600 SCIENCE PROGRESS 



Le Radium, vol. vii, p. 251, 19 10. Cotton and Mouton, Ann. 

 de Chim. et de Phys., 8, vol. xix, p. 155, 19 10.) 



Atomic Magnetism. — The additive nature of the diamagnetic 

 property will now be considered. Pascal (Ann. de Chim. et 

 de Phys., vol. viii, p. 19, 19 10) has shown that in the case of a 

 large variety of organic compounds, the sum of the atomic 

 susceptibilities of the component atoms is equal to the mole- 

 cular susceptibility of the compound, providing the compound 

 does not contain any one of certain types of peculiarity such 

 as the benzene and naphthalene nuclei, an ethylene linkage, 

 etc. In any case, if the atomic susceptibility x of a component 

 element be % a , the molecular susceptibility of the compound 



X™ tnen 



where X is a constant, for a given series of compounds, having 

 a definite peculiarity of structure. 



The atomic susceptibility of hydrogen is — 30-5 x io~ 7 , that 

 of carbon— 59-5 x io~ 7 . It is not a little remarkable that the 

 differences of molecular susceptibilities of a series of hydro- 

 carbons, whose constitutions differ by the group CH 2 , is always 

 very nearly — 120 x io~ 7 , i.e. the sum of the susceptibilities 

 of one carbon and two hydrogen atoms in the elementary 

 state. It appears that the addition of one or more atoms, 

 or of a group of atoms, adds a definite amount of diamagnetism 

 to the compound, at least in so far as the hydrocarbons are 

 concerned. Complications are found, however, in compounds 

 containing oxygen, nitrogen, etc., which behave differently 

 according to the positions of such atoms in the molecule, but 

 even in these cases the effect of carbon, hydrogen or CH 4 , 

 is as described above. The author has shown (Proc. Roy. 

 Soc., A, vol. xcv, p. 58, 191 8) that if the electron in the hydrogen 

 atom rotate with a frequency corresponding to the line H a , 

 then the magnetic moment of the orbit of radius io~ 8 cms., 

 which gives an atomic susceptibility of — 30-5 x io~ 7 , is equal 

 to 16-3 x io" 22 , which is nearly equal to the moment of the 

 magneton, 2 as derived from ferro-magnetic considerations. 

 In the light of these results it seems that in considering the 

 atomic constitution of the molecule due regard must be paid 

 to magnetic effects. (See also a paper by Sir J. J. Thomson, 

 Phil. Mag., April 191 9.) 



1 Atomic susceptibility = susceptibility per gram of element 



x its atomic weight. 

 Molecular susceptibility = susceptibility per gram of compound 



x its molecular weight. 



2 The magneton is an experimentally derived unit of magnetism, obtained 

 by Weiss, from measurements on the saturation intensities of magnetisation 

 of a number of ferro-magnetic elements and alloys. 



