AbstractThe responses of three different soil microbial communities to the experimental application of 2, 4‐dichlorophenoxyacetic acid (2, 4‐D) were evaluated with a variety of molecular genetic techniques. Two of the three soil communities had histories of prior direct exposure to 2, 4‐D, and one had no prior direct application of any herbicide. Dominant 2, 4‐D degrading strains isolated from these soils the previous year were screened for hybridization with three catabolic genes (tfdA,tfdAII, andtfdB) cloned from the well‐studied 2, 4‐D degradative plasmid, pJP4, revealing varying degrees of similarity with the three genes. Hybridization of total community DNA from the three soils with thetfdgene probes also indicated that pJP4‐liketfdgenes were not harboured by a significant percentage of the community. Community level response was evaluated by the comparison of different treatments by Random Amplified Polymorphic DNA (RAPD) fingerprints and by community DNA cross‐hybridization. No differences between treatments within the same soil were detected in any of the RAPD fingerprints generated with 17 primers. Community DNA cross‐hybridization also indicated that the application of 2, 4‐D at the applied rates did not quantitatively affect the structure of the soil microbial communities present in the three soils during the