Larger clutch size increases ~edging success and

Larger clutch size increases ~edging success and 56\ 109Ð105 offspring quality in a precocial species Journal of Animal Ecology 0887\ DENIS LEPAGE\ GILLES GAUTHIER and ANDRE DESROCHERS$ Departement de Biologie and Centre d E tudes Nordiques\ Pavillon Vachon\ Universite Laval "Quebec# G0K 6P3\ Canada^ and $Departement des Sciences du Bois et de la Fore¼t\ Pavillon Abitibi!Price\ Universite Laval "Quebec# G0K 6P3\ Canada Summary 0[ We tested the hypothesis that the ability of parents to raise viable o}spring limits clutch size in the greater snow goose "Anser caerulescens atlanticus L[#\ a precocial bird[ 1[ We manipulated clutch size by exchanging complete clutches between pairs of nests to increase or decrease the clutch size by zero "control#\ one\ two or three eggs in 203 nests over 1 years[ 2[ Pre!~edging survival of goslings increased in enlarged broods and decreased in reduced broods compared to control[ Consequently\ enlarged broods ~edged more o}spring and the reverse was true for reduced broods[ 3[ Size and mass of goslings near ~edging was also higher in enlarged broods than in control\ which suggests that o}spring quality was also enhanced by the manipulation[ This is contrary to the common trade!o} between o}spring numbers and quality[ 4[ Large families were dominant over smaller ones in feeding sites\ which could explain the increased survival and growth of enlarged broods[ 5[ Our results suggest that the ability to raise young does not limit clutch size in this species and that parents could be more successful "i[e[ increase both the number and quality of their o}spring# by laying more eggs[ However\ the time required to lay additional eggs reduces the viability of all o}spring and may explain why females do not lay more eggs[ Key!words] clutch size\ snow goose\ growth rate\ ~edging success\ parental ability Journal of Animal Ecology "0887# 56\ 109Ð105 Ecological Society 109 Introduction The hypothesis that clutch size is optimal with respect to the ability of the parents to recruit young is central to the study of life history strategies[ In altricial birds\ clutch size is generally thought to be limited by the ability of parents to raise young "Lack s optimal clutch size hypothesis# or by the cost incurred by the parents when raising too many young "the cost of repro! duction hypothesis# "Partridge + Harvey 0874^ Pet! tifor et al[ 0877^ VanderWerf 0881^ Jacobsen et al[ 0884#[ To test these hypotheses\ one must manipulate clutch:brood size\ which has seldom been done with precocial birds "Safriel 0864^ Rohwer 0874^ Lessels 0875^ Sandercock 0883#[ Only Safriel "0864# found a negative e}ect of enlarged broods on production of This is Polar Continental Shelf Project contribution No[ 90086[ young in a precocial species\ while others found no e}ect[ In precocial birds\ the hypothesis that the ability of parents to raise young limits clutch size has often been dismissed simply on the ground that these birds do not feed their young "Winklers + Walters 0872#[ However\ precocial birds provide other forms of parental invest! ment shared among o}spring "e[g[ brooding\ Lazarus + Inglis 0875^ Schindler + Lamprecht 0876#[ In geese and shorebirds\ there is some evidence that chick mor! tality increases with brood size "Safriel 0864^ Cooke et al[ 0884#\ which suggests that parental ability to raise young may set an upper limit on clutch size "Winkler + Walters 0872#[ Few studies have examined this hypothesis in precocial birds and they have found mixed results "Safriel 0864^ Winkler + Walters 0872^ Lessels 0875^ Williams et al[ 0883#[ We experimentally tested the hypothesis that the ability of parents to raise viable o}spring may limit clutch size in precocial birds by manipulating clutch

100 size and examining the e}ects on o}spring growth and et al[ 0883b#[ All goslings captured were weighed and D[ Lepage\ survival[ Using such experiments\ one can answer the measured "culmen\ head\ tarsus and ninth primary G[ Gauthier + question] would parents be more successful if they had length#[ A[ Desrochers laid additional eggs< The experiments were conducted During brood!rearing\ we observed social inter! in the greater snow goose Anser caerulescens atlanticus actions between families during a total of 14 h[ Every L[\ a large precocial and colonial bird nesting in the time a family displaced another family we noted the High Arctic[ Methods number of goslings in the winning and losing "dis! placed# families[ Interactions sometimes lasted for more than a minute\ but only the _nal outcome was recorded[ Observations could include both manipu! lated and unmanipulated broods[ STUDY AREA Our study area was located at the Bylot Island colony "62>N 79>W\ North!west Territories\ Canada#[ Much DATA ANALYSES of the island is occupied by mountains and glaciers\ Pre!~edging survival of goslings was estimated by the except for a plain of ¼ 0599 km 1 in its southern part[ recapture of marked individuals[ Probability of recap! Rolling hills\ covered by upland tundra and numerous ture is the product of survival rate and capture rate isolated wetland patches used by brood!rearing geese\ "i[e[ probability of capturing a bird if alive#[ Di}er! dominate the landscape of this plain[ It is the most ences in probability of recapture re~ect di}erences in important breeding site of greater snow geese "16 999 survival\ assuming that capture rate is constant among nesting pairs were censused in 0882^ Reed et al[ 0881^ experimental groups[ Goose families were captured Gauthier et al[ 0885# which occur in several nesting over a wide area at our study site and we have no colonies on the island[ evidence of di}erential dispersal among our exper! imental groups[ We tested the e}ect of Year\ Hatch date\ ICS\ DM\ and DM 1 on the proportion of gos! FIELD METHODS lings recaptured per brood near ~edging with log! In 0882 and 0883\ nests were found and marked during linear models "Proc Catmod of SAS Institute Inc[ laying and incubation[ Their fate was determined by 0877#[ The analysis was restricted to manipulated visiting them regularly until hatching "Lepage et al[ nests "experimental and control#\ and only signi_cant 0885#[ Nesting success was de_ned as the proportion variables were retained in the _nal model[ We used of nests where at least one young left the nest and was DM instead of CCS in this analysis because small calculated with the May_eld method "May_eld 0864^ sample size prevented us of using CCS\ which had too Johnson 0868#[ Nesting success between di}erent many classes[ Even then\ we could not test more than groups was compared using the test suggested by two independent variables simultaneously[ We thus Johnson "0868#\ which allows two by two compari! tested all possible two!variable combinations before sons[ Hatching success was de_ned as the proportion removing a variable from the model[ of eggs that hatched in successful nests "i[e[ those Body measurements of goslings caught were used where at least one egg hatched#[ as indices of growth[ Body size was de_ned as the Shortly before hatching\ in both years\ we ex! _rst component of a principal component analysis changed complete clutches between pairs of nests to combining measurements of tarsus\ head and culmen[ change the clutch size by one\ two or three eggs[ Paired Age of unmarked goslings was estimated by the linear nests had a similar hatch date\ to control for the e}ect relationship between age of marked goslings "from of hatch date on growth rate "Cooch et al[ 0880^ both manipulated and unmanipulated nests# and Lindholm et al[ 0883#[ We also exchanged complete length of their ninth primary feather "0882] n 091\ clutches between pairs of control nests with identical R 1 9[52\ P ³ 9[990^ 0883] n 51\ R 1 9[61\ clutch sizes[ The change in clutch size "CCS# isthe P³9[990#[ Age of goslings when measured ranged di}erence between the number of eggs received from from 17 to 34 days "mean 26[0\ n 2788 young#[ the donor nest and the actual number of eggs laid An initial generalized linear model was _rst used to "from Ð 2 to 2#\ direction of the manipulation "DM#is test the e}ects of Year\ Age and Hatch date on gos! the sign of CCS "reduced clutches] CCS from Ð 2 to Ð lings size and mass using all goslings captured[ To 0\ control] CCS 9 and increased clutches] CCS from control for the e}ects of these variables\ residuals from 0to2#\andtheinitial clutch size "ICS# is the number this initial model were used to test the e}ects of year\ of eggs in the nest before the manipulation[ ICS and manipulation "CCS and CCS 1 # on marked All goslings that hatched from experimental and goslings from manipulated nests only[ control nests\ as well as goslings from a sample of unmanipulated nests\ were individually marked at Ecological Society hatch with numbered web!tags "Alliston 0864#[ Shortly Results Journal of Animal before ~edging\ family groups were captured in mass The mean clutch size was 3[9 eggs "range] 1Ð5 eggs\ Ecology\ 56\ 109Ð105 banding drives in nearby brood!rearing areas "Hughes n 427 nests#[ Clutch size was increased or decreased

101 in 040 nests and left unchanged in 052 control nests[ Since parents with enlarged broods left the nest with Larger clutch Nesting success of reduced nests "58[4) 2 4[6 SE\ more young\ and these young survived signi_cantly increases ~edging n 80# was smaller than nesting success of control better\ it follows that the number of ~edged o}spring success nests "71[2) 2 2[3 SE\ n 052# "P 9[94#[ This in enlarged broods was much higher than in control e}ect could possibly be an artefact of the manipu! "Table 0#[ For instance\ in parents laying a clutch of lation "see discussion#[ However\ nesting success of three eggs "the group with the largest sample size#\ enlarged nests "62[9) 2 5[3 SE\ n 59# did not di}er enlarging the clutch by 0[8 eggs more than doubled from control nests "P 9[08# or from reduced nests the number of o}spring at ~edging compared to "P 9[4#[ Among successful nests\ hatching success control[ For the same reasons\ the number of ~edged after the manipulation did not di}er between reduced o}spring in reduced broods was much lower than in "78[2) 2 1[3 SE\ n 57#\ enlarged "70[2) 2 2[2 SE\ control broods[ It should be reminded\ however\ that n 39# and control nests "78[7) 2 0[5 SE\ n 024^ average number of ~edged o}spring per brood in x 1 4[5\ P 9[951\ d[f[ 1# despite a trend for Table 0 are underestimated for all categories because slightly lower hatchability in enlarged clutches[ Only it is not all broods with live young which are recap! successful manipulated nests "i[e[ those that hatched# tured at ~edging[ were used in subsequent analyses[ Year\ Age\ Hatch date\ and most of their inter! The mean number of goslings leaving the nest 13 h actions had a signi_cant e}ect on the size and mass of after hatching was 2[4 "n 280 nests#[ On average\ all gosling captured near ~edging "Table 1#[ In the manipulations increased the number of goslings leav! second analysis\ which controlled for the e}ects of ing the nest in successful nests by 0[3 in enlarged the previous variables\ Year\ ICS and CCS 1 had no clutches "n 39 nests# and decreased it by 0[5 in signi_cant e}ect on the residual size and mass of gos! reduced clutches "n 57 nests#[ A total of 760 young lings from manipulated broods "P 9[14#\ and these were tagged in 132 manipulated nests\ and 385 were variables were not retained in the _nal model[ tagged in 037 unmanipulated nests[ Near ~edging\ However\ residual size and mass of goslings from 2788 goslings were captured\ including 50 marked manipulated broods were positively related to CCS ones from manipulated broods and 31 from unman! "Table 1#\ indicating that goslings were larger and ipulated broods[ heavier when associated with an enlarged brood The proportion of goslings recaptured per brood "Fig[ 1#[ was positively related to the experimental change in We observed 19 dyadic encounters involving broods clutch size "DM\ x 1 6[95\ P 9[997\ d[f[ 0#\ indi! of unequal size[ Larger families won 07 of these cating that survival of o}spring increased in enlarged encounters "sign test] P ³ 9[990# and were therefore broods and decreased in reduced broods\ compared to controls "Fig[ 0#[ None of the other variables tested "Year\ Hatch date\ ICS and DM 1 # a}ected the pro! portion of goslings recaptured "P 9[2#[ There was dominant over smaller families[ Discussion no di}erence in the average proportion of goslings The higher survival and better growth of goslings in recaptured between manipulated controls "5[4)\ enlarged broods compared to control or reduced n 024 broods# and unmanipulated broods "4[2)\ broods are contrary to the predictions of the hypoth! n 037^ x 1 9[17\ P 9[5\ d[f[ 0#[ esis that parental ability limits clutch size[ Although we have no data on post~edging survival\ and hence recruitment\ body size of goslings at ~edging is posi! tively related with subsequent survival "Owen + Black 0878^ Francis et al[ 0881^ Cooke et al[ 0884#\ recruit! ment "Sedinger et al[ 0884# and perhaps even fecundity "Sedinger et al[ 0884#[ Thus\ by laying more eggs\ females ~edged more o}spring\ which in turn should have better chances of recruitment because of their larger body size at ~edging[ This suggests that the number of eggs laid by geese is not optimal with respect to their ability to raise young and that parents would be more successful if they laid more eggs[ Several studies in altricial birds have also found that enlarged broods ~edge more young in absolute terms "Pettifor et al[ 0877# but this was often obtained at the expense of a lower individual nestling survival or quality "e[g[ lower body mass at ~edging# "see review in Fig[ 0[ Average proportion of goslings recaptured per brood Ecological Society Lessells 0875#[ To our knowledge\ this is the _rst study "2 SE# for the three experimental groups "reduced\ control Journal of Animal and enlarged broods# "ž#\ for unmanipulated broods "T# that experimentally reported an increase in the quality Ecology\ 56\ 109Ð105 and values predicted by the log!linear model "dotted line#[ and survival of individual o}spring in larger broods

102 Table 0[ Number of goslings recaptured per family in control and experimentally reduced or enlarged broods "clutch size D[ Lepage\ decreased:increased on average by 0[8 eggs^ range] 0Ð2# according to initial clutch size "i[e[ before manipulation# G[ Gauthier + A[ Desrochers Initial Reduced Control Enlarged clutch size Mean SE n Mean SE n Mean SE n 0 9[9 09 9[49 9[23 5 1 9[9 1 9[9 6 9[49 9[23 01 2 9[9 7 9[12 9[03 15 9[35 9[22 02 3 9[94 9[11 19 9[19 9[98 45 9[56 9[56 8 4 9[93 9[93 17 9[54 9[12 20 5 9[9 09 9[9 4 Table 1[ Summary of the general linear models[ Model 0] e}ect of Year\ Age and Hatch date on body size "PC0 score^ R 1 9[55\ n 2746# and mass "R 1 9[46\ n 2740# of all goslings captured near ~edging[ Model 1] e}ect of manipulation "CCS# on the residuals of size "R 1 9[97\ n 59# and mass "R 1 9[08\ n 50# obtained from model 0 for goslings from manipulated nests Body size Body mass d[f[ F P F P Model 0 Full model 6 0957[09 ³ 9[990 613[70 ³ 9[990 Year 0 222[49 ³ 9[990 260[47 ³ 9[990 Age 0 0570[53 ³ 9[990 0016[67 ³ 9[990 Year Age 0 297[36 ³ 9[990 255[34 ³ 9[990 Hatch date 0 13[69 ³ 9[990 28[75 ³ 9[990 Hatch date Year 0 8[92 9[992 0[35 9[12 Age Hatch date 0 05[97 ³ 9[990 13[32 ³ 9[990 Age Hatch date Year 0 12[54 ³ 9[990 9[65 9[27 Model 1 "residuals# Full model 0 4[19 9[915 02[43 ³ 9[990 CCS 0 4[19 9[915 02[43 ³ 9[990 independent of parental quality[ This suggests that the Our protocol did not control speci_cally for the common trade!o} reported in altricial birds between heritable component of growth rate and body size or number of o}spring and their quality does not apply for maternal e}ects[ For instance\ if larger parents lay in precocial birds "see also Rohwer 0874 and Lessells larger clutches\ they may transmit this trait to their 0875# and could even be in the opposite direction in o}spring\ which would then grow faster and attain geese "i[e[ laying more eggs may increase both the larger _nal body size[ Although Sedinger et al[ "0884# survival and quality of individual o}spring#[ found a positive relationship between female body The reduction in the number of eggs in the nest size and clutch size in brant geese Branta bernicla L[\ resulted in a lower nesting success[ It is possible that Cooke et al[ "0884# found no such relationship in lesser a reduction of the number of eggs is perceived by the snow geese A[ c[ caerulescens L[ using a larger data parents as an act of predation\ which may reduce set[ Heritability of morphological characters in birds nest attentiveness and create favourable conditions for "i[e[ size of the various parts of the body# has been further predation[ However\ this would be a transient estimated at about 59Ð69) "Boag + van Noordwijk e}ect that is unlikely to persist into the brood!rearing 0876#[ However\ the heritability component is a period[ Moreover\ the manipulation of eggs in itself di.cult measure to obtain in the _eld and most heri! did not appear to have a}ected nesting success[ Ano! tability studies may have underestimated the e}ect of ther limitation of our study is that it was conducted the common environment between parents and o}! during only 1 years[ Although environmental con! spring[ For example\ if large parents have a better ditions during these 1 years tended to be similar access to local food supply\ this would enhance the "except for nest predation which was higher in 0883 growth of their o}spring "Boag + van Noordwijk Ecological Society than 0882# and near average "Lepage et al[ 0885#\ 0876^ Larsson + Forslund 0880#[ This bias is especially Journal of Animal di}erent results could be obtained in extreme years important in geese because of their strong female! Ecology\ 56\ 109Ð105 "e[g[ very late snow!melt#[ biased philopatry[ A recent experiment by Sedinger

103 The egg!production limitation hypothesis states that Larger clutch the number of eggs laid by a female is limited by the increases ~edging amount of reserves accumulated before laying success "Rohwer 0881#[ In lesser snow geese\ females rely almost exclusively on reserves accumulated prior arrival on the breeding grounds for the production of eggs "Ankney + MacInnes 0867^ but see Ganter + Cooke 0885#[ However\ in greater snow geese\ clutch size is unlikely to be limited by nutrient reserves alone[ As in most other precocial birds\ this species acquires large amount of nutrients by feeding during prelaying on the nesting ground "Gauthier + Tardif 0880^ Cho! iniere + Gauthier 0884#\ and hence\ by retarding lay! ing and feeding for a longer period\ they should theor! etically be able to lay more eggs "Drent + Daan 0879#[ According to the cost of reproduction hypothesis\ a survival or fecundity cost could be associated with larger broods "Jacobsen et al[ 0884^ Daan et al[ 0885#[ Families persist up to 0 year in geese\ and the presence of young could interfere with acquisition of body reserves by parents in spring through competition for feeding sites "Turcotte + Bedard 0878#[ However\ no evidence of such cost was found in lesser snow geese "Williams et al[ 0883#\ nor in experimental manipu! lation of reproductive costs in other species of geese Fig[ 1[ The e}ect of experimental change in clutch size "CCS# "Lessels 0875^ Tombre + Erikstad 0885#[ Physio! on the residual size and mass of goslings near ~edging\ after logical costs of producing the eggs may also ultimately controlling for the e}ects of year\ gosling age\ hatch date limit the number of eggs by decreasing future repro! and their interactions[ ductive output[ In an experimental manipulation of reproductive costs\ Heaney + Monaghan "0884# showed that increased e}ort in egg production resulted in a lower ability of parents to raise additional chicks in the common tern "Sterna hirundo L[#\ sug! gesting that there is a cost to laying the eggs[ Similarly\ et al[ "0886# also failed to detect any genetic or there could be a cost to incubating a clutch too large maternal e}ects on seasonal variations in gosling in terms of reduced hatchability or higher predation growth in black brant[ risk[ This could account for the weak trend toward A possible explanation for the enhanced growth lower hatchability found in enlarged clutches[ of goslings in larger broods is that they are socially However\ Rohwer "0874# failed to _nd such costs in dominant over smaller ones and lone pairs "Prop et blue!winged teals "Anas discors L[# incubating al[ 0873^ Black + Owen 0878^ Gregoire + Ankney enlarged clutches[ 0889^ Cooch et al[ 0880#[ Our behavioural obser! Actual clutch size could still be optimal at the indi! vations showed that larger families were indeed domi! vidual level\ according to the laying date and feeding nant over smaller ones during the summer[ Dominant conditions encountered before laying\ as stated by the families may have better access to sites o}ering high individual optimization hypothesis "Pettifor et al[ 0877#[ food quality and the best protection from predators In arctic!nesting geese\ like many other birds\ there is "Prop et al[ 0873^ Hughes et al[ 0883a#[ The number a steep seasonal decline of o}spring viability "Cooke of goslings may either increase the motivation of the et al[ 0873^ Cooke et al[ 0884#[ The number of eggs parents to provide parental care or be used by other laid may thus represent a trade!o} between the pro! individuals as a signal of parental quality[ The dis! duction of more o}spring and the delay imposed by tinction between these two phenomena is important the production of more eggs "Drent + Daan 0879^ only if there is a cost to parents in raising more o}! Choiniere + Gauthier 0884# because the laying inter! spring "Schindler + Lamprecht 0876#[ val between each egg is about 22 h in snow geese If the ability to raise young posthatch does not limit "Schubert + Cooke 0882#[ In great tits "Parus major clutch size and\ on the contrary\ laying more eggs L[#\ although clutches enlarged by one egg were more increases both the number and quality of o}spring\ productive\ this was compensated by the costly delay Ecological Society then other factors must constrain clutch size in greater in laying the additional egg "Pettifor et al[ 0877#[ This Journal of Animal snow geese[ We see three hypotheses that could hypothesis could also explain the limitation of clutch Ecology\ 56\ 109Ð105 explain why geese do not lay more eggs[ size in precocial birds[

104 Acknowledgements Gregoire\ P[E[ + Ankney\ C[D[ "0889# Agonistic behavior and dominance relationship among lesser snow geese dur! D[ Lepage\ We thank E[G[ Cooch\ F[ Cooke\ K[E[ Erikstad\ ing winter and spring migration[ Auk\ 096\ 449Ð459[ G[ Gauthier + M[J[J[E[ Loonen\ N[ Plante\ J[ McNeil\ R[ Pettifor\ J[ Heaney\ V[ + Monaghan\ P[ "0884# A within!clutch trade! A[ Desrochers o} between egg production and rearing in birds[ Pro! Tinbergen and S[ Verhulst for helpful comments on ceedings of the Royal Society of London B\ 150\ 250Ð254[ the manuscript and the many people who helped with Hughes\ J[R[\ Gauthier\ G[ + Reed\ A[ "0883a# Summer _eldwork\ especially J[P[ Tremblay[ Funding was pro! habitat use and behaviour of greater snow geese Anser vided by a Natural Sciences and Engineering Research caerulescens[ Wildfowl\ 34\ 38Ð53[ Council of Canada "NSERC# grant to G[ Gauthier\ Hughes\ J[R[\ Reed\ A[ + Gauthier\ G[ "0883b# Space and habitat use by greater snow goose broods on Bylot Island\ Ducks Unlimited Canada\ the Arctic Goose Joint Northwest Territories[ Journal of Wildlife Management\ Venture "Environment Canada#\ the Fonds pour la 47\ 425Ð434[ Formation de Chercheurs et l Aide a la Recherche Jacobsen\ K[O[\ Erikstad\ K[E[ + S%ther\ B[E[ "0884# An "FCAR\ Ministere de l E ducation du Quebec#\ the experimental study of the costs of reproduction in the Canadian Wildlife Service and the Department of kittiwake Rissa tridactyla[ Ecology\ 65\ 0525Ð0531[ Johnson\ D[H[ "0868# Estimating nest success] the May_eld Indian and Northern A}airs Canada[ The NSERC\ method and an alternative[ Auk\ 85\ 540Ð550[ FCAR and Centre d E tudes Nordiques provided _n! Larsson\ K[ + Forslund\ P[ "0880# Environmentally induced ancial assistance to D[ Lepage[ Logistic support was morphological variation in the barnacle goose\ Branta leu! generously provided by the Polar Continental Shelf copsis[ Journal of Evolutionary Biology\ 3\ 508Ð525[ Project "Natural Resources Canada#[ We are grateful Lazarus\ J[ + Inglis\ I[R[ "0875# Shared and unshared par! ental investment\ parentðo}spring con~ict and brood size[ to the Pond Inlet Hunters and Trappers Association Animal Behaviour\ 23\ 0680Ð0793[ for allowing us to work on Bylot Island[ Lepage\ D[\ Gauthier\ G[ + Reed\ A[ "0885# Site in_delity in Greater Snow Goose] a consequence of constraints on laying date< Canadian Journal of Zoology\ 63\ 0755Ð0764[ References Lessells\ C[M[ "0875# Brood size in Canada geese] a manipu! lation experiment[ Journal of Animal Ecology\ 44\ 558Ð Alliston\ W[G[ "0864# Web!tagging ducklings in pipped eggs[ 578[ Journal of Wildlife Management\ 28\ 514Ð517[ Lindholm\ A[\ Gauthier\ G[ + Desrochers\ A[ "0883# E}ects Ankney\ C[D[ + MacInnes\ C[D[ "0867# Nutrient reserves of hatch date and food supply on gosling growth in arctic! and reproductive performance of female lesser snow geese[ nesting greater snow geese[ Condor\ 85\ 787Ð897[ Auk\ 84\ 348Ð360[ May_eld\ H[ "0864# Suggestions for calculating nest success[ Black\ J[M[ + Owen\ M[ "0878# Agonistic behaviour in bar! Wilson Bulletin\ 76\ 345Ð355[ nacle goose ~ocks] assessment\ investment and repro! Owen\ M[ + Black\ J[M[ "0878# Factors a}ecting the survival ductive success[ Animal Behaviour\ 26\ 088Ð198[ of barnacle geese on migration from the breeding grounds[ Boag\ P[J[ + van Noordwijk\ A[J[ "0876# Quantitative Journal of Animal Ecology\ 47\ 592Ð507[ genetics[ Avian Genetics "eds F[ Cooke + P[A[ Buckley#\ Partridge\ L[ + Harvey\ P[H[ "0874# Costs of reproduction[ pp[ 34Ð67[ Academic Press\ London[ Nature\ 205\ 19[ Choiniere\ L[ + Gauthier\ G[ "0884# Energetics of repro! Pettifor\ R[A[\ Perrins\ C[M[ + McCleery\ R[H[ "0877# Indi! duction in female and male greater snow geese[ Oecologia\ vidual optimization of clutch size in great tit[ Nature\ 225\ 092\ 268Ð278[ 059Ð051[ Cooch\ E[G[\ Lank\ D[B[\ Dzubin\ A[\ Rockwell\ R[F[ + Prop\ J[\ van Eerden\ M[R[ + Drent\ R[H[ "0873# Repro! Cooke\ F[ "0880# Body size variation in lesser snow geese] ductive success of the barnacle goose Branta leucopsis in environmental plasticity in gosling growth rates[ Ecology\ relation to food exploitation on the breeding grounds\ 61\ 492Ð401[ western Spitsbergen[ Norsk Polarinstitutt Skrifter\ 070\ 76Ð Cooke\ F[\ Findlay\ C[S[ + Rockwell\ R[F[ "0873# Recruit! 006[ ment and the timing of reproduction in lesser snow geese Reed\ A[\ Boyd\ H[\ Chagnon\ P[ + Hawkings\ J[ "0881# The "Chen caerulescens caerulescens#[ Auk\ 090\ 340Ð347[ numbers and distribution of greater snow geese on Bylot Cooke\ F[\ Rockwell\ R[F[ + Lank\ D[B[ "0884# The Snow Island and near Jungersen Bay\ Ba.n Island\ in 0877 and Geese of La Perouse Bay] Natural Selection in the Wild[ 0872[ Arctic\ 34\ 004Ð008[ Oxford University Press\ Oxford[ Rohwer\ F[C[ "0874# The adaptive signi_cance of clutch size Daan\ S[\ Deerenberg\ C[ + Dijkstra\ C[ "0885# Increased in prairie ducks[ Auk\ 091\ 243Ð250[ daily work precipitates natural death in the kestrel[ Journal Rohwer\ F[C[ "0881# The evolution of reproductive patterns of Animal Ecology\ 54\ 428Ð433[ in waterfowl[ Ecology and Management of Breeding Water! Drent\ R[H[ + Daan\ S[ "0879# The prudent parent] energetic fowl "eds B[D[J[ Batt\ A[D[ Afton\ M[G[ Anderson\ C[D[ adjustements in avian breeding[ Ardea\ 57\ 114Ð141[ Ankney\ D[H[ Johnson\ J[A[ Kadlec + G[L[ Krapu#\ pp[ Francis\ C[M[\ Richards\ M[H[\ Cooke\ F[ + Rockwell\ R[F[ 375Ð428[ University of Minnesota Press[ Minneapolis[ "0881# Long!term changes in survival rates of lesser snow Safriel\ U[N[ "0864# On the signi_cance of clutch size in geese[ Ecology\ 62\ 0235Ð0251[ nidifugous birds[ Ecology\ 45\ 692Ð697[ Ganter\ B[ + Cooke\ F[ "0885# Pre!incubation feeding activi! Sandercock\ B[K[ "0883# The e}ect of manipulated brood ties and energy budget of snow geese] can food on the size on parental defence in a precocial bird\ the willow breeding ground in~uence fecundity[ Oecologia\ 095\ 042Ð ptarmigan[ Journal of Avian Biology\ 14\ 170Ð175[ 054[ SAS Intitute Inc[ "0877# SAS version 5 edn[ SAS Institute\ Gauthier\ G[\ Rochefort\ L[ + Reed\ A[ "0885# The exploi! Cary\ North Carolina\ USA[ tation of wetland ecosystems by herbivores on Bylot Schindler\ M[ + Lamprecht\ J[ "0876# Increase of parental Island[ Geoscience Canada\ 12\ 142Ð148[ e}ort with brood size in a nidifugous bird[ Auk\ 093\ 577Ð Ecological Society Gauthier\ G[ + Tardif\ J[ "0880# Female feeding and male 582[ Journal of Animal vigilance during nesting in greater snow geese[ Condor\ 82\ Schubert\ C[A[ + Cooke\ F[ "0882# Egg!laying intervals in Ecology\ 56\ 109Ð105 690Ð600[ the lesser snow goose[ Wilson Bulletin\ 094\ 303Ð315[

105 Sedinger\ J[S[\ Flint\ P[L[ + Lindberg\ M[S[ "0884# Environ! VanderWerf\ E[ "0881# Lack clutch size hypothesis*an mental in~uence on life!history traits] growth\ survival and Larger clutch examination of the evidence using meta!analysis[ Ecology\ fecundity in black brant "Branta bernicla#[ Ecology\ 65\ 62\ 0588Ð0694[ increases ~edging 1393Ð1303[ Williams\ T[D[\ Loonen\ M[J[J[E[\ Cooke\ F[ "0883# Fitness success Sedinger\ J[S[\ Lindberg\ M[S[\ Eichholz\ M[ + Chelgren\ N[ consequences of parental behavior in relation to o}spring "0886# In~uence of hatch date versus maternal and genetic number in a precocial species] the lesser snow goose[ Auk e}ects on growth of black brant goslings[ Auk\ 003\ 018Ð 000\ 452Ð461[ 021[ Winkler\ D[W[ + Walters\ J[R[ "0872# The determination of Tombre\ I[M[ + Erikstad\ K[E[ "0885# An experimental study clutch size in precocial birds[ Current Ornithology\ 0\ 22Ð of incubation e}ort in high!arctic barnacle geese[ Journal 57[ of Animal Ecology\ 54\ 214Ð220[ Turcotte\ Y[ + Bedard\ J[ "0878# Shared parental investment\ parent!o}spring con~ict and brood size in greater snow geese[ Animal Behaviour\ 27\ 692Ð695[ Received 08 June 0885^ revision received 01 April 0886 Ecological Society Journal of Animal Ecology\ 56\ 109Ð105