The sample of 850 points within each unit allowed rather precise estimation of the 

 proportion of the surface area of that unit falling within each category. For example, 

 the true proportion of a unit in one ground-cover category would lie between 2 and 8 

 percent (at the 95-percent level of confidence) if the sample estimate was 5 percent, 



RESULTS 



By the end of the second season of growth, units treated with water, fertilizer, 

 and seed had noticeably more grass and other herbaceous vegetation than units that had 

 been treated only with seed (control units) . Differences between the other treatments 

 were recognizable but less obvious. In general, statistical evaluation of the measure- 

 ment data confirmed these observations. 



.-Analysis of variance of the data showed that the covariant, visitor use, was non- 

 significant. Therefore, the means for various treatments did not require adjustment for 

 differences in visitor use between units. The model accounted for 95 percent of the 

 total variation in measurements of vegetation growth (R^) . The majority--85 percent of 

 the total--was attributed solely to the main effects of treatment. In other words, 

 treatment accounted for the major portion of all squared differences from overall mean 

 growth. 



Tests of significant differences between the treatment means by Scheffe Test (Ban- 

 croft 1968) indicated no differences between the sites. That is, treatment responses 

 were not influenced by site. Treatment responses were found to be markedly different 

 from one another at the 95-percent level of significance. 



(1) Most important, the water-fertilizer-seed combination produced significantly 

 more vegetation than any of the other treatments. 



(2) Water and seed produced significantly more than seed alone. 



(3) Fertilizer and seed together were not significantly different from seed alone 

 (control) . 



The average change in proportion of ground cover vegetation for each of the four treat- 

 ments is listed below and presented in figure 4. 



Treatment Growth 



Water, fertilizer, seed 40 

 Water, seed 11 

 Seed, fertilizer 6 

 Seed -1 



The control units, which show a slight decline in ground cover vegetation since the 

 campground was reopened for use in 1968, will probably stabilize at some rather low 

 level of vegetation. 



Change in percent of vegetative cover of total available area. For example, the 

 seed (control) treatment decreased from 8 percent in September 1967 to 7 percent in 

 September 1969. 



7 



