zyxwvutsrqpo zyxwvutsrqp zyxwvuts zyxwv zyxw zy zy Bolanical Journal o f the Linnean Sociely (1993), 112: 293-309. With 6 figures A cladistic analysis of Nassauvia Comm. ex Juss. (Asteraceae, Mutisieae) and related genera SUSANA E. FREIRE, JORGE V. CRISCI AND LILIANA KATINAS Departamento Cient$co de Plantas Vasculares. Museo de La Plata, Paseo del Bosque sln., 1900 La Plata, Argentina zyxwvuts Received April 1992, accepted for publication December 1992 FREIRE, S. E., CRISCI, J. V. & KATINAS, L., 1993. A cladistic analysis of Nassauvia Comm. ex Juss. (Asteraceae, Mutisieae) and related genera. Nassauuia and the most closely related genera Calopappus and Triptilion from the southern Andes and Patagonia of South America, form a rnonophyletic group diagnosed by the following synapomorphies: cypsela trichomes single two-celled, cypsela testa with strengthened cells, pollen grains spheroidal to spheroidal-oblate, colpi membrane with sexine processes, pappus bristles two to six, and pappus deciduous. Furthermore, “Wassauuia, Triptilion, and Calopappus form a group with two other Andean genera, Moscharia and Polyachyrus, diagnosed by occurrence of pseudocephalia and a reduction in the number of flowers to five, three or one. A cladistic analysis of the group was undertaken using 35 characters from morphology, anatomy, and palynology. The monophyletic terminal taxa were the 38 species of Nassauuia, the genus Triptilion, the monotypic genus Calopappus, the genus Polyachyrus; and the genus Moscharia. Character polarity was based on outgroup comparison using Cephalopappus. The analysis resulted in 223 equally parsimonious cladograms, each with 70 steps and a consistency index of0.57. A successive weighting procedure was applied, resulting in 15 cladograms with a consistency index of 0.82. Results of the cladistic analysis support most of the current systematic classification of Nassauuia, with three exceptions: ( 1) Nassauuia (excluding Calopappus) is paraphyletic; ( 2 ) section Masligophorus appears to be a polyphyletic group ( N ,pygmaea does not cluster with the remaining species of the section); (3) section Panargyrum (without N. lagascae = section Caloptilium) appears to be a paraphyletic group. The capitula arranged in cymose conflorescences in Triptilion are regarded as a primitive condition which gave rise to all stages present in Nassauuia (conflorescence spicate, pseudocephalium, capitula solitary), The capitula arranged in pseudocephalia in Moscharia and Polyac/yrus are regarded as a parallel development to the pseudocephalium found in Nassauuia. Nassauuia, subgenus Strongyloma appears as the most primitive taxon, with its spicate conflorescence, whereas section Mastigophorus with its solitary capitulum is thought-derived. These results correspond well with cytological data where species of the subgenus Strongyloma have n = 1 I and the species of section Masfigophorus are tetraploids ( n = 22). ADDITIONAL KEY WORDS:-Cytology ~ evolution of heads ~ phylogeny. CONTENTS Introduction . . . . . . . Material and methods. . . . . Results . . . . . . . . Discussion . . . . . . . Placement of Calopappus . . . Evolution of flowering heads . . Pappus . . . . . . . Cytology . . . . . . . Geographic distribution and ecology 0024-4074/93/080293 + 17 W8.OOjO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 294 298 304 304 305 305 307 307 zyxwvut 293 0 1993 The Linnean Society of London 294 zyxwvut zyxwvutsrq zyxwvu zyxwvutsr zyxwvu S. E. FREIRE E T AL. Acknowledgements References . . . . . . . . . . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . 308 308 zy Nassauvia Comm. ex Juss. is found in the southern Andes and Patagonia of South America. T h e 38 species of Nassauvia (Arroyo & Marticorena, 1988; Cabrera, 1982) have been divided into two subgenera and four sections based on plant growth habit, cypsela morphology, pappus, and leaf shape (Cabrera, 1.c.). Although some hypotheses of relationships between species have been postulated (Cabrera, 1982; Hunziker et al., 1991; Stuessy et al., 1988), their genealogy is still unclear. Nassauvia, with Triptilion, Calopappus, Moscharia and Polyachyrus, represent an interesting group within the subtribe Nassauviinae (Crisci, 1974a, 1980) because of the several disjunct distribution patterns of closely related species and advanced characters within the family, such as the occurrence of secondary capitula or pseudocephalia in several species. Nevertheless, Nassauvia, Triptilion and Calopappus form a monophyletic group based on several synapomorphies (cypsela trichomes single two-celled, cypsela testa with strengthened cells, pollen grains spheroidal to spheroidal-oblate, colpi membrane with sexine processes, pappus bristles two to six, pappus deciduous). This paper presents a cladistic analysis of Nassauvia using a Wagner parsimony algorithm (Farris et al., 1970). To test the monophyly of Nassauvia, we have included the most closely related genera Triptilion, Calopappus, Polyachyrus, and Moscharia as part of the ingroup. All of them have been recently monographed (Nassauvia: Cabrera, 1982; Calopappus: Crisci & Freire, 1986; Triptilion: Katinas et al., 1992; Polyachyrus: Ricardi & Weldt, 1974; Moscharia: Crisci, 1974b). MATERIAL AND METHODS Nassauvia, Triptilion, Calopappus, Moscharia and Polyachyrus belong to the subtribe Nassauviinae (Crisci, 1974a), which is delimited from the rest of the Mutisieae by the following three characters: disc florets bilabiate and style branches truncate a t the apex with a crown of collector hairs. The monophyletic terminal taxa of this analysis are: (1) the genus Polyachyrus (seven species), which is characterized as a monophyletic group by the presence of pseudocephalia composed of up to ten heads disposed at different levels; (2) the genus Moscharia (two species), which can be also delimited from all other genera of the subtribe by the pseudocephalia composed of eight to ten heads attached a t the same level; (3) the genus Triptilion (seven species) which is characterized as a monophyletic group by a pappus formed of four to five paleae plicate along the median line, and expanded and laciniate in the upper part; (4)the monotypic genus Calopappus; (5) the 38 species of Nassauvia. Table 1 shows their geographical distribution. Data from 35 characters were derived from plant habit, leaf morphology, growth characteristics, capitula arrangement, capitula morphology, cypsela morphology, pappus morphology and pollen grain type (Cabrera, 1982; Grau, 1980; Parra & Marticorena, 1972). Table 2 shows the characters and character states used in the cladistic analysis. z zyxwvu zy zyxwvut CLADISTICS OF NASSAUVIA 295 TABLE 1. Taxa studied, acronyms, and geographical distributions Taxa Acronym Geographical distribution (Cabrera, 1982; Crisci, 1974a) Calopappus Cephalopappus Macrachaenium Moscharia CAL CEP MAC MOS central Chile northeastern of Brazil (Bahia) Nothofagus forest central Chile Nassauuia Subgen. Jvassauuia Sect . Caloptilium N . lagascae LAG Argentina (S Mendoza to Santa Cruz), Chile (Aconcagua to Cautin, and Magallanes) Sect . MastigophoruJ N . ameghinoi N . gaudichaudii N. hilii N. juniperina N . pentacaenoides N. pygmaea AME GAU HIL JUN PEN PYG Argentina (S Chubut and Santa Cruz) Argentina (Falkland Islands = Mas Malvinas) Argentina (Neuqutn) Argentina (Chubut, Santa Cruz) Argentina (Santa Cruz) Argentina (Neuqutn to Tierra del Fuego, Isla de 10s Estados), Chile (Ruble to Magallanes) Sect. Nassauuia N. argentea N. argvrophylla N. coronipappa N. chubutensis N. cumingii AGE AGY COR CHU CUM Argentina (Neuqutn), central Chile Argentina (Neuqutn to Chubut) Chile (Ultima Esperanza) Argentina (Chubut) Argentina (S San Juan to Mendoza), Chile (Coquimbo to Santiago) Argentina (Neuqutn, Rio Negro, Chubut), Chile (Llanquihue, Magallanes) Argentina (Neuqutn), Chile ( T a k a to Bio-Bio) Argentina (Rio Negro to S Chubut) Argentina (S Tierra del Fuego, Isla de 10s Estados) Argentina (Santa Cruz, Tierra del Fuego) Argentina (Mendoza, Neuqukn), Chile (Santiago to Curicb) Argentina (S Mendoza, N Neuquin) Argentina (S Neuqutn, NW Rio Negro) Argentina (Mendoza, N Neuqukn), Chile (Santiago to Bio-Bio) central Chile (Linares) Argentina (Mendoza to Santa Cruz), Chile (Aconcagua to Magallanes) Argentina (Mendoza) Argentina (Santa Cruz) Argentina (Falkland Islands = Mas Malvinas) Chile (Curic6 to Malleco), Argentina Argentina (S Mendoza, N Neuqukn) N. dentata DEN N. digitata N . dusmii N . latissima N . magellanica N . pinnigera N . planifolia N . pulcherrina N. pyramidalis N . ramosissima N. revoluta DIG DUS LAT MAG PIN PLA PUL PYR RAM REV N. ruizii N. sceptrum N. serpens N. sprengeiioides N . sublobata RUI SCE SER SPR SUB Sect. Panargyrurn N . aculeata ACU zyx N . darwinii N . glomerata N. looseri N. unijlora DAR GLT LOO UNI Argentina (Mendoza to Tierra del Fuego), Chile (Aconcagua to Magallanes and Aisen) Argentina (Neuqutn to Tierra del Fuego), Chile Argentina (N Mendoza, N Neuqukn), Chile (Santiago to T a k a ) Chile (Santiago) Argentina (S San Juan, N Mendoza), Chile (Aconcagua) Subgen. Stronuloma N . axillaris N. fuegiana N . glomerulosa N . maeuiae N . uticina AX1 FUE GLS MAE ULI S Bolivia to central Chile and Argentina (N Chubut) Argentina (S Mendoza to N Tierra del Fuego) Argentina (S Mendoza to Santa Cruz) Argentina (S Santa Cruz) Argentina (Rio Negro to Santa Cruz) Polyachyru POL S Peru, central Chile Triptilion TRI central Chile, Argentina (Neuqutn to Santa Cruz) 296 zyxwvutsrqpo zyxwvut zyxwvutsrqp zyxwv S. E. FREIRE E T A L . TABLE 2. Characters and character states used in cladistic analysis of Nussauviu and allied genera Characters Character states 1 Habit 2 Life cycle 3 Limb of leaves multicostate 4 Brachyblast 5 Leaves 6 Shape of leaf 7 Shape of leaf base 8 Auricles 9 Cross section of leaf 10 Basal leaves in rosettes 11 Phyllotaxy 12 Type of inflorescence 0. Herbs; I . Shrubs 0. Perennial; 1. Annual 0. Absent; 1. Present 0. Absent; I . Present 0. Entire; 1. Partite 0. Obovate, ovate; 1. Linear-subulate 0. Non-vaginate; 1. Vaginate 0. Absent; 1. Present 0. Aplanate; 1. Circular 0. Present; 1. Absent 0. Spreading; 1. Crowded 0. Conflorescence cymose (paniculate, spicate or corymbose); I . Pseudocephalium; 2. Solitary 0. Absent; 1. Present 0. Conical; 1. Convex 0. Monomorphic; 1. Dimorphic 0. In branched cyme; 1. Solitary 0. 2; 1. I ; 2. 3 0. Outer bracts longer than inner; 1. All the same length 0. More than 5; 1. Less than 5 0. Less than 15 mm; 1. More than 17 mm 0. 5 (rarely 4); 1. 1-2 (rarely 3) 0. White; 1. Violet or pink 0. Present; 1 . Absent 0. Double trichomes ‘Zwillingshaare’; 1. Single trichomes twocelled 0. External layer of the testa with epidermal cells not strengthened; 1. Lateral and basal walls of the testa epidermis strengthened 0. Subprolate to prolate-spheroidal; 1. Spheroidal to spheroidal-oblate 0. Smooth; 1. With sexine processes 0. Infratectum thickness; 1. Tectum and infratectum equal thickness 0. Plumose; 1. Entire or ciliate 0. Completely plumose; I . Plumose on the distal half 0. 10 or more; I . 2-6 0. Flat; 1. Plicate 0. Hairy (less than 0.1 mm wide); 1. Chaffy (0.2 to 1 mm wide); 2. Paleaceous (more than 1.5 mm) 0. Persistent; 1. Deciduous 0. Pappus and corollas (upper lip) equal; 1. Pappus shorter than corollas (upper lip); 2. Pappus reduced or absent 13 Pseudoinvolucre 14 Receptacle of the pseudocephalium 15 Capitula of the pseudocephalium 16 Type of pseudocephalia 17 Number of series of involucre 18 Length of involucral bracts 19 Number of bracts of involucre 20 Height of involucre 21 Number of flowers per capitulum 22 Colour of corollas 23 Cypselas pubescent 24 Cypsela hairs 25 Cross-section of cypsela testa 26 Shape of pollen grains 27 Colpus membrane 28 ’rectum and infratectum 29 Margin of pappus bristles 30 Pappus plumose 3 1 Number of pappus bristles 32 Surface of pappus bristles 33 Type of pappus 34 Pappus duration 35 Length of bristles pappus zyxwvu zyxw Character polarity was determined by outgroup comparison (Humphries & Funk, 1984; Maddison et al., 1984; Watrous & Wheeler, 1981) with Cephalopappus as the outgroup which, according to Crisci (1980), appears at the base of the tree branch of Nassauvia, Triptilion, Calopappus, Moscharia and Polyachyrus. Nassauvia, related genera and Cephalopappus form a monophyletic group diagnosed by a reduction in the size of capitula and the reduction in the size or number of pappus bristles. Because the pappus is absent in Cephalopappus, we used Macrachaenium as a second outgroup (Karis et al., 1992). However, identical results were obtained by using ‘question marks’ for these characters involving the pappus in Cephalopappus. Table 3 contains the data matrix used in this analysis. Four of the 35 characters (12, 17, 33 and 3 5 ) had more than two character states. Characters zyxwvut zyxwvu zyxwv zyxwvutsrq zyxwvuts zyxwvut zyx CLADISTICS OF NASSAUVIA 297 TABLE 3. Data matrix Taxa 1 2 3 4 5 6 7 8 9 10 I 1 1 OUT 2 CAL 3 MOS 4 POL 5 TRI 6 ACU 7 AGE 8 AGY 9 AME 10 AX1 1 1 COR 12 CUM 13 CHU 14 DAR 15 DEN 16 DIG 17 DUS 18 FUE 19 GAU 20 GLS 21 GLT 22 HIL 23 JUN 24 LAG 25 LAT 26 LOO 27 MAE 28 MAG 29 PEN 30 PIN 31 PLA 32 PUL 33 PYG 34 PYR 35 RAM 36 REV 37 RUI 38 SCE 39 SER 40 SPR 41 SUB 42 ULI 43 UNI 12 13 1415 1617 18 19202122232425262728293031 3233 3435 0 0 0 0 0 0 0 0 0 0 0 O ? ? ? ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 1 1 2 ? ? ? ?2 0 0 1 0 0 I ? 1 1 0 1 1 2 0 0 1 0 0 0 1 0 0 1 ? 0 1 0 0 0 1 1 1 1 0 l ? 1 0 1 1 0 0 0 0 1 0 0 0 0 0 1 0 2 0 1 0 0 1 ? 0 1 0 0 0 1 0 0 0 0 1 ? 0 0 1 1 0 0 0 0 1 0 0 0 0 0 1 0 1 0 1 0 0 1 ? 0 0 0 0 0 0 ? ? ? ?0 1 0 0 0 1 0 1 1 1 1 1 0 1 I 1 1 1 1 0 0 0 0 0 1 0 0 0 1 1 0 ? ? ? ? 0 0 0 0 0 0 1 ? 1 I 1 1 0 0 0 0 0 0 0 0010000001 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 I 1 0010000001 1 1 0 0 0 1 0 0 0 0 0 0 I ? 1 1 1 1 1 ? 1 0 1 1 1 0000011011 1 2 ? ? ? ? 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 0 100101100 1 1 O ? ? ? ? 0 0 0 0 0 0 0 1 1 1 1 1 1 ? 1 0 1 1 1 0 0 ? 0 0 0 0 0 ? 1 1 1 0 0 0 1 0 0 0 0 1 0 1 ? 1 1 1 1 l ? 1 1 2 0 1 0 0 ? 0 0 0 0 0 0 I 1 I 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 0010100001 ? O ? ? ? ? 0 0 0 0 0 0 1 ? I 1 1 1 1 ? 1 0 1 1 1 000001000 1 ? 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 0 0 0 0 0 0 1 0010000001 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 0 0 1 0 1 ? 0 0 0 1 1 1 0 0 0 1 0 0 0 0 0 0 l ? 1 1 1 1 1 ? 1 0 1 1 1 001000000 I 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 100101100 1 I O ? ? ? ? 0 0 0 0 0 0 0 1 1 1 1 1 1 ? 1 0 1 1 1 000001101 I I Z ? ? ? ? 0 0 0 0 0 0 1 ? 1 1 1 1 I ? 0 0 1 1 0 100101100 I I O ? ? ? ? 0 0 0 0 0 0 0 1 1 1 1 1 1 2 1 0 1 1 1 0 0 ? 0 1 ? 0 0 ?I I 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 I ? 0 0 0 0 0 000001101 1 I 2 ? ? ? ? 0 0 0 0 0 0 1 ? 1 1 I I I ? 1 0 1 1 0 000001 101 1 I 2 ? ? ? ? 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 I I 0 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 0 0 0 0 0 0 0 0010000001 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 2 1 0 1 1 1 00?00100?1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 0 0 0 O ? 1001011001 l o ? ? ? ? 0 0 0 0 0 0 0 1 1 1 1 1 1 ? 1 0 1 1 1 0010000001 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 000001101 1 1 Z ? ? ? ?0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 0 1 2 1 0 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 I ? 1 1 1 1 1 ? 1 0 I 1 ? 000000100 1 1 1 0 0 0 1 0 0 0 0 0 0 I ? 1 1 1 I 1 ? 1 0 1 1 0 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 I ? 1 1 1 1 1 ? 1 0 1 1 1 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 I 1 1 ? 1 0 1 1 ? 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 001000000 1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 I 1 1 001000?00 1 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 001000000 I 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 l ? 1 0 1 1 1 001000000 I 1 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 1 ? 1 0 1 1 1 00101?0001 I 1 0 0 0 1 0 0 0 0 0 0 1 ? 1 1 1 1 I ? I 0 1 1 1 1001011001 I O ? ? ? ? 0 0 0 0 0 0 0 1 1 1 1 1 I ? I 0 1 1 1 0 0 0 0 1 ? 0 0 0 I I 2 ? ? ? ? 0 0 0 O ? 0 1 ? 1 1 I I I ? 0 0 0 0 0 zyxwvutsrqpo 17, 33 and 35 were treated as non-additive (equivalent to ‘unordered’ in other analyses). Character 12 (type of inflorescence) was treated as additive. Despite the fact that for some authors (e.g. Hauser & Presch, 1991) the hypothesis of order in character changes is unnecessary, we believe that character 12 should be treated as ordered because as was pointed out by Cabrera (1982) the changes from cymose conflorescence to solitary capitula seem to be more parsimonious when it has a transformation series through the pseudocephalium. Characters 6, 13-16, 24, 30 are related to characters 5, 12, 23 and 29, respectively. Therefore we have coded those taxa with partite leaves (character 5, state l ) , without a pseudocephalium (character 12, state 0 or 2), with glabrous cypselas (character 23, state l ) , with entire or ciliate pappus bristles (character 29, state 1) as ‘?’ in characters 6, 13-16, 24 and 30. 298 zyxwvut zyxwvu S. E. FREIRE ET AL. The data were analysed using Farris’s phylogenetic program, Hennig86 version 1.5 (Farris, 1988), applying the options ‘mhennig’ for constructing multiple initial cladograms, followed by ‘bb*’ (branch breaker), an heuristic for determining the shortest tree. We also used the successive approximations weighting procedure in Hennig86, which calculates weights from the last fits to the prior round of most parsimonious trees. The weights are based on the rescaled consistency index ( r c ) , which is the product of character consistency (c) and character retention index ( r ) . The value is rescaled to lie within a range of 0-1 0. T h e weighting procedure is repeated on successively produced trees until the trees no longer change (Farris, 1989). When the analysis yielded more than one tree, we constructed a strict consensus tree using the Nelsen option of Hennig86. zyxw zyxwvuts zyxwvu RESULTS Two hundred and twenty three equally parsimonious cladograms were generated from our data matrix. All have 70 steps and a consistency index of 0.57. If we exclude the autapomorphies for the genera Moscharia (characters 13, 14, 15, 19), Triptilion (characters 18, 30), and Calopappus (character 20), the length is 63 steps and the consistency index is 0.50. The strict consensus tree of the 223 cladograms (Fig. 1) shows that six monophyletic groups appear in all of them: ( 1) MOS-POL (annual, leaves partite, auricles, pseudocephalium, involucre one-seriate, flowers one to three, corollas violet); (2) all taxa except MOS and POL (cypsela hairs single two-celled, testa epidermis strengthened, pollen grains spheroidal to spheroidal oblate, colpi membrane with sexine processes, pappus bristles two to six, pappus deciduous); (3) all taxa except MOS-POL and T R I (basal leaves not in rosette, leaves crowded, pappus bristles entire or ciliate); (4) UNI-GLT-LOO (LAG-DAR-ACU) (leaves non-vaginate, hairy pappus); (5) LAG-DAR-ACU (pappus bristles plumose); ( 6 ) AXI-FUEGLS-MAE-ULI (shrubs, brachyblasts, leaves linear-subulate, leaves vaginate) . When the successive approximations weighting procedure was applied, 15 minimum length trees were obtained after the third round of weighting, with a length of 292 steps (the high value for the length is a function of the weight being scaled up to a value of lo), and the consistency improved to 0.82. The values of the range and number of steps and consistency indices (c), retention indices ( r ) , and weights (rc x 100) for each character are listed in Table 4. After the procedure, the maximum weight of 10 was assigned to characters 4,8-11, 13-20, and 23-26; a weight of 4 to characters 3, 29 and 35; a weight of 3 to character 33; a weight of 2 to characters 2, 6-7, 12, 21-22, 31 and 34; a weight of 0 to characters 1, 5, 27 and 32. The strict consensus tree of the 15 cladograms (Fig. 2) shows seven additional monophyletic groups not included in the strict consensus tree of the 223 cladograms, listed with their synapomorphies: (1) all taxa except MOS-POL, T R I and the species of the subgenus Strongyloma (pseudocephalium solitary, cypselas glabrous); (2) PYG ( ( (HIL-JUN-AME) GAU-PEN) (CAL(UNI(GLT-LOO(LAG-DAR-ACU))))) (leaves vaginate, pappus and corolla of equal long); (3) ( ( (HIL-JUN-AME)GAU-PEN(CAL(UNI(GLT-LOO (LAG-DAR-ACU) ) ) ) ) (leaves linear-subulate, capitula solitary and bristles pappus 10 or more); (4)(HIL-JUN-AME)GAU-PEN (cross section of leaf circular); ( 5 )HIL-JUN-AME (pappus bristles two to six); (6) CAL(UN1-GLO- z OUT MOS POL TRI zyxwvutsrqpon z 7 zyxwvutsrqponmlk AGY COR CUM DEN DIG DUS LAT I - MAG PIN I PUL PYR RAM REV RUI SCE SER SPR SUB CHU PVG HIL JUN AM€ GAU PEN I AGY COR CUM DEN DIG DUS LAT MAG PIN PLA PUL I PYR RAM 3zyxwvutsrqpon zyxwvutsrqponmlk MAE ULI 4 ACU Figure 1. Strict consensus cladogram from the 223 most parsimonious cladograms. Numbers 1-6 show six monophyletic groups that appear in all of the 223 cladograms. Refer to Table 1 for acronyms. Figure 2. Strict consensus cladogram from 15 most parsimonious cladograms after applying the successive weighting procedure. Numbers 1-7 show seven additional monophyletic groups not included in Fig. 1. Refer to Table 1 for acronyms. zyxwvutsrq zyxwvu zyxwvu S. E. FREIRE E'T AL. 300 TABLE 4. Character consistencies (c) and retention indices ( r ) as the best fits on the 15 equally parsimonious trees from Hennig86, used to calculate weights. Final weights were obtained after the third round of the successive weighting procedure in Hennig86. Weights were scaled to integers Character ~ Range of step Number of step Consistency index (c) Retention index ( r ) Weight (rc x 10) Final weight ~ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 I 1 1 1 1 1 1 1 1 1 I 2 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 I 2 zyxwvutsrqpo 2 2 2 1 0.50 0.50 0.50 7 0.14 0.33 0.33 1 1 3 3 1 1 1 1 6 1 I 1 I 2 1 1 1 2 2 1 1 1 1 2 1 2 1 3 3 4 3 4 1 I I 0.33 1 1 1 1 1 1 1 1 0.50 0.50 1 1 1 1 0.50 1 0.50 1 0.33 0.33 0.50 0.33 0.50 0 0.50 0.92 1 0.14 0.77 0.71 I 1 1 1 0.60 1 1 1 1 1 1 1 1 0.50 0.50 1 1 1 1 0 1 0.83 1 0.81 0 0 2 4 10 2 4 4 10 10 10 10 2 10 10 10 10 10 10 10 10 2 2 10 10 10 10 0 10 4 10 4 0 0.71 3 0.80 0.8 I 2 4 0 2 4 10 0 2 2 10 10 10 10 2 10 10 10 10 10 10 10 10 2 2 10 10 10 10 0 10 4 10 2 0 3 2 4 LOO(LAG-DAR-ACU))) (pappus persistent); ( 7 ) GLT-LOO(LAG-DARACU) (pseudocephalium) . The 15 trees obtained after successive weighting can be further simplified to ten topologies without loss of parsimony. The reason for this is that the Hennig86 regards trees as distinct if there are any different character interpretations to distinguish them and producing extra branches in some cladograms (Carr et al., 1990). In five of our 15 cladograms appear simultaneously two groups: (GAUPEN) and (HIL-JUN-AME), which cannot be supported at the same time in the same cladogram. Figure 3 illustrates one of ten equally parsimonious cladograms. Figure 4 shows the ten different configurations for the positions of N . coronipappa, N. chubutensis and HIL-JUN-AME-GAU-PEN for the middle of the cladogram. The differences in the positions of N . coronipappa and N . chubutensis (Fig. 3) result from changes in character 3 (presence or absence of leaves multicostate) and character 12 (presence of pseudocephalium). Character 3 is missing for N . coronipappa (i.e. '?'), so the presence of this character is assumed in Figure 4a, c, e, i, whereas absence is assumed in Figure 4b, d, f-h, j. We have no zy zyxwvutsr zyxwvutsrqponmlk zyxwv zy Figure 3. O n e of the ten most parsimonious cladograms found after applying the successive weighting procedure. Consistency index = 0.82. Character state changes are superimposed onto the cladograms; single lines = synapomorphies; double lines = homoplasies (parallel or convergent evolution), X = reversals. Refer to Table 1 for acronyms. 5 zyxwvutsrqpon -4&zyxwv -v z Y E PEN PEN GAU GAU HlL niL I a PEN - I I HIL JUN AYE HIL GAU JUN AYE zyxwvuts t= I 7 I Econ JUN AME - AYE E L =I -CHU I _--- Y -- PEN - - characters 3, 12 and 31. JV. coronipapfa (COR), N . chubutensis (CHU), JV. gaudichaudii (GAU), JV. pentacaenoides (PEN), N . hiffii (HIL),N.juniperina UUN), and N . ameghinoi (AME), resulting from the changes in Figure 4. Variation in the topologies of the middle of the ten most parsimonious trees found after applying the successive weighting procedure. The ten configurations show the different positions of I - E -+- zy zyxwvutsrqponml zyxwvutsrq CHU zyxwvutsrqponmlkjih - k!E JUN - E t z zyxw ...*. ,.... :.:.. $ *: ....... noc-, zyxwvu zy CLADISTICS OF NASSAUVIA ~ ~~ 303 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA After Character 12 number As coded o + 1 + 2 0 cladograrn construct ion zyxwvu 17 31 32 zyxwvuts 33 34 36 zyxwvutsrqpo Figure 5. Character transformation for selected characters as coded and after cladogram construction. 304 zyxwvut zyxwv zyxwvuts S. E. FREIRE E T A L . reason to choose between the alternative patterns found in character 3. I n character 12, the hypothesis that the pseudocephalium is lost (Fig. 4a-d, f-g) requires a developmentally more complex pattern of changes than the alternative of two parallel presences (Fig. 4e, h-j). We have chosen the last case where N . chubutensis, is isolated from the remaining species of the section Nassauvia, because it is the only species that inhabits the Patagonian steppe, whereas the other species of section Nassauvia are present in the High Andean steppe. The different positions of HIL-JUN-AME-GAU-PEN complex are the result of changes in character 31. Character 31 (number of pappus bristles) show two patterns: one reversal to ten or more pappus bristles with one reduction to two to six pappus bristles (Fig. 4c, f-g, i-j) and independent reversals to ten or more pappus bristles (Fig. 4a-b, d-e, h). We have chosen to illustrate in the reversal followed by reduction in Figure 3. The cladogram chosen (Fig. 3 ) shows that 18 characters remain as originally coded (characters 4, 8-1 1, 13-16, 18-20, 23-26, 28, and 30). Of the remaining 17 characters (Fig. 5), seven show parallelisms (characters 1-3, 5, 21, 22 and 32), five show parallelisms and reversals (characters 6, 7, 12, 33 and 34) and four show reversals (characters 27, 29, 31 and 35). DISCUSSION Our analysis suggests that Nassauvia, is a paraphyletic group. However, if Calopappus is included within the genus, Nassauvia is likely to be monophyletic. The monophyletic groups found in the cladograms of our analysis correspond to those of the current infrageneric classification of Nassauvia (Cabrera, 1982) with two exceptions: (1) section Mastigophorus is polyphyletic because N . pygmaea does not cluster with the remaining species; and (2) section Panargyrum is paraphyletic because N. lagascae ( = section Caloptilium) is not included within the section Panargyrum. The major section of the subgenus Nassauvia, section Nassauvia, is a monophyletic group diagnosed by multicostate leaves. Within section Nassauvia, there is no resolution among the species in this analysis. The other monophyletic group is subgenus Strongyloma defined by four synapomorphies: shrubs with brachyblasts, leaves linear-subulate and vaginate. zyxwvu zyxwv Placement of Calopappus In two studies of the subtribe Nassauviinae (Crisci, 1974a, 1980) Calopappus is considered as a valid genus separated from Nassauvia. According to Crisci (L.c.), Triptilion and Nassauvia form the sister group of Calopappus, and these three genera together form the sister group of Polyachyrus and Moscharia. Calopappus was recognized as a distinct genus (Crisci & Freire, 1986) based on capitula length, 17-25 mm long, pollen morphology (‘type Calopappus’, Crisci, 1974a) and three rows of involucral bracts. However, in light of the results obtained here, Nassauvia constitutes a monophyletic genus only when Calopappus is included in it, diagnosed by the presence of entire or ciliate pappus bristles, crowded and basal cauline leaves not occurring in a rosette. The distinctive appearance of zyxwvu zyx zyxwvutsrq CLADISTICS OF NASSAUVZA 305 zyxwvu zyxwvut Calopappus is due to the possession of several derived characters (autapomorphies) mentioned above. Within Nassauuia, Calopappus is related to species of subgenus Nassauuia by its glabrous cypselas. Within this large subgenus, Calopappus combines the species of section Mastigophorus (excluding N. pygmaea) and the species of section Panargyrum (including section Caloptilium). These three groups have linear-subulate leaves, solitary capitula, and ten or more pappus bristles. Calopappus was previously included in subgenus Mastigophorus (Weddell, 1855) and compared with N. pentacaenoides of section Mastigophorus by Cabrera (1982). Nassauuia pentacaenoides possesses pollen grains of intermediate size as compared with Calopappus and other species of Nassauuia. On the other hand, Calopappus appears to be related to members of section Panargyrum by its persistent pappus. Evolution o f Jowering heads Evolution of capitula in Asteraceae has been the subject of several investigations during the 1960s and 70s (Burtt, 1961, 1977; Leppik, 1960, 1970, 1977; Zohary, 1950). Often observed is a reduction in the number of flowers in the capitula and the aggregation of capitula into capitate secondary inflorescences (pseudocephalium, Troll, 1928). Approximately 70 genera (7% of the genera in the family) have pseudocephalia, and occur in nine tribes (Crisci, 1974b; Good, 1931; Stuessy, 1978). Within the Nassauviinae, Nassauuia and related genera show a trend towards reduction of the number of flowers in the capituIa and the aggregation of capitula into capitate secondary inflorescences, as pseudocephalia (Crisci, 1974a). I n Nassauuza all stages are found between a cymose conflorescence and a solitary capitulum or pseudocephalium. In our cladistic analysis (Fig. 6) the corymbose conflorescence in Trz$tilion is regarded as primitive, from which three different degrees of head condensation were derived: ( 1 ) spicate conflorescence (subgenus Strongyloma, and N. chubutensis); (2) heads arranged in pseudocephalium in cymose or racemose clusters (Moscharia and Polyachyrus); (3) solitary pseudocephalium (section Nassauuia except N . chubutensis, JV. pygmaea, N. lagascae, section Panargyrum except N . unijora and N . aculeata). Within the context of the cladogram presented, the solitary capitula ( N . uniJora, Calopappus, section Mastigophorus except N. pygmaea) are regarded as derived, originating by a reduction in the total number of the capitula from a cymose conflorescence. By the same context, the occurrence of secondary heads in Moscharia and Polyachyrus is regarded as a parallel development from the pseudocephalia found in Nassauuia. zyxwvu Pappus From the cladogram variation in the pappus can be examined as an evolutionary scenario (Fig. 6). From the condition in Macrachaenium of plumose hairy and numerous pappus bristles, there is an increase in the width of pappus bristles (Moscharia, Polyachyrus), followed by a reduction in size (in Moscharia) or number of pappus bristles, becoming deciduous (disfunctional) in the latter case. From the chaffy pappus composed of two to six pappus bristles, two evolutionary lines evolved: one (culminating in Triptilion) where the pappus bristles are plumose only in the upper part, and a second major line that begins with ciliate 306 zyxwvuts zyxwvut zyxwv zyx zy S. E. FREIRE E l AL. d Y m a zyxwvu zyxwvu zyxwvu zyxwvuts CLADISTICS OF NASSAUVZA 307 or entire pappus bristles (subgenus Strongyloma, section Nassauvia, N . pygmaea) and culminates in the plesiomorphic condition (as reversals, i.e. hairy and numerous pappus bristles) becoming again persistent (section Panargyrum including Caloptilium) . T h e paleaceous pappus occurs twice on the cladogram in Nassauvia, in N. coronipappa and N . pentacaenoides and presumably evolved from a chaffy pappus. Within Nassauvia, Cabrera (1982) considered the hairy plumose pappus as the most primitive type and the chaffy or paleaceous pappus as the result of the union of many hairs. Accordingly, he suggested section Panargyrum as the basal complex of the genus Nassauvia which has given rise to sections Mastigophorus, Caloptilium and Nassauvia. I n contrast, the results of our analysis shows the section Panargyrum (including Caloptilium) as the most derived group. Cytology Within the tribe Mutisieae only about 10% of the c. 1000 species have been examined cytologically (Crisci, 1976). Both Crisci (1976) and Cabrera (1977) reported that counts are available for 26 of the 89 genera in the tribe. I n the subtribe Nassauviinae, chromosome numbers have been published for 37 of the 300 species (12%) and ten for the other 22 genera (Crisci, 1976). On the basis of our results we can assess some cytological data within Nassauvia and related genera (Fig. 6). In this group chromosome numbers are known for Moscharia ( n = 20, Crisci, 1974a), and 1 1 species of Nassauvia ( N . darwinii n = 11, Moore in Crisci, 1974a; N . gaudichaudii n = 22, Moore, 1967; N . magellanica n = 11, Moore in Crisci, 1974a; N . serpens n = 11, Moore, 1967; N . axillaris n = 11, Hunziker et al., 1991; N. chubutensis n = 11, Hunziker et al., 1991; N. glomerulosa n = 1 1 , Hunziker et al., 1991; N . lagascae n = 11, Hunziker et al., 1991; N . revoluta n = 11, Hunziker et al., 1991; N . uniJEora n = 11, Hunziker et al., 1991; N . aculeata n = 44, Hunziker et al., 1991). The basic number in Nassauvia is x = 11; species are reported at three ploidy levels: 2x, 4x and 8x (Hunziker et al., 1991). Our results correspond well with this interpretation. If the trend of evolution within Nassauvia has been from an ancestral type such as Cephalopappus (with paniculate conflorescence), then subgenus Strongyloma (with its spicate conflorescence), would presumably be closer in morphology to the ancestral stock than any other extant group within the genus. The presence of diploid chromosome number of n = 11 in N . axillaris and N . glomerulosa supports the hypothesis that subgenus is primitive. According to its solitary capitula the tetraploid N . gaudichaudii ( n = 22) belongs to an advanced section of Nassauvia. Nassauvia aculeata ( n = 44) is placed as the most derived species on the cladogram. zyxwvu zyxwv zyxwv Geographic distribution and ecology The distribution of the taxa (Table 1, Fig. 6) shows that four genera, Moscharia, Polyachyrus, Triptilion and Calopappus, are restricted to central Chile (The ‘Chilean Province’ mentioned in Cabrera and Willink, 1973). However, Pobachyrus encroaches northern Chile and extreme southern Peru (The ‘Desert Province’ of Cabrera and Willink, 1.c.) and Triptilion is found also in the western part of Patagonia (The ‘Subantarctic Province’ of Cabrera and Willink, 1.c.). Excluding Nassauvia, they do not form a monophyletic group in any of our 308 zyxwvut zyxwvutsrqpo zyxwvu S . E. FREIRE E T A L . cladograms. These four genera present several apomorphic character states, such as the annual life cycle ( Triptilion, Moscharia, Polyachyrus) , compacted head clusters or pseudocephalia (Moscharia, Polyachyrus) or monocephalous flowering stems (Calopappus), basally branched scrub shrub (Calopappus), linear-subulate shaped leaves (Calopappus), and succulent leaves (Polyachyrus). Nassauuia is found in two biogeographic provinces (Cabrera, 1982), Patagonian and Altoandean provinces, with different ecological requirements. The Patagonian steppe species of Nassauuia (subgenus Strongyloma; N . chubutensis; section Mastigophorus excluding N . pygmaea and N . gaudichaudii; N . darwinii and N . aculeata of section Panargyrum) grow under dry conditions and most of them possess linear-subulate leaves. Within the Patagonian steppe species there are two species groups, subgenus Strongyloma which is relatively primitive because of its spicate conflorescence and pubescent cypselas, and section Mastigophorus with advanced characters, such as solitary capitula. The Altoandean steppe species of Nassauuia (N.pygmaea; N . lagascae; section Nassauvia except N . chubutensis and N . serpens; section Panargyrum except N . darwinii and N . aculeata) grow under more humid conditions. They have aggregations of capitula into secondary clusters or pseudocephalia (Fig. 6). This head condensation is probably correlated with an increase in elevation and shortening of flowering season (Stuessy et al., 1988). Only two species of Nassauuia grow in the Insular Province. According to the cladogram, one of them ( N . serpens) is related to the Altoandean species, and the other ( N . gaudichaudii) is closely related to the Patagonian species. The complexity of the biota inhabiting the Faikland Islands was pointed out by Morrone (1992), who identified them as a panbiogeographic node (sensu Croizat, 1958). zyxwvu ACKNOWLEDGEMENTS We thank Paul Berry, Angel L. Cabrera, Peter Hoch and Juan J. Morrone for critical review and discussion of the manuscript. We thank Nelly Vittet and Piero Marchionni for technical assistance and Iona Nieva and Hugo Calvetti for help with the illustrations. Finally we are grateful for the valuable review of Christopher Humphries. This research was partially supported by grant number 4662-91 from the National Geographic Society. The continued support of the Consejo Nacional de Investigaciones Cientificas y TCcnicas (CONICET), Argentina to which the authors belong is gratefully acknowledged. zyxwvut zyxwvut zyxw REFERENCES Arroyo MTK, Marticorena C. 1988. A new species of the South American genus Nassauuia (Compositae: Mutisieae) from Chilean Patagonia. B r i ~ o n i a40: 332-334. Burtt BL. 1961. Compositae and the study of functional evolution. Transactions of the Botanical Socieb of Edinburgh 39: 216-232. Burtt BL. 1977. Aspects of diversification in the capitulum. In: Heywood VH, Harborne JB, Turner, BL. eds. The biology and chemirtry o f the Compositae. London: Academic Press, 41-59. Cabrera AL. 1977. Mutisieae. Systematic review. In: Heywood VH, Harborne JB, Turner BL, eds. The biology and chemistry of the Compositae. London: Academic Press, 1039-1066. Cabrera AL. 1982. Revisi6n del gtnero Nassauuia (Compositae). Darwiniana 24(14): 283-379. Cabrera AL, WilliaL A. 1973. Biogeografia de America Latina. Programa regional de Desarrollo Cientifico y Tecnol6gico. OEA. Washington, D.C. Serie Biologia. Monografia no. 13. Carr BL, Crisci JV,Hoch PC. 1990. A cladistic analysis of the genus Gaura (Onagraceae). $sternatic Botany 15: 454-461. zyxwvu zy zyxwvutsr zyxwvutsr zyxwvut zy zyxwvuts zyxw CLADISTICS OF NASSAUVIA 309 Crisci JV. 1974a. A Numerical-Taxonomic study of the subtribe Nassauviinae (Compositae, Mutisieae). Journal Arnold Arboretum. 55: 568-610. Crisci JV. 197413. Revision of the genus Moscharia (Compositae: Mutisieaej and a reinterpretation of its inflorescence. Contribution from the Gray Herbarium of Harvard Universib 205: 163-1 73. Crisci JV. 1976. Numeros cromsomicos en 10s gineros Panphalea y Holocheilus (Compositae, Mutisieae). Hickenia l(4): 17-20. C r i s u JV. 1980. Evolution in the subtribe Nassauviinae (Compositae, Mutisieaej: A phylogenetic reconstruction. Taxon 29(2-3): 2 13-224. Crisci JV, Freire SE. 1986. El ginero Calopappus (Compositae, Mutisieaej. Caldusiu 15(71-75): 57-69. Croizat L. 1958. Panbiogeography. Caracas: published by the author. Farris JS. 1988. Hennig86. Version 1.5. Documentation. Farris JS. 1989. The retention index and the rescaled consistency index. Cladistics 5: 417419. Farris JS, Kluge AG, Eckardt MJ. 1970. A numerical approach to phylogenetic systematics. Systematic ZOO~OQ 34: 280-299. Good RD. 1931. Some evolutionary problems presented by certain members of the Compositae. Journal of Botany 69: 299-305. Grau J. 1980. Die testa der Mutisieae und ihre systematische Bedeutung. Mitteilungen Aus der Botunischen Staatssammlung Munchen 16: 269-332. Hauser DL, Presch W. 1991. The effect of ordered characters on phylogenetic reconstruction. Cladistics 7: 243-265. Humphries CJ, Funk VA. 1984. Cladistic methodology. In: Heywood VH, Moore DM, eds. Current concepts in plant taxonomy London: Academic Press, 323-362. Hunziker JH, WulE AF,Stuessy TF, Crawford DJ, Crisci JV. 1991. Chromosome numbers in the genus Nassauuia (Compositae, Mutisieae). Annals of the Missouri Botanical Garden 78: 512-531. Karis PO, Kallersjo M, Bremer K. 1992. Phylogenetic analysis of the Cichorioideae (Asteraceae), with emphasis on the Mutisieae. Annals of the Missouri Botanical Garden 79: 416427. Katinae L, Crisci JV, Freire SE. 1992. Revision sistemitica y anilisis cladistico del genera Triptilion (Asteraceae, Mutisieaej. Boletin de la Sociedad de Biologia de Concepcidn 63: 79-1 10. Leppik EE. 1960. Evolutionary differentiation of the flower head of the Compositae. Archioum Societatis <oologicae Botantcae Fernnicae ' Vanano' 14: 162-181, Leppik EE. 1970. Evolutionary differentiation of the flower head of the Compositae 11. Annales Botanici Fennici 7: 325-352. Leppik EE. 1977. The evolution of the capitulum types of the Compositae in the light of insect-flower interaction. In Heywood VH, Harborne JB, Turner BL, eds. The biology and chemisty of the Compositae. London: Academic Press, 61-89. Maddison WP,Donoghue MJ, Maddison DR. 1984. Outgroup analysis and parsimony. Systematic Zoology 33: 83-103. Moore DM, 1967. Chromosome numbers of Falkland Islands angiosperms. British Antarctic Suruey Bulletin 14: 69-82. Morrone JJ. 1992. Revisi6n sistemitica, Anilisis cladistico y Biogeografia hist6rica de 10s gineros Falklandius Enderlein y Lanteriella Gen. Nov. (Coleoptera, Curculionidae). Acta Entomoldgica Chilena 17: 157-1 74. Parra 0, Marticorena C. 1972. Granos de polen de plantas chilenas. 11. Compositae Mutisieae. Gayana (Botdnica) 21: 1-107. Ricardi My Weldt E. 1974. Revision del ginero Polyachyrus (Compositae). Gayana (Botdnica) 26 1-34. Stuessy TF. 1978. Revision of Lagascea (Compositae, Heliantheae). Fieldiana Botany 3 8 75-133. Stuessy TF, Pandey AK, Crawford DJ, Crisci JV,Hunziker JH,Arroyo MK. 1988. The evolution of inflorescences in Nassauvia (Compositae, Mutisieae). American Journal of Botany 75 (6, pt. 2): 210 (Abstract). Troll W. 1928. Organisation und Gestalt in Bereich der Bliite. Monographien aus dem Gesemtgebiet der Wissenschaftlichen Botanik 1 (I-2j: 1-413. Watrous L, Wheeler 1981. The outgroup method of character analysis. Systematic <oology 30: 1-1 1. Weddell HA. 1855. Chloris Andina 1: 45-53. Paris: Chez P. Bertrand, (Nassuuuia). Zohary M. 1950. Evolutionary trends in the fruiting head of the Compositae. Euolution 4: 103-109. a