2o8 Roosevelt Wild Life Bulletin 



hold it, and then more water goes out around the ends and the 

 process is repeated. I have but little doubt that many of the long, 

 crooked dams which we see were thus built, not by design but by 

 this continual effort to stop the leaking over and around the dams. 



Not all dams are built in streams. Some are built across gulches 

 or on the sides of the valley, to control the water from springs. The 

 largest pond I examined near Camp Roosevelt lies between Crescent 

 Hill and a low ridge to the east (figure 42). It is 800 feet long by 

 340 feet wide, and is controlled by a comparatively short dam 165 

 feet long. This pond is supplied by springs, no surface water flow- 

 ing into it except in the spring of the year when the snow is melting. 

 There is quite a series of dams and small ponds below the large 

 one, and several hundred feet downstream is a new dam which in 

 time may make a fair sized pond, for the builders have selected the 

 most suitable site, where the rather wide valley narrows somewhat, 

 making it possible for a comparatively short dam to back up con- 

 siderable water. 



Also on Tower Creek, about two miles above Tower Falls, is an 

 extensive series of ponds deriving their water supply from a very 

 large spring on the flat ground scarcely one hundred feet from the 

 stream and but a few feet above its level. A low dam had been built 

 across the lower side of the spring, which is now about fifty feet in 

 diameter, and measures eight feet deep. 



Let us return to the building of the dam. On the lower face are 

 placed many sticks, often those from which the bark has been eaten, 

 or willow branches are used, and I have seen quite good sized logs 

 utilized, whose ends projected high above the dam (figure 56). 

 Whatever the material, these sticks are generally placed up and down 

 the face, not transversely (figure 57). There is invariably, I think, 

 a trail over the dam where the beavers cross, and this is always where 

 the stream is, so that they may go down into the water from above. 



It seems to me likely that in beginning a dam some of the branches 

 are customarily laid across the current, for I have seen a number of 

 dams which appear to have been thus underpinned. Moreover, when 

 a dam is cut through, whether by man or by natural agencies, there 

 are always the ends of sticks showing in the cross section thus 

 exposed, indicating that they are deliberately laid crosswise (figure 

 58). 



Dams are not always built completely across the stream. I recall 

 a series of three dams in Colorado, none of which extended all the 



