FROM ORDINARY DAYS OF THE ELEVEN YEARS 1890 TO 1900. 
207 
S, but in the group or groups with S above 80 the rate of increase seems reduced. 
The exceptionally high monthly values of S occurred sporadically. For instance, the 
extreme value, 129'2 in August 1893, was preceded by 88'8 in July and followed by 
77'9 in September. Thus we should not expect a correspondingly high value of It 
uidess the influence of previous months is negligible. 
In considering the significance of Table XLV. allowance should be made for the fact 
* that some of the groups come entirely from years of sunspot maximum. No value of 
S as high as 60 occurred in the sunspot minimum or intermediate years. Again, 
sunspot minimum years gave no value of S larger than 20‘5, and so their contributions 
were practically confined to the groups 0 to 10 and 10 to 20. Intermediate years 
contributed only two months to the group 0 to 10, and six to the group 10 to 20. 
§ 28. We have hitherto considered only mean monthly values of It', but the way in 
which these values are made up seems also of interest. To make the results more 
complete, D has been included in the investigation. Some data for D were got out 
in a previous paper, # but they are considerably extended here. As a preliminary it 
is well to know something about the extreme values of R/. To this end Tables XLYI. 
and XLVII. give the smallest recorded values of R' in D and H for each month. 
Table XLVI.—Declination. Absolute Daily Ranges. Smallest Values. 
Year. 
January. 
February. 
March. 
April. 
May. 
June. 
July. 
C/3 
3 
b£> 
<D 
o 
2 
<D 
■+3 
cs 
CD 
m 
October. 
November. 
December. 
Year. 
1890 
4-2 
r 
5-6 
5-1 
/ 
8-2 
/ 
6-2 
/ 
6-9 
/ 
8-0 
/ 
8-2 
5-0 
5 • 7 
/ 
5-0 
/ 
3-8 
/ 
3-8 
1891 
31 
4-1 
4-3 
8-1 
8'5 
7-9 
8-6 
9-2 
9-8 
7-9 
7-1 
4-7 
3-1 
1892 
5-7 
8-3 
11-4 
10-9 
10-3 
9-2 
11-4 
10-1 
9-2 
8-5 
3-7 
5-7 
3-7 
1893 
5-0 
5-0 
8-0 
12-5 
11-9 
12-3 
11-6 
11-5 
10-1 
9-3 
5-9 
4-1 
4-1 
1894 
6-8 
7-9 
8-3 
9-6 
10-6 
10-5 
6-8 
9-7 
10-3 
9-0 
4-7 
4-6 
4-6 
1895 
3-9 
6-8 
8-3 
11-9 
10-0 
13-0 
10-8 
8-0 
8-5 
8-6 
5-2 
3-8 
3-8 
1896 
6-0 
7-3 
8-0 
10-9 
9-0 
6-7 
7-9 
9-8 
8-2 
6-6 
3-4 
2-7 
2-7 
1897 
3-2 
4-3 
8-9 
8-3 
7-4 
7-8 
8-0 
9-1 
6-1 
5-8 
4-3 
9 • 9 
—l Ju 
2-2 
1898 
3-1 
3-5 
6-0 
8-0 
7-1 
7-7 
7-2 
8-0 
7-5 
6-5 
4-3 
3-3 
3-1 
1899 
3-0 
3-8 
6-3 
7-9 
7-1 
8-4 
6-4 
8-2 
7-7 
4'2 
3-0 
2-4 
2-4 
1900 
3-2 
3-1 
5-3 
7-2 
5-7 
7-8 
8-0 
6-5 
5-6 
5-1 
2 * 2 
2-5 
2-2 
Whole 1 
period. J 
3-0 
3-1 
4-3 
7-2 
5-7 
6-7 
.6-4 
6-5 
5-0 
4’2 
9 • 9 
-J £J 
2 • 2 
2 * 2 
Corresponding data are not given for V owing to their greater uncertainty. V 
ranges are on the average less than D or H ranges, and when very small they are 
VOL. CCXVI.-A. 
‘Phil. Trans.,’ A, vol. 208, p. 205. 
2 O 
