Wildfwl (1973) 24:144-149 Prximate and ultimate determinants f clutch size in Anatidae PAUL A. JOHNSGARD In a stim ulating analysis, Lack (1967, 1968a) reviewed w aterfw l clutch size and egg size d ata, and cncluded th at interspecific variatins in average clutch size are generally inversely related t thse f relative egg size. Thus, he suggested th at the average clutch size f each w aterfwl species has evlved in relatin t th e average availability f fd t th e fem ale arund th e tim e f nesting. H e hypthesized th at in w aterfw l, relatively large eggs have prbably evlved t prvide the new ly-hatched yung w ith a large fd supply (an idea he subsequently (1968b) questined) r w ith an adequate insulating layer f fat. H e als suggested that, since annual, seasnal, and lcal variatins in clutch size exist, prxim ate factrs, such as the fd supply f individual fem ales, may be m re likely t influence a b ird s date f laying th an either its egg size r clutch size. Since the publicatin f L ack s study, several additinal reviews f avian clutch sizes have appeared, ntably thse f K lm p (1970), vn H aartm an (1971) and Cdy (1971). K lm p cnsidered the pr b lem s psed by th e A natidae in sm e detail, and generally agreed th at th e fd supplies available t the fem ale are prbably the ultim ate factr influencing clutch size in this grup. Because f th e large am unt f infrm a tin available fr w aterfw l, and since Lack did nt cncern him self w ith prxim ate influences n clutch sizes f individuals, I have given these m atters sm e attentin and have reached sm ew hat different cnclusins. N sim ple single explanatin fr inter- and intraspecific variatins in clutch sizes seem s pssible at present, but a sum m ary f th e evidence relating t these phenm ena nevertheless appears w rthwhile. P rxim ate factrs influencing clutch sizes f individual birds It is well established th at intrappulatin variatins in clutch size f w aterfw l d exist, which prbably reflect prxim ate envirnm ental influences th at vary w ith tim e. K lm p (1970) listed thirteen species f A natidae th a t have been reprted t exhibit declines in average clutch size during a single breeding seasn, and Bengtsn (1972) reprted this trend in all ften species f ducks in his study. E xcept fr K skim ies (1957) suggestin th at such individual clutch size variatins m ight be th e result f genetic plym rphism having adaptive value, it has been generally believed th at these seasnal differences are m ainly th e result f prxim ate influences. T he lw ered hrm nal levels after the reprductive peak (H ilden, 1964), r the reduced stim ulating by tem perature r phtperid effects (D ane, 1966), m ight reduce the clutch size f late-nesting birds. T heir sm aller clutch sizes m ight als simply reflect renesting, since repeat clutches are generally sm aller than the riginal nes (Swls, 1959). This, in turn, may result frm exhaustin f the fem ale s reserves (W agner, 1960). A gain, the sm aller, later clutches m ay be prduced by yunger females nesting fr the first tim e and thus nt in peak reprductive cnditin (Lem ieux, 1959). K ssack (1950) prvided sm e data fr th e C anada G se Branta canadensis favuring this view. V ariatins in average clutch f a single ppulatin in different years als ccur in th e A natidae, and clearly m ust be cntrlled by prxim ate factrs. The pssibility th a t these m ight be related t annual variatins in fd availability has recently been supprted by Bengtsn (1971) wh bserved th at during a year f relative fd scarcity, fur species f diving ducks and th e E urpean W igen Anas penelpe prduced significantly sm aller clutches th a n nrm al. This finding clearly runs cunter t L ack s suggestin th at annual variatins in fd supply influence laying date rath er than clutch size, but des supprt his view th at the average clutch size m ay be basically attuned t nrm al fd availability. A nnual variatins in clutch size and nnbreeding, related t w eather cnditins during the egg-laying perid, have als been established fr the A natidae, p articularly am ng arctic breeders. Evidence n this p int has been prvided fr the A t lantic B rent G se Branta bernicla hrta, th e W hite-frnted G se Anser albifrns, Black B rant Branta bernicla rientalis 144
Clutch size f Anatidae 145 and Lesser Snw G se Anser caerulescens caerulescens by B arry (1962), fr th e Lesser Snw and Blue G eese by C ch (1961), and fr the R ss G se Anser rssii by R yder (1970b). If fem ales breeding fr the first tim e tend t have sm aller clutches th an experienced breeders, th en variatins in th e percentage f yung fem ales in a ppulatin might result in betw een year variatins in average clutch size. M endall (1958) suggested this pssibility fr th e R ing-necked D uck Aythya cllaris, w hen he fund a lim ited crreletin betw een p r hatching success in ne year and large average clutches f R ing-necked D uck. H ilden (1964) prvided supprting evidence am ng T ufted D ucks Aythya fuligula and G reater Scaups Aythya marila. Ppulatin density is, theretically, an th er factr cntrlling clutch size. H ilden (1964) fund n evidence fr this am ng the tw Aythya species ju st m entined. The nly exam ple I have encuntered is th at f M arshall (1967), wh reprted th at increased clutch size (but reduced nesting success) was assciated w ith increased nesting densities in the E urpean Eider Sm ateria m. mllissima. R yder (1970b) fund th a t R ss G eese nested earlier in high-density cncentratins, b u t th at clutch size was nt influenced by th e breeding density. Interppulatin variatins in clutch size m ight, f curse, be a reflectin f either prxim ate r ultim ate factrs. If the ppulatins are gegraphically well islated, th e prbability f genetic cntrl fr these differences w uld seem t be higher, and indeed Lack (1967) prvided sm e exam ples am ng varius races f certain ducks. H w ever, cntinuus clines in clutch size m ight well be geared t such prxim ate factrs as p h tperid r tem peratu re gradients. W eller (1964) was unable t find any evidence fr such gegraphic variatin in the clutch size f the C anada G se. T he suggestin f Paynter (1951) th at th e A m erican E ider S.m.dresseri has intraspecific clinal variatin was based n data frm varius investigatrs and, as W eller indicated, is therefre f questinable value as well as being nly ver a range f less th an ten degrees f latitude. A m re general cm parisn f average clutch size f this species (T able 1) is nt indicative f increased clutch size w ith increasing latitude; if anything, th e average clutch size is largest at interm ediate latitudes. T sum m arize, it w uld appear that prxim ate factrs are as likely t influence anatid clutch size as t accunt fr variatins in laying dates, at least in tem perate r arctic species. Ultimate factrs influencing average clutch sizes f anatid species L ack (1967) advanced th e view th at the average clutch size f each anatid species has evlved in relatin t average available fd supplies fr th e fem ale, as influenced by th e size f the species eggs. H e and, earlier, H einrth (1922) have als pinted ut th a t larger-bdied birds channel relatively less fd m aterial int th e prductin f individual eggs, resulting in a negative crrelatin betw een average adult fem ale w eight and the w eight f th e species eggs. An even clearer negative crrelatin Table 1. Breeding latitude and reprted clutch sizes f the Cmmn Eider S. mllisima Lati Average tude Clutch- Ttal ( N) size nests Lcality Authrity 79 2-95 2,993 Kngsfjrd, Spitzbergen Ahlen & Anderssn 1970 65 2-74 42 Iceland Gudmundssn, 1932 64 3-44 1,598 Cape Drset, NW T Cch, 1965 63 46* 193 Valassaaret, Finland Hilden, 1964 62 4-89 89 Yukn-Kuskkwim Delta Lensink (persnal cmmunicatin) 61 4-32 734 Green Is., Que. Lemieux (cited in Cch, 1965) 60 3-13 73 Payne, Bay, Que. Edwards (cited in Cch, 1965) 56 4-47 120 Belcher Is., NWT. Freeman, 1970 51 3-6 60 Amchitka Is., Alaska Kenyn, 1961 48 4-04 1,131 Saquénay C., Que. Lewis, 1939 45 3-53 134 Kent Is., Me. Grss, 1938 44 3-25 44 Penbsct Bay, Me. Paynter, 1951 *Excludes clutches f less than three eggs.
146 Paul A. Jhnsgard m ay be seen- betw een th e average adult fem ale w eight and th e prprtinal w eight f the average ttal clutch. This rati prvides a useful index t the relative energy drain n th e fem ale during laying. L ack s (1968a) tabular data indicate th a t in eight taxa (species r subspecies) f swans, the equivalent f frm 16 t 34% (average 24%) f th eir adult w eight is depsited in an average clutch f eggs. F r sixteen tax a f true geese th e calculated range is frm 20 t 40% (average 28%), fr frty-seven surfacefeeding ducks frm 35 t 106% (average 60%), and fr sixteen sea ducks frm 20 t 109% (average 63%). The larger species, geese and sw ans, w hich have th e greatest available energy reserves f bdy fat, thus actually experience th e sm allest relative energy drain frm egg-laying. T heir generally sm all clutch size is therefre unlikely t be attrib u tab le t lim ited fd supplies during laying. Several hyptheses have been prpsed t explain this prblem. R yder (1970a) has suggested th at, at least in th e R ss G se, th e t tal pre-breeding fd reserves needed by th e fem ale fr bth egg-laying and during incubatin m ight lim it the average clutch size f this unusually sm all gse, w hich usually lays nly fur eggs. T hus, fr this and perhaps ther arctic nesting geese, the fem ale s energy supplies m ust allw the fem ale t prvide m axim um prtectin t the clutch during incubatin and ensure the survival f the yung until they can frage. This thery is essentially an extensin f L ack s basic views n clutch size cntrls. A secnd hypthesis is th a t egg predatin during the egg-laying perid may limit effective clutch size. T hus the dangers f leaving th e nest expsed pri r t incubatin may utw eigh the advantages f adding additinal eggs. The prbability f this ccurring increases w ith increasing clutch size, especially in species vulnerable t preincubatin pred ati n f the entire clutch. Lack has rejected predatin as an ultim ate cntrl f avian clutch size, n the grunds th at natural p redatin levels are nrm ally t lw t be effective and in general this wuld seem t be true. Bengtsn (1972) prvided d ata fr ten Icelandic duck species, shw ing th at pre-incubatin nest predatin caused 55% f th e nest failures in 2,889 nests, r a ttal f abut 19% f the nests under study. This w uld suggest a daily predatin rate f little m re th an 2% per day during a 10-day egg-laying perid, t lw effectively t lim it clutch size. H w ever, C hate s (1967) study f A m erican Eiders indicated th a t 66% f th e entire nest predatin in 1 year (ttalling 58% f 448 nests) ccurred n incm pleted nests, im plying a 12-15% daily nest lss during the 3- r 4-day perid required t cm plete a clutch. If they are at all typical, such predatin rates m ight easily accunt fr the lw average clutch size f this species. Thirdly, it is pssible th at average clutch size m ay be lim ited by decreasing parental effectiveness. M endall (1958) reprted decreased hatching success am ng large clutches f R ing-necked D uck. H ilden (1964) had sim ilar results fr T ufted D ucks and G reater Scaups, and Bengtsn (1972) fr G reater Scaups. R yder (1970a) has cited sm e additinal exam ples. Likewise, larger than nrm al brds m ay suffer relatively higher m rtality rates than nrm al sized nes, as indicated by E ygenraam (1957) fr M allard Anas platyrhynchs. C ch (1961) nted th at Lesser Snw G se gslings hatched frm eggs laid late in large clutches survived less well. Sim ilarly, Parsns (1970) reprted th a t H erring G ulls Larus argentatus hatched frm sm aller eggs, usually the third-laid nes, exhibited th e prest survival. If this trend is general, it might well be a surce f increasing selective disadvantage f r enlarging the clutch size. F urthly, a restricted ptim um breeding perid, r the need t synchrnize hatching w ith an ptim um hatching date, may lim it average clutch size. K skim ies (1957) fund laying date and clutch size t be individually cnstant am ng fem ale V elvet Scters M elanitta fusca, and believed this ensured unifrm hatching tim es f different-sized clutches. C ch (1961) sim ilarly pinted ut that, in the Lesser Snw G se, sm aller clutches are necessary in retarded breeding seasns r tw ards th e end f nrm al nes if reprductin is t be efficient. H ilden (1964) m entined th at, since th e perid f bth egg-laying and hatching are undesirably prlnged in large clutches, and since latehatched brds generally survive prly, selectin m ay lim it clutch size belw the fem ale s physilgical lim its r her incubatin abilities. D ane (1966) sim ilarly pinted ut th at each egg added t a d u ck s clutch delays hatching an th er day, bth increasing th e prblem s f egg and chick survival and shrtening th e available tim e fr m aturatin p ri r t m igratin. T here is cnsiderable evidence frm a variety f w aterfw l species th a t brds hatching relatively late usually survive m re prly than thse hatched fairly early in the seasn (G rice & R gers, 1965; Bengtsn, 1972). This influence f a restricted ptim um
Clutch size f Anatidae 147 breeding perid is m st likely t be exerted in arctic r subarctic areas. If ne cm pares th e average clutch size f a num ber f nrthern hem isphere w aterfw l as given by Lack w ith average July istherm s ( F) representing the nrthern lim it f these species breeding ranges prim arily as reprted by Vus (1960), an interesting relatinship m ay b e seen (Figure 1). Species having sm all average clutches nearly all are arctic r subarctic breeding, w hile thse w ith large average clutches tend t be m re tem perate r subtrpical. If ne calculates clutch w eight as a percentage f adult fem ale w eight fr such species and again plts th e results against July istherm s at the nrthern limits f distributin, a greater taxnm ic spread m ay be seen (Figure 2). A lthugh cllectively th ere is a strng tendency t reduce th e relative am unt f energy put int th e clutch the fu rth er nrth they breed, the trend is clearer in sm e grups than in thers. This m ay be due t restricted breeding perids, pssibly t increased egg-predatin dangers, t ther dem ands n energy reserves assciated w ith breeding in arctic envirnm ents, r t ther factrs. In general, thse anatid species having th e sm allest average clutch size (swans, geese, eiders) are nt likely t encunter fd shrtages at the tim e f laying. N r are their clutch sizes likely t be lim ited by eclgical needs fr relatively large eggs. Increased average fd availability seem t play little p art in th e evlutin f increased average clutch size in w aterfw l. Instead, I w uld suggest that, w here fd supplies are nt a lim iting factr, clutch size is likely t be lim ited by decreasing parental effectiveness, decreased available ptim um breeding tim e, and increased prbability f nest predatin. These fur selective factrs are in turn fu rth er bscured by prxim ate factrs w hich prduce th e individual, seasnal and lcal variatins in clutch size encuntered by field bilgists. 10 1 c - J I I I I I I 1 I I I 85 80 75 70 65 60 55 50 45 40 35 30 July istherm ( F) at nrthern breeding limit Figure 1. The relatinship between average clutch size and nrthern breeding limits in fifty nrthern hemisphere Anatidae.
148 Paul A. Jhnsgard 120 110 Sea Ducks 100 90 80 2 70 \ Dabbling and Perching Ducks \ Whistling Ducks I V \ \ \ \t \ > \ Ia \ 60 50 40 30 20 i Pchards / \ Swans c ~~ \ N \ A + \ \ 0 /.V ' \ J Geese / / + 1 ' s i I 1 \ 1 D 1 1 't ------ \ / - - f / 10 85 80 75 70 65 60 55 50 45 40 35 July istherm ( F) at nrthern breeding limit 30 Figure 2. The relatinship between the ttal average dutch weight relative t adult female weight in the nrthern breeding limits in frtythree nrthern hemisphere Anatidae. Sum m ary References A review f Anatidae clutch-size data and an analysis f David Lack s hypthesis that fd availability t the female arund the time f her nesting can accunt fr the evlutin f average waterfwl clutch-sizes suggest several additins t r mdificatins f his thery. First, in cntrast t Lack s suggestin, prximate factrs d influence anatid clutch-sizes and result in measureable seasnal, yearly and perhaps als interppulatinal variatins in these. Secndly, indirect evidence suggests that several factrs in additin t average fd supplies may have influenced clutch-size evlutin. These include needs fr efficient partitining f energy reserves between the incubating female and her eggs, dangers f pre-incubatin clutch lsses in thse species that are unusually susceptible t nest predatin, and decreasing parental effectiveness as well as pssible decreased available ptimum breeding perids assciated with increases in clutch-sizes. Several r all f these may have placed upper limits n anatid clutch-sizes independently f r in cnjunctin with fd supply effects. Ahlen, I. & Anderssn, A. 1970. Breeding eclgy f an Eider ppulatin n Spitzbergen. Ornis Scand. 1:83-106. Barry, T. 1962. Effects f late seasns n Atlantic Brant reprductin J. Wildl. Mçmt, 26:19-26. Barry, T. 1967. The geese f the Andersn River Delta, N.W.T. Unpublished PhD. thesis, Univ Alberta, Edmntn. Bengtsn, S-A. 1971. Variatins in clutch-size in ducks in relatin t fd supply. Ibis, 113:523-6. Bengtsn, S-A. 1972. Reprductin and fluctuatins in the size f duck ppulatins at Lake Myvatn, Iceland. Oiks, 23:35-58. Chate, J. S. 1967. Factrs influencing nesting success f eiders in Penbsct Bay, Maine. J. Wildl. M gmt, 31:769-77. Cdy, M. L. 1971. Ecngical aspects f reprductin. In: Avian Bilgy. Pp. 461-512. New Yrk: Academic Press. Cch, F. G. 1961. Eclgical aspects f the Blue-snw Gse cmplex. Auk, 78:72-9. Cch, F. G. 1965. The breeding bilgy and
Clutch size f Anatidae 149 management f the N rthern Eider (Smateria mllissima brealis) in the Cape Drset area. Nrthwest Territries. Can. Wildl. Serv., Wildl. M gmt Bull., Ser. 2, N. 10. Dane, C. W. 1966. Sme aspects f the breeding bilgy f the Blue-winged Teal. Auk, 83:389-402. Eygenraam, J. A. 1957. The sex-rati and the prductin f the Mallard, Anas platyrhynchs L. Ardea, 45:177-43. Freeman, M. M. R. 1970. Observatins n the seasnal behavir f the Hudsn Bay Eider. Can. Fid. Nat. 84:145-53. Grice, D. & Rgers, J. P. 1965. The Wd Duck in Massachusetts. Mass. Div. Fisheries & Game, Final Rept. Prj. N. W-19-R. 96 pp. Grss, A. O. 1938. Eider ducks f Kent's Island. Auk, 55:387-100. Gudmundssn, F. 1932. Bebachtungen an isländischen Eiderenten (Smateria m.mllissima). Beitr. Frtpfe. Bil. Vögel, 9: 85-93; 142-47. Haartm en, L. vn 1971. Ppulatin dynamics. In: Avian Bilgy. Vl. 1, pp. 391-459. New Yrk: Academic Press. Heinrth, O. 1922. Die Beziehungen zwischen Vgelgewicht, Eigewicht, Gelegewicht und Brutauer, J. Orn., Lpz. 70:172-285. Hilden, O. 1964. Eclgy f duck ppulatins in the island grup f Valassaaret, G ulf f Bthnia. Ann. Zlgici, 1:153-279. Klmp, O. 1970. The determ inatin f clutchsize in birds. A review. Ardea, 58:1-124. Kskimies, J. 1957. Plymrphic variability in clutch size and laying date f the Velvet Scter, Melanitta fusca (L.). Ornis Fennica, 34:118-28. Kssack, C. W. 1950. Breeding habits f Canada Geese under refuge cnditins. Am. Midi. Nat. 43:627-49. Lack, D. 1967. The significance f clutch-size in waterfwl. Wildfwl Trust Ann. Rep. 18: 125-8. Lack, D. 1968a. Eclgical Adaptatins fr Breeding in Birds. Lndn: Methuen. Lack, D. 1968b. The prprtin f ylk in the eggs f waterfwl. Wildfwl, 19:67-9. Lemieux, L. 1959. The breeding bilgy f the G reater Snw Gse n Bylt Island. Nrthwest Territries. Can. Fid. Nat. 73: 117-28. Lewis, H. F. 1939. Size f sets f eggs f the American Eider. J. Wildl. M gmt, 3:70-3. Marshall, I. K. 1967. The effects f high nesting density n the clutch-size f the Cmmn Eider Smateria mllissima L. (abstract). J. Ecl. 55(3):59pp. Mendall, H. L. 1958. The Ring-necked Duck in the Nrtheast. U. Maine Studies, 2nd Ser., N. 73. U. Maine Bull. 60(16): 1-317. Parsns, J. 1970. Relatinship between egg size and pst-hatching chick m rtality in the Herring Gull (Larus argentatus). Nature, Lnd, 228:1221-2. Paynter, R. A. 1951. Clutch-size and egg m rtality f Kent Island eiders. Eclgy, 32: 497-507. Ryder, J. F. 1970a. A pssible factr in the evlutin f clutch size in Rss s Gse. Wilsn Bull. 82:5 13. Ryder, J. F. 1970b. Timing and spacing f nests and breeding bilgy f Rss Gse. Unpublished Ph.D. dissertatin, Univ. f Saskatchewan, Saskatn. Swls, L. K. 1959. Prairie Ducks. W ashingtn, D.C.: Wildlife Mgmt. Inst. Vus, K. H. 1960. Atlas f Eurpean Birds. New Yrk: Nelsn. Wagner, H. O. 1960. Beziehungen zwischen Umweltfactren under der Brutziet, Zahl der Gelege swie ihres Grösse. Zl. Anz. 16:161-72. Weller, M. W. 1964. The reprductive cycle. In: The Waterfwl f the Wrld. Vl. 4, pp. 35-79. Lndn: Cuntry Life. Paul A. Jhnsgard, The University fn ebraska- Lincln, Departm ent f Zlgy, Oldfather Hall, Lincln, Nebraska 68508, U.S.A.