This study compared fluorosis in the upper central incisors of children from socially diverse backgrounds who had received either 440- or 1,450-ppm F toothpaste from 12 months of age. The children were resident in non-fluoridated districts in the north-west of England. They received either 440- or 1,450-ppm F toothpaste and advice regarding its use until the age of 5-6 years. Dental fluorosis (TF index) was assessed on digital images of dried teeth when the children (n = 1,268) were 8-10 years old. In the less deprived districts the prevalences of fluorosis (TF >or=0) for the 1,450- and 440-ppm F groups were 34.5 and 23.7% (p = 0.006). In the deprived districts the prevalences of fluorosis were 25.2 and 19.5% (p = 0.2). Overall the prevalences of TF >or=2 were 7 and 2.1% for the 1,450- and 440-ppm F groups and 2.2 and 0.2% for TF >or=3. These differences were statistically significant (p < 0.003). There was a strong association between the deprivation status of wards and fluorosis. Only 1 subject with a TF score of 3 was identified in the two most deprived quintiles of the Townsend score. It is concluded that careful targeting of programmes of this type to children living in high caries risk deprived communities carries only a small risk of aesthetically objectionable fluorosis (TF >2) whether low or high fluoride toothpastes are used. High fluoride (1,450 ppm F) toothpastes should not be provided on a community basis to very young children in less deprived communities.