162 



SCIENCE 



[Vol. LV, No. 1416 



which constituted these spectra. It was obvious 

 that if a concave diffraction grating could be 

 substituted for the prisms and lenses wave- 

 length measurements could be made and at 

 the same time the limitations set by the opacity 

 of fluorite could be avoided. 



I attacked the problem a good many years 

 ago and most of the results have been long 

 since in print. The spectrum of hydrogen was 

 extended to the neighborhood of 900 units and 

 its wavelengths were measured to within an 

 error 0.2 of a unit, the accuracy being checked 

 by Wolff through an ingenious application of 

 certain series relations devised by Paschen. In 

 a fruitless search for substances more trans- 

 parent than fluorite the opacity of a large 

 number of solids was next tested. The absorp- 

 tion of the commoner gases was measured and 

 it w-as demonstrated that oxygen partly 

 resumed its transparency with decreasing 

 wavelength, a result which has been recently 

 confirmed as we shall presently see. A con- 

 siderable time was spent in the study of the 

 emission of various substances and in the 

 measurement of their spectra; partly because 

 they occurred as impurities in the hydrogen 

 discharge and partly because of the interest 

 which attached to certain series relations con- 

 necting the radiations of some of them. 



Toward the end of the period covered by 

 these researches W. T. Bovie- of the Harvard 

 Medical School became interested in the abiotic 

 effects of Schumann rays, and carried out sev- 

 eral investigations in that field. At nearly the 

 same time F. H. Palmer^ conducted a study of 

 the volume ionisation produced by ultra-violet 

 light; his experiments afford what I believe is 

 the most direct proof of the effect yet obtained 

 by any one. Later P. E. Sabine* carried out a 

 preliminary investigation of the photo-electric 

 effect in the region of very short wavelengths; 

 and I. C. Gardner^ made a quantitative study 

 of the reflective power of metals and demon- 

 strated the superiority of platinum and silicon 

 over speculum in the extreme ultra-violet. 



"i Botanical Gazette, 59, No. 2, 1915; 60, No. 2, 

 1915; 61, No. 1, 1916, etc., etc. 

 3 Fhy. Rev., 32, p. 1, 1911. 

 * Phy. Bev., N. S. 9, p. 210, 1917. 

 5 Astrophysical J., 45, p. 30, 1917. 



Having completed a reconnaissance of the 

 region discovered by Schumann it was natural 

 that attention should turn to its extension 

 toward the ultra-violet. By the use of helium 

 gas and by the employment of a strong dis- 

 ruptive discharge I succeeded in pushing the 

 spectral limit to the neighborhood of 500 AU 

 in 1917." In the process of this research vari- 

 ous gas spectra including those of argon and 

 nitrogen came under observation; moreover, 

 the existence of the series in hydrogen pre- 

 dicted by Eitz and obeying the law 



F = iVll I was confirmed. 



During the period whose activities have so 

 far been described interest in the Schumann 

 region had been confined to a very few, and 

 the work had been cari'ied on in but a limited 

 number of places. We now enter on an epoch 

 extending to the present time, where the study 

 of this field bids fair to be much more general. 

 In view of this awakening interest it may be 

 well to pause long enough to say something of 

 the technique which is involved in the study 

 of the spectral region under discussion. 



The spectroscopy of the extreme ultra-violet 

 combines all the difficulties inherent in the 

 nice adjustment of delicate optical apparatus 

 with those which accompany the production 

 and maintenance of high vacua in metallic 

 containers of large volume. The first are only 

 to be overcome by skill and patience; the 

 second have been greatly alleviated during 

 recent years by the improvement in mechan- 

 ical vacuum pumps and by the use of charcoal 

 and liquid air. In the design of the vacuum 

 spectroscope, whether it be intended for prism 

 or grating, simplicity is the chief requisite. 

 This I was at some pains to point out years 

 ago. A good example is afforded by the 

 highly successful apparatus of Millikan where 

 grating, plate holder and source are all con- 

 tained in one cylindrical brass tube closed at 

 the ends by simple plates made air tight with 

 rubber gaskets. The improvements which 

 McLennan has introduced have also been 

 directed away from the pernicious ingenuities 

 of the instrument makers. 



6 Astrophy. J., 43, p. 89, 1916. 



Science, 45, p. 187, 1917. 



