Garlic (Allium sativum L.) is an important crop in China. In the spring of 2015, stunted and chlorotic plants in some garlic crops were observed in Jinxiang, Shandong Province, and Feng County, Jiangsu Province, in China. The growth of field symptomatic plants was weaker, with affected plants showing signs of apical chlorosis, wilting, and dieback. In the aerial part of the plant, the symptoms of pink root disease resembled nutrient deficiency or drought. The roots were pink and rotted, with reduced root density. To identify the causal agent of this disease, symptomatic plants were sampled. Samples of pink roots were washed in tap water, surface sterilized in 1.5% sodium hypochlorite for 1 min and then 75% ethanol for 30 s, rinsed with sterilized distilled water three times, and then blotted dry on sterile filter paper. Pieces of infected root tissues (2 to 3 mm) were cut and placed on potato dextrose agar (PDA) at 28°C for 7 days. After the colonies were established on PDA, the fungal strains were purified by the hyphal-tip method. Eight fungal isolates were obtained, and they were morphologically similar to each other, so H5 was chosen for the following experiment. Its colony grew slowly (5 cm in 10 days) on PDA plates, with abundant aerial mycelium, which were initially pale pink but changed to dark red with growth. Microconidia were not observed on PDA plates. For sporulation and pycnidia generation, isolate H5 was cultured on corn meal agar at 28°C under darkness for 7 days and then removed to combined 13-h photoperiod and 11-h darkness conditions for 7 days. It produced dark brown pycnidia (globose to subglobose, 160 to 230 µm) with brown setae 55 to 150 µm long. Pycnidiospores (4.0 to 5.0 × 1.5 to 2.0 µm) were oblong ovoid and oozed from rupture or through the ostiole. The morphological characteristics of H5 were consistent with S. terrestris (Boerema et al. 2004). Total genomic DNA was extracted from mycelium of isolate H5, and the large subunit region was amplified by polymerase chain reaction with the primer pair LROR and LR5 (de Gruyter et al. 2010). The sequence of isolate H5 was deposited in GenBank (MH046778), and a BLASTn search showed 100% similarity with several Setophoma terrestris sequences (GQ387587). The pathogenicity of isolate H5 was tested. Twelve healthy 10-day-old garlic sprouts (about 5 cm) were used to conduct pathogenicity tests by wound inoculation. Nine plants were inoculated with 10 ml of conidial suspension (1.2 × 10⁵ conidia/ml), respectively, by the root irrigation method, and another three plants were inoculated with the same volume of sterile water as a control. After 40 days, the leaves gradually yellowed, similar to leaf symptoms observed during natural infection. The roots showed typical symptoms of pink root, and finally the whole plant wilted and died, whereas the control plants remained healthy. The same colonial fungus was reisolated, thus fulfilling Koch’s postulates. The pathogen associated with pink root of various vegetables was hitherto considered to be Pyrenochaeta terrestris (H.N. Hansen, Gorenz, J.C. Walker & Larson or Phoma terrestris H.N. Hansen). To our knowledge, this is the first report of S. terrestris causing pink root of garlic in China.