98 



SCIENCE. 



[N. S. Vol. XII. No. 290. 



up on the flanks, first attacking them in 

 Vermont. The Ordovician sea followed and 

 its sediments reached points well into the 

 crystalline area. Pursuing the thought 

 further we may raise the query, were the 

 crystallines then reduced to a base-level and 

 did submergence gradually bury them, and 

 did the Ordovician sea and the subsequent 

 Silurian sea go all across from side to side 

 with a continuous mantle of sediments? 

 Or were the crystallines a great island dur- 

 ing all this time and have they remained so 

 with minor faultings and upheavals to the 

 pi-esent? These are questions easy to ask 

 and difficult to answer. The most that we 

 shall say about them now is that they are 

 another story. 



J. F. Kemp. 

 Colombia University. 



ON KATHODE BAYS AND SOME BELATED 

 PSENOMENA. 



II. 



The view here briefly formulated, al- 

 though first suggested by Wiechert, owes 

 its development chiefly to J. J. Thomson. 

 The number of instances in which its con- 

 sequences are at least qualitatively con- 

 fii'med is already surprisinglj' large. Thus 

 it has been known for some time that a wire 

 or carbon filament, when heated to incan- 

 descence in vacuo, sends off negatively 

 charged particles. Thomson * has recently 

 shown that the ratio e/m for such particles 

 is the same as for the kathode rays. Many 

 metals also are capable of giving off' nega- 

 tively charged particles when illuminated 

 by ultra-violet light; at sufficiently high 

 vacua, rays may be produced in this way 

 which possess all the essential properties of 

 the ordinary kathode raj's.f In this case 

 also, the ratio ejm is found to be the same. J 

 In these cases we have an indication that 



*Phil. Slag., 49, p. 547, 1899. 



fMerritt and Stewart, Physikalische Zeitscli., 1, p. 

 338, 1900. 



J Thomson, Phil. 3Iag., 48, p. 547, 1899. 



the corpuscles may be separated from the 

 molecules of a substance by processes differ- 

 ent from those which occur at the kathode. 

 That intense heat, on account of the violent 

 collisions between molecules, should make 

 it easier for the corpuscles to escape, is 

 quite natural. And that the rapid elec- 

 trical vibrations set up bj' light, especially 

 by that of short wave-lengths, should pro- 

 duce a similar effect, agrees equally well 

 with the corpuscular hypothesis. 



If the light radiated by a molecule of gas 

 is due to the vibration or orbital motion of 

 these charged corpuscles, a highly concrete 

 and satisfactory explanation is at once ob- 

 tained of the Zeeman effect. The theory 

 has shown itself capable of accounting not 

 only for the comparatively simple phenom- 

 ena first observed, but also for the more 

 complicated modifications of the spectral 

 lines detected later. The ratio e/m as de- 

 termined from the Zeeman effect is of the 

 same order of magnitude as that determined 

 from observations on the kathode rays. 



Perhaps the strongest confirmation of 

 Thomson's corpuscular hypothesis is that 

 afforded by the recent investigations, of the 

 Becquerel rays. In 1899 it was found that 

 some of these rays, notably those produced 

 by certain preparations of radium, were de- 

 flected in passing through a magnetic field.* 

 More recently, it has been found that the 

 rays are electrostatically deflectedf and that 

 they carry a negative charge. In fact, they 

 behave in all respects like kathode rays. 

 Within the last few months the ratio 

 e/m has been determined by BecquerelJ and 

 found to have approximately the same value 

 as in the case of the Zeeman effect and the 

 kathode rays. 



* Meyer and V. Schweidler, PItys. Zeitsch., November 

 25 and December 2, 1899. Giesel, Wied. Ann., 69, 

 834, 1899. Becquerel, Compies rendus, 129, p. 996, 

 1899. 



t Dorn, Ahhandlungen d. Naturforsch. Gesell., Halle, 

 March 11, 1900. 



t Comptes rendus, 130, p. 809, March 26, 1900. 



