Evolution of biometric and life-history traits in lizards (Gallotia) from the Canary Islands

Males and females of the oriental garden lizard (Calotes versicolor) from a population in Hainan, southern China, differed in head size at hatching: males had larger heads than females. This dimorphism was slightly more pronounced in adults than in hatchlings. Hatchlings had larger heads but shorter tails relative to snout–vent length (SVL) than subadults and adults. When SVL was held constant, tail length of subadults and adults of the same sex was similar. Adult lizards were not sexually dimorphic in body size (SVL), but males possessed longer tails than females at the same SVL throughout their size range. The smallest and largest reproductive females in our sample were 78.9 and 103.8 mm SVL, respectively. Females in our laboratory (Hangzhou, eastern China) laid multiple clutches from April to June. Clutch size, clutch mass, and egg mass varied from 3 to 14 (mean = 9.0) eggs, 1.58 to 6.47 (mean = 4.65) g, and 0.45 to 0.61 (mean = 0.52) g, respectively. Clutch size and clutch mass were both positively correlated with female SVL. Egg mass was not correlated with female SVL even when clutch size was held constant using a partial correlation analysis. Clutch size was not correlated with female condition when female SVL was held constant. Egg length, egg width, and clutch size were not correlated with each other when the remaining variable was held constant. Egg mass was not correlated with clutch size when female SVL was kept constant. Larger-sized C. versicolor females increased reproductive output primarily through production of more eggs (and hence heavier clutches).

Simple SummaryWe review the results of studies performed during the last four decades on the morphology, behaviour and evolution of lizards of the genus Gallotia, from the Canary Islands. We show that there is sexual dimorphism in all species of that genus, with males having larger snout-vent length (SVL), head width (HW) and hind limb lengths (HLLs) than females. The changes in SVL of males and females have been correlated throughout the evolution of these lizards in the islands. In two species, longer HLLs were found in populations from open (less vegetation) rather than closed habitats. In most species, males have a more conspicuous coloration, with blue or green large lateral spots. Blue spots in some species also have their peak reflectance in the ultraviolet part of the spectrum (300–400 nm wavelength). Detailed analysis of Gallotia galloti shows a large repertoire of behaviour patterns, and results of intramale competition trials show that the outcome of fights depends on several morphological patterns and especially on bite frequency. Detailed behavioural analyses of individuals of two endangered lizard species proved to be a useful tool for keeping them while in breeding centres, as well as to train them to recognize local predators before they were reintroduced into natural habitats.We summarize, here, the results from several studies conducted over many years on several endemic species of lizards (genus Gallotia) from the Canary Islands. Quantitative analyses show clear differences both among the species of every island and populations within each species. Sexual dimorphism exists in all analysed species, and a phylogenetic analysis shows that the degree of dimorphism did not change along the evolutionary history of the Canary Islands: species with large and small body sizes have a similar degree of sexual dimorphism, with male body size changes closely following those undergone by females. In G. caesaris (from El Hierro and La Gomera islands) and in G. stehlini (from Gran Canaria), longer hind limb length was correlated with more open habitats. Within most species, males are more conspicuous than females, mainly in terms of body size, behaviour and coloration pattern. Lateral colour spots are blue in most species and green in others. In G. galloti from Tenerife, male lateral spots have larger spot areas and percentage of reflectance in the ultraviolet/blue part of the spectrum than females. This trait shows a monthly variation along April to July, both in males and females, its magnitude being larger in May–June. Behaviour analysis, especially in the last species, shows a great diversity in behaviour patterns, and analysis of intrasexual male competition revealed that contest outcome depends on several morphological and colouration characteristics but mainly on the individual’s behaviour. Detailed behavioural analyses were useful for managing a few captive individuals of the highly endangered G. bravoana from La Gomera island. Experimental analyses of some behaviours in the endemic Hierro island lizard (G. simonyi, in danger of extinction) show that individuals may learn to recognize predator models and increase their running speeds with training.

Reproduction in a Nebraska Sandhills Population of the Northern Prairie Lizard Sceloporus undulatus garmani

We studied ontogenetic shifts in sexual dimorphism and female reproduction in the Reeves's Butterfly Lizard, Leiolepis reevesii (Agamidae), from Hainan, southern China. The smallest reproductive female was 84 mm snout–vent length (SVL). The largest female and male were 136 mm and 166 mm SVL, respectively. Sexual dimorphism in head size and tail size was evident in adults, with adult males having larger heads and longer tails than did adult females of the same SVL. Head size relative to SVL was greatest in hatchlings and smallest in adults, with juveniles in between. Tail length relative to SVL was shorter in hatchlings than in juveniles and adults but did not differ between juveniles and adults. Females produced a single clutch of 2–8 eggs per breeding season stretching from mid-April to mid-July. Clutch size, egg mass, and clutch mass were all positively related to female SVL. Egg mass was negatively correlated with clutch size when holding female SVL constant, and egg length was negatively correlated with egg width when holding egg mass constant. Larger females increased reproductive output primarily through production of more and larger eggs (and hence heavier clutches). Males are the larger sex in L. reevesii presumably because (1) the selective pressure toward increased male size is strong; (2) the selective pressure toward increased female size is comparatively low; and (3) somatic growth is more likely to be constrained by reproduction in smaller females because of a negative correlation between relative clutch mass and SVL.

The difference in body size and/or shape between males and females, called sexual size dimorphism, is widely accepted as the evolutionary consequence of the difference between reproductive roles. To study the mating pattern, female reproduction and sexual size dimorphism in a population of Microhyla fissipes, amplexus pairs were collected, and the snout-vent length of males and females, female reproductive traits and fertilization rate were measured. If the body size of amplexed females is larger than that of amplectant males, this is referred to as a female-larger pair, a phenomenon that was often observed for M. fissipes in this study. Interestingly, snout-vent length of males in male-larger pairs was greater than that in female-larger pairs, however the post-spawning body mass, clutch size, egg dry mass and clutch dry mass did not differ between both types of pairs. Snout-vent length of males was positively related to that of females in each amplexus pair. After accounting for the snout-vent lengths of females, we showed that snout-vent lengths of males in male-larger pairs were greater than those of females in female-larger pairs. The snout-vent length ratio of males and females was not related to fertilization rate in each amplexus pair. The mean fertilization rate was not different between both amplexus pairs. These results suggest that (1) M. fissipes displays female-biased sexual size dimorphism and has two amplexus types with size-assortative mating; (2) the snout-vent length ratio of males and females in each amplexus type was consistent with the inverse of Rensch’s rule, and was driven by the combined effect of sexual selection and fecundity selection; (3) females with a larger body size were preferred by males due to their higher fecundity, while the body size of males was not important for fertilization success.

The nesting ecology of the Allen Cays rock iguana was studied on Leaf Cay and Southwest Allen's Cay (= U Cay) in the northern Exuma Islands, Bahamas, during 2001 and 2002. Mating occured in mid-May, and females migrated 30–173 m to potential nest sites in mid to late June. Females often abandoned initial attempts at digging nest burrows, and average time from initiation of the final burrow to completion of a covered nest was six days. At least some females completely buried themselves within the burrow during the final stages of burrow construction and oviposition. Females defended the burrow site during the entire time of construction, and most continued that defense for at least three to four weeks after nest completion. Nests were completed between mid-June and mid-July, but for unknown reasons timing was seven days earlier on U Cay than on Leaf Cay. Nest burrows averaged 149 cm in length and terminal nest chambers usually angled off the main burrow. Depth to the bottom of the egg chamber averaged 28 cm, and was inversely correlated with shadiness of the site, suggesting that females may select depths with preferred temperatures (mean, 31.4 C in this study). Overall, only about one in three females nested each year, although nesting frequency increased with female size such that the largest females usually nest annually. Nest fidelity was common, despite the potential for observer effects; seven of 13 two-year nesters placed nests within 0.7 m of that constructed the previous year.Nesting females averaged 32 cm snout–vent length (SVL) and 1336 g body mass, and larger, older females nested earlier than smaller, younger ones. Sexual maturity is reached at 26–27 cm SVL, about 750 g body mass, and twelve years of age (nearly twice as old as any previously studied lizard). Longevity of females apparently exceeds 40 years. Clutch size ranged from 1–10 eggs (mean 4.6) and was correlated with female body size and age. Eggs averaged 66 mm in length, 35 mm in width, and 49 g in mass. Egg mass was not correlated with female body size, although egg length was negatively correlated, and egg width was positively correlated with female size. The production of elongate eggs in the smaller females allowed them to invest the same total mass in each egg as a larger female, while being constrained by the limits of the pelvic opening. No trade-off existed between standardized clutch size versus egg size. Relative clutch mass (clutch mass/gravid female body mass × 100) averaged 16.5 and did not vary with female size or age. Hatching apparently occurs in late September and early October after about 80–85 days incubation, with emergence within just a few days. Hatchlings averaged 9.5 cm SVL and 33 g body mass. Survivorship to emergence was 78.9%, and was inversely correlated with soil moisture.The reproductive ecology of other iguanids (sensu strictu) is reviewed for comparison with that of the Allen Cays rock iguana. Comparisons of these data with those available for other rock iguanas of the genus Cyclura suggest that colonization of smaller islands has produced reductions in adult female body size, clutch size, clutch mass, and relative clutch mass, but no change in egg or hatchling mass. Because this pattern is also demonstrated by a population of Cuban iguanas introduced to a small island only 40 years ago, it may primarily be a proximal response to decreased resource availability and/or physiological processability on small islands rather than an evolved response to reduced predation rates or other factors affecting survivorship.

Intrapopulation Variability in Morphology, Coloration, and Body Size in Two Races of the Lacertid Lizard, Gallotia galloti

The remote Tomellin Canyon of southeastern Puebla and northcentral Oaxaca is inhabited by two distinctive gonochoristic whiptail lizards (genus Cnemidophorus). The giant whiptail lizard C. sacki, which is one of the largest species of Cnemidophorus, is associated with brushy arroyos, the margins of river flood plains, and canyon slopes with trees and shrubs. The diminutive whiptail lizard C. parvisocius, which is one of the smallest species of Cnemidophorus, prefers open canyon flats with cactus tangles, agave, and island-like groves of small trees. These species are marginally syntopic only along the interface of mesic habitats and xeric habitats. C. sacki (snout vent length of gravid females, x = 111.9 mm) has one of the largest clutch sizes (x = 5.9, range 3-10, N = 34) and C. parvisocius (snout vent length of gravid females, x = 59.4 mm) has the smallest clutch size (x = 1.8, 1-4, N = 85) reported for species of the sexlineatus group. Clutch size to snout vent length regressions for both species lie along the same regression slope indicating that differences in clutch size can be explained in part on the basis of differences in body size. Large body size in C. sacki is correlated with a reduction in the number of individuals per unit area and an increase in clutch potential. Small body size in C. parvisocius is correlated with an increase in population density and a decrease in clutch potential. The majority of yearly neonates in both species appear to attain sexual maturity in less than one calendar year.

We investigated life history traits (body sizes, sexual maturity, sexual dimorphism, clutch size) and dietary preferences in the sympatric lizards, Oligosoma nigriplantare polychroma and Oligosoma lineoocellatum (Scincidae). One‐hundred and forty O. n. polychroma and 153 O. lineoocellatum were collected for a twelve month period. Our results show that these sympatric species have different life history traits, notably in body size, size at maturity, and reproductive output. O. n. polychroma were smaller (51.72 mm adult males, 56.74 mm adult females) than O. lineoocellatum (70.70 mm adult males, 76.30 mm adult females). Size at maturity was determined by logistic regression models of snout‐vent length (SVL) for both species. O. n. polychroma reached sexual maturity at a relatively smaller SVL (42.21 mm) than O. lineoocellatum (62.11 mm), and had a larger mean clutch size (5.79) than O. lineoocellatum (3.82). There were some differences in diet, although both species consumed a wide range of prey, including spiders, worms, beetles, and millipedes.

In a model group of giant reptiles, we explored the allometric relationships between male and female body size and compared the effects of sexual and fecundity selection, as well as some proximate causes, on macroevolutionary patterns of sexual size dimorphism (SSD). Monitor lizards are a morphologically homogeneous group that has been affected by extreme changes in body size during their evolutionary history, resulting in 14-fold differences among the body sizes of recent species. Here, we analysed data concerning the maximum and/or mean male and female snout–vent lengths in 42 species of monitor lizard from literary sources and supplemented these data with measurements made in zoos. There was a wide scale of SSD from nearly monomorphic species belonging mostly to the subgenus Odatria and Prasinus group of the Euprepriosaurus to apparently male-larger taxa. The variable best explaining SSD was the body size itself; the larger the species, the higher the SSD. This pattern agrees with the currently discussed Rensch’s rule, claiming that the relationship between male and female body size is hyperallometric, i.e. the allometric exponent of this relationship exceeds unity and thus SSD increases with body size in the case of male-larger taxa. All our estimates of the reduced major axis regression slopes of this relationship ranged from 1.132 to 1.155. These estimates are significantly higher than unity, and thus unequivocally corroborate the validity of Rensch’s rule in this reptilian group. In spite of our expectation that the variation in SSD can be alternatively explained by variables reflecting the strength of sexual selection (presence of male combat), fecundity selection (e.g. clutch size and mass) and/or proximate ecological factors (habitat type), none of these variables had consistent effects on SSD, especially when the data were adjusted to phylogenetic dependence and/or body size. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 293–306. ADDITIONAL KEYWORDS: allometry – body size – evolution of SSD – fecundity selection – monitor lizards – sexual selection – Varanidae.

The whiptail lizard Cnemidophorus deppii was studied during late dry season on a tropical beach on the Pacific coast of Nicaragua. Most aspects of the ecology of this species are similar to those of other active foraging lizard species studied. Individual C. deppii spend most of a typical daily activity period on sand moving from vegetation patch to vegetation patch, presumably in search of food. The amount of time spent in the sun is greatest in early morning and at its lowest level at midday. The average rate of movement was 0.048 ± 0.004 m/s. Body temperatures of active lizards averaged 40.0 ± 0.25 °C, and most activity occurred during morning and late afternoon. Body temperatures were significantly lower in whiptails active during the morning than later in the day. Forty-two types of prey were identified in stomachs, with termites, spiders, and various orthopterans accounting for most of the diet volumetrically. There was no correlation between lizard size and prey size. There was a significant negative relationship between prey width and the number of prey in stomachs. Snout–vent length (SVL) at sexual maturity was 60 mm for females and 58 mm for males. Mature females averaged 63.8 ± 0.7 mm SVL and produced clutches varying from 1 to 3 eggs ([Formula: see text]). Oviductal eggs averaged 13.6 ± 0.64 × 7.7 ± 0.21 mm in size. There was no significant relationship between female SVL and clutch size. Relative clutch mass was similar to that for other active foraging lizard species. Sexual dimorphism was apparent in coloration (males brightly colored), body size (males larger), and relative head size (male heads larger independent of body size differences). These differences presumably are due to sexual selection.

Aspects of the reproductive biology of a trogloxene population of the pickerel frog, Rana palustris (Anura: Ranidae), were studied at a cave on the edge of the Ozark plateau in Missouri. Sexual dimorphism in body size was marked; there was almost no overlap in adult body size ranges, and the ratio of mean adult female snout–vent length (SVL) to adult male SVL was 1.26. All males with SVL of >45 mm were sexually mature. Males showed a clear cycle of spermatogenesis, with a peak in midsummer and a decline in spermatogenic activity through autumn and winter. Minimum size at maturity for females was 59 mm SVL. Females completed vitellogenesis during the summer, before their arrival at the study site. Of 28 females above the minimum size at maturity, 27 contained egg clutches. Clutch size, clutch mass, and egg size (mass) show significant positive correlation with SVL. Mean (±SD) clutch size was 1759 ± 548. Fat bodies were present in both males and females in autumn, but were totally depleted before the animals emerged from hibernation. Trogloxene R. palustris do not diverge significantly from patterns of reproductive biology typical of other temperate zone Rana. Current or past patterns of cave use have had no detectable effect on reproductive characteristics of trogloxene R. palustris.

We examined museum specimens to quantify diet, reproductive cycles and sexual dimorphism in body size of the colubrid snake Crotaphopeltis hotamboeia. Females attain sexual maturity at approximately 300 mm snout‐vent length (SVL) and males at approximately 240 mm SVL. Females grow larger than males (maximum SVLs of 830 vs 700 mm SVL), and also have longer and wider heads and wider bodies than do males of the same body length. Males have relatively longer tails than females, but eye size relative to head length shows no sexual dimorphism. Both sexes breed each year. Females commence vitellogenesis in late winter (September) and oviposit from October to January. Clutch size ranged from four to 12 eggs with a mean of 7.58 and was highly correlated with female SVL. The testes of adult males are turgid throughout the year, suggesting a prolonged mating season. Of 73 prey items recorded, 97% were anurans. Six anuran families were represented among the prey items, but bufonids (39%), ranids (29%) and microhylids (25%) comprised most of the 51 anuran food items identified to genus. Larger snakes ate larger prey items, in terms of SVL as well as mass. However, the snake's sex and age (adult vs juvenile) did not affect prey type. Bufonids, microhylids and ranids were consumed by snakes of all ages and both sexes, and were eaten all year except during mid‐winter (July and August).

We studied the reproduction, sexual dimorphism, and diet of Oxyrhopus trigeminus from two sites in southeastern Brazil. Oxyrhopus trigeminus from Irapé Power Plant (IPP) contained vitellogenic follicles and eggs in both rainy and dry seasons and clutch size was not correlated with female snout–vent length (SVL). Sexual dimorphism was evident. Females attain larger SVL but males have longer tails. We found three females from Santa Clara Power Plant (SPP) with vitellogenic follicles, all of them collected in the dry season. Mean SVLs of adult females from IPP and SPP were 717.7 mm and 786 mm, respectively. Mean SVL of adult males from IPP was 553.4 mm and the single adult male from SPP was 507 mm. The diet of O. trigeminus from IPP included rodents (46.7%), lizards (33.3%), and birds (20%). The volume of individual prey items was not correlated with snake SVL. The diet of O. trigeminus from SPP included rodents (37.5%), lizards (37.5%), birds (12.5%), and marsupials (12.5%). It seemed that an ontogenetic shift may occur in individuals of this snake species from IPP.

Aspidoscelis neotesselata (Squamata: Teiidae), the triploid, parthenogenetic lizard endemic to southeastern Colorado, originated from a hybridization event involving normally parthenogenetic Aspidoscelis tesselata and bisexual Aspidoscelis sexlineata. A previous study of a sympatric assemblage of these taxa in the Higbee vicinity, Otero County, Colorado, found that A. neotesselata resembled A. tesselata in snout–vent length (SVL) and the much smaller A. sexlineata in clutch size, a pronounced departure from the clutch size predicted from the mean SVL of gravid females. The phylogenetic constraint hypothesis, that clutch size in A. neotesselata is influenced disproportionately by the genome acquired from A. sexlineata, was proposed to explain this enigma. Use of new characters in recently acquired samples of A. tesselata Pattern Class C (N = 44), A. neotesselata Pattern Classes A, B, and C (N = 132), and A. sexlineata (N = 31), revealed that clutch size in A. neotesselata is genetically constrained by both progenitors—a nonadditive component from A. tesselata (large egg volume) and an additive component from A. sexlineata (reduced body volume). Although mean oviductal egg volume, mean SVL for hatchlings of the year, and maximum observed body size were similar in A. neotesselata and A. tesselata, representatives of A. neotesselata attained reproductive maturity at smaller body sizes, and their oviductal eggs were proportionately longer and narrower than those of both progenitor species. Gravid individuals of A. neotesselata were intermediate to those of A. tesselata and A. sexlineata in body volume (BV) and SVL, but, as previously reported, the mean clutch sizes of A. neotesselata and A. sexlineata were similar and significantly smaller than that of A. tesselata. Statistically, egg volume was the least variable of the four primary reproductive attributes investigated. This would presumably ensure the production of hatchlings of adaptive size by each species, a factor that likely supersedes clutch size and body volume in maximizing individual fitness. Hybrid-derived polyploidy (i.e., the addition of an A. sexlineata genome to the diploid genome of A. tesselata) in A. neotesselata and the postformational origin of geographic Pattern Classes A, B, and C, have resulted in a constellation of adaptations that extend well beyond reproductive characteristics. These adaptations have permitted each of the three triploid pattern classes to exploit mutually exclusive habitats, some at a distance of 173 linear kilometers west-northwest of the northern range boundary of A. tesselata in southeastern Colorado.