200 
DE. H. T. BARNES ON THE CAPACITY FOR HEAT OF WATER 
The following Table contains the results of the comparisons for the chronometer 
obtained by telephonic communication :— 
Table XY. 
Date. 
Gain. 
Date. 
Gain. 
Jan. 3 to Jan. 4, 1899 . . 
>' 4: ,, ,, 5, ,, 
n ^ 5) ?) JJ * 
,, 6 ,, „ 7, ,, . . 
„ 7 ,, „ 9, ,, 
„ 9 „ ., 10, „ . . 
„ 10 „ „ 11, . . 
6 • 5 secs. 
6- 5 „ 
7- 0 „ 
6- 5 „ 
12-5 „ 
7- 0 „ 
7-0 „ 
Jan. 11 to Jan. 14, 1899 . . 
„ 14 „ ,, 16, ,, . . 
„ 16 „ „ 23, „ . . 
)> 1) JO, ,, 
„ 30 „ Feb. 6, ,, . . . 
Feb. 6 „ „ 13, „ 
„ 13 ., „ 20, „ . . 
19-5 secs. 
12-0 „ 
48 - 5 ,, 
49- 0 .. 
46-0 „ 
48-5 ., 
48’5 „ 
Mean gain .... 7‘0 secs, in 24 hours. 
From March 20 to March 21 of the same year (1899) the gain was 7 '0 secs. ; from 
February 9 to February 10, 1900, the gain was 7'0 secs. ; from February 10 to 
February 12 it was 15'0 secs. 
Ill August, 1898, the chronometer was taken to the Observatory for two weeks, and 
daily comparisons were made with the standard clock. Its rate was found to vary 
between 6 and 9 seconds gain per day, which is a somewhat greater irregularity than 
I obtained, although the mean value agrees very well with the later comparisons. 
In rating the clock, 1 considered it safe to assume the rate of the chronometer 
constant to within a second from day to day, and equal to a gain of 7 secs. 
In Table X VI., I give the comparison of the clock with the chronometer from 
May, 1898, to the close of the present work. The seconds represent the loss of the 
clock per day, and are not corrected for the rate of the chronometer. 
In Table XVII., a summary of the previous Table is given, showing the greatest 
and least loss per month, with the mean loss corrected for the chronometer, which is 
obtained by subtracting 7 seconds. As far as possible, the rate of the clock was 
obtained over any day in which an experiment was obtained. 
