ON COLOUES IN METAL GLASSES, ETC. 277 



11. Colour Changes Caused by Heating Metal Films. 



In the Bakerian Lecture* for 1857, FARADAY described a, number of experiments 

 concerning the colours which gold and other metals were, in various conditions, 

 capable of exhibiting. 



Mr. G. T. BEILBY'S investigations on the colour and structure of films of metal are 

 described in his paper on " The Surface Structure of Solids."t 



The average thickness of the gold leaf which FARADAY used in his earlier 

 expei'iments was about 90 /i/i.| Reference to p. 271 above will show that, with a 

 probable error of 2 per cent., the optical properties of such a leaf will be subject to 

 the analysis given in 10 above. 



Thus, for example, if we assume that, in a gold leaf as it leaves the beater, the 

 gold is in an amorphous state, its colour by directly transmitted light is that for 

 which T , as given by equation (23), namely 



T = M u exp {-4ml. riK/X} (23), 



is a maximum. If, further, the metal has its normal specific gravity, so that /j. = 1, 

 the values of nx/K and of M in this equation are those given for p, = 1 in Tables VIII. 

 and IX., or figs. 8 and 10. 



Now, when /JL = 1, the value of n/c/A. is much smaller for blue and green than for 

 yellow and red, and is slightly smaller for blue than for green ; while the value of M 

 is greater for green and blue than for yellow and red, and is considerably greater for 

 green than for blue. Thus both the factors of T in equation (23) are greater for 

 blue and green than for yellow and red. The former colours therefore predominate 

 in the transmitted light. Further, in very thick films, for which n/c/A. is of supreme 

 importance, blue should, in the transmitted beam, predominate over green ; while in 

 thinner films, on account of the greater value of M u for green than for blue, green 

 light should be more intense. 



FARADAY found that all his gold leaf appeared olive-green by transmitted light. 



Again, Table IX., or fig. 12, shows that the reflecting power, R , is, when p = 1, 

 much greater for red and yellow than for green and blue ; and this result is again in 

 accordance with the observed colour of gold leaf by reflected light. 



FARADAY, however, states that gold leaf still appeared green by transmitted 

 light when its thickness was reduced to only 10 JU./A or 5 pp,. Now equation (24) of 

 the former communication, || namely, 



T = l-47rd.n 2 K /\, 



* Reprinted from the 'Phil. Trans.' in his 'Researches in Chemistry and Physics,' p. 391. References 

 will be made to the pages of this reprint." 

 t Glasgow, 1903. 

 t Loc. tit., p. 394. 

 Loc. tit., p. 395. 

 || 'Phil. Trans.,' A, 1904, p. 408. 



