University of New Hampshire University of New Hampshire Scholars' Repository
Doctoral Dissertations Student Scholarship
Winter 1997 Systematics of the Hamamelidaceae based on morphological and molecular evidence Jianhua Li University of New Hampshire, Durham
Follow this and additional works at: https://scholars.unh.edu/dissertation
Recommended Citation Li, Jianhua, "Systematics of the Hamamelidaceae based on morphological and molecular evidence" (1997). Doctoral Dissertations. 1997. https://scholars.unh.edu/dissertation/1997
This Dissertation is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected]. f
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 face, while 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 frir any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NOTE TO USERS
The original document received by UMI contained pages with indistinct print. Pages were filmed as received.
This reproduction is the best copy available.
UMI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SYSTEMATICS OF THE HAMAMELIDACEAE BASED ON MORPHOLOGICAL AND MOLECULAR EVIDENCE
BY
JIANHUA LI B.S. Henan Normal University, Xinxiang, China, 1984 M.S. Central China Normal University, Wuhan, China, 1987
DISSERTATION
Submitted to the University of New Hampshire in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy in Plant Biology
December, 1997
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SYSTEMATICS OF THE HAMAMELIDACEAE BASED ON MORPHOLOGICAL AND MOLECULAR EVIDENCE
BY
JIANHUA LI B.S. Henan Normal University, Xinxiang, China, 1984 M.S. Central China Normal University, Wuhan, China, 1987
DISSERTATION
Submitted to the University of New Hampshire in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy in Plant Biology
December, 1997
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 9819680
UMI Microform 9819680 Copyright 1998, by UMI Company. All 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
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This dissertation has been examined and approved.
Dissertation Director, Dr. A. Linn Bogle Professor of Plant Biology
Dr.//Garrett E s c r o w Professor of Plant Biology
/
Dr. Anita^?. Klein Associate Professor of Biochemistry and Molecular Biology
Dr. Thomas D. Lee Associate Professor of Plant Biology
Dr. Leland S J J&hnke Associate Professor of Plant Biology
'9 Dati
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. THIS DISSERTATION IS DEDICATED TO
my parents and parents-in-laws
Jinsheng and Yuolian Li
&
Jiabao and Yuozhao Zhang
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS
I owe great indebtedness to my advisor, Professor Dr. A.
Linn Bogle. Without his continuing encouragement, unreserved
support, great understanding and consideration, I would not
be able to finish this project.
Dr. Anita S. Klein has been very supportive and generous
to me technically and sometimes financially. I am indebted
to her generosity and guidance.
I would like to extend my sincere thanks to the other
members of my dissertation committee, Dr. Garrett Crow, Dr.
Tom Lee, and Dr. Lee Jahnke, who have made the course of my
doctorate education a rich and rewarding experience.
Thanks are extended to Dr. Tao Sang, Michigan State
University, East Lansing, MI for providing matK primers; Dr.
Bruce Baldwin for technical advice; Donald Padgett for the
conversations and friendship we shared as fellow graduate
students in Plant Systematics; Joselle Germano and Danielle
Friel for sharing lab space and technical assistance.
The following people and their institutions deserve
special thanks for helping me collect materials or sending me
DNA for this study: Patricia Grady and Vinnie Simeone of the
Planting Field Arboretum at Oyster Bay, Long Island, New
York; Dr. Niangjun Chung, Forest Experimental Station,
National University of Taiwan; Dr. Shumiaw Chaw, Botanical
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Institute, Academia Sinica, Taiwan; Dr. Tracy D. Omar,
Washington University Arboretum, Seattle, WA; Chris Strand,
Arnold Arboretum of Harvard University, Boston, MA; Dr. Rob
Nicholson, Smith College, MA; Zhongchun Luo, Forest Bureau of
Xinning County, Hunan, China; Dr. Peter K. Endress, Institute
of Systematic Botany, University of Zurich, Switzerland; Dr.
Yinlong Qiu, Indiana University; Dr. Leng Guan Saw, Forest
Research Institute Malaysia, Kepong, Malaysia; Dr. Richard
Saunders, University of Hong Kong; Dr. Armand Randrianasolo,
Missouri Botanical Garden, St. Louis, MI; Longwood Gardens,
PA; Etienne A. Rakotobe, National Center for Pharmaceutical
Research, Madagascar; Division of State Park, Hawaii.
My sincere thanks also extend to the following Herbaria
for allowing me to access specimens of the Hamamelidaceae, or
loaning me specimens of the Hamamelidaceae for this study:
Harvard University Herbaria, Missouri Botanical Garden,
National Herbarium at the Smithsonian Institution, New York
Botanical Garden, and Hong Kong Herbarium.
My special thanks go to the Department of Plant Biology
and the Graduate School of the University of New Hampshire
for awarding me financial support in the form of Teaching and
Research Assistantships, Travel Funds, Graduate School Summer
Teaching Assistant Fellowships, Graduate Student Research
Enhancement Funds, and a Dissertation Fellowship. Financial
support for research from the Powers Fund is also greatly
appreciated.
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. My heartfelt thanks are extended to my wife Yuling and
daughter Ran for their support, encouragement, and great
companionship.
I would also like to thank Mrs. M. Bogle for her
hospitality and consideration to me and my family. She made
us feel at home and warmly-loved.
Finally, thanks are extended to many unnamed people who
have offered me help in different ways and enriched my life
in the University of New Hampshire.
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
DEDICATION ...... iii
ACKNOWLEDGMENTS ...... iv
LIST OF TABLES ...... ix
LIST OF FIGURES ...... X
LIST OF APPENDICES ...... xiv
ABSTRACT ...... xv
INTRODUCTION
i-1. General Information of the Hamamelidaceae...... 1
i-2. Brief History of Systematics of the
Hamamelidaceae ...... 5
i-3. Systematics Problems of the Hamamelidaceae...... 9
i-4. Parsimony Analysis and Molecular Systematics ....10
i-5. Objectives ...... 11
CHAPTER I. A PHYLOGENETIC ANALYSIS OF THE HAMAMELIDACEAE BASED ON MORPHOLOGICAL DATA
1.1. Introduction ...... 24
1.2. Materials and Methods...... 25
1.3. Results ...... 28
1.4. Discussion ...... 48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER II. PHYLOGENETIC RELATIONSHIPS OF THE HAMAMELIDACEAE INFERRED FROM NUCLEOTIDE SEQUENCES OF THE PLASTID GENE MATK
2.1. Introduction...... 83
2.2. Materials and Methods ...... 87
2.3. Results ...... 91
2.4. Discussion ...... 92
CHAPTER III. PHYLOGENETIC RELATIONSHIPS OF THE HAMAMELIDACEAE INFERRED FROM SEQUENCES OF INTERNAL TRANSCRIBED SPACERS OF NUCLEAR RIBOSOMAL DNA
3.1. Introduction...... 149
3.2. Materials and Methods ...... 152
3.3. Results ...... 158
3.4. Discussion ...... 161
CHAPTER IV. PHYLOGENETIC RELATIONSHIPS OF THE HAMAMELIDACEAE: EVIDENCE FROM MORPHOLOGY AND NUCLEOTIDE SEQUENCES OF ITS AND MATK GENE
4.1. Introduction...... 199
4.2. Materials and Methods ...... 201
4.3. Results ...... 204
4.4. Discussion ...... 205
CHAPTER V. TRANSLATING PHYLOGENY OF THE HAMAMELIDACEAE INTO A CLASSIFICATION SYSTEM
5.1. Introduction...... 251
5.2. Materials and Methods ...... 252
5.3. Results ...... 253
5.4. Discussion ...... 253
CHAPTER VI. GENERAL CONCLUSIONS ...... 271
LITERATURE CITED ...... 273
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES
TABLE 1.1, Morphological character codings in the Hamamelidaceae ...... 63
TABLE 2.1. Sources and vouchers of the species sequenced for matK ...... 104
TABLE 2.2. Location and base composition of amplification and sequencing primers ...... 106
TABLE 2.3. Sequence characteristics of the matK gene in the Hamamelidaceae ...... 107
TABLE 2.4. Codon usage of the matK gene in the Hamamelidaceae ...... 110
TABLE 3.1. Species sequenced for nrDNA ITS ...... 171
TABLE 3.2. Sequence divergence of ITS-1 and ITS-2 (below and above diagonal) in the Hamamelidaceae, calculated in PAUP* (mean distance x 100) ...... 173
TABLE 3.3. Characteristics of the most parsimonious trees obtained by PAUP heuristic searches of ITS-1 sequences aligned using different combination of gap and gap length penalties ..176
TABLE 3.4. Characteristics of the most parsimonious trees obtained by PAUP heuristic searches of ITS-2 sequences aligned using different combination of gap and gap length penalties ..177
TABLE 3.5. Sequence characteristics of the two internal transcribed spacers, separated and combined, in 32 species of the Hamamelidaceae...... 178
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES
FIG.i.l. Geographic distribution of the Hamamelidaceae...... 12
FIG.i.2. Classification system of the Hamamelidaceae proposed by Reinsch (1889) formatted into a c l a d o g r a m ...... 14
FIG.i-3. Generic relationships of the Hamamelidaceae depicted by Tong (1930), formatted into a c l a d o g r a m ...... 16
FIG.i.4. Classification system of the Hamamelidaceae proposed by Harms (1930), formatted into a c l a d o g r a m ...... 18
FIG.i.5. Classification system of the Hamamelidaceae (non-Chinese genera not included, Chang, 1973) formatted into a cladogram ...... 20
FIG.i.6. Classification system of the Hamamelidaceae proposed by Endress (1989b, c), formatted into a cladogram...... 22
FIG.1.1. Stomatal apparatus in the sampled species of the Hamamelidaceae ...... 64
FIG.1.2. Strict consensus of the 12 shortest trees of 219 steps based on morphological data of the Hamamelidaceae ...... 66
FIG. 1.3. 50% majority consensus of the 12 shortest trees based on morphological data of the Hamamelidaceae ...... 68
FIG. 1.4. 50% majority consensus tree of 12 shortest trees based on morphological data, showing the distribution of numbers of characters that change unambiguously on branch ...... 70
FIG.2.1. Approximate relative locations of matK primers ...... 113
FIG.2.2. The Strict consensus of the six most parsimonious trees of 582 steps long based on sequences of the matK gene ...... 115
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIG.2.3. One of the six most parsimonious trees of 582 steps long based on sequences of the matK gene, showing the number of unambiguous substitutions along the branches ...... 117
FIG.3.1. The nrDNA ITS region ...... 179
FIG.3.2. UPGMA phenogram of the Hamamelidaceae generated by analysis of nrDNA sequences aligned using the Optimality method. Jukes-Cantor's one parameter model and gamma rate (1.0) were used in the analysis ..... 181
FIG.3.3. Strict consensus of 90 shortest trees of 1041 steps, generated using nrDNA ITS sequences of the Hamamelidaceae, aligned using the Optimality method ...... 183
FIG.3.4. 50% Majority consensus of 90 shortest trees of 1041 steps, generated using nrDNA ITS sequences of the Hamamelidaceae, aligned using the Optimality method ...... 185
FIG.3.5. Strict consensus of 12 shortest trees of 6747 steps, generated using nrDNA ITS sequences of the Hamamelidaceae, aligned with the Elision method ...... 187
FIG.3.6. One of the shortest tree of 1041 steps using nrDNA ITS sequences of the Hamamelidaceae, aligned using the Optimality method, showing the number of unambiguous character state changes of ITS sequences ...... 189
FIG.4.1. Strict consensus of 203 most parsimonious trees of 185 steps based on morphology of the Hamamelidaceae ...... 217
FIG.4.2. Strict consensus of six most parsimonious trees of 381 steps based on sequences of the matK gene in the Hamamelidaceae ...... 219
FIG.4.3. Strict consensus of 16 most parsimonious trees of 6952 steps based on the ITS grand data set of the Hamamelidaceae ...... 221
FIG.4.4. Strict consensus of morphology and ITS of the Hamamelidaceae obtained using the Consensus approach ...... 223
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIG.4.5. Strict consensus of morphology and matK of the Hamamelidaceae obtained using the Consensus approach ...... 225
FIG.4.6. Strict consensus of ITS and matK of the Hamamelidaceae obtained using the Consensus approach ...... 227
FIG.4.7. Strict consensus of morphology, ITS and matK of the Hamamelidaceae obtained using the Consensus approach ...... 229
FIG.4.8. Strict consensus of two shortest tree of 7161 steps based on the combined data set of morphology and ITS of the Hamamelidaceae ....231
FIG.4.9. Strict consensus of eight shortest trees of 590 steps based on the combined data set of morphology and matK of the Hamamelidaceae ...... 233
FIG.4.10. Single shortest tree of 7338 steps based on the combined data set of ITS and matK of the Hamamelidaceae ...... 235
FIG.4.11. Single shortest tree of 7546 steps based on the combined data set of morphology, ITS and matK of the Hamamelidaceae ...... 237
FIG.4.12. Single most parsimonious tree found in the analysis of the combined data of morphology, ITS and matK. showing distribution of character states of leaf form in the Hamamelidaceae ...... 239
FIG.4.13. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK. showing distribution of the number of ovules per carpel in the Hamamelidaceae ...... 241
FIG.4.14. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK. showing distribution of the character states of petals in the Hamamelidaceae ...... 243
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIG.4.15. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK. showing distribution of pollination types in the Hamamelidaceae...... 245
FIG.4.16. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK. showing distribution of character states of sexuality in the Hamamelidaceae...... 247
FIG.4.17. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK. showing distribution of character states of chromosome base numbers in the Hamamelidaceae ...... 249
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF APPENDICES
APPENDIX 1.1. Herbarium and chemically fixed specimens observed for this s t u d y ...... 72
APPENDIX 2.1. Aligned sequences of the chloroplast gene matK in the Hamamelidaceae ...... 119
APPENDIX 2.2. Amino acid sequences translated from the matK gene sequences ...... 140
APPENDIX 3.1. Aligned nrDNA ITS sequences in the Hamamelidaceae...... 191
xiv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT
SYSTEMATICS OF THE HAMAMELIDACEAE BASED ON
MORPHOLOGICAL AND MOLECULAR EVIDENCE
by
Jianhua Li University of New Hampshire, December, 1997
The Hamamelidaceae are a plant family of 31 genera and
more than 140 species distributed in both the Old and New
worlds. There has long been debate about the subfamilial,
tribal, as well as generic, relationships within the family.
Parsimony analysis was performed to examine the
relationships in the Hamamelidaceae based on morphology, and
DNA sequences of the Internal Transcribed Spacers (ITS) of
nuclear ribosomal DNA, and the chloroplast matK gene. The
morphology-based phylogeny shows weak support for all clades
except for the Altingioideae, and differs greatly from the
molecular phylogenies. Phylogenies based on the matK and ITS
data are basically congruent. The consensus of the most
parsimonious trees generated from separate analyses of the
morphology, ITS and matK data sets resolves few
relationships. However, the analysis using the combined data
set of morphology, ITS and matK gene data resolves all
suprageneric relationships of the Hamamelidaceae.
xv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In the combined phylogeny, the first clade containing
Altinqia and t,iqiHriainhar is sister to the second clade
including the other genera of the Hamamelidaceae, supporting
the recognition of the Altingioideae. Exbucklandia and
Rhodoleia form a clade, suggesting a close relationship
between the two genera. Mytilaria forms its own clade, as
does Disanthus. Paraphyly of Disanthus, Mytilaria and
Exbucklandia does not support the combination of these genera
in one subfamily. Disanthus is the immediate sister taxon to
the clade of the Hamamelidoideae.
A phylogenetically annotated classification system is
presented here that reflects natural relationships of the
genera, facilitates communication of taxon relationships, and
minimizes nomenclatural modifications. This classification
system recognizes six subfamilies, and six tribes in the
subfamily Hamamelidoideae: 1) Altingioideae,
2) Rhodoleioideae, 3) Exbucklandioideae, 4) Mytilarioideae,
5) Disanthoideae, and 6) Hamamelidoideae. Tribes are
1) Corylopsideae, 2) Loropetaleae Trib. Nov.,
3) Hamamelideae, 4) Fothergilleae, 5) Eustimgateae, and
6) Dicorypheae Trib. Nov.
The combined phylogeny manifests that parallel evolution
has occurred for several important morphological characters
that have been used to classify the family, including loss of
petals, wind pollination, and monoecy.
xvi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION
General Information of the Hamamelidaceae
Diagnostic Features
The Hamamelidaceae (Witch-hazel) are a family of plants
in the subclass Hamamelidae (Cronquist, 1981). Plants of
this family can be identified by the following morphological
characteristics. Trees or shrubs. Leaves persistent or
deciduous, generally alternate or rarely opposite, stipulate,
simple or palmately lobed; venation pinnate or palmate;
epidermal trichomes simple, stellate or peltate scales.
Inflorescences racemose, spicate or capitate, sometimes
subtended by showy bracts. Flower morphology very variable,
ranging from complete to incomplete, from bisexual to
andromonoecious, and to monoecious; calyx composed of four or
five united sepals, variable or lacking (absent in Distylium
S.et Z. and Distvliopsis Endress); corolla of four or five
distinct petals, reduced or lacking (absent in Liquidambar
L ., Parrotiopsis Schneider, Fotherqilla Murray,
Semiliquidambar Chang, Distylium. Matudaea Lundell,
Molinadendron Endress, Parrotia C.A.Mey, Sycopsis Oliv. and
Altinqia Nor.); stamens (2)-4, 5, 10-(>20), anthers
tetrasporangiate (bisporangiate in Exbucklandia R.W.Br. and
Hamamelis L.); ovaries 2-carpellate, superior, semi-inferior,
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. or inferior, each locule with one or more ovules. Fruit, a
capsule, with woody exocarp and horny endocarp. Seeds one to
several per carpel, winged or unwinged; seed coat bony,
embryo straight, and endosperm albuminous.
Monophyly of the Hamamelidaceae
The Hamamelidaceae have long been recognized as a
natural group (Bentham and Hook, 1865; Reinsch, 1889;
Niedenzu, 1891; Harms, 1930; Bogle and Philbrick, 1980;
Endress, 1989c). Moreover, the monophyly of this family has
been supported by phylogenetic analyses based on both
morphological (Hufford, 1992) and molecular data (Hoot and
Crane, 1996; Hoot et al., 1997; Li et al., unpublished).
Floristics and Geographic Distribution
There are 31 genera in the Hamamelidaceae as of this
date (Zhang and Lu, 1995). The hamamelidaceous plants are
distributed in every Continent (Fig. i.l ). Whereas most of
the genera are distributed in the Northern Hemisphere,
several genera occur in the Southern Hemisphere, including
Trichocladus Pers., Ostrearia Baill., Neostrearia L.S.Smith,
Noahdendron Endress, Hyland et Tracey, and Dicoryphe Du
Petit-Thouars. Many genera in the Hamamelidaceae are
narrowly endemic to one island or a small geographic area.
For instance, Trichocladus is endemic to southeastern Africa;
Ostrearia. Neostrearia and Noahdendron cure distributed only
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in Queensland, Australia; Mainqaya Oliv. is endemic to
Malaysia, Dicoryphe to Madagascar, Chunia Chang to Hainan,
while Fortunearia Rehd.et Wils., Sinowilsonia Hemsl.,
SemilicpTi-riamhar. and Shaniodendron Deng, Wei et Wang can be
found only in China. However, there are two genera in the
family whose species are widely and disjunctly distributed in
several continents. One of them is Liquidambar. which can be
found in central Asia, southeastern Asia, southeastern North
America and Central America (Bogle, 1968, 1986); the other is
Hamamelis, which contains species distributed in southeastern
Asia, southeastern North America and Central America (Mione
and Bogle, 1990).
There are about 140 species in the Hamamelidaceae (Zhang
and Lu, 1995). However, more than half of the genera in the
Hamamelidaceae are either monotypic (single species) or
oligotypic (several species). Only a few genera contain more
than ten species, e.g., Corylopsis S.et Z. (7-36), Dicoryphe
(10-20), and Distylium (10-18). However, the number of
species in most of the large genera is uncertain, and all
groups need revision.
Zhang and Lu (1995) analyzed the ecogeographical pattern
of the genera in the Hamamelidaceae and recognized two types
and nine subtypes of distribution: I. Tropical distributions
(18 genera), including: 1) tropical Asia, 2) tropical Central
America, 3) tropical Africa, and 4) tropical Australia; and
II. Temperate distributions (13 genera), including: 1) Asian,
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2) western Asian, 3) western Asian-eastern Asian-southeastern
North American, 4) eastern Asian-North American disjuncts,
and 5) southeastern North American. Zhang and Lu (1995) also
made a statistical analysis of the genera and species
distributed in different floristic regions recognized by
Takhtajan (1969). These included Eastern Asiatic Region,
81/16 (species/genera); North-American Atlantic Region, 6/3;
Iran-Turanian Region, 3/3; Sudano-Zambezian Region, 3/1;
Madagascan Region, 13/1; Indo-Chinese Region, 40/12; Malesian
Region, 6/4; Caribbean Region, 8/4; Cape Region, 2/1; and
Northeast Australian Region, 3/3. Several genera are
distributed in more than one region.
Economic Significance
Many species in the Hamamelidaceae are large trees whose
wood has long been used as lumber for different kinds of
construction and furniture making (Chang, 1979). The best
known ones are r.igm'danihar and Altinaia. Several tree
species have long been planted as street trees in many
cities, such as Liouidambar, while many shrubby species are
widely cultivated as ornamentals, including Corylopsis.
Fotherailla. Hamamelis. and Loropetalum R.Br.ex Reichb. A
fragrant gum (storax) used in perfumery, as an expectorant,
inhalant and as a fumigant in treatment of skin diseases is
derived from Liouidambar styraciflua and lu orientalis
(western Asia). Liquidambar species are also frequently
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. grown for their colorful autumn foliage. Hamamelis
virainiana yields the widely used astringent and soothing
lotion for cuts and bruises known as 'Witch Hazel'. Water
diviners ('dowsers') favor Witch hazel twigs for their
dowsing (Mione, 1987).
Brief History of Systematics of the Hamamelidaceae
The name Hamamelidaceae dates back to 1818 when Brown
recognized it as a natural group, including the four genera
Hamamelis L., Dicoryphe Thouars, Dahlia Thunb.
(= Trichocladus Pers.), and Fotherailla L. f. Since then
more genera have been continually added to the Hamamdlidaceae
(See Bogle, 1968). As a result, many botanists have studied
the natural relationships of the taxa in the Hamamelidaceae.
De Candolle (1830) recognized two tribes within the
family: the tribe Hamamelideae (Hamamelis, Dicoryphe. and
Trichocladus) and the tribe Fothergilleae (Fotherqilla).
Gardner (1849) proposed a broad concept of the
Hamamelidaceae, including the segregate families
Altingiaceae, Bruniaceae and Helwingiaceae.
Bindley (1853) divided all the known genera of the
Hamamelidaceae into two groups: Group I, characterized by
solitary ovules, included Dicoryphe. Corylopsis.
Trichocladus. Hamamelis. Loropetalum. Parrotia, Fotherqilla.
and Distylium. Group II, characterized by several ovules in
each locule, consisted of Bucklandia (=Exbucklandia).
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sedaewickia (= Altinqia). Rhodoleia Champ.ex Hook, Eustiama
Gardn.et Champ, and Tetracrypta (= Anisophyllea.
Anisophylleaceae).
Oliver (1862) arranged the uniovulate genera of the
Hamamelidaceae into three groups, based on the absence or
presence, and on the forms of petals. The apetalous group
consisted of Parrotia. Fotherqilla. Distylium. and Sycopsis.
The second group, with linear-elongate, lanceolate to
spathulate petals, was composed of Corvlopsis. Dicoryphe.
Hamamelis. Loropetalum. and Trichocladus. The third group,
with squamiform petals, was made up of Tetrathvrium Benth.
and Eustioma.
Bentham and Hooker (1865) proposed a system of the
Hamamelidaceae which contained 15 genera. They firstly
divided the family into two groups based on the number of
ovules per locule, a uniovulate-locule group and a
multiovulate group. The first group was the same as those
three groups of Oliver (1862, see preceding paragraph), while
the second included Rhodoleia. Bucklandia (= Exbucklandia).
Altinqia. and Liquidambar.
Reinsch (1889) was the first botanist to combine data
from both morphology and anatomy, and to provide a natural
classification system for the family with a hierarchy of
subfamilies. In this system, he recognized three
subfamilies: the Altingioideae, Bucklandioideae, and
Hamamelidoideae (Fig. i.2).
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Niedenzu (1891) revised Reinsch's system by: 1) reducing
subfamilies Altingioideae and Bucklandioideae into the rank
of tribe under the subfamily Bucklandioideae; 2) removing
Mvrothamnus (= Myrothamnaceae) from the family, and 3)
recognizing two tribes in the Hamamelidoideae, Parrotieae,
which included Distylium. Parrotia, Fotherqilla. and
Corvlopsis. and Hamamelideae, which contained Tetrathyrium.
Mainqava. Loropetalum. Eustioma. Sycopsis. Hamamelis.
Dicoryphe. Trichocladus. and Franchetia in the
Hamamelidoideae.
Tong (1930) was the first to devise a classification
system of the Hamamelidaceae explicitly on the basis of
floral evolution. After analyzing the possible evolutionary
course of floral characters, he proposed several criteria for
determining primitive flowers in the Hamamelidaceae: 1)
flowers hermaphroditic, 2) floral organs in definite numbers,
3) sepals only slightly connate, 4) petals well developed, 5)
stamens variable, and 6) ovules in each locule numerous.
Consequently, he revised Niedenzu's system to reflect his
hypothesis of floral character evolution (Fig. i.3).
Harms (1930) made a considerable revision of Niedenzu's
system. In fact, his revised classification scheme had been
considered as the most comprehensive until a more recent one
was proposed by Endress (1989b). In his system of the
Hamamelidaceae, shown in Fig. i.4, Harms (1930) established
five subfamilies, the largest of which, the Hamamelidoideae,
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was further subdivided into five tribes. However, he did not
place the two, then little-known genera, Mytilaria Lecomte
and Ostrearia.
Chang (1948, 1973, 1979) basically adopted Harms's
system, but he erected a new subfamily Mytilarioideae to
include Mytilaria and his new genus Chunia (Fig. i.5).
Schulze-Menz (1964) transferred Sinowilsonia from the
Tribe Distylieae into the Corylopsideae, which consisted of
Corvlopsis and Fortunearia.
More recently, Endress (1989a) reviewed the previous
systems, especially those of Harms and Chang. He recognized
four subfamilies for the Hamamelidaceae, including
Altingioideae, Exbucklandioideae, Rhodoleioideae and
Hamamelidoideae (Fig. i.6). His Exbucklandioideae is a
combination of Exbucklandioideae, Disanthoideae, and the
Mytilarioideae. For the largest subfamily Hamamelidoideae,
Endress (1989b, c) combined two tribes, Fothergilleae and
Distylieae, forming the tribe Fothergilleae sensu lato. He
not only recognized the replacement of Sinowilsonia by
Schulze-Menz but also united Fortunearia and Sinowilsonia
with Eustigma, thus establishing the Eustigmateae sensu lato.
Further, he separated the five genera of the Southern
Hemisphere (Ostrearia, Neostrearia. Noahdendron. Dicoryphe,
and Trichocladus1 from the rest as a subtribe Dicoryphinae in
the tribe Hamamelideae.
Deng et al. (1992a) described a segregate genus
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Shaniodendron and placed it in the Fothergilleae.
Systematic Problems of the Hamamelidaceae
There is no doubt that the systematics of the
Hamamelidaceae has evolved to a better understanding of the
relationships of the taxa due to the accumulation of more
data from studies in a number of different disciplines,
including morphology, anatomy, embryology, floral ontogeny,
and cytology (Baillon, 1871, 1871-1873; Harms, 1930; Tong,
1930; Chang, 1948, 1962, 1973, 1979; Endress, 1967, 1976,
1989a, b, c; Bogle, 1968, 1970, 1986, 1987a, b, 1989, 1990,
1991; Melikian, 1973a, b, 1975; Rao, 1974; Goldblatt and
Endress, 1977; Bogle and Philbrick, 1980; Wisniewski and
Bogle, 1982; Mione and Bogle, 1990; Pan et al., 1991).
However, the Hamamelidaceae are so widely distributed in
several continents of the World that it is almost impossible
to access every genus for many studies, especially for those,
such as embryology and floral ontogeny that require serial
collections of flowers and fruits once every three to four
days for at least one growing season. Both field study and
laboratory research are costly, while financial assistance,
on the other hand, is hard to acquire for systematics
projects. Encouragingly, as sciences and technologies that
are related to systematics develop, systematists get more and
more new characters that might be used to better estimate
phylogenies, thus providing a basis for revisions of previous
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. classification schemes. Therefore, every system reflects
what has been understood up to that time. This can be
attested by the differences among the several suprageneric
systems of the Hamamelidaceae (Harms, 1930; Tong, 1930;
Chang, 1973, 1979; Endress, 1989b, c).
As can be summarized from the controversies concerning
suprageneric relationships of the Hamamelidaceae, several
problems need to be addressed: 1) How many subfamilies should
be recognized in the Hamamelidaceae so that each subfamily
represents a monophyletic group? 2) What are the
relationships among the subfamilies? 3) Within the
Hamamelidoideae, how many monophyletic tribes can be
recognized? 4) Are the defining features in the
Hamamelidaceae derived from one ancestral form? In other
words, how many times have some of the morphological
characters in the family evolved?
Parsimony Analysis and Molecular Systematics
Given the assumption that evolution takes place in a
minimum number of steps, maximum parsimony analysis has been
created to search for the most parsimonious, i.e., the
shortest, phylogenetic trees that should reflect the
evolutionary relationships of organisms. This type of
analysis has been used extensively and proven to be effective
in resolving relationships at various levels in many groups
of organisms (Olmstead and Palmer, 1994).
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Molecular data, especially DNA sequences, have been
widely favored for studying natural relationships among
organisms because DNA sequence differences are free of non-
heritable environmental perturbations that obscure true
genetic relationships (Doyle, 1993).
Objectives
The objectives of this study include the following: 1)
to conduct parsimony analyses using both morphological and
molecular data, thus being able to assess the suprageneric
systems previously proposed for the Hamamelidaceae; 2) to
understand the possible evolutionary directionalities of
important characters in the family; 3) to provide a revised
classification system for the family so that only
monophyletic groups will be recognized.
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. i . 1. Geographic distribution of the Hamamelidaceae (Drawn based on Elias, 1980; Heywood et al., 1993; Zhang and Lu, 1995; and Rakotobe, 1996).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.i.2. Classification system of the Hamamelidaceae proposed by Reinsch (1889) formatted into a cladogram.
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altinqia Altingioideae Myrothamnus Liquidambar Bucklandia Bucklandioideae Disanthus Rhodoleia Eustigma Maingaya Fothergilla (D M H- Loropetalum CL 0 Hamamelis H- CL (D Tetrathyriurn 0) (D Dicoryphe Franchetia Corylopsis Parrotia Trichocladus Distylium Sycopsis
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.i.3. Generic relationships of the Hamamelidaceae depicted by Tong (1930), formatted into a cladogram. Arrows indicate the proposed ancestral taxa in the original diagram.
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bucklandia Rhodoleia
Disanthus
Altingia •Liquidambar
Mytilaria
Distylium
Sycopsis
Parrotia
Parrotiopsis
Fotherqilla
Sinowilsonia
Eustigma
.Corylopsis
Fortunearia
Maingaya
Loropetalum
Hamamelis
Trichocladus
Dicoryphe
Franchetia
17
i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.i.4. Classification system of the Hamamelidaceae proposed by Harms (1930), formatted into a cladogram.
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Disanthoideae Disanthus Altingioideae Liquidambar Altingia
Disylium — Distylieae Sycopsis Sinowilsonia Parrotia Fothergilleae Parrotiopsis Fothergilla Hamamelis — Loropetalum
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. i.5. Classification system of the Hamamelidaceae (non- Chinese genera not included, Chang, 1973), formatted into a cladogram.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Disanthoideae Disanthus
Altingia Altingioideae Semiliquidambar
Liquidambar Rhodoleioideae Rhodoleia Exbucklandioideae Exbucklandia
Mytilarioideae Chunia c Mytilaria
Distylium Distylieae Sycopsis
Sinowilsonia
Parrotia Fothergilleae Parrotiopsis
Fothergilla Hamamelidoideae Corylops ideae Corylopsis c Fortunearia Eustigmateae Eustigma
Hamamelis
Embolanthera Hamamelideae Tetrathyrium
Loropetalum
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. i.6. Classification system of the Hamamelidaceae proposed by Endress (1989b, c), formatted into a cladogram.
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia Altingioideae Semiliquidambar Liquidambar Rhodoleioideae Rhodoleia Mytilaria Exbucklandioideae Chunia Exbucklandia Disanthus Distylium Matudaea Distyliopsis Sycopsis Parrotia Fothergilleae Fothergilla Parrotiopsis Molinadendron Eustigmateae Eustigma Fortuneeuria Sinowilsonia Hamamelidoideae Corylopsis Hamamelis Hamamelidinae Embolanthera Maingaya Tetrathyriurn ■ Loropetalum J
.Dicoryphe Loropetalinae Trichocladus Dicoryphinae Ostrearia 'Neostrearia • Noahdendron
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER I
A PHYLOGENETIC ANALYSIS OF THE HAMAMELIDACEAE
BASED ON MORPHOLOGICAL DATA / INTRODUCTION
Morphology, or the external form of an organism, has
been and still is the type of data used most in plant
classification (Davis & Heywood, 1963).
Today, even though more efforts and financial support
are directed to using molecular markers to study natural
relationships of plants, non-molecular data, including
morphology, anatomy, embryology and floral ontogeny, etc.,
are still invaluable systematically. They produce
phylogenetically informative data for resolving issues such
as homology, and for assessing relationships among taxa at
different levels. This is attested by the fact that in many
taxa whose systematics has been studied on the basis of both
non-molecular and molecular data, the phylogenies produced
from the two sources are to some extent congruent (Donoghue
and Sanderson, 1992).
In the Hamamelidaceae, three major classification
systems have previously been proposed in recent decades
(Harms, 1930; Chang, 1973; Endress, 1989a). It can be seen
from Figs. i.4-i.6 that there are many differences among the
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. three systems. Endress (1989b) also published a polytomous
tree showing his view of the phylogenetic relationships in
the subfamily Hamamelidoideae. However, until the present a
comprehensive phylogenetic analysis based on morphological
data has not been conducted for the family. Therefore, the
objectives of this study are dual: 1) to conduct a parsimony
analysis of the Hamamelidaceae using morphological
characters; and 2) to evaluate the existing systems based
upon the analysis.
MATERIALS AND METHODS
There are 31 genera in the Hamamelidaceae. In this
analysis, however, 30 genera were used. The one excluded
genus is Semiliquidambar. which is morphologically similar to
Licruidambar. and is possibly a hybrid between the latter and
Altinaia (Chang, 1948; Bogle, 1986).
Data collection was carried out mostly by using
literature information, as many observations have been made
by various researchers in the last 250 years (Murray, 1774;
Noronha, 1790; Persoon, 1807; Du Petit-Thouars, 1804; Brown,
1818; Candolle, 1830; Meyer, 1831; Siebold and Zuccarini,
1835; Gardner and Champion, 1849; Bentham, 1861, 1865;
Maximowicz, 1866; Baillon, 1871, 1871-1873; Reinsch, 1889;
Oliver, 1860, 1862, 1872, 1890; Niedenzu, 1891; Schneider,
1905; Hemsley, 1906; Merrill, 1909; Rehder and Wilson, 1913;
Lecomte, 1924; Harms, 1930; Lundell, 1940; Brown, 1946;
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Smith, 1958; Bogle, 1968, 1987; Chang, 1948, 1973, 1979;
Endress, 1969, 1970, 1977, 1978a, 1993; Melikian, 1973a, b;
Endress et al., 1985; Deng et al., 1992a, b; Zhang and Lu,
1995). I also did my own first-hand investigations on some
of the characters, either to test previous observations or to
fill the previously missing data for some taxa. For such,
herbarium specimens were borrowed from Missouri Botanical
Garden (MO), United States National Herbarium at Smithsonian
Institute (US), Hong Kong Herbarium (HK), and Harvard
University Herbaria (GH).
I also dissected chemically fixed specimens of most of
the genera in the family using a Wild Heerbrugg M-5
stereoscope with multiple magnifications. The materials
observed are listed in Appendix 1.1.
To study the leaf epidermis of a dozen genera that had
not previously been observed, leaves from either herbarium or
F.A.A. (formalin: glacial acetic acid: 70% alcoho1=90:5:5)
fixed specimens were bleached with 5% sodium hypochlorite
(Chlorox) for one to two days depending on leaf thickness.
Then, both lower and upper epidermis layers were cleaned with
a pen-brush, stained with safranin, dehydrated via alcohol
series, and mounted with Permount mounting medium (Fisher
Scientific Co., NJ).
Parsimony analyses were conducted using PAUP 3.1.1
(Swofford, 1993) with a Powermac 7200/120 computer.
Cercidiphvllum ~i aponicum was chosen as the outgroup since
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. recent molecular studies (Chase et al., 1993; Hibsch-Jetter
and Soltis, 1996; Hoot and Crane, 1996; Qiu et al.,
unpublished) have all suggested that this species was a
sister group to the Hamamelidaceae.
Given the limit of computer memory and the size of the
data set, the heuristic tree search option was used with the
TBR (Tree Branch Reconnection) branch swapping strategy,
coupled with steepest descent off and multipars on. The
support for each branch of the cladogram was first assessed
by calculating all cladograms one, two, etc., steps longer
than the most parsimonious ones. The corresponding strict
consensus trees show the stability of groups when longer
trees were allowed. This was originally proposed by Farii et
al. (1982) for distance analysis, and by Bremer (1988) for
parsimony analysis (see Gustafsson and Bremer, 1995). This
support was called "Bremer support" or "decay index"
(Donoghue et al., 1992). Bootstrap analysis was also applied
to show the relative strength of branches (Felsenstein,
1985).
Owing to a large number of trees obtained upon the
relaxation of parsimony and the memory capability of the
computer, bootstrap analysis could not get to 100 replicates.
Alternatively, bootstrap analysis was conducted for 1000
replicates with multipars off, and 10 trees were saved in
each bootstrap replicate. To avoid any a priori assumptions,
all characters and their states were unweighted, and every
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. multistate transformation was treated as unordered.
RESULTS
Characters and their states
A total of 70 characters were collected. However, only
52 of these were used in this analysis because either the
information for the other 18 characters was missing for most
of the taxa in the Hamamelidaceae, or the characters were
invariable throughout the family. Table 1.1 lists the
characters and their state coding used for this analysis.
The character numbers are as those used in the following
character descriptions. The delimitations of subfamilies,
tribes, and subtribes for the character discussion followed
the system of Endress (1989c), unless otherwise indicated.
1. Habit - evergreen (0), semi-evergreen (1), deciduous (2 ).
In the Hamamelidaceae, nine genera are deciduous and
distributed in Temperate regions of the Northern Hemisphere,
including Disanthus Maxim., Fortunearia. Fothergilla.
Hamamelis. Liquidambar, Parrotia. Parrotiopsis.
Shaniodendron. and Sinowilsonia (Bogle, 1968, 1970; Endress,
1993; Zhang and Lu, 1995). In Coryloosis. while most species
are deciduous, a couple of species are semi-evergreen, such
as CL multiflora Hance. In contrast, the rest of the genera
are evergreen and mainly distributed in subtropical or
tropical areas (Zhang and Lu, 1995).
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 . Leaf venation - pinnate (0), intermediate (1), palmate
(2 ).
Most of the genera in Hamamelidaceae possess pinnate
venation (e.g., Sycopsis. Distylium. and Distylioosis. while
some other genera have palmate venation, including
Licruidambar. Exbucklandia. Chunia, Mytilaria, and Disanthus.
Intermediate venation pattern (pinnate venation with basal
crowding of secondary veins), however, is found in several
genera, such as Corylopsis. Fotherailla. Hamamelis, Parrotia
Parrotiopsis. and Shaniodendron (Harms, 1930; Endress, 1993;
Zhang and Lu, 1995).
3. Number of prophylls - one (0), two (1), three (2).
In most of the genera of the Hamamelidaceae there is a
solitary prophyll, while two prophylls are found in Matudaea
Molinadendron. Fortunearia. and Sinowilsonia. Corylopsis is
the only genus with three prophylls on the branch (Hufford
and Crane, 1989; Endress, 1993).
4 . Phyllotaxy - alternate (0), spiral (1), opposite (2).
Leaves of the genera of the Hamamelidaceae are
alternately arranged with the exception of Altingioideae,
where a spiral pattern can be observed, and of Trichocladus
and Dicoryphe. where occasionally subopposite or opposite
leaf arrangement occurs (Harms, 1930; Endress, 1993; Zhang
and Lu, 1995).
5. Peltate leaf blade - absent (0), present (1).
Leaves in the Hamamelidaceae are petiolate, and in most
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. genera petioles are basifixed. However, in Molinadendron,
Dicoryphe. and Trichocladus. petioles tend to be peltate
(Baillon, 1871, 1871-1873; Endress, 1969, 1989b, 1993).
6 . Leaf margins - entire (0 ), crenate (1), half serrate
(2), serrate (3), lobed (4), lobes serrate (5).
Three or five-lobed leaves are predominant in
Altingioideae (Altiniaa. Liquidambar) and Exbucklandioideae
(Exbucklandia and Mytilaria). However, in Altingia,
Exbucklandia. and Mytilaria. both simple and lobed leaves can
be found on an individual tree (Harms, 1930; Hutchinson,
1967; Bogle, 1968; Chang, 1979; Endress, 1993; Zhang and Lu,
1995), showing leaf dimorphism in these genera. In
Liquidambar. leaf lobes are serrate. Simple leaves are
typical of the rest of the genera in the Hamamelidaceae, the
evergreen genera of which have entire leaves, but several
taxa have half-serrate leaves, or leaves serrate in the upper
half to 1/3, including Distylium. Distyliopsis. Eustiama.
Svcopsis. and Tetrathyrium. Whereas most of the deciduous
taxa have toothed leaf margins, the leaves of Hamamelis are
somewhat crenate (Bradford and Marsh, 1977).
7. Modal Anatomy - trilacular (0), multilacunar (1).
The genera of the family typically have
trilacunar-three-trace nodes (Sinnott and Bailey, 1914;
Metcalfe and Chalk, 1979; Li et al., 1993) with the exception
of Chunia and Mytilaria. where multilacunar nodal anatomy has
been reported (Bogle, 1990, 1991).
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8. Secretory canals (gum ducts) - absent (0), present
(1).
These structures occur in the stems and petioles of the
Altingioideae, Chunia, Mytilaria, and Ostrearia (Baillon,
1871; Harms, 1930; Huang and Lee, 1982; Huang, 1986).
9. Leaf tooth type - absent (0), Altingioid (1),
Fothergilloid (2).
The Altingioid type of leaf tooth occurs exclusively in
the Altingioideae of the Hamamelidaceae, while the
Fothergilloid type is represented in the rest of the
hamamelidaceous genera (Li and Hickey, 1988).
10. Stomatal apparatus - Paracytic (0), Stephanocytic (1),
Cyclocytic (2), Anomocytic (3).
The Paracytic type of stomatal apparatus prevails in
the Hamamelidaceae (this study, Fig. l.la-1; Metcalfe and
Chalk, 1979; Pan et al., 1990). The Stephanocytic type
occurs only in Mvtilaria and Tetrathyrium. the Cyclocytic
type is found in Exbucklandia and Rhodoleia (Pan et al.,
1990). The Anomocytic type occurs with the Paracytic type in
Corylopsis. Hamamelis. Fotheroilla. and Sinowilsonia.
11. Leaf trichomes - simple (0), glandular (1), stellate
(2), scale (3).
Simple or 2-branched trichomes occur in Altingioideae
(Liquidambar and Altingia) and Exbucklandioideae
(Exbucklandia. none were found in Chunia, and Disanthus)
(Warren Davis, University of New Hampshire, unpublished).
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stellate trichomes are represented in most of the genera of
the Hamamelidoideae with the exception of Distylium.
Distyliopsis, Tteostraari a. Noahdendron. Ostrearia, Sycoosis.
and Rhodoleia. where the scale type of trichome is found (Li
and Hickey, 1988; Fang and Fan, 1993; Warren Davis,
unpublished). Multicellular glandular hairs are found on
petioles and primary veins of several species of Corylopsis.
blade margins of the primary leaves of Matudaea, and stipules
or bracts of Matudaea, Embolanthera Merr., and Tetrathyrium
(Endress, 1989b).
12. Foliar sclereids - absent (0), columnar (1), fusiform
(2), libriform (3), and idiofibrosclereids (4).
Four types of foliar sclereids were defined by Li and
Bogle (1995) in the Hamamelidaceae. No sclereids were found
in Altingioideae, Eustigmateae (excl. Eusticnna). and
Corylopsideae, while the columnar and fusiform types occur in
Hamamelis. the Exbucklandioideae and the Fothergilleae sensu
stricto. Libriform and idiofibrosclereids exist in the
Loropetalinae and Dicoryphinae.
13. Stipule shape - absent (0), linear (1), leafy but not
appressed (2), leafy and appressed (3).
Leaves of the Hamamelidaceae are stipulate, but in
Rhodoleia stipules are not formed in all leaves of a shoot
(Endress, 1978b). Stipules are small and linear in the
following genera: Disanthus. Distylium. Distyliopsis.
Fortunearia. Sinowilsonia. Eusticnna. Trichocladus. Among the
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. genera containing leafy stipules, Chunia, Exbucklandia. and
Mytilari a have stipules that are paired up and appressed to
each other (Endress, 1969, 1989b, c, 1993; Bogle, 1986).
14. Stipule habit - persistent (0), deciduous (1).
Stipules are deciduous in most of the genera of the
Hamamelidaceae, but in Chunia. Dicoryphe. Exbucklandia,
Mytilaria. and Noahdendron stipules are persistent (Chang,
1948, 1973, 1979; Endress, 1989a).
15. Ovary pubescence - absent (0), present (1).
Glabrous ovaries are found in the Altingioideae,
Exbucklandioideae, Rhodoleia, Eustigma, Fortunearia. and a
couple of species of Corylopsis (Baillon, 1871; Harms, 1930;
Tong, 1930; Hutchinson, 1967; Endress, 1967, 1976, 1989a, b,
1993; Bogle, 1968, 1986).
16. Ovary lenticels - absent (0), present (1).
Eustiama and Fortunearia are the only two genera in the
Hamamelidaceae that have lenticels on the ovaries (Chang,
1979; Endress, 1989b, c, 1993).
17. Number of flowers per inflorescence - two (0), more
than two (1).
The number of flowers per inflorescence varies
considerably in the Hamamelidaceae, or even within a genus,
such as from 5 to more than 30 in Corylopsis (Li et al.,
1993). However, two flowers per inflorescence are found
almost constantly in Disanthus (Maximowicz, 1866; Pan et al.,
1991).
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18. Flower parts - five (0)f four (1), variable (2).
The number of flower parts, especially stamens, is 5-
merous in most of the petaliferous genera of the
Hamamelidaceae, while Hamamelis is constantly 4-merous and
several genera have both 5- and 4-merous flowers, such as
Loropetalum. Dicoryphe. and Trichocladus (Baillon, 1871-1873;
Harms, 1930; Tong, 1930; Mione and Bogle, 1990). In
contrast, in most of the apetalous genera, the number of
stamens is variable, such as 15-45 in Shaniodendron (Deng et
al., 1992a, b; Hao et al., 1996), and 21-25 in Parrotiopsis
(Bogle, 1970; Endress, 1976, 1993).
19. Sexuality - bisexual (0), andromonoecious (1),
monoecious (2).
Flowers are both morphologically and functionally
bisexual in petaliferous genera and several apetalous members
of the Hamamelidaceae. However, in several apetalous genera,
both stamen and pistil are present in a flower, but one is
not functional, thus the flower is functionally unisexual.
In this case, both pistillate and staminate flowers are
generally on the same individual, thus giving rise to
monoecy. These taxa include the Altingioideae and
Sinowilsonia (Harms, 1930; Hutchinson, 1967; Bogle, 1968;
Bogle and Philbrick, 1980; Endress, 1993). Andromonoecy
(both bisexual and staminate flowers on the same individual)
occurs in some taxa, including the Fothergilleae (except for
Matudaea and Molinadendron) and Exbucklandia (Bogle, 1970;
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chang, 1979; Endress, 1993).
Fortunearia has long been believed to be monoecious
because the pollen grains were thought to be non-functional
in pistillate flowers (Chang, 1979; Endress, 1993). I
collected fresh pollen grains from a small tree of
Fortunearia cultivated in the greenhouse of the University of
New Hampshire, and conducted a pollen viability test on the
pollen from staminate and pistillate flowers using the
Aniline Blue method (Bradford et al., 1974). More than 500
pollen grains were counted and the results showed that about
84% of the pollen grains were viable in both staminate and
"pistillate" flowers. Therefore, this genus is
andromonoecious instead of being monoecious.
20. Pollination - bird (0), insect (1), wind (2).
Rhodoleia is the only genus that is pollinated by birds
in the Hamamelidaceae (Harms, 1930; Bogle, 1989; Endress,
1989b). Although species with showy petals are insect
pollinated, several hamamelidaceous species that have lost
their petals in the course of evolution continue to have
insect pollination because of the specialization, or
modification of other parts, either floral or vegetative, to
attract insects, such as the whitish, showy stamens in
Fotherqilla. and the large, white inflorescence bracts in
Parrotiopsis (Harms, 1930; Bogle, 1968, 1970; Endress,
1989b). Both Eustiama and Fortunearia have rather reduced
petals, but they share purplish and largely expanded stigmas,
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. especially in Eustiama. It is likely that the showy
stigmatic surfaces have a function as insect optical
attractants.
21. Flower symmetry - symmetric (0), asymmetric (1).
All of the genera in the Hamamelidaceae have symmetric
flowers except for Rhodoleia. where asymmetric flowers are
found. That is, petals are not initiated on the adaxial side
of a flower due to flower crowding and fusion by condensation
of the inflorescence into a capitulum in the course of
evolution (Harms, 1930; Bogle, 1968, 1989; Endress, 1989a).
22. Sepals - absent (0), present (1).
All of the genera in the Hamamelidaceae possess sepals
except for several Altinaia species, Distylium, Distyliopsis.
Matudaea, and Liquidambar acalycina (Endress, 1989a, 1993).
23. Fusion of sepals into a tubular structure - absent
(0), present (1).
Fusion of sepals into a tubular structure that ruptures
at anthesis and/or falls off by means of circumscissile
dehiscence occurs in Mainaaya and Embolanthera of the
Loropetalinae, in Neostrearia. Dicoryphe. Trichocladus
grandiflorus. and 1\. Goetzei of the Dicoryphinae. Sepals are
partly fused in the Fothergilleae, while in Parrotia persica
differences between individuals in the amount of fusion are
considerable (Harms, 1930; Hutchinson, 1967; Endress, 1967,
1976, 1989a, b, 1993; Bogle, 1968, 1970, 1986).
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24. Showy bracts - absent (0), present (1).
This character is present exclusively in Parrotiopsis
(Harms, 1930).
25. Fimbriate glandular hairs - absent (0), present (1).
Bracts with fimbriate glandular hairs are found only in
Matudaea and Embolanthera (Bogle, 1968; Endress, 1989a, 1993;
Zhang and Lu, 1995).
26. Bract fusion - absent (0), present (1).
Fusion of bracts to form a calyx-like structure occurs
in the American genus Matudaea (Bogle, 1970; Endress, 1989b,
c ) .
27. Petals - geniculate with two lateral auricles (0),
geniculate with two dorsal swellings (1), orbicular (2),
spathulate (3), ribbon-like and coiled circinately in the bud
(4), ribbon-like, but not coiled circinately in the bud (5),
ribbon-like and fleshy (6), ribbon-like with expanded broad
base and tapering tip (7), reduced (8), and absent (9).
Petal morphology is fairly diverse in the
Hamamelidaceae. Corylopsis is the only genus having
orbicular petals (Morley and Chao, 1977; Li et al., 1993).
Rhodoleia and Exbucklandia share spathulate petals. Many
genera in the Hamamelidaceae have ribbon-like petals, with
petals coiled circinately in the bud, including Hamamelis.
Loropetalum. Mainaaya. Neostrearia. Noahdendron. Ostrearia,
Tetrathyrium. and Trichocladus. However, Dicoryphe is an
exception, where petals are strap-shaped, but they are not
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. coiled circinately in the bud (Endress, 1989a, b ) . Mytilaria
has ribbon-like petals, but the petals are thick and fleshy.
Petals in Disanthus are broadly lanceolate and glandular at
the base (Maximowicz, 1866; Pan et al., 1991). Both
Fortunearia and Sinowilsonia have reduced petals (Rehder and
Wilson, 1913; Bogle, 1968; Chang, 1979; Endress, 1993). In
Embolanthera and Eustiama. petals have auricles, but the
auricles are lateral and more distinct in the former and
dorsal and minute in the latter. No petals are present in
the Altingioideae, Chunia and the Fothergilleae.
28. Fusion of stamen to petal - absent (0), present (1).
Fusion of petals to stamens to form a tube occurs in
some members of the Hamamelidaceae such as Embolanthera.
especially in E^. qlabrescens (Tardieu-Blot, 1965), and in
Dicoryphe. especially in D^. stipulacea. but not in D.
viticoides (Endress, 1989b).
29. Stamen organization - cyclic (0), centripetal (1),
centrifugal (2).
In most of the genera in the Hamamelidaceae, regardless
of how many stamens occur in a flower, stamens cure cyclically
initiated, however, the centripetal pattern of initiation was
found in Matudaea, while the centrifugal initiation pattern
was seen in Fotherailla (Endress, 1976).
30. Anther orientation - Introrse (0), latrorse (1),
extrorse (2).
Latrorse anthers are prevalent in the Hamamelidaceae.
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. However, in Disanthus, extrorse anthers occur (Baillon,
1871). Gardner and Champion (1849) described the anthers in
Eusticnna as extrorse. However, later observations
(Hutchinson, 1967, and this study) indicated that the anthers
are latrorse instead. Introrse anthers are found in the
Dicoryphinae, Hamamelis and Exbucklandia (Endress, 1989b, c).
31. Showy stamens - absent (0), present (1).
In Fotherailla. the stamen filaments are elongate,
clavate, whitish and showy at anthesis, whereas all the other
members of the Hamamelidaceae have simple filaments (Bogle,
1970; Endress, 1989b).
32. Filament length - equal to or longer than the anther
(0), shorter than the anther (1).
Filament lengths vary greatly in the Hamamelidaceae.
Long filaments can be found in several Corylopsis species,
Chunia, Fothergilleae (excl. Molinadendron). Rhodoleia. and
Hamamelis. while short filaments can be seen in the
Eustigmateae, the Loropetalinae, the Dicoryphinae, Disanthus,
Exbucklandia. and Mytilaria (Bogle, 1968, 1970; Chang, 1979;
Endress, 1989b).
33. Elongate anthers - absent (0), present (1).
Oblong and elongate anthers are found in the
Altingioideae, Exbucklandia. and several genera of the
Fothergilleae (Distyliopsis. Matudaea, Parrotia.
Shaniodendron. and Svcopsis) (Bogle, 1970; Endress, 1989b).
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34. Anther connective protrusion - adaxial (0), axial
(1), absent (2).
The length of the anther connective protrusion is
variable in the Hamamelidaceae (Baillon, 1871; Bogle, 1968,
1970; Chang, 1979; Endress, 1989b). Long protrusions are
found in the Loropetalinae (excl. Embolanthera 1, and in the
Dicoryphinae (excl. Neostrearia\. where protrusions become
highly adaxial, covering the inner gynoecium completely
before anthesis. In several genera, however, protrusions are
long and erect, or axial, including Exbucklandia. Eustigma,
Chunia. Corylopsis. and the Fothergilleae (excl. Parrotiopsis
and Fotherqillal. Anthers show little connective protrusion
in the rest of the Hamamelidaceae, including the
Altingioideae, Disanthus, Mytilaria, Hamamelis, Embolanthera.
Fortunearia. Sinowilsonia. and Neostrearia (Bogle, 1968).
35. Staminodial phyllomes - more than one whorls (0), one
whorl (1), absent (2).
There is a great diversity of sterile phyllomes in the
Hamamelidaceae (Bogle, 1968, 1970, 1986, 1989; Chang, 1979;
Wisniewski and Bogle, 1982; Endress, 1989b). Sterile
structures in most of the genera in the Hamamelidaceae are of
staminodial origin, while those of the Altingioideae possibly
have carpellary characteristic (Wisniewski and Bogle, 1982).
More than two whorls of the staminodial phyllomes can be
found in Corylopsis. Loropetalum. Mainaaya. and Mytilaria.
whereas one whorl has been observed in Exbucklandia.
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hamamelis. Tetrathyrium. Ostrearia. Rhodoleia, and Disanthus.
However, there is no staminodial phyllomes in Chunia,
Embolanthera. the Eustigmateae, the Dicoryphinae (excl.
several Dicoryphe species, and Ostrearia), and the
Fothergilleae.
36. Pollen exine sculpture - coarsely reticulate (0),
finely reticulate (1), foveolate (2), scrobiculate (3).
Coarsely reticulate pollen exines are most prevalent in
the Hamamelidaceae (Bogle and Philbrick, 1980). Finely
reticulate exines are found in Chunia, Distylium and
Parrotiopsis. Rhodoleia is the only genus in the
Hamamelidaceae that has scrobiculate pollen grains, while the
Altingioideae and Matudaea have foveolate pollen exines.
37. Pollen aperture - tricolpate (0), tricolporate (1),
rugate (2), polyporate (3).
Pollen aperture is rather diverse in the Hamamelidaceae.
Tricolpate pollen is the basic type in this family, while
rugate type occurs in several genera, including Chunia,
Matudaea. Molinadendron. and Sycopsis. Rhodoleia is reported
to have tricolporate pollen grains, while polyporate pollen
type is found in the Altingioideae. However, more than one
aperture pattern can be found within a genus. For example,
both tricolpate and tetracolpate pollen grains are reported
in Corylopsis and Hamamelis (Bogle and Philbrick, 1980; Zhang
and Zhang, 1991).
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38. Pollen exine - non-verrucate (0), verrucate (1).
Pollen exines of most of the genera in the
Hamamelidaceae are non-verrucate, while in several apetalous
genera or genera with reduced petals, the pollen exine is
verrucate, including the Altingioideae, Fortunearia.
Parrotiopsis. Parrotia, Sycopsis. Distylium, and
Mo1inadendron (Bogle and Philbrick, 1980).
39. Pollen aperture margins - ragged (0), smooth (1).
Smooth pollen aperture margins are found in several
genera including Dicoryphe. Fortunearia. Sinowilsonia.
Embolanthera. Parrotia. Sycopsis. Molinadendron. and
Distylium. while the ragged aperture margin occurs in the
rest of the genera in the Hamamelidaceae (Bogle and
Philbrick, 1980).
40. Pollen aperture membranes - fine (0), coarse (1).
Coarse pollen aperture membranes are the basic type in
the Hamamelidaceae, while fine pollen aperture membranes
occur in Hamamelis. Fothercrilla. Embolanthera. Mainoaya. and
Mytilaria (Bogle and Philbrick, 1980).
41. Anther dehiscence - four valves for four sacs (0), two
valves for four sacs (1), two valves for two sacs (2), simple
slits (3).
Anthers of the Hamamelidaceae are tetrasporangiate with
the exception of Exbucklandia and Hamamelis. where anthers
are bisporangiate. Anthers in several genera dehisce by
simple slits (Disanthus, Parrotia, Shaniorienrirnn. and
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sycopsis). while in most genera, anthers dehisce by valves
(Baillon, 1871; Bogle, 1968, 1970; Endress, 1989a). There
are three types of anthers with valvate dehiscence (Endress,
1989a). The first type occurs most commonly, where each
theca has two pollen sacs and opens with two valves. In the
second type, each theca has one pollen sac and opens with one
valve. This dehiscence pattern occurs in only two genera:
Hamamelis and Exbucklandia. The third anther dehiscence type
is most unusual and systematically most interesting. It is
somewhat between the two valvate types above. Although each
theca has two pollen sacs, it has only one valve serving both
pollen sacs. The valve opens toward the floral center as in
the second type. This type is known only from the subtribe
Dicoryphinae, whose genera are confined to the southern
Hemisphere, including southeastern Africa (Trichocladus).
Madagascar (Dicoryphe), and northeastern Australia
(Ostrearia, Neostrearia. and Noahdendron).
42. Inflorescences - raceme (0), spike (1), spadix (2),
capitulum (3), complex capitulum (4).
Flowers in the Hamamelidaceae are clustered into
inflorescences even when there are only a couple of flowers
in an inflorescence (such as in Disanthus). These small to
middle-sized flowers, mostly sessile, and rarely with a short
pedicel, are mostly in dense spikes or capitula. The raceme
occurs in Eusticnna. Fortunearj a. Loropetalum. and in some
species of Corylopsis. Dicoryphe. Madudaea. Molinadendron and
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sycopsis. The capitulum type of inflorescence is found in
many taxa such as the Altingioideae, Exbucklandia,
Disanthus, and Trichocladus. among which the Altingioideae
has complex capitulum (Harms, 1930; Tong, 1930; Hutchinson,
1967; Endress, 1967, 1976, 1989a, b, 1993; Bogle, 1968,
1986). Mytilaria and Chunia share the spadix type of
inflorescence (Chang, 1979).
43. Hypanthium tube - Absent (0), present (1).
Hypanthium tubes are formed in Distyliopsis. Sycopsis
and Sinowilsonia enclosing ovaries completely at anthesis
(Harms, 1930; Bogle, 1968; Chang, 1979; Endress, 1989a, b).
However, it is absent in all of the other genera of the
Hamamelidaceae.
44. Ovary position - inferior (0), semi-inferior (1),
superior (2).
The position of the ovary in the Hamamelidaceae ranges
from superior (Sycopsis. Distylium. and Matudaea) to
different levels of fusion of ovary with the hypanthium.
This character is stable within a genus with the exception of
Corylopsis and Fothergilla. where superior to semi-inferior
ovaries occur (Harms, 1930; Bogle, 1968; Chang, 1979;
Endress, 1989b, c). There are several genera possessing
inferior ovaries, including the Altingioideae, Eustigma,
Embolanthera. Loropetalum. and Tetrathyrium (Endress, 1993).
45 . Endocarp - Papery (0), thick but inseparable from
exocarp (1), thick but not ballistic (2), thick and ballistic
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (3).
Some fruits in the Hamamelidaceae show an explosive seed
discharge mechanism (Endress, 1989a). This mechanism is
caused by the ballistic activities of the thick and bony
endocarps. It is restricted to one-seeded carpels of the
Hamamelidoideae. In the Altingioideae, the endocarps are
papery and thus are not ballistic. Thick endocarps are
present in the fruits of Rhodoleia and Exbucklandia. but they
are not separable from the exocarp, thus do not discharge
seeds explosively. The endocarps in Disanthus and Mytilaria
are similar to those of the Hamamelidoideae, but they are not
ballistic (Endress, 1989a).
46. Styles - equal to or longer than the stamens (0),
shorter thcin the stamens (1).
In most of the members of the Hamamelidaceae, styles are
equal to or longer than the stamens, however, the styles are
much shorter than the stamens in several genera, including
Ostrearia. Neostrearia. Noahdendron. and Loropetalinae
(Chang, 1979; Endress, 1989b).
47. Stigmas - unexpanded (0), greatly expanded (1).
Both Eustiama and Fortunearia have expanded purplish
stigmas. However, this characteristic does not occur in any
other members of the Hamamelidaceae (Harms, 1930; Chang,
1979; Endress, 1989b, 1993).
48. Decurrent stigmas - absent (0), present (1).
In both long- and short-styled carpels, the stigma is
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. often decurrent, though in some entomophilous genera it is
more restricted to the apical region, including Rhodoleia
(Bogle, 1986), Corylopsis, and Hamamelis (Endress, 1989b).
49. Humber of ovules - three or more (0), two to three
(1), one (2).
In the Hamamelidoideae, generally, there is only one
seed in each locule of the bicarpellate ovary, although there
are several genera in this subfamily having more than one
ovule in each locule (Bogle, 1968; Chang, 1979; Wisniewski
and Bogle, 1982; Mione, 1987; Endress, 1989a, b; Pan et al.,
1991), including Corylopsis. Dicoryphe. Hamamelis.
Neostrearia. and Noahdendron. In contrast, there are more
than three ovules in each locule in the other subfamilies.
However, in most of these multiovulate subfamilies
(Rhodoleioideae, Exbucklandioideae, Liquidambaroideae), only
one or two ovules enlarge, reach maturity and become seeds,
while the other ovules remain much smaller and sterile. The
only exception is the Disanthoideae, where there are two or
more fertile seeds per carpel.
50. Seed appendage - unwinged (0), winged (1)
Most of the genera in the Hamamelidaceae do not have a
seed appendage, or seed wing. Winged seeds are found in the
Altingioideae, Chunia, and Rhodoleia (Chang, 1948, 1979;
Endress, 1989a; Zhang and Wen, 1996).
51. Capsule dehiscence - ventricidal (0), loculicidal (1),
septicidal (2).
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The capsules are ventricidal in Licruidambar and
septicidal in Disanthus. while the other genera of the
Hamamelidaceae show both ventricidal and septicidal dehiscing
patterns (Baillon, 1871; Chang, 1979).
52. Chromosome numbers (n) - 8 (0), 12 (1), 13 (2), 21
(3).
The chromosome number n=12 is typical of the
Hamamelidoideae and Rhodoleia. Mytilaria has a chromosome
number of n=13 (Pan and Yang, 1994), while n=8 is reported in
the rest of the genera, including Disanthus (2n=16),
Exbucklandia (2n=32), and the Altingioideae (2n=32)
(Goldblatt and Endress, 1977; Morawetz and Samuel, 1989;
Oginuma and Tobe, 1991; Pan and Yang, 1994).
Phylogenetic trees
The parsimony analysis generated 12 shortest trees of
219 steps, and the indices of the trees were 0.45, 0.55, 0.59
and 0.27 for consistency index (Cl), homoplasy index (HI),
retention index (RI), and rescaled consistency index (RC)
respectively. In the strict consensus tree (Fig.1.2),
Altinoia and Liquidambar formed the basal clade with a
bootstrap of 76% and a decay index of one, followed by the
clade of Chunia. Each of the following genera, Exbucklandia.
Mytilaria. Disanthus and Rhodoleia, formed its own clade and
showed paraphyletic relationships consecutively. However,
the bootstrap percentages were less than 50% and the decay
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. indices were one step for each of the successive clades.
The genera of the Hamamelidoideae formed a single clade
in both the strict consensus and the majority consensus trees
(Figs.1.2, 1.3). Within the Hamamelidoideae (Fig.1.3), the
five genera that are distributed exclusively in the Southern
Hemisphere formed a monophyletic group, in which Ostrearia
was the basal taxon followed by Noahdendron and Neostrearia.
and Trichocladus and Dicoryphe were allied in a clade.
Mainqaya, Loropetalum. Tetrathyrium. and Embolanthera were
paraphyletic to one another in the most parsimonious trees
(Figs.1.2, 1.3). The rest of the genera of the
Hamamelidoideae formed a clade, with the three genera of the
Eustigmateae being the basal branch. Sister to the
Eustigmateae was the clade consisting of the Fothergilleae
and Corylopsis-Hamamelis.
The bootstrap values were less than 50%, and decay
indices were only one step for all the clades except for a
few, including Altinqia-Liauidambar; Corylopsis-Hamamelis;
Sinowilsonia- Fortunearia; and ((Distylium. Distyliopsis)-
Sycopsis1 (Fig.1.2).
DISCUSSION
The phylogenetic analysis in this study provides pros
and cons for the hypotheses regarding intergeneric
relationships in the Hamamelidaceae suggested by the three
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. major classification systems (Harms, 1930; Chang, 1979;
Endress, 1989b, c). I will examine the suprageneric
relationships in each of the subfamilies, tribes, and
subtribes of the system of Endress (1989c).
Altingioideae
This subfamily includes three genera: Altingia,
Liquidambar and Semiliquidambar (Chang, 1962, 1979; Bogle,
1968; Wisniewski and Bogle, 1982; Zhang and Lu, 1995).
Morphologically, Semiliquidambar stands as intermediate
between the other two genera of the Altingioideae.
Therefore, it has been considered either as a possible hybrid
(Chang, 1962; Bogle, 1968; Wisniewski and Bogle, 1982), or a
doubtful genus (Endress, 1989c, 1993). Although it has been
widely recognized that Altingioideae are a monophyletic group
(Bogle, 1968; Endress, 1989a, c), it has long been
controversial concerning whether this subfamily should be
raised to familial status. Nakai (1943) proposed that each
of the five subfamilies of Harms (1930) be raised to family
rank, including the Altingioideae. Family recognition of the
Altingioideae was further supported by palynological studies
(Wang, 1992), wood anatomy (Huang 1986), and seed coat
anatomy (Melikian, 1973a, b). However, Pan et al. (1990)
preferred to keep Altingia and Liquidambar in the
Hamamelidaceae based on their study of leaf epidermis in the
family. This view is still reflected in a recent
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. classification system of the Hamamelidaceae (Endress, 1989c).
Takhtajan (1997), however, recognized the family Altingiaceae
and considered it to be closely related to the
Hamamelidaceae.
It should be noticed that most of the characters that
researchers have used to support the separation of the
Altingioideae from the other members of the Hamamelidaceae
are also found in one or more genera in the other
subfamilies. For example, Wang (1992) studied pollen grain
types in the Hamamelidaceae and concluded that the
Altingioideae should be separated from the Hamamelidaceae
proper because of the occurrence of polyporate pollen in the
three genera of the Altingioideae. However, similar types of
pollen grains have been reported in several other genera of
the Hamamelidaceae, including Chunia, Matudaea. Sycopsis and
Distylium (Bogle and Philbrick, 1980). Therefore, using this
one piece of evidence to determine the systematic rank of a
taxon is not convincing, but rather inappropriate, even
though it provides insights for better understanding the
phylogenetic relationships.
The present phylogenetic analysis indicates that the
Altingioideae was supported by five character transformations
(Fig.1.4), these being winged seed, chloroanthoid leaf tooth
type, absence of foliar sclereids, monoecy, and polyporate
pollen grains. This clearly supports the monophyly of the
Altingioideae. However, most of these synapomorphies can
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. also be found in other hamamelidaceous genera.
Recent studies (Hoot and Crane, 1996; Hoot et al., 1997;
Li et al., unpublished) based on both nuclear and chloroplast
gene sequences have also shown that the Hamamelidaceae sensu
lato are monophyletic. Therefore, I prefer to retain these
genera in the Hamamelidaceae as the subfamily Altingioideae.
Exbucklandioideae
This subfamily includes four genera, Exbucklandia
(= Symincrtonia 1. Disanthus, Mvtilaria and Chunia (Endress,
1989c).
Brown (1946) was the first to substitute the name
Exbucklandia for Bucklandia since Bucklandia had been
previously used for a fossil gymnosperm by Presley (See van
Steenis, 1952) and must therefore be rejected as a later
homonym. Later van Steenis (1952) also proposed another
name, Syminqtonia. for Bucklandia, apparently unaware that it
was six years after Brown's Exbucklandia; the latter has
nomenclatural priority and is the valid name for the genus
(Kaul and Kapil, 1974).
Exbucklandia has, in the past, been placed close to
Liquidambar and Altingia (Lindley, 1846, 1853; Gardner and
Champion, 1849; Bentham and Hooker, 1865; Baillon, 1871-
1873), or allied with Rhodoleia (Reinsch, 1889; Niedenzu,
1891). Later, Harms (1930) established the subfamily
Exbucklandioideae, which included only a single genus
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Exbucklandia.
Disanthus was first described by Maximowicz (1866), and
later was placed into the Bucklandioideae (Reinsch, 1889;
Niedenzu, 1891), or treated as a monogeneric subfamily
(Harms, 1930; Pan et al., 1991).
Mytilaria was described by Lecomte (1924) and was placed
by Tong (1930) in the tribe Altingieae. However, Harms
(1930) did not include consideration of this little known
genus. Later, Chang (1948) described Chunia, a relative of
Mytilaria. and erected a new subfamily Mytilarioideae for the
two genera, implying a close relationship between Chunia and
Mytilaria. Bogle (1990, 1991) reported that both Mytilaria
and Chunia have multilacunar nodal anatomy, which is not
found in any other members of the Hamamelidaceae, supporting
the close relationship of the two genera. However, the
evident differences of Chunia and Mytilaria in petiolar
anatomy, floral and pollen morphology led Bogle (1991) to the
suggest that these two genera be placed in the subfamily
Exbucklandioideae, but treated as a distinct tribe, the
Mytilareae.
A cytological study (Pan and Yang, 1994) pointed out the
karyomorphological differences between Mytilaria and the
other members of the Exbucklandioideae and supported the
recognition of the subfamily Mytilarioideae (Karyomorphology
of Chunia was not studied in that study).
Endress (1989c) took a broader view of the subfamily and
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. put all four genera together in his concept of the
Exbucklandioideae. The characters he claimed to define the
subfamily are 5-8 ovules per carpel, simple or tricuspidate
leaves, actinodromous venation, large and persistent
stipules, x=8 for Disanthus and Exbucklandia, and seemingly
bisexual flowers. Most recently, however, Takhtajan (1997)
has reclaimed the subfamily Disanthoideae from the
Exbucklandioideae.
In the present phylogenetic analysis, the four genera
each formed their own clade and were paraphyletic to one
another. This strongly suggests that they do not form a
natural group (Figs.1.2, 1.3). Furthermore, Chunia is very
isolated from the others in the cladogram, as supported by a
group of distinct characteristics, including lack of petals,
winged seeds, and wind pollination.
Rhodoleioideae
Rhodoleia was first described by Hooker (1850). Before
being recognized as belonging to a separate monogeneric
subfamily (Harms, 1930), Rhodoleia was consistently
associated with Exbucklandia and Liquidambar (Reinsch, 1889;
Niedenzu, 1891). Nakai (1943) elevated the Rhodoleioideae to
familial status, which found support from Shaw (1966) and
Wolfe (1973). Nevertheless, this view has not been widely
adopted by others.
Wood anatomy led Tang (1943) to the conclusion that
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rhodoleia is distinct from Exbucklandia and Disanthus. and
should be a separate subfamily. Skvortsova (1960a, b, c)
found Rhodoleia to be comparable to a number of genera of the
Hamamelidoideae in petiole vasculature, leaf venation and
stomatal type, but distinct from Exbucklandia, Mytilaria.
Disanthus, Altinaia. and Liquidambar. Melikian (1971, 1972,
1973a, b, 1975) found the seed coat anatomy of Rhodoleia to
be similar to that of Exbucklandia and Chunia, and concluded
that the three genera could be included in a single
subfamily, Rhodoleioideae, perhaps as tribes Exbucklandieae
and Rhodoleieae. A similar conclusion was reached by Rao
(1974), who studied seed anatomy of the Hamamelidaceae.
In the strict consensus tree generated in this study
(Fig.1.2), Rhodoleia formed its own clade and was sister to
the Hamamelidoideae. The most important characteristic that
allies Rhodoleia and the Hamamelidoideae is the chromosome
number n=12, which is different from any of the other
subfamilies of the Hamamelidaceae, where n=8 or 13 (Goldblatt
and Endress, 1977; Morawetz and Samuel, 1989; Oginuma and
Tobe, 1991; Pan and Yang, 1994). Therefore, this
phylogenetic analysis supports the most widely recognized
treatment, that is, to keep Rhodoleia as a monogeneric
subfamily within the Hamamelidaceae. The Rhodoleioideae are
supported by a combination of several morphological
characteristics, including cyclocytic stomatal apparatus,
tricolporate and scrobiculate pollen grains, asymmetric
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. flowers, long spathulate petals, long stamen filaments, and
presence of staminodial phyllomes.
Hamamelidoideae
The hamamelidaceous taxa had long been divided into a
minimum of two groups (Gardner and Champion, 1849; Lindley,
1853; Bentham and Hooker, 1865; Baillon, 1871-1873), the
largest of which is the Hamamelidoideae, characterized by
uniovulate locules, and the remaining genera, which are
multiovulate. In the phylogeny (Figs.1.2, 1.3), the genera
of the Hamamelidoideae formed a well-defined clade,
suggestive of the monophyly of this taxon. This result
agrees with previous morphological studies (Bogle, 1968;
Bogle and Philbrick, 1980; Endress, 1989b, c; Hufford and
Crane, 1989), and further supported by recent DNA sequence
analyses of the rbcL gene (Qiu et al., unpublished); ITS and
matK (Li et al., 1997a). There are several synapomorphies
defining this natural subfamily, such as the ballistic
mechanism of seed dispersal, several to one ovule per locule,
and the chromosome base number of n=12.
There are about 23 genera in the Hamamelidoideae,
accounting for more than 75% of the genera in the family. A
great morphological diversity can be seen among the genera.
As a result, there has been disagreement about the
intergeneric relationships within the Hamamelidoideae.
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Corylops ide ae
The circumscription of this tribe varies, depending on
the author. In Harms's system (1930), the Corylopsideae
included two genera, Corvlopsis and Fortunearia. which he
believed to be similar in leaf morphology. Schulze-Menz
(1964), however, recognized the similarity of Sinowilsonia
with Fortunearia in both floral structures and leaf
morphology, and thus transferred Sinowilsonia from the
Distylieae into the Corylopsideae, resulting in a three-genus
tribe. Corvlopsis has two unique characteristics that can
not be found anywhere else in the Hamamelidaceae, or even the
subclass Hamamelidae. They are orbicular petals and the
horizontally reflexed styles. It is these peculiarities that
led Endress (1989b, c) to the conclusion that Corylopsis
should constitute a monogeneric tribe.
In the phylogenetic tree produced in this study
(Fig. 1.2), Corylopsis was grouped with Hamamelis and the
clade was supported by four morphological states, including
presence of staminodes, non-decurrent stigma, 2-3 ovules per
carpel, and the semi-palmate venation. The clade of
Corylopsis and Hamamelis, however, did not share immediate
ancestry with the tribe Eustigmateae sensu latp, thus
offering support for a distant relationship between
Corvlopsis and the genera Fortunearia and Sinowilsonia. as
proposed by Endress (1989b, c).
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Eustiamateae
Eustiama was first described by Gardner and Champion
(1849), who noticed the peculiarities of this plant.
However, it was Harms (1930) who recognized Eustiama as
forming a monogeneric tribe because of the its small, bilobed
petals, and its greatly expanded purplish stigmas. The close
relationship between Fortunearia and Sinowilsonia has been
proposed and widely recognized (Schulze-Menz, 1964; Bogle,
1968; Endress, 1969), but the association of Eustiama with
Fortunearia and Sinowilsonia was only recently put forward
(Endress, 1989b,c). The present study offers support for
that proposition (Figs.1.2, 1.3). The unambiguous
morphological characters supporting the relationship include
expanded purplish stigmas (personal observations), and ovary
lenticels (Endress, 1989b).
Distylieae and Fotherai1leae
The tribes Distylieae and Fothergilleae combined include
eight genera that have been treated taxonomically in a number
of ways by various authors (Candolle, 1830; Hallier, 1903;
Harms, 1930; Endress, 1989c).
The Distylieae originally included Distylium and
Svcopsis (Hallier, 1903), to which was added Sinowilsonia
(Harms, 1930), while the Fothergilleae (Candolle, 1830)
contained only Fotherailla. Harms's (1930) Fothergilleae,
however, included Fotherqilla. Parrotia, and Parrotiopsis.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. As described by Harms (1930), the major differences
between the two tribes are: 1) flowers are bisexual in the
Fothergilleae and andromonoecious in Distylieae; 2) leaves
are deciduous in the Fothergilleae, but persistent in
Distylieae.
Lundell (1940) described Matudaea. but it was Walker
(1944) who placed Matudaea in the Distylieae. Walker (1944)
also recognized several Central American species of
Distvlium. Endress (1969), however, separated the Central
American species of Distvlium (D. cruatemalense. D.
hondurense, and D^. sinaloense) from the east Asian species of
Distylium as a new genus, Molinadendron. He also suggested
that Mo1inadendron was closer either to Fotheroilla and
Parrotiopsis. or to Fortunearia and Sinowilsonia. rather than
to the Distylieae. Endress (1970) separated some species of
Svcopsis. e.g., S^. dunni. S. tutcheri, as a new genus,
Distyliopsis. and suggested that Distyliopsis is more closely
related to Distylium than to Sycopsis.
Shaniodendron. the most recently described segregate
genus (Deng et al., 1992a) was originally identified as a
species of Hamamelis (Chang, 1962, 1979). Deng et al.
(1992a, b) placed Shaniodendron in the Fothergilleae.
In the present phylogeny (Fig.1.2), Molinadendron.
Matudaea, Distylium. Distyliopsis. and Svcopsis comprised a
monophyletic clade, while the Fothergilleae did not form a
clade. Svcopsis was sister to the clade of Distvlium-
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Distyliopsis. supporting the close relationship of
Distyliopsis and Distylium. as suggested by Endress (1969).
The tribes Distylieae and Fothergilleae altogether formed a
clade (Fig.1.2), indicating that it is reasonable to unite
the two tribes as the Fothergilleae sensu lato (Endress,
1989b, c). Endress (1989b, c) united the two tribes on the
basis of a fertile, spontaneous intergeneric hybrid between
Svcopsis and Parrotia, which Endress and Anliker (1968)
described as XSycoparrotia.
My observations of floral structures indicate that the
members of the Fothergilleae not only lack petals, as widely
recognized previously (Baillon, 1871; Harms, 1930; Bogle,
1968; Endress, 1993) but also have andromonoecy (except for
Mo1inadendron and Matudaea), as implied by Bogle (1970).
Hamamelideae
The concept of the Hamamelideae has not been changed
much since the establishment of the tribe (Harms, 1930;
Chang, 1979; Endress, 1989b, c), except for the addition of a
few new genera (Smith, 1958; Endress, 1969; Endress et al.,
1985). This tribe is characterized by strap-shaped petals
and 4- or 5-merous flower parts. Endress (1989b, c) did a
revision of the tribe and recognized three subtribes,
Hamamelidinae, Loropetalinae, and Dicoryphinae. The
Hamamelidinae included only one genus, Hamamelis.
characterized by 4-merous flowers and deciduous leaves.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. However, the parsimony analysis placed Hamamelis in a
position close to Corvlopsis (Fig.1.2), which in turn was
sister to the Fothergilleae.
The Loropetalinae contained four genera, including
Tetrathyrium. Mainaaya. Loropetalum, and Embolanthera.
Tetrathyrium and Loropetalum are so closely related that they
might be congeneric species (Endress, 1993). However,
Loropetalum is rather unique in the family by its very small
leaves. Maincraya is endemic to the northwestern part of the
Malay Peninsula (Parak and Penang). It has more than two
whorls of staminodial phyllomes, and the calyx tube falls off
after anthesis, leaving a ring-like scar on the hypanthium.
Embolanthera is distributed in the Philippines, the southern
tip of China, and adjacent Vietnam. Its petals are small and
geniculate with two lateral lobes. Petals and stamens are
fused at the base. The disjunct distributions and
morphological uniqueness of each of the four genera seems to
dictate a 'not-very-closely-related' assessment, even though
they share the character of long, horn-like connective
protrusions in their anther. In Figs.1.2, and 1.3, these
four genera did not form a clade, suggestive of their 'loose'
relationships.
The third subtribe, Dicoryphinae, is composed of five
genera that are exclusively distributed in the Southern
Hemisphere. These genera share a unique anther dehiscence
pattern, i.e., dehiscing by two valves serving four pollen
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sacs. The majority consensus tree (Fig. 1.3) showed that
Dicoryphinae was a monophyletic group, as implied by Endress,
(1989b), even though the strict consensus tree did not
resolve the relationships. However, this suggestion does not
provide support for the attempt to recognize the five genera
as a separate family, as suggested by Zhang and Lu (1995).
In summary, the present phylogenetic analysis recognizes
the monophyly of the subfamilies: Altingioideae,
Exbucklandioideae sensu Harms (1930), Disanthoideae,
Rhodoleioideae, and Hamamelidoideae. Also recognized by the
phylogeny are the following tribes and subtribes: the
Eustigmateae sensu Endress (1989c), Fothergilleae sensu Harms
(1989c), and the Dicoryphinae. Polyphyletic taxa suggested
by the present phylogenetic analysis include
Exbucklandioideae sensu Endress (1989c), Mytilarioideae,
Hamamelideae, and Loropetalinae.
Noticeably, both supporting indices (bootstrap
percentage and decay index) were small for all the clades
except for Licruidambar-Altinqia and Corylopsis-Hamamelis.
Also, the consistency indices were low (CI=0.46, RC=0.27),
and the gl value (skewness test) was only -0.42. These low
indices indicate that there are high homoplasies in this
morphological data set. Therefore, the relationships
obtained in this study should be considered with caution.
That is, some characteristics could be the result of parallel
or reverse evolution. Therefore, other lines of evidence are
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. needed to further evaluate the relationships within the
Hamamelidaceae.
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1.1. Morphological character codings in the Hamamelidaceae. Cereidiphy 1 lum is the outgroup. (Characters and their states are given in the text). a=01, b=02, c=12, d=34, e=03, f=23, g=24, h=14, i=124, j=13, k=39, m=04, ?=missing data.
Character Genus 10 20 30 40 52
Altingia OOOlOdOOlO 1011001222 0000009001 0112221101 040000010110 Chunia 02?00mll00 0d30001002 0100009001 0001212101 0200200101c? Corylopsis 112003002e f021a01001 0100002001 OaOlOOOOOl 0a0c300010cl Dicoryphe OOOblOOOOO 3120101a01 0110005a01 OlOOaOOOll Ia0c30011011 Disanthus 2200000000 0111000001 0100007002 0102100001 330120010020 Distylium 0000020020 3d21101212 0000009001 0all21f111 3102300120cl Distyliopsis 0000020020 3d21101212 0000009001 0all21f111 3112300120C1 Embolanthera 0020030020 3h21101001 0110100101 0102200010 0100310120c? Eustigma 0020020020 3gll011001 0100001001 01012f0001 0000301120c? Exbucklandia 02000m0002 lg30001212 OlOOOOkOOl 0111100001 2301100100C0 Fortunearia 2010030020 3011011011 0100008001 0102200111 0001301120C1 Fothergila 210003002e 3011101211 01a0009021 1002200000 0101300120C1 Hamamelis 210001002e 3a21101101 0100004000 0102100000 2 j01300010cl Liquidambar 2200050110 1011001222 0000009001 0112223101 040000010100 Loropetalum 0020000000 3d21101a01 0100004001 0100100001 0000310120C1 Maingaya 0020000000 3dlll01001 0110004001 0100000000 0 j01310120c? Matudaea 0010000000 £111101202 0000119011 0011222001 0a023001201? Molinadendron 0010100000 3dlll01202 0100009001 0101202111 0a01300120cl Mytilaria 02200mll01 3g30001001 0100006001 0102000000 0201200100C2 Neostrearia 0020000000 4d21101001 0110004000 0102200001 1101310110c? Noahdendron 0020000000 4220101001 0100004000 0100200001 1101310110C1 Ostrearia 0020000100 4d21101001 0100004000 0100100001 1201310120cl Parrotia 2100030020 3221101212 01a0009001 001120blll 3101300120C1 Parrotiopsis 2100030020 3221101211 0101009001 01a2210101 0101300120C1 Rhodoleia 0000000002 42bl001200 1100003001 0002131001 0301100001C1 Shaniodendron 2100030020 3221101212 0100009001 0011200001 3101300120c? Sinowilsonia 201003002e 3011101022 0100008001 0102200011 0111300120C1 Sycopsis OOOOObOObO 4d21101212 0100009001 0all20f111 0al2300120cl Tetrathyrium 0020020021 3hlll01001 0100004001 0100100001 0j00310120c? Trichocladus OOOblOOOOO 3d20101a01 01a0004000 0100200001 1301300120cl Cere idiphy Hu m 2c020c0003 0011001232 0000009001 0011210111 330270010103
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.1.1. Stomatal apparatus in the sampled species of the Hamamelidaceae. All lower epidermis and 200X unless otherwise indicated, a. Dicoryphe stipulacea. Randrianasolo 543; b. Dicoryphe stipulacea (upper epidermis); c. Embolanthera spicata. Soefarto and Fernando 7345 (GH); d. Mainqava malayana (no collection data); e. Matudaea hirsuta. Bogle 848; f. Matudaea hirsuta (upper epidermis); g. Molinadendron sinaloense. Bogle 864; h. Neostrearia fleckeri. Endress s.n. (400X) i. Noahdendron nicholasii. Endress s.n.: j. Ostrearia australiana. Irvine 2081; k. Parrotia persica. Bogle 884; 1. Trichocladus crinitus. Bogle, 1545.
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a
j■
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.1.2. Strict consensus of the 12 shortest trees of 219 steps based on morphological data of the Hamamelidaceae. Branches have a decay value of one and a bootstrap percentage of less than 50% unless otherwise indicated. Numbers below and above the branches are decay indices and bootstrap percentages. CI=0.45, RC=0.27.
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76 Altingia Liquidambar Chunia 66 Corylopsis Hamamelis 67 Distylium Distyliopsis Matudaea Molinadendron -tE Sycopsis Parrotia Shaniodendron Fothergila Parrotiopsis Eustigma Fortunearia £ Sinowilsonia Embolanthera Tetrathyriurn Loropetalum Dicoryphe Trichocladus Maingaya Neostrearia Noahdendron Ostrearia Rhodoleia Disanthus Mytilaria Exbucklandia Cereidiphy Hu m (Outgroup)
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 1.3. 50% majority consensus of the 12 shortest trees based on morphological data of the Hamamelidaceae. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae.
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. -| Altingia -i A^oi^ae Liquidambar -I “ “ Chunia — EXBoideae r— Corylopsis — cOReae— Hamamelis — HAMinae (HAMeae) Distylium j— c Distyliopsis Matudaea 0) Molinadendron (0 Cercidiphyllum (Outgroup)
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 1.4. 50% majority consensus tree of 12 shortest trees based on morphological data, showing the distribution of numbers of characters that change unambiguously on branch.
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i Altingia
i Liquidambar i Chunia
i Corylopsis
Hamamelis D isty liu m Distyliopsis Matudaea Molinadendron Sycopsis P a rro tia Shaniodendron Fothergilla Parrotiopsis Eustigma Fortunearia Sinowilsonia
Embolanthera Tetrathyrium Loropetalum
Maingaya Dicoryphe Trichocladus
t Chars, that change Neostrearia Unambiguously on branch Noahdendron CZ3 o Ostrearia UZ3 1 Rhodoleia ES3 2 Disanthus 3 4 M y tila ria 5 Exbucklandia 6 Cercidiphyllum 7 8 9 or more
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX 1.1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix.1.1. Herbarium and chemically fixed specimens observed for this study. GH, Harvard University Herbaria; MO, Missouri Botanical Garden; HK, Hong Kong Herbarium; US, U.S. National Herbarium at Smithsonian Institution.
HERBARIUM SPECIMENS
Corvlopsis aotoana Makino
JAPAN. Honshu: Mt. Aterayama, Murata and Iwatsuki 908
(MO).
C . glaucescens Hand.-Mazz.
CHINA. Tibet: Rock 11226 (US).
£_«_ himalavaoa Griff.
INDIA. Khasia and Jynteah Hills, Gallatly 627 (US).
Sirhoi: Kinqdon-Ward 17205 (MO).
£_*_ multiflora Hance
CHINA. Guangxi: Yuan Tung Shan, Chinq 5722. 6038 (US).
Guizhou: Fanjing Shan, Jiangkou Xian, Sino-American Botanical
Expedition 113 (MO).
C . pauciflora S.et Z.
JAPAN. Honshu: Sakiya 78 (MO).
SLi. Platvoetala R.et W.
CHINA. Sichuan: Dujiangyan, Boufford and Bartholomew
24584 (MO).
C . sinensis Hemsl.
CHINA. Anhui: Jinzhai, Yao 8899 (MO). Hunan: Xinning, Li
et al. 1823 (MO). Jiangxi: Lu Shan, Steward 2473 (MO); Chiao
18588 (US).
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
C . soicata S.et Z.
JAPAN. Honshu: Sakiya 81 (MO). Shikoku: Osaka-toge,
Nangoku-shi, Kochi-ken, June 20, 1973, Ohashi s.n.(US
willmottiae R.et W.
CHINA. Sichuan: Guan Xian, Wane 870075 (MO).
wilsonii Hemsl.
CHINA. Guangdong: Le Chang Xiang, Tsanq 20757 (MO). C.
vunnanensis Diels. CHINA. Yunnan: Chienchuan, Mekong divide,
Forrest 23588 (MO); Tsangshan Range, west of Talifu, Rock
6376 (US).
P\fftVij.9Pgiff A-flvr-i.f9.lAa (Hemsl.) Endress
CHINA. Yunnan: Lunan Xian, Sino-American Botanical
Expedition 1584 (GH ).
PiStyiiMffi bnxifolium (Hance) Merr.
CHINA. Hubei: Western Hupeh, Wilson 115 (GH); A. Henry,
3626 (GH). Zhejiang: Barchet 170 (US). IK gracile Nakai.
CHINA. Taiwan: Taipeh, Botanical Garden (cultivated), Keng
K1047 (GH); Wilson 11107 (US).
L . Nakai
JAPAN. Tokyo: Ogasawara Isis, Fuiita and Shimizu 72
(MO).
raggmggwm s.et z.
CHINA. Taiwan: Botel Tobago, Taitung county, Chang 2793
(MO); Tienchih, Huang 9378 (MO).
JAPAN. Hondo: Waterfall of Ozaka, Pref. Nara, Faurie
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
12077 (MO). Honshu: Pref. Yamaguchi, Kasayama, Hagi-shi,
Murata 22143 (MO). Kagoshima: Kyushu, Mitsuta. Doei and
Naaamasu 53 (GH); Wilson 6039 (MO). Komaba: Ohashi and
Tateishi 10949 (GH). Kosugidani: Arakawa Dam, Yahara.
Murakami, and Watano 7633 (GH). Linkiu Islands: Wilson 8070
(GH). Okinawa: Isl. Iriomote, along Yutsun river, Taketomi-
cho, Okada. Takahashi and Naaamasu 358 (GH).
Embolanthera spicata Merr.
PHILIPPINES. Palawan: Panacan, Aborlan, Dulit 3725 (GH).
Eustiama oblongifolium Gardn.et Champ.
CHINA. Guangdong: Sin-fung District, Taam 389 (MO);
Luofu Xiang, Chun 40455 (MO). Guangxi: Yao Shan, Wang 39379
(GH); Shang-sze district, Tsanq 22213 (GH). Hainan: Chim Fung
Mt. Kan-en District, Lau 5678 (GH); Wright 35219 (US). Hong
Kong: Peel Rise, Choi 403 (HK); Mt. Gough, Victoria, Hu and
But 21997 (GH); Choi and Hu 12739. 13251 (GH); Wright 186
(US ). Taiwan: Wilson 9959 (GH ).
§_«_ balansae Oliv.
CHINA. Guangdong: Ding-hu Shan, Ting 1569 (GH). Guangxi:
Steward and Cheo 696 (GH).
SsfrttsJslanflia pypvLagfl (R.Br.) R.w.Br.
CHINA. Hong Kong: Sunset Peak, Li 73 (GH). Yunnan: Henry
11068 (GH); Hsien and Wang 72631 (GH); Between Kambaiti and
Tengyueh, Rock 7574 (GH).
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
Fortunearia sinensis R.et W .
CHINA. Anhui: Huangshan, Chow 101 (GH, MO); IP 6339
(US). Jiangsu: I-hinq and Fana 7902 (MO); Bau Hwa Shan, Tsu
1680 (US). Zhejiang: Tihtaishan, Chinq 1550 (US).
P.9tllgr.q341a qardenii Murr.
U.S.A. North Carolina: Scotland Co., just south of the
bridge over Lumber Creek (Hoke-Scotland Co. Line) on US 15-
501, Weaver Jr. 350 (GH); Onslow county, May 11, 1948, Boyce
and Moreland s.n. (GH). South Carolina: Horry county, Griscorn
16547 (GH).
F . maior (Sims) Loddiges
U.S.A. North Carolina: Burke Co., Bell 6564 (GH); Rable
Rock Mountain, Stone and Family 1889 (GH); Burke Co. Shortoff
Mountain, Wilbur 7012 (GH); Polk County, Weaver. Jr. 1333
(GH).
Liquidambar formosana Hance
CHINA. Guangdong: Manchi Shan (Wan Zhi Shan), Jen-hwa
District, Tsanq 26304 (GH); Loh Hoh Tsuen, Ling Yun Ksien,
Steward and Cheo 442 (GH). Guangxi: Liang 69991 (GH).
Guizhou: Ta-ho-Yen, Fan Ching Shan, Steward. Chiao and Cheo
716 (GH). Hong Kong: Hu and But 20064 (GH). Hubei: western
Appendix 1.1. Continued
Hupeh, Wilson 513. 5218 (GH).
JAPAN. Kyoto: Yase, Muroi 4987 (GH).
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
kj_ styraciflua L.
U.S.A. Alabama: Tuscaloosa County, Ingram and Findley 66
(GH). Arkansas: Drew county, Demaree 16661 (GH); Saline
County, Demaree 8517 (GH). Delaware: Newcastle county, Long
32326 (GH).
frgrgpetalEfl chinense (R.Br) Oliv.
CHINA. Anhwei: Ching 3284 (GH). Fukien: Chung 1013. 8548
(GH). Guangxi: Ling-chuan District (Ling Chuan County), Tsang
27845 (GH). Hong Kong: Hu and But 20118 (GH). Jiangxi:
Lushan, Cheng 74 (GH).
INDIA: Khasia and Jaintia Hills, Ruse 47 (GH).
JAPAN. Honshu: Taoda 3680 (GH).
Maingava malavana Oliv.
MALAYSIA. Kepong: Frim Arboretum, van Balgooy 5755 (GH).
Penang: Penang Hill, Viaduct Rd., Lesmv 33667 (GH).
Matudaea triuervia Lundell
COSTA RICA. Hammel 17420 (GH).
HONDURAS. Esperanza and Intibuca: Standley 25512 (GH).
MEXICO. Chiapas: Matuda 3984 (GH). Jalisco: south of
Talpa de Allende, McVauoh 23317A (GH).
Molinadendron guatemalense (Harms) Endress
GUATEMALA. Alta Verapaz: Williams. Terua. Williams, and
Molina 40373 (GH); von Turckheim 187 (US).
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
M . sinaloense (Standi.& Gentry) Endress
MEXICO. Sinaloa: Sierra Surotato, Gentry 7254. (Isotype)
(US).
M. hondurense (Standi.) Endress
HONDURAS. La Mision: Allen 6188 (GH); Molina 7226 (GH).
Mvtilaria laosensis H.Lee
CHINA. Guangxi: Shap Man Taai Shan, Shang-sze District,
Tsana 2498 (GH).
Neostrearia fleckeri L.S.Smith
AUSTRALIA. Queensland: Mosspanan River Gorge, Brass 2140
(GH).
Ostrearia australiana Baill.
AUSTRALIA. Queensland: Upper Parrot Creek, Annan River,
Cape York Peninsula, Brass 20266 (GH); Cook District, Smith
11209 (GH); Sanderson 43 (GH).
PgHT9ti9Pgjg iacquemontiana Schneider
INDIA. Kashmir: Bhadraloah, Rao 9064 (GH). Jhelum
Valley: Stewart 12088 (GH). Punjab: Kabl, Chamba, Stewart
2331 (GH).
PAKISTAN. Swat District: Showa Khuar Kalam, Shah and
Javed 231 (GH).
Sinowilsonia henrvi Hemsl.
CHINA. Hubei: Shennongjia Forest District, Sino-American
Botanical Expedition 1483 (GH); Wilson 584 (GH, US); Henry
6559 (Isotype) (US).
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
Sv c o p s x s sinensis Oliv.
CHINA. Taiwan: Hwalien, Kao 9842 (MO).
CHEMICALLY FIXED SPECIMENS
(Stored in the Laboratory of Plant Morphology and
Anatomy at the University of New Hampshire).
Alttflqia excelsa Nor.
CHINA. Hong Kong: Hong Kong Botanical Garden, Bogle 583.
MALAYSIA. Cameron Highlands: Mentigi F. R., Bogle 313.
C M l U a friLS.fr IflAfliaAdSfi Chang
CHINA. Hainan, Liang 64169.
Corvlopsis veitchiana Bean
U.S.A. Massachusetts: Arnold Arboretum (cultivated),
Li 02.
Picorvphe so.
Madagascar: Dorr 3982.
Disanthus cercidifolius Maxim.
JAPAN. Tokyo: Tokyo University Botanical Garden,
Bogle 1268.
U.S.A. Pennsylvania: Morris Arboretum (cultivated),
Bogle 1474. Washington: University of Washington Arboretum
(cultivated), Bogle 1343.
Pistvlium racemosun S.et Z.
U.S.A. New Hampshire: University of New Hampshire
greenhouse (cultivated), Bogle 1459. Washington: University
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
of Washington Arboretum (cultivated), Boole 1361. 1385.
E.ygt-i.qffifl oblonqifolium Gardn.et Champ.
CHINA. Hong Kong: Aberdeen Rd., ALB 589. Taiwan: Chung
s.n., serial collections every three days for two years,
1996-1997.
Exbucklaodia populnea (R.Br.) R.W.Br.
MALAYSIA. Cameron Highlands: Tanah Rata Vill., Boole
314.
Fortunearia sinensis R.et W.
U.S.A. New Hampshire: University of New Hampshire
greenhouse (cultivated), Boole 1458.
Hamamelis macrophvlla Pursh
U.S.A. Washington: University of Washington Arboretum,
Boole 749.
H^_ mollis Oliv.
U.S.A. New Hampshire: University of New Hampshire campus
(cultivated), Boole 1352.
kjttvteflamfear. gtYracifiua l . U.S.A. Massachusetts: Arnold Arboretum (cultivated), M.
Wisniewski, Oct. 5, and 25, 1979. Pennsylvania: Service St.,
Philadelphia, Boole 605. 1267.
Loropetalnn chinense (R.Br.) Oliv.
JAPAN. Tokyo University Botanical Garden, Boole 719.
U.S.A. New Hampshire: University of New Hampshire
greenhouse (cultivated), ALB 1294.
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
qai-Bgaza malavana Oliv.
MALAYSIA. Kepong: Boole 1314. Kepono Research Institute
25856.
Matudaea ££..
MEXICO. Behucos-Nanchititea: Boole 848.
Mvtilarira laosensis Lee.
CHINA. Guangxi, Liano 6980.
Qgtrear4-a australiana Baill.
AUSTRALIA. Queens land: Gray 1036.
Parrotia persica C.A.Mey
U.S.A. Massachusetts: Harvard University campus
(cultivated), Boole 1488. New Hampshire: University of New
Hampshire greenhouse. Jan. 14, 17 1994, Li s.n.
Parrotiopsis iacouemontana Schneider
U.S.A. Massachusetts: Harvard University campus
(cultivated), Boole 1482. 1498. Washington: University of
Washington Arboretum (cultivated), Boole 1327.
Rhodoleia championii Hook. f.
MALAYSIA. Cameron Highlands: Tanah Rata Vill., ALB 316;
Kuala Lumper: Klang Gates, Boole 276.
Sinowilsonia henrvi Hemsl.
U.S.A. New York: Oyster Bay Planting Field Arboretum,
Long Island (cultivated), Boole 1518.
ffYSAPgiS sinensis Oliv.
U.S.A. Washington: University of Washington Arboretum
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix 1.1. Continued
(cultivated), Bogle 1315. 1367.
Tetrathvriuro subcordatum Benth.
CHINA. Hong Kong: Bowen Rd., Bogle 586.
Trichocladus crinitus Pers.
U.S.A. Pennsylvania: Longwood Gardens (cultivated),
Bogle 1513. 1545.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER II
PHYLOGENETIC RELATIONSHIPS OF THE HAMAMELIDACEAE
INFERRED FROM NUCLEOTIDE SEQUENCES OF THE PLASTID GENE MATK
INTRODUCTION
The Hamamelidaceae, a family of 30-31 widely recognized
genera and about 140 species, are distributed in the
tropical, subtropical and temperate areas in both the Old and
New Worlds (Endress, 1993; Zhang and Lu, 1995). The uniform
characters in the Hamamelidaceae include woody habit,
2-carpelled pistils and multicellular stigmatic papillae
(Endress, 1989a). Other characters are highly diverse. For
example, leaves are stipulate, persistent or deciduous,
simple and pinnately veined or palmately lobed and veined.
Most species are bisexual, but some are andromonoecious (both
staminate and bisexual flowers found in an individual), still
others are monoecious (staminate and pistillate flowers
develop on an individual). Flowers are complete and 5-merous
in most genera, 4-merous in several genera and variable in
others; a few genera have an incomplete perianth or are
naked.
Well-known examples of the Hamamelidaceae are
Witch-hazel (Hamamelis virqinianal, which yields the widely
used astringent and soothing lotion for cuts and bruises;
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Winter-hazel (Corylopsis), which has long been cultivated as
an ornamental; and Sweetgum (Licruidambar). which produces a
fragrant gum (storax) used in perfumery, as an inhalant, and
as a fumigant in treatment of skin diseases.
Both morphological (Hufford, 1992) and molecular (Hoot
et al., 1997; Hoot and Crane, 1996) studies have demonstrated
that the Hamamelidaceae are monophyletic. As for the
systematic position of the Hamamelidaceae, most of the
traditional systems place this family in the "Lower"
Hamamelidae that include Cercidiphyllaceae, Tetracentraceae,
Trochodendraceae, Daphniphyllaceae, Platanaceae,
Myrothamnaceae and Eupteleaceae (Takhtajan, 1980; Cronquist,
1988). Endress (1977) suggested this family to be a
connecting taxon between the "Lower" and the "Higher"
Hamamelidae (e.g. Betulaceae, Fagaceae, and Juglandaceae).
Recently, some members of the Hamamelidaceae have been
considered as linking taxa between the "Lower" hamamelids and
some basal elements of rosids and asterids (Hufford, 1992;
Chase et al., 1993; Endress, 1993; Morgan and Soltis, 1993).
Apparently, a more comprehensive study (more taxa and more
sources of data) is needed to further assess the systematic
position of the Hamamelidaceae.
The phylogeny of the Hamamelidaceae remains unresolved
at both subfamilial and tribal levels. At the end of the
19th century, Reinsch (1889) proposed a three-subfamily
scheme of the Hamamelidaceae [Altingioideae, Bucklandioideae
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (= Exbucklandioideae) and Hamamelidoideae]. Niedenzu (1891),
however, suggested a two-subfamily system [Bucklandioideae
(= Exbucklandioideae) and Hamamelidoideae]. In this century,
the first comprehensive classification system was proposed by
Harms (1930) (Fig.i.4). He recognized five subfamilies
[Disanthoideae, Hamamelidoideae, Rhodoleioideae,
Bucklandioideae (= Exbucklandioideae), and
Liquidambaroideae]. Chang (1973, 1979) reviewed the
hamamelidaceous flora of China, recognizing the five
subfamilies of Harms but also erecting a new subfamily for
Mytilaria and his own new genus Chunia (Fig.i.5). Most
recently, Endress (1989c) provided a suprageneric scheme for
the Hamamelidaceae (Fig.i.6). In this system he combined the
three subfamilies, Disanthoideae, Mytilarioideae and
Exbucklandioideae, thus recognizing four subfamilies in the
Hamamelidaceae (Altingioideae, Rhodoleioideae,
Exbucklandioideae, and Hamamelidoideae).
The controversy concerning tribal or subtribal
delimitations focuses on the Hamamelidoideae since the other
subfamilies have only one or a few genera, and further
subdivision does not seem to be justified. Harms (1930)
divided the Hamamelidoideae into five tribes, including
Cory lops ideae, Distylieae, Fothergilleae, Hamamelideae, and
Eustigmateae. Endress (1989c) revised Harms's (1930) system
and recognized four tribes in the Hamamelidoideae by
including the Distylieae in the Fothergilleae. In that
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. system, he also made some generic rearrangements.
Nucleotide sequences of chloroplast genes have proven to
be very promising for resolving phylogenetic relationships in
angiosperms (Olmstead and Palmer, 1994). However, genes may
have different substitution rates, suggesting that to resolve
phylogeny at different levels, gene(s) of appropriate
evolutionary substitution rates should be employed. The most
widely used and so-called "gene of choice" the rbcL gene that
encodes the large subunit of ribulose-l,5-bisphosphate
carboxylase/oxygenase has been found to be informative mostly
at the familial or higher levels (Chase et al., 1993).
Nevertheless, in the rbcL phylogenies (Chase et al., 1993),
where the Hamamelidaceae were poorly represented, the clade
containing both hamamelidaceous genera and others was not
resolved. Compared to the rbcL, the matK gene, which encodes
a chloroplast maturase and is located in the trnK intron, has
a relatively higher substitution rate (Olmstead and Palmer,
1994), indicating that it may be more appropriate for
systematic studies at lower levels. This proposition has
been confirmed by several cladistic analyses (Johnson and
Soltis, 1994; Steele and Vilgalys, 1994), including a
phylogenetic study of the "Higher" Hamamelidae (Manos and
Steele, 1997).
Therefore, in this study, I chose to use DNA sequences
of the matK gene to investigate phylogenetic relationships
within the Hamamelidaceae and to evaluate the existing
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. classification systems of the family.
MATERIALS AND METHODS
Plant Materials
Thirty hamamelidaceous species from 27 genera were
sampled, representing all of the subfamilies, tribes and
subtribes previously proposed by Harms (1930), Chang (1979),
and Endress (1989c). Sources, vouchers, and sequence GenBank
accession numbers are listed in Table 2.1. The aligned matK
sequences are given in Appendix 2.1.
Molecular Techniques
Total genomic DNAs were extracted from fresh or silica
gel dried leaves using the standard DNA extraction procedures
of Doyle and Doyle (1987). Polymerase chain reaction (PCR™)
amplification was conducted using forward primer matKFl and
reverse primer matKRl kindly provided by Tao Sang (Michigan
State University, East Lansing, MI). The amplification
reaction included 50-100 ng of genomic DNA, 3-4 units of Tag
(Promega, Madison, WI), 3-4 units of Tag Extender
(Stratagene, La Jolla, CA), IX Tag Extender buffer, 2.5 mM
dNTP, and 20 gM primers. Amplifications were conducted in
thin-walled tubes using a MJ thermocycler (Watertown, MA)
following the PCR program of Johnson and Soltis (1994).
The PCR amplified products were purified on 0.8% low
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. melting point agarose gels in IX TBE buffer (pH8.0). The
matK bands identified by lambda HindiII DNA size markers were
excised and agarase-digested for 30 minutes. The purified
PCR products were then used as templates for direct
double-stranded sequencing using dyedideoxynucleotide
terminator cycle sequencing on an Automated Sequencer 373A
(Applied Biosystems, Foster City, CA). The procedures were
carried out according to the manufacturer's instructions at
the University of New Hampshire Sequencing Facility Center.
Sequencing primers, FI, F2, F3 (forward) and Rl, R2 and R3
(reverse) were generously provided by Tao Sang. In addition,
I designed forward primers matKF4. matKF4-2 (for Maingaya),
matKF5, matKF6 (for Liquidambar). and reverse primer matKR2-2
(for Liquidambar) in order to get complete matK gene
sequences. Figure 2.1 shows the relative positions of these
primers; base composition of the primers is listed in Table
2 .2 .
Sequence Analysis
Sequence chromatograms were analyzed using the SEQED
program (Applied Biosystems, Foster City, CA), and the
overlap option was employed to assure correct base-calling.
The analyzed sequences were then exported as EDITSEQ files.
Base composition, translated amino acid sequences, and codon
usage were obtained using the EDITSEQ program. The sequences
were aligned using the MEGALIGN program. Both EDITSEQ and
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MEGALIGN are programs in the DNASTAR software package
(version 3.72, Madison, WI).
Phylogenetic Analyses
Parsimony analyses were conducted using the test version
4.0d55 of PAUP* written by David L. Swofford (Smithsonian
Institution). Given the limit of computer memory and the
sample size, heuristic searches were performed with the other
options of TBR swapping, multipars on, and steepest descent
off. I have tried a 1:1.3 weighting scheme for transition
(Ts) and transversion (Tv) according to the estimated
frequencies of character changes using the MacClade program
(Maddison and Maddison, 1992), but did not find any effect on
the topology of phylogenetic trees. Therefore, I report only
the results of the analyses using unweighted characters and
unordered transformations. Indels in the data matrix were
coded as missing data. However, some indels are
phylogenetically informative and will be mentioned in the
Discussion section of this Chapter.
The trees were rooted by outgroup comparison using
Sullivantia sullivantii (GenBank accession #20130) and
Saxifraqa inteqrifolia (GenBank accession #20131)
(Saxifragaceae) as outgroups. These taxa were two of only a
few taxa whose complete matK sequences were available in the
GenBank. In addition, the possible relationship of the
Hamamelidaceae and Saxifragaceae has been previously proposed
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Hufford, 1992; Chase et al., 1993; Endress, 1993; Morgan and
Soltis, 1993).
Bootstrap analysis (Felsenstein, 1985) of 100 replicates
and the Constraint decay analysis (Bremer, 1988; Morgan,
1997) were performed to obtain the indices of relative
support for individual clades.
Skewness of tree length distributions has been proposed
to be a level indicator of phylogenetic information contained
in a data matrix (Kallersjo, 1982; Huelsenbeck, 1991). The
skewness test was implemented using the Random Tree option of
PAUP* and 10,000 random trees were examined. Another test of
phylogenetic information of a data set is the randomization
test, which produces the permutation tail probability (PTP)
statistics. Data sets with values of PTP<0.01 are considered
to be considerably different from randomized data (Faith and
Cranston, 1991; Plunkett et al., 1997). The permutation test
was performed using the Permutation option of PAUP* with 100
replicates and heuristic searches.
The aligned sequences were imported into the MacClade
computer program, version 3.03 (Maddison and Maddison, 1992)
to estimate transition and transversion ratio, and character
changes in the first, second and third codon using one of the
most parsimonious tree, and to compare the competing
hypotheses concerning generic relationships of the
Hamamelidaceae. The MacClade program was also used to
estimate the number of unambiguous changes along branches.
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RESULTS
Sequence Characteristics
The sequence length of the matK gene in the
Hamamelidaceae ranged from 1506 bases in Pisan thus to 1521
bases in Molinadendron. In most of the other genera, this
gene was 1515 bases long, while in Altinaia and Liquidambar
(Altingioideae), Exbucklandia. Rhodoleia. and Sinowilsonia,
the matK gene was 1512 bases in length. The number of amino
acids ranged from 502 to 507 (Appendix 2.2). In terms of
nucleotide composition, the matK gene was AT rich in the
Hamamelidaceae; the GC contents were ca. 34% (Table 2.3).
Pairwise sequence divergence basically corresponded to
taxonomic levels: 0.5-1% within Corvlopsis. Hamamelis, and
Liquidambar: 2-3% at generic levels, and 4-6% at the
subfamilial level (Table 2.3). Alignment of the sequences in
the Hamamelidaceae required five deletions, two of which were
three bases long, while the other three were two, six, and
seven bases in length (See Appendix 2.1).
The ratio of transitional and transversional changes
obtained using the MacClade program was about 1.3 for the
matK gene sequences in the Hamamelidaceae, and the relative
percentages of character state changes were 32%, 25%, and 43%
for the first, second, and third codon positions
respectively. Codon usage of matK gene in the Hamamelidaceae
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was far from balanced among the codons encoding the same
amino acid (Table 2.4).
Phylogenetic Trees
The parsimony analysis resulted in six equally shortest
trees with a length of 582 steps and a consistency index (Cl)
of 0.85 (strict consensus tree shown in Fig.2.2). Altinqia
and Liquidambar formed the basal clade supported by a
bootstrap percentage (BP) of 100% and a decay index (DI) of
eight. The next clade contained Exbucklandia and Rhodoleia
and the supporting indices were 100% BP and DI of seven.
Mvtilaria formed its own clade followed by the clade of
Disanthus« which was in turn sister to the trichotomous clade
of the Hamamelidoideae. The first clade in the
Hamamelidoideae included Corvloosis and the branch of
Maingava. Loropetalum. and Matudaea, while the second one
contained two groups: Dicoryphinae Endress and the
Eustigmateae sensu Endress (1989c) (incl. Molinadendron).
The third clade was composed of the Fothergilleae sensu
Endress (1989c), but excluding the New World Matudaea and
Molinadendron. and including Hamamelis).
DISCUSSION
MatK Sequence Variation
The matK gene is approximately 1500 bases long (Johnson
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and Soltis, 1995). So far, only a portion of the matK gene
has been sequenced for a variety of plants (Johnson and
Soltis, 1994, 1995; Steele and Vilgalys, 1994; Soltis et al.,
1996b; Hilu and Liang, 1997; Manos and Steele, 1997; Plunkett
et al., 1997), thus it is premature to give an accurate range
of the length variation of the matK gene in the angiosperms.
Nevertheless, on the basis of the available matK sequence in
the Hamamelidaceae (this study), Saxifragaceae (Johnson and
Soltis, 1994), Poaceae (Hilu and Liang, 1997), Nymphaeaceae
(Padgett, 1997), and Brassicaceae (GenBank X04826), the matK
gene in angiosperms is approximately 1500-1600 bases in
length.
Plunkett et al. (1997) studied the relationship between
Apiaceae and Araliaceae using DNA sequences of both the rbcL
gene and two thirds of the matK gene. They found that the
ratio of transition (Ts) to transversion (Tv) was 1.13 for
the matK gene, and codon position substitution ratios were
1.2:1:1.68 for the first, second and third codon
respectively. The matK gene in the Hamamelidaceae shows
similar ratios both for Ts/Tv (1.3:1), and for codon
substitutions (1.3:1:1.74). It is noticeable that more
variation occurs at the third codon position for Apiales
(0.43) (Plunkett et al., 1997), for Polemoniaceae (0.38-0.5)
(Steele and Vilgalys, 1994), and for the Hamamelidaceae
(0.43) (this study).
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Phvloaenetic Usefulness of matK Gene
The matK gene has been increasingly applied in molecular
systematics of plants because of its several characteristics.
First, there is a certain degree of evolutionary constraint
imposed by the function of the maturase it encodes (Neuhaus
and Link, 1987; Wolfe et al., 1992). This property
facilitates sequence alignment. In this study, the
divergence of the matK gene between the hamamelidaceous
genera and the outgroup genera from the Saxifragaceae is 8-
11%, and the sequences are unambiguously alignable (See
Appendix 2.1). This suggests that the matK DNA sequences are
likely to be useful in resolving deep relationships at family
or even order levels. This agrees with recent studies
(Soltis et al., 1996b; Plunkett et al., 1997). Second, the
matK gene is more variable than the rbcL gene that is
informative in resolving relationships at the family or
higher taxonomic levels (Chase et al., 1993). Thus, the matK
gene may be more informative at lower levels. Sequence
divergence within a genus is rather small, ca. 1%, providing
a limited number of informative sites. However, the sites
have been proved very useful in resolving interspecific
relationships in some genera such as T.iguidambar (Li et al.,
1997b). In contrast, informative sites are about 8.6% in the
Hamamelidaceae, giving rise to about 130 informative
characters. As a result, phylogenetic analyses using the
matK data set resolved most of the intergeneric relationships
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in the Hamamelidaceae. In addition, a higher percentage of
phylogenetically-informative sites has been reported in other
families (Plunkett et al., 1997). Therefore, as pointed out
by Hilu and Liang (1997), the matK gene is informative in
reconstructing phylogenetic relationships at different
taxonomic levels and is a promising source of data in
studying the molecular systematics of plants.
Both gi (tree length skewness) statistics and the PTP
(permutation tail probability) test have shown that matK data
matrices contain a large amount of phylogenetic structure in
the Apiales (gi=-0.51, P<0.01, Plunkett et al., 1997) and the
Hamamelidaceae (gi=-1.89, P<0.01, this study), thus
quantitatively indicating the usefulness of matK gene in
these phylogenetic studies.
Phyloaenv of the Altinaioideae
The Altingioideae includes three genera, Altinaia.
Liquidambar. and Semilicruidambar (Chang, 1973, 1979; Bogle,
1986; Endress, 1989c). Semiliquidambar is morphologically
intermediate, and thus has been considered to be a hybrid
between the other two genera (Chang, 1962; Bogle, 1968,
1986). Altinqia is very similar to the species of
Liquidambar in many morphological structures (Harms, 1930;
Tong, 1930; Bogle, 1968, 1986; Melikian, 1973a, b; Rao, 1974;
Chang, 1979; Bogle and Philbrick, 1980; Wang, 1992). The
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. phylogenetic analysis reported here supports a close
relationship between Altinoia and Liquidambar (Fig.2.2 ). As
can be seen in Figure 2.3, 35 unambiguous nucleotide changes
support this clade. Also, the species in the Altingioideae
share a three-base deletion and several synapomorphic amino
acid substitutions (Appendix 2.2). Interestingly, the
sampled species of Altinaia is located within the clade of
the two Liquidambar species, indicating that either one of
the two genera is derived from the other, or both genera are
not monophyletic. Unfortunately, it is impossible to
determine which of the two hypotheses is correct because this
study sampled only one Altinqia species. Therefore, a more
inclusive study of the two genera is needed using sequences
of matK gene or a piece of DNA of higher variation.
The long-standing controversy regarding the
Altingioideae has been whether this group should be
considered as a separate family (Endress, 1989c; Pan et al.,
1991; Wang, 1992; Takhtajan, 1997). I prefer to keep this
clade in the Hamamelidaceae because of the monophyly of the
Hamamelidaceae supported by several recent molecular studies
(Hibsch-Jetter and Soltis, 1996; Hoot and Crane, 1996; Hoot
et al., 1997; Li et al., unpublished). Furthermore, many
morphological characters that have been used to support the
separation of the Altingioideae as a family occur in both the
Altingioideae and some other members of the Hamamelidaceae.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Polyphyly of the Exbucklandioideae
This subfamily, as recently prescribed by Endress
(1989c), consists of four genera, Disanthus, Exbucklandia.
Mvtilaria. and Chunia, the latter was not available for this
analysis. However, each of the other three genera has been
treated as representing a separate subfamily based on
morphological characters, which suggests their distant
relationships (Harms, 1930; Chang, 1948, 1979). The fact
that these genera share several morphological
characteristics, such as palmately veined leaves; large
persistent stipules; and 5-8 ovules in each locule, led
Endress (1989c) to propose a unification of the genera.
Takhtajan (1997), however, moved Disanthus out of the
Exbucklandioideae and treated the genus as representing the
subfamily Disanthoideae.
The matK phylogeny suggests a polyphyletic relationship
among these genera (Fig.2.2). Disanthus and Mytilaria each
forms a monogeneric clade with 35 and 25 unambiguous
character state changes respectively (Fig.2.3), while
Exbucklandia and Rhodoleia are nested with strong support
from both BP and DI (Fig.2.2). Forcing Mytilaria. Disanthus,
and Exbucklandia into a monophyletic clade required 17 more
steps than the most parsimonious trees. Disanthus is the
most closely related genus to the Hamamelidoideae, which is
concordant with previous studies (Hufford and Crane, 1989;
Pan et al., 1991).
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rhodoleioideae
Rhodoleia has long been treated as a monotypic subfamily
(Harms, 1930; Endress, 1989c; Takhtajan, 1997). However, in
the matK-based phylogeny, Rhodoleia is allied with
Exbucklandia in a well-supported clade, suggesting a close
relationship of the two genera. Ten unambiguous character
changes occur along the branch supporting the relationship.
This cladistic pattern agrees with the phylogeny based on
rbcL data (Chase et al., 1993; Qiu et al., unpublished).
Furthermore, splitting Exbucklandia from Rhodoleia required
nine steps more than the minimum character changes.
Interestingly, Reinsch (1889) recognized the Bucklandieae
including Bucklandia (= Exbucklandia) and Rhodoleia based on
his comparative study of floral morphology. The clade of
Rhodoleia and Exbucklandia is located in the phylogenetic
tree (Fig.2.2) as the taxa linking the Altingioideae and the
clades of Mytilaria. Disanthus and the Hamamelidoideae. This
agrees with the implications from floral ontogenetic studies
(Bogle, 1986, 1989).
Tribal and subtribal relationships in the Hamamelidoideae
There is no doubt that the Hamamelidoideae are a
monophyletic group, as shown by previous studies (Bogle and
Philbrick, 1980; Endress, 1989a, b; Hufford and Crane, 1989)
and by this analysis. In the matK-based phylogeny, the
Hamamelidoideae clade is well supported by both BP and DI,
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and by five unambiguous character changes (Fig.2.2).
However, intergeneric relationships within the
Hamamelidoideae have long been debated (Harms, 1930; Schulze-
Menz, 1964; Chang, 1979; Endress, 1989c).
Three clades constitute the Hamamelidoideae (Fig.2.2).
The first clade includes species of Corylopsis and the branch
containing Mainaaya. Loropetalum. and Matudaea. The
relationships of Corylopsis and Mainaaya-Loropetalum-Matudaea
are strongly supported by BP and DI, and characterized by 10
unambiguous substitutions. The fact that Corylopsis was
phylogenetically far from Fortunearia and Sinowilsonia
supports the separation of this genus from the latter two
genera, as hypothesized by Endress (1989c). The well-
supported relationship between Mainaaya. Loropetalum, and
Matudaea is a new and unexpected pattern. Interestingly,
these genera share several morphological characteristics such
as bisexual flowers, long anther connective protrusions, and
valvate anther dehiscence. However, Matudaea differs
strongly from the other two genera in its irregular calyx,
lacking petals and large, variable number of stamens.
Loropetalum and Matudaea form a clade, but the supporting
indices are small (Fig.2.2). In addition, the other two
morphologically similar taxa (Embolanthera and Tetrathyrium)
were not available for this study. Therefore, the result is
not very conclusive.
The second clade in the Hamamelidoideae associates the
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subtribe Dicoryphinae with the tribe Eustigmateae (incl.
Molinadendron). The Dicoryphinae, which includes the genera
exclusively distributed in the Southern Hemisphere remnants
of the former Gondwanaland, forms a monophyletic clade. This
result agrees with the unique anther dehiscence pattern the
five genera share (Endress, 1989a). Considering the
exclusive distribution in the Southern Hemisphere and the
unique pattern of anther dehiscence, Zhang and Lu (1995)
suggested family status for the five genera. However, the
matK-based phylogeny does not support this proposition. In
contrast, the Dicoryphinae is closely nested with the
Eustigmateae and their clade is moderately supported by BP
and DI (Fig.2.2). As described above, Endress (1989c) placed
both Fortunearia and Sinowilsonia in the Eustigmateae, which
originally included one genus, Eusticrma (Harms, 1930).
Fortunearia and Sinowilsonia were put together based on their
similarities in leaf morphology and reduced petals (Schulze-
Menz, 1964), while the association of Fortunearia and
Eustiqma involves several common characteristics including
pedicellate flowers, covering sepals, small petals, sessile
anthers, large lenticellate fruits, and phloem in the
inflorescence axis containing libriform fibre groups
(Endress, 1989b). At the nucleotide level, two character
changes occur along the branch of the four genera (Fig.2.3).
Molinadendron is put in a clade with Sinowilsonia in the
matK-based phylogeny, which agrees with the implication that
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Molinadendron is closer to Fortunearia and Sinowilsonia. or
the Fothergilleae group than to Distylium (Endress, 1967).
My close examination of specimens of Sinowilsonia and
Molinadendron found that the two genera, along with Eustiama
and Fortunearia, share several characteristics including
linear stipules, covering sepals, naked floral buds, and two
prophylls flanking each lateral bud.
The third clade in the Hamamelidoideae is basically
genera of the tribe Fothergilleae, plus Hamamelis (Fig.2.2).
In Harms's system (1930), two tribes were recognized for the
apetalous Hamamelidoideae, one being Distylieae (Distylium
and Svcopsis) and the other the Fothergilleae (Fotherailla.
Parrotia, and Parrotioosis). However, Endress (1989c)
combined the two tribes and recognized the Fothergilleae
sensu lato because of the inherent connection between the two
tribes suggested by the production of a spontaneous hybrid
(XSycoparrotia) between Sycopsis and Parrotia (Endress and
Anliker, 1968). The matK data did not put genera of the
Distylieae and the Fothergilleae sensu stricto into
individual monophyletic clades, but clustered them into one
clade, thus supporting the tribal unification proposed by
Endress (1989c). My observations revealed that flowers are
not strictly bisexual, but andromonoecious in the
Fothergilleae sensu stricto. This further diminishes the
proposed distinctness of separate tribes of Harms (1930), the
Distylieae and the Fothergilleae, and supports the union of
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the genera into one tribe, the Fothergilleae.
As has been pointed out above, Mo1inadendron and
Matudaea do not sort out as members of the Fothergilleae.
The former falls into the clade of the tribe Eustigmateae and
the latter into the Cory lops is -Loropetalum group. These two
genera are apetalous, but bisexual, thus supporting their
separation from the andromonoecious Fothergilleae.
Furthermore, attempting to place Matudaea and Mo 1 inadendron
into the Fothergilleae clade entailed 38 more steps than the
minimum length.
The most interesting genus is Hamamelis. This analysis
placed this genus within the clade of the tribe Fothergilleae
(Fig.2.2). However, Endress (1989c) treated it as a
monogeneric subtribe of the tribe Hamamelideae. Hamamelis is
characterized by strictly 4-merous flowers and bisporangiate
anthers (Endress, 1989b; Mione and Bogle, 1990). The
parsimony analysis with Hamamelis removed from the data set
did not change the DI, but increased BP from 54 to 69 for the
clade of Fothergilleae (tree not shown). This implies a weak
relationship between Hamamelis and the other genera of the
Fothergilleae, and a stronger relationship within the
Fothergilleae.
The only obvious similarity between Hamamelis and the
Fothergilleae (Parrotia. Parrotiopsis. Shaniodendron) is the
presence of the semicraspedodromous venation in both groups.
However, the recent discovery of a hamamelidaceous fossil
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. flower in Upper Cretaceous deposits in Sweden might suggest
an ancient relationship among these genera since this fossil
flower had bisporangiate anthers, as in extant Hamamelis, and
a variable number of stamens, as in the Fothergilleae
(Endress and Friis, 1991).
Loropetalum. Hamamelis and genera of the subtribe
Dicoryphinae were previously grouped in the tribe
Hamamelideae (Harms, 1930; Chang, 1979; Endress, 1989c).
Forcing these taxa into a monophyletic group in the matK-
based phylogeny, However, required 21 steps more than the
minimum character changes.
Parrotia is morphologically similar to Shaniodendron and
Svcopsis (Bogle, 1968, 1970; Endress, 1970, 1993), and a
phylogenetic analysis of the Fothergilleae sensu Endress
(1989c) using nuclear DNA sequences supports the close
relationship of the three genera (Li et al., 1997d).
However, in the matK phylogeny, Parrotia is phylogenetically
far distinct from the other two genera (Fig.2.2). It is
likely that lineage sorting has occurred in the evolutionary
processes of the Fothergilleae. This will be discussed in
detail in Chapter IV.
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. U77092 AF013045 AF013042 AFO13043 GenBank AF013038 AF013039 AF015650 AF015651 AF013037
Hawaii Univ. of New Hampshire campusHampshire New of Univ. AF013046 MA.Arboretum, Arnold Hampshire New of Univ. Greenhouse Arboretum Washington of Univ. AF013047 SourceTaiwan number Tulear, MadagascarTulear, SC. Inc. Woodlanders, AF013040 AF013041 MA.Arboretum, Arnold AF013044 Gard. Bot. Missouri AF013059 Manuka State Roadside Park, Roadside State Manuka China MA. Arboretum, Arnold SC.Inc. Woodlanders, SC. Inc. Woodlanders, U77091 BOGLE QIU 03LI MA. Arboretum, Arnold LI 02 LI J.-H. LI J.-H. LI y.-L. J.-H. A. L. A. A. L.A. BOGLE A. L.A. BOGLE L.A. BOGLE A. L.A. BOGLE A. Randrianasolo 543 Randrianasolo A. Voucher HVI HVE DIR L. A. BOGLE DIC DIS FORFOT J.-H. 04 LI J.-H. LI campus Hampshire New of Univ. EUS N.-J. CHUNG EXB LFO LOR D. T. Omar CSP LSI Taxon Collector and CSI J.-H. accession ALT tutcheri DOT used in this analysis. Hamamelis Hamamelis virginiana Corylopais sinensis Corylopais Distyliopsis Distylium racemosum Corylopsis Corylopsis spicata Dicoryphe stipulacea Fortunearia sinensis Hamamelis vernalis Loropetalum sinense Species Disanthus cercidifolius abbr. majorFothergilla Exbucklandia populnea Exbucklandia Liquidambar formosana Liquidambar orientalis Eustigma Eustigma oblongifolium Altingia excelsa Table 2.1. Sources and vouchers of the species sequenced for matK gene and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Harvard Univ. campus campus Univ. Harvard Hawaii AF013054 PA. Gardens, Longwood AFO13058 Forest Research Institute, Institute, Research Forest Malaysia Kepong, AF025393 Garden Zurich. Botanical of AFO13048 Garden Zurich. Botanical of AF013049 Garden Zurich. Botanical of AF013050 Garden Zurich. Botanical of AF013051 Garden Zurich. Botanical of AF013052 Hampshire New of Univ. Greenhouse AF013053 Honolulu, Arboretum, Lyon U77094 China Jiangsu, MA Arboretum, Arnold AFO13056 AF013055 Guangxi, China Guangxi, U77093 SC. Inc. Woodlanders, AFO13057
BOGLE SAW ENDRESS ENDRESS ENDRESS BOGLE 05 LI BOGLE LUO ENDRESS ENDRESS BOGLE BOGLE QIU P. P. P. J L. A A A. SHASIN Y SYSTRI A PAR PPSRHO A MAT OST MAI MOL MYTNEO Z. NOA P. P. Trichocladus Trichocladus crinitus Ostrearia Ostrearia australiana Sinowilsonia Sinowilsonia henryi Sycopsis sinensis Parrotia persica Parrotia Parrotiopsis jacquemontiana Rhodoleia championi Shaniodendron Shaniodendron subaequale Table 2.1 Continued Neostrearia fleckeri Noahdendron nicholasii Maingaya malayana Matudaea trinervia Molinadendron guatemale Mytilaria laosensis
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.2. Locations and base compositions of amplification and sequencing primers used in this study. * this primer was synthesized with equal parts of "C" and "T" at base position 6.
Designed Primer 5' sequence 3' by Forward matKFl ACT GTA TCG CAC TAT GTA TCA T. Sang matKF2 GTT CACTAATTGTGAAACGT T. Sang matKFA ACC CCA CCC CAT CCA TCT J. Li matKFA-2 TGG TTC AAA CCC TTC GCT ACT J. Li matKF5 TGG AGYCCTTCTTGAGCGA* J. Li matKF6 TCA GTG GTA CGG AAT CAA ATG C J. Li Reverse matKRl GAA CTA GTC GGA TGG AGT AG T. Sang matKR2 TTC ATG ATT GGCCAG ATCA T. Sang matKR2-2 ACG GGGCCATAAGAA AGT CG J. Li matKR3 GAT CCG CTG TGATAA TGA GA T. Sang
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.3. Sequence characteristics of the matK gene (Taxon abbreviations as in Table 2.1). Divergence Species Length G+C(%) ALTCSICSPDIC DIS DIRDOT EUS EXB
ALT 1512 35 -
CSI 1515 34 6.1 -
CSP 1515 34 6.1 0.5 -
DIC 1515 34 4.7 2.2 2 -
DIS 1506 34 6.1 4.4 5 3.1 -
DIR 1515 34 5.2 2.7 2 1.3 4 -
DOT 1515 34 5 2.4 2 1 4 0.3 -
EUS 1515 34 5.3 3 3 1.1 4 2.1 1.8 -
EXB 1512 35 5.4 4.8 5 3.3 5 3.8 3.6 4.1 - FOR 1515 34 5.8 3.2 3 1.5 4 2.2 1.9 1.5 4.5 FOT 1515 34 5 2.3 2 0.9 3 1 0.7 1.5 3.6 HVI 1515 34 4.9 2 2 0.7 3 0.9 0.6 1.5 3.4 HVE 1515 34 5.3 2.4 2 1.1 4 1.3 1 1.8 3.8 LFO 1512 35 0.5 5.9 6 4.6 6 5 4.8 5.2 5.3 LOR 1512 35 0.9 5.8 6 4.4 6 4.9 4.6 5 5 LSI 1515 34 5.8 2.1 2 2 4 2.6 2.3 2.8 4.5 MAI 1515 34 5.8 2 2 2 4 2.1 2.2 2.8 4.5 MAT 1515 34 5.9 2.2 2 2.2 5 2.8 2.5 3 4.9 MOL 1521 34 5.2 2.6 3 0.7 4 1.8 1.5 1.3 3.7 MYT 1515 34 5.7 4.1 4 2.4 4 3.2 2.8 3.2 3.9 NEO 1515 34 5.2 2.6 3 0.5 4 1.7 1.4 1.3 3.7 NOA 1515 34 5 2.4 3 0.4 3 1.6 1.3 1.3 3.6 OST 1515 34 5.3 2.7 3 0.6 4 1.8 1.5 1.6 3.8 PAR 1515 34 5.5 3.4 4 1.8 4 2.2 1.9 2.6 4 PPS 1515 34 4.8 2.1 2 0.6 3 0.7 0.4 1.4 3.4 RHO 1512 34 5.2 4.7 5 3 5 3.7 3.4 3.8 2.4 SHA 1515 34 5 2.4 2 0.9 4 0.6 0.3 1.7 3.5 SIN 1515 35 5.6 2.8 3 1.2 4 1.9 1.5 1.7 4.1 SYS 1515 34 5.2 2.5 2 1.1 4 0.3 0.1 1.9 3.8 TRI 1515 34 5.4 3 3 1.1 4 2.2 1.9 2 4.2 SAX 1518 32 12 12 12 11 . 11 11 11 11 12 SUL 1521 34 8.9 9.1 9 7.5 8 8.1 7.8 8.1 8.3
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.3 Continued
Divergence Species FOR FOT HVI HVE LFO LOR LSI MAI MAT MOL MYT NEO NOA OST ALT CSI CSP DIC DIS DIR DOT EUS EXB FOR
FOT 1.9 -
HVI 1.9 0.5 -
HVE 2.2 0.9 0.4 - LFO 5.6 4.8 4.7 5.1 -
LOR 5.5 4.7 4.6 5 0.7 -
LSI 3.1 2.2 1.8 2.2 5.6 5.4 -
MAI 3.1 2.2 1.8 2.2 5.6 5.5 1.5 -
MAT 3.2 2.4 2.1 2.5 5.7 5.6 1.7 1.8 - MOL 1.7 1.3 1.1 1.5 5 4.8 2.4 2.4 2.6 -
MYT 3.5 2.8 2.6 3 5.4 5.2 3.6 3.6 3.9 3 -
NEO 1.9 1.3 1.1 1.5 5 4.8 2.4 2.4 2.6 1 2.8 -
NOA 1.7 1.1 0.9 1.3 4.8 4.7 2.2 2.2 2.5 1 2.7 0.5 - o 00
OST 2 1.4 1.2 1.6 5.1 5 2.5 2.5 2.8 1 2.9 • 0.7 - PAR 3 1.9 1.7 2 5.3 5.2 3.2 3 3.5 2 3.6 2.2 2 2.3 PPS 1.8 0.3 0.2 0.6 4.6 4.5 1.9 1.9 2.2 1 2.5 1 0.9 1.1 RHO 4.2 3.4 3.2 3.6 5 4.7 4.2 4.4 4.6 3 3.8 3.4 3.3 3.6 SHA 2 0.5 0.5 0.9 4.8 4.6 2.2 2 2.4 1 2.8 1.3 1.1 1.4 SIN 1.6 1.7 1.5 1.9 5.4 5.3 2.8 2.8 3 1 3.2 1.6 1.5 1.7 SYS 2 0.8 0.7 1.1 5 4.8 2.4 2.3 2.6 2 3 1.5 1.4 1.7 TRI 2.4 1.8 1.6 1.9 5.2 5.1 2.9 2.9 3.2 2 3.4 1.3 1.3 1.5 SAX 12 11 10 11 12 12 11 11 12 11 12 11 11 11 SUL 8.7 7.8 7.6 8 8.6 8.6 8.6 8.5 9.1 8 8.2 8 7.8 7.9
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE 2.3 Continued
Divergence Species PAR PPS RHO SHA SIN SYS TRX SAX SUL ALT CSI CSP DIC DIS DIR DOT EUS EXB FOR FOT HVI HVE LFO LOR LSI MAI MAT MOL MYT NEO NOA OST PAR
PPS 1.6 -
RHO 4 3.1 -
SHA 1.7 0.3 3.4 - SIN 2.7 1.4 3.8 1.7 -
SYS 2 0.5 3.6 0.4 1.7 - TRI 2.7 1.5 4 1.8 2.1 2 SAX 11 10 11 10 11 11 11 - SUL 8.4 7.6 8.1 7.8 8.4 8 8.2 8.4
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 8 8 5 8 8 6 6 6 8 8 8 8 7 gag 15 15 7 15 16 16 17 17 16 7 15 7 16 7 15 7 15 8 15 8 19 16 7 16 17 16 7 15 16 6 16 7 15 15 15 16 16 7 16 7 15 8 Glu GAA 3 15 8 4 4 4 16 7 4 4 16 8 4 ______16 14 4 15 5 15 3 15 15 16 13 16 3 16 Gin caacag
7 55 17 3 4 15 3 5 5 5 15 4 7 5 5 15 4 ______1 5 14 4 1 5 15 3 1 4 14 3 1 0 5 15 3 0 0 0 5 15 4 1 5 15 3 0 0 5 15 3 0 6 14 3 0 0 00 5 6 16 4 15 3 0 6 15 4 0 0 00 50 5 5 16 3 16 3 15 3 0 50 15 4 Cys ugcugu
14 16 14 0 6 15 3 1414 1 4 13 5 ______5 14 5 15 5 16 2 5 4 15 4 14 0 5 16 4 5 14 5 144 0 13 5 15 5 13 4 14 5 144 0 5 5 15 5 14 5 14 0 5 5 14 5 14 1 4 5 15 5 6 15 5 14 5 14 5 14 Asp gacgau
23 5 12 26 21 21 23 23 23 23 25 23 24 23 ______7 5 24 5 14 5 23 5 24 5 24 4 16 8 6 23 5 5 23 5 23 5 5 21 35 16 23 5 5 23 5 5 23 5 14 5 5 23 6 6 22 5 13 4 8 205 5 23 5 5 Asn aacaau
4 3 5 5 5 7 5 4 5 5 8 5 5 cgu
5 5 2 2 3 3 8 46 5 4 5 4 8 8 21 5 5 5 4 4 5 5 6 cgg
1 3 4 6 22 1 1 1 3 5 6 23 cgc
______9 1 9 3 8 2 9 1 9 9 10 1 4 10 10 10 1 4 10 1 10 10 1 4 10 10 1 10 1 4 Arg cga
4 8 4 6 5 25 5 4 4 4 3 5 4 9 1 5 5 5 10 0 4 4 9 4 agg
6 6 4 9 1 4 5 5 6 6 6 4 9 5 7 6 4 10 1 4 5 6 4 9 1 5 5 6 5 8 aga 9 7 9 9 9 9 7 8 11 11 10 10 7 10 1111 6 6 5 10 10 6 4 10 5 4 11 1 5 6 10 6 4 9 1 5 5 1 9 9 6 1 1 10 6 4 8 1 4 1 11 2 9 9 3 1 11 210 5310145 524 2 9 6 4 9 1 5 2 10 6 4 2 10 6 3 3 10 6 4 2 2 11 6 4 10 1 3 ______5 2 10 6 4 9 1 5 5 5 2 10 6 4 5 2 5 1 5 1 6 3 55 2 1 5 2 5 2 5 2 10 6 5 2 10 6 4 9 1 Ala 1 1 1 1 1 4 1 1 2 3 2 1 5 2 (Taxon abbr. as in Table 2.1). SAX 4 2 SUL 3 4 SHA 1 SIN SYS TRI 0 5 3 FOR 5 FOT HVI 1 5 PAR 1 4 PPS RHO 3 CSI DISDOT 4EXB 5 1 1 3 HVE 1 5 LOR 1 LSIMAI 2 2 5 NEO 1 5 OST 2 4 DIR CSPDIC 1 3 2 ALT 1 6 EUS 1 5 2 LFO 1MAT 6MOL 1 2 4 MYT 2 5 NOA Species Species GCAGCCGCGGCU Table 2.4. Codon usage of the matK gene in the Hamamelidaceae
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced
Table 2.4 Continued •H JS rH rH J (3 X cu j E ■p 04 M m a) >1 (0 0) 0) u u 0 >i 9 9 U u o 3 3 < 3 3 3 CD a CD CD CD 3 3 3 o < 3 < 3 3 3 a 3 CD 3 3 3 D D < D 3 CD 3 2 < S a < 3 3 3 3 < 3 3 3 < a u 3 3 U O D 3 < < 3 CJ < 3 3 3 3 < 3 n i n t n i n n t f i n t n ^ ^ i n i n i o r o ^ ^ m i n i n t n t n i n t n M O i n i n i n i n t o n ^ H HCMNNHOHMOOMMHHNfOnNHHCnnMfOtMHCMOHM'Ofn nnnnnfnn^^rnnnfOonnconfnNnfn^fnn^nnnmnn cococororocorocMrococMcococococococorococoeocororococoTpcorocNTp on rH rH - r CO rH r** CMCMCMCMCMCMCMCNCMCMCMCMCMCMCMCMCNCMCMCMCMCMPMCMCMCMCMCMCMCMCMCM in^o,00000p-p-000000000000000r*-000r*-00 oDp-0p-0p-p-*p**p-p'p-r-p-p-p-p**p-cop-*p'r-p'*p»p-p-0p-GOp-co0CDOjHHHHHHHHHHHHMHHHHHHHHHHHHHHHONn TP rH rH GO 0 TP aN0p-oDaNr'*oocop-cooooooooNOocococooocNp**p*coaNOoaococooDCOTpp«. ,*»0000 MCM CM P- rH m TP rH 0C 010 10 CO 10 10 HrH rH rH m m m m o o MC CM CM CM on Hr HrH rH rH rH IT CM CM 10 OTP CO H rH 00 10 OC CO CO CO CM rH P* HrH rH CM CO HCM rH o o in rH H H NONON m ON rH HH rH o o o CM in ON HrH rH m rH HH rH o o rH MC CM CM CM MCMCM m CM CM 00 TP oo rH m rH a HH rH H p- m o H nm in H CO HrH rH rH o CM CM 10 PTf TP rH 00 an in rH p- CM CM CM CM -p* p- CO HH rH HrH rH in an ON H rH H *P ONON rH HrH rH nin in o o ON rH CM OTP CO H rH rH PTP TP TP 00 rH P- MCMCM CM in rH V0 NONON rH rH CM rH rH Hr rH rH rH GO rH 0 rH o CM nm in m o HrH rH 111 rH rH CM VO CM CM CO rH rH in p- rH NO TP rH o rH CM CM o CM p- HrH rH rH PTP TP ON 0p» V0 rH r-* H m rH CM CM TP p* rH H 01 rH NP - 0 P-* P* ON H GO rH p* rH 00 HrH rH rH V0 CM CM P** CO CM H no in rH P*» rH CO in Hr Hr HrH rH rH rH rH rH in o o H TP rH rH rH CM CM 10 in CM CM CM rH CO 10 O t - ( O O O O O m O O O O t H O V0 rH rH rH nm in o MC MCM CM CM CM in HrH rH rH 11 CM TP rH P* HrH rH rH CO m rH 0 0 rH o o O 0 0 0 0 0 0 •O' P^ ON ON MCMCM MCM CM rH f T an HrH rH TP rH 0GO 00 0 rH O H HHr rH rH H rH CM CM CM H H H H HHr HrH rH rH H rH ON H CO TP O0 CO 0 0 0 0 0 0 0 0 0 0 0 0 rH rH o o rH CM 0 an 0 0 0 rH rH TP 0 0 0 O O an CM CM CM CM rH H m P-* rH TP rH ON HH rH o MCMCM MCMCM MCM CM 0 CM CM an -ON p- rH TP Hr rH rH rH rH o H CO O' rH rH OC CO CO CO H -0 000 00 00 P- rH O ON © rH MCMCM CM H rH rH TP CM -P* P- H Tp rH ON rH O O CM Hr HrH rH rH rH H 00 H CM rH •H -0C 0 0 CO 0 P- rH an r-» rH rH P- rH co o o o o rH MC CM CM CM CM ^4 rH *O in O' M* 0 rH HrH rH OCO CO HrH rH HrH rH 0 o ON 0 HrH rH CM on rH 0 0 HrH rH ON rH H P- 0 0 O an rH H CM N0CO 0 ON CO 0 0 ON 0 Hr rH rH rH rH HrH rH rH O MCMCM MC CM CM CM CM iH HHrH H rH CM P«» PTP TP rH TP TP rH co o o rH 10 CO CM 00 ON rH TP rH CM 0 H t P CM CM GO 0 ON 0 0 rH rH rH TP rH rH 0 0 rH CO CM TP Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced Table 2.4 Continued j CO > E E a> 0) <0 G 1 > u U U U U Q< h h ^tNCMnfnnn^cNinnnn^inoiNCMnnnnn'fnfnfninnfn^cN VO H ( N ( S C M f S N ( N M H i nn ( M m ( M i M n H H i ( N f ( t N N ^ ( m M H i ( S n N i H ( f M ( l N « O N C M N D f i M O invovovoinvomvoTi* n { N t n ,3'inin*3, i o v n m n n in i m o n v i o v n in ' O in o w v ^ in i o n v m i o n v t o v f o l v i in n r-» i o n v >\o o i n i n i n < o 2 r ' 2 « « o o i ^ c ^ r ^ r ^ © r ^ c o © f l o 2 H ® r ‘ 'C O I ^r ^r v ®®r ^®®p ‘c ^v 0 0Q 'C \ 0 0 O ' \ Q \ C Pioinvoinn-tninininvotnvo\ovovoiov©i©vointninmHrinv©ininintnvoin f " « - O \ g \ C P q v < O 0 0 Q P O \ O \ O \ Q 0 ^ Q P O \ O \ 0 0 G 0 C 0 0 D 0 D t , - * 0 > C 0 O V t * - » V O 0 D N(Nf ' H M M ( f') N N ( N M ( N fN N ( fN N H N M H M C m N M C N N J r**» O M H M J O H M C N M S ( O f H f M r H O \ r H r H O N O O r H r H O O O O N r H r H Ooooooooooooooooooooooooooooooooo N ' - H ' ‘ H 'oooooooooooooooooooooooooooooooo - H r H O r H r H r H O r H O O O O N M ( Nn M n f f o M n ^ M ^ ^ M N c ( » N ) { { N ^ N ' N i n ( ( N n f C M ( M N i C n i N n N ^ ' N t f H ^ M < n N o j C ^ M ,n ^ ( ( n i n M ^ ^ ' { n i M n ^ ' f > l < N N N N N N f M ( N M ( > i ( N r*» rH r t H H H H H H H H N H H H r t H H H H H M N H H H H H r t H H H N n t ~ r ~ r ^ [ ~ r ' v o r ~ r ~ r ~ i o t ^ r ~ c ~ r ~ v o t ~ r ~ r ~ t ~ - r ^ r ~ r ~ r ~ c o [ ~ t ~ t ~ r ~ r ~ i — a > 2 13 00 on o 13 OO o 13 10 on o CO 15 CM - r 0OO 00 rH 14 CO *H o o o 14 10 m on 000 00 14 VO 14 10 00 Hf 000 00 13 VO o o 12 11 non onon CO 16 000 00 HI* © 16 112 o 16 VO VO 00 CM 13 VO r* r-* H 13 000 00 CO o o 15 10 00 16 00 HI* rH 15 10 ON in 00 o 15 VO CM 14 10 ON 000 00 o o 15 on 15 000 00 CO o 14 10 o nM M* M* in 16 10 ON 00 o 16 r*- r— rH 13 as o m 00 16 m ON o o o 14 i i m CO 14 10 ON in 00 17 in rH m * c—C*- HrHrH ON VO rH Fig. 2.1. Approximate relative locations of matK primers (base compositions are listed in Table 2.1, shaded areas are introns). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ______5]______3' trnK j j matK ^ trnK matK FI F2 F4 / F4-2 F5 R3 F6 R2/R2-2 R1 > > > > <><< 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.2.2. The Strict consensus of the six most parsimonious trees of 582 steps long based on sequences of the matK gene. CI=0.85, RI=0.82. Numbers above and below branches are decay indices and bootstrap percentages; boxed numbers denote the three major clades in the Hamamelidoideae. Taxon abbreviations as in Table 2.1. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae. 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I 100L LFO 100 I------LOR- 7 I---- CSI=T 100 I. CSP J 100 l*SI-|LORinae 61 (HAMeae) FOTeae 96 MAI J DIC NEO NOA 58 71 H OST TRI- 1 81 EUS-i •H EUSeae 69 FOR- 85 MOL FOTeae 62 SIN-I DIR-1 70 SYS 92 DOT 91 SHA 66 FOT PPS <50 HVI ea) d 72 DIS-i MYT EXB-1 RHO-RHOideae SAX (Outgroup) SUL 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.2.3. One of the six most parsimonious trees of 582 steps long based on sequences of the matK gene, showing the number of unambiguous substitutions along the branches. Taxon abbreviations as in Table 2.1. 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ALT 35 LFO LOR 14 CSI CSP LSI MAT 11 MAI DIC NEO NOA OST 14 TRI EUS FOR MOL SIN DIR SYS DOT SHA FOT PPS HVI — HVE —21 PAR 35 DIS 25 MYT 10 EXB 14 RHO SAX SUL (Outgroups) 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDICES 2.1-2.2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 80 ...... C ...... T ...... T. T T. T C ...... C T ...... A A A A A A A A A ...... G ...... C C ...... T C . C A C .A...... C . .T. . T...... C ...... T T T T T T T T T T T T T T T T T T T T T T T A T T T A... T A...... Saxifraaa (SAX) and Sullivantia (SUL) are the outgroups. Taxon abbreviations as as in Table and 3.1. Gap31'-', represents same base as in the first taxon. DIC DIC DIS DIR FOR FOR Appendix 2.1. Aligned sequences of the chloroplast gene matK in the Hamamelidaceae. CSP DOT SUL SUL G T PAR PAR PPS ...A SHA SIN SYS SAX EUS EUS EXB FOT HVI HVE CSI CSI RHO RHO TRI ALT ATGGAGGAAT CTCAAGGATA TTTAGAACTA GATAAATCTC GGCAACATGA CTTCCTATAT CCACTTATCT TTCAGGAGTA CCACTTATCT ALT CTTCCTATAT GGCAACATGA GATAAATCTC TTTAGAACTA CTCAAGGATA ATGGAGGAAT NEO NEO NOA OST LFO LFO LOR LSI MAI MAT MOL MYT MYT Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. H0VO10H0H0H010V0H0V0H0HOH0H0H0101OV&VOV0HOV010V0HO10H0H0oooooooooooooooooooooooooooo < CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ oa < CD Eh CJ CJ TJ cm X •H •o C >fc«ocn O >h W O CO < Oh a sc h SYS SYS SAX SUL C C C GAT T A.A..GC ..A CA. .T...C.TA. C .C. CT. [160 T G T.. ..A A C C [160 [160 a* dlCJCJQQQQNHbbSICJriJ £ £ Z Z O Oh Oh a co co TRI A .A . . G.A. C [160 121 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 [240 A. . A. . . ..A ...... A...... T .AG .•. GAACCGTTTG ATTATTTCCG CTAATGATTC TAACCAATAT . .A. . .A. . TGTATCAGCA c ...... c . . .c . .c . .c . .c . . .c . . .C . .c . . .C . . .c . .c . .c . . .c . . .c . .c . .c . ..c . .c . . .c . . .c . . .A . . .c . CATTAATTGTGAAACGTTTAATTACTCGAA DIC DIS DIR FOT SHA CSP DOTEUS .c . HVE LFO HVI CSI SAXSUL .c . EXB.c . LOR LSI MAT PAR PPS.c . SIN.c . TRI MAI MYT Appendix 2.1. Continued ALT FOR SYS MOL NEO NOA.c . OST RHO.c . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 [320 A [320 ...... C ...... G T ...... A A ...... A A ...... [320 G G G G G G G GA G G G G G G G GA G G G G ...... G G G G G G G G G G G G G G G G G G G G G G G G G G G G G ...... A .A GG ...... TGATATC ACAGGGATTT GCAGTCATTG TGGAAATTCC TGGAAATTCC GCAGTCATTG ACAGGGATTT TGATATC [320 ...... T ...... A .A GG ...... ------...... T ...... T T ...... A A A A A A A A ...... ___ ...... ___ ...... ___ ...... ______...... ___ ...... ___ ...... ___ ...... A A A A A A A A A A A A A A A A A A A A A A A A A CG A A TA TA ...... T ...... T T A..T.C... .A A..T.C... AG.. .A AAA ATA ...... T ...... DIC DIC HVI SHA SHA SIN SYS SUL AA DIS DIS A DOT FOR FOT HVE DIR DIR CSP CSP ..T LOR SAX A CSI CSI ..T EUS EUS EXB MOL MOL NOA PAR PPS RHO AT Appendix 2.1. Continued ALT TCTCAAA GGATTTGTAT GGCACAACAA CCATTTTTGG MAT OST TRI LFO LFO NEO NEO MYT MYT * * LSI h £ MAI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o Eh a a § o Eh Eh 2 CJ EH < CD a < Eh CJ CJ CJ CJ CJ a cj CJ CJ CJ CJ CJ CJCJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CD CD CCDCDCDCDCDCDCDCDCDCDCDCD CDCDCDCDCDCDCDCDCDCDCDCDCDCDCD < <* Eh CJ Eh Eh CJ CJ Eh Eh CJ T3 Is CD 0) 3 CD £3 g CJ CJ CJ CJ •H Eh +J < e CD o CJ o g CJ CJ CJ CN CJ Eh Eh K •H Is -o c _>faocQ< 124 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o o o o o o o oo o o o o o o o o o o o o o o o o o o o o o o o 0000000000000000 oooooooooooooooo oooooooooooooooo oooooooooooooooo N* N* *1* N1 N* o C3 T3 CN CJ X < < «: < < < < < < < < CJ CJ < < < < < < < < < •H *o c 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 sfin r [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 [560 .T. ,T. .T. T. . TCTACGAGTA CGATTCTTTC • C. . C. • CTTCTTCTTT GCATTTATTA T...... GTGAAAGACG TCGATACTGG C. .C. C. T GAAATATTGG TTCAAACCCT GAAATATTGG DOT FOR DIC EUS DIS DIR FOT HVI CSP SHA SIN SYS Appendix 2.1. Continued ALT EXB CSI HVE PPS SAX SUL RHO LFO LOR LSI MAI MAT MOL NEO NOA OST PAR TRI MYT Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 [640 ...... A A.. ... A A A ...... A G ...... TTTTTTCAAA AAGGAATCAA AGATTATTCT AAGGAATCAA TTTTTTCAAA CTA CTA CTA CTA CTA CTA CTA CTA .CTA .CTA .CTA CCTA ...CTC ..CTA . ...CTC ...CTA ...CTA ...... CTA ...CTA ...CTA ...CTA ...CTC ...... CTA ...CTA ...... CTA ...... 0 C TCCATTTTCG r TCCAAAGAAA ————— ----- ___ _ c c c c q C-— C———— c c c c c c c . ..CTTCAAC . TTAA ...... ...... TA .TA .TA .TA .TA .TA .TA ..AG...C.A »C...... AG...C.A .TA .TA .TA TCGTAATTGG AATAGTCTTA TCGTAATTGG DOT DIS DIR CSITA • EXB FOR SIN .TA EUS FOT HVI .TA SYS DIC CSP HVE ALT SAX SUL SHA .TA LFO LOR LSI MAT PAR PPS TRI MAI RHO Appendix 2.1. Continued OST MOL MYT NEO NOA 0— Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. oooooooooooooooooooooooooooooooo (NIMNMNtMINNIMNNNOINlMNNPINNNNNNPIIMlNOINNtMN r~ [— r~ r~ [— r^f^r^r^E— r^r^i — r~ c— [— c^c— r~~ aft EH < cj cj < Eh Eh Eh CJ ft CJ Eh Eh CJ Eh Eh CJ Eh 3 CJ CJ 2 Eh CJ CJ Eh CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ Eh • • CJ Et Eh Eh" Eh Eh Eh Eh Eh Eh Eh Eh CJ CJ CJ CJ CJ CJ CJ cj a a cj CJ CJ CJ CJ CJ CJ a a < < CJ Eh Eh CJ Eh CJ CJ < < ft < < ft ft CJ CJ Eh 2 CJ CJ < Eh 3 CJ Eh CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ g CJ CJ CJ CJ < Eh CJ Eh < T3 CJ 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 [800 ...... T T ...... T T T T T TT T T T T T T T T T T T T T T T T T..GG. T T T T T T T T T T T T T T T T T T T.- T T T T TA TA T T T .C T A. T T ...... A ...... T ...... A A A A A ...... T. .C T...... T T G ...... C C C C ...... M ...... A A A ...A ...... T ...... __ ...... T T T T T T C T C T T T T T T T C T T T T T T T T T T T T _ A.T...T. A.. . . A. . A...... DIC DIC SYS SYS SAX SUL .T CSP CSP DIS DIR DOT HVE SHA CSI CSI EUS EUS EXB FOR FOT PAR PAR PPS SIN TRI Appendix 2.1. Continued HVI MAT MAT LFO LFO LOR MOL MOL NEO MAI MAI RHO ALT ACATCTTCTG GAGCCCTTCT TGAGCGAATA TATTTCTATG GAAAAATAAA ACATCTTGTA GAAGCCTTTG CTAATGATTT CTAATGATTT GAAGCCTTTG ALT ACATCTTGTA GAAAAATAAA [800 TATTTCTATG TGAGCGAATA GAGCCCTTCT ACATCTTCTG MYT MYT NOA NOA OST (—* (—* LSI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [880 [880 [880 [880 [880 ...... G G G G [880 G [880 [880 G [880 G [880 [880 G G G [880 [880 G G [880 [880 G G G G [880 [880 [880 G G [880 G G [880 G G G G G [880 G [880 [880 [880 [880 G G [880 G G ___ CC G G [880 G G ..A A G ..A T. [880 A G ..A ...... A A G [880 A G [880 A A A A A A A A A A A A A A A A A A A G [880 ____ G...A ...... A A .... A.CC.. G..TA..T A.CC.. A ...... TA TA TA TA TA TA TA TA TA TA TA TA T TA TA TA TA TA TA TA TA N N.. TA TA TA TA TA TA ...... C. TTA C. G ...... SHA SHA GC. ..A SAX .TA . . A. .CTA.TG. SUL C. . . . SIN SIN N SYS SYS TRI OST OST PAR PPS C RHO RHO DIS DIS DIR FOR HVI HVE HVE CSI CSI EXB DIC DIC FOT DOT DOT BUS ..G ALT TCAGGCCATC CCGTGGTTGT TCAAGGATCC TTTCGTGCAT TATGTTAGGT ATCAAGGAAA ATCAATTCTC GCTTCAAAAG GCTTCAAAAG ATCAATTCTC ALT ATCAAGGAAA TATGTTAGGT [880 TTTCGTGCAT TCAAGGATCC CCGTGGTTGT TCAGGCCATC LSI MOL MAT MAT MYT NOA NOA CSP CSP LFO LOR NEO Appendix 2.1. Continued g MAI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. oooooooooooooooo oooooooo oooooooo VO VO vo vo VO VO VO VO VO vo vO VO vo vo vo vo vooiovovovcKovovoi vo vo vo vo vo vo vo 0\0'\0>0V0\O < CJ 1 CJ eh CJ CO E-i CD CD EH CDEH CD CJ < E h Eh EH e h CD CD CD Eh CD T3 0) I 3 CD c % CJ •H CD -p E h c a o EH u Eh CJ a < % Eh a U a CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD a CD CD CD CD CD CN cj • Eh ••••••••••• CJ CD CD X •SCDCDCDCJCDCDCDCDCDCDCDCD CDCDCDOCDCDCDCDCDCDCDCDCDCDCDEH •H TJ c d) a EHMCUCJCOOSEHCOmoSEHH BOHiHHEHjEHOrtEH_ _ . , _ (KCOOrtJZCOHXPl ^COCQHHHODXPP> .XMOlOrtlHSEH^iSO % SSZZOP hPhKOICOIOECDCO 131 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [1040 [1040 [1040 [1040 [1040 1040 [ [1040 [1040 [1040 [1040 [1040 [1040 [1040 [1040 [1040 1040 [ [1040 [1040 [1040 [1040 [1040 [ 1040 [ [1040 [1040 [1040 [1040 [1040 [1040 [1040 1040 [ [ 1040 [ [1040 ...... c C . C C ....c TAAATCCTTC ____ c c c c c c c c c c c c c c c c c c c c . . . C c ...... AGTGTGGGAT ...... c c ...... A C ...... ---- G TA CG G .TC. CATTCCTTTT ACTTTCTGGG CTATCTTTCG T. . . . T. A. . . . A...... T ...... G. . .G. .G. . .G. .G.. .G. . .G. .G. .G. .G. .G. .G. .G. . .G. . .G. . G. • .G.. . .G. . .G. . .G. .G. .G. .G. . .G. . .G. .G. .G. . .G. GTAAGGATCT ATATAAACCAATTATCCAAT CSI CSPDIC . .G. DIS DIR DOT. .G. SUL PPS. SHA .G. SYS SAX EUS Appendix 2.1. Continued FORFOT G. . HVI • . .G. PAR RHO. .G. TRI NOAOST . .G. SIN MAIMAT. .G. . .G. ALT EXB HVE LFO LOR MAI MAI .G MYT NEO. .G. LSI MOL Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 [1120 C C [1120 ...... A A ...... A A ...... G ...... A A A A A A ..C A. A A A A A A A A...C A A A A A A A A A A...C T A T T A T T A T ...... C C G.R ...... T T A C.. A ...... G G G G G G G G G G G G G G G G G G G G G A .G A .G ...... G ...... DIS DIS ...C CSI CSI CSP CSP DIC DIR DOT S. .G ALT AGTGGTACGG AATCAAATGC TAGAAAATTC GTTTATAATA GATAATGCTA TTAAGAAGTT CGATATCATA GTTCCAATTA GTTCCAATTA CGATATCATA ALT TTAAGAAGTT GATAATGCTA [1120 GTTTATAATA TAGAAAATTC AATCAAATGC AGTGGTACGG EUS EUS EXB G FOR C... .G G C... .G Appendix 2.1. Continued PAR PAR SAX SUL ..C SHA SHA SIN SYS C... .G FOT FOT HVI HVE OST OST PPS RHO LFO LFO LOR MAT MAT MOL MYT NEO NOA C... .G TRI U> U> MAI £ LSI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 [1200 A [1200 .... T ...... T T TT....T... [1200 T T [1200 ...... T T ...... C C C C C C C C C C C C C C C C C C C C C C C C C C T T... c C ---- G....A T. .T G ...... T T T T ...... T ...... C G ...... G C..G ...... G ...... T T ...... T A ...... A.. CA A.. A ...... DIC DIC T.A DIS DIS DIR DOT T.A T.A HVI HVI T.A SUL SUL .A A.. CSP CSP T.A SHA SHA SYS T.A T.A PPS PPS SIN T.A SAX T.A CSI CSI T.A Appendix 2.1. Continued EUS EUS EXB FOR ..G.G FOT T.A HVE T.A T.A T.A NEO NEO T.A TRI T.A PAR PAR RHO T G ALT TTCCTCTGAT TGGATCATTG GCTAAAGCGA AATTTTGTAA CGTATTAGGG CATCCTATTA GTAAGCCGGC CCGGGCCGAT CCGGGCCGAT GTAAGCCGGC ALT CATCCTATTA CGTATTAGGG [1200 AATTTTGTAA GCTAAAGCGA TGGATCATTG TTCCTCTGAT MAT MAT MOL T.A T.A LOR LOR LSI MAI T.A C . MYT T.A. NOA OST A.A T.A T.A LFO LFO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [1280 [1280 [1280 G. G. [1280 ..... A A [1280 A [1280 A A [1280 A [1280 A [1280 [1280 A [1280 A [1280 A [1280 . . . . [1280 CC A [1280 A A [1280 A [1280 A [1280 A C C [1280 C C A [1280 [1280 [1280 [1280 C...CC [1280 A C C C C [1280 [1280 [1280 A A [1280 [1280 A [1280 C C [1280 _____ ...... T T T T T..G T T T T T T T T T T T T T T T T T T T ---- ______---- T T T T T T T ...... G ...... A. A. ..A A ...... G A ...... A ...... T T T T T T T T T T T T T T T T T T T T..W T T T AT ...... C C G ...... A T G...G ...... A ...... A T ...... T T...A ...... G ...... T G G G G ...... A ...... A ...... SAX G SUL SHA SIN SYS TRI DIS DOT DIR EXB FOR MAT NOA PAR PPS C CSI DIC FOT LOR MOL RHO EUS HVI HVE LFO MYT OST C CSP NEO Appendix 2.1. Continued ALT TCATCAGATT CTGATATTAT CGACCGATTT GTGCGTATAT GCAGAAATCT TTCTCATTAT CACAGCGGAT CCTCGAAAAA [1280 ^ MAI i_i i_i LSI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 [1360 ...... A [1360 A A A A A A A A A A A A A A A [1360 A A A A A [1360 A A A A [1360 A..T... [1360 CA C. C. ...A C. ...A ...... T ...... G ...... A ...... C C C G ...... G ...... G ...... TG ...... SHA SIN SYS SAX ...CT SUL PPS TRI PAR RHO OST ...A DIS DIR HVI HVE MYT NOA ...A ALT CSI AAAGAGTTTG TATCGAATAA .M AGTATATACT TCGACTTTCT TGTGCTAGAA CTTTGGCTCG TAAACACAAA AGTCCTGTAC [1360 CSP DIC DOT EXB FOR FOT LSI MOL LFO MAI NEO ...A Appendix 2.1. Continued LOR MAT EUS Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [1440 [1440 [1440 [1440 1440 [ [ 1440 [ [1440 [1440 [1440 [1440 [1440 [1440 1440 [ 1440 [ [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 [1440 T T T. T. T. T. T...... TTTGATCGTC ...... C G. G. G...... GAGGAAGAACAAGTTCTTTC ATTCCTTACG TATTGGAAGA .G. . .G. .G. . .G. .G. . .G. • G. . G. • . G. • . .G. ....T.. .G. GGTTCGGAAT c ...... C..C GTGCTTTTTT GAAAAGATTA DIC DIS ALT CSP HVE DOT EXB FOR CSI DIR FOT SAX SUL Appendix 2.1. Continued HVI SHA SIN SYS TRI EUS LOR PAR OST PPS RHO MAT MOL LFO MYT NOA NEO LSI m CO -o MAI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 [1520 T. T. [1520 ...... T T T T T T T T T T T T T T T T T T T T T T T .T T . .T. . [1520 T. T...T...T [1520 T T. T. .T ...... C C C C C C C C C C C C C C C C C C C ...... G..C G C...C ...... T T T ...... C T ...... C T T T T T T T T K T T T T T C. C. ...T C. C. ...T ...... G...G G A G G G G G G G GG G G G G ...... T G ...... T G G G G G G G ...... T G ...... T G G A .G ...... C C ...... C C ...... A ...... T A. .G A C ...... C.A C . A C . A C . C .A C.A T .T G. .T G. AA. .G ..... C...T .T ...... G T ...... GCTT CTTCTACTTC GCGGAGGTTA TATAGAGGGC GTATTTGGTA TTTGGATATT ATTTGTATCA ACGATCTGGC [1520 ...... T ...... --- .... DIC ...AAA DIR ...AAA DIS ...AAA ALT CSI CCA ...AAA CSP ...AAA DOT EUS ...AAA EXB ...AAA FOR ...AGAFOT ...AAA HVI ...AAA HVE ...AAA ...AAA Appendix 2.1. Continued SUL ...AGA PAR PPS ...AAA ...AAA SHA SIN ...AAA SYS ...AAA ...AAA SAX .AGAA . . LFO LOR ...... MAT MOL ...AAAMYT ...AAA NEO ...AAANOA ...AAA OST ...AAA ...AAA RHO ...AGA TRI ...AAA M LSI ...AAA oo oo MAI ...AAA Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced Appendix 2.1. Continued < 5jcncoHHnoDxoo> U C J' ‘ E H H O < U W B 5 E H m m «u E H1 . . cj < % 0 C5 fopofnmrofororornpororororoforofoforocofororoforofocororororo E ft H H H H lAininioinininintAinmininintninioinininininininioinininininioH «Hp n H rH H H niH m r n m H m H H r H o?H f o r o H n H rH o m r o m r o r o n n m r o m m r n r n r o r o r o r o f n r o r n h t ft ft ^ . E h E h „ w o K h h t t ft ft ft 139 < < E h o JBO(< > p H H H hwo Z Z Z O &<„ _ P CO < cu E p H h B CO H H r t i H h « C« Oa B H C >O O H C0 O 5 E * H 5 0 D 3 C O ft Z CO H X J p H H H H 1533 [60 [60 [60 [60 [60 [60 [60 [60 [60 [60 [60 [60 KSSSLIVKRLKSSSLIVKRL [60 KSSSLIVKRL KSSSLIVKRLKSSSLIVKRL [60 KSSSLIVKRL [60 KSSSLIVKRL [60 KSSSLIVKRL KSSSLIVKRL KSSSLIVKRL [60 KSSSLIVKRLKSSSLIVKRL [60 KSSSLIVKRL [60 KSSSLIVKRLKSSSLIVKRL [60 KSSSLIVKRL KSSSLIVKRLKSSSLIVKRL [60 KSSSLIVKRL KSSSLIVKRLKSSSLIVKRL [60 KSSSLIVKRL [60 NSSSLIVKRL [60 KfSSLIVKRL [60 KfSSLIVKRLKSSSLIVKRL [60 KSSSLIVKRL [60 KSSSLIVKRL [60 KfSSLIVKRL [60 ILLENVGYDN ILLENVCYDN ILLENVGYDN ILLENVGYDN ILLENVGYEN ILLENVGYDN ILLENVGYDN ILLENLGfiDN LAHDHGLNRS LAHDHGLNRS ILLENVGYDN LAHDHGLNRS ILLENLGfiDN LAHDHGLNRS y PLIFQEYIYA LAHDHGLNRS PLIFQEYIYA PLIFQEYIYA LAHDHGLNRS PLLFQEYIYA LAHDHGLNRS ILLENVGYDN PLIFQEYIY PLIFQEYIYA DKSRQHDFLY DKSRQHDFLY PLIFQEYIYA DKSRQHDFLY DKSGQHDFLY DKSRQPDFLY PLIFQEYIYA LAHDHGLNKS ILLENLGYDN HKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVRYDN 10 10 20 30 40 50 60 MEEFQGYLEL MEEFQGYLEL MEEFQGYLEL MEEfiQGYLEL MEEFQGYLEL MEEFQGYLEL DKSRQHDFLY MEEFQGYLEL MEEFQGYLEL DKSRQHDFLY MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN (Taxon abbreviations as in Table 2.1). Note the underlined lower case letters. DIC DIS FORSIN MEEFQGYLELFOT DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS MEEFQGYLEL ILLENVGYDN PPS DQSRQHDFLY PLIFQEYIYA LAHDHGLNRS MEEFQGYLEL ILLENVGYDN DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRSSYS ILLENVGYDN MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRSHVI ILLENVGYDN HVE MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN SHA MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRSDIR ILLENVGYDN MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN OST LFOLOR DKSRQHDFLY LAHDHGLNRS MEEfiQGYLEL PLIFQEYIYfi EXBRHO MEEFQGYLEL DKSRQHYFLY PLIFQEYIYA LAHDHGLNRS MEKFQGYLEL ILLENVGYDN DKSRQYDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN ALT LAHDHGLNRS DKSRQHDFLY MEEfiQGYLEL PLIFQEYIYjf ILLENLGfiDN CSP MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN PAP MKEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRSDOT ILLENVGYDN CSI MEEFQGYLELLSI DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN EUS NOA MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN MYT MEEFQGYLEL DKSRQHYFLY Appendix 2.2. Amino acid sequences translated from the matK gene sequences MATMAI MEEFQGYLEL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS MEEFQGYLELMOL DKSRQHDFLY PLIFQEYIYA LAHDHGLNRS ILLENVGYDN MEEFQGYLEL DQSRQHDFLY PLIFQEYIYA LAHDHGLNRS NEO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 [120 120 [ 120 [ VSSLEGKEIV GSSLEGKEIV VSSLE£KEIV VSSLErKEIV VSSLE£KEIV VSSLEGKERV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV VSSLEGKEIV IVEIPFSLRL IVEIPFfiLRL IVEIPFSLRL IVEIPFSLRL IVEIPFSLRL IVEIPFjaLRL IVEIPFjaLRL IVEIPFSLRS IVEIPFSLRS IVEIPFSLRS IVEIPFSLRS IVEIPFSLRL IVEIPFSLRS IVEIPFSLRL IVEIPFSLRS IVEIPFSLRS IVEIPFSLRS IVEIPFSLRS IVEIPFSLRL IVEIPFSLRL IVEIPFSLRL IVEIPFSLRL IVEIPFSLRL SQMISEGFAV SQMISQGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQMISEGFAV SQIISEGFAV IFLGYNKNLY PFLGHNKNLY PFLGHNNNLYSQIISEGFAI PFLGHNKNLYSQMISEGFAIIVEIPFSLRL TFLGLNKNLYSQMISEGFAIIVEIPFSLQL PFLGHNKNLYSQMISEGFAV PFFGHNKNLYFQMISEGFAIIVEIPFSLRL PFLGHNKNLF PFLGHNKNLY PFLGHNKNLY PFLGHNKNLY PFLGHNKNLYSQMISEGFAV PFLGHNNNLYSQIISEGFAIIVEIPFSLRL PFLGHNKNLF PFLGHNKNLYSQMISEGFAV PFLGHNKNLY PFLGHNKNLY PFLGHNKNLYSQMISEGFAV PFLGHNKNLY PFLGHNKNLYSQMISEGFAV IISANDSNQYPFLGHNKNLY IISANDSNQNPFLGHNKNLY IISVNDSNQN IISANDSNQNPFLGHNKNLYSQMISEGFAVIISVNDFNQN IISANDSNQNPFLGHNKNLY IISVNDSNQN IISVNDFNQN IISVNDSNQNPFLGHNKNLYSQMISEGFAIIVEIPFSLRLIISANDSNQNPFLGHNKNLYSQMISEGFAVIISANDSNQN IISVNDSNQN IISVNDSNQN IISANDSNQN IISANDSNQN IISANDSNQNPFLGHNKNLYSQMISEGFAVIISANDSNQNPFLGHNKNLY IISANDSNQNPFLGHNKNLYSQMISEGFAVIVEIPFSLRLIISANDSNQN IISANDSNQN IISANDSNQN IISANDSNQN IISANDSNQN IISVNDSNQN IISANDSNQN IISANDSNQN IISANDSNQN NISANDSNQN 70 70 80 90 100 110 120 ITRMYQQNHL ITRMYQQ-NL ITRMYQQNQL ITRLYQQNQL ITRMYLQNHL ITRMYQQNHL ITRMYQQNHL ITRMYQQNQL ITRMYQQNQL ITRMYQQNHL DIS Appendix Continued2.2 ALTITRMYQQNRL LORITRMYQQNRLEXBITRMYQQNHLSISANDSNQDRHOITRMYQQNHL FORITRMYQQNQLSIN PPSSHAITRMYQQNHL ITRMYQQNHL LFOITRMYQQNRL MYTIIRMYQQNHLCSICSPITRMYQQNHL ITRMYQQNHL HVE ITRMYQQNHL DIRITRMYQQNHLDOT HVI DIC PAPSYS ITRMYQQNHL LSIMATITRMYQQNHL ITRMYQQNHL FOT MAI EUS ITRMYQQNQL NEOITRMYQQNQLNOA OST MOLITRMYQQNQL Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [178 [180 [178 [178 [178 [178 DASSLDLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF [178 DASSLDLLRF DASSLHLLRF DASSLHLLRF [178 DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF [178 DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRFDASSLHLLRF [178 DASSLHLLRF [178 DASCLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRF DASSLHLLRFDASSLHLLRF [178 LVQTLRYWVK LVQTLRYWVK LVQTLRYWVK LVQTLRYWVK LVQVLRYWVK LVQTLRYWVK LIPHPIHLEI LIPHPIHLEI LIPHPIHLEIWDQTLRYWVK LIPHPIHLEILVQTLRYWVK LIPHPIHLEILVQTLRYWVK LIPHPIHLEILVQTLRYWGK LIPHPSHLEILVQTLRYWVK LIPHPIHLEILVQTLRYWVK LIPHPIHLEILVQTLRYWVK LIPHPTHLEILVQTLRYWVK LIPHPIHLEILVQTLRYWVK LIPHPIHLEI LIPHPIHLEILVQTLRYWVK LIPHPTHLEILVQILRYWVK LIPHPIHLEIWQTLRYWVKLIPHPIHLEI LIPHPIHLEI FLHLNYVSDI FLHLNYVSDI FLHLNYVSDILIPHPIHLEILVQTLRYWLK FLHLNYVSDI FLHLNYVSDI FLHLNYVSDI FLHLNYVSDI FLHLNYVSDILIPHPIHLEILVQTLRYWVK FLHLNYVSDI LLHLNYVSDILIPYPIHLEILVQTLRYWVK FLHLNYVSDI LLHLNYVSDI VLHLNYMSDI HSIFPFLEDK HSPFPFLEDKLLHLNYVSDI HSIFPFLEDK HSIFPFLEDRFLHLNSVSDI HAIFPFLEDN HSIFPFLEDKLLHLNYVSDILIPHPIHLEILVQTLRYWVKHSIFPFLEDKLLHLNYVSDI HSIFPFLEDK HSIFPFLEDK HSIFPFLEDKFLHLNYVSDILIPHPIHLEILVQTLRYWVKHSIFPFLEDKFLHLNYVSDILIPHPIHLEILVQTLRYWVKHSIFPFFEDK HSIFPFFEAK HSIFPFLEDK HSIFPFLEDKFLHLNYVSDILIPHPIHLEILVQTLRYWVK HSVFPFPEDKFLHLNYVSDILIPHPIHLEILVQTLRYWVK HSVFPFLEDK HSIFPFLEDK HSIFPFLEDKFLHLNYVSDI HSIFPFLEDKFLHLNYVSDILIPHPIHLEILVQTLRYWVK HSIFPFFEDKFLHLNYVSDILIPHPIHLEI HSIFPFLEDK HSLFPFLEDKFLHLNYVSDI HSIFPFLEDK HSIFPFLEDK HSIFPFLEDKFLHLNYVSDILIPHPIHLEILVQTLRYWVK HSVFPFLEDKFLHLNYVSDILIPHPIHLEILVQTLRYWVK 130 130 140 150 160 170 180 QSH—NLRSI QSH—NLRSI QSH—NFRSI QSH— NLRSI QSH—NLRSIHSIFPFLEDKFLHLNYVSDIQSH— NLRSI QSH— NLRSI QSH—NLRSI QSH—NLRSI QSH— NLRSI QSH—NLRSI QSH— NLRSI QSH— NLRSI HAIFPFLEDK FLHLNYVSDI QSHTHNLRSI QSH— NLRSI KSH— NLRSI icSH— NLRSI SYS SHA QSH— NLRSI DIR HVE QSH— NLRSI DIC OST DIS QSH— NLRSI PAP RHOMYT CSI QSH— NLRSI DOTHVI QSH— NLRSI QSH— NLRSI NOA QSH— NLRSI EXB MAT FOR SINQSH—NLRSIPPS QSH— NLRSI CSP EUS MAI QSH— NLRSI LSI FOT NEO QSH— NLRSI Appendix 2.2 Continued ALT LFO LOR JtSH— NLRSI MOL N) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [238 [238 [238 [240 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [238 [235 [238 [238 [238 SSYLRSTSSG [238 SSYLRSTSSG [238 SSYLRSTSSG SYHLRSTSSG [237 SSYLRSTSSG SSHLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQ ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQ ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFVFLRNQ EYIFLFFRNQSSYLRSTSSG ESIFVFLRNQSSHLRSTSSG EYIFLFFRNQSSYLRSTSSG ESIFFFFCNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSGESIFLFFRNQSSYLRSTSSG ESIFLFFRNQ ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQ ESIFLFFRNQSSYLRSTSSG ESIFLFFRNQSSYLRSTSSG ESIFVFLRNQ SSHLRSTSSG [237 ESyFVFLRNQ ESyFVFLRNQ SSHLRSTSSG ESvFVFLRNQ SSHLRSTSSG FLYNSHVCEY FLYNSHVCEY F L Y N S H V y E Y FLYNSHVCEY FLYNSHVCEY FLYNSHVCEY FSKRNQRLFLFLYNSHVCEY FSKRNQRLFLFLYNSHVCEY FSKRNQRLFLFLYNSHVCEY FSKRNQRLFLFLYNSHVCEY FSKRNQRLFL FSKRNQRLFLFLYNSHVCEY FSKRNQRLFLFLYNSHVCDYESIFVFLRNQSSYLRSTSSG FSKRNQRLFL FLYNSHVyEY FSKRNQRLFL FSKRNQRLFL FLYNSHVyEY FSKRNQRLFLFLYNSHVCEY FSKRNQRLFLFLYNSHVCEY L-KRNQRLFLFLYNSHVCEY L-KRNQRLFL ; Y --- LITPKKSISI LITPKKSISI LITPKKSISI LITPKKSISIFSKRNKRLFLFLYNSHVCEY LITPKKSISIFSKRNQRLFLFLYNSHVCEYLITPKKSISIFSKRNKRLFLFLYNSHVCEY LITPKKSISIFSKRNKRLFL LITPKKSISIFSKRNKRLFLFLYNSHVCEY LITPKKSISI LITPKKSISI LITPKKSISI LITPKKPISI LITPKKPISI l i n p k k s i fl y i n p k k s i fLINPKKSIFy y LITPKKSISI LITPKKSISIFSKRNQRLFL LITPKKSISI 190 190 200 210 220 230 240 FFYEYYNWNSLITPKKSISIFSKRNQRLFLFLYNSHVCEY FFYEYHNWNS FFYEYHNWNS FFYEYYNWNS FFYEYYNWNS FFYEYYNWNSLITPKKSISIFSKRNKRLFSFLYNSHVCEY FFYEYHNWNSLITPKKSISLFSKGNQRLFLFLYNSHVCEY FFYEYHNWNSLITPKKSISLFSKRNQRLFLFLYNSHVCEY FFYEYYNWNS FFYEYYNWNS FFYEYHNWNSLITPKKSISIFSKRNQRLFLFLYNSHVCEY FLYEYHNWNS FLYEYRNWNS FFYEYYNWNSLITPKKSISLFSKRNQRLFLFLYNSHVCEY FFYEYYNWNS FLYEYRNWNS FLYEYRNWNS FFYEYRNWNSLITPKK FFYEYHNWNS FFYEYYNWNS FFYEYYNWNSLITPKKSISIFSKRNQRLFLFLYNSHVCEY LFYEYYNWNSLITPKKSISIFSKRNQRLFLFLYNSHVCEY HVI FFYEYYNWNS SHA DIR DOT HVE DIC OST FOT CSI LSI LFO DIS FOR SIN PAP EUS PPS FFYEYYNWNS SYS NEOFFYEYHNWNS NOAFFYEYHNWNS EXB RHOFLYEYRNWNS MYT CSP MAT MOLFFYEYHNWNSLITPKKSIPIFSKRNQRLFL MAIFFYEYYNWNS Appendix 2.2 Continued ALT LOR CO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [298 [295 [297 [297 [298 [298 [297 [298 [298 [298 [298 TPLLMNKWKY TPTLMNKWKY LPLLMNKWKY [298 TPLLMNKWKY TPLLMNKWKY [298 QGKSILASKGTPLLMNKWKY QGKSILASKGTPLLMNKWKY QGKSILASKGTPLLMNKWKY QGKAILASKG QGKSILASKG QGKSILASKG TPLLMNKWKYQGKSLLASKG [298 QGKSILASKGTPLLMNKWKY QGKSILASKG TPLLMNKWKY [298 QGKSILASKG TPLLMNKWRY [298 QGKSILASKGTPLLMNKWKY QGKAILASKG APLLMNKWKY [298 QGKSILASKG TPLLMNKWKY [300 QGKSILASKGTPLLMNKWKY QGKSILASKGTPLLMNKWKY QGKSILASKG TPLLMNKWKY [298 QGKSILASKG TPLLMNKWKYQGKSILASKG [298 TPLLMNKWKY [298 KDPFVHYVRY KDPFMHYVRY KDPFMHYVRYQGKSILASKGTPTLMNKWKYKDPFMHYVRYQGKSILASKG KDPFVHYVRY QGKSILASKG TPLLMNKWKY [298 KDPFVHYVRY KDPFVHYVRY KDPFMHYVRY KDPFMHYVRY QGKSILASKG TPLLMNKWKY [298 KDPFMHYVRY QGKSILASKG TPLLMNKWKY [298 KDPFMHYVRY KDPFMHYVRY KDPFMHYVRYQGKSILASKG NDFQASPWLF NDFQAILWLF NDFQAILWLFKDPFMHYVRY NDFQAILWLF NDFQAILWLF NDFQATLWLF NDFQAILWLFKDPFMHYVRY NDFQAIPWLF NDFQAILWLF KDPFVHYVRYNDFRAILWLFKDPFMHYVRY QGKSILASKGNDFQAILWLFKDPFMHYVRY TPLLMNKWKY [298 NDFRPS-YGV NDFQAILWLFKDPFMHYVRY NDFQAILWLF KDPFMHYVRY QGKSILASKGNDFQAILWLFKDPFMHYVRY TPLLMNKWKY [298 KIKHLVEAFA KIKHFVEVFV KIKHLVEVFV KIKHLVEVWV KIKHLVEVFV NDFQAILWLF KDPFMHYVRY QGKSILASKG TPLLMNKWKY [298 KIKHLVEVFVNDFQAILWLF KIKHLVEVFV KIKHLVEVFV KIKHLVEVFVNDFQAILWLF KIKNLVEVFVNDFQAILWLFKDPFMHYVRY KIKNLVEVFVNDFQAILWLF KIKHLVEVFV-DFQAILWLF KIKHLVEVFVNDFQAILWLFKDHFMHYVRY KIKHLVEVFVNDFQAILWLF KIKHLVEVFV NDFQAILWLFKIKHLVEVFVNDFQAILWLF KDPFVHYVRY QGKSILASKGKIKHLVEVFVNDFQAILWLFKDTFVHYVRY TPLLMNKWKY [298 KIKHLVEVFVNDFQPILWLFKDPFMHYVRY KIKHLVEVFVNDFQAILWLFKDPFMHYVRY KIKHLVEVFV NDFQAILWLF KDPFVHYVRY QGKSILASKG TPLLMNKWKY [298 250 250 260 270 280 290 300 ALLERIYFYG ALLERIYFYG ALLERIYFYGKIKHLVEVFA ALLERIYFYGKIKHLVEVFA ALLERIYFYGKIKHLVEAFA ALLERIYFYGKIKHLIEVFA ALLERIYFYGKIKHLVEVFA ALLERIYFYG ALLERIFFYG ALLERIYFYG ALLERIYFYG ALLERIYFYG ALLERIFFYGKIKHLVEVFV ALLERIYFYG ALLERIYFYG ALLERIYFYG ALLERIYFYG ALLERIYFYG ALLERIYFYG ALLERIYFYG ALLERICFYG ALLERIYFYG ALLERIYFYG ALLERIYFYGKIKNLVEVFV ALLERIYFYGKIKNLVEVFVALLERIYFYG ALLERIYFYGKIKHLVEVFV DIS EXB FOR SIN CSI RHO MYT EUS FOT PPS SHA SYS HVE DIC ALLERIYFYG CSP LFO LOR Appendix 2.2 Continued ALT MAI MOL PAPDIR ALLERIYFYG DOT HVI NOA LSI MAT OST NEO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [358 [358 [355 [357 [357 [358 [358 [358 [358 [358 [358 [358 [358 [358 [358 [358 [358 [358 [358 [357 [358 ENSFIIDNAI ENSFIIDNAI ENSFIIDNAI ENSFIIDNAI ENSFIIDNDI ENLFIIDNAI ENSFIIDNAI [358 ENSFIIDNAI ANSFLIDNAI NPSWRSQMLENSFLIDNAI NPSWRSQMLENSFIIDNAI NPSWRnQML NPSWRSQML-NSFIIDNDINPSWRSQML NPSWRSQMLENSFIIDNAI NPSWRSQMLENSFIIDNAI NPSWRSQMLENSFIIDNAI NPSWRSQMLNPSWRSQMLANSFLIDNAI ENSFIIDNAI [360 NPSWRSQMLENSFIIDNAI NPSWRSQML ENSFIIDNAINPSWRSQMLENLFIIDNAI [358 NPSWRSQMLENSFIIDNAI NPSWRSQMLENSFIIDNAINPSWRSQMLENSFIIDNAINPSWRSQMLENSFIIDNAI FLGYLSSVGL FLGYLSSVGL FLGYLSSVGL NPSWRnQML ENSFIIDNAIFLGYISSVGL [358 FLGYLSSVGLNPSWRSQML FLGYLSSVGL FLGYLSSVGL FLGYLSSVGL FLGYLSSVGL FLGYLSSVGL FLGYLSSVGLNPSWRSQMLFLGYLSSVGL NPSW-SQML ENLFIIDNAIFVGNLSSVGL [358 d S-D -- INQLSNHSF^ INQLSNHSL INQLSNHSLGFLGYLSSVGL INQLSNHSLD INQLSNHSFDFLGYLSSVGL INQLSNHSFD INQLSNHSFD FLGYLSSVGL NPSWRnQML ENSFIIDNAI [358 INQLSNHSFDFLGYLSSVGLNPSWRSQMLINQLSNHSFD INQLSNHSFDFLGYLSSVGL INQLSNYSFDFLGYLSSVGL INQLSNHSFD FLGYLSSVGL NPSWRSQML ENSFIIDNAI [358 INQLSNHSFDFLGYLSSVGL INQLSNHSFD INQLSNHSFDFLGYLSSVGL INQLSNHSFDFLGYLSSVGLNPSWRSQMLINQLSNHSFD INQLSNHSFD INQLSNHSFDFLGYLSSVGL LNQLSNHSFDFLGYLSSVGLNPSWRSQML INQLSNHSFDFLGYLSSVGL INQLSNHSFD FLGYLSSVGL NPSWRSQML ENSFIIDNAIINQLSNHSFDFLGYLSSVGLNPSWRSQIL [358 YVWSQPGRIYINQLSNHSLD YVWSQPVRIY YVWSQPVRIY YVWSQPVRIY YVWSQPGRIY YVWSQPGRIH YVWSQPGRVY YVWSQPGRIYINQLSNHSFD YVWSQPGRIY YVWSQPGRIYRNQLSNHSFD YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIY YVWSQPGRIYINQL YVWSQPGRIY YVW-QPGRIYINQLSNHSFD YVWSQPGRIY 310 310 320 330 340 350 360 YLVNFWQCHF YLVNFWQCHF YLVNFWQCHF YLVNFWQCHF YLVNFWQCHF YLVNFWQCHF YLVNFWQCHF YLVNFWQCHF DIS LORYLVNFWQCHFEXBYLVYFWQCHFRHO MYTYLVNFWQCYFCSI LSI YLVNFWQCHF LFOYLVNFWQCHF FORYLVNF-QCHFSIN YLVNFWQCHF ALTYLVNFWQCHF CSPYLVNFWQCHFMATYLVNFWQCHFMAIEUS YLVNFWQCNF FOT YLVNFWQCHF YVWSQPGRIY SHAYLVNFWQCHFPAPSYS YLVNFWQCHF YLVNFWQCHF YVWSQPGRIY MOLYLVNFWQCHF DIR DOT YLVNFWQCHF DIC YLVNFWQCHF YVWSQPGRIY PPS HVE YLVNLWQCHF Appendix Continued2.2 HVI NEO NOAYLVNFWQCHFYVWSQPGRIY OSTYLVNFWQCHFYVWSQPGRIYINQLSNHSFDFLGYLSSVGLNPSWRSQML Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [417 [418 [418 [418 [418 [420 [417 [418 [418 [418 [418 [418 RNLSHYHSGSRNLSHYHSGS [417 RNLSHYYSGS RNLSHYHSGSRNLSHYHSGS [418 RNLSHYHSGS [418 RNLSHYHSGS [418 RNLSHYHSGSRNLSHYHSGS [418 [418 RNLSHYHSGS RNLSHYHSGS -NLSHYHSGS [418 RNLSHYYSGS [418 KNLSHYHSGS RNLSHYYSGSRNLSHYYSGS [418 RNLSHYYSGS [418 RNLSHCHSGS RNLSHYHSGSRNLSHYHSGS [418 RNISHYHSGS [415 RNLSHYHSGS RNLSHYYSGS RNLSHYYSGS [418 RNLSHYHSGS RNLSHYHSGSRNLSHYHSGS [418 [418 RNLSHYHSGS [418 RNLSHYHSGS [418 DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDQFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIY DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC EIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFGRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC DIIDRFVRIC KPARADSSDS KPARADSSDYDIIDRFVRIC KPARADSSDS KPARADSSES FCNVLGHPISKPARADSSDS FCNVLGHPISKPARADSSDS FCNVLGHPISKPARADSSDS FCNVLGHPISKPARADSSDS FCNVLGHPIS FCNVLGHPISNPARADSSDS FCNVFGHPISKPARADSSDS FCNVFGHPISKPARADSSDS PLIGSLAKAKFCNVLGHPISKPARADSSDS PIIGSFAKAK PLIGSLAKAKFCNVLGHPISKPARSDSSDSPLIGSLAKAK PLIGSLAKAK PLIGSLAKAK PLIGSLAKAKFCNVLGHPISKPARADSSDS ALIGSLAKAR 370 370 380 390 400 410 420 KKFDIIVPII KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVLGHPIS KKFDIIVPII KKFDIIVPIIPLIGSLAKAKFCNGLGHPISKPARADSSDSKKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKLARADSSDS KKFDIIVPII KKFDIIVPII KKFDIIVPIIPLIGSLAKAKFCNVLGHPIS KKFDIIVPII KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVSGHPISKPVRADSSDSKKFDIIVPII KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAK KKFDIIVPIIPLIGSLAKAK KKFDIIVPIIPLIGSLAKAK KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS KKFDIIVPIIPLIGSLAKAKFCNVLGHPISKPARADSSDS SYS DOT RHO CSI LSI SIN KKFDIIVPII PLIGSLAKAK FCNVLGHPIS KPARADSSDS SHAKKFDIIVPIIPAP DIR HVI HVE NOAKKFDIIVPIIPLIGSLAKAKFCNVFGHPISKPARADSSDSDIC OST LOR EXB MAT MAI DIS MYT CSP EUS FOR FOT PPS NEOKKFDIIVPIIPLIGSLAKAKFCNVFGHPIS LFO Appendix 2.2 Continued ALTKKFDIIVPII MOL Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [478 [477 [477 [478 [478 [478 [478 [478 [478 [477 [477 VLSLIFPKAS [478 VLSLIFPKAS VLSLIFPKAS [478 VLSLIFPKAS [478 VLSLIFPKAS [478 VLSLIFPKAS VLSLIFPKASVLSLIFPKAS [478 [478 VLSLIFPKAS [478 VLSLIFPKAS [478 VLSLIFPKAS VLSLIFPKAS [478 ALSLIFPKAS [478 VLSLIVP-AS LEEFLTEEEQ LEEFLTEEEQ VLSLIFPKAS [480 LEEFLTEEEQ LEEFLTEEEQ LEEFLTEEEQ LEEFLTEEEQ LEEFLTEEEQ LEEFLTEEEQ VLSLIFPKAS [478 LEEFLTEEEQ LEEFLTEEEQ LEEFLTEEEQ LEEFLTEEEQ LEEFLTEEEQ VLSLIVP-AS [477 LEEFLTEEEQ LEEFLTEDEQ VLSLIFPKAS [475 LEEFLTEEEQ AFLKRFGSGL AFLKRLGSGL AFLKRLGSGLLEEFLTEEEQVLSLIFPKAS AFLKRFGSGL AFLKRLGSgL LEEFLMEEEQ VLSLIFPRAS AFLKRLGSGL AFLKRLGSGL AFLKRLGSGL AFLKRLGSGLLEEFLTEEEQVLSLIFPKASAFLKRLGSGL AFLKRLGLGL AFLKRLGSELLEEFFTEEEQVLSLIFPKASAFLKRLGLGL LEEFLTEEEQ VLSLIFPKAS [478 AFLKRLGSGL AFLKRFGSGL AFLKRLGLGL LEEFLTEEEQAFLKRLGSGL VLSLIFPKAS [478 AFLKR-GLGL AFLKRLGSgL AFLKRLGSgL AFLKRFGSGLLEEFLTEEEQ AFLKRLGLGL LEEFLTEEEQ VLSLIFPKASAFLKRLGSGLLEEFLTEEEQ [478 AFLKRLGSGL AFLKRLGSGLLEEFLTEEEQVLSLIFPKAS AFLKRLGSgL LEEFLTEEEQ VLSLIVP-AS AFLKRLGSGLLEEFLTEEEQVLSLIFPKAS AFLKRLGSGL LARKHKSTVR LARKHKSTVR LARKHKSTVR AFLKRLGSGLLARKHKSTVR LEEFLMEEEQ VLSLIFPRAS [477 LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSPVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHTSTVR LARKHKSTVR LARKHKSTVR LARKHTSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSTVR LARKHKSPVR LARKHKSPVR LARKHKSTVR LARKHKSTLR YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART YILRLSCART 430 430 440 450 460 470 480 SKKKSLYRIK SKKKSLYRIK SKKKSLYRIK SKKKSLYRIKYILRLSCART SKEKSLYRIK SKKKSLYRIKYILRLSCART SQKKSLYRIKYILRLSCART SKKKSLYRIKYILRLSCART SKKKSLYRIK SKKKSLYRIR SKP-SLYRIK SKKKSLYRIKYILRLSCART SKKKSLYRIK SKKKSLYRIK SKKKSLYRIK SIN SQKKSLYRIK YILRLSCART EUS FOR FOTSHASKKKSLYRIKYILRLSCART SKKKSLYRIK PAP PPS SKKKSLYRIK SYS SQKKSLYRIK HVI SKKKSLYRIK YILRLSCART DIS DIRSQKKSLYRIK RHO LSI HVE SKKKSLYRIK DIC SKKKSLYRIK DOT SQKKSLYRIK CSP SQKKSLYRIK LOR Appendix 2.2 Continued LFO MAI MAT MOL MYT EXB CSI NEOSKKKSLYRIKYILRLSCARTOSTSKKKSLYRIK ALT NOASKKKSLYRIK Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DLANHE. [504 IWYLDIVCIN 490 490 500 DIC STSRRLYRGR IWYFDIISINDLANHE. [505 OST STSRRLYRGR IWYFDIISINDLANHE. [505 SIN STSRRLYRGR IWYFDISSINDLANHE. SHA [504 PAP STSRRLYRGR SYS IWYFDIISINDLANHE. STSRRLYRGR IWYFDIISINDLANHE. STSRRLYRGR [505 IWYFDIISINDLANHE. [505 HVI [505 HVE STSRRLYRGR IWYFDIISINDLANYE. STSRRLYRGR IWYFDIISINDLANYE. [505 [505 EUS STSRRLYRGR FOR IWYFDIISINDLANHE. [505 STSRRLYRGR FOT IWYFDILSINDLANHE. PPS [505 STSRRLYRGR IWYFDIISINDLANHE. STSRRLYRGR IWYFDIISINDLANHE. [505 [505 DIR DOT STSRRLYRGR IGYLDIISINDLANHE. STSRRLYRGR IWYFDIISINDLANHE. [505 NEO [505 NOA STSRRFYRGR IWYFDIISISDLANHE. STSRRLYRGR IWYFDIISINDLANHE. [505 [505 DIS SSLRRLYRER IWYFDIICINDLANHE. [502 CSP STSQRLYRGR IWHFDIISINDLANQE. [505 LFO STSRRLYRGR IWYLDIICINDLANHE. [504 RHO STSRRLYRGR CSI STSQRLYRGR LSI IWHFDIISINDLANQE. MAT [505 FTSQRLYRGR MAI IWY-DIISINDLANQE. STSQRLYRGR IWYFDIISINDLANQE. [505 STSQRLYRGR MOL IWYFDIISINDLANQE. [505 [504 STSRRLYRGR IWYFDIISINDLANHE. [507 EXB STSRRLYRGR IWYLDIICINDLSNHE. [504 ALT STSRRLYRGR LOR IWYLDIICINDLANHE. [504 STSRRLYRGR ILYLDIICINDLANHE. [504 MYT STSRRLYRGR IWYFDIICINDLANHE. [505 Appendix 2.2 Continued Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER III PHYLOGENETIC RELATIONSHIPS OF THE HAMAMELIDACEAE INFERRED FROM SEQUENCES OF INTERNAL TRANSCRIBED SPACERS OF NUCLEAR RIBOSOMAL DNA INTRODUCTION Using sequence variation between homologous genes has become an increasingly routine method of molecular systematics to reconstruct the phylogenetic relationships of a taxon, although it is still controversial concerning whether the phylogenetic trees obtained via cladistic analysis are species trees or just gene trees (Doyle, 1993). Many studies have been conducted on phylogenies of organisms at various hierarchical levels, from populations to divisions, using single gene sequences (Chase et al., 1993). Most studies suggested similar relationships to those using other lines of evidence such as morphology and anatomy (Donoghue and Sanderson, 1992; Chase et al. 1993). As expected, discrepancies exist as well, indicating where future research should be concentrated. It has been well known that different genes evolve at different rates (Gaut et al., 1992). Thus, finding appropriate genes with meaningful substitution rates for assessing phylogenetic relationships at various levels has 149 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. been one of the frontiers in molecular systematics. It is often the first step for a thorough study of phylogenetic relationships of a group of organisms. One of the widely used segments of DNA sequence is the Internal Transcribed Spacer (ITS) region of 18S-26S nuclear ribosomal DNA (nrDNA) (Fig.3.1). This region includes three components: two spacers designated ITS-1 and ITS-2 and the 5.8S subunit intercalated between the two spacers. The ITS region is part of the transcriptional unit of nr DNA, but the spacers of the transcript are not incorporated into mature ribosomes. However, the spacers probably function in the maturation of nrRNAs (Musters et al., 1990; van der Sande et al., 1992; van Nues et al., 1994). Therefore, ITS-1 and ITS-2 are under some evolutionary constraints in structure and sequence (Baldwin et al., 1995). Lower length variation and a relatively high base substitution rate indicate that DNA sequences of these spacers might be readily alignable across closely related taxa, yet be sufficiently variable to allow resolution of lower-level phylogenetic questions in angiosperms (Sytsma and Schaal, 1985, 1990; Jorgensen and Cluster, 1988; Yokota et al., 1989; Kim and Mabry, 1991; Baldwin, 1992). There are three advantages for using the ITS region in phylogenetic analyses of angiosperms. First, along with other components of the nrDNA multigene family, the ITS region is highly repeated in the plant nuclear genome. The 150 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. entire nrDNA repeat unit is present in up to many thousands of copies at a chromosomal locus or at multiple loci (Rogers and Bendich, 1987; Hamby and Zimmer, 1992). The high copy number promotes detection amplification, cloning and sequencing of nrDNA (Baldwin et al., 1995). Second, albeit many-copied, the sequences of the spacers do not vary much due to intraspecific homogenization by virtue of concerted evolution (including unequal crossing over and gene conversion) (Zimmer et al., 1980; Dover, 1982; Hillis et al., 1991). Homogenization can even affect sequences of nrDNA of non-homologous chromosomes (Arnheim et al., 1980; Wendel et al., 1995). This property systematically promotes accurate reconstruction of species relationships from the sequences (Hamby and Zimmer, 1992; Sanderson and Doyle, 1992). Third, due to the small size of the ITS region (<700 bp in angiosperms) and the presence of highly conserved sequence flanking each of the two spacers, it is relatively easy to amplify the region using universal eukaryotic primers designed by White et al. (1990) (Fig.3.1). Although this ITS region is mostly used for phylogenetic study of interspecific relationships, many studies have also suggested its informativeness for resolving intergeneric relationships of closely related taxa (Baldwin et al., 1995; Bogler and Simpson, 1996; Downie and Katz- Downie, 1996; Schilling and Panero, 1996; Soltis et al., 1996; Kron and King, 1996). 151 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. As mentioned in the Introduction Chapter, the relationships among the genera of the Hamamelidaceae have been controversial for more than a century (Figs.i.2-i.6). The objective of the research reported here, therefore, is to study the phylogenetic relationships among the hamamelidaceous genera using DNA sequences of the ITS region. MATERIALS AND METHODS Plant Material Material of 32 species was used in this study (Table 3.1), representing 28 of the total 31 genera in the Hamamelidaceae. Unfortunately, three genera were not available for analysis: Chunia, Semiliouidambar. and Embolanthera. Vouchers for the sampled species are deposited in the Hodgdon Herbarium of the University of New Hampshire (NHA). Sequences of both ITS-1 and ITS-2 have been submitted to the GenBank and the accession numbers are given in Table 3.1. The aligned ITS sequences are shown in Appendix 3.1. Molecular Techniques Total genomic DNAs were extracted from fresh or silica gel dried leaves or buds following the protocol of Doyle and Doyle (1987). Polymerase Chain Reaction11*1 (PCR) was conducted in 0.2 ml thin-walled microcentrifuge tubes. Each 50 ^1 reaction included 50-100ng DNA, 5pl of Tag extender buffer 152 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Stratagene, CA), 4 fjl of 2.5 jjm dNTP, 1-4 units of Tag extender (Stratagene, La Jolla, CA) and Tag polymerase (Promega, WI), 1 gl of 20pm primers, and an appropriate amount of UV-treated deionized water. The PCR reaction was conducted with a three-minute 'hot- start' (D'aquila et al., 1991). Tag polymerase and Tag extender were added into the reaction tube at the end of the hot-start, when the temperature was stabilized at 80°C. Then, 30 cycles of PCR reaction were started following the thermocycler program of 105 seconds at 45°C, 115 seconds at 72°C and 30 seconds at 94°C. The last cycle ended at 72°C for 15 minutes of extension. The PCR primers were universal ITS4 and ITS5 of White et al. (1990). The PCR products were loaded on 0.8% LMP (Low Melting Point) agarose gel along with 0X174 Haelll DNA size markers and separated for 2-3 hours at 40 volts in 0.5X TBE buffer. The band identified by comparison to the markers was then excised from the gel, liquified at 650C, and treated with agarase (Sigma Chemical Co., St. Louis, MO) for 30 minutes at 37°C. This gel-purified PCR product was used directly as a sequencing template. Sequencing reactions were carried out using Cycle Sequencing Kit, FS (catalog#: 402119) and following the manufacturer’s protocols (Applied Biosystems, Foster City, CA). The sequencing primers were ITS2, ITS3, ITS4 and ITS5 of White et al. (1990). The cycle sequencing products were then 153 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. separated on 6% polyacrylamide gel using an Automated Sequencer 373A (Applied Biosystems, Foster City, CA) in the Sequencing Facility Center of the University of New Hampshire (UNH). For Tetrathyrium and Maingaya, 0.8% DMSO (Dimethylsulfoxide) was added to both the PCR and cycle sequencing reactions. For Liquidambar formpsana, when the PCR product was used directly as a template for sequencing, I could not acquire reliable sequences. Therefore, the standard pBiuescript gene clone technique (Promega, Madison, WI) was used to clone the PCR product of the ITS region. The PCR primers are ITS4 and ITS5 of White et al. (1990), with BamHl and EcoRl restriction sites added to their 5' ends respectively. Cycle sequencing procedures sure as mentioned above. Sequence from a single clone of the ITS region was used in this analysis. The chromatograms were analyzed using the SEOED program (Applied Biosystems, Foster City, CA). Also, in order to assure correct base-calling, I overlapped sequences generated from adjacent primers of either the same or opposite directions. The boundaries of ITS-1 and ITS-2 were determined by comparing sequences of the 3 ’ 18S, 5.8S and the 5 1 end of the 26S ribosomal genes of Canella winterana (GenBank accession number L03844) 154 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sequence Alignment. The ITS sequences within a genus and among closely related genera were easily aligned by sight. However, ITS sequences among genera of different subfamilies were not readily alignable, thus resulting in some ambiguous regions. According to some authors, these ambiguous sites, as identified by alignability by eye, were eliminated from the data matrix before a phylogenetic analysis was conducted (Downie and Katz-Downie, 1996). This "culled" method tends to create clades where internal relationships are not well resolved (Wheeler et al., 1995; Soltis et al., 1996a). Wheeler et al. (1995) proposed a method called the "Elision" approach for the alignment of ambiguous DNA sequences. Alignments change with gap penalty (GP) and gap length penalty (GLP), thus, setting different GPs and GLPs produces various alignments. The resulting aligned sequences using different penalties are then strung together, giving rise to a combined, grand data set. Elision is essentially a fine character weighting scheme based on character consistency in the sets of alignments (Wheeler et al., 1995). Another way of dealing with alignments with ambiguous sites is to select a so-called optimal alignment. This approach involves consistency indices of the aligned sequences and the phylogenetic trees produced based on the data. Bogler and Simpson (1996) used this "optimality" method to analyze the phylogenetic relationships of the 155 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Agavaceae. Many indices can be used to assess the optimality of sequence alignment, including the number of trees generated, the Consistency Index (Cl), Retention Index (RI), and Rescaled Consistency Index (RC). Bogler and Simpson (1996), however, implied that the higher the RC was, the better the alignment. This seems to be reasonable because the RC index excludes characters that do little to the 'fit' of the tree but inflate the Cl (Wiley et al., 1991). In this study, both the Elision and Optimality methods were employed to generate data matrix of the ITS sequences of the Hamamelidaceae. For the Optimality method, the alignment that created trees with the highest RC index was considered as optimal for both ITS-1 and ITS-2. When two or more alignments having equally highest RC were encountered, the one with the highest Cl was chosen. For the Elision method the procedures were carried out following Wheeler et al. (1995), and Soltis et al. (1996a). Individual alignments were conducted using the CLUSTAL option of the MEGALIGN program of DNA* software package (DNA* Inc., CA). Phylogenetic Analysis The resulting data matrices from both alignments (Optimality and Elision) were imported into the beta-test version of PAUP* 4.0d57 computer program, written by David L. Swofford (1997) at the Smithsonian Institution, for phylogenetic analyses. 156 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Indels were treated as missing data because this coding strategy retains information about substitutions that occur in other taxa in the indel region. However, it does not convey the information regarding the evolutionary event involved in the insertion or deletion, and thus can introduce ambiguities (Platnick et al., 1991; Wojciechowski et al., 1993). To compensate for the disadvantage, attention was paid to the evolutionary significance of some unambiguous indels. All characters and their states were equally weighted in both parsimony and distance analyses. Sequence divergence was analyzed using the pairwise difference obtained from PAUP*. Due to the size of the data set and the limitation of computer memory, the heuristic search option was used to find the shortest trees with TBR (Tree Bisection and Reconnection) branch swapping, mulpars on, and steepest descent off. In the parsimony analysis, Altinoia was used as the outgroup because: 1) both sequence divergence (Table 3.2) and cluster analysis of the ITS sequences (UPGMA, Fig.3.2) indicated that Altinoia. along with Liquidambar. was rather distinct from other Hamamelids; 2) the fossil record has shown that Altinoia is the most ancient member in the Hamamelidaceae (Zhang and Lu, 1995); and 3) both morphology and matK gene phylogenies (See Chapter I and II) suggested that Altinoia. together with Liquidambar. is the basal taxon 157 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in the Hamamelidaceae. Both bootstrap and decay analyses were conducted using the PAUP program to test the relative strength of putative clades (Felsenstein, 1985; Bremer, 1988; Donoghue et al., 1992). MacClade 3.03 (Maddison and Maddison, 1992) was used to trace unambiguous changes along branches and to compare competing hypotheses concerning the relationships. The skewness test (Huelsenbeck, 1991) was implemented using the Random tree option of the PAUP*, and 10,000 random trees were examined to evaluate the phylogenetic information contained in ITS data matrix. Also, the permutation test (Faith and Cranston, 1991; Plunkett et al., 1997) was performed using the Permutation option of the PAUP* with 100 replicates and heuristic searches. RESULTS Sequence Alignment Among the seven pairs of alignment parameters, GPs/GLPs, including 8/8, 8/15, 10/20, 10/10, 15/8, 20/8, and 20/20, the pair of 8/8 resulted in the alignment of the highest RC index for ITS-1 (Table 3.3) and that of 20/20 produced the optimal alignment for ITS-2 (Table 3.4). 158 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■Sequence Characteristics The lengths (bp) of spacers 1 and 2 range from 236 to 277 and from 224 to 241 respectively in the Hamamelidaceae (Table 3.5), and the means are 267 bases for ITS-1 and 236 for ITS-2. Several genera have roughly equal length of ITS-1 and ITS-2 including Fortunearia. Eustioma. Molinadendron. and Sinowilsonia. GC content ranged from 58% to 67% with an average of 63% in ITS-1, and from 61% to 68% with an average of 65% in ITS-2. Statistical analysis (Chi-square test) showed that the differences of nucleotide composition in both ITS-1 and ITS-2 across taxa, and between the two spacers were not significant with P>0.9. Pair-wise sequence divergence in ITS-1 was generally higher than that in ITS-2, with averages of 18.1% and 16.5% respectively. Pair-wise divergence between the two genera of Liquidambaroideae and the other member of the Hamamelidaceae was higher than 30% for both spacers. The percentages of potentially informative sites were 52% and 51% for ITS-1 and ITS-2 respectively (Table 3.5). The alignment of the sequences required 45 and 31 indels in ITS-1 and ITS-2 respectively (Table 3.5). in both spacers, about 70% of the indels were a single base in length. There were three indels of 10 or more bases in the ITS region: two indels (12 and 26 bases) were in ITS-1, and one (10 bases) in ITS-2. The parsimony analysis using the Optimality data of ITS- 159 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 and ITS-2 yielded 90 shortest trees of 1041 steps, with a consistency index of 0.59. Figure 3.3 shows the strict consensus tree. Liquidambar is the first taxon to branch following the outgroup Altincria. Mytilaria. Exbucklandia. and Rhodoleia form a clade, while Disanthus is sister to the clade containing the sampled genera of the Hamamelidoideae. In the Hamamelidoideae, three major clades are recognized: the Corylopsis clade, Hamamelis clade, and the Trichocladus clade, among which the Corylopsis clade is basal. Both Corylopsis and Hamamelis clades are strongly supported, whereas the Trichocladus clade is weakly supported (Fig.3.3). The Corylopsis clade consists of two subclades, the first of which is comprised of three species of Corylopsis: the second includes Loropetalum, Tetrathyrium. Maingaya, and Matudaea. Both subclades receive strong support of a 100% bootstrap and a decay index of more than five steps. In the Trichocladus clade, four branches are recognized: Dicoryphe. Trichocladus. the Eustigmateae sensu Endress (1989b) (incl. Molinadendron). and the three Australian genera. However, in the 50% majority consensus tree, the five Southern Hemisphere (Gondwanaland) genera (Dicoryphe. Trichocladus. Ostrearia. Neostrearia. and Noahdendron) form a single clade (Fig.3.4). Among the three Australian genera, Neostrearia is basally sister to the latter two genera (Figs.3.3, 3.4). In Fig.3.3, Molinadendron and Sinowilsonia are bound 160 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with a bootstrap value of 84% and a decay index of three steps. In both Fig.3.3 and 3.4 the Hamamelis clade is essentially the Fothergilleae sensu Endress (1989c), plus Hamamelis. In Fig.3.3 there are four clades whose relationships are not well resolved, including Hamamelis, Fotherqilla, Parrotiopsis. and the well-supported Distylium group consisting of Distylium. Distyliopsis. Parrotia, Shaniodendron. and Sycopsis. However, in Fig.3.4 Fotherqilla is the basal clade, followed by the clades of Hamamelis species, Parrotiopsis. and the Distvlium group. Twelve most parsimonious trees of 6747 steps were generated based on the grand data set from the Elision alignments. The strict consensus tree (Fig.3.5, CI=0.578) is largely congruent with the phylogeny that was generated using the Optimality alignment data. However, the subtribe Dicoryphinae of Endress (1989c) forms a clade and so does Mytilaria in the Elision phylogeny. DISCUSSION Sequence Characteristics In angiosperms ITS-1 and ITS-2 are each less than 300 bases long: ITS-1 ranges from 187 to 298bp and ITS-2 from 187 to 252bp (Baldwin et al., 1995; Johnson and Soltis, 1995; Downie and Katz-Downie, 1996; Padgett, 1997). The length of 161 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the two spacers in the Hamamelidaceae falls within the above range, but ITS-1 sequences in most of the species sampled are in the high end of the range, except for the four genera (Eustigma, Fortunearia. Molinadendron. and Sinowilsonia) that share two unique deletions (12 and 26 bases). The ITS-2 in the Hamamelidaceae is relatively less variable in length, mostly in the range of 224 to 241, which agrees with a generalization that ITS-2 is more conservative than ITS-1 (Hershkovitz and Lewis, 1996; Liston et al., 1996). A permutation test for the data sets from both the Elision and the Optimality alignments resulted in a probability of 0.01, indicating that the ITS data are significantly phylogenetic, and supporting the appropriateness of both alignments. This is consistent with the skewness of tree length distribution (Table 3.5). Monophvly of the Altinaioideae When Altinoia was used as an outgroup, the first taxon following Altinoia was Liquidambar. indicating a close relationship between the two genera. In the UPGMA tree (Fig.3.2), Altinoia and Liquidambar formed a clade sister to a clade consisting of all the other taxa of the Hamamelidaceae. Additionally, there were some 80 substitutions between the Altingioideae and the other hamamelidaceous taxa (Fig.3.6). Therefore, this study strongly suggests that the Altingioideae are monophyletic. 162 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It has been proposed that the Altingioideae be raised to the family status (Melikian, 1973afb; Huang and Li, 1982; Wang, 1992; Takhtajan, 1997). However, recent molecular studies support inclusion of the Altingioideae in the Hamamelidaceae (Hoot and Crane, 1996; Hibsch-Jetter and Soltis, 1996; Hoot et al., 1997). Moreover, many morphologically diagnostic apomorphies found in the Altingioideae are also represented in other hamamelidaceous genera (Pan et al., 1990; Zhang and Lu, 1995). Thus, the Altingioideae are retained in the Hamamelidaceae as a distinct subfamily. Polvphvletic Exbucklandioideae sensu Endress The Exbucklandioideae sensu Endress (1989c) includes four genera, Chunia, Disanthus, Exbucklandia. and Mytilaria. The morphological synapomorphies of this subfamily are large, persistent stipules, palmate venation, and 4-6 ovules per carpel. However, the phylogenetic analysis based on the ITS DNA sequences provided a different picture of this taxon (Chunia was not available for this study). Exbucklandia was grouped with Rhodoleia, and Disanthus forming its own clade, While Mytilaria was either allied with the clade of Exbucklandia and Rhodoleia (Fig.3.3), or constituted its own clade. This indicates that the Exbucklandioideae sensu Endress (1989c) are not monophyletic. Furthermore, when the three genera (Disanthus, Exbucklandia. and Mytilaria) were 163 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. forced into a monophyletic group, 34 more steps were required. Disanthus and Mytilaria have been previously recognized as subfamilies Disanthoideae and Mytilarioideae, respectively (Harms, 1930; Chang, 1948, 1973, 1979). Mytilaria formed its own clade in the Elision phylogeny (Fig.3.5, but not in the Optimality phylogeny (Figs.3.3, 3.4). However, the alliance of this genus with the clade of Exbucklandia-Rhodoleia in the Optimality tree was not well-supported. Moreover, separating Mytilaria from the Exbucklandia-Rhodoleia group required only one more step, while putting Mytilaria into the clade of the Exbucklandia-Rhodoleia group entailed 13 more steps than the shortest length. Thus, isolation of Mytilaria from Exbucklandia and Rhodoleia is more parsimonious. Therefore, this study recognizes the isolated position of both Disanthus and Mytilaria in the Hamamelidaceae as individual subfamilies. Disanthus is the immediate sister taxon to the Hamamelidoideae in the phylogenetic trees (Figs.3.3, 3.4, 3.5), reminiscent of the conclusions from a previous morphological analysis (Hufford and Crane, 1989; Pan et al., 1991). However, the intermediate systematic position of Disanthus between the Exbucklandia-Rhodoleia clade and the Hamamelidoideae in the ITS-based tree did not provide evidence for the hypothesis that this genus is most primitive in the Hamamelidaceae, as has been suggested by some authors 164 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Reinsch, 1889; Tong, 1930; Cronquist, 1981; Pan et al., 1991; Takhtajan, 1997). Exbucklandia-Rhodoleia Clade Exbucklandia has been placed in its own subfamily (Harms, 1930; Chang, 1979), or in a subfamily that includes Chunia, Disanthus, and Mytilaria (Endress, 1989c). However, the present phylogenetic analysis strongly supports a close relationship of Exbucklandia with Rhodoleia. These two genera also share 22 unambiguous base substitutions (Fig.3.6). Interestingly, this pattern agrees with an earlier taxonomic treatment (Reinsch, 1889). Nevertheless, these two genera are rather different in both vegetative and reproductive structures, suggestive of an ancient divergence. Therefore, it seems reasonable to continue to treat them as belonging to individual subfamilies (Harms, 1930; Chang, 1973, 1979; Endress, 1989b, c). Monophyly of the Hamamelidoideae It has been widely recognized that the Hamamelidoideae are a monophyletic group (Bogle, 1968; Endress, 1989a, b, c; Hufford and Crane, 1989). The present analysis adds strong evidence to this hypothesis because the clade of this subfamily receives a 97% bootstrap value and a decay index of more than six steps, and is further supported by 17 unambiguous base substitutions (Fig.3.6). 165 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The intergeneric relationships within the Hamamelidoideae, however, have been contentious (Harms, 1930; Schulze-Menz, 1964; Endress, 1989b, c). The relationships among the five tribes recognized by Harms (1930) and more recently revised by Endress (1989c) and others, are discussed below. Corvlops ideae The concept of the tribe Corylopsideae has been revised several times in the past decades. Harms (1930) delimited Corylopsideae as including Corylopsis and Fortunearia. while Schulze-Menz (1964) added Sinowilsonia to this tribe. However, Endress (1989b,c) considered Corylopsideae as containing a single genus, Corylopsis. In the ITS-based phylogenies (Figs.3.3, 3.5), the three species of Corylopsis sampled formed a well-supported clade and were sister to the branch comprised of Loropetalum, Tetrathyrium. Mainaaya. and Matudaea, while Fortunearia and Sinowilsonia are phylogenetically distant from Corylopsis. Therefore, Corylopsis is not closely related to either Fortunearia or Sinowilsonia. Hamamelideae In Harms's system (1930), all of the genera in the Hamamelidoideae that have strap-shaped petals were placed in the tribe Hamamelideae and no further divisions were 166 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. attempted. Endress (1989c), however, recognized three subtribes in the Hamamelideae, including the Hamamelidinae, Loropetalinae, and the Dicoryphinae. The Hamamelidinae contained a single genus, Hamamelis: the Dicoryphinae included the five Southern Hemisphere (Gondwanaland) genera (Dicoryphe. Trichocladus. Ostrearia, Neostrearia. and Noahdendron): while the Loropetalinae contained the remaining four genera (Loropetalum. Embolanthera. Maingaya, and Tetrathyrium). The analysis of the ITS data placed the genera of the Hamamelideae into three different clades (Figs.3.3, 3.5), suggesting that the tribe is polyphyletic. Nevertheless, the Dicoryphinae was suggested to be monophyletic in both the majority consensus tree of the Optimality analysis and the strict consensus tree of the Elision analysis. A monophyletic origin agrees with the exclusive Gondwanaland distribution and the unique pattern of anther dehiscence of these five genera (Endress, 1989a, b, c). Eustiqmateae The delimitation of the Eustigmateae has expanded from monogeneric (Harms, 1930) to trigeneric (Endress, 1989c). The small, auriculate petals and greatly enlarged stigmatic surfaces earned Eustiama a tribal status (Harms, 1930). However, the similarities of Eustiama with Fortunearia in other morphological characters such as pedicellate, 167 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lenticellate fruits, and covering sepals led Endress (1989c) to the belief that Eustigmateae should include Fortunearia and Sinowilsonia. The ITS-based trees (Figs.3.3, 3.5) offer strong support for Endress's hypothesis, and further corroborates the implication (Endress, 1967) that Molinadendron and Sinowilsonia are closely related. Interestingly, these four genera share four unambiguous base substitutions (Fig.3.6) and two unique deletions totaling 36 bases (Li et al., 1997c). Combination of Distylieae and Fotherqilleae In the ITS-based phylogeny, one of the three major lineages in the Hamamelidoideae includes the clade of Hamamelis and the branch containing the Distylieae (Distylium. Distyliopsis. and Svcoosis) and Fothergilleae sensu Harms (1930) (Fotherqilla. Parrotia, and Parrotiopsis) (Figs.3.3, 3.5). Neither Distylieae nor Fothergilleae forms a monophyletic clade. Instead, a close but unresolved relationship is discerned among Sycopsis. Parrotia, and Shaniodendron. This agrees with the finding that Svcopsis and Parrotia are interfertile, giving rise to the hybrid taxon, XSvcoparrotia (Endress and Anliker, 1968). Thus, the Fothergilleae sensu Endress (1989c) is recognized. The apetalous genera of the Hamamelidoideae have long been considered as monophyletic (Harms, 1930; Endress, 168 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1989c), but the ITS-phylogenies (Figs.3.3, 3.5) suggest that loss of petals in this subfamily has evolved at least three times independently: Matudaea in the Corylopsis clade, Mo1inadendron in the Trichocladus clade, and the Fothergilleae in the Hamamelis clade. Therefore, the apetalous genera do not belong to a single monophyletic group, and apetaly is homoplasious in this family. Fotherqilla and Parrotiopsis each form their own clades in the ITS phylogenies (Figs.3.3, 3.5). This is consistent with the showy specialized structures they have evolved for insect pollination: long, white, and clavate stamen filaments in Fotherqilla. and large, white subfloral bracts in Parrotiopsis (Bogle, 1970; Endress, 1989a). These specializations do not occur in any other members of the Fothergilleae sensu Endress (1989c) or in the Hamamelidaceae. It is a novel relationship that Hamamelis is grouped with Fothergilleae sensu Endress (excl. Matudaea and Molinadendron). Interestingly, this pattern finds support from leaf venation (Harms, 1930; Chang, 1979; Li and Hickey, 1988). The systematic position of Hamamelis is not well resolved in the ITS phylogenies (Figs.3.3, 3.5), however, this genus differs greatly from the Fothergilleae in a group of floral characters, including showy, strap-shaped petals, bisexuality, nectariferous phyllomes, and insect pollination (Endress, 1989a). Therefore, it should be isolated from the Fothergilleae in taxonomic treatments. 169 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In summary, at the subfamily level, this study recognizes the monophyly of Liquidambaroideae, Exbucklandioideae, Mytilarioideae, Rhodoleioideae, Disanthoideae, and Hamamelidoideae. Furthermore, it suggests a close relationship of Exbucklandia with Rhodoleia, and the paraphyly of Exbucklandioideae sensu Endress (1989c). At the tribal and subtribal levels, the monophyletic groups are the Corylopsideae sensu Endress (1989c), Eustigmateae sensu Endress (1989c, incl. Molinadendron). Fothergilleae sensu Endress (1989c, incl. Hamamelis), Dicoryphinae, and Loropetalinae (incl. Matudaea). The Hamamelideae, however, is polyphyletic. The ITS phylogeny further suggests that some morphological characteristics commonly utilized to classify the Hamamelidoideae have originated several times independently. These characteristics include strap-shaped petals, apetaly, monoecy, and wind pollination. Therefore, whenever a classification system is attempted, this homoplasy should be taken into account so that natural relationships are reflected. 170 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. U65464 U65465 AF015425-26 U65462 AF015423-24 AF015656-57 U65463 AF015436 campus Park, Hawaii Park, Hampshire New of Univ. campus Hampshire New of Univ. Taiwan U65466 Tulear, MadagascarTulear, AF015419-20 China AF015417-18 greenhouse. Gard. Bot. Missouri AF015427-28 Woodlanders, Inc. SC.Inc. Woodlanders, AF015421-22 Roadside State Manuka MA.Arboretum, Arnold U65467 Arnold Arboretum, MA. Arboretum, Arnold MA.Arboretum, Arnold AF015655 BOGLE SC. Inc. Woodlanders, BOGLE BOGLE 04 LI BOGLE SC. Inc. Woodlanders, BOGLE LI 01LI MA. Arboretum, Arnold QIU 03 LI BOGLE Hampshire New of Univ. LI LI 02 MA.Arboretum, Arnold U65461 J.-H. LI J.-H. LI J.-H. L. A. A. L. A. A. L. A. FOT J.-H.HVI LI DIR FOR J.-H. EXB DIC 543 Randrianasolo A. DOTDIM L. A. BOGLE L. A. SC.Inc. Woodlanders, AF019231-2 HVE DIS Taxon Taxon CPA J.-H. GenBank EUS N.-J. CHUNG CPA J.-H. CSI LSI L. A. LFO L. A. ALT Y.-L. Fothergilla majorFothergilla Hamamelis virginiana Corylopsis pauciflora Corylopsis Corylopsis Corylopsis sinensis Distyliumracemosum Fortunearia sinensis Distyliopsis Distyliopsis tutcheri myricoides Distylium Disanthus Disanthus cercidifolius Species Species Corylopsis spicata Abbr. Voucher and Collector Source populnea Exbucklandia Accessions Dicoryphe stipulacea Eustigma oblongifolium Hamamelis vernalis Liquidambar formosana Loropetalum sinense Altingia excelsa Table 3.1. Species sequenced for nrDNA ITS for this analysis. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. U65469 AP015439 AF015441 AFO15443 AF015442 AP015429-30 U65468 AF015438 AP015440 AF015431-32 AF015433-34 AF019233 AF022241 AF015437 AF022242 AF015435 Zurich, Switzerland Zurich, Zurich, Switzerland Zurich, Switzerland Zurich, Zurich, Switzerland Zurich, greenhouse. Zurich, Switzerland Zurich, Botanical Garden of Garden Botanical Univ. of New Hampshire New of Univ. Harvard Univ. campus Univ. Harvard Botanical Garden of Garden Botanical Botanical Garden of Garden Botanical Jiangsu, China Jiangsu, Kong Hong Kepong, Malaysia Kepong, Garden, Botanical China Guangxi, of Garden Botanical Lyon Arboretum, Hawaii Arboretum, Lyon MA Arboretum, Arnold SC. Inc. Woodlanders, PA. Gardens, Longwood Z.-C. LUO Z.-C. P. K. ENDRESS K. P. ENDRESS K. P. ENDRESS K. P. ENDRESS K. P. QIU Y.-L. LI 05 J.-H. P. K. ENDRESS K. P. BOGLE L. A. BOGLE L. A. Saunders R. L. G. SAW G. L. BOGLE L. A. BOGLE L. A. BOGLE L. A. PPS MOL TET SHA NOA OST TRI NEO MYT RHO SIN MAT SYS MAI PAR Tetrathyrium subcordatum Tetrathyrium crinitus Trichocladus Shaniodendron Shaniodendron subaequale henryi Sinowilsonia Sycopsis sinensis Ostrearia Ostrearia australiana Molinadendron Molinadendron guatemalense Neostrearia fleckeri Noahdendron nicholasii Parrotia persica Parrotia Parrotiopsis Parrotiopsis jacquemontiana Rhodoleia championi Matudaea Matudaea trinervia Mytilaria laosensis Table 3.1. Continued malayanaMaingaya - j N) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2. Sequence divergence of ITS-1 and ITS-2 (below and above diagonal) in the Hamamelidaceae. The values are derived from pairwise distances, calculated in PAUP* (mean distance x 100). Taxon abbreviations as in Table 3.1. Taxon CPA CSI CSPDIS DICDIMDIRDOTEUSEXBFOR CPA - 3 3.9 24.9 16.7 15.4 15.8 15.1 15.1 19.5 14.1 CSI 4.1 - 0.9 23.5 15.7 14 14 13.3 14.1 18.5 13.6 CSP 3.7 2.2 - 24 16.6 14.9 14.9 14.1 15 19 14.4 DIS 24.4 26.3 26.2 - 25.1 23.4 23.3 22.5 25.3 28.1 25.2 DIC 14.9 16.8 17.5 24.3 - 11.6 10.7 9.8 8.1 22.9 5.9 DIM 15.3 17.3 17.2 23.6 11.2 - 1.7 2.6 9.9 19.4 8.2 DIR 15.3 17.2 17.2 24 12 1.1 - 1.7 9.1 19.7 8.6 DOT 15.6 17.6 17.5 23.7 12 0.7 0.4 - 8.3 20.3 8.6 EUS 10.9 13.7 13.6 23.2 10.8 8 8.5 8.4 - 21.1 6 EXB 30.6 32.8 32.7 31.9 30.1 30.4 30.6 30.8 28.6 - 20.6 FOR 12.9 14.9 14.9 22.7 9.4 8 8.4 8.4 3.4 28.1 - FOT 14 16.4 16 22.9 11.9 5.3 5.7 5.7 7.6 30.4 8.4 HVE 14.5 16.5 16.4 22.1 11.1 3.7 4.1 4.1 7.7 30.5 7.7 HVI 15.6 17.9 17.9 23.6 11.9 4.5 4.9 4.9 8.5 30.9 8.5 LSI 11.5 12.5 12.8 26.4 18 17.7 17.7 17.7 15.7 30.7 16.8 MAI 15.2 15.8 15 27.1 17.4 18.3 18.6 18.7 17.5 31.7 16.8 MAT 13.7 13.5 13.5 26 18.7 19.6 19.6 19.6 17 30.8 17.3 MOL 12.2 15 14.9 24.4 10.2 8.4 8.4 8.4 3.4 27.7 5.5 MYT 30 32.2 32.1 29.6 28.8 28.5 27.9 28.2 26 25.9 23.8 NEO 13.8 17.2 16.8 25.1 13.1 11.9 12.3 12.3 6.4 31 7.1 NOA 12.9 15.7 15.3 24.8 10.7 11.2 11.5 11.5 7.3 30.9 8 OST 11.6 14.9 14.5 25.1 10.4 10.4 10.8 10.8 7.7 32.5 8.5 PAR 16.8 18.7 18.6 24.8 12 1.9 3 2.6 8.6 29.7 8.4 PPS 16 17.9 17.9 23.6 13 6.3 7.1 7.1 8.5 29 9.3 RHO 28.7 30.8 31.5 28.8 28.2 30.1 29.4 29.7 27.2 19.1 25.7 SHA 16 18.3 18.3 23.6 10.8 2.2 3.3 3 9.4 30.8 8.5 SIN 13 15.9 15.8 26.1 10.7 8.8 9.2 9.2 3.8 30.9 5.1 SYS 17.5 19.4 19.3 26.5 13.5 3.4 4.1 3.7 9.8 31.5 9.8 TET 11.4 13.2 12.8 25.7 16.7 16.9 17.2 17.2 15.4 27.9 16.9 TRI 15.6 16.4 16.4 25.1 11.5 11.6 11.6 11.6 9.4 32.2 9.8 LFO 32.5 34 34.3 34.8 36.2 32.1 31.3 32.1 28.2 38.9 29.1 ALT 35.5 37.4 36.9 36 35.6 32.7 32 32.8 31.4 39.3 30.5 173 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 Continued FOT HVE HVILSI MAIMAT MOLMYT NEO NOA 0ST PAR PPS 15 12 13.3 19 11.2 12.1 15.3 23.1 17.6 14 15 14.6 13.6 14 11.5 12.8 18 9.1 10.8 13.5 23.9 16.6 13 14.4 12.7 12.7 14.9 12.4 13.7 18.9 9.9 11.7 14.4 23.4 17.5 13.9 15.3 13.6 13.5 22.9 23.4 23.6 25.2 25.4 24.2 24.1 31.3 23.5 24.1 25.1 22 23.4 10.7 10.3 10 17.2 15.1 14.8 8.9 26 9.4 8.5 8.1 10.3 9.6 6.4 4.3 4.7 19.1 16.9 17.9 10.3 25.7 13.4 11.6 11.1 2.1 4.8 5.5 5.2 4.7 18.1 15.6 17 9.8 25.2 12.5 10.7 10.2 1.3 3.9 4.7 4.3 3.9 16.7 15.6 17.1 10.8 24.8 12.5 9.9 9.4 1.3 3 8.2 8.6 8.2 17.5 15.6 16.5 8.5 25.8 11.2 7.7 7.7 8.6 8.4 20.7 20.3 20.8 27.5 21.3 23.8 19.7 26.8 24.5 24.5 21.9 19.3 20.6 9.4 8.1 7.8 15.9 14.2 18.2 5.9 26.9 9.8 6.8 8.1 9 7.9 - 4.7 4.3 20.2 16.7 17.4 11.5 25.1 12.4 10.2 9.8 4.3 3.9 3.7 - 2.1 19.5 16.4 17.1 12 25.5 12.9 10.3 9.8 3.9 3.5 4.5 1.5 - 19.1 16.1 16.8 11.6 25.8 13 10.4 9.5 3.4 3 17.5 17.2 18.3 - 19.9 11.2 15.8 31.9 17.3 14.5 17.6 17.6 18.2 18.2 18.2 19.1 11.6 - 14.7 14.5 24.9 16.9 14 15.5 15.1 14.6 19 18.4 19.6 9.7 11.2 - 14.7 27.6 17.5 14.7 15.1 16.6 15.4 7.6 8.1 8.9 16.9 18.2 17.8 - 27.9 12 10.2 9.7 10.3 11.3 28.9 27.1 26.7 30.5 32.7 31 25.6 - 26.8 26.7 26.7 25.1 24.6 11.2 10.5 11.9 19.2 21 21.1 6.8 30 - 9.4 8.1 12.9 12.7 9.6 9.6 10.3 18.2 18.9 20.1 7.6 30.5 6.7 - 5.5 10.3 9.6 9.2 9.2 10 17.6 18.3 19.2 8.1 30.8 7.3 5 - 9.8 10 5.7 3.4 4.1 18.4 19.5 20 8.9 27.5 12 11.9 11.2 - 2.6 6.4 5.6 6.3 18.7 20.5 20.3 8.9 28.8 12.2 12.5 11.5 5.2 - 28.8 28.2 28.6 29.3 29.5 28.3 25.8 23.8 29.1 30.3 28.6 29.1 27 6.1 3 4.5 18 18.3 19.2 9.8 27.9 12.3 12.3 11.6 1.8 6.3 9.3 8.5 9.3 17.4 18.7 19.2 4.7 26.8 8 9 9.4 9.3 10.1 7.5 5.2 5.6 18.7 19.5 21.1 9.4 29.1 14.3 13 12.7 3.4 7.9 17.1 17.2 18.4 5.6 11.2 10.5 17 28.7 19.9 18.9 18.3 16.9 18.3 11.1 10.7 11.5 18.3 18.5 19.1 8.9 30.4 11.5 9.2 9.6 13.1 12.2 33.1 33 33 36.2 36.5 35.9 29 33.4 35.1 33.3 32.5 33.1 33.5 34 32.9 32.9 37 36.5 36.3 31.3 36.8 36.1 33.8 34.3 32.9 34 174 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2 Continued RHOSHA SIN SYSTETTRI LFOALT 21.9 15.7 18.2 15.9 12.5 14.4 39.7 38.8 20 13.9 16.4 14.1 12 14.8 40 39.1 20.4 14.8 17.3 15 12.9 15.7 40.5 39.6 29.7 22.5 26.2 23.3 22.6 23.9 41.5 39.1 24.4 9.9 9.3 10.7 16.8 9.3 37.7 37.9 23.9 2.6 12.9 3.4 18.6 11.2 37.5 36.7 23 1.7 12.9 2.5 17.2 11.1 36 35.2 23.1 1.7 12.1 0.4 17.3 10.3 35.8 35.9 23.6 9.1 9.4 9.1 17.7 10.3 38.2 38.8 15.9 20.5 24.7 20.7 22.4 23.7 37.5 37.8 23 9.4 7.6 9.4 17.5 10.2 37.6 37.8 24.4 5.5 11.9 5.1 18.5 11.6 37.4 37.9 24 5.2 12 5.2 17.8 9.9 37.8 38.3 23.7 4.7 12.1 4.7 17.4 10.3 38.1 38.5 28 17.1 19.6 17.5 14.1 19 35.3 35 22.7 16.3 18.8 16.4 15.1 17.5 40.4 37.8 23.7 17 17.2 17.9 9.4 17.7 37.4 35.6 23 10.7 7.6 11.6 16.7 11 38.5 36.9 30.1 24.5 29.2 25.7 25.4 25.5 39.4 39.3 24.4 12.5 13.6 13.3 15.5 12.4 38.4 38.5 24.3 10.7 10.6 10.7 14.4 11 38.5 38.6 23.4 10.2 10.6 10.2 15 10.2 37.6 37.8 22.6 1.3 12.4 2.1 16.8 10.7 36.5 35.7 23.7 3.9 11.7 3.9 16.8 10.9 36.6 36.7 - 23.3 24.9 23.9 22.8 22.7 39.4 40.2 29.1 - 12.8 2.5 16.2 11.1 35.5 34.7 29.6 10.2 - 12.9 20.5 11.9 40.4 40.5 30.5 3.7 10.6 - 18.1 11.1 36.5 36.7 28.5 16.5 17.9 18.4 - 18.4 38.5 37.1 30.5 13.5 11.1 13.4 18.2 - 38.4 38.5 35 33.4 30.5 34.5 35.8 33.9 - 6.3 36.7 33.7 31.8 34.7 37 35.1 14.6 - 175 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.3. Characteristics of the most parsimonious trees obtained by PAUP heuristic searches of ITS-1 sequences aligned using different combination of gap and gap length penalties. Abbreviations: Consistency Index (Cl), Retention Index (RI), Rescaled Consistency Index (RC), number of steps (Steps), number of trees found (Trees), Number of characters in the matrix (Nchar), the aligned considered optimal is highlighted. Penalty (Gap/gap length) Steps Trees Nchar Cl RI RC 10/10 581 90 291 0.613 0.69 0.423 15/8 582 402 291 0.612 0.695 0.425 20/20 665 1716 284 0.561 0.662 0.371 20/8 586 492 289 0.600 0.689 0.414 8/15 576 234 289 0.611 0.696 0.425 8/20 583 858 288 0.602 0.696 0.419 8/8 548 210 294 0 . 620 0.696 0.432 176 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.4. Characteristics of the most parsimonious trees obtained by PAUP heuristic searches of ITS-2 sequences aligned using different combination of gap and gap length penalties. Abbreviations: Consistency Index (Cl), Retention Index (RI), Rescaled Consistency Index (RC), number of steps (Steps), number of trees found (Trees), Number of characters in the matrix (Nchar), the aligned considered optimal is highlighted. Penalty (Gap/gap length) Steps Trees Nchar Cl RI RC 10/10 435 753 260 0.556 0.623 0.347 15/8 440 180 259 0.566 0.633 0.358 20/20 480 54 251 0. 577 0.635 0.366 20/8 450 162 258 0.573 0.639 0.366 8/15 453 138 255 0.559 0.625 0.349 8/20 463 252 252 0.572 0.634 0.363 8/8 413 81 262 0.557 0.606 0.337 177 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64 61.5-67 471-522 1.4-38.2 Combined 168 168 (30.8) 377 377 (69.2) (ITS-1 (ITS-1 and ITS-2) 156121 365 244 235 501.8 -1.2 -1.0 ITS-2 168 168 (66.9) 45 30 75 294 251 545 266.9 ITS-1 236-277 229-241 1.5-39.3 1.7-40.5 57.5-66.6 61.3-68.2 and combined, in 32 species of the Hamamelidaceae. (gi value (gi for 10,000 random trees) -0.8 Skewness of tree length distribution Transversions (minimum)Transition/transversion 118 1.68 1.29 1.5 Sequence divergence (%) Number of potentially informative sitesNumber of constant sitesTransitions (minimum) 153 (52) 128 (51) 281 (51.5) 85 (28.9) 83 (33.1) 198 Number of variable sites 209 (71.1) G+C content mean (%) 63 64.9 Length range (bp) G+C content range (%) Number of indels Lengthmean (bp) Aligned length (bp) Table 3.5. Sequence characteristics of the two internal transcribed spacers, separated Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 3.1. The ITS region. Arrows indicate orientation and approximate positions of primer sites. Primer names and sequences as in White et al. (1990). 179 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ITS5 ITS3 — > — > 5.8S 18S nrDNA ITS-1 nrDNA ITS-2 26S nrDNA < - - < - ITS2 ITS4 ITS Region 180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.3.2. UPGMA phenogram of the Hamamelidaceae generated by analysis of nrDNA sequences aligned using the Optimality method. Jukes-Cantor's one parameter model and gamma rate (1.0) were used in the analysis. Taxon abbreviations as in Table 3.1. 181 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ALT LFO CPA CSI CSP MAI MAT LSI TET DIC EUS FOR MOL SIN NEO NOA OST DIM DIR DOT PAR SHA SYS FOT HVE HVI PPS TRI DIS EXB RHO MYT 182 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.3.3. Strict consensus of 90 shortest trees of 1041 steps, generated using nrDNA ITS sequences of the Hamamelidaceae, aligned using the Optimality method. Taxon abbreviations as in Table 3.1. Numbers above and below branches are bootstrap percentages and decay values respectively. CI=0.59, and RC=0.39. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae. 183 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ALT — ALToideae CPA-1 100 aat CSI a> Corylopsis dadi >5 99 os CJo 100 \ >5 c s p -j >5 97 LSI-i LORinae 95 >5 TET (HAMeae) >5 62 MAI J 1 MAT— FOTeae DIC — DXCinae (HAMeae) EUS-i 97 80 FOR BUSeae Trichocladus dade >5 84 MOL —FOTeae 72 ^ 4 SIN-1 1 NEO-| J3JL NOA 73 DXCinae HAMoideae OST (HAMeae) TRI-1 66 62 DIM “I 80 <50 DIR DOT (0v PAR 184 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.3.4. 50% Majority consensus of 90 shortest trees of 1041 steps, generated using nrDNA ITS sequences of the Hamamelidaceae, aligned using the Optimality method. Taxon abbreviations as in Table 3.1. Numbers are clade frequencies. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae. 185 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ALT — ALToideae CPA 1 100 100 CSI 100 CSP J LSI 1 100 LORinae 100 TET (HAMeae) 100 MAI -J MAT— FOTeae 60 NEO OlCinae 100 (HAMeae) 100 NOA 100 OST 100 TRI —1 100 EUS EUSeae FOR (HAMeae) 60 HAMoideae 100 MOL FOTeae SIN-1 DIM - 100 100 100 DIR 100 DOT 100 PAR SHA SYS 67 PPS 100 100 100 HVE HVI POT -I DIS— 100 EXB EXBoideae 100 RHO RHOideae MYT. LFO— ALToideae Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.3.5. Strict consensus of 12 shortest trees of 6747 steps, generated using nrDNA ITS sequences of the Hamamelidaceae, aligned with the Elision method. Taxon abbreviations as in Table 3.1. Numbers above and below branches are bootstrap percentages and decay values respectively. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DlCinae=Dicoryphinae. 187 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ALT — ALToideae CPA -i 1001 >6 I 1001 CSI >6 I 100 CSP J >6 99 LSI-1 LORinae 100 >6 TET (HAMeae) >6 56 MAI-I MAT — FOTeae <50 DIC-I DICinae <50 TRI (HAMeae) 100 NEO 100 >6 NOA 100 >6 I 93 >6 OST — 'EUS— | EUSeae 100 , FOR (HAMeae) >6 HAMoideae 1 liooj MOL FOTeae SIN-I 100 DIM 91 >6 DIR 100 >6 DOT 100 PAR >6 SHA SYS 100 FOT 91 >6 >6 100 HVE •H >6 HVI PPS 100 >6 DIS—I EXBoideae 100 EXB >6 RHI RHOideae MYT— LFO— ALToideae 188 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.3.6. One of the shortest tree of 1041 steps using nrDNA ITS sequences of the Hamamelidaceae, aligned using the Optimality method, showing the number of unambiguous character state changes of ITS sequences. Taxon abbreviations as in Table 3.1. 189 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ALT CPA CSI CSP 28 T 190 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX 3.1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 20 40 60 as as in Table 3.1. represents gaps. SHA SIN TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGC-TGC-GGGG-GCGAGGGG-AGGCATGC-CCCSYS TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AA-CAATGC-TAT-GGGG-GCGAGGGG-AGGCATGC-CCC TCGAAACCT-GCCCAGBAGAACGACCCGCGAACACGT-NT-CAATGC-TAC-GGGG-GCGAGGGG-AGGCATGC-CCC Teuton Teuton |ITS-1 -> CPA CSI TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AAAAA-TACAAGT-GGGG-GCGGGGGGGGCSP TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AAAAAATACAAGT-GGGG-GGGGGGGGGCA-TTCGTGCTCDIC TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AAAAAATACAAGT-GGGG-GCGGGGGGGCAATCCGTGCTC TCGAAACCT-GCCCAGCAGAACGACCCGCGAACGAGT-AA-GAAAGC-TAC-GGGG-GGGAGGGG-AGGCATGC-CCC CATGCTC FOT HVE TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGC-TAT-GGGG-GCGAGGGGTGGGCATGC-CCC TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGC-TGT-GGGG-GCGAGGGG-AGGCATGC-CCC PAR PPS TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGC-TAC-GGGG-GCGAGGGG-AGGCATGC-CCCRHO TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGC-TAC-GGGG-GCGGGGGG-AGGCATGC-CCC TCGAAACCT-GCACAGCAGAACGACCCGCGAACCGGTAAAC— ACGT TCGGGGGGGAGCGGG-GGGTGCAAGCCC ALT TCGATGCCCCGCAAACAAAACCAACCCGCACACATATTATTCA CATATCGGGGGGCGGGGGG— GGCGAAACCCC DIS DIM TCGAAACCT-GCCCAGCAGAACGACCCGTGAACATGT-ATA-AACTCCCGTGGGGGAGCGGAGGGGGGGCATGC-CTCDIR TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGC-TAC-GGGG-GCGAGGGG-AGGCATGC-CCCDOT TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGCATAA-GGGG-GCGAGGGG-AGGCATGC-CCCEUS TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AT-CAATGC-TAA-GGGG-GCGAGGGG-AGGCATGC-CCCEXB TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AA-CAATGC-CAT-GGGG-GCGAGGGG-AGGCATGC-CCC TCGAAACCT-GCACAGCAGAACGACCCGCGAACCGGTTACA— ACGT HVI CCGGGGGGAGGGGTGTGGGTGCAAGCCC TCGAAACCT-GCCCAGTAGAACGACCCGCGAACACGT-AT-CAATGA-TAT-GGGG-GCGAGGGG-AGGCATTC-CCCMAI TCGAAACCT-GCACAGCAGAACGACCCGCGAACATGT-AAAAA-TACTGCCCGGTG-GCGAAGGGGCATTCC CCC TRI TCGAAACCT-GCTCAGCAGAACGACCCGCGAACACGT-AA-CAATGC-TAT-GGGG-GTGAGGGG-AGGCATGC-CCT Appendix 3.1. Aligned nrDNA ITS sequences in the Hamamelidaceae. Taxon Abbreviations MYT TCGAAACCT-GCCCAGTAGAACGACCCGCGAATACGTAACACAACGT TTGGGGGGCGAGGGG— GGCGCAGGCCC TET TCGAAACCT-GCACAGCAGAACGACCCGCGAACACGT-AAAAA-CACAAGCCGGGG-GCGG-GGGGGATTCC CCT NEO TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AA-CAATAC-TGT-GGGG-GCGAGGGG-AGGCATGC-CCCOST TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AA-TAATAC-TAT-GGGG-GCGAGGGG-AGGCATGC-CCC FOR TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AA-CAATKC-CAT-GGGG-GCGAGGGG-AGGCATGC-CCC MAT TCGAAACCT-GCACAGCAGAACGACCCGCGAACATGT-AAATA-CACAAGTTGGGG-G-AG-GGGGCATTCC ACC MOL TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AA-CAATGC-TAT-GGGG-GCGAGGAG-AGGCATGC-CCC NOA TCGAAACCT-GCCCAGCAGAACGACCCGCGAACACGT-AA-CAATAC-TAT-GGGG-GCGAGGGG-AGGCATGC-CCC M fo LFO LSI TCGATGCCC-GCAAAGCAGAACGACCCGCGAACACAT-ACGAA TCGAAACCT-GCACAGCAGAACGACCCGCGAACATGT-GAAAA-CACAAGTCGGTG-GCGG-GGGG-ATTTC AACATCGGGGGGGCGGGGG— GGCACGAGCCC CCC Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ------CGGCAA CGGCAA CGGCAA CGGCAA GGGGC-AAGCTTC ------GTCGGGACGCGTCCGGTGCTT GTCGAGACGCGTCCGGTGCTT GTCGAGACGCGTCCGGTGCTT GTCGGGACGCGCTCGGTGCTCGGTCCGCGACCCGTGC-CAC-GTGCTCGG-GAAGCCCCGTCGGGACACGCTCGGTCCCCGGCCCGCATCCCGTGC-CGC-GTGCTCGG-GAAGCCCC ------GTCGAGACGCGTCCGGTGCTC ----- • • • • • MAI CTCGC-CCCC— AGA SIN CTCTCACCCCC-AAA SHA CTCTCTCCCCCCAAA GTCGGGACGCGCTCGGTGCTCGATCCGCCGACCGTGC-CAC-GTGCGCGG-GAAGCACC PAR CTCTCTCCCCCCAAA GTCGGGACGCGCTCGGTGCTCGATCCGCCAACCGCGC-CAC-GTGCGCGG-GAGGCACC SYS TET CTCTCTCCCCC-AGA CTCGC-CCCC— AAA GCCGGGACGCGCTCGGTGCTCGATCCGCCAACCGTGC-CAC-GTGCCCGG-GAAACCCC GTCGGGACGCGCTCGGCGCTCGGTCCGCAGCCCGCGC-CAC-GTGCTCGG-GAAGCCCC HVI CTCTCTCCCCC-AAA GTCGGGACGCGCTCGGTGCCCGATCCGCCAACCGTGC-C-CCGTGCGCGG-GAAGCATC DIC DIS CCCTCGCCCCC-AAG CTCTCTCTCCTCTACCC—DIR GTCGAGACGCGCTCGGAGCTCGATCCGCAGCCCGTGC-C-CCGTGCGCGG-GAAGCTTC ATCGGGATGCGCTTGGCGTCCGACCCGCGACCCGTGT-C— CTCTCTCCCCC-ATA C-TGCACGG-AGAGCATC GTCGGGACGCGCTCGGTGCTCGATCCGCCAACCGTGC-C-CCGTGCACGG-GAAGCACC PPS RHO CTCTCTCCCCC-AAA -CTC-TCTTCC— TCCCGTGTCGGGACGCGCTCGGCAC-CGCCCCGTGAC— GTCGGGACGCGCTCGGTGCTCGACTCGCCAAACGTGC-C-CCGCGCGCGG-GAGGCATC CTGCGCGATGCGTGCAGGGAGCCGGC TRI CTCTCAACCCC-AGA GTCGGGACACGCTCGGTGCACGTTTCGCCACCCGTGC-C-CCGTGTACGG-GAAGCTTC EXB FOR TCTCCTCCTCCACCCCCAAGTCGGGATGCGCATGGCTT-CGCACCGCGCC—FOT CTCTCACCCCC-AAC HVE CCGTGCCCAGTGTGCGGGGAGC— CTCTCTCCCCC-AAA GT CTCTCTCCCCC-AAA GTCGGGACGCGCTCGGTGCTCGATCCGCCAACCGTGC-C-CCGTGCGCGG-GAAGCATC GTCGGGACGCGCTCGGTGCCCGATCCGCCAACCGTGC-C-CCGTGCGCGG-GAAGCATC Taxon 80 CPA CSI CTCTCACCCCC-AAA CSP CTCTCACCCCC-AAA CTCGAGACGCGCTCGGTGCTCGATCAGCAACGCGTGC-CTC-GCGCACGG-GAAGCTTC CTCTCACCCCC-AAA ATCGGGACGCGCTCGGTGCTCGATTAGAAACGCGTGC-CTC-GTGCACGG-GAAGCTTCDIM ATCGGGACGCGCTCGGTGCTCGATTAGCAACGCGTGC-CTC-GTGAACGG-GAAGCTTC CTCTCTCCCCC-AAA 100 DOT GTCGGGACGCGCTCGGTGCTCGATCCGCCAACCGTGC-C-CCGTGCACGG-GAAGCACC CTCTCTCCCCC-ATA GTCGGGACGCGCTCGGTGCTCGATCCGCCAACCGTGC-CACCGTGCACGG-GAAGCACC 120 140 LFO CTTCTCCCTTCC CC— GTCGAGAAGCGGCCAGTGCT-AGCGTGCCCCTCGACCACGG-GT-TAGGGGAAAGCGTC MYT -CTCTCCTTCCCWTC OST GTCGGGATGCGCTCGGTGC-TGCTCCGCACCTCCCATGCCCTGTGCACGGGAAGC— CTCTCACCCCC-AAA GT GTCGAGACACGCTCGGTGCTCGATCCGCCACCCGTGC ALT CTTCTCCCCTCT TC— GTCGGRAAGCGGCCAGTGCG-GGCGT-CCCCTCGCCCACGG-GT-TGGGGGAAAGCGTC EUS CTCTCACCCCT-AAA NEO NOA CTCTCACCCCC-AAA CTCTCACCCCC-AAA GTCGAGACGCGCTATGTGCTCGATCTGCCACCCGTGC-C-ACGCGTGCGGCAAGCAAAA GTCGAGACACGCTCGGTGCTCGATCTGCCACCCGTGC-C-CCGTGTGCGGCAAAGCTTC MAT MOL CTCAC-CCCCA-AAA CCCTCACCCCC-AGA GTCGGGACGCGCTCGGTCTCCGGTTCGCGACCCGGGC-CAC-GTGCTCGG-GAAGCCCC Appendix 3.1 Continued 3.1 Appendix co £ LSI CTCGC-CCCCC-AAA Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CGCCGTGCTCTCGACA—TCACAACGAACCCCGACGCAAAACGCGTCAAGGAAA-CGCAATGCAAGAGCATG CGGCGTGCTCCCGGTA—TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAA-CGCAATGCAAGAGCATG CGCCGTGCTCTCGATA—TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAA-CGCAATGCAAGAGCATG CGTCGTGCTCTCGATA— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAA-CGCAATGCAAGAGCATG • • • • • ------DIS GGCT-CCAACGCATTCCCGATC— ACACAACGAACCCCGGCGCAAAACGCGCCAAGGAA-CTTTAACGCAAGAGCATGDOT GGCA-CCGGCGTGCGCTCGGCA— TCACAACGAACCCCGGCGCAAATCGCGTCAAGGAAA-CGAAACGCAAGAGCTTG FOT HVE GGCA-CCGGCGTGCGCTCGATA—HVI GGCA-CCGGCGTGCGCTCGGCA— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAA-CGCAACGTAAGAGTCTG GGCA-CCGGCGTGCGCTCGGCA— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAA-CGCAACGCAAGAGCCCG TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAA-CGCAACGCAAGAGCCCG SYS GGGAACCGGCGTGGGCTCGGCA—TRI TCACAACGAACCCCGGCGCAAATCGCGTCAAGGAAA-CGAAACGCAAGAGCCCG GGCG-CCAGCATGCTCTCGATA— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAC-CGCAATGCAAGAGCGTG PAR GGCA-CCGGCGTGCGCTCGGCA— TCACAACGAACCCCGGCGCAAATCGCGTCAAGGAAA-CGAAACGCAAGAGCCCGSHA SIN GGCA-CCGGCGTGCGCCCGGCA— TCACAACGAACCCCGGCGCAAATCGCGTCAAGGAAA-CGAAACGCAAGAGCCCG Taxon 160 CSI GGTG-CCGTCGTGCTCTCGATC—DIC TCACAACGAACCCCGGCCCAAAACGCGTCAAGGAATCTAAAATGCAAGAGCGTG GGCG-CCGGCGTGCTCCCGACA—DIM TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAC-CGTAATGCAAGAGCGTGDIR GGCA-CCGGCGTGCGCTCGGCA— 180 GGCA-CCGGCGTGCGCTCGGCA— TCACAACGAACCCCGGCGCAAATCGCGTCAAGGAAAACGAAACGCAAGAGCTTG BUS TCACAACGAACCCCGGCGCAAACCGCGTCAAGGAAA-CGAAACGCAAGAGCTTGFOR 200 220 LFO GGTGCTTGTCGTGTTCTTAGCCC—MAI CATAACGAACCCCAGAGCAAACCGCGCCAAGGAAACT-CAACTCGAGAGCATGMAT GGCG-CCGTCGTGCTCCCGATC— GGCG-CCGTCGTGCTCCCGATC— ACACAACGAACCCCGGCGCAAAACGCGTCAAGGAATCTCAA-CGTAAGAGCGTG TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAATCTCAA-TGCAAGAGCGTG TET GTCC-CCGTCGCGCTCTCGATC— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAATCTCAA-CGCAAGAGCGTG MYT NEO CTGTGCCGGTGCGCTTCTCG-ACCTCACAACGAACCCCGGCGCAAAACGCGCCAAGGAA-CTTGAACSCAAGAGAGCWNOA GGCA-CCGGCATGCTCTCGATA—OST GGCA-CTGGGGTGCTCTCGATA— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAC-CGCAATGCAAGAGCGTG GGCA-TCGGCGTGCCCTCGATA— TCACAATGAACCCCGGCGCAACACGCGTCAAGGAAC-CGCAATGCAAGAGCGTGPPS TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAC-CACAATGCAAGAGCGTGRHO GGCA-CCGGCGTGCGCTCGATA— GAGTGTCGGTGCGCTCCCGATACACAACAACGAACCCCGGCGCAAAACGCGCCAAGGAA-CTCTAACGCAAGAGCACG TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAAA-CGCAACGCAAGAGCCCG ALT CPA GGTGCTTGTCGTGTTCTCGGCCC— GGTG-CCGTCGTGCTCTCGATC— CATAACGAACCCCGGCGCAAACCGCGCCAAGGAAATT-TAACTCGAGAGCATGCSP TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAATCTAAA-TGCAAGAGCGTG GGTG-CCGTCGTGCTCTCGATC— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAATCTAAAATGAAAGAGCGTG EXB TCGTGCCGG— CGCTCTCGAT-C-TAACAATAATCCCCGGCGCAAGACGCGCCAAGGAA-CTCTAACGCAAGAGCATG Appendix 3.1 Continued 3.1 Appendix MOL £ LSI GTCC-CCGTCGTGCTCTCGATC— TCACAACGAACCCCGGCGCAAAACGCGTCAAGGAATCTCAA-TACAAGAGCGTG Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. jlTS-2. DIS DIS CGTATTGCGTTGCCCCCC AGCCCCGTG— CCCCGTTCTCGGGACGCCCGGGGGGTAGTGCAATCTTCGATAT-ATC-T SHA SHA CGTATCGCGTCGCCACCC C-TCCCATC— CCCCGGTCTCGGGTCGTGGTGGGGGCGATGGCATCTTCGATATTAT-AC Taxon Taxon 240 CPA CGTATCGCGTCGCCCCCC CSI C-TCCCATC— CGTATCGCGTCGCCCCCC CCCCGGTCACGGGTTGTGGTGGGGGCAATGTCATCTTCGATATTATT-C CSP C-TCCCATC— CGTATCGCGTCGCCCCCC GCCCGGTCACGGGTTTTGGTGGGGGCAATGTCATCTTCGATATTATT-C DIC C-TCCCATC— CGTATCGCGTCGCCCCCA GCCCGGTCACGGGTTTTGGTGGGGGCAATGTCATCTTCGATATTATT-C C-TCCCATC— CCCCGGTCCCGGGCCGCGGTGGGGGCAATGGCATCTTCGATATTATT-C DIM CGTATCGCGTCGCCACCC DIR C-TCCCATC— CGTACCGCGTCGCCACCC CCCCGGTCTCGGGTCGTGGTGGGGGCGATGGCATCTTCGATATTAT-AC DOT C-TCCCATC— CGTATCGCGTCGCCACCC CCCCGGTCTCGGGTCGTGGTGGGGGCGATGGCATCTTCGATATTAT-AC 260 EUS C-TCCCATC— CGTATCTCGTCGCCCCCA CCCCGGTCTCGGGTCGTGGTGGGGGCGATGGCATCTTCGATATTAT-AC EXB C-TCCCATC— CCCCGGTCTCGGGTCGTGGTGGGGGCAATGGCATCTTCGATATTATT-C CGTATCACGTCCTCCCCC CCCCTGCTCTGCCCCGGTCTCGGGGCGTGTGGCGGGCGATGTCATCTCTGAAA-TCT HVE CGTATCGCGTCGCCACCC C-TCCCATC— CCCCGGTCTCGGGTCGTGGTGGGGGCAATGGCATCTTCGATATTATT-C 280 |-> 300 RHO CCCCCTTGCTGCCCCGATCTCGGGGCGTGCGGGCGGCTGTGTTATCTTCGAAA-TCT RHO CGTATCACGTCGCCCCCA CCCCCTTGCTGCCCCGATCTCGGGGCGTGCGGGCGGCTGTGTTATCTTCGAAA-TCT SIN CGTATCGCGTCGCCCCCA SYS T-TCCCATC— CGTATCGCGTCGCCACCC CCCCGGTCCCGGGTCGTGGTGGGGGCAACAGCATCTTCGATATTATT-C C-TCCCATC— CCCCGGTCTCGGGTCGTGGTGGGGGCGATGGCATCTTCGATATTATT-C TRI CGTATCGCGTCGCCCCCA C-TCCCATC— CCCCGGTCTCGGGACGCGGTGGGGGCAATGGCATCTTCTATATTATT-C Appendix 3.1 Continued 3.1 Appendix OST CGTATCGCGTCGCCCCCA PAR C-TCCCACC— CGTATCGCGTCGCCACCC CCCCGGTCTCGGGCCGTGGCGGGG-CAATGTCATCTTCGATATTATT-C PPS C-TCCCATC— CGTATCGCGTCGCCACCC CCCCGGTCTCGGGTCGTGGTGGGGGCGATGGCATCTTCGATATTAT-AC C-TCCCATC— CCCCGGTCTCGGGTCGCGTTGGGGGCGATGGCATCTTCGATACTATTAC TET CGCATCGCGTCGCCCCCC C-TCCCATC— CCCCGGTCACGGGTGCCGGTGGGGGCAATGGCTTCTTCGATATTATT-C ALT CGCGCCCGTCGTCCCGGTCCCGGGGCGTGCGGGAGGCGGTGCGATCTTGTATGTATACAT CGTCACGCATCGCGTCGC ALT CGCGCCCGTCGTCCCGGTCCCGGGGCGTGCGGGAGGCGGTGCGATCTTGTATGTATACAT FOR CGTATCGCGTCGCCCCCA FOT C-TCCCATC— CGTATTGCGTCGCCACCC CCCCGGTCCCGGGCCGTGGTGGGGGCAATGGCATCTTCGATATTATT-C C-TCCCATC— CCCCGGTCTCGGGTTGTGGTGGGGGCATTGGCATCTTCGATATTATT-C HVI CGTATCGCGTCGCCACCC C-TCCCATC— CCCCGGTCTCGGGTCGTGGTGGGGGCAATGGCATCTTCGATATTATT-C LFO CGCGCCCGCCGCCTCGGTCTCGGGGTGTGCGGGAAGCGGTGCGATCTTGTATGTATACAC CG LFO LSI CGCGCCCGCCGCCTCGGTCTCGGGGTGTGCGGGAAGCGGTGCGATCTTGTATGTATACAC CGCATCGCGTCGCCCCCC CAACCCATCGC C-TCCCATC— CCCCGGTCACGGGTGCCGATGGGGGCAATCGCTTCTTCGATATTATT-C MAT CGTATCGCGTTGCCCCCC MOL C-TCCCATC— CGTATCGCGTCGCCCCCA CCCCGGTCCCGGGTGCTGGTGGGGGCAGTGGCATCTTCGATACTAAT-C C-TCCCATC— CCCCGGTCTCGGGTCGTGGTGGGGGCAGTGGCATCTTCGATATTATT-C NEO CGTATCGCGTCGCCCCCA NOA C-TCCCATC— CGTATCGCGTCGCCCCCA CCCCGGTCTCGGGCCGTGGTGGGGGYAATGTCATCTTCGATATTATTA- C-TCCCATC— CCCCGGTCTCGGGCCGTGGTGGGGGCAATGTCATCTTCGATATTATTCT MYT TGCACCCGCTGCCTCGGTCTCGGGGCCCGCGGGAGSCAGTSCCATCTTCGATACTCT MYT CGTATTGCGTCGCACCCC TGCACCCGCTGCCTCGGTCTCGGGGCCCGCGGGAGSCAGTSCCATCTTCGATACTCT U* U* MAI CGTATCGCGTCGCCCCCC C-TCCCATC— CCCCGGTCCCGGGCGTTGGTGGGGGCAACGGCATCTTCGATATTATT-C vo Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T— GCG -- .... 320 340 360 380 SIN SIN GCG CAAACCCTTTCCCACGTATATCG-TGGCTGGTGGCATTATATGTGGGGGCGGAGATTGGCCTCCCGTGCCC-C— SHA SHA TCG GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— CAAACCCCCGCCCACGCATGCCG-TGATGCGGGGGGTCCT— Taxon Taxon DIM TCG GTGGGGGCGGATATTGGCCTCCCGTGCGC-T— DIR CAA-CCCCCGCACACGCATGCCG-TGATGCGGGGCGTACT— TCG GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— DOT CAA-CCCCCGCCCACGCATGCCG-TGATGCGGGGCGTCCT— TCG GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— CAA-CCCCCGCCCACGTATGCCG-TGATGCGGGGCGTCCT— EXB FOR — GCG ACG AAACCCTGTCCACGG-TGCCCGTGGTGAGGTGCTTTGGGCG GAGCGGATGTTGGCCTCCCGTGCGT-C— FOT CAA— TCG CCCCGCACACGTACACCG-TGGAGAGGGGCATTATATGTGGGGGCGGAGATTGGCCTCCCGTGCAC-C— GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— HVE CAA-CCCCCGTCCACGTATATCG-TGCTGCGGGGCGTCAT— CAA TCG GTGGGGGCGGATATTGGCCTCCCGTGCGCGT— TCCGCACACGTATACCG-TGCTGCGGGGCATCAT— PPS TCG — GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— AACCCCCGCCCACGTACACCG-TGCCGCGGGGCGTCTT— SYS TCG GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— TET CAA-CCCCCGGCCACGTATGCCG-TGATGCGGGGCGTCCT— GCG C— AACCCTGCCCACGCA-AGCCGTGGTGCGGGGCATCGTGCG-GGGG-CGGATACTGGCCTCCCGTGCAC-C— CSP CSP DIC GCG GCG DIS AACCTCGCCCACATCCGACG-TGGTGCGGGGCATTATGCG-GGGG-CGGATATTGGCCTCCCGTGCGA-T— CAA— CCCCGCCCACGTATACCG-TGGCATGGGGCACCATGTGTGGGGGCGGAGATTGGCCTCCCGTGCAC-C— CAA-TCCCCGCCCACGGATCCCG-TGGTGCGTGGCCTT-TCTGTGGGGGCGGATACTGGCCTCCCGTTC EUS EUS GCG CAA— CCCCGCCCACGTATATCG-TGGAGAGGGGCA-CACAGGTGGGGGCGGATATTGGCCTCCCGTGCAC-T— HVI CAA TCG GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— CCCGCACACGTATACCG-TGCTGCGGG-CATCAT— OST GAG CAA— CACCGCCCACGTATATCG-TGGTGCGGGTCATCGTGCGTGGGGGCGGAGATTGGCCTCCCGTGCAC-C— RHO CCG GAGCGGATATTGGTCTCCCGTGCAT-C— CAAAACCTCACCCACGT-TGCTCGTGGTGCGGGGATTTGAGCG TRI CTG CAA— CCCCTCCCACGTATACCG-TGGTGCGGGGCATCATGTGTTGGGGCGGAGATTGGCCTCCCGTGTGC-T— Appendix 3.1 Continued CPA CSI GTG GCG AACCCCGCTCACGTCCAACG-TGGTGCGGGGCATTATGCG-GGGA-TGGATATTGGCCTCCCGTGCGC-T— AACCTCGCCCACGTCCGACG-TGGTGCGGGGCATTATGCG-GGGG-CGGATATTGGCCTCCCGTGCGC-T— ALT ALT CCCCCCGAACCCCGC-CATCCTTT-GGTGGCGTGGGGCTTCGCGGGGAGCGGAGATTGGCCTCCCGTGAACCACGGTG LFO LFO LSI CCCCCCGAACCCCGC-TATCCTTTTGGTGGCGTGGGGCTTTGCAGGGAGCGGAGATTGTCCTCCCGTGAACCATGGTG MAI GCG MAT CCAATCCGGGGGCGGAGATTGCCCTCCCGTGCAC-C— GCG AACCCTTCCCACGAACCCCG-TGGTGATGGGC— GCG MOL AACCCCGCCCGCGTTCGACGTTGGGCCGGGGCACCGTGTG-GGGGGCGGATATTGGCCTCCCGTGCAC-C— CCAAACCCCGCCCGCGCACAGCGTTGGCCGGGGGCATTGTGAG-GGGG-CGGAGATTGGCCTCCCGTGCAC-C— GCG MYT CAA— CCCTTCCCACGCATGCCG-TGGCGAGGGGCATTATATGTGGGGGCGGAGATTGGCCTCCCGTGCAC-C— — TTC GACCCCCGCCCA-ATGTCATCTTGGCGCGGGGCCTCGCGTG GAGCGGATACTGGCCTCCCGTG-GGAT— NOA GCG CAG— CACCGCCCACGTACATCG-TGGTGCGGGGCATCGTATGTGGGGGCGGAGATTGGCCTCCCGTGCAC-T— PAR TCG GTGGGGGCGGATATTGGCCTCCCGTGCAC-T— CAA-CCCCCGCCCACGCATGCCG-TGATGCGGGGCGTCCT— NEO NEO GCG CAA— CGCCGCCCACGTGAATCG-TGGTGCGGGGCTTCATGTGT-GGGGCGGAGATTGGCCTCCCGTGCAC-C— Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ---- CGGACGTCACGATAAGTGGTGGTTTTTAAGCCGCA CAGAACGTCACGATAAGTGGTGGTCTACAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGACCGT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT TAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT TCGAACGTCACGACAAGTGGTGGTTGTTAA-CACCGACCTT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT CGGAACGTCACGACAAGTGGTGGTTGTTAGGCCCCGGCCTC CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCGT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT CAGAACGTCACGATAAGTGGTGGGTGATAAGCCCCGGCCCT CAGAACGTCACGATAAGTGGTGGTTTATAAGCCCCGACCCT CACAACGTCACGATAAGTGGTGGTTTACAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT CG-AACGTCACGACAAGTGGTGGTTGACAGGCCTCGGCCTC CAGAACGTCACGATAAGTGGTGGTTGATAAGCCTCGGCCCT CAGAACGTCACGATAAGTGGTGGTTTATAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTGATAGGCCCCGACCCA CAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT CAGAACGTCACGATAAGTGGTGGT-GATAAGCCCCGGCCCT CAGAACGTCACGATAAGTGGTGGTTTATAAGCCCCGACCCT CAGAACGTCACGATAAGTGGTGGTTTATAAGCCCCGACCCT ------A— AGAACGTCACGATAAGTGGTGGTTAACAACCCCCGATCCG ------ACCAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT ------ACCAGAACGTCACGATAAGTGGTGGTTGATAAGCCCCGGCCCT ------.... 400 400 420 440 460 SHA SHA SIN CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA SYS CAGCGGTTGGCCTAAAAGAGAGCCCCGGGCGA CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCCG TRI TRI CCGCGGTTGGCCTAAAAGTGAGCCCCGGGCGA PPS PPS CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA TET TET CTGCGGTTGGCCTAAAAACGAGCCCCGGGCGA DIS DIS DIM GTGCGGTTGGCTTAAAAACGAGCCCCGAGCG DIR CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA DOT CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA CCGCGGTTGGCCTAAAAGCGAGCCCCGGGC-G FOT CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA HVI CCGCGGGTGGCCTAAAAGCGAGCCCCGGGCGA PAR CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA RHO TTGCGGTTGGCCTAAAAATCGGTCCCGAGCGA CPA CPA CTGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA CSP DIC CTGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA EXB EXB CCGCGGTTGGCTTAAAAGCGAGCCCCGGGCGA OST CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA NEO NEO CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA CSI CSI CTGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA FOR CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA HVE CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA LFO TCGCGGTTGGCTTAAAAGCGTGCCCCAGGCGATGAACGCCAACGTCTCAAGCAGTGGTGGTTACCAAACCCCGGCATC MAT MOL TTGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA NOA CTGCGGTTGGCCTAAAAACGAGCCCCGGGCGA Appendix 3.1 Continued 3.1 Appendix Taxon EUS EUS CCGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA MYT MYT CCGCGGCTGGCCTAAATGCGAGCCCCGAGCGA ALT ALT TCGCGGTTGGCTTAAAAGCGTGCCCCGGGCGACGAACGCCAACGTCTCAAGCAGTGGTGGTTACCAAACCCCGGCATC LSI MAI CTGCGGTTGGCCTAAAAACGAGCCCCGGGCGG TTGCGGTTGGCCTAAAAGCGAGCCCCGGGCGA I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. GGCACAGAGACCCCCGAACGCGTCGCACACG-CGATGT GGCACAGAGACCCCCGAACGCGTCGCATATG-CGGCGT -- -- .... 480 480 500 520 540 TGCGGTGTCGTGCGTTGTCCCGTCGCTCGCT-GGTGCTGCGTGACCCCGATGCTCCGTCGAATGGCGGTGC ------SHA SIN TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCCCGCT-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGATGCSYS TCAA-GTGTCGAGTCGTGCGTTGCGCTGTCGCCTGTT-GGAGCTACGAGACCCC-GATGTGTCGTCCC-GACGATGCTET TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCCCGTT-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGATGC CTCA-GTGTCGTGTCGTGCGTTGCACTGTCGCCCGCTGGAGCCA-CGAGACCCC-GATGCACCGCA-ACCGCGGTGC DIM TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCCCGTCCGGAGCTACGAGACCC— GATGCGTCGTCCC-GACGATGC PPS TCAA-GTGTCATGTCGTGCGTTGCC-TGTCGCTCGTT-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGATGC Taxon FOR TCAA-GTGTCGTGTCGTGCGTTGCGCTGTCGCCTGTC-GGAGCCACGAGACCCC-GATGCGTCGTCCC-GACGATGC HVI TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCTCGTT-GGAGCTACGAGACCCC-GATGCGTCGTCCCCGACGATGC DIC DIS TCGA-GTGTCGTGTCGTGCGTTGTGCTGTCGCCTTCT-GGAGCCACGAGACCC— GATGCGTCGTCCC-CACGATGC RHO AGCG-GTATCGTGTCGTGCGTTTTTCCGTCGCTCGTCGGAGCGAACGAGACCCC-GATGCGTCGC-CCCGACGATGC TRI TCGA-GTGTCGTGTCGTGCGTTTTGCTGTCGCCCGTT-GGAGCTACGAGACCCCCGATGCGTCGTACG-GACGATGC DIR TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCCCGTC-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGATGC NOA OST TAAA-GTGTCGTGTCGTGCGCTGCGCTGTCGCCTGTT-GGAGCCACGAGACCCC-GATGCGTCGTCCC-CACGATGC AAAA-ATGTCGTGTCGTGCGTTGCGCTGTCGCCTGTT-GGAGCTACGAGACCCC-GATGCGTCGTCCC-CACGATGC DOT EUS TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCCCGTT-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGATGCEXB TCAA-GTGTCGTGTCGTGCGTTGCGCTGTCGCCTGTT-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGATGC AGCG-GTATCGTGTCGTGCGTTGCCCCGTCGCTCGTCGGAGCCAACGAGACCCT-GACGCGTCGT-CCCGACGACGCFOT HVE TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCTTGTT-GGAGCTACGAGACCCC-GATGCGTCGTTCC-GACGATGC TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCTCGTT-GGAGCTATGAGACCCC-GATGCGTCGTCCCCGACGATGC CPA CSI TGCA-GTGTCGTGTCGTGCGTTGCACTGTCGCTCGTTGGAGCTA-TGAGACCCC-GATGCGCCGTA-CCGACGGTGCCSP TGCA-GTGTCGTGTCGTGCGTTGCACTGTCGCTCGTTGGAGCTAATGAGACCCC-GATGCGCCGTG-CCGACGGTGC TGCA-GTGTCGTGTCGTGCGTTGCACTGTCGCTCGTTGGAGCTAATGAGACCCC-GATGCGCCGTG-CCGACGGTGC PAR TCAA-GTGTCATGTCGTGCGTTGCCCTGTCGCTCGTT-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGATGC LSI MAI GAAACGTGTCGTGTCGTGCGTTGCACTGTCGCCCGCT-GGAGCCACGAGACCCC-GATGCGCCCCAACC-ACGGTGC TTCA-GTGTCGTGTCGTGCGTCGCGCTGTCGCTCGTCGGAGCTA-CGAGACCCC-GATGCGTCTTG-CTGA-GGCGCMOL MYT TGCA-GTGTCGAGTCGTGCGTTGCGCTGTCGCCTGTC-GGAGCTACGAGACCCC-GATGCGTCGTCCC-GACGGTGCNEO CAAA-GCGTCGTGTCGTGCGTTGTACTGTCGCCTGCC-GGAGCTACGAGACCCC— GTGTCGTATCGTGCGTTGC-CCGTCGCCAACGGGAGCCACACAATACCCCCGTGCGTCGTCATCG-CGACGC GTGCATCTCCCC-CACGATGC Appendix 3.1 Continued 3.1 Appendix LFO GATGATGTTTGCCTTGTGCGTTGCCACGTCGCCCGC ALT GATGATGTTTGCCTTGTGCGTTGCCTCGTCGCCCGC MAT TCCA-GTGTCGTGTCGTGCGTTGCACTGTCGCCCGTTGGAGCCA-CGAGACCCC-GATGCGCCGCG— CCGCGGTGC Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER IV PHYLOGENETIC RELATIONSHIPS OF THE HAMAMELIDACEAE: EVIDENCE FROM MORPHOLOGY AND NUCLEOTIDE SEQUENCES OF ITS AND MATK GENE INTRODUCTION In the Hamamelidaceae, there are 31 genera and more than 140 species (Zhang and Luf 1995). These taxa are rather diverse in morphology. For example, some are trees, while others are shrubs; some species are deciduous, others are evergreen; some are bisexual, others are either andromonoecious or monoecious; some are wind pollinated, while others are insect or bird pollinated. With respect to biogeographic distribution, the hamamelidaceous plants are distributed in southeastern Africa, Madagascar, northeastern Australia, western-, central-, and southeastern Asia, eastern North America, Central America and even northern South America. The fossil record of several hamamelidaceous taxa dates back to the Upper Cretaceous, indicating the long evolutionary history of this family (Endress and Friis, 1991; Zhang and Lu, 1995). Given the ancient diversification of the Hamamelidaceae in both morphology and biogeography, it is not surprising that there has been a great deal of disagreement concerning 199 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the natural classification of the family, especially, since various authors have placed an emphasis on different characters in ascertaining relationships among certain groups of species. Consequently, two, three, four, five, and even six subfamilies have been recognized in the Hamamelidaceae (Reinsch, 1889; Niedenzu, 1891; Harms, 1930; Chang, 1962, 1973; Endress, 1989c). Also controversial are the intergeneric relationships within the largest subfamily, the Hamamelidoideae. The result has been the division of the subfamily into five or four tribes (Harms, 1930; Endress, 1989c). Evidence from several sources, including morphology, palynology, and DNA sequences, etc., contributes to a better understanding of the natural relationships of a particular taxon, but phylogenetic information contained in each data set may differ, thus giving rise to various hypotheses about relationships. Unfortunately, evolutionary history can only recognize one correct phylogeny. Therefore, many relevant data should be taken into account when proposing phylogenetic hypotheses (Tehler, 1995). This is especially important for molecular data, since the phylogeny obtained based on the nucleotide sequences in a single segment of DNA is probably not an organismic tree but a gene tree (Doyle, 1993). In addition, some other biological processes (such as lineage sorting and introgression) may get involved in shaping genetic history, and thus distort some relationships (Ritland 200 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and Eckenwalder, 1992). In contrast, because of different inheritance patterns in nuclear and chloroplast genes, comparison of phylogenies from both data sources may provide insights on the origin of some taxa (Soltis et al., 1996). In this Chapter, therefore, I present a phylogenetic analysis of the Hamamelidaceae using data from morphology, and nucleotide sequences of the ITS (Internal Transcribed Spacers) of nuclear ribosomal DNA, and of the plastid matK gene. MATERIALS AND METHODS Data Sources The morphological characters used in this analysis are those described in Chapter I, and the ITS and matK sequences are those presented in Chapters II and III, respectively. Phylogenetic Analysis There have been great debates over the approaches that should be used in phylogenetic analyses when multiple data sets are available for a taxon (Penny et al., 1982; Cracraft and Mindell, 1989; Kluge, 1989; de Queiroz, 1993; Tehler, 1995). The advocates for the consensus approach argue that this means of analysis gives equal weight to each data set, thus reducing the potential for data sets with relatively large numbers of characters to swamp data with fewer 201 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. characters (Kluge, 1983), and that a conservative estimate should be generated using the consensus method (Hillis, 1987). However, proponents of the combined approach counter that equally weighting data sets of different numbers of characters results in an arbitrary weighting of characters (Cracraft and Mindell, 1989). Consensus trees do not necessarily indicate the most parsimonious pattern of character change (Miyamoto, 1985), and consensus trees can actually contradict combined trees, thus not necessarily showing a conservative pattern (Barrett et al., 1991). However, as noted by de Queiroz (1993), each method is preferred under different circumstances. When different data sets give conflicting trees that cure strongly supported, consensus is preferred over a combined approach. When non independence of characters is suspected within (but not among) data sets, but conflicts among the trees resulting from individual analyses are not strongly supported, consensus should be used in conjunction with a combined analysis. But, when there is not obvious non-independence of characters within data sets, a combined approach is preferred over consensus. As far as defining reasonable limits of support, Kellogg et al. (1996) treated clades as "poorly supported" when their bootstrap values are less than 70%, and corresponding decay indices are one or two. In this study, I applied both the consensus and combined approaches to the data sets. The consensus approach was 202 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. carried out by reconstructing phylogeny from each data set and then determining the consensus tree from all estimates, while the combined method was performed by combining the data sets from the outset. Because the individual data sets of morphology, ITS and matK do not contain exactly the same taxa, I had to reduce the number of taxa so that a common set of taxa is included in the combined data set. Both the morphological analysis using Cercidiphyllum as the outgroup, and the matK gene sequence analysis using two genera of the Saxifragaceae as the outgroups, placed Altinqia in the basal clade of the Hamamelidaceae (See Chapters I and II). In addition, neither the ITS nor the matK sequences of Cercidiphyllum were available for this study. Furthermore, the distinct morphological differences between either Cercidiphyllum or the Saxifragaceae and the Hamamelidaceae cast doubt on the alignability of the ITS sequences between these taxa. Therefore, Altincria was chosen as the outgroup in the combined analysis. Phylogenetic analyses were conducted using the PAUP* test versions 4.0d57-d59 computer program, written by David Swofford of the Smithsonian Institution. The options were Heuristic search, TBR branch swapping, Mulpars on, and Steepest descent off. All characters were unweighted, and all state transformations were unordered. Both bootstrap statistics of 100 replicates (Felsenstein, 1985) and the 203 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Constraint decay analysis (Morgan, 1997) were performed for all data sets to indicate relative supports for individual clades. RESULTS Consensus Approach Strict consensus trees based on each data set are presented in Figs.4.1, 4.2, and 4.3 for morphology, matK, and ITS respectively. When the Consensus approach was applied, the consensus tree of neither morphology+ITS nor morphology+matK resolved many relationships (Figs.4.4, 4.5); however, ITS+matK consensus resolved most of the intergeneric relationships of the Hamamelidaceae (Fig.4.6). There is little resolution in the consensus of morphology+matK+ITS (Fig.4.7). Combined Analysis When the Combined approach was used, the most parsimonious phylogenies of morphology+ITS and morphology+ITS+matK resolved all the intergeneric relationships (Figs.4.8, 4.11), while those of morphology+matK and ITS+matK showed slightly less resolution for intergeneric relationships (Figs.4.9, 4.10). Figure 4.11 is the single tree generated in the combined analysis of morphology+ITS+matK . In this phylogeny, Liquidambar is the 204 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. first clade following the outgroup Altinqia. Exbucklandia and Rhodoleia are clustered in one clade, followed by two paraphyletic clades of Mvtilaria and Disanthus. Within the Hamamelidoideae, the basal clade is Corylopsis-Loropetalum- Mainqaya-Matudaea. followed by a dichotomy of Hamamelis- Fothergilleae and Eustigmateae-Dicoryphinae. The Dicoryphinae Endress forms a monophyletic group, within which Dicorvphe and Trichocladus are of a clade, and its sister clade contains the three Australian genera (Neostrearia, Ostrearia. and Noahdendron). The Eustigmateae sensu Endress (1989c) is a clade, with the inclusion of Molinadendron. Hamamelis is basal in the clade of Hamamelis-Fothergilleae. within which Fotherqilla. Parrotiopsis. Parrotia, and Shaniodendron form a consecutive paraphyletic pattern, while Distylium. Distyliopsis. and Sycopsis are located on the top of the clade. DISCUSSION Consensus Approach Reduces Resolution The trees based on the two DNA sequence data sets (ITS and matK) are mostly congruent to each other (Figs.4.2, 4.3), thus producing a consensus that resolves some relationships of the Hamamelidaceae (Fig.4.6). However, the tree topological patterns based on either of the two DNA sequence data sets rarely agree with those generated using 205 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. morphological data, resulting in a consensus tree of little resolution (Figs.4.4, 4.5, 4.7). This is a drawback of the consensus approach for the data sets. If bootstrap values less than 70% are considered as weak supports (Kellogg et al., 1996), the clades in the morphology-based phylogeny are not well supported except for several closely related genus-pairs (Fig.4.1). As a result, when applied in the consensus approach, which does not differentiate the well-supported and poorly-supported topologies, morphological data offset the resolving power from the other data sets, resulting in a poor overall resolution for the Hamamelidaceae. Thus, the combined approach should be preferred in this situation, as recommended by de Queiroz (1993) and Kellogg et al. (1996). However, some relationships are recognizable when the phylogenetic estimates from each data set are compared. For instance, Eustiama. Fortunearia. and Sinowilsonia are clustered in one group in all three trees (Figs.4.1, 4.2, 4.3), supporting the recognition of the Eustigmateae sensu Endress (1989c). The Combined Approach Increases Resolution Parsimony analyses were conducted for different combinations of the three data sets, including ITS-morphology (Fig.4.8), matK-morpholooy (Fig.4.9), ITS-matK (Fig.4.10), and ITS-matK-morphology (Fig.4.11). It is readily noticeable 206 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that well-supported relationships in one data set overperform the weakly-supported ones in another. For example, the weakly supported relationships produced by morphological data, such as Corylopsis-Hamamelis. and Matudaea- Molinadendron. are rarely shown when the morphology data set is combined with either ITS or matK data, both of which have demonstrated much stronger clade supports. Interestingly, the morphological data set produces few well-supported clades, but it increases resolution for some clades when combined with either the ITS or matK data. For example, in the ITS tree (Fig.4.3), the relationships of Fothergilla. Hamamelis, and Parrotiopsis remain unclear, while the analysis using the combined data set of ITS and morphology resolves the relationships of the genera (Fig.4.8). It seems to be appropriate to combine data sets when there are no conflicting hypotheses that are well-supported by different, independent data sets (Kellogg et al., 1996). However, I believe that performing both the consensus and combined analyses has several advantages. First, the consensus approach often produces highly unresolved phylogeny, but when it does resolve a relationship, greater confidence should be placed in the reality of this pattern, especially when the analysis involves data from multiple independent sources. Second, the combined analysis generally produces a phylogeny with a high resolution when data sets 207 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. are not obviously conflicting to one another. Third, when individual analyses are conducted, and the phylogenetic trees are compared, the discrepancies may shed some new light on biological processes leading to the differentiated pattern, such as lineage sorting and chloroplast capture (Soltis et al., 1996a). Implications of the topological differences between phyloqenies based on nuclear and chloroplast DNA sequences In the ITS-based phylogeny, Parrotia is located next to Sycopsis. suggesting a close relationship (Fig.4.3); however, in the matK tree, Parrotia is the basal taxon of the Fothergilleae, but Sycopsis is at the top of the clade, far away from Parrotia (Fig.4.2). These two genera are known to be interfertile, producing a fertile hybrid (=XSycoparrotia. Endress and Anliker, 1968). This is consistent with their close relationship, as shown in the ITS phylogeny. Parrotia and Svcopsis also share some morphological characteristics, especially floral structures (Bogle, 1968; Endress and Anliker, 1968). Since morphology is not consistent with the chloroplast gene-based tree, and there is a strong differentiation of the matK gene between Parrotia and Sycopsis. it does not make sense to use chloroplast capture to explain the differences in the phylogenetic patterns of ITS and matK for the two genera (Soltis et al., 1996a). Alternatively, lineage sorting may be the cause of the 208 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pattern (Rieseberg and Brunsfeld, 1992; Soltis and Soltis, 1995; Soltis et al., 1996a). High polymorphism of the chloroplast genome may have developed in the common ancestor of these two genera, and their geographical separation since the Late Miocene (Deng et al., 1992b; Zhang and Lu, 1995) may have provided an opportunity for further accumulation of base substitutions. Shaniodendron (Chang) Deng, Wei & Wang is the most recently described segregate genus in the Hamamelidaceae (Deng et al,. 1992a). It is endemic to the eastern ranges of Dabieshan mountains and the northern parts of the Tianmushan mountains of eastern China (Deng et al., 1992a, b; Hao et al., 1996). Because of its rarity, Shaniodenriron has been added to the list of the most endangered plant species in China (Deng et al., 1992a, b). Deng et al. (1992a) suggested that this genus was more closely related to Fothergilleae (incl. Fotherailla. Parrotia. Parrotiopsis) than to Distylieae (incl. Distylium. Matudaea. Molinadendron. and Sycopsis) (See Harms, 1930; Endress, 1989a, b ) . Recently- conducted comparative studies of floral morphology (Hao et al., 1996) and wood anatomy (Fang and Deng, 1996) support Deng et al.'s placement (1992a) of Shaniodendron in the Fothergilleae. Furthermore, the study of floral morphology led Hao et al. (1996) to the conclusion that Shaniodendron is most closely related to Parrotia. In the ITS phylogeny (Fig.4.3), Shaniodendron is placed 209 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. between Parrotia and Sycopsis, while in the matK tree (Fig.4.2), it is closely allied within the clade consisting of Sycopsis and Distylium. but separated from Parrotia. Shaniodendron shares morphological characters with both Sycopsis and Parrotia. but especially with Parrotia. including deciduous leaves, andromonoecy, and elongated stamens. Therefore, the morphological data support the relationships among the three genera shown in the ITS-based phylogeny. As stated above, Sycopsis and Parrotia are able to hybridize, producing a hybrid, XSvcoparrotia. Interestingly, Shaniodendron shows many somewhat intermediate morphological characteristics between Parrotia and Sycopsis, as in parallel with the hybrid genus XSycoparrotia, such as middle-sized lanceolate stipules, slightly elongated anthers, shallow hypanthium tubes, and semi-inferior ovaries (See Deng et al., 1992a, b; Fang and Deng, 1996; Hao et al., 1996). Therefore, it is possible that Shaniodendron could be derived from an intergeneric hybridization event between species of Parrotia and Sycopsis. Furthermore, in the maternally inherited matK gene phylogeny, Shaniodendron is closely related to Sycopsis. but distant from Parrotia. One could interpret this as an indication that Sycopsis might be the maternal parent for this putative hybrid. Parrotia is a monotypic genus endemic to the Caspian Sea region of central Asia (Zhang and Lu, 1995). Shaniodendron is also a monotypic taxon, but narrowly distributed in 210 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. eastern China (Deng et al., 1992a, b; Zhang and Lu, 1995). In contrast, Sycopsis is widely distributed in southeastern Asia from southern Vietnam to central areas of China (Chang 1979; Zhang and Lu, 1995). Therefore, Parrotia is disjunct from both Shaniodendron and Sycopsis. Moreover, the divergence time of Parrotia and Shaniodendron was estimated to be in the late Tertiary (Li et al., 1997d). Thus, the hypothesized hybrid origin of Shaniodendron. if true, must be an ancient event. Phylogenetic relationships in the Hamamelidaceae Most of the relationships suggested by this combined analysis are similar to the ones shown in previous analyses using individual data sets of the matK gene (Fig.2.2, Chapter II) and ITS (Figs.3.3, 3.4, 3.5, 3.6, Chapter III). These relationships are as follows: 1) the monophyly of several groups including the Altingioideae, Exbucklandia-Rhodoleia. and the Hamamelidoideae; 2) polyphyly of Exbucklandioideae sensu Endress (1989c); 3) the Hamamelidoideae containing three major branches: Corylopsis-Loropetalum-Mainqava- Matudaea: Dicoryphinae-Eustigmateae sensu Endress (1989c), but including Molinadendron: and Fothergilleae sensu Endress (1989c), but including Hamamelis. In the combined analysis Hamamelis was placed in the position sister to the Fothergilleae clade. This relationship is morphologically reasonable. 211 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Evolution of Some Morphological Characters The Hamamelidaceae are rather diverse morphologically, thus resulting in controversies about a correct natural classification of the family. A phylogenetic analysis based only on morphological data produced trees with a small consistency index (Cl) of 0.45 (Chapter I), indicating high homoplasy of morphological characters in the Hamamelidaceae. Therefore, a combined analysis of DNA sequence data and morphological data may shed some light on the evolution of some morphological characters, especially the ones that are commonly used in defining groups in previous classification systems. Palmately lobed leaves occur in several paraphyletic clades of the tree based on the combined data set (Fig.4.12), including Liquidambar. Mvtilaria. and Exbucklandia. Therefore, the most parsimonious interpretation of the distribution of the character is that it represents a plesiomorphic feature. In the Hamamelidoideae, most of the genera have only one ovule in each locule of the two-carpelled pistil, while in several genera two to three ovules have been observed per locule, yet only one is fertile and becomes seed. The one- seeded condition is functionally correlated with explosive seed dispersal (Endress 1989b). But in other taxa, such as Exbucklandia. Rhodoleia, Mytilaria. and Disanthus, there are several to many ovules in each locule. Accordingly, these 212 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. genera do not have the ballistic seed discharge mechanism in the fruit. Endress (1989b) described this mechanism in detail and concluded that a one-seeded pistil with explosive seed discharge is more advanced. The combined phylogeny (Fig.4.13) shows this directionality of ovule evolution in the Hamamelidaceae, and that the one-seeded carpel is a synapomorphy of the Hamamelidoideae. The tendency toward wind pollination has been proposed to be a driving force of character evolution in the Hamamelidaceae, such as loss of petals and change of sexuality from bisexual via andromonoecious to monoecious (Tong, 1930; Bogle and Philbrick, 1980). The phylogenetic analysis (Fig.4.14) reveals that loss of petals is distributed in at least five clades in the Hamamelidaceae, including the Altingioideae, Exbucklandia. Fothergilleae, Matudaea. and Molinadendron. Forcing the character to become a single synapomorphy increases the number of tree steps greatly compared to the most parsimonious trees. Therefore, parallel evolution may have occurred in this character state. The same appears to be true for wind pollination, andromonoecy and monoecy (Figs.4.15, 4.16). Within the Hamamelidoideae, three lineages can be recognized (Fig.4.11). Interestingly, each lineage exhibits similar evolutionary trends in certain floral characters adapting to wind pollination. In the first lineage, the bisexual flower is a synplesiomorphy, while petals have 213 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. evolved in the direction of loss from Loropetalum to Matudaea (Fig. 4.14). In the second lineage, flowers change from bisexual in Eustiama and Molinadendron. to andromonoecious in Fortunearia. and monoecious in Sinowilsonia (Fig.4.16). In the third lineage, flowers evolve from insect pollination in Parrotionsis and Fotheroilla to wind pollination in Parrotia, Shaniodendron. Distylium. Distyliopsis. and Svcopsis (Fig.4.15). Possession of showy structures other than petals (stamens in Fotheroilla and inflorescence bracts in Parrotiopsis) has been considered as secondarily evolved in the course of evolution (Endress, 1989b). However, Fotheroilla and Parrotiopsis are located in the basal position of the clade that also includes Distyliopsis, Distylium. Sycopsis. and Shaniodendron. Therefore, these characteristics may be secondarily evolved but in conjunction with the plesiomorphic insect pollination. The haploid chromosome numbers (n) in the Hamamelidaceae exhibit two basic series: 8, 16, 32, and 12, 24, and 36 (Santmour, 1965, 1972; Goldblatt and Endress, 1977; Morawetz and Samuel, 1989; Oginuma and Tobe, 1991). Goldblatt and Endress (1977) suggested that x=7 was the basic chromosome number in the Hamamelidaceae, and that later descending aneuploidy resulted in x=6, which, via polyploidy, gave rise to n=12 group (Rhodoleia and Hamamelidoideae), whereas later ascending aneuploidy produced x=8, which also via subsequent 214 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. polyploidy led to the n=8 alliance (Altingioideae and Exbucklandioideae). However, they admitted that it is also possible that either x=8 or x=6 is the basic chromosome number. Morawetz and Samuel (1989) considered x=8 as basic in the Hamamelidaceae, and suggested that 2n=24 was of triploid origin from x=8 via anorthoploidization. Oginuma and Tobe (1991) agreed with both Goldblatt and Endress (1977) and Morawetz and Samuel (1989), and suggested that x=8 is basic to an evolutionary line of Exbucklandioideae and Altingioideae, while x=6 is basic in the Hamamelidoideae and Rhodoleioideae. Interestingly, a new base number, x=13 (2n=26), was first reported in Liquidambar acalycina (Huang et al., 1985) [no photographs were provided in the paper, thus it needs to be confirmed]. This number (2n=26) was also recently reported in Mytilaria (Pan and Yang, 1994), an isolated genus in the Exbucklandioideae (Endress, 1989c). In the phylogeny based on the combined data set (Fig.4.17), x=8 occurs in the most basal clade, the Altingioideae, followed by the clade of Exbucklandia (x=8) and Rhodoleia (x=12). In the succeeding clades, Mytilaria has x=13, which probably is derived from x=12 via ascending aneuploidy, while Disanthus exhibits the base number of x=8. However, the Hamamelidoideae share x=12 with Rhodoleia. This phylogeny does not support the two evolutionary-line hypothesis (Oginuma and Tobe, 1991). In the evolutionarily more recent clade, the Hamamelidoideae, x is invariably equal 215 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to 12. Therefore, x=8 is probably the ancestral base number, while x=12 is a base number of possibly ancient polyploid origin. The close relationship between Exbucklandia (x=8) and Rhodoleia (x=12) in the phylogenetic analysis seems to support the suggestion that x=12 is of triploid origin of x=8 via anorthoploidization (Morawetz and Samuel, 1989), since it is difficult to believe that a series of cytological events has taken place, including descending and ascending aneuploidy, and polyploidization, in this strongly supported clade. The further ploidy levels, n=24 and 36, may be explained as having been derived directly from the ancient base number, x=12, via polyploidy. 216 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.1. Strict consensus of 203 most parsimonious trees of 185 steps based on morphology of the Hamamelidaceae. CI=0.5, RC=0.31. Branches have bootstrap percentage of less than 50% and decay values of one unless otherwise indicated above and below the branches respectively. 217 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia Corylopsis Hamamelis Dicoryphe Trichocladus Loropetalum Maingaya 2 Neostrearia Noahdendron Ostrearia Disanthus Distylium Distyliopsis Eustigma { Fortunearia Sinowilsonia Exbucklandia Fothergilla Matudaea t Molinadendron Mytilaria Parrotia Parrotiopsis Rhodoleia Shaniodendron Sycopsis Liquidambar 218 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.2. Strict consensus of six most parsimonious trees of 381 steps based on sequences of the matK gene in the Hamamelidaceae. CI=0.87, RC=0.72. Numbers above and below branches are bootstrap percentages and decay indices. 219 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia Corylopsis 100 611---- Loropetalum 96 Matudaea 2 i Maingaya Dicoryphe Neostrearia 58 71 Noahdendron 1 1 ---- Ostrearia 81 Trichocladus 69J— Eustigma 89 5 | 85 Fortunearia 2 62 r “ Molinadendron 2 1--- Sinowilsonia 70 1---- Distylium 92 1 1____ Sycopsis 1 4 62 Distyliopsis Shaniodendron Fothergilla <50 Parrotiopsis 95 Hamamelis Parrotia 100 Disanthus Mytilaria 100 Exbucklandia ■ c Rhodoleia Liquidambar 220 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.3. Strict consensus of 16 most parsimonious trees of 6952 steps based on the ITS grand data set of the Hamamelidaceae. CI=0.6, RC=0.36. Numbers above and below branches are bootstrap percentages and decay indices. 221 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia Corylopsis 100 Loropetalum 17 100 “l2\ 82 Maingaya Matudaea <50 Dicoryphe 55 Trichocladus Neostrearia 100 100 Noahdendron 100 Ostrearia Eustigma 100 Fortunearla 14 10 Molinadendron Sinowilsonia 100 99 Distylium 96 11 Distyliopsis 100 Parrotia 15 51 Shaniodendron Sycopsis 64 M Hamamelis 100 Parrotiopsis Fothergilia 100 20 Disanthus Mytilaria 100 Exbucklandia Rhodoleia Liquidambar 222 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.4. Strict consensus of morphology and ITS of the Hamamelidaceae obtained using the Consensus approach. 223 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia Corylopsis Dicoryphe Disanthus Distylium Distyliopsls Eustigma Fortunearia Exbucklandia Fothergilla Hamamelis Loropetalum Maingaya Matudaea Molinadendron Mytilaria Neostrearia Noahdendron Ostrearia Parrotia Parrotiopsis Rhodoleia Shaniodendron Sinowilsonia Sycopsis Trichocladus Liquidambar 224 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.5. Strict consensus of morphology and matK of the Hamamelidaceae obtained using the Consensus approach. 225 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia Corylopsis Dicoryphe Disanthus Distylium Distyliopsis Eustigma Fortunearia Exbucklandia Fothergilla Hamamelis Loropetalum Maingaya Matudaea Molinadendron Mytilaria Neostrearia Noahdendron Ostrearia Parrotia Parrotiopsis Rhodoleia Shaniodendron Sinowilsonia Sycopsis Trichocladus Liquidambar 226 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.6. Strict consensus of ITS and matK of the Hamamelidaceae obtained using the Consensus approach. 227 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■■ ■■ Altingia — — Corylopsis — —— Loropetalum — ~ Maingaya — — Matudaea ■■ Dicoryphe — Neostrearia — — Noahdendron 11 11 ■ Ostrearia — — Trichocladus — Eustigma 11 Fortunearia — — Molinadendron — — Sinowilsonia —— Distylium ■— — Distyliopsis — 1— Fothergilla — — Hamamelis " Parrotia — — Parrotiopsis — — Shaniodendron ■ Sycopsis 1 Disanthus — — Mytilaria | Exbucklandia ' Rhodoleia — Liquidambar 228 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig. 4.7. Strict consensus of morphology, ITS and matK of the Hamamelidaceae obtained using the Consensus approach. 229 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia Corylopsis Dicoryphe Disanthus Distylium Distyliopsis Eustigma Exbucklandia Fortunearia Fothergilla Hamamelis Loropetalum Maingaya Matudaea Molinadendron Mytllaria Neostrearia Noahdendron Ostrearia Parrotia Parrotiopsis Rhodoleia Shaniodendron Sinowilsonia Sycopsis Trichocladns Liquidambar 230 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.8. Strict consensus of two shortest tree of 7161 steps based on the combined data set of morphology and ITS of the Hamamelidaceae. CI=0.6, RC=0.36. Numbers above and below branches are bootstrap percentages and decay indices. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae. 231 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia — ALToideae Corylopsis _ C0Reae Loropetalum LORinae (HAMeae) 14 75 Maingaya '] Mafcuda&a — FOTeae Dicoiryphe — Trxchocladus DICinae Neostrearla (HAMeae) Noahdendron Ostrearia _ (Q 232 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.9. Strict consensus of eight shortest trees of 590 steps based on the combined data set of morphology and matK of the Hamamelidaceae. CI=0.72, RC=0.49. Numbers above and below branches are bootstrap percentages and decay indices. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae. 233 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 Altingia — ALToideae Corylopsis — coReae 90 Loropetalum -j Maingaya _1 (HAMeae) Matudaea — FOTeae Dicoryphe Neostrearia DICinae 83 77 Noahdendron (HAMeae) Ostrearia Trichocladus— Distylium — 100 Distyliopsis 8| 57 Sycopsis 1 78 Shaniodendron 6 <50 Parrotia 1 52 Parrotiopsis 1 84 Fothergilla— 70 3 Hamamelis — •HAMinae (HAMeae) Eustigma EUSeae Foxrtunearia 95 60 d 234 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.10. Single shortest tree of 7338 steps based on the combined data set of ITS and matK of the Hamamelidaceae. CI=0.62, RC=0.38. Numbers above and below branches are bootstrap percentages and decay indices. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae , EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamame1idoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae. 235 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Altingia — ALToideae Corylopsis — coReae Loropetalum L0Rinae Maingaya J (HAMeae) Matuda&a — FOTeae Dicoryphe — Trichocladus DICinae Neostrearia (HAMeae) Noahdendron Ostrearia _ Eustigma Fortunearia EUSeae Molinadendron -FOTeae Sinowilsonia _ Distylivm — Distyliopsis Fothergilla OJ 0) 5C « a) Hamamelis — Parrotiopsls — Disanthus (d0) d) Mytilaria •HTJ O CQ Exbucklandia —I w* Rhodoleia RHOideae Llquldambar — ALToideae 236 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.11. Single shortest tree of 7546 steps based on the combined data set of morphology, ITS and matK of the Hamamelidaceae. CI=0.61, RC=0.37. Numbers above and below branches are bootstrap percentages and decay indices. Groupings on the right side follow Endress (1989c). ALToideae=Altingioideae, EXBoideae=Exbucklandioideae, RHOideae=Rhodoleioideae, HAMoideae=Hamamelidoideae, COReae=Corylopsideae, HAMeae=Hamamelideae, HAMinae=Hamamelidinae, FOTeae=Fothergilleae, EUSeae=Eustigmateae, LORinae=Loropetalinae, DICinae=Dicoryphinae. 237 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 'Alti.ngi.6i — ALToideae Cory lops is — coReae I O r ° P e t a I u m -|L0R i„ . e Maingaya -I Matudaea — FOTeae Dicoryphe — Trichocladus DXCinae Neostrearia (HAMeae) Noahdendron Ostrearia Eustigma id(U Fortunearia a) •H•a Molinadendron o Sinowilsonia— Distylium — Distyliopsis (00) Sycopsis e o Shaniodendron fa Parrotia Paurrotiopsis v d) 238 I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.12. Single most parsimonious tree found in the analysis of the combined data of morphology, ITS and matK, showing distribution of character states of leaf form in the Hamamelidaceae. 239 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Liquidambar Shaniodendron Hamamel is Hamamel Trichocladus Parrotiopsis Ostrearia Fothergilla Eustigma Sinowilsonia Dicoryphe Noahdendron Molinadendron Neostrearia is Corylops Loropetalum Maingaya Mytilaria iExbucklandia \Disanthus i nDistylium uDistyliopsis nFortunearia a mParrotia a a o aMatudaea in in jaAltingia ■■ Sycopsis in in in mRhodoleia □a □n l=in jpzin in "I*— _r— in [fE * cm K> O Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.13. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morophology, ITS and matK. showing distribution of the number of ovules per carpel in the Hamamelidaceae. 241 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sycopsis Shaniodendron Liquidambar Parrotiopsis Trichocladus Parrotia Hamamelis Sinowilsonia Eustigma Fortunearia Dicoryphe Ostrearia Molinadendron Neostrearia Noahdendron Corylopsis Loropetalum Disanthus Maingaya Exbucklandia Mytilaria Rhodoleia Matudaea \Distylium iDistyliopsis i i \Fothergilla aAltingia i i to to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.14. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK, showing distribution of the character states of petals in the Hamamelidaceae. 243 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. q q tn O 01 O (Q 0} q Vh q <0 M *H 3 q •H <0 01 XJ oi rH -H XJ t? XJ •H XI § Q< q Q, -h oi ^ q o oi q Ih q q q q O oi qqo-H-wqqqoi •q m q -H q q q H ^ *H -H XI tn «s XJ q (h & XJ M M 01 o U 01 1 § <0 'H q q q q q -H ft "H 0) E q -q in Oi XI +J +J o C •H O OII O -w O12. 01 01 o (0 q O *H —I -H •H 'H S, .q q *0 a -h -h w q q Q Q tQ to S! to Q Eh BBBBBBBBQBIBBBQBBBB T Petal unordered lyl&yl auriculate dorsaly orbicular long clawed orbicular ribbon-like coiled circinately ribbon-like but not coiled ribbon-like and fleshy ■ ■ ribbon-like with broad base ■ ■ reduced ■ ■ absent fevAvt uncertain I i equivocal 244 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.15. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK. showing distribution of pollination types in the Hamamelidaceae. 245 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sycopsis Distylium Diatyliopsis Shaniodendron Trichocladus Parrotia Parrotiopsis Fothergilla Hamamelis Sinowilsonia Altingia Liquidambar Dicoryphe Ostrearia Neostrearia Noahdendron Corylopsis Molinadendron Loropetalum Disanthus Exbucklandia Rhodoleia Maingaya Matudaea Mytilaria \Eustigma iFortunearia I I bird P. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.16. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK. showing distribution of character states of sexuality in the Hamamelidaceae. 247 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a lari Shaniodendron Distyliopsis Parrotia Altingia s i Hamamel Trichocladus Parrotiopsis Fothergilla Sinowilsonia Dicoryphe Ostrearia Eusti gma Eusti Fortunearia Neostrearia Molinadendron Corylopsis Noahdendron Disanthus Loropetalum Maingaya Matudaea Rhodoleia Myt i i Myt i nDistylium aSycopsis mLiquidambar in in i o o a in in v d m in L = . r— in l UzzaaExbucklandia j = i n ~~|Lmn PM L h to a> c oi to ss a p> a a> H- H H- 2 H cr (0 a> 3 3 h H- 0 3 (0 2! o O 3 n> o 1 d o 3 0 30 a r H- o e(0 O K> 00 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fig.4.17. Strict consensus of the two most parsimonious trees found in the analysis of the combined data of morphology, ITS and matK, showing distribution of character states of chromosome base numbers in the Hamamelidaceae. 249 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sycopsis Shaniodendron Liquidambar Hamamelis Altingia Trichoc ladusTrichoc Parrotia Fothergilla Dicoryphe Fortunearia Parrotiopsis Neostrearia Corylopsis Eustigma Noahdendron Mol i nadendron i Mol Maingaya Matudaea Exbucklandia Mytilaria Rhodoleia i i aSinowilsonia aOstrearia nDistylium a uDisanthus i i \nLoropetalum •saa •saa — in hzaaDistyliopsis Rssa Rssa razan razan fclrmn fclrmn V j ■'■ -jtri ■'■ j :|tinna :|tinna L L nw unordered rsj cn o Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER V TRANSLATING PHYLOGENY OF THE HAMAMELIDACEAE INTO A CLASSIFICATION SYSTEM INTRODUCTION Since the publication of The Origin of Species by Charles Darwin in 1859, taxonomists have been pursuing classifications that reflect the natural history (phylogeny) of organisms. In recent decades, the development of computer technology and molecular systematics, together with improved techniques for classical studies of morphology, embryology and floral ontogeny, etc. has rejuvenized the field of plant systematics. A number of explicit phylogenetic hypotheses for some plant groups have been proposed to be translated into classification systems (Cantino et al., 1995; Donoghue, 1995; Morgan, 1995; Olmstead, 1995). The exercise of making the necessary taxonomic and nomenclatural changes has been urged for systematists conducting phylogenetic analyses (Judd, 1995). For the Hamamelidaceae, the most recently proposed classification system has largely improved the previous systems in terms of recognizing monophyletic groups (Endress, 1989c); however, my phylogenetic analyses (Chapter IV) based on both morphological and molecular data have shown that some groups are still polyphyletic. In order to reflect 251 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. new understandings of the natural relationships in the Hamamelidaceae, I present in this chapter a new classification system for the Hamamelidaceae. MATERIALS AND METHODS The combined analysis using data from morphology, ITS and matK DNA sequences was conducted as in Chapter IV. The recognition of only monophyletic groups as taxonomic units is widely accepted by systematists; however, there are alternative ways to formally name these groups (Kron, 1995). One approach is the sequencing convention (Wiley et al., 1991), in which sister groups are considered "equivalent" and given the same ending, thus allowing familiar names and endings to be applied. However, the proliferation of names and the possible "loss" of some well-known names becomes a concern, particularly in regions of the tree that are highly asymmetric (pectinate). According to de Queiroz and Gauthier (1995), a phylogenetic systematic approach names clades without the restrictions imposed by the Linnaean hierarchical approach. However, abandonment of the Linnaean hierarchy obscures the inherently hierarchical nature of genealogy, thus creating difficulties in communicating the known phylogenetic relationships (Kron, 1995). Therefore, following the belief that the goal of an evolutionary taxonomist is to develop a classification system that 252 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. recognizes monophyletic groups, but disrupts the present classification as little as possible (Funk, 1985), I translate the combined phylogeny (Fig.4.11) into an ordered and phylogenetically annotated classification system for the family, recognizing monophyletic clades, and keeping well- known names intact whenever it is reasonable to do so. The classification system starts from the basal clade up, with each taxon followed by a brief note indicating its phylogenetic position. RESULTS The phylogeny produced in Chapter IV, based on the combined data set, resolved the intergeneric relationships of the Hamamelidaceae (Fig.4.11). Six monophyletic groups, herein treated as subfamilies, are recognized, including Altingioideae, Rhodoleioideae, Exbucklandioideae, Mytilarioideae, Disanthoideae, and Hamame lido ide ae. Within the Hamamelidoideae, six tribes are recognized. Tribe 1. Corylopsideae, Tribe 2. Loropetaleae, Tribe 3. Eustigmateae, Tribe 4. Hamamelideae, Tribe 5. Fothergilleae, and Tribe 6. Dicorypheae. DISCUSSION It is rarely disputable that defining monophyletic groups should be the mission of taxonomy, and that 253 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. phylogenetic patterns should be embodied in the classification system. In the Hamamelidaceae, several monophyletic groups are recognized in the present study, including Altinaia-Semiliquidambar-Liquidambar. Exbucklandia Rhodoleia, Mytilaria. Disanthus. and the subfamily Hamamelidoideae. The latter group, the Hamamelidoideae, contains three major clades: 11 Corvlopsis-Loropetalum- Maingaya-Matudaea. 2) Eustiama-Fortunearia-Mo1inadendron- Sinowilsonia-Dicoryphinae. and 3) Hamamelis-Fotherqilleae. Some of the monophyletic groups share one or several morphological character states, while others do not. When a clade does not have a common morphological characteristic(s ) treating the major clade as a taxon of whatever rank is not reasonable in terms of communicating phylogeny among systematists. Therefore, it may be preferable to split a monophyletic group so that a practicable definition can be made (Bremer, 1978). Accordingly, I divided certain clades into two (or more) equally-ranked taxonomic units; furthermore, their close relationships are indicated in the classification system by explicitly stating their sister relationships. Altinqia-Semilicruidambar-Liauidambar These three genera are distinct from the rest of the members of the Hamamelidaceae in many morphological characters and nucleotide substitutions (See Chapter I, II, 254 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ill), thus there is no doubt that they are a monophyletic group. However, the controversy has been whether they should be separated from the rest of the Hamamelidaceae and be recognized as an individual family (Nakai, 1943; Melikian, 1973; Huang and Lee, 1982; Pan et al., 1990; Wang, 1992; Zhang and Lu, 1995; Takhtajan, 1997). Molecular studies have shown that Liquidambar and Altingia are closely knit with the other Hamamelidaceae (Hibsch-Jetter, 1996, and pers. comm.; Hoot and Crane, 1996; Hoot et al., 1997). Therefore, these three genera can be treated either as a subfamily Altingioideae, or as a separate family Altingiaceae closely related to the Hamamelidaceae. However, I prefer to retain them in the Hamamelidaceae because many of the charateristics that have been used to define them as a family Altingiaceae can also be found in some genera of the Hamamelidaceae, such as monoecy, polyporate pollen grains, and apetaly. Exbucklandioideae and Rhodoleioideae Exbucklandia and Rhodoleia have long been treated as monogeneric subfamilies (Harms, 1930; Chang, 1948, 1979). However, Endress (1989c) expanded the Exbucklandioideae to include not only Exbucklandia. but Disanthus, Mytilaria. and Chunia as well, based on their having 5-8 ovules per carpel and their possession of persistent stipules. In contrast, the present phylogeny associates Exbucklandia with Rhodoleia, supporting the treatment of putting the two genera into one 255 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subfamily proposed by Reinsch (1889). However, these two genera are rather different morphologically, even though both have complex capitulate inflorescences. Leaves are palmate- veined in Exbucklandia. but pinnatly-veined in Rhodoleia: flowers are andromonoecious, rarely have petals, are symmetric, and possibly wind-pollinated in Exbucklandia. while in Rhodoleia. flowers are bisexual, have asymmetrically distributed showy petals (usually on the abaxial side only), and are bird pollinated. In addition, the chromosome base number is x=8 in Exbucklandia and x=12 in Rhodoleia. Therefore, Exbucklandia and Rhodoleia are treated herein as closely related individual subfamilies. Mytilarioideae Mytilaria was considered as a little known genus by Harms (1930), and not included in his system. Chang (1948) described the genus Chunia, and erected a subfamily, Mytilarioideae, for Mytilaria and Chunia. These two genera are rather similar in leaf morphology, spadix inflorescence (Chang, 1979), presence of resin canals (Huang, 1986), and multilacunar nodal anatomy (Bogle, 1990, 1991). Unfortunately, material of Chunia was not available for molecular study (the genus is rare, and endemic to Hainan). Mytilaria. Chunia, Disanthus. and Exbucklandia have recently been combined into a subfamily Exbucklandioideae by Endress, (1989c). However, the paraphyletic relationships among the 256 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. genera do not support this union of the four genera. Instead, Mytilaria forms its own clade. Therefore, the Mytilarioideae should be recognized as a subfamily. Interestingly, karyotype study has shown that Mytilaria has a chromosome base number of x=13 (the chromosome number in Chunia is unknown), which is different from any other Hamamelidaceae (x=13 has also been reported in Liquidambar (Huang et al., 1985, but no photograph was provided, and the report could be spurious), thus supporting the recognition of this genus as a subfamily. The systematic position of Chunia needs further study, but I place this genus in the subfamily Mytilarioideae for the time being, based on their morphological similarities. Disanthoideae The paraphyly of Disanthus with Mytilaria and Exbucklandia does not provide support for the expanded Exbucklandioideae sensu Endress (1989c). Instead, Disanthus forms its own clade. Disanthus is specialized in many morphological characters, including rounded palmately-veined leaves, strap-shaped petals with broad and glandular bases, capitulate inflorescence with two flowers fused back to back, and two or more seeds in each carpel locule. Therefore, I treat Disanthus as representing a monogeneric subfamily Disanthoideae. This treatment agrees with several studies (Harms, 1930; Chang, 1979; Takhtajan, 1997). Disanthoideae 257 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. have been proposed to be the most primitive taxon in the Hamamelidaceae (Harms, 1930; Pan et al., 1991; Takhtajan, 1997). However, it is intercalated between Mvtilaria and the Hamamelidoideae in the phylogenetic tree, suggesting that it is evolutionarily intermediate in the Hamamelidaceae. Obviously, Disanthoideae are a transitional taxon between the Hamamelidoideae and other more primitive subfamilies. Hamamelidoideae There is no doubt that the Hamamelidoideae are a monophyletic group (Bogle and Philbrick, 1980; Hufford and Crane, 1989; Endress, 1989a, b, c). However, morphologies are so diverse in this subfamily that controversies concerning tribal and subtribal relationships of the Hamamelidaceae exist only for this subfamily. Corylops ideae The tribe Corylopsideae sensu Harms (1930) included Corylopsis and Fortunearia. while Schulze-Menz (1964) transferred Sinowilsonia to this tribe due to his recognition of the similarity of Sinowilsonia and Fortunearia. Recently, Endress (1989c) separated Corylopsis from Fortunearia and Sinowilsonia and recognized it as a monogeneric tribe. The present phylogenetic analysis recommends the monophyletic group of Corylopsis-Loropetalum-Mainqaya-Matudaea (Fig.4.11). However, Corylopsis, as pointed out by Endress (1989b,c), is 258 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the only genus in the family, even the entire subclass Hamamelidae, that has clawed, orbicular petals. This unique characteristic, together with several other attributes, such as presence of seed caruncle and polyembryogeny (Li and Bogle, unpublished), led me to recognize Corylopsis as a monogeneric tribe following Endress (1989c). This tribe is closely related to Loropetalum-Mainqaya-Matudaea. Loropetaleae Loropetalum. Maingaya, and Tetrathyrium. along with a fourth genus, Embol anther a. which was not available for this study, but may soon become available, have been placed in the tribe Hamamelideae (Harms 1930) and subtribe Loropetalinae of the Hamamelideae (Endress, 1989c), while Matudaea is placed in the apetalous group of the Hamamelidoideae, namely, the tribe Fothergilleae (Endress, 1989c). However, the present phylogenetic analysis puts these four genera in a well- supported clade (Figs.2.2, 3.5, 4.11). There cure two morphological characters shared by these four genera, including distinct adaxial anther connective protrusions and bisexual flowers. Nevertheless, Matudaea is apetalous, and its ovaries are almost superior, while Loropetalum. Maingaya, and Tetrathyrium have showy strap-shaped petals, and almost inferior ovaries. Therefore, different subtribes could be recognized. However, I have not done that in this study for several reasons. First, in the phylogeny (Fig.4.11), 259 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Loropetalum is sister to the clade of Maingaya and Matudaea. Thus, if Matudaea is treated as a subtribe, the other genera should not be recognized as the other subtribe because they do not belong to a single clade, and thus are not monophyletic. Second, Embolanthera was not available for this study, and it is possible that inclusion of Embolanthera in a future analysis may change the intergeneric relationships of these genera. Hamamelideae The definition of the tribe Hamamelideae has not been changed since Harms (1930), although Endress (1989c) revised its infratribal relationships. The floral specializations of Hamamelis earned it monogeneric subtribal status (Endress, 1989c). In the present phylogenetic tree, this genus is in a separate clade from the other members of the Hamamelideae, thus recognizing its isolated systematic position. Instead, Hamamelis has a sister relationship with the Fothergilleae sensu Endress (1989c) (excl. Matudaea and Molinadendron). However, while Hamamelis is similar in leaf morphology with several genera of the Fothergilleae sensu Endress (1989c), including Fotherailla. Parrotia, Parrotiopsis. and Parrotia, it is rather different from the Fothergilleae in floral structure. For example, in Hamamelis. flowers are strictly 4-merous, have showy petals, pre-carpel initiation of staminodial nectars, and are bisexual (Mione and Bogle, 260 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1990), while in the Fothergilleae sensu Endress (1989c), flowers are apetalous, andromonoecious, and number of stamen is variable (Bogle, 1970). Therefore, I prefer to recognize two tribes in the Fothergilleae clade (Fig.4.11), one being the Hamamelideae (Hamamelis), and the other the Fothergilleae (Fothercrilla. Parrotiopsis. Parrotia, Shaniodendron, Sycopsis. Distyliopsis, and Distylium). Eusticmateae Eustigmateae sensu Harms (1930) is a monogeneric tribe, while that sensu Endress (1989c) includes three genera, Eustiqma. Fortunearia. and Sinowilsonia. Based on the present phylogeny, Molinadendron also belongs to the Eustigmateae and the grouping is strongly supported (Fig.4.11). Therefore, this study recognizes the Eustigmateae sensu Endress (1989c), but with the addition of Molinadendron. This tribe is supported by a couple of synapomorphies, such as linear stipules and two prophylls on a branch. Within this tribe, Eustiama and Fortunearia are closely related to each other and this pattern is supported by several synapomorphies, including pedicellate and lenticellate capsules, expanded purplish stigmatic surfaces, and reduced petals. On the other hand, Sinowilsonia and Molinadendron are allied in a clade, suggesting a close relationship. The synapomorphies supporting the clade of Molinadendron-Sinowilsonia include wind pollination and great 261 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reduction or absence of petals (Bogle, 1970; Endress, 1969). Dicorypheae It is Endress (1989b,c) who first recognized the possible monophyly of the five genera endemic to the Gondwanaland Continents of the Southern Hemisphere (Africa, Madagascar, Australia). These genera, Dicoryphe. Trichocladus. Ostrearia, Noahdendron. and Neostrearia. share this geographic distribution and a special pattern of anther dehiscence in which anthers are tetrasporangiate and bithecate, but dehisce via one valve per theca. This characteristic does not occur in any other members of the Hamamelidaceae (Endress, 1989a). In the present phylogeny (Fig.4.11), these five genera form a strongly supported clade, within which Dicoryphe and Trichoc ladus. and the three Australian genera form their own clades respectively. Some species of the former two genera tend to have oppositely arranged, peltate leaves, persistent stipules, and stellate foliar trichomes. The three Australian genera are very closely related and share a couple of synapomorphies, including scale trichomes (Warren Davis, per. comm.) and short styles (Endress, 1989a, b). An annotated classification system for the Hamamelidaceae As discussed above, I propose below a detailed, phylogenetically annotated classification system for the 262 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hamamelidaceae. Hamamelidaceae R.Br. Subfam.I. Altingioideae Reinsch (= Liquidambaroideae Harms) - basal clade in the Hamamelidaceae and paraphyletic to Exbucklandioideae-Rhodoleioideae. Liqniriamhar L. Semiliquidambar Chang Altinqia Nor. Sub£am.II. Exbucklandioideae Reinsch (= Bucklandioideae, Symingtonioideae) - sister to Rhodoleioideae. Exbucklandia R.W.Br. (= Bucklandia Griffith, Svmincrtonia Steenis). Subfam.IIl. Rhodoleioideae Harms Rhodoleia Champ.ex Hook. Subfam.IV. Mytilarioideae Chang - paraphyletic to Disanthoideae Mytilaria Lecomte Chunia Chang Subfam.V. Disanthoideae Harms - sister to Hamamelidoideae. Disanthus Maxim. Subfam. VI. Hamamelidoideae Reinsch - tribes I and II as one basal clade. Tribe I. Corylopsideae Harms - sister to Loropetaleae. Corylopsis Sieb. et Zucc. 263 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tribe II. Loropetaleae Li. Trib. Nov. Loropetalum R.Br.ex Reichb. Tetrathyrium Benth. Maingaya Oliv. Embolanthera Merr. Matudaea Lundell Tribe III. Hamamelideae A. DC. - sister to Fothergilleae. Hamamelis L. Tribe IV. Fothergilleae Harms Fotherqilla Murray Parrotiopsis Schneider Parrotia Mey. Shaniodendron Deng, Wei, et Wang Svcopsis Oliv. Distvlium Sieb. et Zucc. Distvliopsis Endress Tribe V. Eustigmateae Harms - sister to Dicorypheae. Eustiqma Gardn. et Champ. Fortunearia Rehd. et Wils. Sinowilsonia Hemsl. Molinadendron Endress Tribe VI. Dicorypheae Li. Trib. nov. Dicoryphe Du Petit-Thouars. Trichocladus Pers. Neostrearia Smith 264 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Noahdendron Endress, Hyland et Tracey Ostrearia Baill. Synopsis of the Hamamelidaceae Family Hamamelidaceae R.Brown, Abel.Narr. Journey China: 374 (1818) ("Hamamelideae"). (Harms, 1930; Chang, 1979; Endress, 1989c; Zhang and Lu, 1995). Trees or shrubs. Leaves persistent or deciduous, generally alternate or rarely opposite, stipulate, simple or palmately lobed; venation pinnate or palmate; epidermal trichomes simple, stellate or peltate scales. Inflorescences racemose, spicate or capitate, sometimes subtended by showy bracts. Flower morphology very variable, ranging from complete to incomplete, from bisexual to andromonoecious, and to monoecious; calyx composed of four or five united sepals, variable or lacking; corolla of four or five distinct petals, reduced or lacking; stamens (2)-4, 5, 10-(>20), anthers tetrasporangiate or bisporangiate; ovaries 2-carpellate, superior, semi-inferior, or inferior, each locule with one or more ovules. Fruit, a capsule, with woody exocarp and horny endocarp. Seeds one to several per carpel, winged or unwinged; seed coat bony, embryo straight, and endosperm albuminous. Subfamily I. Altingioideae Reinsch, Bot. Jahrb. Syst. 11: 389, 1889 (Harms, 1930; Bogle, 1968, 1970; Chang, 1979; 265 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Endress, 1989c; Zhang and Lu, 1995). Evergreen or deciduous trees. Leaves ovate, simple to tricuspidate or palmately lobed, pinnately or palmately veined, margins toothed or rarely entire. Stipules linear and caducous. Flowers unisexual, sessile, aggregated into dense capitula (or spikes). Perianth lacking. Stamens numerous, anthers tetrasporangiate and bithecate. Ovaries bicarpellate, semi-inferior, each carpel with numerous ovules. Seeds, winged, not ejected ballistically. Chromosome base number x=8. Three genera and about 20 species in eastern and western Asia, Central and North America. Subfamily II. Exbucklandioideae Reinsch (= Bucklandioideae), Bot. Jahrb. Syst. 11: 389, 1890 (Harms, 1930; Bogle, 1968; Chang, 1979; Zhang and Lu, 1995). Evergreen trees. Leaves simple, broadly ovate to 3- lobed, palmately veined, margins entire. Stipules large, appressed, elliptic, and persistent. Flowers andromonoecious, aggregated as dense capitula. Petals reduced or lacking. Stamens numerous. Anthers monothecate. Ovaries semi-inferior. Ovules numerous. Seeds winged. One genus and about four species, distributed in southeastern Asia. 266 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Subfamily III. Rhodoleioideae Harms in Engler et Prantl, Nat. Pflanzenfam., 2 ed., 18a: 335, 1930 (Bogle, 1968, 1986; Chang, 1979; Bogle and Philbrick, 1980; Endress, 1978b, 1989c, 1993; Zhang and Lu, 1995). Evergreen trees or shrubs. Leaves simple, ovate, rarely cordate, pinnately veined, margins entire. Stipules mostly lacking. Flowers bisexual and zygomorphic, aggregated into a pseudanthium, surrounded by an involucre of broad bracts, and bird-pollinated. Sepals and petals present. Stamens 6-8, anthers bithecate. Ovaries semi-inferior, bicarpellate, each carpel with 10-20 ovules. Seeds, winged, not ejected out of the capsule. Chromosome base number x=12. One genus and about 10 species in eastern Asia and Malaysia. Subfamily IV. Mytilarioideae Chang, Sunyatsen Univ. Bull. 1: 57, 1973 (Chang, 1948, 1979; Bogle, 1968; Pan and Yang, 1994; Zhang and Lu, 1995). Evergreen tree. Leaves broadly ovate to 3-lobed, palmately veined, margins entire. Stipules large, appressed and persistent. Flowers bisexual, spirally arranged on a spadix. Sepals present or absent, petals strap-like or absent. Stamens 8-13. Ovaries inferior, sunken in spadix. Carpels with 6 ovules. Chromosome base number x=13 (unknown for Chunia). Two genera and 2 species in southern China, Hainan, and northern Laos and northern Vietnam. 267 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Subfamily V. Disanthoideae Harms in Engler et Prantl. Nat. Pflanzenfam. 2 ed., 18a: 316, 1930 (Bogle, 1968; Chang, 1979; Pan et al., 1991; Zhang and Lu, 1995). Deciduous shrubs or small trees. Leaves simple, broadly ovate to orbicular, margins entire, palmately veined. Inflorescence capitulate, two-flowered. Flowers bisexual, 5- merous. Sepals present, petals strap-like, circinately coiled. Ovaries semi-inferior. Carpels with 4-6 ovules. Chromosome base number x=8. One genus and one species in southern Japan and several provinces of eastern and central China. Subfamily VI. Hamamelidoideae Reinsch, Bot. Jahrb. Syst. 11: 389, 1890 (Endress, 1967, 1970, 1989a, b,c, 1993; Bogle, 1968, 1970; Chang, 1979; Zhang and Lu, 1995). Trees or shrubs, evergreen or deciduous. Leaves simple, pinnately veined. Flowers, bisexual, andromonoecious, or monoecious. Inflorescence a spike, raceme, or capitulum. Perianth present or lacking. Ovaries superior to inferior. Carpels mostly with one ovule (rarely one fertile and two sterile). Seed ejected at maturity. Chromosome base number x=12). Twenty three genera and about 100 species widely distributed in Asia, North, Central, and South America, Australia, Africa, and Madagascar. 268 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tribe I. Corylopsideae Harms in Engler et Prantl, Nat. Pflanzenfam., 2 ed., 18a: 325, 1930 (Bogle, 1968; Morley and Chao, 1977; Endress, 1989b, c; Li et al., 1993). Deciduous or rarely semi-evergreen shrubs, or small trees. Flowers 5-merous, bisexual. Sepals present. Petals yellow, orbicular, and clawed. Carpels with 2-3 ovules (one fertile and two sterile). Capsules woody, with curved beaks. One genus and 11 species in southeastern Asia. GENUS TYPICUM: Corylopsis Sieb.et Zucc. Tribe II. Loropetaleae Li. Trib. Nov. Flores bisexuales. Petala lineari-eloncata v. nulla. Antherae basifixae. connectivo producto acuminato. Four genera about 5 species in southeastern Asia. GENUS TYPICUM: Loropetalum R.Br.ex Reichb. Tribe III. Dicorypheae Li. Trib. Nov. Flores bisexuales. Petala lineari-elonaata. Antherae bithecae. valva unica dehiscens. Five genera and about 20 species in eastern and southern Africa, northeastern Australia, Madagascar, and Comoros Islands). GENUS TYPICUM: Dicoryphe Du Petit-Thouars. Tribe IV. Eustigmateae Harms in Engler et Prantl, Nat. Pflanzenfam., 2 ed., 18a: 324, 1930 (Endress, 1989c). Buds naked. Branch prophylls 2. Flowers bisexual or 269 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. unisexual, with covering bracts. Petals reduced, or lacking. Stamens 5 or variable. Four genera and about eight species distributed in eastern Asia and central America. GENUS TYPICUM: Eustiqma Gardn. et Champ. Tribe V. Hamamelideae A. DC. Prodr. Syst. Nat. 4: 268, 1830 (as Hamameleae) (Mione and Bogle, 1990). Flowers 4-merous, bisexual, with ribbon-like, showy petals. Carpels with 2-3 ovules (one fertile and one or two sterile). One genus and about 5 species in eastern Asia, Central and North America. GENUS TYPICUM: Hamamelis L. Tribe VI. Fothergilleae A. DC., Prodr. Syst. Nat. 4: 269, 1830 (incl. Distylieae Harms in Engler et Prantl, Nat. Pf lanzenf am., 2 ed., 18a: 331, 1930) (Endress, 1989c, excl. Molinadendron and Matudaea). Flowers andromonoecious, apetalous (in a few genera also asepalous), Sepals and stamens highly variable. Eight genera and about 20 species in southeastern and western Asia, and eastern North America. GENUS TYPICUM: Fotherqilla Murray 270 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER VI GENERAL CONCLUSIONS The classification system of Endress (1989c) recognized four subfamilies in the Hamamelidaceae, four tribes in the subfamily Hamamelidoideae, and three subtribes in the tribe Hamamelideae. In the present study, parsimony analyses of the Hamamelidaceae were conducted based on both morphological and molecular data to examine the naturalness of the groups in the Endress classification system. The phylogeny based on sequences of the Internal Transcribed Spacers (ITS) of nuclear ribosomal DNA is basically congruent with that based on nucleotide sequences of the chloroplast gene matK. In contrast, the phylogeny based solely on morphology hardly agrees with the molecular phylogenies. A combined analysis using the pooled data of morphology and DNA sequences (ITS and matK) resolves all of the intergeneric relationships in the Hamamelidaceae. The resulting phylogeny recognizes the naturalness of several suprageneric taxa in Endress's system (1989c), including the subfamilies Altingioideae, Rhodoleioideae, and Hamamelidoideae. In the Hamamelidoideae, the tribes Corylopsideae, Eustigmateae, and Fothergilleae, and the 271 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subtribes Dicoryphinae, Hamamelidinae, and Loropetalinae are also natural. However, several taxa are suggested to be polyphyletic, including the subfamily Exbucklandioideae and the tribe Hamamelideae. The combined phylogeny also indicates that the apetalous, wind-pollinated genera in the Hamamelidoideae do not form a single clade, but are distributed in three separate clades, suggesting that parallel evolution has occurred in the family. A revision of the current classification system of the Hamamelidaceae is carried out to reflect the phylogenetic relationships suggested by the present analysis, resulting in a phylogenetically annotated suprageneric classification system. This new classification system recognizes six subfamilies, and six tribes in the subfamily Hamamelidoideae. At the subfamily level, this new classification system agrees with the system of Chang (1979), while at the tribal level, this system recognizes the four tribes in the system of Endress (1989c), with two new tribes Loropetaleae and Dicorypheae. The six subfamilies include the Altingioideae (basal), Exbucklandioideae (sister to Rhodoleioideae), Rhodoleioideae, Mytilarioideae (paraphyletic to Disanthoideae), Disanthoideae (sister to the Hamamelidoi deae), and Hamamelidoideae. The six tribes are Corylopsideae (sister to Loropetaleae), Loropetaleae, Eustigmateae (sister to the Dicorypheae), Dicorypheae, Hamamelideae (sister to the Fothergilleae), and Fothergilleae. 272 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LITERATURE CITED Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Arnheim, N., M. Krystal, G. Wilson, 0. Ryder, and E. A. Zimmer. 1980. Molecular evidence for genetic exchange among ribosomal genes on non-homologous chromosomes in man and apes. Proc. Natl. Acad. U.S.A. 77: 7323-7327. Baillon, H. 1871. Saxifragacdes, Hist. PI. 3: 325-464. rHamamelis series, 389-397; 414-415; Hamamelideae, 456- 461. Liquidambar series, 397-400; 415; Liquidambareae, 461-462.] ______. 1871-1873. Nouvelles notes sur les Hamamdlidees. Adansonia 10: 120-137. Baldwin, B. G. 1992. Phylogenetic utility of the internal transcribed spacers of nuclear ribosomal DNA in plants: An example from the Compositae. Mol. Phylogenetics Evol. 1: 3-16. ______, M. J. Sanderson, J. M. Porter, M. F. Wojciechowski, C. S. Campbell, and M. J. Donoghue. 1995. The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Ann. Missouri Bot. Gard. 82: 247-277. Barrett, M . , M. J. Donoghue, and E. Sober. 1991. Against consensus. Syst. Zool. 40: 486-493. Bentham, G. 1861. Flora Hongkongensis. L. Reeve, London. ______and J. D. Hooker. 1865. Hamamelideae. Gen. PI. 1: 664-669. Bogle, A. L. 1968. Floral vascular anatomy and the nature of the hamamelidaceous flower. Ph.D. thesis, University of Minnesota. ______. 1970. Floral morphology and vascular anatomy of the Hamamelidaceae: the apetalous genera of Hamamelidoideae. J. Arnold. Arb. 51: 310-366. ______. 1986. The floral morphology and vascular anatomy of the Hamamelidaceae: subfamily Liquidambaroideae. Ann. Missouri Bot. Gard. 73: 325-347. ______. 1987. Inflorescence and flower ontogeny in the pseudanthium of Rhodoleia (Hamamelidaceae). Amer. J. Bot. 74: 607-608 (Abstract). 274 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ______. 1989. The floral morphology, vascular anatomy, and ontogeny of the Rhodoleioideae (Hamamelidaceae) and their significance in relation to the "Lower" hamamelids, pp. 201-220. In: P. R. Crane and S. Blackmore, eds., Evolution, Systematics and Fossil History of the Hamamelidae. Vol. 1, Introduction and "Lower" Hamamelidae. Systematics Association Special Volume, no. 40A. Oxford Univ. Press, New York. ______. 1990. Multilacunar nodal anatomy in Mvtilaria (Hamamelidaceae). J. Arnold. Arb. 71: 111-118. ______. 1991. A second multilacunar nodal type in Hamamelidoideae. Harvard Papers in Botany. 3: 65-70. ______and C. T. Philbrick. 1980. A generic atlas of Hamamelidaceous pollens. Contr. Gray Herb. 210: 29-103. Bogler, D. J. and B. B. Simpson. 1996. Phylogeny of Agavaceae based on ITS rDNA sequence variation. Amer. J. Bot. 83: 1225-1235. Bradford, A. E., w. C. Dickison, J. R. Massey, and C. R. Bell. 1974. Vascular plant systematics. Harper & Row, Publishers, Inc., New York. Bremer, K. 1978. Oreolysera and Antithrixia, new and old South African genera of the Compositae. Bot. Notiser 131: 449-453. ______. 1988. The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42: 795-803. Brown, R. 1818. Appendix.-B, pp. 374-379. In: C. Abel. Narrative of a journey in the interior of China. Longman, London. Brown, R. W. 1946. Alterations in some fossil and living floras. J. Wash. Acad, Sci. 36: 344-355. Cantino, P. D., R. G. Olmstead, and S. J. Wagstaff. 1995. Phylogenetic classification of Labiatae and Verbenaceae: coping with poor resolution. Amer. J. Bot. 82 (6, Suppl.): 107 (Abstract). Candolle, A. P. De. 1830. Hamamelideae. Prodromous systematis naturalis regni vegetabilis. [Tribe Hamameleae, 268, Tribe Fothergilleae, 269, 270]. Part 4. Paris. 275 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chang, H.-T. 1948. Additions to the hamamelidaceous flora of China. Sunyatsenia 7: 63-74. ______. 1962. A new genus of Hamamelidaceae in the flora of China. Sunyatsen Univ. Bull. Nat. Sci. 1: 34- 44. ______. 1973. A revision of the Hamamelidaceous flora of China. Sunyatsen Univ. Bull. 1: 54-71. ______. 1979. Florae Reipublicae Popularis Sinicae. 35 (2): 36-116. Science Press of China, Beijing. Chase, M. W., D. E. Soltis, R. G. Olmstead, D. Morgan, D. Les, B. D. Mishler, M. R. Duvall, R. A. Price, H. G. Hills, Y. L. Qiu, D. A. Rron, J. H. Rettig, E. Conti, W. D. Clark, M. Hedren, B. S. Gaut, R. K. Jansen, K. J. KIM, C. F. Wimpee, J. F. Smith, G. R. Furnier, S. H. Strauss, Q.-L. Xiang, G. M. Plunkett, P. S. Soltis, S. M. Swensen, S. E. Williams, P. A. Gadek, C. J. Quinn, L. E. Eguiarte, E. Golenberg, G. H. Learn, S. W. Graham, S. C. H. Barrett, S. Dayanandan, V. A. Albert. 1993. Phylogenetics of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL. Ann. Missouri Bot. Gard. 80: 528-580. Cracraft, J. and D. P. Mindell. 1989. The early history of modern birds: a comparison of molecular and morphological evidence, pp. 389-403. In: B. Fernholm, K. Bremer, and H. Jornvall, eds., The Hierarchy of Life. Elsevier, Amsterdam. Cronquist, A. 1981. An Integrated System of Classification of Flowering Plants. Columbia Univ. Press, New York. ______. 1988. The Evolution and Classification of Flowering Plants. Second Edition. New York Bot. Gard. Bronx, New York. Davis, P. H. and V. H. Heywood. 1963. Principles of Angiosperm Taxonomy. J. J: Van Nostrand. Princeton. de Queiroz, K. 1993. For consensus (sometimes). Syst. Biol. 42: 368-372. ______and J. Gauthier. 1995. Translating phylogenies into classifications: Basic principles. Amer. J. Bot. 82 (6, Suppl.): 108 (Abstract). 276 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Deng, M.-B., H.-T. Wei, and X.-Q. Wang. 1992a. Shaniodendron. a new genus of Hamamelidaceae from China. Acta Phytotax. Sin. 30: 57-61. (In Chinese). ______, , , P. Su, and J.-Y. Xing. 1992b. On the significance of the discovery of Fothergilleae in China. J. Plant Resour. and Environ. 1 (1): 30-35. (In Chinese). Donoghue, M. J. 1995. Phylogeny and phylogenetic taxonomy of Dipsacales. Amer. J. Bot. 82 (6, Suppl.): 108 (Abstract). ______and M. J. Sanderson. 1992. The suitability of molecular and morphological evidence in reconstructing plant phylogeny, pp. 340-368. In: P. S. Soltis, D. E. Soltis, and J. J. Doyle, eds., Molecular Systematics of Plants. Chapman and Hall, New York. ______, Olmstead, R. G . ,Smith, J. F., and J. D. Palmer. 1992. Phylogenetic relationships of Dipsacales based on rbcL sequences. Ann. Missouri Bot. Gard. 79: 333-345. Dover, G. A. 1982. Molecular drive: a cohesive mode of species evolution. Nature 299: 111-117. Downie, S. R. and D. S. Katz-Downie. 1996. A molecular phylogeny of Apiaceae subfamily Apioideae: evidence from nuclear ribosomal DNA internal transcribed spacer sequences. Amer. J. Bot. 83: 234-251. Doyle, J. J. 1993. DNA, phylogeny, and the flowering of plant systematics. BioScience 43: 380-389. ______and Doyle, J. L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19: 11-15. Du Petit-Thouars. 1804. Histoire des Vegetaux. 31. Paris. D'aquila, R. T., L. J. Bechtel, J. A. Viedeler, J. J. Eron, P. Gorczyca, and J. C. Kaplin. 1991. Maximizing sensitivity and specificity of PCR by preamplification heating. Nucleic. Acids Res. 19: 3749. Elias, T. S. 1980. The Complete Trees of North America - Field Guide and Natural History. Times Mirror Magazines, Inc., New York. 277 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Endress, P. K. 1967. Systematische Studie viber die verwandtschaftlichen Beziehungen zwischen den Hamamelidaceen und Betulaceen. Bot. Jahr. Syst. 87: 431-525. ______. 1969. Molinadendron. eine neue Hamamelidaceen Gattung aus Zentralamerika. Bot. Jahrb. 89: 354-359. . 1970. Die Inf lores zenzen der apetalen Hamamelidaceen, ihre grundsatzliche morphologische und systematische Bedeutung. Bot. Jahr. Syst. 90: 1-54. . 1976. Die Androeciumanlage bei polyandrischen Hamamelidaceen und ihre systematische Bedeudung. Bot. Jahrb. Syst. 97: 436-457. . 1977. Evolutionary trends in the Hamamelidales-Fagales-group. PI. Syst. Evol., Suppl. 1: 321-347. . 1978a. Blutenontogenese, Bliitenabgrenzung und systematische Stellung der perianthlosen Hamamelidoideae. Bot. Jahrb. Syst. 100: 249-317. . 1978b. Stipules in Rhodoleia (Hamamelidaceae). PI. Syst. Evol. 130: 157-160. ______. 1989a. Aspects of evolutionary differentiation of the Hamamelidaceae and the lower Hamamelidae. PI. Syst. Evol. 162: 193-211. _ . 1989b. Phylogenetic relationships in the Hamamelidoideae, pp. 227-248. In: P. R. Crane and S. Blackmore, eds., Evolution, Systematics and Fossil History of the Hamamelidae. Vol. 1: Introduction and "Lower" Hamamelidae. Systematics Association Special Vol. No. 40A. Oxford Univ. Press, New York. ______. 1989c. A suprageneric taxonomic classification of the Hamamelidaceae. Taxon 38: 371-376. _ . 1993. Hamamelidaceae, pp. 322-331. In: K. Kubitzki, J. G. Rohwer, V. Bittrich, eds., The Families and Genera of Vascular Plants, II: Flowering Plants. Dicotyledons. Springer-Verlag, Berlin. ______and J. Anliker. 1968. XSycoparrotia semidecidua hybr. nov. (= Parrotia persica C. A. Mey. X Sycopsis sinensis Oliv.). Schweiz. Beitr. Dendrol. 16- 18: 6-28. 278 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ______and E. M. Friis. 1991. Archamamelis. hamamelidacean flowers from the Upper Cretaceous of Sweden. PI. Syst. Evol. 175: 101-114. ______, B. P. M. Hyland, and J. G. Tracey. 1985. Noahdendron. a new Australian genus of the Hamamelidaceae. Bot. Jahrb. Syst. 107: 369-378. Faith, D. and P. Cranston. 1991. Could a cladogram this short have arisen by chance alone? Cladistics 7: 1-28. Fang, Y.-M. and R.-W. Fan. 1993. Variation and evolution of leaf trichomes in the Chinese Hamamelidaceae. Acta Phytotax. Sin. 31: 147-152. ______and M.-B. Deng. 1996. Wood anatomy of Shaniodendron subaequale (Hamamelidaceae) and its systematic implications. J. Plant Resour. and Environ. 5(2): 50-54. Farii, J. S., A. G. Kluge and M. F. Mickevich. 1982. Immunological distances and the phylogenetic relationships of the Ranahnyi i -i species group. Syst. Zool. 31: 479-491. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791. Funk, V. A. 1985. Cladistics and generic concepts in the Compositae. Taxon 34: 72-80. Gardner, G. and J. G. Champion. 1849. Descriptions of some new genera and species of plants, collected in the Island of Hong-Kong by Capt. J. G. Champion, Hooker's J. Bot. Kew Gard. Misc. 1: 321-328. Gaut, B. S., S. V. Muse, W. D. Clark, and M. T. Clegg. 1992. Relative rate of nucleotide substitution at the rbcL locus of monocotyledonous plants. J. Molec. Evol. 35: 292-303. Golblatt, P. and P. K. Endress. 1977. Cytology and evolution in Hamamelidaceae. J. Arnold Arb. 58: 67-71. Gustafsson, M. H. G., and K. Bremer. 1995. Morphology and phylogenetic interrelationships of the Asteraceae, Calyceraceae, Campanulaceae, Goodeniaceae, and related families (Asterales). Amer. J. Bot. 82: 250-265. 279 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hallier, H. 1903. Uber den Umfang, die Gliederung und die Verwandschaft der Familie der Hamamelidaceen. Beih. Bot. Centralbl. 14: 247-260. Hamby, R. K. and E. A. Zimmer. 1992. Ribosomal RNA as a phylogenetic tool in plant systematics, pp. 50-91. In: P. S. Soltis, D. E. Soltis, and J. J. Doyle, eds., Molecular Systematics of Plants. Chapman and Hall, New York. Hao, R.-M., H.-T. Wei, and W.-G. Liu. 1996. Floral morphology of Shaniodendron (Hamamelidaceae) and its taxonomic significance. J. PI. Resour. and Environ. 1 (1): 30-35. Harms, H. 1930. Hamamelidaceae. In: A. Engler and K. Prantl, eds., Die Natiirlichen Pflanzenfamilien. Second Edition. 18A: 303-345. W. Engelmann, Leipzig. Hemsley, W. B. 1906. Sinowilsonia henryi Hemsl. In: Hook. Ic. PI. 29: t. 2817. Hershkovitz, M. A. and L. A. Lewis. 1996. Deep level diagnostic value of the rDNA-ITS region. Mol. Biol. Evol. 13: 1276-1295. Heywood, V. H., D. M. Moore, I. B. K. Richardson, W. T. Stearn. 1993. Flowering plants of the world. Oxford University Press, New York. Hibsch-Jetter, C. and D. E. Soltis. 1996. Phylogenetic analysis of 'Saxifragales' based on nrDNA and cpDNA sequence data (18S, rbcL, and matK). Amer. J. Bot. 83 (6, suppl.): 163. (Abstract). Hillis, D. M. 1987. Molecular versus morphological approaches to systematics. Ann. Rev. Ecol. Syst. 18: 23-42. ______, C. Moritz, C. A. Porter, and R. J. Baker. 1991. Evidence for biased gene conversion in concerted evolution of ribosomal DNA. Science 251: 308-310. Hilu, K. W. and H.-P. Liang. 1997. The matK gene: sequence variation and application in plant systematics. Amer. J. Bot. 84: 830-840. Hooker, W. J. 1850. Rhodoleia championii. Curtis's Bot. Mag. 76: t 4509. 280 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hoot, S. B. and P. R. Crane. 1996. A comparison of results from 18S nuclear ribosomal DNA sequences with two chloroplast gene (aptB and rbcL) within the "Lower" Hamamelididae and Ranunculidae. Amer. J. Bot. 83 (6, Suppl.): 212 (Abstract). ______, S. Magallon-Pubela, and P. R. Crane. 1997. Evolutionary relationships of the "basal" eudicots based on three sequence data sets: atpB. rbcL and 18S ribosomal DNA. Amer. J. Bot. 84 (6, Suppl.): 203 (Abstract). Huang, G.-L. 1982. Anatomical studies of Chunia wood. Acta Bot. Sin. 24: 506-511 (In Chinese). ______. and C.-L. Lee. 1986. Comparative anatomical studies on the woods of Hamamelidaceae in China. Acta Sci. Nat. Univ. Sunyatsen. 1: 22-28. Huang, S.-F. 1985. Preliminary reports of plant chromosome numbers. Asian Forest Science 1: 37-42 (in Chinese). Huelsenbeck, J. P. 1991. Tree-length distribution skewness: an indicator of phylogenetic information. Syst. Zool. 40: 257-270. Hufford, L. D. 1992. Rosidae and their relationships to other nonmagnoliid dicotyledons: A phylogenetic analysis using morphological chemical data. Ann. Missouri Bot. Gard. 79: 218-248. ______and P. R. Crane. 1989. A preliminary phylogenetic analysis of the "Lower" Hamamelidae, pp. 175-193. In: P. R. Creme and S. Blackmore, eds., Evolution, Systematics and Fossil History of the Hamamelidae. Vol. 1: Introduction and "Lower" Hamamelidae. Systematics Association Special Volume, No. 40A. Oxford Univ. Press, New York. Hutchinson, J. 1967. The Families of Flowering Plants. Dicotyledons. Vol. 2. Clarendon Press, Oxford. Johnson, L. A. and D. E. Soltis. 1994. matK DNA sequences and phylogenetic reconstruction in Saxifragaceae sensu stricto. Syst. Bot. 19: 143-156. ______and ______. 1995. Phylogenetic inference in Saxifragaceae sensu stricto and Gilia (Polemoniaceae) using matK sequences. Aim. Missouri Bot. Gard. 82: 149-175. 281 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jorgensen, R. A. and P. D. Cluster. 1988. Modes and tempos in the evolution of nuclear ribosomal DNA: new characters for evolutionary studies and new markers for genetic and population studies. Ann. Missouri Bot. Gard. 75: 1238- 1247. Judd, W. S. 1995. On the need of expedite the transfer of results of phylogenetic analyses into botanical instruction and floristics. Amer. J. Bot. 82 (6, Suppl.): 10-109 (Abstract). Kaul, U. and R. N. Kapil. 1974. Exbucklandia populnea-from flower to fruit. Phytomorphology 24: 217-228. Kallersjo, M., J. S. Farris, A. G. Kluge, and C. Bult. 1982. Skewness and permutation. Cladistics 8: 275-287. Kellogg, E. A. R. Appels, R. Mason-Gamer. 1996. When genes tell different stories: the diploid genera of Triticeae (Gramineae). Syst. Bot. 321-347. Kim, K.-J. and T. J. Mabry. 1991. Phylogenetic and evolutionary implications of nuclear ribosomal DNA variation in dwarf dandelions (Krigia, Lactuceae, Asteraceae). PI. Syst. Evol. 177: 53-69. Kluge, A. G. 1989. A concern for evidence and a phylogenetic hypothesis of relationships among Epicrates (Boidae, Serpentes). Syst. Zool. 38: 7-25. Kron, K. A. 1995. Classification of the Ericaceae, Empetraceae, and Epacridaceae: A preliminary evaluation of the Linnaean hierarchical and phylogenetic systematic approaches. Amer. J. Bot. 82 (6, Suppl.): 109 (Abstract). ______and J. M. King. 1996. Cladistic relationships of Kalmia. Leiophyllum. and Loiseleuria (Phyllodoceae, Ericaceae) based on rbcL and nrlTS data. Syst. Bot. 21: 17-31. Lecomte, H. 1924. Une hamamelidacde nouvelle d'Indochina. Bull. Mus. Natl. Hist. Paris. 30: 503-507. Li, H.-M. and L. J. Hickey. 1988. Leaf architecture and systematics of the Hamamelidaceae sensu lato. Acta Phytotax. Sin. 96-110. 282 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Li, J.-H., K.-Y. Pan, and A.-M. Lu. 1993. A revision of Corylopsis (Hamamelidaceae). (Abstract). Proceedings of the 60th Anniversary of the Chinese Botanical Society (in Chinese). ______and A. L. Bogle. 1995. Foliar crystals and sclereids of the family Hamamelidaceae and their systematic significance. Amer. J. Bot. 82 (6, Suppl.): 145 (Abstract). ______,______, and A. S. Klein. 1997a. A phylogenetic analysis of the Hamamelidaceae using DNA sequences of nuclear ribosomal ITS and chloroplast gene matK. Amer. J. Bot. 84 (6, suppl.): 212 (Abstract). ______, ______, and ______. 1997b. Interspecific relationships and genetic divergence of the disjunct genus Liquidambar (Hamamelidaceae). Rhodora 99, in press. ______, ______, and ______. 1997c. Phylogenetic relationships in the Corvlopsis complex (Hamamelidaceae) based on sequences of the internal transcribed spacers of nuclear ribosomal DNA and morphology. Rhodora 99, in press. ______, ______, ______, and K. Y. Pan. 1997d. Close relationship between Shaniodendron and Parrotia (Hamamelidaceae), evidence from sequences of the internal transcribed spacers of nuclear ribosomal DNA. Acta Phytotax. Sin. 35, in press. Lindley, J. 1846. The Vegetable Kingdom. Bradbury and Evans, London. ______. 1853. The Vegetable Kingdom. Ed.3. Bradbury and Evans, London. Liston, A., W. A. Robinson, J. M. Oliphant, and E. R. Alvarez Buylla. 1996. Length variation in the nuclear ribosomal transcribed spacer region of non-flowering seed plants. Syst. Bot. 21: 109-121. Lundell, C. L. 1940. Three noteworthy new plants from southern Mexico. Lloydia 3: 209-212. Maddison, W. P. and D. R. Maddison. 1992. MacClade: Analysis of Phylogeny and Character Evolution. Version 3.03. Sinauer Associates. Sunderland, MA. 283 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Manos, P. S. and K. P. Steele. 1997. Phylogenetic analysis of "higher" Hamamelididae based on plastid sequence data. Amer. J. Bot. 1407-1420. Maximowicz, C. J. 1866. Diagnoses breves plantarum novarum Japoniae et Mandshuriae. Decas 1. Bull. Acad. Imp. Sci. Saint-Petersbourg 10: 485-490. Melikian, A. P. 1971. Anatomical structure of the spermodermis in the members of the genera Licruidambar L. and Altincia Nor. in relation to their taxonomy. Izv. Biol. Nauki 24: 50-55 (In Russian). ______. 1972. On the place of the genera Rhodoleia. Exbucklandia. and Chunia in the system of the family Hamamelidaceae. Izv. Biol. Nauki 25: 39-45 (In Russian). ______. 1973a. Seed coat types of Hamamelidaceae and allied families in relation to their systematics. Bot. Zurn. (Moscow & Leningrad) 58: 350-359 (In Russian). ______. 1973b. The anatomy of seed coat and systematics of the family Hamamelidaceae. Izv. Biol. Nauki 26: 104-105 (In Russian). ______.1975. Comparative morphological studies on representatives of the family Hamamelidaceae and their phylogenetic relationships, pp. 7-24. In: L. J. Budantsjev, ed., Problems of Comparative Morphology of the Seed Plants. Nauka, Leningrad (In Russian). Merrill, E. D. 1909. New and noteworthy Philippine plants. Philip. J. Sci (Bot.) 3: 263. Metcalfe, C. R. and L. Chalk. 1979. Anatomy of the Dicotyledons. Vol. 1. Clarendon Press, Oxford. Meyer, C. A. 1831. Verzeichniss der Pflanzen, welche 1829-30 im Kaukasus und in den Provinzen am westlichen Ufer des Kaspischen Meeres gefunden und eingesammelt worden sind. St. Petersburg. Mione, T. 1987. Comparative ontogeny of the inflorescences and flowers of Hamamelis virqiniana and Loropetalum chinense (Hamamelidaceae). M.S. thesis, University of New Hampshire. Durham. 284 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ______and A. L. Bogle. 1990. Comparative ontogeny of the inflorescences and flowers of Hamamelis viroiniana and Loropetalum chinense (Hamamelidaceae). Amer. J. Bot. 77: 77-91. Miyamoto, M. M. 1985. Consensus cladograms and general classifications. Cladistics 1: 186-189. Morawetz, W., M. Rosabelle, and A. Samuel. 1989. Karyological patterns in the Hamamelidae, pp. 129-155. In: P. R. Crane and S. Blackmore, eds., Evolution, Systematics and Fossil History of the Hamamelidae. Vol. 1: Introduction and "Lower" Hamamelidae. Systematics Association Special Volume, No. 40A. Oxford Univ. Press. New York. Morgan, D. R. 1995. Molecular data and subfamilial classification of Rosaceae. Amer. J. Bot. 82 (6, Suppl.): 109 (Abstract). ______. 1997. Decay analysis of large sets of phylogenetic data. Taxon 46: 509-517. ______and D. E. Soltis. 1993. Phylogenetic relationships among members of Saxifragaceae sensu lato based on rbcL sequence data. Ann. Missouri Bot. Gard. 80: 633-660. Morley, B. and J. M. Chao. 1977. A review of Corylopsis (Hamamelidaceae). J. Arnold Arb. 58: 382-414. Murray, J. A. 1774. Fotherailla. In: C. Linnaeus, ed., Syst. Veget. Ed. 13: 418. Musters, w . , K. Boon, C. A. F. M. van der Sande, H. van Heerikhuiszan, and R. J. Planta. 1990. Functional analysis of transcribed spacers of yeast ribosomal DNA. EMBO 9: 3989-3996. Nakai, T. 1943. Ordines, familiae, tribi, genera, sectiones, species, varietates et combinationes novae. Imperial Univ., Tokyo. Neuhaus, H., and G. Link. 1987. The chloroplast tRNAiys (UUU) gene from mustard (Sinapsis alba) contains a class II intron potentially coding for a maturase-related polypeptide. Current Genetics 11: 251-257. Niedenzu, F. 1891. Hamamelidaceae, pp. 115-130. In: A. Engler and K. Prantl, eds., Nat. Pflanzenfam. III. 2a. 285 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Noronha, F. 1790. Altinqia excelsa. malaice et javanice Rasamala, Lignum papuanum Rumphii Herbar. Amboin. 2: 57. Verh. Batav. Genootsch. Kunsten Wetensch. 5. 1-20. Nues, R. W. van, J. M. Rientjes, C. A. M. van der Sande, S. F. Zerp, C. Sluiter, J. Vernema, R. J. Planta, and H. A. Raue. 1994. Separate elements within internal transcribed spacer 1 of Saccharomyces cerevisiae precursor ribosomal RNA direct the formation of 17S and 26S rRNA. Nucl. Acids Res. 22: 912-919. Oginuma, K., and H. Tobe. 1991. Karyomorphology and evolution in some Hamamelidaceae and Platanaceae (Hamamelididae; Hamamelidales). Bot. Mag. (Tokyo) 104: 115-135. Oliver, D. 1860. On Sycopsis. Trans. Linn. Soc. Lond. 23: 83- 89. PI. 8. ______. 1862. Note on Hamamelis and Loropetalum: with a description of a new Anisophyllea from Malacca. Trans. Linn. Soc. (Lond.) 23: 457-461. ______. 1872. Description of three new genera of plants in the Malayan herbarium of the late Dr. A. C. Maingay. Trans. Linn. Soc. Lond. 28: 517. T.44. ______. 1890. Svcopsis sinensis Oliv. Hooker's Ic. PI. 20: pi. 1931. Olmstead, R. G. 1995. Phylogeny and classification in the Solanaceae. Amer. J. Bot. 82 (6, Suppl.): 109-110 (Abstract). ______and J. D. Palmer. 1994. Chloroplast DNA systematics: a review of methods and data analysis. Amer. J. Bot. 81: 1205-1224. Padgett, D. J. 1997. A biosystematic monograph of the genus Nuphar Sm. (Nymphaeaceae). Ph.D. thesis. University of New Hampshire, Durham. Pan, K.-Y. and Q.-E. Yang. 1994. Karyotypes of Disanthus and Mytilaria (Hamamelidaceae). Acta Phytotax. Sin. 32: 235-239. ______, A.-M. Lu, and J. Wen 1990. Characters of leaf epidermis in Hamamelidaceae (s.l.). Acta Phytotax. Sin. 28: 10-26. 286 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. , ______, and______. 1991. A systematic study on the genus Disanthus Maxim. (Hamamelidaceae). Cathaya. 3: 1-28. Penny, D., L. R. Foulds, and M. D. Hendy. 1982. Testing the theory of evolution by comparing phylogenetic trees constructed from five different protein sequences. Nature 297: 197-200. Persoon, C. H. 1807. Synopsis Plantarum. II: 597. Paris. Platnick, N. I., C. E. Griswold, and J. A. Coddington. 1991. On missing entries in cladistic analysis. Cladistics 7: 337-343. Plunkett, G. M. , D. E. Soltis, and P. S. Soltis. 1997. Clarification of the relationship between Apiaceae and Araliaceae based on matK and rbcL sequence data. Amer. J. Bot. 84: 565-580. Rakotobe, E. A. 1996. Le Genre endemique Malgache Dicoryphe Du Putit-Thouars (Hamamelidaceae): Repartition et phytogeographie. Biogeographie de Madagascar: 177-182. Rao, M. P. R. 1974. Seed anatomy in some Hamamelidaceae and phylogeny. Phytomorphology 24: 113-139. Rehder, A., and E. H. Wilson. 1913. Hamamelidaceae, pp. 421- 432. In: C. S. Sargent, Plantae Wilsonianae. Vol. 2. Harvard Univ. Press, Cambridge. Reinsch, A. 1889. uber die anatomischen verhaltnisse der Hamamelidaceae mit Rucksicht auf ihre systematische Gruppierung. Engler's Bot. Jahrb. 2: 347-395. Rieseberg, L. H. and S. J. Brunsfeld. 1992. Molecular evidence and plant introgression, pp. 151-176. In: P. S. Soltis, D. E. Soltis, and J. J. Doyle, eds., Molecular Systematics of Plants. Chapman and Hall, New York. Ritland, K. and J. E. Eckenwalder. 1992. Polymorphism, hybridization, and variable evolutionary rate in molecular phylogenies, pp. 404-428. In: P. S. Soltis, D. E. Soltis,and J. J. Doyle, eds., Molecular Systematics of Plants. Chapman and Hall, New York. Rogers, S. O. and A. J. Bendich. 1987. Ribosomal RNA genes in plants: variability in copy number and in the intergenic spacer. PI. Mol. Biol. 9: 509-520. 287 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sande, C. A. F. M. van der, M. Kwa, R. w. vein Nues, H. van Heerikhuizen, H. A. Raue, and R. J. Plant. 1992. Functional analysis of internal transcribed spacer 2 of Saccharomyces cerevisiae ribosomal DNA. J. Molec. Biol. 223: 899-910. Sanderson, M. J. and J. J. Doyle. 1992. Reconstruction of organismal and gene phylogenies from data on multigene families: concerted evolution, homoplasy, and confidence. Syst. Biol. 41: 4-17. Santmour, F. S., Jr. 1965. Chromosome number in Corylopsis. Morris Arboretum Bulletin 16: 7. ______. 1972. Chromosome numbers in Liquidambar. Rhodora 74: 287-290. Schilling, E. E. and J. L. Panero. 1996. Phylogenetic reticulation in subtribe Helianthinae. Amer. J. Bot. 83: 939-948. Schneider, C. K. 1905. Illustriertes Handbuch der Laubholzkunde. Gustav Fischer, Jena. Schulze-Menz, G. K. 1964. Rosales, pp. 196-198. In: H. Melchior, ed., A. Engler's Syllabus der Pflanzenfamilien. Ed. 12. 2. Shaw, H. K. A. 1966. In: J. C. Willis, ed., A dictionary of the flowering plants and ferns (7th ed.). Cambridge Univ. Press, New York. Siebold, P. F. von, and J. G. Zuccarini. 1835. Flora Japonica. 1: Corylopsis. 45; Distylium. 178. Sinnott, E. W . , and I. W. Bailey. 1914. Investigations on the phylogeny of the angiosperms. 3. Nodal anatomy and the morphology of stipules. Amer. J. Bot. 1: 441-453. Skvortsova, N. T. 1960a. The anatomical structure of the conducting system of leaf petioles of representatives of families Hamamelidaceae and Altingiaceae. Dokl. Akad. Nauk SSSR 133: 1231-1233. (In Russian). ______. 1960b. On the types of leaf venation in the members of the family Hamamelidaceae. Trudy Leningrad Chemico-Pharmaceutical Inst. 12: 75-83 (In Russian). ______. 1960c. The structure of epidermis in the members of the family Hamamelidaceae. Bot. Zurn. (Moscow & Leningrad) 45: 712-717 (In Russian). 288 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Smith, L. S. 1958. New species of and notes on Queensland plants. Proc. R. Soc. Queensl. 69: 43-48. Soltis P. S., and D. E. Soltis. 1995. Plant molecular systematics. Inferences from phylogeny and evolutionary processes, pp. 139-194. In: M. K. Hecht et al., eds., Evolutionary Biology 28. Plenum Press, New York. Soltis, D. E., L. A. Johnson, and C. Looney. 1996a. Discordance between ITS and chloroplast topologies in the Bovkinia group. Syst. Bot. 21: 169-187. ______, R. K. Kuzoff, E. Conti, R. Gornall, and K. Ferguson. 1996b. matK and rbcL gene sequence data indicate that Saxifraaa (Saxifragaceae) is polyphyletic. Amer. J. Bot. 83: 371-382. Steele, K. P. and R. Vilgalys. 1994. Phylogenetic analyses of Polemoniaceae using nucleotide sequences of the plastid gene matK. Syst. Bot. 19: 126-142. Steenis, C. G. G. J. van. 1952. Nomenclatural note on the genus Bucklandia R. Br. (Hamamelidaceae). Acta Bot. Neerl. 1: 443-444. Swofford, D. L. 1993. PAUP: Phylogenetic Analysis Using Parsimony, version 3.1.1. User's Manual. Washington, D.C., Laboratory of Molecular Systematics, Smithsonian Institution. ______. 1997. PAUP* beta test version. 4.0d55-59. Sinauer Associates. Sunderland, MA. Sytsma, K. J. and B. A. Schaal. 1985. Phylogenetics of the Lisianthius skinneri (Gentianaceae) species complex in Panama utilizing DNA restriction fragment analysis. Evolution 39: 594-608. ______and ______. 1990. Ribosomal DNA variation within and among individuals of Lisianthius (Gentianaceae) populations. PI. Syst. Evol. 170: 97-106. Tahktajan, A. 1969. Flowering Plants - Origin and Dispersal. Smithsonian Institution Press, Washington, DC. ______. 1980. Outline of the classification of flowering plants (Magnoliophyta). Bot. Rev. 46: 225-359. ______. 1997. Diversity and Classification of Flowering Plants. Columbia University Press, New York. 289 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tang, Y. 1943. Systematic anatomy of the woods of the Hamamelidaceae. Bull. Fan Mem. Inst. Biol. 1: 8-63. Tardieu-Blot, M. L. 1965. Hamamelidaceae, pp. 75-116. In: A. Aubreville, and M. L. Tardieu-Blot. Flore du Cambodge de Laos et du Vietnam, 4. Paris. Tehler, A. 1995. Morphological data, molecular data, and total evidence in phylogenetic analysis. Can. J. Bot. 73 (Suppl. 1): S667-S676. Tong, K. 1930. Studien iiber die Familie der Hamamelidaceae mit besonderer Berucksichtigung der Systematik und Entwicklungsgeschichte von Corylopsis. Bull. Biol. Dept. Sci. Coll. Sun Yat-sen Univ. 2: 1-72. Walker, E. H. 1944. A revision of Distvlium and Svcopsis (Hamamelidaceae). J. Arnold Arb. 25: 319-341. Wang, X.-Z. 1992. Palaeopalynological evidence of phylogeny in Hamamelidaceae. Acta Phytotax. Sin. 30: 137-145. Wendel, J. F., A. Schnabel, and T. Seelanan. 1995. Bidirectional interlocus concerted evolution following alloploid speciation in cotton (Gossypium). Proc. Natl. Acad. U.S.A. 92: 280-284. Wheeler, W. C., J. C. Gatesy, and R. Desalle. 1995. Elision: a method for accommodating multiple molecular sequence alignments with alignment-ambiguous sites. Molec. Phylogenetics Evol. 4: 1-9. 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: M. Innis et al., eds., PCR Protocols: A Guide to Methods and Applications. Academic Press, San Diego, California. Wiley, E. 0., E. Siegel-Causey, D. R. Brooks, V. A. Funk. 1991. The Compleat Cladist-A Primer of Phylogenetic Procedures. Special Publication No. 19. Museum of Natural History, the University of Kansas, Lawrence, Kansas. Wisniewski, M. and A. L. Bogle. 1982. The ontogeny of the inflorescence and flower of Liquidambar styraciflua L. (Hamamelidaceae). Amer. J. Bot. 69: 1612-1624. 290 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wojciechowski, M. F., M. J. Sanderson, B. G. Baldwin, and M. J. Donoghue. 1993. Monophyly of aneuploid Astaaalus (Fabaceae). evidence from nuclear ribosomal DNA internal transcribed spacer sequences. Amer. J. Bot. 80: 711-722. Wolfe, J. A. 1973. Fossil forms of Amentiferae. Brittonia 25: 334-335. Wolfe, K. H., C. W. Morden, and J. D. Palmer. 1992. Function and evolution of a minimal plastid genome from a nonphotosynthetic parasitic plant. Proc. Natl. Acad. Sci. USA. 89: 10648-10652. Yokota, Y., T. Kawata, Y. Iida, A. Kat, and S. Tanifuji. 1989. Nucleotide sequences of the 5.8S rRNA gene and internal transcribed spacer regions in carrot and broad bean ribosomal DNA. J. Mol. Evol. 29: 294-301. Zhang, J.-T. and D.-W. Zhang. 1991. Studies on pollen morphology of the genus Corylopsis. Acta Phytotax. Sin. 347-351. Zhang, Z.-Y. and Lu, A.-M. 1995. Hamamelidaceae: Geographic distribution, fossil history and origin. Act. Phytotax. Sin. 33: 313-339. ______and J. Wen. 1996. The seed morphology in Hamamelidaceae and its systematic evaluation. Acta Phytotax. Sin. 34: 538-546. Zimmer, E. A., S. L. Martin, S. M. Beverley, Y. Kan, and A. C. Wilson. 1980. Rapid duplication and loss of gene coding for the alpha chain hemoglobin. Proc. Natl. Acad. Sci. U.S.A. 77: 2158-2162. 291 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.