276 Mr Harker, Magnetic Disturbances in the Isle of SJcye. 



for some miles. Observations were taken along the northerly 

 ridge of the mountain, avoiding magnetic crags which are not 

 numerous or important. Beginning at the summit, the declination 

 was found to be quite normal for the first half mile, that is, down 

 to about 1650 feet altitude. Here a negative deflection begins, as 

 we approach the boundary of the gabbro, and this negative deflec- 

 tion continues, with a steady magnitude of 5° or 6°, for a mile and 

 a half at least along the lower (granite) portion of the ridge. 

 Unfortunately the observations were not carried beyond this point, 

 but they are sufficient to prove a somewhat wide-spread magnetic 

 disturbance, changing only slowly in magnitude, and extending 

 well beyond the limits of any possible magnetised rocks. 



In the foregoing paragraphs I have, so far as is possible, 

 avoided theoretical considerations, feeling that the data are not 

 sufficient to warrant any strong opinion as to the precise explana- 

 tion of the magnetic disturbances noted, but a few concluding 

 remarks under this head will not be out of place. It may be 

 taken as highly probable, if not certain, that, as regards Skye, 

 these disturbances, small as well as large, are attributable to rock- 

 magnetism. The alternative advocated by Naumann, viz. the 

 effects produced in earth-currents by dislocations of the strata, 

 can scarcely be appealed to in this case, especially in view of the 

 fact that the district affords extremely little evidence of the 

 existence of such dislocations. There remain the rival claims of 

 permanent magnetisation of the rocks and the inductive action of 

 the earth's magnetic field. As regards what I have distinguished 

 as disturbances of the first order, the former cause is clearly indi- 

 cated by the facts recorded ; but whether it supplies a vera causa 

 for the disturbances of lower orders is a more doubtful question. 



Messrs Rticker and Thorpe, in the valuable report already 

 cited, lay stress upon induction ; but it must be remembered that 

 they deal generally with disturbances much smaller and more 

 wide-spread than those described above. They distinguish three 

 different orders, as follows: — 



[iii] (local) affecting only a single station or its immediate 

 neighbourhood ; i.e. having a range less than the average distance 

 between the stations ; 



[iv] (regional) having a greater range, but small compared 

 with the entire area under survey; 



[v] involving the whole, or a considerable fraction of that 

 area. 



I have renumbered these to correspond with the orders of 

 magnitude recognized above, on the assumption that the most 

 local disturbances embraced by this general survey are similar to 



