642 



NATURE 



[February ii, 19 15 



LETTERS TO THE EDITOR. 



[The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature-. No notice is 

 taken .'of anonymous cotnmunications.'] 



A Penalty on Research. 



We are frequently told that there is no difficulty in 

 being accorded the use of small quantities of absolute 

 alcohol, duty-free. While professor at University Col- 

 lege, it was possible for me to procure what I re- 

 quired, duty-free, for a permit had been granted to the 

 college. Now, working in my private laboratory here, 

 I am refused the privilege. Having applied for per- 

 mission to be allowed to buy not more than two gallons 

 a year of absolute alcohol, free of duty, the secretary, 

 in a letter of February i, says : — "The Commissioners 

 regret their inability to grant you the use of pure 

 spirit, duty-free." 



Now, sir, I have lately been doing a good deal of 

 work for various departments of the Government free 

 of charge ; and for some of this, absolute alcohol is, 

 if not a necessity, at least a great convenience. More- 

 over, my laboratory servant, who is, by the way, a 

 teetotaler, is serving; with the colours. 



I regard this action of the Customs Department as 

 a penalty on research. No doubt it will be said, if an 

 exception (exception to what?) is made in one case, 

 why not in all? The privilege might be abused. I 

 think not. Such a permit might be granted only to 

 persons who are known to be actually engaged in 

 research, and whose bona fides is attested. 



The question of free alcohol is,- you will observe, still 

 unsettled; and I hope that some action will be taken 

 to put an end to such folly. William Ramsay. 



Beechcroft, Hazlemere, Bucks. 



The Spectra of Helium and Hydrogen. 



It has become a general supposition that the validity 

 of Bohr's theory of the spectra of hydrogen and helium 

 would be definitely proved if certain faint components 

 very close to the hydrogen lines given by Balmer's 

 formula could be found experimentally. On the other 

 hand, if the components were not found after exhaus- 

 tive search, the theory would be disproved. This sup- 

 jx)sition is quite incorrect, and in view of the remark- 

 able paper by Mr. E. J. Evans in the current number 

 of the Philosophical Magazine, which describes the 

 successful discovery and measurement of these com- 

 ponents, it becomes a matter of urgency to indicate 

 at once why such components can give no test of any 

 theory of spectral production. 



Into the vexed question of the agent producing the 

 line A 4686, and the Pickering series — whether it be 

 hydrogen or helium — it is not necessary to enter, for 

 the conclusions are the same in either case. We have 

 two alternatives to consider : (i) that the lines are 

 due to helium, and (2) that they are due to hydrogen. 

 Let us, in the first place, take the first alternative, 

 which has been rendered so probable by Fowler in his 

 recent Bakerian lecture. In that case, the line A 4686 

 belongs to an enhanced line series with 4N as the 

 Rydberg constant instead of N. It is strictly 

 analogous, as Fowler points out, to his new series of 

 magnesium, commencing with A 4481, and we must 

 expect , an analogous behaviour. Now no spectro- 

 Scopist questions to-day the universal validity of the 

 combination principle in spectra, developed by Ritz and 



NO. 2363, VOL. 94] 



Paschen, and some well-defined types of combination 

 series are known. In particular, there is a well- 

 marked spark series in magnesium formed by com- 

 binations of the type 3/— tn/, in the usual notation, 

 where nf is the variable part of a line in the "4481 " 

 series. In other words, the wave numbers in this 

 series — a comparatively strong- one — are differences of 

 the wave numbers of lines in the "4481" series. 

 Fowler has measured seven successive members of this 

 series (Bakerian Lecture, p. 253). If we calculate the 

 corresponding series for helium from the " 4686 " 

 series, we obtain precisely the Pickering series and the 

 new components measured by Evans. This calculation 

 is independent of any theory of the production of 

 series. One series must necessarily be accompanied 

 by the other. In this particular combination, the 

 result is even independent of any expression, empirical 

 or otherwise, of the series by a formula. The difficulty 

 of observing the lines in no way contradicts this neces- 

 sary origin. They may have another origin in addi- 

 tion, and be really twofold, but since this process must 

 produce them even in the absence of any other, their 

 existence cannot in consequence contribute to the estab- 

 lishment either of Bohr's or. of any other theory. 



The second alternative is to return to Rydberg's 

 view that half the "4686" series is a principal series 

 of hydrogen, and that the other half is a new principal 

 series. The writer must here insist that he is not 

 expressing a belief in Rydberg's view, though, in his 

 opinion, sufficient justice has not been done to this 

 alternative, which, from the point of view of exact 

 formulae, is not really inferior to the other, although 

 the need for the presence of helium in the experimental 

 isolation of the lines is against it. The important point 

 is that even if this now somewhat discredited view is 

 correct, the new components near the hydrogen lines 

 must exist, and in exactly the same calculated, posi- 

 tions as before. They exist again as combination 

 tones, of the type 3P — mP, where P is now the prin- 

 cipal series of hydrogen, and can be calculated at once 

 by subtracting the wave numbers of the "4686" 

 series, or, in fact, of just the half of it which Rydberg 

 designated as the principal series. This combination 

 from the principal series is well known in the alkalis 

 lithium and sodium, just below hydrogen in the 

 periodic table. The lines could only be expected to 

 occur, however, when A 4686 is strong-, as in the ex- 

 periments of Evans. 



While, therefore, the isolation of these lines is a 

 g^reat step in the advancement of our knowledge of 

 hydrogen or helium, it cannot prove that they are 

 helium lines, and although thev were predicted by 

 Bohr's theory, they were, in fact, previously predicted 

 by the very existence of the "4686" series, whatever 

 its interpretation. While sharing the general admira- 

 tion for the great simplicity of Bohr's theory as applied 

 to the Pickering and "4686" series, the writer must 

 nevertheless point out the one, and apoarently the 

 only, method bv which he believes it can be proved or 

 disproved. This is by interference measurements of 

 the first four or five lines in the "4686" series, from 

 which the value of Rydberg's constant can be calcu- 

 lated exactly, and compared with Curtis's value for 

 hydrogen. The lines are not at present measured 

 accurately enough for this purpose. 



J. W. Nicholson. 



L'niversit\' of London, King's College, February 4. 



On the History of a Notation in Trigonometry. 



The interpretation, of formulas in the trigonometry 

 of plane and spherical triangles is greatly simplified 

 by the expedient of designating the sides by the same 

 letters, respectively, as the angles opposite those sides 



