BRITISH ISLES FOR THE EPOCH JANUARY 1, 1915. 
9 
Inclination. 
1913. 
1914. 
1915. 
November 10. 
November 18. 
1 
November 24. 
November 25. 
November 26. 
May 25. 
September 25. 
Cl 
^ ! 
a 
Cambridge survey value . 
67 22-2 
21'-7 
22'-3 
2l'-5 
22-4 
23'-4 
24'-7 
26'-7 
Greenwich value. 
66 52-4 
50-6 
49-4 
50-5 
49-0 
51-4 
53-1 
52-4 
Difference. 
+ 29-8 
31-1 
32-9 
31-0 
33-4 
32-0 
31-6 
34-3 
Mean . 
. +32-0 
Divergence . 
- 2 
-o'-9 
+ 0'-9 
-T-0 
+ T-4 
o-'o 
-o'-4 
+ 2'- 3 
i 
I think the conclusion is that within the limits of experimental error, the magnetic 
elements at Cambridge vary 'pari passu with those at Greenwich. 
The magnetic values obtained at any station refer to a particular day and a 
particular hour. Thus since the magnetic values are continually varying the observed 
values are subject to a correction which will reduce them all to a common standard, 
viz., the normal value characteristic of the stations at the same time—the epoch of 
the survey. The correction may be considered as made up of (l) correction to mean 
for the day, (2) correction for secular change to epoch. 
Rucker and Thorpe regarded the correction to mean for the day as consisting 
of two parts (l) that due to the normal diurnal variation ; (2) that due to dis¬ 
turbance. They assumed that the values of these obtained from the magnetograms 
at Kew were applicable to the whole of the British Isles, that the normal diurnal 
variation depended on local time, but that the disturbance depended on Greenwich 
time. 
None of these assumptions are correct, and it is generally agreed that the only 
satisfactory way is to have a triangle of recording observatories within which the 
survey is to be carried out, so that the interpolation of variations may be made. 
Although we could not arrange for this, a scheme was drawn up for taking a set of 
magnetographs for a few weeks to the North of Scotland and the West of Ireland, 
so that by comparison with Greenwich, factors applicable to the various stations 
might be determined. This plan had, however, to be given up on account of the 
war and I had to depend on the data from Greenwich only. 
Agam the rate of secular change is not the same all over the British Isles. On the 
other hand, preliminary comparison of my results with those of Rucker and Thorpe 
showed that the average rate of change was fairly uniform all over, so that within 
VOL. CCXIX.—A. 
c 
