RESEARCH ARTICLE Seeding Method Influences Warm-Season Grass Abundance and Distribution but not Local Diversity in Grassland Restoration Kathryn A. Yurkonis, 1,2 Brian J. Wilsey, 1 Kirk A. Moloney, 1 Pauline Drobney, 3 and Diane L. Larson 4 Abstract Ecological theory predicts that the arrangement of seedlings in newly restored communities may influence future species diversity and composition. We test the pre- diction that smaller distances between neighboring seeds in drill seeded grassland plantings would result in lower species diversity, greater weed abundance, and larger con- specific patch sizes than otherwise similar broadcast seeded plantings. A diverse grassland seed mix was either drill seeded, which places seeds in equally spaced rows, or broadcast seeded, which spreads seeds across the ground surface, into 24 plots in each of three sites in 2005. In summer 2007, we measured species abundance in a 1 m 2 quadrat in each plot and mapped common species within the quadrat by recording the most abundant species in each of 64 cells. Quadrat-scale diversity and weed abundance were similar between drilled and broadcast plots, suggest- ing that processes that limited establishment and controlled invasion were not affected by such fine-scale seed distribu- tion. However, native warm-season (C 4 ) grasses were more abundant and occurred in less compact patches in drilled plots. This difference in C 4 grass abundance and distribu- tion may result from increased germination or vegetative propagation of C 4 grasses in drilled plots. Our findings suggest that local plant density may control fine-scale het- erogeneity and species composition in restored grasslands, processes that need to be further investigated to determine whether seed distributions can be manipulated to increase diversity in restored grasslands. Key words: broadcast seeding, drill seeding, grassland restoration, heterogeneity, invasion, reconstruction, slot seeding, spatial pattern, tallgrass prairie. Introduction The arrangement of seeds in space may strongly affect the ways newly established communities develop in time. This is an important aspect to consider when planning a restoration where several methods are available for incorporating seeds into a landscape (Wilson 2002). Seed arrangement may have unintentional and long-lasting effects on restoration success (Allison 2002; Bartha et al. 2004) as a result of potential interactions that determine which species persist and their subsequent spatial arrangement (Stoll & Prati 2001; Monzeglio & Stoll 2005; Korner et al. 2008). Although theory predicts the distribution of seeds during establishment may influence fine- scale heterogeneity and species diversity (reviewed in Tilman & Kareiva 1997; Bolker et al. 2003), studies have mostly 1 Iowa State University, Department of Ecology, Evolution, and Organismal Biology, 253 Bessey Hall, Ames, IA 50011, U.S.A. 2 Address correspondence to K. A. Yurkonis, email yurkonis@iastate.edu 3 United States Fish and Wildlife Service, Neal Smith National Wildlife Refuge, PO Box 399, 9981 Pacific Street, Prairie City, IA 50228, U.S.A. 4 United States Geological Survey, Northern Prairie Wildlife Research Center, 100 Ecology Building, 1987 Upper Buford Circle, St. Paul, MN 55108, U.S.A.  2010 Society for Ecological Restoration International doi: 10.1111/j.1526-100X.2009.00617.x considered this question in low-diversity, annual systems (e.g., Stoll & Prati 2001; Lortie et al. 2005; Monzeglio & Stoll 2005). Only a few studies (De Luis et al. 2008; Yurkonis et al. in press) have asked whether the initial spatial arrangement of seeds is related to subsequent vegetation structure in species- rich perennial systems. Here, we ask whether plant community structure, as measured by species diversity, composition, and fine-scale plant distribution, differs between replicated drill and broadcast seeded grassland plots. Our goal in this study is to investigate possible mechanisms that control grassland establishment and fine-scale heterogeneity. Drill and broadcast seeded restorations differ in depth and distribution of seeds at planting and provide an interesting ecological context in which to study the effects of manipu- lating seed distributions in space. In drill seeded restorations, seeds are planted in equally spaced rows with short distances between neighboring seeds and large, uniform spaces between rows (Bufton 1978). In broadcast seeded restorations, seeds are dispersed across the soil surface with potentially longer and more variable distances between neighboring seeds (Packard & Mutel 1997; Skinner 2005). Only a handful of studies have systematically compared these seeding techniques for grass- land restoration and have yielded mixed, often site-specific, 344 Restoration Ecology Vol. 18, No. S2, pp. 344–353 NOVEMBER 2010