PEIRCE. 



MAGNETIZATION IN IRON. 



125 



It is evident that in such measurements of magnetic flux changes 

 as are necessary in the work described in this paper, it is of fundamental 

 importance that the ballistic galvanometers used be correctly cali- 

 brated, and we used a number of standards of mutual inductance, most 

 of them rather larger than those commonly emploA'ed for such purposes, 



fill 



'/. 



Li.U 



D 



m 



Figure 2. Three standards of mutual inductance. 



since our rather slowly moving galvanometer was not very sensitive. 

 We had in all seventeen mutual inductances for our calibrations. Of 

 these five have been measured for us this year at the United States 

 Bureau of Standards, and seven others are of such forms that their 

 ^"alues may be calculated by well known methods. We found one of 

 Doctor Campbell's ^ ariable Standards of Mutual Inductance (which 

 was very kindly lent to us by Professor Kennelly), most useful. It 

 proved to be very accurately calibrated, and it agreed closely at all 

 points with the standards determined for us at Washington. 



An elaborate series of comparisons of our inductances occupied 

 Mr. Coulson and myself for more than two months, because we found 

 that three or four of those which, according to our computations based 

 upon their geometrical forms, should have certain values, seemed to 

 have slightly different values, though they did not seem to be quite 

 constant. This phenomenon puzzled us at first and gave us much 

 trouble, but we believe, after all our work, that the ebonite used as a 

 core in three of them is very slightly susceptible in a strong magnetic 

 field, that the split thick brass tube used as a core for one of our sole- 

 noids is sufficiently paramagnetic to affect the field inside it percep- 

 tibly, though another solenoid constructed in a similar manner seems 



