In this way races arise, which under favorable conditions may be as hereditary as species.  In following these indications, watching opportunities, and breeding only from those individuals which vary most in a desirable direction, man leads the course of variation as he leads a streamlet—­apparently at will, but never against the force of gravitation—­to a long distance from its source, and makes it more subservient to his use or fancy.  He unconsciously strengthens those variations which he prizes when he plants the seed of a favorite fruit, preserves a favorite domestic animal, drowns the uglier kittens of a litter, and allows only the handsomest or the best mousers to propagate.  Still more, by methodical selection, in recent times almost marvelous results have been produced in new breeds of cattle, sheep, and poultry, and new varieties of fruit of greater and greater size or excellence.

It is said that all domestic varieties, if left to run wild, would revert to their aboriginal stocks.  Probably they would wherever various races of one species were left to commingle.  At least the abnormal or exaggerated characteristics induced by high feeding, or high cultivation and prolonged close breeding, would promptly disappear; and the surviving stock would soon blend into a homogeneous result (in a way presently explained), which would naturally be taken for the original form; but we could seldom know if it were so.  It is by no means certain that the result would be the same if the races ran wild each in a separate region.  Dr. Hooker doubts if there is a true reversion in the case of plants.  Mr. Darwin’s observations rather favor it in the animal kingdom.  With mingled races reversion seems well made out in the case of pigeons.  The common opinion upon this subject therefore probably has some foundation, But even if we regard varieties as oscillations around a primitive centre or type, still it appears from the readiness with which such varieties originate that a certain amount of disturbance would carry them beyond the influence of the primordial attraction, where they may become new centres of variation.

Some suppose that races cannot be perpetuated indefinitely even by keeping up the conditions under which they were fixed; but the high antiquity of several, and the actual fixity of many of them, negative this assumption.  “To assert that we could not breed our cart and race horses, long and short horned cattle, and poultry of various breeds, for almost an infinite number of generations, would be opposed to all experience.”

Why varieties develop so readily and deviate so widely under domestication, while they are apparently so rare or so transient in free Nature, may easily be shown.  In Nature, even with hermaphrodite plants, there is a vast amount of cross-fertilization among various individuals of the same species.  The inevitable result of this (as was long ago explained in this Journal [I-7]) is to repress variation, to