the Western Part of Asia Minor. 33 



the approach laborious. The cone itself is more clothed with vegetation than 

 either of the other two, though it does not yield to them in the steepness of its 

 sides, and the general freshness of its appearance. It possesses a very perfect 

 crate'r, between 300 and 400 feet deep, surrounded by a ridge only ten or twelve 

 feet wide at the top, and about half a mile in circumference. 



On the north-west side of the great cone, a small one about half its height has 

 been thrown up, with a regularly formed crater, but no lava appears to have flowed 

 from it. On the west, several other small cones of scoriae have been formed 

 amongst the lava. 



The lava from Kaplan Alan has chiefly issued from the east side, whence, 

 spreading round both flanks of the cone, the two streams again united, and 

 flowed down the plain to the west for about three miles ; then turning south-west 

 the stream descended a narrow valley for several miles, parallel to the Hermus, 

 from which it is separated by a ridge of lacustrine limestone. At length, finding 

 an opening into the valley of the Hermus, it followed the bed of that river along 

 the narrow gorge in the schistose hills, whence, issuing at Adala, it spread for 

 about a mile over the plain of Sardis, and at length terminated at the distance of 

 at least thirteen miles from its source*. We were able to examine closely only 

 the two extremities of this remarkable coulee, and we recommend the future geo- 

 logist to follow its entire course from Kaplan Alan to Adala, as likely to afford 

 him some interesting examples of the action of the river upon the lava. 



The appearance of this rugged mass of lava, issuing from a narrow gorge in the 

 schistose hills, and spreading over the fertile plain at Adala, is very striking. Half a 

 mile above the town is the remarkable scene exhibited in the sketch No. 13 (PI. HI.). 

 It appears that the Hermus, when its course was first impeded by the lava, was 

 compelled to flow over the top of the couUe, the cavities of which have here been 

 filled by river-worn gravel, forming a stratum on the surface of the lava. In 

 course of time, the river has effected a passage between the mica-schist and the 

 lava, and has denuded both rocks to a great depth. It now flows at the base of a 

 cliff" of compact lava about eighty feet high, imperfectly columnar in its lower part 

 and scoriaceous above. Higher up in the gorge, the river has cut a channel com- 

 pletely through the coulee of lava, and changed its course from the right side to 

 the left. Such are the effects which this river has produced in the lapse of ages. 

 Yet so compact is the substance of the lava, that those parts of the coulee, which 

 have escaped the action of the running water, are as rugged and naked as the latest 

 eruptions of Vesuvius, and exhibit not the slightest tendency to decomposition. 



* It is remarkable that this is the precise length of the lava-stream of the Puy de Tartaret in Au- 

 vergne, which has flowed down the narrow valley of the Couze to Nechers, in a manner strictly analogous 

 to the case here described. See ' Scrope on Central France,' p. 11 ?• 



VOL. VI. SECOND SERIES. F 



