1885.J Magnetisation of Steel, Cast Iron, and Soft Iron. 375 



copper wire wound on a copper case of square section just fitting the 

 bars. 



The magnetising current passing through the coil was obtained from 

 a battery of Thomson's tray Daniells, by means of which any desired 

 current above 0'25 of an ampere could be employed. To obtain cur- 

 rents ranging from to the minimum (0*2 5) to be got by the battery, a 

 resistance-box, capable of inserting any resistance up to 10,000 ohms, 

 was placed in the circuit with a single cell. 



The strength of the current was measured by one of Sir William 

 Thomson's graded galvanometers, represented at G, a full description 

 of which will be found in Mr. Andrew Gray's " Absolute Measure- 

 ments in Electricity and Magnetism." 



A reflecting magnetometer, M, of the well-known form devised by 

 Mr. J. T. Bottomley, was used to measure the intensity of magnetisa- 

 tion. . In the experiments on the wires it was placed due magnetic 

 east of the coil, at a distance of 1 metre from the middle point of its 

 axis, and in such a position that if the axis were produced it would 

 pass through the centre of the mirror. In the experiments on the 

 bars its position was due magnetic north of the coil. 



A framework, holding a lamp L, and having a scale S of half- 

 millimetre divisions attached to it, was placed in front of the mag- 

 netometer at such a distance that the scale was exactly 1 metre from 

 the magnetometer needle. The light from the lamp, passing through 

 a tube T in front, in which, a fine wire is vertically fixed, is reflected 

 from the mirror to the scale, and the deflection read by the image of 

 the fine wire in the middle of the spot of light. 



The results in these experiments were got by beginning with a 

 feeble magnetising current, which was increased step by step until the 

 maximum obtainable was reached. It w r as then gradually diminished 

 to zero, when the direction of the current was reversed, and the same 

 process of increasing and diminishing repeated. 



The magnetometer readings taken while the current is flowing repre- 

 sent the effect upon the magnetometer needle of the joint electromag- 

 netic action of the current passing througli the coil, and the magneti- 

 sation it produces in the wire or bar. Hence the effect due to the 

 magnetisation of the wire or bar alone is obtained by subtracting from 

 the total effect the magnetic effect of the coil. This last is propor- 

 tional to the current flowing, and was found by experiment to be for 

 the ' : wire" coil 0*0385, and for the " bar " coil 0*0326 of a division 

 of the magnetometer scale per division of the galvanometer scale. 



The results of the investigation are shown in the accompanying- 

 curves. The abscissae are divisions of the galvanometer scale, and are 

 therefore proportional to the magnetising forces. The ordinates are 

 divisions of the magnetometer scale, and are therefore proportional 

 to the magnetisation produced. The curves marked A represent the 



