252 MESSES. SKEEICHLY AND EINGSMILL ON [May 1895, 



evidently marine in origin, apparently once widely spread, but now 

 principally confined to the neighbourhood of Nanking and Chin-kiang 

 (lat. N. 32° 20', long. E. 119° 30'). 



The ancient rocks of the Yangtse valley are for the most part 

 Carboniferous, but passing down to Devonian and probably Silurian. 

 They were formerly capped by the upper coal-bearing sandstones, 

 but these were in comparatively early times removed by denudation, 

 except in a few isolated patches, and not improbably the red Hang- 

 chow Sandstones are the result. Everywhere along the Tangtse, 

 from Western Szechuen to Chinkiang, the Carboniferous Limestone 

 is the prevailing rock where the Palseozoic strata are exposed, and 

 it preserves throughout a remarkable similarity in structure and 

 composition. Certain of its beds are marked by the occurrence of 

 large masses of chert, the remains of siliceous sponges, and these 

 are usually accompanied by the characteristic shells of the deposit, 

 Sjoirifer, Ehynchonella, etc., as well as by the remains of corals. 



The siliceous gravels, which from their being characteristic at 

 this spot near Nanking we may call the Yii-hwa-t'ai (lat. N. 32° 1', 

 long. E. 119° 5') Gravels, are the remains of the slow denudation, 

 under aqueous conditions, of the Carboniferous Limestone. They 

 are, in fact, entirely composed of the ancient chert-sponges, worn 

 down and rounded, but marked by the contained fossils ; and, 

 though they attain a thickness of several hundred feet, not the 

 smallest particle of the limestone-matrix can be discovered in the 

 mass, it having been entirely removed in solution. 



The period of deposition of these gravels ended with an outburst 

 of volcanic activity in the region of Lower Kiangsu at least. Yiolent 

 eruptions on a large scale threw out wide sheets of basaltic lavas, 

 occasionally assuming columnar structure, while in places volcanic 

 cones still cap the gravels. To this outburst we apparently owe the 

 preservation of such portions of the gravels as still exist. 



Although these later rocks have hitherto proved barren of fossils, 

 the cause is probably inadequate search. They are, however, 

 distinctly marine, and it may be assumed that the greater part of 

 North China was at this period covered by the ocean. We speak 

 with some degree of diffidence, but there seems evidence for con- 

 necting these beds with the pre-loess deposits of Eastern Turkestan 

 and the Upper Oxus. However it happened, the way was prepared 

 for the deposit of the loess : most probably a gradually rising land 

 altered the currents, and converted an open into a more or less 

 land-locked muddy sea. The loess, in Kiangsu at all events, 

 immediately succeeds the basaltic flows, and is to be found capping 

 the thick masses. The beginning of the period was probably, how- 

 ever, accompanied by a sinking which allowed for the formation of 

 the fringing beach already spoken of ; but this sinking was doubtless 

 in no case much more than 1000 feet and in most places much less, 

 and ended in a long period of rest during which the loess was 

 deposited to a depth in Shantung of a little over 500 feet, in 

 Kiangsu to about 400 feet, and in Shanse and the west probably 

 1500 to 2000 feet. 



