191 ] Tracy: White Markings in Birds. 303 



glottos), the magpie (Pica hudsonia), and the shrike (Lanius 

 borealis). These are interesting cases, but little can be inferred 

 from them as to the significance of white markings in the group 

 at large. 



It is in the correlation of special color features with special 

 feeding and breeding ranges that we get the first clear indication 

 of a large underlying principle determining which birds shall 

 possess the white pattern and which shall not. The top-marked 

 finches are seen to be birds of open woods mainly, largely of 

 roving disposition and wide feeding range. Unmarked species 

 of the family are mainly birds of low zones and narrower feeding 

 beat. Only two of the first list are characteristically low rangers 

 and given to covert-seeking: these are our eastern and western 

 white-marked towhees (Pipilo erythrophthalmus and P. macu- 

 latus). On the other hand nearly all of the second list are either 

 confined to close foliage of medium height, or belong to such 

 associations as the rank growth of humid regions, the dark 

 borders of shady swamps, or thickets of the chaparral belt. They 

 are the thrushes, painted in monochrome above ; warblers, 

 lacking top-patterns ; wrens with their finely barred color 

 scheme ; flycatchers with dull olivaceous or other uniform shad- 

 ings. These are correspondences that surely have significance, 

 and require for their interpretation something more than the 

 older theories of coloration could offer. 



Finding such good illustrations of the disruptive effect of 

 white or bright patterns among animals and birds, the authors 

 of "Concealing Coloration in the Animal Kingdom" (Gerald 

 and A. H. Thayer, 1909), have come to believe that no other 

 explanation is needed to account for the presence of top-white in 

 birds that show themselves against the sky, than natural selection 

 working through this means. Their belief accords with the 

 conditions just cited. It does not, however, take account of the 

 fact already mentioned that markings often become conspicuous 

 during the flight of the bird, nor does it take note of the corre- 

 lation that has been shown to exist among open-ground birds, 

 of flight-revealed markings with the flocking habit a condition 

 which we shall also find largely present among the arboreal birds. 

 In the use of these lists of birds to determine the latter point, 



