Author Manuscript Paul Peterson ORCID iD: 0000-0001-9405-5528 Research Article A biogeographical analysis of Muhlenbergia (Poaceae: Chloridoideae: Cynodonteae: Muhlenbergiinae) Running title: Peterson et al.: Biogeography of the Muhlenbergiinae Paul M Peterson1*, Cristina Roquet2, Konstantin Romaschenko1,3, Yolanda Herrera Arrieta4, Alfonso Susanna5 1 Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, DC 20013-7012, USA 2 Systematics and Evolution of Vascular Plants (UAB) – Associated Unit to CSIC, Departament de Biologia Animal, Biologia Vegetal i Ecologia, Facultat de Biociències, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain 3 M.G. Kholodny Institute of Botany, National Academy of Sciences, Kiev 01601, Ukraine 4 Instituto Politécnico Nacional, CIIDIR Unidad-Durango-COFAA, Durango, C.P. 34220, Mexico 5 Laboratory of Molecular Systematics, Botanic Institute of Barcelona (IBB-CSIC- ICUB), Passeig del Migdia s.n., 08038 Barcelona, Spain *Author for correspondence. E-mail: peterson@si.edu. Tel.: 1-202-633-0975. Fax: 1202-786-2653. ORCID (http://orcid.org): PMP, 0000-0001-9405-5528; KR, 0000-0002-7248-4193. Received XX May 2021; Accepted XX XXX 2021; Article first published online xx Month 20XX This is the author manuscript accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/jse.12805. This article is protected by copyright. All rights reserved. Author Manuscript Abstract—A phylogeny based on analysis of six DNA sequence markers (ITS, ndhA intron, rpl32-trnL, rps3, rps16 intron, and rps16-trnK) is used to infer ancestral areas, divergence times, and reconstruct the biogeographical history and evolution of 150 of the 183 (82%) species of Muhlenbergia. Our results suggest the genus originated 9.3 mya in the Sierra Madre (Occidental and Oriental) in Mexico splitting into six lineages: M. ramulosa diverging 8.2 mya, M. subg. Muhlenbergia at 5.9 mya, M. subg. Pseudosporobolus at 5.9 mya, M. subg. Clomena at 5.4 mya, M. subg. Bealia at 4.3 mya, and M. subg. Trichochloa at 1 mya; each of these with a high probability of Sierra Madrean origin. Our results further suggest that founder-event speciation from Sierra Madre to South America occurred independently multiple times in all five subgenera during the Pleistocene and late Pliocene. One long-distance dispersal event most likely originating from Central or Eastern North America to East and Central Asia occurred 1.6−1 mya in M. subg. Muhlenbergia. In our cladogram members of M. subg. Trichochloa show little genetic resolution, suggesting very low levels of divergence among the species, and this may be a consequence of rapid radiation. Key words: biogeography, classification, ITS, molecular phylogeny, Muhlenbergia, plastid DNA sequences 1 Introduction The subtribe Muhlenbergiinae Pilg. (tribe Cynodonteae Dumort.) is a diverse assemblage of 183 species included in a single, monophyletic genus, Muhlenbergia Schreb. (Peterson et al., 2010a,b, 2018a,b; 2016; Soreng et al., 2017). Species within Muhlenbergia are morphologically highly variable and are characterized in having membranous ligules (rarely a line of hairs); paniculate inflorescences that are rebranched or composed only of primary branches; spikelets that are usually solitary but sometimes in pairs or triads, with cleistogenes (self-pollinated flowers that do not This article is protected by copyright. All rights reserved. Author Manuscript open at maturity) occasionally present in the leaf sheaths; one floret (rarely more) per spikelet that is perfect, staminate, or sterile; glumes that are awned or unawned; lemmas 3-nerved, apically awned or unawned; and a base chromosome number of x = 8−10 (Peterson et al., 1995, 1997, 2007a,b, 2018b; Peterson, 2000, 2003, 2007b; Giraldo-Cañas & Peterson, 2009; Peterson & Giraldo-Cañas 2011, 2012; Herrera & Peterson, 2017, 2018). Two subtypes of C4 photosynthesis based on nicotinamide adenine dinucleotide cofactor malic enzyme (NAD-ME) and phosphoenolpyruvate carboxykinase (PCK, known only in M. subg. Muhlenbergia species) have been found in Muhlenbergia; subtypes in most species have been verified by anatomy, and in a few species by biochemical assay (Gutierrez et al., 1974; Brown, 1977; Hattersley & Watson, 1992). Based on seven molecular markers (nuclear ITS and plastid ndhA intron, ndhF, rps16-trnK, rps16 intron, rps3, and rpl32-trnL) Peterson et al. (2010b) provided a phylogeny and classification for 124 species (68%) of the Muhlenbergiinae, recognizing five subgenera within Muhlenbergia: M. subg. Bealia (Scribn.) P.M. Peterson, M. subg. Clomena (P. Beauv.) Hack., M. subg. Muhlenbergia, M. subg. Pseudosporobolus (Parodi) P.M. Peterson, and M. subg. Trichochloa A. Gray. Prior to molecular DNA investigations the Muhlenbergiinae was narrowly interpreted, including only Muhlenbergia s.s. (Pilger, 1956). Other species now included in Muhlenbergia s.l. as treated by Pilger (1956) were: Blepharoneuron Nash and Epicampes J. Presl placed in the Sporobolinae Benth., Aegopogon Humb. & Bonpl. ex Willd.and Schaffnerella Nash placed in the Lappagineae Link, Redfieldia Vasey placed in the Eragrostinae Ohwi, Lycurus Kunth and Pereilema J. Presl placed in the Lycurinae Pilg., and Schedonnardus Steud. placed in the Chloridinae Pilg. Using plastid-mapped DNA restriction sites for a limited sample of species, Duvall et al. This article is protected by copyright. All rights reserved. Author Manuscript (1994) found a monophyletic Muhlenbergiinae that included seven genera (Bealia Scribn., Blepharoneuron Nash, Chabboissaea E. Fourn., Lycurus Kunth, Muhlenbergia, Pereilema J. Presl, and Redfieldia Vasey), although Muhlenbergia with only two species in their study was polyphyletic. Finally, with the inclusion of Aegopogon Humb. & Bonpl. ex Willd., Schaffnerella Nash, and Schedonnardus Steud. in molecular studies and subsequent subsumation, we now recognize a much expanded Muhlenbergia s.l. (Columbus et al., 2010; Peterson et al., 2010b, 2018b). Earlier phylogenetic studies suggested that the Muhlenbergiinae originated in North America since numerous sister groups, i.e., Allolepiinae P.M. Peterson, Romasch. & Y. Herrera, Boutelouinae Stapf, Hilariinae P.M. Peterson & Columbus, Jouveinae P.M. Peterson, Romasch. & Y. Herrera, Kaliniinae P.M. Peterson, Romasch. & Y. Herrera, Monanthochloinae Pilg. ex Potztal, Scleropogoninae Pilg., and Sohnsiinae P.M. Peterson, Romasch. & Y. Herrera, are composed almost entirely of North American species (Peterson et al., 2010b; 2016, 2017; 2018a,b). Ninety-six percent of the species within the Muhlenbergiinae are native to the western hemisphere and more than 80% of these are native to North America (Peterson et al., 2007a; 2010b). Within the Muhlenbergiinae multiple independent radiations to South America have occurred in all five subgenera, and within M. subg. Muhlenbergia there is evidence for a single colonization event to East Asia (Peterson et al., 2010b). The study of American amphitropical disjunctions within the Muhlenbergiinae also indicates a North American origin, and a recent introduction (probably late Pliocene to Pleistocene) into South America has been verified for at least three species (Peterson & Herrera Arrieta, 1995; Sykes et al., 1997; Peterson & Morrone, 1998; Peterson & Ortíz Diaz, 1998; Peterson et al., 2007a). This article is protected by copyright. All rights reserved. Author Manuscript Here we present an updated phylogeny and biogeographical analysis of 150 species (82%) of Muhlenbergia with three new species based on six molecular markers (nuclear ITS and plastid ndhA intron, rpl32-trnL, rps3, rps16 intron, and rps16-trnK). 2 Material and Methods 2.1 Phylogenetic analyses and taxon sampling Detailed methods for DNA extraction, amplification, sequencing and phylogenetic analysis are given in Peterson et al. (2010a,b, 2014a, 2015a,b, 2016). In brief, the phylogeny was estimated among members of Muhlenbergia based on the analysis of six molecular markers −(nuclear ITS 1&2, and plastid ndhA intron, rpl32-trnL, rps3, rps16 intron, rps16-trnK). We sampled most species within subtribe Muhlenbergiinae and included outgroups: Distichlis scoparia (Nees ex Kunth) Arechav. (Monanthochloinae Pilg. ex Potztal), Willkommia sarmentosa Hack. (Traginae P.M. Peterson & Columbus), and Sporobolus indicus L. (Zoysieae Benth., Sporobolinae Benth.) [Peterson et al., 2010a, 2015a, 2014b, 2016; Soreng et al., 2017]. Voucher information and GenBank numbers for all samples (including the new ones) are given in Appendix S1. The resulting plastid and ITS topologies were inspected for conflicting nodes with posterior probabilities (PP) ≥ 0.95 and bootstrap values where 90−100% were interpreted as strong support and 70−89% as moderate. No supported conflict was found so plastid and ITS sequences were combined. Characteristics of the six DNA sequence regions and parameters used as priors in the Bayesian analyses were estimated using GARLI 0.951 (Zwickle, 2006). This article is protected by copyright. All rights reserved. Author Manuscript 2.2 Phylogenetic inference and divergence time estimation Phylogenetic inference was conducted with the maximum likelihood (ML) method implemented in RAxML v8.1.3 (Stamatakis, 2014), applying the GTRCAT model, the rapid hill-climbing algorithm and the optimal partition scheme according to PartitionFinder. S v.1.1.1 (Lanfear et al., 2012). Specifically, we conducted 100 independent ML searches plus 1000 bootstrap replicates (BS); BS values were then added in the best ML tree (i.e. the tree that yielded the highest likelihood). To obtain a time-calibrated phylogeny, we dated the best ML tree with the penalized-likelihood method implemented in treePL (Smith & O’Meara, 2012), fixing the node from which Muhlenbergia diverges from Distichlis to 12.4 mya based on the results of Bouchenak-Khelladi (2010), which estimated that age for the node for the most recent common ancestor (mrca) of Muhlenbergiinae and Monanthochloinae (Fig. S1). We first performed a priming analysis and then ran 10 cross-validations in order to obtain the best smoothing parameter, which corresponded to the value of 0.0001 (Roquet et al., 2009). 2.3 Biogeographic model comparison and ancestral geographic range estimation We used the maximum likelihood method implemented in the R package BioGeoBEARS vers. 1.1.2 (Matzke, 2013) to estimate geographic range evolution in Muhlenbergia and investigate the main biogeographic processes that have shaped the evolution of this genus. To do so, we compared the fit of the three main biogeographical models implemented in BioGeoBEARS: the dispersal–extinction– cladogenesis model (DEC; Ree, 2005); a likelihood implementation of DIVA (Ronquist, 1997), hereafter DIVAlike; and a likelihood implementation of BayArea (Landis et al., 2013), hereafter BAYAREAlike. These three models differ on the processes allowed to occur at cladogenetic events (i.e., branching points on the tree). This article is protected by copyright. All rights reserved. Author Manuscript The DIVAlike model includes two types of cladogenetic events: widespread vicariance, which means that a broad-ranging ancestor gives rise to two daughter lineages that occupy more than one area; and narrow vicariance (one daughter lineage inherits only one of the ancestral areas). The DEC model allows for narrow vicariance (but not widespread), and subset sympatric speciation: a broad-ranging ancestor gives rise to one daughter found only in one ancestral area, and another daughter found in the entire ancestral range, resulting thus in partial sympatry. The BAYAREAlike model only allows for sympatric speciation, i.e., daughter lineages have the same area (or combination of areas) as their ancestor. In all BioGeoBEARS models, dispersal to and extinction in one or more areas are modeled in all cases as anagenetic processes, i.e. they are assumed to occur along branches. In addition, we also fit a more complex version of each of the three mentioned models through the addition of the jump dispersal parameter J (i.e., DEC+J, DIVAlike+J, BAYAREAlike+J), which depicts founder-event speciation events (i.e., range switch that occurs at a lineage-splitting event, resulting in one daughter lineage in a new range and the other daughter lineage retaining the ancestral range; Matzke, 2014). We compared the resulting six models with the AIC and the Akaike information criterion weights (AICwt), the latter provides a relative probability for each model. As input for the BioGeoBEARS analyses, we used the time-calibrated tree after removal of the outgroup taxa, and study species were coded as present or absent in nine geographical regions: (A) Sierra Madre Occidental, (B) Great Basin and Northern Rockies, (C) Sierra Madre Oriental and Central Mexico, (D) Central and Eastern North America, (E) Central America, (F) Northern South America, (G) Central South America, (H) Southern South America, and (I) East and Central Asia. This article is protected by copyright. All rights reserved. Author Manuscript We set the maximum number of areas for any lineage to six, which is the highest number of areas occupied by the most widespread taxon in our study. 3 Results 3.1 Phylogenetic analyses Eighteen new sequences are reported in GenBank and we include three additional species [M. decumbens Swallen, M. eriophylla Swallen, and M. sylvatica (Torr.) Torr. ex A. Gray] in our current phylogeny of 150 (150/183 = 82%) species of Muhlenbergia (Appendix S1) [Peterson et al., 2010b; 2018]. Total aligned characters, number of sequences, number of new sequences, likelihood score, model type, substitution rate, and character state frequency are noted for each DNA marker in Table S2. A maximum likelihood tree from combined plastid and ITS regions is given in supplement (Fig. S3). 3.2 Biogeographic analyses Comparison among biogeographic models with AIC and AICwt indicated that BAYAREAlike+J is the model that best fits the data with more than 99% of the model weight and an AIC > 2 points lower than the second best model (Burnham & Anderson, 2002.). For the three main models, the more complex version with the parameter +J added performed best in all cases. Concerning anagenetic processes or gradual speciation, the inferred parameters show that range contraction (e = 0.13) had a larger contribution than range expansion (d = 0.031). According to our divergence time (Fig. S4) and biogeographic estimations (Fig. 1A,B), Muhlenbergia originated 9.3 mya in Sierra Madre Occidental and Sierra Madre Oriental (hereafter, Sierra Madre, labeled AC). The genus split in two main lineages, one constituted by the mrca of M. ramulosa (which diverged 8.2 mya from This article is protected by copyright. All rights reserved. Author Manuscript the rest of taxa) and the subgenera Bealia and Trichochloa (which diverged from each other 5.5 mya), and another one formed by subgenus Muhlenbergia plus the sister subgenera Clomena and Pseudosporobolus. Subgenus Muhlenbergia split 8.7 mya from the mrca of subgenera Clomena and Pseudosporobolus, the latter two split 6.8 mya. According to BioGeoBEARS estimations with the BAYAREAlike+J model, all subgenera originated in the Sierra Madre (AC), which is also the combination of areas recovered as the most probable for the majority of deep and intermediate nodes. Muhlenbergia subg. Pseudosporobolus started to diversify c. 5.9 mya. Founderevent speciation from Sierra Madre to South America likely occurred at least four times in this group: 2.6 mya, (ancestor of M. monandra Alegría & Rúgolo), 1.8 mya (ancestor of M. palmirensis Grignon & Lægaard), 0.8 mya [ancestor of M. fastigiata (J. Presl) Henrard], and 0.4 mya (ancestor of M. atacamensis Parodi). Muhlenbergia subg. Clomena originated c. 5.4 mya in Sierra Madre, and most of the ancestors within this clade were distributed there. At least three founder-event speciation occurrences are estimated within this subgenus, one to Central America and two to Western North America, all of them in very recent times and from ancestors distributed in Sierra Madre (> 1 mya). Muhlenbergia subg. Muhlenbergia started to diversify c. 5.9 mya either in Sierra Madre (50% of relative probability), or in Sierra Madre and Central America (45% of relative probability). Expansions to South America occurred several times: in the ancestor of M. romaschenkoi P.M. Peterson and M. tenuifolia (Kunth) Kunth, from Sierra Madre and Central America (0.9 mya); in the mrca of M. tenella (Kunth) Trin. and M. ciliata (Kunth) Trin., which expanded from Sierra Madre to Central America at some point between 4.9−0.5 mya (dispersal occurred during anagenesis) and then the daughter lineages of M. tenella and M. ciliata expanded to South America; the mrca of M. bryophilus (Döll) P.M. Peterson This article is protected by copyright. All rights reserved. Author Manuscript and M. uniseta (Lag.) Columbus also expanded from Sierra Madre to Central and South America between 3−1.5 mya; and the species M. microsperma (DC.) Kunth and M. diversiglumis Trin. expanded independently to Central and South America from an ancestor distributed only in Sierra Madre Occidental. In addition, at least one colonization event to East and Central Asia occurred between 1.6 and 1 mya in the subclade defined by the mrca of M. sobolifera (Muhl. ex Willd.) Trin. to M. ramosa (Hack. ex Matsum.) Makino. Concerning Muhlenbergia subg. Bealia, which started to diversify 4.3 mya, most of the ancestors of this clade were distributed either in Sierra Madre or only in Sierra Madre Occidental. Within this clade, there was recent range expansions towards South America at least twice (ancestor of M. caxamarcensis Lægaard & Sánchez Vega, 1.3 mya; mrca of M. torreyi (Kunth) Hitchc. ex Bush and M. arenicola Buckley, 0.9 mya). Muhlenbergia subg. Trichochloa is of very recent origin (it started to diversify 1 mya) and highly speciose. We can state that within M. subg. Trichochloa there was at least one expansion (although more events could also have occurred) to South America given that this clade holds several species found in this continent (at least two endemics plus others distributed there and in other regions); however, given the low bootstrap support the phylogenetic resolution for the relationships among the majority of its taxa, and biogeographic range estimation for nodes within this clade, should be taken with great caution. 4 Discussion 4.1 Geographic range evolution This article is protected by copyright. All rights reserved. Author Manuscript We found strong support for a biogeographic model in which the only range-changing process, i.e., when lineages split (cladogenesis), is founder-event speciation, wherein dispersal to different areas often leads to a speciation event. Past extinction events not visible may have occurred, and if so would impede our full understanding of specific evolutionary pathways. However, given our models, vicariance apparently did not play a significant role in the biogeographic history of Muhlenbergia. Range expansions from North America to South America occurred independently and at multiple times in all five subgenera primarily during the Pleistocene but also in the Pliocene. The timing of connection between these two continents has been the subject of an intense debate, but a recent extensive review reaffirmed that the land bridge of Panama isthmus formed recently, c. 3 mya (O’Dea et al., 2016). Also, the Panama Arc was quite near to South America in the last 20 mya (O’Dea et al., 2016). Such proximity could have facilitated dispersal through floating seeds or rafts transporting seeds (Nathan et al., 2008); however, several studies suggest that this type of dispersal of terrestrial organisms was likely impeded by strong interoceanic currents between North and South America (Bartoli et al., 2005; von der Heydt, 2005; Schneider et al. 2006). This perhaps could explain why Muhlenbergia only started to expand its range to another continent only 3 mya. Apart from that, Muhlenbergia reached another continent - Eurasia, thanks to a colonization event that occurred between 1.6 and 1 mya. Beringian dispersal is the most likely route for colonization from North America to East and Central Asia, and this path is thought to have opened during one of the Mid-Pleistocene cooling events in the last million years (900 kyrevent; Kender et al., 2018). The crown node for Asian species of Muhlenbergia is 0.9 mya, coinciding exactly with significant reduction in global CO2 and lowering sea levels leading to the closure of the Bering Strait (Kender et al., 2018). Patis racemosa This article is protected by copyright. All rights reserved. Author Manuscript (Sm.) Romasch., P.M. Peterson & Soreng and Ptilagrostis porteri (Rydb.) W.A. Weber have both been suggested to have migrated from Asia to North America via the Bering land bridge at approximately the same time (Romaschenko et al., 2013). The most likely vector is bird-mediated dispersal since grass propagules, i.e., florets, rhizomes etc., can be transported over long distances attached to feathers, and seeds can be easily consumed and remain viable after passing through the gut (Popp et al., 2011; Viana et al., 2016). 4.2 Phylogeny and morphology Of particular interest in our study is the single colonization event of East and Central Asia, where M. ramosa (native to E Asia) is sister to eight eastern North American species plus M. schreberi (native to North America and type species of the genus), M. japonica Steud., M. hakonensis (Hack. ex Matsum.) Makino, M. huegelii Trin., and M. himalayensis Hack. ex Hook. F (latter four species all native to E Asia). The placement in our tree of M. schreberi rendering M. ramosa in a grade is not strongly supported and earlier analysis of six plastid DNA markers suggest the former species is basal to eight species in the eastern North American clade (Peterson et al., 2010b). The Asian clade and M. ramosa clearly share a mrca with M. schreberi, a weedy species that has been introduced in South America, temperate Asia, Europe, the Caucuses (Jogan, 2014; Clayton et al., 2016). However, alignment of these species could change with inclusion of two not yet studied Asian species: M. curviaristata (Ohwi) Ohwi and M. duthieana Hack. Morphologically, species of the M. subg. Muhlenbergia clade have broad, flat leaf blades with PCK leaf anatomy, fan- to shield-shaped bulliform cells that do not form a column of colorless cells from the adaxial to the abaxial surface, generally with four or more secondary and/or tertiary vascular bundles between consecutive primary vascular bundles. In addition, most This article is protected by copyright. All rights reserved. Author Manuscript have well-developed, scaly, and creeping rhizomes, and panicles that are usually narrow at maturity (Peterson & Herrera Arrieta, 2001; Peterson et al. 2010b, 2018b). PCK-like leaf anatomy appears to have arisen once in the evolution of the Muhlenbergia, and this morphology is linked to species that occupy slightly more mesic habitats (Peterson & Herrera Arrieta, 2001; Peterson et al., 2010b). Although the clade of species representing the M. subg. Trichochloa is strongly supported in our analyses (Fig. 1 & 2), there is little resolution among members, suggesting very low levels of genetic divergence among the species in this subgenus, which may be a consequence of rapid radiation. Within Muhlenbergia, species in this group are by far the most difficult to distinguish because there are very few morphological differences among the taxa and discrete (nonplastic) characteristics are few. Species of M. subg. Trichochloa consist of robust perennials up to 3 meters tall with compressed-keeled or rounded basal sheaths, 1-veined glumes, and unequal rectangular or obovate/elliptic secondary and tertiary vascular bundles with welldeveloped sclerenchyma girders, these usually with sclerosed phloem (Peterson & Herrera Arrieta, 2001; Peterson et al., 2010b, 2018b). The Peruvian endemic Muhlenbergia caxamarcensis shares a mrca with M. ligularis (Central and South America) plus M. filiformis (Thurb. ex S. Wats.) Rydb. and M. vaginata Swallen; the former two are the only members of M. subg. Bealia to have expanded to South America. Species of M. subg. Bealia are strongly caespitose, never rhizomatous, annuals or perennials with pubescent margins or midveins at least on the lower ½ of the lemma [only M. ligularis (Hack.) Hitchc. is without pubescence], and round, equal primary, secondary, and tertiary vascular bundles without well-developed sclerenchyma (Peterson & Herrera Arrieta, 2001; Peterson et al., 2010b, 2018b). This article is protected by copyright. All rights reserved. Author Manuscript Species of M. subg. Clomena have 3-veined upper glumes that are often 3toothed, densely caespitose non rhizomatous culms with lower leaf sheaths that are often flat and somewhat papery at maturity, and lemmas with flexuous awns [only M. jonesii (Vasey) Hitchc. lacks an awn, apex is mucronate] (Herrera Arrieta, 1998; Peterson et al., 2010b, 2018b). The wide-ranging and most likely apomictic, M. peruviana (P. Beauv.) Steud. is the only species of this subgenus that reaches South America (Reeder, 1968; Peterson & Annable, 1991). At least nine species in M. subg. Pseudosporobolus occur in South America, four of these are restricted to South America (M. atacamensis in Bolivia and Peru; M. fastigiata in Argentina, Bolivia, Chile, Colombia, and Peru; M. monandra in Peru; and M. palmirensis in Ecuador) and five species are found in North America extending to South America [M. alopecuroides (Griseb.) P.M. Peterson & Columbus, M. asperifolia (Nees & Meyen ex Trin.) Parodi, M. implicata (Kunth) Trin., M. paniculata (Nutt.) Columbus, and M. phalaroides (Kunth) P.M. Peterson not included in our analysis] (Peterson et al., 2001). Members of M. subg. Pseudosporobolus usually have plumbeous (lead colored) spikelets, well-developed adaxial and abaxial sclerenchyma in their primary vascular bundles, narrow to loosely open panicles, unawned, mucronate or short-awned lemmas (long-awned in M. implicata and M. seatonii Scribn.), and the plants are rhizomatous when perennial (Peterson & Annable, 1992; Peterson & Herrera Arrieta, 2001; Peterson et al., 2010b, 2018b). Acknowledgements We thank the Smithsonian Institution’s Restricted Endowment Fund, the Scholarly Studies Program, Research Opportunities, Atherton Seidell Foundation, Biodiversity Surveys and Inventories Program, Small Grants and the Laboratory of Analytical Biology; the National Geographic Society for Research and Exploration (Grant No. This article is protected by copyright. All rights reserved. Author Manuscript 8848-10, 8087-06); Robert J. Soreng and Neil Snow for suggesting improvements to the manuscript. References Bartoli G, Sarnthein M, Weinelt M, Erlenkeuser H, Garbe-Schönberg D, Lea DW. 2005. Final closure of Panama and the onset of northern hemisphere glaciation. Earth and Planetary Science Letters 237: 33–44. Bouchenak-Khelladi Y, Verboom GA, Savolainen V, Hodkinson TR. 2010a. Biogeography of the grasses (Poaceae): A phylogenetic approach to reveal evolutionary history in geographical space and geological time. Botanical Journal of the Linnean Society 162: 543−557. Brown W. 1977. The Kranz syndrome and its subtypes in grass systematics. Memoirs of the Torrey Botanical Club 23: 1−97. Burnham KP, Anderson DR. 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretical Approach. New York: Springer-Verlag Columbus JT, Peterson PM, Refulio Rodríguez NF, Cerros Tlatilpa R, Kinney MS. 2010. Phylogenetics of Muhlenbergiinae (Poaceae, Chloridoideae, Cynodonteae) based on ITS and trnL-F DNA sequences. Pp. 477−495. In: Seberg O, Petersen G, Barfod AS, Davis JI (eds.). Diversity, Phylogeny, and Evolution in the Monocotyledons. Denmark: Aarhus University Press. Duvall MR, Peterson PM, Christensen AH. 1994. Alliances of Muhlenbergia within the New World Eragrostideae are identified by phylogenetic analysis of mapped restriction sites from plastid DNA's. American Journal of Botany 81: 622−629. Giraldo-Cañas D, Peterson P. 2009. El género Muhlenbergia (Poaceae: Chloridoideae: Cynodonteae: Muhlenbergiinae) en Colombia. Caldasia 31: 269−302. This article is protected by copyright. All rights reserved. Author Manuscript Gutierrez M, Gracen VE, Edwards GE. 1974. Biochemical and cytological relationships in C4 plants. Planta 119: 279−300. Hattersley PW, Watson L. 1992. Diversification of photosynthesis. In: Chapman GP (Ed) Grass evolution and domestication. Cambridge University Press, Great Britain, 38−116. Herrera Arrieta Y. 1998. A revision of the Muhlenbergia montana (Nutt.) Hitchc. complex (Poaceae: Chloridoideae). Brittonia 50: 23−50. Herrera Arrieta Y, Peterson PM. 2017. Subfamily Chloridoideae, Parte 1 (Muhlenbergia). Flora del Bajío de regiones adyacentes 199: 1−130. Herrera Arrieta Y, Peterson PM. 2018. Grasses of Chihuahua, Mexico. Smithsonian Contributions to Botany 107: 1−380. Jogan N. 2014, Muhlenbergia schreberi J.F. Gmel. (Poaceae), a new naturalized species in Croatia. Acta Botanica Croatia 73(2): 465−470. Kender S, Ravelo AC, Worne S, Swann GEA, Leng MJ, Asahi H, Becker J, Detlef H, Aiello IW, Andreasen D, Hallet IR. 2018. Closure of the Bering Strait caused Mid-Pleistocene Transition cooling. Nature Communications 9, 5386. https://doi.org/10.1038/s41467-018-07828-0 Landis MJ, Matzke NJ, Moore BR, Huelsenbeck JP. 2013. Bayesian analysis of biogeography when the number of areas is large. Systematic Biology 62: 789-804. Lanfear R, Calcott B, Ho SYW, Guindon S. 2012. PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses. Molecular Biology and Evolution 29:1695–1701 Matzke, N.J. 2013. BioGeoBEARS: biogeography with Bayesian(and likelihood) evolutionary analysis in R scripts. R package. This article is protected by copyright. All rights reserved. Author Manuscript Matzke, N.J. 2014. Model selection in historical biogeography reveals that founderevent speciation is a crucial process in island clades. Systematic Biology 63: 951– 970. Nathan R, Schurr FM, Spiegel O, Steinitz O, Trakhtenbrot A, Tsoar A. 2008. Mechanisms of long-distance seed dispersal. Trends in Ecology and Evolution 23: 638–647. doi:10.1016/j.tree.2008.08.003 O'Dea A, Lessios HA, Coates AG, Eytan RI, Restrepo-Moreno SA, Cione AL, Collins LS, de Queiroz A, Farris DW, Norris RD, Stallard RF, Woodburne MO, Aguilera O, Aubry M-P, Berggren WA, Budd AF, Cozzuol MA, Coppard SE, Duque-Caro H, Finnegan S, Gasparini GM, Grossman EL, Johnson KG, Keigwin LD, Knowlton N, Leigh EG, Leonard-Pingel JS, Marko PB, Pyenson ND, RachelloDolmen PG, Soibelzon E, Soibelzon L, Todd JA, Vermeij GJ, Jackson JBC. 2016. Formation of the Isthmus of Panama. Science Advances 2 (8): e1600883. DOI: 10.1126/sciadv.1600883 Peterson PM. 2000. Systematics of the Muhlenbergiinae (Chloridoideae: Eragrostideae). In: Jacobs SWL, Everett J eds. Grasses: Systematics and Evolution. CSIRO, Melbourne, 195−212. Peterson PM. 2003. 17.33 Muhlenbergia Schreb. In: Barkworth ME, Capels KM, Long S, Piep MB eds. Magnoliophyta: Commelinidae (in part): Poaceae, part 2 Flora of North America North of Mexico, Vol 25. Oxford University Press, New York, 145−201. Peterson PM, Annable CR. 1991. Systematics of the annual species of Muhlenbergia (Poaceae-Eragrostideae). Systematic Botany Monographs 31: 1−109. Peterson PM, Annable CR. 1992. A revision of Chaboissaea (Poaceae: Eragrostideae). Madroño 39: 8−30. This article is protected by copyright. All rights reserved. Author Manuscript Peterson PM, Columbus JT, Pennington SJ. 2007a. Classification and biogeography of New World grasses: Chloridoideae. Aliso 23: 580−594. Peterson PM, Giraldo-Cañas D. 2011. Las especies de Muhlenbergia (Poaceae: Chloridoideae) de Argentina. Caldasia 33: 21−54. Peterson PM, Giraldo-Cañas D. 2012. Muhlenbergia Schreb. In: Zuloaga FO, Rugolo ZE, Anton AMR eds. Flora Argentina, Flora Vascular de la Republica Argentina, Vol. 3, Tomo 1, Monocotyledoneae, Poaceae: Aristidoideae a Pharoideae. IBODA-IMBIV, CONICET, Buenos Aires, Argentina, 140−154. Peterson PM, Herrera Arrieta Y. 1995. Allozyme variation in the amphitropical disjunct, Chaboissaea (Poaceae: Eragrostideae). Madroño 42: 427−449. Peterson PM, Herrera Arrieta Y. 2001. A leaf blade anatomical survey of Muhlenbergia (Poaceae: Muhlenbergiinae). Sida 19: 469−506. Peterson PM, Herrera Arrieta Y, Romaschenko K. 2018a. Phylogeny of Muhlenbergia subg. Pseudosporobolus, including M. spatha (Poaceae, Chloridoideae, Cynodonteae, Muhlenbergiinae) now found in Zacatecas, Mexico. PhytoKeys 103: 83−94. Peterson, PM, Morrone O. 1988. Allelic variation in the amphitropical disjunct Lycurus setosus (Poaceae: Eragrostideae). Madroño 44: 334−346. Peterson PM, Ortiz-Diaz JJ. 1998. Allelic variation in the amphitropical disjunct Muhlenbergia torreyi (Poaceae: Muhlenbergiinae). Brittonia 50: 381−391. Peterson PM, Romaschenko K, Herrera Arrieta Y. 2015a. A molecular phylogeny and classification of the Eleusininae with a new genus, Micrachne (Poaceae: Chloridoideae: Cynodonteae). Taxon 64: 445−467. Peterson PM, Romaschenko K, Herrera Arrieta Y. 2015b. Phylogeny and subgeneric classification of Bouteloua with a new species, B. herrera-arrietae (Poaceae: This article is protected by copyright. All rights reserved. Author Manuscript Chloridoideae: Cynodonteae: Boutelouinae). Journal of Systematics and Evolution 53: 351−366. Peterson PM, Romaschenko K, Herrera Arrieta Y. 2017. Four new subtribes: Allolepiinae, Jouveinae, Kaliniinae, and Sohnsiinae in the Cynodonteae (Poaceae: Chloridoideae). Phytoneuron 2017-44: 1–9. http://www.phytoneuron.net/2017Phytoneuron/44PhytoNChloridoidSubtribes.pdf Peterson PM, Romaschenko K, Herrera Arrieta Y. 2016. A molecular phylogeny and classification of the Cynodonteae (Poaceae: Chloridoideae) with four new genera: Orthacanthus, Triplasiella, Tripogonella, and Zaqiqah; three new subtribes: Dactylocteniinae, Orininae, and Zaqiqahinae; and a subgeneric classification of Distichlis. Taxon 65: 1263−1287. Peterson PM, Romaschenko K, Herrera Arrieta Y, Saarela JM. 2014b. A molecular phylogeny and new subgeneric classification of Sporobolus (Poaceae: Chloridoideae: Sporobolinae). Taxon 63: 1212−1243. Peterson PM, Romaschenko K, Johnson G. 2010a. A classification of the Chloridoideae (Poaceae) based on multi-gene phylogenetic trees. Molecular Phylogenetics and Evolution 55: 580−598. Peterson PM, Romaschenko K, Johnson G. 2010b. A phylogeny and classification of the Muhlenbergiinae (Poaceae: Chloridoideae: Cynodonteae) based on plastid and nuclear DNA sequences. American Journal of Botany 97: 1532−1554. Peterson PM, Romaschenko K, Soreng RJ. 2014a. A laboratory guide for generating DNA barcodes in grasses: a case study of Leptochloa s.l. (Poaceae: Chloridoideae). Webbia 69: 1−12. This article is protected by copyright. All rights reserved. Author Manuscript Peterson PM, Sánchez Vega I, Romaschenko K, Giraldo-Cañas D, Refulio Rodriguez NF. 2018b. Revision of Muhlenbergia (Poaceae, Chloridoideae, Cynodonteae, Muhlenbergiinae) in Peru: classification, phylogeny, and a new species, M. romaschenkoi. PhytoKeys 114: 123–206. http://dx.doi.org/10.3897/phytokeys.114.28799 Peterson PM, Soreng RJ, Davidse G, Filgueiras TS, Zuloaga FO, Judziewicz EJ. 2001. Catalogue of New World grasses (Poaceae): II. subfamily Chloridoideae. Contributions from the United States National Herbarium 41: 1−255. Peterson P, Valdés-Reyna J, Herrera Arrieta Y. 2007b. Muhlenbergiinae (Poaceae: Chloridoideae: Cynodonteae): from northeastern México. Journal of the Botanical Research Institute of Texas 1: 933−1000. Peterson PM, Webster RD, Valdés-Reyna J. 1995. Sub tribal classification of the New World Eragrostideae (Poaceae: Chloridoideae). Sida 16: 529−544. Peterson PM, Webster RD, Valdes Reyna J. 1997. Genera of New World Eragrostideae (Poaceae: Chloridoideae). Smithsonian Contributions to Botany 87: 1−50. Pilger R. 1956. Gramineae II. Pp. 1−168. In: Engler A, Prantl J. (eds.) Die Natürlichen Pflanzenfamilien, 2nd ed., vol. 14. Berlin, Germany: Duncker and Humblot. Popp M, Mirré V, Brochmann C. 2011. A single Mid-Pleistocene long-distance dispersal by a bird can explain the extreme bipolar disjunction in crowberries (Empetrum). Proceedings of the National Academy of Sciences 108 (16): 6520−6525. This article is protected by copyright. All rights reserved. Author Manuscript Ree RH, Moore BR, Webb CO, Donoghue MJ. 2005. A likelihood framework for inferring the evolution of geographic range on phylogenetic trees. Evolution 59: 2299–2311. Reeder JR. 1968. Notes on Mexican grasses VIII. Miscellaneous chromosome numbers─2. Bulletin of the Torrey Botanical Club 95: 69−86. Romaschenko K, Garcia-Jacas N, Peterson PM, Soreng RJ, Vilatersana R, Susanna A. 2013. Miocene−Pliocene speciation, introgression, and migration of Patis and Ptilagrostis (Poaceae: Stipeae). Molecular Phylogenetics and Evolution 70: 244−259. http://dx.doi.org/10.1016/j.ympev.2013.09.018 Ronquist, F., 1997. Dispersal–vicariance analysis: a new approach to the quantification of historical biogeography. Systematic Biology 46: 195–203. Roquet C, Sanmartín I, Garcia-Jacas N, Sáez L, Susanna A, Wikström N, Aldasoro JJ. 2009. Reconstructing the history of Campanulaceae with a Bayesian approach to molecular dating and dispersal-vicariance analyses. Molecular Phylogenetics and Evolution 52: 575‒587. Schneider B, Schmittner A. 2006. Simulating the impact of the Panamanian seaway closure on ocean circulation, marine productivity and nutrient cycling. Earth and Planetary Science Letters 246: 367–380. Smith SA, O’Meara BC. 2012. treePL: divergence time estimation using penalized likelihood for large phylogenies. Bioinformatics 28: 2689−2690. doi:10.1093/bioinformatics/bts492. Soreng RJ, Peterson PM, Romaschenko K, Davidse G, Teisher JK, Clark LG, Barberá P, Gillespie LJ, Zuloaga FO. 2017. A worldwide phylogenetic classification of the Poaceae (Gramineae) II: An update and a comparison of two 2015 classifications. Journal of Systematics and Evolution 55: 259−290. This article is protected by copyright. All rights reserved. Author Manuscript Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and postanalysis of large phylogenies. Bioinformatics 30: 1312−1313. Sykes GR, Christensen AH, Peterson PM. 1997. A chloroplast DNA analysis of Chaboissaea (Poaceae: Eragrostideae). Systematic Botany 22: 291−302. Viana DS, Santamaría L, Figuerola J. 2016. Migratory birds as global dispersal vectors. Trends in Ecology and Evolution 30 (10): 763−775. Von der Heydt A, Dijkstra HA. 2005. Flow reorganizations in the Panama seaway: A cause for the demise of Miocene corals? Geophysical Research Letters 32: L02609. Zwickl DJ. 2006. Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Austin: University of Texas. Fig. 1A & B. Estimated ancestral biogeographic areas under the model BAYAREAlike+J plotted on the time-calibrated phylogeny with the highest likelihood. Pie charts reflect the relative probability of estimated ancestral areas. Values below branches indicate bootstrap support. Letters above nodes indicate ancestral area with the highest relative probability (indicated only at nodes with BS equal or higher than 70%). World map shows the 9 areas defined for the biogeographical analysis, colored squares show the color legend for combinations of areas. Capital letters inside rectangles next to species names indicate known present geographic distribution for each taxon. This article is protected by copyright. All rights reserved. Author Manuscript This article is protected by copyright. All rights reserved. Author Manuscript This article is protected by copyright. All rights reserved.