1885.] Magnetisation of Steel, Cast Iron, and Soft Iron. 377 



needle may be regarded as at its centre c. Hence the total attractive 

 force exerted on the pole s is — 



-n,»f 1 11 w 4iar , , , . - 2Mr 



We find similarly that the pole exerts an eqnal repulsive force 

 on the pole n. The needle is therefore acted on by a " couple " — 



¥ -— cos e, that is^^p 



To balance this we have another "couple," /H.2&sin0, that is, 

 mH sin 0. Hence, equating these two couples, we get — 



^_Htan0 (1.) 



2r 



The value of H at this point was found to be 0'16, by comparison 

 with a particular spot in the laboratory, for which, by the method 

 fully described in Mr. Thomas Gray's paper on " The Experimental 

 Determination of Magnetic Moments in Absolute Measure " (" Philo- 

 sophical Magazine," November, 1878), the value of H had already 

 been very accurately determined. Half the distance between the 

 poles of the wire, represented by <x, may be taken as half its length, 

 that is, 15'5 cm. The angle through which the magnetometer 

 needle is deflected is measured on the half -millimetre scale ; and for 

 small angles tan 6 = -J tan 20 ; so tan is got by dividing the scale 

 reading by 2r in half -millimetres, that is by 4000. By substituting 

 these values in equation (1), we find that the magnetic moments are 

 obtained by multiplying the readings of the magnetometer scale, 

 which in the curves are represented by the ordinates, by the factor 

 19110125. The magnetic moments per gram are therefore got by 

 multiplying the ordinates in — 



Curves I and V by '4795. 



II 



„ VI 



„ 0-479. 



III 



„ VII 



„ -4605. 



IV 



„ VIII 



„ 0-4964. 



The position of the bar with regard to the magnetometer in the 

 experiments on the bars, is shown on fig. 3. 



The pole N of the bar attracts the pole s of the needle with a force 



and the pole S repels the pole s with an equal force. The 

 r iJ ra 2 ^ 



VOL. XXXIX. 2 c 



