At a secondary smelter (24 smelting workers, 13 refinery workers), data on exposure to lead were collected by systematic observation of hygienic behaviour, a questionnaire, personal sampling of lead dust in ambient air (PbA) and determination of lead in the blood level (PbB). The smelting workers showed a negative relation between PbA and PbB. The 53% variance in PbB levels in the smelting workers can be explained by the combination of PbA, the percentage of time an air-stream helmet is worn, the frequency of cigarette smoking at the workplace and the amount of spitting. Air-stream helmets and spitting contribute to a lower PbB, whereas smoking contributes to a higher PbB. Moreover, expected PbB levels were computed by using several regression equations for the relation between PbA and PbB, as suggested in the Final OSHA Standard for Occupational Exposure to Lead (OSHA 1978). As the percentage of time an air-stream helmet is worn increases, the deviation from the expected PbB falls substantially. The refinery workers showed the expected weak positive correlation coefficient between PbA and PbB. No direct relation between the PbB level and the observed hygienic behaviour could be established. However, there was a positive relation between the level of education and the level of PbB. Moreover, the level of education was related to the frequency of eating at the workplace and negative as far as the percentage of time gloves are worn is concerned. Our conclusion is that hygienic behaviour is a major factor that modifies the relation between PbA and PbB in groups of workers.(ABSTRACT TRUNCATED AT 250 WORDS)