In studying the effects of the carcinogenic hydrocarbons these substances are most commonly applied to the skin or injected subcutaneously for obvious reasons. No doubt there are variations in the sensitivity of different animals, as well as of the different tissues of individual animals; as examples may be cited the extreme reactivity of mouse skin and of mouse and rat subcutis, which are in sharp contrast to the high resistance characteristic of rat skin. Nevertheless, as proof of the general effectiveness of these agents in the production of neoplasia, the literature abounds in examples of tumors induced in practically all tissues. But the ability of the carcinogenic hydrocarbons to produce new growths in the liver has been seriously questioned by a number of investigators, who found minimal or nonspecific reactions in this organ after exposure for considerable periods (5, 8, 9, 12). Shear, Stewart, and Seligman (9) state that this apparent resistance of the liver does not depend upon absorption or removal of the chemicals, for pellets that had lain in the mouse liver for as long as 16 months without inducing malignant disease were found intact, and surrounded by only a mild tissue reaction. Still, their protocols include a description of an adenocarcinoma that was found in the liver of a mouse of the C57 black strain, in which spontaneous hepatic neoplasms do not occur, 12 months after the insertion of a thread coated with dibenzanthracene; the authors favor an origin of this induced tumor in the biliary ducts or gall bladder because of its histology and location. ()berling and the" Gu&ins (7), on the other hand, recorded 2 examples of sarcoma of the rat liver in animals examined 16 and 25 months after an implant of benzpyrene crystals. Strong (10) succeeded in producing carcinoma of the liver in 19 of 1,367 N H O mice at an average age of 406.2 days, by injecting methylcholanthrene subcutaneously at 60 days of age; spontaneous hepatic tumors do not occur in this strain. Recently White and Eschenbrenner (11) reported the occurrence of multiple nodules, interpreted as hepatomas, in the cirrhotic livers ot~ 2 of 6 rats that survived 14 months on a basal diet containing 60 mgm. of 1,2-benzanthracene per 100 mgm. of food. This parent hydrocarbon possesses only negligible carcinogenicity when injected subcutaneously or painted on the skin. That the liver is endowed with no general resistance to carcinogenic agents such as might be linked with its extensive detoxifying powers is clearly evident from the ease of production of parenchymal cell neoplasia in the rat with azo compounds administered orally or parenterally, although they are ineffective when applied directly to the liver (6); or of sarcoma in this animal with Cysticercus larvae (1). There is much of value in the observation of Dunning, Curtis, and Bullock (2) who, on the basis of a study of subcutaneous and hepatic tumors induced by the carcinogenic hydrocarbons and Cysticereus disease respectively in rats, concluded that "the potency for malignancy must be a universal cell characteristic and that the histogenesis of these tumors was determined by the fortuitous exposure to the irritant of the various types of cells." The present note records the induction of 7 hcpatic sarcomas in rats bearing implants of methylcholanthrene or benzpyrene in the liver. The cxperiments were performed between 1941 and 1943, and were not reported earlier because of the absence of the author in military service.