Six Bradyrhizobium (lupin) strains were evaluated for their ability to produce in vitro siderophores using four chemical assays. Bradyrhizobium strains WPBS 3201 D and 3211 D gave positive reactions with the chrome azurol S assay (CAS) and produced hydroxamate-type siderophores. The other four strains (USDA 3040, 3041, 3042 and CB 2272) gave negative results for siderophore production with the four assays. The generation time, growth yield and hydroxamate production of strain WPBS 3201 D were affected by the iron concentration of the culture medium and the previous culture history of the cells. Resuspension of washed cells grown previously in media supplemented with 0 and 20 μmol·L−1 Fe into differing iron regimes (0 and 20 μmol·L−1 Fe) suggest that the extent of hydroxamate production was dependent on the growth history of the cells. Cells pre-grown in 20 μmol·L Fe produced a high amount of hydroxamates compared with cells pre-grown in iron-free medium when resuspended in medium containing up 4 μmol·L−1 Fe. Cells pre-grown in 20 μmol·L−1 Fe were also more sensitive to iron repression than those pre-grown in 0.5 μmol·L Fe. Mannitol was the best carbon source for siderophore production. Siderophore synthesis was inhibited by 4-chloromercuribenzenesulfonic acid, 2,4-dinitrophenol, sodium azide and MgCl2 suggesting that an energized membrane and a mercapto group are essential and required for hydroxamate synthesis in Bradyrhizobium (lupin) strain WPBS 3201 D.