Quaternary Deposits, Etc. — Mills. 359 
tent washed away and terraced and otherwise carried by waves 
and currents, and having its quartz and other hard materials 
strewn along the shores in beaches and sand ridges. 
Richtofen's theory of the deposition of loess by wind fails 
also to meet the conditions here. The region is near the sea, 
and is in the pathway of the tradewind blowing from the sea, 
which now comes laden with moisture, and there is no reason 
to think that there has been within Quaternary or recent time 
a dry region to the windward where the wind could be 
charged with dust to be dropped here. 
I think that facts from the Quaternary history of the region 
already stated afford a full explanation of the origin of the 
loess. The region had been above sea-level for long periods, 
and had been worn down to a " base level of erosion " so long 
that a very thick layer of decomposed rock had been formed at 
the outcrop of all but the most resisting rocks. It was there- 
fore a low-lying region, and its streams were sluggish 
and deep, and their waters nearly up to the level of 
the general surface of the land, and with water clear except 
as colored and rendered turbid by organic matter; for 
the slopes of the region had become so low that erosion 
was reduced to a minimum and furnished very little 
sediment to the streams. Then began the Quaternary up- 
lifting. The uprise was greatest along axes of former 
uplifting that is, along the divides at the heads of the 
streams. As soon as portions of the region began to rise 
erosion there was increased and on account of the softened 
condition of all but the most resisting rocks, was very rapid. 
The streams became charged with sediment then, and as they 
were flowing away from the lines of greatest uplift, and their 
slope and velocity decreased as they flowed on, their power of 
transportation decreased relatively to their load of sediment, 
and when they reached the still undisturbed low-lying por- 
tions of the region they dropped the coarser materials of their 
load — the sands and gravels — upon their beds. An overloaded 
stream cuts away its banks, because it drops its load unevenly 
on its bed, and so obstructs and deflects its current, and each 
deflection starts a series of undulations from side to side, and 
the stream cuts into its bank right and left, and so takes up 
more sediment to be deposited near by. This is the origin of 
the more locally derived sediment. In time the beds of the 
