378 E. HUNTINGTON GLACIAL PERIOD IN NON-GLACIATED REGIONS 



uplands. Under other conditions reed beds flourished. During a single 

 prolonged interlacustrine epoch many minor variations of climate appear 

 to have taken place, so that the strata exhibit much variety. The varia- 

 tions can hardly have been due to movements of the earth's crust, for 

 there is no sign of erosion or of the unconformities which would probably 

 have been formed imder such circumstances. Moreover, the fact that in- 

 spite of highly complex changes, involving constant repetition of a varied 

 series of events, the history of Turfan agrees with that of distant regions, 

 such as Lop and Seyistan, argues in favor of some widespread cause such 

 as worldwide changes of climate. 



Throughout the epochs shown in figure 14, the climate seems at all 

 times to have been moist enough so that plants grew vigoroiisly. Even 

 when coal or carbonaceous claj^s were not deposited, there were plants 

 enough to leach out from the deposits their oxidized iron as fast as it was 

 formed. The absence of red color seems to demand this explanation. 

 During later thesial epochs, however, red strata Avere deposited, as we 

 find in the almost concealed layers lying above those shown in the section. 

 Hence we may infer that during some of the more severe thesial epochs 

 the climate of Turfan became so dry that plants ceased to flourish and* 

 iron oxide accumulated with nothing to take it out. 



COMPARISON OF THE PLEISTOCENE VEGETAL LAYERS OF TURFAN WITH 



ME80Z01C COAL BEDS 



The very recent interlacustrine vegetal layers described above are not 

 the only ones at Turfan. At the l)ase of the faulted and tilted strata 

 exposed in the gorges of the Fire mountains workable coal is exposed. 

 It is probably of Cretaceous age, biit this is not certain. The section, as 

 seen in the Lemjin gorge, a few miles north of Lukchun, is as follows : 



(10) A great thickness of fiue-grn iiied, sandy, red silt or shale, full of 

 lenses, mud cracks, and other signs of snhaeriul origin (see ]>late 3(5, 

 figure I, and plate 39, figure 1). 



(9) Transitional beds, showing a gradual change from the underlying pure 

 green shales to the overlying sandy red shales. The transitional 

 beds consist of an intimate mixture of layers of red' and green and 

 purple of all thicknesses from an inch to 5 or 10 feet. At the bot- 

 tom, green beds predominate, and at the top, red. 



(8) Several hundred feet of green shale. 



(7) Thin coal seams. 



(6) 30 feet of greou shale. 



(5) 20 feet of coal measures, with seams of bituminous coal from 1 to 8 

 inches thick, some of which are worked. The rest of the 20 feet 

 consists of carbonaceous shales and of thin partings of iron ore. 



(4) 100 to 200 feet of green shale, 



(3) A workable layer of coal. 



(2) Soft whitish shales. 



(1) Yellowish gi"een shales. 



