3. — Vigilante Experimental Range and 

 Monument Ridge in the Gravelly Range, Bea- 

 verhead National Forest, southwestern Mon- 

 tana. Elevations of the 84 study plots varied 

 between 7,000 and 9,500 feet. This is cattle 

 and sheep range dominated in many parts by 

 Idaho fescue (Festuca idahoensis) and in others 

 by native forbs or seeded grasses (mainly 

 Agropyron desertorum). Soils are mostly silt 

 loams and silty clay loams derived from red 

 shale, siltstone-shale, and glacial till. 



4. — Diamond Mountain Cattle Allotment 

 near Flaming Gorge, Ashley National Forest, 

 eastern Utah. This is an experimental grazing 

 area where portions of the native sagebrush- 

 grass vegetation has been replaced by intro- 

 duced grass species. Soils are loams and sandy 

 loams derived from sandstone. Elevations of 

 the 34 plots were about 8,000 feet. 



5. — The basalt study area north of Seven 

 Devils, Nezperce National Forest, central 

 Idaho. Forty -five study plots were located in 

 grassy openings in open ponderosa pine stands 

 at about 5,000 feet elevation. Soils are loams 

 and silt loams derived from basalt. 



6. — Coolwater Ridge, Nezperce National 

 Forest, central Idaho. Elevation of the 15 plots 

 was about 6,000 feet. Vegetation on this deteri- 

 orated subalpine range is predominantly forbs 

 of low value insofar as palatability and protec- 

 tion are concerned. The granitic soils are sandy 

 loams and loams. 



7. — Trinity Mountains, Boise National For- 

 est, southern Idaho. Forty study plots were lo- 

 cated in large and small openings in coniferous 

 forest at elevations of about 7,000 feet. The 

 granitic soils, typical of much of the Idaho 

 Batholith, are sandy loams and loamy sands. 



Uniform procedures were followed on all 

 460 plots involved in this analysis. The plots 

 were 20 inches wide and 30.5 inches long; ap- 

 proximately 1/10 milacre in area. Dortignac's 

 (1951) rain simulator was used to apply 2.5 

 inches of water to these plots at a constant in- 

 tensity of 5 inches per hour for 30 minutes. The 

 raindrops produced by this simulator tend to 

 be larger than those of actual thundershowers, 

 but have lower impact velocities. All runoff 

 from the plots was collected and measured. All 

 eroded mineral and organic material, including 

 that deposited in the collecting trough at the 

 bottom of the plot frame and that suspended in 

 the runoff, was ovendried and weighed. 



Density of cover on each plot was measured 

 with a point frame (Levy and Madden 1933). 

 First strikes at 100 evenly-spaced points were 



recorded as plants by species, litter, stone, or 

 bare soil. A day or two after the simulated rain 

 test, vegetation was clipped at the soil surface 

 and litter removed. Vegetation and litter were 

 air-dried and weighed. 



The following soil parameters were among 

 those measured: 



1. — Antecedent moisture content of the 



surface 2 inches of soil; 



2. — bulk density and capillary porosity of 



the surface 6 inches of soil; 



3. — organic matter content of the surface 



inch of soil by the dichromate method 

 (Peechl947); 



4. — particle-size distribution (Bouyoucos 



1962) and aggregation (Middleton 

 1930) in the surface inch of soil. 



3 



