A cross-sectional survey on the prevalence of rampant caries in infants and a case-control study were executed to determine the prevalence of rampant caries and its etiology. A total of 2192 infants aged between 1 and 4.0 yr from nine regions of Tanzania comprised the sample of the prevalence survey. The overall prevalence of rampant caries resembling nursing caries was 6.8%. The prevalence varied considerably between 1.5% and 12.8% in the different regions and was closely associated with the varying prevalence of linear hypoplasia but not with fluoride content in drinking water. The relative risk of developing rampant caries in the presence of linear hypoplasia was 14.5. The case-control study comprised 116 cases, infants with rampant caries and 243 controls, without the disorder. Risk factors such as breast feeding frequency and duration during day and night, bottle feeding and weaning habits were assessed through a structured interview with the mother. Bottlefeeding was performed in only a minority of the infants for a short period (1.1 months) and in a low exposure frequency (0.4 daily) and was not associated with rampant caries. The frequency of sugar moments in the supplementary diet was on average 1.5 times per day and was not associated with rampant caries. The duration and frequency of breast feeding during the day was also not associated. The habit of allowing infants to sleep with the breast nipple in their mouths and the presence of linear hypoplasia were strongly associated with rampant caries. Odds ratios for these variables were 17.8 and 15.6 respectively. Based on these associations it is concluded that rampant caries in this sample was caused by the habit of allowing infants to sleep with the breast nipple in their mouths, whereas linear hypoplasia is considered to be a predisposing factor.