1896.] On the Mississippi Valley Unimuim. 381 



very little fall either way. To the northeast this channel opens 

 out into an ancient lake, and at the southwest it touches bed 

 rock at Huntington, and then descends more rapidly. 3 



It will be noticed on the map that the St. Josephs, St. 

 Mary's, and Auglaize Rivers, tributaries of the Maumee, flow 

 in the direction of the Wabash, that the two former join at 

 Fort Wayne and flow partly backward as the Maumee; the 

 whole looking like a tree with its branches broken down, and 

 hanging against its trunk. If the river was continued into 

 the Wabash, and the water all flowed to the southwest it would 

 form a natural looking system. It is quite within the bounds 

 of probability that there were old overflows from the St. Law- 

 rence drainage to the eastward of this through the Oswego 

 River into the Mohawk, or by way of the Sorel into the Hud- 

 son, and possibly through eastern Lake Erie into the Alle- 

 ghany system. 



Now if the water from this region north of the Height of 

 Land flowed over into the Mississippi drainage area at various 

 places it would be almost certain that the Unionidse of this 

 system would migrate up these overflows and into the northern 

 lakes, that in this region they would obtain a foothold and 

 flourish, for the reason that at the time of their entrance it is 

 quite probable that all freshwater life of this area was destroyed 

 by the grinding and crushing of the great ice cap. It is 

 possible that a few of the Naiades of the eastern drainage sys- 

 tem might have survived in the St. Lawrence Valley but it is 

 more likely that such as are now found there have since 

 reached that region by migration from the overflows through 

 the Mohawk and Oswego Rivers, or the Sorel. There has 

 probably been at some time since the close of the Glacial Epoch 

 a connection between the Hudson River and Lake Champlain, 

 as the latter is largely peopled with Mississippi Valley Naiad, s. 

 These forms, most likely, entered Lake Erie through the old 

 Maumee Channel, or by some connection with the Upper Ohio 

 system, passed into Lake Ontario, thence through the Oswego 



3 See a paper "On the Ancient Outlet of Lake Michigan," by Prof., W. M. 

 Davis. Pop. Science Monthly, XLVI, No. 2, p. 217. Also a paper on this old 



