Sorbus ursina, Don, 1832

Chen, Xin, 2018, Taxonomic and nomenclatural notes on Sorbus foliolosa and related taxa (Rosaceae), Phytotaxa 382 (3), pp. 297-300 : 298-299

Sorbus ursina and S. ursina var. wenzigiana

Sorbus ursina was originally described as Pyrus ursina based on Wall. Cat. 675 ( Don 1832). Wenzig treated this taxon as a variety of S. foliolosa , S. foliolosa var. ursina , and cited Wallich 675 and 675A from Kumaon as original material ( Wenzig 1874: 75). Wenzig provided the Latin diagnosis, cited the basionym as “ Pirus ursina Wallich herb. (1824) N. 675 ”, but he failed to cite Don’s publication where the name was validly published. According to Art. 41.8 (d) ( Turland et al. 2018), Wenizg (1874) validly published a name at new rank. Hedlund (1901) neglected Don’s publication and elevated Wenzig’s S. foliolosa var. ursina to species status as “ S. ursina (Wenzig) Hedlund ”. Hedlund’s treatment was followed by Gabrielian (1978) and Lu & Spongberg (2003). However, S. foliolosa var. ursina itself is a name at new rank based on Wallich’s nomen nudum, and “ S. ursina (Wenzig) Hedlund ” is therefore a later isonym without nomenclatural status. Schneider (1906) annotated that Wallich 675 is heterogeneous, treated the fruiting part as S. ursina , and segregated Wallich 675A as a new variety, S. ursina var. wenzigiana ( Schneider 1906) together with other five collections. Koehne elevated the variety to species rank, S. wenzigiana (Schneider) Koehne (1912: 516) .

Gabrielian (1978) designated the lectotype of S. ursina by citing “Typus (lectotypus): Kam[a]oon, (Ind. Orient.), 1829, Wallich, Cat. 675 (G! photo ERE, iso. BM! E! K! LE! P!)” and that of S. ursina var. wenzigiana by citing “Typus: Kumaon, Wallich 675 A ”. He regarded the two as separate taxa, and his treatment was followed by Lu & Spongberg (2003). Long (1987a) erroneously assumed that the two names had never been formally lectotypified. He designated lectotypes of the two names superfluously and cited both as synonyms of P. foliolosa . Long’s treatment was followed by McAllister (2005). For the priority consideration, Gabrielian’s lectotypifications must be followed.

Wallich 675 comprises specimens collected from three different localities: Kumaon, Srenaghur and Gosain Than ( Wallich 1828), and the collections from Kumaon were lectotypified by Gabrielian. Seventeen sheets belong to six gatherings at B, BM, LE, E, G, K, L, M and P were downloaded and examined. The evidence showed that three gatherings collected by Wallich are from Kumaon, others were collected by Webb at Srenaghur, Blinkworth at Kumaon, and Wallich at Gosain Than, respectively.

For the name Sorbus ursina , G-00437497, a specimen collected at Kumaon in 1929 and with Gabrielian’s handwritten “ Lectotypus ”, is confirmed here as lectotype. But, three specimens, BM-000522004, E-00262154 and P-02143295, cited by Gabrielian as isolectotypes and with his determination labels, are not the same gatherings as the lectotype and should be excluded. BM-000522004 was collected at Gosain Than, E-00262154 at Srenaghur, P-02143295 at Kumaon in 1831 and represented obviously S. ursina var. wenzigiana . Six specimens, LE-00013490, K-000758179, K-001111569, L-0019873 (upper part), L-0019874 (lower left part) and M-0213796 (left part), belong to the same gathering as G-00437497, and are regarded as isolectotypes here.

For the name Sorbus ursina var. wenzigiana, Gabrielian merely indicated that the type was a Wallich 675 A specimen from Kumaon. Nevertheless, Wallich 675.a (capital “A” in the protologue is really lowercase “a” on the specimens) comprises two gatherings from Kumaon. Seven flowering specimens, B-10-0278149, E-00010885, K-001111570 (left part), L-0019873 (lower part), M-0213796 (upper right part), P-02143295 (upper part) and P-02143296, belong to one gathering collected in 1831. Four fruiting specimens, E-00262152, E-00262153, L-0019874 (upper and lower right part) and P-02143295 (lower part, fruit branch), belong to another gathering collected in 1830. B-10-0278149, a flowering specimen and the only sheet with Schneider’s determination label “ S. ursina var. wenzigiana ”, was designated as the second-step lectotype, other six specimens of the same gathering are regarded as isolectotypes.

Sorbus ursina with leaflets serrated from nearly the base and light brown hairs on abaxial side, was obviously not the same taxon as S. ursina var. wenzigiana , which with leaflets serrate apically and rusty brown hairs on the abaxial side just like S. foliolosa . So, it is proposed to treat S. ursina as a separate species and S. ursina var. wenzigiana as a synonym of S. foliolosa .