If it is not necessary to see in the actual phenomena of differential sensitiveness a series of trials which rectify the errors, and if one considers the stops, turnings, and recoils which characterize these phenomena as primitive and automatic reactions, under the same title as those of tropisms, we do not believe that these phenomena can be adapted to the variations which they provoke.

In fact, on the contrary, tropic reactions which are always well determined, the responses of differential sensitiveness, lend themselves to different modalities, and these modalities may be tributaries of experience.

We have seen (page 74) the bedbug direct itself toward the light which it ordinarily flees, at the simple approach of a lamp interposed from behind, and the Spirostomum ambiguum recoil from a noxious drop placed behind its body. These two responses seem false, they place the animal in peril, which perfectly conforms to the law of differential sensitiveness. But in both cases the organism finds itself in the presence of experimental variations which never occur in the environment in which it normally passes its exist* enee. In cases where several modalities are possible, the insect seems to react by chance and often to adopt the wrong modality. This is the case in the following experiment, carried on by Bohn (1906): When the bedbug starts away from the bright window, a black or white screen is placed in its path, which produces a sudden variation in the intensity of the light ; the animal hesitates, and then, combining the rotation of differential sensitiveness with the acquired impulse, goes in an. oblique direction, either to the side opposite the screen, or toward the screen itself. It pursues its march in the same direction if it meets the screen on the left side. Its adaptation is null ; between the two modalities (direction to the right or to the left) which differential sensitiveness permits, the animal has no choice.

Quite different is the habit of these animals under natural conditions in moving in an environment where light and shade coexist. The animals are accustomed to screens and in the presence of these invariably choose one direction, always the same for each screen. One might say that they have learned to choose. This has been very well shown by the experiments of Bohn on the little gasteropods of the genus Littorina, and those of the same biologists upon the caterpillar of Tyria jacobœœ (see page 70). With these animals, and surely with all those whose organization has at least reached a certain point, the responses of differential sensitiveness to normal variations have been adapted and consequently are the result of a natural apprenticeship. This apprenticeship has been served by all the actual individuals who possess differential sensitiveness from birth, it 11 can have taken place, ' ' says Bohn, '1 only in a far distant past" in the course "of many fruitless trials," made already by the species under conditions of environment in which it has accomplished its evolution. When it cannot take place, as with the bugs and the Spirostomas presented with abnormal conditions, the phenomena of differential sensitiveness are produced by chance, without following a precise modality, without adaptation.