20 



Bulletin oi* nut 



range extended beyond its intluence." 1 see no reason why the class of 

 birds just mentioned should have been so restricted. Kirtlundi may well 

 be taken as a representative of the class of birds, just referred to. It is a 

 bird which breeds in the northern coniferous /.one. In general, it seems safe 

 to infer that those migratory birds, which now have their breeding grounds 

 in the north^ were among the early migrants which pushed back with the 

 retreat of the ice. Can it be that the evident northern breeding area of 

 Kirtlandi indicates that it was one of these carlv invaders? And can its 



Fig. 3. Lines of Glacial Drainage or Shore Lines. To show rela= 

 tion of those topographic features to bird migration routes. 



Border of the last ice sheet v Wisconsin), indicated by 

 dashes. 



apparent adherence to glacial highways indicate that its dispersal dates back 

 to early post-glacial times? Habits of migration may preserve records when 

 fossils are lacking. If tliis was one of the early species to push north, 

 it is but natin-al that it should follow such highways, as it is along such 

 valleys and shore lines, at that time, that the vegetation would make its 

 most rapid extension northward. A comparison of maps (Figures 1 and ?>) 

 is suggestive in this connection. The ma]) of the Prothonotary Warbler 

 (Figure 1) shows the present distribution of a distinctly southern type 

 of bird, which is extending its range northward, and suggests the method 

 and stage which Kirllandi has long ago passed through as it extended its 

 breeding range northward with the amelioration of the glacial climate. These 

 valleys and shores, (Figure .3) as has been mentioned, were in early post- 

 glacial times, as to-day, highways of dispersal. The breeding range of 



