R. S. Tarr — Origin of Terraces in Glaciated Regions. 59 



Art. YIII. — A Hint with respect to the Origin of Terraces 

 in Glaciated Regions / by Ralph S. Tare. 



The exact method of formation of terraces in rivers flooded 

 by the melting of the ice sheet, such rivers for instance as the 

 Connecticut, does not appear to be definitely agreed upon by 

 geologists. By some they are supposed to mark recent erosion 

 in a drift-filled valley. That is, the supposition is that the 

 glacially fed rivers were overloaded and actually built up their 

 beds' to the height of the higher terraces, and in this exten- 

 sive deposit the present terraced valley has been carved. The 

 other important theory is that the terraces mark high stages of 

 floods — that they are flood plains at various stages of flooding. 



It is not my purpose to enter into the subject in a critical 

 manner and state the reasons pro and con which have been 

 advanced by the advocates of these theories, but, rather, to 

 record some observations in an entirely different region where 

 the terraces of glaciated regions are being imitated, and where 

 the general conditions are quite similar to those attending the 

 formation of the terraces in glacial regions, as I understand 

 those conditions. I refer to the valley of the Colorado in 

 Central Texas. 



The river is here superimposed upon a hard Silurian barrier 

 which is effectually retarding its downcutting although the 

 river is still well above base level. One of the effects of this 

 retardation of development on the region upstream from .the 

 barrier is that a temporary base level is produced and the Colo- 

 rado itself and the side stream are for a certain distance actu- 

 ally building up their beds. The effect of the barrier is thus 

 felt for forty or fifty miles ; but above this, the flow is rapid 

 and the river is degrading its channel. Along the entire 

 course the side streams are rapidly at work, and from these 

 two sources much sediment is being furnished. Owing to the 

 many soft beds of Cretaceous, Permian and Carboniferous 

 through which these streams are flowing the amount of sedi- 

 ment supply is very great. 



Another point of importance in this connection is the pecu- 

 liarity of rainfall. The immediate region is sub-humid, the 

 extreme head waters are in a truly arid region. Consequently, 

 the water supply, during a great part of the year, is small in 

 amount ; but heavy rains, which are of annual occurrence, 

 and often of greater frequency, bring to the river vast floods 

 of water which the ordinary channel is totally unable to hold. 

 Almost the entire rainfall has to be carried off; for the barren 

 soil holds but little, and the violence of the rain speedity forms 



