52 On HYGROMETERS. 
more humid than it is in reality, and the contrary in a 
change from warm to cold. 
But if fuch a fuddenly changing inftrument could be 
freed from thefe imperfections, yet when the defign is to 
difcover the different degrees of humidity in the air of dif- 
ferent countries, | apprehend the quick fenfibility of the 
inftrument to be rather a difadvantage; fince, to draw the 
defired conclufions from it, a conftant and frequent obfer- 
vation day and night in each country will be neceflary for 
a year or years, and the mean of each different fet of ob-- 
fervations is to be found and determined. After all which 
fome uncertainty will remain refpeCting the different de- 
grees of exactitude with which different perfons may have 
made and taken notes of their obfervations. 
For thefe reafons, I apprehend that a fubftance which, 
though capable of being diftended by moifture and con- 
tracted by drynefs, is fo flow in receiving and parting with 
its humidity that the frequent changes in the atmo{phere 
have not time to affect it fenfibly, and which therefore 
fhould gradually take nearly the medium of all thofe 
changes and preferve it conftantly, would be the moft pro- 
per fubftance of which to make fuch an hygrometer. 
Such an inftrument, you, my dear fir, though without 
intending it, have made for me; and I, without defiring 
or expecting it, have received from you. It is therefore 
with propriety that I addrefs to you the following account 
of it; and the more, as you have both a head to contrive 
and a hand to execute the means of perfecting it. And I 
do this with greater pleafure, as it affords me the oppor- 
tunity of renewing that antient correfpondence and ac- 
quaintance with you, which to me was always fo pleafing 
and fo inftructive. 
You may poflibly remember, that in or about the year 
1758, you made for me a fet of artificial magnets, fix in 
number, each five and a half inches long, half an inch 
broad, and one eighth of aninch thick, Thefe, with two 
pieces 
