THE PREGLACIAL DRAINAGE OF OHIO. 35 



valley is about one-half mile wide, and so corresponds to the 

 present valley of Sugar Creek at Bellbrook. Similarly it har- 

 monizes with the Miami at the place of junction with that 

 stream. These relations show plainly that Sugar Creek formerly 

 flowed through this old valley, and thence northward in the 

 valley of the present Miami. 



The Miami valley below the point of junction with Sugar 

 creek was occupied by a stream which flowed east to Spring 

 Valley where it united with another stream which will be dis- 

 cussed later. 



Now the question how was the change from these conditions 

 to the present produced? The answer is not difBcult to find. 

 It is one of the many changes produced by the great ice-sheet 

 which formerly covered the northern half of the continent. The 

 existence of a terminal moraine across the valley at Alpha 

 shows that the ice front once stood at that place. This effec- 

 tually blocked the course of the north flowing Sugar creek. 

 The waters were ponded in front of the ice forming a long and 

 narrow but deep lake. The waters rose higher and higher until 

 they overflowed the divides, thus starting the streams in their 

 present courses. The rapid flowing silt laden waters soon low- 

 ered the divides thus draining the lake, but not until its bed had 

 been rapidly silted with drift. After the withdrawal of the 

 glacier the streams found it easier to continue in their new 

 channels than they did to clear out the drift deposits and resume 

 their preglacial courses. 



From the point of junction of Sugar creek and the Miami, 

 the latter flows east to Spring Valley and thence making a sharp 

 turn runs due south for a few miles. The valley widens until 

 just opposite Mt. Holly where it is a mile or more in width, not 

 being exceeded in this respect by any part of the valley except 

 just above Cincinnati. Everywhere in this section of the river 

 there is a heavy mantle of drift. About ten years ago a deep well 

 was sunk at Spring Valley, and according to the best evidence 

 now obtainable 170 feet of drift were found. This well it should 

 be noted was on the north side of the valley in the angle made 

 by the sharp bend in the stream. More recently two deep wells 



