236 RESEARCH IN CHINA. 
existed a large river, the initial Huang-ho, which meandered in the graded 
valley on the Huang-t’u; or, as a result of tilting, a large stream may 
have developed by rapid headwater erosion and repeated capture, taking 
a consequent course on the tilted surface. The latter, in view of the 
present course of the river, from the highlands of Mongolia toward the 
lowest point of the downthrow of the great fault-system, seems to be the 
more probable origin of the Huang-ho. The stream, being once estab- 
lished, has maintained itsway in spite of later warpings of the surface, 
and has adjusted itself in its lower course to the single exit provided by 
the great east and west fault of the Ta-hua-shan. 
PHYSIOGRAPHIC STAGES. 
In the former of the two preceding sections on the physiography of 
northwestern China, a descriptive statement of observations of physio- 
graphic facts has been given. The second has reference to the growth 
of the rivers, which among physiographic facts are commonly the oldest 
and always the most expressive features of the country. In describing 
what we saw, frequent reference has been made to the youth or maturity 
of the phases of topography and the great differences in this respect noted 
among closely adjacent features of the region. On the summits of the 
Wu-t’ai-shan, at altitudes of 8,000 to 10,000 feet, 2,400 to 3,000 meters, 
above sea, there is an ancient topographic surface which has the aspect 
of a peneplain. In that same mountain mass we recognized in the high 
ridges below the broad old summits a surface of mature erosion, and this 
phase we have found to be widely distributed. We saw it among the 
foothills of the mountains along the western margin of the Great Plain 
in the vicinity of T’ang-hién; we saw it in the loess basins about Wu-t’ai- 
hién; and it was observed in the valley of the Fén-ho, between T’ai-ytian-fu 
and P’ing-yang-fu. This mature surface is generally more or less covered 
by the Huang-t’u formation, except in the Wu-t’ai-shan, where, being 
deeply dissected, it is represented only by rocky ridges, from which any 
deposit that may have rested upon them has been eroded. The epoch 
during which the Huang-t’u formation was spread over the surface suc- 
ceeded that of mature erosion and is distinct from it. The Huang-t’u 
formation being a deposit resulting from unusual conditions of aggrada- 
tion, the corresponding epoch is distinguished by that fact from those 
epochs during which denudation was the prevailing activity of the region. 
Still another, a fourth, phase of physiographic sculpture was seen along 
our route, and it is one which offers the most striking features in the scen- 
ery of the region. It is the canyon aspect of much of the mountainous 
area. All of the streams whose valleys traverse districts outside of the 
