LOESS OF CHINA—BARBOUR 293 
tection of vegetation can hold it. How far to such desert-borne 
material must be added the decay products of rock decomposition in 
the locality can only be a matter of conjecture. 
Schlosser, referring to Doctor Andersson’s observations on this 
very point, quotes the latter’s remark that “he had not seen real 
loess, at least in larger masses, on the Mongolian Plateau.” Schlosser 
agrees that the absence of loess may be due to its removal by north- 
erly winds as fast as it was produced by rock decay. Once across 
the Chinese border, however, the mountains protect it from the ex- 
treme effects of the violent winds. Schlosser?* believes that the 
chief material which yields the loess by decomposition is the Hip- 
parion clay, which may thus grade up into the loess locally, as 
happens with the Miocene “ Flinz”’ near Munich, where the transi- 
tion is so gradual as to defy demarcation of the two formations. 
The scouring out of immense quantities of fine dust from deserts 
has probably been a much more common occurrence than we are apt 
to think. It has doubtless affected vast stretches of Central Asia 
and Africa and other areas which in former days had more arid 
climates. Much of the adobe soil of southwestern North America 
has the same origin. Keyes** thinks this factor has been seriously 
underestimated in the case of the deserts of the western United 
States and the deposits of the middle western plains. 
Such vast quantities of the finest products of rock decay clearly 
call for conditions very unlike those of to-day. During the whole 
of Pliocene and Early Pleistocene times, Mongolia and southern 
Siberia stood at a much lower altitude than at present. Relatively 
moist conditions favored extensive decay due to the attack of mete- 
oric waters, and this proceeded to considerable depths, as the relief 
of the land was low and the rivers did not carry off the disintegrated 
material. Pumpelly has discussed the relation of this secular dis- 
integration to the development of loess and allied deposits (8). 
Widespread uplift of the central plateau in Pleistocene times was 
accompanied by increased aridity and high winds, which found a 
ready supply of material awaiting ablation. 
When the dry steppe conditions gave place to moister times, dust 
was lifted more seldom and carried for shorter distances. Its surface 
was periodically planed off by the wash of heavy rains, which also 
helped to distribute it further. During this later time of the piling 
up of “redeposited loess,” it is likely that the new supply from the 
desert was small, the bulk of the loess being shifted only a short dis- 
tance and then “ redeposited.” 
18 Max Schlosser, G. 8. C. Palaeontologica Sinica, Ser. C., Vol. I, fase. 1, “ Tertiary 
vertebrates from Mongolia,’ p. 104 (seen in proof through the courtesy of the Director, 
Geological Survey). 
14 Pan-American Geologist, XLII, 8, 1924, p. 225. 
