76 GEOLOGICAL RESEARCHES IN 



of products of this class in place and as scattered fragments at many points on the 

 route across the plateau, and finally the information derived from Chinese authorities 

 concerning the existence within historical times of active volcanoes, among the 

 mountains of Manchuria to the east, and in the Tienshan of the west, all point to 

 a development of volcanic activity, which was formerly coextensive with the area 

 of the present table-land. The remains of this action still make themselves felt in 

 the violent earthquakes that from time to time shake the districts of northern Chihli 

 and the shores of Lake Baikal. 



The greater flows of lavas seem to have been predetermined by the fissures of 

 dislocation, formed along the borders of the area that was subsequently to be ele- 

 vated. Such a fissure we have seen marked by a great fault south of the Lakes 

 Kirnoor and Tehai. 



In the present state of our knowledge of this vast region, it is, I believe, impos- 

 sible to say whether, at the time of the eruption of these rocks, the present depres- 

 sion of the Gobi was or was not under water. That a portion of the southern edge 

 of the plateau was not submerged appears from the fact that where the bottom of 

 the lava formation was visible it was found to rest immediately on the old granitic 

 and metamorphic rocks. This, however, does not preclude the possibility of the 

 existence of undisturbed deposits under the steppe sandstones of the Gobi. 



The sea in which the great steppe deposit was precipitated was studded with 

 islands now represented by the ridges and peaks that rise above the plains. The 

 surface of the plains rises everywhere toward these former islands, partly because the 

 deposit in its formation adapted itself partially to the original surfaces of the vaUeys 

 it fills, and partly fi-om its thickness being increased by the tributary detritus of 

 the islands. The effect of such a combination of circumstances upon the form of 

 the surface, has been discussed in treating of the lake deposits of Northern China. 

 It seems not improbable that the same causes may have operated here as there, in 

 forming many of those lake valleys, the beds of which rest upon the steppe deposit. 



The age of this extensive deposit is a question of much interest. If it is con- 

 temporaneous with the steppes and terraces of the valley system of the Orkhon and 

 Angara, it seems probable that the sea which left this deposit over nearly all of 

 what is now the plateau, was also contemporaneous, Avithin certain limits, with that 

 great body of water which, extending from the polar ocean to the Caspian, occupied 

 all Western Siberia. 



The fact, to which Baron v. Humboldt^ has called attention, that seals, identical 

 in species, inhabit the fresh waters of the lakes Baikal andOron (lat. 55° N., long. 

 119° E.) and the Caspian Sea, seems to refer to that period. The Oron lake is a 

 tributary of the Vitim, and through this of the Lena, in which no seals occur. This 

 circumstance points very clearly to a former water communication between these 

 far separated localities, and the time at which the seals of the Oron became isolated 

 from those of the Baikal and the Caspian falls, perhaps, in the same period with the 

 emergence of the great plains of Northern and Western Siberia, the deposits of 



* Hnraboldt, Kosnios, IV, p. 456, Stuttgart uijcl Tubingen, 1858. Pallas. Zoograpliia Rosso-Asia- 

 tica, 1818, p. 115, 



