[ 28z ]

posed to fill the bore of the pipe ; a less quantity, with aless velocity, must leave a vacancy, which will suffer partof the air to escape ; and a greater quantity, with a greatervelocity, must have some part of it spent ineffectually, forwant os sufficient room to spread. Some experiments men-tioned hereafter afford a clear proof of this.

The force of the air may be determined in an easierand more simple method, by means of a glass pipe, openat both ends, with one end fixed in a bason of water.The bason may be hung in the upper part of the tub orair-veflel of our water machines, and the glass pipe letinto it through a hole in the top, what space may remainbetween the pipe and the hole being properly closed :the pressure of the air on the surface of the fluid in thebason, forces part of it up into the pipe ; and this ascentwill always be the measure of the power or density of theair. Water is here greatly preferable to the quicksilverused in the same intention on other occasions, as it dis-covers smaller variations in the force ; for being fourteentimes less ponderous than quicksilver, an equal pressureforces it fourteen times higher in the pipe : the wholeascent of quicksilver, by the pressure of the air in bellows,is so small, as frequently not to exceed that part of thepipe which is inserted into the tub. Instead of a glasspipe, a copper or iron one may be used j and the ascentof the water measured, either by occasionally dipping arod in it, or by means of a hollow copper ball, or otherfloating body, with a stem standing out of the pipe, anda proper weight below to keep it upright. It mustbe observed, that the height of the water in the pipe is tobe estimated from the surface of the water in the bason :whence the pipe ought to be of small bore in proportionto the bason, that the water may not fall considerably in thebason by the loss of that which rises in the pipe.

Dr.