394 MICHIGAN SURVEY, 1908. 



is a closer genetic relationship between forms along the same general 

 migration route than between those of verj' distinct routes and histories.. 



2. Post-Glacial Origin of the Fauna. — The geographic affinities of 

 the mammal fauna of Isle Eoyale have been shown to be with those of the 

 region north of Lake Superior, and representative of the coniferous 

 forest region of central and eastern Canada. There now remains to be 

 considered the approximate post-Glacial geographic origin ol this north- 

 ern fauna. But before this subject can be understood, special attention 

 should be directed to the fact that an extensive barrier in the form of 

 a series of Glacial and post-Glacial lakes and even the Champlain Sea 

 (cf. Taylor, '05, pp. 103, 106 and 107) stood between the advancing fauna 

 from the south and Canada. All these barriers were not contempor- 

 aneous, yet some of them, generally several extensive ones, have been 

 present since the decline of the Wisconsin ice sheet. This barrier was 

 onlj' interrupted, as far as many mammals have been concerned, by 

 narrow streams, such as, the Saint Clair, Detroit, Niagara and St. 

 Lawrence rivers. Even these must have retarded many forms, except 

 during the winter, if they were not amphibious or flying species. The 

 significance of this barrier seems to have been generally overlooked, 

 but a moment's reflection will show its important influence upon the 

 post-Glacial origin of the biota of eastern Canada. 



On account of the presence of the ice sheet on both sides of Hudson 

 Bay, and its longer duration at the Labradorian center, we may safely 

 dismiss the question of the fauna under consideration as being of imme- 

 diate northern origin. On the other hand we have much positive infor- 

 mation which shows that there were centers of preservation of biotic 

 types soutli of the ice margin in the United States. For these reasons 

 our problem becomes one of tracing the probable northern and perhaps 

 eastern migration routes from these centers of preservation to the region 

 vacated by the retreating ice sheet. 



Therefore, keeping in mind the major interruptions of the water 

 barrier and the fauna] affinities of Canada east of the Great Plains, it 

 appears that the major routes into Canada have been, in the east, up 

 the Hudson and Champlain valleys, along the Appalachian range up the 

 Hudson and Mohawk valleys and thence around both ends of Lake 

 Ontario — routes for the coastal and Appalachian types. The Ohio val- 

 ley types invaded Ontario around both ends of Lake Erie, especially some 

 of the more recent southern and Mississippi forms, around the western 

 end. Perhaps a limited number of western forms have entered Ontario 

 through the Upper Peninsula of Michigan and a very large number of 

 Mississippi valley, and to a lesser degree western types, around the 

 western end of Lake Superior. The Mackenzie basin seems to have been 

 invaded largely up the Mississippi and down the Eed River valleys, tlie 

 Plains also sending their quota. These routes are largely shown by the 

 affinities of the present biota and have in all probability functioned 

 throughout post-Glacial times, because there have been no marked 

 changes in the major routes, with the exception perhaps of the drainage 

 changes which have influenced the fresh-water life. 'SMth such general 

 relations in mind, we are in a position to consider the geographic origin 

 of the northern Ontario fauna. 



In considering the post-Glacial invasion of northern Ontario from 



