Abstract

Four commercial procedures for inoculating soybean (Glycine max [L.] Merr.) with Bradyrhizobium japonicum were compared at three sites in New South Wales. Three procedures involved applying inoculant to the seed shortly before sowing (seed inoculation) as a slurry in water or gum arabic adhesive or by sprinkling inoculant on the seed in the seed box of the planter. In the fourth, inoculant was suspended in water and sprayed directly into the seed bed alongside and beneath the seed (liquid inoculation).With seed inoculation substantial losses of viability of inocula (94%-99.95%) occurred between inoculation and sowing. Some of the loss was attributable to separation of inoculant and seed, as it passed through the machinery; no such loss of viability occurred with liquid inoculation. Colonization of the rhizosphere relative to the numbers of rhizobia delivered into the seed bed was poorer with liquid inoculation than with the slurry treatments. This was attributed to the uniform distribution of rhizobia applied as liquid inoculant in the seed bed, in contrast to the concentration of rhizobia applied as slurries in the vicinity of each seed. Slurry treatments generally promoted earlier nodulation than liquid inoculation, but as the crops aged, the differences disappeared and were not reflected in seed yield or seed nitrogen. Application of inoculant to seed in the seed box, although better than an uninoculated control, was consistently inferior to the other inoculation procedures.Given favourable environmental conditions both at sowing and afterwards, slurry inoculation and liquid inoculation were equally likely to initiate a successful symbiosis. Slurry inoculation may be superior when there is limited soil moisture after sowing, whereas liquid inoculation may be better when seed is sown into relatively dry soil with good moisture beneath.