This in vitro study evaluated the amount of fluoride released from fluoride-containing materials over a period of 28 days. Six disk samples (2.06 +/- 0.06 cm2) were prepared of each material and divided at random into seven groups: Group 1: Chelon-Fil; Group 2: Chelon-Silver; Group 3: VariGlass; Group 4: Dyract; Group 5: Vitremer; Group 6: Vitremer + Scotchbond Multi-Purpose; Group 7: Fuji II LC. The cements were mixed according to the manufacturers' recommendations, placed in plastic molds, and pressed between two glass plates. Paraffined dental floss was incorporated into the cements during setting to serve as attachments. The materials in Groups 3, 4, and 5 were light cured (Heliolux) in two different positions for 40 seconds each. In Group 6, the adhesive was light cured in two different positions on both sides for 10 seconds each. The samples were stored at 100% relative humidity for 24 hours. Each sample was then suspended in individual plastic tubes containing 5 ml of deionized water and submitted to constant agitation at 25 degrees C. The water was changed every 24 hours. Fluoride release was determined at 1, 2, 3, 4, 5, 7, 14, and 28 days after buffering the solution with equal volume of TISAB. Fluoride release was measured with a fluoride ion-specific electrode (Orion 96-09) and an ionanalyzer (Orion EA 940) previously calibrated with standard solutions containing 0.05 to 5.00 micrograms F/ml. Fluoride release was expressed as ppm in solution and micrograms F/cm2. ANOVA and Student-Newman-Keuls tests were used to evaluate the data. The results revealed that Chelon-Fil released significantly (P < 0.001) more fluoride for the first 7 days than all the other products. This was followed by Fuji II LC, which exhibited significantly more fluoride release than the rest of the materials for the same 7 days. At days 14 and 28, Chelon-Fil, Dyract, and Fuji II LC released similar amounts of fluoride that were significantly greater than the other products. Group 6 (Vitremer + Scotchbond Multi-Purpose) released significantly less fluoride than the other materials at all time intervals. Fluoride release for all products at days 1 and 2 was significantly greater than the rest of the time intervals, except for Chelon-Silver, which released similar amounts of fluoride for days 2, 3, 4, and 5. Although significance for the remaining time intervals varied for all materials, all fluoride release decreased from day 1 to day 28.