the Light emitted by Incandescent Zinc Oxide. 27 



most probable that it indicates a peculiarity of the radiation 

 which begins to show itself at that temperature and which 

 causes the sudden rise in the carves of figs. 1 to 6. In 

 fig. 10 the third method of plotting was used. Platinum at 



Fig. 10. — Radiations from Zinc Oxide at 1013°, as a function of the 

 time. Ordinates give the intensity of each wave-length in terms of 

 that of the radiation from platinum (wave-length for wave-leno-th) 

 at the same temperature. 



1013°, instead of the comparison-lamp, was the reference- 

 standard, and the ordinates of the curves give the ratio of the 

 intensity of each region of the spectrum of the zinc oxide to 

 that of the corresponding wave-length in the spectrum of the 

 glowing metal. 



We think it evident from these experiments that in zinc 

 oxide one has to do with a material which, in addition to the 

 ordinary incandescence due to temperature, is highly lumi- 

 nescent above 880°. The phenomenon is of the class wdnch 

 Becquerel and other of the earlier students of the subject 

 have described as " phosphorescence by heat." Owing to the 

 presence of the ordinary incandescence at such temperatures, 

 this property of zinc oxide seems hitherto to have escaped 

 investigation ; although it has of course been noticed by every 

 one acquainted with blowpipe reactions that the oxide, when 

 heated, emits a light which is very different from that of 

 carbon, or of most substances with the incandescence of which 

 we are familiar. 



There are good reasons for supposing that other of the 

 metallic oxides will show peculiarities of radiation when 



