Feathers rr 



are almost always due to j)igmeiit or colouriii<r-mattor in 

 the shaft or vane. If we take a black feallier and liold 

 it to the light, it will still look black; if we ])()und it witli 

 a hammer, it will not change. 



Green is never found as a pigment except in ilie 

 feathers of a small family of birds called plantain-eaters 

 or turacous, which inhabit West Africa. For some time 

 it was thought that the natives dyed the birds artificiallv, 

 as when tliese birds were kept captive, the magnificent 

 scarlet patcli on the wing would graduall}- fade and 

 become a dull grav. It is a fact that this (-(jlouriuir- 

 matter washes out wlien the feather is washed in alkahne 

 water. Even or'dinary water will be shghtly tinged if the 

 feather is soaked in it. The pigment contains al)out t(Mi 

 per cent of copper, and this can be extracted chemically 

 in the form of a metallic powder. The ])lumage of almost 

 all brightly coloiu'ed birds will fade in the cotirse of years, 

 if the feathers are left exposed to direct sunlight ; but, like 

 photographic ])lates, the hues of some birds are more sen- 

 sitive than others to the light. The delicate i-eds and 

 yellows on tlie lower parts of Mexican Trogons are j)ar- 

 ticularly evanescent, and the rose-pink of the African 

 Fairy Warbler disaj:)pears a short time after death. 



We might speak of a third class of colours, which are 

 due to both pigment and structure. For instance, no 

 blue pigment is known to exist in the feathers of birds, 

 but blue feathers contain a l^rown or yellowish pignienl 

 which is encased in the horny coating of llie feather. 

 Between this outer sheath and tlie luiderlying pigment 

 is a layer of many-sided cones or small })rojections which 



