542 Annals of the Carnegie Museum. 



O. Bangs (Proc. Biol. Soc. Wash., XIX, 1906, 105) seems to me to 

 best represent the real state of affairs. M. Eugene Simon also says 

 that he agrees with this same arrangement, except that he recognizes 

 O. pectoralis as a good species, to which neither Mr. Bangs nor 

 I can agree. As Mr. Bangs says, O. leucaspis is antedated by O. 

 castaneiventris Gould, and O. calolcema is clearly only subspecifically 

 distinct from it, true castaneiventris being restricted to Chiriqui. 



As to O. pectoralis, I believe that the few specimens which have 

 been taken (certainly not more than four or five males, the females 

 not entering into the question at all, being indistinguishable from 

 females of O. calolcema) and called that, are nothing more than 

 peculiar cases of partial melanism in O. calolcema, or rather one of 

 those peculiar color aberrations which are so frequent in humming- 

 birds. Very few specimens have ever been taken. It occurs in 

 the same districts as 67. calolcema, and is so rare that such inde- 

 fatigable collectors as Zeledon, Cherrie, Alfaro and Underwood have 

 never taken it. In view of all these facts I do not think it can be 

 considered as a normal form. 



It is a well-known fact that in feathers having a structure similar to 

 that found in hummingbirds and other groups where brilliant metallic 

 tints are present, the color of the feather is not due to pigments, but 

 to physical structure, causing it to absorb all the colors of the spectrum 

 except that which it reflects, and that the slightest derangement of this 

 structure following certain lines will cause the feather to lose its bright 

 metallic sheen and appear dull black. A striking case in point is a 

 specimen in the collection of Mr. Bangs, that of a melanistic individ- 

 ual of Panterpe insignis. This bird still retains the " color pattern " 

 of the normal bird, but the brilliant, glowing metallic sheens which 

 are so much in evidence in that species have been entirely lost, and 

 the whole bird has a very dull, sooty-black appearance, slightly pur- 

 plish in some lights. This bird was taken among numerous normal 

 individuals, and its color can be attributed to no cause other than that 

 stated above. 



O. castaneiventris calolcema is found throughout the highlands of 

 central and northern Costa Rica, from about 3,000 feet upwards, but 

 does not extend into the southern part of the country, its place there 

 seeming to be taken by its near relative O. cinereicauda. It is a bird 

 of the forest, but also frequents open woodland, scattering trees in 

 pastures, etc., and I do not think it would be found in dense unbroken 

 forest far from any clearing as is O. cinereicauda. 



