Taxonomy and pathogenicity of rusts from Allium species in the U.K
Comparison of seven isolates of rust from leek (Allium porrum), three from chives (Allium schoenoprasum), one from A. scorodoprasum and one from A. babingtonii on the basis of telial and uredinial morphology showed there to be three morphologically distinct species. Application of these criteria to herbarium material confirmed these findings and showed the rust on leeks in the U.K. to be the same species found on European leeks, garlic and certain wild Allium spp. from the Mediterranean. It is suggested that the following names be adopted, viz. Puccinia allii (DC.) Rud. for the rust on leeks, Puccinia mixta Fuck. for the rust on chives and Uromyces ambiguus (DC) Lev. for the rust on A. babingtonii. Infection studies on isolates of each rust supported the morphological evidence, and showed the three species to have different, extensive but overlapping host ranges within the genus Allium. In the ampeloprasum complex, A. kurrat accessions were highly susceptible to leek rust whereas some accessions of A. ampeloprasum and A. babingtonii had high levels of quantitative resistance. There was no evidence of 'hypersensitive-type' resistance in the complex. Tests within one leek cultivar (Musselburgh) showed older plants to be more resistant than seedlings in at least two quantitative components. However leaf tissue appeared to become more susceptible to infection with age, except in the leaf tips, which did not change in susceptibility over time. Inoculation of 16 leek cultivars with leek rust isolates from different geographical areas, and subsequent analysis during the disease cycle of several components of resistance (viz; latent period, pustule density and pustule length) showed that some cultivars performed consistently better against all isolates. However, in most cases there was a considerable and complex cultivar-isolate-component interaction. There was no evidence of physiologic specialisation in the isolates, but low levels of specialisation could have been hidden by the high level of variation in the experiments. Comparison of field cultivars of garlic with equivalent virus-free material using an isolate of leek rust gave inconclusive results, and further study of the rust-garlic-virus interaction is recommended. Major trends in the infection/resistance studies included a high level of environmentally-dependent variation and a lack of 'hypersensitive-type' resistance, even in host species quite distantly related to the normal host.