( 157) 



VI. — On the Colours of Thin Plates. By Lord Rayleigh. (Plate X.) 



(Received 13th July 1886. Revised Aug. 1886.) 



Introduction. 



The first impression upon the mind of the reader of the above title will 

 probably be, that the subject has long since been exhausted. The explanation 

 of these colours, as due to interference, was one of the first triumphs of the 

 wave-theory of light ; and what Young left undone was completed by Poisson, 

 Fresnel, Arago, and Stokes. And yet it would be hardly an exaggeration 

 to say that the colours of thin plates have never been explained at all. The 

 theory set forth so completely in our treatises tells us indeed how the com- 

 position of the light reflected depends upon the thickness of the plate, but what 

 will be its colour cannot, in most cases, be foretold without information of an 

 entirely different kind, dealing with the chromatic relations of the spectral 

 colours themselves. This part of the subject belongs to Physiological Optics, 

 as depending upon the special properties of the eye. The first attempt to deal 

 with it is due to Newton, who invented the chromatic diagram, but his represen- 

 tation of the spectrum is arbitrary, and but a rough approximation to the 

 truth. It is to Maxwell that we owe the first systematic examination of the 

 chromatic relations of the spectrum, and his results give the means of predicting 

 the colour of any mixed light of known composition. Almost from the time 

 of first reading Maxwell's splendid memoir, I have had the wish to undertake 

 the task of calculating from his data the entire series of colours of thin plates, 

 and of exhibiting them on Newton's diagram. The results are here presented, 

 and it is hoped may interest many who feel the fascination of the subject, and 

 will be pleased to see a more complete theory of this celebrated series of 

 colours. 



The diagram (Plate X.) explains many things already known from observa- 

 tion, such as the poverty of the blue of the first order and of the green of the 

 second order. For good blues we must look to the second and third orders, and 

 for good greens to the third and fourth. The point in which the diagram dis- 

 agrees most with descriptions by former observers, e.g., Herschel, relates to the 

 precedence of the reds of the first and second orders. The first red has usually 

 been considered inferior, but the reason appears to lie in its feeble luminosity, 

 and consequent liability to suffer from contamination of white light. This and 

 other questions are further discussed in the sequel. 



VOL. XXXIII. PART I. X 



