238 
DR. C. CHREE: DISCUSSION OF KEW MAGNETIC DATA 
There is a regularity in both sets of figures for H which can hardly be accidental. 
They agree in making the successive monthly values all positive from May to August, 
and all negative from September to February. Both periods show a clear maximum 
in the summer months, but one puts the minimum in winter, the other in equinox. 
The inequality obtained by combining the two periods, allowing them equal weights, 
makes the winter and equinoctial values practically equal. It shows a range of 12'5y, 
the maximum coming in June, and the minimum in November. 
Table XVIII.—Annual Inequality. 
Month 
Horizontal force. 
Inclination. 
Vertical force. 
and 
season. 
First 
period. 
Second 
period. 
Mean. 
First 
period. 
Second 
period. 
Mean. 
First 
period. 
Second 
period. 
Mean. 
7 
7 
7 
' 
r 
' 
7 
7 
7 
January .'. . 
-2-0 
- 1-8 
-1-9 
+ 0-19 
o-oo 
+ 0-09 
+ 1-9 
- 4-4 
-D2 
February . . . 
-o-i 
-1-6 
-0-9 
+ 0-03 
+ 0-06 
+ 0-04 
+ 0-8 
- 1-8 
-0-5 
March. 
-4-5 
■+■ 3 * 5 
-0-5 
+ 0-41 
-0-07 
-4-0-17 
+ 4-1 
+ 6-0 
+ 5-0 
April. 
-2-3 
+ 0-5 
-0-9 
+ 0-08 
+ 0-09 
+ 0-08 
-2-6 
+ 4-4 
+ 0-9 
May. ..... 
*3-1 
+ 5-6 
+ 4-3 
-0-08 
-0-17 
-0-12 
+ 4-7 
+ 7-3 
+ 6-0 
June. 
+ 8-8 
+ 7-0 
+ 7-9 
-0-56 
-0-16 
-0-36 
+ 1-0 
+ 10-9 
+ 5-9 
July ..... 
+ 2-6 
+ 2-3 
+ 2-5 
-0-26 
-0-18 
-0-22 
-3-0 
- 0-8 
-1-9 
August .... 
+ 4-4 
+ 0-5 
+ 2-5 
-0-23 
-0-07 
-0-15 
+ 2-2 
- 1-3 
+ 0-4 
September . . 
-4-2 
-2-7 
-3-4 
+ 0-12 
-0-06 
+ 0-03 
-5-6 
- 8-5 
-7-1 
October .... 
-3-4 
-3-7 
-3-6 
+ 0-25 
+ 0-14 
+ 0-19 
+ 0-8 
- 4-0 
- 1-6 
November. . . 
-1-9 
- 7-3 
-4-6 
+ 0-21 
+ 0-36 
+ 0-29 
+ 3-0 
- 4-4 
-0-7 
December. . . 
-0-9 
-1-7 
- 1-3 
-0-16 
+ 0-06 
-0-05 
-8-0 
_ 9*9 
jLJ tJ 
-5-1 
Winter .... 
- 1-2 
-3-1 
_ 2 • 2 
+ 0-07 
+ 0-12 
+ 0-09 
-0-6 
- 3-2 
-1-9 
Equinox. . . . 
-3-6 
-0-6 
-2-1 
+ 0-22 
+ 0-03 
+ 0-12 
-0-8 
- 0-5 
-0-7 
Summer. . . . 
+ 4*7 
+ 3-8 
+ 4-3 
-0-28 
-0-14 
-0-21 
+ 1-2 
+ 4-0 
+ 2-6 
In the case of I the two periods agree in showing a distinct minimum in summer, 
but the one places the maximum in winter, the other in equinox. Combining the 
two periods, allowing them equal weight, we get a comparatively smooth inequality 
with a range of 0 /- 65,the maximum coming in November, the minimum in June. The 
type is on the whole fairly similar to that of results that have been published for 
Parc St. Maur and Potsdam, and for the northern hemisphere as a whole by Liznar 
and Hanr, but the range shown in Table XVIII. is a good deal less than in the cases 
quoted. 
Assuming that the phenomena are not of instrumental origin, it seems reasonably 
certain, after allowing for a secular change progressing at a uniform rate throughout 
the year, that TI is higher and I lower in summer than in winter. Also it seems 
probable, though more open to doubt, that V displays the same phenomena as H. This 
agrees with the results obtained from quiet days alone in a previous paper. 
