PHENOMENA TYPICALLY SHOWN BY ACTINOCYCLUS RALFSII. 23 



could, I doubt not, be readily worked out for each case, though 

 I will not weary you with this now.* 



There is only one other fact which I should like to draw your 

 attention to, and, although it does not immediately bear on colour 

 phenomena, it is a rather interesting matter, and I do not think 

 it has been recorded. 



The ways referred to in which interference effects are produced 

 do not exhaust the number of possible ways in which an extinction 

 of light may occur. -Figs. 9 and 10 show another way. Here 

 both the rays travel through the silex, and meet in the point P — the 

 one ray having been reflected from the side of the wall at s. As 

 is easily seen, both have travelled just the same length of path, but 

 here comes the point to be noted. Whenever a wave is reflected 

 it is well known that half a wave-length retardation occurs, and 

 therefore when they meet at p the light is extinguished. Now 

 this will occur irrespective of the colours of the light, and that 

 seems to be the cause why a circular hole in a diatom is always 

 surrounded by a little dark ring. The rings are easily seen on 

 photographs of A. RaJfsii. The thickness of the little ring will 

 depend on the thickness of the diatom and on the relative 



* The reason why the phenomenon does not occur more frequently is 

 even a simpler matter than appeared at the time the paper was read. 



Calculations which I have since made, and can here only summarise, 

 show that for diatoms mounted in any specific medium, a minimum and 

 a maximum limit can be assigned for the production of any appreciable 

 colour effect. A great many diatom valves are of a thickness below the 

 necessary limit, or only slightly exceed it, and assume a pale yellow-brown 

 tint (that being the first colour in the series) as, for instance, Plcurosigma 

 angulatnm, mounted dry. The limits vary with the difference in refractive 

 index of the diatom silex and the medium in which it lies. The greater 

 the difference, the smaller is the minimum limit for the occurrence of 

 colour. 



Taking diatom silex to have a refractive index of 1"43, as stated in 

 Carpenter, the minimum limits would be roughly : — 



In Air 2i --millionths of an inch. 



,, Monobromide of Naphthaline 44 ,, „ ,, 



,, Styrax ..... 66 ,, mm 



„ Canada Balsam . . .111 ,, „ „ 



But if the refractive index of the diatom silex varies, so must these 

 limits change, and I have found, by having, A. Ralfsii mounted in a 

 medium of just 1-43, that it does vary considerably from this refractive 

 index. Many other diatoms varied also from this, but none so markedly 

 as A. Ralfsii. This therefore will also help to explain, on the theory 

 indicated, why the interference effects are more accentuated than usual. 



