This study was carried out to obtain reliable prospective data on the incidence, home management and duration of infantile diarrhoeal disease in a developing and rapidly urbanizing community. Fourteen hundred and seventy six infants from 4 different communities of Lahore, representing different stages of urbanization, were followed prospectively from birth to 2 years of age. Each child had an average of 3.6 episodes of diarrhoea during one year; 4.3 episodes per child per year in the village, 4.4 in the periurban slum, 3.4 in the urban slum and 1.4 in the upper middle (Um) class control group. The maximum incidence of diarrhoeal episodes occurred in children between 9-10 months of age. No sex difference was found. The seasonal variation showed a peak incidence during April-June. Bloody diarrhoea, in 3-5% of the episodes, was predominantly seen in winter months and in older infants. The maximum use of ORS was seen in the Um class (53%). Antibiotics were used in 75% of all diarrhoeal episodes maximally in the Um class, and morphine derivatives in 1% of all cases. The overall incidence of persistent diarrhoea was 14% of all acute cases, being highest in the village (18%), followed by the periurban slum (14%), Um class (10%) and urban slum (8%) with peak incidence occurring in June and July. Our results confirm the magnitude of the problem of diarrhoea.(ABSTRACT TRUNCATED AT 250 WORDS)