COLOURS IN METAL GLASSES AND IN METALLIC FILMS. 401 



many metal particles to a wave-length is satisfied. We have shown that when the 

 metal is gold such glasses should be pink (cf. column 3 of Table II.) by transmitted 

 light ; and that the small gold spheres should send up the microscope light whicli is 

 pre-eminently yellow or green (cf. columns 4 and 5) ; and we have remarked that for 

 the same reason that explains the polarisation of sky light, such small spheres send 

 no light directly up the microscope tube when the electric vector of the incident light 

 is in that direction, so that in this case the cone of light as examined with the low- 

 power objective will be cut off (cf. column 5), although the large numerical aperture 

 of the Zeiss -j^-th oil immersion lens will allow some light to go up the tube, but so 

 as to leave a black spot in the centre of the focal plane of the microscope as shown 

 in fig. 6. 



All these deductions from our analysis are confirmed in every detail by the three 

 glasses F, G, H (Table II.). And it is these very glasses, of all the glasses in that 

 table, for which, according to the numbers there given, the particles are both smallest 

 and closest together. 



G. Let us now briefly notice the remaining glasses of Table 1 1. For these glasses 

 the number of metal particles to a wave-length, measured by (gold content) -=- si/.e 

 of particle, as determined from the Gth and 8th or from the 7th and 9th columns of 

 that table, is smaller than for the glasses F, G, H, which show the regular pink 

 colour. For the glasses A to E this number is greatest for the glasses Cc and E, of 

 which the former and parts of the latter do show the regular pink colour. 



Even glasses which do not satisfy the condition of many particles to a wave-length, 

 and which consequently do not exhibit the "regular" (pink) colour of gold glass, 

 have many of their properties co-ordinated by the results we have obtained for 

 regular glasses. 



Take, for instance, the glasses A and B (Table II.). Comparison of the gold 

 content /x with the size of the observed particles shows that tlio.se particles at any 

 rate are so far apart as not to satisfy our condition. The fact that glass A is 

 colourless shows that if there are also minute spheres present which escaped 

 observation, they also lie so far apart as not to be many to a wave-length. On the 

 other hand the pink colour of glass B suggests the presence of minute unobserved 

 spheres which are sufficiently close together to satisfy our condition, the absorption 

 of the glass being proportional to that small part of the gold content (/i) which is 

 associated with the minute spheres. 



In both glasses the large particles reflect much more light than is emitted by the 

 minute spheres. The colour of this reflected light is the usual yellow-red metallic 

 reflection from gold. Therefore the colour of the cone of light should be gold- 

 yellow (i). 



When the Nicol is introduced parallel to the plane of incidence, presumably half 

 the incident light is cut off. Consequently . the large particles send only half the 

 yellow-red light up the tube that they previously sent. Owing, however, to the fact 



VOL. com. -A. 3 F 



