380 



Mr. J. W. Gemmell. 



[Dec. 10 r 



inferior to the first in respect of magnetisability. The malleable iron 

 bar exhibits a very much higher magnetisability than the cast iron 

 bars ; and its residual magnetisation was so low that it could not be 

 observed with the same arrangement of apparatus. 



These are the main points in a comparison of the results of the 

 experiments. A study of the curves, however, reveals many points 

 of interest, one or two of which I may here indicate. 



The curve beginning from the zero of magnetisation was not obtained 

 in all the experiments, the wires having been previously magnetised 

 in a preliminary test ; but in those cases in which the smallest mag- 

 netising forces were employed, the curve of results is seen to be for 

 a short distance concave towards OY. 



Returning again to the small magnetising forces after having pro- 

 ceeded to the limit, we find the curve first becoming concave towards 

 OY, and then convex just before it crosses that line. On the negative 

 side of OY, it remains concave for but a short distance, and is convex 

 when it crosses the line of zero magnetisation, remaining so both in 

 the direct and the return curves until we again near the zero of 

 magnetising force, when it becomes concave for a short distance up 

 to zero. To show these points clearly, the central portion on an 

 enlarged scale has been affixed to each set of curves. 



Turning to the curves for the residual magnetisation, an interesting 

 point at once presents itself. This is a loop between the direct and 

 return curves, more or less marked in most of the diagrams, but best 

 seen in No. IV. A similar loop is seen in the curves of total magnet- 

 isation in Nos. V and VII, and there seems to be a tendency to form 

 such a loop in all these curves. Regarding that part of the positive 

 return curve which represents the effects of the small magnetising 

 forces, we see that the residual magnetisation first begins to take a 

 greater value, and then diminishes again just before the zero of 

 magnetising force is reached. 



I shall defer any discussion of these anomalies until I have made a 

 further observation of them under conditions more suitable for their 

 special investigation. 



