Diet selection and plant nutritional quality in Attwater's pocket gopher (Geomys attwateri)

Many above-ground and fossorial rodents are generalized herbivores, either ingesting various types of monocots and dicots to balance nutrients in their diet or avoiding some plants because they contain secondary plant metabolites. Attwater's pocket gopher (Geomys attwateri) is a common rodent in bunchgrass-annual forb communities in Texas whose diet contains both monocot and dicot plants, with a preference for perennial dicots (as compared to their availability). We reduced availability of dicots with an herbicide and predicted that, because of their importance in the diet, the proportion of dicots in the diet of G. attwateri would not be affected. We found no difference in dicot content in the diet of animals on plots with dicots reduced (DR plots) compared to plots with natural abundance of dicots (DP plots), and that preferences for dicots (compared to their availability) increased on DR plots while preferences for perennial monocots (compared to their availability) decreased. Dicot roots contained more lipid, soluble carbohydrate, Mg, K, and Ca than monocot roots. During the summer non-breeding season, dicot shoots contained more protein, lipid, soluble carbohydrate, Na, Mg, P, and Ca than monocot shoots. During the winter breeding season, content of protein, soluble carbohydrate, Na, Mg, P, K, and Ca was higher in dicot shoots than monocot shoots. Low protein content and high neutral detergent fiber content in diet plants may limit reproduction especially during the summer non-breeding season. We hypothesized that increased use of extant burrow systems, extension of burrow systems on plots with reduced availability of dicots, or extension of tunnels into dicot areas off DR plots may have enabled G. attwateri to maintain the concentration of dicots in the diet. Larger burrow systems would increase intraspecific interactions and could explain lower residency time and decreased density of G. attwateri reported by other studies on plots with low abundance of dicots. We suggest that, in spite of similarity in abundance of dicots in diets of animals on DR and DP plots, ingestion of inadequate amounts of dicots or mixtures of dicot species that did not provide sufficient concentrations of nutrients may have led to a reduction in reproduction on DR plots reported in other studies.