We have been investigating the multiple controls on community assembly and restoration in 4.5 ha grassland restoration. The site is a former hayfield and it was treated with two applications of glyphosate herbicide prior to fall planting. The planting mixtures included three C3 graminoids and three C4 grasses randomly chosen from a pool of 10 species. We factorially added zero, eight, or 16 forbs (from a pool of 20) and zero, four, or eight legumes (from a pool of 10) to yield nine functional mixtures planted into 45 0.1 ha plots. We nested a factorial combination of chlorothalonil fungicide (which reduced MF colonization by 40%) and nitrogen (as ammonium nitrate). After three growing seasons, native and non-native species richness was positively correlated at both 1 m2 and 246 m2 grain size. MF suppression and nitrogen decreased the richness of both native and non-native species at both grain sizes. They also decreased the cover by both native species and target (planted) species. Target legumes rarely established. Even though target forbs had strong establishment, there was no major difference between planting 8 or 16 species. This led to an overall reduced floristic quality in the MF suppressed and added nitrogen communities. In these communities, functional composition was strongly dominated by non-native c3 grasses, with relatively greater cover by native forbs, legumes, and woody plants in the control plots. One C4 grass (Sorghastrum nutans) had increased establishment when MF were suppressed.