Agroforestry-based biofuel production has recently been proposed as a rural development strategy in the South. However, there exists a complete lack of empirical evidence on farmer adoption rates and determinants for these novel systems. This study describes adoption rates of oilseed tree mixtures on smallholder farms in Hassan district, South India, and quantifies how these rates are determined by a biofuel extension program (BP) and farm(er) characteristics. This is done through a set of regression-based analyses, addressing various forms of selection bias. The findings reveal that although 60% of the farmers cultivate oilseed trees, oilseed collection rates are generally low (13%), and the adoption of both practices is driven by different determinants. More specifically, BP activities are found to stimulate tree cultivation and therefore agroforestry establishment, but not seed collection and biofuel production. This calls for a better understanding of adoption profitability in function of the opportunity costs of land, labour and capital involved, and conditional on these results for intensifying BP activities and value chain development. Acknowledgement : Thanks go to the UAS Bangalore Department of Agricultural Extension and all individual enumerators for their assistance in data collection and data entry. We would also like to thank the Biofuel Park Hassan staff for their scientific support and for their assistance in data collection. This work was supported by the Research Foundation Flanders (FWO) [Aspirant PhD grant].