Plot size . The rectangular sample plots measuring 27.5 feet X 13.5 feet 

 (8.4 m X 4.1 m) and covering areas of 371 feet^ (34.5 m^) represented com- 

 promises in sampling technique. Although elongate plots are superior to 

 square plots for sampling (Clapham 1932), a problem arises when the plot 

 cannot be viewed as a whole (Daubenmire 1968). Also, although these rather 

 large plots cannot provide the precision available when using many smaller 

 plots, they can be sampled quickly and enclose many plants with low coverage. 

 Coverage can De accurately estimated for these plots when cover classes are 

 used. 



Site factors . Investigators recorded general descriptors such as lo- 

 cation, plot number, aerial photograph number, dominants, and observations 

 at each sample plot. A color photograph of each plot was taken with the 

 plot number appearing in the picture. These slides proved to be a useful 

 reference during classification. 



The investigators sampled the upper 4 inches (10 cm) of the soil at 

 three locations in each plot using a 1.5 inch (4 cm) diameter tubular sam- 

 pling device. The three samples from each plot were grouped into a single 

 sample. A soil scientist later determined the texture of each sample by 

 the feel method. Mechanical analysis of forty-six soil samples indicated 

 that the accuracy of the feel method was adequate. 



This determination of surface texture leaves much to be desired. A 

 soil profile description of textures to the rooting depth would be pre- 

 ferable. However, work on the soil component of the baseline study has not 

 been initiated to date. 



The slope aspect of each plot was measured by facing in the direction in 

 which water would run off the plot and shooting azimuth with a compass. 



Slope was measured in degrees normal to the contour using a clinometer 

 or a abney level . . • . 



The horizontal configuration was recorded as convex, straight, undulating, 

 or concave. 



Slope position was recorded as bottom, low, mid, upper, or top. 



Grazing pressure, which is simply a subjective estimate of intensity or 

 current livestock use and does not necessarily reflect past use, was record- 

 ed at each plot. Ratings of light, moderate or heavy were planned, but the 

 additional categories of light moderate, moderate-heavy, and severe appeared 

 on some data forms, reflecting a continuum of grazing intensities. 



The distance in air kilometers livestock would actually have to travel 

 to get water under average precipitation conditions was measured for each 

 plot. 



Precipitation was to be taken from 1:250,000 precipitation maps, but it 

 was not used because it seems that, in the study area, precipitation extremes 

 have a greater effect on vegetation than average precipitation. Annual pre- 

 cipitation over the study area is highly erratic as Figure 1 shows. 

 Lommasson (1947) documents the variation of precipitation in eastern Montana. 



n 



