An epidemiological study was conducted in the market town of March, Cambridgeshire, to assess the quantitative importance of cooking and table salt to total dietary salt intake by the use of a fused mixture of lithium carbonate and sodium chloride. Men and women aged 20-60 participated in a 12 day study with sequential 24 h urine collections to assess salt sources over a 7 day period. Total salt consumption estimated from urinary chloride excretion amounted to 10.6 +/- 0.55 (SEM) g in 33 men and 7.4 +/- 0.29 (SEM) g in 50 women. The cooking salt eaten was only 0.45 +/- 0.09 (SEM) g in men and women, with men eating more table salt (0.77 g/day) than women (0.46 g/day). Discretionary sources, i.e. cooking and table salt use, contributed only 15% to the total intake. Salt from manufacturing foods and catering in purchased food therefore provided on average 85% of total salt intake. These results are consistent even when an allowance is made for the slightly poorer pouring quality of the lithium-tagged salt. The importance of food as a source of salt was reflected in the significant relationship between the weight of the individual and the amount of salt eaten (for males P less than 0.05 and for females P less than 0.001). Cooking salt consumption did not relate to the amount of salt derived from purchased food nor did table salt use relate to the amount of salt in cooked foods.(ABSTRACT TRUNCATED AT 250 WORDS)