And yet I am assailed by doubts. To break down the screen that still separated it from the outer air, the larva has expended its last reserves of fluid. The flask is dry; and there is no way of replenishing it in the absence of a live root. My suspicion of failure is well-founded. For three days I see the entombed one wasting itself in efforts without succeeding in rising an inch higher. The materials removed refuse to stay in position for lack of anything to bind them; they are no sooner pushed aside than they slip down again under the insect’s legs. The labour has no perceptible result and has always to be done all over again. On the fourth day, the creature dies. With the water-can full, the result is quite different. I subject to the same experiment an insect whose work of self-deliverance is just beginning. It is all swollen with urinary humours which ooze out and moisten its whole body. This one’s task is easy. The materials offer hardly any resistance. A little moisture, supplied by the miner’s flask, converts them into mud, sticks them together and keeps them out of the way. The passage is opened, very irregular in shape, it is true, and almost filled up at the back as the ascent proceeds. It is as though the larva, recognizing the impossibility of renewing its store of fluid, were saving up the little which it possesses and spending no more than is strictly necessary to enable it to escape as quickly as possible from its unfamiliar surroundings. This economy is so well arranged that the insect reaches the surface at the end of ten days. CHAPTER III THE CICADA: THE TRANSFORMATION The exit-gate is passed and left wide open, like a hole made with a large gimlet. For some time the larva wanders about the neighbourhood, looking for some aerial support, a tiny bush, a tuft of thyme, a blade of grass or the twig of a shrub. It finds it, climbs up and, head upwards, clings to it firmly with the claws of the fore-feet, which close and do not let go again. The other legs take part in sustaining it, if the position of the branch make this possible; if not, the two claws suffice. There follows a moment of rest to allow the supporting arms to stiffen into an immovable grip. First, the mesothorax splits along the middle of the back. The edges of the slit separate slowly and reveal the pale-green colour of the insect. Almost immediately afterwards, the prothorax splits also. The longitudinal fissure reaches the back of the head above and the metathorax below, without spreading farther. The wrapper of the skull breaks crosswise, in front of the eyes; and the red stemmata appear. The green portion uncovered by these ruptures swells and protrudes over the whole of the mesothorax. We see slow palpitations, alternate contractions and distensions due to the ebb and flow of the blood. This hernia, working at first out of sight, is the wedge that made the cuirass split along two crossed lines of least resistance.