Parnassia nubicola Wallich ex Royle (1835: 227)

Shu, Yumin, Yu, Huiying, Mu, Xianyun & Zhang, Zhixiang, 2017, Checklist and typification of Parnassia (Celastraceae), Phytotaxa 309 (1), pp. 1-22 : 9-10

Parnassia nubicola Wallich ex Royle (1835: 227)

TYPE:— NEPAL. Gossain-than (Gossain-kund), Wallich Cat. n. 1246a (lectotype: W 1889-0305721, designated by Schönbeck-Temesy

1966; isolectotypes: E00301173 , E00301174 , GZU000100103 , NY00186147 , BM000947005 , BM000521923 , L0281535 ,

K 000739481, K 000739482) Parnassia nubicola var. cordata Drude (1875: 316) ≡ Parnassia cordata (Drude) Jien ex T.C. Ku (1987: 38) TYPE:— INDIA. Kumaon, Blinkworth s.n., Wallich Cat. n. 1246b (lectotype: L 0281535 designated here by marker II) Notes: Parnassia cordata was an accepted name in the latest treatment by Ku & Hultgard (2001), but the type material was related to that of P. nubicola , so we arranged it here. Parnassia nubicola was formally described based on Wallich

Cat. n. 1246, which contains two gatherings, one by Wallich from Gossain Than (Wallich Cat. n. 1246a) and the other by Blinkworth from Kumaon (Wallich Cat. n. 1246b). Several specimens of both gatherings were traced, and we found that both gatherings are mixed in some individuals that should be more accurately determined as P. wightiana . Schönbeck-Temesy (1966) selected a sheet of Wallich Cat. n. 1246a from W as the lectotype. Based on these specimens collected from Kumaon, Drude described P. nubicola var. cordata , but no holotype was indicated. In the protologue, he mentioned ‘foliis cordatis 3cm longis et paulo latioribus petalisque ad basin magis fimbriatis diversa; haec ad Parnassiam Wrightianam propius accedit’. From this description, we can speculate that Drude had seen the mixed specimens and named the individuals with fimbriate petals and cordate leaves as var. cordata , and he also noted that these materials may belong to P. wightiana . Eight sheets of Wallich Cat. n. 1246b have been found, in which most of the materials have ovate leaves and subentire petals, thus do not match the descriptions. A scape marked II from L 0281535 has a cordate leaf and proximally fimbriate petals, which should be determined as P. wightiana , matches the description well, and this should be what Drude intended to describe. Since no specimens labeled with Wallich’s number have been found from GOET, where Drude worked when he described this new variety, we designated the individual marked II from L 0281535 as the lectotype of var. cordata .