286 Paul F. MORRIS 1 , Mary S. CONNOLLY 1 and Dina A. ST CLAIR 2  Department of Biological Sciences, Bowling Green State University, Bowling Green, Ohio 43403, USA.  Department of Vegetable Crops, University of California, Davis, California 95616–8746, USA. Accepted 11 March 1999. Black mould lesions were caused by Alternaria alternata in 76 % of 228 tomato fruit with characteristic sunken black lesions collected from fields of processing tomatoes in California. Analysis of 29 RAPD primers revealed a high level of genetic diversity among the 69 isolates tested. Two major phenetic groups (Group 1 with 55 isolates and Group 2 with 14) were identified independently by PCA and by UPGMA of Jaccard similarity coefficients. Only 34 of 137 RAPD markers were monomorphic for all isolates and the genetic similarity between the two groups was 50 %. Co-infection of black mould lesions by genetically distinct strains of A. alternata occurred in two of 10 isolates tested. There was no evidence for geographic clustering of isolates with high levels of genetic similarity, suggesting that isolates are widely dispersed across California. Only one isolate was identified which also caused stem canker disease on a susceptible tomato cv., suggesting that these strains play a minor role in causing black mould on processing tomatoes in California. This isolate and two other known stem canker isolates were clustered together with 11 other isolates in Group 2. Group 2-specific bands were also identified in a survey of seven additional isolates known to produce host-specific toxins. Alternaria alternata (Fr.) Keissl. is the principal causative agent of black mould, a disease characterized by sunken black lesions on ripe tomato fruit (Pearson & Hall, 1975). A. alternata was found to be the most prevalent fungus causing ripe fruit rot in California processing tomatoes (Butler, 1959). A recent survey indicated that 5 % of all fruit harvested from four counties in California had visible symptoms of black mould (Davis et al., 1997). Infection of ripe tomato fruit by this airborne pathogen is promoted by high humidity and cool temperatures, dew, or rainfall (Pearson & Hall, 1975). Alternaria alternata f. sp. lycopersici Grogan, Kimble & Misaghi (AAL) produces stem cankers, foliar symptoms and lesions on green fruit of tomato cultivars which do not contain the incompletely dominant allele Asc in the homozygous condition (Grogan, Kimble & Misaghi, 1975 ; Gilchrist & Grogan, 1976). These disease symptoms result from host- specific toxins produced by AAL strains (Gilchrist & Grogan, 1976 ; Chen et al., 1992; Caldas et al., 1994). AAL isolates are also capable of causing black mould on ripe tomatoes, but it is not known whether they are a major cause of black mould in the Central Valley of California. This region produces the majority of tomatoes for the tomato processing industry in the USA. The potential for breeding resistance against A. alternata isolates that cause black mould was examined using a sample of California isolates on two tomato populations derived from crosses of cultivated Lycopersicon esculentum with a related wild species, L. cheesmanii (Cassol & St. Clair, 1994). Resistance to black mould in L. cheesmanii f. sp. typicum accession LA 422 is controlled by at least two genes and exhibits low (16 %) heritability (Cassol & St. Clair, 1994). This study suggested that breeding for resistance to A. alternata is possible. Knowledge of pathogen diversity, however, and its relation to aggressiveness, if any, is needed to determine how useful any source of resistance would be in controlling black mould. There is considerable morphological variation of spore size, shape, and ornamentation among A. alternata isolates (Simmons, 1978) and this is also reflected in estimates of genetic variation. Petrunak & Christ (1992) identified 22 genotypes in a survey of 96 isolates collected from across the United States which were clustered at a genetic distance of 035. Adachi et al. (1993) used a rDNA probe to assess genetic variation of the Japanese pear pathotype in central and western regions of Japan. Their results indicated that there was also a high level of genetic variation in this pathotype. Hybridization of DNA from several Alternaria spp. with a rRNA probe indicated that isolates producing host-specific toxins were genetically distinct from other Alternaria spp. but not from isolates of A. alternata (Kusaba & Tsuge, 1994). Sequence analysis of spacer regions of rRNA also supported the conclusion that isolates which produce host-specific toxins are variants of A. alternata (Kusaba & Tsuge, 1995). We have found RAPD markers to be a reliable first step for the analysis of genetic variation in plants (Williams & St Clair, 1993) and fungi (Francis & St Clair, 1993 ; Francis, Gehlen & St Clair, 1994). RAPD markers are randomly distributed throughout the genome and can be efficiently and reliably sampled using established procedures (Williams et al., 1990; Mycol. Res. 104 (3) : 286–292 (March 2000). Printed in the United Kingdom. Genetic diversity of Alternaria alternata isolated from tomato in California assessed using RAPDs