Let us hark back for a moment to the gymnastic feat which enables the Cicada to leave his scabbard. At first retained by the tip of the abdomen, which is the last part to remain in its case, the Cicada turns over perpendicularly, head downwards. This somersault allows him to free his wings and legs, after the head and chest have already made their appearance by cracking the armour under the pressure of a hernia. Now comes the time to free the end of the abdomen, the pivot of this inverted attitude. For this purpose, the insect, with a laborious movement of its back, draws itself up, brings its head to the top again and hooks itself with its fore-claws to the cast skin. A fresh support is thus obtained, enabling it to pull the tip of its abdomen from its sheath. There are therefore two means of support: first the end of the belly and then the front claws; and there are two principal movements: in the first place the downward somersault, in the second place the return to the normal position. These gymnastics demand that the larva shall fix itself to a twig, head upwards, and that it shall have a free space beneath it. Suppose that these conditions were lacking, thanks to my wiles: what would happen? That remained to be seen. I tie a thread to the end of one of the hind-legs and hang the larva up in the peaceful atmosphere of a test-tube. My thread is a plumb-line which will remain vertical, for there is nothing to interfere with it. In this unwonted posture, which places its head at the bottom at a time when the near approach of the transformation demands that it should be at the top, the unfortunate creature for a long time kicks about and struggles, striving to turn over and to seize with its fore-claws either the thread by which it hangs or one of its own hind-legs. Some of them succeed in their efforts, draw themselves up as best they can, fasten themselves as they wish, despite the difficulty of keeping their balance, and effect their metamorphosis without impediment. Others wear themselves out in vain. They do not catch hold of the thread, they do not bring their heads upwards. Then the transformation is not accomplished. Sometimes the dorsal rupture takes place, leaving bare the mesothorax swollen into a hernia, but the shelling proceeds no farther and the insect soon dies. More often still the larva perishes intact, without the least fissure.