At the same time the other extreme must equally be avoided--"anthropophobia," which at all costs desires to see in every living organism a "machine," forgetting that a "machine" which lives, that is to say, which grows, takes in nutriment, and strikes a balance between income and expenditure, which, in a word, continually reconstructs itself, is not a "machine," but something entirely different. In other words, it is necessary to steer clear of two dangers. We must avoid (1) identifying the mind of an insect with our own, but, above all, (2) imagining that we, with what knowledge we possess, can reconstruct the mind by our chemical and physical laws. On the other hand, we have to recognise the fact that this mind, and the sensory functions which put it on its guard, are derived, just as with our human selves, from the primitive protoplasmic life. This life, so far as it is specialised in the nervous system by nerve irritability and its connections with the muscular system, is manifested under two aspects. These may be likened to two branches of one trunk. (_a_) _Automatic_ or _instinctive_ activity. This, though perfected by repetition, is definitely inherited. It is uncontrollable and constant in effect, adapted to the circumstances of the special life of the race in question. It is this curious instinctive adaptation--which is so intelligent when it carries out its proper task, so stupid and incapable when diverted to some other purpose--that has deceived so many scientists and philosophers by its insidious analogy with humanly constructed machines. But, automatic as it may appear, instinct is not invariable. In the first place, it presents a racial evolution which of itself alone already demonstrates a certain degree of plasticity from generation to generation. It presents, further, individual variations which are more distinct as it is less deeply fixed by heredity. Thus the divergent instincts of two varieties, _e.g._, of insects, present more individual variability and adaptability than do those instincts common to all species of a genus. In short, if we carefully study the behaviour of each individual of a species of insects with a developed brain (as has been done by P. Huber, Lubbock, Wasmann, and myself, among others, for bees, wasps, and ants), we are not long in finding noteworthy differences, especially when we put the instinct under abnormal conditions. We then force the nervous activity of these insects to present a second and plastic aspect, which to a large extent has been hidden from us under their enormously developed instinct.