II] OF LIGHT AND COLOUR 55 



blue of a blue jay, an Indian roller or a macaw; to the other belong 

 the iridescent hues of mother-of-pearl, of the humming-bird, the 

 peacock and the dove: for the dove's grey breast shews many 

 colours yet contains but one — colores inesse plures nee esse plus uno, 

 as Cicero said. The jay's blue feather shews a layer of enamel-like 

 cells beneath a thin horny cuticle, and the cell-walls are spongy 

 with innumerable tiny air-filled pores. These are about 0-3 /it in 

 diameter, in some birds even a little less, and so are not far from 

 the hmits of microscopic vision. A deeper layer carries dark-brown 

 pigment, but there is no blue pigment at all; if the feather be dipped 

 in a fluid of refractive index equal to its own, the blue utterly 

 disappears, to reappear when the feather dries. This blue is like 

 the colour of the sky; it is ''Tyndall's blue," such as is displayed 

 by turbid media, cloudy with dust-motes or tiny bubbles of a size 

 comparable to the wave-lengths of the blue end of the spectrum. 

 The longer waves of red or yellow pass through, the shorter violet 

 rays are reflected or scattered; the intensity of the blue depends 

 on the size and concentration of the particles, while the dark pigment- 

 screen enhances the effect. 



Rainbow hues are more subtle and more complicated ; but in the 

 peacock and the humming-bird we know for certain* that the 

 colours are those of Newton's rings, and are produced by thin plates 

 or films covering the barbules of the feather. The colours are such 

 as are shewn by films about J [x thick, more or less ; they change 

 towards the blue end of the spectrum as thei hght falls more and 

 more obliquely; or towards the red end if you soak the feather 

 and cause the thin plates to swell. The barbules of the peacock's 

 feather are broad and flat, smooth and shiny, and their cuticular 

 layer sphts into three very thin transparent films, hardly more than 

 1 jjL thick, all three together. The gorgeous tints of the humming- 

 birds have had their places in Newton's scale defined, and the 

 changes which they exhibit at varying incidence have been predicted 



* Rayleigh, Phil. Mag. (6), xxxvii, p. 98, 1919. For a review of the whole 

 subject, and a discussion of its many difficulties, see H. Onslow, On a periodic 

 structure in many insect scales, etc., Phil. Trans. (B), ccxi, pp. 1-74, 1921; 

 also C. W. Mason, Journ. Physic. Chemistry, xxvii, xxx, xxxi, 1923-25-27; 

 F. Suffert, Zeitschr. f. Morph. u. Oekol. d. Tiere, i, pp. 171-306, 1924 (scales of 

 butterflies); also B. Reusch and Th. Elsasser in Journ. f. Ornithologie, lxxiti, 

 1925; etc. 



