13. Mr Leslie, who has devoted some attention to this instrument, has proposed a modification, which, as he himself remarks, has perhaps carried the hygroscope thus formed to as high a state of improvement as such an imper fect instrument admits. The shell of ivory is turned, in his construction, into an elongated spheroid, about an inch and a quarter in length, and reduced so thin as weigh only eight or ten grains. At its greatest expansion it con tains about 300 grains of mercury. The upper end, which is adapted to the body by means of a delicate screw, has a slender glass tube inserted into it, six or eight inches long, and a bore nearly -Nth part of an inch in diameter. The point of extreme humidity is determined, as in the case of De Luc's instrument, by immersing the bblb in water, or surrounding it with a wet bit of cambric. The divisions of the scale, however, are determined somewhat differently : Mr Leslie distinguishes the tube into spaces, which corre spond to the thousandth part of the entire cavity, and each of which contains about of a grain of mercury. The ordinary range of the scale includes about 70 of these di visions. The upper extremity of the tube is covered with a small ivory cap, which admits the air, but prevents the escape of the mercury, thus rendering the instrument por table. Mr Leslie remarks, that the contraction of the mercury corresponding to equal increase in the dryness of the air, is six times greater at the beginning of the scale than at the 70th division ; and that it seems to be in gene ral inversely as the number of hygrometric degrees, reck oning below. He therefore places another scale on the opposite side of the tube, the interval between zero and 70° being divided into 100°, and corresponding to the un equal portions, from the number 20 to 120 in a logarithmic line, (see Plate CCCXX V. Fig. 5. By extending the lo garithmic divisions farther in conformity with the base of the scale, 320 of its degrees would correspond to 108 of the equable divisions, or a contraction of 108 parts in a thousand, with respect to the capacity of the bulb. At the dryness, however, of 300 of his own hygrometer, Mr Leslie never found the contraction of the ivory to exceed 105. It would have been more satisfactory, if the temperature at which the observations were made had been given, as we shall afterwards show that 300, on Mr Leslie's hygrometer, may correspOnd to very cilfelent A'2 portions of moisture in the medium to which it is ex posed.
14. The instrument we have described, though very un fit for delicate observations, may nevertheless be used in i certain cases with advantage. The slowness of its ndica tions, when its scale has once been compared with that of a more accurate instrument, is well fitted to point out ge ral results, corresponding to considerable intervals of time between the observations. Mr Leslie has suggested, that on this account, it may be usefully employed to ascertain the degree of humidity which prevails in the higher re gions of the atmosphere, and to determine the hygroscopic state of certain kinds of goods, such as grain, wool, cotton, See. For the latter purpose, all that is necessary is to thrust the instrument among the substances, whose condi tion, with respect to moisture, we wish to determine, and to observe the degree which it indicates after it has been al lowed to remain a suitable length of time.
15. Under hygroscopic instruments constructed on the principle of a general dilatation by moisture, we may briefly notice the hygrometer of Chiminello, to whom the prize proposed in 1783 by the Academy of Sciences at Manheim for the best comparable hygrometer was adjudged. The
substance lie employed was the banel of a quill, fitted up in the same manner as the ivory hygrometers, already de scribed. The graduation of the scale was determined by means of two fixed points. The point oicxtreme moisture was obtained by immersien in water ; that of extreme dry ness, by exposing the instrument, for the space of four hours, before a moderate fire, at a temperature equal to 25° of Reaumur, or of Fahrenheit. The quill, by ex posure to heat, becomes somewhat contracted ; and though the contraction is not so great as would be produced by ex treme dryness, Chiminello considered it sufficiently uni form in different quills to serve as a fixed point in his scale. It is obvious, however, that an instrument graduat ed in so vague a manner is totally unfit for any philosophi cal purpose.
16. Having alluded to this instrument, it may be worthy of remark, that it is probably owing to the hygroscopic pro perty of their feathers that birds are enabled to judge of approaching rain or fair weather. For it is easy to con ceive, that an animal having a thousand hygrometers inti mately connected with its body, must be liable to be power fully affected, with regard to the tone of its organs, by very slight changes in the dryness or humidity of the air; particu larly when it is considered, that many of the feathers contain a large quantity of blood, which must thus be alternately propelled into the system, or withdrawn from it, to their contraction or dilatation by dryness or moisture. This view of the subject seems to afford a satisfactory ex planation of the extreme sensibility which birds in general shew to coming changes in the weather.
17. An hygrometer constructed upon similar princples, but much more delicate in its indication's, has lately linen proposed by Mu Wilson of Dublin. The substance he employs is rat's bladder; which, besick s having an exten sive range of dilaiatien, is affected slight changes In the 1 ygroscopic state of the air. TI.e scale is gr adulard by exposing the instrument to air saturated with moisture for the point of extreme humidity ; and by afterwards in closing it in a receiver, over mercury, with a quantity of concentrated sulphuric acid, for the point of extreme dry ness. The interval between the range of these two points is then divided into 100 equal parts. These points, how ever, must obviously vary with the temperature of the mer cury ; and though this objection, which is applicable to all these mercurial hygrometers, may be obviated to a certain extent by enlarging the diameter of the tube, it cannot be entirely removed without diminishing the delicacy of the instrument, unless the precautions, adopted by De Luc, be observed in the construction of the scale. The rat's blad der hygrometer is liable to another objection, which, on ac count of the difficulty attending the construction of the in strument, is of considerable force ; the elevation of the mer cury in the tube, by the contraction of the membranous substance, must occasion a pressure in the bladder, which, in some cases, may amount to nine or ten pounds on the square inch, according to the range of the scale. The dis tension occasioned by such a pressure cannot fail to affect the accuracy of the instrument, and even to expose it to destruction. According to the observations of Lord Gray, this instrument corresponds pretty nearly in its indications with the whale-hone hygrometer of De Luc.