Most Probable Number Method
This assay measures presence or absence of mycorrhizae in a dilution series, with the results interpreted as a probability estimate of propagule numbers from a statistical table. Even though the assay is indirect (absolute numbers of propagules are not measured), it has the advantage of providing a single number that can be compared directly with other tests in the same assay. The disadvantage is that the number has a high 95% confidence intervals, even with a low dilution factor and many replicate tubes. Environmental factors, type of inoculum, and manipulations of the inoculum all will affect results (Wilson and Trinick, 1982), so that comparisons are not possible between different types of inocula or trials carried out at different times by different people. All relevant comparisons must be made in one experiment (Adelman and Morton, 1986). This leads into the next most important consideration: dilution factor and number of tubes per dilution. The lower the dilution and the more tubes per dilution reduce the 95% confidence intervals, thus increasing accuracy of the test. Logistic considerations often require a compromise. Five tubes per dilution generally is adequate. Two-fold dilutions are optimal, but this often leads to many dilutions to insure that the end-point is reached. Five- or 10-fold dilutions often are used when inoculum is perceived to be highly infective, but some accuracy will be sacrificed.
The time of harvest depends on host, growing conditions, and size of growth container; but it must await infiltration of roots throughout the growth medium. When setting up the assay, it is useful to have several extra sets of tubes which can be harvested at 4 and 6 weeks to obtain some idea of the pattern and degree of root growth. At harvest, roots are stained and examined for presence or absence of colonization. In the dilutions with less inoculum, roots must be scanned carefully because only localized infection units are likely to be found. If using field soils as inoculum, beware of root colonization by water molds and other lower fungi which can be easily mistaken for mycorrhizal colonization under a dissecting microscope. It is worth a little extra time verifying mycorrhizal structures under a compound microscope, especially in roots from the dilutions near the end-point (which will be the main source of error in scoring). The presence of detached hyphae or germination spores does not count in scoring procedures.
The MPN value is determined from counts of positive tubes (those with mycorrhizae) in different dilutions. The first dilution to be measured is the lowest one with the maximum number of positive tubes (usually 5 in an assay with 5 tubes per dilution), along with the next two dilutions. The number of positive tubes in each of these three dilutions are applied to a table of most probable numbers (Alexander, 1965). The number of progagules is estimated by multiplying the most probable number from the table by the dilution factor of the middle reading. Confidence intervals are calculated from another table (Alexander, 1965).
Adelman, M. J. and J. B. Morton. 1986. Infectivity of vesicular-arbuscular mycorrhizal fungi: Influence of host-soil diluent combinations on MPN estimates and percentage colonization. Soil Biol. Biochem. 18: 7-13.
Alexander, M. 1965. Methods of Soil Analysis, Part 2, Agronomy Series #9, ASA, Madison, Wisconsin, pp. 1467-1472.
Wilson, J. M. and M. J. Trinick. 1982. Factors affecting the estimation of numbers of infective propagules of vesicular arbuscular mycorrhizal fungi by the most probable number method. Aust. J. Soil Res. 21: 73-81.