The sand or earth in the zinc pan at the bottom of the breeding cage should be kept constantly moistened, and in the case of hibernating pupae the constant adding of water to the top of the earth or sand causes it to become very hard and compact. To overcome this objection it was suggested in the _Entomologists' Monthly Magazine_ for June, 1876, page 17, that the base should be made with an inner perforated side, the water to be applied between it and the outer side, and I have for some years employed a similar double-sided base, which answers the purpose admirably (See Figure 124). It is substantially the same as that made for the Department by Prof. J. H. Comstock in 1879. It consists of a zinc tray _a_, of two or three inches greater diameter than the breeding cage, which surrounds the zinc pan proper containing the earth, and the tube _d_ for the reception of the food-plant. The lower portion of the inner pan _b_ is of perforated zinc. Zinc supports, _c c_, are constructed about halfway between the bottom and the top of this pan, on which the breeding cage rests. In moistening the earth in the cage, water is poured into the tray, which enters the soil slowly, through the perforations in the zinc pan. I have found this modification of very decided advantage and use it altogether in the work of the Division, and heartily recommend it. [Illustration: FIG. 123.--Insect breeding-cage or vivarium.] The base of the vivarium or breeding cage should never be made of tin, but always of zinc. If made of tin, it will soon rust out. Galvanized iron may be used in place of the zinc, and will doubtless prove equally satisfactory. "A dozen such cages will furnish room for the annual breeding of a great number of species, as several having different habits and appearance, and which there is no danger of confounding, may be simultaneously fed in the same cage. I number each of the three parts of each cage to prevent misplacement and to facilitate reference, and aside from the notes made in the notebook, it will aid the memory and expedite matters to keep a short open record of the species contained in each cage, by means of slips of paper pasted on the glass door. As fast as the different specimens complete their transformations and are taken from the cage the notes may be altered or erased, or the slips wetted and removed entirely. To prevent possible confounding of the different species which enter the ground, it is well, from time to time, to sift the earth, separate the pupae and place them in what I call 'imago cages,' used for this purpose alone and not for feeding. Here they may be arranged with references to their exact whereabouts. [Illustration: FIG. 124.--Improved base for breeding-cage (original).]