Behavioral and Morphological Adaptations by Galapagos Land Iguanas (Conolophus subcristatus) to Water and Energy Requirements of Eggs and Neonates 1

Transcription

1 AMER. ZOOL., 25: (1985) Behaviral and Mrphlgical Adaptatins by Galapags Land Iguanas (Cnlphus subcristatus) t Water and Energy Requirements f Eggs and Nenates 1 HOWARD L. SNELL 2 AND C. RICHARD TRACY Department f Zlgy and Entmlgy, Clrad State University, Frt Cllins, Clrad SYNOPSIS. We examine the imprtance f bth the hydric envirnment f naturally incubating reptilian eggs and the energetic needs f hatchlings via an investigatin f reprductin in Galapags land iguanas (Cnlphus subcristatus). Hatching success f egg clutches and the size f subsequent hatchlings are bth psitively crrelated with the water ptential f natural nests, as predicted frm previus labratry experiments. Water ptentials representing ptimal incubatin envirnments are available fr nly a brief perid in nature, and depend upn the same seasnal rainfall as des fd abundance fr the emerging hatchlings. The tempral placement f the reprductive seasn f Cnlphus subcristatus balances these cnflicting needs fr water by eggs and energy by hatchlings. Ovipsitin ccurs slightly befre suitable water ptentials are reached and hatchling emergence ccurs after the peak in fd abundance. Mrphlgical adaptatins by Cnlphus subcristatus t their precarius reprductive phenlgy include greater amunts f albumen in their eggs, and greater energy reserves in emergent hatchlings than mst ther lizard species. These adaptatins lessen the severity f an arid envirnment where water becmes available fr perids t shrt t allw bth vipsitin and hatching t be temprally placed in an ptimal manner. INTRODUCTION The rle f energy in the reprductive bilgy f reptiles has received cnsiderable attentin, nearly all f which fcuses n energetic requirements f adults and/ r the energy cntained in clutches f eggs (see Cngdn et al., 1982, fr an extensive review). A cmparatively few studies have prvided infrmatin n energy surces f nenates (Mcllhenny, 1934; Fitch, 1960; Platt, 1969; Clark, 1970; Ewert, 1979; Kraemer and Bennett, 1981; Andrews, 1982; Cngdn et al., 1983&; and Tryer, 1983). In this paper we examine the interactin between water requirements f incubating lizard eggs and the energetics f recently emerged nenates. Labratry experiments have demnstrated the ptential imprtance f the hydric envirnment fr naturally incubating reptilian eggs (Tracy et al., 1978; Packard et al., 1979; Tracy, 1980; Packard and Packard, 1980; and Tracy and Snell, 1985). 1 Frm the Sympsium n Animal Energetics: Amphibians, Reptiles, and Birds presented at the Annual Meeting f the American Sciety f Zlgists, December 1983, at Philadelphia, Pennsylvania. * Present address: Department f Bilgy, Memphis State University, Memphis, Tennessee Theretical mdeling f the factrs influencing water transprt between eggs and the nest medium prvide predictins abut a variety f behaviral and mrphlgical adaptatins that culd be used by reptiles t mdify the effects f detrimental envirnmental cnditins (Tracy, 1982; Tracy and Snell, 1985). Hwever, tests f such predictins in natural cnditins are lacking. The results frm experiments with Galapags land iguanas (Cnlphus subcristatus) have shwn that the water ptential f the incubatin medium influences bth the size f hatchlings and the frequency f mrtality f eggs incubated under artificial cnditins (Tracy and Snell, 1985). T investigate the effects f sil-water ptential n hatchling size and hatching success under natural cnditins, and the effect f the timing f the reprductive seasn n the energetics f hatchlings, we studied Galapags land iguanas n Isla Plaza Sur, in the Galapags Archipelag. T gain insight int the pssible use f energy reserves (ylk and fat) by hatchlings f varius lizard species, we cmpare hatchlings f Galapags land iguanas frm a ppulatin in which hatching ccurs after the end f the rainy seasn, when fd abundance

2 1010 H. L. SNELL AND C. R. TRACY is lw, t hatchlings f ther iguanid lizards whse reprductive seasns prduce hatchlings when fd is abundant. PREDICTIONS AND TESTS Predictins frm ur experimental studies (Tracy and Snell, 1985) include: 1) In natural nests, the size f hatchlings and the hatching success f eggs will be influenced by the water ptential f the nest medium. 2) The reprductive seasn f land iguana ppulatins frm small, arid islands (small islands in the Galapags Archipelag are typically drier than larger islands due t the small islands' lack f influence n lcal weather patterns, see Alpert, 1963, and Wiggins and Prter, 1971) will be timed s that egg incubatin ccurs during a perid crrespnding t the greatest sil-water ptential. 3) If adjusting the reprductive seasn causes hatchlings t emerge t cnditins f relatively lw fd abundance, such hatchlings will pssess energy reserves greater than thse pssessed by hatchlings f species that emerge t mre abundant fd. Tests f the first tw predictins are relatively simple. We will cmpare the sizes f land iguana hatchlings f ne ppulatin amng years differing in the amunt f rainfall during the incubatin perid. We als present an analysis f the size f hatchlings frm individual nests, within ne year, in relatin t the water ptential f sil frm the nests. The predictin that water ptential will affect the hatching success f eggs in nests is tested by regressin. A rigrus test f the secnd predictin wuld invlve cmparing the tempral placement f the reprductive seasn between large and small islands in cnjunctin with detailed cmparisns f the yearly patterns f sil-water ptential. We are unable t cnduct such a test, but we will cmpare the yearly patterns f silwater ptential and the reprductive seasn f ne ppulatin f land iguanas. Testing the third predictin is mre cmplex. First, the presence f energy reserves in hatchlings will be demnstrated. Previus research has either cnsidered ttal lipid levels f nenates (Cngdn et al., 1983a, b), r cncentrated n ylk that remains after hatching and is absrbed by the nenate (Mcllhenny, 1934; Fitch, 1960; Clark, 1970; Platt, 1969; Ewert, 1979; Kraemer and Bennett, 1981; Andrews, 1982; and Tryer, 1983) as imprtant energy reserves f nenatal reptiles. We will demnstrate that, in the cntext f a lng-term energy-strage medium, absrbed ylk is nt as imprtant as abdminal fat bdies (prduced frm egg ylk during embrynic develpment) fr the hatchlings f several species f lizards. We will then cmpare the amunts f abdminal fat in the hatchlings f five species f lizards, ne whse hatchlings demnstrably emerge t abundant fd {Iguana iguana; Rand and Greene, 1982), three whse hatchlings apparently emerge during perids f high fd abundance (Scelprus undulatus, Phrynsma crnutum, and Crtaphytus cllaris; Duvall et al., 1982) and a fifth whse hatchlings will be shwn t emerge t an envirnment in which fd abundance is declining (Cnlphus subcristatus). MATERIALS AND METHODS We studied Galapags land iguanas frm Isla Plaza Sur, a 12 hectare island supprting a ppulatin f apprximately 350 adult iguanas (Snell et al., 1984). The reprductive activity f this ppulatin was studied during 1979, 1980, 1981, and Nests were marked as they were laid, and emerging hatchlings were captured in wire cages. After a nest hatched, it was excavated and hatched and unhatched eggs were cunted and sil samples cllected. Nests that failed t prduce hatchlings were excavated 140 days after laying (the lngest bserved incubatin perid f a successful nest) and the same data were cllected. Rainfall was measured during the mnths f incubatin (January t July) in 1979, 1980, and 1981, and until April in Because this perid includes the rainy seasn, such a seven mnth measurement accunts fr the vast majrity f rainfall within a year (Alpert, 1963). In 1979, 1981,

3 REPRODUCTIVE ADAPTATIONS OF LAND IGUANAS 1011 and 1982 a sample f nests was excavated within a day f vipsitin, and the masses f the eggs measured n an Ohaus "Dial-a- Gram" balance (accurate t 0.01 g). Samples f sil frm nesting areas were cllected peridically at nest depths thrughut the incubatin perid. All sil samples cllected were sealed (with rubber stppers fllwed by wax) in glass vials and the water ptential was measured six t eight mnths later with a Wescr C52 Sample Chamber. Cmparisns f water cntent between samples stred in glass vials (fr the same perid) and thse measured immediately after cllectin shwed n lss f water during strage. Wet masses f hatchlings, their absrbed ylk and abdminal fat were recrded fr Cnlphus subcristatus frm Plaza Sur; Scelprus undulatus and Phrynsma crnutum frm Guadelupe Muntains, Texas; and Crtaphytus cllaris frm Manhattan, Kansas. Dr. Katherine Tryer kindly prvided mean values fr Iguana iguana frm the Gatun Lake area f Panama. These masses were either recrded within ne day f hatching (t be used fr interspecific cmparisns f reserves), r at specified psthatching intervals (t cmpare use f the different reserves). We measured grwth and mass lss in five land iguana hatchlings held withut fd in simulated nest chambers at 30 C (this temperature is within the range f natural nest-chamber temperatures) t further investigate the use f energy reserves. Cnlphus subcristatus hatchlings n Plaza Sur use parts f apprximately seven species f perennial plants and all available annuals as regular fd items. T measure patterns f fd abundance during the year n Plaza Sur we used a mdified line-intersect technique t estimate the numbers f leaves f tw species f perennials (Parkinsnia aculeata and Grabwskia berhaaviaeflia), a direct cunt t estimate the number f flwers f ne species f perennial (Opuntia echis), and the pint-quarter methd t sample annuals, befre, during, and after the rainy seasn in 1981 and In sampling vegetatin, we first randmly selected ten pints n the island. We then lcated the clsest individual f each perennial 102 "E E "~ 2 E 98 i > 94 t t n CONOLOPHUSSUeCRISTAWS I 11 TT l 1982 n-18 I\ n-14 * 1981 n*224 PLAZA SUR O RAINFALL DURING INCUBATION (mm) GALAPAGOS ^ n>t92 FIG. 1. Cmparisns f the mean sizes f hatchlings frm fur years as a functin f the amunt f rainfall during the incubatin perid. Bars are 95% cnfidence limits n the means using the pled standard deviatin frm the entire sample. See text fr explanatin f statistical treatment. species t be sampled t these pints and used thse same individuals in each subsequent sampling sessin. Each f the ten randmly selected pints was used as a pint fr the pint-quarter sampling f annuals, and tw ther pints (2.5 m nrth, and 2.5 m east f the randm pint) were als sampled. The same pints were sampled in all subsequent measures f annual abundance. The cunts fr each f the individual perennial plants were nrmalized t the highest cunt frm that individual, and the densities f annuals at each f the ten randmly selected pints were nrmalized t the highest density f annuals fund at that pint. Thus, we culd examine patterns in relative abundance f the varius types f vegetatin ver the year. Fr 1979 and 1980 we made subjective measures f the same plant types n a bimnthly basis frm January until July. RESULTS Rainfall during the incubatin perid clearly affected the mean size f hatchlings in the fur years they were studied (Fig. 1). A Student-Newman-Keuls Multiple Range Test shwed that mass differed significantly^ < 0.05) amng all years except 1979 and 1980, and 1979 and 1982; and lengths f individuals differed significantly (P < 0.05) between all years except 1979 and 1982, and 1980 and Overall, hatchlings with greater mass and length were prduced in years in which mre rain

4 1012 H. L. SNELL AND C. R. TRACY TABLE 1. Cmparisn f mean egg masses between years in Cnlphus subcristatus.* Year Mean egg mass (g) Sample size * The results f an analysis f variance (F (s 1M) = 2.59, 0.1 < P < 0.05) indicate n significant difference between years in mean egg-mass, at a 0.05 level f significance. fell during the incubatin perid than in years with less rainfall. A pssible alternative hypthesis t accunt fr these bserved differences in size is that the initial size f the eggs was greater fr the years with larger hatchlings. Previus labratry research (Packard and Packard, 1980; Tracy, 1982) has shwn that the mass f eggs at vipsitin is a significant cvariate f hatchling size in turtles and lizards. We were unable t d a similar analysis f cvariance with hatchling land iguanas because we fund, by cmparing the hatching success f disturbed and undisturbed nests in which at least ne egg had hatched, that the simple actin f pening a nest and measuring the mass f the eggs has a significant effect n the hatching success f that nest (mean hatching success f disturbed nests = 35%; mean hatching success f undisturbed nests = 87%; n( disturbed n(. st5 ) = 4; n (undisturb<. d ne«s) = 26; Mann-Whitney statistic = 443, P = 0.016). As hatching success is an imprtant variable fr ther parts f this study, such an effect had t be avided. Therefre, we cmpared masses f eggs (frm nests nt used in the rest f the study) between years. If differences in egg size are respnsible fr differences in hatchling size, then years with large hatchlings shuld als be years with large eggs. Hwever, egg mass did nt vary significantly between 1979, 1981 and 1982 (Table 1). Cnsequently, egg size differences cannt accunt fr the variatin in hatchling size between years, and the alternative hypthesis can be rejected. The sizes f hatchlings were significantly crrelated with the water ptential within the nests (Fig. 2) as were the hatching successes f clutches (Fig. 3). Because these j 2 10 >» Ml X 3 O < i e WATER POTENTIAL OF NEST IMP.) FIG. 2. Linear regressins f mean snut-vent length (SVL) and mass f nestmates (SIBS) against the silwater ptential f their individual nests. Bth regressins are highly significant: F (1-<4) = 8.18,0.005 < P < 0.01 fr mass, and F (144) = 8]68, < P < 0.01 fr SVL. cmparisns were made n data frm within ne year (1981), they primarily reflect differences in the sil-water ptential f different sil types rather than differing amunts f water available t each nest. Again, we were unable t perfrm an analysis f cvariance n these data with the initial egg size. Hwever, we did cmpare initial masses f eggs frm nests laid in sils characterized by lw water ptential with initial masses f eggs frm sils characterized by high water ptential, and fund n significant difference (mean egg mass frm "dry" sil nests = g, mean egg mass frm "wet" sil nests = g; t i0 = 0.635, P = ). Hence, there is n clinearity evident between initial egg mass and sil-water ptential f the nest. Cnlphus n Plaza Sur lay their eggs slightly befre the beginning f the rainy seasn and mst f incubatin ccurs during the perid f greatest water ptential (Fig. 4). Hatching ccurs subsequent t the rainy seasn, and hatchlings emerge t a perid f increasing scarcity f fd, which lasts until the rainy seasn f the next year (Fig. 4). While the fd abundance and sil misture curves shwn in Figure 4 are fr

5 REPRODUCTIVE ADAPTATIONS OF LAND IGUANAS c CO m CO X 40 u < I 20 0 * "> WATER POTENTIAL OF NEST (MPa) FIG. 3. Linear regressin f hatching success f clutches against sil-water ptential f their nests. Open circles represent water ptential values less than - 11 MPa (the limit f the instrument used). Thse pints were nt used in the regressin, althugh including them at arbitrary values f 11 MPa des nt change the significance f the regressin. The regressin is highly significant: F (1, 8 j, = 36.4, P < ne year, 1981, the pattern was essentially the same frm 1979 t 1982 (Snell, unpublished data). The main difference between years is that in sme years mre rain falls befre March, shifting the fd abundance (exclusive f Opuntia) and water ptential curves t a pint slightly earlier in the year. Only in El Nin years is fd abundance likely t be greater later in the year than shwn here (Alpert, 1963; Wiggins and Prter, 1971). Relative energy stres f Cnlphus hatchlings are greater than thse f Iguana iguana, Scelprus undulatus, Phrynsma crnutum, and Crtaphytus cllaris hatchlings, regardless f the frm f the energy stre (Table 2). All f these species differ significantly in the relative amunt f absrbed ylk pssessed by hatchlings (except fr Scelprus and Phrynsma), but nly Cnlphus and Crtaphytus each differ significantly frm ther species treated here in the relative amunt f abdminal fat in hatchlings (Table 2). Hatchlings f Cnlphus subcristatus, Phrynsma crnutum, and Scelprus undulatus metablized absrbed ylk much mre rapidly than abdminal fat, even thugh the amunt f absrbed ylk was always greater initially (Figs. 5-7). This resulted in reserves f abdminal fat remaining after all r almst all f the absrbed ylk was gne. All f the Cnlphus hatchlings held withut fd grew in length. Once grwth in length stpped, lss f mass accelerated significantly (Fig. 8 illustrates grwth in length and lss f mass fr a representative hatchling). Using equatins frm the literature (mdified frm Vitt, 1978), we cnverted the mass lst t energy lst, with cnversin factrs fr ylk during the perid f grwth, and fr bdy tissues after grwth stpped. The resultant line f energy lss against time is straight ver bth the perids f grwth and n grwth, with n significant differences in slpe (t = , 0.5 < P < 0.9).

6 1014 H. L. SNELL AND C. R. TRACY OVIPOSITION TABLE 2. Cmparisns f energy stres pssessed by hatchlings f fur species f lizards. * 0.8 Species Scelprus undulatus Phrynsma crnutum Crtaphytus cllaris Iguana iguana Cnlphus subcristalus Percent ab- Percent abdm- Sample srbed ylk inal fat size 1.19* 0.45* * 0.501* * * Analyses f variance, perfrmed n arcsine transfrmed data (Skal and Rhlf, 1969), shw significant difference amng the means f bth the percent f ttal hatchling mass represented by absrbed ylk ( F <j.s7) = P < 0.001), and the percent f ttal hatchling mass represented by abdminal fat (F (557) = P < 0.001). Means with asterisks (*) are nt significantly different (P > 0.05) as determined by Student-Newman-Keuls Multiple Range tests n arcsine transfrmed data. The size f the Iguana iguana sample is estimated frm Tryer (1983), and is nt influential in the statistical results. J F M A M J J MONTH FIG. 4. Representative fluctuatins in sil-water ptential and fd abundance frm January t July 1981 n Isla Plaza Sur. Fur types f plants are presented: squares = Opuntia, regular triangles = Grabwskia, inverted triangles = Parkinsnia, and circles = annuals. These fur plant types are amng the mst preferred fd items fr hatchling and yearling Cnlphus (persnal bservatin) and are representative f all plants n Plaza Sur in terms f seasnal abundance. See text fr explanatin f plant parts mnitred and nrmalizatin prcedures. The dtted line between samples in April and May represents a subjective interpretatin f general fd abundance during that time, based n nn-quantitative bservatins. Analyses f variance shw that all plant species differ in abundance amng mnthly samples (Opuntia, F(4.44> = 135.3; Grabwskia, F (435) = ; Parkinsnia, F (M5) = 34.2; and annuals, F (4, 4S) = 20.72; P < fr all types). Bars n the sil-water ptentials represent 95% cnfidence limits arund the means fr each mnth, based upn the pled standard deviatin. Fr the mnths f March and April, the bars are truncated at water ptential values f 0, because values higher than 0 are, by definitin, impssible. Analysis f variance indicates significant differences amng mnthly means (F (M7) = 6.03, P < 0.001). The water ptential prfile is frm ne f apprximately fur different sil types present n Isla Plaza Sur. This particular sil type, a sandy gravel, is the type mst used by nesting females. The annual pattern is the same fr ther sil types, but the sil-water ptential values wuld pssibly be lwer (Snell and Tracy, unpublished data). in S UJ a. 100 e PLAZA SUR, GALAPAGOS CONOLOPHUS SUBCRISTATUS i 20 AT HATCHING AT EMERGENCE ABDOMINAL FAT D ABSORBED YOLK 0 BODY FIG. 5. Percent f ttal bdy mass represented by abdminal fat, absrbed ylk, and bdy tissues in Cnlphus subcristatus hatchlings at hatching and at emergence frm the nest. Land iguana hatchlings apparently spend apprximately 10 t 14 days between hatching and emerging frm the nest. The amunt f absrbed ylk is significantly less at emergence than at hatching (/, = 43.83, P < 0.001), while the amunt f abdminal fat is nt significantly changed (U = ,0.5 <P < 0.9).

7 REPRODUCTIVE ADAPTATIONS OF LAND IGUANAS 1015 c 3.0 < 2 i i I X Q til CC 2.0 O c CO u 1.0 ai UJ cc z 2 O a CD \ \ \ \\\ \ * DAYS SINCE HATCHING FIG. 6. Use f absrbed ylk and abdminal fat by Phrynsma crnutum hatchlings. Lines cnnect the means f the samples at each age. DISCUSSION The hatching success f Galapags land iguana clutches in natural nests, as well as the size f hatchlings, are psitively crrelated with the water ptential f the nest medium, bth within and between years, in accrdance with ur predictin ne. Water ptentials yielding the largest hatchlings and greatest hatching successes are thse greater than 3 MPa (Figs. 2, 3). Spradic, seasnal rainfall in the Galapags islands, and the general aridity f the archipelag (Alpert, 1963), permit water ptentials greater than 3 MPa nly during and immediately after the rainy seasn n small islands like Isla Plaza Sur (Fig. 4). The abundance f fd available t new hatchlings als depends upn the ccurrence f rains, and is highest during the latter part f the rainy seasn. If hatchlings emerge when there is insufficient fd t meet maintenance needs and t increase fat stres, then they may nt be able t survive later perids f lw fd availability. Water availability, then, appears t be an imprtant factr in the evlutin f the timing f the reprductive seasn via tw ppsing selectin pressures: incubatin must cincide with sufficient sil-water ptential, and hatchling emergence must ccur during perids f suitable fd abundance. Because f the length f the incubatin perid in Cnlphus (apprximately 97 t 140 days) bth f these cnditins cannt be met ptimally (Fig. 4). Hence, land iguanas vipsit befre sil-water cnditins are ptimal, their eggs incubate thrugh the perid f greatest sil-water ptential, and their hatchlings emerge t an envirnment characterized by decreasing amunts f fd. This pattern agrees with predictin tw. Land iguanas apparently have evlved a balance between detrimental effects f their arid envirnment thrugh the timing f their reprductive seasn. Specifically, if land iguanas frm Plaza Sur laid eggs later in the year, the eggs culd encunter suitable nest mistures,

8 1016 H. L. SNELL AND C. R. TRACY i.a F n>36 p<000l 2 4 DAYS SINCE HATCHING "ABDOMINAL FAT "ABSORBED YOLK , p<0025 FIG. 7. Use f absrbed ylk and abdminal fat by Scelprus undulatus hatchlings. Absrbed ylk is used significantly faster than abdminal fat. but hatchlings wuld face even less fd upn emergence. If hatchlings emerged earlier in the year, they wuld have cmparatively abundant fd, but the eggs frm which they hatched wuld have t be laid lng befre adequate sil-water ptentials had been reached and, cnsequently, hatching success and hatchling size wuld decrease. Hatchling size has been shwn in ther species t be an imprtant characteristic in that large hatchlings survive the first year f life mre ften than smaller hatchlings in the lizard Uta stansburiana (Fx, 1978; Fergusn and Bhlen, 1978), and we have bserved a similar pattern in Cnlphus (Snell, Tracy, and Snell, in preparatin). The phenlgy f reprductin in iguanine lizards has been variusly reprted as a respnse t the fllwing: availability f fd fr hatchlings, availability f fd fr reprductive females, thermal cnditins f the nest site, water relatins f the nest site and cmpetitin amng cnspecific females fr nesting sites, as well as a variety f ther factrs (see Muth, 1980; Wiewandt, 1982; Rand and Greene, 1982; Christian and Tracy, 1982; Werner, 1982; PLAZA SUR, GALAPAGOS CONOLOPHUS SUBCRISTATUS 95 *- ^ M A S S, SVL 33 ^ x^ MASS;, j S COMPARISON OF MASS LOSS SLOPES: m,'-0.i607 m t«5.041 p< AGE (DAYS) FIG. 8. Grwth and mass lss in a representative land iguana hatchling held withut fd (see text fr details). The vertical dtted line indicates the pint at which grwth stps. The rates f mass lss befre and after that pint are significantly different. 25

9 REPRODUCTIVE ADAPTATIONS OF LAND IGUANAS 1017 Duvall et al., 1982; and their cited references). Tracy (1982) presented a summary f selectin pressures n eggs, clutches, and females frm which the pattern we have demnstrated here, where the actual timing f reprductin is a tradeff between different selectin pressures (in this case water availability and energy needs), can be predicted. The current identificatin f selective frces mlding the reprductive seasnality f a ppulatin f Galapags land iguanas represents a psitive test f the predictin that interppulatin variatin in reprductive phenlgy f these lizards is a respnse t lcal selective pressures rather than a phylgenetic artifact (SneWetal, 1984). What cnstitutes an energy strage medium fr hatchling lizards? We have demnstrated that abdminal fat bdies, while initially smaller in mass than the absrbed ylk, remain in the nenate lizard after the absrbed ylk has been cmpletely metablized. Additinally, ur grwth experiment with Cnlphus indicates that ylk supprts early grwth by the hatchling, while abdminal fat remains essentially unused. Nenatal snakes have als been shwn t grw in length, using absrbed ylk as their energy surce, while kept withut fd (Fitch, 1960); unfrtunately, the cnditin f the fat bdies in these specimens is unknwn. Abdminal fat bdies play an imprtant rle in the strage f energy fr hibernatin and reprductin (see Gregry, 1982; and Duvall et al., 1982, fr reviews), and their use by nenates as a lng-term energy strage medium during perids f fd-stress is demnstrated here. Others have prpsed that abdminal fat bdies are f insignificant size, cmpared t absrbed ylk, t serve as energy stres in green iguanas (Tryer, 1983) and green sea turtles (Kraemer and Bennett, 1981). Clearly, a cmparative verview f the imprtance f abdminal fat in nenate reptiles must await further data. Cnlphus have adapted mrphlgically t their reprductive pattern in tw nticeable ways. First, they have greater amunts f albumen in their eggs at vipsitin than mst ther lizards (Tracy and Snell, 1985). This is prbably an adaptatin t prvide the develping embry with water during the initial perid f incubatin when sil-water ptentials are dangerusly lw (Fig. 4). Secnd, they prvide greater energy reserves t their hatchlings than lizards whse hatchlings emerge t mre favrable cnditins. Bth f these adaptatins apparently lessen the severity f the effects f an arid envirnment in which water becmes available fr perids t shrt t allw bth vipsitin and hatching t be temprally placed in an ptimal manner. ACKNOWLEDGMENTS We thank Sctt Lacur, Sctt Steckbauer, Sylvia Harcurt, Randy and Sue Jennings, Warwick Reed, Katie Belt, and Bettina Vn Hagen fr their invaluable and untiring help in the field. Keith Christian graciusly prvided the gravid females f Scelprus undulatus and Phrynsma crnutum frm which we btained hatchlings. Heidi Snell deserves special thanks fr her help with data analysis and the drawing f figures, and fr accepting Isla Plaza Sur as her hme fr nearly three years. The Charles Darwin Research Statin and the Servici Parque Nacinal Galapags prvided encuragement, permissin, and lgistical and financial assistance fr this research; fr that and a multitude f ther things we thank them. Additinal financial assistance fr varius aspects f this research came frm grants t H.L.S. frm the Ecuadrian Gvernment, the Smithsnian Institutin, the Wrld Wildlife Fund (#1544), and the United States Peace Crps; a Guggenheim Fellwship t C.R.T., and the Department f Zlgy and Entmlgy at Clrad State University. REFERENCES Alpert, L The climate f the Galapags Islands. In Galapags Islands: A unique area fr scientific investigatins, pp Occas. Pap. Califrnia Acad. Sci. 44. Andrews, R. M Patterns f grwth in reptiles. In C. Gans and F. H. Pugh (eds.), Bilgy f the Reptilia, Vl. 13, Physilgy D, Physilgical eclgy, pp Academic Press, Lndn.

10 1018 H. L. SNELL AND C. R. TRACY Christian, K. A. and C.R.Tracy Reprductive behavir f Galapags land iguanas, Cnlphus pallidus, n Isla Santa Fe, Galapags. In G. M. Burghardt and A. S. Rand (eds.), Iguanas f the wrld, their behavir, eclgy, and cnservatin, pp Nyes Publicatins, Park Ridge, New Jersey. Clark, D. R., Jr Eclgical study f the wrm snake Carphphisverrms (Kennictt). Univ. Kansas Publ. Mus. Nat. Hist. 19: Cngdn, J. D., A. E. Dunham, and D. W. Tinkle Energy budgets and life histries f reptiles. In C. Gans and F. H. Pugh (eds.), Bilgy f the Reptiha, Vl. 13, Physilgy D, Physilgical eclgy, pp Academic Press, Lndn. Cngdn.J. D., D. W. Tinkle, and P. C. Rsen. 1983a. Egg cmpnents and utilizatin during develpment in aquatic turtles. Cpeia 1983: Cngdn, J. D., J. W. Gibbns, and J. L. Greene. 1983i. Parental investment in the chicken turtle (Deirchelys reticularia). Eclgy 64: Duvall, D., L. J. Guillette, and R. E. Jnes Envirnmental cntrl f reptilian reprductive cycles. In C. Gans and F. H. Pugh (eds.), Bilgy f the Reptilia, Vl. 13, Physilgy D, Physilgical eclgy, pp Academic Press, Lndn. Ewert, M. A The embry and its egg: Develpment and natural histry. In M. Harless and H. Mrlck (eds.), Turtles: Perspectives and research, pp Jhn Wiley and Sns, New Yrk. Fergusn, G. W. and C. H. Bhlen Demgraphic analysis: A tl fr the study f natural selectin f behaviral traits. In N. Greenberg and P. D. MacLean (eds.), Behavir and neurlgy f lizards, an interdisciplinary cllquium, pp U.S. Department f Health, Educatin, and Welfare, Washingtn, D. C. Fitch, H. S Auteclgy f the cpperhead. Univ. Kansas Pub. Mus. Nat. Hist. 13: Fx, S. F Natural selectin n behaviral phentypes f the lizard Uta stansburiana. Eclgy 59: Gregry, P. T Reptilian hibernatin. In C. Gans and F. H. Pugh (eds.), Bilgy f the Reptilia, Vl. 13, Physilgy D, Physilgical eclgy, pp Academic Press, Lndn. Kraemer, J. E. and S. H. Bennett Utilizatin f psthatching ylk in lggerhead sea turtles, Caretta caretta. Cpeia 1981: Mcllhenny, E. A Ntes n incubatin and grwth f alligatrs. Cpeia 1934: Muth, A Physilgical eclgy f desert iguana (Dipssaurus drsalis) eggs: Temperature and water relatins. Eclgy 61: Packard, G. C, T. L. Taigen, T. J. Bardman, M. J. Packard, and C. R. Tracy Changes in mass f sftshell turtle (Trinyx spinerus) eggs incubated n substrates differing in water ptential. Herpetlgica 35: Packard, M. J. and G. C. Packard Water balance f the eggs f a desert lizard (Callisaurus dracnides). Canad. J. Zl. 58: Platt, D. R Natural histry f the hgnse snakes Heterdn platyrhins and Heterdn nasicus. Univ. Kansas Pub. Mus. Nat. Hist. 18: Rand, A. S. and H. W. Greene Latitude and climate in phenlgy f reprductin in the green iguana, Iguana iguana. In G. M. Burghardt and A. S. Rand (eds.), Iguanas f the wrld, their behavir, eclgy, and cnservatin, pp Nyes Publicatins, Park Ridge, New Jersey. Skal, R. R. and F. J. Rhlf Bimetry: The principles and practice f statistics in bilgical research. W. H. Freeman and Cmpany, San Francisc. Snell, H. L., H. M. Snell, and C. R. Tracy Variatin amng ppulatins f Galapags land iguanas (Cnlphus): Cntrasts f phylgeny and eclgy. Bil. J. Linn. Sc. 21: Tracy, C. R On the water relatins f parchment-shelled lizard (Scelprus undulatus) eggs. Cpeia 1980: Tracy, C. R Biphysical mdeling in reptilian physilgy and eclgy. In C. Gans and F. H. Pugh (eds.), Bilgy f the Reptilia, Vl. 12, Physilgy C, Physilgical eclgy, pp Academic Press, Lndn. Tracy, C. R., G. C. Packard, and M. J. Packard Water relatins f chelnian eggs. Physil. Zl. 51: Tracy, C. R. and H. L. Snell Interrelatins amng water and energy relatins f reptilian eggs, embrys, and hatchlings. Amer. Zl. 25: Tryer, K Psthatching ylk energy in a lizard: Utilizatin pattern and interclutch variatin. Oeclgia 58: Vitt, L. J Calric cntent f lizard and snake (Reptilia) eggs and bdies and the cnversin f weight t calric data. J. Herp. 12: Werner, D. I Scial rganizatin and eclgy f land iguanas, Cnlphus subcristatus, n Isla Fernandina, Galapags. In G. M. Burghardt and A. S. Rand (eds.), Iguanas f the wrld, their behavir, eclgy, and cnservatin, pp Nyes Publicatins, Park Ridge, New Jersey. Wiewandt, T. A Evlutin f nesting patterns in iguanine lizards. In G. M. Burghardt and A. S. Rand (eds.), Iguanas f the wrld, their behavir, eclgy, and cnservatin, pp Nyes Publicatins, Park Ridge, New Jersey. Wiggins, I. L. and D. M. Prter Flra f the Galapags Islands. Stanfrd University Press, Stanfrd.