118 Dr. W. Watson. On the Sensibility of the [June 12, 



change in the colours of the spectrum observed by Brewster, and which led 

 him to suppose erroneously that there were three kinds of solar light. 



In conclusion, when special means are taken to have as pure a spectrum 

 as possible, I can find no method which will enable me to distinguish as 

 distinct colours the wave-lengths in a monochromatic region. I therefore 

 regard the appearance of the monochromatic region as a fundamental physio- 

 logical fact, as I stated over 20 years ago. 



Note on the Sensibility of the Eye to Variations of Wave-length. 

 By W. Watson, D.Sc, F.R.S. 

 (Received June 12, — Read June 29, 1911.) 



In a recent communication to the Royal Society, Dr. Edridge-Green has 

 suggested that the reason Lord Rayleigh found he was able to distinguish a 

 difference in hue between two monochromatic patches of yellow (D) light, 

 when they differ in wave-length by about the distance between the sodium 

 lines (06 /xfj), is that (a) the spectrum used was not pure, and hence the 

 patches were not monochromatic ; and (b) that the difference in wave-length 

 was apparent because of admixture with white light. Some experiments 

 made by the author seem so conclusively to show that at any rate the second 

 of the above reasons cannot be correct that it seems worth while to put them 

 on record. 



By means of Sir William Abney's double spectrum apparatus,* two 

 patches of monochromatic light were thrown side by side on a magnesium 

 carbonate screen, and matters were so arranged that no line of separation 

 was observable when the patches were of the same colour. Each patch was 

 9 mm. by 18 mm., and the observer was at a distance of 60 cm. The 

 intensity of the illumination on the screen was throughout 3'5 candle-metres. 

 The slit in the second spectrum apparatus was kept at a fixed point in the 

 spectrum, while that in the first spectrum was moved by means of a micro- 

 meter screw, the movement being read on a scale on which a millimetre 

 represents in the yellow a difference in wave-length of 3 '7 fi/M. 



By cutting off the light from one slit and placing a short focus lens in 

 front of the other an enlarged image of the slit would be formed on the 

 screen. Thus by watching this image and gradually opening the slit the 

 width of the Edridge-Green monochromatic patch could be determined. 

 * ' Phil. Trans.,' A, 1905, vol. 205, p. 333. 



