I fixt a pipe of an inch and a half diameter to a pair of double bellows three feet diameter; which pipe reached about one foot under the surface of the water in the boiler; to the end of which are fixed horizontally two branches, each about eight feet long, tapering from one inch diameter to about ¼ of an inch. These branches are bent in a circular manner, as in the plan, to answer the form of the concave, and are perforated with small holes about four inches distant at the thickest part, and decreasing gradually in distance, to within ¼ of an inch, towards the small end. The reason of these branches being made taper, and the distance between the holes decreasing to the small end, was in order to give the greater power to the air forced by the bellows to discharge the water lodged in such a length of pipe; and I observed by this method, that the water was gradually forced thro’ the holes to the end of each branch, and seemed to throw an equal quantity of air thro’ the water. The length of the pipe, to which the branches are fixed horizontally, is about 18 feet to the nosle of the bellows: notwithstanding which length, the steam, that passed thro’ the pipe into the bellows, was so hot before the water boiled, as to force thro’ the leather: but this I easily remedied, by fixing a brass cock of one inch and a half diameter to the pipe, which hindered the steam from ascending, until the engine was ready to work; and being opened, the air continually keeps it cold until the engine has done working; then the cock must be shut again. The bellows is worked by means of a small lever, and pullies applied to the great lever of the fire-engine, which keeps a continual blast whilst the engine works; the strength of which is increased or diminished, by adding or taking off the weights on the bellows. The effect produced, according to the best observations I could make, was, first, a very visible alteration for the better in the working of the engine. When the fire was stirred, as it must be every time fuel is added, the steam generally became too fierce, which occasioned great irregularity, and sometimes, if not watched, great damage to the engine; and when the fire abated, the stroke became immediately much shorter, or stopped intirely, if fuel was not soon added: whereas, by blowing air thus thro’ the water, it keeps, with any moderate care, an equal stroke to its full length, from the beginning to the end; and by that means discharges a considerably greater quantity of water. A proof of which was very evident, tho’ I could not ascertain the exact quantity: for the engine, before this improvement, supplied but two main pipes at once, which conveyed the water to the houses served by them; but since could not take off the quantity of water thrown up, part of which was obliged to be discharged into a third main.