road with sufficient dispatch to occasion but little delay in the movement of the column. After the first 150 miles had been accomplished but few creeks or rivulets into the river, although the gulches were more numerous and the bluffs approached the river more frequently than before, forcing the route into the hills. Aside from sand the road was good. When near the heard of the Loup wood entirely out, and the command was forced to resort to buffalo chips, of which a very limited quantity could be found. Bog peat was also as a substitute for wood, of which there was considerable quantities found on this part of the route.

I had many misgiving as to the practicability of moving a command as large as mine across the country from the head of the Loup to the Niobrara River, as my information was, in effect that the region to be traversed was a barren, sandy desert, without fresh water, wood, or grass. To be positive, I lay in camp one day for the purpose of scouting for best route. I sent two parties, one on Lieutenant Warren's trail, who returned reporting that route impassable for trains, with but little grass and ponds of brackish water. The other party sent to the north or most direct line to Nebraska River, reported a practicable route, with excellent grass and water in abundance. I adopted the new route, and by slowly my way with as I went, discovered a number of lakes and small streams, on the banks of which were excellent camping-grounds. The topography of the country consists of a series of sand ridges, sparsely covered with vegetation, and broad, intervening valleys, densely covered with grass. The latter all had lakes, streams, or marshes in them, yet forming no material obstacle to the movement of the command. In crossing these sand ridges, which varied from 200 to 500 feet in height, the deep, loose sand made the draught heavy on the mules, more particularly in the rear of the trains, necessitating the frequent use of the men, dismounted and pulling with drag-ropes. Scarcely and wood was found in this section until reaching Snake River, where there was an abundance of cedar. This is a rapid-flowing stream of excellent water, a tributary of the Niobrara River. The hills on the north side of Snake River are low and the face of the country grows gradually flatter until it becomes a level plain of eight to ten miles in breadth, ending at the crest of the bluffs on the Niobrara River. Where we struck the Niobrara timber was very plenty, the bottom land having the usual cotton wood and box elder, with scattering pines and cedars along the bluffs and in the canons. Grass was very scant, and with my large number of animals it was difficult to obtain sufficient for one night.

From the Niobrara out route lay over a gently rolling prairie toward the bluffs at the head of Wounded Knee Creek; from thence down the valley of that stream to White Earth River. The first night's camp was made on the head of Antelope Creek, a small marsh from which in a number of springs a considerable amount of water rises. These had to be dug out, and with a few small ponds adjacent furnished sufficient water for the entire command. Here there was no wood. On the Wounded Knee wood and water were abundant, but grass very scarce. In this vicinage dry beds of streams had to be bridged. This plan was resorted to when timber could be had in preference to cutting through embankments. The column was delayed for hours whilst roads for the trains were cut through the bluff, hard clay banks of the frequent gulches in this vicinity, leaving behind us as we passed a very good road, which, with but little repairs might, if desired, be again made available. Upon reaching White River we struck an American