Herpes-zoster is caused by the reactivation of varicella-zoster virus (VZV). In this paper different hypotheses of how this re-emergence of virus comes about are reviewed and discussed. From these hypotheses, and epidemiological data describing the initial transmission of the virus, a mathematical model of primary disease (varicella) and reactivated disease (zoster) in developed countries is derived. The steady-state age distributions of zoster cases predicted by this model are compared with the observed distribution, derived from a review and analysis of published epidemiological data. The model allows differentiation between published hypotheses in which age of host may or may not influence the probability of viral reactivation. The results indicate that the probability of reactivation must increase with age to allow the observed pattern of zoster cases. The basic mathematical model presented provides a conceptual framework, which may be extended to assess possible control programmes.