The assumptions of the Random Predator equation are discussed and the problems associated with parameter estimation and prediction are pointed out.
Experiments are described in which the functional response characteristics of a mesostigmatid mite Pergamasus crassipes, preying upon a collembolan, Onycbiurus armatus, were observed over four successive time intervals. In a series of separate experiments the handling-time, Th, was estimated directly.
A modification of the random predator equation is proposed in which a searching time sub-model is incorporated. Predictions from this model suggest that the attack rate is not a constant but is a function of prey density. An attack rate sub model is formulated and the resultant functional response model is shown to be more accurate and more readily usable for prediction than the random predator equation.