ORIGINAL ARTICLE M. Zabed Hossain • Atsushi Okubo • Shu-ichi Sugiyama Effects of grassland species on decomposition of litter and soil microbial communities Received: 28 April 2009 / Accepted: 11 September 2009 / Published online: 16 October 2009 Ó The Ecological Society of Japan 2009 Abstract Decomposition of litter is greatly influenced not only by its chemical composition but also by activ- ities of soil decomposers. By using leaf litter from 15 plant species collected from semi-natural and improved grasslands, we examined (1) how interspecific differences in the chemical composition of litter influence the abundance and composition of soil bacterial and fungal communities and (2) how such changes in microbial communities are related to the processes of decomposi- tion. The litter from each species was incubated in soil of a standard composition for 60 days under controlled conditions. After incubation, the structure of bacterial and fungal communities in the soil was examined using phospholipid fatty-acid analysis and denaturing gradient gel electrophoresis. Species from improved grasslands had significantly higher rates of nitrogen mineralization and decomposition than those from semi-natural grass- lands because the former were richer in nitrogen. Litter from improved grasslands was also richer in Gram-po- sitive bacteria, whereas that from semi-natural grass- lands was richer in actinomycetes and fungi. Nitrogen content of litter also influenced the composition of the fungal community. Changes in the composition of both bacterial and fungal communities were closely related to the rate of litter decomposition. These results suggest that plant species greatly influence litter decomposition not only through influencing the quality of substrate but also through changing the composition of soil microbial communities. Keywords Decomposition Æ Denaturing gradient gel electrophoresis (DGGE) Æ Litter chemistry Æ Microbial community Æ Phospholipid fatty acids (PLFA) Introduction Plant species differ greatly in the decomposability of their litter (Cornelissen 1996; Grime et al. 1996; Corn- well et al. 2008), which is largely influenced by the quality of its substrate, which in turn is determined mostly by its chemical composition (Wardle et al. 2003; Bardgett 2005). High levels of nitrogen (N) and phos- phorus (P) enhance rates of microbial decomposition of litter and its mineralization (Enriquez et al. 1993) be- cause such litter contains nutrients in quantities surplus to what microbes require. In contrast, anti-herbivory chemicals such as phenolic compounds and tannins slow down the decomposition (Ha¨ttenschwiler and Gasser 2005) because they block the action of decomposing enzymes. Besides litter quality, the rate of decomposition (DR) is also affected by microbial communities (Hector et al. 2000). Orwin et al. (2006) showed experimentally that sources of carbon in the substrate and their diversity alter the structure of the soil bacterial community. Therefore, litter quality can influence the DR of litter in soil in two ways: (1) directly, through its decompos- ability as a substrate, which is a function of its compo- sition, and (2) indirectly, through altering the structure and functioning of decomposer communities. Although many studies have examined the direct effects of litter quality on decomposition (e.g., Cornelissen 1996; Aerts and de Caluwe 1997), few have addressed the indirect effects—those of litter quality on soil microorganisms, and of the consequent changes in soil biota—on the DR. To explore the relationships among litter chemistry, soil microbes, and decomposition processes, leaf litter from 15 herbaceous plants from semi-natural and im- proved grasslands was incubated in soil and subsequent changes in litter decomposition and microbial commu- M. Z. Hossain Æ A. Okubo Æ S. Sugiyama (&) Faculty of Agriculture and Life Science, Hirosaki University, 3 Bunkyo-cho, Hirosaki, Aomori 036-8561, Japan E-mail: sugi@cc.hirosaki-u.ac.jp Present address: M. Z. Hossain Department of Botany, University of Dhaka, Dhaka 1000, Bangladesh Ecol Res (2010) 25: 255–261 DOI 10.1007/s11284-009-0648-8