S8S Mr Anderson’s Corrections for the effects of Humidity 
succession of inclined planes, the lengths and inclinations of 
which were carefully measured, and connected together at their 
extremities by levelling, where the irregularity of the ground 
rendered it necessary. The measurement was repeated, by 
employing a different series of lines of ascent, and the result of 
both was such as to satisfy me that I had determined the differ- 
ence of elevation, between the two stations, within a few inches 
of the truth. To remove, however, as much as possible, the 
suspicion of error, I measured with great care, on the meadow 
below, a fine base line, having a slight inclination, and ly- 
ing in the same vertical plain with the summit of the hill ; and 
then, with an excellent theodolite, constructed by Adie, which 
read off to a minute of a degree, I took the angles of elevation 
at each extremity of the base line, and from the data thus ob- 
tained, calculated the -height geometrically. The result, with a 
due allowance for refraction and curvature, agreed within two 
feet of the former measurement ; — a coincidence as near as could 
be expected, from the graduation of the theodolite. 
Having thus determined the difference of elevation between 
the stations I had fixed upon, with sufficient accuracy for my 
purpose, I continued from day to day, for a course of several 
weeks, to make observations in every state of the weather which 
the season presented in the months of August and September ; 
and, on comparing the results deduced by the formula, correct- 
ed as I have proposed, with those furnished by the formula of 
Sir G. Shuckburgh and General Roy, as well as by that of La- 
place, no doubt remained on my mind with respect to the ne- 
cessity of employing the new coefficients, under some form such 
as I have proposed ; by applying them, at least, under the form 
I have given them, the coincidence of the results which were 
deduced from observations made in very different states of the 
air, in point of temperature and humidity, with the heights de- 
termined geometrically by the methods I have described, was 
the more entitled to confidence, as the altitudes computed by the 
common formula deviated, in many cases (particularly when 
the air was in an extreme state of dryness or humidity,) up- 
wards of SO feet from the true height. 
Not having an assistant to make contemporaneous observa- 
tions at the lower station, at the instant of time I was engaged 
