Narbon bean (Vicia narbonensis L.) shows promise as a fodder, green manure and grain crop in south-western Australia. This study examines the effect of time of sowing (2 experiments), plant density (3 experiments) and reaction to herbicides (4 experiments on tolerance to herbicides and 1 experiment on removing narbon bean from a wheat crop) in 10 separate field experiments sown at 4 locations in the mallee region of Western Australia from 1998 to 2001. Narbon bean was found to be unresponsive to changes in sowing date with yield maintained until the first week of June. The optimum plant density (90% of fitted maximum) for seed yield was found to be 31 plants/m², equivalent to sowing rates in the range of 75–100 kg/ha. A wide range of herbicides applied either before sowing or immediately after sowing and before emergence had no significant effect on grain yield. These included simazine (750 g a.i./ha), cyanazine (1.25 kg a.i./ha) and diuron (500 g a.i./ha), which were applied immediately before sowing, and imazethapyr (29 g a.i./ha), which was applied after sowing, before emergence. Diflufenican (75 g a.i./ha) was found to be the only available option for post-emergence control of broadleaf weeds. The use of the non-selective herbicides glyphosate (450 g a.i./L) and Sprayseed 250 (paraquat 135 g a.i./L and diquat 115 g a.i./L) as post-emergence herbicides was found to be unpredictable at a range of application rates. Results ranged from a yield loss of 47% to a yield increase of 23%. In an experiment to test a range of herbicides for the selective control of narbon bean within a wheat crop, numerous herbicides were found to effectively remove volunteer narbon bean indicating that narbon bean is unlikely to become a weed in most cereal cropping systems.