INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter 6ce, wdiile others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Infoimation Company 300 North Zed) Road, Ann Aibor MI 48106-1346 USA 313/761-4700 800/521-0600
ORIGIN AND EVOLUTION OF WOODY SONCHUS
AND FIVE CLOSELY RELATED GENERA (ASTERACEAE: LACTÜCEAE)
IN MACARONESIA
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Seung-Chul Kim, M.S.
*****
The Ohio State University
1997
Dissertation Committee:
Professor Daniel J. Crawford, Advisor Approved by
Professor Andrea Wolfe
Professor Keith Davis
Advisor
Plant Biology Department UMI Number: 9 801723
UMI Microform 9801723 Copyright 1997, by UMI Company. AU rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 ABSTRACT
Woody species of Sonchus (i.e., subgenus
Dendrosonchus) , which are composed of 19 species, and five
related genera (i.e., Babcockia. Taeckholmia. Sventenia.
Lactucosonchus, and Prenanthes pendula) are restricted
primarily to the archipelago of the Canaries in the
Macaronesian phytogeographical region. These plants
(referred to as woody Sonchus alliance) display great
morphological, ecological, and anatomical diversity, and
have been regarded as an outstanding example of adaptive
radiation in angiosperms. Despite this diversity, the
occurrence of natural and artificial interspecific and
intergeneric hybrids suggests genetic cohesiveness within
the alliance.
Sequences from the internal transcribed spacer region
(ITS) of nuclear ribosomal DNA (nrDNA) suggest that, despite the extensive morphological and ecological diversity of the plants and the geographical proximity of the Macaronesian
islands to a continental source area, the entire alliance was derived from a single colonization event. The ITS sequence data also suggest that the woody members of Sonchus were derived from ancestors similar to allied genera now
ii present on the Canary Islands, and that the alliance probably occurred in the Canary Islands during the late
Miocene or early Pliocene. The ITS phylogeny also provides evidence that inter-island colonization events to similar ecological habitats and intra-island differentiation have played important roles in the evolution of the alliance.
An enzyme electrophoretic study was also conducted to assess genetic diversity within and divergence among species of the alliance. Results show high genetic identities among the alliance, and this further supports the genetic cohesiveness of the alliance and its single origin on the
Macaronesian Islands. The results also suggest early colonization, radiation, and divergence of the woody Sonchus alliance on older islands followed by subsequent colonization to younger islands. This study indicates that lineage sorting played a role in divergence among species and suggests that time is a factor for divergence at allozyme loci. Both ITS and allozyme data suggest that the early divergence and rapid radiation of the alliance apparently took place during the Late Tertiary on either
Gran Canaria or Tenerife of the Canaries.
Ill Dedicated to my parents and wife, Min-ju
IV ACKNOWLEDGMENTS
I am grateful to my advisor. Dr. Daniel J. Crawford, for continual advice, support, encouragement, and discussions during the course of this study. Special thanks go to two crazy Spainards, Drs. Javier Francisco-Ortega and Arnoldo
Santos-Guerra, for their continual encouragement, endless support, and introducing me to the woody sow-thistle from the Canary Islands during the course of this study. Some of the greatest moments of my life were spent with Javier and
Arnoldo during my trips to the Canary Islands and I am grateful for their friendship. I wish to thank colleagues and friends in Dr. Crawford's laboratory: Mesfin Tadesse,
Mary Beth Cosner, Betsy Esselman, and Tao Sang. I would like to thank Dr. Tod Stuessy for his helpful comments and suggestions during the early stages of this study. Special thanks also to Dr. Robert K. Jansen for his support, encouragement, and help in publishing some of the results. I am indebted to people from the Canaries, United States,
Africa, Europe, and New Zealand, who helped in providing plant materials and DNAs for this study (see acknowledgments from each chapter) . This work couldn't have been
V accomplished without their help. I also thank Bette
Hellinger and Vicki Payne for their great help in various ways during five years of study in the Plant Biology department at OSÜ. I especially wish to thank my wife, Min- ju, for her love and strong, endless, unconditional support, and encouragement during five years of this study. I also thank my parents who devoted their lives to their children and gave me a chance to accomplish this work. I dedicate this dissertation and my Ph. D. degree to my wife and my father on his 70th birthday.
VI VITA
Januany 06, 1968 ...... Born - Seoul, Korea
1989 ...... B.S. Sung Kyun Kwan University, Seoul, Korea
1990-1992 ...... M.S. Kent State University, Kent, Ohio Garduate Teaching Associate
19 9 2-present ...... Graduate Teaching and Research Associate, The Ohio State University, Columbus, Ohio
PUBLICATIONS
Kim, S.-C., S. H. Graham, and A. Graham. 1994. Palynology and pollen dimorphism in the genus Laaerstroemia (Lythraceae). Grana 33: 1-20.
Sang, T., D. J. Crawford, S.-C. Kim, and T. F. Stuessy. 1994. Radiation of the endemic genus Dendroseris (Astearceae) on the Juan Fernandez Islands: Evidence from sequencing of the ITS region of nuclear ribosomal DNA. American Journal of Botany 81(11): 1494-1501.
Kim, S.-C., D. J. Crawford, J. Francisco-Ortega, and A. Satos-Guerra. 1996. The common origin for woody Sonchus and five related genera in the Macaronesian Islands: Molecular evidence for extensive radiation. Proceedings of the National Academy of Sciences, USA 93: 7743-7748.
Kim, S.-C., D. J. Crawford, and R. K. Jansen. 1996. Phylogenetic relationships among the genera of the subtribe Sonchinae (Lactuceae: Asteraceae): Evidence from ITS sequences. Systematic Botany 21(3): 417-432.
Vll Lee, N.-S., T. Sang, T. F. Stuessy, S.-H. Yeau, and S.-C. Kim. 1996. Molecular divergence between disjunct taxa in eastern Asia and eastern North America. American Journal of Botany 83(10): 1373-1378.
Abstracts presented to Meetings
Kim, S. -C., D. J. Crawford, J. Francisco-Ortega, and A. Santos-Guerra. 1994. Preliminary molecular data and the evolution of Sonchus (Asteraceae: Lactuceae) in the Canary Islands. Amer. J. Hot. (suppl.) 81: 165.
Sang, T., D. J. Crawford, S. -C. Kim, and T. F. Stuessy. 1994. Radiation of the endemic genus Dendroseris (Asteraceae) on the Juan Fernandez Islands: Evidence from sequecning of the ITS region of nuclear ribosomal DNA. Amer. J. Hot. (suppl.) 81: 184.
Crawford, D. J., T. Sang, S. -C. Kim, and T. F. Stuessy. 1994. The origin and evolution of Dendroseris (Lactuceae) on the Juan Fernandez Islands: insights from DNA. COMPOSITAE. Internation Compositae conference, Kew, England.
Kim, S. -C., D. J. Crawford, J. Francisco-Ortega, and A. Santos-Guerra. 1995. Phylogenetic relationships in subtribe Sonchinae (Asteraceae) based on ribosomal DNA sequences from the ITS region. Amer. J. Hot. (suppl.) 82: 141.
Kim, S. -C., D. J. Crawford, J. Francisco-Ortega, and A. Santos-Guerra. 1995. The origin and radiation of Sonchus (Lactuceae: Asteraceae) in the Macaronesian islands: Evidence from ITS sequences. Amer. J. Hot. (suppl.) 82: 141.
Mesfin, T., D. J. Crawford, and S. -C. Kim. 1995. A cladistic analysis of morphological features in Bidens L. and Coreopsis L. (Compositae-Heliantheae) with notes on generic delimitation and systematics. Amer. J. Bot. (suppl.) 82: 166.
Kim, S. -C., D. J. Crawford, and R. K. Jansen. 1996. Phylogeny of the subtribe Sonchinae (Asteraceae: Lactuceae): additional information from the noncoding region of cpDNA. Amer. J. Bot. (suppl.) 83: 168.
Kim, S. -C., D. J. Crawford, E. J. Esselman, and M. Tadesse. 1996. Phylogenetic relationships in Coreopsis (Asteraceae) based on the ITS sequences of nrDNA. Amer. J. Bot. (suppl.) 83: 167-168.
Vlll Kim, S. -C., D. J. Crawford, J. Francisco-Ortega, and A. Santos-Guerra. 1996. The origin and evolution of Sonchus and allied genera in the Macaronesia: Molecular evidence for rapid radiation. Symposium "Fauna and flora of the Atlantic islands". Las Palmas de Gran Canaria, Canary Islands, Spain, p.55.
Francisco-Ortega, J., J. Fuertes-Aguilar, A. Santos-Guerra, S. -C. Kim, D. J. Crawford, and R. K. Jansen. 1996. Morphological and molecular evidence for the evolution of Crambe L. (Brassicaceae) in Macaronesia. Symposium "Fauna and flora of the Atlantic islands". Las Palmas de Gran Canaria, Canary Islands, Spain, p.41.
Kim S. -C. , D. J. Crawford, J. Francisco-Ortega, A. Santos- Guerra, and M.-J. Kim. 1997. Adaptive radiation and genetic differentiation in the woody Sonchus alliance (Asteraceae: Lactuceae) in the Canary Islands. Amer. J. Bot. (suppl.)
Francisco-Ortega, J., A. Santos-Guerra, J. Fuertes-Aguilar, S.-C. Kim, D. J. Crawford, and R. K. Jansen. 1997. Molecular evidence for the evolution of Crambe sect. Dendrocrambe (Brassicaceae) in the Macaronesian islands. Amer. J. Bot. (suppl.)
FIELDS OF STUDY
Major Field: Plant Biology
IX TABLE OF CONTENTS
ABSTRACT ...... Ü
DEDICATION ...... iv
ACKNOWLEDGMENTS ...... V
VITA ...... vii
LIST OF TABLES ...... xii
LIST OF FIGURES ...... xiii
CHAPTER PAGE
1. INTRODUCTION ...... 1
LITERATURE CITED ...... 11
2. PHYLOGENETIC RELATIONSHIPS AMONG THE GENERA OF THE SUBTRIBE SONCHINAE (ASTERACEAE) : EVIDENCE FROM ITS SEQUENCES (Published in Systematic Botany, 1996, 21(3): 417-432, co-authored with D. J. Crawford and R. K. Jansen) ...... 20
ABSTRACT ...... 20 INTRODUCTION...... 21 MATERIALS AND MTETHODS ...... 24 RESULTS ...... 27 DISCUSSION ...... 29 LITERATURE CITED ...... 37
3. A COMMON ORIGIN FOR WOODY SONCHUS AND FIVE RELATED GENERA IN THE MACARONESIAN ISLANDS: MOLECULAR EVIDENCE FOR EXTENSIVE RADIATION (Published in Proceedings of the National Academy of Sciences, USA, 1996, 93: 7743-7748, co-authored with D. J. Crawford J. Francisco-Ortega, and A. Santos-Guerra) .... 43
X ABSTRACT ...... 43 INTRODUCTION ...... 44 MATERIALS AND METHODS ...... 48 RESULTS ...... 49 DISCUSSION ...... 50 LITERATURE CITED ...... 59
4. ADAPTIVE RADIATION AND GENETIC DIFFERENTIATION IN THE WOODY SONCHUS ALLLIANCE (ASTERACEAE: SONCHINAE) IN THE CANARY ISLANDS (manuscript in preparation for submission to Plant Systematics and Evolution; co-authored with D. J. Crawford J. Francisco-Ortega, and A. Santos-Guerra) ... 64
ABSTRACT ...... 64 INTRODUCTION ...... 66 MATERIALS AND METHODS ...... 69 RESULTS ...... 72 DISCUSSION ...... 75 LITERATURE CITED ...... 87
5. THE USE OF NON-CODING REGION OF CHLOROPLAST DNA IN PHYLOGENETIC STUDIES OF THE SUBTRIBE SONCHINAE (ASTERACEAE: LACTUCEAE) (manuscript in preparation for submission to Plant Systematics and Evolution; co-authored with Daniel J. Crawford and R. K. Jansen) ...... 96
ABSTRACT ...... 96 INTRODUCTION ...... 97 MATERIALS AND METHODS ...... 100 RESULTS ...... 102 DISCUSSION ...... 106 LITERATURE CITED ...... 116
LIST OF REFERENCES ...... 125
XI LIST OF TABLES
TABLE PAGE
1. Subtribal classifications of the Lactuceae .... 141
2. Sources of plants for ITS sequences ...... 142
3. Aligned ITS sequences from the subtribe Sonchinae and four outgroup genera ...... 145
4. Populations and species of the woody Sonchus alliance studied for isozymes ...... 152
5. Nei's genetic identités and distances for genera of the woody Sonchus alliance ...... 154
6. Gene diversity statistics for the woody Sonchus alliance ...... 155
7. Sources of plant materials for sequence comparisons for the chloroplast intergenic spacer between psbA and trnH ...... 156
8. Summary of parameters calculated for each of the four data sets ...... 158
9. Comparisons of average pairwise sequence divergence between psbA-trnH intergenic spacer and ITS for several genera of Sonchinae ...... 159
Xll LIST OF FIGURES
FIGURE PAGE
1. ITS sequence phylogeny of subtribe Sonchinae .. 160
2. (a) The dashed line enclose the biogeographical region of Macaronesia, which includes northwestern Africa and five archipelagos .... 162
(b) The canary archipelago and the oldest published radiometric ages from subaerial volcanics of each island in parentheses ...... 162
3. ITS phylogeny of subtribe Sonchinae (redrawn from Kim S.-C. et al. 1996) 164
4. Preferred ITS phylogeny of the woody Macaronesian Sonchus and their alliance ...... 166
5. The Canary archipelago and the woody Sonchus alliance taxa studied for isozymes ...... 168
6. UPGMA tree based on Nei's genetic distances for populations of the woody Sonchus alliance included in this study ...... 170
7. One of the 8952 equally parsimonious trees of the sbutribe Sonchinae based on intergenic spacer sequences between psbA and trnH of chloroplast DNA ...... 172
8. ITS sequence phylogeny of subtribe Sonchinae (modified from Kim S.-C. et al. 1996) 174
9. One of 360 equally parsimonious trees based on combined data sets ...... 176
Xlll CHAPTER 1
INTRODUCTION
Plants endemic to oceanic islands have attracted the attention of evolutionary biologists since the time of
Darwin (1859) and Wallace (1880) because they are often very distinct morphologically from possible ancestors in continental areas and congeners show greater morphological variation among themselves than do continental relatives
(Skottsberg 1956; Carlquist 1965, 1974; Bramwell 1975; Carr
1985; Helenurm and Ganders 1985; Sanders et al. 1987; Wagner et al. 1990; Mayer 1991). Although the reasons for the seemingly greater morphological diversity of plants on islands are not known, it may be the result of rapid adaptive radiation into the open habitats available after volcanic activity (Carlquist 1965, 1970, 1974). The interfertility of several endemic congeners and their high similarity at allozyme loci are concordant with relatively recent divergence relative to many continental species (Carr and Kyhos 1981; Lowrey and Crawford 1985; Crawford et al.
1987, 1992a; Mayer 1991). Several basic and important questions about island
plants center on determining their continental ancestors
(sister groups) and reconstructing their phylogeny on the
islands (see Stuessy et al. 1984; Sanders et al. 1987;
Wendel and Per ci val 1990; Baldwin et al. 1991). It is also
of interest to address how plants come to exist on remote
islands and why there are so many unique insular species as
contrasted with continental settings. As discussed by
Carlquist (1965), island plants can serve as model systems
for studying plant evolution because islands are often
considered as natural laboratories. Islands are relatively
small, isolated, and simple systems compared to most
continental situations. Furthermore, the age of an island
can be determined by potassium-argon dating (e.g., Stuessy
et al 1984) , a method that indirectly places a maximum age
limit on any endemic plants that evolved in situ.
There are certain limitations with using morphology to study the origin and radiation of insular endemics. First, the endemics may be so different from continental relatives that determining continental progenitors is difficult if not impossible with morphology. Second, among congeners, the adaptive value of morphological features may confound their use for reconstructing phylogeny. In other words, there is some difficulty in distinguishing features shared by common ancestry as opposed to parallelisms. Molecular markers, however, have proven valuable for examining the origin. monophyly, and evolution of island endemics (Baldwin et al.
1990, 1991; Baldwin 1992; Sang et al. 1994, 1995; Francisco-
Ortega et al. 1996b) . In addition, molecular sequences have
proven useful in the study of insular groups because,
contrasted with morphology, base substitutions may be
neutral or nearly so. Furthermore, molecular data sets tend
to be much larger than morphological ones, offering many
more characters for phylogenetic analysis (Soltis and Soltis
1995) .
Most studies of island plant groups have involved
several archipelagos from the Pacific Ocean, such as the
Hawaiian, Galapagos, and the Juan Fernandez Islands, which
are very isolated from possible source areas (Carr and Kyhos
1981; Lowrey and Crawford 1985; Helenurm and Ganders 1985;
Witter and Carr 1988; Wendel and Percival 1990; Mayer 1991).
Also, islands in the Pacific are rather young, although sea mounts may be older (Carlquist, 1970). Fewer investigations have been done on insular endemics from other areas of the world, where islands may be much closer to source areas and be of a greater variety of geological ages. These situations may provide the chance for multiple introductions from nearby source areas (i.e., groups are not monophyletic) and also some of the taxa could be much older than others depending on the different ages of the islands.
The phytogeographical region of Macaronesia is composed of five oceanic archipelagos (i.e., Azores, Desertas- Madeira, Selvagens, Canaries, and Cape Verde) , with an
enclave of Morocco (Sunding 1979). These islands exhibit a
broad range of variation both in ecology and geology.
Geological ages vary from about 0.8 Myr for El Hierro (The
Canaries) and 21 Myr for Fuerteventura (The Canaries) . There
are several ecological zones, including two major ones
distinguished because of the combination of latitudinal
gradients and influence of northeastern trade winds. The
zones under the effect of the trade winds are situated on
the northern slopes of the islands between 400 and 1200 m
altitude; these include the humid lowland scrub, laurel
forest, and heath belt. The arid zones are not directly
influenced by trade winds and they include the coastal
desert, arid lowland scrub, pine forest, and high altitude desert (Bramwell 1972). High habitat diversity and insular
isolation have been the main factors responsible for the rich flora of Macaronesia. At least 831 species and 40 genera are endemic to the region (Humphries 1979; Hansen and
Sunding 1993; La Roche and Rodriguez-Pinero 1994). There has been a long controversy over whether some of the woody
Macaronesian endemics are relict elements of a flora that extended along the Mediterranean basin during the Tertiary period or are recent derivatives from continental ancestors
(Carlquist 1962, 1974; Bramwell 1972, 1975, 1976; Sunding
1979) .
The Canary Islands, which are composed of seven islands, are of volcanic origin and form an approximately
linear island chain (McDougall and Schmincke 1976-77; Banda
et al. 1981). These islands are very close to possible
source areas; one of the eastern islands, Fuerteventura, is
only 100 km from the African coast (Carracedo 1994) . Also, the geological ages of these oceanic islands vary from 0.8 to 20 Myr (Adbel-Monem et al. 1971, 1972; Cantagrel et al.
1984; Ancochea et al. 1990; Coello et al. 1992). The flora of the Canary Islands is exceedingly rich in endemic species. Estimates on the size of the flora are between 1600 and 1700 species (Bramwell 1976) . About 470 of these are endemic to the islands, while another 110 or so are also found on other islands of the Macaronesian region (i.e.,
Macaronesian endemics). Generic endemism is a notable feature of the Canary Islands flora, 19 genera being found only in the Canarian Archipelago and 12 more shared with other islands of Macaronesia (Bramwell 1976).
The genus Sonchus belongs to the sunflower family,
Asteraceae (Stebbins 1953; Stebbins and Walter 1953; Jeffrey
1966), and has been studied extensively by Boulos, who divided it into the three subgenera, Sonchus. Dendrosonchus. and Oriaosonchus (Boulos 1960, 1967a, 1972, 1973, 1974a,
1974b; Roux and Boulos 1972; Pons and Boulos 1972) . Subgenus
Sonchus (21 species) includes widely distributed weedy species, while subg. Oricrosonchus (14 species) occurs exclusively in Africa. Subgenus Dendrosonchus. with 17 species and 10 subspecies, is composed of woody plants
endemic to the Canaries, Madeira Islands, and the Cape Verde
archipelago with one species in Western Morocco (Aldridge
1977, 1978) . Two genera, Babcockia (monotypic, the Canary
Islands) and Taeckholmia (seven species in the Canary
Islands), are segregate genera (Boulos 1965, 1967a, 1968,
1974b) . The two monotypic genera from the Canaries,
Sventenia and Lactucosonchus. are also closely related to
subg. Dendrosonchus (Aldridge 1975). All species of subg.
Dendrosonchus and several closely related genera in the
Canaries are diploid (n=9, 2n=18) and the evolution of subg.
Dendrosonchus in the Canary Islands is usually explained by
adaptive radition. Many authors, especially Boulos (1967a,
1967b, 1972, 1974a, 1974b) and Aldridge (1975, 1976a, 1976b,
1977, 1978, 1979), have made extensive contributions to the
taxonomy of woody, Macaronesian species of subg.
Dendrosonchus.
The origin of subg. Dendrosonchus in the Macaronesian
Islands and its relationships to the other subgenera are
controversial. Boulos (1960, 1967a) proposed that it evolved
from subg. Oricrosonchus (endemic in Africa) , which he
considered the most primitive group of Sonchus. According to
Boulos, subg. Sonchus was in turn derived subgenus and
evolved from the Macaronesian subg. Dendrosonchus. Thus, in his view, subg. Dendrosonchus is not a primitive group within Sonchus. Later, this hypothesis was supported by pollen morphology (Saad 1961) . Boulos's hypothesis about the origin of subg. Dendrosonchus in Macaronesia is in agreement with Darwin's hypothesis (1872) and with Carlquist's idea of paedomorphosos (1962).
In contrast, Aldridge (1975, 1978, 1979) suggested that subg. Dendrosonchus is primitive and that the subgenera
Oriaosonchus and Sonchus are derived from it. Her hypothesis is basically congruent with Takhtajan's view (1969) that the endemic flora of the Canary Islands is mostly quite old with the majority of taxa either paleoendemics or their derivatives. Later, Bramwell (1972) provided several lines of evidence to support this view. It is of interest to test these two conflicting hypotheses concerning the relationships of subg. Dendrosonchus. In addition, the answer to this question may provide a different perspective on the evolutionary history of plants in the Atlantic
Oceanic Islands and contribute to a better understanding of the origin and evolution of plants on oceanic islands.
Aldridge (1979) proposed two major evolutionary lineages in subg. Dendrosonchus. represented by sections
Atlanthus and Dendrosonchus. Aldridge did not think that these two types arose from a common ancestor within the
Canary Islands, but that they may have diverged during or before the Tertiary period before they migrated to the
Canaries. Thus, she hypothesized that her subg.
Dendrosonchus is not monophyletic. Boulos (1972, 1974a), however, suggested that his subg. Dendrosonchus is
monophyletic based on its woody habit and geographical
distribution. In a strict cladistic sense, however, subg.
Dendrosonchus of Boulos is paraphyletic because subg.
Sonchus was derived from it. More rigorous cladistic study will, however, answer the basic biological questions of whether subg. Dendrosonchus is the result of one or more
introductions to the Canaries. Although Aldridge's hypothesis (1979) was based on differnt lines of
information, she investigated only subg. Dendrosonchus species and did not include any members of subg.
Orioosonchus and subg. Sonchus. The various hypotheses about the origin and evolution of subg. Dendrosonchus need to be tested by more explicit methods using macromolecular data.
It is also of critical that members of subg. Oriaosonchus and subg. Sonchus be included in the analyses.
Enzyme electrophoresis has proven useful for studying generic diversity within and divergence among congeneric species endemic to islands (Lowrey and Crawford 1985;
Helenurm and Ganders 1985; Witter and Carr 1988; Crawford et al. 1987, 1992b). The general pattern (with some notable exceptions) that has emerged is one of low allozyme diveristy within endemics (see compilation in DeJoode and
Wendel 1992) and high identities among species. Recent enzynme electrophoretic studies from Macaronesia provide results similar in some but different in other respects from
8 Pacific plants. For example, Francisco-Ortega et al. (1996a)
found high genetic identities between taxa (average of
0.893) in Arovranthemum (Asteraceae), a result similar to many species endemic to Pacific islands. However, they detected 50% higher allozyme diversity within populations
(0.098) in Arovranthemum than the mean total diversity
(Hy=0.064) for species endemic to other oceanic islands (de
Joode and Wendel 1992). Relatively high genetic diversity has also been reported for other Macaronesian endemics in
Poaceae, Fabaceae, and Asteraceae (see Francisco-Ortega et al. 1996a). Although several electrophoretic studies from
Macaronesian Island plants, primarily the Canary Island plants, agree with this generally recognized pattern in several aspects, we do not know levels of isozyme variation within and divergence among species in subg. Dendrosonchus and closely related genera compared to results from other island endemics, both in the Pacific and Atlantic. It is also of interest to know whether the radiation of subg.
Dendrosonchus in Macaronesia has involved a relatively recent explosive radiation from a single introduction or is the result of differentiation from more than one ancestral introduction over a much longer period of time. These questions can be addressed using enzyme electrophoresis because a single rapid radiatin would be favored if the species have high identities, whereas multiple introductions over long time period should result in lowered identities among certain taxa.
In this dissertation, I discuss first phylogenenetic
relationships among the genera of the subtribe Sonchinae
(Asteraceae: Lactuceae) to which the genus Sonchus belongs.
This was the initial study of this project not only to
assess phylogenetic relationships among genera (and
subgenera in Sonchus) within the Sonchinae, but also to
determine the phylogenetic position of the Macaronesian
endemics. In the following chapters, I demonstrate a common
origin of the woody members of Sonchus and five closely
related genera in the Macaronesian Islands (i.e., the woody
Sonchus alliance). I also discuss the role of colonization
and adaptive radiation in the evolution of the alliance and
also examine the origin of woody members of Sonchus in
Macaronesia. In chapter IV, I estimate levels of genetic
variation within and divergence among species of the woody members Sonchus and several closely related genera and discuss their implications for the origin and evolution of the woody Sonchus alliance. The isozyme results are also used to make comparisons with relationships based on morphology and internal transcribed spacers of nuclear ribosomal DNA (ITS of nrDNA). Finally, the phylogenetic relationships among the genera within the Sonchinae are assessed independently using non-coding sequences of chloroplast DNA (cpDNA). The utility of non-coding sequence of cpDNA is also discussed in this chapter.
10 LITERATURE CITED
Abel-Monem, A., N. D. Watkins, and P. W. Cast. 1971.
Potassion-argon ages, volcanic stratigraphy, and
geomagnetic polatiry of the Canary Islands; Lanzarote,
Fuerteventura, Gran Canaria, and La Gomera. Amer. J.
Sci. 271: 490-521.
______, ______, and ______. 1972. Potassium-argon
ages, volcanic stratigraphy, and goemagnetic polarity
history of the Canary Islands: Tenerife, La Palma, and
Hierro. Amer. J. Sci. 272: 805-825.
Aldridge, A. 1975. Taxonomic and anatomical studies in
Sonchus L. Subgenus Dendrosonchus Webb ex Schults Bip.
and related genera. Ph. D. Dissertation. University of
Reading, England.
______. 1976a. Macaronesian Sonchus subgenus Dendrosonchus
s.l. (Compositae-Lactuceae), including a reappraisal of
the species concept and new combinations. Bot. Mac. 2:
81-93.
______. 1976b. A critical reappraisal of the Macaronesian
Sonchus subgenus Dendrosonchus (Compositae-Lactuceae).
Bot. Mac. 2: 25-57.
. 1977. Anatomy and evolution in the Macaronesian
Sonchus subgenus Dendrosonchus (Compositae-Lactuceae).
Nodal and petiolar vascular patterns. Bot. Mac. 3: 41-
59.
11 1978. Anatomy and evolution in the Macaronesian
Sonchus subgenus Dendrosonchus (Compositae-Lactuceae).
Bot. J. Linn. Soci. 76; 249-285.
. 1979. Evolution within a single genus: Sonchus in
Macaronesia. In "Plants and Islands" (ed. D. Bramwell) ,
Academic Press, New York, NY.
Anacochea, E . , J. M. Fuster, E. Ibarrola, A. Cendrero, J.
Coello, F. Hernan, J. M. Cantagrel, and C. Jamond.
1990. Volcanic evolution of the island of Tenerife
(Canary Islands) in the light of new K-Ar data. J.
Volcanol. Geotherm. Res. 44: 231-249.
Baldwin, B. G. 1992. Phylogenetic utility of the internal
transcribed spacers of nuclear ribosomal DNA in plants:
An example from the Compositae. Mol. Phy. and Evol. 1:
3-16.
______, D. W. Kyhos, and J. Dvorak. 1990. Chloroplast DNA
evolution and adaptive radiation in the Hawaiian
silversword alliance (Asteraceae-Madiinae). Ann.
Missouri Bot. Gard. 77: 96-109.
______, ______, , and G. D. Carr. 1991.
Chloroplast DNA evidence for a Norht American origin of
the Hawaiian silversword alliance (Asteraceae) . Proc.
Natl. Acad. Sci. USA 88: 1840-1843.
Banda, E. J. J. Danobeitia, E. Surinach, and J. Ansorge.
1981. Features of crustal structure under the Canary
Islands. Earth Planet Sci. Lett. 55: 11-24.
12 Boulos, L. 1960. Cytotaxonomic studies in the genus Sonchus.
2. The genus Sonchus. a general systematic treatment.
Bot. Not. 113: 400-420.
______. 1965. Babcockia. un nouveau genre de Composées des
iles Canaries. Bull. Jard. Bot. Etat. 35: 63-66.
______. 1967a. Taeckholmia. a new genus of Compositae from
Canary Islands. Bot. Not. 120: 95-108.
. 1967b. Nomenclatural changes and new taxa in
Sonchus from the Canary Islands. Nytt Mag. Bot. 14: 7-
18.
. 1968. The genus Sonchus and allied genera in the
Canary Islands. Cuad. de Bot. 3: 19-26.
. 1972. Revision systématique du genre Sonchus L.
s.l. I. Introduction et classification. Bot. Not. 125;
287-305.
. 1973. Revision systématique du genre Sonchus L.
s.l. IV. Soud-genre I. Sonchus. Bot. Not. 126: 155-196.
. 1974a. Revision systématique du genre Sonchus L.
s.l. V. Dendrosonchus. Bot. Not. 127: 7-37.
. 1974b. Revision systématique du genre Sonchus L.
s.l. VI. Sous-genre 3. Oriaosonchus. genres Emberaeria.
Babcockia. et Taeckholmia. Species exclusae dt dubiae.
Index. Bot. Not. 127: 402-451.
Bramwell, D. 1972. Endemism in the flora of the Canary
Islands. In "Taxonomy, Phytogeography, and Evolution",
ed. D. H. Valentine. Academic Press, New York, NY.
13 1975. Some morphological aspects of the adaptive
radiation of Canary Islands Echium species. Anal. Inst.
Bot. Cavanilles 32: 241-254.
. 1976. The endemic flora of the Canary Islands;
Distribution, relationships and phytogeography. In
"Biogeography and Ecology in the Canary Islands, ed. G.
Kunkel, 99 207-240. Dr. W. Junk Publishers: Hague.
Cantagrel, J. M., A. Cendrero, J. M. Fuster, E. Ibarrola,
and C. Jamond. 1984. K-Ar chronology of the volcanic
eruptions in the Canarian Archipelago: Islands of La
Gomera. Bull. Volcanol. 47: 597-609.
Carlquist, S. 1962. A theory of paedomorphosis in
dicotyledonous wood. Phytomorphology 12: 30-45.
_ . 1965. Island Life. New York, New York, The Natural
History Press.
_ . 1970. Hawaii, A natural history. New York, New
York, Natural History Press.
. 1974. Island Biology. New York and London.
Columbia University Press.
Carr, G. 1985. Monographs of the Hawaiian Madiinae
(Asteraceae) : Aroyranthemum. Dubautia. and Wilkesia.
Allertonia 4: 1-123.
______and D. W. Kyhos. 1981. Adaptive radiation in the
Hawaiian silversword alliance (Compositae-Madiinae). I.
Cytogenetics of spontaneous hubrids. Evolution 35: 543-
556.
14 Carracedo, J. C. 1994. The Canary Islands: an example of
structural control on the growth of large oceanic-iland
volcanoes. J. of Volcanology and Geothermal Research
60: 225-241.
Coello, J., J. -M. Cantagrel, F. Hernan, J. -M. Fus term E.
Ibarrola, E. Ancochea, C. Casquet, C. Jamond, J. -R.
Diaz de Teran, and A. Cendrero. 1992. Evolution of the
eastern volcanic ridge of the Canary Islands based on
new K-Ar data. J. Volcanol. Geotherm. Res. 53: 251-274.
Crawford, D. J., T. F. Stuessy, and M. Silva O. 1987.
Allozyme divergence and the evolution of Dendroseris
(Compositae: Lactuceae) on the Juan Fernandez Islands.
Syst. Bot. 12: 432-443.
______, , M. B. Cosner, D. W. Haines, M. Silva 0.,
and M. Baeza. 1992a. Evolution of the genus Dendroseris
(Asteraceae: Lactuceae) on the Juan Fernandez Islands:
Evidence from chloroplast and ribosomal DNA. Syst. Bot.
17: 676-682.
Darwin, C. 1859. On the origin of species by means of
natural selectoins. Reprint of first edition, 1950.
Watts and Co., London.
______1872. On the origin of species by means of natural
selection or the preservation of favored races in the
struggle for life. 6th ed. Murray, London.
De Joode, D. R. and J. F. Wendel. 1992. Genetic diversity
and origin of the Hawaiian Island endemic, Gossvm'um
15 tomentosum. Amer. J. Bot. 79: 1311-1319.
Francisco-Ortega, J., D. J. Crawford, A. Santos-Guerra, J.
A. Carvalho. 1996a. Isozyme differentiation in the
endemic genus Aroyranthemum (Asteraceae: Anthemideae)
in the Macaronesian Islands. Pi. Syst. Evo. 202: 137-
152.
______, R. K. Jansen, and A. Santos-Guerra. 1996b.
Chloroplast DNA evidence of colonization, adaptive
radiation, and hybridization in the evolution of the
Macaronesian flora. Proc. Natl. Acad. Sci. USA 93:
4085-4090.
Hansen, A. and P. Sunding. 1993. Flora of Macaronesia. Check
list of vascular plants, 4th revised ed. Sommerfeltia
17: 1-296.
Helenurm, K. and F. Ganders. 1985. Adaptive radiation and
genetic differentiation in Hawaiian Bidens. Evolution
39: 753-765.
Humphries, C. J. 1979. Endemism and evolution in
Macaronesia. In "Plants and Islands", ed. D. Bramwell,
pp 171-199. London: Academic Press.
La Roche, F. and J. C. Rodriguez-Pihero. 1994. Aproximaciôn
al numéro de tâxones de la flora vascular silverstre de
los archipiélagos macaronésicos. Revista de la Academia
Canaria de Ciencias 6: 77-98.
Lowrey, T. K. and D. J. Crawford. 1985. Allozyme divergence
and evolution in Tetramolooium (Compositae: Astereae)
16 on the Hawaiian Islands. Syst. Bot. 10: 64-72.
Mayer, S. S. 1991. Artificial hybridization in Hawaiian
Wilkstroemia (Thymelaeaceae). Amer. J. Bot. 78: 122-
130.
McDougall, I. and H. U. Schmincke. 1976-77. Geochronology of
Gran Canaria, Canary Islands: Age of shield building
volcanism and other magmatic phases. Bull. Volcanol.
40: 57-77.
Pons, A. and L. Boulos. 1972. Revision systématique du genre
Sonchus L. s.l. III. Etude palynologique. Bot. Not.
125: 310-319.
Roux, J. and L. Boulos. 1972. Revision systématique du genre
Sonchus L. s.l. II. Etude caryologique. Bot. Not. 125:
306-309.
Saad, S. I. 1961. Pollen morphology in the genus Sonchus.
Pollen et Spores 3: 247-260.
Sanders, R. W., T. F. Stuessy, C. Marticorena, and M. Silva
O. 1987. Phytogeography and evolution of Dendroseris
and Robinsonia. tree-Compositae of the Juan Fernandez
Islands. Opera Bot. 92: 195-215.
Sang, T., D. J. Crawford, S.-C. Kim, and T. F. Stuessy.
1994. Radiation of the endemic genus Dendroseris
(Asteraceae) on the Juan Fernandez Islands: Evidence
from sequences of the ITS region of nuclear ribsomal
DNA. Amer. J. Bot. 81: 1494-1501.
______, ______, T. F. Stuessy, and M. Silva O. 1995. ITS
17 sequences and the phylogeny of the genus Robinsonia
(Asteraceae). Syst. Bot. 20: 55-64.
Skottsberg, C. 1956. Derivation of the flora and fauna of
the Juan Fernanez and Easter Islands; part 1. The Juan
Fernandez Islands. Nat. Hist. Juan Fernandez an dEaster
Islands 1: 193-405.
Soltis, P. S. and D. E. Soltis. 1995. Plant molecular
systematics: Inferences of phylogeny and evolutionary
processes, eds. M. K. Hecht et al. Evolutionary Biology
28: 139-194.
Stebbins, J. L. 1953. A new classification of the tribe
Cichorieae. Madrono 12: 65-81.
______and M. S. Walter. 1953. Chromosome and phylogeny in
the Compositae, tribe Cichorieae. Univ. Calif. Publ.
Bot. 26: 401-430.
Stuessy, T. F ., K. A. Poland, J. F. Sutter, R. W. Sanders,
and M. Silva O. 1984. Botanical and geological
significance of potassium-argon dates from the Juan
Fernandez Islands. Science 225: 49-51.
Sunding, P. 1979. Origins of Macaronesian flora. In "Plants
and Islands", ed. D. Bramwell. pp 13-40. London:
Academic Press.
Takhtajan, A. 1969. Flowering plants: origin and dispersal.
Edinburgh.
Wagner, W., D. R. Herbst, and S. H. Sohmer. 1990. Manual of
the flowering plants of Hawaii. University of Hawaii
18 Press and Bishop Museum Press: Honolulu. 2 vols. 1853
pp.
Wallace, A. R. 1880. Island Life. Macmillan and Co., London.
Wendel, J. F. and A. E. Percival. 1990. Molecular divergence
in the Galapagos Isalnds-Baja California species pair
Gossvpium klotzschianum and G. davidsonii (Malvaceae) .
PI. Syst. Evol. 171: 99-115.
Witter, M. S. and G. D. Carr. 1988. Adaptive radiation and
genetic differentiation in the Hawaiian silversword
alliance (Compositae: Madiinae). Evolution 42: 1278-
1287.
19 CHAPTER 2
PHYLOGENETIC RELATIONSHIPS AMONG THE GENERA OF THE SUBTRIBE
SONCHINAE (ASTERACEAE) : EVIDENCE FROM ITS SEQUENCES
ABSTRACT
Sequences from the internal transcribed spacer region
(ITS) of nuclear ribosomal DNA were used to assess relationships among genera of subtribe Sonchinae
(Lactuceae). The data suggest that Sonchinae is paraphyletic, and that the Dendroseridinae should not be recognized as a distinct subtribe. Several Macaronesian genera, along with woody Macaronesian species of Sonchus
(all diploid, 2n = 18), appear to have been derived from a common ancestor. Prenanthes pendula is more closely related to this Macaronesian group than to other members of
Prenanthes. suggesting that Prenanthes. like Sonchus. is polyphyletic. The Juan Fernandez Islands endemic genus
Dendroseris (2n = 36) is monophyletic, but its phylogenetic relationships to other Sonchinae remain uncertain.
Reichardia. which forms the sister genus of the Sonchinae,
20 is also monophyletic. The genus Launaea. by contrast, is
considered to be paraphyletic. The monotypic genus
Aetheorhiza (2n = 18) is sister to the annual weedy species .
of Sonchus. rather than being closely related to Launaea.
Two monotypic endemic genera from New Zealand, Emberqeria
(2n = 36) and Kirkianella (2n = 90, 126) , are sister genera,
and together they form the sister clade to Sonchus arvensis-
S. maritimus. This implies that these Pacific genera were
derived either from section Arvenses or Maritimi of subgenus
Sonchus. rather than from section Apseri.
INTRODUCTION
Bremer (1993, 1994) proposed a revised subtribal classification of the Lactuceae in which he recognized a new subtribe Sonchinae from the Launaea-Sonchus group of
Stebbins (1953) (Table 1). He also divided the subtribe
Crepidinae of Stebbins (1953) into the four subtribes, i.e.,
Crepidinae sensu stricto, Hieraciinae, Lactucinae, and
Sonchinae (Table 1). Bremer (1993, 1994) suggested that
Launaea and Sonchus are closely related and that the entire
Launaea-Sonchus clade is monophyletic. In contrast, Jeffrey
(1966) separated Sonchus and Launaea into two groups, i.e., the Crepis L. subgroup and Sonchus group (Table 1).
The Sonchinae as recognized by Bremer (1993) comprises
21 11 genera and approximately 130 species, and consists of two alliances: (1) Launaea and the related genera Aetheorhiza and Reichardia; and (2) Sonchus and a group of closely related genera, i.e., Actites. Babcockia. Emberaeria.
Kirkianella, Lactucosonchus. Sventenia. and Taeckholmia.
Most species occur in the Mediterranean region, Africa, and the Canary Islands, while several genera are found in the
Pacific islands. Several genera of the Sonchus group, including Babcockia. Lactucosonchus. Sventenia. and
Taeckholmia. are restricted to the Canary Islands. Two monotypic genera, Kirkianella and Emberaeria. are endemic to
New Zealand, and Actites occurs exclusively in Australia. In contrast, Sonchus and Launaea are widely distributed in
Africa, Macaronesia, and Eurasia. The 54 species of Sonchus are divided into three subgenera (Boulos 1972) :
Dendrosonchus (19 spp.), Oriaosonchus (14 spp.), and Sonchus
(21 spp.). Subgenus Dendrosonchus consists of woody plants endemic to Macaronesia, subg. Oriaosonchus occurs exclusively in Africa, and subg. Sonchus includes widely distributed weedy species. The monotypic genus Aetheorhiza is distributed locally throughout the Mediterranean and western Europe (Babcock and Stebbins 1943) . Reichardia. which is composed of eight species, occurs in the
Mediterranean and Europe (Gallego et al. 1980) .
In Jeffrey's (1966) informal classification of the
Lactuceae, Dendroseris and Thamnoseris Phil., which are
22 endemic to the Juan Fernandez and San Ambrosio Islands,
respectively, were placed with Sonchus in his Sonchus group
(Table 1) . Stebbins (1953), however, placed Dendroseris in
its own subtribe (Dendroseridinae) and Thamnoseris in the
subtribe Stephanomeriinae. Bremer (1994) retained the
Dendroseridinae with the addition of Thamnoseris.
Furthermore, he indicated uncertainty with regard to the
position of the two genera in Lactuceae (see also Sanders et
al. 1987).
Chloroplast DNA (cpDNA) restriction site data provided
several insights into phylogenetic relationships among
genera of the Lactuceae (Whitton et al. 1995) . For example,
the Crepidinae of Stebbins (1953) was suggested to be highly
polyphyletic, while Dendroseris. and thus his subtribe
Dendroseridinae, appeared monophyletic. Monophyly of
Dendroseris is concordant with the results of Sang et al.
(1994) . The cpDNA restriction site data of Whitton et al.
(1995) also suggested that the Sonchinae of Bremer (1994)
was paraphyletic because Dendroseris was nested within it.
The cpDNA data indicated that Dendroseris was closely
related to Sonchus and Sventenia. Restriction site data
implicated Reichardia as the sister group to a monophyletic
clade of Dendroseris. Sonchus. Sventenia. and Aetheorhiza.
Although the cpDNA data of Whitton et al. (1995) resolved
several phylogenetic issues, broader sampling within the
Sonchinae is needed to assess relationships more rigorously.
23 The internal transcribed spacers (ITS) of nuclear
ribosomal DNA (nrDNA) have proven useful for elucidating
phylogenetic relationships 'among congeneric species and
closely related genera (see Baldwin et al. 1995). The
primary purpose of this study was to use ITS sequences to
assess phylogenetic relationships among genera within
subtribes Sonchinae and Dendroseridinae. More specifically,
we wished to: (1) assess the monophyly of Sonchinae; (2)
test the monophyly of several genera within this subtribe;
(3) determine phylogenetic relationships among genera of the
Sonchinae, especially of Dendroseris and other small Pacific genera, with the largely Macaronesian, African, and
Mediterranean Sonchus.
MATERIALS AND METHODS
Complete sequences of the ITS region were generated for
36 accessions representing 11 genera and 31 species of subtribe Sonchinae (Bremer 1994), three species of
Prenanthes, one species of Taraxacum. and one species of
Lactuca (Table 2). Sequences from nine species of
Dendroseris and the four outgroup genera Kriaia.
Pvrrhopappus. Microseris. and Lactuca. were obtained from
Sang et al. (1994) and Kim and Jansen (1994) , respectively.
Four genera, Kriaia. Microseris. Pvrrhopappus. and Lactuca.
24 were chosen as outgroups based on the cpDNA restriction site data of Whitton et al. (1995) and the advice of Charles
Jeffrey (Royal Botanical Gardens, Kew) and Kâre Bremer
(Uppsala University). Two additional species of Prenanthes were also included in this study to assess the phylogenetic relationships of the Canary Island endemic P. pendula (see
Discussion). No material was available for Actites (subtribe
Sonchinae) from Australia and Thamnoseris (subtribe
Dendroseridinae) from the San Ambrosio Islands in the South
Pacific off the coast of northern Chile.
Total genomic DNA was isolated from leaf tissue using the CTAB method of Doyle and Doyle (1987), and purified further by ultracentrifugation with CsCl/ethidium bromide gradients (Sambrook et al. 1989). Double-stranded DNAs of the entire ITS regions including the 5.8s coding region were amplified directly by 30 cycles of symmetric PCR using universal primers (White et al. 1990). The initial PCR reaction was 3 min at 95°C for dénaturation, 1 min at 50®C for annealing, and 1 min at 72°C for primer extension. Each of the 30 cycles consisted of 1 min at 95°C, 1 min at 50°C, and 45 sec at 72®C. PCR products were purified by agarose gel (IX TAE buffer) electrophoresis and the concentrated
DNAs were recovered using glass powder (U.S. Bioclean, U.S.
Biochemical).
Double-stranded PCR products were directly sequenced using the Sequenase Version 2.0 (United States Biochemical
25 Corp.) dideoxychain-termination method, employing two
forward (ITS3 and ITS5) and two reverse (ITS2 and ITS4) primers (White et al. 1990). Modifications to the Sequenase protocol included dénaturation of the double-stranded DNA by boiling the DNA/primer mix for 3 min, followed by snap- chilling the annealing mixture for 7 min in an ice water bath. In addition, 1 nl DMSO was added to both the labeling and termination reactions to reduce the effects of DNA secondary structure (Cosner et al. 1994).
DNA sequences were separated in 6% acrylamide gels using wedge-shaped spacers. Both short (3.5 hr at 1500 volts) and long (7.5 hr at 1500 volts) gels were run, in order to read the entire ITS region. Gels were fixed for 30 min in 10% acetic acid, transferred to 3-MM Whatmann filter paper, dried under vacuum for 2.5 hr at 80°C, and exposed to
Kodak XAR x-ray film for 12-72 hr.
The boundaries of the ITS and rDNA coding regions were identified by comparison to known sequences (Yokota et al.
1989; Ramon et al. 1990; Baldwin 1992, 1993; Kim and Jansen
1994). Sequences were aligned using the Clustal W program
(provided by D. Higgins). Several gap opening and extension penalties were used to align the entire sequences. The sequences aligned with Clustal W were then adjusted manually in order to align several regions of conserved sequences.
Both the small size and number of indels made manual adjustments feasible (Table 3).
26 Phylogenetic analyses using Fitch parsimony were performed employing PAÜP (version 3.1.1; Swofford 1993) using the HEURISTIC search option with TBR branch swapping and MULPARS on. To search for multiple islands of trees
(Maddison 1991), 100 replications of "random" taxa addition were conducted. Bootstrap analysis (Felsenstein 1985) was performed with 100 replications (maxtree = 100) to provide a measure of support for the clades. Decay analysis (Bremer
1988; Donoghue et al. 1992) was also performed to assess the robustness of the monophyletic groups. Trees up to five steps longer were examined.
RESULTS
Length Variation and Base Composition of the ITS
Region. The length of ITS 1 varied from 233 bp to 255 bp, and ITS 2 varied from 220 bp to 226 bp. One 18 bp deletion in ITS 1 was detected in the monotypic genus Aetheorhiza.
The length of ITS 1 and ITS 2 of the Sonchinae is within the size range of other Asteraceae (Baldwin et al. 1995). The
G+C content in ITS 1 varied from 42.0% fTaeckholmia heterophvlla) to 48.2% (Sonchus luxurians), while ITS 2 varied from 46.8% (Sonchus aryensis) to 52.7% (Sonchus palustris).
No evidence of multiple rDNA repeat types in any of the
27 taxa analyzed was observed. All of the double-stranded PCR
products obtained appeared as sharp, single bands on 1.0%
agarose gels. Furthermore, polymorphism at individual
nucleotide sites was not commonly encountered (except for
two taxa, Taeckholmia heterophvlla and Lactucosonchus ; Kim
et al. 1996) . Therefore, ITS sequence data in this study
provide no evidence for different ITS length variants or
major sequence variants within individual samples.
Phvlocenetic Analvses. A total of 330 variable sites
was found among all taxa examined, with 264 of them
phylogenetically informative. There were 274 variable sites
in the ingroup; 188 were cladistically informative. Only 64
of the 16170 characters (0.4%) were scored as missing or
ambiguous sites. Unambiguous gap positions were treated as
missing and they were not scored as separate characters in
the phylogenetic analyses. The heuristic search identified
144 equally parsimonious trees with a length of 898, a
consistency index (Cl) of 0.580 (0.526 excluding
uninformative changes), and a retention index (RI) of 0.743.
The Cl is close to the regression line of Sanderson and
Donoghue (1989) for an analysis of 50 taxa.
The ITS tree (Fig. 1) suggests that Reichardia and
Launaea are basal in the Sonchinae. The tree does not
support the monophyly of the Sonchinae because Dendroseris and Prenanthes pendula are nested within the subtribe. The
28 monophyly of the woody members of Sonchus and their close
relatives in Macaronesia (Fig. 1; clade A) is very strongly
supported (100% bootstrap value and a decay index of >5) .
Prenanthes pendula. which is endemic to the Canary Islands,
is clearly grouped within this clade. Bootstrap analysis
also shows strong support (>85%) for the monophyly of
several genera and groups of genera including Reichardia.
Dendroseris. subg. Oriaosonchus. and
Emberaeria-Kirkianella. Several genera such as Launaea and
Taeckholmia appear to be paraphyletic. Both Prenanthes and
Sonchus are polyphyletic.
DISCUSSION
Phylogenetic Relationships in the Sonchinae.
Phylogenetic analysis of ITS sequences provides several
insights into evolutionary relationships within the
Sonchinae (Fig. 1) . The subtribe as delimited by Bremer
(1993, 1994) is not monophyletic. The ITS data instead provide support for Jeffrey's view that Dendroseris is closely related to Sonchus and its relatives (Jeffrey 1966; see Table 1) . In the ITS tree Dendroseris is nested within subtribe Sonchinae, which indicates that the Dendroseridinae of Stebbins (1953) should not be recognized as a distinct subtribe. This relationship is also concordant with cpDNA
29 restriction site data (Whitton et al. 1995). The
Launaea-Sonchus line (Table 1) recognized by Stebbins (1953) is not supported by the ITS data. The two major alliances of
Bremer within the Sonchinae, i.e., Launaea with related genera, and Sonchus and a group of closely related genera, are not distinguished in the ITS phylogeny because
Aetheorhiza is closer to certain members of Sonchus than to
Launaea and Reichardia (Fig. 1) . However, the two genera
Reichardia and Launaea are not part of the main radiation of
Sonchinae, which is congruent with cpDNA restriction site data (Whitton et al. 1995).
The ITS sequence data also offer the opportunity to examine the monophyly of several genera. The ITS tree (Fig.
1) supports the monophyly of both Reichardia and
Dendroseris. while Taeckholmia and Launaea appear paraphyletic. However, the paraphyly of Launaea (ca. 50 species) needs to be examined further with wider taxonomic sampling. In addition, the ITS phylogeny provides strong evidence that Sonchus and Prenanthes are polyphyletic.
Within Sonchus. subg. Oriaosonchus is monophyletic, while the two subgenera Dendrosonchus and Sonchus are not.
Sonchus and Related Genera in Macaronesia. Boulos
(1972) recognized three subgenera in Sonchus. He suggested that all three subgenera are natural groups and proposed that subg. Dendrosonchus in Macaronesia evolved from subg.
30 Oriaosonchus in Africa, which he considered the most
primitive in the genus (Boulos 1967). Furthermore, he
hypothesized that subg. Sonchus was derived from woody
members of Sonchus. i.e., subg. Dendrosonchus, in
Macaronesia. Neither monophyly of each of the three
subgenera nor the phylogenetic relationships among them as
proposed by Boulos (1967, 1972) is supported by the ITS
data. For example, subg. Dendrosonchus is closer to several
genera from Macaronesia than to either subg. Sonchus or
subg. Oriaosonchus (Fig. 1; clade A ) . In addition, subg.
Dendrosonchus is paraphyletic and members of subg. Sonchus were derived independently several times.
The ITS phylogeny strongly supports the monophyly of
subg. Dendrosonchus and its close relatives, including the herbaceous tuberous perennial Sonchus tuberifer and the woody Prenanthes pendula. all of which occur in Macaronesia
(Fig. 1; clade A ) . This suggests that members of subg.
Dendrosonchus. along with Sventenia. Babcockia. Taeckholmia.
P. pendula. S. tuberifer. and Lactucosonchus. were derived from a common ancestor in the Macaronesian Islands (Kim et al. 1996). Within this clade, Lactucosonchus. which is monotypic and endemic to La Palma in the Canaries, is sister to the rest of the taxa. The ITS tree also suggests that
Sonchus palustris. a member of subg. Sonchus that occurs widely in Europe, is the sister taxon to the Macaronesian clade (Fig. 1). This relationship, however, is weakly
31 supported.
The Juan Fernandez Islands Endemic Dendroseris. Our
results, like those of Sang et al. (1994) and Whitton et al.
(1995), support the monophyly of Dendroseris (Fig. 1; clade
B), and further suggest that, while two subgenera
(Dendroseris and Phoenicoseris) are monophyletic, subg. Rea
is paraphyletic. The subg. Dendroseris is weakly supported
as sister to the remainder of the genus. The relationship of
Dendroseris to other genera of Sonchinae remains uncertain
beyond being part of the large clade that contains
everything except Launaea. and Reichardia.
The Pacific Genera Emberoeria and Kirkianella. The
monotypic genus Emberaeria. a tetraploid (2n = 36), is
endemic to the Chatham Islands of New Zealand, and has been
viewed as a survivor from Pleistocene glaciation (Wardle
1963; Lander 1976; Webb et al. 1988). Lander (1976)
hypothesized that during the late Pliocene subg. Sonchus migrated to New Zealand where Emberaeria subsequently
originated. Cytological, palynological, and morphological
data suggested subg. Sonchus. especially sect. Asperi (S. kirkii), as the most probable ancestor of Emberaeria (Boulos
1967; Pons and Boulos 1972; Roux and Boulos 1972). The monotypic genus Kirkianella. a pentaploid (2n = 90) and heptaploid (2n = 129) (Beuzenberg and Hair 1984), is also
32 endemic to New Zealand and is morphologically variable. This genus has been considered closely related to either Sonchus
or Launaea (Allan 1961). Glenny (pers. comm.) also proposed
a close relationship to Youncria Cass, and Crepis L., genera
of Lactuceae outside Sonchinae (not sampled in this study).
In the ITS tree, the two New Zealand endemics are sister to each other and form a monophyletic group with two species of
Sonchus subg. Sonchus (Fig. 1; clade C). The presence of this strongly supported clade is somewhat unexpected because
Sonchus kirkii has been considered ancestral to Emberaeria. yet it is sister to the clade of S. asper and S. oleraceus
(Fig. 1; clade D) . Therefore, the ITS phylogeny does not support previous hypotheses about Emberaeria. Rather, it seems more likely that Emberaeria and Kirkianella are most closely related to S. arvensis (sect. Arvenses) and S. maritimus (sect. Maritimi).
Aetheorhiza. Babcock and Stebbins (1943) suggested that
Aetheorhiza of the Mediterranean and western Europe is closely related to the Launaea and Sonchus groups.
Aetheorhiza shares several features with Launaea. including four broad, round-ribbed, and obcompressed achenes, long corolla tubes with conspicuous white pubescence on their upper parts as well as on the base of the ligule, and a mixture of relatively broad and narrow setae in the pappus.
Several other features of Aetheorhiza. especially those of
33 the corolla and involucre, are shared with Sonchus. The
distinctive habit of Aetheorhiza was the sole basis for
Babcock and Stebbins' (1943) recognition of the genus; new
rosettes of leaves form along elongated rhizomes in
Aetheorhiza. The ITS tree shows that Aetheorhiza is clearly
part of the lineage that includes no species of Launaea and
all species of Sonchus (Fig. 1). It seems likely that
Aetheorhiza diverged as a part of the radiation of the
Sonchus group, which presumably occurred after the elements
classified as Launaea had already diverged.
Prenanthes pendula. The phylogenetic relationship of
the Canary Island endemic P. pendula to other species of
Prenanthes is problematic. It is the only species of the
genus from the Canary Islands that is locally common in
mountain cliffs on the south and north sides of Gran
Canaria. It has woody stems, small heads (about 2 mm across)
with 5-6 yellow florets, and simple pappus hairs (Bramwell
and Bramwell 1974). A close relationship between P. pendula
and African Prenanthes species was suggested by Bramwell
(1985) . In the ITS tree P. pendula is in a clade that is
strongly supported by high bootstrap (100%) and decay values
(>5) and that includes only plants from Macaronesia (Fig. 1;
clade A) . This finding is concordant with the results from
Perez de Paz (1976), who suggested that P. pendula is closely related to Sventenia and Sonchus. Also, a naturally
34 occurring intergeneric hybrid between Sventenia bupleuroides and P. pendula was reported by Sventenius (1960). Therefore, the ITS tree agrees with other results in suggesting a close relationship between P. pendula and woody Sonchus in
Macaronesia. The ITS tree further suggests that P. pendula is not closely related to other species of Prenanthes. For example, P. altissima and Taraxacum officinale are sister taxa, and P. purpurea is a sister taxon to the clade that includes those taxa and the remainder of the subtribe. Thus, the ITS phylogeny suggests that Prenanthes may be polyphyletic.
Although ITS sequence data have been useful in proposing phylogenetic hypotheses at and above the generic level in the subtribe Sonchinae, five issues remain unresolved: 1) relationships between subg. Dendrosonchus and its close relatives in the Macaronesian islands; 2) elucidation of the position of Prenanthes pendula; 3) confirmation of the paraphyly of Launaea and its relationships to other genera; 4) relationships of
Dendroseris and Thamnoseris to each other and to other
Sonchinae; 5) investigation of the relationship of Actites
(endemic to Australia) to other genera, especially
Emberaeria. Other data, especially from cpDNA, would provide additional data to resolve these issues.
35 ACKNOWLEDGMENTS
We are grateful to Javier Francisco-Ortega, Arnoldo
Santos-Guerra, Aguedo Marrero, Pedro Ortega-Machin, and
Francisco Jose Gonzalez Artiles for assistance in the field.
We also thank Eric Knox, Tom Myers, David Glenny, P. J.
Garnock-Jones, Loutfy Boulos, J. L. S. Kessing, and Joan
Pedrola i Monfort for providing plant and DNA materials. We
are indebted to Charles Jeffrey and Kâre Bremer for their
helpful suggestions about outgroups. We are extremely
grateful to Jeff Doyle for his comments and suggestions
during review of the manuscript. This paper represents a
portion of a doctoral dissertation by S.-C. K. submitted to
the Ohio State University. This work was supported by ASPT
Grants for Graduate Student Research, Tinker Foundation
(Latin American Studies Program, Ohio State University),
Beatley Herbarium Award (Ohio State University), and Sigma
Xi Grants-in-Aid of Research to S. -C. K, National Science
Foundation Doctoral Dissertation Improvement Grant DEB-
9521017 to S. -C. K. and D. J. C., and NSF grant BSR 87-
08246 to R. K. J. This study was also partially supported by the Centro de Investigacion y Tecnologla Agrarias de
Canarias through the advice of Manuel Fernàndez-Galvàn. We dedicate this paper to the memory of our friend and colleague Mary Elizabeth Cosner.
36 LITERATURE CITED
ALLAN, H. H. 1961. Flora of New Zealand, vol.l. Wellington,
New Zealand: Government Printer.
BABCOCK, E. B. and G. L. STEBBINS, JR. 1943. Systematic
studies in the Cichorieae. University of California
Publications in Botany 18: 227-240.
BALDWIN, B. G. 1992. Phylogenetic utility of the internal
transcribed spacers of nuclear ribosomal DNA in plants:
An example from the Compositae. Molecular Phylogenetics
and Evolution 1: 3-16.
______. 1993. Molecular phylogenetics of Calvcadenia
(Compositae) based on ITS sequences of nuclear
ribosomal DNA: Chromosomal and morphological evolution
reexamined. American Journal of Botany 80: 222-238.
______, C. S. CAMPBELL, J. M. PORTER, M. J. SANDERSON, M.
F. WOJCIECHOWSKI, AND M. J. DONOGHUE. 1995. Utility of
nuclear ribosomal DNA internal transcribed spacer
sequences in phylogenetic analyses of angiosperms.
Annals of the Missouri Botanical Garden 82: 247-277.
BEUZENBERG, E. J. AND J. B. HAIR. 1984. Contributions to a
chromosome atlas of the New Zealand flora - 27
Compositae. New Zealand Journal of Botany 22: 353-356.
BOULOS, L. 1967. Taeckholmia. a new genus of Compositae from
Canary Islands. Botaniska Notiser 120: 95-108.
______. 1972. Révision systématique du genre Sonchus L.
37 s.l. I. Introduction et classification. Botaniska
Notiser 125: 287-305.
BRAMWELL, D. 1985. Contribucion a la biogeografia de las
islas Canarias. Botanica Macaronesica 14: 3-34.
______AND Z. BRAMWELL. 1974. Wild flowers of the Canary
Islands. London: Stanley Thornes (Publishiers) Ltd.
BREMER, K. 1988. The limits of amino acid sequence data in
angiosperm phylogenetic reconstruction. Evolution 42:
795-803.
______. 1993. New subtribes of the Lactuceae. Novon 3;
328-330.
______. 1994. Asteraceae-Cladistics and classification.
Portland, Oregon; Timber Press.
COSNER, M. B. , R. K. JANSEN, AND T. G. LAMMERS. 1994.
Phylogenetic relationships in the Campanulales based on
rbcL sequences. Plant Systematics and Evolution 190:
79-95.
DONOGHUE, M. J. , R. G. OLMSTEAD, J. F. SMITH, AND J. D.
PALMER. 1992. Phylogenetic relationships of Dipsacales
based on rbcL sequences. Annals of the Missouri
Botanical Garden 79: 333-345.
DOYLE, J. J. AND J. L. DOYLE. 1987. A rapid DNA isolation
procedure for small quantities of fresh leaf material.
Phytochemical Bulletin 19: 11-15.
FELSENSTEIN, J. 1985. Confidence limits on phylogenies: An
approach using the bootstrap. Evolution 39: 783-791.
38 GALLEGO, M. J. , S. TALAVERA, AND S. SILVESTRE. 1980.
Revision del genero Reichardia Roth (Compositae).
Lagascalia 9: 159-218.
JEFFREY, C. 1966. Notes on Compositae, I. The Cichorieae in
East Tropical Africa. Kew Bulletin 18: 427-486.
KIM, K. -J. AND R. K. JANSEN. 1994. Comparisons of
phylogenetic hypotheses among different data sets in
dwarf dandelions (Krioia) : additional information from
internal transcribed spacer sequences of nuclear
ribosomal DNA. Plant Systematics and Evolution 190:
157-185.
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 radition. Proceedings
of the National Academy of Sciences USA. 93: 7743-7748.
LANDER, N. S. 1976. Actites. a new genus of Compositae from
Australia. Telopea 1: 129-135.
MADDISON, D. R. 1991. The discovery and importance of
multiple islands of most-parsimonious trees. Systematic
Zoology 40: 315-328.
PEREZ DE PAZ, J. 1976. Observaciones sobre la biologia y
relaciones de Sventenia bupleuroides F. Q. Botanica
Macaronesica 1: 51-65.
PONS, A. AND L. BOULOS. 1972. Révision systématique du genre
Sonchus L. s.l. III. Étude palynologique. Botaniska
39 Notiser 125: 310-319.
RAMON, A. T., M. CANAL, AND V. HEMLEBEN. 1990. GC balance in
the internal transcribed spacers ITS 1 and ITS 2 of
nuclear ribosomal RNA genes. Journal of Molecular
Evolution 30: 170-181.
ROUX, J. AND L. BOULOS. 1972. Révision systématique du genre
Sonchus L. s.l. II. Étude caryologique. Botaniska
Notiser 125: 306-309.
SAMBROOK, J. , E. F. FRITSCH, AND R. MANIATIS. 1989.
Molecular cloning, a laboratorv manual. Second Edition.
Plainview, New York: Cold Spring Harbor Laboratory
Press.
SANDERS, R. W. , T. F. STUESSY, C. MARTICORENA, AND M. SILVA
0. 1987. Phytogeography and evolution of Dendroseris
and Robinsonia. tree-Compositae of the Juan Fernandez
Islands. Oper Botanica 92: 195-215.
SANDERSON, M. J. AND M. J. DONOGHUE. 1989. Patterns of
variation in levels of homoplasy. Evolution 43: 1781-
1795.
SANG, T., D. J. CRAWFORD, S. -C. KIM, AND T. F. STUESSY.
1994. Radiation 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 8: 1494-1501.
STEBBINS, G. L. JR. 1953. A new classification of the tribe
Cichorieae, family Compositae. Madrono 12: 65-81.
40 SVENTENIUS, E. R. S. 1960. Additamentum ad floram
Canariensem. Madrid, Spain: Institute Nacional de
Investigaciones Agronômicas Ministerio de Agriculture.
SWOFFORD, D. L. 1993. PAUP: Phvloqenetic analysis usina
parsimony, yersion 3.1.1. Champaign: Illinois Natural
History Suryey.
WARDLE, P. 1963. Evolution and distribution of the New
Zealand flora, as affected by Quaternary climates. New
Zealand Journal of Botany 1: 3-17.
WEBB, C. J . , W. R. SYKES, AND P. J. GARNOCK-JONES. 1988.
Flora of New Zealand, vol. IV. Christchurch: Botany
Division DSIR.
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.
WHITE, T. J. , T. BRUNS, S. LEE, AND J. TAYLOR. 1990.
Amplification and direct sequencing of fungal ribosomal
RNA genes for phylogenetics. Pp. 315-322 in PCR
protocols: A guide to methods and amplification. eds.
M. Innis, D. Gelfand, J. Sninsky, and T. White. San
Diego: Academic Press.
YOKOTA, Y. , T. KAWATA, Y. IIDA, A. KATO, and S. TANIFUJI.
1989. Nucleotide sequences of the 5.8s rRNA gene and
internal transcribed spacer regions in carrot and broad
41 bean ribosomal DNA. Journal of Molecular Evolution 29;
294-301.
42 CHAPTER 3
A COMMON ORIGIN FOR WOODY SONCHUS AND FIVE RELATED
GENERA IN THE MACARONESIAN ISALNDS: MOLECUIAR
EVIDENCE FOR EXTENSIVE RADIATION
ABSTRACT
Woody Sonchus and five related genera fBabcockia.
Taeckholmia. Sventenia. Lactucosonchus, and Prenanthes) of
the Macaronesian islands have been regarded as an
outstanding example of adaptive radiation in angiosperms.
ITS sequences were used to demonstrate that, despite the
extensive morphological and ecological diversity of the
plants, the entire alliance in insular Macaronesia has a
common origin. The sequence data place Lactucosonchus as
sister group to the remainder of the alliance and also
indicate that four related genera are in turn sister groups to subg. Dendrosonchus and Taeckholmia. This implies that the woody members of Sonchus were derived from an ancestor similar to allied genera now present on the Canary Islands.
It is also evident that the alliance probably occurred in
43 the Canary Islands during the late Miocene or early
Pliocene. A rapid radiation of major lineages in the
alliance is consistent with an unresolved polytomy near the
base and low ITS sequence divergence. Increase of woodiness
is concordant with other insular endemics and refutes the
relictural nature of woody Sonchus in the Macaronesian
islands.
INTRODUCTION
Ascertaining the origin and evolution of plants endemic to oceanic islands is both fascinating and frustrating.
Endemics may assume the typical insular woody habit and become very distinct in other suites of characters such that determining their continental relatives is difficult if not impossible with morphology (1-4) . Once a colonizer becomes established on an oceanic island, extensive diversification may occur as plants move into a variety of open habitats.
This process has been viewed as a good example of adaptive radiation (2-9) , and once it occurs, the different descendent lines may exhibit a wide array of characters, making it difficult to determine whether they evolved from a common ancestor. Thus, difficult problems posed by insular endemics include whether they result from a single introduction, estimating the time of radiation, and
44 identifying the continental relative(s) and source area(s) of the original colonizer(s) (10-12). As noted above, comparative morphology may be of limited value for addressing these questions because of the difficulty in distinguishing features shared by common ancestry as opposed to parallelisms. Molecular sequences, however, have proven useful in the study of some insular groups because, contrasted with morphology, base substitutions of determined regions (mutations) may be neutral or nearly so. (11, 13-
16) .
One of the basic assumptions in the study of the origin and evolution of endemics in oceanic islands is that the remoteness of the islands from possible source areas
(combined with their small land masses) acts like a sieve and the arrival of propagules to the islands is a very rare event. This in turn makes it more likely that related endemics on remote archipelagos have a common origin, i.e., represent a monophyletic group. Unlike more remote volcanic islands in the Pacific (Juan Fernandez, Hawaiian, and
Galapagos Islands) where most evolutionary studies of insular endemic plants have been carried out, the
Macaronesian islands (Fig. 2a, b) are very close to possible source areas and exhibit a broad range of geological ages
(17-2 2) (In this paper, "Macaronesian" will refer to the islands of this area, and the Canary Islands will be specified when referring to that archipelago alone). The
45 proximity of the islands to the African continent and their
different geological ages make it much more likely that
multiple colonization events could have occurred for some
closely related taxonomic groups, and that some taxa could
be much older than others. This means, of course, that
closely related endemic taxa may not be monophyletic. There
also has been a long controversy over whether some of the
woody Macaronesian endemics are relict elements of a flora
that extended along the Mediterranean basin during the
Tertiary period or they are recent derivatives from
continental ancestors (3, 6, 19, 23-26).
There are 34 taxa of ca. 130 taxa in the subtribe
Sonchinae (Asteraceae) endemic to Macaronesia (27) , with
most of them in the Canary Islands. A previous phylogenetic
analysis of the Sonchinae (28, Fig. 3) recognized a
Macaronesian clade that includes the woody members of
Sonchus and five related genera (hereafter for convenience
often referred to as the woody Sonchus alliance), but this
was based on limited taxon sampling from Macaronesia. This
alliance is composed primarily of 19 species of woody
Sonchus (i.e., subg. Dendrosonchus1. the genera Babcockia.
Taeckholmia. Sventenia. Lactucosonchus. one species of
Sonchus subg. Sonchus (i.e., S. tuberifer). and one species
of Prenanthes. Two taxa, Lactucosonchus and S. tuberifer.
are the only members of the group which do not have a true woody habit; they are herbaceous perennials with tuberous
46 roots. As may be inferred by the recognition of six genera, these taxa display great morphological, ecological, and anatomical diversity (29-30). Despite this diversity, the fertility of several intergeneric hybrids suggests genetic cohesiveness within the alliance (38, 39) . The basic question is whether this remarkable array of plants results from a single introduction followed by extensive radiation and diversification, or from several introductions from nearby and closely related source areas over a long period of time. It has been shown, for example, that the three genera of the very diverse silversword alliance in the
Hawaiian Islands came from a single dispersal event (10,
34), but the extreme isolation of Hawaii contrasts with the situation in Macaronesia where a source area is within one hundred km of certain islands (Fig. 2).
In this study, sequences from the internal transcribed spacer region of the nuclear ribosomal DNA (ITS) were used to assess further the monophyly of the woody Sonchus alliance in the Macaronesian islands. The sequence data were also used to elucidate phylogenetic relationships within the group, the role of colonization and adaptive radiation in the evolution of the alliance, and to examine the origin of woody members of Sonchus in Macaronesia.
47 MATERIALS AND METHODS
A phylogenetic analysis of subtribe Sonchinae was
conducted previously, with limited sampling from the
Macaronesian taxa (28, Fig. 3). Following this analysis, we
examined the ITS sequences of 35 additional accessions
representing 16 species of subg. Dendrosonchus. five species
of Taeckholmia. and the four related genera endemic to the
Canary Islands. Total DNA was isolated from leaf tissues using the CTAB method (35) , and purified in CsCl/ethidium bromide gradients. Methods for PCR amplification, purification of PCR products, and sequencing reactions of the ITS regions are given in detail elsewhere (11, 28) . Both coding and noncoding strands were read and all sequences were manually aligned using MacClade ver 3.0 (36) . Aligned sequences are available upon request from the first author.
Variable nucleotide sites were analyzed by unweighted
Wagner parsimony using PAUP 3.1.1 (37). On the basis of previous analyses of the Sonchinae (28), Sonchus palustris was used as an outgroup. The heuristic algorithm with stepwise addition option was used to find the shortest trees. Bootstrap (100 replicates) and decay analyses were performed to assess the relative support for clades found in the analysis (38). Pairwise sequence divergences were calculated by the Jukes and Cantor one-parameter method using PHYLIP 3.52c (39). To assess rate homogeneity over
48 different lineages, the relative-rate test (40, 41) was
conducted using the genus Lactucosonchus as the reference
taxon because this lineage is sister to the remainder of the
alliance.
RESULTS
Size of ITS, sequence divergence, and relative-rate
test. Lengths of ITS 1 and ITS 2 in the woody Sonchus alliance fall within the size range reported for Asteraceae
(42). Percent pairwise sequence divergence between species ranges from 0.0 - 4.8, and average divergences within subg.
Dendrosonchus and Taeckholmia are 1.6% and 1.0%, respectively. Average sequence divergence between subg.
Dendrosonchus (excluding Taeckholmia) and other genera in the alliance varies from 2.5% to 3.4%.
Results of the relative rate tests indicate no significant differences for any of the lineages, and thus the molecular clock can not be rejected.
Phylogenetic analyses of ITS sequences. Following the previous phylogenetic study of the Sonchinae, which suggests the monophyly of the woody Sonchus alliance in Macaronesia
(Fig. 3), additional phylogenetic analyses of ITS sequences with a much broader sampling within the group were carried
49 out. The heuristic search option for 35 accessions found
8123 equally most parsimonious trees, one of which is shown
in Fig. 4. This tree is identical to the 50% majority-rule
consensus tree and suggests that Sventenia. Babcockia.
Prenanthes pendula. and Sonchus tuberifer are in turn sister
groups to subg. Dendrosonchus sensu Aldridge. It also
suggests that Lactucosonchus. an endemic genus from the
island of La Palma (Fig. 2b), is sister to the remaining
members of the woody Sonchus alliance (Fig. 4).
DISCUSSION
Origin and evolution of the woody sonchus alliance in
Macaronesia. The molecular data strongly confirm and extend
considerably our previous preliminary work and indicate
that, despite the extensive morphological and ecological
diversity of the plants and the geographical proximity of
the Macaronesian islands to a continental source area, the
entire alliance was derived from a single colonization event
(Fig. 3, 4). Both bootstrap and decay analyses support
strongly the monophyly of the group. The presence of
intergeneric hybrids (32, 33) and results of crossing experiments (A. Aldridge, unpublished data) likewise suggest genetic cohesiveness despite the considerable morphological diversity of the alliance. These results are similar to
50 those for other insular groups such as the genus
Arovrantheitium in the Macaronesian islands (43) , and the
silversword alliance (10, 34) and Tetramolopium (44) in
Hawaii where highly fertile interspecific hybrids can be
produced despite pronounced morphological differences.
Within the woody Sonchus group, the ITS tree suggests that
Lactucosonchus diverged first, followed by the radiation of the four other genera, as well as subg. Dendrosonchus sensu
Aldridge (Fig. 4). It is also likely that, based on the previous study of the Sonchinae (28, Fig. 3), the entire group was derived from a single dispersal event from a more widely distributed European taxon, such as S. palustris. The low average sequence divergence and the polytomy in the ITS tree suggest a rapid radiation of major lineages early in the history of the alliance after a single introduction.
Adaptive radiation connotes the process by which a monophyletic group of organisms adapts to a broad diversity of habitats (3, 4), and it has been of major importance in the evolution of woody Sonchus and its close relatives. The
ITS phylogeny suggests that eight ecological shifts from mesic to dry habitats and three shifts from dry to coastal habitats have occurred during the diversification of the group (Fig. 4) . Two lineages of Taeckholmia (excluding two species of Sonchus) have radiated exclusively into dry habitats, whereas a major clade of Sonchus has radiated primarily into mesic habitats with two ecological shifts to
51 dry habitats. The ITS phylogeny also indicates that, during
adaptive radiation, rosette-shrub/subshrub and rosette-tree
habits have been quite successful in colonizing and adapting
to the various habitats of the islands. In contrast, two
lineages with tubers have only one species each, i.e., S.
tuberifer and Lactucosonchus. and have not been successful
in radiating and speciating in the Macaronesian islands
(Fig. 4).
The ITS phylogeny also provides evidence that inter
island colonization events to similar ecological habitats
and intra-island differentiation have played important roles
in the evolution of the alliance (Fig. 4). Note, for
example, that S. canariensis occurs on two islands, as do S. hierrensis, S. acaulis. S. cooestus. and T. pinnata as well
as other species (Fig. 4) . Further, long distance dispersal has been responsible for the origin of Sonchus species in other archipelagos (Fig. 4). For example, S. daltonii in the
Cape Verde archipelago likely results from a recent dispersal event from the western Canary Islands. Also, three species of Sonchus in Madeira appear to be derived from a single colonization event, probably from Tenerife or Gran
Canaria of the Canary Islands (Fig. 4).
Biogeographical implications. The molecular data suggest that, after the initial divergence of Lactucosonchus from a common ancestor, the two genera Sventenia and
52 Babcockia as well as Prenanthes pendula and S. tuberifer
each radiated early in the Canary Islands. These taxa are
confined to mountains or mountain cliffs of the geologically
oldest areas of Gran Canaria and Tenerife (with the
exception of P. pendula which is also locally common in the
south and west of Gran Canaria) , and suggests that the woody
Sonchus group arose in the Canary Islands. Further, it
indicates that subg. Dendrosonchus and Taeckholmia probably
originated in the geologically oldest areas of Gran Canaria
or Tenerife; remarkably these two islands are also the
center of diversity for subg. Dendrosonchus and Taeckholmia
(45) .
Because the molecular clock cannot be rejected, it is
possible to estimate the time of the radiation of the woody
Sonchus alliance. The genus Dendroseris. which is endemic to
the Juan Fernandez Islands, was used to estimate the rate of
nucleotide substitution in ITS because both cpDNA
restriction sites (46) and ITS sequence data (28) suggest that it is closely related to the Macaronesian group. In addition, Dendroseris is an insular endemic and has life
forms similar to subg. Dendrosonchus (i.e., paliform and rosette trees to rosette shrubs). Thus, generation-time effect on the substitution rate of the ITS sequences should be minimized (16). Two different average rates of nucleotide substitutions per site per year in ITS in the genus
Dendroseris. r = (3.94 ± 0.10) x 10'® and (6.06 ± 0.15) x 10'
53 were used to estimate the time of radiation. The slower
rate assumes radiation shortly after the formation of the
island of Masatierra (one of the Juan Fernandez Islands)
whereas the faster rate assumes a later radiation and was
estimated from cpDNA divergence. The slower rate for the
sequences using Dendroseris as the standard was 0.78% per
million years (11) . The average sequence divergence between
Lactucosonchus and the other Macaronesian taxa is 3.34%,
indicating that the divergence between them may have
occurred 4.2 million years ago (Mya) . The average sequence divergence between subg. Dendrosonchus (including
Taeckholmia) and allied genera is 2.8%. Thus, the divergence of subg. Dendrosonchus from the other genera may have occurred approximately 3.6 Mya. Therefore, the origin of genera in the woody Sonchus alliance may have taken place about 4.2 million years ago or earlier on the Canary
Islands, and the radiation of subg. Dendrosonchus accordingly took place between 4.2 and 3.6 Mya on Gran
Canaria or Tenerife. However, if we use the faster rate, i.e., 1.20% per million years, then the origin of the alliance and the radiation of subg. Dendrosonchus may have occurred 2.8 Mya or earlier and 2.3 Mya, respectively.
Although all the colonization events of Sonchus species in the Macaronesian islands postdate the geological origin of the islands (47), the calculated time of occurrence of
Lactucosonchus in La Palma predates the origin of the island
54 (not older than 1.5 Mya, Fig. 2b). This suggests that the
immediate ancestor of Lactucosonchus evolved for some time on another island in the Canaries, either Tenerife or Gran
Canaria, followed by extinction of Lactucosonchus or its ancestor on such a source island.
The estimated divergence times for the allied genera and the radiation of subg. Dendrosonchus are long after the formation of the Canary Islands, except the two westernmost ones. La Palma and El Hierro. Even assuming some uncertainty in dating the ages of the older islands, it seems highly likely that their ages are still considerably older than the calculated time of radiation for Sonchus and its relatives.
One question then is why Sonchus was so successful if radiation occurred after the islands were several million years old and presumably open habitats were not plentiful.
One hypothesis is that mass extinctions in the Canaries and northwestern Africa during the first glaciation in the
Northern Hemisphere (2.8 Mya; 48) and the beginning of
Sahara desertification (2.5 Mya; 49) may have provided many open habitats for the radiation of the alliance. The estimated divergence times, using the faster mutations rate for Dendroseris in the calculations, coincide closely with glaciation and desertification.
The origin of the woody Sonchus in Macaronesia and insular woodiness. The origin of woody Sonchus in
55 Macaronesia is controversial. Boulos (50) proposed that
subg. Dendrosonchus evolved from subg. Oriaosonchus (endemic
to Africa), which he considered the most primitive group of
Sonchus. This hypothesis is supported by pollen morphology
(51). In contrast, Aldridge (45) suggested that subg.
Dendrosonchus is very primitive and that the two subgenera
Oriaosonchus and Sonchus were derived from it. Her
hypothesis is congruent with Takhtajan's view (52), and also
agrees with Bramwell (24, 25). Neither of these two hypotheses is supported by the ITS phylogeny. The ITS phylogeny indicates that subg. Dendrosonchus is a relatively derived group but was not derived directly from subg.
Oriaosonchus (Figs. 3, 4). Rather, it is more likely that it originated from an ancestor somewhat like one of its allied genera in the Canary Islands (Fig. 4).
The woody life-forms in the Canaries in genera such as
Sonchus. Echium. Arayranthemum. Pericallis. and Crambe have been considered by several authors to represent Tertiary relicts (6, 19, 24, 53-55). Bramwell (24) suggested that taxonomic, cytological, morphological, palaeobotanical, distributional, and phytogeographical data are consistent with the endemic flora being of considerable age and probably ancestral to many modern Mediterranean genera and species. Meusel (53) viewed the woody Macaronesian species as ancestral forms of modern Mediterranean herbaceous species, with the herbaceous forms derived by reduction in
56 lignification and adaptation to more extreme conditions. In contrast, Carlquist (2-4) argued that the endemic,
frutescent species found on many oceanic islands are the result of an increase in woodiness in response to the uniformity of insular climates. More specifically, Carlquist
(3) claimed that a group of true shrubby Sonchus in the
Macaronesian Islands evolved from plants similar to the
European weedy sow thistle (S. oleraceus) . Thus, he did not consider these plants to be relicts but rather secondary derivatives of herbaceous ancestors. The ITS tree clearly shows that neither subg. Sonchus nor subg. Oriqosonchus was derived from the woody members of Sonchus in Macaronesia
(Fig. 3) . Rather, it is evident that the woody Sonchus and its relatives were derived ultimately from continental herbaceous perennial ancestors (Fig. 4). Further, the ITS tree shows a general trend toward increased woodiness. For example, it is likely that the ancestor of the entire alliance was an herbaceous perennial, and then there was evolution towards caudex perennials, shrubs, and trees in different lineages during radiation in the Macaronesian islands (Fig. 4). In particular, it seems reasonable to assume that the immediate ancestors of the three Madeiran
Sonchus species and the lineage consisting of most of the woody Sonchus species were probably either herbaceous or caudex perennials, and then became trees after dispersal to
Madeira and various Canary Islands (see asterisked clade in
57 Fig. 4) . This trend is a well-known feature of adaptive
radiation in many insular plants (4) . Thus, the molecular
data do not support subg. Dendrosonchus being relictually
woody plants.
In conclusion, the woody Sonchus alliance in the
Macaronesian islands apparently resulted from a single
introduction. This founder event was likely followed by
several radiations. The ITS phylogeny refutes the relictural
nature of the woody life-form of Sonchus in Macaronesia and
suggests that it represents secondary derivation from
herbaceous ancestors. The ITS sequence data also suggest
that evolution of the woody Sonchus alliance may have
occurred during Late Tertiary (i.e., the late Miocene or early Pliocene) from Gran Canaria or Tenerife in the Canary
Islands. The rapid radiation of major lineages in the alliance is consistent with an unresolved polytomy at the base of the cladogram and low sequence divergence.
ACKNOWLEDGMENTS
We thank Aguedo Marrero, Pedro Ortega-Machin, and
Francisco Jose Gonzalez Artiles for assistance during a field trip in the Canary Islands. Also, Instituto Canario de
Investigaciones Agrarias in Tenerife, through the advice of
Manuel Fernândez-Galvan, provided financial support for
58 field studies to one of us (JFO) . We are indebted to Charles
Jeffrey and Kâre Bremer for helpful suggestions about outgroups, and are grateful to Robert Jansen, Eric Knox, Tom
Myers, David Glenny, P.J. Garnock-Jones, Loutfy Boulos, and other people for providing plant and DNA materials. We thank
Angela Aldridge for her helpful discussion about this project. We are especially indebted to Sherwin Carlquist for many helpful comments and suggestions on an earlier version of this paper. This work was supported by Sigma Xi Grants- in-Aid of Research, Graduate Student Research Grant from
American Society of Plant Taxonomists, Tinker Foundation
(Latin American Studies Program, Ohio State University) ,
Janice Beatley Herbarium Award (OS), and National Science
Foundation Doctoral Dissertation Improvement grant DEB-
9521017 to DJC and KSC.
LITERATURE CITED
1. Carlquist, S. (1965) Island Life (The Natural History
Press, New York).
2. Carlquist, S. (1970) Hawaii: A Natural History (The
Natural History Press, New York).
3. Carlquist, S. (1974) Island Bioloav (Columbia University
Press, New York).
4. Carlquist, S. (1995) in Hawaiian Bioaeoaraphv: Evolution
59 on a Hot Spot Archipelago. eds. Wagner, W. & Funk V.
(Smithsonian Inst. Press, Washington D.C.) pp. 1-13.
5. Lack, D. (1947) Darwin's Finches (Cambridge University
Press, New York).
6. Bramwell. D. (1975) Anal. Inst. Bot. Cavanilles 32, 241-
254.
7. Baldwin, B. G., Kyhos, 0. W., & Dvorak, J. (1990) Ann.
Missouri Bot. Gard. 77, 96-109.
8. Carr, G. D. & Kyhos, D. W. (1981) Evolution 35, 543-556.
9. Helenurm, K. & Ganders, F. (1985) Evolution 39, 753-765.
10. Baldwin, B. G., Kyhos, D. W., Dvorak, J. & Carr, G. D.
(1991) Proc. Natl. Acad. Sci. USA 88, 1840-1843.
11. Sang, T., Crawford, D. J., Kim, S. -C. & Stuessy, T. F.
(1994) Amer. J. Bot. 81, 1494-1501.
12. Sanders, R. W., Stuessy, T. F., Marticorena, C. & Silva,
O. M. (1987) Oner. Bot. 92, 195-215.
13. Givnish, T. J., Sytsma, K. J., Smith, J. F. & Hahn, W.
J. (1994) Proc. Natl. Acad. Sci. USA 91, 2810-2814.
14. Sang, T., Crawford, D. J., Stuessy, T. F. & Silva O. M.
(1995) Svst. Bot. 20, 55-64.
15. Baldwin, B. G. (1992) Mol. Phvloaenet. Evol. 1, 3-16.
16. Soltis, P. S. & Soltis, D. E. (1995) in Evolutionarv
Biology, ed. Kecht, M. K. et al. (Plenum Press), pp.
139-194.
17. McDougall, I. & Schmincke, H. U. (1976-1977) Bull.
Volcanol. 40, 57-77.
60 18. Banda, E., Danobeitia, J. J., Surinach, E. & Ansorge, J.
(1981) Earth Planet Sci. Lett. 55, 11-24.
19. Bunding, P. (1979) in Plants and Islands, ed. Bramwell,
D. (Academic Press, New York) pp. 13-40.
20. Carracedo, J. C. (1994) J. Volcanol. Geotherm. Res. 53,
251-274.
21. Abdel-Monem, A., Watkins, N. D. & Cast, P. W. (1971).
Amer. J. Sci. 271, 490-521.
22. Abdel-Monem, A., Watkins, N. D. & Cast, P. W. (1972).
Amer. J. Sci. 272, 805-825.
23. Carlquist, S. (1962) Phvtomorphology 12, 30-45.
24. Bramwell, D. (1972) in Taxonomy. Phvtoaeoaraphv. and
Evolution, ed. Valentine, D. H. (Academic Press, New
York) pp. 141-159.
25. Bramwell, D. (1976) in Bioaeocrraphv and Ecology in the
Canary Islands. ed. Kunkel, G. (Dr. W. Junk Publisher,
Hague) pp. 207-240.
26. Bohle, U. -R., Hilger, H., Cerff, R. & Martin, W. F.
(1994) in Molecular Ecology and Evolution: Approaches
and Application, eds. Schierwater, G. P., Streit, B.,
Wagner, G. P. & DeSalle, R. (Bukhauser Verlag Basel,
Switzerland) pp. 391-403.
27. Bremer, K. (1994) in Asteraceae: Cladistics &
Classification (Timber Press, Oregon) pp. 157-201.
28. Kim, S. -C., Crawford, D. J. & Jansen, R. K. (1996)
Svst. Bot. 26:
61 29. Aldridge, A. (1977) Bot. Mac. 3, 41-59.
30. Aldridge, A. (1978) Bot. J. Linn. Soci. 76, 249-285.
31. Aldridge, A. (1976) Bot. Mac. 2, 25-57.
32. Pérez de Paz, J. (1976) Bot. Mac. 1, 51-65.
33. Hansen, A. & Bunding, P. (1993) Sommerfeltia 17, 1-295.
34. Baldwin, B. G. & Robichaux, R. H. (1995) in Hawaiian
Biooeoaraphv: Evolution on a Hot Spot Archipelago. eds.
Wagner, W. & Funk, V. (Smithsonian Inst. Press,
Washington D.C.) pp. 259-287.
35. Doyle, J. J. & Doyle, J. L. (1987) Phvtochem. Bull. 19,
11-15.
36. Maddison, W. P. & Maddison, D. R. (1992) MacClade.
Version 3.0 (Sinauer Associates, Inc., Massachusetts).
37. Swofford, D. L. (1993) PAUP, Phvloaenetic Analysis Using
Parsimony. Version 3.1.1 (Smithsonian Inst. Press,
Washington, D.C.).
38. Felsenstein, J. (1985) Evolution 39, 783-791.
39. Felsenstein, J. (1986-1993) PHYLIP, Phvlogenv Inference
Package. Version 3.52c (The University of Washington,
Seattle, WA).
40. Li, W. -H. & Tanimura, M. (1987) Nature 326, 93-96.
41. Wu, C. -I. & Li, W. -H. (1985) Proc. Natl. Acad. Sci.
USA 82, 1741-1745.
42. Baldwin, B. G., Campbell, C. S., Porter, J. M . ,
Sanderson, M. J., Wojciechowski, M. F. & Donoghue, M.
J. (1995) Ann. Missouri Bot. Gard. 82, 247-277.
62 43. Francisco-Ortega, J., Jansen, R. K., Crawford, D. J. &
Santos-Guerra, A. (1995) Svst. Bot. 20, 413-422.
44. Lowrey, T. K. (1995) in Hawaiian Bioqeoaraphv: Evolution
on a Hot Spot Archipelago. eds. Wagner, W. & Funk, V.
(Smithsonian Inst. Press, Washington D.C.) pp. 195-220.
45. Aldridge, A. (1979) in Plants and Islands, ed. Bramwell,
D. (Academic Press, New York), pp. 279-290.
46. Whitton, J., Wallace, R. S. & Jansen, R. K. (1995) Can.
J. Bot. 73, 1058-1073.
47. Mitchell-Thome, R. C. (1985) Bol. Mus. Mun. Funchal 37,
52-85.
48. Flint, R. R. (1971) in Glacial and Ouaternarv Geology,
(John Wiley & Sons, New York).
49. Williams, M. A. J. (1982) in Ouaternarv Environments in
Northern Africa, eds. Williams, M. S. J. & Adamsom, D.
A. (Balkema, Rotterdam) pp. 13-22.
50. Boulos, L. (1967) Bot. Not. 120, 95-108.
51. Saad, S. I. (1961) Pollen et Spores 3, 247-260.
52. Takhtajan, A. (1969) Flowering Plants; Origin and
Dispersal (Smithsonian Inst. Press, Washington D. C.).
53. Meusel, H. (1953) Flora 139, 333-393.
54. Takhtajan, A. (1978) The Floristic Region of the World
(Nauka, Leningrad, USSR).
55. Cronk, Q. C. B. (1992) Biol. J. Linn. Soc. 46, 91-103
63 CHAPTER 4
ADAPTIVE RADIATION AND GENETIC DIFFERENTIATION IN
THE WOODY SONCHUS ALLIANCE (ASTERACEAE: SONCHINAE)
IN THE CANARY ISLANDS
ABSTRACT
The woody Sonchus alliance consists of 19 species of
Sonchus and five closely related genera fBabcockia.
Sventenia. Taeckholmia. Lactucosonchus. Prenanthes pendula. and Sonchus tuber if er^ and is restricted primarily to the archipelago of the Canaries in the Macaronesian phytogeographical region. An enzyme electrophoretic study, including 13 loci, was conducted to assess genetic diversity within and divergence among species of the alliance. Nei's genetic identities (distances) between genera (and subgenera) show a wide range of values from 0.490 (0.714) to
0.980 (0.013), and pairwise comparisons of all populations show relatively high genetic identities, with a mean of
0.804. The high identities further support the genetic cohesiveness of the alliance and its single origin on the
64 Macaronesian islands. Species in the alliance also shows
about 50% higher total genetic diversity (H^) than the mean
for other oceanic endemics. This study also suggests that
there is greater divergence between endemics on the older
islands or between species on the older islands compared to
those on younger islands, which suggests that time is a
factor for divergence at allozyme loci. Furthermore,
populations on older islands have higher total genetic
diversities and lower identities than conspecific
populations on younger islands. These resluts imply early
colonization, radiation, and divergence of the woody Sonchus
alliance on older islands followed by subsequent
colonization to younger islands. The taxonomic distribution
of alleles in the alliance indicates lineage sorting also
played a role in divergence among species. Lineage sorting may also produce nonconcordance with either taxonomic
designation or the pattern of variation obtained from other molecular markers such as ITS sequences of nrDNA. Timing for the origin and radiation of the alliance agrees with the estimate based on ITS sequences, and suggests that the early divergence and rapid radiation apparently took place during the Late Tertiary on either Gran Canaria or Tenerife.
65 INTRODUCTION
The woody Sonchus alliance (Asteraceae; Sonchinae), endemic to the Macaronesian islands, is composed of 19 species of primarily woody members of Sonchus (subg.
Dendrosonchus), seven species of Taeckholmia, one species of subg. Sonchus (S. tuberifer), three monotypic genera
Babcockia. Lactucosonchus. Sventenia. and one species of
Prenanthes. P. pendula (KIM & al. 1996a, b). The alliance has been regarded as an outstanding example of adaptive radiation among angiosperme in Macaronesia (ALDRIDGE 1975,
1979). These taxa are all endemic to the Canary Islands
(except three species of Dendrosonchus in Madeira and S. daltonii in the Cape Verde archipelago) and display extensive morphological, ecological, and anatomical diversity (ALDRIDGE 1977, 1978). Despite this diversity, all taxa have a uniform chromosome number (i.e., n=9, 2n=18 ;
ARDÊVOL GONZALES & al. 1993), and the high fertility of frequent interspecific and intergeneric hybrids suggests genetic cohesiveness within the alliance (ALDRIDGE 1976;
PEREZ DE PAZ 1976; HANSEN & SUNDING 1985). DNA sequence data from the internal transcribed spacer region of nuclear ribosomal DNA (ITS of nrDNA) and noncoding region of chloroplast DNA (cpDNA) suggest that the considerable morphological and ecological diversity results from a single colonization event (KIM & al. 1996a, b; KIM S.-C., unpubl.).
66 These previous studies also hypothesized that the radiation
of the alliance took place in the Canary Islands during the
late Miocene or early Pliocene.
The Canary archipelago, where most of the alliance
occurs, is located in the Atlantic Ocean and consists of
seven islands (Fig. 5). These islands are of volcanic origin
and form an approximately linear chain (MCDOUGALL &
SCHMINCKE 1976-1977; BANDA & al. 1981). In contrast to
several remote archipelagos in the Pacific, such as the
Hawaiian, Galapagos, and Juan Fernandez Islands, the
Canaries have two unique features from a biogegraphical
perspective. The proximity of the islands to the African
continent (i.e., the eastern most island, Fuertuventura, is
only about 100 km distance from the west coast of Morocco)
suggests that colonizers could reach the islands easily, and thus multiple introductions may have occurred for some closely related taxonomic groups. In addition, the broad range of geological ages of the archipelago, from 0.8 to 20 million years (Mya), raises the possibility that some elements of the Canarian flora are much older in origin than others (FERNANDEZ-PALACIOS & ANDERSON 1993; FOSTER & al.
1993; CARRACEDO 1994).
There are numerous studies concerning patterns of genetic variation in insular endemics from the Pacific islands using isozymes as markers; Bidens (HELENURM &
GANDERS 1985) , Tetramolopium (LOWREY & CRAWFORD 1985) , and
67 the silversword alliance (WITTER & CARR 1988) from the
Hawaii; Dendroseris (CRAWFORD & al. 1987a) and Robinsonia
(CRAWFORD & al. 1992) from the Juan Fernandez Islands; and
Oossypium in Hawaii (WENDEL & PERCIVAL 1990) and the
Galapagos (DE JOODE & WENDEL 1992). This previous work found
generally low levels of genetic variation within and
divergence between taxa, with the exception of certain
species in the silversword alliance and Robinsonia. This low
diversity within species may result from founder events
associated with establishment in the islands and genetic
drift and inbreeding in small populations. Low allozyme
divergence among insular endemics has usually been
attributed to the young age of most taxa (i.e., recent
spéciation) (HELENURM & GANDERS 1985; LOWREY & CRAWFORD
1985; CRAWFORD & al. 1987a). These generalizations about
isozymes in island plants have come almost exclusively from
Pacific islands. Recent results from Macaronesia give results similar in some but different in other respects from
Pacific plants. For example, FRANCISCO-ORTEGA & al. (1996)
found high genetic identities between taxa (average of
0.893) in Arovranthemum (Asteraceae). a result similar to many species endemic to Pacific islands. However, they detected 50% higher allozyme diversity within populations
(0.098) in Arovranthemum than the mean total diversity (H-r=
0.064) for species endemic to other oceanic islands (DE
JOODE & WENDEL 1992) . Relatively high genetic diversity has
68 also been reported for other Macaronesian endemics in Avena
fPoaceae) (MORIKAWA & LEGGETT 1990), Chamaecvtisus
fFabaceae) (FRANCISCO-ORTEGA & al. 1992), Dactylis fPoaceae)
(SAHUQUILLO & LUMARET 1995) , and Tolpis (Asteraceae)
(FRANCISCO-ORTEGA & al., unpubl.).
One purpose of the present study was to determine how
the levels of isozyme variation within and divergence among
species in the woody Sonchus alliance compare to results
from other island endemics, both in the Pacific and
Atlantic. Another objective in this study is to ascertain whether similarity among species at allozyme loci
corresponds with proposed relationships based on morphological and ITS sequences of nrDNA.
MATERIALS AND METHODS
Plant material. A total 478 plants representing 49 populations of the 22 species was examined (Table 4). All accessions, except Sventenia, were from wild populations.
Three species of woody Sonchus from Madeira and S. daltonii from Cape Verde archipelago were not included in this study because very few plants were available and they were not from natural populations. Young leaves were collected from natural populations, placed on ice, and returned to the laboratory. Seeds from several capitula from individual
69 plants were also collected separately from natural populations, germinated, and grown in the greenhouse of the
Ohio State University. One progeny (about three weeks old) from each individual plant in nature was subjected to enzyme electrophoresis. Voucher specimens are deposited at the Ohio
State University Herbarium (OS).
Enzvme electrophoresis. Enzymes were extracted from young fresh leaves in a cold pestile and mortar using a buffer of 0.1 M tris-HCL (pH 7.5), 14 mM 2-mercaptoethanol,
1 mM EDTA (tetrasodium salt) , 10 mM MgClg, 10 mM KCL, and 5-
10 mg polyvinyIpolypyrro1idone per 0.5 ml of buffer
(GOTTLIEB 1981a). Three enzymes, glutamate dehydrogenase
(GDH, E.G. 1.4.1.2), asparate aminotransferase (AAT, E.G.
2.6.1.1), and alcohol dehydrogenase (ADH, E.G. 1.1.1.1), were resolved in polyacrylamide gels as described by
GRAWFORD & al. (1987a) . The following two buffer systems were employed to separate the remaining enzymes in 12% starch gels: (l) electrode buffer of 0.5 M tris, 0.65 M boric acid, 0.02 M EDTA, pH 8.0 diluted 1:9 for the gel buffer (phosphoglucomutase, PGM, E.G. 5.4.2.2; triose- phosphate isomerase, TPI, E.G. 5.3.1.1); (2) an electrode buffer of 0.04 M citric acid adjusted to pH 6.1 with N-(3- aminopropyl)-morpholine, the gel buffer a 1:19 dilution of the electrode buffer (malate dehydrogenase, MDH, E.G.
1.1.1.37; shikimate dehydrogenase, SKD, E.G. 1.1.1.25;
70 phosphogluconate dehydrogenase, PGD, B.C. 1.1.1.44).
Staining protocols for all enzymes followed WENDEL & WEEDEN
(1989) .
Data analysis. The genetic bases of the banding
patterns were inferred from variatin reported for other
plants (reviewed by WEEDEN & WENDEL 1989). These include the
active subunit composition of the enzymes and the minimal
number of loci for each enzyme found in diploid plants. In
all instances, only loci readily interpretable were scored,
and the most conservative interpretation was used.
Allelic frequencies were determined for each population, and these frequencies were used to calculate genetic identities and standard genetic distances for populations of each species, and for pairwise comparisions of the taxa (NEI 1972) . Polymorphism indices and gene diversity statistics were also calculated (NEI 1973, 1987).
The GeneStat-PC 3.3 (LEWIS 1993) package was used to calculate these statistics. A tree based on Nei's distance was constructed by the UPGMA (unweighted pair-group method using arithmetic averages) of the Phylip package (version
3.52c, FELSENSTEIN 1986-1993)
71 RESULTS
A total of 13 loci was resolved; Aat-1. Aat-2, Adh-1.
Adh-2. Gdh. Mdh-1. Mdh-2. Mdh-3 Pad-1. Pad-2. Pom-1, Skd,
Tpi. In all instances, the number of isozymes detected and scored was the same or fewer than expected for diploid plants (GOTTLIEB 1982; WEEDEN & WENDEL 1989). Additional
loci were expressed for most of the enzymes but were not
included because of weak activity and/or poor resolution of bands; these included one locus for LAP, two loci for IDH, and one additional locus for PGM. Two of the loci, Mdh-1 and
Mdh-2. were monomorphic. The number of alleles per locus varied from one for Mdh-l and Mdh-2 to seven for Pad-2. with an average of 3.5. A table of allelic frequencies for all populations was generated and is available from the first author upon request.
Nei's genetic identities and distances for the genera of the woody Sonchus alliance are shown in Table 5, and those for all conspecific populations and among all the species are available from the first author upon request.
Genetic diversity statistics, mean number of alleles per locus and per polymorphic locus, and proportion polymorphic loci are given in Table 6. Genetic identities (distances) between genera (or subgenera) of the alliance range from
0.490 (0.714) for Babcockia and Sventenia to 0.988 (0.013) for Lactucosonchus and Sonchus tuberifer (Table 5). About
72 70% of them are above 0.70, and 50% of the pairwise
comparisons are higher than 0.80. The mean genetic identity
for all populations of the alliance is 0.804. Mean genetic
identities (distances) for populations of each species vary
from 0.778 (0.265) for Sonchus pinnatifidus to 1.000 (0.000)
for S. acaulis. Pairwise comparisons of genetic identities
for all populations of Sonchus. subg. Dendrosonchus. range
from 0.445 to 1.000, with an average of 0.824. The genus
Taeckholmia shows slightly higher identities, ranging from
0.600 to 0.999, with a mean of 0.846.
Total genetic diversity (H^) within species of the
entire alliance varies from 0.000 in Sonchus bornmuelleri
and S. wildnretii to 0.415 in S. pinnatif idus. with an
average of 0.100. The diversity within populations (Hg) ranges from 0.000 in S. bornmuelleri and S. wildpretii to
0.253 in S. pinnatif idus. with a mean value of 0.067 for the entire alliance (Table 6) . The total diversity (H^) found for all populations of subg. Dendrosonchus and Taeckholmia are 0.206 and 0.165, respectively. The total diversity found for all populations of the alliance is 0.256. The number of unique alleles (u) found in subg. Dendrosonchus and
Taeckholmia was nine and three, respectively. Only one unique allele was found in the monotypic genus Babcockia.
Both Sonchus brachvlobus and S. pinnatifidus have two unique alleles, while S. conaestus and Taeckholmia oinnata each have one. Ggy values, which indicate proportion of total
73 diversity among populations, range from 0.009 (S. acaulis) to 0.773 (S. hierrensis).
The phenogram of all populations based on Nei's distance and the UPGMA method is shown in Figure 6. Four major groups may be recognized. Cluster 1 contains two populations of the monotypic genus Babcockia. endemic to
Gran Canaria, whereas cluster 2 has only Sonchus wildpretii from La Gomera. The large cluster 3 contains several single island endemics, inlcuding Sonchus palmensis. S. bornmuelleri. S. brachylobus. S. fauces-orci. Taeckholmia microcarpa. T. canariensis. T. capillaris. Prenanthes pendula, and monotypic genus Sventenia. Also, cluster 3 inlcudes one population of S. canariensis from Gran Canaria,
T. pinnata from Tenerife, and one population of S. pinnatifidus from Lanzarote. The largest cluster 4 consists of the monotypic genus Lactucosonchus. Sonchus tuberifer
(subg. Sonchus), and six species of subg. Dendrosonchus
(i.e., S. conqestus. S. ortunoi. S. hierrensis. S. acaulis.
S. qonzalezpadronii. S. gummifer). This cluster also includes all but one population of S. canariensis. two populations of T. pinnata from Tenerife and Gran Canaria, and one population of S. pinnatif idus from Lanzarote.
74 DISCUSSION
The woody Sonchus alliance shows quite high genetic
identities despite the considerable morphological and ecological diversity. The mean genetic identity for all populations of the alliance is 0.804, which is still higher than the mean of 0.65-0.70 reported for congeneric species of flowering plants as a whole (GOTTLIEB 1981b; CRAWFORD
1990). About 70% of pairwise comparisions of genetic identities between genera (or subgenera) are above 0.70 and
50% of them are higher than 0.80. These high values for the woody Sonchus alliance further support the genetic cohesiveness of the group and its single origin on the
Macaronesian islands. These results are congruent with high fertility and frequent interspecific and intergeneric hybrids within the alliance (ALDRIDGE 1976; PEREZ DE PAZ
1976; HANSEN & SUNDING 1985). Furthermore, this result is concordant with sequence data from ITS of nrDNA and noncoding region of chloroplast DNA, suggesting that the group is the result of a single introduction (KIM & al.
1996a, b; KIM & al., unpubl.)
In general, island plant populations are less variable genetically than mainland ones (HAMRICK & al. 1979; DE JOODE
& WENDEL 1992). Several studies from the Pacific, such as
Dendroseris. Tetramolopium. and Bidens. showed low allozyme variation (HELENURM & GANDERS 1985; LOWREY & CRAWFORD 1985;
75 CRAWFORD & al. 1987a), and this has been attributed to
genetic bottlenecks associated with long distance dispersal
and subsequent establishment on islands. Within an island
archipelago, additional founder events and genetic drift in
small populations could likewise reduce genetic variation by
loss of alleles. The breeding system, i.e., selfers and
inbreeding in small populations, could also be factors in keeping diversity low. Species of the woody Sonchus alliance
in the Macaronesian islands have approximately 50% more genetic variation than the mean for insular endemics, and the mean total diversity within each species of the alliance
(i.e., 0.100) is similar to that of endemics overall (see
Table 6; HAMRICK & GODT 1990; DE JOODE & WENDEL 1992). The total within-species diversity in the alliance is considerably higher than other Compositae from the Pacific islands, such as Bidens (HELENURM & GANDERS 1985) and
Tetramolopium (LOWREY & CRAWFORD 1985) , but it is only slightly higher than that of the silversword alliance
(WITTER & CARR 1988) (H^ = 0.075). The low genetic diversity in Bidens and Tetramolopium probably results from the small population sizes with highly localized distributions, and their inbreeding in small population and some degree in selfing (LOWREY 1986; CARR & al. 1986) . Some species of the silversword alliance are self-incompatible, abundant, and ecologically dominant species in their communities, thus contributing to high genetic diversity (WITTER & CARR 1988).
76 The high overall genetic variation in Sonchus compared
to endemics from the Pacific islands might at first glance
be attributed to the large population sizes and relatively
wide geographic distributions of several species. There is
not, however, a consistent correlation between population
size and genetic variability in the alliance. For example,
Sonchus acaulis. which is locally frequent in Gran Canaria
and widespread in forest and xerophytic zones of Tenerife,
has very low variation (Hy=0.030) (Table 6). In contrast, S.
canariensis. which is rare in both Tenerife and Gran
Canaria, has relatively high genetic diversity (Hy=0.092).
In some cases, however, species with wide geographical
distributions and large population sizes have high
diversity. For example, Taeckholmia pinnata. which is very
frequent with large population sizes in Tenerife and locally
frequent in Gran Canaria, has very high variation
(Ht=0.204). It is also likely that self-imcompability of the woody Sonchus alliance promotes high genetic diversity by obligate outcrossing (ALDRIDGE 1975). Total diversity in the alliance is comparable to that of Arcfvranthemum. the largest endemic genus in the Macaronesian islands. Francisco-Ortega
& al. (1996) found that average allozyme diverisity within populations of Arovranthemum (0.098) is 50% higher than the mean total diversity (all populations) for species endemic to oceanic islands (0.064), and that the total diversity for all populations of all species of Arovranthemum is 0.230,
77 which is similar to that of the woody Sonchus alliance
(0.256). High allozyme diversities have been reported for
other Macaronesian endemics from a wide range of taxonomic
families (including monocots and dicots) (MORIKAWA & LEGGETT
1990; FRANCISCO-ORTEGA & al. 1992; SAHUQUILLO & LUMARET
1995; FRANCISCO-ORTEGA & al. unpubl.). It is not known what
factors promote high diversities for Macaronesian endemics
compared to other oceanic endemics in general. Nevertheless,
the woody Sonchus alliance displays the same pattern of high
genetic diversities which seemingly characteristic of
Macaronesian endemics. Thus, the question of why the
Macaronesian endemics have higher diversity than many
endemics in other oceanic archipelagos has to be addressed.
Pairwise comparisons of genetic variation for
conspecific populations from geologically different islands
demonstrate that time could be a factor promoting genetic diversity in populations of the alliance. For example,
several populations of Sonchus canariensis from Gran Canaria
(ca. 14 Mya) have higher mean genetic variation (Hy = 0.071) than conspecific populations from the younger island of
Tenerife (ca. 11 Mya) (i.e., Hy = 0.051). This trend can also be seen in other species of Sonchus. such as S. acaulis. which is one of the most abundant species in the
Canaries. Populations on the older island of Gran Canaria have twice as much diversity (Hy = 0.043) as conspecific populations on the younger island of Tenerife (H^ = 0.021).
78 Furthermore, Tenerife populations of S. acaulis have a subset of alleles found in Gran Canaria populations at the loci Adh-1. Aat-1. and Aat-2. suggesting that Tenerife populations were colonized by geologically older Gran
Canaria populations.
The low levels of allozyme divergence among congeneric insular endemics despite morphological and ecological diversity have frequently been explained as the result of the combined effects of genetic bottlenecks associated with colonization, small population sizes, and recent spéciation on the relatively young islands (HELENURM & GANDERS 1985;
LOWREY & CRAWFORD 1985, CRAWFORD & al. 1987a, b). The accumulation of genetic differences between species can be explained by one or a combinations of two processes: the first is the divergence in frequencies of alleles shared due to common ancestry, and the second is the accumulation of newly arisen mutations among independent lineages (WITTER &
CARR 1988). The second process is presumably the slower of the two. If the ancestor(s) of insular endemics brought high allozyme diversity to the islands and little of it was lost during establishment, then divergence could occur through changes in allelic frequencies during lineage divergence and spéciation. By contrast, if the ancestor carried very little allozyme diversity to the islands or it was lost subsequent to establishment, then divergence could occur only through mutation. WITTER & CARR (1988) argued that the accumulation
79 of new mutations has been important in divergence observed
at allozyme loci among species of the silversword alliance
in the Hawaiian Islands because species in older islands
show higher divergence, and this is what would be expected
with the gradual accumulation of mutations with time. For
example, species of Dubautia on the oldest island of Kaui
have lower genetic identities (mean of ca. 0.69) than those
species on younger islands (0.95) in the archipelago.
Therefore, the Hawaiian silversword alliance offers a clear
illustration that allozyme divergence among genetically
depauperate taxa depends on the slow accumulation of new
genetic varition. However, this is not the case in the genus
Arovranthemum in Macaronesia (FRANCISCO-ORTEGA & al. 1996) .
Given the much older ages of certain Macaronesian Islands compared to those in the Pacific, and their wide range of geological ages, it might be expected that some species would be much older than others and, at least some of the taxa endemic to the older islands would be more divergent from taxa on younger islands. FRANCISCO-ORTEGA & al. (1996), however, failed to detect any evidence either of greater divergence between endemics on the older islands or between species on the older islands compared to those on younger islands. By contrast, the present study of Sonchus suggests early colonization, radiation and divergence on older islands followed by subsequent colonization to younger islands where spéciation has been much more recent. For
80 example, for those species of subg. Dendrosonchus sensu
ALDRIDGE (including Taeckholmia; ALDRIDGE 1975) there is a correlation between age of islands and genetic identities
(Fig. 5) . Those species endemic to the older island of Gran
Canaria (ca. 14 Mya) have a mean genetic genetic identity of
0.74, while the value is 0.77 for La Gomera (12.5 Mya) , 0.81 for Tenerife (ca. 11 Mya), and 0.85 for La Palma (1.5 Mya).
A similar correlation between island age and mean genetic identity is also revealed with other members of the woody
Sonchus alliance. For example, several taxa from Gran
Canaria, such as Babcockia. Sventenia. Prenanthes pendula. and Sonchus (including Taeckholmia), have the lowest mean identity of 0.66 for any island, while taxa from La Palma
(i.e. , Sonchus and Lactucosonchus ) have the highest mean identity (0.84) for any island. Two islands of intermediate age. La Gomera and Tenerife, show intermediate mean identities of 0.77 and 0.81, respectively (Fig. 1). This trend can also be seen in conspecific populations of S. canariensis and S. acaulis. which occur both in Gran Canaria and Tenerife. Several populations of S. canariensis from
Gran Canaria have lower mean identity (1=0.908, range of
0.85-0.99) than conspecific populations from the younger island of Tenerife (1=0.989, range of 0.98-0.99). However, in the case of S. acaulis. there is almost no difference in mean identity: 0.99 and 1.00 in Gran Canaria and Tenerife, respectively. These results suggest that time is a factor
81 for genetic differentiation during the radiation of the
Sonchus alliance in Macaronesia.
In contrast to Dubautia and certain other insular endemics, the woody Sonchus alliance may not have diverged at allozyme loci simply by the accumulation of mutations.
The taxonomic distribution of certain alleles suggests
lineage sorting during radiation. For example, Gdh* occurs in morphologically divergent taxa, such as most species of subg. Dendrosonchus sect. Dendrosonchus. the genera
Babcockia and Lactucosonchus. and Sonchus tuberifer (a member of subg. Sonchus) (BOULOS 1972) . In contrast, Gdh** occurs in all species of subg. Dendrosonchus sections
Pinnati and Brachvlobi. Prenanthes pendula, and most of
Taeckholmia. All species of the alliance except two are fixed for Adh-1**. while Adh-1* is fixed in the two morphologically distinct taxa Sonchus fauces-orci and
Babcockia. Furthermore, Aat-l** occurs in most of the alliance, while Aat-1** and Aat-1* occur in morphologically distinct pairs of taxa Sonchus brachvlobus and Sventenia. and Babcockia and Prenanthes. respectively. Lineage sorting could be responsible for these allellic distributions. The relatively high genetic identities (i.e., a mean of 0.804) and total diversity (Hy = 0.100) of the woody Sonchus alliance compared to other insular endemics together with the taxonomic distribution of alleles could be interpreted as evidence of ancestral allozyme polymorphisms in the
82 alliance.
As indicated in the above discussion, there are several cases of nonconcordance between allozyme data and sectional assignment. For example, all but three species belonging to sect. Dendrosonchus of subg. Dendrosonchus are grouped in cluster 4 with two morpholgically distinct species, S. canariensis (sect. Pinnati) and Taeckholmia pinnata (Fig.
6) . Cluster 4 also includes two distinct elements of the alliance, S. tuberifer (subg. Sonchus) and Lactucosonchus which are the only herbaceous perennials with tuberous roots. Cluster 3 contains all but two populations of
Taeckholmia. the sole member of sect. Brachvlobi. three species of sect. Dendrosonchus. and S. palmensis (sect.
Pinnati) together with one population of S. canariensis
(sect. Pinnati). This cluster also includes two Gran Canaria endemics, Prenanthes pendula and Sventenia (Fig. 6) .
Clearly, there is little concordance between allzoyme similarity and sectional assignment in cluster 3.
The UPGMA tree based on allozyme data can also be compared to the tree generated from ITS sequences of nrDNA
(KIM & al. 1996b, Fig. 4; KIM & al., unpubl.), and there are several major incongruences. For example, in the ITS phylogeny, Lactucosonchus diverges early in the alliance, and several other segregate genera such as Sventenia.
Babcockia, Prenanthes pendula. and Sonchus tuberifer. represent early radiations and are closely related to each
83 other. These genera are in turn sister groups to subg.
Dendrosonchus sensu ALDRIDGE (including Taeckholmia), which represents a second radiation (Fig. 3 of KIM & al. 1996b) .
However, the phenogram based on allozyme data shows
Babcockia distantly related from the rest of the alliance
(Fig. 6) . It also suggests that Sventenia and P. pendula from Gran Canaria are more closely related to S. brachvlobus. also from Gran Canaria. Lactucosonchus and S. tuberifer were grouped with other species from cluster 4
(Fig. 6). Also, in the ITS tree, two species of Sonchus. S. ortunoi and S. qonzalezpadroni from La Gomera (including S. qummifer from Tenerife in unpublished data, KIM & al.) were clustered with most species of Taeckholmia. except T. arborea. This clade is rather strongly supported by both bootstrap (98%) and decay values (3). In contrast, in the allozyme tree, these three species are rather closely related to each other, but are not grouped closely with any other species of Taeckholmia. All but two populations of
Taeckholmia are grouped in cluster 3 with two other species of Sonchus, S. palmensis and S. bornmuelleri. which have morphological characterstics toward Taeckholmia (Fig. 6) .
However, in the ITS tree, the two Sonchus species were clustered in another distinct lineage with most other species of Sonchus (KIM & al. 1996b; KIM & al., unpubl.).
These are some examples of nonconcordance between allozyme and other molecular data. The reasons are not known,
84 however, several possibilities can be mentioned. First, as
discussed above, populations in the Sonchus alliance have
relatively high genetic identities, suggesting rather recent divergence between populations of the same and different taxa. Repeated genetic bottlenecks associated with the founding of new populations during radiation of the alliance may result in lineage sorting of ancestral allozyme polymorphisms. Therefore, populations of different taxa could be more similar to each other than to populations of the same taxa. This could also explain why species from the same island are sometimes more similar than populations of the same species from a different island. Second, we do not know how extensive hybridization was during radiation of the group, but based on field observations and previously reported occurrence of spontaneous interspecific and intergeneric hybrids, hybridization may well have been (and is) extensive (SVENTENIUS 1960; ALDRIDGE 1975; HANSEN &
SUNDING 1985) . It is also likely that this may cause nonconcordance among morphology, allozyme data, and ITS sequences.
Timing the radiation of the alliance can be estimated based on allozyme data by using Nei's equation (1987) where time is equal to D/2a, with D the standard genetic distance and a the substitution rate per locus per year. If we assume the value of a is 10'^, then t may be calculated as (5 X
10®) D (CRAWFORD & al. 1992). The times for divergence for
85 all pairwise comparisions of species pairs range between
0.06 million years (Mya) and 3.57 Mya. These estimated times
postdate the origin of the Canary Islands, except for La
Palma and El Hierro. The earliest divergence within the
alliance can be estimated from taxa endemic to Gran Canaria and Tenerife (i.e., between 3.4 and 3.5 Mya) and this is
long after the formation of the islands (14 Mya and 11.6 Mya
in Gran Canaria and Tenerife, respectively; CARRACEDO 1994) .
These two islands are considered the center of diversity for woody members of Sonchus (subg. Dendrosonchus) including
Taeckholmia (ALDRIDGE 1979). Therefore, initial divergence in the alliance probably started as early as 3.6 mya ago when all the islands were formed, except La Palma and El
Hierro, and rapid radiation apparently took place during
Late Tertiary on Gran Canaria and Tenerife. The timing for the origin and radiation of the alliance agrees roughly with estimates based on ITS sequences; KIM & al. (1996b) suggested that the origin of the genera in the alliance may have occurred about 4.2 mya or earlier (2.8 Mya or earlier when a faster rate is assumed) on the Canary Islands, and the radiation of subg. Dendrosonchus accordingly took place between 4.2 and 3.6 mya (2.8 and 2.3 mya when a faster rate of ITS sequence evolution) ago on Gran Canaria or Tenerife.
Furthermore, the estimated divergence and radiation times coincide closely with glaciation in the Northern Hemisphere
(2.8 Mya; FLINT 1971) and beginning of Sahara
86 desertification (2.5 Mya; WILLIAMS 1982). These events may have caused mass extinctions in the Canaries and northwestern Africa providing many open habitats, and allowed rapid radition of the alliance.
ACKNOWLEDGMENTS
We wish to thank AGUEDO MRRERO, PEDRO ORTEGA-MACHIN,
FRACISCO JOSE GONZALEZ ARTILES, and several people from
Garajonay National Park (La Gomera) for assistance during the field work in the Canaries. Thanks also go to P. O.
LEWIS for help in using Gene Stat-PC 3.3 and MIN-JU KIM for assisting in preparation of figures. This study represents part of Ph. D. dissertation by the first author submitted to the Ohio State University. This work was supported by ASPT
Grants for Graduate Student Research, Tinker Foundation (The
Ohio State University), Sigma Xi Grants-in-Aid Research, and
Beatly Herbarium Award (OS) to S.-C. K, and National Science
Foundation Doctoral Dissertation Improvement Grant DEB-
9521017 to D. J. C. and S.-C. K.
LITERATURE CITED
ABDEL-MONEM. A., WATKINS. N. D. , GAST. P. W. , 1971:
Potassium-argon ages, volcanic stratigraphy, and
87 geomagnetic polarity history of the Canary Islands:
Lanzarote, Fuerteventxira, Gran Canaria, and La Gomera.
- Amer. J. Sci. 271: 490-512.
______1972: Potass ium-argon ages,
volcanic stratigraphy, and geomagnetic polarity history
of the Canary Islands: Tenerife, La Palma, and El
Hierro. - Amer. J. Sci. 272: 805-825.
ALDRIDGE. A. 1975: Taxonomic and anatomical studies in
Sonchus L. subg. Dendrosonchus Webb ex Schultz Bip. and
related genera. A Ph.D. dissertation of the University
of Reading, England, pp 1-467.
______1976: A critical reappraisal of the Macaronesian
Sonchus sugenus Dendrosonchus s.l. fCompositae-
lactuceae). - Hot. Mac. 2: 25-57.
______1977: Anatomy and evolution in the Macaronesian
Sonchus subg. Dendrosonchus (Compositae-Lactuceae^. Nodal and petiolar vascular patterns. - Hot. Mac. 3:
41-59.
1978: Anatomy and evolution in the Macaronesian
Sonchus subg. Dendrosonchus (Compositae-Lactuceae^. -
Hot. J. Linn. Soc. 76: 249-285.
1979: Evolution within a single genus: Sonchus in
Macaronesia. - In BRAMWELL. D., (Ed.) : Plants and
island, pp. 249-285. -London: Academic Press.
ANOCHEA. E. , FÛSTER. J. M. , IBARROLA. E. , CENDRERO. A.,
COELLO. J. , HERMAN. F. Ç. , CANTAGREL. J. M. , JAMOND. Ç.
88 1990: Volcanic evolution of th eisland of Tenerife (Canary Islands) in the light of new K-Ar data. - J.
Volcanol. Geotherm. Res. 231-249.
HERMAN. F., CENDRERO. A., CANTAGREL. J. M. , FÛSTER.
J. M., IBARROLA. E., COELLO. J., 1994: Constructive and
destructive episodes in the building of a young oceanic
island. La Palma, Canary Islands, and genesis of the
Caldera de Taburiente. - J. Volcanol. Geotherm. Res.
60; 243-262.
ARDÉVOL GONZALEZ. J. F. , BORGEN. L. , PÉREZ DE P ^ , P. L. ,
1993: Checklist of chromosome numbers counted in
Canarian vascular plants. - Sommerfeltia 18: 1-59.
BANDA. E. , DANOBEITIA. J, J. , SURINACH. E. , ANSORGE. J. ,
1981: Features of crustal structure under the Canary
Islands. - Earth Planet Sci. Let. M : 11-24.
BAVERSTOCK. P. R. , COLE. S. R. , RICHARDSON. B. J. , WATTS. Ç.
H. S., 1979: Electrophoresis and cladistics. - Syst.
Zool. 28: 214-219.
BOULOS. P. L., 1972: Révision systématique du genre Sonchus
L. s.l. I. Introduction et classification. - Bot.
Notiser 125: 287-305
CARR. G. D. , POWEL. E. A., KYHOS. D. W. , 1986: Self-
incompability in the Hawaiian Madiinae fCompositael: an
exception to Baker's rule. - Evolution 4^: 430-436.
CANTAGREL. J. M. , CENDREDO. A., FOSTER. J. M. , IBARROLA. E,
JAMOND. Ç . , 1984: K-Ar chronology of the volcanic
89 eruptions in the Canarian Archipelago: isalnd of La
Gomera. - Bull. Volconol. 47: 597-609.
CARRACEDO. J. Ç., 1994: The Canary Islands: an example of
structural control on the growth of large oceanic-
island volcanoes. - J. Vole, and Geoth. Res. 225-
241.
COELLO. J., CANTAGREL. J. M., HERNAN. F., FOSTER. J. M.,
IBARROLA. E., ANCOCHEA. E., CASOÜET. Ç. , JAMOND. Ç. ,
DIAZ DE TERAN. J. R., CENDRERO. A., 1992: Evolution of
the eastern volcanic ridge of the Canary Islands based
on new K-Ar data. - J. Volcanol. Geotherm. Res. 53;
251-274.
CRAWFORD. D. J., 1990: Plant Molecular Systematics:
Molecular approaches. John Wiley & Sons, New York, NY.
______STUESSY. T. F., SILVA O.. M., 1987a: Allozyme
divergence and the evolution of Dendroseris (Compositae- Lactuceae) on the Juan Fernandez
Islands. - Syst. Bot. 12.: 435-443.
______WHITKUS. R., STUESSY. T. F., 1987b: Plant evolution
and spéciation on oceanic islands. - In URBAKSKA. K . ,
(Ed.): Differentiation patterns in higher plants, pp.
183-199. - London: Academic Press.
______HAINES. D. W. , COSNER. M. B., SILVA O. .
M., P. LOPEZ.. 1992: Allozyme diversity within and
divergence among four species of Robinsonia
fAsteraceae: Senecioneae), a genus endemic to the Juan
90 Fernandez Islands, Chile. - Amer. J. Bot. 79; 962-966.
DE JOODE. D. R., WENDEL. J. F., 1992: Genetic diversity and
origin of the Hawaiian Islands cotton, Gossvnium
tomentosum. - Amer. J. Bot. 79: 1311-1319.
FELSENSTEIN. J., 1986-1993: PHYLIP, Phylogeny Inference
Package, Version 3.52c. University of Washington,
Seattle, WA.
FERNANDEZ-PALACIOS. J. M., ANDERSON. Ç., 1993: Species
composition and within archipelago co-occurrence
patterns in the Canary Islands. - Ecography 16: 31-36.
FLINT. R. F., 1971: Glacial and quaternary geology. John
Wiley and Sons, New York, NY.
FRANCISCO-ORTEGA. J. , JACKSON. M. T., CATTY. J. P., FORD-
LLOYD. B. V., 1992: Genetic diversity in the
Chamaecvtisus proliferus complex fFabaceae: Genisteael
in the Canary Islands in relation to in situ
conservation. - Genet. Resources Crop. Evol. 22: 149-
158.
______CRAWFORD. D. J. , SANTOS-GUERRA. A., CARVALHO. J.
A., 1996: Isozyme differentiation in the endemic genus
Arovranthemum (Asteraceae: Anthemideae) in the
Macaronesian Islands. - PI. Syst. Evol. 202: 137-152.
FOSTER. J. M., HERNAN. F., CENDERO. A., COELLO. J.,
CABTAGREL. J. M., ANCOCHEA. E. , IBARROLA. E., 1993:
Geochronology de la isla de El Hierro (Islas Canarias).
- Bol. Soc. Esp. Hist. Nat. 88: 1-4.
91 GOTTLIEB. L. D., 1981a: Gene number of species of Asteraceae
that have different chromosome number. - Proc. Natl.
Acad. Sci. USA 78: 3726-3729.
______1981b: Electrophoretic evidence and plant
populations. -Prog. Phytochem. 7: 1-46.
______1982: Conservation and duplication of isozymes in
plants. - Science 216: 3726-3729.
HAMRICK. J. L. , LINHART. Y. B. , MITTON. J. B. , 1979:
Relationships between life history characteristics and
electrophoretically detectable genetic variation in
plants. - Ann. Rev. Ecol. Syst. 10: 173-200.
______GODT, M. J. W., 1990: Allozyme diversity in plant
species. - In BROWN. H. D. (Ed. ) : Plant population
genetics, breeding, and genetic resources, pp. 43-63. -
Sunderland: Sinauer Associates.
HANSEN. A., SUNDING. P., 1985: Flora of Macaronesia.
Checklist of vascular plants. 3. revised edition. -
Sommerfeltia l: 26- 29.
HELENURM. K., GANDERS. F. R., 1985: Adaptive radiation and
genetic differentiation in Hawaiian Bidens. - Evolution
39: 753-765.
KIM. S.%C., CRAWFORD. D. J. , JANSEN. R. K. , 1996a:
Phylogenetic relationships among the genera of the
subtribe Sonchinae (Asteraceae): Evidence from ITS
sequences. - Syst. Bot. 21: (in press).
______FRANCISCO-ORTEGA. J. , SANTOS-GUERRA . A. ,
92 1996b: A common origin for woody Sonchus and five
related genera in the Macaronesian islands: Molecular
evidence for extensive radiation. - Proc. Natl. Acad.
Sci. USA 93: 7743-7748.
LEWIS. P., 1993: GeneStat-PC version 3.3. North Carolina
State University, Raleigh, NO.
LOWREY. T., 1986: A biosystematic revision of Hawaiian
Tetramo 1 opium fCompositae: Astereae) . - Allertonia 4:
203- 265.
______CRAWFORD. D. J., 1985: Allozyme divergence and
evolution in Tetramo lop ium ( Compos itae-Astereae ) on the
Hawaiian Islands. - Syst. Bot. 10: 64-72.
MCDOUGALL. I., SCHMINCKE. H. U., 1976-1977: Geochronology of
Gran Canaria, Canary Islands: age of shield building
volcanism and other magmatic phases. - Bull. Volcanol.
40: 57-77.
MORIKAWA. T., LEGGETT. J. M., 1990: Isozyme polymorphism in
natural populations of Avena canariensis from the
Canary Islands. - Heredity 403-411.
NEI. M., 1972: Genetic distance between populations. - Amer.
Naturalist 106: 283-293.
______1973: Analysis of gene diversity in subdivided
populations. - Proc. Natl. Acad. Sci. USA 70: 3 321-
3323.
______1987: Molecular and evolutionary genetics. Columbia
University Press, New York, NY.
93 PEREZ DE PAZ, J., 1976: Observaciones sobre la biologie y
relaciones de Sventenia bupleuroides F. Q. - Bot. Mac.
1: 51-65.
SAHUOUILLO. E., LUMARET. R., 1995: Variation in the
subrtropical group of Dactvlis alomerata L.- I.
Evidence from allozyme polymorphism. - Biochem. Syst.
Ecol. 23.: 407-418.
SVENTENIUS. E. R. S., I960: Additamentum ad floram
Canariensem. Madrid, Spain: Instituto Nacional de
Investigaciones Agronomicas Ministerio de Agriculture.
WEEDEN. N. F., WENDEL. J. F., 1989: Genetics and plant
isozymes. In SOLTIS. D. E., SOLTIS. P. E., (Eds):
Isozymes in plant biology, pp. 46-72. - Portland,
Dioscorides Press.
WENDEL. J. F., WEEDEN. N. F., 1989: Visualization and
interpretations of plant isozymes. In SOLTIS. D. E. ,
SOLTIS. P. E., (Eds): Isozymes in plant biology, pp. 5-45. -
Portland, Dioscorides Press.
______PERCIVAL. A. E., 1990: Molecular divergence in the
Galapagos Islands-Baja California species pair,
Gossvpium klotzschanum and G. davidsonii (Malvaceae). -
PI. Syst. Evol. 171: 99-115.
WILLIAMS. M. A. J. , 1982: Quaternary environments in
Northern Africa. In WILLIAMS. M. A. J., ADAMSOM. D. A.,
(Eds): A land between two Niles, pp. 13-22. -
Rotterdam, BaIkema.
94 WITTER. M. S., CARR. G. D., 1988: Adaptive radiation and
genetic differentiation in the Hawaiian silversword
alliance (Compositae: Madiinae) . - Evolution 1278-
1287.
95 CHAPTER 5
THE USE OF A NON-CODIHG REGION OF CHLOROPLAST DNA
IN PHYLOGENETIC STUDIES OF THE SUBTRIBE
SONCHINAE fASTERACEAE; LACTUCEAE^
ABSTRACT
The systematic utility of sequences from a non-coding region of chloroplast DNA (cpDNA) between osbA and trnH^°“°^ was examined by assessing phylogenetic relationships in subtribe Sonchinae (Asteraceae: Lactuceae). Primers constructed against highly conserved regions of tRNA genes were used for PCR amplification and sequencing. The psbA- trnH intergenic spacer contains several insertions and deletions (indels) in Sonchinae with the length varying from
385 to 450 bp. Sequence divergence ranges from 0.0% to 7.54% within Sonchinae. with an average of 2.4%. Average sequence divergence in Sonchus subg. Sonchus is 2.0%, while the mean for subg. Dendrosonchus and its close relatives in
Macaronesia (i.e., the woody Sonchus alliance) is 1.0%. Our results suggest that this region does not evolve rapidly
96 enough to resolve relationships among closely related genera
or insular endemics in the Asteraceae. The phylogenetic
utility of psbA-trnH sequences of the non-coding cpDNA was
compared to sequences from the ITS region of nuclear
ribosomal DNA. The results suggest that ITS
sequences evolve four times faster than psbA-trnH intergenic
spacer sequences. Furthermore, the ITS sequences provide
more variable and phylogenetically informative sites and
generate more highly resolved trees with more strongly
supported clades. Thus, ITS sequences are more suitable for
phylogenetic comparisons at lower taxonomic levels than
intergenic chloroplast sequences.
INTRODUCTION
Chloroplast DNA (cpDNA) has been used extensively to
investigate phylogenetic relationships at a wide range of taxonomic levels in plants. Chloroplast genes are now used routinely to infer phylogenies because direct sequencing of polymerase chain reaction (PCR) products makes it relatively easy to obtain sequence data. DNA sequences from several chloroplast genes, such as rbcL, matK. ndhF. atpB. and rps2 . have been used to estimate phylogenetic relationships at higher taxonomic levels. For example, rbcL has been widely used at the family level or above (PRICE & PALMER 1993;
97 SMITH & al. 1993; SCOTLAND & al. 1995; SOLTIS & al. 1995;
PLUNKETT & al. 1995). Several studies have also shown that
rbcL can be used at the generic and infrageneric levels in
some groups (GADEK & QUINN 1993; CONTI & al. 1993; PRICE &
PALMER 1993; SOLTIS & al. 1993; KRON & CHASE 1993; PAX & al.
1997) , but is sometimes too conserved to resolve
phylogenetic relationships between closely related genera
(DOEBLEY & al. 1990; GAUT & al. 1992; KIM K.-J. & al. 1992;
XIANG & al. 1993; SOLTIS & al. 1993; SMITH & al. 1993; KIM
Y.-D. & JANSEN 1996). The maturase encoding gene matK. which
is located in the intron of the transfer RNA gene for lysine
(trnK) . evolves faster (average of 2.5 times) than rbcL
(NEUHAUS fie LINK 1987; WOLFE & al. 1992; JOHNSON & SOLTIS
1995; XIANG Q.-Y., personal communication), and has been used successfully for generic-level phylogenetic reconstruction in several angiosperm families (JOHNSON fié
SOLTIS 1994; STEELE fie VILGALYS 1994). The ndhF gene, which encodes the ND5 protein of chloroplast NADH dehydrogenase, has also been shown to be useful for resolving phylogenies within and among plant families that have experienced recent and rapid radiation (KIM K.-J. fié JANSEN 1995; OLMSTEAD fié
REEVES 1995; CLARK fié al. 1995).
It is widely recongized that molecular phylogenetic studies should include multiple markers to assure that the gene trees are an accurate representation of the species phylogeny (DOYLE 1992). The most widely used markers in
98 plants are from chloroplast genes and the internal
transcribed spacer (ITS) regions of nuclear ribosomal DNA
(nrDNA) . Direct sequencing of the ITS has proven useful for
elucidating phylogenetic relationships at or below the
generic level (BALDWIN & al. 1995). Most cpDNA coding
regions do not evolve fast enough to resolve relationships
at these lower taxonomic levels (DOEBLEY & al. 1990; GAUT &
al. 1992). Several non-coding regions of cpDNA have
potential utility at lower taxonomic levels (GIELLY &
TABERLET 1994; HAM & al. 1994; MANEN & al. 1994; EHRENDORFER
& al. 1994; BOHLE & al. 1994; MANEN & NATALI 1995; NATALI & al. 1995; GIELLEY & TABERLET 1996; KIM J.-H. & al. 1996).
For example, GIELLY & TABERLET (1994) showed that introns and intergenic spacers evolve 1.93 to 11.72 times faster than rbcL and that insertions/deletions (indels) occur as often as nucleotide substitutions. Their later study (GIELLY fit TABERLET 1996) based on intron sequences from the chroloplast trnL gene showed that the ITS sequences have two to three times higher sequence divergence than the trnL intron. They concluded that ITS sequences are more appropriate for estimating phylogenetic relationships at the intrageneric level.
We sequenced the spacer between chloroplast genes psbA and trnH<°“°> to evaluate its phylogenetic utility at lower taxonomic levels in the Asteraceae. Phylogenetic trees based on ITS sequences for the Sonchinae (KIM S.-C. & al. 1996a,
99 b) enable direct comparisions of sequences of an intergenic
cpDNA spacer and the ITS region. In addition, the maternally
inherited chloroplast sequences provide an independent
assessment of phylogenetic relationships within subtribe
Sonchinae (Asteraceae: Lactuceae).
MATERIALS AND METHODS
Total genomic DNA was isolated from leaf tissue using
the CTAB method of DOYLE & DOYLE (1987), and purified in
CsCl/ethidium bromide gradients. Double-stranded DNAs of the non-coding region between psbA and trnH genes were amplified directly by 30 cycles of symmetric PCR using two primers.
The primers were designed with T. Sang (Michigan State
University) using the alignment of previously published sequences of dicots (Fabaceae. Brassicaceae. and Asteraceae;
SHAPIRO & TEWARI 1986; REITH & STRAUS 1987; AMBROSINI & al.
1992) . The primer sequences, which were used for both DNA amplification and sequencing, are: 5'-GTT ATG CAT GAA CGT
AAT GCT C-3' and 5'-CGC GCA TGG TGG ATT CAC AAT C-3'.
Methods for PCR amplification, purification of PCR products, and sequencing reactions are given in SANG & al. (1994) and
KIM S.-C. & al. (1996b). Sequences were aligned manually and this was feasible because the small and few large indels (up to 25 bp) present allowed unambiguous alignment. All
100 sequences were deposited in GenBank (Table 7) and the
aligned sequences are available from the first author upon
request. Complete sequences of the psbA-trnH intergenic
region were generated for 30 accessions representing nine
genera and 24 species of subtribe Sonchinae (from a total of
11 genera and about 130 species; BREMER 1993, 1994) and one
species of Taraxacum (Table 7) . Two species of Dendroseris
(subtribe Dendroseridinae of STEBBINS 1953) were also
sequenced because this genus was clearly nested within
subtribe Sonchinae in the ITS phylogeny (KIM S.-C. & al.
1996b) . Three species of Prenanthes were also sequenced to
determine the taxonomic position of P. pendula (see
Discussion).
Fitch parsimony was performed using PAUP version 3.1.1
(SWOFFORD 1993) with all changes weighted equally and
ACCTRAN, MULPARS, and TBR options. Multiple islands of
equally parsimonious trees (MADDISON 1991) were searched for
by performing 100 random entries in heuristic searches.
Prenanthes purpurea was used as an outgroup based on the
previous phylogenetic study of ITS sequences (KIM S.-C. &
al. 1996a, b) . Indels were treated as missing data.
Bootstrap analysis using HEURISTIC search, with simple addition, ACCRTAN, MULPARS, and TBR options, was conducted with 100 replicates (maxtree= 100) (FELSENSTEIN 1985) to evaluate the amount of support for monophyletic groups.
Decay analysis (BREMER 1988; DONOGHUE & al. 1992) was also
101 performed using same HEURISTIC search and trees up to five
steps longer were examined. Pairwise sequence divergences
were calculated by the Kimura two-parameter method using
PHYLIP version 3.52c (FELSENSTEIN 1986-1993). To compare the
phylogenetic utility of ITS and the psbA-trnH intergenic
spacer sequences in the Sonchinae. the original ITS data
matrix for 49 taxa of Sonchinae (KIM S.-C. & al. 1996a, b)
was reduced to 30 taxa (same taxa as in cpDNA data set) . The
same options were used to find the shortest trees for the
reduced ITS data set. Parsimony analysis was also performed
on the combined data sets. The same options were used to
find the shortest trees using the combined data.
RESULTS
Length and sequence divergence of psbA-trnH spacer. The
length of the intergenic spacer between psbA and trnH varies from 385 bp fReichardia picroides) to 450 bp (Sonchus palustris) in the subtribe Sonchinae. The aligned sequences are 506 bp and contain 23 indels (insertions/deletions). The direction of evolutionary change of indels was hypothesized by outgroup comparison. A total of 10 insertions and 13 deletions was detected; approximately 50% are shorter than four base pairs. The longest insertion of 25 bp occurred only in S. palustris. Of the seven putative insertions
102 longer than four base pairs, six were found to be direct
repeats or part of direct repeat motifs. A five base pair
deletion was shared by all three species of Reichardia. All
genera of Sonchinae involved in the main radiation, except
two basal genera Reichardia and Launaea. have a 15 bp
insertion.
Sequence divergence in the psbA and trnH intergenic
spacer ranged from 0.0 to 7.54% (R. oicroides and Emberaeria arandifolia) in the Sonchinae. with an average of 2.4%. No
sequence divergence was detected between 12 species pairs:
R. lioulata - R. tinaitana; S. asoer - S. kirkii; S. schweinfurthii - S. luxurians: S. acaulis - S. aonzalezpadroni; T. caoilaris - S. aonzalezoadroni; T. caoilaris - S. acaulis; Sventenia bupleuroides - T. capilaris; D. litoralis - T. caoilaris; D. litoralis -
Sventenia ; D. marainata - T. caoilaris; Sventenia - D. marainata; and D. litoralis - D. marainata. Mean sequence divergence within subg. Sonchus was 2.0%, while the woody
Sonchus alliance in Macaronesia was 1.0% (Table 9). Within
Reichardia. sequence divergence ranged from 0.0% to 3.5%, with an average of 2.3%.
Phvloaenetic analysis of the psbA-trnH spacer sequences. A total of 105 variable sites of the 506 aligned nucleotides (21%) was found in subtribe Sonchinae and the outgroup, with 48 of them (about 10%) phylogenetically
103 informative. Parsimony analyses found 8952 equally
parsimonious trees with a length of 118, a consistency index
(Cl) of 0.679 (excluding autapomorphies) and a retention
index (RI) of 0.787 (Fig. 7). These trees do not support the
monophyly of the Sonchinae as delimited by BREMER (1993,
1994) because Dendroseris and Prenanthes pendula are
embedded within the subtribe (Fig. 7) . The genus
Dendroseris. endemic to the Juan Fernandez Islands in the
Pacific, is closely related to the woody Sonchus alliance in
Macaronesia. There are no synapomorphies supporting the
monophyly of the woody Sonchus alliance in Macaronesia. The
cpDNA tree also suggests that Reichardia and Launaea are
basal within the Sonchinae. Both Sonchus as a whole and
Prenanthes are polyphyletic, whereas Reichardia and Sonchus
subg. Oriaosonchus are monophyletic.
Phvloaenetic analysis of the reduced ITS data (30 taxa)
and combined data sets. There are 252 variable sites (50%)
in the reduced ITS data set, 182 (36%) of which are
phylogenetically informative. Parsimony analysis found 90
equally parsimonious trees (not shown) with a length of 570
(Cl = 0.599, RI = 0.748). These trees, like the one based on the original data set of KIM S.-C. & al. (1996b), suggest that the Sonchinae of BREMER (1993) is not monophyletic, and that the clade including Reichardia and Launaea represent the basal split in the subtribe. The monophyly of the woody
104 Sonchus alliance in Macaronesia is still supported strongly
in the reduced ITS tree (bootstrap value of 99%, decay value
of >5). Finally, the reduced ITS phylogeny suggests that
Sonchus palustris is no longer the sole sister group to the
alliance, and that several Pacific genera, such as
Dendroseris. Kirkianella. and Emberaeria (i.e., clade B and
C of KIM & al. 1996b) , are more closely related to
Aetheorhiza and several species of subg. Sonchus (clade D)
than to the woody Sonchus alliance (clade A) . These
relationships are supported weakly by both bootstrap and
decay values.
Parsimony analysis of combined data sets (ITS and psbA-
trnH spacer sequence) was also conducted. A total of 360
equally parsimonious trees was found, one of which is shown
in Fig. 3. This tree is topologically almost identical to
the one based on the reduced ITS data set, except for the
relationships in the woody Sonchus alliance (i.e., clade
"A") .
A summary of ten parameters calculated for the four different data sets is shown in Table 8. The total number of
aligned sequences in the reduced ITS and psbA-trnH regions are nearly indentical; 501 and 506 in ITS and psbA-trnH, respectively. However, the ITS sequences provide approximately four times more variable sites (252 vs. 83) and phylogenetically informative positions (182 vs. 48) than the chloroplast intergenic spacer. Furthermore, the ITS
105 trees provide more resolved nodes and higher decay and
bootstrap values. All four data sets have high G, values,
indicating that there is a strong phylogenetic signal in the
data.
DISCUSSION
Phvloaenetic implications of c p DNA intergenic spacer sequences. Phylogenetic analysis of the psbA-trnH intergenic spacer provides several insights into relationships among genera of the subtribe Sonchinae. The results indicate that the Sonchinae as circumscribed by BREMER (1993, 1994) is not monophyletic, and that Dendroseris from the Juan Fernandez
Islands and Prenanthes pendula from the Canary Islands should be included in the subtribe. Both Dendroseris and P. pendula are clearly nested within the Sonchinae (Fig. 7) .
These results are in agreement with phylogenies generated from ITS (KIM S.-C. & al. 1996a, b) and cpDNA restriction site (WHITTON & al. 1995) data. Therefore, the
Dendroseridinae of STEBBINS (1953) should not be recognized as a distinct subtribe. The psbA-trnH spacer seguences were also used to examine the monophyly of the three subgenera of
Sonchus. The results suggest that subg. Oriaosonchus is monophyletic, whereas the subgenera Sonchus and
Dendrosonchus (sensu BOULOS 1972) are not (Fig. 7).
106 Furthermore, Sonchus is polyphyletic, as previously
suggested from the ITS phylogeny (KIM S.-C. & al. 1996b) .
The cpDNA tree (Fig. 7) indicates that Reichardia and
Launaea are basal in the Sonchinae and are not part of the main radiation of the subtribe. This relationship is congruent with the ITS tree (KIM S.-C. & al. 1996b) and also supported by sharing a 15 bp deletion in the cpDNA sequences. However, in the cpDNA tree Launaea is sister to the clade containing Reichardia species, and these two genera are sister to the remainder of the Sonchinae (Fig.
7) .
Two large lineages within the major radiation of
Sonchinae can be recognized in the cpDNA phylogeny. One includes Aetheorhiza. certain members of sections Sonchus and Asperi of subg. Sonchus. and two genera (Emberaeria and
Kirkianella) endemic to New Zealand. The other lineage includes African species of Sonchus. the woody Sonchus alliance from Macaronesia, and Dendroseris from the Juan
Fernandez Islands (Fig. 7).
In the ITS tree (Fig. 8), the New Zealand endemics
Emberaeria and Kirkianella are sister genera and are closely related either to sections Arvensis or Maritimi of subg.
Sonchus. The cpDNA tree also suggests that these two genera are closely related to sect. Maritimi. However, in the ITS tree the closest relatives of the New Zealand taxa and sections Arvenses and Maritimi (i.e., clade "C") was
107 suggested to be either Dendroseris (clade "B" of Fig. 8) or
Aetheorhiza and certain members of Sonchus (clade "D"). In contrast, the cpDNA intergenic spacer tree (Fig. 7) indicates that clade "C" is more closely related to clade
"D" (Aetheorhiza and several species of subg. Sonchus) than to Dendroseris (clade "B").
All woody members of Sonchus and allied genera in
Macaronesia appear to be derived from a common ancestor in the ITS tree (clade "A”, Fig. 8). Furthermore, S. palustris. which is widespread in Europe, is sister to the Macaronesian clade. Subgenus Oriaosonchus from Africa was sister to the rest of the clade that includes all of the Macaronesian taxa and Dendroseris. The monophyly of the woody Sonchus alliance in Macaronesia is supported strongly by bootstrap (100%) and decay values (more than five steps), while the sister group relationship between S. palustris and this alliance is supported weakly (42% bootstrap value, decay value of 2)
(Fig. 8). In contrast, S. palustris is sister to subg.
Oriaosonchus and this entire clade is sister to the woody
Sonchus alliance and Dendroseris in the cpDNA tree (Fig. 7).
The sister group relationship between S. palustris and subg.
Oriaosonchus is moderately supported (70% bootstrap value and decay index of 1).
Both ITS and cpDNA phylogenies support the close relationship of Aetheorhiza to sections Sonchus and Asperi
(subg. Sonchus. clade "D", Figs. 7 and 8) and also suggest
108 that Aetheorhiza originated after the Sonchus group diverged from a common ancestor with the Reichardia and Launaea lineage.
Prenanthes pendula. the only species in the genus endemic to the Canary Islands, was suggested by ITS data to be closest to the Macaronesian woody Sonchus alliance, especially to Sventenia and Babcockia (KIM S.-C. & al.
1996a, b) . In the cpDNA phylogeny (Fig. 7), P. pendula is clearly nested within a clade containing the woody Sonchus alliance and Dendroseris. The relationships indicated in the
ITS and psbA-trnH trees are congruent with a naturally occurring intergeneric hybrid between Sventenia and P. pendula was reported by SVENTENIUS (1960) and HANSEN &
SUNDING (1985) . The psbA-trnH spacer of cpDNA does not provide sufficient resolution to determine the phylogenetic relationship of P. pendula within the Macaronesian clade.
Nevertheless, the occurrence of natural hybrids and the placement of P. pendula in the Macaronesian clade in both nuclear and cpDNA trees suggests that P. pendula shares a more recent common ancestor with members of the woody
Sonchus alliance than it does with other members of
Prenanthes. Thus, Prenanthes as now delimited is not monophyletic. More detailed studies of Prenanthes. including
African (i.e., P. subpeltata) , Asian, and North American species, are needed for the circumscription of the genus.
The ITS phylogeny suggests that Dendroseris is closely
109 related to the two other Pacific taxa fKirkianella and
Emberaeria) and members of Sonchus subg. Sonchus (clade "C”)
(Fig. 8) , although support for this relationship is weak
(45% bootstrap value and dacay index of 1) . The cpDNA
phylogeny, however, suggests that Dendroseris is more
closely related to the woody Sonchus alliance and that the
Kirkianella-Emberqeria clade is more closely related to
clade "D" (Fig. 7) . Again, support for this group is very
weak (40% bootstrap and decay index of 1) .
Phvloaenetic analvsis of combined data sets. A
phylogenetic hypothesis for subtribe Sonchinae derived from
combining two data sets shows both incongruence and
congruence with those derived from independent data sets
(Fig. 9) . The four major clades (A-D) in the individual ITS
and psbA-trnH trees are present in the combined tree. The
woody Sonchus alliance from Macaronesia (clade A) forms a
strong monophyletic group (100% bootstrap value), and
further confirms a common origin for the morphologically and
ecologically diverse woody members of Sonchus and five
allied genera in Macaronesia (KIM S.-C. & al. 1996a).
However, the sister group relationship between S. palustris
and the woody Sonchus alliance, which was suggested by ITS
sequence data (Fig. 8) , was not present in the combined or
chloroplast trees (Figs. 7 and 9) . Two of the other three major clades (B and C) are supported strongly.
110 The tree based on combined chloroplast and ITS data has
a nearly identical topology to the ITS tree based on reduced
data. This may be the result of the fact that there are
three times as many phylogenetically informative characters
in the ITS region than in cpDNA intergenic spacer. In some
instances, phylogenetic analysis of the combined data
reinforces the relationships that are supported weakly in
the ITS phylogeny. For example, the clade that includes all
members of the woody Sonchus alliance except Lactucosonchus
has more character support (3 vs. 2) and higher bootstrap values (72% vs. 66%) in the combined tree than the ITS tree
(Figs. 8-9).
Phvloaenetic utilitv of the psbA-trnH interaenic spacer. The summary of tree statistics for the psbA-trnH spacer and the ITS region indicate that ITS sequences are more useful for resolving relationships in the Sonchinae
(Table 8). The ITS sequences produced fully resolved trees at the inter- and infrageneric levels, while many unresolved nodes (due to lack of synapomorphies) at infrageneric (e.g.,
Sonchus) or intergeneric levels of insular endemics (e.g., the woody Sonchus alliance in Macaronesia; Fig. 7).
Furthermore, ITS sequences produced fewer equally parsimonious trees (with slightly higher homoplasy) and thus resulted in trees with higher bootstrap and decay values.
There are approximately four times as many variable sites
111 and about three times as many informative sites in the ITS
sequences than in psbA-trnH sequences. This shows that the
non-coding sequences of nrDNA are more suitable than short
cpDNA non-coding sequences for resolving relationships in
subtribe Sonchinae.
Average sequence divergence of the intergenic spacer
between psbA and trnH is 2.4% (Tables 8 and 9) , whereas an
average of 9.9% sequence divergence was estimated for ITS
sequences (KIM S.-C. & al. 1996b) . This indicates that the
psbA-trnH intergenic spacer of cpDNA evolves approximately
four times slower than the ITS region of the nuclear
ribosomal DNA. In all cases, sequence divergence of non
coding regions of cpDNA is lower than in the ITS (Table 9) .
Our results are congruent with the GIELLY & al. (1996)
comparison between the trnL intron of cpDNA and nrDNA ITS in the genus Gentiana fGentianaceae) . They suggested that ITS
sequences have two to three times higher sequence divergence than trnL intron sequences. They also found that the ITS- based phylogeny displays higher bootstrap values and slightly higher homoplasy, suggesting that the trnL intron is more useful at the intergeneric level. These studies indicate that the substitution rate of intergenic spacer of cpDNA is comparable to that of cpDNA introns and that short non-coding sequences (e.g., about 500 bp) of cpDNA are more suitable for phylogenetic study at the generic rather than at the interspecific levels. Comparison of the sequences
112 from two non-coding regions of cpDNA indicates that the psbA-trnH intergenic spacer in Sonchinae evolves at least twice as fast as the trnL-trnF intergenic spacer in
Sonchinae (KIM S. -C., unpublished). The results of HAM & al. (1994) suggested that the trnL-trnF intergenic spacer is useful for studying relationships among distantly related genera. This region, however, seems not to evolve rapidly enough to resolve relationships among closely related genera or insular endemics in Asteraceae (KIM S.-C., unpublished).
In contrast, the psbA-trnH spacer seems to be better suited for inferring phylogenies between closely related genera.
However, this spacer alone is still not evolving rapidly enough to resolve phylogenetic relationships among recently evolved insular endemics.
Rates of nucleotide substitutions appear to vary considerably among different plant groups as well as among different non-coding regions of chloroplast DNA. At the infrageneric level, the psbA-trnH noncoding region seems to be useful for several genera (Table 9) . For example, the average sequence divergence of subg. Dendrosonchus in the
Macaronesian islands is approximately 0.9% and that of the woody Sonchus alliance is 1.0%. In subg. Sonchus. the pairwise sequence divergence ranges from 0.0 to 3.6%, with an average of 2.0%. These values are comparable to the average divergence for the trnT^"^") -trnL^"**' intergenic spacer of Echium species in Macaronesia (BOHLE & al. 1994) . In
113 subg. Sonchus. the psbA-trnH intergenic region evolves much
faster than the trnT-trnL or trnL-trnF non-coding regions.
However, this region alone may not be useful for resolving
relationships among closely related insular endemic groups
(Fig. 7), especially ones that apparently diverged rapidly
by adaptive radiation. Restriction site analysis of the
entire chloroplast genome are more suitable for congeneric
species (JANSEN & al. 1997) or radiated insular endemics
(FRANCISCO-ORTEGA & al. 1996) than are short intergenic
spacer sequences. Restriction site ananlyses of the entire
chloroplast genome can provide more information if a
sufficient number of restriction sites are examined using
frequent cutting enzymes (JANSEN & al. 1997).
In conclusion, the intergenic spacer between psbA and
trnH seems to be relatively well-suited for inferring
phylogenies at the intergeneric or in some instances at the
infrageneric levels in the Asteraceae. The small size of
this region requires the use of only two primers for
amplification and sequencing. Long stretches of As or Ts, which may complicate sequencing, are relatively rare
(GIELLEY & TAJBERLET 1994) . However, this region may not be
long enough to provide sufficent numbers of characters for resolving relationships among congeneric species or recently radiated groups in Asteraceae. Preliminary sequence comparisons should be performed before undertaking large-
114 scale taxonomic surveys due to the heterogeneity of
evolutionary rates of non-coding sequences of cpDNA. Other
intergenic spacers are currently being sequenced but
evaluation of their phylogenetic utility awaits further
study.
ACKNOWLDEGMENTS
We thank Javier FRANCISCO-ORTEGA, ARNOLDO SANTOS-
GUERRA, AGUEDO MARRERO, PEDRO ORTEGA-MACHIN, FRANCISCO JOSE
GONZALAEZ ARTILES, for assistance during the field work in the Canary Islands; ERIC KNOX, TOM MYERS, DAVID GLENNY, and
P. J. GARNOCK-JONES for providing plant and DNA materials.
Special thanks go to JAVIER FRANCISCO-ORTEGA for his continual encouragement and help, and introducing us to
Macaronesian plants during the course of this study. This study represents a partial fulfillment of Ph.D dissertation submitted by the first author to The Ohio State University.
This work was supported by a NSF Doctoral Dissertation
Improvement grant DEB-9521017 to D. J. C. and S.-C.K. and a
NSF grant (DEB-9318279) to R.K.J.
115 LITERATURE CITED
ALDRIDGE. A., 1975: Taxonomic and anatomical studies in
Sonchus L. subg. Dendrosonchus Webb ex Schultz Bip. and
related genera. A Ph.D. Dissertation of the University
of Reading, England, pp 1-467.
AMBROSINI. M. , CECI. L. R. , FIORELLA. S., GALLERANI. R. ,
1992: Comparison of regions coding for tRNA (His) genes
mitochondrial and chloroplast DNA in sunflower: a
proposal concerning the classification of 'CP-like'
tRNA genes. - Plant Mol. Biol. 20_: 1-4.
BALDWIN. B. G. , CAMPBELL. Ç. S., PORTER. J. M. , SANDERSON.
M. J., WOJCIECHOWSKI. M. F. , DONOGHUE. M. J. , 1995:
Utility of nuclear ribosomal DNA internal transcribed
spacer sequences in phylogenetic analysis of
angiosperms. - Ann. Missouri Bot. Gard. 247-277.
BOHLE. U. zR., HILGER. H. , CERFF. R. , MARTIN. W. F., 1994:
Non-coding chloroplast DNA for plant molecular
systematics at the infrageneric level. - In:
SCHIERWATER. B. , STREIT. B., WAGNER. G. P., DESALLE. R.
(Eds) Molecular Ecology and Evolution: Approaches and
Applications, pp. 391-403. - Switzerland: Bukjauser
Verlag Basel.
BOULOS. L. , 1972: Revision systématique du genre Sonchus L.
s.l. I. Introduction et classification. - Bot. Not.
125: 287-305.
116 BREMER. K., 1988: The limits of amino acid sequence data in
angiosperm phylogenetic reconstruction. - Evolution 42:
795- 803.
______1993 : New subtribes of the Lactuceae. - Novon 2:
328-330.
______1994: Asteraceae - Cladistic and classification. -
Oregon: Timber Press.
CLARK. L. G. , ZHANG. W . , WENDEL. J. F. , 1995: A phylogeny of
the grass family based on ndhF sequence data. - Syst.
Bot. 20: 436-460.
CONTI. E., FISCHBACH. A., SYTSMA. K. J. , 1993: Tribal
relationships in Onaaraceae: implications from rbcL
sequence data. - Ann. Missouri Bot. Gard. 80: 672-685.
DOEBLEY. J. , DURBIN. M. L. , GOLENBERG. E. M. , CLEGG. M. T.,
MA . D. P., 1990: Evolutionary analysis of the large subunit
of carboxylase frbcL) nucleotide sequence among the
grasses fGramineae). - Evolution 4^: 1097-1108.
DONOGHUE. M. J. , OLMSTEAD. R. G. , SMITH. J. F. , PALMER. J.
D. 1992: Phylogenetic relationships of Dopsacales based
on rbcL sequences. - Ann. Missouri Bot. Gard. 29: 333-
345.
DOYLE. J. J., 1992: Gene trees and species trees: molecular
systematics as one-character taxonomy. - Syst. Bot. 17:
144-163.
_ DOYLE. J. L. , 1987: A rapid DNA isolation procedure
for small quantities of fresh leaf materials. -
117 Phtochem. Bull. 19: 11-15.
EHRENDORFER. F. , MANEN. J. , NATALI. A., 1994: cpDNA
intergene sequences corroborate restriction site data
for reconstructing Rubiaceae phylogeny. - Pi. Syst.
Evol. 190: 245-248.
FELSENSTEIN. J., 1985: Confidence limits on phylogenies: an
approach using the bootstrap. - Evolution 39.: 783-791.
______1986-1993: PHYLIP: Phylogenetic Inference Package.
Version 3.5c. - Seattle, WA: The University of
Washington.
FRANCISCO-ORTEGA. J. , JANSEN. R. K. , SANTOS-GUERRA. A.,
1995: Chloroplast DNA evidence of colonization,
adaptive radiation, and hybridization in the evolution
of the Macaronesian flora. - Proc. Natl. Acad. Sci. USA
4085-4090.
GADEK. P. A., QUINN, Ç. J., 1993: An analysis of
relationships within the Cupressaceae sensu stricto
based on rbcL sequences. - Ann. Missouri Bot. Gard. 80:
581-586.
GAUT, B. S., MUSE. S. V., CLARK. W. D., CLEGG. M. T., 1992:
Relative rates of nucleotide substitutions for the
establishment of monocotyledonous plants. - J. Mol.
Evol. 15: 292-303.
GIELLY. L. , TABERLET. P., 1994: Chloroplast DNA polymorphism
at the intrageneric level: implications for the
establishment of plant phylogenies. - C. R. Acad. Sci.
118 Life Sci. 317; 685-692.
______1996: A phylogeny of the European gentians
inferred from chloroplast trnL^"*** intron sequences. -
Bot. J. Linn. Soc. 120; 57-75.
YUAN. Y. rM. , KUPFER. P., TABERLET. P., 1996:
Phylogenetic use of noncoding regions in the genus
Gentiana L.: Chloroplast trnL(UAA) intron versus
nuclear ribosomal internal transcribed spacer
sequences. - Mol. Phy. & Evol. 5; 460-466.
HAM. R. Ç. H. J. VAN, HART HI. T., MES, T. H. M. , SANDBRINK.
J. M., 1994: Molecular evolution of noncoding regions
of the chloroplast genome in the Crassulaceae and
related species. - Curr. Genet. 25: 558-566.
HANSEN. A., SUNDING. P., 1985: Flora of Macaronesia.
Checklist of vascular plants. 3. revised edition. -
Sommerfeltia 1: 26-29.
JANSEN. R. K., WEE. J. L., MILLIE. D., 1997: Comparative
utility of chloroplast DNA restriction site and DNA
sequencing data for phylogenetic studies in plants. -
In: SOLTIS, D., SOLTIS, P., DOYLE, J. J. (Eds)
Molecular Systematics of Plants, 2nd ed., -Chapman
Hall, New York. In press.
JOHNSON. L. A., SOLTIS. D. E. , 1994: matK DNA sequences and
phylogenetic reconstruction in Saxifraaaceae sensu
stricto. - Syst. Bot. 19.: 143-156.
______1995: Phylogenetic inference in
119 Saxifraaaceae sensu stricto and Gilla fPolvmoniaceae)
using matK sequences. - Ann. Missouri Bot. Gard. 82;
149-175.
KIM. K. zJ. , JANSEN. R. K. , WALLACE. R. S., MICHAELS. H. J. ,
PALMER. J. D., 1992: Phylogenetic implications of rbcL
sequence variation in the Asteraceae. - Ann. Missouri
Bot. Gard. 79: 428-445.
______1995: ndhF sequence evolution and the major
clades in the sunflower family. - Proc. Natl. Acad.
Sci. USA 92: 10379-10383.
KIM. J. rH. , HART. HI. T., T. H. M. , 1996: The
phylogenetic position of East Asian Sedum species
(Crassulaceae) based on chloroplast DNA trnL^"**^-
trnp<°^> intergenic spacer sequence variation. - Acta
Bot. Neerl. 45: 309-321.
KIM. S. %C. CRAWFORD. D. J. , FRANCISCO-ORTEGA. J. , SANTOS-
GUERRA. A. , 1996a: A common origin for woody Sonchus
and five related genera in the Macaronesian islands:
Molecular evidence for extensive radiation. - Proc.
Natl. Acad. Sci. USA 93: 7743-7748.
______JANSEN. R. K., 1996b: Phylogenetic
relationships among the genera of the subtribe
Sonchinae (Asteraceae) : Evidence from the ITS
sequences. - Syst. Bot. 21: 417-432.
KIM. Y. %D.y JANSEN. R. K., 1996: Phylogenetic implications
of rbcL and ITS sequence variation in the
120 Berberidaceae. - Syst. Bot. 21: 381-396.
KRON. K. A., CHASE. M. W - / 1993: Systematics of the
Ericaceae. Empetraceae. Epacridiaceae and related taxa
based upon rbcL sequence data. - Ann. Missouri Bot.
Gard. M : 735-741.
MADDISON. D. R., 1991: The discovery and importance of
multiple islands of most-parsimonious trees. -Syst.
Bot. 40: 315-328.
MANEN. J-F.. NATALI. A. EHRENDORFER. F. , 1994: Phylogeny of
Rubiaceae-Rubiae inferred from the sequence of a cpDNA
intergene region. - PI. Syst. Evol. 190: 195-211.
______1995: Comparison of the evolution of
ribulose-1,5-biphosphate carboxylase frbcL) and atpB-
rbcL non-coding spacer sequences in a recent plant
group, the tribe Rubiae (Rubiaceae). - J. Mol. Evol.
41: 920-927.
NATALI. A., MANEN. J-F.. EHRENDORFER. F., 1995: Phylogeny of
the Rubiaceae-Rubioideae. in particular the tribe
Rubieae: Evidence from a non-coding chloroplast DNA
sequences. - Ann. Missouri Bot. Gard. 8%: 428-439.
NEUHAUS. H. , LINK. G. , 1987: The chloroplast tRNA*-''® (UUU)
gene from mustard. - Curr. Genet. 11: 251-257.
OLMSTEAD. R. G., REEVES. P. A., 1995: Evidence for the
polyphyly of the Scrophulariaceae based on chloroplast
rbcL and ndhF sequences. - Ann. Missouri Bot. Gard. 82:
176-193.
121 PAX. D. L., PRICE. R. A., MICHAELS. H. J., 1997;
Phylogenetic position of the Hawaiian geraniums based
on rbcL sequences. - Amer. J. Bot. 72-78.
PEREZ DE PAZ. J., 1976: Observaciones sobre la biologia y
relaciones de Sventenia bupleuroides F. Q. - Botanica
Macaronesica 1: 51-65.
PLUNKETT. G. M. , SOLTIS. D. E. , SOLTIS. P. S., BROOKS. R.
E., 1995: Phylogenetic relationships between Juncaceae
and Cvperaceae: insights from rbcL sequence data. -
Amer. J. Bot. 82,: 520-525.
PRICE. R. A., PALMER. J. D., 1993: Phylogenetic
relationships of the Geraniaceae and Geraniales from
rbcL sequences comparisons. - Ann. Missouri Bot. Gard.
80; 661-671.
REITH. M. E., STRAUS. N. A., 1987: Nucleotide sequence of
the chloroplast gene responsible for triazine
resistance in canola. - Theor. Appl. Genet. 72: 357-
363.
SANG. T., CRAWFORD. D. J., KIM, S. %C., STUESSY. T. F. ,
1994: Radiation of the endemic genus Dendroseris
(Asteraceae) on the Juan Fernandez Islands: Evidence
from sequences of the ITS region of nuclear ribosomal
DNA. - Amer. J. Bot. 81: 1494-1501.
SCOTLAND. R. W. , SWEERE. J. A. , REEVES. P. A. , OLMSTEAD. R.
G., 1995: Higher-level systematics of Acanthaceae
determined by chloroplast DNA sequences. - Amer. J.
122 Bot. 82: 266-275.
SHAPIRO. D. R. , TEWARI. K. K., 1986: Nucleotide sequences of
transfer RNA genes in the Pisum sativum chloroplast
DNA. Plant Mol. Biol. 6: 1-12.
SMITH. J. F., KRESS. W. J. , ZIMMER. E. A., 1993:
Phylogenetic analysis of the Zinaiberales based on rbcL
sequences. - Ann. Missouri Bot. Gard. 8^: 620-630.
SOLTIS. D. E., MORGAN. D. R. , GRABLE. A., SOLTIS. P. S.,
KUZOFF. R., 1993: Molecular systematics of
Saxifraaaceae sensu stricto. - Amer. J. Bot. 8^: 1056-
1081.
SOLTIS. D. E., XIANG. O.-Y.. HUFFORD. L., 1995:
Relationships and evolution of Hvdranaeaceae based on
rbcL sequence data. - Amer. J. Bot. 8,2: 504-514.
STEBBINS. G. L. JR., 1953: Subfamilial nomenclature of
Compositae. - Taxon 12: 229-235.
STEELE. K. P., VILGALYS. R., 1994: Phylogenetic analyses of
Polemoniaceae using nucleotide sequences of the plastid
gene matK. - Syst. Bot. 19: 126-142.
SVENTENIUS. E. R. S., 1960: Additamentum ad floram
Canariensem. Madrid, Spain: Instituto Nacional de
Investigaciones Agronomicas Ministerio de Agriculture.
SWOFFORD. D. L., 1993: PAUP: Phylogenetic Analysis Using
Parsimony. Version 3.1.1. - Washington, D. C.:
Smithsonian Institution.
WHITTON. J., WALLACE. R. S., JANSEN. R. K . , 1995:
123 Phylogenetic relationships and patterns of character
change in the tribe Lactuceae (Asteraceae) based on
chloroplast DNA restriction site variation. - Can. J.
Bot. 72: 1058-1073.
WOLFE. K. H. , MORDEN. Ç. W. , PALMER. J. D. , 1992: Function
and evolution of a minimal plastid genome from a
monphotosynthetic parasitic plant. - Proc. Natl. Acad.
Sci. USA 98: 10648-10652.
XIANG. 0-Y. . SOLTIS. D. E. , MORGAN. D. R. , SOLTIS. P. S.,
1993: Phylogenetic relationships of Cornus L. sensu
lato and putative relatives inferred from rbcL sequence
data. - Ann. Missouri Bot. Gard. 80: 723-734.
124 LIST OF REFERENCES
ABDEL-MONEM, A., WATKINS, N. D. AND CAST, P. W. 1971. Potassion-argon ages, volcanic stratigraphy, and geomagnetic polatiry history of the Canary Islands : Lanzarote, Fuerteventura, Gran Canaria, and La Gomera. Amer. J. Sci. 271: 490-521.
ABDEL-MONEM, A., WATKINS, N. D. AND GAST, P. W. 1972. Potassion-argon ages, volcanic stratigraphy, and geomagnetic polarity history of the Canary Islands: Tenerife, La Plama, and Hierro. Amer. J. Sci. 272: 805- 825.
ALDRIDGE, A. 1975. Taxonomic and anatomical studies in Sonchus L. Subgenus Dendrosonchus Webb ex Schults Bip. and related genera. Ph. D. Dissertation. University of Reading, England.
ALDRIDGE, A. 1976. A critical reappraisal of the Macaronesian Sonchus subgenus Dendrosonchus s. 1. (Compositae-Lactuceae). Bot. Mac. 2: 25-57.
ALDRIDGE, A. 1976. Macaronesian Sonchus subgenus Dendrosonchus s.l. ( Compos itae-Lactuceae ) , including a reappraisal of the species concept and new combinations. Bot. Mac. 2: 81-93.
ALDRIDGE, A. 1977. Anatomy and evolution in the Macaronesian Sonchus subg. Dendrosonchus (Compositae-Lactuceae). Nodal and petiolar vascular patterns. Bot. Mac. 3: 41- 59.
ALDRIDGE, A. 1978. Anatomy and evolution in the Macaronesian Sonchus subgenus Dendrosonchus (Compositae-Lactuceae). Bot. J. Linn. Soci. 76: 249-285.
ALDRIDGE, A. 1979. Evolution within a single genus : Sonchus in Macaronesia. In Plants and Islands. ed. Bramwell, D., Academic Press, New York. pp. 279-290.
ALLAN, H. H. 1961. Flora of New Zealand, vol.1. Wellington, New Zealand: Government Printer.
125 AMBROSINI, M. , CECI, L. R. , FIORELLA, S., AND GALLERANI, R. 1992. Comparison of regions coding for tRNA (His) genes mitochondrial and chloroplast DNA in sunflower: a proposal concerning the classification of 'CP-like' tRNA genes. Plant Mol. Biol. 20: 1-4.
ANACOCHEA, E. , J. M. FUSTER, E. IBARROLA, A. CENDRERO, J. COELLO, F. HERNAN, J. M. CANTAGREL, AND C. JAMOND. 1990. Volcanic evolution of the island of Tenerife (Canary Islands) in the light of new K-Ar data. J. Volcanol. Geotherm. Res. 44: 231-249.
ANACOCHEA, E. , HERNAN, F. , CENDRERO, A., CANTAGREL, J. M. , FÛSTER, J. M., IBARROLA, E., AND COELLO, J. 1994. Constructive and destructive episodes in the building of a young oceanic island. La Palma, Canary Islands, and genesis of the Caldera de Taburiente. J. Volcanol. Geotherm. Res. 60: 243-262.
ARDÉVOL GONZALEZ, J. F. , BORGEN, L. , AND PÉREZ DE PAZ, P. L. 1993. Checklist of chromosome numbers counted in Canarian vascular plants. Sommerfeltia 18: 1-59.
BABCOCK, E. B. AND G. L. STEBBINS, JR. 1943. Systematic studies in the Cichorieae. University of California Publications in Botany 18: 227-240.
BALDWIN, B. G. 1992. Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: An example from the Compositae. Molecular Phylogenetics and Evolution 1: 3-16.
BALDWIN, B. G., KYHOS, D. W . , AND DVORAK, J. 1990. Chloroplast DNA evolution and adaptive radiation in the Hawaiian Silversword alliance (Asteraceae-Madiinae). Ann. Missouri Bot. Gard. 77: 96-109.
BALDWIN, B. G. 1993. Molecular phylogenetics of Calvcadenia (Compositae) based on ITS sequences of nuclear ribosomal DNA: Chromosomal and morphological evolution reexamined. American Journal of Botany 80: 222-238.
BALDWIN, B. G. , KYHOS, D. W. , DVORAK, J. AND CARR, G. D. 1991. Chloroplast evidence for a North American origin of the Hawaiian Silversword alliance (Asteraceae) . Proc. Natl. Acad. Sci. USA 88: 1840-1843.
BALDWIN, B. G . , C. S. CAMPBELL, J. M. PORTER, M. J. SANDERSON, M. F. WOJCIECHOWSKI, AND M. J. DONOGHUE. 1995. Utility of nuclear ribosomal DNA internal transcribed spacer sequences in phylogenetic analyses of angiosperms. Annals of the Missouri Botanical Garden
126 82: 247-277.
BALDWIN, B. G. AND ROBICHAUX, R. H. 1995. Historical biogeography and ecology of the Hawaiian Silversword alliance (Asteraceae) : New molecular phylogenetic perspectives. In Hawaiian Bioaeoaraohv: Evolution on a Hot Spot Archipelago. eds. Wagner, W. & Funk, V., Smithsonian Inst. Press, Washington D.C. pp. 259-287.
BANDA, E. , DANOBEITIA, J. J. , SURINACH, E. AND ANSORGE, J. 1981. Features of crustal structure under the Canary Islands. Earth Planet Sci. Lett. 55: 11-24.
BAVERSTOCK, P. R. , COLE, S. R. , RICHARDSON, B. J. , AND WATTS, C. H. S. 1979. Electrophoresis and cladistics. Syst. Zool. 28: 214-219.
BEUZENBERG, E. J. AND J. B. HAIR. 1984. Contributions to a chromosome atlas of the New Zealand flora - 27 Compositae. New Zealand Journal of Botany 22: 353-356.
BOHLE, U. -R. , HILGER, H. , CERFF, R. AND MARTIN, W. F. 1994. Non-coding chloroplast DNA for plant molecular systematics a the infrageneric level. In Molecular Ecoloov and Evolution: Approaches and Application, eds. Schierwater, G. P., Streit, B., Wagner, G. P. & DeSalle, R. , Bukhauser Verlag Basel, Switzerland, pp. 391-403.
BOULOS, L. 1960. Cytotaxonomic studies in the genus Sonchus. 2. The genus Sonchus. a general systematic treatment. Bot. Not. 113: 400-420.
BOULOS, L. 1965. BabcQckia. un nouveau genre de Composées des iles Canaries. Bull. Jard. Bot. Etat. 35: 63-66.
BOULOS, L. 1967. Nomenc 1 atura 1 changes and new taxa in Sonchus from the Canary Islands. Nytt Mag. Bot. 14: 7- 18.
BOULOS, L. 1967. Taeckholmia. a new genus of Compositae from Canary Islands. Botaniska Notiser 120: 95-108.
BOULOS, L. 1968. The genus Sonchus and allied genera in the Canary Islands. Cuad. de Bot. 3: 19-26.
BOULOS, L. 1972. Revision systématique du genre Sonchus L. s.l. I. Introduction et classification. Bot. Not. 125: 287-305.
BOULOS, L. 1972. Révision systématique du genre Sonchus L. s.l. I. Introduction et classification. Botaniska
127 Notiser 125: 287-305.
BOULOS, L. 1973. Revision systématique du genre Sonchus L. s.l. IV. Soud-genre I. Sonchus. Bot. Not. 126: 155-196.
BOULOS, L. 1974. Revision systématique du genre Sonchus L. s.l. V. Dendrosonchus. Bot. Not. 127: 7-37.
BOULOS, L. 1974. Revision systématique du genre Sonchus L. s.l. VI. Sous-genre 3.Oriaosonchus. genres Emberaeria. Babcockia. et Taeckholmia. Species exclusae dt dubiae. Index. Bot. Not. 127: 402-451.
BRAMWELL, D. 1972. Endemism in the flora of the Canary Islands. In Taxonomy. PhvtoaeocrraphY. and Evolution. ed. Valentine, D. H. , Academic Press, New York. pp. 141-159.
BRAMWELL. D. 1975. Some morphological aspects of the adaptive radiations of Canary Islands Echium species. Anal. Inst. Bot. Cavanilles 32: 241-254.
BRAMWELL, D. 1976. The endemic flora of the Canary Isalnds; Distribution, relationships, and phytogeography. In Bioaeooraphv and Ecology in the Canary Islands, ed. Kunkel, G., Dr. W. Junk Publisher, Hague, pp. 207-240.
BRAMWELL, D. 1985. Contribucion a la biogeografla de las islas Canarias. Botanica Macaronesica 14: 3-34.
BRAMWELL, D. AND Z. BRAMWELL. 1974. Wild flowers of the Canary Islands. London: Stanley Thornes (Publishiers) Ltd.
BREMER, K. 1988. The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42: 795-803.
BREMER, K. 1993. New subtribes of the Lactuceae. Novon 3: 328-330.
BREMER, K. 1994. Asteraceae-Cladistics and classification. Portland, Oregon: Timber Press.
CANTAGREL, J. M. , A. CENDRERO, J. M. FUSTER, E. IBARROLA, and C. JAMOND. 1984. K-Ar chronology of the volcanic eruptions in the Canarian Archipelago: Islands of La Gomera. Bull. Volcanol. 47: 597-609.
CARLQUIST, S. 1962. A theory of paedomorphosis in dicotyledonous wood. Phytomorphology 12: 30-45.
128 CARLQUIST, S. 1965. Island Life. The Natural History Press, New York.
CARLQUIST, S. 1970. Hawaii: A Natural History. The Natural History Press, New York.
CARLQUIST, S. 1974. Island Biology. Columbia Uniyersity Press, New York.
CARLQUIST, S. 1995. Introduction. In Hawaiian Bioaeooranhy: Eyolution on a Hot Spot Archipelago, eds. Wagner, W. & Funk V. Smithsonian Inst. Press, Washington D.C. pp. 1- 13.
CARR, G. 1985. Monographs of the Hawaiian Madiinae (Asteraceae): Argyranthemum. Dubautia. and Wilkesia. Allertonia 4: 1-123.
CARR, G. D. AND KYHOS, D. W. 1981. Adaptiye radiation in the Hawaiian Si Iyer sword alliance (Compositae-Madiinae) . I. Cytogenetics of spontaneous hybrids. Eyolution 35: 543- 556.
CARR, G. D. , POWEL, E. A., AND KYHOS, D. W. 1986. Self- incompability in the Hawaiian Madiinae (Compositae): an exception to Baker's rule. Evolution 40: 430-436.
CARRACEDO, J. C. 1994. The Canary Islands : an example of structural control on the growth of large oceanic- island volcanoes. J. Volcanol. Geotherm. Res. 53: 251- 274.
CLARK, L. G. , ZHANG, W. , AND WENDEL, J. F. 1995. A phylogeny of the grass family based on ndhF sequence data. Syst. Bot. 20: 436-460.
COELLO, J., J. -M. CANTAGREL, F. HERNAN, J. -M. FUSTERM E. IBARROLA, E. ANCOCHEA, C. CASQUET, C. JAMOND, J. -R. DIAZ DE TERAN, AND A. CENDRERO. 1992. Eyolution of the eastern yolcanic ridge of the Canary Islands based on new K-Ar data. J. Volcanol. Geotherm. Res. 53: 251-274.
CONTI, E., FISCHBACH, A., AND SYTSMA, K. J. 1993. Tribal relationships in Onagraceae: implications from rbcL sequence data. Ann. Missouri Bot. Gard. 80: 672-685.
COSNER, M. B. , R. K. JANSEN, AND T. G. LAMMERS. 1994. Phylogenetic relationships in the Campanulales based on rbcL sequences. Plant Systematics and Eyolution 190: 79-95.
CRONK, Q. C. B. 1992. Relict flowers of Atlantic islands:
129 pattern assessed. Biol. J. Linn. Soc. 46: 91-103
CRAWFORD, D. J., 1990: Plant Molecular Systematics: Molecular approaches. John Wiley & Sons, New York, NY.
CRAWFORD, D. J. , T. F. STUESSY, AND M. SILVA O. 1987. Allozyme divergence and the evolution of Dendroseris (Compositae: Lactuceae) on the Juan Fernandez Islands. Syst. Bot. 12: 432-443.
CRAWFORD, D. J. , WHITKUS, R. , AND STUESSY, T. F. 1987b. Plant evolution and spéciation on oceanic islands. In URBAKSKA. K., (Ed. ) : Differentiation patterns in higher plants, pp. 183-199. London; Academic Press.
CRAWFORD, D. J. , T. F. STUESSY, M. B. COSNER, D. W. HAINES, M. SILVA 0., AND M. BAEZA. 1992a. Evolution of the genus Dendroseris (Asteraceae: Lactuceae) on the Juan Fernandez Islands: Evidence from chloroplast and ribosomal DNA. Syst. Bot. 17: 676-682.
CRAWFORD, D. J. , T. F. STUESSY, HAINES, D. W. , COSNER, M. B. , SILVA 0., M. , AND P. LOPEZ. 1992. Allozyme diversity within and divergence among four species of Robinsonia (Asteraceae: Senecioneae), a genus endemic to the Juan Fernandez Islands, Chile. - Amer. J. Bot. 79: 962-966.
DARWIN, C. 1859. On the origin of species by means of natural selectoins. Reprint of first edition, 1950. Watts and Co., London.
DARWIN, C. 1872. On the origin of species by means of natural selection or the preservation of favored races in the struggle for life. 6th ed. Murray, London.
DE JOODE, D. R. and J. F. WENDEL. 1992. Genetic diversity and origin of the Hawaiian Island endemic, Gossvoium tomentosum. Amer. J. Bot. 79: 1311-1319.
DOEBLEY, J. , DURBIN, M. L. , GOLENBERG, E. M. , CLEGG, M. T., AND MA, D. P. 1990. Evolutionary analysis of the large subunit of carboxylase (rbcL) nucleotide sequence among the grasses (Gramineae). Evolution 44: 1097-1108.
DONOGHUE, M. J. , R. G. OLMSTEAD, J. F. SMITH, AND J. D. PALMER. 1992. Phylogenetic relationships of Dipsacales based on rbcL sequences. Annals of the Missouri Botanical Garden 79: 333-345.
DOYLE, J. J. 1992. Gene trees and species trees: molecular systematics as one-character taxonomy. - Syst. Bot. 17:
130 144-163.
DOYLE, J. J. AND J. L. DOYLE. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf material. Phytochemical Bulletin 19: 11-15.
EHRENDORFER, P., MANEN, J. -P., AND NATALI, A. 1994. cpDNA intergene sequences corroborate restriction site data for reconstructing Rubiaceae phylogeny. PI. Syst. Evol. 190: 245-248.
FELSENSTEIN, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783-791.
FELSENSTEIN, J. 1986-1993. PHYLIP, Phvloaenv Inference Package. Version 3.52c. The University of Washington, Seattle, WA.
FERNANDEZ-PALACIOS, J. M. , AND ANDERSON, C. 1993. Species composition and within archipelago co-occurrence patterns in the Canary Islands. Ecography 16: 31-36.
FLINT, R. R. 1971. Glacial and Ouaternarv Geology. John Wiley & Sons, New York.
FRANCISCO-ORTEGA, J. , JACKSON, M. T., CATTY, J. P., AND FORD-LLOYD, B. V. 1992. Genetic diversity in the Chamaecvtisus oroliferus complex (Fabaceae: Genisteae) in the Canary Islands in relation to in situ conservation. Genet. Resources Crop. Evol. 39: 149- 158.
FRANCISCO-ORTEGA, J. , JANSEN, R. K., CRAWFORD, D. J. AND SANTOS-GUERRA, A. 1995. Chloroplast DNA evidence for intergeneric relationships of the Macaronesian endemic genus Argyranthemum (Asteraceae). Syst. Bot. 20: 413-422.
FRANCISCO-ORTEGA, J. , D. J. CRAWFORD, A. SANTOS-GUERRA, J. A. CARVALLO. 1996. Isozyme differentiation in the endemic genus Argyranthemum (Asteraceae: Anthemideae) in the Macaronesian Islands. PI. Syst. Evo. 202: 137- 152.
FRANCISCO-ORTEGA, J. , R. K. JANSEN, and A. SANTOS-GUERRA. 1996. Chloroplast DNA evidence of colonization, adaptive radiation, and hybridization in the evolution of the Macaronesian flora. Proc. Natl. Acad. Sci. USA 93: 4085-4090.
FUSTER, J. M. , HERNAN, P., CENDERO, A., COELLO, J. , CABTAGREL, J. M. , ANCOCHEA, E. , AND IBARROLA. E. 1993.
131 Geochronology de la isla de El Hierro (Islas Canarias). Bol. Soc. Esp. Hist. Nat. 88: 1-4.
GADEK, P. A., AND QUINN, C. J. 1993. An analysis of relationships within the Cupressaceae sensu stricto based on rbcL sequences. Ann. Missouri Bot. Gard. 80: 581-586.
GALLEGO, M. J. , S. TALAVERA, AND S. SILVESTRE. 1980. Revision del genero Reichardia Roth (Compositae). Lagascalia 9: 159-218.
GAUT, B. S., MUSE, S. V., CLARK, W. D. , AND CLEGG, M. T. 1992. Relative rates of nucleotide substitutions for the establishment of monocotyledonous plants. J. Mol. Evol. 35: 292-303.
GIELLY, L. , AND TABERLET, P. 1994. Chloroplast DNA polymorphism at the intrageneric level: implications for the establishment of plant phylogenies. C. R. Acad. Sci. Life Sci. 317: 685-692.
GIELLY, L. , AND TABERLET, P. 1996. A phylogeny of the European gentians inferred from chloroplast trnL^"^' intron sequences. Bot. J. Linn. Soc. 120: 57-75.
GIELLY, L. , YUAN, Y. -M. , KUPFER, P., AND TABERLET, P. 1996. Phylogenetic use of noncoding regions in the genus Gentiana L. : Chloroplast trnL(UAA) intron versus nuclear ribosomal internal transcribed spacer sequences. Mol. Phy. & Evol. 5: 460-466.
GIVNISH, T. J., SYTSMA, K. J. , SMITH, J. F. AND HAHN, W. J. 1994. Thorn-like prickles and heterophylly in Cvanea: Adaptations to extinct avian browsers on Hawaii? Proc. Natl. Acad. Sci. USA 91: 2810-2814.
GOTTLIEB, L. D. 1981. Gene number of species of Asteraceae that have different chromosome number. Proc. Natl. Acad. Sci. USA 78: 3726-3729.
GOTTLIEB, L. D. 1981. Electrophoretic evidence and plant populations. Prog. Phytochem. 7: 1-46.
GOTTLIEB, L. D. 1982. Conservation and duplication of isozymes in plants. Science 216: 3726-3729.
HAM, R. C. H. J. VAN, HART H'. T., MES, T. H. M. , AND SANDBRINK, J. M. 1994. Molecular evolution of noncoding regions of the chloroplast genome in the Crassulaceae and related species. Curr. Genet. 25: 558-566.
132 HAMRICK, J. L., LINHART, Y. B. , AND MITTON, J. B. 1979. Relationships between life history characteristics and electrophoretically detectable genetic variation in plants. Ann. Rev. Ecol. Syst. 10: 173-200.
HAMRICK, J. L., AND GODT, M. J. W. 1990. Allozyme diversity in plant species. In BROWN. H. D. (Ed. ) : Plant population genetics, breeding, and genetic resources, pp. 43-63. Sunderland: Sinauer Associates.
HANSEN, A. AND SUNDING, P. 1985. Flora of Macaronesia. Checklist of vascular plants. 3. revised edition. Sommerfeltia 1: 26- 29.
HANSEN, A. AND SUNDING, P. 1993. Flora Of Macaronesia. Checklist of vascular plants. Sommerfeltia 17: 1-295.
HELENURM, K. AND GANDERS, F. 1985. Adaptive radiation and genetic differentiation in Hawaiian Bidens. Evolution 39: 753-765.
HUMPHRIES, C. J. 1979. Endemism and evolution in Macaronesia. In "Plants and Islands", ed. D. Bramwell, pp 171-199. London: Academic Press.
JANSEN, R. K. , WEE, J. L. , AND MILLIE, D. 1997. Comparative utility of chloroplast DNA restriction site and DNA sequencing data for phylogenetic studies in plants. In: SOLTIS, D., SOLTIS, P., DOYLE, J. J. (Eds) Molecular Systematics of Plants, 2nd ed., Chapman Hall, New York. In press.
JOHNSON, L. A., AND SOLTIS, D. E. 1994. matK DNA sequences and phylogenetic reconstruction in Saxifragaceae sensu stricto. Syst. Bot. 19: 143-156.
JOHNSON, L. A., AND SOLTIS, D. E. 1995. Phylogenetic inference in Saxifragaceae sensu stricto and Gilla (Polymoniaceae) using matK sequences. Ann. Missouri Bot. Gard. 82: 149-175.
JEFFREY, C. 1966. Notes on Compositae, I. The Cichorieae in East Tropical Africa. Kew Bulletin 18: 427-486.
KIM, J. -H. , HART, H'. T., AND MES, T. H. M. 1996. The phylogenetic position of East Asian Sedum species (Crassulaceae) based on chloroplast DNA trnL^"**)- ^^p(GAA) intergenic spacer sequence variation. Acta Bot. Neerl. 45: 309-321.
KIM, K. -J. AND R. K. JANSEN. 1994. Comparisons of phylogenetic hypotheses among different data sets in
133 dwarf dandelions (Krigia) : additional information from internal transcribed spacer sequences of nuclear ribosomal DNA. Plant Systematics and Evolution 190: 157-185.
KIM, K. -J., JANSEN, R. K. , WALLACE, R. S., MICHAELS, H. J. , AND PALMER, J. D. 1992. Phylogenetic implications of rbcL sequence variation in the Asteraceae. Ann. Missouri Bot. Gard. 79: 428-445.
KIM, K. -J., JANSEN, R. K., 1995: ndhF sequence evolution and the major clades in the sunflower family. - Proc. Natl. Acad. Sci. 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 radition. Proceedings of the National Academy of Sciences USA. 93: 7743-7748.
KIM, S. -C., CRAWFORD, D. J. AND JANSEN, R. K. 1996. Phylogenetic relationships among the genera of the subtribe Sonchinae (Asteraceae: Lactuceae): Evidence from ITS sequences. Syst. Bot. 26:
KIM, Y. -D. AND JANSEN, R. K. 1996. Phylogenetic implications of rbcL and ITS sequence variation in the Berberidaceae. Syst. Bot. 21: 381-396.
KRON, K. A. AND CHASE, M. W. 1993. Systematics of the Ericaceae, Empetraceae, Epacridiaceae and related taxa based upon rbcL sequence data. Ann. Missouri Bot. Gard. 80: 735-741.
LA ROCHE, F. AND J. C. RODRIGUEZ-PINERO. 1994. Aproximacion al nûmero de tâxones de la flora vascular silverstre de los archipiélagos macaronésicos. Revista de la Academia Canaria de Ciencias 6: 77-98.
LACK, D. 1947. Darwin^s Finches. Cambridge University Press, New York.
LANDER, N. S. 1976. Actites. a new genus of Compositae from Australia. Telopea 1: 129-135.
LEWIS, P., 1993 : GeneStat-PC version 3.3. North Carolina State University, Raleigh, NC.
LI, W. -H. AND TANIMURA, M. 1987. The molecular clock runs more slowly in man than in apes and monkeys. Nature 326: 93-96.
134 LOWREY, T. 1986. A biosystematic revision of Hawaiian Tetramo1opium (Compositae: Astereae). Allertonia 4: 203- 265.
LOWREY, T. K. AND D. J. CRAWFORD. 1985. Allozyme divergence and evolution in Tetramolop ium (Compositae: Astereae) on the Hawaiian Islands. Syst. Bot. 10: 64-72.
LOWREY, T. K. 1995. Phylogeny, adaptive radiation, and biogeography of Hawaiian Tetramolopium (Asteraceae: Astereae). In Hawaiian Bioaeoaraphv: Evolution on a Hot Spot Archipelago. eds. Wagner, W. & Funk, V., Smithsonian Inst. Press, Washington D.C. pp. 195-220.
MADDISON, D. R. 1991. The discovery and importance of multiple islands of most-parsimonious trees. Systematic Zoology 40: 315-328.
MADDISON, W. P. AND MADDISON, D. R. 1992. MacClade. Version 3.0. Sinauer Associates, Inc., Massachusetts.
MANEN, J-F., NATALI, A., AND EHRENDORFER, F. 1994. Phylogeny of Rubiaceae-Rubiae inferred from the sequence of a cpDNA intergene region. PI. Syst. Evol. 190: 195-211.
MANEN, J.-F. AND NATALI, A. 1995. Comparison of the evolution of ribulose-1,5-biphosphate carboxylase (rbcL) and atpB-rbcL non-coding spacer sequences in a recent plant group, the tribe Rubiae (Rubiaceae). J. Mol. Evol. 41; 920-927.
MAYER, S. S. 1991. Artificial hybridization in Hawaiian Wilkstroemia (Thymelaeaceae). Amer. J. Bot. 78: 122- 130.
MCDOUGALL, I. AND SCHMINCKE, H. U. 1976-1977. Geochronology of Gran Canaria, Canary Islands: age of shield building volcanism and other magmatic phases. Bull. Volcanol. 40: 57-77.
MEUSEL, H. 1953. Über Wuchsformen, Verbreitung, und Phylogenie der einiger mediterran-mitteleuropaischen Angiospermen Gattungen. Flora 139: 333-393.
MITCHELL-THOME, R. C. 1985. Radiometric studies in Macaronesia. Bol. Mus. Mun. Funchal 37: 52- 85.
MORIKAWA, T. AND LEGGETT, J. M. 1990. Isozyme polymorphism in natural populations of Avena canariensis from the Canary Islands. Heredity 64: 403-411.
NATALI, A., MANEN, J-F., AND EHRENDORFER, F. 1995. Phylogeny
135 of the Rubiaceae-Rubioideae, in particular the tribe Rubieae; Evidence from a non-coding chloroplast DNA sequences. Ann. Missouri Bot. Gard. 82: 428-439.
NEI, M. 1972. Genetic distance between populations. Amer. Naturalist 106: 283-293.
NEI, M. 1973 Analysis of gene diversity in subdivided populations. Proc. Natl. Acad. Sci. USA 70: 3321- 3323.
NEI, M. 1987. Molecular and evolutionary genetics. Columbia University Press, New York, NY.
NEUHAUS, H. AND LINK, G. 1987. The Chloroplast tRNA’-^® (UUU) gene from mustard. Curr. Genet. 11: 251-257.
OLMSTEAD, R. G. AND REEVES, P. A. 1995. Evidence for the polyphyly of the Scrophulariaceae based on chloroplast rbcL and ndhF sequences. Ann. Missouri Bot. Gard. 82: 176-193.
PAX, D. L. , PRICE, R. A., AND MICHAELS, H. J. 1997. Phylogenetic position of the Hawaiian geraniums based on rbcL sequences. Amer. J. Bot. 84: 72-78.
PEREZ DE PAZ, J. 1976. Observaciones sobre la biologia y relaciones de Sventenia bupleuroides F. Q. Botanica Macaronesica 1: 51-65.
PLUNKETT, G. M. , SOLTIS, D. E. , SOLTIS, P. S., AND BROOKS, R. E. 1995. Phylogenetic relationships between Juncaceae and Cyperaceae: insights from rbcL sequence data. Amer. J. Bot. 82: 520-525.
PRICE, R. A. AND PALMER, J. D. 1993. Phylogenetic relationships of the Geraniaceae and Geraniales from rbcL sequences comparisons. Ann. Missouri Bot. Gard. 80: 661-671.
PONS, A. AND L. BOULOS. 1972. Révision systématique du genre Sonchus L. s.l. III. Étude palynologique. Botaniska Notiser 125: 310-319.
RAMON, A. T., M. GANAL, AND V. HEMLEBEN. 1990. GC balance in the internal transcribed spacers ITS l and ITS 2 of nuclear ribosomal RNA genes. Journal of Molecular Evolution 30: 170-181.
REITH, M. E. AND STRAUS, N. A. 1987. Nucleotide sequence of the chloroplast gene responsible for triazine resistance in canola. Theor. Appl. Genet. 73: 357- 363 .
136 ROUX, J. AND L. BOULOS. 1972. Révision systématique du genre Sonchus L. s.l. II. Étude caryologique. Botaniska Notiser 125: 306-309.
SAAD, S. I. 1961. Pollen morphology in the genus Sonchus. Pollen et Spores 3: 247-260.
SAHUQUILLO, E. AND LUMARET, R. 1995. Variation in the subrtropical group of Dactvlis alomerata L.- I. Evidence from allozyme polymorphism. Biochem. Syst. Ecol. 23: 407-418.
SAMBROOK, J., E. F. FRITSCH, AND R. MANIATIS. 1989. Molecular cloning, a laboratorv manual. Second Edition. Plainview, New York: Cold Spring Harbor Laboratory Press.
SANDERS, R. W. , T. F. STUESSY, C. MARTICORENA, AND M. SILVA 0. 1987. Phytogeography and evolution of Dendroseris and Robinsonia. tree-Compositae of the Juan Fernandez Islands. Oper Botanica 92: 195-215.
SANDERSON, M. J. AND M. J. DONOGHUE. 1989. Patterns of variation in levels of homoplasy. Evolution 43: 1781- 1795.
SANG, T., D. J. CRAWFORD, S. -C. KIM, AND T. F. STUESSY. 1994. Radiation 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 8: 1494-1501.
SANG, T., CRAWFORD, D. J. , STUESSY, T. F. AND SILVA 0. M. 1995. ITS sequences and the phylogeny of the genus Robinsonia (Asteraceae). Syst. Bot. 20: 55-64.
SCOTLAND, R. W. , SWEERE, J. A., REEVES, P. A., AND OLMSTEAD, R. G. 1995. Higher-level systematics of Acanthaceae determined by chloroplast DNA sequences. Amer. J. Bot. 82: 266-275.
SHAPIRO, D. R. AND TEWARI, K. K. 1986. Nucleotide sequences of transfer RNA genes in the Pisum sativum chloroplast DNA. Plant Mol. Biol. 6: 1-12.
SKOTTSBERG, C. 1956. Derivation of the flora and fauna of the Juan Fernanez and Easter Islands; part 1. The Juan Fernandez Islands. Nat. Hist. Juan Fernandez an dEaster Islands 1: 193-405.
SMITH, J. F. , KRESS, W. J. , AND ZIMMER, E. A. 1993. Phylogenetic analysis of the Zingiberales based on rbcL
137 sequences. Ann. Missouri Bot. Gard. 80: 620-630.
SOLTIS, D. E., MORGAN, D. R. , GRABLE, A., SOLTIS, P. S., AND KUZOFF, R. 1993. Molecular systematics of Saxifragaceae sensu stricto. Amer. J. Bot. 80: 1056- 1081.
SOLTIS, D. E., XIANG, Q.-Y., AND HUFFORD, L. 1995. Relationships and evolution of Hydrangeaceae based on rbcL sequence data. Amer. J. Bot. 82: 504-514.
SOLTIS, P. S. AND SOLTIS, D. E. 1995. Plant molecular systematics: Inferences of phylogeny and evolutionary processes. In Evolutionary Biology, ed. Kecht, M. K. et al. Plenum Press, pp. 139-194.
STEBBINS, G. L. JR. 1953. A new classification of the tribe Cichorieae, family Compositae. Madrono 12: 65-81.
STEBBINS, G. L. JR., AND M. S. WALTER. 1953. Chromosome and phylogeny in the Compositae, tribe Cichorieae. Univ. Calif. Publ. Bot. 26: 401-430.
STEELE, K. P. AND VILGALYS, R. 1994. Phylogenetic analyses of Polemoniaceae using nucleotide sequences of the plastid gene matK. Syst. Bot. 19: 126-142.
STUESSY, T. F. , K. A. FOLAND, J. F. SUTTER, R. W. SANDERS, AND M. SILVA 0. 1984. Botanical and geological significance of potassium-argon dates from the Juan Fernandez Islands. Science 225: 49-51.
SUNDING, P. 1979. Origin of the Macaronesian flora. In Plants and Islands, ed. Bramwell, D., Academic Press, New York. pp. 13-40.
SVENTENIUS, E. R. S. 1960. Add it amentum ad floram Canariensem. Madrid, Spain: Institute Nacional de Investigaciones Agronomicas Ministerio de Agriculture.
SWOFFORD, D. L. 1993. PAUP: Phylogenetic analysis usina parsimony, version 3.1.1. Champaign: Illinois Natural History Suryey.
TAKHTAJAN, A. 1969. Flowering Plants: Origin and Dispersal. Smithsonian Inst. Press, Washington D. C.
TAKHTAJAN, A. 1978. The Floristic Region of the World Nauka, Leningrad, USSR.
WAGNER, W. , D. R. HERBST, and S. H. SOHMER. 1990. Manual of the flowering plants of Hawaii. Uniyersity of Hawaii
138 Press and Bishop Museum Press: Honolulu. 2 vols. 1853 pp.
WALLACE, A. R. 1880. Island Life. Macmillan and Co., London.
WARDLE, P. 1963. Evolution and distribution of the New Zealand flora, as affected by Quaternary climates. New Zealand Journal of Botany 1: 3-17.
WEBB, C. J., W. R. SYKES, AND P. J. GARNOCK-JONES. 1988. Flora of New Zealand, vol. IV. Christchurch: Botany Division DSIR.
WEEDEN, N. F. AND WENDEL, J. F. 1989. Genetics and plant isozymes. In SOLTIS. D. E., SOLTIS, P. E., (Eds); Isozymes in plant biology, pp. 46-72. Portland, Dioscorides Press.
WENDEL, J. F. AND WEEDEN, N. F. 1989. Visualization and interpretations of plant isozymes. In SOLTIS. D. E., SOLTIS. P. E., (Eds): Isozymes in plant biology, pp. 5- 45. Portland, Dioscorides Press.
WEDNEL, J. F. AND A. E. PERCIVAL. 1990. Molecular divergence in the Galapagos Isalnds-Baja California species pair Gossvpium klotzschianum and G. davidsonii (Malvaceae). PI. Syst. Evol. 171: 99-115.
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.
WHITE, T. J., T. BRUNS, S. LEE, AND J. TAYLOR. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. Pp. 315-322 in PCR Protocols: A guide to methods and amplification, eds. M. Innis, D. Gelfand, J. Sninsky, and T. White. San Diego: Academic Press.
WITTER, M. S. AND G. D. CARR. 1988. Adaptive radiation and genetic differentiation in the Hawaiian silversword alliance (Compositae: Madiinae). Evolution 42: 1278- 1287.
WILLIAMS, M. A. J. 1982. Ouaternarv Environments in Northern Africa. eds. Williams, M. S. J. & Adamsom, D. A., Balkema, Rotterdam, pp. 13-22.
WITTER, M. S., CARR, G. D., 1988: Adaptive radiation and genetic differentiation in the Hawaiian silversword
139 alliance (Compositae: Madiinae). - Evolution 42: 1278- 1287.
WOLFE, K. H., MORDEN, C. W., AND PALMER, J. D. 1992. Function and evolution of a minimal plastid genome from a monphotosynthetic parasitic plant. Proc. Natl. Acad. Sci. USA 98: 10648-10652.
WU, C. -I. AND LI, W. -H. 1985. Evidence for higher rates of nucleotide substitution in rodents than man. Proc. Natl. Acad. Sci. USA 82: 1741-1745.
YOKOTA, Y. , T. KAWATA, Y. IIDA, A. KATO, AND S. TANIFUJI. 1989. Nucleotide sequences of the 5.8s rRNA gene and internal transcribed spacer regions in carrot and broad bean ribosomal DNA. Journal of Molecular Evolution 29: 294-301.
XIANG, Q-Y., SOLTIS, D. E., MORGAN, D. R., AND SOLTIS, P. S. 1993. Phylogenetic relationships of Cornus L. sensu lato and putative relatives inferred from rbcL sequence data. Ann. Missouri Bot. Gard. 80: 723-734.
140 Jeffrey(1966) Stebbins(1953) Bremer(1993, 1994)
Scolvrnus orouo S c o l y m i n a e S c o l v r n u s
Tolois qroup Cichoriinae S i Ç h o c & y m Tolpis subaroup Microseridinae Catananchinae Catananche subaroup stephanomeridiinae Microseridinae steohanomeria subaroup Thamnoseris stephanomeridiinae Microseris subaroup Malacothricinae
Sonchus qroup Dendroseridinae Dendroseridinae S o n c h u s : Dendroseris Dendroseris S u b a . S o n c h u s Thamnoseris Subq. Oriaosonchus Suba. Oendrosonchus Thamnoseris. Dendroseris
HvDochaeris aroup Hypochaer idinae Hypochaeridinae HvDochaeris subaroup Scorzonerinae Scorzonerinae Scorzonera subaroup
Cichorium group Crepidinae C r e p i d i n a e Cichorium subaroup Dubvaea-Soroseris line Crépis subaroup Launaea-Sonchus line: Sonchinae: Launaea, Sonchus. Dubvaea series Launaea. Sonchus. Reiçhardis, AethepFhias, Prenanthes series ReicbSCdiS, ÊâtheoEhtZA Actites. Babcockia. Crépis series Hieracium-Andrvala line Emberaeria. Kirkianella, Taraxacum series Prenanthes-Lactuca line Lactucosonchus. Sventenia, Chondrilla series younaia-Ixeris line Taeckholmia Launaea series: Crépis line Hieraciinae Launaea. Reichardia. L a c t u c i n a e Aetheorhiza
Table 1. Subtribal classifications of the Lactuceae. Taxa Voucher 'Accession Numbers
Aetheorhiza Cass. A. pulbosa (L.) Cass. Jansen 1105 (TEX) L48135, L48136 Babcockiq Boulos B, platylfepis (Webb) Boulos Kim S.-C. et al. 1028 (OS) L48137, L48138 Dendroseris D. Don.* D. litoralis Skottsb. Stuessy et al. 11973 (OS) L49508, L 4 9 5 0 9 D. marainata (Bert. & Dene.) Stuessy et al. 11999 (OS) L49510, L49511 Hook, & Arn. D. pacrantha (Bert. & Dene.) Skottsb. Stuessy et al. 5149 (OS) L49512, L49513 D. micrantha Hook. & Arn. Stuessy et al. 11582 (OS) L49514, L49515 D. pruinata (Johow) Skottsb. Stuessy et al. 11349 (OS) L 4 9 5 1 6 , L 4 9 5 1 7 D. neriifolia Hook. & Arn. Stuessy et al. 11534 (OS) L 4 9 5 1 8 , L 4 9 5 1 9 D, pinnata (Bert. & Dene) Hook. & Arn. Stuessy et al. 11334 (OS) L 4 9 5 2 0 , L 4 9 5 2 1 D. berteroana (Done) Hook. & Arn. Stuessy et al. 11589 (OS) L49522, L49523 p. pçoia Skottsb. Landero & Ruiz 9316 (OS) L49524, L49525 Emberaeria Boulos E. qrandifolia (T. Kirk) Boulos Atkinson 118/85 (OS) L 4 8 1 3 9 , L 4 8 1 4 0 Kirkianella Allan K. novae-zelandiae (Hook, f.) Allan D.Glenny 4910 (OS) L48141, L48142 Kriaia Schreber. K. jnontana (Michx.) Nutt. Kim K.-J. 10141 (TEX) L13946 L a c t q c a L . L. perennis L. Bonn Bot. Gard. s.n. L 4 8 1 4 3 , L 4 8 1 4 4 L. sativa L. No Voucher (cultivated) L 1 3 9 5 7 Lactucosonchus (Sch. Bip.) Svent. L. webbii (Sch. Bip.) Svent. Kim S.-C. et al. 1033 (OS) L 4 8 1 5 9 , L 4 8 1 6 0 Launaea Cass. L. arborescens (Batt.l Murb. Kim S.-C. et al. 1040 (OS) L 4 8 1 4 5 , L 4 8 1 4 6 L. nudicaulis (L.) Hook. f. Kim S.-C. et al. 1053 (OS) L 4 8 1 4 7 , L 4 8 1 4 8 Microseris D. Don M. laciniata Sch.-Bio. Chambers 5369 (OSC) L 1 3 9 5 4
Table 2. Sources of plants for ITS sequences. Voucher specimens are deposited in Ohio State University Herbarium (OS) and the Herbarium of University of Texas (TEX). All sequences are deposited in GenBank; ‘sequences obtained from Sanq et al. (1994); "sequences obtained from Kim and Jansen (1994). Taxa Voucher 'Accession number
Prenanthes L. P. altissima L. Mehrhoff s.n. (TEX) L48149, L 4 8 1 5 0 p. pendula Sch. Bip. Kim S.-C. et al. 1051 (OS) L48155, L 4 8 1 5 6 Kim S.-C. et al. 1052 (OS) L 4 8 1 5 7 , L 4 8 1 5 8 P. Durpurea L. Kim S.-C. 1049 (OS) L48151, L48152 Pvrrhopappus DC." P. arandiflous Nutt. Kim K.-J. 10508 (TEX) L13953 Reichardia Roth R. picroides (L.) Roth Belgium Bot. Gard. 2871 L48153, L 4 8 1 5 4 R. tinqinata (L.) Roth KEW 223-70-02 090 L48163, L48164 R. liqulata (Vent.) Kunkel & Sundinq Kim S.-C. et al. 1044 (OS) L48165, L 4 8 1 6 6 S o n c h u s L . Suba. Oendrosonchus Sch. Bip. ex Boulos S. canariensis (Sch. Bio.) Boulos Kim S.-C. et al. 1021 (OS) L 4 8 2 9 1 , L 4 8 2 9 2 S. conaestus Willd. Kim S.-C. et al. 1000 (OS) L 4 8 1 7 3 , L 4 8 1 7 4 S. fruticosus L. Fil. Kim S.-C. et al. 1046 (OS) L48125, L48126 S. qonzalezpadroni Svent. Kim S.-C. et al. 1037 (OS) L48127, L48128 S. ortunoi Svent. Kim S.-C. et al. 1036 (OS) L 4 8 1 2 9 , L 4 8 1 3 0 Subq. Oriaosonchus Boulos S. schweinfurthii Oliv. et Hiern Knox 2560 (OS) L 4 8 2 9 5 , L 4 8 2 9 6 s. luxurians (R. E. Fries) c. Jeffrey Knox 2559 (OS) L48297, L48298 Subq. Sonchus S. kirkii (T. Kirk) Allan Silbury s.n. (OS) L 4 8 2 9 9 , L 4 8 3 0 0 s. asper L. Hill Jansen 1109 (TEX) L48301, L48302 S. oleraceus L I. Regk s.n. (OS) L48303, L48304 S. bourqeaui Sch. Bip. Kim S.-C. 1035 (OS) L48305, L48306 S. arvensis L. Jansen 1103 (TEX) L48307, L48308 S. maritimus L. L. Vilar s.n. (OS) L 4 8 3 0 9 , L 4 8 3 1 0 S. oalustris L. Kim S.-C. 1050 (OS) L 4 8 3 1 1 , L 4 8 3 1 2 s. tuberifer Svent. Kim S.-C. et al. 1045 (OS) L 4 8 3 1 3 , L 4 8 3 1 4 Sventenia Font Ouer S. bupleuroides Font Ouer Kim S.-C. et al. 1041 (OS) L48315, L48316 Taeckholmia Boulos T. pinnata (L. Fil.) Boulos Kim S.-C. et al. 1006 (OS) L48319, L48320
Table 2. Continued. Taxa Voucher 'Accessi on number
T. canariensis Boulos Kim S.-C. et al. 1043 (OS) L 4 8 3 2 3 , L 4 8 3 2 4 T. heterophvlla Boulos Kim S.-C. et a l . 1 0 3 7 (OS) L48333, L 4 8 3 3 4 T. qboreg (DC.) Boulos Kim S.-C. et a l . 1 0 4 7 (OS) L48325, L48326 Taraxacum Weber T. officinale Weber Jansen 1107 (TEX) L48337, L48338
Table 2, Continued. ->ITS 1 10 20 30 40 SO 60 70 KRIG TCCAACCCTG CAAAGCAGA GACGACCCGC GAACTTGTAC CCAT-AATCC GGAGTCAGGC ATA-TTGCCT CT-GTC PYRR TCGAACCCTG CAAAGC-CA GACGACCCCC GAACATGTAC ATA-CAATCG GCTTTGATGC ATA-TTGACT CT-GGC HICR TCCAACCCTG CAAAGG-GA GACTACCCGC GAACAGGTAC CCA-AATATC GCAGTTCGGC ATA-TTGCCT TT-GGT TARA TCGAACCCTG CAAGCC-AG AACGACCTGT GAACACGTAA ATA-CAACTG CGTCATGGGG AGA-TGGATC TT-GCT PRAL TCGAACCCTG CAAGGC-AG ACCGACCCGT GAACACGTAA ATA-CAACTT GGTGACGCGG AGA-TCGGCC TT-CGT PRPU TCGAACCCTG CAAGGC-AG AACGACCTGT GAACATGTAA ATA-CAACCC GGTGATCTGG AGTC-GGCCC TT-GGT LAPE TCGAACCCTG CAA-GC-AG AACGACCTGT GAACATGTAA ACA-CAACTG GGTGACAGGG AAA-TGGCAA TT-GGT LASA TCGAACCCTG CAA-GC-AG AAACCCCCGT GAACATGTAA CCA-CAACGG GGTCACCGTG ATAA-CGCCC TC-GGT REPI TCGAACCCTG CAATGC-AG AACGACCTGT GAACATGTAA AT-TCAACTC GGTGTTGGTC AAA-TGGGCC CAAGCT RETI TCGAACCCTG CAACGC-AG AACGACCTGT GAACATGTAA AT-TCAACTC GGTGTTGGTG AAAAT-CCCC TAAGGT RELI TCGAACCCTG CAACGC-AG AACGACCTGT GAACATGTAA AT-TCAACTC CCTGTTCGTG AAAAT-GCCC TAAGGT LAAR TCGAACCCTG CAAGGC-AG AACGACCCCT GAACATGTAA A-A-CAACTT GCTGCTCTTC ACA-TTGCCT TTAGGT LAND TCGAACCCTG CAAAG—AG AACGACCTGT GAACATGTAA -T7CCA-CTT GGTGCTCCTA AGG-TGGCTC TTACGC AETH TCGAACCC— GA-C-TGTAA -T—CA-CTT GCTGCTCTTC ACA-TCGCCC TTAGGT SVEN TCGAACCCTG CAAAGC-AG AACCCCCTGT GAACATGTAA ATA-CAACTC GCTGTTCTTC AGACIGGGCC TTAGGT LACT TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC CCTGTTCTTC AGA-TCGGCC TTAGGT PRPEl TCGAACCCTG CAAAGC-7- AACGACCCCT GAACATGTAA ATA-CAACTC GGTGTTGCTC AGA-TGGCCC TTAGGT PRPE2 TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GCTGTTGCTC AGA-TCGGCC TTAGGT SOKI TCCAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTT CCTCCTGTTC TGA-TGGCCT TTATGT SOAS TCGAACCCTG CAAAGG-?- AACGACCCCT GAACATGTAA AT7-CAAGTT GCTGCTCTTC TGA-TCCCCT TTATGT SOBO TCGAACCCTG CAAACC-AG AACGACCCCT GAACATGTAA ATA-CAAGTT GCTGCTCTTC AGA-TGGCCC TTATGT SOAR TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTT GCTGCTCTTC TCA-TGGCCC TTAGGT SOMA TCGAACCCTG TAAAGC-AG AACGACCTGT GAACATGTAA ATA-CAACTT GCTGCTCTTC TCA-TGCGCC TTAGGT SOPA TCGAACCCTG CAAAGC-A- AACGACCTGT GAACATGTAA ATT-CAACTC CGTGTTCTTT AGA-TGGCCC TTAGGT SOOL TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAAGTT GCTGCTCTTC TGA-TGCGCT TTATGT SOTU TCGAACCCTG CAAAGC-7- AACGACCCCT GAACATGTAA ATA-CAACTC GGTGTTCTTC ACA-TGCGCC TTAGGT SOLU TCGAACC-TG CAAACC-AC AACGACCCCT GAACATGTAA ATA-CAACTC GCTGCTCTTC ACA-TGGACC TTAGGT SOSC TCGAACC-TG CAAAGC-AG AACGACCCCT GAACATGTAA TTA-CAACTC GGTGATCTTG AGA-TCGGCC TTAGGT SOCA TCGAACCCTG CAAAGC-AG AACGACCTGT GAACATGTAA ATA-CAACTC GGTGTTCTTC AGA-TGGGCC TTAGGT SOCO TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GCTGTTCTTC AGA-TCGGCC TTAGGT SOFR TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GGTGTTCTTC ACA-TGCGCC TTAGGT SOCO TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GGTGTTCTTC AGA-TGGGCC TTAGGT SOOR TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GGTCTTCTTC AGA-TGGGCC TTAGGT BABC TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GGTGTTCTTC AGA-TGG7CC TT7GGT TAPI TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATATCAACTC GGTCTTCTTC AGA-TGGGCC TTAGGT TAAR TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GGTCTTCTTC A7A-TG7GCC TTAG7T TACA TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GGTCTTCTTC AGA-TGGCCC TTAGGT TAME TCGAACCCTG CAAACC-7- AACGACCCCT GAACATGTAA ATA-CAACTC GGTCTTCTTC AGA-TGGGCC TTAGGT KIRK TCGAACCCTG CCAACC-AG AACGACCCCT GAACATGTAA ATA-CAACTT CCTCCICTTC AGA-TTGGCC TTAGGT EKBE TCGAACCCTG CAAAGC-AG AACGACCTGT GAACATGTAA ATA-CAACTT GCTCCrCTTC AGA-TGGGCC TTAGGT DELI TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GCTGCTCTTC AGA-TGGGCC TTAGGT DEMA TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC CCTCCICTTC AGA-TGGGCC TTAGGT DEMO TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC CCTCCtCTTC AGA-TGGGCC TTAGGT DEMI TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC GCTGCTCTTC AGA-TGGGCC TTAGGT DEPR TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC CCTCCTCTTC AGA-TGGGCC TTAGGT DENE TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-YAACTC CCTCCTCTTC AGA-TGGCCC TTAGGT DEPI TCGAACCCTG CAAAGC-AG AACCACCYCT GAACATGTAA ATA-CAACTC CCTCCTCTTC AGA-TGGGCC TTAGGT DEBE TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC CCTCCTCTTC AGA-TGGGCC TTAGGT DERE TCGAACCCTG CAAAGC-AG AACGACCCCT GAACATGTAA ATA-CAACTC CCTCCTCTTC AGA-TGGGCC TTAGGT Table 3. Aligned ITS sequences from the subtribe Sonchinae and four outgroup genera. A dash represents a gap. Ambiguous data are Indicated by -?-. Polymorphic sites are represented by following symbols: R « A/C, Y = C/T, S « C/G, V • A/C/G. Species abbreviations are: Krlola montana (KRIG); Pvrrhooaoous multlcaulls (PYRR); Microseris laciniata (HICR); Taraxacum officinale (TARA); Prenanthes altissima (PRAL); P. ourourea (PRPO); Lactuca perenis (LAPE); L. sativa (LASA); Reichardia picroides (REPI); g. tinqinata (RETI); R. liqulata (RELI); Launaea arborescens (LAAR); t. nudlcaulIs (LANU); Aetheorhiza (AETH); Sventenia (SVEN); Lactucosonchus (LACT); P. pendula (PRPEl); P. pendula (PRPE2); Sonchus Iclrkll (SOKI); S. asper (SOAS); S. bouroeaul (SOBO); S. arvensis (SOAR); S. maritimus (SOMA); S. oalustris (SOPA); S. oleraceus (SOOL); S. tuberifer (SOTU); S. luxurians (SOLU); S. schweinfurthii (SOSC); S. canariensis (SOCA); S. conoestus (SOCO); S. fruticosus (SOFR); S. qonzalezpadroni (SOCO); S. ortunoi (SOOR); Babcockia (BABC); Taeckholmia pinnata (TAPI) ; T. arborea (TAAR) ; T. canariensis (TACA); T. heterophvlla (TAHE); Kirkianella (KIRK); Emberoerla (EMBE); Dendroseris litoralis (DELI); D. mardlnata (DEMA); g. macrantha (OEMC); D. micrantha (DEMI); fi. pruinata (DEPR); D. nerrlfolla (DENE); E- Pinnata (DEPI); D. berteroana (DEBE); E- reola (DERE).
145 80 90 100 110 120 130 140 150 KRIG CTTTG TCCCTGACAC CCTGTCGGCA TATGTTT-GT GCTGCC-CCC TTAGGAATGC CACGGATGTT TATGTCGGCC PÏRR CTTTG TCCATGGAGC CCTGTCGGCA TACGTTT-GT GGTCAT-CCG TTCGGAACGA CACGAAAGTC -ATGCCGGCG HICR CTCTC TCCTTGCACA CCCTTCGGCA TGTGTTT-GT GGTGCC-CCG ATCGGGGTGC CATGTATGTC -ATGTTTGCA TARA TCTGA TCCT-CAACA CCTCCTAGCG TGCCTGC-AT GCTT7C-TCT TTTCGGCTAT CATGCATGTA T-TGTTGGAA PRAL CCTGA ■TCCT-CAACA CCTCCCGGCG TGCTTGT-GT GCTGTC-TCT TTT-GGGCAC CATGGATGTC T-CGTTGGAT PRPU CCTTA TACCC— ATC CCTACCGGÏG TGTGTTT-GT GCTGCC-TCT TTTGGGGTGC CATAGATCCA T—GCTGGAC LAPE CYTGA TCCCCCAACC CCTTCTGAYG TGTATTT-CT GGTCCCTTCT TTT-GGGCAT CATGGATCCT T—GTCAGAC LASA CCTTA GCCCCTAACA CTTCCCGACG TGAGTTC-GT GGTGTC-TTT TTTGGGGCAT CATGGAT—T -CCGTTGGAC REPI TCTGA TTAC-CAACA CCTCTTGGTG TGTTTTC-AT GGTAT-AGCT TTTGCTGTAT CATGGATGAT -CCATCAGAC RETI CCTGA TTAC-CAACA CCTCCTGGTG TGTTTTC-AT 7GTATT-GCT TTTGCTGTAT CATGGATGGC -CCATCAGAC RELI CCTGA TTAC-CAACA CCTCCTGGTG TGTTTTC-AT GCTATT-GCT TTTGCTGTAT CATGGATGCC -CCATCAGAC LAAR CTTGA TGAG-CAGCA CATCCCGGTG TGTTTTC-AT GGTCTCACCT T—GTGGTAC CATGGATGTC -CCATCGGAC LANU CTGGA TTAG-CAATA CCTCCCGTTG TGCGTTC-AT GGTTCCTATC TTT—GGTAG CATGGATGTC -CCATCGCAC AETH TCTTA TCAG-CAACT CCATYCGGTG TGTTTCC-AT GGTATTCTCT TTTGCGGTAC CATGGATGTC -CCATCGGGC SVEN TTTGA TCAG-CAATA CCACCCGGTT TGTTTCC-AT GGTATCTTCT TTTATGGTAC CATGGATGTC -ACATCGGAT LACT TTTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GCTATC-TCT TTTGTGGTAC CATGGCTGTC -ACATCGGAT PRPEl TCTSA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GCTATC-TCT TTTATGGTAC CATGGATSTC -ACATCGGAT PRPE2 TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GGTATC-TCT TTTATGGTAC CATGGATGTC -ACATCGGAT SOKI TTAGA TCAG-CAACG CCATCCGGTG TGTTTTC-AA GGTATC-TCT TTTGTGGTAC CATGGATGTC -CCATCGGAC SOAS TTAGA TCAG-CAACG CCATCCGCTG TGTTTTC-AA GGTATC-TCT TTTGTGGTAC CATGGATGTC -CCATCGCAC SOBO TTGGA TCAG-CAAC- CCATCCGGTG TGTTTTC-AT GGTATC-TCT TTTGTGGTAC CATTGATCTC -CCATTGGAC SOAR TTTGA TCAG-CAACA CCATCCGGTG TGTTTCT-AT GCTATC-TCT TTTGTTGTGC CATGGATGTC -CCACTGGAC SOMA TTTGA TCAG-CAACA CCATCCGGTG TGTTTCT-AT GGTATC-TCT TTTGTTATGC CATGGATGTC -CCACTGGAC SOPA TCTGA TAAG-CAACA CCATCCGGTG TGTTTCC-AT GGTATC-TCT TTTGTGGTAC CATGGATGTC -CCATCGGAC SOOL TTAGA TCAG-CAACG CCATCCGGTG TGTTTTC-AA GCTATC-TCT TTTGTGGTAC CATGGATGTC -CCATCGGAC SOTU TCTGA TCAG-CAATA CCATCYG7TT TGTTTCC-AT GGTATC-TCT TTTATGGTAC CATGGATGTC -ACATCGGAT SOLU TTTGA TCAG-CAAC- CCATCCGGTG TGTTTCC-CT GCTATC-TCT TTTGTGGTAC CATGGATGTC -CCATCGGAC SOSC TTTGA TCAG-CAAC- CCATCCGGTG TGTTTCCAAT GGTATC-TCT TTTGTGGTAC CATGGATGTC -CCATCGGAC SOCA TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GGTATC-TCT TTT7TG7TAC CATGGATGTC -AA7TCCGAT SOCO TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GGTATC-TCT TTTATGGTAC CATGGATGTC -AAATCGGAT SOFR TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GGTATC-TCT TTTATGGTAC CATGGATGTC -AAATCGGAT SOGO TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GGTATC-TCT TTTATGGTAC CATGGCTGTC T-AATCGGAT SOOR TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GGTATC-TCT TTTATGGTAC CATGGVTGTC T-AATCGGAT BABC TCTGA TCAG-CAATA CCATCCGGCT TGTTTCC-AT GGTATC-TCT GTYATGGTAC CATGGATGTC -ACATCGGAT TAPI TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GCTATC-TCT TTTATGGTAC CATGGATGTC -AAATCGGAT TAAR TYTGA TCAG-CAACA CCATCCGGTT TGTTTCC-AT GGTATC-TCT TTTATGGTAC CATGGATGTC -AAATCGGAT TACA TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GCTATC-7CT TTTATGGTAC CATGGATGTC -AAATCGGAT TAHE TCTGA TCAG-CAATA CCATCCGGTT TGTTTCC-AT GCTATC-ACT TTTATGGTAC CATGGATGTC -AAATCGGAT KIRK TGTCA TCAG-CAACA CTATTCCGTG TGTTTCT-AT GCTATC-TCT TTTGTGGTAC CATGGATGTC -CCATCGGAC EMBE TCTGA TCAG-CAACA TCATCCCGTC TGTTTCT-AT GGTATC-TCT TTTGTAGTAC CATGGATGTC -CCATCGGAC DELI TCTTA TCAG-CAACA CCATCCGGTG TGTTTCC-AT GCTATC-TCT GTTGTGGTAC CATGGATGTC -CCATCGGAC DEMA TCTTA TCAG-CAACA CCATCCGGTG TGTTTCC-AT GGTATC-TCT GTTGTCCTAC CATGGATGTC -CCATCGGAC DEMC TCTTA TCAG-CAACA CCATCCGGTG TGTTTCC-AT GCTATC-TCT GTTGTGGTAC CATGGATGTC -CCATCGGAC DEMI TCTGA TCAG-CAACA CCATCCGGTG TGTTTCC-AT GGTATC-TCT GTTGTGGTAC CATGGATGTC -CCATCGGAC DEPR TCTGA TCAG-CAACA CCATCCGGTG TGTTTCC-AT GGTATC-TCT GTTGTGGTAC CATGGATGTC -CCATCGGAC DENE TCTGA TCAG-CAACA CCATCTGGTG TGTTTCC-AT GGTATC-TCT GTTGTGGTAC CATGGATGTC -CCATCGGAC DEPI TCTGA TCAT-CAAAA CCATCCGATG TGTTTCC-AT GGTATC-TCT GTTGTCGTAC CATGGATGTC -CCATCGGAC DEBE TCTGA TCAT-CAAAA CCATCCGATG TGTTTCC-AT GGTATC-TCT GTTGTCATAC CATGGATGTC -CCATCGGAC DERE TCTGA TCAT-CAAAA CCATCCGATG TGTTTCC-AT GGTATC-TCT GTTGTCGTAC CATGGATGTC -CCATCGGAC Table 3. Continued.
146 160 170 180 190 200 210 220 KRIG CATT-AACAA A-CCCCG-CA C-GCAATGTG CCAAGAAAAA CAAAAAACTG AGAAGCACGC GTCC— AATT TTGCC PYRR CCAT-AACAA A-CCCCG-CA C-GGAATGTG CCAAGGAAAA C-GAAATATG AGAAGCGCAT GTCC— ATTA TCGCC HICR CATT-AACAA A-CCCCG-CA C-GGACTGTG CCAAGGAAAA TATTAAACTG AGAAGGACGC GTCC— AATA TCGCC TARA TTTT-AACAA A-CCCCGGCA C-GCCATGTG CCAAGGAAAA C-AATAAACC AGAAGGACTC GACC— TGTT ATGCC PRAL C-T-AAACAA A-CCCCGGCA C—GAATCT- C-AAGGAAAA C-AAATAATG AGAAGGACTC GTC— TTGTT ATCTC PRPU CAT— AACAA AACCCCGGCA C-GCCATCTG CCAAGGAAAA C-AAAAA-TG AGAAGGACTC AAACC— GTG TTGCC LAPE CAT— AACAA A-CCCCGGCA C—GCATGTS —AAGGAAAA C-AAAAAATG AGAAGGACAC TTAC— TGTA TTGCC LASA CAT— AACAA AACCCCGGCA C-GCTATCTG CCAAGGAAAA C-AAAAA-TG AGAAGGACAC TACCA— GTT TCGCC REPI T-T— AACAA A-CCCCGGCA C—GTATGT- C-AAGGAAAA C-AAGAAATC AGAAGGTATC GACC— TCAT TTGCC RETI T-T— AACAA A-CCCCGGCA C-GGCATGTC CCAAGGAAAA C-AAAAA-TG AGAAGGTATC GACGACTTGA ATGCC RELI T-T— AACAA A-CCCCGGCA C—GCATGTG C-AAGGAAAA C-AAAAA-TG AGAAGGTATC GAC— TTGAA TTGCC LAAR TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA G-AAAAAATG AGAAGATATA GACC— TGTT TTGTC LANU AAT— AACAA A-CCCCGGCA C-GG-ATGT- CCAAGGAAAA C-AAAACATT GGAAGGTATC GACC— TGTT TTGCC AETH TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAA AGAAGGTATA TACC— TYAT TTVCC SVEN TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAG— TTGAT TTGCC LACT TAT— AACAA A-CCCCGGCA C-GCCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC PRPEl TAT— AACAA A-CCCCGGCA C—GVATGTG C-AAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC PRPE2 TAT— AACAA A-CCCCGGCA C—GAATGTG C-AAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC SOKI TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA TT-AAATAAA AGATGCTATT TAC— TTGAT CTGCC SOAS TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA TT-AAATAAA AGATGCTATT TAC— TTGAT TTGCC SOBO TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAA AGATGCTATC TAC— TTGAT CTGCC SOAR CAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATACG AGAAGGTATC TAC— TTGAT TTGCC SOMA CAT— AACAA A-CCCCGGCA C—GCATGTG C-AAGGAAAA CTGAAATAAG AGAAGGTGTC TAC— TTGAT TTGCC SOPA TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA T-GAAATAAG TGAAGGTATC TAC— TTGAT TTGCC SOOL TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA TT-AAATAAA AGATGCTATT TAC— TTGAT TTGCC SOTU TAT— AACAA A-CCCCGGCA CAGGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC SOLU TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGTAGGTATT CAC— TTGAT TTGCC SOSC TAA-TAACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAC AGTAGGTATC CAC— TTGAT TTGCC SOCA TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC SOCO TAT— AACAA A-CCCCGGCA C—GCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC SOFR TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC SOGO TA-C-AACAA A-TCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC SOOR TATC-AACAA A-TCCCCGCA C-GGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC BABC TAT— AACAA A-CCCCGGCA C—GCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC TAPI TAT— AACAA A-TCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC TAAR TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATATA AGAAGGTATC TAC— TTGAT TTGCC TACA TAT—AACAA A-TCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAT? AGAAGGTATC TAC— TTGAT TTGCC TAHE AAT 7CAA A-TCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATATG AGAAGGTATC TAC— TTGAT TTGCC KIRK TATATAACAA A-CCCCGGCA C—GCATGTC G-AAGGAAAA A-AAAATAAG AGAAGGTATC TACC— TGAT TTGCC EKBE TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-AAAATATG AGAAGGTATC TAC— TTGAT TTGCC DELI TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGAAGGTATC TAC— TTCAT TTGCC DEMA TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGAAGGTATC TAC— TTCAT TTGCC DEMC TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGAAGGTATC TAC— TTCAT TTGCC DEMI TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGAAGGTATC TAC— TTCAT TTGCC DEPR TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGAAGGTATC TAC— TTCAT TTGCC DENE TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-TAAATAAG AGAAGGTATC TAC— TTCAT TTGCC DEPI TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-7AAATAAG AGAAGGTATC TTC— TTCAT TTGCC DEBE TAT— AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGAAGGTATC TAC— TTCAT TTGCC DERE TAT—AACAA A-CCCCGGCA C-GGCATGTG CCAAGGAAAA C-GAAATAAG AGAAGGTATC TAC— TTCAT TTGCT Table 3. Continued.
147 ->ITS 2 230 240 250 260 270 280 290 300 KRIG CCC-T TTCCGCTCTG CTTG TTC CGT-GGCC-T CCTTGAAATC ACATC-GC— TCGCCCCC-A ACCATGCATT PYRR CCG-T ACGCGGTGTG CGTG-TTGTG TGACCTCC-T CCTTGAAATC ACATC-RC— TCGCCCCC-A ACCAAGCATC MICR CCG-T TCGCGCTGTG CTTG— TCGC CGT-GTCC-T CCTTGAAATC ACATC-AC— TTGCCCCCCA -CCATGCATC TARA CCG-T TTGTGG7GTG CATTC-TGAG CGT-CTCCTC CCTTG-AATC ACATC-GC— TCGCCCCCCA TC-ATACTTC PRAL CCG-T TCGCGCTGTG CATAC-TGGT CGC-GGCC-T CCTTGGAATC ACATC-GC— TCGTCCCCCA -CCATACTTC PRPO CTG-T TTGC7GTGTG CATGC7-CTT CGT-GGAC-T CTTTGTAATT ACATC-GC— TCGCCCCC-A TC-ATACAAC LAPE CCG-T TTGCGGTGTG C7TGC--GCT TGT-GGCCTT CCTTGGAATC ACATC-GCG- TCGCTCCTAA ACCATGCTTC LASA CCG-T TTGCGCTGTG CGTACA-GGT CGT-GGCC-T CCTTGGAATC ACATC-GC— TCGCTCCCCA -CCATACCTC REPI CCGAT TTATGGTGTG CATG-ATGCT TGT-ATCC-T CCTTGAATTT AAATC-GCG- TCGCCCCCCA TC-AAACAAC RETI CCGAT TTACGGTGTG CATG-ATGGT TGT-ATCC-T ACTTTAATTT AAATC-GC— TCGCCCCC-A TC-AAACACC RELI CCGAT TTACGCTGTG CATGCATGGT TGT-ATCC-T ACTTTAATTT AAATC-GTCG TCGCCCCC-A TCCAAACAGC LAAR CCGTT TTTCGGTGTG CACACA-GGC TCT-AGC—T CCTTGAAATT ACATC-GTGT TGCCCCCC-A GG-AAACATC LANU CCGTT TTGCGGTGTG CATGCA-GGA TAT-AGCC-T CCATGAAAAT ACATC-GCG- TCGCCCCTT- GCCAACAGTA AETH C7GTT TTGCGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAC ACATT-GCG- TCGCCCCCCA -CCGAACATC SVEN CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAAAT ACATC-GTG- TCGCCCCCC- GCCAAACATC LACT CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAAAT ACATC-GCG- TCGCCCCCC- GCCAAACAT- PRPEl CCGTT TTGCGGTGTG CATGCA-GGT GCTCAGCA-T TCTTTAAAAT ACATC-GCG- TCACCCCCC- GCCAAACATC PRPE2 CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAAAT ACATC-GTG- TCACCCCCC- GCCAAACAT- SOKI CCGTT TTACGGTGTG CATACA-GGT GGT-AGCC-T TCTTTAAAAC ACATCTGCG- TCGCCCCCT- GCCACACAT- SOAS TVGTT TTACGGTGTG CATACA-GGT GGT-AGCC-T TCTTTAAAAC ACATCT-CG- TCGCCCCCT- GCCACACAT- SOBO CCGTT TAGCGGTGTG CATACATG-T GGT-GGCA-T TCTTTAAAAC ACATC-GCGC TCGCCCCCC- GCCACACAT- SOAR CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCC-T CTTTAAAATC AAATC-GCTC TGSCCCC-T- GCCAAACAT- SOMA CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCC-T CCTTAAAAAC AAATC-GCG- TCGCCCC-T- GCCAAACAT- SOPA CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAT ACATC-GCG- TCGCCCCCC- GCCAAACATC SOOL YCGTT TTACGGTGTG CATACA-GGT GGT-AGCC-T TCTTTAAAAC ACATCT-C7- TCGCCCCYT- GCCACACAT- SOTO CCGTT TT7CGCTGTG CATGCA-GGT GGT-AGCA-7 TCTTTAAAAT ACATC-GC— TCGCCCCCC- GCCAAACATC SOLO CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAC ACATC-GCG- TCGCCCCCC- GCCAAACATC SOSC CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAC ACATC-GCG- TCGCCTCCC- GCCAAACATC SOCA CYGTT TTACGGTGTG CATGCA-GGT GGT-AGCA-A TCTTTAAATT ACATC-GC— TCGCCCCCC- GCCAAACATC SOCO CCGTT TTACGGTGTG CATGCA-GGT GGT-AGCA-A TCTTTAAATT ACATC-GC— TCGCCCCCC- GCCAAACATC SOFR CYGTT TTACGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAA-T ACATC-GTG- TGGCCCCCCA -CCAAACATC SOGO CCGTT TTACGGTGTG CATGCATG-T GGT-AGCA-T TCTTTAAAAT ACATC-GCG- TCGCCCCCC- GCCAAACATC SOOR CCGTT TTACGGTGTG CATGCATG-T GGT-AGCA-T TCTTTAAAAT ACATC-GCG- TCGCCCCCC- GCCAAACATC BABC CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTCAAAT ACATC-GTG- T7SCCCCTC- GCCAAACATC TAPI CCGTT TTACGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAAAT ACATC-GC— TCGCCCCCC- GCCAAACATC TAAR CCGTT TTACGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAAAT ACATC-GS— TCGCCCCCC- GCCAAACATC TACA CCGTT TTACGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAAAT ACATC-GC— TCGCCCCCC- GCCAAACATC TAHE CCGTT TTACGGTGTG CATGCA-GGT GGT-AGCA-T TCTTTAAAAT ACATC-GCG- TCGCCCCCC- GCCAAACATC KIRK CTGTT TTGCGGTGTG CATTCA-GGT GGT-AGCC-T CCTTAAAAAC ACATC-ACG- TCGCCCCCC- GCCAAACATC EMBE CCGTT TTGCGGTGTG CATGCA-GGT GGT-TGCC-T CCCTAAAAAC ACATC-GCG- TCGCTCCCT- GCCAAACATC DELI CCGTT TT-CGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAC ACATC-CCG- TCGCCCC-T- GCCAAACATC DEMA CCGTT TT-CGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAC ACATC-GCG- TC7CCCC-T- GCCAAACATC DEMC CCGTT TT-CGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAC ACATC-CCG- TCGCCCC-T- GCCAAACATC DEMI CCGTT TTGCGGTGTG CATG7A-GTT GGT-AGCC-T ACTTTAAAAC ACATC-GCG- TCGCCCC-T- GCCAAACATC DEPR CCGTT TTGCGGTGTG CATGCA-GTT GGT-AGCC-T CCTTTAAAAC ACATC-GCG- TCGCCCC-T- GCCAAACATC DENE CCGTT TTGCGGTGTG CATGCA-GGT GGT-AGCC-T CCTTTAAAAC ACATC-GCG- TCGCCCC-T- GCCAAACATC DEPI CCGTT TTGCGGTGTG CATGAA-GAT GGT-TGCC-T CCTTTAAAAC ACATC-GCG- TCGCCCC-T7 GCCAAACATC DEBE CCGTT TTGCGGTGTG CATGAA-GAT GGT-TGCC-T CCTTTAAAAC ACATC-CCG- TCGCCCC-T- GCCAAACATC DERE CCGTT TTGCGGTGTG CATCAA-GGT GGT-TGCC-T CCTTTAAAAC ACATC-GCG- TCGCCCCCT- GCCAAACATC Table 3. Continued.
148 310 320 330 340 350 360 370 KRIG C-TCAT-GGG ATG— CTTGG CATCGGGG— CGGAGATTCG CCTCCCGTCC CTTTGGTGTG GTTGGCCTAA ATCGG PYRR CTT-AC-GGG ATG— CTTGG CATTCGGG— CCGAGATTGC CCTCCCGTGC TTTTGGTGCG GTTGGCCTAA ACTGG MICR CTATTT-GGG ATG— ATTGG CATCGGGG— CGGATATTCG ACTCCCGTCC CTTTGGTGTG GTTGTCCTAA ACCTG TARA CCTTAA-GGG TAGT-CGTGG TGATTGGGAG CGGAGATTCG CTTCCCGTGC TTGTTCTGCC GTTGGTCAAA ATAGC PRAL CATGAT-GCT TAGT-CATGG TCTTTGGGGG CGGAGATTCG CCTCCCGTCC TTCTGGTGCC GTTGGCTTAA ACAAG PRPU CCAACC— GG TTGT-CATGG TGAT-GGG-G CGGAGATTCG TCTCCCCTAC TTGTT—CCG GTTGGCC7AA AAAAG LAPE CCTAAC-GGG TTGT-CATGG TGTTAGG—G CGGATACTCG CCTCCCCTTC TTATGTTTCG GTTGGCCTAA ATAGG LASA CCT-AC— GG TTGG-CATGG TGTTGGGG— CGGATAATGG CCTCCCGTCC TTGTGTTTCG GTTGGCCTAA ATAAG REPI CC-CAT-GCG TAAG-TATCG TGATGGGG-G CG-AAATTCG CCTCCCCTTC TT—CCTGCG GTTGGCCTAA ATATG RETI CCTCAT-GGG TAAG-TTTGG TGATCCGG— CGGAAATTCG TTG-CGTGCG GTTGGCCTAA ATAGG RELI CCTCAT-GGG TAAG-TTTAG TGAT—GCAG CGGAAATTGG CCTCCCCTTC TTG-CCTGCG GTTGGCCTAA ATAGG LAAR CCTAAC-CGG TAAT-CCTGG TGATGCGG-G CGGAAATTCG CCTCCCCTTC TTG-CGTGCG GTTGGCCTAA ATAGG LANU CC-ATC-GGG TAAT-CCTGG TGATCGGG-G CGGAAACTCG TCTCCTGTTC TTA-CCTGCA GTTGGCCTAA ATATG AETH CTTAAAAGGC TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGATCG GTTGGCCCAA AGATG SVEN CCCTAA— GG TAAT-CATGG TGAIGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TCTCTG GTTGGCCCAA AGATG LACT CCCAAA-GGG TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCCG GTTGGCCTAA AGATG PRPEl CCCAAA-GGG TAATTC7TGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG PRPE2 CCCAAA-GGG TAAT-CGTGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG SOKI CCTAAAGGTG TAAT-AATGG TGATGGTG-G CGGAAATTGG CCTCCCCTTC TTG-TGTTTG GTTGGCCTAA AGATG SOAS CCTAAAGGTG TAAT-AATGG TCATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTTTG GTTGGCCTAA AGATG SOBO CCTAAA-GGG TAAT-CGTGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGGTG SOAR CCTTAATGGG TAAG-CATGG TGATGGGG— CGGAAAATGG CCTCCCCTTC TTG-TGTCCG GTTTGCCTAA ATAAG SOMA CCTTAATGGG TAAG-CATGG TGATCCGG— CGGAAAATCG CCTCCCCTTC TTG-TGTTCG GTTTCCCTAA ATAAG SOPA CCGGAA— GG TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-CGTCCG GTTGGCCTAA AGAGG SOOL -CTAAAGGTG TAAT-AATGG TCATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTTTG GTTGGCCTAA AGATG SOTU CCCAAA— GG — ATT-A-GG TGATCCGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG SOLU CCGAAA—-C TAGT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTTCG GTTGGCCTAA AGATG SOSC CCCAAA G TAGT-CATGG TGATCGGG-G CGGAAATTCG CCTCCCCTTC TTG-TGTCCG GTTGGCCTAA AGATG SOCA CCCAAA— GG TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG SOCO CCCAAA— GG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC -TG-TGTCTG GTTGGCCTAA AGATG SOFR CCCAAA— GG TAAT-AATGG TGAAGGGG-G CGGAAATTGG — TCCCCTTC TTG-TGTCCG GTTGGCCTAA AGATG SOGO CCCAAA— GG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG SOOR CCCAAA— GG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTGA TG BABC CCGAAA-GGG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCTTAA AAATG TAPI CCCAAA-GGG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG TAAR CCYAAA-GGG TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-TCTCTG GTTGGCCTAA AGATG TACA CCCAAA— GC TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG TAHE CCCAAA— GC TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCTG GTTGGCCTAA AGATG KRIK CTGAGA-GGG TAAT-CATGG TGATCGGG-G CGGAAAATGG CCTCCCCTTC TTG-TGTCCG GCTGGCCTAA AGAAG EMBE CTGAGA-GGG TAAT-CATGG TGATGGGG-G CGGAAAATGG CCTCCCCTTC TTG-TGTCCG GTTGGCCTAA AGAAG DELI CTGAAA-GGG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCG7TC TTG-TGTTCG GTTGGCCTAA AGATG DEMA CTGAAA-GGG TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCC7TC TTG-TGTTCG GTTGGCCTAA AGATG DEMC CTGAAA-GGG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCG7TC TTG-TGTTCG GTTGGCCTAA AGATG DEMI CTGAAA-GGG TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCG7TC TTG-TGTCCG GTTGGCCTAA AGATG DEPR CTGAAA-GGG TAAT-CATGG TCATGGGG-G CGGAAATTGG CCTCCCG7TC TTG-TGTCCG GTTGGCCTAA AGATG DENE CTGAAA-GGG TAAT-CATGG TGATGGGG-G CGGAAATTGG CCTCCCG7TC TTG-TGTCCG GTTGGCCTAA AGATG DEPI CTGATA-GGG TAAT-CATGG TGATGGGG-G CGGAAATTGC CCTCCCG7TC TTG-TGTCCG GTTGGCCTAA AGATG DEBE CTGATA-GGG TAAT-CATGG TGATCGGG-G CGGAAATTCG CCTCCCG7TC TTG-TGTCCG GTTGGCCTAA AGATG DERE CTGCTA-GGG TAAT-CATGG TGATCGGG-G CGGAAATTGG CCTCCCCTTC TTG-TGTCCG GTTGGCCTAA AGATG Table 3. Continued.
149 380 390 400 410 420 430 440 450 KIRG AGT-A CC—TTC-GG TGGACCCACG ACTAGTGGTG GTTGAAAAGA CCCTCGTCCT GTGTTGTGCG T-CCTAAGCT PÏRR AGTC- C-- TTC-GG TGGACGCACG ACTAGTGGTG GTTGAATAGA CCCTCGTCTT ATGTTGTGCG TT-GTAAGCT MICR AGTCA CC—TT-GGG TTGACGCACG ACTAGTGGTG GTTGAATAGA CCCTCGTCCT AAGGTGTGCG T-CGTAAGCC TARA AGTCC C-- TTC-GG TGGACACACG GCATGTGGTG GTTGTAAAGA CCCTTTTCTT CTGCTGTGTG TT-CTGAGCT PRAL AGTCC CC—TTC-GG TGGATACACG GCTAGTGGTG GTTGTATAGA CTCTCTTCTT GTGTCCTGTA T-CGTGAGCT PRPU AGTCC CC—TTCA-G TGGACACACG ACTAGTGGTG GTTTAACAGA CCCTTCTCTT T-ATCGTGTG TT-ATGAGCT LAPE AGTTC CC— TTCA-G CGGACACACA ACTAGTGGTG GTTGAACAGA CCTTCCTCTT GGGTTGTGTG T-CGTGAGCT LASA AGTTC C-- TTC-GG CGGACGCACG ACTAGTGGTG GTTGAATAGA CCCTCGTCTT TTCTTGCGTG T-CGTGAGCT REPI AGTCC CC— TTT-GG TGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTTGTCTT TTGTTGTGTG T-CATGAGCT RETI AGTCC CC—TTT-GG TGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT TTGTTGTGTG T-CATGAGCC RELI AGTCC CC—TTT-GG TGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT T-CATGAGCC LAAR AGTCC C-- TTC-GG TGGATGCACA ACTAGTGGTG GTTGAACTGA CCCTCGTCTT GTCTTGTGTG TT-GTTAGCT LANU AGTCC C-- TTC-GG CGGATGCACA ACTAGTGGTG GTTGATCAGA CCCTCGTCTT GTGTTGTGTG T-CTTGAGCT AETH AGTCC CC—TTTAGG TGGATGCACA ACAAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SVEN AGTCC CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT LACT AATCC CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CATGAGCT PRPEl AGTCC YC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT PRPE2 AGTCC CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOKI AGTCC CC-TTAG-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOAS AGTCC CC-TTAG-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOBO AGTCC C—TTAC-GG TGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTA T-CGTGAGCT SOAR AGTCC CCCTTAT-GG TGGATGCACA ACTAGTGGTG GTTGAAAAGA CCTTCGT-TT GTGTTGTGTG TT-GTTAGCT SOMA AGTCC CCCTTAT-GG TGGATGCACA ACTAGTGGTG GTTGAAAAGA CCTTCGTTTT GTGTTGTGTG TT-GTTAGCT SOPA AGTCC CC—TAT-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOOL AGTCC CC-TTAG-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOTU AAGYC CC-TTAY-GG TGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOLU AGTCC CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOSC AGTCC CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOCA AGTCC CC-TTAT-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOCO AGTCC C—TTAC-GG TGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOFR AGTCC CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOGO AGTCT C-- TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT SOOR AGTCT C-- TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT BABC AGTCC 7C— TAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-C7TGAGCT TAPI AGTCT C-- TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT TAAR ATGCC CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG TTCGTGAGCT TACA AGTCT CC— TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT TAHE AGTCT CC—TAC-GG CGGATGCACA ACTAGTGGTG GTTGAACAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT KIRK AGTCC CC-TTAT-GG TGGATGCACA ACTTGTGGTG GTTGAAAAGG CCTTCCTCTT GTGTTGTGTG T-CGTCATCT EMBE AGTCT TC-TTAT-GG TGGATGCACA ACTAGTGGTG GTTGAAAAGA CCTTCCTCTT GTGTTGTGTG T-CGTCAGTT DELI AGCCC CCC-TAC-GG TGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT DEMA AGCCC CCC-TAC-GG TGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT DEMC AGCCC CCC-TAC-GG TGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT DEMI AGTCC CCC-TAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT DEPR AGTCC CCC-TAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGSGCT DENE AGTCC CCC-TAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTATT GTGTTGTGTG T-CGTGAGCT DEPI AGTCC CCC-TAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT DEBE AGTCC CCC-TAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT DERE AGTCC CCC-AAC-GG CGGATGCACA ACTAGTGGTG GTTGAATAGA CCCTCGTCTT GTGTTGTGTG T-CGTGAGCT Table 3. Continued.
150 460 470 480 490 500 510 KIRG GTGAGGGAGG CCCTT— CAT GAAGACCCCA AT-GTGGTGT CTTGCGACGA TGCTT-CGAC PYRR GTGAGGTGGG CCCTT— GAT GAACACCCCT AT-GTATCGT CATCTCACGA TGCTT-CGAC MICR GTAAGGGA-G GCCTT T GAAGACCCCA AC-GTGA-GT CTTGCGACGA TGCTT-CGAC TARA GCTAGGGAAA CCCTC— AAA AAAGAACCCA AT-GTATCGT TCTAGGACCA TGCTT-CGAC PRAL GCTAGGGAAA CCCTC— ATC AAAGACCCCA AT-GTATT?- CTTGCGACGA TGCTT-CGAC PRPO GCTAGGGAAG -CCTC— ATC AA-GACCCCA TC-GTATCGT TTTACCACGC TGCTT-CGAC LAPE GTGAGGGAAG CCCTC— ATC AATGACCCCT TT-GTATCGT CTTCGCACGG TGCTT-CGAC LASA GTAAGGGTAG CCCTC— ATC AAAGACCCCA TT-GTATCGT CTTCGGATGA TGCTT-CGAC REPI ATTAGGGAAG GCCTT— TTA TAAGACCCCA TT-GTATCGT TATAAAACGG TATAT-CGAC RETI CTTTGGGAAG GCCTT— ATA TATGACCCCA TT-CTATCGT TATAAAACGG TATAT-CGAC RELI GTTTGGGAAG GCCTT— ATA TATGACCCCA TT-CTATCGT TATAAAACGG TATAT-CGAC LAAR GTGAGGGAAG -CCTA-CTTT TACGACCCCA C7-GTATCGT TAT7AGACGA TATAT-CGAC LANU GTTAGGGAAG -TCTC— ATT TAAGACCC-A TT-GTATCGT TATAAGGCGA TATAT-CGAC AETH GCGAGGGGAG CTCTT— ATT TTAGACCCCA TT-GTATCTT TAAAAAACGA TACAT-CGAC SVEN GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAAACACGA TATAT-CGAC LACT GTAAAGGAAA TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAAAGATGA TATAT-CGAC PRPEl GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAAAGACGG TATAT-CGAC PRPE2 GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAAAGACGG TATAT-CGAC SOKI GTGAGGGAAA TTCTCTCAAT TTAGACCCCA CT-GTATCGT TAAAAAACGA TATAT-CGAC SOAS GTGAGGGAAA TTCTC— AAY TTAGACCCCA CT-GTATCGT TAAAAAACGA TATAT-CGAC SOBO GTGAGGGAAA TTCTC— ATT TTAGACCCCA CT-GTATCGT TTGAAAACGA TACAT-CGAC SOAR GTGAGGGAAG CTCTC— ATT TTAGACCCTA CT-GTATCGT TAAAA-ATGA TATAT-CGAC SOMA GTGAGGGAAG CCCTC— ATT TTAGACCCTA CT-GTATCGT TAAAA-ATGA TATATT-GAC SOPA GTGAGGGAAG CTCTC— TTT TCAGACCCTA AT-ATATCGT TTACAGACGA TATAT-CGAC SOOL GTGAGGGAAA TTCTC— AAT TTAGACCCCA CT-GTATCGT TAAAAAACGA TATAT-CGAC SOTO GTGAAGGAAG TTCTC— ATT TCAGACCCTT TT-CTATCGT TAAAAGACGA TATAT-CGAC SOLO GCGAGGGAAG CTCTC— ATT TTAGACCCTA AT-GTATCGT TAAAAGACGA TATAT-CGAC SOSC GTGAGGGAAG CTCTC— ATT TTAGACCCTA AT-GTATCGT TAAAAGACGA TATAT-CGAC SOCA GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT -GAAAGACGA TATAT-CGAC SOCO GTGAAGGAAG TTCTC— ATC TCAGACCCTA CY-GTATCGT TAAAAGATGA TATAT-CGAC SOFR GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAAAGACGA TATAT-CGAC SOGO GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAATGACCA TATAT-CGAC SOOR GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAATGACCA TATAT-CGAC BABC GTGAAGGAAG TTGTC— ATT TCAGACCCTA CT-GTATCGT TAAAAGACGA TATAT-CGAC TAPI GTGAAGGAAG TTCTA— ATT TCAGACCCTA CT-GTATCGT TAAATGACCA TATAT-CGAC TAAR GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAAAGACGA TATAT-CGAC TACA GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAATGACCA TATAT-CGAC TAHE GTGAAGGAAG TTCTC— ATT TCAGACCCTA CT-GTATCGT TAAATGACCA TATAT-CGAC KIRK GTGAGGGAAG CTCTC— ATT TTAGACCCCA CT-GTATCGT TTAAAAATGA TATATTGATC EMBE GTGAGGGAAG CTCTC— ATT TTAGACCCCA CTAGTATTGT TAAAAAATGA TATAT-CGAC DELI GCGAGGGAAG CTCTC— ATT CTAGACCCTA CT-GTATCGT TAAAAAATGA TATAT-CGAC DEMA GCGAGGGAAG CTCTC— ATT CTAGACCCTA CT-GTATCGT TAAAAAATGA TATAT-CGAC DEMC GCGAGGGAAG CTCTC— ATT CTAGACCCTA CT-GTATCGT TAAAAAATGA TATAT-CGAC DEMI CAGAGGTAAG CTCTA— ATT TTAGACCCCA CT-GTATCGT TAAAAAATGA TATAT-CGAC DEPR CAGAGGTAAG CTCTA— ATT TTAGACCCCA CT-GTATCGT TAAAAAATGA TATAT-CGAC DENE GGGCGGCAAC CTCTC— ATT TTAGAACCCA CT-GTATCGT TAAAAAATGA TATAT-CGAC DEPI GCGAGGGAAG TTCTC— ATT TTAGACCCCA CT-GTATCGT TAAAAAATGA TATAT-CGAC DEBE GCGAGGGAAG TTCTC— ATT TTAGACCCCA CT-GTATCGT TAAAAAATGA TATAT-CGAC DERE GCTAGGGAAG TTCTC— ATT CTAGACCCCA CT-GTATCGT TATAAAATGA TATAT-CGAC Table 3. Continued.
151 Taxon and Population H I s l a n d s L o c a l i t y
Babcockia olatvleois fWebbI Boulos* f6 1 0 GC T e j e d a # 5 4 1 0 G C A y a c a t a Lactucosonchus webbii fSch. B i p . ) S v e n . " 1 0 P A San Antonio Prenanthes oendula fWebb) Sch. Bio. # 9 1 0 GC Fataga (south) # 1 0 1 0 GC Fataga (north) S o n c h u s L . Suba. Oendrosonchus Sch. Bio. ex Boulos S. acaulis Dum.-Cours. # 2 9 1 0 TE #8, Puerto de Erjos # 3 0 9 TE A r t i c o s a # 3 1 9 GC V a l l e s e c o # 3 2 8 GC Degollada Tasarte S. bornmuelleri Pitard" # 4 3 1 2 PA Bajamar cliffs S. brachvlobus Webb 6 Berth." # 3 3 1 0 GC Cuesta de Silva # 3 4 7 G C A n d e V e r d e # 1 1 9 6 GC Cuesta de Silva S. canariensis fSch. Bio.) Boulos #38 18 TE Carretera Chio-Boca # 3 9 1 4 TE A n a g a # 4 0 8 GC Valleseco #41 8 TE Puerto Erjos # 5 1 0 GC Ande Verde # 1 2 1 0 GC San Nicolas S. conaestus Willd #19 10 TE Free way exit 14 # 2 0 1 0 T E G a r a c h i c o #21 10 TE Icod # 2 2 5 T E Anaga, El Bailadero # 2 3 5 GC Ladera barranco de la Virgen sobre Molino Chico
S. fauces-orci Knoche' # 7 1 0 T E M a s c a S. aonzalezoadronii Svent.' # 4 1 0 GO San Sebastian-Hermigua # 4 8 5 GO Hermigua valley S. aummifer Link’ #49 1 2 TE Ladera de Güimer # 5 1 8 TE Punta de Teno # 5 2 1 0 TE Punta de Hidelgo S. hierrensis fPitard1 Boulos # 4 5 1 5 P A Las cabezadas # 4 6 1 GO La Laguna Grande S. ortunoi Svent.' # 2 7 6 GO L a s R o s a s # 2 8 2 GO Puerto de Hermigua
Table 4. Populations and species of the woody Sonchus alliance studied for isozymes. Abbreviations of islands are as follows; LA, Lanzarote; FU, Fuerteventura; GC, Gran Canaria, TE, Tenerife; GO, La Gpmera; PA, La Palma; HI, El Hierro. N represents number of individuals per population. species endemic to a single island.
152 Taxon and Population N I s l a n d s L o c a l i t y
# 3 1 0 G O B/t Roque la Zarzota & Roque de Ojila s. oalmensis fSch. Bio . ) B o u l o s # 2 4 1 0 PA between Franceses & B a r l o v e n t o # 2 5 1 0 PA San Bartolomé # 2 6 1 0 PA Las Cabezades S. oinnatifidus Cav. # 3 6 1 0 LA F a m a r a # 2 1 0 LA Martaina Canada S. wildoretii U. & A. Reifenberaer 1 0 GO Roque de Ojila Suba. Sonchus L. S. tuberifer Svent.' 1 0 TE M a s c a Sventenia buoleuroides Font Ouer. 1 0 GC Jardin Canario Genus Taeckholmia Boulos T. canariensis Boulos* 1 0 GO El Camello (San S e b a s t i n ) T. caoillaris fSvent.) Boulos* 1 0 TE M a s c a T. microcaroa Boulos* 10 TE Malpais de Güimar T. oinnata CL. Fil.) Boulos # 1 5 1 5 GC Cuesta de Silva # 1 7 1 0 T E El Boqueron # 4 2 8 TE El Boqueron
Table 4. Continued.
153 DENDR BABCO SVENT TUBER PRENT LACTU TAECK
DENDR 0. 600 0.840 0.950 0.887 0.972 0.900 BABCO 0.511 0. 490 0.494 0.495 0.508 0.503 SVENT 0.174 0.714 0.667 0 . 6 8 6 0 . 7 2 6 0 . 8 6 5 TUBER 0.054 0.706 0.405 0 . 7 1 1 0 . 9 8 8 0 . 7 4 5 PRENT 0.120 0.704 0.121 0.342 0.777 0.923 LACTU 0. 028 0.678 0.320 0.013 0 . 2 5 2 0 . 8 9 4 TAECK 0.105 0.688 0.146 0.294 0.080 0 . 1 1 2 1
abbreiviations: subg. Dendrosonchus. DENDR; Babcockia platvlepis. BABCO; Sventenia platvlepis. C/1 SVENT; Sonchus tuberifer. TUBER; Prenanthes pendula. PRENT; Lactucosonchus webbii. LACTU; Taeckholmia. TAECK. P T a x a N H, H r D,T G,T A A p
B a b c o c k i a 2 0 . 0 5 9 0.127 0.068 0.538 1.128 2.000 0.182 Lactucosonchus 1 0.051 0.051 0.000 0.000 1.250 2.000 0.250 P r e n a n t h e s P. pendula 2 0. 057 0.079 0.021 0.271 1 . 3 0 0 2 . 0 0 0 0 . 3 0 0 Sonchus Suba. Dendrosonchus S . a c a u l i s 4 0.030 0.030 0.000 0.009 1 . 3 8 5 2 . 6 6 7 0 . 2 3 1 s. bornmuelleri 1 0 . 0 0 0 0 . 0 0 0 0 . 0 0 0 N A 1 . 0 0 0 NA 0 . 0 0 0 s. brachvlobus 3 0.059 0 . 1 2 6 0 . 0 6 8 0.535 1.500 2.200 0.417 s. canariensis 6 0.028 0.092 0.064 0.696 1.417 2.000 0.417 S. conaestus 5 0.026 0.052 0.026 0.501 1.167 2.000 0.167 s. fauses-orci 1 0.184 0.184 0.000 0.000 1 . 5 0 0 2 . 0 0 0 0 . 5 0 0 S. aonzalezpadroni 2 0.056 0.079 0.023 0 . 2 9 4 1.300 2.000 0.300 s. aummifer 3 0.032 0.037 0.005 0.142 1.182 2.000 0.182 S. hiecrensis 2 0 . 0 3 8 0 . 1 6 7 0 . 1 2 9 0 . 7 7 3 1 . 2 2 2 2 . 0 0 0 0 . 2 2 2 S . o r t u n o i 3 0 . 0 7 5 0 . 0 8 8 0 . 0 1 3 0 . 1 4 3 1 . 3 6 4 2 . 0 0 0 0 . 3 6 4 y; S. oalmensis 3 0.052 0.087 0.035 0.401 1.182 2 . 0 0 0 0 . 1 8 2 s. oinnatifidus 2 0 . 2 5 3 0 . 4 1 5 0 . 1 6 3 0.392 2. Ill 2 . 1 1 1 1 . 0 0 0 S. wildoretii 1 0.000 0.000 0 . 000 N A 1 . 0 0 0 NA 0.000 Sonchus suba. Sonchus s. tgberifer 1 0.047 0.047 0.000 0.000 1. Ill 2.000 0.111 S v e n t e n i a 1 0.057 0.057 0.000 0 . 0 0 0 1 . 2 0 0 2 . 0 0 0 0 . 2 0 0
T. canariensis 1 0 . 0 6 8 0 . 0 6 8 0 . 0 0 0 0 . 0 0 0 1 . 2 7 3 2 . 5 0 0 0 . 1 8 2 T. capiiiaris 1 0. 154 0.154 0 . 0 0 0 0 . 0 0 0 1 . 5 0 0 2 . 3 3 3 0 . 3 7 5 T. micrpcarpe 1 0.054 0.054 0.000 0 . 0 0 0 1 . 2 3 1 2 . 5 0 0 0 . 1 5 4 T . o i p n a t a 3 0 . 0 9 5 0 . 2 0 4 0 . 1 1 0 0 . 5 3 6 1 . 6 9 2 2 . 1 2 5 0 . 6 1 5 M e a n s 0 , 0 6 7 9 , 1 0 9 1 , 3 3 1 0 , 3 8 9
Island endemic means (DE JOODE & WENDEL 1992) 0 . 0 6 4 1 . 3 2 0 . 2 5 Endemics (HAMRICK £. GODT 1990) 0. 063 0.096 1 . 8 0 0 . 4 0
*N number of populations examined; H, ■» diversity within populations; Hy = total diversity; D,ST diversity among populations; G,y = proportion of total diversity among populations; A = mean number of alleles per locus; Ap = mean number of alleles per polymorphic locus; p proportion of polymorphic loci. T a x a V o u c h e r Accession Numbers
Aetheorhiza Cass. A. bulbosa (L.) Cass. Jansen 1105 (TEX) Dendfoseti^ D. Don D. litoralis s)cottsb. Stuessy et al. 11973 (OS) D. marainata (Bert. & Dene.) Stuessy et al. 11999 (OS) Hook. & Arn. Embergeria Boulos E. qrandifolia (T. Kirk) Boulos Atkinson 118/85 (OS) Kirkianella Allan K. novae-zelandize (Hook.f.) Allan D. Glenny 4910 (OS) Lactucosonchus (Sch. Bip.) Svent. L. webbii (Sch. Bip.) Svent. Kim et al. 1033 (OS) Launaea Cass. L.larborescens fBatt.) Murb. Kim et al. 1040 (OS) Prenanthes L. P.'altissima L. Mehrhoff s.n. (TEX) Lf, P. pendula Sch. Bip. Kim et al. 1051 (OS) ON P.- pMTPUrsa L. Kim et al. 1049 (OS) Reichardia Roth R. liqulata (Vent.) Kundel & Sunding Kim et al. 10474 (OS) R. oicroides (L.) Roth Belgium Bot. Gard. 2871 R. tinqinata (L.) Roth KEW 223-70-02 090 S o p c b u s L . Subq. Dendrosonchus Sch. Bip. ex Boulos S. acaulis Dum.-Cours. Kim et al. 1027 (OS) S. canariensis (Sch. Bio.) Boulos Kim et al. 1021 (OS) S. fruticosus L. Fil. Kim et al. 1046 (OS) S. qonzalezpadroni Svent. Kim et al. 1037 (OS) S. aummifer Link Kim et al. 1022 (OS) Subq. Oriaosonchus Boulos S. luxurians (R.E.Fries) c. Jeffrev Knox 2559 (OS) S. schweinfurthii Oliv. et Hiern Knox 2560 (OS) Subg. Sonchus L. S. asoer L. Hill Jansen 1109 (TEX) S. bourqeaui Sch. Bip. Kim 1035 (OS) s. kirkii (T. Kirk) Allan Silbury s.n. (OS) S. martimus L. L. Vilar s.n. (OS) S. oalustris L. Kim 1050 (OS)
Table 7. Sources of plant materials for sequence comparisions of chloroplast intergenic spacer between osbA and trnH. Voucher specimens are deposited in Ohio state University Herbarium (OS) and the Herbarium of University of Texas (TEX). 'All sequences are deposited in GenBan)c. T a x a V o u c h e r 'Accession Numbers
S. tuberifer Svent. Kim et al. 1045 (OS) Sventenia Font Quer S. buoleuroides Font Ouer Kim et al. 1041 (OS) Taeckholmia Boulos T. caoillaris (Svent.^ Boulos Kim et al. 1019 (OS) T. oinnata (L. Fil.) Boulos Kim et al. 1006 (OS) Taraxacum Weber Jansen 1107 (TEX) 1 Table 7. Continued.
LA Original ITS Reduced ITS p g b A - t r n H C o m b i n e d ( 4 9 t a x a ) ( 3 0 t a x a ) ( 3 0 t a x a ) ( 3 0 t a x a )
Number of variable sites 330(65%) 252(50%) 83(16%) 335(33.3%) Number of informative sites 264(52%) 182(36%) 48(9.5%) 230(22.8%) Tree length 898 570 1 1 8 6 9 8 Number of the shortest trees 144 9 0 8 9 5 2 3 6 0 Consistency index (excluding autapomorphies) 0 . 5 2 6 0 . 5 9 9 0 . 6 7 9 0 . 6 0 0 Average bootstrap value" 8 0 . 8 % 76.9% 66.4% 79.3% Average decay value 2.9 3.1 2.2 NA Number of resolved nodes' 2 5 2 0 1 3 1 7 Average sequence divergence NA 9 . 9 % 2 . 4 % NA Ln Skewness of tree-length o o distribution (gl) - 0 . 8 1 6 8 - 0 . 8 2 4 6 - 0 . 7 7 6 2 - 0 . 8 5 5 2
Table 8. Summary of parameters calculated for each of the four data sets. NA not applicable; "values are based on strict consensus tree. psbA-trnH ITS
Reichardia 2.3% 4.7% Sonchus subq. Sonchus 2.0% 9.7% Sonchus suba. Dendrosonchus 0.9% 1.7% The woody Sonchus alliance in Macaronesia 1.0% 2.5% Subtribe Sonchinae 2.4% 9.9%
Table 9. Comparisons of average pairwise sequence divergence between psbA-trnH intergenic spacer and ITS for several genera of Sonchinae (BREMER 1993, 1994).
159 Figure 1. ITS sequence phylogeny of subtribe Sonchinae.
This tree is one of the 144 equally most parsimonious trees with a length of 898. Consistency index = 0.526 (excluding autapomorphies). Retention index = 0.743. Dashed lines indicate branches that collapse in the strict consensus tree. Numbers above lines represent the number of base substitutions followed by the decay values in parentheses.
The bootstrap support (%) is shown below the nodes.
160 Krigia montana Micmsaris ladniata 33 _12_ Pyrrhopappus mulHcaulis Outgroups 20g _ia&5L Lactucapergnis 8 7 X Lactuca saliva 1 ^ J55. Prenanthes puipurea Prenanthes I00« 17(S> Taraxacum officinale Taraxacum 112(4: 77% P renanüm afttsdm a P renanthes 66% _2SWL Reichardia pkm ides 4495 100% ReicttanSa tinginata Reichardia Reichardia ttgufata ] fjum aea arborescens 43 Launaea 100% Laimaea nurOcauBs 3 Aetheorhlza buBxjsa Aetheorhlza _zm_ Sonchus fdrldi 02(2: Sonchusasper Sonc/iussubg. 72% U2_ 100% 13 Sonchus oleraceus S o n ch u s Sonchus bourgeaui ] Sventenia bupleuroides I Sventenia [sm Babcodda platylepis Babcockia 5 6 % k [ S Prenanthes penduiafN) Prenanthes 96% I----- Prenanthes pendulai(S) Sonchus tuberffer Sonchus subg. 3i21 Sonchus canadensis S o n ch u s 559 [679 5 Sonchus congestus Sonchus fruBcosus S o , n c h u ssubg. 1Dendrosonehus r K r Sonchus gonzaiezpadroni I _j__ Sonchus ortunoi 100%] 100 &r 9871 Taedrholmia pinnata —i % Taetikhoimia canadensis I Taeckholmla pUZL. Taeckhobnia heterophytia I 42% Taeckholmla arborea —I Lectucosonchus webbS — LMCtucosonchus Sonchus palustris Sonchus arvensis Sonchus subg. Sonchus madtimus ]_ S o n ch u s KiddaneBa novae-zelandiae KIrklanella Embergeda grandi folia Embergerla Dendroseds BtoraBs _&za_ D endroseds subg. 100% Dendroseds marginata E Dendroseds macrantha D endroseds 12(21 4 21% Dendroseds micrantha 86% ■d: Dendroseds pruinata D endroseds subg. 12 0 1 Dendroseds nedifoBa Rea 5 3 % Dendroseds pinnata j i r H ± Dendroseds ivrteroana D endroseds subg. 99%l___ 6 Dendroseds regia ]Phoenlcoseds Sonchus luxudans S o n c h u s subg. - ± Sonchus schweinfurthii3 Odgosonchus
Figure 1
161 Figure 2. (a) The dashed line encloses the biogeographical région of Macaronesia (19) , which includes northwestern Africa and five archipelagos, (b) The Canary archipelago and the oldest published radiometric ages from subaerial volcanics of each island in parenentheses (17-22)
Abbreviations of islands are: LA (Lanzarote, 15.5 Mya)^ FU
(Fuerteventura, 20.7 My a) ^ GO (Gran Canaria, 13.9 My a) , TE
(Tenerife, 11.6 My a) , GO (La Gomera, 12.5 Mya), PA (La
Palma, 1.5 Mya), and HI (Hierro, >0.7 Mya).
162 Azores Islands
Europe
Madeira Islands
Selvagens Islands Africa
Canary Islands
Tropic of Cancer -
Cape Verde Islands
(b) CANARY ISLANDS LA(15.5) ^
TE(1I.6)
PA(i.5) O \J X? 00(12.5) FU(20.7)
GC(13.9) HI(>0.7)
0 , lOOKm 1 - = I f
Figure 2
163 Figure 3. ITS phylogeny of subtribe Sonchinae (redrawn
from 28). This is one of the 144 equally most parsimonious
trees (Cl = 0.526, RI = 0.743). Dashed lines indicate branches that collapse in strict consensus tree. Numbers above and below branches represent decay and percentage of bootstrap values, respectively. Arrow indicates the clade of woody Sonchus alliance.
164 Krigia montana 38 Microseris laaniata 33 43 Pynhopappus multicaulis Outgroups 20(2, I«>S1 Lactuca perenis 50% 87% L a_ Lactuca sath/a (»S) 35 JL Prenanthes purpurea Prenanthes 00% 17f51 Taraxacum officinale • Taraxacum 77% L2I. 6G% 13 Prenantttes altissima • Prenanthes (2) Z9bSl. —jI — 4 ReichanSa pictoiries 49^ 100% 110 rr~ ReichanJia tinginata Reichardia X 100%'-^ ReichanJia figulata ] Launaea artxirescense _i2_ Launaea 100% Launaea nudicauEs ] Aetheorhiza bufbosa Aetheorhiza -ZÜL Sonchus kirfdi 112(2: Sonchus asper Sonchus subg. 72% Sonchus oferaceus Sonchus Sonchus bourgeaui Ï Sventenia bupfeurokJes • Babcockia platyfepis Prenanthes pâvJuta(N) 96%1- Prenanthes pencJula(S) Sonchus tutjerifer J £ 2 i Sonchus canariensis The woody 55^ Sonchus congestus rS n 5 Sonchus alliance Sonchus fruticosus In Macaronesia \ Sonchus gonzaJezpatSroni Sonchus ortunoi Kz&l 1 100% 100 52% Taeckholmia pinnata % 3 Taeckholmia canariensis 2(2L Taeckholmia heterophyiia 42% 3 Taeckholmia arborea Lactucosonchus webbii —* _li_ Sonchus paiusiis Sonchus arvensis Sonchussubg. Sonchus marWmus ] _Sonchus Kirkianella novae-zelandiae Kirtdanella Embergeria grandifotia Eml>ergeria Dendroseris liloraBs 100% Dendroseris marginata Dendroseris macrantha Dendroseris micrantha Dendroseris Dendroseris pruinata (The Juan Dendroseris neriifotia Fernandez Dendroseris pinnata Islands, Chile) Dendroseris tjerteroana ; Dendroseris regia _i£21 Sonchus luxurians S o n ch u s subg. 95% Sonchus schweinfurthii Origosonchus
Figure 3
165 Figure 4. Preferred ITS phylogeny of the woody
Macaronesian Sonchus and their alliance. This is one of the
8123 equally most parsimonious trees (CI= 0.821, RI= 0.821) and is identical to the 50% majority-rule consensus tree.
Dashed lines and numbers above and below branches as in legend Fig. 2. Abbreviations of islands as in legend Fig. 1.
Abbreviations of habits are hp, herbaceous perennial; cp, caudex perennial; s, rosette subshrub; t, rosette tree/shrub. Abbreviation of habitats are M, mesic; D, dry,
C, coastal.
166 Sventenia bupleuroides (GC; cp) Babcodda platylepis (GC; s) Prenanthes pertduta (GC-N; s) -C Prenanthes penduta (GC S; s) Sontdjus tuberifer (TE; hp) — Sonchus subg. Sonchus S. pinnatiSdus (LA; t) ”{ZS. brachykJbus (GC; s) S. dsftort/» (CAPE VERDE; cp) -- 1 S. hierrensis (GO; t) 63% -- S. hierrensis (PA; t ) 55% 1 54% S. gandogeri (HI; t) S.palmensis (PA; t) Sonchus .P_J----- S. anariensis (TE; t) subg. S. canariensis (GC; t) Dendro- S. acautis (GC; cp) sonchus
58% S. acaulis (TE; cp) S. congestus (TE.GC; t) — 70% S.pinnatus (MADEIRA; t) Sonchus 52% _ |A c : S. fruticosus (MADEIRA; t) subg. S. ustulatus (MADEIRA; cp) Dendro- \ sonchus S. lauces-ord (TE; cp) 100^ sensu S. gonzaiezpadroni (GO; cp) Aldridge S. ortunoi (GO; cp) T. pinnata (GC.TE; s) ” r. pinnata var. microcarpa (TE; s) r. canariensis (GO; s)
T. heterophyiia (GO-1; s) Taeckholmia 2 r. capillaris (TE; s) 40% T. heterophyiia (GO-2; s) r. arborea (TE; I ) { = r. artX3rea (PA; I) _ Lactucosonchus webbii (PA-1; hp) Lactucosonchus webbii (PA-2; hp) Sonchus palustns (EUROPE; hp)
Figure 4
167 Figure 5. The Canary archipelago and the woody Sonchus alliance taxa studied for isozymes. The oldest published radiametric ages from subaerial volcanoes of each island are indicated in parentheses (ABDEL-MONEN & al. 1971, 1972;
MCDOUGALL & SCHMINCKE 1976-1977; CANTAGREL & al. 1984;
ANCOCHEA & al. 1990, 1994; COELLO & al. 1992; FUSTER & al.
1993; CARRACEDO 1994). Average Nei's genetic identities are also shown in each island. Taxa with asterisk represent single island endemic.
168 ~ S. tubenfer’ 1-0.81 T. apiUeris'r 1 T. pinnata J 1-0.89 S. bommuellen' T. microcarpa' S. ptlmensis' ^ 1-0.85 S. canariensis — S. hierrensis Lanzarote( 1 S.SMya) S. congestus Laauœsonchus' S. gammifer' 1-0.80 S. pinnatindus S. fauces-ora‘ t— Babcockia’ Sventenia' _ S. acaulis _ Prenanthes pendulauia- Tenerifet 11.6Mya) T. pinnata & canariensis 1-0.66 S. acaulis 1-0.74 La Palma(l.SMya) & brachylobus S. congestus ]“îl s O — 28-N La Comcra( 1 Z.SMya) Puerlevenlural 20.7 Mya) Q Hierro(>0.7Mya) r 7: wildpretil' “ I l o R4 T. canariensis' _ l S. gomalerpadronii' —i 1-0.77 Cran Canaiiaf 13.9Mya)
for populations of the woody Sonchus alliance included in
this study. Bolded taxa represents segregate genera (or
subgenus) of woody members of Sonchus. i.e, subg.
Dendrosonehus. Abbreviations of islands as in Table 1.
Abbreviations of sectional classification of subg.
Dendrosonehus are as follows: D, Dendrosonehus; B,
Brachvlobi; P, Pinnati (BUOLOS 1972).
170 s e;-,Vi«us iTZ. 19» 0 S. cen;«scus (1t.o
'. erCunoi t00.27| O fC^orcunojorcu (C3.2I) o ^— S.s. /ii«rr#si« (00.441 0 s. results (CC.Sll 0 (0C.32* 0 L— g.ttibTlgtrg.cubTie* (TC) S. «c«uUs (1E.M) 0 _Æ::— _ ^ é u i i $ n t . J o i 0 l — S.gcmjl9tp*dronii (00.41 0
£« ceu ce< o aeb a« (PA} 0
s. 9 u n i/« r (TC.SU o
H::: 9 U M if« r (TC.S21 0 p . f. ortunoi (G0.3J 0 C l u o t o r i L S, gotuéUipÊdeonil (CO.di o
S. eong^tcus (TC.21I 0
S. conçttcut (GC.23) 0 S. comgwcw* (1E.22J o S. pianseUidus (LA .2) O I— S. cânsriMtt (1t. 3#: P I J. cjn4«*n*i5 (1t. 311 P L - S. cdturimsls ( 1 t.4 1 ) p
S. coA«ri«nsio (CC.401 P i r C : 0UOBi/«r (1t.411 0 L- g. €4florl«nsis (CC.Si P
•> t . Aiorrm sip (PA.4H o
— - r. p i / n ê Cé (OC.lî»
— r. pirvMCo (1t.42l
S. polM /uis (PA.24I p
f . t o m o u « ii « r i (PA. 43) o
J poiOORSiS (FA.34) P
L- r. «icrocArpo iTY.
p t/n s ed ( 1 t . (7 . r. c4piJ/4ric . lei
" t e r ftifur(*ns:> ^J.ui
p é l M n g l t (PA.iS) P
6rJCftv/o6«fs ICC. )4 . t h:Pr*a«ACA«* p«Adui4 (O:. ^ S. ctturucsi* (CC.12) P
Prunaneiic* p tn d u l* (CC.1)
5. pinndCtltduM (LA.3(1 O
6r4cfty(ecvs (CC.3ÎI 0 or4cfty;coui (cc.)it a
F v « n e« n i« b w p i« u re id « c (CCi
. S- i«wi.*es*erct (1t. ?* 0
T w tiO p ro C Ii (Col
gèbcockim p3acyi*p&# (GC.(t C l u o c o r 1 -C B * te o c k ia piaCy(*p(g (0C.S4I ]
Figure 6
171 Figure 7. One of the 8952 equally parsimonious trees of
the subtribe Sonchinae (BREMER 1993, 1994) based on
intergenic spacer sequences between psbA and trnH of
chloroplast DNA (consistency index of 0.679 excluding
autapomorphies and retention index of 0.787). Three tax of
Prenanthes are in bold. Dashed lines indicate branches that collapse in the strict consensus tree. Number above nodes represent base substitutions followed by the decay values in parentheses and the bootstrap support(%) is shown below the nodes.
172 Prenanthes purpurea Outgroup .SID, Taraxacum ofSdnale Taraxacum 90% C Prenanthes altissim a Prenanthes im| Reichardia tingitana r86%L_ Reichardia ligulata Reichardia 61% 12 Reichardia picroides 15 1 Launaea arborescens Launaea Aetheorhiza bulbosa Sonchus kifkii -1ÜL Clade 0 4 5 % 66% 1_____ Sonchusasper 3 51% Sonchus bourgeaui 6 Sonchus maritimus 3(4) 62% Kirkianella novae-zelandiae I Clade 0 m l 3 Embergeria grandifolia Sonchus paluslris — Sonchus subg. Sonchus 4M1 3111 71% 7 0% Sonchus luxurians Sonchus subg. / Sonchus schweinfurthii ]Origosonchus 15 bp Sonchus tuberifer Insertion Sonchus canariensis !COl Sonchus fruticosus 27% im Sonchus gummifer The woody Sonchus .IfQ) .... .UQ).. Sonchus gonzaiezpadroni 15% 12% alliance in c Sonchusacaulis Macaronesia Taeckholmia pinnata (clade A) Prenanthes penduia Taeckholmia capillaris 2111 46% Sventenia bupleuroides Lactucosonchus webbii Dendroseris litoralis Dendroseris (clade B) Dendroseris marginata The Juan Fernandez ]Islands
Figure 7
173 Figure 8. ITS sequence phylogeny of subtribe Sonchinae
(modified from KIM S.-C. & al. 1996b). This is one of the
144 equally most parsimonious trees with a length of 898
(consistency index of 0.526 excluding autapomorphies and retention index of 0.743) . The bootstrap support is also shown below the nodes.
174 Krigia montana Microseris laaniata Pyrrhopappus multicaulis Oulgroups Lactuca perenis Lactuca sativa Prenanthes purpurea Prenanthes Taraxacum oflidnale Taraxacum Prenanthes altissima Prenanthes ReicharrTta picroides Reichardia tinginata Reichardia Reichardia ligulata ] Launaea artxtrescens Launaea nucScaulis ] Launaea Aetheorhiza bullxrsa Sonchus kirldi Sonchus asper Clade D Sonchus oleraceus Sonchus bourgeaui 1 Sventenia bupleuroides ' Babcoch'a platylepis Prenanthes pendula(N) Prenanthes pendula(S) Sonchus tuberifer Sonchus canariensis Sonchus congestus The woody Sonchus fruticosus Sonchus Sonchus gonzaiezpadroni alliance In Sonchus ortunoi Macaronesia Taecktiolmia pinnata (Clade A) Taeckholmia canariensis Taeckholmia heterophyiia Taeckholmia artjorea Lactucosonchus wet)bii - Sonchus (iaiustris Sonchus arvensis —, Sonchus maritimus I Clade C 90% 3 Kirkianella novae-zelandia^ Embergeria grandifolia Dendroseris litoralis 100% Dendroseris marginata Dendroseris macrantha Dendroseris micrantha Dendroseris Dendroseris pruinata (The Juan Dendroseris neriifotia Fernandez Dendroseris pinnata Islands, Chile) Dendroseris berteroana (Clade B) Dendroseris regia Sonchus luxurians 95% Sonchus schweinfurthii
Figure 8
175 Figure 9. One of 360 equally most parsimonious trees based on combined data sets, i.e., reduced ITS and psbA-trnH intergenic spacer sequences (Cl = 0.60; RI = 0.742). Number above nodes represent base substitutions.
176 Prenanthes purpurea — Outgroup Taraxacum officinale Prenanthes aftissima Reichardia tinginata 37 RekdtartSa ligulata 100% ReicharcCa picroides Aetheorhiza txrlbosa Sonchus kirldi 53 % Clade D Sonchusasper Sonchus txmrgeaui Sonchus maritimus 9 4 % Kirkianella novae-zelandiae | Clade C 3 5 % Embergeria grandifolia ] Dendroseris litoralis 21 Clade B 100% Dendroseris marginata Sonchus luxurians 99% Sonütus schweinfurthii Sonchus tuberifer Sventenia bupleuroides Prenanthes pendula Sonchus canariensis Sonchus fruticosus 55% Sonctnjs acaulis Clade A Sonchus gummifer 100»/ Sonchus gonzaiezpadroni Taeckholmia pinnata Taeckholmia capillaris Lactucosonchus webbii Sonchus palustris 33 Launaea artrorescens
Figure 9
177