We know that rabbits do not breed kittens, nor eagles geese, nor apples oranges, nor negroes whites. We know that in all cases where the breed is pure the descent is pure; and we understand that where a black sheep is born into a white flock, or a fair child is born of dark fore-parents, the "sport," as it is called, is due to atavism, or breeding back. Somewhere, near or far, the breed has been "crossed." But there is one question that has caused a good deal of doubt and perplexity, and, as the answer to that question is not obvious, we will consider it here. A "sport" is "an individual departure from a type." A sport is a "freak of nature." A genius is a "sport"; and the question we are to answer here is: How does heredity account for genius? To make the matter quite clear, and to meet all doubts, we will split our question into two: 1. How is it that genius does not always beget genius? 2. How is it mediocrity does sometimes beget genius? Take the first question. How is it that genius does not always beget genius? Mr. Galton has disposed of the objection that clever men do not have clever sons by showing that clever men often do have clever sons. But the fact remains that such men as Shakespeare, Plato, Cæsar, and Socrates never have children as great as themselves. And it has been claimed that this fact belies heredity. But to those who know even a very little about heredity it is quite obvious that we ought not to expect the son of a very great genius to be equal to his father. Such a recurrence is rendered almost impossible by the law of variation. A great man is a lucky product of heredity and environment. He is a fortunate, and accidental, blending of several qualities which make greatness possible. But the great man's son is not born of the same parents as his father. His blood is only half of it drawn from the families which produced his father's greatness; the other half is from another family, which may contain no elements of greatness. Thus so far from its being strange that genius does not beget genius, we see that it would be strange if genius did beget genius. The children of Shakespeare would not be Shakespeareans: they would be half Shakespeare and half Hathaway; and it is quite possible that their intellectual qualities might come chiefly from the mother's side. Now, if Ann Hathaway's family were not intellectually equal to Shakespeare's family, how could we expect the children of those two to be equal to the child of the superior breed? We should not expect a mixture of wine and water to be all wine; nor the foal of a blood horse and a half-bred mare to be a thoroughbred horse. So much for the first question. Those who ask such a question have lost sight of the law of variation. Now for the second question. How is it that mediocrity breeds genius? The answer to that is that mediocrity does _not_ breed genius. Let us take a case that is often cited: the case of the great musician, Handel.