The concentration of lead in tibia (Pb-T) was measured in vivo by a 109Cd K-shell X-ray fluorescence technique in 123 workers from a primary lead smelter (age: mean, 45 years; range, 30-61; duration of employment: mean, 20 years; range, 7-45). Their cumulative blood lead index (CBLI) was also calculated on the basis of the blood lead (Pb-B) records available from the company's medical files. Geometric mean for Pb-T was 49 micrograms Pb/g bone mineral (range, 15-167). The company's health surveillance programs, implemented since 1945, resulted in Pb-B values which rarely exceeded 70 micrograms Pb/dl whole blood. Pb-B at the time of Pb-T measurement averaged 31 micrograms Pb/dl (range, 6-62) and the geometric mean for CBLI amounted to 803 micrograms Pb/dl x year (range, 220-2130). Despite various assumptions and uncertainties inherent in the assessment of the cumulative lead dose through Pb-T measurement or CBLI calculation, the relation between both variables in the present lead smelter populations is very strong (rpearson = 0.80, P < 0.0001; age explained at the most 9.5% of the variance). The slope of the regression equation of log Pb-T vs log CBLI showed that a doubling of CBLI also corresponds to doubling of Pb-T. It may be concluded that a sound calculation of CBLI represents a valid surrogate for estimating the life time integrated dose of lead as assessed by the measurement of cortical bone lead (e.g., in tibia).