SURVEY OF THE BRITISH ISLES FOR THE EPOCH JANUARY 1, 1886. 
G7 
of the Kew Committee for 1887-8 was not published when the observations made in 
1888 were reduced, we used the same Table as for the 1887 observations. 
The question as to whether we should apply similar corrections to the other 
elements has been carefully considered, and we have on the whole decided to do so. 
Though we have worked up the observations on the Horizontal Force so that each 
experiment gives two nearly independent results, the times at which these have been 
taken are practically identical, and they do not serve as a test of the advantage of 
applying corrections. In 18 cases we have repeated either the deflection or the 
vibration experiment (generally the latter) in the fleld on the same day. The 
differences between the results when uncorrected and corrected for diurnal variations 
and disturbance are as follows. They are expressed in terms of O’0001 metric or 
O’OOOOl C.G.S. unit. 
Table I. 
Station. 
Uncorrected. 
Cori-ected. 
Loch Aylort. 
- 5 
0 
Ayr. 
- 4 
- 4 
Stornoway. 
- 3 
- 5 
+ 3 
+ 2 
Cambridge. 
+ I 
+ 1 
Northampto 2 i 
+ 18 
- 7 
Stoke . 
- 5 
- 6 
Swansea . . . . 
+ 3 
- 1 
Taunton. 
+ 17 
+ 16 
Thirsk. 
0 
+ 1 
Tunbridge Wells 
+ 17 
- 3 
Worthing. 
+ 8 
- 2 
Ballywilliam. 
0 
- 1 
Bangor. 
+ 4 
- 6 
Coleraine. 
- 7 
-11 
Drogheda. 
+ 9 
+ 3 
Kildare. 
-10 
+ 5 
Londonderry .... 
+ 6 
+ 5 
Probable difference -( ^ncoiiected 
[ Corrected . 
. 5-9 
. 4-0 
The values of H obtained by us at Kew in 1887 also afford a test of the method of 
correction. The diurnal variation and disturbance are expressed in the following 
Table in terms of O’OOOl metric unit 
