The provisions, therefore, even though they have to last but three or four weeks, are liable to dry up before that time and to become uneatable. When, instead of the soft bread of its first meal, the unhappy grub finds nothing to stay its stomach but a horrible crust, hard as a pebble and tooth-proof, it is bound to perish of hunger. And it does actually so perish. I have found numbers of these victims of the August sun which, after eating plentifully of the fresh food and digging themselves a cell in it, had succumbed, unable to continue biting into provisions that had become too hard. There remained a thick shell, a sort of closed oven, in which the poor thing lay baked and shrivelled up. While the grub dies of hunger in a shell which has dried into stone, the full-grown insect that has completed its transformations dies there too, for it is incapable of bursting the prison and freeing itself. I shall come back later to the question of the final emergence and will say no more about it for the present. Let us confine our attention to the troubles of the grub. The drying-up of the victuals is, I have said, fatal to it. This is proved by the larvæ found baked in their oven; it is also proved, in a more definite fashion, by the following experiment. In July, the period of active nidification, I place in wooden or cardboard boxes a dozen pears unearthed that morning from their native burrows. These boxes, carefully closed, are put away in the dark, in my study, where the same temperature prevails as outside. Well, in none of them is the infant reared: sometimes the egg shrivels; sometimes the worm is hatched, but very soon dies. On the other hand, in tin boxes or glass receptacles, everything goes well: not one attempt at rearing fails. Whence do these differences arise? Simply from this: in the high temperature of July, evaporation proceeds apace under the permeable wooden or cardboard screen; the food-pear dries up; and the unfortunate worm dies of hunger. In the impermeable tin boxes, in the carefully-sealed glass receptacles, there is no evaporation; the provisions retain their softness; and the grubs thrive as well as in their native burrow.