FISHER: RIVER TERRACES. 
39 
early stage when the river began its degrading work. The general 
appearance and character of the deposit would suggest that the high 
area is one of local out-wash from the higher ground during the period 
of aggradation. The area may be considered as a delta surface, built 
up in standing water, perhaps the flooded West River. As we descend 
from this delta plain and its lobate slopes, we pass successively younger 
and younger terrace patterns. The older and higher ones show deeply 
incised valleys, rounded edges, and long wasting slopes of erosion. 
At the bottom of the series, in contrast, we have sharp terrace edges 
and steep terrace scarps with few or no evidences of erosion. 
The terrace patterns of the West River are ideal young and middle- 
stage patterns, showing free and defended cusps. The strong defend¬ 
ing rock ledges south of the Retreat farm buildings, at the pumping 
station, and the strong rock barrier east of the meadow are good illus¬ 
trations of the influence of such accidental barriers. The wandering 
belt of the river has been constantly narrowing by reason of the discov¬ 
ery in its swinging of rock barriers nearer and nearer the axis of the 
valley. The flood plains of the West River valley are both meander- 
built and partition-built. The 262-foot plain on the left side of the 
river is a typical meander-built plain, and the meadow is a good exam¬ 
ple of a partition plain. 
It is evident that in the future the river, if undisturbed, will abandon 
the left-hand course about its large sand-bar island for the deeper right- 
hand course. It will thus swing to the right unless another deflection 
causes a change in the direction of the current. At present flood- 
carved gullies cut deeply into that left-hand flood plain, and their 
further extension through the plain will produce a short-cut thereby 
preventing the right-hand incision of the meander and thus straight¬ 
ening the course of the river. 
Conclusion. 
Obviously the conclusion to be deduced from the foregoing study is 
the one already suggested,— that Millers theory of defending ledges 
affords a better explanation than any other for the terraces of our New 
England valleys. These terraces are not in every instance defended, 
though all the flights of terraces exist in consequence of defending 
ledges. A lessening of the river volume and interrupted uplifts appar- 
