The survival of genetically altered Pseudomonas putida strains harboring an inducible plasmid, pR0101, or a constitutive plasmid, pR0103, was compared. These plasmids encode for the degradation of 2,4-dichlorophenoxyacetate (TFD) to 2-chloromaleylacetate, and the maintenance of either plasmid did not alter survival of P. putida PPO301(pRO101) or PPO301(pRO103) in an unamended agricultural soil. Moreover, in TFD-amended soil, survival of PPO301, PPO301(pRO101), and PPO301(pRO103) was statistically the same after 50 days. Reapplication of TFD to soil 50 days after the original application did not change the numbers of PPO301(pRO101) or PPO301(pRO103), which cannot use TFD as a sole source of carbon. However, a mutant strain, PPO301KS(pRO101), which is able to use TFD as a sole source of carbon, was stimulated by the second addition of TFD:PPO301KS(pRO101) cfu/g soil increased by greater than 20-fold. Although the microbiota indigenous to the study soil was capable of degrading TFD, the addition of plasmid-bearing PPO301 had a dramatic effect on TFD degradation. In a parallel study, Raphanus sativus (radish) seeds failed to germinate in uninoculated and PPO301-inoculated soil amended with 500 ppm TFD. Seed germination was 53 and 80% in soils inoculated with PPO301(pRO101) and PPO301(pRO103), respectively (P>0.001). However, the difference in the rate of TFD degradation between the native soil and soil inoculated with plasmid-bearing P.putida was probably related to the relatively high inoculum density of P.putida strains (10 to the eighth power cfu) and the relatively low population density of TFD metabolizers indigenous to the soil.