PHYSICAL DEVELOPMENT OF CENTRAL ASIA 641 



lating on these heights feed perennially the few larger and countless 

 smaller streams that flow toward the central basin region. Without 

 these, Turkestan would be an absolutely desert and practically lifeless 

 region. 



I imagine that the general trend of climatic conditions over the cen- 

 tral continental area was from the beginning toward aridity. The 

 mountains that separated it from the ocean were of slow growth, and 

 mountains of moderate altitude are compatible with a moderate amount 

 of precipitation over the interior region beyond them. The grassy 

 plains of Mongolia and of our central western states are illustrations of 

 this. 



The early condition of Turkestan and northern Persia during much 

 of Pliocene time may well have been one in which at first forests existed, 

 at least on the piedmont hills and plains, while the rest of the region, 

 that was not still occupied by the residuary seas, consisted of broad, 

 grassy steppes extending to Europe and of interior areas of deserts. 



Parallel with the growing elevation of the moisture-intercepting 

 mountains progressed the regional desiccation. The progressive effect 

 of this would be continued shrinkage of the water areas, conversion of 

 much of the central plains into deserts, narrowing of the grass-covered 

 zones toward the mountains, and change in the character and extent of 

 the forested areas. 



After the Miocene sea had been shut off from the ocean, it dried up, 

 as is shown in the Sarmatic strata by the widespread deposits of gypsum 

 and salts resulting from the evaporation or the saline waters. That the 

 basin was reoccupied more than once by a more or less extensive land- 

 locked sea is shown in successive formations characterized by changes in 

 organic forms and by old beach and water lines. 



There is little doubt that these expansions of the water area record 

 the climatic changes that mark the advent and phases of the glacial 

 period. An effect of these changes, which were of mundane extent, was 

 doubtless an increase of precipitation over a larger part of the central 

 region. 



In the Glacial period a large part of Eussia west of the Ural moun- 

 tains was covered to a depth of several thousand feet by ice, a large part 

 of which in melting went toward filling the central basin. Our explora- 

 tion in 1903, as shown in the reports of Professor Davis and Messrs 

 Huntington and E. W. Pumpelly, have proved the existence of several 

 successive glacial epochs in the mountains of high Asia during the 

 glacial period, and that glaciers existed on a greatly extended scale 



