Yol. 54.] THE STET7CTURE OF THE DAVOS VALLEY. 283 



places the 30-metre contour- lines on the map are only J millimetre 

 distant. The slopes of the Seehorn dip straight into the lake ; those 

 of the Riicken disappear in the Lareter gorge ; but the steepness 

 of the Hornli mass is suddenly broken at the Drusatch Alp, and 

 a long, low, pine-clad ridge stretches from that point transversely 

 acro^ the valley to "Wolfgang. 



On the west rise the livid green and purple wastes of the Todtalp^ 

 a great mass of intrusive serpentine which justifies its name by the 

 weird colouring of its weathered slopes and the terrible sterility of 

 its upper plateaux. Its eastern face is scored by hundreds of deep- 

 cut trenches, and the vast talus at its foot spreads fan-like down- 

 ward, overlapping the end of the Drusatch- Wolfgang ridge. 



Most of the streams which are so rapidly denuding it combine 

 with the Stutzbach to form the Lareter Bach and thus discharge 

 to the north. This is not the case with the southernmost, the 

 Todtalp Bach itself : rising far back under the Mittelgrat, and 

 running at first north and east, the stream then bends round the 

 southern edge of the talus-fan and falls into the lake. It seems 

 impossible to regard this as the original course. Everything appears 

 to suggest that it formerly drained northward, but has been gradu- 

 ally pushed over by the accumulating talus. This leads one to 

 consider what determines the watershed of the whole area and what 

 is the real nature of the bridge between the Todtalp and the Hornli. 

 Its position and contour, its sharp distinction from the lateral 

 mountain-slopes, and its situation at the end of a lake all suggest 

 the idea of a moraine. 



It is, I believe, impossible to find solid rock in situ anywhere 

 between Drusatch and the Todtalp. The steep Hornli slopes are 

 separated by nearly 2 miles from the solid serpentine, and though 

 the latter is prolonged down just to the west of Laret, there is even 

 here a wide interval filled with superficial deposit between it and 

 the schists on the east. The Drusatch-Wolfgang ridge is composed 

 mainly of blocks of serpentine from the Todtalp, and this has 

 naturally led to its being mapped as such. Mixed with these, how- 

 ever, are large erratics of granites, gneisses, and schists, as well as 

 fragments of limestone and verrucano. It will be suggested that 

 there may still be a transverse solid rock- ridge below ; but it would 

 be difficult, if not impossible, to account for the formation of such 

 a ridge, and still more difficult to explain the lake. 



The maximum height of the mass above the present lake-level is 

 200 metres, and there seems no reason to doubt a similar thickness 

 below. The northward extension is also considerably larger than 

 one would at first suspect. It forms the plain round Laret and 

 the Schwarzsee, as also the rising ground of Weiden between the 

 railway and the Klosters road ; and then is prolonged in a sloping 

 tongue down between the Lareter Bach and the Eiedloch Bach, 

 Only at Seifranga, above Klosters, is the solid rock (limestone and 

 serpentine) met with in the river-bed. From below Laret to this 

 point the river runs between the cliffs of gneissose rocks on the east, 

 and the mass of detritus between them and the Schwarzsee. The 



