SUMMARYWe describe two models explaining the increase in tissue nitrogen isotope ratios ( Supplementary material available online at

SUMMARY Supplementary material available online at

Calculating body water through isotope dilution has become a useful way to nondestructively estimate body composition in many species. The most accurate estimates using this method require calibration against proximate chemical analysis of body composition for individual species, but no studies to our knowledge have calibrated this method on a hibernating mammal that seasonally undergoes dramatic changes in body composition. We use deuterium oxide to estimate total body water in captive arctic ground squirrels, Urocitellus parryii, and compare two approaches of calculating lean mass from total body water, both calibrated against lean mass based on proximate analysis. The first method uses a single tissue hydration constant to calculate lean mass from total body water; the second method uses a predictive equation to calculate lean mass from total body water and body mass. We found that the predictive equation performs better over the large range of body composition common to this species. Distillation of blood samples did not affect lean mass estimates from either calculation method. These findings indicate that isotope dilution using a predictive equation should work well as an alternative to destructive methods in other small mammals that undergo radical changes in body composition across their annual cycle.

Stable carbon isotope ratios (δC) in breath show promise as an indicator of immediate metabolic fuel utilization in animals because tissue lipids have a lower δC value than carbohydrates and proteins. Metabolic fuel consumption is often estimated using the respiratory exchange ratio (RER), which has lipid and carbohydrate boundaries, but does not differentiate between protein and mixed fuel catabolism at intermediate values. Because lipids have relatively low δC values, measurements of stable carbon isotopes in breath may help distinguish between catabolism of protein and mixed fuel that includes lipid. We measured breath δC and RER concurrently in arctic ground squirrels (Urocitellus parryii) during steady-state torpor at ambient temperatures from -2 to -26°C. As predicted, we found a correlation between RER and breath δC values; however, the range of RER in this study did not reach intermediate levels to allow further resolution of metabolic substrate use with the addition of breath δC measurements. These data suggest that breath δC values are 1.1‰ lower than lipid tissue during pure lipid metabolism. From RER, we determined that arctic ground squirrels rely on nonlipid fuel sources for a significant portion of energy during torpor (up to 37%). The shift toward nonlipid fuel sources may be influenced by adiposity of the animals in addition to thermal challenge.