﻿and Absorption by Resonating Gas Molecules. 711 



of a metal was being gradually thrown down upon it, and this 

 increase of the reflecting power was sufficient to develop 

 the diffraction halo due to the slight roughness of the inner 

 wall. The fact that the contracting glow usually had 

 a definite though irregular shape, and was frequently 

 accompanied by neighbouring patches of lesser intensity, 

 made me suspicious of the phenomenon in the beginning, 

 but it was not until I noticed the slight devitrification of the 

 surface that I was able to give a satisfactory explanation of 

 the thing. In PL XL fig. 6, the last impression was taken 

 after the mercury arc had been burning for half a minute, 

 and no longer emitted the exact frequency for which the 

 mercury molecules responded. The reflexion in this case is 

 due wholly to the quartz, and we see only the two minute 

 points of light previously referred to. 



The real phenomenon is shown in PL XL fig. 7, made 

 with the clean bulb. A and B were taken at room tempera- 

 ture, the former with a five-second exposure, the latter with 

 one of half a second, to bring out better the relative 

 intensities of the two images reflected from the inner and 

 outer walls. I threw these out of focus a little, «o that their 

 relative intensities could be better determined. They appear 

 expanded into circles of light due to zonal errors of the lens. 

 It is clear that at room temperature the lower circle is much 

 brighter than the upper, which was the one reflected from 

 the inner wall. The other pair of circles are reflected from 

 the back wall, and disappear in the other pictures owing 

 to the failure of the radiation to penetrate to the back. 

 Pictures and D were made with a pressure of mercury 

 vapour of about 20 mm. The glow is now confined to that 

 portion of the wall which receives the direct rays, and the 

 image formed by the outer wall is still brighter than that 

 formed by the inner, the ratio being about the same. E 

 and F were made with a pressure of very nearly one atmo- 

 sphere. The diffuse glow has entirely disappeared, and there 

 remain only the two regularly reflected images, the upper 

 circle now being distinctly brighter than the lower. That 

 this image (the upper circle) really came from the inner 

 wall was proved in the following way. The heated bulb was 

 cooled on the illuminated side by a blast of air. This caused 

 a very fine " dew " of mercury globules to condense on the 

 wall, practically silvering it on the inside. A photograph 

 was immediately taken and the upper circle of light was 

 found to be the brighter of the two, as was the case with the 

 dense mercury vapour. The diffuse reflexion begins to 



