THE DETERMINING INFLUENCE OF TAR, BENZPYRENE, AND METHYLCHOLANTHRENE ON THE CHARACTER OF THE BENIGN TUMORS INDUCED THEREWITH IN RABBIT SKIN.

The benign tumors of rabbit skin which appear in response to benzpyrene and methylcholanthrene are nearly all of the same kinds that tar produces, namely frill horns, papillomas, and carcinomatoids; but the collateral effects of these agents render many of them very different from the tar tumors. The connective tissue of the corium is so slowly excited by the pure hydrocarbons that for a long while it furnishes to most of the growths only a scanty stroma, when any, and in consequence they remain small,-low, dry, indolent, bas-relief affairs, made up almost entirely of the neoplastic epithelium,-not florid, fleshy excrescences with a large connective tissue component, such as tarring calls forth. The frill horns usually desquamate instead of building up like those due to tar, and some of their cells undergo a dyskeratotic change with result in spherical, homogeneous, deep-staining bullet-like elements, which give to the growths a singular aspect. In a frill horn due to benzpyrene numerous inclusion bodies were come upon which would seem to have been the result of intercurrent infection with a "passenger virus." Benzpyrene and methylcholanthrene produce papillomas, carcinomatoids, and frill horns in very different proportionate numbers from those obtaining when tar is the carcinogen. Tar gives rise much more frequently to carcinomatoids, -papillomas urged on to mimic cancers,-as would follow from its pronounced stimulating influence; yet it seldom produces frill horns whereas the pure hydrocarbons do so frequently. All three agents cause many more cells to become potentially capable of forming tumors than do so ordinarily, but the latent neoplastic elements on which tar exerts a promoting influence, causing them to form visible growths, are in general not the same as those which respond to benzpyrene and methylcholanthrene. Yet the relatively rare occurrence of frill horns due to tar cannot be wholly explained in this way and it becomes necessary to suppose that tar seldom changes normal cells into frill horn cells. Benzpyrene and methylcholanthrene give rise now and again to sebaceous adenomas, as tar does not. But tar does away with the sebaceous glands at an early period whereas the other two carcinogens cause them to increase notably in number. No evidence has been obtained that tar, benzpyrene, methylcholanthrene, arsenical preparations, or benzene,-which produces tumors of rabbit skin occasionally (1),-bring about any neoplastic changes peculiar to them individually when they act upon cells of a single sort, those of the stratum germinativum of rabbit epidermis. Yet the experimental findings make plain that these agents exert no inconsiderable influence on the morphology of the benign cutaneous growths they call forth and on the frequency with which this or that kind occurs. In appraising the action of carcinogens one must take into account not only the capacity of these agents to induce neoplastic change and to promote, or perhaps suppress, tumor growth but an ability to condition to no inconsiderable extent both the kind of tumor arising and its structure.