Plant P1. Syst. Evol. 184:259-283 (1993) Systematics and Evolution © Springer-Verlag I993 Printed in Austria A cladistic analysis of the tribe Astereae (Asteraceae) with notes on their evolution and subtribal classification ZHANG XIAOPING and K&RE B~EMER Received April 22, 1992; in revised version September 8, 1992 Key words: Angiosperms, Asteraceae, Astereae.- Cladistics, evolution, phylogeny, classification. Abstract: The Astereae were surveyed and the genera arranged in 23 informal groups. The generic groups were used to sample representative genera for a cladistic analysis based on morphological characters. The resulting cladogram was used for discussion of evolution and subtribal classification within the tribe. The lower basic chromosome numbers x = 4, 5, 6, and 8 are interpreted as reductions from a primitive x = 9. The subtribe Grangeinae occupies a phylogenetically basal position as sister group to the rest of the tribe. This may be divided into two large groups, largely corresponding to the homochromous Solidagininae and to the heterochromous Asterinae sensu lato, i.e. including the Bellidinae, Hinterhuberinae, Conyzinae, and Baccharidinae. The latter four subtribes are derived within the Asterinae, and hence reduced to synonymy. Several intercontinental relationships indicate that a geographical subdivision of the tribe should be avoided, although in our analysis most of the groups proved to be restricted to one of five major regions. Several of the major tribes of the Asteraceae have recently been reclassified at the subtribal level, for example, Heliantheae sensu lato (ROBINSON 1981), Senecioneae (JEFFREY & CHEN 1984), Eupatorieae (KING & ROBINSON 1987), Inuleae sensu lato (ANDEe.BERG 1989, 1991 a, b, c), and Anthemideae (BREMER & HUMeHRmS 1993). In contrast, cladistic interrelationships within the Astereae, a worldwide tribe with c. 170 genera and more than 2800 species, remain badly understood. Although m u c h progress has been made in the generic classification, there have been few attempts towards a revised subtribal classification. The phylogeny o f the tribe is u n k n o w n and intercontinental interrelationships between genera and groups of genera are largely unclear. The old subtribal classification of the Astereae has been rejected as largely artificial, and generic interrelationships are today investigated according to a geographical rather than a taxonomic subdivision of the tribe, following GRAU (1977 a). Most workers seem to consider the tribe regionally, usually from one continent. There is a need for a more complete analysis of the Astereae, with new hypotheses of generic interrelationships between groups from different continents. 260 ZHANG XIAOPING & K. BREMER: We present a cladistic analysis of generic interrelationships throughout the tribe. We make provisional generic groupings in order to select representative genera from all areas and we use classical morphological characters as well as micromorphological characters partly investigated by ourselves. We do not attempt to solve all generic relationships within the tribe and we do not attempt to cover all relevant morphological information, but we present the result of our analysis as a hypothesis for further investigations of evolution and phylogeny within the Astereae. Review of subtribal classification The existing subtribal classification was provided by BENTHAM(1873 a) and adopted almost unchanged by HOFFMANN (1890), except for the names of the subtribes (Table 1). BENTHAM (1873b) noted that the tribe was "not being divisible into distinct subtribes", so the subtribal classification that he actually proposed must be regarded as provisional. BENTHAM (1873 a) distinguished six subtribes. Most genera are classified either in the Homochromeae (Solidagininae) or the Heterochromeae (Asterinae), depending on ray floret colour, either yellow as in the disc florets, or white-blue-violet-purple contrasting with the yellow disc florets. On the face of it, this appears to be purely artificial. Yet, in the present paper we shall reach the conclusion that the homochromous/heterochromous division of the tribe is not such a bad disposition as generally assumed, although there are many exceptions to the colour character and many necessary generic rearrangements to be made. The Grangeinae are characterized by disciform capitula, with outer filiform female florets in several series. In this character they are similar to the Conyzinae, but distinct from that subtribe in the reduced pappus. The Grangeinae are usually considered a well-characterized and isolated group within the Astereae (FAYED 1979). As we shall see below, in our selected cladogram it assumes the position as sister group to the rest of the tribe. The Bellidinae are characterized by a reduced pappus, which is absent or reduced to a minute crown or a few short bristles. Several of the Bellidinae genera indeed seem closely related to each other. The subtribe appears to be derived within the Asterinae (DE JoNa 1965), and its retention would probably leave the Asterinae as an unnatural, paraphyletic grade. The same certainly applies to the remaining two subtribes, the Conyzinae and the Baccharidinae. The Baccharidinae presumably have their sister group in the Table 1. Subtribal classification of the Astereae from BENTHAM (1873 a) and HOFFMANN(1890) BENTHAM 1873 HOFFMANN 1890 Homochromeae BENTH. Solidagininae HOFFM. Grangeinae BENTH. Bellidinae BENTH. Asterinae DUMORT. Conyzinae BENTH. Baccharidinae LESS. Grangeineae BENTH. Bellideae BENTH. Heterochromeae BENTH. Conyzeae BENTH. Baccharideae LESS. Cladistic analysis of the Astereae 261 Conyzinae and together the two subtribes have their sister group in the Erigeron complex of the Asterinae. The Conyzinae are distinguished from the Asterinae by their disciform capitula, with outer filiform female florets in several series. The Baccharidinae are further specialized in their frequently dioecious capitula, also with filiform female florets. This latter subtribe is certainly monophyletic (HELLWIa 1990), but its retention would leave the Conyzinae and Asterinae as unnatural, paraphyletic grade groups. GRAU (1977 a) stressed the unnatural distinctions between the subtribes. He pointed out the many exceptions to the homochromous and heterochromous conditions present within genera, obscuring the distinction between BENTHAM'S tWO major subtribes Homochromeae (Solidagininae) and Heterochromeae (Asterinae). GRAU also stressed, as did BENTHAM himself as well as several later authors (e.g., CRONQUIST 1947), the gradual transition from Aster through Erigeron to Conyza, obscuring the distinction between the Asterinae and the Conyzinae. Two of the subtribes were considered natural groups by GRAU, the Grangeinae and the Baccharidinae. Because of the problems in delimiting the other subtribes, GRAU appealed for new characters, and pending more information he abandoned the subtribal classification altogether and arranged the genera geographically in his review of the tribe. A seventh subtribe, the Hinterhuberinae, was proposed by CUATRECASAS(1969, 1989). The genus Hinterhubera has disciform capitula, with distinctly lobed outer female florets. In its disciform capitula it is superficially similar to the Conyzinae, but its lobed outer florets differ from the filiform outer florets of that subtribe. Hence CUATRECASAS(1969) placed it in a subtribe of its own. Later CUATRECASAS (1989) emended the Hinterhuberinae to include also Blakiella, Westoniella, and his new genus Flosmutisia. He had earlier (CUATRECASAS1977) suggested a relationship between Blakiella and Westoniella. The four genera of the Hinterhuberinae are characterized by glandular corollas and functionally male disc florets. The Hinterhuberinae were proposed without any consideration of subtribal classification in general. Many small subtribes could be proposed for similar reasons, yet hardly improving our understanding of intergeneric relationships within the tribe. Recently NESOM (1991 i) has proposed a very narrow delimitation of the Solidagininae, essentially comprising the Solidago, Petradoria, and Gutierrezia groups (to be discussed below). However, NESOM'S study concentrated on American genera, and we have attempted to evaluate subtribal classification of the entire tribe. Selection of taxa We surveyed all generally accepted genera of the Astereae, and tried to identify their alleged close relatives, as proposed by earlier authors. Subsequently, we grouped most of the genera in 23 informal generic groups, listed in Table 2. In addition, there are 17 isolated genera, which could not unequivocally be assigned to any of the generic groups. Most of the generic groups are presumably monophyletic, though there are probably several exceptions. Some of the groups, e.g., the Aster group, may be paraphyletic. However, the generic groups are not intended as a formal classification. We present them only as an attempt to an overview of the entire tribe and as a basis for sampling genera for the cladistic analysis. In order to facilitate the review of the generic groups we have arranged them geographically below and in Table 2. We applied no geographical constraints in distinguishing the groups, and were careful to look for possible relatives on other continents. Nevertheless, Table 2. Informal generic groups in the Astereae, geographically arranged, with references to major revisions. The positions of the isolated genera are unknown to us Eurasian and widespread genera Aster group: Arctogeron, Aster (JONES 1980, JONES& SMOGOR1983, JONES& YOUNG 1983), Asterothamnus, Callistephus, Doellingeria, Galatella, Gymnaster, Heteropappus, Kalimeris, Krylovia, Linosyris, Tripolium, Turczaninowia Erigeron-Conyza group: Brachyactis, Chamaegeron, Conyza (NEsoM 1990C), Erigeron (CRONQUIST1947; SOLBRIG1960, 1962; NESOM1989 a, c), Heteroplexis, Lachnophyllum, Psychrogeton (GRIERSON1967), Trirnorpha (NEsoM 1989b) Bellis group: Bellis, Bellium (SIMON 1978), Rhynchospermum Isolated genera: Formania, Nannoglottis (LING & CHEN 1965), Thespis, Tolbonia North American and pan-American genera Erigeron-Conyza group: Archibaccharis (JACKSON1975, NESOM 1991 g), Baccharis (BLAKE 1924, HELLWIG1990) Boltonia group: Aphanostephus (TURNER 1984), Astranthium (DE JONG 1965), Boltonia (ANDERSON1987), Dichaetophora, Egletes (SHINNERS1949), Townsendia(BEAMAN1957) Chaetopappa group: Chaetopappa (NESOM 1988 a), Pentachaeta (VANHORN 1973), Rigiopappus (RoBTNSON& BRETTELL1973 a), Tracyina Chrysopsis group (NESOM 1991 f): Chrysopsis (SEMPLE& al. 1980, SEMPrE 1981, NESOM 1991 b), Croptilon (SMITH1965, 1981), Heterotheea (SEMPLE& al. 1980, NESOM 1990 d), Osbertia (TURNER & SUNDBERG 1986, NESOM 1991 e), Pityopsis (SEMPLE& al. 1980, SEMELE& BOWERS1985), Tomentaurum Corethrogyne group: Corethrogyne, Lessingia (HowELL 1929), Machaeranthera (TURNER 1987, HARTMAN1990), Xylorhiza (WATSON 1977) Erieameria group: Chrysothamnus (HALL & CrEMENTS 1923, ANDERSON1966), Ericameria (NEsoM 1990 a) Grindelia group: Grindelia (STEYERMARr¢1934), Isoeoma (NEsoM 1991 a), Olivaea (DE JONG & BEAMAN1963), Prionopsis, Stephanodoria, Xanthoeephalum (LANE 1983) Gutierrezia group (LANE 1985, NESOM 1991 i): Amphiachyris (LANE 1979), Gundlachia, Gutierrezia (LANE 1985, SOLBR~G1966), Gymnosperma, Thurovia Haplopappus group: Hazardia (CLARK 1979), Oonopsis, Pyrroeoma Petradoria group: Acarnptopappus (LANE 1988), Amphipappus (PORTER 1943), Columbiadoria, Eastwoodia, Hesperodoria (NESOM1991 h), Petradoria (ANDERSON1964), Stenotus (NEsoM 1989 d), Tonestus (NESOM & MORGAN 1990, NESOM 1991 C), Vanelevea (ANDERSON& WEBERG 1974) Solidago group: Bigelowia (ANDERSON1970, 1972, 1977), Euthamia (SmREN1981), Solidago, Xylothamia Isolated genera: Benitoa, Chloracantha (SUNDBERG1991), Chrysoma, Geissolepis (ROBINSON & BRETTELL1972), lonactis, Monoptilon, Xanthisma South and Central American genera Erigeron-Conyza group: Conyzanthus, Darwiniothamnus, Heterothalamus, Hysterionica (CABRERA 1946, ARIZAESPINAR1980), Laennecia (ZARDXNI1981, NESOM 1990b), Oritrophium, Sommerfeltia (CHEBATAROFF1981) Brachyscome group: Floscaldasia, Laestadia, Plagiocheilus (ROBINSON& BRETa'ELL1973 b) Chiliotrichum group (CABRERA 1954): Aylacophora, Chiliophyllum, Chiliotrichum, Chiliotrichiopsis, Diplostephium, Lepidophyllum, Nardophyllum (CABRERA1954), Paleaepappus, Parastrephia Chrysopsis group: Noticastrum (ZARD~NI1985) Haplopappus group: Haplopappus (HALL 1928, BROWN& CLARK 1982), Llerasia (CuATRECASAS 1970) Hinterhubera group (CuATRECASAS1989): Blakiella, Flosmutisia, Hinterhubera (CABRERA 1954), Microgynella, Podocoma, Westoniella Zr~AN~ XIAOPIN~ & K. BREMER: Cladistic analysis of the Astereae 263 Table 2 (continued) African and tropical Asian genera Erigeron-Conyza group: Mieroglossa, Nidorella (WILD 1969), Psiadia, Psiadiella, Rochonia, Sarcanthemum Amellus group: Amellus (RoMMEL 1977, 1979), Chrysocoma (BAYER 1981), Felicia (GRAu 1973), Gymnostephium, Heteromma (HILLIARD & Bt:RTT 1973), Jeffreya, Mairia, Nolletia, Poecilolepis, Polyarrhena (GRAu 1970) Apodocephala group: Apodocephala, Vernoniopsis Commidendron group: Commidendron, Melanodendron Engleria group: Engleria, Pteronia (HuTcHINSON & PHILLIPS 1917) Grangea group (FAYED 1979): Ceruana, Colobanthera, Cyathocline, Dacryotrichia, Dichrocephala, Grangea, Grangeopsis, Grauanthus, Gyrodoma Isolated genera: Psednotrichia, Rhamphogyne Australasian and Pacific genera Brachyscome group: Brachyscome (DAvis 1948), Calotis (DAvis 1952), Erodiophyllum, Lagenifera (CABRERA 1966, CUATRECASAS 1982), Myriatics, Piora, Solenogyne (ADAMS 1979) Olearia group: Celmisia (GIVEN 1969, 1980, 1984), Damnamenia, Olearia, Paehystegia, Pleurophyllum, Remya (WAcNER & H~RBSa" 1987) Vittadinia group: Camptacra, Dichromochlamys, Dimorphocoma, Etachanthus, Eurybiopsis (BuRBIDG~ 1982), Ixiochlamys (DUNLOP 1980), Kippistia (LANDER & BARRY 1980 a), Minuria (LANDER & BARRY 1980 b), Tetramolopium (LOWREY 1986), Vittadinia (BURBIDCE 1982) Isolated genera: Achnophora, Apostates, Ceratogyne, Phacellothrix most of the groups proved to be restricted to one of five major regions (Table 2), although there are exceptions. For example, the Brachyscome group occurs in Australasia, the Pacific, and South America, and the Erigeron-Conyza group is almost worldwide. For the cladistic analysis we have sampled usually one genus from each generic group. The large Erigeron-Conyza group was represented by four genera, Erigeron, Conyza, Archibaccharis, and Psiadia. In general we have chosen the large core genera within each group, such as Aster, Erigeron, and SoIidago. The genus Aster was coded sensu stricto, allowing for variation only in its narrow circumscription. We avoided genera with disciform or discoid capitula, since they lack information on ray florets. Thus, Arehibaccharis (at least sometimes radiate) was sampled rather than Baecharis (always disciform or discoid), and Engleria (radiate) rather than Pteronia (discoid). Eurasian and widespread genera. The genus Aster is differently circumscribed in various floras. Some Old World floras adopt a narrow circumscription, and recognize several segregate genera (e,g., KOMAROV 1959, DAVIS 1975, RECHIN~ER 1982). In other floras the segregate genera are included in Aster (e.g., Tua-IN & al. 1976). We have assembled them together with Aster sensu stricto in the Aster group (Table 2). The genera Callistephus, Gymnaster, Heteropappus, and Kalimeris are closely related to each other as well as to Aster. Gymnaster differs from Kalimeris mainly by the absence of pappus, and the two genera are sometimes united. Heteropappus differs from Aster mainly by its zygomorphic disc florets. The large Erigeron-Conyza lineage comprises a major part of the tribe. A gradual trend from Aster with uniseriate long ray fiorets through Erigeron with pluriseriate rays to Conyza with more or less tubular outer female florets has been postulated by several authors (e.g., 264 ZHANG XIAOPING (~¢ K. BREMER: CRONQUIST 1947). Taken to its extreme, this lineage ends with the highly specialized subtribe Baccharidinae (Archibaccharis, Baccharis, Heterothalamus). We have included all relatives of Erigeron and Conyza, as well as the Baccharidinae within the Erigeron-Conyza group. Throughout the group the female florets are pluriseriate, whether radiate as in Erigeron or minutely radiate to tubular as in Conyza and the Baccharidinae. We have sampled four genera from the Erigeron-Conyza group, in order to test its possible monophyly. Asian members include Brachyactis (LIN~ & CHEN 1973), Chamaegeron, Lachnophyllum, and Psychrogeton (GRIERSON 1967), all close relatives of Erigeron. We add Heteroplexis to the group, since it has more female outer florets than hermaphroditic central florets. The northern hemisphere Trimorpha is usually included in Erigeron, but kept distinct by NESOM (1989 b). Strictly African genera of the Erigeron-Conyza group are discussed below. The Bellis group is not equivalent to the subtribe Bellidinae, but we restrict it to Bellis and Bellium, as well as to the possibly related Rhynehospermum. The sister genera Bellis and Bellium are seapose herbs with pauciseriate involucral bracts. Rhynchospermum has similar, compressed cypselas. North American and pan-American genera. Currently there is much research undertaken on North American Astereae, not least in generic delimitation and intergeneric relationships (e.g., NESOM 1990a-d, 1991a-i). We have tried to identify groups based on hitherto published information on morphological data. There is also a growing body of information on morphological data. There is also a growing body of information on generic interrelationships of North American Astereae based on chloroplast DNA restriction site analyses (e.g., SUH & SIMPSON 1990). However, hitherto published information on molecular data involves a very limited sample of species and we have excluded molecular data from our analysis. This will allow for a future independent molecular-based test of our morphologically derived classification. The Boltonia group was essentially recognized already by StuNNERS (1946), who indicated the relationship between Aphanostephus, Astranthium, Boltonia, Diehaetophora, and Egletes, on the basis of the conical receptacle which these genera have in common. Townsendia was added to this group by BEAMAN(1957). There seems to be general agreement about the close relationship between the four genera Aphanostephus, Astranthium, Dichaetophora, and Townsendia (DE JONG 1965, TURNER 1984). However, DE JON~ (1965) stated that Boltonia is more closely related to Erigeron and Aster. We keep Boltonia together with the other genera of this group, suggesting that the alleged affinity to Erigeron and Aster may be a case of symplesiomorphy. The position of Egletes is more uncertain. In his revision of the genus, SHINNERS(1949) removed Egletes from the Boltonia group, and placed it in subtribe Grangeinae (also DE JONC 1965). However, FAYED (1979) did not accept Egletes in his revision of that subtribe. Following FAYED, we provisionally retain Egletes in the Boltonia group, noting that it may be misplaced and could be a true member of the Grangeinae. The Chaetopappa group consists of Chaetopappa, Pentachaeta, Rigiopappus, and Tracyina. They are small, usually annual herbs with linear-oblanceolate leaves. Chaetopappa and Pentaehaeta were once united (KEcK 1958), but VAN HORN (1973) has convincingly demonstrated that the two genera are distinct. Although VAN HORN stressed the dissimilarities between Chaetopappa and Pentachaeta, he obviously did so to demonstrate their generic status rather than to disprove their relationship. Rigiopappus and Traeyina are closely related to each other as well as to Pentaehaeta (BLAKE 1937, VAN HORN 1973, ROBINSON 8¢ BRETTELL1973 a). In the Chrysopsis group, Chrysopsis, Heterotheea, and Pityopsis have usually been associated and even merged into a single genus, as reviewed by SEMPLE& al. (1980). SEMPLE & al. also concluded that Croptilon and Xanthisma are closely related to Chrysopsis. The Chrysopsis group was further emended by NESOM(1991 f) with the inclusion of Tomentaurum (NESOM 1991 d) and Osbertia (NEsoM 1991 e) and the exclusion of Xanthisma (NEsoM Cladistic analysis of the Astereae 265 1991 f). The South American genus Noticastrum also belongs in this predominantly North American group (TURNER & SUNDBERG 1986). NESOM (1991 f) listed several micromorphological characters supporting the Chrysopsis group as monophyletic. Corethrogyne and Lessingia of the Corethrogyne group seem to be sister genera, as noted by HOWELL (1929) and several later authors. Furthermore, the two genera are generally considered close relatives of Machaeranthera (HOWELL1929). Xylorhiza, the fourth member of our Corethrogyne group, has been resurrected from Machaeranthera (WATSON 1977). Despite their generic distinction, WATSONnoted that the two genera are close relatives. Xylorhiza has also been considered related to several sections of Haplopappus. Most of these sections are now removed from Haplopappus and included in Machaeranthera, however (TURNER 1987, HARTMAN 1990). There is a well supported close relationship between Ericameria and Chrysothamnus (ANDERSON 1964, NESOM 1990 a, NESOM & al. 1990), which together constitute the Ericameria group. Grindelia, Olivaea, and Xanthocephalum form the majority of the Grindelia group, characterized by resinous leaves and involucres. The close relationship between these genera has been convincingly demonstrated by DE JONG & BEAMAN(1963) and LANE (1982). To this group NESOM (1991 a) has added Isocoma, Prionopsis, and Stephanodoria. The six genera all have a base chromosome number of x = 6, and form NESOM'S (1991 a) "Xanthocephalum group". The Gutierrezia group consists of Amphiachyris, Gutierrezia, Gymnosperma, and Thurovia as circumscribed by LANE (1982) in her review of the "Xanthocephalum complex". These genera have rather small corymbose-paniculate capitula and a chromosome base number x = 4, 5, 6, or 8. NESOM (1991 i) expanded this group with Gundlachia, as well as Bigelowia, Euthamia, and XyIothamia to form his "Gutierrezia lineage". We retain the Gutierrezia group in LANE'S more restricted sense. As circumscribed by HALL (1928), Haplopappus is generally regarded as an unnatural, polyphyletic assemblage. Consequently, most of HALL'Ssections of Haplopappus have now been moved to separate genera (see NESOM & MORGAN 1990). We arrange most of them in other generic groups, whereas others are retained in the Haplopappus group. The latter include Hazardia, Llerasia (in South America), Oonopsis, and Pyrrocoma. Some of them are no doubt closely related to Haplopappus sensu stricto, i.e. Hazardia (CLARK 1979, BROWN & CLARK 1982), whereas others require further investigation; their placement with Haplopappus must be regarded as provisional. The Petradoria group was identified and characterized by NESOM (1991 h). It comprises Acamptopappus, Amphipappus, Columbiadoria, Eastwoodia, Hesperodoria, Petradoria, Stenotus, Tonestus, and Vanclevea (see also ANDERSON 1964, ANDERSON & WEBERG 1974). The Solidago group consists essentially of the large genus Solidago and a few satellite genera closely associated with Solidago and with each other, i.e. Bigelowia, Euthamia, and Xylothamia (ANDERSON 1970, SIEREN 1981, NESOM & al. 1990). They all have numerous small capitula in corymbs or elaborate panicles. NESOM (1991 i) placed Chrysoma near the Solidago group, although he noted that its position must be regarded as isolated. South and Central American genera. There are a number of South American members of the Erigeron-Conyza group, related either to Erigeron or to Conyza. The genus Conyzanthus was described in Flora SSSR but according to its author it is indigenous to South America (TAMAINSCHYAN1959). Conyzanthus and Laennecia are close relatives of Conyza (TAMAINSCHYAN1959, ZARDINI 1981). Darwiniothamnus is a segregate from Erigeron, and doubtfully distinct from it (NEsoM 1980). Hysterionica, Oritrophium, and Sommerfeltia all appear to be related to each other, as well as to Erigeron (CABRERA 1946, CHEBATAROFF 1981). Heterothalamus has been classified in the Baccharidinae, and belongs together with Baccharis in this group (see above). Another South American genus of the Baccharidinae, the monotypic Baccharidiopsis (BARROSO 1975), is nothing but a specialized species of 266 ZHANG XIAOPING& K. BREMER; Baccharis and should not be recognized at the generic level; it would leave Baccharis as paraphyletic. Yet another South American genus of the Baccharidinae, Baeeharidastrum, was synonymised with Baeeharis by NESOM (1988 b). The circumpacific though mainly Australian Brachyscome group (see further below) comprises a number of South American representatives, i.e. Floscaldasia, Laestadia, Lagenifera, and Plagiocheilus. The Pacific genus Lagenifera has species in South America as well as in Australasia (CABRERA 1966). CUATRECASAS(1969) described Flosealdasia and noted a possible relationship with Laestadia. He further compared both genera with Lagenifera. ROBINSON & BRETTELL(1973 b) stressed the same relationship, adding Plagiocheilus, Solenogyne (also BRUHL & QVINN 1990), and other Australian genera; all are members of the Brachyscome group. The Chiliotrichum group consists of shrubs with mostly densely set, coriaceous and often dorsally tomentose leaves. Most of the genera have paleate receptacles. The Chiliotrichum group was circumscribed by CABRERA(1954), with a later addition of Paleaepappus (CABRERA1969). Aylacophora and Paleaepappus are very similar, and possibly sister groups, Diplostephium was not discussed by CABRERA,but we place it in the same group. Diplostephium shares the same foliage characters. The Hinterhuberinae of CUATRECASAS(1989) comprise Hinterhubera, Flosmutisia, Westoniella, and Blakiella. The latter genus was segregated from Podoeoma; yet the two genera seem related because of their similarly beaked cypselas. Mierogynella has unbeaked cypselas but they are nevertheless similar to the flattened and marginally ribbed cypselas of Podoeoma. For these reasons we include Podoeoma and Microgynella together with the Hinterhuberinae in a Hinterhubera group. African and tropical Asian genera. A number of African (Nidorella; WILD 1969), paleotropical (Microglossa), and Madagascan-Mascarene genera (Psiadia, Psiadiella, Rochonia, Sarcanthemum) all appear to be close relatives of Conyza, in the Erigeron-Conyza group. Most African genera of the Astereae form the Amellus group, as circumscribed essentially by GRAU (1973) in his treatment of Felicia. Most of the genera in this group were by GRAU considered derivatives of Felicia, presumably with sister groups within Felicia. African "Brachyscome", included in this group by GRAU, have been removed to a separate genus, Jeffreya (~,VILD1974). Poecilopsis is another later addition (GRAU 1977 b). Two groups, the Apodocephala group and the Commidendron group, were left without representatives in the cladistic analysis because of insufficient information. Apodocephala and the related genus Vernoniopsis are usually associated with Psiadia and Rochonia in the Erigeron-Conyza group, and Commidendron and Melanodendron from St Helena are usually considered specialized representatives of Aster. The genus Pteronia is isolated among other African Astereae. However, Engleria (ENGLER 1888) has been proposed as a possible sister group, and we have assembled the two genera in the Engleria group. The Grangea group is essentially circumscribed as the Grangeinae in FAYED'S (1979) revision of this subtribe. However, we have added three genera, Colobanthera, Dacryotrichia, and Gyrodoma. The two former were not considered by FAYED, but appear to be relatives of Grangea (cf. WILD 1973). Gyrodoma was explicitly excluded from the Grangeinae by FAYED because of its possession of ray florets and specialized cypsela hairs. These may represent plesiomorphic and autapomorphic features within the Grangeinae, and we include Gyrodoma in the subtribe in agreement with WILD (1974). Australasian and Pacific genera. We consider the two large genera Brachyscome and Calotis in the Brachyscome group related, and to these we provisionally add Erodiophyllum. These genera have variously incised leaves and very variable, although frequently compressed cypselas. Presumably related are Lagenifera, Myriactis, Piora, and Solenogyne. That these genera form a (monophyletic) group is well known (CABRERA 1966, KOSTER 1966, Cladistic analysis of the Astereae 267 ADAMS 1979). There are also several South American members of this group, as discussed above. The large shrubby genus Olearia comprises most species in the Olearia group. GIVEN (1973) has convincingly argued for the relationship between Olearia and Celmisia, Damnamenia, and Pleurophyllum. Two other small genera related to Olearia are Pachystegia (CHEESEMAN1925) and Remya (WAGNER• HERBST 1987). The genera of the Vittadinia group are all related, and some sister genera have been established. Thus Dichromochlamys and Ixiochlamys are sister groups (DunLoP 1980), as well as Kippistia and Minuria (LANDER & BARRY 1980 a, b). Camptacra and Eurybiopsis are segregates of Vittadinia, and Eurybiopsis is considered more closely related to Minuria (BURB~DGE 1982). Tetramolopium has its closest relatives in Camptacra and Vittadinia (LOWREY 1986). The morphologically reduced Dimorphocoma and Elaehanthus, which are often regarded as isolated, are presumably also members of the Vittadinia group. Selection of characters The characters are listed in Table 3. We surveyed characters generally used at and above the generic level. In a more detailed study, there are certainly many more micromorphological characters that may be included. Many such characters are known for particular genera, e.g., Petradoria ANDERSON (1964), but had to be omitted due to lack of information in most of the genera. Genera with variable characters were coded as such, unless one of the states could be hypothesized as ptesiomorphic within the genus. The Anthemideae were used as outgroup. This tribe is closely related to the Astereae according to morphological data (PER OLA KARIS, pers. comm.) as well as cpDNA restriction site data (JANSEN & aI. 1990). Some characters are variable in the outgroup, and were consequently left unpolarized, indicated by a dash in the Anthemideae row of the data matrix (Table 4). Other presumably related tribes of subfam. Asteroideae are the Senecioneae, Calenduleae, or Gnaphalieae. However, they are similar to the Anthemideae in many of the characters used, and their use as outgroups is not likely to result in a dramatically different result. Another tribe, the Plucheeae (ANDERBERG 1991 b) differs from the Anthemideae in many of our characters, and is similar to the Erigeron-Conyza group in some floral structures. Analysis of the Astereae with the Plucheeae as outgroup will result in a very different cladogram, with the Erigeron-Conyza group as a basal paraphyletic group within the Astereae. We note this because tribal interrelationships within the subfamily Asteroideae are still uncertain (BREMER & al. 1992). However, there is presently no indication that the Astereae and the Plucheeae are closely related, and hence the use of Plucheeae as outgroup seems unjustified. Habit. The distinction between annual and perennial herbs is mainly at the specific level, within genera, and was thus omitted from this analysis. On the other hand, the difference between a herbaceous and shrubby habit seems applicable at the generic level, since there are several consistently herbaceous or consistently woody genera. However, there are a number of genera with both herbaceous and shrubby species. These genera were coded as variable. Leaves. The leaves are usually alternate along the stems, more rarely opposite (e.g., Engleria) or rosulate (e.g., Bellis). The latter two conditions represent uninformative autapomorphies in our sample of genera. In the outgroup, Anthemideae, the leaves are generally pinnatifid-pinnatisect or at least serrate-dentate. Within the Astereae the leaves are usually entire or serrate-dentate only. The more or less pinnatifid-pinnatisect condition appears to be characteristic mainly of Grangea and its relatives. Several American genera have entire and mid-ribbed or parallel-veined leaves. Most of the other genera were coded as variable for the leaf division character and plesiomorphic for the leaf venation character. 268 ZHANG XIAOPING • K. BREMER: Table 3. Characters and states used in the analysis of the Astereae 1. (0) Herbs or half-shrubs. (1) Shrubs or small trees. 2. (0) Leaves pinnatifid-pinnatisect or at least distinctly serrate-dentate. (1) Leaves entire or sparsely serrate-dentate only. 3. (0) Leaves pinnately veined. (1) Leaves mid-ribbed to tri-nerved or parallel-nerved. 4. (0) Capitula many-flowered, solitary or in lax inflorescences. (1) Capitula few-flowered and numerous in elaborate inflorescences. 5. (0) Involucral bracts very unequal and strongly imbricate. (1) Involucral bracts subequal and slightly imbricate. 6. (0) Involucral bracts not glutinous. (1) Involucral bracts very glutinous. 7. (0) Involucral bracts herbaceous, straight. (1) Involucral bracts chartaceous with squarrose tips. 8. (0) Receptacle epaleate. (1) Receptacle paleate. 9. (0) Receptacle flat, concave or slightly convex. (1) Receptacle conical or globular. 10. (0) Ray florets present, with expanded lamina. (1) Outer female florets present, more or less tubular. (2) Outer female florets present, bilabiate to deeply and irregularly lobed. 11. (0) Ray florets or outer female florets in 1-2 series. (1) Ray florets or outer female florets in 3 series or more. 12. (0) Ray florets or outer female florets yellow or whitish. (1) Ray florets or outer female florets white, pink, purple, violet, or blue. 13. (0) Ray florets nocturnally moving upwards, becoming erect. (1) Ray florets nocturnally rolling downwards, becoming revolute. 14. (0) Ray floret epidermis cells short with straight end walls. (1) Ray floret epidermis cells long with obtuse end walls. 15. (o) Disc florets perfect. (1) Disc florets functionally male. 16. (0) Disc floret corolla abruptly ampliate from tube to limb. (1) Disc floret corolla gradually ampliate without distinct tube and limb. 17. (0) Anther base obtuse or rounded. (1) Anther base cordate or sagittate (caudate). 18. (0) Style branches long, linear. (1) Style branches short, oblong. 19. (0) Style branch appendages shorter than stigmatic areas. (1) Style branch appendages longer than stigmatic areas. 20. (0) Style branch hairs short, obtuse-rounded. (1) Style branch hairs slender, acute. 21. (0) Cypselas angular or terete. (1) Cypselas flattened. (2) Cypselas winged. 22. (0) Cypselas 10 (8-)-ribed. (1) Cypselas 2-5 (7)-ribbed. 23. (0) Pappus elements few. (1) Pappus elements numerous. 24. (0) Pappus of bristles only. (1) Pappus of bristles and scales. (2) Pappus of scales only. Cladistic analysis of the Astereae 269 Table 3 (continued) 25. (0) (1) (2) (3) (4) 26. (0) (1) Basic chromosome number x = 9. Basic chromosome number x = 8. Basic chromosome number x = 6. Basic chromosome number x = 5. Basic chromosome number x = 4. Embryo sac development monosporic. Embryo sac development tetrasporic. Table 4. Data matrix of the Astereae. Unknown and variable states are coded as ? and -, respectively Character no 12345 67891 0 11111 12345 11112 67890 22222 12345 2 6 Anthemideae Grangea Solidago Engleria Ericameria Petradoria Chrysopsis Haplopappus Corethrogyne Grindelia Gutierrezia Amellus Chaetopappa Bellis Brachyscome Boltonia Erigeron Vittadinia Olearia Chiliotrichum Hinterhubera Aster Conyza Archibaccharis Psiadia 00000 00000 0-010 0-000 11000 01100 01100 --000 0-000 --000 -1110 0-000 011-0 01001 --001 01001 --000 0-000 1-000 11100 11100 --000 --010 --010 1~010 00-00 000-1 00000 00000 10000 000000000 00000 01000 11000 11000 00100 00000 00010 00010 000-0 00000 00000 00000 00100 00002 00000 00001 00001 00001 0-000 104 00? 00 00??0 00??0 00 ? ? 00??0 00?00 01??0 00000 00??0 01110 01110 01010 01110 01 ? ? 0 11-10 11??0 01? 10 01??0 004 01010 11--0 11--1 10--1 -0000 00100 00000 00010 00011 00011 00011 00001 00011 00011 00011 00000 00000 10100 10000 10000 -0000 00000 01000 110-0 1-000 10000 10000 10--0 10--0 0 0 000-0 00100 0011? 00100 00100 00113 -0103 00103 000-2 00024 00011 --0 11000 21000 21010 0-100 01100 01100 0-100 -1100 -0100 00100 00100 00100 0 0 0 ? ? ? ? ? ? 0 ? 9 ? 0 0 0 1 1 ? ? ? 0 0 ? ? Capitula. Most capitula are rather wide and many-flowered, solitary or arranged in lax inflorescences. There are also several genera with consistently few-flowered, slender capitula in elaborate panicles or corymbs. The involucral bracts are usually unequal in size and imbricate in several rows, whereas in some genera they are subequal and arranged in one or two rows. Some American genera have very sticky (glutinous) bracts, sometimes chartaceous with squarrose tips, typically in Grindelia. 270 ZHANG XIAOPING & K. BREMER: The receptacle is paleate in a few genera, and strongly convex-conical to globular in a few other genera. Ray florets. An effort was made to select radiate or at least disciform genera, rather than discoid genera for the analysis, in order to avoid unknown character states in the data matrix. It is assumed that ray florets and tubular outer female florets are homologous; in some genera there is a continuous variation from ray florets with expanded laminas through very shortly radiate outer florets to entirely tubular-filiform outer florets. The ray florets/outer female florets may be arranged in 1-2 series, or frequently in the Astereae, in several series. The distinction between yellow versus white/pink/purple/violet/blue ray florets is clearly applicable at the generic level. Most genera consistently have one of the two colour kinds, although there may be occasional, presumably secondarily colour-apomorphic species in some of the larger genera. The nocturnal movements of the ray florets were surveyed by STIRTON (1983). During nighttime the ray florets move upwards and assume an erect position in some Astereae, whereas they roll downwards and become revolute in other Astereae. Both types were found in different species of Erigeron. In the Anthemideae several types of nocturnal movements were observed, including the erect type but excluding the revolute type. The epidermis cells of the ray florets were investigated by BAAGOE(1978; see also LANE 1988). She distinguished two groups of genera within the Astereae, characterized by rather short epidermis cells with straight (truncate) end walls, and long epidermis cells with obtuse end walls, respectively. BAAGOEalso noted some correlation with floret colour and cuticle patterns. The short-celled type occurs in predominantly yellow-rayed genera and the longcelled type in white/pink/purple-rayed genera. All the Astereae have transversely rugose to striate cuticle patterns and according to BAAGOEthe short-celled epidermis cells never have an entirely striate cuticle, whereas the long-celled epidermis cells never have an entirely rugose cuticle. The cuticle patterns are largely overlapping, and are thus omitted in our data matrix. The outgroup, Anthemideae, has a different type of epidermis cells, more or less papillose and with a longitudinally striate cuticle pattern. The cells are usually short with obtuse end walls, as in one of the two Astereae types. Disc florets. The disc florets are perfect in most genera, although some of the central florets may be functionally male. Consistently functionally male disc florets occur in a few genera only. The corolla is usually shallowly 5-lobed. Deeply lobed, 4-lobed, and somewhat zygomorphic corollas occur in certain genera representing autapomorphies or characters applicable at the species level. The distinction between tube and limb (Fig. 1 A) is variable and provides a character for the data matrix. Anthers. The anther bases are normally rounded or obtuse and ecaudate, that is without sterile tails (Fig. 1 B). Some genera have cordate or sagittate anther bases (Fig. 1 C) provided with short sterile tails (caudate; ROBINSON 1983, THIELE 1988). Styles. As in many other tribes, the properties of the disc floret style branches have proved to be important, notably the length of the entire style branch as well as the ratio between the appendage and the stigmatic portion (Fig. 1 D-F). Most style branches are comparatively long, but they are distinctly shorter in Bellis and Grangea among the sampled genera (Fig. 1 D). A style branch appendage longer than the stigmatic area occurs mainly in the North American genera, and interestingly, this character seems to correlate with a type of more slender and acute style branch hairs (Fig. 1 E). This type of sweeping hair is different from the rather thick and obtuse-rounded type found in other Astereae and the outgroup. We noted the distinction in surveying micromorphological features in floret preparations. The slender and acute type of sweeping hair is somewhat similar but not identical to the very acute sweeping hairs found in most members of subfam. Cichorioideae as well as in the tribe Inuleae sensu stricto (BREMER 1987, ANDERBERG 1989). NESOM Cladistic analysis of the Astereae 271 .A D E iii!iiiiii i:iiii:i !: Fig. 1. Some floral characters used in the analysis of the Astereae. Bars: 1 mm. A Abruptly ampliate disc floret corolla in Aster ageratoides Tt~RCZ. B Obtuse anther bases in Aster tataricus L. F. C Caudate anther bases in Olearia myrsinoides (LABILL.) F. MUELL. D Short disc floret style branches in Bellis perennis L. E Long disc floret style branches with appendages longer than stigmatic areas and slender acute sweeping hairs in Chrysopsis gossypma (MicI~x.) ELL. F Long disc floret style branches with appendages shorter than stigmatic areas and short obtuse-rounded sweeping hairs in Aster sedifolius L. (1991 i) has made similar use of stylar variation in his studies of the Gutierrezia group and related Solidagininae. Cypselas. Much of the cypsela variation in the Astereae pertains to the specific level or constitutes autapomorphies for single genera. Some cypsela features characterize generic groups, but are autapomorphic and uninformative in our sample of genera. Two cypsela characters were selected for coding in the data matrix. One character represents the sequence from angular-terete cypselas through flattened to winged cypselas. The other cypsela character in the matrix distinguishes between the number of cypsela fibs. 272 ZHANG XIAOPINO t~ K. BREMER: Pappus. A distinction has been made as to the quantity and quality of the pappus elements. Most genera have a pappus of numerous scabrid bristles. A reduction to fewer bristles, sometimes entirely absent, occurs, for example, in Bellis. A distinction between soft and rigid or between white and brownish bristles was not possible to maintain at the generic level. A pappus composed of bristles as well as scales or of scales only occurs in a few genera (e.g., Grindelia,Gutierrezia).Plumose pappus bristles are rare and autapomorphic for a few genera, not included in the analysis. Chromosome number. The base number x = 9 is by far the most widespread in the Astereae, in the outgroup Anthemideae, and in many other Asteraceae tribes (SOLBRI~ 1977). Lower base numbers are characteristic of many American genera, for example, Grindelia (x = 6), Chrysopsis, Corethrogyne, Haplopappus (x = 5), and Gutierrezia (x = 4). Embryology. HARL~NG(1951) investigated the embryology of several Astereae species. Most species had a monosporic embryo sac development, whereas a tetrasporic embryo sac development was found in Vittadinia and Erigeron (except in sections Coenotus and Trimorphaea, however both now transferred to the genera Conyza and Trimorpha, respectively). Cladistic analysis The data matrix (Table 4) was analysed with the parsimony program Hennig 86 (FARRIS 1988), The options used where mhennig* and bb*, representing the initial tree building algorithm and a branch-swapping ("branch-breaker") feature in search for equally parsimonious solutions, respectively. This is currently the most forceful combination in a heuristic search for equally parsimonious cladograms (FARRIS 1988, SANDERSON 1990). The cladograms were rooted using the Anthemideae as outgroup. Multistate characters were treated as non-additive, without a priori assuming any transformation series. As evident from the data matrix, all analyses were made excluding uninformative autapomorphies and ingroup synapomorphies. Hennig 86 found 319 equally parsimonious cladograms. The cladograms are 58 steps long, with a consistency index of 0.55 and a retention index of 0.69 (FARRIS 1989). The strict consensus tree is shown in Fig. 2 A. Several of the clades are present in all 319 cladograms, as shown by the strict consensus tree. The different solutions are mainly due to various arrangements within the Engleria to Gutierrezia clade and the trichotomies of the cladogram. By way of an example, one cladogram is shown in Fig. 3. The combinable component consensus tree (BREMER1990) is shown in Fig. 2 B. In all 319 cladograms Grangea is either sister group to the rest or part of a trichotomy with Grangea, Solidago, and the rest of the genera. This trichotomy is unresolved in some cladograms, but mostly resolved, frequently with Solidago as sister group to the Engleria to Gutierrezia clade, but never with Solidago as sister group to the Amellus to Psiadia clade. Discussion The result of the cladistic analysis must be regarded as a very tentative hypothesis of Astereae phylogeny. Detailed conclusions on generic interrelationships should certainly be avoided. However, Figs. 2-4 may serve as a reference for some hypotheses and notes on character evolution and classification of the tribe. Cladistic analysis of the Astereae 273 Grangea -l [ [ A Grangea Solidago I Engleria EricameHa Petradoria [----cbrysops~ Haplot~ppus Coretbrogyrw Grindelia Gutierretzia Solidago Engleria , , E*qcameria Petradoria Chrysopsis Haplopappus Corethrogyne _ _ _ _ ~ Grindelia Gutierretzia B Fig. 2. A Strict consensus tree of all 319 equally parsimonious cladograms of the Astereae found from data matrix in Table4. B Combinable component consensus tree of all 319 cladograms The Astereae are mainly herbs with solitary or corymbose capitula, sometimes arranged in elaborate corymbose-paniculate inflorescences. Our results indicate that the shrubby habit (character 1) has evolved independently several times, also within genera. Likewise, elaborate infiorescences with slender, few-flowered capitula (character 4), evolved independently in several lineages, as indicated by the optimization on the cladogram. Two characters widely used in Asteraceae classification are the presence of receptacular paleae and the structure of the pappus. Evolution of the Astereae pappus (characters 23 and 24) is very homoplasious, as indicated by the many parallelisms and reversals on the cladogram. Receptacular paleae (character 8) are interpreted as independently evolved in Amellus and Chiliotrichum. This character was left unpolarized in the data matrix, and two parallel gains of paleae are more parsimonious than several secondary losses from a primitive, paleate condition. This is in agreement with the interpretation of a paleate receptacle as an evolutionary novelty within the family rather than being a relictual structure (KEn. & Sa-uEssY 1981, BP.EME~. 1987). Chromosomal evolution in the Astereae is differently interpreted depending on the cladogram chosen. In Fig. 3 reduction from a basic number of x = 9 to x = 5 characterizes the Chrysopsis to Gutierrezia clade, with a subsequent increase to x = 6 in Gutierrezia and decrease to x = 4 in Grindelia. In other cladograms the four lower numbers, x = 4, 5, 6, and 8, may be interpreted as unique and independent reductions from the plesiomorphic number x = 9. With respect to the phylogeny and subtribal classification, the perhaps most interesting and unexpected result of our analysis is the maintained distinction 274 ZHANG XIAOPING & K. BREMER: " i", i; ran+e+ /er/a Ericameria 191 ,,1"16 I Petradoria irl Ix -, H+ opapvus Co etbrogy.e 6 2a L..4t-R..II..~ Gutierrezia • '3"4~4~225 4 Cbaetopappa oltonia I ill 26 ~..-Erigeron I tl I ] .... rllz"X-Hinterbubera [..~ a ls ' ~1 Chiliotricbum I ii [" ' Olearia 1 I ,,,, I Aster """ ~ o n y z a 16 22 ~fi6qi ~ l _ _ ~ A r c b i b a c c h a r i s is ~ X Psiadia 12 Fig. 3. One of the 319 equally parsimonious cladograms of the Astereae found from the data matrix (Table 4). Solid bars = non-homoplasious synapomorphies. Open bars= homoplasious synapomorphies with reversals. Double bars =parallelisms. Crosses = reversals between the Solidagininae and the Asterinae. The two subtribes were originally distinguished on the technical character of the homochromous or heterochromous ray florets, which have been regarded as largely artificial. Although there are obvious exceptions, such as the heterochromous Corethrogyne group being nested within the homochromous Solidagininae, the homochromous/heterochromous division of the tribe is rather well maintained. Our analysis suggests that the Solidagininae may be retained as a possible monophyletic group, polythetically defined by two putative, albeit homoplasious synapomorphies, i.e. by the numerous pappus elements and long style branch appendages (characters 23 and 19). Slender and acute sweeping hairs (character 20) also characterize a large group of the Solidagininae. The Asterinae may be maintained sensu lato, i.e. including Bellidinae, Hinterhuberinae, Conyzinae, and Baccharidinae, hypothesized as monophyletic by the mostly heterochromous ray florets and the long ray floret epidermis cells (characters 12 and 14). Cladistic analysis of the Astereae 275 Grangea (Grangeinae) GRANGEINAE Solidago (Solidagininae) | ----J [-----Engleria (Solidagininae) I Ericameria (Solidagininae) SOLIDAGININAE I I ' Petradoria (Solidagininae) " I [---- Ch rysopsis (Solidagininae) .... I I Haplopappus(Solidagininae) ---7 Co reth rogyne (Asteri nae) ~-4 Grindelia (Solidagininae) -- ~ ~ Gutierrezia (Solidagininae) s Amellus (Asterinae) Chaetopappa (Asterinae) Boltonia (Asterinae) Brachyscome (Bellidinae) I I m ~ ] I I I ' ~ I I • (Bellidinae) ASTERINAE Vittadinia (Asterinae) Erigeron (Asterinae) t-----4 Hinterhubera (Asterinae) ' Chiliotrichum (Asterinae) ' Olearia (Asterinae) J Aster(Asterinae) I ~ Conyza (Conyzinae) I I Archibaccharis (Baccharidinae) ' Psiadia (Conyzinae) Fig. 4. The cladogram of the Astereae in Fig. 3 with subtribal classifications superimposed. The subtribes of BENTHAM(1873 a) and HOFFMANN(1890) are given in brackets after each genus and the three subtribes accepted here (Table 5) are printed in capitals As expected, the cladogram supports the well-known assumptions that the Bellidinae and the Conyzinae + Baccharidinae are derived within the Asterinae, and that the Baccharidinae are derived within the Conyzinae. In order to avoid obviously paraphyletic subtribes, the Bellidinae, Conyzinae, and Baccharidinae are best included in the Asterinae. This applies also to the Hinterhuberinae, represented by Hinterhubera. They are also nested within the Asterinae, and consequently best included therein. The isolated position of the Grangeinae is demonstrated in the cladogram by its position as sister group to the rest of the tribe. The Grangeinae are characterized by several putative synapomorphies, although with parallelisms within other parts of the tribe, i.e. reduced ray florets in several series and short style branches (characters 10, 11, and 18). The Grangeinae also differ from the rest of the tribe in their pinnatifid-pinnatisect leaves (character 2), a plesiomorphic feature in our analysis. Pending further analyses resulting in more robust cladograms of the Astereae, we propose to restrict the subtribes of the Astereae to Grangeinae, Solidagininae, and Asterinae sensu lato (Table 5). The subtribal name Heterochrominae is illegit- 276 ZHANG XIAOPING• K. BREMER: Table 5. Revised subtribal classification of the Astereae. The generic groups are circumscribed as in Table 2 Grangeinae BENTH. Grangea group (Africa and Tropical Asia) Solidagininae HOFFM. Chrysopsis group (North and South America) Corethrogyne group (North America) Engleria group (Africa) Ericameria group (North America) Grindelia group (North and South America) Gutierrezia group (mainly North and South America) Haplopappus group (North and South America) Petradoria group (North America) Solidago group (mainly North and South America) Isolated genera (North and Central America): Benitoa, Chrysoma, Xanthisma Asterinae DUMORT. Aster group (mainly Eurasia) Erigeron-Conyza group (Eurasia, North and South America, Africa and Tropical Asia) Amellus group (Africa) Apodocephala group (Madagascar) Bellis group (Eurasia) Boltonia group (North and South America) Brachyscome group (Australasia, South America, and the Pacific) Chaetopappa group (North America) Chiliotrichum group (South America) Commidendron group (St Helena) Hinterhubera group (South America) Olearia group (mainly Australia and New Zealand) Vittadinia group (Australia and the Pacific) Isolated genera (Asia): Formania, Nannoglottis, Thespis, Tolbonia Isolated genera (North America): Chloracantha, Geissolepis, Ionactis, Monoptilon Isolated genera (Africa and Tropical Asia): Psednotrichia, Rhamphogyne Isolated genera (Australia and the Pacific): Achnophora, Apostates, Ceratogyne, Phacellothrix imate and goes into synonymy under Asterinae. The subtribal name Homochrominate is based on the generic name Homoehroma and is as such legitimate, but in our classification it as a synonym under Asterinae, since Homochroma is a generic synonym of Mairia in the Amellus generic group. Our analysis indicates that there are several intercontinental relationships between generic groups. The Solidagininae are mainly American, but they include also the African Engleria group. The Asterinae are worldwide in distribution and include a number of Pacific distribution patterns (cf. BREMER 1993). Apparently, there is a good reason to consider genera from several continents in suprageneric studies of the Astereae. We thank ARNE A. ANDERBERG,JAMES S. FARRIS, CHRISTOPHERJ. HUMPHRIES,PER OLA KARIS, MARI K~LLERSJO, MEREDITH A. LANE, and especially GuY L. NESOM for Cladistic analysis of the Astereae 277 m a n y comments and much useful information; naturally, they do not necessarily agree with our conclusions based on this information. The study was supported by a Chinese State Education Commission grant to ZHAN6 XIAOPING for a research visit to Sweden 1990-1991 and a Swedish Natural Science Research Council grant to KARE BREMER for Asteraceae phylogeny. References ADAMS, L. G., 1979: A review of the genus Solenogyne (Asteraceae) in Australia and New Zealand. - Brunonia 2: 43-65. ANDERBERG, A. A., 1989: Phylogeny and reclassification of the tribe Inuleae (Asteraceae). - Canad. J. Bot. 6"/: 2277-2296. 1991 a: T a x o n o m y and phylogeny of the tribe Inuleae (Asteraceae). - P1. Syst. Evol. 176: 75-123. - 1991 b: T a x o n o m y and phylogeny of the tribe Plucheeae (Asteraceae). - P1. Syst. Evol. 176: 145-177. - 1991 c: T a x o n o m y and phylogeny of the tribe Gnaphalieae (Asteraceae). - Opera Bot. 104: 1-195. ANDERSON, L. C., 1964: Studies on Petradoria (Compositae): anatomy, cytology, taxonomy. Trans. Kansas Acad. Sci. 66: 632-684. - 1966: Cytotaxonomic studies in Chrysothamnus (Astereae-Compositae). - Amer. J. Bot. 53: 204-212. - 1970: Studies on Bigelowia (Astereae-Compositae). 1. Morphology and taxonomy. Sida 3: 451465. - 1972: Studies on Bigelowia (Astereae, Compositae). II. Xylary comparisons, woodiness, and paedomorphosis. - J. Arnold Arbor. 53: 499-514. - 1977: Studies on Bigelowia (Asteraceae). III. C y t o t a x o n o m y and biogeography. - Syst. Bot. 2: 209-218. 1987: Boltonia apalachieolensis (Asteraceae): a new species from Florida. - Syst. Bot. 12: 133-138. - WEBERG, P. S., 1974: The a n a t o m y and t a x o n o m y of Vanclevea (Asteraceae). - Great Basin Naturalist 34:151-160. ARIZA ESPINAR, L., 1980: Las especies centroargentinas de Hysterionica (Compositae). Darwiniana 22: 537-549. BAAGOE, J., 1978: Taxonomical application of ligule microcharacters in Compositae. II. Astereae. - Bot. Tidsskr. 72: 129-131. BARROSO, G. M., 1975: Baeeharidiopsis- u m genero novo da substribo Baccharidinae HOFFMANN (Tribo Astereae). - Sellowia 27: 95-101. BAYER, E., 1981' Revision der Gattung Chrysocoma L. (Asteraeeae-Astereae). - Mitt. Bot. Staatssamml. Mfinchen 17: 259-392. BEAMAN, J. H., 1957: The systematics and evolution of Townsendia (Compositae). - Contr. G r a y Herb. 183: 1-151. BENTHAM, G., 1873 a: Compositae. - In BENa'HAM, G., HOOKER, J. D., (Eds): Genera plantarum 2, pp. 163-533. - London. 1873 b: Notes on the classification, history, and geographical distribution of the Compositae. - J. Linn. Soc. Bot. 13: 335-577. BLAKE, S. F., 1924: Hemibaccharis, a new genus of Baccharidinae. - Contr. U.S. Natl. Herb. 20: 543-554. - 1937: Tracyina, a new genus of Asteraceae from Northern California. - Madrofio 4: 73-77. BREMER, K., 1987: Tribal interrelationships of the Asteraeeae. - Cladistics 3: 210-253. 1990: Combinable component consensus. - Cladistics 6: 369-372. 278 ZHANG XIAOPING& K. BREMER: 1993: Intercontinental relationships of African and South American A s t e r a c e a e - a cladistic biogeographic analysis. - In GOLDBLAa'a',P., (Ed.): Biological relationships between Africa and South America. - New Haven, Connecticut: Yale University Press (in press). - HUMPHRIES,C. J., 1993: Generic monograph of the Anthemideae (Asteraceae). - Bull. Brit. Mus. (Nat. Hist.)Bot. (in press). -- JANSEN, R. K., KARIS, P. O., K.~LLERSJO, M., KEELEY, S. C., KIM, K.-J., MICHAELS, H. J., PALMER,J. D., WALLACE,R. S., 1992: A review of the phylogeny and classification of the Asteraceae. - Nordic J. Bot. 12: 141-148. BROWN, G. K., CLARK, W. D., 1982: Taxonomy of Haplopappus sect. G y m n o e o m a (Compositae). - Syst. Bot. 7: 199-213. BRUHL, J. J., QUINN, C. J., 1990: Cypsela anatomy in the "Cotuleae" (Asteraceae-Anthemideae). - Bot. J. Linn. Soc. 102: 37-59. BURBIDGE, N. T., 1982: A revision of Vittadinia A. Rich. (Compositae) together with reinstatement of Eurybiopsis DC. and description of a new genus, Camptacra. - Brunonia 5: 1-72. CABRERA, A. L., 1946:E1 g6nero Hysterionica en el Uruguay en y la repflblica Argentina. Notas Mus. La Plata, Bot. 11: 349-359. 1954: Las especies del g6nero "Nardophyllum". - Notas Mus. La Plata, Bot. 17: 5566. - 1966: The genus Lagenophora (Compositae). - Blumea 14: 285-308. 1969: Compuestas nuevas de Patagonia. - Bol. Soc. Argent. Bot. 11: 271-75. CHEBATAROFF, P. J., 1981: Una nueva especie del g6nero Sommerfeltia (Compositae). Hickenia 1: 233-235. CHEESEMAN, T. F., 1925: Manual of the New Zealand Flora, 2nd edn. - Wellington: Government Printer. CLARK, W. D., 1979: The taxonomy of Hazardia (Compositae-Astereae). - Madrofio 26: 105-127. CRONQUIST, A., 1947: Revision of the North American species of Erigeron, North of Mexico. Brittonia 6: 121-302. CUATRECASAS,J., 1969: Prima Flora Colombiana, 3. Compositae-Astereae. - Webbia 24: 1-335. - 1970: Reinstatement of the genus Llerasia (Compositae). - Biotropica 2: 39-45. - 1977: Westoniella, a new genus of the Astereae from the Costa Rican paramos. Phytologia 35: 471-487. 1982: Miscellaneous notes on neotropical flora, XV. New taxa in the Astereae. Phytologia 52: 166-177. 1989: Un g6nero nuevo de Astereae, Compositae, de Colombia. - Anales Jard. Bot. Madrid 42: 415-426. DAVIS, G. g., 1948: Revision of the genus Brachyeome CAss. -- Proc. Linn. Soc. New South Wales 73: 142-241. 1952: Revision of the genus Calotis R. BR. -- Proc. Linn. Soc. New South Wales 77: 146-188. DAvis, P. H., (Ed.), 1975: Flora of Turkey 5. - Edinburgh: The University Press. DE JONG, D. C. D., 1965: A systematic study of the genus Astranthium (CompositaeAstereae). - Michigan, East Lansing: Publication of the Museum, Michigan State University. BEAMAN, J. H., 1963: The genus Olivaea (Compositae-Astereae). - Brittonia 15: 8692. DUNLOP, C. R., 1980: A revision of L,ciochlamys (Asteraeeae-Astereae). - J. Adelaide Bot. Gard. 2: 241-252. ENCLER, A., 1888: Plantae Marlothianae. - Bot. Jahrb. Syst. 10: 242-285. - - - - - - - - - Cladistic analysis of the A s t e r e a e 279 FARRIS, J. S., 1988: Hennig 86. Version 1.5. - Port Jefferson Station, New York: Published by the author. 1989: The retention index and the rescaled consistency index. - Cladistics 5: 417.419. FAYED,A., 1979: Revision der Grangeinae (Asteraceae-Astereae). - Mitt. Bot. Staatssamml. Miinchen 15: 425-576. GIVEN, D. R., 1969: Taxonomic notes on the genus Celrnisia (Compositae). A synopsis of infrageneric categories in Celmisia (Astereae-Compositae). - New Zealand J. Bot. 7: 389-418. - 1973: D a m n a m e n i a gen. nov. A new subantarctic genus allied to Celmisia CAss. (Astereae-Cornpositae). - New Zealand J. Bot. 11: 785-796. 1980: A taxonomic revision of Celmisia coriacea WORST.F.) HOOK.V. and its immediate allies (Astereae-Compositae). - New Zealand J. Bot. 18: 127-140. - 1984: A taxonomic revision of Celmisia subgenus Pelliculatae section Petiolatae ( C o m positae-Astereae). - New Zealand J. Bot. 22: 139-158. GRAU, J., 1970: Die Gattung P o l y a r r h e n a CASS. (Asteraceae-Astereae). - Mitt. Bot. Staatssamml. Mfinchen 7: 347-368. - 1973: Revision der Gattung Felicia (Asteraceae). - Mitt. Bot. Staatssamml. M/inchen 4: 195-705. - 1977 a: A s t e r e a e - s y s t e m a t i c review. - In HEYWOOD,V. H., HARBORNE,J. B., TURNER, B. L., (Eds): The biology and chemistry of the C o m p o s i t a e 1, pp. 539-565. - London: Academic Press. 1977 b: Poecilolepis - Eine neue Gattung der Asteraceae-Astereae. - Mitt. Bot. Staatssamml. Miinchen 13: 243-254. GRIERSON, A. J. C., 1967: The genus P s y c h r o g e t o n (Compositae). - Notes Roy. Bot. Gard. Edinburgh 27: 101-147. HALL, H. M., 1928: The genus Haplopappus. A phylogenetic study in the Compositae. Washington: Carnegie Institution of Washington. - CLEMENTS,F. C., 1923: The phylogenetic method in taxonomy. The North American species of Artemisia, C h r y s o t h a m n u s , and Atriplex. - Publ. Carnegie Inst. Washington 326: 1-355. HARLING, G., 1951: Embryological studies in the C o m p o s i t a e III: Astereae. - Acta Horti Berg. 16: 73-120. HARTMAN, R. L., 1990: A conspectus of M a c h a e r a n t h e r a (Asteraceae: Astereae). - Phytologia 68: 439-465. HELLWI6, F. H., 1990: Die Gattung Baccharis L. (Compositae-Asteraceae) in Chile. Mitt. Bot. Staatssamml. Mfinchen 29: 1-456. HILLIARD, O. M., BURTT, B. L., 1973: Notes on some plants of Southern Africa chiefly from Natal: I I I . - Notes Roy. Bot. Gard. Edinburgh 32: 364--365. HOFFMANN, O., 1890: Tubuliflorae-Astereae. - In EN~LER, A., PRANTL, K., (Eds): Die natfirlichen Pflanzenfamilien 4(5), pp. 142-172. - Leipzig: Wilhelm Engelmann. HOWELL, J. T., 1929: A systematic study of the genus Lessingia CHAM. -- Univ. Calif. Publ. Bot. 16: 1-44. HUTCHINSON, J., PHILLIPS,E. P., 1917: A revision of the genus Pteronia (Compositae). Ann. South African Mus. 9: 277-329. JACKSON, J. D., 1975: A revision of the genus Archibaccharis HEERIN6 ( C o m p o s i t a e - A s tereae). - Phytologia 32: 81-105. JANSEN, R. K., HOLSINGER, K. E., MICHAELS, H. J., PALMER, J. D., 1990: Phylogenetic analysis of chloroplast DNA restriction site data at higher taxonomic levels: an example from the Asteraceae. - Evolution 44: 2089-2105. JEFFREY, C., CHEN, Y. L., 1984: Taxonomic studies on the tribe Senecioneae (Compositae) of Eastern Asia. - Kew Bull. 39: 205.446. - - - 280 ZrtANQ XIAOPING t~; K. BREMER: JONES, A. G., 1980: A classification of the new world species of Aster (Asteraceae); Data on chromosome numbers in Aster (Asteraceae), with comments on the status and relationships of certain North American species. - Brittonia 32: 230--261. SMOGOR, A., 1983: Chromosome counts of and notes on some old world Asters (Asteraceae). - Phytologia 53: 429-431. - YOUNG, D. A., 1983: Generic concepts of Aster (Asteraceae); a comparison of cladistic, phenetic, and cytological approaches. - Syst. Bot. 8: 71-84. KECK, D. D., 1958: Taxonomic notes on the California flora. - Aliso 4: 102-103. KEIL, D. H., STUESSY, T. F., 1981: Systematics of Isocarpha (Compositae." Eupatorieae). Syst. Bot. 6: 258-287. KING, R. M., ROBINSON, H., 1987: The genera of the Eupatorieae (Asteraceae). - Lawrence, Kansas: Allen Press. KOMAROV, V. L., (Ed.), 1959: Flora S.S.S.R. 25. - Moscow, Leningrad: Akademija Nauk. KOSTER, J. TH., 1966: The Compositae of New Guinea I. - Nova Guinea 24: 524--612. LANDER, N. S., BARRY, R., 1980a: Reinstatement of the genus Kippistia F. MUELL. (Asteraceae-Astereae). - Nuytsia 3:215-219. 1980b: A review of the genus Minuria DC. (Asteraceae-Astereae). - Nuytsia 3: 221-237. LANE, M. A., 1979: Taxonomy of the genus Amphiachyris (Asteraceae-Astereae). - Syst. Bot. 4: 178-189. - 1982: Generic limits ofXanthocephalum, Gutierrezia, Amphiachyris, Gymnosperma, Greenella, and Thurovia (Compositae-Astereae). - Syst. Bot. 7: 405-416. - 1983: Taxonomy of Xanthocephalum (Compositae-Astereae). - Syst. Bot. 8: 305-316. - 1985: Taxonomy of Gutierrezia (Compositae-Astereae) in North America. - Syst. Bot. 10: 7-28. 1988: Generic relationships and taxonomy of Acamptopappus (Compositae: Astereae). - Madrofio 35: 247-265. LINa, Y., CHeN, Y. L., 1965: Genera nova del minus cognita familiae Compositarum, II. Cavea W. W. SMITH et SMALL et Nannoglottis MAXIM. -- Acta Phytotax. Sin. 10: 91104. 1973: Notulae de genere Erigeron L. generibusque affinibus florae Sinicae. - Acta Phytotax. Sin. 11: 399-430. LOWREY, T. K., 1986: A biosystematic revision of Hawaiian Tetramolopium (CompositaeAstereae). - Allertonia 4: 204-264. NESOM, G. L., 1980: A revision of the epappose species of Erigeron (Compositae-Astereae). - Ph.D. dissertation, Univ. of North Carolina, Chapel Hill. 1988a: Synopsis of Chaetopappa (Compositae-Astereae) with a new species and the inclusion of Leucelene. - Phytologia 64: 448-456. - 1988 b: Baccharis sect. Baccharidastrum (Cornpositae-Astereae), including two monoecious and one dioecious species. - Phytologia 65: 169-173. - 1989 a: Taxonomy of Erigeron sect. Polyactis (Compositae: Astereae). - Phytologia 66: 415-455. - 1989 b: The separation of Trimorpha (Compositae-Astereae) from Erigeron. - Phytologia 67: 61-66. - 1989 c: Infrageneric taxonomy of New World Erigeron (Compositae-Astereae). - Phytologia 67: 67-93. 1989 d: A new combination in Stenotus (Compositae: Astereae). - Phytologia 67:113114. - 1990 a: Taxonomic summary of Ericameria (Asteraceae: Astereae), with the inclusion of Haplopappus sects. Macronema and Asiris. - Phytologia 68: 144-155. - 1990b: Taxonomy of the genus Laennecia (Asteraceae: Astereae). - Phytologia 68: 205-228. - - - - Cladistic analysis of the Astereae - 28 l 1990 c: Further definition of Conyza (Asteraceae: Astereae). - Phytologia 68: 229-233. 1990d: Taxonomy of Heterotheca sect. Heterotheea (Asteraceae-Astereae) in Mexico, with comments on the taxa of the United States. - Phytologia 69: 282-294. - 1991 a: Taxonomy of Isocoma (Compositae." Astereae). - Phytologia 70: 69-114. - 1991 b: Union of Bradburia with Chrysopsis (Asteraceae." Astereae), with a phylogenetic hypothesis for Chrysopsis. - Phytologia 71: 109-121. - 1991 c: Transfer of Aster kingii to Tonestus (Asteraceae: Astereae). - Phytologia 71: 122-127. 1991 d: Tomentaurum (Asteraceae." Astereae), a new genus of goldenaster from Chihuahua, Mexico. - Phytologia 71: 128-131. - 1991 e: Transfer of Heterotheca bartletii to Osbertia (Asteraceae." Astereae). - Phytologia 71: 132-135. - 1991 f: A phylogenetic hypothesis for the goldenasters (Asteraceae: Astereae). - Phytologia 71: 136-151. - 1991 g: Two new species of Archibaccharis (Asteraceae: Astereae) from Mexico with a reevaluation of sectional groupings in the genus. - Phytologia 71: 152-159. 1991 h: Redefinition of Hesperodoria (Asteraceae: Astereae) and the segregation of Columbiadoria, a new monotypic genus from the western United States. - Phytologia 71: 244-251. - 1991 i: Morphological definition of the Gutierrezia group (Asteraceae." Astereae). Phytologia 71: 252-262. MORGAN, D. R., 1990: Reinstatement of Tonestus (Asteraceae: Astereae). - Phytologia 174-180. - SuH, Y., MORGAN, D. R., SIMPSON, B. B., 1990: Xylothamia (Asteraceae-Astereae), a new genus related to Euthamia. - Sida 14:101-116. PORTER, C. L., 1943: The genus Arnphipappus TORR. and GRAY. -- Amer. J. Bot. 30: 481483. RECHINGER, K. H., (Ed.), 1982: Flora Iranica 154. - Graz: Akademische Druck- und Verlagsanstalt. ROBINSON, H., 1981: A revision of the tribal and subtribal limits of the Heliantheae (Asteraceae). - Smithsonian Contr. Bot. 51: 1-102. 1983: A generic review of the tribe Liabeae (Asteraceae). - Smithsonian Contr. Bot. 1-69. - BRETTELL, R. D., 1972: Tribal revision in the Asteraceae. I. The relationships of Geissolepis. - Phytologia 24: 299-301. 1973 a: Tribal revisions in the Asteraceae. V. The relationship of Rigiopappus. Phytologia 26: 6%70. 1973 b: Tribal revisions in the Asteraceae. X. The relationship of Plagiocheilus. Phytologia 26: 159-162. ROMMEL, A., 1977: Die Gattung Amellus L. (Asteraceae-Astereae). Systematischer Teil. Mitt. Bot. Staatssamml. Miinchen 13: 579-728. 1979: Die Gattung Amellus L. (Asteraceae-Astereae). Allgemeiner Teil. - Mitt. Bot. Staatssamml. Miinchen 15: 243-328. SANDERSON, M. J., 1990: Flexible phylogeny reconstruction: a review of phylogenetic inference packages using parsimony. - Syst. Zool. 39: 414420. SEMPL~, J. C., 1981: A revision of the goldenaster genus Chrysopsis (NUTT.) ELL. nora. cons. (Compositae-Astereae). - Rhodora 83: 323-384. BLOK,V. C., HEIMAN,P., 1980: Morphological, anatomical, habit, and habitat differences among the goldenaster genera Chrysopsis, Heterotheca, and Pityopsis (CompositaeAstereae). - Canad. J, Bot. 58: 147-163. - BOWERS,F. D., 1985: A revision of the goldenaster genus Pityopsis NUTT. (CompositaeAstereae). - Univ. Waterloo Biol. Ser. 29: 1-34. - - - 68: - 54: - - - 282 ZHANG XIAOPING& K. BREMER: SHrNNERS, L. H., 1946: Revision of the genus Aphanostephus DC.; The genus Dichaetophora A. GRAY and its relationships. - Wrightia 1: 90-121. 1949: Revision of the genus Egletes CASSlNI North of South America. - Lloydia 12: 239-250. SIEREN, D. J., 1981: The taxonomy of the genus Euthamia. - Rhodora 83: 551-579. SIMON, V. C., 1978: Bellium eorsicum n. sp. sowie Notizen fiber die Gattung Bellium (Asteraeeae). - Bauhinia 6: 279-284. SMITH,E. B., 1965: Taxonomy o f Haplopappus, section Isopappus (Compositae). - Rhodora 67: 217-238. 1981: New combinations in Croptilon (Compositae-Astereae). - Sida 9: 59-63. SOLBRIG, O. T., 1960: The South American sections of Erigeron and their relation to Celmisia. " Contr. Gray Herb. 188: 65-85. - 1962: The South American species of Erigeron. - Contr. Gray Herb. 191: 3-79. - 1966: The South American species of Gutierrezia. - Contr. Gray Herb. 197: 3-42. - 1977: Chromosomal cytology and evolution in the family Compositae. - In HEYWOOD, V. H., HARBORNE, J. B., TURNER, B. L., (Eds): The biology and chemistry of the Compositae 1, pp. 267-281. - London: Academic Press. STEYERMARK,J. A., 1934: Studies in Grindelia. I. New species, varieties, and combinations of Grindelia. - Ann. Missouri Bot. Gard. 21: 227-230. STmTON, C. H., 1983: Nocturnal petal movements in the Asteraceae. - Bothalia 14: 10031006. SUH, Y., SIMPSON,B. B., 1990: Phylogenetic analysis of chloroplast DNA in North American Gutierrezia and related genera (Asteraceae: Astereae). - Syst. Bot. 15: 660-670. SUND~ERG, S. D., 1991: Infraspecific classification of Chloracantha spinosa (BENTH.)NESOM (Asteraceae) Astereae. - Phytologia 70: 382-391. TAMAINSCFIYAN, S. G., 1959: Conyzanthus. - In KOMAROV,V. L., (Ed.): Flora S.S.S.R. 25, pp. 185-188, 583. - Moscow, Leningrad: Akademija Nauk. THIELE, E.-M., 1988: Bau und Funktion des Antheren-Griffel-Komplexes der Compositen. Berlin: J. Cramer in Gebriider Borntraeger. TURNER, B. L., 1984: Taxonomy of the genus Aphanostephus (Asteraeeae-Astereae). Phytologia 56: 81-101. 1987: Taxonomic study of Machaeranthera, sections Maehaeranthera and Hesperastrum (Asteraceae). - Phytologia 62: 207-266. SUND~ERG,S. D., 1986: Systematic study of Osbertia (Asteraceae-Astereae). - P1. Syst. Evol. 151: 22%-239. TUTIN, T. G., HEYWOOD, V. H., BURGES, N. A., MOORE, D. M., VALENTINE, D. H., WALTERS,S. M., WEBB,D. A., (Eds), 1976: Flora Europaea 4. - Cambridge: Cambridge University Pre~s. VAN HORN, G. Si, 1973: The taxonomic status of Pentachaeta and Chaetopappa with a revision of Pentachaeta. - Univ. Calif. Publ. Bot. 65: 1-41. WAGNER, W. L., HERBST, D. R., 1987: A new species of R e m y a (Asteraceae-Astereae) on Kauai and a review of the genus. - Syst. Bot. 12: 601-608. WATSON, T. J., 1977: The taxonomy of Xylorhiza (Asteraeeae-Astereae). - Brittonia 199-216. WELD, H., 1969: The genus Nidorella CASS. -- Bol. Soc. Brot. 43: 209-243. 1973: A new genus of Compositae (Astereae) from the flora Zambesiaca area. - Garcia de Orta 1: 67-68. 1974: New and interesting Compositae from South-central Africa, 2. - Kirkia 9: 293300. ZARDINI, E. M., 1981: Contribuciones para una monografia del grnero Conyza LESS. II. Rehabilitaci6n del grnero Laennecia CASS. -- Darwiniana 23: 159-169. - - - - - 2 - - 9 : Cladistic analysis of the - 283 Astereae 1985:Revision del grnero N o t i c a s t r u m Secc. Bot. 13: 313-424. (Compositae-Astereae). - Revista Mus. La Plata, Addresses of the authors: KARE BREMZR,Department of Systematic Botany, Uppsala University, P.O. Box 541, S-751 21 Uppsala, Sweden. - ZHAN6 XIAOPING, Department of Biology, Anhni Normal University, Wuhu 241000, People's Republic of China. Accepted September 8, 1992 by F. EHRENDORF~R Verleger: Springer-Verlag KG, Sachsenplatz 4-6, A-1201 Wien. - Herausgeber: Prof. Dr. Friedrich Ehrendorfer, Institut f/Jr Botanik und Botanischer Garten der Universitht Wien, Rennweg 14, A-1030 Wien. - Redaktion: Rennweg 14, A-1030 Wien. - Hersteller: AdolfHolzhausens Nfg., Kandlgasse 19-21, A-1070 Wien. - Verlagsort: Wien. - Herstellungsort: Wien, - Printed in Austria.