In insects the chromosomes appear to be predominant throughout life. In vertebrates, however, they seem to play their chief rôle in early development, ending by building up either a male or a female gonad in the early embryo. This, once produced, takes over what remains of the task of sex-determination. It secretes a specific internal secretion which in a male acts so as to encourage the growth of male organs and instincts, to suppress those of females; and vice versa in a female. As a result of this difference we find that castration in insects, even followed by engrafting of a gonad of opposite sex, produces no effect upon other sexual characters; whereas it exerts a profound effect upon mammals or birds. As a second result, we find that in vertebrates the gonads form part of what has been called the chemical directorate of the body--the interlocking system of endocrine glands, each of which is exerting an effect upon the rest. The importance of this is seen in the experiments of Steinach, Sand, Voronoff, and others, who have been able to obtain a rejuvenating effect in senile mammals by increasing, by various methods, the amount of secreting reproductive organ in the body.[32] To what then has our rapid survey led us? The actual origin of sex is lost to us in the mists of a time inconceivably remote. Its preservation once in existence, and its present all-but-universal distribution seem to be definitely associated with the biological advantage of the plasticity which it confers. Later, the primary difference between male and female--their power of producing different sorts of reproductive cells--leads on to secondary differences. These differences may be biologically speaking non-significant, mere accidents of the primary difference. Or they may be in the nature of a division of labour between the sexes, this division of labour usually concerning the protection of the embryo or the care of the young, or more rarely the preservation of the individual itself. Or, finally, they may concern the more efficient union of the gametes; such differences may merely affect the ducts and apertures of the reproductive system, and be more or less mechanical; or they may concern the use of these systems, in the form of still mechanical instincts, or they may be concerned in some way or other with the emotional side of the animals, and consist in characters and actions which stimulate the emotions of the other sex, characters which we have termed epigamic. It is only in higher groups that these emotion-stimulating sexual characters arise, for only in them has mind reached a sufficient degree of perfection. But even though detailed study reveals in a bird or a mammal a mental life of a complexity far more considerable than the average man would imagine, yet on the whole it is straightforward and its currents run fairly direct from stimulus to fulfilment.