﻿336 Intelligence and Miscellaneous Articles. 



To the Editors of the Philosophical Magazine. 



Gentlemen, — Freiburg i. Br., 22 Dec, J 911. 



In vol. xxii. of the Philosophical Magazine, Dec. 1911, p. 922, 

 Mr. Rossi, starting from Ritz's Theory {Ann. der Phys. 1908), 

 makes some interesting observations on a relation between the 

 atomic volumes and the spectra of elements. May I be allowed 

 to state that I obtained similar results in 1904, which were pub- 

 lished in the PhysiJcalische Zeitschrift, v. p. 302. From the experi- 

 ments of Kayser and Runge, it was shown that the limits v 

 of the branch series for each group of the periodic system are 

 shifted further towards the red end of the spectrum the greater 

 the atomic weight. I started with the supposition that the limits 

 were not only dependent on the atomic weight, but also on the 

 atomic volume (i. e. lineal dimensions of the atoms), and I was 

 surprised to find that the limits can be represented as functions of 

 the atomic volume alone. So far my results are in perfect 

 agreement with those of Mr. Rossi. 



The following which I obtained was not observed by him, but 

 it is well shown in fig. 1 of his paper : that the monovalent 

 elements Cu and Ag fall roughly into the same line as the mono- 

 valent alkali elements, with which they are nearly related in the 

 periodic system. Similarly the divalent elements Zn, Cd, Hg 

 also fall into the same line with those of the divalent elements of 

 the alkaline earths, viz. Mg, Ca, Sr. 



It is scarcely necessary for me to say that Mr. Rossi's result, 

 that the log of the limits of the branch are lineally connected 

 with the log of the atomic volumes, was already known to me, 

 forming the starting-point of my calculations, as can be seen 

 in my figure. I should, however, like to state that Mr. Rossi's 

 interesting result on the different inclinations of the lines of the 

 nearly related elements was not remarked by me, and that in 

 consequence I regard his work as a considerable addition to our 

 knowledge of the subject. 



Finallv, I may add that in 1903 W. Ritz has published another 

 theory of the spectrum-lines (Ann. der Phys. xii, p. 264). From 

 this work also the influence of the lineal dimensions of the atom 

 on the limits r can be inferred, as Ritz imagined we have to do 

 with vibrations of square surfaces where the frequencies depend, 

 amongst other things, on the length of the sides of the squares. 

 In rav work, however, I started from more general considerations. 



I am, 



Yours sincerely, 



M. Reixgaxt^:. 



