554 LECTURE LVII. 



whalebone, which have been employed by Saussure* and Deluc,t appear 

 to be more certain and accurate in their indications. The hair hygrometer 

 acquires more speedily the degree corresponding to any given state of the 

 air, but it seems to reach the utmost extent of its scale before it arrives at 

 perfect humidity : while the whalebone hygrometer appears to express a 

 greater change upon immersion in water than from the effect of the moist- 

 est transparent air, which has also been considered by some as an imper- 

 fection. Both these instruments are impaired by time, and acquire contrary 

 errors, so that a mean between both is more likely to be correct than either 

 separately. Their indications are at all times widely different from each 

 other, and the mean appears to approach much nearer to a natural scale 

 than either of them. Mr. Leslie J employs a very delicate thermometer, of 

 which the bulb is moistened, for measuring the drjoiess of the air, by the 

 cold produced during evaporation, when the thermometer is exposed to it ; 

 but this mode of estimating the quantity of moisture appears to be liable 

 to considerable uncertainty. (Plate XLI. Fig. 581.) 



In order that the scale of a hygrometer should be perfectly natural, it 

 ought to express, at all temperatures, the proportion of the quantity of 

 moisture in the air to that which is required for its saturation ; thus, at 

 100 degrees, it should imply that the slightest depression of temperature 

 would produce a deposition ; at 50 degrees, that the air contains only half 

 as much water as would saturate it, or, supposing the thermometer at 52, 

 that a deposition would be produced in it by a depression of 17. And 

 if we know the actual temperature, and the temperature at which the depo- 

 sition takes place, we may find the height of the natural hygrometer, by 

 the proportion of the corresponding elasticities of steam. The mean 

 height of the natural hygrometer in London is probably about 80 ; that 

 of Deluc's hygrometer, with proper corrections, being nearly 70: so 

 that a depression of 6 must usually be sufficient to cause a deposition of 

 moisture. 



The quantity of water actually contained in a cubic foot of air, satu- 

 rated with moisture, appears to be about 2 grains at the freezing point, 

 4 grains at 48, 6 at 60, and 8 at 68 ; and the density of the vapour, 

 thus mixed with air, is, according to Saussure's experiments, about three 

 fourths as great as that of the air itself ; so that moist air is always a little 

 lighter than dry air ; and the more so as the air is warmer, provided that 

 it be saturated with moisture by means of the presence of water. It follows 

 from the properties of moisture thus determined, that if any two portions 

 of perfectly humid air, at different temperatures, be mixed together, there 

 must be a precipitation : thus, a cubic foot of air at 32 being mixed with 

 another at 60, their common temperature must be 46 ; if they are satu- 



* Essai sur 1'Hygrometrie, Neuch. 1783. Jour, de Phy. xxxii. 24, 98. 



t Ph. Tr. 1791, 1, 389. Jour, de Ph. xxx. 437 ; xxxii. 132. 



J Nich. Jour. iii. 401. A short Account of Instruments depending on the Rela- 

 tions of Air to Heat and Moisture, Edin. 1813. Descrip. of Instrs. for Improving 

 Meteor. Obs. Edin. 1820. See on this subject Forbes's Supplementary Report on 

 Meteor. Brit. Ass. 1840, p. 95. 



This phrase must not be supposed to imply any combination between the air 

 and vapour. 



