no PATAGONIAN EXPEDITIONS: NARRATIVE. 



haps a quarter of a mile in width. On the opposite side there appeared 

 a steep but gradual slope from the bottom of the canon to the high pampa, 

 where we should be out of the limits of the basalt. In front was a practi- 

 cal route by which we could descend to the bed of the canon, so that in 

 a few moments we found ^ourselves at the bottom, which opened out 

 above into such a splendid meadow of beautiful waving grass, fresh and 

 green, with an abundance of pure sweet water, that we could not resist 

 the temptation to turn out and pass the remainder of the day and night 

 here, where our horses could be on such excellent feed and water and find 

 time to recuperate after their hard pull up the steep, rocky slope of the 

 river bluff strewn with broken fragments of basalt. Moreover I wished to 

 spend a few hours in observing the basalts and studying their relations 

 to the underlying sedimentary rocks. I had noticed during the morning 

 that between the valley of the river and the high pampa which lay some 

 miles to the northward there was a series of terraces, and that each of 

 these down to a certain level was capped with basalt and appeared like 

 successive basaltic platforms. I was interested to know whether these 

 represented successive flows at different intervals, or one flow which had 

 taken place at a comparatively late period after the river and its tributaries 

 had carved their valleys down at least to the level of the lowermost of 

 the lava-capped benches. I discovered that in this locality at least there 

 had been but the one flow, and at different points along the sides of the 

 small canon where sufficient erosion had been effected at opportune places, 

 the different layers of lava could be seen as they flowed over the crest 

 and down the slopes of the various inclines. By following up the canon 

 for the distance of a mile the little park in which we were camped was 

 seen to terminate suddenly, The walls contracted on either side until 

 the park was reduced to a deep, narrow canon only a few yards in width. 

 At the summit of this extremely narrow gorge there were two narrow, 

 basalt-covered platforms extending back to the slopes of the broader main 

 canon above. It was very evident that the position of the basaltic layer 

 at the top of the lowermost narrow gorge indicated the extent to which 

 erosion had taken place at the time of the outflow, while the gorge 

 beneath had been entirely cut subsequent to that time. On ascending to 

 the summit of the high lava-capped plain to the westward, I found the 

 surface deeply riven, presenting an extremely rough and broken sur- 

 face, with deep crevices and great yawning caverns. Nowhere, however, 



