Abstract Identity; physical and chemical properties: The commercial term asbestos refers to a group of fibrous serpentine and amphibole minerals that have extraordinary tensile strength, conduct heat poorly, and are relatively resistant to chemical attack. The principal varieties of asbestos used in commerce are chrysotile, a serpentine mineral, and crocidolite and amosite, both of which are amphiboles. Anthophyllite, tremolite, and actinolite asbestos are also amphiboles, but they are rare, and the commercial exploitation of anthophyllite asbestos has been discontinued. Other natural mineral fibres that are considered potentially hazardous because of their physical and chemical properties are erionite, wollastonite, attapulgite, and sepiolite. Environmental levels and exposure: Asbestos is ubiquitous in the environment because of its extensive industrial use and the dissemination of fibres from natural sources. Once in the environment, fibres are mainly transported and distributed vie air and water. Available data using currently-accepted methods of sampling and analysis indicate that fibre levels (fibres : 5 um in length) at remote rural locations are generally below the detection limit (less than l fibre/litre), while those in urban air range from < l to l0 fibres/litre or occasionally higher. Airborne levels in residential areas in the vicinity of industrial sources have been found to be within the range of those in urban areas or occasionally slightly higher. Non-occupational indoor levels are generally within the range found in the ambient air. Occupational exposure levels vary depending on the effectiveness of dust control measures; they may be up to several hundred fibres/ml in industry or mines without or with poor dust control, but are generally well below 2 fibres/ml in modern industry. Reported concentrations in drinking-water range up to 200 x 106 fibres/litre (all fibre lengths). Toxicological effects on animals: Fibrosis in many animal species, and bronchial carcinomas and pleural mesotheliomas in the rat, have been observed following inhalation of both chrysotile and amphibole asbestos. In these studies, there were no consistent increases in tumour incidence at other sites, and there is no convincing evidence that ingested asbestos is carcinogenic in animals. Data from the inhalation studies have shown that shorter asbestos fibres are less fibrogenic and carcinogenic. Few data are available concerning the pathogenicity of the other natural mineral fibres. Fibrosis in rats has been observed following inhalation of attapulgite and sepiolite; a remarkably high incidence of mesotheliomas occurred in rats following inhalation of erionite. Long-fibred attapulgite induced mesotheliomas following intrapleural and intraperitoneal administration. Wollastonite also induced mesothelioma after intrapleural administration. Erionite induced extremely high incidences of mesotheliomas following inhalation exposure and intrapleural and intraperitoneal administration. The length, diameter, and chemical composition of fibres are important determinants of their deposition, clearance, and translocation within the body. Available data also indicate that the potential of fibres to induce mesotheliomas following intrapleural or intraperitoneal injection in animal species is mainly a function of fibre length and diameter; in general, fibres with maximum carcinogenic potency have been reported to be longer than 8 um and less than 1.5 um in diameter. Effects on man: Epidemiological studies, mainly on occupational groups, have established that all types of asbestos fibres are associated with diffuse pulmonary fibrosis (asbestosis), bronchial carcinoma, and primary malignant tumours of the pleura and peritoneum (mesothelioma). That asbestos causes cancers at other sites is less well established. Gastrointestinal and laryngeal cancer are possible, but the causal relationship with asbestos exposure has not yet been firmly established; there is no substantial supporting evidence for cancer at other sites. Asbestos exposure may cause visceral and parietal pleural changes. Cigarette smoking increases the asbestosis mortality and the risk of lung cancer in persons exposed to asbestos but not the risk of mesothelioma. Generally, cases of malignant mesothelioma are rapidly fatal. The observed incidence of these tumours, which was low until about 30 years ago, has been increasing rapidly in males in industrial countries. As asbestos-related mesothelioma became more widely accepted and known to pathologists in western countries, reports of meso- thelioma increased. The incidence of mesothelioma prior to, e.g., 1960, is not known. Mesotheliomas have seldom followed exposure to chrysotile asbestos only. Most, but not all, cases of mesothelioma have a history of occupational exposure to amphibole asbestos, principally crocidolite, either alone or in amphibole-chrysotile mixtures. There is strong evidence that one non-asbestos fibrous mineral (erionite) is carcinogenic in man. This fibrous zeolite is likely to be the cause of localized endemic mesothelioma in Turkey. Non-malignant thickening of the visceral pleura is frequently associated with asbestosis. Thickening of the parietal pleura, sometimes with calcification, may occur in the absence of detectable asbestosis. It is seen in those occupationally exposed to asbestos and also occurs endemically in a number of countries, but the causes have not been fully established. Tremolite fibre has been implicated as an etiological agent in some regions.