This in vitro study evaluated the amount of fluoride released from glass ionomer-lined amalgam restorations over a period of 1 year. Class V cavities (2 x 2 x 7 mm) were prepared on the facial and lingual surfaces of 50 extracted human molars randomly distributed into 5 groups: Group 1: No restorations; Group 2: Dispersalloy amalgam alone; Group 3: same as Group 2 except 1 mm of Ketac-Silver was placed on the axial wall before amalgam insertion; Group 4: GC Lining/amalgam; and Group 5: Miracle Mix/amalgam. After restoration, each tooth was thermocycled (100x) at 5 degrees C and 55 degrees C with a dwell time of 30 seconds for baseline fluoride release levels. The teeth were placed in a polyethylene vial containing 4 ml of deionized water. At weekly intervals, each tooth was transferred to a fresh vial. Fluoride release was measured with a fluoride ion specific electrode for 10 consecutive weeks and then again at the end of 1 year. Calibration curves for low level measurements were prepared so the readings could be expressed in micrograms/ml F. At 1 year, fluoride released in micrograms/ml was: Group 1: less than 0.08; Group 2: less than 0.08; Group 3: 0.28; Group 4: 0.68; Group 5: 1.12. An ANOVA was used to evaluate the statistical difference between the groups. At the end of 1 year, measurable amounts of fluoride were recorded for all glass ionomer-lined groups with Miracle Mix and GC Lining releasing significantly more fluoride than Ketac-Silver (P less than 0.002).