592 W. J. Hamilton, Esq., on the 



form one of the most striking features of these valleys, which have been ex- 

 cavated in every direction in the volcanic pumiceous sand, to the depth of 

 three or four hundred feet, by weathering- and the action of water. 



To the eastward of the volcanic cone, whence the basalt of Tatlar flowed, 

 another table-land of cellular basalt extends about a mile, to the deep ravine of 

 Alajahscheher, the castle of which stands upon the edge of the basaltic cliff. 

 The basalt on the west side of the ravine overlies the horizontal pumiceous 

 tuff, which differs from that of Tatlar only in being of a pinker colour, and 

 in consisting almost entirely of pumice, but containing, in some of the beds, 

 numerous small fragments of black obsidian. 



Thick beds of this tufaceous deposit, varying in hardness, continue, with 

 occasional exceptions, almost the vvhole way to the foot of Mount Argasus. 

 But the compactness or hardness of the beds increases considerably near the 

 trachytic base of that mountain, from which they are separated by a narrow 

 and irregular plain or valley. 



A few miles to the eastward of Alajahscheher, and before reaching Nemb- 

 scheher, a mass of red trachytic rock rises to a considerable height through 

 the pumiceous tuff, which appears to have been quietly deposited against it, 

 as the tuff presents no trace of having been upheaved from its original hori- 

 zontality. 



A short distance from Nembscheher, a deep ravine exhibits a section of red 

 and yellow marls underlying the pumiceous deposit, and dipping considerably 

 to the south-east. The marls contain many veins and masses of white fibrous 

 gypsum, the whole of which must have been disturbed previously to the accu- 

 mulation of the tuff. I have already mentioned that selenite occurs in the red 

 sandstone and m'arls near Kodj-hissar, to which formation these marls ought 

 also to be referred, as well as the red sands beyond the volcanic table-lands 

 to the north. In a deep ravine to the north of Nembscheher several beds of 

 basalt, alternate with the tuff, and were probably derived from some high hills 

 to the south. Basalt also occurs in the bed of the torrent which flows down 

 from these mountains, on the east side of the town. 



The same tufaceous rock, in horizontal beds, extends from Nembscheher to 

 Urgub; at which place, and at Utch-hissar, half-way between the two, it has 

 been excavated into deep valleys, by weathering and the agency of two small 

 streams. From the unequal manner in which the rock has been acted upon, 

 the most extraordinary phenomena have been produced. From the edge of 

 the plateau of Utch-hissar (PI. XLVIII., fig. 5.) we looked down upon a deep 

 vaUey,in places three or four miles wide,and filled for several miles up and down, 

 with an infinite number and variety of sugar-loaved cones, from 150 to 300 



