38 
and that in the sun during the six years 1859-1864, and 
the monthly and annual means of the entire series : — 
1859 
1860 
1861 
1862 
1863 
1864 
Monthly 
Means 
o 
o 
O 
o 
o 
o 
D 
J armary. . . 
7-3 
5-7 
5-7 
4-2 
4-9 
5-3 
5-51 
February.. 
10-9 
11-5 
90 
6-8 
9-2 
7.6 
916 
March . . . 
11-7 
11-2 
10-4 
7-6 
10-2 
9-4 
10-08 
April 
13-2 
16-1 
154 
11-6 
11-8 
11-3 
13-23 
May 
150 
13-5 
13-5 
11-6 
11-5 
12-3 
12-90 
June 
16-0 
12-3 
14-4 
10-1 
13"2 
11-6 
12-93 
July 
17-8 
12-8 
12-4 
13-4 
14-2 
12-3 
13-81 
August ... 
16-3 
12-5 
151 
13-3 
12-5 
14-5 
14-03 
September 
16-3 
11-8 
14-1 
12-5 
12-8 
14-8 
13-71 
October . . . 
111 
8-7 
11-0 
8-8 
8-4 
100 
9-66 
November 
12-4 
8-4 
7-9 
8-2 
5-4 
6-5 
8-00 
December. 
62 
4-9 
6-0 
4‘4 
3-3 
2-2 
4-50 
Annual 1 
Means ) 
12-85 
10-78 
11-24 
9-37 
.9-78 
9-81 
j f mam j j a $ owmm 
A projection of the mean monthly values gives the curve 
A in diagram A, from which it will he 
seen that the maximum amount of 
radiation occurs in August, or about 
a month later than the time of 
maximum temperature; whilst the 
minimum occurs in December, about 
a month earlier than the time of 
minimum temperature. Th ere is also 
a slight secondary maximum in April. This curve, therefore, 
differs from that of any other element of temperature; but 
observing that the time of minimum corresponded exactly with 
that of maximum in the curve laid down from the numbers 
given in a table which 1 communicated to the Physical section 
on the 5th of March, 1863, showing the monthly sums of the 
oscillations of mean daily temperature at Greenwich during 
the thirteen years 1848-60, and also the mean daily values 
for the different months; and that the time of maximum 
agreed nearly with that of minimum disturbance of mean 
daily temperature, it occurred to me that the two phenomena 
might be closely connected with and dependent upon each 
