Arthur E. Dunham

The Temperature‐Size Rule in Ectotherms: Simple Evolutionary Explanations May Not Be General

In many organisms, individuals in colder environments grow more slowly but are larger as adults. This widespread pattern is embodied by two well‐established rules: Bergmann’s rule, which describes the association between temperature and body size in natural environments, and the temperature‐size rule, which describes reaction norms relating temperature to body size in laboratory experiments. Theory predicts that organisms should grow to be larger in colder environments when growth efficiency decreases with increasing environmental temperature. Using data from 97 laboratory experiments, including 58 species of ectotherms, we found little evidence that growth efficiency is negatively related to environmental temperature within the thermal range that is relevant to the temperature‐size rule. Instead, growth efficiency was either positively related or insensitive to environmental temperature in the majority of cases (73 of 89 cases for gross growth efficiency and 18 of 24 cases for net growth efficiency). Two possibilities merit consideration. First, high temperatures may impose constraints on growth that only arise late during ontogeny; this simple and potentially general explanation is supported by the fact that thermal optima for growth efficiency and growth rate decrease as individuals grow. Alternatively, the general explanation for relationships between temperature and body size may not be simple. If the latter view is correct, the best approach might be to generate and test theories that are tailored specifically to organisms with similar behavior and physiology.

Interfaces between Biophysical and Physiological Ecology and the Population Ecology of Terrestrial Vertebrate Ectotherms

Physiological and biophysical processes interact with a suite of environmental factors to produce important patterns in the population ecology of terrestrial vertebrate ectotherms. We develop a mechanistic approach to understanding the relative contributions of these interactions. Our approach requires a distinction between a life history and a life-history phenotype and allows the incorporation of system-specific trade-offs and constraints in a manner that facilitates the generation of testable predictions for specific populations or sets of populations. We define operant sources of selection and the effects of these averaged over the lifetime of individuals exposed to different environmental factors (available foraging microhabitats, mate availability, preferred egg sites), and synthesize a probabilitistic definition of a life history. We apply this approach to understanding geographic variation in several life-history characters in the saxicolous, iguanid lizard Sceloporus merriami.

Food Availability as a Proximate Factor Influencing Individual Growth Rates in the Iguanid Lizard Sceloporus Merriami

Von Bertalanffy, logistic-by-length, and logistic-by-weight models of individual growth for 2 populations of Sceloporus merriami were estimated from 3 yr (1974-1976) of mark-recapture data using nonlinear regression. The logistic-by-weight model was found to provide the best fit to observed growth rates. Abundance of arthropod prey was estimated by 2 techniques and found to be serially correlated with seasonal precipitation. Prey abundance was also correlated with annual precipitation. Individual foraging success was found to be correlated with estimated prey abundance. Parameters of logistic-by-weight growth models were compared for the 3 yr of the study and revealed that variation in food availability resulted in predictable variation in individual growth rates. Individual growth rates were positively correlated with active-season and annual precipitation and with estimates of food availability. The limitations of using stationary growth models to describe a potentially nonstationary process such as individual growth are discussed.

Bergmann's Clines in Ectotherms: Illustrating a Life-History Perspective with Sceloporine Lizards

The generality and causes of Bergmann’s rule have been debated vigorously in the last few years, but Bergmann’s clines are rarely explained in the context of life‐history theory. We used both traditional and phylogenetic comparative analyses to explore the causes of latitudinal and thermal clines in the body size of the eastern fence lizard (Sceloporus undulatus). The proximate mechanism for larger body sizes in colder environments is delayed maturation, which results in a greater fecundity but a lower survival to maturity. Life‐history theory predicts that a higher survivorship of juveniles in colder environments can favor the evolution of a Bergmann’s cline. Consistent with this theory, lizards in colder environments survive better as juveniles and delay maturation until reaching a larger body size than that of lizards in warmer environments. We expect similar relationships among temperature, survivorship, and age/size at maturity exist in other ectotherms that exhibit Bergmann’s clines. However, life‐history traits of S. undulatus were more strongly related to latitude than they were to temperature, indicating that both abiotic and biotic factors should be considered as causes of Bergmann’s clines. Nonetheless, analyses of the costs and benefits of particular body sizes in different thermal environments will enhance our understanding of geographic variation.

Elevational covariation in environmental constraints and life histories of the desert lizard Sceloporus merriami

We examine environmental constraints on life history characters among three elevationally distinct populations of the desert lizard Sceloporus merriami in west Texas. We show how environmental gradients in temperature and food abundance interact to constrain body temperatures, daily activity times, growth rates, and age-specific body size. We suggest that these differences resulted in opposite responses from males and females with respect to their size and age at first reproduction. The highest elevation site, Maple Canyon (1609 m), has more rainfall, higher food availability, and cooler temperatures compared with the lower elevation sites, Grapevine ( 1036 m) and Boquillas (560 m). Lizards were active throughout the day at Maple Canyon, but were restricted to midday inactivity at the lower sites. Body temperatures (Tb) were higher during inactivity and late-afternoon activity at the lower sites. We suggest that low food availability, fewer hours to forage, and higher Tb constrained the energy budgets of lizards at Boquillas. Consequently, these lizards had low activity rates and low daily growth rates. Paradoxically, lizards at Maple Canyon, the most food-rich site, also exhibited low daily growth rates. This may have resulted from high daily activity expenditure (longer activity period) and lower food processing rate (low inactive Tb). Grapevine lizards had the highest individual activity and growth rates. On a seasonal basis, the pattern of activity time was reversed, and Maple Canyon lizards had a shorter annual active season by 1 mo compared with Grapevine and 2 mo compared with Boquillas. Despite low growth rates at Boquillas, a 2-mo longer activity season proxi- mally resulted in the largest yearlings by May-June. Maple Canyon yearlings grew at a similar rate, but over a shorter time period and were the smallest. At Grapevine, high summer growth rates resulted in large yearlings by late summer and the largest adults, whereas low summer growth rates at Boquillas and Maple Canyon led to smaller adults. The resulting difference in body size between yearling and older age classes was great at Maple Canyon and at Grapevine, which forced yearling males to delay reproduction until they had grown to a size to compete successfully for breeding territories. However, Boquillas yearlings and age > 1 yr lizards were much more similar in size, and many yearlings were able to acquire breeding territories. This suggests that male age of first reproduction was lowest at the low elevation site. In contrast, more yearling females reproduced earlier at the highest elevation site than at other sites. At Grapevine and Boquillas, females similar in size to breeding females at Maple Canyon did not breed in May-June and instead grew and were reproductively active in July-August. We hypothesize that the brief growing season at Maple Canyon may have constrained females to breed early to allow sufficient time for themselves, as well as for their offspring, to amass winter stores. Also, this would increase the likelihood that their offspring were large enough to breed as yearlings. Results suggest an interaction between resource levels and biophysical constraints that may greatly influence differences among populations in important life history character- istics. Although these responses are hypothesized to be proximally induced by environ- mental constraints, the resulting life history differences in age-specific resource allocation to growth, storage, and reproduction may significantly affect fitness.

Variation in locomotor performance in demographically known populations of the lizard Sceloporus merriami

We examined correlates (sex, size, age) of variation in maximal burst speed (racetrack) and in stamina (time until exhaustion at 0.5 km · h⁻¹) in two populations of the iguanid lizard Sceloporus merriami in Big Bend National Park, Texas. Because these populations were known demographically and were studied in several years, we were able to examine the effects of adult age on performance, the between-year repeatability of individual performance, and environmental correlates of yearly variation in performance. Variation in locomotor capacities was striking. Males were faster than females, and lizards from the Grapevine Hills were faster than those from Boquillas (lower elevation). However, sex and population did not affect stamina. Young adults (1-yr-olds) were slightly faster but had slightly less stamina than did older adults (2-3-yr-olds). Adult size (mass, snout-vent length, hind limb length) had little effect on performance. Average performance varied significantly among years, but stamina varied more than did speed. Stamina was highest in a summer preceded by relatively wet fall and winter. Despite the between-year variation in average performance, individual speed and stamina are both significantly repeatable over at least 1 yr. Neither tail loss nor toe loss affected sprint speed, but body temperature had a marked effect on speed.

The relationship of body size to survivorship of hatchling snapping turtles (Chelydra serpentina): an evaluation of the “bigger is better” hypothesis

Abstract In many organisms, body size is positively correlated with traits that are presumably related to fitness. If directional selection frequently favors larger offspring (the “bigger is better” hypothesis), the results of such selection should be detectable with field experiments. We tested the “bigger is better” hypothesis in hatchling snapping turtles (Chelydra serpentina) by conducting one long-term and three short-term experiments on the University of Michigan E.S. George Reserve in southeastern Michigan. In the fall of 1995 and 1996, we released hatchlings at artificial nests separated from the nearest wetland by fences. We recorded the proportion of hatchlings recaptured, the time it took hatchlings to move to fences from artificial nests 45, 55, and 80 m away, and dispersion along the fence. We determined whether the response variables and probability of recapture at fences were associated with hatchling body size. During 1995, average travel times of hatchlings from the experimental nests were not related to distance from the fence; however, time to recapture was positively correlated with dispersion from the zero point on the fence, and the maximum time to reach the fence was almost twice as long for hatchlings from the 80-m nest compared to those from the 45-m nest. Sixty-seven percent of the hatchlings reached the fence and the proportions doing so from each nest were not different. Body size was not significantly related to probability of recapture in either of the 1995 experiments. In 1996, 59% of released hatchlings were recaptured. Time to recapture was not related to dispersion from the zero point or to body size. Cubic spline analysis suggested stabilizing selection on body size. We also conducted a set of long-term hatchling release experiments between 1980–1993 to compare the survival of hatchlings released at nest sites to that of hatchlings released directly into marshes, and we looked for relationships between survivorship and hatchling body size. During 7 years in which more than 30 hatchlings were released, 413 hatchlings were released directly into the marsh and 262 were released at nests: their probability of survival did not differ. Over all years, for both release groups combined and for each group separately, survival was not related to body size. In 1983 alone, survival was also not related to body size for either group or for both groups combined. In our three short-term experiments and one long-term experiment, we found no evidence to support the “bigger is better” hypothesis. When selection on body size did occur, selection was stabilizing, not directional for larger size.

Optimal Diet as a Function of Absolute Abundance, Relative Abundance, and Relative Value of Available Prey

We have presented a model that specifies the optimal diet of a predator faced with a range of kinds of potential prey. The optimal diet is defined to be the set of kinds of prey which if eaten whenever encountered will maximize the intake of food value per unit time. The parameters required to determine the optimal diet are the absolute abundance of potential food, the relative values to the predator of potential prey types, and the relative abundances of the potential prey types. Predictions of the model are that: no diet can be optimal if it excludes a prey type more valuable than any of the included prey types; number of prey types eaten by a predator which is feeding optimally for a fixed relative value and relative abundance of prey types decreases as the absolute abundance of all potential food increases; determination of optimal diet is not independent of the relative abundances of the possible prey; but a small change in relative value is more effective in changing the optimal diet than is a small change in relative abundances of the potential prey. Relative abundance of the potential food types is not very important in determining the optimal diet; relative value of prey types is quite important; but absolute abundance of food is the overriding parameter in the determination of optimal diet.

Size at Metamorphosis and Water Loss in a Desert Anuran (Scaphiopus couchii)

We measured net water flux and tolerance to dehydration as a function of size at metamorphosis in the desert anuran Scaphiopus couchii (Anura: Pelobatidae). Experimental conditions were based on extensive sampling in the field of environmental conditions available to and encountered by recently metamorphosed toadlets. Size had little effect on dehydration tolerance. Nevertheless, because larger toadlets lost water at a lower mass-specific rate, they had an advantage in the time available before reaching a critical level of dehydration. Field studies also provide a context for interpreting physiological differences in water loss. In the desert, moist soil has a patchy distribution in space and time. Larger metamorphs, because they can survive longer in the intervening dry areas, may have more time to locate suitable microhabitats.

Life History and Demographic Variation in the Lizard Sceloporus Graciosus: A Long‐Term Study

An 1 1-yr study of life history and demographic variation in the sagebrush lizard Sceloporus graciosus was carried out on two study areas (Rattlesnake Ridge and Ponderosa Flat) in the Kolob Mesa Section of Zion National Park, Utah. Two primary objectives of this mark-recapture study were to: (1) quantify variation in age structure, age and size at maturity, age-specific survivorship and fecundity, and individual growth rates, and (2) conduct a series of density reduction experiments designed to elucidate the effects of density on growth rates and survival of posthatchling lizards. In addition, we examined the relationships of variation in population density and deviation from long-term average precipitation and temperature to variation in individual growth, reproduction, and demography. At both sites the active season was - 160 d, extending from early April to mid-September. Reproduction occurred during a 50-d period between mid-May and early July. Mean clutch size was 3.7 eggs and most females produced their first clutch in the 2nd yr of life (their third active season) at an age of - 22-24 mo and a minimum snout-vent length of - 50 mm. Most mature females produced two clutches of eggs per year, and there was no statistically significant variation in either mean clutch size or body-size-adjusted clutch size among the 11 yr of study. Clutch size was significantly correlated with body size. Relative clutch mass averaged 0.247 and was not significantly correlated with body size. Since hatchlings first appeared in early to mid-August, their first growing season was -2 mo long. There was no significant sexual dimorphism in growth rate or body size in either population. There was great variation in estimates of egg-yearling survival among years. Egg-yearling survival probability varied from 0.12 to 0.59 with a mean of 0.28. At Ponderosa Flat, the survival of yearling males (0.38) was significantly lower than that of yearling females (0.47). Survival of yearling males (0.45) and females (0.43) at Rattlesnake Ridge was not significantly different. There were no other significant differences in the survival of males and females (X = 0.56 for both sexes) within any age class in any year of the study. However, the survival of yearlings was significantly lower than that of older lizards in both populations. Mean posthatchling survival over all years was 0.45, and there was significant heterogeneity in posthatchling survival among years. Average annual survival of immigrants (0.32) was significantly lower than that of residents (0.44). There was a significant negative linear relatonship between yearling body size in late June and total density of posthatchling lizards. A stepwise linear regression model revealed significant effects of both rainfall (and presumably resource availability) and population density on the growth of yearlings. This model explained 78% of the annual variation in yearling growth. Rank correlation analysis revealed that survivorship of hatchlings was negatively correlated with density of nonspecific lizards. The negative correlation implies direct density dependence of hatchling mortality rates and is a potentially important mechanism of population regulation. Removals of almost all yearling and older age lizards from the study sites resulted in significant increases in growth rates of hatchlings in the year of the removal and yearlings during the following year. Four results from this study combine to suggest substantial resource limitation of S. graciosus on the Kolob Mesa. (1) Snout-vent lengths attained by yearling lizards were positively correlated with deviations from long- term mean rainfall values. (2) Body sizes attained by yearlings were greatest in the years following density reductions. (3) Body size attained by yearlings was negatively correlated with density of conspecifics. And (4), in a year in which a density reduction followed a warm, wet spring, more yearling females reached maturity than in all other years of the study combined. Finally, we compared data from the Kolob Mesa populations to those from other populations of S. graciosus.