Evolutionary history of Chrysosplenium L. (Saxifragaceae): a molecular phylogenetic approach to dispersal and diversification processes

Evolutionary history of Chrysosplenium L. (Saxifragaceae): a molecular phylogenetic approach to dispersal and diversification processes | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Evolutionary history of Chrysosplenium L. (Saxifragaceae): a molecular phylogenetic approach to dispersal and diversification processes Tsubasa Toji, Takayuki Sawai, Yuichi Isaka This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7444744/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The diversification of plant lineages is often governed by intricate interactions among geography, climate, and lineage-specific traits. The genus Chrysosplenium (Saxifragaceae), a herbaceous group with a disjunct distribution across the Northern Hemisphere and parts of South America, serves as an excellent model for exploring the mechanisms underlying plant diversification and historical biogeography. In this study, we reconstructed a comprehensive phylogeny of Chrysosplenium by integrating plastid and nuclear DNA markers. We also estimated divergence times, inferred ancestral geographic regions and distributable climate classes, and quantified colonization rates among ancestral regions and distributable climate classes to better understand the processes shaping the genus’s present-day distribution and regional diversification hotspots. Our analyses suggest that Chrysosplenium originated in the Eocene within cold climate zones of East Asia, from which it subsequently dispersed into Europe, North America, and southern parts of Asia. The genus underwent episodic diversification likely driven by climatic fluctuations that promoted both range expansion and lineage divergence. Ancestral distributable climate classes reconstructions indicate that early species had already adapted to a spectrum of cold and temperate conditions. However, despite the existence of climatically suitable areas beyond their current ranges, most species remain geographically localized. This pattern implies that factors beyond macroclimatic suitability—such as physiological constraints like water-use efficiency and habitat specialization—may restrict range expansion and foster allopatric speciation. Overall, our findings highlight the value of combining phylogenetic, biogeographic, and ecological perspectives to uncover the evolutionary processes shaping plant diversity. ancestral trait reconstruction Asia distributable climate divergence time estimation paleoclimate Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction The genus Chrysosplenium L. (Saxifragaceae), commonly known as golden saxifrages, comprises over 85 species (POWO 2025; WFO 2025). These small perennial herbs are typically found in moist habitats and are predominantly distributed across the cold and temperate regions of the Northern Hemisphere. The genus shows particularly high diversity in Asia, where species have radiated into a broad range of ecological niches (Folk et al. 2021 ; Soltis et al. 2001 ; Yang et al. 2023 ). Chrysosplenium also presents a notable South American–East Asian disjunction, with two species ( C. valdivicum and C. macranthum ) occurring in southern South America (Argentina and Chile), though the biogeographic history of this pattern remains incompletely understood (Hara 1957 ). Asia harbors a major portion of the genus’s diversity, offering a unique setting to investigate diversification processes. More than 30 species are narrowly distributed in continental Asia—especially in China, which hosts over 20 endemic species. In addition, 13 island endemics are found in Japan (11 species) and Taiwan (2 species) (POWO 2025). To date, various studies have explored phylogenetic relationships, biogeographic patterns, and diversification within Chrysosplenium (e.g., Deng et al. 2015 ; Folk et al. 2021 ; Nakazawa et al. 1997 ; Soltis et al. 2001 ; Yang et al. 2023 ). A recent study by Yang et al. ( 2023 ), which analyzed chloroplast genomes and partial nuclear gene sequences from 48 species, confirmed the monophyly of Chrysosplenium and identified three main clades. This study also found that species with opposite phyllotaxis form a monophyletic group, whereas those with alternate phyllotaxis are polyphyletic, confirming earlier morphological classifications (Hara 1957 ; Nakazawa et al. 1997 ). Previous molecular studies, including phylogeographic analyses (Deng et al. 2015 ; Soltis et al. 2001 ), suggest that the genus originated in East Asia. Moreover, metadata analyses indicate that the subtribe Chrysospleninae, which includes Chrysosplenium , is a specialist of wet woodlands, with transitions to arctic habitats occurring throughout its evolutionary history (Folk et al. 2021 ). The diversification of East Asian plant lineages has been linked to the development and intensification of the East Asian climate (Wen et al. 2021 ), highlighting the role of past climate change in shaping plant evolution. Additionally, the formation of the Japanese archipelago likely influenced the diversification and distribution of island endemics (Deng et al. 2015 ). Thus, the climatic and geological history of East Asia may have played a crucial role in driving the evolutionary dynamics of Chrysosplenium . Recently, integrating molecular phylogenetics with paleoclimatic reconstructions has enhanced our understanding of such processes (e.g., Isaka et al. 2025 ). While previous research has made important contributions to the taxonomy and evolutionary biology of Chrysosplenium , further insight can be gained by expanding taxon sampling using open-access molecular data (e.g., GenBank: https://www.ncbi.nlm.nih.gov/genbank/ ). Additionally, recent advances in angiosperm phylogenomics and divergence time calibration (ranging from 154 to 247 million years ago [Mya] at the crown node of angiosperms; Zuntini et al. 2024 ) allow for more precise evolutionary timeline reconstructions. Furthermore, integrating updated global floristic regionalization (Liu et al. 2023 ) and paleoclimatic distribution modeling enables a refined understanding of both ancestral area and climatic niche evolution. In this study, we perform a molecular phylogenetic analysis to reconstruct the evolutionary history of Chrysosplenium . Using sequence data, distribution and climatic information, and paleoclimate reconstructions from open-access resources, we address the following objectives: (1) to infer phylogenetic relationships among Chrysosplenium species; (2) to assess which divergence timeline better explains the genus’s evolutionary history; (3) to elucidate patterns of distributional shifts with respect to climatic variables; and (4) to identify the center of diversification within the genus. Materials and methods Taxon sampling The genus Chrysosplenium includes 86 recognized species (POWO 2025). We retrieved sequence data for 57 Chrysosplenium species and one outgroup species, Peltoboykinia tellimoides , from GenBank ( https://www.ncbi.nlm.nih.gov/genbank/ ). Peltoboykinia , comprising two species, is considered the sister group to Chrysosplenium (Yang et al. 2023 ; Zuntini et al. 2024 ). Details of all sampled taxa and accession numbers are listed in Appendix 1 and Table S1 . Phylogenetic analyses A concatenated dataset comprising four chloroplast DNA regions (t rnL–trnF , rbcL , and matK ) and the nuclear ITS region was assembled. Sequence alignment for each region was performed using MAFFT v.7.490 (Kuraku et al. 2013 ) with default parameters and manually corrected where necessary. PartitionFinder v.2.1 (Lanfear et al. 2016) was used to select the best-fit substitution model for each partition (Table 1 ). Bayesian inference was conducted in MrBayes v.3.2.7a (Ronquist et al. 2012 ), employing partition-specific models under the Bayesian Information Criterion (BIC). The Markov Chain Monte Carlo (MCMC) Metropolis–Hastings algorithm was run for 10 million generations, sampling every 100 generations. Two output log files were evaluated based on the effective sample size (ESS) greater than 200 after removing the 25% burn-in using Tracer v.1.7.2 (Rambaut et al. 2018 ), with the first 25% of samples discarded as burn-in. Clade support was evaluated using Bayesian posterior probabilities (PPs). Table 1 The best-fitted substitution model for each region using Bayesian and maximum likelihood phylogenetic reconstruction. Data set Regions No. of sequences No. of sites Substitution model* chlDNA trnL–trnF 44 1101 GTR + G rbcL 47 1428 GTR + I + G matK 53 1650 GTR + G nrDNA ITS 51 6223 GTR + I + G Combined** 58 10402 GTR + I + G *: Same substitution models were selected for both Bayesian inference and maximum likelihood phylogenetic analyses. **: substitution model for BEAST analysis To validate the topology, maximum likelihood (ML) analysis was performed in RAxML-NG v.1.2.0 (Kozlov et al. 2019 ) using substitution models selected under Akaike Information Criterion (AIC). Branch support was assessed via 1,000 bootstrap replicates. Phylogenetic trees were visualized using FigTree v.1.4.4 (Rambaut 2018 ). Divergence time estimation and temporal-based diversification analyses Divergence times were estimated using BEAST v.2.7.6 (Bouckaert et al. 2014 ) with the optimized relaxed clock model and tree prior based on the Yule model (pure-birth model, Yule 1925). The combined partition scheme from Table 1 was used. Subclade B2, as identified in Bayesian analysis (see Results; Fig. 1 ), was constrained as monophyletic. Two calibration scenarios were applied for the Chrysosplenium – Peltoboykinia splitting: an "older" estimate (tree height = 80.83 Mya) and a "younger" estimate (tree height = 44.87 Mya), both from Zuntini et al. ( 2024 ). MCMC chains ran for 100 and 70 million generations, respectively, with sampling every 1,000 generations. All output files were evaluated based on the ESS (> 200) after the removal of 10% burn-in and were combined using Tracer v.1.7.2 (Rambaut et al. 2018 ). The output tree files from the two independent runs were pooled into a single combined file after removing the initial 10% trees as burn-in using LogCombiner v.2.7.6 (in the BEAST package) and the maximum clade credibility tree with mean node heights was summarized using TreeAnnotator v.2.7.6 (in the BEAST package) before visualization with FigTree v.1.4.4 (Rambaut 2018 ). Phylogenetic support was assessed by Bayesian PPs. To evaluate variation in evolutionary rates within Chrysosplenium , we conducted diversification analyses using RevBayes v.1.2.4 (Höhna et al. 2016 ), based on a BEAST-derived phylogenetic tree calibrated with the younger age prior. We assessed changes in net diversification rates across the evolutionary history of the genus by applying an episodic diversification model. This analysis followed the online tutorial ( https://revbayes.github.io/tutorials/divrate/ebd ), with the sampling fraction set to ρ = 0.663 (57 of 86 known species) and a chain length of 30000 generations, sampling every 200 generations. Visualizations of rate variation were generated using the R package RevGadgets v.1.2.1 (Tribble et al. 2022 ). To identify potential rate shifts along specific lineages, we also estimated branch-specific diversification rates, following the approach outlined in the RevBayes tutorial ( https://revbayes.github.io/tutorials/divrate/branch_specific ) with ρ = 0.663 (57 out of 86 total samples) and a chain length of 30000 generations, sampling every 200 generations. Again, we used RevGadgets v.1.2.1 (Tribble et al. 2022 ) for visualizing shifts in diversification rates within Chrysosplenium .. Ancestral region and ancestral distributable climate reconstruction analyses To reconstruct the historical biogeography of Chrysosplenium based on the inferred phylogeny, we conducted ancestral region reconstruction using the R package BioGeoBEARS v.1.1.3 (Matzke 2013 ) in R v.4.3.2 (R Core Team 2023). BioGeoBEARS allows comparison among multiple probabilistic models for historical range evolution and identifies the best-fitting scenario. Distribution data for each species (Table S1 ) were compiled primarily from the Plants of the World Online (POWO 2025), and biogeographic regions were delimited following Liu et al. ( 2023 ): (1) Asian region (Central Asia, Russia, East Asia, and Norway); (2) Chile-Patagonian region (Chile and southern Argentina); (3) European region (Europe and Turkey); (4) Indian region (India, Sri Lanka, Indo-China Peninsula, and Hainan Island); (5) Malaysian region (Taiwan, Philippines, Malaysia, and New Guinea); (6) North American region (United States and Canada); and (7) Saharo-Arabian region (Western Sahara, Morocco, Algeria, Tunisia, Egypt, Libya, and Saudi Arabia). These regions are illustrated in Fig. 1 . We evaluated six models implemented in BioGeoBEARS: DEC (dispersal-extinctioncladogenesis), DEC + J (including founder-event speciation), DIVALIKE (a likelihood version of dispersal-vicariance), DIVALIKE + J, BAYAREALIKE (a likelihood version of the Bayesian inference of historical biogeography for discrete areas), and BAYAREALIKE + J. Model comparisons were conducted using corrected Akaike Information Criterion with correction (AICc) scores to identify the most plausible scenario. The model with the lowest AICc value was selected as the best fit for explaining the historical biogeographic patterns of Chrysosplenium . To infer the historical climatic preferences of Chrysosplenium species, we performed ancestral state reconstruction of distributable climate classes using BayesTraits v.4 (Pagel et al. 2004 ). Distributable climate data for each species (Table S1 ) were derived from species occurrence records and a global climate classification system (Beck et al. 2023 ). Occurrence data were primarily obtained from the Global Biodiversity Information Facility (GBIF; https://www.gbif.org ), and each species’ climatic distribution was determined by overlaying its distribution range with the climate classification map. We initially assigned each distribution point to the one of 18 Köppen-Geiger climate classes following Beck et al. ( 2023 ): arid-desert-cold (BWk), arid-steppe-cold (BSk), tempelate-dry summer-warm summer (Csb), temperate-dry winter-hot summer (Cwa), temperate-dry winter-warm summer (Cwb), temperate-without dry season-hot summer (Cfa), temperate-without dry season-warm summer (Cfb), temperate-without dry season-cold summer (Cfc), cold-dry summer-warm summer (Dsb), cold-dry summer-cold summer (Dsc), cold-dry winter-hot summer (Dwa), cold-dry winter-warm summer (Dwb), cold-dry winter-cold summer (Dwc), cold-dry winter-very cold summer, (Dwd), cold-without dry season-hot summer (Dfa), cold-without dry season-warm summer (Dfb), cold-without dry season-cold summer (Dfc), and Polar-tundra (ET). These classes were then grouped into nine broader climatic categories for the ancestral distributable climate classes reconstruction: BW (arid-desert), BS (arid-steppe), Cs (temperate-dry summer), Cw (temperate-dry winter), Cf (temperate-without dry season), Ds (cold-dry summer), Dw (cold-dry winter), Df (cold-without dry season), and ET (polar-tundra). Ancestral distributable climate classes reconstruction was conducted using the multistate reverse jump MCMC algorithm with model reduction in BayesTraits. We applied the HyperPrior All command to set a uniform prior between 0 and 10. Each analysis ran for 11 million iterations, with samples taken every 1,000 iterations. The output files from the two independent runs were pooled into a single combined file after removing the initial one million iterations were discarded as burn-in. Convergence and mixing of the chains were assessed by checking that ESS (> 200). Ancestral states at each node were summarized using a 95% credibility threshold. When multiple states together accounted for 95% of the PP, all such states were reported. For example, if the PPs for states A, B, C, and D were 0.60, 0.33, 0.05, and 0.02, at node X respectively, states A, B, and C were included as probable ancestral states for that node. In these analyses, we used a molecular phylogenetic tree reconstructed using MrBayes (see Section 3; Fig. 1 ). Although phylogenetic trees were also inferred using BEAST (see below), we selected the MrBayes tree for downstream analyses because it provided higher resolution of interspecific relationships within Chrysosplenium . Different substitution models were applied in each case: partition-specific models in MrBayes, and a single model for the combined partition in BEAST. To visualize the historical extent of distributable climates, we reconstructed paleoclimate maps. The classification of climate zones followed the Köppen-Geiger system as described by Peel et al. ( 2007 ) and Beck et al. ( 2023 ). Paleoclimate data—including monthly mean near-surface air temperature (°C) and monthly total precipitation (mm)—from 81 to 11 Mya were obtained from Valdes et al. ( 2021 ). Estimation of Species Colonization and climate preference in each clade To investigate shifts in colonization rates across geographic regions and climate classes, we applied a rolling estimation method previously used by Xing et Ree (2017). This approach quantifies the relative contribution of each state (i.e., region or climate class) to the diversification of Chrysosplenium . Colonization rates through time were calculated using the formula d ij ( t ) = c ij ( t )/n i (t − 1), where c ij ( t ) represents the number of inferred colonization events from state i to state j at time t . For each node, the sum of colonization probabilities across states satisfies ∑c ij ( t ) = 1. Based on ancestral state reconstructions using BioGeoBEARS (for geographic regions) and BayesTraits (for climate classes), colonization scores were assigned to each node, and the cumulative colonization score for each region and climate class was computed. To assess variation in climate preferences among clades, we performed a non-metric multidimensional scaling (nMDS) analysis using the metaMDS function from the R package vegan (Oksanen et al. 2025). This analysis was conducted using the Jaccard index on a presence/absence species–climate matrix (Table S1 ), with a maximum of 20 random starts and two dimensions. To compare climate class preferences between species in clades B and C, we conducted a permutational multivariate analysis of variance (PERMANOVA) using the adonis2 function and a permutational analysis of multivariate dispersion (PERMDISP) using the permutest function in vegan, both based on the Jaccard index with 5000 permutations. These analyses were restricted to 18 Köppen–Geiger climate classes (Beck et al. 2023 ) for which distribution data of Chrysosplenium species were available. Comparisons involving clade A were not conducted due to its small sample size (n = 2 species). Results Molecular phylogeny, divergence time estimation and temporal-based diversification analyses Bayesian consensus phylogenetic analysis, using 10402 aligned nucleotide sites, revealed three major monophyletic clades (clades A–C) with high posterior probabilities (PP > 0.92; Fig. 1 ). Clade A consisted of Chrysosplenium microspermum and C. sedakowii . Clades B and C included 25 and 30 species, respectively, and each was further divided into three well-supported subclades (B1–3 and C1–3; PPs > 0.88; Fig. 1 ). Species in clades A and B exhibited alternate leaf arrangement (phyllotaxis), while those in clade C showed opposite phyllotaxis. Notably, C. valdivicum , which occurs disjunctly in the Chile-Patagonian region, located a basal position within clade C. The ML analysis yielded a tree with a nearly identical topology (Fig. S1 ). Divergence time estimates based on both younger and older calibrations (Fig. 1 , Fig. S2 , Table S2 ) suggest that the split between Chrysosplenium and Peltoboykinia (node 1) occurred between 46.32–43.20 Mya (95% highest posterior density [HPD]) under the younger calibration, and between 82.33–79.20 Mya under the older calibration (95% HPD). The splitting of clade A from the remaining Chrysosplenium lineages (node 2) was estimated 43.91–24.82 Mya (younger) and 79.90–45.98 Mya (older). The split between clades B and C (node 4) occurred between 44.37–22.72 Mya (younger) and 79.96–41.86 Mya (older). Within these, the diversification of subclades B1–3 (node 5 and 7) and C1–3 (node 10 and 12) began after approximately 22 Mya and 31 Mya under the younger calibration, and after around 40 Mya and 56 Mya under the older calibration, respectively. The RevBayes analyses based on the BEAST tree calibrated with the younger constraint successfully converged, as evidenced by high effective sample sizes (ESS = 5401 for the likelihood of the episodic diversification rate model and ESS = 16351 for the branch-specific diversification rate model). Although no significant diversification rate shifts were detected at specific nodes within the phylogeny, the estimated net diversification remained stable at approximately 0.10 (Fig. 2 ). Results from the analysis using the BEAST tree with the older calibration are not presented here, as our study primarily focuses on the outcomes derived from the younger calibration (see Discussion). Ancestral states reconstruction The ancestral region reconstruction conducted using BioGeoBEARS identified the DEC model as the best-fitting scenario, exhibiting the lowest AICc value (LnL = − 98.42, AICc = 201.05, AICc weight = 0.64; Table S3 ). Furthermore, the reconstruction of ancestral distributable climate classes performed with BayesTraits showed good convergence, as indicated by a high average effective sample size (mean ESS = 3818.7). The results of ancestral state reconstructions (Fig. 1 ; Figs. S3 and S4; Table S2 ) suggest that Chrysosplenium split from Peltoboykinia in the Asian region, with a range of climate classes including BW, BS, Cw, Cf, Ds, Dw, Df, and ET (node 1 in Fig. 1 ). Among these, the Dw climate class had the highest posterior probability (PP = 0.52) at this node. The crown node of Chrysosplenium that means as the splitting between clades A and the B + C lineage and splitting between clades B and C were inferred to be in the Asian region with DW climate class (nodes 2 and 4). The crown nodes of clades A, B (node as splitting subclade B1) and C (node as splitting subclade C1) were all inferred to be in the Asian region, primarily with the Dw climate class (nodes 3, 5 and 10). The splitting of subclade B2 and B3, as well as their respective crown nodes, were also situated in Asia with Dw as the dominant climate (nodes 7, 8 and 9). The split between clades C2 and C3 occurred in both the Asian and Chile–Patagonian regions, with Dw climate class (node 12). The crown node of subclade C1 was located in Asia and most strongly associated with the Df as the prevailing climate class (node 11). The crown node of subclade C2 was inferred to be in Asia with Dw as the prevailing climate class (node 13), while the crown node of subclade C3 was reconstructed in both Asia and the Chile–Patagonian regions, also with Dw-dominated conditions (node 14). A node within clade C3 excluding C. valdivicum was inferred to have occurred in Asia with Dw-dominated climate class (node 15). Estimation of species colonization and climate preference in each clade Rolling estimates of regional colonization rates revealed that the Asian region served as the primary center of diversification for Chrysosplenium . However, several species and/or their ancestors subsequently diversified after dispersing from Asia to other regions, notably Europe, India, and North America (Fig. 3 A). Similarly, colonization rate estimates across distributable climate classes highlighted three major climatic hotspots: Cf (temperate-without dry season), Dw (cold-with dry winter), and Df (cold-without dry season). These climate classes acted as key sources for species expanding into other climate zones. Moreover, diversification events involving migrations among these three climate types also played an important role in the colonization history of Chrysosplenium (Fig. 3 B). The nMDS analysis yielded a stress value of 0.11, indicating an acceptable fit. Although overlaps in climate preferences were observed among clades, the analysis suggested some clade-specific trends: clade B tended to occur in Dfa (old-without dry season-hot summer) and Dfb (old-without dry season-warm summer) climates, while clade C was more often associated with Dwc (cold-without dry season-cold summer) and ET (Polar-tundra) climates (Fig. 4 ). Results of PERMANOVA and PERMDISP analyses showed that clades B and C differed significantly in climate preference, although the dispersion of those preferences was statistically homogeneous (PERMANOVA: F = 2.39, P = 0.024; PERMDISP: F = 0.61, P = 0.44). Discussion Molecular phylogeny of genus Chrysosplenium Our Bayesian phylogenetic analysis resolved three major clades within Chrysosplenium , each strongly supported by high PP. Most subclades within clades B and C also received strong support. Clade A and B were defined by species exhibiting alternate phyllotaxis, whereas clades C was associated with alternate and opposite phyllotaxis, respectively. These findings are broadly consistent with the previous phylogenetic work by Yang et al. ( 2023 ). However, while the overall tree topology matched that earlier study, our analysis further resolved three distinct subclades within clade C—substructure not recovered by Yang et al. ( 2023 ). Additionally, our study placed C. valdivicum within subclade C3, which differs from its placement in Yang et al. ( 2023 ), where it was recovered as sister to a clade containing C. alpinum , C. oppositifolium , C. dubium , C. nepalense , C. sinicum , and C. grayanum . This discrepancy likely results from differences in the quantity and proportion of cpDNA and nrDNA data used. Specifically, our study analyzed approximately 4 kb of cpDNA and 6 kb of nuclear nrDNA, while Yang et al. ( 2023 ) used about 154 kb of cpDNA and 7 kb of nrDNA. In our cpDNA-only phylogeny, C. valdivicum was placed in a clade corresponding to subclade C1 (Fig. S5 ), whereas the nrDNA-based phylogeny in Yang et al. ( 2023 ) aligned C. valdivicum with a group equivalent to our clade C3. Therefore, the phylogenetic position of C. valdivicum remains uncertain. Future studies incorporating a larger and more diverse set of nuclear markers may help resolve this ambiguity and improve phylogenetic resolution across the genus. Moreover, our results suggest that the Japanese endemic species C. macrostemon and C. fauriei are conspecific, a finding consistent with Yang et al. ( 2023 ). This highlights the need for taxonomic revision of these species. Adaptive timeline to the evolutionary history of Chrysosplenium In this study, we conducted two divergence time estimation analyses using alternative calibration ages representing younger and older scenarios (Table S3 ). Based on ancestral state reconstruction, the timeline inferred from the younger calibration appears to align better with paleoclimate reconstructions. For instance, the ancestral state of node 11 (Fig. 1 ) corresponds more closely with the environmental conditions shown in the 21 Mya paleoclimate map than those in the 40 Mya map (Fig. S6). As we employed a secondary calibration approach—using calibration points derived from previous molecular dating studies—it is important to acknowledge the potential for bias toward younger divergence estimates (Schenk 2016 ). Consequently, the results must be interpreted cautiously, with an awareness of the limitations inherent in such methods. Based on this suggestion, across multiple comparisons between estimated node ages and paleoclimate maps of corresponding or slightly earlier ages, the divergence times inferred from the younger calibration points generally showed better congruence than those from the older calibration. This trend supports the suitability of the younger timeline for interpreting the evolutionary history of Chrysosplenium in the context of past climatic changes. Second, when we focus on the node representing the split between C. kitoense and its sister clade—composed mainly of Japanese endemic species located in clade C3—the estimated divergence times under the younger and older calibration scenarios were 12.17 Mya (95% HPD: 17.77–6.97 Mya) and 22.07 Mya (32.12–12.87 Mya in 95% HPD), respectively. Additional ancestral region reconstruction using BioGeoBEARS, in which Japan was treated as a separate region from continental Asia, indicated that the ancestral area for this node was Japan (Fig. S6). Geological evidence suggests that the Japanese archipelago began to separate from the Asian continent around 20 Mya (Isozaki et al. 2010 ). If C. kitoense , which is distributed in Cf (temperate-without dry season) and Df (cold-without dry season) climate zones, had diverged at 22 Mya as suggested by the older calibration, it is likely that it would have also been distributed on the continent, where such climatic zones were present at that time (Fig. S7). This biogeographic mismatch suggests that the divergence estimate based on the younger calibration is more consistent with geological and climatic evidence, supporting its suitability for interpreting the evolutionary history of Chrysosplenium . Previous studies estimated the origin of Chrysosplenium at approximately 25 Mya or 45 Mya (Deng et al. 2015 ; Folk et al. 2019 ), with the latter being more consistent with our results. The differences in these estimates primarily stem from the calibration strategies employed. Both studies used Itea (Iteaceae) as an outgroup, calibrating the stem node of Saxifragaceae (including Chrysosplenium and Iteaceae) using fossil-based ages of 48 Mya (Deng et al. 2015 ) and 89 Mya (Folk et al. 2019 ). In our study, the younger calibration scenario (Zunteni et al. 2024) is largely consistent with that of Folk et al. ( 2019 ), which explains the similarity in divergence time estimates. The origin of Chrysosplenium Our findings shed light on the dispersal and diversification history of Chrysosplenium , revealing that its diversification was initially stable and concentrated in Asia, followed by extensive geographic expansion primarily into the Palearctic and Nearctic regions. Furthermore, our results support a novel hypothesis that Chrysosplenium species predominantly diversified within Cf (temperate-without a dry season), Dw (cold-with dry winter), and Df (cold-without a dry season) climate classes. The results of our ancestral region reconstruction analyses suggest that Chrysosplenium originated in Asia. This finding is consistent with previous studies (Deng et al. 2015 ; Folk et al. 2021 ; Nakazawa et al. 1957; Soltis et al. 2001 ), which have similarly proposed East Asia as the center of origin and diversification for the genus. Furthermore, the subfamily Micrantheae—comprising Micranthes and sister to Chrysosplenieae (which includes Chrysosplenium and Peltoboykinia )—is also thought to have originated in Asia (Folk et al. 2021 ). For example, ancestral species of Micranthes (e.g., M. merkii in western Beringia, M. nudicaulis , and M. tolmiei distributed on both sides of Beringia) exhibit distribution patterns consistent with an Asian origin (PWPO, 2025; Stubbs et al. 2020 ). Additionally, the earliest-diverging subfamily Darmereae—which includes genera such as Astilboides , Bergenia , Darmera , Mukdenia , Oresitrophe , and Rodgersia —is sister to the clade comprising Chrysosplenieae and Micrantheae (Folk et al. 2021 ). Most species in Darmereae are distributed in northeastern China, with Darmera being the notable exception, occurring in western North America (PWPO 2025). Our additional ancestral region reconstruction analysis identified the splitting node between Chrysosplenium and Peltoboykinia in the region of Asia and Japan, further supporting a northeastern Asian origin for Chrysosplenium (Fig. S6). Moreover, distributable climate class reconstructions indicated that the Dw (cold-dry winter) climate zone inferred from ancestral state reconstructions was primarily located in northeastern Asia (Fig. S7). Together, these lines of evidence reinforce the hypothesis that Chrysosplenium originated in northeastern Asia. Following its origin, Chrysosplenium diversified primarily in the Northern Hemisphere, particularly within subclades that first radiated in northeastern Asia under the Dw climate class. Dispersal and diversification scenario of Chrysosplenium Most diversification likely occurred through dispersal within Asia, particularly in China. However, several species— C. carnosum , C. forrestii , and C. lanuginosum in subclade B3; C. nepalense in C1; and C. delavayi in C2—appear to have speciated following dispersal into southern China, where the Cw (temperate-with dry winters) climate class. In addition, C. iowense distributed in the North America shown in subclade B1, C. alpinum distributed in the Europe, and C. dubium distributed in the western Asia, Europe, and Saharo-Arabia in subclade C1 likely diversified after long-distance dispersal events to their respective regions. C. alternifolium in subclade B1 and C. oppositifolium in subclade C1 seem to have expanded westward through areas characterized by Df (cold-without a dry season) climate class. Conversely, C. rosendahlii and C. wrightii in subclade B1 and C. americanum in C1 likely spread eastward under the same Df climatic conditions, crossing the Bering land bridge during the Neogene or Quaternary periods—a route known to facilitate floristic exchange from East Asia to western North America (Wen et al. 2016 ). C. tetrandrum is notable for its bidirectional expansion, both eastward and westward, under the Df climate regime. Several species endemic to Japan, including C. maximowiczii , C. tosaense , and the ancestral lineages of C. album and C. rhabdospermum , likely originated from continental ancestors and underwent allopatric speciation after migrating to the Japanese archipelago. Additionally, species such as C. kamtschatium located in clade C3 found within the same clade as C. kitoense , appear to have migrated into the Sakhalin and the Kamchatka Peninsula following diversified in Japan (Fig. S6). According to the results of ancestral distributable climate reconstruction, along with nMDS, PERMANOVA, and PERMDISP analyses, both ancestral and extant species of Chrysosplenium could be distributed across most of the climate classes where current species are found (see node 1 in Fig. 1 and Fig. S3 ). Donoghue ( 2008 ) suggested that ecological traits influencing species distributions tend to evolve gradually. For instance, the transition from tropical climates to conditions involving freezing temperatures and high seasonality is considered a slow evolutionary process. Therefore, certain traits enabling plant dispersal in response to climate change may have evolved early in the lineage (Donoghue 2008 ). Our findings possibly suggest that Chrysosplenium species had already adapted to a wide range of climatic conditions and this characteristic leads stable diversification in mainly Cf (temperate-without a dry season), Dw (cold-with dry winter), and Df (cold-without a dry season) climate classes within genus Chrysosplenium . Both China and Japan are known for having many endemic species of Chrysosplenium . Although our study does not fully resolve the reasons for the high endemism in these regions, the Japanese endemic species likely originated via allopatric speciation in mountainous regions (Kubota et al. 2015 ), a pattern that may also apply to China. Thus, many endemic Chrysosplenium species may have diversified through habitat fragmentation and isolation. Additionally, members of Chrysosplenieae are considered habitat specialists of moist woodlands and are commonly found in mid-elevation mountain zones (Folk et al. 2021 ). Such environments may have promoted allopatric conditions more effectively than open, lowland habitats like grasslands. In this study, we were unable to determine the precise process underlying the disjunct distribution of C. valdivicum . Ancestral state reconstruction suggests that this species likely migrated from Asia to the Chile-Patagonian region under Cf (temperate-without a dry season) and Dw (cold-dry winter) climate classes (Fig. 1 ). However, such a direct, long-distance dispersal from Asia to South America seems unlikely. Although this distribution pattern represents an amphitropical disjunction in the Americas, most comparable plant migrations into South America are believed to have originated from North America (Simpson et al. 2017 ). Liu et al. ( 2023 ) defined both Asia and North America as parts of the Holarctic region, suggesting that C. valdivicum or its ancestral species may have first spread into North America—potentially via the Bering land bridge—before reaching South America under Df (cold-without dry season) climate class or similar conditions. Interestingly, two additional Chrysosplenium species are found in South America: C. aulacocarpum in Colombia and Venezuela, and C. macranthum , which, like C. valdivicum , exhibits a disjunct distribution between Argentina and Chile. Further investigation of these species could provide deeper insights into the biogeographic history and dispersal routes of Chrysosplenium in South America. This study suggests that plant species have historically migrated into suitable habitats and undergone diversification in response to past climate changes. The ancestral climate reconstruction indicates that Chrysosplenium species had already begun adapting to Cf (temperate-without dry season), Dw (cold-dry winter) and Df (cold-without dry season) climate classes since their early evolutionary history. However, a key question remains: why are most species still confined to limited geographic areas despite the existence of climatically suitable habitats elsewhere? To address this, further research is needed to uncover the factors shaping plant distribution. Physiological studies may offer valuable insights into this issue. Previous findings show that Chrysosplenium species occupy a wide range of environmental conditions, such as variation in mean annual precipitation (Folk et al. 2021 ), suggesting that they differ in physiological traits related to water use. These traits may play a crucial role in determining current distribution patterns and the speciation processes within the genus. Declarations Competing Interests: None Funding: This research was financially supported by the JSPS KAKENHI (23K14240). Acknowledgements We would like to express our gratitude to Dr. Kenji Izumi (Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences) for his contribution to the reconstruction of paleoclimate maps in this study. This research was financially supported by the JSPS KAKENHI (23K14240). This study was conducted using an existing dataset; therefore, no fieldwork requiring specific permissions was conducted, and no permits were necessary. We sincerely thank the researchers of previous studies for providing the datasets used in this research. Data availability The data utilized in this study are accessible through GenBank ( https://www.ncbi.nlm.nih.gov/genbank/ ) and the Plants of the World Online website ( http://www.plantsoftheworldonline.org/ ) and the World Flora Online Plant List ( https://wfoplantlist.org ). References Beck HE, McVicar TR, Vergopolan N, Berg A, Lutsko JL, Dufour A, Zeng Z, Jiang X, van Dijk AIJM, Miralles DG (2023) High-resolution (1 km) Köppen-Geiger maps for 1901–2099 based on constrained CMIP6 projections. Sci Data 10:724. 10.1038/s41597-023-02549-6 Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu CH, Xie D, Suchard MA, Rambaut A, Drummond AJ (2014) BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput Biol 10:e1003537. 10.1371/journal.pcbi.1003537 Deng JB, Drew BT, Mavrodiev EV, Gitzendanner MA, Soltis PS, Soltis DE (2015) Phylogeny, divergence times, and historical biogeography of the angiosperm family Saxifragaceae. Mol Phyl Evol 83:86–98. 10.1016/j.ympev.2014.11.011 Donoghue MJ (2008) A phylogenetic perspective on the distribution of plant diversity. PNAS 105:S11549–S11555. 10.1073/pnas.0801962105 Folk RA, Stubbs RL, Mort ME, Cellinese N, Allen JM, Soltis PS, Soltis DE, Guralnick RP (2019) Rates of niche and phenotype evolution lag behind diversification in a temperate radiation. PNAS 116:10874–10882. 10.1073/pnas.1817999116 Folk RA, Stubbs RL, Engle-Wrye NJ, Soltis DE, Okuyama Y (2021) Biogeography and habitat evolution of Saxifragaceae, with a revision of generic limits and a new tribal system. Taxon 70:263–285. 10.1002/tax.12450 Hara H (1957) Synopsis of the genus Chrysosplenium L. J Fac Sci Univ Tokyo Sect III Bot 7:1–90 Höhna S, Landis MJ, Heath TA, Boussau B, Lartillot N, Moore BR, Huelsenbeck JP, Ronquist F (2016) RevBayes: Bayesian Phylogenetic Inference Using Graphical Models and an Interactive Model–Specification Language. Syst Biol 65:726–736. 10.1093/sysbio/syw021 Isaka Y, Nakamura A, Izumi K, Schneider H (2025) Molecular phylogenetic analyses reveal the diversification and dispersal processes of the tribe Shoreae (Dipterocarpaceae). J Biogeo e15166. 10.1111/jbi.15166 Isozaki Y, Aoki K, Nakama T, Yanai S (2010) New insight into a subduction-related orogen: Reappraisal on geotectonic framework and evolution of the Japanese Islands. Gondwana Res 18:82–105. 10.1016/j.gr.2010.02.015 Kozlov AM, Darriba D, Flouri T, Morel B, Stamatakis A (2019) RAxML-NG: a fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference. Bioinformatics 35:4453–4455. 10.1093/bioinformatics/btz305 Kubota Y, Shiono T, Kusumoto B (2015) Role of climate and geohistorical factors in driving plant richness patterns and endemicity on the east Asian continental islands. Ecography 38:639–648. 10.1111/ecog.00981 Kuraku S, Zmasek CM, Nishimura O, Katoh K (2013) aLeaves facilitates on-demand exploration of metazoan gene family trees on MAFFT sequence alignment server with enhanced interactivity. Nucl Acids Res 41:W22–W28. 10.1093/nar/gkt389 Lanfear R, Frandsen PB, Wright AM, Senfeld T, Calcott B (2017) PartitionFinder 2: new methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Mol Biol Evol 34:772–773. 10.1093/molbev/msw260 Liu Y, Xu X, Dimitrov D, Pellissier L, Borregaard MK, Shrestha N, Su X, Luo A, Zimmermann NE, Rahbek C, Wang Z (2023) An updated floristic map of the world. Nat Commun 14:2990. 10.1038/s41467-023-38375-y Matzke NJ (2013) Probabilistic historical biogeography: new models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing. Dissertation, University of California Nakazawa M, Wakabayashi M, Ono M, Murata J (1997) Molecular phylogenetic analysis of Chrysosplenium (Saxifragaceae) in Japan. J Plant Res 110:265–274. 10.1007/BF02509315 Oksanen J, Blanchet FG, Kindt R et al (2011) vegan: community ecology package. R package version 2.6–6.1. https://github.com/vegandevs/vegan Pagel M, Meade A, Barker D (2004) Bayesian estimation of ancestral character states on phylogenies. Syst Biol 53:673–684. 10.1080/10635150490522232 Peel MC, Finlayson BL, McMahon TA (2007) Updated World Map of the Köppen-Geiger Climate Classification. Hydrol Earth Syst Sci 11:1633–1644. 10.5194/hess-11-1633-2007 POWO (2024) Plants of the World Online. Facilitated by the Royal Botanic Gardens, Kew. https://powo.science.kew.org/ . accessed 20 April 2025 Rambaut A (2018) ) FigTree v.1.4.4. https://github.com/rambaut/Figtree/releases/tag/v1.4.4 Rambaut A, Drummond AJ, Xie D, Baele G, Suchard MA (2018) Posterior summarization in Bayesian phylogenetics using Tracer 1.7. Syst Biol 67:901–904. 10.1093/sysbio/syy032 R Core Team (2024) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing v.4.3.2. Vienna, Austria: R foundation for Statistical Computing. https://www.R-project.org/ Ronquist F, Teslenko M, van Der Mark P, Ayres DL, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst Boil 61:539–542. 10.1093/sysbio/sys029 Schenk JJ (2016) Consequences of secondary calibrations on divergence time estimates. PLoS ONE 11:e0148228. 10.1371/journal.pone.0148228 Simpson MG, Johnson LA, Villaverde T, Guilliams CM (2017) American amphitropical disjuncts: perspectives from vascular plant analyses and prospects for future research. Ame J Bot 104:1600–1650. 10.3732/ajb.1700308 Soltis DE, Tago-Nakazawa M, Xiang QY, Kawano S, Murata J, Wakabayashi M, Hibsch‐Jetter C (2001) Phylogenetic relationships and evolution in Chrysosplenium (Saxifragaceae) based on matK sequence data. Ame J Bot 88:883–893. 10.2307/2657040 Stubbs RL, Folk RA, Xiang CL, Chen S, Soltis DE, Cellinese N (2020) A phylogenomic perspective on evolution and discordance in the alpine-arctic plant clade Micranthes (Saxifragaceae). Front Plant Sci 10:1773. 10.3389/fpls.2019.01773 Tribble CM, Freyman WA, Landis MJ, Lim JY, Barido-Sottani J, Kopperud BT, Höhna S, May MR (2022) RevGadgets: an R package for visualizing Bayesian phylogenetic analyses from RevBayes. Met Ecol Evol 13:314–323. 10.1111/2041-210X.13750 Valdes PJ, Scotese CR, Lunt DJ (2021) Deep Ocean Temperatures Through Time. Clim Past 17:1483–1506. 10.5194/cp-17-1483-2021 Wen J, Nie ZL, Ickert-Bond SM (2016) Intercontinental disjunctions between eastern Asia and western North America in vascular plants highlight the biogeographic importance of the Bering land bridge from late Cretaceous to Neogene. J Syst Evol 54:469–490. 10.1111/jse.12222 Wen J, Xie DF, Price M, Ren T, Deng YQ, Gui LJ, Guo XL, He XJ (2021) Backbone phylogeny and evolution of Apioideae (Apiaceae): New insights from phylogenomic analyses of plastome data. Mol Phyl Evol 161:107183. 10.1016/j.ympev.2021.107183 WFO (2024) The World Flora Online. https://worldfloraonline.org . Accessed 20 April 2025 Xing Y, Ree RH (2017) Uplift-driven diversification in the Hengduan Mountains, a temperate biodiversity hotspot. PNAS 114:E3444–E3451. 10.1073/pnas.1616063114 Yang T, Wu Z, Tie J, Qin R, Wang J, Liu H (2023) A comprehensive analysis of chloroplast genome provides new insights into the evolution of the genus Chrysosplenium . Int J Mol Sci 24:14735. 10.3390/ijms241914735 Yule GU, Willis JC, Philos FRS (1925) Trans R Soc Lond 213: 21–87. doi: 10.1098/rstb.1925.0002 Zuntini AR, Carruthers T, Maurin O et al (2024) Phylogenomics and the rise of the angiosperms. Nature 629:843–850. 10.1038/s41586-024-07324-0 Supplementary Files ChrysoFigSs.pdf ChrysospleniumTableS1.xlsx ChrysospleniumTableS2.xlsx ChrysospleniumTableS3.xlsx SIContentsChrysosplenium.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7444744","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":516226838,"identity":"000f18f4-3081-4835-aa85-6d47fbf8723e","order_by":0,"name":"Tsubasa Toji","email":"","orcid":"","institution":"Nihon University: Nihon Daigaku","correspondingAuthor":false,"prefix":"","firstName":"Tsubasa","middleName":"","lastName":"Toji","suffix":""},{"id":516226839,"identity":"2e203e79-ac3f-4dcd-a4ce-ab6b7e0724b2","order_by":1,"name":"Takayuki Sawai","email":"","orcid":"","institution":"Shinshu University: Shinshu Daigaku","correspondingAuthor":false,"prefix":"","firstName":"Takayuki","middleName":"","lastName":"Sawai","suffix":""},{"id":516226840,"identity":"ac58c348-2870-4074-bd9c-9b6d408cc774","order_by":2,"name":"Yuichi Isaka","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-2220-5635","institution":"Xishuangbanna Tropical Botanical Garden","correspondingAuthor":true,"prefix":"","firstName":"Yuichi","middleName":"","lastName":"Isaka","suffix":""}],"badges":[],"createdAt":"2025-08-24 07:51:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7444744/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7444744/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92193384,"identity":"d0f0a6e0-4b6f-4293-b876-2bf147d6d4c6","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"xml","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8121,"visible":true,"origin":"","legend":"","description":"","filename":"jpreJPRED2500828.xml","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/f4340e15b484d002e3f0621e.xml"},{"id":92193386,"identity":"a34dd8e8-4b58-4a46-9e01-0f16999ad3f9","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1169,"visible":true,"origin":"","legend":"","description":"","filename":"JPRED250082816324.go.xml","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/7c8775aa18d2e3193d5c2053.xml"},{"id":92194704,"identity":"e0633352-f920-41a8-a793-2f00a8bba606","added_by":"auto","created_at":"2025-09-25 15:41:37","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":857,"visible":true,"origin":"","legend":"","description":"","filename":"JPRED2500828Import.xml","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/bcd19dd2655197b4a35ebd0f.xml"},{"id":92194708,"identity":"b862e169-dd1b-40e7-97c8-c0a6bf82d324","added_by":"auto","created_at":"2025-09-25 15:41:37","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":123646,"visible":true,"origin":"","legend":"","description":"","filename":"JPRED25008280enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/ae51ef4b0deaeb81089b2e3a.xml"},{"id":92193397,"identity":"96b1cdbb-3e33-4cea-8830-07866b318d07","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"emf","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1787548,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.emf","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/a04eeffd8792e7a57683e1fb.emf"},{"id":92193393,"identity":"a0de7817-a782-4562-9fd8-e9f7933f71e1","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25456,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/660802c185f5c64ec8f8f38a.png"},{"id":92193401,"identity":"68c75142-1753-4e22-93cf-a8642dffcfe2","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"emf","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":311940,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.emf","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/cec849a58994d7b6f8f1e0cd.emf"},{"id":92194707,"identity":"9750cb44-bc7b-4882-86e3-4ecc4a6af386","added_by":"auto","created_at":"2025-09-25 15:41:37","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54768,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/660a4389419b4db3e4e646d9.jpeg"},{"id":92193399,"identity":"cd5fb043-09a3-4983-bf6a-b2afc7946a68","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":121480,"visible":true,"origin":"","legend":"","description":"","filename":"JPRED25008280structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/5d45ba6e60aa052fc04b0538.xml"},{"id":92193395,"identity":"663b3037-ddf8-4fcb-a1f0-da5e137710ff","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":134579,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/b63ff9afedbb43bc805c0a05.html"},{"id":92194703,"identity":"237b2050-36d1-4c76-a885-2c4ac723f335","added_by":"auto","created_at":"2025-09-25 15:41:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":43352,"visible":true,"origin":"","legend":"\u003cp\u003eBayesian molecular phylogenetic tree of \u003cem\u003eChrysosplenium\u003c/em\u003ereconstructed using molecular data from partial cpDNA, \u003cem\u003etrnL–trnF\u003c/em\u003e, \u003cem\u003erbcL\u003c/em\u003e, and \u003cem\u003ematK \u003c/em\u003eand partial nrDNA \u003cem\u003eITS\u003c/em\u003e regions. Each Bayesian posterior probability is shown near each branch. Pie charts on a node show the results of ancestral region (left) and distributable climate class (right) reconstruction based on the Bayesian phylogeny (color legend in the bottom). Present distribution and climate class are denoted on the right side of the species name. Chinese, Japanese, Korean, and Russian region-specific species are denoted by stars, triangles, double circles, and crosses, respectively. Non-region-specific species are denoted by squares (colors in the symbols correspond to colored regions in the map). Distributable climate class of each species is denoted by circles (colors in the symbols correspond to colored legend in the bottom right). The map delineates regions defined (Liu et al. 2023) in BioGeoBEARS: Asian (Central Asia, Russian, East Asia, and Norway); Chile-Patagonian (Chile and southern Argentina); European (Europe and Turkey); Indian (India, Sri Lanka, Indo-China Peninsula, and Hainan Island); Malaysian (Taiwan, Philippines, Malaysia, and the islands of New Guinea); North American (The United States and Canada); Saharo-Arabian (Western Sahara, Morocco, Algeria, Tunisia, Egypt, Libya, and Saudi Arabia). Climate classes defined (Beck et al. 2023) in BayesTraits4: BW (arid-desert); BS (arid-steppe); Cs (temperate-dry summer); Cw (temperate-dry winter); Cf (temperate-without dry season); Ds (cold-dry summer); Dw (cold-dry winter); Df (cold-without dry season); ET (polar-tundra). Boxes and bars shown in right side are indicating grouping among \u003cem\u003eChrysosplenium\u003c/em\u003especies. Asterisk indicates a calibration node. Estimated divergence time with 95% highest probability density (HPD) on each node with younger calibration is shown in middle left. Mya, million years ago.\u003c/p\u003e","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/8e0d48c716edfc721024a05c.png"},{"id":92193382,"identity":"1c84de25-c83e-4b8c-a3f3-fbcfa973a60f","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":10593,"visible":true,"origin":"","legend":"\u003cp\u003eResults of diversification analyses of \u003cem\u003eChrysosplenium\u003c/em\u003e. Molecular phylogenetic tree was reconstructed by BEAST with younger calibration. The graph displays a net-diversification rate resulted from the episodic birth-death model implemented in RevBayes, where the red line represents the net-diversification rate and the blue area indicates its 95% confidence intervals. Color pattern illustrates the variation in net-diversification rates of \u003cem\u003eChrysosplenium\u003c/em\u003e. This colored phylogeny was based on the scenario with no diversification rate shift. Mya, million years ago.\u003c/p\u003e","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/556a1177fb0f186316b4c209.png"},{"id":92194702,"identity":"3775067c-faee-4120-973e-4ba822585041","added_by":"auto","created_at":"2025-09-25 15:41:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":20179,"visible":true,"origin":"","legend":"\u003cp\u003eAssembly of \u003cem\u003eChrysosplenium\u003c/em\u003e species across different (a) regions and (b) climate classes, with colonization rates (\u003cem\u003er\u003c/em\u003e) calculated by the rolling estimates through time. Numbers within circles represent colonization rates per state within specific regions or climate classes, while numbers on arrows denote colonization rates associated with dispersal between regions or between climate classes (thin numbers: 0 ≤ \u003cem\u003er\u003c/em\u003e \u0026lt; 1, medium: 1 ≤ \u003cem\u003er\u003c/em\u003e \u0026lt; 10, thick: 10 ≤ \u003cem\u003er\u003c/em\u003e). Regions: Asian (Central Asia, Russian, East Asia, and Norway); Chile-Patagonian (Chile and southern Argentina); European (Europe and Turkey); Indian (India, Sri Lanka, Indo-China Peninsula, and Hainan Island); Malaysian (Taiwan, Philippines, Malaysia, and the islands of New Guinea); North American (The United States and Canada); Saharo-Arabian (Western Sahara, Morocco, Algeria, Tunisia, Egypt, Libya, and Saudi Arabia). Climate classes: BW (arid-desert); BS (arid-steppe); Cs (temperate-dry summer); Cw (temperate-dry winter); Cf (temperate-without dry season); Ds (cold-dry summer); Dw (cold-dry winter); Df (cold-without dry season); ET (polar-tundra).\u003c/p\u003e","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/53b2000200f4303ea9c6b142.png"},{"id":92193392,"identity":"d1f2c386-8a73-4673-bb4e-eaad55fda514","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":24770,"visible":true,"origin":"","legend":"\u003cp\u003eNon-metric multi-dimensional scaling (NMDS) ordination among \u003cem\u003eChrysosplenium\u003c/em\u003e species based on their present distributed climate class using the jaccard index. Square, circles and triangles are indicating species belonging to clade A, B, and C showed in Bayesian phylogeny (Fig. 1), respectively. As well as, polygons colored with red and blue are belonging to clade B1–3 and C1–3 in Bayesian phylogeny (Fig. 1), respectively. Polygon indicating Clade A was not shown as it is composed of two species. Cwa (temperate-dry winter-hot summer); Cfa (temperate-without dry season-hot summer); Dwb (cold-dry winter-warm summer); Dwc (cold-dry winter-cold summer); Dfa (cold-without dry season-hot summer); Dfb (cold-without dry season-warm summer); Dfc (cold-without dry season-cold summer); ET (Polar-tundra).\u003c/p\u003e","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/7ac8dd20e8b3d010ce4f9592.png"},{"id":96604256,"identity":"a6fd83dc-9b70-446f-807f-adee2eee3f51","added_by":"auto","created_at":"2025-11-24 09:13:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":979746,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/213a980e-2223-40ff-9fed-30c8e6712205.pdf"},{"id":92193400,"identity":"dd1405a4-9ada-4ba9-be55-006a2723eb83","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":6763992,"visible":true,"origin":"","legend":"","description":"","filename":"ChrysoFigSs.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/234341bbec1ad4de39128c9f.pdf"},{"id":92193389,"identity":"8242b9bd-50d1-498b-8305-e3a27a6add3f","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":22783,"visible":true,"origin":"","legend":"","description":"","filename":"ChrysospleniumTableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/ad5912172df86e0e03a039a3.xlsx"},{"id":92195273,"identity":"3f63226b-d40d-43d0-90bd-8a8075e117a1","added_by":"auto","created_at":"2025-09-25 15:49:37","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":12480,"visible":true,"origin":"","legend":"","description":"","filename":"ChrysospleniumTableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/9ab0dd74958cd2343da16b7f.xlsx"},{"id":92193388,"identity":"9a1a58b9-4ef3-4531-9839-f9f074e2e915","added_by":"auto","created_at":"2025-09-25 15:33:37","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":11227,"visible":true,"origin":"","legend":"","description":"","filename":"ChrysospleniumTableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/e5d79c21edb365db9f5767c5.xlsx"},{"id":92194705,"identity":"f506f59c-ca83-4bd8-aaa2-9060ca7ec77d","added_by":"auto","created_at":"2025-09-25 15:41:37","extension":"docx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":16800,"visible":true,"origin":"","legend":"","description":"","filename":"SIContentsChrysosplenium.docx","url":"https://assets-eu.researchsquare.com/files/rs-7444744/v1/8a2bba556e946916f50aafa9.docx"}],"financialInterests":"","formattedTitle":"Evolutionary history of Chrysosplenium L. (Saxifragaceae): a molecular phylogenetic approach to dispersal and diversification processes","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe genus \u003cem\u003eChrysosplenium\u003c/em\u003e L. (Saxifragaceae), commonly known as golden saxifrages, comprises over 85 species (POWO 2025; WFO 2025). These small perennial herbs are typically found in moist habitats and are predominantly distributed across the cold and temperate regions of the Northern Hemisphere. The genus shows particularly high diversity in Asia, where species have radiated into a broad range of ecological niches (Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Soltis et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Yang et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). \u003cem\u003eChrysosplenium\u003c/em\u003e also presents a notable South American\u0026ndash;East Asian disjunction, with two species (\u003cem\u003eC. valdivicum\u003c/em\u003e and \u003cem\u003eC. macranthum\u003c/em\u003e) occurring in southern South America (Argentina and Chile), though the biogeographic history of this pattern remains incompletely understood (Hara \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1957\u003c/span\u003e). Asia harbors a major portion of the genus\u0026rsquo;s diversity, offering a unique setting to investigate diversification processes. More than 30 species are narrowly distributed in continental Asia\u0026mdash;especially in China, which hosts over 20 endemic species. In addition, 13 island endemics are found in Japan (11 species) and Taiwan (2 species) (POWO 2025).\u003c/p\u003e\u003cp\u003eTo date, various studies have explored phylogenetic relationships, biogeographic patterns, and diversification within \u003cem\u003eChrysosplenium\u003c/em\u003e (e.g., Deng et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Nakazawa et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Soltis et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Yang et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). A recent study by Yang et al. (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), which analyzed chloroplast genomes and partial nuclear gene sequences from 48 species, confirmed the monophyly of \u003cem\u003eChrysosplenium\u003c/em\u003e and identified three main clades. This study also found that species with opposite phyllotaxis form a monophyletic group, whereas those with alternate phyllotaxis are polyphyletic, confirming earlier morphological classifications (Hara \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1957\u003c/span\u003e; Nakazawa et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). Previous molecular studies, including phylogeographic analyses (Deng et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Soltis et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), suggest that the genus originated in East Asia. Moreover, metadata analyses indicate that the subtribe Chrysospleninae, which includes \u003cem\u003eChrysosplenium\u003c/em\u003e, is a specialist of wet woodlands, with transitions to arctic habitats occurring throughout its evolutionary history (Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe diversification of East Asian plant lineages has been linked to the development and intensification of the East Asian climate (Wen et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), highlighting the role of past climate change in shaping plant evolution. Additionally, the formation of the Japanese archipelago likely influenced the diversification and distribution of island endemics (Deng et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Thus, the climatic and geological history of East Asia may have played a crucial role in driving the evolutionary dynamics of \u003cem\u003eChrysosplenium\u003c/em\u003e. Recently, integrating molecular phylogenetics with paleoclimatic reconstructions has enhanced our understanding of such processes (e.g., Isaka et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). While previous research has made important contributions to the taxonomy and evolutionary biology of \u003cem\u003eChrysosplenium\u003c/em\u003e, further insight can be gained by expanding taxon sampling using open-access molecular data (e.g., GenBank: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/genbank/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/genbank/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Additionally, recent advances in angiosperm phylogenomics and divergence time calibration (ranging from 154 to 247\u0026nbsp;million years ago [Mya] at the crown node of angiosperms; Zuntini et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) allow for more precise evolutionary timeline reconstructions. Furthermore, integrating updated global floristic regionalization (Liu et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and paleoclimatic distribution modeling enables a refined understanding of both ancestral area and climatic niche evolution.\u003c/p\u003e\u003cp\u003eIn this study, we perform a molecular phylogenetic analysis to reconstruct the evolutionary history of \u003cem\u003eChrysosplenium\u003c/em\u003e. Using sequence data, distribution and climatic information, and paleoclimate reconstructions from open-access resources, we address the following objectives: (1) to infer phylogenetic relationships among \u003cem\u003eChrysosplenium\u003c/em\u003e species; (2) to assess which divergence timeline better explains the genus\u0026rsquo;s evolutionary history; (3) to elucidate patterns of distributional shifts with respect to climatic variables; and (4) to identify the center of diversification within the genus.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eTaxon sampling\u003c/h2\u003e\u003cp\u003eThe genus \u003cem\u003eChrysosplenium\u003c/em\u003e includes 86 recognized species (POWO 2025). We retrieved sequence data for 57 \u003cem\u003eChrysosplenium\u003c/em\u003e species and one outgroup species, \u003cem\u003ePeltoboykinia tellimoides\u003c/em\u003e, from GenBank (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/genbank/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/genbank/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). \u003cem\u003ePeltoboykinia\u003c/em\u003e, comprising two species, is considered the sister group to \u003cem\u003eChrysosplenium\u003c/em\u003e (Yang et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Zuntini et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Details of all sampled taxa and accession numbers are listed in Appendix 1 and Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003ePhylogenetic analyses\u003c/h3\u003e\n\u003cp\u003eA concatenated dataset comprising four chloroplast DNA regions (t\u003cem\u003ernL\u0026ndash;trnF\u003c/em\u003e, \u003cem\u003erbcL\u003c/em\u003e, and \u003cem\u003ematK\u003c/em\u003e) and the nuclear \u003cem\u003eITS\u003c/em\u003e region was assembled. Sequence alignment for each region was performed using MAFFT v.7.490 (Kuraku et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) with default parameters and manually corrected where necessary. PartitionFinder v.2.1 (Lanfear et al. 2016) was used to select the best-fit substitution model for each partition (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Bayesian inference was conducted in MrBayes v.3.2.7a (Ronquist et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), employing partition-specific models under the Bayesian Information Criterion (BIC). The Markov Chain Monte Carlo (MCMC) Metropolis\u0026ndash;Hastings algorithm was run for 10\u0026nbsp;million generations, sampling every 100 generations. Two output log files were evaluated based on the effective sample size (ESS) greater than 200 after removing the 25% burn-in using Tracer v.1.7.2 (Rambaut et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), with the first 25% of samples discarded as burn-in. Clade support was evaluated using Bayesian posterior probabilities (PPs).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eThe best-fitted substitution model for each region using Bayesian and maximum likelihood phylogenetic reconstruction.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eData set\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eRegions\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eNo. of sequences\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eNo. of sites\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSubstitution model*\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003echlDNA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cem\u003etrnL\u0026ndash;trnF\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1101\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eGTR\u0026thinsp;+\u0026thinsp;G\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cem\u003erbcL\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1428\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eGTR\u0026thinsp;+\u0026thinsp;I\u0026thinsp;+\u0026thinsp;G\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cem\u003ematK\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1650\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eGTR\u0026thinsp;+\u0026thinsp;G\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003enrDNA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cem\u003eITS\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e6223\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eGTR\u0026thinsp;+\u0026thinsp;I\u0026thinsp;+\u0026thinsp;G\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCombined**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e10402\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eGTR\u0026thinsp;+\u0026thinsp;I\u0026thinsp;+\u0026thinsp;G\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003e*: Same substitution models were selected for both Bayesian inference and maximum likelihood phylogenetic analyses.\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003e**: substitution model for BEAST analysis\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTo validate the topology, maximum likelihood (ML) analysis was performed in RAxML-NG v.1.2.0 (Kozlov et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) using substitution models selected under Akaike Information Criterion (AIC). Branch support was assessed via 1,000 bootstrap replicates. Phylogenetic trees were visualized using FigTree v.1.4.4 (Rambaut \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eDivergence time estimation and temporal-based diversification analyses\u003c/h3\u003e\n\u003cp\u003eDivergence times were estimated using BEAST v.2.7.6 (Bouckaert et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) with the optimized relaxed clock model and tree prior based on the Yule model (pure-birth model, Yule 1925). The combined partition scheme from Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e was used. Subclade B2, as identified in Bayesian analysis (see Results; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), was constrained as monophyletic. Two calibration scenarios were applied for the \u003cem\u003eChrysosplenium\u003c/em\u003e\u0026ndash;\u003cem\u003ePeltoboykinia\u003c/em\u003e splitting: an \"older\" estimate (tree height\u0026thinsp;=\u0026thinsp;80.83 Mya) and a \"younger\" estimate (tree height\u0026thinsp;=\u0026thinsp;44.87 Mya), both from Zuntini et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). MCMC chains ran for 100 and 70\u0026nbsp;million generations, respectively, with sampling every 1,000 generations. All output files were evaluated based on the ESS (\u0026gt;\u0026thinsp;200) after the removal of 10% burn-in and were combined using Tracer v.1.7.2 (Rambaut et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The output tree files from the two independent runs were pooled into a single combined file after removing the initial 10% trees as burn-in using LogCombiner v.2.7.6 (in the BEAST package) and the maximum clade credibility tree with mean node heights was summarized using TreeAnnotator v.2.7.6 (in the BEAST package) before visualization with FigTree v.1.4.4 (Rambaut \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Phylogenetic support was assessed by Bayesian PPs.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo evaluate variation in evolutionary rates within \u003cem\u003eChrysosplenium\u003c/em\u003e, we conducted diversification analyses using RevBayes v.1.2.4 (H\u0026ouml;hna et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), based on a BEAST-derived phylogenetic tree calibrated with the younger age prior. We assessed changes in net diversification rates across the evolutionary history of the genus by applying an episodic diversification model. This analysis followed the online tutorial (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://revbayes.github.io/tutorials/divrate/ebd\u003c/span\u003e\u003cspan address=\"https://revbayes.github.io/tutorials/divrate/ebd\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), with the sampling fraction set to ρ\u0026thinsp;=\u0026thinsp;0.663 (57 of 86 known species) and a chain length of 30000 generations, sampling every 200 generations. Visualizations of rate variation were generated using the R package RevGadgets v.1.2.1 (Tribble et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). To identify potential rate shifts along specific lineages, we also estimated branch-specific diversification rates, following the approach outlined in the RevBayes tutorial (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://revbayes.github.io/tutorials/divrate/branch_specific\u003c/span\u003e\u003cspan address=\"https://revbayes.github.io/tutorials/divrate/branch_specific\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) with ρ\u0026thinsp;=\u0026thinsp;0.663 (57 out of 86 total samples) and a chain length of 30000 generations, sampling every 200 generations. Again, we used RevGadgets v.1.2.1 (Tribble et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) for visualizing shifts in diversification rates within \u003cem\u003eChrysosplenium\u003c/em\u003e..\u003c/p\u003e\n\u003ch3\u003eAncestral region and ancestral distributable climate reconstruction analyses\u003c/h3\u003e\n\u003cp\u003eTo reconstruct the historical biogeography of \u003cem\u003eChrysosplenium\u003c/em\u003e based on the inferred phylogeny, we conducted ancestral region reconstruction using the R package BioGeoBEARS v.1.1.3 (Matzke \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) in R v.4.3.2 (R Core Team 2023). BioGeoBEARS allows comparison among multiple probabilistic models for historical range evolution and identifies the best-fitting scenario. Distribution data for each species (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) were compiled primarily from the Plants of the World Online (POWO 2025), and biogeographic regions were delimited following Liu et al. (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e): (1) Asian region (Central Asia, Russia, East Asia, and Norway); (2) Chile-Patagonian region (Chile and southern Argentina); (3) European region (Europe and Turkey); (4) Indian region (India, Sri Lanka, Indo-China Peninsula, and Hainan Island); (5) Malaysian region (Taiwan, Philippines, Malaysia, and New Guinea); (6) North American region (United States and Canada); and (7) Saharo-Arabian region (Western Sahara, Morocco, Algeria, Tunisia, Egypt, Libya, and Saudi Arabia). These regions are illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eWe evaluated six models implemented in BioGeoBEARS: DEC (dispersal-extinctioncladogenesis), DEC\u0026thinsp;+\u0026thinsp;J (including founder-event speciation), DIVALIKE (a likelihood version of dispersal-vicariance), DIVALIKE\u0026thinsp;+\u0026thinsp;J, BAYAREALIKE (a likelihood version of the Bayesian inference of historical biogeography for discrete areas), and BAYAREALIKE\u0026thinsp;+\u0026thinsp;J. Model comparisons were conducted using corrected Akaike Information Criterion with correction (AICc) scores to identify the most plausible scenario. The model with the lowest AICc value was selected as the best fit for explaining the historical biogeographic patterns of \u003cem\u003eChrysosplenium\u003c/em\u003e.\u003c/p\u003e\u003cp\u003eTo infer the historical climatic preferences of \u003cem\u003eChrysosplenium\u003c/em\u003e species, we performed ancestral state reconstruction of distributable climate classes using BayesTraits v.4 (Pagel et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Distributable climate data for each species (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) were derived from species occurrence records and a global climate classification system (Beck et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Occurrence data were primarily obtained from the Global Biodiversity Information Facility (GBIF; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gbif.org\u003c/span\u003e\u003cspan address=\"https://www.gbif.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and each species\u0026rsquo; climatic distribution was determined by overlaying its distribution range with the climate classification map. We initially assigned each distribution point to the one of 18 K\u0026ouml;ppen-Geiger climate classes following Beck et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e): arid-desert-cold (BWk), arid-steppe-cold (BSk), tempelate-dry summer-warm summer (Csb), temperate-dry winter-hot summer (Cwa), temperate-dry winter-warm summer (Cwb), temperate-without dry season-hot summer (Cfa), temperate-without dry season-warm summer (Cfb), temperate-without dry season-cold summer (Cfc), cold-dry summer-warm summer (Dsb), cold-dry summer-cold summer (Dsc), cold-dry winter-hot summer (Dwa), cold-dry winter-warm summer (Dwb), cold-dry winter-cold summer (Dwc), cold-dry winter-very cold summer, (Dwd), cold-without dry season-hot summer (Dfa), cold-without dry season-warm summer (Dfb), cold-without dry season-cold summer (Dfc), and Polar-tundra (ET). These classes were then grouped into nine broader climatic categories for the ancestral distributable climate classes reconstruction: BW (arid-desert), BS (arid-steppe), Cs (temperate-dry summer), Cw (temperate-dry winter), Cf (temperate-without dry season), Ds (cold-dry summer), Dw (cold-dry winter), Df (cold-without dry season), and ET (polar-tundra).\u003c/p\u003e\u003cp\u003eAncestral distributable climate classes reconstruction was conducted using the multistate reverse jump MCMC algorithm with model reduction in BayesTraits. We applied the HyperPrior All command to set a uniform prior between 0 and 10. Each analysis ran for 11\u0026nbsp;million iterations, with samples taken every 1,000 iterations. The output files from the two independent runs were pooled into a single combined file after removing the initial one million iterations were discarded as burn-in. Convergence and mixing of the chains were assessed by checking that ESS (\u0026gt;\u0026thinsp;200). Ancestral states at each node were summarized using a 95% credibility threshold. When multiple states together accounted for 95% of the PP, all such states were reported. For example, if the PPs for states A, B, C, and D were 0.60, 0.33, 0.05, and 0.02, at node X respectively, states A, B, and C were included as probable ancestral states for that node.\u003c/p\u003e\u003cp\u003eIn these analyses, we used a molecular phylogenetic tree reconstructed using MrBayes (see Section 3; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Although phylogenetic trees were also inferred using BEAST (see below), we selected the MrBayes tree for downstream analyses because it provided higher resolution of interspecific relationships within \u003cem\u003eChrysosplenium\u003c/em\u003e. Different substitution models were applied in each case: partition-specific models in MrBayes, and a single model for the combined partition in BEAST.\u003c/p\u003e\u003cp\u003eTo visualize the historical extent of distributable climates, we reconstructed paleoclimate maps. The classification of climate zones followed the K\u0026ouml;ppen-Geiger system as described by Peel et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) and Beck et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Paleoclimate data\u0026mdash;including monthly mean near-surface air temperature (\u0026deg;C) and monthly total precipitation (mm)\u0026mdash;from 81 to 11 Mya were obtained from Valdes et al. (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eEstimation of Species Colonization and climate preference in each clade\u003c/h3\u003e\n\u003cp\u003eTo investigate shifts in colonization rates across geographic regions and climate classes, we applied a rolling estimation method previously used by Xing et Ree (2017). This approach quantifies the relative contribution of each state (i.e., region or climate class) to the diversification of \u003cem\u003eChrysosplenium\u003c/em\u003e. Colonization rates through time were calculated using the formula d\u003csub\u003e\u003cem\u003eij\u003c/em\u003e\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e)\u0026thinsp;=\u0026thinsp;c\u003csub\u003e\u003cem\u003eij\u003c/em\u003e\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e)/n\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e (t\u0026thinsp;\u0026minus;\u0026thinsp;1), where c\u003csub\u003e\u003cem\u003eij\u003c/em\u003e\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e) represents the number of inferred colonization events from state \u003cem\u003ei\u003c/em\u003e to state \u003cem\u003ej\u003c/em\u003e at time \u003cem\u003et\u003c/em\u003e. For each node, the sum of colonization probabilities across states satisfies \u0026sum;c\u003csub\u003e\u003cem\u003eij\u003c/em\u003e\u003c/sub\u003e (\u003cem\u003et\u003c/em\u003e)\u0026thinsp;=\u0026thinsp;1. Based on ancestral state reconstructions using BioGeoBEARS (for geographic regions) and BayesTraits (for climate classes), colonization scores were assigned to each node, and the cumulative colonization score for each region and climate class was computed.\u003c/p\u003e\u003cp\u003eTo assess variation in climate preferences among clades, we performed a non-metric multidimensional scaling (nMDS) analysis using the metaMDS function from the R package vegan (Oksanen et al. 2025). This analysis was conducted using the Jaccard index on a presence/absence species\u0026ndash;climate matrix (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), with a maximum of 20 random starts and two dimensions. To compare climate class preferences between species in clades B and C, we conducted a permutational multivariate analysis of variance (PERMANOVA) using the adonis2 function and a permutational analysis of multivariate dispersion (PERMDISP) using the permutest function in vegan, both based on the Jaccard index with 5000 permutations. These analyses were restricted to 18 K\u0026ouml;ppen\u0026ndash;Geiger climate classes (Beck et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) for which distribution data of \u003cem\u003eChrysosplenium\u003c/em\u003e species were available. Comparisons involving clade A were not conducted due to its small sample size (n\u0026thinsp;=\u0026thinsp;2 species).\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003eMolecular phylogeny, divergence time estimation and temporal-based diversification analyses\u003c/h2\u003e\u003cp\u003eBayesian consensus phylogenetic analysis, using 10402 aligned nucleotide sites, revealed three major monophyletic clades (clades A\u0026ndash;C) with high posterior probabilities (PP\u0026thinsp;\u0026gt;\u0026thinsp;0.92; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Clade A consisted of \u003cem\u003eChrysosplenium microspermum\u003c/em\u003e and \u003cem\u003eC. sedakowii\u003c/em\u003e. Clades B and C included 25 and 30 species, respectively, and each was further divided into three well-supported subclades (B1\u0026ndash;3 and C1\u0026ndash;3; PPs\u0026thinsp;\u0026gt;\u0026thinsp;0.88; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Species in clades A and B exhibited alternate leaf arrangement (phyllotaxis), while those in clade C showed opposite phyllotaxis. Notably, \u003cem\u003eC. valdivicum\u003c/em\u003e, which occurs disjunctly in the Chile-Patagonian region, located a basal position within clade C. The ML analysis yielded a tree with a nearly identical topology (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eDivergence time estimates based on both younger and older calibrations (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e, Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e) suggest that the split between \u003cem\u003eChrysosplenium\u003c/em\u003e and \u003cem\u003ePeltoboykinia\u003c/em\u003e (node 1) occurred between 46.32\u0026ndash;43.20 Mya (95% highest posterior density [HPD]) under the younger calibration, and between 82.33\u0026ndash;79.20 Mya under the older calibration (95% HPD). The splitting of clade A from the remaining \u003cem\u003eChrysosplenium\u003c/em\u003e lineages (node 2) was estimated 43.91\u0026ndash;24.82 Mya (younger) and 79.90\u0026ndash;45.98 Mya (older). The split between clades B and C (node 4) occurred between 44.37\u0026ndash;22.72 Mya (younger) and 79.96\u0026ndash;41.86 Mya (older). Within these, the diversification of subclades B1\u0026ndash;3 (node 5 and 7) and C1\u0026ndash;3 (node 10 and 12) began after approximately 22 Mya and 31 Mya under the younger calibration, and after around 40 Mya and 56 Mya under the older calibration, respectively.\u003c/p\u003e\u003cp\u003eThe RevBayes analyses based on the BEAST tree calibrated with the younger constraint successfully converged, as evidenced by high effective sample sizes (ESS\u0026thinsp;=\u0026thinsp;5401 for the likelihood of the episodic diversification rate model and ESS\u0026thinsp;=\u0026thinsp;16351 for the branch-specific diversification rate model). Although no significant diversification rate shifts were detected at specific nodes within the phylogeny, the estimated net diversification remained stable at approximately 0.10 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Results from the analysis using the BEAST tree with the older calibration are not presented here, as our study primarily focuses on the outcomes derived from the younger calibration (see Discussion).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eAncestral states reconstruction\u003c/h3\u003e\n\u003cp\u003eThe ancestral region reconstruction conducted using BioGeoBEARS identified the DEC model as the best-fitting scenario, exhibiting the lowest AICc value (LnL\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;98.42, AICc\u0026thinsp;=\u0026thinsp;201.05, AICc weight\u0026thinsp;=\u0026thinsp;0.64; Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). Furthermore, the reconstruction of ancestral distributable climate classes performed with BayesTraits showed good convergence, as indicated by a high average effective sample size (mean ESS\u0026thinsp;=\u0026thinsp;3818.7).\u003c/p\u003e\u003cp\u003eThe results of ancestral state reconstructions (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e; Figs. S3 and S4; Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e) suggest that \u003cem\u003eChrysosplenium\u003c/em\u003e split from \u003cem\u003ePeltoboykinia\u003c/em\u003e in the Asian region, with a range of climate classes including BW, BS, Cw, Cf, Ds, Dw, Df, and ET (node 1 in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Among these, the Dw climate class had the highest posterior probability (PP\u0026thinsp;=\u0026thinsp;0.52) at this node. The crown node of \u003cem\u003eChrysosplenium\u003c/em\u003e that means as the splitting between clades A and the B\u0026thinsp;+\u0026thinsp;C lineage and splitting between clades B and C were inferred to be in the Asian region with DW climate class (nodes 2 and 4). The crown nodes of clades A, B (node as splitting subclade B1) and C (node as splitting subclade C1) were all inferred to be in the Asian region, primarily with the Dw climate class (nodes 3, 5 and 10). The splitting of subclade B2 and B3, as well as their respective crown nodes, were also situated in Asia with Dw as the dominant climate (nodes 7, 8 and 9). The split between clades C2 and C3 occurred in both the Asian and Chile\u0026ndash;Patagonian regions, with Dw climate class (node 12). The crown node of subclade C1 was located in Asia and most strongly associated with the Df as the prevailing climate class (node 11). The crown node of subclade C2 was inferred to be in Asia with Dw as the prevailing climate class (node 13), while the crown node of subclade C3 was reconstructed in both Asia and the Chile\u0026ndash;Patagonian regions, also with Dw-dominated conditions (node 14). A node within clade C3 excluding \u003cem\u003eC. valdivicum\u003c/em\u003e was inferred to have occurred in Asia with Dw-dominated climate class (node 15).\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eEstimation of species colonization and climate preference in each clade\u003c/h2\u003e\u003cp\u003eRolling estimates of regional colonization rates revealed that the Asian region served as the primary center of diversification for \u003cem\u003eChrysosplenium\u003c/em\u003e. However, several species and/or their ancestors subsequently diversified after dispersing from Asia to other regions, notably Europe, India, and North America (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Similarly, colonization rate estimates across distributable climate classes highlighted three major climatic hotspots: Cf (temperate-without dry season), Dw (cold-with dry winter), and Df (cold-without dry season). These climate classes acted as key sources for species expanding into other climate zones. Moreover, diversification events involving migrations among these three climate types also played an important role in the colonization history of \u003cem\u003eChrysosplenium\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe nMDS analysis yielded a stress value of 0.11, indicating an acceptable fit. Although overlaps in climate preferences were observed among clades, the analysis suggested some clade-specific trends: clade B tended to occur in Dfa (old-without dry season-hot summer) and Dfb (old-without dry season-warm summer) climates, while clade C was more often associated with Dwc (cold-without dry season-cold summer) and ET (Polar-tundra) climates (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Results of PERMANOVA and PERMDISP analyses showed that clades B and C differed significantly in climate preference, although the dispersion of those preferences was statistically homogeneous (PERMANOVA: F\u0026thinsp;=\u0026thinsp;2.39, P\u0026thinsp;=\u0026thinsp;0.024; PERMDISP: F\u0026thinsp;=\u0026thinsp;0.61, P\u0026thinsp;=\u0026thinsp;0.44).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003cb\u003eMolecular phylogeny of genus\u003c/b\u003e \u003cb\u003eChrysosplenium\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOur Bayesian phylogenetic analysis resolved three major clades within \u003cem\u003eChrysosplenium\u003c/em\u003e, each strongly supported by high PP. Most subclades within clades B and C also received strong support. Clade A and B were defined by species exhibiting alternate phyllotaxis, whereas clades C was associated with alternate and opposite phyllotaxis, respectively. These findings are broadly consistent with the previous phylogenetic work by Yang et al. (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, while the overall tree topology matched that earlier study, our analysis further resolved three distinct subclades within clade C\u0026mdash;substructure not recovered by Yang et al. (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAdditionally, our study placed \u003cem\u003eC. valdivicum\u003c/em\u003e within subclade C3, which differs from its placement in Yang et al. (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), where it was recovered as sister to a clade containing \u003cem\u003eC. alpinum\u003c/em\u003e, \u003cem\u003eC. oppositifolium\u003c/em\u003e, \u003cem\u003eC. dubium\u003c/em\u003e, \u003cem\u003eC. nepalense\u003c/em\u003e, \u003cem\u003eC. sinicum\u003c/em\u003e, and \u003cem\u003eC. grayanum\u003c/em\u003e. This discrepancy likely results from differences in the quantity and proportion of cpDNA and nrDNA data used. Specifically, our study analyzed approximately 4 kb of cpDNA and 6 kb of nuclear nrDNA, while Yang et al. (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) used about 154 kb of cpDNA and 7 kb of nrDNA. In our cpDNA-only phylogeny, \u003cem\u003eC. valdivicum\u003c/em\u003e was placed in a clade corresponding to subclade C1 (Fig. \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e), whereas the nrDNA-based phylogeny in Yang et al. (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) aligned \u003cem\u003eC. valdivicum\u003c/em\u003e with a group equivalent to our clade C3. Therefore, the phylogenetic position of \u003cem\u003eC. valdivicum\u003c/em\u003e remains uncertain. Future studies incorporating a larger and more diverse set of nuclear markers may help resolve this ambiguity and improve phylogenetic resolution across the genus. Moreover, our results suggest that the Japanese endemic species \u003cem\u003eC. macrostemon\u003c/em\u003e and \u003cem\u003eC. fauriei\u003c/em\u003e are conspecific, a finding consistent with Yang et al. (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This highlights the need for taxonomic revision of these species.\u003c/p\u003e\u003cp\u003e\u003cb\u003eAdaptive timeline to the evolutionary history of\u003c/b\u003e \u003cb\u003eChrysosplenium\u003c/b\u003e\u003c/p\u003e\u003cp\u003eIn this study, we conducted two divergence time estimation analyses using alternative calibration ages representing younger and older scenarios (Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). Based on ancestral state reconstruction, the timeline inferred from the younger calibration appears to align better with paleoclimate reconstructions. For instance, the ancestral state of node 11 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) corresponds more closely with the environmental conditions shown in the 21 Mya paleoclimate map than those in the 40 Mya map (Fig. S6). As we employed a secondary calibration approach\u0026mdash;using calibration points derived from previous molecular dating studies\u0026mdash;it is important to acknowledge the potential for bias toward younger divergence estimates (Schenk \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Consequently, the results must be interpreted cautiously, with an awareness of the limitations inherent in such methods. Based on this suggestion, across multiple comparisons between estimated node ages and paleoclimate maps of corresponding or slightly earlier ages, the divergence times inferred from the younger calibration points generally showed better congruence than those from the older calibration. This trend supports the suitability of the younger timeline for interpreting the evolutionary history of \u003cem\u003eChrysosplenium\u003c/em\u003e in the context of past climatic changes.\u003c/p\u003e\u003cp\u003eSecond, when we focus on the node representing the split between \u003cem\u003eC. kitoense\u003c/em\u003e and its sister clade\u0026mdash;composed mainly of Japanese endemic species located in clade C3\u0026mdash;the estimated divergence times under the younger and older calibration scenarios were 12.17 Mya (95% HPD: 17.77\u0026ndash;6.97 Mya) and 22.07 Mya (32.12\u0026ndash;12.87 Mya in 95% HPD), respectively. Additional ancestral region reconstruction using BioGeoBEARS, in which Japan was treated as a separate region from continental Asia, indicated that the ancestral area for this node was Japan (Fig. S6). Geological evidence suggests that the Japanese archipelago began to separate from the Asian continent around 20 Mya (Isozaki et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). If \u003cem\u003eC. kitoense\u003c/em\u003e, which is distributed in Cf (temperate-without dry season) and Df (cold-without dry season) climate zones, had diverged at 22 Mya as suggested by the older calibration, it is likely that it would have also been distributed on the continent, where such climatic zones were present at that time (Fig. S7). This biogeographic mismatch suggests that the divergence estimate based on the younger calibration is more consistent with geological and climatic evidence, supporting its suitability for interpreting the evolutionary history of \u003cem\u003eChrysosplenium\u003c/em\u003e.\u003c/p\u003e\u003cp\u003ePrevious studies estimated the origin of \u003cem\u003eChrysosplenium\u003c/em\u003e at approximately 25 Mya or 45 Mya (Deng et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Folk et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), with the latter being more consistent with our results. The differences in these estimates primarily stem from the calibration strategies employed. Both studies used \u003cem\u003eItea\u003c/em\u003e (Iteaceae) as an outgroup, calibrating the stem node of Saxifragaceae (including \u003cem\u003eChrysosplenium\u003c/em\u003e and Iteaceae) using fossil-based ages of 48 Mya (Deng et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) and 89 Mya (Folk et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In our study, the younger calibration scenario (Zunteni et al. 2024) is largely consistent with that of Folk et al. (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), which explains the similarity in divergence time estimates.\u003c/p\u003e\u003cp\u003e\u003cb\u003eThe origin of\u003c/b\u003e \u003cb\u003eChrysosplenium\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOur findings shed light on the dispersal and diversification history of \u003cem\u003eChrysosplenium\u003c/em\u003e, revealing that its diversification was initially stable and concentrated in Asia, followed by extensive geographic expansion primarily into the Palearctic and Nearctic regions. Furthermore, our results support a novel hypothesis that \u003cem\u003eChrysosplenium\u003c/em\u003e species predominantly diversified within Cf (temperate-without a dry season), Dw (cold-with dry winter), and Df (cold-without a dry season) climate classes.\u003c/p\u003e\u003cp\u003eThe results of our ancestral region reconstruction analyses suggest that \u003cem\u003eChrysosplenium\u003c/em\u003e originated in Asia. This finding is consistent with previous studies (Deng et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Nakazawa et al. 1957; Soltis et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), which have similarly proposed East Asia as the center of origin and diversification for the genus. Furthermore, the subfamily Micrantheae\u0026mdash;comprising \u003cem\u003eMicranthes\u003c/em\u003e and sister to Chrysosplenieae (which includes \u003cem\u003eChrysosplenium\u003c/em\u003e and \u003cem\u003ePeltoboykinia\u003c/em\u003e)\u0026mdash;is also thought to have originated in Asia (Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). For example, ancestral species of \u003cem\u003eMicranthes\u003c/em\u003e (e.g., \u003cem\u003eM. merkii\u003c/em\u003e in western Beringia, \u003cem\u003eM. nudicaulis\u003c/em\u003e, and \u003cem\u003eM. tolmiei\u003c/em\u003e distributed on both sides of Beringia) exhibit distribution patterns consistent with an Asian origin (PWPO, 2025; Stubbs et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Additionally, the earliest-diverging subfamily Darmereae\u0026mdash;which includes genera such as \u003cem\u003eAstilboides\u003c/em\u003e, \u003cem\u003eBergenia\u003c/em\u003e, \u003cem\u003eDarmera\u003c/em\u003e, \u003cem\u003eMukdenia\u003c/em\u003e, \u003cem\u003eOresitrophe\u003c/em\u003e, and \u003cem\u003eRodgersia\u003c/em\u003e\u0026mdash;is sister to the clade comprising Chrysosplenieae and Micrantheae (Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Most species in Darmereae are distributed in northeastern China, with \u003cem\u003eDarmera\u003c/em\u003e being the notable exception, occurring in western North America (PWPO 2025). Our additional ancestral region reconstruction analysis identified the splitting node between \u003cem\u003eChrysosplenium\u003c/em\u003e and \u003cem\u003ePeltoboykinia\u003c/em\u003e in the region of Asia and Japan, further supporting a northeastern Asian origin for \u003cem\u003eChrysosplenium\u003c/em\u003e (Fig. S6). Moreover, distributable climate class reconstructions indicated that the Dw (cold-dry winter) climate zone inferred from ancestral state reconstructions was primarily located in northeastern Asia (Fig. S7). Together, these lines of evidence reinforce the hypothesis that \u003cem\u003eChrysosplenium\u003c/em\u003e originated in northeastern Asia. Following its origin, \u003cem\u003eChrysosplenium\u003c/em\u003e diversified primarily in the Northern Hemisphere, particularly within subclades that first radiated in northeastern Asia under the Dw climate class.\u003c/p\u003e\u003cp\u003e\u003cb\u003eDispersal and diversification scenario of\u003c/b\u003e \u003cb\u003eChrysosplenium\u003c/b\u003e\u003c/p\u003e\u003cp\u003eMost diversification likely occurred through dispersal within Asia, particularly in China. However, several species\u0026mdash;\u003cem\u003eC. carnosum\u003c/em\u003e, \u003cem\u003eC. forrestii\u003c/em\u003e, and \u003cem\u003eC. lanuginosum\u003c/em\u003e in subclade B3; \u003cem\u003eC. nepalense\u003c/em\u003e in C1; and \u003cem\u003eC. delavayi\u003c/em\u003e in C2\u0026mdash;appear to have speciated following dispersal into southern China, where the Cw (temperate-with dry winters) climate class. In addition, \u003cem\u003eC. iowense\u003c/em\u003e distributed in the North America shown in subclade B1, \u003cem\u003eC. alpinum\u003c/em\u003e distributed in the Europe, and \u003cem\u003eC. dubium\u003c/em\u003e distributed in the western Asia, Europe, and Saharo-Arabia in subclade C1 likely diversified after long-distance dispersal events to their respective regions. \u003cem\u003eC. alternifolium\u003c/em\u003e in subclade B1 and \u003cem\u003eC. oppositifolium\u003c/em\u003e in subclade C1 seem to have expanded westward through areas characterized by Df (cold-without a dry season) climate class. Conversely, \u003cem\u003eC. rosendahlii\u003c/em\u003e and \u003cem\u003eC. wrightii\u003c/em\u003e in subclade B1 and \u003cem\u003eC. americanum\u003c/em\u003e in C1 likely spread eastward under the same Df climatic conditions, crossing the Bering land bridge during the Neogene or Quaternary periods\u0026mdash;a route known to facilitate floristic exchange from East Asia to western North America (Wen et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). \u003cem\u003eC. tetrandrum\u003c/em\u003e is notable for its bidirectional expansion, both eastward and westward, under the Df climate regime. Several species endemic to Japan, including \u003cem\u003eC. maximowiczii\u003c/em\u003e, \u003cem\u003eC. tosaense\u003c/em\u003e, and the ancestral lineages of \u003cem\u003eC. album\u003c/em\u003e and \u003cem\u003eC. rhabdospermum\u003c/em\u003e, likely originated from continental ancestors and underwent allopatric speciation after migrating to the Japanese archipelago. Additionally, species such as \u003cem\u003eC. kamtschatium\u003c/em\u003e located in clade C3 found within the same clade as \u003cem\u003eC. kitoense\u003c/em\u003e, appear to have migrated into the Sakhalin and the Kamchatka Peninsula following diversified in Japan (Fig. S6).\u003c/p\u003e\u003cp\u003eAccording to the results of ancestral distributable climate reconstruction, along with nMDS, PERMANOVA, and PERMDISP analyses, both ancestral and extant species of \u003cem\u003eChrysosplenium\u003c/em\u003e could be distributed across most of the climate classes where current species are found (see node 1 in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig. \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). Donoghue (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) suggested that ecological traits influencing species distributions tend to evolve gradually. For instance, the transition from tropical climates to conditions involving freezing temperatures and high seasonality is considered a slow evolutionary process. Therefore, certain traits enabling plant dispersal in response to climate change may have evolved early in the lineage (Donoghue \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Our findings possibly suggest that \u003cem\u003eChrysosplenium\u003c/em\u003e species had already adapted to a wide range of climatic conditions and this characteristic leads stable diversification in mainly Cf (temperate-without a dry season), Dw (cold-with dry winter), and Df (cold-without a dry season) climate classes within genus \u003cem\u003eChrysosplenium\u003c/em\u003e. Both China and Japan are known for having many endemic species of \u003cem\u003eChrysosplenium\u003c/em\u003e. Although our study does not fully resolve the reasons for the high endemism in these regions, the Japanese endemic species likely originated via allopatric speciation in mountainous regions (Kubota et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), a pattern that may also apply to China. Thus, many endemic \u003cem\u003eChrysosplenium\u003c/em\u003e species may have diversified through habitat fragmentation and isolation. Additionally, members of \u003cem\u003eChrysosplenieae\u003c/em\u003e are considered habitat specialists of moist woodlands and are commonly found in mid-elevation mountain zones (Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Such environments may have promoted allopatric conditions more effectively than open, lowland habitats like grasslands.\u003c/p\u003e\u003cp\u003eIn this study, we were unable to determine the precise process underlying the disjunct distribution of \u003cem\u003eC. valdivicum\u003c/em\u003e. Ancestral state reconstruction suggests that this species likely migrated from Asia to the Chile-Patagonian region under Cf (temperate-without a dry season) and Dw (cold-dry winter) climate classes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). However, such a direct, long-distance dispersal from Asia to South America seems unlikely. Although this distribution pattern represents an amphitropical disjunction in the Americas, most comparable plant migrations into South America are believed to have originated from North America (Simpson et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Liu et al. (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) defined both Asia and North America as parts of the Holarctic region, suggesting that \u003cem\u003eC. valdivicum\u003c/em\u003e or its ancestral species may have first spread into North America\u0026mdash;potentially via the Bering land bridge\u0026mdash;before reaching South America under Df (cold-without dry season) climate class or similar conditions. Interestingly, two additional \u003cem\u003eChrysosplenium\u003c/em\u003e species are found in South America: \u003cem\u003eC. aulacocarpum\u003c/em\u003e in Colombia and Venezuela, and \u003cem\u003eC. macranthum\u003c/em\u003e, which, like \u003cem\u003eC. valdivicum\u003c/em\u003e, exhibits a disjunct distribution between Argentina and Chile. Further investigation of these species could provide deeper insights into the biogeographic history and dispersal routes of \u003cem\u003eChrysosplenium\u003c/em\u003e in South America.\u003c/p\u003e\u003cp\u003eThis study suggests that plant species have historically migrated into suitable habitats and undergone diversification in response to past climate changes. The ancestral climate reconstruction indicates that \u003cem\u003eChrysosplenium\u003c/em\u003e species had already begun adapting to Cf (temperate-without dry season), Dw (cold-dry winter) and Df (cold-without dry season) climate classes since their early evolutionary history. However, a key question remains: why are most species still confined to limited geographic areas despite the existence of climatically suitable habitats elsewhere? To address this, further research is needed to uncover the factors shaping plant distribution. Physiological studies may offer valuable insights into this issue. Previous findings show that \u003cem\u003eChrysosplenium\u003c/em\u003e species occupy a wide range of environmental conditions, such as variation in mean annual precipitation (Folk et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), suggesting that they differ in physiological traits related to water use. These traits may play a crucial role in determining current distribution patterns and the speciation processes within the genus.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompeting Interests:\u003c/strong\u003e\u003cp\u003eNone\u003c/p\u003e\u003ch2\u003eFunding:\u003c/h2\u003e\u003cp\u003eThis research was financially supported by the JSPS KAKENHI (23K14240).\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eWe would like to express our gratitude to Dr. Kenji Izumi (Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences) for his contribution to the reconstruction of paleoclimate maps in this study. This research was financially supported by the JSPS KAKENHI (23K14240). This study was conducted using an existing dataset; therefore, no fieldwork requiring specific permissions was conducted, and no permits were necessary. We sincerely thank the researchers of previous studies for providing the datasets used in this research.\u003c/p\u003e\u003ch2\u003eData availability\u003c/h2\u003e\u003cp\u003eThe data utilized in this study are accessible through GenBank (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/genbank/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/genbank/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and the Plants of the World Online website (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.plantsoftheworldonline.org/\u003c/span\u003e\u003cspan address=\"http://www.plantsoftheworldonline.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and the World Flora Online Plant List (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://wfoplantlist.org\u003c/span\u003e\u003cspan address=\"https://wfoplantlist.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBeck HE, McVicar TR, Vergopolan N, Berg A, Lutsko JL, Dufour A, Zeng Z, Jiang X, van Dijk AIJM, Miralles DG (2023) High-resolution (1 km) K\u0026ouml;ppen-Geiger maps for 1901\u0026ndash;2099 based on constrained CMIP6 projections. Sci Data 10:724. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41597-023-02549-6\u003c/span\u003e\u003cspan address=\"10.1038/s41597-023-02549-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBouckaert R, Heled J, K\u0026uuml;hnert D, Vaughan T, Wu CH, Xie D, Suchard MA, Rambaut A, Drummond AJ (2014) BEAST 2: a software platform for Bayesian evolutionary analysis. PLoS Comput Biol 10:e1003537. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pcbi.1003537\u003c/span\u003e\u003cspan address=\"10.1371/journal.pcbi.1003537\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeng JB, Drew BT, Mavrodiev EV, Gitzendanner MA, Soltis PS, Soltis DE (2015) Phylogeny, divergence times, and historical biogeography of the angiosperm family Saxifragaceae. Mol Phyl Evol 83:86\u0026ndash;98. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ympev.2014.11.011\u003c/span\u003e\u003cspan address=\"10.1016/j.ympev.2014.11.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDonoghue MJ (2008) A phylogenetic perspective on the distribution of plant diversity. PNAS 105:S11549\u0026ndash;S11555. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.0801962105\u003c/span\u003e\u003cspan address=\"10.1073/pnas.0801962105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFolk RA, Stubbs RL, Mort ME, Cellinese N, Allen JM, Soltis PS, Soltis DE, Guralnick RP (2019) Rates of niche and phenotype evolution lag behind diversification in a temperate radiation. PNAS 116:10874\u0026ndash;10882. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.1817999116\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1817999116\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFolk RA, Stubbs RL, Engle-Wrye NJ, Soltis DE, Okuyama Y (2021) Biogeography and habitat evolution of Saxifragaceae, with a revision of generic limits and a new tribal system. Taxon 70:263\u0026ndash;285. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/tax.12450\u003c/span\u003e\u003cspan address=\"10.1002/tax.12450\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHara H (1957) Synopsis of the genus \u003cem\u003eChrysosplenium\u003c/em\u003e L. J Fac Sci Univ Tokyo Sect III Bot 7:1\u0026ndash;90\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eH\u0026ouml;hna S, Landis MJ, Heath TA, Boussau B, Lartillot N, Moore BR, Huelsenbeck JP, Ronquist F (2016) RevBayes: Bayesian Phylogenetic Inference Using Graphical Models and an Interactive Model\u0026ndash;Specification Language. Syst Biol 65:726\u0026ndash;736. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/sysbio/syw021\u003c/span\u003e\u003cspan address=\"10.1093/sysbio/syw021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIsaka Y, Nakamura A, Izumi K, Schneider H (2025) Molecular phylogenetic analyses reveal the diversification and dispersal processes of the tribe Shoreae (Dipterocarpaceae). J Biogeo e15166. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/jbi.15166\u003c/span\u003e\u003cspan address=\"10.1111/jbi.15166\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIsozaki Y, Aoki K, Nakama T, Yanai S (2010) New insight into a subduction-related orogen: Reappraisal on geotectonic framework and evolution of the Japanese Islands. Gondwana Res 18:82\u0026ndash;105. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.gr.2010.02.015\u003c/span\u003e\u003cspan address=\"10.1016/j.gr.2010.02.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKozlov AM, Darriba D, Flouri T, Morel B, Stamatakis A (2019) RAxML-NG: a fast, scalable and user-friendly tool for maximum likelihood phylogenetic inference. Bioinformatics 35:4453\u0026ndash;4455. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/btz305\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btz305\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKubota Y, Shiono T, Kusumoto B (2015) Role of climate and geohistorical factors in driving plant richness patterns and endemicity on the east Asian continental islands. Ecography 38:639\u0026ndash;648. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/ecog.00981\u003c/span\u003e\u003cspan address=\"10.1111/ecog.00981\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKuraku S, Zmasek CM, Nishimura O, Katoh K (2013) aLeaves facilitates on-demand exploration of metazoan gene family trees on MAFFT sequence alignment server with enhanced interactivity. Nucl Acids Res 41:W22\u0026ndash;W28. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkt389\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkt389\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLanfear R, Frandsen PB, Wright AM, Senfeld T, Calcott B (2017) PartitionFinder 2: new methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Mol Biol Evol 34:772\u0026ndash;773. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/molbev/msw260\u003c/span\u003e\u003cspan address=\"10.1093/molbev/msw260\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu Y, Xu X, Dimitrov D, Pellissier L, Borregaard MK, Shrestha N, Su X, Luo A, Zimmermann NE, Rahbek C, Wang Z (2023) An updated floristic map of the world. Nat Commun 14:2990. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-023-38375-y\u003c/span\u003e\u003cspan address=\"10.1038/s41467-023-38375-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatzke NJ (2013) Probabilistic historical biogeography: new models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing. Dissertation, University of California\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNakazawa M, Wakabayashi M, Ono M, Murata J (1997) Molecular phylogenetic analysis of \u003cem\u003eChrysosplenium\u003c/em\u003e (Saxifragaceae) in Japan. J Plant Res 110:265\u0026ndash;274. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/BF02509315\u003c/span\u003e\u003cspan address=\"10.1007/BF02509315\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOksanen J, Blanchet FG, Kindt R et al (2011) vegan: community ecology package. R package version 2.6\u0026ndash;6.1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/vegandevs/vegan\u003c/span\u003e\u003cspan address=\"https://github.com/vegandevs/vegan\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePagel M, Meade A, Barker D (2004) Bayesian estimation of ancestral character states on phylogenies. Syst Biol 53:673\u0026ndash;684. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/10635150490522232\u003c/span\u003e\u003cspan address=\"10.1080/10635150490522232\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeel MC, Finlayson BL, McMahon TA (2007) Updated World Map of the K\u0026ouml;ppen-Geiger Climate Classification. Hydrol Earth Syst Sci 11:1633\u0026ndash;1644. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5194/hess-11-1633-2007\u003c/span\u003e\u003cspan address=\"10.5194/hess-11-1633-2007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePOWO (2024) Plants of the World Online. Facilitated by the Royal Botanic Gardens, Kew. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://powo.science.kew.org/\u003c/span\u003e\u003cspan address=\"https://powo.science.kew.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. accessed 20 April 2025\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRambaut A (2018) ) FigTree v.1.4.4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/rambaut/Figtree/releases/tag/v1.4.4\u003c/span\u003e\u003cspan address=\"https://github.com/rambaut/Figtree/releases/tag/v1.4.4\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRambaut A, Drummond AJ, Xie D, Baele G, Suchard MA (2018) Posterior summarization in Bayesian phylogenetics using Tracer 1.7. Syst Biol 67:901\u0026ndash;904. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/sysbio/syy032\u003c/span\u003e\u003cspan address=\"10.1093/sysbio/syy032\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eR Core Team (2024) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing v.4.3.2. Vienna, Austria: R foundation for Statistical Computing. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.R-project.org/\u003c/span\u003e\u003cspan address=\"https://www.R-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRonquist F, Teslenko M, van Der Mark P, Ayres DL, Darling A, H\u0026ouml;hna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Syst Boil 61:539\u0026ndash;542. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/sysbio/sys029\u003c/span\u003e\u003cspan address=\"10.1093/sysbio/sys029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchenk JJ (2016) Consequences of secondary calibrations on divergence time estimates. PLoS ONE 11:e0148228. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0148228\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0148228\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSimpson MG, Johnson LA, Villaverde T, Guilliams CM (2017) American amphitropical disjuncts: perspectives from vascular plant analyses and prospects for future research. Ame J Bot 104:1600\u0026ndash;1650. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3732/ajb.1700308\u003c/span\u003e\u003cspan address=\"10.3732/ajb.1700308\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoltis DE, Tago-Nakazawa M, Xiang QY, Kawano S, Murata J, Wakabayashi M, Hibsch‐Jetter C (2001) Phylogenetic relationships and evolution in \u003cem\u003eChrysosplenium\u003c/em\u003e (Saxifragaceae) based on \u003cem\u003ematK\u003c/em\u003e sequence data. Ame J Bot 88:883\u0026ndash;893. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2307/2657040\u003c/span\u003e\u003cspan address=\"10.2307/2657040\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStubbs RL, Folk RA, Xiang CL, Chen S, Soltis DE, Cellinese N (2020) A phylogenomic perspective on evolution and discordance in the alpine-arctic plant clade Micranthes (Saxifragaceae). Front Plant Sci 10:1773. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpls.2019.01773\u003c/span\u003e\u003cspan address=\"10.3389/fpls.2019.01773\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTribble CM, Freyman WA, Landis MJ, Lim JY, Barido-Sottani J, Kopperud BT, H\u0026ouml;hna S, May MR (2022) RevGadgets: an R package for visualizing Bayesian phylogenetic analyses from RevBayes. Met Ecol Evol 13:314\u0026ndash;323. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/2041-210X.13750\u003c/span\u003e\u003cspan address=\"10.1111/2041-210X.13750\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eValdes PJ, Scotese CR, Lunt DJ (2021) Deep Ocean Temperatures Through Time. Clim Past 17:1483\u0026ndash;1506. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5194/cp-17-1483-2021\u003c/span\u003e\u003cspan address=\"10.5194/cp-17-1483-2021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWen J, Nie ZL, Ickert-Bond SM (2016) Intercontinental disjunctions between eastern Asia and western North America in vascular plants highlight the biogeographic importance of the Bering land bridge from late Cretaceous to Neogene. J Syst Evol 54:469\u0026ndash;490. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/jse.12222\u003c/span\u003e\u003cspan address=\"10.1111/jse.12222\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWen J, Xie DF, Price M, Ren T, Deng YQ, Gui LJ, Guo XL, He XJ (2021) Backbone phylogeny and evolution of Apioideae (Apiaceae): New insights from phylogenomic analyses of plastome data. Mol Phyl Evol 161:107183. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ympev.2021.107183\u003c/span\u003e\u003cspan address=\"10.1016/j.ympev.2021.107183\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWFO (2024) The World Flora Online. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://worldfloraonline.org\u003c/span\u003e\u003cspan address=\"https://worldfloraonline.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 20 April 2025\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXing Y, Ree RH (2017) Uplift-driven diversification in the Hengduan Mountains, a temperate biodiversity hotspot. PNAS 114:E3444\u0026ndash;E3451. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1073/pnas.1616063114\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1616063114\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang T, Wu Z, Tie J, Qin R, Wang J, Liu H (2023) A comprehensive analysis of chloroplast genome provides new insights into the evolution of the genus \u003cem\u003eChrysosplenium\u003c/em\u003e. Int J Mol Sci 24:14735. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/ijms241914735\u003c/span\u003e\u003cspan address=\"10.3390/ijms241914735\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYule GU, Willis JC, Philos FRS (1925) Trans R Soc Lond 213: 21\u0026ndash;87. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1098/rstb.1925.0002\u003c/span\u003e\u003cspan address=\"10.1098/rstb.1925.0002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZuntini AR, Carruthers T, Maurin O et al (2024) Phylogenomics and the rise of the angiosperms. Nature 629:843\u0026ndash;850. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41586-024-07324-0\u003c/span\u003e\u003cspan address=\"10.1038/s41586-024-07324-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"ancestral trait reconstruction, Asia, distributable climate, divergence time estimation, paleoclimate","lastPublishedDoi":"10.21203/rs.3.rs-7444744/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7444744/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe diversification of plant lineages is often governed by intricate interactions among geography, climate, and lineage-specific traits. The genus \u003cem\u003eChrysosplenium\u003c/em\u003e (Saxifragaceae), a herbaceous group with a disjunct distribution across the Northern Hemisphere and parts of South America, serves as an excellent model for exploring the mechanisms underlying plant diversification and historical biogeography. In this study, we reconstructed a comprehensive phylogeny of \u003cem\u003eChrysosplenium\u003c/em\u003e by integrating plastid and nuclear DNA markers. We also estimated divergence times, inferred ancestral geographic regions and distributable climate classes, and quantified colonization rates among ancestral regions and distributable climate classes to better understand the processes shaping the genus\u0026rsquo;s present-day distribution and regional diversification hotspots. Our analyses suggest that \u003cem\u003eChrysosplenium\u003c/em\u003e originated in the Eocene within cold climate zones of East Asia, from which it subsequently dispersed into Europe, North America, and southern parts of Asia. The genus underwent episodic diversification likely driven by climatic fluctuations that promoted both range expansion and lineage divergence. Ancestral distributable climate classes reconstructions indicate that early species had already adapted to a spectrum of cold and temperate conditions. However, despite the existence of climatically suitable areas beyond their current ranges, most species remain geographically localized. This pattern implies that factors beyond macroclimatic suitability\u0026mdash;such as physiological constraints like water-use efficiency and habitat specialization\u0026mdash;may restrict range expansion and foster allopatric speciation. Overall, our findings highlight the value of combining phylogenetic, biogeographic, and ecological perspectives to uncover the evolutionary processes shaping plant diversity.\u003c/p\u003e","manuscriptTitle":"Evolutionary history of Chrysosplenium L. (Saxifragaceae): a molecular phylogenetic approach to dispersal and diversification processes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-25 15:33:32","doi":"10.21203/rs.3.rs-7444744/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"30f0efb7-b0f9-4d07-80b4-34aecbfd0575","owner":[],"postedDate":"September 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-22T06:42:56+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-25 15:33:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7444744","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7444744","identity":"rs-7444744","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}