When attempting to re-establish native plant communities on degraded lands, seeds are sown or planted using "well-mixed" or uniform mixtures. However, such seed mixtures rarely occur in natural plant communities, where most plants occur near conspecifics due to aggregated seed dispersal. Additionally, soil conditions often vary in unknown ways potentially influencing the establishment success of sown plant species. We tested how seed distributions (uniform versus aggregated) and soil heterogeneity influence species richness in plant communities in a grassland in south-central Kansas. We established 96, 4x4.8 m plots each divided into 120 large-scale patches (0.4x0.4 m) or 480 small-scale patches (0.2x0.2 m). We then excavated, mixed, and redistributed soil within each plot to create either homogenous or heterogeneous soil conditions. Seeds were sown with either one species per patch, to create aggregated species distributions within the plot, or were sown uniformly throughout the plot. After two years, we found that species richness did not vary between plots with homogenous or heterogeneous soils and that the responses were similar in plots with large or small patch sizes. However, plots with aggregated seed sowing had higher species richness than plots with uniform sowing. These results suggest that utilizing aggregated rather than uniform seed dispersal may be important to enhance local species richness when restoring prairie ecosystems.