284 PLEISTOCENE OF INDIANA AND MICHIGAN. 



of a great river, and it is tentatively suggested as the possible correlative of the Lum channel 20 

 miles farther north. 



Certain facts strongly suggest this possibility. In the first place, both the Rochester and the 

 Lum channels lie close against the front of the Birmingham moraine, and it seems clear that if 

 this moraine had drawn back a little from its present position between Romeo and Lum it would 

 have permitted a continuous open passage between the two places. 



In the second place the weak and greatly modified part of the Birmingham moraine lies 

 just in the interval between the north end of the Rochester channel and the south end of the 

 Lum channel, as might be expected if the channel had been closed in this interval by a read- 

 vance of the ice. The facts are so simple that it] seems impossible to doubt the reality of this 

 event. 



But if these channels are related in this way the altitudes of their floors are such that it 

 would be necessary to suppose that the land around the Lum channel stood somewhat lower 

 with reference to the Fort Wayne outlet than it did later at the time of the Imlay outlet, 

 and hence that after the Lum channel was closed and before the Imlay was opened there was a 

 slight elevation of that region. That such an elevation took place is not inconsistent with the 

 later history of this district. 



Regarding the Rochester-Liim channel as a former outlet of Lake Maumee, it may be noted 

 that its head was near Birmingham and that if allowance is made for the later sandy outwash 

 that covers the channel floor between Rochester and Birmingham the head appears not to have 

 been measurably affected by this early uplift, for the channel still stands in normal relations 

 to the highest beach of Lake Maumee at Birmingham. It appears to follow that the uplift 

 which elevated the Lum section before the opening of the Imlay channel did not affect the 

 land at Birmingham; nor, so far as indicated by the highest beach of Lake Maumee, did it 

 affect any of the lands bordering on this lake farther south. 



When compared with other parts of the shore the inner slope of the moraine between 

 Rochester and Romeo appears to be steepened more than would be expected from wave action, 

 and from this it is inferred that as the ice front drew back from the Birmingham moraine the 

 waters of Lake Maumee rushed northward through a narrow passage between the ice and the 

 inner side of the moraine and cut away the base of the latter. At Romeo the middle beach of 

 Lake Maumee runs along the crest of a small, narrow morainic ridge — apparently the last 

 stand of the ice during the building of the Birmingham moraine. Between this and the high 

 moraine there is a trough one-fourth mile wide which seems to have been scoured out when the 

 ice stood at the position of the middle beach. These two features are distinctly developed in the 

 south part of the village of Romeo. Later, when the ice had withdrawn eastward the waves of 

 the lake at the level of the first or highest beach made a shore line along the base of the hill a few 

 feet above this channel floor. The inner slope south of Rochester is not steepened in this way. 



WATER-LAID MORAINES AND BOWLDERY STRIPS. 



The water-laid moraines and bowldery strips are less definite than the land-laid ones, and 

 their correlations are more or less uncertain. It is thought best, however, to describe briefly such 

 of their features as seem to throw light on the recession of the ice sheet from the southeastern 

 part of Michigan. There is more room for difference of opinion here than there is in regard to 

 the well-defined glacial features, and much that is here presented must be regarded as tentative. 



DETROIT INTERLOBATE MORAINE. 



The broad, smooth ridge running directly down the slope from Birmingham to Detroit (p. 292 ; 

 also fig. 13, p. 485) is not to be regarded as the normal continuation of the Birmingham mo- 

 raine, for such an assumption would involve a violation of the fundamental principles of the 

 adjustment of a semiplastic ice mass to topography and of the law governing liquid or semiliquid 

 bodies moving on the line of least resistance under the force of gravity. 



The Birmingham moraine appears to continue southwestward beyond Birmingham as a 

 bowldery strip with the same normal altitude relation that it has between Romeo and Binning- 



