Do Changing Moisture Levels during Incubation ... - Semantic Scholar

2 downloads 0 Views 107KB Size Report
May 24, 2005 - and hatchling traits, in the wall lizards (Podarcis muralis). Comp Biochem Physiol A 124:205–213. Lillywhite H.B. and R.A. Ackerman. 1984.
524

Do Changing Moisture Levels during Incubation Influence Phenotypic Traits of Hatchling Snakes (Tropidonophis mairii, Colubridae)? Gregory P. Brown Richard Shine* Biological Sciences A08, University of Sydney, New South Wales 2006, Australia Accepted 11/8/2004; Electronically Published 5/24/2005

ABSTRACT Phenotypic traits (e.g., size, strength, speed) of hatchlings in many reptile species are influenced by hydric conditions in the nest. Previous experiments have focused on comparisons between eggs maintained under constant (but different) conditions, but eggs in natural nests frequently experience strong temporal shifts in soil water content during incubation. Keelback snakes (Tropidonophis mairii) in the Australian wet-dry tropics nest over most of the year, so early nests experience decreasing water availability during development, late nests experience increases, and others (midyear) remain stable in this respect. We mimicked these three conditions and incubated 54 eggs (nine from each of six clutches) in a split-clutch design to maintain the same average water content but with differing trajectories through incubation. The experimental treatments significantly affected the total amount of water taken up by the eggs (and thus final egg mass), but incubation periods were unaffected. Hatchling size but not strength showed minor but statistically significant effects of incubation regimes. The ability of keelback eggs to take up excess water whenever it becomes available (either early or late in development) and to retain it even when conditions change buffers embryogenesis effectively (but not completely) against fluctuations in soil water conditions.

Introduction The evolution of the amniotic egg is widely viewed as one of the major events in vertebrate evolution and has been ascribed * Corresponding author; e-mail: [email protected]. Physiological and Biochemical Zoology 78(4):524–530. 2005. 䉷 2005 by The University of Chicago. All rights reserved. 1522-2152/2005/7804-4049$15.00

a central role in allowing early reptiles to reproduce away from water and thus to invade a range of semiarid and arid environments not accessible to amphibians (Romer 1966; Hildebrand 1995). There can be no doubt that the eggs of modernday reptiles are indeed better able to develop successfully under relatively dry conditions than can those of most amphibians, but it would be a gross overstatement to describe reptile embryogenesis as independent of substratum water content (Packard et al. 1977; Packard 1991). Although some reptiles deposit calcareous-shelled eggs that can develop successfully even in extremely arid conditions, most depend on water uptake from the nest environment to sustain embryogenesis (Packard 1991). For example, permeabilities of snake eggshells are at least two orders of magnitude greater than for bird eggshells (Lillywhite and Ackerman 1984). Thus, females generally choose moist substrata for egg deposition, and eggs that are exposed to very dry conditions typically lose water, and the embryos within die (Muth 1980; Packard and Packard 1988). The amount of water needed from the substratum varies considerably, even among closely related species (hatching success can be compromised by too much or too little water; Badham 1971), and we will need detailed studies on a wide range of taxa from a wide range of environments before we can frame meaningful generalities about the water requirements of reptile embryos. Water availability is important not simply in terms of whether an egg is able to hatch but also for the phenotypic traits of the resultant hatchling. Laboratory studies show that the phenotypic traits of hatchling reptiles can be influenced by physical conditions inside the nest (Burger 1989; Andrews et al. 2000; Flatt et al. 2001; Deeming 2004). Although temperature regimes have attracted most study in this respect, the hydric environment can be important also (Gutzke and Packard 1987; Ji and Bran˜a 1999; Deeming 2004). The emphasis in most of these studies has been on steady state experiments, with eggs kept at one condition throughout incubation compared with other eggs kept at some different condition (Shine and Brown 2002; Brown and Shine 2004). In natural nests, however, extensive variation in both temperature and moisture occurs over many temporal scales. Recent experimental work has shown that such variation can affect developmental rates and hatchling phenotypes. For example, gradually increasing versus decreasing nest temperatures (as frequently occurs in the wild, especially in temperate-zone habitats) affect hatchling traits of the scincid lizard Bassiana duperreyi (Shine 2002, 2004). The same

Developmental Plasticity in Snakes 525 question seems never to have been addressed in the case of the hydric environment, but again we might expect that soil moisture often either increases or decreases over the course of incubation and, hence, that phenotypic responses to incubation conditions might depend on such shifts as well as on mean values for soil water potential averaged over the entire incubation period. One of the most intensively studied cases where soil moisture exerts biologically significant (i.e., fitness-relevant) effects on hatchling phenotypes involves small snakes (keelbacks, Tropidonophis mairii) that live in the wet-dry tropics of northern Australia. The climate in this area is hot year-round (monthly mean maxima 31⬚–35⬚C, minima 15⬚–24⬚C), but rainfall is highly seasonal. More than 75% of the 1,394-mm annual rainfall occurs during a brief wet season (December to March), when monsoonal rains cause widespread flooding. Water content of the soil at nesting sites in this study area peaks at the end of wet season rainfall (March), falling from about 10% at this time to 2% by late dry season (October). It then begins to increase with early wet season rains (Fig. 1). Female keelbacks produce multiple clutches of eggs from April to November (Brown and Shine 2002), and thus the eggs from these clutches experience a wide range of conditions in terms of not only mean moisture levels but also changes from the beginning to the end of incubation. Eggs from clutches laid early in the season will be deposited into moist soil that will dry out significantly during the 2-mo incubation period. In contrast, eggs from late clutches will be deposited in relatively dry soil that progressively becomes wetter with the onset of the monsoons. In previous work, we have shown that water potential of the nest significantly influences many phenotypic traits of hatchling keelbacks, notably body size (wetter substrata produce larger hatchlings), and that, in turn, these larger hatchlings are more likely to survive (Brown and Shine 2004). However, temporal shifts in water potential remain unexplored. This situation provides an ideal model system to examine whether temporal fluctuations in soil moisture (independent of mean levels) during incubation significantly affect the phenotypic traits of hatchling reptiles. Because fluctuating hydric conditions are widespread, any sensitivity to such conditions could have effects on a diverse array of reptile species from temperate-zone as well as tropical habitats. Material and Methods Study Species and Area Keelbacks (Tropidonophis mairii) are nonvenomous colubrid snakes, morphologically similar to other natricines in North America (garter snakes Thamnophis; water snakes Nerodia), Asia (Rhabdophis), and Europe (grass snakes Natrix; Madsen 1983; Fukada 1992; Weatherhead et al. 1995; Rossman et al. 1996; Luiselli et al. 1997). Keelbacks are found throughout tropical and subtropical areas of Australia and New Guinea,

Figure 1. Seasonal variation in soil water content at Fogg Dam. These values are based on sampling of soil at potential and actual nest sites during 28 mo, with samples taken at five sites each time and at three depths per site. The error bars show SEs.

usually in swampy habitats (O’Shea 1996; Cogger 2000). Adult body sizes average 47.1 cm snout-vent length (SVL) for males and 55.4 cm for females in our study population. We study the ecology of keelbacks at Fogg Dam, an artificial impoundment on the Adelaide River floodplain 60 km east of the city of Darwin, in the Northern Territory (12⬚34⬘S, 131⬚18⬘E). Our primary study area is the wall of Fogg Dam, bordered on one side by water and on the other by a periodically inundated floodplain. The area and climate are described in more detail in previous publications (Brown et al. 2001; Webb et al. 2001; Brown and Shine 2002). Methods We captured gravid female snakes by hand during nightly surveys along the dam wall. These snakes were then maintained in captivity at our nearby laboratory in individual plastic boxes (40 cm # 30 cm # 20 cm) with a water bowl and a shelter/ nesting box containing moist vermiculite (300% water by mass). They were housed indoors in a room kept between 21⬚ and 31⬚C (mean p 25.9⬚C, SD p 2.1) until they oviposited 2– 8 d later (mean p 5.3 d, SD p 2.3). Females were released at their site of capture immediately after oviposition. Eggs were weighed !12 h after they were laid in moist vermiculite. We retained nine eggs from each of six clutches for this experiment for a total of 54 eggs. Each egg was incubated at room temperature (means per clutch 23.9⬚–24.4⬚C) in a separate 135mL container (plastic vial, 46 mm high, 60 mm diameter, with a small perforation in the lid to allow air exchange) in moist vermiculite (same mass of dry vermiculite as egg mass, plus

526 G. P. Brown and R. Shine

Figure 2. Weekly changes in mass of eggs of keelback snakes exposed to three incubation treatments involving increasing, constant, or decreasing water availability through the incubation period. Vertical dashed lines indicate timing of changes between moisture regimes. The error bars show SEs.

appropriate water as described below). All eggs were kept in the same room in which the females had been housed. Three eggs from each clutch were assigned to each of three treatment groups differing in moisture conditions experienced during incubation. These treatment groups mimicked the following natural conditions: (1) early nests (wet-dry): eggs laid early in the season in wet conditions and subjected to progressively drier soil; (2) middle nests (stable): eggs laid midway through the nesting season and thus exposed to relatively constant (intermediate mean, gradual drying) moisture conditions; and (3) late nests (dry-wet): eggs laid late in the season when the soil was dry and exposed to progressively moister soil throughout the incubation period. The moisture content of vermiculite (the incubation medium) for each treatment was based on measured changes in soil moisture levels in the field over these periods (see Fig. 1). Because the relationship between water content of the substratum and water availability to eggs (as measured by their rates of water uptake) differ substantially between soil and vermiculite, we used data from calibration trials to generate moisture conditions equivalent to those experienced by keelback eggs in natural nests (Shine and Brown 2002). To establish the relative water availability to eggs incubating in soil versus vermiculite, we used data on mass increase (moisture uptake) of (1) 286 eggs incubated in vermiculite ranging

in moisture content from 6% to 600% by mass (for more information on these experiments, see Brown and Shine 2004) and (2) 91 eggs incubated in soil ranging from 3% to 15% moisture content. We then compared the mean water uptake 8 wk into incubation (i.e., shortly before hatching) of the eggs in these treatments. Eggs kept under conditions of soil moisture similar to those in the driest month (October; 2.1% water in the soil) gained as much water as eggs kept in vermiculite with 1.8% water by mass. However, this similarity in the egg’s response to percent water content in the two incubation media disappeared as water content increased, with 4.8% soil water (as in June) corresponding to 23.2% water in vermiculite, and 10.7% water by mass in the soil (as in the wettest month, March) equivalent to 277.7% water by mass in vermiculite. Using mean values of soil moisture content per month and then calculating their equivalents for water availability in vermiculite, we selected the following levels for our experimental treatments. Eggs in the wet-dry treatment, mimicking nests laid early in the season (laid in May, hatched in July), were incubated under wet conditions (70% water : vermiculite by mass) for 3 wk, then medium (30%) for 2 wk, then dry (10%) for 3 wk. Eggs in the stable treatment, mimicking nests laid in the middle of the season (laid in July, hatched in September), were incubated continuously under medium water content conditions (30% water : vermiculite). Eggs in the dry-wet treatment, mimicking nests laid late in the season (laid in November, hatched in January), were incubated under dry conditions (10% water : vermiculite) for 3 wk, then medium conditions (30%) for 2 wk, then wet conditions (70%) for 3 wk. Eggs were weighed weekly throughout incubation. Eggs were checked daily for hatching, and all hatchlings were immediately removed, measured (SVL, head length, tail length), and weighed. Hatchling muscular strength was tested by allowing them to pull seven times in quick succession against a 50-g pesola scale clipped to the tail while the forebody was restrained (for details of this method, see Shine and Brown 2002). From these data, we extracted three measures of muscular performance: maximum strength, mean strength, and consistency in performance (SD of the mean). Hatchlings were then marked and released in the field at their mother’s site of capture. These data were analyzed using two-factor mixed-model ANOVAs with incubation treatment (increasing, constant, or decreasing substratum water content) as a fixed factor and clutch identity and the interaction between clutch identity and incubation treatment as random factors, on a series of dependent variables. When ANOVA or ANCOVA indicated significant effects, we used Tukey’s honestly significant difference tests on least squares means to determine which groups differed from one another. Fixed-factor models were analysed using Proc GLM in SAS, and mixed-model analyses were carried out using Proc MIXED (SAS Institute 1999). Proc MIXED uses restricted maximum likelihood to estimate variance components of ran-

Developmental Plasticity in Snakes 527

Figure 3. Phenotypic traits (a) snout-vent length, (b) tail length, (c) head length, and (d) mass of hatchling keelback snakes exposed to three incubation treatments involving increasing, constant, or decreasing water availability through the incubation period. The error bars show SEs.

dom factors. The significance of these variance estimates is then assessed using a Wald test rather than F-tests (Steele et al. 1997). All models were initially run with interaction terms included. However, no interactions were statistically significant (all P 1 0.10), and thus we report only main effects. Results Hatching Success and Incubation Periods All eggs hatched in all three treatments. Incubation periods were all within the range of 60–70 d; all eggs within the same clutch hatched within the same day, and incubation periods did not vary significantly as a result of substratum moisture (F2, 51 p 0.000, P p 1.00). Changes in Egg Mass Eggs from all treatments were similar in size at the beginning of the experiment (F2, 46 p 0.20, P p 0.82; clutch effect Z p 1.56, P p 0.06), and all increased in mass during incubation

(Fig. 2). However, the magnitude of their mass increase (pwater uptake) differed among treatments (Fig. 2). At the time of hatching, eggs in the wet-dry (early nest) treatment were significantly lighter than eggs in either of the other groups (F2, 46 p 31.77, P ! 0.0001; clutch effect Z p 1.43, P p 0.08). Eggs in the stable (middle nest) group increased mass at an intermediate rate throughout incubation, whereas the wet-dry (early nest) group gained and then lost mass. The dry-wet (late nest) group lost mass initially but then gained water rapidly toward the end of incubation (Fig. 2).

Effects on Phenotypic Traits of Hatchlings Despite the substantial differences in egg mass at the end of incubation (Fig. 2), hatchlings in all three groups were similar in terms of body measurements (Fig. 3) and muscular strength. Nonetheless, statistical analysis detected significant differences (Table 1). The marginally nonsignificant (0.07 ! P ! 0.09 ) maternal (among-clutch) effects in these analyses presumably are partly due to among-clutch differences in initial egg mass. To

528 G. P. Brown and R. Shine Table 1: Effects of increasing, constant, or decreasing substratum water content on phenotypic traits of eggs and hatchlings of keelback snakes, Tropidonophis mairii

Trait Hatchling snout-vent length Hatchling mass Hatchling tail length Hatchling head length Maximum strength Mean strength Strength variance

Fixed Effect Incubation Treatment

Random Effect Clutch Identity

F2, 46

Wald Z

4.5 3.0 3.5 1.7 1.3 1.0 3.1

P .016 .06 .04 .19 .28 .37 .05

1.5 1.5 1.5 1.3 1.0 1.0 1.3

P .07 .07 .07 .09 .16 .16 .11

Note. Results are for a series of dependent variables from two-factor mixedmodel ANOVAs with incubation treatment (increasing, constant, or decreasing substratum water content) as a fixed factor and clutch identity as a random factor. Statistically significant results (P ! 0.05) are shown in bold.

clarify whether sources of variation unrelated to egg size were important, we repeated the analysis but with initial egg mass included as a covariate. Table 1 provides results from ANOVAs on individual variables. Hatchlings from the early nest (wet-dry) treatment had shorter bodies than did hatchlings from the late nest (dry-wet) treatment (Fig. 3). Hatchlings from the stable (middle nest) treatment had longer tails than did wet-dry (early nest) hatchlings. Including initial egg mass as a covariate in these analyses increased the effect of incubation treatment on hatchling mass to a statistically significant level (Table 2). Discussion Although our experimental treatments generated substantial (21%) differences in egg mass at hatching, they exerted only minor (albeit statistically significant) influences on the phenotypic traits of the hatchlings that emerged from those eggs. The magnitude of change in mean values as a result of our treatments was !10% for most traits (e.g., 3% for SVL, 5% for tail length, 1.6% for head length, 5% for mass, 6% for mean strength, 6% for maximum strength). In a previous study, we found that incubation at constant moisture levels of 25% versus 50% resulted in larger effects, including a 12% difference in mean hatchling SVL (Brown and Shine 2004). This comparison implies that although seasonal shifts in water content of the soil per se do modify hatchling phenotypes, the effects are so minor that they are unlikely to affect hatchling fitness. For example, field data on the relationship between offspring body size and survival rates (Brown and Shine 2004) suggest that the disparity in mean SVL generated by our treatments (approximately 0.5 cm SVL; Fig. 3) would correspond to only a

1% difference in the probability that a hatchling keelback would survive through its first year of life. Our previous studies have provided strong evidence that mean water content of the incubation medium substantially influences fitness-relevant phenotypic traits of hatchling keelbacks, with wetter substrata yielding larger, fitter offspring (Brown and Shine 2002, 2004). Thus, the amount of water uptake during embryogenesis presumably is important, and the range of moisture treatments used in this study was similar to that in our previous work. Why then did fluctuations in soil water content have so little influence on offspring phenotypes? The likely answer is that keelback eggs are very adept at obtaining moisture from the surrounding soil whenever it is available. The most telling example comes from eggs in the wet-dry (early nest) treatment, which experienced relatively dry substrata through the latter part of development at a time when water is likely to be critically important for development (Deeming 2004). Sustained incubation at these water potentials strongly reduces size (and thus viability) of hatchlings, because a significant proportion of the original yolk stores are left in the desiccated egg at hatching rather than being taken up into the neonate’s body (Brown and Shine 2004). Nonetheless, the eggs in our wet-dry treatment were able to take up water early in development (when it was available) and sequester it so effectively that the subsequently drying environment did not reduce the embryo’s water stores to a dangerous level. Similarly, eggs in the dry-wet treatment were able to tolerate restricted availability of water early in development and take advantage of more favorable conditions as incubation proceeded. This ability to buffer embryogenesis Table 2: Effects of increasing, constant, or decreasing substratum water content on phenotypic traits of eggs and hatchlings of keelback snakes, Tropidonophis mairii, after standardizing all dependent variables for initial egg mass Fixed Effect Incubation Treatment

Random Effect Clutch Identity

Trait

F2, 10

P

Wald Z

Hatchling snout-vent length Hatchling mass Hatchling tail length Hatchling head length Maximum strength Mean strength Strength variance

6.0 5.1 3.5 2.1 1.4 1.1 3.2

.005* .01 .04 .14 .25 .34 .05

1.2 1.4 1.3 1.3 .9 .9 1.2

P .11 .09 .09 .11 .20 .19 .12

Note. Results are for a series of dependent variables from two-factor mixedmodel ANCOVAs with incubation treatment (increasing, constant, or decreasing substratum water content) as a fixed factor and clutch identity as a random factor. Statistically significant results (P ! 0.05) are shown in bold. * Indicates results that are significant following Bonferroni correction for multiple tests.

Developmental Plasticity in Snakes 529 against fluctuations in ambient hydric conditions by sequestering water whenever it becomes available confers a remarkable independence of developmental trajectories from otherwise harmful variation in water availability. The major finding from our study is that as long as the eggs of keelbacks obtain access to sufficient substratum moisture at some stage during incubation, developmental trajectories can proceed with little interruption. Thus, even a brief period of wet conditions, regardless of when it occurs, may be enough to overcome earlier or later episodes of dryness. One implication of this result is that female keelbacks that lay their eggs into very moist nests (as they do; Brown and Shine 2004) may enable the egg to obtain sufficient water very early in development to withstand later fluctuations. More generally, our data show that the ability to sequester water reserves above and beyond those needed for current embryogenesis enables reptile eggs to survive periods when incubation conditions fall well outside the range required for successful long-term incubation. Thus, experimental studies that maintain eggs under constant conditions throughout incubation may substantially underestimate the range of physical conditions that eggs can tolerate for shorter periods. The same may be true for incubation temperatures but will apply to a much lesser degree because eggs are unable to store thermal reserves in the way that they can sequester moisture. Thus, a recent study that exposed lizard eggs to gradual increases or decreases in mean incubation temperature during incubation reported significant effects of this treatment on phenotypic traits of the hatchlings (Shine 2004). More generally, a survey of published studies indicates that embryonic responses to incubation temperature shift during development; for example, a temperature increase accelerates developmental rates more in early-stage embryos than in late-stage embryos (Andrews 2004). In nature, seasonal shifts in incubation conditions may well involve both moisture and temperature (and may involve variances in both parameters as well as mean values; Packard and Packard 1988). However, the embryo is much more able to buffer such variation in hydric conditions than it can in thermal conditions, and hence the phenotypic consequences of such abiotic variation may often be greater for thermal than for hydric regimes in many habitat types. Clearly, however, we will need studies on a broad range of taxa in a broad range of nest conditions before we can develop valid generalizations about the ecological significance of temporal fluctuations in the physical conditions that a reptilian egg experiences over the course of incubation.

Acknowledgments We thank C. Shilton, M. Elphick, and S. Dusty for encouragement and support, an anonymous reviewer for suggesting mixedmodel analysis, and the Australian Research Council for funding.

Literature Cited Andrews R.M. 2004. Patterns of embryonic development. Pp. 75–102 in D.C. Deeming, ed. Reptilian Incubation: Environment, Evolution and Behaviour. Nottingham University Press, Nottingham. Andrews R.M., T. Mathies, and D.A. Warner. 2000. Effect of incubation temperature on morphology, growth, and survival of juvenile Sceloporus undulatus. Herpetol Monogr 14: 420–431. Badham J.A. 1971. Albumen formation in eggs of the agamid Amphibolurus barbatus barbatus. Copeia 1971:543–545. Brown G.P. and R. Shine. 2002. Reproductive ecology of a tropical natricine snake, Tropidonophis mairii (Colubridae). J Zool (Lond) 258:63–72. ——. 2004. Maternal nest-site choice and offspring fitness in a tropical snake (Tropidonophis mairii, Colubridae). Ecology 85:1627–1634. Brown G.P., R. Shine, and T. Madsen. 2001. Responses of three sympatric snake species to tropical seasonality in northern Australia. J Trop Ecol 18:549–558. Burger J. 1989. Incubation temperature has long-term effects on behaviour of young pine snakes (Pituophis melanoleucus). Behav Ecol Sociobiol 24:201–207. Cogger H. 2000. Reptiles and Amphibians of Australia. 6th ed. Reed New Holland, Sydney. Deeming D.C. 2004. Post-hatching phenotypic effects of incubation on reptiles. Pp. 229–251 in D.C. Deeming, ed. Reptilian Incubation: Environment, Evolution and Behaviour. Nottingham University Press, Nottingham. Flatt T., R. Shine, P.A. Borges-Landaez, and S.J. Downes. 2001. Phenotypic variation in an oviparous montane lizard (Bassiana duperreyi): the effects of thermal and hydric incubation environments. Biol J Linn Soc 74:339–350. Fukada H. 1992. Snake Life History in Kyoto. Impact Shuppankai, Tokyo. Gutzke W.H.N. and G.C. Packard. 1987. Influence of the hydric and thermal environments on eggs and hatchlings of bull snakes Pituophis melanoleucus. Physiol Zool 60:9–17. Hildebrand M. 1995. Analysis of Vertebrate Structure. 4th ed. Wiley, New York. Ji X. and F. Bran˜a. 1999. The influence of thermal and hydric environments on embryonic use of energy and nutrients, and hatchling traits, in the wall lizards (Podarcis muralis). Comp Biochem Physiol A 124:205–213. Lillywhite H.B. and R.A. Ackerman. 1984. Hydrostatic pressure, shell compliance and permeability to water vapor in flexibleshelled eggs of the colubrid snake Elaphe obsoleta. Pp. 121– 136 in R.S. Seymour, ed. Respiration and Metabolism of Embryonic Vertebrates. W. Junk, Dordrecht. Luiselli L., M. Capula, and R. Shine. 1997. Food habits, growth rates, and reproductive biology of grass snakes, Natrix natrix (Colubridae) in the Italian Alps. J Zool (Lond) 241:371–380.

530 G. P. Brown and R. Shine Madsen T. 1983. Growth rates, maturation and sexual size dimorphism in a population of grass snakes, Natrix natrix, in southern Sweden. Oikos 40:277–282. Muth A. 1980. Physiological ecology of desert iguana (Dipsosaurus dorsalis) eggs: temperature and water relations. Ecology 61:1335–1343. O’Shea M. 1996. A Guide to the Snakes of Papua New Guinea. Independent Publishing, Port Moresby, Papua New Guinea. Packard G.C. 1991. Physiological and ecological importance of water to embryos of oviparous reptiles. Pp. 213–228 in D.C. Deeming and M.W.J. Ferguson, eds. Egg Incubation: Its Effects on Embryonic Development in Birds and Reptiles. Cambridge University Press, Cambridge. Packard G.C. and M.J. Packard. 1988. The physiological ecology of reptilian eggs and embryos. Pp. 523–605 in C. Gans and R.B. Huey, eds. Biology of the Reptilia. Vol. 16. Liss, New York. Packard G.C., C.R. Tracy, and J.J. Roth. 1977. The physiological ecology of reptilian eggs and embryos, and the evolution of viviparity within the class Reptilia. Biol Rev 52:71–105. Romer A.S. 1966. Vertebrate Paleontology. 3rd ed. University of Chicago Press, Chicago. Rossman D.A., N.B. Ford, and R.A. Seigel. 1996. The Garter Snakes: Evolution and Ecology. University of Oklahoma Press, Norman.

SAS Institute. 1999. Statistical Analysis Software. Version 8. SAS Institute, Cary, NC. Shine R. 2002. Eggs in autumn: responses to declining incubation temperatures by the eggs of montane lizards. Biol J Linn Soc 76:71–77. ——. 2004. Seasonal shifts in nest temperature can modify the phenotypes of hatchling lizards, regardless of overall mean incubation temperature. Funct Ecol 18:43–49. Shine R. and G.P. Brown. 2002. Effects of seasonally varying hydric conditions on hatchling phenotypes of keelback snakes (Tropidonophis mairii, Colubridae) from the Australian wet-dry tropics. Biol J Linn Soc 76:339–347. Steele R.G.D., J.H. Torrie, and D.A. Dickey. 1997. Principles and Procedures of Statistics. 3rd ed. McGraw-Hill, New York. Weatherhead P.J., F.E. Barry, G.P. Brown, and M.R.L. Forbes. 1995. Sex ratios, mating behavior and sexual size dimorphism of the northern water snake, Nerodia sipedon. Behav Ecol Sociobiol 36:301–311. Webb J.K., G.P. Brown, and R. Shine. 2001. Body size, locomotor speed and antipredator behaviour in a tropical snake (Tropidonophis mairii, Colubridae): the influence of incubation environments and genetic factors. Funct Ecol 15:561– 568.