In winter 2007, disease symptoms were observed in stored yellow onion bulbs (Allium cepa) grown in New York (NY) in 2006. Similar symptoms were observed in bulbs produced in 2007, 2008, and 2009. Symptoms were associated with one to three bulb scales near the midsection. Infected scales were light brown to brown, not macerated, and lacking foul odors typical of onion bulbs infected with Burkholderia cepacia. Onion grower-packers located in Orange County, NY were concerned that onion lots were rejected following grading by inspectors who cut bulbs to check market quality. Extent of the problem statewide is not currently clear. Isolation attempts were made from symptomatic tissues onto nutrient agar plates (3), with incubation for 24 h at 26 to 28°C, and PA-20 (2), a semiselective medium for the isolation of Pantoea ananatis, with similar incubation for 4 to 6 days. Most strains that grew on PA-20 were gram negative and yellow pigmented with dark centers. Isolated strains were tentatively identified as P. ananatis on the basis of growth on PA-20, a positive indole and negative oxidase test, positive tests for catalase, fermentation of glucose, Voges-Proskauer, and citrate utilization; negative for phenylalanine deaminase, urease, nitrate reductase, methyl red tests, and hypersensitive response induction in tobacco. The BIOLOG (Hayward, CA) system indicated that all presumptive strains of P. ananatis utilized d-mannose, d-cellobiose, d-melibiose, l-inositol, d-arabinose, cellulose, glycerol, d-arabitol, and sucrose, but not glycogen, N-acetyl-d-galactosamine, malonic acid, l-fucose, or xylitol. Strains of P. ananatis recovered from diseased onions in Georgia (GA) (1) were included in all tests as positive controls. We used PCR primers suggested by R. D. Gitaitis (University of Georgia): PanITS1 (5'-GTC TGA TAG AAA GAT AAA GAC-3') and AS2b (5'-TTC ATA TCA CCT TAC CGG CGC-3'). Together, they amplify the 16S-23S rDNA internal transcribed spacer region of 398 bp; the nucleotide sequences of six NY and three GA strains are identical to each other and 99.3% identical to P. ananatis LMG 20103 (GenBank CP001875) and 93.3% identical to P. stewartii (AJ311838). Pathogenicity tests were done in onion leaves. For inoculation, strains were grown on nutrient agar for 24 h and bacterial suspensions of ~108 CFU/ml were prepared in sterile water. Tips of healthy, greenhouse-grown onion leaves were cut and inoculum was applied to the cut surfaces with cotton swabs. Plants were incubated in a greenhouse for up to 2 weeks. Plants mock inoculated with water were symptomless. Bacteria were recovered from all lesions induced by artificial inoculation with the presumptive strains of P. ananatis. Recovered bacteria had characteristics of P. ananatis. Pathogenic strains from NY and GA produced off-white lesions that extended the length of the leaf, which was consistent with previous studies of the pathogenicity of P. ananatis (1). On the basis of microbiological and molecular analyses and pathogenicity tests, 14 NY strains, each isolated from a different diseased bulb, were identified as P. ananatis. To our knowledge, this is the first published report of P. ananatis causing a disease of onion in New York. References: (1) R. D. Gitaitis et al. USA Crop Prot. 21:983, 2002. (2) T. Goszczynska et al. J. Microbiol. Methods. 64:22, 2006. (3) N. W. Shaad et al, eds. Laboratory Guide for Identification of Plant Pathogenic Bacteria. 3rd ed. The American Phytopathological Society, St. Paul, MN, 2000.