Rocks aruJ Regional Magnetic Disturbances. 517 



A specimen of Canna basalt was placed in the left-hand test-tube, 

 and liquid abstracted until the volume was the same as before. 



r 9-t 



...i 94 



I 9-* 



9-8 



leadings . . . <( 9'9 Mean = 9'83 = 



8 



therefore x l = (^r^ = O28. t 



Equal volumes of the liquid of strength 1/5 were now taken. 



rlO-15 



9'9 

 Zero readings < ' Mean = 10'06 = z 2 . 



LlO-2 



The same specimen of rock was now placed in the left-hand test- 

 tube, after having been carefully washed and dried. 



Headings . . J Mean = 10'08 = r 2 . 



rlO'O 

 J ^ 

 LlO'l 



x 2 = (z 2 -r 2 ) = 0-02. 

 Hence, by the formula given above, 



c = 0-001227. 



On the whole then, we think that the various tests which have 

 been applied to it prove that the method employed fulfils the 

 required conditions very satisfactorily. It is not capable of giving 

 results of the last degree of accuracy, but it enables us to measure 

 quickly and certainly, with only a small percentage error, the per- 

 meabilitiiies of rock specimens without the labour and expense 

 involved in shaping them into definite geometrical forms. 



The method, too, has the advantage that, when once the permeabili- 

 ties of the standard liquids are determined, the apparatus can be 

 used anywhere. If therefore it were desirable to institute a close 

 comparison between the magnetic disturbances and the magnetic 

 permeabilities of the rocks in a given district, and it were important 

 that the investigator should become at once acquainted with his 

 results, it would be quite practicable to transport the apparatus 

 required to the scene of the investigation, and to determine the mag- 

 netic properties of the specimens in any convenient room within a 

 few hours of their collection. 



Our observations on rook specimens may be divided into three 



