If I want to expel air from any liquid, I nearly fill a phial with it, and having a cork perforated, I put through it, and secure with cement, a glass tube, bended in the manner represented at _e_ fig. 1. I then put the phial into a kettle of water, which I set upon the fire and make to boil. The air expelled by the heat, from the liquor contained in the phial, issues through the tube, and is received in the bason of quicksilver, fig. 7. Instead of this suspended bason, I sometimes content myself with tying a flaccid bladder to the end of the tube, in both these processes, that it may receive the newly generated air. In experiments on those kinds of air which are readily imbibed by water, I always make use of quicksilver, in the manner represented fig. 8, in which _a_ is the bason of quicksilver, _b_ a glass vessel containing quicksilver, with its mouth immersed in it, _c_ a phial containing the ingredients from which the air is to be produced; and _d_ is a small recipient, or glass vessel designed to receive and intercept any liquor that may be discharged along with the air, which is to be transmitted free from any moisture into the vessel _b_. If there be no apprehension of moisture, I make use of the glass tube only, without any recipient, in the manner represented _e_ fig. 1. In order to invert the vessel _b_, I first fill it with quicksilver, and then carefully cover the mouth of it with a piece of soft leather; after which it may be turned upside down without any danger of admitting the air, and the leather may be withdrawn when it is plunged in the quicksilver. In order to generate air by the solution of metals, or any process of a similar nature, I put the materials into a phial, prepared in the manner represented at _e_ fig. 1, and put the end of the glass tube under the mouth of any vessel into which I want to convey the air. If heat be necessary I can easily apply to it a candle, or a red hot poker while it hangs in this position. When I have occasion to transfer air from a jar standing in the trough of water to a vessel standing in quicksilver, or in any other situation whatever, I make use of the contrivance represented fig. 9, which consists of a bladder, furnished at one end with a small glass tube bended, and at the other with a cork, perforated so as just to admit the small end of a funnel. When the common air is carefully pressed out of this bladder, and the funnel is thrust tightly into the cork, it may be filled with any kind of air as easily as a glass jar; and then a string being tied above the cork in which the funnel is inserted, and the orifice in the other cork closed, by pressing the bladder against it, it may be carried to any place, and if the tube be carefully wiped, the air may be conveyed quite free from moisture through a body of quicksilver, or any thing else. A little practice will make this very useful manoeuvre perfectly easy and accurate.