Statistical analyses suggest that 25-50% of the salt intake of Western populations is derived from the discretionary use of cooking and table salt. Yet direct estimates of discretionary salt use by a lithium technique show that in one community in Britain this source contributed only 15% to total intake. The estimates of discretionary salt use in Finland, the United States, and Britain have been exaggerated because salt losses in cooking water were not considered. Only about a quarter of cooking salt actually enters the consumed food; allowance for this in statistical calculations makes data on dietary intake similar to those assessed from urinary sodium excretion. Daily salt intake in Britain averages about 10.7 g for adult men and 8.0 g for women, figures similar to those from countries in northern Europe. The natural salt content of food provides about 10% intake, the remaining 75% being derived from salt added by manufacturers; drinking water provides a negligible amount. Any programme for reducing the salt consumption of a population should therefore concentrate primarily on a reduction in the salt used during food processing.