Sorbus tapashana C. K. Schneid.

Chen, Xin, Hou, Wen-Xiang, Hu, Jun-Ling, Dong, Meng-Die & Tan, Bao-Mei, 2024, Resurrection of Sorbus tapashana (Rosaceae) based on molecular and morphological evidence, PhytoKeys 247, pp. 29-38 : 29-38

Sorbus tapashana C. K. Schneid. in Bull. Herb. Boissier, sér. 2, 6: 313 (1906)

≡ Pyrus tapashana (C. K. Schneid.) M. F. Fay & Christenh. in Global Fl. 4: 123 (2018).

Type.

China • Shaanxi: the summit of Taibai Shan , 10–20 September 1897, Giraldi 5126 (lectotype, designated here: A [ A 00046062 (the upper left part)] image!) .

Note.

Sorbus tapashana was first described by Schneider (1906). It was later transferred to the genus Pyrus L. s. l. along with other species from Sorbus sensu lato by Christenhusz et al. (2018). However, Pyrus s. l. has proven to be overly inclusive and polyphyletic ( Liu et al. 2019, 2022; Tang et al. 2022; Jin et al. 2024; Wang et al. 2024).

When Sorbus tapashana was described, Schneider (1906) cited three syntypes, Giraldi 5126, 5127, and 1082, deposited in the Berlin Herbarium (B). However, none of them could be located at B. In the herbarium of the Arnold Arboretum, Harvard University (A), duplicates of these gatherings are present. At A, we found the original materials of these gatherings on Plate no. 2296. The sheet, with barcode A 00046062 (image available at: HUH - Databases - Specimen Search (harvard. edu )), bears a photo of specimen “ Giraldi n. 5126 ” at B, and three fragments belonging to Giraldi 5126, 5127 and 1082 respectively (Fig. 1 A ). The photo of Giraldi 5126 is a fruit specimen, with scale bars and three labels on it. One of the three is the determination label of Schneider, with “ Sorbus tapashana , an var. S. pohuashanensis ? ” on it, the same as those that had been stated in the protologue. Giraldi 5126 is a good candidate to serve as lectotype because: (1) there is a photo of complete specimen from herbarium B which bears Schneider’s annotation; (2) the fragmentary material of Giraldi 5126 has a leaflet (clearly showing the white tomentose persistent along the mid-vein abaxially) and two fruits. Therefore, we designate Giraldi 5126 (the upper left part of A 00046062) as the lectotype for the name.

In the protologue, it is implied that Sorbus tapashana may be a variety of S. pohuashanensis (Hance) Hedl. However , it is distinguished from the latter in having much larger flowers (1.5–2 cm in diameter), deciduous linear-lanceolate stipules (vs. relatively small flowers 5–8 mm in diameter, persistent broadly ovate or semi-orbicular stipules). Furthermore, S. tapashana is resolved in clade II while S. pohuashanensis in subclade I. Therefore, S. tapashana and S. pohuashanensis are separate species.

Figures

Figure 1. Phylogenetic tree of Sorbus s. s. resulting from the maximum likelihood analysis and Bayesian inference of the plastome data set. Numbers below the branches indicate bootstrap values (≥ 70 %) of the ML analyses and the posterior probability (≥ 0.95) of Bayesian analyses.