Next, I suppose, physiological selection, or Romanes’s factor, came into operation. After the introduction of sex, it became necessary that the individuals of some varieties should be isolated in some way, so as to prevent the swamping of varietal characters, as fast as formed, in a common stock, by _cross-breeding_. In very low forms, with slow locomotion, such isolation might easily take place accidentally. Even in higher forms, changes in physical geography or accidental dispersion by winds and currents would often produce geographical isolation, and thus, by preventing crossing with the parent stock, secure the formation of new species from such isolated varieties. But, in order to insure in all cases the preservation of commencing species, _sexual isolation_, or partial or complete infertility of some varieties with other varieties and with the parent stock, was introduced, as I suppose, later. The process by which this takes place has already been explained. According to Romanes, natural selection alone, with cross-breeding, tends to _monotypal_ evolution; isolation of some kind is necessary for polytypal evolution. The tree of evolution, under the influence of natural selection alone, grows, palm-like, from its _terminal bud_; isolation of varieties was necessary for the starting of _lateral buds_, and thus for the profuse ramification which is its most conspicuous character. Next, I suppose, was introduced _sexual selection_, or contest among the males, by battle or by display, for possession of the females, and the success of the strongest or the most attractive; and the perpetuation and increase of these superior qualities of strength and beauty in the next generation. This, I suppose, was later, because connected with a higher development of the psychical nature. This is especially true where splendor of color or beauty of song determines the selection. As might be supposed, therefore, this factor is operative only among the highest animals, especially birds and mammals.[18]