The paper examines the factors which generate various patterns of dispersion in the distribution of parasites within their host populations. Particular emphasis is placed on the role played by chance elements in the growth and decay of parasite populations and on the influence of different types of demographic processes. It is argued that observed distributions are dynamic, rather than static, entities generated by opposing forces, some acting to create over-dispersion and others acting to generate under-dispersion. Monte Carlo simulation experiments, based on probability models of the growth and decay of host and parasite populations, are used to study the dynamics of parasite dispersion. Attention is specifically focused on the role played by parasite-induced host mortality. It is shown that, for certain types of host-parasite associations, convex curves of mean parasite abundance in relation to age (age-intensity curves), concomitant with a decline in the degree of dispersion in the older age, classes of hosts, may be evidence of the induction in host mortality by parasite infection. Empirical evidence is examined in light of this prediction. In general, however, simulation studies highlight the technical difficulties inherent in establishing clear evidence of parasite-induced host mortality from ecological studies of hosts and parasites in their natural habitats.