Molecular Evidence for a Mediterranean Origin of the Macaronesian Endemic Genus Argyranthemum (Asteraceae)

by Javier Francisco-Ortega, Arnoldo Santos-Guerra, Ayelet Hines, Robert K. Jansen
Citation
Title:
Molecular Evidence for a Mediterranean Origin of the Macaronesian Endemic Genus Argyranthemum (Asteraceae)
Author:
Javier Francisco-Ortega, Arnoldo Santos-Guerra, Ayelet Hines, Robert K. Jansen
Year: 
1997
Publication: 
American Journal of Botany
Volume: 
84
Issue: 
11
Start Page: 
1595
End Page: 
1613
Publisher: 
Language: 
English
URL: 
Select license: 
Select License
DOI: 
PMID: 
ISSN: 
Abstract:

MOLECULAR

EVIDENCE FOR A MEDITERRANEAN ORIGIN OF THE MACARONESIAN

ENDEMIC GENUS ARGYRANTHEMUM

(ASTERACEAE)~

JAVIERFRANCISCO-ORTEGA,'~~

ARNOLDOSANTOS-GUERRA," AYELETHINES,?AND ROBERTK. JAN SEN'^"

2Department of Botany, University of Texas, Austin, Texas 78713-7640; and
'Jardin de Aclimatacidn de La Orotava, Calle Retama Numero 2, E-38040, Puerto de La Cruz,
Tenerife, Canary Islands, Spain

The internal transcribed spacers (ITS) of nuclear ribosomal DNA were sequenced for 52 species from 32 genera and eight subtribes of Anthemideae. Phylogenetic analyses of ITS data generated trees that are largely incongruent with the recent classification of Anthemideae; most of the subtrihes examined are not resolved as monophyletic. However, ITS trees are congruent with morphological. isozyme, phytochemical, and chloroplast DNA (cpDNA) restriction site data in supporting a Mediterranean origin for Argyranthernum, the largest endemic genus of the Atlantic oceanic islands. A combined analysis of ITS sequences and cpDNA restriction sites indicates that Argyraizthemum is sister to the other three genera of Chrysan- theminae (i.e., Chrysanthemum, Heteranthemis, and Ismelia). Times of divergence of Argyrunthemum inferred from the ITS sequences ranged between 0.26 and 2.1 million years ago (mya) and are lower than values previously reported from isozyme and cpDNA data (1.5-3.0 mya). It is likely that rate heterogeneity of the ITS sequences in the Anthemideae accounts for the low divergence-time estimates. Comparison of data for 20 species in Argymnthemunz and Chrysantheminae indicates that the cpDNA restriction site approach provided much more phylogenetic information than ITS sequences. Thus, restriction site analyses of the entire chloroplast genome remain a valuable approach for studying recently derived island plants.

Key words: Anthemideae: Chrysantheminae; ITS sequence data; Macaronesia; oceanic islands; phylogeny; plant evo-

lution.

Nearly one-third (32) of the 108 genera of Anthemi- deae (Asteraceae) are endemic to the Mediterranean and Macaronesian regions (Heywood and Humphries, 1977; Bremer and Humphries, 1993). Although many of these genera are monotypic, several contain multiple species, including Rhodantheinurn from Spain, Morocco, and Al- geria, with 12 species, and Argyranthemum from Maca- ronesia, with 24 species (Bremer and Humphries, 1993; Hansen and Sunding, 1993). The latter genus is restricted to the archipelagos of Canaries, Madeira, Desertas, and Selvagens where it provides one of the most spectacular cases of rapid speciation of Atlantic Ocean island endem-

I Manuscript received 4 November 1996; revision accepted 28 March 1997.

The senior author thanks Aguedo Marrero-Rodriguez for his friend- ship and help during germplasm collection in Gran Canaria. Our grat- itude also extends to J. Fuertes-Aguilar, Miguel A. Martin, and E. Rico for their help in tracing most of the endemic genera from southwestern Europe; to Y. Uozumi and A. Yokoyama for collecting Nipponantlzenurrn from Japan: to S.-C. Kim for advice concerning performance of relative rate tests; to Linda Watson for providing DNA samples of Cymbopappus, Glossopapprts, and Le~icanthem~tm,

and to J. Barber, T. Bark- man, D. J. Crawford, J. Fuertes-Aguilar, H.-G. Kim, s.-C. Kiln, and B.

B. Simpson for critically reading an early draft of the manuscript. This paper is dedicated to the memory of Israel Bello GonzBlez, major gar- dener at the Jardin de Aclimatacidn de La Orotava, Tenerife. Israel together with the late E. Sventenius conducted field surveys and col- lected plant materials that were the basis for the most important floristic studies of the Canary Islands of this century. This project was funded by grants from the MEC, Spain (PF92 42044506 to JFO) and the NSF(DEB 9318279 to RKJ). The ICIA, Canary Islands through the advice of M. Fernfindez-Galvfin provided a specific contract to JFO to conduct this project during a postdoctoral leave.

Author for correspondence.

ics (Howard, 1973; Hansen and Sunding, 1993; Francis- co-Ortega, Jansen, and Santos-Guerra, 1996).

Morphological, phytochemical, isozyme, and chloro- plast DNA (cpDNA) restriction site data supported a close relationship among Argyrantlzemum and three other continental genera from the Mediterranean basin: Chrysanthemum, Ismelia and Heteranthemis (Greger, 1977; Christensen, 1992; Bremer and Humphries, 1993; Fran- cisco-Ortega et al., 1995a, b). Ismelia is a monotypic ge- nus endemic to Morocco. Heteranthemis is also mono- typic with a disjunct distribution in southern Iberia and northern Morocco. Chrysanthemum has two species that are believed to have a Mediterranean origin (Heywood, 1976; Heywood and Humphries, 1977). Argyrantlzemum species are woody perennials, whereas taxa of the other three genera are annuals. In addition, Argyranthemum is the only genus of the Anthemideae examined with bis- poric embryo sacs (Harling, 1954; Borgen, 1972).

Morphological cladistic analyses (Bremer and Hum- phries, 1993) indicated that Argyraizthetnutn, Chrysanthe- mum, Heteranthemis, and Ismelia form a monophyletic group, which was circumscribed as the subtribe Chrysan- theminae. The subtribe Anthemidinae with only two gen- era (i.e.. Aathemis and Nananthea DC.) was sister to the Chrisantheminae. Furthermore, ~remer and Humphries (1993) suggested that the Achilleinae, Matricariinae, and Leucantheminae-Thaminophyllinae formed an unresolved tetrachotomy with the Chrysantherninae-Anthemidinae clade.

Taxonomic treatments prior to Bremer and Humphries (1993) considered the four genera of the Chrysanthemi- nae as a monophyletic assemblage in the "Clzrysanthemum-Leucanthemum complex" and that neither Anthemis nor Nananthea were closely related to them (Harling, 1950, 1954, 1960; Humphries, 1976a; Greger, 1977; Hey- wood and Humphries, 1977; Soreng and Cope, 1991; Vogt, 1991; Christensen, 1992). Vogt (1991) suggested Lepidophorum and Coleostephus as the closest relatives to Chrysantheminae. Humphries (1976a) regarded Zsmelia as the continental genus most closely related to Argyranthemum. Others suggested that Argyranthemum does not have a Mediterranean origin and that its closest relatives might be South ~frican "Chrysanthemums" such as Cymbopappus (Hutchinson, 1917; Bramwell, 1976; Takhtajan, 1986) or the Japanese genus Nipponanthemum (Kitamura, 1978).

Argyranthemum has been referred to as an active epi- biotic, i.e., a taxon derived from a relictual continental ancestor that became extinct on the mainland following major climatic changes in the late Tertiary (Humphries, 1976b). It has been postulated that in Macaronesia active epibiotics have gone through a secondary evolutionary episode from relictual groups (Bramwell, 1972a, 1976; Cronk, 1992). Other genera such as Aeonium Webb & Berth. (Crassulaceae), Echium L. (Boraginaceae), Euphorbia L. (Euphorbiaceae), and Sonchus L. (Asteraceae) have also radiated extensively in Macaronesia and may represent relicts of some present Mediterranean taxa (Bramwell, 1972a, b, 1976; Aldridge, 1979).

The woody habit of Argyranthemum and the high lev- els of isozvme diversitv between the senus and its closest relatives kpport the hypothesis thatlit may be derived from a relictual group that became restricted to the Ma- caronesian islands in the late Tertiary (Bramwell, 1972a, 1976; Aldridge, 1979; Sunding, 1979; Francisco-Ortega et al., 1995a. b). In contrast, the woody habit of most Macaronesian endemics has been suggested to be a de- rived feature that represents an adaptive response to the seasonal uniformity of the climate of the insular environ- ment (Carlquist, 1962, 1965, 1974).

In this paper we examine the origin of Argyranthemum within the framework of the evolutionary history of the tribe Anthemideae. All putative relatives of Argyranthemum were examined. We also sampled 26 genera from seven of the 12 subtribes of Anthemideae (sensu Bremer and Humphries, 1993). The five objectives were to: (1) examine the origin of Argyranthemum in relation to the major speciation events in Anthemideae in the Mediter- ranean, South African, and Far East regions; (2) test the monophyly of Chry santheminae (i.e., Argy raizthemum, Chrysanthemum, Heteranthemis, and Ismelia); (3) determine whether molecular and morphological phylogenies are congruent, especially with regard to the sister group relationship of the subtribes Anthernidinae and Chrysan- theminae; (4) evaluate whether Argyranthemum represents a relictual member of the Macaronesian flora; and

(5) compare the phylogenetic utility of ITS sequences with that of cpDNA restriction site data.

These objectives were accomplished by performing phylogenetic analyses of nucleotide sequences of the in- ternal transcribed spacers (ITS) of nuclear ribosomal DNA (nrDNA). Sequences of these regions have proven utility in tracing the origin and evolution of island en- demics in other plant groups, such as Dendroseris D. Don (Asteraceae) (Sang et al., 1994) and Robinsonia DC. (As- teraceae) (Sang et al., 1995) in the Juan Fernandez Is- lands; Crambe L. sect. Dendrocrambe DC. (Brassicaceae) (Francisco-Ortega et al., 1996a), Lavatera L. (Malvaceae) (Ray, 1995), Aeonium (Mes, 1995), and Sonchus subgen. Dendrosonchus Sch. Bip. in Macaronesia (Kim et al., 1996; Kim, Crawford, and Jansen, 1997), and the silversword alliance (Baldwin, 1992; Baldwin and Rob- ichaux, 1995) in the Hawaiian islands.

MATERlALS AND METHODS

Sampling-Sequences of the ITS region were determined for 52 taxa from 32 genera of Anthemideae (Table 1). This includes all five genera that have been regarded as putative relatives of Argyranrhemum (i.e., Cymbopappus, Chrysanthemum, Heterarzthemis, Ismelia, and Nipponanthemum), and 16 Argyranrhemum taxa from seven islands and all sections of the genus. In addition, the ingroup includes the following 13 genera of the "Leucanthemum-Chrysanthemum" complex (as de- fined by Humphries, 1976a; Heywood and Humphries, 1977; and Vogt, 1991): Coleosrephus, Dendranthema, Glossopappus, Hymenostemma, Lepidophorum, Leucanthemum, Leucanthemella, Leucanthemopsis, Mauranthem~im, Phalacrocarpum, Prolongoa, Rhodanthemum, and Tanacerum. Furthermore, 12 representative genera of the subtribes Achilleinae (i.e., Achilles, Anacyclus, Cladanthus, and Sanrolina), Anthemidinae (Anrhemis), Artemisiinae (i.e., Ajania, Arctanrhem~im, and Artemisia),and Matricariinae (i.e., Lonas, Matricaria. Tripleurospermum, and Otospermum) were considered as part of the ingroup. Ursinia (Ursiniinae), which has been suggested as one of the basal lineages in Anthemideae (Bremer and Humphries, 1993; Kim and Jansen, 1995; Watson, 1996), was selected as the outgroup.

DNA isolation, PCR amplz~cation, and sequencing-Most of the DNA isolations were from leaves collected from plants cultivated in the greenhouses at the University of Texas at Austin. Sources of seeds for the taxa examined are in Table 1. The CTAB technique of Doyle and Doyle (1987) was used for DNA isolations from living material. DNA extractions from herbarium specimens were performed according to a modification of the CTAB technique developed by Loockerman and Jansen (1996).

Primers ITS4 and ITS5 of White et al. (1990) were used for double- stranded DNA amplification of the ITS region by polymerase chain reaction (PCR). The amplification strategy followed the protocol in Kiln and Jansen (1994). The amplified product was purified with glass milk after electrophoresis through 1.0% agarose gel in 1X TAE buffer.

Amplification of the ITS region of Tanacetum vulgare was conducted by nested PCR (Albert and Fenyo, 1990). The sequences of the forward and reverse PCR primers for the initial amplification were 5'-GTAAGCGCGAGTCATCAGCTCG-3' and 5'-CATCTTTCCCTCGCGGT- ACTTG-3', respectively. Following this first amplification a second am- plification was performed in a 50 FL reaction using primers ITS4 and ITS5 and 1.0 pL of the nested PCR product.

Direct sequencing of double-stranded DNA was accomplished using Sequenase Version 2.0 (United States Biochemical Corp., Cleveland, OH) and the snap-chill method (Winship, 1989). Two forward (ITS3 and ITS5) and two reverse (ITS2 and ITS4) primers (White et al., 1990) were utilized to sequence both strands. Sequencing reactions, acryla- mide gel electrophoresis, and autoradiography were performed as de- scribed in Kim and Jansen (1994).

Sequence analysis-Published sequences of nrDNA of other genera of Compositae (Baldwin, 1992; Kim and Jansen, 1994; Bain and Jansen, 1995) were used to determine the boundaries of the ITS1 and ITS2 regions. DNA sequences were aligned by means of multiple alignment using the CLUSTAL W computer software package with the penalty parameters of 10.00 for opening of gaps and of 5.00 for extension of gaps (Higgings, Bleasby, and Fuchs, 1991). Only minor corrections

TABLE 1. List of taxa and source of plant material. Taxonomy follows TABLE 1. Continued Bremer and Humphries (1993), and Vogt and Oberprieler (1995). Institute codes follow Holmgren, Holmgren, and Barnett (1990)

Inst~tute and

with the exceptions of: BPE: Bluestone Perennials, Madison, OH,

Taxon acceiilon number.

USA; HUD: J. L. Hudson Seedsman, Redwood City, CA, USA; MAI: Germplasm Seed Bank at Escuela TCcnica Superior de In- A. webbii Sch. Bip. ORT-2 1 genieros Agr6nomos, Madrid, Spain; MAR: Seed Collection of

A. winreri (Svent.) Humphries ORT-191 Chrysanthemum L. (2) Marseille Botanic Garden, Marseille, France; NCP: Germplasm

C. coronarium L. MAR- 1449 Seed Bank at North Central Regional Plant Introduction Station

C. segerum L. BORD-00393(USDA), Ames, IA, USA; TFN: Town Farm Nursery, Cleveland,

Hereranthemis Schott (1)

England, Great Britain; THM: Thompson and Morgan Inc., Jack- H. viscidehirta Schott MAI-667584

son, NJ, USA. Accession numbers refer to either germplasm seed Zsmelia Cass. (1)

bank, seed collection, or voucher number in botanical institute. I. carinata (Schousb.) Sch. Bip. GAT-38825
Number of species in each genus is indicated in parentheses.

Leucantheminae Coleosrephus Cass. (3)

GenBank Institute and accesslon C. myconis (L.)Rchb. f. ORT-s.n. Taxon accesslon number number

Glossopappus Kunze (1)

G. macrotus (Durie) Briq. MU, L. Watson,

Achilleinae

95-26CAnacyclus L. (12)

Hymenosremmu (Kunze) Willk. (1)

A. pyrethrum (L.) Lag. var. depres

H. pseudoanthemis (Kunze) Willk. MAI-527478 sus (Ball) Maire

Lepidophorum Cass. (1)

Achillea L. (1 15)

L. repandum (L.) DC. SALA, E. Rico,

A. millefolium L. TEX, Kim 9 1 - June-I -1984 207 Leucanthemella Tzvelev (2)
Cladanthus Cass. (1)

L. serotina (L.)Tzvelev NCP-

C. arabicus (L.) Cass.

PI50226 1

Sanrolina L. (8)

Leucanrhemopsis (Giroux) Heywood (9)

S. chamaec~parissusL

L. alpina (L.) Heywood UPS, Watson

s.n.

Anthemidinae

Leucanthemlrm Mill. (33)

Anthemis L. (21 1) L. maximum (Ramond) DC. UPS, Warson

A. arvensis L. NCY-466 s.n.

A. crerica L. NCY-1148 Mauranthemum Vogt & Oberprieler (3)

A. maririma L. BORD- 10192 M. paludosum (Poir.) Vogt & Ober-SALA, E. Rico,

A. rincroria L. TO-s.n.

prieler April-16-1993 Artemisiinae

Nippananthemum Kitam. (1) N.nipponicum (Maxim.) Kitam. TEX, J. Yokoy-

Ajania Polj. (34) ama

A. pacijica (Nakai) K. Bremer & NCP-

951 00701

Humphries PI479350 Phalacrocarpum (DC.) Willk. (2) Arctanthemum (Tzvelev) Tzvelev (4)

P. hoffmannseggii (Brot.) Lainz SALA, Casase-

A. arcticum (L.) Tzvelev NCP-

cu, April-1 l

PI50226

1977

Artemisia L. (388)

Prolongoa Boiss. (1)

A. canariensis (Besser) Less ORT, Santos er

P. hispanica G. L6pez & C. E. Jar- SALA, Rico, er

al., July-10-

vis al., 8-Julio- 1995 1985

Dendranthema (DC.) Des Moul. (37) Rhodanthemum R. H. Wilcox, K. Bre-
D. coreanum (H. LCv. & Vaniot) mer & Humphries (12)
Vorosch.

R. gayanum (Cosson & Durieu) R. TFN-s.n.

H. Wilcox, K. Bremer & Hum-Chrysantheminae phriesArgyranthemum Sch. Bip. (24) Matricariinae

A. adauctum (Link) Humphries ORT- 1
subsp, adauctum Cj~mbopappusB. Nord. (4)

A. adauctum (Link) Humphries ORT-78 C. adenosolen (Harv.) B. Nord MU, L. Watson

subsp. canariense (Sch. Bip.) 94-79. Humphries Lonas Adans. (1)

A. brozrssonerii (Pers.) Humphries ORT-84 L. annua (L.) Vines & Druce THM-6478 subsp. gomerensis Humphries Matricaria L. (7)

A. coronopifolium (Willd.) Humphries ORT-6 M. recutita (L.) Rauschert HUD-MATI2

A. dissectum (Lowe) Lowe ORT-M2 Otospermum Willk. (1)

A. jlifolium (Sch. Bip.) Humphries ORT-77 0. glabrum (Lag.) Willk. SALA, Casase-

A. foeniculaceum (Willd.) Sch. Bip. ORT-9 ca et al.,

A. frutescens (L.) Sch. Bip. subsp. ORT- 10 1 April-10-1980

frutescens Tripleurospermum Sch. Bip (38)

A. frurescens (L.) Sch. Bip. subsp. ORT-69 T. peforatum (Merat) Lainz THM-6 13 1 succulentum Humphries

"Tanacetinae"

A. haouarytheum Humphries & ORT- 184

Tanacetum L. (152)

Bramwell

T. microphyllum DC. MA-s.n.

A. pinnar$dum (L. f.) Lowe subsp. ORT-M8

T. vulgare L. TEX, Bohusldpinnati'dum vek 989

A. pinnari'dum (L. f.) Lowe subsp. ORT-M7
montanum Rustan Ursiniinae

A. tenerifae Humphries ORT- 19 Ursinia Gaertn. (1 14)

A. thalassophilum (Svent.) Humphries ORT-M9 U. anthemoides (L.) Poir. THM-73 19

GenBank accessron number

L77797

L77795 L77741 L77742 L77761 L77764

L77758

L77760 L77762 L77765 L77766 L77767 L77768 L7777 1 L77772 L77775

L77776

L77777

L77759

L77769 L77770 L77774

L77782

L77780 L77781

L77783 were necessary to adjust the output alignment of CLUSTAL W. The software package MEGA (Kumar, Koichiro, and Nei, 1993) was used to calculate nucleotide sequence divergence based on Kimura's two-parameter model (Kimura, 1980) using both complete and pairwise de- letion of gaps and missing data. The G + C content of both ITSl and ITS2 was also obtained using MEGA. MacClade (version 3; Maddison and Maddison, 1992) was used to calculate transitionltransversion ratios from the most parsimonious trees.

Phylogenetic analysis-Fitch parsimony was performed using PAUP version 3.1.1. (Swofford, 1993) with ACCTRAN, MULPARS, and TBR options. Multiple islands of equally parsimonious trees (Maddison, 1991) were searched for by performing 100 random entries in all heu- ristic searches. Gaps in aligned sequences were coded in three ways:

(1) all gaps were treated as missing data, (2) gaps were scored as miss- ing and each gap was coded as a binary character (present or absent), and (3) gaps were eliminated from the matrix.

Character-state weighted parsimony was conducted with all three codings of the gaps. Transversions were weighted over transitions by a

1.5:l ratio by means of the USERTYPE STEPMATRIX command of PAUP A weight of 1.5 was selected because this was the transition1 transversion ratio obtained from MacClade using the most parsimonious tree obtained in unweighted initial analysis.

Support for monophyletic groups was evaluated using 100 bootstrap replicates (Felsenstein, 1985). These analyses were conducted using PAUP with the ACCTRAN, MULPARS, and TBR options and a heu- ristic search with simple addition sequence of taxa. For weighted par- simony and analyses in which indels were coded as presentlabsent lim- itations of computer memory necessitated that the number of random entries of taxa was increased to 500, and the maximum number of trees saved constrained to five in each replicate.

The g, statistic (Hillis and Huelsenbeck, 1992) was calculated by computing the tree-length distribution of 100 000 random parsimony trees using the RANDOM TREES command of PAUP Consistency In- dex (CI) (Kluge and Farris, 1969) and Retention Index (RI) (Farris, 1989) were also computed.

Parsimony analyses (branch and bound search) were also conducted on a combined data matrix of ITS sequences and previously published cpDNA restriction site data (Francisco-Ortega, Jansen, and Santos-Guerra, 1996). For these analyses gaps were treated as missing data. The ingroup comprised seven taxa (i.e., Argyranthemum foeniculaceum, Chrysanthemum coronarium, Chrysanthemum segetum, Heteranthemis viscidehirta, Ismelia carinata, Leucanthemum maximum, and Santolina chamaecyparissus). Four Anthemis species were included in the out- group (A. awensis, A. cretica, A. tinctoria, and A. maritima). Anthemis was selected as the outgroup based on phylogenetic analyses of ITS sequences of Anthemideae (see below, Figs. 1-2).

Relative rate tests (Wu and Li, 1985; Li and Tanimura, 1987) using divergence values from the Kimura two-parameter method were used to test for a molecular clock within Anthemideae. Six taxa (i.e., Ajania pacijca, Anthemzs cretica, Argyranthemum foeniculaceum, Prolongoa hispanica, Rhodanthemum gayanum, and Tripleurospermum perj%ra- tum) from the major lineages identified in the ITS tree were used. Rate tests were also conducted using Argyranthemum, Chrysanthemum, Het- eranthemis, and Ismelia. Ursinia was used as the reference taxon for all relative rate tests. Standard errors of estimated relative rate differ- ences between lineages were calculated following Li and Tanimura (1987) and Kimura (1980), and were utilized for significance tests.

RESULTS

Length variation and base composition-Aligned ITS sequences are 503 bp long (268 bp for ITSl and 235 for ITS2, Table 2, Appendix). The longest sequence (479 bp) is in the outgroup Ursinia anthemoides; the shortest (440 bp) is in Prolongoa hispanica. The number of indels after alignment is 70, and most are 1 and 2 bp long. However, Prolongoa hispanica has two deletions of 11 and 5 bp in ITSl and ITS2, respectively. In addition, an indel of 17- 19 bp is present in ITS2, between positions 293 and 3 11 (Appendix). This indel marks a major division in An- themideae; all taxa restricted to South Africa (i.e., Cym- bopappus and Ursinia) and the Far East (Ajania, Arctan- themum, Dendranthema, and Nipponanthemum) have an insertion in this region relative to the rest of the taxa examined. This insertion is also present in the eastern European Leucathemella serotina and in the only repre- sentative of Artemisia. The G + C content in Anthemi- deae ranges from 46.7 to 55% with a mean of 49.6%.

Sequence divergence-A total of 283 variable sites was detected in ITSl and ITS2. There were 168 and 220 phylogenetically informative and unique changes, respec- tively. Sequence divergence among genera (Table 3) ranges between 0.2% (Argyranthemum and C. coronar- ium, and Ismelia and C. coronarium) and 27% (Ursinia and Prolongoa). The five taxa with the lowest divergence in comparisons with Argyranthemum were: Chrysanthe- mum coronarium (0.2%), Ismelia (0.5%), C. segetum (1.6%), Santolina (3.3%), and Heteranthemis (5.3%). The five taxa with the highest divergence with Argyranthe- mum were Ursinia (16.5%), Hymenostemma (12.7%), Prolongoa (1 2.7%), Leucanthemopsis (1 1.8 %), and Lon- as (10.8%).

The highest nucleotide divergence value within Argyr- anthemum is 1.1 1% between A. thalassophilum and A. haouarytheum. Only seven variable sites were detected within Argyranthemum, three of which are phylogeneti- cally informative (Table 4).

Phylogenetic analysis of ITS sequences-Fifteen of the 16 taxa of Argyranthemum were excluded from the phylogenetic analysis because only three informative changes were detected in the genus. Unweighted parsi- mony analysis of the remaining 37 taxa generated a sin- gle tree (744 steps) when indels were treated as missing data (Fig. 1). The tree has a CI of 0.466 (excluding au- tapomorphies) and a RI of 0.632. The g, statistic for 100 000 random trees is -0.746, indicating that the ITS data are skewed significantly from random (P < 0.01 for g, = -0.09 for 250 variable sites and 125 taxa). Thus, there is a substantial amount of phylogenetic signal in the ITS sequences (Hillis and Huelsenbeck, 1992).

A monophyletic group including Ismelia carinata and Chrysanthemum coronarium is sister to Argyranthemum. The monophyly of these three taxa is supported by a low bootstrap value of 63% and this group is sister to Chry- santhemum segetum. The Argyranthemum-Chrysanthemum-Zsmelia clade has only 68% bootstrap support and a relatively derived position in Anthemideae. This lineage of three genera is sister to a weakly supported group (29% bootstrap value) that includes Heteranthemis and five other genera of the subtribes Leucantheminae (Hy- menostemma, Leucanthemopsis, Prolongoa, Lepidopho- rum) and Matricariinae (Lonas).

None of the subtribes examined are monophyletic in the ITS tree except Anthemidinae. However, most Med- iterranean genera- (except Phalacrocarpum and Anacy- clus) form a weakly supported clade (33% bootstrap val-

1 1

Argyranthernum foeniculaceum

Chrysanthemum coronarium hrysantheminae

Chrysanthemum segeturn

28 20

Heteranthemisviscidehirta

7 12

Hymenostemma pseudoan fhemts Leucanthemapsis alpina

Leucantheminae

2

5 Lepidophorum repandum

I

Lonm annua = Matricariinae Cludanthus arabicus Sun tolina chamuecyparissus

a

9 Rhodanthemum gayanum Leucanthemum maximum Glossopappus macrotus

Leucanthetttinae

16

Muuranthenrumpaludosum

7

Coleostephus nzyconis

i

l-e 8 13 .-• Otospermum glabrum 4 16

Mutricaria recutita IMatricariinae

TripleurospermumperforatumAnacyclus pyrethrum

28 11 Achill einae

-Achillea millefolium

Deiehon

28 Phalacrocurpum hofinnseggii =Leucantheminae

1

\ = Tanacetinae

2 I

77 14

Anthemidinae 21

4 13

Anthemis tinctorin

12 7

Tanacetum microphyllum ITanacetinae 13

D C~mbopap~us~Matricariinae

udenosolen Leucanfhemellaserotrna Nrpponanthemum nlpponlcum

I

Dendranthema coreanum

Artemisiinae

Arctunthemum arctrcum

12

Artemrsra canartensrs

I

37

I) Ursrn~aanthernordes Ursiniinae

Fig. 1. Single most parsimonious Wagner tree (744 steps: CI = 0.466, without autapomorphies; RI = 0.632) from the ITS nucleotide sequence data of the 37 taxa included in the present study. The number of nucleotide changes is indicated along each branch. Bootstrap values are in large boldface. Distributions of genera are coded as follows: closed circle = Mediterranean, open circle = predominantly Mediterranean distribution, open square = predominant distribution in Europe and East Asia, gray square = Eastern Asian distribution, triangles = South African. Branches

in boldface are given for those terminal taxa that were examined to test the molecular clock hypothesis within the Anthemideae. Anow indicates clade that has an -17-bp deletion in ITS2.

ue). Only nine of the 16 nodes found within this clade tained in these four additional analyses ranged between have bootstrap values >50%. Similar results were also three (indels coded and weighted parsimony) and 168 obtained for the other four cladistic analyses involving (indels removed and weighted parsimony). Trees from

weighted and unweighted parsimony with indels coded three of these analyses support the monophyly of Argyras binary characters or deleted (Fig. 2A-D). The strict artthemum, Chrysarzthemum, and Isrrzelia (Fig. 2B-D). consensus tree using weighted parsimony and coding in-All analyses are concordant in showing a weakly supdels as binary states (Fig. 2B) is identical to the weighted ported monophyletic group of Mediterranean genera, in-parsimony tree that treated indels as missing data (not cluding Argyranthem~trnand the three other genera of shown). The number of equally parsimonious trees ob-Chrysantheminae.

Argyranthemum foeniculaceum Chrysantemum coronarium Ismelia carinata Chrysanthemum segerum Heteranthemis visadehirta e--------.I

lfymenostemma pseudoan themis

Leucan themopsis alpina Prolon aoa his~anica

I urn~e~idophirume-1iependum Lonas annua 2n

Cladanthus arabicus 85 Santolina chamaecyparissus Rhodan themum gayanum Leucan themum maximum

:-ZIl-1

Glossopappus macrotus Mauran themum pal udosum

87

Coleostephus myconis Otospermum glabrum Phalacrocarpum hoffmannseggii P81,

Ma micaria recutita Tanacetum vulgare An themis cretica

An themis maritima An themis arvensis An themis tinctoria

Anacyclus pyrethrum Tripleurospermum perforatum ----'

Achillea millefolium Tanacetum microphyllum Cymbopappus adenosolen Leucanthemella serotina Lkndran thema coreanum Ajania pacifica Arctan themum arcticum Artemisia canariensis Nipponathemum nipponicum

Ursinia anthemoides

Argyranthemum foeniculaceum Chrysantemum coronarium Ismelia carinata

Chrysanthemum segetum Heteranthemis viscidehirta Hymenstemma pseudoanthemis Leucanthemopsis alpina Prolongoa hispanica Lepidophorum rependum Lonas annua Cladanthus arabicus

San tolina chamaecyparjssus Rhodan themum gayanum Coleostephus myconis

~Mauranthemum paludosum Glossopappus macrotus Leucan themum maximum Otospennum glabrum

Matricaria recutita 82 Phalacrocarpum hoffmannseggii Tanacetum vulpare 100 An themis cretca 100 An themis maritima

I

I I An themis arvensis An themis tincton'a Anacyclus pyrethrum Achillea millefolium Tripleurospermum perforatum

Tanacetum microphyllum Leucan themella serotina Lkndran thema coreanum

Ajania pacifica Arctan themum arcticum h'ipponathemum nipponicum

Artemisia canariensis Cymtopappus adenosolen Ursinia anthemoides

Fig. 2, Four strict consensus trees from the ITS nucleotide sequence data obtained after unweighted and weighted parsimony analyses and after considering indels as missing data or coded as binary states. Transversions were weighted over transitions by a 1.5:1 factor. Bootstrap values >50% are indicated along nodes. Closed circle = genera endemic to Medite~~aneanor with a predominant Mediterranean distribution. (A) Unweighted parsimony analysis with indels coded as binary characters (867 steps; CI = 0.454, without autapomorphies; RI = 0.624, number of most parsimonious trees: 55); (B) Weighted parsimony analysis with indels coded as binary characters (8792 steps; CI = 0.389, without autapomorphies; RI = 0.556; number of most parsi~nonioustrees: 3); (C) Unweighted parsimony analysis with indels removed from (589 steps; CI = 0.461, without autapomorphies; RI = 0.640; number of most parsi~nonioustrees: 90); (D) Weighted parsimony analysis with indels removed (6825 steps; CI = 0.350, without autapomorphies; RI = 0.640; number of most parsimonious trees: 168).

.

gaps as missing data. g, value calculated for 100 000 random trees.

ITS 1

Length range (bp)
241-255
Length mean (bp)
236.9 Aligned length (bp)
268
G + C content range (%)
43.4-54.4
G + C content mean (70)
48.5

Sequence divergence (%) Conlplete deletion of gaps and missing data 0.0-22.1 Pairwise deletion of gaps and nlissing data 0.0-24.91

Number of indels 1 bp 18

2 bp
9

3 bp
4

4 'JP
1

5 bp
0

11 bp
1

17 bp
0

18 bp
0

19 bp
0

*
Number of variable sites 148
*
Number of informative changes 92
*
Number of constant sites 120
*
Number of autapomorphic sites 56
*
Transitions (average) 257
*
Transversions (average) 142

* Transitions/transversions 1.80

* Skewness of tree-length distribution (g,) -0.670

Phylogenetic analyses of combined ITS and cpDNA data-Chloroplast DNA restriction site and ITS sequence data are available for 11 taxa. Phylogenetic analyses of the combined data generated two most parsimonious trees of 412 steps (Fig. 3A) with a RI of 0.831, CI of 0.684 (excluding autapomorphies), and g, value (100 000 ran- dom trees) of -0.555. Bootstrap values for all nodes are higher than 89%. The combined tree is congruent with the single most parsimonious tree obtained from the cpDNA restriction site data alone (Fig. 3B) in showing Argyranthemum as a distinct lineage sister to the rest of Chrysantheminae. Phylogenetic analyses of ITS data from the 11 taxa produced six equally parsimonious trees. The strict consensus tree is virtually unresolved and thus does not identify which of the genera of the Chrysan- theminae was sister to Argyranthemum (Fig. 3C).

Phylogenetic utility of ITS sequences and cpDNA re- striction site data-The phylogenetic utility of ITS sequence and cpDNA restriction site data was compared for the 20 taxa of Chrysantheminae (Table 4). Both data sets include 16 species of Argymnthemum and the four outgroup species from Ismelia, Chrysanthemum, and Heteranthemis. A number of tree statistics (Table 4) were compared in parsimony analyses of the two data sets (trees not shown). The results clearly indicate that the cpDNA restriction site approach identified many more variable and informative changes than ITS sequencing. Thus, phylogenetic analyses of the cpDNA data gener- ated many fewer trees (12 vs. 598) with many more re- solved nodes (12 vs. 2). The CIS and RIs were similar for both data sets, but the g, value was three times higher for the cpDNA data. The higher number of variable po-

ITS2 ITSl & ITS2

196-224 440-479

207.7 458.3

235 503
45.6-55.8 46.7-55.0

5 1 .0 49.6

0.0-24.8 0.3-22.5 0.0-25.63 0.4-24.20

23 41
7 16
1 5

2 3
1 1
0 1
1 1
1 1
1 1

135 283
76 168
100 220
59 115
194 450
154 294

1.28 1.53 -0.681 -0.746

sitions can be explained mainly by the much larger num-
ber of base pairs examined by the cpDNA restriction site
approach (14 716 vs. 503). The ITS sequences are more
divergent in most comparisons than the chloroplast se-
quences (Table 5).

Relative rate tests-Eight of the 15 pairwise compar- isons of ITS sequences indicate significantly different rates of sequence divergence (Table 6). Therefore, a mo- lecular clock can be rejected in almost half of the com- parisons in Anthemideae. Most instances in which the clock is rejected involve Argyranthemum and Prolongoa. All relative rate differences between Argyranthemum and the five other genera are significant at the 1% level, ex- cept the comparison made with Ajania paci'ca. In con- trast, the only pairwise comparison that is not significant with Prolongoa involved Anthemis cretica.

Pairwise comparisons that involved Argyranthemum, Chrysanthemum, Heteranthemis, and Ismelia indicate significant rate differences at the 1% level between Heteranthemis and the three other genera (data not shown). The highest value is between Heteranthemis and Chrysanthemum coronarium (standard error of 5.999, a rela- tive rate of 0.059). Therefore, a molecular clock within Chrysantheminae was rejected whenever Hetera~zthemis was included in the analysis.

DISCUSSION

The fact that phylogenetic analyses of ITS sequences generated a single most parsimonious tree (Fig. 1) with reasonably high CI (0.466), RI (0.632), and g, (-0.746) values indicates that the data contain substantial phylo-

TARI.E3. Pairwise divergence estimates of the 37 taxa included in the phylogenetic analysis (Fig. 1). Nucleotide divergence values based on Kimura's two-parameter model are above diagonal. Numbers of nucleotide substitutions are shown below diagonal. Divergences were calculated after complete deletion of gaps and missing data. Taxa are coded in the Appendix.

ARFO llEVI ISCA CHCO CHSF CL4R PHHO TAVIJ TAMI

ARFO HEVI 19 ISCA 2 CHCO 1 CHSE 6 CLAR 19 PHHO 25 TAVU 21 TAMI 21 ANCR 36 ANAR 25 ANTI 31 ANMA 36 ROGA 26 COMY 23 GLMA 30 LEMA 23 MAPA 25 SACH 12 ANPY 31 ACMI 25 HYPS 43 LERE 35 PRHI 43 LEAL 40 TRPE 32 MARE 22 OTGL 22 LOAN 39 CYAD 33 LESE 26 DECO 32 AJPA 32 ARAR 31 NINI 25 ARCA 35 URAN 54

genetic signal. The ITS tree of Anthemideae provides several insights into the systematics and evolution of the tribe. However, low bootstrap values for many of the nodes (Fig. 1) and rate heterogeneity (Table 6) suggest that we should be cautious in using the ITS tree alone to make strong conclusions about the origin and evolution of Argyranthemum. Fortunately, our discussions of the systematic and evolutionary implications of the ITS data can incorporate the results from previous morphological (Bremer and Humphries, 19931, isozyme (Francisco-Or- tega et al., 1995a, 1996b), and cpDNA restriction site (Francisco-Ortega et al., 1995b, Francisco-Ortega, Jan- sen, and Santos-Guerra, 1996) studies. We believe that these four data sets together provide a strong basis for clarifying the origin and systematic relationships of the largest genus endemic to the Atlantic oceanic islands.

Systematic relationslzips in Anthemideae-The ITS phylogeny (Figs. 1-2) suggests that the Mediterranean taxa Chrysanthemum coronarium and Ismelia carinata are the closest continental relatives to Argyranthemunz. In addition, the ITS tree is congruent with previous cpDNA restriction site and isozyine data (Francisco-Ortega et al.,

ANCR ANAR ANTI ANMA ROGA COMY GLMA IFMA MAPA

1995a, b) in supporting Ismelia and C. coronarium as

sister taxa.

Several taxonomic treatments considered Argyranthemum congeneric with Chrysanthemum (e.g., Harling, 1954; Sventenius, 1960; Jeffrey, 197 1 ; Borgen, 1972). The ITS tree (Fig. 1) provides weak support of this view. In contrast, cpDNA restriction site data clearly support Argyranthernum as a distinct, monophyletic genus that is sister to Heteranthemis, Ismelia, and Chrysanthemum (Francisco-Ortega et al., 199%). Phylogenetic analyses of combined ITS and cpDNA data (Fig. 3A) agree with the restriction site data in supporting Argjranthenzum as a distinct lineage. Thus, we recommend that Argymnthemum continues to be recognized as a separate genus with- in the subtribe Chrysantherninae.

The ITS tree also suggests a close relationship between Chrysantheminae and five other Mediterranean genera (i.e., Hymenostemma, Lepidophorum, Le~tcanthemopsis, Lonas, and Prolongoa). They form a weakly supported monophyletic group with the core genera of Chrysan- theminae. These genera are sister to Heteranthemis in most trees (Figs. 1-2) except the unweighted parsimony analysis with indels coded as binary states. The sister- group relationship of these five Mediterranean genera to Chrysantheminae contradicts the Bremer and Humphries (1993) hypothesis that Anthemidinae and Chrysanthem- inae are sister subtribes.

SACII ANPY ACZII HYPS LERE PRITI LELA TRPE MARE OTGL LOAN CYAD LESE DECO AJPA ARAR NlNI ARC.4 IJRAN
  -    
  0.090 0.071 0.127 0.102   0.093 0.062 0.062 0.108 0.096 0.074 0.093 0.093 0.090 0.071 0.102 0.165
  0.140 0.117 0.157 0.130   0.140 0.111 0.114 0.133 0.137 0.127 0.134 0.134 0.130 0.117 0.150 0.221
  0.096 0.077 0.131 0.108   0.099 0.068 0.065 0.1 11 0.102 0.080 0.099 0.099 0.096 0.077 0.105 0.162
  0.093 0.074 0.127 0.105   0.096 0.065 0.062 0.108 0.099 0.077 0.096 0.096 0.093 0.074 0.102 0.166
  0.096 0.074 0.141 0.114   0.093 0.065 0.059 0.11 1 0.105 0.083 0.102 0.102 0.099 0.080 0.1 11 0.172
  0.099 0 086 0.154 0.140   0.102 0.080 0.074 0.140 0.1 11 0.096 0.121 0.121 0.118 0.099 0.121 0.196
  0.096 0.074 0.154 0.161   0.093 0.093 0.102 0.141 0.109 0.109 0.142 0.142 0.138 0.109 0.125 0.194
  0.059 0.042 0.140 0.137   0.050 0.045 0.080 0.131 0.083 0.081 0.106 0.105 0.102 0.084 0.105 0.187
  0.065 0.045 0.137 0.127   0.069 0.056 0.083 0.118 0.068 0.078 0.096 0.096 0.093 0.081 0.089 0.172
  0.115 0.092 0.199 0.170   0.090 0.080 0.130 0.174 0.137 0.121 0 151 0.151 0.147 0.128 0.154 0.231
  0.077 0.044 0.143 0.150   0.07 1 0.062 0.098 0.144 0.080 0.090 0.121 0.121 0.118 0.099 0.108 0.201
  0.084 0.065 0.153 0.161   0.080 0.074 0.1 11 0.168 0.102 0.106 0.132 0.131 0.128 0.1 15 0.124 0.194
  0.11 1 0.089 0.195 0.160   0.086 0.080 0.133 0.170 0.134 0.124 0.151 0.150 0.147 0.131 0.154 0.231
  0.099 0.097 0.157 0.154   0.116 0.089 0.077 0.141 0.122 0.112 0.118 0.118 0.115 0.115 0.124 0.200
  0.093 0.087 0.158 0.131   0.103 0.077 0.059 0.144 0.122 0.099 0.115 0.115 0.112 0.099 0.115 0.203
  0.134 0.111 0.170 0.153   0.121 0.102 0.089 0.170 0.144 0.111 0.144 0.144 0.141 0.118 0.134 0.232
  0.087 0.078 0.147 0.131   0.096 0.077 0.062 0.121 0.103 0.087 0.096 0.096 0.093 0.084 0.096 0.188
  0.106 0.087 0.161 0.144   0.116 0.093 0.065 0.138 0.119 0.109 0.129 0.128 0.125 0.106 0.102 0.192
  0.096 0.071 0.147 0.118   0.099 0.068 0.062 0.127 0.102 0.086 0.108 0.108 0.105 0.090 0.115 0.183
  -0.065 0.161 0.164   0.072 0.071 0.105 0.161 0.112 0.106 0.125 0.125 0.122 0.109 0.118 0.197
  23 -0.141 0.137   0.057 0.056 0.083 0.138 0.087 0.090 0.115 0.115 0.1 12 0.093 0.105 0.202
  53 47 -0.157   0.175 0.137 0.160 0.186 0.164 0.157 0.189 0.196 0.193 0.168 0.182 0.240
  54 46 5 2 -   0.154 0.130 0.140 0.164 0.164 0.154 0.164 0.164 0.161 0.151 0.178 0.252
  54 44 2 1 54   0.164 0.130 0.160 0.177 0.173 0.164 0.184 0.184 0.181 0.153 0.181 0.270
  47 43 12 49   0.164 0.117 0.150 0.161 0.160 0.154 0.178 0.178 0.174 0.157 0.167 0.244
  25 20 57 5 1   -0.077 0.108 0.165 0.112 0.112 0.128 0.128 0.125 0.119 0.138 0.225
  25 20 46 44   27 -0.083 0.130 0.108 0.087 0.105 0.105 0.102 0.090 0.112 0.175
  36 29 53 47   3 7 29 -0.127 0.118 0.099 0.118 0.121 0.118 0.095 0.121 0.191
  55 48 62 54   56 46 44 -0.151 0.144 0.150 0.150 0.147 0.140 0.160 0.218
  38 30 54 54   3 8 37 40 50 -0.062 0.084 0.084 0.081 0.072 0.071 0.166
  36 31 52 5 1   3 8 30 34 48 22 -0.053 0.053 0.050 0.025 0.047 0.172
  42 39 61 54   43 36 40 50 29 19 -0.005 0.002 0.045 0.062 0.187
  42 39 63 54   43 36 41 50 29 19 2 -0.002 0.044 0.062 0.190
  41 38 62 53   42 35 40 49 28 18 1 1 -0.042 0.059 0.187
  37 32 55 50   40 31 33 47 25 9 16 16 15 -0.050 0.173
  40 36 59 5 8   46 38 41 53 25 17 22 22 21 18 -0.176
  63 64 75 78   70 57 61 69 54 56 60 61 60 56 57 -
  -    

TABLE4. Comparison of phylogenetic information generated by ITS

Although the primary systematic focus of this study

(this paper) and cpDNA restriction site (Francisco-Ortega, Jansen,

was on Argyranthemum, there are several implications for

and Santos-Guerra. 1996) data in Argyranthemum. The analysis

Anthemideae as a whole. Most subtribal circumscriptions

included the 20 Chrysantheminae taxa listed in Table 1. Hete~unthemis, Ismeliu, and the two Chrysanthemum species were selected and relationships suggested by Bremer and Humphries' as outgroup. Number of variable and informative positions within (1993) morphological cladistic analysis and the "groups" Argyrunthemun~are given in brackets. g, value calculated for proposed by Heywood and Humphries (1977) are not 100 000 random trees.

congruent with the ITS phylogeny. Examples of incon- gruence between the ITS phylogeny and Bremer and

cpDNA restriction

Parnrneter s~tedata ITS sequesiclrlg Humphries' (1993) cladograms concern the lack of sup- port for monophyly of the subtribes Achilleinae, Leucan-

Number of variable positions Number of informative positions theminae, Matricariinae, and Tanacetinae, However, Bre- Number of base pairs examined mer and Humphries did suggest that the Matricariinae Number of shortest trees and Tanacetinae may be paraphyletic. Furthermore, Tan-

Length of shortest tree

acetum and Dendranthema, which were considered as

Number of resolved nodes

part of the "Chrysanthemum complex" by Heywood and

Consistency Index (excluding autapomorphies) Humphries (1977) are not related in the ITS tree. Retention Index The ITS phylogeny, however, does show that a major Skewness of tree-length division in the tribe is diagnosed by a large, -17-bp de-

distribution (g,)

letion in ITS2 (Fig. 1, Appendix). This deletion was not

Average bootstrap value

detected in any taxa restricted to the Far East and South

A

Argyranthemumfoeniculaceum --32 Heteranthemis viscidehirta

Chnlsanthemum coronarium 8 L chrysanthemum segetum

41 Santolina chamaecyparissus

7

B

Argyranthemumfoenicuiac~um Hetermthemis viscidehirta Ismelia carinata

100

Chrysanthemum coraarium 37 Chrysanthemum segetum 29 Santolina chamaecyparissus

Leucanthemum maximum

l6 Anthemis maritima

l6 Anthemis tinctoria

Argyrmthemumfoenicuhceum
24 Heteranthemis viscidehirta
Ismelk carinata
Chrysanthemum coronarium
100       Chrysanthemum segetum
19       22 Santolina chamaecyparissus
        Leucanthemum mavimum
  r I 81 5 lool 23 Anthemis cretica p, Anthemis maritima lo Anthemis arvensis I 9 l2 Anthemis tinctoria 1

Fig. 3. Three phylogenetic reconstructions obtained after parsimony analysis of: (A) ITS nucleotide sequences and cpDNA restriction sites (based on Francisco-Ortega et al., 199%) (412 steps; CI = 0.684, with- out autapomorphies; RI = 0.831, number of most parsimonious trees = 2). (B) cpDNA restriction site data (based on Francisco-Ortega et al., 1995b) (223 steps; CI = 0.695, without autapomorphies; RI = 0.848, number of most parsimonious trees: 1). (C) ITS nucleotide sequences (185 steps: CI = 0.694, without autapomorphies; RI = 0.829, number of most parsimonious trees = 6). Bootstrap values are indicated along nodes in large boldface. Nurnber of steps is also indicated along branch- es. They are also given in the strict consensus tree (A and C) for those branches fully resolved in all of the most parsimonious trees.

Africa nor in Artemisiinae. Conclusions regarding gener- ic relationships in Anthemideae should be viewed as ten- tative because sampling was incomplete and many of the basal nodes in the ITS tree are weakly supported. Mo- lecular systematic studies of the entire tribe currently un- derway by L. Watson (personal communication, Miami University) will provide a more comprehensive under- standing of generic relationships.

Evolutionary implications-Plants endemic to Atlan- tic oceanic islands have been frequently regarded as rel- icts (Engler, 1879; Sunding, 1979; Cronk, 1992). Paleo- botanical evidence and the high frequency of woodiness (70%) among Macaronesian endemics have been used to support this hypothesis (Bramwell, 1972a, 1976; Al-

TABLE 5. Nucleotide sequence divergence expressed as percentage

from cpDNA restriction site (above diagonal from Francisco-Or-

tega, Jansen, and Santos-Guerra, 1996) and ITS data (below diag-

onal). Taxon abbreviations in Appendix 1 except for ARAD: Ar

gyrnnthernztm ndtutcturn subsp. tzdauctctm, ARFR: A. ,frzttescens

subsp. frutescens, ARPI: A. pi~zizatijdzmnl subsp. pinnntijdum,

ARTE: A. tenerifne.

Taxon ISCA CHCO CHSE. HEVl ARAII ARFR ARPI ARTE

ISCA CHCO CHSE HEVI ARAD ARFR ARPI ARTE

dridge, 1979). Tertiary fossils have been identified in Eu- ropean sites of at least six plant genera [i.e., Apollonias Nees (Lauraceae), Clethra L. (Clethraceae), Dracaena L. (Dracaenaceae), Ocotea Aubl. (Oleaceae), Persea Mill. (Lauraceae), and Picconia DC. (Lauraceae)] that do not currently occur in Europe. However, all of these genera have representatives in Macaronesia, and they provide the strongest evidence supporting the relictual nature of the Macaronesian flora (Saporta, 1865; Depape, 1922; Sun- ding, 1979). The ITS phylogeny of Anthemideae placed Argyranthemum in a derived position among 17 Medi- terranean genera, although support for this placement was very weak (Fig. 1) and collapsed in some trees generated by different methods of coding gaps (Fig. 2A, D). Thus, the ITS data could not resolve the position of Argyranthemum and its continental relatives. However, cpDNA restriction site data alone and in combination with the ITS data provide strong evidence that Argyranthemum was as old as the other three genera of Chrysantheminae (Fig. 3A-B). This suggests that either (1) Chrysantheminae originated in Macaronesia and a founder coloniza- tion from this region into the continent gave rise to the other three genera, (2) Argyranthemum was originally on the continent and in Macaronesia but became restricted to the Canaries and Madeira following extinction in the Mediterranean region, or (3) two independent radiations from a common extinct ancestor took place simultaneous- ly both in Macaronesia and in the Mediterranean basin. Under the third hypothesis, the first radiation led to the three continental genera, whereas the second radiation only involved differentiation of Argyranthemum in the Macaronesian islands. Ar~vranthemumdoes not occur in

",

the Mediterranean basin, but there are several taxonomic groups [i.e., Sonchus subgenus Dendrosonchus, and Cheirolophus Cass. (Asteraceae)] with a high number of species in Macaronesia and that also occur in southern Iberia and/or Western Morocco (Peltier, 1973; Boulos, 1974; Dostal, 1976). Molecular phylogenies of the pre- dominantly Macaronesian genus Aeonium (Crassulaceae) show that the insular species are basal to the African species (Mes and Hart, 1996), suggesting that dispersal from Macaronesia to the mainland is possible. Thus, we cannot rule out dispersal of Argyranthemum from Ma- caronesia to the continent in the evolutionary history of the genus.

The position of Argyranthemum in Chrysantheminae

TABLE6. Evaluation of molecular clock hypothesis between six selected taxa of the Anthemideae (indicated in bold line in Fig. 1). Relative rate test values and variances (brackets) are shown above diagonal. Significance levels of evolutionary rate differences are shown below the diagonal; not significant differences (ND), significant at 1% level (***). Taxon abbreviations are in the Appendix.

Ta~on ARE0 PRHl ROGA

ARFO 0.105 (0.02) 0.035 (0.01) PRHI *** 0.070 (0.02)

:b * * ***

ROGA

*** :b *:b

TRPE ND

***

ANCR ND ND AJPA ND *** ND

is also concordant with the patterns of isozyme (Francis- co-Ortega et al., 1995a) and cpDNA restriction site (Fran- cisco-Ostega et al., 1995b) variation. Isozyme diversity and number of unique alleles are higher in Argyranrhe- mum than in any of the other three genera of Chrysan- theminae (Francisco-Ostega et al., 1995a). Interspecific cpDNA divergence is as high within Argyranthemum as between other genera in Chrysantheminae (Francisco-Or- tega et al., 1995b, Francisco-Ortega, Jansen, and Santos- Guerra, 1996). These results are congruent with the hy- pothesis that Argyranthemum may represent an old lin- eage within Chrysantheminae that has subsequently spe- ciated in the Macaronesian islands (Francisco-Ortega et al., 1995a, 1997).

Molecular phylogenies of several Macaronesian groups suggest that the island taxa may be derived rather than basal when compared with closely related continental taxa (Warwick and Black, 1993: Badr, Martin, and Jen- sen, 1994; Knox and Palmer, 1995; Kuss and Wink, 1995; Mes and Hart, 1996; Susanna et al., 1995; Whitton, Wal- lace, and Jansen, 1995; Kim et al., 1996; Bohle, Hilger, and Martin, 1996; Carvalho and Culham, in press; Kim, Crawford, and Jansen, 1996). Three notable examples are the Sonchus alliance [i.e., Sventenia Font Quer, Taeck- holmia Boulos, Babcockia Boulos, Lactucosonchus (Sch. Bip.) Svent., Sonchus subg. Dendrosonchus] (Kim et al., 1996; Kim, Crawford, and Jansen, 1996), the endemic species of Echium (Bohle, Hilger, and Martin, 1996), and genera of the subfamiliy Sempervivoideae (Crassulaceae) (i.e., Aichpson Webb & Berth., Aeonium, and Monanthes Haw.) (Van Ham, 1994; Van Ham et al., 1994; Mes and Hart, 1996). These studies indicate that (1) many of the Macaronesian taxa had a Mediterranean origin and (2) none of the above Macaronesian taxa are clearly basal to the continental genera. There are, however, two notable exceptions to these results. First, the endemic genus Per- icallis D. Don (Asteraceae) appears closely related to the Mesoamerican genus Roldana La Llave & Lex. (Knox and Palmer, 1995). Second, Lavatera phoenicea Vent. (Malvaceae), a rare endemic species restricted to the old- est pasts of the island of Tenerife, occurs in a basal po- sition in the ITS phylogeny of the "Lavatera-Malva com- plex" (Ray, 1995).

An estimate of divergence times from isozymes sug- gests that Argyranthemum differentiated from the rest of Chrysantheminae -2.5-3.0 mya (Francisco-Ortega et al., 1995a). Similar estimates can be made from the pub- lished nucleotide sequence divergence values of the cpDNA restriction site data (Table 2, in Francisco-Ortega, Jansen, and Santos-Guerra, 1996). The cpDNA tree (Fig. 3B) does not support a sister-group relationship between Argyranthemum and any one of the genera of Chrysan- theminae. Therefore, estimations of divergence times were based on the average value of all the pairwise com- parisons of nucleotide sequence divergence between Ar- gyranthemum and the thee remaining genera. Divergence values ranged between 0.298 (A. haouarytheum vs. Chrysanthemum segetum) and 0.155 (A. haouarythecim vs. Heteranthemis), with an average value of 0.217. Pre- vious estimates of substitution rates for the chloroplast genome were -nucleotide substitutions per site per year (Zurawski, Clegg, and Brown, 1984; Zurawski and Clegg, 1987). Parks and Wendel (1990), based on fossil evidence, suggested 0.1 % cpDNA divergence per million years. Using this value divergence times between Argyr- anthemum and the rest of Chsysantheminae would range between 1.5 and 3 mya, with an average value of 2.2 mya.

TRPE ANCR AJPA
0.060 (0.02) 0.066 (0.02) 0.025 (0.02)
0.270 (0.02) 0.039 (0.03) 0.080 (0.03)
0.025 (0.02) 0.03 1 (0.02) 0.010 (0.02)
  0.006 (0.02) 0.035 (0.02)
ND   0.041 (0.02)
ND ND  

The rate of change of ITS sequences in Heteranthemis was significantly higher than in Argyranthemum, Chry- santhemum, and Zsmelia. Thus, reliable estimates of di- vergence times are not possible between Argyranthemum and this genus. The accelerated rate of ITS sequence evo- lution in Heteranthemis is surprising because both iso- zyme (Francisco-Ortega et al., 1995a) and cpDNA re- striction site data (Francisco-Ortega et al., 1995b, Fran- cisco-Ortega, Jansen, and Santos-Guerra, 1996) indicate that Heteranthemis is the least divergent genus from Ar- gq~ranthemum.

The average rate of nucleotide substitutions in ITS has been estimated for Dendroseris (Sang et al., 1994), Gos- sypium L. (Malvaceae) (Wendel, Schnabel, and Seelanan, 1995), Betulaceae (Savard, Michaud, and Bousquet, 1993), and Winteraceae (Suh et al., 1993). We select the average value of 0.78% sequence divergence per million years proposed by Sang et al. (1994) for Dendroseris for estimates involving Argyranthemum, Chrysanthemum, and Zsmelia for three reasons: (1) the genera are in the same family (Asteraceae), (2) Dendroseris and Argyr- anthemum are endemic to oceanic islands, and (3) species of both genera are woody perennials, thus the effect of generation time on the substitution rate of the ITS sequences would be minimized (Soltis and Soltis, 1995). Estimated divergence times of Argyranthemum vs. Chry- santhemzim and Zsmelia range between 0.26 mya (Argyr- anthemum vs. C. coronarium) and 2.1 mya (Argyranthe- mum vs. C. segetum) with an average value of 0.98 mya.

Times of divergence from the isozyme data (2.5-3.0 mya) and cpDNA restriction site data (1.5-3.0 mya) are not in agreement with those obtained from the ITS (0.26-

2.1 mya). Although there is some overlap, ITS diver- gence times are lower and only the date from the pairwise

JOURNAL OF BOTANY

comparison between Argyranthemum and C. segetum (2.1 mya) is similar to those obtained for the other two mark- ers. Rejection of a molecular clock for many comparisons in Anthemideae (Table 6) probably accounts for this dis- agreement. Sequence evolution of the ITS was also slow in Argyranthemum, Zsmelia, and C. coronarium. This ex- plains why virtually no ITS variation was detected within Argyranthemum, despite the high levels of isozyme (Francisco-Ortega et al., 1995a, 1997) and cpDNA (Fran- cisco-Ortega, Jansen, and Santos-Guerra, 1996) variation.

European Tertiary fossils and disjunct distributions be- tween Macaronesia and East Africa [i.e., Campylanthus Roth (Scrophulariaceae), Canarina L.(Campanulaceae), Drusa DC. (Apiaceae)] suggest an ancient origin for some endemics [reviewed by Bramwell (1972a, 1976, 1985)l. Further studies, especially in the families Cleth- raceae, Dracaenaceae, Lauraceae, and Oleaceae and in genera with disjunctions between East Africa and Ma- caronesia, are needed to test previous hypotheses that some members of the Macaronesian flora are relictual. The recent discovery of 13-mya-old plant fossils in southwestern Gran Canaria (Garcia-Talavera, SAnchez- Pinto, and Socorro-HernQndez, 1996) will provide new insights into time of origin and relationships of the Ma- caronesian flora with the pre-Pleistocene Mediterranean flora.

Phylogenetic utility of ITS and cpDNA restriction site data in oceanic island groups-DNA sequences from the ITS have been widely used for phylogenetic compar- isons of plants endemic to oceanic islands (Sang et al., 1994, 1995; Baldwin and Robichaux, 1995; Mes, 1995; Francisco-Ortega et al., 1996a; Kim et al., 1996; Kim, Crawford, and Jansen, 1997). Comparisons of levels of sequence divergence in taxa from the Hawaiian and Juan Fernandez Islands have indicated higher levels of varia- tion in the ITS region than in the chloroplast genome. Furthermore, in most island plant groups studied there has been sufficient divergence to use this marker to re- solve phylogenetic relationships. Thus, many recent stud- ies have focused on ITS sequences to the exclusion of cpDNA restriction sites for phylogenetic comparisons in recently evolved island groups. However, within Argyranthemum the ITS is essentially useless for examining evolutionary relationships (Table 4). We detected only 41 variable positions in the ITS regions of the 16 species of Argyranthemum and four outgroups sequenced, with only seven of these changes being phylogenetically informa- tive. In contrast, the cpDNA restriction site approach identified 122 variable positions and 72 informative changes for the same 20 taxa. The primary reason that more variation was detected by the cpDNA approach was that we sampled -30 times more nucleotides (14 716 for restriction sites and 503 for ITS). Levels of sequence di- vergence were actually higher for the ITS than the chlo- roplast genome in most pairwise comparisons in the sub- tribe Chrysantheminae (Table 5). Thus, the ITS region evolved more rapidly than the chloroplast genome in Argyranthemum and in other island groups examined (Sang et al., 1994, 1995; Baldwin and Robichaux, 1995).

Our results in Argyranthemum may seem surprising in view of previous studies where both cpDNA restriction sites and ITS sequences have been compared (Sang et al.,

1994, 1995). We would argue that the results are in fact not incongruent with previous studies and that restriction site analysis of the entire chloroplast genome will in gen- eral provide more phylogenetic information than ITS se- quences using the approach that we employed in Argyranthemum (Jansen. Wee. and Millie. in mess). This in-

, A ,

volved sampling numerous restriction sites with very fre- quent-cutting enzymes including four-base cutters, running very long gels (up to 18 cm) to maximize sep- aration of fragments, and using smaller hybridization probes so that fewer fragments are visualized during each round of hybridization. This same approach was used by Mason-Gamer, Holsinger, and Jansen (1995) in Coreopsis grandiflora and they identified more variation within a single species than a previous study identified among all nine species in the section (Crawford, Palmer, and Ko- bayashi, 1990) to which this species belongs. We are not arguing that the ITS region is not useful in island groups because there is ample evidence from several studies (i.e., Sang et al., 1994, 1995; Baldwin and Robichaux, 1995; Francisco-Ortega et al., 1996a; Kim et al., 1996, Kim, Crawford, and Jansen, 1996) that phylogenetic informa- tion can be sometimes obtained using this marker. How- ever, we believe that the cpDNA restriction site approach should not be abandoned because it will usually yield more phylogenetic information than ITS sequencing even in recently evolved groups. Our preliminary results in three other Macaronesian groups [Cheirolophus, Perical- lis, and Tolpis Adans. (Asteraceae), J. Francisco-Ortega and R. K. Jansen, unpublished data] confirm the pattern observed in Argyranthemum.

LITERATURE CITED

ALBERT,J., AND M. FENYO. 1990. Simple, sensitive and specific detec- tion of human inmmunodeficiency virus type 1 in clinical speci- mens by polymerase chain reaction with nested primers. Joc~rnal of Clinical Microbiology 28: 1560-1564.

ALDRIDGE,A. E. 1979. Evolution within a single genus: Sonchus in Macaronesia. In D. Bramwell [ed.], Plants and islands, 279-291. Academic Press, London.

BADR, A., W. MARTIN, AND U. JENSEN. 1994. Chloroplast DNA restric- tion site polymorphism in Genisteae (Leguminosae) suggests a common origin for European and American lupines. Plant System- atics and Evolution 193: 95-106.

BAIN,J., AND R. K. JANSEN. 1995. A phylogenetic analysis of the au- reoid Senecio (Asteraceae) complex based on ITS sequence data. Plant Systematics and Evolution 195: 209-219.

BALDWIN,B. G. 1992. Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: an example from the Compositae. Molecular Plzylogenetics and Evolution 1: 3-1 6.

, AND R. H. ROBICHAUX. 1995. Historical biogeography and ecology of the Hawaiian silversword alliance (Asteraceae). New molecular phylogenetic perspectives. In W. L. Wagner, and V. A, Funk [eds.], Hawaiian biogeography: evolution on a hot spot ar- chipelago, 259-287. Smithsonian Institution Press, Washington, DC.

BOHLE,U.-R., H. H. HILGER, AND W. F. MARTIN. 1996. Island coloni- zation and evolution of the insular woody habit in Echium L. (Bor- aginaceae). Proceedings of the National Academy of Sciences, USA

93: 11740-1 1745.

BORGEN, L. 1972. Embryology and achene morphology in endemic Canarian species of Chrysanthemum (L.) Hoffm. subgenus Argyranthemum (Webb) Harling. Norwegian Journal of Botany 19: 149-

170. Bou~os,L. 1974. RCvision systkmatique du genre Sonchus s. 1. V.

Sous-genre 2. Dendrosonclzus. Botaniska Notiser 127: 7-37. BRAMWELL, 1972a. Endemism in the flora of the Canary Islands. In

D.

D. H. Valentine [ed.]. Taxonomy, phytogeography and evolution, 141-159. Academic Press, London. . 1972b. A revision of the genus Echium in Macaronesia. Lagascalia 2: 37-1 15.

, 1976. The endemic flora of the Canary Islands; distribution, relationships and phytogeography. In G. Kunkel [ed.], Biogeogra- phy and ecology in the Canary Islands, 207-240. W. Junk, The Hague.

. 1985. Contribucidn a la biogeografia de las Islas Canarias. Botdnica Macaronksica 14: 3-34.

BREMER.K., AND C. J. HUMPHRIES. 1993. Generic monograph of the Asteraceae-Anthemideae. Bulletin of the Natural History M~tseum of London (Botany) 23: 71-177.

CARLQLIIST, 1962. A theory of paedomorphosis in dicotyledonous

S. woods. Phytomorphology 12: 30-45. . 1965. Island life: a natural history of the islands of the world. Natural History Press, New York, NY. . 1974. Island biology. Columbia University Press, New York,

NY. CARVALHO, In press. Conservation status and

J. A., AND A. CULHAM. phylogenetics of Isoplexis (Lindl.) Benth. (Scrophulariaceae): an endemic Macaronesian genus. Bulletin do Museo Municipal do Funchal.

CHRISTENSEN, 1992. Acetylenes and related compounds in An-

L. I?

themideae. Phytochemistry 3 1 : 7-49. CRAWFORD,D. J., J. D. PALMER, AND M. KOBAYASHI.

1990. Chloroplast DNA restriction site variation and the phylogeny of Coreopsis section Coreopsis (Asteraceae). American Journal of Botany 77: 552-

558. CRONK,Q. C. B. 1992. Relict floras of Atlantic islands: Patterns as- sessed. Biological Journal of the Linnean Society 46: 91-103. DEPAPE, G. 1922. Recherches sur la flora pliocgne de la Vall&e du

RhBne. Annales des Sciences Narurelles Series I0 4: 73-265. DOSTAL.J. 1976. Cheirolophus. Cass. In T. G. Tutin, V. H. Heywood,

N. A. Burges, D. M. Moore, D. H. Valentine, S. M. Walters, and

D. A. Webb [eds.], Flora europaea, vol. 4, 249-250. Cambridge University Press, Cambridge.

DOYLE,J. J., AND J. L. DOYLE. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf material. Phytochentical Bulletin

19: 11-15.

ENGLER,A. 1879. Versuch einer Entwicklungsgeschichte der Pfianzen- welt, insbesondere der Florengebiete seit der Tertiarperiode. I. Die extratropischen Gebiete der nordlichen Hemisphare. W. E. Verlag, Leipzig.

FARRIS,J. S. 1989. The retention index and homoplasy excess. Systematic Zoology 38: 406-407. FELSENSTEIN, 1985. Confidence limits on phylogenies: an approach

J.

using bootstrap. Evohction 39: 783-791. FRANCISCO-ORTEGA, A. SANTOS-GUERRA, AND

J., D. J. CRAWFORD, S. SA-FONTINHA.1995a. Genetic divergence among Mediterranean and Macaronesian genera of the subtribe Chrysantheminae (Aster- aceae). Anzerican Journal of Botany 82: 1321-1328.

, R. K. JANSEN, D. J. CRAWFORD,AND A. SANTOS-GUERRA. 1995b. Chloroplast DNA evidence for intergeneric relationships of the Macaronesian endemic genus Argyranthetnum (Asteraceae). Systematic Botany 20: 4 13-422.

--,AND A. SANTOS-GUERRA.

1996. Chloroplast DNA ev- idence of colonization, adaptive radiation, and hybridization in the evolution of the Macaronesian flora. Proceedings of the National Academy of Sciences, USA 93: 4085-4090.

,J. FUERTES-AGUILAR, A. SANTOS-

S.-C. KIM, D. J. CRAWFORD, GUERRA,AND R. K. JANSEN. 1996a. Molecular evidence for the origin, evolution, and dispersal of Cranzbe (Brassicaceae) in the Macaronesian islands. In Anonymous [ed.], Second symposium on fauna and flora of the Atlantic Islands, 12-16 February 1996, 41. Universidad de Las Palmas de Gran Canaria, Gran Canaria, Canary Islands (Abstract).

, D. J. CRAWFORD, AND

A. SANTOS-GUERRA, J. A. CARVALHO. 1996b. Isozyme differentiation in the endemic genus Argyranthetnum (Asteraceae: Anthemideae) in the Macaronesian islands. Plant Systematics and Evolution 202: 137-152.

, AND R. K. JANSEN. 1997. Origin and evolu- tion of Argyranthemum (Asteraceae: Anthemideae) in Macaronesia.

In T. Givnish, and K. Sytsma [eds.], Molecular evolution and adap- tive radiation, 407-431 Cambridge University Press, Cambridge.

GARCIA-TALAVERA, AND S. SOCORRO-HERNANDEZ.

I?, L. SANCHEZ-PINTO, 1996. Fossil plants in the thrachytic-syenitic complex of Gran Ca- naria. In Anonymous [ed.], Second symposium on fauna and flora of the Atlantic Islands, 12-16 February 1996. 233. Universidad de Las Palmas de Gran Canaria, Gran Canaria, Canary Islands (Ab- stract).

GREGER,H. 1977. Anthemideae-Chemical review. In V. H. Heywood,

J. B. Harborne, and B. L. Turner [eds.], The biology and chemistry of the Compositae, vol. 2, 899-941. Academic Press, London. HANSEN,A,, AND l? SUNDING.1993. Flora of Macaronesia. Checklist of vascular plants, 4th ed. Sommefeltia 17: 1-296. HARLING.G. 1950. Embryological studies in the Compositae. Part I. Anthemideae-Anthemidinae. Acra Horti Bergiani 15: 135-168. . 1954. Embryological studies in the Compositae. Part 11. Anthemideae-Chrysantheminae. Acta Horti Bergiani 16: 1-56.

. 1960. Further embryological and taxonomical studies in Anthemis L. and some related genera. Sveilsk Botanisk Tidskrift 54: 571-590.

HEYWOOD, 1976. Chrysanthemum L. In T. G. Tutin. V. H. Hey-

V. H. wood, N. A. Burges, D. M. Moore, D. H. Valentine, S. M. Walters, and D. A. Webb [eds.], Flora europaea, vol. 4, 168-1 69. Cambridge University Press, Cambridge.

, AXD C. J. HUMPHRIES. 1977. Anthemideae-Systematic review. In V. H. Heywood, J. B. Harborne, and B. L. Turner [eds.], The biology and chemistry of the Compositae, vol. 2, 851-898. Academic Press, London.

HIGGINGS,D. G., A. J. BLEASBY,

AND R. FUCHS. 1991. CLUSTAL V: improved software for multiple sequence alignment. Computer Ap- plications in the Biosciences 8: 189-191.

HILLIS, D. M., AND J. F? HUELSENBECK.

1992. Signal, noise, and reli- ability in molecular phylogenetic analyses. Journal of Heredit). 83: 189-195.

HOLMGREN, AND L. C. BARNETT.

I? K., N. H. HOLMGREN, 1990. Index herbariorum. Part 1: the herbaria of the world, 8th ed. Regnum Vegetabile 120: 1-693.

HOWARD,R. A. 1973. The vegetation of the Antilles. In A. Graham [ed.], Vegetation and vegetational history of northern Latin Amer- ica, 1-38. Elsevier, Amsterdam.

HUMPHRIES,

C. J. 1976a. A revision of the Macaronesian genus Argyranthemum Webb ex Schultz-Bip. (Compositae-Anthemideae). Bulletin of the British Museum (Natural History) Botany 5: 147-240.

. 1976b. Evolution and endemism in Argyranthemum Webb ex Schultz Bip. (Compositae: Anthemideae). Botanica Macaronksica

1: 25-50. HUTCHINSON, 1917. Notes on African Compositae: IV. Kew Bulletin

J.
1917: 11 1-1 18.

JANSEN,R. K., J. L. WEE, AND D. MILLIE. In press. Comparative utility of chloroplast DNA restriction site and sequence data for phylo- genetic studies in plants. In S. Soltis, D. S. Soltis, and J. J. Doyle [eds.], Molecular systematics of plants. 2d ed. Chapman Hall, New York, NY.

JEFFREY,C. 197 1. Chrysanthemum ptarmicij'lorum. Curtis' Botanical Magazine 178: 583.

KIM, K.-J., AND R. K. JANSEN. 1994. Comparisons of phylogenetic hy- pothesis among different data sets in dwarf dandelions (Krigia): additional information from internal transcribed spacers sequences of nuclear DNA. Plant Systematics and Evolution 190: 157-185.

, AND -, 1995. ndhF sequence evolution and the major clades in the sunflower family. Proceeditzgs of the National Acad- emy of Sciences, USA 92: 10379-10383.

KIM, S.-C., D. J. CRAWFORD,

J. FRANCISCO-ORTEGA,AND A. SANTOS- GUERRA. 1996. A common origin for woody Sonchus and five related genera in the Macaronesian Islands: molecular evidence for extensive radiation. Proceedings of the National Academy of Sci- ences, USA 93: 7743-7748.

--, AND R. K. JANSEN. 1996. Phylogenetic relationships among the genera of the subtribe Sonchinae (Asteraceae): evidence from ITS sequences. Systematic Botany 21: 4 17-432. KIMURA,M. 1980. A simple method for estimating evolutionary rates of base substitution through comparative studies of nucleotide se- quences. Journal of Molecular Evolution 16: 11 1-120.

KITAMURA,S. 1978. Dendranthemum et Nipponanthemum. Acta Phy- totaxonomica et Geobotanica, Kyoto 29: 165-170. KLUGE, A. G., AND J. S. FARRIS. 1969. Quantitative phyletics and the evolution of anurans. Systematic Zoology 18: 1-32.

KNOX,E. B., AND J. D. PALMER. 1995. The origin of Dendrosenecio within the Senecioneae (Asteraceae) based on chloroplast DNA ev- idence. American Journal of Botany 82: 1567-1573.

KUMAR,S., T. KOICHIRO, AND M. NEI. 1993. MEGA: molecular evo- lutionary genetics analysis, version 1.0. Pennsylvania State Uni- versity, University Park, PA.

Kuss, E., AND M. WINK. 1995. Molecular phylogeny of the Papilio- noideae (family Leguminosae): rbcL genes sequences versus chem- ical taxonomy. Botanica Acta 108: 149-162.

LI, W.-H., AND M. TANIMURA.

1987. The molecular clock runs more slowly in man than in apes and monkeys. Nature 326: 93-96. LOOCKERMAN, R. K. JANSEN. 1996. The use of herbarium

D. J., AND material for DNA studies. In T E Stuessy, and S. H. Sohmer [eds.], Sampling the green world, 205-220. Columbia University Press, New York, NY.

MADDISON, 1991. The discovery and importance of multiple is- D. R.

lands of most parsimonious trees. Systematic Zoology 40: 315-328. MADDISON, 1992. MacClade, version 3.

W. P., AND D. R. MADDISON.

Sinauer, Sunderland, MA. MASON-GAMER, AND R. K. JANSEN.

R. J., K. E. HOLSINGER, 1995. Chlo- roplast DNA haplotype variation within and among populations of Coreopsis grandijlora (Asteraceae). Molecular Biology and Evo- lution 12: 371-381.

MES, T. H. M. 1995. Origin and evolution of the Macaronesian Sem- pervivoideae (Crassulaceae). Ph.D. dissertation, University of Utrecht, Utrecht.

, AND H. 'T HART. 1996. The evolution of growth-forms in the Macaronesian genus Aeonium (Crassulaceae) inferred from chlo- roplast DNA RFLPs and morphology. Molecular Ecology 5: 351-

q<9

JUJ.

PARKS,C. R., AND J. E WENDEL. 1990. Molecular divergence between Asian and North American species of Liriodendron (Magnoliaceae) with implications of fossil floras. American Journal of Botany 77: 1243-1256.

PELTIER,J. R. 1973. EndCmiques macaronCsiknnes au Maroc. Inven- taire bibliographique et problkmes taxonomiques. Monographiae Biologicae Canariensis 4: 134-142.

RAY, M. E 1995. Systematics of Lavatera and Malva (Malvaceae, Mal- veae)-A new perspective. Plant Systematics and Evolution 198: 29-53.

SANG, T., D. J. CRAWFORD,

S. C. KIM, AND T. E STUESSY. 1994. Radi- ation of the endemic genus Dendroseris (Asteraceae) on the Juan Fernandez Islands: evidence from sequences of the ITS regions of nuclear ribosomal DNA. American Journal of Botany 81: 1494- 1501.

--, T. E STUESSY,AND M. SILVA 0. 1995. ITS sequences and the phylogeny of the genus Robinsonia (Asteraceae). Systematic Botany 20: 55-64.

SAPORTA,G. 1865. Etudes sur la vCgttation du sud-est de la France B l'epoque tertiaire. Annales des Sciences Naturelles Series 5 4: 5-

264.

SAVARD,L., M. MICHAUD, AND J. BOUSQUET. 1993. Genetic diversity and phylogenetic relationships between birches and alders using ITS, 18s rDNA, and rbcL gene sequences. Molecular Phylogenet- ics and Evolution 2: 1 12-1 18.

SOLTIS,P. S., AND D. E. SOLTIS. 1995. Plant molecular systematics. inferences of phylogeny and evolutionary processes. Evolutionary Biology 28: 139-194.

SORENG,R. J., AND E. A. COPE. 1991. On the taxonomy of cultivated species of the Chrysanthemum genus-complex (Anthemideae; Compositae). Baileya 23: 145-165.

SUH, Y., L. B. THIEN, H. E. REEVE, AND E. A. ZIMMER. 1993. Molecular evolution and phylogenetic implications of internal transcribed spacer sequences of ribosomal DNA in Winteraceae. American Journal of Botany 80: 1042-1055.

SUNDING, 1979. Origins of the Macaronesian flora. In D. Bramwell P.

[ed.], Plants and islands, 13-40. Academic Press, London. SUSANNA, D. E. SOLTIS,

A,, N. GARC~A-JACAS, AND P. S. SOLTIS. 1995. Phylogenetic relationships in tribe Cardueae (Asteraceae) based on ITS sequences. American Journal of Botany 82: 1056-1068.

SVENTENIUS,E. 1960. Additamentum ad floram canariensem I. Instituto

Nacional de Investigaciones Agronbmicas, Madrid. SWOFFORD,

D. L. 1993. PAUP: Phylogenetic analysis using parsimony,

version 3.1.1. Illinois Natural History Survey, Champaign, IL. TAKHTAJAN, 1986. Floristic regions of the world (translation by C.

A.
Jeffrey). Oliver and Boyd, Edinburgh.

VAN HAM, R. C. H. J. 1994. Phylogenetic implications of chloroplast DNA variation in the Crassulaceae. Ph. D. dissertation, University of Utrecht, Utrecht.

, H. 'T HART, T. H. M. MES, AND J. M. SANDBRINK.

1994. Mo- lecular evolution of noncoding regions of the chloroplast genome in the Crassulaceae and related species. Current Genetics 25: 558-

566. VOGT, R. 1991. Die Gattung Leucanthemum Mill. (Compositae -An

themideae) auf der Iberischen Halbinsel. Ruizia 10: 1-261. , AND C. OBERPRIELER.

1995. Mauranthemum, a new name for Leucoglossum B. H. Wilcox & al. non S. Imai (Compositae, An- themideae). Taxon 44: 377-378.

WARWICK,S. I., AND L. D. BLACK. 1993. Molecular relationships in subtribe Brassicinae (Cruciferae, tribe Brassiceae). Canadian Jour- nal of Botany 7 1: 906-9 18.

WATSON.L. E. 1996. Molecular svstematics of the tribe Anthemideae. In D. J. N. Hind, and H. J. Beentje [eds.], Compositae: Systematics, vol. 1. Proceedings of the International Compositae Conference, Kew, 1994, 341-348. Royal Botanic Gardens, Kew.

WENDEL,J. F, A. SCHNABEL, AND T. SEELANAN. 1995. An unusual ri- bosomal DNA sequence from Gossypium gossypioides reveals an- cient, cryptic, intergenomic introgression. Molecular Phylogenetics and Evolution 4: 298-3 13.

WHITE, T. J., T. BRUNS, S. LEE, AND J. TAYLOR. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylo- genetics. In M. Innis, D. Gelfand, J. Sninsky, and T. White [eds.], PCR protocols: a guide to methods and application, 3 15-322. Ac- ademic Press, San Diego, CA.

WHITTON,J., R. S. WALLACE, AND R. K. JANSEN. 1995. Phylogenetic relationships and patterns of character change in the tribe Lactuceae (Asteraceae) based on chloroplast DNA restriction site variation. Canadian Journal of Botany 73: 1058-1073.

WINSHIP,P. R. 1989. An improved method for direct sequencing PCR amplified material using dimethyl sulphoxide. Nucleic Acid Research 17: 1266.

Wu, C.-I., AND W.-H. LI. 1985. Evidence for higher rates of nucleotide substitution in rodents than in man. Proceedings of the National Academy of Sciences, USA 82: 1741-1745.

ZURAWSKI,G., AND M. T. CLEGG. 1987. Evolution of higher-plant chlo- roplast DNA- encoded genes: implications for structure-function and phylogenetic studies. Annual Review of Plant Physiology 38: 391-419.

--, AND A. H. D. BROWN. 1984. The nature of nucleotide sequence divergence between barley and maize chloroplast DNA. Genetics 106: 735-749.

APPENDIX1. Aligned ITS sequences from the taxa studied. Gaps are indicated by "-". Uncertain sites are shown as "?". Polymorphic sites are as follows: K = GIT, M = A/C, R = NG,S = C/G,Y = C/T, V = NCIG, W = NT,. Species abbreviations are as follows: Achillea millefolium (ACMI); Ajania paci$ca (AJPA); Anthemis lawensis (ANAR); A. cretica (ANCR); A. maritima (ANMA); A. tinctoria (ANTI); Anacyclus pyrethrum (ANPY); Arctanthemum arcticum (ARAR); Artemisia canariensis (ARCA); Argyrantlzemum foeniculaceum (ARFO); Chrysantlzemum coronarium (CHCO); C. segetum (CHSE); Cladanthus arabicus (CLAR); Coleosteplzus myconis (COMY); Cymbopappus adenosolen (CYAD); Dendranthema coreanum (DECO); Glossopappus macrotus (GLMA); Heteranthemis viscidelzirta (HEVI); Hymenostemma pseu- doantlzemis (HYPS); Ismelia carinata (ISCA); Leucantlzemopsis alpina (LEAL); Leucanthemum maximum (LEMA); Lepidophorum repandum (LERE); Leucanthemella serotina (LESE); Lonas annua (LOAN); Matricaria chamomilla (MARE); Mauranthemum paludosum (MAPA); Nipponanthemum nipponicum (NINI); Otospermum glabrum (OTGL); Plzalacrocarpon hoffmannseggii (PHHO); Prolongoa Izispanica (PRHI); Rhodanthemum gayanum (ROGA); Santolina chamaecyparissus (SACH); Tanacetum microphyllum (TAMI); T. vulgare (TAVU); Tripleurospermum perforatum (TRPE); Ursinia antlzemoides (URAN). The seven variable positions in Argyrantlzemum are in bold. The three informative positions in Argyranthemum are underlined.

-----
ARFO TCGAACCSTG CAAAGCAGA- ACGACCCGTG AACACGTAAT AATAACCGAG CgCGAG--T
HEVI TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAT AATAACCGAG CVCCGAG--T
ISCA TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAT AATAACCGAG CRCCGAG--T
CHCO TCGAACCCTG CAAAGCAGA- ASGACCCGTG AACACGTAAT AATAACCGAG CGCCGAG--T
CHSE TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACAYGTWAT AATAACCGAG CACCGAG--T
CLAR TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG C-TCGAG--T
TAW TCGAACCCTG CAAAGCAGA- ACGACCCGCG AACACGTAAA AACAACCGAG CGTCGAG--T
TAMI TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG CGTCGAG--T
ANCR TCGAACCCTG CAAAGCAGA- ACGACCCGCG AACACGTAAA A-CAACAGAG CATCGAG--T
ANAR TCGAACCCTG CAAAGCAGA- ACGACCCGCG AACACGTAAA AACAACCGAG CATCGAG--T
ANTI TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG CGTCGAG--T
ANMA TCGAACCCTG CAAAGCAGA- ACGACCCGCG AACACGTAAA A-CAACAGAG CATCGAG--T
ROGA TTGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAT A-CAACCAAG CGTCGAG--T
COW TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAT A-CAATCGAG CGTCGAG--T
GLMA TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAT A-CAATCGAG CGTCGAG--T
LEMA TCGAACCCTG CAAAGCA?A- ACGACCCG?G AACA?GTAAA A???ACCGAG C??CGAG--T
MAPA TCGAACCCTG CAAAGCAGAT ACGACCCGTG AACACGTAAT A?CAATCGAG CGTCGAG--T
SACH TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG CGTCGAG--T
ANPY TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG YGTCGAG--T
ACMI TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACTGAG CGTCGAG--T
PHHO TCGAACCCTG CAAAGCATA- ACGACCCGTG AACATGTAAA AACAATCGGG CGTCAAT--T
HYPS TCGAACCCTG CAAAGCA??- ?CGACCCGCG AACACGTAAA A-CAAACAAG CGTCGAG--T
LERE TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACAAGTAAA A-CAAACGAG CGTCGAG--C
PRHI TCAAACCCTG CAAAGCAGA- ACGACCCGCG AACACGTAAA A-CAAACGAG CGTCGAG--T
LEAL TCGAACCCTG CAAAGCAGA- ACGACCCGCG AACACGTAAA A-CAAACGAG CGTCGAG--T
TRPE TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACTGAG CGTCGAGAGT
MARE TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG TGTCGAG--T
OTGL TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAT A-CAACTGTG TGTCGAG--T
LOAN TCGAACCCTG CAAAGCAGA- AACATCTGTG AACATGTAAA AATGAC-GAG CGTTGAG--T
CYAD TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG TGTCGAT--T
LESE TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACYGAG TGTTGAT--T
DECO TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG TGYTGAG--A
AJPA TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG TGTTGAG--A
ARAR TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG TGTTGAG--A
NINI TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACACGTAAA AACAACCGAG TGTTGAT--T
ARC A TCGAACCCTG CAAAGCAGA- ACGACCYGYG AACACGTAAA AACAAY-GAG CGTCGAT--T
URAN TCGAACCCTG CAAAGCAGA- ACGACCCGTG AACATGTAAC AACAACTGAG CGTCGGG--T

-----...................................................................

1610     AMERICANJOURNALOF BOTANY [Vol. 84
APPENDIX1. Continued.      
    APPENDIX1. Continued.  
70 80 90 100 110 120
ARFO GGGTTAAGCG CTT-TGTTTG A-TCCTCTCG GTGCTTT-GT CGATG-TGCA -TTTACTCGA
HEVI GGGTTAAGCA CTT-TGTTTG A-TCCTCTCG GTGCTTTTGT CGATG-TACA -WTTACTCGA
ISCA GGGTTAAGCG CTT-TGTTTG A-TCCTCTCG GTGCTTT-GT CGATG-TGCA -TTAACTCGA
CHCO GGGTTAAGCG YTT-TGTTTG A-TCCTCTCG GTGCTTT-GT CGATG-TGCA -TTAACTCGA
CHSE GGGTTAAGCA CTT-TGTTTG ?-TCCTATCG GTGCTTT-GT CGATG-TGCA -TTTACTCGA
CLAR GGGTTAAGCA CTT--GTTTG A-TCCTCTCG ATGCTTT-GT CGATG-TGCA -CTTACTCGA
TAW GGATTAAGCA CTT--GTTTG A-TCCTCTCG ATGCTTT-GT CGATG-TGCA -TTTACTTGT
TAMI GGATTAAGCA CTT--GTTTG A-TCCTCTCG ATGCTTT-GT CGATG-CGCA -TTTACTTGT
ANCR GAATTAAG-G TTT--GTTTG A-TACACTCG ATGCTTC-GT CGATG-TGCA -TTTACTTTT
ANAR GGGTTAAGCG CTT--GTTTG A-TCCTCTCG ATGCTTT-GC CGATG-TGCA -TTTACTTGT
ANTI GGATTAAGCA CTT--GTTTG A-TCCTCGCG ATGCTTT-GT CGATG-CGCA -TTTACTTGT
ANMA GAATTAAG-G TTT--GTTTG A-TACACTCG ATGCTTC-GT CGATG-TGCA -TTTACTTTT
ROG A GGGTTAAGCG CTT--GTTTG A-TCCTCTCG ATGCTTT-GC CGATG-CACA -TTAACTTGA
COW GGGTTAAGCG CTT--GTTTG A-TCCTCTCG ATGCTTTTGT CGATG-TGCA -TTCACTTGA
GLMA GGGTTAAGCG CTT--GTTTG A-TCCTCTCG ATGCTTT-GT CGATG-TGCA -TTCACGTGA
LEMA G??TTAAGCG CTA--GTTTG A-TCCTCTCG ATGCTTT-GT CGATG-TGCA -TTCACTTGA
MAPA GGGTTAGGCT CTT--GTTTG A-TCCTCTCG ATGCTTT-GT CGACG-TGCA -TTCACTTGA
SACH GGATTGAGCA CTT--GTTTG A-TCCTCTCG ATGCTTT-GT CGATGATGCA -TTTACTCGA
ANPY GGATTAAGGA ATT-- TTTG A-TCCTCTCA ATGCTCT-GT CGATG-CGCA -TTTACTTTT
ACMI GGATTAAGCA TTT--GTTTG A-TCCTCTCG ATGCTTT-GT CGATG-TGCA -TTTACTTGT
PHHO GGATTAAGCA CTT--GTTTG A-TCCTCTTG GTGCTTT-GT CGATG-TGCA CTTTACTAGA
HY PS GGATTAAGCA TTT--GTTTG A-TCCGCTTG ATGCTTT-GC CGATG-TGCA -TTCACTTGA
LERE --ATTAGGCA CTT--GTTTG G-TCCTCTTG ATGCTTT-GT CGATG-TGCA -TTTACTTG-
PRHI GGGATTAA-- ---------G A-TCCGCTTG ACGCTTT-GT CGATG-TGCA -TTCGGTTGA
LEAL GGATTAAGCA CTT--GTTTG A-TCCGCTTG ATGCTTT-GT CGATG-CGCA -TTCACTTGA
TRPE GGACCAAGGA ATT-- TTTG AATACACTCG ATGCTTT-GT CAATG-TGCA -TTTACCTTT
MARE GGGTTGAGCA CTT--GTTTG GATCCTCTCG ATGCTTT-GT CGATG-TGCA -TTTGCTTGT
OTGL TGGTTAAGCA CTT--GTTTG A-TCATCTCG ATGCTTT-GT CGATG-TGCA -TTCACTCGA
LOAN GGATTAAGCA CTA--GATTG A-TCCTCTCG GCGCTTT-GT CGATG-AGCA -CTGRCTTGG
CYAD GGACTAAGCA CTAGTGTTTG A-TCCTCTCG ACGCTTT-GC CGATG-CGCA -TCTACTCGT
LESE GGATTAAGCG CTT--GTTTG A-TCCTCTCG ACGCTTT-GT CGATG-TGCA -TTTACACGA
DECO GGACCAAGCT CCT--GTTTG A-TCCTCTCG ACGCTTT-GT CGATG-CGCA -TTTACTCGA
AJPA GGACCAAGCT CCT--GTTTG A-TCCTCTCG ACGCTTT-GT CGATG-CGCA -TTTACTCGG
ARAR GGACCAAGCT CCT--GTTTG A-TCCTCTCG ACGCTTT-GT CGATG-CGCA -TTTACTCGA
NINI GGATTAAACG CTT--GTTTG A-TCTTCTCG ACGCTTT-GT CGATG-TGCA -TTTACTCGA
ARCA GGATTAGGCG CTT--GTTTG A-TCCTCTCG ACGCTTT-GT CGACG-CGCA -TTCACTCG-
URAN TTATTGGGCA CAT--GCCTG G-TCTTCCCG GTGCTTA-GC CGATG-CGCG CCTGTCTCGT

APPENCIx 1. Cont lnued.

.......................................................

130 140 150 160 170 ARFO GTCCTTTT-- GGGCCTTGTG AGTGTGTCAT TGGCG-CAAT AACAACCCCC HEVI GTCCTTTT-- GGGCCTTGCG AGTGTGTCAT TGGCG-CAAT AACAACCCCC ISCA GTCCTTTT-- GGGCCTTGTG AGTGTGTCAT TGGCG-CAAT AACAACCCCC CHCO GTCCTTTT-- GGGCCTTGTG AGTGTGTCAT TGGCG-CAAT AACAACCCCC CHSE GYCCTTTT-- GGGTCTTGTG AGTGT--CAT TGGCG-CAAT AACAACCCCC CLAR CTGGTTTT-G GAGTCTTGTG AGTGTGTCGT TGGCG-CAAT AACAACCCCC TAW GTTCTTTT-- GGACATGGTG AATGTGTCAT TGGCG-CAAT AACAACCCCC TAMI GTTCTTTT-- GGACCCGGTG AATGCGTCAT TGGCG-CAAT AACAACCCCC ANCR GTTCTTTT-- TGACATGGTG AATGTT?CAT AGGCGGCAAT AACCAACCCC ANAR GTTCTTTT-- GGACATGATG AATGCGTCGT TGGCGCCAAT AACAACCCCC -GGCACAA-T ANTI GTTCTTTT-- GGACACGGTG AATGTGTCGT TGGACGCAAT AACAACCCCC -GGCACAA-T ANMA GTTCTTTT-- TGACATGGTG AATGTTTCAT AGGCGGCAAT AACCAACCCC -GGCACAA-T ROGA GTCCTCTT-- GGGCCTTGTG AATGTGTTGT TGGCG-CAAT AACAACCCCC -GGCACAA-C COMY GTCCTCTT-- GGGCTTCGTG AATGTGTCGT TGGCG-CAAT AACAACCCCC -GGCACAA-C GLMA GTCCTATT-- GGGCTTTGTG AATGTGTTGT CGGCG-CAAT AACAACCCCC -GGCACAA-T LEMA GTCCTCTT-- GGGCCTTGTG AATGTGTCGT TGGCG-CAAT AACAACCCCC -GGCACAA-C MAPA GTCCTCTT-- GGTCCTTGTG AATGTGTCGT TGGCG-CAAT AACAACCCCC -GGCACAA-T SACH GTCCTTTT-G GGGCCTTGGG AGTGCGTCAT CGGCG-CAAT AACAACCCCC -GGCACAA-T ANPY GTTCTTTT-- GGACTTGGTA AATGTGTCAT TGACG-CAAT AACAACCCCC -GGCACAA-T ACMI GTTCTTTTTA GGACCTGGTG AATGTGTCAT TGACG-CAAT AACAACCCCC -GGCACAA-T PHHO GTTTTTTT-- GGACCTTGTG AGTGTTCCAT TGGCAGCAAT AACAACCCCC AGGCACAA-T HYPS GTTCTTTC-- GGACCATTTG GGAGTATCGT TGGCG-TAAT AACAACCCCC -GGCACAA-T LERE GTCCTTTT-- GGGCTGTGTG AGTGTGTCAT TGGCG-CAAT AACAACCCCC -GGCACAA-T PRHI GTTCTTTT-- GGACCATTTG AGTGTGTCAT TGGCG-CAAT AACAACCCCC -GGCACAA-T LEAL GTTCTTTT-- GGACCATTTG AGTGTGTCGT TGGCG-CAAT AACAACCCCC -GGCACAA-T TRPE GTTCTTTT-- GGACTTGGTG AATGTGTCAT TGGCG-CAAT AACAACCCCC -GGCACAA-T MARE GTTCTTTT-- GGACATGGTG AATGTTTCAT TGGCG-CAAT AACAACCCCC -GGCACAA-T OTGL GTCCTTAAGT GACGCTTGTG AATGTGTCAT CGGCG-CAAT AACAACCCCC -GGCACAA-T LOAN GTCCCTTT-G GTCCCTTGTG AGTGT--CAT TGGCG-CAAT AACAACCCCC -GGCACAA-T CYAD GTTCTTTT-- GGACCTTGTG GATGCGTCGT TGGCG-CAAT AACAACCCCC -GGCACAA-T LESE GTTCTTTT-- TGACCTTGTG GATGTGTCGT CGGCG-CAAT AACAACCCCC -GGCACAA-T DECO GTCCTTTC-- GGACCTTGTG AATGTGTCAT TGGCG-CATT AACAACCCCC -GGCACAA-C AJPA GTCCTTTT-- GGACCTTGTG AATGTGTCAT TGGCG-CATT AACAACCCCC -GGCACAA-C ARAR GTCCTTTT-- GGACCTTGTG AATGTGTCAT TGGCG-CATT AACAACCCCC -GGCACAA-C NINI GTTCTTTT-- GGACCTTGTG AATGTGTCAT TGGCG-CAAT AACAACCCCC -GGCACAA-T ARCA -TTCTTTT-- GGACCTTGTG A??????CGT TGGCG-CATT AACAACCCCC -GGCACAA-T UR AN GTCTTTCG-G GGACGCTGTG GACGTGTCAT TGGCA-TAAT AACAAACCCC -GGCACAA-T

------..................................................................

November 19971 FRANCISCO-ORTEGAET AL.-ORIGIN OFARGYRANTHEMUM

APPENDIX1. Continued.  
    APPENDIX1. Continued.
    ........................................................................
    190 200 210 220 230 240
    ARFO GCGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTTTCA TGTTT-GCCC CGTTCGCGGT
    HEVI GCGTGCCAAG GAACACTAAA -CTTAAGAAG GCTTGTTTCA TGTTT-GCGC CGTTTGCGGT
    ISCA GCGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTTTCA TGTTT-GCCC CGTTCGCGGT
    CHCO GCGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTTTCA TGTTT-GCCC CGTTCGCGGT
    CHSE GCGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTTTCA TGTTT-GCCC CGTTCGCGGT
    CLAR GTGTGCCAAG GAAAACTAAA --GTAAGATG GCTTGTCTCA TGTA--GTCC CGTTCGCGGT
    TAW GTGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTAGTTTTA TGTT--GCCC CGTTCGCGGT
    TAMI GTGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTCTCG TGTT--GCCC CGTTTGCGGT
    ANCR GCGTGCCAAG GAAAACTAGA -CATAAGAAG GCTTGTTTTG TGTTC-GCCC CGCTCGCGGT
    ANAR GTGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTTTAA TGTT--GCCC CGTTCGCGGT
    ANTI GTGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTCTTA TGAT--GCCC CGTTCGCGGT
    ANMA GCGTGCCAAG GAAAACTAAA -CATAAGAAG GCTTGTTTTG TGTC--GCCC CGTTCGCGGT
    ROGA GTGTGCCAAG GAAAACTAAA -CTTGAGAAG GCTCGTCTCA TTTT--GCCC CGTTCGCGGT
    COW GTGTGCCAAG GAAAACTAAA -CTTGAGAAG GCTTGTTTCA TGGT--GCCC CGTTCGCGGT
    GLMA GTGTGCCAAG GAAAACTAAA -CTTGAGAAG GCTTGTCTTA TGTTGTGCAC CGTTCGCGGT
    LEMA GTGTGCCAAG GAAAACTAAA -CTTGAGAAG GCTTGTCTCA TGTT--GCCC CGTTCGCGGT
    MAPA GTGTGCCAAG GAAAACTAAA -CTTGAGAAG GCTTGTCTCG TGKT--GCCC CGTTCGCGGT
    SACH GTGTGCCAAG GAAAACTAAA -CGTAAGAAG GCTTGTTTCA TGTTT-GCCC CGTTCGCGGT
    ANPY GTGTGCCAAG GAAAACTAAA -CTTGAGAAT GCTTGTTTCA TGTT--GCCC CGTTCGCGGT
    ACMI GTGTGCCAAG GAAAACTAAA -CTTGAGAAG GCTTGTTTCA TGTT--GCCC CGTY---GGT
    PHHO GTGTGCCAAG GNQACTAAA -CTTAAGAAG GCTTGTTTCA TGTT---CCT CGTTTGCGGT
    HYPS GTGTGCCAAG GAAAACTAAA -CTTTAGAAT GCTTGTTGCA TGTT--GCCC CGTTCGCGGT
    LERE GCGTGCCAAG GAAAACAAAA -CGAAAGAAG GCTTGTTTCA TGTC--GCCC CGTTCGCGGT
    PRHI GTGTGCCAAG GAAAACTAAA -CTTGAGAAT GCTCGTTGCA TGAT--GGCC CGTTCGCGGT
    LEAL GTGTGCCAAG GAAAACTAAA -CTTGAGAAT GCTTGTTGCA TGTT--GCCC CGTTCGCGGT
    TRPE GTGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTTGTTTTA TGTT--GCCC CGTTTGCGGT
    MARE GTGTGCCAAG GAAAACTAAA -CTTGAGAAG GCTTGTTTCA TGTT--GCCC CGTTCGCGGT
    OTGL GTGTGCCAAG GAAAACAAAA -CTTGAGAAG GCTTGTTTCA TGTTT-GCCC CGTTCGCGGT
    LOAN GCGTGCCAAG GAAAACAAAA ACTTAAGAAG GCTTGTTCCA TGTT--GCCC CGTTCGCGGT
    CYAD GTGTGCCAAG GAAAACTAAA -CTTAAGAAG GCTCGTTTCA TGTT--GCCC CGTTCGCGGT
    LESE GTGTGCCAAG GAAAACTAAA -CTCAAGAAG GCTCGTTTCA TGTT--GCCC CGTTCGCGGT
    DECO GTGTGCCAAG GAAAACTAAA -CTCAAGAAG GCTCGTTTCA TGAT-GCCCC CGTTCGCGGT
    AJPA GTGTGCCAAG GAAAACTAAA -CTCAAGAAG GCTCGTTTCA TGAT-GCCCC CGTTCGCGGT
    ARAR GTGTGCCAAG GAAAACTAAA -CTCAAGAAG GCTCGTTTCA TGAT-GCCCC CGTTCGCGGT
    NINI GTGTGCCAAG GAAAACTAAA -CTCAAGAAG GCTCGTTTCA TGTT--GCCC CGTTCGCGGT
    ARCA GTGTGCCAAG GAAAACTAAA -CTCTAGAAG GCTMGTTTCA TGTT-GCACC CGTTCGCGGT
    URAN GTGTGCCAAG GAAATCTAAA -CTTAAGAAG GCCCGTCTCA TGTT?GCCCC CGTTCGCGGT
    ........................................................................
    APPENDIX1. Continued.
    ->ITS2
    250 260 270 280 290 300
    ARFO GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCTTCGCCC CCCGCAAATC TAT-------
    HEVI GTGCTCATGG GACGTGGCTT CTTTATAAAT CGTC--GC-C CCCTCAAATC TAT-------
    ISCA GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCT-CGCCC CCCGCAAATC TAT-------
    CHCO GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCT-CGCCC CCCGCAAATC TAT-------
    CHSE GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCT-CGCCC CCCACAAATA TTT-------
    CLAR GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCMACAAATC TTT-------
    TAW GTGCTCATGA GACGTGGCTT CTTTATAAAT CGCGTCGCCC CCAACAAATC TTT-------
    TAMI GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTYGCCC CTAACAAATC TTAT------
    ANCR GTGCTCATGA GACGTGGCTT CTTTATAAAT CGCGTCGCCC CCAACAAATT ACT-------
    ANAR GTGCTCATGA GACGTGGCTT CTTTATAAAT CGCGTCGCCC CCAACAAATC TAT-------
    ANTI GTGCTCATAG GACGTGGCTT CTTTATAAAT CGCGTCGCCC C?AACAAATC TTT-------
    ANMA GTGCTCATGA GACGTGGCTT CTTTATAAAT CGC?TCGCCC C?AACAAATT ACT-------
    ROGA GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CC?ACAAATA TTT-------
    COW GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAAGTA TTT-------
    GLMA GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAATTA TTT-------
    LEMA GTGCTCATGG GATGTGGCTT CTTTATAAAT CGC?TCGCCC CCAACAATTA TTT-------
    blAPA GTGCTCATGG GATGCGGCTT CTTTATAAAT CGCGTCGCCC CCCGCAATTA TTT-------
    SACH GTGCTCATGG GATGTGGCTT CTTTATAAAT CG--TCGCCC CCCGCAAATC TAT-------
    ANPY GTGCTCCTGG GATTTGGCTT CTTTATAAAT CGCGTCGCCC CCAACAAATC TTT-------
    ACMI GYGCTCATGG AATGTGGCTT CTTTTTAAAT CGCGTCGCCC CCAACAAATA TCT-------
    PHHO GTGCTCATGA GACTTGGCTT CTTTATAAAT CGCGTCGCCC CCCGCAAATC TTT-------
    HYPS GTGCTCGTGT GACATGGCTT CTTTTTAAAT CGCGTCGCCC CCCGCAAAAC TTT-------
    LERE GTGCGCGTGG AATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCGCAAATA TAT-------
    PRHI GTCCTCATGT GACGTGGCTT CTTATTAAAT CGCGTCGCCC CCCGCAAAAC TTT-------
    LEAL GTGCTCGTGT GACACGGCTT CTTTTTAAAT CGCGTCGCCC CC?GCAAAAC ATT-------
    TRPE GTGCTCATGG AACGTGGCTT CTTTATAAAT CGCATCGCCT CCAACAAATA TTT-------
    MARE GTGCTCATGG GATGTGGCTT CTTTATAAAT CCCGTCGCCC CCAACAAACT AT--------
    OTGL GTGCTCATGG GAYGTGGCTT CTTTATAAAT CGCGTCGCCC CC?ACAAGTA TTT-------
    LOAN GTGCTCATGA GACGCGACTT CTTTTTAAAT CGCGTCGCCC CCCACCAAAT CTTT------
    CYAD GTGCTCGTGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACATTTC TCTGTCA-AG
    LESE GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAAATC TCCGTAA-AG
    DECO GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAATTC TCCGTAA-AG
    AJPA GTGCTCATGG GACGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAATTC TCCGTAA-AG
    ARAR GTGCTCATGG GACGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAATTC TCCGTAA-AG
    NINI GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAATTC TCCGTAA-AG
    ARCA GTGCTCATGG GATGTGGCTT CTTTATAAAT CGCGTCGCCC CCCACAGTTC TCCGCAA-AG
    UR AN GTGCTCGTGG GATGTGGCGT CTTTGTAAAT CGCGTCGCCC CCCGAATCCG CTCCTTTTAA

------..................................................................

APPENDIX1. Continued.  
    APPENDIX 1. Continued.
    ........................................................................
    310 320 330 340 350 360
    ARFO -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC - - - - - - - - - -
    HEVI -GTT-GGGGG CGGAT--ATT GGTYTCCCGT GATATT-GTC G-CGGTTGGC - - - - - - - - - -
    ISCA -GCT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC ---- -- -- --
    CHCO -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC -- --- - -- --
    CHSE -GAT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC -- -- -- ----
    CLAR -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC ---- -- -- --
    TAW -GTT-CGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGT G-TGGTTGGC - - - - - - - - - -
    TAMI -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAC-GGT G-TGGTTGGC - - - - - - - - - -
    ANCR -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGT G-TGGTTGGC - - - - - - - - --
    ANAR -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGT G-TGGTTGGC - - - - - - - - - -
    ANTI -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCCCAW-GGT G-TGGTTGGC - - - - - - - - - -
    ANMA -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGT G-TGGTTGGC - - - - - - - - - -
    ROGA -GTA-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC - - - - - - - - - -
    corn -GTC-GAGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC GATGGTTGGC -- -- -- ----
    GLMA -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-CGC T-CGGTTGGC - - - - - - - - --
    LEMA -GTTAGGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC - - - - - - - - - -
    MAPA -GTC-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC - - - --- - - - -
    SACH -GTC-GGGGG C--AT--ATT GGTCTCCCGT GCTCAC-G-C G-TGGTTGGC - -- -- ---- -
    wpy -GTT-GGGAG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC ----------
    ACMI -GTT-GGGG- CGGAT--ATT GGTCTCCCGT GCCCAT-GGT G-TGGTTGGC - - - - - - - - - -
    PHHO -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGT G-TGGTTGGC - - - - - - - - - -
    ~yps -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GTCACC-GGC G-CGGCTGTC -- -- -- -- - -
    LERE -GTG-GGGGG CGGATTAATT GGTCTCCCGT GATATT-GTC G-TGGTTGGC - - - - - - - - - -
    PRHI -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GTCACT-GGT G-CGGTTGTC - -- -- -- -- -
    LEAL -GTT-GGGGG CGGAT--ATT GGTCTCCCGT GTCACC-GGC G-CGGTTGTT ---- - - - - - -
    TRPE -GTT-GGGGA CGGAT--ATT GGTTTCCCGT GCTCAT-GGT G-TGGTTGGC -- - -- - - - - -
    MARE -GTT-GGGG- CGGAT--ATT GGTCTCCCGT GCTTAT-GGC G-TGGTTGGC - - - - - - - - - -
    OTGL -GTC-GGGG- CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC - - - - - - - - - -
    LOAN -GTTGGGGGG CGGAT--ATT GGTCTCCCGT GCTATT-GGC G-TGGTTGGC - - - - - - - - - -
    CYAD AGA-CTTGTG TTTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAC-GGT G-TGGTTGGC
    LESE GGAATTTGTG TTTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC
    DECO GGAACATGTG TTTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC
    AJPA GGAACATGTG TTTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC
    ARAR GGAACATGTG TTTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC
    NINI GGAACTTGTG TTTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCAT-GGC G-TGGTTGGC
    ARCA GGAACTGTTG TTTT-GGGGG CGGAT--ATT GGTCTCCCGT GCTCATTGGC G-TGGTTGGC
    URAN AGGATGCGAT GTCT-GGGGG CGGAT--ATT GGTCTCCCGT GCTTAC-GGC G-TGGTTGGC
    -x1. Continued.
    370 380 390 400 410 420
    ARFO CAAAATA-GG AGTCCCTX- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    HEVI CAAAATACGT AGTCCTTTC- GATGGACGCA TAAACTAGTG GTGGTCGTAA AAACCCTCGT
    ISCA CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    CHCO CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    CHSE CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    CLAR CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    TAW CAAAATA-GG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    TAM1 CAAAATAAGG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    ANCR CAAAATA-GG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGATCTAA AAACCCTCGT
    ANAR CAAAATA-GG AGTCCCTTC- GATGGACACA CGAACTAGTG GTGGACGTAA AAACCCTCGA
    ANT I CAAAATA-GG AGTTCCTTY- GATGGACGCA CGAACTGGTG GTGGATGTAA AAACCCTCGT
    ANMA CAAAATA-GG AGTCCCTTC- GATGGATGCA CGAACTAGTG GTGGATCTAA AAACCCTCGT
    ROGA CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    COW CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    GLMA CGAAATA-AG AGTCCCTTC- GTTGGAAGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    LEMA CGAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    MAPA CAAAATA-GG AGTCCCTTCC GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    SACH CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    ANPY CAAAATA-GG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    ACMI CAAAATA-AG AGTCCCTTC- GATGGACACA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    PHHO CAAAATA-AG AGTCCCTTC- GACGGACGCA CTAACTAGTG GTGGTCGTAA AAACCCTCGT
    HYPS CAAAATA-TG TTTTCCTTC- -ATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    LERE CAAAATA-TG GTTTCCTCT- GATGGATGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    PRHI CAAAATA-TG TTTTCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    LEAL CAAAATA-TG TTTTCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    TRPE CAAAATA-AG AGTCCCTTC- GATGGACGCA TGAACTAGTG GTGGTCGTAA AAACCCTCGT
    MARE CAAAATA-GG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    OTGL CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    LOAN CAAAATA-GG AGTCCCTTC- GATGGACGCA CAAACAAGTG GTGGTCGTAA AAACCCTCGT
    CYAD CAAAATA-GG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    LESE CGAAATA-GG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    DECO CGAAATA-GG AGTCCTTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    AJPA CGAAATA-GG AGTCCTTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    ARAR CGAAATA-GG AGTCCTTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    NINI CGAAATA-GG AGTCCTTTC- GATGGACGCA CAAACTAGTG GTGGTCGTAA AAACCCTCGT
    ARCA CGAAATA-GG AGTCCCTTC- GATGGACGCA CGAACTAGTG GTGGTCGTAA AAACCCTCGT
    UR AN CAAAATA-GG AGTCCCCTTC GATGGACGCA CGA-CTAGTG GTGGTCGTAA AAACCCTCGT
    -----

November 19971 FRANCISCO-ORTEGA OF ARGYR~~NTHEMG'M 1613

ET AL.-ORIGIN

APPENDIX1. Continued.  
    APPENDIX 1. C0nt hued .
    ........................................................................
    430 440 450 460 470 480
    ARFO TCTTTGTTTT GTGTC-GTTC GTCGCAATGG AACA-CTCTA TAAAAACCCC -AATGTGTTG
    HEVI TCTTTGTTTA CAGTC-GTCT CTCGCAAGGG AACA-GTATA TCAAAA-CCC -AACGTGTTG
    ISCA TCTTTGTTTT GTGTC-GTTC GTTGCAATGG AACA-CTCTA TAAAAACCCC -AATGTGTTG
    CHCO TCTTTGTTTT GTGTC-GTTC GTTGCAATGG AACA-CTCTA TAAAAAACCC -AATGTGTTG
    CHSE TCTTTGTTT- GTGTC-GTTC GTTGCAAKGG AACA-CTCTA TAAAAACCCC -AATGTGTTG
    CLAR TCTTTGTTT- GTGTC-GCAC GTCGATAGGA AACA-CTCTC TAAATACCCC -AATGTGTTG
    TAW TCTTTGTTCT GTGTT----A GTCGCAAGGA AAAA-CTCTT CAAATACCCC -AATGTGTTA
    TAMI TCTTTGTTCT GTGCT-GTTT GTCGCAAGGG AAAA-CTCTT CAAATACCCT -AACGTGTTG
    ANCR TCTTTGTTTC GTGTTAGTTA GTCGCAAGAA TAAA-CTCTC TGAATACCCC CAATGTGTTG
    ANAR TCTTTGTTCT GTGTT----A GTCGCAAGGA AAAA-CTCTT CGAATACCCC -AACGTGTTG
    ANTI TCTTTGTTCT GTGTT----A GTCGCAAGGA AAAA-CTCTT CAAATACCCC -AATGTGTTG
    ANMA TCTTTGTTCT GTGTT----T GTCGCAAGGA TAAA-CTCTT TGAATACCCC -AATGTGTTG
    ROGA TCTTTGTTTT GTGCT-GACG GTCGCAAGGA AACA-CTCTT CAAATACCCC -AATGTGTTG
    COW TCTTTGTTTT GTGCT-GATA GT--GAAGGA AACA-CTCTT AAATAACCCC -AATGTGTTG
    GLMA TCTTTGTTTT GTGCC-GACA GCTGCAAGGG TACA-CTCTT CAAAAC-CCC -AATGTGTTG
    LEMA TCTTTGTTTT GTGCC-GACA TTTGCAAGGA AACA-CTCTT TTCTTTACCT -AATGYGTTG
    MAPA TCTTTGTTTT GTGCC-GAAA GTCGGAAGGA AACA-CTCTT CAAATACCCC -AACGTGTTG
    SACH TCTTTGTTTT GTGTC-GTAC GTCGCTAGGA AACA-CTCTC TAAATACCCC -AACGTGTTG
    ANPY TCTTTGTTCT GCGTT----A GTCGTAAGGA AAAG-CTCTT CAAATACCCC -AATGTGTTG
    ACMI TCTTTGTTCT GTGTT----A GTCGCAAGGA AAAA-CTCTT CAAATACCCC -AACGYGTTG
    PHHO TCTTTGTTAT GTGTT-GTTA GTTGCAAGGA AACA-CTCTT CAAATACCCC -AACGTGTTG
    HYPS TCTTTGTTT- GTGT---GGT ATCGCAAGGG ACT--CTT-- ---AAACCCT -AATGCGTTG
    LERE TCTTTGTTTT GTGCT-GTAT CTCTCATTAG GAA--CTC-- ---TAATCCC -AATGTGTTG
    PRHI TCTTTGTTT- GTGTT-TGGT ATCGCAAGGG ACT--CTT-- ---AAACCCT -AATGTGTTG
    LEAL TCTTTGTTT- GTGTT-TGGT ATCGCAAGGG ACT--CTTT- ---AAACCCT -AATGTGTTG
    TRPE TCTTTGTTCT GTGTT----A GTCGTAAGGA AAAA-CTCTC TAAATACCC- -AATGTGTTG
    MARE TCTTTGTTTT GTGTC-GTCG GTCGCAAGGA TAAG-CTCTS TAAAAACCCC -AATGTGTTG
    OTGL TCTTTGTTTT GTGCC-GATA GTCGCAAGGA AACA-CTCTT CAAATAACCC -AATGTGTTG
    LOAN TCTTTGTTTT GTGCC-GTCT CTCGCAAGGG AACA-CTCTT CAAGTAACCC -AGCGCATCG
    CYAD CTTTTGTTCT GTGCC-GTTA CTCGCAAGGG AAAG-CTCTT CAAGTACCCT -AACGTGTTG
    LESE CTTTTGTTTC GTGCT-GTTA GTCGCAAGGG AAAG-CTCTT TAAAAAACCC -AATGTGTCG
    DECO CTTTTGTTTC GTGCT-GTTG CTCGCAAGGT AAA--CTCTT TAAAAACCCC -AATGTGTCG
    AJPA CTTTTGTTTC GTGCT-GTTG CTCGCAAGGT AAA--CACTT TAAAAACCCC -AATGTGTCG
    ARAR CTTTTGTTTC GTGCT-GTTG CTCGCAAGGT AAA--CTCTT TAAAAACCCC -AATGTGTCG
    NINI CTTTTGTTTC GTGCT-GTTA GTCGCAAGGG AAA--CTCTT TAAAAACCCC -AATGTGTCG
    ARCA CTTTTGTTGT GTGCC-GTTA GTCGCAAGGG AAA--CTCTT -AAAAACCCC -AACGTGTCG
    URAN CTTTTGTCGT GCGTT-GTTA GTCGCAAGGG AACAGCTCTT CAAGTACCCC --AATCGTCG
    ........................................................................
    APPENDIX 1. Continued.
    ................................
    490 500
    ARFO
TCTTAG-GAT GACGCTTCGA CCG
    HEVI
TCTTAG-GAT GACGCTTCGA CCG
    ISCA
TCTTAG-GAT GACGCTTCGA CCG
    CHCO
TCTTAG-GAT GACGCTTCGA CCG
    CHSE
TCTTAG-GAT RACGCTTCGA CCG
    CLAR
TCTTAG-GAT GACGCTTCGA CCG
    TAW
TCTTAG-GAT GACGCTTCGA CCG
    TAMI
TCTTTG-GAT GACGCTTCGA CCG
    ANCR
TCTTCT-GAT GACGCTTCGA CCG
    ANAR
TCTTCG-GAT GACGCTTCGA CCG
    ANTI
TCTTCG-GAT GACGCTTCGA CCG
    ANMA TCTTCT-GAT GACGCTTCGA CCG
    ROGA
TCTTAG-GAC GATGCTTCGA CCG
    COMY
TCTT?G-AAC GACGCTTCGA CCG
    GLMA
TCTTAG-GAT GACTCTTCGA CCG
    LEMA
TCTTAG-GAC GATGCTTCGA CCG
    MAPA
TCTTAG-GAC GGCGCTTCGA CCG
    SACH
TCTTAG-GAT GACGCTTCGA CCG
    ANPY
TCTTAG-GAC GATGCTTCGA CCG
    ACMI
TCTTAG-GAT GACGCTTCGA CCG
    PHHO
TCTTAG-GAT GATGCTTCGA CCG
    HYPS
TCCTTG-GAT GACGCTTCGA CCG
    LERE
TTTTAG-GAT GACGCTTCGA CCG
    PRHI
TCTTTG-GAT GACGCTTCGA CCG
    LEAL
TCTTTG-GAT GACGCTTCGA CCG
    TRPE
TCTTTG-GAT GATGCTTCGA CCG
    MARE
TCTTAG-GAT GACGCTTCGA CCG
    OTGL
TCTTAG-GAT GACGCTTCGA CCG
    LOAN
TCTAAG-GAT GATGCTTCGA CCG
    CYAD
TCTTTC-GAT GACGCTTCGA CAG
    LESE
TCTTTTTGAC GACGCTTCGA CCG
    DECO
TCTCTT-GAC GACGCTTCGA CCG
    AJPA
TCTCTT-GAC GACGCTTCGA CCG
    ARAR
TCTCTT-GAC GACGCTTCGA CCG
    NINI
TCTTTT-GAC GACGCTTCGA CCG
    ARCA
TCTCTT-GAC GAGGCTTCGA CCG
    URAN
TCTATT-GGC GACGCTTCGA CCG
    ................................
Comments
  • Recommend Us