The objective of this paper is to review published data on the carcinogenicity of asbestos fibres with regard to the elucidation of a potential risk originating from exposure to man-made vitreous fibres (MMVF). Steps in the comparison of the two fibre classes are characterization of the fibres, pulmonary deposition, biodurability and biopersistence and a review of the cancer risk from asbestos fibres after inhalation in rats and humans. Various dust samples of chrysotile, crocidolite, and amosite were used as reference materials in studies with experimental animals. These fibres are normally thinner and shorter than MMVF. These differences in dimensions cause differences in the deposition in the airways. In addition, significant dissimilarities exist in the deposition pattern between rats and humans. Data from biopersistence studies show that focusing only on fibres longer than 20 microm and using weighted half-time for a characterization of risk may be misleading. Inhalation experiments with rats need fibre exposure concentrations over 100 times higher to match the lung cancer risk of asbestos workers, and about 1,000 times higher to reach the same mesothelioma risk. Also, the striking difference between the low lung burden of amphibole fibres of asbestos workers with mesothelioma and the more than 1,000 times higher lung burden of rats with a low mesothelioma risk demonstrates the low sensitivity of the inhalation test model for the carcinogenic potency even of crocidolite fibres. It can be concluded that the rat inhalation model is also not sensitive enough to predict the cancer risk of other fibre types for humans.