Chap. XV.] COLOR AND NIDIFICATION. 165 



Even in the groups in which, according to Mr. Wal- 

 lace, the females, from being protected during nidification, 

 have not had their bright colors eliminated through natu- 

 ral selection, the males often differ in a slight, and occa- 

 sionally in a considerable degree, from the females. This 

 is a significant fact, for such differences in color must be 

 accounted for on the principle of some of the variations 

 in the males having been from the first limited in their 

 transmission to the same sex ; as it can hardly be main- 

 tained that these differences, especially when very slight, 

 serve as a protection to the female. Thus all the species 

 in the splendid group of the Trogons build in holes ; and 

 Mr. Gould givesfigures 20 of both sexes of twenty-five spe- 

 cies, in all of which, with one partial exception, the sexes 

 differ sometimes slightly, sometimes conspicuously in col- 

 or — the males being always more be'autiful than the 

 females, though the latter are likewise beautiful. All the 

 species of kingfisher build in holes, and with most of the 

 species the sexes are equally brilliant, and thus far Mr. 

 Wallace's rule holds good ; but in some of the Australian 

 species the colors of the females are rather less vivid than 

 those of the male ; and, in one splendidly-colored species, 

 the sexes differ so much that they were at first thought to 

 be specifically distinct. 21 Mr. R. B. Sharpe, who has es- 

 pecially studied this group, has shown me some American 

 species (Ceryle) in which the breast of the male is belted 

 with black. Again, in Carcineutes, the difference between 

 the sexes is conspicuous : in the male the upper surface is 

 dull-blue banded with black, the lower surface being part- 

 ly fawn-colored, and there is much red about the head ; in 

 the female the upper surface is reddish-brown banded 

 with black, and the lower surface white with black 



20 See his ' Monograph of the Trogonidae,' first edition. 

 51 Namely Cyanalcyon. Gould's ' Hand-book of the Birds of Austra. 

 Ha,' vol L p. 133; see, also, pp. 130, 136. 



