Integrating new soybean varieties for soil fertility management in smallholder systems through participatory research: Lessons from western Kenya M. Misiko a,c, * , P. Tittonell b,c , J.J. Ramisch c,d , P. Richards a , K.E. Giller b a Technology and Agrarian Development (TAD), Social Sciences, Wageningen University, Hollandseweg 1, 6706KN Wageningen, The Netherlands b Plant Production Systems (PPS), Plant Sciences, Wageningen University, Postbus 430, 6700 AK Wageningen, The Netherlands c Tropical Soil Biology and Fertility Institute of the International Centre for Tropical Agriculture (TSBF-CIAT), United Nations Avenue, Gigiri, P.O. Box 30677-00100, Nairobi, Kenya d International Development and Globalisation, Social Sciences, University of Ottawa. Tabaret Hall, 550 Cumberland Street, Ottawa, Ontario, Canada K1N 6N5 Received 11 March 2007; received in revised form 25 September 2007; accepted 12 October 2007 Available online 4 January 2008 Abstract The aim of this paper was to understand the process of selecting soybean (Glycine max [L.] Merr.) promiscuous varieties by small- holders for soil fertility management in western Kenya. Eight varieties were screened on 2.5 m · 3 m plots that were managed according to farmers’ practices and evaluated through participatory monitoring and evaluation approaches. Farmers selected preferred varieties and explained their reasons (criteria) for making the selections. Seven promiscuous varieties had better yields than a local one. Farmers’ selection criteria fell into three broad categories relating to yield, appearance and labour. Selection criteria were not primarily aimed to improve soil fertility. This created a challenge to embed the new varieties within the local farming systems for soil fertility improvement. This study shows that farmer criteria for selecting varieties overlapped with scientific procedures. We propose co-research activities tar- geted to strengthen farmer experimentation skills, their understanding on N addition, and the role of P. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Farmer selection criteria; Participatory variety screening; Farmer experimentation; Grain legumes 1. Introduction In addition to their direct food provision, grain legumes are an appropriate way of targeting soil fertility improve- ment on smallholder farms due to their capacity to fix atmospheric N 2 in symbiosis with rhizobial bacteria (e.g. Muhr et al., 2001; Snapp and Silim, 2002; Ojiem, 2006). In response to rapidly declining soil fertility and resultant low crop yields in sub-Saharan Africa, breeders have focused on developing improved soybean (Glycine max [L.] Merr.) varieties that fix a high proportion of nitrogen from the atmosphere. These cultivars have the capacity to fix N 2 in symbiosis with rhizobia occurring in the natural soil flora without inoculation with specific strains, and are known as promiscuous varieties (Giller, 2001). Promis- cuous varieties can also produce large amounts of biomass and are often referred to as dual-purpose because they not only provide grain and/or cash but also leave a net amount of N in the soil that can benefit subsequent crops (Mpeper- eki et al., 2000; Sanginga, 2003). Soybean grows well in tropical climates; it constitutes a source of high quality but inexpensive crop protein and can be used to produce cooking oil. In addition to improving soil fertility through biomass and N 2 -fixation, promiscuous www.elsevier.com/locate/agsy 0308-521X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.agsy.2007.10.002 * Corresponding author. Address: Technology and Agrarian Develop- ment (TAD), P.O. Box 8130, 6700 EW Wageningen, The Netherlands. Tel.: +31 317 485029; fax: +31 317 485616. E-mail addresses: michael.misiko@wur.nl, m.misiko@cgiar.org (M. Misiko). Available online at www.sciencedirect.com Agricultural Systems 97 (2008) 1–12