Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2006 May;97(5):705–714. doi: 10.1093/aob/mcl029

Phylogeny of Rhaponticum (Asteraceae, Cardueae–Centaureinae) and Related Genera Inferred from Nuclear and Chloroplast DNA Sequence Data: Taxonomic and Biogeographic Implications

ORIANE HIDALGO 1,*, NÚRIA GARCIA-JACAS 1, TERESA GARNATJE 1, ALFONSO SUSANNA 1
PMCID: PMC2803413  PMID: 16495316

Abstract

• Background and Aims The precise generic delimitation of the Rhaponticum group is not totally resolved. The lack of knowledge of the relationships between the basal genera of Centaureinae could imply that genera whose position is as yet unresolved could belong to the Rhaponticum group. On the other hand, the affinities among the genera that are considered as members of this group are not well known. The aim of the study is to contribute to the phylogenetic and generic delineation of the Rhaponticum group on the basis of molecular data.

• Methods Parsimony and Bayesian analyses of the combined sequences of one plastid (trnL-trnF) and two nuclear (ITS region and ETS) molecular markers were carried out. The results of these analyses are discussed in the light of the biogeographic history.

• Key Results The Rhaponticum group appears as monophyletic, and closely related to the genus Klasea. The results confirm the preliminary generic delimitation of the Rhaponticum group, with the new incorporation of the genus Centaurothamnus. Ochrocephala is supported as a separate genus from Rhaponticum and, contrary to this, Acroptilon and Leuzea appear as merged into the genus Rhaponticum. Several nomenclatural rearrangements are made in Klasea and Rhaponticum.

• Conclusions The new molecular evidence is consistent with the morphological and karyological data, and suggests particularly coherent biogeographic routes of migration and speciation processes for the genus Rhaponticum. The biogeographic inference proposes a Near East and/or Caucasian origin for the genus. Furthermore, representatives of Rhaponticum could have reached Europe in two different ways: (1) expansion across central Asia to eastern Europe, and (2) expansion through the Near East, North Africa and then to the Iberian Peninsula and the Alps.

Keywords: Acroptilon, biogeography, Callicephalus, Centaurothamnus, ETS, ITS, molecular phylogeny, Myopordon, Ochrocephala, Oligochaeta, Rhaponticum, trnL-trnF

INTRODUCTION

One of the main problems which persists in the subtribe Centaureinae (Asteraceae, Cardueae) is that the phylogenetic relationships between the early branching genera are unresolved. Monographers (Dittrich, 1977; Wagenitz and Hellwig, 1996; Garcia-Jacas et al., 2001; Hellwig, 2004; Susanna and Garcia-Jacas, 2006) have described informal groups of genera. One of these is the Rhaponticum group which comprises about seven genera and approx. 40 species. Apart from the genus Rhaponticum Vaill. (=Stemmacantha Cass.; cf. Greuter et al., 2005) of approx. 25 species, it includes Acroptilon Cass. (two species), Callicephalus C. A. Mey. (one species), Leuzea DC. (one species), Myopordon Boiss. (five species), Ochrocephala Dittrich (one species) and Oligochaeta (DC.) K. Koch. (four species).

The classic morphological approach limited itself to associate Rhaponticum and Acroptilon (Dittrich, 1977), and pointed out generic delimitation problems: Holub (1973) concluded that Leuzea and Rhaponticum should be merged, and Dittrich (1983) that Rhaponticum imatongensis (Phillipson) Soják should be segregated to constitute a new monotypic genus, Ochrocephala. Later, the molecular approach allowed the addition of Callicephalus, Oligochaeta (Garcia-Jacas et al., 2001) and Myopordon (Susanna et al., 2006).

These genera show the symplesiomorphic characters common to the basal genera of the Centaureinae (absence of radiant peripheral florets, lack of crystals in the phyllaries, basal hilum, no bolster cells, absence of hairs on the achene and pollen of the Serratula type), and some morphological traits which characterize the group: (a) a peculiar type of involucral bract with a big, soft scarious, entire or lacerate and usually silvery-white appendage (an exclusive character of the group which, unfortunately, not all the species show); (b) dimorphic achenes (the outer dorsiventrally compressed and the inner laterally compressed); and (c) the double pappus typical of all the Centaureinae, but with the peculiarity that the inner bristles are wider and longer than the outer.

The geographical distributions, environmental requirements and life cycles are very diverse in Rhaponticum and related genera. They are naturally distributed in North Africa (including the Canary Islands), temperate Eurasia, Siberia and the Far East, Caucasus, central and eastern Asia and eastern Australia. They grow in deserts or mountains, and are either widely distributed or narrow endemics. They can be perennial or annual, and their habit is shrubby, or hemicryptophyte from 10 cm to >1 m in height, or acaulescent. Several species are endangered to the verge of extinction, but one taxon, Acroptilon, is considered to be an invasive weed in America and Australia. The Rhaponticum group includes the only species of Centaureinae indigenous to Australia, Rhaponticum australe (Gaudich.) Soskov (Wagenitz and Hellwig, 1996). Some representatives of the group show medicinal properties which were already known in Roman culture (Plinius, 77), and various species are being marketed due to their anabolic and adaptogenic properties.

The main goals of the present study in establishing a combined molecular phylogeny were to (a) elucidate the relationships between the basal groups of Centaureinae, with the purpose of determining the taxa most closely related to the Rhaponticum group; (b) verify the generic delimitation of the Rhaponticum group and the relationships between its genera; and (c) link the findings to the group's biogeographic history.

MATERIALS AND METHODS

Plant material

The sampling includes representatives of all the genera of the Rhaponticum group, and all the species of the genus Rhaponticum (except R. namanganicum Iljin). The outgroups have been selected according to previous works by Garcia-Jacas et al. (2001) and Susanna et al. (2006) to represent most of the basal Centaureinae which could have phylogenetic affinities with the study group: Centaurothamnus Wagenitz & Dittrich, Cheirolophus Cass., Klasea Cass., Plagiobasis Schrenk, Psephellus Cass., Rhaponticoides Vaill., Serratula L. and Stizolophus Cass. The purpose of representing numerous outgroups is to be able, without forcing the topology, to test how good is the assignation of a taxon as an outgroup or as an ingroup, and to define the taxa which are most closely related to the Rhaponticum group. Both previously published and the 111 new sequences (31 ITS, 47 ETS, 33 trnL-trnF) were used in the analyses. The origin of the samples and GenBank sequence accession numbers are given in Table 1.

Table 1.

Origin of the materials, herbaria where the vouchers are deposited and GenBank accession numbers (new sequences are indicated by bold type)

Species Voucher ITS accession ETS accession trnL-F accession
Acroptilon australe Iljin Mongolia: V. Grubov 301 et al. (LE) DQ310942 DQ310990 DQ310909
Acroptilon repens (L.) DC. Uzbekistan: Susanna 2046 et al. (BC) AY826223 DQ310989 AY772268
Callicephalus nitens (M. Bieb. ex Willd.) C. A. Mey. Armenia: Susanna 1578 et al. (BC) AY826237 DQ310972 AY772281
Centaurothamnus maximus Wagenitz & Dittrich Yemen: Molero s. n. (BC) AY826259 DQ310971 AY772301
Cheirolophus mauritanicus (Font Quer) Susanna Morocco: Romo 4617 et al. (BC) AY826261 DQ131087 AY772303
Cheirolophus teydis (C. Sm.) G. López spain: Susanna 1429 et al. (BC) AY826262 DQ131092 AY772304
Klasea algida (Iljin) Hidalgo Tajikistan: Susanna 2558 &Romashchenko (BC) DQ310929 DQ310968 DQ310895
Klasea biebersteiniana (Iljin ex Grossh.) Hidalgo Armenia: Susanna 1493 et al. (BC) DQ310928 DQ310967 DQ310894
Klasea cerinthifolia (Sm.) Greuter & Wagenitz Iran: Susanna 1700 et al. (BC) DQ310924 DQ310963 DQ310890
Klasea chartacea (C. Winkl.) L. Martins Tajikistan: Susanna 2467 &Romashchenko (BC) DQ310927 DQ310966 DQ310893
Klasea coriacea (Fisch. & C. A. Mey. ex DC.) Holub Armenia: Susanna 1558 et al. (BC) DQ310926 DQ310965 DQ310892
Klasea grandifolia (P. H. Davis) Greuter & Wagenitz Iran: Susanna 1709 et al. (BC) DQ310930 DQ310969 DQ310896
Klasea kuzhistanica (Mozaffarian) Mozaffarian Iran: Mozaffarian 70181 (TARI) DQ310925 DQ310964 DQ310891
Klasea serratuloides (DC.) Greuter & Wagenitz Armenia: Susanna 1569 et al. (BC) AY826295 DQ310962 AY772334
Leuzea berardioides Batt. Morocco:Hidalgo & Romo 12749 (BC) DQ310948 DQ310998 DQ310915
Leuzea conifera (L.) DC. Spain: Font s. n. (BC) AY826298 DQ310996 AY772337
Leuzea fontqueri Sauvage Morocco:Hidalgo & Romo 12621 (BC) DQ310947 DQ310997 DQ310914
Myopordon aucheri Boiss. Iran: Carls s.n. (W) AY826299 DQ310977 AY772338
Myopordon hyrcanum (Bornm.) Wagenitz Iran: Koelz 16395 (W) AY826300 DQ310975 AY772339
Myopordon persicum Boiss. Iran: Remandieri s.n. (W) AY826301 DQ310976 DQ310898
Ochrocephala imatongensis (Phillipson) Dittrich Ethiopia:Fantahun Simon 9163 et al. (K) DQ310931 DQ310970 DQ310897
Oligochaeta divaricata (Fisch. & C. A. Mey.) K. Koch Armenia: Susanna 1583 et al. (BC) AY826306 DQ310973 AY772344
Oligochaeta minima (Boiss.) Briq. Uzbekistan: Botanical Garden of Tashrent (BC) AY826307 DQ310974 AY772345
Plagiobasis centauroides Schrenk Kazakhstan: Susanna 2130 et al. (BC) AY826312 DQ310956 DQ310887
Psephellus persicus (DC.) Wagenitz Iran: Susanna 1716 et al. (BC) AY826316 DQ310957 AY772352
Psephellus pulcherrimus (Willd.) Wagenitz Armenia:Susanna 1492 et al. (BC) AY826317 DQ310958 AY772353
Rhaponticoides hajastana (Tzvelev) M. V. Agab. & Greuter Armenia: Susanna 1587 et al. (BC) DQ310922 DQ310959 DQ310888
Rhaponticoides iconiensis (Hub.-Mor.) M. V. Agab. & Greuter Turkey: Ertugrul 1761 (BC) DQ310923 DQ310960 DQ310889
Rhaponticum acaule (L.) DC. Algeria: J. M. Montserrat 2331 et al. (BC) AY826334 DQ310995 AY772369
Rhaponticum aulieatense Iljin Kyrgyzstan: Sheremetova & Lazkov (LE) DQ310936 DQ310983 DQ310903
Rhaponticum australe (Gaudich.) Soskov Australia: Funk 12203 (BC) AY826335 DQ310978 AY772370
Rhaponticum canariense DC. Spain:Carqué Álamo s. n. (BC) DQ310954 DQ311004 DQ310921
Rhaponticum carthamoides (Willd.) Iljin Russia:Botanical Garden of Sibiricus Centralis, Novosibirsk 2003/2004–1062 (BC) DQ310933 DQ310980 DQ310900
Rhaponticum cossonianum (Ball) Greuter Morocco: Gómiz 17-IV-2003 (BC) DQ310949 DQ310999 DQ310916
Rhaponticum cynaroides Less. Spain: Hidalgo 504 et al. (BC) DQ310946 DQ310994 DQ310913
Rhaponticum exaltatum (Willk.) Greuter Spain: Garcia-Jacas & Susanna 2434 (BC) DQ310953 DQ311003 DQ310920
Rhaponticum heleniifolium Godr. & Gren. Botanical Garden of Minsk, Bielorussia 1/303–2000 (BC) DQ310945 DQ310993 DQ310912
Rhaponticum insigne (Boiss.) Wagenitz Iran:Archibald 2034 (K) DQ310944 DQ310992 DQ310911
Rhaponticum integrifolium C. Winkl. Tajikistan: Makhmetov & R. Kamelin 344 (LE) DQ310934 DQ310981 DQ310901
Rhapontikum karatavicum Iljin Kazakhstan: Kamelin (LE) DQ310940 DQ310987 DQ310907
Rhaponticum longifolium (Hoffmanns. & Link) Dittrich Portugal: Garcia-Jacas & Susanna 2436 (BC) DQ310950 DQ311000 DQ310917
Rhaponticum lyratum C. Winkl. ex Iljin Tajikistan:Konnov16-VII-1965 (LE) DQ310935 DQ310982 DQ310902
Rhaponticum nanum Lipsky Tajikistan:Kochkariova 5834 (DUSH) DQ310939 DQ310986 DQ310906
Rhaponticum nanum Lipsky ssp. pellucidum (Rech. f.) Dittrich Afghanistan:Renz 120 (W) DQ310938 DQ310985 DQ310905
Rhaponticum nitidum Fisch. Russia:Kalibernova 5676 et al. (LE) DQ310937 DQ310984 DQ310904
Rhaponticum pulchrum Fisch. & C. A. Meyer Russia:Popova 326 et al. (LE) DQ310943 DQ310991 DQ310910
Rhaponticum scariosum Lam. Slovenia:Botanical Garden of Universitatis Labacensis Slovenia 1994–180 (BC) DQ310952 DQ311002 DQ310919
Botanical Garden of Minsk, Bielorussia 1/304–2000 (BC) DQ310951 DQ311001 DQ310918
Rhaponticum serratuloides (Georgi) Bobrov Botanical Garden of Cluj-Napoca, Romania 636-2001 (BC) DQ310941 DQ310988 DQ310908
Rhaponticum uniflorum (L.) DC. Mongolia: Vallès 13–2003 (BC) DQ310932 DQ310979 DQ310899
Serratula coronata L. Botanical Garden of Vienna, Austria (BC) AY826327 DQ310961 AY772362
Stizolophus coronopifolius Cass. Turkey: Ilarslan 4303 (ANK) AY826337 DQ310955 AY013516
Open in a new tab

DNA extraction, amplification and sequencing

Total genomic DNA was extracted following the miniprep procedure of Doyle and Doyle (1987) as modified by Soltis et al. (1991) and Cullings (1992) from herbarium material, silica gel-dried leaves collected in the field, or fresh leaves of plants cultivated in the Botanic Institute of Barcelona. DNA of old herbarium material was extracted using the DNeasy Plant Mini Kit (Qiagen Inc., Valencia, CA, USA).

nrDNA ITS region strategies

ITS1, 5·8S gene and ITS2 (the ITS region) were amplified and sequenced together with 1406F (Nickrent et al., 1994), ITS1 (White et al., 1990) and 17SE (Sun et al., 1994) as forward primers, and ITS4 (White et al., 1990) and 26SE (Sun et al., 1994) as reverse primers, referring to the protocol described in Soltis and Kuzoff (1993). PCR products were purified using the QIAquick PCR Purification Kit (Qiagen Inc.). Direct sequencing of the amplified DNA segments was performed using the BigDye Terminator Cycle Sequencing v3·1 (PE Biosystems, Foster City, CA, USA). Nucleotide sequencing was carried out at the Serveis Cientificotècnics of the University of Barcelona on an ABI PRISM 3700 DNA analyser (PE Biosystems).

nrDNA ETS region strategies

The ETS region was amplified and sequenced with ETS1f as forward primer and 18S-2L as reverse primer (Linder et al., 2000), referring to the PCR procedure described in the same publication. Purification and direct sequencing of the amplified DNA segments were performed as for the ITS region.

cpDNA trnL-trnF region strategies

The trnL-trnF region includes the trnL intron, the 3′ trnL (UAA) exon, and the intergenic spacer between trnL (UAA) and trnF (GAA), which were amplified and sequenced together. The universal primers trnL-c, forward, and trnL-f, reverse (Taberlet et al., 1991) were used to amplify and sequence the trnL-F region. For old material, the region was amplified and sequenced in two parts using the two precedent primers and the trnL-e, forward, and trnL-d, reverse, of the same author. The PCR procedure includes a warm start at 95 °C for 1 min 35 s, followed by 80 °C during which the polymerase (Ecotaq, Ecogen S.R.L., Barcelona, Spain) is added, and 34 cycles of 1 min denaturation at 93 °C, 1 min annealing at 58 °C, 1 min extension at 72 °C, and a final 10 min extension at 72 °C. Purification and direct sequencing of the amplified DNA segments was performed as for the ITS region.

Phylogenetic analyses

Nucleotide sequences were edited with Chromas 1·56 (Technelysium Pty, Tewantin, Australia). DNA sequences were aligned visually by sequential pairwise comparison (Swofford and Olsen, 1990).

Parsimony analysis

Parsimony analysis involved heuristic searches conducted with PAUP version 4·0b10 (Swofford, 1999) using tree bisection recognition (TBR) branch swapping with character states specified as unordered and unweighted. All most-parsimonious trees (MPT) were saved. To locate islands of most-parsimonious trees (Maddison, 1991), 100 replicates were performed with random taxon addition, also with TBR branch swapping. Trees lengths, consistency index (CI) and retention index (RI) are always given excluding uninformative characters. Bootstrap (BS; Felsenstein, 1985) was carried out to obtain support estimates of the nodes of the trees selected. Bootstrap analysis was performed using 1000 replicates of heuristic search with the default options. ACCTRAN (accelerated transformation) character-state optimization was used for all trees illustrated. To conserve the phylogenetic information of insertions–deletions and, at the same time avoiding over-estimation of lengthy indels, ‘missing data’ were used and the indels coded as presence–absence characters added to the end of the matrix.

Bayesian analysis

Data sets were analysed using Mr Modeltest 2.2 (Nylander, 2004) to determine the sequence evolution model that best described the present data. This model was used to perform a Bayesian analysis using the program Mr. Bayes 3.1.1 (http://morphbank.ebc.uu.se/mrbayes/; Huelsenbeck et al., 2001). Four Markov chains were run simultaneously for 1000  000 generations, and these were sampled every 100 generations. Data from the first 1000 generations were discarded as the ‘burn-in’ period, after confirming that likelihood values had stabilized prior to the 1000th generation. The 50 % majority rule consensus phylogeny and posterior probability (PP) of nodes were calculated from the remaining sample.

Biogeographic distributions

The distributions were mapped on the tree using the Farris double-pass method (Farris, 1970) which provides the hypothesized distributions of the deep branches and nodes. Each taxon branch and internode was coloured as to its distribution using Adobe Illustrator.

RESULTS

Some of the ETS sequences have repeats in the 5′ end of the region, as found in other groups (Baldwin and Marcos, 1998; Linder et al., 2000). In most cases, these repetitions constitute autapomorphic events or characterize a group without alignment or homology problems. Conversely, the majority of Klasea species present a region with a different number of repetitions. Because of these repetitions, it has not been possible to obtain the entire sequence for some species; for the others, their alignment was impeded by the difficulty of establishing the homology of the repetitions. For these reasons, the unalignable sequence area, exclusive of Klasea, was removed from the matrix.

Parsimony analysis

Four indel characters were included in the data matrix. The results from the combined ITS, ETS and trnL-trnF parsimony analysis are given in Table 2.

Table 2.

Results from the combined ITS, ETS and trnL-trnF parsimony analysis (the consistency and homoplasy indexes are calculated excluding uninformative characters)

Data set Combined analyses
Total characters 2950
(ITS: 669, ETS: 1354, trnL-trnF: 927)
Informative substitutions 549
(ITS: 149, ETS: 375, trnL-trnF: 25)
Number of MPTs 558
Number of steps 1582
Consistency index (CI) 0.48
Retention index (RI) 0.72
Mean pairwise distances, ingroup (%) From 0 (Acroptilon australe/A. repens; Leuzea conifera/L. fontqueri) to 37 (Acroptilon australe, A. repens/Callicephalus nitens
Open in a new tab

Bayesian analysis

The GTR+G+I model was found to be the most efficient model for optimizing sequence evolution of the data set.

The two methods of phylogeny reconstruction lead to congruent results (there is no discordance for strongly supported branches) that lead to only the tree obtained with the Bayesian inference, shown in Fig. 1, being presented. This tree indicates both the bootstrap values (calculated by the parsimony analysis), the posterior probability (calculated by the Bayesian inference) and biogeography mapping.

Fig. 1.

Fig. 1.

Open in a new tab

Majority-rule consensus tree based on Bayesian analysis. Numbers above branches indicate Bayesian clade-credibility values (posterior probability); numbers below branches indicate parsimony bootstrap percentages. Subgenera of Rhaponticum: CES, Cestrinus; ELE, Eleutherochaetum; FOR, Fornicium; LEU, Leuzea; RHA, Rhaponticina; RHL, Rhaponticella; STE, Stemmacantha.

DISCUSSION

This new molecular phylogeny makes it possible to hypothesize regarding the evolutionary history of the Rhaponticum group. Among all the genera selected to constitute the outgroup, the genus Klasea appears as the most closely related to the Rhaponticum group (PP = 97 %, BS = 81 %; Fig. 1). Furthermore, the results contradict the election of Centaurothamnus as outgroup, and suggest that it should be included in the Rhaponticum group. On the other hand, all the taxa previously considered as members of the Rhaponticum group appear in the ingroup (PP = 98 %; Fig. 1).

Klasea is sister to the Rhaponticum group

Klasea (≡ Serratula section Klasea) comprises approx. 65 species distributed from the western Mediterranean to China and the western part of the Himalayas. The phylogenetic distance within the genus Serratula between Serratula sensu stricto (S. coronata L. and S. tinctoria) and section Klasea has been demonstrated by different authors, on morphological (Wagenitz and Hellwig, 1996) and on molecular bases (Martins and Hellwig, 2005). Therein, those authors were in accordance with Cassini (1825) who placed Klasea as a genus distinct from Serratula. The present results confirm these findings and clearly define Klasea as closely related to the Rhaponticum group (PP=97 %, BS=81 %; Fig. 1). This proximity was previously suggested by Lessing (1832), who considered Klasea Cass. as a subgenus of Rhaponticum. Moreover, the observation of the achene morphology has brought to light for the first time the fact that Klasea shows heterocarpy like the Rhaponticum group. The Klasea species constitute a monophyletic group (PP = 99 %, BS = 100 %; Fig. 1). Among them, Serratula algida Iljin, S. biebersteiniana (Iljin ex Grossh.) Takht. and S. kuzhistanica (Mozaffarian) Mozaffarian have not been recombined as Klasea until now; this has lead to the establishment of new nomenclatural combinations as detailed in the Appendix.

The monophyly of the Rhaponticum group is supported by the Bayesian inference (PP = 98 %; Fig. 1), but the parsimony analysis shows a polytomy consisting of Klasea, the Centaurothamnus plus Ochrocephala clade, and the rest of the genera. Even though the parsimony analysis is equivocal in the placement of Klasea as sister to the Rhaponticum group, this hypothesis is, nonetheless, supported both by karyological and morphological data. Representatives of the genus Klasea have a base chromosome number of x = 15 (Löve and Löve, 1961; Cantó, 1982, 1984; Garcia-Jacas et al., 1998a, b), while Centaurothamnus and Callicephalus have x = 14 [Wagenitz et al. (1982) for the former; Chouksanova et al. (1968), Hellwig (1994) and Garcia-Jacas et al. (1998a) for the latter] and Acroptilon, Leuzea and most of the Rhaponticum species present x = 13 (Wagenitz and Hellwig, 1996). Descending dysploidy being one of the main evolutionary mechanisms in plants and in particular in the Centaureinae (Fernández Casas and Susanna, 1986; Garcia-Jacas et al., 1996; Vilatersana et al., 2000), this could suggest that x = 15 is more primitive than x = 14. Then, in the case of the present focus of study, this could suggest that Klasea should be sister to the Rhaponticum group. Furthermore, although Klasea shows heterocarpy like the Rhaponticum group, it does not exhibit the two other morphological apomorphies present in the other two clades, namely the typical involucral bracts (present in Ochrocephala) and the characteristic pappus (present in Centaurothamnus and Ochrocephala).

Centaurothamnus and Ochrocephala

The present results confirm that Ochrocephala is more appropriately treated as a monotypic genus (Dittrich, 1983), rather than considered as Rhaponticum imatongensis. Furthermore, this study allows the systematic position of the genus Centaurothamnus to be defined for the first time. This genus was placed with the genera of ‘uncertain position’ by Wagenitz and Hellwig (1996), Garcia-Jacas et al. (2001) and Hellwig (2004). With the new molecular evidence, Centaurothamnus appears to be closely related to the genus Ochrocephala (PP = 99 %, BS = 100 %; Fig. 1). This result is not surprising, because these two monotypic genera are geographically very close: Centaurothamnus maximus Wagenitz & Dittrich grows in south-western Arabia, in Yemen, and Ochrocephala imatongensis (Phillipson) Dittrich in eastern Africa (Ethiopia, Sudan and Congo). Morphologically, these two taxa share the same shrubby habit, an exclusive trait of the group. The shrubby port is uncommon within the Centaureinae, and outside Centaurothamnus and Ochrocephala it is only know from the genus Centaurodendron Johow, Centaurea ptosimopappa Hayek and the genus Cheirolophus. It corresponds, probably, to a secondary adaptation, this phenomenon being particularly evident for the insular taxa such as Centaurodendron and Cheirolophus (Böhle et al., 1996; T. Garnatje, unpubl. res.). Centaurothamnus and Ochrocephala are genetically and morphologically distinguished: their molecular divergence for the three markers considered is 13·8 %; Ochrocephala shows the typical involucral bract appendages of the Rhaponticum group, while these are not present in Centaurothamnus. A new question introduced by these results is: Should Centaurothamnus and Ochrocephala be more appropriately maintained as distinct genera or should they should be merged to constitute a single genus?

Callicephalus nitens

The genus Callicephalus includes a single species, Callicephalus nitens (M. Bieb. ex Willd.) C. A. Mey., from the middle and low mountains of the Caucasus, central Asia and the Near East. It appeared within the Rhaponticum group in the molecular analysis of Garcia-Jacas et al. (2001), but with weak statistical support. The present results strongly support the fact that Callicephalus belongs to the group of genera related to Rhaponticum (PP = 99 %, BS = 86 %; Fig. 1). Because of its annual nature, this species might show increased mutation rates, which could have induced an anomalous result in the parsimony analysis. However, the Bayesian inference method, less affected by long branch attraction, leads to an identical result. Furthermore, this hypothesis is reinforced by morphological features such as the structure of the inner pappus or the tuberculate pericarp (Garcia-Jacas et al., 2001). Callicephalus has no closely related taxa and appears as isolated in the phylogeny. Thus, this genus may be one of the numerous ‘relict’ taxa that grow in the Caucasus. The abundance of relict and endemic plant species in this area seems largely due to the fact that it was spared glaciation during the most recent ice ages.

The rest of the ingroup belongs to a strongly supported clade (PP = 99 %, BS = 100 %, Fig. 1) which includes the genera Acroptilon, Leuzea, Myopordon, Oligochaeta and Rhaponticum, placed in three different groups. The relationships between these three groups are not resolved either in the parsimony or the Bayesian inference.

Oligochaeta and Myopordon

Myopordon, a small genus with five perennial species from the Near East which had been placed in the Carduinae, and Oligochaeta, another genus composed of four annual species from the Near East, Caucasus, Afghanistan and India, and related to Rhaponticum, have apparently nothing in common. However, evidence that they are closely related was provided by the molecular study of Susanna et al. (2006), and it is also confirmed in this analysis (PP = 99 %, BS = 96 %; Fig. 1), whose sampling of the Rhaponticum group species is much more complete. In spite of the morphological review of Myopordon by Wagenitz (1958), several questions remain open. One consists of the generic delimitation of Myopordon and Oligochaeta: the present analyses support the monophyly of Oligochaeta (PP = 100 %, BS = 100 %; Fig. 1) but not that of Myopordon. More studies are necessary to verify whether Myopordon and Oligochaeta are independent taxa or whether they should be merged. Other questions concern the morphological traits of this clade, focusing especially on their palynological characteristics, which are baffling. While all the species of the Rhaponticum group show a Serratula-type pollen, Oligochaeta presents a reduced form of Serratula-type pollen grain (Martín and Garcia-Jacas, 2000), and Myopordon exhibits three different pollen types: Jacea, Centaurea scabiosa and Serratula (Wagenitz, 1958). It is perplexing that these two specialized and divergent taxa can be so narrowly related to Rhaponticum up to the point that our three molecular markers are not able to segregate them.

The genus Rhaponticum

The genus Rhaponticum does not appear as monophyletic in the phylogeny established for the following two reasons. (1) Its segregation from the clade of Myopordon plus Oligochaeta is not statistically supported (Fig. 1). Fortunately, the resolution within the genus Rhaponticum is better, and it shows two strongly supported clades, one ‘oriental’ mostly composed of central Asian species (PP = 99 %, BS = 100 %; Fig. 1) and the other ‘occidental’ including predominantly species from North Africa and Europe (PP = 99 %, BS = 92 %; Fig. 1). (2) The genera Acroptilon and Leuzea are firmly nested in the genus Rhaponticum, the first in the oriental clade, and the second in the occidental clade (Fig. 1), which leads to the paraphyly of Rhaponticum in its present circumscription. This implies some nomenclatural changes to reconcile the delimitation of the genus with this new evidence. Other evidence for the placement of Acroptilon and Leuzea in the genus Rhaponticum is that the three taxa share the same chromosome number x = 13, which is uncommon within the Centaureinae.

The comparison between the more compehensive infrageneric classification of Rhaponticum (Holub, 1973) and the molecular phylogeny shows numerous incongruities (Fig. 1). Only two of the seven subgenera described are natural groups: the subgenus Rhaponticella (Soskov) Holub (PP = 99 %; Fig. 1) and the subgenus Leuzea DC. (PP = 99 %, BS = 99 %; Fig. 1). The present results suggest that the more appropriate division of the genus Rhaponticum would be into only two subgenera, these corresponding to the oriental and the occidental clades, but it has not been possible to detect any character that defines either group on morphological grounds.

The Rhaponticum oriental clade

The first clade (PP = 99 %, BS = 100 %; Fig. 1) consists mostly of central Asian species, but includes species from middle and eastern Asia, Australia and eastern Europe. These species have relatively restricted areas of distribution, except for two groups of taxa that have wider areas.

One group extends from western to eastern Europe, and comprises Rhaponticum serratuloides (Georgi) Bobrov and Acroptilon (PP = 99 %, BS = 100 %; Fig. 1). The incorporation of Acroptilon in Rhaponticum had never been mentioned before, and necessitates the new nomenclatural combinations detailed in the Appendix. The most recent classifications do not recognize Acroptilon australe Iljin as a species separate from A. repens (L.) DC. (Hellwig, 2004; Susanna and Garcia-Jacas, 2006), while on the contrary, Soskov (2001) considers them to be two well-defined species. It is not possible from the present results to come to a verdict, and more studies are necessary to clarify the status of A. repens. This is the reason why we have preferred to abstain from making a new combination for this taxon. Acroptilon is considered to be an invasive weed in America and Australia, where it adversely affects agronomic harvests. It is aggressively competitive and exhibits allelopathic effects. It differs from the other species of the group, most of them endemics restricted to unfavourable environments where the competition with other species is less notable, as for example the mountain screes. The structure of capitula, achenes and the type of ramification are basically the same as Rhaponticum, but Acroptilon shows secondary adaptations due to its colonizing strategy: it is a hemicryptophyte like Rhaponticum but, instead of presenting few stems weakly or not ramified, this species generates numerous strongly branched stems in spring, due to its extensive root and rhizome system. Therefore, vegetative multiplication is favoured, although it also produces numerous capitula and achenes.

The other group extends from central to eastern Asia and Australia and is composed of Rhaponticum australe and R. uniflorum (L.) DC. The close relationship between these two species (PP = 100 %, BS = 97 %; Fig. 1) is a logical result, considering that R. uniflorum is the only species of the genus which has reached eastern Asia. From a geographic point of view, this was the best candidate to be sister to the Australian species. The fact that R. australe is the only species of Centaureinae indigenous to Australia is surprising because nothing explains such a long dispersal distance of the achenes of a Rhaponticum species. This lead Susanna and Garcia-Jacas (2006) to hypothesize that the species was doubtfully native in Australia. There is a considerable genetic divergence between R. uniflorum and R. australe for the three regions studied (8·7 %), and this means there is no possibility of a recent introduction from R. uniflorum. The colonizing event would have taken place during the period of lowest sea levels (between 50 000 and 84 000 years ago), from the coasts of South Asia. Was it the Aborigines that introduced the plant, and were they motivated by its medicinal properties? Had the species, on the other hand, reached Australia without human intervention, but how would this have been achieved?

The Rhaponticum occidental clade

This second clade within Rhaponticum (PP = 99 %, BS = 92 %; Fig. 1) embraces species distributed in North Africa, the Canary Islands, Europe and the Near East. Rhaponticum pulchrum Fisch. & C. A. Meyer, from Iran–Afganistan and the Caucasus, is situated as sister of the remainder of this group (Fig. 1), which suggests that the occidental clade originated in the Near East. A characteristic of the occidental clade is that a grouping of the North-African species is not seen in one subclade with the European species in another, but, on the contrary, various subclades combining species from North Africa, Europe and/or a mixed distribution are seen (Fig. 1). This suggests several independent passages from one continent to the other during the evolutionary history of the group.

(1) The association Rhaponticum heleniifolium Godr. & Gren. plus R. cynaroides Less. (PP = 99 %, BS = 100 %; Fig. 1) is the exception because the former one is endemic to the Alps and the latter to the Pyrenees. The two species present the particularity of exhibiting ramified inflorescential stems, according to the authors' observations. Few species of Rhaponticum show this character, and these are always <50 cm in height, while R. heleniifolium and R. cynaroides reach 1 m.

(2) The group including the Leuzea species (PP = 99 %, BS = 99 %; Fig. 1) shows a mixed distribution between North Africa and Europe. The present study confirms, for the first time on a molecular basis, that Leuzea and Rhaponticum should be fused, as previously suggested by Holub (1973) for morphological reasons. Leuzea berardioides Batt., endemic to the High Atlas (Morocco), appears as clearly segregated from L. conifera (L.) DC. (the molecular divergence for the regions studied between the two species is 7 %). This fact contradicts its consideration as a synonym or as a subspecies of L. conifera by Susanna (2002), and Greuter (2003), respectively, and implies a new nomenclatural combination of L. berardioides as Rhaponticum, as detailed in the Appendix. Another taxon, L. fontqueri, had been described by Sauvage (1968) as closely related to L. berardioides. The present results suggest that L. fontqueri is more closely related to L. conifera (PP = 100 %, BS = 100 %; Fig. 1), the only molecular divergences observed between these two taxa concerning polymorphic positions. Leuzea conifera presents a wide distribution area (western Mediterranean and Portugal) and a high morphological variability. In this sense, more studies are necessary to determine whether the differences observed with L. fontqueri are included in the natural variability of L. conifera, or if this endemic of the Chefchaouène Mountains (Morroco) merits the status of species. Meanwhile, the new nomenclatural combination for L. fontqueri as Rhaponticum is proposed in the appendix. Leuzea conifera had been combined previously as Rhaponticum by Greuter (2003). Rhaponticum acaule (L.) DC., is positioned at the base of the Leuzea group but without statistical support.

(3) Rhaponticum canariense DC., the only representative of Rhaponticum from the Canary Islands and seriously threatened with extinction, appears closely related to R. exaltatum (Willk.) Greuter, a species from central Spain and north-east Portugal (PP = 99 %, BS = 100 %; Fig. 1). Although floras do not usually indicate it, R. exaltatum could also be present in Morocco, in the Rif Atlas, according to a voucher from the Herbarium of Montpellier ['montagnes de Ketama', Sennen & Mauricio, VI-1934 (MPU); the determination of the herbarium sample was established and confirmed respectively by Maire in 1936 and Dittrich in 1976].

(4) There is another subclade which associates R. scariosum Lam. and R. longifolium (Hoffmanns. & Link) Dittrich (PP = 99 %, BS = 100 %; Fig. 1). Rhaponticum cossonianum (Ball) Greuter is positioned as sister of these two species, but with weak statistical suppport (Fig. 1).

An important outcome of this study is that the two representatives of Rhaponticum growing in the Alps, R. scariosum and R. heleniifolium, considered by several authors as subspecies of R. scariosum (Briquet, 1902; Rouy, 1905; Burnat, 1931; Holub, 1973; Dostál, 1976), do not appear as sisters in the phylogeny (Fig. 1). This implies inter alia that the colonization of the Alps took place in, at least, two independent events. In the same order of things, the biogeographic inference suggests that the two species of Rhaponticum indigenous to the east of Europe (R. serratuloides and R. scariosum), could have reached this region in two different ways: one expansion across central Asia to eastern Europe generating R. serratuloides; the other expansion through the Near East, North Africa and then to the Iberian Peninsula, thence on to the Alps, generating R. scariosum.

In view of this hypothesis suggested by the present analysis, it is regrettable that some nodes of the Rhaponticum occidental clade are weakly supported, and it would be interesting to perform more studies to get a better understanding of its biogeographic history.

Conclusions

This study confirms the main expectations of the study. It defines the genus Klasea as being probably the group of taxa most closely related to the Rhaponticum group. The generic delimitation of the Rhaponticum group would include the genera Callicephalus, Centaurothamnus, Myopordon, Ochrocephala, Oligochaeta and Rhaponticum (including Acroptilon and Leuzea). The new molecular evidence is consistent with the karyological and morphological data, and suggests particularly coherent biogeographic routes of migration and speciation processes for the genus Rhaponticum.

Acknowledgments

The authors thank two anonymous reviewers for critical comments on the manuscript, E. Carqué, V. Funk, F. Gómiz, G. Lazkov, K. Romashchenko, A. Romo and J. Vallès for their assistance with collections. We also thank M. Veny for keeping the collections of living plants, C. Roquet for technical support and R. Vilatersana for helpful comments. In addition, we thank S. Garcia and S. Pyke for considerably improving the English of the text. The collaboration of the botanical gardens and herbaria listed in Table 2 is also acknowledged. This work was subsidized by the Dirección General de Enseñanza Superior, Spain (Project PB 97/1134), Ministerio de Ciencia y Tecnologia, Spain (Projects PB BOS2001-3041-C02-02 and PB BOS2002-11856-E) and Generalitat de Catalunya ('Ajuts a grups de recerca consolidats' 1999SGR00332 and 2001SGR00125). T. Garnatje and O. Hidalgo benefited from a post-doctoral and a pre-doctoral grant, respectively, from the Ministerio de Educación y Ciencia (Spanish government).

APPENDIX: NEW NOMENCLATURAL COMBINATIONS

Klasea algida (Iljin) Hidalgo, comb. nov. Basionym: Serratula algida Iljin, Repertorium Specierum Novarum Regni Vegetabilis 35: 357 (1934).

Klasea biebersteiniana (Iljin ex Grossh.) Hidalgo, comb. nov. Basionym: Serratula radiata ssp. biebersteiniana Iljin ex Grossh., Flora Kavkaza 4: 194 (1934) ≡ Serratula biebersteiniana (Iljin ex Grossh.) Takht. in Takhtajan & Fedorov, Flora Erevana: 323 (1945).

Klasea khuzistanica (Mozaffarian) Mozaffarian, comb. nov. Basionym: Centaurea khuzistanica Mozaffarian, Iranian Journal of Botany 5(2): 84 (1992) ≡ Serratula khuzistanica (Mozaffarian) Mozaffarian in Garcia-Jacas et al., Botanical Journal of the Linnean Society 128: 420 (1998).

Rhaponticum berardioides (Batt.) Hidalgo, comb. nov. Basionym: Leuzea berardioides Batt., Contributions à la Flore Atlantique: 55 (1919) ≡ Rhaponticum coniferum subsp. berardioides (Batt.) Greuter, Willdenowia 33: 61 (2003).

Rhaponticum fontqueri (Sauvage) Hidalgo, comb. nov. Basionym: Leuzea fontqueri Sauvage, Collectanea Botanica (Barcelona) 59, 7(2): 1100 (1968).

Rhaponticum repens (L.) Hidalgo, comb. nov. Basionym: Centaurea repens L., Species Plantarum ed. 2: 1293 (1763) ≡ Acroptilon repens (L.) DC., Prodromus 6: 663 (1838).

LITERATURE CITED

  1. Baldwin BG, Markos S. 1998. Phylogenetic utility of the external transcribed spacer (ETS) of 18S-26S rDNA: congruence of ETS and ITS trees of Calycadenia (Compositae). Molecular Phylogenetics and Evolution 10: 449–463. [DOI] [PubMed] [Google Scholar]
  2. Böhle UR, Hilger HH, Martin WF. 1996. Island colonization and evolution of the insular woody habit in Echium L. (Boraginaceae). Proceedings of the National Academy of Sciences of the USA 93: 11740–11745. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Briquet J. 1902. Monographie des Centaurées des Alpes Maritimes. In: Burnat E, ed. Matériaux pour servir à l'histoire de la flore des Alpes maritimes. Geneve, Basel.
  4. Burnat E. 1931. Centaurea Rhapontica L. (Composées Cynaroïdées). In: Briquet J, Cavillier FG, eds. Flore des Alpes maritimes 7: 97–100, Genève, Basel. [Google Scholar]
  5. Cantó P. 1982. Números cromosómicos en algunos taxones del género Serratula L. (Asteraceae). Lazaroa 3: 189–195. [Google Scholar]
  6. Cantó P. 1984. Revisión del género Serratula L. (Asteraceae) en la Península Ibérica. Lazaroa 6: 7–80. [Google Scholar]
  7. Cassini AHG. 1825. In: King RM, Janaske PC, Lellinger DB. 1995. Cassini on Compositae II. Monographs in Systematic Botany from the Missouri Botanical Garden 54: [i]–xii, 1–190.
  8. Chouksanova NA, Sveshnikova LI, Alexandrova TV. 1968. Data on karyology of the family Compositae Giseke. Citologija 10: 198–206. [Google Scholar]
  9. Cullings KW. 1992. Design and testing of a plant-specific PCR primer for ecological and evolutionary studies. Molecular Ecology 1: 233–240. [Google Scholar]
  10. Dittrich M. 1977. Cynareae – systematic review. In: Heywood VH, Harborne JB, Turner BL, eds. The biology and chemistry of the Compositae. London: Academic Press, 999–1015.
  11. Dittrich M. 1983. Ochrocephala, eine neue afrikanische Compositengattung der Subtribus Centaureinae. Botanischer Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie 103: 467–480. [Google Scholar]
  12. Dostál J. 1976. Leuzea DC. In: Tutin TG, Heywood VH, Burges NA, Moore DM, Valentine DH, Walters SM, et al., eds. Flora Europaea, Vol. 5. Cambridge: Cambridge University Press, 252–253.
  13. Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15. [Google Scholar]
  14. Farris JS. 1970. Methods of computing Wagner trees. Systematic Zoology 19: 83–92. [Google Scholar]
  15. Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791. [DOI] [PubMed] [Google Scholar]
  16. Fernández Casas J, Susanna A. 1986. Monografía de la sección Chamaecyanus Willk. del género Centaurea L. Treballs de l'Institut Botànic de Barcelona 10: 5–174. [Google Scholar]
  17. Garcia-Jacas N, Susanna A, Ilarslan R. 1996. Aneuploidy in Centaureinae (Compositae): is n = 7 the end of the series? Taxon 45: 39–42. [Google Scholar]
  18. Garcia-Jacas N, Susanna A, Mozaffarian V. 1998b New chromosome counts in the subtribe Centaureinae (Asteraceae, Cardueae) from West Asia. III. Botanical Journal of the Linnean Society 128: 413–422.
  19. Garcia-Jacas N, Susanna A, Vilatersana R, Guara M. 1998a New chromosome counts in the subtribe Centaureinae (Asteraceae, Cardueae) from West Asia. II. Botanical Journal of the Linnean Society 128: 403–412.
  20. Garcia-Jacas N, Susanna A, Garnatje T, Vilatersana R. 2001. Generic delimitation and phylogeny of the subtribe Centaureinae (Asteraceae): a combined nuclear and chloroplast DNA analysis. Annals of Botany 87: 503–515. [Google Scholar]
  21. Greuter W. 2003. The Euro+Med treatment of Cardueae (Compositae)—generic concepts and required new names. Willdenowia 33: 49–61. [Google Scholar]
  22. Greuter W, Aghababian M, Wagenitz G. 2005. Vaillant on Compositae—systematic concepts and nomenclatural impact. Taxon 54: 149–174. [Google Scholar]
  23. Hellwig FH. 1994. Chromosomenzahlen aus der Tribus Cardueae (Compositae). Willdenovia 24: 219–248. [Google Scholar]
  24. Hellwig FH. 2004. Centaureinae (Asteraceae) in the Mediterranean—history of ecogeographical radiation. Plant Systematics and Evolution 246: 137–162. [Google Scholar]
  25. Holub J. 1973. Contribution to the taxonomy and nomenclature of Leuzea DC. and Rhaponticum auct. Folia Geobotanica et Phytotaxonomica, Praha 8: 377–395. [Google Scholar]
  26. Huelsenbeck JP, Ronquist F, Nielsen R, Bollback JP. 2001. Evolution—Bayesian inference of phylogeny and its impact on evolution biology. Science 294: 2310–2314. [DOI] [PubMed] [Google Scholar]
  27. Lessing CF. 1832. Synopsis generum Compositarum. Berlin: Dunker and Humblot.
  28. Linder CR, Goertzen LR, Vanden Heuvel B, Francisco-Ortega J, Jansen RK. 2000. The external transcribed spacer of the rDNA repeat: a new nuclear region for low-level taxonomic analysis of the Asteraceae and closely allied families. Molecular Phylogenetics and Evolution 14: 285–303. [DOI] [PubMed] [Google Scholar]
  29. Löve Á, Löve D. 1961. Chromosome numbers of central and northwest European plant species. Opera Botanica 5: 1–581. [Google Scholar]
  30. Maddison DR. 1991. The discovery and importance of multiple islands of most-parsimonious trees. Systematic Zoology 40: 315–328. [Google Scholar]
  31. Martín J, Garcia-Jacas N. 2000. Pollen studies in subtribe Centaureinae (Asteraceae): the Jacea group analysed with electron microscopy. Botanical Journal of the Linnean Society 133: 473–484. [Google Scholar]
  32. Martins L, Hellwig FH. 2005. Systematic position of the genera Serratula and Klasea (Cardueae, Asteraceae) inferred from ETS and ITS sequence data and new combinations in Klasea. Taxon 54: 632–638. [Google Scholar]
  33. Nickrent DL, Schuette KP, Starr EM. 1994. A molecular phylogeny of Arceuthobium (Viscaceae) based on nuclear ribosomal DNA internal transcribed spacer sequences. American Journal of Botany 81: 1149–1160. [Google Scholar]
  34. Nylander JA. 2004. Mr Modeltest V. 2. Program distributed by the author. Evolutionary Biology Centre, Uppsala University.
  35. Plinius G. 77. In: Teubner. 1897. Naturalis Historia 27: 574.
  36. Rouy G. 1905. Centaurea sous-genre Eu-Rhaponticum. In: Flore de France ou Description des plantes qui croissent spontanément en France, en Corse et en Alsace-Lorraine, Vol. 9. Tours: Société des sciences naturelles de la Charente-Inférieure, 114–115.
  37. Sauvage 1968. Notes de floristique rifaine. Collectanea Botanica 59: 1100–1104. [Google Scholar]
  38. Soltis DE, Kuzoff RK. 1993. ITS sequence homogeneity within and among populations of Lomatium grayi and L. laevigatum (Umbelliferae). Molecular Phylogenetics and Evolution 2: 166–170. [DOI] [PubMed] [Google Scholar]
  39. Soltis DE, Soltis PS, Collier TG, Edgerton ML. 1991. The Heuchera group (Saxifragaceae): evidence for chloroplast transfer and paraphyly. American Journal of Botany 78: 1091–1112. [Google Scholar]
  40. Soskov JD. 2001. Rhaponticum Adans. (Compositae Tribes Cynareae and Mutisieae). In: Bobrov EG, Cerepanov SK, eds. Flora of URSS, Vol. 28. Leningrad: Academiae Scientiarum URSS, 305–319.
  41. Sun Y, Skinner DZ, Liang GH, Hulbert SH. 1994. Phylogenetic analysis of Sorghum and related taxa using internal transcribed specers of nuclear ribosomal DNA. Theoretical and Applied Genetics 89: 26–32. [DOI] [PubMed] [Google Scholar]
  42. Susanna A. 2002. Leuzea DC. In: Valdés B, Redjali M, Achhal El Kadmiri A, Jury SL, Montserrat JM, eds. Catalogue des plantes vasculaires du Nord du Maroc, incluant des clés d'identification, Vol. 2, Madrid: CSIC, 715.
  43. Susanna A, Garcia-Jacas, N. 2006. Tribe Cardueae (Compositae). In: Kadereit J, Kubitzki K, eds. The families and genera of vascular plants. Heidelberg: Springer-Verlag (in press).
  44. Susanna A, Garcia-Jacas N, Hidalgo O, Vilatersana R, Garnatje T. 2006. The Cardueae (Compositae) revisited: insights from a ITS, trnL-trnF and matK nuclear and chloroplast DNA analysis. Annals of the Missouri Botanical Garden 93 (in press).
  45. Swofford DL. 1999. PAUP*. Phylogenetic analysis using parsimony (*and other methods), Version 4. Sunderland, MA: Sinauer Associates.
  46. Swofford DL, Olsen GJ. 1990. Phylogeny reconstruction. In: Hillis DM, Moritz C, eds. Molecular systematics. Sunderland, MA: Sinauer Associates, 411–501.
  47. Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal primers for amplification of three non–coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105–1109. [DOI] [PubMed] [Google Scholar]
  48. Vilatersana R, Susanna A, Garcia-Jacas N, Garnatje T. 2000. Karyology, generic delineation and dysploidy in the genera Carduncellus, Carthamus and Phonus (Asteraceae). Botanical Journal of the Linnean Society 134: 425–438. [Google Scholar]
  49. Wagenitz G. 1958. Die Gattung Myopordon Boiss. (Compositae-Cynareae). Berichte der Deutschen Botanischen Gesellschaft (Stuttgart) 71: 271–277. [Google Scholar]
  50. Wagenitz G, Hellwig FH. 1996. Evolution of characters and phylogeny of the Centaureinae. In: Hind DJN, Beentje HG, eds. Compositae: systematics. Proceedings of the International Compositae Conference, Kew, 1994. Kew: Royal Botanic Gardens, 491–510.
  51. Wagenitz G, Dittrich M, Damboldt J. 1982. Centaurothamnus, eine neue Gattung der Compositae-Cardueae in Arabien. Candollea 37: 101–115. [Google Scholar]
  52. White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis M, Gelfand D, Sninsky J, White T, eds. PCR protocols: a guide to methods and applications. San Diego, CA: Academic Press, 315–322.

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES