meafuring Heights with the Barometer. 713 
from the zero of Fahrenheit, as far as 1 5 2 0 or 172°; 
and even to the boiling point, in thofe which I efteem the 
heft experiments. By adhering to the mean refult it will 
appear that air, however moift, having that moifture 
condenfed or feparated from it by cold, its expanlion dif- 
fers not fenlibly from that of dry air. Thus the rate for 
32 0 below freezing 2.22799, is nearly the fame as in dry 
air; but no fooner doth the moifture begin to diflblve 
and mix with the air, by the addition of 20° of heat, 
than the difference is perceptible : for inftead of 2.4667 5, 
the rate for 20° above 32 0 in dry air, we have 2.588 for 
that which is moift. In the next ftep of 20°, the rate for 
dry air is 2.5809; whereas that for moift is 2.97. In 
this manner the progreflion goes on continually encreaf- 
ing, fo as to give 7.86854 for the mean rate on each de- 
gree of the 2 1 2 0 , which is near three times and a half 
the expanlion of dry air. And laftly, the rate for the 
20° between 192 0 and 212 0 , is twice and one-half the 
mean rate, and about nine times that which correfponds 
to the zero of the fcale : but if the comparifon is drawn 
from the mean of the fifth, fixth, and ninth experi- 
ments, as being probably neareft the truth, the total ex- 
panfion of moift, will be more than four times that 
of dry air ; and the rate for the temperature at boiling, 
4 Y 2 will 
