If it be true, as we cannot doubt, that the social Hymenoptera are derived from solitary species, we can imagine that the evolution of the curious instincts which are shown in the two groups in the division of sexes follows.
In the beginning males and females were of about the same size, with the solitary species and the nesting species, so that the cells built by these insects were equally spacious, provisioned in the same way, and divided by chance. But, on account of the proterandry (early appearance of the males), which is of very ancient date with most articulates, this division of the cells had the fault of exposing species either to the disarrangement of the cells by reason of the premature issuing of the males or to the disadvantage of direct fecundation, since the males would have to await the issuing of the females nearer the orifice. So that selection should favor the nesting Hymenoptera with which the male eggs were laid after the female eggs.
But this rhythm in the laying could be established regularly only with forms in which sexual dimorphism appeared, doubtless by mutation. Sexual dimorphism is advantageous for nest-makers. It should extend under the influence of natural selection, and it is found, in fact, in almost all species of the group. It brought about as a consequence the building of two kinds of cells, the larger ones for the larger sex, and the others smaller and easier to construct. On account of proterandry and to avoid a mixing up in the architecture, the female cells were constructed before the male cells, and it is probably due to a psychic reflex produced by tbe sight of the cells that the mother acts upon her seminal receptacle so as to place in each one of the latter an egg of the appropriate sex.
Without doubt, social wasps came from predatory nesting species in which this rhythm was not yet well established, for after the middle of the summer they lay by chance male eggs and neuter eggs in the ordinary cells. With the honey-bee, on the contrary, the rhythm shows a remarkable regularity, and it is only in the especial cells that the queen lays male eggs a little while before swarming.
In any case, with the honey-bee as with the social wasps, the female eggs are fertilized, while the male eggs are not. It is perhaps the same with ants and with the solitary bees. If future investigations justify this hypothesis, we may believe that the social Hymenoptera have inherited this curious habit from their solitary nesting ancestors, who have at the same time bequeathed to them the rhythm in the laying of the male eggs. As to the abnormal laying of the workers, this seems to be a return to the functions of the solitary species, and, rather than otherwise, vain, for these forms seem incapable of being fertilized, and consequently produce only males.