WITH MAGNETISM AND ELECTRICITY. l65 



true internal cause, where, besides the known external cause, 

 we become (through its actions) acquainted with the diamag- 

 netism which is the resultant of both causes. Following up the 

 way here indicated, and combining the known magnetizing force 

 with the ideal distribution deduced from the observed actions, 

 for iron as well as for bismuth, we learn that the same force of 

 magnetization causes opposite ideal distributions in the cases of 

 iron and bismuth ; or, inversely, the same ideal distribution in 

 iron and bismuth corresponds to oppositely directed magnetizing 

 forces. The reason why opposed external causes produce the 

 same effect in iron and bismuth must be referred to different 

 internal causes within the iron and the bismuth themselves. To 

 determine more accurately the difference between the internal 

 causes of iron and bismuth, it is necessary to classify all possible 

 internal causes which can produce effects explainable by the 

 ideal distribution, and then to ascertain whether, among all that 

 we can reckon, such are embraced as will enable us to render an 

 account of the above antithesis between magnetic and diamag- 

 netic bodies, subjected to the same external influences. 



Classification of the internal causes which may be assumed as the 

 sources of the actions explainable by an ideal distribution. 



We can adduce fotir essentially different kinds of internal 

 causes which are capable of producing effects explainable by an 

 ideal distribution : — 



1 . The internal cause of such effects may be referred to the 

 existence of two magnetic fluids, which are more or less inde- 

 pendent of the ponderable matter which carries them. 



2. They may be due to the existence of two magnetic fluids, 

 which are only capable of moving in connexion with their pon- 

 derable carriers (rotatory molecular magnets). 



3. They may be due to the existence of permanent molecular 

 currents formed by the electric fluids, and which rotate with the 

 molecules. 



4. They may be due to the existence of electric fluids which 

 can be thrown into molecular currents. 



These four possible internal causes of the actions explainable 

 by an ideal distribution on the surface are the only ones which 

 are known, and which can be submitted to examination. The 



SCI EN. MEM.— .Va/. Phil. Vol. I. Part II. N 



