THE COWBIKDS. 613 



1 aiay lueutiou tliat twice 1 have seen birds of tliis species attempting 

 to build nests, and that on both occasions they failed to complete the 

 work. So universal is the nest-making instinct that one might safely 

 say tlie M. honariensis had once possessed it, and that in the cases I 

 have mentioned it was a recurrence, too weak to be eliicient, to the 

 ancestral habit. Another interesting circumstance may be adduced as 

 strong presumptive evidence that M. honariensis once made itself an 

 open exposed nest as M. hadius occasionally does, viz, the difference 

 in color of the male and female, for while the former is rich purple 

 the latter possesses an adaptive resemblance in color to nests and to 

 the shaded interior twigs and branches on which nests are usually 

 built. How could such an instinct have been lost? To say that the 

 Cowbird occasionally dropped au egg in another bird's nest, and that 

 the young hatched from these occasional eggs possessed some (hypo- 

 thetical) advantage over those hatched in the usual way, aud that the 

 parasitical habit so became hereditary, sui)planting the original one, is 

 an assertion without anything to support it, and seems to exclude the 

 agency of external conditions. Again, the want of correspondence in 

 the habits of the young parasite and its foster parents would in reality 

 be a disadvantage to the former; the unfitness would be as great in 

 the eggs and other circumstances, for all the advantages the parasite 

 actually possesses in the comparative hardness of the eggshell, rapid 

 evolution of the young, etc., already mentioned, must have been 

 acquired little by little through the slowly accumulating process of 

 natural selection, but subsequently to the formation of the original 

 parasitical inclination and habit, I am inclined to believe that M, 

 honariensis lost the nest-making instinct by acquiring that semipara- 

 sitical habit common to so many South American birds of breeding in 

 the large covered nests of the Dendrocolaptidw. We have evidence 

 that this semiparasitical liabit does tend to eradicate the nest-making 

 one. The SynaJlaxes build great elaborate domed nests, yet we have 

 one species [S. wgithaloides) that never builds for itself, but breeds in 

 the nests of other birds of the same genus. In some species the nest- 

 making habit is in a transitional state, Machetornis rixosa sometimes 

 makes an elaborate nest in the angle formed by twigs and the bough of 

 a tree, but prefers, and almost invariably makes choice of, the coverec; 

 nest of some other species or of a hole in the tree. It is precisely the 

 same with our Wren, Troglodytes furvus. The Yellow House Sparrow, 

 Sycalis lyehelnii, invariably breeds in a dark hole or covered nest. The 

 fact that these three species lay colored eggs, and the -first aud last 

 very darkly colored eggs, inclines one to belive that they once invari- 

 ably built exposed nests, as M. rixosa still occasionally does. It may 

 be- added that those species that lay colored eggs in dark places con- 

 struct and line their nests far more neatly than do the species that 

 breed in such places but lay white eggs. As with M. rixosa and the 

 Wren, so it is with the Bay- winged Molothrus; it lays mottled eggs, and 



