Quality control test for legume inoculants containing specific Rhizobium involves strain identification to verify that the product contains the right type of bacteria.
It will depend largely on which set of techniques one has access to. If you can use molecular genetics, one of the fingerprinting techniques such as BOX or ERIC should be useful. If this is not feasible, you may think about whole protein electrophoresis. Finally, at low end of the technical spectrum, you may use intrinsic antibiotic resistance, or carbohydrate usage patterns.
For all of these techniques, though, the first step would be to develop a data bank of characteristics of the possible strains, so you can compare you unidentified strain to the data bank, and later identify it.
These techniques can be grossly ranked for these characteristics
Precision
molecular>electrophoresis = antibiotic
Speed
molecular>electrophoresis>antibiotic
Labor demand
antibiotic>electrophoresis>molecular
Equipment cost
molecular>electrophoresis>antibiotic
Chemical costs
molecular>electrophoresis>antibiotic
One major point that explains the difference in labor and speed of this set of techniques is that while molecular may be conducted directly on the nodules or pure inoculant, this is not possible for the antibiotic technique, which will need to be done in pure cultures.