The origin of tumors from cells whose limits of growth are already defined is rendered probable from the absence, entire or in great part, of indifferent cells from certain tumors, and the direct continuity of the latter with a similar normal tissue of the body. Various tumors show such an intimate relation, and there is no sharply defined border-line between the normal tissue and that which represents the tumor. The occasional presence of islets of well characterized tissue at points more or less remote from the normal position of such tissue at the time of their discovery suggests a feasible source for an eventual tumor. Virchow long ago called attention to isolated nodules of cartilage within bones in the vicinity of epiphyseal cartilages, probably detached from the latter, which might serve as the origin of a cartilaginous tumor in this region. This inclusion of tissue is also suggested by the frequency of certain tumors in certain regions where the developmental conditions are favorable. Lücke[79] mentions the frequency of dermoid cysts near the median line of the head, the vicinity of the eye, and the side of the neck. Such regions are those where fissures exist during foetal life, with normal involutions of the outer germinal layer; which involutions may become irregular, and eventually included or shut in, as the fissures become closed. A similar explanation is offered for the frequent occurrence of cartilaginous tumors at the angle of the jaw, it being thought probable that bits of embryonal cartilage, during the formation of the ear, become included in the salivary glands. [Footnote 79: _Volkmann's Sammlung klinischer Vorträge_, xcvii. 819.] In like manner, Cohnheim explains the frequent occurrence of certain epithelial tumors at the orifices of the body--the cervix uteri and the vicinity of the tracheal bifurcation--not through the exposure of these parts to injury, but because they are regions in which embryonal irregularities of development are likely to arise. That congenital, local peculiarities are an important element in the origin of tumors has already been strongly advocated by Virchow. Not only are children born with tumors, but instances of growths eventually arising from birth-marks, and the occurrence of certain tumors in the same locality in successive generations of the same family, are sufficiently familiar.