SURVEY OP THE BRITISH ISLES FOR THE EPOCH JANUARY I, 1886. 259 
Local and Eegional Disturbances. 
We now come to the consideration of the local and regional disturbances which exist 
in many j^arts of the United Kingdom, and which we have investigated much more 
fully than has hitherto been done. 
As the problem is one of difficulty, it is necessary that every step should be carefully 
considered. It may be attacked in three ways, which are not independent, but each 
of which is attended with special advantages and disadvantages, and which, when 
combined, afford in many cases the means of arriving at definite conclusions. 
(1.) If the true isomagnetic curves are drawn as accurately as possible, without any 
attempt to smooth the irregularities, they present in disturbed districts distorted 
forms which enable us to judge of the nature and magnitude of the disturbance. The 
great advantage of this method is, tliat it is independent of calculation. It is not 
affected by errors possibly introduced by the method of determining tlie terrestrial 
curves. The conclusions arrived at are based directly upon the observations. 
The objections that may be raised to it are, that it can only be used with effect in 
the case of considerable disturbances, and that it may tend to exaggerate the importance 
of those which, though of great local mtensity, are of small range. A curve may 
be carried many miles from its true position, in order to pass through a station, 
the disturbance at which dies out within a very short distance. 
When the stations are as numerous as ours, and when the curves are drawn with a 
due regard to the possibility of isolated maxima and minima, we do not think that 
the risk of error on this account is as great as it might at first sight appear, but in so 
far as it exists, it may be checked by the second method. 
(2.) If the disturbances of the elements, i.e., the differences between the observed 
values and those calculated from the general equations given above, are plotted down 
on a map, it is found that they are not scattered haphazard, but that certain districts 
exhibit definite peculiarities, such as that the observed value is always too large or 
too small. From a study of such maps deductions can be made as to the nature of the 
disturbing forces. 
This method is open to the objection that the peculiarities in question may not 
correspond to physical realities, but may be due to the inadequacy of the formulae. 
The calculated rate of increase of the declination with longitude, for instance, may be 
made a little too rapid in one part of the country, and a little too slow elsewhere. The 
observed declinations may, therefore, appear too large in the one distiict and too small 
m another, and thus a mere mathematical error may create a fictitious attractive or 
repulsive force. 
To this objection, it may be answered that the observed disturbances are too large 
and too irregular to be thus explained. Taking, for instance, the English and Welsh 
stations which lie between long. 3 W. and long. 5 W., the mean disturbances of the 
Declinations for the groups indicated, are as follows ;— 
