656 SCIENCE PROGRESS 



cent, of the incident light if there are no more than thirty-two alternations 

 of crystal directions. But the problem is of greater importance at the 

 present time, because it gives a good illustration of the general effect of 

 multiple reflexions in connection with X-rays and crystal structure. It 

 must be admitted that there are considerable differences between the two 

 phenomena. Whereas in the case of chlorate-of-potash crystals and ordinary 

 light a ray reflected from a lower layer is bound to be caught by an upper 

 layer, yet it is otherwise with X-rays. Owing to the shortness of their 

 wave-length, there is much more chance of a wave passing through the inter- 

 spaces between the atoms without being caught. Nevertheless, a careful 

 perusal of Rayleigh's problem is exceedingly instructive in connection with 

 applications to X-rays. 



At various times photographic questions attracted Lord Rayleigh. In 

 No. 359 he takes up the general problem of photographic reproduction, with 

 suggestions for enhancing gradation originally invisible. The original object 

 is itself taken to be a transparency, the fraction of light transmitted at a 

 given point being t {i.e. the transparency). Similarly, the transparency of 

 the negative is t', and since this transparency depends upon the light that 

 had acted upon the negative we may write t' = J{t). When the operation 

 is repeated, using the negative as " source," and a positive is taken, the 

 transparency of the last t" is given by 



Complete photographic reproduction may be considered to demand that at 

 every point t" = t. This requires that if the relation between t and t' be 

 written F {f,t') = o, then F must be a symmetrical function of t and t'. 

 So far all is beyond question ; at any rate, if the negative and positive are 

 on the same kind of plate and are similarly developed. In the particular 

 examples Rayleigh does not hit on cases that approach sufficiently near to 

 practical conditions to be of use. The question has recently been thrashed 

 out in greater detail, with a close approach to practical applications. 



In Article No. 381 Rayleigh takes up the case of the diffraction of light 

 by (dielectric) spheres of relative refractive index differing little from unity. 

 The object of the fresh attack was to extend the calculations to particles 

 larger, compared with the wave-length, than had previously been done. 

 Unfortunately, they are limited, for simplicity, to cases for which the dielectric 

 constant of the particles differs little from that of the surrounding medium. 

 The present writer had previously shown experimentally that if bright light 

 is examined through a sulphur suspension set free from a " hypo " solution 

 by adding acid, the light becomes redder and weaker in the early stages, 

 (forming an artificial setting sun), but later on, if observations are continued, 

 the transmitted light grows again in intensity and is now deep blue, gradually 

 changing to green, greenish- white, and white. No explanation was forth- 

 coming from Rayleigh's calculations ; nor does it seem likely that an increase 

 in the value of the assumed index would materially alter the character of 

 the results. Calculations which are now being made by Captain Talbot 

 Paris (and which are not yet published) indicate that the explanation is to 

 be found in another direction. 



In No. 427 the remarkable colours, variable with the angle of observation, 

 which are so frequent in beetles, butterflies, and feathers are stated to be 

 due probably to a periodic structure in the diffracting particles, so that the 

 dielectric constant has a different value at different distances from their 

 centres. Reference may also be made to No, 438 on the optical character of 

 some brilliant animal colours. 



Of the papers indexed under the head of Miscellaneous, the one to which 

 this title fits best is Rayleigh's Presidential Address to the Society of 

 Psychical Research in 1919 (No. 443). The , following extracts illustrate 



