\ 



418 Professor A. W. Biicker [April 17, 



are, as I think, weighty reasons for believing that rock magnetism is 

 often the principal cause, I propose to discuss them. 



Permanent magnetisation of the rocks may perhaps be discarded 

 as the cause of disturbances which extend over large areas. Basaltic 

 columns are often strongly magnetised, but then magnetisation is 

 irregular. At a short distance the opposite poles would neutralise 

 each other's effects, and widespreading effects are most likely due to 

 the inductive influence of the earth acting on widespread masses. 



The evidence for this may be summed up as follows : — In some 

 cases, as in that of the cliffs on the Hudson Eiver and at Snake Hill 

 (New Jersey), the mass is apparently polarised at the upper and 

 lower extremities, as it would be if magnetised by the earth. Clear 

 indications of this fact are, however, often difficult to obtain, as they 

 are masked by local permanent magnetism. 



A more certain test can be applied in the case of less strongly 

 magnetic rocks if the instruments can be placed in their neighbour- 

 hood, but on non-magnetic soil. These conditions are satisfied 

 by the Malvern Hills, and they are found to attract the north 

 pole of a magnet, which is consistent with the view that they are 

 magnetised inductively. Again, over large districts, the centres 

 of which are marked by the outcrop of basaltic rocks, the magnetic 

 forces tend towards the centre, which is again what would occur if 

 the rocks which appear on the surface are the uppermost portions of 

 a much larger mass magnetised by the inductive influence of the 

 earth. This state of things is observed in the south and west of 

 Scotland and in Antrim. 



Lastly, Captain Creak, F.E.S., has shown that islands in the 

 northern and southern hemispheres attract the north and south poles 

 of the magnet respectively. 



The only question that remains is whether the presence of rocks 

 similar to those which exist on the surface would suffice to account 

 for the observed surface disturbances. To test this the magnetic 

 permeabilities of a number of specimens of basalt kindly supplied by 

 Professor Judd have been determined. Assuming (1) that magnetite 

 becomes non-magnetic at the same temperature as iron, (2) that its 

 magnetic properties are not affected by great pressures, (3) that the 

 temperature at which iron ceases to be magnetic is reached at a depth 

 of 12 miles, (4) that large sheets of magnetic rock exist between the 

 surface and this depth, the areas of which are of the same dimensions 

 as those of the regions of high vertical force which exist in the 

 United Kingdom, (5) that the magnetic susceptibilities of these rocks 

 vary between the mean of those of 13 sj)ecimens from the Island 

 of Mull (0' 00163), and of 34 sj)ecimens from the west of Scotland 

 (0* 00271), it is found that a fair agreement exists between the results 

 of calculation and of observation, and that there is no doubt that the 

 calculated and observed disturbances are of the same order of 

 magnitude. 



All these facts then accord well with the theory that local and 



