A mildly acidic (pHCa 4.79, 0–10 cm depth) red-brown earth soil (Chromosol) at Condobolin in central-western New South Wales was cultivated and limed (once only) at six rates (range 0–4 t/ha) and sown with field peas (Pisum sativumL.) with and without inoculation (once only) with Rhizobium leguminosarum bv. viciae – the rhizobium for peas. The soil already contained a very small population of pea rhizobia (<4 per g soil). The experiment embraced two parallel rotations, each over 4 years: (1) year 1, inoculated peas; year 2, wheat; year 3, wheat; year 4, uninoculated peas; and (2) year 1, inoculated peas; year 2, wheat; year 3, inoculated chickpeas; year 4, uninoculated peas. The objectives of the work were to establish whether liming had any immediate and residual benefits for rhizobia and plants and, if so, to determine if the two events were linked. Liming had an immediate effect on soil pH (0–10 cm depth). Increases in pH were greater per unit of lime at lower rates of application than at higher rates. Although lime effects existed for the duration of the experiment (four seasons of cropping), there was a small decline in soil pH over time (mean decline in unlimed plots 0.16 pHCa units, mean decline in limed plots 0.47 pHCa units). In the first year (pea crop), there was a very large and highly significant response to inoculation on populations of rhizobia in soil and rhizosphere. The number of rhizobia that occurred naturally in uninoculated plots increased rapidly in high-lime plots until, by the third year, they were substantial and, by the fourth year, equal to those in the inoculated treatment. By the end of the experiment, the mean population of rhizobia in the 4 t/ha lime treatment was 7250 per g soil, compared with <4 rhizobia per g in the nil lime treatment. It was noteworthy that, in those years in the rotations when peas were not grown, populations of R. leguminosarum bv. viciae were sustained by their ability to colonise the rhizospheres of wheat and chickpea. In the first pea crop, eight parameters of plant production responded overwhelmingly to inoculation, while there was an underlying response to liming in two of those parameters. The positive effect of inoculation on peas in the first year carried over to the wheat crop of the second year, which was interpreted as a consequence of increased soil N in the inoculated plots. By the third and fourth years, soil populations of pea rhizobia in the plus inoculation and minus inoculation treatments were approximately equal, and inoculation was no longer a determinant of crop production. On the other hand, application of lime, which had only an underlying effect on pea production in the first year, significantly enhanced several parameters of the symbiosis and growth of the chickpea and pea crops, including legume nodulation and percentage nitrogen in the seed. R. leguminosarum bv. viciae, legumes and cereals each responded differently to increasing rates of lime application. Populations of rhizobia in soil and plant rhizospheres increased with each additional rate of liming. Legume productivity responded to additional lime up to 2 t/ha. There was no significant evidence that liming per se had any effect at any time on wheat production. The practical implications of these results are discussed.