Canada thistle [Cirsium arvense [L.] Scop. # CIRAR] is a major weed problem in birdsfoot trefoil (Lotus corniculatus L. ‘Norcen’) seed production in northern Minnesota. Several systemic herbicides applied with selective applicators (roller and ropewick) were evaluated for Canada thistle control in birdsfoot trefoil. Picloram (4-amino-3,5,6-trichloro-2-pyridinecarboxylic acid), clopyralid (3,6-dichloro-2-pyridinecarboxylic acid), and glyphosate [N-(phosphonomethyl)glycine] controlled Canada thistle better than MCPA [(4-chloro-2-methylphenoxy)acetic acid] and dicamba (3,6-dichloro-2-methoxybenzoic acid). Glyphosate and MCPA did not injure birdsfoot trefoil foliage or bloom, whereas, picloram, clopyralid, and dicamba injured both. Although selective applications of glyphosate controlled Canada thistle for a short term with the least birdsfoot trefoil injury, long-term Canada thistle control in birdsfoot trefoil does not appear feasible with selective herbicide applications.

Greenhouse and field experiments were conducted with a roller applicator at Lincoln, Nebraska, during 1979 and 1980. Glyphosate [N-(phosphonomethyl)glycine] concentrations of 5, 10, and 20% and carpet saturations of 50 and 75% controlled shattercane [Sorghum bicolor (L.) Moench] when applied to the top 30 cm of the plant in greenhouse research. In the field, glyphosate concentrations of 5 to 20% with a carpet saturation of 50% controlled shattercane acceptably in soybeans [Glycine max (L.) Merr.], but a concentration of 2.5% with 25% carpet saturation did not. Weed control was comparable whether speed of application was 3.2, 6.4, or 9.6 km/h. Shattercane control in grain sorghum [Sorghum bicolor (L.) Moench.] was excellent at glyphosate concentrations of 5, 10, and 20% and at carpet saturations of 50 and 75%, and sorghum injury was minimal at 25 and 50% carpet saturations. The roller applicator was compared to a ropewick applicator for shattercane control in sorghum. Excellent weed control (90% or greater) with minimal crop injury was obtained with the roller applicator at glyphosate concentrations of 10 and 20% at application speeds of 3.2 and 6.4 km/h and with the ropewick applicator with glyphosate concentrations of 35 and 50% applied at 3.2, 6.4, and 9.6 km/h.

Three applications/yr of glyphosate [N-(phosphonomethyl)glycine] with a ropewick applicator (RWA), a roller applicator (RA), and a recirculating sprayer (RCS) were compared to hoeing and cultivation for the control of johnsongrass [Sorghum halepense (L.) Pers. ♯3 SORHA] in cotton (Gossypium hirsutum L. ‘Acala SJ-2′) in six field experiments from 1979 to 1981. Averaged over the 3 yr and dates of application, plots treated with glyphosate contained 78% fewer live johnsongrass shoots (15/m2) than cultivated-control plots (68/m2). Although applications of glyphosate with the RWA and RA improved or tended to improve yields of cotton, only plots treated with the RWA yielded significantly more seed cotton than cultivated control plots in all years. However, yields of plots treated with the RWA still averaged 38% less than hoed plots. Plots treated with the RA produced significantly more cotton than cultivated-control plots only in 1981.