Consequences of diet choice by a small generalist herbivore

We examined the energy, nutrient, time, and habitat consequences of diet ch oice by a generalist herbivore, the hispid cotton rat (Sigmodon hispidus) O n the Texas coastal prairie. Benefit-cost and linear programming models wer e developed and used to test several specific hypotheses about diet choice, foraging strategy, and habitat use by cotton rats. Within cotton rat habit at, there were significant differences among both seasons and habitat patch es in the abundance and biomass of foods available. Monocot patches typical ly were higher in standing crops of energy and nutrients than were either d icot or mixed patches. Relative to available biomass and nutrients in dicot foods, this cotton rat habitat was both spatially and temporally heterogen eous and dynamic. Although the abundance of dicot foods in a habitat patch did influence costs of foraging time, time costs of cotton rats foraging fo r suitable foods did not decrease when standing crops of energy and nutrien ts in these habitats increased. Higher requirements for energy and nutrient s needed for lactation by reproductive cotton rats resulted in higher forag ing costs than for nonreproductive female rats. A mixed diet incorporating both monocot and dicot foods was nutritionally superior to either a monocot -only diet or a dicot-only diet. A diet of only monocots would meet minimum daily energy requirements and would not exceed either maximum daily digest ive capacity or maximum daily foraging time, but it would not meet minimum daily requirements for protein and phosphorus during reproduction. Converse ly, a diet of only dicots would meet nutrient requirements for reproduction but would exceed maximum daily foraging time. Mixing of monocot and dicot plant parts in the diet met both net energy and nutritional requirements, a nd increased foraging lime, as compared to a diet of only monocots. Diets o f cotton rats foraging to maximize net energy intake each season differed f rom the observed diets. Likewise, diets of cotton rats foraging to minimize total daily foraging time differed from the observed diets. Thus, we concl ude that reproductive females of this generalist herbivore neither maximize d their energy nor minimized their time, but rather, consumed a mixed diet to balance intake of requisite nutrients. Accordingly, a diet choice that i ncluded a mixture of monocot and dicot foods was most efficient when foragi ng occurred in a habitat having a mixture of monocot and dicot plants. Mixe d patches provided higher energy and nutrient benefits to foraging cotton r ats than did either monocot or dicot patches. Finally, we observed that rep roductive female cotton rats concentrated their activities in mixed-patch h abitats.