328 M. Guyot on the 



more crystalline forms, and a general tint of deeper yellow. 

 Mont Mine also yields cliloriteous gneiss, but it is rather 

 syenites and various rocks rich in amphibole, which predomi- 

 nate there. 



More to the east, the arkesino and chloriteous gneiss dis- 

 appeai", and seem to be wanting in the chain of the Weiss- 

 horn. At least, M. Guyot scarcely met with any in the val- 

 leys of Anniviers and Tourtemagne, which descend from this 

 great chain ; and the numerous blocks of these two species 

 which he noticed on the glacier of Zmutt, and on the heights 

 which overlook its left bank, evidently come from the Dent 

 Blanche. 



"We thus see that the chlorites, the chloriteous gneiss, and 

 the arkesines, as well as the green granites, syenites, and 

 other amphibolic rocks, belong to the most central, and most 

 elevated part, — in a word, to the axis of this high chain of the 

 Pennine Alps situate between the bottom of the val de Bagnes 

 and the col d'Erin. It is in these almost inaccessible peaks, 

 and in the bosom of the glaciers that descend from them, 

 that we must seek for the rocks from which have been de- 

 tached the erratic masses, at once the most numerous, most 

 colossal, and most widely distributed on the surface of the 

 basin of the Rhine. How surprising that their primitive site 

 should have remained so long unknown .' 



The group of erratic species to which M. Guyot, with a 

 just title, has given the name of Pennine rocks, likewise be- 

 longs to this elevated chain. These rocks have issued from 

 it by two principal channels only, the valley of Erin and that 

 of Bagnes. The valley of Viege furnishes only a small num- 

 ber of them, for this may be considered an accidental outlet 

 for these rocks, while it is the principal, indeed the only 

 canal, by which the rocks of Mont Rosa have been brought 

 doAvn to the valley of the Rhone. 



The group of the Mont Rosa rocks contains species only 

 whose place of origin was previously in some degree deter- 

 mined. The principal site of the serpentines is the region 

 lying- between the great Cervin and the Lyskamm. The chain 

 of the RifFel, likewise composed of serpentines and which is 

 prolonged eastward into the Saasgrat, may be regarded as 



