1Institute of Field and Vegetable Crops, 21000 Novi Sad, Maksima Gorkog 30, Yugoslavia; 2Agricultural Research lnstitite ‘Serbid’, Belgrade Centre for Small Grains, 3400OKragujevac, Save Kovadevi6a 31, Yugoslavia

Background and objectives
Successful wheat breeding for resistance to causal agent of powdery mildew is based on the identification of virulence and changes in the pathogen population. In our country, continuous population surveys started in 1961, when physiological races were identified and have been continued. In early 1970s, the conventional investigations of physiological races were improved by introducing the survey of the pathogen population virulence, based on the interaction between resistance alleles and virulence according to the ‘gene-for-gene’ theory [1].

Materials and methods
The sexual part of the population was studied for eight years (1990-97). Wheat samples containing cleistothecia of the fungus were collected in 233 locations in Serbia. The cultures of fungi were produced by a standard method . The pathogenicity of the 1383 isolates (cultures) of Erysiphe graminis tritici was determined by isogenic lines and wheat varieties. Reaction of seedlings was assessed 8-0 ;days after the inoculation, on a 0-9 scale.

Results and conclusions
The 1383 isolates analysed rendered 881 virulence formulae. Complex genotypes are formed by sexual reproduction. Most of the genotypes possessed 8-14 virulent genes. It is important to know individual frequency of virulence alleles, because of recombination in sexual reproduction. The results showed that numerous genotypes had Vi, V7, V8, V6, V5 and V3c virulence alleles. The relationship between individual virulence alteles is not accidental. Most Vl and V2 isolates were in the combination with V3a, V3c, V5, V6, V7 and V8. This indicates that most cultivated varieties possess some of Pm 1, Pm 2, Pm 3a, Pm 3c, pm 5, Pm 6, Pm 7 and Pm 8. The high frequency of V6, V8 and V 5 is undefined because Pm 6 resistance gene originated from T. timpheevi, Pm 8 from rye and pm 5 is a recessive gene.

1. Flor HH, 1955. Phytopathology 45, 680-685.