To define the relationships between chrysotile exposure in fibre terms and death from specific cancers, and pneumoconiosis, all 11,379 persons born 1891-1920 who had worked in the asbestos mines and mills of Quebec for a month or more before 1967 were followed to the end of 1975. Among the 10,939 men, there had been 4,463 deaths, 634 from these causes. For each death, referents were randomly selected from among men in the cohort born in the same year as the case and known to have survived to a greater age. For each case and his referents, exposures accumulated up to nine years before the death of the case had been obtained as (million particles per cubic foot) x years. Fibre counts were estimated for each work-place so that all exposures could be expressed in (fibres/ml) x years. The ratio of the means for all 2,586 accumulated exposures was 3.46 (f/ml)/mpcf. Relative Risks (RR) were related to exposure by matched analysis. For pneumoconiosis and lung cancer, RR = 1 + b.(f/ml).y fitted well, with b estimated as 0.00647 and 0.00038, respectively. For cancers of upper and of lower G.I. tract, severe exposure was associated with elevated RRs, but rather unclearly. For other abdominal cancers, and laryngeal cancer, risks and exposure were not positively associated. The asbestos-smoking interaction in lung cancer was closer to multiplicative than to additive.