A double-blind clinical trial was conducted in a non-fluoridated community to determine the effect on enamel fluoride and caries experience of daily rinsing in school with 1,000 parts/10(6) solutions of ammonium fluoride or sodium fluoride at pH 4.4. Subjects were 10- to 12-year-old children (n approximately equal to 200/group at baseline), about one-half of whom reported the usage of fluoride supplements. Dental caries (DFS index) and enamel fluoride (in vivo biopsy) were evaluated at baseline, 12 months, and 24 months. Supplement users had higher enamel fluoride levels and less caries experience initially, as well as generally lower caries increments over the study. In year 1, the overall caries reductions (supplement users and non-users combined) were 23 % (ammonium fluoride) and 33 % (sodium fluoride), P less than 0.01. For year 2, treatment effects were significantly greater: 54 % (ammonium fluoride) and 47 % (sodium fluoride). In newly erupted teeth, the effects of the ammonium fluoride (70 % DFS reduction) was significantly greater (P = 0.013) than that of the sodium fluoride (48 % DFS reduction). Enamel fluoride levels at the end of 2 years were 3,124 parts/10(6) (ammonium fluoride), 2,771 parts/10(6) (sodium fluoride), and 2,603 parts/10(6) (placebo), P = 0.025.