Dr. Tyndall on Diamagnetism and Magnecrystallic Action, 167 



During an investigation on the magneto-optic properties of 

 crystals*, which I had the pleasure of conducting in connexion 

 with Professor Knoblauch, I had repeated opportunities of ob- 

 serving phsenomena exactly similar to those observed by M. 

 Pliicker with the cherry-tree bark ; but a close study of the sub- 

 ject convinced me that the explanation of these phsenomena by 

 no means necessitated the hypothesis of two forces acting in the 

 manner described. Experiment further convinced me, that a 

 more delicate apparatus than the balance used by M. Pliicker 

 would be better suited to the measurement of such feeble mani- 

 festations of force. 



An exact acquaintance with electro-magnetic attractions ap- 

 peared to be a necessary discipline for the successful investiga- 

 tion of diamagnetic phsenomena; and pursuing this idea, an 

 inquiry was commenced last November into the action of an 

 electro-magnet upon masses of soft iron. I was finally led to 

 devote my entire attention to the attraction of soft iron spheres, 

 and the results obtained were so remarkable as to induce me to 

 devote a special memoir to them alonef. 



In this investigation it was proved, that a ball of soft iron> 

 separated by a small fixed distance from the pole of an electro- 

 magnet, was attracted with a force exactly proportional to the 

 square of the exciting current. Now this attraction is in each 

 case the product of two factors, one of which represents the mag- 

 netism of the magnet, and the other the magnetism of the ball. 

 For example, if the magnetism of the magnet of any given mo- 

 ment be represented by the number 4, and that of the ball by 3, 

 the attraction, which is a consequence of their reciprocal action, 

 is represented by the number 12. If we now suppose the mag- 

 netism of the magnet to be doubled by a current of double 

 strength, the ball will have its magnetism also doubled, and the 

 attraction resulting will be expressed by the number 48. Thus 

 we see that a doubling of the power of the magnet causes four 

 times the attraction ; and that while the attraction increases as 

 the square of the current, the magnetism of the ball increases in 

 the simple ratio of the current itself 



Our way to a comparison of magnetism and diamagnetism is 

 thus cleared. We know the law according to which the magnet- 

 ism of an iron ball increases, and we have simply to ascertain 

 whether the diamagnetism of a bismuth ball follows the same 

 law. For thie investigation of this question I constructed the 

 following apparatus. 



In two opposite sides of a square wooden box two circular 

 holes were sawed about four inches in diameter. The holes were 



* Philosophical Magazine, July 1850. 



t Phil. Mag., April 1851. Poggendorff's Annalen, May 1851. 



N2 



