N 2 fixation in three perennial Trifolium species in experimental grasslands of varied plant species richness and composition Georg Carlsson Æ Cecilia Palmborg Æ Ari Jumpponen Æ Michael Scherer-Lorenzen Æ Peter Ho ¨gberg Æ Kerstin Huss-Danell Received: 28 October 2008 / Accepted: 12 March 2009 / Published online: 27 March 2009 Ó Springer Science+Business Media B.V. 2009 Abstract This study is the first to investigate quantitative effects of plant community composition and diversity on N 2 fixation in legumes. N 2 fixation in three perennial Trifolium species grown in field plots with varied number of neighbouring species was evaluated with the 15 N natural abundance method (two field sites, several growing seasons, no N addition) and the isotope dilution method (one site, one growing season, 5 g N m -2 ). The proportion of plant N derived from N 2 fixation, pNdfa, was generally high, but the N addition decreased pNdfa, especially in species-poor communities. Also follow- ing N addition, the presence of grasses in species-rich communities increased pNdfa in T. hybridum and T. repens L., while legume abundance had the opposite effect. In T. repens, competition for light from grasses appeared to limit growth and thereby the amount of N 2 fixed at the plant level, expressed as mg N 2 fixed per sown seed. We conclude that the occurrence of diversity effects seems to be largely context dependent, with soil N availability being a major determinant, and that species composition and functional traits are more important than species richness regarding how neighbouring plant species influence N 2 fixation in legumes. Keywords BIODEPTH Biodiversity Clover Functional traits 15 N natural abundance 15 N isotope dilution Introduction Biological N 2 fixation in symbioses between legumes and rhizobia provide valuable inputs of N that becomes available to neighbouring as well as succeeding plants (Høgh-Jensen and Schjørring 1997; Mulder et al. 2002; Spehn et al. 2002; Stopes et al. 1996; Temperton et al. 2007). Results from biodiversity manipulations in natural meadow G. Carlsson C. Palmborg A. Jumpponen K. Huss-Danell Department of Agricultural Research for Northern Sweden, Swedish University of Agricultural Sciences (SLU), 901 83 Umea ˚, Sweden A. Jumpponen P. Ho ¨gberg Department of Forest Ecology and Management, Soil Science Section, Swedish University of Agricultural Sciences (SLU), 901 83 Umea ˚, Sweden A. Jumpponen Division of Biology, Kansas State University, 433 Ackert Hall, Manhattan, KS 66506, USA M. Scherer-Lorenzen ETH Zu ¨rich, Institute of Plant Sciences, LFW C55.2, 8902 Zu ¨rich, Switzerland G. Carlsson (&) INRA-SupAgro, UMR 210 Eco&Sols, 2 Place Pierre Viala, 34060 Montpellier Cedex 1, France e-mail: carlsson@supagro.inra.fr 123 Plant Ecol (2009) 205:87–104 DOI 10.1007/s11258-009-9600-9