February 17, 1922] 



SCIENCE 



163 



Time does not permit the discussion of 

 details of design but it is useful to note that 

 McLennan has emphasized the advantage of 

 horizontal surfaces for those joints which must 

 be often made and broken; I venture to add 

 that I still find the gasket formed of a string 

 of soft wax useful in such joints even when 

 the highest vacuum is required. Moreover, I 

 have found that a carefully cut screw of say 

 % mm. pitch working in a nut about 6 cm. 

 long will serve to communicate motion to elec- 

 trodes within the apparatus without introdvicing 

 a leak provided the screw and nut are sealed 

 together with wax after each adjustment. 



There is one fault in vacuum spectroscope 

 design so common that it merits particular 

 attention. It is concerned with the question 

 of angular aperture in those cases where a 

 lens cannot be employed. For example, if the 

 slit is placed on the circle whose diameter is 

 the grating's radius of curvature, and to which 

 the plate should conform, it is mechanically 

 diflicult to place the source of light sufficiently 

 near the slit to fill the grating. Millikan has 

 overcome the trouble, as we have just seen, by 

 placing the source of light within the body of 

 the spectroscope, but this process presents 

 some inconveniences when vacuum tube spectra 

 are to be examined. 



The grating affords the only means at our 

 disposal for analyzing light of the very short- 

 est wave lengths spectroscopically, but its use 

 is accompanied by several disadvantages, not 

 the least of which arises from the fact that an 

 amount of tarnish which would be quite harm- 

 less in the ordinary part of the spectrum 

 proves fatal in the region under discussion. 

 To the favored few who have a ruling engine 

 at their command the difficulty is overcome by 

 the simple exjjedient of ruling a new gi'ating! 

 The majority are not so fortunate, they 

 may therefore be interested in the results of 

 some experiments of my own. Guided by the 

 work of Gardner, I have tried covering a tar- 

 nished grating with a thin cathode deposit of 

 either platinum or silicon. The results seem 

 favorable and I have good hopes that the 

 method affords a means of rejuvenating diffrac- 

 tion gratings especially if silicon be employed. 



We may now return to a consideration of 



the results of the last half dozen years. In 

 1914, Saunders,' working at Tiibingen, fol- 

 lowed the spectra of calcium and zinc in a 

 vapor lamp to the neighborhood of 1000 A.U. 

 He also confirmed my observations on the 

 hydrogen lines of the Ritz series. 



About the same period L. and E. Bloch^ 

 began their investigations; their work, inter- 

 rupted by the war, has recentlj' been resimied. 

 Their vacuum spectrometer contains a train of 

 fluorite and includes the novel feature of a 

 constant deviation prism. They have measured 

 the spark spectra of sixteen elements, to the 

 neighborhood of 1400 A.U. Their tables con- 

 tain not only their own results but also those 

 of other investigators in the same field. The 

 determinations of wavelength rest on my values 

 for certain lines in aluminium, hj'drogen, mer- 

 cury and nitrogen. It may be noted paren- 

 thetically that wherever possible a direct com- 

 parison with the spectrum of hydrogen should 

 be used in measurements between 2000 and 

 1000 A.U. 



McLennan," in collaboration with his stu- 

 dents, has measured both the arc and spark 

 spectra of a considerable number of substances. 

 He has employed both prism and grating 

 instruments and with the latter he has pushed 

 his results to the neighborhood of 500 A.U. 

 His researches have emphasized the importance 

 of the suppression of water vapor in the spec- 

 troscope; to this end he has added ample 

 drying tubes to his apparatus. It is particu- 

 larly interesting to note that he has succeeded 

 in obtaining radiations down to the neighbor- 

 hood of 1020 A.U. through an atmosphere of 

 helium over two meters long at a pressure of 

 29 cm., thus confirming the transparency of 

 the gas in the Schumann region. 



His most recent work revives the dis- 

 cussion as to the existence of the series 

 in helium corresponding to the fonnula 



F = 4iV I I . I have presented^" some 



evidence for the existence of the two first mem- 



' Astrophy. Jour., 40, p. 377, 1914. 

 s Jour. d. Phy. et le Sadiuin, 2, p. 229, 1921. 

 aProc. Eoy. Soc, 95, p. 258, 95, p. 316, 1919; 

 98, pp. 95-123, 1920. 



10 Science, 50 p. 481, 1919. 



