432 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1911. 



brilliancy by holding a piece of glass over the smoke of an arc formed 

 between metal electrodes, iron, nickel, silver, and copper being used. 

 In the most successful trials the rings were brighter than the colors 

 of soap films, and, as might be expected, the intensity of the color 

 increased with the angle of incidence, but the rings did not show 

 with normal reflection nor until the incidence exceeded 30°. 



In the spectroscopic examination of the color, it was found conven- 

 ient to focus the much enlarged image of the surface on the slit of 

 the spectroscope. By this means, and using the sun or an arc lamp, 

 there was sufficient light to show the spectrum from a small part 

 of a single scale. 



When cutting thin sections of chitin or feathers, it is important 

 that the embedding material should be of the same consistence and 

 hardness as the object to be cut. For this purpose shellac gave 

 the best results. The hardness could be regulated by the time 

 allowed for drying. 



The compression tests were carried out by placing the specimens 

 on a slide under a convex lens of about a foot radius. The objects 

 compressed were generally thin enough to allow of the Newton 

 rings of the higher orders being recognized between the lens and 

 slide before any compression occurred, and by centering the object 

 in these the pressure could be applied in the right place. 



Since writing the above I have examined the colors of some of 

 the iridescent Diptera (chiefly of the genus Lucilia), using the pres- 

 sure apparatus. It was found that with them, as with the scales 

 and feathers, the color disappeared under compression, and it seems 

 probable, therefore, that interference of one kind or another is the 

 true cause of natural iridescent color in all cases. It may be remarked 

 that the intensity and composition of the light reflected from the 

 integument of the flies is such as would be accounted for by the 

 interference of a single film or pair of surfaces. 



DESCRIPTION OF PLATES 



Fig. 1. — Scale of Entimus imperialis. X 490. 



2. — The same partly permeated with celluloid solution. X 490. 



3.— The same. X 1750. 



4. — Three scales of same completely permeated. X 490. 



5. — Cross section of scale. X 1750. 



6. — Iridescent scale of Ornithoptera Poseidon. X 1170. 



7. — Black scale of same. X 1170. 



