Page:Popular Science Monthly Volume 83.djvu/134

130 in arithmetic in school children, finds, as one sure principle of explanation, the likeness of children to parents—and this even in subtle traits and relations between traits, of whose very existence the parents were not aware, and which the parents would not have known how to nurture had they known of their existence.

Dr. Keyes has recently made an elaborate study of various possible causes of the rate of progress of a child through the elementary school. He traces the effects of defective vision, of sickness, of moving from one school to another, and so on, but finds nothing of great moment until he happens to trace family relationships. Then it appears that certain families are thick with "accelerates," or pupils who win double promotions, whereas other families are thick with retarded pupils, who require two years to complete a normal year's work. Of 168 families, only 30 contain both an "accelerated" and a "retarded" pupil, whereas 138 show either two or more accelerates or two or more retarded pupils. The differences in home training are here not allowed for, but, in view of what has been found in other cases, it appears certain that the rate at which a child will progress in school in comparison with his fellows is determined in large measure before he is born.

In intellect and morals, as in bodily structure and features, men differ, differ by original nature, and differ by families. There are hereditary bonds by which one kind of intellect or character rather than another is produced. Selective breeding can alter a man's capacity to learn, to keep sane, to cherish justice or to be happy.

 The upper surface being taken to represent the existing distribution of Intellect, the lower surface represents what might be expected from, say, ten or twenty generations of breeding exclusively from the apparently best tenth of human Intellects.