PHENOTYPIC DIVERSITY IN 2 POPULATIONS OF PUCCINIA-RECONDITA F SP TRITICI IN CANADA DURING 1931-1987

The Canadian eastern and prairie populations of Puccinia recondita f. sp. tritici from race surveys conducted during 1931-1987 were assessed for phenotypic diversity within each population and phenotypic overlap between the two populations using four different indexes based on infection type on the modified Unified Numeration differentials. The indexes showed that the two populations had similar levels of absolute diversity during 1931-1959. Diversity in the prairie population was consistently lower than in the eastern population during 1960-1975, increased during 1976-1984, and decreased during 1985-1987. Levels of phenotypic diversity in the eastern population changed relatively little from the initial years. The degree of phenotypic overlap as measured by the Rogers index was highest during 1931-1946, after which the two populations diverged phenotypically. The changes in phenotypic diversity in the prairie population could be attributed to the directional changes in race frequencies due to the use of resistant cultivars in this region. The diversity indexes were regressed on the variables' sample size, number of races, and evenness of race frequencies to determine the proportion of variation in the diversity indexes that could be accounted for by the independent variables. The Gleason index was the most sensitive to number of races; the Shannon and Simpson indexes were the most sensitive to evenness of race frequencies. In general, the three indexes of absolute diversity showed little variation due to sample size. Differences in absolute phenotypic diversity within, and phenotypic overlap among, the eastern, prairie, and Pacific populations of P. r. tritici in 1988 were enhanced by using the 12 differentials in the Prt nomenclature.