TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum Molecular phylogeny of Acanthophyllum (Caryophyllaceae: Caryophylleae), with emphasis on subgeneric classification Atefeh Pirani,1,2 Shahin Zarre,1 Bernard Pfeil,2 Yann J.K. Bertrand,2 Mostafa Assadi3 & Bengt Oxelman2 1 Department of Plant Biology, and Center of Excellence in Phylogeny of Living Organisms, School of Biology, College of Science, University of Tehran, P.O. Box 14155-6455 Tehran, Iran 2 Department of Biological and Environmental Sciences, University of Gothenburg, Box 461, 40530 Göteborg, Sweden 3 Research Institute of Forests and Rangelands, P.O. Box 13185-116, Tehran, Iran Authors for correspondence: Atefeh Pirani, pirani@khayam.ut.ac.ir; Shahin Zarre, zarre@khayam.ut.ac.ir DOI http://dx.doi.org/10.12705/633.38 Abstract Despite being one of the larger genera of Caryophyllaceae with about 60 cushion-forming subshrubby species, Acanthophyllum is represented poorly in previous molecular phylogenetic studies. The genus is an important component of the subalpine steppe flora in Central to Southwest Asia. Although the placement of Acanthophyllum in the tribe Caryophylleae and a close relationship to Allochrusa has already been suggested, the monophyly of the genus and its subgeneric taxa, as well as its relation to other closely related genera, have not been addressed. We have assembled datasets of nuclear ribosomal internal transcribed spacer (ITS) sequences and intron sequences of the chloroplast gene rps16 for 47 Acanthophyllum species and 63 species of 12 additional genera from Caryophylleae. Phylogenetic analyses were performed using maximum parsimony, maximum likelihood and Bayesian methods. Our analysis suggests that Allochrusa, Diaphanoptera, Ochotonophila and Scleranthopsis are nested within Acanthophyllum but that the traditionally recognized sections of Acanthophyllum are monophyletic after reassignment of a few species. Emarginate petals may be a synapomorphy for one of the two basal clades of Acanthophyllum. Moreover, non-monophyly of the genera Gypsophila and Diaphanoptera is suggested by the present study. The age of the crown clade of Acanthophyllum s.l. is estimated to be 11.1 Ma by *BEAST species tree analysis. Keywords Acanthophyllum; Caryophyllaceae; Irano-Turanian; ITS; molecular phylogeny; rps16; species tree INTRODUCTION Acanthophyllum C.A.Mey., with ca. 60 species, is a mainly Irano-Turanian genus (Bittrich, 1993; Ghaffari, 2004) that inhabits areas between Syria and western China (Ghaffari, 2002). The diversity center of the genus is the Khorassan-Kopet dagh floristic province in NE Iran and neighboring areas in Afghanistan and Turkmenistan (Ghaffari, 1989, 2004; MahmoudiShamsabad & al., 2012). Acanthophyllum species are small, shrubby, cushionforming perennials with spiny leaves. They grow in exposed habitats on sandy or stony hills and rocky slopes (Fig. 1A–B). They are important components of the steppe (Zohary, 1973) and mountain vegetation in Central and Southwest Asia. Several biologically active triterpene saponins have been reported from different species of Acanthophyllum (Gaidi & al., 2000, 2004). The genus could be medicinally interesting due to the highly cytotoxic properties of many saponins. It has been shown that cytotoxic compounds have potential antitumor activity (Sparg & al., 2004). Saponins of some Acanthophyllum species have been suggested as substitutes for synthetic surfactants in shampoo (Aghel & al., 2007). Although Acanthophyllum is one of the larger genera in the Caryophyllaceae, it has been subjected to few systematic and phylogenetic studies and no inclusive monograph of the genus is available. The number of species was estimated to 56 in Bittrich (1993), based on Flora Iranica (Schiman-Czeika, 1988), but 14 species should be added based on Flora of the U.S.S.R. (Schischkin, 1936), Flora of Uzbekistan (Vvedensky, 1953) and Flora of Tajikistan (Ovchinnikov, 1968). With a few recently described species (Aytaç, 2001; Mahmoudi-Shamsabad & al., 2012; Pirani & al., 2013), the species number exceeds 70. On the other hand, several synonymies have been suggested by regional studies in Iran and Pakistan (Ghazanfar & Nasir, 1986; Basiri-Esfahani & al., 2011). Clearly, the taxonomy of Acanthophyllum is in need of revision. Boissier (1867) recognized five groups within Acanthophyllum, delimited mainly on the basis of inflorescence and floral features. The basic framework of his classification was followed in the next three major taxonomic treatments of the genus by Golenkin (1893), Schischkin (1936) and SchimanCzeika (1988). Golenkin (1893) recognized 19 species, including the genus Allochrusa Bunge as a sixth section. Schischkin (1936) classified the genus into two subgenera (Euacanthophyllum and Allochrusa), recognizing two sections in subg. Allochrusa (Bunge) Schischk. Schiman-Czeika (1988) transferred sect. Pseudacanthophyllum (Boiss.) Rech.f. from Gypsophila L. to Acanthophyllum and a new section, Scapiflora Received: 2 Jul 2013 | returned for first revision: 6 Sep 2013 | revision received: 11 Feb 2014 | accepted: 13 Feb 2014 | not published online ahead of inclusion in print and online issues || © International Association for Plant Taxonomy (IAPT) 2014 Draft version for proof purposes only. 1 Schiman-Czeika, was introduced. Acanthophyllum sect. Scapiflora includes four species, among which two were transferred from Gypsophila and Saponaria L. A comparison of the four major subgeneric classifications of Acanthophyllum (with type for each section) is given in Table 1. Hitherto used characters in infrageneric classifications of Acanthophyllum are mostly related to inflorescence, shape and texture of bracts, and number of ovules. The general inflorescence type in Acanthophyllum is dichasial cymes arranged in terminal heads and/or axillary verticillasters. The exception is Fig.  (in two parts). A–B, Natural habitats of Acanthophyllum; C–E, A. speciosum (sect. Oligosperma); F–G, A. gracile (sect. Macrostegia); H, A. bracteatum (sect. Macrostegia); I–K, A. crassifolium (sect. Acanthophyllum); L–M, A. glandulosum (sect. Pleiosperma); N, A. spinosum (sect. Pleiosperma); O–P, A. caespitosum (sect. Oligosperma). — Photos by Hamid Moazzeni & Atefeh Pirani. 2 Draft version for proof purposes only. red indicates changes by me check and confirm TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum sect. Macrodonta Boiss. that includes species with 1-flowered (rarely up to 3-flowered) terminal inflorescences. Among inflorescence characters applied in sectional classification are (1) 1-flowered versus multi-flowered inflorescences, (2) terminal versus elongated inflorescences and (3) richly branched versus poorly branched inflorescences (richly branched inflorescences are composed of several partial dichasial units distributed along the stem or confined to several apical nodes of the stem, whereas poorly branched inflorescences are composed of one, or rarely up to three condensed dichasial unit(s) confined to only one terminal node of the stem; Fig. 2A–B). Inflorescence variation in A. sect. Acanthophyllum, sect. Macrostegia Boiss., sect. Oligosperma Schischk. and sect. Pleiosperma Boiss. is shown in Fig. 1C–P. As with many other genera in the family (Kurtto, 2001; Oxelman & al., 2001; Fior & al., 2006), the limits of Acanthophyllum in relation to closely related genera are controversial. There have been several transfers of species among Acanthophyllum, Gypsophila, Saponaria and Dianthus L. by different authors. Moreover, the relationships to the small genera, Scleranthopsis Rech.f., Allochrusa, Ochotonophila Gilli, Diaphanoptera Rech.f., and Kuhitangia Ovczinn. need to be elucidated. Draft version for proof purposes only. 3 TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum Acanthophyllum has been included with few samples in a small number of molecular studies of Caryophyllaceae. In a phylogenetic analysis based on the chloroplast DNA matK region, Allochrusa versicolor Boiss. is sister to Acanthophyllum sordidum Bunge ex Boiss. (Fior & al., 2006), but as only one Acanthophyllum species was included, it is not possible to judge whether Allochrusa should be retained or not. Greenberg & Donoghue (2011) showed that All. versicolor, A. sordidum and A. paniculatum Regel & Herder make a monophyletic clade, reinforcing the idea that the two genera are closely related. Allochrusa was sister to A. sordidum, thus rendering Acanthophyllum paraphyletic. The latter work is the most comprehensive molecular study of Caryophylleae to date. In the few divergence time studies that included Acanthophyllum, the approximate age of the Acanthophyllum-Allochrusa clade was estimated to be 2.88 Ma (Frajman & al., 2009) or between 3 and 9 Ma (Valente & al., 2010). Since only one species each of Allochrusa and Acanthophyllum were sampled in these studies, there is a need for a focused study using a wider sampling. The main goals of this study are: (1) to test the monophyly of Acanthophyllum and its infrageneric taxa using DNA sequence data; (2) to evaluate the infrageneric classification of Acanthophyllum in light of molecular data; (3) to check the evolution of morphological characters on the background of the molecular phylogeny; (4) to estimate the divergence time of the crown node of Acanthophyllum; (5) to investigate the phylogenetic position of Acanthophyllum within tribe Caryophylleae. Table . Major infrageneric classifications of Acanthophyllum including types for sectional taxa. not in lit. cit. why do some empty cells do not have a dash? Schischkin (1936) Subg. Euacanthophyllum (Boiss.) Schischk.a Subg. Allochrusa (Bunge) Schischk. — Boissier (1867) Golenkin (1893) Schiman-Czeika (1988) Type Sect. Macrodonta Sect. Macrodonta — Sect. Macrodonta Boiss. A. grandiflorum Stocks Sect. Macrostegia Sect. Macrostegia Boiss. Sect. Macrostegia Boiss. Sect. Macrostegia Boiss. A. bracteatum Boiss. Sect. Turbinaria Sect. Turbinaria Boiss. Sect. Turbinaria Boiss. Sect. Acanthophyllum A. mucronatum C.A.Mey. Sect. Pleiosperma Sect. Pleiosperma Boiss. Sect. Pleiosperma Boiss. Sect. Pleiosperma Boiss. A. spinosum (Desf.) C.A.Mey. Sect. Euacanthophylla Sect. Euacanthophylla Sect. Oligosperma Schischk. Sect. Oligosperma Schischk. A. squarrosum Boiss. — — — — Sect. Pseudacanthophyllum A. laxiflorum (Boiss.) Rech.f. Boiss. Syn.: Gypsophila sect. Pseudacanthophyllum Boiss. — — — — Sect. Scapiflora Schiman-Czeika A. scapiflorum (Akhtar) Schiman-Czeika — — Sect. Versicoloria Schischk. — A. versicolor Fisch. & C.A.Mey. — Sect. Paniculata Golenk. Sect. Paniculata Golenk. — A. paniculatum Regel & Herder a 4 Acanthophyllum when described by Meyer (1831) included only one species which should automatically serve as the type of the genus. Boissier (1867) made the first infrageneric classification of the genus including five sections (indicated by “§”). He was not aware of typification rules and assigned the type to the genus in sect. Turbinaria. Most of earlier taxonomists ignored to put the type of the genus in section Acanthophyllum (or sect. Euacanthophylla). Schiman-Czeika (1988) recognized the problem and defined section Acanthophyllum with A. mucronatum as the type and including formerly described sect. Turbinaria. Draft version for proof purposes only. check all tables for conversion errors TAXON — 26 May 2014: 16 pp. Sampling strategy and plant material. — All genera from Caryophylleae represented in former phylogenetic studies plus four small genera allied to Acanthophyllum were included. The genus Silene L. was chosen as outgroup (Sileneae is well established as sister group to Caryophylleae; Harbaugh & al., 2010; Greenberg & Donoghue, 2011). New sequences were obtained from specimens deposited at FUMH, GAZI, GB, M, MSB, TARI, TMRC and TUH; a total of 108 sequences representing 47 species of Acanthophyllum, 4 of Diaphanoptera, 2 of Allochrusa, 1 of Ochotonophila, 1 of Scleranthopsis and 1 of Gypsophila. All sections of Acanthophyllum recognized by Schiman-Czeika (1988) and Schischkin (1936), covering the whole morphological variation within the genus, have been included. Amplification of the rps16 region for A. lilacinum Schischk. and D. afghanica Podlech failed. There was no material of A. paniculatum available, therefore only and ITS sequence for this species could be obtained from GenBank. Other ITS and rps16 sequences representing 21 Gypsophila, 3 Saponaria, 23 Dianthus, 4 Petrorhagia (Ser.) Link, 1 Velezia L., 1 Psammosilene W.C.Wu & C.Y.Wu, 1 Vaccaria Medik., 1 Bolbosaponaria Bondarenko and 17 Silene species were obtained from GenBank. In several cases only the ITS or the rps16 sequence was available for a species, so the ITS and rps16 datasets do not strictly match. The final dataset for ITS contained 117 sequences representing 117 taxa and for rps16 72 sequences representing 72 taxa. Voucher information is listed in Appendix 1. DNA extraction, PCR, and sequencing. — Genomic DNA was extracted from herbarium specimens using the E.Z.N.A. SP Plant DNA Mini Kit (company, city, state country) according to the manufacturer’s protocol, or using a modified Carlson/Yoon method (Oxelman & Lidén, 1995; Rautenberg & al., 2010). The nuclear ribosomal internal transcribed spacer (ITS) region was amplified using primer pairs P17/26S-82R (Popp & Oxelman, 2001; Kool & al., 2012). The complete intron of the plastid rps16 gene was amplified using primer pairs rpsF/rpsR2R (Oxelman & al., 1997; Petri & Oxelman, 2011; Kool & al., 2012) or rpsF/ rpsR3R. Multiscreen PCR (Millipore, city, state, country) was used to purify amplification products, according to the manufacturer’s instructions. The ITS region was sequenced using primer pairs P16/ITS4 (Eggens & al., 2007; Popp & Oxelman, 2007), whereas the rps16 region was sequenced using primer pairs rpsF2a/rpsR3R (Popp & al., 2005). Sequencing was performed by Macrogen (Amsterdam, Netherlands). Sequence assembly, alignment and analysis. — Sequence editing, contig assembly and alignments were performed using Geneious v.5.5.8. For multiple alignments, we used the MUSCLE plug-in with default settings. The alignments were checked and adjusted manually. Indels were coded using SeqState v.1.4.1 (Müller, 2005), under the simple indel-coding option (Simmons & Ochotorena, 2000). The best substitution model for each alignment was selected using jModelTest v.0.1.1 (Posada, 2008), under the Bayesian information criterion (BIC). The GTR + G model was determined as the best-fit model for both nuclear and chloroplast markers. The binary data model (Lewis, 2001), as implemented in MrBayes, was used for indel characters. Bayesian inference (BI) of the individual gene analyses was performed using MrBayes v.3.1.2 (Huelsenbeck & Ronquist, 2001), with default prior settings, for ten million MCMC generations in four parallel runs, each with four parallel Fig. . A, richly branched inflorescence; B, poorly branched inflorescence; C, dentate petal; D, emarginate petal; E, bifid petal; F, entire petal; G, fimbriate petal; H, sinuate petal. — A–B after Schiman-Czeika, 1988; C–H after Castroviejo & al. (1990). Draft version for proof purposes only. 5 not in lit.cit. check wording MATERIALS AND METHODS you must submit alignment files either to TreeBase or provide as suppl. data Pirani & al. • Molecular phylogeny of Acanthophyllum Pirani & al. • Molecular phylogeny of Acanthophyllum 6 Ochotonophila allochrusoides A. pleiostegium Sect. Acanthophyllum A. schugnanicum A. glandulosum 6x; ovules 8–12(–16) Sect. Pleiosperma A. crassinodum A. spinosum A. longicalyx Sect. Macrodonta Diaphanoptera lindbergii Diaphanoptera stenocalycina A. kandaharicum A. gracile A. bracteatum 2x; ovules 4 Sect. Macrostegia A. pachycephalum A. leucostegium A. andarabicum A. heratense A. pachystegium A. borsczowii A. squarrosum A. ejtehadii A. diezianum A. speciosum A. heterophyllum Sect. Oligosperma 2x; ovules 4 A. laxiusculum A. yasamin-nassehiae A. stocksianum A. andersenii A. subglabrum A. kabulicum A. adenophorum A. korshinskyi Scleranthopsis aphanantha A. macrodon $JUDQGLÀRUXP Sect. Macrodonta A. xanthoporphyranthum A. anisocladum Sect. Scapiflora A. raphiophyllum A. stewartii Sect. Pseudacanthophyllum $OD[LÀRUXP $VFDSLÀRUXP Sect. Scapiflora Diaphanoptera ekbergii A. sordidum Sect. Pleiosperma 4x; ovules (6–)8(–10) Allochrusa versicolor Allochrusa bungei A. acerosum A. verticillatum A. mucronatum A. microcephalum Sect. Acanthophyllum 4x; ovules 4–8 A. kurdicum A. crassifolium $RSSRVLWLÀRUXP A. caespitosum Sect. Oligosperma 2x; ovules 4 0.84 Clade E Clade G 0.81 0.74 2YXOHV• 0.82 0.83 0.63 0.98 Clade H Clade C Membranous bracts 0.7 0.92 0.97 Clade F 0.54 Clade K 0.66 Clade L 0.61 0.96 Clade J Draft version for proof purposes only. CLADE I 0.46 0.77 0.94 Solitary ÀRZHUV 0.99 0.56 Clade D 0.51 0.88 Clade A 0.96 Poorly branched inflorescence 0.91 0.99 CLADE II 10 7.5 MIOCENE 0.5 0.99 Emarginate petals 12.5 0.71 5 0.76 0 2.5 PLIOCENE QUATERNARY Fig. 3. *BEAST species tree of Acanthophyllum with node bars representing 95% HPD intervals for the root, Clade I and Clade II nodes. Bayesian posterior probabilities are indicated above branches. Values below 50% are not shown. TAXON — 26 May 2014: 16 pp. Note: text of all figures retyped by me - check all figs thoroughly Clade B 0.93 TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum chains for rps16, and four runs each with eight chains under increased temperature of t = 0.1 for the ITS dataset. Tracer v.1.5 (Rambaut & Drummond, 2009) was used to check convergence of runs and diagnose MCMC chains. Trees were summarized in TreeAnnotator v.1.6.1 (Drummond & Rambaut, 2007) with the 25% first generations discarded as burn-in. Maximum likelihood (ML) analyses for individual nuclear and chloroplast datasets were conducted using RAxML-HPC2 v.7.3.2 on the Cipres Science Gateway (Miller & al., 2010) under the GTRCAT model with 1000 bootstrap replicates. Bootstrap proportions on the ML tree were summarized using Sumtrees from the Python Dendropy library (Sukumaran & Holder, 2010). Maximum parsimony (MP) analyses were performed on both datasets after excluding outgroup taxa, using PAUP* v.4.0b10 (Swofford, 2003), which was also used to calculate consistency and retention indices, and tree length. Parameters for the parsimony heuristic searches included 100 random addition replicates, tree bisection-reconnection branch swapping and multrees off. Trees were visualized using FigTree v.1.3.1 (Rambaut, 2009). Species tree inference and dating. — The *BEAST method as implemented in BEAST v.1.7.3 (Drummond & al., 2006; Drummond & Rambaut, 2007) was used to estimate the species tree for 52 species (covering those species common to both plastid and nuclear datasets except for Gypsophila cerastioides D.Don). The data matrix was composed of two unlinked partitions: (1) ITS sequences, (2) rps16 sequences, using the GTR + G as substitution model for both partitions, and an uncorrelated relaxed lognormal clock. One MCMC chain was run for 20 million generations, with tree sampling every 1000 steps. Since there is no known fossil record for the study group, we used published ITS and chloroplast substitution rates. The prior probability of the clock rate was set to a truncated normal distribution with a mean of 2.15 × 10−3, ranging from 0.38 × 10−3 to 7.83 × 10−3 substitutions per site per Ma for ITS and a truncated normal distribution with a mean of 0.67 × 10−3, ranging from 0.5 × 10−4 to 2.6 × 10−3 substitutions per site per Ma for rps16. ITS substitution rates were set according to previous estimates for perennial woody plants (Kay & al., 2006). The rps16 rates could have been independently estimated, but in order to speed up convergence (particularly by excluding highly suboptimal starting rates) we instead provided rate priors where the lower and upper rates of ITS were divided by three (substitution rates in the chloroplast genome are typically slower by similar magnitudes compared to rates in the nuclear genome in plants; Wolfe & al., 1987; Gaut & al., 1996; Yue & al., 2010). The rps16 prior standard deviation was set to 0.7 × 10−3 that fits in the truncated normal distribution. The value of standard deviation is larger than the mean and unlikely to bias its posterior estimate. Then the estimated rates were checked and modified to the values that allowed the rps16 posteriors to be able to fluctuate within these priors. The ploidy level for rps16 was set to haploid. Tracer was used to check the convergence and mixing of MCMC chains and the effective sample sizes for all parameters to be above 200. Trees were summarized in TreeAnnotator with the 10% first generations discarded as burn-in, and visualized using FigTree. RESULTS Species tree inference and dating. — Two major clades are retrieved within Acanthophyllum in a wide sense (see below) by the *BEAST species tree analysis (Fig. 3). These clades are referred to as Clade I and Clade II hereafter. Two clades are recognized within Clade I (PP 0.94): (1) clade A (PP 0.91), including A. sect. Pseudacanthophyllum and sect. Scapiflora, with Diaphanoptera ekbergii Hedge & Wendelbo as sister to these sections; (2) clade B (PP 0.93), including A. sect. Macrodonta, sect. Macrostegia, sect. Pleiosperma and sect. Oligosperma as well as Ochotonophila allochrusoides Gilli, Scleranthopsis aphanantha (Rech.f.) Rech.f., Diaphanoptera stenocalycina Rech.f. & Schiman-Czeika and D. lindbergii Hedge & Wendelbo. A single species of A. sect. Acanthophyllum (A. pleiostegium Schiman-Czeika) also nests within clade B. Clade II (PP 0.99) largely corresponds to A. sect. Acanthophyllum. Allochrusa bungei Boiss. and All. versicolor form a subclade within Clade II with strong support (PP 0.99). Acanthophyllum sordidum of A. sect. Pleiosperma and A. caespitosum Boiss. of A. sect. Oligosperma also resolve within Clade II. This analysis estimates the age of the Acanthophyllum clade (crown node) to be 11.1 Ma (3.8–32.4 Ma, 95% HPD interval). Clades I and II are estimated to be 8.2 Ma (3.2–22.2 Ma, 95% HPD interval) and 3.9 Ma (1.7–11.1 Ma, 95% HPD interval), respectively. Phylogenetic analyses. — BI, ML and MP analyses of individual nuclear and plastid markers yielded mostly congruent trees within each marker, with no strongly supported differences. Therefore, the results of BI are presented and discussed in detail, whereas those of ML and MP analyses including tree length, invariable characters, number of informative characters, number of indels, consistency and retention indices, and likelihood bootstraps are summarized in Table 2 and Figs. 4–5. Table . Models chosen by ModelTest, sequence and parsimony statistics. ITS rps16 Number of sequences (number of ingroup taxa) 56 53 BIC model choice GTR + G GTR + G Sequence length [bp] 674 839 Number of coded indels 63 40 Invariable characters [bp] 446 742 Variable sites [bp] 228 97 Variable sites [%] 33.8 11.6 Parsimony-informative characters [bp] 140 43 Parsimony-informative characters [%] 20.7 5.1 Consistency index, excl. uninformative characters 0.57 0.78 Retention index 0.8 0.94 Draft version for proof purposes only. 7 TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum Two major well-supported clades are recognized within Acanthophyllum in both the ITS (Fig. 4) and the rps16 (Fig. 5) trees. These clades are congruent with the Clades I and II obtained by *BEAST species tree analysis. The genera Ochotonophila, Scleranthopsis, Diaphanoptera (pro majora parte) and one species of Gypsophila are nested within Acanthophyllum. However, Allochrusa is sister to Acanthophyllum in the ITS phylogeny (PP 0.86; Fig. 4), whereas it nests within Acanthophyllum Clade II in the rps16 phylogeny (PP 1.0; Fig. 5). Clade I is composed of A. sect. Oligosperma, sect. Pleiosperma, sect. Scapiflora, sect. Pseudacanthophyllum, sect. Macrostegia and sect. Macrodonta. Clade I also contains representatives of the genera Diaphanoptera, Ochotonophila, Scleranthopsis and Gypsophila. Moreover, monophyly of A. sect. Oligosperma, Silene 1 Vaccaria hispanica 1 99 1 98 Gypsophila 1 100 Diaphanoptera afghanica Bolbosaponaria bucharica Gypsophila Dianthus 1 100 Velezia rigida 1 100 1 Petrorhagia Gypsophila muralis Gypsophila cerastioides Diaphanoptera ekbergii 1 86 Ochotonophila allochrusoides 1 1 1 1 99 1 1 68 0.86 Clade I 1 1 100 Acanthophyllum s.l. Sect. Acanthophyllum A. pleiostegium A. longicalyx Sect. Macrodonta A. schugnanicum 1 1 91 A. crassinodum Sect. Pleiosperma 0.66 A. glandulosum 72 A. spinosum Diaphanoptera lindbergii 1 91 Diaphanoptera stenocalycina A. pachycephalum A. bracteatum A. leucostegium Sect. Macrostegia 1 A. kandaharicum 100 A. gracile 1 A. stocksianum 89 A. subglabrum A. andersenii A. squarrosum A. heterophyllum A. borscziwuu A. pachystegium 0.94 A. lilacinum A. heratense A. speciosum Sect. Oligosperma A. diezianum 1 A. yasamin-nassehiae 98 A. kabulicum 1 A. laxiusculum 53 A. ejtehadii A. korshinskyi A. adenophorum A. andarabicum Sect. Paniculata A. paniculatum Scleranthopsis aphanantha A. xanthoporphyranthum 1 A. macrodon Sect. Macrodonta 100 A. grandiflorum A. scapiflorum 1 Sect. Scapiflora 87 A. anisocladum A. laxiflorum A. stewartii A. raphiophyllum A. oppositiflorum A. acerosum A. crassifolium A. verticillatum A. mucronatum A. caespitosum A. kurdicum A. microcephalum A. sordidum Allochrusa versicolor Allochrusa bungei 71 1 100 Clade II 1 84 1 89 1 100 1 Psammosilene tunicoides 1 100 Sect. Pseudacanthophyllum Sect. Acanthophyllum Sect. Oligosperma Sect. Acanthophyllum Sect. Pleiosperma Saponaria Fig. 4. Majority-rule consensus tree inferred from Bayesian analysis of ITS data. Posterior probability values are indicated above branches, ML bootstraps values below. Values below 50% are not shown. 8 Draft version for proof purposes only. TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum sect. Macrodonta and sect. Macrostegia is not obtained in the rps16 tree. Clade II includes all examined species of A. sect. Acanthophyllum (except for A. pleiostegium) as well as A. sordidum (sect. Pleiosperma) and A. caespitosum (sect. Oligosperma). The Allochrusa species are included in this clade in the rps16 tree. The resolution within Clade II is low in both datasets. DISCUSSION Phylogenetic position of Acanthophyllum within tribe Caryophylleae. — Acanthophyllum has gained little attention in previous phylogenetic analyses of tribe Caryophylleae. The most comprehensive molecular study included only two species (Fior & al., 2006; Harbaugh & al., 2010; Valente & al., Silene 1 97 1 100 Dianthus Petrorhagia saxifraga Gypsophila cerastioides Ochotonophila allochrusoides A. pleiostegium Sect. Acanthophyllum A. spinosum A. schugnanicum Sect. Pleiosperma A. crassinodum A. glandulosum 0.85 0.94 1 79 1 100 0.98 64 1 93 1 100 1 93 1 95 Diaphanoptera lindbergii Diaphanoptera stenocalycina A. longicalxy Sect. Macrodonta A. gracile A. kandaharicum A. pachycephalum A. bracteatum Sect. Macrostegia A.leucostegium A. heratense A. squarrosum A. kabulicum A. pachystegium A. heterophyllum A. subglabrum A. korshinskyi A. yasamin-nassehiae A. laxiusculum 1 79 A. borsczowii A. ejtehadii 1 90 A. speciosum A. andersenii A. adenophorum A. stocksianum A. diezianum A. andarabicum Scleranthopsis aphanantha A. macrodon 0.97 A. grandiflorum 66 A. xanthoporphyranthum 1 100 Acanthophyllum s.l. 1 100 Clade I Sect. Oligosperma Sect. Macrodonta A. raphiophyllum Sect. Pseudacanthophyllum 0.99 A. stewartii 60 0.9 A. anisocladum 63 1 A. laxiflorum 87 0.68 54 A. scapiflorum Sect. Scapiflora Sect. Pseudacanthophyllum Sect. Scapiflora Diaphanoptera ekbergii Fig. 5. Majority-rule consensus tree inferred from Bayesian analysis of rps16 data. Posterior probability values are indicated above branches, ML bootstraps values below. w. Values V below 50% are not shown. A. oppositiflorum A. crassifolium A. kurdicum A. acerosum 1 86 Sect. Acanthophyllum 0.94 1 100 Clade II 1 93 A. verticillatum A. mucronatum A. microcephalum 0.96 Allochrusa versicolor 72 1 95 Allochrusa bungei A. sordidum A. caespitosum Sect. Pleiosperma Sect. Oligosperma Gypsophila paniculata 1 100 Draft version for proof purposes only. Saponaria 9 TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum 2010; Greenberg & Donoghue, 2011) and the small genera allied to Acanthophyllum, i.e., Scleranthopsis, Ochotonophila and Diaphanoptera, were ignored. In our trees these small genera are embedded in Acanthophyllum s.l. (Figs. 4–5). Cushionforming subshrubby to somewhat woody habit is a character that distinguishes the Acanthophyllum s.l. clade (Acanthophyllum, Scleranthopsis, Allochrusa, Ochotonophila, Diaphanoptera p.p.) from the rest of tribe Caryophylleae. We conclude that the morphological characters used for delimiting Acanthophyllum s.str., Allochrusa, Scleranthopsis, Ochotonophila and Diaphanoptera are either not congruent with the phylogenetic patterns or need to be reassessed. Therefore, we suggest a new classification for this group of taxa based on more reliable characters and the evidence from molecular phylogenetics (see below under Circumscription of Acanthophyllum). The analyses of nuclear and plastid loci show that the Acanthophyllum s.l. clade (including Allochrusa, Ochotonophila, Diaphanoptera p.p. and Scleranthopsis), (PP 1.0; Figs. 4–5) is sister to a clade holding Dianthus and Petrorhagia (and Velezia only in ITS tree) (PP 1.0; Figs. 4–5). This is in agreement with previous studies (Fior & al., 2006; Harbaugh & al., 2010; Valente & al., 2010; Greenberg & Donoghue, 2011). Gypsophila cerastioides nests within Acanthophyllum s.l. The placement of G. muralis L. as sister to the DianthusPetrorhagia clade in the present ITS tree is consistent with the results of Greenberg & Donoghue (2011). Circumscription of Acanthophyllum. — This study reveals that Acanthophyllum as currently circumscribed (SchimanCzeika, 1988; Bittrich, 1993; Takhtajan, 2009) is not monophyletic. The circumscription of Acanthophyllum is discussed in relation to its closely related genera residing within Acanthophyllum s.l. clade. Scleranthopsis. — The monospecific Scleranthopsis is distributed in SW, E and central Afghanistan. The position of this genus has been uncertain and was discussed by Rechinger (1957, 1967). It was first described as Acanthophyllum aphananthum Rech.f. and classified in the monotypic sect. Aphanantha Rech.f. (Rechinger, 1957). It was later raised to generic level, Scleranthopsis, based on the following characters: non-clawed petals, short petals enclosed within the calyx, and unequal calyx teeth (Rechinger, 1967). Our study supports returning Scleranthopsis to Acanthophyllum and re-establishing the sect. Aphanantha. Allochrusa. — Allochrusa has approximately seven species distributed in Armenia, Turkey, NW Iran, NE Afghanistan, Tajikistan and Turkmenistan. It was first proposed as distinct from Acanthophyllum (Boissier, 1867) based on non-spiny leaves and enclosed stamens but was included in Acanthophyllum by Golenkin (1893) and Schischkin (1936). Golenkin made Allochrusa a section, whereas Schischkin (1936) treated it as a subgenus. He divided subg. Allochrusa into two sections, sect. Versicoloria Schischk. and sect. Paniculata Golenk. (Table 1). The species of the latter section have a short calyx and petals, as well as loose inflorescences. Barkoudah (1962) followed Schischkin. However, Allochrusa has later been accepted as a separate genus (Yukhananov, 1974; Schiman-Czeika, 1987; Bittrich, 1993; Takhtajan, 2009). 10 Allochrusa versicolor and All. bungei form a strongly supported clade within Acanthophyllum s.l. (Figs. 3–5). The Bayesian analysis of the plastid marker as well as the results of the *BEAST species tree analyses are congruent in resolving this clade within Clade II. However, Bayesian analysis of ITS suggests that the Allochrusa clade is sister group to Acanthophyllum (PP 1.0, but note PP only 0.86 for monophyly of the latter). We cannot resolve the reason for this incongruence at the moment. Emarginate petals is a possible synapomorphy for Clade II, including Allochrusa bungei and All. versicolor (= Allochrusa s.str.). Our ITS dataset also included Acantophyllum paniculatum, which was classified in subg. Allochrusa sect. Paniculata by Schischkin (1936) and in the genus Allochrusa by Schiman-Czeika (1988). Acanthophyllum paniculatum which has entire petals, is placed in Clade I in the ITS tree, together with other species with entire petals. Interestingly, in Clade II, All. versicolor and bungei that are distributed in NW Iran and Armenia come together with Acanthophyllum species with a mainly western distribution pattern, i.e., Syria, Turkey, Iraq and W Iran, whereas the more eastern A. paniculatum (Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, Uzbekistan) is placed in Clade I, which includes taxa with an eastern distribution. Our study supports the inclusion of Allochrusa in the Acanthophyllum s.l. clade. Ochotonophila. — Ochotonophila comprises two species inhabiting central and E Afghanistan (Gilli, 1956; Hedge & Wendelbo, 1963; Schiman-Czeika, 1988). They are lowgrowing multi-stemmed perennials with a woody base. In the original description by Gilli (1956) the genus (with the single species O. allochrusoides) was considered intermediate between Acanthophyllum and Gypsophila. Hedge & Wendelbo (1963) described O. eglandulosa Hedge & Wendelbo, again emphasizing similarities to Acanthophyllum and Gypsophila. Later work by Schiman-Czeika (1988) suggested that Ochotonophila is more closely related to Acanthophyllum, Scleranthopsis and Allochrusa. The tubular calyx and the presence of bracteoles are features shared by Ochotonophila, Allochrusa, Scleranthopsis and Acanthophyllum but not Gypsophila. Ochotonophila differs from Acanthophyllum by its non-spiny leaves, enclosed stamens and, non-appressed bracteoles. Ochotonophila allochrusoides is placed close to A. sect. Pleiosperma, A. pleiostegium (A. sect. Acanthophyllum) and A. longicalyx Hedge & Wendelbo (A. sect. Macrodonta) in our analyses (Figs. 3–5). An important feature shared by Ochotonophila and A. sect. Pleiosperma is the high number of ovules (10–21). We suggest reducing Ochotonophila to a section under Acanthopyllum s.l. Diaphanoptera. — Diaphanoptera has six species and is distributed in NE Iran, Turkmenistan and Afghanistan (Schiman-Czeika, 1988). They are perennial multi-stemmed plants with a woody base. The genus was described by Rechinger (1940) with the single species D. khorassanica Rech.f., distinguished from Acanthophyllum by diaphanous wings on the calyx (Rechinger, 1940; Hedge & Wendelbo, 1963). However, the other five currently accepted species of the genus lack diaphanous wings. Lack of bracteoles and a non-tubular calyx also separate Diaphanoptera from Acanthophyllum (Schiman-Czeika, 1988). Diaphanoptera afghanica Draft version for proof purposes only. still in prep.? equal or more ? TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum (included only in the ITS dataset) is sister to Bolbosaponaria, a genus closely related to Gypsophila s.str., whereas Diaphanoptera stenocalycina, D. lindbergii and D. ekbergii, nest inside Acanthophyllum Clade I in all trees (Figs. 3–5). Diaphanoptera lindbergii and D. stenocalycina have 10–12 ovules in each ovary, in accordance with its placement in clade E (PP 0.82; Fig. 3) that contains taxa with more than eight ovules. However, the support for clade E is not strong and is due only to rps16 (PP 0.98; Fig. 5). The placement of D. ekbergii as sister to A. sects. Pseudacanthophyllum and Scapiflora is likewise poorly supported in the species tree (PP 0.91; Fig. 3) and the rps16 tree (PP 0.68; Fig. 5) and not at all in the ITS tree (Fig. 4), where it is resolved as sister to Gypsophila cerastioides. We were not successful in generating high-quality sequences for D. khorassanica, the type of the genus. Therefore, the delimitation of Diaphanoptera from Acanthophyllum, Gypsophila and Ochotonophila cannot be ascertained, but it is clear that Diaphanoptera as currently circumscribed contains unrelated elements. Gypsophila. — Gypsophila has about 150 species and is distributed in temperate Asia and Europe, Egypt, Australia, North America and China (Amini & al., 2011). The delimitation of Acanthophyllum from Gypsophila has long been controversial. General habit, calyx shape, petal shape in terms of developing a claw, leaf texture, duration of bracts and bracteoles, calyx shape and mode of capsule dehiscence (Barkoudah, 1962; Gilli, 1964; Schiman-Czeika, 1987) are the most commonly used characters for discriminating the two genera. Emphasis on different characters has led to several transfers of species. Acanthophyllum sect. Pseudacanthophyllum is one of the most controversial groups transferred from Gypsophila to Acanthophyllum (Oliver, 1859; Bentham & Hooker, 1862; Boissier, 1867; Williams, 1889; Barkoudah, 1962; Schiman-Czeika, 1988). The examined species of this section nest within Acanthophyllum in our analyses (Figs. 3–5). In both analyses of ITS and rps16 (Figs. 4–5) G. cerastioides falls within Acanthophyullum s.l. Gypsophila muralis resolves as sister to the Dianthus-Petrorhagia clade in the ITS tree (consistent with Greenberg & Donoghue, 2011), and the remaining species of Gypsophila are found in two additional clades. This shows that Gypsophila is non-monophyletic. A broad molecular study (Zarre & al., in prep.) shows that apart from G. cerastioides no other species of Gypsophila appear to belong to the Acanthophyllum clade. Therefore, the poor sampling of Gypsophila is not likely to affect our taxonomic conclusions. Subgeneric classification of Acanthophyllum s.str. — The sections in individual ITS and rps16 trees do not fully match. Acanthophyllum sect. Pseudacanthophyllum and sect. Scapiflora are intermingled in the plastid tree, whereas they are sister sections in the nuclear tree. Acanthophyllum sect. Macrostegia appears monophyletic only in the ITS tree. Acanthophyllum sect. Pleiosperma is monophyletic only in rps16 tree, except for one species. The large section Oligosperma forms a clade in both nuclear and plastid trees, except for one and two species, respectively. Acanthophyllum sect. Macrodonta is monophyletic in both gene trees if A. longicalyx is excluded. Most species of A. sect. Acanthophyllum form a clade in the ITS tree; only one species is excluded from this clade while one species of A. sect. Oligosperma is nested within it. In rps16, A. sect. Acanthophyllum is also monophyletic except for two species. The differences between the rps16 and ITS trees can be explained by different numbers of informative characters in plastid and nuclear datasets. It can also be attributed to stochastic effects, hybridizations, or incomplete lineage sorting. The implementation of the multispecies coalescent model in *BEAST takes the latter into account (Heled & Drummond, 2010), and by using the information from all datasets simultaneously, the stochastic effects can be expected to decrease. Therefore, we will from now on discuss the phylogenetic status of the taxa in relation to the *BEAST species tree (Fig. 3). Emarginate petals appear to be a synapomorphy for Clade II, whereas Clade I holds taxa with mainly entire petals (some other petal types, e.g., dentate-sinuate in A. pleiostegium and deeply bifid in O. allochrusoides are rarely seen in Clade I). The common petal shapes in outgroups are entire, bifid, emarginate, fimbriate and sinuate. Some common petal shapes in Caryophylloideae are shown in Fig. 2C–H. Clade I. — Acanthophyllum sect. Scapiflora, sect. Pseudacanthophyllum, sect. Macrodonta, sect. Pleiosperma, sect. Macrostegia and sect. Oligosperma together form Clade I (PP 0.94; Fig. 3). Acanthophyllum sects. Scapiflora and Macrodonta are not resolved as monophyletic. A single species of A. sect. Acanthophyllum, A. pleiostegium, also nests within Clade I. Acanthophyllum sect. Scapiflora, sect. Pseudacanthophyllum, sect. Macrodonta, sect. Pleiosperma, sect. Macrostegia and sect. Oligosperma correspond to the subclades A, D, G, H and K, respectively. Clade A and clade B, the latter including clades D, E and H, have posterior probabilities from 0.82 to 0.99, and are in the following discussed in relation to putative synapomorphies. The other clades have low support and are not discussed. Clade A. – Clade A (PP 0.91; Fig. 3) includes A. sect. Pseudacanthophyllum and sect. Scapiflora and D. ekbergii. A putative synapomorphy for Acanthophyllum sections in this clade is poorly branched inflorescence (Fig. 2B) of one (rarely up to three in A. laxiflorum Boiss.) condensed unit(s) confined to only one terminal node of the stem and composed of multiflowered partial cymes (with more than seven flowers). These sections are almost confined to Afghanistan. Acanthophyllum anisocladum Schiman-Czeika and A. scapiflorum (Akhtar) Schiman-Czeika of sect. Scapiflora and A. laxiflorum, A. raphiophyllum (Rech.f.) Barkoudah and A. stewartii (Thoms. ex Edgew & Hook.f.) of A. sect. Pseudacanthophyllum constitute clade A. The two sections do not form separate clades in the *BEAST tree (Fig. 3), despite morphological differences that make them easily recognizable. Species of A. sect. Scapiflora have elongated, leafless peduncles whereas species of A. sect. Pseudacanthophyllum have a condensed habit with the inflorescence set close to the leaves. The two sections are, however, resolved as monophyletic in the ITS tree (Fig. 4). It will require sampling of other nuclear loci to resolve whether the incongruence should be attributed to stochasticity or needs a biological explanation. Draft version for proof purposes only. 11 TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum Clade B. – Clade B (PP 0.93; Fig. 3) includes A. sect. Macrodonta (clade D), sect. Pleiosperma (clade G), sect. Macrostegia (clade H) and sect. Oligosperma (clade K), a single species of A. sect. Acanthophyllum, as well as Diaphanoptera, Ochotonophila and Scleranthopsis. No morphological synapomorphy for this clade could be found. Clade D. – Within clade B, subclade D is strongly supported (PP 0.99; Fig. 3) and holds all examined species of A. sect. Macrodonta except A. longicalyx. Clade D is characterized by solitary flowers and four ovules. Acanthophyllum longicalyx has one to three terminal flowers and eight ovules and resolves within clade E. This species also possesses the longest calyx and the shortest calyx teeth among the taxa assigned to sect. Macrodonta (Hedge & Wendelbo 1963; Schiman-Czeika 1988). Clade E. – Clade E (PP 0.82; Fig. 3) holds Ochotonophila allochrusoides, A. sect. Pleiosperma (except A. sordidum), A. pleiostegium (sect. Acanthophyllum), A. longicalyx (sect. Macrodonta) and Diaphanoptera lindbergii and D. stenocalycina. Apart from A. pleiostegium with 4–6 ovules in each ovary, the rest of the taxa in this clade produce eight or more ovules (up to 21 ovules in O. allochrusoides). Clade H. – Clade H (PP 0.92; Fig. 3) corresponds to A. sect. Macrostegia. Broadly ovate to lanceolate membranous bracts and bracteoles constitute a synapomorphy for this clade. It is widespread in Iran, Iraq, Turkmenistan, Tajikistan, Kyrgyzstan, Uzbekistan, Kazakhstan, Afghanistan and Pakistan. Acanthophyllum sect. Oligosperma. — Clade K (PP 0.66; Fig. 3) has poor support but is more or less identical to the large and taxonomically complicated A. sect. Oligosperma (except for A. caespitosum which is resolved as member of Clade II in all analyses). However, without A. andarabicum Podl. ex Schiman-Czeika, the rest of the clade (clade L) has strong support in all trees. Only the ITS phylogeny (Fig. 4) has strong support (PP 1.0) for clade K. Schischkin (1936) divided A. sect. Oligosperma into five series. Series Elatiora Schischk. (including A. elatius Bunge and A. borsczowii Litv.) was later elevated to sectional rank (Zakirov & Musaeva, 1981). However, Schiman-Czeika (1988) included A. elatius and A. borsczowii within sect. Oligosperma. We can confirm the latter opinion, as A. borsczowii falls within the core group of A. sect. Oligosperma in the species (PP 0.96; Fig. 3), ITS (PP 1.0; Fig. 4) and rps16 (PP 1.0; Fig. 5) trees. Clade II. — Clade II (PP 0.99; Fig. 3) is characterized by emarginate petals. A large part of this clade corresponds to A. sect. Acanthophyllum which is recognised by branched elongated inflorescences of white flowers (bright pink flowers are rarely seen in some populations of A. acerosum Sosn.). However, the section is not resolved as monophyletic, with A. oppositiflorum Aytaç falling outside this group including A. acerosum, A. verticillatum Hand.-Mazz., A. mucronatum C.A.Mey., A. microcephalum Boiss., A. kurdicum Boiss. & Hausskn. ex Boiss. and A. crassifolium Boiss. Acantophyllum oppositiflorum possesses elongated inflorescences with only two opposite flowers at each node, whereas the other species of A. sect. Acantophyllum have inflorescences with multiflowered verticillasters. 12 Acanthophyllum sect. Acanthophyllum shows a mainly western distribution and inhabits Syria, Iraq, Turkey, Armenia, Azerbaijan, Iran and Turkmenistan. The Afghan species A. pleiostegium was assigned to sect. Acanthophyllum by Schiman-Czeika (1988), but belongs in clade E (see above). Acanthophyllum caespitosum (sect. Oligosperma) and A. sordidum (sect. Pleiosperma) also nest in Clade II. They are the only species of their sections that have emarginate petals. Acanthophyllum caespitosum is an odd member of A. sect. Oligosperma also in its procumbent densely branched habit, extremely reduced stems and imbricate leaves. The other taxa of A. sect. Oligosperma have tall stems and non-imbricate leaves. Further, A. caespitosum is the only known species of Acantophyllum with a haploid base chromosome number of x = 14 instead of x = 15 (Ghaffari, 2004). Ghaffari discussed these morphological and base chromosome number differences and suggested that A. caespitosum should form a section of its own. Exclusion of A. caespitosum from A. sect. Oligosperma is corroborated by our analyses and it is placed together with the species of A. sect. Acanthophyllum (Figs. 3–5). However, A. caespitosum differs from A. sect. Acanthophyllum in 4-ovulate ovaries, reduced stems and reduced inflorescences. Furthermore, all examined representatives of sect. Acanthophyllum are tetraploids (Ghaffari, 2004), whereas A. caespitosum is diploid. Taking these complexities into account, it is not possible to decisively judge the taxonomic placement of A. caespitosum in A. sect. Acanthophyllum. Acanthophyllum sordidum has been assigned to A. sect. Pleiosperma (Boissier, 1867; Schischkin, 1936; Golenkin, 1893; Schiman-Czeika, 1988; Basiri-Esfahani & al., 2011), but our analysis shows that it is remotely related to the rest of the section, which resides within Clade I (Figs. 3–5). The non-clawed linear emarginate petals of A. sordidum are not seen in any other taxa of A. sect. Pleiosperma. The ploidy level reported for A. sordidum is tetraploid (Ghaffari 2004), the same as for A. sect. Acanthophyllum, whereas the rest of the examined species of A. sect. Pleiosperma are hexaploid (Ghaffari, 2004). However, non-elongated inflorescences and non-clawed petals of A. sordidum separate it from A. sect. Acanthophyllum. We conclude that A. sordidum should be excluded from A. sect. Pleiosperma. Some conclusions with regard to the circumscription and subgeneric classification of Acanthophyllum s.l. and the geographical distribution of its subgeneric taxa can be derived from our study: (1) The majority of the morphological characters used in subgeneric classifications of Acanthophyllum show low levels of homoplasy, and are thus useful in interpreting phylogenetic patterns within the genus. We conclude that the value of petal morphology has been underestimated. Emarginate petals, for instance, characterizes one of the two major clades. However, the state of “clawed petal” which is frequent among the studied group, is lost in several places in the trees to form the “non-clawed petals” in A. aphananthum (see above under Scleranthopsis), O. allochrusoides (see above under Ochotonophila), A. pachycephalum Schiman-Czeika, A. pleiostegium, A. sordidum and A. oppositiflorum; (2) Increase in ploidy level is accompanied by increase in the number Draft version for proof purposes only. of ovules (see Fig. 3). Acanthophyllum sect. Macrostegia and sect. Oligosperma, which have diploid chromosome numbers, always produce four ovules per ovary, whereas tetraploid A. sect. Acanthophyllum produces 4–8 ovules, and hexaploid A. sect. Pleiosperma produces 8–16 ovules; (3) The number of Acanthophyllum species is estimated to amount to about 80–90. Acanthophyllum sect. Acanthophyllum constitutes the westernmost limit of the genus. The diversity center of this section is in NW Iran and neighboring areas in Turkey and Iraq. The examined species of this section reside within Clade II in our trees. The easternmost limit is set by A. pungens Boiss. (sect. Oligosperma), which inhabits Central Asia to W China. Acanthophyllum sect. Oligosperma is otherwise most diverse in NE Iran and adjacent areas of Afghanistan and Turkmenistan, corresponding to the diversity center of the genus as a whole. Acanthophyllum sect. Oligosperma falls into Clade I in our analyses. Generally, most Acanthophyllum species are in Clade I, found in NE Iran or further east. Afghanistan harbors the highest number of Acanthophyllum sections, including three endemic sections (A. sect. Scapiflora, sect. Ochotonophila and sect. Aphanantha; see below). Divergence time of Acanthophyllum. — Valente & al. (2010) obtained an approximate age of 5 Ma (3–9 Ma, 95% HPD interval) for the Acanthophyllum-Allochrusa clade. BEAST analysis of matK data by Frajman & al. (2009) estimated the age to 2.88 Ma. These studies included only one species each of Acanthophyllum and Allochrusa: A. sordidum and All. versicolor. In our *BEAST species tree (Fig. 3) Allochrusa nests within Acanthophyllum and A. sordidum is resolved as sister to All. versicolor and All. bungei). Although weakly supported, the A. sordidum–Allochrusa clade has an approximate age of 2.48 Ma (0.1–7.7 Ma, 95% HPD interval) in our analysis. The species tree analysis suggests a somewhat younger age for the A. sordidum–Allochrusa clade, compared to the age estimated by Valente & al. (2010) using gene trees, although the credibility intervals of both estimates largely overlap. The age of 2.48 Ma estimated by the present species tree analysis for the AllochrusaAcanthophyllum clade is very close to the age (2.88 Ma) gained by the BEAST analysis of matK (Frajman & al., 2009). The *BEAST analysis estimated the age of the Acanthophyllum s.l. crown clade as 11.1 Ma (3.8–32.4 Ma, 95% HPD interval), a time span covering the Oligocene, Miocene and Pliocene (Fig. 3). There are some fossils assigned to Caryophyllaceae (Jordan & Macphail, 2003; Huang & al., 2012), but we did not use them as the taxa were phylogenetically far from our focal group, which would either introduce long branches to our tree or require an extremely increased taxon sampling. Therefore, we used only clock rate calibration analysis. TAXONOMIC IMPLICATIONS Acanthophyllum C.A.Mey., Verz. Pfl. Casp. Meer.: 210. 1831 – Type: A. mucronatum C.A.Mey. = Ochotonophila Gilli in Feddes Repert. Spec. Nov. Regni Veg. 59: 169. 1956, syn. nov. – Type: O. allochrusoides Gilli. = Allochrusa Bunge in Boissier, Fl. Orient. 1: 559. 1867 ≡ Acanthophyllum subg. Allochrusa (Bunge) Schischk., Fl. U.S.S.R. 6: 800. 1936 – Type: All. versicolor (Fisch. & C.A.Mey.) Boiss. New names and combinations Acanthophyllum sect. Ochotonophila (Gilli) Pirani, comb. & stat. nov. ≡ Ochotonophila Gilli in Feddes Repert. Spec. Nov. Regni Veg. 59: 169. 1956 – Type: A. allochrusoides (Gilli) Pirani Perennial low-growing plants, multi-stemmed. Leaves non-spiny. Bracteoles distant from flowers. Ovary stipitate, ovules 10–18; capsule many-seeded. at least this type has been designated - provide according information Pirani & al. • Molecular phylogeny of Acanthophyllum Acanthophyllum allochrusoides (Gilli) Pirani, comb. nov. ≡ Ochotonophila allochrusoides Gilli in Feddes Repert. Spec. Nov. Regni Veg. 59: 169. 1956 – Holotype: AFGHANISTAN. In montibus calc. NE Bamian, 3000 m, date? Gilli 1269 (W [photo!]). Acanthophyllum eglandulosum (Hedge & Wendelbo) Pirani, comb. nov. ≡ Ochotonophila eglandulosa Hedge & Wendelbo in Aarbok Univ. Bergen, Mat.-Naturvitensk. Ser. 28: 21. 1963 – Holotype: AFGHANISTAN. Kabul, in decl. orientalibus jugi Shibar, 2750 m, date? Hedge & Wendelbo W-4239 (E n.v.; isotypes: BG n.v., W [photo!]). Section to be reinstated Acanthophyllum sect. Aphanantha Rech.f. in Oesterr. Bot. Z. 104: 174. 1957 ≡ Scleranthopsis Rech.f. in Ann. Naturhist. Mus. Wien 70: 37. 1967 ≡ Scleranthopsis Rech.f., Ann. Naturhist. Mus. Wien 70: 37. 1967 – Type: Acathophyllum aphananthum Rech.f. in Oesterr. Bot. Z. 104: 174. 1957. ACKNOWLEDGEMENTS We thank Magnus Lidén and an anonymous reviewer as well as the Associate Editor Sigrid Liede-Schumann and the Managing Editor Michael Pirie for valuable suggestions on the manuscript. We are grateful to Hamid Moazzeni, Anna Petri, Vivian Alden, Claes Persson, Zeynep Aydin and Zekí Aytaç for their kind help during different steps of this work. We also wish to thank curators at herbaria B, BM, FUMH, G, GB, JE, K, KEW, LD, M, MSB, TARI, TMRC, TUH and WU for the loan and permission to study plant material used in this study. The herbarium of RBG Edinburgh (E) is thanked for providing photos of their Acanthophyllum specimens. All analyses (except ML) were performed on the bioinformatics computer cluster Albiorix at the Department of Biological and Environmental Sciences, University of Gothenburg. This paper is a part of Ph.D. thesis conducted by AP at the University of Tehran. Financial support provided by the University of Tehran is appreciated. BO and BEP are supported by grants from the Swedish Research Council, the Royal Swedish Academy of Sciences, Lars Hiertas Minnesfond, The Royal Physiographic Society in Lund, Helge Ax:son Johnsons fond and Lundgrenska fonden. Draft version for proof purposes only. 13 why twice? TAXON — 26 May 2014: 16 pp. 14 Draft version for proof purposes only. not cited? Aghel, N., Moghimipour, E. & Raies Dana, A. 2007. Formulation of a herbal shampoo using total saponins of Acanthophyllum squarrosum. Iran. J. Pharm. Res. 6(3): 167–172. Amini, E., Zarre, S. & Assadi, M. 2011. Seed micro-morphology and its systematic significance in Gypsophila (Caryophyllaceae) and allied genera. Nordic J. Bot. 29: 1–10. http://dx.doi.org/10.1111/j.1756-1051.2011.01208.x Aytaç, Z. 2001. A new species of Acanthophyllum (Caryophyllaceae) from central Anatolia, Turkey. Nordic J. Bot. 21: 263–266. http://dx.doi.org/10.1111/j.1756-1051.2001.tb00766.x Barkoudah, Y.I. 1962. A revision of Gypsophila, Bolanthus, Ankyropetalum and Phryna. Wentia 9: 1–203. Basiri-Esfahani, Sh., Bidi, B., Assadi, M. & Rahimi-Nejad, M.R. 2011. A taxonomic study of Acanthophyllum C.A.Mey. (Caryophyllaceae) in Iran. Iran. J. Bot. 17: 24–39. Bentham, G. & Hooker, W.J. 1862. Genera plantarum, vol. 1. London: Reeve. http://dx.doi.org/10.5962/bhl.title.747 Bittrich, V. 1993. Caryophyllaceae. Pp. 206–236 in: Kubitzki, J. (ed.). The families and genera of vascular plants, vol. 2. Berlin: Springer. Boissier, E. 1867. Flora orientalis, vol. 1. Basel: Georg. http://dx.doi.org/10.5962/bhl.title.20323 Drummond, A.J. & Rambaut, A. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. B. M. C. Evol. Biol. 7: 214. http://dx.doi.org/10.1186/1471-2148-7-214 Drummond, A.J., Ho, S.Y.W., Phillips, M.J. & Rambaut, A. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biol. 4: e88. http://dx.doi.org/10.1371/journal.pbio.0040088 Eggens, F., Popp, M., Nepokroeff, M., Wagner, W. L. & Oxelman, B. 2007. The origin and number of introductions of the Hawaiian endemic Silene species (Caryophyllaceae). Amer. J. Bot. 94: 210–218. http://dx.doi.org/10.3732/ajb.94.2.210 Fior, S., Karis, P.O., Casazza, G., Minuto, L. & Sala, F. 2006. Molecular phylogeny of the Caryophyllaceae (Caryophyllales) inferred from chloroplast matK and nuclear rDNA ITS sequences. Amer. J. Bot. 93: 399–411. http://dx.doi.org/10.3732/ajb.93.3.399 Frajman, B., Eggens, F. & Oxelman, B. 2009. Hybrid origins and homoploid reticulate evolution within Heliosperma (Sileneae, Caryophyllaceae): A multigene phylogenetic approach with relative dating. Syst. Biol. 58: 328–345. http://dx.doi.org/10.1093/sysbio/syp030 Gaidi, Gh., Miyamoto, T., Rustaiyan, A., Laurens, V. & LacailleDubois, M.A. 2000. Two new biologically active triterpene saponins from Acanthophyllum squarrosum. J. Nat. Prod. (Lloydia) 63: 1497–1502. http://dx.doi.org/10.1021/np000212+ Gaidi, Gh., Miyamoto, T., Rustaiyan, A. & Lacaille-Dubois, M.A. 2004. Glandulosides A-D, triterpene saponins from Acanthophyllum glandlosum. J. Nat. Prod. (Lloydia) 67: 1114–1118. http://dx.doi.org/10.1021/np040001v Gaut, B.S., Morton, B.R., McCaig, B.C. & Clegg, M.T. 1996. Substitution rate comparisons between grasses and palms: Synonymous rate differences at the nuclear gene Adh parallel rate differences at the plastid gene rbcL. Proc. Natl. Acad. Sci. U.S.A. 93: 10274– 10279. http://dx.doi.org/10.1073/pnas.93.19.10274 Ghaffari, S.M. 1989. Center of diversity and center of origin of the genus Acanthophyllum from a cytotaxonomical point of view. J. Sci. Univ. Tehran 18: 65–74. Ghaffari, S.M. 2002. Biosystematic study of some Acanthophyllum species. Dissertation, University of Tehran, Iran. Ghaffari, S.M. 2004. Cytotaxonomy of some species of Acanthophyllum (Caryophyllaceae) from Iran. Biologia (Bratislava) 59/1: 53–60. Ghazanfar, S.A. & Nasir, Y.J. 1986. Acanthophyllum. Pp. 102–108 in: Nasir, E. & Ali, S.I. (eds.), Flora of Pakistan, vol. 175. Islamabad: Pakistan Agricultural Research Council. Gilli, A. 1956. Neue Caryophyllaceen aus Afghanistan. Feddes Repert. Spec. Nov. Regni Veg. 59: 169–170. Gilli, A. 1964. Die Gattungsumgrenzung von Gypsophila und Acanthophyllum. Oesterr. Bot. Z. 111: 285–290. http://dx.doi.org/10.1007/BF01373771 Golenkin, M. 1893. Revision of the genus Acanthophyllum. Trudy Imp. S.-Peterburgsk. Bot. Sada 13: 77–87. Greenberg, A.K. & Donoghue, M.J. 2011. Molecular systematics and character evolution in Caryophyllaceae. Taxon 60: 1637–1652. Harbaugh, D.T., Nepokroeff, M., Rabeler, R.K., McNeill, J., Zimmer, E.A. & Wagner, W.L. 2010. A new lineage-based tribal classification of the family Caryophyllaceae. Int. J. Pl. Sci. 171: 185–198. http://dx.doi.org/10.1086/648993 Hedge, I. & Wendelbo, P. 1963. Studies in the flora of Afghanistan I. Aarbok Univ. Bergen, Mat.-Naturvitensk. Ser. 18: 21. Heled, J. & Drummond, A.J. 2010. Bayesian inference of species trees from multilocus data. Molec. Biol. Evol. 27: 570–580. http://dx.doi.org/10.1093/molbev/msp274 Huang, Y., Liu, Y., Jacques, F.M.B., Su, T., Xing, Y. & Zhou, Z. 2012. First discovery of Cucubalus (Caryophyllaceae) fossil, and its biogeographical and ecological implications. Rev. Palaeobot. Palynol. 190: 41–47. http://dx.doi.org/10.1016/j.revpalbo.2012.11.011 Huelsenbeck, J.P. & Ronquist, F. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17: 754–755. http://dx.doi.org/10.1093/bioinformatics/17.8.754 Jordan, G.J. & Macphail, M.K. 2003. A Middle-Late Eocene inflorescence of Caryophyllaceae from Tasmania, Australia. Amer. J. Bot. 90: 761–768. http://dx.doi.org/10.3732/ajb.90.5.761 Kay, K.M., Whittall J.B. & Hodges S.A. 2006. A survey of nuclear ribosomal internal transcribed spacer substitution rates across angiosperms: An approximate molecular clock with life history effects. B. M. C. Evol. Biol. 6: 36. http://dx.doi.org/10.1186/1471-2148-6-36 Kool, A., Perrigo, A. & Thulin, M. 2012. Bristly versus juicy: Phylogenetic position and taxonomy of Sphaerocoma (Caryophyllaceae). Taxon 61: 67–75. Kurtto, A. 2001. Caryophyllaceae. Pp. 83–84 in: Jonsell, B. (ed.), Flora Nordica, vol. 2, Chenopodiaceae to Fumariaceae. Stockholm: Bergius Foundation, Royal Academy of Sciences. Laínz M., Garmendia, F.M. 1990. Caryophyllaceae. Pp. 98–464 in: Castroviejo, S., Laínz, M., González, G.L., Montserrat, P., Garmendia, F.M., Paiva, J. & Villar, L. (eds.) Flora iberica: Plantas vasculares de la Península Ibérica e Islas Baleares, vol. 2. Madrid: Real Jardin Botanico, CSIC. Lewis, P.O. 2001. A likelihood approach to estimating phylogeny from discrete morphological character data. Syst. Biol. 50: 913–925. http://dx.doi.org/10.1080/106351501753462876 Mahmoudi-Shamsabad, M., Vaezi, J., Memariani, F. & Joharchi, M.R. 2012. A new species and a new record of Acanthophyllum C.A.Mey. (Caryophyllaceae) from northeast of Iran. Iran. J. Bot. 18: 59–63. Miller, M.A., Pfeiffer, W. & Schwartz, T. 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: Proceedings of the Gateway Computing Environments Workshop (GCE), 2010. http://dx.doi.org/10.1109/GCE.2010.5676129 Müller, K. 2005. SeqState—primer design and sequence statistics for phylogenetic DNA data sets. Appl. Bioinf. 4: 65–69. Oliver, D.J. 1859. Observations on the structure of certain species of the natural order Caryophyllaceae. Trans. Linn. Soc. London 22: 289. http://dx.doi.org/10.1111/j.1096-3642.1856.tb00099.x Ovchinnikov, P.N. 1968. Acanthophyllum. Pp. 615–623 in: Ovchinnikov, P.N. (ed.), Flora Tadzhikskoi SSR, vol. 3. Leningrad: Izdatelstvo Nauka. Oxelman, B. & Lidén, M. 1995. Generic boundaries in the tribe Sileneae (Caryophyllaceae) as inferred from nuclear rDNA sequences. Taxon 44: 525–542. http://dx.doi.org/10.2307/1223498 Oxelman, B., Lidén, M. & Berglund, D. 1997. Chloroplast rpsl6 intron phylogeny of the tribe Sileneae (Caryophyllaceae). Pl. Syst. Evol. 206: 393–410. http://dx.doi.org/10.1007/BF00987959 provide correct title and page nos. give issue nos. only if paginated separately part 1 of vol. 1? LITERATURE CITED provide correct page nos. TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum Pirani & al. • Molecular phylogeny of Acanthophyllum Oxelman, B., Lidén, M., Rabeler, R.K. & Popp, M. 2001. A revised generic classification of the tribe Sileneae (Caryophyllaceae). Nordic J. Bot. 20: 743–748. http://dx.doi.org/10.1111/j.1756-1051.2000.tb00760.x Petri, A. & Oxelman, B. 2011. Phylogenetic relationships within Silene (Caryophyllaceae) section Physolychnis. Taxon 60: 953–968. Pirani, A., Joharchi, M.R. & Memariani, F. 2013. A new species of Acanthophyllum (Caryophyllaceae) from Iran. Phytotaxa 92: 20–24. http://dx.doi.org/10.11646/phytotaxa.92.1.3 Popp, M. & Oxelman, B. 2001. Inferring the history of the polyploidy Silene aegaea (Caryophyllaceae) using plastid and homoeologous nuclear DNA sequences. Molec. Phylogen. Evol. 20: 474–481. http://dx.doi.org/10.1006/mpev.2001.0977 Popp, M., Eggens, F., Erixon, P. & Oxelman, B. 2005. Origin and evolution of a circumpolar polyploid species complex in Silene (Caryophyllaceae) inferred from low copy nuclear RNA polymerase introns, rDNA, and chloroplast DNA. Syst. Bot. 30: 302–313. http://dx.doi.org/10.1600/0363644054223648 Popp, M. & Oxelman, B. 2007. Origin and evolution of North American polyploid Silene (Caryophyllaceae). Amer. J. Bot. 94: 330–349. http://dx.doi.org/10.3732/ajb.94.3.330 Posada, D. 2008. jModelTest: Phylogenetic model averaging. Molec. Biol. Evol. 25: 1253–1256. http://dx.doi.org/10.1093/molbev/msn083 Rambaut, A. 2009. FigTree, version 1.3.1. http://tree.bio.ed.ac.uk/software/figtree/. Rambaut, A. & Drummond, A.J. 2009. Tracer, version 1.5.0. http:// beast.bio.ed.ac.uk/Tracer. Rautenberg, A., Hauthaway, L., Oxelman, B. & Prentice, H.C. 2010. Geographic and phylogenetic patterns in Silene section Melandrium (Caryophyllaceae) as inferred from chloroplast and nuclear DNA sequences. Molec. Phylogen. Evol. 57: 978–991. http://dx.doi.org/10.1016/j.ympev.2010.08.003 Rechinger, K.H. 1940. Plantae novae iranicae I. Repert. Spec. Nov. Regni Veg. 48: 42–43. Rechinger, K.H. 1957. Caryophyllaceae novae Afghanicae. Oesterr. Bot. Z. 104: 173–174. http://dx.doi.org/10.1007/BF01289124 Rechinger, K.H. 1967. Scleranthopsis, eine neue CaryophyllaceenGattung aus Afghanistan. Ann. Naturhist. Mus. Wien. 70: 37–39. Schiman-Czeika, H. 1987. Notes on the capsule dehiscence in Acanthophyllum (Caryophyllaceae) and allied genera. Pl. Syst. Evol. 155: 67–69. http://dx.doi.org/10.1007/BF00936289 Schiman-Czeika, H. 1988. Acanthophyllum. Pp. 253–329 in: Rechinger, K.H. (ed.), Flora Iranica, vol. 163. Graz: Akademische Druck- u. Verlagsanstalt. Schischkin, B.K. 1936. Acanthophyllum. Pp. 781–802, 892–896 in Komarov, V.L. & Shishkin., B.K. (eds.), Flora of the U.S.S.R., vol. 6. Moskva-Leningrad: Izdatel’stvo Akademii Nauk SSSR. Simmons, M.P. & Ochotorena, H. 2000. Gaps as characters in sequence-based phylogenetic analyses. Syst. Biol. 49: 369–381. http://dx.doi.org/10.1093/sysbio/49.2.369 Sparg, S.G., Light, M.E. & van Staden, J. 2004. Biological activities and distribution of plant saponins. J. Ethnopharmacol. 94: 219–243. http://dx.doi.org/10.1016/j.jep.2004.05.016 Sukumaran, J. & Holder, M.T. 2010. DendroPy: A Python library for phylogenetic computing. Bioinformatics 26: 1569–1571. http://dx.doi.org/10.1093/bioinformatics/btq228 Swofford, D.L. 2003. PAUP*: Phylogenetic analysis using parsimony (*and other methods), version 4.0b10 for 32-bit Microsoft Windows. Sunderland, Massachusetts: Sinauer. Takhtajan, A. 2009. Flowering plants, 2nd ed. Dordrecht: Springer. http://dx.doi.org/10.1007/978-1-4020-9609-9 Valente, L.M., Savolainen, V. & Vargas, P. 2010. Unparalleled rates of species diversification in Europe. Proc. Roy. Soc. London, Ser. B, Biol. Sci. 277: 1489–1496. http://dx.doi.org/10.1098/rspb.2009.2163 Vvedensky, A.I. 1953. Acanthohpyllum. Pp. 409–414 in: Korovin, E.P. & (ed.), Flora of Uzbekistan, vol. 2. Tashkent: izdatelstvo Akademii nauk Uzbekskoi SSR. Williams, F.N. 1889. A revision of the genus Gypsophila. J. Bot. 27: 321–329. Wolfe, K.H., Li, W.H. & Sharp, P.M. 1987. Rates of nucleotide substitution vary greatly among plant mitochondrial, chloroplast, and nuclear DNAs. Proc. Natl. Acad. Sci. U.S.A. 84: 9054–9058. http://dx.doi.org/10.1073/pnas.84.24.9054 Yue, J.X., Li, J., Wang, D., Araki, H., Tian, D. & Yang, S. 2010. Genome-wide investigation reveals high evolutionary rates in annual model plants. B. M. C. Pl. Biol. 10: 242. http://dx.doi.org/10.1186/1471-2229-10-242 Yukhananov, D.KH. 1974. Taxonomic relations of genera Acanthophyllum C.A.Mey. and Allochrusa Bunge. Bot. Zhurn. (Moscow & Leningrad) 59: 414–417. Zakirov, K.Z. & Musaeva, M. 1981. O biomorfologii roda Acanthophyllum C.A.Mey. (Caryophyllaceae). Uzbeksk. Biol. Zhurn. 5: 28–30. Zohary, M. 1973. Geobotanical foundations of the Middle East, 2 vols. Stuttgart: Fischer. Appendix . Voucher information: species name, geographical origin, collector(s), voucher (herbarium), GenBank accession numbers for ITS and rps16, respectively (* indicates sequences new for this study). Acanthophyllum acerosum Sosn., Iran, W Azarbaijan, Zarre & al. 41900 (TUH), *KF924680, *KF924732; Acanthophyllum adenophorum Freyn, Iran, Khorassan, Pirani & Moazzeni 1947 (TMRC), *KF924679, *KF924731; Acanthophyllum andarabicum Podl. ex Schiman-Czeika, Afghanistan, Baghlan, Podlech 10985 (MSB), *KF924678, *KF924730; Acanthophyllum andersenii Rech.f. & Schiman-Czeika, Iran, Khorassan, Anders & Peterson 320 (GB), *KF924677, *KF924729; Acanthophyllum anisocladum Schiman-Czeika, Afghanistan, Lugar, Ekberg W9180 (GB), *KF924676, *KF924728; Acanthophyllum borsczowii Litv., Iran, Khorassan, Zarre & al. 41034 (TUH), *KF924675, *KF924727; Acanthophyllum bracteatum Boiss., Iran, Kohghiluyeh & Boyerahmad, Pirani & Moazzeni 2104 (TMRC), *KF924674, *KF924726; Acanthophyllum caespitosum Boiss., Iran, Hamadan, Zarre & al. 41903 (TUH), *KF924673, *KF924725; Acanthophyllum crassifolium Boiss., Iran, Mazandaran, Pirani & Moazzeni 41702 (TUH), *KF924672, *KF924724; Acanthophyllum crassinodum Yukhan. & J.R.Edm., Iran, Kerman, Pirani & Moazzeni 2923 (TMRC), *KF924671, *KF924723; Acanthophyllum diezianum Hand.-Mazz., Iran, Khorassan, Zarre & al. 41040 (TUH), *KF924670, *KF924722; Acanthophyllum ejtehadii Mahmoudi & Vaezi, Iran, Khorassan, Pirani & Moazzeni 2181 (TMRC), *KF924669, *KF924721; Acanthophyllum glandulosum Bunge ex Boiss., Iran, Khorassan, Zarre & al. 41037 (TUH), *KF924668, *KF924720; Acanthophyllum gracile Bunge ex Boiss., Iran, Semnan, Pirani & Moazzeni 2989 (TMRC), *KF924667, *KF924719; Acanthophyllum grandiflorum Stocks, Afghanistan, Bamian, Podlech 1340 (MSB), *KF924666, *KF924718; Acanthophyllum heratense Schiman-Czeika, Iran, Isfahan, Pirani & Moazzeni 2152 (TMRC), *KF924665, *KF924717; Acanthophyllum heterophyllum Rech.f., Iran, Kerman, Pirani & Moazzeni 2921 (TMRC), *KF924664, *KF924716; Acanthophyllum kabulicum Schiman-Czeika, Afghanistan, Ghazni, Frey s.n. (GB), *KF924663, *KF924715; Acanthophyllum kandaharicum Gilli, Iran, Khorassan, Joharchi & Zangouei 36245 (FUMH), *KF924662, *KF924714; Acanthophyllum korshinskyi Schischk., Iran, Khorassan, Pirani & Moazzeni 2123 (TMRC), *KF924661, *KF924713; Acanthophyllum kurdicum Boiss. & Hausskn. ex Boiss., Iran, Ilam, Hamzehee & Lashkarbolooki 1756 (TARI), *KF924660, *KF924712; Acanthophyllum laxiflorum Boiss., Afghanistan, Lugar, Ekberg W9184 (GB), *KF924659, *KF924711; Acanthophyllum laxiusculum Schiman-Czeika, Iran, Qom, Pirani & Moazzeni 1941 (TMRC), *KF924658, *KF924710; Acanthophyllum leucostegium Schiman-Czeika, Iran, Bandar Abbas, Ghahreman & Mozaffarian 5656 (TUH), *KF924657, *KF924709; Acanthophyllum lilacinum Schischk., Afghanistan, Badghis, Podlech & Jarmal 29855 (MSB), *KF924656, –; Acanthophyllum longicalyx Hedge & Wendelbo, Afghanistan, Jawzjan, Freitag 6580 (MSB), *KF924655, *KF924708; Draft version for proof purposes only. 15 ??? provide correct page nos. Your citations are inaccurate. You must check all. TAXON — 26 May 2014: 16 pp. Pirani & al. • Molecular phylogeny of Acanthophyllum TAXON — 26 May 2014: 16 pp. Appendix . Continued. Acanthophyllum macrodon Edgew., Afghanistan, Kandahar, Hedge & al. W7641 (GB), *KF924654, *KF924707; Acanthophyllum microcephalum Boiss., Iran, Tehran, Rajamand & Bazargan 32055 (TARI), *KF924653, *KF924706; Acanthophyllum mucronatum C.A.Mey., Iran, W Azarbaijan, Assadi & Olfat 68668 (TARI), *KF924652, *KF924705; Acanthophyllum oppositiflorum Aytaç, Turkey, Sivas, Aytaç 7476 (GAZI), *KF924651, *KF924704; Acanthophyllum paniculatum Regel & Herder, JN589016, –; Acanthophyllum pachycephalum Schiman-Czeika, Iran, Tehran, Ganjalizadeh 6106 (TUH), *KF924650, *KF924703; Acanthophyllum pachystegium Rech.f., Afghanistan, Badakhshan, Hedge & Wendelbo W9282 (GB), *KF924649, *KF924702; Acanthophyllum pleiostegium Schiman-Czeika, Afghanistan, Kataghan, Grey-Wilson & Hewer 1320 (GB), *KF924648, *KF924701; Acanthophyllum raphiophyllum (Rech.f.) Barkoudah, Afghanistan, Kapisa, Podlech 12548 (MSB), *KF924647, *KF924700; Acanthophyllum scapiflorum (Akhtar) Schiman-Czeika, Afghanistan, Kabul, Podlech 31232 (MSB), *KF924646, *KF924699; Acanthophyllum schugnanicum Schischk., Afghanistan, Bamian, Wendelbo & Ekberg W9796 (GB), *KF924645, *KF924698; Acanthophyllum sordidum Bunge ex Boiss., Iran, Isfahan, Pirani & Moazzeni 2147 (TMRC), *KF924644, *KF924697; Acanthophyllum speciosum Rech.f. & Schiman-Czeika, Iran, Khorassan, Pirani & Moazzeni 2186 (TMRC), *KF924643, *KF924733; Acanthophyllum spinosum C.A.Mey., Iran, Isfahan, Pirani & Moazzeni 2150 (TMRC), *KF924642, *KF924696; Acanthophyllum squarrosum Boiss., Iran, Semnan, Pirani & Moazzeni 2974 (TMRC), *KF924641, *KF924695; Acanthophyllum stewartii (Thoms. ex Edgew. & Hook.f.) Barkoudah, Afghanistan, Khost, Anders 8994 (MSB), *KF924640, *KF924694; Acanthophyllum stocksianum Boiss., Afghanistan, Kandahar, Toncev s.n. (MSB), *KF924639, *KF924693; Acanthophyllum subglabrum Schischk., Afghanistan, Nangarhar, Hedge & al. W7483 (GB), *KF924638, *KF924692; Acanthophyllum verticillatum Hand.-Mazz., Iran, Markazi, Mozaffarian & Sardabi 42175 (TARI), *KF924637, *KF924691; Acanthophyllum xanthoporphyranthum Hedge & Wendelbo, Afghanistan, Herat, Hedge & al. W8003 (GB), *KF924636, *KF924690; Acanthophyllum yasamin-nassehiae Joharchi & Pirani, Iran, Khorassan, Memariani & Zangouei 41448 (FUMH), *KF924635, *KF924689; Allochrusa bungei Boiss., Iran, E Azarbaijan, Rechinger 43834 (M), *KF924634, *KF924688; Allochrusa versicolor Boiss., Turkey, Kars, Nydegger 43597b (MSB), *KF924633, *KF924687; Bolbosaponaria bucharica Bondarenko, JN589057, –; Dianthus anatolicus Boiss., MA-690057, GU440777, –; Dianthus armeria L., –, FJ404903; Dianthus capitatus J.St.-Hil., GU440792, –; Dianthus carthusianorum L., –, EF674194; Dianthus charidemi Pau, GU440795, –; Dianthus ciliatus Guss., GU440798, –; Dianthus corymbosus Sibth. & Sm., GU440801, –; Dianthus costae Willk., GU440802, –; Dianthus crassipes De Roem., GU440803, –; Dianthus crinitus Sm., GU440805, –; Dianthus cyri Fisch. & C.A.Mey., GU440808, –; Dianthus diffusus Sm., GU440811, –; Dianthus erinaceus Boiss., GU440814, –; Dianthus gracilis Sm., JN589061, –; Dianthus hyssopifolius L., GU440826, –; Dianthus laricifolius Boiss. & Reut., GU440831, –; Dianthus microlepis Boiss., GU440840, –; Dianthus micropetalus Ser., GU440841, –; Dianthus orientalis Adams, GU440847, –; Dianthus pyrenaicus Bernh. ex Steud., GU440854, –; Dianthus serratifolius Sm., GU440858, –; Dianthus thunbergii S.S.Hooper, GU440872, –; Dianthus turkestanicus Preobr., GU440876, –; Dianthus versicolor Fisch. ex Link, GU440878, –; Diaphanoptera afghanica Podlech, Afghanistan, Baghlan, Podlech 21075 (MSB), *KF924632, –; Diaphanoptera ekbergii Hedge & Wendelbo, Afghanistan, Takhar, Podlech 11848 (MSB) & 11760 (MSB), *KF924631, *KF924686; Diaphanoptera lindbergii Hedge & Wendelbo, Afghanistan, Fariab, Hedge & al. W8336 (GB), *KF924630 *KF924685; Diaphanoptera stenocalycina Rech.f. & Schiman-Czeika, Iran, Golestan, Attar & Mehdigholi 24422 (TUH), *KF924629, *KF924684; Gypsophila arrostii Guss., JN589043, –; Gypsophila aucheri Boiss., JN589077, –; Gypsophila bicolor Grossh., JN589151, –; Gypsophila capituliflora Rupr., JN589143, –; Gypsophila cephalotes (Schrenk) Raikova, JN589105, –; Gypsophila cerastioides D.Don., Pakistan, Hazar, Ewald & Zetterlund 6227 (GB), *KF924628, *KF924683; Gypsophila curvifolia Fenzl, JN589159, –; Gypsophila desertorum Fenzl, JN589021, –; Gypsophila elegans M.Bieb., JN589130, –; Gypsophila fastigiata L., JN589144, –; Gypsophila heteropoda Freyn, JN589110, –; Gypsophila montserratii Fern.Casas, JN589155, –; Gypsophila muralis L., JN589037, –; Gypsophila paniculata L., JN589150, FJ404908; Gypsophila patrinii Ser., JN589076, –; Gypsophila pilulifera Boiss. & Heldr, JN589132, –; Gypsophila pinifolia Boiss. & Hausskn. ex Boiss., JN589050, –; Gypsophila scorzonerifolia Ser., JN589100, –; Gypsophila silenoides Rupr., JN589049, –; Gypsophila steveni Besser, JN589022, –; Gypsophila violacea Fenzl, JN589068, –; Ochotonophila allochrusoides Gilli, Afghanistan, Bamian, Wendelbo & Ekberg W9801 (GB), *KF924627, *KF924682; Petrorhagia prolifera (L.) P.W.Ball & Heywood, GU440883, –; Petrorhagia saxifrage Link, –, FJ404930; Petrorhagia thessala (Boiss.) P.W. Ball & Heywood, GU440885, –; Petrorhagia velutina (Guss.) P.W. Ball & Heywood, AY857974, –; Psammosilene tunicoides W.C.Wu & C.Y.Wu, JN589122, –; Saponaria ocymoides L., AY936271, FJ404936; Saponaria officinalis L., AY594313, FJ404937; Saponaria pumila Hayek, AY594311, –; Scleranthopsis aphanantha (Rech.f.) Rech.f., Afghanistan, Kabul, Rechinger 31265 (M), *KF924626, *KF924681; Silene ampullata Boiss., EF060223, –; Silene austroiranica Rech.f., Aellen &Esfand., EF060204, EF061364; Silene campanulata Pers., –, DQ908812; Silene cariensis Boiss., EF060205, EF061365; Silene conoidea L., FN821101, –; Silene corinthiaca Boiss., EF060206, EF061366; Silene dioica (L.) Clairv., –, FN821276; Silene echinosperma Boiss. & Heldr., X86845.1, Z83196; Silene fruticosa L., X86865, Z83188.1; Silene latifolia Poir., –, Z83171; Silene martyi Emb. & Maire, EF060213, EF061373; Silene mentagensis Coss., EF060236, EF061396; Silene nana Kar. & Kir, EF060217, EF061377; Silene nutans L., EF061361; Silene reticulata Desf., EF060216, Ef061376; Silene vulgaris (Moench) Garcke, –, EF674192; Vaccaria hispanica (Mill.) Rauschert, AY857969, –; Velezia rigida L., GU440888, –. 16 Draft version for proof purposes only.