Field experiment was conducted during the rice fallow season of 1999 to determine the most effective integrated methods of weed control. The experiment was laid out in randomised block design with four replications with six treatments viz. pendimethalin @ 1.0 kg/ha, fluchloralin @ 1.0 kg/ha, pendimetha1in @ 1.0 kg + one hand weeding on 20 DAS, fluchloralin @ 1.0 kg + one hand weeding on 20 DAS, two hand weeding on 20 and 40 DAS and weedy check. Integrated weed management combining pendimethalin @ 1.0 kg/ha + one hand weed on 20 DAS was most effective method of weed control and produced the maximum seed yield followed by fluchloralin @ 1.0 kg−1 ha + one hand weeding. Weedy condition for the entire crop season reduced the seed yield by 35% as compared to twice hand weeding. Integration of one herbicide with one hand weeding provided better growth, yield attributes and consequently higher yield. In respect of nodulation twice hand weeding treatment recorded the highest nodule number and nodule weight followed by penldimetha1in @ 1.0 kg with one hand weeding treatment.