496 PLEISTOCENE OF INDIANA AND MICHIGAN. 



At Ecorse a bend, much like that above Amherstburg, has the same radius and slightly 

 greater length, but is not so easy to explain, though it seems to be partly due to Ecorse River, 

 which enters at this point, and partly to the fact that the undivided stream of Detroit River, 

 since it sank to its present deep and relatively narrow bed, strikes the bank at this point with 

 full force. 



It is to be remembered that St. Clair River and the Detroit River above the head of the 

 estuary flow at a slightly lower level to-day than they ever flowed before while carrying the full 

 discharge of the upper lakes. It seems certain from this fact that their beds in the immediate 

 past were not so deeply trenched as they are now, and this is the same as saying that they have 

 been in recent times and almost certainly are still deepening and shaping their channels. These 

 rivers bear many evidences of the fact that they are of recent origin. One is their straightness 

 and the absence of any but very slight curves or meanders, like those of the Mississippi and other 

 great rivers. (It must be remembered, however, that the Detroit and St. Clair rivers are not 

 sxibject to floods and do not vary more than 5 or 6 feet between extreme stages, and are thus 

 less likely to develop meanders.) The only features suggesting lateral cuttings and the develop- 

 ment of meanders are the bends above Amherstburg, at Ecorse, and in the rapids at the head 

 of the St. Clair River. This latter bend, however, is due to a different cause (p. 483). 



The sinking of these rivers into their present narrow, constricted channels in clay is very 

 recent. It is certainly not long since their courses were marked by shallow expansions which 

 have now disappeared. Lake Rouge in particular is of this class and Lake St. Clair was much 

 more expanded, mainly toward the east, south, and north. Almost the entire delta, as known 

 to-day, was submerged when Lake St. Clair stood 3 feet higher. With the exception of the 

 bowldery and partly rocky barrier at Trenton and Amherstburg, the entire course of these rivers 

 is in soft sediments. If the resistant barrier west of Limekiln crossing had been absent the 

 sinking of the rivers into the clay plain would have been more rapid. But the barrier yielded 

 only very slowly and the rivers have in consequence got down into their present beds so recently 

 that they have not had time to develop the normal characters of graded rivers flowing through 

 soft sediments; lateral swing has barely begun to find expression. The cutting of the new 

 Livingstone channel through the rock floor east of Stony Island will tend to lower the surface 

 and increase the rate of flow and also the rate of channel deepening in the rivers above. 



FLOODING OF TRIBUTARIES. 



CAUSE OF FLOODING. 



All the streams — even the very small ones — which are tributary to this connecting river 

 system show certain characters which are abnormal and are not reconcilable with the ordinary 

 course of stream development. Their mouths and lower courses are all drowned in precisely 

 the same way as are the mouths of the streams that enter the western parts of Lakes Ontario 

 and Erie. In the lakes the drowning was formerly supposed to be due entirely to recent uplift 

 at their outlets, causing the water to back up and overflow the southwestern shores. But it is 

 due also in considerable part to large increase in the volume of discharge at the outlet, amount- 

 ing to 700 per cent, as shown on page 462. 



The backing up of the waters in Lake Erie probably deadened the flow of Detroit River a 

 little, but evidently only a very little, for gentle rapids still exist at the Limekiln crossing, and 

 the river is lower and Lake Erie higher than formerly. The drowning in these rivers is due 

 chiefly to the return of large volume after a long period of abandonment and very small volume. 

 During the dry stage the river channels became long, narrow valleys, like the dry coulees of 

 the West, but smaller and . shallower, in which the small streams were lost. Meanwhile all the 

 tributaries, great and small, worked away, deepening their beds to the lower base-level, which 

 may be taken as the bottom level of the shallower parts of the great channels. The holes in 

 the great river beds were probably ponds and the intervening stretches were swampy flats. 

 The floor of the whole connecting river system was probably in this condition. 



