18 



ON THE DIAMAGNETIC CONSTANTS OF BISMUTH AND 

 CALC-SPAK IN ABSOLUTE MEASUKE 



[American Journal of Science [3], XVIII, 360-371, 1879] 



PART I. BY H. A. ROWLAND 



Since my experiments on the magnetic constants of iron, nickel and 

 cobalt, I have sought the means of determining those of some diamag- 

 netic substances, and to that end have described a method in this 

 Journal for May, 1875 (vol. ix, page 357). As Mr. Jacques, Fellow of 

 the University, was willing to take up the experimental portion, I have 

 here worked up the subject more in detail and brought the formulae 

 into practical shape. No experiments have been made on this subject 

 so far, but some rough comparisons with iron have been made by 

 Becquerel, Plucker and Weber. But as iron varies so greatly, and as 

 the methods of experiment are inexact, we cannot be said to know 

 much about the subject. As, however, the relative results of these 

 experiments and those of Faraday can be accepted as reasonably exact 

 for diamagnetic substances and weak paramagnetic ones, it is only 

 necessary to make a determination of one substance such as bismuth, 

 and then the rest can be readily found. But as bismuth is very crys- 

 talline it is necessary to make our formulae general, unless we use bis- 

 muth in a powder, which would introduce error. 



The general method of experiment has been indicated in the paper 

 before referred to, but I may here state that it consists in counting 

 the number of vibrations made by a bar hung in the usual manner 

 between the poles of an electromagnet. The distribution of the mag- 

 netic force in the field being known, we can then calculate the force 

 acting on the body, and the comparison of thi? with the time of vibra- 

 tion gives us the means of determining the constant sought. But I 

 will leave the more exact description to be given by Mr. Jacques in the 

 experimental part. 



