ATTENTION: THIS PAPER IS BASED UPON AN ORAL
PRESENTATION AND ABSTRACT PRESENTED
Formal taxonomic transfers have been published in Arnaldoa 10(1): 45-60. 2003
CLASSIFICATION AND PHYLOGENY OF THE SOUTH AMERICAN GNAPHALIEAE (ASTERACEAE)
[Revised 1 Oct 2012]
Michael O. Dillon
IntroductionThe tribe Inuleae (s. ampl.) has been the focus of recent attempts to establish classifications reflective of putative phylogenetic relationships. In the traditional interpretation, the tribe contains between 180-200 genera and over 2100 species, and has worldwide distribution with centers of generic diversity in Australia, Eurasia, South Africa, and South America. The tribe has recently been redefined and no South American genera are placed in the Inuleae (s.s.), rather they are relegated to two new segregate tribes Gnaphalieae Benth. and Plucheeae Anderb. In South America, the Gnaphalieae consists of ca. 20 genera and over 100 species with highest diversity in the tropical and subtropical Andean Cordillera. The majority are Neotropical endemics, but some are cosmopolitan or pantropical genera, such as Achyrocline, Gamochaeta, and Pseudognaphalium, with their greatest species diversity in South America. Due to conflicting hypotheses of putative relationships among genera within the subtribes of the Gnaphalieae (Dillon 1988, 1990), the phylogeny of selected South American genera was reinvestigated.
South American GnaphalieaeDrury (1970) examined the relationships between the "cudweeds" of New Zealand (Gnaphalieae) and adopted a largely phenetic system with most taxa placed under a broad generic concept for Gnaphalium. Hilliard and Burtt (1981) attempted to clarify generic boundaries within the South African Gnaphaliinae, but their results were largely intuitive and untestable. They provided an "orientation diagram" illustrating the distribution of various character states they considered important, including leaf type and orientation, phyllaries, and achenial trichome types. Merxmüller et al. (1977) recognized several loose "groups" of potentially related taxa. Bremer (1976a, 1976b, 1978a, 1978b) and later Anderberg (1986, 1988) produced revisions of various genera.
Anderberg (1991) prepared a cladistic analysis of the Gnaphalieae which included 72 genera and utilized 82 characters to establish five subtribes and many monophyletic groups. The paper ultimately treated 146 genera, including formation of 6 new genera, 16 resurrected generic names, 166 new combinations, and 4 new names to replace homonyms. He placed the 20 South American genera into three subtribes: Cassiniinae Anderb., Gnaphaliinae Less., and Loricariinae Anderb. The subtribes Relhaniinae and Angianthinae, established for Old World genera, have no representatives in South America.
Subtribe Cassiniinae Anderb. in South America
Anderberg (1991) established the subtribe Cassiniinae for a group of genera with worldwide distribution and diagnosed as follows: often dioecious or subdioecious shrubs or herbs, pholem fibers absent, phyllaries with opaque laminae, hermaphroditic florets with truncate styles possessing trichomes their abaxial surfaces, achenes usually with two vascular bundles, and pappus bristles with clavate apical cells. Among the 16 genera placed there were Anaphalis, Antennaria, Gnaphaliothamnus, and Chionolaena. Anderberg admitted that the subtribe was largely diagnosed by a string of homoplastic characters, but argued that its members could hardly be included in any of the other subtribes. Within the subtribe, the South American genera were placed in an informal gathering called the "Anaphalis group". Chionolaena (South America) and Gnaphaliothamnus (Central America and Mexico) were diagnosed by revolute leaf margins, and by "Leaf orientation," a character with two states "(1) Basal leaves deflexed" and "(0) Basal leaves not deflexed" (Anderberg 1991, p. 15). Anaphalis and Antennaria (along with Ewartia) formed the sister group to Chionolaena and Gnaphaliothamnus. The monotypic Colombian genus, Pseudoligandra M.O.Dillon & Sagást. based upon Oligandra chrysocoma Wedd. (Dillon & Sagástegui 1990, Fig. 1), was placed under the synonymy of Chionolaena with a statement that its recognition would cause Chionolaena to be paraphyletic. Another segregate, Parachionolaena Dillon & Sagást. established for Chionolaena columbiana S. F. Blake (Dillon & Sagástegui 1991b, Fig. 2), was unknown to Anderberg. However, a recent monograph (Freire 1993) treated it as congeneric with Chionolaena, along with the several species of Gnaphaliothamnus.
Nesom (1990b,c) treated Gnaphaliothamnus as a distinct genus with undoubted relationships with Chionolaena. Nesom (1994) provided a critique of the various classifications of Gnaphaliothamnus and stated that in his opinion the Mexican and Central American taxa formed a monophyletic group potentially related to, but distinct from, Chionolaena. [NOTE: Nesom (2001) has now transfered the Mexican and Central American species into Chionolaena.] If the segregates proposed by Dillon and Sagástegui are to be included in Chionolaena, the diagnosis must be modified to reflect the various morphological traits not shared by the majority of the genus (Dillon & Sagástegui, 1990, 1991b).
Anderberg (1992) published a "revised" data matrix to correct a printing error that omitted several important genera. The results of Anderberg's matrix has not changed, rather the author states that the 1992 matrix allows for the results in Anderberg 1991 to be comprehensible.
Fig 1. Pseudoligandra chrysocoma. A. Achene [1125 µm long]. B. Achene surface enlargement. (Voucher: Rangel et al. 1914, COL).
Fig. 2. Parachionolaena columbiana. A. Achene [1000 µm long]. B. Achene surface enlargement. C. Pappus bristle apex enlargement. (Voucher: Barclay 6927, COL).
Subtribe Loricariinae Anderb. in South AmericaAnderberg (1991) established the subtribes Loricariinae and Relhaniinae as sister taxa in a clade largely made up of woody genera with leaves possessing involute margins and pubescence on adaxial surfaces. The Relhaniinae contained 19 woody African genera with discolorous ray florets and rod-like achenes. The Loricariinae were diagnosed as compact, often dioecious shrubs without fibers in the phloem, crowded leaves with adaxial pubescence, achenes with more than two vascular bundles, and dimorphic pappus (i.e., apical cells of bristles acute in the pistillate florets and clavate in the hermaphroditic florets). Anderberg included Pterygopappus (Tasmania), Psychrophyton (New Zealand), and two Andean endemics, Loricaria and Mniodes in the original cladistic analysis. Later, he added Raouliopsis (Colombian endemic) and Sinoleontopodium (China) to the clade a posteriori. Preliminary results (Dillon, unpubl.) suggest that Mniodes and Raouliopsis share closer relationships with members of the "Lucilia group" and in particular the "Belloa clade" and bear only superficial resemblance to the New Zealand and Tasmanian genera. The greatly condensed and compacted leafy stems of Mniodes and Raouliopsis are similar to highly reduced Luciliocline taxa. Authentic material of Sinoleontopodium has not been examined, but the generic description could refer to subshrubby species of Anaphalis common to the Himalayas.
Subtribe Gnaphaliinae Benth. in South AmericaThe subtribe Gnaphaliinae (sensu Anderberg) was established for a group of genera with worldwide distribution diagnosed as annual to perennial herbs (occasionally shrubs), capitula with marginal pistillate florets greatly outnumbering the central hermaphroditic or staminate florets, and oblong achenes with pubescence of short, clavate trichomes. The subtribe encompasses 47 genera, with centers of diversity in Africa, Asia, and the Neotropics, while no fewer than 14 genera were partially or wholly distributed in South America. This subtribe included several informal groups. The "Helichrysum group", including Pseudognaphalium and Achyrocline, was largely diagnosed on the basis of phyllaries with divided stereomes, yellow florets, and papillose achenial pubescence. Individual genera were defined by the ratio of pistillate to hermaphroditic florets. Stenophalium was placed with this group after the analysis. There exists a grade of floret ratios beginning with Stenophalium, where the number of pistillate florets is reduced to one or two per capitulum and the fertile hermaphroditic florets are typically only five. Achyrocline, has 1-11(-23) pistillate florets and 1-4(-6) fertile, hermaphroditic florets. Pseudognaphalium has (25-) 40-130 pistillate florets and 5-10 (-25) functionally staminate hermaphroditic florets. Finally, in Helichrysum, the central hermaphroditic florets far out number the pistillate florets. Should the character of the floral ratio of capitula be discarded and these weakly defined genera be combined, the oldest valid name available would be Helichrysum.
The "Filago group" comprises an essentially Nearctic clade with African, Eurasian, and North American elements. Psilocarphus is a predominately North American genus represented in South America by one Chilean endemic (Cronquist 1950). Both Anderberg (1991) and Morefield (1992) related Psilocarphus to Stylocline. Anderberg (1994) suggested that the closest relative of Micropsis was obscure, though it was most frequently associated with genera of the "Filago group." An alternative position for Micropsis in the "Lucilia group" is presented below.
Anderberg (1991) established the "Lucilia group" diagnosed as herbs or subshrubs, with polychromous phyllaries, pistillate florets with generally yellow corollas, hermaphroditic florets with generally purple corollas, and the corolla veins ending below the apex of the lobes. The group initially contained Lucilia, Belloa, Chevreulia, Jalcophila, Cuatrecasasiella, Berroa, and Facelis. Subsequently, Anderberg and Freire (1991) considered Gamochaeta as the sister taxon to the "Lucilia group" and Micropsis and Stuckertiella were included in it subsequent to their cladistic analysis. Freire (1986, 1987) published species-level cladistic analyses where Belloa and Lucilia were combined to form an expanded Lucilia. Anderberg and Freire (1991) published another cladistic analysis where the expanded Lucilia was dismantled and its constituent taxa placed in four genera: (1) a revised Belloa (9 spp.), (2) a revised Lucilia (8 spp.), and two new genera, (3) Gamochaetopsis Anderb. & Freire (1 sp.), and (4) Luciliocline Anderb. & Freire (5 spp.). To test the various hypotheses of relationship presented by Anderberg (1991) and Anderberg and Freire (1991), an independent analysis of the "Lucilia group" was undertaken.
Cladistic Analysis of the Lucilia Group
|All cladistic analyses were conducted using 35 discrete characters (cf. Appendix 1 for character list and data matrix) analyzed with PAUP 3.0 (Swofford 1991) and examined using MacClade (Maddison & Maddison 1992). The initial analyses were conducted with individual species of each genus included (cf. Table 1), with the exception of Gamochaeta where representative species were scored. Ultimately, the branches were collapsed into the monophyletic taxa shown in the strict concensus tree Figure 3. The genus Chionolaena (including Leucopholis) was used as the outgroup. A Branch and Bound analysis produced 24 equally parsimonious cladograms with a length of 57 steps, a consistency index of 0.84, a retention index of 0.81, and a rescaled consistency index of 0.68. The strict consensus tree indicates that there are monophyletic groups present in all 24 trees.|
Fig. 3. Strict consensus tree for 24
ResultsThe results of the current analysis suggest quite different relationships within the boundaries of "Lucilia group" than those suggested by previous analyses. Both Lucilia and Gamochaetopsis form a basal grade with no unambiguous characters to distinguish them. The outgroup, Chionolaena (including Leucopholis), though considered by Anderberg (1991) to be a member of the Cassiniinae, shares synapomorphies with Lucilia, including biseriate trichomes (Zwillingshaares, Hess 1938) with enlarged adaxial basal cells. In Chionolaena (Fig. 4), Gamochaetopsis, and Lucilia (Fig. 5), the apical cells of achenial trichomes are elongate and in excess of 150 µm in length, thick-walled, acute, and never rupturing in water. Lucilia (12 species) is left with its original circumscription (Table 2, sensu Anderberg & Freire 1991), plus a suite of caespitose Andean taxa previously included in Belloa or reduced to synonymy (e.g., Lucilia araucana, L. conoidea, L. kunthiana, L. nivea).
Fig. 4. Chionolaena arbuscula. A. Achene
Fig. 5. Lucilia conoidea. A. Achene [660 µm long].
| These cushion-form species are here considered sister taxa to the predominately
eastern Argentine and Brazilian members of Lucilia and represent
the consummation of habit reduction, a trend evident in L.recurvata
and L. flagelliformis. The superficial resemblance of these reduced
members to some Luciliocline taxa (e.g., Luciliocline longifolia,
L. radiata, and L. schultzii) has lead to erroneous groupings.
|The identity of Gamochaetopsis Anderb. & Freire (1991) is problematic. Its type was considered by Cabrera (1961) to be congeneric with Lucilia and stated that it could be confused vegetatively with Gamochaeta nivalis (Phil.) Cabr., but was easily distinguished from the latter taxon by the sericeo-pubescent achenes (i.e., elongate trichomes). In 1971, Cabrera considered Lucilia alpina distinct from both L. araucana and L. nivea. However, all three taxa were described as possessing sericeo-pubescent achenes. Anderberg and Freire (1991) based their new genus, Gamochaetopsis, upon Laennecia alpina Poepp. & Endl., a species originally collected in austral Chile; however, it has not been established whether the type material was examined by the authors and whether it supports the diagnosis. The genus is said to be isolated and contain three autapomorphies (p. 183): "lanate adaxial leaf surfaces, divided stereome [phyllaries], and capselas with short, clavate twin hairs."|
|An examination of type material of Laennecia alpina collected by Poeppig (889, WU) shows it to possess achenes with short clavate trichomes identical to those found in Belloa. The achenial trichomes described as "short, clavate twin hairs" are only known from Belloa chilensis (Hook. & Arn.) Remy (= Lucilia chilensis Hook. & Arn.), and it is possible that material referable to Belloa chilensis was taken for L. alpina. Whether Gamochaetopsis should continue to be recognized will await further study, but it may well prove congeneric with Belloa s.s.|
Facelis, Berroa, and Micropsis always group together in a clade diagnosed as annual herbs with variable capitulescences and achenes with elongate (>150 µm), myxogenic trichomes (Fig. 6). In all these taxa, the achenial trichomes rupture through terminal pores in the apical cells when hydrated. Berroa and Micropsis are sister taxa and are distinguished from Facelis by achenial trichomes with twisted apical cells. Micropsis is further defined by autapomorphies, including paleate involucres where the outer phyllaries enclose the pistillate florets, and achenial trichomes with unfused apical cells.
Fig. 6. Facelis lasiocarpha. A. Achene [1580 µm long].
Fig. 7. Belloa chilensis. A. Achene
The other clade contains Belloa chilensis in a basal position to a trichotomy containing Luciliocline (Belloa p.p. sensu Cabrera 1958), the Jalcophila (sensu Dillon & Sagástegui 1986b)/ Jalcophilaboliviensis/Chevreulia/Cuatrecasasiella clade, and the Gamochaeta / Stuckertiella clade. The "Belloa clade" is diagnosed as possessing achenial trichomes with the apical pair of cells reduced to 50 µm or less in length. The topology of all trees supports the distinction of this clade from the remainder of the "Lucilia group."
Belloa is here considered monotypic and restricted to austral Chile and adjacent Argentina (Table 1). The lectotype collections (Hooker & Arnott 342, K) have been examined and, just as Hooker and Arnott (1835) stated, their material contained only old capitula lacking florets or achenes. The distinctness of this species was recognized by DeCandolle (1838) when he established Lucilia section Luciliodes based upon Lucilia chilensis. Remy (1847), in establishing Belloa, apparently did not examine type material, but clearly stated the primary difference between his Belloa and Lucilia was in the achenial trichomes, with the former genus possessing achenes "papulosa, non villosa" whereas, Lucilia possessed densely villous achenes. This difference in achenial pubescence was again stressed by Cabrera (1958). After examining material annotated by Cabrera, it seems that the achenial trichomes of Belloa (Fig. 7) are not shared by either Lucilia nor Luciliocline.
| Luciliocline consists of 13 species distributed in high-elevation
habitats throughout the Andean Cordillera from Venezuela to Chile and Argentina.
It is diagnosed by heterogamous capitula, pappus bristles fused at the base,
style branches of hermaphroditic florets rounded or obtuse, and achenes
glandular with biseriate, multicellular, capitate-glandular trichomes (Fig.
8). It is notable that this distinctive type of achenial trichome is also
found in Mniodes (Fig. 9). The species composition of Luciliocline
is expanded over that of Anderberg & Freire (1991) and the taxa accepted
here are listed in Table 2.
Fig. 9. Mniodes pulvinulata. A. Achene [970 µm
In the strict consensus tree (Fig. 1), Luciliocline forms an unresolved trichotomy with the Gamochaeta/Stuckertiella clade and the Jalcophila/Chevreulia/Cuatrecasasiella clade. Gamochaeta and Stuckertiella are consistently sister taxa in a clade diagnosed by truncate style branches and achenial pubescence of sessile, paired myxogenic cells. Gamochaeta contains approximately 80 species distributed primarily in the warmer regions of the New World, but with several species adventive in the Old World. Cabrera (1961) resurrected Gamochaeta and described several new species from Argentina. Drury (1970, 1971) analyzed the gnaphaloid elements in New Zealand and reduced Gamochaeta to sectional status. Gamochaeta was accepted by Holub (1976) in Flora Europaea, but Merxmüller et al. (1977) once again treated the genus as a section of Gnaphalium. Anderberg (1991) accepted Gamochaeta and stated correctly that it had little to do with Gnaphalium (s.s.). A listing of South American species can be found in Anderberg (1991) and Dillon and Sagástegui (1991b). Stuckertiella has achenes identical to Gamochaeta, but is diagnosed by several autapomorphies, including involute leaf margins, 4-merous florets, and clavate pappus apices. The 4-merous floret character has been observed in Gamochaeta as well (Díaz-Piedrahíta, pers. comm.). These two genera form the sister group to the Jalcophila (s.l.) /Chevreulia / Cuatrecasasiella clade.
The two northern Andean Jalcophila species (J. peruviana and J. ecuadoriense) were sister taxa in all cladograms. However, the analysis provided several alternative positions for Jalcophila boliviensis Anderb. & Freire (= Lucilia hypoleuca Wedd.), a recently described taxon from southern Bolivia (Anderberg & Freire, 1990). This highly reduced species is here questionably referred to Jalcophila, since it lacks the phyllaries, pappus, achene shape, and achenial trichomes that distinguish J. ecuadoriense and J. peruviana (Fig. 10). Further, the Bolivian species has a single capitulum on an elongate pedicle a character shared with the two northern taxa, but the large capitula with over 40 florets is aberrant in Jalcophila. This species is most likely a Gamochaeta, a genus diagnosed by all of the characters that distinguish J. boliviensis. A population of this rare species has been discovered (J. C. Solomon 4925, MO) and should provide a source of material for molecular studies.
Fig. 10. A. Jalcophila peruviana achene [830 µ m long]. Voucher: Sagastegui et al. 11131, F.
B. Jalcophila ecuadoriense achene [875 µ m long]. (Voucher: Holm-Nielsom 5362, F).
Chevreulia and Cuatrecasasiella are diagnosed as herbs with opposite, distichous leaves, and a persistent pappus. Chevreulia contains six species primarily in austral South America with three reaching the northern Andean Cordillera and is diagnosed with fusiform achenes contracted into a filiform rostrum, barbellate pappus bristles, and biseriate, myxogenic trichomes approximately 40 µm long, with bulbous apical cells. Cuatrecasasiella is diagnosed as dioecious herbs with glabrous achenes; its two species represent closely related North/South disjuncts. |
Historical BiogeographyA large portion of the South American Gnaphaliinae form a monophyletic group that displays a biogeographic pattern reflecting austral origins and subsequent south to north dispersal. The general pattern exhibited by the "Lucilia group" reflects the independent development of northern distributions from austral South American stock. Chionolaena is most diverse in the Atlantic Coastal mountains and adjacent habitats in eastern Brazil, with disjunct taxa in high-elevation sites in Colombia and Venezuela but with no representatives in the intervening Andean Cordillera. The distribution of the Achyrocline / Pseudognaphalium / Stenophalium clade is also congruent with an austral origin and subsequent invasion and radiation in the Andean Cordillera. Achyrocline has its greatest diversity in central and southern South America (Dillon & Sagástegui 1986a) but with secondary centers of speciation in Central America and Mexico (Nesom 1990a). Pseudognaphalium is most speciose in South America, followed by Mexico and the United States. The Gamochaeta clade also finds the greatest species diversity in austral South America with radiation both in the Neotropics and elsewhere.
In all cladograms, the basal lineages in Lucilia are restricted either to Brazil or Argentina and sister group relations occur between taxa in the southern or northern Andean region. It can be hypothesized that as ancestral stocks were forced down slope by recurring glacial cycles, the fragmentation of pre-existing stocks through climate shift and uplift lead to northern and southern vicariance. The retreat of glacial conditions greatly changed the zonation in the alpine environments and the re-establishment of the Humboldt current formed an arid corridor to prevent distribution along the western versant. That Micropsis, Facelis and Berroa are highly derived members of the Lucilia stock is reflected by the results of Anderberg (1991) and the current analysis. These annual herbs represent diversification in seasonally dry habitats which developed repeatedly during Pleistocene and Holocene periods. More recently, they have dispersed throughout Andean South America and ultimately North America, South Africa and Australia, often through the activities of commerce. In other genera, such as Luciliocline, the center of species diversity is in the Andean Cordillera, where typically occur species pairs with sister-species relationships between northern and southern Andean taxa (e.g. L. lopezmirandae and L. catamarcensis) suggesting vicariant speciation events.
The connection of South American taxa to those of New Zealand, Australia, or Tasmania is a recurring pattern in many plant families. It is here suggested that similar selection on independent stocks has lead to convergent habits, but dissimilar floral morphologies. The relationship of South American Gnaphalieae to those of Australasia will require further analysis to support these hypotheses.
References (* added)Anderberg, A. A. 1986. The genus Pegolettia (Compositae, Inuleae). Cladistics, 2: 158- 186.
________. 1988. The genus Anisothrix O. Hoffm. (Compositae-Inuleae). Bot. Jahrb. Syst. 109: 363-372.
________. 1989. Phylogeny and reclassification of the tribe Inuleae (Asteraceae). Canadian J. Bot. 67: 2277-2296.
________. 1991. Taxonomy and phylogeny of the tribe Gnaphalieae (Asteraceae). Opera Bot. 104: 1-195.
________. 1992. Cladistics of the Gnaphalieae, Additional data. Comp. Newsl. 20/21: 35.
________. 1994. Tribe Gnaphalieae. In K. Bremer, Asteraceae, Cladistics and Classification, pp. 304-364. Timber Press, Portland, Oregon.
________. & S. Freire. 1990. Jalcophila boliviensis, a new species of South American Asteraceae (Gnaphalieae). Brittonia, 42: 138-141.
________.& ________. 1991. A cladistic and biogeographic analysis of the Lucilia group (Asteraceae, Gnaphalieae). J. Linn. Soc. Bot. 106: 173-198.
Bentham, G. 1873. Notes on the classification, history, and geographical distribution of the Compositae. J. Linn. Soc. (Bot.) 13: 335-577.
Bremer, K. 1976a. The genus Relhania (Compositae). Opera Bot. 40: 1-85.
________. 1976b. The genus Rosenia (Compositae). Bot. Notiser, 129: 97-111.
________. 1978a. The genus Leysera (Compositae). Bot. Notiser, 131: 369-383.
________. 1978b. Oreoleysera and Antithrixia, new and old South African genera of the Compositae. Bot. Notiser, 131: 449-453.
________. 1987. Tribal interrelationships of the Asteraceae. Cladistics, 3: 210-253.
________. 1994. Asteraceae, Cladistics and Classification. Timber Press, Portland, Oregon. 752 pp.
Cabrera, A. L. 1958. El género Belloa Remy. Bol. Soc. Argent. Bot. 7: 79-85.
________. 1961. Observaciónes sobre las Inuleae-Gnaphalineae (Compositae) de América del sur. Bol. Soc. Argent. Bot. 9: 359-386.
________. 1978. Compositae. Flora de la Provincia de Jujuy. Colecc. Cient., Inst. Natl. Tecn. Agropec. 13(10): 275-288.
________. 1971. Compositae. Flora Patagónica. Colecc. Cient., Inst. Natl. Tecn. Agropec. 8(7): 1-451.
________. 1978. Compositae. Flora de la Provincia de Jujuy. Colecc. Cient., Inst. Natl. Tecn. Agropec. 13(10): 294-301.
Cronquist, A. 1950. A review of the genus Psilocarphus. Res. Stud. State Coll. Wash. 18(2): 71-89.
DeCandolle, A. P. 1838. Lucilia. Prodr. 7(1): 46.
Dillon, M. O. 1988. Generic limits and microcharacters in the South American Gnaphaliinae (Asteraceae-Inuleae). Amer. J. Bot. Abst. 75: 456.
________. 1990. A tale of two genera: the implication of character choices and outgroup selection in cladistic analysis. Amer. J. Bot. Abst. 77: 129.
________. & A. Sagástegui A. 1986a. A new species of Achyrocline (Inuleae- Asteraceae). Phytologia, 59: 107-110.
________. & ________. 1986b. Jalcophila, a new genus of Andean Inuleae (Asteraceae). Brittonia, 38: 162-167.
________. & ________. 1990. Oligandra Less. revisited and the need for a new genus, Pseudoligandra (Asteraceae: Inuleae). Taxon, 39: 125-128.
________. & ________. 1991a. Family Asteraceae: Part V. Tribe Inuleae. In, J. F. Macbride & Collaborators, Flora of Peru, Fieldiana: Botany, N.S. 26, 1-70.
________. & ________. 1991b. Synopsis de los géneros de Gnaphaliinae (Asteraceae-Inuleae) de Sudamérica. Arnaldoa, 1: 5-91.
Drury, D. G. 1970. A fresh approach to the classification of the genus Gnaphalium with particular reference to the species present in New Zealand (Inuleae-Compositae). New Zealand J. Bot. 8: 222-248.
________. 1971. The American spicate cudweeds adventive to New Zealand. New Zealand J. Bot. 9: 157-185.
Freire, S. E. 1986. Revisión del género Lucilia (Compositae, Inuleae). Darwiniana, 27: 431-490.
________. 1987. A cladistic analysis of Lucilia Cass. (Compositae, Inuleae). Cladistics, 3: 254-272.
________. 1989. Oligandra Less. is Lucilia Cass. (Compositae, Inuleae). Taxon, 38: 298-299.
________. 1993. A Revision of Chionolaena (Compositae, Gnaphalieae). Ann. Missouri Bot. Gard. 80: 397-438.
Hess, R. 1938. Vergleichend Untersuchungen ¸ber die Zwillingshaare der Compositen. Bot. Jahrb. Syst. 68: 435-496.
Hilliard, O. M., and B. L. Burtt. 1981. Some generic concepts in Compositae - Gnaphaliinae. J. Linn. Soc. Bot. 82:181-232.
Holub, J. 1976. Gamochaeta. In, Tutin, T. G., et al. (eds.), Flora Europaea, 4: 127.
Hooker, W. J. & G. A. W. Arnott. 1835. Contributions towards a Flora of South America and the Islands of the Pacific. Comp. Bot. Mag. 1: 102-111.
Jansen, R. K., K. E. Holsinger, H. J. Michaels, & J. D. Palmer. 1990. Phylogenetic analysis of chloroplast DNA restriction site data at higher taxonomic levels: an example from the Asteraceae. Evolution, 44: 2089-2105.
________., H. J. Michaels, & J. D. Palmer. 1991. Phylogeny and character evolution in the Asteraceae based on chloroplast DNA restriction site mapping. Syst. Bot. 16: 98-115.
Keeley, S. C., and R. K. Jansen. 1991. Evidence from chloroplast DNA for the recognition of a new tribe, the Tarchonantheae, and the tribal placement of Pluchea (Asteraceae). Syst. Bot. 16: 173-181.
Maddison, W. P., and D. R. Maddison. 1992. MacClade, Analysis of Phylogeny and Character Evolution, Ver. 3.0, Sinauer Associates, Inc., Sunderland, Mass.
Merxmüller, H., P. Leins and H. Roessler. 1977. Inuleae--Systematic review. In V. H. Heywood et al. (eds.), The Biology and Chemistry of the Compositae, pp. 577-602. Academic Press, London.
Morefield, J. D. 1992. Evolution and systematics of Stylocline (Asteraceae: Inuleae). unpubl. diss., 73 pps.
Nesom, G. L. 1989. New species, new sections, and a taxonomic overview of American Pluchea (Compositae: Inuleae). Phytologia, 67: 158-167.
________. 1990a. Taxonomy of Achyrocline (Asteraceae: Inuleae) in Mexico and Central America. Phytologia, 68: 181-185.
________. 1990b. Taxonomy of Gnaphaliothamnus (Asteraceae: Inuleae). Phytologia, 68: 366-381.
________. 1990c. An additional species of Gnaphaliothamnus (Asteraceae: Inuleae) and further evidence for the integrity of the genus. Phytologia, 69: 1-3.
________. 1994. Comments on Gnaphaliothamnus (Asteraceae: Inuleae). Phytologia, 76: 185-191.
*________. 2001. New combinations in Chionolaena (Asteraceae: Gnaphalieae). Sida 19: 849-852.
Raven, P. H. and D. I. Axelrod. 1974. Angiosperm biogeography and past continental movements. Ann. Missouri Bot. Gard. 61: 539-673.
Remy, J. 1847. Compositae. In C. Gay, Historia Física y Política de Chile, Botánica, 3: 257-482.
Swofford, D. L. 1991. Phylogenetic Analysis Using Parsimony (PAUP), version 3.0s. Illinois Natural History Survey, Champaign.
Turner, B. L. 1977. Fossil history and geography. In V. H. Heywood et al. (eds.), The Biology and Chemistry of the Compositae, pp. 21-39. Academic Press, London.
Appendix 1. Character and Data MatrixA wide array of morphological characteristics have been examined, including habit, capitulescence type, and capitular and floral morphology (Dillon & Sagástegui 1991). Air-dried herbarium collections provided material for various types of microscopy. Material for light microscopy (LM) was treated with a wetting agent (Aerosol OT, 10%) and permanent slide mounts prepared using Hoyer's medium. Measurements were taken from the material prepared for LM. Material utilized for scanning electron microscopy (SEM) was mounted on stubs and gold coated without pretreatment. Voucher specimens are all housed at F unless otherwise noted.
1. Duration: (0) perennials, (1) annuals. 2. Lignification: (0) subshrubs, (1) herbs. 3. Leaf arrangement: (0) alternate, (1) opposite. 4. Leaf distribution: (0) evenly cauline, (1) basal rosette. 5. Leaf ranking: (0) non-distichous, (1) distichous. 6. Leaf margins: (0) revolute, (1) plane, (2) involute. 7. Dorsal leaf surfaces: (0) non canaliculate, (1) canaliculate. 8. Leaf pubescence: (0) both tomentose, (1) abaxial tomentose, (2) adaxial tomentose. 9. Apical tuft on leaf: (0) absent, (1) present. 10. Foliar trichome apical cells: (0) swollen, (1) straight. 11. Capitulescences: (0) solitary, (1) glomerules, (2) spiciform, (3) panicles, (4) racemes. 12. Pedicles: (0) static, (1) elongating. 13. Capitular sexuality: (0) perfect (heterogamous), (1) polygamo-dioecious (heterogamous), (2) dioecious (homogamous), (3) perfect (homogamous). 14. Gynoecia (hermaphrodite): (0) fertile, (1) sterile. 15. Receptacles: (0) epaleate, (1) paleate. 16. Outer phyllaries and florets: (0) not enclosing, (1) enclosing. 17. Hermaphrodite florets: (0) 5-merous, (1) 4-merous. 18. Hermaphrodite style apices: (0) attenuate, (1) obtuse to rounded, (2) truncate. 19. Hermaphrodite styles dorsally: (0) pubescent or papillose, (1) glabrous. 20. Anther apical appendages: (0) isomorphic, (1) dimorphic. 21. Pappus bristles (pistillate): (0) acute to attenuate, (1) obtuse to rounded, (2) clavate. 22. Pappus bristles (hermaphrodite): (0) acute to attenuate, (1) obtuse to rounded, (2) clavate. 23. Trichome sizes: (0) elongate, (1) elongate-globose, (2) capitate-globose, (3) sessile. 24. Pappus condition: (0) deciduous, (1) persistent. 25. Achene shape: (0) oblong to oval, (1) rosulate. 26. Achene surface: (0) trichomes, (1) glabrous 27. Achenial trichome apices: (0) elongate, (1) capitate, (2) sessile. 28. Achenial trichome apical cells: (0) non-myxogenic, (1) myxogenic. 29. Achenial trichome apical cells: (0) untwisted, (1) twisted. 30. Achenial trichomes: (0) non-dehiscent, (1) dehiscent by cell splitting, (2) dehiscent by apical pores. 31. Achenial trichome bases: (0) enlarged myxogenic c-cell, (1) enlarged cells absent. 32. Achenial trichome types: (0) monomorphic, (1) dimorphic. 33. Trichome apical cells: (0) fused unequal, (1) fused equal, (2) unfused equal. 34. Carpopodium: (0) multiseriate, (1) 1-3-seriate. 35. Chromosome number: (0) x = 7, (1) x = 6.
|Subtribe||"Group"||Genus||Total Number||No. in SA||Distribution|
|Cassiniinae||n/a (~"Lucilia")||Anaphalis||24-30 (~65)||1||Holarctic, mainly Asia|
|Cassiniinae||n/a (~"Lucilia")||Antennaria||ca. 50||3||Holarctic|
|20||15||Central & South America|
|Loricariinae||n/a||Loricaria||18||18||Andean South America|
|Loricariinae||n/a||Mniodes||4||4||Andean South America|
|Gnaphaliinae||"Helichrysum"||Achyrocline||ca. 30||23||Central & South America, Africa|
|Gnaphaliinae||"Fliago"||Psilocarpus||5||1||North & South America|
|Gnaphaliinae||"Lucilia"||Berroa||1||1||Austral South America|
|Gnaphaliinae||"Lucilia"||Chevreulia||6||6||Austral and Andean South America|
|Gnaphaliinae||"Lucilia"||Cuatrecasasiella||2||2||Andean South Aamerica|
|Gnaphaliinae||"Lucilia"||Facelis||4||4||South America (adventive)|
|Gnaphaliinae||"Lucilia"||Gamochaeta||80||40||South America (adventive)|
|Gnaphaliinae||"Lucilia"||Jalcophila||2(3)||2(3)||Andean South America|
|Gnaphaliinae||"Lucilia"||Luciliocline||13||13||Andean & Austral South America||Gnaphaliinae||"Lucilia"||Lucilia||12||12||Andean & Austral South America|
|Gnaphaliinae||"Lucilia"||Gamochaetopsis||1||1||Austral South America|
|Gnaphaliinae||"Lucilia"||Micropsis||5||5||Austral South America (adventive)|
|Genus||fide Anderberg & Freire (1991)||fide Dillon (2000)|
|Belloa Remy in Gay
Type: Lucilia chilensis Hook. & Arn.
[=Belloa chilensis Remy in Gay]
|Belloa chilensis (Hook. & Arn.) Remy in Gay
B. kunthiana (DC.) Anderb. & S.E.Freire
B. lehmannii (Hieron.) Anderb. & S.E.Freire
B. longifolia (Cuatrec. & Aristeg.) Sagást. & M.O.Dillon
B. pickeringii (A.Gray) Sagást. & M.O.Dillon
B. piptolepis (Wedd.) Cabr.
B. plicatifolia Sagást. & M.O.Dillon
B. radians Sagást. & M.O.Dillon
B. schultzii (Wedd.) Cabrera
|Belloa chilensis (Hook. & Arn.) Remy in Gay|
|Luciliocline Anderb. & Freire
Type: Belloa lopezmirande Cabr.
[= Luciliocline lopezmirande
(Cabrera..) Anderb. & Freire]
|L. burkartii (Cabrera) Anderb. & S.E.Freire
L. catamarcensis (Cabrera) Anderb. & S.E.Freire
L. lopezmirandae (Cabrera) Anderb. & S.E.Freire
L. santanica (Cabrera) Anderb. & S.E.Freire
L. subspicata (Wedd.) Anderb. & S.E.Freire
|L. burkartii (Cabr.) Anderb. & S.E.Freire
L. catamarcensis (Cabr.) Anderb. & S.E.Freire
L. longifolia (Cuatrec. & Aristeg.) comb. nov.*
L. lopezmirandae (Cabr.) Anderb. & S.E.Freire
L. pickeringii (Gray) comb. nov. *
L. piptolepsis (Wedd.) comb. nov.*
L. plicatifolia (Sagást. & M.O.Dillon) comb. nov.*
L. radians (Benth.) comb. nov.*
L. schultzii (Wedd.) comb. nov.*
L. spathulifolia (Sagást. & M.O.Dillon) comb. nov.*
L. subspicata (Wedd.) Anderb. & S.E.Freire
L. turnerii (Sagást. & M.O.Dillon) comb. nov.*
Type: Serratula acutifolia Poir.
[=Lucilia acutifolia (Poir.) Cass.]
|L. acutifolia (Poir.) Cass.
L. eriophora Remy
L. ferruginea Bak.
L. linearifolia Bak.
L. lycopodioides (Less.) Freire
L. nitens Less.
L. recurvata Wedd.
L. tomentosa Wedd.
|L. acutifolia (Poir.) Cass.
L. araucana Phil.
L. conoidea Wedd.
L. eriophora Remy
L. ferruginea Bak.
L. kunthiana (DC.) Zardini
L. linearifolia Bak.
L. lycopodioides (Less.) Freire
L. nitens Less.
L. nivea (Phil.) Cabr.
L. recurvata Wedd.
L. tomentosa Wedd.
Figure 2. Parachionolaena colombiana. A. Achene [1000 µm long]. B. Achene surface enlargement. C. Pappus bristle apex enlargement. (Voucher: Barclay 6927, COL).
Figure 3. Strict consensus tree for 24 equally parsimonious cladograms.
Figure 4. Chionolaena arbuscula. A. Achene [700 µm long]. B. Achenial trichomes enlargement. C. Carpopodium [190 µm diameter]. (Voucher: Barreto 9186, F). Chionolaena (Leucopholis) capitata. D. Achene [2300 µm long]. E. Achenial trichomes enlargement. F. Carpopodium enlargement. (Voucher: Brade 17012, F).
Figure 5. Lucilia conoidea. A. Achene [660 µm long]. B. Achenial trichome enlargement. (Voucher: Dillon et al. 1082, F).
Figure 6. Facelis lasiocarpha . A. Achene [1580 µm long]. B. Fused pappus base. C. Plumose pappus bristle enlargement. (Voucher: Macbride & Featherstone 446, F).
Figure 7. Belloa chilensis. A. Achene [1150 µm long]. B. Achene surface enlargement. C. Carpopodium enlargement. (Voucher: Teillier et al. 2020, F).
Figure 8. Luciliocline longifolia. A. Achene [1280 µm long]. B. Achenial trichomes enlargement. (Voucher: Sagástegui et al. 12841, F).
Figure 9. Mniodes pulvinulata. A. Achene [970 µm long]. B. Achenial trichome enlargement. (Voucher: Sánchez 2558, F).
Figure 10. A. Jalcophila peruviana achene [830 µ m long]. Voucher: Sagastegui et al. 11131, F. B. Jalcophila ecuadoriense achene [875 µ m long]. (Voucher: Holm-Nielsom 5362, F).
Table 2. Comparison of Species-level Composition of Belloa, Lucilia, and Luciliocline. (Species not recognized by Anderberg & Freire 1991 are in boldface; *new combinations)