The taxonomic placement of Pleurospermum lecomteanum, previously synonymized with Pleurospermum wilsonii, was carefully examined using herbarium specimens and molecular evidence. The results showed that Pleurospermum lecomteanum is distinguished from P. wilsonii by several morphological characters. Its phylogenetic position is separate from P. wilsonii in the ML tree. Therefore, Pleurospermum lecomteanum should be restored as a distinct species.

Apiaceae, Pleurospermum, resurrection, synonym, taxonomy

Of the four major worldwide distribution centers of Apiaceae, China has the highest taxonomic diversity at the species level (614–657 species), and represents approximately 1/5 of all species recognized within the family (Sheh and Shu 1987; Sheh et al. 2005; Pimenov 2017). However, numerous species in Chinese Apiaceae remain rather enigmatic and have not been investigated adequately, even morphologically, because of their remote distribution and inadequate number of collections (Zhou et al. 2008; Pimenov 2017). Therefore, in the past few years, large efforts have been devoted to field investigations and examination of herbarium specimens towards a comprehensive understanding of the species of Chinese Apiaceae.

Until recently, Pleurospermum Hoffm. was treated as comprising about 50 species widely distributed in northern Asia and East Europe, of which 39 were in China (Pan and Watson 2005). Taxonomic changes by Pimenov and Kljuykov (2000a, b), Pimenov (2017), Pimenov et al. (2000) and Zhou et al. (2020a, 2021), have reduced that number considerably. However, relationships among some synonymous species are still ambiguous.

This statement also belongs to the widely accepted Pleurospermum wilsonii H.de Boissieu. It was described by H.de Boissieu based on the collections from western China in 1906. In the past years, several taxa have been included within it, namely Physospermopsis lalabhduriana Farille & S.B.Malla, Pleurospermum cnidiifolium H.Wolff, P. crassicaule H.Wolff, P. lecomteanum H.Wolff, P. tanacetifolium H.Wolff and P. thalictrifolium H.Wolff (Shan and Sheh 1979; Pan and Watson 2005; Pimenov 2017). Pleurospermum tanacetifolium and P. thalictrifolium were later merged with Pleurospermum davidii Franch. and Pleurospermum astrantioideum (H.de Boissieu) K.T.Fu et Y.C.Ho, respectively (Pimenov 2017). Pimenov and Kljuykov (2000a) have since transferred Pleurospermum wilsonii, P. davidii and P. astrantioideum to Hymenidium, but these taxonomic novelties need further confirmation based on the extensive molecular analysis (Zhou et al. unpublished data).

During our fieldwork in western China, we discovered several populations with morphological characters that are different from Pleurospermum wilsonii (10–25 cm tall, stem sometimes shortened vs. 15–60 cm tall, stem elongated; 2–3-pinnatisect, ultimate segments narrowly ovate or lanceolate, entire or 2–3-lobed vs. 1-pinnate, or subbipinnatisect with ultimate segments ovate or suborbicular, base cuneate, margins irregularly serrate to deeply lobed; rays 8–15, unequal or equal vs. 10–25, subequal). After consulting the relevant protologues (Boissieu 1906; Wolff 1925, 1926, 1929; Farille et al. 1985) and type specimens for each of the names, we consider that the population from Chayu County, Tibet, is identical with P. wilsonii, while the populations from Sichuan and Qinghai provinces correspond to P. lecomteanum, based on the morphology. A further analysis of comparative DNA sequences is presented here to clarify the taxonomic relationships between P. lecomteanum and P. wilsonii, and to identify their potential close relatives within the molecular framework of Apiaceae subfamily Apioideae.

The phylogenetic results revealed that all accessions of Pleurospermum wilsonii allied together, and constituted a sister group relationship with the clade of Hymenidium huzhihaoi Pimenov & Kljuykov and P. lecomteanum (Fig. 3, the complete tree containing all sampled representatives is available upon request). The whole clade fell within the tribe Pleurospermeae, and showed close relationship with the clade of Hymenidium lindleyanum (Klotzsch) Pimenov & Kljuykov, Hymenidium stellatum (D.Don) Pimenov & Kljuykov and Trachydium roylei Lindley. The pairwise sequence divergence value between P. wilsonii and P. lecomteanum was 3.44%.

Figure 3. 

The consensus tree obtained from maximum likelihood analysis of 2920 nrDNA ITS sequences from Apiaceae subfamily Apioideae showing the tribe Pleurospermeae, with support values (≥50%) provided next to the branches. The nine newly sampled accessions are shown in bold.

Recently, the circumscription of Pleurospermum was reduced to comprise only two species (the type species P. austriacum L., and P. uralense Hoffm.), while the other species were referred to Aulacospermum, Hymenidium, Hymenolaena, Physospermopsis, and Pterocyclus (Pimenov and Kljuykov 2000a, 2000b). However, only two of these genera, Aulacospermum and Hymenolaena, were supported as monophyletic groups in the molecular phylogenetic study by Valiejo-Roman et al. (2012). Hymenidium seems to be non-monophyletic, with its members assigned to the Acronema clade, the East-Asia clade, the Sinodielsia clade, the Pleurospermopsis clade and the Pleurospermeae (Zhou et al. 2009, 2020b; Wei 2020). Furthermore, the species of Hymenidium did not ally as monophyletic within three of the above clades. Pimenov & Kljuykov (2000a) indicated that Hymenidium in current circumscription is a genus with ambiguous taxonomy, and probably comprised several more distinct species groups. Recently, we conducted molecular phylogenetic studies for Pleurospermum and related genera, in which we sorted members of Pleurospermum into major clades of Apioideae, assessed relationships of these members to other apioid taxa within each of these major clades (Wei 2020), transferred Pleurospermum bicolor (Franch.) C.Norman ex Z.H.Pan & M.F.Watson into the genus Pleurospermopsis as Pleurospermopsis bicolor (Franch.) J.Zhou & J.Wei (Zhou et al. 2020a), confirmed the status of Pterocyclus as a separate genus with four species (Pterocyclus angelicoides (Wallich ex DC.) Klotzsch, P. rotundatus (DC.) Pimenov & Kljuykov, P. forrestii (Diels) Pimenov & Kljuykov, and a restored species, P. wolffianus Fedde ex H.Wolff; Zhou et al. 2021). All of these studies have enhanced our understanding of Pleurospermum and related genera, and brought us one step closer towards a more natural classification system.

Pleurospermum lecomteanum was described by H.Wolff based on collections from China in 1929. In Flora Reipublicae Popularis Sinicae, along with P. cnidiifolium, P. tanacetifolium and P. thalictrifolium, it was synonymized with P. crassicaule (Shan and Sheh 1979). All of the above species, plus Physospermopsis lalabhduriana, were included within P. wilsonii in the Flora of China (Pan and Watson 2005). After consulting the types of P. tanacetifolium (GB0048823) and P. thalictrifolium (GB0048825), we consider that it is reasonable to merge them with P. davidii and P. astrantioideum, respectively as proposed by Pimenov (2017). Physospermopsis lalabhduriana was described based on specimens from Nepal in 1985 (Farille et al. 1985). The morphology of the leaf and bracteoles of the isotype (E00000214) is different from that of P. wilsonii. It should be regarded as a distinct species that is not distributed in China (Pimenov 2017). Among the names treated as synonyms of P. wilsonii, Pleurospermum crassicaule and P. cnidiifolium were each described by H.Wolff in 1925 and 1926, respectively. We have examined their type materials (holotype P00834555 for P. crassicaule and syntype PE00033257 for P. cnidiifolium), and consider that they cannot be separated from P. wilsonii and should be merged into a single species. However, we found a set of morphological characteristics, including the stem length, shape and division of leaves and pinnae, as well as the number and length of rays (Table 1), distinguished P. lecomteanum from P. wilsonii. In our field investigation in Qinghai and Sichuan provinces, we collected several specimens of P. lecomteanum, whose morphology is exactly the same as the type (P00834556 and P00834557), that led us to observe its morphology more carefully and reassess the status of this taxon. In our molecular analysis, the accessions of P. lecomteanum allied as monophyletic, and comprised a sister group relationship with Hymenidium huzhihaoi.

Hymenidium huzhihaoi was a species recently described by Pimenov and Kljuykov (Pimenov and Kljuykov 2004). It was distinguished from P. lecomteanum by being subacaulescent, umbellules compact, and apex of bracteoles 3–10-lobed, or rarely entire, and by its 1.23% nucleotide divergence. Therefore, both morphological and molecular evidence support recognition of P. lecomteanum as a distinct species. Since Hymendium is a polyphyletic genus need to be further revision, and its type species (H. brunonis (DC.) Lindl.) does not fall into the Pleurospermeae to which P. lecomteanum belongs, so we here merely restore it as a distinct species without further taxonomic treatment.

We are grateful to reviewers for helpful comments on the manuscript and thank David E. Boufford from Harvard University Herbaria for language polishing. This work was supported by the National Natural Science Foundation of China (no. 31960048 and 31872649), the Ten Thousand Talents Program of Yunnan (YNWRQNBJ-2019-208), the department of Science and Technology of Yunnan Province (202201AT070118), the Hundred Talents Program of Kunming Medical University (no. 60118260127) and Gaoligong Mountain, Forest Ecosystem, Observation and Research Station of Yunnan Province (no. 202205AM070006).