68 
MR. GEORGE W. WALKER ON THE MAGNETIC RE-SURVEY OF THE 
that comparatively simple ridge lines could be traced in this way for considerable 
distances. The results of the re-survey led me to rather different inferences. At all 
events it appears to me that the combined information supplied by a knowledge of 
the horizontal and vertical disturbances is required to determine the origin of the 
disturbances, and that the independence of the horizontal and vertical force tests is 
somewhat illusory. 
It is impossible to enter on a detailed discussion of every part of the survey. But 
we may select a few illustrative cases in order to bring out the points of resemblance 
and difference in the data and the inferences. 
Consider the region from Portsmouth to Beading. The horizontal forces from the 
two surveys are in good agreement and suggest a ridge of convergence running along 
the meridian 1° W. The vertical forces differ considerably in the two surveys. 
Starting in the south with Byde, Chichester and Worthing, the old survey gives 
— 18, -M30, -1-24, so that a pronounced downward maximum is indicated near 
Chichester. The new survey gives —106, —34, —48. Thus the maximum at 
Chichester is not nearly so sharply defined, and its value is 160y less than Bucker 
and Thorpe’s. 
Next take Salisbury, Alresford, Haslemere and Horsham. The old survey gives 
-1-43, 6, -1-87, 4-20. This gives a fairly pronounced maximum at Haslemere, and a 
secondary one at Salisbury. The new survey gives —98, -h37, +12, —88, so that 
the maximum is indicated near Alresford. 
Again, take Swindon, Wallingford, Beading and Windsor. The old survey gives 
+ 38, +145, +235, +136, so that the maximum is near Beading. The new survey 
gives +59, +47, +168, —34, so that Beading is again indicated as the maximum. 
It is to be remarked that the old values are all positive, and in the District Table 
Bucker and Thorpe’s values are about lOOy higher than mine. The addition of lOOy 
to the above values for the new survey would give a better general agreement with 
the values for the old survey in this particular district, although some discrepancies 
would remain. Thus the line of maximum values in the old survey, viz., Chichester 
( + 130), Haslemere ( + 87), Beading ( + 235), and that of the new survey, Chichester 
( — 34), Alresford ( + 37), Beading ( + 168), indicate a difference which is not to be 
explained by uniform shift of the datum plane. The new survey thus gives no 
suggestion of a peak at Chichester, but the maxima continually increase towards 
Beading. The question what is the correct datum plane from which to reckon 
disturbance is, as already explained, a very difficult one, and one that does not at 
present admit of exact solution. But having started with the principle that we are 
to make the residuals as small as possible, we are not at liberty to alter the datum 
plane in any particular region unless we have reached an impasse. 
No necessity arises in the present case for such a change in the datum plane. The 
general character of the data in this region appears to admit of explanation by the 
supposition of an underground disc of material magnetised vertically. The boundary 
