EFFECTS OF 3 SOIL-APPLIED HERBICIDES ON POPULATIONS OF PLANT-DISEASE SUPPRESSING BACTERIA IN THE COTTON RHIZOSPHERE

The potential impact of three widely used herbicides, pendimethalin, p rometryn, and trifluralin, on populations of five plant disease suppre ssing bacterial isolates (three isolates of Pseudomonas fluorescens an d two isolates of Burkholderia cepacia) in the rhizosphere of cotton s eedlings was investigated. All isolates are efficient cotton root colo nizers and each is capable of suppressing a plant disease. In microcos m experiments, application of each of the test herbicides at the rates of 1, 2, and 4 mu g active ingredient (a.i.) g(-1) soil caused signif icant (p<0.05) reductions in populations of most of the isolates in th e rhizosphere, 14 days after the release of bacteria into the soil by seed coating. The responses of the isolates to the herbicides varied d epending on the isolate and the type and concentration of the herbicid es. In microcosm experiments the impact of pendimethalin, prometryn, a nd trifluralin at the respective concentrations of 2.4, 3.6, and 1.8 m u g a.i. g(-1) soil on the population of isolate D1 in the cotton rhiz osphere declined with time during a four week period of monitoring fol lowing the release of the isolate into the soil by seed coating. The i mpact of soil applied test herbicides on the population sizes of D1 in cotton rhizosphere was also studied in two field experiments (Safford and Tucson, Arizona) where the bacteria were added as a soil drench. In the Safford experiment pendimethalin and prometryn, but not triflur alin, caused significant (p<0.05) reductions in the population of the bacterium 15 days after sowing. In the Tucson experiment a significant (p<0.05) reduction in the population of the bacterium was observed 15 and 25 days after sowing in soils treated with pendimethalin and prom etryn and 25 days after sowing in soils treated with trifluralin.