http://orcid.org/0000-0003-2794-539X
Figures
Abstract
Panicum sensu stricto is a genus of grasses (Poaceae) with nearly, according to this study, 163 species distributed worldwide. This genus is included in the subtribe Panicinae together with Louisiella, the latter with 2 species. Panicum and subtribe Panicinae are characterized by including annual or perennial taxa with open and lax panicles, and spikelets with the lower glume reduced; all taxa also share a basic chromosome number of x = 9 and a Kranz leaf blade anatomy typical of the NAD-me subtype photosynthetic pathway. Nevertheless, the phylogenetic placements of many Panicum species, and the circumscription of the genus, remained untested. Therefore, phylogenetic analyses were conducted using sequence data from the ndhF plastid region, in an extensive worldwide sampling of Panicum and related genera, in order to infer evolutionary relationships and to provide a phylogenetic framework to review the classification of the genus. Diversification times, historical biogeography and evolutionary patterns of the life history (annual vs. perennial) in the subtribe and Panicum were also studied. Results obtained provide strong support for a monophyletic Panicum including 71 species and 7 sections, of which sections Arthragrostis and Yakirra are new in the genus; 7 new combinations are made here. Furthermore, 32 species traditionally assigned to Panicum were excluded from the genus, and discussed in other subtribes of Paniceae. Our study suggested that early diversification in subtribe Panicinae and Panicum occurred through the Early-Mid Miocene in the Neotropics, while the subsequent diversification of its sections mainly occurred in the Late Miocene-Pleistocene, involving multiple dispersals to all continents. Our analyses also showed that transition rates and changes between annual and perennial life history in Panicum were quite frequent, suggesting considerable lability of this trait. Changes of the life history, together with C4 photosynthesis, and the multiple dispersal events since the Mid Miocene, seem to have facilitated a widespread distribution of the genus. All these findings contribute to a better understanding of the systematics and evolution of Panicum.
Citation: Zuloaga FO, Salariato DL, Scataglini A (2018) Molecular phylogeny of Panicum s. str. (Poaceae, Panicoideae, Paniceae) and insights into its biogeography and evolution. PLoS ONE 13(2): e0191529. https://doi.org/10.1371/journal.pone.0191529
Editor: William Oki Wong, Indiana University Bloomington, UNITED STATES
Received: October 19, 2017; Accepted: January 5, 2018; Published: February 21, 2018
Copyright: © 2018 Zuloaga et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files. The aligned matrix, XML files for BEAST analyses and the trees obtained are available from the Dryad Digital Repository: http://datadryad.org/review? doi=doi:10.5061/dryad.286gn.
Funding: This study was supported in part by ANPCyT (Agencia Nacional de Promoción Científica y Técnica, Argentina), grant 1220-2013 to Fernando Omar Zuloaga and by CONICET, grant 0207 to Fernando Omar Zuloaga and by the National Geographic Society, grants 8862-10 and 3938-13 to Fernando Omar Zuloaga. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Within flowering plants, including grasses, reproductive characters have traditionally formed the backbone of hierarchical classifications. This scheme in many cases conflicted with molecular phylogeny research, which produced a new classification system; [1–6] in grasses. This is particularly true for huge genera, such as Panicum L., which in its broad sense is non-monophyletic, as well as Senecio [7], Acacia [8–9]), and Aster [10–11]. In this regard, Panicum is still maintained as a polyphyletic genus by some authors [12–17]), while others, summary in [16–17], treated species of the genus in other tribes, subtribes, and genera of subfamily Panicoideae or established new taxa to accommodate these segregate species. Panicum L., as traditionally circumscribed, was one of the largest genera of the Poaceae [12], with nearly 450 species distributed worldwide and inhabiting habitats from sea level to approximately 2500 m [18]. The main character placing species in the genus was the spikelet structure, with a lower glume present, usually shorter than the upper glume and lower lemma, the latter subequal, a lower flower present or absent, the upper anthecium indurate and abaxially convex, and a caryopsis with a punctiform to oblong hilum; nevertheless these characters also appear in other members of tribes Paniceae and Paspaleae. Panicum s. l. also exhibits differences in inflorescence types, developmental patterns of spikelets, including nervation of the glumes, and texture and ornamentation of the upper anthecium. Furthermore, physiological, anatomical and cytological diversity is present in Panicum s.l.: all known photosynthetic types found in grasses, occur in the genus, with many non-Kranz species gathered together with all Kranz variants of C4 physiology, i.e., NADP-me, NAD-me and PEP-ck subtypes; also, some species are intermediate between the C3 and C4 pathways [19–21]. In addition, two basic chromosome numbers were reported for the genus, with some species x = 9 and others x = 10.
Phylogenetic studies, based on morphological and molecular characters, have demonstrated that Panicum in its traditional sense [13–15] is not monophyletic [22–29] and that it should be restricted to a set of species all using the C4 NAD-me photosynthetic subtype. These studies implied several changes during the shift from schemes based exclusively on morphological data [12, 30] to those based on molecular data, with new delimitations within the Panicoideae. In the new classification scheme proposed by [16], and [5–6], species traditionally grouped in Panicum s.l. were included under three different subtribes of tribe Paspaleae (Arthropogoninae, Paspalinae, and Otachyriinae) and five subtribes of Paniceae: Boivinellinae, Cenchrinae, Dichanthelliinae, Melinidinae, and Panicinae, [31–51], Table 1.
Subtribe Panicinae includes ca. 165 species, distributed worldwide, of Panicum s. str., and two American and African species of its sister genus Louisiella C.E. Hubb. & J. Léonard. Although over the last decade several grass phylogenies have been published for Panicoideae [1, 3, 23–26, 29, 52–55], species of Panicum s. str. were underrepresented; consequently, a study including a comprehensive sampling of the genus and the small genera related to Panicum, i.e., Louisiella, Arthragrostis Lazarides, Whiteochloa C.E. Hubb., and Yakirra Lazarides & R.D. Webster, is still needed. [5–6, 29, 53].
The aims of this study are to reconstruct the molecular phylogeny of subtribe Panicinae and Panicum s. str., using sequence data from the ndhF plastid region with an extensive sampling of Panicum and related genera, in order to test whether the current classification agrees with the phylogenetic history of the group, and to identify robust clades within the genus. Additionally, we also explore the divergence times for the subtribe and its members, the biogeographical events occurring over its diversification, and the evolutionary patterns exhibited by the life history (annual vs. perennial). Results obtained here are used to propose a new subgeneric classification for Panicum, and to elucidate different evolutionary insights from its diversification.
Materials and methods
Taxon sampling and DNA sequencing
In this study, we inferred a ndhF phylogeny because this marker has provided a robust and strong phylogeny of the Panicoideae and it has proven to be a useful tool to resolve different phylogenetic lineages of plants ([25–26, 29] and several other treatments summarized in Table 1), confirmed by phylogenies based in other genes [1, 3, 23–24, 28, 52, 54–55]. It is important to mention, however, that single-locus phylogenies (gene tree) can be discordant with the species tree, due to different processes such as lineage sorting, introgression, gene duplication, and strong positive selection. Additional multilocus analyses are needed to confirm results obtained here. Nevertheless, our analyses represent the first study to include an extensive sampling of Panicum. The ndhF matrix analyzed here consisted of 214 sequences, 70 of which were generated for this study to maximize the representation of Panicum species and allied genera (57 Panicum, 3 Yakirra, 6 Whiteochloa and 4 Arthragrostis). The remaining 134 sequences were selected from Genbank based on the Panicoid matrix from [26], with the addition of Panicum and related species from [43, 50, 55–58]. In case of potentially uncertain or unexpected positions, two or more vouchers per species were analyzed [i.e. P. laetum Kunth, Whiteochloa capillipes (Benth.) Lazarides]. Information on specimen vouchers for the new sequences obtained and Genbank accessions for all species analyzed are provided in S1 Appendix.
Total genomic DNA was extracted from silica-dried leaves (7 taxa) and from herbarium specimens (63 taxa). DNA of silica samples was extracted with a CTAB protocol [59], while with herbarium material, the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) was used. The complete ndhF gene (ca. 2100 bp) was amplified using primers specified by [26, 60]. For silica-dried samples, three pairs of primers were used (5F-972R, 972F-1666R and 1666F-3R). For herbarium samples, five smaller fragments were amplified using the pairs of primers 5F-536R, 536F-972R, 972F-1666R, 1666F-1821R and 1821F-3R. PCR reactions were performed in 25 ul of final volume with 50–100 ng of template DNA, 0.2 uM of each primer, 25 uM of dNTP, 5 mM MgCl2, and 0.3 units of Taq polymerase provided by Invitrogen Life Technologies. PCR was carried out using the following parameters: one cycle of 94°C for 5 min, 39 cycles of 94°C for 30 s, 48°C for 1 min, and 72°C for 1 min 30 s, and a final extension cycle of 72°C for 10 min. For the species that failed this protocol, primer concentrations were varied. In addition, a variety of PCR additives and enhancing agents (bovine serum albumin, dimethyl sulfoxide) have been used to increase the yield, specificity and consistency of PCRs of herbarium samples. PCR products were run out on a 1% TBE agarose gel stained with SYBR Safe DNA gel stain (Invitrogen) and visualized in a blue-light transilluminator. Automated sequencing was performed by Macrogen, Inc.
Alignment was manually performed using BioEdit ver. 5.0.9 [60]. The aligned matrix is available online from the Dryad Digital Repository: doi:10.5061/dryad.286gn
Phylogenetic analyses and molecular dating
First, the best-fitting codon partition scheme and model of sequence evolution for the ndhF dataset were determined using the Bayesian Information Criterion (BIC) in PartitionFinder 2.1.1 [61]. Three partition schemes corresponding with the codon positions were selected: 1st pos. (TVM+I+G), 2nd pos. (TRN+I+G), and 3rd pos. (TVM+I+G). Maximum likelihood (ML) analyses were conducted in RAxML 8.2.4 [61–62] using nonparametric bootstrap (BS) analysis and searches for the best-scoring ML tree in a single run [63]. We performed 1000 rapid bootstrap inferences and, thereafter, a thorough ML search under the GTRCAT model across codon positions.
Additionally, Bayesian inference (BI) analysis was performed using BEAST 1.8.4 [64]. Phylogenetic analyses were conducted in BEAST using the nucleotide substitution models unlinked across codon position and a Birth-Death model with incomplete sampling as tree prior [65]. To determine the model of rate variation among tree branches we first compared the performance of the strict clock and the uncorrelated lognormal clock model using Bayes Factor (BF) in BEAST. Model comparison was performed through a marginal likelihood estimation (MLE) using path sampling (PS) and stepping-stone sampling (SS) with 100 steps of one million iterations each. The uncorrelated lognormal clock model best explained our data (BFps = 267, BFss = 273) and was used in the final calibrated analyses.
To estimate divergence times, we used two alternative calibration schemes based on the results of [66]. Because Poaceae has a limited fossil record, and the use of phytolits microfossils [67] strongly affected estimated ages, yielding significantly older estimates, [66] tested two alternative calibration schemes: (1) a calibration based only on external angiosperm fossils (eudicots and non-grass monocots), and (2) a calibration including these fossils together with the controversial phytolith microfossils of Poaceae. The authors concluded that the inclusion of phytolith fossils strongly affect estimated ages and they should be considered only as an alternative to the external calibration, at least until more evidence about their placement becomes available. Based on these results, we used median ages and the 95% high posterior density (HPD) reported by [66] under the two calibration schemes as secondary calibrations in normal prior distributions for the following six crown nodes: subfamily Panicoideae (scheme 1: mean = 38.18 Mya, SD = 3.86) (scheme 2: mean = 48.09.18 Mya, SD = 4.94), most recent common ancestor (MRCA) of supertribes Andropogonodae–Panicodae (mean = 30.31 Mya, SD = 3.27) (mean = 36.65 Mya, SD = 3.64), supertribe Andropogonodae (mean = 28.5 Mya, SD = 3.33) (mean = 34.29 Mya, SD = 3.73), tribe Andropogoneae (mean = 11.79 Mya, SD = 2.95) (mean = 14.45 Mya, SD = 2.71), tribe Paspaleae (mean = 22.6 Mya, SD = 3.13) (mean = 26.25 Mya, SD = 3.61), and supertribe Panicodae (mean = 25.46 Mya, SD = 7.76) (mean = 30.74 Mya, SD = 2.97) (Supertribes and Tribes following classification by [6] Fig 1). BEAUti 1.8.4 was used to generate input files for the analyses, in which substitution models were edited manually on the xml file to fit the models selected using PartitionFinder. We conducted three independent runs of 100 million generations, sampling every 50,000. The first 25% of each run was discarded as burn-in after checking for convergence and effective sample size (ESS) > 200 in Tracer v1.6 [68]. Trees of different runs were then combined using LogCombiner 1.8.4 (http://beast.bio.ed.ac.uk/logcombiner) and the maximum clade-credibility tree (MCC tree) was calculated using TreeAnnotator 1.8.4 (http://beast.bio.ed.ac.uk/treeannotator). Phylogenetic trees were visualized in Figtree v1.4.2. The XML files for BEAST analyses and the trees obtained are available from the Dryad Digital Repository: doi:10.5061/dryad.286gn. All RAxML and BEAST analyses were conducted in the CIPRES Science Gateway v3.3 (http://www.phylo.org/) [69].
Red boxes indicate phylogenetic placement of Panicum species recovered outside subtribe Panicinae. Maximum likelihood bootstrap ≥ 70% and Bayesian posterior probability ≥ 0.9 are shown above/below the branches, respectively. Horizontal bars on the nodes indicate the 95% HPD of ages. Black circles to the right of taxon names indicate new sequences generated for this study. Subtribe Panicinae are shown in detail in Fig 2. Mya, million years ago; Pli, Pliocene; Plei, Pleistocene. Results from divergence time estimation using the calibration based in the external angiosperm fossils plus grass phytoliths (scheme 2) are shown in Supporting Information S1 Fig.
Biogeographic analyses
For biogeographical analyses of Panicinae, we identified seven major areas, modified after [70] and important for the subtribe: (1) North America; (2) South America, including Central America and the West Indies; (3) Eurasia, including Europe, Mediterranean Africa, and temperate Asia; (4) Sub-Saharan Africa, including Madagascar; (5) Southeast Asia, including India, Indo-China, the Malaysian Peninsula, the Philippines, Sumatra, Borneo and the Inner Banda Arc; and (6) Australia, including New Guinea, New Caledonia and New Zealand. Species occurrence data were compiled mainly from extensive examination (conducted by F.O. Zuloaga) of herbarium specimens deposited at B, BA, BAA, BAB, BAF, BR, BRI, C, COL, CORD, CTES, F, G, GH, K, LE, LIL, MA, MEXU, MO, NY, P, SI, US, VEN, W, WIS, herbarium abbreviations from [71], and from the literature, mainly taxonomic revisions, floras, and online databases (GBIF, TROPICOS). Analyses were conducted using the package BioGeoBEARS 0.2.1 [72] implemented in R 3.3.1 [73], which allows comparison of different models of ancestral-area reconstruction. Each model allows for a subset of different biogeographical possibilities, such as dispersal, vicariance and extinction. These biogeographical processes are implemented in an ML framework as free parameters that are estimated from the data [74–75]. We used six different models: DEC, DEC+J, DIVA, DIVA+J, BayArea, BayArea+J (J models include a j parameter controlling founder-event speciation), the maximum number of areas was restricted to the maximum number of regions observed among extant taxa (three) and dispersal probabilities among areas were weighted using a dispersal probability matrix (S1 Table, supplementary material). We did not include temporal stratification in the analyses because divergence of Panicinae was dated from the Miocene and there are not substantial changes in the continental configuration at this time for selected areas [70]. Reconstructions were calculated on the MCC tree inferred in BEAST, pruned to include only the subtribe Panicinae and one specimen per species (except for P. fluviicola Steud. and P. phragmitoides Stapf, since they presented alternative phylogenetic placements with PP ≥ 90%). Fit of the models was compared using the Akaike information criterion corrected for sample size (AICc). In addition, and in order to estimate the number and type of biogeographical events (e.g. within-area speciation, vicariance, and dispersal), we used biogeographic stochastic mapping (BSM) [76] under the best fit model (BayArea+j, see results). Event frequencies were estimated by taking the mean and the standard deviation of event counts from 1000 BSMs.
Evolution of life history
We examined the evolutionary patterns associated with life history in subtribe Panicinae coding the life forms as annual (ie. semelparous) or perennial (ie. iteroparous). Data were obtained from the examination of herbarium specimens, the taxonomic literature, and online databases cited above. Transition rate estimation and ancestral character reconstruction were performed in BayesTraits 2.0.2 [77]. Analyses were conducted employing a continuous-time Markov model of trait evolution with two instantaneous rates representing all possible state changes (qannual→perennial and qperennial→annual). Ancestral state reconstructions were executed using the reversible-jump Markov chain Monte Carlo (rjMCMC) method, allowing the analyses to move among different classes of models (for binary traits, five possible models). A reversible-jump hyper prior was set with an exponential prior between 0 and 100, and two independent analyses were run for ten million generation and sampled every 5000 iterations, using 1000 trees randomly subsampled from the posterior distribution of chronograms obtained in BEAST analyses, and pruned to include only the subtribe Panicinae. The first million generations were discarded as burn-in and ESS > 200, while the remaining samples were checked using the R package CODA 0.19–1 [78]. Ancestral states were reconstructed for the MRCA of main sections and clades within Panicinae using the AddMRCA command. Additionally, we compared two models: one in which the rates q01 and q10 were free to vary and another in which rates were constrained to be equal. Fit of the models was evaluated using BF calculated using SS with 100 samples and 10000 iterations per sample.
Numbers of transitions in the life form within Panicinae were estimated using stochastic character mapping (SCM) [79] in phytools 0.6–20 [80] on the 1000 subsampled posterior trees under the best-fitting model (q01 = q10, ‘ER’ model, see results), 100000 simulations (100 SCM on each of the 1000 trees), and sampling the values of the transition matrix (Q) from its posterior distribution.
Additionally, phylogenetic signal in life history was studied using the method proposed by [81] for discrete (binary) characters, and implemented in the R package caper 0.5–2 [82]. The D-value is estimated as the sum of state changes along branches for a binary trait, with smaller values indicating fewer state changes and supporting the hypothesis that a trait is phylogenetically conserved. We compared the estimated D-value to alternative D values generated with simulated data based on the Brownian evolution threshold model (presence of phylogenetic signal) and the white noise model (no phylogenetic signal). The estimated D-value was then scaled according to the simulated values, such that a D-statistic of 0 indicates the trait conservatism expected under Brownian motion and a value of 1 indicates a random distribution. P values are calculated to determine if the D-statistic is significantly different from simulated D values under the Brownian motion and WN models. We estimated D-values for the 1000 subsampled posterior trees and assessed its significance through 1000 permutations.
Results
Phylogeny and divergence times of Panicinae
The analyzed ndhF matrix consisted of 214 taxa and 2084 characters, 440 (21%) of which were parsimony informative. The phylogenetic trees recovered from ML and BI analyses were highly congruent (Figs 1–2 and S1–S2 Figs, supplementary material) and recovered subtribe Panicinae [86% bootstrap support (BS), 1.00 posterior probability (PP)] including two main clades: one composed of Louisiella [L. elephantipes (Nees ex Trin.) Zuloaga and L. fluitans C.E. Hubb. & J. Léonard] and the other including Panicum s. str., Yakirra, and Arthragrostis (Fig 2). The genus Whiteochloa was recovered outside the subtribe Panicinae, in a strongly supported clade (96% BS, 1.00 PP) within subtribe Cenchrinae. Moreover, 32 species previously assigned to Panicum were placed outside Panicinae (Fig 1): four of them (P. pygmaeum R. Br., P. comorense Mez, P. andringitrense A. Camus, and P. robynsii A. Camus) in subtribe Boivinellinae; P. trichocladum Hack. ex K. Schum., sister to Megathyrsus maximus (Jacq.) B.K. Simon & S.W.L. Jacobs in subtribe Melinidinae; P. antidotale Retz. remains in Cenchrinae; and the remaining 26 species appear distributed in the Sacciolepis-Trichanthecium-Kellochloa clade of tribe Paniceae. Finally, within the Panicum s. str. clade of Panicinae (Fig 2) seven well supported groups were recovered, representing different sections of Panicum: Rudgeana (Hitchc.) Zuloaga (BS: 90, PP: 0.99), Hiantes Stapf (BS:76, PP: 1.00), Panicum (BS: 67, PP: 1.00), Dichotomiflora (Hitchc.) Hitchc. & Chase ex Honda (BS: 89, PP: 1.00), Repentia Stapf (BS: 70, PP: 1.00), and genera Arthragrostis (BS: 71, PP 0.99) and Yakirra (BS: 74, PP: 1.00) (Fig 2 and S2 Fig of supplementary material).
A. Maximum clade credibility (MCC) tree of Panicoideae obtained from BEAST analyses with ndhF sequences using the uncorrelated lognormal relaxed clock model and secondary calibrations based only on external angiosperm fossils. Only subtribe Paniceae is shown in detail; for the remaining clades see Fig 1. Maximum likelihood bootstrap ≥ 70% and Bayesian posterior probability ≥ 0.9 are shown above/below the branches, respectively. Horizontal bars on the nodes indicate the 95% HPD of ages. Vertical bars indicate sections within Panicum. Paniceae 1 and Paniceae 2 refer the “Dichanthelinae+ Boivinellinae” clade and the “Incertae sedis genera” clade, respectively. B. Divergence time estimations for crown nodes (MRCA) of subtribe Panicinae, Louisiella, Panicum, and sections of Panicum, based only on external angiosperm fossils (black bars), or angiosperm fossils plus grass phytoliths (red bars). Bars show the 95% HPD of estimated ages, while the squares on bars indicate the median value. Black circles to the right of taxon names indicate new sequences generated for this study. Mya, million years ago; Pli, Pliocene; Plei, Pleistocene.
Divergence time analyses using the calibration scheme based on the external angiosperm fossils of [66] (scheme 1) recovered younger estimates than analyses including the phytoliths (scheme 2) (Table 2, Fig 2 and S2 Fig of supplementary material). However, results from both schemes dated the crown node of subtribe Panicinae principally in the Early Miocene (scheme 1: 17.55 Mya, 95% HPD 21.68–13.98; scheme 2: 21.04 Mya, 95% HPD 25.82–16.73). Within Panicinae, the MRCA of Panicum was estimated around Early-Mid Miocene (15.27Mya, 95% HPD 19.07–12.03; 18.15 Mya, 95% HPD 23.04–14.51). MRCAs of sects. Hiantes and Panicum were estimated around Mid-Late Miocene, while MRCAs of the remaining sections diversified principally during the late Miocene to Pliocene. Table 2 and Fig 2 and S2 Fig provide node ages (median and 95% HPD) for main clades in Panicinae. Subsequent studies were conducted with results obtained from the analyses under scheme 1.
Biogeographical analyses
Of the six biogeographical models evaluated using BioGeoBears, the BayArea+j model resulted the best supported (AICcwt ~ 1, Table 3). The inclusion of the “jump dispersal” parameter j significantly improved all models (BayArea+j, DEC+j, and DIVA+j) (Table 3), suggesting for Panicinae that the models without founder-event speciation (only accounting for dispersal via anagenetic range expansion) are not adequate to account for all movements to new areas. Ancestral range estimation under the BayArea+J model (Fig 3 and S3 Fig of supplementary material) suggests the Neotropics as the most probable ancestral area of the MRCA of the Panicinae (p = 0.78) and its early diversification during the Early-Mid Miocene, including the MRCA of Panicum s. str. (p = 0.76). Subsequent diversification of main clades from the Mid-Miocene to Pliocene involves four primary biogeographical routes: 1) Neotropic- Sub-Saharan Africa, 2) Neotropic-North America, 3) Sub-Saharan Africa- Southeast Asia, 4) Australia-Old World. Clades representing the genus Louisiella, Panicum incertae sedis, and sects. Rudgeana and Hiantes diversified primarily in the Neotropics and Sub-Saharan Africa, with several dispersal events between these two areas (Fig 3). Diversification in sect. Panicum mostly occurred between the Americas (North America and Neotropics). In the remaining clades of Panicum s. str. the Americas were poorly-represented. Diversification in sect. Dichotomiflora involved Sub-Saharan Africa with dispersals to Southeast Asia, whereas in sect. Repentia and genera Yakirra and Arthragrostis ancestral areas were mainly in Australia, with subsequent dispersal to Sub-Saharan Africa, Southeast Asia, or Eurasia.
A. Ancestral range estimation (ARE) on the Panicinae chronogram using the BayArea+J model in BioGeoBEARS. States at nodes (squares) represent the area with highest ML probability before the instantaneous speciation event, whereas those on branches represent the state of the descendant lineage immediately after speciation. Squares with more than one letter refer to ancestral areas composed of more than one biogeographical area. Branch labels have been removed to reduce overlap in cases where they are identical to the state at both the ancestral and the descendant node. Boxes to the left of taxon names indicate areas of tip species. S2 Fig of supplementary material provides all ARE per node and corner with pie charts representing probability of each ancestral area. B. Results from 1000 biogeographic stochastic mapping (BSM) under the BayArea+J model in BioGeoBEARS. Numbers of dispersal events (range-expansion dispersals plus cladogenetic founder/jump dispersal) among areas for Panicinae. Counts of dispersal events were averaged across the 1000 BMSs and are presented here with standard deviations in parentheses. Colour temperature indicates the frequency of events. The sum and corresponding percentages of events involving each area, either as a source for dispersal (the rows) or as a destination (the columns). Map on the left shows main dispersal routes recovered in the BSM analyses. Thick arrows correspond to more frequent dispersal routes.
Log-likelihood ln(L), Akaike information criterion corrected for sample size (AICc), difference in AICc value compared with the best model (ΔAICc), and the Akaike weights (ωi) showing the relative likelihood of each model.
BSM analyses revealed that biogeographical events in the Panicinae comprise within area-speciation (63% of total events) and dispersals (37%), of which 12% correspond to range-expansion dispersals (anagenetic dispersal) and 25% to cladogenetic dispersals (cladogenetic founder/jump dispersal) (S2 Table, supplementary material). Within area-speciation was greater in Africa and Australia and lower in Southeast Asia and North America. Regarding dispersal events, the highest number of dispersals involved interchanges between the Neotropics and Sub-Saharan Africa (Fig 3B), mainly within Louisiella, sect. Rudgeana, and sect. Hiantes, followed by movements, mostly in sect. Panicum, from the Neotropics to North America. Overall, the Neotropics were the most common source for the estimated dispersal events (ca. eleven of 31 events, 36%), whereas Sub-Saharan Africa resulted the largest destination (approx. eight events, ~26%) (Fig 3B).
Evolution of life history
Analyses of habit evolution in Panicinae using the rjMCMC showed that the model with the highest marginal probability was an Equal Rates model (p = 0.97), with the probability of change from perennial to annual the same as the probability of reversal. This model (q01 = q10) was also strongly supported by the BF over the two-rates model (BFq01 = q10/ q01,q10 = 8.24). Ancestral state reconstruction (Fig 4) favored the annual habit for the MRCA of genera Yakirra (p = 0.96), Arthragrostis (p = 0.93) and section Dichotomiflora (p = 0.85); and perennial habit in MRCA of Louisiella (p = 0.77), and sects. Rudgeana (p = 0.93), Hiantes (p = 0.70) and Repentia (p = 0.79). Reconstructions for the MRCA of subtribe Panicinae, Panicum s. str., and section Panicum were ambiguous. The transition count between the two states over the 100,000 SCMs for Panicinae recovered a median of 42 total changes in the life history, with 21 from annual to perennial and 22 for the reverse shift. SCM analyses also indicate that the mean total evolutionary time of Panicinae associated with the annual and perennial habit was similar (47% and 53%, respectively).
The colour of edges in the tree gives the posterior probability (computed as the relative frequency across stochastic maps) of each habit type through the branch of the clade. Red indicates high posterior probability of perennial habit. Pie charts on the main nodes of the Panicineae show character-state probability (blue, annual; red, perennial) from reconstructions in BayesTraits using the 1000 subsampled posterior trees and the rjMCMC method.
Phylogenetic signal estimation for the life form over the 1000 subsampled posterior trees using the Fritz and Purvis’ D statistic resulted in a mean value of 0.18 (95% quantiles 0–0.36), recovering a significant phylogenetic signal (100% of the trees rejected the white noise model with p<0.01), and not significant differences from the distribution expected under a Brownian threshold model (only 0.3% of trees rejected the BM model with p<0.05).
Discussion
Although our analysis was only based on the ndhF chloroplast gene, the results are completely congruent with previous multilocus plastid phylogenies of this group [15, 52, 55]. Due to the potential pitfalls related to single-locus analyses (see materials & methods), new multilocus phylogenies, both with nuclear and plastid data, are needed to confirm results obtained here. Nevertheless our findings represent a major step in understanding the systematics and evolution of Panicum.
Implications for systematic and taxonomy
Results of our study stress that subtribe Panicinae includes only two genera, Louisiella and Panicum, in agreement with [6]. Species of Whiteochloa remaining in subtribe Cenchrinae, together with “P.” antidotale Retz.; one species, “P.” trichocladum, in the Melinidinae; and non-Kranz “Panicum” species in different positions of subtribes Boivinellinae, and in the incertae sedis clade of the Paniceae. From now on “P” or “Panicum” species designates what we consider non Panicum species.
Kranz genera and species excluded from Panicum.
Whiteochloa. This genus includes six species and was classified within subtribe Cenchrinae [3, 5–6, 29], with the analyzed species sister, with low branch support, to the genera Pseudoraphis Griff and Chamaeraphis R. Br. [53], in a study of Panicoideae based on plastome phylogenomics, discussed the position, with maximum support, of W. capillipes within the Panicinae. Nevertheless, these authors were cautious and did not reclassify the genus since they analyzed only one species of the genus. During the current analysis, we studied all species of Whiteochloa, which appeared in a strongly supported clade within subtribe Cenchrinae, confirming previous studies and its position in this subtribe [3, 5–6, 29].
“Panicum” trichocladum Hack. ex K. Schum. This species belongs to subtribe Melinidinae, where it is related, in a strongly supported clade, to Urochloa, Eriochloa, and Megathyrsus. Most species of Eriochloa and the genus Megathyrsus were included as synonyms of Urochloa by [16], while [6] considered valid, as traditionally circumscribed, the three genera. Therefore, and taking into account that the limits of the included genera in Melinidineae are far from clear [16], we do not attempt to place “P." trichocladum in a particular genus within the Melinidinae.
Non-Kranz species excluded from Panicum.
Although many species of Panicum s.l. were studied in previous treatments (see literature cited in Table 1) we have added more than ten additional taxa. As a result, “P.” pygmaeum, “P.” robynsii, “P.” comorense and “P.” andringitrense belong to subtribe Boivinellinae (mostly C3), consistent with new results by [15], who also considered within this clade other species, such as “P.” ibitense A. Camus, “P.” cupressifolium A. Camus, “P.” andringitrense, “P.” vohitrense A. Camus, and “P.” malacotrichum Steud. Relationships of these taxa within this subtribe are still unclear and need further study. Another group of non Kranz species (classified in Group 1 [20], author who analyzed the anatomy of “P.” laticomum Nees, “P.” monticola Hook. f., “P.” heterostachyum Hack., “P.” aequinerve Nees, and “P.” brevifolium L.) are, with strong support but also in need of a complete analysis, in the incertae sedis clade of the Paniceae composed of Sacciolepis Nash, Trichanthecium Zuloaga & Morrone, Kellochloa Lizarazu, M.V. Nicola & Scataglini, and “Panicum” sect. “Monticolae” [16], all C3. Within this clade, “P." notatum Retz., “P". gardneri Thwaites, “P." monticola, “P." pleianthum Peter, and “P." laticomum are in an unresolved position, while “P.” glandulopaniculatum Renvoize, “P.” subhystrix A. Camus, “P.” issongense Pilg., and “P.” capuronii A. Camus appeared in a strongly supported clade with no clear morphological synapomorphies [16], together with “P.” brevifolium, “P.” hirtum Lam., and “P.” heterostachyum. Finally, several species are also in a strongly supported clade embedded in the genus Trichanthecium; they are “P.” perrieri A. Camus, “P.” acrotrichum Hook. f., P. calvum Stapf, “P.” aequinerve, “P.” chionachne Mez, “P.” inaequilatum Stapf & C.E. Hubb., “P.” eickii Mez (the latter four species mentioned in this position by [17]), while “P.” pusillum Hook. f. is a sister species of Kellochloa. A similar result was presented [15] for “P.” perrieri A. Camus and “P.” ambositrense A. Camus. The inclusion of these species in Trichanthecium will require a recircumscription of the genus, a goal beyond the scope of this contribution.
Subtribe Panicinae and Panicum s. str.
Subtribe Panicinae forms a strongly supported clade; morphologically, its taxa are defined mainly by having an open to lax panicles, spikelets dorsiventrally compressed, upper anthecium indurate and convex, a basic chromosome number of x = 9, with all species Kranz of the NAD-me subtype. This subtribe comprises the genera Louisiella and Panicum, the latter also including as synonyms Arthragrostis and Yakirra.
Louisiella. Both species of Louisiella are aquatic perennials with spongy culms, leaves lanceolate, flat, inflorescence lax and open, with spikelets lanceolate, the lower glume 1/6 its length, nerveless to 1-nerved, lower flower absent, and caryopsis with a linear hilum. Both species have the outer parenchymatous bundle sheaths with centrifugally arranged specialized chloroplasts, a feature also distinguishing species of Panicum sect. Dichotomiflora [20]; they also include four molecular synapomorphies at positions 279, 540, 1,260 and 1,431 of the ndhF matrix [50].
Panicum sect. Repentia This clade includes Panicum repens L. and another six species, mostly cespitose perennials with stout rootstocks, spikelets clustered toward the branch tips, with the lower glume ¼ to 1/3(-1/2) the length of the spikelet, and lower palea and lower flower usually present. Species of sect. Repentia were treated within group Virgata [83–84], in sect. Repentes Stapf (including P. repens), and sect. Coloratae Stapf [85] with P. coloratum L. [86] considered in sect. Repentia species of this clade, i.e., P. repens, together with species of clade Hiantes; later, [18] placed sect. Virgata Nees in synonymy of sect. Repentia (see comments under clade Hiantes), although he mentioned that species of Virgata and Repentia could be classified in two different groups. Recently, [26] classified P. repens and allied species within sect. Dichotomiflora.
Panicum sect. Yakirra This is the strongly supported Yakirra lineage, congruent with several molecular studies that have found Yakirra to be nested within Panicum [3, 16, 50, 57]. This genus was segregated from Panicum by [87], based on Panicum pauciflorum R.Br. [Yakirra pauciflora (R. Br.) Lazarides & R.D. Webster], in order to include a group of Australian species that were previously transferred from Panicum to Ichnanthus [88]. Yakirra was distinguished by the presence of a stipitate upper anthecium, as well as a prominent rachilla between the upper glume and lower lemma. The stipe at the base of the upper anthecium in Yakirra has a pair of winged appendages [57], described by [89] as a swollen stipe with two acute lobes. As [57] pointed out, a stipe and elongate rachilla are also present in species of Panicum s.str., such as those of Panicum sect. Rudgeana, and also, as previously mentioned, in species now considered within sect. Arthragrostis. Our analyses confirm the inclusion [16] of Yakirra within Panicum s.str.
Panicum sect. Arthragrostis. We find strong support for a lineage corresponding to species previously treated in the genus Arthragrostis. This genus was established by [90], based on Panicum deschampsioides Domin [Arthragrostis deschampsioides (Domin) Lazarides]. This author characterized the genus by the disarticulation of the inflorescence into component divisions, the stipitate upper anthecium, and the presence of a conspicuous rachilla between the lower and upper glume. Our study grouped seven species, four of them previously treated as Arthragrostis, A. aristispicula (= Panicum aristispiculum), A. brassiana (= P. brassianum), A. clarksoniana (= P. clarksonianum), and A. deschampsioides in a strongly supported clade together with P. seminudum, P. chillagoanum and P. robustum, while P. mitchelii, and P. pilgerianum appeared as the sister taxa of this clade; P. mitchelii is a perennial species with the rachilla internodes elongated between the glumes but no stipe present at the base of the upper anthecium, while P. pilgerianum is an African annual and aquatic species, with lanceolate spikelets, the lower glume 1/6 to 1/5 the length of the spikelet, without a morphological relationship with species of Arthragrostis. The morphological characters that defined Arthragrostis also appear in some species of Panicum. The inflorescence disarticulates at maturity in as P. bergii Arechav. and P. olyroides Kunth; similarly, several species, i.e., P. ligulare Nees ex Trin., P. cervicatum Chase, among others, have a conspicuous internode between the glumes and below the upper anthecium. Therefore, we are considering Arthragrostis as a section within Panicum, including seven Australasian species, characterized as being annuals with open and lax inflorescences, disarticulating or not at maturity, with spikelets ovoid to lanceolate, glabrous, occasionally pilose, the rachilla conspicuous between the lower and upper glume, and upper anthecium stipitate or not, indurate.
Panicum sect. Dichotomiflora. We refer to this lineage as the P. dichotomiflorum Michx. complex, since this species is the most wide-ranging member of this group in the New and Old Worlds. All species in this clade share the following morphological characteristics: annuals, occasionally perennial, with soft culms, blades flat, lanceolate; inflorescence with the branchlets short and appressed, spikelets glabrous, the lower glume 1/5 to ¼ the length of the spikelet, the upper glume and lower lemma subequal, (5-)7-9(-11) nerved, growing in wet and open places. [20] classified species, here arranged in sect. Dichotomiflora, i.e., P. schinzii Hack., P. subalbidum Kunth, in a group of species with PEP-ck type anatomy, i.e., with a Kranz outer bundle sheath in which the specialized chloroplasts are centrifugally arranged. Taxa here considered were treated in group Dichotomiflora [83–84], including P. elephantipes Nees ex Trin. (= Louisiella elephantipes), and subsequently in sect. Dichotomiflora [18, 91]. On the other hand, [26] grouped in sect. Dichotomiflora some species, i.e., P. repens and P. coloratum, that are placed here in sect. Repentia and differ from sect. Dichotomiflora by having conspicuous rootstocks and distichous, involute leaves. [20] also pointed out that P. repens differs anatomically from species here classified in sect. Dichotomiflora, by cross-sectional anatomy and epidermal structure. Species in this clade are widely distributed in southeast Asia, Africa, and North and South America.
Panicum sect. Rudgeana. This is a strongly supported clade which includes three species of sect. Rudgeana [92], P. ligulare Nees ex Trin., P. cervicatum Chase, and P. rudgei Roem. & Schult. Morphologically, they are cespitose plants with erect culms, membranous-ciliate ligules, blades lanceolate to linear-lanceolate, inflorescence a terminal, open and lax panicles, spikelets pilose or glabrous, with the lower glume ½ to ¾ the length of the spikelet, and upper anthecium stipitate, the stipe membranous ventrally and indurate dorsally. All five species of this section grow in open places in Central and South America and differ morphologically from species of sects. Arthragrostis and Yakirra: species of sect. Arthragrostis have a conspicuous rachilla between the lower and upper glume and a homogeneous and slender stipe is present below the upper anthecium. On the other hand, species of sect. Yakirra present a swollen stipe with two acute lobes below the upper anthecium [57, 87].
As sister species of this clade we found P. queenslandicum Domin, an Australian perennial species with a small stipe at the base of the upper anthecium.
Panicum sect. Panicum. We find strong support for a lineage corresponding to the P. miliaceum complex, treated here as P. sect. Panicum as recognized by previous authors [3, 16, 18, 26, 29]. Within this clade, three species from Hawaii, P. carteri Hosaka, P. fauriei Hitchc., and P. nephelophilum Gaudich. grouped together in a supported clade. Morphologically, species of sect. Panicum are characterized by being annual or cespitose perennials, with culms erect, inflorescences open and lax, bearing terminal spikelets with the lower glume (1/3-)1/2-3/4(-4/5) the length of the spikelet, 3-5(-9) nerved, and upper glume and lower lemma (5-)7-9(-15) nerved, the spikelets without elongated or modified rachilla internodes. [20] described P. miliaceum, and other “true” Panicum species, as having double bundle sheaths and centripetally located Kranz chloroplasts. This lineage is widely distributed in the New World, Africa, Asia, Australia, and the Pacific.
Panicum sect. Hiantes. This clade includes a group of American and African and Asian species with strong support; morphologically, they are characterized as annual or perennial, cespitose species, with a terminal, lax and open to contracted panicles, spikelets gaping at maturity, with the lower glume ¾ to 4/5 the length of the spikelet, and lower palea and lower flower present. Species of this clade traditionally were classified in groups Urvilleana and Virgata [83–84], together with species of Repentia), in sects. Hiantes and Dura [85] in Africa, in sects. Repentia and Urvilleana [18], with P. olyroides Kunth as an ungrouped species of Panicum) in the New World, and in sects. Urvilleana and Virgata [26]. Recently [53] also considered sect. Urvilleana within sect. Virgata (= Hiantes). Panicum chloroleucum Griseb., P. racemosum (P. Beauv.) Spreng., and P. urvilleanum Kunth, American species previously grouped in sect. Urvilleana, are all perennial species with conspicuous rhizomes, and spikelets pilose, with the upper lemma with long macrohairs toward its base. Panicum olyroides Kunth and P. mystasipum Zuloaga & Morrone, previously ungrouped species within Panicum, are also, in a strongly supported clade, within sect. Hiantes, with P. curviflorum Hornem., an Asian species, as sister taxon of this clade.
Panicum incertae sedis clade. This clade consists of three species, two of them, P. voeltzkovii A. Camus and P. cinctum Hack., endemic to Madagascar; both are erect cespitose perennials, with linear to lanceolate, flat, blades, inflorescence open and terminal, and spikelets ovoid, with the lower glume ½ its length and the upper glume and lower lemma subequal, 5-7-nerved. Our result agrees with that of [15], who showed both species, together with P. luridum Hack. in a strongly supported clade. Our analysis differs by the presence, in this clade, of P. pinifolium Chiov., a species morphologically similar to P. repens with linear to aciculate, distichous leaves, and lower glume reduced, nerveless to 1-nerved.
Panicum incertae sedis species. Panicum laetum appears ungrouped and sister to the clade including sects. Repentia, Arthragrostis and Yakirra. This is an annual species growing in Africa and Asia characterized by its open and lax panicles, spikelets with the lower glume ½ or more the spikelet length, lower flower absent, and upper glume and lower lemma 7-9-nerved.
Spatio-temporal diversification of Panicum
Results obtained here suggest that early diversification of Panicum occurred through the Early-Mid Miocene in the Neotropics, and principally during the warm period of the Mid-Miocene climatic optimum [93]. Divergence time analyses [66] did not included members of subtribe Panicinae; nevertheless our age estimations for other panicoid groups were in general consistent with those reported by them and other authors [27–28, 54, 94]. [15] recovered the crown node of Panicum s. str. around the Mid-Miocene (~ 12 Mya), while estimates of [27–28] placed it in the Late Miocene (6–8 Mya), after the mid-Miocene climatic optimum, when the global climate became cooler. However, Panicum s. str. and subtribe Paniceae in these later phylogenies are poorly represented (below 5%).
Dispersal events seem to have played an important role in the biogeographic diversification of Panicum. The importance of dispersal in panicoid and other grasses was reported in the extensive biogeographical analyses on grass diversification in Madagascar presented by [15]. These authors concluded that the extant grass flora in Madagascar was the result of multiple overseas dispersals. In Panicum s. str., dispersals were recovered as the most frequent biogeographic event for range change, both involving anagenetic dispersal (i.e., range expansion) and cladogenetic dispersal (i.e., founder-event speciation) [75]. Diversification of sections Hiantes and Panicum, for which a Mid-Miocene origin was estimated, were characterized by two main dispersal routes from the Neotropics, to Sub-Saharan Africa for the former, and North America for the latter. The second group of sections/genera recovered within Panicum, including sects. Rudgeana, Repentia, Dichotomiflora, and genera Arthragrostis and Yakirra, exhibited younger divergences, with their crown node ages recovered mainly through the Late Miocene–Pliocene, their diversification associated with gradual global cooling. In these groups, interchanges with America are infrequent, with the exception sect. Rudgeana. Section Dichotomiflora seems to have dispersed from the Neotropics to Africa by the end of the Late Miocene (around 5.6 mya), exhibiting subsequent dispersions principally to Southeast Asia in the Pliocene. For the MRCA of sect. Repentia, and genera Arthragostis and Yakirra, the Australian continent was recovered as the most likely ancestral area, involving the oldest dispersal from the Neotropics in the Panicinae, most likely around the Mid-Miocene (~13 Mya).
Evolutionary patterns of the life history in Panicum
Our analyses show that rates and changes between annual and perennial life history in Panicum s. str. were quite frequent and similar, suggesting considerable lability of life history and the absence of strong evolutionary constraints. Evolutionary labile traits related to niches have been associated with different intrinsic factors including genetic variation, biophysical constrains, epistatic interactions, and complexity of the new phenotypes [95–98]. This evolutionary lability of the life history in the subtribe Panicinae and Panicum s. str., added to the presence of C4 photosyntesis, could have facilitated repeated shifts between habitats and the colonization of new areas. Evidence reported by [99] suggests that changes from C3 to C4 photosynthesis among panicoid grasses promoted niche expansion into hotter climates, and also into more arid climates for tribe Paniceae. These traits, together with the numerous dispersal events since the Late Miocene, could have generated the widespread distribution of the group.
Transitions between annual and perennial growth habit are reported to be associated mainly with temperature. Annuals are favored in hot conditions with highly variable and unreliable precipitation patterns, and in disturbance regimes, both of which can adversely affect adult perennial plants [100–102]. Perennials are favored in colder environments with short growing seasons [64, 103–104]. Thus, annuals are common in desert floras and are apparently better adapted than perennials to lowland areas with greater temperatures, while perennials are better adapted to cooler environments, principally in alpine habitats [105–106]. However, Panicum species, both annuals and perennials, are distributed in tropical and temperate lowlands, and they rarely occur at latitudes or elevations with short growing seasons. Therefore, further empirical analyses using georeferenced specimen data and aridity index values together with potential evaporation (PET), and other variables related to the ecological niche, should be conducted within Panicum s. str. to detect the ecological correlates of life history traits in this group.
Concluding remarks
Our analyses support the circumscription of subtribe Panicinae as comprising two genera: Louisiella and Panicum, while Arthragrostis and Yakirra are treated as synonyms of Panicum. Also, this study supports the recognition of seven sections in Panicum. Nearly 40 non-Kranz species belong in other subtribes of Paniceae, and one Kranz species goes to Cenchrinae and another to Melinidineae. Evidence obtained here suggest that the early diversification of Panicum s. str. occurred in the Early to Mid-Miocene, while subsequent diversification of its sections took placed mainly through the Late Miocene–Pliocene. Recurrent dispersals, together with the considerable lability of the life-form, along with the advantages of C4 photosynthesis, seem to have favored the widespread distribution and diversification of the genus in latitudes with hot dry, and warm wet, long growing seasons.
Taxonomic treatment
- Subtribe Panicinae Fr., Fl. Scan.: 195. 1835. Type. Panicum L.
Annual or perennial; ligules membranous-ciliate to ciliate. Blades oblong-lanceolate to linear-lanceolate. Inflorescence an open and lax panicle. Spikelets dorsiventrally compressed, the lower glume reduced or up to the full length of the spikelet; upper glume and lower lemma subequal; upper anthecium indurate, abaxially convex, with simple or compound papillae toward the apex. Caryopsis with a linear to punctiform hilum. Basic chromosome number x = 9. Photosynthetic pathway; C4 subtype, NAD-me.
Including two genera, Louisiella and Panicum, distributed worldwide mainly in tropical and subtropical areas.
- Louisiella C.E. Hubb. & J. Léonard, Bull. Jard. Bot. État Bruxelles 22: 316. 1952. Type species. Louisiella fluitans C.E. Hubb. & J. Léonard
Aquatic perennials, with succulent culms and internodes spongy. Blades oblong-lanceolate to linear-lanceolate. Spikelets lanceolate, glabrous, lower glume reduced, hyaline, nerveless to 3-nerved, upper glume and lower lemma longer than the upper anthecium, (5-)7-9 nerved, lower palea reduced or absent, lower anthecium flower absent; upper anthecium not stipitate, shiny. Caryopsis with a linear hilum.
Genus with two species, present in tropical and subtropical areas of America, L. elephantipes, and Africa, L. fluitans.
- Panicum L., Sp. Pl.: 55. 1753. Type species. Panicum miliaceum L.
Annual or perennial, mostly cespitose, with culms erect to decumbent and rooting and branching at the lower nodes. Blades lanceolate to linear-lanceolate. Inflorescence a terminal open panicle, axillary inflorescences occasionally present. Spikelets pedicelled on second or third-order branches, pilose or glabrous, the rhachilla conspicuous or not between the bracts, with the lower glume ¼ to 4/5 the length of the spikelet, 3-9-nerved; upper glume and lower lemma usually subequal, (5-)7-11(-13) nerved; lower anthecium palea and lower flower present or absent; upper anthecium stipitate or not, indurate, often textured, shiny. Caryopsis with a punctiform hilum.
A pantropical genus with nearly 163 species worldwide and classified in seven sections (Table 4). Of these sections, one is endemic to Australia, another is present in Australia and southeast Asia, one is restricted to the Neotropics, and the other four are pantropical. Non Panicum species are listed on Table 5.
Abbreviation: STA (Species tentatively accepted).
Ph-p. refers to photosynthetic pathway; U: Unknown.
- 1. Panicum sect. Arthragrostis (Lazarides) Zuloaga, comb. nov. Arthragrostis Lazarides, Nuytsia 5(2): 285. 1984. Type species: Panicum deschampsioides Domin, Biblioth. Bot. 20(85): 320. 1915.
Annual or perennials, erect to decumbent and rooting at the lower nodes; internodes hollow, glabrous. Ligules membranous-ciliate. Blades lanceolate, flat, pilose, the margins usually ciliate. Inflorescence an open and lax panicle. Spikelets ovoid to narrowly ovoid or ellipsoid, glabrous, less frequently covered by tuberculate hairs; lower glume less than ½ the length of the spikelet, separated by a distinct internode from the upper glume, 5-7-nerved; upper glume and lower lemma subequal, awned to acuminate or acute, 7-11-nerved, membranous, with a manifest rhachilla between the bracts; lower palea reduced, lower flower absent; upper anthecium shorter than the upper glume and lower lemma, stipitate or not, indurate, pale to dark, glabrous. (Fig 5).
Panicum deschampsioides. A. Habit. B. Detail of ligule. C. Detail of the inflorescence. D. Spikelet, upper glume view. E. Spikelet, lateral view. F. Upper anthecium, dorsal view, and lower lemma. G. Upper anthecium, palea view. H. Upper palea with lodicules and stamens.
Including ten Old World species, four of which require combinations in Panicum. [107] classified the genus Arthragrostis as endemic to Australia. Nevertheless, two species are present in the Philippines, P. caudiglume Hack. and P. mindanaense Merr., the former also growing in Java.
- Panicum aristispiculum (B.K. Simon) Zuloaga, comb. nov. Arthragrostis aristispicula B.K. Simon, Austrobaileya 2(3): 238. 1986. Type: Australia. Queensland: Brisbane District: Almaden-Petford road, 4 km from Almaden, 10 Mar, 1980, B. K. Simon & J. R. Clarkson 3598 (holotype, BRI!).
- Panicum brassianum (B.K. Simon) Zuloaga, comb. nov. Arthragrostis brassiana B.K. Simon, Austrobaileya 8(2): 188. 2010. Type: Australia. Queensland: Cook District: crest of Western Scarp of Great Dividing Range, 12 mi E of The Lynd, 11 July 1954, S.T. Blake 19478 (holotype, BRI!; isotypes, AD, CANB!, DNA, K!, L, MO!, PERTH, PRE).
- Panicum brassianum var. minutiflorum (B.K. Simon) Zuloaga, comb. nov. Arthragostis brassiana var. minutiflora B.K. Simon, Austrobaileya 8(2): 188. 2000. Type: Australia. Queensland: Cook District: Lockerbie, 10 mi W of Somerset, 4 May 1948, L.J. Brass 18637 (holotype, BRI!; isotype, A).
- Panicum clarksonianum (B.K. Simon) Zuloaga, comb. nov. Arthragrostis clarksoniana B.K. Simon, Austrobaileya 3(4): 585. 1992. Type: Australia. Cook District: 16 km from Meripah homestead on road to the south, 13°49'S, 142°22'E, 11 May 1987, J.R. Clarkson 7149 & B.K. Simon (holotype, BRI!; isotypes, CNS!, MBA, NSW!).
- 2. Panicum sect. Dichotomiflora (Hitchc.) Hitchc. & Chase ex Honda, J. Fac. Sci. Univ. Tokyo, Sect. 3, Bot. 3(1): 244, 246. 1930. Panicum [unranked] Dichotomiflora Hitchc., N. Amer. Fl. 3(2): 200, 202. 1915. Panicum group Dichotomiflora Hitchc. & Chase, Contr. U.S. Natl. Herb. 15: 28, 47. 1910, nom. inval. Type species: Panicum dichotomiflorum Michx.
Annual, occasionally perennials, with culms erect to decumbent, rooting and branching at the lower nodes. Blades lanceolate. Inflorescence a terminal and open, diffuse to contracted, panicle. Spikelets ellipsoid to lanceolate, glabrous; lower glume 1/5 to 1/3 the length of the spikelet, 1-3-nerved; upper glume and lower lemma subequal, 5-7(-9) nerved; upper anthecium smooth, indurate. (Fig 6).
Panicum impeditum. A. Habit. B. Detail of ligule. C. Apex of blade. D. Spikelet, lower glume view. E. Spìkelet, upper glume view. F. Upper anthecium and lower lemma. G. Lower palea, ventral view. H. Upper anthecium, palea view. I. Upper palea with lodicules and anthers. J. Gynoecium.
This section includes four species in America, approximately twelve in the Old World, with P. dichotomiflorum widely distributed worldwide. They are frequent in humid and wet, open areas, usually present in river banks.
- 3. Panicum sect. Hiantes Stapf, Fl. Trop. Afr. 9: 640, 644. 1920. Type species: P. phragmitoides Stapf.
- Panicum sect Durae Stapf, Fl. Trop. Afr. 9: 640, 648. 1920. Type species: Panicum turgidum Forssk., lectotype here designated.
- Panicum sect. Urvilleana (Hitchc.) Pilger, Notizbl. Bot. Gart. Berlin-Dahlem 11(104): 244. 1931. Panicum group Urvilleana Hitchc. & Chase, Contr. U.S. Natl. Herb. 15: 28, 132. 1910, nom. inval. Panicum [unranked] Urvilleana Hitchc., N. Amer. Fl. 3(2): 200, 205. 1915.
- Panicum sect. Virgata Hitchc. & Chase ex Pilg., Nat. Pflanzenfam. (ed. 2), 14e: 22. 1940. Panicum group Virgata Hitchc. & Chase, Contr. U.S. Natl. Herb. 15: 29, 84. 1910, nom. inval. Panicum [unranked] Virgata Hitchc., N. Amer. Fl. 3(2): 200, 203. 1915.
Annual or cespitose perennials, culms simple, erect or geniculate ascending. Blades linear to lanceolate, flat or involute. Inflorescence an open, oblong to ovate panicle. Spikelets gaping at maturity, silky villous to pilose or glabrous; lower glume ¾ to the full length of the spikelet, upper glume and lower lemma subequal, longer than the upper anthecium, (5-)7-9(-11) nerved; lower palea conspicuous and lower flower male; upper anthecium indurate, smooth, shiny. (Fig 7).
Panicum phragmitoides. A. Habit. B. Detail of ligule. C. Spikelet, lateral view. D. Spikelet, upper glume view. E. Upper anthecium and lower lemma. F. Upper anthecium, lemma view. G. Upper anthecium, palea view. H. Upper palea with lodicules and anthers. I. Lodicules, stamens and gynoecium.
The section includes fourteen perennial American species, and nearly 24 species in the Old World, five of them annual; they are usually found in open and dry or mesophytic environments.
4. Panicum sect. Panicum
- Panicum sect. Capillare (Hitchc.) Fernald, Rhodora 21(246): 110. 1919. Panicum [unranked] Capillaria Hitchc., N. Amer. Fl. 3(2): 200, 206. 1915. Panicum group Capillaria Hitchc. & Chase, Contr. U.S. Natl. Herb. 15: 28, 54. 1910, nom. inval.
- Panicum group Diffusa Hitchc., Contr. U.S. Natl. Herb. 15: 29, 71. 1910, nom. inval. Panicum [unranked] Diffusa Hitchc., N. Amer. Fl. 17(3): 200, 203. 1915.
- Panicum sect. Miliaceae Stapf, Fl. Trop. Afr. 9(4): 640, 646. 1920.
Annual or cespitose perennials, with culms erect, occasionally decumbent. Blades oblong-lanceolate to filiform, flat or with involute margins. Inflorescence an open and lax terminal panicle, axillary panicles occasionally present. Spikelets ovoid to long-ellipsoid, glabrous or pilose, lower glume (1/3-)1/2-3/4(-4/5) the length of the spikelet, 3-5(-9) nerved, upper glume and lower lemma (5-)7-9(-15) nerved; lower palea present or reduced to absent; lower flower male or absent; upper anthecium indurate, smooth, shiny, with simple or compound papillae toward the apex or simple papillae all over its surface. (Fig 8).
A. Habit. B. Spikelet, lateral view. C. Spikelet, lower glume view. D. Spìkelet, upper glume view. E. Lower palea. F. Upper anthecium, dorsal view. G. Upper anthecium, palea view. H. Caryopsis, embryo view. I. Caryopsis, hilum view.
The section includes 30 American species, and ca. 28 growing in Africa, India, Asia, islands of the Pacific and Australia. They are most commonly found in dry and open areas.
Panicum venosum Swallen, a species transferred to the genus Urochloa [108], belongs to sect. Panicum and is strongly related in this analysis to P. alatum.
- 5. Panicum sect. Repentia Stapf, Fl. Trop. Afr. 9: 640, 648. 1920. Type species: Panicum repens L.
- Panicum sect. Coloratae Stapf, Fl. Trop. Afr. 9: 641, 648. 1920.
Perennials, occasionally annuals, with stout rootstocks, culms erect. Blades oblong-lanceolate to lanceolate, flat to involute. Inflorescence a terminal, open to contracted panicle. Spikelets long-ovoid to ellipsoid, glabrous; lower glume ¼ to 1/3(-1/2) the length of the spikelet, 1-5(-7) nerved, upper glume and lower lemma 9–11 nerved; lower palea present, lower flower male or absent; upper anthecium indurate, smooth and shiny. (Fig 9E–9H).
Panicum majusculum. A. Spikelet, lower glume view. B. Spikelet, upper glume view. C. Upper anthecium, palea view with stipe. D. Upper anthecium, lemma view. Sect. Repentia. Panicum repens. E. Spikelet, upper glume view. F. Spikelet, lower glume view. G. Upper anthecium, palea view. H. Upper anthecium, lemma view. I. Upper palea and lodicules.
The section includes two species of America (P. coloratum and P. repens introduced), and 18 species growing in the Old World, that inhabit mesophytic environments.
- 6. Panicum sect. Rudgeana (Hitchc.) Zuloaga, Ann. Missouri Bot. Gard. 74: 470. 1987. Type species: Panicum rudgei Roem. & Schult.
- Panicum [unranked] Rudgeana Hitchc., N. Amer. Fl. 17(3): 200, 205. 1915.
Annual or cespitose perennials, with erect culms. Blades lanceolate to linear-lanceolate. Inflorescence a single, terminal and lax panicle, axillary inflorescences occasionally present. Spikelets obovoid to ellipsoid, pilose to glabrous, upper glume and lower lemma subequal, 5-9(-11) nerved; lower palea present, lower flower male or absent; upper anthecium stipitate, indurate, smooth, and with compound papillae at the apex. (Fig 10).
Panicum ligulare. A. Habit. B. Spikelet, lateral view. C. Spikelet, lower glume view. D. Spikelet, upper glume view. E. Lower palea, dorsal view. F. Lower palea, ventral view. G. Upper anthecium dorsal view, with stipe. H. Upper anthecium, ventral view. I. Upper anthecium, lateral view.
The section includes five species, two, P. cayennense Lam. and P. rudgei Roem. & Schult., widespread from Central to South America, and P. cervicatum, P. ligulare and P. campestre Nees ex Trin. growing in savannas of South America, ranging from sea-level up to near 1500 m elevation.
- 7. Panicum sect. Yakirra (Lazarides & R.D. Webster) Zuloaga, comb. nov. Yakirra Lazarides & R.D. Webster, Brunonia 7(2): 292. 1985. Type species: Panicum pauciflorum R.Br. [= Yakirra pauciflora (R. Br.) Lazarides & R.D. Webster]
Annual or perennial, the culms ascending or erect, branching at the lower nodes; internodes hollow, glabrous. Ligules membranous-ciliate. Blades linear to lanceolate, flat. Inflorescence an open, usually diffuse, panicle. Spikelets glabrous, with the rachilla manifest between the bracts and below the stipitate upper anthecium; lower glume 3-5-nerved, ½ or more the length of the spikelet; upper glume and lower lemma subequal, 7-9-nerved; lower palea reduced and lower flower absent; upper anthecium indurate, smooth, glabrous. (Fig 9A–9D).
The section includes seven species and one variety, mostly growing in open and dry areas of Australia (with one species, P. foliolosum, also present in Myanmar). Combinations for three taxa not previously named in Panicum are made below.
- Panicum australiense Domin var. intermedium (R.D. Webster) Zuloaga, comb. nov. Yakirra australiense (Domin) Lazarides & R.D. Webster var. intermedia R.D. Webster, Austral. Paniceae: 266. 1987. Type: Australia. Western Australia, near Lucky Hill, 23 km NNE of Dunham River, 13 Mar 1978, M. Lazarides 8547 (holotype, CANB!; isotype, PERTH!).
- Panicum nullum (Lazarides & R.D. Webster) Zuloaga, comb. nov. Yakirra nulla Lazarides & R.D. Webster, Brunonia 7(2): 295. 1985. Type: Australia. Northern Territory: Darwin & Gulf District: 8 miles NE of Adelaide River township, 17 Mar 1965, M. Lazarides & E. D. Adams 262 (holotype, CANB!; isotypes, DNA!, K!, L!, NT!).
- Panicum websterii (B.K. Simon) Zuloaga, comb. nov. Yakirra websteri B.K. Simon, Austrobaileya 3(4): 602, Fig 9. 1992. Type: Australia. Queensland: Mitchell District: 93 km N of Langlo Crossing, 1 Jul 1975, G. R. Beeston 1361C (holotype, BRI-AQ 268164!; isotypes, BRI!, CANB!, K!, NSW!).
Supporting information
S1 Appendix. List of taxa of the molecular analysis and GenBank accession numbers.
New sequences are denoted by * and the voucher information is given (DOCX).
https://doi.org/10.1371/journal.pone.0191529.s001
(DOCX)
S1 Table. Areas and dispersal probabilities used in BioGeoBEARS analyses of subtribe Panicinae.
Area names in rows and columns are: A, North America; B Central and South America; C, Eurasia + Mediterran + North Africa; D, Tropical and South Africa; E, Southern Asia; F, Australia. For dispersal events the ancestral areas (where the lineage dispersed from) are given in the row, and the descendent areas (where the lineage dispersed to) are given in the column. (DOCX)
https://doi.org/10.1371/journal.pone.0191529.s002
(DOCX)
S2 Table. Number of biogeographical events estimated in the history of the subtribe Panicinae using biogeographical stochastic mapping.
Event counts were averaged across 1000 BSMs and are presented here with the standard deviations in parentheses. Total event counts are given for range-expansion events (anagenetic dispersal) (a), founder events (cladogenetic dispersal) (b), and Sympatry speciation events (c). For dispersal events the ancestral areas (where the lineage dispersed from) are given in the row, and the descendent areas (where the lineage dispersed to) are given in the column. The percentages of events involving each area either as a source (the rows) or as the destination (the columns), are given on the margins. Area names in rows and columns are: A, North America; B Central and South America; C, Eurasia + Mediterran + North Africa; D, Tropical and South Africa; E, Southern Asia; F, Australia. (DOCX)
https://doi.org/10.1371/journal.pone.0191529.s003
(DOCX)
S1 Fig. Maximum clade credibility (MCC) tree of Panicoideae obtained from BEAST analyses with the ndhF sequences, using the uncorrelated lognormal relaxed clock model and secondary calibrations based on external angiosperm fossils together with the phytolith microfossils of Poaceae (calibration scheme 2).
Red boxes indicate phylogenetic placement of Panicum species recovered outside subtribe Panicinae. Posterior probability ≥ 0.9 are shown on the branches and horizontal bars on the nodes indicate the 95% HPD of ages. Subtribe Panicinae are shown in detail in S2 Fig Mya, million years ago; Pli, Pliocene; Plei, Pleistocene. Results from divergence time estimation based only on external angiosperm fossils calibration are shown in Fig 1. (PDF)
https://doi.org/10.1371/journal.pone.0191529.s004
(PDF)
S2 Fig. Divergence time estimations for subtribe Panicinae.
Maximum clade credibility (MCC) tree of Panicoideae obtained from BEAST analyses with ndhF sequences using the uncorrelated lognormal relaxed clock model and secondary calibrations based on external angiosperm fossils together with the phytolith microfossils of Poaceae (calibration scheme 2, see materials and methods). Only subtribe Paniceae is shown in detail; for the remaining clades see S1 Fig Posterior probabilities ≥ 0.9 are shown on the branches and horizontal bars on the nodes indicate the 95% HPD of ages. Vertical bars indicate sections within Panicum. Paniceae 1 refers to tribe Paniceae excluding subtribes Cenchrinae, Melinidinae, and Panicineae. Mya, million years ago; Pli, Pliocene; Plei, Pleistocene. Results from divergence time estimation based only on external angiosperm fossils calibration are shown in Fig 2. (PDF)
https://doi.org/10.1371/journal.pone.0191529.s005
(PDF)
S3 Fig. Ancestral range estimation (ARE) on the Panicinae chronogram using the BayArea+J model in BioGeoBEARS.
Pie charts at nodes show the relative probability of the possible states (areas or combination of areas) before the instantaneous speciation event, whereas those on branches represent probability of the descendant lineage immediately after speciation. Boxes to the left of taxon names indicate areas of tip species. (PDF)
https://doi.org/10.1371/journal.pone.0191529.s006
(PDF)
Acknowledgments
We are grateful to Vladimiro Dudás, Marcelo Moreno and Francisco Rojas for preparing the illustrations. The authors sincerely thank the editor of this manuscript, Rob Soreng and an anonymous reviewer for comments that greatly strengthened the manuscript.
References
- 1. GPWG. Phylogeny and subfamilial classification of the Poaceae. Ann. Missouri Bot. Gard. 2001; 88: 373–457.
- 2. Sánchez-Ken JG, Clark LG, Kellogg EA, Kay EE. Reinstatement and emendation of subfamily Micrairoideae (Poaceae). Syst. Bot. 2007; 32: 71–80.
- 3. GPWG II. New grass phylogeny resolves deep evolutionary relationships and discovers C4 origins. New Phytol. 2012; 193: 304–312. pmid:22115274
- 4. Vorontsova MS, Simon BK. Updating classifications to reflect monophyly: 10 to 20 percent of species names change in Poaceae. Taxon. 2012: 61: 735–746.
- 5. Soreng RJ, Peterson PM, Romaschenko K, Davidse G, Zuloaga FO, Judziewicz E, et al. A Worldwide Phylogenetic Classification of the Poaceae (Gramineae). J. Syst. Evol. 2015; 53(2): 117–137.
- 6. Soreng RJ, Peterson PM, Romaschenko K, Davidse G, Teisher JK, Clark LG, et al. A worldwide phylogenetic classification of the Poaceae (Gramineae) II: an update and a comparison of two 2015 classifications. J. Syst. Evol. 2017;
- 7. Pelser PB, Nordenstam B, Kadereit JW. Watson LE. An ITS phylogeny of tribe Senecioneae (Asteraceae) and a new delimitation of Senecio L. Taxon. 2007; 56: 1077–1104. https://doi.org/10.2307/25065905.
- 8. Luckow M, Hughes C, Schrire B, Winter P, Fagg C, Fortunato R, et al. Acacia: The case against moving the type to Australia. Taxon. 2005; 54: 513–519. https://doi.org/10.12705/633.2.
- 9. Miller JT, Seigler D, Mishler BD. A phylogenetic solution to the Acacia problem. Taxon. 2014; 63: 653–658. https://doi.org/10.12705/633.2.
- 10. Nesom GL. Review of the taxonomy of Aster sensu lato (Asteraceae: Astereae), emphasizing the New World species. Phytologia. 1994; 77: 141–297.
- 11. Li WP, Yang FS, Jivkova T, Yin GS. Phylogenetic relationships and generic delimitation of Eurasian Aster (Asteraceae: Astereae) inferred from ITS, ETS and trnL-F sequence data. Ann. Bot. (Oxford). 2012; 109: 1341–1357. https://doi.org/10.1111/boj.12193.
- 12. Clayton WD, Renvoize SA. Genera graminum. Grasses of the world. Kew Bull., Addit. Ser. 1986; 13: 1–389.
- 13.
Clayton WD, Vorontsova MS, Harman KT, Williamson H. GrassBase—The online World grass flora: The Board of Trustees, Royal Botanic Gardens [online]. Available from http://www.kew. org/data/grasses-db.html [accessed 20 July 2017].
- 14.
Simon BK. GrassWorld [online]. Available from http://grassworld.myspecies.info/[accessed 10 April 2017].
- 15. Hackel J, Vorontsova M, Nanjarisoa O, Hall R, Razanatsoa J, Malakasi P, et al. Grass diversification in Madagascar: in situ radiation of two large C3 shade clades and support for a Miocene to Pliocene origin of C4 grassy biomes. J. Biogeogr- In press.
- 16.
Kellogg EA. Flowering plants, Monocots, Poaceae. In: Kubitski K., editor, The families and genera of vascular plants. Heidelberg: Springer International. 2015; 13:1–416.
- 17. Nicola MV, Lizarazu MA, Scataglini MA. Phylogenetic analysis and taxonomic position of section Verrucosa of Panicum and its relationship with taxa of the Sacciolepis-Trichanthecium clade (Poaceae: Panicoideae: Paniceae). Pl. Syst. Evol. 2015; 301: 2247–2260. https://doi.org/10.1007/s00606-015.
- 18.
Zuloaga FO. Systematics of New World Species of Panicum (Poaceae: Paniceae). In: Soderstrom TR, Hilu KW, Campbell CS, Barkworth ME, editors. Grass Systematics and Evolution. Smithsonian Institution Press. Washington, DC; 1987. pp. 287–306.
- 19. Brown WV. The Kranz syndrome and its subtypes in grass systematics. Mem. Torrey Bot. Club. 1977; 23: 1–97.
- 20. Ellis RP. Leaf anatomy and systematics of Panicum (Poaceae: Panicoideae) in Southern Africa. Monogr. Syst. Bot. Missouri Bot. Gard. 1988; 25: 129–156.
- 21. Acosta J, Scataglini MA, Reinheimer R, Zuloaga FO. A phylogenetic study of subtribe Otachyriinae (Poaceae, Panicoideae, Paspaleae). Pl. Syst. Evol. 2014; 300: 2155–2166.
- 22.
Gómez-Martínez R, Culham A. Phylogeny of the subfamily Panicoideae with emphasis on the tribe Paniceae: evidence from the trnL-F cpDNA region. In: Jacobs SWL, Everett JE, editors. Grasses: Systematics and Evolution. Commonwealth Scientific and Industrial Research Organization (CSIRO) Publishing, Victoria, Collingwood; 2000. pp. 136–140.
- 23. Duvall MR, Noll JD, Minn AH. Phylogenetics of Paniceae (Poaceae). Amer. J. Bot. 2001; 88: 1988–1992.
- 24. Duvall MR, Saar DE, Grayburn WS, Holbrook GP. Complex transitions between C3 and C4 photosynthesis during the evolution of Paniceae: a phylogenetic case study emphasizing the position of Steinchisma hians (Poaceae), a C3—C4 intermediate. Int.l J. Pl. Sci. 2003; 164: 949–958.
- 25. Giussani LM, Cota-Sánchez JH, Zuloaga FO, Kellogg EA A molecular phylogeny of the grass subfamily Panicoideae (Poaceae) shows multiple origins of C4 photosynthesis. Amer. J. Bot. 2001; 88: 1993–2012.
- 26. Aliscioni SS, Giussani LM, Zuloaga FO, Kellogg EA. A molecular phylogeny of Panicum (Poaceae: Paniceae). Tests of monophyly and phylogenetic placement within the Panicoideae. Amer. J. Bot. 2003; 90: 796–821. https://10.3732/ajb.90.5.796.
- 27. Christin PA, Besnard G, Samaritani E, Duvall MR, Hodkinson TR, Savolainen V, et al. Oligocene CO2 Decline Promoted C4 Photosynthesis in Grasses. Curr. Biol. 2008; 18(1):37–43. pmid:18160293
- 28. Vicentini A, Barber JC, Aliscioni SS, Giussani LM, Kellogg EA. The age of the grasses and clusters of origins of C4 photosynthesis. Global Change Biol. 2008; 14(12):2963–2977.
- 29. Morrone O, Aagesen L, Scataglini MA, Salariato DL, Denham SS, Chemisquy MA, et al. Phylogeny of the Paniceae (Poaceae: Panicoideae): integrating plastid DNA sequences and morphology into a new classification. Cladistics. 2012; 28: 333–356.
- 30.
Watson L, Dallwitz MJ. The grass genera of the world: descriptions, illustrations, identification, and information retrieval; including synonyms, morphology, anatomy, physiology, phytochemistry, cytology, classification, pathogens, world and local distribution, and references, 23rd April 2010 edn. Wallingford: CAB International. 1992 onward. http://delta-intkey.com/grass/
- 31. Grande Allende JR. Novitates agrostologicae, IV. Additional segregates from Panicum incertae sedis. Phytoneuron. 2014; 22: 1–6.
- 32. Zuloaga FO, Giussani LM, Morrone O. Hopia, a new genus segregated from Panicum (Poaceae: Panicoideae: Paniceae). Taxon. 2007; 56: 145–156.
- 33. Sede SM, Zuloaga FO, Morrone O. Phylogenetic studies in the Paniceae (Poaceae-Panicoideae): Ocellochloa, a new genus from the New World. Syst. Bot. 2009; 34: 684–692. http://dx.doi.org/10.1600/036364409790139655
- 34. Sede SM, Morrone O, Giussani LM, Zuloaga FO. Phylogenetic studies in the Paniceae (Poaceae): a realignment of section Lorea of Panicum. Syst. Bot. 2008; 33: 284–300. http://dx.doi.org/10.1600/036364408784571626
- 35. Zuloaga FO, Morrone O, Vega AS, Giussani LM. Revisión y análisis cladístico de Steinchisma (Poaceae: Panicoideae: Paniceae). Ann. Missouri Bot. Gard. 1998; 85: 631–656.
- 36. Zuloaga FO, Giussani LM, Morrone O. On the taxonomic position of Panicum aristellum (Poaceae: Panicoideae: Paniceae). Syst. Bot. 2006; 31: 497–505. http://dx.doi.org/10.1043/05-56.1
- 37. Soreng R. Coleataenia Griseb. (1879): the correct name for Sorengia Zuloaga & Morrone (2010) (Poaceae: Paniceae). J. Bot. Res. Inst. Tex. 2010; 4: 691–692.
- 38. Scataglini M A, Aliscioni SS, Zuloaga FO. On the taxonomic position of Panicum scabridum (Poaceae, Panicoideae, Paspaleae). Phytotaxa. 2014; 163(1): 1–15. http://dx.doi.org/10.11646/phytotaxa.163.1.1
- 39. Scataglini MA, Zuloaga FO. Cyphonanthus, a new genus segregated from Panicum (Poaceae: Panicoideae: Paniceae) based on morphological, anatomical and molecular data. Taxon. 2007; 56: 521–532.
- 40. Zuloaga FO, Soderstrom TR. Classification of the outlying species of New World Panicum (Poaceae: Paniceae). Smithsonian Contr. Bot. 1985; 59: 1–63.
- 41. Zuloaga FO, Scataglini MA, Morrone O. 2010. A phylogenetic evaluation of Panicum sects. Agrostoidea, Megista, Prionitia and Tenera (Panicoideae, Poaceae): two new genera, Stephostachys and Sorengia. Taxon. 2010; 59: 1535–1546.
- 42. Wills KE, Whalley RDB, Bruhl JJ. Systematic studies in Paniceae (Poaceae): Homopholis and Whalleya gen. et sp. nov. Austral. Syst. Bot. 2000; 13(3): 437–468.
- 43. Zuloaga FO, Morrone O, Scataglini MA. Monograph of Trichanthecium, a new genus segregated from Panicum (Poaceae, Paniceae) based on morphological and molecular data. Syst. Bot. Monogr. 2011; 94: 1–101.
- 44. Zuloaga FO, Salomón L, Scataglini MA. Phylogeny of sections Clavelligerae and Pectinatae of Panicum (Poaceae, Panicoideae, Paniceae). Pl. Syst. Evol. 2015; 301: 1693–1711.
- 45. Scataglini MA, Zuloaga FO. Morronea, a new genus segregated from Panicum (Paniceae, Poaceae) based on morphological and molecular data. Syst. Bot. 2013; 38(4): 1–11.
- 46. Morrone O, Denham SS, Aliscioni SS, Zuloaga FO. Parodiophyllochloa, a new genus segregated from Panicum (Paniceae, Poaceae) based on morphological and molecular data. Syst. Bot. 2008; 33: 66–76. http://dx.doi.org/10.1600/036364408783887393.
- 47. Lazarides M. The genus Whiteochloa C.E. Hubbard (Poaceae, Paniceae). Brunonia. 1978; 1: 69–93.
- 48. Bess E, Doust ANL, Davidse G, Kellogg EA. Zuloagaea, a new genus of neotropical grass within the “Bristle Clade” (Poaceae: Paniceae). Syst. Bot. 2006; 31: 656–670.
- 49. Simon BK, Jacobs SWL. Megathyrsus, a new generic name for Panicum subgenus Megathyrsus. Austrobaileya. 2003; 6: 571–574.
- 50. Scataglini MA, Lizarazu MA, Zuloaga FO. A peculiar amphitropical genus of Paniceae (Poaceae, Panicoideae). Syst. Bot. 2014; 39(4): 1108–1119.
- 51. Hubbard CE, Léonard J. A new genus of Gramineae from tropical Africa. Bull. Jard. Bot. État Bruxelles. 1952; 22: 312–318.
- 52. Triplett JK, Wang Y, Zhong J, Kellogg EA. Five Nuclear Loci Resolve the Polyploid History of Switchgrass (Panicum virgatum L.) and Relatives. PloS One 2012; 7(6): 1–19.
- 53. Burke S., Wysocki WP, Zuloaga FO, Craine JM, Pires JC, Edger PP, et al. Evolutionary relationships in panicoid grasses based on plastome phylogenomics. BMC Plant Biol. 2016; 16: 140. pmid:27316745
- 54. Christin PA, Salamin N, Kellogg EA, Vicentini A, Besnard G. Integrating phylogeny into studies of C4 variation in the grasses. Pl. Physiol. 2009; 149: 82–87.
- 55. Zimmermann T, Bocksberger G, Brüggemann W, Berberich T. Phylogenetic relationship and molecular taxonomy of African grasses of the genus Panicum inferred from four chloroplast DNA-barcodes and nuclear gene sequences. J. Pl. Res. 2013; 126: 363–371.
- 56. Salariato DL, Zuloaga FO, Giussani LM, Morrone O. Molecular phylogeny of the C4 –PCK photosynthetic subtype group (Poaceae: Panicoideae: Paniceae) and evolutionary trends in the homogenization of inflorescences. Molec. Phylogen. Evol. 2010; 56: 355–369.
- 57. Silva C, Snak C, Schnadelbach AS, van den Berg C, Oliveira RP. Phylogenetic relationships of Echinolaena and Ichnanthus within Panicoideae (Poaceae) reveal two new genera of tropical grasses. Molec. Phylogen. Evol. 2015; 93: 212–233.
- 58. Vorontsova MS, Besnard G, Forest F, Malakasi P, Moat J, Clayton WD, et al. Madagascar’s grasses and grasslands: anthropogenic or natural? Proc. R. Soc. London B: Biol. Sci. 2016; 283: 20152262.
- 59. Doyle JJ., Doyle JL. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. Bot. Soc. Amer. 1987; 19: 11–15.
- 60. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. (Oxf.) 1999; 41: 95–98.
- 61. Lanfear R, Frandsen PB, Wright AM, Senfeld T, Calcott B. PartitionFinder 2: new methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Molec. Biol. Evol. 2016; pmid:28013191
- 62. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 2014; 30: 1312–1313. pmid:24451623
- 63. Stamatakis A., Hoover P., Rougemont J. A rapid bootstrap algorithm for the RAxML web-servers. Syst. Biol. 2008; 57, 758–771. http://dx.doi.org/10.1080/10635150802429642. pmid:18853362
- 64. Drummond AJ, Suchard MA, Xie D., Rambaut A. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molec. Biol.Evol. 2012; 29: 1969–1973. pmid:22367748
- 65. Stadler T. On incomplete sampling under birth–death models and connections to the sampling-based coalescent. J. Theor. Biol. 2009; 261: 58–66. pmid:19631666
- 66. Christin PA, Spriggs E, Osborne CP, Strömberg CAE, Salamin N, Edwards EJ. Molecular dating, evolutionary rates, and the Age of the Grasses. Syst. Biol. 2014; 63: 153–165. pmid:24287097
- 67. Prasad V, Strömberg CAE., Leaché AD, Samant B, Patnaik R, Tang L, et al. Late Cretaceous origin of the rice tribe provides evidence for early diversification in Poaceae. Nature Commun. 2011; 2: 480.
- 68.
Rambaut A, Drummond AJ. Tracer version 1.6. Computer program and documentation distributed by the author. 2013. http://beast.bio.ed.ac.uk/Tracer.
- 69.
Miller MA, Pfeiffer W, Schwartz T. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In Gateway Computing Environments Workshop (GCE), 2010; IEEE. pp. 1–8.
- 70. Buerki S, Forest F, Alvarez N, Nylander JA, Arrigo N, Sanmartín I. An evaluation of new parsimony‐based versus parametric inference methods in biogeography: a case study using the globally distributed plant family Sapindaceae. J. Biogeogr. 2011; 38: 531–550.
- 71.
Thiers B [continuously updated, accessed 2017] Index Herbariorum: a global directory of public herbaria and associated staff. New York Botanical Garden’s Virtual Herbarium, http://sweetgum.nybg.org/ih.
- 72.
Matzke NJ BioGeoBEARS: BioGeography with Bayesian (and Likelihood) Evolutionary Analysis in R Scripts. University of California, Berkeley. 2013.
- 73.
R Core Team. R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. 2016. Available from https://www.R-project.org/
- 74. Matzke NJ. Founder-event speciation in BioGeoBEARS package dramatically improves likelihoods and alters parameter inference in Dispersal-Extinction-Cladogenesis (DEC) analyses. Front. Biogeogr. 2012; 4: 210.
- 75. Matzke NJ. Model selection in historical biogeography reveals that founder-event speciation is a crucial process in island clades. Syst. Biol. 2014; 63: 951–970. pmid:25123369
- 76. Dupin J, Matzke NJ, Särkinen T, Knapp S, Olmstead RG, Bohs L, et al. Bayesian estimation of the global biogeographical history of the Solanaceae. J. Biogeogr. 2017; 44: 887–899.
- 77.
Pagel M Meade A. BayesTraits version 2.0.2. 2016. Software <http://www.evolution.rdg.ac.uk/BayesTraitsV2.0.2.html >.
- 78. Plummer M, Best N, Cowles K Vines K. CODA: Convergence diagnosis and output analysis for MCMC. R news 2006; 6, 7–11.
- 79. Huelsenbeck JP, Nielsen R. Bollback JP. Stochastic mapping of morphological characters. Syst. Biol. 2003; 52: 131–158. pmid:12746144
- 80. Revell LJ. phytools: an R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 2012; 3: 217–223.
- 81. Fritz SA, Purvis A. Selectivity in mammalian extinction risk and threat types: a new measure of phylogenetic signal strength in binary traits. Conserv. Biol. 24: 1042–1051. pmid:20184650
- 82.
Orme D., Freckleton R, Thomas G, Petzoldt T, Fritz S, Isaac N, et al. Caper: Comparative Analyses of Phylogenetics and Evolution in R. 2013. R package version 0.5.2. https://CRAN.R-project.org/package=caper
- 83. Hitchcock AS, Chase A. 1910. The North American species of Panicum. Contr. U.S. Natl. Herb. 1910; 15: 1–396.
- 84. Hitchcock AS, Chase A. Tropical North American species of Panicum. Contr. U.S. Natl. Herb. 1915; 17: 459–539.
- 85.
Stapf O. Panicum. In Flora of Tropical Africa, Gramineae (Maydeae-Paniceae), Prain D (editor) England: L. Reeve & Co. Ltd. Kent. 1920; 9: 638–738.
- 86. Zuloaga FO. El género Panicum en la República Argentina. II. Darwiniana. 1981; 23(1): 233–256.
- 87. Lazarides M Webster RD. Yakirra (Paniceae, Poaceae), a new genus for Australia. Brunonia. 1984; 7(2): 289–296.
- 88. Hughes DK. The genus Panicum of the Flora Australiensis. Bull. Misc. Inform. Kew. 1923; 1923(9): 305–332.
- 89. Berg RY. Spikelet structure in Panicum australiense (Poaceae): taxonomic and ecological implications. Austral. J. Bot. 1985; 33: 579–583.
- 90. Lazarides M. New taxa of tropical Australian grasses (Poaceae). Nuytsia. 1984; 5(2): 273–303.
- 91. Honda M. Monographia poacearum japonicarum, bambusoideis exclusis. J. Fac. Sci. Univ. Tokyo, Sect. 3, Bot. 1930; 3(1): 43–55.
- 92. Zuloaga FO. A revision of Panicum subgenus Panicum section Rudgeana (Poaceae: Paniceae). Ann. Missouri Bot. Gard. 1987; 74(3): 463–478.
- 93. Zachos J, Pagani M, Sloan L, Thomas E Billups K. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science. 2001; 292: 686–693. pmid:11326091
- 94. Pessoa-Filho M, Martins AM Ferreira ME. Molecular dating of phylogenetic divergence between Urochloa species based on complete chloroplast genomes. BMC Genomics. 2017; 18: 516. pmid:28683832
- 95. Ogburn RM Edwards EJ. Anatomical variation in Cactaceae and relatives: trait lability and evolutionary innovation. Amer. J. Bot. 2009; 96: 391–408.
- 96. Osborne CP, Beerling DJ, Lomax BH Chaloner WG. Biophysical constraints on the origin of leaves inferred from the fossil record. Proc. Natl. Acad. Sci. U.S.A. 2004; 101: 10360–10362. pmid:15240879
- 97. Christin PA, Osborne CP, Chatelet DS, Columbus JT, Besnard G, Hodkinson TR, et al. Anatomical enablers and the evolution of C4 photosynthesis in grasses. Proc. Natl. Acad. Sci. U.S.A. 2013; 110: 1381–1386. pmid:23267116
- 98. Christin PA, Arakaki M, Osborne CP Edwards EJ. Genetic enablers underlying the clustered evolutionary origins of C4 photosynthesis in angiosperms. Molec. Biol. Evol. 2015; 32: 846–858. pmid:25582594
- 99. Aagesen L, Biganzoli F, Reinheimer R, Bena J, Zuloaga FO. Macro-climatic distribution limits show both Niche expansion and Niche specialization among C4 Panicoids. PloS One. 2016; 11(3): e0151075; pmid:26950074
- 100. Barbier P, Morishima H, Ishihama A. Phylogenetic relationships of annual and perennial wild rice: probing by direct DNA sequencing. Theor. Appl. Genet. 1991; 81: 693–702. pmid:24221388
- 101. Evans ME, Hearn DJ, Hahn WJ, Spangle JM, Venable DL. Climate and life-history evolution in evening primroses (Oenothera, Onagraceae): a phylogenetic comparative analysis. Evolution. 2005; 59: 1914–1927. pmid:16261729
- 102. Friedman J, Rubin MJ. All in good time: understanding annual and perennial strategies in plants. Amer. J. Bot. 2015; 102: 497–499.
- 103. Karl R, Koch MA. A world-wide perspective on crucifer speciation and evolution: phylogenetics, biogeography and trait evolution in tribe Arabideae. Ann. Bot. (Oxford) 2013; 112: 983–1101.
- 104. Ogburn RM, Edwards EJ. Life history lability underlies rapid climate niche evolution in the angiosperm clade Montiaceae. Mol. Phylogenet. Evol. 2015; 92, 181–192. pmid:26143714
- 105. Cole LC. The population consequences of life history phenomena. Q. Rev. Biol. 1954; 29: 103–137. pmid:13177850
- 106.
Schaffer WM, Gadgil M. Selection for optimal life histories in plants. In: Cody M., Diamond J(editors). The ecology and evolution of communities. Cambridge (MA): Harvard University Press. 1975. pp. 142–157.
- 107. Simon BK. Gondwanan grasses in the Australian Flora. Austrobaileya. 1990; 3(2): 239–260.
- 108. Morrone O, Zuloaga FO.Sinopsis del género Urochloa (Poaceae: Panicoideae: Paniceae) para México y América Central. Darwiniana. 1993; 32(1–4): 59–75.