Effects of oxygen on responses to heating in two lizard species sampled along an elevational gradient.

Thermal tolerance is an important variable in predictive models about the effects of global climate change on species distributions, yet the physiological mechanisms responsible for reduced performance at high temperatures in air-breathing vertebrates are not clear. We conducted an experiment to examine how oxygen affects three variables exhibited by ectotherms as they heat-gaping threshold, panting threshold, and loss of righting response (the latter indicating the critical thermal maximum)-in two lizard species along an elevational (and therefore environmental oxygen partial pressure) gradient. Oxygen partial pressure did not impact these variables in either species. We also exposed lizards at each elevation to severely hypoxic gas to evaluate their responses to hypoxia. Severely low oxygen partial pressure treatments significantly reduced the gaping threshold, panting threshold, and critical thermal maximum. Further, under these extreme hypoxic conditions, these variables were strongly and positively related to partial pressure of oxygen. In an elevation where both species overlapped, the thermal tolerance of the high elevation species was less affected by hypoxia than that of the low elevation species, suggesting the high elevation species may be adapted to lower oxygen partial pressures. In the high elevation species, female lizards had higher thermal tolerance than males. Our data suggest that oxygen impacts the thermal tolerance of lizards, but only under severely hypoxic conditions, possibly as a result of hypoxia-induced anapyrexia.

Oxygen concentration affects upper thermal tolerance in a terrestrial vertebrate.

We tested the oxygen limitation hypothesis, which states that animals decline in performance and reach the upper limits of their thermal tolerance when the metabolic demand for oxygen at high temperatures exceeds the circulatory system's ability to supply adequate oxygen, in air-breathing lizards exposed to air with different oxygen concentrations. Lizards exposed to hypoxic air (6% O2) gaped, panted, and lost their righting response at significantly lower temperatures than lizards exposed to normoxic (21% O2) or hyperoxic (35% O2) air. A greater proportion of lizards in the hyperoxic treatment were able to withstand body temperatures above 44°C than in the normoxic treatment. We also found that female lizards had a higher panting threshold than male lizards, while sex had no effect on gaping threshold and loss of righting response. Body size affected the temperature at which lizards lost the righting response, with larger lizards losing the response at lower temperatures than smaller lizards when exposed to hypoxic conditions. These data suggest that oxygen limitation plays a mechanistic role in the thermal tolerance of lizards.

Body size impacts critical thermal maximum measurements in lizards.

Understanding the mechanisms behind critical thermal maxima (CTmax; the high body temperature at which neuromuscular coordination is lost) of organisms is central to understanding ectotherm thermal tolerance. Body size is an often overlooked variable that may affect interpretation of CTmax, and consequently, how CTmax is used to evaluate mechanistic hypotheses of thermal tolerance. We tested the hypothesis that body size affects CTmax and its interpretation in two experimental contexts. First, in four Sceloporus species, we examined how inter- and intraspecific variation in body size affected CTmax at normoxic and experimentally induced hypoxic conditions, and cloacal heating rate under normoxic conditions. Negative relationships between body size and CTmax were exaggerated in larger species, and hypoxia-related reductions in CTmax were unaffected by body size. Smaller individuals had faster cloacal heating rates and higher CTmax, and variation in cloacal heating rate affected CTmax in the largest species. Second, we examined how body size interacted with the location of body temperature measurements (i.e., cloaca vs. brain) in Sceloporus occidentalis, then compared this in living and deceased lizards. Brain temperatures were consistently lower than cloacal temperatures. Smaller lizards had larger brain-cloacal temperature differences than larger lizards, due to a slower cloacal heating rate in large lizards. Both live and dead lizards had lower brain than cloacal temperatures, suggesting living lizards do not actively maintain lower brain temperatures when they cannot pant. Thermal inertia influences CTmax data in complex ways, and body size should therefore be considered in studies involving CTmax data on species with variable sizes.