550 . Prof. J. W. Nicholson on the High-frequency 



periodic character or cumulative ? If they are cumulative, 

 we can discover the fact by treating them as small, and 

 finding that they actually are large, somewhat after the 

 usual manner of testing the stability of a system, although 

 of course in this case the problem is not one of stability, but 

 of the ordinary " steady motion " of the atom, not involving 

 any radiation of energy. 



It would not be logical, on any view, to apply Newton's 

 Second Law only to the circular steady state, and not to any 

 other. If the component of radial force is equal to the mass- 

 acceleration in the one case, as is necessary for Bohr's 

 spectral formula, it must be also in the other. Bohr's only 

 deviation from ordinary dynamics, apart from the quantum- 

 character of the radiation, occurs in relation to the com- 

 ponent acceleration of an electron in its plane, transverse to 

 the line joining it to the nucleus, — the equation for which is 

 replaced by the law of constancy of angular momentum. 

 The necessity for this law seems to point to something in 

 the nature of a tube of force, constituting the mechanism of 

 " binding,'''' which compels the tube to have a definite angular 

 momentum in the plane of the ring, but not a definite length, 

 for the electron at the end takes up the corresponding- 

 position determined by the forces and Newton's law, as 

 Bohr assumes throughout his paper. Transverse to the 

 plane, also, Newton's law can be used even for vibrations. 



The Atom of Lithium. 



The configuration for the atom of lithium, suggested by- 

 Bohr's theory, consists of a ring, approximately circular, of 

 two electrons rotating round a nucleus 3e, with a solitary or 

 valency electron outside. Bohr chooses this configuration in 

 spite of the fact that the collection together of a single ring 

 of three electrons would correspond to a greater emission of 

 energy, on his mode of calculation. The chemical valency 

 of lithium is definitely unity, and the system at first sight is 

 very attractive. We must examine the deviations of the 

 orbits from the circular form, in order to ascertain whether 

 the system can be permanent. Bohr's method of treatment, 

 by which his approximations are obtained, would evidently 

 fail if the deviations were cumulative. If they are small, 

 the substitution of a uniformly electrified ring for a ring of 

 electrons will give a very accurate idea of the effect of one 

 ring on another, and in fact an examination of Bohr's Table*, 

 which the writer has made, shows that it gives surprisingly 



* L. c. p. 485. 



