PHYSIOGRAPHY OF NORTHWESTERN CHINA. 247 
of that sea during late Tertiary time, he concludes that an arid climate 
ensued. At the present time we may supplement his hypothesis by adding 
to the continental causes connected with mountain growth those general 
causes of climatic change, which are believed to have initiated the cold 
climate of the Pleistocene.* 
Believing that, toward the close of the Tertiary, there was a notable 
change of climate resulting in pronounced aridity in the interior of Asia, 
we find therein a sufficient explanation for the destruction of the vegeta- 
tion and the removal of the mantle of decayed rock. The change is thought 
to have been one from a mild moist climate to a cold arid climate. Glacia- 
tion is excluded by the absence of any deposits, such as would undoubtedly 
have remained as records if glaciers had developed far beyond the high 
mountain ridges in which they are now found. ‘The degree of dessication 
is thought to have been sufficient to give the wind that power which it 
now possesses as a sorting agent, in those regions where vegetation does 
not clothe the ground. It does not appear that this necessarily implies 
a desert condition, since at the present time, in northern China, where 
crops are successfully grown in many districts every year and in most 
districts three years out of five, the dust storms produced by the wind 
have a notable effect in resorting and redistributing the loess. The degree 
of aridity essential to efficiency of wind action is not inconsistent with 
the continuance of constant streams in larger watersheds. Under these 
conditions of effective wind action during a longer or shorter arid season 
of each year, and of effective erosion and transportation by water during 
the corresponding season covering the remainder of the year, that is, under 
climatic conditions similar to those now existing throughout a large part 
of eastern and northern Asia, both wind and water must have taken part 
in handling the available detrital material. During the dry season wind 
would be most effective, and the transported product would be that which 
was fine enough to be taken up by moving air. During the wet season 
torrential rains would gather the coarser products of rock decay, and, in 
brief hours of activity, transport them in large quantities to alluvial cones 
at the mouths of gulches. In the wide valleys, as on the mountains, the 
alluvium of the streams would be sorted by the winds and distributed, 
and the dust accumulations, spread by the winds in the basins and moun- 
tain slopes, would be taken up by the waters and redistributed as alluvium. 
These are the processes which are now interacting in the Loess Basins of 
Shan-si, and the results are modern deposits of the Huang-t’u formation, 
which are indistinguishable in lithologic character and structure from the 
earliest deposits which we saw. 

* Chicago Journal of Geology, T. C. Chamberlin, vol. vir, 1899. 
