212 GEOLOGICAL SURVEY OF THE TERRITORIES. 



retains no wind-marks. At the angle of 38^ the sand will rest for an 

 instant, but apparently holds no i)ermanent position Avhen the angle of 

 slope is greater than 33°. The upper and lower edges of the dunes 

 are marked by dense thickets of sage-brush. Above the dunes the 

 base of the mountain shows solid foundations of trachytes, covered with 

 considerable groves of cedar, from 12 to 15 feet high. Then follow cov- 

 ered slopes, thickly beset with J erf^ey tea, {Cmnothus velufinus) and sumac, 

 [BJms glaher,) and spotted with blossoms of Geranium Richardsonii, and 

 various Oompositse and Umbelliferaj, and, finally, irregular masses of 

 coarselj'-porous trachyte cap the crests at the height of about 1,200 feet 

 above the valley. The range is a double one, the southern j)ortion being 

 the longer and larger, which runs about northeast and southwest for 

 perhaps ten miles, with a moderate' southern slope, much broken into 

 long spurs, and a steep northern one, descending almost unbroken into 

 a valley perhaps 200 feet above the plain, and about a mile wide, which 

 opens westward, and has thence been filled with the barren drifting 

 sands. These have been blown into drifts reaching to the very crest of 

 the more northern ridge, which attains the same elevation as the other, 

 with generally steep southern slopes, but apparently somewhat spurred 

 on the northern, where it descends to the plain. The range, as a whole, 

 has no apparent connection with any other, the rocks have no definite 

 bedding, and I am inclined to believe that the two main ridges, facing- 

 each other with steep slopes, may be fragments of the bounding walls 

 of a huge crater, whose fires were extinguished long before the eruj)tion 

 at the Crater Buttes commenced, and even before the eruption of at 

 least the later of the basalt-layers which floor the plain. 



Looking sonthwestward from these crests, we see the rough, almost 

 impassable, basalt-plain, thickly overgrown with sage-brush, stretching 

 away toward ^larket Lake. To the west and northwest si)read the bar- 

 ren sands of the dunes, which are several miles in width ^\here crossed 

 by the stage-road, near Sand-Holes Station. These also extend north- 

 ward, and their eastern edge forms a ridge from 100 to 150 feet high, 

 joining the Sand- Hill Mountains to the main divide of the Eocky Mount- 

 ains near Camas Creek, and forming the western border of the valley 

 of Henry's Fork. These absorbent sands, and the cracked and cavern- 

 ous basalt with sand and gravel foundations together, may well account 

 for the fact that, from near the head of Henry's Fork to the Malade 

 Eiver, which enters below the Great Shoshone Falls, a distance of fully 

 three hundred miles by the river, no stream joins the Snake by a sur- 

 face channel, though several good-sized ones reach the plain from the 

 Salmon Eiver Mountains. Along the bank of the Snake, however, at 

 several points, there are large cold springs escaping from the basalt ; 

 and, in the lower part of the region named, several large streams are 

 said to leap boldly from the walls of the canon, thus escaping from the 

 subterranean channels to which they have been confined for many miles. 



The low plain bordering Henry's Fork on the west is from two to 

 eight or ten miles in width, partly well gTassed, though with many dense 

 patches of sage-brush. About halfway between the Sand-Hill Mountains 

 and the river, there is a low hill, in the shape of a horseshoe, opening 

 to the south, which apparently was once a crater; but it has now been 

 so much worn away and covered with soil that no rock is .visible. 



Nearly opposite the Sand-Hill Mountains we crossed Henry's Fork, at 

 Eagle-Nest Ford. The river is here, perhaps, three hundred yards wide, 

 with from one to three feet of water, with i)ebbly bottom, a bluff bank 

 on the west side and a low one on the east, which is overflowed during 

 the spring freshets, so as to double the width of the stream. At that 



