port, Goddard and Rosanoff, which they interpret as evidence that original imbecility is due to the absence of a single determiner, and that an originally neurotic, unstable mental organization is explainable almost as simply. It is with regret that I must assure you that these observations are susceptible of a very different interpretation. Much as I should like to believe that these burdens on man's nature are each carried in heredity in a single package, which selective breeding can shuffie off in a generation or so, I can not. A eugenics that assumes that intellect, morality, sanity and energy are so many single niches in the germs which selective breeding can, by simple transfers, permanently fill, is, I fear, doomed to disappointment and reaction. I dare to believe that the time will come when a human being idiotic by germinal defect will be extinct like the dinosaur—a subject for curious fiction and for the paleontology of human nature; but I have no hope that such a change can be made with the ease with which we can change short peas to tall, curly-haired guinea pigs to sleek, or plain blossoms to mottled ones.
There is another fundamental question whose answer is needed for the most economical selective breeding of human nature, a question which time permits me only to mention, not to describe clearly. Stated as a series of questions, it is this: Do the germs which a man produces—his potential halves of offspring—represent a collection peculiar to him, or only a collection peculiar to some line, or strain, or stock, or variety, of mankind of which he is one exemplar?
Suppose a hundred men and a hundred women to exist, each with identical germinal constitutions, so that, say, in every case one tenth of the germs (or ova) would be of quality 5; one fifth, of quality 6; two fifths, of quality 7; one fifth, of quality 8; and one tenth, of quality 9. Suppose that they mated and had five hundred offspring. Suppose that the best fifty of this second generation married exclusively among themselves; and similarly for the worst fifty. Would the offspring of these two groups differ, the children of the best fifty being superior to the children of the worst fifty? Or would this third generation revert absolutely to the condition of the grandparental stock whence they all came; and be alike, regardless of the great difference in their parentage?
Does the selection of a superior man pay because his superiority is, in and of itself, a symptom of probable excellence in his germs; or only because his superiority is a symptom that he is probably of a superior "line" or strain?
That the second answer of each pair may be the true one, is a natural, though not, I think, an inevitable, inference from the work of Johanssen, Jennings and others. They have found selective breeding within any one pure line futile, save when some peculiar and rare variations have taken place within it. Their work is of very great importance and
