Influence of Environment and Stage of Growth on Honey Mesquite (Prosopis glandulosa) Response to Herbicides

Honey mesquite (Prosopis glandulosaTorr. # PRCJG) was treated with four herbicides at 1.1 kg ae/ha at 14 dates from April 30 through September 23 over a 2-yr period. Overall, clopyralid (3,6-dichloro-2-pyridinecarboxylic acid) was the most effective herbicide, killing 80% or more of the honey mesquite at most dates. Picloram (4-amino-3,5,6-trichloro-2-pyridinecarboxylic acid), triclopyr {[(3,5,6-trichloro-2-pyridinyl)oxy] acetic acid}, and 2,4,5-T [(2,4,5-trichlorophenoxy)acetic acid] were less effective than clopyralid. Honey mesquite control was correlated with Julian day of year sprayed, 15 plant, 11 environmental, and 8 herbicide concentration variables. Most effective control was negatively correlated with Julian day of year sprayed and new xylem ring thickness. Control was positively correlated with total daily apparent photosynthesis, upward methylene blue dye movement in the xylem, and herbicide concentration in the xylem 3 days after spraying. The only environmental variable of 11 measured that was significantly correlated with honey mesquite control was soil water and that was only with picloram. The concentration of herbicide in the stem was highest for clopyralid, followed by picloram, and then by 2,4,5-T and triclopyr, which were about equal. Variables most useful in regression equations for predicting control with herbicides were total daily apparent photosynthesis and rate of upward movement of methylene blue dye. Variables most incorporated into regression equations relating honey mesquite control with herbicide content were herbicide concentrations in the xylem 3 or 30 days after treatment.