Diet Breadth of Mammalian Herbivores: Nutrient versus Detoxification Constraints

Two hypotheses, nutrient constraints and detoxification limitation, have been proposed to explain the lack of specialists among mammalian herbivores. The nutrient constraint hypothesis proposes that dietary specialization in mammalian herbivores is rare because no one plant can provide all requisite nutrients. The detoxification limitation hypothesis suggests that the mammalian detoxification system is incapable of detoxifying high doses of similar secondary compounds present in a diet of a single plant species. We experimentally tested these hypotheses by comparing the performance of specialist and generalist woodrats (Neotoma) on a variety of dietary challenges. Neotoma stephensi is a narrow dietary specialist with a single species, one-seeded juniper, Juniperus monosperma, comprising 85-95% of its diet. Compared with other plants available in the habitat, juniper is low in nitrogen and high in fiber, phenolics, and monoterpenes. The generalist woodrat, N. albigula, also consumes one-seeded juniper, but to a lesser degree. The nutrient constraint hypothesis was examined by feeding both species of woodrats a low-nitrogen, high-fiber diet similar to that found in juniper. We found no differences in body mass change, or apparent digestibility of dry matter or nitrogen between the two species of woodrats after 35 days on this diet. Moreover, both species were in positive nitrogen balance. We tested the detoxification limitation hypothesis by comparing the performance of the generalist and specialist on diets with and without juniper leaves, the preferred foliage of the specialist, as well as on diets with and without α-pinene, the predominant monoterpene in juniper. We found that on the juniper diet, compared with the specialist, the generalist consumed less juniper and lost more mass. Urine pH, a general indicator of overall detoxification processes, declined in both groups on the juniper diet. The generalist consumed half the toxin load of the specialist yet its urine pH was slightly lower. Moreover, the generalist consumed significantly less of the treatment with high concentrations of α-pinene compared to the control treatment, while the specialist consumed the same amount of food regardless of α-pinene concentration. For both groups, urine pH declined as levels of α-pinene in the diet increased. The generalist produced a significantly more acidic urine than the specialist on the treatment with the highest α-pinene concentration. Our results suggest that