The effects of seasonality and other temporal patterns on the occurrence of rotavirus diarrhoea were studied among hospitalised cases at Pune, India from July 1992 to June 1996. The well-accepted Box-Jenkins methodology based on modelling was employed for the analysis. This is the first presentation of such analysis for rotavirus diarrhoea. The model suggests strong influence of climatic changes on the incidence of the disease. Further study of weather parameters not only confirms that daily minimum temperature is the principal factor but also reveals that easterly wave, a characteristic feature of tropical weather, is useful in predicting the peak of hospital admissions and the geographical sequence of outbreaks of the disease in tropical India.