Mouth rinses with 7 ml 0.05% NaF containing 51Cr-EDTA (2 μCi) were performed by 11 dental staff members. The water-soluble non-absorbable 51Cr-EDTA was added to enable estimations of the amount of fluid swallowed and the quantity of fluoride retained and/or absorbed in the oral cavity. Experiment 1. 2-min rinses with 7 ml test solution were followed by three 1-min rinses with 10 ml water. The recoveries of 51Cr-EDTA and fluoride in the first expectorate were 86.3 ± 4.0% and 83.9 ± 4.1% (n = 11), respectively. After the succeeding water rinses, a difference of 1.8 ± 0.9% remained, which represents the fraction of fluoride being absorbed from and/or retained more firmly in the oral cavity. The amount of 51Cr-EDTA not recovered from the expectorates after this rinsing procedure (5.3 ± 3.0%) indicates the fraction of the test solution swallowed. Experiment 2. This was similar to experiment 1, except that the rinsing with test solution lasted only 30 sec. No significant difference in the total recoveries was observed (0.0 ± 1.0%), and only 1.7 ± 1.6% was estimated to have been swallowed. The findings indicate that (1) negligible parts of the retained fluoride is absorbed from the oral cavity, (2) 2-min vs. 30-sec rinsing periods increased the fraction of fluid swallowed, (3) after 2-min rinses, in contrast to 30-sec rinses, the oral cavity retained a significant fraction of fluoride not removable by the water rinses.