PEIRCE. MAGNETIZATION IN IRON. 139 



each of the other pieces about 0.10%. The Drill Rod has about 

 1.10% of carbon; the specimen, which is of very fine grain, has no 

 slag and shows simply sorbite, pearlite, and some cementite in fine 

 net work. 



Nos. 1, 2, 20, 21, 22, 24, and 25 contain considerable slag, and this 

 is in comparatively large masses in portions of the "Refined Iron. 

 The relatively small amount of slag in the Norway Irons is in fine 

 particles distributed through the mass. Nos. 11, and 12 seem to be 

 simply f errite. 



Table VI gives some corresponding values of H and I for the speci- 

 men of American Ingot Iron, known as No. 12. 



It is well known that if a relatively stout rod of soft ii*on be exposed 

 to a strong field in a solenoid and if the magnetizing current be very 

 suddenly broken, the direction of the residual magnetism in the rod 

 may be opposite in sign to what it was when the current was running. 

 If the rod be enclosed in a thick walled copper tube within the solenoid, 

 this reversal never takes place and the sign of the residual magnetism 

 is always normal. The moment of the rod under the new field is 

 often larger when the current is suddenly reversed than when it is 

 slowly reduced to zero through a constantly growing resistance before 

 the switch is thrown over and then gradually brought to its new 

 strength, or when the change is made less violent by eddy currents 

 induced in a thick copper shell around the specimen. Though it 

 seemed possible that the results given in this paper might be slightly 



