On dead Platyzosteria novae-zelandiae (det. Mrs B. May) on Agathis australis North Auckland Omahuta Forest, 19.VI.1963, PDD 21230 (DAOM 157650 )

Our population genomics-based phylogenetic results agree with previous taxonomic treatments of the genus Metarhizium (Bischoff et al. 2009; Kepler et al. 2014; Mongkolsamrit et al. 2020) in the sense that the M. flavoviride complex is a sister clade to the M. anisopliae species complex and that, within the M. anisopliae complex, M. acridum and M. globosum form a sister clade to the rest of the complex. We recognized three new species within the M. anisopliae complex, namely M. neoanisopliae, M. hybridum, and M. parapingshaense. Metarhizium neoanisopliae and M. hybridum are distinguished from M. anisopliae s. str. based on multiple lines of evidence. First, the genomics data segregated M. neoanisopliae and M. hybridum from M. anisopliae s. str. Second, the virulence data showed M. anisopliae s. str. to be less virulent than M. neoanisopliae and M. hybridum in laboratory conditions. Finally, the metabolomics analysis also revealed the differences in secondary metabolite production between the two former taxa and the latter. These support the distinct species status of M. neoanisopliae and M. hybridum from M. anisopliae s. str.

The original description with illustration of M. anisopliae by Metchnikoff (1879) is in Russian, and currently not accessible. The oldest account of the morphology of this species, that we could find, was that of Delacroix (1893), who stated that he had examined specimens of Metchnikoff and described the length of M. anisopliae’s conidia as 7–15 µm. Veen (1968) and Tulloch (1976) made a reference to Metchnikoff’s description of M. anisopliae as having conidia of 4.8 µm long and 1.6 µm wide. The examination of putative M. anisopliae strains by Veen (1968) and Tulloch (1976) resulted in the conidia length being respectively at 4.6–11.5 µm and 3.5–9 µm. These measurements are smaller than those of Delacroix (1893) and of M. anisopliae s. str. as interpreted in our study, but more within the range of M. neoanisopliae. Overall, the conidial dimension of M. anisopliae s. str. with CBS 170.71 as the neotype (Mongkolsamrit et al. 2020) better fits the account of Delacroix (1893). Taken with the fact that CBS 170.71 came from the same original locality of Metchnikoff’s species, it is justifiable to accept this isolate as the neotype of M. anisopliae s. str. (Mongkolsamrit et al. 2020). This reclassification means that ARSEF 7450 and many strains previously identified as M. anisopliae should be reidentified as M. neoanisopliae.

With CBS 170.71 accepted as the ex-neotype culture of M. anisopliae s. str., M. lepidiotae should be considered an objective synonym of M. anisopliae s. str. This proposition is supported by the clustering of the ex-type strain of M. lepidiotae (ARSEF 7488) with CBS 170.71. Initially described as a variety of M. anisopliae by Driver et al. (2000), M. lepidiotae was later elevated to species rank by Bischoff et al. (2009), primarily based on molecular phylogeny. Notably, the spore dimensions of M. lepidiotae (conidia 7.3–10.6 × 3–4.1 µm; Driver et al. 2000) falls within the range of M. anisopliae s. str.