134 G. H. Stone — Glacial S c <7 line/it* of Maine. 



or superficial, I regard the broad channels as having been open 

 on the top to the air. 



Several of the longer osar-systems change to the form of 

 osar-plains for a distance of five to forty miles, and then 

 return to their original form of the two-sided ridge M r ith 

 arched cross-section. Thus the two developments — narrow 

 and broad osar — alternate in the course of the same gravel 

 system. The topographical relations of the two are the same. 



The diagnosis of the osar-plain has come slowlj' to me. In 

 the earlier years of my exploration all alluvium found in the 

 bottom of valleys was at once classified as valley drift, and 

 was supposed to have no direct bearing on the question of the 

 glacial sediments. It was almost shocking to begin to suspect 

 that the credentials of the valley drift had not all been written 

 out, so great is our veneration for geological terms. The 

 object of this paper is not to deal with original data. But the 

 osar-plains are so important a class of glacial sediments that it 

 may be well to depart from the rule and give a brief descrip- 

 tion of a concrete sample. 



A nearly continuous osar begins in the vicinity of the 

 Sebois lakes and passes south and eastward past Patten to 

 Sherman, where for a time the gravel nearly disappears on the 

 top of the divide between a stream flowing northward into the 

 Mattawamkeag River, at Island Falls, and the Molunkus 

 stream which flows nearly south into the Mattawamkeag at 

 Kingman. The proof is positive that a large glacial river 

 flowed from the north to the head of the Molunkus valley. 

 Naturally this river ought to have flowed clown the valley. 

 For ten miles below Sherman we find the Molunkus bordered 

 by a low plain of sand and fine gravel. This plain extends 

 across the bottom of the valley from hill to hill, just like a 

 sheet of ordinary river alluvium. "We shall notice, however, 

 that the stones of the gravel are more worn and rounded than 

 is common in the beds of Maine streams except among the 

 mountains where there is a fall of seventy-five feet per mile or 

 more. As we tramp this plain for about ten miles the appear- 

 ance of the deposit remains like valley drift and probably we 

 shall become more and more convinced that it is valley drift 

 and nothing more. Then we shall find our gravel plain going 

 obliquely up out of the bottom of the valley and for the next 

 ten miles taking the form of a terrace up to one-fourth of a 

 mile wide, situated on the east side of the river at a height of 

 fifty to seventy-five feet above it. This gravel terrace is on 

 the average at least one-fourth of a mile from the river. 

 From the point where the gravel plain leaves the river and 

 goes up on to the hillsides there is no gravel along the Molun- 

 kus all the way to Kingman, only silt. Nor is there any 

 gravel terrace on the west side of the river corresponding to 



