Skip to main content

Ancient allopatry and ecological divergence act together to promote plant diversity in mountainous regions: evidence from comparative phylogeography of two genera in the Sino-Himalayan region



How geographical isolation and ecological divergence act together to promote plant diversity in mountainous regions remains largely unknown. In this study, we chose two genera comprising a small number of species distributed in the Sino-Himalayan region, Megacodon (Gentianaceae) and Beesia (Ranunculaceae), which both exhibit a fragmented distribution pattern and are found across a wide range of elevations. By summarizing their common patterns of speciation and/or divergence processes, we aim to understand how environmental changes accelerated lineage diversification in the Sino-Himalayan region through ancient allopatry and ecological divergence.


Using ddRAD-seq, chloroplast genome sequences, and specific molecular markers, we studied the phylogenetic relationships, population structure, and historical biogeography of Beesia and Megacodon. Both genera began to diverge from the late Miocene onwards, with ancient allopatry at lower elevations formed narrow-range species or relict populations. Mantel tests between genetic distance and climatic, elevational, or geographic distance revealed an isolation-by-distance pattern in Beesia and Megacodon stylophorus. Megacodon showed two clades occupying entirely different altitudinal ranges, whereas Beesia calthifolia exhibited a genetic divergence pattern along an elevation gradient. Furthermore, we conducted morphological measurements on Beesia calthifolia and found that different elevational groups had distinct leaf shapes.


The regional disjunctions of plant groups in the Sino-Himalayan region are drastic and closely related to several biogeographic boundaries. As a consequence of major geological and climate change, ecological divergence when different elevations are colonized often happens simultaneously within plant groups. Although habitat fragmentation and parapatric ecological divergence each spur speciation to different extents, a combined effect of these two factors is a common phenomenon in the Sino-Himalayan region.

Peer Review reports


A plant species usually occupies a specific habitat type because of niche conservatism [1, 2] and tracks suitable ecological niches into a new geographical area (i.e., disperses). Therefore, niche conservatism promotes habitat fragmentation by limiting adaptation to new environments differing from its ancestral niche [3]. However, plant species can also colonize new ecological niches through adaptive changes, resulting in ecological divergence [4]. Mountains provide diverse habitat types and high habitat heterogeneity in topographically complex regions. The patterns of divergence of a plant species in mountainous regions are, therefore, closely related to habitat fragmentation in its ancestral distribution range and its ability to adapt to a novel ecological niche [5, 6]. Habitat reduction or fragmentation in mountainous regions is directly associated with geological and climatic events under natural circumstances [7, 8]. These events may have created a geographical or climatic barrier and can cause local extinctions, but relict populations may be isolated in relatively stable environments (i.e., refugia), resulting in an interruption of gene flow and giving rise to allopatric speciation. Ecological divergence is closely related to environmental gradients which mountain regions can provide, and it can also be promoted by geological and climatic changes. Although both habitat fragmentation and parapatric ecological divergence have been proposed to spur speciation, disentangling the relative effects of those processes in mountain regions is still challenging [9, 10].

The Sino-Himalayan region is a temperate biodiversity hotspot with a high level of species endemism [11,12,13]. It is also a modern distribution and diversification center for temperate flora in the northern hemisphere. The high level of vascular plant richness and endemism in this region have been enhanced by tectonic movement and climatic oscillation since the Miocene [12, 14,15,16]. Allopatric isolation has been proven as an important mechanism for promoting speciation or intra-specific divergence of many plant groups in this region (e.g., [17,18,19]). Several important biogeographic lines have been proposed in this region [20,21,22]. In addition to highly fragmented habitats, this region is also characterized by alpine valleys (2000–6000 m a.s.l.) with highly heterogeneous environments per a very small geographical range. These can promote ecological divergence and local adaptation of different populations of a species. However, few studies have focused on the role of the ecological shift in this region’s speciation or intra-specific divergence [23, 24]. In addition, the same elevation range in different mountain chains in the Sino-Himalayan region may have completely different environments. For example, an area at an elevation of 2000 m in the Sino-Himalayan region can be a dry hot valley, but it can also be a very humid forest. In the heterogenization process of the environment in this region, plants might have to migrate vertically to find suitable habitats to avoid local extinction, giving rise to allopatric divergence and ecological differentiation simultaneously between populations in stable (or ancestral) environments and populations in a variable environment. But until now, this hypothesis has not been fully tested in this region, although some studies have found a simultaneous allopatric and ecological divergence between sister species (or lineages) (e.g., [25,26,27]).

Endemic genera or an endemic clade within genera in a mountain region may be an ideal system for studying specific speciation or divergence processes. Phylogeographic studies of endemic species or genera could reveal their distribution ranges and ecological niche evolution over evolutionary timescales, thereby helping us to understand and explore the relationship between speciation and the environmental changes that took place in a given geographical area [28]. Here, we present population assessments of the evolutionary history of two genera that are mainly distributed in the Sino-Himalayan region: Megacodon (Hemsl.) H. Smith and Beesia Balf. f. & W.W. Sm. These two genera are representative temperate groups that favor a humid environment in or near temperate forests. Megacodon belongs to the tribe Swertiinae of Gentianaceae and contains three species, M. stylophorus, M. lushuiensis, and M. venosus [29,30,31], while Beesia belongs to tribe Cimicifugeae in Ranunculaceae and has two species: B. calthifolia and B. deltophylla. [32,33,34,35,36,37]. These two genera show similarities in their whole biogeographic distribution ranges (Fig. 1a and c), and both offer a fragmented distribution pattern. These two genera are distributed in regions with relatively stable forest ecosystems, such as central China and mountain regions around Sichuan Basin which harbor many relict genera or species. They also both occur in the Hengduan Mountains region, which has experienced drastic tectonic movement and climatic oscillation. Another similarity is that these two genera both occur at a wide range of elevations and grow in (or near) the understory of different temperate forest types.

Fig. 1
figure 1

Geographic distribution and phylogenetic chronogram. (a) and (c), geographic distribution based on records of specimens and Flora of China for Beesia and Megacodon respectively. Shaded areas indicate the distribution areas. Phylogenetic chronogram of Beesia (b) and Megacodon (d) produced by BEAST. The Beesia data were based on 78 chloroplast CDSs, while the Megacodon data were based on ITS, atpB-rbcL and trnL-trnF sequences. Red circle indicates the fossil calibration, green circle indicates the secondary calibration. Blue dots are on well-supported nodes (posterior probabilities > 0.95), grey blue dots are on moderately supported (posterior probabilities ≥ 0.85), and grey dots are on weakly supported (posterior probabilities < 0.85)

The sister species (or clade) pairs in Megacodon occupy very different elevation ranges, grow within completely different vegetation types, and show significant differentiation in their microhabitat. Megacodon stylophorus usually grows at an elevation of above 3600 m and can occur in a variety of alpine habitats, but the two lowland species (M. lushuiensis and M. venosus) usually occur at elevations of below 2000 m, which are greatly affected by human activities. In contrast, the two Beesia species have no obvious differentiation in elevation ranges and microhabitat (both grow in the understory of moist forests). However, B. calthifolia occupies a very large elevation range (from 1700 to 3900 m) and can grow in the understory of different forest types. Therefore, ecological divergence may occur within the species B. calthifolia. Although different plant groups have their own characteristics, different plant groups may share a similar divergence pattern because they share the same topographic and climatic histories in the Sino-Himalayan region.

In this study, we tried to uncover the biogeographic histories of Megacodon and Beesia using various data (ddRAD-seq, chloroplast genome, molecular fragment, and morphological data) to identify similarities in the evolutionary histories of the two genera. The ecological divergence in Megacodon is evident according to our previous study [30]. In order to compare the ecological divergence of these two genera, we perform a study about whether the divergence process in B. calthifolia may have been affected by niche divergence. With this study, we want to test the hypothesis that both habitat fragmentation and ecological adaptation widely influence speciation or divergence processes in the Sino-Himalayan region and their divergence processes are closely related to environmental heterogeneity in this region.


Phylogenetic relationships and molecular dating

The length of whole chloroplast genome of Beesia were ranged from 156,961 to 158,274 (details in Additional file 2: Table S11). The alignment of the 78 non-duplicated chloroplast coding genes for Beesia was 69,256 bp in length. According to plastid phylogenomics, Beesia deltophylla diverged from B. calthifolia in the late Miocene to early Pliocene [5.46 Ma; 95% highest posterior density (HPD): 2.69–7.54 Ma Fig. 1b]. The BC1 population from the western edge of the Sichuan Basin represents the earliest evolutionary branching in B. calthifolia (Fig. 1b). Further genetic divergences of B. calthifolia took place in the late Pliocene to early Pleistocene (Fig. 1b).

Megacodon may have originated during the early Miocene (stem age: 18.9 Ma, HPD: 10.67–28.14 Ma) and diverged into two clades which occurred at different elevations during the Middle to Late Miocene (crown age: 6.28 Ma, HPD: 2.60–12.87 Ma; Fig. 1d). Megacodon stylophorus further diverged into two geographical lineages at 2.56 Ma (1.09–4.77 Ma), and M. venosus and M. lushuiensis also diverged at a similar time (2.77 Ma, HPD: 0.75–6.17 Ma).

Phylogeographical structure

For 119 individuals of Beesia (82 individuals of 12 populations of Beesia calthifolia, 37 individuals of B. deltophylla), a total of 1,279,331,182 clean reads were generated after filtering, with a sequencing depth of 6.02 to 15.49 X (Additional file 2: Table S4). A total of 666 SNPs were obtained for further analysis of Beesia (dataset B119). For all the individuals of the three Megacodon species, a total of 881,394,581 clean reads were generated after filtering, with a sequencing depth of 6.73 to 22.14 X (Additional file 2: Table S3). After SNP discovery using Stacks, 278 SNPs were obtained for all the three Megacodon species (dataset M155). 1777 SNPs were obtained for M. stylophorus (dataset M131).

On the basis of the results of STRUCTURE analysis, we divided the populations of Beesia into five groups (Fig. 2a–c and f). The result of the delta-K distribution showed that the best K value for Beesia was 2 based on the B119 dataset (Additional file 1: Fig. S7). Three populations of Beesia deltophylla formed one group (BD_EH group) consistently from K = 2 to K = 4. Population BC1 from 1700 m a.s.l was identified as being of admixed ancestry between B. deltophylla and B. calthifolia at K = 2 and 3 but it was identified as a unique evolutionary group (BC_QCS group) at K = 4 (Fig. 2a-c). The populations (BC2–BC5) from lower elevation (1800–2200 m a.s.l) and those (BC7–BC12) from higher elevation (2700–4000 m a.s.l) formed as two separate groups (BC_SC and BC_HM group) when K = 3 and 4 (Fig. 2g). Interestingly, samples collected from Lushui (population BC6) at 2660 m a.s.l were identified as of admixed ancestry (BC_LS group) between group BC_QCS and BC_SC at K = 4 (Fig. 2c and g). BC1 and BC6 contained the highest proportion of private alleles among the Beesia species (Additional file 2: Table S5). Furthermore, the BD_EH and BC_QCS groups were divergent from other populations according to the first two axes of the PCA plot, which explained 26.8% and 19.5% of the total variation (Fig. 2d). The third principal component (explaining 8.9%) showed substructure within the BC_SC, BC_LS, and BC_HM groups (Fig. 2e). The results were confirmed by phylogenetic inference (Additional file 1: Fig. S1).

Fig. 2
figure 2

Population delimitation of Beesia (a-g) and Megacodon (h-m). (a-c) Genetic structure of Beesia based on, respectively, K = 2–4 for dataset B119. (d-e) PCA analysis based on dataset B119. (f), geographic distribution of genetic structure of Beesia based on K = 4. Dashed circles in (f) encompass groups inferred by STRUCTURE. A representative leaf from each Beesia group is shown. (h) Genetic structure of Megacodon based on K = 2 for dataset M155. (i) and (j) are genetic structure based on, respectively, K = 2 and K = 3 for dataset M131. (k) PCA analysis based on dataset M155 (outer) and M131 (inner). (l), geographic distribution of genetic structure of Megacodon based on K = 4 for dataset M155. A representative flower from each Megacodon species is shown. Dashed circles in (l) encompass groups of M. stylophorus inferred by STRUCTURE at K = 4. Schematic views of altitudinal distribution of genetic structure and phylogenetic relationships of Beesia (g) and Megacodon (m). Genetic structure for Beesia in (g) were inferred from B119 dataset. Orange dashed line and numbers in (g) show the elevation of BC_HM (> 2700 m) and BC_SC groups (1800–2200 m). Genetic structure for Megacodon in (m) were inferred from M155 datasets. Blue dashed lines and numbers in (m) show elevation of M. stylophorus (> 3690 m), M. lushuiensis (1700 m) and M. venosus (1100 m). Phylogenetic trees at bottom are simplified versions from BEAST results presented in Fig. 1. The estimated geographic regions of occurrence are estimated by BioGeoBEARS. Pie charts in (g) and (m) on nodes represent the relative probabilities of occurrence areas for the Beesia and Megacodon corresponding to (f) and (l) respectively. Reconstructions resulting in more than two possible regions are shown in dark gray. EH, East Himalayas; HM, Hengduan Mountains; Basin: Sichuan Basin; black dashed line: mountains around Sichuan Basin

The results of genetic diversity within populations of Beesia showed that the variation range of nucleotide diversity and observed heterozygosity of B. deltophylla (0.013  π  0.026, 0.019 Hobs 0.023) was lower than that of B. calthifolia (0.017  π  0.051, 0.017 Hobs 0.062; Additional file 2: Table S5). Among the B. calthifolia, the populations of BC_QCS (Private = 14.11%) and BC_LS (Private = 6.91%) contain higher proportions of private alleles than other populations (Additional file 2: Table S5).

Populations of Megacodon formed two clusters in STRUCTURE based on dataset M155. The result of the delta-K distribution showed that the best K value was 2 for both Megacodon and M. stylophorus based on the M155 and the M131 dataset respectively (Additional file 1: Fig. S8). Megacodon venosus and M. lushuiensis, from lower elevations (1100 and 1700 m respectively), formed one group (Fig. 2h and l), while M. stylophorus from higher elevations (3690–4000 m) formed another group (Fig. 2h). The STRUCTURE results based on dataset M131 showed that M. stylophorus could be divided into two major evolutionary clades (MS_EH and MS_HM) and four subclades (MS_EH1, MS_EH2, MS_HM1, and MS_HM2; Fig. 2i and j). The relationship was further confirmed by PCA and phylogenetic inference (Fig. 2k and Additional file 1: Fig. S2).

The results of genetic diversity within species of Megacodon showed that the nucleotide diversity of M. stylophorus (π = 0.101) was higher than that of M. lushuiensis (π = 0.084) and M. venosus (π = 0.051), while the observed heterozygosity of M. stylophorus (Hobs = 0.062) was lower than other two species based on dataset M155 (Additional file 2: Table S6). Among the M. stylophorus, all populations have similar nucleotide diversity and observed heterozygosity (Additional file 2: Table S7). The proportions of private alleles in M. venosus, M. lushuiensis, and MS_EH2 subclades were greater than 10% (Additional file 2: Tables S6 and S7).

Historical range reconstruction and historical gene flow

The biogeographic model selection conducted through BioGeoBears using the BEAST tree and Anemonopsis from Japan as the outgroup, identified DIVALIKE + J as the most supported model, accounting for 58.17% of the relative weight (Additional file 2: Table S16). Biogeographical reconstructions showed that the common ancestor of B. calthifolia and B. deltaphylla was inferred to have established in the low elevations in Southwest China and the east Himalayan region (> 75%; Fig. 2g). Vicariance events likely played a significant role in the subsequent diversification and formation of each group of Beesia (Fig. 2g). High-level gene flows (Nm > 5) were found to only between BC_LS group and two widely distributed groups (i.e., BC_HM and BS_SC groups) as per the results of the MIGRATE-N analysis (Additional file 2: Table S12).

In the biogeographic analysis of Megacodon using the BEAST tree (excluding outgroups), the DIVALIKE + J model emerged as the predominant model with 41.98% relative weight (see Additional file 2: Table S17). The data showed that vicariance events from the East Himalayas to central China significantly influenced the diversification of Megacodon groups (Fig. 2m). The MIGRATE-N analyses showed that no significant historical gene flow among the three Megacodon species (Additional file 2: Table S13). Additionally, for M. stylophorus, high-level historical gene flows were only identified between MS_EH2 and other subgroups (Additional file 2: Table S14).

Correlations between spatial genetic structure and geographic, elevation, and climatic factors

Pairwise FST values between populations of Beesia ranged from 0.13 to 0.88 (Additional file 1: Fig. S3). The population BC1 showed the highest genetic differentiation from other populations of B. calthifolia, while BC6 had the second highest value (Additional file 1: Fig. S3). According to a Mantel test based on dataset B119, genetic differentiation showed significant correlation with geographic distance (isolation-by-distance, IBD, r = 0.403, P-value = 0.004) and elevational distance (isolation-by-elevation, IBE, r = 0.238, P-value = 0.037), but not climatic distance (isolation-by-climate, IBC, r = 0.108, P-value = 0.236, Fig. 3a-c, Additional file 1: Fig. S5). Partial Mantel tests further supported this result (Additional file 2: Table S8). After excluding B. deltophylla and the highly diverged population BC1, the genetic distances within BC_HM, BC_LS, and BC_SC groups increased significantly with elevational distance (r = 0.547, P-value = 0.001) and climatic distance (r = 0.393, P-value = 0.006 Fig. 3d-f, and Additional file 2: Table S10). However, climatic distance is not an independent variable but exhibits a partial matrix correlation with elevational distance (r = 0.213, P-value = 0.121, Additional file 2: Table S8).

Fig. 3
figure 3

Correlations between pairwise genetic differentiation (pairwise FST) based on B119 or M131 dataset and geographic (isolation-by-distance; a, d, and g), elevational (isolation-by-elevation, b, e, h), and climatic distance (isolation-by-climate, c, f, i). Numbers in the right lower corner are Mantel’s r and significance (P-value). The red font represents significantly strong correlation (Mantel’s r > 0.3 and P-value < 0.01)

Pairwise FST values between populations of M. stylophorus ranged from 0.11 to 0.78 (by dataset M131), indicating a high level of genomic differentiation in this species (Additional file 1: Fig. S4). The highest FST was found between M15 and other populations. There was a significant correlation between pairwise FST and geographic distance for all the populations (r = 0.86, P-value = 3e-05, Fig. 3g and Additional file 2: Table S8), while elevational (r = 0.082, P-value = 0.3, Fig. 3h) and climatic distance (r = 0.145, P-value = 0.1795, Fig. 3i) were not correlated with pairwise FST.

Adaptive and ecological divergence of Beesia calthifolia across elevation in the Hengduan Mountains

Ecological divergence can lead to clinal variation, which in turn can result in genetic differentiation among populations. To further investigate the relationship between clinal variation and genetic differentiation, we first conducted a measurement on 258 specimens of B. calthifolia, which were collected at elevations ranging from 800 to 3900 m. Of these specimens, 79 individuals were collected at elevations higher than 2700 m, which is similar to the BC_HM group (Additional file 2: Table S15 and Additional file 3: Table S20). Our results indicate that leaf traits exhibit significant variability among all individuals, with coefficients of variation (CV) greater than 20% considered high (Additional file 2: Table S15). Among the eight traits studied, three traits (leaf length, leaf shape, and leaf teeth density) were found to differ significantly between high elevation (specimens ≥ 2700 m) and low elevation (specimens < 2700 m) specimens (Additional file 1: Fig. S9 and Additional file 2: Table S15). Additionally, three leaf traits (leaf length, leaf teeth density, and the number of teeth) were found to have a weak correlation with elevation (0.15 < |r| 0.3, P-value < 0.01, Fig. 4). However, the ratio of leaf length to width was found to have a strong negative correlation with elevation (r = -0.53, P-value < 2.2e-16, Fig. 4). Furthermore, elliptic Fourier descriptors and principal component analysis (EF-PCA) of 58 intact leaves revealed that the ratio of leaf length to width could explain 75.31% of the variation in leaf shape (PC1) (Additional file 1: Fig. S6). These findings suggest that leaf shape variation in B. calthifolia is significantly influenced by elevation, and that the ratio of leaf length to width is a key factor in this variation.

Fig. 4
figure 4

Linear regression analysis of correlation between leaf traits for B. calthifolia and elevation based on 258 specimens. (a): leaf length (cm), (b): number of teeth, (c): leaf width (cm), (d): perimeter (cm), (e): leaf shape (represented by ratio of leaf length to leaf width), (f): tooth density (calculated by number of teeth divided by perimeter), (g): leaf area (cm2), (h): petiole length (cm). The numbers in the left upper corner are correlation coefficient (r), with its significance (P-value), between each trait and elevation. The red font represents significantly strong correlation (|r| > 0.3 and P-value < 0.01)

The aim of the ecological niche modeling (ENM) analysis is to understand whether B. calthifolia from two elevational ranges will respond differently to past climate change. The niche identity test was used to detect the niche divergence between B. calthifolia from two elevational ranges. All ENM models exhibit a strong predictive ability with area under the curve (AUC) exceeding 0.95. The predicted distribution range under current conditions matched the actual distributions of B. calthifolia occurring at both elevational ranges (Additional file 1: Fig. S10). The ENM results showed that the geographic ranges of both high- and low-elevation populations shrank during the LIG compared to their current ranges (Additional file 1: Fig. S10). The predicted distribution of high-elevation populations during the LGM were southward expansion (Additional file 1: Fig. S10). In contrast, the predicted distribution of low-elevation populations has tended to migrate to Sichuan Basin during the LGM (Additional file 1: Fig. S10). Discrepancies between the expected and observed values of Schoener’s D and Warren’s I suggest significant niche divergence between populations from the two elevational ranges (p < 0.01; Additional file 1: Fig. S11).


Allopatric divergence in Beesia and Megacodon

Through reconstructing phylogenetic relationships and population structure in Beesia, we revealed an unexpectedly complex process of allopatric divergence in this genus. The genus most closely related to Beesia is Anemonopsis, which is endemic to Japan. However, the genetic divergence between B. calthifolia and B. deltophylla occurred at the western end of the total geographic range of Beesia around the late Miocene (Figs. 1b and 2g). Subsequently, the BC1 (Dujiangyan) population of B. calthifolia on the opposite side of the range diverged from other conspecific populations around the middle of the Pliocene (Fig. 2g). The result of BioGEOBears showed that all these divergences are the result of vicariance events, suggesting there may have been a widespread ancestral species for Beesia that existed at relatively lower elevations from the eastern Himalayas to central China (Fig. 2g). The Dujiangyan population was found to retain shared ancestral polymorphisms with B. deltophylla, as evidenced by K = 2 and 3 in our STRUCTURE analyses (Fig. 2a and b). A very small historical gene flow (Nm) was detected between the Dujiangyan population and B. deltophylla and these two groups are isolated geographically, so the genetic similarity of these two groups may result from retention of ancestral polymorphisms of this widespread ancestral species at relatively lower elevations (Additional file 2: Table S12). In addition, population BC_LS (= 2640 m) was found to retain a greater proportion of shared ancestral polymorphisms with the group BC_SC (< 2200 m) than its adjacent populations in group BC_HM (> 2700 m) (Fig. 2b and c). The genetic similarity of these geographically isolated lineages may have resulted from ancient fragmentation of the habitat of a widespread Beesia ancestor at relatively lower elevations in the Hengduan Mountains.

A very similar distribution pattern was also found in allopatric divergence of two Megacodon species at lower elevations, M. venosus, and M. lushuiensis, formed in two isolated areas [30], but these two species still grow in similar habitat. These relict lineages isolated in different lowland regions by retaining the niche of their most recent common ancestor. These ancient allopatric lineages contain very large numbers of private alleles (Additional file 2: Table S5) due to limited gene flow. Therefore, genetic drift may have played an important role in their divergence processes.

Allopatric divergence also played an important role in speciation and/or divergence history of Megacodon stylophorus (Fig. 2m). This species diverged from the other two lowland Megacodon species in the late Miocene (6.28 Ma; Figs. 1d and 2h-m). Although M. stylophorus occupies a much higher elevation, this species has a distribution largely allopatric to those of the other two species and very limited historical gene flow was found between them, so the initial separation of this species may have resulted from allopatric divergence. After the initial divergence, significant intra-specific divergence in M. stylophorus was found between the Hengduan Mountains and the East Himalayas (Fig. 2h-m).

Because the same habitat type is highly fragmentized in the whole Sino-Himalayan region, relict lineages or species could be preserved in different isolated regions. We found that the allopatric divergence within these two genera occurred mainly within the same elevation range and among three separate sites within the Sino-Himalayan region: the mountain areas around the Sichuan Basin, the northwestern Yunnan (especially Gaoligong Mountain and its nearby areas), and the East Himalayas. These areas may all have stable humid climates in relatively low elevations in the last few million years because of the modern Asian monsoon system and their special geographical locations. These three regions are isolated from each other by peripheral high mountains, hot dry valleys, or regions that experience seasonal drought at lower elevations, which could have accelerated the allopatric divergence of many temperate taxa that favor humid climates [38, 39].

The historical fragmentation of humid region and the drastic geological changes (e.g., the landform formation of alpine valleys) in intermediate region may give rise to a phenomenon that early-diverged species (lineages) or sister species (lineages) were persistently isolated at opposite ends of the distribution area. The allopatric divergence in Beesia occurred successively at opposite ends of the distribution areas. The BC_HM group in the central distribution area is a young evolutionary lineage in Beesia and it occupies the highest elevations (Fig. 1b and 2g). Megacodon stylophorus also occurs in area between two Megacodon species at lower elevations and occupies the highest elevations in the genus. This pattern can also be found in the allopatric divergence of another East Asian endemic genus, Rodgersia [40], and an important tree group which favor humid habitat, Tsuga [41]. In these genera, the species (lineages) in the central region (e.g., several mountain chains in the Hengduan Mountains and the Yunnan Plateau) underwent adaptive changes because of the drastic geological and climatic changes [15], which may have created young species or lineages.

The drastic geological changes and vegetation evolution in the central region can also cause local extinction. Some temperate plant species or groups that favor humid habitats, such as Davidia involucrata, Dickinsia hydrocotyloides, Dicentra macrantha, and Asteropyrum peltatum [42], all have the same fragmented disjunction on Gaoligong Mountain or in nearby areas. However, the main distribution ranges of these species are all humid regions around the Sichuan Basin. This disjunction pattern in these species may be related to widespread areas of unsuitable habitat (e.g., pinewood) and seasonal drought in central Yunnan. The historical vegetation in central and west Yunnan may be closely related to the strength of the Asian monsoon [43]. When the Asian monsoon was weaker in late Pliocene, the humid habitats may become more fragmentized, resulting in local extinction or adaptive change of temperate taxa that favor humid climates.

Another conspicuous disjunction pattern within Beesia and Megacodon was found between the East Himalayas and the Hengduan mountains (between two Beesia species or within the species M. stylophorus). Many other temperate taxa, such as Maddenia, Rodgersia [40], and Roscoea [19], showed similar regional disjunctions or fragmented distributions. Interestingly, even some drought-tolerant species can also show a fragmented distribution between the East Himalayas and the Hengduan mountains [44]. We believe that an important biogeographic boundary once existed between the Nujiang and the Yarlung Zangbo River. This boundary may have been closely related to geological events in this region, such as the historical river capture of the Yarlung Tsangpo river [45, 46]. In addition, an ecological barrier between the Himalayas and the Hengduan Mountains was proposed in the study of Roscoea [19].

Ecological divergence in Beesia and Megacodon

While ecological adaptation can play a role in the divergence of species over time, it may not be the primary cause of initial divergence processes. Instead, it may be a consequence that emerges after prolonged periods of allopatric divergence. But ecological divergence can reduce the effect of subsequent gene flow, whether under the circumstance of sympatry or secondary contact [47]. Although we have no evidence about ecological processes trigger incipient speciation or divergence directly in Megacodon and Beesia, ecological divergence indeed promotes subsequent lineage or species diversification in these two genera.

In our study, M. stylophorus and two lowland species occupy two distinctly different ecological niches and have fairly different morphological characters [30]. Although they are not sympatrically distributed, M. stylophorus and M. lushuiensis can co-occur in the Nushan Mountains. Habitat differentiation may be an important factor in the persistent divergence between M. stylophorus and the species at lower elevations. Our previous study found that many genes with functions in response to water deprivation and other external stimuli were under positive selection between M. stylophorus and M. lushuiensis [30]. According to the molecular dating analyses, the initial divergence of Megacodon took place in the late Miocene (Fig. 1d). This time coincides with a period when the East Asian monsoon was intense [48, 49] and there was heightened tectonic activity in the Hengduan Mountains [50,51,52,53,54,55,56]. This tectonic activity and the emergence of alpine valleys in the Hengduan Mountains made many lowland areas turn into hot dry valleys that were not suitable for the survival of Megacodon [57, 58]. At the same time, because monsoon intensification brings abundant moisture at high elevations, many species were able to colonize the high-elevation environment through adaptive evolution [15].

Habitat differentiation should play an important role in the persistent divergence of the BC_QCS group and the other groups of Beesia in the Hengduan mountains. The BC_QCS group and the populations in the group BC_SC can co-occur in the Qionglai mountains and can even be found in a very short distance, but they occupy different elevations and occur in different vegetation types. As we mentioned above, the BC_QCS population grows in relatively lower elevation and mainly grow in the understory of evergreen broad-leaf forest, which is similar to those of B. deltophylla. The other groups in the Hengduan mountians grow in the understory of deciduous or coniferous forests. Except for the BC_QCS group, the divergence of other three B. calthifolia groups was impacted by ecological divergence in altitudinal gradient, although Beesia showed an isolation-by-distance pattern across the whole distribution range (Fig. 4). We identified four groups in Beesia calthifolia based on STRUCTURE and PCA analysis (Fig. 2a-e). Of these, BC_HM, BC_LS, and BC_SC are three independent lineages based on all the datasets (Figs. 1 and 2 and Additional file 1: Fig. S1). The results of the Mantel test and partial Mantel test between genetic distance and climatic, geographic, or elevational distance showed that isolation-by-elevation was the main cause of genetic divergence among BC_HM, BC_LS, and BC_SC groups (Fig. 3e). We also found significant niche divergence between high and low elevation populations, as evidenced by niche identity test (Additional file 1: Fig. S11).

Changes in biotic and abiotic factors along altitudinal gradients can result in strong selective pressures and thus lead to clinal variation in morphological characteristics [59,60,61,62]. Leaves, being a key component of plant architecture and a medium for light capture, gas exchange, and thermoregulation [63], are particularly susceptible to such changes. Studies have shown that leaf shape and chloroplast evolution can contribute to the adaptation of plants to high elevations [63,64,65]. Our measurements of specimens revealed that leaf shape in different genetic lineages is strongly correlated with altitudinal gradients (Fig. 4e). Specifically, we found that groups at different elevations have distinct leaf shapes (Fig. 4e and Additional file 1: Fig. S6). Leaf traits, being classic functional characters that reflect changes along elevation or latitude gradients, are often directly linked to the local climatic environment [62, 66]. This relationship, as well as the underlying mechanisms, have been extensively studied in molecular biology and large-scale ecology research [67,68,69]. Therefore, the gradual changes in leaf shape across elevation gradients may be indicative of adaptation to varying environmental conditions. Additionally, as leaf shape is a plastic trait, we undertook a preliminary common garden transplant at the Kunming Institute of Botany to observe changes in leaf shape in the BC_HM, BC_LS, and BC_SC groups. We observed that the edges of individual leaves in the BC_HM group became concave when transplanted, while the leaves of the BC_SC and BC_LS groups grew normally. This suggests that leaf trait divergence in B. calthifolia is likely to be genetically based. However, for a more definitive conclusion, rigorous common garden or transplant experiments still need to corroborate this initial finding in the future.

According to molecular dating, the BC_HM group diverged from the BC_LS group in the early Pleistocene (Fig. 2). The Pleistocene era was marked by numerous glacial, which prompted the vertical migration of plants within mountainous regions [8, 58]. As a result, isolated populations of B. calthifolia may experience varying selection pressures during these glacial cycles. Meanwhile, the very different peripheral environments of the B. calthifolia clade at different elevations also led to a different response to the rapid climatic changes. From the result of ecological niche modelling, we found an apparent southward expansion of the distribution of the high-elevation clade during the LGM, while low-elevation populations tended to migrate to lower elevations (into the Sichuan Basin). This continuous isolation during glacial cycles may have further strengthened the divergence between high and low-elevation clade (Additional file 1: Fig. S10 and Fig. S11).

The IBD + IBE synthesis effect in the Sino-Himalayan region

We found that the groups at higher elevations are largely allopatric with groups at relatively lower elevations in both Beesia and Megacodon. The loss of suitable habitat at the lower elevations may have forced these populations to migrate upward. Some populations probably colonized moist regions at high elevations during the interglacial period or the stage of the intensive Asian monsoon, but in areas where lower elevations are still suitable, relict populations may be preserved at lower elevations. A simultaneous allopatric and altitudinal isolation may have promoted speciation or intra-specific divergence of temperate plant taxa in the Sino-Himalayan region (Fig. 5), particularly those plant taxa with very specific habitat requirements. We believe that the complex divergence pattern observed in Beesia and Megacodon is applicable in many other temperate plant taxa in Sino-Himalayan region. However, the divergence process and geographical range evolution of a plant group are shaped by many factors. Different plant groups may have different life histories, habitat preference, demographics, and experience different tectonic and climatic events [70]. Even Beesia and Megacodon also have markedly different in their divergence processes, such as an earlier and sudden ecological divergence in Megacodon. Sometimes, the divergence events in some taxa may dominate by allopatric divergence because of strong niche conservatism, such as plant groups in extreme habitats. Furthermore, some taxa may involve more complex evolutionary histories, such as radiative diversification because of extensive hybridization [71], polyploidization or other unrevealed processes. Our study emphasizes the fact that habitat fragmentation and ecological adaptation at different elevations can work together in driving the divergence of plant groups in mountain regions. Plant groups in such regions are more likely to be isolated by environmental heterogeneity and physical barriers, resulting in habitat fragmentation and allopatric divergence (IBD). Different isolated populations may experience different selection pressures and also have different ecological opportunities, resulting in them colonizing new ecological niches (IBE). Sometimes, habitat fragmentation and ecological divergence interact. The effect of one process may be magnified by another. Different populations or groups in mountain regions can establish reproductive isolation quickly because of strong genetic drift and divergent selection. This IBD + IBE synthesis effect may be widespread as a factor leading to the divergence of plant groups in the Sino-Himalayan region.

Fig. 5
figure 5

Schematic view of external factors influencing ecological divergence in Sino-Himalayan regions. The diagram at the top shows initial allopatric divergence. The two diagrams on the left show the vertical movement of plants caused by climate dynamics in the Pleistocene. The two on the right show tectonic activities and geological events including uplift, erosion, and historical river capture which can form new and isolated habitats leading to ecological divergence. Different color layers indicate elevation-related vegetation zones. Blue regions indicate ice sheets. Icons of small flower represent populations. Different colored icons represent different species


After conducting this comparative study between Beesia and Megacodon, it became evident that plant groups in mountain regions can diverge due to a combination of habitat fragmentation and ecological adaptation at different elevations. The presence of environmental heterogeneity and physical barriers in such regions can lead to isolation, resulting in habitat fragmentation and allopatric divergence. Furthermore, isolated populations can experience diverse selection pressures and opportunities, leading to the colonization of new ecological niches. These processes can interact and magnify each other, ultimately resulting in the establishment of reproductive isolation through strong genetic drift and divergent selection. Although the relative contributions of geographical isolation and parapatric ecological divergence vary among different plant taxa, a combination of these two factors is common in the Sino-Himalayan region.


Fieldwork and sampling

In order to study the population genetics of Beesia and Megacodon, a total of 119 samples were collected from 12 sites of Beesia calthifolia and three sites of B. deltophylla, as well as 10 individuals from one site of M. venosus, 14 individuals from one site of M. lushuiensis, and 131 individuals from 13 sites of M. stylophorus (Additional file 2: Table S1 and Table S2). The samples were collected to ensure that the distance between any two individuals sampled from each site was greater than 10 m. Leaf samples were collected and dried in silica gel for DNA extraction, and vouchers of each site were deposited at Herbarium of the Kunming Institute of Botany, Chinese Academy of Sciences (KUN). All samples were used for the population genetic analysis by ddRAD-seq (double digest restriction site associated DNA sequencing).

Molecular dating based on molecular markers and chloroplast genomes

Since there is no prior information on the mutation rate for both genera before this study, we selected 16 individuals of Beesia and 16 individuals of Megacodon and combined previous published data for molecular dating analysis. To estimate divergence times among different populations or species, we used a combination of fossil records and secondary calibration. For Megacodon, the dataset used for molecular dating was obtained from [29] and [30], which used the nuclear ribosomal internal transcribed spacer (nrITS) and two chloroplast markers, atpB-rbcL, and trnL-trnF.

For Beesia, thirteen individuals from ten populations of B. calthifolia and three individuals from three populations of B. deltophylla were used to confirm phylogenetic relationships. We assembled chloroplast genomes for these Beesia samples and extracted protein-coding gene sequences (CDSs) for phylogenetic analysis and molecular dating, following the methods outlined in [35]. Paired-end 150-bp libraries were constructed and sequencing for Beesia was carried out on an Illumina HiSeq 2000 platform, and approximately 3 Gb data for each sample was generated. After filtering out low-quality reads, the GetOrganelle v1.7.1 pipeline [72] was used to assemble the whole chloroplast genome sequence. In this pipeline, plastid genome reads were extracted from total sequencing data and assembled with SPAdes version 3.10 [73]. The kmer values were set as 127, 115, 105, and 85. The sequences were further annotated in the PGA software package [74] using a published whole chloroplast genome from Beesia calthifolia (MK253467) [35] as reference and then checked manually in Geneious 8.2.4 [75]. We also downloaded chloroplast sequences of the Cimicifugeae from [35] as outgroups (Additional file 2: Table S9). Geneious 8.2.4 was then used for extraction of chloroplast CDSs. Alignment of matrix was performed by using MAFFT v 7.308 [76].

BEAST v1.84 [77] was used for molecular dating analysis. The secondary calibration setting for Beesia followed the results of [35], constraining the crown age of the Cimicifugeae (26.57 Ma) and the stem age of Beesia and Anemonopsis (6.4 Ma). The best-fit model of nucleotide substitution was GTR + I + R, which was estimated using jMODELTEST v 2.1.7 [78]. For Megacodon, the fossil calibration settings followed Favre et al. (2016) [29], constraining the stem age of Lisianthius (35.21 Ma) and the crown age of the Gentiana sect. Cruciata (6.25 Ma). The best-fit model of nucleotide substitution in BEAST analysis of Megacodon was the same as that given by [30]. A birth-death incomplete sampling tree prior was used, and a burn-in of the initial 10% cycles was carried out in BEAST. The analysis was run for 108 Markov Chain Monte Carlo (MCMC) steps. MCMC samples were imported into TRACER v1.5 (available from to inspect the sampling adequacy and convergence of the chains to a stationary distribution. The results were visualized and plotted in FIGTREE v1.3.1 (available from

Population genetics analysis based on RAD-seq

Genomic DNA was extracted using a Plant Genomic DNA kit (Tiangen Biotech, Beijing, China) following the kit protocol. Restriction site-associated DNA library preparation followed a universal and simplified ddRAD library method for angiosperm plants named MiddRAD [79]. Total genomic DNA (20 ng) was digested by MspI (New England Biolabs, NEB, #R0106L) and AvaII (New England Biolabs, NEB, #R0153L) at 37 ℃ for 3 h in a 20 µl reaction volume buffered with 1x CutSmart buffer (New England Biolabs, NEB, #B7204S). A fragment library with inserts of 400–600 bp was constructed strictly according to the protocol for MiddRAD. The Illumina HiSeq X Ten (San Diego, CA, USA) sequencing platform (PE150) was used for sequencing at Novogene Co. (Beijing, China).

Filtered clean reads were processed by the Stacks 2.0 Beta 8 software pipeline [80,81,82]. First, reads were demultiplexed by P1 barcode and low-quality reads were removed (-c, -q, -r). Then Fastp [83] was used to trim the low-quality bases. The first reads were trimmed by 11 bp at the 3’ end and 10 bp at the 5’ end, and the second reads were trimmed by 20 bp at the 3’ end and 10 bp at the 5’ end. Then these trimmed reads were assembled using The parameter m represents the minimum number of reads to create a stack; the M represents the maximum number distance allowed between stacks; the n represents the maximum number of mismatches allowed between loci. The various parameter combinations by systematically varying M and n from 1 to 6 (M = n) and setting m at a constant value of 3 were explored. The values maximized the number of polymorphic loci found in 80% of the individuals were used (parameter settings: m = 3, M = 4, n = 4 for Megacodon; m = 3, M = 5, n = 5 for Beesia). The datasets for downstream population genetic analyses were generated by the POPULATIONS module and PLINK v.1.07 [84]. Each RAD locus presents in at least 50% of the individuals (-r 0.5) of each population and that appeared each population (-p 15 for dataset B119, -p 12 for dataset M155, -p 10 for dataset M131) was retained, and for each locus, unlinked SNP were retained (--write_single_snp). Finally, we used PLINK to filter out loci with minor allele frequency < 0.01and missing genotypes > 0.3 (--maf 0.01 –geno 0.3). Input files were converted by the POPULATIONS program implemented in Stacks and PGDSpider [85] for further analysis.

We used STRUCTURE 2.3.4 [86] to detect the underlying population structure among samples. Ten independent runs of each K-value in the range from 1 to 10 were performed with 5 x 105 burn-in lengths and 106 iterations. STRUCTURE HARVESTER [87] was performed to choose the optimal K-values by the delta-K method. We averaged the coefficients over the ten independent runs by CLUMPP version 1.1 [88] and plotted the output with DISTRUCT [89]. Principal component analysis (PCA) was performed by the Bioconductor package SNPRELATE [90]. Phylogenies were reconstructed using a maximum likelihood approach in RAxML 8.2.12 [91] with 1000 bootstrap replicates. The substitution model of GTR + GAMMA model with acquisition bias correction (--asc-corr = lewis) were assigned. Pairwise FST and genetic diversity (including private sites; observed heterozygosity, Hobs; nucleotide diversity, π; FIS) were estimated by the POPULATIONS program in Stacks.

We used the MIGRATE-N 3.6.11 software to estimate the historical gene flow (Nm) between populations and species [92]. This software utilizes a Bayesian inference strategy to infer the migration rates and divergence times of populations based on genetic data. For our analysis, we used three SNP datasets and set the following parameters: 500,000 visited parameter values (recorded steps = 5000, increment = 100, number of concurrent chains = 1), with a burn-in period of 10,000.

Biogeography analysis

Ancestral range reconstructions for Beesia and Megacodon were conducted using BioGeoBEARS [93]. We employed three analyses: dispersal-extinction-cladogenesis (DEC), a likelihood version of BayArea (BAYAREALIKE), and a likelihood version of dispersal-vicariance analysis (DIVALIKE). Each analysis was performed both with and without the J-parameter, utilizing the time tree obtained from the aforementioned BEAST analysis. For Megacodon, outgroups were excluded due to the uncertainty of the Megacodon’s outgroup. Anemonopsis was retained because this genus is most closely related to Beesia. We defined four biogeographic regions based on the current distribution of Megacodon: (A) East Himalayas; (B) Lushui in Nushan; (C) Hengduan Mountains; and (D) Central China. In addition to these four regions, two more biogeographic areas were defined for Beesia and Anemonopsis: (E) Dujiangyan near the Sichuan Basin; and (F) Japan. The best-fit biogeographical model was determined using the AIC criterion.

Isolation by distance (IBD), isolation by elevation (IBE), and isolation by climatic distance (IBC) analyses

To investigate the drivers of population genetic differentiation for Beesia and Megacodon, we conducted a Mantel test between genetic distance (represented by pairwise FST) and climatic, elevational, or geographic distance using the function mantel of vegan [94] in R [95]. We also conducted a partial Mantel test, controlling these distance matrixes using the Pearson coefficient. The permutations were all set as 99,999. The geographic distance was generated by Geographic Distance Matrix Generator v. 1.2.3 ( based on the GPS reading at each sampling site. The climatic distance was calculated based on the Euclidean distance of the first two PCs from PCA analysis of 19 bioclimatic factors, which were extracted from the WorldClim database ( by the raster package in R. The elevational distance was represented by the absolute value of elevation difference between every pair of populations.

Ecological niche modelling and niche identity test of Beesia calthifolia

We classified our collection and the herbarium records of B. calthifolia into two groups based on an elevation threshold. The group has a high elevation distribution (above 2700 m) correspond to the BC_HM group in our study, while the group distributed at relatively lower elevation correspond to the BC_SC group, the BC_LS group and the BC_QCS group. After eliminating overlapping sites, we identified 16 habitat locations in the high-elevation group and 26 in the low-elevation group.

We estimated the potential distribution ranges of B. calthifolia during the LIG, the LGM, and the present using the bioclimatic variables from WorldClim. To mitigate the effects of collinearity, we excluded bioclimatic variables that had a Pearson’s correlation coefficient of r ≥ 0.85 [96]. Upon exclusion of bioclimatic variables with high Pearson correlation, eight variables were retained for the high-elevation group (i.e., BIO1, BIO2, BIO3, BIO7, BIO12, BIO14, and BIO19; Additional file 2: Table S18) and six for the low-elevation group (i.e., BIO1, BIO2, BIO4, BIO12, BIO13, and BIO15; Additional file 2: Table S19). We employed MaxEnt 3.4.4 [97] for distribution estimation, using default settings and specific parameters: random test (25%), training (75%), regularization multiplier (1), maximum iterations (5000), convergence threshold (0.00001), maximum background points (10,000), and 10 bootstrap replications. Model accuracy was gauged using the AUC of receiver operating characteristic (ROC) curve. Outputs were reclassified for visualization, with high suitability set at > 0.4.

For niche identity test of Beesia calthifolia distribution at different elevations, we used Schoener’s D [98] and Warren’s I [99] indices, which range from 0 (no overlap) to 1 (equal suitability). Using ENMTools v.1.4.4 [100], a niche identity test with 100 pseudo-replicates provided expected index values. We applied a nonparametric Monte Carlo test to compare observed and expected indices, with p < 0.01 indicating non-identical environmental niches between regions.

Morphometrics of Beesia calthifolia

We downloaded specimen images of Beesia with elevation information from CVH ( and measured phenotypes using ImageJ. We then measured the 259 largest complete leaves (leaf width, leaf length, petiole length, leaf area, and leaf perimeter) from each specimen image. Statistical analysis was carried out using R version 3.5.2 [95].

In order to examine the effects of elevation on the principal component scores of leaf shape among the varieties, EF-PCA was performed by SHAPE software [101] which includes ChainCoder, Chc2Nef, PrinComp, and PrinPrint programs. Fifty-eight intact leaf images of specimens from different elevations were selected to convert into bitmap format using Windows Paint. Then we followed the SHAPE manual. We selected normalization based on the longest radius for normalizing the shape manually in Chc2Nef. The leaf shape was approximated by PrinComp and PrinPrint from the first 20 harmonics of the coefficients. The PCA was implemented in R [95].

Data Availability

All genomic data are available from the National Genomics Data Center ( under the accession number PRJCA007438. For any inquiries regarding the data, please contact Junchu Peng at



Double digest restriction site associated DNA sequencing


Principal component analysis


Nucleotide diversity

F IS :

Inbreeding coefcient


Nuclear ribosomal internal transcribed spacer


Protein-coding gene sequence


Markov Chain Monte Carlo

F ST :

Genetic differentiation

H obs :

Observed heterozygosity

N m :

Historical gene flow








Last Glacial Maximum


Last Interglacial


Coefficients of variation


Elliptic Fourier descriptors and principal component analysis


Whole chloroplast genome


  1. Harvey PH, Pagel MD. The comparative method in evolutionary biology. Oxford: Oxford university press; 1991.

    Book  Google Scholar 

  2. Wiens JJ, Graham CH. Niche Conservatism: integrating evolution, ecology, and conservation biology. Ann Rev Ecol Evol S. 2005;36:519–39.

    Article  Google Scholar 

  3. Pyron RA, Burbrink FT. Lineage diversification in a widespread species: roles for niche divergence and Conservatism in the Common Kingsnake, Lampropeltis getula. Mol Ecol. 2009;18:3443–57.

    Article  CAS  Google Scholar 

  4. Funk DJ, Nosil P, Etges WJ. Ecological divergence exhibits consistently positive associations with reproductive isolation across disparate taxa. Proc Natl Acad Sci U S A. 2006;103(9):3209–13.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Badgley C, Smiley TM, Terry R, Davis EB, DeSantis LR, Fox DL, et al. Biodiversity and topographic complexity: modern and geohistorical perspectives. Trends Ecol Evol. 2017;32(3):211–26.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Antonelli A, Kissling WD, Flantua SG, Bermúdez MA, Mulch A, Muellner-Riehl AN, et al. Geological and climatic influences on mountain biodiversity. Nat Geosci. 2018;11(10):718–25.

    Article  CAS  Google Scholar 

  7. Comes HP, Kadereit JW. The effect of quaternary climatic changes on plant distribution and evolution. Trends Plant Sci. 1998;3(11):432–38.

    Article  Google Scholar 

  8. Hazzi NA, Moreno JS, Ortiz-Movliav C, Palacio RD. Biogeographic regions and events of isolation and diversification of the endemic biota of the tropical Andes. Proc Natl Acad Sci U S A. 2018;115(31):7985–90.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. McCormack JE, Zellmer AJ, Knowles LL. Does niche divergence accompany allopatric divergence in Aphelocoma jays as predicted under ecological speciation? Insights from tests with niche models. Evolution. 2010;64(5):1231–44.

    PubMed  Google Scholar 

  10. Nunes LA, Raxworthy CJ, Pearson RG. Evidence for ecological processes driving speciation among endemic lizards of Madagascar. Evolution. 2022;76(1):58–69.

    Article  PubMed  Google Scholar 

  11. Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J. Biodiversity hotspots for conservation priorities. Nature. 2000;403(6772).

  12. Sun H, Zhang J, Deng T, Boufford DE. Origins and evolution of plant diversity in the Hengduan Mountains, China. Plant Divers. 2017;39(4):161–6.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Sun W, Ma X, Zhang J, Su F, Zhang Y, Li Z. Karyotypes of nineteen species of Asteraceae in the Hengduan Mountains and adjacent regions. Plant Divers. 2017;39(4):194–201.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Chen Y-S, Deng T, Zhou Z, Sun H. Is the east Asian flora ancient or not? Natl Sci Rev. 2018;5(6):920–32.

    Article  Google Scholar 

  15. Ding W-N, Ree RH, Spicer RA, Xing Y-W. Ancient orogenic and monsoon-driven assembly of the world’s richest temperate alpine flora. Science. 2020;369(6503):578–81.

    Article  CAS  PubMed  Google Scholar 

  16. Wen J, Zhang J-Q, Nie Z-L, Zhong Y, Sun H. Evolutionary diversificatons of plants on the Qinghai-Tibetan Plateau. Front Genet. 2014;5.

  17. Li Y, Zhai S-N, Qiu Y-X, Guo Y-P, Ge X-J, Comes HP. Glacial survival east and west of the ‘Mekong-Salween divide’ in the Himalaya-Hengduan Mountains region as revealed by AFLPs and cpDNA sequence variation in Sinopodophyllum hexandrum (Berberidaceae). Mol Phylogenet Evol. 2011;59(2):412–24.

    Article  PubMed  Google Scholar 

  18. Luo D, Yue J-P, Sun W-G, Xu B, Li Z-M, Comes HP, Sun H. Evolutionary history of the subnival flora of the Himalaya-Hengduan Mountains: first insights from comparative phylogeography of four perennial herbs. J Biogeogr. 2016;43(1):31–43.

    Article  Google Scholar 

  19. Li DB, Ou XK, Zhao JL, Li QJ. An ecological barrier between the Himalayas and the Hengduan Mountains maintains the disjunct distribution of Roscoea. J Biogeogr. 2019;47(2):326–41.

    Article  Google Scholar 

  20. Zhang XL, Wang YJ, Ge XJ, Yuan YM, Yang HL, Liu JQ. Molecular phylogeny and biogeography of Gentiana sect. Cruciata (Gentianaceae) based on four chloroplast DNA datasets. Taxon. 2009;58(3):862–70.

    Article  Google Scholar 

  21. Fan DM, Yue JP, Nie ZL, Li ZM, Comes HP, Sun H. Phylogeography of Sophora davidii (Leguminosae) across the ‘Tanaka-Kaiyong line’, an important phytogeographic boundary in Southwest China. Mol Ecol. 2013;22(16):4270–88.

    Article  CAS  PubMed  Google Scholar 

  22. Ma X-G, Zhao C, Wang C-B, Liang Q-L, He X-J. Phylogenetic analyses and chromosome counts reveal multiple cryptic species in Bupleurum commelynoideum (Apiaceae). J Syst Evol. 2015;53(1):104–16.

    Article  Google Scholar 

  23. Li JL, Zhong LL, Wang J, Ma T, Mao KS, Zhang L. Genomic insights into speciation history and local adaptation of an alpine aspen in the Qinghai–Tibet Plateau and adjacent highlands. J Syst Evol. 2020;59(6):1220–31.

    Article  Google Scholar 

  24. Zhao J-L, Gugger PF, Xia Y-M, Li Q-J. Ecological divergence of two closely related Roscoea species associated with late quaternary climate change. J Biogeogr. 2016;43(10):1990–2001.

    Article  Google Scholar 

  25. Liu J, Möller M, Provan J, Gao L-M, Poudel RC, Li D-Z. Geological and ecological factors drive cryptic speciation of yews in a biodiversity hotspot. New Phytol. 2013;199(4):1093–108.

    Article  PubMed  Google Scholar 

  26. Yang L-Q, Hu H-Y, Xie C, Lai S-P, Yang M, He X-J, Zhou S-D. Molecular phylogeny, biogeography and ecological niche modelling of Cardiocrinum (Liliaceae): insights into the evolutionary history of endemic genera distributed across the sino-japanese floristic region. Ann Bot. 2017;119(1):59–72.

    Article  PubMed  Google Scholar 

  27. Wang Z-M, Meng S-Y, Rao G-Y. Two species of the Rhodiola yunnanensis species complex distributed around the Sichuan Basin of China: speciation in a ring? J Syst Evol. 2022;60(5):1092–108.

    Article  Google Scholar 

  28. Favre A, Packert M, Pauls SU, Jahnig SC, Uhl D, Michalak I, Muellner-Riehl AN. The role of the uplift of the Qinghai-Tibetan Plateau for the evolution of tibetan biotas. Biol Rev Camb Philos Soc. 2015;90(1):236–53.

    Article  PubMed  Google Scholar 

  29. Favre A, Michalak I, Chen C-H, Wang J-C, Pringle JS, Matuszak S, et al. Out-of-Tibet: the spatio-temporal evolution of Gentiana (Gentianaceae). J Biogeogr. 2016;43(10):1967–78.

    Article  Google Scholar 

  30. Peng J-C, Ma X-G, Wang Y-H, Sun H. New insights into the evolutionary history of Megacodon: evidence from a newly discovered species. Plant Divers. 2020;42(3):198–208.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Chassot P, Nemomissa S, Yuan YM, Küpfer P. High paraphyly of Swertia L. (Gentianaceae) in the Gentianella-lineage as revealed by nuclear and chloroplast DNA sequence variation. Plant Syst Evol. 2001;229(1):1–21.

    Article  CAS  Google Scholar 

  32. Yang Q-E. Correction of karyotype of diploid Beesia calthifolia and discovery of a tetraploid cytotype. J Syst Evol. 1999;37(1):1–9.

    Google Scholar 

  33. Yang Q-E, Gu Z-J, Sun H. The karyotype of Beesia deltophylla and its systematic significance. J Syst Evol. 1995;33(3):225–9.

    Google Scholar 

  34. Yuan Q, Yang Q-E. Tribal relationships of Beesia, Eranthis and seven other genera of Ranunculaceae: evidence from cytological characters. Bot J Linnean Soc. 2006;150(3):267–89.

    Article  Google Scholar 

  35. Zhai W, Duan X, Zhang R, Guo C, Li L, Xu G, Shan H, Kong H, Ren Y. Chloroplast genomic data provide new and robust insights into the phylogeny and evolution of the Ranunculaceae. Mol Phylogenet Evol. 2019;135:12–21.

    Article  CAS  PubMed  Google Scholar 

  36. Compton JA, Culham A. Phylogeny and circumscription of tribe Actaeeae (Ranunculaceae). Syst Bot. 2002;27(3):502–11.

    Google Scholar 

  37. Ro KE, Han HY, Lee S. Phylogenetic contributions of partial 26S rDNA sequences to the tribe Helleboreae (Ranunculaceae). Korean J Biol Sci. 1999;3(1):9–15.

    Article  CAS  Google Scholar 

  38. Zhang R. Dry and hot valleys of Henduan Mountains. Beijing: Science Press; 1992.

    Google Scholar 

  39. Zheng D, Zhang Q, Wu S. Mountain geoecology and sustainable development of the Tibetan Plateau. Volume 57. Springer Science & Business Media; 2000.

  40. Ma X, Sun W, Zhu W, Sun H. Resolving the phylogenetic relationships and evolutionary history of the east Asian endemic genus Rodgersia (Saxifragaceae) using multilocus data. Perspect Plant Ecol Evol Syst. 2017;25:20–8.

    Article  Google Scholar 

  41. Feng Y-Y, Shen T-T, Shao C-C, Du H, Ran J-H, Wang X-Q. Phylotranscriptomics reveals the complex evolutionary and biogeographic history of the genus Tsuga with an east asian-north American disjunct distribution. Mol Phylogenet Evol. 2021;157:107066.

    Article  PubMed  Google Scholar 

  42. Cheng S, Zeng W, Wang J, Liu L, Liang H, Kou Y et al. Species delimitation of Asteropyrum (Ranunculaceae) based on morphological, molecular, and ecological variation. Front Plant Sci 2021;12.

  43. Su T, Liu Y-SC, Jacques FM, Huang Y-J, Xing Y-W, Zhou Z-K. The intensification of the east Asian winter monsoon contributed to the disappearance of Cedrus (Pinaceae) in southwestern China. Quat Res. 2013;80(2):316–25.

    Article  Google Scholar 

  44. Rana HK, Luo D, Rana SK, Sun H. Geological and climatic factors affect the population genetic connectivity in Mirabilis Himalaica (Nyctaginaceae): insight from phylogeography and dispersal corridors in the Himalaya-Hengduan biodiversity hotspot. Front Plant Sci. 2020. 10.

  45. Sun Y, Lu G, Li J, Dang T, Xue C, Liu J, Yan Z. Multimedia distribution and trophic transfer of PPCPs in the middle and lower reaches of the Yarlung Zangbo River. Environ Pollut. 2021;271:116408.

    Article  CAS  PubMed  Google Scholar 

  46. Yue LL, Chen G, Sun WB, Sun H. Phylogeography of Buddleja crispa (Buddlejaceae) and its correlation with drainage system evolution in southwestern China. Am J Bot. 2012;99(10):1726–35.

    Article  PubMed  Google Scholar 

  47. Antunes B, Velo-Antón G, Buckley D, Pereira RJ, Martínez-Solano I. Physical and ecological isolation contribute to maintain genetic differentiation between Fire salamander subspecies. Heredity. 2021;126(5):776–89.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Wan SM, Li AC, Clift PD, Stut JBW. Development of the east Asian monsoon: mineralogical and sedimentologic records in the northern South China Sea since 20 ma. Palaeogeogr Palaeoclimatol Palaeoecol. 2007;254:561–82.

    Article  Google Scholar 

  49. An ZS, Kutzbach JE, Prell WL, Porter SC. Evolution of Asian monsoons and phased uplift of the Himalayan Tibetan plateau since late miocene times. Nature. 2001;411(6833):62–6.

    Article  CAS  Google Scholar 

  50. Clark MK, House MA, Royden LH, Whipple KX, Burchfiel BC, Zhang X, Tang W. Late cenozoic uplift of southeastern Tibet. Geology. 2005;33(6):525–8.

    Article  Google Scholar 

  51. Wang E, Kirby E, Furlong KP, Van Soest M, Xu G, Shi X, Kamp PJ, Hodges K. Two-phase growth of high topography in eastern Tibet during the Cenozoic. Nat Geosci. 2012;5(9):640–5.

    Article  Google Scholar 

  52. Wang P, Scherler D, Liu-Zeng J, Mey J, Avouac J-P, Zhang Y, Shi D. Tectonic control of Yarlung Tsangpo Gorge revealed by a buried canyon in Southern Tibet. Science. 2014;346(6212):978–81.

    Article  CAS  PubMed  Google Scholar 

  53. Meng K, Wang E, Wang G. Uplift of the Emei Shan, western Sichuan basin: implication for eastward propagation of the tibetan plateau in early Miocene. J Asian Earth Sci. 2016;115:29–39.

    Article  Google Scholar 

  54. Sun B-N, Wu J-Y, Liu Y-SC, Ding S-T, Li X-C, Xie S-P, et al. Reconstructing Neogene vegetation and climates to infer tectonic uplift in western Yunnan, China. Palaeogeogr Palaeoclimatol Palaeoecol. 2011;304:328–36.

    Article  Google Scholar 

  55. Su T, Spicer RA, Li S-H, Xu H, Huang J, Sherlock S, et al. Uplift, climate and biotic changes at the eocene–oligocene transition in south-eastern Tibet. Natl Sci Rev. 2018;6(3):495–504.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Spicer RA, Farnsworth A, Su T. Cenozoic topography, monsoons and biodiversity conservation within the Tibetan Region: an evolving story. Plant Divers. 2020;42(4):229–54.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Fahrig L. Effects of habitat fragmentation on biodiversity. Annu Rev Ecol Evol Syst. 2003;34:487–515.

    Article  Google Scholar 

  58. Rahbek C, Borregaard MK, Antonelli A, Colwell RK, Holt BG, Nogues-Bravo D, et al. Building mountain biodiversity: geological and evolutionary processes. Science. 2019;365(6458):1114–9.

    Article  CAS  PubMed  Google Scholar 

  59. Halbritter AH, Fior S, Keller I, Billeter R, Edwards PJ, Holderegger R, et al. Trait differentiation and adaptation of plants along elevation gradients. J Evol Biol. 2018;31(6):784–800.

    Article  PubMed  Google Scholar 

  60. Gonzalo-Turpin H, Hazard L. Local adaptation occurs along altitudinal gradient despite the existence of gene flow in the alpine plant species Festuca Eskia. J Ecol. 2009;97(4):742–51.

    Article  Google Scholar 

  61. Nosil P. Ecological speciation. Oxford: Oxford University Press; 2012.

    Book  Google Scholar 

  62. Peppe DJ, Royer DL, Cariglino B, Oliver SY, Newman S, Leight E, et al. Sensitivity of leaf size and shape to climate: global patterns and paleoclimatic applications. New Phytol. 2011;190(3):724–39.

    Article  PubMed  Google Scholar 

  63. Chitwood DH, Sinha NR. Evolutionary and environmental forces sculpting leaf development. Curr Biol. 2016;26:R297–306.

    Article  CAS  PubMed  Google Scholar 

  64. Demmig-Adams B, Muller O, Stewart JJ, Cohu CM, Adams WW 3rd. Chloroplast thylakoid structure in evergreen leaves employing strong thermal energy dissipation. J Photochem Photobiol B. 2015;152(Pt B):357–66.

    Article  CAS  PubMed  Google Scholar 

  65. Sun M, Su T, Zhang S-B, Li S-F, Anberree-Lebreton J, Zhou Z-K. Variations in leaf morphological traits of Quercus guyavifolia (Fagaceae) were mainly influenced by water and ultraviolet irradiation at high elevations on the Qinghai-Tibet Plateau, China. Int J Agric Biol. 2016;18:266–73.

    Article  Google Scholar 

  66. Schmerler SB, Clement WL, Beaulieu JM, Chatelet DS, Sack L, Donoghue MJ, et al. Evolution of leaf form correlates with tropical–temperate transitions in Viburnum (Adoxaceae). Proc R Soc B. 2012;279(1744):3905–13.

    Article  PubMed  PubMed Central  Google Scholar 

  67. Campitelli BE, Stinchcombe JR. Natural selection maintains a single-locus leaf shape cline in Ivyleaf morning glory, Ipomoea hederacea. Mol Ecol. 2013;22(3):552–64.

    Article  CAS  PubMed  Google Scholar 

  68. Ichihashi Y, Aguilar-Martinez JA, Farhi M, Chitwood DH, Kumar R, Millon LV, et al. Evolutionary developmental transcriptomics reveals a gene network module regulating interspecific diversity in plant leaf shape. Proc Natl Acad Sci U S A. 2014;111(25):E2616–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. Midolo G, De Frenne P, Hölzel N, Wellstein C. Global patterns of intraspecific leaf trait responses to elevation. Glob Chang Biol. 2019;25(7):2485–98.

    Article  PubMed  Google Scholar 

  70. Mas-Peinado P, García-París M, Jiménez-Ruiz Y, Valdeón A, Recuero E, Martínez-Solano I, et al. Geology-based and ecological processes of divergence between and within species of wingless darkling beetles. J Biogeogr. 2022;49(12):2281–95.

    Article  Google Scholar 

  71. Ma Y, Mao X, Wang J, Zhang L, Jiang Y, Geng Y, et al. Pervasive hybridization during evolutionary radiation of Rhododendron Subgenus hymenanthes in mountains of southwest China. Natl Sci Rev. 2022;9(12):nwac276.

    Article  PubMed  PubMed Central  Google Scholar 

  72. Jin JJ, Yu W-B, Yang JB, Song Y, Depamphilis CW, Yi TS, et al. GetOrganelle: a fast and versatile toolkit for accurate de-novo assembly of organelle genomes. Genome Biol. 2020;21(1):1–31.

    Article  Google Scholar 

  73. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol. 2012;19(5):455–77.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  74. Qu X-J, Moore MJ, Li D-Z, Yi T-S. PGA: a software package for rapid, accurate, and flexible batch annotation of plastomes. Plant Methods. 2019;15(1):50.

    Article  PubMed  PubMed Central  Google Scholar 

  75. Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, et al. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012;28(12):1647–9.

    Article  PubMed  PubMed Central  Google Scholar 

  76. Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol. 2013;30(4):772–80.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  77. Drummond AJ, Rambaut A. BEAST: bayesian evolutionary analysis by sampling trees. BMC Evol Biol. 2007;7(1):214.

    Article  PubMed  PubMed Central  Google Scholar 

  78. Darriba D, Taboada GL, Doallo R, Posada D. jModelTest 2: more models, new heuristics and parallel computing. Nat Methods. 2012;9(8):772.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  79. Yang GQ, Chen YM, Wang JP, Guo C, Zhao L, Wang XY, et al. Development of a universal and simplified ddRAD library preparation approach for SNP discovery and genotyping in angiosperm plants. Plant Methods. 2016;12:39.

    Article  PubMed  PubMed Central  Google Scholar 

  80. Catchen JM, Amores A, Hohenlohe P, Cresko W, Postlethwait JH. Stacks: building and genotyping loci de novo from short-read sequences. G3: genes. Genomes Genet. 2011;1(3):171–82.

    CAS  Google Scholar 

  81. Catchen J, Bassham S, Wilson T, Currey M, O’Brien C, Yeates Q, et al. The population structure and recent colonization history of Oregon threespine stickleback determined using restriction-site associated DNA-sequencing. Mol Ecol. 2013;22(11):2864–83.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  82. Catchen J, Hohenlohe PA, Bassham S, Amores A, Cresko WA. Stacks: an analysis tool set for population genomics. Mol Ecol. 2013;22(11):3124–40.

    Article  PubMed  PubMed Central  Google Scholar 

  83. Chen S, Zhou Y, Chen Y, Gu J. Fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884–90.

    Article  PubMed  PubMed Central  Google Scholar 

  84. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira M, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  85. Lischer HEL, Excoffier L. PGDSpider: an automated data conversion tool for connecting population genetics and genomics programs. Bioinformatics. 2012;28(2):298–9.

    Article  CAS  PubMed  Google Scholar 

  86. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155(2):945–59.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  87. Earl DA, von Holdt BM. Structure harvester: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Cons Genet Resour. 2012;4(2):359–61.

    Article  Google Scholar 

  88. Jakobsson M, Rosenberg NA. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics. 2007;23(14):1801–6.

    Article  CAS  PubMed  Google Scholar 

  89. Rosenberg NA. Distruct: a program for the graphical display of population structure. Mol Ecol Notes. 2004;4(1):137–8.

    Article  Google Scholar 

  90. Zheng X, Levine D, Shen J, Gogarten SM, Laurie C, Weir BS. A high-performance computing toolset for relatedness and principal component analysis of SNP data. Bioinformatics. 2012;28(24):3326–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  91. Stamatakis A. Bi oinformatics. 2014;30(9):1312–3. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies.

  92. Beerli P, Palczewski M. Unified framework to evaluate panmixia and migration direction among multiple sampling locations. Genetics. 2010;185(1):313–26.

    Article  PubMed  PubMed Central  Google Scholar 

  93. Matzke N. BioGeoBEARS: Biogeography with bayesian (and likelihood) evolutionary analysis with R scripts (v1.1.1). Zenodo. 2018.

    Article  Google Scholar 

  94. Dixon P. VEGAN, a package of R functions for community ecology. J Veg Sci. 2003;14:927–30.

    Article  Google Scholar 

  95. Core Team R. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2012.

    Google Scholar 

  96. Dormann CF, Elith J, Bacher S, Buchmann C, Carl G, Carré G, et al. Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography. 2013;36(1):27–46.

    Article  Google Scholar 

  97. Phillips SJ, Dudík M. Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography. 2008;31(2):161–75.

    Article  Google Scholar 

  98. Schoener TW. The Anolis lizards of Bimini: resource partitioning in a Complex Fauna. Ecology. 1968;49(4):704–26.

    Article  Google Scholar 

  99. Warren DL, Glor RE, Turelli M. Envi Environmental niche equivalency versus Conservatism: quantitative approaches to niche evolution. Evolution. 2008;62(11):2868–83.

    Article  PubMed  Google Scholar 

  100. Warren DL, Glor RE, Turelli M. ENMTools: a toolbox for comparative studies of environmental niche models. Ecography. 2010;33(3):607–11.

    Article  Google Scholar 

  101. Iwata H, Ukai Y. SHAPE: a computer program package for quantitative evaluation of biological shapes based on elliptic fourier descriptors. J Hered. 2002;93(5):384–5.

    Article  CAS  PubMed  Google Scholar 

Download references


We appreciate the help of Jianwen Zhang, Zhuo Zhou, Yang Niu, Yi Yang, Lu Sun, Zhe Chen, Hongliang Chen, Lishen Qian, Jipei Yue, Yongzeng Zhang, Bo Xu (Southwest Forestry University), Bo Xu (Chengdu Institute of Biology), Sirong Yi, Wei Zhang, and Daigui Zhang during fieldwork. Thanks to Dong Luo, Hantao Qin, Ticao Zhang, and Yanting Hu for useful advice on experiments and data analysis.


This study was supported by the Second Tibetan Plateau Scientific Expedition and Research (STEP) program (2019QZKK0502), the National Natural Science Foundation of China (31971393), and the Yunnan Provincial Science and Technology Leading Talent Project and Scientist Studio Project.

Author information

Authors and Affiliations



HS and XGM conceived and designed experiments. JCP performed field investigations and analyzed the data. JCP and XGM wrote the manuscript. XGM and HS revised the manuscript.

Corresponding author

Correspondence to Hang Sun.

Ethics declarations

Ethics approval and consent to participate

This study did not involve any human tissue materials or animal tissue materials. It did not require ethical approval. We declare that all plant materials used in this study were collected in accordance with relevant institutional, national, and international guidelines and legislation. Relevant permits, permissions, and licenses were obtained prior to the collection and use of plant materials. All steps in the study were performed in accordance with the relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Supplementary Material 3

Rights and permissions

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 The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Peng, J., Ma, X. & Sun, H. Ancient allopatry and ecological divergence act together to promote plant diversity in mountainous regions: evidence from comparative phylogeography of two genera in the Sino-Himalayan region. BMC Plant Biol 23, 572 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: