552 PROFESSOR DANIELL ON THE WATER-BAROMETER ERECTED 
The most striking result of this comparison is, the almost exact coincidence 
in the first ten observations of the elasticity of the aqueous vapour, derived 
from the experiment, with the amount as determined from calculation in a 
range of temperature from 58° to 74° ; the differences in the eleventh column 
being much less than I should have anticipated, even from the necessary un¬ 
certainty in ascertaining the temperature by the thermometers. 
The remaining series exhibit larger and rather increasing differences, but 
such only as might fairly be supposed to come within the limits of errors of 
observation. It must also be observed that they were taken at greater inter¬ 
vals apart, a circumstance which I shall presently show may have had a con¬ 
siderable influence upon the results. The differences in the last column are, 
however, all, except the first, marked with the positive sign -j-, denoting that 
the depression from observation is invariably greater than that which would 
have resulted from the calculated elasticity of the vapour. This would rather 
indicate some constant error in some of the data of the calculation than the 
necessarily fluctuating errors of observation; and we should only have to 
assume the specific gravity of mercury as 13’590 instead of 13‘624, and the 
mean difference would disappear. There can, therefore, I think, be no hesi¬ 
tation in coming to the conclusion that, considering the difficulty and com¬ 
plexity of the several adjustments, and the variety of the necessary corrections 
applied to the observations, the whole arrangement was even more perfect than 
could have been expected, up to the time of this first register. 
It was a principal object with me, as soon as possible to obtain a good and 
uninterrupted series of observations during a long period, taken at least once 
a day at some fixed hour; and for this purpose I engaged a careful workman 
of Mr. Newman’s, who had been instructed in the reading of the different 
instruments, to keep a register of their indications at 7 a.m. in the summer 
months, and 7^ a.m. in the winter. By a careful comparison of his readings 
with those of others, he was found to be fully competent to the task. The 
following register contains these observations for one year and a half, com¬ 
mencing in October 1830, and ending in March 1832. They have been cor¬ 
rected in the same way as the last, and the same kind of comparison insti¬ 
tuted. The depression of the water-barometer has been worked out daily for 
the first two and the last months; but for the intermediate months I have 
satisfied myself with making the calculation for the monthly mean results. 
