^54 Mr Anderson’s Corrections for the Effects of Humidity 
From this result it appears, that the intermixture of vapour 
with air, provided the density of both fluids decreased in the 
same ratio, should have no effect upon the expression already 
obtained for h, beyond rendering necessary a slight accommoda- 
tion of the coefficient to make it apply to the diminished 
density of the air. This at least would be the case, so far as 
pressure is concerned ; though it will afterwards appear, that the 
presence of vapour requires another correction, on account of 
the dilatation of the air which it produces, — a circumstance which 
has been entirely overlooked by all who have written on baro- 
metrical measurements. 
The supposition, however, that the density of atmospheric 
vapour diminishes in the same ratio as dry air, is far from be- 
ing conformable to observation. The diminution of tempera- 
ture as we ascend in the atmosphere ; the frequent precipitation 
of moisture from the strata of it contiguous to the surface of the 
earth, by which the ascent of moisture is continually checked, — 
and even the very form of the atmospherical columns, which 
may be regarded as pyramidal frusta, having the earth’s centre 
for a common apex, — all conspire to render the more elevated 
strata of the atmosphere absolutely, as well as relatively, drier, 
than those which press upon the surface of the ocean. 1 have 
elsewhere * endeavoured to shew, by general considerations, that 
the absolute humidity of the air is reduced one half, by an as- 
cent of about 4500 feet ; and by some observations which I 
lately made, under very favourable circumstances, from the base 
to the summit of Bengloe, one of the loftiest mountains of Perth- 
shire, I ascertained that the line of humidity was a species of 
logarithmic curve, according to 'which it appeared, that the 
density of atmospheric vapour, (regarded as ordinates to the 
heights, as abscissae), accorded nearly with the deductions I 
obtained from theory. Various observations which I made, 
about the same time, on the sides of Benvracky, the lofty hill 
which overlooks the Pass of Killikrankie, and terminates the 
northern extremity of the vale of Athole, afforded results which 
agreed almost exactly with the conclusions which had been 
deduced from the observations made on Bengloe, — a result the 
* Edin. Encyc. Art» Ilygrometr^ Sec. 97. 
