JR. S. Tarr — Origin of Terraces in Glaciated Itegicms. 61 



terrace first mentioned, which is reached by a steep ascent of 

 fifteen or twenty feet. 



ISTor are these two terraces the only ones. Above the upper 

 flood plain terrace is a third or still higher one which is much 

 less distinct and more irregular and coarser in composition. It 

 is formed by the wash of material from the bordering hill-side 

 and the detritus brought in by the side streams. This, in the 

 case of exceptionally high floods, is partly worked over so as 

 to form an indistinct upper terrace. 



I ask any student of the subject if this is not a description 

 which would apply almost equally well to terraces in glacial 

 regions. As I know the terraces of the Connecticut, there are 

 one or two lower terraces sometimes present on one, sometimes 

 on both sides of the river. Above these is a higher, broader, 

 flood plain terrace, often of great breadth, and frequently ex- 

 tending over low divides, — just such a terrace as would be 

 expected from a great flood which the ordinary river was 

 absolutely unable to take care of. Still above this is an upper 

 terrace, irregular in distribution, and in form, often of a coarse 

 nature, and particularly so below the mouths of side streams 

 where it is delta-like in form. This is a terrace comparable, on 

 a larger and more perfectly developed scale, with the upper- 

 most terrace of similar origin in the Colorado. 



When I saw the terraces of the Colorado two years ago I 

 was immediately impressed with the resemblance to the terraces 

 of the Connecticut both in form and in cause, the difference 

 being only in the source of the floods. Later I have examined 

 the Connecticut terraces in Massachusetts with this resemblance 

 in mind and it is so striking that I desire to put it on record, 

 and to call attention to the fact that terraces are being formed 

 on a smaller though not by any means an insignificant scale, 

 which imitate the terraces of glacial regions in form and in 

 general cause. In the one case the floods and sediment supply 

 arise chiefly from the climatic accident of desiccation, while in 

 the other they are the result of glacial accident. Otherwise 

 the resemblance seems to be quite perfect. 



