sign in sign up
<<
Home , Anvillea, Asteriscus (plant), Asteriscus aquaticus, Asteriscus sericeus, Blepharispermum, Blumea

ARTICLE IN PRESS

Organisms, Diversity & Evolution 5 (2005) 135–146 www.elsevier.de/ode

Evolutionary relationships in the tribe (incl. ) evidenced by DNA sequences of ndhF; with notes on the systematic positions of some aberrant genera Arne A. Anderberga,Ã, Pia Eldena¨ sb, Randall J. Bayerc, Markus Englundd aDepartment of Phanerogamic Botany, Swedish Museum of Natural History, P.O. Box 50007, SE-104 05 Stockholm, Sweden bLaboratory for Molecular Systematics, Swedish Museum of Natural History, P.O. Box 50007, SE-104 05 Stockholm, Sweden cAustralian National , Centre for Biodiversity Research, GPO Box 1600 Canberra ACT 2601, dDepartment of Systematic Botany, University of Stockholm, SE-106 91 Stockholm, Sweden

Received27 August 2004; accepted24 October 2004

Abstract

The phylogenetic relationships between the tribes Inuleae sensu stricto andPlucheeae are investigatedby analysis of sequence data from the cpDNA gene ndhF. The delimitation between the two tribes is elucidated, and the systematic positions of a number of genera associatedwith these groups, i.e. genera with either aberrant morphological characters or a debated systematic position, are clarified. Together, the Inuleae and Plucheeae form a monophyletic group in which the majority of genera of Inuleae s.str. form one clade, and all the taxa from the Plucheeae together with the genera Antiphiona, , , Ondetia, Pechuel-loeschea, ,andIphionopsis from Inuleae s.str. form another. Members of the Plucheeae are nestedwith genera of the Inuleae s.str., andsupport for the Plucheeae clade is weak. Consequently, the latter cannot be maintained and the two groups are treated as one tribe, Inuleae, with the two subtribes Inulinae andPlucheinae. The genera , Chrysophthalmum, , , , , and are demonstrated to be non-monophyletic. and Iphionopsis are foundto belong to the same clade as the taxa of the former Plucheeae. is shown to be a close relative of and of the Inuleae-Inulinae. The genera and Zoutpansbergia belong to the major clade of the family that includes the tribes sensu lato andInuleae (incl. Plucheeae), but their exact position remains unresolved.The is not part of the Inuleae, but is either part of the unresolvedbasal complex of the paraphyletic Cichorioideae, or sister to the entire . r 2005 Gesellschaft fu¨ r Biologische Systematik. Publishedby Elsevier Gmbh. All rights reserved.

Keywords: Asteraceae; Inuleae; Inulinae; Plucheinae; Phylogeny; ndhF

Introduction papers (Bremer, 1987; Anderberg, 1989; Karis et al., 1992; Kim andJansen, 1995 ; Eldena¨ s et al., 1999; The subdivision of the former tribe Inuleae into three Panero andFunk, 2002 ). The separation of tribe smaller groups has gainedsupport in a number of from the rest of the Inuleae complex is well supported and evidently correct, but the delimita- ÃCorresponding author. tion of the other genera of the Inuleae, referredto as E-mail address: [email protected] (A.A. Anderberg). tribes Inuleae s.str. andPlucheeae by Anderberg (1989,

1439-6092/$ - see front matter r 2005 Gesellschaft fu¨ r Biologische Systematik. Publishedby Elsevier Gmbh. All rights reserved. doi:10.1016/j.ode.2004.10.015 ARTICLE IN PRESS 136 A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146

1991a, b), have not been fully understood. Early group with other taxa of Gnaphalieae, with Inuleae analyses of molecular data based on ndhF sequence s.str., or with the Plucheeae. The genus , which data (Kim andJansen, 1995 ; Eldena¨ s et al., 1999) with great hesitation was acceptedin the Plucheeae by showedthat the Inuleae andPlucheeae are sister groups, Anderberg (1994), has been foundto belong in the andthat they together constitute the sister group of Heliantheae (Baldwin and Wessa, 2000), andwas not Heliantheae s. lat., Eupatorieae, andthe African investigatedfurther by us. Anisopappus, and Blepharispermum, three genera formerly included in the Inuleae (Merxmu¨ ller et al., 1977). Only three genera from Inuleae s.str. andPlucheeae Material and methods were included in the study by Kim andJansen (1995) , andthis small sample size didnotallow for any Molecular methods statements on whether the two representedmonophy- letic groups, or one was derived within the other. To DNA was extractedfrom taken from herbar- clarify this issue, Eldena¨ s et al. (1999) investigatedthe ium specimens or from material dried in silica gel. relationships between the two tribes with a larger sample Voucher information for new ndhF sequences is of taxa. They showedthat the tribe Plucheeae formed presentedin the AppendixA, other sequences are those one clade and that the majority of taxa from the Inuleae usedby Eldena¨ s et al. (1999), or more recent depositions s.str., as circumscribedby Anderberg (1989, 1991a), in GenBank (Asteriscus spp., spp., , formedanother. However, their studyfailedto resolve Vierea, Xerolekia). For many extractions, leaves were the positions of three genera from the Inuleae s.str., i.e. groundin liquidnitrogen with mortar andpestle, and Antiphiona, Geigeria, and Pegolettia, in relation to the DNA subsequently extractedwith the methodof Plucheeae andInuleae clades,andthus left tribal Saghai-Maroof et al. (1984) as modified by Doyle and delimitation somewhat unclear. In the same paper, Doyle (1987). For others, leaves were groundwith a Eldena¨ s et al. (1999) described the presence of a Mini-BeadBeater (BioSpec Products,Bartlesville, Ok- characteristic 3-bp insertion in ndhF (CCT in position lahoma, USA) andsubsequently treatedwith the 1588 from 50 endof the Inula sequence in GenBank DNEasy plant DNA extraction kit from Qiagen (Qiagen Accession No. L39453) as characteristic of taxa belong- Inc., Valencia, , USA), following the manu- ing to the Inuleae s.str. clade. The same genera that were facturer’s protocol. All PCR reactions were performed shown to have this 3-bp insertion are also characterized with 10 mmol/l primers in 25-ml reactions using ‘‘Ready- by a large, rhomboidoxalate crystal in each cell of the to-go’’ PCR beads from Pharmacia Biotech (Amersham cypsela epidermis (Anderberg 1989, 1991a). The 3-bp Pharmacia Biotech, Uppsala, Sweden), following the insertion is absent from all other Asteraceae, including manufacturer’s standard protocol and generally with the all the genera of the Plucheeae clade, and notably also thermal cycling profile, 95 1C for 5 min, followedby 45 from the three genera of unclear systematic position or 55 cycles of 95 1C for 30 s, 45 1C30s,721C 2 min, and mentionedabove ( Antiphiona, Geigeria, and Pegolettia). finally 72 1C for 8 min. Primers usedare those of Eldena¨ s Our present investigation includes a still larger sample et al. (1999). of taxa, aiming at a better understanding of the For sequencing reactions in some taxa, the Thermo relationships between Inuleae andPlucheeae in order Sequenase Fluorescent Sequencing Kit from Pharmacia to resolve their tribal status. Apart from elucidating (Amersham Pharmacia Biotech AB, Uppsala, Sweden) tribal interrelationships, we have also triedto shedlight was used. Fragments were sequenced using fluorescently on the issue of monophyly of some of the larger genera, labeled(CY-5) primers. Fragments were separatedon an such as Blumea, Pluchea, Inula, and Pulicaria,by ALF-Express (Pharmacia Biotech AB, Uppsala, Swe- including several representatives of each in the analysis. den). For most taxa the ‘‘Big Dye Terminator Sequen- Furthermore, we have included a number of genera cing’’ kit (AppliedBiosystems, Warrington, Cheshire, whose systematic positions have been debated, or which UK) was used, and fragments were separated on an have an aberrant morphology comparedto typical ABI377 from AppliedBiosystems. Primers usedfor members of these tribes. Such genera are Caesulia, PCR andfor sequencing are the same as those usedby Callilepis, Cratystylis, Gymnarrhena, Iphionopsis, Rho- Eldena¨ s et al. (1999). Sequences were assembledwith the dogeron, , Zoutpansbergia, andalso Dielitzia. Staden software (Staden et al., 1998), andalignedwith The latter was described as a member of Inuleae, but the BioEdit software (Hall, 1999). Two sequences, was included in the Gnaphalieae by Anderberg (1991c). unpublishedat the time of our analysis, from Rhodoger- Like Dielitzia, many Australian Gnaphalieae have more on coronopifolius (GenBank Acc. No. AY226799)and or less cartilaginous without the papery lamina from Sachsia polycephala (AY226800), were kindly that is typical of most representatives of that tribe, andit provided by Javier Francisco-Ortega, Interna- was of interest to test if a taxon such as Dielitzia would tional University. All 63 new ndhF sequences have been ARTICLE IN PRESS A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146 137 submittedto GenBank (Accession Nos. AY780811– Boopis Barnadesia Dasyphyllum AY780873; see Appendix A). 95 100 Chuquiraga 98 Doniophyton 78 Schlechtendahlia Phylogenetic analyses Mutisia Adenocaulon Gochnatia The alignment of ndhF sequences resultedin a data Onoseris Stifftia matrix with 184 taxa and2280 characters of which 660 Leibnitzia Gerbera 100 were informative for phylogenetic analysis. The data 93 Piloselloides was analyzedwith parsimony jackknifing ( Farris et al., Nassauvia 100 Trixis 1996), using the computer software ‘‘Xac’’ (Farris, 1997) 57 Perezia1 100 Perezia2 with the following settings: 1000 replications, each with Ainsliaea branch-swapping and ten random-addition of se- Tarchonanthus Echinops quences. For the analysis, Boopis (Calyceraceae) was Atractylodes 99 Carlina usedas outgroup, as this family constitutes the sister Cardueae p. p. maj. 95 group of the Asteraceae. Gaps foundin the ndhF 62 Gymnarrhena Lactuceae 100 sequences were treatedas missing information; gaps 69 Arctoteae Liabeae 100 were few, andthe alignment was unproblematic. Vernonieae 97 Senecioneae Analyses were performedwith all codon positions 100 98 Anthemideae 99 included, follwing the recommendations of Ka¨ llersjo¨ 67 100 Calenduleae 81 87 et al. (2000). Gnaphalieae

100 Asteroideae Zoutpansbergia 100 Callilepis 99 Athroisma 100 Blepharispermum Results 84 Anisopappus smutsii 100 Anisopappus latifolius 68 Heliantheae s. l. 98 continued in fig. 2 Inuleae – Inulinae The results of our investigation (Fig. 1) show the same 100 100 basic topology andrelationship between tribes as those continued in fig. 3 Inuleae – Plucheinae 51 foundearlier with a smaller taxon sample ( Eldena¨ set al., 1999), but give a much more detailed picture of Fig. 1. Asteraceae (partly abbreviated; continued in Figs. 2 generic relationships within the Inuleae-Plucheeae com- and3 ). basedon jackknife analysis of 184 DNA plex. The two tribes form a highly supported(100% sequences of ndhF; only nodes with jackknife support450% jackknife) monophyletic group, which is composedof are resolved; several tribes are shown abbreviated into single two main clades: Inuleae s.str. with robust support lines since they are not subjects of the main discussion in the present paper. (100%), andPlucheeae together with some genera from Inuleae s.str. with low support (51%). In Inuleae s.str. (Fig. 2), the basal bifurcation is between a group formedby Duhaldea, Blumea, and verbascifolia (99%). The thirdclade(86%) has Pulicaria Caesulia (93% jackknife support) andthe rest of the canariensis, P. dysenterica, Dittricha and in a genera (98%). In the first clade, Duhaldea is sister to a group (80%) that is sister to one formedby , group (100%) that has Caesulia as sister to five Asteriscus, Ighermia, and Pallenis (99%). and of Blumea (78%). is here sister to B. Jasonia form a monophyletic group (65%) within a riparia, B. saxatilis, B. integrifolia,andB. psammophila, paraphyletic Pulicaria, with P. dysenterica as sister to and B. integrifolia is sister to B. psammophila (100%). In Dittrichia+Jasonia and P. canariensis as sister to P. the clade comprising the rest of the Inuleae s.str., there dysenterica, Dittrichia, and Jasonia. Anvillea is sister to a are three main monophyletic groups. The first of these group formedby members of the genera Asteriscus, (82%) comprises Schizogyne, Vierea, Pulicaria crispa, Pallenis, and Ighermia (64%). Ighermia has low support andtwo species of (99%). The secondclade (52%) as sister to the two Asteriscus species, A. (61%) has Xerolekia as sister to (99%) imbricatus and A. schultzii (99%), whereas the remain- andthese two genera are then sister to a large group ing six Asteriscus species form another group (83%). (86%) containing Inula oculus-christi (GenBank Acces- The Plucheeae clade (Fig. 3) has a basal polytomy of sion No. L39453, under the synonym I. sericea), I. five smaller monophyletic groups, three of which are peacockiana, Pentanema alanyense as sister to Chry- formedby genera usually associatedwith Inuleae s.str. sophthalmum gueneri (98%), as sister to One group is that with Geigeria and Ondetia (100%), the (98%), anda group with Amblyocarpum, secondhas 94% support for a group with Calostephane Chrysophthalmum montanum, Pentanema glanduligerum, and Mollera (79%) as sister to the three Pegolettia Inula bifrons, I. britannica, I. , I. germanica and I. species (100%). The thirdgroup has Laggera decurrens ARTICLE IN PRESS 138 A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146

IN Duhaldea cappa PL campestre1 ? Caesulia axillaris 100 PL Stenachaenium campestre2 93 IN Blumea balsamifera IN Geigeria ornativa 100 IN Blumea riparia 100 IN Ondetia linearis 78 IN Blumea saxatilis PL Laggera decurrens IN Antiphiona fragrans 100 IN Blumea integrifolia 100 IN Antiphiona pinnatifida 100 IN Blumea psammophila 100 IN IN Calostephane marlothiana IN Vierea laevigata 79 IN Mollera angolensis continued IN Pegolettia lanceolata IN Pulicaria crispa 94 82 from fig. 1 51 IN Pegolettia senegalensis IN Rhanterium adpressum 100 continued 86 IN Pegolettia oxydonta from fig.1 100 99 IN ? Cratystylis conocephala IN Xerolekia speciosissima IN? Iphionopsis rotundifolia 99 IN PL sphacelatum IN Inula oculus-christi PL Rhodogeron coronopifolius IN Inula peacockiana 99 PL Sachsia polycephala1 61 IN Pentanema alanyense 87 PL Sachsia polycephala2 98 IN Chrysophthalmum gueneri IN Pechuel-loeschea leubnitziae IN Carpesium divaricatum 54 PL commersonii IN Telekia speciosa PL Doellia bovei 86 98 IN Chrysophthalmum montanum PL cunninghamii 98 IN Inula britannica PL caspica IN Pentanema glanduligerum PL Laggera pterodonta PL Pluchea dioscoroides IN Inula bifrons 59 IN Inula conyza PL Pluchea dunlopii 99 PL ssp.yemenensis IN Inula germanica PL Pluchea nogalense IN Inula verbascifolia PL Pluchea obovata IN Amblyocarpum inuloides PL IN Pulicaria canariensis PL Pluchea tetranthera1 IN Pulicaria dysenterica 50 80 PL Pluchea tetranthera2 100 IN PL Pluchea yucatanensis 65 IN Jasonia tuberosa PL Porphyrostemma grantii 86 IN Anvillea garcinii PL viscosa IN PL angolensis IN Pallenis hierochuntica PL integrifolia PL liatroides 99 100 IN IN Ighermia pinifolia 52 PL Pluchea dentex IN Asteriscus imbricatus 52 PL Coleocoma centaurea 52 PL Pluchea arabica 64 IN Asteriscus schultzii 99 PL Pluchea lucens IN PL Pluchea littoralis IN Asteriscus daltonii 62 PL Pluchea sarcophylla IN Asteriscus graveolens 76 PL Pluchea elliptica 83 IN Asteriscus intermedius IN IN Fig. 3. Inuleae-Plucheinae (continuedfrom Fig. 1). Tree based on jackknife analysis of DNA sequences of ndhF; only nodes Fig. 2. Inuleae-Inulinae (continuedfrom Fig. 1). Tree based with jackknife support 450% are resolved; IN ¼ Inuleae s.str. on jackknife analysis of DNA sequences of ndhF; only nodes (Anderberg, 1991a), PL ¼ Plucheeae (Anderberg, 1991b). with jackknife support450% are resolved; IN ¼ Inuleae s.str. (Anderberg, 1991a). as sister to the two species of Antiphiona (100%). One of that differ from the typical everlastings of that tribe. The the two remaining clades has two samples of Stenachae- position of this genus is not shown in the tree, andwill nium (100%), andthe final cladein the polytomy not be discussed further in the present context. comprises the remaining genera (54%). In that large group, the genera Iphionopsis, Cratystylis, Pterocaulon, a group with Rhodogeron and Sachsia (99%), andthe remaining Plucheeae genera (59%) constitute an un- Discussion resolvedpentatomy. The core Plucheeae group has Pechuel-loeschea from Inuleae s.str. as sister to the rest Our present results (Fig. 1) corroborate the hypoth- of the taxa (50%), a group formedby Coleocoma, eses of Bremer (1987) and Anderberg (1989), that two Cylindrocline, Doellia, Epaltes, Karelinia, Laggera p.p., separate monophyletic groups can be identified within Pluchea, Porphyrostemma, Pseudoconyza, Sphaeranthus, the Inuleae-Plucheeae complex, but extendthose hy- Streptoglossa, and Tessaria. In the larger (50%) group, potheses to show that Inuleae andPlucheeae together Pluchea arabica, P. kelleri, P. littoralis, P. lucens, and P. form a monophyletic group. Most taxa placedin Inuleae sarcophylla form a group (62%), andweak support s.str. by Anderberg (1991a) form a clade (Fig. 2), but (52%) is also obtainedfor a group indicating a close some of its genera belong to the same clade as the taxa relationship between Coleocoma, Pluchea dentex, and of Plucheeae (Anderberg, 1991b) andconstitute part of Streptoglossa. The other relationships among the taxa of an unresolvedancestral complex in that clade( Fig. 3). this group are unresolved. Therefore we propose that Plucheeae henceforth be The genus Dielitzia in our analysis was confirmedto treatedas a part of Inuleae. Although support for this belong in Gnaphalieae andwas placednear Antennaria clade is low (51%), there are subtribal names already (100%). This means that it belongs in the core available for the two clades: Inuleae-Inulinae and Gnaphalieae, in spite of its many specializedfeatures Inuleae-Plucheinae, respectively. ARTICLE IN PRESS A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146 139

Inuleae-Plucheinae scales at the base of the pappus bristles. Apart from this, P. lanceolata only shares the widened corolla base, a few The tribes Lactuceae, Arctoteae, Liabeae, andVer- vague pollen wall characteristics, andpresence of several nonieae, i.e. the tribes constituting the so-calledLALV- small crystals in the cypsela epidermis with the rest of complex (Kim andJansen, 1995 ), are the Cichorioideae the species of the genus. tribes that are most closely relatedto the subfamily On the next higher level of the phylogeny the relation- Asteroideae to which Inuleae belongs. In the LALV- ships are still somewhat unresolvedandthe genera belong complex, taxa generally have style branches with acute to one of five clades in a pentatomy. The first is formed sweeping hairs, which are also characteristic of the by the mainly Caribbean genera Rhodogeron and Sachsia, genera in Inuleae-Inulinae, thus being a plesiomorphic which are sisters. The genus Sachsia has pollen with character state. In contrast, most genera belonging to branching bacules (Leins, 1971), andin this differs from the Plucheeae, as circumscribedby Anderberg (1991b), pollen of other Plucheinae. Both genera were placedin have styles with obtuse sweeping hairs reaching below the Sachsia group of Inuleae-Inulineae s. ampl. by the style bifurcation. However, also several taxa of the Merxmu¨ ller et al. (1977), andin spite of their styles Inuleae-Plucheinae, as circumscribedhere, have styles providedwith acute sweeping hairs they were treatedas with acute sweeping hairs ending above the style anomalous members of the Plucheeae by Anderberg bifurcation, i.e. Geigeria, Laggera decurrens, Ondetia, (1991b), who also hypothesized Rhodogeron to be a taxon Antiphiona, Pegolettia, Calostephane, Mollera, Sachsia, derived within Sachsia proper. A recent study by Hong and Rhodogeron. Mollera differs from Calostephane et al. (2004) using ITS sequence data reconstructed the mainly in being devoid of a pappus, the former name phylogeny of the group andreinstated Rhodogeron as a was thus reduced to a synonym of the latter by separate genus, as it was shown to be the sister of Sachsia Anderberg (1991a), andthe close relationship gains andnot a derivedingroup. support also from our present study. All the genera just The pentatomy described above also includes the mentioned, as well as Laggera decurrens, also have genus Cratystylis, another taxon for which the systema- endothecial tissue with polarized wall thickenings. tic position has been debated. It is an Australian genus Laggera decurrens thus shares two character states with of four species of dioecious having white, deeply Antiphiona that differ from the condition in other lobedflorets, styles almost devoidofsweeping hairs, and Laggera species, which have obtuse sweeping hairs a stigmatic area covering almost the entire inner surface reaching below the style bifurcation, endothecial tissue of the style branches. Other Inuleae are not dioecious, with radial wall thickenings, and belong to the core have short-lobedflorets, distinct sweeping hairs, anda Plucheeae clade together with Pluchea andother genera. stigmatic area separatedin two lateral bandsthat are All taxa of the Inuleae-Plucheinae clade with acute confluent distally. Merxmu¨ ller et al. (1977) placedthe sweeping hairs on the style branches are positionedat genus with uncertain affinity in the so-called Cratystylis lower nodes in the tree than those with obtuse sweeping group of Inuleae-Inulineae s. ampl., but Zdero et al. hairs extending down on the style, e.g. Pluchea. (1988) considered the phytochemical components they All taxa of the Inuleae-Plucheinae clade, as it appears foundin Cratystylis to indicate a relationship with here, are devoid of the large oxalate crystal in the genera such as Epaltes, Pluchea, and Sphaeranthus cypsela epidermis that is characteristic of the Inuleae- (Plucheinae), andnot with Inula (Inulinae), thus similar Inulinae, andthey also lack the 3-bp insertion (CCT) in to our present results. Anderberg et al. (1992) described ndhF that is another characteristic of the taxa belonging the pollen morphology of Cratystylis conocephala as to that group. Anderberg (1991a) considered both being caveate with irregular baculate sexine, andwith a Antiphiona, Pechuel-loeschea, and Pegolettia to be smaller exine/pollen diameter ratio than what is typical members of Inuleae s.str. because of their acute style of genera belonging to any of the Asteroideae tribes. hairs, but notedthat they lackedthe oxalate crystals. They also foundthe pollen to be differentfrom that of However, in our present study these genera are shown to other Inuleae andPlucheeae, but with some resemblance be part of the Inuleae-Plucheinae. It is also interesting to to pollen from the genus Stenachaenium. Basedon a note that Pegolettia lanceolata is sister to P. oxyodonta number of features in Cratystylis that are normally and P. senegalensis; except for P. lanceolata, the species foundin the Cichorioideae,they consequently hypothe- of this genus share many distinct synapomorphies, and sizeda position for the genus in that subfamily, and the only uncertainty regarding the monophyly of the hence Bremer (1994) included Cratystylis as a genus of genus (Anderberg, 1986) was whether or not P. uncertain position within subfamily Cichorioideae. lanceolata is part of this genus, or not. The molecular Later, Bayer andCross (2003) foundthat Cratystylis data give robust support for P. lanceolata as sister to the actually is part of the plucheoidgroup of genera in the other species of the genus, thus corroborating Ander- subfamily Asteroideae, in spite of its somewhat differing berg’s results. All species of Pegolettia have a pappus morphological andpalynological features, andour with bristles andscales, andalso P. lanceolata has small molecular data corroborate that conclusion. ARTICLE IN PRESS 140 A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146

The thirdpart of the pentatomy is another genus for data. We included 14 taxa of Pluchea, the New World which the systematic position has been debated, viz. the species P. sericea and P. yucatanensis, the Australian P. African Iphionopsis. Two of the species were described dentex, P. dunlopii, and P. tetranthera (two sequences), as members of (Inuleae-Inulinae) but subse- andthe OldWorld P. arabica, P. dioscorides, P. indica, quently transferredto a new genus of uncertain P. kelleri, P. lucens, P. littoralis, P. nogalense, P. systematic position that has been associatedwith the obovata,andP. sarcophylla. These Pluchea species do Inuleae, but without a well-founded opinion of where it not form a clade, and their relationships to other genera belongs (Anderberg, 1985, 1991a). The species of of the clade are almost unresolved. King-Jones (2001) Iphionopsis have homogamous capitula andcharacter- placed Streptoglossa outside of Pluchea, andhad P. istic resin ducts in the mid-portion of the involucral dentex embedded between other Pluchea species. In our bracts andin the corolla, andhas an unsclerifiedfruit results, the Australian P. dentex takes a position closer wall. The stigmatic area covers almost the entire inner to the genera Coleocoma and Streptoglossa than to any surface of the style branches. In our results Iphionopsis is of the other Pluchea species. The support for this is low; part of Inuleae-Plucheinae, andit is interesting in this thus the question whether or not Pluchea is mono- context that Pluchea serra Franch. recently has been phyletic remains to be solved, and this is true also for the foundto be a synonym of I. rotundifolia (Thulin, 2003). generic status of Karelinia, Streptoglossa,andTessaria The fourth clade of the pentatomy is represented by in relation to Pluchea. Another taxonomic problem is the genus Pterocaulon, a mainly New Worldgenus, with the phylogeny of Laggera, in our study represented by a few species in Australia andadjacentareas, here the two species L. pterodonta, and L. decurrens. Whereas representedonly by the Australian P. sphacelatum. The L. pterodonta has a position in the Pluchea clade, L. diagnostic characters for Pterocaulon are the hetero- decurrens was foundto be sister to Antiphiona in the gamous capitula with filiform female florets, generally basal complex of Inuleae-Plucheinae. As mentioned wingedstems, andcapitula congestedinto spherical or above, Laggera s. str. has the same kindof obtuse elongated secondary heads. The monophyly of the genus sweeping hairs extending below the style bifurcation as has not been tested, but there is some documented in Pluchea, Epaltes, andother Plucheinae, but L. variation in morphology (number of disc-florets, num- decurrens differs by having acute sweeping hairs not ber of row of pappus bristles, trichome structure) reaching below the bifurcation. Formerly, this species between the Australian-Oceanic species (sometimes has been placedin Blumea (syn. B. gariepina), but it is referredto as genus Monenteles) andthe American apparent that it can be included neither in Blumea, nor ones, which may indicate that the genus is non- in Laggera. The genus Pseudoconyza (syn. Laggera monophyletic. Some species have distinct spherical or aurita, Blumea viscosa) did not group with any of the ellipsoid secondary heads, whereas others have a rather Laggera species in our analysis. The three species loose, spike- or raceme-like arrangement of capitula that Laggera gariepina, Pseudoconyza viscosa, and Doellia hardly can be referred to as secondary heads. Pterocau- bovei (syn. Blumea bovei) have at some point been lon has acute sweeping hairs not reaching below the style included in Blumea, but this genus, notedto be bifurcation, as in the genera at the next deeper level of heterogeneous by Anderberg (1991a), belongs to In- the clade (see above), but the endothecial tissue has uleae-Inulinae. The phylogenetic relationships among radial wall thickenings. the genera of Inuleae-Plucheinae is part of our ongoing The last group in the pentatomy comprises the core research andwill be the topic of a forthcoming paper. plucheoidtaxa, i.e. Coleocoma, Cylindrocline, Doellia, Epaltes, Karelinia, Laggera p.p., Pechuel-loeschea, Plu- chea, Porphyrostemma, Pseudoconyza, Sphaeranthus, Inuleae-Inulinae Streptoglossa,andTessaria. Most taxa in this group have styles with obtuse sweeping hairs extending below The taxa of this clade, except for Caesulia, are the bifurcation. Capitula are generally disciform with characterizedby the presence of one large oxalate filiform female outer florets andmale or hermaphroditic crystal in each cypsela epidermis cell, and furthermore central florets, but some species have discoid capitula they all share the unique 3-bp CCT insertion in ndhF with only hermaphroditic florets. The Australian Strep- mentionedabove. The subtribe has two well-supported toglossa differs in having conspicuously radiate capitula main monophyletic groups, i.e. Duhaldea together with with distinct purple ray florets, but otherwise conforms the Caesulia-Blumea group, andits sister group com- to the normal condition in the group. Anderberg prising all the taxa from the remaining genera, including (1991b) found Pluchea to be heterogeneous andbelieved the species rich Inula and Pulicaria (Fig. 2). the genus wouldprove to be paraphyletic, whereas Anderberg (1989, 1991a) foundthat Blumea shares King-Jones (2001) considered Old World and Australian several characteristic features with genera of Inuleae species of Pluchea to form a homogeneous monophyletic s.str., rather than with those of the Pluchea group, and group, a conclusion that is not supportedby our present movedit from a position near Pluchea to Inuleae s.str. A ARTICLE IN PRESS A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146 141 close relationship between the type of the genus (B. epidermis crystals. The fruits are unique within Inuleae, balsamifera)andDuhaldea was also hypothesizedfrom andin appearance more closely resemble those of many Anderberg’s analysis of morphological data. Among Heliantheae, albeit without the black phytomelanin other considerations, the presence in Blumea of the content. Anderberg (1989) did not accept Caesulia in Inula-type cypsela crystals andacute stylar sweeping either Inuleae or Plucheeae, because of its atypical style hairs were interpretedas indicationsof a position in andstigma andthe lack of other characters diagnosing Inuleae s.str. Anderberg also found Blumea heteroge- genera of these tribes. Later treatments by Anderberg neous in these character states, andtherefore put (1991b) and Bremer (1994), included Caesulia as an forwardthe ideathat the genus was not monophyletic. unplacedmember of subfamily Asteroideaeawaiting Later, Anderberg (1995) placedtwo Blumea species with more conclusive evidence regarding its tribal position. styles provided with obtuse sweeping hairs extending far Pandey et al. (2000) described the fruit anatomy of below the bifurcation in the genus Doellia, here foundto Caesulia axillaris andfoundthis to be more complex belong in the Pluchea clade of Inuleae-Plucheinae (see than hadbeen assumedbefore. What appears to be the above). As mentionedabove, two other former Blumea fruit was shown to be a structure formedby two species were foundto belong in Plucheinae as well. In strongly andlaterally flattenedbracts, which at maturity our present study we have included five species of completely conceal the pericarp. With increasing age the Blumea s. str., all of which group together with the type tissue of this ‘involucral cup’ dominates as it becomes andfall within the same monophyletic group that also hard and provided with resiniferous ducts, whereas the includes Duhaldea, thus corroborating Anderberg’s true pericarp only consists of a single cell layer, a (1995) view. sclerifiedepidermis. Pandey et al. (2000) referredto the A surprising result is that the molecular data strongly complex fruit structure as a pseudopericarp. A pappus is suggest that the morphologically aberrant genus Caesu- missing in Caesulia, but insteadthe pseudopericarp is lia belongs to the clade that includes Blumea and crownedby two scale-like outgrowths, which are likely Duhaldea, being a derived member of Inuleae-Inulinae. to be the tips of the involucral bracts. The very reduced The relationships andtribal position of Caesulia have pericarp lacks large oxalate crystals, andin this respect been unclear so far because of its many odd morpho- Caesulia differs from all other members of the Inuleae logical character states. The cypselas of Caesulia do not clade. The most parsimonious interpretation, given the have the large oxalate crystal in the epidermis cells that present phylogeny, is that the fruit of the ancestor of is typical of other genera of this clade, but the ndhF Caesulia was provided with cypsela crystals like in the sequence from Caesulia has the same characteristic 3-bp other taxa of Inuleae s.str., but that these were lost as a insertion (CCT, discussed above) as the other genera result of the increasing reduction of the pericarp. of Inuleae-Inulinae, an insertion that is not present in The sister group of the Duhaldea-Caesulia-Blumea ndhF of any sequencedAsteraceae genera outsidethe clade is the large group formed by the remaining genera, subtribe. including the Inula and Pulicaria complexes. These Hoffmann (1889) included Caesulia in Inuleae- complexes were hypothesizedby Anderberg (1991a) to Angianthinae (now part of tribe Gnaphalieae) because be paraphyletic residue groups devoid of the synapo- of its secondary heads consisting of one-flowered morphies diagnosing a number of smaller descendant capitula surrounded by two hyaline bracts. Besold genera, andseparatedby technical characters, the (1971) showedthat the style andpollen in Caesulia are presence (Pulicaria) or absence of small scales in a row not structuredlike in that group, andonly with outside the barbellate pappus bristles (Inula). Anderberg hesitation did Leins (1971) place Caesulia near the anticipatedthat a number of the smaller genera would genus Sphaeranthus (now Inuleae-Plucheinae). Merx- prove to have their closest relatives in either of these two mu¨ ller et al. (1977) followedLeins, but notedthat the genera, andin this respect our results corroborate his chromosome number (2n ¼ 14) andthe synflorescence conclusion. structure of Caesulia was aberrant from that of the other The large clade including the Inula and Pulicaria members of the Sphaeranthus group. The style in complexes has an unresolvedtrichotomy at the base. Caesulia has a stigmatic area that seems to cover the One clade in that trichotomy includes the genera entire inner surface of the style-branches, whereas the Rhanterium, Vierea,andSchizogyne, but also one typical Inuleae style has two separate stigmatic bands species of Pulicaria, i.e. P. crispa, which is often placed that merge apically. The sweeping hairs are also of in the genus Francoeuria. The two other species of intermediate shape, neither distinctly acute nor obtuse. Pulicaria belong to another monophyletic group, and Like Duhaldea the anthers have polarizedendothecial this supports the previous supposition that Pulicaria is tissue andwell-developedbranchedtails, but the apical not monophyletic. Vierea has an Inula-like appearance, anther appendix is not truncate as in Duhaldea. The but unlike Inula it has a pappus of bristles of unequal fruits in Caesulia are strongly flattenedwhen mature, length arrangedin several rows. In Rhanterium the crownedby two hyaline scales, andhave no visible pappus bristles are broadandflattenedwith an aspect ARTICLE IN PRESS 142 A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146 that is clearly different from the ordinary kind of surprising considering their similarity in general appear- capillary pappus bristles foundin other genera. The ance. Among the tall, large-headed, large-leaved species second clade within the trichotomy includes Amblyo- of Inuleae-Inulinae, Telekia is readily recognized by its carpum, Carpesium, Chrysophthalmum, Inula, Pentane- paleate receptacles andlack of a pappus. In our analysis ma, and Telekia. This clade has Buphthalmum-Xerolekia Telekia is closest to the morphologically derived genus as sister group, but only with low support. The third Carpesium; both are epappose but otherwise very clade in the trichotomy includes the genera Anvillea, different. Carpesium differs from all other Inuleae in Asteriscus, Ighermia,andPallenis, where Anvillea is its capitulum morphology andthe cypsela shape, anda sister to the others. These genera all have paleate position near Inula or Telekia is not self-evident. receptacles, andexcept for Anvillea have a pappus of However, cypselas of Carpesium have one large, large scales only. Two species of Pulicaria, i.e. P. characteristic, elongatedcrystal in each epidermiscell, canariensis and P. dysenterica, group together with just like the other Inuleae s.str., andits ndhF sequence Jasonia and Dittrichia, with P. dysenterica sister to also has their characteristic CCT insertion in the same Jasonia+Dittrichia. All except for Dittrichia have a position. The sister relationship with Telekia is surpris- double pappus consisting of both bristles and scales. ing, but there are a few morphological character states The pappus in Dittrichia consists of barbellate bristles in support of this, e.g. long rod-shaped cypselas and the unitedat the base in a cupule-like structure that is lack of a pappus. Anderberg (1991a) hadhypothesized unique in the tribe. Its ellipsoidcypselas are, however, that the long cypselas andlack of a pappus were similar in shape to those of most taxa of Pulicaria, and synapomorphies uniting Carpesium and Amblyocarpum, the fruits also have a collar of glandular hairs distally. but this postulatedrelationship does not receive any support here, although both genera are part of the same larger clade, and Amblyocarpum seems to be a derived Paraphyly of Inula and Pulicaria epappose relative of some Inula species. Except for Inula peacockiana and I. oculus-christi, the Anderberg (1991a) put forwardthat Inula and species of Inula belong to the same (99%-jackknife) Pulicaria were likely to be paraphyletic andmorpholo- clade, but this clade also includes Chrysophthalmum gically heterogeneous assemblages, andthat more montanum, Pentenema glanduligerum,andAmblyocar- detailed analyses would lead to a new circumscription pum inuloides. Within the larger (86%) Inula clade, the of these genera. Earlier, Dittrichia hadbeen separated genera Pentanema and Chrysophthalmum also seem to from Inula, and Anderberg demonstrated Duhaldea to be paraphyletic. Pentanema is heterogenous andis be only distantly related to Inula, andmore closely merely diagnosed by having few pappus bristles, which relatedto Blumea. The simple, bristly pappus in seems to be a technical character. The dwarf P. combination with the absence of more specific char- glanduligerum andthe perennial herb P. alanyense do acteristics of other genera has been the only feature not group with each other, the latter insteadforms a diagnosing Inula, andAnderberg proposedthat several clade with Chrysophthalmum gueneri. Duman and smaller genera couldhave their closest relatives within Anderberg (1999) haddoubts about the generic assign- Inula. Likewise, the double pappus with bristles and ment of P. alanyense when the species was described. scales is the only characteristic diagnosing Pulicaria,if Due to its few pappus bristles andlanate resting buds the smaller andmore derivedgenera with the same kind the choice of genus was Pentanema, a decision also of pappus structure are kept separate. In the present influencedby the assumption that Inula, the alternative study we found that all the species of Inula belong to the possibility, was heterogeneous andparaphyletic. Penta- same clade, but that other genera are also part of this nema alanyense and Chrysophthalmum gueneri show clade, i.e. Amblyocarpum, Carpesium, Chrysophthalmum, some similarities in vegetative morphology, but are Pentanema p.p., and Telekia. All taxa belonging to this distinct in capitula morphology and floral structure. The clade have a pappus of bristles only, or are completely former has epaleate capitula anda pappus of capillary devoid of pappus bristles. They may be with or without bristles, the latter is paleate without a pappus. The paleae, but most taxa are epaleate. At the base of the possibility of DNA contamination is ruledout because Inula clade is a pentatomy including Inula (Codonoce- the Pentanema alanyense DNA was extractedand phalum) peacockiana, I. oculus-christi, a well-supported sequencedlong before the work on Chrysophthalmum (98% jackknife) group with Chrysophthalmum gueneri gueneri was initiated. An interesting parallel is that the and Pentanema alanyense, another robust (98%) group two species of Chrysophthalmum included in the analysis with Carpesium and Telekia, anda 99%-jackknife clade do not form a monophyletic group, as the second species including Amblyocarpum, Chrysophthalmum montanum, (C. montanum) is closer to a group of Inula species Pentanema glanduligerum, andfive species of Inula. than to C. gueneri. The small genus Chrysophthalmum Francisco-Ortega et al. (2001) founda close relationship is diagnosed by the lack of a pappus and by between Telekia and Inula , andthis is hardly paleate capitula, but both these character states are ARTICLE IN PRESS A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146 143 homoplasious, as evidenced by our results. Considering species we have included shows that the circumscription also that the epappose genera Amblyocarpum, Carpe- of many of its genera may have to be revisedto obtain sium, and Telekia are part of the Inula clade, where monophyletic genera. A more detailed study of the members are normally characterizedby a pappus of phylogeny of Inuleae-Inulinae is another ongoing part capillary bristles, parallel losses or gains of pappus of our present research on the Asteraceae (Englundand bristles seem to have occurred. The absence or presence Anderberg in prep.). of paleae is evidently a highly homoplasious character in Inuleae s.str. The secondlarge genus of Inulinae, Pulicaria, also is Excluded genera paraphyletic as presently circumscribed, with P. crispa more closely relatedto Rhanterium, Vierea, and Schizogyne than to P. dysenterica and P. canariensis. Callilepis DC. and Zoutpansbergia Hutch With a different sample of taxa, Francisco-Ortega et al. These two South African genera, nowadays generally (2001) also presenteda tree topology where Pulicaria treatedas congeners, have been associatedwith Inuleae species were found to belong in two different clades, but (Merxmuller et al., 1977). Our present results show that the bootstrap support for these clades was less than ¨ Callilepis and Zoutpansbergia are sisters (Fig. 1), andthe 50%, andthe nodeswere saidto collapse in the strict notion that the latter may be a derived Callilepis has consensus tree. Pulicaria crispa sometimes is placedin thus gainedfurther support. Their styles are inuloid,but the genus Franceouria together with a few similar their strange cypselas andpaleae are not matchedby any species. Our present results indicate that recognition of Franceouria for P. crispa may be warranted. Except for taxon in Inuleae, and Leins (1971) showedthat these taxa have a type of pollen wall different from that found not having a pappus of bristles andscales, Schizogyne in, e.g., Inula and Pluchea. The pollen wall is more resembles some species of Pulicaria subg. Platychaete, similar to that in Gnaphalieae; Callilepis (incl. Zout- thus it will be interesting to learn the relationships of pansbergia) was included in an informal basal complex that yet unsampledsubgenus when more databecome of the Gnaphalieae by Anderberg (1991c). At present available. there is nothing that wouldindicatethat Callilepis and Like the Inula clade, the taxa of the Pulicaria- Zoutpansbergia are closer to the Inuleae clade than to Asteriscus clade may be with or without paleae. Eldena¨ s et al. (1998) foundmultiple losses of paleae to be less the Heliantheae s. ampl. (incl. Athroismeae and Eupatorieae), or that insteadthey are sister to both parsimonious than multiple independent gains. This these clades. The position of Callilepis-Zoutpansbergia is interpretation is supportedby the variable shape of the not clear, andfuture research basedon a combination of paleae, which ranges from robust structures subtending sequences from several genes may shedmore light on the florets to thin, linear, bristle-like structures between this issue. For the time being we therefore treat Callilepis the florets. and Zoutpansbergia as unclassifiedto tribe, pending Anvillea, which is epappose but with a paleate more conclusive results on their relationship. , is sister to Asteriscus, Ighermia, and Pallenis, all of them genera with paleate receptacles anda pappus of scales only. The only other two taxa of Inuleae- Gymnarrhena Desf. Inulinae with the same combination of character states are Xerolekia and Buphthalmum. The former has been This genus, formerly associatedwith Inuleae, was associatedwith Telekia, but Telekia differs in having included in Cichorioideae but not assigned to tribe by linear, subulate paleae, a less developed scaly pappus Bremer (1994). It is a morphologically derived, stemless that is often missing, andlong, branchedanther tails. taxon provided with glomerules of heterogamous The opinion that Xerolekia is relatedto Telekia is capitula bearing filiform female florets andfunctionally rejected. Both Xerolekia and Buphthalmum have folded male central florets. The male florets have an undivided paleae andalmost ecaudateanthers. In contrast to style andno visible stigmatic surface, vestigial cypselas, results of an earlier analysis of morphological data anda pappus of ciliate scales. The capitula structure of (Anderberg, 1991a), now there is goodsupport that Gymnarrhena is complicated, and its morphology has Buphthalmum speciosissimum (Xerolekia) is sister to made the genus difficult to place, except for the Buphthalmum, andtherefore the two taxa shouldbe indication that Gymnarrhena was misplacedin the treatedas congeners again. These genera are, however, vicinity of Inuleae, with pollen suggesting a position not closely relatedto Asteriscus andits allies, but near Cardueae (Skvarla et al., 1977). Apart from pollen, together form a small monophyletic group that is sister the calcarate anthers are almost the only feature that to the clade comprising Inula andits closest relatives. indicates a relationship with the Cichorioideae tribes. In The relationships among the taxa of Inuleae-Inulinae our analysis, Gymnarrhena was foundto be one branch still are not fully understood, but the small sample of of a polytomy also including Lactuceae, Arctoteae, ARTICLE IN PRESS 144 A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146

Liabeae, Vernonieae, andthe subfamily Asteroideae.In Callilepis salicifolia: South , Bayer SAF-01009 other words, the position of Gymnarrhena in relation to (S); [AY780851]. the four tribes and Asteroideae was unresolved. It is Calostephane marlothiana: , Merxmu¨ ller & either part of the unresolvedbasal complex of the Giess 28120 (S); [AY780819]. paraphyletic Cichorioideae, or sister to the entire Carpesium divaricatum: Taiwan, Chung & Anderberg Asteroideae. Our results are implicitly similar to those 1422 (S); [AY780820]. of Panero andFunk (2002) , who found Gymnarrhena to Chrysophthalmum gueneri: , Duman 7072 (S); be even more basal – sister to both the Asteroideae and [AY780848]. the LALV-complex – andhence proposedit as the sole Coleocoma centaurea: Australia, Albrecht 10563 (S); member of subfamily Gymnarrhenoideae. [AY780852]. Cratystylis conocephala: Australia, Nordenstam & Anderberg 604 (S); [AY780821]. Acknowledgements Dielitzia tysonii: Australia, Nordenstam & Anderberg 394 (S); [AY780822]. Dittrichia viscosa: , Trift et al. 22 (S); Financial support was receivedas a grant (to A.A.) [AY780823]. from the Swedish Research Council. We are grateful to Doellia bovei: , Kilian & Hein NK 4724 (S); Arun K. Pandey for providing material of Caesulia [AY780858]. axillaris, to Dave Albrecht for help with obtaining Duhaldea cappa: , Luo 0253 (HAST); material of Coleocoma, to Mats Thulin for material of Pluchea, to Alexander Sennikov for providing material [AY780827]. Epaltes cunninghamii: Australia, Nordenstam & An- of Amblyocarpum and Karelinia, to Marinda Koeke- derberg 972 (S); [AY780824]. moer for material of Zoutpansbergia caerulea, to Javier Gymnarrhena micrantha: , Anderberg 1326 (S); Francisco-Ortega for providing a Sachsia anda [AY780825]. Rhodogeron sequence, to the US National Plant Inula bifrons: France, Barbezat s. n. (S); [AY780840]. Germplasm System (NPGS) for providing seeds of Inula britannica: Sweden, Svensson AS02100 (S); Stenachaenium, andto Bodil Cronholm andMaria [AY780826]. Arvidsson for laboratory assistance. The curators of the Inula conyza herbaria B, CANB, HAST, andTEX are thankedfor : Bulgaria, Anderberg & Anderberg B01- 24 (S); [AY780841]. allowing sampling from herbarium specimens. We are Inula germanica: Macedonia, Segelberg 30661/22 (S); indebted to Bruce Baldwin and one anonymous reviewer [AY780842]. for useful comments on the manuscript. Inula peacockiana: , Rechinger & Rechinger 4956 (S); [AY780839]. Inula verbascifolia ssp. methanea: Greece, Rechinger Appendix A 20377 (S); [AY780843]. Iphionopsis rotundifolia: , Thulin & Warfa Taxa investigated, vouchers sequenced, and new 5914 (UPS); [AY780853]. GenBank accession numbers [AY780811–AY780873] Jasonia tuberosa: Spain, Dittrich 3247 (S); [AY780844]. Amblyocarpum inuloides: Azerbejdjan, Bobrov & Karelinia caspia: , Nikulina s.n. (LE); Tzvelev 194 (LE); [AY780849]. [AY780854]. Antiphiona pinnatisecta: Namibia, Nordenstam & Laggera decurrens: Yemen, Kilian et al. NK 4562 (S); Lundgren 678 (S); [AY780811]. [AY780861]. Anvillea garcinii ssp. radiata: , Anderberg 576 Laggera pterodonta: Yemen, Kilian & Hein NK 4705 (S); [AY780812]. (S); [AY780862]. Blumea balsamifera: Taiwan, Peng & Chen 10780 Mollera angolensis: Zambia, Robinson 6557 (B); (HAST); [AY780814]. [AY780863]. Blumea integrifolia: Australia, Western Australia, Ondetia linearis: Namibia, Emanuelsson 1021 (S); Telford6307 (CANB); [ AY780813]. [AY780828]. Blumea psammophila: Australia, Northern Territory, Pechuel-loeschea leubnitziae: Namibia, Bremer 455 Wightman & Dunlop 1287 (CANB); [AY780816]. (S); [AY780829]. Blumea riparia: Taiwan, Peng et al. 15102 (HAST); Pegolettia lanceolata: Swaziland, McCallum & Balk- [AY780815]. will 186 (B); [AY780860]. Blumea saxatilis: Australia, Northern Territory, Cow- Pegolettia oxyodonta: Namibia, Emanuelsson 1032 ie 6734 (CANB); [AY780817]. (S); [AY780830]. Caesulia axillaris: , Pandey 3021 (S); Pentanema alanyense: Turkey, Duman 5504 (S); [AY780818]. [AY780831]. ARTICLE IN PRESS A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146 145

Pentanema glanduligerum: , Rechinger Anderberg, A.A., 1986. The genus Pegolettia (Compositae, 18595 (S); [AY780832]. Inuleae). Cladistics 2, 158–186. Pluchea arabica: Yemen, Thulin et al. 9555 (UPS); Anderberg, A.A., 1989. Phylogeny and reclassification of the [AY780864]. tribe Inuleae (Asteraceae). Can. J. Bot. 67, 2277–2296. Pluchea dentex: Australia, New South Wales, Craw- Anderberg, A.A., 1991a. and phylogeny of tribe ford1076 (CANB); [ AY780833]. Inuleae (Asteraceae). Plant. Syst. Evol. 176, 75–123. Pluchea dioscorides: Eritrea, Ryding & Ermias 1279 Anderberg, A.A., 1991b. Taxonomy and phylogeny of the tribe Plucheeae (Asteraceae). Plant. Syst. Evol. 176, 145–177. (UPS); [AY780855]. Anderberg, A.A., 1991c. Taxonomy and phylogeny of the tribe Pluchea dunlopii : Australia, Western Australia, Hun- Gnaphalieae (Asteraceae). Opera Bot 104, 1–195. ger & Kilian 3948 (B); [AY780867]. Anderberg, A.A., 1994. Tribe Plucheeae. In: Bremer, K. (Ed.), Pluchea indica ssp. yemensensis: Yemen, Kilian et al. Asteraceae. Cladistics & Classification. Timber Press, NK 4601 (S); [AY780871]. Portland, pp. 292–303. Pluchea kelleri: Somalia, Thulin & Mohamed6953 Anderberg, A.A., 1995. Doellia, an overlookedgenus in the (UPS); [AY780872]. Asteraceae-Plucheeae. Willdenowia 25, 19–24. Pluchea littoralis: Somalia, Thulin et al. 10634 (UPS); Anderberg, A.A., Karis, P.O., El-Ghazaly, G., 1992. Cratys- [AY780856]. tylis, an isolatedgenus of the Asteraceae-Cichorioideae. Pluchea lucens: Somalia, Thulin 10775 (UPS); Aust. Syst. Bot. 5, 81–94. [AY780865]. Baldwin, B.G., Wessa, B.L., 2000. Phylogenetic placement of Pluchea nogalensis: Somalia, Thulin & Warfa 6150 Pelucha andnew subtribes in sensu stricto (UPS); [AY780868]. (Compositae). Syst. Bot. 25, 522–538. Pluchea obovata: Yemen, Socotra, Thulin & Gifri Bayer, R.J., Cross, E.W., 2003. A reassessment of tribal affinities of Cratystylis and Haegiella (Asteraceae) basedon 8831 (UPS); [AY780857]. three chloroplast DNA sequences. Plant Syst. Evol. 236, Pluchea sarcophylla: Somalia, Thulin et al. 10040 207–220. (UPS); [AY780866]. Besold, B., 1971. Pollenmorphologische Untersuchungen an Pluchea tetranthera: Australia, New South Wales, Inuleen (Angianthinae, Relhaniinae, Athrixiinae). Diss. Lyne 1797 (CANB); [AY780834]. Bot. 14. J. Cramer, Lehre. Pluchea tetranthera: Australia, Western Australia, Bremer, K., 1987. Tribal interrelationships of the Asteraceae. Hunger & Kilian 3797 (B); [AY780873]. Cladistics 3, 210–253. Pluchea yucatanensis: USA, Mississippi, Jones & Bremer, K. (Ed.), 1994. Asteraceae. Cladistics & Classification. Jones 12656 (TEX); [AY780870]. Timber Press, Portland. Pseudoconyza viscosa: Yemen, Kilian et al. NK 4607 Doyle, J.J., Doyle, J.S., 1987. A rapidDNA isolation (S); [AY780859]. procedure for small quantities of fresh leaf tissue. Phyto- Pulicaria crispa: Iran, Karis 734 (S); [AY780845]. chem. Bull. 19, 11–15. Pulicaria dysenterica: Bulgaria, Anderberg & Ander- Duman, H., Anderberg, A.A., 1999. An undescribed species of Pentanema berg B01-23 (S); [AY780846]. Cass. (Asteraceae-Inuleae) from Turkey, with notes on the phylogenetic status of the genus. Bot. J. Linn. Rhanterium epapposum: , Nilsson et al. 16436 Soc. 129, 333–338. (S); [AY780847]. Eldena¨ s, P., Anderberg, A.A., Ka¨ llersjo¨ , M., 1998. Molecular Sachsia polycephala: , Gutie´rrez & Nilsson 6 (S); phylogenetics of the tribe Inuleae s. str. (Asteraceae), based [AY780835]. on ITS sequences of nuclear ribosomal DNA. Plant Syst. Stenachaenium campestre: , Myndel Peder- Evol. 210, 159–173. sen 3014 (S); [AY780836]. Eldena¨ s, P., Ka¨ llersjo¨ , M., Anderberg, A.A., 1999. Phyloge- Stenachaenium campestre: US National Plant Germ- netic placement andcircumscription of tribes Inuleae s.str. plasm System (NPGS), Accession No. PI 312840; origin andPlucheeae (Asteraceae): evidencefrom sequences of (S); [AY780869]. chloroplast gene ndhF. Mol. Phyl. Evol. 13, 50–58. Telekia speciosa: Sweden, Thora´n 4100 (S); Farris, J.S., 1997. ‘‘Xac.’’ Computer program andmanual. [AY780838]. Swedish Museum of Natural History, Stockholm. Tessaria integrifolia: , Daly et al. 6392 (S); Farris, J.S., Albert, V.A., Ka¨ llersjo¨ , M., Lipscomb, D., Kluge, [AY780837]. A.G., 1996. Parsimony jackknifing outperforms neighbor- joining. Cladistics 12, 99–124. Zoutpansbergia caerulea: , Koekemoer Francisco-Ortega, J., Park, S.-J., Santos-Guerra, A., Benabid, 2259 (PRE); [AY780850]. A., Jansen, R.K., 2001. Origin andevolution of the endemic Macaronesian Inuleae (Asteraceae): evidence from the internal transcribedspacers of nuclear ribosomal DNA. References Biol. J. Linn. Soc. 72, 77–97. Hall, T.A. (Ed.), 1999. BioEdit: a user-friendly biological Anderberg, A.A., 1985. The genus Iphionopsis (Compositae- sequence alignment editor and analysis program for Inuleae). Nord. J. Bot. 5, 51–56. Windows 95/98/NT, Nucl. Acids. Symp. Ser. 41, 95–98. ARTICLE IN PRESS 146 A.A. Anderberg et al. / Organisms, Diversity & Evolution 5 (2005) 135–146

Hoffmann, O., 1889. Inuleae-Angianthinae. In: Engler, A., Pandey, A.K., Jha, S.M., Dhakal, M.R., 2000. Seed coat, Prantl, K. (Eds.), Die natu¨ rlichen Pflanzenfamilien IV(5). pericarp andpseudopericarpin Caesulia axillaris Roxb. Engelmann, Leipzig, pp. 192–195. (Asteraceae). Compositae Newslett 35, 37–43. Hong, L., Trusty, J., Oviedo, R., Anderberg, A., Francisco- Panero, J.L., Funk, V.A., 2002. Towarda phylogenetic Ortega, J., 2004. Molecular phylogenetics of the Caribbean subfamilial classification for the Compositae (Asteraceae). genera Rhodogeron and Sachsia (Asteraceae). Int. J. Plant Proc. Biol. Soc. Wash. 115, 909–922. Sci. 165, 209–217. Saghai-Maroof, M.A., Soliman, K.M., Jorgensen, R.A., Ka¨ llersjo¨ , M., Bergqvist, G., Anderberg, A.A., 2000. Generic Allard, R.W., 1984. Ribosomal DNA spacer-length poly- realignment in primuloidfamilies of the Ericales s.l.: a morphisms in barley: Mendelian inheritance, chromosomal phylogenetic analysis basedon DNA sequences from three location and population dynamics. Proc. Natl. Acad. Sci. chloroplast genes andmorphology. Am. J. Bot. 87, 1325–1341. USA 81, 8014–8018. Karis, P.O., Ka¨ llersjo¨ , M., Bremer, K., 1992. Phylogenetic Skvarla, J.J., Turner, B.L., Patel, V.C., Tomb, A.S., 1977. analysis of the Cichorioideae (Asteraceae), with emphasis Pollen morphology in the Compositae andin morphologi- on the Mutisieae. Ann. Missouri Bot. Gard. 79, 416–427. cally relatedfamilies. In: Heywood,V.H., Harborne, J.B., Kim, K.-J., Jansen, R.K., 1995. ndhF sequence evolution and Turner, B.L. (Eds.), The Biology and Chemistry the major clades in the sunflower family. Proc. Natl. Acad. of the Compositae 1. Academic Press, London, Sci. USA 92, 10379–10383. pp. 141–265. King-Jones, S., 2001. Revision of Pluchea Cass. (Compositae, Staden, R., Beal, K., Bonfield, J.K., 1998. The Staden package. Plucheeae) in the OldWorld.Englera 23, 1–136. In: Misener, S., Krawets, S.A. (Eds.), Bioinformatics Leins, P., 1971. Pollensystematische Studien an Inuleen. I. Methods and Protocols, Computer Methods in Molecular Tarchonanthinae, Plucheinae, Inulinae, Buphthalminae. Biology 132. Human Press, Totowa, pp. 115–130. Bot. Jahrb. Syst. 91, 91–146. Thulin, M., 2003. Notes of Pluchea (Asteraceae) in Somalia Merxmu¨ ller, H., Leins, P., Roessler, H., 1977. Inuleae- andEthiopia. Nord.J. Bot. 22, 659–666. systematic review. In: Heywood, V.H., Harborne, J.B., Zdero, C., Bohlmann, F., Haegi, L., King, R.M., 1988. Turner, B.L. (Eds.), The Biology and Chemistry of the Eudesmanolides and other constituents from Cratystylis Compositae 1. Academic Press, London, pp. 577–602. conocephala. Phytochemistry 27, 865–868.