Evening Glow and analogous Phenomena. 277 



millim. (that is, equal to the wave-length of yellow sodium-light 

 — line D in the solar spectrum), the minimum for this colour 

 will be at the extreme edge of the surface of the image ; there 

 will be no minimum for red light; for violet, however (line H), 

 a minimum will occur with a diffraction-angle of 42° 18'. Inas- 

 much, then, as the more refrangible rays approach their respec- 

 tive minima more rapidly than the less refrangible, the latter 

 will predominate the more in diffracted light the further we are 

 from the middle of the image — that is, from the place where 

 the image of the source of light appears. In the above-men- 

 tioned example the ratio of the intensities of violet light (Fraun- 

 hofer's line H, wave-length = 0-0003963) and of red (line B, 

 wave-length =0*0006897 mil.), for the angles of diffraction 30', 

 1°, 5°,10°, would be expressed by the numbers 0*9997, 0-9987, 

 0*9635, 0*8597*. These numbers show that the action is very 

 feeble, especially for the diffracted rays nearest the direct ones ; 

 we shall, however, subsequently become acquainted with circum- 

 stances by which it is increased. We may, however, maintain 

 that a white luminous point viewed through a very narrow aperture 

 appears white indeed, but surrounded by an aureole of diffracted 

 light which shows a reddish tint, though feeble and perhaps almost 

 imperceptible. 



2. If in a dark screen several equal apertures be arranged in 

 irregular order, each of them will furnish in each direction of 

 diffraction a resultant ray ; all resultants of the same direction 

 are equal to each other as regards intensity and mixture of 

 colour; the phenomenon of interference which they produce 

 upon the image will be occasioned solely by the differences of 

 path which their inclination to the direct rays imparts to them. 

 Owing to the irregularities in the arrangement of the apertures, 

 the resultants will not, however, be able to neutralize one another 

 in any direction of diffraction ; and just as little will there be any 

 direction in which they completely strengthen each other ; there 

 is no ground for supposing that in any one direction the degree 

 of their concordance will be greater or less than in another. The 

 phenomenon of diffraction which a screen perforated several times 

 produces is therefore essentially the same as that which one of its 

 apertures would produce, though the phenomenon occurs with an 

 intensity increased proportionally to the square of the number of 

 apertures f. 



* These numbers are calculated from the well-known expression 



(sin7r6sin^A. _1 \ a 

 7r&sin\//'\ -1 ' 

 where b denotes the Breadth of the slit. 



f The theory of sun and moon halos is, as is well known, founded on 

 this reasoning. 



