grasses. The apparent absence of slender wheatgrass 

 from the study plots in year 1 may have resulted from 

 sampling error; seedlings were extremely difficult to iden- 

 tify during the first year, and slender wheatgrass may 

 have been misidentified or placed in an Agropyron species 

 category. Spike trisetum was not observed in the sample 

 plots in year 1 or year 3. Both muttongrass and tufted 

 hairgrass were present in a high number of sample plots 

 in year 3. The seeded native forbs all occurred in the 

 study plots by year 3, but only western yarrow was found 

 in a high number of sample quadrats. Establishment of 

 the seeded native shrubs was poor with only a few indi- 

 viduals of Saskatoon serviceberry and Woods rose occur- 

 ring in the sample plots. Colonizer dominants in year 1 

 were alkali sandspurry, spreading groundsmoke, and 

 prostrate knotweed. By year 3 individuals of winter 

 bentgrass, draba, common fireweed, aster, common dan- 

 delion, and lodgepole pine were also recorded. 



In general, the effects of fertilizer on individual species 

 were the same as those observed in the combined species 

 analyses. Standing crop biomass of most species was 

 higher in fertilized than in unfertilized plots in year 1. In 

 year 3 standing crop biomass was either less in fertilized 

 than in unfertilized plots or exhibited a relatively lesser 

 increase in fertilized than in unfertilized plots. 



Cover 



Results of the cover data analysis followed the same 

 trends as those of the standing crop biomass analysis 

 (fig. 3). There were no significant differences in the cover 

 classes that were attributable to grass seeding density. 

 Fertilization resulted in higher vegetation cover and 

 lower bare ground cover in the combined 3-year analysis 

 (both, p < 0.001). Differences existed among years for all 

 of the cover classes, except vegetation (p < 0.05 to 

 p < 0.001). The response to fertilizer was not consistent 

 over time. In treatments that were unfertilized, vegeta- 

 tion cover increased between years 1 and 3, while litter 

 and bare ground remained unchanged and rock increased 

 (p < 0.05). Treatments that were fertilized showed signifi- 

 cant yearly decreases in vegetation cover and increases in 

 litter, bare ground, and rock (p < 0.05). 



Soil Properties 



Two years after the study plots had been fertilized, 

 there were no significant differences in either soil N0 3 -N 

 or exchangeable K between fertilized and unfertilized 

 treatments (table 3). However, significantly higher levels 

 of available P existed in fertilized than in unfertilized 

 treatments (p < 0.01). 



8 °- * 



E2 VEGETATION 



I 1 BARE GROUND 



^3 LITTER 



ROCK & GRAVEL 



4 



NO NO NO 



FERTILIZER FERTILIZER FERTILIZER FERTILIZER FERTILIZER FERTILIZER 



NATIVES ONLY 



NATIVES + LOW GRASS 



NATIVES + HIGH GRASS 



Figure 3— Percentage cover by cover class for 

 1985, 1986, and 1987 in research plots that 

 were seeded with natives and a zero, low, or 

 high level of introduced grasses and then 

 either fertilized or not fertilized. Values are 

 mean ± S.E., n = 30. 



6 



