336 
lost to history. The larger valleys generally remain sufficiently unobseured 
to enable geologists to trace their courses, either continuously or at inter- 
vals close enough together to enable a safe inference to be made concerning 
their previous courses. The larger the valley the better chance it had in 
general to leave behind itself traces of its former course, for, occupying the 
lowest part of the surface and carrying great quantities of water, it was 
automatically kept open by drainage from the melting ice. Yet even the 
largest river trenches were in imminent danger of defacement. Such an 
instance is found in Jay and Adams counties, Indiana, where there are 
signs of a huge valley whose bottom is buried beneath nearly 400 feet of 
drift and no traces left of its existence on the surface. Another case is 
that of the preglacial Mississippi where it turns southeastward to the 
[Ilinois valley just below Clinton, Lowa. 
The map shows large hiatuses wherein there are no preglacial streams 
indicated, but they certainly exist buried in several hundred feet of drift. 
West and south of the basin of Lake Michigan and between that basin and 
the Lake Erie depression in northern Indiana and Michigan no details are 
shown, and only a few larger courses suggest probabilities of preglacial 
existence. The depth of the drift and the absence of deep-seated natural 
resources do not encourage the digging of a sufficiently large number of 
deep wells to permit the construction of a topographic map of the preglacial 
surface. Enough, however, is known to assure us that the ancient drainage 
lines were quite different in many details from the present systems. 
Without further preliminaries we shall discuss the pros and cons re- 
garding the Claims of the streams shown on the maps to a preglacial an- 
cestry. For the sake of convenience of treatment, the area is divided 
according to the several smaller drainage basins which make up the 
greater Mississippi basin. This will be found convenient because there 
are wide elements of correspondence between the present and the pre- 
glacial drainage basins, as a glance at the generalized map will show. The 
basin of the Great Lakes, which seems to cut out a portion of the Mis- 
sissippi basin, and which is separated by a very low primary divide, over 
which the lakes drained in the Ice Age, is discussed briefly. 
THE PREGLACIAL DRAINAGE OF THE UPPER MISSISSIPPI 
BASIN. 
The preglacial divide of the northern side of the Upper Mississippi 
basin is not definitely determined. It can be pretty definitely located at 
