A time-course study was conducted on the effect of synthetic ABA and phloem exudate on root nodulation, nitrogenase activity, and dry matter content of pouch-grown soybean [Glycine max (L.) Merr.]. A hypernodulating mutant (NOD1-3) derived from the cultivar Williams was evaluated. The normally-nodulating cultivar Williams 82 served as the control. All phases of nodulation (nodule initiation, development and function) were inhibited by 1-50 μM synthetic ABA treatment of roots of pouch-grown soybean. The ABA effect was more pronounced for the treatment made 24 h prior to inoculation. The magnitude of inhibition of nodulation by ABA and by phloem exudate from inoculated Williams 82 was decreased when application was delayed to 48 h post-inoculation. Synthetic indoleacetic acid and benzyladenine (1 and 5 μM) also inhibited the number of nodules per plant, and the magnitude of inhibition was greater in the normally-nodulating Williams 82 cultivar than in the NOD1-3 hypernodulating mutant. Changes in endogenous free ABA in cotyledons, leaves, and xylem sap of seedlings were monitored at 24-h intervals after inoculation in both lines, and appeared to respond to inoculation. Addition of crude phloem exudate from Williams 82 or hypernodulating NOD1-3 was inhibitory to root nodulation of NOD1-3 - this effect was concentration dependent. Partially-purified (acidic aqueous fraction) phloem exudate from Williams 82 suppressed nodulation in NOD1-3 by 54% (averaged over a 7-d time-course), while partially-purified phloem exudate (acidic aqueous fraction) from NOD1-3 was much less inhibitory (23%). ABA levels in the phloem fractions applied were nearly identical (15.6 vs 16.0 ng mL-1), which indicates that ABA was not directly affecting nodulation control. Partially-purified extract (acidic aqueous fraction) from leaves of nodulated NOD1-3 was much less inhibitory to nodulation of NOD1-3 seedlings than was the comparable fraction from Williams 82. Although free ABA applied exogenously was capable of altering nodulation, it does not appear that endogenous ABA levels directly account for the autoregulatory control of nodulation in soybean. This implies that there may be factor(s) moving in the phloem other than ABA that possibly impact expression of root nodulation. Alternatively, it may be that plant development is responsive to interactions among plant growth regulators mediated by subtle changes in individual hormone concentrations. The possibility remains that normally-nodulating and hypernodulating lines have differential sensitivity to ABA, and that this explains altered nodulation control at similar levels of endogenous ABA.