324 
MR. A. W. RUCKER AND DR. T. E. THORPE ON A MAGNETIC 
These may be checked in future by means of the observations at Kew, Greenwich, 
Falmouth, Stonyhurst, and Valentia, but we do not think a comparison of two obser¬ 
vatories only leads to trustworthy laws of rate of change of secular variation with 
geographical position. There can be no question that Declination change on the whole 
increases with the latitude, yet the secular coefficient is less at present at Stonyhurst 
than at Greenwich and Kew. It is therefore very much to be regi'etted that the obser¬ 
vations so sedulously carried on by the late Provost Lloyd, at Trinity College, 
Dublin, have of late years been interrupted. 
We may point out that the best position for a magnetic observatory would be in 
the centre of a widespread region of low Vertical Force. The disturbing rocks would 
23robably jDroduce less effect in such a situation, and all the phenomena might be 
expected to be more normal. Unfortunately Kew and Greenwich are within the 
range of tlie Reading disturbance, and Stonyhurst is in a region where the Vertical 
Force changes very rapidly both to the east and west of that jfface. 
2. Inclination. 
We have also in the case of the Dqt found the secular change between 1842‘5 and 
1886 by comparing Table X, with Sir E. Sabine’s map for the earlier epoch. The result 
confirms the conclusions jireviously arrived at, viz., that the secular change diminishes 
with latitude and increases with west longitude. As, however, the north-westerly 
stations are on the line of minimum change they probably do not form real exceptions 
to the general rule. Partly by this method and jiartly by the ina}) of assumed values 
(j). 85) we have selected those given in Table XII. 
3. Horizontal Force. 
This element when treated in the same way gives since 1842'5 a greater rate of 
secular change in the south than in the north, and in the west than in the east. The 
latter conclusion is in accord with that of M. Moureaux. 
We are now inclined to think that the high value of the secular change between 
the years 1883-85 at Greenwich ('0028) led us to assume rather too high a value 
('0022) for the south of England. 
The mean annual change for Greenwich between the years 1883-87 is '0020 and for 
Kew '0017. The values for Greenwich in 1857 and 1887 are 1'769 and 1’818, but in 
1861 a new instrument was introduced, and to make the two values comparable we 
must subtract 0'0i6 from the first. This leads to a difference of 0'U65 or a secular 
change of 0'0022. 
Though this is a little greater tlian the present rate, we do not think the evidence 
is sufficient to justify us in altering the rates obtained for Scotland by a comparison 
for the interval 1857-87, and we assume that they are valid at the jn’esent time. 
For the neighbourhood of London we take O'OOlO, and by the aid of Sabine’s and 
Welsh’s papers we have arrived at the values given in the following Table :— 
