For experiments in which air will bear to be confined by water, I first used an oblong trough made of earthen ware, as _a_ fig. 1. about eight inches deep, at one end of which I put thin flat stones, _b. b._ about an inch, or half an inch, under the water, using more or fewer of them according to the quantity of water in the trough. But I have since found it more convenient to use a larger wooden trough, of the same general shape, eleven inches deep, two feet long, and 1-1/2 wide, with a shelf about an inch lower than the top, instead of the flat stones above-mentioned. This trough being larger than the former, I have no occasion to make provision for the water being higher or lower, the bulk of a jar or two not making so great a difference as did before. The several kinds of air I usually keep in _cylindrical jars_, as _c_, _c_, fig. 1, about ten inches long, and 2-1/2 wide, being such as I have generally used for electrical batteries, but I have likewise vessels of very different forms and sizes, adapted to particular experiments. When I want to remove vessels of air from the large trough, I place them in _pots_ or _dishes_, of various sizes, to hold more or less water, according to the time that I have occasion to keep the air, as fig. 2. These I plunge in water, and slide the jars into them; after which they may be taken out together, and be set wherever it shall be most convenient. For the purpose of merely removing a jar of air from one place to another, where it is not to stand longer than a few days, I make use of common _tea-dishes_, which will hold water enough for that time, unless the air be in a state of diminution, by means of any process that is going on in it. If I want to try whether an animal will live in any kind of air, I first put the air into a small vessel, just large enough to give it room to stretch itself; and as I generally make use of _mice_ for this purpose, I have found it very convenient to use the hollow part of a tall beer-glass, _d_ fig. 1, which contains between two and three ounce measures of air. In this vessel a mouse will live twenty minutes, or half an hour.