Altruism and Related Phenomena, Mainly in Social Insects

In what sense can the self-sacrificing sterile ant be considered to "struggle for existence" or to endeavor to maximize the numbers of its descendants? Since the founding of the theory of evolution by natural selection, most biologists have evaded this question by focusing attention exclusively on the colony as the reproducing unit. There is a powerful precedent for this. Darwin himself took this course. He saw only a "minor" difficulty in the evolution of sterility, and he passed over it in a few lines as he proceeded to discuss the "great" difficulty of how the special aptitudes of the workers could be passed on in latent form by their fertile sisters (26). A difficulty over sterility exists, nevertheless, and it is the more surprising that Darwin should have passed over it in that he discussed-but left unsolved-a parallel one raised by the social virtues (courage and self-sacrifice) in man (25). He saw that such qualities would be promoted in intergroup selection but counter-selected within each group. Perhaps the possible avenues of indiscipline in social insects had been so little reported in Darwin's time that the problems they raised were easily overlooked. Darwin's inadequate understanding of heredity may, likewise, have helped to keep the problem out of focus. With better knowledge of heredity and with more facts regarding the social insects to draw upon, Weismann (154) recognized the possible conflict between intergroup and intragroup selection in the evolution of worker attributes. He made the perceptive comment that, "Obviously the workers must be more rapidly improved when all in a hive are progeny of one queen-i.e. they are all alike or almost alike." But this comment was made in the course of discussion of another topic, and he did not pursue the matter. Soon Mendelian genetics resolved Darwin's difficulty of latency, apart from details of mechanism. But the disappearance of this problem does not seem to have given greater prominence to the other, and the question of how worker sterility comes to be selected continued to receive only occasional comment for a long time. Sturtevant (142) in 1938 again outlined it with admirable clarity and with special reference to multiqueened (polygynic) organization, which was by then well known. Rau (118) in 1940 also briefly touched on this crux when he noted how small might be the step separating workerlike behavior in auxiliary Polistes queens 193