Mesothelioma is rare except among the many miners, shipbuilders, and insulation workers and their families who had a large exposure to asbestos within the period from 1940 to the early 1960's, when this link was found and occupational inhalation was reduced. This cancer first becomes detectable twenty to forty years after the exposure. Thus its incidence has been increasing and will continue to be responsible for 2 to 3,000 death per year until at least the end of the century. Mesothelioma does not respond to current types of treatment. Very little is known about its biology, nor is the mechanism of asbestos-induced transformation or the reason for the long latency period for tumor growth understood. I have recently identified the specific requirements of normal human mesothelial cells for rapid, long-term growth in culture, and have discovered that these cells can undergo an extraordinary, but reversible dedifferentiation during rapid growth. Surprisingly, most mesotheliomas do not grow in the culture conditions optimal for normal mesothelial cells. I propose a three-pronged approach to understanding malignant transformation of mesothelial cells: 1) experimental carcinogenesis of normal human cells with asbestos or with X-rays, selecting for growth factor-independent variants, 2) study of the tumorigenicity of cell lines cultured from normal or asbestos-treated rats, and 3) identification of additional substratum components required by mesothelioma cells for growth in culture. In order to assess the malignant or premalignant behavior of experimental transformants and natural mesotheliomas, we will inoculate cells intraperitoneally into nude mice and assess the ability of the cells to replace the resident mouse mesothelial cell population and/or to invade the underlying stroma. Our results will provide specific information about this human cancer, and, more generally, about the role of EGF- and hydrocortisone-sensitive systems in the regulation of epithelial cell growth.