Liangia H. Yu, Y.B. Wang, Y. Wang, Z.H. Chen & Zhu L. Yang, gen. nov.

Index Fungorum number: MB 833107; Facesoffungi number: FoF14503

Etymology: In honor of Prof. Zong-Qi Liang, acknowl- edging his contributions to our knowledge of cordycipitoid fungi.

Colonies on PDA slow-growing, effuse or stellate, white, usually raising dome-shaped mycelial density with a sunken zone at the centrum, verrucose around the margin. Conidiophores not observed. Phialides lanceolate, occurring directly from the prostrate hyphae, solitary, gradually attenuated toward the apex. Two types of macro- and microconidia, aseptate, smooth-walled, one-celled, both of them existing singly or in pairs at the apex of phialides. Macroconidia positioned at a right angle to the apex of phialides, straight, oblong-oval to fusiform. Microconidia oval to ellipsoidal.

Type species: Liangia sinensis H. Yu, Y.B. Wang, Y. Wang, Z.H. Chen & Zhu L. Yang.

Notes: –  Liangia sinensis, isolated from an entomopathogenic fungus B. yunnanensis, represents a well-supported monophyletic lineage in the family Cordycipitaceae (Fig. 1). The new genus Liangia with Lecanicillium-like asexual morph is proposed for the type species Lia. sinensis based on its phylogenetic placement. In this study, it appears more closely related to C. piperis (J.F. Bisch. & J.F. White) D. Johnson et al. and L. psalliotae clades by the five-gene phylogenetic analyses. The genus Liangia is morphologically similar to these two clades which possess asexual morph of Lecanicillium (Zare and Gams 2001; Bischoff and White 2004). However, it differs from the latter two groups by the shape and size of its colonies, phialides and conidia.

Fig. 1 Phylogenetic tree of Cordycipitaceae and related families inferred from a multigene dataset (nrLSU, nrSSU, tef-1α, rpb1 and rpb2) based on maximum likelihood (ML) and Bayesian inference (BI) analyses. Families of Hypocreales and genera of Cordycipitaceae are differentiated with different colors. Phylogenetic relationships among families of Hypocreales are strongly supported by ML boot- strap proportions and BI posterior probabilities (BP ≥ 70%, PP ≥ 95%)