The aim of the present study was to test the caries-preventive effect of a bis-GMA type fissure sealant (Concise Enamel Bond System) in a public dental health service for children. The study compared one group of children, in which all sound permanent teeth were sealed, and a control group in which no sealants were placed. During the study period all children received systematic restorative care, while sealing was performed on the basis of diagnoses made at subsequent examinations by one of the authors. After 2 years a total of 187 children were available for re-examination. Sealant retention was comparable to that obtained in previous studies, but no statistically significant reduction in overall caries increment could be demonstrated. Further analysis showed that only 223 sites had been sealed, while the majority of all erupting sites (394) were either carious or filled before they were screened for the purpose of sealing. It is therefore assumed that only teeth less susceptible to occlusal decay were sealed, thereby explaining the lack of benefit derived from the sealing program.