676 PROCEEDINGS OF THE AMERICAN ACADEMY. 



fully, clear down into the green, and, while fairly transparent to the rays 

 of high luminosity, was preferable to either of the glasses just described 

 for reducing the painful glare of powerful illuminants. 



Figure 9, Plate 1, taken at the end of this series, shows again the spec- 

 trum of the bare quartz tube with an exposure of 15 seconds. The plate 

 therefore, in this case, received just five per cent of the total energy re- 

 corded on Figure 1, Plate 1. The line near w. 1. 391 fx^i appears faintly 

 in the negative, but is lost in the reproduction. The line near w. 1. 

 405 nfj. does not appear on this plate since, in this case, as in some of 

 the others, the plateholder had been shifted by a small amount in order 

 to reach down a little further into the ultra-violet. A comparison of 

 this plate with those showing the effects of the absorbing glasses gives 

 a vivid idea of the extent of the absorption produced by the glasses, 

 especially those having even slight coloration. 



In connection with these figures of Plate 1 should be considered 

 Figure 8 of Plate 2, exposed under parallel conditions for 15 minutes. 

 This shows the extreme ultra-violet portion of the quartz lamp spec- 

 trum, the last line visible in the reproduction being at w. 1. 230 nfi. 

 With the exception of the doublet at w. 1. 313 hij,, this portion of the 

 ultra-violet was obliterated by the absorption of the glass in all the 

 spectrograms shown. The relatively great richness of the spectrum in 

 the extreme ultra-violet portion down to about w. 1. 230 nn, where at- 

 mospheric absorption became very powerful, is conspicuous in this 

 figure. In the lower portion of this spectrum are indicated the so- 

 called " abiotic " radiations, which are particularly active in cell destruc- 

 tion, and as the quartz lamp here tested was put to use for bactericidal 

 experiments, this portion of the ultra-violet was of special interest. 

 As Cernovodeanu and Henri 7 have shown, the bactericidal effects are 

 practically confined to wave lengths below 270, and increase beyond 

 this point with enormous rapidity, an effect very possibly corresponding 

 to the powerful ionization produced in this part of the extreme ultra- 

 violet. 



Figures 1 to 7, inclusive, of Plate 2, are not strictly comparable with 

 the figures of Plate 1, since they were taken at a later period, after the 

 spectrograph had been in use for other work. The apparatus w^as set 

 up anew. The grating was turned over, giving a somewhat more 

 brilliant first order spectrum, and the brightest portion of the quartz 

 tube was focussed upon the slit. The exposure in this series was, 

 except when noted, 5 minutes, as before, but the intensities were not- 

 ably higher. 



' Elektrotechnik u. Maschinenbau, 28, 69. 



