36 Very — On a Possible Limit to Gravitation. 



tion increases as the radius diminishes), which seems a 

 bit paradoxical, since it is natural to suppose that 

 increased mass comes from the addition of new acces- 

 sions, and this could happen if we suppose that the 

 minute positive electrons are symmetrically arranged in 

 the form of a thin shell. It is also in accordance with 

 all of the electrical analogies that the static part of the 

 electrification should seek the surface. The rapidly 

 moving electrons are elements of electric current, but this 

 does not prevent them from also playing the part of 

 static charge in potency, even though this charge may be 

 neutralized. 



General Scheme of Electronic Motions within the Nucleus. 



In considering the motions of the negative electrons, 

 some modifications are needed. The Larmor-Bohr- 

 Eutherford atom is likened to a solar system in which 

 the negative electrons, like the planets, move substan- 

 tially in a single plane. This may be true of certain 

 satellite electrons concerned in chemical action, but I 

 would suggest that not all of the orbital planes need coin- 

 cide. The larger part of them may be parallel, but dis- 

 tributed on the spherical surface of the atom. The 

 resultant effect will be the same in many respects as if 

 all were in one plane. This leads to a consistent scheme 

 of motions according to the following plan: 



Assume as a preliminary conception that all of the 

 electrons in a meridional section of the atom are either 

 ring-shaped, or else approximately oblate spheroidal 

 particles of a uniform size (equatorially expanded by 

 centrifugal pressure of the spin) and with a common (let 

 us say, clockwise) rotation over a given half-section as 

 seen from our point of view; and that they are revolving 

 rapidly in planes parallel to the equator of the atom 

 (with the electrons juxtaposed, pole to pole, or forming 

 virtual vortex-filaments) in definite, but not necessarily 

 invariable orbits, since they may pass into spirals whose 

 radii vectores alternately expand and contract when the 

 elastic surface of the atom is disturbed by a collision. 

 The electrons having like rotation mutually repel one 

 another, except in the direction of the spiral vortex-fila- 

 ment; and this may compel some motion of the orbital 

 planes and change of configuration of the orbits during 

 radiation. Juxtaposed electrons on either side of an 



