326 PROFESSOR J. A. EWING AND MR. G. C. COWAN 



straight piece of the wire, from 300 to 400 diameters long, was hung in a vertical 

 position with its upper end due east (magnetically) of a small mirror magnetometer. 

 Over the wire a tube was slipped on which the magnetising solenoid was wound, and 

 on the same tube there was another solenoid in which a constant current was kept 

 up, of just sufficient strength to neutralise the vertical component of the earth's 

 magnetic force. The field within the tube was therefore wholly due to the current in 

 the magnetising solenoid, and was exactly reversed when that current was reversed. 

 The strength of the magnetising current was adjusted by a slide resistance consisting 

 of a column of solution of sulphate of zinc, with two fixed blocks and one sliding block 

 of amalgamated zinc for terminals ; and a revolving commutator between the slide 

 and the solenoid allowed the current to be rapidly reversed. The slide was used to 

 vary the current slowly in studying the relation of magnetism to magnetising force, 

 and, in conjunction with the rapid commutator, it served also to demagnetise the 

 wire under examination by the method of reversals (that is, by numerous successive 

 reversals of a magnetising force which decreased slowly from a strong value to zero). 

 The magnetising current was taken from a secondary battery of ample capacity, and 

 was measured by a mirror galvanometer, calibrated to read the current in absolute 

 measure, which was included in the circuit of the magnetising solenoid. The current 

 which served to neutralise the vertical component of the earth's field was adjusted by 

 substituting for the nickel wire a wire of soft iron, and subjecting that to demagneti- 

 sation by reversals ; it was only when the earth's field was very exactly balanced that 

 this process gave complete demagnetisation. In all the results given below the 

 magnetising force $ and the intensity of magnetism 3 are expressed in absolute 

 c.g.s. units. 



Cyclic Magnetisation of Nickel. 



Fig. 1, Plate 15, exhibits the cyclic magnetisation of a piece of nickel wire, 

 0'068 cm. in diameter and 25'4 cm., or 374 diameters long, in the hard-drawn state, 

 as supplied by Messrs. JOHNSON and MATTHEY. Previous to this test the wire had 

 been strongly magnetised and demagnetised by reversals, and at the beginning there 

 was a little residue which this process had failed to wipe out. Nickel, like hard steel, 

 is much less easily demagnetised to a perfectly neutral state than soft iron. The 

 following are the observed values of f% and 3 during the first part of the experiment, 

 and the values of K, the susceptibility, which is 



