The aim of this study was to investigate whether differences in surface treatment prior to repair influenced the seal of a resin fissure sealant placed on the occlusal surfaces of permanent molar teeth. In-vitro study. One hundred and twenty-eight extracted human first and second molars were randomly allocated to one of four groups of 32 teeth each. A light cured, unfilled, opaque resin fissure sealant (Delton) was placed on their occlusal surface following cleaning by prophylaxis and acid etching. Following storage in artificial saliva (Saliva Orthana) for a week, duplication of sealant failure was carried out. The teeth were then subjected to one of four different surface treatments: Group 1: with a slow-speed prophylaxis brush followed by acid etching (control method); Group 2: a slow-speed bur and acid etching; Group 3: air abrasion and acid etching; and Group 4: acid etching and application of a bonding agent. Following a further week of storage in artificial saliva two layers of impermeable varnish were applied to the nonocclusal surfaces of the teeth; their apices were then sealed with wax and the teeth were immersed in 1% methylene blue for 48 h. The teeth were then sectioned (ISOMET 1000) to achieve three cuts resulting in a maximum of four blocks, i.e. six surfaces per tooth. A total of 715 sections from 126 teeth were scored for microleakage on the intact and repaired side of the fissure sealant. Statistical analysis did not demonstrate any one single method of repair to be superior to the control method for reapplication of the sealant. All four techniques compared in this study seem to be acceptable for replacing or repairing lost or fractured fissure sealants. As prophylaxis with a brush rotating at slow speed followed by acid etching, which probably represents current practice, is also the simplest technique that can be practised on children, it is therefore recommended.