At a bottomland hardwood forest near Athens, GA, two non-soil-active herbicides (glyphosate and triclopyr) were tested at various rates (0, 1.5, 3.0, 4.5, and 6.0 lbs. a.e. per acre), and timings (April, June, August, October, and December 2000), to identify cost-effective methods for controlling Chinese privet (Ligustrum sinense Lour.) A second study tested different formulations (Accord SP vs. Roundup Pro Dry for glyphosate and Garlon 3A in water vs. Garlon 4 in JLB improved plus oil for triclopyr). A third study tested for the effects of trenching to separate root linkages on herbicide efficacy on privet. Immediately prior to and approximately one year following each treatment, privet cover (%) was estimated visually in each plot. The data from each study were subjected to analysis of covariance with pre-treatment cover as a covariate. Multiple comparisons of covariate-adjusted means ( =0.05) were performed. The first year results from the study testing herbicides, rates and timing showed that timing and herbicide interaction was significant. For glyphosate, October and December timings were more effective. For triclopyr, the December treatment was as effective as glyphosate. Glyphosate was more effective than triclopyr at all timings. In December, glyphosate reduced the privet cover as low as 0% in some replications. triclopyr was almost as effective as glyphosate in December when it reduced the privet cover to about 2%. The interaction occurred because glyphosate was much better in all months of application other then December. The different rates were not significantly different suggesting that lower rates were as effective as higher rates, which is an important finding from an economic perspective. Neither the formulation study nor the trenching study showed any significant differences.