Fall applications of 0.14 kg/ha of picloram (4-amino-3,5,6-trichloropicolinic acid) to musk thistle (Carduus nutans L.) rosettes controlled musk thistle stems better and more consistently than fall applications of 2.2 kg/ha of 2,4-D amine [(2,4-dichlorophenoxy)acetic acid], 0.6 kg/ha of dicamba (3,6-dichloro-o-anisic acid), or a mixture of 0.3 kg/ha dicamba + 1.1 kg/ha of 2,4-D. Dry conditions in 1975 and cool conditions in November, 1976, hampered control with all herbicides except picloram. In spring applications when growing conditions were favorable, herbicide differences were usually not significant. For all herbicides, treatment of bolted stems did not control musk thistle satisfactorily. Picloram was superior for reducing stems 12 to 18 months after application in 2 of 3 yr. Residual control resulted principally because the herbicide killed many seedlings and first-year rosettes at the time of application.