Herbicide sensitivity of desmanthus (Desmanthus virgatus) at different stages of development

Desmanthus (Desmanthus spp.) is a native, naturalised or sown legume in many tropical and subtropical areas, including Australia where its potential use as a companion legume is extending to more southern latitudes. Desmanthus pastures are commonly sown into former cropping paddocks which have a significant weed burden. In these situations, weeds can be an issue during the establishment phase as well as in established desmanthus pastures. Three field experiments were conducted in northern New South Wales to assess the sensitivity of D. virgatus ‘Marc’ to herbicides at different stages of pasture development. At sowing, imazethapyr [0.41 kg active ingredient (a.i./ha)] and trifluralin (0.85 kg a.i./ha) were suitable for pre-emergent use. In a 9-week-old seedling pasture, bromoxynil (0.14–0.56 kg a.i./ha) and flumetsulam (0.02–0.08 kg a.i./ha) caused little plant damage and are suitable for use, while 2,4-D amine (0.54–2.16 kg a.i./ha) and MCPA (0.49–1.95 kg a.i./ha) caused extensive damage, reducing herbage production and delaying flowering. In a 10-month-old established pasture, bromoxynil (0.14–0.56 kg a.i./ha), imazethapyr (0.04–0.14 kg a.i./ha) and terbuthylazine (0.44–1.75 kg a.i./ha) caused minor short-term plant damage. Isoxaflutole (0.04–0.15 kg a.i./ha) resulted in temporary plant foliage bleaching and reduced production, without on-going effects following cutting. Established plants of desmanthus were defoliated by paraquat (0.2–0.8 kg a.i./ha) but regrew without phytotoxic effects. All these herbicides could be used on desmanthus plants. In established desmanthus pastures 2,4-DB (0.35–1.4 kg a.i./ha) caused significant damage but not plant death. There were no ongoing effects of the herbicide on plant productivity following cutting.