338 RESEARCH IN CHINA. 
may certainly name the Han-kiang and Yang-tzi-kiang as antecedent. 
Among the consequent streams we would place all those tributaries of the 
Hei-shui-ho and the Wei-ho, of the Han and of the Yang-tzi, which flow 
in essentially parallel courses across the structure and down the warped 
surfaces. 
We must note, however, that in all of these tributary valleys, even to 
the highest part of the Ts’in-ling-shan and the Kiu-lung-shan, we were 
able to trace the features of old valley levels which, in general, coincide in 
course with the existing canyons. Hence it is evident that these conse- 
quent streams developed their present lines of flow at an early stage of 
warping, and that their work during the Yang-tzi epoch has been chiefly 
in sinking their canyons without having greatly modified their drainage 
basins, so far as we have been able to observe. The capture which has 
been accomplished by the Nan-kiang at the Ki-sin-ling is apparently an 
exceptional instance. 
CORRELATION OF PHYSIOGRAPHIC STAGES. 
The distinctive names which are here given to physiographic stages 
recognized in northern China and in the south central provinces are but 
convenient terms which designate parallel sequences of events. It would, 
perhaps, be simpler to adopt one name to cover each of the epochs which 
may be recognized as common to both regions, but in so doing we should 
force a correlation which can not be made with precision, at least not until 
the physiographic study shall have been carried out on many lines of obser- 
vation instead of along one route only. It seems appropriate, however, 
to suggest such correlation as the present knowledge indicates. 
The Pei-t’ai stage of North China was clearly recognized only in the 
Wu-t’ai-shan, in a limited area, where it is identified by characteristic 
topographic features and by the occurrence of residual soil which in that 
district is unusual. The survival of features of such antiquity is rarely 
general, and, in view of the development of later physiographic phases, 
was not expected. It is therefore natural that surfaces of that ancient 
stage should not have been discovered in the southern provinces. 
In North China the succeeding T’ang-hién stage was a time of develop- 
ment to a condition of advanced maturity, when valleys became very wide 
and hills relatively low and isolated; it was interrupted by the Hin-chéu 
stage of aggradation which resulted in the wide-spread Huang-t’u forma- 
tion, and the Fén-ho epoch of mountain growth followed. 
To correlate the two stages recognized in Central China with those 
just enumerated, it is best to reason from the later back to the earlier. 
In North China as in Central the latest epoch is one of very pronounced 
mountain growth. Its characteristic topographic form is the canyon, and 
