Validation of a serial-dilution, plant-infection test for enumerating Rhizobium leguminosarum bv. viciae and its application for counting rhizobia in acid soils

Abstract

Rhizobium leguminosarum bv. viciae resident in soil cannot be distinguished from other root-nodule bacteria by cultural means so indirect methods of identification and enumeration are used in ecological studies. A dilution-nodulation frequency technique (plant-infection test) is described whereby Vicia hirsuta, a small-seeded species related symbiotically to pea, lentil, faba bean, other vetch and chickling, is grown in test tubes plugged with flexible polyurethane foam for counting R. leguminosarum bv. viciae. This plant-infection test was compared with counting of pure cultures in broth and peat on agar spread-plates. Comparisons were also made between counts of pure cultures in broth and plant-infection counts of soil-rhizobia mixtures prepared from those cultures. Provided that the V. hirsuta test plants were grown in vermiculite, not agar, both the plant-infection test and plate counting gave similar results with broth culture. With peat culture and soil-rhizobia mixtures, the estimates of rhizobial numbers obtained with plant-infection tests were usually higher than those obtained with spread-plates; in about 23% of instances the differences, although small bacteriologically, were significant ( P < 0.05). The discrepancy may be attributable to more efficient dispersion of clumps of cells resulting from the operations used in preparation of serial dilutions for the plant-infection tests. In all other respects the data showed that the plant-infection test, utilizing V. hirsuta as a test plant, was an appropriate and valuable technique for enumeration of R. leguminosarum bv. viciae in the field. The plant-infection test was used in a survey of populations of R. leguminosarum bv. viciae resident in 62 acid soils in New South Wales where inoculated field peas had grown just once. The populations were generally small (geometric mean 23 g −1 soil) and, in 19 soils, undetectable. There was a significant correlation ( P < 0.001) between population size and degree of soil acidity but no relationship between population and time since growth of peas. It is concluded that the number of R. leguminosarum bv. viciae surviving in soil after a pea crop depends on the degree of acidity, not on how recently the crop was grown.