Spiranthes hachijoensis (Orchidaceae), a new species within the S. sinensis species complex in Japan, based on morphological, phylogenetic, and ecological evidence

The systematics of the Old World Spiranthes sinensis (Pers.) Ames species complex (Orchidaceae) has been complicated by its wide distribution and morphological variations. Within the species complex, S. australis Lindl. has been generally accepted as the only Spiranthes Rich. species distributed on the Japanese mainland. The present study provides morphological, phylogenetic, and ecological evidence for the recognition of S. hachijoensis Suetsugu as a new species of the S. sinensis species complex on the Japanese mainland. Spiranthes hachijoensis is morphologically similar to S. hongkongensis S.Y. Hu & Barretto and S. nivea T.P. Lin & W.M. Lin, sharing a degenerated rostellum, pollinia without a viscidium, and distinctly trilobed stigma. However, the taxon can be morphologically distinguished from S. hongkongensis by its glabrous rachis, ovaries, and sepals, and from S. nivea by its papillate labellum disc, larger papillate basal labellum callosities, and glabrous rachis, ovaries, and sepals. The autogamy and flowering phenology (i.e., earlier flowering) of S. hachijoensis are most likely responsible for premating isolation from the sympatric S. australis. A MIG-seq-based high-throughput molecular analysis indicated that the genetic difference between S. hachijoensis and its putative sister species S. sinensis is comparable to, or even greater than, the genetic difference between pairs of other species within the S. sinensis species complex. Our multifaceted approach strongly supports the recognition of S. hachijoensis as a morphologically, phenologically, phylogenetically, and ecologically distinct species. Supplementary Information The online version contains supplementary material available at 10.1007/s10265-023-01448-6.


Introduction
The genus Spiranthes Rich. (Orchidaceae) includes approximately 50 species that are widely distributed across the tropical and temperate regions of the Americas, Eurasia, and Australia (Dueck et al. 2014;Pace et al. 2019;Surveswaran et al. 2017Surveswaran et al. , 2018. Nonetheless, the delimitation of closely related Spiranthes species based on morphology alone is hindered by phenotypic plasticity, convergent morphological features, and hybridization (Dueck et al. 2014;Pace et al. 2019;Suetsugu et al. 2020;Surveswaran et al. 2017Surveswaran et al. , 2018Tao et al. 2018). In particular, the systematics of the Old World S. sinensis (Pers.) Ames species complex has been complicated by the wide distribution and morphological variation of the group (Hsu and Chung 2014;Hu and Barretto 1976;Lin and Lin 2011).
Recently, the systematics of the S. sinensis species complex has partially been resolved using molecular approaches (Pace et al. 2019;Surveswaran et al. 2017Surveswaran et al. , 2018. Working under an integrative phylogenetic species concept, several recent studies have performed extensive molecular phylogenetic sampling of the S. sinensis species complex Surveswaran et al. 2017Surveswaran et al. , 2018. The combination of this molecular evidence with phenological and morphological data has facilitated the recognition of seven distinct taxa within the complex (Pace et al. 2019;Surveswaran et al. 2018). However, although the recognition of seven distinct taxa within the complex is plausible,, there is some controversy over the most appropriate scientific names for the seven taxa due to their complex morphological features (Pace and Cameron 2020;Surveswaran et al. 2020). Here, we basically followed , with slight modification as detailed by , accepting the following seven species (S. australis Lindl., S. flexuosa Lindl., S. hongkongensis S.Y. Hu & Barretto, S. maokensis M.C. Pace, S. nivea T.P. Lin & W.M. Lin [replacing S. suishaensis in Pace et al. (2017)], S. sinensis, and S. sunii Boufford & Wen H. Zhang) within the S. sinensis species complex. Nonetheless, given that phenological divergence could have facilitated speciation without other observable or measurably significant morphological differences (other than pubescence; Pace et al. 2019) and that some taxa within the S. sinensis complex have recently been recognized (Pace et al. 2019;Surveswaran et al. 2017), it is likely that more species remain unrecognized and misidentified within widespread taxa such as S. australis and S. sinensis.
Three species (S. australis, S. sinensis, and S. hongkongensis) have been recognized in Japan. It should be noted that, although Pace et al. (2019) reported S. flexuosa from Japan, this was likely the result of an incorrect assignment, as discussed previously (Surveswaran et al. 2020). Our intensive herbarium and field surveys failed to provide evidence of the occurrence of S. flexuosa in Japan. Spiranthes sinensis can be characterized by its phenology (spring flowering) and glabrous rachis and ovaries (Hayakawa et al. 2013;Pace et al. 2019;Suetsugu and Hayakawa 2016). Meanwhile, S. australis can be characterized by its early summer or autumn flowering phenology and glandular pubescent rachis, stems, and ovaries (Hayakawa et al. 2013;Pace et al. 2019;Suetsugu and Hayakawa 2016;Tsukaya 1994Tsukaya , 2005. Finally, S. hongkongensis, with glandular pubescent rachis, floral bracts, ovaries, and sepals, can be distinguished from both S. sinensis and S. australis by its degenerated rostellum, pollinia without a viscidium, shorter column, and distinctly trilobed stigma (Hsu and Chung 2014;Hu and Barretto 1976;Suetsugu and Hayakawa 2019). These species also show some geographic segregation, at least in Japan and Taiwan. Spiranthes australis is distributed on mainland Japan (northern Ryukyu Islands and northward), whereas S. sinensis and S. hongkongensis are restricted in the more southern areas (central and southern Ryukyu Islands and Taiwan [S. sinensis] and Ishigaki Island, southern Ryukyu, and Taiwan [S. hongkongensis]) (Maekawa 1971;Satomi 1982;Suetsugu and Hayakawa 2019). Although glabrous Spiranthes individuals have been reported from mainland Japan on rare occasions, molecular analyses have suggested that most of these individuals are variants of S. australis, rather than a range extension of S. sinensis (Hayakawa et al. 2013;Suetsugu and Hayakawa 2016).
During extensive field surveys focusing on Japanese Spiranthes individuals for phylogeographic studies (Suetsugu, unpublished data), multiple populations of the unknown Spiranthes taxon with glabrous rachis and ovaries have been found on mainland Japan. Interestingly, the unknown taxon often co-occurs with S. australis but flowers approximately 1 month earlier, thereby isolating the taxa reproductively. Given that the features "pubescence" and "flowering time" are often used as diagnostic traits in Spiranthes (Hayakawa et al. 2013;Pace et al. 2019;Suetsugu and Hayakawa 2016), the glabrous individuals may reflect a range extension of S. sinensis. Meanwhile, the geographical separation of S. sinensis and the unknown taxon suggests that the glabrous individuals are variants of S. australis (Maekawa 1971;Hayakawa et al. 2013;Satomi 1982;Suetsugu and Hayakawa 2016). However, additional morphological curiosities, such as a weakly open flower and a weakly recurved labellum (usually recurved by < 270°), which are not typical variations of either S. sinensis or S. australis (Surveswaran et al. 2017), suggest that the individuals represent an overlooked species. Accordingly, we have used an integrative taxonomic approach to determine whether the unknown taxon represents a distinct entity within the S. sinensis complex. Species delimitation that explicitly considers ecological and phylogenetic differences has critically advanced the current understanding and evaluation of biodiversity (Barrett and Freudenstein 2011). Over the last two decades, these integrative taxonomic approaches have provided more robust estimates of biodiversity among taxonomically challenging species (Barrett and Freudenstein 2011;Barrett et al. 2022;Botes et al. 2020;Pace et al. 2019;. The present study provides morphological, phylogenetic, and ecological evidence that supports the recognition of S. hachijoensis Suetsugu, named after its type locality, which sustains the largest population of the taxon.

Morphological observations
Morphological characteristics of the S. sinensis species complex were compared using herbarium specimens from CBM, KYO, KPM, TAIF, TI, and TNS and living individuals of S. australis, S. flexuosa, S. hachijoensis, S. hongkongensis, S. nivea (including S. nivea var. papillata), and S. sinensis collected in Laos, Japan, and Taiwan during fieldwork between 2012 and 2022 (Table S1). Morphological characters were visually observed under a Leica M165C stereomicroscope and measured using a digital caliper. The dissected flowering specimens were photographed using an Olympus OM-D E-M1 Mark II digital camera equipped with an Olympus 30 mm macro lens or a Leica MC170 HD digital camera attached to a Leica M165C stereo microscope. Morphological differences among species of the S. sinensis complex, including S. maokensis and S. sunii which we could not obtain living materials for analysis, were investigated by reviewing relevant literature and online digitized plant collections such as JSTOR Global Plants (http:// plants. jstor. org/). At least one voucher specimen for newly collected samples from each population during our field survey was deposited in KYO, TAIF, TNS, RYU, and SPMN (Table S1). The herbarium acronyms follow the Index Herbariorum (Thiers 2022).

Reproductive biology
The reproductive biology of a S. hachijoensis population in Ichihara-shi (Chiba Pref., Japan) was investigated from mid-May to early-June, 2016. Because previous studies suggested that closely related Spiranthes species are pollinated by bees (e.g., Megachile and Apis species; Iwata et al. 2012;Suetsugu and Abe 2021;Tao et al. 2018), floral visitors were surveyed during the daytime (10:00-16:00) for a total of 18 h.
To investigate the breeding system of the species, handpollination experiments were performed using five treatments. Because flowers in the upper of the inflorescence often do not bear fruit even when pollinated due to resource limitations, 5 flowers from the base of the inflorescence on each individual were used to accurately determine the effect on pollination. For the autonomous autogamous treatment, flowers were bagged before anthesis using a fine mesh to exclude pollinators (20 flowers from 4 individuals). For the manually autogamous treatment, pollinaria were removed and used to hand-pollinate the same flowers, after which the flowers were bagged with a fine mesh (20 flowers from 4 individuals). For the manually allogamous treatment, pollinaria were removed and used to hand-pollinate flowers on plants at least 1 m away, after which flowers on the recipient plant were bagged with fine mesh (20 flowers from 4 individuals). For the open treatment, flowering plants were randomly tagged and allowed to develop fruit under natural conditions (40 flowers from 8 individuals). The experimental plants were monitored intermittently for fruit development over the following 3 weeks. The proportion of seeds with at least one well-developed embryo was then evaluated by screening 50 randomly selected seeds from each capsule. In addition, because polyembryonic seeds have been associated with agamospermy (vs. monoembryonic seeds with sexual reproduction) in Spiranthes (Catling 1982;Sun 1996), the occurrence of agamospermous seed development was investigated by examining the number of embryos in a seed.
Finally, the fruit set among the treatment groups was compared using the Fisher's exact test. The effects of pollination treatment on seed viability were evaluated using ANOVA.

MIG-seq-based high-throughput genomic analysis
Eight S. australis individuals, twenty-eight S. hachijoensis individuals, and six S. sinensis individuals collected throughout Japan were used for multiplexed inter-simple sequence repeat genotyping (MIG)-seq analysis. Five S. sinensis individuals from Taiwan, four S. hongkongensis individuals from China and Taiwan, three S. nivea individuals (including two S. nivea var. papillata individuals) from Taiwan, and a single S. flexuosa individual from Laos were included in the comparative study (Table S1). Genomic DNA was extracted from silica-dried leaves using the CTAB method (Doyle and Doyle 1990). A MIG-seq library for the 55 Spiranthes samples was prepared as described by Suyama et al. (2022) and sequenced using a MiSeq system (Illumina, San Diego, CA, USA) and MiSeq Reagent Kit v3 (150 cycle; Illumina). The raw MIG-seq data were deposited in the DDBJ Sequence Read Archive (SRA accession number PRJNA907989).
After removing primer sequences and low-quality reads , 15,943,788 reads (289,887 ± 6998 reads per sample) were obtained from 17,873,310 raw reads (324,969 ± 7906 reads per sample). Stacks 2.60 pipeline was used for de novo single nucleotide polymorphism (SNP) discovery (Rochette et al. 2019), with the following parameters: minimum depth of coverage required to create a stack (m) = 3, maximum distance allowed between stacks (M) = 2, and number of mismatches allowed between sample loci while building the catalog (n) = 2. SNP sites with high heterozygosity (Ho ≥ 0.6) were removed, and SNP sites with fewer than three minor alleles were filtered out. Then, a SNP was excluded if the number of samples shared by the SNP was below the reference value R (the minimum proportion of samples that retained a SNP). We used four conditions referring to the threshold for the minimum number of samples that retained a SNP to determine the robustness of the results. Only SNPs retained by 6 (R = 0.1), 17 (R = 0.3), 28 (R = 0.5), and 44 (R = 0.8) or more samples were included in the datasets. Finally, 13,809 SNPs in 6630 loci, 7972 SNPs in 3438 loci, 5275 SNPs in 1976 loci, and 538 SNPs in 203 loci were included for subsequent analysis.
A SNP-based maximum-likelihood (ML) phylogeny was inferred using RAxML v. 8.2.10 (Stamatakis 2014), with a GTR substitution model with Lewis' ascertainment bias correction and 1000 bootstrap replicates. Additionally, a Neighbor-Net network was constructed using the uncorrected p distance matrix and ignoring ambiguous sites in SplitsTree4 4.14 (Huson and Bryant 2006).

Phylogenetic distinctness of Spiranthes hachijoensis
A ML phylogenetic tree reconstructed from MIG-seq data indicated that S. hachijoensis is more closely related to the allopatric S. sinensis than to the sympatric S. australis ( Fig. 1 and Figs. S1, S2). It should be noted that the choice of different values for the parameter R resulted in some variations in the topology of the phylogenetic tree.

S. australis
The ML phylogenetic tree using R = 0.1 suggested that S. hachijoensis forms a relatively well-supported clade with S. sinensis (Fig. 1), while the phylogenetic tree using R = 0.5 or R = 0.8 suggested that S. hachijoensis forms a clade with S. hongkongensis + S. flexuosa, despite weak bootstrap support (Fig. S1). Regardless of the discrepancy among phylogenetic trees reconstructed using different numbers of loci, all the phylogenetic analyses indicated that the genetic difference between S. hachijoensis and S. sinensis is similar to, or even greater than, the genetic differences observed between pairs of other species within the complex (Figs. 1, 2 and Figs. S1, S2). Therefore, all molecular data support the recognition of S. hachijoensis as an independent species.
Although S. hachijoensis is morphologically most similar to S. hongkongensis, the ML and Neighbor-Net phylogenetic analyses also indicated that S. hachijoensis represents a distinct genetic cluster from S. hongkongensis (Figs. 1, 2 and Figs. S1, S2). The molecular results are consistent with our preliminary karyological study showing 2n = 30 for S. hachijoensis (K. Suetsugu & N. Nakato, unpublished data). In contrast, the chromosome counts, analysis of isozyme loci, and molecular phylogenetics indicated that S. hongkongensis is an allotetraploid (2n = 60) resulting from a hybridization event between S. sinensis and S. flexuosa (Pace et al. 2019;Sun 1996;Surveswaran et al. 2018). Furthermore, even in the localities where S. hachijoensis and S. australis grow sympatrically (Toyofusa-shi, Chiba Pref. Remarkably, even though S. hachijoensis is somewhat polymorphic in flower coloration and floral bract length (Fig. 3), the species showed relatively weak genetic variation within and between populations (Figs. 1, 2 and Figs. S1, S2), which may be due to its predominantly autogamous breeding system. Both the ML and Neighbor-Net phylogenetic analyses also indicated that S. nivea var. nivea and S. nivea var. papillata show little genetic differentiation (Figs. 1, 2 and Figs. S1, S2). The phylogenetic data supported S. nivea var. papillate as an intraspecific taxon, despite the morphological distinctness (Hsu and Chung 2016). Overall, the present study indicates that MIG-seq data can provide valuable insight into species delimitation within morphologically ambiguous groups.

Ecological distinctness of Spiranthes hachijoensis
The pollination experiments revealed that S. hachijoensis consistently achieves a high fruit set, even in the absence of pollinators. Therefore, S. hachijoensis reproduction is not pollinator-limited under natural conditions. In addition, the proportion of seeds having a well-developed embryo was not significantly affected by pollination treatment (Table 1). All the examined seeds were monoembryonic or lacked embryos, which suggested that the high fruit set of the species was due to autogamy rather than agamospermy. Further investigation revealed that degenerated rostellum tissue, which allows contact between pollen masses and the upper portion of the stigmatic surface, was the most likely mechanism responsible for autonomous self-pollination in this species.
During floral visitor surveys, syrphid flies (Sphaerophoria sp.) and sweat bees (Lasioglossum sp.) were observed landing and spending time on S. hachijoensis flowers but not entering the inner part of the flowers. Consequently, no pollen grains were introduced to the flowers by the visitors, and the pollinia remained intact within the anther. Thus, no effective pollinators were documented. Notably, S. hachijoensis does not have a viscidium; therefore, its pollinaria exhibit almost no adhesive properties, thereby limiting the attachment of pollinaria to potential pollinators. This provides further evidence that autogamy is the dominant (if not exclusive) reproductive strategy of this species. Furthermore, because autogamy has been recognized as a mechanism of reproductive isolation between sympatric outcrossing and autogamous species (Suetsugu 2022;Sun 1996), the predominantly autogamous breeding system of S. hachijoensis likely facilitates the reproductive isolation of the species from the sympatric and outcrossing species S. australis.
Spiranthes hachijoensis and S. australis are also isolated by their phenology. Such phenological isolation has substantial potential to facilitate reproductive isolation, since asynchrony in flowering time can reduce heterospecific pollen deposition, thereby promoting conspecific mating (Lowry et al. 2008;Pace et al. 2019;Sun 1996). In warm-temperate regions of Japan, S. australis blooms from mid-June to early-July, i.e., approximately 1 month after S. hachijoensis, which flowers from early-May to early-June. Therefore, it is likely that both the autogamy and early flowering phenology of S. hachijoensis contribute to the premating isolation and are responsible for the absence of natural hybrids between S. hachijoensis and S. australis, even where the two species co-occur. Notably, S. hachijoensis and its putative outcrossing ancestral species S. sinensis are also isolated by their disjunct distribution and differential flowering phenology. Spiranthes hachijoensis is distributed at latitudes higher than the Ryukyu Islands, whereas S. sinensis is restricted to the more southern areas (whose northern limit is Amami-Ohshima Island, Ryukyu Islands) (Maekawa 1971;Satomi 1982;Suetsugu and Hayakawa 2019). Moreover, S. sinensis blooms in Japan from late-February to mid-April, i.e., at least 1 month earlier than S. hachijoensis. Accordingly, based on the biological species concept, which defines a species as members of populations that interbreed in nature, S. hachijoensis is also distinct from S. sinensis, due to geographical, flowering phenological, and reproductive isolation. Pace et al. (2019) reported that the component taxa of the S. sinensis species complex exhibit varying degrees of autogamy and suggested that autogamy has contributed to intraspecific morphological variability and, in some instances, speciation. For example, flowers of S. australis in New Zealand (also known as S. novae-zelandiae Hook. f.), which lack a rostellum and viscidium, form a weakly supported phylogenetic subclade within S. australis (Frericks et al. 2018;Pace et al. 2019). However, since similar levels of molecular variation have been documented in other Spiranthes species without morphological divergence , the recognition of S. novae-zelandiae as a distinct taxon has not yet been warranted (Pace et al. 2019). In contrast, the autogamous species S. nivea and S. hongkongensis are usually treated as distinct taxa because they exhibit distinct molecular and morphological characteristics (Pace et al. 2019;Surveswaran et al. 2018). Therefore, given not only reproductive isolation (for the biological species concept) but also genetic isolation (for the phylogenetic species concept) and morphological distinction (for the morphological species concept; see later section), we consider S. hachijoensis to be a separate species. Table 1 Effects of pollination treatment on fruit set, seed mass and proportion of seeds having an embryo in Spiranthes hachijoensis Different superscript letters indicate significant differences (P < 0.05) between treatments. The proportion of seeds having an embryo are expressed by mean ± SD Agamospermy

Morphological distinctness of Spiranthes hachijoensis
The presence or absence of pubescence and viscidium and the shape of the stigma and labellum basal callosities have been considered consistent morphological characteristics for the delimitation of species within the S. sinensis species complex (Pace et al. 2019). Spiranthes hachijoensis is easily distinguishable from the sympatric S. australis by its papillate (vs. glabrous) basal labellum callosities and glabrous (vs. densely pubescent) rachis, ovaries, and sepals (Figs. 3,4,5,6,7,8). Furthermore, S. hachijoensis can be distinguished from its putative parental species, S. sinensis, by its modified floral morphology associated with its autogamy. Specifically, S. sinensis exhibits a larger flower, a strongly recurved labellum that almost bends 360°, a suborbicular stigma, a well-developed rostellum that separates the stigma and pollinarium, and pollinia with a viscidium (Fig. 9). In contrast, S. hachijoensis has a smaller flower, a weakly recurved labellum that typically bends less than 270°, a crescent-shaped stigma, a degenerated rostellum, and pollinia without a viscidium (Figs. 3,4,5,6,7,8). In addition to these morphological differences that result from distinct breeding systems, S. hachijoensis differs from S. sinensis by possessing globose (vs. conical to clavate) labellum basal callosities (Figs. 3,4,5,6,7,8,9). It is also noteworthy that the labellum basal callosities of S. hachijoensis are larger in proportion to the width of the base of the labellum compared to those of S. sinensis (Figs. 3,4,5,6,7,8,9). Consequently, S. hachijoensis is morphologically similar to S. hongkongensis and S. nivea that are also autogamous (Lin and Lin 2011;Pace et al. 2019;Surveswaran et al. 2017). However, S. hachijoensis can be distinguished from S. hongkongensis by its glabrous (vs. densely pubescent) rachis, ovaries, and sepals. Spiranthes hachijoensis also differs from S. nivea in terms of its papillate (vs. nearly glabrous) labellum disc, larger papillate (vs. smaller glabrous) basal labellum callosities, and glabrous (vs. sparsely pubescent) rachis, ovaries, and sepals. Spiranthes hachijoensis appears to also be morphologically similar to S. neocaledonica, which has frequently been treated as a synonym of S. sinensis (Pace et al. 2019). Given that S. neocaledonica was described as having a short column and suppressed rostellum (Schlechter 1906), it is probably autogamous. Unfortunately, the type specimen of S. neocaledonica deposited at B (ZE Botanischer Garten und Botanisches Museum, Freie Universität Berlin) was destroyed during World War II, leaving insufficient reliable data to fully clarify the identity of this species. Nevertheless, according to the protologue (Schlechter 1906), S. hachijoensis differs from S. neocaledonica by having larger basal labellum callosities (0.5-0.7 mm long, versus 0.3 mm long). Moreover, it should be noted that the other autogamous lineages within the S. sinensis complex (i.e., S. hongkongensis and S. nivea) generally exhibit narrow geographical ranges (Pace et al. 2019;. Therefore, given the great geographical disjunction (S. neocaledonica in New Caledonia and S. hachijoensis in Japan) and the somewhat discrepant morphology, it is less likely that S. hachijoensis and S. neocaledonia belong to the same autogamous lineage. Overall, the molecular phylogeny reconstructed from MIG-seq data together with morphological and ecological analyses support the separation of S. hachijoensis as an independent species.
It is also noteworthy that the flower color of S. hachijoensis is highly polymorphic, varying from purple-pink to white (Figs. 3,4,5,6,7,8), although species in the S. sinensis species complex often have pink flowers, and the members with entirely white flowers were traditionally recognized as a separate species only due to their white flowers . Similarly, even though bract length is a diagnostic character for certain Spiranthes species such as S. nivea (Lin and Lin 2011), S. hachijoensis exhibits tremendous variations in bract length (Figs. 3,4,5,6,7,8). However, S. hachijoensis is a phylogenetically unified group with limited intraspecific genetic variation. Furthermore, the molecular phylogenetic analysis also confirmed that S. nivea var. papillata is a variant of S. nivea. However, it is morphologically distinguishable from S. nivea var. nivea by its more densely pubescent rachis and ovaries (vs. sparsely pubescent rachis and ovaries), narrower sepals with white tinged with pink or purple at apex (vs. wider and entirely white sepals), and papillose labellum disc (vs. almost glabrous labellum disc). These results indicate the need to reconsider the diagnostic characteristics of species delimitation in the S. sinensis species complex. As partly suggested by Pace et al. (2019), our molecular results suggested that floral color, bract length, and degree of pubescence should only be considered secondarily important features when investigating the systematics of the S. sinensis species complex, and that other characteristics, such as regional phenology, pubescence (presence/absence, not degree), and labellum and column morphology, should be given more emphasis. In particular, callosity papillae are likely to be a useful morphological feature that has been hitherto overlooked.

Taxonomic treatment
Spiranthes hachijoensis Suetsugu,sp. nov. (Figs. 3,4,5,6,7,8  with inconspicuous pedicel, pale green, ellipsoid-obovoid, 3.7-6.7 mm long, 1.6-2.7 mm wide, glabrous. Flower resupinate, horizontal or nodding, weakly opening. Dorsal sepal entirely white or pinkish purple with a basal white part, narrowly triangular, 3.2-4.3 mm long, 0.9-1.6 mm wide, glabrous, apex obtuse or rarely acute, connivent with petals and forming hoods over column. Lateral sepals entirely white or pinkish purple with a basal white part, narrowly triangular to lanceolate, slightly oblique approximately halfway along length, 3.7-4.2 mm long, 0.7-1.6 mm wide, glabrous, apex obtuse or acute. Petals entirely white or pinkish purple with a basal white part, linear to oblanceolate, occasionally slightly oblique approximately halfway along length, 3.1-3.5 mm long, 0.7-0.9 mm wide, glabrous, apex rounded. Labellum entirely white sometimes tinged with pink or purple at apex, recurved downward approximately two-thirds from claw to labellum apex, oblong to slightly constricted near reflection and then dilating below, centrally papillate near apex, 3.2-3.5 mm long, 1.4-1.8 mm wide below callosities, 1.4-1.9 mm wide at widest point below recurvature; margin entire to slightly undulating from base until area of recurvature, below point of recurvature margin becoming ruffled and lacerate; two basal callosities transparent, globose, papillate, 0.5-0.7 mm long, 0.5-0.8 mm wide. Column translucently white to pale green dorsally, pale green ventrally, clavate, 1.3-2.0 mm long; anther cap yellow-brown, ovate, partly embedded on upper part of column, 0.9-1.1 mm long; pollinia 2, each 2-partite, yellow, granular-farinaceous, without viscidium at narrower end, 0.9-1.1 mm long; rostellum narrow, typically lacking; stigma crescent-shaped, distinctly trilobed filled with viscid liquid, 0.6-1.3 mm long. Fruit for their support during the field studies and/or for providing specimens and photos for this study. We are grateful to Takako Shizuka and Kazuma Takizawa for technical assistance. We also appreciate the curators of CBM, KPM, KYO, TAIF, TI, TNS, RYU, and SPMN for herbaria access. We also thank Hirokazu Tsukaya and Kenta Fujii for valuable discussions on Spiranthes taxonomy. We would like to thank Editage (www. edita ge. com) for English language editing. The line drawings were prepared by Kumi Hamasaki. This study was financially supported by PRESTO (JPMJPR21D6, KS) from the Japan Science and Technology Agency and by the Environment Research and Technology Development Fund (#4-2001, KS and YS) from the Ministry of Environment, Japan.
Author contributions KS planned and designed the research. KS, HH, SF, and T-CH collected the materials. KS, SF, T-CH, and MI obtained the morphological data. KS investigated the reproductive biology. KS, HH, and SKH conducted the molecular experiments. SKH, KS, and YS performed the molecular analyses. KS wrote the article with input from all the authors. All the authors approved the final version of the manuscript.
Funding Open access funding provided by Kobe University.
Data availability statement MIG-seq data are deposited at the NCBI Sequence Read Archive (accession number: PRJNA907989).

Conflict of interest The authors declare no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.