Most epidemiological models assume that disease is the inevitable outcome of infection (see Bailey, 1957). Yet as Dubos (1965) has said; ‘Throughout nature, infection without disease is the rule rather than the exception’. There are, in fact, many diseases whose distribution cannot be explained solely by a consideration of the probabilities of host parasite encounters. In these cases, a diseased state is only one possible outcome of an interaction between parasite phenotypes, host phenotypes and the external environment. Haemonchosis is an example of such a disease and has been studied extensively in quantitative terms (see Whitlock & Georgi, 1968).