A field experiment was conducted in order to evaluate the impacts of previous nonmycorrhizal (Brassica juncea/Raphanus sativus) and mycorrhizal (Crotalaria juncea) crops on the post-transplant growth of mycorrhizal and nonmycorrhizal green onion seedlings. The extent of arbuscular mycorrhizal (AM) fungal colonization of onion roots was not affected by previous crops or by AM fungal inoculation. Solution P concentration of the rhizosphere soil increased as a function of time, and previous crop had significant influence on the variable. However, onion leaf P status was unaffected by soil solution P concentration, AM fungal inoculation, or previous crop. Arbuscular mycorrhizal inoculation significantly reduced yield of onion initially if onion was grown in the field previously planted to mustard and white raddish but not to crotalaria. Yield determined at the termination of the experiment was significantly stimulated if onion followed crotalaria compared to when it followed the non mycorrhizal crops. Final yield was not significantly influenced by AM fungal inoculation. Our data suggest that the presence of appreciable level of AM fungal colonization at solution P concentrations sufficient for mycorrhiza-free growth of onion was responsible for the parasitic effect of Glomus aggregatum observed during the earlier phase of growth. The subsequent disappearance of this parasitism is explained by the suppression of AM fungal colonization by higher solution P concentrations. The significantly higher yield of onion following crotalaria in contrast to that following the brassicas may be due to the better N nutrition in the former field.