Assessment of the fluoride content of water in wells selected randomly and after examining schoolchildren for dental fluorosis.

The random sampling of well water for fluoride assumes that all the wells in an area are equally accessible and that there are no marked local variations in the concentration of fluoride. Both these assumptions were found to be incorrect. Another sampling method is described in which the teeth of schoolchildren are examined for evidence of dental fluorosis and the water drunk by those with different grades of dental mottling is collected for analysis. This method is more reliable than that of random sampling in that wells with a high concentration of fluoride are unlikely to be missed.