Contact with a solid body favours the formation of stolons. Fig. 36 shows a piece of a stem of _Pennaria_ another hydroid, which was lying on the bottom of an aquarium and which formed stolons at both ends _a_ and _b_. In _Margelis_, another hydroid, the writer observed that without any operation the apical ends of branches which were in contact with solid bodies continued to grow as stolons, while those surrounded by sea water continued to grow as stems. [Illustration: FIG. 36] Herbst discovered a very interesting form of heteromorphosis in certain crustaceans; namely, that in the place of an eye which was cut off, an entirely different organ could be formed, namely, an antenna. He showed that the experimenter has it in his power to determine whether the crustacean shall regenerate an eye or an antenna in place of the eye. The latter will take place when the optic ganglion is removed with the eye, the former when it is not removed. These experiments were carried out successfully on _Palæmon_, _Palæmonetes_, _Sicyonia_, _Palinurus_, and other crustaceans. The influence of gravitation is very familiar in plants; in stems of _Bryophyllum_ placed horizontally the roots usually come out from the lower end of the callus. Such phenomena are not often found in animals but they exist here too as the following observation shows. [Illustration: FIG. 37] If we cut a piece _a b_ (Fig. 37), from the stem _s s_ of _Antennularia antennina_ (Fig. 38), a hydroid, and put it into the water in a horizontal position, new stems _c d_ (Fig. 37) may arise on its upper side. The small branches on the under side of the old stem _a b_ begin suddenly to grow vertically downward.[180] In appearance and function these downward-growing elements are entirely different from the branches of the normal _Antennularia_; they are roots. In order to understand better the transformation which thus occurs in these branches, it may be stated that under normal conditions they have a limited growth (see Fig. 38), are directed upward, and have polyps on their upper side. The parts which grow down (Fig. 37) have no polyps, but attach themselves like true roots to solid bodies. Thus the changed position of the stem alone, without any operation, suffices to transform the lateral branches, whose growth is limited, into roots with unlimited growth. The lateral branches on the upper side of the stem do not undergo such a transformation into roots except in the immediate surroundings of the place where a new stem arises. It seems that the formation of a new stem also causes an excessive growth of roots, possibly because the formation of new branches causes the removal of substances which naturally inhibit the formation of roots. If a piece from the stem be put vertically into the water with top downward, the uppermost point may continue to grow as a stem, while the lowest point may give rise to roots. In this case, therefore, a change