Greenhouse and laboratory experiments were conducted to investigate mechanisms of glyphosate resistance in horseweed populations from Mississippi, Arkansas, Delaware, and Tennessee. A nondestructive leaf-dip bioassay was developed to confirm resistance and susceptibility in individual test plants. A single leaf was excised from each plant, and the petiole and bottom one-fourth of leaf was dipped in a 600 mg ae L⁻¹ glyphosate solution for 2 d followed by visually estimating the injury on a scale of 0 to 10. Plants were classified as resistant (R) if the score was 2 to 3 and susceptible (S) if the score was 5 to 6. ¹⁴C-glyphosate solution was applied on the adaxial surface of a fully expanded leaf of the second whorl of four-whorl rosette plants. Plants were harvested 48 h after treatment and radioactivity was determined in treated leaf, other leaves, crown, and roots. Absorption of ¹⁴C-glyphosate was similar (47 to 54%) between R and S plants from within and among the four states, suggesting absorption is not involved in glyphosate resistance. The amount of radioactivity translocated from the treated leaf was reduced in R plants compared with S plants. The reduction in translocation of ¹⁴C-glyphosate ranged from 28% in Mississippi-R biotype to 48% in Delaware-R biotype compared with their respective S biotypes. Epicuticular wax mass ranged from 6 to 80 μg cm⁻² among horseweed biotypes, with no differences between R and S biotypes within each state. Treating two leaves with glyphosate solution at the field use rate (0.84 kg ae ha⁻¹) killed S plants but not R plants (38 to 58% control) regardless of state origin. These results suggest that a simple bioassay can be used to screen biotypes for suspected resistance and that reduced translocation of glyphosate plays a major role in glyphosate resistance in R biotypes of horseweed.

Nomenclature: ¹⁴C-glyphosate; glyphosate; horseweed, Conyza canadensis (L.) Cronq.