10 



SCIENCE 



[N. S. Vol. XXV. No. 627 



only fair warning that the forces here in- 

 volved are possibly not to be limited to 

 those of purely electrical origin; but it is 

 not obvious what experiment can be devised 

 to answer these qiiestions. 



6. Passing next to that somewhat motley 

 group of phenomena which we have classed 

 under the head of Multiple Spectra, I am 

 not certain that there is a single fact in 

 the entire group that can be predicted 

 from the structure of the atom which has 

 been assumed. 



In his Royal Institution lecture^' of this 

 year, J. J. Thomson has suggested an atom 

 with certain capacities for receiving and 

 spending energy, and has described this 

 atom by simple differential equations, in- 

 volving certain atomic constants; but the 

 connection between these constants and the 

 Saturnian atom are by no means clear. 



So that while he accomplishes the ex- 

 planation of the phenomena for which this 

 new atom is introduced, the explanation 

 can hardly be said to hinge upon the atomic 

 structiire, which has so highly recommend- 

 ed itself in other directions. 



At present, we seem to be justified in 

 going little farther than to say that rapid- 

 ity of change of electric field seems to be a 

 (not the) determining factor in nearly 

 every case. The main difference between 

 the arc and the spark appears to be con- 

 fined to the earlier stages of the spark. 

 The oscillograph shows that spark lines are 

 introduced into the metallic arc when the 

 break is quickened; the interruption of an 

 intermittent are is very much hastened by 

 a hydrogen atmosphere; and one might 

 think that, therefore, the action of hydro- 

 gen in introducing the spark lines is com- 

 pletely accounted for, and perhaps justly 

 so, if it had not been discovered by Hale, 

 Adams and Gale^^ that an arc fed by a 



"Thomson, Chemical News, 94, 197, 1906. 



'■' Hale, Adams and Gale, Astroph. Jour., 24, 

 213, 1906. 



small steady current and surrounded by 

 hydrogen also yields spark lines. But this 

 calls for an examination of the steady ( ? ) 

 current by means of the oscillograph. 



The effect of a very minute current in 

 introducing spark lines probably also rests 

 upon the greater rapidity with which the 

 small current is interrupted. The effect 

 of self-induction is to retard the break, and 

 hence obviously to obliterate spark lines. 

 The effect of a parallel capacity is, of 

 course, to increase the speed of break. The 

 spark lines obtained in arc under water are 

 apparently special eases of the effect of a 

 hydrogen atmosphere and a consequent 

 rapidity of break. 



A most valuable research at the present 

 time would be one which would determine 

 whether for a 'quickness of break' as a 

 unifying principle we should substitute 

 'rate at which energy is delivered' or, as 

 Thomson suggests, 'rate at which energy is 

 delivered cowibined ivith rate at which en- 

 ergy is spent.' 



Until more definite information upon 

 this point is obtained, it remains almost 

 impossible to say how the Saturnian atom 

 must be modified in order to explain mul- 

 tiple spectra. 



7. As to the relation between spectral 

 phenomena and atomic masses, we must, I 

 think, all feel the most hearty admiration 

 for the manner in which Professor J. J. 

 Thomson has succeeded in picturing the 

 periodic law as an almost immediate con- 

 sequence of the atomic structure which he 

 has proposed. The way in which the nor- 

 mal atom, by the addition of a few sub- 

 atoms, becomes electro-positive or electro- 

 negative is especially attractive. The 

 achievements of Runge and Precht^" in 

 closely approximating if not accurately 

 determining the atomic mass of radium is 



" Runge & Preoht, Physikal. Zeitschrift , 4, 285- 

 287, 1903. 



