Finding missing diversity from synonyms of Haplopteris (Pteridaceae)

Abstract Although taxonomists target the remote wild regions to discover new species, taxa lacking a comprehensive and modern systematic treatment may be the new hotspot for biodiversity discovery. The development of molecular systematics integrated with microscopic observation techniques has greatly improved the ability of taxonomists to identify species correctly. Vittariacentrochinensis Ching ex J.F. Cheng, regarded as a synonym of Haplopterisfudzinoi (Makino) E.H.Crane, remained hidden from the eyes of fern taxonomists for more than 20 years. Herein, we collected several population samples of V.centrochinensis by performing molecular phylogenetic analysis of five cpDNA regions (rbcL, atpA, matK, ndhF, and trnL-trnF) and through micromophological observation of specimens which differs from H.fudzinoi by lamina width and exospores. Considering the differences in morphology, geographical range, and genetic distance between these two species, we formally recognized V.centrochinensis as an authentic species and proposed a new combination Haplopteriscentrochinensis (Ching ex J.F.Cheng) Y.H.Yan, Z.Y.Wei & X.C.Zhang, comb. nov. Our findings demonstrate that several taxa in synonyms are missing, and nowadays taxonomy should also include re-evaluation of the past taxonomy.


Introduction
The question "How many species are there on earth?" is one of the top 125 questions in science, and exploring it is considered equivalent to imagining the number of stars in the sky (Kennedy and Norman 2005). To understand the biodiversity of species, taxonomists should not only explore new species but also re-evaluate the published species' names that are considered synonymous with older species' names. According to the stasis of the web of TPL (The Plant List 2013), more than onethird of species names are unclear and approximately one-third of species names are considered synonymous. Unfortunately, once a species name is treated as a synonym, it remains in the pile of synonyms forever. With the development in molecular phylogeny research, DNA barcoding, and detailed taxonomic observations, an increasing number of species names have been re-established from the checklist of synonyms, which ranges from ferns (Shu et al. 2017;Shu et al. 2018;Wei et al. 2018;Wang et al. 2020) to spermatophyte (Luo et al. 2016;Jin et al. 2017;Wang et al. 2018). Consequently, we found that the synonym database could be a new hotspot for biodiversity discovery.
Accurate specimen identification through sequencing of the type specimens or samples from type locality is the key to solving questions regarding taxonomic synonyms. In addition, a clear understanding of the taxonomic status and barcoding database of the species suspected of being independent is required. Haplopteris C.Presl is a genus of vittarioid ferns, long treated as a synonym of Vittaria Sm. (Kramer 1990;Wu and Ching 1991;Chen et al. 2013). Recently, it has been widely adopted and followed that the Old World Vittaria species were transferred to Haplopteris (Chen et al. 2013). Because of limited morphology exhibited by members of the Haplopteris as well as convergent and parallel evolution of morphological characteristics, the circumscription of species in the genus has been ambiguous. Vittaria centrochinensis Ching ex J.F.Cheng, just one of a long sleeping suspicious species in the synonym list of Haplopteris, was initially published in "Flora of Jiangxi" (Xu 1993) as a new species and subsequently considered as a synonym of V. fudzinoi (Zhang 1999). To date, it has been regarded as a synonym of Haplopteris fudzinoi (Makino) E.H.Crane (Zhang and Gilbert 2013;Yan et al. 2016). Haplopteris fudzinoi is a species originally described in Japan, and then used to refer to a Chinese fern (Zhang 1999). During our field investigation in Wuyi Mountain (Jiangxi, China), we collected some population samples of V. centrochinensis and found some obvious differences between V. centrochinensis and H. fudzinoi.
In this study, we analyzed morphological characteristics and geographic distribution along with the molecular phylogeny to confirm the identity of V. centrochinensis and phylogenetic affinities of this species with H. fudzinoi. We hope that this study can provide a paramount example of re-evaluating of synonyms for new insights into biodiversity discovery.

Morphological analyses and geographical distribution
For morphology, the H. centrochinensis was compared with similar species by analyzing photographs of type specimens and field photos. The features of rhizome scales were obtained using Nikon SMZ-1500 (Japan). The morphology of spores was observed with a Quanta 250 scanning electron microscope (FEI, USA), and spore size was measured using ImageJ software (Collins 2007). The descriptions of spore ornamentation abided by Wang and Dai (2010) and Ranker et al. (1993). The map of the geographical distribution of two species, namely H. centrochinensis and H. fudzinoi, was obtained through field investigation and National Specimen Information Infrastructure (NSII). The specimens of H. centrochinensis in this study were deposited in Shanghai Chenshan Herbarium (CSH).

Phylogenetic analyses
The total genomic DNA was extracted from silica-dried leaves by using a plant total genomic DNA kit (Tiangen, Beijing, China), according to the manufacturer's instructions. The primers used for amplification and sequencing were shown in Table 1. Sequencing was performed with an ABI 3730xl DNA analyzer (Applied Biosystems, Foster City, CA, USA). The cpDNA sequences of the three samples of H. centrochinensis were submitted to GenBank under accession numbers: MW810047-MW810061 (Table 2). In addition, five cpDNA regions of nine species were downloaded from GenBank (Table 3). Of these, the unavailable data (Table 3) were treated as missing data when they were concatenated. The newly generated sequences were assembled and edited using SeqMan (Burland 1999). Subsequently, all sequences were aligned and manually adjusted on MEGA software (v7.0) (Kumar et al. 2016), with default alignment parameters. Alignments of five cpDNA regions were concatenated using PhyloSuite . Then the matrix was used to construct phylogenetic trees with maximum likelihood (ML) and MrBayes. Maximum likelihood analysis was conducted using IQ-TREE (Nguyen et al. 2015) integrated in PhyloSuite with standard bootstrap and TVM+F+G4 model. Bayesian analysis was performed using MrBayes (v3.2.6) (Ronquist et al. 2012) with the GTR+F+G4 model. Four Markov chains were  run 1,000,000 generations, with the sampling frequency of 100. The standard deviation of split frequencies was set to less than 0.01 to achieve the convergence of the independent runs. A majority-rule consensus tree was constructed to estimate the posterior probabilities (PP); the first 25% of samples were discarded as the burn-in phase.

Morphological comparisons and geographical distribution
The morphological and micromorphological characters of H. centrochinensis and H. fudzinoi are presented in Figure 1 and Table 4. The lamina of H. centrochinensis was shorter and wider than that of H. fudzinoi (Fig. 1A, B, F); in H. fudzinoi costa it was raised adaxially with two prominent long grooves beside the costa on adaxial surface (Fig. 1D). The rhizome scales were yellow-brown, margin denticulate, linear-lanceolate, and clathrate (Fig. 1E left); scale margins of H. centrochinensis were toothed and those of H. fudzinoi were subentire to minutely denticulate at lower margin and upper part, respectively. The scales length of H. centrochinensis were visibly longer than that of H. fudzinoi (Fig. 1E). Spores were monolete for both the species. Spore ornamentation observed in H. centrochinensis was scabrate and rugate (Fig. 1G, I), whereas it was laevigate or inconspicuous-granulate in H. fudzinoi (Fig. 1H, J). Additionally, sori position was distinct between the two species; the soral line in H. fudzinoi was located close to the edge of lamina and immersed in groove (Fig. 1D), whereas it was immersed between the frond  costa and margin in H. centrochinensis (Fig. 1C). The geographical distribution for these two species was obtained on the basis of the information of the specimens. The result indicated that most distributions are shared by H. centrochinensis and H. fudzinoi (Fig. 3).
Haplopteris centrochinensis is an independent species according to molecular data obtained using phylogenetic analysis The two phylogenetic analyses (BI, ML) recovered congruent topologies, with Antrophyum parvulum and Antrophyum sessilifolium as outgroups (Fig. 2). The results revealed strong support for the monophyly of H. centrochinensis (PP = 1.0, BS = 100) (Fig. 2), and it was strongly supported as a sister to another four species (i.e., H. fudzinoi, H. doniana, H. taeniophylla, and H. linearifolia) of Haplopteris (PP = 1.0, BS = 100) (Fig. 2). The genetic distance between the H. centrochinensis and these four Haplopteris species ranged from 0.073 to 0.120, and the intraspecific distances of these four species were 0 and 0.001 (Table 5).

Re-evaluation of synonyms for new insights into biodiversity discovery
Synonym is the first concern in the estimation of the total number of species in one taxon, and only after its resolution can one ask the next question regarding how many additional species there are in the taxon (Joppa and Pimm 2011). Surprisingly, near-ly two-thirds of the plant names are synonymous or recorded as unresolved in TPL (2013), which consists of 26,000 additional synonyms that were not listed in its earlier version (v.1.0). The increase in the number of newly discovered species has been consistent in line with the use of molecular evidence; however, information on synonyms is meager. For instance, International Plant Names Index (IPNI 2020) provides information on nomen novum, combination nova, and taxa nova, but it provides no information on new species resurrected from the established synonyms. Although many species of ferns Morigengaowa et al. 2018;Shu et al. 2017;Shu et al. 2018;Sigel et al. 2014;Wang et al. 2020;Wei et al. 2018) and seed plants (Tkach et al. 2015;Hu et al. 2015) have been reinstated as independent species, this type of study has been rarely reported. The extent of biodiversity hidden in taxonomic literature is an interesting question to explore. To understand the biodiversity of our planet and for efficient conservation of valuable natural resources, the prime objective of taxonomists should be to correctly identify our planet biodiversity by using modern taxonomic facilities. Various taxa, especially widely distributed ones, still require a comprehensive systematic treatment that also involves evaluating their nomenclature. Then, if cryptic taxa or misunderstood species have to be segregated, naming these taxa needs first to be evaluated against synonymy as potential sources of the needed name, otherwise a new name needs to be proposed. However, the number of taxonomists has significantly declined (Haas and Hauser 2005), and young taxonomists do not pay enough atten- tion either to the evaluation of synonymous names already listed in the taxonomy or to the assessment of thousands of different species names and their type specimens. Therefore, the number of species on earth remains uncertain. Fortunately, technological advancement has considerably affected taxonomy. According to the IPNI data, two obvious turning points in taxonomy have affected the trends in new combinations and new taxa (Fig. 4), and those are related to the development of electron microscopy and molecular phylogeny in the 1970s and the 1990s, respectively (Endess et al. 2000). Although the new taxa have been displaying a steady or even a downward trend, the new combination is expected to display an upward trend in future, with the application of molecular biology in taxonomy (Fig. 4). Unsurprisingly, new combinations will continue to occur for a long time because of the abundance of listed synonyms and suspected species names, which are equivalent to the new species in the wild. Thus, synonyms and suspected species will serve as the new biological diversity hotspot for the exploration of new unknown species.

Integrative taxonomy contributes to clarifying species delimitation
The reason for numerous synonyms existing only in books may be the lack of sufficient morphological judgments made in the past. In the present study, the phylogeny (Fig. 2) based on the 5-locus dataset revealed strong support for the monophyly of H. centrochinensis (PP = 1.0, BS = 100) (Fig. 2) and strong support for H. centrochinensis as sister to H. fudzinoi (PP = 1.0, BS = 100) (Fig. 2). Although our result was different from that of a study by Zhang and Gilbert (2013) that indicated distinction between the two species, no further research was performed merely because of limited conditions at that time. Moreover, genetic distance in line with the K2P model showed an obvious hereditary difference between the H. centrochinensis and another three Haplopteris species (Table 5). For morphological comparisons, several traits were observed. Of these, the most  (1970-2020). unambiguous differences between the two species (H. centrochinensis and H. fudzinoi) were the wider lamina, longer rhizome scale, and shorter stipe in H. centrochinensis. In addition, H. fudzinoi costa was raised adaxially with two prominent long grooves besides the costa on adaxial surface. Furthermore, the scabrate and rugate ornamentation of spores observed in H. centrochinensis was found to be laevigate or inconspicuousgranulate in H. fudzinoi. To sum up, monophyletic clade, long genetic distance, stable morphological differentiation, and independent geographical distribution form the basis of establishing H. centrochinensis as a divergent species or an independent species, and therefore, it should not be considered synonymous with H. fudzinoi.