The aim of the present investigation was to determine the caries-protective potential of non-fluoridated and fluoridated milk and to compare the efficacy of different compounds of fluoride as additives to milk. OM rats were maintained in three experiments in a frequency-controlled feeding machine of König et al. (1968) or, during one study, in metabolic cages with diet MIT 200 for three weeks. They received (1) milk with Ca-Fluoride [solubilized by KA1-Sulfate], (2) NaF, (3) NaMFP, and (4) Na-Silicofluoride. Controls were supplied with water or NaF solution of the same concentration of 10 or 15 ppm F. In addition, raw milk was provided ad libitum for the rats in a preliminary test. At the beginning and the end of the experiments, the pH of milk, its fluoride concentration, the body weight gain, the caries score, the fluoride concentration of the outermost enamel surface, the percentage of the interproximal bacteria, and the fluoride ingestion and excretion were determined. The raw milk significantly reduced the animal caries score by around 40%. This effect was lower but reproducible under programmed feeding with milk of a reduced fatty content (1.5%). The addition of Ca-Fluoride, which was not totally ionized (6.5 ppm F), reduced the caries score again by around 40%. Increasing concentrations of NaF (5, 10, 15 ppm F), Na-Silicofluoride, or NaMFP showed similar caries-inhibiting effects without remarkable influence of the fluoride dosage used. The percentage of streptococci ranged from 30 to 60 in the fluoridated milk and control groups as well. The increasing fluoride deposition in the enamel reflected the various fluoride dosages offered. The rats receiving non-fluoridated milk or distilled water had a significantly higher incidence of dental caries than those receiving fluoridated milk. The permanent availability of fluoride during the animal tests caused a higher caries-inhibiting effect than in clinical human studies reported.