In insanity and epilepsy we have probably the last refuge and almost only valid instance of the old belief in the remorseless heredity of disease. But even here the part played by heredity is probably only a fraction of that which it is popularly, and even professionally, believed to play. It is, of course, obvious that diseases which tend quickly to destroy the life of the patient, especially those which kill or seriously cripple him before he has reached the age of reproduction, or prevent his long surviving that epoch, will not, for mechanical reasons, become hereditary. The Black Death, or the cholera, for instance, could not "run in a family." Supposing that children were born with a special susceptibility to this disease, there would obviously soon be no family left. The same is true in a lesser degree of milder or more chronic diseases. The family which was hereditarily predisposed to scarlet fever, measles, smallpox, or tuberculosis would not last long, and in fact the whole progress of civilization has been a continuous process of the weeding out of those who were most susceptible and the survival of those who were least so. But when we come to deal with certain conditions, fortunately rare, such as functional disturbances of the nervous system, which neither seriously unfit their possessor for the struggle of life nor prevent him from reproducing his kind, then it becomes possible that a tendency to such disease may be transmitted through several successive generations. Such is the case with insanity, with epilepsy, with _hemophilia_, or "bleeders," and with certain rare and curious disturbances of the nervous system, such as the hereditary _ataxias_ and "tics" of various sorts. However, even here the only conditions on which these diseases can continue to run in a family for more than one or two generations is either that they shall be mild in form or that only a comparatively small percentage of the total family shall be affected by them. If, for instance, two-thirds, one-half, or even a third of the descendants of a mentally unsound individual were to become insane, it would only need a few generations for that family to be crushed to the wall. While the descendants of insane persons are distinctly more liable to become insane than the rest of the community, yet, on account of their fewness, this tendency probably does not account for more than a small fraction of the total insanity. We should, by all means, prevent the marriage of the insane and discourage that of their children, and the development of any well-defined form of insanity should act at once, _ipso facto_, as a ground and cause of divorce.