176 



Cliamplaiu period. The more difficult problem is the explanation of their 

 persistence. 



A consideration of a few other species will serve to emphasize the 

 point in mind. 



The Larch or Tamarack (Larix laricin<i (DuRoi) Koch) is found in 

 Porter, Marshall. Kosciusko. Noble. ►Steuben. DeKalb and Blackford. An 

 examination of the older shore lines of Lake Michigan gives a sufficient 

 explanation. Even the Blackford County citation, which seems well to the 

 south, is made clear when the ancient bay of Lake Michigan extending 

 southward through Allen County in the neighborhood of Fort Wayne is 

 recalled. 



Thus also the eastern Peachdeaved Willow (Salix amyffdaloides An- 

 derson) found in Lake and Kosciusko finds ready interpretation, as does 

 also the case of the Wild Red Cherry (Prwrws Permsylvanica Linnaeus fils) 

 occurring in Lake, Porter and Kosciusko counties. 



Any one who maps some of the more widely ranging species of the 

 State will be immediately impressed by the close relations existing be- 

 tween the distribution of the species and the course of waterways. In 

 soine instances the distribution follows a single waterway, in others it 

 seems to follow not merely the main stream but also all of the tributaries. 

 Indeed, by far the most striking feature in the series of one hundred 

 twenty-six maps is the definite way in winch this relationship stands out. 

 The most cursory inspection of the maps reveals it and serves to suggest at 

 least a possible causal relation. 



In the opinion of the writer the occurrence of given species in widely 

 separated localities without intervening stations will be found to be due to 

 the existence at some time in the past of practically continuous waterways 

 connecting these now separated localities. Further, that such connections, 

 in so far as the region under consideration is concerned, are mainly to be 

 found in the Champlain period, althougli perhaps in some cases this con- 

 nection was furnished by the northward stretching arm of the Gulf of 

 Mexico. If the shore lines of streams and lakes of the Champlaiu period 

 could be drawn upon our present maps many of our problems in Phytogeog- 

 raphy would solve themselves. In confirmation of this view is the dominat- 

 ing influence of continuous waterways or of streams in the distribution of 

 species clearly shown by any careful study of present range extensions. 



In the main, widely ranging species, at least among trees, do not have 



