Increased fluoride levels in plaque and saliva have been associated with improved protection against dental caries for dentifrices which contained sodium monofluorophosphate (Duckworth et al., 1992). The main aim of the present work was to test whether oral fluoride retention depended on F source after use of dentifrices containing either NaF or Na2FPO3. In study 1, plaque samples were collected from 474 subjects who had been using one of six test dentifrices for two years, and analyzed by F extraction with water. The dentifrices contained 1000 or 1500 micrograms F/g as either NaF or Na2FPO3. Significantly more fluoride was found in plaque from subjects who were using the NaF dentifrices than in plaque from subjects who were using Na2FPO3 dentifrices of the same F content. Subsets of plaque samples were large enough to divide into two parts for extraction by both acid and water. No significant difference was found between mean fluoride contents, indicating that the majority of fluoride retained in plaque from these conventional dentifrices appears to be relatively labile. The results of two small-scale human enamel studies showed that NaF dentifrices gave elevated F concentrations in plaque and saliva, respectively, compared with Na2FPO3 dentifrices of equivalent F content, consistent with the main plaque study 1. These findings demonstrate that oral F retention from dentifrices is dependent on the source of ionic fluoride and support the view that NaF dentifrices may be more clinically effective than dentifrices which contain the same amount of F as Na2FPO3.