498 



NATURE 



[August 21, 19 19 



spectrum. There are evidently two quite different 

 mechanisms of interaction possible between the atoms 

 of the two gases — a problem I commend to the chemist 

 for investigation. 



But it is not necessary to mix one gas with another 

 in order to produce the energy transfer. It can be 

 achieved otherwise, as some further slides I have here 

 will suffice to show. We have made many measure- 

 ments of intensity, more especially in the spectrum ot 

 pure helium, of the lines from a pure gas as dependent 

 on the part of the tube they arise from, and on the 

 conditions of excitation. We shall only consider one 

 or two of the more interesting results which arise from 

 a comparison of three spectra of helium : (i) tht' 

 "ordinary" spectrum, or the spectrum given bj the 

 capillary of a vacuum tube containing helium at about 

 a millimetre pressure, excited by the discharge from 

 an induction coil without capacity or spark-gap; 

 (2) the bulb spectrum, obtained by putting a small con- 

 denser and a very small spark-gap in parallel in the 

 circuit; (3) the capillary spectrum with a spark-gap 

 and a strong condensed discharge. In both (2) and (3) 

 the transfer of energy to the more refrangible members 

 of a series takes place very stronglj;. In the diffuse 

 series the transferred energy goes in (3) mainly towards 

 increased breadth of the line, but in (2) mainly towards 

 enhanced central intensity— two quite distinct effects. 

 The sharp and principal series show the same transfer 

 quite definitely, though on a smaller scale, and the 

 effect is in these cases more closely confined to enhance- 

 ment of the central intensity. 



The most striking enhancements produced by the 

 condensed discharge in helium occur with the lines 

 A 4472 and A 4388, which are precisely the helium lines 

 . apt to be found abnormally strongly in the spectra of 



some of the planetary nebulae. Some other experiments 

 we have made, on the spectrum of helium at very low 

 pressure, indicate that these lines, together with the 

 line A 5015 more frequently quoted, are the strongly 

 enhanced lines also in these circumstances. If the two 

 sets of circumstances occur together, A 5015 is not 

 especially strong, but the others are enhanced for both 

 reasons. We have, in fact, been able to demonstrate 

 that the peculiar "nebular " spectrum of helium could 

 be produced in the laboratory by a combination of the 

 condensed discharge with an extremely; low pressure. 



I shall not discuss the spectra of gases as dependent, 

 in their intensity relations, on pressure. The time 

 required would be prohibitive, and my object is to 

 indicate the range of work now open to precise investi- 

 gation, rather than to give any complete account of 

 the phenomena which the method has yet indicated or 

 elucidated. One remark must, however, be made in 

 connection with high-pressure spectra. We investi- 

 gated the intensity distribution in a helium tube at 

 the extraordinary pressure of 42 mm. Except for the 

 trace of hydrogen which came out of the electrodes 

 during the discharge, the helium was pure. Yet the 

 hydrogen spectrum was nevertheless predominant on 

 the plate, and fourteen members of the Balmer series, 

 instead of the usual six or seven at most, could be 

 seen visually as very sharp lines. This phenomenon 

 incidentally cannot be reconciled with the current quan- 

 tum theory of the hydrogen spectrum — perhaps not an 

 unexpected fact to those conversant with the hydrogen 

 spectrum. No atomic theory as yet has begun to inter- 

 pret any of this spectrum except the Balmer series, 

 and many have done this. No spectroscopist can, in 

 fact, accept a theory which can give no hint of the 

 origin of the so-called "secondary spectrum " of hydro- 

 gen, known to arise mainly from the atom, and, in the 

 laboratory at least, the most important and extensive 

 part of the spectrum. The elucidation of this spectrum 

 is in many ways the most fundamental problem of 

 NO. 2599, VOL. 103] 



physics, and far more fundamental than the Balmer 

 series problem. 



Many of the .. ' nroblems of interest, which the 

 possession of an acc- " method of intensity deter- 

 mination in spectra enables us to attack, are mainly 

 of astrophysical importance. There may be variations 

 of intensity in the Fraunhofer lines accompanying 

 other more readily perceived solar phenomena, for ex- 

 ample, but of more urgent importance is the need for 

 a series of photographic registers of the intensity across 

 the whole spectrum of a new star at different stages 

 of its history. It has not often been possible even to 

 determine the actual number of component radiations, 

 in an apparent broad band with a structure, emitted 

 from such a star — at least with any real certainty. A 

 method which automatically sifts out such bands and 

 gives peaks on a photograph at all the maxima of in- 

 tensity in the band may well be expected to contribute 

 greatly to the elucidation of the phenomena taking 

 place, which must in any case be totally different from 

 anything known by our terrestrial experience. 



The only other class of phenomenon depending for 

 its elucidation on precise measures of intensity in spec- 

 trum lines, to which I shall refer with further illustra- 

 tive slides, is the variation which takes place in the 

 spectrum from a helium tube as we recede from the 

 cathode. The slides serve to show the considerable dif- 

 ferences which take place in the distribution of the 

 various series, which are all emitted most strongly at 

 unequal distances from the cathode. One very extra- 

 ordinary result, shown clearly on the last slide, is the 

 fact that there exists a narrow region of the tube in 

 which the characteristic spark line A 4686 is emitted 

 simultaneously with the helium band spectrum — a cir- 

 cumstance which necessitates some readjustment of 

 preconceived ideas. 



UNIVERSITY AND EDUCATIONAL 

 INTELLIGENCE. 



Cambridge. — Mr. S. Lees, who was recently ap- 

 pointed University lecturer in thermodynamics, has 

 been re-elected a fellow of St. John's College. 



London. — The following appointments have been 

 made in connection with the newly instituted school 

 of librarianship at University College, of which, as 

 already stated, Dr. E. A. -Baker is the director : — 

 Bibliography, Mr. A. Esdaile; Cataloguing and 

 Library Routine, Mr. W. R. B, Prideaux ; Classifica- 

 tion, Mr. W. C. B. Sayers; Public Library Law, 

 Mr. H. W. Fovargue; Library Organisation, Mr. 

 B. M. Headicar; Literary History, Dr. A. W. 

 Chambers ; Literary History and Book Selection, 

 Dr. E. A. Baker; Palaeography and Archives, Mr. H. 

 Jenkinson ; assistant to the director, Mr. L. F. New- 

 combe. The work of the school is to begin on 

 October i, but the formal opening will take place on 

 October 8, at i, o'clock, at the hands of Sir F. G. 

 Kenyon, the director and principal librarian of the 

 British Museum. 



Dr. S. W. Patterson has been appointed director of 

 the Eliza Hal! Institute of Research, in connection with 

 the Melbourne Hospital. 



Mr. W. H. N. James has been appointed head of the 

 electrical engineering department of the Bradford 

 Municipal Technical College. 



Mr. J. A. R. Marriott has intimated to the vice- 

 chancellor of the University of Oxford his intention to 

 resign the secretaryship to the University Extension 

 Delegacy in March next. 



