"Whatever we may define instinct to be, it is evidently some form of mental manifestation," says Wallace in his "Contributions to Natural Selection" (1871). We know little of the senses of animals; some animals may even possess senses which we have not, and by which stores of knowledge of the outside world may be opened that are closed to us. We do not know certainly, for instance, what is the office of the little stalked balls that are the sole remnants of hind wings in flies, or what is the office of the third joints of the antennae in the same insects, though both these evidently correspond to some sense. How can we pretend to fathom the profound mystery of the mental nature of animals, and decide what or how much they can perceive or remember, reason or reflect? Defining instinct, then, as "the performance by an animal of complex acts, absolutely without instruction," Wallace refuses to accept the theory of such action, in any case where all other modes of explanation have not been exhausted; for "a point which can be proved should not be assumed, and a totally unknown power should not be brought in to explain facts, when known powers may be sufficient." He maintains that there is a possibility, for instance, of the instruction of young birds by old in the art of nest-building. It is quite likely that birds remember the form, size, position, and materials of the nest in which they were hatched, as it is also probable that young birds often pair with old ones who have experience in nest-building. Man's architecture is also chiefly imitative. "Birds brought up from the egg in cages do not make the characteristic nest of their species, even though the proper materials are supplied them, and often make no nest at all, but rudely heap together a quantity of materials." "No one has ever yet obtained the eggs of some bird which builds an elaborate nest, hatched those eggs by steam or under quite a distinct parent, placed them afterwards in an extensive aviary or covered garden, where the situation and the materials of a nest similar to that of the parent-birds may be found, and then seen what kind of nest these birds would build. If under these rigorous conditions they choose the same materials, the same situation, and construct the nest in the same way and as perfectly as their parents did, instinct would be proved in their case; now it is only assumed.... So no one has ever carefully taken the pupae of a hive of bees out of the comb, removed them from the presence of other bees, and loosed them in a large conservatory with plenty of flowers and food, and observed what kind of cells they would construct. But till this is done no one can say that, with every new swarm there are no bees older than those of the same year, who may be the teachers in forming the new comb."[24] "Young birds never have the song peculiar to their species if they have not heard it, whereas they acquire very easily the song of