This paper uses relatively simple and deterministic mathematical models to examine the impact that different immunization policies have on the age-specific incidence of rubella and measles. Following earlier work by Knox (1980) and others, we show that immunization programmes can, under some circumstances, increase the total number of cases among older age groups; the implications for the overall incidence of measles encephalitis and of congenital rubella syndrome are examined, paying attention both to the eventual equilibrium and to the short-term effect in the first few decades after immunization is initiated. Throughout, we use data (from the U.K., and U.S.A. and other countries) both in the estimation of the epidemiological parameters in our models, and in comparison between theoretical predictions and observed facts. The conclusions defy brief summary and are set out at the end of the paper.