C Linde1, GHJ Kemp2, and MJ Wingfield3

1Infruitec, Infruitec Center for Fruit Technology, Research Institute for Fruit, Wine and Vine, Plantbiotechnology and Pathology, Private Bag X5013, Stellenbosch, 7599, South Africa; 2Tree Improvement Center, PO Box 1515, White River, 1240, South Africa; 3Forest and Agricultural Biotechnology Institute (FABI), University of Pretoria, Pretoria, 0002, South Africa.

Background and objectives
Phytophthora cinnamomi is an oomycetous, soilborne fungal pathogen of many ornamental and woody plants. It is heterothallic with two mating types, A1 and A2. Although the center of origin of P. cinnamomi is unknown, isozyme studies have indicated that it was introduced into South Africa [1] and Australia. As a pathogen of forest trees, it is most significant in die-back of Eucalyptus marginata in Australia. P. cinnamomi is also important as a root and root collar pathogen of commercially planted Eucalyptus species in South Africa. Here, Eucalyptus species most seriously affected by P. cinnamomi include E. fastigata, E. fraxinoides and E. smithii.

The long-term success of breeding and selection programs for resistance/tolerance to P. cinnamomi on Eucalyptus species in South Africa, will be partly dependent on variation in pathogenicity of the pathogen population. The aims of the current investigation were, therefore; (1) to assess variation in levels of pathogenicity under field conditions using artificial inoculations on E. smithii; (2) investigate the influence of climate on pathogenicity in the field; (3) study the influence of culture age on pathogenicity; (4) explore the relationship between growth rate in vitro and levels of pathogenicity; (v) compare levels of pathogenicity between P. cinnamomi populations from different geographic regions; and (vi) test for differences in levels of variation for pathogenicity, between mating type populations and those with different multilocus isozyme genotype backgrounds.

Materials and methods
Fifty-nine (39 A2 and 20 A1) South African P. cinnamomi isolates were tested for pathogenicity to E. smithii in the field. Isolates were collected between 1977 and 1993 from the Cape and Mpumalanga regions of South Africa. Based on isolation dates, isolates were divided into two sub-populations, referred to as 'old' and "new". All isolates were inoculated into stems of 3-year-old E. smithii trees at two localities, (Piet Retief and Richmond). Trees were inoculated during summer and winter during 1994 and inoculations were repeated in 1995. Twenty trees were inoculated with each isolate and a control inoculation, using a randomized block design. Lesion lengths in the secondary phloem were measured 8 ;weeks after inoculation. Stem lesion data were analysed using ANOVA.

Results and conclusions
P. cinnamomi isolates differed significantly in levels of pathogenicity to E. smithii in the field. Levels of pathogenicity were influenced by season of inoculation. Culture age, determined by storage in the laboratory, had a significant negative effect on pathogenicity. Significant differences in levels of pathogenicity were found for different multilocus isozyme genotypes. Geographic origin and mating type of P. cinnamomi isolates had no significant effect on levels of pathogenicity. A strong positive correlation was found between growth rate in vitro and levels of pathogenicity. Levels of variation for pathogenicity within A1 mating type isolates were significantly lower than for A2 mating type isolates. Results of this study provide valuable information on selection of P. cinnamomi isolates for future resistance/tolerance screening assays of Eucalyptus germplasm in South Africa.

1. Linde C, Drenth A, Kemp GHJ, Wingfield MJ, von Broembsen SL, 1997. Phytopathology 87, 822-827.<.p>