Vol. 51.] THE LOESS AND OTHER DEPOSITS OF SHANTUNG. 243 



dition, spread out in large indurated masses, shows that at that 

 time the coast of the Yellow Sea was washed by comparatively 

 clear water, in marked contrast with its present mud-laden 

 condition. 



Prior to this, during the deposition of the Yang-chow Marine 

 Sands, the waters of the Yellow Sea must have stood at a level 

 about 200 feet higher than is now the case, and have covered the 

 lowlands of Shantung, much of Northern Kiaugsu, a considerable 

 extent of Honan, and the entire eastern portion of Chihli. In places 

 north of the Yangtse, as near Ting-yuan-hyien (lat. rT. 32° 32', 

 long. E. 117° 30') in Anhwei, near Fung-yang-fu (lat. N. 32° 40', 

 loug. E. 117° 20'), the boundary of the loess eastward is as well 

 marked by ranges of old bluffs as the subsequent coast-line at 

 Chushan during the earlier part of the Yangtse delta- forming 

 period. 



Doubtful deposits of this age, with marine shells, occur in Hong- 

 kong Island, and at Amoy, Swatow, and near Takow in Formosa. 

 These we have seen, but not since we worked out the present 

 scheme. 



3. Old River-gravels. The old river-gravels of Shantung form a 

 very interesting set of deposits. They were first seen by one of the 

 writers (T. W. K.) in 1887, near I-chow-fu (lat. N. 35°' 5', long. E. 

 118° 30') in Southern Shantung. Here, as elsewhere in the 

 province, they constitute a conglomerate of limestone-pebbles in a 

 calcareous cement, the component fragments varying from fine 

 gravel-stones to heavy shingle. More than 90 per cent, of the 

 pebbles are Carboniferous Limestone, and indeed when the gravels 

 were first seen they were looked upon as a conglomerate of Car- 

 boniferous age. Their true character soon made itself known, when 

 they were traced to their origin in the heads of valleys in the 

 Carboniferous Limestone-ranges of Tsi-nan-fu. 



The conglomerate is exceedingly hard, and hence has resisted the 

 denudation that swept away the looser material which doubtless 

 originally accompanied it. The beds often staud out as bosses and 

 banks upon the loess-plain, just as, and for similar reasons, many of the 

 ancient gravels of Cambridgeshire and Norfolk rise above the Chalk. 

 So eminently calcareous are they, that in places where no limestone 

 is locally available, as at Chow Ts'un (lat. N. 36° 48', long. E. 

 118° 3'), they are used for lime-burning. From their very nature 

 they are fragmentary, ranging from 1 or 2 to over 6 feet in thickness, 

 sometimes only a few feet in length, sometimes continuous for 

 hundreds of yards, and from 8 to 30 yards in width. So compact 

 is the mass that fragments 6 or 7 feet wide and 20 to 30 feet long 

 are often seen projecting like shelves from the loess, and at one 

 place we rode through a chasm 15 feet wide spanned by one of these 

 masses hardly a foot thick. It is frequently used as a rough 

 building-stone for retaining-walls. 



In the great Carboniferous Limestone district which flanks the 

 sacred T'ai Shan range (lat. 1ST. 36°, long. E. 117°), these relics of 

 prehistoric rivers are still to be seen clinging to the valley-sides. 



