A single application of a filled chemically initiated BIS-GMA sealant to permanent first molars of children ranging in age from 5 to 8 years covered the original sites in 52.4% of the 185 teeth examined after four years. Retention of sealant is greater in mandibular (64.7%) than in maxillary (42.0%) molars. The rate of loss of sealant was greatest during the first year (20%), then less than 10% per year at the second, third, and fourth recalls. In comparison with two similar studies, the retention of sealant was 20% to 30% greater. Using the parameters of percent effectiveness, net gain, and DMF scores per 100 test and control teeth, the data from the current study indicate a continued positive effect substantially greater than that in other similar studies. Complete loss of sealant does not appear to predispose that surface to caries any more than its contralateral paired surface. However, a partial loss of sealant because of abrasive wear that results in the exposure of the terminal ends of a fissure is a potential failure in that it creates an environment conducive to caries. Thus, even the well-applied sealant does not necessarily constitute permanent obturation of pits and fissures. Periodic clinical observation is necessary to determine the success or potential failure of the sealant treatment.