380 The American Naturalist. [May? 



the Hudson River, in some places mingling with the forms 

 belonging to the Atlantic drainage area proper, in others 

 occupying the waters exclusively. 



I think we may safely take it for granted that the only way 

 in which the Mississippi Valley Unionidx could have entered 

 these northern and north-eastern river systems was by migrat- 

 ing along connecting fresh water. As there is no such con- 

 nection to-day between these systems the question as to how 

 they reached their present distribution becomes an extremely 

 interesting one. 



If the theory of the Ice Age as held by most glacialists is 

 a true one I think it will fully explain the present remarkable 

 distribution of these extra-limital Mississippi Valley Naiades. 

 And at the same time I believe the evidence of these fresh 

 water mussels is strongly corroborative of the glacial theory. 

 It is held that at the close of the Ice Age a great cap of ice of 

 immense thickness covered North America east of the Rocky 

 Mountains, down to about Latitude 40°. That with the com- 

 ing on of warm weather it gradually melted away at its south- 

 ern extremity, and that when this thawing was continued 

 north of the height of land great lakes were formed whose 

 southern shores were the slope of the land which raised 

 towards the south, and whose northern borders were the slowly 

 dissolving wall of ice. On account of the ice to the north- 

 ward this water could only drain into the Mississippi system, 

 or to the Southeastward, and several old channels are found 

 through which it is believed that it flowed. One of these is 

 the Red River of the North, which almost connects by means 

 of Traverse Lake at its head with Big Stone Lake at the head 

 of the Minnesota River. There is still a broad channel near 

 the western end of Lake Superior which connects with the St. 

 Croix River, and at Chicago there was no doubt an overflow 

 from Lake Michigan into the Des Plaines River, and Lake 

 Erie is believed to have had its outlet into the Wabash through 

 the Maumee which nearly connects with it. The two streams 

 are connected over a very flat country by an old channel not 

 less than a mile and a half wide, and having an average depth 

 of 20 feet. For 25 miles this character continues, and there is 



