Dodge.J 26!^ [March 21, 
to liiul ill a tt'rr.icoil valley, eilluu- .simple or coiuposite terraces, 
according- as the terraces were foriuetl by tlie river in one uninter- 
rin)te(l geograpl)ic cycle or in more tlian one cycle. But 
alluvial terraces are formed by other i)rocesses than those 
de])ending solely upon the development of the river, and as such 
terraces are often not clearly traceable to one cause, they may well 
be called com|>lex: or involved terraces. In each of tliese classes 
there are several subdivisions, which I shall consider later. For 
instance, among simple terraces we may have normal terraces, due 
to the normal development of a river ; subnormal terraces, due to 
slight but not abnormal variations in the development of the river ; 
and accidental terraces, due to accidental variations in develop- 
ment. I will consider these in detail. 
Normal Alluvial Terkaces. 
During maturity a river would wear down its whole drainage 
area nearly to base-level, so that by the beginning of old age its 
load would in the majority of cases be in excess of its carrying 
power, due to decreased grade, and it would necessarily even 
aggrade its valley forming an alluvial plain. Afterwards, to 
quote Professor Davis', "In the later and (juieter old age of a 
river system the waste of the land yielded slowly by reason of the 
diminishing slopes of the valley' sides, then the headwaters 
deliver less detritus to the main channel, which, thus relieved, turns 
to the postponed task of carrying away its former excess of load 
to the sea, and cuts terraces in its flood plain preparatory to 
carrying it away." At first it would appear that such would be 
the work of the river in its old age. A more careful consideration, 
however, would seem to show that terraces would not be formed in 
old age on account of this very (juietness and the slowness of the 
work at that [)eriod of that river's history. The cutting and wast- 
ing would then be so extremely slow and evenly balanced that the 
river could not sink deep enough into its alluvial deposits to form 
a terrace. If, however, such a terrace could be formed by a river, 
it would be the simplest and most natural terrace we could possi- 
bly have, according to a classification of terraces based on the 
1 Nat. s^poi;. iiia^;., v. 1 , p. '205. 
