58 



SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



atomic volumes in the case of those elements in which the compres- 

 sion is reasonably supposed to be so great that the envelope has 

 practically vanished (see §15). Such are Iron and Platinum. Divid- 

 ing the atomic volumes of these by Avogadro's constant (6.0 X io 23 ), 

 we get 11.4X io -24 and 14.6 X io~ 24 as values for V in the two cases. 

 Since their respective magneton numbers are 32 and 80, the two 

 values 2.2 xio -9 and 1.8 xio -9 are got for the magneton's radius. 

 That Iron should give the larger value was to be expected, for the 

 assumption that the envelope has vanished is less nearly true for it 

 than for Platinum, as their relative compressibilities show. Allowing, 

 then, for the presence of some envelope even in metallic Platinum, 

 we may take the radius of the magneton to be about 1.5 x io~ 9 cm. 

 This is only about one-tenth of the radius of the total atomic space 

 usually found in a solid or liquid, where the envelope may occupy a 

 large part of it. 1 



Now since the velocity at the circumference of the magneton has 

 been assumed equal to c (the velocity of light), we have for the 

 moment of the magneton : 



Moment = area x current 



ec 

 271-r 



= 7ir 2 X 



rec 



~~ 2 



= (l.5XIO" 9 ) X (1.55 XIO" 20 ) X (3XIO 10 )-^2 



= 3.5Xio~ 19 E. M. U. 



1 The radii of some atomic spaces may be given for comparison with those 

 of the corresponding positive spheres, and that of the magneton. It should 

 be noticed that the values of (R'—R) got from the table below give the 

 volume of the envelope. 



R' (of atomic space) K (of positive sphere) r (of magneton) 



H (solid) 1.7 X io" 



He (liquid) 2.2 



C (solid) 1.1 



O " 



Na " 



K " 



Fe " 



Pt " 



1-7 



2.1 

 2.6 

 i-4 



i-5 



36 X io-s 



72 



83 



85 



9i 



98 



1 



5 



.15 X io" 8 



The values given for R must all be too great because they are derived from 

 the assumption that the envelope has quite disappeared in solid Platinum. In 

 the case of H there is also a factor that would tend to make it too small — 

 the lack of internal magnetic compression has been left out of account. In a 

 subsequent paper, evidence will be adduced from the phenomena of electrolytic 

 dissociation to show that the positive sphere of this atom has a radius of about 

 .50 X io-s. 



