August 24, 1900.] 



SGIENGK 



297 



become more complicated than triplets as 

 well as some that are apparently unaffected; 

 moreover the separation is very far from 

 varying as the square of the wave-length, 

 and recent work has shown that in some 

 cases at least it is not proportional to the 

 strength of the field.* In spite of these 

 inconsistencies, however, we do not feel 

 called upon to abandon the theory of 

 electrified ions, for we must bear in mind 

 that Lorentz's expression was deduced from 

 assumptions which can hardly be realized 

 in nature. He assumed a molecule of the 

 simplest possible kind, consisting of a 

 single positive or a negative ion acted upon 

 by a central force proportional to its dis- 

 placement and an electromagnetic force 

 due to the external field equal in magnitude 

 and direction to that which would act on a 

 conductor carrying a current equal to the 

 product of the velocity of the ion by the 

 charge which it carries. Now it seems 

 reasonable to suppose that the central 

 force varies directly as the first power of 

 the displacement because if it varied as any 

 other power the period of vibration would 

 change with the amplitude, and the spec- 

 tral lines would change their position when 

 the source of light became brighter, which 

 has never been observed. The assumption 

 that the same forces act on a particle carry- 

 ing a charge e with a velocity v as would 

 act on a conductor carrying a current of 

 strength ev in the same direction is justified 

 for comparatively low velocities by Eow- 

 land's experiment in Berlin in 1876. It 

 seems utterly impossible, however, that a 

 molecule should consist of a single ion, 

 for in very few cases does the spectrum of 

 an element contain less than twenty lines 

 in the visible spectrum, and in the iron- 

 spectrum there are thousands of them. A 

 molecule which can vibrate in so many dif- 

 ferent periods must be exceedingly compli- 

 *Shedd, Phys. iJei)., July, 1899, p. 1; Aug., 1899, p. 

 86. Eeese, J. H. U. Circulars, June, 1900. 



cated. It is not surprising, then, that our 

 simple theory is inadequate to account for 

 the facts. Lorentz, in fact, knew this and 

 instituted * a theoretical research on more 

 general grounds before its insufficiency had 

 been shown by the discovery of the quad- 

 ruplet and other complications. He found 

 that if the molecule naturally possessed 

 more than three equivalent modes of vibra- 

 tion — that is, if it could vibrate in more 

 than three ways with the same period — 

 then the single spectral line corresponding 

 to this period would become more than 

 three-fold under the influence of magnetic 

 force. Professor Lorentz does not regard 

 this explanation as satisfactory, owing to 

 the difficulty in conceiving a system having 

 this property. 



More recently Voigtf has proposed a 

 theory which accounts for all the observed 

 phenomena and is especially interesting in 

 that by it he predicted cases of asymmetry 

 found by Zeeman and others. Unfortu- 

 nately the theory does not give any me- 

 chanical conception of the subject, merely 

 consisting of the introduction into the equa- 

 tions of motion of terms of arbitrary form, 

 which have no apparent justification. 



It is comparatively easy to treat the case 

 of a molecule composed of two ions carry- 

 ing equal charges of opposite signs, and, in 

 fact, Professor Rowland has lately given 

 such a treatment before his students at 

 Johns Hopkins University, but it leads to 

 no new results as regards the Zeeman ef- 

 fect. Any case more general than this is 

 very difficult. Heebeet M. Reese. 



EUROPEAN APPLE TREE CANKER IN 

 A3IERICA. 



Shoetlt after bulletin Fo. 163 of this 

 station, entitled ' A New York Apple Tree 



* Wied. Annal., vol. 63, p. 278. Asiroph. Jour., 

 vol. 9, p. 37. 



t Wied. Annalen, No. 2, 1899, p. 345 ; No. 6, 1899, 

 p. 352 ; No. 9, 1899, p. 290 ; No. 2, 1900, p. 376, and 

 p. 389. 



