Effects of grass and browse consumption on the winter mass dynamics of elk

Much of the research into herbivore ecology and evolution has focused on patterns and mechanisms of niche partitioning in the diets of specialized grazers and browsers, but the significance of the balance of grazing and browsing within intermediate feeders present unique questions that have received less attention. We explored the nutritional effects of varying the balance of grass and browse in the winter diets of elk (Cervus elaphus). We compiled published data from three similar studies that monitored the mass dynamics of captive elk fed diets of pure grass, pure browse or 1 of 14 mixed diets in winter feeding trials. Elk lost mass (up to 22% of initial body weight) in 29 of 33 feeding trials, similar to wild elk in winter. We used regression models of mass dynamics, considering the linear, quadratic or logarithmic effects of the proportion of the diet that was grass ( approximately 1 - proportion of the diet that was browse) and the additive and interactive effects of nitrogen intake. Diet composition had strong effects on mass dynamics, and all models explained > or =73% of the variation (adjusted r (2)) in mass dynamics. Nitrogen intake had uniformly positive effects on mass balance, and increasing grass intake caused improved maintenance of body mass up to a point, but further increases in the proportion of grass in the diet had neutral or negative effects on body mass. Overall, the data suggest that elk are adapted to consuming mixed diets. Nonetheless, data on the foraging behavior and diet selection of wild elk in many populations show that elk often consume grass-dominated diets and sometimes consume browse-dominated diets, but rarely consume mixed diets. Physiological adaptations to mixed diets may place unique spatio-temporal constraints on diet selection in intermediate feeders and impose large penalties for a sub-optimal balance of grazing and browsing.