430 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1911. 
Where the scales are transparent, white light is transmitted, but 
with feathers, where the color film is generally backed by dark pig- 
ment, the pigment color appears untinged with the complementary 
to the color naturally reflected. 
With many feathers the color returns when the pressure is taken 
off, but with insect scales the structure seems to be permanently 
injured by compression, and though when allowed to expand again 
the material is not colorless the brilliancy which belonged to the 
uninjured scale is gone, and the color in general changed. 
The facts above mentioned seem to offer stronger reasons in favor 
of interference than the polarization phenomena referred to by 
Michelson and Walter! do against it. 
The ellipticity, etc., found in the reflected beams may, although 
functions of the wave length, accompany the production of color 
without being necessary to it—that is, they may depend on the 
molecular, while the colors depend on the mechanical structure. 
All Lepidopterous scales, whether colored by pigment or giving 
metallic reflection, are traversed by a series of fine lines or dots 
arranged in lines and very evenly spaced, and the universality of 
these lines on all such scales, whether with or without color, is a 
good reason for not regarding diffraction as an explanation of the 
metallic colors. 
In many insects these lines are as close as 36,000 to 40,000 per inch, 
and when light is transmitted through a single scale or a few scales 
placed side by side very fine diffraction spectra are formed, but no 
corresponding effect is seen by reflection, what effect there may be 
being masked by the other form of color production. 
The beetle scales which I have examined were, as a rule, without 
linear markings, and where lines existed they were not very regular. _ 
The surface was always mapped out into unequal polygonal areas, 
and cross sections (pl. 2, figs. 5a, 5b) showed that the scale consisted of 
a flattened sac of transparent material containing a cellular structure 
in which the color originated. 
When an unbroken scale is immersed in cedar oil, the outer walls 
prevent the fluid reaching the color-producing layer and but little 
change results either in the reflected or transmitted ight; but when 
the scale is broken or has a piece cut off the oil penetrates the interior 
and all trace of color disappears. 
Occasionally when a viscous fluid is employed the penetration 
is not complete and the character of the cellular layer is then indi- 
cated by the parts which still show color. 
Figures 1 to 4, plates 1, 2, illustrate this. Figure 1 is an unbroken 
scale of Entimus imperialis showing the polygonal areas. Figure 2 
1 Polarized light should be used for this observation, 
