One hundred and fifty-seven strains belonging to Aspergillus section Nigri were tested for ochratoxin A production using three different methods: a relatively new immunochemical method based on an enzyme-linked immunosorbent assay (ELISA), thin-layer chromatography (TLC) and high-performance liquid chromatography (HPLC). The monoclonal antibody-based ELISA technique was successfully used to screen for low levels of ochratoxin A in the black Aspergilli without concentrating the culture filtrates. The results were confirmed by TLC and HPLC analysis and chemical derivatization. These latter methods required concentrated filtrates. Ochratoxin A was detected in the culture filtrates of five of the 12 A. carbonarius strains, none of the 45 A. japonicus strains and three of the 100 isolates in the A. niger aggregate (A. foetidus, A. awamori and A. niger.

Ochratoxin production was tested in 172 strains representing species in sections Fumigati, Circumdati, Candidi, and Wentii of the genus Aspergillus by an immunochemical method using a monoclonal antibody preparation against ochratoxin A. Ochratoxin A was detected in Aspergillus ochraceus, A. alliaceus, A. sclerotiorum, A. sulphureus, A. albertensis, A. auricomus, and A. wentii strains. This is the first report of production of ochratoxins in the latter three species. Ochratoxin production by these species was confirmed by highperformance thin-layer chromatography and by high-performance liquid chromatography. The chemical methods also indicated the production of ochratoxin B by all of the Aspergillus strains mentioned above.

Sixty-one isolates and collection strains of Aspergillus fumigatus were compared for their phenotypic (morphological features and isoenzyme profiles) and genotypic (restriction enzyme-generated mitochondrial DNA and ribosomal DNA profiles and random amplified polymorphic DNA patterns) features. The examined strains exhibited highly variable colony morphologies and growth rates at different temperatures, but their micromorphologies and conidial diameters were characteristic of the species. Of the isoenzymes studied, the ␤-arylesterase and phosphatase patterns were the most divergent, and the 61 strains could be classified into seven groups. The glucose 6-phosphate dehydrogenase and catalase isoenzyme patterns displayed only a limited variability, while the profiles of superoxide dismutase, lactate dehydrogenase, and glutamate dehydrogenase were highly conserved. The HaeIII-generated mitochondrial DNA patterns and SmaI-digested repetitive DNA and ribosomal DNA hybridization patterns of almost all strains were also invariable. The level of variation was much higher when random amplified polymorphic DNA analysis was applied. Although the patterns of the strains were very similar with most of the primers, the application of some primers made it possible to cluster the A. fumigatus isolates into several groups. The results indicate that the random amplified polymorphic DNA technique could be used more efficiently than isoenzyme analysis for typing A. fumigatus isolates. A good correlation was found between the dendrograms obtained from the isoenzyme and random amplified polymorphic DNA data, but the isoenzyme and amplified DNA patterns did not correlate with the pathogenicity, pigment production, or geographical origin of the strains. One ''A. fumigatus'' strain (strain FRR 1266) exhibited unique isoenzyme, mitochondrial DNA, ribosomal DNA, and random amplified polymorphic DNA patterns; it is proposed that this strain represents a new species of the section Fumigati.

Isolates (178) belonging to Aspergillus sections Fumigati, Candidi, Clavati, and Circumdati were tested for the presence of double-stranded RNA (dsRNA) genomes. Altogether, 5.6% of the Aspergillus strains examined were infected with dsRNAs. dsRNA segments indicative of mycovirus infection were observed for the first time in Neosartorya hiratsukae, Neosartorya quadricincta, Petromyces alliaceus, and Aspergillus clavatus strains. Correlation was not observed between ochratoxin production and dsRNA content of the strains. This is the first report on the detection of naturally occurring dsRNAs in Aspergillus species that are able to reproduce sexually. The detection of dsRNA in sexual aspergilli gave us a chance to examine the transmission of these segments through ascospores. A Neosartorya hiratsukae strain transmitted the dsRNAs efficiently through sexual spores, while the stromata embedding the asci in Petromyces alliaceus did not transmit one of the dsRNA segments. The 0.6-kb dsRNA segment that was present in the single-stromatal cultures was found to be located in the mitochondrial fraction of this strain. This observation indicates that some mechanisms exist in aspergilli to exclude cytoplasmically located dsRNA molecules from stromatal structures.

Isolates (178) belonging to Aspergillus sections Fumigati, Candidi, Clavati, and Circumdati were tested for the presence of double-stranded RNA (dsRNA) genomes. Altogether, 5.6% of the Aspergillus strains examined were infected with dsRNAs. dsRNA segments indicative of mycovirus infection were observed for the first time in Neosartorya hiratsukae, Neosartorya quadricincta, Petromyces alliaceus, and Aspergillus clavatus strains. Correlation was not observed between ochratoxin production and dsRNA content of the strains. This is the first report on the detection of naturally occurring dsRNAs in Aspergillus species that are able to reproduce sexually. The detection of dsRNA in sexual aspergilli gave us a chance to examine the transmission of these segments through ascospores. A Neosartorya hiratsukae strain transmitted the dsRNAs efficiently through sexual spores, while the stromata embedding the asci in Petromyces alliaceus did not transmit one of the dsRNA segments. The 0.6-kb dsRNA segment that was present in the single-stromatal cultures was found to be located in the mitochondrial fraction of this strain. This observation indicates that some mechanisms exist in aspergilli to exclude cytoplasmically located dsRNA molecules from stromatal structures.