466 THE ZOOLOGIST. 



examining from both sides. The blue is only seen by reflected 

 or incident light ; by transmitted light the cap is a rich orange- 

 brown, with no trace of blue. It is figured as the cell nucleus 

 by Fatio, and described as the pigment by Gadow. Collectively, 

 the cell-caps give the rich brown colour we see when we look 

 through a Kingfisher's feather (especially slightly diagonally) at 

 a strong light. The following striking experiment proves this 

 orange — and sometimes blue — cap to be neither a coloured 

 nucleus nor a pigmented cell- wall. 



When a single cap (or, better still, a connected series) is 

 mounted dry and examined by incident light, the blue colour 

 can only be described as a glare. If we allow a drop of a mixture 

 of xylol and Canada balsam to flow down the slide, the very 

 instant it reaches the cell-cap the brilliant light is snapped off 

 as suddenly and as completely as in the switching off of an 

 electric lamp ; and the most delicate adjustment of the micro- 

 scope fails to reveal any further trace of the fragment of feather. 

 By transmitted light, by which the cap resembles a flake of 

 orange shellac, the xylol and balsam act in causing an oblitera- 

 tion almost as complete. When the fluid reaches it, the rich 

 orange vanishes instantaneously, and all that remains is a thin 

 and only just visible plate of perfectly transparent ceratin. 

 Sometimes the disappearance is complete, and the cap is not 

 seen again ; but previous treatment with some such substance 

 as fuchsin enables it to be kept always in view. 



This experiment paves the way for a solution of the problem 

 of the blue colour in the Kingfisher and many other birds. It 

 is clear that the colour is not due to pigment, either directly (as 

 yellow is produced in a Canary) or indirectly, as Dr. Gadow has 

 said. The absence of strise — and above all the behaviour of the 

 colour under different arrangements of the light — enables us 

 to dismiss the suggestion of " Gitterfarben." Leaving out of 

 account polarization colours, which are impossible in feathers, 

 we have but three remaining explanations to consider. The first 

 is prismatic colour. If the grains on the surfaces of the cell-caps 

 are prisms, we should certainly see under the microscope at 

 least the sparks of some other colour besides blue — the tiny 

 gleams of yellowish white are from a known source that it is not 

 necessary to explain here. The case of the Eclectus, presently 



