Worberg & al. • Phylogenetic position of Huerteales TAXON 58 (2) • May 2009: 468–478 Huerteales sister to Brassicales plus Malvales, and newly circumscribed to include Dipentodon, Gerrardina, Huertea, Perrottetia, and Tapiscia Andreas Worberg1, Mac H. Alford2, Dietmar Quandt1,3 & Thomas Borsch1,4 1 Nees-Institut für Biodiversität der Pflanzen, Rheinische Friedrich-Wilhelms-Universität Bonn, Meckenheimer Allee 170, 53115 Bonn, Germany. andreas@worberg.de (author for correspondence) 2 Department of Biological Sciences, University of Southern Mississippi, 118 College Drive 5018, Hattiesburg, Mississippi 39406, U.S.A. 3 Institut für Botanik, Technische Universität Dresden, Zellescher Weg 20b, 01217 Dresden, Germany 4 Current address: Botanischer Garten und Botanisches Museum Berlin-Dahlem und Institut für Biologie, Freie Universität Berlin, Königin Luise-Str. 6–8, 14195 Berlin, Germany Sequence data from the matK gene, the trnK group II intron, the trnL group I intron and the trnL-F spacer were analysed for a broad sampling of the rosids and other eudicots. For the first time all putative genera of Dipentodontaceae and Tapisciaceae (Dipentodon, Huertea, Perrottetia, Tapiscia), as well as the recently described Gerrardinaceae were included in a molecular phylogenetic dataset. All genera were found in a well supported Huerteales clade. Moreover, with the rapidly evolving and non-coding plastid sequence data we were able to resolve the Huerteales clade to branch after Sapindales, and to be sister to a Brassicales-Malvales clade. Increased resolution and support among the malvids underscore the potential of plastid introns and spacers as well as the matK gene as phylogenetic markers in rosids. KEYWORDS: Huerteales, matK, rosids, trnK, trnL-F INTRODUCTION The large scale phylogenetic analysis of angiosperms based on rbcL, atpB, and 18S sequence data of Soltis & al. (2000) included one genus that was unresolved in either Brassicales, Malvales, or Sapindales in the eurosid II (malvid) clade. This genus, Tapiscia, was later inferred to be sister to Dipentodon (Dipentodontaceae), a relationship strongly supported (94% JK) based on combined analyses of matR, rbcL, and 18S DNA data (Peng & al., 2003) and matR alone (Zhu & al., 2007). The order Huerteales was thus described by Doweld (2001) to include the genus Dipentodon (Dipentodontaceae) and the two small genera Tapiscia and Huertea (Tapisciaceae), formerly placed together as subfamily Tapiscioideae of Staphyleaceae. Wu & al. (2002) later recognized the Dipentodontales as a new order in their vascular plant classification system, consisting of Dipentodon sinicus alone. All three genera are poorly known, but toothed leaf margins, 1–2 ovules per carpel, and relatively small embryos seem to be shared morphological characters (Stevens, 2001 onwards). Recently, an analysis based on several plastid and nuclear genes by Zhang & Simmons (2006) also placed Perrottetia (formerly Celastraceae) with Tapiscia and Dipentodon. Later, an analysis based on rbcL and atpB by Alford (2006) indicated that the genus Gerrardina (formerly Flacourtiaceae) was unplaced in the eurosids II (malvids) but potentially related to the Huerteales based on having two 468 ovules per carpel. However, analyses of rbcL and atpB datasets remained inconclusive, and Gerrardina could not be placed unambiguously in Huerteales. In the latest update of their angiosperm classification Thorne and Reveal (Thorne, 2007) grouped all three families into Huerteales within the newly described superorder named Huerteanae. No study, though, has included all of the genera potentially placed in Huerteales, thus underscoring the need for an analysis with complete taxon sampling. Pre-phylogenetic placements of genera that are now candidates as members of a monophyletic order Huerteales varied considerably among diverse angiosperm orders. Dipentodon sinicus Dunn is a small tree native to southern China, Burma and northern India and can be recognized by stipulate, serrate leaves, umbelliform inflorescences, and essentially free-central placentation (Takhtajan, 1997). This taxon was formerly included in Santalales (Cronquist, 1981) because of its placentation or in Violales (now Malpighiales) because of perianth morphology, wood anatomy, and pollen (Dahlgren, 1980; Thorne, 1992; Takhtajan, 1997). Dunn (1911) originally placed Dipentodon in Celastraceae, whereas a series of subsequent authors hypothesized it to be a member of Flacourtiaceae (including Samydaceae) mostly based on floral features (Sprague, 1925; Fischer, 1941; Loesener, 1942; Metcalfe & Chalk, 1950; Lobreau, 1969). Tapisciaceae consist of the two genera Tapiscia Ruiz & Pav. and Huertea Oliv. (Takhtajan, 1987). Tapiscia sinensis is native to China, TAXON 58 (2) • May 2009: 468–478 and the four known Huertea species are distributed from the West Indies and Central America through the Andes to Peru. They comprise small trees with spirally arranged compound leaves that are odd-pinnate or trifoliolate and have serrate margins (Krause, 1942; Stevens, 2001 onwards). Whereas Tapiscia was originally described in Sapindaceae by Oliver (1890), Huertea and Tapiscia were then mostly included together in Staphyleaceae (Bean, 1909; Diels, 1909; Schneider, 1912: p. 1026 and Fig. 607; Dickison, 1986) and placed in Sapindales (Cronquist, 1981). Takhtajan (1987) later described a separate family Tapisciaceae that remained classified in Sapindales. Perrottetia comprises about 15 species and is distributed in Asia from China to Malesia and southwards to northeastern Australia but also occurs in the Neotropics from Mexico to Peru. The floral structure of Perrottetia was studied in some detail by Matthews & Endress (2005), whereas that of the other genera is poorly known. All species of Perrottetia share two carpels per ovule and spirally arranged leaves (Stevens, 2001 onwards), characters also present in Dipentodon and Tapiscia. Gerrardina Oliv. consists of two species, Gerrardina eylesiana and Gerrardina foliosa, both of which are small trees or shrubs native to southern Africa (Malawi and Tanzania south to the Eastern Cape Province of South Africa). Species of Gerrardina have simple, alternate, stipulate leaves with serrate margins, stamens opposite the petals, and two apical ovules per carpel. Formerly Gerrardina was placed in Flacourtiaceae (including Samydaceae; Oliver, 1870; Warburg, 1894; Gilg, 1925). Sequences of rapidly evolving plastid spacers and introns have been very successfully used in phylogenetic analyses to infer deep relationships among families and orders of angiosperms (Borsch & al., 2003, 2005; Löhne & Borsch, 2005; Müller & al., 2006; Wanke & al., 2007; Worberg & al., 2007; Korotkova & al., 2009), and they were expected to provide some resolution to the above-mentioned phylogenetic ambiguity. Well known examples of such rapidly evolving genomic regions that also can be amplified using universal primers are the so-called trnT-trnF-region (composed of the trnT-L and trnL-F spacers and the group I intron in trnL; Taberlet & al., 1991; Borsch & al., 2003; Quandt & al., 2004), and the group II introns within trnK (Johnson & Soltis, 1994; Müller & Borsch, 2005a, b), rpl16 (Kelchner, 2002) and petD (Löhne & Borsch, 2005; Korotkova & al., 2009). In addition to high phylogenetic structure per nucleotide sequenced (Müller & al., 2006) these non-coding datasets yield additional indel matrices that usually further improve resolution and support of the reconstructed trees (Worberg & al., 2007; Borsch & al., 2007). Combining such intron and spacer sequence data with sequences of the rapidly evolving matK gene (Hilu & al., 2003) has been shown to lead to further improved resolution and Worberg & al. • Phylogenetic position of Huerteales support of phylogenetic hypotheses, for example, in early branching angiosperms (Borsch & al., 2005; Müller & al., 2006) or early branching eudicots (Worberg & al., 2007). Our approach was therefore to first combine matK sequence data with trnK group II intron sequence data. The matK gene is located within domain V of the trnK intron, and therefore both markers can be co-amplified easily. In the context of work aimed at understanding the mutational dynamics and phylogenetic signal in non-coding plastid regions in eudicots, a large dataset is currently being assembled for the rosids, a clade which comprises more than 70,000 species in ca. 140 families. One of the most prominent results is that we found strong support for the Huerteales clade in a well resolved eurosid II ( = malvid) clade, based on a combined matK/trnK dataset. Moreover, trnL-intron and trnL-F spacer data confirmed the monophyly of Huerteales and provided molecular evidence for the inclusion of Huertea, for which any kind of sequence data were previously unavailable. Considering the still unclear phylogenetic position of Gerrardinaceae (Alford, 2006) and the hitherto unresolved position of Dipentodontaceae and Tapisciaceae (see APG II, 2003), our study aims at contributing to the ongoing discussion by first clarifying the composition of the Huerteales clade, and by second resolving the phylogenetic relationships of Gerrardinaceae and Huerteales, respectively. MATERIAL AND METHODS Taxon sampling. — The dataset comprises 54 taxa from core eudicots with three representatives of Gunnerales as outgroup. Representatives for all rosid orders recognized by APG II (2003) were sampled with a focus on the eurosids II (malvid) clade to establish the framework for positioning Huerteales. Moreover, the sampling of this study represents major lineages also found in a more extensive combined trnK/matK plus trnL-F rosid analyses by the authors ( > 180 taxa, not shown here). All genera with putative affinities to Huerteales are included in this study. Taxa with complete matK sequences already available (Worberg & al., 2007) were completed by adding the trnK intron, what could easily be achieved through further sequencing reactions on already existing amplicons. In addition, some species with published completely sequenced plastid genomes were added. Material for further taxa came from the living collections at the Botanical Gardens Bonn, Dresden, and Berlin (FU-BGBM). A DNA extraction of Gerrardina foliosa was provided from the DNA Bank of the Royal Botanic Gardens, Kew, U.K. A list of all sampled taxa, their origin and voucher information is given in Appendix 1. 469 TAXON 58 (2) • May 2009: 468–478 Worberg & al. • Phylogenetic position of Huerteales DNA isolation, amplification and sequencing. — DNA was isolated from fresh or silica gel-dried plant material, using a CTAB method with three extractions (Borsch & al., 2003), designed to yield high amounts of genomic DNA. Fresh leaves were generally dried in silica gel before extraction. Dry tissue was ground to a fine powder using a mechanical homogenizer (Retsch MM200) with 5 mm beads at 30 Hz for 2 min. To identify mutational hotspots complete sequences of spacers and introns are necessary. We amplified with primers that were located sufficiently far away from the actual region under study. Sequencing was performed with either the universal primers already used for amplification or with additional internal primers, some of which were newly designed using SeqState v1.36 (Müller, 2005b). The matK gene was amplified within the trnK intron, either entirely or in two overlapping halves. Primers annealing to the trnK exons were trnKFbryo forward (Wicke & Quandt, in press) and trnK2R (Johnson & Soltis, 1994). To amplify two overlapping fragments, internal primers with higher taxon specificity primers were placed roughly 600 nt (reverse) and 450 nt (forward) downstream from the matK-start codon. Internal primers for respective families were either automatically selected from the eudicot matK primer database (www.eudicots.de) or designed newly with SeqState v1.36. The trnL intron and the trnL-F spacer (in the following called trnL-F region) were co-amplified using primers c and f of Taberlet & al. (1991) and sequenced with primers d of Taberlet & al. (1991) and trnL460F (Worberg & al., 2007) that is a forward primer universal for angiosperms annealing within the trnL intron. All primers used in this study are listed in Appendix 2 (in the online version of this article). Amplification and sequencing reactions were performed in a T3 Thermocycler (Biometra, Göttingen, Germany). Amplicons were purified with an Avegene gel extraction kit (Avegene) after running them out on a 1.2% agarose gel and excising the bands. The BeckmannCoulter DTCS QuickStart Reaction kit was used for direct sequencing. Temperature profiles and PCR reaction conditions for trnK/matK followed Müller & Borsch (2005a) and Borsch & al. (2003) for trnL-F. Extension products were run on BeckmannCoulter CEQ 8000 automated sequencers in Bonn or Dresden. Alternatively, sequences were generated on ABI377 autosequencers. Sequences were edited manually with PhyDE v0.995 (Müller & al., 2005). Alignment, indel coding and phylogenetic analysis. — The presence of microstructural changes, such as deletions, single sequence repeats, other insertions, and inversions, necessitates special attention to the alignment of sequences. Alignment was carried out by eye using PhyDE v0.995, applying the rules outlined in Borsch & al. (2003) and Löhne & Borsch (2005), as there are still no automated alignment algorithms that can handle length-variable DNA 470 regions adequately (Simmons & al., 2007). These alignment rules are based on recognizing sequence motifs that result from microstructural changes (Kelchner & Clark, 1997; Kelchner, 2000, 2002) rather than globally applying fixed gap costs. Inclusion of ambiguously alignable regions into character matrices (Lutzoni & al., 2000; Aagesen, 2004) were not pursued here. Mutational hotspots with unclear primary homology (Borsch & al., 2003) were excluded from tree inference in order to achieve maximum accuracy. To utilize indel characters we applied the simple indel coding method of Simmons & Ochoterena (2000) via SeqState v1.36 (Müller, 2005b). The resulting indel matrix was directly combined with the nucleotide-sequence matrix for parsimony and Bayesian Inference, as an increase of resolution and support by adding indel information from matK/trnK could be expected based on previous evidence (Müller & Borsch, 2005b; Worberg & al., 2007). To infer most parsimonious trees we used the Parsimony Ratchet (Nixon, 1999) as implemented in the program PRAP (Müller, 2004). Ratchet settings were 20 random addition cycles of 200 ratchet replicates and upweighting 25% of the characters. If more than one shortest tree was found, strict consensus trees were created. Nodes were evaluated by jackknifing in PAUP* 4b10 (Swofford, 2001) with 36.79% deletion of characters and 10,000 replicates saving only one tree per replicate. This approach follows studies on the reliability of jackknife percentages (Freudenstein & Simmons, 2004; Müller, 2005a). Bayesian inference (BI) was performed using the program MrBayes v3.1 (Ronquist & Huelsenbeck, 2003) applying the GTR + Γ + I model for sequence data and the adaption of the restriction site model (“F81”) for presence/absence for the indel matrix. Four runs (106 generations each) with four chains each were run simultaneously, starting from random trees. Chains were sampled every 10th generation. Stationarity of the runs was reached within the first 250,000 generations (burn-in). Calculations of the consensus tree and the posterior probability (PP) of clades were done based upon the trees sampled after the burn-in. Only PP’s of 90 and higher were considered significant (alpha = 0.1). Trees were drawn using TreeGraph (Müller & Müller, 2004). RESULTS The combined trnK and matK dataset, excluding mutational hotspots, comprises 3,334 substitution characters, of which 1,402 were parsimony-informative, 757 variable but not parsimony-informative, and 1,175 constant, and an additional 348 indel characters. The parsimony ratchet analysis yielded four most parsimonious trees with a length of 8,697 steps (consistency index, CI = 0.420; retention index, RI = 0.410; rescaled consistency index, TAXON 58 (2) • May 2009: 468–478 Worberg & al. • Phylogenetic position of Huerteales RC = 0.172; homoplasy index, HI = 0.580). The strict consensus tree is depicted in Fig. 1 and displays a topology with Gerrardina (Gerrardinaceae) well supported as part of the Huerteales clade (100% JK, 100 PP). Gerrardina is then sister to Tapisciaceae plus Dipentodontaceae (100% JK, 100 PP), Tapisciaceae, including Huertea and Tapiscia (100% JK, 100 PP) are sister to Dipentodontaceae (99% JK, 100 PP), and Dipentodon is sister to the 100 100 53 51 100 100 100 100 50 - 94 94 68 54 100 100 100 100 94 93 100 100 malvids 99 98 100 100 100 100 99 99 100 100 54 - 100 99 100 100 66 100 100 97 96 99 99 77 76 100 100 rosids 97 89 100 100 50 51 fabids 69 57 54 - 69 69 100 100 93 77 100 100 67 64 90 75 100 99 56 57 100 100 55 61 58 65 100 100 51 51 72 70 100 100 100 98 70 70 100 100 two species of Perrottetia sampled here (100% JK, 100 PP). The Huerteales are a highly supported clade (100% JK, 100 PP). They are well positioned within the eurosids II ( = malvids), which are overall highly supported (99% JK, 100 PP). Within malvids, Sapindales are sister to Huerteales + Brassicales + Malvales (94% JK, 100 PP), and Huerteales are sister to Brassicales + Malvales (94% JK, 100 PP). Separate analyses of trnK and matK (trees not BIX Bixa CIST Cistus MUNT Muntingia MALV Gossypium BOMB Durio THYM Daphne TROP Tropaeolum CARIC Vasconcellea RESED Reseda BRASS Lobularia DIPEN Perrottetia long. DIPEN Perrottetia ovat. DIPEN Dipentodon TAP Tapiscia TAP Huertea GERR Gerrardina RUTA Citrus MELI Melia SIMA Ailanthus SAPIN Acer ANAC Schinus PARN Parnassia CELAS Salacia OXAL Oxalis SALIC Populus PASS Passiflora EUPH Euphorbia RHIZO Kandelia CORI Coriaria CUCUR Cucumis JUG Juglans ROSAC Malus ROSAC Spiraea URTI Urtica FAB Glycine POLY Polygala GERAN Erodium ZYGO Larrea MELAS Melastoma ONAG Oenothera COMBR Combretum MYRT Eucalyptus STACH Stachyurus VIT Vitis LEEA Leea AQUI Ilex BALS Impatiens API Daucus ARALI Panax SAX Chrysosplenium AEX Aextoxicon MYRO Myrothamnus flab. MYRO Myrothamnus mosc. GUNN Gunnera Malvales Brassicales Huerteales Sapindales Celastrales Oxalidales Malpighiales Cucurbitales Fagales Rosales Fabales Geraniales Zygophyllales Myrtales Crossosomatales Vitales Asterids Gunnerales Outgroup Fig. 1. Combined strict consensus tree based on substitutions and indels of trnK and matK, inferred with maximum parsimony. Bold values above branches are Jackknife percentages with indels coded, plain ones below are Jackknife values without simple indel coding applied. 471 TAXON 58 (2) • May 2009: 468–478 Worberg & al. • Phylogenetic position of Huerteales shown) found identical topologies as depicted here but with generally lower node support. The Bayesian Analysis (Fig. 2) yielded nearly the same topology as Maximum Parsimony. Minor differences concern the unsupported position of the Geraniales and Zygophyllales, i.e., the monophyly of the fabids (compare Figs. 1 and 2). Whereas Geraniales and Zygophyllales are resolved within the fabid clade and share a moderate supported sister group relation under parsimony, the position is unresolved under a Bayesian framework. However, the remaining fabid clade reaches significance in the Bayesian analyses if information obtained from indels was included in the analyses. Moreover, two strong supported clades appear: (1) Malpighiales, Oxalidales and Celastrales, and (2) Rosales, Fabales, Cucurbitales, and Juglans (Fagales), with the former being resolved with high support under parsimony as well. To further clarify the composition of Huerteales we sampled trnL-F data for a smaller dataset (26 taxa). Considering the high statistical confidence of the matK/trnK trees with larger sampling, we restricted the analysis of trnL-F data BIX Bixa CIST Cistus THYM Daphne 100 MALV Gossypium 100 BOMB Durio 100 MUNT Muntingia TROP Tropaeolum CARIC Vasconcellea 100 RESED Reseda 100 100 BRASS Lobularia DIPEN Perrottetia long. 100 100 DIPEN Perrottetia ovat. 100 DIPEN Dipentodon TAP Tapisci malvids 100 100 TAP Huertea 100 GERR Gerrardina RUTA Citrus 100 SIMA Ailanthus 100 MELI Melia SAPIN Acer 100* ANAC Schinus PARN Parnassia 100 CELAS Salacia 100 OXAL Oxalis SALIC Populus 100 98 PASS Passiflora 100 EUPH Euphorbia rosids 100 92* RHIZO Kandelia CORI Coriaria 100 CUCUR Cucumis ROSAC Malus 100 100 ROSAC Spirea URTI Urtica 100 FAB Glycine 100 POLY Polygala JUG Juglans ZYGO Larrea GERAN Erodium MELAS Melastoma 99 MYRT Eucalyptus 100 COMBR Combretum ONAG Oenothera 100 STACH Stachyurus VIT Vitis 100 LEEA Leea API Daucus 100 100 ARALI Panax 100 AQUI Ilex 100 BALS Impatiens AEX Aextoxicon SAX Chrysosplenium MYRO Myrothamnus flab. 100 MYRO Myrothamnus mosc. GUNN Gunnera 0.1 100 472 Malvales Brassicales Huerteales Sapindales Celastrales Oxalidales Malpighiales Cucurbitales Rosales Fabales Fagales Zygophyllales Geraniales Myrtales Crossosomatales Vitales Asterids Gunnerales Fig. 2. Bayesian phylogram based on the combined trnK + matK matrix. Posterior probabilities are depicted above branches. * indicates that the respective node did not reach significance ( > 90) in the analysis after exclusion of the indel matrix. TAXON 58 (2) • May 2009: 468–478 Worberg & al. • Phylogenetic position of Huerteales to a malvid subset that included all six putative Huerteales genera (Fig. 3). The matrix comprised 1,231 characters of which 274 were parsimony-informative, and an additional 183 indel characters. Parsimony analysis yielded six shortest trees of 1,168 steps (CI = 0.622, RI = 0.440, RC = 0.273). The Huerteales clade was resolved with 88% JK, again rendering Gerrardina as sister to the remaining genera. Huertea was found as sister to Tapiscia (Fig. 3). Notably, the rather small trnL-F dataset also yielded good support for the support within all other malvid orders, Brassicales, Malvales, and Sapindales. BOMB Bombax BIX Bixa 89 90 MUNT Muntingia 58 76 82 80 97 94 CARIC Vasconcellea BRASS Lobularia TROP Tropaeolum 99 96 75 77 malvids 87 78 TAP Huertea DIPEN Perrottetia long. 99 100 88 85 TAP Tapiscia DIPEN Dipentodon 81 76 Brassicales CIST Cistus 53 - Huerteales 64 76 MALV Gossypium Malvales 100 100 DIPEN Perrottetia ovat. GERR Gerrardina 100 99 100 100 RUTA Citrus MELI Melia ANAC Schinus SIMA Brucea Sapindales 100 100 SAPIN Acer ZYGO Larrea GERAN Erodium MYRT Eucalyptus CORI Coriaria CELAS Salacia Outgroup 62 77 ROSAC Malus STACH Stachyurus Fig. 3. Combined strict consensus tree based on substitutions and indels of trnL-intron and trnL-F intergenic spacer region for a smaller sampling, inferred with MP. Bold values above branches are Jackknife percentages with indels coded, plain ones below are Jackknife values without simple indel coding applied. DISCUSSION Although morphological features and previous analyses of DNA sequence data could not confidently place Gerrardina (Alford, 2006), this study using matK/ trnK sequence data provides unambiguous support for a placement of Gerrardinaceae sister to a Dipentodontacae + Tapisciaceae clade (100% JK, 100 PP). We have thus enlarged the Huerteales to include Gerrardina. Moreover, phylogenetic analysis of trnK/matK and trnL-F sequence data shows for the first time with molecular data that Huertea is the sister group of Tapiscia. Furthermore, our results provide strong evidence for the interrelationships of the orders of the eurosids II (malvids). Brassicales and Malvales are sister, which are then sister to Huerteales, which are then sister to Sapindales. This branching pattern has strong support, regardless of the analytical approach. Our results are a further example of the effectiveness of matK/trnK and trnL-F as phylogenetic markers at deeper levels in eudicots. High performance of partial matK was earlier shown by Hilu & al. (2003) for angiosperms. In addition to larger quantities of informative sites in such rapidly evolving and non-coding genomic regions, Müller & al. (2006) found increased phylogenetic signal per informative character of rapidly evolving or non-coding markers like matK or trnT-F in contrast to slow evolving regions like rbcL. Complete matK sequences alone (tree not shown) inferred a topology similar to the complete angiosperm analysis made by Hilu & al. (2003; Huerteales not included) with high support for most nodes. This study is the first to apply trnK group II intron sequences at deeper level in angiosperms. Similar to the addition of petD group II intron sequence data to a matK + trnL-F matrix in eudicots (Worberg & al., 2007), the addition of trnK drastically improved the signal, especially for deeper nodes (e.g., raising the JK support for rosids from 88% to 97%). The results presented here were based on analyses of all putative genera of the Huerteales clade. Gerrardina is consistently inferred as sister to all other genera (Figs. 1–3) whereas the detailed relationships within the clade comprised of Perrottetia, Dipentodon, Huertea and Tapiscia await additional sequence data. Both Huertea and Tapiscia share a large number of indels in trnL-F and trnK/matK that further underscores their close relationship. Whereas trnK/matK depict Dipentodon and Perrotetia as sisters with 99% JK (100 PP) support, trnL-F signal is inconsistent (75% JK) and therefore not conclusive to this question. Perrottetia, which previously had been classified within Celastraceae, does not belong in this family anymore. Matthews & Endress (2005) found substantial floral structural differences between Perrottetia and Celastrales. While reconstructing Celastrales phylogeny, Zhang & Simmons (2006) found Perrottetia among the outgroup taxa of their taxon set. Zhang & Simmons (2006) further 473 Worberg & al. • Phylogenetic position of Huerteales investigated the position of Perrottetia by including newly generated 18S, atpB and rbcL sequences into the 567-taxon dataset of Soltis & al. (2000) and found Perrottetia as sister to the monotypic Tapiscia with 100% JK support. Our study supports these results, placing the two Perrottetia species with high statistical confidence into the Dipentodontaceae clade (JK 98%, 100 PP) sister to Dipentodon sinicus. Alford (2006) recovered Gerrardinaceae in the malvid clade using rbcL+atpB sequences from Soltis & al. (2000) with the addition of Gerrardina foliosa and Dipentodon sinicus to the sampling. Although Gerrardinaceae were recovered in a clade of Huerteales + Malvales + Brassicales (with 70% JK), its relationships among those orders were unresolved. While ndhF-sequences (3 end only), for which less comparative data are available, showed an affinity to Tapiscia, Dipentodon, Brassicales, and Malvales, analyses of atpB and rbcL sequences did not unambiguously reveal the relationship of Gerrardinaceae within the eurosids II. Phylogenetic relationships in the malvids ( = eurosids II) are depicted in our study as Brassicales being sister to Malvales, with Huerteales being sister to the Brassicales + Malvales clade. Finally, Sapindales are shown to be sister of all remaining malvids. The latter topology was already hypothesised by partial matK sequences (Hilu & al., 2003), although no member of the Huerteales was represented. The three gene analysis of Soltis & al. (2000) could not resolve relationships among Malvales, Sapindales, Brassicales and Tapiscia using maximum parsimony. Huerteales were represented by Tapiscia alone in Soltis & al. (2000). However, in their subsequent Bayesian analysis of the same dataset, Soltis & al. (2007) arrived at a different malvid topology, with Tapiscia sister to Brassicales (unsupported), and this clade being sister to a well-supported Malvales + Sapindales clade. Similarly, the recent addition of mitochondrial matR sequences to a rbcL + atpB + 18SrDNA dataset (Zhu & al., 2007) indicated a position of Tapiscia sister to Brassicales, albeit without support. The presented analyses, however, clearly resolve the Huerteales sister to Malvales and Brassicales irrespective of the optimization criterion with strong support. As a taxonomic order, Huerteales is a recent creation based almost entirely on analyses of DNA sequence data (Soltis & al., 2000; Doweld, 2001; Peng & al., 2003). The taxa of Huerteales, in general, are poorly known and have been considered as peripheral members of various families, often shuffled among notoriously diverse families such as Celastraceae and Flacourtiaceae. Much of the detailed morphological and anatomical work for adequate comparisons, however, remains to be done. In particular, Gerrardina is lacking study of wood anatomy, and data on seed anatomy and secondary chemistry for all of the genera are scarce. Thus, one would not be surprised to learn that morphological synapomorphies for Huerteales are not yet clear. Furthermore, since the composition of 474 TAXON 58 (2) • May 2009: 468–478 Huerteales has been modified drastically in its short lifespan, phylogenetic studies utilizing morphological data or mapping morphological features onto molecular trees have not incorporated all of the relevant genera (cf. Soltis & al., 2005; Ronse de Craene & Haston, 2006). The highly supported resolution resulting from this study now permits the formation of preliminary hypotheses about the evolution of the order and highlights types of data that may be useful in further clarifying morphological evolution within the eurosids II. Table 1 summarizes the current state of knowledge on morphology and anatomy of Huerteales taxa and is the basis for our discussion. Within Huerteales, Dipentodon and Perrottetia are sister, a relationship suspected even in morphological and pre-phylogenetic works (e.g., Dunn, 1911; Fischer, 1941; Liu & Cheng, 1991). They both have stipulate, toothed leaves, poorly differentiated perianth (sepals and petals similar), a nectary disk, and similar wood anatomy. Zhang & Simmons (2006) recognize them together in the family Dipentodontaceae. Huertea and Tapiscia are not obvious relatives of Dipentodon and Perrottetia morphologically, but they do share scalariform perforation plates in the vessels (Carlquist & Hoekman, 1985), septate fibers (lacking in Perrottetia), stamen number equal to the petal number, and stamens alternating with the petals. Unlike Dipentodontaceae, Huertea and Tapiscia have compound leaves, a calyx tube, and clearly differentiated calyx and corolla. Tapiscia also lacks a nectary disk. Previously recognized as a subfamily of the Staphyleaceae, Huertea and Tapiscia are now recognized as the family Tapisciaceae. Rarely, the two genera have been placed in monogeneric families, Tapisciaceae and Huerteaceae (Doweld, 2001). The final member of Huerteales is the monogeneric family Gerrardinaceae. Like other Huerteales, Gerrardina has toothed leaves, cymose inflorescences, a hypanthium (short in Tapiscia and Huertea), a unilocular ovary with 1–2 ovules per carpel, and small embryos. Although not present in all members of Huerteales, Gerrardina also has stipules (lacking in Huertea), mucilage cells (lacking in Dipentodon), and a nectary disk (lacking in Tapiscia). Gerrardina differs from the other members of Huerteales in having stamens opposite the petals. Comparison of the features of Huerteales and the other orders of the eurosids II (Sapindales, Malvales, Brassicales) may provide clues for uncovering potential synapomorphies for Huerteales. Wood anatomy seems a good place to start. Sapindales, Malvales, and Brassicales usually have simple perforation plates and axial parenchyma in the wood, but Huerteales have scalariform plates (rarely simple in Perrottetia; unknown for Gerrardina) and lack axial parenchyma (except Perrottetia, where it is paratracheal; again, unknown for Gerrardina). Although nectary disks are common in Sapindales, they are not usually associated with a hypanthium (albeit short in Tapisciaceae), as in Huerteales. TAXON 58 (2) • May 2009: 468–478 Worberg & al. • Phylogenetic position of Huerteales Table 1. Representative anatomical and morphological features of the taxa included in Huerteales, as well as predominant states for the related orders Sapindales, Malvales, and Brassicales. Feature Gerrardina Tapiscia Huertea Dipentodon Perrottetia Sapindales Malvales Vessel perforation plates ? Scalariform Scalariform Scalariform Scalariform (rarely simple) Simple Simple Simple Septate fibers ? Present Present Present Absent Often present Absent Absent or present (then plesiomorphic) Axial parenchyma in wood Brassicales ? Absent Absent Absent Present Present Present Present Mucilage cells Present Present Present Absent Present Sometimes present Present (except Cistaceaee) Absent Leaf complexity Simple Simple Simple Simple or Simple or compound compound Stipules Present Present (modified)/ absent Present Present Absent Present Usually absent Imbricate +/– valvate Valvate Open to valvate Imbricate Valvate Mostly imbricate Sepal aestivation Imbricate Compound Compound Present Simple or compound Hypanthium (floral cup) Present Present Present (but short) Present Present Usually absent Absent Usually absent Nectary disk Present Absent Present Present Present Present Absent Absent Multilocular Usually multilocular Usually multilocular Ovary Unilocular Unilocular Unilocular Unilocular Unilocular (with basal (with apical septum) septum) Carpels 2 2 2 3 2 2–5(–6) 2–5(–many) 2–6(–12) Ovules/carpel 2 1 1 2 2 1–2 1–∞ 1–2(–∞) Placentation Apical Basal Axile Axile or parietal Embryo Straight Straight Folded Straight or curved Axile/ Free-central Basal and Axile, basal-axile lateral basal-axile, apical-axile Straight Straight ? Straight or curved General information compiled from Metcalfe & Chalk (1950), Cronquist (1981), Takhtajan (1997), Stevens (2001 onwards), and Ronse de Craene & Haston (2006). Specific data for Gerrardina from Alford (2006); for Tapiscia and Huertea from Krause (1942), Carlquist & Hoekman (1985), and Dickison (1986); for Dipentodon from Dunn (1911), Merrill (1941), Liu & Cheng (1991), and Zhang & Gao (1995); and for Perrottetia from Matthews & Endress (2005). The nectary disks in Huerteales are non-vascularized or only supplied by phloem (Dickison, 1986; Matthews & Endress, 2005). Huerteales also share unilocular ovaries, although this feature may occasionally occur in the other orders. Some characters show remarkable diversity within Huerteales. Leaf complexity, presence/absence of stipules, and position of stamens have already been mentioned, but the order also includes taxa with imbricate, valvate, and open sepal aestivation, uniseriate and homocellular to multiseriate and heterocellular wood rays, and apical, basal, partially axile, and free-central placentation. APG (2003) treat Tapisciaceae as an unplaced family in malvids. In addition to insufficient tree resolution, earlier analyses were also incomplete in taxon sampling, thereby hindering the development of a strongly supported ordinal classification. We therefore recognise the order Huerteales undoubtedly as a monophyletic group of families with ordinal level that has a distinct position within the malvid grade with high support. Thus we support an angiosperm classification system with the order Huerteales included into the rosids. ACKNOWLEDGEMENTS We acknowledge support by the Deutsche Forschungsgemeinschaft (DFG) for the project “Mutational dynamics of non-coding genomic regions and their potential for reconstructing eudicot evolution” (grants BO1815/2 to T.B. and QU153/2 to D.Q.). Most of the material was provided by the living collections of the Botanical Gardens of Bonn University and Dresden Technical University. The authors like to thank Peter 475 Worberg & al. • Phylogenetic position of Huerteales Stevens for helpful comments on the manuscript. For various kinds of support and helpful discussions we are grateful to Wilhelm Barthlott (Bonn), Mark W. Chase (K), Jay Horn (FTG), Nadja Korotkova (Bonn), Xia Lei (Cornell University), Anna-Magdalena Barniske (Dresden), Kai Müller (Bonn). We would like to express our sincere thanks to Armando Urquiloa (Jardín Botánico de Pinar del Río), Ramona Oviedo (HAC) and Werner Greuter (B) in organizing to collect fresh material of Huertea cubensis, and to Angela T. Leiva Sánchez, Jardín Botánico Nacional, Havana, for arranging the permission to export material. Helpful comments of three anonymous reviews are gratefully acknowledged. T.B. acknowledges a Heisenberg scholarship by the Deutsche Forschungsgemeinschaft (DFG) that greatly facilitated his work on angiosperm phylogenetics. This study is in partial fulfilment for the requirements to obtain a Ph.D. from Bonn University by the first author. LITERATURE CITED Aagesen, L. 2004. The information content of an ambiguously alignable region, a case study of the trnL intron from the Rhamnaceae. Org. Divers. Evol. 4: 35–49. Alford, M.H. 2006. Gerrardinaceae: a new family of African flowering plants unresolved among Brassicales, Huerteales, Malvales, and Sapindales. Taxon 55: 959–964. APG II. 2003. An update of APG classification for the orders and families of flowering plants. Bot. J. Linn. Soc. 141: 399–436. Bean, W.J. 1909. Garden notes on new trees and shrubs. Bull. Misc. Inform. Kew 1909: 355–356. Borsch, T. 2000. Phylogeny and Evolution of the Genus Nymphaea. Ph.D. thesis, University of Bonn, Bonn. Borsch, T., Hilu, K.W., Quandt, D., Wilde, V., Neinhuis, C. & Barthlott, W. 2003. Non-coding plastid trnT-trnF sequences reveal a well supported phylogeny of basal angiosperms. J. Evol. Biol. 16: 558–576 Borsch, T., Hilu, K.W., Wiersema, J.H., Löhne, C., Barthlott, W. & Wilde, V. 2007. Phylogeny of Nymphaea (Nymphaeaceae): evidence from substitutions and microstructural changes in the chloroplast trnT-trnF region. Int. J. Pl. Sci. 168: 639–671. Borsch, T., Löhne, C., Müller, K., Hilu, K.W., Wanke, S., Worberg, A., Barthlott, W., Neinhuis, C. & Quandt, D. 2005. Towards understanding basal angiosperm diversification: recent insights using rapidly evolving genomic regions. Nova Acta Leop. 92: 85–110. Carlquist, S. & Hoekman, D.A. 1985. Wood anatomy of Staphyleaceae: ecology, statistical correlations, and systematics. Flora 177: 195–216. Cronquist, A. 1981. An Integrated System of Classification of Flowering Plants. Columbia University Press, New York. Dahlgren, R. 1980. A revised system of classification of the angiosperms. Bot. J. Linn. Soc. 80: 91–124. Dickison, W.C. 1986. Floral morphology and anatomy of Staphyleaceae. Bot. Gaz. 147: 312–326. Diels, L. 1909. Staphyleaceae. Bot. Jahrb. Syst. 29: 448. Doweld, A.B. 2001. Prosyllabus Tracheophytorum. Tentamen systematis plantarum vascularium (Tracheophyta). Geos, Moscow. 476 TAXON 58 (2) • May 2009: 468–478 Dunn, A. 1911. Dipentodon, a new genus of uncertain systematic position. Bull. Misc. Inform. Kew 1911: 310–313. Fischer, C.E.C. 1941. Contributions to the flora of Burma: XVIII. Kew Bull. 1940: 282–294. Freudenstein, J.V. & Simmons, M.P. 2004. Relative effects of increasing genetic distance on alignment and phylogenetic analysis. Cladistics 20: 83–83. Gilg, E. 1925. Flacourtiaceae. Pp. 377–457 in: Engler, A. & Prantl, K. (eds.), Die natürlichen Pflanzenfamilien, ed. 2, vol. 21. Duncker & Humblot, Berlin. Hilu, K.W., Borsch, T., Müller, K., Soltis, D.E., Soltis, P.S., Savolainen, V., Chase, M., Powell, M., Alice, L.A., Evans, R., Sauquet, H., Neinhuis, C., Slotta, T.A., Rohwer, J.G., Campbell, C.S. & Chatrou, L. 2003. Angiosperm phylogeny based on matK sequence information. Amer. J. Bot. 90: 1758–1776. Johnson, L.A. & Soltis, D.E. 1994. MatK DNA sequences and phylogenetic reconstruction in Saxifragaceae sensu stricto. Syst. Bot. 19: 143–156. Johnson, L.A. & Soltis, D.E. 1995. Phylogenetic inference in Saxifragaceae sensu stricto and Gilia (Polemoniaceae) using matK sequences. Ann. Missouri Bot. Gard. 82: 149–175. Kelchner, S.A. 2000. The evolution of noncoding chloroplast DNA and its application in plant systematics. Ann. Missouri Bot. Gard. 87: 482–498. Kelchner, S.A. 2002. Group II introns as phylogenetic tools: structure, function, and evolutionary constraints. Amer. J. Bot. 89: 1651–1669. Kelchner, S.A. & Clark, L.G. 1997. Molecular evolution and phylogenetic utility of the rpl16 intron in Chusquea and the Bambusoideae (Poaceae). Molec. Phylog. Evol. 8: 385–397. Korotkova, N., Schneider, J.V., Quandt, D., Worberg, A., Zizka, G. & Borsch, T. 2009. Phylogeny of the eudicot order Malpighiales — analysis of a recalcitrant clade with sequences of the petD group II intron. Pl. Syst. Evol. doi 10.1007/s00606-008-0099-7. [Special issue “Molecular evolution and phylogenetic utility of non-coding DNA – applications from species to deep level questions.”] Krause, J. 1942. Staphyleaceae. Pp. 255–321 in: Engler, A. (ed.), Die natürlichen Pflanzenfamilien, ed. 2, vol. 20b. Duncker & Humblot, Berlin. Liang H. & Hilu K.W. 1996. Application of the matK gene sequences to grass systematics. Canad. J. Bot. 74: 125–134. Liu, J. & Cheng, C. 1991. On the systematic position of genus Dipentodon Dunn. J. Wuhan Bot. Res. 9: 29–39. Lobreau, D. 1969. Les limites de l’ “ordre” des Celastrales d’apres le pollen. Pollen & Spores 11: 499–555. Loesener, T. 1942. Celastraceae. Pp. 87–197 in: Engler, A. (ed.), Die natürlichen Pflanzenfamilien, ed. 2, vol. 20b. Duncker & Humblot, Berlin. Löhne, C. & Borsch, T. 2005. Molecular evolution and phylogenetic utility of the petD group II intron: a case study in basal angiosperms. Molec. Biol. Evol. 22: 317–332. Lutzoni, F., Wagner, P., Reeb, V. & Zoller, S. 2000. Integrating ambiguously aligned regions of DNA sequences in phylogenetic analyses without violating positional homology. Syst. Biol. 49: 628–651. Matthews, M.L. & Endress, P.K. 2005. Floral structure and systematics in Celastrales (Celastraceae, Parnassiaceae, Lepidobotryaceae). Bot. J. Linn. Soc. 149: 129–194. Matthews, M.L. & Endress, P.K. 2006. Floral structure and systematics in four orders of rosids, including a broad TAXON 58 (2) • May 2009: 468–478 survey of floral mucilage cells. Pl. Syst. Evol. 260: 199– 221. Merrill, E.D. 1941. The Upper Burma plants collected by Captain F. Kingdon Ward on the Vernay-Cutting Expedition, 1938–1939. Brittonia 4: 20–188. Metcalfe, C.R. & Chalk, L. 1950. Anatomy of the Dicotyledons: Leaves, Stem, and Wood in Relation to Taxonomy with Notes on Economic Uses, 1st ed., vols. 1–2. Clarendon Press, Oxford, U.K. Müller, K. 2004. PRAP-computation of Bremer support for large data sets. Molec. Phylog. Evol. 31: 780–782. Müller, K. 2005a. The efficiency of different search strategies in estimating parsimony jackknife, bootstrap, and Bremer support. BMC Evol. Biol. 5: 58. Müller, K. 2005b. SeqState: primer design and sequence statistics for phylogenetic DNA datasets. Appl. Bioinformatics 4: 65–69. Müller, K. & Borsch, T 2005a. Phylogenetics of Amaranthaceae based on matK/trnK sequence data — evidence from parsimony, likelihood, and Bayesian analyses. Ann. Missouri Bot. Gard. 92: 66–102. Müller, K. & Borsch, T. 2005b. Phylogenetics of Utricularia (Lentibulariaceae) and molecular evolution of the trnK intron in a lineage with high substitutional rates. Pl. Syst. Evol. 250: 39–67. Müller, K., Borsch, T. & Hilu, K.W. 2006. Phylogenetic utility of rapidly evolving DNA at high taxonomical levels: contrasting matK, trnT-F, and rbcL in basal angiosperms. Molec. Phylog. Evol. 41: 99–117. Müller, J. & Müller, K. 2004. TREEGRAPH: automated drawing of complex tree figures using an extensible tree description format. Molec. Ecol. Notes 4: 786–788. Müller, J., Müller, K., Quandt, D. & Neinhuis, C. 2005. PhyDE — www.phyde.de Nandi, O. I., Chase, M.W. & Endress, P.K. 1998. A combined cladistic analysis of angiosperms using rbcL and non-molecular data sets. Ann. Missouri Bot. Gard. 85: 137–212. Nixon, K.C. 1999. The parsimony ratchet, a new method for rapid parsimony analysis. Cladistics 15: 407–414. Oliver, D. 1870. Gerrardina foliosa Oliv. T. 1075 in: Hooker, J.D. (ed.), Hooker’s Icones Plantarum, vol. 11. Longman, Rees, Orme, Brown, Green, & Longman, London. Oliver, D. 1890. Tapiscia sinensis. T. 1928 in: Hooker, J.D. (ed.), Hooker’s Icones Plantarum, vol. 20. Longman, Rees, Orme, Brown, Green, & Longman, London. Peng, Y., Chen, Z., Gong, X., Zhong, Y. & Shi, S. 2003. Phylogenetic position of Dipentodon sinicus: evidence from DNA sequences of chloroplast rbcL, nuclear ribosomal 18S, and mitochondrial matR. Bot. Bull. Acad. Sin. 44: 217–222. Quandt, D., Müller, K., Stech, M., Frahm, J.-P., Frey, W., Hilu, K.W. & Borsch, T. 2004. Molecular evolution of the chloroplast trnL-F region in land plants. Pp. 13–37 in: Goffinet, B., Hollowell, V. & Magill, R. (eds.), Molecular Systematics of Bryophytes. Monographs in Systematic Botany from the Missouri Botanical Garden 98. Missouri Botanical Garden, St. Louis. Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574. Ronse de Craene, L.P. & Haston, E. 2006. The systematic relationships of glucosinolate-producing plants and related Worberg & al. • Phylogenetic position of Huerteales families: a cladistic investigation based on morphological and molecular characters. Bot. J. Linn. Soc. 151: 453–494. Schneider, C.K. 1912. Illustriertes Handbuch der Laubholzkunde, vol. 2. Gustav Fischer, Jena. Simmons, M.P., Müller, K.F. & Norton, A.P. 2007. The relative performance of indel-coding methods in simulations. Molec. Phylog. Evol. 44: 724–740. Simmons, M.P. & Ochoterena, H. 2000. Gaps as characters in sequence-based phylogenetic analyses. Syst. Biol. 49: 369–381. Solereder, H. 1932. Über die Staphyleaceengattung Tapiscia Oliv. Ber. Deutsch. Bot. Ges. 10, 545-551. Soltis, D.E., Gitzendanner, M.A. & Soltis, P.S. 2007. A 567taxon data set for angiosperms: the challenges posed by Bayesian analyses of large data sets. Int. J. Pl. Sci. 168: 137–157. Soltis, D.E., Soltis, P.S., Chase, M.W., Mort, M.E., Albach, D.C., Zanis, M., Savolainen, V., Hahn, W.H., Hoot, S.B., Fay, M.F., Axtell, M., Swensen, S.M., Price, L.M., Kress, W.J., Nixon, K.C. & Farris, J.S. 2000. Angiosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequences. Bot. J. Linn. Soc. 133: 381–461. Soltis, D.E., Soltis, P.S., Endress, P.K. & Chase, M.W. 2005. Phylogeny and Evolution of Angiosperms. Sinauer Associates, Sunderland. Sprague, T.A. 1925. Samydaceae. J. Bot. 63: 9–13. Steele, K.P. & Vilgalys, R. 1994. Phylogenetic analyses of Polemoniaceae using nucleotide sequences of the plastid gene matK. Syst. Bot. 19: 126–142. Stevens, P.F. 2001 onwards. Angiosperm Phylogeny Website, vers. 9, June 2008. www.mobot.org/MOBOT/research/ APweb/ Swofford, D.L. 2001. PAUP*: Phylogenetic Analysis Using Parsimony (*and Other Methods). Sinauer, Sunderland. Taberlet, P., Gielly, L., Pautou, G. & Bouvet, J. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Pl. Molec. Biol. 17: 1105–1109. Takhtajan A.L. 1987. System of Magnoliophyta. Academy of Sciences U.S.S.R., Leningrad. Takhtajan, A.L. 1997. Diversity and Classification of Flowering Plants. Columbia University Press, New York. Thorne, R.F. 1992. An updated phylogenetic classification of the flowering plants. Aliso 13: 365–390. Thorne, R.F. 2007. An updated classification of the class Magnoliopsida (“Angiospermae”). Bot. Rev. 73: 67–182. [“with many nomenclatural additions by James L. Reveal.”] Wanke, S., Gonzalez, F. & Neinhuis, C. 2006a. Systematics of pipevines: combining morphological and fast-evolving molecular characters to investigate the relationships within subfamily Aristolochioideae (Aristolochiaceae). Int. J. Pl. Sci. 167: 1215–1227. Wanke, S., Jaramillo, M.A., Borsch, T., Samain, M.-S., Quandt, D. & Neinhuis, C. 2007. Evolution of Piperales — combined matK gene and trnK intron sequence data reveal lineage specific resolution contrast. Molec. Phylog. Evol. 42: 477–497. Wanke, S., Samain, M.S., Vanderschaeve, L., Mathieu, G., Goetghebeur, P. & Neinhuis, C. 2006b. Phylogeny of the genus Peperomia (Piperaceae) inferred from the trnK/ matK region (cpDNA). Pl. Biol. 8: 93–102. Warburg, O. 1894. Flacourtiaceae. Pp. 1–56 in: Engler, A. & Prantl, K. (eds.), Die natürlichen Pflanzenfamilien, ed. 1, vol. 3(6a). Engelmann, Leipzig. 477 Worberg & al. • Phylogenetic position of Huerteales Wicke, S. & Quandt, D. In press. Universal primers for the amplification of the plastid trnK/matK region in land plants. Anales Jard. Bot. Madrid 66. Worberg, A., Quandt, D., Barniske, A.-M., Löhne, C., Hilu, K.W. & Borsch, T. 2007. Phylogeny of basal eudicots: insights from non-coding and rapidly evolving DNA. Org. Divers. Evol. 7: 55–77. Wu, Z.-Y., Lu, A.-M., Tang, Y.-C., Chen, Z.-D. & Li, D.Z. 2002. Synopsis of a new “polyphyletic-polychronicpolytopic” system of the angiosperms. Acta Phytotax. Sin. 40: 298–322. TAXON 58 (2) • May 2009: 468–478 Zhang, L.-B. & Simmons, M.P. 2006. Phylogeny and delimitation of the Celastrales inferred from nuclear and plastid genes. Syst. Bot. 31: 122–137. Zhang, X.-Y. & Gao, X.-Z. 1995. Anatomical studies on secondary xylem and secondary phloem of Dipentodon sinicus. Acta Bot. Sin. 37: 534–538. Zhu, X.-Y., Chase, M.W., Qiu, Y.-L., Kong, H.-Z., Dilcher, D.L., Li, J.-H. & Chen, Z.-D. 2007. Mitochondrial matR sequences help to resolve deep phylogenetic relationships in rosids. BMC Evol. Biol. 7: 217 Appendix 1. Taxa and GenBank/EMBL/DDBJ accession numbers of the sequences used in this study. Species in alphabetical order and their corresponding family following APG II (2003), voucher information, garden or field origin and GenBank/ EMBL/DDBJ accession numbers of matK, trnL-F and trnK with corresponding citations. New sequences generated in this study are labeled with “this study”. Dashes indicate missing data. Acer campestre L., Sapindaceae, EMBL/GenBank, –, AJ438793 Bittkau & Mueller-Starck (2002), AF401189 Tian & al. (2001), AJ438793 Bittkau & MuellerStarck (2002); Aextoxicon punctatum Ruiz & Pav., Aextoxicaceae, T. Borsch 3459 (BONN), BG Bonn, DQ182342 Müller & al. (2006), – , DQ182342 this study; Ailanthus altissima Swingle, Simaroubaceae, A. Worberg 025 (BONN), BG Bonn, FM179922 this study, – , FM179922 this study; Bixa orellana L., Bixaceae, A. Worberg 040 (BONN), BG Bonn, FM179929 this study, FM179540 this study, FM179929 this study; Bombax malabaricum DC., Bombacaceae, EMBL/GenBank, – , – , AY328149 Yuan & al. (2003), – ; Brucea javanica (L.) Merr., Simaroubaceae, EMBL/GenBank, – , – , AB365015, AB365024 Tanaka & al. (2007), – ; Chrysosplenium alternifolium L., Saxifragaceae, T. Borsch s.n. (BONN), Germany, AM396496 Worberg & al. (2007), –, AM396496 this study; Cistus ladanifer L., Cistaceae, A. Worberg s.n. (BONN), BG Bonn, FM179939 this study, FM179538 this study, FM179939 this study; Citrus sinensis Osbeck, Rutaceae, EMBL/GenBank, – , NC_008334 Bausher & al. (2006), NC_008334 Bausher & al. (2006), NC_008334 Bausher & al. (2006); Combretum molle R. Br. ex G. Don, Bombacaceae, IPEN-xx-0-B-1560480, BG Berlin, FM179938 this study, – , FM179938 this study; Coriaria myrtifolia L., Coriariaceae, T. Borsch 3415 (BONN), BG Bonn, AF542600 Worberg & al. (2007), AM397179 Worberg & al. (2007), AF542600 this study; Cucumis sativus L., Cucurbitaceae, EMBL/GenBank, – , DQ119058 Kim & al. (2006), – , DQ119058 Kim & al. (2006); Daphne bholua Buch.-Ham. ex D. Don, Thymeleaeceae, A. Worberg 026 (BONN), BG Bonn, FM179927 this study, – , FM179927 this study; Daucus carota L., Apiaceae, EMBL/GenBank, – , DQ898156 Ruhlman & al. (2006), – , DQ898156 Ruhlman & al. (2006); Dipentodon sinicus Dunn, Dipentodontaceae, EMBL/GenBank, – , AJ429397 Bremer & al. (2002), AJ430865 Bremer & al. (2002), – ; Durio zibethinus Murr., Bombacaceae, EMBL/GenBank, – , AY321188 Nyffeler & al. (2005), – , AY321188 Nyffeler & al. (2005); Erodium cicutarium (L.) L’Hér, Geraniaceae, T. Borsch 3483 (BONN), Germany, Eifel, AM396500 Worberg & al. (2007), AM397178 Worberg & al. (2007), AM396500 this study; Eucalyptus globulus Labill., Myrtaceae, EMBL/GenBank, – , NC_008115 Steane (2005), NC_008115 Steane (2005), NC_008115 Steane (2005); Euphorbia milii Desmoul., Euphorbiaceae, A. Worberg 002 (BONN), BG Bonn, FM179936 this study, – , FM179936 this study; Gerrardina foliosa Oliv., Gerrardinaceae, Balkwill & al. 11983, KEW, FM179924 this study, FM179535 this study, FM179924 this study; Glycine max Merr., Fabaceae, EMBL/GenBank, – , DQ317523 Saski & al. (2005), – , DQ317523 Saski & al. (2005); Gossypium hirsutum L., Malvaceae, EMBL/GenBank, – , NC_007944 Lee & al. (2006), NC_007944 Lee & al. (2006), NC_007944 Lee & al. (2006); Gunnera tinctoria (Molina) Mirb., Gunneraceae, N. Korotkova 50 (BONN), BG Bonn, AM396506 Worberg & al. (2007), –, AM396506 this study; Huertea cubensis Griseb., Tapisciaceae, R. Oviedo & al. s.n. (B, HAC), Cuba, FM179926 this study, FM179539 this study, FM179926 this study; Ilex aquifolium L., Aquifoliaceae, T. Borsch 3419 (BONN), BG Bonn, AF542607 Worberg & al. (2007), – , AF542607 this study; Impatiens noli-tangere L., Balsaminaceae, T. Borsch 3485 (BONN), BG Bonn, AF542608 Worberg & al. (2007), – , AF542608 this study; Juglans microcarpa Berland., Juglandaceae, EMBL/GenBank, – , AF118034 Stanford & al. (2000), – , AF118034 Stanford & al. (2000); Kandelia candel Druce, Rhizophoraceae, EMBL/GenBank, – , AF105090 Shi & al. (1999), – , – ; Larrea tridentata Coult., Zygophyllaceae, A. Worberg 012 (BONN), BG Bonn, AM396502 Worberg & al. (2007), AM397180 Worberg & al. (2007), AM396502 this study; Leea coccinea Planch., Leeaceae, T. Borsch 3418 (BONN), BG Bonn, AM396497 Worberg & al. (2007), – , AM396497 this study; Lobularia maritime L. (Desv.), Brassicaceae, EMBL/GenBank, – , AP009375 Hosouchi & al. (2007), AP009375 Hosouchi & al. (2007), AP009375 Hosouchi & al. (2007); Malus trilobata C.K. Schneid., Rosaceae, EMBL/ GenBank, – , DQ860463 Campbell & al. (2006), DQ863235 Campbell & al. (2006), DQ860463 Campbell & al. (2006); Melastoma septemnervium Jacq., Melastomataceae, T. Borsch s.n. (BONN), BG Bonn, FM179940 this study, –, FM179940 this study; Melia azedarach L., Meliaceae, A. Worberg 035 (BONN), BG Bonn, FM179921 this study, FM179536 this study, FM179921 this study; Muntingia calabura L., Muntingiaceae, EMBL/GenBank, –, –, AY328166 Yuan & al. (2003), – ; Muntingia calabura L., Muntingiaceae, D. Quandt s.n. (BONN), BG Bonn, FM179930 this study, – , FM179930 this study; Myrothamnus flabellifolia Welw., Myrothamnaceae, A. Worberg 011 (BONN), BG Bonn, AM396507 Worberg & al. (2007), – , AM396507 this study; Myrothamnus moschata Baill., Myrothamnaceae, E. Fischer & W. Höller s.n. (BONN), BG Bonn, AF542591 Worberg & al. (2007), – , AF542591 this study; Oenothera elata Kunth, Onagraceae, EMBL/GenBank, – , NC_002693 Hupfer & al. (2000), – , NC_002693 Hupfer & al. (2000); Oxalis hedysaroides Kunth, Oxalidaceae, N. Korotkova 55 (BONN), BG Bonn, FM179935 this study, – , FM179935 this study; Panax ginseng C.A. Mey., Araliaceae, EMBL/GenBank, – , NC_006290 Kim & Lee (2004), – , NC_006290 Kim & Lee (2004); Parnassia palustris L., Parnassiaceae, EMBL/GenBank, – , AY935911 Zhang & Simmons (2006), – , – ; Passiflora quadrangularis L., Passifloraceae, N. Korotkova 56 (BONN), BG Bonn, FM179937 this study, – , FM179937 this study; Perrottetia ovata Hemsl., Dipentodontaceae, EMBL/GenBank, – , AY935916 Zhang & Simmons (2006), AY935771 Zhang & Simmons (2006), – ; Perrottetia longistylis Rose, Dipentodontaceae, EMBL/GenBank, – , AY935915 Zhang & Simmons (2006), AY935770 Zhang & Simmons (2006), – ; Polygala californica Nutt. ex Torr. & A. Gray, Polygalaceae, EMBL/GenBank, – , AY386842 Wojciechowski & al. (2004), – , – ; Populus alba L., Salicaceae, EMBL/GenBank, – , AP008956 Okumura & al. (2005), – , AP008956 Okumura & al. (2005); Reseda lutea L., Resedaceae, A. Worberg 027 (BONN), BG Bonn, FM179932 this study, – , FM179932 this study; Salacia lehmbachii Loes., Celastraceae, T. Borsch 3549 (BONN), BG Bonn, AF542599 this study update, FM179534 this study, AF542599 this study; Schinus molle L., Anacardiaceae, EMBL/GenBank, – , – , AY640463 Yi & al. (2004), – ; Schinus molle L., Anacardiaceae, A. Worberg s.n. (BONN), BG Bonn, FM179923 this study, – , FM179923 this study; Stachyurus chinensis Franch., Stachyuraceae, A. Worberg s.n. (BONN), BG Bonn, AM396501 Worberg & al. (2007), – , AM396501 this study; Stachyurus chinensis Franch., Stachyuraceae, EMBL/GenBank, – , – , AB066335, Ohi & al. (2003), – ; Spiraea thunbergii Siebold ex Blume, Rosaceae, A. Worberg s.n. (BONN), BG Bonn, FM179934 this study, – , FM179934 this study; Tapiscia sinensis Oliver, Tapisciaceae, T. Borsch s.n. (BONN), BG Bonn, FM179925 this study, FM179541 this study, FM179925 this study; Tropaeolum majus L., Tropaeolaceae, A. Worberg s.n. (BONN), BG Bonn, FM179931 this study, – , FM179931 this study; Urtica cannabina L., Urticaceae, A. Worberg s.n. (BONN), BG Bonn, FM179933 this study, – , FM179933 this study; Vasconcellea parviflora A. DC., Caricaceae, A. Worberg 038 (BONN), BG Bonn, FM179928 this study, FM179537 this study, FM179928 this study; Vitis riparia A. Gray, Vitaceae, T. Borsch 3458 (BONN), BG Bonn, AF542593 Worberg & al. (2007), – , AF542593 this study. 478 TAXON 58 (2) • May 2009: E1 Worberg & al. • Phylogenetic position of Huerteales Appendix 2. List of primers used in this study. Primers used for the amplification of matK along with their sequences and the taxa for which they were designed. References are given for primers that were not designed for this study. Primer name MG15 MG1 trnKFbryo trnK3914Fdi trnK2R psbA-R NYmatK480F ARmatK660R ARmatK1200F ARmatK2400R ASmatK460F ACmatK873R ACmatK100F ACmatK1400R ACmatK634R ACmatK1746F ACmatK2231R ACmatK499F PImatK1060F PImatK1480F ROSmatK655R ROSmatK530F DIDYmatK1035R DIDYmatK1107F DIDYmatK570F EUPTmatK1006F NYmatK980R Sequence ATC TGG GTT GCT AAC TCA ATG AAC TAG TCG GAT GGA GTA GAT GGG TTG CTA ACT CAA TGG TAG AG GGG GTT GCT AAC TCA ACG G AAC TAG TCG GAT GGA GTA G CGC GTC TCT CTA AAA TTG CAG TCA T CAT CTG GAA ATC TTG STT C AYG GAT TCG CAT TCA TA TTC CAA AGT CAA AAG AGC G ATT TTC TAG CAT TTG ACT CC TAC TTC CCT TTT T(ACT)G AGG ATA TAC TCC TGA AAG AGA AGT GG CTC GAC TGT ATC AAC AGA ATC GGA TTC GTA TTC ACA TAC A ATA GGA ACA AGA ATA ATC T CTT CAA AAG GGA CAT CTC TTC TG TCT TCC AAA AAT TCT GAA CCT CCT CKT CTT TGC ATT TAT TAC G ACT T(AG)T GGT CTC AAC (CT)G TCG TAA ACA (CT)AA AAG TAC GGA TTC GTA TTC ACA TAC AT AGA TGC CTC TTC TTT GC CGA ACT CGT AAA GAC TCG A CAA TTA CTC CTC TGA TTG GAT C CGA GTA TCA GAA CTG GAG GGC TAT CTT TCA AGT GTA CG GGT TAG AAT CAT TAG CRG Taxa Angiosperms Angiosperms Land plants Angiosperms Angiosperms Angiosperms Angiosperms Angiosperms Angiosperms Angiosperms Angiosperms Eudicots Proteaceae Aquifoliaceae Basal eudicots Aextoxicaceae Cercidiphyllaceae Stachyuraceae Trochodendraceae Papaveraceae Rosids Rosids Didymelaceae Didymelaceae Didymelaceae Eupteleaceae Rosids Reference Liang & Hilu (1996) Liang & Hilu (1996) Wicke & Quandt (in press) Johnson & Soltis (1995) Johnson & Soltis (1995) Steele & Vilgalys (1994) Borsch (2000) Wanke & al. (2006a) Wanke & al. (2006a) Wanke & al. (2006a) Wanke & al. (2006a) Müller & Borsch (2005) Müller & Borsch (2005) Müller & Borsch (2005) Müller & Borsch (2005) Müller & Borsch (2005) Müller & Borsch (2005) Müller & Borsch (2005) Wanke & al. (2006b) Wanke & al. (2006b) Worberg & al. (2007) Worberg & al. (2007) Worberg & al. (2007) Worberg & al. (2007) Worberg & al. (2007) Worberg & al. (2007) Worberg & al. (2007) Primers used for the amplification of trnK along with their sequences and the taxa for which they were designed. References are given for primers that were not designed for this study. Primer name Sequence trnK-f bryo trnK-2R MG15 psbAR GGG AAC ATC CGC TTG TAG TGG GTC CTA TCG GTT TCT ACT GAT GCT CTA CAA GGA AAC AAA TGG GTA TCA TTG TAG AG G ATG CAG TCA T Taxa Reference Angiosperms Angiosperms Angiosperms Angiosperms Wicke & Quandt (in press) Johnson & Soltis (1995) Liang & Hilu (1996) Steele & Vilgalys (1994) Primers used for the amplification of trnL-F region along with their sequences and the taxa for which they were designed. References are given for primers that were not designed for this study. Primer name Sequence trnTc trnTf trnTd trnL110R trnL460F CGA ATT GGG GAT GAG AAT TGA GAT TTG AAT CGG ACT AGA GCT AAA TAG GGT GGG CAG GAT ACG GAC ACT GAT AGA CTA ACG TGA TGC GTC CG AG AC CC C Taxa Reference Angiosperms Angiosperms Angiosperms Angiosperms Angiosperms Taberlet & al. (1991) Taberlet & al. (1991) Taberlet & al. (1991) Borsch & al. (2003) Worberg & al. (2007) E1