is still 4260 English feet above the level of the sea. A few wretched huts, surrounded by the same wool or cotton-trees (Bombax discolor) which we had first seen on the banks of the Amazons, were called an Indian village. The scanty vegetation of the valley bears some resemblance to that of the province of Jaen de Bracamoros, but we missed the red groves of Bougainvillæa. This valley is one of the deepest with which I am acquainted in the chain of the Andes: it is a true transverse valley directed from east to west, deeply cleft, and hemmed in on the two sides by the Altos de Aroma and Guangamarca. In this valley recommences the same quartz formation which we had observed in the Paramo de Yanaguanga, between Micuipampa and Caxamarca, at an elevation of 11720 English feet, and which, on the western declivity of the Cordillera, attains a thickness of several thousand feet, and was long an enigma to me. Since von Buch has shown us that the cretaceous group is also widely extended in the highest chains of the Andes, on either side of the Isthmus of Panama, the quartz formation which we are now considering, which has perhaps been altered in its texture by the action of volcanic forces, may be considered to belong to the Quadersandstein, intermediate between the upper part of the chalk series, and the Gault and Greensand. On quitting the mild temperature of the Magdalena valley we had to ascend again for three hours the mountain wall of 5120 English feet, opposite to the porphyritic group of the Alto de Aroma. The change of climate in so doing was the more sensible, as we were often enveloped in the course of the ascent in a cold fog.
