88 EXPERIMENTS WITH THE DISPLACEMENT INTERFEROMETER. 



which gives a rough reproduction of the facts, the aperture of a corona of 

 average wave-length when the particles approximate io~ 4 cm. must soon 

 exceed 90, so that the colors are spread out along the axis of a shallow cone 

 close to or within the fog-chamber. It would require a similarly divergent cone 

 of light to make these colors seem axial. Now, in producing coronas from a 

 single small source of light, the rays which strike the chamber are quite 

 inadequately divergent. 



Finally, by putting the electric lamp with an annular screen close to the 

 fog-chamber, thus avoiding all lenses, the phenomena were produced much 

 more brilliantly, but without new results. 







8 10 

 FIG. 49. 



16 18 



51. Coronas from a point source. The best evidence obtained bearing on 

 the nature of the disk colors is probably found in my own earlier experiments 

 with strictly homogeneous light. The green light of a mercury lamp is used 



