Enhanced biodegradation of organic xenobiotic compounds in the rhizosphere is frequently recorded although the specific mechanisms are poorly understood. We have shown that the mineralization of 2,4-dichlorophenoxyacetic acid (2,4-D) is enhanced in soil collected from the rhizosphere of Trifolium pratense[e.g. maximum mineralization rate=7.9 days-1 and time at maximum rate (t1)=16.7 days for 12-day-old T. pratense soil in comparison with 4.7 days-1 and 25.4 days, respectively, for non-planted controls). The purpose of this study was to gain a better understanding of the plant-microbe interactions involved in rhizosphere-enhanced biodegradation by narrowing down the identity of the T. pratense rhizodeposit responsible for stimulating the microbial mineralization of 2,4-D. Specifically, we investigated the distribution of the stimulatory component(s) among rhizodeposit fractions (exudates or root debris) and the influence of soil properties and plant species on its production. Production of the stimulatory rhizodeposit was dependent on soil pH (e.g. t1 for roots grown at pH 6.5 was significantly lower than for those grown at pH 4.4) but independent of soil inorganic N concentration. Most strikingly, the stimulatory rhizodeposit was only produced by T. pratense grown in non-sterile soil and was present in both exudates and root debris. Comparison of the effect of root debris from plant species (three each) from the classes monocotyledon, dicotyledon (non-legume) and dicotyledon (legume) revealed that legumes had by far the greatest positive impact on 2,4-D mineralization kinetics. We discuss the significance of these findings with respect to legume-rhizobia interactions in the rhizosphere.