On the left side of the tube there is another strip of brass, with another set of numbers, so situated as to comprise two decrees of the scale above mentioned, in one. Agreeably to this enumeration the height of the tubes is, by the aid of a corre spondent graduation on the vernier, divided into 1100 parts or degrees. A small strip of sheet-tin k is let into a kerf in the wood, supporting the tubes, in order to indicate the commencement of the scale ; and the depth to which the orifices of the tubes must extend. At distances from this of 1000 parts and 2000 parts, (commensurate with those of the scale,) there are two other indices, TT, to the right hand tube. Let a small vessel containing water be made to receive the lower end of the tube, by the side of which the scale is situated ; and a similar vessel of any other fluid, whose gravity is sought, be made to receive the lower end of the other tube ; so that the end of the one tube may be covered by the liquid in question, and the end of the other tube by the water. The bag being compressed, a great part of the contained air is expelled through the tubes, and rises through the liquid in the tumblers. When the bag is allowed to resume its shape, the consequent rarefaction allows the-liquids to rise into the tubes, in obedience to the greater pressure of the atmosphere without- If the liquid to be assayed be heavier than water (as, for instance, let it be con centrated sulphuric acid,) it should be raised a little , above the first index, at the distance of 1000 degrees from the common level of the orifices of the tubes. The vessels holding A the liquids being then removed, so 2 that the result may be uninfluenced by any inequality in the height of the liquids, the column of acid must be lowered, until its upper surface co incide exactly with the index 1000. Opposite the upper surface of the column of water the two first numbers of specific gravity of the acid will then be found ; and, by duly adjusting and inspecting the vernier, the third figure will be ascertained. The liquids should be at the temperature of 60°. If the liquid under examination be lighter than water, as in the case of pure alcohol, it must be raised to the upper index. The column of water, measured by the scale of 1000, will then be found at 800 nearly ; which shows that 1000 parts of alcohol are, in weight, equivalent to 800 parts of water ; or, in other words, 800 is ascertained to be the specific gravity of the alcohoL The sliding rod and tube at r, between the cock and the glass tubes, facilitates the adjustment to the index of the column of liquid in the right hand glass tube.
When the rod is pushed in as far as possible, it causes a small leak, by which the air enters ; and the columns of the liquids, previously raised too high by the beg, may be allowed to fall, till the liquid which is to be assayed is near the index ; then, by pushing the rod in, they may be gradually lowered, and adjusted to the proper height, with great accuracy. A rod of this kind, graduated, might answer the purpose of a vernier.
Meikles Syphon Hydrometer, represented in the annexed engraving, it will be seen, is upon this same principle, but of a much simpler construction. This instrument consists of a glass tube, open at both ends, and bent into a kind of double syphon, having four parallel legs; so that the open ends are pointed in the same direction, or upwards, as in the annexed engraving. The manner of using it is very simple: let one of the ends be stopped with a finger or cock, and water be poured into the fluid will only rise a small way into the second leg, because of the included air : next, stop the other orifice, and open the one first closed; and having poured into the latter the liquid whose specific gravity is to be tried, open the to of the water tube ; then the instrument being held upright, the two liquids will arrange themselves so as to press equally on the included air. This pressure will be measured by the difference in the heights of the two columns of either liquids multiplied by its specific gravity, so that by dividing the difference of the two columns of water by the difference of those of the other liquid, we obtain the specific gravity of the latter; that of water being unity. The difference between the columns may be measured by applying any scale of small equal parts, or the glass may be attached to a graduated plate furnished with verniers, &c. The longer the columns of liquid employed, the more accurate the process. The expansion of the glass or its capillary action cannot affect the result, nor is it influenced by the expansion of the scale; the only correction required will be to reduce the observations to one temperature.