| Abstract: | Seed inoculation is frequently essential for annual Medicago establishment in Mediterranean dryland farming systems. As post-planting soil inoculation is often practiced when seed inoculation fails, the effect of delayed inoculation was investigated. Roots of Medicago rigidula (L.). All selection 716, and Medicago rotata Boiss. selection 2132, were initially pre-exposed to Rhizobium meliloti strains. Subsequently, roots were exposed to a secondary inoculum after 6, 48 and 168 h to simulate delayed inoculation. and subsequent establishment of others strains in the nodules were investigated. Combinations of highly effective and host compatible ICARDA strains (M15 and M53) and effective-ineffective strains (M15 and M28) were used to evaluate proportional nodulation responses. Plants were harvested after 6 weeks of growth. Nodules were assessed for distribution in the root system and for occupancy based on their differential resistance to kanamycin and streptomycin, and, in the case of ICARDA M28, on nodule characteristics. The strain ICARDA M15 was a better competitor than ICARDA M53 when applied in equal density in M. rotata, at zero time. While forming nodules with M. rigidula, ICARDA M53 was equally competitive under the same conditions. With ICARDA M15 as the primary inoculum, and ICARDA M53 delayed for 6, 48 and 168 h, the incidence of ICARDA M15 nodules increased on M. rigidula from 43% (at zero time) to 86, 94, and 97% and, on M. rotata, from 78% (at zero time) to 88%, 95%, and 98% for the three time delays, respectively. Pre-exposure of 2-day old M. rigidula seedlings to the ineffective strain ICARDA M28 as the primary inoculant was followed by significant nodule number increases (P 0.01) as compared with ICARDA M15. Nonetheless, when ICARDA M15 was the primary inoculum, M28 was able to produce about 35% of the nodules when applied at the 6 and 48 h time delays. However, with ICARDA M28 as the primary inoculum, followed by ICARDA M15, nodule occupancy of ICARDA M28 in M. rigidula increased from 40 (at zero time) to 75%, 80%, and 95% for the three time delays, respectively. The percentage of total nodulation by M28, applied at the three delay times, was markedly lower (10%, 5% and 1%, respectively) when M. rotata was pre-exposed to ICARDA M15. This suggested a host preference for ICARDA M15, even if applied as a late inoculum. Results indicate that the early events in the nodulation process of annual medics coupled with host-specificity factors are perhaps the most critical for competition among R. meliloti strains for nodule formation. |
