412 DR. H. GADOW ON THE COLOUR OF FEATHERS. [May 2, 
can actually be done. If we press one of the deep-blue feathers 
of a Maccaw between two hard planes, so as to squash or smash the 
stratum of prismatic cones, or if we hammer it carefully, the blue 
immediately disappears, and the injured part looks grey or brownish 
according to the underlying pigment. The same is the case with 
the beautifully blue feathers of Artamia. Green parrot-feathers, when 
treated in a similar way, become yellow, since this is the colour of 
their pigment. Thus structural or optical colour may, so to speak, 
be knocked out of a feather. (Fatio observed that blue disappears 
after injuring the surface by scratching off some of the enamel.) 
This explains the dark appearance of the abraded parts of feathers 
of Parrots and other vividly coloured birds. Again, red, orange, 
brown, black, and most of the yellow feathers (7. e. such which owe 
their colour directly to pigment) do not lose or change their colour 
under any physical treatment. 
The explanation of the due colour is the most difficult of all in 
those feathers where the blue is independent of the position of the 
eye, 7. e. in which the blue does not change. In most cases the blue 
is confined to the rami, which, for imstance in Cereba and in 
Artamia, in the blue parts of the feather are devoid of cilia and radii, 
and are broader and flattened out (cf. Fatio). 
With a magnifying-power of about 640, we first observe that the 
whole ramus is covered by a transparent, slightly yellowish, or per- 
haps quite colourless, sheath or coating, the thickness of which is not 
more than 0°0014 of a millimetre. The surface of this sheath is 
uneven and granulated. Immediately under this sheath we find one 
continuous layer of prismatic polygonal (frequently hexagonal) cells 
or cones. Most of these cones are broadest at their apices, and be- 
come smaller towards their bases; others have nearly parallel walls 
or may be broadest below. (This layer of cones has been called 
by Fatio, its discoverer, ‘‘émail.’”) The space between their apices 
seems to be filled up with the same matter as the coating. The 
colour of the cones is pale yellowish, or, if this is only the reflection 
of the underlying pigment, they are colourless. The distance 
between the middle of two neighbouring apices I found equal to 
0:0050 of a millimetre ; this would also be their breadth at the 
base. Their height seems to be slightly larger. No actual mea- 
surement, however, could be obtained, as I did not succeed in 
getting a clear side view of them. As to the structure of these 
little cones themselves, it is very difficult to arrive at a satisfactory 
conclusion, considering the minuteness of the subject. However, in 
Pitta moluccensis and in Artamia I observed a system of extremely 
fine lines running parallel with the long axis of the cones, i. e. trans- 
verse or vertical to the long axis or surface of the ramus. These 
lines themselves do not seem to be straight, but irregularly waved. 
The breadth of each bar I calculated to be less than 0°0006 of a 
millimetre. 
Below this stratum of polygonal prisms or cones lies brownish- 
yellow pigment, near the middle of the barb; where the layer of 
pigment is thicker it looks black-brown. This pigment, of course, 
