Again, there is the generalisation of heredity formulated by Mr. Galton, and named by him the _law of regression towards mediocrity_. It asserts that the children of those who are above or below the mean of a race, tend to return towards that mean. The children of the born criminal will be probably somewhat less criminal in tendency than he, though more criminal than the average citizen. The children of the man of genius, if he has any, will probably be nearer mediocrity than he, though on the average possessing greater talent than the average citizen. It is thus not in the nature of sheer genius to reproduce on its own level. It is only the critics who are wholly ignorant of the elementary facts of heredity that attribute to the eugenist an expectation of which no one knows the absurdity so well as he does. On the other hand, it is impossible to question that the hereditary transmission of genius or great talent does occur. One may cite at random such cases as that of the Bach family, Thomas and Matthew Arnold, James and John Stuart Mill: and the reader who is inclined to believe that there is no law or likelihood in this matter, must certainly make himself acquainted with Mr. Galton's _Hereditary Genius_, and with such a paper as that which he printed in _Sociological Papers_, 1904, furnishing an “index to achievements of near kinsfolk of some of the Fellows of the Royal Society.” There is, of course, the obvious fallacy involved in the possibility that not heredity but environment was really responsible for many of these cases. It must have been a great thing to have such a father as James Mill. But it would be equally idle to imagine that the evidence can be dismissed with this criticism. A Matthew Arnold, a John Stuart Mill, could not be manufactured out of any chance material by an ideal education continued for a thousand years.