■^rj, \v. \|. DAVIS — GLACIAL SAND PLAINS. 



New Hampshire, is a large example of this kind. The sand and gravel 

 plain of Rock river in southern Wisconsin appears, from its descriptions and 

 from the brief sight that I have had of it, to be another. As a natural con- 

 f the change from the conditions of their formation to their present 

 conditions it follows thai valley flood-plains are now commonly terraced by 

 the Btreams that formed them, while sand plains proper are nearly always 

 avoided by streams. 



The other deposits that simulate the glacial sand plains are the stream 

 delta-, formed normally in the ponds that temporarily fringed the ice front. 

 Following Upham again, I have been led t<> an excellent illustration of these 

 deposits in the ( lontoocook valley in southern New Hampshire, where a lake 



of< siderahle size was ponded hack by the ice. The streams that enter this 



valley tunned deltas of several acres in extent when they entered the hike, 

 and these deltas are now found perched up. at an accordant altitude, on the 

 p hill-slopes that enclose the valley : and at thesame level, stony benchi 3, 

 sandy beaches, and linear bars may easily be traced lor many miles. Emer- 

 has described similar stream deltas in the Connecticut valley. Like 

 the valley flood-plains, these normal Btream deltas are now commonly cut 

 through by the streams that made them. Like the glacial sand plains, they 

 present Btrongly marked frontal slopes, but, unlike them, they are built out 

 from a solid hind support, againsl which they still rest, while the support 

 from which the glacial -and plain- grew ha- vanished away. 



Points needing further Observation. — The search for structural features of 

 sand plains 'in which I have been aided by several students, especially 

 Messrs. Ropes and Stone, of the class of L889 at Harvard College, and by 



Mr. Gage, a Bpecial Btudent) has not yet discovered any example- of the 

 superposition of -and plain beds on their foundation of till ; the statement 



already made to the effect that such is the order of deposit is based partly 

 on the doI infrequent protrusion of glaciated rocky knobs above the surface 

 of a plain, and partly on the apparent overlapping of till slope- by sand- 

 plain lobes. Nor has the point of change from back-set to fore-set beds been 



found in any cut yet visited: hut the occurrence of fore-sel beds close up 

 to the head of several plain- -how- clearly enough that very little room 



can he left for the back-sets. No Section of the slopes of a pit within the 



plain ha- vet been discovered; the nearest approach to this was the finding 

 of a -mall huried pit aboul fifteen feet in diameter— that is, a pit that had 

 been filled by subsequent deposit of top-set beds; this showed distinct down- 

 faulting of the marginal beds, a- if some local Bupporl had been withdrawn 



from below them, and thifl i- interpreted as indicating the melting away of 



a -mall ice block, after some of the top-sets had been Bpread over it, and 

 before the building of the plain had ceased. 



Whei r search ha- been carried further, we shall attempt a fuller state- 

 ment of the case, with detailed illustration of many sections and various 

 Band plain 



