ), and 1 0 o ther
nationally-threatened species. The best single habitat predictor for bird richness and abundance w as canopy
cover . The st udy pro duc ed model s of the ef fect s of habi tat features on r ich ness and ab und ance o f threat ened
species of birds, applicable to the refinement of certification criteria for biodiversity-friendly,
enviro nment ally -f rien dly , o r bi rd- frien dly cof fee.

180      Wightman
* Patt erns associat ed w it h br eedin g hab it at qual it y in an i ncreasing p opu lat ion of Peregrine Falc ons in c ent ral
west Greenland. CATHERINE S. WIGHTMAN, Rapto r Research Center, Boise State Univ. , Boise, ID.
          I evaluated patterns of distr ibution and nest-site selection, and characterized the physical features of
quality nest sites in an increasing population of Peregrine Falcons in central West Greenland. I used occupancy
and p rod uctivit y data from 19 72 - 199 9, pro vided b y t he Green land Peregrine Falc on Su rv ey, to evaluat e
distribution and habitat selection patterns. Variance in productivit y w as significantly low er at traditionally
rather than recently occupied cliff sites, which suggests that traditionally occupied cliff sites are better quality.
This also indicated that Peregrine Falcons occupy breeding sites according to a pattern of despotic distribution.
Consistent ly oc cupied cli ff sites pro vided hig her produc tiv ity than in consist ently occupi ed clif f sit es, and th us,
these sites were also better quality. I measured habitat features at 67 occupied cliff sites in summer 1998 and
1999 and created two models using these features to predict traditional and consistent occupancy at cliff sites.
Features of good quality habitat included the height or elevation gain of the cliff and protection from w eather on
the eyrie ledge. Spacing of suitable and occupied cliff sites also w as an important f eature, and the best cliff s
gener ally w ere more i solated. I suggest this associat ion w it h isolat ion resu lt s f rom the ben efit s of decr eased
intra-specific competition from neighbors and floaters.

181      Fletcher
* Understanding spatial distributions of an area-sensitive passerine near different edge types. ROBERT J.
FLETCHER, Jr., Dept. A nimal Ecol., Iow a State Univ. A mes, IA .
          With on goi ng f ragm ent ation and h uman dev elop ment, t here i s gr ow ing con cern abo ut how spec ies
respond to human-altered environments. I investigated spatial distribution patterns of Bobolinks near 3 edge
types (agriculture, road, and woodland) in n. Iowa. Specifically, I w as interested if t his area-sensitive passerine
w as less abun dant near ed ges, and i f so, w hat pot ent ial m echan ism s explai ned t hese patterns . I inv est igat ed
four alternative hypotheses for edge avoidance: 1) habitat structure gradient, 2) conspecific attraction, 3)
territ ory size, and 4 ) act ive avoidan ce. Based on sur vey dat a, b obo lin ks w ere less abundan t near al l edg e typ es
(w ithin 75 m) than near the interior of grassland habitats. There was no evidence for t he habitat gradient
hypothesis explaining abundance near any edge type. How ever, territory mapping revealed diff erent
mechanisms for t hese similar abundance patterns. There was w eak evidence for conspecific attraction
occurring near road edges. Territory size tended to increase near road edges, but showed no consistent pattern
near agriculture and woodland edges. Bobolinks actively avoided road and woodland edges, but not agriculture
edges, w here t errit ory bou ndar ies w ere di splaced aw ay f rom the edg e. In add it ion , i solated bi rds tended to bi as
their locations tow ard the interior when near woodland edges. I conclude that bobolinks do indeed actively
avoid some edge types, but t he mechanisms underlying these spatial patt erns are complex and diff erent for
different edge types.

182      Cederbaum
* Effect s of alt ernativ e agricult ural cropp ing met hods on f ield use by song birds. SANDRA B. CEDERBAUM ,
War nell Sch. Forest Res., Uni v. Georgia, A thens, GA.
         Significant agricultural advancements over the past 50 years have altered the landscape and negatively
affected the avian populations associated w ith early successional habitat. A mong the major agricultural crops
in t he Sou theast , c ot ton i s generall y c onsider ed t o pr ov ide t he least suitable h abit at for m ost early suc cession al
songbi rds . High insect ici de and herb ici de use, an d in tensive t ill age ch aracteristic o f cot ton p rod uce a barr en
land scap e for w ild lif e. New er crop pin g sy st ems such as use of con serv ation till age and st rip cover cropp ing of fer
hope for improving the value of cott on fields to songbirds. Stripcover cropping refers to the practice in w hich a
row of cover crop, in t his case clover, is left t o grow betw een crop row s throughout t he early portion of t he
grow ing season. During 199 9 and 200 0, w e examined the eff ects of strip-cov er planting, in conservation
tillage, versus conventionally grow n cott on, in both conv entional and conservation tillage, on the avian and
arthropod species composition and field usage in Jefferson County, Georgia. Surveys of avian populations
w ere performed w eekly on each field wit h distance sampling methods used to determine species richness and
density. A rthropod populations w ere sampled bi-weekly on each field using pitf all cups. Preliminary arthropod
results suggest greater insect abundance in stripcover f ields. Stripcover and conservation tillage fields
cont ained hig her n umbers of ind ividu al species, w hil e st rip cover field s show ed hi gher trend s f or av ian
detections. Our findings suggest that both conservation tillage and strip-cropping systems will improve
conditions for birds in cotton fields, with strip-cropped fields attracting greater bird densities.

183      Saab & Vierling
Role o f non-equilibrium dy nam ics in creat ing so urce habitats f or cav it y-nesting birds. V ICTORIA A. SAAB,
Rocky Mountain Res. Sta., Boise, ID; and KERRI T. VIERLING, Dept. Chem., South Dakota Sch. Mines & Tech.,
Rapid City, SD.
          Many cav it y-nest ing bir ds ar e associat ed w it h po nder osa pine (Pinus ponderosa) forests of w estern
North Am erica. Woodpeckers, particularly, are dependent on t hese fire-prone landscapes for port ions of their
lif e history . A lt hou gh p ond eros a pin e fores ts w ere t yp ical ly main tained by frequ ent , l ow -sev erit y surf ace f ires
(equilibrium dynamics), recent evidence indicates that large, stand-replacing crow n fires (non-equilibrium
dynamics) occurred historically on an infrequent but regular basis. We determined the relative importance of
crow n-burned, ponderosa pine f orests t o Lewi s’ s Woodp eckers by evaluat ing t heir demographic s in f orests
affected by equilibrium and non-equilibrium conditions. We compared their reproductive success, productivity,
and potential source-sink status in crow n-burned, ponderosa pine forests and cott onw ood (Populus fremontii)
forest patches. Daily nest survival rates w ere significantly low er in cottonw ood compared to burned pine
for ests. Nest ing succ ess w as 46% (n = 65 ) in cot ton w ood f orests and 7 8% (n = 28 3) in b urned pine f orests.
Proportion of nests destroyed by predators w as significantly higher in cott onw ood forests (34 %) compared to
bur ned p ine f ores ts (1 6% ). We consist ent ly foun d cr ow n-burn ed f ores ts t o be p ot ent ial sour ce hab it at,
w hereas cottonw ood sites were more often predicted to be potential sink habitat. Cott onwood f orests were
surrounded primarily by agriculture w here the composition and abundance of nest predators w as likely diff erent
than t he pr edat or as semb lage o ccupy ing a larg e-scale bu rn i n a rel atively nat ural land scap e. Prescribed
und erst ory fire i s t he pr evailin g managem ent tool for r est ori ng p ond eros a pin e ecos ystems. Cond it ion s cr eated
by crow n fire may be equally important in maintaining ponderosa pine systems, including the conservation of
nesting habitats for Lewis’ s Woodpecker.

184      Kim, Chavez-Ramirez & Slack
Impacts of w oody st ruct ure on patch use by t hree win tering gr assland raptor species. DANIEL H. KIM* Dept.
Wild. & Fish. Sci., Texas A&M Univ., College Station, TX and CKWRI, Texas A& M Univ. -Kingsville, TX , FELIPE
CHAVEZ-RAMIREZ, World Wildlife Fund, Mexico, and R. DOUGLAS SLACK, Dept . W ild . & Fish. Sci., Texas
A&M Univ.
        We measur ed t he im pact s of art if ici al per ches upon p atch use by Logger head Sh rik es, American
Kestrels and Northern Harriers along the coastal prairie of s. Texas in the wint ers of 199 8 - 19 99 and 19 99 -
200 0. Study sites were Matagorda Island and the Norias division of the King Ranch. Perch manipulations
consisted of patches wit h 25 perches, w hile controls were native vegetation wit h and wit hout natural w oody
structure. Manipulations differed by height in 1998 - 1999 and by canopy structure in 199 9 - 200 0. Patch
use was influenced by differences in migratory chronology , interspecific interactions and both fine grain and
course grain habitat composition. Harriers arrived to the study areas last, and used plots in accordance to t heir
availability. Patch use by kestrels was dictated by course grain habitat composit ion (at the landscape level) and
fine grain habitat composition (at t he patch level). Finally, patch use by shrikes was dictated by interference
competition f rom Kestrels, and predatory pressure by Harriers. Ultimately shrikes were limited by the
availability of escape cover.

185      Stein, W illiams, del Rio & Place
Seasonal and age-related modul ation o f di gestiv e func tio n in mig rating Western Sandpip ers. R. W. STEIN, T. D.
WILLIAMS, Dept. Bi ol. Sci., Sim on Fraser Univ, Burnaby, BC, C. M. DEL RIO, Dept. Zool., Univ. W yoming,
Laramie, WY, and A. R. PLACE, Inst. Marine Biotech., Univ. Maryland, Baltimore, MD.
          The current understanding of t he functional significance of digestive phenotypic f lexibility is largely the
resu lt of cou pli ng o pt imal di gest ion models w it h empir ical st udy . This bod y of inf orm ation demonstrat es t hat
dig est ive tract volum e, d igestive enzy me conc ent rat ion s, micro-nut rien t trans por ters, and d igesta passage rat e
are modulated in response to changes in energy demands and/or diet composition. The diet of the Western
Sandpiper is composed of invertebrates; how ever, the species and their energy composition are not well
characterized. This presents a challenge to understanding the role of diet in meeting t he energetic demands of
migrat ion . To assess int rasp ecif ic v ariat ion in m igr atory perf orm ance i n West ern Sandp iper s, w e measu red
morph olo gic al ch aracteristics of the di gest ive tract and ac tivit ies of three d igestive enzy mes. Season al
comparisons were made for adults and age-related comparisons were and during fall. Results show that in
spr ing adul ts hav e a smal ler p rov ent ric ulu s, and h ave high er chit inase, am ino pept idase-N, and m alt ase act ivit ies
than in fall. The elevated levels of digestive enzymes in spring are indicative of a higher diet quality.
Modulating digestive enzymes is an efficient means of increasing energy extraction when diet quality is high.
Thi s suggests t hat diet qual it y inf luen ces m igr atory perf orm ance. During fall m igr ation juv enil es hav e a lig ht er
pro ventric ulu s and a low er chit inase act ivit y t han ad ult s; how ever , j uv enil es also have a lon ger an d heav ier
small intestine. A longer small intestine indicates that juveniles ingest a larger volume of food t o meet their
energy demands. However, enzyme activit ies suggest that there is no substantial age-related diff erence in Fall
diet composition. Therefore, juveniles may have lower foraging proficiency.

186      Hobson
Linking br eeding and w int ering popul ations o f Loggerhead Shrik es using st able isotop es. KEITH A. HOBSON,
Canadian Wildlif e Service, Saskatoon , SK.
          The ability t o link breeding and wint ering sites of declining or threatened species is fundamental to t heir
effect ive con serv ation , b ut prev iou s co nv ent ion al met hod s such as bandin g hav e pro ven larg ely unsuccessful at
the population level. New approaches to tracking migratory birds in North A merica using stable isotope
measurements of feathers can assist in the determination of origins w here feathers w ere grown, and thereby
aid in delineating origins of migrants on w intering grounds or stopover sites. The Loggerhead Shrike is
decl ini ng t hro ugh out most of it s ran ge bu t var iou s f act ors may oper ate on d if ferent subpop ulat ion s. We used
the st able-carbon and h yd rog en isot ope v alues in t he ou ter t ail f eather o f Logg erhead Shrikes w int erin g in Texas
(n = 70), Florida (n = 121 ), Georgia (n = 24), 1 997 - 1998 , and n. Mexico (n = 72), 1 999 , in order to
ascertain the minimum proportion of w intering birds that w ere northern migrants. A pproximately 10 % of
w int erin g shrik es in Florida w ere no rt hern migran ts co mpared w it h 8 % for t he M exican s ample and onl y 4 % of
the Tex as popul ation . W e det ect ed no nor thern migran ts in the Geor gia p opu lat ion . Car bon isotope v alues
varied considerably indicating a broad range of C-3 to C-4 based foodw ebs used by shrikes. Our study
indicates that the stable isotopic technique answ ers a number of longstanding and fundamental questions
concerni ng t he breeding or nat al origins of w int ering popul ations o f Loggerhead Shrik es, and th ereby prov ides a
templ ate fo r isot opic appro aches to ecolo gical inv estigat ions of oth er migrat ory sp ecies.
187     Yen, O' Hara & Lank
The eff ect of ult raviolet radiation on primary flight feathers of Western Sandpipers: An unappreciated cost of
wintering closer to the equator. PEGGY P. W. YEN* , PATRICK O'HARA and DAVID LANK, Dept. Biol., Simon
Fraser Un iv., Burnaby, BC.
         Ult rav iol et radi ation (UV R) degrades k erat in, the f ibr ous pro tein m olec ule t hat is t he main struc tural
mat erial in f eathers . Si nce UVR int ensity is higher in t he t rop ics , w e expect rat es of UVR-induced f eather w ear
to be h igh er f or b ird s w int erin g cl oser to t he equ ator. We presen t a met hod for i ndex ing the amoun t of feather
w ear potentially att ributable to UVR, and preliminary results from Western Sandpipers. We compared 9th
primary feathers collected at a southw ard migratory stopover site in the Fraser River Delta, Canada, and
w intering sites in n. Mexico and Panama. We examined feathers midway along the vane, 2 mm in from t he
barb ends. We indexed "dif ferential wear" by assigning a wear class score to the outer part of the vein, w hich
is usually exposed to UVR, and subtracted the score of the inner vane, w hich is usually shaded from UVR by
the ov erlapping adjacent p rimary . Dif ferent ial w ear increased during t he season, for primaries f rom juv eniles (p
= 0.0018) and adults (p = 0.005), and differed among countries for juveniles (p = 0.0007), but not adults.
Overall, f eather w ear increases wit h th e tot al UV t o w hich f eathers have been exposed. Our t echnique of fers a
novel approach tow ards quanti fy ing t he w ear attr ibut able to UV , and event ually assessing t he diff erential cost s
of spending more time in areas with higher solar intensity.

188     Bonter & Donovan
Temporal, spatial and competition f actors inf luencing migration st opover site quality. DAVID BONTER* ,
Vermont Coop. Fish & Wildl. Res. Unit, Univ. Vermont, Burlington, VT and Braddock Bay Bird Observ.,
Rochester, NY, and THERESE M. DONOVAN, Vermont Coop. Fish & Wildl. Res. Unit.
        The ab ili ty of a mig rant to im pro ve its ener get ic c ond it ion dur ing st opo ver may be used as a su rrogat e
measure of stopover site quality. Regressing the energetic condition of migrants on capture time is one method
for assessing the rate of energetic condit ion change in migrants. Analyzing diff erences in these rates under the
influence of covariates provides information on how various factors may affect stopover site quality. We
examined the influence of bird density, t ime of season, season, year, and location on energetic condit ion and
rates of energetic condit ion change during stopover. Data from over 31 ,00 0 landbirds captured during
migration at 2 research sites on the south shore of Lake Ontario during 1999 and 2000 w ere analyzed. A
density-dependent effect w as detected during spring and fall migration, w ith birds captured on low migration
volume (density) days being in significantly superior energetic condition w hen compared to birds captured on
high-density days. In addition, t he hourly rate of energetic condition change was significantly higher on
low -density days for the Hermit Thrush. Annual, seasonal, and site-specific dif ferences in stopover site quality
w ere also detected. While migrants require high-quality st opover sites for successful migration, numerous
factors cont ribute to variability in habitat quality at a given site. Identification of optimal stopover locations for
broad-front landbird migrants is complicated by this variability.

189     Payne, Harrington & Temple
Are trends in shorebird populations correlated wit h trends in spatial use during migration? Implications for
management. LAURA X. PAYNE,* Dept. Wildl. Ecol., Univ. Wisconsin, Madison, WI; BRIAN A. HARRINGTON,
Manomet Center Cons. Sci., Manomet, MA; and STANLEY A. TEMPLE, Dept. Wildlife Ecol., Univ. Wisconsin.
          The recent (200 0) U.S. Shorebird Conservation Plan (USSCP) identified significant (n = 16) or apparent
(n = 8) declines in 24 of the 50 shorebird species that migrate and breed in the U.S. An import ant driver of
population decline is believed to be loss or disturbance of w etland habitat, alt hough this relationship is not well
quantified. To reverse declines, practical management must identify c auses, and target groups (e.g., guilds
based on habitat requirements) rather than single species. One index of response to habitat loss is a change in
spat ial di spersi on. Using Int ernat ional Shorebird Surv ey co unt s of 30 shor ebird species at > 10 00 U.S.
w etlands during autumn migration (19 78 - 1 990 ), w e quantif ied spatial dispersion for each species using site
aggregation indices. We then examined temporal trends in aggregation indices and compared species, guilds
and years using ANOVA/regression. Twenty -tw o (63% ) species showed slight t o strong tr ends in aggregation.
Of 18 declining species, 56% have become more aggregated, 11 % are more dispersed, and 33% show ed no
trend. In 3 guilds (mud, w ading, upland), no trends w ere apparent at t he guild level; w ithin t he oceanfront
gui ld, 5 o f 7 speci es bec ame more ag greg ated. Our resu lt s suggest that chan ges in spatial d ispersion m ay
reflect losses in habitat, pot entially indicating decline-prone populations; additionally, t hough practical,
guild-based management m ay mask broader-scale requirement s of so me species.

190      Holberton, Wilson & Hunter
The Blackpoll Warbler: Little bird - big journey. REBECCA L. HOLBERTON,* Dept . Biol. Sci., Univ. Maine,
Orono, ME; C. MORGAN WILSON and MEREDITH J. HUNTER, Dept. Biol., Univ. Mississippi, Oxford, MS.
          We are currently investigating t he behavior, ecology and physiology of the Blackpoll w arbler, a small
bird that makes one of the longest round trip songbird migrations in the Western Hemisphere. Thus far, we
have found that t he Blackpoll prepares for it s initial overland departure from t he breeding grounds at Churchill,
Manitoba, in a similar way that the co-occurring shorter-distance migrant, the Yellow-rumped Warbler does. At
Churchill, we f ound no difference between the 2 species in fat reserves, energetic condition, and patterns of
corticost erone secretion and triglyceride levels. However, at coastal Maine, w here many Blackpolls prepare for
a 4 - 5 d , non -stop trans-oc eanic flig ht t o Sout h Am erica (oft en doubling their l ean body mass t o do so),
Blackpolls had higher baseline corticosterone and triglyceride levels than they expressed at Churchill;
Yellow-rumps show ed no change. In the laboratory, Blackpolls retained greater fat reserves and higher baseline
cort ico st eron e and t rig lyceride l evels lat er in to t he aut umnal m igr atory peri od t han Yello w -ru mps did. Studi es
on Blackpoll migratory orientation patterns related to weather, geographic barriers and energetic condition are
also und erw ay. In summary, the Blackpoll ' s abilit y t o un derg o extreme mig rat ion s mak e it an excell ent syst em
for investigating many behavioral and physiological capabilities and constraints underlying bird migration.

191      Carey
* Shorebird habitat requirements at an inland stopover site. CORINNE A. CAREY, Dept. Biol. Univ. North
Dakota, Grand Forks, ND.
          Large numbers of shorebird species migrate through the interior of North America each year.
Dependabl e, n ut rien t ric h in land sit es are essent ial f or s urv ival of migrat ory shorebi rds . W hil e there h as been an
emergence of shorebird research, litt le is known of t he consequences of shorebird habitat management to the
invertebrate prey population or other biota. As a prerequisite to development of a migratory shorebird
management plan and to provide a baseline for comparison of pre-management and post-management
populations, I documented the migration chronology, species composition and relative abundances of shorebirds
at Kellys Slou gh Natl. Wildl. Refug e, an inland st opover sit e in the prairi e potho le region of Nort h Dakota. I also
samp led sever al po ols to do cument num bers and d iversity of the in ver tebrat e prey . I plan to manip ulat e w ater
levels in a po ol an d do cument the ef fect s of the manip ulat ion on shor ebir d nu mbers an d t he in ver tebrat e prey
base. My hypothesis is that an interspersion of moist soil mudf lat and shallow w ater wit h scattered pockets of
deep water will provide optimal shorebird habitat and increase the number of shorebirds that utilize the area. I
w ill doc ument the ch anges in t he in ver tebrat e pop ulat ion and v eget ation that occ ur as a result of w ater lev el
manipulations. I w ill also document the associated use of t he manipulated pool by other bird species like ducks
and geese. These data w ill provide shorebird managers the information necessary to t ime pool impoundments
or drawdow ns w ith local peak shorebird numbers and seasonal migrations. Ultimately, long-t erm management
of mud f lats and shallow w ater will insure availability of required habitat and contribute to t he overall survival of
migratory shorebirds.

192      Nebel
* Western Sandpipers during t he non-breeding season: Spatial segregation on a hemispheric scale. SILKE
NEBEL, Dept . Bio Sci. , Simon Fraser Un iv., Burnaby BC.
       The Western Sandpiper is a migratory shorebird that breeds in Alaska and overwint ers on the Pacific
and Atlantic coast of the Americas. Based on 19 unpublished data sets (British Columbia to Ecuador on the
American w est coas t and So ut h Caro lin a to V enezuel a on t he east ) t he di fferent ial d ist rib ut ion of age and sex
classes of Westerns Sandpipers during the non-breeding season is documented. With increasing distance from
the breeding grounds, females become more abundant relative to m ales. The distribution of juv eniles along the
Pacif ic c oast is u-shaped, w it h t he major it y of juv enil es w int erin g f urt her n ort h t han ad ult s. Thr ee fact ors that
could influence the differential distributions are discussed. 1. A n increase in UV-induced feather wear closer to
the equator could affect juveniles more strongly than adults, due to differences in moulting strategies. 2.
Females ar e pred ict ed t o be m ore v uln erabl e to pr edat ion , and pr edat or ab und ance i s pr edic ted t o be l ow er at
southern latitudes, resulting in the higher proportion of females in the south. 3. Densities of invertebrates living
in t he upper int ertidal lay er are predicted t o be low er in th e south c ompared t o th e north due to t he intense
competition by cr abs. Consequently, females, w hich have a longer bill than males, would be better suited to
feed at southern latitudes than males.

193      McGraw , V onnegut, Dale & Hauber
Do different plumage colors reveal different information? How nutrit ional stress affects t he expression of
melanin- and structurally based ornamental coloration. KEVIN J. McGRAW* , EMIKO A. VONNEGUT, JAMES
DALE and MARK E. HAUBER, Dept. Neurobiol . Behav., Cornell Univ. , It haca, NY.
         Evolut ion ary bio log ist s hav e show n gr eat recent int erest in t he t yp es of inf orm ation that can b e signaled
by avian plumage colors. In many species of bird, t he brightness of carotenoid-based plumage reflects the
health and condition of indiv iduals and is used in mate selection. The information cont ained in melanin- and
struct urally based ornamental colors in birds is less well resolved. We subjected captive male House Sparrows
and Brow n-headed Cow birds to stressful nutr itional conditions during molt t o test t he hypothesis that melanin-
and structurally based plumage colors are nutrit ionally condition-dependent. We restricted food access for
treatment males during randomized 6-hr periods on 4 d/w k, w hile allow ing control birds ad lib access to food
throughout the course of molt. W e found that t he size and darkness of the melanin-based throat badges in
male House Sparrow s w ere no t affect ed by nut rit ion al st ress. Si milarl y, there w ere no dif ferenc es bet w een
treatment and control male cowbirds in t he size or darkness of the melanin-based brow n hood. How ever, the
struct urally based iridescent plumage of cow birds was indicative of t he nutritional condition of m ales during
molt. Nut ritionally stressed cowbirds grew significantly less colorful plumage than did males receiving ad lib
access to food. These results are consistent w ith observations in other avian species that diff erent types of
plu mage col or c ommunic ate dif ferent set s of inf orm ation . M elani n or nament s are less sen sit ive to nu trit ion al
condi tion s durin g molt and i nst ead may r eflect the ho rmonal st atus and/ or c ompet it ive abil it y of males , w hereas
struct ural coloration appears to be an honest signal of health and condition.

194      Berres
BADGER: A new bioinformatic tool for high-resolution genotyping. MARK E. BERRES, Dept. Zool. and Zool.
Mus. , Univ. Wisconsin, Madison, WI.
        PCR-based DNA fing erpr int ing techn olo gies have great pot ent ial t o pr ov ide an unp recedent ed nu mber
of genet ic m arkers u sef ul f or p opu lat ion -lev el anal yses. How ever , t he major it y of these m ethod olo gies gener ate
dom inan t markers: dom inan t hom ozy got e and h eterozyg ot e geno typ es can not be di st ing uished f rom each o ther
using conventional electrophoretic methods. In contr ast, co-dominant markers allow unambiguous distinct ion
among each genotype and therefore permit estimation of allele frequencies. In principle, the genotype of a
PCR-derived dominant marker may be inferred from the molar mass of DNA present in the marker assay,
enabling co-dominant information to be t eased out of data that apparently confound homozygot e and
heterozygote genotypes. How ever, due to various sources of random variation attributed primarily to t he
marker assay it self and t he det ect ion pro cedu re, measu rement s of DNA molar m ass at a giv en lo cus differ
greatly among sampled individuals, requiring subtle transformations of t he data. Using an existing
dominant-marker assay (AFLP; Am plified Fragment-Length Polymorphism), I demonstrate an analytical method,
BADGER (Bi-Allelic Discrimination of Genotypes wit h Enhanced Resolution; patent pending), designed to assign
co-dominant geno typ es - all elic inf orm ation onc e prev iou sly lost in t radi tion al pr esenc e/ab senc e assay s. The
BADGER genotyping procedure is applicable to a broad class of genetic markers and marker-detection
techn ologies and requires onl y a one- or t w o-dimension al vecto r of spectr al intensit ies (representing DNA
abundance) sampled at regular t ime int ervals. A s an example of t his t echnology applied to natural po pulati ons,
I also pr esent evidenc e of signif ican t dif ferenc es in the po pul ation -genetic s truc ture o f Whit e-bearded an d Blue-
backed Manakins and W hit e-tailed Sabrew ings.

195     Proudfoot, Slack & Honeycutt
Comparison of Ferruginous Pygmy-Owl mtDNA at local and international scales. GLENN A. PROUDFOOT* , R.
DOUGLAS SLACK and RODNEY HONEYCUTT, Dept. Wildl. & Fish., Texas A& M Univ. , College Station, TX.
         To assess p hy log eogr aphi c v ariat ion among Ferrug ino us Pygmy-Ow ls and evaluat e neces sit y f or s peci al
management of t his "endangered" species in the U.S., w e extracted, amplified, and analyzed 899 base pairs
(bp) of mit ochond rial deoxy ribonu cleic acid (mt DNA). mt DNA w as extract ed from tissu e (n = 70 ) or blood (n
= 29) w ith a modified Chelex extraction. Samples w ere obtained from ow ls in Texas (n = 18), Arizona (n =
18), and 11 States in Mexico (n = 63). Target gene regions w ere amplified by polymerase chain reaction
(PCR). Using pri mers design ed f or o w ls, a 11 00 bp p iece o f the cy tochro me b (cyt b) gene was amplified for
comparative analysis. Double-strand manual sequencing was performed with the same pairs used in PCR.
Ligh t and h eavy st rand s w ere sequenc ed f or c yt b. Standar d 6 % acrylamid e gels w ere us ed and the sequenc es
w ere v isualized w it h an au tomated sequenc er (Perk in Elm er® 37 7). Seq uenc es w ere compared for s imilar it ies
and differences. A t otal of 2 9 haplotypes w as observed that diff ered by only 1 to 10 bp each, resulting in
levels of haplotype divergence from 0.1 % t o 1.0 %. Haplotype comparisons revealed low levels of variation in
the Ferruginous Pygmy-Ow l. How ever, phylogenetic analysis of haplotypes shows that the Texas and Arizona
populations are distinct entities because there are 2 discrete groups of haplotypes corresponding to each area.
Hence, separate management may be warranted for populations in Arizona and Texas. Given the extreme
similarity of the 29 haplotypes, it is possible that the population as a whole underw ent tw o bottlenecks
sometime in the not-too-distant past, ow ing to t he eff ects of low population number, inbreeding, or both,
possibly advocating consideration as a population in peril. To advocate special management for t he Ferruginous
Pygm y-Ow l po pul ation s at this time, h ow ever , m ay be ill -adv ised, u nless ot her g enet ic m arkers such as
micro-satellites show low level genetic variation.

196     Barrowclough & Groth
Phylogeny o f Strigiformes based on a large nuclear exo n. GEORGE F. BA RROWCLOUGH* and JEFFREY G.
GROTH, Am . M us. Nat . Hist ., New York, NY.
        We at tempted t o ob tain a r obu st phy log eny of the ow ls by sequen cin g geo grap hic ally w idespread
repr esent atives of all av ailab le gen era f or a l arge f raction (about 2. 9 k b) of the nu clear RAG-1 exon. Boot st rap
analysis of both parsimony and parameter-rich likelihood approaches yield strong support f or the existence of
three clades of strigid ow ls; these include a basal clade composed of Ninox, and 2 large, multigeneric
monophyletic groups. These include 1) a clade composed of Glaucidium, At hene, and Aegolius, and 2) a clade
composed of Otus, Strix, Bubo, and Asio. The genus Otus appears to be paraphyletic and is composed of
separate New World and Old World radiations. Phodilus is sist er to t he genus Tyto.

197     Brito, Barrowclough & Groth
Phy logeography co nt radict s cur rent Taw ny Ow l t axonom y. PATRICIA H. BRITO* , GEORGE F.
BARROWCLOUGH and JEFF G. GROTH, Am. Mus. Nat. Hist ., New York, NY.
         The subspecific taxonomy of the Tawny Ow l (Strix aluco) is com posed of a European gr oup (nominat e
aluco) including up to six subspecies and an Asian (or nivicola) gr oup w it h 4 subspec ies. The c harac ters u sed
in this classification w ere plumage color and w ing length. Due to considerable variation in plumage coloration,
subspecies boundaries have been difficult to define and authors often did not agree in their classifications.
Additionally, some authors have suggested that some of the described subspecies may represent distinct
spec ies. A phy log eogr aphy st udy w as undert aken w it h w idespread samp les f rom West ern Eur ope an d a f ew
ind ividu als f rom North A fric a and A sia. The r esults show the exist ence o f 3 c lades in Europe, 1 i n Nort h A fric a,
and another in Asia. These six evolutionary lineages are not congruent w ith current subspecies taxonomy. For
example, the English and Iberian birds (which are considered a single taxon by all authors) do not share the
same evolutionary history. Each of these populations had its origin in diff erent glacial refugia and evolved in
allopatry. Alt ernate taxonomic treatments will be suggested that are consistent w ith t he evolutionary history of
these organism s.

198     Chandler & Mulvihill
Age-related variation in wing morphology: Adaptation or constraint? C. RAY CHANDLER* , Dept. Biol., Georgia
Southern Un iv , Statesbo ro, GA ; and ROBERT S. MULVIHILL, Pow derm ill Nat ur e Reserve, Rector , PA.
         Adult songbirds usually have longer wings than young birds. This may be adaptive because short
w ings enhance maneuverability in inexperienced young birds, or shorter w ings in immatures might be the
nonadaptive result of nut ritional constraints. We tested these hypotheses by taking advantage of the molt
pattern in hatching-year Song Sparrows. Some hatching-year Song Sparrow s replace a variable number of
remiges and r ect ric es durin g t he f irs t preb asic molt . This var iat ion raises t w o po ssibil it ies. Young b ird s t hat
repl ace ju venal pr imaries should have few er constrain ts on feather g row th bec ause t hey are reg row ing pri maries
sequent iall y. Thu s, molt ing immat ures may gro w feathers of adul t leng th, imply ing a rol e for n ut rit ion al
constraints. A lternatively, immatures that molt their juvenal primaries may retain the w ing length of young
birds, implying short w ings are advantageous. We tested these alternatives in a sample of Song Sparrows f rom
w estern Pennsylvania. We also carried out a broader comparison of w ing morphology betw een age and molt
classes. Young birds that molted t heir juvenal primaries achieved wing lengths similar to adults, imply ing a role
for const raints. They also had w ing spans and tail lengths similar to adults. However, young Song Sparrows
that molt ed ju venal pr imaries had a n ov el w ing tip p lacem ent (differin g f rom ot her ag e/m olt clas ses) an d ret ained
a wing loading similar to young birds that did not replace primaries. Our results support t he constraint
hypothesis, but f actors affecting w ing length in passerines are probably more complex than implied by the
adaptati on-or-co nstr aint hy pot hesis.

199      Haig, Gratto-Trevor & Mullins
Populat ion genet ic st ruct ure and conser vat ion unit desi gnation in Piping Plover s. SUSA N M . HAIG,
USGS-Forest & Rangeland Ecosyt em Sci. Ctr, Cor valli s, OR; CHERI L. GRAT TO-TREVOR, Canadian Wildlife
Servi ce, Sask ato on, SK; and THOMAS D. MULLINS, USGS-Forest & Rangeland Ecosytem Sci . Ct r.
          Piping Plovers (Charadrius melodus) are endang ered t hro ugh out much of their rang e and ar e con sidered
the top priority species for recovery in the Canadian Shorebird Conservation Plan and U.S. National Shorebird
Plan. The species has been described as having 2 subspecies: eastern (C. m. melodus) and western (C. m.
circumcinctus), w it h no clear ind icat ion of the id ent it y of Great Lakes bird s (t he most severely endan gered
pop ulat ion ). Cont rary t o it s st atus as an end anger ed species, t here h as been no assessm ent of subspec ies
designat ion s nor gen etic s truc ture an d st atus. We addres sed t hese issues via sam pli ng o f the mit ochon dri al
control region (800 bp) from 11 breeding populations across the species range. Results suggests that there are
significant differences among the subspecies and Great Lakes birds appear to be members of the w estern
subspecies. The critical status of t hese Great Lakes birds may be somewhat alleviated by translocation of
ind ividu als f rom ot her p opu lat ion s. But unt il t hese analyses, it w as nev er clear i f it w as saf e to t rans loc ate
from eastern or w estern populations. Subspecific markers also facilitate identification of eastern and western
birds in t he w int er. Given our lack of know ledge regarding w int er locati ons of most Piping Plovers, t his is a
significant breakthrough in monitoring the status of specific breeding populations throughout the annual cycle.
Finally, t here was not significant diff erentiation among populations w ithin subspecies suggesting gene flow is
adequ ate at this time.
200      Dial
On the origin and ontogeny of avian flight: Wing-assisted incline running. KENNETH P. DIAL, Div. Biol. Sci.,
Univ. M ontana, Missoula, M T.
          Recent studies on locomotor behavior of post-hatchling precocial ground birds during ontogeny provide
novel insight into possible processes associated w ith t he origin of avian flight. Discussions on the origin of
avian fl ight fall int o tw o philoso phical camps (arbo real and cursorial hy pot heses) that f ail to adequat ely address
log ical inc rement al adap tive st ages necessary to ach ieve full y developed flig ht mechani cs. I pr opo se a new
model that off ers a solution to t he impasse of previous scenarios on the origin of avian flight . Daily progress of
locomotor performance (e.g., vertical and horizontal accelerations, maximum dist ances during flapping flight,
and terrestrial incline running) and morphometrics of w ing development (e.g., w ing loading, feather growt h) of
3 species (Chukar Partridge, n = 10; Japanese Quail, n = 10; and Ring-necked Pheasant, n = 2) from
hatching to adult stage w ere obtained using high-speed video (60 - 2 50 Hz) and Doppler radar. Throughout
ontogeny, partially for med wings develop significant aerodynamic forces ensuring traction for t he aerobically
pow ered hind limb s, enabling bi rds t o eventu ally " run v ertic ally" thu s achieving sub stant ial height s against
various surfaces (rocks, cliffs, and tree branches and trunks). This "Wing Assisted Vertical Running" (WAVR)
hypothesis appears consistent w ith f ossil evidence and provides incremental adaptive plateaus, as revealed by
ontogenetic trajectories, necessary to achieve fully developed avian flight mechanics as observed in living
spec ies. It is f urt her p rop osed that prec ocial d evelopm ent is phy log enet ical ly pri mitive for b ird s and that
elevated nest building and parental care (feeding, protection) are derived features requisite for improved flight
styles of avian species w ith altricial development.

201      Loos
* Clutch size limitation in precocial birds: Beyond energetics. ELIZABETH R. LOOS, Dept. Biology, Univ.
Loui siana, Laf ayet te, LA .
          A cent ral q uest ion of evolut ion ary theor y inv olv es t he li mits t o rep rod uctive out put . In pr ecoc ial speci es
events during the egg-laying stage appear to limit reproductive output , but t he constraint on clutch size remains
unr esolved. In w aterf ow l, effect ive inc ubat ion dur ing the lay ing st age result s in a 2 - 3 d develop mental sk ew
at the end of the laying period among embryos wit hin a clutch. How ever, this skew in development is largely
overcom e during t he incubat ion period, so hatc h of indiv iduals w ith in a nest occ urs syn chrono usly in only 6 - 8
hr. Hatching synchrony is crucial for precocial birds because young leave the nest in 24 hr and late hatching
indiv iduals may be abandoned at t he nest. Giv en the high cost of asy nchron ous hatc h fo r w aterf ow l, I pr opose
and t est the no vel hypot hesis t hat clu tch size m ay be lim it ed by som e max imum develop mental sk ew that
cann ot be ov ercome to ach ieve synchro nou s hat ch. I test ed t his hypot hesis by inc reasing d evelopm ent al sk ew
at the end of the laying period in wild nests of Blue-winged Teal by replacing 2 mid-sequence eggs wit h 2
fresh ly laid eggs eit her 2 d or 4 d after t he f emale l aid h er last egg. A Chi-square t est show ed t hat manipu lat ed
eggs in b ot h 2 -d and 4 -d t reat ment nest s had signif ican tly low er hat chabil it y (58 and 1 4% , r espec tively) t han
manipulated eggs in control nests (86%, X 2 = 47. 04, df = 2, P < 0.0 01). This is consistent wit h the
pred ict ion of the hat ching synchro ny hypot hesis and suggest s t hat the co nst rain t of synchro nou s hat ch may
impose an up per l imit on c lut ch size in w aterf ow l. I am c urr ent ly exam ini ng t he pr edic tion that duc kli ngs that
do hatch from delayed eggs have diminished survival.

202      Brook
* Modeling the dynamics of a boreal forest Lesser Scaup population. RODNEY W. BROOK, Dept. Biol. Univ.
Saskatchew an, Saskatoon, SK.
        The relationship between vital rates and 8 (asymptotic grow th rate) is important f or understanding the
dynamics of avian populations. I parameterized a Leslie matrix model and used analytic and simulation
techniques to compare the relative importance of vital rates to 8 for a boreal forest population of Lesser Scaup.
8 calculated from t he matrix indicates the population is sharply declining (8 = 0.7 9). Further, perturbation
analysis w as conducted to determine the sensitivit y and elasticity of 8 to matrix sub-elements (vital rates). High
sensitivity and elasticity were associated with breeding season survival and non-breeding season survival.
Also, n est ing suc cess had h igh sensit ivit y. Simulat ion s indic ated variat ion in b reedi ng season sur vival explain ed
45% of t he variation in 8 , w hereas var iat ion in n on-breed ing season surv ival and nest ing suc cess explain ed
35% and 11% , respectively. The importance of breeding season survival and nesting success suggests female
mortality during the breeding season and nest f ailure are important factors for 8 . Pred ation of breed ing females
and eggs may be important factors limiting boreal scaup populations. Similarly, female non-breeding season
mortality may also limit boreal scaup populations. Data gaps are recognized and furt her research is
recommended to improve estimates of vit al rates, covariation among vital rates and determine how density
dependence functions in the population. Further, independent tests of scaup models and their assumptions are

203     Devlin
* Pieces of a population: Arctic Terns on Machias Seal Island, Bay of Fundy. CATHERINE M. DEVLIN, Dept.
Biol. , Univ. New Brunsw ick, Frederict on, NB.
         Mac hias Seal Island i s loc ated at t he mout h of th e Bay of Fundy sit uat ed on t he bor der bet w een U.S.
and Canada. It is the largest known A rctic Tern colony in the region and is designated as a Migratory Bird
Sanctuary. Before understanding the dynamics of a regional population, it is necessary to closely the
characteristics of local populations. Betw een 199 5 - 19 98 ov er 100 nesting adult t erns were banded on the
island and since the initiation of a regional study of Arct ic Terns involving the use of f ield readable bands in
199 9, ov er 500 adult t erns and 700 chicks have been banded. In the long term w e will establish a population
of know n age banded birds in the region, however, in the short t erm, information can be gathered about the
sur vival rat es and movem ent pat terns of adul t terns . In t he 2 00 1 f ield season w e w ill obt ain o ur f irs t est imate
of adult survival based on birds wearing both standard BBL bands and field readable bands. Additionally we
w ill have collected tw o estimates of inter-colony movement to t his island. I w ill discuss the implications of
these data as well as records of historical banding. This island w as never completely abandoned during the
declines of seabird populations on the eastern seaboard of North America in the late 19th and early 20th
centuries. It is thought this island may have served as one of a few refuge populations for t he Gulf of Maine
and the Bay of Fundy. Recent estimates of movement patterns in combination w ith observations of
productivit y indicate that t his colony may currently be a "sink" in the regional population.

204     Sandercock, Martin & Hannon
Life history variation in extreme environments: The comparative demography of arctic and alpine ptarmigan.
BRETT K. SANDERCOCK,* Div. Biol. , Kansas State Univ., Manhatt an, KS, KATHY MARTIN, Dept. Forest Sci.,
Uni v. British Columbia, Vancouver, BC, and SUSAN J. HANNON, Dept. Biol. Sci., Univ., A lberta, Edmonton,
         The stringent ecological conditions of arctic and alpine habitats are expected to have important
consequences for the evolution of lif e history traits in vertebrates. We compared the demography of 3
populations of ptarmigan breeding along a gradient of environmental conditions. Female Willow Ptarmigan and
White-tailed Ptarmigan breeding at alpine and subalpine sites had smaller clutches, and lower probabilities of
nesting success, fledging success, and renesting than Willow Ptarmigan at a low elevation arctic site.
Reprod uctive out put and ad ult mort alit y r ates w ere ranked : alpin e < subalpi ne < arctic, w it h li ttle ov erlap
among sites. Age structure of v ital rates also differed among sites, wit h no significant age differences at the
arctic site, age differences in components of reproduction among subalpine birds, and age-specific variation in
reproduction and adult survival at t he alpine site. Matrix models predicted stable population numbers for both
Willow Ptarmigan popu lation s (8 = 1.0 ) and declines in White-tailed Ptarmigan (8 = 0. 7). J uv enil e sur vival had
the highest elasticity at all sites, and rates of population change were most sensitive to variation in this rate.
However, f ecundity rates of 1-year breeders had high elasticity in Willow Ptarmigan, w hereas survival rates of
older females w ere more important in Whit e-tailed Ptarmigan. These diff erences suggest t hat the effects of
global warming, harvesting and other perturbations will dif fer among arctic and alpine vertebrates, possibly in
predict able w ays.
205      Beissinger & Sandercock
Sex-biased and fitness based dispersal complicate source-sink dynamics in a parrot metapopulation. STEVEN
R. BEISSINGER, Ecosys. Sci. Div., Univ. California, Berkeley, CA and BRETT K. SANDERCOCK, Div. Biol.,
Kansas St ate Un iv. , M anhat tan , KS.
         A dominant paradigm to describe spatially structured populations is the "source-sink" concept. Source
populations are self-sustaining and net exporters of individuals, whereas sink populations have negative
population grow th rates that require immigration to sustain them. Source-sink dynamics are often inferred but
rarely demonstrat ed. We presen t resu lt s f rom 10 year s of demogr aphi c st udi es of Green-rum ped Parr ot let s
(Forpus passerinus) in Venezuela that demonstrate large differences in dynamics of 2 populations separated by
only 0. 5 km. Upland and lowland sites differ in elevation by 1-2 m and soil type, but not in rainfall or
temperature. Despite their proximity , these sites function as distinct breeding populations connected by
dispers ing juv enil es. Upland f emales pro duc ed 1 more y oun g/ yr than l ow land bir ds becaus e they had h igh er
nesting success and laid more clutches. Multi-st ate mark-recapture models indicated that local survival of
breeding and nonbreeding adults did not diff er betw een sites. However, juvenile females emigrated at much
higher rates than males, and juvenile dispersal w as highly biased tow ard the upland site. Source-sink status
dif fered am ong sit es and by sex w hen r ates w ere combin ed in to a matrix model: females decl ined in b ot h sites
w ithout immigration and were rescued by immigrants each year, w hereas upland males increased and lowland
males decl ined w it h or w it hou t immigrat ion . Sen sit ivit y analyses (l if e table r esponse ex peri ment) indic ated t hat
dif ferenc e in j uv enil e sur vival caus ed t he up land sit e to be a sour ce f or m ales and a sink for f emales , w hereas
dif ference in sit e fidelit y caused t he upland sit e to be a source and t he low land sit e a sink for males. Thus,
birds move from the sink to the source instead of vice versa, as predicted by theory.

206      Jodice, Roby & Turco
Is kittiw ake reproduction constr ained by food quality in the northern Gulf of A laska? Revisiting the Junk Food
Hypothesis. PATRICK G. R. JODICE* , DANIEL D. ROBY, USGS - Oregon Coop. Fish & Wildl. Res. Unit,
Oregon St ate Uni v. , Cor valli s, OR, and KATHY R. TURCO, Inst. Marine Sci., Univ. A laska, Fairbanks, A K.
           We tested the " junk food hypot hesis", w hich states that diet quality (i.e., lipid content) limits
pro duc tivit y of apex marine p redat ors in t he n. Gulf of Alask a. W e inv est igat ed rel ation ships betw een di et
composition , f ood deli ver y r ates, and en ergy pro visionin g rat es t o Blac k-l egged Kit tiw ake br ood s at 6 c olo nies
during the 1990 s. Differences in lipid content and energy density among and w ithin forage fishes were
suf fici ent to po tent iall y limit energ y av ailab ili ty t o gr ow ing nest lin gs, thereb y limit ing fledg ing suc cess. Three
high-lipid forage fishes (sand lance, Pacific herring, and capelin), how ever, accounted for > 82% of biomass in
kitt iw ake chick diets. The proportion of low -lipid fish in kitt iw ake diets (e.g., pollock) averaged < 1% among
all colonies and years. Nevertheless, nearly 50 % of the variation in kittiw ake productivity w as explained by
variation in energy provisioning rates to nestlings. Diff erences in energy provisioning rates to kitt iw ake broods
were, however, influenced more by meal delivery rate and meal size than energy density of the diet.
Productivity of kitt iw akes at colonies in the n. Gulf of Alaska was primarily dependent upon availability of the
three af orem ent ion ed hi gh-lip id f orag e fishes. Sw it ching to alt ernat ive low -lipid prey (e.g. , p oll ock) w hen
hig h-lipi d f orag e fish w ere sc arce does not appear to be ad apt ive. For kit tiw akes in t he n. Gulf of Alask a,
factors aff ecting prey availability appear to be more important t han differences in energy density among
high-lipid forage fish.

207      Sillett , Holmes & Rodenhouse
Can intraspecific agonism regulate population size of a territ orial bird during the breeding season? T. SCOTT
SILLETT*, RICHARD T. HOLMES, Dartmo uth Coll., Hanover, NH; and NICHOLAS L. RODENHOUSE, Wellesley
Coll., Wellesley, MA.
        The mechanisms that control, or regulate, the size of bird populations are poorly understood. One
hypothesis is that agonistic interactions betw een territorial conspecifics increase wit h population density. This
crowding mechanism would be regulatory if it lowers reproductive success and thus creates a negative
feedbac k on pop ulat ion gro w th. To t est if a crow din g mechan ism cou ld r egul ate loc al abu ndan ce of a territ ori al
bird, w e experimentally manipulat ed local densit y of Black-thro ated Blue Warblers durin g th e 199 7 - 1 99 9
breeding seasons. We compared demographic and behavioral traits between 2 density treatments: a
high-density control, and a second treatment where warbler density w as reduced. Fledging success, territory
size, and proportion of time males spent foraging w ere all greater in the reduced density treatment. How ever,
parental feeding rates of nestlings, fledgling mass, and male-male interactions, represented by countersinging
rates, did not clearly differ betw een density treatments. These results indicate that some agonistic int eractions
bet w een in div idu als are densit y-depen dent , an d suppo rt the hy pot hesis t hat crow din g is one m echan ism that
can regulate populations of territorial animals.

208      Rohwer & Richkus
Forty years of recruit ment data for millions of duc ks, but w hat accuracy? FRANK C. ROHWER and KENNETH
D. RICHKUS, For. , W ild l. , & Fish. , Lo uisiana St ate Univ. , Bat on Rou ge, LA.
         Since 19 61 the USFWS has annually collect ed about 1 20 ,0 00 w ings f rom hun ter-ki lled ducks t o
estimate age ratios. These age ratios have been used for decades as our best measure of annual recruitment
and have recently been incorporated in Adaptive Harvest Management models. How ever, the accuracy of t he
techniques used to categorize age and sex of the wings has never been tested during the operational survey.
We collected wings from hunt er-killed ducks w here we could positively age each duck using multiple
characteristics like tail molt, presence or absence of a bursa, or condition of the oviduct, ovary, or penis. We
then sen t these kn ow n-age w ings (n = 63 74 ) t o t he USFWS in a blind test of their sys tem t o cat egor ize
species, age, and sex. Accuracy varied betw een species; 39 .4% of Green-w inged Teal were incorrectly
categori zed, w hile only 1. 3% of A merican Wig eon w ere incorrect ly assigned t o age/sex grou p. For M allards,
20 .8 % of our w ing s w ere in correct ly cat egor ized t o age o r sex cat egor ies. Ther e w ere no errors in speci es
identification. Error rates were generally highest for juvenile males and adult females. Unfortunately, t he errors
are not rando m; ju veniles are overestim ated and adult s underestim ated result ing in in flat ed age ratios. W orse
yet, t he error rates appear to vary from one year to the next, so it may not be possible to use a standard
correction factor f or all years. By collecting both w ings from many ducks w e discovered that errors are not
rand om, b ut harv est dat e and t rait s of ind ividu al du cks increase t heir pro babi lit y of bein g in correct ly cat egor ized
to age group.

209      Rodewald
Floristics and avian community structure: Implications for regional changes in eastern forest composition.
AMANDA D. RODEWALD, School Nat . Res., Ohio St ate Univ. , Columb us, OH.
          Because many forest stands are increasingly dominated by maple in eastern North America, there will
likely be a relative scarcity of mature oak forests wit hin the next several decades. In an eff ort t o understand
how avian community struct ure may be affected by a regional change in forest composition f rom oak to maple
dominance, I compared bird communities in both oak- and maple-dominated forest stands in three seasons
(w inter, spring, and fall). In 199 8 and 199 9, birds w ere surveyed and habitat characteristics w ere measured in
six mat ure f ores t w ood lot s in cent ral Penn sylvania. Tot al abu ndan ce of bir ds w as gr eater w it hin oak- than
maple-dominated stands in all seasons, whereas species richness was greater in oak stands only in the spring.
Resident spec ies, lon g-dist ance m igr ant s, w ood peck ers, and b ark-gleaners w ere more ab und ant w it hin oak t han
maple stands. Eleven individual species also were more abundant w ithin oak than maple stands in at least one
season. Diff erences in avian community struct ure between oak and maple stands were greatest in t he fall
w hen o aks p rod uced har d mast and i n t he sprin g w hen i nsec tivoro us bird s gleaned ar thro pod s f rom tree
substrates. This pattern is consistent w ith expectations based on mast characteristics and tree physiognomy of
maples and oak s. These resu lt s suggest that a regi onal chan ge f rom oak- to maple-dom inat ed f ores ts may
st ron gly affect avian co mmuni ty struc ture an d po ssibly pop ulat ion s of som e com mon bi rd s peci es asso ciat ed
w ith eastern deciduou s for ests.
210     Snell-Rood & Cristol
Avian communities of created forested wetlands are not the same as those of natural forested wetlands.
EMILIE C. SNELL-ROOD* and DANIEL A. CRISTOL, Dept. Biol., Coll. William & Mary, Williamsburg, VA.
         The avian communities of 6 created forested w etlands were compared to natural w etlands to determine
w het her c reat ed w etland s pr ov ide r eplac ement habi tat for t he avian c ommunit ies lost w hen n atural fores ted
w etland s are dest roy ed. Creat ed w etland s w ere similar t o nat ural w etland s of the sam e age in tot al avian
abundance, abundance of most guilds, and overall wetland dependency of the bird species present. How ever,
created w etland s had signif ican tly low er av ian d iversity, due t o an u nder -rep resen tation of passerin es. Creat ed
w etland s lac ked ex pect ed nu mbers of neot rop ical migran ts and spec ies f eedin g at hig h t rop hic levels, as w ell as
passerine species dependent on w etlands or wit h high habitat specificity. Avian communit y succession on
created wetlands appeared to be retarded as 1) comparisons to natural wetland communities ranging from 1 -
25 y r old indicated that the 7 - 11 yr old created wetlands w ere most similar to t he youngest age group of
nat ural w etland s (less t han 6 yr old ) and 2) a trajec tory of expect ed avian c ommunit y developm ent show ed t hat
created w etland s lagged b ehin d nat ural w etland s by at least 15 yr . W hereas regen erat ing nat ural w etland s can
replace the avian communities of mature w etlands aft er approximately 25 yr, created w etlands will apparently
take much lo nger . Because t he avian c ommunit ies of created w etland s w ere no t com parab le t o t hose of nat ural
w etlands in the same developmental stage, the use of created forested wetlands to replace destroyed mature
w etland s can not cur rent ly be co nsidered a con serv ation st rat egy for p rev ent ing a net loss of w etland habi tat
for birds.

211      Huettmann, Cam, Bradley, McFarlane-Tranquilla, Lougheed, Lougheed & Cooke
Nesting h abitat selection o f M arbled Mur relets in Desolat ion Sound, British Colu mbia FALK HUETTMA NN* ,
LOUGHEED and FRED COOKE, Cent re f or W ildlif e Ecolog y, Simo n Fraser Univ ., Burnaby BC.
          Quantitative landscape descriptions of nesting habitat f or Marbled Murrelets (Brachyramphus
marmorat us), a species of conservat ion con cern, are poor ly kn ow n. For t his st udy 8 4 M arbled Mur relet nest s
w ere used, w hich w ere located by radio-telemetry done from a helicopter in Desolation Sound, British Columbia
Canada. These nest s w ere geo -ref erenc ed, and f or t he st udy area ov erlai d in a Geogr aphi c In form ation System
(ArcView). For overlays we used a Digital Elevation Model (slope, elevation and aspect) and compiled 1:20,000
scale Forest Cover maps. M ult ivariat e generalized linear models w ere built u sing M arbled Mur relet nest s
stratif ied by forest cover (Old Forest > 140 yr old). The habitat characteristics corresponding to nest locations
w ere compared to t hose of an equ al nu mber of rand omly draw n lo cat ion s f rom the sam e st rat a. This w as
repeated over 1 00 0 t imes to obt ain robust results f or t he heterogeneous landscape of the st udy area. For t hese
model runs we applied an automatized model selection using AIC (Akaike's Information Criterion) to identif y the
best models, their predictors and coeff icients of predictors. Results and implications from t his approach are
pres ent ed and discussed regar din g t he nesting habi tat of Marbl ed M urr elet s in Brit ish Colu mbia.

212      Reinking, W iedenfeld, W olfe, Sherrod & Jenkins
Vegetation structur e, avian relative abundance, and avian nest success responses to tallgrass prairie
SHERROD, and M. ALAN JENKINS, Sutt on Av ian Research Center, Bartlesvill e, OK.
          Tallgras s pr airi e is a dyn amic , d ist urb ance- depen dent habi tat in w hic h veget ation st ruc ture m ay dif fer
dramatic ally f rom one y ear to t he next. Drought , f ire, and grazing are t he predominat e forc es shaping tallgr ass
prairie veget ation, but even their absence result s in veget ativ e succession ov er tim e. These changes to
vegetation structure, w hether through disturbance or through lack of disturbance, may affect bird populations
over time in measurable ways. We monitored vegetation height and density indices, along with avian relative
abundance and nest success, on 18 study plots of 1 6.2 ha each in size from 1992 - 19 96 in Oklahoma
tallg rass prai rie. Vegetation densit y and maximum heigh t inc reased each y ear f or t he f irs t 4 y r af ter a f ire, after
w hich time litt le if any increases occurred. The number of Grasshopper Sparrows declined each year after the
first year post-burn, Dickcissels increased slightly, and Eastern Meadowlarks showed little change in number.
The trend in the number of nests found on plot s roughly paralleled the point count relative abundance trends for
Grasshopper Sparrow and Dickcissel, but not f or Eastern Meadow lark. Several specific vegetation
charact eristic s w ere associated w ith nest suc cess of Grasshopp er Sparrow s and Dickcissels, but overall nest
suc cess on study plo ts w as not relat ed t o pl ot treat ment, sugg est ing that suitable n est ing microhab it at w as
available for t hese species even wit hin burn ed and grazed plots.

213     Ralph, Hogoboom & Frey
Landscape-level landbird population responses to riparian habitats and alterations by f ire in northern California
and southern Oregon. C. JOHN RALPH* , WILLIAM C. HOGOBOOM, and ROBERT I. FREY. Redw ood Sci.
Lab., U. S. For est Ser., A rcat a, CA .
          More than 10 ,00 0 point count s over the Klamath and Siskiyou Mount ains, including the entire drainage
of t he Klamath River, w ere included in the study. Our landscape-level analyses determined the effects of
riparian habitats, as well as historical fires on bird populations. We found in several species a continuum of
habi tat associat ion pat terns . So me f ew w ere ripar ian (or u plan d) speci alists, and m any that w ere hab it at
generalists. The eff ects of f ires, plotted over the past 100 yr, helped determine the effect of chronic fires on
habitat ty pes, as w ell as on the bir d species associated w ith the result ing veget ation c onf igurat ions. No
species, or assemblage of species, appeared to be markedly affected by f ire. We have explored the interactions
of site, v egetation, and various other landscape att ributes. We had expected profound influences of fire upon
either t he habitat s or t he bird com munit ies, but fou nd only relativ ely modest eff ects.

214     Rodewald & Brittingham
Stopover habitat use by songbirds during spring: The roles of habitat st ructure, leaf development, and food
availability. PAUL G. RODEWALD* and MARGARET C. BRITTINGHAM, Ecol. Prog. and School For. Res.,
Pennsylv ania St at e Un iv ., Univ ersity Park, PA.
          Despi te imp ort ance o f songbi rd s topo ver habi tat for a s uccessful m igr ation , spat ial an d t emporal
patterns of inland habitat use and potential mechanisms are not w ell understood. In A pr - Jun 199 7 - 19 99 in
cent ral Penn sylvania, w e sur vey ed bi rds w it hin 5 f ores t habi tats: fores t int erio r, fores t-agr icu lt ural edge,
suburban forest, pole-stage forest, and shrub/sapling-stage forest. W e used abundance, species richness, and
food availabi lit y (assessed by forag ing behav ior ) t o id ent if y import ant migran t habi tats and measu red l eaf
develop ment and m icr ohab it at to assess import ant habi tat char act eristics . Pat terns of habi tat use v aried
stro ngly b y year. Mat ure edge-dominat ed suburban f orest and f orest-agri cult ural edge habitat s (and for est
int erior t o a lesser extent ) had higher abundance of transient species and guilds and hig her tr ansient ric hness
than shrub/sapling-stage and pole-stage forests. Locally breeding migrants w ere more abundant in
edge-dominated habitats t han in breeding periods, suggesting that transients of t hese species also primarily
used edge h abit ats durin g passage. Higher p rey attack rat e and g lean r ate in f ores t-agr icu lt ural edge sugg est ed
that food availabi lit y and qu alit y w as higher in some edge-dom inat ed f ores ts. Tran sient abun danc e w as
positively associated w ith numbers of large trees (> 38 cm dbh) and percent shrub cover, both characteristics
of mature edge-dominated forests. Early season understory/ midstory leaf development was most advanced in
suburban forests and was positively associated w ith early season transient abundance in all 3 yr. Findings
suggest that habitat diversity is import ant for migrant songbirds and > 2 yr st udies are needed to account for
temporal var iat ion in stopo ver habi tat use. Where app rop riat e, improv ing or m aint aini ng ex ist ing edge h abit ats
may benefi t m igrant songbird s.

215     Plummer & Conroy
Habitat-association models of three Neotropical migrants in a fire-managed southern piedmont landscape. W.
TODD PLUMMER* and MICHAEL J. CONROY, U.S. Fish & Wildl. Ser. Co-op Unit, Warnell School For. Res.,
Uni v. Georgia, A thens, GA.
        We model ed t he rel ation ship bet w een hab it at var iabl es and abun danc e of 3 n eot rop ical migran t spec ies
on a national w ildlife refuge. The refuge in the southern Piedmont of central Georgia is managed largely for the
federally listed Red-cockaded Woodpecker. The management regime relies heavily on prescribed burning and
thinning over portions of the refuge to provide areas of mature pine savannah habitat required by the
Red-co ckaded W ood peck er. The h abit at-asso ciat ion models w ill be used t o explo re t he rel ation ship bet w een
prescribed burning and thinning treatments and the abundance of selected bird species in various habitat types.
We used line transect surveys to estimat e abundance of Hooded Warblers, Prairie Warblers, and Wood
Thrushes in 12 management compartments wit hin the refuge during breeding seasons from 1998 t o 2000. We
used point count data from sites separate from t he transect sit es in 2000 to validate our models and refine
parameter values. Results of our preliminary analysis indicate that prescribed burns result in no effect to slight
increases in population density in compartments aft er a burn for Wood Thrushes and Prairie Warblers, and no
eff ect f or Hooded W arblers.

216     Driscoll & Donovan
Landscape an d edg e effect s on the di st rib ut ion , ab und ance, and n est ing suc cess of Wood T hru shes in c ent ral
New York. MELANIE J. L. DRISCOLL* , Dept. Env. & For. Biol., SUNY Coll. Env. Sci. & For. , Syr acuse, NY;
and THERESE M. DONOVAN, Vermont Coop. Fish & Wildl. Res. Unit, USGS, Burlington, VT.
         Habitat fragmentation changes landscape composition and creates edges betw een forest and
hum an-dominat ed lan dsc ape t yp es. Forest -interio r sp ecies may avoid f ores t fragm ent s and edges, o r may
experien ce reduc ed rep rod uctive suc cess in t hese altered h abit ats. Some resear chers h ave hyp ot hesized t hat
landscape composition moderates the strength of edge effects. I documented distribution, abundance, and
nesting success for Wood Thrushes in edge and core habitats in fragmented and contiguous landscapes in
cent ral New Yor k and test ed t he hy pot hesis t hat land scap e com posit ion moderat es edge effect s. Neit her
landscape composition nor proximity to edge aff ected Wood Thrush distribution or abundance. How ever,
average number of young fledged/nest was lower in fragmented (0.80) than contiguous (1.55) landscapes.
Daily survival estimates were higher in core (0.971) than in edge habitats, in fragmented landscapes only. My
results indicated that the relationship betw een survival parameters, landscape composition, and edge eff ects are
complex and not easily interpreted.

217     Steeves, Kim, M cNally, Anderson & Friesen
How import ant are physical barriers to gene flow in the diversific ation of t ropical seabird species? TAM MY
STEEVES* , MICHAEL KIM, HEATHER McNALLY, Dept. Biol ., Queen' s Univ ., Kings to n, ON; DA VID
ANDERSON, Dept. Biol., Wake Forest Univ. , Wi nsto n-Salem, NC; and VICKI FRIESEN, Dept. Biol., Queen' s
         Genetic studies of arctic and temperate seabird species indicate that many populations diverged in
allopatric glacial refugia during the Pleistocene. Some populations, how ever, appear to have diverged in the
absence of any obv iou s physical barr iers to gen e flow . The rel ative imp ort ance o f phy sic al and non -physical
barriers in t ropical seabird species is po orly u nderst ood. In th is stu dy, w e examined mit ochond rial DNA
(cy tochro me b) sequence variation in Brow n and Red-footed booby samples collected from islands in the
At lantic and the eastern and central Pacific Oceans. Sequences from t he Atlantic w ere distinct f rom those in
the Pacific for both species. These results are concordant w ith a recent study conducted on masked boobies.
It appears that t he Isthmus of Panama is likely an eff ective physical barrier to (female-mediated) gene flow in
these 3 booby species. The situation w ithin t he Pacific is less clear. We found that Red-footed Boobies in the
east ern an d cen tral Pacif ic s hared the sam e sequence. Thi s pat tern i s co ncord ant w it h t hat observ ed in Masked
Boobies and suggests t hat gene f low is not rest ric ted bet w een t he east ern an d cen tral Pacif ic i n t hese 2
spec ies. Brow n Booby sequences from the east ern Pacif ic, how ever , w ere di st inc t from those in t he cen tral
Pacific. Genetic analyses using nuclear markers are required to determine whether gene flow in this species is
restricted by physical or non-physical barriers; these studies are underway.
218     Anderson
Compon ent s of repr odu ct ive imp airm ent in a c ont amin ated West ern/ Clark ’ s Grebe pop ulat ion from cent ral
California. DANIEL W. ANDERSON, Dept . W ildl. Fish & Cons. Biol. , Un iv. Calif orn ia, Davi s, CA.
         The Western Grebe complex (Aechmophorus sp.) at Clear Lake, CA, made ecotoxicological history
w hen Hunt & Bisch off (19 60 , California Fish & Game 46: 91-1 06) f irst reported, anyw here, "bioaccumulation"
by lipophilic contaminants, in this case, the insecticide p,p'DDD. And, DDD soon aft er was also determined to
be the proximate cause of severely depressed populations at Clear Lake; and Rachel Carson and Robert L. Rudd
described the Clear Lake situation in their well-known books. Now, some 50 yr later, grebe populations have
improv ed, but stil l suf fer repro duct ive impairm ents f rom a suit e of pro ximat e causes. Here, I evaluate t he most
impor tant mult ipl e-st ressors (t he or gano chlor ines DDD and DDE, met hy l mercury , h abit at chan ge, and h uman
disturbance), and I estimate the relative roles of each in depressing current grebe reproduction. Since DDD is
not know n to be converted to DDE in natural systems, the source of DDE has to be from the past use of the
insecticide, DDT. Clear Lake grebes still suffer chronic DDE-induced eggshell thinning, but DDD levels, although
still present as a distinct " Clear Lake fingerprint", are probably no longer important toxicologically and have
declined by several orders of magnitude in all ecosystem components measured. Methyl mercury, now know n
to h ave been present in Clear Lake grebes at t he tim e of t he DDD studies, is sti ll present at levels close t o th ose
know n to have negative eff ects on developing embryos and adult liver function. Yet, human disturbances in
the last 10 yr hav e been an over-riding f actor depressing reprodu ct ion. Yet , t he suite of cont aminants h as also
resulted in a chronically low eff ect on reproduction compared to other sites, so that w ith all factors combined,
Clear Lake still has to be considered a population " sink" relative to com parison "source" populations at Eagle
Lake and Tulelake.

219     Sharbaugh & Hahn
Use of radio-t ransmit ters t o locat e w int er roost sites of Black-capped Chickadees in interio r Al aska. SUSAN M .
SHARBAUGH* and THOMAS P. HAHN, Inst. Arctic Biol., Univ. Alaska Fairbanks, Fairbanks, AK.
         Small bi rds overw int erin g at hig h lat it udes enco unt er a nu mber of enviro nment al ch allen ges. Limi ted
daylight severely rest rict s for aging oppor tun iti es at a time w hen chroni c subf reezing temperat ures increase
energy demand. Black-capped Chickadees in interior Alaska meet these challenges through a variety of
phy siolog ical and b ehavior al adj ust ments. These inc lud e noc turn al hy pot herm ia, rapi d dai ly fattenin g, inc reased
insulation, and food caching. An important factor in maintaining energy balance is to minimize energy loss.
Alask an ch ick adees are in r oost for 1 6 - 17 hr o f a w int er day . Theref ore, the qu alit y of w int er roost sit es
w ould great ly inf luence daily energy expendit ure. Small (0 .5 g) radio-t ransmit ters (Holo Hil Syst ems Ltd, LB-2)
w ere us ed t o f oll ow chi ckadees t o ro ost sit es durin g t he w int er of 20 00 - 200 1. Thi s pilot st udy det ermined
length of daily activit y, roost site location and roost site fidelity. Birds roosted singly in birch cavities. Carrying
radio-transmitters did not appear to inf luence the chickadees' energy balance or mobility. Birds with
transmitters maintained body mass and fat depots even during periods of low ambient temperatures (-30°C).
This study demonstrates the practicality of using t elemetry t o explore overwintering behavior and energetics of
these t iny passerines even under ex treme envi ronment al condit ions.

220     Zharikov
* Eastern Curlews do not improve their gastric function prior to migration. YURI ZHARIKOV. Dept. Zool., Univ.
Queensland, Brisbane, Queensland, Aust rali a.
         Eastern Cur lew s w int erin g in Aust rali a feed on low org anic con tent prey and h ave t o reg urg it ate pell ets
cont aining indi gestibl e parts regularly . Hyp oth eses that eastern c urlew s improv e (1) gastr ic digest ion and/o r (2)
the rat e of gast ric trans por t of food mat ter pr ior to migr ation have been i nv est igat ed in this st udy . It w as
shown that neither extraction of organic matter or energy in the stomach, nor the rate of transport improve
dur ing the mont h pr ecedi ng t heir nor thw ard d epart ure. The r ate of trans por t, h ow ever , i ncreased w it h t he rat e
of int ake. It w as concluded that w ith t he food supply apparently superabundant t he rate of energy intake in the
birds is limited by the speed at w hich food is michanically processed in the stomach (gizzard) and energy
content of the prey.
221      Seavy
* Physiological correlates of habitat association in East African sunbirds. NATHANIEL E. SEAVY, Dept. Zool.,
Univ. Florida, Gainesville, FL.
          Evidence of physiological adaptions by birds and mammals to specific climatic conditions is widespread.
Such adap tion s may also be associat ed w it h hab it at selection , especi ally if adjac ent habi tats are c harac terized
by dras tical ly dif ferent phy sic al co ndi tion s. I investigat ed t he ph ysiol ogi cal ec olo gy of a sunbir d (Nec tarin iid ae)
communi ty in a t rop ical East A fric an f ores t/g rassland land scap e. This com munit y c onsist s of 17 spec ies
belo ngi ng t o 2 gener a (Ant rhep etes and Nectarinia); some are associated w ith forest habitats, w hereas others
are found primarily in open grassland habitats. Open habitats are characterized by more extreme daily
temperat ure an d hu midit y f luc tuat ion s t han f ores t habi tats. Usin g f low thro ugh resp iro met ry , I measu red b asal
rat es of met abol ism of 10 sunbir d species. BRM w as influen ced b y body mass, b ut there w as no clear
dif ference bet w een forest and open habit at species. For 3 open habit at species and 4 f orest speci es I
measured metabolism and body temperature as a function of ambient t emperature from 1 0 t o 35º C. During
these experiments most sunbirds maintained a constant body temperature at low ambient temperature. Body
temperat ure w as not inf luen ced b y body mass or h abit at associat ion . The in crease in met abol ism in r esponse t o
decr easing ambien t temperat ure w as measured by therm al co ndu ct ance. Ther mal c ond uctance w as asso ciat ed
w ith body m ass and habitat association ; open habit at species had low er therm al conduct ance than f orest
species. This difference suggests that phy siological characteristics related to t he thermal environment may be
associated with habitat selection by birds in heterogeneous landscapes.

222      Netter
* Loss of genetic variation in Greater Prairie Chickens. M. RENEE NETTER, Dept. Biol., Univ. Wisconsin
Milw aukee, Milw aukee, WI.
          Population genetic theory predicts the loss of alleles and reduction in heterozygosity during a population
bott leneck, how ever, this has rarely been shown in natural populations. Greater Prairie Chickens in Wisconsin
underwent a population bott leneck during the last century and are currently listed as threatened. In 19 30, their
estimated population size was 55, 000 birds. Over the next 60 y r prairie chickens declined to 2,5 00 birds
concentrated near Stevens Point, W I. I investigated genetic variation in historic (19 51, n = 47) and
present-day (199 6 - 19 99, n = 87) populations using 6 microsatellite loci. Population mean heterozygosity and
number of alleles/locus declined significantly aft er the population bot tleneck. Sixteen alleles w ere detected in
the historic but not in the present-day population and, thus, w ere apparently lost during the population
bottleneck. Mean number of alleles/locus declined from 9.17 (range: 5 - 16 alleles/locus) in the historic to 7.00
(ran ge: 3 - 12 allel es/loc us) in t he cu rrent pop ulat ion . A ver age het erozyg osity ac ros s all loc i w as also redu ced
signif icantl y (f rom 0 .7 71 to 0 .5 57 ).

223      Elchuk
* Food and predation-risk as fact ors inf luencing f oraging locati ons of Nor thern Flickers. CA NDACE L. ELCHUK,
Dept. Biol., Univ. Saskatch ewan. , Saskatoon, SK.
         At tribut es of foraging sites chosen by birds may be related to increased foraging eff iciency, higher food
availability and abundance, and protection f rom predation. Within home ranges of 2 9 radio-tagged adult
Northern Flickers in British Columbia, we att empted to identify preferred foraging habitat during t he nestling
period. We compared habitat characteristics of flicker foraging locations to randomly selected locations in their
home range. Foraging plots were located in grassland and were characterized by a lack of tall vegetation, a
hig h per cent age of bare g rou nd, and a h igh densit y of smal l ant hil ls. On av erage, forag ing plo ts had 5 t imes
more anthills than random plots. Foraging plots were also closer to trees and forest edges than random plots.
The random plo ts had sig nif icantl y mor e medium and t all vegetat ion t han for aging plot s, and lit ter (dead grass)
w as the most abundant t ype of cover. The sparse ground co ver of for aging plot s likely in creases access to
food, w hile foraging in close proximity to t rees decreases avian predation. The selectivity of foraging areas by
flickers suggests that foraging efficiency could be negatively aff ected by land use practices that alter the
structure of ground cover.

224     Hamel & Parrish
Do radio-t ransmit ters inf luence behavior? Behavioral comp arisons of tagged and unt agged Common M urres.
NATHALIE J. HAMEL* and JULIA K. PARRISH, School Aquatic & Fish. Sci., Univ. Washington, Seattle, WA.
          Radio-telemetry is a valuable method for assessing animal habits, and has contribut ed to the
conservation of many species. Alt hough radio-telemetry is w idely used, the eff ect of t ransmitters on the
behav ior of ind ividu als is rar ely assessed. In t his st udy , w e com pare t he beh avior o f nest ing Common M urr es
tagged wit h small VHF radio-transmitters relative to unt agged murres. In 199 9 and 200 0, w e tagged 1 adult in
each of 23 pairs on Tatoosh Island, WA. Tagged murres, their untagged mates, and control pairs, w ere
monitored from blinds. For half hour intervals, we recorded the dominant behavior of each individual every 5
min, as preening, sitting, sleeping, standing. We also recorded instances of brooding duty switches. Finally,
provisioning behavior w as recorded as individual returns wit h or w ithout fish. The distribution of routine
behavior of t agged birds relative to controls was not diff erent. How ever, tagged birds brooded chicks more
oft en than their untagged mates, sw itching f rom brooder to fisher only 28 % of the time. M urres in control
pair s sw it ched du ties 4 8% of the t ime, o n average. Moreo ver , t agged ind ividu als return ed w it h f ew er f ish than
their unt agged mat es (5 9% ver sus 87 % of ret urn s respect ively), suc h t hat pair s w it h a t agged murre ret urn ed
few er fish than control pairs (77% versus 95% of returns, respectively, had fish). How ever, fish size did not
diff er significantly among groups. Fledging success was not significant ly diff erent between tag and control
pairs. Although behavioral eff ects of t agging were demonstrable, they did not t ranslate, in the short-t erm, into
a reproductive cost to the pair.

225     Nott, DeSante & Siegel
The influence of El Niño/Southern Oscillation and forest insect outbreaks on avian productivit y in the Pacific
Northwest. M. PHILIP NOTT*, DAVID F. DeSANTE, and RODNEY B. SIEGEL, Inst. Bird Populations, Point
Reyes St ati on, CA.
         Local reproduct ive success of several landbird species has recent ly been show n to vary annu ally as a
funct ion of global climate indices. Here w e show t hat regional productivit y indices for landbird species in the
Pacific Northw est, measured as the ratio of young to adult birds captured at 36 Monit oring Avian Productivity
and Survivorship (MAPS) stations, varied annually as functions of indices relating to seasonal global climate and
forest defoliation by irrupt ive insects. The seasonal El Niño/Southern Oscillation Precipitation Index (ESPI) for
the mont hs M ar t o M ay, pri or t o and inc lud ing spr ing migrat ion , b est explain ed t he rel ation ship bet w een ann ual
variation in productivit y and the climate cycle. For most species productivity w as higher follow ing El Niño
w inters that also correlated w ith cooler, w etter pre-migration conditions along the entire Pacific slope of
Mexico. This effect w as mo re pr ono unc ed among spec ies t hat overw int er pr imaril y sout h of n. Mexico, t han
among species that overw inter in the more temperate climates of California and the Pacific Northw est. The
Jun-J ul larval bio mass of 2 defoli ating m oth species also correlated po siti vely w ith reproduc tiv e success,
espec iall y f or t hose bir d species kno w n t o pr ey upon them. On aver age, the co mbined effect of these 2
parameters accoun ted f or about 60 % o f t he annual variatio n in prod uct ivit y det ected f or 2 2 species.

226     Schiegg, Pasinelli, W alters & Daniels
Inb reeding and ex perience af fect resp onse t o clim ate change in Red-cockaded Woodpeckers. KA RIN SCHIEGG,
GILBERTO PASINELLI, JEFFREY R. WALTERS* and SUSAN J. DANIELS, Dept. Biol., Virginia Polytech. Inst. &
State Univ., Blacksburg, VA.
          Recent paper s hav e doc ument ed adv ances in eg g-laying d ates of several speci es of bir ds breedi ng at
nor th t emperat e lat it udes in r esponse t o ch angi ng c lim ate. W e exam ined egg-laying dat es of Red-co ckaded
Woodpeckers over the past 19 y ears in the North Carolina Sandhills, and over the past 13 yr at Camp Lejeune
in coastal North Carolina. In both populations egg-laying dates have become earlier, and these shifts are
correlated with w armer temperatures in the Sandhills and with increased precipitation at Camp Lejeune. We
determined that the changes evident at the population level w ere driven by shift s in the timing of egg-laying
made by ind ividu al bi rds . Indi vidual s t hat w ere abl e to respon d t o ch anges in c lim ate by laying earli er enj oy ed
higher reproductive success. However, not all individuals were able to make these adjustments. Specifically
females breeding for t he first t ime, females that changed mates and inbred birds did not exhibit the shifts in t he
timing of egg-laying that other birds did. Our data suggest that individual Red-cockaded Woodpeckers that are
able t o adj ust the t iming of their egg-laying may act uall y benef it from cli mat e chan ge, w hereas those unab le t o
make these adjustments are adversely affect ed. Our results provide another example of the reduced ability of
inb red an imals t o han dle en viron mental ch allen ges, in t his case g lob al cl imate chan ge.

227     Robinson, Angehr & Robinson
Climate change and bird exti nct ions on a Neotr opical island. W. DOUGLAS ROBINSON* , Dept. Biol. Sci.,
Auburn Univ., Auburn, AL; GEORGE ANGEHR, Smithsonian Tropical Res. Inst., Panama; and TARA R.
ROBINSON, Auburn Univ.
         Barro Colorado Island (BCI), Panama, is a 160 0-ha hilltop isolated during construction of the Panama
Canal. Since the first ornithological investigations in 19 23, 35% of t he avifauna has disappeared despite the
cont inu ed exist ence o f pop ulat ion s of all speci es in nearb y m ainl and f ores ts. We analyzed t he eco log ical
characteristics of all species lost from BCI and found t hat most species lost occupied narrow ranges wit hin the
w ettest f orests in central Panama. We compared the bird community composition of BCI through time w ith t he
species comp osition in each of 12 forested areas o n t he mainland. Imm ediat ely aft er iso lat ion, t he BCI
community w as most similar to the wet forest communit ies. Now, how ever, the community most closely
resembles a dry forest bird community. A century-long drying trend has been documented in the region, w hich
appears to be related to a global climate cycle rather than a result of deforestation or other local processes. We
hy pot hesize that dry ing of the island has r educ ed availab ili ty of reso urc es t hat w et fores t spec ies requi re,
leading to t heir local extinction. Eff orts t o conserve tropical diversity in isolated reserves like BCI may be
impeded w hen those reserves are exposed to global climatic changes that alter the microclimatic prof ile of the

228     Hussell
Climate change, spring t emperatures and timing of breeding in Tree Swallow s in southern Ontario. DA VID J. T.
HUSSELL, Ontario Min istr y of Natural Resources, Peterborou gh, ON.
          Several European studies have indicated advances in breeding dates of birds in the last 30 years, w hich
authors attributed to climate change. In North America, Dunn & Winkler (1999, Proc. R. Soc. Lond. 266B:
248 7-24 90) reported a continent-w ide advance of 5 - 9 d in breeding dates of Tree Swallows betw een 195 9
and 1991 . I present results of an intensive study of Tree Swallow s breeding on and near Long Point, Ontario,
at 4 sit es mo nit ored 19 69 - 200 1, 19 77 - 198 6, 19 78 - 200 1 an d 1 98 7 - 20 01 . Lo cal ai r t emperat ures var ied
among sites, according to distance from Lake Erie. Median dates of clut ch initiation by f emales 2+ yr old
varied among years by up to 19 d ov erall and by 15 d w ithin sites; and diff ered by 0 - 10 d betw een sites in
the same year. Th ere were no sig nif icant d iff erences in average timin g of breeding among t he 19 70 -19 80 ,
198 0 - 19 90 and 19 90 - 2 000 periods. Nevertheless, there w as a significant trend t ow ards earlier laying in
19 90 - 200 0, w hic h w as at trib ut able t o 3 exc ept ion ally early year s in 19 98 - 200 0. Across all sit es, median
dates of clut ch init iation w ere strong ly co rrelated w ith average daily maxim um t emperatures in t he first fou r 5-d
periods in May, but clut ches at one site w ere initiated an average of 3.1 d earlier than predicted by
temperatures alone. This analysis indicates that climate w arming of 5°C in May w ill likely result in an average
advance of about 7 d in the median date of laying of Tree Sw allows at t hese sites. I w ill discuss the
differences betw een my findings and those of Dunn & Winkler.

229      Mays & Hopper
Behavioral responses of Yellow -breast ed Chat s t o male and f emale mo del pr esent ati ons. HERMA N L. M AYS,
Dept . Biol. Sciences , A ubu rn Univ ., Aubu rn, AL; and KEVI N HOPPER* , Biol. Sciences, Univ. Kentucky
Lexington Comm . Coll., Lexington, KY.
           Inter- and intrasexual interactions wit hin a species reveal insights into mating sy stem evolution, but
such interactions are oft en difficult to observe consistently in the field. To study these interactions in a w ild
pop ulat ion of the social ly monog amous y ello w -br east ed ch at, w e pres ent ed br eedin g pai rs o n est abli shed
terri tor ies w ith 3 mo dels: a conspecif ic male, a consp ecific female, and a het erospecif ic (Carolina w ren)
control. W e measured vocal and behavioral responses of ten resident pairs to each model. There was no
significant eff ect of t reatment on the number of songs resident males sang or on our measure of male
aggression, although there were trends for males to sing more in the presence of t he female model and to be
more aggressive tow ard the male model. Male mating behavior responded significantly. M ales directed mating
behavior exclusively tow ard the female model, giving courtship displays in 40% of t hose observations and
even mount ing the model in 3 0% . Fin ally , t here w as a highl y sign if ican t effect on f emale ag gres sion. Females
direct ed more cont act calls and alarm c alls tow ard the f emale model and only at tacked f emale models. Thus,
although yellow-breasted chats are socially monogamous, the males we observed would engage in extra-pair
copulations when t he opportunity presented itself. How ever, the strong female-female aggression we
observed may act ually lim it su ch oppo rtu nit ies and thereby h elp maintain m onogamy in t his species.

230     Mumme
Scare tactics in a neotropical warbler: How w hite tail feathers enhance flush-pursuit f oraging performance in
the Slate-throated Redstart. RONALD L. MUMME, Dep t. Bio l. , Allegheny Coll., Meadville, PA.
         Myioborus redstarts are noted for t heir animated foraging displays in w hich birds expose their w hite
outer tail feathers and flush insects t hat are pursued and captured in flight. I examined the role of the whit e tail
feathers in flush-pursuit foraging by means of a field experiment on the Slate-throated Redstart (Myioborus
miniatus) in Monteverde, Costa Rica. Aft er locating nests cont aining young 5 - 7 d old, I captured the male and
female and randomly selected one for experimental plumage manipulation. For each experimental bird I used a
black marking pen to darken the white tips of its outer retrices. Its mate served as a control; I used the same
marking pen to sham-darken the tips of t he naturally black inner retrices. Experimental birds wit h darkened tails
fed their nestlings about 5 times/hr less than did their sham-darkened mates, primarily because of a decrease in
flush-pursuit foraging performance; for experimental birds, only 8% of hops in the fanned-tail posture w ere
fol low ed by an att ack on a fl ying i nsect, compared t o 21 % o f ho ps fo r cont rols. These experimental result s,
similar to those obtained recently by Piotr Jablonski for t he congeneric Painted Redstart, indicate that w hite tail
feathers are critically important in startling potential prey and raise questions about geographic variation in
plumage patt ern in t he genus.

231     Marzluff & Neatherlin
Responses of American Crows, Common Ravens, and Steller's Jays to human sett lement and recreation: More
dissimilar than alike. JOHN M. MARZLUFF and ERIK A. NEATHERLIN* , Coll. Forest Res., Univ. Washington,
Seattle, WA.
         Corvids have increased in many parts of t he world in response to human activity. In w estern
Washington, human populations are increasing rapidly which w as hypothesized to juxtapose increasing corvid
populations against forest-dependent endangered species. We tested this hypothesis from 1 995 - 200 0 on t he
Olympic Peninsula by surveying corvids, monitoring their home range movements and demography, and
assessing their rol e as nest pred ators on t he t hreat ened M arbl ed M urr elet . In t his paper w e desc rib e American
Crow, Common Raven and Steller's Jay response to human settlement and recreation on the Olympic
Peninsula. Crows responded most strongly to human settlement and recreation, w hereas jays displayed no
significant response for any parameter measured. Crows increased their abundance by a reduction in home
range size allowing them to attain higher densities near humans. Ravens appeared unable to compress home
ranges enough to allow density t o increase near humans. Crow s and ravens both capitalized on human food
sources, but this appeared to be an ephemeral, rather than sustained, population response for ravens. Ravens
nesting f ar from h umans accessed human food sources inf requently w ith li tt le effect on home range size. In
contrast, cr ow s nesting far from humans traveled daily to reliable human food sources increasing home range
size five-fold. Proximit y to humans positively inf luenced survivorship for crow s and ravens, but did not
signif ican tly affect fecun dit y f or c row s, rav ens or jay s on the Olym pic Peninsula.

232     Patricelli, Uy & Borgia
Interactive communication during mate choice in the Satin Bowerbird: An experimental test using robotic
females. GAIL L. PATRICELLI* , J. ALBERT C. UY and GERALD BORGIA, Dept. Biol., Univ. M aryland, College
Park, MD.
          Sexual selection in many species involves communication during mate choice, as males display their
secondary sex traits t o assessing females. Male display traits are generally assumed to remain constant
throughout t he mate choice process, but t his may not be so. Male success may require the ability t o produce
attractiv e traits and the ability t o modify t hem in response to female signals. In a field experiment using robotic
females, w e show that such interactive communication occurs during mate choice in the Satin Bowerbird
(Ptilonorhynchus violaceus). Males w it h hi gh i nt ensity displays are pref erred as m ates, but int ense displays can
frig ht en f emales early in t he mate cho ice p roc ess. As f emales assess pot ent ial m ates, they signal t he di splay
int ensity t hat they w ill accept , w hic h in creases as m ate cho ice p rog resses. Males t hat effect ively modul ate
their display intensity in response to f emale signals, startle females less often and improve their courtship
suc cess. Sex ual select ion for i nt ense m ale di splays, w hic h can threat en f emales in o ne co nt ext and at tract them
in an ot her, may facil it ate the evolu tion of int eractive com munic ation bet w een t he sex es; favorin g f emales that
signal the level of display intensity t hey will accept, and males that modify t heir displays in response to f emale
signals. Our results suggest that int eractive communication is important for eff ective courtship in satin
bow erbi rds and m ay have a m ore s ign if ican t rol e in shapi ng t he ou tcome of sexual select ion than h as been

233     Langham
Rufous-tailed Jacamar predation patterns on novel Heliconius butterflies. GARY M. LANGHAM, Dept. Ecol. &
Evol. Bio., Cornell Univ. , It haca, NY.
         Rufous-t ailed jacamars,Galbu la ru ficau dai, are but terf ly predat ors and t hus an excellent stud y org anism
to estimate predation rates on aposematic butterflies. Heliconius butterflies are toxic and 2 species, H. erat o
and H. melpomene, mimic each other throughout the lowland Neotropics. It is widely assumed that birds drive
the evolution of but terfly m imicry via increased predation pressure on rare forms. This frequency dependent
selection is often thought to maintain color patterns within and betw een large regions of similar wing patterns.
Using 3 large outdoor cages (2 x 2 x 4 m) in the Andean piedmont of Venezuela, I tested 80 jacamars by
presenting t hem wit h modified local Heliconius. To create novelty, I modified the red forew ing patch w ith
either a red or black Sharpie pen; the third but terfly w as unmodified. In particular, I w as investigating sampling
dif ferenc es bet w een sex es, ages, w it hin breed ing season, etc. Result s indic ate signif ican t dif ferenc es in males
and females sampling behavior: 21 of 40 males and 8 of 40 females sampled one of the novel butterflies. I
retested 46 birds from 4 to 6 85 d later. 1 4 of 18 birds t hat sampled the first time did so again, w hile 3 of 28
non-samplers sampled the second time. These results suggest that jacamars play an important role in the
evolution and maintenance of Heliconius butterfly mimicry.

234     Schuetz
Reducing the risk of n est predat ion: adapt ive use of carnivo re scat by Common W axbills (Estrilda astrild).
JUSTIN G. SCHUETZ, Dept. Ecol. Evol. Biology, Cornell Univ., Ithaca, NY.
        Common Waxbill s build an enc losed gr ass nest w hic h t hey lin e (inside and o ut ) w it h a rat her u nusual
material: fur excavated from carnivore scat. This behavior has not previously been studied in the wild though
captive birds have been reported to drape dead nestlings and insect carcasses across the top of their nests.
Research ers h ave sugg est ed t hat suc h nest ' deco rat ion s' func tion to red uce the ri sk o f nest pred ation . I test ed
this hypot hesis by baiting artif icial nests w ith f inch eggs and lining half of them w ith f ur-laden scat. Over the
course of 16 d (t he typical laying and incubation period for w axbills) scat-lined nests suffered lower predation
than unlined nests. This result suggests that use of carnivore scat by Common Waxbills is an adaptive
response to p redation p ressure. Furt her research is required t o ident ify pot ential nest predator s and to est ablish
w hether scat activ ely deters pr edators, or rat her, con ceals the scent of eggs o r birds i n th e nest.

235     Cooper, Dececco, Mullin & Wilson
Nes t site select io n stud ies an d n est pr edat or s: What are w e measur in g and w hy ? ROBERT J. COOPER* ,
JENNIFER A. DECECCO, School For . Res. , Univ. Georgia, A thens, GA; STEPHEN J. MULLIN, Dept. Biol. Sci.,
Eastern Illinois Univ ., Charleston , IL; and R. RANDY WILSON, U.S. Fish & Wil dl. Serv. , V ick sbur g, MS.
         Most of t he studi es that have sought to i dentif y f actor s that inf luence the lik elihood of songbird nest
depredation have compared numerous variables betw een successful and unsuccessful nests. The majority have
foun d no dif ferenc e bet w een t he 2 gro ups. W e took a sim ilar appr oach and f oun d no dif ferenc e bet w een
suc cessful an d un suc cessful n est s of Acadi an Fly cat chers in a bo ttomland fores t. Yet w e also foun d t hat 1
tree species, Nuttall oak, had high selectivity as a nest substrate, and that nests in Nuttall oaks tended to be
suc cessful m ore t han i n ot her t ree sp ecies. This trend w as co nsist ent for 3 out of 4 y r. We cond ucted an
experim ent w it h cap tive gray rat snak es t o assess t heir abil it y t o cl imb Nuttall o ak and 2 o ther c ommon t ree
species. We found that none of the 24 snakes could climb large (> 30 cm dbh) Nuttall oaks because of t he
smoothness of the bark, and that vines decreased the amount of time needed for rat snakes to climb 2 m. We
propose that, rather than focusing on the nest alone, researchers interested in nest depredation should focus
more at tent ion on p redat ors , an d let their behav ior s suggest meaning ful v ariab les t o measur e in n est ing habi tat

236     Johnson & Oring
The use of predator exclosures in research, management, and conservation of ground nesting birds. J.
MATTHEW JOHNSON* and LEWIS W. ORING, Dept. Environ. & Resources Sci., Univ. Nevada, Reno, NV.
          Researchers, managers, and conservation biologists oft en wish t o identify f actors influencing
reproductive success. Avian, mammalian, and/or reptilian egg predators may limit breeding bird populations.
High lev els of egg p redat ion hin der s tudy of avian rep rod uctive bio log y. Cont rol lin g f or p redat ion bias
associated w ith investigator presence is essential in many studies. Although activ e control of pot ential egg
predators through lethal methods is sometimes practical, these methods are time consuming, diff icult, and
controversial. During 1999 and 2000, w e studied depredation of Killdeer nests in the nw. Great Basin. We
experimentally demonstrated that predator exclosures placed around Killdeer nests significantly decreased daily
mort alit y r ates co mpared t o nests w it hou t exc losures (p < 0. 00 1). A lt hou gh p redat or ex clo sur es may be used
to increase nest success for Killdeer in the nw. Great Basin, it is necessary to identif y w hat predators at a site
are impacting nest success prior to exclosure design. Predator exclosures are unlikely to be equally successful
among sites, and variation in exclosure design is likely to be required across the range of a species and/or
betw een seasons at a single site. The predator exclosure w e used does not deter small birds, mammals or
reptiles; and mustelids kill incubating birds. However, t his exclosure is still effectiv e at decreasing daily
mortality rates of nests even when small nest predators are present, as exclosures deter larger predators
(ravens, gu lls, c anids). A lth ough w e have show n predator exclosur es reduced daily mor talit y rat es of nest s,
this benefit is likely off set by reproductive loss resulting f rom adults killed on the nest. Use of predator
exclosures requires detailed knowledge of the source of egg predation.

237      López & Torres
Fidelity and divorce in the Blue-footed Booby. I. LÓPEZ* and R. TORRES, Dept. Ecologí a Evolutiv a,
Ecologí a, UNAM, MÉ   XICO.
           Alt hough there are benefit s associated wit h staying w ith t he same partner for successive breeding
events (Familiarity Hypothesis), pair-bond duration in monogamous birds varies within and betw een species.
The Incompatibility Hypothesis states that birds divorce because they have low reproductive success although
the characteristics of each individual per se are not negative. We evaluated benefits from f idelity in the
blu e-f oot ed bo oby and t est ed pr edic tion s f rom the In compatibi lit y Hyp ot hesis in this seabird. Expected
advantages from t he Familiarity Hypot hesis were supported by our data. Faithful pairs laid their first egg 8 d
earli er in the seaso n t han d ivorc ed pai rs, and h ad gr eater t han av erage h atching suc cess. W hen d ate w as
considered, hatching success of faithf ul pairs increased from 60 to 7 6% , w hile success of divorced pairs
decreased from 84 to 1% . There w as a positive correlation betw een the pair duration and the number of chicks
fledged (GLIM, P 2 = 21. 81, df = 7, p = 0.0 02), and the proportion of divorces decreased wit h the duration of
the pair-bond (P 2 = 45. 51, df = 1, p < 0.0 001 ). Predictions from the Incompatibility Hypot hesis were not
support ed. The probabil ity of d ivorc e w as not relat ed to pr evious repro duct ive success ( P 2 = 1.29, p = 0.25,
n = 95), and breeding success before and after the divorce did not differ betw een sexes (Wilcoxon W = 115,
p = 0.32, n = 34; W = 166, p = 0.12, n = 32). Thus, the Blue-footed Booby benefits from fidelity,
althou gh t he Incompat ibilit y Hyp oth esis does not seem t o be the mechanism of d ivorc e in thi s bird species.

238      D’Alba & Torres
Resource allocat ion t o eggs and surv ival of Blue-foot ed Booby chic ks. LILIANA D' ALBA* and ROXANA
TORRES, Dept. Ecolog ia Evolout iva, Inst . Ecolog ia, UNA M. MEXICO.
          In the Blue-footed Booby (Sula nebouxii) different ial r esource allo cat ion to egg s can be expect ed du e to
marked s exual size dim orp hism and oc currence o f bro od r educ tion thro ugh aggr essive behavior bet w een
siblin gs. We ev aluat ed di fferent ial r esource allo cat ion in eg gs of dif ferent sex and l aying o rder , an d anal yzed
relationships betw een egg volume/mass and offspring survival. Volume and initial mass of 1 30 eggs of know n
laying order, and survival of chicks t o fledging w as recorded. No differences in volume and mass w ere found
betw een eggs wit h female and male embryos. Overall, first- and second-laid eggs did not differ in volume and
initial mass. However, there was a seasonal decrease in relative allocation (only in mass) to first - and
second-laid eggs (R = 0.2 9, P = 0.0 25, n = 57). Eggs wit h greater mass and volume had higher probability
of survival at hatching ( P 2 = 8.6 3, P = 0.0 03, n = 127 ) and until chicks were 10 d old ( P 2 = 3.96, P =
0.0 466 ; n = 84). Larger and heavier eggs produced heavier chicks at hatching (R = 0.4 1, P = 0.0 002 , n =
82 ). Thu s, var iat ion in r esource allo cat ion to egg s is relat ed t o ch ick sur vival. In t his siblic idal bir d, seasonal
variati on in relat ive allocat ion t o fi rst- and seco nd-laid eggs may be adapt ive, in creasing t he probabilit y t o raise
bot h ch ick s in favorabl e repr odu ct ive con dit ion s (early bro ods), and f avorin g eli minat ion of seco nd-hat ched
chick in late broods, when reproductive conditions are deteriorated.

239      Clifford & Anderson
Food limitation contributes to clutch size variation in the Nazca Booby. LESLIE D. CLIFFORD* and DAVID J.
ANDERSON, Dept . Bio l., W ake Forest Univ ., Win st on-Salem, NC.
         Natural selection is expected to optimize clutch size, but int rapopulation variation is maintained in many
bird species. We investigated intrapopulation clutch size variation in the Nazca Booby, which lays 1 or 2 eggs.
The Nazca Booby provides a simple system in w hich to investigate clutch size evolution because clutch size and
brood size are decoupled due to obligate siblicide. When both eggs of a 2-egg clutch hatch, t he younger sibling
is eliminated through sibling aggression, and brood size is invariably 1. Thus, no indirect effect of brood size on
clutch size evolution exists in t his species. Second eggs provide insurance against the failure of t he first egg or
early death of the f irst hatchling and therefore increase reproductive success, but a substantial fraction of
females lay only 1 egg. We tested the hypothesis that 1-egg clutches result from f ood limitation by prov iding
female Nazca boobies with supplemental food. A higher proportion of supplemented females produced 2-egg
clutches than did control f emales. Supplemented females produced larger second-laid eggs than did control
females, but not first-laid eggs. Layin g dat e and lay ing int erval w ere not aff ect ed. Comp ariso ns of clu tch size
and eg g volu me betw een years indi cat ed t hat the suppl emental f eedin g experi ment w as not con duc ted in a year
w ith a poor natural food supply. Thus supplemented females produced larger clutch sizes despite apparently
normal natural food levels. This experiment indicated that f ood limitation and the costs of egg-laying should be
considered carefully in studies of clutch size evolution.

240     Morgan, Anderson & Ashley-Ross
Nazca Booby feet prov ide heat to eggs during incu bati on. STEPHANI E M. M ORGAN* , DA VID J . A NDERSON
and MIRIAM A. ASHLEY-ROSS, Dept . Bio l., W ake Forest Univ ., Win st on-Salem, NC.
          Lacking brood patch es, Nazca Boobies (Sula grantii) cover their eggs with t heir webbed feet during
incubation. Whether birds add heat to eggs through w ebs or passively direct heat f rom abdomen to eggs is
unconfirmed and controversial. Also, no studies to date have examined incubation temperature in Nazca
Boobies. Incubation temperatures were tracked for individual Nazca Boobies nesting in a colony on the island
of Española, Galápagos, Ecuador. Plastic Easter eggs, identical in size to actual eggs, containing wat er and a
temperature data logger to record egg temperature, w ere sealed wit h w hite acrylic paint to closely resemble
natural eggs. Natural eggs in 52 1 -egg clutches were replaced w ith logger eggs at various points of the
inc ubat ion peri od. Bird s w ere all ow ed t o in cubat e the lo gger eggs for 3 d. All eg g t emperat ures remained
abov e ambient temperat ure d uri ng i ncubat ion . Ov er t he 3 -d int erv al, mean egg t emperat ures w ere 3 7°C (ran ge:
2 - 3°), w hile ambient temperature fluctuated cyclically throughout each 24-hr period from 22°C to 35°C. Very
larg e eggs w it h isolat ed t herm ocoup les position ed t o rec ord foot temperat ure s eparat ely from abdo men
temperature indicated that f eet provide heat to eggs independently of t he abdomen.

241     Bowie, Fjeldsa, Hackett & Crowe
Systematics of the endemic Double-collared Sunbirds (Nectariniidae) of the eastern arc mountains of Tanzania:
A com bined morpholog ical and molecular approach. RAURI C. K. BOWIE* , Percy Fitzpatrick Inst., Univ. Cape
Town, South Africa, JON FJELDSA, Zool. Mus., Univ. Copenhagen, Denmark, SHANNON J. HACKETT, Field
Mus. , Chicago, IL; and TIMOTHY M. CROWE, Percy Fit zpatrick Inst.
         The 'Double-collared Sunbirds' of the eastern arc mountains of Tanzania and se. Kenya are
characterised by regi onal var iat ion in m orp hol ogy and p lum age pat terns that has r esulted in much disput e over
their taxonomic status and the delineation of range boundaries. It has been suggested that Moreau's sunbird
(Nectarinia moreaui) is an atypical phenotype that has arisen from a hybridisation event between the more
widely distributed Eastern Double-collared Sunbird (EDC; N. mediocris) and the nar row ly distrib ut ed Lov erid ge' s
Sunbird (N. lov erid gei). A discriminant analysis of 6 standard morphological characters indicates Moreau's
sunbir d is distinc tive in i ts mo rph olo gy , an d is int ermediat e in shape an d size bet w een EDC and Lo ver idg e' s
sunbir ds. Ther e is great er ov erlap bet w een f emale M oreau ' s and EDC than bet w een males, w hil e Loverid ge' s
show s lit tle ov erlap w ith the ot her 2 species. Conversely, molecular analyses of 72 0 bp o f m tDNA (ND3 &
control region) shows Moreau's sunbird to be monophyletic, and the sister species to Loveridge's sunbird.
High sequenc e div ergen ce (4 %) bet w een t hese 2 taxa supp ort their st atus as separ ate spec ies. Furt herm ore,
EDC haplotypes can be divided into 3 distinct clades separated from each other by 9 - 10 % sequence
div ergen ce. Addi tion al in vestigat ion w ill be req uir ed t o det ermine w het her eac h of these c lades w arrant spec ies
st atus. Thi s co mbined morph olo gic al and molecu lar ap pro ach d emonst rat es t hat morph olo gic al similari ties m ay
be dependent upon similar ecolog ical requirement s, and may n ot n ecessarily corr elate w ith genetic r elatedness.

242     Prum & Williamson
Theoretic al models of the gro w th and ev olut ion of w ith in-f eather pigment ation p att erns. RICHARD O. PRUM* ,
Dept. Ecol & Evol. Biol., Nat. Hist. Mus. , Univ. Kansas, Law rence, KS, and SCOTT WILLIAMSON, Dept. Ecol &
Evol. Biol., Univ. Kansas.
        Feathers ar e com plex , b ranc hed k erat in struc tures that exhibi t div erse p igm ent ation pat terns . Feat her
pig ments are t rans ferred in to dev elop ing feather k erat ino cy tes by pig ment cell s t hat migrat e int o t he t ubu lar
feather germ from the dermis. Within f eather pigment patt erns are created by the diff erential pigmentation of
keratinocytes wit hin multiple, independent barb ridges. Little is known about the developmental mechanisms
that determine w hich feather keratinocyt es receive pigment. W e present reaction-diffusion models of the
grow th of w ithin f eather pigment patt erns. The models accurately simulate the grow th of a diversity of t he
w ithin f eather pigmentation patt erns found in real feathers, including a central patch, a ' hollow ' central patch,
concent ric cent ral p atches, bars , c hevron s, a cent ral c irc ular spot, r ow s of pair ed spot s, and ar ray s of of fset
dots. The models can also simulate the complex transitions betw een distinct pigmentat ion patterns observed in
various avian plumages, including transitions from bars to chevrons, bars to paired dots, and bars to arrays of
dots. The congruence betw een the developmental dynamics of the simulated and observed feather patterns
indicates that the reaction-diffusion models provide a realistic and accurate description of t he determination of
pigment pattern within avian feather follicles.

243     Allen
Genetics of the Baltimore Oriole x Bullock's Oriole hybrid zone. EVA SANDERS ALLEN, Dept . Biol. and Cen ter
for Integr ativ e Stud. Ani m. Behav., Indiana Univ. , Bloomin gto n, IN.
         Baltimore Oriol es (Icterus galbula) and Bullock' s Orioles (I. bullockii) hav e lon g been kno w n t o in terbr eed
w here their ranges overlap in the Great Plains of North A merica. The persistence of t he hybrid zone raises the
question of possible selective pressures that maintain the separateness of the species. To investigate this
question, I assessed variation in mitochondrial (mt) DNA haplotypes and plumage characters across transects of
the hybrid zone. Two transects surveyed in the early 1990s by Rising (1996, Condor 98 : 6 58 -663 ) co ver ed
river sy stems t hat dif fer gr eatly in the put ativ e age of cont act of the 2 species. I surv eyed one of t hese
transects again in 2000, which allowed me to ask whether haplotype frequencies have changed over time.
Plumage ch aracters an d mtDNA hapl ot yp es varied cli nall y ac ros s t he hy bri d zon e for al l t rans ect s. Ther e w as
no significant correlation betw een plumage and haplotype, w hich indicates the presence of backcross and
late-generation hybrid individuals, and demonstrates that hybrid individuals reproduce successfully. The
haplotype frequencies across transects w ere quite similar, whet her compared betw een river systems or wit hin a
single river system over time, w hich suggests that the struct ure of this hybrid zone is not affected by duration
of cont act betw een the species. My findings support earlier conclusions based on phenotypic data that t he
oriole hybrid zone is maintained by ecological selection pressures, specifically rainfall during the breeding
season (Rising 1983, Auk 10 0: 88 5-89 7). A fine-scal e analy sis of plu mage cli nes is underw ay, w hic h may
reveal differences in the strength of selection on various plumage traits.

244     Jones, Dieni & Gouse
Composit ion an d product ivi ty of a grassl and bi rd c omm unit y in north easter n M ont ana. STEPHANIE L. JONES,
USFWS, Den ver, CO, J . SCOTT DIENI, Redst art Consult ing, Evergreen, CO, and PAULA J. GOUSE, USFWS,
Malta, M T.
         We describe the reproductive ecology of a breeding bird community in northern mixed-grass prairie.
Our study area w as located in northeastern Montana, w here we conducted intensive nest searching during the
summers of 19 96 - 20 00. Overall, 25 species nested within the study area, including 8 w aterfow l, 5 shorebird
and 7 passerine species. Six passerine species accounted for 97% of 68 3 landbird nests located from 199 7 -
20 00 . Ch est nut -co llar ed Lon gspurs (n = 31 7 n est s) w ere t he most abun dant passerin e, f oll ow ed by Savan nah
Sparrow s (n = 13 0), Western M eadow larks (n = 68 ), Baird' s (n = 53 ) and Grasshopper sparrow s (n = 46 ),
and Sp ragu e' s Pipits (n = 46 ). Tot al nests amo ng t his gro up d ecli ned f rom 11 3 p er 1 00 ha in 19 97 to 6 1 p er
100 ha in 2000 . How ever, relative contribut ions among the six species were consistent across years. Average
annu al M ayf ield nest suc cess rang ed f rom 28 % for Sprag ue' s Pipits t o 3 3% for Baird' s Spar row s, and show ed
considerable annual variation w ithin species. Nest depredation was commonly observed, averaging from 28%
for Grasshopper Sparrows to 5 7% for Sprague' s Pipits. Nest parasitism w as low, averaging < 11% for each
species, except Western Meadowlarks, which incurred an average parasitism rate of 40% across all years.
Returns for ad ult Savan nah Sp arrow s (0 .0 %, n = 16 ) and Baird' s Spar row s (5 .8 %, n = 69 ) w ere f ew ,
suggesting low sit e fidelity. A nnual variation in relative abundance and nesting success, coupled wit h a low
ret urn rat es, pro vide f urt her i ndi cat ion of the variab le co ndi tion s underl ying m ixed-grass prai rie.
245      Buehler & Baker
Unraveling the global popu lation st ruct ure of t he Red Knot. DEBORAH M . BUEHLER and ALLAN J . BAKER,
Royal Ontario M us. and Dept . Zool. , Univ . Tor ont o, Tor ont o, ON.
         Red Knot s (Calidris canutus) are long distance migrant shorebirds whose population biology and
physiology in relation to their circumpolar migrations have long fascinated ornithologists. Population structure
based on morphological features recognizes 5 subspecies: C. c. islandica, C. c. canutus, C. c. rogersi, C. c.
roselaari, and C. c. rufa. Alt hough knots can be recognized morphologically on their migration routes and in
their w intering grounds, litt le is known about t heir genetic population struct ure. To resolve their population
genetic struct ure knots were sampled in their staging and wint ering areas and mtDNA and microsatellites were
exam ined for g enet ic v ariab ili ty. Sequen cin g of the mtDNA con trol regi on u sing Lon g Tem plat e PCR conf irm ed
the mitochondrial DNA origin of previous sequences which indicated very litt le variation. This conf irms the
general low variability and lack of diff erentiation among subspecies in the knot cont rol region. Sequencing of
the mit ochon dri al pr ot ein c odi ng g enes ND2 an d A TPase 6 and 8 also show ed lo w var iat ion . This dat a
indicates that the mtDNA molecule is bottlenecked. In addition 3 micr osatellite primers w ere located wit hin the
nuc lear g enom e. A ll o f these loci hav e w ere mono morph ic i ndi cat ing low var iabi lit y in t he nu clear geno me as
w ell as in mit ochond rial DNA. Future research w ill incl ude the co mplet e sequencing of the mi toc hondri al DNA
cont rol region and the develop ment o f several addit ional micr osatellit e markers and AFLP polymorph isms t o
screen the nuclear genome.

246      Whittier, Leslie & van den Bussche
Lack of genet ic variat ion bet w een subspecies of Least Tern: Implicat ions f or conserv ation. JOA NNA
WHITTIER* , DAVID M. LESLIE, Jr., Oklahoma Coop. Fish & Wildl. Res. Unit, Oklahoma State Univ., Stillw ater,
OK; and RONALD A. VAN DEN BUSSCHE, Zool. Dept. , Oklahoma State Univ ., St illw ater, OK.
          DNA sequence variat ion f rom int rons w ith in nuclear genes and part of t he mit ochond rial cyt ochrom e-b
gene w ere used to evaluate population struct ure wit hin and among the 3 breeding populations of Least Terns in
the U. S. (Calif orn ia [Sterna antillarum brow nii], Int erior [S. a. athalassos], Eastern [S. a. antillarum]). Sequence
variation was highest for nuclear intron XI (Gadp) within the Glyceraldehyde-3-phosphate dehydrogenase gene.
The second nuclear intron was fixed for the same allele in all breeding populations. Fixation indices, FST and
Phi ST, for Gadp i nd ic at ed genetic di vergenc e bet w een on ly the Calif or ni a and i nt erio r b reedin g p op ul at io ns (P <
0. 05 ). How ever , m it ochon dri al DNA genet ic i ndi ces d id n ot dif fer among breed ing pop ulat ion s. These dat a
may indicate gene flow via female dispersal.

247      Griffiths & Bates
Genetic data and the value of voucher specimens. CAROLE S. GRIFFITHS, Biol. Dept., Long Island Univ.,
Brooklyn , NY and Dept . Ornit hol., Am . M us. Nat . Hist ., New York NY, and JOHN BATES, Bird Dept. , Field Mus.
Nat . Hist ., Chic ago, IL.
         The genus Cathart es w as rev ised in 1 96 4 (Wet more1 96 4, Smithson. M isc. Coll. 146 , no. 6 ) and
subdivided into 3 species, Turkey Vulture, Lesser Yellow-headed Vulture and Greater Yellow-headed Vulture.
The Greater Yellow -headed Vulture w as designated as a separate species based on slight diff erences in overall
color of t he plumage, and mainly by t he larger size. The distinct ive measurements are the measurements of
retrices. These measurements and color differences are difficult to use for identifying these birds in the field.
In the field, identification is based on habitat. The Lesser Yellow -headed Vulture occurs in open grasslands and
marshes, w hereas the Greater Yellow -headed Vulture occurs in forested areas. As part of an analysis of t he
family Cathartidae, samples of Greater Yellow -headed and Lesser Yellow -headed Vultures were sequenced. The
sequence of one specimen was initially identified as a Lesser Yellow -headed because it w as collected in Brazil
flying low over w et grassland near a small Gallery forest, but 10 - 1 5 km f rom Terra Firma forest. How ever, the
sequence o f this spec imen w as ident ical to Great er Yellow -headed Vult ure s equen ces an d di ffered f rom ot her
Lesser Yellow -headed sequences by 4%. Based on measurements of t he voucher specimen, we could confirm
that t his was a Greater Yellow -headed Vulture. This w ould not have been possible if t he specimen had not
been collected and deposited as a voucher.

248     Rowe, Gehlbach & Kim
Phylogeography of North American screech owls, Otus asio and Otus kennicottii, b ased on cy tochro me b
sequences. DIANE L. ROWE* , Dept. Wildl. & Fish. Sci., Texas A&M Univ., College Station, TX; FREDERICK R.
GEHLBACH, Dept. Biol., Baylor Univ., Waco, TX; and DANIEL H. KIM, Caesar Kleberg Inst., Texas A&M Univ.,
Kingsville, TX.
         Species designations f or New World Scops Ow ls (Screech Ow ls, genus Otus) have been problematic,
ow ing to cl inal var iat ion in m orp hol ogi cal f eatures and p lum age co lor . Fur therm ore, geog raph ic as semb lages
w ithin each species have previously been defined and redefined according to diff erences in wing cord, bill
length, and mass. In this study, w e present phylogenetic relationships of New World Otus species based on
cy tochro me b sequences. In addition, Otus asio and Otus kennicottii, each displaying strong clinal variation in
size and plumage across their entire ranges, are analyzed phylogeographically. Several representatives of all six
of t he recognized subspecies of O. asio are included in analyses of how genetic variation is apportioned wit hin
and among designated subspecies. The genetic data are compared to morphological subspecific classifications.
Preliminary analyses are presented for O. kennicottii, but are limited in scope due to dif ficult y in sampling
southern geographic assemblages.

249     McCracken & Sorenson
Phylogeography and species limit s of t he stif ft ail ducks (Nomonyx - Oxyura). KEVIN G. McCRACKEN* , Inst.
Arctic Biol. and Dept. Biol. & Wildl., Univ. Alaska Fairbanks, Fairbanks, AK; and M ICHAEL D. SORENSON,
Dept. Biol., Boston Univ., Boston, MA.
         Nomonyx-Oxyura stiff tail ducks are a group of small-bodied diving ducks, w ell known f or their
promiscuous mating activity , large testes, and long, spiny penises. Half of the w orld's t rue stifft ails are
endem ic t o Sou th A merica an d adj acent part s of North A merica (Nomonyx, Oxy ura f errugi nea, O. andina, O.
vittata), and some of these are sympatric in So ut h A merica (O. ferrugi nea and O. vittata at the base of t he
Andes and Nomonyx and O. vittata in the sub-tropics). One each of the other 4 stiff tails are endemic to North
America (O. jamaicensis), Eurasia (O. leucocephala), Afric a (O. maccoa), and A ust rali a (O. australis). To b etter
understand phylogeographic and population level relationships, w e sequenced and aligned most of the cont rol
region and part of t he adjacent phenylalanine tRNA genes (bp 78-12 51) of 4 Nomonyx and 77 Oxyura.
Nomonyx, O. vittata, and O. jamaicensis are the basal-most stifftails. The Andean Ruddy Duck, O. ferrugi nea,
is m ono phy let ic an d po ssibly nest ed w it hin nom inat e O. jamaicensis, suggesting a secondary invasion of the
Andean Cordillera by Ruddy Ducks from North America. Old World species, O. leucocephala, O. maccoa, and
O. australis, form a more recently derived clade. Monophy ly of each species is robustly supported for all but
nom inat e O. jamaicensis.

250     Pearce, Talbot, Pierson & Scribner
Genetic struc ture o f King Eiders: M igr ato ry behav ior s durin g summer and w int er. JOHN M . PEARCE* ,
SANDRA L. TA LBOT, BARBARA J . PIERSON, U.S. Geol. Sci. Center, Alaska Biol. Sci. Center, Anchorage, AK;
and KIM T. SCRIBNER, Mich igan State Univ. , Dept. Fish. & W ildl., East Lansing, MI.
          The degree of male and female sea duck site fidelity t o w intering areas is not w ell known, nor is t he
amount of intraspecific admixture during wint er by disparate breeding groups. Using nuclear microsatellite loci
and m it ochon dri al DNA sequence f rom cy tochro me b and the control region, w e examined the genetic struct ure
of breed ing and w int erin g gr oup s of King Eiders f or ev iden ce of breed ing and w int er ph ilo pat ry and w int er
admixture. We obser ved lit tle gen etic d ivergen ce among 5 ar ct ic b reedi ng ar eas of King Eiders w it hin Russia,
Alaska, and Canada. However, w e did find diff erences among wint ering areas. The lack of genetic
dif ferent iat ion among b reedi ng ar eas co upl ed w it h gen etic h eterog eneit y betw een w int erin g areas suggest s t hat
multiple w intering areas may be mixing on the breeding grounds and that w inter site fidelity occurs w ithin bot h

251      Bridge
The aerody namic and physiological effect s of sim ulated primary m olt on flig ht in Rock Dov es. ELI S. BRIDGE,
Dept. Ecol. Evol. Behav., Univ. Min nesota, St . Paul, M N.
         Understanding the costs and benefit s of flight feather replacement is essential for explaining the
evolutionary significance of molting strat egies. I investigated the effects of simulated molt gaps on the flight
ability of Rock Doves by examining flight perf ormance before and after removing measured lengths of either t he
innermost or outermost 3 primaries. All f light sessions were videotaped and took place in a large metabolic
chamber, allow ing for simultaneous measurement of flap rates, flight v elocities, and metabolic rates. Wing gaps
of bot h types increased w ing loading by approximately 10 % and caused an increase in flap rate of about 7 %.
Metabolic rate w as not significantly affected by the presence of molt gaps. There was no detectable difference
betw een the 2 w ing trimming t reatments wit h regard to aerodynamic or metabolic measurements. In general,
the det rim ent al ef fect s of w ing gaps w ere mino r, w hic h len ds supp ort to t he no tion that the gr adual , sequen tial
molt st rat egy prev alent among b ird spec ies is an adapt ation for s pread ing out the ener get ic c ost of feather
synt hesis rather t han a means of mini mizing t he aerodynamic im pediment of m olt gaps.

252      Pereira & Baker
Molec ular phy log eny and b iog eogr aphy of cur assow s (Cracid ae, Gallif orm es). SERGIO L. PEREIRA* and A LLAN
J. BAKER, Centr e Biodiv ers. & Conserv . Biol ., Royal Ont ario M us., To ront o, ON.
         Cracids comprise a family of large gallinaceous birds that occur in the Neotropics, and about 50 % of
species are endangered. How ever, their systematics are in urgent need of revision to bett er understand their
phylogenetic relationships and to help preserve this group of seed-dispersing birds from extinction. We
sequenced several mitochondrial genes from 13 of the 14 c urassow species and estimated their phylogenetic
relationships using maximum likelihood and maximum parsimony. Trees constructed wit h both methods are
congru ent and recov er a clade made up of all t he curassow s of t he genus Crax and another clade with Mitu,
Pauxi and Not hoc rax , although the positioning of Not hoc rax w as not w ell supp ort ed. One of the Pauxi spec ies
w as nested w ithin Mitu, making both genera polyphyletic. The general geographic pattern of curassow
divergence can be att ributed to vicariance due to geographic barriers such as the Andes and some major rivers
in t he Amazon Forest, and by t he form ation o f t he Brazilian savanna betw een the Am azon and At lantic Forests.
Differentiation among very closely related species in some cases can be att ributed to adaptation to and isolation
in highl and habitat s near the A ndes. Grants: FAPESP (Brazil) and NSERC (Canada).

253      Kondo & Omland
Genetic diversity of the Baltimore and Black-backed orioles: Analysis of a long and short distance migrant
species pair. BEATRICE KONDO* and KEVIN E. OMLAND, Dept. Biol. Sci., Univ. Maryland, Baltimore Co.,
Baltimore, MD.
          A recent phylo geny of New World Orioles (Icterus) yield ed some su rpr ising r elat ion ships at the species
and subspecies levels (Omland et al. 1999, Mol. Phylo. Evol. 12: 224 -239 ). Particularly striking w ere the
grouping of long and short distance migrant species or subspecies as each other’ s closest neighbors. This
occurred w it h Balt imore Orio le (Icterus galbula) and Black -bac ked Orio le (I. abeil lei), w it h Bulloc k’ s Ori ole (I.
bullockii) and Streak-bac ked Orio le (I. pustulatus), and w it h Orchard Orio le (I. spurius spurius) and Ochre Oriole
(I. spur ius f uertesi). These are 3 separate instances in which a long distance migrant is most closely related to
a short distance migrant, yielding an excellent model system for examining the gain or loss of migration. We
are ex amin ing mitochod rial cy tochro me b and control region genes of multiple individuals from throughout the
ranges of the Baltimore and Black-backed orioles to compare genetic diversity within and betw een species.
Preliminary results indicate that Black-backed is very homogeneous, displaying only 1 character difference in
40 0 b p of cy tochro me b, which difference occurred in only 1 of the 8 Black-backed sequenced so far. We have
also found < 0.5% divergence betw een Black-backed and Baltimore in the samples sequenced to date.
Compari son of the f ull cy tochro me b and control region sequences in this pair of closely related species should
yield more insight int o the time scales of speciation and gain or loss of long distance migration.

254     Ibanez-Hernandez
Geogr aphi c v ariat ion in m orp hol ogy of the Blue-cr ow ned M ot mot spec ies c omplex . P. GABRIELA
IBANEZ-HERNANDEZ. Mus. Zool., Fac. Ciencias, UNAM, Mexico.
          The Blue-crowned Mot mot (Momotus momota) is a highly polyt ipic taxon that is w idespread along the
Neotropics, from eastern Mexico to n. A rgentina. Patterns of geographic variation in external morphology and in
plumage coloration are complex, and there is a possibility that t he taxon is composed of several evolutionary
ind epend ent uni ts (sp ecies). To c ont rib ut e for t he un ders tandi ng o f spec ies limit s w it hin the t axon, I analyzed
geog raph ic v ariat ion in m orp hom etric and c olo rat ion of the po pul ation s along t he w hol e geog raph ic r ange.
Statist ical uni var iat e and m ult ivariat e analy ses su ggest the exist ence o f several g rou ps, each c learl y delim it ed
by a unique combination of characters and by geographic range. The relevance of t his data to systematics of
the taxon is discussed.

255     Latch & Rhodes
Isolation and characterization of polymorphic microsatellite loci in the Wild Turkey. EMILY K. LATCH* and
OLIN E. RHODES, Jr., Dept. For. & Nat. Res., Purdue Univ ., West Lafay ett e, IN.
          The Wild Turkey was effectively eliminated from most of its original range by the early 1900s.
Extensive efforts to bring this species back from the brink of extinction have primarily involved translocations.
Many different strategies have been used in these translocations, despite little information regarding the relative
eff ectiv eness of each. Sinc e the ult imate goal of translo cation is to establish self -sustaini ng popul ations,
indicators of population health are needed to measure success. Populations that retain large amounts of
genet ic d iversity ar e thou ght to hav e a bet ter chanc e of sur vival over a gi ven peri od o f time; h ence,
characterization of genetic variation in reestablished Wild Turkey populations provides a measure of
translocation success. In this preliminary research, polymorphic microsatellite loci were developed for t he Wild
Turkey. I scanned approximately 800 sequences published for domestic t urkey and found 137 w ith simple
microsatellite repeats. I designed primers for 107 of t hese sequences, and screened them for polym orphism in
the Eastern subspecies of the Wild Turkey. Of t he 75 loci that amplified in the Wild Turkey, 16 w ere
polymorphic in this subspecies. These loci w ere combined w ith 6 poly morphic loci identified in a previous study
to cont rast the genetic variability w ithin and among 3 translocated populations of Eastern Wild Turkey in
Ind iana, each o f w hic h or igi nat ed f rom a dif ferent source. A ddi tion ally , a sample of dom est ic t urk eys w as
analyzed, so comparisons between wild and domestic populations are made.

256     Erdman
A phylogenetic analysis of morphological traits in neotropical frugivorous birds. JOSHUA A. ERDMAN, Dept.
Zool., Univ. W isconsin, M adison, WI.
        Moermond & Denslow (1985, Ornithol. Monogr. 3 6: 86 5-89 7) foun d rel ation ships betw een
morphological flight characteristics (including wing characteristics such as wing area and aspect ratio, flight and
leg muscle mass, and tail area) and feeding modes (aerial versus perching) in neotropical frugivorous birds and
suggested that different aerial modes may also show dif ferences. I have reanalyzed these data to see if the
phylogenetic relationships betw een these birds have any influence upon these correlations. I used Monte Carlo
simulations of evolution of t hese traits across a phylogeny to generate phylogenetically correct null distribut ions
for A NCOVA analy sis. M y r esults show that that the di fferenc es in these m orp hol ogi cal t rait s bet w een modes
previously reported are no different than expected through random evolution of these traits. Type of migratory
behavior and type of aerial feeding mode also failed to be any diff erent than random. My study illustrates the
importance of taking phylogeny into account during st atistical comparisons and discusses the usefulness of t his
method of analysis. It also suggests that t he requirements for flight in birds cannot be modified to any large
degree for use by species with different behaviors.

257     Demos & Tello
Patt erns of advertisement call evolution in Trogon (A ves: T rogonidae). TERRENCE C. DEMOS an d J OSE G.
TELLO, Dept . Biol. Sci., Univ. Illino is at Chic ago an d Div. Bird s, Field M us. , Ch icag o, IL.
         The genus Trogon includes 17 species distributed from t he extreme s. U.S. to n. Argentina, reaching its
greatest abun dance in th e humid region s of Cent ral Am erica. In mo st species, the advert isement call c onsist s
of loud, multiple repetitions of the same note(s) at fairly regular intervals, with relatively little change in pitch.
Phylogenetic analysis w as used to investigate the evolution of advertisement calls w ithin t he genus. A matrix
of str uctural characters of Trogon species was derived from the analysis of tape-recorded vocalizations.
Similarity among structural call elements was evaluated using tw o criteria: (1) acoustic struct ure of the element;
and (2) posit ion of the elem ent w it hin the cal l. These char act ers w ere op timized o nt o a molec ular based
phylogeny (Espinoza de los Monteros 1998, Auk 115 : 93 7-95 4) to f ind the most parsimonious hypothesis for
the evolution of each character. Patterns of advertisement call evolution w ithin t he genus are discussed in
relation to t his phylogenetic framework.

258     Feinstein
Entire mitochondrial genomes of two species of terns in the same genus: Sterna hirundo and Sterna dougallii.
JULIE FEINSTEIN, Am. Mus. Nat. Hist ., New York, NY.
         The entire mit ochond rial genomes of 2 species in t he genus Sterna have been sequenced: the Common
tern, S. hirundo and the Roseate tern, S. dougallii. The genomes are, respect ively, 16 ,7 20 and 16 ,7 58 base
pairs long. The diff erence in length is due mainly to dif ferent numbers of a repeated motif, CAACAA A, at t he
3' end of the control region. The gene order in terns was found t o be the same as in Gallus gallus, and many
of the features described for G. gallus are present. A nov el start codon, ATC, not found in G. gallus, begins
the ND3 gene of both terns. Comparison of the entire sequence of t he 2 terns shows an overall divergence of
only 4. 3% . Ten of the 22 tRNAs are identical. The most variable gene, ND6, w as over 7% divergent. The
least variable genes, the rRNAs, w ere both 2% divergent. A n anomaly w as encountered in S. hirundo. Some
primer pairs amplify anomalous sequences that contain inserts and a junction w herein the control region
changes abru pt ly to cy tochro me b. Presumptive mitochondrial sequences w ere isolated by long PCR, and
comparison of the resulting products from S. dougallii, S. hirundo, and G. gallus. A PCR from ND5 to 12S
results in bands of canonical length f or the 3 species, which cont ain the traditional arrangement of this gene
region. Another band, of approximately 6 kb, containing the anomalous sequences, occurs in S. hirundo. The
presence of indistinguishably similar copies of mit ochondrial sequences at diff erent loci in the genome of terns
suggests a problem for the interpretation of molecular systematics studies in this genus.

259     Feldman & Krannitz
Bird-habitat relationships in the Garry oak (Quercus garryana) ecosystem of southeastern Vancouver Island,
British Columbia. RICHARD E. FELDMAN* , Cen. App. Cons. Biol. , Univ. British Columbia, Vancouver, BC; and
PAM G. KRANNITZ, Can. Wil d. Serv., Delt a, BC.
         The Gar ry oak ec osy st em of Brit ish Colu mbia is a remnant mosaic of meadow and w ood land that has
been fragmented by urbanization and degraded by conifer encroachment and exotic shrub and grass invasion.
We used data from Breeding Bird Survey and Christmas Bird Count records of Vict oria and Chemainus/Duncan,
British Columbia to make comparisons betw een population sizes of the 19 90' s and the 1970 ' s. The results
show t hat the majority of breeding birds are decreasing in population but w inter populations are increasing.
Furthermore, the decline is greater in Victoria, the more urbanized region. During the spring and summer of
200 0, w e conducted bird and vegetation surveys on 2 transects in 7 Garry oak patches and their adjacent
habitat (urban or forest) in or near the city of Victoria. A Canonical Correspondence Analysis and MANOVA
indicat ed t hat sever al spec ies w ere associat ed w it h each habitat t ype. W it hin Garry oak p atc hes, a CCA
showed t hat conifer encroachment and Scot' s broom cover had the greatest influence on composition of the
bir d co mmuni ty. Our resu lt s suggest that Garry oak f ragm ent s w it hin the ur ban c ont ext have low er v alue t o
bird communities than fragments adjacent t o large patches of conifer forest; how ever, encroachment of
conifers into adjacent Garry oak patches will have to be managed to maintain habitat values important to bird
species choosing oak dominat ed habitat s.

260     Spellman & Klicka
Does morphol ogy accu rately t rack phy logeny in emberizid sparrow s? GARTH M. SPELLMAN and J OHN
KLICKA, Marjorie Barrick Mus. Nat. Hist., Univ. Nevada, Las Vegas, NV.
           Morphological characters have been essesntial tools in traditional taxonomy and systematics. However,
it is oft en the case that these characters demonstrate high degrees of homoplasy. We evaluate the utilit y of 3 6
morph olo gic al ch aracters in t rackin g t he ph ylogen y of emberizi d sparrow s. We construc ted a mo lecu lar
phy log eny of full cy tochro me b sequences from 47 emberizid sparrow species. The morphological characters
w ere mapped onto our best supported molecular phylogeny. Mapping the morphological characters onto t he
molecular ph ylogeny suggests t hat t hese characters of emberizid sparrow s show a high degree of ho moplasy
and may be misleading when used as taxonomic and systematic tools.

261     Reichart, Nastase, Deppe & W estneat
Molecular analysis of alternative mating strategies in Cliff Swallows. LETITIA M. REICHART* , ANTHONY J.
NASTASE, Dep t. Bio l. , Indi ana Un iv . of Pennsylv ania, Indi ana, PA; JILL L. DEPPE, Dept . Biol. , Univ . Cal if orn ia,
Rivers ide, CA; and DAVID F. WESTNEAT, Center Ecol., Evol. & Behav., Univ. Kent ucky, Lexington, KY.
         Patt erns of maternity and paternity w ere studied in a Cliff Swallow population located in nw.
Pennsylvania. In 1998 , birds w ere banded and blood samples w ere taken from t he brachial vein for parentage
analysis. Microsatellite analysis of 1 5 Cliff Swallow families was used to estimate the frequency of
intraspecific brood parasitism (ISBP) and extra-pair fertilizations (EPF). Parental exclusions, suggesting off spring
w ere t he res ult of an ISBP event , o ccurr ed in 23 .3 % of the t ot al nu mber of of fspr ing test ed. Our resu lt s
support ISBP estimates obtained from allozyme data on Cliff Swallow f amilies located in sw. Nebraska. We
found 1 c ase w here only the male was excluded, which w ould most likely represent a case of ISBP in which the
parasitic f emale had alleles matching the social mother. Several other microsatellite loci are being tested to
increase the probability of exclusion and decrease the number of undetected ISBP and EPF.

262     Mertz, Barrowclough, Groth, Cane & Gutierrez
Phy logeography of Blu e Grouse b ased on m tDNA co nt ro l r egion sequenc es. LISA A. MERTZ* , GEORGE F.
BARROWCLOUGH, JEFFREY G. GROTH, W. PARKER CANE, Am . M us. Nat . Hist ., New York, NY, and R. J.
GUTIERREZ, Dept. Wildl. , Univ . M innesot a, St. Paul, M N.
         We inv est igat ed t he ph ylogeo grap hy and speci es st atus of Blue Grouse (Dendragapus obscurus) using
mt DNA sequenci ng o f pop ulat ion samp les f rom all 3 of the major subspec ies grou ps. The D. o. obscurus and D.
o. fuliginosus groups are found t o be reciprocally monophyletic w ith high bootst rap support w ith t he exception
of a small sample from s.-central British Columbia. However, t here is no evidence of substantial or continuing
gene flow betw een these taxa. There is substantial differentiation betw een the southern (D. o. obscurus) and
northern (D. o. richardsonii) Rocky Mt n. groups, and evidence of reduced gene flow , but t hese taxa are not
reciproc ally mon ophyl etic. Gene flow appears to be subst antial t hrough out the range of the Pacific coast
(fuliginosus) populat ions.

263     Baumgarten, Kohlrausch & Miyaki
Genet ic var iabilit y i n booby species of the Brazil ian coast . M ELINA M. BA UM GARTEN* , A DRIA NA B.
KOHLRAUSCH and CRISTINA Y. MIYAKI, Depto. Biol., Univer. de Sáo Paulo, Brazil.
         Three species of boobies breed on the Brazilian coast: Sula l euco gast er (Brown), Sula dactylatra
(Masked) and Sula sula (Red-f oot ed). Thi s w ork w as done at 4 b reedi ng c olo nies alon g t he Brazi lian coas t: Sáo
Pedro and Sáo Paulo Archipelago- ASPSP, Fernando de Noronha Island, Rocas Reef and Moleques do Sul
Island. We est imated t he gen etic v ariab ili ty (GV) based on t he M ean Band Sharing Coef fici ent (BSC) bet w een
individuals from these 3 species using DNA fingerprinting. The GV found was very low compared to other bird
species. The BSC in Brown Booby for the minisatellites probes 33.6 and 33.15 were respectively, 0.726 ±
0.08 and 0.571 ± 0.13 f or Moleques do Sul, 0.688 ± 0.09 and 0.608 ± 0.07 f or Rocas Reef and 0.887 ± 0.04
and 0.81 4 ± 0 .05 for A SPSP. Even though t he GV w as low in this species, the populations are statistically
different. For the Masked Booby from Fernando de Noronha Island, the BSC was 0.384 ± 0.09, using 33.6
probe, and 0.30 9 ± 0 .11 , using 33. 15 probe. For the Red-footed Booby from Fernando de Noronha Island, the
BSC was 0.3 43 ± 0.1 2, using 33 .6 probe, and 0.1 8 ± 0 .09 , using 33. 15 probe. The BSC values found in
these booby species, specially for the Brown booby, are only found in species that have gone through a severe
bott leneck or in extremely endangered species. A bot tleneck event and recent population expansion could
explain these unexpected values. Grants: FAPESP, CNPq, CAPES, Brazilian Navy.

264     Porzecanski & Cracraft
Biogeographic area-relationships of the Neotropical arid areas of endemism based on raw distributions of birds.
ANA LUZ PORZECA NSK I* and J OEL L. CRACRAFT, Am . M us. Nat . Hist ., New York, NY.
          Many hypot heses have been proposed to explain the high levels of diversity and endemism found in the
Neotropical forest biota, and hypotheses of area relationships have been proposed for the forest areas of
endemism based on diff erent vertebrate groups. By comparison, the endemic biota of t he aridlands has seldom
been studied. In this study, biogeographic relationships among Neotropical avian areas of endemism were
investigated using a Hierarchical Parsimony Analysis of Endemicity and distributional information f rom a large
avian dataset (Stotz et al. 1996. Neotropical birds, Univ. Chicago Press) in order to: (a) generate a testable
hypothesis of area relationships, including the arid areas, from raw dist ributional information, and (b) examine
the degree of biogeographical noise introduced by dispersal and lack of diff erentiation. A cladistic analysis of
the dist ribut ion of 42 0 bir d tax a in 24 areas resulted in a single most parsimonio us tr ee wit h 2 m ain clades: (1)
Chaco+ Pampas+ Caatinga, and (2) Cerrado+ At lantic Forest+ Amazonia + others. In this t opology the 3 main
arid areas of Central Sout h Am erica, t he Caatin ga, Cerrado, and Chaco, do not for m a clade, w hich sug gests
that their avifaunas may have had independent histories. The results obtained are congruent in several aspects
w ith hypot heses derived from cladistic biogeographic analyses, suggesting that at t his large scale plesiomorphic
distrib ut ion s and lon g-dist ance d ispersal do n ot con trib ut e suf fici ent noi se as t o eras e the hi st ori cal s ign al
present in raw distributional information.

265      Chubb
Explori ng t he ph ylogen etic u tili ty of the nu clear gene ZENK: Relat ion ships w it hin passerin es and apod if orm s as
case studies. ALISON L. CHUBB, Mus. Vert. Zo ol. , Un iv. Calif orn ia, Berkeley , CA .
         In o rder to explo re t he ph ylogen etic u tili ty of the nu clear marker ZENK, g ene t rees o f the major lin eages
of t he orders Passerif ormes and A podif ormes w ere constr uct ed separately using ZENK sequences, and th ese
trees were compared to existing hypot heses of relationships wit hin these groups based on other molecular and
morphologic data. 170 0 base pairs of 18 passerine species and 17 apodiform species were used as ingroups in
the tw o separate analyses. The ZENK gene proved to be useful at this deep level of phylogenetic inquiry f or
the follow ing reasons: 1) an absence of saturation of transitions and transversions, 2) no significant departure
from base composition homogeneity across taxa, 3) resolution of relationships among most lineages, w hich
mirro r exist ing phy logenetic hypot heses based on other dat a, and 4) hi gh boot strap sup port for many nod es.
Used al one o r po ssibly in c ombin ation w it h mtDNA , ZENK show s pr omise f or p rov idi ng a m uch sough t-af ter
nuclear marker f or st udyin g phyl ogenetic r elationshi ps among major lineages w ith in avian orders.
266     Joseph, Bermingham, W ilke & Alpers
Molecular systematics, historical biogeography and intraspecific phylogeography of Myiarchus
tyrant-flycatchers. LEO JOSEPH* , Acad. Nat . Sci. , Philadelp hi a, PA; ELDREDGE BERMINGHAM, Smithsonian
Trop. Res. Inst., Panama; THOMAS WILKE and DERYN ALPERS, Acad. Nat. Sci.
         Prompted by an interest in using molecular data to explore the evolution of migration in Swainson' s
Flycatcher Myiarchus sw ainsoni and the biogeography of West Indian species, we have developed a
mtDNA-based phylogeny of Myiarchus based on ATPase 8 and 6 sequences from 120 individuals representing
all but 3 of t he ca 22 currently recognized species. Tw o major radiations w ithin Myiarchus are apparent, one
almost entirely within South America and the other in the Caribbean-North America-Central America region.
The Brow n-c rest ed Fly cat cher M. t yrannulus, w hich is widespread throughout South, Central and North
America, is not part of t he endemic South American radiation and may have dispersed through its range there
recently. Geographical and phylogenetic structure of mt DNA in M. sw ainsoni is at variance w ith t he current
view t hat it comprises 4 or 5 subspecies and suggests that migration has independently evolved tw ice and
pro babl y at ver y dif ferent times in this parap hy let ic as semb lage. Dif ferin g deg rees o f div ergen ce w it hin several
spec ies w idespread across Sout h A merica s ugg est that Myiarchus has evolved its current distribution there
thro ugh temporal and spat ial l ayer ing of taxa.

267      Causey
Discovery of Kenyon's Shag on Attu Island, Alaska. DOUGLAS CAUSEY, Mus. Comp. Zool., Harvard Univ.,
Cambridge, MA.
          Three species of Cliff Shags are known from breeding sites around the periphery of the North Pacific:
Pelagic, Red-faced, and Kenyon's Shag (Stictocarbo pelagicus, S. urile, and S. kenyoni). The l ast spec ies w as
desc rib ed f rom skel etons and ar cheolo gic al mat erial obt ained from Amchit ka Island , b ut unt il n ow it s external
morphology and genetics have been unknow n. I report here the external morphology and selected gene
sequences of this species on the basis of 6 individuals in breeding condition collected on Att u Island, Alaska, in
May 1 99 8. The b ird s looked un lik e ot her k now n shags, but possessed all o f the di agno st ic o st eolo gic al
characters for Kenyon's Shag. This cryptic species differs from other shags by its small size; its lighter,
yellow -or ange b ill w it h a dar k culm en; in b reedi ng, a mand ibl e col ored bri ght scar let ; an d nu merous ot her
characters. Morphometric analysis revealed that Kenyon' s Shag diff ers complexly f rom other shags by
mult ivariat e shape and size.

268      Ruegg & Smith
Post glacial expansion and the evolution of migratory routes in the Swainson's Thrush, a Neotropical migrant.
KRISTEN C. RUEGG* and THOMAS B. SMITH, Cent er f or T rop ical Research , San Francisc o St ate Univ. , San
Franci sco, CA .
          The influence of Pleistocene vicariant events and glacial cycling on avian diversity and genetic diversity
is a su bjec t of ong oin g deb ate. A n equ ally impor tant , y et und erst udi ed area o f resear ch is t he ef fect of glac ial
events on cur rent pat terns of behav ior al di ver sit y. Here w e pro vide eviden ce t hat Pleist ocene po st glac ial
expansions influenced phylogeographic patterns and the evolution of migration routes in the Swainson' s
Thrush, a Neotropica migrant. The phylogeographic struct ures of the Swainson' s Thrush was determined using
100 0 base pairs of mt DNA control region sequence. Analysis of samples from 1 0 breeding populations from
across the breeding range revealed tw o genetically distinct clades, a western coastal clade (found w est of t he
coast and Cascade mountains from Juneau, Alaska, to Big Sur, California) and a continental clade (found
throughout the remainder of t he breeding range). Analyses of North A merican Banding data are consistent w ith
our mol ecular genetic f indings and i ndicat e that w estern coast al birds mig rate along t he pacific coast t o
w int ering sit es in Central A merica and Mex ico w hile w estern con tin ental bird s fly tho usands of kilomet ers east,
retracing what appears to be a historical expansion route, before flying t o their w intering grounds in South
America. While f urt her s tudi es are needed t o det ermine i f spec ies w it h similar m igr atory rou tes exhib it sim ilar
patterns of genetic struct ure, our data suggest that vicariant events were likely fundamental to the evolution of
migrat ory r out es in some Nort h Am erican passerines.
269     Staley, Blanco, Wildt, Dufty & Monfort
Noni nvasiv e fec al st eroid measures for assessing g onadal and adrenal f unc tion in r apt ors. A IRICA M .
STALEY* , Dept. Biol., Boise Stat e Univ., Boise, ID; JUAN M. BLANCO, Centro de Estudios d e Rapaces Ibericas,
Sevilleja de la Jara, Spain; DAVID E. WILDT, Conserv. & Res. Center, Natl. Zool. Park, Smithsonian Inst., Front
Royal, VA; ALFRED M. DUFTY, Jr., Dept. Biol., Boise State Univ., and STEVEN L. MONFORT, Conserv. & Res.
         We validated radioimmunoassays to permit noninvasive assessments of fecal steroid metabolites in 2
captive raptor species, Golden Eagles and Peregrine Falcons. Longitudinal steroid excretion patt erns of
testosterone (T), total estrogens (Et), progesterone (P), and corticosterone (B) w ere compared to major breeding
stages in the circannual reproductive lifecycle. Female hormone trends showed increased excretion in P, Et, B
during courtship and egg lay, and B also increased during incubation. In males, T and B excretion patterns
increased during courtship and nestling care. Male and female B excretion also increased after disturbances.
Addit ionally, w e validated noninvasive hormone measures through intramuscular administration of exogenous
hormones (adrenocorticotropic hormone (ACTH), T, E, and P) or saline to 6 birds of each species in a rotating
repeated measures design. Demonstration of a direct cause-and effect relationship betw een ACTH
administration (2 5IU/kg) and fecal B excretion. Peak B excretion typ ically occurred w ithin 1 6 hr of ACTH
admini st rat ion , b ut som e bir ds ex hib it ed bi phasic B ex cretion pat terns . So me sal ine-treat ed bi rds exc ret ed
increased B, which may document handling stress. Administration of unlabeled steroids (0.5-1.0 ug/bird T, E,
and P) was used to determine excretion lag times for corresponding steroid metabolites. The results w ere
variable, but t he excretion lag time w as approximately 16 hr post-steroid hormone administration. Overall,
these data confirm t he utility of noninvasive fecal steroid hormone monitoring for evaluating gonadal and
adrenal status in male and female golden eagles and peregrine falcons.

270     Benowitz-Fredericks & Kitaysky
Pot ent ial for synchronous f ledging among neighboring Common M urre chicks. MORGAN
BENOWITZ-FREDERICKS* and ALEXANDER KITAYSKY, Dept Zool., Univ. Washington, Seattle, WA.
          Comm on M urr es are co lonial seabird s t hat experience higher repr odu ct ive success i f they syn chr onize
their reproductive effort s w ith neighbors. How ever, eggs are often laid asynchronously. Parents of late-laid
eggs should benefit if t hey produce chicks capable of grow ing quickly and fledging synchronously wit h their
neighbors. We looked for diff erences in growt h between murre chicks from early-laid eggs and their neighbors
from late-laid eggs. All t he eggs from a single ledge of a murre colony on Gull Island, Alaska were taken,
artificially incubated and hand-reared. All chicks hatched wit hin 6 d, and were fed ad libitum for 10 d. We
foun d a signi fican t effect of hat ch dat e on g row th par amet ers. Chic ks t hat hat ched lat er at e more and grew
more quickly than chicks that hatched earlier. When the youngest chicks w ere 2 d old, there was a significant
positive relationship betw een age and mass of all chicks. However late-hatched chicks grew quickly enough
that 3 d later this relationship disappeared - older chicks were no significantly heavier than younger chicks. We
also exam ined hor monal c orr elat es of gro w th and hat ch dat e. W e con clu de t hat giv en a 6 d lag in l ay date,
parents can pr oduce chic ks capable of f ledging at the same t ime and mass as neighbors.

271     Lipar, Ketterson & Nolan
Yolk t est ost eron e inf luen ces t he dev elop ment of the hat ching muscl e in t he Red-w ing ed Blackbird . J OSEPH L.
LIPAR* , Sch. Biol. Sci., Washington State Univ., Pullman, WA; ELLEN D. KETTERSON and VAL NOLAN Jr.,
Dept. Biol., Indiana Univ., Bloomington, IN.
         Hatchi ng asy nch ron y in avian sp ecies o ften resu lt s in the f orm ati on o f a hat chi ng-ord er-dependent size
hier archy among n est lin gs. Because access t o par ent ally -delivered f ood is largel y depend ent on t he size-rel ated
competitive abilities of nestlings, the youngest siblings in a brood are often at a significant disadvantage. I
found t hat in the Red-winged Blackbird the concentration of testosterone in the yolks of eggs increases with
laying order. I investigated the relationship betw een yolk testost erone concentration and the development of the
hatching muscle (musculus complexus), a dorsally-located neck muscle involved in the dorsal flexion of the
head and neck during hatching and begging. I found t hat the mass of the complexus increased significantly w ith
laying ord er in bot h new ly hat ched and 3-d ol d nestlin gs and t hat yolk t est ost eron e con cent rat ion w as
positively correlated with complexus mass at both ages. In addition, direct injections of testosterone into yolks
resulted in an increase in complexus mass, while injections of f lutamide, a testosterone antagonist, resulted in a
decr ease in com plex us m ass. Neit her y olk test ost eron e con cent rat ion nor com plex us m ass dif fered b etw een
male and female nestlings. These results support the hypothesis that yolk testosterone has anabolic effects
upon the development of the complexus and suggest that maternally-derived yolk testosterone is deposited in
greater concentrations in later-laid eggs to mitigate the effects of hatching asynchrony on the competitive
abilities of last-hatched off spring. Current research is focused on understanding the funct ional importance of a
larger complexus during both hatching and begging.

272     Hayward & Wingfield
Maternal corticosterone implants change yolk hormone composition in Japanese Quail. LISA HAYWA RD* and
JOHN C. WINGFIELD, Dept. Zool., Univ. Washington, Seattle, WA.
         Maternal sex steroids are deposited in avian egg yolk to varying degrees that may correspond to
changes in offspring growt h rates, dominance and survival. Stress hormones such as glucocorticosteroids are
molecu larl y similar to and rog ens and also li pid solubl e, so may be sim ilar ly depo sit ed in to yolk . Fur therm ore,
glucocorticoid exposure during development has been shown t o permanently alter off spring phenotype in
rodents and fish. To determine whether we could experimentally manipulate corticosterone levels in yolk, I
increased maternal stress steroid levels in Japanese Quail (Coturnix japonica) by implanting laying females w ith
silastic tubes of corticost erone (cort) and measured the effects on yolk hormone composition. Cort implants
keep p lasm a cor t levels high for p rol ong ed per iod s, mimicking the ef fect s of chr oni c st ress. Si x days after
implantation cort levels were higher in the yolk of experimental birds than controls (w hich w ere given empty
implants). Our results suggest that experimental manipulation of yolk cort c oncentrations can be used to
invest igate t he eff ects of glucoc ort icoids on development and off spring ph enoty pe in birds.

273     Richardson & W ingfield
Cort ico st eron e supp ressi on in t he Grey -crow ned Rosy Finch . M AT THEW I. RICHARDSON* and J OHN C.
WINGFIELD, Dept. Zool., Univ. Washington, Seattle, WA.
          The Grey-crowned Rosy Finch is an alpine specialist w hose breeding environment is characterized by
low primary productivit y, large daily temperature fluctuations and unpredictable storms. A large portion of the
Rosy Finch diet is temporally and spatially patchy insect f allout. The arctic is also subjected to unpredictable
storms, has low primary productivit y, and has many other similarities to the alpine environment; how ever, it
diff ers by exhibiting more constant temperatures and a more predictable food supply during the breeding
season. Some arctic s peci alists show a suppres sion of cor tico st eron e release w hen ex posed t o an
unpredictable stressor during the breeding season. This suppression serves to enhance reproductive fit ness in
an extreme enviro nment . W e hypot hesized t hat bir ds breedi ng at hig h alt it ude w oul d also di splay similar
physiological mechanisms for successful breeding due to the similarities in the tw o environments. We
conduc ted a prelimi nary inv estigat ion of the st ress physiol ogy of the Grey-cr ow ned Rosy Finch. The st ress
resp onse of males is low est dur ing the w int er (1 6. 1 ± 5. 2 n g/ ml) and molt (18. 0 ± 2. 7 n g/ ml), int ermediat e
w hen they are feeding fledglings (21. 7 ± 3 .5 ng/ ml) and highest during pre-breeding (30.0 ± 4. 9 ng/ml) and
w hen females are incubating eggs (31.67 ± 3. 77 ng/ ml). Females also suppress their stress response during
pre-breeding (1 4. 9 ± 5. 6 ng/ ml) and molt (25 .6 ± 4 .9 ng/m l), but actually upregulat e their st ress response
during incubation (31.5 ± 4. 0 ng/ml) and w hile feeding fledglings (30.2 ± 5. 2 ng/ml). In cont rast, arctic
specialists tend to decreased their stress response during the breeding season. We are currently investigating
possible reasons for these differences.

274     Vezina & W illiams
The met aboli c co st of egg pr oduct ion i n European Star ling s. FRANCOIS VEZINA * and TONY D. W ILLIAMS,
Dept . Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
         The en erget ic c ost s of egg p rod uction are po orl y under st ood in b ird s. Most st udi es hav e been b ased
simply on the energy content of eggs or clutches and little is known about additional physiological costs, e.g.,
energy expenditure for nutrient t ransport and/or uptake, yolk precursor synthesis, or maintenance cost(s) of
repr odu ct ive org ans. We measur ed res ting met abol ic r ate (RMR) of free li ving f emale European Starli ngs over
the breeding season from t he pre-breeding stage (prior to the development of t he reproductive organs) to the
complet ion of the cl ut ch. RM R was af fect ed by foll icu lar d evelopm ent and w as f oun d t o be ab out 30 % hig her
at the 1 egg stage (point of maximal predicted energy investment) than at the pre-breeding stage. Plasma
levels of yolk precursors vitellogenin (VTG) and very low density lipoprotein (VLDL) w ere positively correlated to
RMR in egg producing females. Body composition analysis revealed a strong effect of lean dry oviduct mass on
RMR w hich together w ith lean dry heart mass and lean dry pectoral muscle mass explained 44% of the
variation in RMR. This suggests that phy siological adjustments associated w ith egg production, most ly the
development and energy consumption of the oviduct , are substantial and cause the elevation in RMR.

275      Love, Shutt, Bird & Silfies
Effects of di et ary PCB ex po sure on the adreno co rt ic ol respon se of capt iv e A meri can Kestrels. OLIVER P.
LOVE* , Avian Sci . & Conserv. Centre, McGill Uni v., St e-A nn e-d e-Bellevue, PQ; LAIRD J. SHUTT, Canadi an
Wildl . Ser ., Nat l. Wildl. Res. Centre, Hull, PQ; DAVID M. BIRD, Avian Sci. & Conserv. Centre, McGill Univ.; and
JOEL SILFIES, Dept. Biol. Sci. , Cornell Univ. , It haca, NY.
          An experi ment w as co ndu ct ed t o examin e the ef fect s of diet ary exposure t o PCBs on the adr enoc ort ical
response in American Kestrels. Nine captive male American kestrels previously exposed to elevated levels of a
mixture of PCB congeners (1248: 125 4:1 260 ; 1: 1:1 ) in their diet underwent a standardized capture, handling
and r est rain t pro tocol d esigned t o res ult in an inc rease in ci rculat ing cor tico st eron e. A sim ilar pro tocol h as been
applied to a w ide range of avian species and w as used here to evaluate the response of PCB-exposed and
control kestrels to a defined physical stressor. Both baseline and maximum levels of corticosterone were
signif icantly low er in PCB-exposed birds w hen c omp ared w it h co nt rol bir ds of the same age. I n addit ion, PCB
bir ds had signi fican tly low er cort ico st eron e levels as t ime inc reased dur ing the adr enoc ort ico l respon se w hen
compared w it h co nt rol bir ds, ind epend ent of bod y c ond it ion . These r esults suppo rt several r ecent st udi es
w hich report decreased levels of circulating corticosterone in PCB-exposed birds. The results presented here
provide the first evidence that exposure to environmentally relevant levels of PCBs can impair avian HPA axis
funct ion, possibly increasing the susceptibility of birds t o environmental stressors such as severe w eather and
pred atory and h uman di st urb ance.

276      Pravosudov, Kitaysky, Wingfield & Clayton
The eff ect of phot operiod and long-term unpredictable food supply on baseline levels of cort icosterone and on
the adrenocortical stress response in Mountain Chickadees. VLADIMIR V. PRAVOSUDOV* , Neuro., Physiol. &
Behav. , Un iv. Calif orn ia, Davi s, CA, ALEXANDER S. KITAYSKY, JOHN C. WINGFIELD, Dept. Zool., Univ.
Washington, Seattle, WA, and NICOLA S. CLAYTON, Dept. Exp. Psychol. , Univ. Cambridge, Cambridge, UK.
         Birds respond to environmental changes by modulating their levels of corticosterone, a stress hormone
produc ed by t he adrenal glands. Baseline levels of cor tic ostero ne and the magnit ude of adrenoc ort ical response
to acute stress are know n to vary seasonally. There are several factors that might trigger elevated levels of
baseline corticosterone observed during the w inter: unpredictable food, short days, and low ambient
temperatures. In tw o separate experiments, w e examined the eff ects of long-term limit ed and unpredictable
food supply and photoperiod on baseline and acute-stress induced levels of corticost erone in adult Mount ain
Chickadees. Long-term unpredictable food supply resulted in a significant increase in baseline levels of
cort icost erone, w hereas it had no effect on th e magnitude of adrenocortic al response to acute str ess. In
cont rast , p hot oper iod had n o signi fican t effect on b aseline lev els of plasma cort ico st eron e, b ut bir ds sw it ched
from short days t o lo ng d ays reached m aximum levels of st ress-induc ed co rt ico st eron e signif ican tly fast er t han
birds maint ained on short days. In bot h experiment s, f emales had signific antly stro nger adrenocort ical response
to acute stress than males. Our results suggest that food supply has a direct eff ect on baseline levels of
cort icost erone, w hereas photop eriod migh t be dir ectly invol ved in t he regulation of adrenoc ort ical str ess
resp onse.

277     Roauf, Smith, Brown, Wingfield & Brown
Cort icost erone level in Clif f Sw allow s is af fec ted by c olon y si ze and exp eriment al parasite r emov al. SA MRRAH
A. RAOUF* , Dept. Zool., Univ. Washington, Seattle, WA; LINDA C. SMITH, Richard Stock ton Coll., NJ; MARY
BOMBERGER BROWN, Dept. Biol . Sci ., Univ . Tu lsa, Tulsa, OK; JOHN C. WINGFIELD, Dept. Zool., Univ.
Washington; and CHARLES R. BROWN, Dept. Biol. Sci. , Univ . Tulsa.
          Cliff Sw allows nest in colonies ranging in size from 2 nests to over 30 00. At our study sit e along the
Platte River in sw. Nebraska, ectoparasitism by the swallow bug (Cimicidae: Oeciacus vicarius) is the primary
cause of nest f ailure and is a major cost of living. Fumigation of colonies significantly reduces parasite load on
both adults and nestlings. During the summer of 2 000 , w e sampled adults nesting in colonies that ranged in
size from 7 5 nests to 3 000 nests to measure corticost erone (Cort) and testosterone levels. We found that in
nonfumigated colonies adults had Cort levels of 1 2ng/mL during both t he egg-laying and nestling-rearing
stages. However, adults living in fumigated colonies had significantly low er overall Cort levels, and Cort levels
during the egg-laying stage (2.9 ng/m L) were lower than Cort levels during the nestling-rearing stage
(9.6 ng/mL). Among nonf umigated colonies adults had the largest amount of Cort in t he largest colonies but
they had the low est testosterone levels. How ever, there was no effect of colony size on hormone level for
adults l iving in f umigat ed colonies. These results sugg est t hat inc reased cortic ostero ne levels in large groups,
oft en interpreted to result f rom greater social stress, may instead be attribut able to heavier parasite loads in
tho se large groups.

279     Salvante
* Hormonal control of egg production: Effects of exogenous corticosterone on circulating yolk precursor levels
and reproductive effort. KATRINA G. SALVANTE, Dept . Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
           Egg formation involv es major changes in protein and lipid metabolism, including the up-regulation of
tot al li pop rot ein syn thesis and the pr odu ct ion of the t w o avian eg g-y olk prec urs ors , a speci fic y olk -t arget ed
very-low density lipoprotein (VLDL) and vitellogenin (VTG). Corticosterone (B) is know n to be involved in
regulating many aspects of lipid metabolism in non-reproductive animals, including increased fat deposition and
hypertriglycedemia. I therefore investigated w hether B is involved in regulating or modulating lipoprotein
pro duc tion dur ing yolk f orm ation , an d t he ef fect of B tr eatment on r epro duc tive out put (egg size, clu tch size,
and tim ing of laying), in Zebra Finches (Taeni opy gia g ut tata). Egg-lay ing females had sign if ican tly low er
stress-induced plasma B levels compared w ith non-breeders (8.8 v s. 18 .1 ng/ ml), although there was no
diff erence in basal B levels (3.2 vs. 3.3 ng/ml). Egg-laying females were treated with exogenous corticosterone
using silastic implants, and this elevated plasma B to high physiological levels (22. 8 ± 5 .8 ng/ ml). B treatment
decreased breeding propensity by almost 50% and delayed initiation of egg laying by an average of 7 days.
However, t he B-treated females that did initiate egg laying had comparable egg and clutch sizes to control
females. The effects of B on plasma levels of VTG and VLDL will also be described.

280      Cason, Carr & Meyer
p-Nonylphenol: Incorporation of an environmental estrogen into the eggs of Japanese Quail and effects on male
reproduction. DARLA A. CASON* , Dept. Biol., A ustin Coll., Sherman, TX; ANDREW CARR, Dept. Chem.,
Austin Coll.; and WAYNE E. MEYER, Dept. Biol., Aust in Coll.
        p-Nonylphenol (NP), an environmental estrogen, may cause reproductive abnormalities in wildlif e. The
incorporation of NP into the eggs of female Japanese Quail intubated weekly w ith 50 or 100 mg NP/kg body
w eight or corn oil vehicle was investigated using proton NMR of organic yolk material. Eff ects of embryonic
exposure to NP on cloacal protrusion area (CPA), copulatory behavior, and testes w eight w ere studied to
assess the effect of NP on male reproduction. Body mass was also recorded. No NP w as found in eggs of
exposed females above a detection limit of 3 -4 mM NP. NP or an estrogenic metabolite w as likely present in
the egg s below the det ect ion lim it , b ecaus e cop ulat ory behav ior w as signi fican tly dim ini shed in t reat ed males
(p< 0.0 5). No dosage eff ect w as observed. NP exposure did not alter CPA or t estes weight (p> 0.0 5). Control
birds were heavier than treated birds (p< 0.0 5). Diff erential sensitivity t o estrogens in the medial preoptic
nucleus of the hypothalamus and the peripheral reproductive organs may account for diminished copulatory
behavior in t he absence of CPA or t estes alterat ion. Embry onic exp osure to NP may cause a marked decrease
in reprodu ct ive success in J apanese Quail and pot entially in ot her avian species.

281     Marks, Capparella & Hackett
Relation ships among Neot ropical areas of endemi sm: Phylo geography o f t he Wedge-billed Wood creeper (Av es:
Dendrocolapt idae). BEN D. M ARKS, Dept . Biol. Sci. an d M us. Nat . Sc i. , Lo uisiana St ate Univ. , Bat on Rou ge,
LA, A NGELO P. CAPPARELLA, Dept. Biol. Sci. , Illinois St ate Univ., Normall, IL, and SHANNON J. HACKETT,
Field Museum, Chicago, IL.
          Studi es of the di st rib ut ion of Sout h A merican taxa have ident if ied sever al areas of endem ism that may
have contributed to t he historical diversification of the region. We constructed a phylogeny of Glyphorynchus
spirurus (Aves: Dendrocolaptidae) populations using mtDNA sequence data from portions of cy tochrome b,
NADH dehydrogenase subunit II (ND2), and complete NADH dehydrogenase subunit III (ND3). Using this
phylogeny w e evaluate five previous hypotheses of area-relationships, 2 based on phylogenetic st udies of
morphological characters in birds and 3 based on parsimony analysis of endemism in birds and primates.
Maximum likelihood and maximum parsimony analyses recovered 2 phylogenetic hypot heses that diff ered in the
placement of 4 of the areas. Wit hin each of the areas of endemism , t he 2 analyses suppor t t he same clades.
Neither of the phylogenetic hypotheses for Glyphorynchus exactly matches any of the 5 previous hypotheses of
area-r elat ion ships, alt hou gh am big uou s suppo rt exist s f or o ne of them. Fou r areas – Cent ral A merica,
Inambari, Napo and Rondô nia – are supported as composites with com ponent taxa having phylogenetic
affinit ies wit h > 1 area. Data reported here also indicate high levels of sequence divergence w ithin
Glyphorynchus. Genetic breaks wit hin Glyphorynchus are only partially congruent with subspecific taxonomy.
The r egio nal samplin g design used makes t his st udy the lar gest scal e genet ic as say of a w idespread Neot rop ical
avian taxon to date.

282     Meddle, Owen-Ashley, Hayward, Wada & Wingfield
Corticosterone has no effect on t erritorial aggressive behavior in the Snow Bunting: A n arctic-breeding
WINGFIELD, Dept. Zool., Univ. Washington, Seattle, WA.
           For Arctic breeding birds, there is conflict betw een the need to establish territories and pair bond and to
cope w it h extreme w eather an d f ood short ages in t he br ief Arctic b reedi ng season. Co rt ico st eron e (B), released
in response to harsh conditions, t riggers physiological and behavioral responses that allow birds to cope w ith
stressful situations. However, the stress response has an established inhibitory effect on reproductive function.
Present evidence suggests that arctic-breeding passerines have evolved a mechanism by w hich they are
desensitized t o th e behavioral eff ects of B. This insensit ivit y t o gluco cort icoids is t hought to i ncrease
repr odu ct ive suc cess by avoidi ng p arent al car e int errupt ion s. We determined w het her t he adr enoc ort ical
resp onse to st ress is m odu lat ed in the arc tic b reedi ng Sn ow Bunt ing in Barrow AK, 71 °N. Males implant ed
w it h silast ic c apsules f ill ed w it h B (n = 7) had sign if ican tly hig her p lasm a B level s (P < 0. 05 ) and pect oral
muscle wastage than control birds implanted wit h empty capsules (n = 7). The behavioral response to a
simulated territorial intrusion 2 to 4 d follow ing implantation w as not significantly diff erent between treatments
(P > 0. 05 for all behavioral parameters: number of songs, att acks, f light s, t ime spent w ith in 5 m and closest
approach). This insensitivity to behavioral effects of high glucocorticoid levels that w ould otherwise distract
the bi rd f rom repr odu ct ive and t errit ori al beh aviors may be adv ant ageou s f or s peci es w it h sev ere t emporal
const raints o n th eir breeding schedules.
283     Olson & Ricklefs
Ontog enic changes in met abolic capabilit ies and shivering t hermogenesis in t he Willow Ptarmigan. J OHN M.
OLSON* , Dep t. Bio l. , Villano va Un iv ., Villano va, PA, and ROBERT E. RICKLEFS, Dept . Biol., Un iv. Mi ssouri, St.
Louis, MO.
         The willow ptarmigan is a gallinaceous resident of arctic and subarctic regions. These birds are
rout inely subj ected t o cold, w et con dit ions, ev en during t he short breeding season. A s a precocial species,
new ly hat ched and young pt armigan chi cks f orag e on t heir ow n soon af ter hat ching . Thus, i t is reasonable t o
hypothesize that even the youngest chicks possess substantial capabilities for regulatory thermogenesis.
Met abolic rat e and (in some cases) body t emperature (T b) of individual chicks (1-d - 24-d old) and adults were
measured cont inuously during ex posure f irst to g radually decreasing ambi ent t emperature (T a ) follow ed by
gradually increasing Ta. Ta w as dec reased unt il t he bi rd s tart ed t o go hypot herm ic an d/ or t he metabol ic r ate
decreased persistent ly. Restin g metabol ic rat es at th ermoneut ral temperat ures (RMRs) were similar t o th ose
predicted for precocial species. Younger chicks maintained low er Tb at t herm oneutrality (~ 38 .5 ° - 40. 5° C)
than older chicks and adults (~41° - 42° C) . All birds measured were capable of regulatory thermogenesis.
Maximal metabolic rate of the youngest chicks ranged betw een ~2- and 3.5-fold higher than RMR.
Surprisingly, this factor ial scope did not change significantly through day 24 , and w as only slightly higher in
adults. How ever, the range of T a toler ated in creased duri ng d evelopm ent , o w ing pri marily to lo w er
conductance and lower critical temperatures in older birds. Often, increases in metabolic rate during cooling,
especially in younger (1-d - 16-d) chicks, demonstrated bouts of higher metabolic activity; independent
measurements of electomyographical activity in selected skeletal muscles substantiated this bursting patt ern of
therm ogen esis. These t herm oreg ulat ory capab ili ties allow pt armigan chi cks t o f orag e ind epend ent ly from an
early age.

284     Williams & Miller
Ability t o lay supranormal clutches, in response to egg removal, is resource-dependent in the Zebra Finch
(Taeni opy gia g ut tata). TONY D. WILLIAMS and MELINDA A. MILLER, Dept. Biol. Sci., Simon Fraser Univ.,
Burnaby, BC.
          The ab ili ty of bir ds t o lay suprano rmal cl ut ches in resp onse to egg remov al du rin g lay ing is often ci ted
as eviden ce in suppor t of a low phy siolog ical or en erget ic c ost of egg p rod uction . How ever , r ecent st udi es
(e.g., Nager et al. 2000, Ecology 81: 133 9-135 0) have shown t hat although birds can readily produce
additional eggs, and these eggs may be of normal size, there can be diff erences in size independent egg
composition of later-laid eggs which have important (negative) consequences for offspring viability. We
investigated the physiological changes associated with laying of supranormal clutches by removing eggs from
laying female zebra finches and measuring plasma levels of the yolk precursors: yolk-targeted very-low density
lipoprotein (VLDL) and vitellogenin (VTG). Females w ere maintained on tw o diets: 1) a high quality, egg-food
supplemented diet (n = 10), and 2) a low-quality seed diet (n = 10). Eggs were removed up to the 16t h egg
and females w ere then allow ed to lay a normal clutch, and egg composition w as determined for eggs 3, 10 and
16. Response to egg removal w as highly diet-dependent, but also varied markedly among individuals on each
diet. Egg-diet females laid on average 12.8 extra eggs (range 5 - 17), but seed-diet females laid only 4.1
additional eggs (range 1 - 12 ). This suggests that individual females are not equally able to support t he costs of
cont inuous egg f ormat ion, and t hat t his variat ion is resour ce-dependent. There w as no evidence fo r a decrease
in egg size for later-laid eggs, even in birds laying 20+ eggs (4 times the normal clutch size). The effects of
egg removal on egg composition and plasma yolk precursor levels will also be described.

285     Lovell
Song variation in a population of Alder Flycatchers in southwestern Alberta. SCOTT F. LOVELL, Dept. Biol.
Sci ., Uni v. Calgary , Calgary , A B.
         Song variation has been studied extensively over the past 50 yr, but almost entirely in oscine
passerines. It is perhaps surprising that few studies have investigated song variation in suboscine passerines.
Learning is an essential component of song development in most, if not all oscines, w hile there is no evidence
for song learning in suboscines. This raises the possibility t hat the pattern and amount of individual and
geographic variation may differ betw een these groups as w ell. I examined individual variation in the " fee-bee-o"
song in a population of Alder Flycatchers in sw . Alberta. Songs of individual males were recorded during the
breed ing season in 19 96 . I measu red t emporal and f requ ency var iabl es of songs using SIGNAL bio acou st ical
analysis softw are and conducted multivariate statistical analyses (PCA and DFA) to characterize variation w ithin
and among individuals. There w as little variation among the songs of an individual male during single recording
sessions and across recordings made over the breeding season. However, there w as sufficient variation among
individual males to identify them statistically and, potentially, to permit individual discrimination by the birds.
Future ex peri ments w ill test neig hbo r-strang er di scr iminat ion and i ndi vidual recogn it ion in a p opu lat ion of Alder
Flycatch ers.

286     Keller & Yahner
Avian use of fores t edges dur ing spr ing and f all m igr ation . GREGORY S. KELLER* , Int ercoll ege Grad . Degree
Pro g. in Ecolo gy , Pennsylv ania St at e Un iv ., Univ ersity Park, PA; and RICHARD H. YAHNER, School Forest Res.,
Pennsylvania State Univ.
         We compared habi tat use by m igr atory bir ds am ong dif ferent typ es of edges in a l andscape d ominat ed
by agriculture (Gett ysburg National Military Park and Eisenhower National Historic Site) and in a landscape
dominated by residential and urban development (Valley Forge National Historical Park) in 2000 - 200 1. Our
objectives were to determine the effects of f orest-edge type (i.e., nort h- versus south-facing edges and full-sun
versus shaded edges) and distance from edge (i.e., edge interface versus interior forest) on habitat use by birds
and to d etermine t emporal changes in edge use th rougho ut sp ring and f all migrat ion. Species ric hness,
particularly long-distance migrants, was significantly affected by distance from edge during spring migration.
During fall migration, migrants did not respond to edge type or distance from edge; however, migrants were
significantly inf luenced by vegetation measures. Birds appear to respond more to site characteristics during fall
migration versus landscape characteristics during spring migration.

287     Tautin, Doherty & M etras
Development of a bird banding recapture database. JOHN TAUTIN* , USGS Bird Banding Lab., Laurel, MD;
PAUL F. DOHERTY, Jr., USGS Georgia Coop. Fish & Wildl. Res. Unit, A thens, GA; and LUCIE METRAS,
Canadian Wi ldl. Ser., Bird Bandin g Of fi ce, Hu ll, PQ.
         Recaptures (and resightings) constitute the vast majority of post-release data from banded or
otherw ise marked nongame birds. A pow erful suite of cont emporary analytical models is available for using
recapture data to estimate population size, survival rates and other parameters, and many banders collect
recapture data for t heir project specific needs. However, despite widely recognized, broader programmatic
needs for m ore an d bet ter dat a, b ander s' recapt ure d ata are no t cent rall y r eposit ed and made availabl e
for use by others. To address this need, the US Bird Banding Laboratory, the Canadian Bird Banding Office
and the Georgia Cooperative Fish and Wildlife Research Unit are developing a bird banding recapture
database. In this poster w e discuss the critical steps in developing the database, including: determining
exactly w hich recapture data should be included; developing a standard record format and struct ure for
the database; developing electronic means for collecting, vett ing and disseminating the data; and
most importantly, developing metadata descriptions and individual data set profiles to facilitate the user's
selection of appr opr iat e analy tical models. W e pro vide examples of ind ividu al dat a set s t o be i nclud ed
in t he dat abase, and w e assess the f easibili ty of develop ing a prescribed pro gram for o bt aini ng r ecapt ure d ata
from banders who do not presently collect t hem. It is expected that the recapture database eventually will
cont ain million s records made available public ly f or a variet y of avian research and management pur poses.

288     Dodenhoff
Observations of helping among Downy Woodpeckers. DANIELLE J. DODENHOFF, Dept. Evol. Ecol. & Organ.
Biol., Ohio Stat e Univ., Columbus, OH.
        I observed, at 4 of 13 cavit ies, a female Downy w oodpecker helping a monogamous pair during the
199 8 - 20 00 breeding seasons, Kraus Woods, Delaware Co., OH. Assistance in duration of incubation and the
int erv al of food deli ver y v aried among f emale h elper s. Mont ior ing of signal u se indic ated hel pers had
significantly low er rates of ratt le and pik calls versus mated females. In addition, helper females did not drum
w hil e assis ting monog amous pairs. A n anal ysis of call s exchanged bet w een hel pers and m ated pai rs i ndi cat ed
that previously described calls were used in these interactions; novel calls were not observed. I used multilocus
minisatellite DNA f ingerprinting to t est for relatedness of individuals wit hin groups. The average proportion of
band s shared b etw een f emales and h elper s w as 0.2 2, w hil e the average p rop ort ion of band s shared b etw een
males and helpers was 0.4 1. These proportions indicated that helpers w ere not first order relatives. The
proportion of bands shared betw een mates, and betw een males and non-mated, non-helper females was 0.15.
Since the mated females were the previous years territorial occupant, and males were not observed the
previou s year it i s unknow n w hy helpers shared mo re bands w ith the males.

289      Mallett & Pepperberg
Identifying birds f rom their songs using signal processing and adaptive reasoning. JACKY MALLETT* and
IRENE PEPPERBERG, Media Lab., Massachusetts Inst. Tech., Cambridge, MA.
         The species distinctive and highly stereotyped nature of most bird calls makes bird song an obvious
candidate for comput ational analysis. Using techniques from music recognition and auditory analysis, this
project has built a database of bird call characteristics based on frequency and magnitude variations over time.
These characteristics are then used to identify bird songs according to their species and individual song types.
A set of tools has been developed to extract and display this data from bird song recordings, allowing t he
dat abase t o be u sed as an in form ation tool as w ell as for r ecog nit ion pur poses. Work is prog ressing on r eal
time recognition, w hich w ould allow f ield work w ith artif icial responses to be automated.

290      Stark
Is there a phylogenic or phenotypic component to woodpecker drums? ROBERT D. STARK. Dept. Evol., Ecol.
& Or gan. Biol. , Ohio St ate Univ. , Columb us, OH.
         There has been little research conducted on woodpecker drumming, even less concerning possible
phylogenic or phenotypic influences on this signal across species. Previous research noted that phenotypically
similar species have divergent drums (i.e., Nuttall' s versus Ladder-backed w oodpeckers, Hairy versus Downy
w oodpeckers), how ever, it w as unknown w hether this observed patt ern applied across species. I construct ed a
bin ary phen ogr am using 2 4 m orp hol ogi cal c harac teristics for al l 2 1 n earctic s peci es, and u sed a hierar chical
cluster analysis to construct a dendrogram of divergence betw een species. Using the means/species/variable
for drumming, I used Mantel tests to determine there was a significant correlation betw een the drum cadence
(strikes/s), duration, and inter-strike interval versus nearctic phylogenic relationships (Short 1982, Woodpeckers
of the w orld). Given t his, I d econst ruct ed the phy logeny by sympat ric species f or 6 major nearct ic habit ats,
and tested (using Mantel tests) w hether drums were significantly correlated by either phylogenic or phenotypic
divergences in sympatry. Results indicated that desert species (and boreal species for duration) w ere
significantly correlated to divergence in drum duration and cadence versus phenology. Yet, southern pine
fores t spec ies diverge i n dr um dur ation ver sus phy log eny, n ot phen olo gy . W est ern c oni ferou s f ores t spec ies
w ere significantly correlated to bot h. Western oak woodlands and eastern deciduous forest species w ere not
correlat ed. Furt her t est ing ind icat ed t hat overall phy log enic signif ican ce v ersu s dr umming w as due to
diff erences at the genus, not species, level.

291      Green & Leberg
The eff ect of w hite plumage and group size on flocking decisions in Ciconiiformes. M. CLAY GREEN* and
PAUL L. LEBERG, Dept . Biol. , Univ . Lo uisiana, Lafayet te, LA .
         Cico nii form es of ten f orm larg e mix ed-speci es f loc ks f or b ot h f orag ing and n est ing . W hit e plu mage may
funct ion as a signal for the formation of w ading bird aggregations. Under certain solar conditions, w hite
plu mage may be more o r less cr yptic t hereb y af fect ing the nec essity f or u tili zing floc ks. We hyp ot hesized t hat
based on the visibility of w hite plumage, birds will land more often at flocks of w hite birds than flocks of dark
birds. Additionally, differences in flock size and solar conditions may further affect selection of whit e or dark
floc ks. To t est these hyp ot heses, w e ut ili zed 5 adjac ent experim ent al po nds (0.4 0 h a). Each po nd w as
randomly assigned either dark or whit e heron decoys. Ponds received either a single decoy or group of 5
decoys; an empty pond served as the control. Each observation ceased aft er the first bird landed at one of t he
pon ds, at w hic h t ime the t reat ments w ere reallo cat ed among pon ds. We used c ategor ical modeli ng f or d ata
analysis. Wading birds landed at ponds w ith 5 decoys (84 %) more oft en than at ponds wit h 1 (14 %) or no
deco ys (2%) (P = 0. 00 5). There w as a sig nif ican t int eraction of gro up size and deco y c olo r (P = 0. 00 2), as
w ading birds c hose ponds w ith 5 w hit e decoys (7 0% ) more of ten t han ponds w ith 5 dark d ecoys (1 4% ), singl e
w hite decoy (5%), or single dark decoy (9%). White plumage appears to play an important role in attracting
wading birds to an area, but this affect is influenced by group size.

292     Neudorf, McGlothlin, Casto, Nolan & Ketterson
The effect of elevated testosterone on female mate choice in Dark-eyed Juncos. DIANE L. NEUDORF* , Dept.
Biol. Sci., Sam Houston State Univ. , Huntsv ille, TX; JOEL W. McGLOTHLIN, Dept. Biol., Vanderbilt Univ.,
Nashville, TN; JOSEPH M. CASTO, VAL NOLAN, Jr., and ELLEN D. KETTERSON, Dept. Biol., Indiana Univ.,
Bloomingt on, IN.
            Previ ous st udi es hav e show n t hat experim ent ally elevated t est ost eron e (T) c auses male Dark-eyed
Juncos to be more attractive to f emales, to occupy larger home ranges, and to achieve more extra-pair
fert ili zation s (EPFs). Based on at tractiveness and EPF frequen cy , elevat ed male T ough t to be f avored b y sexual
selection unless some counteracting force constrains its evolution. We considered the possibility that if such
selection proceeded and if T levels were correlated in the sexes, then elevation of the hormone in females might
be a constraint on evolution in males. To test this possibility w e examined the eff ects of experimentally
elevated male and female T on mate choice by female juncos. We increased T levels in both sexes and
performed trials in which cont rol (C-) or testosterone-implanted (T-) females w ere presented wit h dyads
consisting of a control (C-) male and a testosterone-implanted (T-) male matched for age, size, and plumage
coloration. We measured the behavior of males and females and time spent by females with T- and C-males.
We hypothesized that t he masculinizing eff ects of t estosterone might lead females to be less discriminating,
nullifying t he greater attract iveness of T-males and thus constraining the evolution of high T in males. Our
predict ion w as support ed: T-f emales spent as much t ime in mat e choice as C-females, but , unli ke C-females,
they spent nearly equal amounts of time w ith T- and C-males. To our knowledge, this is the first study of a wild
passerin e spec ies t o show that a hor monal c orr elat ion bet w een t he sex es co uld con st rain sexuall y select ed
trait s in males.

293     Whittaker, Thusius, W hittingham & Dunn
Male parental care and paternity in a double-brooded species wit h brood div ision. KA RA A. W HITTAKER* ,
KEVIN J. THUSIUS, LINDA A. WHITTINGHAM and PETER O. DUNN, Dept. Biol. Sci., Univ. Wisconsin,
Milwaukee, WI.
        The r elat ion ship bet w een male par ent al car e and p aterni ty has been inv est igat ed in a num ber o f avian
species, but in many cases the influences of confounding factors, such as variation in male and territory
quality, w ere not addressed. These sources of variation can be controlled for by making wit hin-male
comparisons betw een successive broods or w ithin-brood comparisons betw een groups of fledglings in a
divided brood. We studied the relationship betw een male parental care and paternity in t he Common
Yellowt hroat and proposed three hypotheses: first, males in double-brooded pairs should provide relatively
more parental care to broods in which t hey have higher paternity; second, after fledging and brood division,
males should provide more care to related offspring; and finally, among all broods in the population, paternity
should be related positively to male parental care. Brood division occurred in the many of broods st udied;
how ever, bro ods w ere not div ided accordin g to fledgl ing size or patern ity . Furt hermore, w ith in div ided broods,
males fed w ithin-pair and extra-pair fledglings at similar rates. For sequential broods of the same pair, male
feeding rates w ere not associated wit h diff erences in paternity betw een broods. Among all broods in the
population, males did not provide relatively less care to broods containing unrelated young. The lack of a
relationship betw een male parental care and paternity suggests t hat either males cannot assess their paternity
or t he cost s of r educing male parent al care outw eigh th e benefit s.

294      Budden & Wright
Less is more? The effect of f ood delivery ratio on begging performance and digestive efficiency in Southern
Grey Shrikes. AMBER E BUDDEN* and JONATHAN WRIGHT, Sch. Biol. Sci. , Univ. Wales, Bangor, UK.
           One of the most critical assumptions of models of nestling begging is that the intensity of such begging
is related to t he (nutritional) requirements of t he offspring. Indeed, numerous studies have demonstrated a
positive relationship betw een nestling begging intensity and nestling need. However, rather than purely
ref lect ing dif ferenc es in " short -t erm" or " lon g-t erm" need, nest lin g beg gin g may r eflect phy siolog ical resp onses
to food already present in the digestive system. In this way, begging may also reflect levels of digestive
efficiency, and therefore represent " need" in a more intermediate physiological sense. The aim of this
experiment w as to test the follow ing predictions. First, t hat the nature of f ood intake influences the efficiency
of digestion, w hich may be mediated by the passage rate of digesta, and second, that begging intensity of
nestlings experiencing the same overall rate of food delivery w ill differ according size of prey items received and
the interval betw een their delivery. The results showed that nestlings hand-fed small prey frequently had more
protein present in t heir faeces than nestlings provisioned the same tot al volume of food in large infrequent
packages. These diff erences in digestive eff iciency w ere reflected in nestlings fed small prey demonstrating a
higher begging intensity t han their sibling aft er a period of f ood deprivation. Such results suggest that subt le
dif ferences in t he feeding hist ory o f an ind ividu al can infl uence its " need" , and t he potent ial for nestlin gs to
adjust their digestion must be considered in relation to t he competitive nest environment and the evolution of
honest signals of need.

295      Peluc, Ghalambor & Martin
Nestling prov isioning rat es under the immediate risk of nest predation: Why less care is bett er? SUSANA I.
PELUC* , CA MERON K . GHALA MBOR, Dept . Bio l., Un iv. Calif orn ia, Riverside, CA, and THOM AS E. MA RTIN,
Mont ana Coop. Wildl. Res. Unit, Univ. M ontana, Missoula, M T.
          Variation in parental care is thought to result f rom diff erences in the fitness costs and benefit s
associated w ith providing care. In birds wit h dependent young in the nest, parents must w eigh the benefits of
pro vidin g f ood to t heir of fspr ing again st the co st s of attracting the at tent ion of nest pred ators . Thus, w hen
faced wit h an immediate risk of nest predation, parents should temporarily decrease their nestling provisioning
rat e to red uce the pr obab ili ty at tracting attent ion to t he nest. Here, w e pres ent a con cept ual m odel that
pred ict s how spec ies t hat dif fer in their ris k of nest pred ation should alt er t heir nest lin g f eedin g rat es w hen
faced wit h an immediate risk of predation. We then tested this model experimentally by increasing the
immediate risk of nest predation in 5 phylogenetically paired species occurring in Arizona and Argentina.
Vocalizations of a common nest predator were broadcast near the nest to simulate the presence of predator.
We f oun d st ron g suppo rt for p redi ct ion s f rom the model ; 1 ) all spec ies reduc ed t heir nest lin g pr ov isioni ng r ate
in response to t he experiment, and 2) variation among species in response to the experiment w as highly
correlat ed w it h di fferenc es in the ri sk o f nest pred ation across spec ies. These resu lt s suggest that dif ferenc es
in vul nerabilit y t o nest pr edation can accu rately p redict the behavio ral response of parents t o a potent ial nest

296      Diaz, Torres & M acias
Relativ e parental invest ment o f m ales and females in the V ermilion Flyc atcher. MA RIANA DIA Z* , ROXANA
TORRES and CONSTANTINO MACIAS, Dept. Ecologia Evolutiv a, Inst. Ecologia, UNAM , Mexico.
         Parent al in vestment may dif fer accordi ng t o social and i ndi vidual con dit ion s. In speci es w it h bi paren tal
care, the amount and quality of care a parent can provide is an important trait during mate selection because it
may be related to reproductive success of the pair. The Vermilion flycatcher is a monogamous sexually
dimorphic passerine wit h biparental care. Males spend energy and time making costly courtship displays. We
evaluated (1) relative contribution of males and females during incubation and provisioning of the young and
(2) w hether variation in parental investment is related to phenotypic t raits of males (duration of courtship
display s, size and mass) an d f emales (size and m ass). We f oun d t hat males and f emales play dif ferent rol es
during parental care. Females built nests and incubated eggs, while both parents fed young. During
incubation, both parents w ere involved in nest defense and in 11 pairs males fed females. Before fledging,
feeding rate by females was 30% higher than males', however the differences were not significant (t = 1.43,
p = 0.17, n = 14); after fledging, female's feeding rates were 78% higher than males (t = 2.31, p = 0.06, n
= 7; pow er of test: 0 .41 ). Reproductive success increased w ith t he feeding rate of both parents and
decreased wi th l aying dat e. Heavier (but not larger) males had greater f eeding rates and reprodu ct ive success.
In cont rast, light er and smaller-beaked females had higher f eeding rates and reprodu ct ive success.

297     DuVal
Cooperative display behavior in the Lance-tailed Manakin. EMILY H. DuVAL, Dept. Integrative Biol., Univ.
Calif orn ia Berkel ey, Berkeley , CA .
          Manakins (Family Pipridae) are small neotropical passerines known for their complex courtship behavior.
Manakins in the genus Chiroxiphia cou rt mat es in a uni que c oop erat ive syst em. Tw o or more males display
together for 1 female, performing duet songs and complex coordinated dance displays. Only the dominant male
of t he pair appears to mate, and the adaptive advantage of cooperation for t he subordinate male remains
unclear. This study investigates cooperative display behavior in the Lance-tailed Manakin (Chiroxiphia
lanc eolat a), a previously unstudied member of the genus. Subordiante males may benefit by direct
reproduction, helping relatives, or inheriting t he pair' s display perch. Genetic t ests of relatedness and paternity
w ill be combined w ith behavioral observations at display sites to t est these possibilities. This poster presents
preliminary data f rom t he first tw o fi eld seasons.

298     Blinn
* Selection and use of brood-rearing habitat of Common Eider on Grand Manan Island, New Brunswick.
BRENDA BLINN, Atlantic Coop. Wildl. Ecol. Res. Netw ork, Uni v. New Brunsw ick, Frederict on, NB.
          The survival rate of eider ducklings in the Bay of Fundy has decreased from > 14% of t he adult
breeding population prior to 19 91 t o 4 - 8% betw een 199 5 and 1998 . Broods may travel several kilometers
from nesting islands to coastal duckling rearing areas w ithin 24 hr post-hatch. Eider ducklings and att ending
females feed on invertebrates commonly associated w ith rockw eed. Commercial boating (rockw eed harvesting,
fishing activit y), recreational boating, indust rial development (aquaculture cages) and public use of coasts could
increase risk of duckling predation by Great Black-backed Gulls and Bald Eagles. My objective is to quantify the
relative influence of various factors on habitat selection and use of eider broods. I will examine food
availabi lit y, ext ent of pred ation and h uman di st urb ance, pro ximity t o nesting islands and d egree o f exposure t o
wind and waves. I will look at spatial and temporal patterns of distribution and abundance of eider broods.
Rockweed harvested and unharvested sites will be visited from t he period of 22 May to 3 Aug 20 01. The
follow ing data will be recorded at 2 0 min int ervals for a period of 15 hr daily: number and behaviour of
ducklings and females, age-class of ducklings, number, species and age of predators, predation attempts or
successes, and presence of human activity. A brood's reaction following a disturbance will also be recorded.

299     Neville
Division of parental care in Western Sandpipers: Evidence for male biased effort. JULIETTE A. NEVILLE, Dept.
Biol. & Wildl. , Univ. Alaska Fairbanks, Fairbanks, AK.
        There is considerable interspecific variation in the division of parental care betw een the sexes w ithin t he
family Scolopacidae. I investigated whether male and female Western Sandpipers contribute equal amounts of
parental care during the breeding season, near Nome, Alaska (64° N) during 1998 and 19 99. I repeatedly
observed which parent w as present at the nest during the incubation period and w hich parent remained with
the brood during the brood care period. Incubation data were collected for 30 mated pairs, 23 of w hich
successfully hatched eggs. Males spent more time incubating than females (57% vs. 43%, P < 0.05).
Females incubated mostly at night (18:00 - 06:00 ADT), while males incubated mostly during the day (06:00 -
18: 00 A DT). The daily incubation pattern betw een the sexes did not change during the 3 w k incubation period
(P = 0.6 33). Brood care data were collected for 2 4 mated pairs. Males remained with broods longer than did
females (P < 0.0 1). Nests that hatched later in the season received significantly less bi-parental care during
the br ood care peri od (r = - 0.6 6, P < 0. 00 1). Result s indic ate that males pro vide signi fican tly more par ent al
care than females during both the incubation and brood care periods. These findings enhance the interpretation
of com parat ive relat ion ships am ong clo sely relat ed species to det ermine t he evolu tion ary history of paren tal
care in t he Scolo paci dae.

300     Vigallon
Predation on artificial ground nests by Steller's Jays. STACEY VIGALLON, Coll. For. Res., Univ. W ashington,
Seattle, WA.
         I examined predation by Steller's Jays on artificial ground nests in n. California. During 2 10-d trials, I
compared t he tim e to predat ion on an ini tial nest to t hat of a subsequent nest placed nearby (3 m aw ay) 10 d
later, w hich w as also then monitored for 10 d. I also examined the frequency of v isitation and the effect of
cano py and g rou nd c ov er on time to pr edat ion . In t he f irs t 10 -d t rial I f oun d eviden ce of avian pr edat ion at
74% of nests, and at 82 % during t he second 10-d t rial. There w as a significant diff erence in days to predation
bet w een the 2 tr ials. Of n ests th at w ere prey ed on, th e mean t ime t o predation f or t he f irst tr ial w as 4 d (SD
= 2.7 , n = 28), and 89% of t hese nests were visited more than once. The mean time to predation f or the
second trial was 2.4 d (SD = 2.2 , n = 31), and 90% of t hese nests were visited more than once. I found no
relationship between ground and canopy cover and time to predation. These data lend support to t he theory
that Steller' s Jays may use an area-restri ct ed search method to f ind nest s.

301     Krakauer
Structure and function of cooperative Wild Turkey coalitions. ALAN H. KRAKAUER, Mus. V ert. Zool. and Dept.
Int . Bio l., Un iv. Calif orn ia, Berkeley , CA .
           Male Wild Tur key s f orm breed ing coal it ion s in w hic h gr oup s of 2 t o 4 males display joi nt ly to f emales
and interact wit h other male groups. The functions of these coalitions, along w ith t he fitness benefit s for
subordinate group members are not know n. Subordinate males may gain through direct reproduction, indirect
reproduction through helping of closely related coalition members, or queuing for the dominant position. I w ill
test predictions derived from reproductive skew t heory in order to identify w hich of these three benefits are
important in st ructuring male display groups. I w ill observe a marked population of Wild Turkeys at the
Hastings Nat ural Hist ory Reservation , Carmel Valley, CA, and r elate behavioral dat a to mi crosat ellite DNA
characterization of male relatedness and reproductive success. Here I present preliminary data on the structure
and fun ct ion of male display gro ups. A ge struc tur e, stabil ity , int ra- and inter-gr oup behavior s prov ide the f irst
data f or underst anding t he evolut ion of cooperat ive male display in Wild Tu rkeys.

302     Moody, Wilhelm, Cameron, W alsh & Storey
An analysis of div orce in the alcid seabird Common Mur re on Great Island, Newf oundland. ALLISON T.
MOODY, Dept. Biol., Memor ial Univ. , St .J ohn' s, NF, SABINA W ILHELM, M AUREEN CAMERON, CAROLYN
WALSH, Dept. Biopsych, Memorial Univ., and ANNE STOREY, Dept. Psych., Memorial Univ.
       This study was done to try t o determine if Common Murres divorce to obtain a bett er mate or a more
compatibl e mat e, t w o hy pot heses bro ugh t forw ard t o explai n di vo rces in ot her b ird spec ies. A bet ter mate
w ould be an inherently superior mate for any individual, w hile a more compatible mate w ould be more similar or
balance some aspect (for instance compatible incubation routines), leading to an increase in reproductive
success. The field work was done during May and Jul from 1997 to 2000, on Great Island, Witless Bay, NF.
Thirty-three nest sites were followed for this time and various life history events were recorded and analyzed.
It w as f oun d t hat the di vo rce rat e on t he pl ot w as 5.7 3% , t he mort alit y r ate w as 8.0 3% . The bi rd t hat ini tiat ed
the divorce (chooser) usually improved its reproductive success while its partner (victim) did not . The timing of
div orc es seemed t o be r elat ed t o vacanc ies c reat ed by w ido w ing s. We f oun d t hat the bi rds that div orc ed had
low er reproductive success than reunited pairs. Younger birds were more likely than older birds to have poor
reproductive success and were more likely to div orce compared to old pairs. I found t hat the Bett er Option
Hypothesis provided the more likely explanation f or w hy divorces occurred in Common Murres. This
hypothesis was supported by our finding that the chooser improved reproductive success aft er the divorce
while the victim did not.

303      Gergen, Hunt & W ingfield
Plumage variation correlates wit h aggressive behavior in Lapland Longspur during the mate guarding stage of
the breeding season. THOMAS G. GERGEN* , KATHLEEN E. HUNT and JOHN C. WINGFIELD, Dept. Zool.,
Univ. Washington, Seattle WA.
         The role of plumage variat ion in w ild Lapland Longspurs of n. A laska w as studied w ith regards to
territorial behavior. Some males are entirely black in the breast, t hroat and crown, w hile others show flecks of
w hit e mix ed in w it h t he bl ack feathers in t hose regi ons. Si mulat ed Ter rit ori al in trus ion s (STI) w ere per form ed
on fr ee-living territ orial males (TM) during t he mate guarding stage of the breeding season. During t he STI a
live caged decoy w as placed in the approximate center of the area defended by a TM. Tw o decoy types were
used in the experiment: dark decoy and light decoy. Each TM in the experiment received a STI from a dark
decoy and a light decoy, w hich allow ed for a paired t-test comparison. TM' s were significantly more
aggressive towards the lighter decoys. Post-experiment trend analyses showed that darker TM w ere more
aggressive to w ards light er decoys, but light er TM w ere not mo re aggressive to w ards light er decoys. These
data suggest that plumage might be used as a badge to commu nicate rank i n male Lapland Longspurs.

304      Brewer & Vilina
Parental care behavior and double-brooding of Coscoroba Swan in central Chile. GWEN BREWER, Dept. Biol.,
Frostbur g Stat e Univ., Frostbur g, M D, and YERKO VILINA, Univ. Santo Tomas, Santiago, Chile.
         Tw elve families of Coscoroba Sw an (Coscoroba coscoroba) were observed in the El Yali wetland, 200
km sw . o f Sant iago , Ch ile, over 1 39 hr f rom 24 Oct to 1 3 Nov 20 00 and 6 - 18 J an 2 00 1. Broo d sizes ran ged
from 1 t o 8, w ith initial brood ages from several days to over 1 mon. Both male and female parents w ere
vigilant, although males w ere vigilant more often in 8 of 12 families. Cygnets and females spent more time
feeding than males. Both parents w ere aggressive toward conspecifics, and in one location, territories were
defended throughout the season. Tw o females began incubating a second clutch 4 0 - 50 d after successfully
hatching broods of 7 and 8 cy gnets, w ith second broods containing 3 and 7 cygnets, respectively. In both of
these f amil ies, the male acc ompani ed t he f irs t bro od w hil e the f emale i ncubat ed t he sec ond clu tch. In on e case,
the first and second broods formed one large family accompanied by both parents, and in the other case, both
parents were aggressive to the first brood and preferentially accompanied the second brood. Relatively high
levels of aggression were characteristic of both sets of parents, although the cohesive family spent more time
feeding and less time in alert postures than the separated family. These observations are interesting in light of
the rarity of double-brooding in w ild populations of w aterfow l, and the suggested basal evolutionary position of
this species of swan.

305      Kellam
Hom e rang e size and ov erlap among m ale and female Dow ny Woodp eckers d uri ng w int er. JA MES S. KELLAM ,
Dept. Biol. Sci. , Purdue Univ. , West Lafayett e, IN.
          Many non-migratory f orest birds, including most w oodpecker species, defend individual or group
territories even during non-breeding periods. Often, a male and female will share or partition a mutual territory.
However, t here is little published evidence that Downy Woodpeckers defend territories, and little is know n
about how paired males and females might share overlapping or adjacent spaces. The purpose of t his study
w as to determine if overlap betw een male and female home ranges during w inter could be used as an indicator
of pair bond maintenance. Using radio telemetry, I recorded home range use and behavior of 17 male and 19
female w ood peck ers d uri ng t he mont hs J an - A pr i n 1 99 8 - 20 01 . Home rang e size w as calculat ed in Arc-View
using k ernel s based on 75 % isoplet hs. The f requ ency at w hic h males and f emales w ere conc urr ent ly det ect ed
w it hin 40 m of each o ther w as also record ed at the t ime hom e rang e dat a w ere t aken. Female home rang es
w ere much larger than male home ranges. All individuals' home ranges overlapped w ith > 1 w oodpecker of the
opp osite sex, an d most overlap ped w it h at least 1 o ther i ndi vidual of the sam e sex. The magni tude o f overlap
betw een males and females did not consistently matc h the frequency at w hich pair members w ere concurrently
detected near each other. This study is t he first t o document home range use by Downy Woodpeckers using
radio telemetry.

306     Bradley
Inferring demographic parameters through radio telemetry in a cryptically breeding threatened seabird.
RUSSELL W. BRADLEY, Cent re Wi ldl. Ecol ., Dept . Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
            Radio and satellite telemetry has allowed wildlife ecologists to st udy questions which w ould be very
dif ficu lt to add ress w it hou t the adv ent of these t echn olo gies. M ost telemet ry resear ch inv olv es st udy of anim al
habi tat use, movem ent , an d surv ival. Only a f ew st udi es hav e used telemet ry dat a to di rectly inf er v it al
demographic parameters (Green et al. 1997, J. Zool. Soc. Lond. 24 3: 81 -115 ). Thi s t yp e of appr oach has
applications for t he study of cry ptically breeding species, w hose basic ecology and life history can be difficult
to d etermine w ith out the abilit y t o monit or reprodu ct ive eff ort . M anagement of thr eatened species w ith these
sor ts of repr odu ct ive st rat egies can b e dif ficu lt w it hou t kno w ledg e of basic br eedin g par amet ers. Here I
describe remote, inferred determination of breeding stage, status, and success in the Marbled Murrelet, a
threatened forest nesting A lcid. Using spatial location data from presence and absence of radio marked birds
on the ocean and in the forest, w e were able to determine initiation of breeding events and reproductive
success up to late chick rearing. These inferred results are compared w ith ground t ruthed data from inspecting
nest sites at the end of the breeding season.

307     Mika & White
Prey base diff erences and reproduct ive out put of b reeding Flammulated Ow ls in nort hern Utah. MA RKUS
MIKA* and CLAYTON M. WHITE, Zool. Dept. Brigham Young Univ., Provo, UT.
          Populations of t he Flammulated Ow l, a small insectivorous forest ow l, have been found in deciduous
forests throughout n. Utah dominated by quaking aspen (Populus tremuloides). In the summer of 2000, We
compared r epro duc tive out put w it h rel ative insect prey abun danc e and p rey deliver y r ates in 3 p opu lat ion s near
Ogden and Logan, Utah. Results suggest that noct urnal insect abundance had an eff ect on reproductive output
during this particularly dry year. Reproductive output did not correlate wit h insect prey delivery rates to the
nest. Delivery rates follow ed a pattern during different segments of the night wit h peaks after dusk and before
dawn. Young in nests w ith low er prey abundance showed increased w eight asynchrony. In 6 nests w ith
st ron g asy nchro ny , t he sm allest of the nestlin gs w as f oun d dead dur ing the nesting peri od. Dry w eather
conditions may have increased mortality among nestlings w hich is relatively low during w et years. Certain land
use practices lik e grazing and r ecreation cou ld h ave indi rectly inf luen ced av ailab le in sect prey for Flammulat ed

308     Aitken
Nest-site selection and availability in cavity-nesting birds. KATHRYN E. H. AITKEN, Centre Appl. Cons. Biol.,
Univ . Brit ish Columbi a, V anco uver, BC.
          Nest-site selection in cavity-nesters may be influenced by proximity t o foraging habitat, predation risk,
competition, nest-site fidelity, nest-site availability and specific cavity att ributes such as microclimate. Suitable
nest-sites are essential for reproduction in most cavity -nesting species and competition f or these sites may be
intense. Aggressive species may usurp cavities occupied by other individuals, including woodpeckers.
Characteristics of holes used for nesting may help to maximize the reproductive success of the individuals using
them, or t hey may simply be more common in the landscape. I examined use and availability of 37 7 cavities on
10 st udy sit es in cent ral Br it ish Colu mbia in 20 00 . Ov erall occ upan cy rat es w ere 4 1% . New ly exc avat ed
cav it ies w ere pr eferred t o ol d cav it ies f or n est ing , b ut w ere so mew hat lim it ed in the env iro nment . W hil e 95 %
of n ew cavit ies w ere occupied in 2 00 0 (n = 77 ), just 53 % o f ol d cavit ies w ere used (n= 30 6). Hatch suc cess
w as signi fican tly hig her i n new cav it ies (9 5% , n = 57 ) t han i n ol d (7 9% , n = 10 3). W hil e occ upi ed cav it ies
w ere closer t o dr y w ood land edge an d w ere lo cat ed in liv e rat her t han d ead t rees, these c avities did n ot appear
to be limited and were not used more than predicted from t heir availability. How ever, there was selection f or
nest trees wit h only 1 cavit y and avoidance of cavities > 100 m from dry edge. There were no significant
diff erences betw een newly excavated and old cavities in any nest-site variables measured.

309      Whitfield, Lynn & Cohen
South w estern Wi llow Flycatch er postf ledging behavio r and surv ival. M ARY J. W HITFIELD* , J ANET C. LYNN
and EMILY B. COHEN, Sout hern Sierr a Res. St a., Weld on, CA.
            The post fledgi ng period is t hought to b e the t ime of a bird' s lif ecycle w hen mort ality is th e greatest
(Lack 1954, Regulation of animal numbers; McCabe 1991, Little green bird). How ever, because of the
diff icult ies involved in f inding birds d uring t his tim e, it is on e of t he least st udied stages of a bird' s lifecy cle. In
199 9 and 2000, w e studied the postfledging behavior and survival of 39 Willow Flycatcher fledglings from 16
nest s. We f oun d t hat the br ood st ayed toget her an d t he f emales fed t he f ledg lin gs f or a l ong er per iod after
fledg ing than t he males. The f ledg lin gs w ere f ed f or at least 11 d af ter leav ing the nest, sometimes as lo ng as
19 d. Surv ival f or t he first 11 d w as 89. 7% (85 .7 % f or 1 3 d in 20 00 and 92 % f or 1 1 days i n 19 99 ). These
results differ from other studies of migratory passerines w hich report a high mortality rate the first w eek out of
the nest (Anders et al. 1997, Cons. Biol. 11: 698-707), Vega Rivera et al. 1998, Condor 100 :69 -78). Thus, it
appears that su rviv orship is h igh f or Wil low Flycatch ers during t he tim e they are st ill dependent on t he adults;
however survivorship after independence is unknown and may prove to be a time of high mortality.

310      King, DeGraaf & Smith
Habitat selection and habitat-specific survival of fledgling Ovenbirds. DAVID I. KING,* R. M. DeGRAAF, USDA
Forest Serv. Northeast. Res. Station, Amherst, MA; and MARIE-LOUISE SMITH, USDA Forest Serv. Nort heast.
Res. St ation, Durham, NH.
          Studi es of the hab it at use of passerin e bir ds duri ng t he per iod bet w een f ledg lin g and dispers al in dic ate
that the pr eferenc es of som e bir d species chan ge af ter f ledg ing , an d t hus, o ur u nder st andi ng o f their habi tat
requirements based on surveys conducted during the nesting season may be incomplete. We studied the
post-fledging habitat selection of Ovenbirds in n. New Hampshire during 2 breeding seasons using radio
telemet ry . Habit at at sit es used by 41 radi o-m arked f ledg lin g Ov enbi rds w ere charac terized by great er v ert ical
struct ure 0 - 3 m above ground than Ovenbird nest sites. Similarly, habitat at sit es used by fledgling Ovenbirds
w ere charac terized by great er v ert ical st ruc ture, hig her s hru b den sit y, and f ew er mediu m and lar ge t rees t han
unu sed sit es. Tw elve of 15 mort alit ies (8 0% ) w e observ ed w ere du e to pr edat ion , an d f ledg lin g surv ival w as
positively significantly related to vegetation struct ure. Our observations that f ledgling Ovenbirds are selective in
their habi tat use, that they select dif ferent habi tat features than ad ult Oven bir ds select for n est ing , an d t hat
fledg lin g surv ival is posit ively associat ed w it h t hese habit at features , sugg est s t hat assessments of habi tat
quality based on abundance of adult birds are incomplete.
311     Rising, Jackson & Fokidis
Geographic variat ion in pl umage color and p att ern of Savannah Sparrow s. J. D. RISING* , D. A. JA CKSON,
and H. B. FOKIDIS, Dept. Zool., Univ. Toront o, Tor ont o, ON.
          The Savannah Sparrow is one of the most abundant and widespread songbirds of North America. At
pres ent , 1 7 subspeci es of Savan nah Sp arrow s are r ecog nized . These have been del imit ed and named
principally on the basis of putativ e differences in coloration and color patt ern. We examined geographic
variati on in co lorati on among po pulati ons of the 1 0 of these t hat are migrat ory. To quant itat ively assess
patterns of color and pattern variation we examined 1020 specimens of adult male Savannah Sparrows from 4 1
localities from virtually t hroughout t heir breeding range. For each bird, 11 quantitative mult istate plumage
characters w ere assign ed a sc ore o f 1 t o 6 (low est to hi ghest). The d egree o f sim ilar it y f or t he co mbined
characters betw een each pair of birds w as determined using Gow er's coeff icient of similarity, and the
complexity and dimensionality of the resulting matrix was reduced using Principal Coordinates Analysis (PCoA).
PCo1 ordinated the birds on an axis of relatively dark to relatively pallid coloration. The birds from Sable Island,
Nova Scotia ("Ipsw ich" Sparrows) are significantly dif ferent from all others on this axis; how ever, there is
substantial overlap among all other localities, and variation is clinal, wit h the most pallid birds being found
w here the air temperature is highest (and in the west). The variation on all of t he other PCo axes similarly is
clinal. With t he exception of t he Ipswich Sparrows, delimiting additional subspecies on the basis of plumage
coloration or pattern would appear to be arbitrary.

312     Gergen, M anuwal & Gara
Food abundance and fledging success of Dark-eyed Juncos in managed forests of the eastside of the Cascade
Mountains, Washington. THOMAS G. GERGEN* , DAVID.A. MANUWAL and ROBERT I. GARA, Coll. For. Res.,
Univ.Washington, Seattle, WA.
         Art hro pod s w ere capt ured at the f eedin g lo cat ion s of nest lin g-f eedin g adu lt Dark -eyed Junc os. Samples
in the vicinity of 2 8 nests w ere collected during 1999 and 2000 using a sweep net. Samples were dried and
w eigh ed. Ther e w ere signi fican t dif ferenc es in the dr ied b iom ass of art hro pod s cap tured bet w een nests t hat
were depredated, and nests that successfully fledged at least 1 nestling. There was also a significant positive
relationsh ip betw een the number of fledgl ings prod uced and t he biomass of arthr opods.

313      Dufour, Brua & Clark
Survival, recruitment, and local population growth in two contiguous sub-populations of Mallards: An
application of reverse-time capture-recapture analysis. KEVIN W. DUFOUR,* ROBERT B. BRUA, Dept. Biol.,
Univ . Sask atc hew an, Sask ato on, SK; and ROBERT G. CLARK, Canadian Wildl. Ser., Saskatoon, SK.
          We used data f rom an 1 1 y r fi eld stud y of indiv idually-m arked female mallards (A nas platy rhyn chos) t o
explor e pot ent ial d if ferenc es in key pop ulat ion pro cesses bet w een 2 loc al sub-popu lat ion s: bir ds breedi ng o n a 2
ha island at the core of our study area, and those breeding on the mainland. Specifically, w e used a
combination of convent ional and reverse-time capture-recapture analysis to characterize each sub-population
w ith respect to rates of apparent survival, recruitment, and local population grow th ( 8 ). An init ial analysis
indicated that island-nesting females survived at an appreciably higher annual rate than mainland-nesting
females. Year-specific estimates of 8 based on reverse-time capture-recapture analysis w ere consistently
posit iv e for isl and-nest ing birds, s uggesting su st ained grow th of this su b-p opulation during t he study. By
cont rast, the mainland su b-populat ion experienced perio dic pop ulatio n declines ( 8 < 1) corresponding to years
of poor breeding habitat condit ions. Over all years, the proportional contribution of adult survival to 8 w as
greatest among island-nesting females, w hereas population increases on the mainland w ere influenced more
heavily by recrui tment (in sit u rep rod uction + immigrat ion ). An anal ysis of dispers al pat terns of marked
duc kli ngs pro vided n o eviden ce of demogr aphi c rescu e bet w een t he 2 sub-popul ation s. We su spec t that
pop ulat ion chan ge on the mainl and w as influen ced m ost heavily by immigrat ion from out side the st udy area.
Regardless, our results indicate substantial spatial variation in demographic processes operating on a relatively
loc al geo grap hic scal e.
314     Trainer, McDonald & Learn
The development of c oordin ated singin g in duet tin g manakins. JILL M. TRAINER* , Dept. Biol., Univ. Northern
Iow a, Cedar Falls, IA; DAVID B. McDONALD, Dept. Zool. & Physiol. , Univ. Wyom ing, Laramie, WY; and
WILLIAM A. LEARN, Dept. Computer Sci., Univ. Northern Iow a.
        Long-tailed manakin s (Chiroxiphia linearis) hav e an un usual social syst em in w hic h t eams o f tw o males
sing and dance cooperatively during courtship, a normally competitive situation. Especially puzzling is the fact
that only t he older, more socially dominant of the tw o males mates wit h visiting f emales. One benefit of
performing as a nonmating partner might be to gain experience as an “apprentice” to improve the performance
of the complex duet song and joint dance. Testing this possibility, w e found that singing performance
improved with age; older partners’ singing was less variable that that of younger partners, and frequency
matching increased w ith t he age of the younger partner. Furthermore, w e found that w ell established,
successful teams had better matched songs than teams we simulated by randomly pairing males. The
develop ment of frequ ency mat ching may be a lo ng-term st rat egy to in crease mat ing suc cess w hen “ appr ent ice”
males inherit display sit es from older males.

315     Wells, Lusk, Guthery & Fuhlendorf
Lark Sparrow nest -site selection and success in a sout hern mixed-grass prairie. KIM BERLY SUEDKAM P
WELLS* , JEFFREY J. LUSK, FRED S. GUTHERY, Dept. For. , Okl ahoma St ate Uni v. , St illw ater, OK; and
SAMUEL D. FUHLENDORF, Dept. Plant & Soil Sci., Oklahoma State Univ.
         Lark Sparrow s are declining t hrough out most of t heir range, but relativ ely lit tle inf ormat ion about nest
site selection is known. We studied factors aff ecting nest-site selection and nest success in a southern
mixed-grass prairie of Oklahoma. We used a neural netw ork technique to discriminate between nest and
random locations and bootstrapping w ith 95 % conf idence intervals to compare successful versus unsuccessful
nests. Habitat featu res measured during t he breeding seasons of 1 99 9 - 2 00 0 inc luded: li tt er depth (m m),
distance to visual obstruction (m), screening cover, emergent plant height (cm), distance to nearest w oody
homologue (m), and percent cover of bare ground, litter, grasses, forbs, and woody plants. The area under a
ROC plot indicated that 91 % of the time our neural netw ork correctly classified nest locations. Percent woody
cover, dist ance to nearest w oody h omolog ue, percent bare ground, and percent l itt er cover w ere the most
important v ariables in neural netw ork. We conducted a simulation analysis using these variables to determine
transition thr esholds betw een random and suitable nest sites. When distance to nearest w oody homologue
w as = 0. 28 m or percent w oody co ver < 10 %, locations became unsuit able for Lark Sparrow nests. If a
locati on had > 80 % l itt er cover or > 90 % b are ground it w as classified as a random poi nt. Successfu l nests
had shor ter di st ances to visual ob st ruc tion , l ow er li tter dep ths, less bare g rou nd, and l ess w ood y c ov er t han
unsuccessful n est s. Unsu ccessf ul n est s w ere associat ed w it h short er emergen t plan t heig ht s and w ere closer
to w oody plants than successful nests. Our results have important implications for land managers interested in
managin g f or Lar k Sparrow nest habi tat.

317     Yuan & Shen
Cooperative breeding of Taiwan Yuhinas. HSIAO-WEI YUAN and SHENG-FENG SHEN* , Dept. Forestry,
National Taiw an Univ., Taipei, Taiwan.
          Taiw an Yuhinas (Yuhina brunneiceps), which are endemic to Taiwan, always breed cooperatively. We
studied their breeding ecology at Mei-Fung, in the mountains of central Taiwan, during 1995, and 1997 - 2001.
Each parental group of 3 - 7 adults w as comprised of 1 t o 3 monogamous pairs (joint-nesting), some wit h 1
help er. The m ost com mon (3 9% ) par ent al gr oup size w as 4 (n = 24 ). Parent al gr oup s of 4 an d 6 had g reat er
breeding success (1.0 7 and 1.0 0 f ledglings/adult, respectively) than parental groups of other sizes. During egg
inc ubat ion , t he amoun t of w ork perf orm ed du rin g t he day by dif ferent ind ividu als in a par ent al gr oup dif fered
signif icantl y (X 2 test , p < 0. 05 ). In 4 nest s, daytime inc ubat ion w as per form ed pr imaril y by t he alp ha f emale,
and in 2 nests by the alpha male. However, all members of t he parental group contributed equally to feeding
the ch icks (X 2 test, p> 0.0 5). There are no studies of joint-nesting passerines. Among 14 species of
joint-nesting non-passerines, incubation by males, especially at night, w as thought to be an essential precursor
to the evolution of cooperative breeding. However, Yuhina females incubated the eggs at night in 4 of 6 nests.
Because juv enil es disper se readily and f ood is abund ant , w e thin k Tai w an Yuhin as br eed co oper atively to of fset
the high rate of nest failure (66%) caused by bad weather and predation. Cooperative breeding may have
evolved to spread risk, with fewer eggs/female/nesting attempt, and to accelerate the process of renesting.

318     Brown, Thomsett & Oliver
Mass m ort alit y of Eurasian Whi te St orks due t o predat ion in Keny a, east Af rica. JESSI L. BROWN* , SIM ON
THOMSETT and PETER OLIVER, The Peregrine Fund, Boise, ID.
         The Eurasian White Stork (Ciconia ciconia) is a palearctic migrant of ten sighted in east Af rica en route
to and from it s w intering grounds. On 28 Feb 2001 w e observed the aftermath of a mass mortality event of a
flock of White Storks roosting in a degraded area of t he Athi Plains of Kenya, some 30 km out side of Nairobi.
We f oun d rem ains of 27 ind ividu als w it hin 30 m of the per imeter of a 1 h a season al po nd. Thr ee addi tion al
bir ds w ere f oun d ali ve but inj ured , w it h co mpoun d f ractures of tarsi and/ or w ing bon es. Necr opsies perf orm ed
on 4 birds show ed injuries consistent w ith bit es inflicted by large carnivores. We concluded that locally
commo n Spot ted Hy enas (Crocuta crocuta) had att acked and killed many birds at night. W e also observed many
oth er species scavenging t he stork s' remains, inc luding Black-back ed Jackals (Canis meso melas ), Taw ny Eagles
(Aqui la rap ax), A fri can Whit e-backed Vult ures (Gyps africanus), and Lappet-f aced Vult ures (Aegypius
tracheliotus). Similar events mu st happ en w it h reg ular it y: how ever , as larg e areas in east Afric a are co nv ert ed
from t he original vegetation cover to a more open and degraded scrub, relative levels of migrant mort ality due
to pr edat ion may chan ge. Posit ive chan ges in po lic y and pr act ice i n Euro pe and Asia m ay be cou nt erbal anced
by events in Af rica. This may have significant ramifications on the success of attempts t o conserve palearctic

319     Folk & Hepp
Sources of variation in incubation constancy of Wood Ducks. TRAVIS H. FOLK and GARY R. HEPP, School
For. & Wildl . Sc i. , A ubu rn Univ ., AL.
          Incubat ing bird s must maint ain a certain level of const ancy (i. e., t ime spent o n nest) f or embry os to
develop suc cessfull y. For so me w aterf ow l species, f act ors suc h as age, b reedi ng ex peri ence, and h abit at
qual it y have been show n t o in fluen ce in cubat ion con st ancy . For Wood Ducks, inc ubat ion con st ancy var ies
among f emales but is not inf luen ced b y body mass of females at the st art of inc ubat ion or m ass loss of females
dur ing inc ubat ion . In t his st udy , w e test ed w het her v ariat ion in h abit at use and mov ement pat terns of females
dur ing inc ubat ion recesses i nf luen ced av erage d aily inc ubat ion con st ancy . Incubat ing females pref erred beav er
pon ds, creeks, ephem eral w etland s, and l ake in fluen ced w etland s, and h abit at pref erenc es w ere no t inf luen ced
by year, age, or initiation date of incubation. Mean distance traveled during incubation recesses, standard
deviation of distances, variation in angle of flight from nest box t o recess locations, and early incubation body
mass did not explain var iat ion in i ncubat ion con st ancy . Females t hat nest ed earl y in t he seaso n had hig her
const ancies than t hose that nested lat er, and incu bation const ancy w as negatively related t o th e amount o f use
of pref erred hab it ats. Result s illustrat e that inc ubat ion con st ancy of female W ood Duc ks w as not con st rain ed
by location of available habitats, but seasonally changing climatic conditions may influence incubation behavior
of W ood Duck s.

320     Caff rey & Peterson
Eff ect s of food lim it ati on o n nest ing in Cali forn ia Least Terns. CAROLEE CAFFREY* and CHA RLES C.
PETERSON, Dept. Zool. , Oklahoma State Univ ., St illw ater OK.
          Yearly variation in local environmental conditions affects reproductiv e performance, and food availability
has been demonstrated to exert proximate control over the nesting characteristics of many bird species.
Californi a Least Terns (Sterna antillarum brow nii) arr ive each spri ng f rom their w int erin g gr oun ds t o br eed at
sites along t he coast, fro m t he border in t he south to San Francisco Bay. A t on e site (Venice Beach),
behav ior al ob serv ation s, and c hic k mort alit y, necr opsy, and eg g aban don ment dat a ind icat ed a mo derat e to
severe shortage of prey in 3 of the 6 yr of t his study. The eff ects of food limitation (adjusted, w here
necessary, for seasonal effects) on several measured nesting parameters w ere egregious. In years w hen food
w as apparently in short supply : 1 ) the num ber of d ays betw een populatio n arrival and t he init iation of t he first
nests was greater, 2) mean clutch size w as significantly smaller, 3) inter-egg intervals for first and second eggs
w ere greater, and limited data suggest the same for second and third eggs, and 4) incubation periods, for bot h
firs t and secon d egg s, w ere signi fican tly great er. Clut ch size and t he len gt hs of inc ubat ion peri ods cor relat ed
w it h f ledg ing suc cess. Thus, f or t his endan gered spec ies, the pr ecip it ous decl ines in f ledg lin g-t o-pair r atios that
occur in El Niño years appear to be a function of bot h chick starvation and the effects of food limit ation on the
egg stage of the nesting cycle.

321      Mays, Yao & Yuan
Breeding biology and antiphonal dueting in a Taiw anese endemic passerine, Steere's Babbler, Liocichla steerii.
HERMAN L. MAYS, Jr.* , Dept. Biol., Auburn Univ., Auburn, AL; CHENG-TE YAO, Taiw an Endem ic Speci es
Res. Inst., Jiji, Taiw an, and HSIAO-WEI YUAN, Dept. Forestry, Natl. Taiwan Univ., Taipei, Taiwan.
          Avian breeding biology has primarily been studied in temperate zone species or to a lesser degree in
Neot rop ical spec ies. Comparat ively lit tle is kno w n abo ut avian mating syst ems i n ol d w orl d t rop ical /subt rop ical
bir ds. Here w e pres ent dat a on t he gen eral b reedi ng b iol ogy of an end emic Taiw anese passerine, Steere' s
babbler, Liocichla steerii. Steere's Babbler like many tropical species are year round residents and exhibit
ant iph onal duet ing ini tiat ed by the male. Alon g w it h do cumenting basic br eedin g bi olo gy w e test the id ea that
female answ er rates are contingent upon habitat characters wit h females answering male song more often in
visually occ lud ed hab it at to f acil it ate clo se co nt act . In co mparin g t o sites, one h igh ly visually occ lud ed
und ist urb ed f ores t and t he a ot her a m ore o pen h uman di st urb ed agr icu lt ural area, w e foun d t hat female an sw er
rate w ere signifi cantly higher on un dist urbed visu ally occ luded habit ats.

322      McFarlane-Tranquilla & Bradley
Does c atching Marbl ed M urr elet s disru pt breed ing ? LAURA M cFA RLANE TRANQUILLA* and RUSSELL
BRADLEY, Cent re Wi ldl. Ecol ., Simo n Fraser Univ ., Burnaby, BC.
          In any research project, it is difficult to assess the impact of capture on the study subjects. However,
it is often di fficu lt to det ermine b asic avian bi olo gy w it hou t hand lin g bi rds . In a species lik e the M arbl ed
Murrelet, elusive behaviour and non-colonial breeding habits have made describing reproductive status of
individuals and populations difficult . This study determines breeding status using radiotelemetry and
physiological techniques concurrently, to give as much information as possible on individual birds.
Physiological methods allow identification of egg-producing females caught at sea, away from the nest.
Incubation behaviour of t hese females is then follow ed using radiotelemetry. The number of females producing
eggs vs. the number of females that show subsequent incubation behaviour is then compared. Results show
that many egg-pro ducing females do not go on t o incub ate eggs (‘ failed’ breeders?), and that many of tho se
w ho d o in cubat e, d o so much lat er t han p redi ct ed. How ever , an alysis of a st ress hor mone, cor tico st eron e,
suggests that the stress response of birds to capture and handling is higher in ' failed' breeders, as compared to
successful or non-breeders. Thus, there may be a predisposition in this group of ‘ failed’ breeders to breeding
disruption by research activit ies. Although breeding may be disturbed by research activities, the amount and
intensity of disturbance that w ill cause abandonment of a breeding att empt is variable, depending on the
at-capture physiological state of each individual.

323      Aspbury, McDaniel & Gibson
Ecological mechanisms of mixed species lek format ion in prairie grouse. A NDREA S. A SPBURY* , Dept. Biol.
Sci., Univ. Nebraska, Lincoln, NE; LEONARD McDANIEL, U.S. Fish & Wildl. Serv., V alentine, NE; an d ROBERT
M. GIBSON, Dept . Biol. Sci., Univ. Nebraska.
         Mixed species leks , w hic h oc cur regul arly bet w een Great er Prairie Ch ick ens (GPC) and Sh arp- tailed
Grou se (STG), are in terest ing becau se t hey are no t easily ex plai ned b y c urr ent ly pop ular hypot heses for l ek
evolut ion . GPC and STG lek s hav e been c oun ted and mapped an nual ly since 1 98 3 at the Valent ine Nation al
Wildlife Refuge in north-central Nebraska. Previously w e have shown t hat mixed-species leks form w hen small
num bers of GPCs join an ex ist ing STG lek , an d t hat this requ ires bot h spatial p rox imit y betw een t he 2 spec ies
and low local density of GPCs. Here w e explore the ecological underpinnings of spatial proximity . We show
that GPC and STG m ales lek in sit es t hat dif fer in topo grap hy and v eget ation al co ver . GPC mal es pr efer sites at
low er elevations and in “w etter” plant communities. How ever, the species' elevational ranges overlap and,
STG leks located at lower elevations are in closer proximity to GPC males than those at higher elevations. As
the population density of GPCs increased, the mean elevation of GPC leks rose, indicating expansion into
marginal habit at. Ho w ever, t his did no t in crease proximi ty to STG leks.

324     Wisti-Peterson & W Ingfield
Chang es in andr ogen recept or ex pres sion are c orr elat ed w it h br eedin g in a w ild songbi rd, the Whit e-cr ow ned
Sparrow. DEBORAH WISTI-PETERSON* and JOHN C. WINGFIELD, Dept Zool. , Univ . W ashingt on, Seattle,
          Environ mental cu es, part icu larl y increasing d ay lengt h, trig ger i ncreases in p lasm a sex st eroi d ho rmones
that , in t urn, orchest rate mor phologi cal, phy siologic al and behavioral changes associat ed w ith breeding in bir ds.
The ef fect s of one sex steroi d, test ost eron e (T), and i ts met abol it e, 5 a-dihy dro test ost eron e (DHT), are mediat ed
through binding to an intracellular receptor protein, the androgen receptor (AR). When AR binds T or DHT, AR
enters the cell nucleus where it binds directly to specific regions in the genome, thereby altering gene
expression. Changes in gene expression are correlated with observed changes in physiology, morphology and
behavior. How ever, a precise understanding of how w ild songbirds detect and respond to diff erent T
concent ration s cannot be full y kno w n unt il seasonal AR expression patt erns are described. To ex amine these
questions, portions of t he White-crowned Sparrow (WCSP) AR were cloned and used to probe northern blots
from v arious tissue types and in situ brain sections of m ale and female WCSP to identif y seasonal changes in
dist ribut ion pat terns and nu mbers. Th ese studies reveal seasonal changes in AR expression in o vary and t estes,
and in particular regions of the brain of WCSPs, thereby supporting the hypot hesis that AR expression levels
change seasonally and are correlated w ith observed changes necessary f or breeding in WCSPs.

325     Edwards, WIngfield & Thompson
When is low weight detrimental to breeding terns and gulls? ANN E. EDWARDS* , JOHN C. WINGFIELD, Dept.
Zool., Univ. Washington, Seattle WA, and CHRISTOPHER W. THOMPSON, Wash. Dept. Fish & Wildl. , Mill
Creek, WA.
         Studies of breeding birds show t hat adult fat scores correlate with reproduct ive success in some but
not all cases. This raises the questions 1) when are a breeding bird’s reproductiv e decisions most affected by
its body condition, and 2) how do species differ in their sensitivity to body condition when making reproductive
decisions? We used circ ulatin g levels of cort icost erone (as a measure of p erceived pot ential f or st ress) to
indicate when low fat reserves were detrimental to reproduction in sympat rically breeding Caspian Terns and
Glaucous-winged/Western Gull hybrids. In breeding birds, elevated plasma levels of cort icosterone frequently
cause reduced parental investment or tot al nest abandonment, behaviors that favor adult survival. Fat score
w as negatively correlated with cort icosterone level in terns but not gulls, suggesting that terns may be more
dependent on fat st ores than gulls to successfully raise chicks. The bases for interspecific variation may include
diff erences in flight eff iciency, extent of diet specialization, or degree of overlap in timing of breeding, molt and
migration. The strength of t he correlation betw een fat and corticost erone varied wit h breeding sub-stage in
terns. This suggests that fat reserves, and thus food availability, may be more critical at some reproductive
stages t han oth ers.
326     Brightsmith
Origin and maintenance of nest site diversity in tropical cavity nesting birds: Evidence from termitarium nesters
in Peru. DONALD BRIGHTSMITH, Dept. Biology, Duke Univ., Durham, NC, and Tambopata Research Center,
Puerto M aldonado, Peru.
         Cavity nesting birds that use old tree cavities (secondary cavity nesters) frequently suf fer from nest sit e
limitation and subsequent competit ion for nest sites. For this reason I hypothesized that competit ion w ould
favor t he use of alternative nesting substrates such as termite mounds. Alt ernatively, predation could f avor the
use of no vel nesti ng subst rates by c avit y nest ing bird s. Parsimony analy ses of pub lished phy logenies suggest
that nesting in arboreal term ite mo unds by Neotrop ical parakeets (genus Brotogeris) and tro gons (genus Trogon)
is a derived trait. Evidence from nearly pristine lowland tropical forests in se. Peru suggests that tree cavity
availability is not limiting reproductive output of small subcanopy nesting species and should not be favoring
the use of alternative nesting substrates. Instead data from natural and artificial nests suggest that predation
rates in secondary cavity nests are much higher than predation rates on nests in termite mounds. This work is
one of few to look at cavit y nesting in a nearly pristine environment and suggest that predation, not
competit ion , i s mo ldi ng n est sit e selection among t hese t rop ical taxa.

327     Brightsmith & Nycander
Conservati on of large Macaw s (genus Ara): Devel opi ng t echn iqu es t o in crease reprod uctive out put . DONA LD
BRIGHTSMITH* , Dept. Biol., Duke Univ., Durham, NC and Tambopata Research Center, Pto. M aldonado, Peru,
and EDUARDO NYCANDER, Rainforest Expeditions, Lima, Peru.
         Large macaws are endangered or declining throughout t he Neotropics due to habitat loss, hunting and
collection for t he pet t rade. Natural reproductive rates of these species are low and there is often a lack of
suitable nest sit es. As a result even w hen protect ed areas are established populat ion recovery w ill be slow . In
southeastern Peru we have used nest boxes, hand raising and hand feeding to increase the reproductive output
of Scarlet and Red-and-green Macaws. For nest boxes, occupancy and survival of nestlings to fledging is high,
but hatching rate is below t hat of natural nests. In addition this project has developed tw o techniques for
counteracting brood reduction and facilitating the survival of additional chicks. Hand raising saved 32 chicks
(Scarlet and Red-and-green Macaw s combin ed) that w ere released in th e early 19 90 ' s. A t least 12 of t hese
(38% ) hav e sur vived for 8 or m ore y ears and 6 of these have paired w it h w ild mat es and bred . Su ppl emental
feeding t echniques have also saved chic ks and allow ed them t o fl edge w ith their p arents. In sum t hese
techn iqu es hav e led t o subst ant ial an nual inc reases in n est ing suc cess and m ay be appl ied t o more en dang ered
members of the genus Ara in ot her areas.

328     Silva, Mackay, Edwards & Boersma
Relative contributions of egg size and parental quality on reproductive success of Thin-billed Prions breeding in
the Falkland Islands. MONICA C. SILVA* , Dept. Zool., Univ. Washington, Seattle WA; STUART MA CKAY,
Seattle, WA; SCOTT V. EDWARDS and P. DEE BOERSMA, Dept. Zool., Univ. W ashington.
         By assuming that egg size may play an im por tant rol e in d etermini ng c hic k gr ow th and sur vival in av ian
species one may overlook the confounding eff ect of v ariability in parental quality, w hich may influence both egg
size and reproductive success. A clutch transfer experiment was conducted in order to assess the relative
impor tance o f paren tal qu alit y and egg size per se on Thin-billed Prion reproductive success. This species, as in
all petrels, lays a single egg. Small-egg clutches w ere cross-fostered wit h clutches of bot h small and large eggs
and v ice v ersa. Egg size w as f oun d no t to af fect tarsu s lengt h, mass and b ody con dit ion of chi cks t hat hat ched
suc cessfull y. How ever chi cks t hat sur vived, h atched f rom signif ican tly larg er egg s t han c hic ks t hat died at
hatching o r short ly aft erw ards. Egg size also did not sign ific antly in fluenced chic k grow th or fledgling size. In
contrast, chicks raised by high-quality parents were not only skeletally larger but also heavier at hatching w hich
indicates that hatchlings were fed shortly aft er eclosion. They also gained mass at a higher rate during the
linear grow th period and achieved higher peak weights. Parental quality effect s became evident early on the
nestling stage and w ere more important t han egg size in determining chick growt h and survival. Correlations
found in the past betw een egg size and fledging success may be due to confounding effects of parental quality.
Our data will be put into context with results found for other seabird species.

329     Green, Lombardo, Thorpe & Czarnowski
Patt erns of sperm depletio n in Tree Swallow s. M ICHELLE L. GREEN, M ICHAEL P. LOMBARDO* , PATRICK A.
THORPE, Dept. Biol., Grand Valley St. Univ., Allendale, MI; and MATTHEW R. CZARNOWSKI, Dept. Ecol.,
Evol. & Nat. Res., Rutgers Univ., New Brunswick, NJ.
         A male bird's ability t o fertilize eggs is positively correlated w ith his sperm count. Therefore there
should be strong selection pressure on male birds to produce large numbers of sperm to avoid sperm depletion
w hen t heir mat ing syst ems ar e char act erized by int ense sperm com pet it ion . Tree Sw allo w s are c harac terized
by intense sperm competition because mated pairs copulate frequently, f emales store sperm, and extra-pair
copulations are common. We examined patt erns of sperm depletion in Tree Swallow s by repeatedly sampling
15 males w hil e their mat es w ere in cubat ing eggs. Sem en sam ples w ere coll ect ed w hen m ales w ere capt ured
and then again hourly for 4 hr. The proportion of infertile samples increased from 6.7 % to betw een 26.7 -
33 .3 % over t he co urs e of the sam pli ng p erio d. 40 % of males pro vided at least one i nf ert ile sample. The t ot al
sperm obtained from males over the sampling period varied by orders of magnitude from hundreds of thousands
to hundreds of millions of sperm cells. Log scores for tot al sperm obtained were not correlated wit h date or the
num ber o f fert ile f emales availabl e at the st udy sit e. These r esults suggest that dif ferenc es in sper m su ppl ies
w ere du e to in trin sic dif ferenc es amo ngst males and t hat males w oul d suffer f rom sper m deplet ion if they
copulated several times in short succession. Sperm depletion by males has important implications for the
copulatory strategies of male and female Tree Swallows.

330     Stern, Lauten, Castelein, Popper & Fukuda
Imp act assessment of oil spilled from the New Carissa in February 199 9 on t he Western Snowy Plover along
and JULIE FUKUDA, Oregon Nat. Herit age Prog., The Nature Conser vancy, Portl and, OR.
         On 4 Feb 199 9 the 33 ,17 6-T freighter New Carissa ran aground on the Oregon coast approximately 4
km north of the Coos River, Coos Co., and beginning on 7 Feb and thereafter spilled an estimated 76,000 -
530 ,00 0 l of f uel oil into the ocean. The spilled oil w ashed onto nearby nesting and wintering beaches of a
thr eatened populat ion of Western Snow y Plovers (Charadrius alexandrinus nivosus). Betw een 10 Feb and 16
Apr 1 99 9, 62 % (n = 73 ) of the in div idu ally marked Snow y Plo ver s observ ed alo ng t he Oregon coas t bet w een
Cape Blanco and Heceta Head were sighted wit h some oiling. A majority of oiled plovers occurred wit hin the
first 3 w k of the spill, and nearly all of the plovers that incurred more severe oiling did so in that same period.
Seventeen plovers were oiled to such an extent that t hey were trapped, cleaned, rehabilitated and subsequently
released. The abundance and productivit y of breeding plovers along the Oregon coast in 19 99 w as not overtly
impacted by this incident, as population levels w ere comparable in 199 9 w ith previous years, and productivity
as measured by the nu mber of young fledg ed w as higher than i mmediat ely prec edin g years and co mparabl e to
peak years of 199 4 and 199 5. There may, how ever, have been impacts to specific sites and specific
indiv iduals. W e report o n th e details of these f indings.

331     Mahony & Krannitz
Productivity of Brewer's Sparrows at the northern edge of their range. NANCY A. MAHONY* , Centre Appl.
Conserv. Biol. , Univ. British Columbia, Vancouver, BC; and PAMELA G. KRANNITZ, Canadi an Wild lif e Serv ice,
Delt a, BC.
         Using measures of individual nest success to assess productivity in songbirds is misleading for
multi-brooded species. Seasonal fecundity, t he number of chicks fledged over the w hole breeding season
ult imately det ermines breed ing suc cess, althou gh i t is seldo m assessed. From 19 98 - 200 0 w e exam ined
seasonal fecundity of Brewer's Sparrows at the edge of their range at 4 sites in s. British Columbia. We
follow ed 125 colour-banded females and determined the number of nesting attempts during the breeding
season and the ou tcome of each n est . 4 9% of females attempted 2 nest s, 9% attempted 3 and 1 made 4
attempts. 17% of f emales fledged 2 broods and 2 f ledged 3 broods in a season. Average seasonal fecundity
w as 3.09 (SE = 0.19) chi ck s over the co ur se of the st ud y. Seasonal f ecun di ty varied b et w een sites (P =
0. 00 8) and t here w as a site* year int eraction (P = 0. 00 04 ) as, fr om year t o year, t he most and least
productive sites alternated. Seasonal fecundity w as positively correlated w ith t he number of successful
nests/female, and negatively correlated wit h the proportion nests depredated, how ever, correlates of f ecundity
varied betw een sites. Season length w as positi vely c orrelated w ith number of att empts, althou gh early nest s
w ere at higher risk of loss due to storms. If global warming increases season length, product ivity could
increase although t he eff ect of early season storms may be detrimental.

332     Roper
Urban living, brood reduction, accidental experiments and the Rufous Hornero. JAMES J. ROPER, Universidade
Federal do Paraná, Curitiba, Paraná, Brazil.
          The Rufous Hornero (Furnarius rufus) is common in cities of Brazil, Paraguay, Uruguay, A rgentina and
Bolivia. Originally it w as found in savanna habitats in these countries. Now it is found in many cities and its
range is increasing due to agriculture, deforestation, and city parks. With range expansion, the Rufous Hornero
has also found new sources of f ood and nest-construct ion materials. The new food is garbage and pet food,
w hile mud is f ound y ear-round in t he w ell-w atered law ns. These have comb ined to increase nesting su ccess
and to facilitate range expansion. Experiments, both planned and accidental, were used to t est the influence of
these sources of f ood and building m aterial on nesting succ ess, brood reduct ion and off spring qualit y. In
experiment al territ ories (th ose w ith anthro pogenic sou rces of foo d and mud), brood r educt ion never oc curs,
w hile in control territories (with ambient food and mud) brood reduction varies from none to the entire clutch
(usually a maximum of 4 eggs). In experimental territories all the young may survive, but as they grow larger,
sibling competition may be great due to size-symmetry and health of the nestlings. In control nests however,
size diff erences betw een siblings may be large, and smaller birds often starve due to sibling competition in
w hich the largest of the nestlings survive. When one or few young survive, t he reduced number of nestlings
may be much easier for parents to feed, and so the survivors may fledge in very good condition. Surprisingly,
theref ore, of fspr ing qualit y (measured as size or body c ondit ion at f ledgling) may be greatest in cont rol nest s
w ith t he greatest mort ality of y oung, w hile competition betw een many survivors of experimental territories wit h
more abundant food may result in more young of lower quality.

333     Valdez
Raptor communities in disturbed and non-disturbed areas of Manu Biosphere Reserve, southeastern Peru.
          Birds of prey (raptors) in the neotropics, have been scarcely studied and little is know n about their
ecology, habitat requirements and effects of human disturbance on them. This study f ocuses on the
composition of diurnal birds of prey in human disturbed and pristine areas of Manu Biosphere Reserve (MBR).
During 8 mon (betw een 199 7 and 199 8), I conduct ed censuses of raptors in 4 localities of MBR (2 pristine and
2 d ist urb ed on es) under 3 m ethod olo gies (open areas, f rom the can opy , p layback s). I co mpared s peci es
compo siti on, sp ecies richness, sp ecies diversit y and relati ve abundance of raptor s among localit ies. I also
performed Canonical Correspondence Analysis (CCA) to examine whether forest raptors show affinit ies for
dif ferent suc cesi onal st ages of the veget ation . From a tot al of 46 rapt or s peci es rep ort ed f or M BR I regist ered
36 (7 8% ) combining all methods and all sites. Species composition diff ered among the localities evaluated, but
percentage of similarity w as higher w ithin t he 2 non-disturbed and w ithin t he 2 disturbed localities (85. 1% and
85. 8% , respectively). I recorded higher number of species in pristine localities (26 and 28) and low er numbers
in dist urbed ones (1 9 and 1 4). Species diversit y w as also higher at pri stin e localiti es (H' = 2. 56 and H' =
2.4 6) than in disturbed ones (H' = 1.7 1 and H' = 1.1 8). Relative abundance of raptor species differed among
the 4 localities in somew hat complex patt erns, although many species were consistently more common or more
rare in each of these sites. Results of the CCA showed that forest raptors did not seem to have specific
associat ion s w it h successional habi tat typ es, but appar ent ly for c ert ain p hy siogno mic f eatures of the hab it at
(density of palms, canopy cover and others). This attempt to characterize the patterns of raptors abundance
and community composition in sit es wit h contrasting levels of human disturbance at MBR is the first step in the
underst anding of raptor commu nit ies in th e Peruvian rainf orest.

334     Gratto-Trevor
Sex-biased brood care in Willets and Marbled Godwits. CHERI L. GRATTO-TREVOR, Canadi an Wild lif e Serv ice,
Env. Can., Saskatoon, SK.
          Brood care in bi-parentally incubating sandpipers is often male-biased. Most females desert broods
earlier than their mates, and desertion is usually earliest at late-hatching nests. Brood care in Willets nesting in
s. Alberta followed this pattern, and only 17% of females remained wit h the brood until fledging. In the same
area, relatively more female Marbled Godw its (49 %) remained with t heir broods until fledging. Since shorebirds
energetically defend broods from predators, male-biased brood care may result in lower mort ality rates for
deserting females, and therefore a higher survival rate of a male's mate if he cares for the brood alone.
Increased survival of a male's mate may result in increased productivity of that male in subsequent years.
How ever, t his does not appear to explain t hese brood care dif ferences bet w een Willets and M arbled Godw its,
since local survival rates of male and female Willets are similar, while low er in female godw its compared to
males . In add it ion , m ate fidel it y is gr eater and ' div orc e' (both members o f the pai r return but do n ot reun it e)
rates lower in godw its than Willets. To some extent t he differences may be phylogenetic: in most bi-parentally
inc ubat ing Calid rid sandpip ers f emales deser t bro ods early , w hil e many Trin gid ae spec ies show a mix ed
resp onse, w it h Numenini (inclu din g go dw it s) apparen tly demonstrat ing lit tle sex -bias in bro od c are.

335     Stanback, Seifert, Tuscano & Riley
Eastern Bluebirds as environmental indicators on North Carolina golf courses. MARK STANBACK* , MEG
SEIFERT, AMBROSE TUSCANO and CARRIE RILEY, Biol. Dept ., Davi dson Coll., Dav idson, NC.
         Analysis of reproductive parameters of birds nesting on golf courses can provide valuable information
on the biological eff ects of golf er and turf maintenance activities on course wildlif e. Because Eastern Bluebirds
nesting on fairw ays feed their young on insects collected primarily on managed turf, they may be especially
vulnerable to t he chemicals used there. For 3 seasons (1999 - 200 1) w e monitored 100 nestbox es in control
habi tat (hay field s and past ures ) and 20 0 o n go lf cou rses near Davidson, NC f or f irs t egg d ate, c lut ch size,
brood size, and n estling condit ion. Habit at (golf vs. non-golf ) per se had no signif icant ef fect s on t hese
parameters, although there w ere some significant interactions w ith ot her factors (year, nest date, etc). Overall
our dat a suggest t hat gol f c ourses prov ide adequate habitat for Eastern Bluebird s.

336     Carey
Decline in rate of extra-pair fertilizations with successional changes in a population of Field Sparrows.
MICHAEL CAREY, Dept. Biol., Univ. Scranton, Scranton, PA.
        Since 1987, breeding biology in a population of Field Sparrows in ne. Pennsylvania has been closely
monitored. A s part of t his long-term study, f rom 19 90 t hrough 200 1, blood samples have been collected from
all breeding adults and 5-day old nestlings in the population. DNA ext racted from the blood has been used to
determine genetic parentage of 8 92 nestlings (19 90 - 2 000 breeding seasons). 88 nestlings (9.9 %) w ere the
result of extra-pair fertilizations (EPFs). Over the course of t he study, t he habitat on t he site has changed from
open, brushy fields at the onset to areas that are now heavily covered by small-medium trees and shrubs.
Along wit h these successional changes in habitat structure, the number breeding territories has declined from a
maximum of 3 7 in 19 91 t o 10 in bot h 199 9 and 200 0. A s the breeding population has declined, so has the
frequency of EPFs: maximum annual EPF rate was 25 .0% in 199 0; minimum 0 % 1 998 and 2000 . Possible
underlying ultimate causes of this decline in EPF frequency w ith t he decline in breeding population size will be
337     Hatch
Repeatability of provisioning rates in a population of House Sparrows. MARGRET I. HATCH, School Biol. Sci.,
Univ. Kent ucky, Lexington, KY.
         Measuring the repeatability of a behavior serves at least 2 purposes (Lessells & Boag 1987, Auk 104:
11 6-1 21 ). The f irst is to determin e the reliabilit y of mult iple observat ions on 1 indiv idual. Th e second is to
determine if the behavior is a consistent trait of individuals, wit h greater variance betw een rather than wit hin
individuals. I examined the repeatability of nestling provisioning rates by male and female House Sparrows
breeding on a farm in Lexington, Kentucky. Trips to t he nest/nestling w ere highly correlated betw een morning
and aft ernoon observations, but not betw een observations at early and middle nestling stages. Number of trips
w as also not repeatable wit hin a season for males or females, but average provisioning rate was significantly
repeatable betw een years for females. Potential explanations for this difference will be discussed.

338     Keane, Fraser & Buckley
Effects of gulls on Piping Plover nest site selection at Monomoy National Wildlife Refuge, Massachusetts.
SHANNON E. KEANE* , JAMES D. FRASER, Dept. Fish. & Wildl. Sci., Virginia Tech, Blacksburg VA, and P. A.
BUCKLEY, U.S.G.S.-Patuxent Wildl. Res. Center, Grad. School Ocean., Univ. Rhode Island, Narragansett, RI.
          We examined the effects of Herring and Great Black-backed Gulls on Piping Plover nest site selection on
South Monomoy Island, MA, f rom 19 98 t o 200 0. We compared Piping Plover behavior and nest site selection
in a gull-free area to a gull area, and compared Piping Plover nesting area characteristics to areas not used by
plo ver s. We f oun d no dif ferenc e in t he f requ ency of disturb ance b y gull s t o pr e-nesting adul t plo ver s bet w een
the tw o areas. We found few er gulls near pre-nesting adults than near random points, and few er gulls in Piping
Plover nesting areas than in areas not used by plovers. Proximity to prime f oraging habitats and available
nesting habitat (wide stretches of open vegetation) may be more important t o Piping Plover nest site selection
than t he presence of gul ls.

339     Dodici, Koford & Zenner
Habitat selectio n by M allards nesting in nort hern Iow a. GIAN DODICI* , Dept. Animal Ecol., Iowa State Univ.,
Ames, IA ; ROLF R. KOFORD, USGS-BRD, Iowa Coop. Unit, Dept. Animal Ecol., and GUY ZENNER, Iow a
D.N.R., Clear Lake, IA .
          Mallards are attracted to breeding habitat in t he Prairie Pothole and Parkland regions of the n. U.S. and
Canada because of t heir high density of w etlands. Intensive agriculture in Iow a has resulted in a reduction in
the number of w etlands and in the amount of upland habitat available to breeding mallards. Habitat selection
occurs in a hierarchical fashion, with f irst order selection of a geographic landscape, then selection of home
range and third order selection of components w ithin a home range. Waterfow l habitat-use studies typically
investigate third order components of habitat selection without considering the second-order habitat matrix.
We radio-tracked female mallards during the breeding season in 199 8-20 00 t o investigate habitat selection by
female mallards in a 124 km 2 st udy area in n. Iow a. A t the f irs t-or der, females pref erent iall y nest ed in lin ear
right-of-w ay habitat over other types of nesting habitat, although nest success is lower in linear habitat then in
non -linear h abit at. W e w ill also pres ent habi tat selection dat a at dif ferent spat ial scales and i nclud e habi tat
spec if ic n est suc cess and f emale s urv ival dat a.

340     Welstead & Krannit z
The influence of tree encroachment on predation rat es of artif icial and real Brew er' s Sparrow nests. KYM E.
WELSTEAD* , CACB, Uni v. British Columbia, Vancouver, BC, and PAM G. KRANNITZ, Canadian Wildlife
Service, Delt a, BC.
         Fire su ppr ession an d in creased precipi tation in t he in terio r of s. Brit ish Colu mbia has resu lt ed in an
expansion of Ponderosa pine and Douglas-fir into shrub-steppe habitat. Brewer's Sparrow is a provincially
red-listed species that maybe adversely affect ed by increased density of trees. Predation is the predominant
cause of nest f ailure in Brew er's Sparrows in t he South Okanagan Valley and w e hypothesised that encroaching
trees provide perch and nesting sites for avian predators. We predicted that nests near trees would be more
likely to be preyed upon than nests far away from trees. To test this, an experiment was conducted by w iring
realistic grass nests cont aining tw o clay eggs that mimicked Brewer' s Sparrow nests into shrubs along a
dist ance gradient (2 5 m i ncrement s) fro m pot ential perch sites (con ifer t rees). Real Brewer' s Sparrow nests
w ere monitored to validate the results from artif icial nests, in terms of predation rates and patterns. Preliminary
resu lt s show ed a signi fican t decr ease in pred ation of art if ici al nests w it h in creasing di st ance f rom trees. Real
nests reflected the same trend but the relationship was not significant. Unless land management reverses the
trend , c ont inu ing tree enc roac hment w ill make it inc reasingl y dif ficu lt for g rassland bir ds t o select nest ing sit es
that are far from t rees. This w ill potentially increase nest predation and have detrimental consequences for
already declining grassland bird populations

341      Thompson
Female Blue Tit mass loss du ring t he breeding cyc le: Testi ng t hree hypot heses. CHARLES F. THOMPSON,
Dept. Biol. Sci. , Illinois St ate Univ., Normal, IL.
         Female passerines often lose body mass during the course of t he breeding cycle, w ith particularly rapid
mass loss occurring during the first f ew days after t heir eggs hatch. I used a food-supplementation experiment
on a Blue Tit (Parus caeruleus) population breeding at Loch Lomond, Scotland, to test 3 hypotheses that have
been proposed to explain this mass loss. The (i) reserve-mobilization, (ii) energy-deficit, and (iii) w ing-loading
hy pot heses each m ake di fferent pred ict ion s dependi ng o n w hen supp lemen tal f ood is prov ided to t he f emale,
w hich in this experiment was at the beginning or end of the incubation period. The results of the experiment
are consist ent w ith the w ing-loading hypot hesis.

342      Hastings
Effects of prescribed burns and mechanical treatment on secondary cavity nesting songbirds. ROBERT V.
HASTINGS, Environ . St ud. Dept., Prescot t Col l., Prescot t, AZ.
         I studied the effects of prescribed burns and the mechanical removal of vegetation on nesting behavior
and success of seco ndar y c avity nest ing songbi rds in c ent ral A rizona. I co nst ruc ted 9 0 n est box es and plac ed
30 in a chipped chaparral treatment, 30 in a burned oak-ponderosa pine treatment, and 30 in an untreated,
control area. I monitored the boxes during the 200 0 nesting season to determine if these forest management
practices would have a positive eff ect on nesting behavior and nesting success. There w as no significant
dif ferenc e bet w een t he sites; the co nt rol had 9 suc cessful n est s out of 11 attempts; the ch ipp ed t reat ment had
8 successful n est s out of 10 attempts; the bu rn t reat ment had 6 suc cessful n est s out of 13 attempts. Furt her
studies are in order to determine the effects of increased management on flora and fauna if land managers are
going to continue to actively manipulate forest ecosystems.

343      Canário & Soler
Weath er con dit ions i nf luence t he Azu re-w inged M agpie' s cooperat ive br eeding sy stem. FILIPE CANÁ RIO* ,
Faculdade d e Ciê  ncias da Univer sidade de Li sboa, Por tugal ; and MANUEL SOLER, Facultad de Ciencias de la
Universidad de Granada, Spain.
         The A zure-w ing ed M agpi e (Cyanopica cyanus) has a remarkably disjunct distribut ion occurring only in
south-w estern Europe, in the Iberian Peninsula, and in Eastern Asia. This colonial species breeds cooperatively
in both areas. Helpers contribut e to nest building, feeding incubating and brooding females, feeding nestlings
and fledglings, removing faecal sacs and mobbing predators. Breeding success and provisioning rates were
studied in Barrancos, sw. Portugal, in 2 consecutive breeding seasons wit h varying climate conditions. On the
first season, w ith f ew rain periods, breeding success w as high. Few individuals acted as helpers, probably
becau se mo st bir ds w ere abl e to br eed in depen dent ly. The f ew nest s w it h hel pers did not experien ce in creased
breeding success. During second breeding season the magpies endured severe w eather with long heavy rain
peri ods. In t his season, breed ing suc cess w as low er, main ly becau se of nest lin g st arvation , althou gh h elper
presence was higher and helpers contribut ed significantly to increase the provisioning rates and breeding
suc cess. Helpi ng b ehavior in t his spec ies seems t o be a " making the best of a bad j ob" st rat egy, so t hese data
support the ecological constraints hypothesis for the evolution of cooperative breeding.

344     Payne & Payne
Begging for parental care from another species: Specialization and generalization in brood-parasitic finches.
ROBERT B. PAYNE* and LAURA L. PAYNE, Mus. Zool. and Dept. Biol., Univ. Michigan, Ann Arbor, MI.
         Af rican indi gobird s (Vidua, 10 species) are species-specific brood parasites on host estrildid finches.
Alt hough the mouth patt erns of nestlings mimic those of their host nestlings, the nestling begging calls are not
host specific, and in only some indigobird species are the begging calls like those of t heir host. Adult male
indigobirds mimic calls and songs of their host species, and mimicry song has tw o kinds of begging calls, one
an in nat e call lik e that used by young bir ds in t he car e of their fost er par ent s; and t he ot her a c all l earned w hen
males imprint on t heir foster species then mimic the begging calls in their song. We recorded young and adult
indigobirds in the field, and begging calls of young and adult song mimics that w e reared under alternative
foster species. The innate begging call in all 10 indigobird species matches that of only certain (3) firefinch
Lagonosticta host species, even in indigobirds that normally parasitize other hosts. This innate call is used in
parental care. Both kinds of begging calls (innate and learned imitative) are used in mate choice, and the
host -specifi c begging c alls in th e mimetic songs of adult m ale indigobird s prov ide cues for females to assess a
mate. T he innate begging call prov ides a clue (mtDNA sequences give anoth er) to t he ancient parasit e-host
species association s.

345      Armour
Nest success of upl and n est ing duc ks an d A merican Bit terns in a w aterf ow l managem ent area. CASEY L.
ARMOUR, Biol. Dept., Univ. North Dakota, Grand Forks, ND.
         Nest success of upland nesting ducks and American bitterns w as measured over a 3 yr period. The
study w as performed in an area managed to promote w aterfow l use. Management regimes included burning,
grazing, m ow ing and allow ing f ields to grow (unmanipul ated). A tot al of 1 03 duck nest s and 42 bit tern nest s
w ere found from 19 98 - 2 000 . Mallards and Blue-wing Teal were the predominate upland nesting ducks in the
area. Nest success was calculated using Mayfield methods. Nest success of w aterfowl ranged from 8% to
52 %. Nest suc cess of bit terns rang ed f rom 50 % to 1 00 %. Both t eals and mallar ds utili zed f ield s t he f irs t year
aft er manipulat ion. Bit terns pr eferred dense nestin g cov er w hich inc luded unmanip ulated areas and fields t w o
years aft er manipulation. Alt hough American Bitt erns at t his site experienced high levels of nest success, this
study revealed several areas where current wat erfow l management practices could be modified to benefit
bitterns nesting in the same area. By waiting to mow fields until the end of Aug it allows time for bitt ern chicks
to f ledge. Allow ing some areas to grow for several years provides the dense nesting cover preferred by
bitt erns. Using a variety of management practices can provide habitat t hat can be used by both w aterfow l and
Am erican Bitt erns.

346      Pratt
Social monogamy in the Kalij Pheasant: Male parental care, cooperative breeding, and skewed sex ratios.
THANE K. PRATT, USGS Pacifi c Is. Ecosystems Research Center, Volc ano, HI.
        Kalij Pheasants (Lophura leucomelanos) belong to a large genus, the gallopheasants, w hich is sexually
dichromatic and currently regarded as polygynous. Native to the Himalayas, Kalij were introduced in 1962 to
the Island of Hawaii, w here they are now common. I studied the social system of Kalij Pheasants by recording
the age and sex composition of groups in and around Hawaii Volcanoes National Park and by daily observing
individually identified Kalij in my neighborhood. I found no evidence for polygyny or promiscuity . Instead,
socially monogamous pairs stayed together year round for one or more years. Intolerance of ot her Kalij by both
sexes and site fidelity of pairs suggested that breeding birds may be territorial. Males constantly att ended their
mate and her brood until t he chicks reached adult size. Males provided parental care by feeding and brooding
chick s, guardi ng t hem fr om predat ors, t ending t hem w hen their m oth er w as temporaril y absent, and most
importantly by driving conspecifics from f ood sources. Breeding w as seasonal w ith chicks evident from J un
through Sep. By Oct, young resembled their parents. Female chicks voluntarily dispersed in Oct - Nov, w hile
male chicks remained w ith their p arents unt il driv en away by t heir fat hers in M ar - Apr. In a few famili es,
fathers expelled certain sons but retained a few ot hers, which remained tightly bonded to t heir parents and
helped raise the next brood. Sex ratios of broods and all adults w ere both highly skew ed tow ards males. In
additi on t o Kalij, ot her lesser-know n gallopheasants may be socially m onogamous.

347     Chu, Koenig, Godinez, M cIntosh & Fleischer
Social and genetic m onogamy in t errit orial and loosely colonial po pulati ons of the Phainopepla. M IYOKO CHU,
WALTER D. KOENIG, Mus. Vert. Zo ol. and Dep t. Int egrat ive Biol. , Un iv. Calif orn ia, Berkeley , CA ; ALVARO
GODINEZ, CARL E. McINTOSH and ROBERT C. FLEISCHER* , Molec. Genet. Lab., Smithsonian Inst.,
Wash ington, DC.
         We obser ved cou rt ship and cop ulat ory behav ior in 2 pop ulat ion s of Phaino pepl as w it h co nt rast ing soc ial
systems and used multilocus DNA f ingerprinting to assess rates of extra-pair fertilization (EPF). No extra-pair
copulations were observed in either population. Copulations were infrequent (< 2 t imes/d) and mate-guarding
w as not observ ed. We f oun d no evidenc e of EPFs in 4 8 n est lin gs f rom 25 nest s. Thu s, Phaino pepl as appear
to be both socially and genetically monogamous. This result is surprising because Phainopeplas are sexually
dim orp hic and b reed s yn chron ously, c harac teristics associat ed w it h hi gh EPF rates in ot her s peci es. Ecologic al
factors experienced by Phainopeplas, such as compressed breeding seasons, geographically shift ing food
resources, and widespread breeding failures, may favor monogamy during any given breeding attempt w hile
allow ing oppo rtu nit ies to sw itc h mates bet w een breeding localit ies and years.

348     Stai
Patterns of spatial use and their role in the mating system of wild Muscovy ducks. SARAH M. STAI, Dept.
Biol., Univ. Miami, Coral Gables, FL.
            Behavioral observat ions have indi cated t hat t he social mati ng syst em of w ild M uscov y duc ks (Cairina
mosch ata) is best classified as promiscuity. Data from the Pantanal of Brazil demonstrate that male-female
pairs do not predominate in the study population and that associations betw een known males and females are
inconsistent through t ime and not exclusive. The absence of a male-female pair bond, w hich is so predominant
in most other waterfowl (Anatidae), suggests that another mechanism for encountering mates must exist.
Results on spatial use indicate that, although males are not strict ly territorial, some males maintain areas of
“ dominion” w ithin t heir home range, i.e., areas w here they are aggressive to, and dominant over, intruding
males . Ot her m ales adopt a st rat egy of “ float ing ,” i. e., visit ing var iou s dominio ns, w here t hey are somet imes
tolerated, as well as spending time in unoccupied areas. Female home ranges overlap male home ranges, w ith
som e females observ ed in mult ipl e dom ini ons and t hus in assoc iat ion w it h mult ipl e males , an d ot her f emales
observed in only one dominion. Therefore it appears that males encounter mates either by seeking them out or
attracting t hem to dominions, and females encounter mates by visiting mult iple dominions (to compare mates or
ut ili ze reso urc es) or by goi ng d irec tly to t he do minio n of a kno w n male.

349      Kosciuch
* A simulation of the population dynamics and colony growt h of Cliff Swallows. KARL L. KOSCIUCH, Dept.
Wildl. & Fish. Sci., Texas A&M Univ., College Station, TX.
         Clif f Sw allo w s are c olo nial bir ds nest ing in c ulv ert s and und er br idg es t hro ugh out the U. S. I ndi vidual
colonies are exposed to different factors that create significantly different reproductive success among colonies.
Danchin et al. (1999, Ecology 79 : 2 41 5-24 28 ) hav e hypot hesized (t he co nspecif ic- perf orm ance b ased
attraction hy pothesis: CBAH) that birds w ill assess annual reproductive success (ARS) of conspecifics and will
nest the following year in those areas where reproductive success was collectively highest. In this poster, I
describe a model of Cliff Sw allow reproduc tio n and use the mo del to si mulat e populati on dyn amics in respon se
to colony choice by sw allows under the CBAH. The model consists of 4 submodels. Each submodel represents
a colony of Cliff Swallow s. Tw o submodels are free of predators, a snake is present in one and fire ants are
present in t he other colony. Simulation results indicate that severity of predation and mortality af fect A RS of
the di fferent Clif f Sw allo w col oni es. Whether o r no t bir ds c hoo se t o ch ange c olo nies depen ds on sev eral
factors. Predation and mortality influence colony choice but more importantly minimum ARS for return aff ects
colony size. The model reveals a sensitivity t hreshold for colony choice based on minimum ARS. We know
that birds choose colonies based on highest overall ARS but what criteria they use to determine when to move
are unknown. Understanding the ARS threshold may lend insight int o how birds choose colonies in globally
versus loc ally poor or good years.

350     Anders & Horne
Relationships betw een habitat characteristics and demographic parameters of Golden-cheeked Warblers.
ANGELA D. ANDERS* and JON S. HORNE, The Nature Conservancy, Fort Hood, TX.
           Underst andi ng t he rel ation ship bet w een avian d emograp hic param eters an d hab it at char act eristics can
help to foc us conserv ation ef for ts on high-qualit y habit ats. I n th is stu dy, w e quantif ied pairing suc cess,
proportion of territories producing fledglings, and age of territorial male Golden-cheeked Warblers on Fort Hood,
Texas. For each territory, w e measured habitat characteristics including canopy height; tree density, diameter,
and species composition; vegetation cover at 5 height intervals; and indices of Ashe juniper (Ju nip erus ashei)
maturity . We assessed the relationship between demographic and vegetation parameters using multiple logistic
regr ession. Rates o f pair ing suc cess and p rod uctivit y in ou r st udy pop ulat ion w ere hi gh, w it h 9 5. 0% of males
pair ed, and 9 2. 6% of pair s pr odu cin g f ledg lin gs. Such hi gh r ates of pair ing suc cess and p rod uctivit y preclud ed
iden tif icat ion of habi tat char act eristics relat ed t o t hese 2 var iabl es. How ever , w e did iden tif y habit at
characteristics associated w ith male age. Territories of after-second-year males contained greater oak (Quercus
spp.) cov er above 2 m and a greater number of l arge-diameter junipers than did second-year male territ ories. If
w e assume t hat old er males out compete young er males f or p ref erred br eedin g hab it at, t hese r esults indic ate
that a high percentage of oak cover and the presence of large junipers are associated wit h high-quality breeding
habi tat. Fut ure s tudi es of forag ing suc cess and b ody con dit ion of Gold en-c heeked Warbler s in dif ferin g hab it at
ty pes may help t o fu rth er quantif y habit at qualit y f or t his species.

351     Dearborn & Ryan
A test of the Darwin-Fisher theory for the evolution of male secondary sexual traits in monogamous birds.
DONALD C. DEARBORN* , Dept. Biol., Sout hw estern Univ., Georgetow n, TX; and MICHAEL J. RYAN, Sec.
Integrative Biol., Univ. Texas, Austin, TX.
        The Darw in-Fisher hypothesis proposes that the presence of male secondary sexual traits in
monogamo us birds is select ed for b y early-season breeding of females th at are in good c ondit ion. T hese
early-breeding females have high fecundity because of t heir good condition, and they select mates based on
secondary sex traits. We t ested whether this hypot hesis may be responsible for male sexual ornaments in the
Great Frigatebir d, a soc iall y and gen etical ly monog amous seabir d. Consist ent w it h t he Darw in-Fisher
hypothesis, we found that reproductive success declined over the season. However, males wit h more
exaggerat ed or nament s w ere no t cho sen as mat es earlier i n t he seaso n t han w ere males w it h less exagger ated
ornaments, and selection gradients on these ornaments w ere not significantly diff erent from zero.

352     Peak
* An experiment al test of the nest co ncealment hypot hesis wi th t he American Goldfin ch. REBECCA G. PEAK,
Dept. Fish. & Wild l. Sc i., Univ . M issouri, Columbia, M O.
        Predation is a major cause of nest failure for birds that build open-cupped nests and many studies have
inv est igat ed t he ef fect of nest con cealm ent on n est pred ation . The res ult s of resear ch exam ini ng n atural
variation in nest concealment are inconclusive; some studies have implicated poor concealment in high
predation rates w hile other investigators have failed to link nest concealment and nest predation. M anipulation
of t he amount of v egetatio n surrou nding nest s is one approach t o experiment ally t est t he eff ect of nest
concealmen t on n est pred ation . I con duc ted a veget ation remov al experi ment in c ent ral I lli noi s using American
Goldfinch nests to test t he hypothesis that poorly concealed nests experience higher predation rates than w ell
concealed nests. Eligibility of n ests f or experim ental manipul ation w as determined usin g nest v isibilit y scor es,
stage of nest cycle, and proximity t o other active nests. Eligible nests w ere paired and randomly assigned to
manipulated or control t reatments. Manipulated nests (n = 34) w ere on average 70% more visible than control
nest s (n = 34 ). All n est s w ere moni tored unt il f ledg ing or f ailu re oc curred, but onl y t hose kno w n t o hav e failed
due to predation were used in the analyses. Daily predation rate was 0.0454 for manipulated nests and 0.0403
for cont rol nest s. Predation rat es did not dif fer bet w een treatment s (P 2 = 0. 10 , 2 df, P = 0. 75 ) or years ( P 2 =
0.2 0, 2 df, P = 0.6 5). Logistic regression models showed that nests active later in the breeding season had a
low er pr obab ili ty of nest pred ation . These ex peri mental result s indic ate that nest con cealm ent is not an
import ant f actor in determ ining pr edation of Am erican Goldfi nch nest s in cent ral Illinois.

353     O'Brien & Hau
Food stim ulatio n alone can induce gon adal grow th i n tr opical Spot ted A ntb irds. SARA O' BRIEN* , Dept. Biol.,
Univ. Il linois, Urbana-Champaign, IL; and MICHAELA HAU, Dept. Ecol. & Evol. Biol., Princeton Univ., Princeton,
           Many organisms regulate the timing of lif e history events, like reproduction, in accordance w ith t heir
enviro nment using en viron mental cu es. In seasonal br eeders, l ike m ost temperat e bir ds, pho toper iod act s as
the primary cue to initiate gonadal development. Only aft er reproductive activit y has been initiated by
photoperiodic cues can additional cues like food or social stimuli influence rate of gonadal growt h. Some
tro pical species like Spot ted A ntb irds (Hylop hy lax n. naevioi des) are apparently seasonal breeders and can use
photoperiod to initiat e gonadal growt h. Subsequently, these birds can then respond to an increase in food
qual it y and qu ant it y t o reg ulat e rat e of gon adal d evelopm ent . Using cap tive Spot ted A nt bir ds w e test ed
w het her t hey are t rue s easonal br eeders by det ermini ng t heir repr odu ct ive resp onse to f ood st imulat ion alon e.
Birds with f ully regressed gonads w ere separated into 3 groups: phot o/food st imulated, food stimulated and
control. The food stimulation consisted of an increase in food quantity and quality while remaining on a 12:12
light :dark (LD) schedule. The phot o/f ood st imulat ion con sisted of an increase in daylengt h by 1 hr (13 :1 1 LD)
w hil e also main taini ng an inc rease in f ood quan tit y and qu alit y. The c ont rol gro up w as kep t on a 1 2: 12 LD
schedule w ith adequate, but not abundant f ood. Wit hin 3 w k gonad size among the photo/f ood and food
stim ulated gro ups signif icantl y changed, but not in t he cont rol gro up. Th ese results suggest that in cont rast t o
st ric t seasonal b reeder s, Spot ted A nt bir ds c an in it iat e gon adal g row th in resp onse to f ood st imulat ion alon e,
refining our current view of t heir seasonal organization. Future work should elucidate the neurochemical and
neurophysiological mechanisms by w hich a food cue is perceived and incorporated to influence changes in life
history events.

354      Greenberg, Pravosudov, Sterling, Kozlenko & Kontorshchikov
The ef fect of the agr icu lt ural rev olu tion on f ores t bir d co mmuni ties: The c ase of the Chaf finc h. RUSSELL
GREENBERG* Smith sonian M igrato ry Cent er, Washing ton , DC, VLADIMIR PRAVOSUDOV Dept. Zool., Univ.
Calif if orn ia, Davi s, CA, JOHN STERLING, Jones and Stokes, Sac ramen to, CA, ANNA KOZLENKO, Cleveland
Height s, OH and VITALLY KONTORSHCHIKOV, Moscow Russia.
           We describe t w o major pat terns of abundance of dif ferent ty pes of c anopy f oliage-gleaning birds acr oss
the Nearctic and Palearctic boreal forest. First, because of t he extreme numerical dominance and breadth of
habitat use of the Chaffinch (Fringilla coelebs), the European Russian sites stand as outliers to t he relationship
bet w een t he nu mber of spec ies and t he abu ndan ce of foli age-gleani ng b ird s. The Ch affinc h is one o f the f ew
canopy foliage-gleaning birds in temperate forests t hat is abundant in agricultural habitats during the
non-breeding season, w hich probably allowed a rapid expansion of it s range and abundance w ith t he advent of
farming in Europe. Outside of Europe, the proportional abundance and the number of species of w arblers or
w arbl er-like b ird s w ere bo th po sit ively relat ed t o ov erall abun danc e of foli age gl eaners. This suggest s t hat
warblers prosper when there is high resource productivity during the breeding season. Fringilla may be
func tion ally equi valent to w arbl ers an d hav e essent iall y r eplac ed w arbl ers t o ach ieve their hig h abu ndan ce.

355      Poirier
Mate switching in House Wrens. NICOLE E. POIRIER, Dept . Biol. Sci., Univ. Wisco nsin-M ilw aukee, Milw aukee,
           We invest igated t he causes of w ith in-season mate sw itc hing in t he house w ren. Over t he 19 98 - 20 00
breeding seasons, 53% of males switc hed mates. In this study , the possible influence of f irst brood failure on
mat e sw it ching w as co nt rol led f or s inc e nest box es w ere pr ot ect ed by pred ator g uard s and depr edat ion w as
prevented. A study of House Wrens in Minnesota found that males decreased the interval betw een first and
seco nd b roo ds w hen t hey sw it ched mates. Males t hat sw it ched mates began t heir seco nd b roo ds 5 .6 d earl ier
than w rens that st ayed w it h t he sam e part ner. In c ont rast , w e foun d t hat sequent iall y m ono gamous males that
switched mates began their second nest an average of 4.7 d later than males that stayed with the same female.
This suggests that reduced inter-nest interval is not a benefit t o males that sw itch mates in our population of
House Wr ens. Males m ay benef it from mat e-sw it ching if their mat e is in po or c ond it ion and n est lin g surv ival at
the f irs t bro od i s co mprom ised, o r t heir pat erni ty in t he f irs t bro od i s red uced. Males m ay be more li kely to st ay
w ith t he same mate when they have high paternity in t he first brood, w hereas males w ith low paternity in the
first br ood may be more likely to sw itch mates. Paternity w ill be assessed using 4 polymorphic microsatelle

356      Lloyd
* Proximate causes of life history variation in Chestnut-collared Longspurs. JOHN D. LLOYD, Mont ana Coop.
Wildl. Res. Unit, Univ. Mont ana, Missoula, MT.
         Populations of migrat ory birds that breed in different habitats oft en show phenotypic v ariation in life
history t raits. Determining the causes of habitat-specific life history v ariation w ill enhance our understanding of
the factors t hat shape life-history phenotypes, and w ill also help us to understand how modif ication of t he
environment can change the selective pressures faced by individuals. I found substantial variation in life history
traits betw een Chestnut -collared Longspurs breeding in native mixed-grass prairie and those breeding in
monoc ult ures of t he intr oduced grass Agropyron crist atum. Longspurs in the exotic habitat began nesting 10 d
later than longspurs in native prairie, and laid significantly smaller clutches. Predation rates w ere high but
similar in both habitats. Despite clutch size diff erences, among all nests the mean number of young
fledged/nest w as similar. Among successful nests, longspurs in the exotic habitat fledged significantly more
young, suggesting higher nestling attrition rates in native prairie. Size at fledging was similar betw een habitats.
Variation in predation rates do not appear to be a proximate cause of life history variation, but the biomass of
potential prey w as lower in the exotic habitat. Dif ferences in insect biomass may explain life history variation,
but the rel ation ship bet w een f ood availabi lit y and co mponen ts of fit ness is c omplex , an d may int eract w it h age-
or condition-specific patterns of habitat selection.

357      Rotella, Clark & Afton
Size-r elat ed rep rod uctive and surv ival pat terns in Lesser Sc aup. JA Y J. ROTELLA , Ecol. Dept., Montana State
Univ., Bozeman, MT; ROBERT G. CLA RK* , Can. W ildl . Serv ., Saskat oon, SK; and A LAN D. A FTON,
USGS-Coop. Fish W ild l. Res. Un it , Lo uisiana St ate Univ. , Bat on Rou ge, LA.
          We ex amin ed rep rod uctive and surv ival pat terns of ind ividu ally -mar ked f emale Lesser Sc aup; hereaf ter
scaup) from 2 study sit es for 5 and 10 y r, respectively. Larger female scaup produced larger eggs, and
duc kli ngs from larg er egg s surv ived bet ter t han t hose hat ching from smal ler eg gs, so w e test ed w het her
selective mort alit y ac ts mai nly in f avor of smal ler f emale s ize, cou nt eracting repr odu ct ive advant ages ass ociat ed
w ith larger female size. Survival w as estimated using mark-recapture methods. Structurally larger adult
females generally had lower survival than smaller individuals and heavier juveniles had lower survival than did
lig ht er ju veniles; t he lat ter f ind ing con trast s result s of several i nt ersp ecif ic s tudi es of bod y m ass-surv ival
relationships. Female survival also declined with indices of increasing breeding effort on both study sites.
Consequently, t he cost of breeding might explain age-related patterns of nesting propensity in scaup. Overall,
results suggest that the advantage of large body size during breeding may be offset by selection favoring
sur vival of smal ler ad ult females and p ossibl y juv enil es. Thi s maj or f ind ing reaf firm s pr evious con tent ion s t hat
analyses of interspecific patterns may incorrectly portray int raspecific relationships between body size and
surviv al in birds.

358      Staplet on
* Consistency of female mate choice. MARY K. STAPLETON, Dept. Biol., Queen's Un iv, Kingst on, ON.
        It is oft en assumed in studies of female mate choice that f emale preference for certain males is
repeatable. Repeatability (r) is a measure of t he wit hin individual-consistency of a trait, how ever, the
repeat abil it y of mat e cho ice h as been exam ined in f ew avian species and t he res ult s are equiv ocal. Tree
Swallow s have the highest level of extra-pair mating among North Am erican birds. I tested w hether female
Tree Swallow s w ere consistent in choosing extra-pair males wit hin a single breeding season by experimentally
inducing them to renest and, thus, choose w hether to remate wit h the same or diff erent males. Extra-pair
young in the first and replacement broods w ere identified using parentage analysis based on microsatellite
DNA. Extra-pair yo ung w ere present in 8 5% of b roods w ith 74 % o f f emales producing young fro m at least
tw o extra-pair sires. Females showed significant repeatability of proport ion of extra-pair young in both broods
and 83% of experimental females were consistent in whether they produced extra-pair young in both broods.
Repeatability of particular extra-pair sires will also be addressed.

359      Basile & Islam
Habitat selectio n and territ ory size of Cerulean Warblers in southern Indi ana. CYNTHIA BASILE* and KAM AL
ISLA M, Dept. Biol., Ball Stat e Univ., Munc ie IN.
        In Indiana, the Cerulean Warbler is listed as a "species of special concern," suffering from extensive
loss of breed ing habi tat. Co nser vat ion and m anagem ent effort s directed t ow ard p rot ect ing fores ted lan dsc apes
and identi fy ing habit at requir ements on t he breeding groun ds are paramount . Yet, lit tle quant itat ive data exist s
on the important vegetative components necessary for successful breeding. We will present a preliminary
report on the Geographic Information System analysis of landscape features associated with 23 male territories.
Territories were monitored during the summer of 2000 in Yellowwood and Morgan-Monroe State Forests. As
expected, canopy gaps were present in all territories, and perch trees were among the tallest in each territory.
How ever, con trary to c urrent lit erature, t his species w as located almost exclusiv ely along rid ges.

360      Owen-Ashley, Meddle, Richardson & W ingfield
Ejaculate quality declines when copulation rat es are high in the polygynandrous Smit h' s Longspur. NOAH T.
OWEN-ASHLEY* , Dept. Zool., Univ. Washington, Seattle, WA, SIMONE L. MEDDLE, Dept. Biomed. Sci., Univ.
Edinburgh Med. Sch., Edenburgh, UK, MATTHEW I. RICHARDSON and JOHN C. WINGFIELD, Dept. Zool.,
Univ. Washington.
         We investigated whether ejaculate quality varies in relation to breeding stage in a free-living population
of the po lygy nand rou s Smi th' s Longspur, an arc tic- breed ing passerin e w it h on e of the hi ghest cop ulat ion rat es
documented in birds. Sperm samples w ere collected by cloacal manipulation f rom breeding males at Toolik
Lake, AK (1999 , 20 00). Three measures were used to assess ejaculate quality: (1) volume, (2) sperm
concentration, and (3) tot al number of sperm/ejaculate. We hypothesized that sperm quality w ould decline
w hen c opu lat ion frequ ency is high . Ejaculat e volum e and size, but not con cent rat ion , w ere signi fican tly low er
in m ales capt ured dur ing the f emale f ert ile p erio d, w hen c opu lat ion rat es are high , c ompared to males capt ured
before female arrival on the breeding grounds, when no copulations occur. This suggests that frequent
copulations, rather than a decrease in sperm production, contribut e to low sperm quality during t he female
fert ile period. Even t hough select ion f avors large ejaculate sizes and sperm reserves in po lygy nandrous species,
frequent matings to maximize paternity are costly because they lower ejaculate quality.

361     Nuechterlein, Buitron, Sachs & Hughes
From solit ary t o semi-colon ial: Nestin g dispersion i n Red-necked Grebes. GARY L. NUECHTERLEIN* ,
DEBORAH BUITRON, JOEL L. SACHS and COLIN R. HUGHES, Dept. Biol., Univ. Nort h Dakota, Fargo, ND.
         We studied a unique breeding population of 140 -160 Red-necked Grebes located on Lake Osakis in
cent ral M inn esot a. A lon g t he shorel ine o f this lake, many pairs est abli shed tradi tion al, territ ori es spaced o ver
100 m apart, w hile others established semi-colonial nests on large, floating cat tail mats that had broken off
from t he shoreline and become anchored over deeper w ater. We examined the spacing, timing, and fates of
nest s est abli shed on m at ver sus shoreli ne nests. Alt hou gh p airs of ten ret urn ed t o t he sam e area year af ter
year, blood analysis showed no tendency for neighbors to be closely related. The mats appeared to provide
ideal, but t emporally and spatially limited nesting habitat, t hat w as available for nest establishment earlier in the
season t han t he emergen t shoreli ne veget ation . Duri ng m ost year s, pair s t hat est abli shed nest s on the mats
hat ched earl ier an d had larg er clut ch and b roo d sizes. How ever , t he ov erall suc cess of the t w o st rat egies
varied from year to year, depending on the nesting habitat available, w ater levels, timing of w ind-storms and
pred ator ac tivit y. Thei r beh avior f lexibi lit y m ay allow Red-necked Gr ebes t o use unstable em ergen t habi tats
that are in constant flux f rom year to year.

362      Dufty & Teall
Interspecific, but not intraspecific, egg rejection by Yellow -headed Blackbirds. ALFRED M. DUFTY, Jr.* , and
MONICA TEALL, Dept. Biol., Boise State Univ. , Boise, ID.
          Yellow-headed Blackbirds typically are not parasitized by Brown-headed Cowbirds. How ever, one
population near Boise had approximately 20% of its nests parasitized in each of 2 yr, and ejected foreign eggs.
We examined 4 additional marshes near Boise during egg laying to determine the extent of brood parasitism of
Yellow-headed Blackbirds. In addition, at some nests a host egg w as replaced w ith an artificial cow bird egg to
sim ulat e int ersp ecif ic b roo d par asit ism . A t ot hers , sing le ho st eggs w ere sw it ched bet w een nests t o simulat e
intraspecific parasitism. At control nests 1 host egg was removed and immediately put back into the same
nest . Egg rejec tion , d efined as rem ov al of a parasit ic eg g or aband onm ent of a parasit ized n est , w as noted ov er
5 d. A lthough w e heard cowbirds at each marsh, we observed no natural brood parasitism. None of 10
sw it ched ho st eggs and 0 /2 8 c ont rol eggs w ere reject ed. In c ont rast , 4 /2 2 c ow bir d egg s w ere reject ed, 1 at
each marsh; 2 by ejection and 2 by clut ch abandonment. Thus, although t here was no direct evidence of
interspecific brood parasitism of Yellow -headed Blackbirds in these populations, indirect evidence, the rejection
of artif icial cowbird eggs but not of transferred conspecific eggs, suggests that cow bird brood parasitism is (or
was) sufficiently common to produce and sustain rejection behavior.

363      Wasson, Ardia & W inkler
Detecting the effects of insect availability on Tree Swallow egg characteristics using new methods for
quan tif ying eg g shape and yolk s ize. MATTHEW F. WASSON* , DANIEL R. A RDIA and DAVID W. WINKLER,
Dept. Ecol. & Evol. Bio., Cornell Univ. , It haca, NY.
          We developed a method to measure egg size and shape and to non-destructiv ely estimate yolk size for
small eggs. We fabricated a dark box w ith guides to facilitate standard positioning of t he egg and a hole in the
top t o allow a digital picture to be taken from a standard distance. To estimate yolk size, w e "Candled" t he
eggs by shining a bright light directly underneath; thus giving a tw o-dimensional impression of the yolk.
Replicate yolk size estimat es were highly repeatable. Images were batch-analyzed w ith a macro w ritt en for NIH
Image softw are that measured the area of the yolk, the length and breadth of the egg, and the shape of the egg
using methods developed by F. W. Preston and D. Baker. Studies w ere conducted on Tree Swallow eggs in
w hich w e collected some eggs and measured the yolk volume directly t o calibrate the yolk size estimating
procedure. Finally this technique allow ed us to analyze the eff ects of insect availability 1 to 6 d prior to t he
laying date of an egg on its size, shape and composition.

364     McGow an & McGowan
American Crow group size as an indication of social behavior in the Maritime Provinces of Canada. KEVIN J.
McGOWAN, Dept. Ecol. & Evol. Biol., Cornell Univ. , It haca, NY; and JAY W. McGOWAN, Dryden, NY.
          American Crows are social birds that are known t o be cooperative breeders throughout much of their
range. However, helpers are unknow n in the most nort herly parts of the range, at least in the Great Plains. The
soc ial system of crow s in the M arit ime pro vinces is unk now n. We recor ded g rou p sizes of all c row s observ ed
w hile traveling through New Brunswick, Nova Scotia, Prince Edw ard Island, and Newfoundland during Aug
200 0. We compared these data to comparable data collected in central New York, w here crows are known to
liv e in ex tended famil y grou ps and hav e a hig h in cid ence o f coo perat ive breed ing . Gr oup size dif fered b etw een
New York and the M arit imes, w it h M arit ime crow gro ups slight ly smal ler. These resu lt s suggest that
cooperative breeding may be less frequent in crows in the Maritime Provinces than in other parts of the range.

365     Snyder, McGowan & Clougherty
Nest sit e selection of suburb an A merican Crow s. HA NNA H SNYDER, KEVIN J . M cGOW AN and JILL
CLOUGHERTY, Dept. Ecol. & Evol. Biol., Cornell Univ. , It haca, NY.
         American Crows are common birds in central New York and have recently moved into urban and
suburban areas to breed. They maintain a breeding territory year-round and build a new nest each year. We
attempted t o di scer n w hic h, if any, v ariab les t hey use t o select a nest sit e. W e measu red a n umber o f habi tat
characteristics at randomly selected sites in Cayuga Heights, NY, and compared these w ith the same
characteristics from kno w n nest sit es. Size of tree, amount of shr ub c ov er, and d ist ance t o t he near est hou se,
lawn or road all did not differ between nest and random sites. Only the distance from the nest tree to the next
nearest tree, and t he t yp e of nest tree it self dif fered. Act ual n est trees t ended to be c loser t o ot her t rees t han
the t rees at the ran dom sit es. Crow s pr eferred co nif ers o ver deci duo us t rees: more t han n inet y percent of over
50 0 n est s w ere in con if erou s t rees, signif ican tly dif ferent from the t rees av ailab le. Crow s seem to hav e few
requ irem ent s in a nest sit e, m ainl y a c oni ferou s t ree t o nest in w it h per haps a few trees c lose by. Su ch a bro ad
tolerance of nest sites may be why they have been successful invaders of the urban environment.

366     Christian
Nesting success, and growt h rates of natural and experimentally manipulated broods, in an Amazonian lowland
forest of Ecuador. DANIEL G. CHRISTIAN, Mus. Nat . Sc i. , Lo uisiana St ate Univ. , Bat on Rou ge, LA.
        Amazonian lowland tropical forests constit ute the majority of low land Neotropical forests yet t he
minority of nest predation studies come from t his region. Comparing tropical forest bird life histories w ith
temperate forest birds is difficult because of lack of studies in the tropics. The nests of 24 2 individuals of 85
species w ere found and monitored for nest success in a tropical forest of low land Ecuador. Growt h rates of 20
species in 9 families were recorded. Broods were manipulated in 6 species to compare nest predation and
growt h rates. Nesting success for some nest ty pes was similar to those from another Amazonian study (Oniki
1979, Biotropica 11: 60-69 ). Nest t ype and position was correlated to the level of nest predation, similar to
the pattern found in Temperate Zone species. Reduced brood nests had similar predation rates to controls.
Growt h rates of supplemental broods w ere slow er than control broods. Addit ional analyses of geographic
patterns will be provided.
367     Fletcher & Koford
Lessons learned from estimating parameters needed to determine seasonal fecundity in grassland birds.
ROBERT J. FLETCHER, Jr.* , Dept. A nimal Ecol., Iow a State Univ., Ames, IA and ROLF R. KOFORD, Iow a
Coop. Fish & W ildl. Res. Unit, USGS-BRD, Iow a State Univ., Ames, IA .
         Seasonal fecundity, or t he number of young fledged/female during the breeding season, is an important
demographic component f or population dynamics of migratory birds. A lthough t he importance of understanding
seasonal f ecun dit y is of ten no ted, in p ractice i t is rarely est imated, and m ost st udi es t hat do estimate season al
fecundity m ake strong assumptions about parameters required for estimation. W e used radio-telemetry on
breed ing female Bo bol ink s (n = 38 ) and Dickcissels (n = 25 ) t o det ermine p arameters r equi red f or s easonal
fecun dit y esti matio n. Daily surviv al rates fo r radio-marked and un marked nests w ere similar fo r bot h species,
but w ere generally quite low . We documented few Bobolinks (2 of 2 2 possible females) and Dickcissels (3 of
17 possible females) renesting in our study area. Renesting occurred at distances of 428 m and 8600 m for
Bobolinks, and 36, 52 , 12 3 m for Dickcissels. Renesting intervals (i.e., number of days between nest failure
and first egg in subsequent nest) was 9. 5 - 10 d for Bobolinks and 6 - 8 d for Dickcissels. Based on overall
nest initiation dates and radio-telemetry, bobolinks do not renest oft en, whereas Dickcissels likely renest more
frequ ent ly. W e con clu de t hat these speci es t end t o emigr ate from gras sland f ield s af ter nes t failu re and that
females do not alw ays re-nest after failure. Modeling approaches for seasonal fecundity and population grow th
rate estimation in grassland birds should incorporate variance estimates for t he likelihood of renesting and
nesting success in different fields and different landscapes.

368     Zanette, Smith, van Oort, Budden & Clinchy
Population, genetic and behavioural effects of f ood addition and predator removal: Results from a large-scale
field experiment on Song Sparrows. LIANA ZANETTE* , Dept. Zoo l., Univ . W estern Ont ario, London , ON;
Brit ish Columbi a, V anco uver, BC.
          Hundreds of laboratory studies have show n that animals balance the risk of predation against time
spent forag ing . A lt hou gh i t is int uit ively clear that suc h beh aviour al dec isions cou ld h ave popu lat ion -lev el
consequenc es in the f ield , t his has only recent ly been d ocument ed in mammals and h as nev er bef ore b een
demonstrated in birds. We are currently conducting a large-scale, long-term, 2 x2, manipulative food addition
plus mensurative predator removal experiment on 14 populations of Song Sparrow s on s. Vancouver Island and
several small islands in the Haro Strait. From Mar to A ug in 200 0 and 200 1 f ood w as to 4 of 8 sites on the
small islands, w here predators are rare or absent, and 3 of 6 sites on Vancouver Island, w here predators are
very abun dant. Data on clut ch size, egg w eight, nestlin g grow th and su rviv al are collected t o assess
population-level eff ects. In 2001 , w e began conducting ' pulsed' predator (Cooper' s Haw k) exposure
experiments of t he type more commonly conducted in the lab, w ith t he aim of directly c omparing how
anti-predator and foraging behaviours diff er betw een treatments. We are also evaluating the longer-term
consequences of these manipulations by examining the eff ects of local population turnover on the genetic
structure of our study populations.

369      Wagoner & Walters
Helper effect s on reproduct ion in t he Red-cockaded Woodpecker. A LEXA L. WA GONER* and JEFFREY R.
WALTERS, Dept. Biol., Virginia Tech. Univ., VA.
        A st atist ical analy sis w as ru n on 20 yr of repr odu ct ive dat a from rou ghl y 2 20 gro ups of Red-co ckaded
w ood peck ers in t he Nort h Caro lin a Sandh ill s. Thr oug h t he use of mult i-linear regr ession an alysis it w as
determined that helpers increased the survival of young during every stage of development. Helpers did not
influence whether a group re-nested after fledging nor did they influence the weight of nestlings at banding.
Breeding p air q uali ty, territ ory qual it y and env iro nment al f luc tuat ion s w ere all accoun ted f or w it h gen erali zed
estimating equations. The odds of nest initiation f or groups w ith 1 helper w ere 35% more than for gr oups wit h
0 helpers. Groups w ithout a helper were 2 times more likely to experience whole brood loss than groups wit h a
helper. The number of helpers in a group was the only factor t hat influenced survival of young from t he egg
st age t o ban din g at 6 - 10 d ol d. The o dds of young sur viving f rom band ing to f ledg ing in g rou ps w it h 1 help er
w ere 1.5 times the odds of surviv al in groups with 0 helpers. Unaffiliated birds, termed floaters, decreased the
survival of young between the ages of fledging and 1 yr.

370      Grecian
A field technique to estimate concealed razorbill nests. V. DEDREIC GRECIAN, At lantic Coop. W ildl. Ecol. Res.
Net w ork, Frederict on, NB.
          Razorbills are colonial seabirds that breed on islands. Machias Seal Island (MSI) 44°30 ' N, 67 °06' W is
the lar gest col ony of razorbi lls near t he sout hern lim it of their rang e. On MSI, an acc urat e pop ulat ion est imate
is difficult , as some nests are not ' visible' t o surveyors due to the heterogeneous habitat. The objective of t his
study w as to develop a survey method that w ill estimate the number of unseen nests. Typically, correct ion
factors (k-values) are calculated based on the ratio of adults count ed in a plot t o the number of active nests in
that plot. Radio telemetry was proposed as a non-invasive method to develop a specific k-value for MSI.
Radio-transmitters were attached to 24 breeding adult razorbills (inferred from presence of brood patch) w hich
w ere representatives of all birds breeding on MSI. All radio-tagged birds were tracked to their nest site and the
sit e sco red as visible or unseen t o a surv eyor. A com plet e nest ing habi tat sur vey w as co ndu ct ed bet w een peak
lay and peak hatch to count all visible nests. A correction factor w as generated from the number of
radio-tagged birds found in concealed nests. Seven radio-tagged birds were found in nests that w ere not
count ed. A pplyi ng t he correct ion f actor to t he tot al number of nests co unt ed (35 1) result ed in a correct ed nest
num ber o f 43 9. The r esults of this st udy suggest this met hod may be useful i n est abli shing ap pro pri ate
correction factors for calibrating counts of other cavity-nesting seabirds.

371      Brua, Clark & Leach
Family matt ers: Eff ects of siblings and m oth ers on natal disp ersal of Gadw all and Mallard. ROBERT B. BRUA* ,
Dept . Biol., Un iv. Saskat chew an, Sask ato on, SK; ROBERT G. CLARK and STEVE W. LEACH, Can. Wildl. Serv.,
Saskatoon, SK.
          In speci es w it h f emale-biased ph ilo pat ry , n atal di sper sal and hab it at use m ay be inf luen ced b y innat e
cues, early experience, and social eff ects of mot hers and sisters. However, inf ormation about natal dispersal of
w aterf ow l, and speci fical ly, ef fect s of relat ives on nat al di sper sal patterns , i s very lim it ed. Thu s, w e evaluat ed
natal dispersal and habit at use in f emale Gadw all and Mallard in r elation t o natal hat ching d ate, age at f irst
(det ect ed) breedi ng, and b ehavior of their sisters, using 1 6 y r of dat a on i ndi vidual ly-mar ked d uckli ngs. M edian
nat al di sper sal dist ance w as 528 m f or Gad w all an d 6 46 m f or M allar ds. Yearling females show ed gr eater
affinit y to t heir natal habitat than expected by chance. For each species, distance dispersed was not related to
duckling hatch date or wetland abundance (an index of habitat quality) in the year of return. How ever, for each
species, a strong positive association existed betw een distance dispersed betw een full siblings, suggesting the
exist ence o f eit her s ubstant ial h erit able v ariat ion in d ispersal distance o r st ron g at traction among sist ers w hen
choosin g nest sit es.

372      Krebs, Norment & Joule
Breeding ecology of Henslow' s Sparrows at Ft. Drum, New York. ROBIN E. KREBS, CHRISTOPHER J.
NORMENT,* Dept. Biol. Sci. , SUNY Brockpor t, Brockport , NY; and STEVEN JOULE, Dept . Envi ronment , U. S.
Arm y, Fort Drum, NY.
        We studied the breeding ecology of Henslow' s Sparrow s at Fort Drum, New York, an active military
base. Abundance of Henslow' s Sparrow s averaged 37 pairs/yr, based on data from 4 8 10 0-m-radius point
counts and four 7 - 1 2 ha gridded study plots. Henslow' s Sparrow s w ere detected at only 4% of t he point
counts during all 6 yrs of t he study. From 199 8 t o 200 0, w e used a combination of banding, territory mapping,
behavioral observations, and vegetation analyses to conduct detailed investigations of Henslow ' s Sparrow
productivit y and habitat relationships. We color banded 84 individuals, including 56 males, 12 f emales, and 16
juv enil es. Annu al adu lt sit e ret urn rat es averaged onl y 1 8. 5% . In add it ion , o nly 25 % of territ ori al mal es
successfully fledged young. Multiv ariate analyses indicated that Henslow ' s Sparrow s selected habitat w ith low
shrub cover and a deep litter layer. Henslow' s Sparrow s declined by 9.5% /yr in the Eastern Breeding Bird
Survey region betw een 196 6 - 19 99, in part due to habitat loss. How ever, military activit y at Fort Drum
pot ent iall y prev ent s gr assland h abit at from rev ert ing to f ores t habi tat, an d t hus help s mai nt ain h abit at that
support s part of the largest Henslow ' s Sparrow breeding popul ation i n th e Northeast .

373      Norment & Halbritter
Suitabili ty of sw itc hgrass (Panicum) stands as habitat for grassland birds in New York. CHRISTOPHER J.
NORMENT* and HEATHER HALBRITTER, Dept. Biol. Sci. , SUNY College Brockport , Brockport , NY.
          We stud ied abundance and breeding biolo gy of nongame birds i n monot ypic stands of sw itc hgrass,
Panicum virgatum, in w. New York betw een 1995 and 2000. In 199 5 and 1996, point count data from 48
fields t hat inc luded a range of habit at t ypes indic ated t hat f ew obligat e grassland birds occ urred in sw itc hgrass
fields. Obligate grassland bird species richness and abundance w as highest in larger fields with relatively low
vegetation height and density. In 2 000 , w e detected almost no obligate grassland birds in 24
Panicum-dominat ed fields rangi ng f rom 1 to 3 1 ha. Tw o species of m anagement conc ern in t he North east,
Nort hern Harrier and Sedge Wrens, oc curred v ery sporadic ally in f ields includ ed in th e study . The most
common nongame birds in sw itchgrass fields were Swamp Sparrows, Song Sparrows, and Common
Yellowt hroats; none of these species show declining Breeding Bird Survey trends in the Northeast. We found
relatively f ew nests in sw itchgrass fields, although the proportion of successful nests in sw itchgrass fields
(0.3 5) w as significantly less than in cool season grasslands (0.71); an earlier study show ed that there was no
significant diff erence in predation rates for artif icial nests in w arm and cool season grasslands. Results of our
study indicate that swit chgrass fields provide poor habitat f or nongame grassland birds, and that managers
seeking to encourage grassland birds in the Northeast should avoid establishing fields w ith t all, dense stands of
w arm season grasses.

374      Chong & Given
Imm uno competence r esponses in ex tra-p air an d w it hin -pair of fspr ing of the Red-bill ed Gul l f rom Kaik our a, New
Zealand. NICOLA L. CHONG* and ANDREW D. GIVEN, Centre Biodiv. & Conserv . Biol. , Royal Ont ario Mu s,
Toront o, ON.
         The Red-billed Gull (Larus novaehollandiae scopulinus) f rom Kaik our a has been t he subjec t of a 39 -year
study creating one of the largest pedigrees for a natural population. The species forms socially monogamous
pair -bonds but ext ra-p air c opu lat ion (EPC) has been observ ed. Behaviou ral o bser vat ion s of EPC has sugg est ed
only a 3% success rate (Mills 1994, Behaviour 128 :41 -64). How ever recent molecular work using
microsatellite loci has revealed this rate to be approximately 20%. Extra-pair paternity is common in many bird
species (Petrie & Kempenaers 1998, Trends Ecol. Evol. 13 : 52 -58). The reasons for this are unclear but there
are many theor ies. The b enef it s of EPC must out w eigh the po tent ial c ost s. A num ber o f theor ies sugg est that
females choose mates to gain genes that increase fit ness in the offspring. The Kaikoura colony has a shortage
of breeding males and thus the females may engage in EPC to obtain genetic benefits for their off spring which
are no t ot herw ise av ailab le t o t hem. To det ermine w het her ex tra-p air o ffspr ing have bet ter T-cell mediat ed
immun e responsiveness th an their half -siblings; immun ocomp etence test s using phy toh aemagglutin in (PHA)
w ere carried out in 94 Red-billed gull families. Microsatellite loci were then used to determine which of fspring
w ere the result of ext ra-pair fertilization and the magnitude of response compared. The responses appear to be
variable both w ithin and betw een broods. Comparison of half-siblings raised in the same nest removes both
enviro nment al and mat ernal genet ic ef fect s and thus any dif ferenc e in r esponse is solely due t o t he pat ernal
genetic cont ribution. Previous studies have shown an increased immune responsiveness in extra-pair off spring
w hen c ompared to t heir mat ernal half -sibli ngs, sugg est ing a possible ben efit for ex tra-p air f ert ili zation (John sen
et al. 2000, Nature 406:296-299, Saino et al. 1997, Proc. Natl. Acad Sci. USA 9 4: 54 9-5 52 ).
375     Brook & Clark
Patterns of reproductive investment in Lesser Scaup. RODNEY W. BROOK, Dept. Biol., Univ. Saskatchew an,
SK; and ROBERT G. CLARK, Can. Wildl. Serv. Saskatoon, SK.
         Birds breeding at northern latitudes may face greater reproductive constraints than conspecifics
breeding farther south. Timing of breeding, and clutch and egg characteristics, w ere recorded for Lesser Scaup
nesting near Yellowknife, NT (northern boreal forest, 1999 and 2000) and Saskatoon, SK (prairie, 1989 to
200 0). Clutch initiation date (CID) and egg size w ere similar betw een sites, but date-adjusted total clutch
inv est ment w as low er in bor eal f ores t. A mong-clu tch egg size v ariat ion also w as low er at the bo real sit e,
perh aps consist ent w it h " ind ividu al op timizat ion " of clu tch size-egg size all ocation s. On t he pr airi es, CID w as
earlier in years w ith higher May wetland numbers, but no relationship w as found betw een total clutch
investment and wet lands. Clutch size declined seasonally on both sites, a decrease which could not be
explain ed by w etland num bers on t he pr airi e sit e. So me patterns of repr odu ct ive inv est ment dif fered
latitudinally for Lesser Scaup (clutch size, egg size variation), w hereas others were remarkably similar (CID and
mean egg size). Because breeding propensity is age-dependent, and interacts w ith w etland conditions, age
should be co nsidered in f utu re studi es of repro duct ive allocat ion pat terns f or t his species.

376      Revels & Amlaner
Effect of Protocalliphora (Diptera: Calliphoridae) parasitism on Hooded Warbler breeding behavior and
reproductive success. MIA R. REVELS* , Biol. Dept., Nort heastern St ate Uni v. , Tahl equah, OK; an d CHA RLES
J. AMLANER, Jr., Dept. Lif e Sciences, Indiana Stat e Univ., Terre Haute, IN.
        The eff ect of Protocalliphora parasitism on the behavior and reproductive success of Hooded Warblers
w as examined. Act ivity budgets of males, females, and nestlings at parasitized and unparasitized nests were
compared. Two behaviors, feeding rate and probing rate, w ere explored in greater detail. Finally, t he
relationship betw een parasitism and reproductive success w as examined. 53 nests w ere filmed for a total of
approximately 300 hr, then collected upon completion of the nesting attempt and examined for parasites.
Male, female, and nestling activit y budgets as measured did not seem to be affected by parasitism. However,
both probing and feeding rates were significantly higher in nests parasitized by Protocalliphora. A ssuming t hat
nest probing is a nest maintenance behavior which serves to limit parasite numbers, probing activity w ould be
expect ed t o in crease if the bi rds det ect paras it ism . The in creased f eedin g rat es may have been a p arent al
response w hich ameliorated the effect of nestling blood lost to Protocalliphora larv ae. Fin ally , p arasitized
bro ods had sign if ican tly low er f ledg ing suc cess than u npar asit ized b roo ds, and m ort alit y w as pr imaril y due t o
predation. This association might reflect a cause and eff ect relationship, wit h parasitism increasing feeding
trips which in turn increased nest apparency.

377      Moore
Parent Common Ter ns r earin g enl arged bro ods deli ver more by alt erin g t he all ocation of energ y betw een
self-maintenance and reproduction. DAVID J. MOORE, Dept . Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
         Model s of opt imal ener gy allo cat ion pred ict that a paren t pro visionin g a rel atively larg e bro od m aximizes
fitness by reducing its own self-feeding rate and delivering a greater proportion of captured prey to offspring.
To test this idea, I manipulated brood size to create reduced, control, and enlarged broods containing 1, 3, and
5 nestlings, respectively. This manipulation served to modify both the costs of provisioning and the relative
reproductive value of the current brood. Parental allocation decisions were assessed by comparing changes in
parental body mass (intake) w ith t he mass of delivered prey, measured concurrently during individual foraging
trip s. As pr edic ted, paren ts respon ded t o in creased deliv ery requ irem ent by redu cin g t he amoun t of food they
consumed/foraging trip, t hereby allocating a greater proportion of captured energy to nestlings. In spite of
consuming l ess f ood /t rip , p arent s rear ing enlar ged b roo ds m ade more p rov isioni ng t rip s, resu lt ing in a h igh er
rate of energy intake. The estimated parental net energy balance (intake - expenditure) did not diff er among
brood size treatments, indicating t hat parents adjusted their self-feeding rate in relation to their daily energy
expendi ture. Parent s pr ov isioni ng en larg ed br ood s w eigh ed less at the end of the experi ment, sugg est ing that
they may be un able t o co mpensat e com plet ely for an inc reased rat e of expendi ture o ver a lon ger t ime scale.
378     Brown & Jiang
Testing sex allocation theory in a communally rearing bird. JERRAM L. BROWN* and JIANQING JIANG, Dept.
Biol. Sci., Univ. at Albany, NY.
         In the Mexican Jay males at any age bring food t o nestlings more often than do females, suggesting
that young mot hers mig ht pref er sons t o daught ers. Therefo re, w e tested f or an excess of males u sing PCR to
amplify CHD genes to identify t he sexes of nestlings (15 d old). Young mothers (4 - 5 y r old) had significantly
more male than female nestlings (P < 0.0 5). Older mothers had more female offspring. How ever, the overall
nestling sex ratio w as 54% ( 15 6 males: 12 8 females), w hich was not significantly diff erent from 5 0% . There
w ere significantly more male than female nestlings in first nests of a mother in a given year ( P < 0.0 5). There
w ere more female nestli ngs in t he second nests, but the dif ference w as not sig nif icant (P > 0. 05 ). First
hatch ed nestling s in a brood (ju dged by w eight at banding) w ere more likely t o be male than f emale (60 males,
35 f emales, P < 0.0 5), but later nestlings w ere not. There w ere more male (121 ) than female (90) nestlings in
the first half of t he breeding season (P < 0.0 5), but not in t he second. There was a significant excess of
males in broods of fiv e-nestlings (P < 0.0 5), but not in smaller broods. The reasons for these biases are not

379      Blair & Reale
Artificial nest studies: Are they based in reality? ROBERT B. BLAIR and JOSEPH A. REALE, Dept. Zool., Miami
Univ. , Oxf ord, OH.
         In recent years, the use of artificial nests to measure predation pressure has become a growt h industry
for ornithologists. How ever, these studies have been criticized on 2 front s: 1) for using eggs that are
inappropriately sized and 2) for not reflecting the true conditions of nests which are protected by vigilant
parents. To address these concerns, w e studied nesting success of American Robins and Northern Cardinals
alon g an u rban grad ient in sw . Oh io an d co mpared i t to pr edat ion rat es on art if ici al nests in the sam e sit es ov er
a 4-yr peri od. Based on lo gistic r egres sion, w e foun d t hat the in tensity of pred ation on r eal egg s and art if ici al
nest s dec reased as t he in tensity of urb anizat ion inc reased. How ever , t he in tensity of pred ation on ar tif ici al
nests was significantly higher than that on real nests. Addit ionally, predation rates on hatchlings showed no
relationship to t he level of urbanization and overall nesting failure actually increased wit h increasing
urbanization. The decrease in nesting success wit h increased urbanization coincided w ith an increase in the
num ber o f mult ipl e nest ing spec ies in t he co mmuni ty. These resu lt s suggest that art if ici al nests pr ov ide an
adequate measure of predation pressure but seriously question their validity as a means of assessing nesting

380      Sustaita & Hertel
Modeling changes in morphology and populat ion ecology o f the J uan Fernan dez Fir ecrow n. DIEGO
SUSTAITA* and FRITZ HERTEL, Biol. Dept ., Calif orn ia St ate Univ . No rt hri dge, CA.
         The Juan Fernandez Firecrown (Sephanoides fernandensis) is an endangered endemic hummingbird from
the Juan Fernandez Islands, off the coast of Chile. Analysis of historical and present data revealed that t his
species has undergone considerable changes in both beak morphology and population size. Reduction of t he
island’s endemic forest due to habitat destruct ion and degradation, coupled wit h invasion of exotic plant
spec ies, have resu lt ed in chan ges in select ive pres sur es t hat pres umably are res pon sible f or t he ob serv ed
decr ease in beak l engt h and in p opu lat ion size over the past 12 8 y r. These chan ges in f lor al co mposition may
also have influenced competit ive interactions between S. fernandensis and a coexisting mainland congener, the
Green-backed Firecrown (Sephan oid es sephano ides), whose population size has greatly increased over the same
period of time. STELLA 6.0. 1 w as used to constr uct a model in order to track changes in the beak length of S.
fernandensis and population size, wit h respect t o changes in the relative proportions of endemic versus exotic
plants, and potential competition w ith S. sephan oid es. The model w as based on estimates of contemporary
rates of change in the aforementioned parameters, and consists of dif ferent scenarios that predict changes in
morphology and demography. The ultimate objective of the model was to depict the changes necessary to
prolong the extinct ion and enhance the population size of the endemic S. fernandensis.

381      Hahn, Pereyra, Sharbaugh, Morton, Katti & M acDougall-Shackleton
Relationships betw een photorefractoriness and reproductive flexibility in Cardueline finches. T. P. HAHN* , M.
E. PEREYRA, S. M . SHARBAUGH, M . L. M ORTON, Inst. Arctic Biol., Univ. Alaska, Fairbanks, AK; M . KA TTI,
Dept. Biol., Arizona State Univ., Tempe, AZ; and S. A. MacDOUGALL-SHACKLETON, Psychol. Dept., Univ.
Toronto, Mississauga, ON.
         We used photoperiod manipulations and measured changes in gonadal condition and gonadotropic
hormones to study phot orefractoriness in 7 taxa of cardueline finches that vary f rom seasonal to opportunistic
breed ers. Experiment al exposure t o very lon g day s (2 4L: 0D) durin g pu tative ref ractori ness, an d pr olo nged
exposure to constant long days, indicate that these taxa vary in whether or not t hey become refractory. At one
extreme of flexibility , crossbills neither regress gonads on constant long days, nor cease to respond to long
days after gonadal regression occurs on declining day length; they do not become absolutely refractory. A t t he
other extreme, Carpodacus finches regress gonads on constant long days, and cease to respond to long days
once gonadal regression occurs; they do become absolutely refractory. Among close relatives, flexible taxa
(e.g., Pine Siskins) do not become refractory, w hile seasonal taxa (e.g., American Goldfinches) do. Further, the
oldest lineage tested, Gray-crow ned Rosy-finch, appears not to become refractory , despite breeding seasonally
in an extreme environment. These data suggest that Cardueline finches possess physiological characteristics
that predispose them to temporal reproductive flexibility. Relatively rigid seasonality in some taxa, w ith
development of absolute photorefractoriness, may have evolved recently in this group.

382      Hahn, Ottinger & Igl
Role of maternal t estost erone in comp etit ive grow th and b ehavior of parasitic cow bird of fspr ing. D. C. HAHN* ,
Hatfield, USGS-PWRC, Laurel, MD; M . A . OTTI NGER, Dept. Poultry Sci., Univ. Maryland, College Park, MD;
and L. D. IGL, USGS-NPRC, J amestow n, ND.
         We investigated the physiological basis of competition by parasitic Brown-headed Cowbird offspring.
Cowbird nestlings are widely reported to be aggressively competitive and dominant over host off spring in
begging for food. We examined the role of yolk t estosterone in effecting nestling cowbird behavior, since
elevated lev els direc tly con trol nest lin g gr ow th and begg ing vigor in song bir ds, and sinc e females in m any avian
species apparently distribute yolk testosterone strategically to create systematic wit hin-clutch patterns. We
hypothesized that female cowbirds may similarly manipulate the competitive success of their offspring by a
strategic inter-species pattern of yolk testosterone deposition. We measured yolk testosterone levels in cowbird
eggs col lect ed f rom four geog raph ical areas r epres ent ing ances tral an d in vaded ran ges, and w e foun d t hat
cow bird eggs had sim ilar levels of yolk test ostero ne regardless of geogr aphical or subsp ecies origin. We also
collec ted egg s of cow bir d ho st spec ies and co mpared t he rel ative yolk t est ost eron e levels. Host eggs of three
species had significantly higher levels of yolk testosterone than cowbirds: Red-winged Blackbird (North
Dakota), Song Sparrow (Washington), and House Sparrow (Virginia). Future work will examine the relative
levels of cow bird yolk testosterone w ith host eggs of different laying order.

383      Curnutt
Grassland fire d yn amic s and management for t he end anger ed Cape Sab le Seaside Spar row . J OHN L.
          Fire dynamics of grasslands in Everglades National Park were identified by vegetation community and by
spat ial d ist rib ut ion . Fir e frequ ency , r eturn time, an d in tensity (i. e., spat ial ex tent ) w ere compared bet w een sites
dom inat ed by Muhl enber gia sp. (pr eferred sparrow nest ing habi tat) and ot her h abit at typ es. Fire return time w as
most inf luen ced b y soil dept h, season of bur n (dry vs. w et), and i nv erse d ist ance f rom agri cult ural /r esident ial
boundaries. Sparrow breeding response t o fi re has been investigat ed by Werner (1 97 5, USFWS Rept,
Everg lades Nat l. Park , Homest ead, FL) and Curnutt et al. (1998, Animal Conser. 1: 1 1-21). Current fire
management practices result in loss of pot ential breeding areas by allowing sub-optimum fire regimes. Results
of t his study show that proactiv e fire management of Everglades grasslands could increase available breeding
habitat for the sparrow.

384      Bisson & Stutchbury
Evolutio nary ori gin and geneti c consequences of avian migrat ion in a t ropical ki ngbird . ISA BELLE-ANNE
BISSON* and BRIDGET J. M. STUTCHBURY, Dept. Biology, York Univ ., Toront o, ON.
          The origin of avian migration is one of the most significant questions in ornithology. The Evolutionary
Precurso r Hy pot hesis pr opo ses t hat migrat ion bet w een t he Tr opi cs o f Cancer and Cap ric orn (intrat rop ical
migration) was the first ancestral step to long-distance Nearctic migration. Our primary goal is to reconstruct
the evol uti onary sequence of events t hat lead t o migrat ion out of t he trop ics by buildin g a mit ochond rial DNA
phylogeography for populations of the Cassin' s Kingbird. Cassin’s Kingbird is ideal for t his study because the
species has recently diverged in migratory behavior at 3 levels: 1) sedentary (south-central Mexico), 2) Nearctic
migrant (southwestern U.S.), and 3) intratropical migrant populations (north-central Mexico). We are
additionally assessing the genetic consequences of migratory behavior by measuring the genetic variation and
comparin g morp hol ogy and d aw n song am ong Cassin' s Kingb ird pop ulat ion s. We have r ecent ly com plet ed
blood sampling, morphological measurements and daw n song recordings for a total of 15 Cassin' s Kingbird
populations (88 blood samples, 83 morphological samples, 136 dawn song recordings) across the U.S. and
Mexico. Analyses on morphological and daw n song structure for populations sampled across U.S. and Mexico
w ill be pr esent ed. Molec ular analy ses using mtDNA con trol regi on an d micr osat telit es are c urr ent ly und erw ay
in Steve Lougheed' s laborator y at Queen' s Universit y (Kingst on, ON).

385      Wikelski, Spinney, Schelsky, Scheuerlein & Gw inner
Slow pace of life in tropical birds. MARTIN WIKELSKI* , LAURA SPINNEY, Dept. Ecol. & Evol. Biol., Princeton
Univ. , Princet on, NJ ; WENDY SCHELSKY, Univ. Il linois, Urbana, IL; ALEXANDER SCHEUERLEIN, Univ.
Missouri, St Louis, MO; and EBERHARD GWINNER, Res. Center f or Ornit hol., Max-Planck Soc. , A ndechs,
          Many trop ical bir ds live lon ger t han t heir temperat e zone c oun terpar ts. It has been hy pot hesized t hat
diff erences in energy metabolism underlie such differences in life history. We tested whether innate rates of
energy expenditure increase with latitude and/or migratory propensity among populations of one bird species.
Nestling St onechats (Saxi cola t orq uat a) w ere obtained f rom equat orial Kenya (00 ° N), Ireland (51 .5 0° N),
Aust ria (47 .5 0° N) and Kazakh st an (5 1. 50 ° N). They w ere raised and subsequen tly kept und er id ent ical
conditions in the laboratory. We then measured their basal metabolic rates (BMR) at an age of about 15 m on
during molt (Aug) and 20 mon during the quiescent period in Jan. Kenyan birds had lowest rates of molt and
BMR in both seasons. Long-dist ance migrat ing Kazakhstani bi rds had highest molt rates, as w ell as highest
energy turnover during molt. Metabolic rates of Aust rian, Irish and Kasakhstani birds during wint ering were
low er t han d uri ng m olt , an d in disting uishabl e bet w een po pul ation s. Our dat a demonstrat e for t he f irs t time that
even in birds kept in the same environment t he “pace of lif e”, as indicated by metabolic turnover, is low est in
sedentary t ropical, and highest in nort h-temp erate migrat ory in divid uals of o ne and the same species.

386      Dahdul & Horn
Energy allocation and postnatal growth in fish-eating seabirds: Feeding trials wit h captive Elegant Tern chicks
raised on high vs. low energy diets. WASILA M. DAHDUL* and MICHAEL H. HORN, Dept. Biol. Sci., California
State Univ ., Fullerton, CA.
        The interval betw een hatching and fledging may require more energy than at any other time in a bird' s
life. This period may be particularly critical for piscivorous seabirds because their fish prey varies in energy
content. This variation in f ood quality, coupled w ith current changes in ocean conditions that aff ect the
abundance of their preferred high-energy prey, means that seabirds are likely to prov ision their chicks on
alternat ive, lo w er quality prey. The diet of the Elegant Tern at t he Bolsa Chica Ecologic al Reserve in s.
California consists of > 60% energy-rich clupeoids, fishes that have fluctuated in abundance historically and
recently. To investigate the effects of diet quality on chick grow th, w e raised Elegant Tern chicks in the
laborator y on 3 diets: (1) energy-r ich nor thern anchovy , a clupeoid , at t he average daily int ake rate, (2 )
energy-poor topsmelt at t he average daily intake rate, and (3) topsmelt at an amount equal in energy to t he
anchovy diet. A s expected, chicks raised on equal-energy diets of anchovy and topsmelt had higher rates of
mass gain than chicks raised on the low -energy diet of t opsmelt. How ever, all 3 groups had similar rates of
w ing grow th, indicating t hat chicks preferentially allocate energy tow ard structural grow th. Body composition
analysis revealed that chicks did not diff er in lean dry mass, and that f at content of the chicks w as greater wit h
increased energy intake. The variability in grow th exhibited by Elegant Tern chicks may increase the likelihood
of suc cessful f ledg ing und er cond it ion s of redu ced en ergy int ake.

387     Anderson, Roby, Lyons & Collis
Foraging patt erns of Caspian Terns in r elation t o colon y size and marine inf luence. SCOTT K. A NDERSON* ,
DANIEL D. ROBY, DONALD E. LYONS, Oregon Coop. Fish & Wildl. Res. Unit, USGS; Dept. Fish. & Wildl.,
Oregon St ate Uni v. , Cor valli s, OR; and KEN COLLIS, Real Tim e Research, Bend, OR.
         We tr acked radio-t agged Caspian terns nestin g at East Sand Island in 19 99 (1, 40 0 nest ing pairs), East
Sand Island in 2000 (9 097 nesting pairs), and Rice Island in 199 9 (80 96 nesting pairs) to investigate the
effects of colony size and location on foraging patterns. Terns nesting at East Sand Island w ere found more in
marine environments and consumed 41% few er out-migrating juvenile salmonids than those nesting at Rice
Island. The median distance of off -colony detections for t erns breeding at t he large East Sand Island colony in
200 0 (18 km) w as similar to that of terns breeding at the large Rice Island colony in 199 9 (16 km), but much
greater than that of t erns breeding at t he smaller East Sand Island colony in 1999 (7 km). We detected nesting
terns > 70 km from their East Sand Island nest sites in 2000; the previously published record was 62 km.
These data suggest that colony size was the primary factor influencing foraging distance, not colony location.
Caspian t erns nest ing at larg e col oni es apparent ly experien ce den sit y-depen dent effect s on food availabi lit y near
the co lon y, con sistent w it h A shm ole' s halo. Alt hou gh t erns nest ing at East Sand Island forag ed at great er
distances in 2000 c ompared to 199 9, salmonid consumption remained the same (44% of t he diet in bot h

388     Pearson, Levey & Greenberg
Assessing the importance of insects, fruit and seeds to over-wintering birds. SCOTT F. PEARSON* , Nat ural
Areas Program, Washington Dept. Nat. Res., Olympia, WA; DOUGLAS J. LEVEY, Zool. Dept ., Univ. Florida,
Gainesv ille, FL; and CATHRYN H. GREENBERG, USDA Forest Service, Sout hern Res. St ati on, As heville, NC.
         Many spec ies of bir ds sw it ch from an in sect ivoro us diet dur ing the br eedin g season t o a di et that
includes at least some fruit and/or seeds during the non-breeding season. Yet, the relative importance of
insects, fruit and seeds to birds during the non-breeding season remains virtually unexplored. We used stable
isotopes to assess the relative importance of f ruit, insects and seeds to 20 species of omnivorous
over-wint ering birds at t he Savannah River Site in South Carolina. Dietary items diff ered in both carbon and
nitrogen isotopes: * 15 N values of insects w ere higher than that of f ruit and seeds and * 13 C values of C4 plant
seeds were higher than fruit or insects. For several bird species, * 15 N values of w hol e blo od d ecreased
betw een fall and wint er, indicating a change in diet f rom one dominated by insects t o one dominated by fruit
and/or seeds. Some species wit h low * 15 N blood values during the wint er also had low * 13 C values -- indicating
a diet dominated by f ruit and/or seeds -- w hile others had high * 13 C blood values, suggesting a diet cont aining
C4 plant seeds. T hese data coupled w ith dietary inf ormat ion gat hered fro m f eces of t he same birds, sugg est
that f ruit and grass seeds are critical w inter food resources for at least some over-wint ering bird species at our
sit e.
389     Bloxton, Marzluff & V arland
Potential in flu ences of La Niña on spati al use and foraging behavior of Nort hern Goshaw ks. TOM D.
BLOXTON,* JOHN M. MARZLUFF, Coll. Forest Res., Univ. Washington, Seattle, WA; an d DA NIEL E.
VA RLAND, Rayonier, Hoquiam, WA.
         We describe breeding season (15 M ar - 15 Aug) spatial use patt erns of 14 adult Nort hern Goshawks in
managed forests of w . Washington betw een 199 6 - 20 00. Mean home range size (3915 ha [100 % minimum
convex polygon] and 37 91 ha [9 5% fixed kernel estimator]) w as substantially larger than in other regions, w ith
the exception of se. Alaska. Kernel estimates of non-nesting birds (4538 ha; including those that failed early)
w ere lar ger t han b reeder s (3 26 8 h a). A sim ilar dif ferenc e w as seen b etw een males (3 95 9 h a) and females
(3005 ha). Range size more than doubled following the La Niña winter of 1998 - 1999 (2374 vs. 5051 ha).
Abundance indices of some key goshaw k prey declined follow ing the La Niña event and goshawk survival and
produc tiv ity estimat es also declined in t he 2 yr fol low ing t he event. While many birds did not breed in th e post
La Niña peri od, even those that did had u nusuall y large h ome rang es (5 77 9 h a). Core ar eas (5 0% fixed ker nel
estimators) nearly doubled follow ing the La Niña winter as well, suggesting that goshaw ks needed to expand
their ent ire range to f ind prey and w ere not just making oc casional w ide hunt ing f orays.

390     Oleyar, M arti, Bechard & Munger
Impacts of Olympic Ski Area development and recreational activit ies on breeding Flammulated Owls in northern
Utah. M. DAVID OLEYAR* , Dept. Biol. and Raptor Res. Center, Boise State Univ., Boise, ID, CARL D. MA RTI,
Raptor Res. Cent er, MARC J. BECHARD, Dept . Biol. and Rapt or Res. Cent er, and JAMES C. MUNGER, Dept.
Biol., Boise State Univ.
         Flamm ulat ed Ow l rep rod uction w as measured to in vestigat e possible im pact s of Oly mpic s ki ar ea
development at t he Snow Basin ski resort in n. Utah. Flammulated Owls nesting near the development site
fledg ed signi fican tly few er y oun g t han o w ls at the co nt rol sit e, an d at ot her ar eas in the w . U.S. Reduced
productivit y at Snow Basin was attribut able to significantly low er hatching rates. Ski area development, as w ell
as high l evels of recreat ion al use, w ere t he most lik ely caus es of this redu ced p rod uctivit y. Dist urb ance m ay
possibly affect hatching rates by interrupt ing normal courtship causing an increase in infertile eggs, or by
forcing incubating females to flush more often causing an increase in egg mortality.

391     Wehtje
The range expansion of the Great-tailed Grackle in the western United States. WALTER WEHTJE, Dept. Earth
Sci. , Un iv. Calif orn ia, Riverside, CA.
         The current range expansion of t he Great-tailed Grackle, Quiscalus mexicanus, provides an opportunity
to determine whether t axonomic and morphological techniques can be used to increase our understanding of a
spatial process. Since 1900, w hen its U.S. range w as restrict ed to south Texas, this species has expanded its
breeding range to include 18 st ates, from Oregon to Louisiana. In the western states, t he range expansion
involves two morphologically and phenotypically distinctive subspecies, Q. m. nelsoni and Q. m. monsoni.
These taxa invaded the U.S. from diff erent source areas and first came into secondary contact in Arizona in the
mid 195 0s. Since then, the tw o subspecies have spread throughout Arizona, California, Nevada, Utah, and
parts north. Birds w ere collected from these four states and assigned to either subspecies based upon
morphological features and plumage coloration. The results of t his classification have allowed the mapping of
the range expansion of each subspecies, the degree of int erbreeding betw een the 2 t axa, and eff ect of
environmental variables upon the spread of either subspecies in the w. U.S. By paying attent ion to t he
subspec ies relat ion ships w it hin the Great -t ailed Grack le, it has been po ssible t o map t his rang e expansion at a
much finer scale than if the species been treated as a single unit.

392     Bednarz, Cannon & Blakely
Distr ibut ion and habit at use of Sw ainson' s Warblers in A rkansas. JA MES C. BEDNARZ* , BRIAN J. CANNON,
and WILLIAM W. BLAKELY, Dept. Biol. Sci., Arkansas State Univ., Jonesboro, AR.
          The se. U.S. has experienced a severe decline in bott omland hardw ood forests impacting many
pop ulat ion s of neot rop ical migrat ory bir ds includ ing the Sw ainson' s War bler (SWWA ). We cond ucted
systematic surveys of f orest habitats/age classes in the White River National Wildlife Refuge, St. Francis
National Forest, Buffalo National River, and Bayou Meto Wildlife Management Area to document the occurrence
of breeding SWWAs. In addition, w e quantit atively described vegetation structure and characteristics at all
documented SWWA sites. We detected 26 SWWAs in bottomland hardwood forests and 7 in canebrakes.
When all study areas were combined, shrub cover (mean = 9.0 %), leaf litt er cover (mean = 65. 9% ), shrub
stem density (mean = 12.5/m 2 ), and c anop y c ov er (m ean = 83 .2 %) w ere signi fican tly great er in occ upi ed t han
in random sites (mean shrub cover = 2.4%; leaf litter cover = 44.7%; stem density = 3.7/m 2 ; c anop y c ov er
= 71.9% ). Whereas green foliage (mean = 29.4% ) and forb cover (mean = 19.0% ) were lower at SWWA
use locations than at random sites (mean green foliage cover = 61. 3% ; mean forb cover = 37. 5% ). Base on a
prel iminar y r adio telemet ry st udy , h ome rang e size f or 2 male SWW As w as 9.2 ha and 14 .7 ha. Our dat a
strongly suggested that SWWAs prefer habitats with dense understory and a closed canopy. Furthermore, we
suggest integrating fire into bott omland systems to restore canebrakes that are important to t he conservation of
SWWA s in A rkansas.

393     Hale
Usin g sat elli te imag ery to model and m ap Bic knell' s Thrus h di st rib ut ion in New Hamp shire. STEPHEN R. HALE,
Complex Syst ems Res. Center. , Univ . New Hampshire, Dur ham, NH.
          Landsat-7 Enhanced Thematic Mapper satellite imagery was used to model Bicknell's Thrush distribut ion
and habitat in t he White Mountains of New Hampshire. Proof-of-concept w as established for using satellite
imagery in speci es-habit at modeli ng, w here i magery spec tral f eatures w ere us ed t o est imate a species-h abit at
model variable. The habitat model w as derived using logistic regression, w hich produced the probability of
detect ing a th rush based on t he patt ern of m odel cov ariates. The mo del predict ed rising pr obabilit ies of t hrush
presence w ith decreasing Dominant Vegetat ion Height , inc reasing Elevati on, and d ecreasing Distance t o Nearest
Fir Sapling cover type. Model solution required regressor estimates at every pixel, which w ere not available for
the Dominant Vegetation Height and Elev ation var iabl es. Top ogr aphi call y norm alized imagery features
Normalized Difference Vegetation Index and Band 1 (blue) were used to estimate dominant vegetation height
using m ult ipl e lin ear regres sion; and a Digi tal Elevat ion Model w as used t o est imate elevation . Dist ance t o
nearest Fir Sapling cover type was obtained for each pixel from a land cover map specifically constructed for
this project. Using independent t hrush data for model validation revealed the model application using satellite
imagery accurately estimated thrush presence at pixels within probability deciles ranging from 0 to < 0.4 0 and
from 0. 50 to < 0. 60 (max imum). Pixels w it h t hru sh presen ce pr obab ili ties f rom 0. 40 to < 0. 50 held great er
num bers of thru shes than ex pect ed (sign if ican tly und erest imated). A map pro duc t resu lt ed t hat inc lud ed
pred ict ed pr obab ili ties of BITH p resen ce at a spat ial scale o f 28 .5 m, over a geo grap hic ext ent of 91 ,0 00 ha.

394     Hames, Rosenberg, Lowe, Barker & Dhondt
The effects of acid rain on the Wood Thrush in eastern North America. RALPH S. HAMES* , KENNETH V.
ROSENBERG, JAMES D. LOWE, SARA E. BARKER, and ANDRE A. DHONDT, Cornell Lab. Ornithol., Cornell
Univ., Ithaca, NY.
           Despite little demonstrated sensitivit y to habitat f ragmentation, the Wood Thrush has shown significant
population declines across most of it s range, including the Northeast and Mid-Atlantic regions where cowbird
paras it ism is relat ively low . These r egio ns ar e, h ow ever , zo nes of inc reased depo sit ion of pol lut ant s such as
acid ions, particularly at higher elevations. We report the results of an analysis of t he eff ects of acid rain on the
Wood T hru sh in e. North A merica, using d ata from the Cor nell Labor atory of Ornit hol ogy ' s Birds in Forest ed
Landscape p rog ram. W e used a Geogr aphi c In form ation System t o co mbine t hese data w it h abu ndan ce dat a
from t he Breeding Bird Survey, data from t he National At mospheric Deposition Program, and data on soil pH
from t he National Resource Conservation Service to control f or covariation in these variables. Af ter controlling
for t he eff ects of t he other variables, we demonstrate a strong and significant negative effect of acid rain on
the pr obab ili ty of attempted br eedin g by the Wood Thr ush, as w ell as a signif ican t int eraction bet w een
measures of fragmentation and acid ion deposition. These findings highlight the need to consider large-scale
fact ors ot her t han h abit at fragm ent ation w hen i nv est igat ing the dec lin es of bir d po pul ation s, and t hey
demonstrat e the po w er of con tinen t-w ide, amat eur- collec ted dat aset s on bir d di st rib ut ion and ab und ance t o
address these questi ons.

395     Stew art, Westneat, Hanshu, Hatch & Rambo
Correlates of extr a-pair paternit y in Hou se Sparrow s. IA N STEWA RT* , DA VID WESTNEAT, RICHARD
HANSCHU, MEG HATCH and BIRCH RAMBO, T. H. M organ School of Biol., Univ . Kentuc ky, Lexington, KY.
         Many different factors have been linked to the extent of ext ra-pair fertilisations (EPFs) wit hin bird
species. We studied a Kentucky population of House Sparrows, a socially monogamous, sexually dimorphic
passerine, w ith t he males possessing a variably sized ‘badge’ of black throat feathers. We used an array of
micro satellit e markers to m easure the incidence of extr a-pair paternit y ov er 4 breeding seasons (1 99 5 - 1 99 8),
w ith specific reference to the influence of parental phenotype (including badge size), breeding density and
breeding synchrony. Extra-pair fertilisations (EPFs) accounted for 16% of nestlings (60/ 360 ), occurring in
37. 7% of broods (40 /10 6). M ale phenotype had a minor influence on paternity, since males wit h deeper bills
had a l ow er pr opo rt ion of EPFs in t heir bro ods, althou gh o ther v ariab les such as body size and bad ge size had
no ef fect . There w ere no signif ican t relat ion ships betw een f emale m orp hol ogy and EPF rates. EPF rat es
increased wit h breeding density (number of nestboxes per study barn) but did not covary w ith breeding

396     Etterson, Etterson & Cuthbert
Forest regeneration and area-sensiti ve birds: If w e grow it, they w ill com e. MA TTHEW A. ETTERSON* ,
Smith sonian M igrato ry Bird Cent er, Washing ton , DC; JULIE R. ETTERSON, Dept. Biol., Univ. Virginia,
Charlottesville, VA, and FRANCESCA J. CUTHBERT, Dept. Fish. & Wildl. , Univ . M innesot a, Mi nneapolis, M N.
         Patt erns observed in contemporaneous forest patches at different seral stages offer specific t estable
hypotheses about the manner in which bird communities are expected to change with forest succession.
Diversity and trophic specialization are expected to increase. Also, area sensitive species should become more
abun dant . W e develop ed a met a-analyt ical techn iqu e that allo w s us to co mpare bi rd s urv ey data col lect ed
using different survey protoc ols over an 82 yr span at the University of M ichigan Biological Station, Pellston,
MI. In general, diversity appeared to increase wit h succession, though resolution and statistical pow er were
low . Neither trophic guild struct ure nor specialization changed significantly w ith succession. Canopy-nesting
spec ies and area-sensit ive spec ies increased signif ican tly. This st udy lend s suppo rt to t he hy pot hesis t hat
area-sensit ive species may be limi ted by available habitat in nort hern temp erate for ests. T his is t he first
demonstration that habitat regeneration at a single site can help restore populations of neotropical migrant
songbirds. Finally, the meta-analytical model we describe should prove useful for analyzing many long-term
processes for w hich standardized data are lacking.

397     Graham
Spatial modeling to investigate patterns and process in species distribution for conservation planning.
CATHERINE H. GRAHAM, Mus. Vert. Zool., Univ. California, Berkeley, CA.
         The distribution of biodiversity and the processes that influence this distribution are poorly understood.
In the face of incomplete knowledge, I explore the use of predictive modeling as a surrogate of biodiversity and
a way t o test hypot heses about processes. I combined current environmental data (i.e. vegetation,
temperature, raw radar data) and point localities to create predictive species distribution models for birds in
Cameroon, West Af rica. I used GIS models to create these distributions and evaluated models wit h independent
point locality data. I used these predictive models as a surrogate of biodiversity and evaluated areas of high
species diversity and endemism. To examine processes that may have generated this distribution of
biodiversity, I used predictive models in combination w ith palaeoclimate information. For example, in
Cameroon, sister species occur on either side of the "A damaoua Cliff," a hypothesized geographical barrier.
Predictive species models showed that suitable habitat for a sister species exists on either side of the
hypothesized geographic barrier, w hich indicates that the allopatric species patt ern may be a result of the
histori cal b arrier an d no t a chan ge in habi tat.

398     Rotenberry, Kristan, Price & Lynam
Alt ernativ e causes of edge-abundance relation ships in bi rd com munit ies of Calif ornia coast al sage scrub. J OHN
T. ROTENBERRY* , WILLIAM B. KRISTAN III, MARY V. PRICE, Dept . Bio l., Un iv. Calif orn ia, Riverside, CA; and
ANTONY J. LYNAM, Wildl. Conserv. Soc., Bangkok, Thailand.
         Changes in the distribution and abundance of birds at urban-w ildland edges can be caused by different
factors. If edge effects are caused by attract ion or aversion to edges they will aff ect populations even if the
habitat near the edge is unchanged. However, because urban edges alter adjacent w ildland vegetation, edge
eff ects co uld be produ ced indirect ly v ia an animal' s habitat association s in t he absence of a behavioral response
to t he edg e it self . W e st udi ed edg e effect s of bir ds in s. Calif orn ian c oast al-sag e-scrub , an d assessed w het her
edge ef fect s w ere du e to di rect behav ior al respon ses t o edg es or to ch anges in v eget ation at edges com bin ed
w ith habitat selection by the avian species. Vegetation species composition and structure varied with distance
from edge, largely in site-specific w ays. Because vegetation characteristics were confounded w ith distance
from edge, the ef fect s of habi tat w ere ex plo red b y c alib rat ing vegetation measu rement s t o t he hab it at affini ties
of each b ird spec ies using in depen dent ly-der ived model s of the species' habi tat associat ion s. Five b ird spec ies
diff ered in occurrence among distances to edge, independent of the habitat. Habitat restoration at edges would
be in effect ive for t hese speci es. In addit ion , 2 spec ies of bir ds r esponded to veget ation grad ient s t hat
coincided wit h distance from edge, such that the effect of edge on these species was expressed via potentially
reversible habitat degradation.

399      Ramenofsky, Landys & Ramfar
Influence of environmental conditions on spring migratory behavior in captive White-crowned Sparrows.
MARILYN RAMENOFSKY* , META LANDYS and TRISHA RAMFAR, Dept. Zool, Univ. Washington, Seattle, WA.
         Migratory birds held in captivity display diel patterns of oriented restless behavior, i.e. migratory
restlessness (MR), that represent an individuals predisposition to migrate. Endogenous circannual rhythms and
pho toper iod are know n t o co nt rol ini tiat ion of spr ing migrat ory act ivit y in some sp ecies. Co rrelat ive st udi es
have shown that environmental factors may regulate activity en-route (Richardson 1978, Oikos 30: 224-272).
How ever , f ew empiri cal s tudi es hav e test ed ho w pro ximat e fact ors affect migrat ory behav ior . W e pro duc ed an
ethogram of diel activity displayed by captive White-crow ned Sparrow s (WCS) during the period of spring
migration (Agatsuma et al, in prep). In late afternoon, daytim e locomotory activit y ceased and birds began to
rest quietly coined as the "quiescent phase" by Morton (1967, Condor 69 : 4 91 -51 2). MR began w ith darkness
and continued until dawn, at w hich time birds commenced foraging. We tested the eff ects of t he follow ing
proximate factors: moonlight , temperature, and food restriction, on migrator y activit y. M R increased wit h
moonlight int ensity (0 to 1 .0 lux ), decreased wit h low overnight temperature (5° C), but show ed no change
w ith food restriction. Thus, w e suggest that environmental factors influence migratory activity in spring. Once
bir ds have init iat ed migr atory flig ht from w int erin g areas, l ocal env iro nment al co ndi tion s may act as
supplementary cues to either accelerate or decrease the speed of travel, ultimately aff ecting timing of arrival on
the breeding g rounds.

400      Kaltenecker, Carlisle & Bechard
Correlati on of s ongbird and ac cipit er num bers d uring f all mi grat ion i n sou th w est I daho. GREGORY S.
KALTENECKER, Idaho Bird Observatory, Boise, ID; JAY D. CARLISLE, Dept . Biol. , Univ . So ut h Dakot a,
Verm illi on, SD; and MARC J. BECHARD, Dept. Biol., Boise State Univ. , Boise, ID.
          We simultaneously mistnett ed songbirds and conducted a hawkw atch daily during fall migration from
199 7 - 20 00 at Lucky Peak, located in southwest Idaho. We compared raptor numbers/8-count -hr and
songbird captures/50 net-hr. For all years, migration timing of accipiters (Sharp-shinned and Cooper's Hawks
co mb in ed) w as co rr elat ed w it h o cc ur renc e of shor t-distance mi gr ant song bi rd species (r = 0.31 - 0 .57, P =
0.0 3 - 0. 000 1). A ccipiter passage was correlated with t he peak in total songbird abundance, regardless of
songbird species. Sharp-shinned Haw k passage was correlated with occurr ence of short-distance migrant
songbirds during all years, while Cooper' s Hawk passage was only correlated wit h songbird numbers during
19 97 and 1 99 8. Thi s annual v ariat ion ind icat es t hat fact ors ot her t han ab und ance o f prey (ie. , w eather
conditions) also may influence migration timing of r aptors. Neotropical migrant songbird species passed well
before accipiters, though both share the same winter ranges.

401     Benson & W inker
A hig h-latit ude c irc us: Tim e-energy jug gli ng an d t he app earanc e of super ph enot yp es amo ng l ong -dist ance av ian
migrant s. A NNA-M ARIE BENSON* , Alaska Bird Observatory, Fairbanks, AK, and KEVIN WINKER, Univ. Alaska
Mus. , Fairbanks AK.
          At high latitudes, avian migrants must molt, fatt en, and migrate under temporal constraints that are
severe relative t o populat ions at l ow er latit udes. We examined w hether hig h-latit ude tim e const raints c ause
some passerine migrants to overlap energetically demanding events. Data were collected on molt and
subcutaneous fat stores in Fairbanks, Alaska (64°50' N 147°50' W), during autumn migration from 1992 t o
200 0. Some individuals of all 14 species examined overlapped energetically demanding events in interior
Alaska. We found a strong negative relationship between species-level autumn passage dates and the amount
of molt -migration and molt-fatt ening overlap. Wit hin species, no individual completely overlapped peak molt
w it h maximum f at st orag e, b ut several i ndi vidual s approac hed t his theor etical max imum. Sev en species
pro duc ed in div idu als w it h more t han 6 0% overlap of molt and f attenin g: Sw ainson' s Thrus h, Orange-c row ned
Warbler, Yellow Warbler, Yellow -rumped Warbler, Wilson' s Warbler, Whit e-crowned Sparrow, and Lincoln's
Sparrow . W e suggest these indi vidual s display a super phen ot yp e. The f requ ency and i nt ensity of overlap
w it hin ind ividu als at our st udy sit e has not been show n in prev iou s st udi es documenting molt -f attenin g ov erlap
or m olt -mi grat ion overlap . These f ind ing s st and i n st ark con trast to t he co mmonl y held assum pt ion that
passerine birds usually segregate energetically demanding events during the annual cycle.

402     Long
Linking the wintering and breeding seasons in Hermit Thrush: Winter diet and the timing of spring migration.
JENNIFER A. LONG, Dept . Biol. , So ut heast ern Lo uisiana Univ ., Hamm ond , LA .
          The timing of spring migration is an important aspect of the annual cycle in migratory songbirds, since
timing of arrival on the breeding grounds influences reproductive success. This study is concerned wit h how
w int er di et inf luen ces p hy sic al co ndi tion and spri ng m igr atory act ivit y of Herm it Thr ushes. Herm it Thr ushes
w intering in se. Louisiana rely on abundant f ruit resources for w inter survival, but f orage exclusively on
arthropods just prior to spring migration. To t est w hether this dietary swit ch affects t he timing of migration,
19 Herm it Thr ushes w ere kept in i ndi vidual cages in an out doo r av iary from Feb to A pr i n 2 00 0 an d f ed eit her
an exclusive arthropod diet or a combination of fruit and arthropods. Body mass and furcular fat w ere
measured every 3 d. Nocturnal migratory activit y, or Zugunruhe, w as measured by resting the low er perch on
2 microsw itches w ired to a counter. I found that birds on an exclusive arthropod diet attained greater amounts
of subcutaneous fat and gained more mass than fruit -eating birds. Birds wit h greater body mass, how ever, did
not t end to display more migratory act ivity . Birds on an arthropod-only diet did tend to init iate Zugunruhe
earlier than fruit eating birds. Age or sex did not explain the large amount of individual variation in migratory
activity in t his species. I believe that food availability may have to be extremely low to aff ect migratory
403      Tallman & Tallman
Summary of 20th Century South Dakota bird banding. D. A. TALLMAN* and E. J. TALLMAN, Dept. Biol.,
Nort hern Stat e Univ., Aberdeen, SD.
        Since the early 192 0' s through Aug 1 999 , the Bird Banding Laboratory of t he U.S. Fish and Wildlife
Service has 8171 2 banding recovery records for 1 53 species of birds found in South Dakota. Alt hough,
comparing species, a slightly higher percentage of passerines than nonpasserine species have recovery records
(54% vs. 5 6% ), the vast majo rit y of ind ividu al rec ord s are of non passerin es (9 6% ). Thi s inequal it y is due to
the vast effort in banding waterfow l made by the Federal Wildlife Refuges. Of all recovery records
(nonpasserines and passerines), 9 0% belong t o th e family Anat idae. Three species of du cks (M allard -- 39 %;
Canada Goose -- 34% ; Snow Goose -- 13% ) constitut e the majority of t he Anatid recoveries (86% , or 35 % of
all banding recov eries).

404      Murphy
Agri cult ural land use practices and avian population t rends in the eastern Unit ed States. M ICHAEL T. M URPHY,
Dept. Biol ., Portl and St ate Uni v. , Port land, OR.
           Habitat loss is gener ally argu ed t o be t he cau se f or o ngo ing pop ulat ion decl ines of many bird spec ies
that breed in grassland or early successional habitats t hat are typical of the agricultural landscapes of t he e. and
cent ral U.S. How ever , d irec t evidenc e exist s f or o nly a hand ful o f spec ies. I at tempted t o rel ated avian
population trends on a state-by-state basis (using Breeding Bird Survey population trend estimates: 1980 -
19 99 ) t o ch anges in l and u se inform ation that w ere der ived f rom U.S. Depart ment of Agri cult ure Cen suses
(198 7 - 19 97). Roughly ¾ (42 of 54) of the species exhibited at least one statistically significant relationship
bet w een po pul ation trend s and chan ges in lan d use. Past ure ac reage w as t he singl e-mo st impor tant var iabl e:
15 of 54 species exhibited a significant relationship wit h change in the amount of unharvested (but grazed)
past ure ( 11 of 15 posit ive). In o ther w ord s, pop ulat ion decl ines tended to f oll ow losses of unh arv est ed
pastur e. Oth er import ant v ariables included changes in t he acreage of farm " w oods" , cov er crops,
conservation reserve program land, and row c rops, and the number of cow s (number of significant relationships
w ith each was 6, 6 , 4, 4, and 5, respectively). These data provide evidence of a strong link betw een the
availabilit y of breeding habit at and popu lation trends f or bot h short - and long-dist ance migrant s, and sugg est
that changes in agricul tur al practi ces are affect ing popu lation trends o f m any species.

407      Boughton
* Post-wildfire effects on bird species in an Australian wet schlerophyl forest. RAOUL BOUGHTON, Dept. Biol.,
Univ. Memph is, M emphis, T N.
           The impact of w ildfire on birds has been under studied and the long-term effects are generally
unknow n. Long-term pre- and post-fire studies that monit or avian populations allow insight s into the impact of
w ild fire o n bi rd s peci es. In 1 98 3 a w ild fire b urn ed a 2 00 ha area o f w et sch lero phy l f ores t w it hin the Barr en
Grounds Nature Reserve, New South Wales, Australia. A banding study to monit or species composition,
movements and longevity w as underway and allows pre- and post-fire comparisons of species composition.
Between 1982 and 1999 banding data were collected from over 12,000 birds comprising 72 species at a 0.4
ha site adjacent t o the Barren Grounds Bird Observatory. Birds were captured using mistnets stationed along
fores t tracks. Relati ve f requ ency , w eigh ted w it h cap ture ef fort , w as calculat ed f or al l species each q uart er
annum. The 17 yr of data have provided insight into t he habitat use of many bird species in relation t o a
wildfire event. For example, species that are typically associated with dry woodlands and heathlands, e. g.,
Chestnut -rumped Heathw ren, Flame Robin, and Whit e-eared Honeyeater, used thi s w et f orest sit e for t he first 5
yr po st-f ire. Habit at generalist s, such as Whit e-brow ed Scrubw rens, Eastern Yellow Robins, Eastern Spinebills,
and New Holland Honeyeaters w ere able to use the site continuously over the 17 yr. These generalists are the
most com mon species across all h abit ats of the res erv e. A t 14 yr post-f ire, w hen t he bu rned area had been
re-colonized by rainforest plants that reached 2 m in height, ground- dw elling insectivorous species of the
rain fores t reapp eared. For ex ample, Bassian Thr ush, So ut hern Logr unn er and Yellow -t hro ated Scrub w ren
becam e resident at 15 yr post-f ire. These dat a iden tif y t he di fferent requ irem ent s of repr esent ative bir d species
in r elat ion to t he im pact s of w ild fire an d aler t land managers abou t the im por tance o f selecting fire r egim es t hat
best pro mot e a suite of species.

408      Millikin
Tracking individual nocturnal migrant birds as they land. RHONDA L. MILLIKIN, Royal Military College of
Canada, Kingsto n, ON.
           The most crit ical feat ure of l andbird mi gratio n is w here the bird s land to rest and f eed. Yet t hese
habitats are not known and therefore cannot be protected. To effectively manage critical stopover habitat, w e
must monit or s hor t-ran ge f lig ht behav iou r and iden tif y indi vidual s t o species. This resear ch w as co ndu ct ed at
Prince Edw ard Point, a projection of land on the north shore of Lake Ontario, important for f all migrants. The
objective of this w ork w as identif y individual landing birds so to provide an unbiased and efficient t ool for t he
selection of critical stopover habitat. By combining the all weather tracking capability of marine radar w ith t he
species identif icatio n capabilit y of an array of micro phones, n ight fli ght calls w ere matched w ith radar track s to
identify landing birds. This method w as compared to standard methods of the Canadian Migration Monit oring
Netw ork (mist nett ing and census) for an independent measure of landing birds. This method overcomes the
problems of obtaining a 3 dimensional position of t he bird using basic (i.e., affordable) marine radar and the
diff iculties of locating a moving sound source (i.e., a call from a flying bird). This w ork builds from t he
pioneering efforts of W.R. Evans and radar enthusiasts like K. P Able, S. A. Gauthreaux, R. P. Larkin, W. J.
Richardson, C. R. Vaugh and T. C. Williams. This experiment is the first successful attempt t o identify tracks
of noct urnal migrant birds using the fusion of radar and acoustic data. This has potential benefit to
conservation and aviation safety, by providing an expeditious and unbiased means of determining areas of
crit ical impor tance.

409      Brua, Clark, Howerter, Anderson & Devries
Wit hin-year habit at f idelit y of nesting female Mallards. ROBERT B. BRUA* , Dept. Biol., Univ. Saskatchew an,
Saskat oon, SK; ROBERT G. CLA RK, Canadian W ildl if e Service, Sask ato on, SK; DAV ID W. HOWERTER,
MICHAEL G. ANDERSON, and JAMES H. DEVRIES, Inst . W etland & Wat erf ow l Res. , Stonew all, M B.
            Many factors purportedly affect avian breeding dispersal, but little is known about within-year fidelity to
nest ing habi tats, pri marily due t o ou r in abil it y t o det ect all n est ing attempts. Ther efore, w e used uni que d ata
collected from radio-marked female Mallards to: (1 ) estimate wit hin-year habitat f idelity among competing
habi tat typ es and (2) elucid ate fact ors affect ing w it hin -year nest-site habi tat fidel it y. Dist ance d ispersed
follow ing nest failure did not vary among habitat t ypes. However, cont rary to expectation, f idelity to nesting
habitats w as positively associated wit h dispersal distance, and was unrelated to previous nest duration.
Apparent nesting habitat fidelity w as greater for planted nesting cover than for any other habitat; how ever,
fidel it y t o pl ant ed co ver w as not great er t han ex pect ed by chan ce. At 2 spat ial scales, t he use of plan ted
nesting cover w as positi vely relat ed to t he amount of p lanted cov er surround ing t he previous nest . These
results h ave import ant im plicat ions f or conf iguring landscapes as part of prairie habit at rest oratio n init iativ es.

410      Dietrich, Melvin & Parrish
Seabird bycatch in Alaska longline fisheries: A closer look. KIMBERLY S. DIETRICH* , School Aquatic & Fish.
Sci., Univ. Washington, Seattle, WA, ED MELVIN, Washington Sea Grant, and JULIA K. PARRISH, School
Aquatic & Fish. Sci., Univ. W ashington.
        Inc iden tal t ake (o r by cat ch) of seabirds in l ong lin e fisheries has been li nked t o t he dec lin e of several
seabird p opu lat ion s w orl dw ide. In A lask an w aters, more t han 2 ,0 00 vessels in d emersal long lin e fisheries set
over 200 million hooks/yr. M anagement efforts t o reduce bycatch are driven by concern for all North Pacific
albatross, especially the highly endangered Short-tailed Albatross. How ever, the observed bycatch is
dominated by Northern Fulmars (60 - 75% ), follow ed by gulls (14 - 18% ), albatross spp. (3 - 10% ) and
shearw ater spp. (2 - 6% ). Bycatch has been characterized on a gross scale but meso-to-fine scale patt erns are
unexplored. Using data from t he NMFS Groundfish Observer Program, w e contrast 199 5 - 20 00 cat ch rates for
the Gulf of A laska and Bering Sea/A leutian Islands f isheries by seabird t axa (5), g eographic region (> 10 ),
vessel t yp e (2), target fish spec ies (5 ) and season (> 10 ). Alt hou gh b yc atch rat es f luc tuat e w ild ly in t ime,
trend s per sist. Cat cher pr ocessors w hic h pr ocess f ish on b oard have con sistent ly hig her b yc atch rat es t han
catcher vessels which deliver fish t o land-based processors regardless of fish species targeted. In the Pacific
cod f ishery, the major ity of b ycat ch is f ulmars w hereas in the halibut and sablefish f isheries, it is albatross.
Vessel type and target species differences may be driven by species-specific geographic variation including:
proximity to colony, use of the marine environment and migration rout es. These patt erns suggest that fine
scal e ext rapo lat ion takin g geo grap hic area and season in to acc oun t may more accurat ely est imate bycat ch rat es
and fine-tune management.

411     Jamieson
Energy reserves of two w intering populations of Common Eiders. SARAH E. JAMIESON, ACWERN, Dept. Biol.,
Uni v. New Brunsw ick, Frederict on, NB.
        I examined and compared the w inter endogenous reserves of Common Eiders in southw est Greenland,
Denmark and the Belcher Islands, Nunavut, Canada. Birds were collected betw een 9 Feb and 20 Mar 2001.
Breast muscle, l eg muscle, abdo minal f at pad and f emoral fat pad w ere extract ed and w eighed. Body m ass
and ex ternal morph ometric measu rement s w ere also rec ord ed f or eac h bi rd i n or der t o co rrect for b ody size.
Eiders over-w int erin g in the Belcher I slands had more r eserv es t han c onspeci fics in Gr eenlan d. The p oly ny as
and floe edges of t he Belcher Islands are only 1 of 2 areas wit h open w ater in Hudson Bay during t he w inter. It
is thought that in years when these w aters freeze-over large numbers of ducks starve to death. This leads me
to believe that eiders of t he Belcher Islands carry more reserves because in years of heavy ice they are unable
to relocate and must rely on these endogenous reserves while attempting to wait out extreme ice conditions.
Eiders over-w int erin g in the w aters surr oun din g Green land have the op tion of moving t o areas of open w ater
during years of h eavy ice. Reserves can be costl y t o maint ain and transpo rt, theref ore for Greenland eiders,
there may b e litt le benefit of st oring larg e amounts o f reserv es throu ghout the w int er mont hs.

412     Agatsuma, Ramenofsky & Landys
Definit ions of migrat ory b ehavior and eff ects of foo d restr ict ion on c aptiv e Whit e-crow ned Sparrow s. RENEE
AGATSUMA* , MARILYN RAMENOFSKY and META LANDYS, Dept. Zool., Univ. Washington, Seattle, WA.
          Piersma (1987, Limosa 60: 185 -194 ) has classified migratory species according to their migration
strat egies: jumpers, migrant s that undert ake long bout s of f light betw een refueling st ops and hoppers,
migran ts t hat fly onl y hour s bef ore s topp ing to ref uel. In jumpers , m igr atory rest lessness (M R) in capt ivit y has
been defined as "w hirring" (Gwinner and Czeschlik 1978, Oikos 30 : 3 34 -372 ), but to dat e has not been
defined in a hopper. Gambels White-crowned Sparrows (GWCS) are an example of a hopper type of m igrant.
We developed a complete ethogram for the migratory behavior of GWCS during the autumn and spring
migratory periods. Even though GWCS spent most of their time resting, daytime and nighttime behaviors
during both migratory seasons were qualitatively dif ferent. M R occurred only at night and w as characterized by
looking up and look-up flight, a behavior similar to "w hirring" in which a bird looks up and flaps its wings.
Daytime activities primarily included jumping and feeding. Because MR increases in response to food restriction
in warblers, which use the jumping migration strategy (Gwinner et al 1988, Oecologica 77: 321-326), we
tested the effects of food restrict ion on the behaviors associated w ith M R (as defined in this study) in a
hopping migrant. In contrast to previous studies, w e found that food restriction in GWCS increased daytime
activity , but did not affect M R. These results suggest that species using the hopping strategy of migration,
suc h as t he GWCS, m ay increase f orag ing dur ing the day to suppo rt the ener get ic d emands of a relat ively fixed
migratory program.

413     Stotz, Willard & Hennen
The effect s of ligh ts on c ollisions w ith bu ildings by m igrant bir ds: a case study. DOUGLAS STOTZ, DAV ID
WILLARD and MARY HENNEN, Field Mus., Chicago, IL.
          An estimated 100 ,00 0,0 00 birds die annually follow ing collisions wit h the w indow s of buildings. Many
of these deaths are birds migrating at night that strike large buildings, such as skyscrapers. Evidence from a
variety of sources indicates that lights on these buildings confuses the migrating birds and increases the
number of birds killed. However, there are relatively f ew st udies with hard data demonstrating the effect of
lights on bird strikes. We have been collecting birds that have collided with a Chicago lakefront building since
197 8. The building is checked daily during migration in both spring and fall. During the early years of t he
study, lights w ere alw ays on in the building. Over the last 2 y r the light regime has varied across days, and
betw een parts of t he buildings. We have recorded the number, and location around the building of all birds
foun d dead . A lt hou gh t here i s subst ant ial d ay t o day var iat ion in t he nu mber of bir ds k ill ed, it is c lear t hat
lights being off reduce the kill be 75% or m ore. When the lights or partially on, dead birds are concentrated in
areas w it h il lum inat ion , sugg est ing that even part ial r educ tion of lig ht s w ill decr ease t he nu mber of bir ds k ill ed
at buil dings.

414      Revis, Dyer & W aller
Possible bactericidal and f ungicid al action o f ant chemicals on c ommon micro bes of avi an feathers. HANNA H
C. REVIS* , BRYAN DYER and DEBORAH A. WALLER, Dept. Biol., Old Dominion Univ., Norfolk, VA.
          Ant ing behavior in songbirds involves birds applying ants to t heir feathers. This may be a grooming
mechanism employed as a type of self-medication. It has been suggested that chemicals secreted by ants
dur ing ant ing behav ior may act as bac teric ides and f ung ici des, redu cin g t he nu mber of bact erial and f ung al
colon ies in f eathers . W e test ed t his hypot hesis w it h micr obi al in hib it ion trial s co ndu ct ed on bact erial and f ung al
colonies commonly found on avian feathers. Whole w orker ants, homogenized ants, and hexane extracted ant
chemicals as w ell as 88% for mic acid w ere tested f or inhib iti on against Bacillus subtilis (mo tile, gram -posit ive,
rod bacteria), Chaetomium globulosum (Ascomycetes), and Penicillium chrysogenum (Deuteromycetes). Five
ant genera w ithin t he Myrmicinae and Formicinae subfamilies produced no inhibition in any trials. Formic acid,
w hic h is pro duc ed by som e ant spec ies, inh ibi ted all microbi al gr ow th w it hin an average zo ne of 30 .7 mm after
8 hr for B. subtilis, 32 .0 mm aft er 2 d for C. globulosum, and for 22. 0 mm after 2 d f or P. chrysogenum

415      Scheuerlein & Ricklefs
Prevalence of blood parasites in European Passeriformes: The effects of habitat, population size and
geographical distribution. ALEXANDER SCHEUERLEIN and ROBERT E. RICKLEFS, Dept. Biol., Univ. M issouri-
St. Louis, St. Louis, MO.
        The factors explaining differences in prevalence of blood parasites in birds are still poorly known. We
simultaneously assessed the eff ects of habitat, population size, geographical distribution and life history
param eters o n pr evalence o f plasmodiu m, hemopr ot eus, leuc ocyt ozoo n and try pano som a in European
passeriformes. We found t hat most of the variation in parasite prevalence resided on the species level wit hin
genus and family. M oreover, unlike a previous study on European raptors, we f ound no effect of breeding
habitat on prevalence. Likew ise, migratory status of t he birds (resident, partial migrant, long-distance migrant)
w as unable to explain variation in prevalence data. The only effect discernible w as a geographical trend: birds
breeding in t he nort h generally have high er prevalence levels th an birds t hat breed in sou thern regions.
Inconsistencies of our findings with previous studies are discussed.

416      Wikelski, Martin, Bernier, W alters, Pater, Raim & Cochran
Examining f oraging behavior of Bicolored Ant birds using mic rophone-assisted telemetry . M ARTIN WIKELSKI,
LYNN MARTIN II*, ELIZABETH BERNIER, ANNIKA WALTERS, Dept. Ecol. & Evol. Biol., Princeton Univ.,
Illinois, Urb ana-Champaign IL.
          Bicolored A ntb irds (Gymnopithys leucaspis) forage on arthropods disturbed by army ant swarms, a
highly mobile and patchily-distributed food source. To maximize food intake, individual antbirds must therefore
trade-off t erritorial defense wit h optimization of f oraging position on an army ant sw arm. Preliminary data from
a new microphone-assisted radio telemetry technique indicate at least 3 diff erent strategies for t his foraging
behavior on the Limbo Plot on Pipeline Road, Gamboa, Panama: i) stationary, w here an individual follow s 1 ant
sw arm all d ay, ii) mobil e, w here an ind ividu al mo ves bet w een sev eral an t sw arms in a day, an d ii i) int ermediat e,
w here an ind ividu al visits f ew er sw arms t han a m obi le bi rd, but more t han t he st ation ary typ e. A ssuming t hat
song ser ves the sam e func tion on an t sw arms as it does for t errit ori al def ense in ot her s peci es, song rat es
should be gr eatest for s tation ary bir ds c hoo sing t o f oll ow (def end) 1 p art icu lar sw arm. Indeed , w e foun d t hat
song rat e of the st ation ary bir d on sw arms w as gr eater t han t he in termediat e bir d, w hic h in turn w as gr eater
than t he mobi le bi rd. Song rat es, how ever , d id c hang e w it h du rat ion on sw arm for i ndi vidual bir ds; song rat es
of stat ionary birds decreased, w hile song rates of intermediate and mobile birds increased. Our findings
support the loose territoriality and w ithin-sw arm hierarchical organization of foraging bicolored antbirds initially
proposed by Willis & Oniki (1978, Ann. Rev. Ecol. Syst. 9: 24 3-26 3) and sugg est that microph one-assist ed
radio telemetry may enhance fut ure studies of avian behavior in the field.

417     Schreiber, Schenk & Doherty
The ef fect of chemical w eapo ns incineration on t he survi val rates of Red-t ailed Tropicbirds. E. A. SCHREIBER* ,
Bird Divs. , Natl . M us. Nat . Hist ., Washingt on, DC, GARY A. SCHENK, Alexandria, VA and PAUL F. DOHERTY,
Jr., Bird Banding Lab., USGS, Laurel MD.
         In 1 99 2, th e Joh nst on A to ll Chem ical A gent Dispo sal Sys tem (JA CADS) began incinerat ing U.S.
chemical weapons stockpiles on Johnston A toll (Pacific Ocean) where about 50 0,0 00 seabirds breed, including
Red-tailed Tr opicbir ds (Phaethon rubricauda). We hy pot hesized that su rviv al rates of birds w ere low er in th ose
nest ing dow nw ind of the in cin erat or s mokest ack com pared to t hose upw ind , an d t hat bir ds m igh t move aw ay
from t he area. From 19 92 - 2 000 w e monitored survival and movements betw een areas upwind and
dow nw ind from t he JACADS facility. We used a multi-strata mark recapture approach to model survival,
probability of recapture and movement. Probability of recapture w as significantly higher for birds in dow nw ind
areas (owing t o greater recapture effort) and thus w as an important ' nuisance' parameter to t ake into account
in modeling. We found no differences in survival betw een birds nesting upwind ( 0. 858 8) and downw ind
(0.8 55 0). There w as no con sistent dif ferenc e in m ov ement rat es bet w een up w ind or d ow nw ind areas f rom year
to year: diff erences found may be attributed to dif fering vegetation grow th and human activities betw een the
areas. Our resu lt s suggest that JA CADS has had n o do cumentable i nf luen ce on the surv ival and year to year
movement of Red-tailed Trop icbird s.

418     Iverson
Assessing plu mage char act eristics and b urs al dep th as indi cat ors of age cl ass in Sur f Scoters. SAMUEL A.
IVERSON, Dept . Bio l., Simon Fraser Un iv., Burnaby, BC.
         Accurat e det erminat ion of the age c lasses is of crit ical impor tance i n avian d emograp hic and b ehavior al
studies. Post-juvenile molt can provide a useful method for separating age classes, as there is often a color,
pattern, or shape difference betw een sub-adult and adult feathers. Surf Scoters are a delayed breeding seaduck
of conservation interest, w hich exhibit unique sub-adult and adult plumages. From Oct 199 9 t hrough Jul 20 01,
I detailed the sequential changes in body feathers and w ing feathers, as well as in bill, foot, and eye color,
w hich allow Surf Scoter gender and age classes to be distinguished. The process by which immature Surf
Scoters acquire adult plumage is a gradual one, and is complete by 14 mon of age. The timing and duration of
different plumages is known to vary w ith hatch date, physical condition, diet, and environment. Further,
sampling error may be comm on w hen trait s vary su btl y. In order t o gauge the accur acy of age class
determinations, I assessed the agreement betw een estimations based upon plumage characteristics wit h
est imation s based up on t he deg ree of bur sal inv olu tion . M easur ements of the bu rsa o f Fabriciu s (an
immunosuppressive organ that forms as a sac in the cloaca, which is present in juveniles and regresses as the
bird matures) and plumage scores were recorded for Surf Scot ers captured in floating mist net s during the
winter of 2000 - 2001 (n= 90). Analysis suggests age class determinations made by the tw o techniques are in
strong agreement.

419     Pogue & Schnell
Effect s of l ocal and landscape habit at charact eristic s on prairie bir ds. DA RRELL W. POGUE* , Dept. Biol., Univ.
Texas at Tyler, Tyler, TX; and GARY D. SCHNELL, Sam Noble Mus. Nat. Hist. and Dept. Zool., Univ.
Oklahoma, Norman, OK.
         We studied the influence of local- and landscape-level habitat factors on breeding-bird communities in a
prairie landscape in sw. Oklahoma. Local habitat characteristics were sampled at study plot transects, and
land scap e measu res w ere comput ed f rom a land -co ver map dig it ized f rom aerial pho togr aphy . Habit at
regr ession m odel s using loc al and land scap e habi tat com pon ent s explai ned 7 9% of the variat ion in b ird spec ies
ric hness and 41 % of the variat ion in ab und ance. Bird spec ies rich ness w as best explain ed by land scap e habi tat
variables, w hile overall bir d abundance w as related to a com binati on of local and landscape variables.
Abun danc e of w ood land and g rassland /savanna species w as acc oun ted f or b y local hab it at var iabl es
repr esent ing vegetation and a slop e grad ient . M odel s f or t yp ical w ood land spec ies Tuf ted Ti tmice, Red-eyed
Vireos, Great Crested Flycatchers, and Northern Cardinals were explained primarily by local vegetation
charact eristic s (r2 = 0.5 2 - 0. 74). A combination of local and landscape habitat f eatures accounted for the
variation (local, 42 - 10 0% ; landscape, 0 - 6 0% ) in abundance of 4 t ypical grassland/savanna species: Field
Sparrow , East ern M eadow lark , Gr asshopp er Sparrow , an d Dickcis sel. In g eneral , av ian-communi ty indi ces
w ere relat ed bo th t o lan dsc ape and loc al hab it at char act eristics . Ecolog ical gro ups of bir ds and in div idu al
species were accounted for primarily by local habitat characteristics.

420     Collias
Foods of the Red Junglefowl in its native home in Asia. NICHOLAS E. COLLIAS, Dept. Organ. Biol., Ecol. &
Evol., Univ California, Los Angeles, CA.
         The Red Junglefowl (Gallus gallus) is the ancestor of the Domestic Fowl w hich is nutritionally t he
w orld' s most studied animal, but there has been litt le systematic study of t he foods of the Red Junglefow l in its
native habitat. Forty-one Red Junglefowl w ere collected in west central and peninsular Thailand in wint er and
spring and their crop and gizzard contents analyzed. Alt hough not t he harvest season, the principal plant food
w as rice (2 2% of bir ds). Ric e is t he main c rop of slash-burn agri cult ure w it h w hic h Red Jung lef ow l are o ften
associated in southeast Asia. They eat a great variety of ot her plant foods providing a reliable and ready source
of energy from carbohydrates and fats. Various animal foods including insects, spiders, and small lizards
furnish a good source of protein and essential amino acids. Termites (66% of birds) and ants (61%) w ere
eaten most frequent ly. In th e crop of 1 co ck col lected at Tasek Bera, Pahang Prov. , M alaya, almost 10 00
termi tes (Macrotermes carbonarius) w ere count ed. Termit es are also comm only eat en by dow ny ch icks w hose
numbers t end to c oincide w ith the onset of t ermit e fligh ts, as w e saw in India. Fragment s of sn ail shell (29 %)
gave a good source of calcium.

421     Scheiman, Bollinger & Johnson
Eff ect s of leaf y spur ge in fest ation on b reedi ng b ird s of the Shey enne Nation al Grassland, ND. DA NIEL M.
SCHEIMAN* , ERIC K. BOLLINGER, Dept. Biol. Sci. , Eastern Illinois Univ., Charleston, IL; and DOUGLAS H.
JOHNSON, Nort hern Prairie Wild. Res. Ctr ., Jamesto w n, ND.
          Leafy spur ge (Euphorbia esula) is an exot ic i nv asiv e w eed in the no rt hern Great Plains. W e exam ined
the effects of leafy spurge infestation on densities and nest success of breeding birds in grasslands on the
Sheyenne National Grassland, ND, during 19 99 and 20 00. We categorized spurge-infested grasslands into 3
levels of inf est ation : l ow , m ediu m, and h igh . Of the 8 most abun dant gras sland bi rds , o nly Grasshop per
Sparrow and Savannah Sparrow densities w ere significantly low er on high-spurge survey points than on low-
and medium-spurge points. However, none of t hese species occurred at highest densities on high-spurge
points. Bobolinks and Savannah Sparrows w ere significantly negatively correlated with spurge infestation.
Low-spurge plots contained more nests (57) t han high-spurge plots (34), how ever, there were no consistent or
monotonic relations betw een daily mortality rate and spurge level. Birds may choose microhabitats based more
on c harac teristics of vegetation st ruc ture t han o n par ticu lar p lant spec ies. Assuming spur ge can alt er
vegetation structure t o the detriment of grassland birds, infestation may not be high enough to show a strong
or w idespread negative ef fect on bird c ommun ity parameters.

422     Dunn, Hussell, Hobson & Wassenaar
Deuterium isot opes of f eathers help ident ify natal areas of f all migrant s. ERICA H. DUNN* , ERICA H. DUNN* ,
Canadian Wi ldlife Serv ice, Ott aw a, ON; DAVID J. T. HUSSELL, Ministry of Natural Resources, Peterborough,
ON; KEITH A. HOBSON and LEN I. WASSENAAR, Environm ent Canada, Saskatoon, SK.
         Bird observatories across southern Canada conduct st andard migration counts t o monitor t rends in
boreal populations of songbirds which breed north of t he Breeding Bird Survey coverage area. However,
information on w hat population is monitored by each station is fragmentary, and is needed to determine
adequacy of sampling and to pinpoint problem areas when important declines are detected. Ratios of the stable
isotopes deuterium:hydrogen in the feathers of birds reflect ratios in rainfall of the region w here feathers were
gro w n. Isotopes in f eathers col lect ed f rom Hat ch-Year migran ts in fall w ere anal yzed for 5 w arbl er sp ecies at 5
sites (1 each in British Columbia, Manitoba and Nova Scotia, and 2 in Ontario). Contours of equal isotope
ratios run largely E-W (although w ith a larger N-S component in western Canada), so power to discriminate bird
populations longitudinally is rather low. Nonetheless, there are consistent diff erences in isotope profiles among
stations. Stations west of t he Great Lakes sample fall migrants moving primarily southeast, w hile southeastern
Ontario may also get migrants moving south and southw est. The Nova Scotia site samples a higher proportion
of bir ds f rom Quebec t han w as pr eviously recogn ized. When co mbined w it h ban d rec ov eries and o ther
inf ormat ion, i soto pe ratios c an provid e valuable new inf ormat ion on br eeding ground origin o f m igrant s.

423     King & Byers
Massachusett s pow erline corridors suppor t sour ce populations of an early-successional shrubland bir d. DA VID
I. KING* , USDA For. Serv. Northeast. Res. Station, Amherst, MA; and BRUCE E. BYERS, Dept. Biol., Univ.
Massachusetts, Amherst, MA.
         Recent population declines among early-successional shrubland birds have been associated w ith
declines in habitat availability. Forest succession has eliminated shrublands in many locations, but pow erline
corridors constitut e a potential reservoir of shrubland breeding habitat. To evaluate the potential conservation
value of these c orr ido rs, w e st udi ed a rep resen tative early -successional shr ubl and b ird , t he Chestnut -sided
Warbler, in powerline corridors at 2 sites in w. Massachusett s. Our goals were to (1) measure the survival and
reproductive success of birds nesting in pow erline corridors, (2) test f or edge-related increases in nest predation
that might compromise the health of bird populations in powerline corridors, and (3) evaluate the viability of
populations nesting in powerline corridors. Nest survival was significantly lower near edges in one of tw o years
for w hich distances from nests to edge were measured; how ever, reproductive success w as sufficient at both
sites to maintain the populations. We conclude that powerline corridors can support viable populations of
early-successional shr ubland bird s.

424     Gillies
* Lost birds: Test ing c orr idor eff icacy us ing b ird hom ing i n t he dry fo rest s of Cost a Rica. CA MERON S.
GILLIES, Dept. Bi ol. Sci., Uni v. Albert a, Edmont on, A B.
         Tropical forest destruction and fragmentation may be disrupting the movement of forest birds.
Corridors hav e been w idely adv ocated as a w ay of mitigating t hese impacts in the t ropics and elsew here. By
relocati ng t errit orial indiv iduals of 2 species of for est bird s 1 - 2 km along r iparian corr idors, along f encerow s,
and across open fields, and then follow ing their return w ith radio-telemetry, I w as able to test t he eff icacy of
riparian corridors and fencerow s in facilitating movement. Thirt een birds were follow ed up to 12 km on their
return wit h as many as 70 recorded locations in each path. Birds returned faster and were more likely to return
in r ipar ian c orr ido rs t han f encer ow s but the suit abil it y of fencer ow s f or f acil it ating movem ent appear s t o di ffer
betw een species. Rufous-naped Wrens (Campylorhynchus rufinucha), w hich are less dependent on for ests,
used fencerow s in their returns and the corresponding times to return w ere intermediate betw een their return in
riparian cor ridor and o pen field r elocation s. By con trast , t he Barred Ant shrikes (Thamnophilus doliatus) avoid ed
using fencerows for their returns in favor of longer routes in riparian corridors, that also had few er gaps in
forest cover. These preliminary results suggest that to f acilitate the movement of the most f orest-dependent
species, fencerow s may not be sufficient and large forested corridors may be required.

425      Mat his, Colwell, Leeman & Leeman
Spatial distributions of non-breeding Long-billed Curlews in intertidal habitats: Scale-dependent patterns.
Humboldt State Univ ., Ar cat a, CA .
          Understanding a species' spatial distribution is important in applied ecology and key insights derive from
analyses conducted at several spatial scales. Consequently, w e quantified low-tide distributions (uniform,
rand om and ag greg ated) of non -br eedin g Lon g-bill ed Cur lew s at 3 spat ial scales, w it hin ind ividu al ho me ranges
(1 - 8 ha), in a local estuary (~50 ha), and regionally in intertidal habitats of Humboldt Bay (62 km2 ), CA.
Individuals (n = 8) most of ten (75% ) foraged in territories in a manner that produced a uniform distr ibution;
patterns tended tow ard random (16% ) and aggregated (8% ) w hen individuals moved over larger areas. In the
estuary, curlew s often (73% ) were distributed uniformly, although patt erns became random (27% ) when more
curlews w ere present during summer and autumn. At this scale, uniform patterns w ere a consequence of
territoriality. Across Humboldt Bay, curlews w ere consistently aggregated in certain areas and absent from
oth ers, suggest ing t hat habit ats varied in qualit y. At each spatial scale, f ood pro bably had t he stron gest
influence on these patterns; predation risk played a relatively minor role in determining dispersion patterns.
Overall, w e estimate that 2 50 - 3 00 cur lews w inter at Humboldt Bay, w hich represents approximately 1% of
the species' population.

426      Styring
* Logging and its effects on a diverse guild of Malaysian woodpeckers. ALISON R. STYRING, Dept. Biol. Sci. &
Mus. Nat . Sc i. , Lo uisiana St ate Univ. , Bat on Rou ge, LA.
         Woodp eckers ar e con sidered s ensitive to lo ggi ng an d def ores tation becau se t hey nest in t ree cavities
and often forage on large trunks or standing dead trees. The diverse w oodpecker community found in t he
low land rainforests of Peninsular Malaysia was investigated in 5 and 10 yr-old selectively logged forest and
virgin forest using census, foraging, and habitat availability data. Relative abundance diff ered significantly
betw een logged and unlogged forest and also between logged stands of different ages. Although w oodpeckers
did not respond uniformly t o logging, they w ere generally less abundant in the 10 y r-old forest. This trend,
supported by an earlier study in 45 y r-old managed forest, reflects a lack of heterogeneity in managed stands
w ith relatively few snags and treefall gaps. Important f oraging substrates were identif ied, and, combined with
subst rate availabilit y dat a, w ere effect ive in predic tin g changes in abundance of indiv idual species across f orest
typ es. The c urr ent prac tice o f main taini ng p atches of virgi n f ores t w it hin log gin g co ncessions seems to be an
effectiv e way of conserving a diverse community of w oodpeckers and other cavity-nesting birds. Maintaining a
por tion of snags and " ov ermature" trees in lo gged st ands may help inc rease t he abu ndan ce of spec ies
dependent on dead wood for foraging or nesting throughout the reserve.

427      Stout & Cooke
Timing and location of molt in Horned, Red-necked and Western grebes in North A merica. BONNIE E. STOUT*
and FRED COOKE, Dept . Bio l. Sci. , Simon Fraser Un iv., Burnaby, BC.
         We review and extend information on timing and location of molt in Horned, Red-necked, and Western
Grebes, focusing on molt and regrowt h of remiges (when grebes are flight less). We add observations and
analysis of grebe phenology relative to molt t iming and location f or each species. We used data from t he
literature, and surveys of Boundary Bay, BC, conducted by Bird Studies Canada volunteers. We also used our
ow n observations from Boundary Bay and near Yellowknif e, NT, and observations from A laska contribut ed by
James G. King. M olting adult Horned Grebes were located on large ponds near Yellow knife, NT, in Aug 1 999 ,
and hundreds of flightless Horned Grebes were captured on lakes in interior Alaska during late Jul-Aug duck
banding operations in 195 6 - 19 66. Boundary Bay, BC, w as identif ied as a molt site for Red-necked and
West ern Gr ebes, but w as not used as a w int erin g lo cat ion by these 2 spec ies. From inf orm ation to dat e w e
surmise that North American Horned Grebes move to invertebrate-rich interior lakes to undergo the w ing-molt,
while Red-necked and Western Grebes move to certain coastal areas near or within their winter range to molt.
Further surv ey eff ort s are needed across Nort h Am erica to i dentif y imp ort ant mo lt l ocatio ns fo r th ese species so
that grebe molt strategies can be better understood and molt locations safeguarded.

428     DeVault, Stephens, Reinhart & Rhodes
Iden tif ying sour ces o f error f or g rou nd-based and aer ial t elemet ry for w ide-rang ing bir ds. TRAV IS L.
DeVAULT* , Dept. For. & Nat. Resources, Purdue Univ., W . Lafayett e, IN; WA RREN L. STEPHENS, BRADLEY D.
REINHART, Savannah River Ecol ogy Lab., Ai ken, SC; and OLIN E. RHODES, Jr., Purdue Univ.
           We evaluated accuracy and examined sources of error for ground-based and aerial telemetry at the
heavily f orested Savannah River Site. Radio transmitters (beacons) were dispatched at 2 5 locations throughout
the 7 80 00 ha site. A t each loc ation, beacons w ere placed at 1 m and 1 0 m f rom gr ound level. At least 2
bearings were taken wit h a hand-held antenna for both beacon heights at each location. Average absolute
bearing error for 1 m and 10 m beacons was 13. 4 and 11.6 degrees, respectively. The 95% confidence
ellipses for 1 m and 10 m beacons averaged 1132 and 1072 ha in size. Regression analysis revealed that the
density of deciduous trees negatively influenced bearing accuracy for 1 m beacons, w hereas the distance
betw een beacons and receiver stations, the proximity of overhead powerlines, and line-of-site obstructions due
to elevation differences had no significant effects on accuracy of ground telemetry for either beacon height.
Additionally, w e used a Cessna 172 equipped w ith w ing-mounted antennas to locate 1 m beacons. Unlike
gro und telemet ry met hod s, aerial telemet ry w as very accurat e. A ver age li near er ror for aer ial t elemet ry w as
191 m, and the 95% confidence circles averaged 47 ha in size. Habitat type also influenced aerial accuracy:
linear error bet w een estimated and act ual beacon locat ions w as greater fo r beacons placed in decid uous f orests
than i n ot her h abit at typ es. The p rox imit y of pow erli nes had no signif ican t effect on t he acc urac y of aerial
telemetry. In heavily forested habitats, aerial telemetry appears to be much more accurate and practical for
radio-t racking w ide ranging bir ds.

429     Fraser & Stutchbury
Forest fr agmentation effects on the home-range movements of Scarlet Tanagers: Insights from radio-tracking
efforts. GAIL S. FRASER and BRIDGET J. M. STUTCHBURY* , Biol. Dept ., York Univ ., Toront o, ON.
         Territ ory size in bir ds is usually measured by w here a male bird sings and reacts t o playback r ecordings.
How ever , f or m any bir d species, h ome-range size, o r t he ent ire ar ea w here an ind ividu al mo ves (somet imes
w ithout singing) is hard to measure, particularly in canopy dw elling species like Scarlet Tanagers. Using radio
telemetry, w e mapped the home-range movements of Scarlet Tanagers on a large (150 ha) forest plot and on
smal ler (0. 8 t o 3 8 h a) f ores t fragm ent s. All m ales on t he lar ge pl ot w ere pai red an d t heir movem ent s inclu ded
foray s onto nei ghb ori ng t errit ori es and movem ent s pr obab ly relat ed t o t he availab ili ty of frui ting bushes. Males
tracked on forest fragments w ere both paired and unpaired. Home-range movements of paired males included a
male (during chick rearing phase) leaving his fragment to f orage in other areas nearby (~ 10 0 m) and a male
shifting his t erritory (and consequently his home-range) mid-season into a neighboring forest. M ovements of
unpaired (sub-adult) males also proved interesting: some males moved through t he landscape singing in multiple
spots, adop ting a float er st rat egy. Cl early , t hese r adio -t rackin g dat a have pro vided n ew insigh ts t o a species
that is ty pically c onsidered limi ted t o ' int erior' for ests. W e w ill discu ss the co nservat ion imp licati ons of these
fin dings.

430      Doran
* A landscape approach to the causes and consequences of spatial variation in abundance forest birds.
PATRICK J. DORAN, Dept. Biol. Sci., Dartmouth Coll., Hanover, NH.
          Spat ial v ariat ion in t he abu ndan ce of org anisms m ay r eflect spat ial p atterns in t he eco log ical fact ors that
inf luen ce repro duc tive suc cess and surv ival. We used using avian c ensus t echn iqu es t o in vestigat e spat ial
variation in abundance of individual bird species wit hin a 3000 ha forest characterized by strong elevational and
vegetation gradients. Using geostatistical techniques, w e described the spatial distribution of breeding bird
species, vegetation structure and composition, and other ecological factors hypothesized to influence bird
abundance (e.g., food, predators). We then compared distributional patterns of several of t he most common
breeding birds to these ecological factors. Results suggest that species' abundance patterns are influenced by
spatial patterns in the ecological variables considered. For example, in the Black-throated Blue Warbler: 1)
distribution maps show ed variable abundance w ithin t he 3000 ha forest, 2) variograms indicated that 72% of
the sample population variance w as related to spatial autocorrelation over a range of 175 m, and 3) high
blac k-t hro ated bl ue w arbl er abu ndan ce w as asso ciat ed w it h hi gh u nder st ory shr ub d ensity, a param eter
associated w ith variation in reproductive success. These findings illustrate that the variation in abundance of
the Black-throated Blue Warbler can be partially attr ibuted to variation in environmental factors operating locally
w ithin t he 3000 ha forest. Understanding spatial patters in the abundance of organisms and determining the
mechanisms that inf luence such patterns represents a vital step in eff ective conservation and management

431      Calmé& Osorio Dí az
Modeling Ocellated Turkey distribution and abundance using human population data and vegetation land cover.
SOPHIE CALMÉ and MARIO OSORIO DÍ AZ, División Conservación de la Biodiversidad, El Colegio de la
Frontera Sur, Chetumal, Quintana Roo, Mexico.
          Ocellated Turkey is an endemic species from the Yucatan Peninsula. Its distribution and abundance have
been greatly reduced during the last 20 y ears. Principal threats to t he species are linked to t he dynamics of the
region both in terms of t ropical forest degradation and human population grow th. W e modeled the distribution
and abundance of the species based on data on its present distribution and abundance, vegetation land cover,
habi tat con nect ivit y and hu man pop ulat ion densit y. Dat a on d ist rib ut ion and ab und ance w ere in a rast er f orm at
(10 x 10 km grid), and were obtained in 1999 - 200 0 t hrough a survey directed to subsistence hunters (n =
39 4). V eget ation land cov er w as obtained from the 1 99 9 Nation al Forest Inv ent ory , w it h some habit at clas ses
merged to meet our needs. Human population density w as obtained from t he 2000 National Population Count.
Using the kernel method, w e generated areas of varying densities that w ere assumed to reflect anthropogenic
influence around villages w ithin a 6-km radius. The abundance of birds w as modeled through multiple
regr ession u sing veget ation land cov er t yp es, habi tat con nect ivit y (measu red b y a prox imit y index ), and h uman
population density. Habitat connectivit y, population density and amount of f orest habitats in the landscape all
accounted for about 60 % of the variation in the distribution and abundance of t he species. The model could
help pred ict how Ocellat ed Tu rkey popu lat ion s w ill evolve in t he co nt ext of rapi d f ores t dest ruc tion and h uman
pop ulat ion inc rease. Similar models cou ld b e easily adapt ed f or o ther en dang ered l arge f ores t bir ds, suc h as
Crested Guan and Great Curassow.

432      Wiedenfeld, Capparella, Harper, Reinking & Frick
Pestic ide residues present in w int ering Dickc issels fro m t w o localit ies in Venezuela. DAV ID A. WIEDENFELD* ,
Sut to n A vian Res. Cent er, Bart lesville, OK , A NGELO P. CAPPARELLA, Illinois Stat e Univ., Normal, IL, R. GIVEN
HA RPER, Illinois Wesleyan Univ. , Bloomingt on, IL, DAN L. REINKING, Sutton Avian Res. Center, and JEFFREY
A. FRICK, Illinois Wesleyan Univ.
           A large majority of Dickcissels winters in the llanos of central Venezuela. There the Dickcissels
congregate in large flocks of up to 1 .5 million birds, and can become crop pests, especially on rice and
sorghum . A long w ith oth er techni ques, V enezuelan farm ers sometim es use pesticides int ended for i nsect p ests
to kill Dickcissels and prevent the birds from depredating t heir crops. Because of t his inappropriate use of
pesticides, we analyzed pesticide residues in a sample of 26 Dickcissels collected at two sites in Venezuela.
The specimens were tested for presence and amount of 17 organochloride pesticides and breakdown
compounds. Only p,p'-DDE w as detected and measured. Males and females showed no significant diff erence
in t he presence of DDE (P 2 = 0.2 5, df = 1, P= 0.6 19). No individuals from 1 of t he 2 sites contained any
DDE residues, and the difference in presence betw een the 2 sites w as significant ( P 2 = 5.99, df = 1, P =
0.0 14). Factors accounting f or a site difference are difficult to disentangle, but may be a result of dif ferent age
classes or locality of origin in North A merica of birds from the tw o localities, different of land use patterns
bet w een t he 2 sit es, or o f ot her f act ors . A smal l sam ple o f 7 Dick cis sels col lect ed in North A merica (Kansas
and I lli noi s) w as pr eviously analy zed. The p resen ce of DDE residues in t hose bir ds occurr ed at great er
frequ ency than i n t he Venezuel an sam ple ( P 2 = 4.9 3, df = 1, P = 0.0 26). There may be an effect of age on
the presence of the DDE residues in Dickcissels.

433      Fisher, Flemming, Roberts & Warkentin
Habitat quality, foraging behaviour and nesting of the Northern Waterthrush in western Newfoundland.
ALLISON L. FISHER, Env. Sci., Memor ial Univ. , Corner Brook, NF; STEPHEN P. FLEMM ING, Gros Morne
National Park, Rocky Harbo ur, NF; SHAWN E. ROBERTS and IAN G. WARKENTIN* , Env. Sci., M emorial Univ.
         The riparian zone is perceived by w ildlife managers to be optimal habitat f or dispersal, foraging and
repr odu ct ion by a w ide v ariet y of taxa. Cent ral t o t his prem ise is t he assumpt ion that accessible w ater and the
associat ed hab it ats are in li mited suppl y. The r elat ively moist marit ime cli mat e in w . New foun dlan d means t hat
w ater may not be a limiting resource in these island forests and that t here may be more extensive use of
upland habitats by traditionally riparian species in this region. We examined the habits of Northern
Watert hrush at six sit es across w . New fou ndland t o assess the relativ e values of ri parian and upland habit ats
in areas wit h and wit hout recent logging activities. Over 3 summers, w e identif ied on average 38 st reamside
and 26 upland territories each year across 6 sites. Average date of first egg laid was 6 Jun producing 3.2
fledgling/successful nest. Aquatic, aerial and arboreal arthropod populations were assessed at all sites wit h
results indicating lower prey availability out side of intact riparian forest stands wit h a sharp decrease in the
biomass of several arthropod t axa in previously harvested forest sites. However, w e detected no diff erences in
the foraging behaviour of w aterthrush occupying intact v ersus previously harvested habitat. A nalyses of
physical and biological features around nest and random locations suggest that concealment provided by high
densities of small diameter deciduous vegetation and steep surfaces upon which to place the nest are
important factors; larger-scale features seem less relevant.

434      Owen & M oore
The effect of migratory activity on the avian immune system. JEN C. OWEN* and FRANK R. MOORE, Dept.
Biol. Sci., Univ. Southern Mississippi, Hattiesburg, MS.
         The concept of biological trade-offs rests on t he fact that energy spent on one activit y cannot be spent
on other activities w hen energy or time is limited. If so, measures of immune funct ion should be inversely
relat ed t o in vestment in, say, f orag ing , b reedi ng ac tivit ies, or m igr atory act ivit y, if reso urc es f or o ne are t raded
against resources spent on another. Migration is a period of exceptional energetic demand and an ideal
context w ithin which to study the effect of prolonged, strenuous exercise on the avian immune system. We
tested the hypothesis that migrants t rade-off cell-mediated immune function in response to t he energetic
demand of migratory activit y. Sw ainson’s Thrushes, captured in fall, w ere held in an indoor aviary until Jan,
w hen migratory disposition, including subcutaneous fat deposition, body mass increase, and nocturnal activit y
[zu gun ruh e], w as induc ed in half the bi rds w it h art if ici ally pro lon ged d aylengt hs. Cell-mediat ed im munit y w as
measured by intradermal injection of phytohemagglutinin in the wing patagium. Birds exhibiting migratory
restlessness had a reduced immune response compared to control birds. Our results suggest t hat birds are
making a trade-off betw een the energetic demands of migration and their immune system. Therefore, a
migrant faced with a long-distance, non-stop flight over an ecological barrier is likely to become
immunosuppressed, which may have consequences for its survival and reproductive success.

435     Vitale, Cooper & Whiffen
Model ing Neot rop ical migrat ory bir d hab it at using Bay esian st atist ics and l ogi st ic r egres sion in a geog raph ical
information system. ABIGAIL C. VITALE* , ROBERT J. COOPER and HELEN J. WHIFFEN, Warnell School For.
Res., Uni v. Georgia, A thens, GA.
         The n eed f or ef fect ive reso urc e management has becom e imp erat ive. Land manager s are often
confront ed wit h disparate mandates, such as timber production and w ildlife preservation, and are required to
meet these objec tives w it h li mited f inan cial reso urc es. Wildl if e models that use satelli te imag ery and o ther
spatial data may eliminate the need for costly and time-consuming measurements by providing data that is
amenable an d app lic able t o bo th aspect s of management prac tice. Underst andi ng w hic h GIS v ariab les c orr elat e
w ith biological variables for a particular species still needs exploration; as well as, the development of methods
that allow a model to be applicable across a theme-by-theme basis. There is also a question of w hich types of
models and statistics are best able to predict the occurrence of a species across a landscape, species diversity,
or identify critical habitats. The objectives of t his proposed research are 1) investigate which GIS variables are
associated w ith dif ferent species of migratory birds and 2) use either a Bayesian statistics model or a logistic
regression model (dependent on sample size) wit hin a GIS to map potential habitat for the birds. Our methods
include bird census points that w ere conducted during the summer of 20 00 and 20 01. We extracted relevant
values of GIS variables to create either a Bayesian or Logistic regression probability model for species presence.
Probability maps generated in the models used data sets such as land history, w atershed data, Ikonis and
Landsat TM imagery, etc.

436     Iko, Kester, Bern, Stendell, Landis & Rye
Assessment of White-tailed Kite population changes related to prey: Implications from f eather stable isotope
analysis. WILLIAM M. IKO* , USGS, M idc ont inent Ecol. Sci. Cent er, Ft. Collins, CO; CYNTHIA L. KESTER,
CARLETON R. BERN, USGS Stable Isot ope Lab. , Den ver, CO; REY C. STENDELL, USGS, Midcont inent Ecol.
Sci. Cen ter, GARY P. LANDIS and ROBERT O. RYE, USGS Stable Isotope Lab.
         The White-tailed Kite has demonstrated large population fluctuations over the last 150 y r. Once
common in California, Texas, and the se. U.S., kit e numbers declined to very low levels in the 190 0s. By the
19 30 s, the species w as co nsidered on t he verge o f ext inc tion . But in t he 1 94 0s , a t rend tow ards recov ery w as
apparent. Today, t he whit e-tailed kite is a fairly common resident in California. These dramatic population
changes may be related to the conversion of native wetlands to agriculture, which may in turn support a more
div erse r oden t pop ulat ion . To in vestigat e the ef fect s of prey base on ki te pop ulat ion s ov er t ime, w e col lect ed
tissue samples from current kite populations in California to compare against tissue samples from museum
specimens collected before and after the kite decline. Since museum samples are usually limited to small
samples of feathers, our initial efforts have focused on characterizing C, N, and S isotopic variation in feathers
from different locations on the kite. Preliminary results show: 1) little variability in C (± 1.0%) and N (± 0.5%)
isotope values for wing and body feathers; 2) small differences betw een adults and juveniles wit h respect t o
w ithin w ing and wit hin bird variability; 3) a large systematic variation in adult S (± 3.0 %) isotope values among
w ing f eathers; and 4 ) signif icant v ariations i n average values among indiv iduals f rom dif ferent locati ons.

437                            rez
        Rí os-Muñoz, M orales-Pé & Navarro
The extinct birds of Mexico : What do w e know about t hem? CESAR A. RÍ OS-MUÑOZ, PATRICIA
MORALES-PÉ    REZ and ADOLFO G. NAVARRO, Mus. Zool. Fac. Ciencias, UNAM, M exico.
         Mexico holds an astonishing bird diversity of > 100 0 species, that are distribut ed in a w ide variety of
habi tats and geog raph ic r egio ns presen t in t he co unt ry . How ever , an impor tant part of that bio div ersity has
been lost recently due to of human-related factors. Those birds that have become extinct, or presumably
ext inc t, w ere in gener al restric ted-r ange t axa w hose pop ulat ion s w ere in capab le of recov er af ter ex ot ic s peci es
introduct ions and habitat modif ication. In this cont ribution, w e compiled information on t he extinct taxa of
Mexican b ird s, analy zing it s t axonom ic s tatus, caus es of ext inc tion , an d t he im pli cat ion s of this kno w ledg e to
the preservation of endangered species in similar conditions.

438      Barber, Monroe & Goodrich
Distribution and abundance of forest birds: Effects of roads, trails and openings. DAVID R. BARBER* , Haw k
Mountain Sanctuary, Kempton, PA; MARK MONROE, Eastern Kentuck y Univ. , Richmond, KY ; and LAURIE J.
GOODRICH, Hawk Mount ain Sanctuary.
          Many forested landscapes are fragmented by roads and natural or human made openings, yet how
these openings affect the distribution and abundance of breeding birds has been poorly studied. Moreover,
hik ing trails may also f ragm ent habitats, but on a sm aller scale. W e used f ixed-radius point cou nt s t o analyze
the distribution and abundance of birds w ith respect to roads (paved and dirt), hiking trails and openings on
Hawk Mountain Sanctuary, a 964 ha, second growt h forest in the central Appalachian Mountains.
Great-crested Flycat chers, A merican Robins, and Nor thern Cardinals w ere more abundant n ear roads (< 10 0 m),
and Black-and-whit e Warblers w ere more abundant aw ay from roads (> 100 m). Abundance near dirt roads
w as similar t o abu ndan ce near paved ro ads f or f lycat chers an d car din als, low er f or r obi ns, and i nt ermediat e
bet w een pav ed ro ads and aw ay f rom road s f or w arbl ers. Red-bellied Woodp eckers, Black -t hro ated Green
Warbler s, and c ardi nals w ere more ab und ant near h iki ng t rail s, and Co mmon Yello w thro ats, Rose-breast ed
Grosbeaks, and Baltimore Orioles w ere more abundant away from trails. Yellow -billed Cuckoos and Eastern
Tow hees w ere more ab und ant near o peni ngs and Hairy Woodp eckers, Eastern Wood -Pew ees, Wood T hru shes
and Ovenbirds were more abundant aw ay from openings. Fragmentation of f orested habitat by roads and
openings can negatively aff ect the abundance of some forest interior species. Moreover, the influence of trails
even on comm on species suggest s that for est f ragmentat ion may o ccur at fin er scales.

439      Lott, Sturt evant , Tow nsend & Seagle
Use of GIS in sampling h abitat variables fo r fo rest int erior passerines. KEITH D. LOTT* , BRIAN R.
STURTEVANT, PHILIP A. TOWNSEND and STEVEN W. SEAGLE, Appalachian Lab., Univ. Maryland, Frostburg,
          Geographic information systems (GIS) are oft en used to link coarse-scale habitat att ributes w ith
songbi rd d ist rib ut ion s in a spat ial en viron ment, b ut exam ini ng f ine-scal e habi tat selection pres ent s some sp ecial
challen ges. These chal leng es inclu de mism atches in reso lut ion /ac curacy of songbi rd o bser vat ion s as compared
w it h ras terized (i. e., cell -based) habit at dat a, an d un certaint y in ho w territ ory bou ndar ies reflect act ual h abit at
use. We address these issues by investigating how fine-scale territory placement relates to spatial data in the
central Appalachians. Digital spot-mapping data collected using handheld field computers were used to
construct territories of singing males using minimum convex polygons. Territories w ere compared wit h terrain
indices constructed from a 30m resolution digital elevation model and vegetation st ructure samples taken wit hin
the study sit es. Spatial analysis w as conducted on a cell by cell basis wit hin Arc/Inf o GRID. Preliminary results
suggest a relat ionship b etw een topogr aphic variables and placement of t errit ories fo r cert ain songbird species.
With f urt her t est ing , o ur m ethod olo gy should pro vide a useful t ool for c ompari ng f ine scale av ian t errit ory dat a
with other spatial databases.

440      Withey & Marzluff
What does a Christmas Bird Count say about population trends for a mobile species? JOHN C. WITHEY* and
JOHN M. MARZLUFF, Coll. Forest Res., Univ. Washington, Seattle, WA.
         Christmas Bird Count (CBC) results are used across North America to examine population t rends in a
variety of species. Winter counts, how ever, may not accurately reflect population changes for mobile species if
bir ds m ov e in o r ou t of the co unt cir cle b etw een seaso ns. In Seat tle, Wash ing ton, CBC dat a suggest an
increase in the American Crow population of 9% /year. We take 3 approaches to examine how w inter counts
relate to local population changes: 1) conducting monthly CBC-style crow surveys in the Seattle count circle, 2)
comparing age ratios of captured crows during w inter versus summer, and 3) radiotracking juvenile crow s born
at dif ferent distances from the Seat tle co unt cir cle. The c row /p art y-hr i ndex for eac h mont hly cou nt bet w een
Oct 2000 and Apr 2001 w as within 17% of the 83 crows/party-hr recorded during the Dec 2000 Seattle CBC.
The j uv enil e:ad ult rat io o f capt ures dur ing w int er w as approx imately 3: 1 an d w ill be co mpared t o summer
captures. We radiotagged 30 fledgling crows during Jul and Aug 2000. To date, 8 radiotagged crows have
died, 8 are found regularly in their natal areas, and 6 transmitters have failed. 3 have moved into the Seattle
CBC circle from > 10 km away, another 3 have dispersed > 10 km from natal areas but remained outside the
CBC circle, and 2 born in Seatt le moved outside the circle. We suggest that w hile some crow s moving only
temporarily into t he circle are likely counted during the CBC, count results should reflect population t rends in
American Crow s in the Seat tle area.

441      Hannah & Schmiegelow
Habit at q uality f or W hit e-thro ated Sparr ow s in a recently logged landsc ape: Result s not so clearcut ? KEVIN C.
HANNAH and FIONA K. A. SCHMIEGELOW, Dept. Ren. Res., Uni v. Albert a, Edmont on, A B.
         Forest harvesting can lead to reductions in the amount of interior forest habitat, incr eases in the amount
of edge, and the degradat ion and isolat ion of remaining habit at. Ul tim ately, these changes may decr ease
habi tat qual it y f or m any passerin e bir ds, w hic h may not be ref lect ed in chan ges in den sit y. We measur ed
habitat qualit y and t errit ory d ensity for breeding Whi te-t hroat ed Sparrow s in int erior f orest, at clearcut /f orest
edges, and in clearcuts, in a recently logged landscape near Calling Lake, Alberta. Density w as measured using
spot-mapping; habitat quality w as assessed using indices of body condition, behavioural time-budgets, and
through measures of reproductive success. The territory density of Whit e-throated Sparrow s w as not
significantly dif ferent betw een habitat treatments. How ever, clearcuts consistently represented poor quality
habitat for t his species w ith respect to body condit ion, pairing success, and reproductive success. Edge and
interior forest represented better quality habitat. How ever, predation of White-throated Sparrow nests did not
increase in any one habitat type follow ing harvest. Alt ernatively, results from this st udy suggest a combination
of redu ced p airi ng success, p oor bod y c ond it ion , an d po ssible b ehaviou ral d if ferenc es of ind ividu als betw een
habitats may result in reduced reproductive success in poorer quality habitats.

442      Rohila, Alberti, Coe, Hill, Marzluff & Pflugh
Selecting st udy sit es f or b ird resear ch in u rban areas: Usin g GIS t o qu ant if y urban pat terns . TINA ROHILA * ,
Coll. For. Res., Univ. Washington, Seattle WA; MARINA ALBERTI, STEFAN COE, Dept. Urban Design &
Planning, Univ. Washington; KRISTINA HILL, Dept. Landscape Architecture & Design, Univ. Washington; J OHN
MARZLUFF, Coll. For. Res., Univ. W ashington; and DOUG PFLUGH, Dept. Urban Design & Planning, Univ.
          We il lustrat e here a un ique met hod ology used f or d escribing u rban pat tern m etr ics that w e hypot hesize
are linked to ecological pr ocesses in urbanizing landsc apes in the Puget Sound r egion. W e selected metri cs t o
quan tif y urban form , u rban int ensity, and l and c ov er het erog eneit y in or der t o seek cor relat ion s w it h bi olo gic al
metrics. Urban form represents a range of clumped to dispersed housing, calculated using mean patch size of
urban land cov er. Urban int ensity is represented by dominance, based on the land co ver t ype w ith the hig hest
tot al percentage. Land cover heterogeneity is calculated using contagion, w hich quantifies the degree of mixing
across land use categories. These urban patt ern metrics have been quantified from Landsat satellite imagery
and parcel data, using ArcINFO and GRASS GIS. We selected 42 sites of 1 km x 1 km and 300 m x 300 m
using landscape classes representing degrees of dominance, mean patch size, and contagion. Researchers are
using these sites for studies of bird communit y composition and productivit y, snag densities and
characteristics, and aquatic macro-invertebrate biodiversity.
443     Kaltenecker, Carlisle, Hamilton & Bechard
The Idaho Bird Observatory. GREGORY S. KALTENECKER, Idaho Bird Observ., Boise, ID; J AY D. CARLISLE,
Dept . Biol., Un iv. Sout h Dakota, Verm illi on, SD; SARAH L. HAMILTON, Dept. Biol., Univ. Id aho, Mo scow , ID;
and MARC J. BECHARD, Dept. Biol., Boise State Univ. , Boise, ID.
         The Idaho Bird Observatory is a cooperative research and education organization that conducts
long-term studies of w estern migratory bird populations and promotes environmental education at high school,
community , undergraduate, and graduate levels. We have conducted yearly fall migration hawkw atch and
raptor banding programs since 1994 . Annual fall hawk counts range from 35 00 t o 800 0 raptors, and banding
programs have identified wintering areas in s. California and western Mexico for many species. The Idaho Bird
Observatory conduct s a fall migration songbird mistnett ing program at Lucky Peak, an Important Bird Area in
Idah o and a kno w n migr ation st opo ver sit e. So ngb ird mist net ting and m igr ation cou nt s hav e iden tif ied
important stopover habitats along the Boise Ridge, and have resulted in over 16,000 birds banded since 1997.
We conduct the only f ull-time ow l migration banding program in the w . U.S. Northern Saw-w het and
Flammulated Owls are targeted, and this project has identified timing and age struct ure of forest ow l migration
in the West for t he first t ime. Environmental education includes a program w hich targets local high school
students, a community volunteer program, undergraduate and graduate course work, and opportunities for
graduate and undergraduate research.

444      Piaskowski & Albanese
Resour ce sam pli ng o f art hro pod s in all v eget ation st rat a and c orr elat ion w it h art hro pod s ident if ied i n f ecal
samples of in sectiv orous w arblers at a spring migrat ion st opover sit e. VICTORIA D. PIASKOW SKI* and GENE
ALBA NESE, Zool. Soc. of M ilw aukee, Milw aukee, WI.
        Insectivorous w arblers (Parulidae) have been captured in high numbers during spring and fall migration
monitoring research at the Birds Without Borders-Aves Sin Fronteras Pewaukee, WI study site. To assess the
foods available to these birds during spring migration stopover and to determine the plants and/or habitats w ith
w hich arthropods w ere associated, sampling w as conducted tw ice weekly from 0 .5 m above the ground to
canopy level (20 m) using branch clippings for sessile arthropods and Tanglefoot ® coated boards for f lying
insects. Fecal samples were collected from birds at t he time of banding and the arthropod parts present w ere
correlated with those sampled at the site. Of the 22 plant species sampled, Quercus and Salix species had the
highest numbers of sessile arthropods. A bundance of f lying insects showed tw o peaks during the spring
migration sampling period and numbers detected demonstrated the importance of sampling all height strata.
Art hropod parts of 12 orders were detected in fecal samples of insectivorous w arblers. In fecal samples of
Nashville and Tennessee warblers, the predominant arthropod parts w ere those of Lepidoptera larvae. Magnolia
Warbler fecal samples contained comparable percentages of both Lepidoptera larvae and Diptera parts. The
conservation implications of these findings will be discussed.

445      Nott & DeSante
Model ing avian pr odu ct ivit y using bi rd b andi ng d ata and l andscape an alyses of the USGS Nation al Land Cov er
Dataset. M. PHILIP NOTT* and DAVID F. DeSANTE, Inst . Bir d Populat ions, Point Reyes St ati on, CA.
         We investigated the utility of the USGS National Land Cover Dataset in predicting the numbers of adults
and young for 9 bird species captured during the breeding season at six Monit oring Avian Productivity and
Survivo rsh ip (MAPS) st ation s operat ed 1 99 4 - 19 99 on J efferso n Proving Gr oun d, Ind iana. We f oun d t hat
patterns of landscape struct ure detected within a 4 km r adius area of each station w ere good predictors not
only of t he numbers of birds captured, but also more importantly, their productivit y levels. The numbers of
adults and young and also the ratio of y oung to adults increased as a function of mean w oodland/forest patch
size for Ovenbird, Acadian Flycatcher, Wood Thrush, and Kentucky Warbler. Although these species are
know n t o be m ore o r less dependen t upo n f ores t int erio rs f or s uccessful b reedi ng, this st udy pro vides
quantitative t hreshold estimates of t he mean w oodland/forest patch size required to maximize productivit y
levels. Conv ersely, for 5 spec ies norm ally associat ed w it h w ood land /f ores t edge o r su ccessional habi tat,
Northern Cardinal, Gray Catbird, White-eyed Vireo, Indigo Bunting, and Common Yellowthroat , the numbers of
adul ts cap tured inc reased as a f unc tion of the amoun t of w ood land /f ores t edge i n t he surroun din g lan dsc ape.
The numbers of young of t hese species correlated positively w ith either the amount of edge or other metrics
such as the amount of development, t he mean distance betw een w oodland/forest patches, or the amount of
successional habitat. W e compared these results w ith t hose of a similar but regionally scaled landscape analysis
for Wilson' s Warbler in nat ional f orests of the Pacific Nort hw est.

446      Weston & Caf frey
Ind ividu al co nt rib ut ion s t o nesting in c oop erat ively breed ing crow s. TIFFANY J . W ESTON* and CA ROLEE
CA FFREY, Dept. Zool. , Oklahoma State Univ ., St illw ater, OK.
          American Crows breed cooperatively. In Stillw ater, OK, group size during the nesting season of 20 01
ranged fr om 2 - 9 ind ividu als. A uxill iaries included ind ividu als of bo th sex es ranging in age fr om 1 to at least 4
yr old. Several auxilliaries are know n to have moved in from neighboring and more distant families. Not all
auxilliaries helped at the nest; breeding group size ranged from 2 - 7 individuals. Helpers in this population are
know n to cont ribute to t he building of nests, feeding the incubating f emale, and feeding nestlings and
fledglings. We documented the contributions of all members of breeding groups during the nest building,
incubation, and nestling stages of the nesting season. Feeding contributions w ill be analyzed to examine
investment decisions as a function of breeding status, sex, age, and relatedness to nestlings.

447      Jordan, Otis, Costa & Gauthreaux
Hatching failure and embryonic mortality within a Red-cockaded Woodpecker population in South Carolina.
NANCY E. JORDAN*, DAVID L. OTIS, South Carolina Coop. Fish & Wildl. Res. Unit, Clemson Univ., Clemson,
SC; RALPH COSTA, Red-cock aded Wo odpecker Field Off ice, U. S. Fish & Wil dl. Ser., Clemson, SC; and
SIDNEY A. GAUTHREAUX, Dept . Bio l. Sci. , Clemson Uni v. , Clemson, SC.
          Hat ching failu re in bir ds r epres ent s a sig nif ican t repr odu ct ive loss. Unt il r ecent ly, m ost eggs that failed
to hatch w ere thought t o be infertile even though egg contents w ere not examined. Causes of hatching failure
include inbreeding depression, breeder quality, t erritory quality, inclement weather, tox ic chemicals and
interruption of incubation. Hatching failure due to infertility or embryonic mortality can potent ially be important
for determining reproductive success in birds (Koenig 1982, Auk 99: 526-536; Ojanen 1983, Acta Univ. Oul.
A 15 4. Biol. 2 0). The Red-cockaded Woodp ecker is a coo perat ive breed er end emic to t he lo ngl eaf
pine-wiregrass ecosystem of t he se. U.S., and w as listed as an Endangered Species in 1970 . It is t he only
North Am erican woodpecker to excavate cavities in living trees. Using new video technology, daily nest visits
w ere conducted on 38 groups of Red-cockaded Woodpeckers in South Carolina to examine patt erns of hatching
failu re du rin g t he 2 00 0 - 20 01 breed ing seasons. In t he 2 00 0 b reedi ng season, 3 3 f ailed eggs w ere coll ect ed
from 2 1 groups at the tim e of banding young (7 d after hatching). Contents of eggs w ere examined and put
into 4 general categories of time w hen embryonic mort ality occurred. 23 % of collected eggs showed no
discernible signs of embryonic development and most embryonic mort ality occurred during the first 4 d of
develop ment. Daily nest visit s indic ate less t han 7 % of failed eggs w ere t hro w n ou t by adul ts; theref ore,
recording the number of failed eggs at time of banding young (7 d) could be a good estimator of hatching
failure in Red-cockaded Woodpeckers.

448      Willson-Hillman
An nual survi val rat es of a guild of obligate ant -fo llow ing birds i n sout heast Peru. SUSAN K. WILLSON
HILLMAN, Div. Biol. Sci. , Uni v. Mi ssour i, Columb ia, M O.
        Survival rat e com pari sons amo ng t rop ical resident bir ds and w it h t heir temperat e cou nt erpar ts has been
hampered by met hod olo gic al bi ases in trod uced by var ying m easur ement techn iqu es. Mark-resigh ting st udi es
have been u nder taken l ess f requ ent ly than m ist -net st udi es, but may pro vide more ac curat e loc al surv ival rat es
due to intensive observation of banded individuals over time. Here, I present survival estimates from a 4-yr
capture-resight study of a guild of obligate army-ant follow ing birds, undertaken in Manu National Park, se. Peru
during 19 97 - 20 01 . I generated sex-sp ecific surviv al rates fo r 4 ant bird species (Phlego psis nigro maculat a,
Rhegmatorhina melanosticta, Myrm eciza fort is, and Gymnopithys salvini) and 1 woodcreeper (Dendrocincla

449      Fields
* Patterns of habitat use in wintering sparrows. WILLIAM R. FIELDS, Dept. Anim. Ecol., Iowa State Univ.,
Ames, IA.
        Community structure, predation risk, optimal foraging theory, and the arrangement of landscapes have
all been studied as factors aff ecting the foraging behavior of sparrows. This study w as designed to examine
the effects of severe w eather on winter sparrow com munities and the eff ects of dif ferent habitat ty pes on
sparrow abundance. Surveys w ere conducted from Jan through Mar 20 00 in f allow, m ow ed, and corn fields
and again from Feb through M ar 2001 in fallow fields. Observations of sparrow s w ere placed into 1 of 3
categories based on their distance from the nearest field edge: short ( < 4 m), medium ( 4 - 10 m), and long (
> 10 m ) distances. Sparrows w ere found almost exclusively in fallow fields in 200 0 but w ere nearly absent
from t he study area in 200 1 as a result of severe wint er weather. In 200 1 t emperatures w ere colder and
central Iowa set a record for the number consecutive days of snow cover in a wint er. Song Sparrows and
American Tree Sp arrow s w ere f oun d in the short distance s ect ion s of the f ield more of ten t han ex pect ed based
on availability of different sections of fields. American Tree Sparrow s were more likely to be found at short
distances from field edges, w hereas Song Sparrow s w ere more l ikel y t o be f oun d at medium or l ong distances
from edges. Food availability is similar across diff erent sections of f ields, so different responses to predation
risk may be responsible f or patt erns of h abitat use w ith in f ields.

450      Martin, Peterjohn & Koneff
Access to bird population data. ELIZABETH MARTIN* , U.S. Geol. Sur., Reston, VA; BRUCE G. PETERJOHN,
USGS Patuxent Wildl. Res. Center, Laurel, MD; and MARK D. KONEFF, U.S. Fish & Wildl. Ser., Laurel, M D.
         Access to bird population data is critical for effective conservation planning and implementation.
Alt hough a tremendous volume of baseline data exists, it is oft en diffusely distribut ed and inaccessible to the
resource manager and decision maker. A m echanism that f acilitates assembly, documentation and delivery of
avian data in a user-f riendly m anner is needed in order t o int egrate bird-r elated inf ormat ion resourc es across
agencies and organizations. To address this fundamental need, the National Biological Information
Infrastruct ure (NBII), in partnership with t he U.S. Geological Survey's Patuxent Wildlife Research Center and the
U.S. Fish and Wil dlif e Service, is developi ng a w eb-based interact ive syst em that w ill f ocus on p rovid ing access
to bir d populatio n and habitat dat a used in bird management and conservation. This syst em, know n as the NBII
Bird Conservation Node, will support planning and evaluation of bird conservation activities within the context
of t he North American Bird Conservation Initiativ e (NABCI), a framew ork for collaboration among organizations
interested in bird conservation across North America. Initial development of t he NBII Bird Conservation Node
w ill focus on creating a prototype mapping application t hat w ill provide interactive access to data from t he
North Am erican Breeding Bird Survey, t he Colonial Waterbird Survey, t he Breeding Waterfow l Population and
Habitat Survey, and the Atlantic Flyway Mid-winter Waterfowl Survey. This prototype mapping application, to
be available on-line at ht tp:/ /w w w by Sep 2001, w ill lay the foundation for establishment of a
Migratory Bird Data Center at Patuxent W ildlife Research Center, and w ill provide an opportunit y for linking t o
and establi shing p art nership s w it h ot her s our ces o f bir d po pul ation and h abit at dat a availabl e over t he In ternet .

451      Arnold, Brault & Croxall
Albatross populations in peril? A long-term population trajectory f or Black-browed Albatross nesting at Bird
Island, South Georgia. JENNIFER M. ARNOLD* , SOLANGE BRAULT, Dept. Biol., Univ. M assachusetts, Boston,
MA; and JOHN P. CROXALL, British A ntarct ic Surv. , Nat. Environ. Res. Council, Cambridge, UK.
        As w ith many albatross species, the Black-brow ed Albatross population at Bird Island is currently
declining. Fluctuations in Black-browed Albatross populations may reflect fishing activities around their
breeding grounds and wintering areas, as well as relate to the availability of krill, their primary food source.
Declines in ad ult and j uv enil e sur vival rat es in the earl y 1 99 0s may be li nked t o in creased m ort alit y due t o
by-catch associated w ith t he Patagonian toot hfish fishery during the breeding season or the long-line tuna and
hake f isheries that oper ate in t he w int erin g gr oun ds. While b y-cat ch-related mort alit ies in reg ulat ed f isheries
have declined in recent years, impacts of illegal fishing are still likely to be substantial. Unfort unately, data on
the im pact of ill egal ac tivit ies is scarce. Fur ther h igh fisheries-related mort alit y c ombin ed w it h t he po tent ial
decline in krill abundance as a result of global warming poses a serious threat to the persistence of this species.
Here w e present a mixed age-/stage-classified matrix model that examines the population trajectory of
Black-browed Albatross for South Georgia given a range of potential scenarios for fisheries related mortality.
We use 30 y r of demographic data from long-term studies at Bird Island to analyze the sensitivit y of population
grow th t o changes in demographic parameters of individual life stages, and the effects of uncertainty in
demographic parameters on pr ojected po pulati on grow th and ex tin ct ion risk for thi s species.

452     Arnold, Hatch & Nisbet
Effects of egg mass, parental quality, and hatch-date on growth and survival on Common Tern chicks.
JENNIFER M. ARNOLD, JEREMY J. HATCH, Dept. Biol., Univ. Massachusetts, Boston, MA, and I. C. T.
NISBET, I. C. T. Nisbet and Co., North Falmouth, MA.
         In many bird species, chicks hatching from larger eggs have higher probabilities of survival; how ever, it
is unclear w hether this is attributable to egg size, parental quality, or timing of breeding. To examine the
eff ects of these f actor s on grow th and su rviv al of Comm on Tern chi cks, w e exchanged clut ches among nest s
(n= 56) of birds initiating laying 13 M ay to 9 J un 199 8. There was no relationship betw een egg mass and
lay-dat e dur ing this peri od. Clut ches w ere ex changed bet w een nests mat ched f or l ay-dat e so t hat dif ferenc es
bet w een masses o f seco nd-laid eggs from don or an d rec ipi ent nest s ran ged f rom -4.4 to + 6. 1 g . W e used
regr ession t echn iqu es t o examin e the ef fect s of egg m ass (f ost er egg ), qual it y of the social paren ts (maternal
egg mass, parental age, male head-length, female head-length), and hatch-date on hatching success, and
growth and survival of second hatched-chicks during 3 stages of growth (Quadratic: days 0-3; Linear: days
3-12 ; Asy mptot ic: days 17 - 2 5). Both f oster egg mass (P< 0.0 01) and hatch date (P< 0.0 5) w ere positively
correlat ed t o, and i mport ant pred ict ors of , m ass at hat ching . Hatchlin g mass w as t he best pred ict or o f sur vival
through t he quadratic period (P< 0.0 8). During the linear period, hatch-date was the only significant predict or
of gro w th (inv erse r elat ion ship; P< 0. 00 5). Ou r results suggest a tradeo ff bet w een earl y and lat e breed ing that
is modif ied by egg mass.

453      Bowman
Demography of Florida Scrub-Jays in a suburban matrix: implications f or reserve design and spatially-explicit
modeling. REED BOWMAN, Archbo ld Biol. Sta., Lake Placid, FL.
          Few er t han 4 ,0 00 breed ing pair s of Florida Sc rub -Jays occ ur t hro ugh out their rang e, a d ecli ne of over
90% from estimated historical numbers. Many of t hese jays occur in habitats degraded by fire suppression, but
these habitats can be restored via application of prescribed fire. However, over 30% of extant scrub-jays
occupy scrub patches in a suburban matrix, for w hich litt le amelioration exists. Urbanization has multiple
impacts on the demography of scrub-jays. Juvenile recruitment is reduced through increased nestling brood
redu ct ion and i ncreased post-f ledg ing mort alit y. Alt hou gh m any bir ds begin breed ing at an earl ier ag e, surv ival
of bot h no n-breedi ng an d br eedin g adu lt s is low er t han t hat of jays in uni mpact ed sc rub s. Stochast ic i ndi vidual
population models developed using these demographic data suggest suburban jay populations are demographic
sinks, in w hich mortality exceeds recruitment. Surprisingly, the strength of these eff ects w as independent of
the density of human residences w ithin suburbs; jays in low -density suburbs suffered as much as jays in
high-density suburbs. This suggests that urbanization has a negative effect on scrub-jays in preserves that are
adjac ent to subur bs. Model ing suggest s t hat these ef fect s may be st ron g eno ugh to t urn demogr aphi c so urc es
into demographic sinks, but these effects vary depending on the size of the jay population within the preserve
and t he pr opo rt ion of jays w it hin the pr eserv e w hic h suffer t he af fect s of urb anizat ion . These data suggest that
the matrix habitats surrounding scrub preserves may have a profound impact on the long-term viability of
scrub-jay populations.
454     Engstrom & Baker
Long-t erm ef fects of fi re excl usion on an avian com muni ty . R. TODD ENGSTROM, Tall Timbers Research
St at ion, Tal lahassee, FL; and W. WILSON BAKER, Tallahassee, FL.
           Fire is a critical disturbance that determines plant species composition and forest structure in upland
pin e w ood land s in the se. U.S., espec iall y t he lo ngl eaf pin e (Pinus palustris) ecosystem. A struct ural analogue
to the longleaf pine ecosystem that is dominated by loblolly (P. taeda) or short leaf (P. echinata) pines may occur
in o ld f ield s w hen f ire h as co nt inu ed t o be u sed f or l and m anagem ent . W e sur vey ed t he br eedin g-season avian
community of a 8.9 ha study plot (NB66) annually from 1967 to 198 1 (WWB and R. L. Crawf ord) and every 5
yr (RTE; 1996 and 2001) after fire was excluded from an open-structured old-field pine woodland in 1967.
During t he 35 yr af ter t he last f ire, 4 6 bir d species tot al have been detected and t he avian commu nit y has lost
1 species every 3 yr on average. Plant succession and the changes in the avian community may be typical of
widespread habitat alteration throughout the Southeast as the use of prescribed fire has declined.

452     DeSante, Nott & O' Grady
Identifying t he proximate demographic cause(s) of population change by modeling spatial variation in
pro duc tivit y, sur vivo rsh ip, and p opu lat ion trend s. DA VID F. DeSANTE* , M . PHILIP NOTT and DANIELLE R.
O' GRADY, Inst . Bir d Populat ions, Point Reyes St ati on, CA.
         A t echnique for identif ying the proximate demographic cause(s) of population change by modeling
spatial variation in vit al rates (productivit y and survivorship) as a funct ion of spatial variation in population
trends is described and evaluated at tw o spatial scales. Productivit y indices and time-constant annual adult
surv ival rat e esti mat es w ere mod eled f rom th e Mo nit oring A vian Produc ti vit y and Surv ivo rshi p (M APS)
Pro gr am. For the larg er scale, Gray Cat bi rd vit al rates du ri ng 1992 - 1 998 w ere mod eled fro m d at a f ro m M APS
stations in areas comprised of physiographic strata where 1992 - 199 8 Breeding Bird Survey (BBS) population
trends w ere either significantly positiv e or negative. Adult survival rate estimates w ere area dependent w hile
pro duc tivit y indi ces w ere in depen dent of area. The d if ferenc e bet w een areas in p opu lat ion chan ges m odel ed
from t he vital rates agreed well wit h the diff erence in BBS population t rends, suggesting that low survival of
adults w as the proximate demographic cause of catbird decline in strata w here they were declining. At the
smal ler scale, vital rat es durin g 1 99 4 - 19 99 for Car oli na Chi ckadee, Gray Catbir d, Oven bir d, Yellow -br east ed
Chat, and Field Sparrow w ere modeled from data from MA PS stations on military installations in the western
and eastern Midw est as a funct ion of spatial variation in 1994 - 199 9 t rends in adult captures. We w ere able
to identif y the proximate demographic cause(s) of population change for each species. Moreover, t he
regr ession o f trend in ad ult capt ures on m odel ed po pul ation chan ge f or t he 5 spec ies in t w o areas w as
significantly positiv e. We conclude that the approach is indeed useful and that an optimal approach would
include mo deling t emporal as w ell as spatial variat ion in v ital rat es and populati on t rends.

456     Rosenberg & Barker
Conservation strategy f or a non-endangered species, the Cerulean Warbler. KENNETH V. ROSENBERG* and
SA RA E. BA RKER, Cornell Lab. Ornithol. , It haca, NY.
        The Cerulean Warbler has declined sharply in most of its range, suffering perhaps a 75% decline overall
since 19 66 . A n eff ectiv e conservat ion st rategy to st abilize populati ons and reverse declines is essenti al must
begin wit h identification of present-day breeding concentrations, designation of crit ical habitats, and
assessment of thr eats to know n sit es. These w ere the goals of the Cerulean Warbler A tlas Project (CEWA P),
w hich recorded nearly 10, 000 Cerulean Warblers at over 2, 000 sites in 24 stat es, Ontario, and Quebec from
199 7 - 19 99. Largest contiguous concentrations are in West Virginia and the Cumberland Mountains of
Tenn essee, w it h ot her s ign if ican t pop ulat ion s (2 00 + pair s) at w idel y scat tered s it es t hro ugh out the ran ge.
Cerulean Warblers showed great variation in habitat and tree-species use, wit h most birds occupying
oak-hickory forests on dry ridgetops, sycamore- and cottonw ood-dominated bottomlands, or cove forests in the
s. Appalachians. Habitat specialization and area sensitivity may be reduced in the Northeast, w here some
populations are expanding. Because important concentrations exist on federal, state, and private lands, a
cooperative partnership among government, indust ry, and academic scientists is needed to focus on research,
management, and land protection issues. Recovering the Cerulean Warbler without the regulatory pow er of the
Endangered Species Act represents an important t est case for t he non-adversarial approach to bird conservation
promoted by Partners in Flight and the North American Bird Conservation Initiative.

457      Foster, Tw eed, W oodworth, Camp & Adler
Avian populatio n trends in th e Alakai Swamp, Kauai: Native declines and exotic expansions? JEFFREY T.
FOSTER* , Prog. Ecol. Evol. Biol., Univ. Illinois, Urbana, IL; ERIK J. TWEED, BETHANY L. WOODWORTH,
RICHARD J. CAMP and COREY D. ADLER, USGS-BRD Kilauea Field Stn. , Haw aii National Park, HI.
          The Hawaiian avifauna has declined severely since human colonization, with losses exceeding 80 native
bir d species. Kauai' s A lakai Sw amp, t he on ly remaini ng n ative fores t bir d hab it at on t he island , h as not been
immun e to t he problems plagui ng Haw aii' s birds, w here 5 nati ve bird speci es have gone extin ct w ith in t he last
30 y r. We conducted avian surveys of the entire Alakai Swamp in Mar and Apr 20 00 and compared them to a
similar survey from 197 3 t o determine long-term population changes of native and introduced species. Our
w ork rev ealed unchanged or incr easing bird popu lation s for 5 of the 7 most commo n nativ e species. At least 4
int rod uced species inc reased in n umber, 1 w as extirp ated, and 1 new spec ies inv aded. Addi tion ally , w hen
compared to surveys conducted in the interior of the sw amp in 1981 , 19 89, and 1994 , only 2 native species in
the 200 0 survey show ed significant population declines. Thus, most nativ e bird populations in the interior and
periphery of t he Alakai Swamp remain healthy despite numerous limiting f actors. Nonetheless, the avifauna of
Kauai remains extremely vulnerable and active management of introduced plants, animals, and disease is
essential in prev enting fur ther habit at degradati on and nati ve bird lo ss.

458      Chen & Peterson
Prioritization of China for conservation using modeled geographic distributions of birds. GUOJUN CHEN* and
A. TOWNSEND PETERSON, Nat. Hist. Mus., Univ. Kansas, Lawrence, KS.
         We developed distributional models for t he 90 endangered bird species in China, and used heuristic
complemen tarit y algor it hms t o pr ior it ize areas for c onserv ation . The pi xel-based op timizat ion pri ori tized 2 0 ar eas
for prot ection, w hich covered all species analyzed. Optimizing inclusion of endangered species based on the
exist ing bio sphere reserv e syst em in clu ded o nly 37 - 62 speci es in 11 reser ves, l eaving 2 8 - 53 spec ies
unprotected. We used the distributional hypotheses to identify optimal additions to the reserve system, which
w ould improve its effectiveness substantially. The pixel-based optimization includes species much more
efficiently on a per-area basis, and thus offers exciting perspectives to improve protection of the country s
endan gered avifauna.

459      Calvert & Robertson
Using multiple population size estimators to infer population t rends of At lantic Puffins. ANNA M . CALVERT,
Dept. Zoo l., Univ . Guelph, Guelph, ON; and GREGORY J. ROBERTSON* , Can. Wi ldl. Serv. Mount Pearl, NF.
         Burrow nesting auks are diff icult t o census and obtaining information on population trends can be
particularly challenging. In the summer of 20 00, w e estimated the number of At lantic Puffins using a specific
study slope on Gull Island, Newfoundland, w ith 5 dif ferent techniques. To estimate the number of breeding
birds on the slope we grubbed burrows; f rom this method w e estimated that 5 16 adult puf fins bred on the
study slope. Visual counts of puf fins att ending the slope consistently underestimated the number of birds
breeding on the slope, and likely has limited utilit y. Since 199 7, 5 35 puf fins have been individually
colou r-b anded on t his slope. A clo sed popu lat ion est imator w it h cap ture h eterog eneit y estimated 3 77 marked
puf fins used the slope i n 2 00 0. Based on rec apt ures of puf fins w it h f oot noo ses and assuming 37 7 m arked
puff ins, a corrected Lincoln-Peterson index estimated 17 44 (125 6 - 22 32 ; 95 % CI) puff ins used the slope. Two
different mark-resight population size estimators were also calculated. An estimate of 2983 (2657-3399; 95%
CI) puf fins w as obtained based on b anded to un band ed rat ios, w hil e 35 25 (349 9 - 36 49 ; 9 5% CI) puf fins w as
calculated with t he Bow den estimator. W e projected an age-based projection matrix using literature values for
vital rates, and extracted the ratio of non-breeding to breeding birds at stable age distribution. To obtain the
large numbers of non-breeding birds seen in our study the population is expected to be 1) increasing, 2) have
large numbers of breeding age puff ins t hat do n ot b reed and 3) att racti ng you ng birds f rom ot her coloni es.
Coupled w ith evidence of breeding slope degradation, these results suggest a growing puff in population that is
pot ent iall y r eachi ng t he li mits of the availab le br eedin g hab it at.

460     Morgan & Smith
Seabird bycatch mortalities in gill net fisheries on Canada's w est coast. K. H. MORGAN, Canadian Wildl. Ser.,
Sidney , BC. and J. L. SMITH,* Vancou ver, BC.
           Seabirds are accidentally caught during commercial fishing and, depending on gear and location, many
individuals, especially albatrosses, fulmars and shearwaters, cormorants, gulls and auks are killed. Responding
to t he UN Food and Agriculture Organization' s National Plan of Act ion initiative, some governments and fishing
indust ries are beginning t o assess the bycat ch level in lo ngline f isheries and to adop t reduc tio n measures.
How ever, w ith f ew exc eptions (Washingt on State) lit tle eff ort h as focussed on t he impact of net f isheries. In
Canada, there are tw o net fisheries wit h onboard observer programs, but litt le is known about t heir seabird
bycatch rates. To begin to address that know ledge gap, Environment Canada and Fisheries and Oceans Canada
ini tiat ed a jo int pro gram to assess and moni tor t he seabird bycat ch level in Canada' s w est coas t com mercial
gillnet and longline fisheries. This paper presents a preliminary analysis of the incidental take of seabirds in
gillnet fisheries.

461     Raphael, Bentivoglio, Baldwin, M ax, Miller, Ralph, Strong & Thompson
An estimate of offshore population size of the Marbled Murrelet in Washington, Oregon, and northern
California. MARTIN G. RAPHAEL*, Pacific Northw est [PNW] Research Station, Olympia, WA; NAOMI
BENTIV OGLIO, U.S. Fish & Wildl. Ser., Hon olulu, HI; JAMES BALDWIN, Pacific Southw est [PSW] Research
Stati on, Al bany , CA ; TIM MAX, PNW Research St atio n, Port land, OR; SHERRI MILLER, C. JOHN RALPH, PSW
Research St ati on, Ar cat a, CA , CRAIG STRONG, Crescent Coastal Research, Ast oria, OR; and CHRIS
THOMPSON, Washington Dept. Fish & Wildl., Olympia, WA.
         Estimating population trend of the threatened Marbled Murrelet is critical to recovery planning. We
designed a monitoring protocol t o estimate population size and trend and have completed the first year of
samp lin g un der t his design. Our design u ses a st rat if ied r ando m sam ple o f trans ect s dist rib ut ed acr oss coas tal
w aters betw een the Canadian border of Washingt on and San Francisco Bay, Califor nia. We use
distance-sampling methods to estimate murrelet density in each of 5 recovery zones during the breeding season
from mid-May to mid-Jul. Our baseline estimate for year 2001 is 17,400 birds (95% C.I = 12,300 to 22,500
birds). Our current estimates of population size in each state are similar to previously published values in
Washington and California; our estimate for Oregon is similar or lower depending on source of data compared.
Pow er anal ysis of these data and ex ist ing est imates of annu al variat ion ind icat es it w oul d t ake 9 yr to det ect an
annual decrease of 5% w ith " = 0.1 0 and $ = 0.90.

462     Smith, Brooks, Jahn, M ehlman & Roca
Conservati on prior iti es for bi rds at r isk in Latin Am erica. KIM BERLY G. SMITH* , Dept. Biol. Sci., Univ.
Arkansas, Fayetteville, AR; THOMA S BROOKS, Conserv ation Int ernat .; ALEX JAHN, Univ. Arkansas; DA VID
MEHLMAN, Wings of Amer. Prog., The Nature Conservancy; and ROBERTO ROCA, Conserv ation Int ernat .
          The distribution of 1 ,28 9 species of permanent resident species of birds were examined simultaneously
using W ORLDMA P sof tw are t o pr odu ce a map of con serv ation pri ori ties f or Lat in A merica. That map w as
compared t o a map of hum an po pul ation s t o det ermine t he rel ation ship bet w een bi rds at ris k and hum an
density. The idea of complimentary conservation areas using a trunctated GAP approach is demonstarted for
the country of Paraguay. A CD containing all bird distribut ions and WORLDMAP softw are is available free from
the f irst author (kgsmit
463 Nakazaw a, Garcia-Trejo, Honey, Lira-Noriega, Pasquetti, Sanchez-Gonzalez, Navarro
& Peterson
Seasonal changes in ecological niche of Neotropical migrants: A predictive modelling approach. YOSHINORI J.
Mexico; and A. TOWNSEND PETERSON, Nat. Hist . M us. , Un iv. Kansas, Law rence, KS.
          A great amount of resear ch is devoted t o id ent if y t he cau ses of the dec lin ing num bers of Neot rop ical
migrants. Many st udies have concluded that habitat loss in the wint ering grounds in Latin America is the main
cause for su ch declines, and not habitat modif icatio n, f ragmentat ion, o r urbanizati on on t he breeding groun ds.
Much of the discussion lies in the understanding of seasonal variation in species' ecological niches: breeding,
w int erin g and migrat ion . W e com pared ecol ogi cal n ich es of a set of Neot rop ical migran t spec ies of several
taxonom ic g rou ps t hat breed w it hin North A merica an d w int er mainl y sout h t o n. Midd le A merica. Model led
distributions w ere obtained using the Genetic Algort ithms f or Rule-set Prediction (GARP) for each season, and
those predictions were translated into combinations of ecological conditions. Based on these models, we w ere
able to document patt erns of specialization versus generalization in diff erent seasons. The relevance of this
kind of approach in migrant bird conservation is discussed.

464     Osterhaug, Richter & Houck
Five-year bird usage o f a w etland rest oration and creat ion si te in an urban set ting. KA RIN OSTERHAUG* ,
KLAUS RICHTER and CATHERINE HOUCK, King County Dept. Nat. Res., Seattle, WA.
         Wetlands p rovid e resting, feeding, and breeding habit at f or a w ide diversit y of birds. Despite t his,
thousands of acres of wetlands are filled, drained, or otherw ise destroyed annually in the U.S. As a means of
combating wetland destruction, agencies and organizations are required to mitigate for projects that harm
w etlands by restoring or enhancing existing w etlands. King County Department of Natural Resources in
Washington State is completing the monitoring phase of a 5-yr experimental wetland mitigation banking project
on t he Issaquah pl atueau, w hic h is one o f the most rapi dly develop ing areas in t he reg ion . For decad es t he 4
ha site had been used for agricultural and grazing. The upland was dominated by pasture grasses and existing
w et areas w ere composed pr imaril y of Spir ea dou glassi. Among o ther g oals, p roj ect land scap ing w as design ed
to improve bird habitat by introducing structural complexity through planting of native wetland emergents, a
variety of shrubs and deciduous and coniferous trees. Thirty-fiv e bird species were detected during baseline
surveys in spring 1996 . 53 species w ere detected in year 2000 ; this approaches the 61 species historically
observed at a nearby reference wetland. Four common non-native species have remained present t hroughout
the 5 yr of monitoring. W e hypothesize that native bird diversity w ill continue to increase because of increasing
fol iage height co mplexit y att ribut able to mat uring shr ubs and t rees. As t he w etland and it s buf fer mat ures,
some species found in historical open habitats (e.g., Savannah Sparrow, Killdeer) may disappear from the site.
Changes in surrounding land-use may also have an effect on species presence and must be taken into
consideration when predicting long-term bird use and comparisons of findings against reference sites.

465     Bismanis & Kessler
Forest bird communit y responses to partial-cutting in British Columbia's boreal forest. ANDRA BISMA NIS* and
WINIFRED KESSLER, Forest ry Prog ., Uni v. Nor thern Brit ish Columbia, Prin ce Geo rge, BC.
         Current interest in partial cutting reflects the " management hypot hesis" that bird com munities should
cope bett er with f orest harvest approaches that more closely resemble eff ects of natural disturbance. We are
testing t his idea by documenting bird communit y responses follow ing partial cutting in boreal forest stands of
the Fort Nelso n Forest Dist ric t, n . Brit ish Colu mbia. As w ell, w e are evaluat ing the abi lit y of spec ies/ habi tat
models developed in Alberta to predict avian occurrence in the boreal forests of British Columbia. Tw o bird
survey protocols, point -counts and fixed-w idth t ransects, w ere employed in 4 partial-cut and 2 old-growt h
(control) stands. Each stand w as surveyed 4 times/season over 2 breeding seasons. We will attempt to
calibrate the transect data to the point-count data. If successful, w e will supplement the partial-cut data wit h a
larg e dat a set col lect ed 1 99 2 - 19 95 by the t rans ect pro tocol i n t he Daw son Creek Forest Dist ric t for t he model
evaluat ion phase of the st udy . Pred ict ed abu ndan ce of selected bi rd s peci es w ill be co mpared t o ob serv ed
abundance as found for the Fort Nelson and Daw son Creek data sets. A tot al of 51 and 52 species w ere
observed across all stands in 1999 and 2000, respectively. Mean species richness in treated stands was 32.7
and 30.0, respectively, in 1999 and 2000 compared to 29 and 32.5 in control stands. Some species (e.g.,
Magnolia Warbler, Tennessee Warbler, Winter Wren and White-throated Sparrow) are represented in all stands
to varying degrees. Others (Cape May Warbler and the Canada Warbler) appear restrict ed to either the
partial-cut stands or the control stands, respectively. Remaining phases of the project include model evaluation
and development of r ecommendat ions f or ecologi cal management in b oreal for ests.

466     Robert, Savard & Benoit
The Barrow ' s Goldeneye in eastern Nort h Am erica: Dist ribut ion, p opulat ion size and thr eats. M ICHEL
ROBERT* , JEAN-PIERRE L. SAVARD, and RÉ           JEAN BENOIT, Canadian Wi ldlife Serv ice, Saint e-Foy, PQ.
         In 2 00 1, the east ern Nort h A merican pop ulat ion of the Barr ow ' s Goldeney e w as design ated as Sp ecial
Concern by the Committ ee on the Status of Endangered Wildlife in Canada (COSEWIC). We review our st ate of
know ledge on this small population, estimated to num ber 4,50 0 individuals or 1,4 00 breeding pairs (adult
females comprise about 30% of the population). The first documented breeding records for eastern North
America w ere ob tained in 1 99 8, in Quebec . Si nce then, using satelli te telemet ry , w e have iden tif ied sever al
breedi ng, molti ng, and w int ering sit es. The cor e breedin g area is locat ed on t he hig h plat eaux nort h of th e St.
Law renc e Estuary and Gu lf . M ales m igr ate nor th t o molt in Hudson, Ungava, and Frobi sher Bays, and i n a f ew
coastal inlets of n. Labrador. Females probably molt close to the breeding areas. Nearly all of t he Barrow' s
Gold eneyes in east ern Nort h A merica w int er alo ng t he St . Law renc e cor rid or, most ly in Quebec and, to a lesser
extent, in the Maritimes. A large part of the population congregates in a few areas of the St. Lawrence corridor,
increasing its vulnerability. A lthough w e do not have precise data to document a trend, w e believe that the
population has probably declined during the 20th Century. Current identif ied threats include oil spills, logging,
fish introduct ions in breeding lakes, hunting, and cont amination by toxic substances. Alt hough a great deal of
effort has been made during the past few years to document the status and ecology of the eastern population
of the Barrow ' s Goldeneye, much remains to be learned. Evidently, opt imal allocation of monitoring, research
and management efforts w ill be necessary to ensure this population' s conservation.

467     Johnson
* Conservation genetics of the Greater Prairie Chicken. JEFFERY A. JOHNSON, Dept. Biol. Sci., Univ.
Wisconsin-M ilw aukee, Milw aukee, WI.
         The Greater Prairie Chicken (Tympanuchus cupido pinnatus) is threat ened w it h extinc tion in sever al
midw estern states. Population sizes have decreased due to habitat loss and fragmentation. The reduction of a
large widespread population to several small isolated populations may lead to a reduction of heterozygosity,
loss of allel es due to ran dom genet ic d rif t, an d a red uction in g ene f low . Genetic v ariat ion w as examin ed
throughout much of t he range of greater prairie chickens using 6 microsatellite DNA loci. Significant population
struct ure was identified across the species range and w ithin 4 subpopulations in Wisconsin. Prairie chickens in
Wisco nsin hav e few er all eles/ loc us and lo w er het erozyg osity t han p opu lat ion s in Nebraska, Minn esot a, Kansas
and Misouri. These results provide baseline information on genetic variation that w ill be useful for both fut ure
monitoring of populations and decisions about translocation of individuals to restore genetic diversity.

468     Bechtoldt & Stouffer
Henslow' s Sparrows w intering in Louisiana savannas: Habitat quality declines with time after burning.
CATHERINE L. BECHTOLDT* and PHILIP C. STOUFFER, Dept. Biol. Sci., Southeastern Louisiana Univ.,
Hamm ond , LA .
          Henslow' s Sparrow is a habitat specialist that w inters in the longleaf pine savanna ecosystem of the
southeastern gulf coastal plain. Periodic fire is the most important factor in maintaining the characteristic
st ruc ture o f these savannas. The t iming of these perio dic bur ns m ay be imp ort ant in m aint aini ng h abit at
characteristics favorable to Henslow' s Sparrow s. We censused, captured, and radiotagged Henslow ' s
Sparrows w intering on six savannas differing in time since burn. All burns took place during the grow ing
season in 1998, 1999, or 2000. In standardized observations we recorded all species within 6.25 ha plots.
We also attempted to capt ure all Ammodramus in each plot . Teams of 4 - 8 people w alked unif ormly across
the pl ot s, flushin g sparrow s into mist net s. We radio tagged 12 Hensl ow ' s Spar row s w int erin g in savannas
burned in 1999 and 2000 . Mov ements were recorded 1-2 t imes daily for about 3 w k. Vegetation analysis
inc lud ed measur ements of densit y, perc ent cov er, and seed availab ili ty. Abun danc e of overall gras sland species
and Ammodramus sparrows w as highest on the most recently burned sites. Henslow' s Sparrow density
declined w ith increasing time since burn, declining from 24 .4/ 10-ha in savannas burned in 2000 to 9 .3/ 10-ha in
savannas burned in 1998 . Radiotagged Henslow' s Sparrow s maintained regular territories. Vegetation density
and diversity varied with t ime since burn. The most recently burned savannas were the least dense, but held
the most gras s species. Seed availabi lit y decli ned o ver time after bu rni ng. Prelim inar y data ind icat e the need
for f requent burns to m aintain high densities of w intering Henslow ' s Sparrow .

469     Bertelli
A phylogenetic study of tinamous (Aves: Palaeognathae) based on morphology. SARA BERTELLI, Univesidad
Nacional de Tu cuman, A rgen tina.
          Tinamous are terrestrial birds endemic to t he neotropical region. Interest in this group derives from the
fact t hat tinamous are basal among extant birds and very litt le is known about t he phylogenetic relationships
w ithin t he family. In this study , a cladistic analysis w as conducted including 31 species of tinamous in a matrix
of 90 ost eolo gic al and 15 my olo gic al ch aracters. The p rog rams EST and NONA pro duc ed 1 5 d ist inc t opt imal
trees of 325 steps. The phylogenetic hypot eses here presented do not recover the Tinaminae (forest-dwelling
tin amous) as originally proposed by oth er author s (Tinamus + Nothocercus + Crypturellus). Tinaminae is
instead paraphyletic, and its ecological niche is plesiomorphic. Among the forest-dwelling tinamous, Tinamus
and Nothocercus are monophyletic groups and the monophyly of Crypturellus is not supported. The genus
Nothocercus has a basal position w ithin Tinamidae. The present analysis, in contrast, supports t he monophyly
of Nothurinae (open areas tinamous, Taoniscus + Nothura + Nothoprocta + Eudromia + Tinamotis +
Rhynchotus). Among the open areas tinamous, Taoniscus is basal to Nothurinae, and Tinamotis and Eudromia
appear to be more closed related to Rhynchotus than to Nothura and Nothoprocta. Except for Nothura, the
monophyly of the genera in the group of open area tinamids is recovered. These hypotheses agree w ith t he
resu lt s of ot her m orp hol ogi cal s tudi es based on int egum ent ary char act ers. Compari son among b ot h anal yses
show s signi fican t con gru ence.

470     Giannini & Bertelli
Netw ork relationships among extant penguins. NORBERTO GIANNINI and SARA BERTELLI, Univer sidad
Nacional de Tu cuman, A rgen tina.
          A preliminary phylogeny of all the 17 currently recognized, extant species of penguins was estimated on
the basis of 62 int egum ent ary char act ers, inc lud ing bil l shape, part s, and c olo r; plu mage pat tern o f the head
and body; podoteca; iris color; plumage patt ern of the first and second plumage of t he chick; and plumage
pattern of t he juvenile. The analysis w as developed as a netw ork w ithout an outgroup, given the diff iculty of
comparison o f this typ e of char act ers w it h t he cl osest alli es of peng uin s. A pars imony analy sis w it h im pli ed
w eights of characters was done using Pee-Wee program by Goloboff (19 93 distribut ed by author). Netw ork
resolution is complete and very well supported; a similar analysis not using weights yielded a similar topology.
If t he root is placed on Eudyptula as suggested by fossil records, all genera but Pygoscelis are monophyletic.
Megadyptes is the sister taxa to all Eudyptes. Spheniscus mendiculus (t he Galapagos peng uin ) is the basal
form of the gen us. The r elat ion s w it hin gener a are bet ter su ppo rt ed t han t he rel ation s amo ng g enera.

Shared By: