G. H. Stone — Glacial Sediments of Maine. 185 



this on the east side. At Macwawhoc our gravel terrace 

 becomes a two-sided plain and soon expands into a plexus of 

 reticulated ridges enclosing kettle-holes. The ridges here are 

 composed of very round cobbles and bowlderets. Approaching 

 Kingman the gravel plain narrows to a two-sided ridge of sand. 



It thus becomes certain that the sand and gravel situated on 

 the east side of the Molunkus stream for the lower twenty 

 miles of its course could not have been deposited by any ordi- 

 nary stream and is of glacial origin. The argument now 

 stands as follows : A large river flowed in an ice channel from 

 the north to the head of the Molunkus valley. It is also 

 proved that a glacial river flowed through the lower part of 

 this valley. Such a stream cannot appear or disappear sud- 

 denly by accident. The stream in the lower twenty miles of 

 the valley must have been the same that flowed to the head of 

 the valley.. But in the upper ten miles the only gravel is that 

 plain in the bottom of the valley looking like valley drift. 

 The shape of the gravel stones is now explained and we see 

 why the alluvium of the valley changes when the osar-plain 

 goes up on to the hills. The alluvial plain of the Molunkus 

 for the ten miles below Sherman is not valley drift but an 

 osar-plain happening to occupy the bottom of a valley. 



It may be added that this gravel system extends from King- 

 man near eighty miles southward to Columbia. We are 

 considering a very long and large glacial river. 



From this and numerous other examples it can be confi- 

 dently affirmed that osar-plains are often found extending 

 across the bottoms of valleys like river alluvium. There are 

 in Maine special facilities for proving these plains to be of gla- 

 cial not fluviatile origin. Thus the courses of the longer 

 osars and osar-plains lead most of them across several valleys 

 of natural drainage. It can thus be easily proved that there 

 is a great change in the character of the alluvium of these val- 

 leys at the points where the gravel systems enter or leave the 

 valleys. The shapes of the gravel stones and various other 

 phenomena also furnish tests, but it requires a considerable 

 study of the till, the glacial gravels, and the valley drift of a 

 region to safely apply these tests. In long north and south 

 valleys, like that of the Connecticut, where the course of the 

 glacial nearly coincided with that of the post-glacial river, some 

 of these tests cannot be applied, and it will be much more 

 difficult to distinguish osar-plains from valley drift than it is 

 in Maine. 



The osar-plains I esteem of more geological significance 

 than perhaps any other form of glacial sediments because of 

 the light they throw on the mystery of the valley drift. 



It is not probable that the narrow channels of the osars 

 broadened to those of the osar-plains by lateral erosion and 



