In order to investigate the characteristics of hepatitis A virus excretion, 372 children in semi-closed day-care centres were examined at the time of the seasonal increase in hepatitis incidence. Enzyme immunoassay and immune electron microscopy were used to detect virus in faeces. Ten communities were surveyed and, for the purposes of this paper, were divided into four groups on the basis of the observed pattern of virus excretion. The first group consisted of 4 communities with one hepatitis case in each; in this group, there was a relatively low rate of virus excretion throughout the study. The second group comprised 2 communities with one or two hepatitis cases; this group was characterized by an increase in the rate of virus excretion some time after isolation of the patient. In the communities of the third group, there were numerous hepatitis cases and a high rate of virus excretion during the first weeks of observation, followed by a significant decrease. There were no hepatitis cases in the fourth group, but some hepatitis A virus shedding was detected.The ratio of the number of cases to the number of virus carriers was 1:11.4 in the first two groups, and 1:3.4 in the third. Administration of normal serum immunoglobulin did not significantly affect the rate of HAV excretion. The cause of the differences in shedding of hepatitis A virus in the communities is unknown, but may possibly be related to the size of the immune section of the population.