The idea of optimisation in animals: Uses and dangers

The idea that animals may, in some sense, be usefully regarded as optimising systems is discussed. The application of this idea to the problems of the prediction of food intake and diet selection is outlined. It is concluded that it provides a useful framework within which to theorise. Feeding systems are described and their problems identified. Within the system of a single food given ad libitum the problem is to predict the rate at which the animal will eat the food that it is given. The conventional framework within which this problem is approached for ruminants, particularly when on foods of lower quality, is described. Recent criticisms of this framework are outlined and discussed. A new framework has recently been proposed. This takes the view that a ruminant animal will eat a particular food at the rate that maximises the efficiency with which it will use oxygen. This theory is briefly described and criticised. In three important ways the predictions of the theory are found to be contradicted by experimental evidence. These are: (i) the failure to deal with the changes in intake of a given food with degree of maturity of a particular animal; (ii) the necessary, but apparently false, assumption of a diminishing marginal efficiency in energetic efficiency above maintenance; and (iii) the very small rate of change in the efficiency of using oxygen as dry matter intake changes around that which gives the maximum efficiency. It is concluded that there are both opportunities and dangers in using the idea of optimisation.