Three experiments were undertaken to investigate the mechanisms used by the annual medic species, Medicago murex (Murex medic) to achieve nodulation more rapidly in acidic soils than the perennial species Medicago sativa (lucerne). In experiment 1, numbers and locations of root hairs on the primary roots of medic and lucerne were determined from plants grown in soil of pH 4.3 and 7.0. Experiment 2 enumerated the numbers of Sinorhizobium medicae (rhizobia) associated with the roots of medic and lucerne when grown in an acidic soil. Experiment 3 used a GFP-marked strain of rhizobia to determine the localised distribution of rhizobia along various zones of the primary roots of medic and lucerne in soil of pH 4.3 and 7.0. When grown in an acidic soil, medic produced 60% more root hairs/mm root along the primary root axis compared with lucerne, from 7 days after sowing. In soil of low pH, medic also had higher numbers of rhizobia associated with its entire root system than lucerne, with about 10³ cfu/cm root throughout the 24-day period of the experiment, compared with lucerne for which rhizobial numbers decreased from 10² to 11 cfu/cm root over the same time period. When the intensity of fluorescence emitted by a GFP-marked transconjugant of rhizobia was measured at localised zones along the primary root of medic and lucerne grown in the acidic soil, it was 1.8-fold higher from the roots of medic than lucerne at 7 days after sowing, indicating higher numbers of rhizobia along medic roots. Hence, the numbers of rhizobia associated with the entire root system, and those colonised on localised zones of the primary root of medic were both higher than that of lucerne. These fundamental differences between the 2 species, combined with the earlier finding that lucerne tends to acidify its rhizosphere more than medic, allowed us to construct conceptual models that attempt to explain the increased symbiotic competency of medic compared with lucerne, and the different nodulation responses between medic and lucerne at low pH.