A couple of days after leaving Cottonwood Creek—­where we had spent several days—­we camped at the Yellowstone Canon below Tower Falls.  Here we saw a second band of mountain sheep, numbering only eight—­none of them old rams.  We were camped on the west side of the canon; the sheep had their abode on the opposite side, where they had spent the winter.  It has recently been customary among some authorities, especially the English hunters and naturalists who have written of the Asiatic sheep, to speak as if sheep were naturally creatures of the plains rather than mountain climbers.  I know nothing of old world sheep, but the Rocky Mountain bighorn is to the full as characteristic a mountain animal, in every sense of the word, as the chamois, and, I think, as the ibex.  These sheep were well known to the road builders, who had spent the winter in the locality.  They told me they never went back on the plains, but throughout the winter had spent their days and nights on the top of the cliff and along its face.  This cliff was an alternation of sheer precipices and very steep inclines.  When coated with ice it would be difficult to imagine an uglier bit of climbing; but throughout the winter, and even in the wildest storms, the sheep had habitually gone down it to drink at the water below.  When we first saw them they were lying sunning themselves on the edge of the canyon, where the rolling grassy country behind it broke off into the sheer descent.  It was mid-afternoon and they were under some pines.  After a while they got up and began to graze, and soon hopped unconcernedly down the side of the cliff until they were half way to the bottom.  They then grazed along the sides, and spent some time licking at a place where there was evidently a mineral deposit.  Before dark they all lay down again on a steeply inclined jutting spur midway between the top and bottom of the canyon.

[Illustration:  Mountain sheep at close quarters.]

Next morning I thought I would like to see them close up, so I walked down three or four miles below where the canyon ended, crossed the stream, and came up the other side until I got on what was literally the stamping ground of the sheep.  Their tracks showed that they had spent their time for many weeks, and probably for all the winter, within a very narrow radius.  For perhaps a mile and a half, or two miles at the very outside, they had wandered to and fro on the summit of the canyon, making what was almost a well-beaten path; always very near and usually on the edge of the cliff, and hardly ever going more than a few yards back into the grassy plain-and-hill country.  Their tracks and dung covered the ground.  They had also evidently descended into the depths of the canon wherever there was the slightest break or even lowering in the upper line of basalt cliffs.  Although mountain sheep often browse in winter, I saw but few traces of browsing here; probably on the