Pulse vaccination, the repeated application of vaccine over a defined age range, is gaining prominence as a strategy for the elimination of childhood viral infections such as measles and polio. However, unlike routine or continuous mass infant immunization, epidemiological understanding of this control method is in its infancy. This paper develops initial work by Agur et al. (1993) using simple steady-state and age-structured dynamic models to extend the theory of the mechanism of action of pulse vaccination, and to explore the relationship between the maximum permitted interval between pulses and key epidemiological, demographic and vaccination variables. Initially, a conceptual model is presented to illustrate the principles of pulse vaccination and to make comparison with routine immunization procedures. An ordinary differential equation model, which assumes homogeneous mixing, is then used to derive equilibrium expressions for the pulse interval in relation to (i) different demographic profiles, (ii) population growth characteristics (stationary or exponen tially increasing), (iii) combined routine and pulse immunization, and (iv) the age range vaccinated. Finally, simulations using age-structured compartmental determin istic models illustrate complex epidemiological dynamics associated with pulse vaccination, particularly where there is age heterogeneity in contact rates in the population. The resultant uncertainty in defining an optimal pulse interval raises concerns of a practical nature.