Modelling exposure to selected temperature during pregnancy: the limitations of squamate viviparity in a cool-climate environment

Similar documents
EQUAL THERMAL OPPORTUNITY DOES NOT RESULT IN EQUAL GESTATION LENGTH IN A COOL-CLIMATE SKINK AND GECKO

Consequences of Extended Egg Retention in the Eastern Fence Lizard (Sceloporus undulatus)

posted online on 19 July 2016 as doi: /jeb

Accessory Publication

HERPETOLOGICA VOL. 68 JUNE 2012 NO. 2 LIN SCHWARZKOPF 1,3 AND ROBIN M. ANDREWS 2

School of Zoology, University of Tasmania, PO Box 252C-05, Tas, 7001, Australia

Effects of nest temperature and moisture on phenotypic traits of hatchling snakes (Tropidonophis mairii, Colubridae) from tropical Australia

phenotypes of hatchling lizards, regardless of overall mean incubation temperature

Evolution of viviparity in warm-climate lizards: an experimental test of the maternal manipulation hypothesis

Seasonal Shifts in Reproductive Investment of Female Northern Grass Lizards ( Takydromus septentrionalis

Incubation temperature and phenotypic traits of Sceloporus undulatus: implications for the northern limits of distribution

DOES VIVIPARITY EVOLVE IN COLD CLIMATE REPTILES BECAUSE PREGNANT FEMALES MAINTAIN STABLE (NOT HIGH) BODY TEMPERATURES?

Maternal and environmental influences on reproductive success of a viviparous grassland lizard

MATERNAL NEST-SITE CHOICE AND OFFSPRING FITNESS IN A TROPICAL SNAKE (TROPIDONOPHIS MAIRII, COLUBRIDAE)

A comparison of placental tissue in the skinks Eulamprus tympanum and E. quoyii. Yates, Lauren A.

ARTICLE IN PRESS. Zoology 113 (2010) 33 38

Is Parental Care the Key to Understanding Endothermy in Birds and Mammals?

PHYSIOLOGICAL AND ECOLOGICAL CONSTRAINTS ON THE EVOLUTION OF VIVIPARITY IN SCELOPORINE LIZARDS. Scott L. Parker

Social and Thermal Cues Influence Nest-site Selection in a Nocturnal Gecko, Oedura lesueurii

Effects of Incubation Temperature on Growth and Performance of the Veiled Chameleon (Chamaeleo calyptratus)

Geographical differences in maternal basking behaviour and offspring growth rate in a climatically widespread viviparous reptile

Maternal Thermal Effects on Female Reproduction and Hatchling Phenotype in the Chinese Skink (Plestiodon chinensis)

COMPARING BODY CONDITION ESTIMATES OF ZOO BROTHER S ISLAND TUATARA (SPHENODON GUNTHERI) TO THAT OF THE WILD, A CLINICAL CASE

Phenotypic Effects of Thermal Mean and Fluctuations on Embryonic Development and Hatchling Traits in a Lacertid Lizard, Takydromus septentrionalis

Who Cares? The Evolution of Parental Care in Squamate Reptiles. Ben Halliwell Geoffrey While, Tobias Uller

Geographic variation in lizard phenotypes: importance of the incubation environment

THE concept that reptiles have preferred

Influence of Incubation Temperature on Morphology, Locomotor Performance, and Early Growth of Hatchling Wall Lizards (Podarcis muralis)

Sheikh Muhammad Abdur Rashid Population ecology and management of Water Monitors, Varanus salvator (Laurenti 1768) at Sungei Buloh Wetland Reserve,

Offspring performance and the adaptive benefits of. prolonged pregnancy: experimental tests in a viviparous lizard

Natural history of Xenosaurus phalaroanthereon (Squamata, Xenosauridae), a Knob-scaled Lizard from Oaxaca, Mexico

Habitats and Field Methods. Friday May 12th 2017

Impact of colour polymorphism and thermal conditions on thermoregulation, reproductive success, and development in Vipera aspis

Do TSD, sex ratios, and nest characteristics influence the vulnerability of tuatara to global warming?

FEMALE PHENOTYPE, LIFE HISTORY, AND REPRODUCTIVE SUCCESS IN FREE-RANGING SNAKES (TROPIDONOPHIS MAIRII)

Is cool egg incubation temperature a limiting factor for the translocation of tuatara to southern New Zealand?

Reproductive modes in lizards: measuring fitness. consequences of the duration of uterine retention of eggs

A Population Analysis of the Common Wall Lizard Podarcis muralis in Southwestern France

University of Canberra. This thesis is available in print format from the University of Canberra Library.

Temperature Relationships of Two Oklahoma Lizards

Embryonic responses to variation in oviductal oxygen in the lizard Sceloporus undulatus from New Jersey and South Carolina, USA

Variation of Chicken Embryo Development by Temperature Influence. Anna Morgan Miller. Rockdale Magnet School for Science and Technology

reproductive life History and the effects of sex and season on morphology in CRoTALus oreganus (northern PaCifiC RATTLESNAKES)

When a species can t stand the heat

Wen SHEN 1, Jianchi PEI 2, Longhui LIN 3* and Xiang JI Introduction

Effect of Tail Loss on Sprint Speed and Growth in Newborn Skinks, Niveoscincus metallicus

Thermal adaptation of maternal and embryonic phenotypes in a geographically widespread ectotherm

Thermal and fitness-related consequences of nest location in Painted Turtles (Chrysemys picta)

CHOOSING YOUR REPTILE LIGHTING AND HEATING

Natural history of Hoplodactylus stephensi (Reptilia: Gekkonidae) on Stephens Island, Cook Strait, New Zealand

Motuora island reptile monitoring report for common & Pacific gecko 2017

Nest-site selection in Eastern hognose snakes (Heterodon platirhinos) Casey Peet-Paré

NOTES ON THE ECOLOGY AND NATURAL HISTORY OF TWO SPECIES OF EGERNIA (SCINCIDAE) IN WESTERN AUSTRALIA

LIZARDS OBSERVED DURING A VISIT TO THE CAVALLI ISLANDS, DECEMBER 1978 TO JANUARY by R.A. Hitchmough SUMMARY

Station 1 1. (3 points) Identification: Station 2 6. (3 points) Identification:

When a species can t stand the heat

Climate affects embryonic development in a viviparous snake, Vipera aspis

Reproductive physiology and eggs

Survival of captive-bred skinks following reintroduction to the wild is not explained by variation in speed or body condition index

RESEARCH ARTICLE Potentially adaptive effects of maternal nutrition during gestation on offspring phenotype of a viviparous reptile

Motuora island reptile monitoring report for common & Pacific gecko 2016

Sunny side up: lethally high, not low, nest temperatures may prevent oviparous reptiles from reproducing at high elevations

A NOVEL PATTERN OF EMBRYONIC NUTRITION IN A VIVIPAROUS REPTILE

ABSTRACT THE IMPORTANCE OF PRE- AND POSTNATAL THERMAL CONDITIONS IN DETERMINING GROWTH TRAJECTORIES IN THREE VIVIPAROUS GRASSLAND SNAKES

Sprint speed capacity of two alpine skink species, Eulamprus kosciuskoi and Pseudemoia entrecasteauxii

JEZ Part A: Comparative Experimental Biology. An experimental test of the effects of fluctuating incubation temperatures on hatchling phenotype

Short-term Water Potential Fluctuations and Eggs of the Red-eared Slider Turtle (Trachemys scripta elegans)

Egg environments have large effects on embryonic development, but have minimal consequences for hatchling phenotypes in an invasive lizard

BEHAVIORAL THERMOREGULATION OF THE TUATARA, SPHENODON PUNCTATUS, UNDER HYDRIC AND DIGESTIVE CONSTRAINTS

Bio4009 : Projet de recherche/research project

Weaver Dunes, Minnesota

Do operational sex ratios influence sex allocation in viviparous lizards with temperature-dependent sex determination?

INDIVIDUAL IDENTIFICATION OF GREEN TURTLE (CHELONIA MYDAS) HATCHLINGS

SELECTED BODY TEMPERATURE AND THERMOREGULATORY BEHAVIOR IN THE SIT-AND-WAIT FORAGING LIZARD PSEUDOCORDYLUS MELANOTUS MELANOTUS

Climate change impacts on fitness depend on nesting habitat in lizards

Offspring size number strategies: experimental manipulation of offspring size in a viviparous lizard (Lacerta vivipara)

Cold climates and the evolution of viviparity. produce poor-quality offspring in the lizard, in reptiles: cold incubation temperatures

Plestiodon (=Eumeces) fasciatus Family Scincidae

Thermal constraints on embryonic development as a proximate cause for. elevational range limits in two Mediterranean lacertid lizards

J.-F. LE GALLIARD, M. LE BRIS and J. CLOBERT

Rubber Boas in Radium Hot Springs: Habitat, Inventory, and Management Strategies

Habitats and Field Techniques

Amniote Relationships. Reptilian Ancestor. Reptilia. Mesosuarus freshwater dwelling reptile

Lizard malaria: cost to vertebrate host's reproductive success

Gecko Monitoring FIELD GUIDE for Motuihe Island

PHENOTYPES AND SURVIVAL OF HATCHLING LIZARDS. Daniel A. Warner. MASTER OF SCIENCE in Biology

Impact of colour polymorphism in free ranging asp vipers

SEXUAL DIMORPHISM IN BODY SHAPE WITHOUT SEXUAL DIMORPHISM IN BODY SIZE IN WATER SKINKS (EULAMPRUS QUOYII)

Testing the Persistence of Phenotypic Plasticity After Incubation in the Western Fence Lizard, Sceloporus Occidentalis

Species Fact Sheets. Order: Gruiformes Family: Cariamidae Scientific Name: Cariama cristata Common Name: Red-legged seriema

Phenotypic Plasticity in Embryonic Development of Reptiles: Recent Research and Research Opportunities in China

DECREASED SPRINT SPEED AS A COST OF REPRODUCTION IN THE LIZARD SCELOPORUS OCCIDENTALS: VARIATION AMONG POPULATIONS

Maturity and Other Reproductive Traits of the Kanahebi Lizard Takydromus tachydromoides (Sauria, Lacertidae) in Mito

A Comparison of morphological differences between Gymnophthalmus spp. in Dominica, West Indies

Interpopulational variation in costs of reproduction related to pregnancy in a viviparous lizard

Like mother, like daughter: inheritance of nest-site

Reptilian Physiology

ACTIVITY #6: TODAY S PICNIC SPECIALS ARE

Analysis of Sampling Technique Used to Investigate Matching of Dorsal Coloration of Pacific Tree Frogs Hyla regilla with Substrate Color

Latent Effects of Egg Incubation Temperature on Growth in the Lizard Anolis carolinensis

Transcription:

Biological Journal of the Linnean Society, 2009, 96, 541 552. With 6 figures Modelling exposure to selected temperature during pregnancy: the limitations of squamate viviparity in a cool-climate environment JONATHON R. HARE*, KARINA M. HOLMES, JACKIE L. WILSON and ALISON CREE Department of Zoology, University of Otago, Box 56, Dunedin, 9054, New Zealand Received 31 March 2008; accepted for publication 11 August 2008 Behavioural thermoregulation is important for the success of cool-climate lizards, and a basis of the cold-climate hypothesis for the evolution of viviparity in squamate reptiles. The temperature (T sel) selected by pregnant females in a thermal gradient is considered to be optimal for embryonic development; however, exposure to T sel throughout pregnancy has been difficult to estimate in small-bodied lizards as temperature-sensitive telemetry is impractical. In addition, the value of maternal thermophily during pregnancy is controversial: some studies have shown elevated T sel, whereas others have found lowered T sel or no change during pregnancy. We estimated indirectly the overall exposure to T sel during the 4 5 months of pregnancy of the cool-climate, sub-alpine species Oligosoma maccanni (McCann s skink, 3 6 g) from southern New Zealand. The thermal environment available to skinks was modelled using temperature loggers inside validated copper models in basking and retreat sites. Pregnant skinks were able to achieve mean T sel (28.9 C) in the field very infrequently (4 15% of each month during the final 4 months of pregnancy). In field thermoregulatory studies, pregnant females did not bask more frequently and did not show altered field body temperature compared with non-pregnant adults, suggesting that all skinks (whether pregnant or not) thermoregulate maximally whenever conditions allow. Further research on cool-climate lizards should address the significance for offspring phenotypes of low and variable exposure to T sel during pregnancy, as well as the significance of temperatures for embryos in maternal bodies (viviparity) versus nest sites (oviparity) arising from differences in maternal body size. 2009 The Linnean Society of London, Biological Journal of the Linnean Society, 2009, 96, 541 552. ADDITIONAL KEYWORDS: copper model field body temperature lizard maternal thermophily Scincidae temperature modelling thermoregulation. INTRODUCTION Behavioural thermoregulation is central to the success of cool-climate lizards, including the development of embryos in viviparous species. The coldclimate hypothesis states that viviparity evolved *Corresponding author. Current address: Department of Fisheries and Oceans Canada, University of Manitoba, 501 University Crescent, Winnipeg, Canada. E-mail: jonathon.hare@dfo-mpo.gc.ca Current address: Department of Conservation, Otago Conservancy, Box 5244, Dunedin, New Zealand. Current address: Department of Biochemistry, University of Otago, Box 56, Dunedin, New Zealand. in squamate reptiles because it enables embryonic development in utero at more thermally suitable (i.e. warmer or more stable) temperatures than those available in potential nest sites (Andrews, 2000; Shine, 2004a; Ji et al., 2007). Some advantage to embryos may result from retention inside pregnant females that thermoregulate normally (as when non-pregnant). An additional thermal advantage may exist in species that exhibit maternal thermophily (for a review, see Shine, 2006). In these species, females alter their exposure to warmth during pregnancy in ways assumed to be beneficial for embryos: for example, by selecting higher or more stable body temperatures on a thermal gradient (selected 541

542 J. R. HARE ET AL. temperature, T sel; Rock, Andrews & Cree, 2000) or basking for longer periods (Schwarzkopf & Shine, 1991; Blázquez, 1995). Much circumstantial evidence exists in support of the cold-climate hypothesis in reptiles; for instance, viviparous species form a larger proportion of species at higher altitudes or latitudes, and recent origins of viviparity are associated with recent invasions of higher latitudes and altitudes (Hodges, 2004; Shine, 2004a). To experimentally test the cold-climate hypothesis, accurate comparisons between temperatures experienced by viviparous reptiles throughout pregnancy and in potential nest sites are required. Direct measurements of maternal body temperature (T b) are feasible in large-bodied species using temperaturesensitive radiotelemetry (Charland, 1995) or intracoelomically implanted temperature loggers (Taylor, DeNardo & Malawy, 2004). Small, externally attached devices are also available to log surface temperatures in medium-sized species (Robert & Thompson, 2003). These devices, however, are still relatively large and unsuitable for use with very small (< 10 g) lizards with extended pregnancies, especially crevice-dwelling species. For such species, it is more appropriate to use models to predict T b. In this article, we demonstrate a method to estimate T b experienced by a small, heliothermic and crevice-dwelling lizard throughout pregnancy, including the extent to which T sel is achieved in a thermally limiting environment. McCann s skink (Oligosoma maccanni, Patterson & Daugherty, 1990) is a small (3 6 g), non-threatened skink endemic to the South Island of New Zealand, with a pregnancy lasting for 4 5 months over the summer (Holmes & Cree, 2006). We used validated copper models to estimate the operative temperature of live skinks in basking positions. Operative temperature (T e) is the body temperature of an animal in equilibrium with its environment in the absence of metabolic heating or evaporative cooling (Dzialowski, 2005), and copper models are a recognized means for estimating T e in heliothermic lizards (Vitt & Sartorius, 1999; Shine & Kearney, 2001; Dzialowski, 2005). Validation is necessary because models may heat and cool differently from live lizards (Dzialowski, 2005). Retreat-site temperatures were measured by placing data loggers under rocks, representing T b of O. maccanni overnight and in inclement weather. We determined T sel for pregnant skinks in the laboratory, made field observations to determine the effect of weather conditions on basking, and tested whether pregnant skinks were more likely than non-pregnant skinks to be seen basking or to attain higher T b when basking. We used all of these data sources to estimate the overall exposure of pregnant females to temperatures assumed to be optimal for, or supportive of, embryonic development. We hypothesized that copper models could be constructed to provide an accurate measure of field T b, that pregnant skinks would bask more frequently than non-pregnant skinks and achieve higher mean field T b, and that T sel of pregnant skinks is achieved only sporadically in the field. MATERIAL AND METHODS STUDY SITE AND SPECIES Fieldwork occurred at Cloverdowns Farm at Macraes Flat, Otago (45 28 S, 170 28 E, 500 700 m above sea-level), a region of rolling hills and gullies. Large tors of fissured schist scatter the landscape and are used as basking and retreat sites by O. maccanni. Native snow tussock (Chionochloa rigida) and exotic pasture grasses dominate the surrounding vegetation. Two other species of lizard (the common skink O. nigriplantare polychroma and the common gecko Hoplodactylus maculatus) occur at the site. The weather is highly variable, with unpredictable rain, snow and fog. The air temperature (monthly mean) ranges from 15 C in the summer (January February) to 3 C in the winter (Cree & Guillette, 1995). Females reproduce annually, ovulating during late September to early October and giving birth during late January to early February to a clutch averaging 2.8 ± 0.2 neonates (range 1 6; Holmes & Cree, 2006). Mature males, identifiable by the presence of hemipenes, overlap in snout-to-vent length (SVL) with females, but are significantly shorter on average. Mean SVLs [±standard error (SE)] in the present study were as follows: pregnant females, 60.9 ± 0.5 mm (range, 50 74 mm), N = 117; males, 57.2 ± 0.6 mm (range, 50 67 mm), N = 49 (two-tailed t-test for samples with equal variances, t = 3.611, d.f. = 164, P < 0.001). VALIDATION OF COPPER MODELS Copper models were made from 120 mm lengths of 15 mm diameter copper pipe, flattened to a width of 18 mm. One end was crimped closed, and the other end, with a temperature probe (Stowaway XTI, Onset, USA) inserted, was sealed with electrical tape. Temperature probes were suspended within the models by a plastic collar, such that they did not touch the sides. Models were painted greyish-brown ( Café Royale, Resene, Wellington, New Zealand), similar to the dorsum of live O. maccanni. These models were somewhat larger and had a less complex colour pattern than live O. maccanni; however, modest differences in size and colour do not have a large effect on temperatures reached by copper models within a biologically relevant range (for example, < 5% difference in the time spent at 20 C

COOL CLIMATES AND VIVIPARITY IN LIZARDS 543 in the experiments of Shine & Kearney, 2001), and validation confirmed that our models accurately predicted T b of O. maccanni in the same position (see Results ). Copper models were validated both indoors against a live skink and outdoors against a dead skink (adult females from a laboratory colony of O. maccanni). A live skink was used only in the indoor validation, where the cage confines, overhead spot lamp and lack of wind kept the skink reliably basking alongside the model. The copper model with temperature probe inside (recording the temperature every minute) was placed on a terracotta saucer (diameter, 17 cm) in a plastic cage beneath a 60 W heat lamp set to give a spot temperature of 28 C. Both model and live skink began the experiment at the ambient temperature (T a) of 14 C. After the heat lamp was turned on, the skink was placed on the saucer alongside the model; it voluntarily remained there for the duration of the experiment. The body temperature (T b) over the lower abdomen was recorded by an infrared thermometer (Raytek, Raynger, ST80 ProPlus, Santa Cruz, CA, USA) when the skink first went onto the saucer, and then intermittently until further measurements showed no increase in T b. After 105 min, the heat lamp was turned off, and T b was recorded intermittently until it reached ambient temperature. This experiment was repeated with a substrate spot temperature of 22 C. Separate tests on a thermal gradient showed that the infrared thermometer produced comparable measurements for T b with a cloacal thermocouple in this species (P = 0.173; Hare, Whitworth & Cree, 2007). For the outdoor validation, temperature loggers were inserted into a copper model glued by epoxy to a schist rock, and into the cloaca of a skink euthanized by halothane overdose less than 1 h previously. The skink was attached to a schist rock with electrical tape over two legs, preventing the wind from altering its position. Both the model and skink, oriented north, were placed on an outdoor patio. Loggers recorded the temperature every minute from 10:15 to 17:45 h, a period that started in the shade, included a period of sun and ended in the shade. FIELD MICROHABITAT TEMPERATURES IN BASKING POSITIONS AND RETREAT SITES Temperature loggers were placed at the field site to record the hourly temperature from 16 October 2004 to 23 February 2005. The external probes of four loggers were placed inside exposed copper models in order to estimate T e available to basking skinks. Each copper model, glued with epoxy to a flat schist rock, was placed where a pregnant female had been seen basking, and was oriented pointing north. Copper models were thus placed to maximize solar exposure during the early morning and late afternoon, as McCann s skinks appear to do. To estimate temperatures in retreat sites, the external probes of four more temperature loggers were placed under rocks where pregnant females had been found. FIELD THERMOREGULATORY BEHAVIOUR OF PREGNANT AND NON-PREGNANT SKINKS Measurements of field T b and thermoregulatory behaviour were compared between pregnant and nonpregnant skinks [mostly males, plus a few sub-adult (46 49 mm SVL) or spent females]. Data were pooled across several years spanning the duration of pregnancy (26 September 2002 to 25 January 2003; 8 14 October 2004; 11 14 November 2005; 19 January to 4 February 1999; 24 January to 13 February 1998), with sampling days biased towards dry, sunny conditions offering the potential for basking. We searched for skinks between 09:00 and 17:00 h, first by looking for those that were basking and then by turning over rocks and examining crevices. In most cases, the micro-location of skinks (basking versus under rock), sun conditions (sun, intermittent sun or cloud) and T a (measured by thermocouple at 1 m height in shade) were recorded; in October 2005, the wind level (none, moderate or strong, assessed qualitatively by the same observer throughout) was also recorded. Skinks were captured by hand or noose. On capture, T b was measured immediately (within 10 s) to the nearest 0.1 C, using a cloacal thermocouple (Digi-Sense model 8528-40, Cole-Parmer, Niles, IL, USA) calibrated to within 0.4 C against a reference thermometer (SAMA-CT40, Eveready Thermometer Co., West Patterson, NJ, USA). SVL was measured with a ruler to the nearest 1 mm. The abdomen of each skink lacking hemipenes was gently palpated to distinguish pregnant from non-reproductive females (Holmes & Cree, 2006). Skinks chased for more than 1 min before capture were excluded from the calculation of the mean T b. Samples for pregnant females between 26 September and 12 October may have included some in pre-ovulatory condition, as preovulatory and early-pregnant conditions can be hard to distinguish by palpation (Holmes & Cree, 2006); indeed, some skinks were palpated as pre-ovulatory on one side (i.e. firm, spherical structures clustered in inferred ovaries), but pregnant on the other (i.e. soft, ovoid structures lined up in inferred oviducts). Thus, ovulation was occurring during late September to early October, but by mid-october virtually all mature-sized females were pregnant. All skinks were released at the field site within a few minutes of capture, apart from a subset of pregnant females collected for laboratory studies (Holmes & Cree,

544 J. R. HARE ET AL. 2006), including the determination of T sel in early pregnancy, as described below. SELECTED TEMPERATURE OF PREGNANT SKINKS Nine pregnant O. maccanni collected between 8 and 14 October 2004 were held individually in cloth bags in the shade for 5 h, and then transported by vehicle (c. 2 h trip) to the University of Otago. They were immediately placed on a thermal gradient, within individual lanes (1.0 m 85 mm), where each animal was free to move and select a temperature (Rock et al., 2000). An angled piece of hardboard ran the length of each runway to provide cover. The gradient was covered by plexiglass (Perspex ), and room windows provided a natural photoperiod (approximately 14 h light, 10 h dark). The metal baseplate of the gradient was covered with clean brown paper; one end was kept at 16 C by copper pipes carrying refrigerant, and the other at 34 C by an electrical heating strip shielded by styrofoam. After overnight equilibration, T sel was determined during the morning (10:00 h), afternoon (15:00 h) and night (22:00 h) using a cloacal thermocouple [paired measurements made with an infrared thermometer are reported by Hare et al. (2007) to validate the latter method]. A red-light headlamp was used for measurements at night. FREQUENCY WITH WHICH T SEL CAN BE ACHIEVED IN THE FIELD Data collected in the previous sections were used to model the frequency and duration with which pregnant females can achieve T sel in their natural habitat. Temperatures from the copper models in basking positions were used to calculate the proportion of hours per month, and days per month, that T sel could be achieved in the field. A frequency distribution of estimated skink T b was constructed for each month by combining temperature data from copper models and under-rock positions. We assumed: (1) that copper models or under-rock probes accurately modelled T b of skinks in each location; (2) that skinks would emerge to bask when the temperature of copper models exceeded the temperature of retreat sites; (3) that once a basking skink had achieved T sel, it would remain at T sel for as long as possible by shuttling between basking and shaded positions (unlike copper models, which continued to increase in temperature); and (4) that skinks would withdraw to retreat sites when these offered warmer temperatures than copper models (e.g. overnight). Observations supporting these assumptions are addressed in Results and Discussion. STATISTICAL ANALYSES For parametric tests, Kolmogorov Smirnov tests were used to confirm a normal distribution. For validations of copper models, regression analyses were used to determine how well the models predicted skink T b (in the outdoor test, the data set was restricted to 50 200 min after recording began in order to meet the assumption of normality). Field microhabitat temperatures recorded each hour by data loggers were averaged across days in each month for each logger, and then analysed with a repeated measures analysis of variance (ANOVA). Position (exposed versus under rock) and month were specified as factors and time of day was the repeated measure. Mauchly s assumption of sphericity was not met (P < 0.001), so the lower bound (most conservative) epsilon correction was used to identify the significance. For skinks captured in October 2004, we examined the effect of wind level on: (1) the proportion of skinks seen basking using a chi-squared test, and (2) the magnitude of difference between T b and T a on sunny days using a one-way ANOVA followed by post hoc (Tukey) tests (family error rate of 0.05). Over the entire course of pregnancy (late September until mid- February), the proportion of skinks seen basking was compared between pregnant and non-pregnant skinks using a chi-squared test, and the mean T b (unadjusted for variation in weather conditions) was compared using a t-test after confirming equal variances. In addition, in a subset of data for sunny days when basking was possible, T b was adjusted for variation in log T a before comparing between pregnant and non-pregnant skinks [analysis of covariance (ANCOVA), after confirming equality of error variances with Levene s test]. Variation in T sel of pregnant skinks with time of day was examined using repeated measures ANOVA (Mauchly s assumption of sphericity was met). Subsequently, the average T b from 10:00 and 15:00 h was considered as daytime T sel for each skink, and the mean and central 50% bounds were calculated for all skinks. Data from temperature loggers were examined to determine the proportion of time each month that pregnant skinks could attain their mean daytime T sel. The proportion of days in which T sel was reached for at least one hourly recording was averaged across loggers and then compared between months using a chi-squared test. The total proportion of time (number of hours per month) in which mean T sel was reached was also compared between months using a chisquared test. Analyses were performed using Minitab 14 (Minitab Inc.), SPSS11.0.4 (for Mac) or SPSS11.5 (SPSS Inc.). Results are reported as means ± SE, and significance was assumed when P < 0.05.

COOL CLIMATES AND VIVIPARITY IN LIZARDS 545 RESULTS VALIDATION OF COPPER MODELS Indoors on a substrate of 28 C, the skink basked alongside the model and closely resembled it in heating and cooling profiles (Fig. 1). The mean temperature difference between the skink (T skink) and the model (T model) was 0.36 ± 0.15 C, and the regression between the two was significant (r 2 = 0.973, F 1,39 = 1386.803, P < 0.001; Fig. 2A). Results with a substrate temperature of 22 C were also significant, although slightly more variable (r 2 = 0.847, F 1,24 = Figure 1. Heating and cooling profiles for a copper model (T model) and female McCann s skink (Oligosoma maccanni, T skink). For the indoor validation (squares), a heat lamp provided a substrate temperature of 28 C (on a terracotta saucer). Maximum T b of the live skink and model was reached by 105 minutes; the heat lamp was then turned off. For the outdoor validation (circles), a freshly-euthanised skink and copper model were placed on a schist rock (data for the final 190 min at 12 15 C are not shown). T skink ( C) y R P x 133.134, P < 0.001; data not presented), probably reflecting the more variable behaviour of the skink (it sometimes climbed on the model). Outdoors, the copper model closely matched the dead skink over a temperature range of c. 15 31 C (Fig. 1), with a significant linear relationship between 50 and 200 min where the data were normally distributed (r 2 = 0.984, F 1,149 = 8885.439, P < 0.001; Fig. 2B). The mean difference between T skink and T model over the entire test was 0.04 ± 0.63 C, and 98.9% of paired measurements were within 1.0 C. FIELD MICROHABITAT TEMPERATURES IN BASKING POSITIONS AND RETREAT SITES One under-rock logger failed completely and, owing to downloading difficulties, data were incomplete between 15 December and 9 January for six loggers and between 28 December and 9 January for one logger. Temperature profiles for the remaining period between mid-october and late February revealed a cool thermal environment overall for skinks (Fig. 3). The time of day had a significant effect on temperature (F 1,25 = 498.423, P < 0.001), and this effect varied between logger positions (time position, F 1,25 = 94.575, P < 0.001). As expected, exposed copper models were generally warmer by day (c. 08:00 17:00 h) and cooler by night than retreat sites under rocks. The effect of time of day did not vary significantly between months (time month, F 4,25 = 2.330, P = 0.084), nor was there a significant interaction between time of day, position and month (F 4,25 = 0.710, P = 0.593). December 2004 was unexpectedly cool (Fig. 3C): the highest mean daily temperature (at 14:00 h, in exposed copper models) reached only 20.3 C (cf. 24.1 24.4 C in October November and 29.8 30.6 C in January February). Across sampling months, under-rock loggers reached mean daily T skink ( C) y R P x T model ( C) T model ( C) Figure 2. Regression between copper model temperature (T model, x) and skink body temperature (T skink, y) using data from Figure 1 for Oligosoma maccanni. A: Indoor validation with a live skink (substrate temperature 28 C). B: Outdoor validation with a dead skink (data for minutes 50 200).

546 J. R. HARE ET AL. N N Figure 3. Field microhabitat temperatures (mean ± SE) recorded by data loggers in copper models exposed in basking positions, and in retreat sites under rocks, in habitat of the skink Oligosoma maccanni. Measurements spanned the pregnancy season (mid-october 2004 late February 2005). Copper models were placed in locations where pregnant skinks had been seen basking and were oriented north. Under-rock loggers were placed in retreat sites where skinks had been caught. Hourly data in each month (NZST) were first averaged across the number of days indicated for individual loggers; mean data were then averaged within each month. Temperatures generally warmed as expected over the austral spring-summer, except that December 2004 was unexpectedly cool. maxima of 14 22 C during the early afternoon, and mean daily minima of 0 13 C overnight. FIELD THERMOREGULATORY BEHAVIOUR OF PREGNANT AND NON-PREGNANT SKINKS During spring (October 2004), skinks basked only on sunny days and were more often seen basking when wind levels were moderate or low (c 2 = 8.8, d.f. = 2, P = 0.016). The elevation of T b above T a was strongly affected by wind (e.g. mean difference of 11.9 ± 0.9 C in no wind vs. 3.1 ± 0.7 C in strong wind; F 2,62 = 43.1, P < 0.001). Among skinks sampled across the course of pregnancy (late September until mid-february), the proportion seen basking did not differ between pregnant skinks (34.5%, N = 87) and non-pregnant skinks (25.0%, N = 56; c 2 = 1.438, d.f. = 1, P = 0.230). There was also no difference in mean T b, nor variance in T b, between pregnant skinks (22.9 ± 0.6 C, N = 117) and non-pregnant skinks (21.5 ± 0.7 C, N = 89; t = 1.564, d.f. = 204, P = 0.119; Fig. 4A). Daytime body temperatures reached 32.2 C (pregnant) and 29.5 C (non-pregnant) for those seen basking, but ranged between 7.2 32.7 C (pregnant) and 7.1 31.7 C (non-pregnant) for those found under rocks. On sunny days when basking was possible (October to February), log T a had a significant and positive effect on T b (r 2 = 0.453, F 1,75 = 62.054, P < 0.001). Using ANCOVA to adjust for variation in T a, no difference was observed between pregnant and non-pregnant skinks in either slopes or intercepts (P 0.250; Fig. 4B). Differences between T b and T a were up to 17.3 C in pregnant skinks and 16.1 C in non-pregnant skinks. The lowest T a at which basking was seen was 10.3 C, but differences of up to 12.8 C between T b and T a at the lowest T a sampled (7.7 C) suggest that basking also occurred at this air temperature. SELECTED BODY TEMPERATURE DURING PREGNANCY There was no effect of time of day on T sel in early pregnancy (P = 0.463; Fig. 5). Mean daytime T sel (average of values at 10:00 and 15:00 h) was 28.9 ± 0.7 C, and the central 50% bound lay between 26.4 and 30.2 C.

COOL CLIMATES AND VIVIPARITY IN LIZARDS 547 T T Figure 4. Field body temperatures (T b) of the skink Oligosoma maccanni during the austral spring-summer over several years (samples for pregnant females up until mid-october may include a few pre-ovulatory females; non-pregnant skinks are mainly males, plus a few subadult or spent females). A: Data for all days, regardless of whether basking was possible. Mean T b (shown in box, ±SE) did not differ significantly between pregnant (n = 117) and non-pregnant skinks (n = 89). B: T b adjusted for log T a on days on which T a was measured and basking was possible (sunny and no-moderate wind). T b adjusted for variation in log T a using ANCOVA (adjusted mean ± SE, in box) did not differ significantly between pregnant (n = 44) and non-pregnant skinks (n = 33). Figure 5. Mean T sel (±SE) of Oligosoma maccanni in early pregnancy (October) when placed on a thermal gradient (n = 9). Mean T sel did not vary with time of day (repeated measures ANOVA, P = 0.463). FREQUENCY WITH WHICH T SEL CAN BE ACHIEVED IN THE FIELD Based on the temperatures recorded by copper models, mean daytime T sel for pregnant females (28.9 C) was achievable in the field, for one or more hourly recordings, on 49% of days between mid- October and late February. Although the percentage did not vary significantly between months, December was noticeably cool (only 29% of days allowed mean T sel to be reached). Data collected by copper models (by day) and in under-rock positions (by night) were combined to estimate the overall temperature frequency distributions that could be achieved by skinks in each month (Fig. 6). The proportion of time that mean daytime T sel of pregnant females was achievable varied significantly between months (c 2 = 49.7, d.f. = 4, P < 0.05). The lowest value (4%, in December, equivalent to 0.9 h per day on average) was significantly lower than in all other months combined (c 2 = 28.0, d.f. = 1, P < 0.05). In the warmest months of January and February, the mean daytime T sel of pregnant females was achievable only 14 15% of the time (3.4 3.6 h per day on average, with the four individual models ranging between 12 and 17%). Daytime T sel was sometimes achievable under rocks. The warmest under-rock logger position reached T sel frequently in January and February (11 12% of the time), but other under-rock positions reached T sel in no more than 3% of any month. Temperatures 16 C (which may represent a minimum for embryonic development; see Discussion ) were achievable only 21% of the time in December, reaching a maximum of 54 58% of the time in January and February (Fig. 6). DISCUSSION Our results demonstrate that pregnant females of a small, viviparous lizard (O. maccanni) in a sub-alpine T

548 J. R. HARE ET AL. Figure 6. Estimated frequency with which the skink Oligosoma maccanni reaches different body temperatures in the field during the months of pregnancy. Symbols show the percentage of time that T b is expected to be at each temperature, based on temperatures recorded by copper models and under-rock loggers in 2004/5 (skinks are assumed to select the warmest location). Values reaching or exceeding mean T sel of pregnant females (28.9 C, arrowhead) are combined at this value, on the assumption that females would not voluntarily expose themselves to a higher T b. Dark shading shows the central 50% bound for T sel. Light shading shows temperatures below T sel in which embryonic development may still be possible (see Discussion). Temperatures generally increase as summer progresses, but December was unexpectedly cool. environment have infrequent access to T sel, a temperature considered to be suitable if not optimal for embryonic development (Beuchat, 1988; Mathies & Andrews, 1997; Ji et al., 2007). Field observations provided no evidence for maternal thermophily, in contrast with reports of altered thermoregulation during pregnancy in many lizard species. VALIDATION OF COPPER MODELS AS ESTIMATES OF T E Both indoors and outdoors, the copper model represented skink T b very well, showing nearly identical heating and cooling profiles, and maintenance of similar plateau temperatures through periods of both sun and cloud. Rapid warming of basking skinks was also demonstrated: for example, T b increased from 14.9 to 26.8 C within 25 min. The use outdoors of a freshly killed skink had the advantage of eliminating behaviour as a variable. Supporting this approach, Ibargüengoytía (2005) found that dead lizards (Phymaturus patagonicus) had the same thermal properties as live ones. In our validations, skink and model temperatures were limited to the range of biological interest ( 32 C, the maximum T b voluntarily experienced by wild skinks). Copper models reached higher temperatures in the field, but skinks retreated to cooler locations once T sel was exceeded. Although some studies have found copper models to have greater errors (Dzialowski, 2005), our results highlighted the usefulness of validated models for estimating the field T b of a skink in the same position. FIELD MICROHABITAT TEMPERATURES IN BASKING POSITIONS AND RETREAT SITES Temperature probes placed inside basking copper models and in retreat sites under rocks confirm that conditions in the sub-alpine habitat at Macraes Flat are often cool, even for diurnal, basking skinks. These data support and extend earlier descriptions of temperatures available for sympatric and primarily nocturnal thigmothermic geckos, H. maculatus (Rock et al., 2000; Rock, Cree & Andrews, 2002). Our data reveal stochastic conditions during the summer: in December 2004, the mean temperatures of basking models averaged only 20.3 C at the daily maximum, cooler by 4 10 C than in the preceding or following 2 months. Although our temperature records were incomplete during December for some loggers, cool and wet conditions prevailed throughout the Otago region for this month. Shade air temperature in December 2004 was 2.7 C lower (11.7 vs. 14.4 C) and rainfall was 28.2 mm higher (68.5 vs. 40.3 mm) than averages for the previous 4 years (data from the National Institute of Water and Atmospheric Research, NIWA). FIELD THERMOREGULATORY BEHAVIOUR OF PREGNANT AND NON-PREGNANT SKINKS Wind reduced basking behaviour, and the ability of skinks to elevate T b above T a. Although McCann s skinks elevated T b above T a on sunny days (by up to 17.3 C), this elevation was smaller than in the montane lizard Phrynosoma douglassi, in which T b values up to 30 C were attained at air temperatures as low as 1.5 C (Christian, 1998). We found no evidence that pregnant O. maccanni bask more frequently or attain higher or more stable T b than non-pregnant skinks, contrary to the generalization that pregnant squamates often choose higher temperatures to hasten gestation and minimize the cost/risk of reproduction (Schwarzkopf & Shine, 1991; Blázquez, 1995; Rock et al., 2000). Other lizards (several Sceloporus species) show reduced T b during pregnancy (Beuchat, 1988; Mathies & Andrews, 1997; Shine, 2006 for more references), but

COOL CLIMATES AND VIVIPARITY IN LIZARDS 549 field T b is often relatively high in non-pregnant individuals of these species (c. 34 C), and reduced T b may be beneficial for offspring size and viability (Beuchat, 1988; Mathies & Andrews, 1997). The lack of difference in T b between pregnant and non-pregnant O. maccanni is mirrored in several species, including skinks in the genus Mabuya from a lowland tropical environment (Vrcibradic & Rocha, 2004) and the Patagonian lizard Liolaemus elongatus (Ibargüengoytía & Cussac, 2002). In contrast with the prediction that evidence of maternal thermophily will be greatest in cool-climate species (Shine, 2006), our result for O. maccanni suggests that, where the thermal environment is limiting, all skinks (whether pregnant or not) thermoregulate intensively, at least for most of the summer months. During late summer, when temperatures available to basking O. maccanni are often well over 30 C and under-rock retreats also sometimes reach T sel, elevated T b may be achieved for longer periods of time by pregnant than non-pregnant skinks; however, our data do not show this. As locomotory ability may be reduced during pregnancy (e.g. Lampropholis guichenoti; Shine, 2003), pregnant lizards may alter their basking behaviour by becoming more cryptic (Lacerta vivipara; Bauwens & Thoen, 1981) and allowing closer approach distances (Platysaurus intermedius; Braña, 1993; Lailvaux, Alexander & Whiting, 2003). Our study may have underestimated the frequency of basking of both pregnant and non-pregnant McCann s skinks, as at least some skinks (with high T b) must have been basking just prior to capture in retreat sites. Although it is possible that pregnancy in O. maccanni causes subtle, undetected differences in behaviour, our measurements of T b provide no support for such differences having thermoregulatory significance. SELECTED BODY TEMPERATURE DURING PREGNANCY Daytime T sel of O. maccanni in early pregnancy (28.9 C) is similar to the lower end of values reported for female lizards during the reproductive season, e.g. L. elongatus (29.5 C) and P. patagonicus (31.9 C; Ibargüengoytía, 2005), Mabuya multifasciata (29.0 C when pregnant; Ji et al., 2007) and the sympatric gecko H. maculatus (27 28 C in early pregnancy; Rock et al., 2000). Unlike H. maculatus and some other lizards (for example, Daut & Andrews, 1993; Cortés et al., 1994; Angilletta, Montgomery & Werner, 1999), pregnant O. maccanni show no significant fall in T sel at night, thus resembling the geckos Goniurosaurus kuroiwae kuroiwae and Eublepharis macularius during the post-reproductive season (Werner et al., 2005). We suggest that this is further evidence of a thermally limiting habitat: when the opportunity to achieve T sel at night is available (as on a thermal gradient), pregnant skinks take advantage of it. Therefore, the selection of warm overnight retreat sites may also be a priority. We did not measure T sel in non-pregnant skinks or skinks in later stages of pregnancy; however, basking non-pregnant skinks allowed field T b to reach 29.5 C and females basking in late pregnancy allowed field T b to reach 32.2 C, suggesting that their daytime T sel is not any lower than measured here for early pregnancy. It is possible that night-time T sel falls to lower values in non-pregnant skinks, or later in pregnancy when warm daytime temperatures are more reliably available. Daytime T sel is clearly labile to recent thermal opportunities in both gravid (eggs in oviducts) and non-gravid individuals of the oviparous skink Bassiana duperreyi (Shine, 2006). FREQUENCY WITH WHICH T SEL CAN BE ACHIEVED IN THE FIELD We estimate that pregnant O. maccanni are able to achieve daytime T sel only 4 15% of the time during each month of pregnancy. To our knowledge, our study is the first to estimate the amount of time that T sel is attainable for such a small-bodied, cool-climate lizard throughout pregnancy. These results support a growing recognition that opportunities for maternal thermoregulation in cool climates (and hence the ability of pregnant lizards to buffer embryos from environmental fluctuations) are not as great as once thought (Andrews, 2000; Lourdais et al., 2004; Shine, 2004a, b). Our model can also be used to estimate exposure to non-optimal temperatures that might still support embryonic development (Fig. 6). Temperature development norms are not available for O. maccanni; however, studies of several oviparous lizards from Australia suggest that embryonic development is unlikely to be successful below about 16 C (Shine & Harlow, 1996). Limited information for two oviparous species from New Zealand, the skink O. suteri (Hare, Daugherty & Cree, 2002) and the rhynchocephalian tuatara Sphenodon punctatus (Thompson, 1990), is broadly consistent. Assuming a similar minimum temperature of 16 C for embryonic development in O. maccanni, pregnant females spend only 21 58% of each month at temperatures at which embryonic development can progress (Fig. 6). Our approach for estimating exposure to different body temperatures has several advantages over direct methods, such as telemetry: no assumptions about the effects of physical burdens or surgical modifications on thermoregulatory behaviour need to be made, and lizards do not need to be recaptured, or restrained in arenas. The validity of our approach nonetheless depends on several assumptions about

550 J. R. HARE ET AL. the behaviour of McCann s skinks. Exposed copper models estimate T b well for O. maccanni basking at the same site, but skinks can move around, perhaps accessing better basking sites, and must engage at times in behaviours other than basking (e.g. predator avoidance). Nonetheless, the open nature of the habitat, the relatively sedentary behaviour of O. maccanni when emerged, and the avidity with which skinks often re-emerge after attempts at capture suggest that the effects of these behaviours on T b may be minor. We assumed that skinks would bask when copper models were warmer than retreat site temperatures (until T sel was reached). This prediction appears to be true, because the temperatures at which copper models became warmer than retreat sites during the mornings (c. 7 13 C, Fig. 3) corresponded well with the lowest air temperatures at which basking was observed, or assumed from elevated T b to occur (7 10 C). Our assumption that skinks retreat to shaded or under-rock positions to avoid higher temperatures when basking positions exceed T sel also appears to be true (90% of T b values for pregnant skinks were below 30.2 C, the upper limit of the central 50% bound for daytime T sel in early pregnancy). IMPLICATIONS OF LOW OR VARIABLE EXPOSURE TO SELECTED TEMPERATURE DURING PREGNANCY Our observation of low exposure to T sel during pregnancy in McCann s skinks raises questions of ecological and evolutionary interest. 1. What are the implications for reproductive success in O. maccanni from low and variable exposure to T sel during pregnancy? Gestation length and success are clearly temperature dependent in viviparous lizards (Beuchat, 1988; Rock & Cree, 2003), and so O. maccanni should be able to achieve shorter and perhaps more successful pregnancies at sites offering greater exposure to T sel. However, even at our study site, offspring are born in the warmest months of the year (January February), suggesting that offspring survival is not necessarily compromised by the current time of birth. There is growing evidence that temperature regimes experienced by embryos of viviparous reptiles affect phenotypic outcomes. Affected features include body size, body shape, locomotor speed, scale pattern and even sex (for a review, see Shine, 2004a; Ji et al., 2006, 2007). Evidence from oviparous lizards suggests that early developmental stages are more susceptible to these effects (Shine & Elphick, 2001; see also Andrews, 2004). Although we know of no experimental studies investigating stage-specific effects in viviparous lizards, a field study of the viviparous snake Vipera aspis close to its cool-temperature limits suggests that variation in the number of ventral scales in the offspring is sensitive to temperatures during early pregnancy (Lourdais et al., 2004). Thus, effects on time of birth or on offspring phenotypes in O. maccanni from the sustained period of low temperatures during December 2004 [at approximately stage 32 in the 40-point staging scheme of Dufaure & Hubert (1961), slightly before the rapid growth phase] would not necessarily be the same as if such unseasonably cool temperatures occurred earlier or later in pregnancy. Information is needed on the effects of low exposure to T sel in O. maccanni, and on the stage-specific effects of low exposure to T sel in viviparous lizards in general. 2. How do potential nest site temperatures for oviparous reptiles compare with those we have estimated for viviparous O. maccanni? This question must be answered to address the hypothesis that viviparity enhances the thermal environment for embryos over those available with oviparity (Andrews, 2000; Shine, 2004b). Currently, no oviparous lizards occur in southern New Zealand, although the rhynchocephalian tuatara (Sphenodon), representing the sister group of squamates, occurred in broad sympatry with O. maccanni as recently as a few hundred years ago (Worthy & Holdaway, 2002). This evidence indicates that reptilian oviparity was possible in climates similar to those inhabited by O. maccanni today. Unlike O. maccanni, tuatara are large-bodied (female S. punctatus reach about 500 g on Stephens Island), burrow dwelling and partly nocturnal, although basking to attain body temperatures up to 30 C (Barwick, 1982). We suggest that, as a consequence of the large body mass (Seebacher & Shine, 2004), female tuatara would heat more slowly than O. maccanni, and thus derive less thermal advantage from rapid but short-term improvements in weather. A large body mass would also inhibit tuatara from sheltering in rock crevices, reducing their ability to access warm rock surfaces in the late afternoon. Under the scenario of a large body mass, soil temperatures in shallow nests in southern New Zealand might offer a more favourable thermal environment for development than inside the body of a viviparous female, a prediction that remains to be tested. ACKNOWLEDGEMENTS This study was performed under approvals from the New Zealand Department of Conservation and the University of Otago Animal Ethics Committee,

COOL CLIMATES AND VIVIPARITY IN LIZARDS 551 following consultation with the University of Otago Ngai Tahu Consultation Committee and Kati Huirapa Runaka ki Puketeraki. We thank Keith and Margaret Philip for land access and hospitality; many helpers, especially Anne Besson, Joanne Connolly, Clément Lagrue, Jack Mace and Claudine Tyrrell, for field assistance; staff of the Department of Zoology for technical assistance; and Anne Besson, Mandy Chamberlain and Kelly Hare for comments on the manuscript. Funding was assisted by a University of Otago Research Grant. REFERENCES Andrews RM. 2000. Evolution of viviparity in squamate reptiles (Sceloporus spp.): a variant of the cold-climate model. Journal of Zoology, London 250: 243 253. Andrews RM. 2004. Patterns of embryonic development. In: Deeming DC, ed. Reptilian incubation: environment, evolution and behaviour. Nottingham: Nottingham University Press, 75 102. Angilletta MJ, Montgomery LG, Werner YL. 1999. Temperature preference in geckos: diel variation in juveniles and adults. Herpetologica 55: 212 222. Barwick RE. 1982. Observations on active thermoregulation in the tuatara, Sphenodon punctatus (Reptilia: Rhynchocephalia). In: Newman DG, ed. New Zealand herpetology. Wellington: New Zealand Wildlife Service Occasional Publication No. 2, 225 236. Bauwens D, Thoen C. 1981. Escape tactics and vulnerability to predation associated with reproduction in the lizard Lacerta vivipara. Journal of Animal Ecology 50: 733 743. Beuchat CA. 1988. Temperature effects during gestation in a viviparous lizard. Journal of Thermal Biology 13: 135 142. Blázquez MC. 1995. Body temperature, activity patterns and movements by gravid and non-gravid females of Malpolon monspessulanus. Journal of Herpetology 29: 264 266. Braña F. 1993. Shifts in body temperature and escape behaviour of female Podarcis muralis during pregnancy. Oikos 66: 216 222. Charland MB. 1995. Thermal consequences of reptilian viviparity: thermoregulation in gravid and non-gravid garter snakes (Thamnophis). Journal of Herpetology 29: 383 390. Christian KA. 1998. Thermoregulation by the short-horned lizard (Phrynosoma douglassi) at high elevation. Journal of Thermal Biology 23: 395 399. Cortés A, Báez C, Rosenmann M, Pino C. 1994. Body temperature, activity cycle and metabolic rate in a small nocturnal Chilean lizard, Garthia gaudichaudi (Sauria: Gekkonidae). Comparative Biochemistry and Physiology 109A: 967 973. Cree A, Guillette LJ Jr. 1995. Biennial reproduction with a fourteen-month pregnancy in the gecko Hoplodactylus maculatus from southern New Zealand. Journal of Herpetology 29: 163 173. Daut EF, Andrews RM. 1993. The effect of pregnancy on thermoregulatory behaviour of the viviparous lizard Chalcides ocellatus. Journal of Herpetology 27: 6 13. Dufaure JP, Hubert J. 1961. Table de développement du lézard vivipare: Lacerta (Zootoca) vivipara, Jacquin. Archives d Anatomie Microscopique et de Morphologie Expérimentale 50: 309 328. Dzialowski EM. 2005. Use of operative temperature and standard operative temperature models in thermal biology. Journal of Thermal Biology 30: 317 334. Hare JR, Whitworth E, Cree A. 2007. Correct orientation of a hand-held infrared thermometer is important for accurate measurement of body temperatures in small lizards and tuatara. Herpetological Review 38: 311 315. Hare KM, Daugherty CH, Cree A. 2002. Incubation regime affects juvenile morphology and hatching success, but not sex, of the oviparous lizard Oligosoma suteri (Lacertilia: Scincidae). New Zealand Journal of Zoology 29: 221 229. Hodges WL. 2004. Evolution of viviparity in horned lizards (Phrynosoma): testing the cold-climate hypothesis. Journal of Evolutionary Biology 17: 1230 1237. Holmes KM, Cree A. 2006. Annual reproduction in females of a viviparous skink (Oligosoma maccanni) in a subalpine environment. Journal of Herpetology 40: 141 151. Ibargüengoytía NR. 2005. Field, selected body temperature and thermal tolerance of the syntopic lizards Phymaturus patagonicus and Liolaemus elongatus (Iguania: Liolaemidae). Journal of Arid Environments 62: 435 448. Ibargüengoytía NR, Cussac VE. 2002. Body temperatures of two viviparous Liolaemus lizard species, in Patagonian rain forest and steppe. Herpetological Journal 12: 131 134. Ji X, Lin C-X, Lin L-H, Qiu Q-B, Du Y. 2007. Evolution of viviparity in warm-climate lizards: an experimental test of the maternal manipulation hypothesis. Journal of Evolutionary Biology 23: 1037 1045. Ji X, Lin L-H, Luo L-G, Lu H-L, Gao J-F, Han J. 2006. Gestation temperature affects sexual phenotype, morphology, locomotor performance, and growth of neonatal brown forest skinks, Sphenomorphus indicus. Biological Journal of the Linnean Society 88: 453 463. Lailvaux SP, Alexander GJ, Whiting MJ. 2003. Sex-based differences and similarities in locomotor performance, thermal preferences and escape behaviour in the lizard Platysaurus intermedius wilhelmi. Physiological and Biochemical Zoology 76: 511 521. Lourdais O, Shine R, Bonnet X, Guillon M, Naulleau G. 2004. Climate affects embryonic development in a viviparous snake, Vipera aspis. Oikos 104: 551 560. Mathies T, Andrews RM. 1997. Influence of pregnancy on the thermal biology of the lizard, Sceloporus jarrovi: why do pregnant females exhibit low body temperatures? Functional Ecology 11: 498 507. Patterson GB, Daugherty CH. 1990. Four new species and one new subspecies of skinks, genus Leiolopisma (Reptilia: Lacertilia: Scincidae) from New Zealand. Journal of the Royal Society of New Zealand 20: 65 84. Robert KA, Thompson MB. 2003. Reconstructing thermocron ibuttons to reduce size and weight as a new technique

552 J. R. HARE ET AL. in the study of small animal thermal biology. Herpetological Review 34: 130 132. Rock J, Andrews RM, Cree A. 2000. Effects of reproductive condition, season and site on selected temperatures of a viviparous gecko. Physiological and Biochemical Zoology 73: 344 355. Rock J, Cree A. 2003. Intraspecific variation in the effect of temperature on pregnancy in the viviparous gecko Hoplodactylus maculatus. Herpetologica 59: 8 22. Rock J, Cree A, Andrews RM. 2002. The effect of reproductive condition on thermoregulation in a viviparous gecko from a cool climate. Journal of Thermal Biology 27: 17 27. Schwarzkopf L, Shine R. 1991. Thermal biology of reproduction in viviparous skinks, Eulamprus tympanum: why do gravid females bask more? Oecologia 88: 562 569. Seebacher F, Shine R. 2004. Evaluating thermoregulation in reptiles: the fallacy of the inappropriately applied method. Physiological and Biochemical Zoology 77: 688 695. Shine R. 2003. Locomotor speeds of gravid lizards: placing costs of reproduction within an ecological context. Functional Ecology 17: 526 533. Shine R. 2004a. Adaptive consequences of developmental plasticity. In: Deeming DC, ed. Reptilian incubation: environment, evolution and behaviour. Nottingham: Nottingham University Press, 187 210. Shine R. 2004b. Incubation regimes of cold-climate reptiles: the thermal consequences of nest site choice, viviparity and maternal basking. Biological Journal of the Linnean Society 83: 145 185. Shine R. 2006. Is increased maternal basking an adaptation or a pre-adaptation to viviparity in lizards? Journal of Experimental Zoology 305A: 524 535. Shine R, Elphick MJ. 2001. The effect of short-term weather fluctuations on temperatures inside lizard nests, and on the phenotypic traits of hatchling lizards. Biological Journal of the Linnean Society 72: 555 565. Shine R, Harlow PS. 1996. Maternal manipulation of offspring phenotypes via nest-site selection in an oviparous lizard. Ecology 77: 1808 1817. Shine R, Kearney M. 2001. Field studies of reptile thermoregulation: how well do physical models predict operative temperatures? Functional Ecology 15: 282 288. Taylor EN, DeNardo DF, Malawy MA. 2004. A comparison between point and semi-continuous sampling for assessing body temperature in a free ranging ectotherm. Journal of Thermal Biology 29: 91 96. Thompson MB. 1990. Incubation of eggs of tuatara, Sphenodon punctatus. Journal of Zoology, London 222: 303 318. Vitt LJ, Sartorius SS. 1999. HOBOs, Tidbits and lizard models: the utility of electronic devices in field studies of ectotherm thermoregulation. Functional Ecology 13: 670 674. Vrcibradic D, Rocha CFD. 2004. Field body temperatures of pregnant and nonpregnant females of three species of viviparous skinks (Mabuya) from southeastern Brazil. Journal of Herpetology 38: 447 451. Werner YL, Takahashi H, Mautz WJ, Ota H. 2005. Behavior of the terrestrial nocturnal lizards Goniurosaurus kuroiwae kuroiwae and Eublepharis macularius (Reptilia: Eublepharidae) in a thigmothermal gradient. Journal of Thermal Biology 30: 247 254. Worthy TH, Holdaway RN. 2002. The lost world of the moa: prehistoric life of New Zealand. Bloomington, IN: Indiana University Press.