Abstract Over time, attention to health effects induced by inorganic fibres has increased, resulting not only from non-occupational exposure to asbestos ( i.e . environmental, either natural or anthropogenic), but also of exposure to non-asbestos inorganic fibres. The International Agency for Research on Cancer classified all forms of asbestos (chrysotile, crocidolite, amosite, tremolite, actinolite and anthophyllite) as carcinogenic to humans (Group 1). The most effective source of information on the health effects of environmental contaminants in humans would be humans themselves as research subjects. Obviously, for ethical and other reasons this is rarely possible. Carcinogen bioassays, such as the ones performed by the Ramazzini Institute (RI), are currently the most predictive non-human model for studying the carcinogenicity of a substance. An alternative method for exposure assessment consists of using a non-experimental animal model represented by populations of animal sentinel systems (ASS). Even if at present there are relatively few studies in this field, the presence of fibres in samples of lung tissue of these animals shows that it is possible to use ASS as indicators of environmental background exposure. However, caution must be taken in extrapolating results from ASS and considering them in terms of human health. The physical mechanisms rather than chemical reaction between inorganic fibres and cells are still unclear. Joint research conducted by a team of biologists and mineralogists in order to investigate the mechanisms of action of the fibres in biological tissues of rodents in vivo have been undertaken by the Department of Chemical and Geological Sciences, University of Modena and Reggio Emilia in collaboration with the Cesare Maltoni Cancer Research Centre of the Ramazzini Institute CMCRC/RI. When planned for precise purposes with attention to the key, general, and specific requisites, and when conducted by standardized methods, experimental studies represent an important tool capable of providing useful information and reducing the uncertainties in terms of risk assessment.