Optimal Food Selection: A Graphic Model

Abstract

A graphic model for optimal food selection has been developed. The model as analyzed in the present paper applies to herbivores feeding on abundant food items. Contrary to several current approaches to optimal feeding behavior, our model does not require an explicit ranking of food items. The currency to be optimized may be any nutritive element or a combination thereof. The substitution of specific currencies represents potentially rejectable hypotheses. Due to the availability of data the model is most easily applicable with digestable energy as currency. For each food type, an absolute distribution curve (measured in density units) is constructed over the range of food values, i.e., the currency to be optimized. Based on this distribution function, another distribution function is constructed over the same range of food values. This latter function represents the potentially extractable net amount of nutrients as a function of food value. A broad food niche is predicted whenever the optimization function (i.e., the distribution of the total potentially extractable net amount of nutrients from all food types as a function of food value) has a broad peak. In a stable environment, a stable polymorphism is predicted whenever the optimization function has two peaks of equal height. Discussions of how search images may alter the optimization function are presented. Cyclically occurring changes of resources may also produce a polymorphic population. Shifts of diet as a result of changing intra- and interspecific competition as well as changes in the food supply due to other factors are discussed. The conclusions are compared with similar conclusions deduced on the basis of other theories on food selection. The major disagreements between our conclusions and other conclusions reported in literature, are as follows. 1. The optimal diet is determined both by the total food density and the relative food density. Thus, a food item not in the optimal diet may be added to the diet by becoming very abundant. 2. MacArthur and Pianka (1966) deduced that food items worth exploiting in the absence of competition are always worth eating if competition for the item is present. Based on our model, this does not seem to follow; instead the opposite may occur. Other conclusions regarding optimal diets, food niche breadth, etc., are consistent with current thought on the subject.