bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1 Towards a global perspective for Salvia L: Phylogeny, diversification, and floral evolution 2 Fatemeh Moein*1, Ziba Jamzad2, Mohammadreza Rahiminejad1, Jacob B. Landis4,5,Mansour 3 Mirtadzadini3, Douglas E. Soltis*6,7,8,9 and Pamela S. Soltis7,8,9 4 1 Department of Biology, Faculty of Science, University of Isfahan, Iran; 5 2 Department of Botany, Research Institute of Forest and Rangelands, Tehran, Iran; 6 3 Department of Biology, Faculty of Science, Shahid Bahonar Univ., PO Box 76169‐133, 7 Kerman, Iran 8 4 9 Cornell University, Ithaca, NY 14583, USA School of Integrative Plant Science, Section of Plant Biology and the L.H. Bailey Hortorium, 10 5 BTI Computational Biology Center, Boyce Thompson Institute, Ithaca, NY 14853, USA 11 6 Department of Biology, University of Florida, Gainesville, FL 32611, USA 12 7 Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611, USA 13 8 The Genetics Institute, University of Florida, Gainesville, Florida 32610, USA 14 9 The Biodiversity Institute, University of Florida, Gainesville, Florida 32611, USA 15 16 *Author for correspondence 17 18 1 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 19 Abstract 20 Premise of this study: Salvia is the most species-rich genus in Lamiaceae, encompassing 21 approximately 1000 species distributed all over the world. We sought a new evolutionary 22 perspective for Salvia by employing macroevolutionary analyses to address the tempo and 23 mode of diversification. To study the association of floral traits with speciation and extinction, 24 we modeled and explored the evolution of corolla length and the lever-mechanism pollination 25 system across our Salvia phylogeny. 26 Methods: We reconstructed a multigene phylogeny for 366 species of Salvia in the broad 27 sense including all major recognized lineages and numerous species from Iran, a region 28 previously overlooked in studies of the genus. Our phylogenetic data in combination with 29 divergence time estimates were used to examine the evolution of corolla length, woody vs. 30 herbaceous habit, and presence vs. absence of a lever mechanism. We investigated the timing 31 and dependence of Salvia diversification related to corolla length evolution through a 32 disparity test and BAMM analysis. A HiSSE model was used to evaluate the dependency of 33 diversification on the lever-mechanism pollination system in Salvia. 34 Key Results: Based on recent investigations and classifications, Salvia is monophyletic and 35 comprises ~1000 species. Our inclusion, for the first time, of a comprehensive sampling for 36 Iranian species of Salvia provides higher phylogenetic resolution for southwestern Asian 37 species than obtained in previous studies. A medium corolla length (15-18mm) was 38 reconstructed as the ancestral state for Salvia with multiple shifts to shorter and longer 39 corollas. Macroevolutionary model analyses indicate that corolla length disparity is high 40 throughout Salvia evolution, significantly different from expectations under a Brownian 41 motion model during the last 28 million years of evolution. Our analyses show evidence of a 42 higher diversification rate of corolla length for some Andean species of Salvia compared to 43 other members of the genus. Based on our tests of diversification models, we reject the 44 hypothesis of a direct effect of the lever mechanism on Salvia diversification. 45 Conclusions: Using a broader species sampling than previous studies, we obtained a well- 46 resolved phylogeny for southwest Asian species of Salvia. Corolla length is an adaptive trait 47 throughout the Salvia phylogeny with a higher rate of diversification in the South American 2 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 48 clade. Our results suggest caution in considering the lever-mechanism pollination system as 49 one of the main drivers of speciation in Salvia. 50 Key words: Salvia, phylogeny, diversification, corolla, pollination, lever mechanism 51 1. Introduction 52 Integrating molecular data with organismal traits can be used to address a major question in 53 biology, “Is higher species diversity related to the presence of specific traits in that lineage?” 54 (Pyron and Tubrin, 2014). Recently developed model-based approaches for estimating 55 divergence times (BEAST: Drummond and Rambaut 2007; treePL: Smith and O’Meara 2010), 56 diversification rates (MEDUSA: Alfaro et al., 2009; BAMM: Rabosky et al., 2014), and the effect 57 of traits on diversification (FitzJohn et al., 2012; Beaulieu and O’Meara 2016; Caetano et al., 58 2018; Landis et al., 2018; Han et al., 2020) provide new opportunities to address this question. 59 These methods have the advantage of providing estimates of the origin, divergence time, rate 60 of diversification, and drivers of diversification among species. 61 There has been considerable recent interest in studying the association of floral traits and 62 species richness in flowering plants (Vamosi et al., 2011; Van der Niet and Johnson 2014; Soltis 63 & Soltis 2014; Saquet et al., 2017; Landis et al., 2018; Onstein 2019; Hernández and Wiens 64 2020). Interactions between flowers and their pollinators have spurred speciation and the 65 evolution of novel floral variation (e.g., Stebbins 1970; Dodd et al., 1999; Crane et al., 1995; 66 Crepet 2000; Soltis and Soltis 2004; Soltis et al., 2008; Ambruster 2014; Fenster et al., 2004; 67 Smith 2010; Van der Niet and Johnson 2014). Some floral traits such as spur length, corolla 68 shape, corolla length, and number of flowers are more often influenced by selection than 69 other floral features (Yoshioka, 2007; Kacrowski et al., 2012; Landis et al., 2016). Floral 70 specialization could potentially promote diversification by the evolution of adaptive floral 71 traits through the establishment of reproductive isolation (Kay and Sargent 2009, Armbuster 72 2014; Serrano-Serrano et al., 2015). Several studies have also shown a correlation between 73 flower specialization and rate of diversification (Fernández-Mazuecos et al., 2013; Ogutcen et 74 al., 2014; Lagomarsino et al., 2016). For example, in genera of Neotropical Gesneriaceae 75 including Codonanthopsis Mansf, Codonanthe (Mart.) Hanst, and Nematanthus Schard, 76 species with hummingbird pollination syndromes have higher rates of diversification than 77 close relatives pollinated by insects (Serrano-Serrano et al., 2015). 3 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 78 Lamiaceae (the mints) are the sixth largest family of flowering plants with over 7000 species 79 distributed worldwide (Harley et al., 2004). Recently, Li et al. (2016, 2017) subdivided 80 Lamiaceae into ten subfamilies and four unplaced genera based on a large-scale, plastid- 81 based phylogenetic analysis, and this topology was largely corroborated by analysis of nuclear 82 transcriptomes (Mint Evolutionary Genomics Consortium 2018). Within Lamiaceae, 83 Nepetoideae is the largest subfamily with 105 genera and 3600 species, including well-known 84 genera such as Thymus L. (thyme), Ocimum L. (basil), Nepeta L. (catnip), Salvia L. (sage), and 85 Lavandula L. (lavender) (Harley et al., 2004). 86 Salvia, the largest genus in Lamiaceae as currently defined, includes approximately 1000 87 species, more than half of which are distributed in North and South America (Alziar, 1988- 88 1993). Morphologically, Salvia is highly diverse, particularly regarding specialized floral traits 89 such as corolla color, corolla and tube length, flower shape and stamen structure (Wester and 90 Claßen-Bockhoff, 2007; Reith et al., 2007; Will and Claßen-Bockhoff, 2015). Traditionally, 91 Salvia was separated from other genera in Lamiaceae by possessing two fertile stamens with 92 an elongated connective tissue. More that 80% of Salvia species are characterized by a special 93 pollination system referred to as a lever mechanism (Walker et al., 2004; Harely et al., 2004; 94 Claßen-Bockhoff et al., 2004; Walker and Sytsma, 2007). transfer to the stigma. The lever 95 mechanism has the advantage of promoting successful pollination. In addition, this approach 96 is efficient in pollen allocation and does not allow the pollinator to collect all of the pollen in 97 one visit (Claßen-Bockhoff et al., 2003; Reith et al., 2007; Celep et al., 2014). A staminal lever 98 is an advantage in Salvia due to the precise placement of pollen on bees while they are 99 accessing the restricted nectar (Claßen-Bockhoff et al., 2004; Zhang et al. (2011) showed that 100 removing the lever arms in Salvia cyclostegia resulted in lower fruit and seed set. Previous 101 morphological studies of Salvia pollinators and floral traits hypothesized that the lever 102 mechanism might play a role as a key innovation in promoting adaptive radiations (Claßen- 103 Bockhoff et al., 2004; Will and Claßen-Bockhoff 2014). Based on phylogenetic results, the 104 lever mechanism evolved in parallel in the Eastern and Western Hemispheres (Walker and 105 Sytsma, 2007). 106 Since the initial phylogenetic study on Menthineae (Wagstaff and Olmstead, 1995), several 107 studies have been performed based on nuclear and plastid regions with increasing taxonomic 108 sampling of Salvia species (Walker and Sytsma, 2007; Takano and Okado 2011; Will and 4 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 109 Claßen-Bockhoff, 2014; Drew and Sytsma, 2012; Will et al., 2015; Will and Claßen-Bockhoff, 110 2017; Hu et al., 2018; Drew et al., 2017; Fragoso-Martinez et al., 2017; Kriebel et al., 2019; 111 Wu et al., 2021). In the first molecular study of Salvia (based on rbcL and the trnL-trnF 112 regions), Walker and Sytsma (2004) found that Salvia is not monophyletic and recognized 113 three clades: clade I includes many species of Salvia from the Eastern Hemisphere along with 114 a Western Hemisphere lineage (8 species from former sect. Heterosphacea and subgen. 115 Salviaostrum), clade II comprises North and South American species and includes subgen. 116 Calosphace Benth. and subgen. Audbertia Benth, and clade III comprises species from eastern 117 North Africa and southwestern Asia. Rosmarinus L. and Perovskia Karel. were placed as sisters 118 to clade I, while Dorystaechas Boiss. & Heldr. ex Benth, distributed in Turkey, was placed with 119 clade II. 120 Walker and Sytsma (2007), with increased taxon sampling for Salvia and related genera in 121 Menthineae, found that Meriandra Benth and Dorystaechas formed the sister clade to North 122 and South American species of Salvia (clade II). They referred to Zhumeria Rech.f & Wendelbo 123 (a monotypic genus endemic to Iran) along with southwest and East Asian Salvia as clade III. 124 Will and Claßen-Bockhoff (2014) excluded the East Asian Salvia species from Walker and 125 Sytsma’s (2007) clade III and considered them to represent an independent lineage (Clade IV). 126 Will and Claßen-Bockhoff (2017) suggested breaking the large Salvia group into six genera: 127 Salvia sensu stricto, Ramonia Raf., Lasemia Raf., Glutinaria Raf., Pleudia, and Polakia. 128 However, they did not provide a taxonomic revision. Drew et al. (2017) embedded these five 129 genera into a broadly defined Salvia and treated each as a subgenus. In recent phylogenetic 130 studies of Salvia, Hu et al. (2018) and Kriebel et al. (2019) followed and updated the Drew et 131 al. (2017) classification of Salvia, recognizing 11 subgenera. In this study, to maintain stability 132 in taxonomic definition and nomenclature, we follow the broad definition of Salvia (Drew et 133 al., 2017; Hu et al., 2018; Kriebel et al., 2019). A schematic diagram of changes in Salvia 134 delimitation based on previous phylogenetic studies is provided in Figure 1. 135 Frequent endemism and enormous morphological diversity have made interpretation of the 136 evolutionary patterns within Salvia challenging, particularly given the limited taxon sampling 137 for some areas, such as southwestern Asia. To improve taxon sampling for southwestern Asia 138 and to clarify patterns of morphological evolution and species diversification, we generated 139 new sequences for 50 Iranian species of Salvia and reconstructed a phylogeny for 351 species 5 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 140 overall. Notably, other recent phylogenetic analyses of Salvia differ in scope and emphasis 141 from our investigation. Kriebel et al. (2019) studied the effect of biome shifts and pollinators 142 on the radiation of Salvia. They found that shifts in pollination system are not correlated with 143 species diversification, except in subgen. Calosphace in the Western Hemisphere where 144 species are pollinated by hummingbirds. Kriebel et al. (2020) showed that the respective floral 145 morphospaces of the Western and Eastern Hemisphere Salvia are different. They inferred 146 that these differences in flower morphology are linked with shifts from bee to bird pollination. 147 In another recent study, Wester et al. (2020) found that shifts from bee- to bird-pollinated 148 Salvia are mostly associated with floral structure rather than floral colors. 149 Despite valuable contributions, the relationship between the evolution of floral traits and 150 patterns of Salvia diversification is not well understood. We used our new phylogenetic tree 151 for Salvia to trace patterns of both character evolution and diversification. We primarily focus 152 on the role of corolla length as one of the putative characters involved in Salvia diversification. 153 This is also the first attempt to trace the evolutionary history of corolla length in Salvia and 154 its association with diversification. Furthermore, we reconstructed the ancestral state for 155 lever mechanism and habitat with greater taxon sampling than in previous work (Will and 156 Claßen-Bockhoff 2014). In addition, we shed new light on the role of the pollination system 157 in Salvia diversification. We statistically examine the longstanding hypothesis that the lever 158 mechanism in Salvia flowers is correlated with high diversity and species richness. 159 2. Materials & Methods 160 2.1. Taxon Sampling 161 In total, 366 taxa representing 351 species covering all major areas of the geographic 162 distribution of Salvia were used to reconstruct the phylogeny. As noted, we considered Salvia 163 in the broad sense and included Zhumeria, Meriandra, Rosmarinus, and Perovskia (Drew et 164 al., 2017, Kriebel et al., 2019). Following Drew and Sytsma (2012), we selected Melissa and 165 Lepechinia as outgroups. We generated new sequences for many Iranian species of Salvia, 166 including 50 species (59 accessions) for the external transcribed spacer (ETS) region of nuclear 167 ribosomal DNA, 46 species (47 accessions) for ITS, and 35 species for the ycf1-rps15 region of 168 the plastome. The remaining sequences used here (representing 216 species) were obtained 169 from GenBank. We concatenated all sequences for the three plastid regions (rpl32, trnL-trnF, 6 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 170 ycf1-rps15) and two nuclear regions (ITS, ETS); the plastid and nuclear data sets were each 171 analyzed separately and then combined, given the highly similar topologies obtained for each. 172 That is, there was no strongly supported incongruence or conflict (hard incongruence sensu 173 Seelanen et al., 1997) between nuclear and plastid trees. The newly generated sequences 174 were deposited in GenBank. Corresponding information for each voucher specimen is 175 provided in Table 1. 176 2.2. DNA extraction, amplification, and sequencing 177 Total DNA was extracted from herbarium and silica-dried material using a modified CTAB 178 method (Doyle & Doyle 1987) in which, to break down secondary metabolites, the mixture of 179 ground leaf tissue and CTAB solution was kept at room temperature for 24 hours. ITS, ETS, 180 and ycf1 regions were amplified using the polymerase chain reaction (PCR) with each sample 181 prepared in 25-μl volumes with the following components: 1 μl of DNA solution (20 ng), 2.5 182 μl of reaction buffer, 2 μl dNTP mix (0.2 mM), 1 μl of each primer (10 uM), 1 μl of MgCl2, and 183 1.5 μl of Taq DNA polymerase. The PCR conditions for the nuclear regions for most species 184 were: 95°C for 2 minutes, 32 cycles of denaturation for 20 seconds at 94°C, primer annealing 185 for 20 seconds at 50°C, and 2 minutes extension at 72°C, with a final extension of 7 minutes 186 at 72°C. For the ycf1 region, we modified the annealing temperature to 52°C for 1 minute 187 (PCR optimization was set based on personal communication with B. Drew). High-quality PCR 188 products were sequenced on an ABI 3730 DNA Analyzer (Applied Biosystems, Inc.) at the 189 University of Florida Interdisciplinary Center for Biotechnology Research (ICBR). 190 191 2.3. Alignment and phylogenetic analysis 192 All consensus DNA sequences were generated using Geneious Pro v. 10.22 (Biomatters, 193 Auckland, New Zealand). Alignments were performed with the MAFFT plugin in Geneious with 194 manual adjustment. Maximum likelihood analysis was performed using the CIPRES Science 195 Gateway with RAxML HPC v.8 on XSEDE using the GTRGAMMA model with _Fa (rapid 196 bootstrapping analysis/search for the best ML tree) with 1000 iterations for bootstrapping. 197 Default settings were used for other options. Phylogenetic analyses were conducted for 1) all 198 plastid loci (rpl32-trnl, trnl-trnf, and ycf1-rps15), 2) both nuclear loci (ITS and ETS), and 3) a 199 combined data set of plastid and nuclear loci. 200 7 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 201 2.4. BEAST analysis (divergence time estimation) 202 We estimated divergence times using BEAST version 2.2.0 (Bouckaert et al., 2014) under the 203 uncorrelated lognormal model. Priors for the branch rate were assumed as a Yule process. A 204 node prior was calibrated for the most recent common ancestor (MRCA) of Melissa and 205 Lepechinia (28.4 Ma with a mean of 1.5 and a SD of 0.5; Drew and Sytsma 2012; Kriebel et al., 206 2019) with a lognormal distribution. The BEAST analysis was performed with two independent 207 runs of Markov Chain Monte Carlo. Each run was performed for 2*108 generations, with 208 parameters logged every 1000 generations. We used Tracer v. 1.6 to evaluate the ESS 209 (Effective Sample Size) to assure that the chains were run sufficiently long. An ESS > 200 210 indicates that the two independent runs were adequate. Tree Annotator was used to find the 211 maximum clade credibility reporting median node ages after discarding the first 10% of the 212 generations as burn-in. 213 214 2.5. Ancestral state reconstruction 215 Two characters with discrete states were scored: mode of lever mechanism (present / absent) 216 and habit (woody / herbaceous). We treated shrubs and subshrubs as woody; however, 217 distinguishing woody from herbaceous is not always straightforward because some mostly 218 herbaceous plants may become woody in special climatic situations (FitzJohn et al., 2014; 219 Zanne et al., 2014). Therefore, we treated a species as woody if it is considered a shrub or 220 subshrub in the literature or if it was defined as having a woody rootstock. In addition, the 221 continuous character corolla length was measured from the joint of the calyx to the end of 222 the upper lip. 223 The relevant data for the discrete and continuous traits were collected from the literature: 224 Flora of China (www.efloras.org/flora_page.asp? flora_id = 2), California Salvia (Epling, 1983), 225 Flora of USSR (Pobedimova, 1954), Flora of Turkey and the East Aegeans (Hedge, 1982), Flora 226 Iranica (Hedge, 1982), Flora of Southern Africa (Codd, 1985), Flora of Madagascar (Hedge, 227 1992), Flora dels Paiso Catalans (Bolos and Vigo, 1995; Wester and Claßen-Bockhoff, 2011), 228 and Flora of Iran (Jamzad, 2012). Additionally, we used online resources (www.gbif.org; 229 www.tropicos.org) as sources of data. For some species, the corolla length was measured 230 using the digitized type specimen available on JSTOR’s Global Plants database 231 (http:/plants.jstor.org). 8 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 232 For the discrete data, we used maximum likelihood to define the best model fitting our data 233 using the function ‘ace’ implemented in the R package ape v5. 3 (Paradis et al., 2004). We 234 tested “ER” (Equal Rates) and “ARD” (All Rates Different) on our data, and the best model was 235 selected based on the Akaike Information Criterion (AIC) (Akaike, 1974). We used the Akaike 236 weight using aic.w function in the R package geiger v2.0.6 to select the best model for those 237 data wih low delta AIC between ER and ARD models. To reconstruct ancestral states, we used 238 stochastic character mapping with 1000 iterations using the make.simmap function in the 239 phytools v0.7.78 package (Revell, 2012). We also reconstructed the ancestral state of corolla 240 length using the lik.anc in phytools to calculate the likelihood of each ancestral state. 241 Ancestral states of corolla length and 95% confidence intervals were evaluated using the 242 function anc.ML with an OU (Ornstein-Unlenbeck) model in phytools. 243 2.6. Macroevolutionary patterns within corolla length 244 We focused on corolla length as one of the most important morphological traits that might 245 influence pollinator-flower interactions (Fernández- Mazuecos et al., 2013; Gómez et al., 246 2016; Landis et al., 2018). To investigate the evolutionary dynamics of corolla length 247 throughout Salvia phylogeny, we applied three quantitative approaches based on the time- 248 calibrated phylogeny as follows: 249 250 2.6.1. Diversification model 251 We examined three evolutionary models with different patterns of phenotypic evolution 252 using the R package geiger v2.0.6 (Harmon et al., 2008) following three different models. 1) 253 The Brownian Motion model (BM): This model describes a “random walk” of evolution for 254 continuous characters. 2) The Ornstein-Uhlenbeck model (OU): This model describes the local 255 occurrence of stabilizing selection in which the trait is drawn toward optimal fitness (Hansen 256 1997). 3) The Early Burst model (EB): This model is known as a classic model of adaptive 257 radiation, in which the initial stage of morphological evolution is rapid with decreasing 258 morphological evolution after ecological spaces are filled (Harmon et al., 2010). Based on a 259 recent model of diversification reconstructed by Aguilée et al. (2018), after an initial phase of 260 geographic adaptive radiation, diversification rates can be affected not only by ecological 261 niches, but also by genetic processes, competition, and landscape dynamics. The best model 262 for explaining diversification of Salvia was selected based on the AIC (Akaike, 1974). 9 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 263 264 2.6.2. Disparity Through Time 265 Disparity Through Time (DTT) of the corolla length was modeled using the R package geiger 266 v2.0.6 (Harmon et al., 2008). This analysis uses corolla length of extant Salvia species to 267 reconstruct ancestral corolla length values and model disparity between species. This 268 approach estimates the pairwise Euclidean distance of the trait over time and compares it 269 with the expected value under a null model of Brownian motion by iterative simulation. 270 Phenotypic disparity refers to the phenotypic variation among related species (Harmon, 271 2003). We simulated corolla length evolution with 10,000 generations across the 272 phylogenetic tree built from the combined data set of plastid and nuclear sequences. The 273 Morphological Disparity Index (MDI) was calculated, and the average disparity of corolla 274 length from the real and simulated data was plotted. Negative MDI shows lower disparity of 275 the trait than expected, and positive MDI indicates strong overlap in morphospace and 276 higher disparity within subclades (Donoso et al., 2015). 277 278 2.6.3. Diversification rate 279 To assess variation in rates of diversification of corolla length across Salvia, we used the 280 phenotypic trait module in BAMM. We simulated 20,000,000 generations, and the priors 281 were set using the function “SetBAMMpriors” in the R package BAMMTools v.2.1.6 (Rabosky 282 et al., 2014). We specified the sampling fraction by accounting for the number of samples for 283 each of the four major clades. Sampling fractions were set as: 0.47 (clade I), 0.26 (clade II), 284 0.95 (clade III) and 0.56 (clade IV). We performed MCMC simulation with 20,000,000 285 generations by sampling every 1000 generations. We discarded the first 25% of runs as burn- 286 in. Effective Sample Size (ESS) > 200 was used to evaluate the convergence of four Markov 287 Chain Monte Carlo chains. The BAMM output was analyzed using BAMMtools. 288 289 2.8. Lever-mechanism-dependent diversification 290 To examine whether the diversification rate in Salvia is correlated with the presence of the 291 lever mechanism, we applied HiSSE (Hidden State Speciation and Extinction) implemented in 292 the R package hisse v2.1.1 (Beaulieu and O’Meara 2016), which is a modified method of BiSSE 10 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 293 (Binary State Speciation and extinction) (Maddison et al., 2014). Rabosky (2014) argued that 294 the BiSSE method suffers from type Ι and type II errors. In those cases, traits that are not 295 biologically correlated with speciation rates show significant effects on diversification 296 (Goldberg and Rabosky, 2015). In other words, rejecting the null hypothesis in BiSSE does not 297 mean the alternative is true. 298 Compared with the BiSSE model, the HiSSE model considers more free parameters and 299 assumes a hidden state for each of the observed states that potentially have independent 300 rates of diversification (0A, 1A, 0B, 1B). The Character Independent Diversification (CID) 301 models, which assume independent evolution for binary characters, were also implemented. 302 The CID models explicitly test that the evolution of a binary character is independent of the 303 diversification process without forcing the diversification process to be constant. Different 304 subsets of the HiSSE model that differ in speciation, extinction, and transition rate 305 parameters, along with standard BiSSE models, were estimated (cf. Harrington and Reeder, 306 2017). We accounted for incomplete taxon sampling in our phylogeny by assigning the 307 sampling frequency of each state as 0.256 (presence of the lever mechanim) and 0.056 308 (absence of the lever mechanism). The model average of ancestral state and diversification 309 of all fitted models was plotted using the function “plot.hisse.state”. The advantage of this 310 function is that it accounts for both state and rate uncertainty of the models along plotted 311 branches. We also used FiSSE (Fast intuitive State-dependent Speciation Extinction) as a non- 312 parametric test for the lever-mechanism-dependent speciation rate. This method does not 313 depend on the character state, but considers the distribution of branch lengths (Rabosky and 314 Goldberg 2017). 315 316 2.9. Diversity-dependent diversification 317 We also used the R package DDD v2.7 (Etienne and Haegman, 2012) to test whether 318 diversification in Salvia is dependent or independent of diversity. DDD uses a hidden Markov 319 model to calculate the likelihood of phylogenetic history under a diversity-dependent birth- 320 death model of diversification. DDD estimates “K”, the maximum number of species that a 321 clade can have in a given environment; a value of K near the number of extant species 322 suggests that a clade is close to its ecological limit. The other two models are a density- 323 dependent logistic (DDL+E) model and a density-dependent exponential (DDE+E) model. The 324 model with the lowest AICc was selected as the best model. We also calculated the maximum 11 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 325 likelihood evolutionary history pattern with both a Yule model and a constant rate birth-death 326 (CrBD) model. The Maximum Clade Credibility (MCC) of the BEAST output was used to 327 perform this analysis. We also examined four model fits as an alternative method using the R 328 package laser v2. 4 (Rabosky 2006). 329 330 3. Results 331 3.1. Characteristics of the phylogenetic data matrix and phylogenetic analysis 332 In this study, 143 new DNA sequences were generated for 50 Iranian Salvia species, including 333 50 species (59 accessions) for the ETS region, 46 species (47 accessions) for ITS, and 34 species 334 for ycf1. 335 Maximum likelihood analyses of all three data combinations were conducted: 1) plastid loci 336 (rpl32-trnl, trnl-trnf, and ycf1-rps15), 2) nuclear loci (ITS and ETS), and 3) a combined data set 337 of plastid and nuclear loci. The overall topologies of the nuclear loci, plastid loci, and the 338 combined data set are smilar in recovering major clades. Based on both nuclear and plastid 339 regions, the phylogenetic relationships among most of the Eurasian species in clade I are 340 unresolved. This result is not surprising given that we had more missing data in the plastid 341 partition than other partitions. The nuclear and combined data sets provide higher support 342 for most of the clades than plastid data. For example, clade III was recovered as fully resolved 343 based on nuclear regions and combined data, but based on plastid data, the relationship 344 between S. majdaea and S. macilenta with the remain group in clade III was unresolved. Trees 345 based on the nuclear and combined data sets were highly similar, with only some minor 346 differences in support values for terminal clades. However, the combined data recovered a 347 more resolved phylogney. For example, in clade I within Subgen. Heterosphace, the 348 relationship between the S. verticillata group with the remaining taxa was resolved based on 349 the combined data but not the nuclear data. As a result, we used the results from the 350 combined data set in subsequent analyses and in our discussion below. 351 We recovered four major clades of Salvia species: clade I (Eurasian and southern African 352 Salvia), clade II (South and North American Salvia), clade III (southwestern Asian and northern 353 African Salvia), and clade IV (Southeast Asian Salvia). With more taxon sampling, we provide 354 a new phylogeny for clade III with species that are primarily distributed in southwestern Asia. 355 However, we mostly focus here on newly recovered relationships for Iranian species of Salvia 12 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 356 and the clades to which they belong rather than on Salvia phylogeny as a whole. For more 357 straightforward comparisons, we also used both clade names provided in previous 358 phylogenetic studies of Salvia (Will and Claßen-Bockhoff, 2017) along with the recent 359 classification (Drew et al., 2017; Kriebel et al., 2019) in Suppl. 1. 360 Clade I 361 In this clade, species are mainly distributed in Europe, Central Asia, western Asia, and 362 southern Africa. This clade contains 140 species (170 accessions) out of the 250-300 species 363 described for these areas. They fall into four distinct subclades (subclades I-A, I-B, I-C, and I- 364 D) of Will and Claßen-Bockhoff (2017) including subgenera Salvia Benth., Sclarea Benth., and 365 Heterosphace Benth (Kriebel et al., 2019). For the first time, resolution within the S. 366 verticillata group was obtained. Salvia taraxacifolia was recovered as the sister to the S. 367 verticillata group (BS = 81%) consisting of S. verticillata, S. judaica, and S. russellii. This clade 368 was in turn sister (BS = 95%) to the southern African clade. 369 Subclade I-C and I-D: 26 species of Iranian Salvia sampled here were recovered as members 370 of subclades I-C, and six species were placed in subclade I-D, including subgen. Sclarea and 371 Salvia based on the updated subgeneric classification of Salvia (Will and Claßen-Bockhoff 372 2017; Drew et al., 2017; Kriebel et al., 2019). These subclades include Salvia distributed in 373 western Asia (Afghanistan, Iran, Iraq, and Turkey), Central Asia, Europe, and the Canary 374 Islands. Based on the combined data set of nuclear and plastid loci, the phylogenetic 375 relationships among most of the taxa were unresolved. However, several groups were 376 identified within this polytomy: 1) S. jamzadaei, S. macrochlamys, S. bracteata, and allied 377 species; 2) two species endemic to Iran (S. leriifolia and S. hypochionaea) along with S. 378 montbretia, S. daghestanica, and S. phlomoides; 3) S. spinosa, S. sclareopsis, S. macrosiphon, 379 S. reuterana, S. perspolitiana, and S. palaestina (a clade with BS = 60%); and 4) S. nemorosa 380 and S. virgata from southwestern Asia, the Caucasus, and Europe along with S. deserta and 381 Salvia × sylvestris. 382 Clade II 383 Salvia species in this clade are endemic to South and North America, and clade II includes 384 more than half of all Salvia species, with approximately 600 species in subg. Calosphace and 13 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 385 19 species in subg. Audibertia. Clade II is recovered with BS= 97%; however, relationships 386 among taxa in this clade are not fully resolved. Clade I is sister to clade II with BS= 74%. 387 Clade III 388 This clade contains species of Salvia from northern Africa and southwestern Asia. We provide 389 the most comprehensive taxon sampling for clade III to date by generating 13 new sequences 390 for members of this group from the region of Iran. Clade III was recovered with 100% BS 391 support with a well-resolved phylogeny in trees from nuclear and combined data. Based on 392 the combined nuclear and plastid tree, S. majdae, which was placed in subgen. Zhumeria 393 (Drew et al., 2017), was found instead to be sister with high support (BS = 97%) to the S. 394 aristata group, which includes S. pterocalyx (from northeastern Afghanistan) and S. vvedenskii 395 and S. margaritae (from Central Asia). Salvia majdae and the S. aristata group were placed as 396 the sister clade to a trichomy of S. aegyptiaca, S. macilenta, and S. eremophila. 397 Clade IV 398 Salvia species in this clade are restricted to eastern Asia, with the exception of S. glutinosa 399 and S. plebeia. We provide new sequence data for S. glutinosa, which is distributed in the 400 northern part of Iran and some parts of Europe. Salvia plebeia is reported from Iran and 401 Afghanistan and extends to Southeast Asia. Based on our results, S. glutinosa forms a clade 402 with S. nubicola, S. koyamae, S. glabrescense, and S. nipponica with BS = 88%. 403 404 3.2. Divergence times 405 Divergence times were estimated using the combined data set of nuclear and plastid loci. The 406 results are congruent with previous results (Drew et al., 2017; Kriebel et al., 2019). Our BEAST 407 analysis (Fig. 3) suggests that Salvia originated in the Oligocene ~34 Ma. Divergence time 408 estimation showed that the split between clade I and the rest of Salvia occurred 409 approximately 31 Ma (95% HPD = 37.6-27.5 Ma). In clade I, the North American clade diverged 410 from the African and Mediterranean clade approximately 15 Ma (95% HPD = 20.06-10.69) 411 during the middle Miocene. The age of the MRCA of most of the Iranian Salvia species in clade 412 I was estimated as 14.3 Ma (95%HPD = 18.74-10.47 Ma) near the end of the early Miocene. 413 Clade II diverged from clade III (mostly from southwestern Asia) in the early Miocene (95% 14 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 414 HPD = 27.8-17.8 Ma). The split between clade IV (eastern Asia) and clade III (southwestern 415 Asia and northern Africa) is estimated to have occurred during the late Oligocene (95%HPD = 416 31.5-20.26 Ma). 417 3.3. Ancestral character reconstruction 418 Corolla length varies from 4 mm in S. aegyptiaca in clade III to 51 mm in S. patens in clade II. 419 The most recent common ancestor for Salvia was reconstructed as having a corolla length of 420 approximately 15-18 mm (Fig. 4). Corolla length of ~20 mm was inferred as the ancestral state 421 for clade I. In clade I, within subg. Sclarea, multiple shifts from a corolla length of 20 mm to 422 smaller corollas occurred, but in subg. Salvia, all the species evolved corolla lengths longer 423 than 20 mm. In subg. Heterosphace, including bird-pollinated species of Salvia from southern 424 Africa (S. africana-lutea, S. lanceolata, S. thermarum), the corolla length is more variable, 425 ranging from ~7-41 mm. In clade II, shifts in the range of corolla size were much higher than 426 in other clades, especially in subg. Calosphace. In clade III, the ancestral state of corolla length 427 was recovered as ~15 mm. Within this clade, species of Salvia have small flowers (4-9 mm) 428 with shifts to larger flowers in subg. Zhumeria and the S. aristata group. 429 The ER model was selected as the best model for the evolution of the lever mechanism based 430 on the AIC value. However, the difference between the ER (Equal Rates) and the ARD (All 431 Rates Different) model was minimal (ER 168.46 vs. ARD 169.68). The Akaike weight for the ER 432 model (0.65) was higher than for the ARD model (0.32). Therefore, we reconstructed the 433 ancestral state of the lever mechanism based on the ER model. The ancestral state of the 434 lever mechanism for Salvia was equivocal (Fig. 5). In clade I, the ancestral state of the lever 435 mechanism was also equivocal, but the ancestral state of subg. Salvia, Sclarea, and 436 Heterosphace is an active lever mechanism. In clade II, the ancestral state is equivocal with 437 several shifts in subg. Calosphace from an active lever to a non-active lever. In clade III, Salvia 438 species lack an active lever mechanism, and the ancestral state for the clade is equivocal. 439 The ARD model was moderately suggested as the best model (Delta AIC = 3.345) for inferring 440 the ancestral state of habit across the Salvia phylogeny. The ancestral state of habit for all of 441 Salvia, as well as major subclades, was found here to be equivocal (Fig. 6). In clade I, species 442 of subgen. Sclarea are mostly herbs with a few shifts to shrub forms, but subgen. Salvia and 15 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 443 Heterosphace are mostly shrubs with several shifts to herbaceousness. In clade II, especially 444 in subgen. Calosphace, shifting from herb to shrub was more frequent than in other clades. 445 446 3.4. Tempo and mode of corolla length evolution 447 The analysis of disparity through time showed that the rates of diversification in corolla length 448 among subclades of Salvia are higher than expected under a null hypothesis of Brownian 449 motion (MDI = +0.21). Therefore, Salvia subclades have diversified greatly in corolla length. 450 The corolla length decreased during the Miocene between approximately 17-15 Ma (within 451 the 95% CI calculated from simulations of corolla length disparity), but showed a remarkable 452 increase during the last ~10 Ma, during which the relative disparity of corolla length is higher 453 than the 95% DTT range of simulated data (Fig. 7). 454 The model-based analysis of corolla length diversification determined "OU" as the best 455 approximation model of this trait across Salvia phylogeny. Hence, our results suggest that 456 corolla length evolution underwent stabilizing selection towards a median value (Table 2). 457 3.4.1. Corolla length evolutionary rates 458 To assess whether the MCMC output of the BAMM analysis for corolla length has converged, 459 we checked the effective sample sizes of the log-likelihood and the number of shift events. 460 Based on ESSNumber 461 converged. The phylorate plot confirmed heterogeneous rates of evolution of corolla length 462 in Salvia. The best distinct shift configuration with the highest posterior probability was 463 detected at the MRCA of core Calosphace in clade II (Fig. 8). of shifts = 1139.461 and ESSLoglike= 1458.028, the MCMC simulation 464 465 3.5. Diversity-dependent diversification 466 The maximum likelihood analysis of lineage diversification showed that among four fitted 467 models, with two models dependent on diversity, the Yule model was selected as the best 468 model based on AIC values for explaining Salvia diversification through time. Hence, the 469 evolutionary pattern of Salvia diversification is independent of diversity (Table 3). The 470 estimated carrying capacity (K>3000), which refers to the potential number of species that a 16 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 471 clade can sustain, is higher than the number of extant species of Salvia (~1000). Rejection of 472 the diversity-dependent diversification model implies that Salvia has not reached its 473 ecological limit in terms of number of species and that speciation has not yet started to 474 decline due to increased species competition or fewer ecological resources. 475 476 3.6. Lever-mechanism-dependent diversification 477 We used HiSSE (Beaulieu and O’Meara, 2016) and FiSSE (Rabosky and Goldbeg., 2017) 478 methods for analyzing the effect of the active lever mechanism on diversification. We found 479 that the HiSSE model with an equal irreversible transition rate among states (q0B1B=0, 480 q1B0B=0, all other q's equal) is the best model for explaining the effect of the lever 481 mechanism on Salvia diversification. Better performance of the HiSSE model than the BiSSE 482 model indicates a signal of lever-mechanism-dependent diversification as well as a signal of 483 other unobserved or unmeasured traits. Therefore, we infer that the lever mechanism is 484 indirectly responsible for Salvia diversification (Table 4). Based on FiSSE two-tailed 485 parameters, the P-value = 0.69; therefore, the null hypothesis of a close association between 486 the lever mechanism and Salvia diversification is rejected. The average tip rate of 487 diversification for an active lever mechanism is λ 1 = 0.26 and for a non-active lever mechanism 488 is λ0 = 0.22. 489 490 4. Discussion 491 4.1. Phylogeny 492 By including 50 species of Salvia from the Iran region, the limited taxon sampling for the 493 Eastern Hemisphere encountered in previous studies (Drew and Sytsma, 2012; Will and 494 Claßen-Bockhoff, 2014, 2017; Drew et al., 2017) was remedied to some extent. To reconstruct 495 the phylogenetic tree of Salvia as comprehensively as possible, our newly generated data 496 were combined with relevant sequences from previous studies (Walker and Sytsma, 2007; 497 Drew and Sytsma, 2012; Takano and Okado 2011; Will and Claßen-Bockhoff, 2014; Fragoso- 498 Martineze, 2017). 17 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 499 Our results for Salvia as a whole are similar to those reported in previous phylogenetic results 500 in recovering four major well-supported clades comprising six subgenera were recovered for 501 Salvia (Will and Claßen-Bockhoff 2017; Drew et al., 2017; Kriebel et al., 2019). The resultant 502 trees for all the data sets largely agree with each other. Nevertheless, some discrepancies in 503 phylogenetic relationships within subclades were observed. In the paragraphs that follow, we 504 summarize and discuss the major phylogenetic results of this study and also compare our 505 findings to other studies of Salvia. 506 Clade I 507 This clade includes species from three subgenera, Heterospahce, Salvia, and Sclarea (Drew et 508 al., 2017; Kriebel et al., 2019). Salvia species in this clade are distributed in small areas of 509 North America, southern Africa and Madagascar, western Asia, Europe, and the Canary 510 Islands. Although the phylogenetic relationships among most of the taxa in clade I are still not 511 well resolved, our use of more nuclear and plastid regions was helpful in recovering the 512 African clade with higher bootstrap support than previously reported (Will and Claßen- 513 Bockhoff, 2017). 514 Subgen. Heterosphace 515 This group comprises three supported subclades (supplementary 1). 1) Subclade I-A includes 516 species from both northern and southern Africa. Salvia nilotica and S. somalensis, two species 517 distributed in Tanzania and Ethiopia, respectively, were recovered as successive sisters to the 518 southern African clade. 2) Subclade I-B comprises several species of Salvia from North 519 America, formerly classified in section Salviastrum Scheele. Kriebel et al. (2019) argued that 520 dispersal to eastern North America (sect. Salviastrum of subgen. “Heterosphace”) from the 521 Eastern Hemisphere lineage occurred during the mid-Miocene. 3) The S. verticillata group: 522 Salvia taraxacifolia was recovered as sister to the S. verticillata group (BS = 81%) consisting of 523 S. verticillata, S. judaica, and S. russellii. The latter clade was sister (BS = 95%) to the southern 524 African clade. Previous studies (Will and Claßen-Bockhoff, 2014; Will and Claßen-Bockhoff, 525 2017) failed to resolve the phylogenetic position of S. taraxacifolia (Mediterranean element) 526 or with low support (kriebel eta l., 2019) within the S. verticillata group. Our results from the 527 combined nuclear and plastid data recovered S. taraxacifolia as sister to the S. verticillata 528 group with high support. S. nilotica placed as sister to the southern African clade species. 18 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 529 Most of the Salvia species in clade I (Europe, Madagascar, Central Asia), along with Iranian 530 species of Salvia, are placed in subclades I-C and I-D and are classified as subgen. Sclarea and 531 Salvia following recent treatments (Drew et al., 2017; Kriebel et al., 2019). Although 532 phylogenetic relationships among species of Salvia in these subclades are mostly unresolved, 533 our increased taxon sampling was helpful in determining the phylogenetic position of Iranian 534 Salvia species and provides evolutionary insights and rationale for improving the taxonomy 535 of Salvia. 536 Clade II 537 As noted above, clade II comprises species from North and South America, within which 538 approximately half of all Salvia species are distributed. Because relationships among species 539 from this geographic region were not the focus of this study, our sampling from this region is 540 more limited than from central and western Asia, Europe, and Africa. Relationships are 541 therefore largely unresolved. For more detail about relationships among species of Salvia 542 from the Americas, we refer the reader to (Walker et al., 2015; Fragoso-Martínez 2018; Kriebel 543 et al., 2019) 544 Clade III 545 Clade III encompasses species distributed in northern Africa and southwestern Asia. Within 546 this clade, species are mostly dwarf shrubs with smaller flowers than those of other Salvia 547 clades. In the Iran flora, most species are distributed in the southern region of Iran (25-27 N°). 548 We present here the best-resolved phylogeny obtained for clade III with the most 549 comprehensive taxon sampling to date; our nuclear data set fully resolved relationships with 550 BS = 100%. Salvia majdae, formerly classified as the monotypic genus Zhumeria, is an endemic 551 aromatic shrub in southern Iran. Based on a recent study (Soltanipour et al., 2020), S. majdae 552 is reported as an endangered species on the IUCN Red List based on Extent of Occurrence 553 (EO) and Area Of Occupancy (AOO). Salvia aristata is another endemic species of Iran in this 554 clade placed as the sister species of S. majdae. This species has a different habit (herabaceous 555 perennial) from the remaining species in clade III, as well as a larger distribution in Iran. 556 Clade IV 19 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 557 Clade IV is restricted to eastern Asia. Notably, S. glutinosa, which is distributed in northern 558 Iran and western Europe, forms a clade with species from eastern Asia. Salvia glutinosa also 559 shares similar traits with S. nubicola in corolla color (yellow with brown-purple spots on the 560 lower lip) and with a clade of S. koyamae, S. glabrescense, and S. nipponica in leaf form and 561 flower shape (Hu et al., 2018). Thus, S. glutinosa may have historically had a larger distribution 562 than currently displayed. It is likely that S. glutinosa is a relict Arcto-Trertiary element and 563 that the Euxine-Hyrcanian province (western Europe, northern Iran) was a refugium for this 564 species (Browicz 1987, Akhani et al., 2010). Ecological niche modeling projecting into the past 565 may enable a more complete view of the past distribution of S. glutinosa. A more detailed 566 view of the phylogeny of Salvia from eastern Asia is found in (Hu et al., 2018; Hu et al., 2020xx. 567 4.2. Divergence times 568 Our estimate for the date of origin of Salvia (31 Ma, 95% HPD = 37.6-27.5 Ma) is consistent 569 with previous studies (Drew et al., 2012; Drew et al., 2017), which is not surprising given that 570 the calibration used here was based on Drew et al. (2012) from analysis on a larger taxonomic 571 scale of Nepethoideae based on a fossil fruit of Melissa from the Early-Middle Oligocene. 572 The Qinghai-Tibetan Plateau (QTP) underwent four periods of uplift: 25-17 Ma, 15-13 Ma, 8- 573 7 Ma, and 3.5-1.6 Ma. The major radiation for Salvia in eastern Asia in clade IV is estimated 574 at 8-10 Ma, which coincides with the QTP uplift in the late Miocene. Our estimate for the 575 crown age of eastern Asian Salvia (~12 Ma) is consistent with that of Drew et al. (2017), but 576 is younger than that reported in another recent study (Hu et al., 2018) on eastern Asian Salvia 577 with an estimated date of ~17 Ma. This inconsistency might be because of different taxon 578 sampling, placement of calibrations, or prior distribution of the calibration node among the 579 studies. Our data suggest that the QTP uplift played an important role in local diversification 580 of Salvia, as it has for other plant genera in eastern Asia (Yao et al., 2016; Malik et al., 2017; 581 Hu et al., 2018). 582 The Arabia-Eurasian collision in the Oligocene-early Miocene led to the emergence of the 583 Alborz and Zagros Mountains in the Middle Miocene (15-12 Ma) in the Iranian plateau 584 (Manafzadeh et al., 2016). The main stage of crustal thickening from the collision was ~25 Ma, 585 and the uplift of the Iranian plateau took place ~15-12 Ma, with further uplift ~5 Ma (Djamali 586 et al., 2012; Manafzadeh et al., 2016). The emergence of these mountains coincides with the 20 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 587 age of the MRCA of Iranian Salvia species. Formation and uplift of mountains can play an 588 important role in evolutionary diversification through providing heterogeneous niches and 589 landscapes. Therefore, we postulate that the emergence and uplift of the Iranian Mountains 590 during the last ~12 Ma, along with subsequent aridification (Manafzadeh et al., 2016; Folk et 591 al., 2020), provided new ecological opportunities and habitat for Salvia diversification in Iran. 592 593 4.3. Ancestral state reconstruction 594 4.3.1. Corolla length 595 Across Salvia phylogeny, there were multiple shifts from a corolla length longer than 25 mm 596 to shorter corollas within and among clades. The MRCA of subg. Calosphace had a corolla 597 length less than 25 mm, but multiple shifts from short (~4 mm) to long (~45 mm) corollas 598 occurred in subclades of Calosphace. Based on the current study and previous reports (Wester 599 and Claßen-Bockhoff, 2011; Wester et al., 2020), most Salvia species with an average corolla 600 length of 22.3±6.5 mm are visited by bees, while bird pollinators are more attracted to flowers 601 with an average corolla length of 31±9.5 mm. Floral construction is associated with the type 602 of pollinators in Salvia (Wester et al., 2020), and an overall correlation between flower size 603 and pollinator is not expected across all Salvia lineages. For instance, S. blepharochlaena, 604 which is distributed in Turkey, is melittophilous, but has a long corolla. Salvia purpurea in 605 subgen. Audibertia has an intermediate flower (pollinated by bees and hummingbirds); S. 606 purpurea has a long corolla (19-36 mm) and long flower tube that is characteristic of 607 hummingbird-pollinated Salvia species, but the flower has the wide landing platform of a bee- 608 pollinated flower (Wester and Claßen-Bockhoff, 2011). Special flower traits like a short flower 609 tube cause a phenotypic trade-off and adaptation to birds and bees, but if a short tube is 610 combined with a narrow corolla opening, this combination of floral traits can generalize to 611 both pollinators (Ohashi et al., 2021). We do not imply that corolla length is the only trait 612 involved in Salvia–pollinator interactions; other factors, such as flower shape, tube length, 613 and color, may also be involved in pollinator attraction and adaptation (Landis et al., 2018; 614 Wessinger et al., 2019; Kriebel et al., 2020). 615 4.3.2. Lever mechanism 21 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 616 Salvia species with an active lever mechanism are characterized by modified stamens. The 617 lever is formed by elongation of the connective tissue that widens and separates the two 618 thecae from each other. Levers have evolved several times in parallel both within and 619 between clades I and II (Drew and Sytsma, 2012). In this study, we inferred the ancestral state 620 based on two models considering different rates of evolution. Based on the ER model (Equal 621 Rates), the ancestral state for Salvia is equivocal, and two alternative hypotheses may explain 622 the distribution of the lever mechanism across Salvia. First, the MRCA of Salvia may have had 623 a non-active lever mechanism, and a lever evolved independently multiple times in separate 624 lineages. Alternatively, the ancestor of Salvia may have had an active lever mechanism, and 625 several losses and reversals took place throughout the clade. Additionally, the HiSSE analysis, 626 with the preference of the irreversible model for lever mechanism diversification, suggests 627 that changes from non-active to active lever is not plausible or at the very least evolutionarily 628 difficult. Therefore, we argue that a Salvia ancestor with an active lever is more probable than 629 a non-active lever. 630 4.3.3. Habit 631 The ancestral habit in Salvia is reconstructed here as equivocal. Several shifts from woody to 632 herbaceous occurred within the main clades. This ambiguity might be due to diverse clades 633 that transition frequently between woody and herbaceous, making it difficult to infer the 634 state of the MRCA of Salvia. 635 4.4. Corolla length evolution 636 4.4.1. Disparity Through Time 637 The value obtained (MDI = 0.21) for the disparity of corolla length reflects a high rate of 638 diversification and morphological lability in related species. The positive MDI value for 639 disparity shows that most of the variation in corolla size is within subclades, while a negative 640 MDI indicates higher disparity among subclades, which is traditionally interpreted as adaptive 641 radiation (Harmon et al., 2010). Increasing disparity in corolla length during the last 10 M 642 years of Salvia evolution coincides with a number of geological events, including the Andean 643 uplift, Mexican vulcanization (clade II), the uplift of the QTP (clade IV), and the uplift of the 644 Zagros Mountains (occupied by species in clade I; Ferrari et al., 2012; Yao et al., 2016; 22 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 645 Manafzadeh et al., 2016), all major geological events that may have profoundly shaped Salvia 646 evolution worldwide. 647 Our positive value of MDI and support for the OU model contrast with traditional 648 interpretations of adaptive radiation in which MDI is negative through phenotypic 649 diversification with the Early Burst (EB) model of diversification (Harmon et al., 2010). In 650 addition, the DDD analyses do not support density-independent lineage diversification, and 651 the Yule model was selected as the best model with no apparent slowdown in Salvia 652 diversification. In the classic definition of adaptive radiation, the rate of diversification first 653 increases due to access to new niche space, followed by slow diversification as niche space 654 fills (Rabosky, 2013; Gillespie et al., 2020). However, Augilee et al. (2018) argued that 655 ecological niche filling as an explanation for negative-dependent diversity should be treated 656 with caution because biotic (competition) and abiotic factors (landscape dynamics) can 657 correspond to species diversity in different stages of a clade’s history. 658 4.4.2. Corolla length diversification 659 Floral traits have played a key role in enhancing angiosperm diversification (e.g., Stebbins, 660 1970; Fenster et al., 2014; Armbruster, 2014; Van der Neit and Johnson, 2014), and some 661 floral characters (corolla length, corolla tube length, corolla shape, and flower color) are 662 associated with pollinator interactions. The positive effects of certain floral traits on the 663 effectiveness of one group of pollinators relative to others occurs most often in bilaterally 664 symmetrical flowers (Ollerton, 2009; Armbruster, 2014; Wester et al., 2020). The rate of 665 evolution of corolla length in one clade of Calosphace (clade II) was significantly higher than 666 in other clades. Detection of correlated rate shifts in this clade implies that changes in corolla 667 length may have enabled an adaptive radiation in this clade. Species in this clade are mostly 668 distributed in South and Central America, including Bolivia, Mexico, Peru, and Argentina, and 669 include hummingbird-pollinated species with several shifts to bee pollination (Fragoso- 670 Martinez et al., 2017; Kriebel et al., 2019). Therefore, corolla size may be a putatively adaptive 671 trait that facilitated pollinator-flower interactions in this clade. 672 4.5. Lever-mechanism-dependent diversification 673 The special lever-mechanism pollination system in Salvia has been hypothesized to have 674 played a major role in Salvia diversification (Claßen-Bockhoff et al., 2004; Drew and Sytsma, 23 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 675 2012). The functionality and structure of the lever mechanism were tested through field 676 investigation and biomechanical experiments (Claßen-Bockhoff et al., 2004; Wester and 677 Claßen-Bockhoff, 2004; Reith et al., 2007; Drew and Sytsma, 2012; Zheng et al., 2015). 678 However, the actual effect of the lever mechanism on diversification has not been previously 679 investigated. We examined this hypothesis across our phylogeny using a Hidden Markov 680 Model implemented in the HiSSE package. The best model fitted was the HiSSE model with 681 irreversible transitions among states. The lever mechanism likely has an important impact on 682 pollination success (Classen-Bockhoff et al., 2004; Zheng et al., 2015; Kriebel et al., 2019) and 683 may have influenced diversification, but we did not find any evidence for a direct association 684 of lever mechanism with Salvia diversification. Characters not measured here, including 685 flower shape features that are associated with the observed state, were likely influential as 686 well (Kriebel et al., 2020). Based on the HiSSE analysis, we suggest that emphasis on the lever 687 mechanism alone as the key promotor of diversification in Salvia may be misplaced and that 688 other phenotypic characters, especially other floral traits, should also be considered and 689 examined across the phylogeny. We should take into account that there might be 690 shortcomings and insufficient information in macroevolutionary models and that trees for 691 extant species may not permit the precise reconstruction of historical diversification (Louca 692 and Pennell, 2020). However, Helmsetter et al. (2021) argue that recent more complex 693 models can provide additional information and overcome the problems of relying on time 694 trees for extant species. An important issue for future studies in understanding Salvia 695 evolutionary history is assessing the effect of other floral traits on diversification via the 696 reconstruction of more robust phylogenetic trees using more genes and species. 697 Acknowledgement 698 We are thankful to Matthew Gitzendanner and Evgeny Mavrodiev, Florida Museum of 699 Natural History, for help with analyses and general lab assistance. 700 References 701 702 Aguilée, R., Gascuel, F., & Lambert, A. (2018). abiotic controls of speciation and extinction rates. Nature Communications, 1–13. https://doi.org/10.1038/s41467-018-05419-7. 703 704 Akaike, H. 1974.Anewlook at the statistical model identification. IEEE Trans. Autom. Control 19:716–723. 24 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 705 706 707 Akhani, H., Djamali, M., Ghorbanalizadeh, A., & And, E. R. (2010). Plant biodiversity of Hyrcanian relict forests, N Iran: An overview of the flora, vegetation, palaeoecology and conservation. Pakistan Journal of Botany, 42: 231–258. 708 709 710 Alfaro, M.E., Brock, C.D., Banbury, B.L. et al. Does evolutionary innovation in pharyngeal jaws lead to rapid lineage diversification in labrid fishes?. BMC Evol Biol 9, 255 (2009). https://doi.org/10.1186/1471-2148-9-255 711 712 713 Armbruster, W. S. (2014). Floral specialization and angiosperm diversity: Phenotypic divergence, fitness trade-offs and realized pollination accuracy. AoB PLANTS, 6, 1–24. https://doi.org/10.1093/aobpla/plu003 714 715 716 Beaulieu, J. M., & O’Meara, B. C. (2016). Detecting hidden diversification shifts in models of trait-dependent speciation and extinction. Systematic Biology, 65(4), 583–601. https://doi.org/10.1093/sysbio/syw022 717 718 Bolòs, O. de 1924-2007. (1990). Flora manual dels països catalans. https://doi.org/10.1017/CBO9781107415324.004 719 720 721 Bouckaert, R., Heled, J., Kühnert, D., Vaughan, T., Wu, C.-H., Xie, D., Drummond, A. J. (2014). BEAST 2: A Software Platform for Bayesian Evolutionary Analysis. PLOS Computational Biology, 10(4), 1–6. https://doi.org/10.1371/journal.pcbi.1003537 722 723 724 Browicz, K. (1989). Chorology of the Euxinian and Hyrcanian element in the woody flora of Asia. Plant Systematics and Evolution, 162(1), 305–314. https://doi.org/10.1007/BF00936923 725 726 727 Caetano, D. S., O’Meara, B. C., & Beaulieu, J. M. (2018). Hidden state models improve statedependent diversification approaches, including biogeographical models. Evolution, 72(11), 2308–2324. https://doi.org/https://doi.org/10.1111/evo.13602 728 729 730 Celep, F., Atalay, Z., Dikmen, F., Doğan, M., & Classen-Bockhoff, R. (2014). Flies as pollinators of melittophilous Salvia species (Lamiaceae). American Journal of Botany, 101(12), 2148– 2159. https://doi.org/10.3732/ajb.1400422 731 732 733 734 Claßen-Bockhoff, R., Speck, T., Tweraser, E., Wester, P., Thimm, S., & Reith, M. (2004). The staminal lever mechanism in Salvia L. (Lamiaceae): A key innovation for adaptive radiation? Organisms Diversity and Evolution, 4(3), 189–205. https://doi.org/10.1016/j.ode.2004.01.004 735 736 737 Claβen-Bockhoff, R., Wester, P., & Tweraser, E. (2003). The Staminal Lever Mechanism in Salvia L. (Lamiaceae) - a Review. Plant Biology, 5(1), 33–41. https://doi.org/10.1055/s2003-37973 738 739 Crane, P. R., Friis, E. M., & Pedersen, K. R. (1995). The origin and early diversification of angiosperms. Nature, 374(6517), 27–33. https://doi.org/10.1038/374027A0 740 25 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 741 742 743 744 Crepet, W. L. (2000). Progress in understanding angiosperm history, success, and relationships: Darwin's abominably “perplexing phenomenon.” Proceedings of the National Academy of Sciences, 97(24), 12939 LP-12941. Retrieved from http://www.pnas.org/content/97/24/12939.abstract 745 746 747 748 Deng, T., Nie, Z., Drew, B. T., Volis, S., & Kim, C. (2015). Does the Arcto-Tertiary Biogeographic Hypothesis Explain the Disjunct Distribution of Northern Hemisphere Herbaceous Plants ? The Case of Meehania ( Lamiaceae ), 1–18. https://doi.org/10.1371/journal.pone.0117171 749 750 751 752 753 754 Djamali, M., Baumel, A., Brewer, S., Jackson, S. T., Kadereit, J. W., López-vinyallonga, S., Simakova, A. (2012). Review of Palaeobotany and Palynology Ecological implications of Cousinia Cass . ( Asteraceae ) persistence through the last two glacial – interglacial cycles in the continental Middle East for the Irano-Turanian fl ora. Review of Palaeobotany and Palynology, 172, 10–20. https://doi.org/10.1016/j.revpalbo.2012.01.005 755 756 Dodd, M. E., Silvertown, J., Chase, M. W. (1999). Analysis of trait evolution and species diversity variation among angiosperm families. Evolution, 53(3), 732–744. 757 758 759 Drew, B. T., & Sytsma, K. J. (2012). Phylogenetics, biogeography, and staminal evolution in the tribe Mentheae (Lamiaceae). American Journal of Botany, 99(5), 933–953. https://doi.org/10.3732/ajb.1100549 760 761 762 Drew, B. T., González-gallegos, J. G., Xiang, C., Kriebel, R., Drummond, C. P., Walker, J. B., & Sytsma, K. J. (2017). Salvia united : The greatest good for the greatest number. TAXON, 66(1), 133–145. 763 764 Drummond, A. J., & Rambaut, A. (2007). BEAST : Bayesian evolutionary analysis by sampling trees, 8, 1–8. https://doi.org/10.1186/1471-2148-7-214 765 766 Epling, C. (1938-1939) A revision of Salvia, subgenus Calosphace. Feddes Rep. Beith., 110, 1383. 767 768 769 Etienne, R. S., & Haegeman, B. (2012). A conceptual and statistical framework for adaptive radiations with a key role for diversity dependence. American Naturalist, 180(4). https://doi.org/10.1086/667574 770 771 772 Fenster, C. B., Armbruster, W. S., Wilson, P., Dudash, M. R., & Thomson, J. D. (2004). Pollination syndromes and floral specialization. Annual Review of Ecology, Evolution, and Systematics, 35, 375–403.https://doi.org/10.1146/annurev.ecolsys.34.011802.132347 773 774 775 Fernández-Mazuecos, M., Blanco-Pastor, J. L., Gómez, J. M., & Vargas, P. (2013). Corolla morphology influences diversification rates in bifid toadflaxes (Linaria sect. Versicolores). Annals of Botany, 112(9), 1705–1722. https://doi.org/10.1093/aob/mct214 26 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 776 777 778 Ferrari, L., Orozco-Esquivel, T., Manea, V., & Manea, M. (2012). The dynamic history of the Trans-Mexican Volcanic Belt and the Mexico subduction zone. Tectonophysics, 522– 523, 122–149. https://doi.org/10.1016/j.tecto.2011.09.018 779 780 781 FitzJohn, R. G. (2012). Diversitree : comparative phylogenetic analyses of diversification in R. Methods in Ecology and Evolution, (3), 1084–1092. https://doi.org/10.1111/j.2041210X.2012.00234.x 782 783 784 FitzJohn, R. G., Pennell, M. W., Zanne, A. E., Stevens, P. F., Tank, D. C., & Cornwell, W. K. (2014). How much of the world is woody ?, (Theophrastus 1916), 1266–1272. https://doi.org/10.5061/dryad.63q27/2 785 786 787 788 Fragoso-Martínez, I., Martínez-Gordillo, M., Salazar, G. A., Sazatornil, F., Jenks, A. A., García Peña, M. del R., Granados Mendoza, C. (2018). Phylogeny of the Neotropical sages (Salvia subg. Calosphace; Lamiaceae) and insights into pollinator and area shifts. Plant Systematics and Evolution, 304(1), 43–55. https://doi.org/10.1007/s00606-017-1445-4 789 790 791 Folk, R. A., C. M. Siniscalchi, D. E. Soltis. 2020. Angiosperms at the edge: Extremity and diversity. Plant, Cell & Environment 43: 2871– 2893. https://doi.org/10.1111/pce.13887 792 793 794 Gillespie, R. G., Bennett, G. M., De Meester, L., Feder, J. L., Fleischer, R. C., Harmon, L. J Wogan, G. O. U. (2020). Comparing Adaptive Radiations Across Space, Time, and Taxa. Journal of Heredity, 111(1), 1–20. https://doi.org/10.1093/jhered/esz064 795 796 797 798 Gómez, J. M., Torices, R., Lorite, J., Klingenberg, C. P., & Perfectti, F. (2016). The role of pollinators in the evolution of corolla shape variation, disparity and integration in a highly diversified plant family with a conserved floral bauplan. Annals of Botany, 117(5), 889–904. https://doi.org/10.1093/aob/mcv194 799 800 801 Givnish, T. J. (2015). Adaptive radiation versus “ radiation ” and “ explosive diversification ”: why conceptual distinctions are fundamental to understanding evolution, New Phytologist, 207, 297–303.. 802 803 804 805 Han, T.-S., Zheng, Q.-J., Onstein, R. E., Rojas-Andrés, B. M., Hauenschild, F., Muellner-Riehl, A. N., & Xing, Y.-W. (2020). Polyploidy promotes species diversification of Allium through ecological shifts. New Phytologist, 225(1), 571–583. https://doi.org/https://doi.org/10.1111/nph.16098 806 807 Hansen, T. F. (1997). Stabilizing selection and the comparative analysis of adaptation. Evolution, 51(5), 1341–1351. https://doi.org/10.1111/j.1558-5646.1997.tb01457.x. 808 809 810 Harder, L. D., & Johnson, S. D. (2009). Darwin’ s beautiful contrivances : evolutionary and functional evidence. New Phytologist, 183, 530–545. https://doi.org/10.1111/j.14698137.2009.02914.x 811 812 Harley, R. M., Atkins, S., Budantsev, A. L., Cantino, P. D., Conn, B. J., Grayer, R., Upson, T. (2004). Labiatae. In J. W. Kadereit (Ed.), Flowering Plants·Dicotyledons: Lamiales 27 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 813 814 815 816 817 818 819 820 821 822 823 (except Acanthaceae including Avicenniaceae) (pp. 167–275). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-18617-2_11 Harmon, L. J., Schulte, J. A., Larson, A., & Losos, J. B. (2003). Tempo and Mode of Evolutionary Radiation in Iguanian Lizards. Science, 301(5635), 961–964. https://doi.org/10.1126/science.1084786 Harmon Luke J, Jason T Weir, Chad D Brock, Richard E Glor, and Wendell Challenger. 2008. GEIGER: investigating evolutionary radiations. Bioinformatics 24, 129-131. Harmon, L. J., Losos, J. B., Jonathan Davies, T., Gillespie, R. G., Gittleman, J. L., Bryan Jennings, W., Mooers, A. T. (2010). Early bursts of body size and shape evolution are rare in comparative data. Evolution, 64(8), 2385–2396. https://doi.org/10.1111/j.15585646.2010.01025.x 824 825 826 827 Harrington, S., & Reeder, T. W. (2017). Rate heterogeneity across Squamata, misleading ancestral state reconstruction and the importance of proper null model specification, 30, 313–325. https://doi.org/10.1111/jeb.13004 828 829 Hedge, I.C. (1982a). Salvia L. In: Davis PH (ed.) Flora of Turkey and the East Aegean Islands, 7, 400-461. Edinburgh Univ. Press. 830 831 Hedge, I.C. (1982b). Salvia L. In: Rechinger KH (ed.), Flora Iranica 150, 403-476, Akademische Druck und Verlagsanstalt, Graz. 832 833 834 835 Helmstetter, A. J., Glemin, S., Käfer, J., Zenil-Ferguson, R., Sauquet, H., de Boer, H. Condamine, F. L. (2021). Pulled Diversification Rates, Lineages-Through-Time Plots and Modern Macroevolutionary Modelling. BioRxiv. https://doi.org/10.1101/2021.01.04.424672 836 837 838 Hernández-Hernández, T., & Wiens, J. J. (2020). Why are there so many flowering plants? A multiscale analysis of plant diversification. American Naturalist, 195(6), 948–963. https://doi.org/10.1086/708273 839 840 841 Hu, G.-X., Takano, A., Drew, B. T., Liu, E.-D., Soltis, D. E., Soltis, P. S., Xiang, C.-L. (2018). Phylogeny and staminal evolution of Salvia (Lamiaceae, Nepetoideae) in East Asia. Annals of Botany, 122(4), 649–668. https://doi.org/10.1093/aob/mcy104 842 843 844 845 846 Hu, G.-X., Liu, E.-D., Wu, Z.-K., Sytsma, K. J., Drew, B. T., & Xiang, C.-L. (2020). Integrating DNA Sequences with Morphological Analysis Clarifies Phylogenetic Position of Salvia grandifolia (Lamiaceae): An Enigmatic Species Endemic to Southwestern China. International Journal of Plant Sciences, 181(8), 787–799. https://doi.org/10.1086/709134 28 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 847 848 849 Kaczorowski, R. L., Seliger, A. R., Gaskett, A. C., Wigsten, S. K., & Raguso, R. A. (2012). Corolla shape vs. size in flower choice by a nocturnal hawkmoth pollinator. Functional Ecology, 26(3), 577–587. https://doi.org/10.1111/j.1365-2435.2012.01982.x 850 851 852 853 Koutroumpa, K., Warren, B. H., Theodoridis, S., Coiro, M., Romeiras, M. M., Jiménez, A., & Conti, E. (2021). Geo-Climatic Changes and Apomixis as Major Drivers of Diversification in the Mediterranean Sea Lavenders (Limonium Mill.). Frontiers in Plant Science .https://doi.org/10.3389/fpls.2020.612258 854 855 856 857 Kay, K. M., & Sargent, R. D. (2009). The Role of Animal Pollination in Plant Speciation: Integrating Ecology, Geography, and Genetics. Annual Review of Ecology, Evolution, and Systematics, 40(1), 637–656. https://doi.org/10.1146/annurev.ecolsys.110308.120310 858 859 860 861 862 Kriebel, R., Drew, B. T., Drummond, C. P., González-Gallegos, J. G., Celep, F., Mahdjoub, M. M., Sytsma, K. J. (2019). Tracking temporal shifts in area, biomes, and pollinators in the radiation of Salvia (sages) across continents: leveraging anchored hybrid enrichment and targeted sequence data. American Journal of Botany, 106(4), 573–597. https://doi.org/10.1002/ajb2.1268 863 864 865 866 Kriebel, R., Drew, B., González-Gallegos, J. G., Celep, F., Heeg, L., Mahdjoub, M. M., & Sytsma, K. J. (2020). Pollinator shifts, contingent evolution, and evolutionary constraint drive floral disparity in Salvia (Lamiaceae): Evidence from morphometrics and phylogenetic comparative methods. Evolution, 74(7), 1335–1355. https://doi.org/10.1111/evo.14030 867 868 869 870 Lagomarsino, L. P., Condamine, F. L., Antonelli, A., Mulch, A., & Davis, C. C. (2016). The abiotic and biotic drivers of rapid diversification in Andean bellflowers (Campanulaceae). New Phytologist, 210(4), 1430–1442. https://doi.org/10.1111/nph.13920 871 872 873 874 Landis, J. B., Bell, C. D., Hernandez, M., Zenil-Ferguson, R., McCarthy, E. W., Soltis, D. E., & Soltis, P. S. (2018). Evolution of floral traits and impact of reproductive mode on diversification in the phlox family (Polemoniaceae). Molecular Phylogenetics and Evolution, 127(June), 878–890. https://doi.org/10.1016/j.ympev.2018.06.035 875 876 877 878 Landis, J. B., Toole, R. D. O., Ventura, K. L., & Gitzendanner, M. A. (2016). The Phenotypic and Genetic Underpinnings of Flower Size in Polemoniaceae The Phenotypic and Genetic Underpinnings of Flower Size in Polemoniaceae, 6(November), 1–20. https://doi.org/10.3389/fpls.2015.01144 879 880 881 882 Li, B., Cantino, P. D., Olmstead, R. G., Bramley, G. L. C., Xiang, C. L., Ma, Z. H., Zhang, D. X. (2016). A large-scale chloroplast phylogeny of the Lamiaceae sheds new light on its subfamilial classification. Scientific Reports, 6(October). https://doi.org/10.1038/srep34343 883 884 Li, P., Qi, Z. C., Liu, L. X., Ohi-Toma, T., Lee, J., Hsieh, T. H., Qiu, Y. X. (2017). Molecular phylogenetics and biogeography of the mint tribe Elsholtzieae (Nepetoideae, 29 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 885 886 Lamiaceae), with an emphasis on its diversification in East Asia. Scientific Reports, 7(1), 1–12. https://doi.org/10.1038/s41598-017-02157-6 887 888 Louca, S., & Pennell, M. W. (2020). Extant timetrees are consistent with a myriad of diversification histories. Nature, 580(7804), 502–505. 889 890 891 Maddison, W. P., Midford, P. E., & Otto, S. P. (2007). Estimating a Binary Character’s Effect on Speciation and Extinction. Systematic Biology, 56(5), 701–710. Retrieved from http://dx.doi.org/10.1080/10635150701607033 892 893 894 Malik, S., Vitales, D., Qasim Hayat, M., Korobkov, A. A., Garnatje, T., & Vallès, J. (2017). Phylogeny and biogeography of Artemisia subgenus seriphidium (Asteraceae: Anthemideae). Taxon, 66(4), 934–952. https://doi.org/10.12705/664.8 895 896 897 Manafzadeh, S., Salvo, G., & Conti, E. (2013). A tale of migrations from east to west : the Irano-Turanian floristic region as a source of Mediterranean xerophytes, 1–14. https://doi.org/10.1111/jbi.12185. 898 899 900 Manafzadeh, S., Staedler, Y. M., & Conti, E. (2016). Visions of the past and dreams of the future in the Orient : the Irano-Turanian region from classical botany to evolutionary studies, 9. https://doi.org/10.1111/brv.12287 901 902 903 McGuire, J. A., Witt, C. C., Remsen, J. V., Corl, A., Rabosky, D. L., Altshuler, D. L., & Dudley, R. (2014). Molecular phylogenetics and the diversification of hummingbirds. Current Biology, 24(8), 910–916. https://doi.org/10.1016/j.cub.2014.03.016 904 905 906 907 908 Ogutcen, E., Hamper, B., & Vamosi, J. C. (2014). Diversification in monkeyflowers: an investigation of the effects of elevation and floral color in the genus Mimulus. International Journal of Evolutionary Biology, 2014, 382453. https://doi.org/10.1155/2014/382453 909 910 911 Ohashi, K., Jürgens, A., & Thomson, J. D. (2021). Trade-off mitigation : a conceptual framework for understanding fl oral adaptation in multispecies interactions, 6. https://doi.org/10.1111/brv.12754 912 913 914 Onstein, R. E. (2019). Darwin’s second “abominable mystery”: trait flexibility as the innovation leading to angiosperm diversity. New Phytologist. https://doi.org/10.1111/nph.16294 915 916 917 Paradis, E., Claude, J., & Strimmer, K. (2004). APE: Analyses of phylogenetics and evolution in R language. Bioinformatics, 20(2), 289–290. https://doi.org/10.1093/bioinformatics/btg412 918 919 920 Pavlopoulos, G. A., Hooper, S. D., Sifrim, A., Schneider, R., & Aerts, J. (2011). Medusa: A tool for exploring and clustering biological networks. BMC Research Notes, 4(1), 384. https://doi.org/10.1186/1756-0500-4-384 30 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 921 922 923 Phylogenomic Mining of the Mints Reveals Multiple Mechanisms Contributing to the Evolution of Chemical Diversity in Lamiaceae. (2018). Molecular Plant, 11(8), 1084– 1096. https://doi.org/https://doi.org/10.1016/j.molp.2018.06.002 924 925 926 927 928 Pincheira-Donoso, D., Harvey, L. P., & Ruta, M. (2015). What defines an adaptive radiation? Macroevolutionary diversification dynamics of an exceptionally species-rich continental lizard radiation. BMC Evolutionary Biology, 15(1), 153. https://doi.org/10.1186/s12862015-0435-9 929 930 Pobedimova, E.G. (1954). Salvia L. - In: Schischkin, B.K. (ed.), Flora of the USSR 21, 178-260. [ translated from Russian] Israel Prog. Sci. Transel., Jerusalem. 931 932 933 Pyron, R. A., & Burbrink, F. T. (2013). Phylogenetic estimates of speciation and extinction rates for testing ecological and evolutionary hypotheses. Trends in Ecology & Evolution, 28(12), 729–736. https://doi.org/10.1016/j.tree.2013.09.007 934 935 936 Rabosky, D. L. (2006). LASER: A Maximum Likelihood Toolkit for Detecting Temporal Shifts in Diversification Rates from Molecular Phylogenies. Evolutionary Bioinformatics, 2, 247250. https://doi.org/10.1177/117693430600200024. 937 938 939 940 Rabosky, D. L. (2017). Phylogenetic tests for evolutionary innovation: The problematic link between key innovations and exceptional diversification. Philosophical Transactions of the Royal Society B: Biological Sciences, 372(1735). https://doi.org/10.1098/rstb.2016.0417 941 942 943 Rabosky, D. L., & Goldberg, E. E. (2017). FiSSE: A simple nonparametric test for the effects of a binary character on lineage diversification rates. Evolution, 71(6), 1432–1442. https://doi.org/https://doi.org/10.1111/evo.13227 944 945 946 Rabosky, D. L., & Goldberg, E. E. (2015). Model Inadequacy and Mistaken Inferences of TraitDependent Speciation. Systematic Biology, 64(2), 340–355. https://doi.org/10.1093/sysbio/syu131 947 948 949 Rabosky, D. L., Grundler, M., Anderson, C., Title, P., Shi, J. J., Brown, J. W., Larson, J. G. (2014). BAMMtools : an R package for the analysis of evolutionary dynamics on phylogenetic trees, 701–707. https://doi.org/10.1111/2041-210X.12199 950 951 Rabosky, Daniel L. 2014. “Automatic Detection of Key Innovations, Rate Shifts, and DiversityDependence on Phylogenetic Trees.” PLoS ONE 9(2). 952 953 954 Reith, M., Baumann, G., Claßen-Bockhoff, R., & Speck, T. (2007). New insights into the functional morphology of the lever mechanism of Salvia pratensis (Lamiaceae). Annals of Botany, 100(2), 393–400. https://doi.org/10.1093/aob/mcm031 31 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 955 956 957 Revell, L. J. (2012). phytools : an R package for phylogenetic comparative biology ( and other things ), Methods in Ecology and Evolution, 3: 217–223. https://doi.org/10.1111/j.2041210X.2011.00169.x 958 959 Ricklefs, R. E. (2007). Estimating diversification rates from phylogenetic information. Trends in Ecology and Evolution, 22(11), 601–610. https://doi.org/10.1016/j.tree.2007.06.013 960 961 962 Sauquet, H., Balthazar, M. Von, Magallón, S., Doyle, J. A., Endress, P. K., Bailes, E. J., … Schönenberger, J. (2017). The ancestral flower of angiosperms and its early diversification. Nature Communications, 8. https://doi.org/10.1038/ncomms16047 963 964 965 Seelanan, T., Schnabel, A., & Wendel, J. F. (1997). Congruence and Consensus in the Cotton Tribe (Malvaceae). Systematic Botany, 22(2), 259–290. https://doi.org/10.2307/2419457 966 967 968 969 Serrano-Serrano, M. L., Perret, M., Guignard, M., Chautems, A., Silvestro, D., & Salamin, N. (2015). Decoupled evolution of floral traits and climatic preferences in a clade of Neotropical Gesneriaceae. BMC Evolutionary Biology, 15(1), 1–12. https://doi.org/10.1186/s12862-015-0527-6 970 971 972 Smith, S. A., & O’Meara, B. C. (2012). treePL: divergence time estimation using penalized likelihood for large phylogenies. Bioinformatics, 28(20), 2689–2690. Retrieved from http://dx.doi.org/10.1093/bioinformatics/bts492 973 974 Smith, S. D. (2010). Using phylogenetics to detect pollinator-mediated floral evolution. New Phytologist, 354–363. 975 976 977 Smith, S. D., & Kriebel, R. (2018). Convergent evolution of floral shape tied to pollinator shifts in Iochrominae (Solanaceae)*. Evolution, 72(3), 688–697. https://doi.org/10.1111/evo.13416. 978 979 980 Soltanipour, M.A, Jamzad, Z., Jalili, A., Mahmoodi, M (2020). The conservation status of Zhumeria majdae, an endemic species of Iran. Iran Nature. https://dx.doi.org/10.22092/irn.2021.121273 981 982 983 Soltis, D. E., Bell, C. D., Kim, S., & Soltis, P. S. (2008). Origin and Early Evolution of Angiosperms. Annals OF THe New York A Cademy of Sciences, 25, 3–25. https://doi.org/10.1196/annals.1438.005 984 985 Soltis, P.S. & D.E. Soltis. 2004. Phylogeny and evolution of the angiosperms. Am. J. Botany 91: 1614–1626. 986 987 988 989 Soltis, D. E., Mort, M. E., Latvis, M., Mavrodiev, E. V., O'Meara, B. C., Soltis, P. S., Burleigh, J. G., & Rubio de Casas, R. (2013). Phylogenetic relationships and character evolution analysis of Saxifragales using a supermatrix approach. American journal of botany, 100(5), 916–929. https://doi.org/10.3732/ajb.1300044 32 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 990 991 992 Soltis PS, Soltis DE. 2014.Flower diversity and angiosperm diversification. In:Riechmann JL, Wellmer F, eds.Flower development: methods and protocols. NewYork, NY, USA: Springer, 85–102. 993 994 Stebbins, G. L. (1970). Adaptive radiation of reproductive characteristics in angiosperms, 4012 i: pollination mechanisms. Annual Review of Ecology and Systematics, 1, 307–326. 995 996 997 Takano, A. (2017). Taxonomic study on Japanese Salvia (Lamiaceae): Phylogenetic position of S. akiensis, and polyphyletic nature of S. lutescens var. intermedia. PhytoKeys, 80, 87–104. https://doi.org/10.3897/phytokeys.80.11611 998 999 1000 Takano, A., & Okada, H. (2011). Phylogenetic relationships among subgenera , species , and varieties of Japanese Salvia L . ( Lamiaceae ), 245–252. https://doi.org/10.1007/s10265010-0367-9 1001 1002 1003 1004 The Angiosperm Phylogeny Group III. (2009). An update of the Angiosperm Phylogeny Group Classification for the orders and families of flowering plants: APG III. Botanical Journal of the Linnean Society, 161(2), 105–121. https://doi.org/10.1111/j.10958339.2009.00996.x 1005 1006 1007 Vamosi, J. C., Vamosi, S. M., Van der Niet, T., & Johnson, S. D. (2011). Factors influencing diversification in angiosperms: At the crossroads of intrinsic and extrinsic traits. American Journal of Botany, 98(3), 460–471. https://doi.org/10.3732/ajb.1000311 1008 1009 1010 Van der Niet, T., & Johnson, S. D. (2012). Phylogenetic evidence for pollinator-driven diversification of angiosperms. Trends in Ecology and Evolution, 27(6), 353– 361.https://doi.org/10.1016/j.tree.2012.02.002 1011 1012 1013 Van Der Niet, T., Peakall, R., & Johnson, S. D. (2014). Pollinator-driven ecological speciation in plants: New evidence and future perspectives. Annals of Botany, 113(2), 199–211. https://doi.org/10.1093/aob/mct290 1014 1015 1016 Wagstaff, S. J., Olmstead, R. G., & Cantino, P. D. (1995). Parsimony analysis of cpDNA restriction site variation in subfamily Nepetoideae (Labiatae). American Journal of Botany, 82(7), 886–892. https://doi.org/10.1002/j.1537-2197.1995.tb15705.x 1017 1018 1019 Wagstaff, S. J., Olmstead, R. G., & Cantino, P. D. (1995). Parsimony analysis of cpDNA restriction site variation in subfamily Nepetoideae (Labiatae). American Journal of Botany, 82(7), 886–892. https://doi.org/10.1002/j.1537-2197.1995.tb15705.x 1020 1021 1022 Walker, J. B., & Sytsma, K. J. (2007). Staminal evolution in the genus Salvia (Lamiaceae): Molecular phylogenetic evidence for multiple origins of the staminal lever. Annals of Botany, 100(2), 375–391. https://doi.org/10.1093/aob/mcl176 1023 1024 1025 1026 Walker, J. B., Sytsma, K. J., Treutlein, J., & Wink, M. (2004). Salvia (Lamiaceae) is not monophyletic: Implications for the systematics, radiation, and ecological specializations of Salvia and tribe Mentheae. American Journal of Botany, 91(7), 1115–1125. https://doi.org/10.3732/ajb.91.7.1115 33 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1027 1028 1029 1030 Walker, J. B., Drew, B. T., & Sytsma, K. J. (2015). Unravelling Species Relationships and Diversification within the Iconic California Floristic Province Sages (Salvia Subgenus Audibertia, Lamiaceae). Systematic Botany, 40(3), 826–844. https://doi.org/10.1600/036364415X689285 1031 1032 1033 Wester, P., & Claßen-Bockhoff, R. (2007). Floral diversity and pollen transfer mechanisms in bird-pollinated Salvia species. Annals of Botany, 100(2), 401–421. https://doi.org/10.1093/aob/mcm036 1034 1035 1036 Wester, P., & Claßen-Bockhoff, R. (2011). Pollination Syndromes of New World Salvia Species with Special Reference to Bird Pollination 1. Annals of the Missouri Botanical Garden, 98(1), 101–155. https://doi.org/10.3417/2007035 1037 1038 1039 1040 Wester, P., Cairampoma, L., Haag, S., Schramme, J., Neumeyer, C., & Claßen-Bockhoff, R. (2020). Bee exclusion in bird-pollinated salvia flowers: The role of flower color versus flower construction. International Journal of Plant Sciences, 181(8), 770–786. https://doi.org/10.1086/709132 1041 1042 1043 1044 1045 Wessinger, C. A., Rausher, M. D., & Hileman, L. C. (2019). Adaptation to hummingbird pollination is associated with reduced diversification in PenstemLandis, J. B., Bell, Charles D.Wessinger, C. A., Rausher, M. D., & Hileman, L. C. (2019). Adaptation to hummingbird pollination is associated with reduced diversifi. Evolution Letters, 3(5), 521–533. https://doi.org/https://doi.org/10.1002/evl3.130 1046 1047 1048 Will, M., & Claßen-Bockhoff, R. (2014). Why Africa matters: Evolution of Old World Salvia (Lamiaceae) in Africa. Annals of Botany, 114(1), 61–83. https://doi.org/10.1093/aob/mcu081 1049 1050 1051 Will, M., & Claßen-Bockhoff, R. (2017). Time to split Salvia s.l. (Lamiaceae) – New insights from Old World Salvia phylogeny. Molecular Phylogenetics and Evolution, 109, 33–58. https://doi.org/10.1016/j.ympev.2016.12.041 1052 1053 Will, M., Schmalz, N., & Classen-bockhoff, R. (2015). Towards a new classification of Salvia s. l.: (re) establishing the genus Pleudia Raf ., 1–15. https://doi.org/10.3906/bot-1405-34 1054 1055 1056 Wu, H., Ma, P.-F., Li, H.-T., Hu, G.-X., & Li, D.-Z. (2021). Comparative plastomic analysis and insights into the phylogeny of Salvia (Lamiaceae). Plant Diversity, 43(1), 15–26. https://doi.org/https://doi.org/10.1016/j.pld.2020.07.004 1057 1058 1059 1060 Yao, G., Drew, B. T., Yi, T. S., Yan, H. F., Yuan, Y. M., & Ge, X. J. (2016). Phylogenetic relationships, character evolution and biogeographic diversification of Pogostemon s.l. (Lamiaceae). Molecular Phylogenetics and Evolution, 98, 184–200. https://doi.org/10.1016/j.ympev.2016.01.02 1061 1062 1063 1064 Zanne, A. E. D. C. Tank, W. K. Cornwell, J. M. Eastman, S. A. Smith, R. G. FitzJohn, D. J. McGlinn, B. C. O'Meara, A. T. Moles, P. B. Reich, D. L. Royer, D. E. Soltis, P. F. Stevens, M. Westoby, I. J. Wright, L. Aarssen, R. I. Bertin, A. Calaminus, R. Govaerts, F. Hemmings, M. R. Leishman, J. Oleksyn, P. S. Soltis, N. G. Swenson, L. Warman, J. M. 34 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1065 1066 Beaulieu. 2014. Into the cold - three keys to radiation of angiosperms into freezing environments. Nature doi:10.1038/nature12872 1067 1068 1069 Zhang, B., Claßen-Bockhoff, R., Zhang, Z. Q., Sun, S., Luo, Y. J., & Li, Q. J. (2011). Functional implications of the staminal lever mechanism in Salvia cyclostegia (Lamiaceae). Annals of Botany, 107(4), 621–628. https://doi.org/10.1093/aob/mcr011 1070 1071 1072 1073 1074 1075 35 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1076 1077 1078 1079 1080 Fig 1: Phenotypic diversity in Iranian Salvia. A: Salvia sclarea (clade I), B: Salvia aegyptiaca (clade III) C: Salvia aristata (clade III), D: Salvia macrosiphon (clade I), E: Salvia bracteata (clade I), F: Salvia verticillata (clade I). A-E: Photos by M. Mirtajzadini, F: Photo by K. Safikhani. 1081 1082 1083 1084 36 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Walker and Sytsma (2007) Will and Claßen -Bockhoff (2017) Drew et al. (2017), Hu et al. (2018) Clade I Clade II Clade III Not determined Clade IV 1085 1086 1087 1088 1089 1090 1091 1092 1093 Fig 2. Schematic trees provide a summary of changes in Salvia delimitation based on previous phylogenetic studies (Walker and Sytsma, 2007; Will and Bockhoff, 2017; Drew et al., 2017). Those species that are classified under Salvia infrageneric delimitations are shown in red. Distinct genera from Salvia are indicated in black. Walker and Systma (2007) recognized three distinct clades for Salvia phylogeny embedded within five genera (Perovskia, Rosmarinus, Dorystaechas, Meriandra and Zhumeria). Will and Bockhoff (2017) identified just part of clade I as Salvia sensu stricto and split Salvia into six genera. Drew et al. (2017) maintained Salvia in the broad sense and treated the five genera in Walker and Sytsma (2007) as subgenera of Salvia. Hu et al. (2018) treated clade IV (from eastern Asia) as subg. Glutinaria. 1094 1095 1096 1097 1098 1099 37 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 Clade I Clade II Clade III Clade IV Outgroup 1119 1120 1121 1122 1123 1124 Fig. 3: Maximum Clade Credibility (MCC) obtained from BEAST analysis based on five combined nuclear and plastid spacer regions. The map is colored based on four identified clades in Salvia. The x-axis represents the age range of extant Salvia lineages. The star indicates divergence of the southwestern Asia clade I (Turkey and Iran) including subgenera Sclarea and Salvia from Subgenus Heterosphace 1125 1126 1127 38 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1128 1129 1130 1131 1132 1133 1134 Fig. 4: Ancestral reconstruction of corolla length in Salvia on a dated phylogeny using maximum likelihood in the phytools R package. The legend indicates the range of corolla length in mm by branch color in Salvia. Four distinct clades in Salvia are identified by relevant colors on the circumference of the tree. 1135 1136 1137 39 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1138 1139 1140 1141 Fig. 5: Ancestral reconstruction of the lever mechanism trait (present/ absent) across Salvia phylogeny based on likelihood state with ARD model. 1142 1143 1144 1145 1146 40 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1147 1148 1149 1150 1151 Fig. 6: Ancestral reconstruction of habit (herb/shrub) across Salvia phylogeny using stochastic mapping in phytools. 1152 . 1153 1154 1155 1156 41 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1157 Disparity (Corolla Length) 1158 1159 1160 Geological time 1161 1162 1163 1164 1165 Fig. 7: The mean subclade Disparity Through Time (DTT) for corolla length compared with the median subclade disparity under a Brownian motion model. The solid line shows the observed disparity, and the dashed line is the mean disparity of 1000 simulations of corolla length disparity over the phylogenetic tree. The grey shade indicates the 95% confidence interval of DTT. 1166 1167 1168 1169 1170 1171 1172 1173 42 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1174 1175 clade1176 I clade II clade1177 III clade IV 1178 Outgroup 1179 1180 1181 Fig. 8: Corolla length evolution across Salvia phylogeny based on a BAMM analysis. The best shift was detected in clade II in Calosphace clade 1182 1183 43 Table 1: Plant materials used in this study with their accession numbers. Newly generated sequences are shown in bold. TARI (National Herbarium of Iran, Research Institute of Forests and Rangelands) Lepechinia chamaedryoides (Balb.) Epling (outgroup) Lepechinia leucopylloides (Ramamoorthy, Hiriart & Medrano) B.T.Drew, Cacho & Sytsma, comb. nov. (outgroup) Melissa officinalis L. (outgroup) S. africana-caerulea L. Voucher 1, 5) Drew and Sytsma 2011; 2) Walker and Sytsma 2007; 4) Walker et al., 2004. 1, 2, 4, 5) Drew and Sytsma 2011; 3) Drew and Sytsma 2012. 1, 2, 4, 5) Drew and Sytsma 2011; 3) Drew and Sytsma 2012. Will and Claßen-Bockhoff 2014. ETS ITS Gen Bank rpl32-trnl trnl-trnf Ycf1 JF301317 DQ667231 / AY570459 JF289031 JF301327.1 JF301354 JQ669348 JF301390 JF289047 JF301325 JF301353 KJ584204 JQ669335 Kj7472171 JF301386 / JF289042 / Salvia absconditiflora Greuter & Burdet (syn. S. cryptantha Montbret & Aucher ex Benth.) Will and Claßen-Bockhoff 2017 / KU563839 KU578211 / // Salvia acerifolia B.L.Turner. 2,4) Fragoso-Martinez et al., 2017. / MF664540 / MF663939 / Salvia adenocaulon H.P.Davis Salvia adenophora M.Martens & Galeotti Salvia adenophylla Hedge and Hub.-Mor. Will and Claßen-Bockhoff 2017. / KU563828 KU563828 / / 2,4) Fragoso-Martinez et al., 2017. / MF622100 / MF663940 / Will and Claßen-Bockhoff 2017. / KU563789 KU578218 / / Salvia aegyptiaca L. Iran: Bushehr, TARI (26835). MK204892 / / / / Salvia aegyptiaca. Salvia aethiopis L. Iran: Hormozgan, TARI (102853). Iran: TARI (6604). MK204891 MK204890 MK256969 / / / / / MK240102 Salvia agnes 2,4) Fragoso-Martinez et al., 2017. Salvia akiensis A.Takano, T.Sera et Kurosaki. 1,2,4,5) Takano and Akiyama 2017. MF66454 1 LC060826 44 LC060279 MF663941 / LC124188 LC060530 Salvia albicaulis Benth. Salvia albimaculata Hedge and Hub.-Mor. Salvia amethystina Sm. 1,2,3) Will and Claßen-Bockhoff 2014. KJ584257 KJ584206 KJ747274 / / Will and Claßen-Bockhoff 2017. / KU563790 KU578219 / / 2) Fragoso-Martinez et al., 2017. / MF664545 / / / Salvia amplexicaulis Lam. Salvia anatolica Hamzaog˘lu and Duran Will and Claßen-Bockhoff 2017. / KU563829 KU578151 / / Will and Claßen-Bockhoff 2017. / / KU563840 KU578221 / KP852935 DQ667214 KJ747321 KP852890 KP853066 / KU563791 KU578247 / / Salvia aramiensis Rech.f. 1,4,5) Walker et al., 2015; 2) Walker et al., 2004; 3) Will and Claßen-Bockhoff 2014. Will and Claßen-Bockhoff 2017 Salvia arbuscular Fernald 2,4) Jenks et al., 2010 / HQ418846 / HQ418949 / Salvia argentea L. / KJ584164 KJ747299 / / Salvia aristate Aucher ex Benth. Will and Claßen-Bockhoff 2014. 1) Drew and Sytsma 2011; 2,4) Walker and Sytsma 2007; 5) Drew and Sytsma 2011 / DQ667465 JF289059 Salvia aristata Salvia aspera Fernald. Iran: Isfahan, TARI (12495). 2,4) Fragoso-Martinez et al., 2017. MK204889 / / MF664547 / / MF663948 MK240103 / Salvia atrocyanea Epling. 2,4) Walker and Sytsma 2007. / DQ667270 / DQ667456 / Salvia atropatana Bunge. Iran, TARI (88803) MK204887 MK213193 / Salvia atropatana. Iran, TARI (29283 2,4) Walker and Sytsma 2007; Will and Claßen-Bockhoff 2017 Will and Claßen-Bockhoff 2014. 2,4) Walker and Sytsma 2007; 3) Will and Claßen-Bockhoff 2014. MK204888 / / DQ667286 KU578248 DQ667471 / KJ584261 KJ584218 KJ747276 / / / DQ667323 KJ747261 DQ667512 / 1,3,5) Drew and Sytsma 2011. JF301330 MF664549 MF663950 JF301330 2,4) Walker and Sytsma 2007. / DQ667317 JQ669366. 1 / DQ667317 / specimen_voucher="SBAI / JQ934103 / / / 2,4) Jenks et al., 2010. / HQ418849 / HQ418952 / Salvia apiana Jeps. Salvia aucheri Benth. Salvia aurita L. f. Salvia austriaca Jacq. Salvia axillaris Moc. and Sessé ex Benth. Salvia azurea Michx. ex Vahl. Salvia baimaensis S.W.Su & Z.A.Shen Salvia ballotiflora Benth. D667280 JF301336 45 Salvia bariensis Thulin. Salvia bazmanica Rech.f. & Esfand. Salvia blepharochlaena Bedge and Hub.-Mor. Salvia blepharophylla Hedge & Hub.-Mor. Will and Claßen-Bockhoff 2014. KJ584262 / KJ747316 / Iran, TARI (43049). MK204886 / / / Will and Claßen-Bockhoff 2017. / KU578210 KU578210 / / 2) Jenks et al., 2010. / HQ418850 / HQ418953 / Salvia brachyantha Will and Claßen-Bockhoff 2017 / KU563844 KU578154 / / Salvia brachysiphon Stapf. Iran, TARI (3162). MK204885 MK213194 / / MK240104 Salvia brachysiphon Iran, TARI (1145). MK204884 MK213195 / / MK240105 Salvia bracteata Banks & Sol. Iran, TARI (16642). MK204883 / / MK240107 Salvia brandegeei Munz. Salvia breviflora Moc. & Sessé ex Benth. Salvia brevipes Benth. 1,2,4,5) Walker and Sytsma 2005. KP852949.1 KP852783 / / KP852896 2,4) Fragoso-Martinez et al., 2017. / MF664551 / MF663952 / 2,4) Fragoso-Martinez et al., 2017. / MF664552 / MF663953 / Salvia broussonetii Benth. Will and Claßen-Bockhoff 2014. KJ584263 KJ584225 KJ747293 / / Salvia bucharica M.Popov. Will and Claßen-Bockhoff 2017. / KU563794 KU578222 / / Salvia bulleyana Diels. Will and Claßen-Bockhoff 2017 / / KU578203 / / Salvia cabulica Benth. / DQ667287 / DQ667472 / / HQ418851 / DQ667259 / Salvia cadmica Boiss. Walker and Sytsma 2007. 2) Jenks et al., 2010; 4) Walker and Sytsma 2007. Will and Claßen-Bockhoff 2017. / KU563795 KU578223 / / Salvia caespitosa Montbret & Aucher. Salvia californica Brandegee. Iran, TARI (85247). Drew and Sytsma 2015. MK204882 KP852951 MK213197 DQ667213 / / / DQ667424 / KP853068 Salvia candelabrum Boiss. Salvia candicans M.Martens and Galeotti. Slavia candidissima Vahl. Will and Claßen-Bockhoff 2014. / KJ584190 KJ747255 / / 2,4) Fragoso-Martinez et al., 2017. / MF664557 / MF663958 / Walker and Sytsma 2007 / DQ667261 / DQ667447 Salvia carduacea Benth. Walker et al., 2015. / KP852785 / KP852900 KP853069 Salvia carnea Kunth. 2,4) Jenks et al., 2010. / HQ418854 / HQ418957 / Salvia cacaliifolia Epling 46 / KP853067 Salvia cassia Sam. ex Rech.f. / KU563845 KU578190 / / / KU563781 KC473232 / KC414279 KC414280 / KP852953 DQ667228 / JQ888128 KP853070 MK204881 MK213198 / / MK240107 / HQ418855 / AY570471 / Salvia chamelaeagnea Berg. Will and Claßen-Bockhoff 2017. 2) Will and Claßen-Bockhoff 2017; 4) Wang et al., 2013. 1,4) Wang et al., 2013. 1,5) Walker et al., 2015; 2) Walker and Sytsma 2007; 4) Murphy and Bola 2012. Iran, TARI, 17238B. 2) Jenks et al., 2010; 4) Walker et al., 2004 Will and Claßen-Bockhoff 2014 KJ584268 KJ747289 / / Salvia chienii E. Peter. Will and Claßen-Bockhoff 2014 / KJ584210 KJ584250 Salvia chinensis Benth. PS0121MT04 / FJ883503 / / / Salvia chionantha Boiss. Will and Claßen-Bockhoff 2017 1,2,3) Will and Claßen-Bockhoff 2014; 4) Walker et al., 2004; 5) Walker et al., 2015. / KU563846 KU578155 / / KJ747318 KJ584188 KJ747318 AY570472 KP853071 Salvia chloroleuca Rech.f. & Aellen. Iran, TARI (36026). MK204879 MK212199 / / MK240109 Salvia chloroleuca. Iran, TARI (12691). MK204880 / / / MK240108 Salvia chorassanica Bunge. Salvia chrysophylla Stapf. Salvia cinnabarina M.Martens and Galeotti. Iran, TARI (5354). Will and Claßen-Bockhoff 2017. MK204878 / MK213200 KU563848 / KU578157 / / MK240110 / 2,4) Fragoso-Martinez et al., 2017. / MF664559 / MF663960 / KP853032 MF664560 / MF663961 / / MF664561 / MF663962 / / AY506651 / KC414281 / Salvia castanea Diels Salvia cavaleriei H.Lév. Salvia cedrosensis Greene. Salvia ceratophylla L. Salvia chamaedryoides Cav Salvia chionopeplica Epling. Salvia clevelandii (A. Gray). Salvia clinopodioides Kunth. Slavia coccinea Buc'hoz ex Etl. 1) JBW 3079; 2,4) Fragoso-Martinez et al., 2017. 2,4) Fragoso-Martinez et al., 2017 2) Trusty et al., 2004; 4) Wang et al.,2013. KJ747322 Salvia columbariae Benth. 4) Walker et al.,2015 KP852960. 1 KP852793 / KP852905 KP853073 Salvia compressa Vent. Salvia concolor Lamb. ex Benth. Iran: Hormozgan, TARI (102856) MK204877 MK213201 / / / 2,4) Jenks et al., 2010 / HQ418858 / HQ418961 / 47 Salvia confertiflora Pohl. 2,4) Jenks et al., 2010. / Salvia confertispicata. 2,4) Fragoso-Martinez et al., 2017 / Slavia congestiflora Epling. 2,4) Fragoso-Martinez et al., 2017. / MF664563 MF664564 Salvia connivens Epling. 2,4) Fragoso-Martinez et al., 2017 2) Fragoso-Martinez et al., 2017; 4) Walker et al., 2004. / MF664565 2,4) Fragoso-Martinez et al., 2017. / MF664568 / MF663970 / Will and Claßen-Bockhoff 2017. / KU563849 KU578158 / / 2,4) Wang et al., 2013. / KC473274 / KC414282 Salvia cynica Dunn. Walker and Sytsma 2007. / DQ667332 / DQ667521 Salvia dabieshanesis J.Q.He. PS1723MT01 FJ883505 / / / Salvia daghestanica Sosn. Walker and Sytsma 2007. / KJ584187 KJ747308 DQ667444 / Salvia densiflora Benth 2,4) Fragoso-Martinez et al., 2017. / MF664570 / MF663972 / Salvia deserta omit. Will and Claßen-Bockhoff 2017. / KJ584176 KJ747263 Salvia deserti Decne. Salvia dianthera Roth ex Room, and Schult. 2,3) Will and Claßen-Bockhoff 2014. 2, 4) Walker and Sytsma 2007; 5) Drew and Sytsma 2011. KJ584270 / KJ747312 / / JF301326.1 DQ667329 / DQ667518 JF289044 Salvia dichroantha Stapf. Will and Claßen-Bockhoff 2017 / KU578159 KU56830 Salvia digitaloides Diels. 2) Walker & Sytsma 2007; 4) Walker & Sytsma 2004. / / DQ667255 AY570477 / 2,4) Jenks et al., 2010. 1,2,3) Will and Claßen-Bockhoff 2014; 4) Walker et al., 2004. 2,3) Will and Claßen-Bockhoff 2017; 4) Fragoso-Martinez et al., 2017. / HQ418860 / HQ418963 / KJ584271 KJ584179 KJ747296 AY570478 / / KU563882 KU578197 MF663975 / Will and Claßen-Bockhoff 2017 / / KU578226 / / Salvia corrugata Vahl. Salvia cuspidata (Benth.) J.R.I.Wood. Salvia cyanescens Boiss. and Bal. Salvia cyclostegia E. Peter. Salvia discolor Sessé & Moc. Salvia disermas L. Salvia disjuncta Fernald. Salvia divaricata Montbret & Auch. ex Benth. HQ418859 / HQ418962 / / MF663964 / / MF663965 / / MF663966 / MF622122 48 AY570476 Salvia divinorum Epling & Játiva. / HQ418861 / DQ667440 / KJ584274 DQ667322 KJ747290 DQ667511 / / HQ418862 / HQ418965 / 2,4) Jenks et al., 2010 1,4,5) Walker et al., 2015; 2) Walker and Sytsma 2007 2) Will and Claßen-Bockhoff 2014; 3) Drew and Sytsma 2012; 4) Walker et al., 2004; 5) Drew and Sytsma 2011 / HQ418863 / HQ418966 / KP853037 DQ667229 / KP852907 KP853074 KJ584257 KJ584284 JQ669302 AY570454 JF289014 Salvia dracocephaloides Boiss. Salvia elegans Vahl. Iran, TARI (30192). 2,4) Fragoso-Martinez et al., 2017. MK204876 / Mk213202 MF622127 / / / MF663978 MK240111 / Salvia engelmannii A. Grey. Will and Claßen-Bockhoff 2017 / KU563870 Ku578163 / / Salvia eremophila Boiss. Iran, TARI (41741). 1,4,5) Walker et al., 2015; 2) Walker and Sytsma 2007. MK204875 / / / MK240112 KP853039 DQ667232 / KP852910 KP853075 Will and Claßen-Bockhoff 2017 / KU563850 KU578167 Jenks et al., 2010. / HQ418864 / HQ418967 / Dizkirici et al., (2015). / KM519756 KU578227 KM519770 / / KJ584251 KJ747323 FJ593462.1 / / MF622128 / MF663979 / / FJ546871 / AY570479 / / MF664576 / MF663980 / Salvia dolomitica Codd. Salvia dombeyi Epling. Salvia dorisiana Standl. Salvia dorrii (Kellogg.) Abrams. Salvia dorystaechas B. T. Drew Salvia eremostachya Jeps. Salvia eriophora Boiss. and Kotschy. Salvia erythrostoma Rusby. Salvia euphratica Montbret and Aucher ex Benth. 2,4) JBW 2330 1,3) Will and Claßen-Bockhoff 2014; 2,4) Walker and Sytsma 2007; 2,4) Jenks et al., 2010 Salvia filifolia Ramamoorthy. 2,3) Will and Claßen-Bockhoff 2014; 3) Zhong et al., 2010. 2,4) Fragoso-Martinez et al., 2017 2) Chen et al., 2010; 4) Walker et al., 2004. 2,4) Fragoso-Martinez et al., 2017. Salvia flocculosa Benth. 2,4) Fragoso-Martinez et al., 2017. / MF664578 / MF663982 / Salvia formosa L'Hér. Salvia frigida Boiss. 2) Fragazo –Martinez et al., 2017. Will and Claßen-Bockhoff 2017. / / MF622131 KU563851 / KU578168 / / / / Salvia fruticosa Miller. Will and Claßen-Bockhoff 2014. / KJ584195 KJ747256 / Salvia evansiana Hand.-Mazz. Salvia exserta Grieseb. Salvia farinacea Benth. 49 Salvia fulgens Cav. Salvia funerea M. E. Jones. 2) Walker and Sytsma 2007; 4) BenitezVieyra et al., 2014. Walker et al., 2015. / MF622133 / KJ473988 / KP853041 KP852812 / KP852911 KP853076 Salvia galloana B.L.Turner. Salvia garipensis E.Meyer ex Benth Salvia geminata Thulin. Salvia gesneriiflora Lindl. & Paxton. Salvia glabrescens var. repens (Koidz.) Kurosaki. 2,4) Fragoso-Martinez et al., 2017. / MF664581 / Walker and Sytsma 2007. / DQ667281 / DQ667466 / 1) Will and Claßen-Bockhoff 2014. KJ584276 / / / / 2,4) Fragoso-Martinez et al., 2017. / MF622133 / MF663986 / Sudarmono and Okada 2007. LC060829.1 / AB295089 LC060533 Salvia glutinosa L. Iran, TARI (21565). 2) Will and Claßen-Bockhoff 2014; 4) Walker et al., 2004; 5) Drew and Sytsma 2011. MK204873 MK213203 / / Mk240113 / KU563774 / AY570480 JF289061 2.4) Walker and Sytsma 2007. / DQ667276 / DQ667461 / / HQ418868 / HQ418971 / JF301331 DQ667215 JF289062 AY570481 JF289062 Salvia greggii A.Gray. 2,4) Jenks et al., 2010. 1,5) Drew and Sytsma 2011; 2) Walker and Sytsma 2007; 3) Drew and Sytsma 2012; 4) Walker et al., 2004. 2,4) Jenks et al., 2010. / HQ418870 / HQ418972 / Salvia grewiifolia S. Moore. 2,4) Jenks et al., 2010. / HQ418871 / HQ418973 / Salvia grossheimii Sosn. Salvia guadalajarensis Briq. Iran, TARI (84031). 2,4) Fragoso-Martinez et al., 2017. MK204832 / MK213204 MF664584 / / / MF663989 / / Salvia guaranitica Briq. 2,4) Wang et al., 2013. / KC473237 / KC414285 Salvia handelii E.Peter. SHAN JQ934124 / / / Salvia hayatana Makino ex Hayata. Salvia heerii Regel. Salvia heldreichiana Boiss. ex Bentham 2,4) Sudarmono and Okada 2007; 5) Takano 2017. 2,4) Fragoso-Martinez et al., 2017. / AB295099 / AB295084 / / MF664587 / / KU563799 KU578246 Salvia glutinosa. Salvia gracilliramulosa Epling & Játiva. Salvia gravida A.Gray. Salvia greatae Brandegee. Will and Claßen-Bockhoff 2017. 50 AB295104 MF663985 / / / / Salvia henryi Gray. Will and Claßen-Bockhoff 2017. / KU563875 KU578165 / / Salvia herbacea Benth. Salvia herbanica A.Santos and M.Ferna´ndez. Salvia heterochroa E. Peter. 2,4) Fragoso-Martinez et al., 2017. / MF664589 / MF663994 / Will and Claßen-Bockhoff 2014. KJ584278 KJ584246 KJ747313 / / 2,3) Will and Claßen-Bockhoff 2014. / KJ584252 KJ747324 / / Salvia heterochroa E.Peter. Will and Claßen-Bockhoff 2017 / KJ584252 KJ747324 / / 2,4) Fragoso-Martinez et al., 2017. / MF664590 / MF663995 / Salvia heterofolia Epling & Mathias. Salvia hians Royle ex Benth. 4) Walker et al., 2004. / DQ763239 / AY570483 / Salvia hidalgensis Miranda. 2,4) Fragoso-Martinez et al., 2017. / MF664591 / MF663996 / Salvia hintonii Epling. 2,4) Fragoso-Martinez et al., 2017. / MF664592 / MF663997 / Salvia hirtella Vahl 2,4) Walker and Sytsma 2007. / DQ667326 / DQ667515 / Salvia hispanica L. 2,4) Fragoso-Martinez et al., 2017. / MF664593 / MF663998 / Salvia honaniaL.H.Bailey. PS1722MT01 / FJ883513 / / / Salvia huberi Hedge. Will and Claßen-Bockhoff 2017. / KU563800 KU578228 / / Salvia hylocharis Diels Salvia hypargeia Fisch. and Mey. Wang et al., 2013. / KC414286 / KC473238 / Will and Claßen-Bockhoff 2017. / KU563876 KU578196 / / Salvia hypochionaea Boiss. Iran, TARI (30437). MK204870 MK213206 / / / Salvia hypoleuca Benth. Iran, TARI (54151). MK204869 MK213207 / / MK240115 Salvia indica L. Salvia interrupta Schousb. Iran, TARI (90002). Will and Claßen-Bockhoff 2014. MK204868 / MK213208 KJ584191 / KJ747265 / / MK240116 Salvia isensis Nakai ex Hara. Takano and Akiyama 2017. LC060831.1 LC060730 / LC124190 LC060535 Salvia jamzadii Mozaff. Salvia japonica f. longipes (Nakai) Sugimoto. Salvia judaica Boiss. Iran, TARI (61992). 1,4,5) Sudarmono and Okada (2007); 2) Takano and Akiyama 2017. Will and Claßen-Bockhoff 2017. MK204867 MK213209 LC060835.1 / AB266239 KJ584241 / KU578160 LC124191 / Salvia jurisicii Košanin. Will and Claßen-Bockhoff 2017 / KU563831 KU578173 / / Salvia karwinskii Benth. 2,4) Fragoso-Martinez et al., 2017 / MF622144 / MF664003 / Salvia kiaometiensis H. Lév. 2,4) Wang et al., 2013 / KC473239 / KC414287 / 51 MK240117 LC060537 / Salvia Koyamae Makino. 2,4) Takano and Okada 2011 AB541114 AB541142 / LC060540 Salvia kronenburgii Rech. f. Salvia kurdica Boiss. and Hohen ex Benth. S. Slavia lasiantha Benth. Will and Claßen-Bockhoff 2017 / KM519759 KU578245 KM519773 / Will and Claßen-Bockhoff 2017 / KU563821 KU578212 / / 2,4) Walker and Sytsma 2007. / DQ667300 / DQ667486 / Salvia lachnocalyx Hedge. Salvia lachnostachys Benth. Iran, TARI (83023). 2,4) Fragoso-Martinez et al., 2017 MK204866 / MK213210 MF664598 / MF664005 MK240118 / Salvia lanceolata Lam. 2,3) Will and Claßen-Bockhoff 2014 / KJ584201 KJ747277 / / Salvia leptostachys Benth. 2,4) Fragoso-Martinez et al., 2017 / MF664603 / MF664010 / Salvia leriifolia Benth. Salvia leucantha Cav. Iran, TARI (35583). 2,4) Jenks et al., 2010. MK204865 / MK213211 HQ418875 / HQ418977 / / Salvia leucodermis Baker. Will and Claßen-Bockhoff 2014. KJ747280 KJ584220 KJ747280 / / Salvia leucophylla Greene. 2) Walker and Sytsma 2007; 4) Walker et al., 2004 / Salvia limbate C.A.Mey. Iran, TARI (27761). MK204862 MK213212 Salvia limbata Iran, TARI (30364). MK204864 MK213213 Salvia limbata Salvia littae Vis. Salvia longispicata M.Martens & Galeotti. Salvia lophanthoides Fernald. Iran, TARI (85267). 2,4) Fragoso-Martinez et al., 2017. MK204863 / / MF622150 / / / MF664014 / / 2,4) Jenks et al., 2010. / HQ418876 / HQ418978 / 2,4) Fragoso-Martinez et al., 2017. / MF664607 / MF664017 / Salvia lutea L. Salvia lutescens var. lutescens Koidz. Will and Claßen-Bockhoff 2014. Takano and Akiyam 2017; Sudarmono and Okada 2007. KJ747273 KJ584205 KJ747273 / / LC060845.1 / AB266232 / Salvia lyrata L. Will and Claßen-Bockhoff 2017. / AB266232 KU563873. 1 KU578166 / / Salvia macilenta Boiss. Salvia macrochlamys Boiss. & Kotschy Salvia macrophylla Bent. Iran, TARI (102851). MK204861 MK213214 / / MK240122 Iran, TARI (102852) 2,4) Jenks et al., 2010. MK204860 / MK213215 HQ418877 / / / HQ418979 MK240123 DQ667210 / / KP852913 / KP853077 MK240120 Mk24011 2,4) 52 Salvia macrosiphon Boiss. Salvia madrensis Seem. Salvia majdae (Rech.f. & Wendelbo) Sytsma. Salvia marashica Ilçim, Celep and Dogan. Salvia margaritae Botsch. Iran, TARI (58399). 2,4) Jenks et al., 2010. Iran: Hormozgan, Geno Mirtajzadin 201. MK204859 / MK213216 HQ418878 / / / HQ418980 MK240124 / MK204858 MK256967 / / ‫ا‬/ Will and Claßen-Bockhoff 2017. / KU563802 KU578230 / / 2,3) Will and Claßen-Bockhoff 2017. KU563880 KU578201 / / Salvia maximowicziana Hemsl. 1,3,4) Deng et al., 2015. KM886617 KM886851 KM886650 Salvia maymanica Hedge. Will and Claßen-Bockhoff 2017. / / PS1730MT 01 KU563805 KU578231 / / PS1719MT01 / FJ546867 / / / Salvia meilienis S.W.Su. mountain, KP852989 HQ418879 JQ669368 KP852916 JF289064 Salvia merjamie Forsk. 1,4) Walker et al., 2015; 2) Jenks et al., 2011; 3) Drew and Sytsma 2012; 5) Drew et al., 2011 Will and Claßen-Bockhoff 2014. KJ584286 KJ584184 KJ747297 / / Salvia mexicana L. 2,4) I. Fragoso-Martinez 79 (FCME). / MF664611 / MF664021 / Salvia microphylla Kunth Salvia microstegia Boiss. & Balansa Salvia minarum Briq. Salvia mirzayanii Rech.f. & Esfand. Salvia misella Kunth. 2,4) Fragoso-Martinez et al., 2017. / MF664022 / MF663986 / Iran, TARI (5397). 2,4) Fragoso-Martinez et al., 2017 MK204857 / MK213216 MF664613 / / / MF664023 / / MK204856 / MK213217 MF664614 / / / / / / / MF664615 / DQ667459 / / KU563856 KU578175 / / KP852997 DQ667212 / KP852920 KP853078 Salvia moniliformis Fernald. Iran, TARI (41724). 2,4) Fragoso-Martinez et al., 2017 2) Fragoso-Martinez et al., 2017; 4) Walker and Sytsma 2007 Will and Claßen-Bockhoff 2017 2) Walker and Sytsma 2007; 1,4,5) Walker et al., 2015 2,3) Will and Claßen-Bockhoff 2017 / KU563884 KU578198 / / Salvia montbretii Benth. Will and Claßen-Bockhoff 2017 / KU563869 KU578195 / / Salvia mellifera E. Greene. Salvia mocinoi Benth. Salvia modesta Boiss. Salvia mohavensis E. Greene. 53 Salvia muirii L.Bolus. Will and Claßen-Bockhoff 2014. KJ584287 KL584208 KJ747283 / / Salvia multicaulis Vahl. / MK213218 / / / KP853000 DQ667224 / MF664026 KP853079 Salvia namaensis Schinz. Iran, TARI (102845) 2) Walker and Sytsma 2007; 1,5) Walker et al., 2015; 4) Fragoso-Martinez et al., 2017. Will and Claßen-Bockhoff 2014. KJ584289 KJ584200 KJ747284 / / Salvia nana Kunth. 2,4) Fragoso-Martinez et al., 2017 / MF664618 / MF664029 / Salvia nemorosa L. Salvia nervosa Benth. Iran, TARI (43572) 2,4) Fragoso-Martinez et al., 2017. 1,2,3) Will and Claßen-Bockhoff 2014; 4) Walker and Sytsma 2004 1,5) Takano and Akiyam 2017; 2) Takano and Okada 2011 MK204855 / MK213219 MF664619 / / / MF664031 MK240125 / KJ747281 KJ584229 KJ747258 AY570487 / LC060848 AB295101 / / 2,4) Fragoso-Martinez et al., 2017 / MF664620 / MF664032 / KU563786 KU578205 / / Salvia munzii Epling. Salvia nilotica Juss. ex Jacq. Salvia nipponica Miq. var. kisoensis Salvia nitida (M.Martens & Galeotti) Benth. Salvia nubicola Wall. ex Sweet 2,4) Will and Claßen-Bockhoff 2017 LC060552 Salvia nutans L. Will and Claßen-Bockhoff 2017 / KU563832 KU578176 / / Salvia nydeggeri Hub.-Mor. Will and Claßen-Bockhoff 2017 / KU563803 KU578233 / / Salvia oaxacana Fernald. Salvia occidentalis Sw. 3) Will and Bockhoff 2017. / / / HQ418882 KU578199 / HQ418983 HQ418984 / / JF301332 DQ667225 JF301398 JF289065 Iran: Zanjan, Mirtajzadini Voucher: SOME 1,5) Takano 2017; 2,4) Takano and Okada 2011 MK204853 / MK256968 JQ934139 EU200176 / MK240126 / LC060852.1 AB353205 / AB353195 LC060557 2,4) Jenks et al., 2010 / HQ418883 / HQ418985 / Salvia officinalis L. Salvia oligophylla Aucher ex Benth. Salvia omeiana E. Peter. Salvia omerocalyx Hayata var omerocalyx Salvia oppositiflora Ruiz & Pav. 2,4) Jenks et al., 2010. 1, 4, 5) Drew and Sytsma 2011; 2) Walker and Sytsma 2007 54 2) Fragoso-Martinez et al., 2017; 4) Walker and Sytsma 2007 / DQ667279 / MF664033 / 4) Walker and Sytsma 2007 / DQ667315 / DQ667502 / 2,4) Jenks et al., 2010 1,5) Walker et al., 2015; 2) Walker and Sytsma 2007 Will and Claßen-Bockhoff 2014 2,4) Fragoso-Martinez et al., 2017 / HQ418884 / HQ418986 / KP853004 / / DQ667431 KP853080 / / KJ584175 MF664623 KJ74304 / / MF664035 / / / KC473252 / KC414297 / JF301333 HQ418885 JQ669370 DQ667442 JF289066 Salvia pauciflora Kunth. Salviapentstemenoides K.Koch and C.D.Bouché. Salvia perlonga Fernald. 2,4) Wang et al., 2013 1,5) Drew and Sytsma 2011; 2) Jenks et al. 2010; 3) Drew and Sytsma 2012; 4) Walker and Sytsma 2007 2,4) Wang et al., 2013. 3) Will and Claßen-Bockhoff 2014; 2,4) Walker and Sytsma 2007. 2,4) Fragoso-Martinez et al., 2017. / KC473254 / KC414299 / / DQ667221 KU578162 AY570489 / / MF664627 / MF664040 / Salvia personata Epling. 2,4) Walker and Sytsma 2007. / DQ667269 / DQ667455 / Salvia perspolitana Boiss. Salvia phlomoides Asso. Iran, TARI (102854). Will and Bockhof 2014. MK204852 / MK213220 KJ584186 KJ747309 / MK240127 / Salvia pinnata L. Will and Claßen-Bockhoff 2017. / KU563798 KU578217 / / Salvia platystoma 2,4) Walker and Sytsma 2007. 1,5) Takano 2017; 2) Will and ClaßenBockhoff 2017; 3) Wang et al., 2013. 2,3) Will and Claßen-Bockhoff 2017. / DQ667277 / DQ667462 / LC060859.1 / KU563788 KU563787 / KU578207 KC414300 / LC060563 / Iran, TARI (102848). 1,5) Drew and Sytsma 2011; 2,4) Fragoso-Martinez et al., 2017; 3) Drew and Sytsma 2012. MK204851 MK213221 / / / JF301334 MF664631 JQ669371 MF664044 JF289067 Will and Claßen-Bockhoff 2017. / KU563807 KU578235 / / Will and Claßen-Bockhoff 2017. / KU563835 KU578180 / / Salvia orbignaei Benth. Salvia ovaliifolia A.St.-Hil. ex Benth. Salvia oxyphora Briq. Salvia pachyphylla Munz. Salvia palaestina Benth. Salvia pallida Benth. Salvia paohsingensis C.Y.Wu. Salvia patens Cav. Salvia plebeia R.Br. Salvia plectranthoides Griff. Salvia poculata Nábelek. Salvia polystachya Cav. Salvia potentillifolia Boiss. and Heldr. ex Benth. Salvia pratensis L. 55 Salvia prattii Hemsl. 2,3) Will and Claßen-Bockhoff 2017. / KU578206 / KU563784 / Salvia priontis Hence. specimen_voucher="PS1711MT01 / FJ883527 / / / Salvia procurrens Benth. / MF664633 / MF664046 / / DQ667275 / MF664048 / Salvia przewalskii Maxim. Salvia pterocalyx Hedge. 2,4) Fragoso-Martinez et al., 2017. 2) Walker and Sytsma 2007; 4) FragosoMartinez et al., 2017. 1,5) Drew and Sytsma 2011; 2,4) Walker and Sytsma 2007; 3) Drew and Sytsma 2012. 3) Will and Claßen-Bockhoff 2017. JF301339.1 / DQ667254 / JQ669372 KU578200 DQ667443 / JF289068 / Salvia pubescens Benth. 2,4) Walker and Sytsma 2007. / DQ667296 / DQ667482 / Salvia purpurea Sessé & Moc. 2,4) Benitez-Vieyra et al., 2014. 1, 5) Takano and Akiyama 2017; 2) Takano and Okada 2011; 4) Sudarmono and Okada 2007. / MF664636 / KJ473981 / AB295083 LC060558 LC060854.1 AB541126 / specimen_voucher="SQIM" / / / / Will and Claßen-Bockhoff 2017. / KU563808 KU578249 / / KJ584293 KJ584180 / / / / AB287375 / AB287374 LC060560 Salvia repens Burch. ex Benth. 1,2) Will and Claßen-Bockhoff 2014. 2,4) Sudarmono and Okada (2007); 5) Takano 2017. Will and Claßen-Bockhoff 2014. KJ584295 KJ584231 KJ747282 / / Salvia retinervia Briq. 2,4) Fragoso-Martinez et al., 2017 / MF664058 / MF664058 / Salvia reuterana Boiss. Iran, TARI (102849). MK204850 MK213222 / / Salvia reuterana. Iran, TARI (102842). MK204849 / / / / Salvia rhytidea Benth. Iran, TARI, (102850). MK204848 MK213223 / / MK240129 Salvia rhytidea. Salvia ringens Sm. Iran, Bahonar University (16297). Will and Claßen-Bockhoff 2017. MK204847 / / KU563810 / KU578213 / / MK240128 / Salvia roborowskii Maxim. 2)Walker and Sytsma 2007. 1,5) Drew and Sytsma 2011; 2) walker and Sytsma 2007. / DQ667289 / DQ667474 / JF301340 DQ667211 / / / Salvia prunelloides Kunth. Salvia pygmaea Matsum. Salvia qimenensis S.W.Su & J.Q.He. Salvia quezelii Hedge and Afzal-Rafi Salvia radula Benth. Salvia ranzaniana Makino. Salvia roemeriana Scheele. 56 JQ934155 Salvia rosifolia Sm. Salvia rosmarinus (L.) Schield., Handb. Med.-Pharm Salvia rufula Kunth. Will and Claßen-Bockhoff 2017 / 1,5) Drew and Sytsma 2011; 2) Trusty et al. 2004; 3) Drew and Sytsma 2012 4) Walker et al., 2004. 2,4) Fragoso-Martinez et al., 2017. KU578209 / / / JF30328 AY506649 JQ669364 AY570465 JF289058.1 / MF622179 / MF664063 / DQ667475 Salvia rugosa 2,5) Walker & Sytsma 2007 / DQ667290 / Salvia rusbyi Britton ex Rusby. 2,4) Walker and Sytsma 2007. / DQ667266 / DQ667452 / Salvia russellii Benth. Salvia rypara Briq. Salvia rzedowskii Ramamoorthy Salvia sagittata Ruiz & Pav. Salvia sahendica Boiss. & Buhse Iran, TARI (86040). 2,4) Walker and Sytsma 2007. MK204846 / MK213224 DQ667266 / / / DQ667452 MK240130 / 2,4) Fragoso-Martinez et al., 2017. / MF664650 / MF664067 / 2,4) Walker and Sytsma 2007. / DQ667446 / DQ667260 / Iran, TARI (73990). MK204845 MK213225 / / / Salvia santalonifolia Boiss. Salvia scabiosifolia Lam. Iran, TARI (102846). MK204844 / MK213226 Will and Claßen-Bockhoff 2017 / KU578237 / / / / Salvia scabra L. Will and Claßen-Bockhoff (2014) KJ584297 KJ747285 / / Salvia scapiformis Hance 2,4) Deng et al., 2015 Salvia schimperi Benth. Will and Claßen-Bockhoff 2014 KJ584298 KJ584233 KM886675 1 KJ584174 / / / Salvia schlechteri Briq, Will and Claßen-Bockhoff 2014 2) Walker and Sytsma 2007; 3,5) Drew and Sytsma 2012 KJ584299 KJ584235 KJ747286 / / / DQ667222 JQ669373 / JQ669265 Iran, TARI (69493). MK204843 MK213227 / / MK240131 Iran, TARI (3533). MK204842 MK213228 / / MK240131 DQ667330 JQ669352 DQ667519 JF289051 Salvia sclarea L. Salvia sclarea. Salvia sclareopsis Bornm. ex Hedge Salvia scrophularifolia (Bunge) B. T. Drew Salvia scutellaroides Kunth. 2,4) Walker and Sytsma 2007; 3) Drew and Sytsma 2012; 5) Drew and Sytsma 2011 2,4) Walker and Sytsma 2007 / / 57 KU563811 DQ667327 / KM886852 DQ667516 Salvia sericeo-tomentosa Rech. F. Salvia sessei Benth. Will and Claßen-Bockhoff 2017 / KU563822 KU578238 / / 2,4) Jenks et al., 2010 / HQ418889 / HQ418991 / Salvia sessilifolia Baker. 3) Walker and Sytsma 2007; KJ584303 DQ667282 DQ667467 / / Salvia sinica Migo. SS201301 / KJ397257 / / / Salvia somalensis Vatke. Will and Claßen-Bockhoff 2014 1,4) Walker and Sytsma 2007; 2,5) Walker et al., 2015 2,4) Fragoso-Martinez et al., 2017 KJ584304 KJ584240 KJ747311 / / KP853008 DQ667218 / DQ667426 KP853081 / MF664655 / MF664073 / Iran, TARI (102844) MK204841 MK213229 / / MK240132 Iran, TARI (1506) MK204872 MK21204 2,4) Fragoso-Martinez et al., 2017 / MF622186 / MF664076 / 2,4) Walker and Sytsma 2007 / DQ667267 / DQ667453 / Iran, TARI (51624) MK204840 MK213230 / / / Will and Claßen-Bockhoff 2014 KJ584305 KJ584237 KJ747260 / / 4) Jenks et al., 2010 / HQ418891 / HQ418993 / / / / / Salvia sonomensis E. Greene. Salvia sophrona Briq. Salvia spinosa L. Salvia spiraeifolia Boiss. & Hohen. Salvia splendens Sellow ex Schult. Salvia stachydifolia Benth. Salvia staminea Montbret & Aucher ex Benth. Salvia stenophylla Burch. ex Benth. Salvia styphelus Epling Salvia suffruticosa Montbret & Aucher ex Benth. Salvia summa A.Nelson. Iran, TARI (86522) 1) Will and Claßen-Bockhoff 2014; 2,3) Will and Claßen-Bockhoff 2017. Salvia superba voucher="PS0134MT01 Salvia syriaca L. Salvia×sylvestris L. Salvia taraxacifolia Hook.f. Salvia tebesana Bunge. Salvia texana (Scheele) Torr. / MK213223 KJ584307 / KU563874 FJ546849 KU578164 / Iran, TARI (16699). 2) Will and Claßen-Bockhoff 2017; 3,4) Wang et al., 2013. 1,2,3) Will and Claßen-Bockhoff 2014; Walker et al., 2004. MK204839 MK213232 / / KJ584177 KJ747292 KC414323 KJ584308 KJ584228 KJ747270 AY570497 / Iran, TARI (84759). Will and Claßen-Bockhoff 2017. MK204838 / MK213233 KJ584199 / KJ747267 / DQ667510 / / / 58 / Salvia thymoides Benth P. Wester and R. Claßen-Bockhoff 336 (MJG 041398). 2,4) Walker and Sytsma 2007. Salvia thyrsiflora Benth. 2,4) Fragoso-Martinez et al., 2017. / MF664668 / MF664088 / Salvia tiliifolia Vahl. 2,4) Fragoso-Martinez et al., 2017. / MF664669 / MF664089 / Salvia tobeyi Hedge. Will and Claßen-Bockhoff 2017. / KU563865 KU578188 / / Salvia tomentosa Mill. Salvia tonaticensis Ramamoorthy ex Lara, Bedolla et Zamudio. Salvia tortuosa Urb. Will and Claßen-Bockhoff 2017. / KU563816 KU578214 / / 2,4) Fragoso-Martinez et al., 2017. / MF664670 / MF664090 / 2,4) Jenks et al., 2010. / HQ418893 / HQ418995 / Salvia trichocalycina Benth. Walker & Sytsma 2007 / DQ667283 / DQ667468 / Salvia trichoclada Benth. Will and Claßen-Bockhoff 2017. / / KU578243 / / Salvia trichostephna Epling. 2,4) Fragoso-Martinez et al., 2017. / MF664671 / MF664091 / Salvia tricuspis Franch. China: Sichuan / EF373633 / EU220730 / / KU563779 / KC414322.1 / / HQ418894 / HQ418996 / Salvia thermarum Van Jaarsv. KJ584309 / KJ747288 / / / DQ667273 / DQ667458 / Salvia tubifera Cav. 2) Will and Claßen-Bockhoff 2017; 4) Wang et al 2013 2,4) Fragoso-Martinez et al., 2017. Salvia tubiflora Sm. 2,4) Fragoso-Martinez et al., 2017. / MF664672 / MF664092 / Salvia uliginosa Benth. 2,4) Jenks et al., 2010. / HQ418895 / HQ418997 / Salvia urmiensis Bunge. Salvia vaseyi (Porter) Parish. Iran, TARI (19586). 4)Walker et al., 2015. MK204837 KP853013 MK213234 / / / / KP852932 / KP853083 Salvia veneris Hedge Will and Claßen-Bockhoff 2017. / KJ584170 KJ747306 / / Salvia venulosa Epling 2,4) Jenks et al., 2010. / HQ418896 / HQ418998 / Salvia verbascifolia M.Bieb. Salvia verbenaca L. Salvia vermifolia Bedge and Huber-Morath. Iran, TARI (88803). Will and Claßen-Bockhoff 2014. MK204831 / MK213235 KJ584183 KJ747298 / / / Will and Claßen-Bockhoff 2017. / Ku563866 KU578192 / / Salvia verticillata L. Iran, TARI (2765). MK204836 MK213236 / / / Salvia virgata Jacq. Iran, TARI (3772). MK204835 MK213237 / / / Salvia trijuga Diels. 59 Salvia viridis L. Salvia viscosa Jacq. Iran, TARI (13283). Will and Claßen-Bockhoff 2017. MK204834 / MK213238 KU563838 / KU578186 / / MK240135 / Salvia vvedenskii Nikitina . 2,3) Will and Claßen-Bockhoff 2017. 1,2,3) Will and Claßen-Bockhoff 2014; 4) Walker and Sytsma 2007. / KU563879 KU578202 / / Iran, TARI (69728). 1,5) Drew and Sytsma 2011; 2) Will and Claßen-Bockhoff (2014); 3) Drew and Sytsma 2011. MK204833 MK213239 / JF301328 KJ584242 JQ669352 / DQ132866 KT210283 Salvia whitehousei Alziar Salvia xanthocheila Boiss. ex Benth. Salvia yangii B. T. Drew Salvia yunnanensis C.H.Wright. 2) YunN0309-2; 3) G.X. Hu & al., QT001, 4) H.F. Guo 2017257(PE) 60 KJ584311 KJ584198 KJ747268 DQ667509 / / MK240136 JF289051` EF014356 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Table 2: Rate and modes of Salvia corolla length diversification. Fitted models are Brownian motion (BM), Ornstein-Uhlenbeck (OU) and Early Burst (EB). The best-fit model is estimated based on the lowest bias-corrected Akaike Information Criterion. Model Model Parameters LogL AICc ∆AIC BM OU EB α=0 α = 0.15 α = -0.00001 -1233 -1198.020364 2470 2402 -1233.21 2472 68 0 70 61 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Table 3: Rates of Salvia diversification examining multiple evolutionary models. Models fitted include diversity-dependent linear speciation and extinction (DDL + E), diversity-dependent exponential speciation and extinction (DDE+E) and two constant-rate diversification models: a purebirth (Yule) model and birth-death (crBD) model. λ = speciation rate (Ma/lineage); µ = extinction rate (Ma/lineage); K = carrying capacity; AIC = Akaike Information Criterion (AIC) for testing model fit. The capacity for the potential number of Salvia species is higher than the number of extant species (~1000 spp.), suggesting that current Salvia diversification is independent of diversity. Model Yule CrBD DDE+E DDS+E lambda 0.246 0.290 0.6682 0.3579 mu 0 0.079 0.2331 0.14616 K ------------------------36504.2 3315.75 62 loglik -989.739 -987.888 -993.5864 -988.042 AIC 1977.782 1979.768 1993.169 1982.86 bioRxiv preprint doi: https://doi.org/10.1101/2021.12.16.473009; this version posted December 17, 2021. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Table 4: HiSSE model subsets that were fitted for study of the effect of the lever mechanism on diversification across Salvia phylogeny. The best-fit model shown in bold was selected based on a bias-corrected Akaike Information Criterion (AICc). Active lever mechanism Model lnLik CID-4: q's equal AIC ΔAIC -948.8211 1915.642 77.136 -909.27 1828.541 16.943 -907.778 1825.556 18.575 HiSεε: q0B1B=0, q1B0B=0, All other q's equal -899.79 1811.598 0 HiSSE: τ0A=τ1A, ε's equal, q0B1B=0, q1B0B=0, All other q's equal -906.91 1827.827 16.229 -895.706 1823.521 11.923 HiSSE: τ0A=τ0B, ε0A=ε0B, q0B1B=0, q1B0B=0, All other q's equal -899.44 1812.896 1.3 HiSSE: τ0A=τ0B, ε's equal, q0B1B=0, q1B0B=0, All other q's equal -897.76 1813.521 1.923 -908.375 1828.751 16.377 -905.77 1829.55 17.152 CID-2: q's equal and ε's equal -906.591 1821.312 9.714 HiSSE: τ0A=τ0B, ε0A=ε0B -904.426 1822.853 11.255 HiSSE: τ0A=τ1A, ε0A=ε1A, q's equal -905.968 1825.937 14.339 HiSSE: τ0A=τ1A=τ0B, ε's and q's equal -909.93 1827.878 16.28 HiSSE: τ0A=τ0B, ε's and q's equal -907.96 1825.93 14.332 HiSSE: τ0A=τ1A, ε's and q's equal -908.76 1826.255 14.657 CID-2: q's equal 906.591- 1823.183 11.585 HiSSE: τ0A=τ1A=τ0B, ε0A=ε1A=ε1B, q's equal -907.552 1825.044 13.446 HiSSE: 25 τ0A=τ1A=τ0B, ε0A=ε1A=ε0B, q0B1B=0, q1B0B=0, All other q's equAl BiSSE modεl: q's equal -907.778 1825.556 13.958 -914.4616 1838.923 27.325 -927.820 -.906.786 1867.641 56.043 1841.536 29.938 -913.091 1834.183 22.585 -910.34 1828.69 17.092 -916.701 1843.409 11.255 HiSSE: τ0A=τ1A, ε0A=ε1A, q0B1B=0, q1B0B=0, All other q's equal HiSSE: ε's equal, q0B1B=0, q1B0B=0, All other q's equal HiSSE full modεl HiSSE: q's and ε's equal HiSSE: q's equal CID-4: ε's and q's equal BiSSE modεl: q's equal, ε0=ε1 BiSSE modεl: All fτee HiSSE τ0A=τ1A=τ0B, ε's equal, q0B1B=0, q1B0B=0, All other q's equAl BiSSE modεl: ε0=ε1 63