28 Professor J. D. Kverett’s Description of a Method 
commenced with some other month instead of January (e.g., 
if we had written the temperatures of the months February to 
July in the 1st column, and those of August to January in the 
2d), would the results have been affected by the change ? 
I have made the trial in the case of the Edinburgh and 
Pictou temperatures above given, and have obtained the fol- 
lowing results :— 
a. é. 
For Edinburgh, commencing with Dec. 10°79 53° 29’ 
Fyctl rE | Jans. LO-78 Sa... 27" 
i # Feb. 10°78 113° 28’ 
For Pictou, . Dec. 23°07 48° 18/ 
ne 3 Jan. 23:06 78° 13’ 
. af Feb, 23:05 108° 13° 
The successive differences in the value of e ought to be pre- 
cisely 30°, since a month corresponds to 30°; and this condi- 
tion is exactly fulfilled for Pictou, while for Edinburgh the 
greatest discrepancy amounts to only 2’, or about a thirtieth 
part of a day, a difference which may be neglected in com- 
paring the phases of annual temperature. The differences of 
the values found for a are also very small, not exceeding 0-02. 
These results may therefore be pronounced sufficiently coinci- 
dent. If we had commenced with any other month, the arith- 
metical process would have been throughout the same as in 
one of the above three cases. 
II. What is the amount of error produced by assuming the 
calendar months to be all of equal length ? 
In order to test this point, I found from Principal Forbes’ 
table of Edinburgh daily temperatures the mean temperatures 
of January, February, and March— 
1st, When the last two days of January and first two days 
of March are reckoned part of February, giving February 
33 days, and leaving January and March only 29 days each. 
2d, When the last three days of February are reckoned part 
of March, so that January will have 381 days, February 26, 
and March 34. 
3d, When the last day of January and first of March are 
reckoned part of February, so that January will have 30 days, 
February 31, and March 30; the resulting values of A, a, and 
e, are as under— | 
