Five preemergence herbicides were applied to seven herbaceous perennials to evaluate weed control efficacy and phytotoxicity. Different species were used each year. The species used during 1992 were coneflower (Rudbeckia fulgida Ait. `Goldstrum'), common foxglove (Digitalis purpurea L. `Excelsior'), Shasta daisy (Leucanthemum ×superbum Bergmans `Alaska'), Stokes's aster (Stokesia laevis Greene `Blue Danube'), and avens (Geum Quellyon Sweet `Mrs. Bradshaw'). The species used in 1993 were woolly yarrow (Achillea tomentosa L.) and woolly thyme (Thymus pseudolanuginosus Ronn.). The herbicides and rates were napropamide (Devrinol 10G) at 4 and 8 lb a.i./acre; metolachlor (Pennant 5G) at 4 and 8 lb a.i./acre; oxyfluorfen+oryzalin (Rout 3G) at 3 and 12 lb a.i./acre; trifluralin (Treflan 5G) at 4 and 8 lb a.i./acre; and oxadiazon (Ronstar 2G) at 4 and 8 lb a.i./acre. Plants were grown in no. 1 containers and weed seeds were sown onto the substrate surface. Two control treatments, no herbicides but with weeds (weedy control), and no weeds or herbicides (weed-free control) also were evaluated. Weed control was effective and similar for all herbicides tested. Napropamide at 8 lb a.i./acre caused stunting in foxglove (20% to 45% less growth compared to weed-free control). Oxyfluorfen + oryzlain at 12 lb a.i./acre caused severe phytotoxicity (≈80% to 95% of plant injured) and stunted the growth (70% to 80% less growth, sometimes plant death) of woolly yarrow. Woolly thyme was stunted by all herbicides when applied at the recommended rates (42% to 97% less growth compared to control) except for oxadiazon and oxyfluorfen + oryzlain. Woolly thyme appeared to be more susceptible to phytotoxicity due to its less-vigorous growth habit and shallow, adventitious roots that were in contact with the herbicide.