4:64: /Scientific Intelligence. 



cipitate under the proper conditions for the copper test. The in- 

 terference of the mercury may be overcome by the addition of 

 sodium chloride, while the zinc compound may be prevented from 

 precipitating by increasing the acidity of the liquid. The reagent 

 is characterized as more specific for copper than anj' heretofore 

 proposed, and it possesses the advantage over the ferrocyanide 

 test in the fact that small quantities of iron do not interfere with 

 it. — Jour. Amer. Client. iSoc, xxxvii, 1471. h. l. w. 



5. A Theory of Magnetism. — Since the older theories of mag- 

 netism are not in complete accord with one another and since no 

 single theory accounts for all of the experimental facts, much 

 interest and importance attach to a theory which has been recently 

 worked out in detail by R. Gans and which seems to agree with 

 all observed phenomena. The fundamental model made use of 

 in this theory is termed the " magneton " and is defined by Gans 

 as a rigid system of electrons or negative electric charges. The 

 magneton is supposed to be situated in the interior of a sphere 

 which is uniformly charged with positive electricit} 7 , the total 

 positive charge being equal numerically to the total negative 

 charge constituting the magneton. Moreover, the center of mass 

 of the magneton is assumed to coincide with the center of the 

 containing sphere. The magneton itself does not have to be geo- 

 metrically a figure of revolution. For example, a prism of square 

 cross-section is used by Gans as an illustration of a form which 

 possesses two equal principal moments of inertia but no geometric 

 axis of revolution (*' Figurenachse "). The results of the mathe- 

 matical analysis may now be summarized. 



When the magneton has no geometric axis of revolution, the 

 excitation in a magnetic field is always diamagnetic. 



The diamagnetic susceptibility is independent of the field 

 strength and of the temperature when the three principal moments 

 of inertia are all equal to one another. This phenomenon has 

 been frequently observed. 



If the principal moments of inertia have not all the same value 

 then the diamagnetic susceptibility does depend upon both the 

 field and the temperature. With the exception of very low 

 temperatures, this has been found experimentally by Honda to 

 be true. 



A quantitative comparison of the theory with experimental 

 data will give the number of magnetons in a gram molecule and 

 also the values of the principal moments of inertia of the 

 magneton. 



The excitation is paramagnetic when the magneton possesses a 

 geometric axis of revolution. The magnetic moment remains 

 constant (as experience requires), since the magneton does not 

 radiate, and it does not depend upon the temperature. The last 

 theoretical deduction is in complete accord with the experimental 

 discoveries of Weiss. In particular, at the absolute zero of tem- 

 perature the magnetic moment of the magneton has a finite value 

 (not zero). This too agrees with the experimental facts at 



