During the summer and fall of 2006, leaf anthracnose samples were collected from fields of cucumber (Cucumis sativus L.), calabash gourd (Lagenaria siceraria (Molina) Standley), and luffa (Luffa cylindrica (L.) M. Roem.) in southern Taiwan. On cucumber leaves, spots start as water-soaked areas and expand into brown spots. Leaf lesions on calabash gourd and luffa begin as water soaked and then become light brown-to-reddish spots. Centers of lesions sometimes fall out; giving infected leaves a shot-hole appearance. Small pieces (approximately 2 × 2 mm) of diseased leaf tissue from margins of individual lesions were surface disinfected in 1% sodium hypochlorite solution for 1 min, rinsed in sterile water, plated on water agar, and incubated at 25°C. After 4 days, mycelium was isolated, transferred to potato dextrose agar (PDA), and then incubated at 25°C in a 12-h light/darkness regimen. Fast-growing colonies on PDA were white to orange or pink with abundant acervuli but no perithecium. One-celled conidia were ovoid to oblong and 12 to 20 × 4 to 6 (15.9 × 5.0) μm. The morphological traits were identical to those of Colletotrichum magna (teleomorph Glomerella magna Jenkins & Winstead) and clearly distinct from those of C. orbiculare (Berk. & Mont.) Arx (synonym C. lagenarium (Pass.) Ellis & Halst. (conidia were mostly oblong, measuring 7 to 11 × 2 to 6 [9.3 × 4.2] μm, with slow-growing gray colonies) (2,3). Koch's postulates were performed to verify that the isolates were capable of causing anthracnose on cucurbitaceous crops. Pathogenicity tests were conducted in the greenhouse at 25°C under natural daylight conditions. Isolate C0604 was grown on PDA for 14 days and a spore suspension was made (106 spores/ml). Three 14-day-old seedlings at the two- to three-leaf stage of muskmelon (Cucumis melo L. var. reticulatus Naud., cv. Sapphire), squash (Cucurbita moschata Duch., cv. Achen), calabash gourd (cv. Huapu), and luffa (cv. 623) were sprayed with the spore suspension and then covered with plastic bags. Control treatments were sprayed with sterile water. After 2 days, the bags were removed. Typical anthracnose symptoms developed on all inoculated seedlings 7 days after inoculation. G. magna was reisolated from inoculated leaves following the protocol used for the original isolation. Control seedlings developed no symptoms. To confirm the identity of the fungus, PCR amplification and DNA sequencing of the internal transcribed spacer 1 (ITS1)-5.8S-ITS2 of rRNA gene of the isolate C0604 was performed by using ITS1/ITS4 as the PCR and sequencing primers. Sequence analysis of the 558-bp PCR product (GenBank Accession No. GU358453) showed 100% identity to the rRNA sequence of G. magna (GenBank Accession No. DQ003103) (1). PCR amplification of the ITS region was also carried out using species-specific primer GmF (5'- GTG AAC ATA CCT CAA ACG TTG CC -3')/GmR (5'- GGA GGG TCC GCC ACT GTA TTT CG -3') designed in this study. A DNA fragment of approximately 378 bp was amplified from nine isolates of G. magna, whereas no amplification products were obtained from reference cultures of C. gloeosporioides (Penz.) Penz. & Sacc. and C. orbiculare. To our knowledge, this is the first report of G. magna causing anthracnose on cucurbitaceous crops in Taiwan. References: (1) M. Du et al. Mycologia 97:641, 2005. (2) S. F. Jenkins, Jr. and N. N. Winstead. Phytopathology 54:452, 1964. (3) T. A. Zitter et al., eds. Compendium of Cucurbit Diseases. The American Phytopathological Society, St. Paul, MN, 1996.