Page:Popular Science Monthly Volume 62.djvu/433

Rh families have done. Some exceptions to this may occur, as among the descendants of Jonathan Edwards and the famous musician, Bach, but in these cases the mental qualities were perpetuated.

In the lower forms of animal life we know by actual experimentation that slight changes in the environment occasion the greatest difference in results, still in spite of the strange modifications that may be occasioned in the developing fish or frog by external mechanical or chemical means, the question resolves itself under ordinary conditions to the nature of the primary germ-cells.

If a naturalist were stocking two tanks, one for fishes and one for frogs, and had eggs of both to use for that purpose, the first practical question for him would be which are the eggs of fishes and which are the eggs of frogs?

It is just so in the development of the human mind. As far as the practical results are concerned, the one bit of knowledge, the possession of which will best enable us to predict the fully developed adult, is an answer to the same sort of question as that we would first wish to know in the case of the fishes and the frogs. What is the nature of the primary germ-cells? Since for obvious reasons we can not know this nature, the next' best thing to know is its theoretical probabilities as derived from a proper study of the ancestry.

It would seem from the facts here studied that the probabilities will be roughly as given below. Quality possessed by entire ancestry is almost sure to appear. Quality possessed by one parent and half the ancestry is likely to appear with almost equal force, in one out of every two descendants. Quality possessed by one parent only, and not present in the ancestry, has one chance in about four for its appearance in the progeny. Quality not possessed by either parent, but present in all the grandparents and most of the remaining ancestry, would also have about one chance in two for its appearance in one of the children. If only one of the great grandparents possessed the quality in question, then the chances of its appearance in any one of the grandchildren of this ancestor would be only about one chance in sixteen. It would be, however, very unlikely that some of the remote ancestry had not also the quality in question, so the chances would be raised in a greater or less degree according to the proportionate amount of this remote influence.

There does not seem to be the least reason for assuming that the male side is any more or less potent than the female side in the transmission of family characteristics, nor does there seem to be any grounds for the fancied belief that sons tend to resemble their mothers and daughters their fathers, or the more generally accepted scientific belief to the contrary. No figures have been compiled on this subject