THE PIRACY OF THE YELLOWSTONE 265 



by distance into a simple roadside ditch. With this view, it is 

 easy to see that the canyon is not all the same age. The north 

 half of it is distinctly older than the south or upper half. In 

 the north half the shoulders are markedly rounded, the walls less 

 steep, the stream at bottom has long ago found an axial equi- 

 librium with the material it has to handle, and is not deepening 

 or widening its bed in any striking way. It is a surprise to 

 notice, too, that Broad Creek, which empties into the Yellow- 

 stone River just at the east foot of Mt. Washburne, has a canyon 

 every whit as wide, as deep, and with shoulders as rounded as 

 has the main canyon at this point. 



One cannot help wondering why the Yellowstone Canyon is 

 so young only above this point ; why the deep stratified clays in 

 Hayden Valley ; why the terrace and cliffs at the high level in 

 Hayden Valley. Why did the Yellowstone Lake abandon a 

 good outlet at Overlook Mountain, and flow off to the north ? 

 The explanation may be read from the correlation of the availa- 

 ble data as follows. 



The Yellowstone Canyon for five miles or so below the falls 

 is extremely young, the occupation by the river representing only 

 a fraction of postglacial time. On the recession of the ice from 

 the region, the plateau of rhyolite stretched untouched by the 

 river action, from the south base of Mt. Washburne southeast 

 across the site of the present canyon, at the general plateau level 

 of about eight thousand feet. There was no canyon, and no 

 Yellowstone River there. The two depressions in the plateau, 

 Hayden Valley, and the present lake basin, if they existed in 

 preglacial time, outflowed by some other route, at present 

 unknown. On the recession of the ice from the region, these 

 basins overflowed to the west, over available cols. Possibilities 

 of such drainage lines, besides the one mentioned on the road to 

 the " Thumb," may be suspected at A, B, C, and D, on the map, 

 Fig. 1. But the one which established itself for greatest perma- 

 nence was the one described at Overlook Mountain. 



Now taking the topographic map and supplying a shore line 

 for a lake outflowing at this channel, the surprising fact is shown 



