teenth share of black blood by any possible test save documentary evidence.
Here, then, we have a clear, physical, and almost mathematically demonstrable case, showing that, so far as regards bodily peculiarities at least, the child is on the average just equally compounded of traits derived from both its parents. Among hundreds and hundreds of mulatto and quadroon children whom I have observed, I have never known a single genuine instance to the contrary. Heredity comes out exactly true; you get just as much of each color in every case as you would naturally expect to do from a mixture of given proportions. In other words, all mulattoes are recognizably different from all quadroons, and all quadroons from all octoroons or all sambos.
This simple fact, I venture to think, gives us at once the real key to the whole complex problem of idiosyncrasy and character. Every child on the average represents one half its father and one half its mother. It is a Jones in this, and in that a Robinson. Here it takes after its grandfather the earl, and there it resembles its grandmother the washerwoman. These traits it derives from the distinguished De Montmorencies, and those from the family of the late lamented Mr. Peace the burglar. But, on the whole, however diversely and curiously the various individual peculiarities may be compounded, it is at bottom a Robinson-Jones, a complex of all its converging strains, its diverse noble and ignoble ancestors. It represents a cumulative effect of antecedent causes, all of which it shares equally on the average with every one of its brothers and sisters.
How does it happen, then, suggests the easy objector, that two brothers or two sisters, born of the same father and mother, twins it may even be, "are often more unlike each other in character and mental qualities than any two ordinary strangers"? Well, the answer simply is, it doesn't happen. Make sure of your facts before you begin to philosophize upon them. Children of the same parents are always very much like one another in all essential fundamentals; they may differ a good deal among themselves, but their differences are really and truly as nothing compared with the vast complexity of their resemblances. The case of twins, in fact, is a peculiarly unfortunate one to allege in this respect, for Mr. Galton has collected an immense mass of evidence tending to show that just as twins usually resemble one another, almost indistinguishably, in face and feature, so do they resemble one another almost as narrowly in character and intellect. I know an instance myself of two twin sisters, one of whom has lived all her life in India, and the other in England, but who, in spite of this difference in circumstances, preserve so entirely their original identity of form and nature that I do not myself in the least discriminate between them in any way, mentally or physically, though they happen to be members of my own family. It does not at all matter to me whether it was Polly who said a thing or Lucy. I regard it in either