Food Resource Partitioning by Sympatric Ungulates on Great Basin Rangeland

Abstract

The usefulness of a conceptual framework for understanding food selection by ungulates, based on four morphological parameters (body size, type of digestive system, rumino-reticular volume to body weight ratio, and mouth size), was tested by applying discriminant analysis to 194 monthly diet determinations based on microhistological fecal analysis for five sympatric species of ungulates in northeastern California and northwestern Nevada. In each season, the group means were located in the hypothesized order along the axis described by the first discriminant function: feral horse, domestic cow, domestic sheep, pronghorn, mule deer. Horse and cow diets consisted primarily of grasses. and mule deer diets consisted primarily of browse. Sheep diets were intermediate. Four browses (Artemisia spp., Cercocarpus ledifolius, Purshia tridentata, and Juniperus occidentalis) were selected as the most useful species for discriminating between animal species. The data and analyses support the hypothesized food selection framework. A conceptual framework for understanding the nutritional basis for food selection by ungulates has been outlined (Hanley 1982) and is based on four ungulate morphological parameters: (1) body and (2) type of digestive system (cecal or ruminant) determine the overall time-energy constraints within which the ungulate may forage selectively; (3) rumino-reticular volume to body weight ratio determines the type of forage the ruminant is most efficient in processing; and (4) mouth determines the ability of the ungulate to harvest selectively plant parts or individuals. Each of these parameters may be considered an important factor in determining an ungulate's relative position along a gradient ranging from highly selective browse diets to less selective graminoid diets. The framework was presented as a means of further understanding the diet selection process of large, generalist herbivores. An understanding of the reasons why ungulates select the kinds of foods that they do will yield predictive insight into problems involving competition and food resource partitioning in ungulate communities. On the basis of that framework, one should be able to predict not only the types of foods that will be selected by the members of a given array of ungulate species but also the rank of each species along a gradient reflecting the differences in degree of selectivity and types of foods selected. We were able to test the theory in this way, using diet composition data collected from five species of sympatric North American ungulates (Table 1). The specific hypothesis tested was that when ordered along a mathematically defined gradient based on plant species composition of their diets, the five ungulate species would appear in the following order: feral horse (Equus caballus), domestic cow (Bos spp.), domestic sheep (Ovis aries), pronghorn antelope (Antilocapra Authors were with the College of Forest Resources, University of Washington, Seattle, at the time of the research. Thomas Hanley's current address is U.S. Dep. Agr. Forest Service, Pacific Northwest Forest and Range Experiment Station, Forestry Sciences Laboratory, P.O. Box 909, Juneau, Alaska 99802. Kathleen Hanley's current address is 136 Behrends Ave., Juneau, Alaska 99801. The authors wish to express their sincere appreciation to C. Rex Clearly, district manager, Susanville District Office, Bureau of Land Management, for his support and interest throughout this investigation. Manuscript received July 31, 1980. americana), mule deer (Odocoileus hemionus). The reasons are as follows. From Table 1, it can be seen that we had two basic groups of animals in ternms of body weight, very large animals (horse and cow) and relatively small animals (sheep, pronghorn, and mule deer). Furthermore, the horse is a cecal digestor, while the others are ruminants. The time-energy constraints should be much more restrictive for the horse and cow than for the other species, especially so for the horse. The horse should have the least amount of time to forage selectively, followed by the cow and then the three small ruminants, which also have small mouths that enable them to be selective of plants parts and individuals consumed. Cattle and domestic sheep have very large rumens in proportion to their body weight (rumino-reticular volume to body weight ratio of about 0.250) and therefore should be well adapted to digesting a high cellulose (i.e., grass) diet. and mule deer, on the other hand, have relatively small rumens (rumino-reticular volume to body weight ratio of about 0.1 10) and therefore should be well adapted to digesting a high cell soluble and/ or high lignin (i.e., forb and browse) diet. For these reasons, sheep diets should be more similar to cattle and horse diets than should pronghorn or mule deer diets. and mule deer diets should be very similar. However, due to differences in stomach structure, pronghorn were expected to be more similar to sheep than mule deer were. Pronghorn antelope have a rumen somewhat more similar to sheep than that of deer in relative shape and size (Church 1975:24). The predicted ordering of the ruminants along a food selection gradient can be related to their ordering along a gradient of ratio of rumino-reticular volume (1) to metabolic body weight (Wkg075 ). Metabolic body weight incorporates both the absolute and relative costs of body weight; and if mouth is correlated with body weight via the time-energy constraints, then effects of mouth also are taken into account with the use of this ratio. Ruminoreticular volume to metabolic body weight ratios for the four ruminants are 1.120, 0.665, 0.338, and 0.289, for cow, sheep, pronghorn, and mule deer, respectively. Thus, the very large cecal digestor (horse) and the small ruminant with the small ruminoreticular volume (mule deer) should be positioned at opposite ends of a less selective graminoid diet to highly selective browse diet gradient, respectively, with the cow, sheep, and pronghorn positioned between these extremes. This hypothesis was tested with data from 194 determinations of plant species composition of diets selected by these 5 species of ungulates.