After a couple of hours of marching and countermarching, the fragmentary processions, comprising each a score of caterpillars, reach the foot of the wall. Here the soil is powdery, very dry, easy to burrow in, although made somewhat firmer by tufts of grass. The caterpillar at the head of the row explores with his mandibles, digs a little, investigates the nature of the ground. The others, trusting their leader, follow him with docility, making no attempts of their own. Whatever the foremost decides will be adopted by all. Here, in the choice of a matter so important as the spot whereat the transformation shall take place, there is no individual initiative. There is only one will, the leader’s. There is only one head, so to speak; the procession may be compared with the chain of segments of an enormous worm. Finally some spot is recognized as propitious. The leading caterpillar halts, pushes with his head, digs with his mandibles. The others, still in a continuous line, arrive one by one and likewise come to a halt. Then the file breaks up into a swarming heap, in which each of the caterpillars resumes his liberty. All their backs are joggling pell-mell; all their heads are plunged into the dust; all their feet are raking, all their mandibles excavating the soil. The worm has chopped itself into a gang of independent workers. An excavation is formed in which, little by little, the caterpillars bury themselves. For some time to come, the undermined soil cracks and rises and covers itself with little mole-hills; then all is still. The caterpillars have descended to a depth of three inches. This is as far as the roughness of the soil permits them to go. In looser soil, the excavation would attain a much greater depth. The greenhouse shelf, supplied with fine sand, has provided me with cocoons placed at a depth of from eight to twelve inches. I would not assert that the interment might not be made still lower down. For the most part, the burial is effected in common, by more or less numerous clusters and at depths which vary greatly, according to the nature of the soil. A fortnight later, let us dig at the point where the descent underground was made. Here we shall find the cocoons assembled in bunches, cocoons of sorry appearance, soiled as they are with earthy particles held by silken threads. When stripped of their rough exterior, they are not without a certain elegance. They are narrow ellipsoids, pointed at both ends, measuring twenty-five millimetres in length and nine millimetres [21] in thickness. The silk of which they are composed is very fine and of a dull white. The fragility of the walls is remarkable when we have seen the enormous quantity of silk expended on the construction of the nest.