It is important to evaluate the tolerance of cassava varieties under new weed chemical control alternatives. Thus, this study aimed to evaluate the selectivity of herbicides, applied at pre-emergence, for two important cassava varieties grown in the state of Paraná, Brazil. The herbicides and respective doses (g a.i. ha-1) were: diuron (400 and 800), metribuzin (360 and 720), isoxaflutole (60), atrazine (720), S-metolachlor (1,920) and mixtures ametryn+clomazone (1,350+1,900), ametryn + trifluralin (1,500+1,350), isoxaflutole + metribuzin (60+320), isoxaflutole + diuron (60+400), combined with the use of a dual checkl adjacent to each treatment. The cultivars used were Fécula Branca and Fibra. Only herbicide S-metolachlor applied on both cultivars and metribuzin (360 g a.e. ha-1) applied on Fibra did not cause any injury. Atrazine reduced plant density only for Fécula Branca at 60 DAS (days after seeding) but no difference in plant height was detected. Atrazine reduced plant density for cultivar Fécula Branca at 60 DAS, but no plant height reduction was detected. Both atrazine (for both cultivars) and diuron at a dose of 800 g a.i. ha-1 (for Fécula Branca) affected root productivity. Thus, atrazine was considered non-selective for both cultivars and the highest diuron dose was also considered non-selective for the cultivar Fécula branca. Differences in tolerance were verified among the cultivars used in this experiment, with Fibra in general being more tolerant to the herbicides evaluated.
varieties; diuron; isoxaflutole; metribuzin; atrazine; S-metolachlor
