Glufosinate at 1.1 and 2.2 kg/ha injured ‘Koshihikari’ rice lines that were transformed with the BAR gene from 0 to 53%. However, transgenic ‘Gulfmont’ rice was not injured. Rice yields of transgenic ‘Gulfmont’ lines and six of nine ‘Koshihikari’ lines were not affected by 2.2 kg/ha glufosinate. In field studies, flooding reduced the efficacy of glufosinate in controlling red rice, and greenhouse tests determined that glufosinate efficacy was reduced when red rice was submerged between 25 and 50% of its height. Plant heights and dry weights of red rice increased as flood water depth increased at all rates of glufosinate.

Glufosinate at 2.2 kg ai/ha injured rice transformed with the BAR gene more when applied to one- to two-leaf (23 to 26%) than to three- to four-leaf (13 to 19%) plants. Visible injury was least when applications were made at boot stage (3 to 14%). However, applications at boot stage caused an average grain yield reduction of 16%. Most treatments did not influence rice plant height. Among single applications (0.3, 0.4, 0.6, 0.8, and 1.1 kg/ha), 1.1 kg/ha glufosinate at three- to four-leaf stage of red rice resulted in greater control (91%) than at panicle initiation (74%) or at boot stage (77%). Injury to red rice was two to 11 times greater than the injury to BAR-transformed rice depending on glufosinate rate and application timing.