THE COLOUB OF THE KINGFISHER. 467 



to be described, hardly agrees with a prismatic solution. Many 

 animal colours, especially in molluscs and insects, are due to 

 thin plates; and probably many metallic bird colours are formed 

 in the same manner.* 



The microscopical appearance of the surfaces of the cell-caps, 

 their behaviour in various media, and the fact that no other 

 colour but blue is produced, leads us to dismiss the question of 

 thin plate or interference colours in the feathers of the King- 

 fisher ; but the greatest objection is to be mentioned later. 



We have thus disposed of every explanation except one, and 

 that is the theory of the production of blue by the reflection of 

 light from small particles, and of orange or red by the trans- 

 mission of light through small particles. I suppose everybody 

 has noticed the bright blue colour of the smoke from a wood 

 fire when seen against a dark background of woodland or moun- 

 tain ; seen against the bright sky, the smoke is no longer blue 

 but orange. It is the same with a mixture of milk and water, 

 where the minute globules of fat reflect a blue colour but trans- 

 mit orange — for the fluid is decidedly yellow when held to the 

 light. The blue of the human eye is said to be due to the effect 

 of light on the particles in the substance of the iris ; and it is 

 well known to physicists that blue may be produced by con- 

 densing various gases, or more simply by dropping an alcoholic 

 solution of mastic info a vessel of pure water. Yet we get the 

 phenomenon on the grandest scale almost daily in the blue of 

 the sky. By reflected light the colour is pure blue, but as the 

 sun sinks to the horizon we see the atmosphere by transmitted 

 light, and it is then orange. This is hardly the place to describe 

 matters that are treated at length in every elementary book on 

 light, but a couple of lines or so may be given. In white light 

 the red waves are twice as big as the blue. When light impinges 

 on a collection of tiny points a great deal is reflected intact as 

 white light, but many of the red waves get broken up into 



* I know this is not the usually accepted explanation of the metallic 

 tints of feathers. If we take a blue feather from the neck of a Peacock, and 

 cautiously wipe it with a solution of caustic potash, with the effect of 

 thinning the outer layers of ceratin, the range of green and blue tints gives 

 way to one of bronze-browns and reds. I cannot see how this fits in with 

 the prismatic explanation; and there are other serious objections. 



