Main Botanical Garden RAS, Botanicheskaya St. 4, 127276 Moscow, Russia

Background and objectives
The invasion of a host by rust and powdery mildew fungi requires the development of specialized infection structures from the germinating spore. Their formation is genetically programmed and the form and shape of these structures is characteristic of the species under optimal envirornmental conditions. Several observations suggest that phenomena associated with the initial stages of fungal contact on host surfaces play a critical role in host recognition, fungal differentiation, and success of infection. Our research is devoted to studying various peculiarities of the differentiation of rust and powdery mildew fungi on the cereal varieties with different resistance. The task is to reveal the correlation between the morphology of pathogen primary infection structures and the resistance of plants to their penetration.

Materials and methods
Six-day-old seedlings of barley and wheat varieties with different resistance types were inoculated with Erysiphe graminis f.sp. hordei, E. graminis f.sp. tritici, and Puccinia graminis f.sp. tritici. Leaf samples were taken 12, 24 and 48 h after inoculation, conventionally fixed in phosphate-buffered glutaraldehyde/osmium tetroxide, dehydrated in ethanol and critical point-dried. The specimens were observed with a Jeol scanning electron microscope. The data were morphometrically processed from negatives with a KONTRON image analyser system.

Results and conclusions
On susceptible varieties conidia of E. graminis hordei and E. graminis tritici form very compact primary infection structures with short germ and appressorial tubes, and oval appressoria with two or three short lobes. Uredospores of P. graminis tritici form long germ tubes and oval appressoria above the stoma or anticlinal walls of epidermal cells. Scanning electron microscopy showed disturbed adhesion between the appressorial lobes and the surface of the epidermal cells in resistant wheat and barley varieties, which led to sliding of infection structures from the surface of resistant plants, making penetration impossible. Various kinds of deviation from the normal were seen in morphology of infection structures of the pathogen, and a genetically conditioned pathway of development could be observed on resistant varieties. The genetic effects were expressed in bolting of the germ tubes and formation of elongated, weakly filled appressoria. Appressoria were not formed on non-host plants, with rare exceptions, and spores germinated several germ tubes of varying length. Morphometric studies of E. graminis and P. graminis development during the pre-penetration stage and on the reproductive stage have revealed positive correlations between normal primary infection structures and the amount of pustule formation. These studies suggest universal reactions of obligate pathogens to contact with the plants resistant to fungal penetration. Our experiments showed that a variety can be considered resistant to penetration of the pathogen when less than 30% of normal infection structures are formed. Hence, it is possible to use the relative amount of normal primary infection structures of the fungus as an index of resistance of the plants to penetration during obligate parasitism on cereals. The use of this method is most expedient for characterization of the donors of resistance from native flora and promising hybrid lines. Development of criteria for plant resistance at different stages of interaction between pathogens and plants could markedly increase the reliability of genetic control of resistance of hybrid forms.

This study was supported by the Russian Foundation for Fundamental Investigations, project N 96-04-48581.