Twelve soybean varieties were tried for their nodulation status, mycorrhiza colonization, and grain yield potential, in four different sites that represent four agro ecological zones of central Kenya. In each of the sites, ten soybean varieties (TGx and some local varieties) were planted during the long rains of the year 2005. Four different treatments (control, phosphorous, phosphorous and lime, phosphorous lime and nitrogen) were applied to each of these varieties. The experiment was carried out in a strip plot block design with the soil amendments as the main factor and varieties as sub factor. The nodulation, biomass, and mycorrhiza were investigated at 50% podding. Ten plants per plot were cut at the point of the first node dried and biomass determined. The roots were carefully dug out and the nodules detached and counted. A sub sample of the fine roots was preserved in formyl acetic acid alcohol (FAA) for analysis of mycorrhiza colonization. The bacteria nodulating various soybean varieties were isolated in YEMA. Nodulation test was carried out on these isolates in the green house with soybeans as the test crop. Means were analyzed using ANOVA and separated by Student Newman Keul’s test. Nodulation with the indigenous Rhizobia was apparent in all sites and for all varieties. Promiscuous varieties had higher nodulation than the specific variety (Nyala) in all sites and for all treatments. The duration of growth had effect on nodulation with the late maturing varieties having better nodulation than the early maturing ones. Different agro ecological zones had different varieties that were superior in terms of nodulation, grain yield, biomass, and mycorrhiza colonization. In Mitunguu there was a significant difference (P>0.05) in yield per acre from the various treatments and a highly significant difference (P>0.01) between the varieties while in Tharaka there was no significant difference (P>0.01) in yield resulting from the different treatments but a highly significant difference (P>0.01) from variety to variety. In Embu there was no significant difference in yield at (P>0.05) level. This is in contrast to Maragua where there was a highly significant difference in yield for both treatment and variety. Mitunguu and Embu had comparable and higher yields in kg ha-1 compared to Tharaka and Maragua which had comparable but lower yields in kg/ha. Inoculation was beneficial in Maragua but not in Mitunguu. In Tharaka and Embu there seems to be enough nitrogen in the soil and infection was very low (3 nodules/plant in Tharaka and <5nodules/plant in Embu).