In Mr. Leslie 's method this source of error is avoided, and some of the results obtained in consequence are extremely curious. The instrument employed consists of a glass tube ac, about three feet long, and open at both ends ; the wide part a b is about fonr tenths of an inch in diameter; the part b c about two-tenths.
The two parts communicate at b by an extremely fine slit, which suffers air to pass, but retains sand or powder. The mouth at a is ground smooth, and can be shut so es to be air light by • smell glass plate 1 The substance whose specific gravity we wish to find is but reduced to powder, which is then put into the wide part of the tube a b, which may either be filled or not. The tube being then held in a vertical position has the narrow part immersed in mercury, contained in an open vessel z, till the metal rises within to the gorge b. The lid is then fitted on airtight at a. In this state it is evident there is no air in the tube, except that mixed with the powder in the cavity a b. Suppose the barometer at the time to stand at 30 inches, and that the tube is lifted perpendicularly upwards, till the mercury stands in the inside of b at apoint e 15 inches, (or one half of i 30,) above its surface, in the open vessel% it is evident, then, that the air in the inside of the tube is subjected to a pressure of exactly half an atmosphere: and of course it dilates and fills precisely twice the space it originally occupied. It follows, too, that since the air is dilated to twice its bulk the cavity a b contains just half what it did at first ; and the cavity b e now containing the other half, the quantity of air in each of these parts of the tube is equal. In other words, the quantity of air in b e is exactly equal to what is mixed with the powder in a b, and occupies precisely the same space which the whole occupied before its dilatation. Let ns now suppose the powder to be taken out, and the same expe riment repeated, but with this difference, that the cavity a b is filled with air only. It is obvious that the quantity being greater it will, when dilated to double the bulk under a pressure of fifteen inches, occupy a larger space, and the mercury will rise, let us suppose, only to d. But the attenuated air in the
narrow tube always occupies exactly the space which the whole occupied at ordinary atmospheric pressure ; and this space is therefore, in the one case, the cavity b e, and in the other b d. Hence it follows that the cavity e al, which is the difference between these, is equal to the bulk of the solid matter in the sand. Now by marking the number of grains of water held by the narrow tube b e on a graduated scale attached to it, we can find at once what is the weight of a quantity of water, equal in bulk to the solid matter in the sand; and by comparing this with the weight of the sand, we have its true specific gravity. Aware that some solid bodies, such as charcoal, hold much condensed air in their pores, and that probably they retain part of this even when reduced to powder, Professor Leslie obviates the chances of error arising from this source by comparing the dilatation which takes place under different degrees of pressure, under 10 inches and 20 for instance, or 71 and 15. Charcoal, from its porosity, is so light that its specific gravity, as assigned in books, is generally under 0.5 less than half the weight of water, or one seventh the weight of diamond; taken in powder by the above instrument it exceeds that of diamond, is one half greater than that of wbinstone, and is, of course, more than seven times heavier than has usually been supposed. Mahogany is generally estimated at 1.36 ; but mahogany sawdust proves by the instrument to be 1.68 ; wheat flour is 1.46; pounded sugar 1.83 ; and common salt 2.15 ; the last agrees very accurately with the common estimate. Writing paper rolled hard by the hand had a specific gravity of 1.78, the solid matter present being less than one third of the space it apparently filled. One of the most remarkable results was with an apparently very light specimen of volcanic ashes, which was found to have a specific gravity of 4.4. These results are, however, given as approximations merely by the first instrument constructed.