Stable isotope analysis is frequently applied as a tool to examine dietary patterns in animals. However, some of the underlying assumptions associated with using this approach are increasingly being questioned. We carried out a controlled diet experiment on the southern African spiny mouse (Acomys spinosissimus Peters, 1852) to test a number of aspects relating to these assumptions and also examine the hypothesis that stable isotopes, especially δ15N, can be used to provide evidence of nutritional stress. We compared the δ13C and δ15N values of livers and blood from animals that were fed ad libitum with animals undergoing a 10% reduction in food supply. Food-restricted animals showed no significant difference in δ15N; however, δ13C values of both liver and blood were depleted. Restricted animals also had a significantly lower C:N ratio. We examined the role of lipids and found following lipid extraction that both livers and lipids still showed the same separation in carbon values. Tissue–diet discrimination values were also calculated and found to be higher for both Δ13C and Δ15N compared with other mice species. Empirical values for discrimination rates were then compared with values calculated using an alternative method based on employing generic values and were found to be dissimilar, suggesting the use of generic values are not always appropriate. Our results highlight the need for greater understanding of the assumptions associated with using stable isotope analysis to examine diet and we suggest that studying a single species under captive conditions presents an ideal method to begin to test these hypotheses.
The influence of food quantity on carbon and nitrogen stable isotope values in southern African spiny mice (Acomys spinosissimus)