DISPARATE DETERMINANTS OF SUMMER AND WINTER DIET SELECTION OF A GENERALIST HERBIVORE, OCHOTONA PRINCEPS

The North American pika, Ochotona princeps, is a generalist herbivore that simultaneously selects two distinct diets. one consumed immediate ly (summer diet), the other harvested, transported, and stored for lat er consumption (winter diet). I investigated factors influencing diet selection at two sites on the West Knell of Niwot Ridge, Boulder Count y, Colorado during 1991 and 1992. The composition of summer and winter diets differed significantly from each other as well as from the rela tive abundance of food items in the environment. Thus, pikas were not foraging randomly for either diet. To explore winter and summer diet s election, I tested two existing hypotheses: (1) that plant morphology restricts the winter diet breadth to plants that are easily harvested and large, and thereby maximizes the amount collected per foraging eff ort, or (2) to compensate for nutrients lost during storage, pikas bia s their winter diet with high-nutrient species. I also tested the hypo thesis that plant secondary compounds may be higher in the winter diet either because they function as preservatives or because pikas delay consumption of these species until the toxins degrade. For individual dietary items, I measured energy, nitrogen, water, fiber, total phenol ic, condensed tannin, and astringency contents. There was little evide nce to suggest that morphology excluded plants from the winter diet. P lant size was not a good predictor of abundance in the winter diet. Ev en after harvesting costs had been experimentally removed, cushion pla nts were still not included in the winter diet. There was weak support for an effect of nutrients on winter diet selection; in three of four cases, the winter diet was significantly lower in water and higher in total energy content as predicted by the nutrient compensation hypoth esis. However, other nutrients exhibited no consistent pattern. Nutrie nts were not reliable predictors of the winter diet in multiple regres sion analyses. There was strong support for the hypothesis of manipula tion of secondary compounds. The winter diet was significantly higher in total phenolics and astringency. Total phenolics were consistent pr edictors of the winter diet in multiple regression analyses. The winte r diets of six additional pika populations contained plant species hig h in secondary compounds. The results suggest that pikas preferentiall y select plants with high levels of secondary compounds for their wint er diet, possibly because the presence of such compounds promotes pres ervation of the cache. This behavior may also enable the exploitation of an otherwise unusable food resource, i.e., toxic plants.