First, we should be careful not to use the phrase, "of advantage to the species," vaguely and indefinitely, but should in all cases endeavour clearly to indicate wherein lies the particular advantage, and how its possession enables the organism to escape elimination; next, we must remember that the advantage must be immediate and present, prospective advantage being, of course, inoperative; then we must endeavour to show that the advantage is really sufficient to decide the question of elimination or non-elimination; lastly, we must distinguish between indiscriminate and differential destruction, between mere numerical reduction by death or otherwise and selective elimination. (1) In illustration of the first point, we may select a passage from the writings of even so great a biologist as Professor Weismann. As is well known, Professor Weismann believes that senility and death are no part of the natural heritage of animal life, but have been introduced among the metazoa on utilitarian grounds. In his earlier papers, he attributed the introduction of death, and the tissue-degeneration that precedes it, to the direct action of natural selection.[CJ] More lately, he attributes it to the cessation of selection.[CK] Concerning this later view, we shall have somewhat to say presently; we may now consider the former as an example of too indefinite a use of such phrases as "of advantage to the species." "Worn-out individuals," says Professor Weismann, "are not only valueless to the species, but they are even harmful, for they take the places of those which are sound. Hence, by the operation of natural selection, the life of our hypothetically immortal individual would be shortened by the amount which was useless to the species. It would be reduced to a length which would afford the most favourable conditions of existence of as large a number as possible of vigorous individuals at the same time." This may be so, but, as it stands, the _modus operandi_ is not given, and is not obvious. We start with a hypothetically immortal metazoon. Barring accidents, it will go on existing indefinitely. But you cannot bar accidents for an indefinite time; hence, the longer the individual lives, the more defective and crippled it becomes. There is neither natural decay nor natural death here. The organism is gradually crippled through accident and injury. But the crippled individuals are harmful to the species, because they take the places of those which are sound. Therefore, says Professor Weismann, natural decay and death step in to take them off before they have time to become cripples. Now, the point I wish to notice is that there is no definite statement how or why natural decrepitude should thus be introduced. We must remember that it is not until a late stage in evolution that, through the association of its members, groups of organisms compete with other groups. In the earlier stages, when we must suppose decrepitude and death to arise on Professor Weismann's