Clematis connata var. trullifera (Franchet) Wang (1998: 170)

Yuan, Qiong & Yang, Qin-Er, 2020, The identity of Clematis dongchuanensis (Ranunculaceae) from northeastern Yunnan, China, Phytotaxa 430 (1), pp. 9-16 : 10-14

Clematis connata var. trullifera (Franchet) Wang (1998: 170) . C. buchananiana var. trullifera Franchet (1889: 3) . C.

trullifera (Franchet) Finet & Gagnepain (1903: 547). Figs. 1–3 , 5 , 6 .

Type:— CHINA. Yunnan, Heqing , “in silvis ad Ta-long-tan prope Tapin-tze, alt. 1800 m.; fr. 20 jan. 1887 ”, J. M. Delavay s.n. (lectotype P00145081 !, designated here, isolectotype P00145082 !; see below for discussion). Fig. 6 .

= Clematis coriigera Léveillé (1913: 281) .

Type:— CHINA. Yunnan: Tong-tchouan (= Dongchuan), 2500 m, September 1912, E. E. Mair e s.n. (lectotype, first step, Rehder 1936: 318; second step, designated here, E00081543 !, isolectotype E00081544 !; see below for discussion). Fig. 3 .

= Clematis connata var. sublanata Wang (1957: 373) .

Type:— CHINA. Yunnan: Binchuan, 25 October 1946, T. N. Liou 22236 (holotype PE!, isotype IBSC!).

= Clematis dongchuanensis Wang (2019: 287) , syn. nov.

Type:— CHINA. Yunnan: Dongchuan, Yinmin, Xiaoshannao, 2290 m, on cliff, 11 September 1984, Shun-bin Lan 217 (holotype PE 01896341!, isotype PE 00387866!). Fig. 1 .

Phenology:—Flowering from August to October; fruiting from October to January of next year.

Distribution and habitat:— Clematis connata var. trullifera is distributed in western Guizhou, western Sichuan, and central and northern Yunnan (Fig. 7). It grows in forests or scrub or along streams at elevations of 1800−3200 m.

Additional specimens examined:— CHINA. Guizhou: Weining, Y. Tsiang 9128 (NAS, PE), Weining Exped. 522427140506193LY (GZTM), Weining Exped. 522427140426438LY (GZTM). Sichuan: Butuo, Anonymous 906 (SM); Ganluo, Anonymous 879 (SM); Huidong, J.S. Yue et al. 64197 (NAS); Kangding, Anonymous 392 (SM), Z.P. Huang et al. 1825 (IBSC, KUN, WUK), X.C. Zhang & L. Shi 934 (PE); Luding, K.C. Kuan et al. 1712 (PE); Meigu, Sichuan Econ. Plant Exped. 1045 (PE), Sichuan Veg. Exped. 13012 (CDBI); Mianning, Anonymous 685 (SM), S.F. Zhu 20330 (PE); Muli, Anonymous 665 (SM), W.L. Chen et al. 5674 (PE), Sichuan Econ. Plant Exped. 3456 (PE), Sichuan Veg. Exped. 408 (CDBI, NAS, PE), M.Z. Wen & W.Z. Zeng 151 (151), S.K. Wu 2444 (CDBI), Q.S. Zhao et al. 7000 (CDBI), Q.S. Zhao et al. 8255 (CDBI), Q.S. Zhao et al. 8367 (CDBI); Puge, Puge Exped. 618 (SM), Puge Exped. 872 (SM), Sichuan Med. Resource Exped. 25208 (SM), Sichuan Med. Resource Exped. 25727 (NAS, SM); Yanyuan, Qinghai-Xizang Exped. 12836 (PE), Q.E. Yang 93029 (PE), Yanyuan Exped. 488 (SM), Yanyuan Exped. 550 (SM); Yuexi, Sichuan Veg. Exped. 13967 (CDBI, PE). Yunnan: Binchuan, T.N. Liou 22106 (IBSC, PE), T.N. Liou 22233 (PE), T.N. Liou 22236 (IBSC, PE); Dongchuan, E.E. Maire s.n. (P); Fumin, B.Y. Qiu 596053 (KUN, LBG, PE); Gongshan, T.T. Yu 19894 (KUN, PE); Heqing, J.M. Delavay 3127 (P), J.M. Delavay 3264 (P), J.M. Delavay 3572 (P); Kunming, K.M. Feng 10356 (KUN, PE, SZ), B.Y. Qiu 55129 (BJM, IBK, KUN, PE); Lijiang, K.M. Feng 2584 (KUN, PE); Luquan, Y.P. Chang 448 (PE), P.Y. Mao 1823 (IBK, PE); Qiaojia, Anonymous s.n. (HITBC).

Notes:—The taxonomic treatment of Clematis connata var. trullifera has been a controversial matter. Franchet (1889) first described C. buchananiana var. trullifera and then Finet & Gagnepain (1903) raised it to the specific rank as C. trullifera . This treatment was accepted by Rehder & Wilson (1913), Handel-Mazzetti (1931), Rehder (1936), Tamura (1962, 1968), Fang (1980), Li (1986), Li (1993) and Johnson (2001), but Handel-Mazzetti (1939), Grierson (1984), Grey-Wilson (2000) and Elliott (2016) did not recognize the independent specific status of C. trullifera and placed it in synonymy with C. connata Candolle (1824: 4) , a species occurring in Bhutan, China, India and Nepal,

and readily distinguishable by, among other characters, often having strongly dilated and connate petiole bases. Wang (1998) pointed out that C. trullifera differed from C. connata only by the leaflets often 3-lobed (vs. undivided) and the anthers abaxially sparsely puberulous (vs. glabrous) on connective and thus he reduced C. trullifera as a variety of C. connata , i.e. C. connata var. trullifera , a treatment accepted by Wang (2000, 2001) and Wang & Bartholomew (2001). According to the two authors, C. connata var. trullifera is different from C. connata var. connata in the leaflets often 3-lobed (vs. undivided), abaxially often more or less densely appressed sericeous-puberulous (vs. sparsely puberulous on veins) and the anthers abaxially sparsely puberulous (vs. glabrous) on connective, and is distributed only in western Guizhou, western Sichuan, and central and northern Yunnan, China, not in the Himalayan region. Here we adopt for the time being the concept of C. connata var. trullifera as construed by Wang (1998, 2000, 2001) and Wang & Bartholomew (2001) pending a further study on the relationship between C. connata var. connata and C. connata var. trullifera . Tamura (1962, 1968) reported the occurrence of C. trullifera in Bhutan and Nepal, and Fang (1980) and Johnson (2001) also recorded its distribution in these two countries. The relevant specimens checked by Tamura (1962, 1968), which have leaflets 3-lobed but abaxially subglabrous, may better be referred to C. connata var. connata . Such specimens have also been collected from southern and southeastern Xizang in China and have been identified as C. connata on the determination slips. These include R.F Huang 123 (HNWP) and FLPH Tibet Exped. 12-1512 (PE) from Zayu, and FLPH Tibet Exped. 12-0689 (PE) and Qinghai-Xizang Exped. 74-2807 (PE) from Cuona (not shown here).

Wang (2001) cited the collection J.M. Delavay 3264 (P; Fig. 5 ), and Elliott (2016) specified a sheet of this collection (P00145083; Fig. 5A ), as the holotype of Clematis connata var. trullifera . Elliott (2016) wrongly gave the collection date as 20 January 1887 and the collection number as J.M. Delavay 3127. In the protologue of C. buchananiana var. trullifera , the basionym of C. connata var. trullifera , however, Franchet (1889) did not cite J.M. Delavay 3127 and J.M. Delavay 3264; he cited only a single collection, i.e. J.M. Delavay s.n. (20 January 1887) ( Fig. 6 ). The holotype citation by Wang (2001) and Elliott (2016) is therefore not correct. It seems that Wang (2001) and Elliott (2016) did not check carefully the protologue of C. buchananiana var. trullifera and were misled by the three sheets of the collection J.M. Delavay 3264 at P (two are shown in Figure 5 ), one of which (barcode P00145083) was annotated as type, and the other two (barcode P00145084 and barcode P00145085, respectively) were annotated as isotypes. The collection J.M. Delavay s.n. (20 January 1887) ( Fig. 6 ) cited by Franchet (1889) under C. buchananiana var. trullifera includes two sheets at P (barcode P00145081and barcode P00145082). The two specimens, both in fruiting, agree with the original description (it is strange that in the original description floral characters were given but no flowering collection was cited). As there is no evidence that only one was used to prepare the original description, the two specimens are syntypes. Here, in conformity with ICN Art. 9.10 ( Turland et al. 2018) and as discussed by McNeill (2014), we have chosen the sheet with the barcode P00145081 of the collection J.M. Delavay s.n. (20 January 1887) as the lectotype of C. connata var. trullifera , for this sheet is much better preserved.

When Rehder (1936) reduced Clematis coriigera to the synonymy of C. trullifera , he cited E.E. Mair e s.n. as the holotype of C. coriigera but did not state the lodgment of herbarium. Wang (2001) cited E.E. Mair e s.n. (E), and Elliott (2016) further specified the E sheet with the barcode E00081543, as the holotype of this name. The collection E.E. Mair e s.n. includes two sheets at E ( Fig. 3 ). The two specimens, both of which are well preserved and agree with the original description, are syntypes as there is no evidence that only one was used to prepare the original description. Here, in conformity with ICN Art. 9.10 ( Turland et al. 2018) and as discussed by McNeill (2014), the holotype citation by Rehder (1936) is corrected above to lectotype. Furthermore, in order to more precisely typify the name C. coriigera , a second step lectotypification is undertaken according to ICN Art. 9.10 ( Turland et al. 2018). The E sheet of the collection E.E. Mair e s.n. (barcode E00081543) has been chosen as the lectotype of D. coriigera , and the duplicate of this collection at E (barcode E00081544) has been cited as an isolectotype. It is worth mentioning that according to ICN Art. 7.11 ( Turland et al. 2018), Wang (2001) and Elliott (2016) did not effectively published the typifications of the name C. coriigera because their typification statements did not include the phrase “designated here” (hic designatus) or an equivalent. The holotype citations by them, therefore, cannot be considered the first step lectotypification for C. coriigera .

Figures

FIGURE 1. Holotype (A) and isotype (B) sheets of Clematis dongchuanensis (= C. connata var. trullifera). Arrows indicate the slightly dilated petiole bases.

FIGURE 2. Four specimens of Clematis connata var. trullifera (E.E. Maire s.n., P) (A−D) from Dongchuan in northeastern Yunnan, China, the type locality of C. dongchuanensis (= C. connata var. trullifera).

FIGURE 3. Lectotype (A) and isolectotype (B) sheets of Clematis coriigera (= C. connata var.trullifera) from Dongchuan in northeastern Yunnan, China.

FIGURE 5. Two specimens of a collection of Clematis connata var. trullifera (A, B), J.M. Delavay 3264 (P), from Heqing in northwestern Yunnan China. Note that Wang (2001) wrongly cited this collection, and Elliott (2016) wrongly specified a sheet of this correction (P00145083; A), as the holotype of C. connata var. trullifera.

FIGURE 6. Lectotype (A) and isolectotype (B) sheets of Clematis connata var. trullifera.