Phylogeny and Character Evolution in the

Genus Crepis L. (Cichorieae, Compositae)

Dissertation zur Erlangung des akademischen Grades des

Doktors der Naturwissenschaften (Dr. rer. nat.)

eingereicht im Fachbereich Biologie, Chemie, Pharmazie

der Freien Universität Berlin

vorgelegt von

Neela Enke

aus Berlin

Oktober, 2008
Durchführung der Arbeit von Oktober 2005 bis Oktober 2008 an der Zentraleinrichtung
Botanischer Garten und Botanisches Museum Berlin - Dahlem, Freie Universität Berlin unter
der Leitung von Herrn Prof. Dr. T. Borsch.

1. Gutachter: Prof. Dr. T. Borsch

2. Gutachter: Prof. Dr. W. Greuter

Disputation am: 06.04.2009

 INDEX

Table of Contents

Chapter Title Page

1 Introduction 1
1.1 The Objective: The Influence of Karyotype Evolution on Species 1
Formation
1.1.1 Karyotype Alterations 1
1.1.2 Karyosystematics and Phylogeny 2
1.1.3 The Objective 4
1.2 The Object: Crepis as Model Group 6
1.3 Comments on the Structure of the Presented Study 11

2 Babcock Revisited: New Insights into Generic Delimitation and 13

Character Evolution in Crepis L. (Compositae: Cichorieae) from
ITS and matK Sequence Data
2.1 Introduction 14
2.2 Material and Methods 16
2.3 Results 19
2.4 Discussion 22
2.5 Summary and Conclusion 26
2.6 Acknowledgements 27
2.7 Literature Cited 27
2.8 Appendix 30

3 Shrinking Genomes? Evidence from Genome Size Variation in 33

Crepis L. (Compositae)
3.1 Introduction 34
3.2 Material and Methods 37
3.3 Results 40
3.4 Discussion 43
3.5 Summary and Conclusion 47
3.6 Acknowledgements 48
3.7 Literature Cited 48
3.8 Appendix 52
 INDEX

Chapter Title Page

4 In the Search of Additional Characters Supporting Systematic 55
Delimitation in Crepis L. and Related Genera in the Subtribe
Crepidinae
4.1 Introduction 56
4.2 Material and Methods 57
4.3 Results 59
4.4 Discussion 64
4.5 Summary and Conclusion 68
4.6 Acknowledgements 68
4.7 Literature Cited 70
4.8 Appendix 72

5 Guideline to a New Infrageneric System in Crepis L. 75

(Compositae/Cichorieae)
5.1 Introduction 76
5.2 Discussion 77
5.3 Summary and Conclusion 83
5.4 Literature Cited 84

6 Afroalpine Dianthoseris actually a congener of Crepis s.str. 87

(Compositae, Cichorieae, Crepidinae)
6.1 Introduction 88
6.2 Material and Methods 89
6.3 Results 92
6.4 Discussion 95
6.5 Taxonomy 98
6.6 Acknowledgements 101
6.7 References 101

7 Outlook 105

8 Summary 109
8.1 Summary 109
8.2 Zusammenfassung 111

9 Literature Cited 113

 INDEX

Chapter Title Page

10 Appendix 125
10.1 Publication List 125
10.1.1 Publications in Journals 125
10.1.2 Talks and Posters at National and International Conferences 125
10.1.2.1 Talks 125
10.1.2.2 Posters 126
10.2 Erklärung über den persönlichen Anteil an den Publikationen 127
10.3 Acknowledgements 129
INDEX
 1. INTRODUCTION: OBJECTIVE AND OBJECT

Chapter 1

Introduction

1.1 THE OBJECTIVE: THE INFLUENCE OF KARYOTYPE EVOLUTION ON SPECIES FORMATION

1.1.1 Karyotype Alterations

Karyotypes are the "phenotypic appearance of the somatic chromosomes, in contrast to their
genetic contents" (Levitsky, 1924, 1931). Number, shape, size and symmetry of
chromosomes, the distribution of hetero- and euchromatic regions as well as the
arrangement of genes and linking groups on the chromosomes define the karyotype of a
given organism. Chromosomal differences account for interbreeding barriers between
species leading to the formation of new species.

An important and often encountered mode of karyotypic alteration is polyploidisation,

especially in plants. Approximately 70% of the angiosperms have experienced at least one
episode of polyploidy in their ancestry (Masterson, 1994), whereas in animals polyploidy is
comparatively rare. Two major causes of polyploidisation are observed, one results from the
hybridisation of two different chromosome sets (allopolyploids) and the other is caused by a
multiplication of one chromosome set (autopolyploids) (Kihara & Ono, 1926).

Series of aneuploid changes of chromosome number are found in many genera throughout
the Angiosperms (e.g. Crepis, Babcock & Jenkins, 1943; Brachyscome, Watanabe et al.,
1999; Hypochaeris, Cerbah et al., 1998). Multiple translocations can cause such aneuploid
differences in chromosome number (Stebbins, 1950). Robertsonian fissions and fusions (the
breaking and rejoining of two nonhomologous acrocentric chromosomes at the centromer, in
which process the short arms can be lost), have also been proposed as mechanism which
can influence the chromosome number of species (Robertson, 1916; Jonson, 1998).
However, an increasing understanding of the genetic control of chromosome pairing (e.g.
Riley et al., 1959; Schifino & Moraes Fernandes, 1987; Naranjo & Corredor, 2004) and the
phylogenetic mapping of chromosome numbers (e.g. Watanabe et al., 1999; Cerbah et al.,
1998) suggest, that species relationships are not well portrayed by chromosomal data (Levin,
2002).

Substantial differences in genome size between closely related species indicate possible
reproductive isolation (Greilhuber & Ehrendorfer 1988; Bennett & Leitch, 2005). How genome
size variation influences speciation is still unclear. High mutational rates have been reported

1
 1. INTRODUCTION: OBJECTIVE AND OBJECT

for the Genlisea/Utricularia clade of the Lentibulariaceae, which comprise species with
ultrasmall genomes (e.g. 1C-value 0.065 pg in Genlisea aurea), whereas the sister group,
Pinguicula, shows neither ultrasmall genomes nor accelerated mutational rates (Greilhuber
et al., 2006). The “large genome constraint hypotheses” assumes that genera with large
genomes are less likely to speciate as accumulation and replication of “junk” DNA is
associated with evolutionary costs (Knight et al., 2005). Transposable elements, which
account for a high fraction of heterochromatic DNA, compete with other DNA sequence
segments for “resources”, e.g. material for replication (Gregory, 2005). This leads to the
hypothesis that genome size variation influences speciation.

Genome size is not correlated to the organisational level of an organism: more complex
organisms do not necessarily feature more DNA (“C-value Paradox”, Thomas, 1971; also
e.g. Vendrely, 1955; MacLean, 1973; Gall, 1981); the actual number of genes required for
normal development is similar in most plants (Flavell, 1980). Variation in genome size is
mostly due to the amount of heterochromatin (e.g. Flavell et al., 1977; Flavell, 1986; Barakat
et al., 1997). Questions as why there is such a considerable variation in non-coding DNA,
how it is distributed among taxa and how it developed are still unanswered (“C-value
enigma”, Gregory, 2001). Because of the self-replicating DNA elements, genomes are often
considered to have a “one way ticket to obesity” (Bennetzen & Kellogg, 1997). But recent
studies suggest that genomes can also “shrink” (e.g. Wendel et al., 2002). The reasons for
the variation of genome size are manifold: increases in DNA content can be due to the
accumulation of retroelements (e.g. SanMiguel & Bennetzen, 1998; Bennetzen, 1996, 2000)
and repeated circles of polyploidy (e.g. Soltis & Soltis, 1999; Otto & Whitton, 2000; Wendel,
2000). Decreases can be due to e.g. the loss of whole chromosomes or parts thereof (Dart et
al., 2004) or through intrachromosomal homologous recombination between LTR’s (long
terminal repeats, Vicient et al., 1999; Ma et al., 2004).

Gregory (2005) stated that genomes represent a distinct and legitimate level of biological
organisation, with their own inherent properties and unique evolutionary histories. The variety
of levels on which genomic or chromosomal changes occur renders karyotype and genome
evolution highly complex. Many facets of genomic evolution are up to now only poorly
understood.

1.1.2 Karyosystematics and Phylogeny

Karyotype similarities and differences, especially chromosome number have often been used
as criteria to infer species relationships (e.g. Babcock et al., 1937; Babcock, 1947a,b;
Gokhman, 2007; Guerra, 2008). Extensive contributions in the field of karyosystematics have
been made by E.B. Babcock and G.L. Stebbins through thorough cytological studies within
the Compositae subtribe Cichorieae (e.g. Babcock & Stebbins, 1937; Babcock & Jenkins,

2
 1. INTRODUCTION: OBJECTIVE AND OBJECT

1943; Babcock, 1947a,b; Stebbins et al., 1953). The cooperation of the two botanists led to
the first pre-cladistic “phylogenetic” hypotheses for confined plant groups: Based on a
phenetic multi-evidence approach by considering chromosomal data, geographical
distribution, and morphology, Babcock (1947a,b) postulated phylogenetic coherences within
the genus Crepis L (Compositae), later followed by a publication on phylogenetics within the
whole Compositae subtribe Cichorieae (Stebbins, 1953; Stebbins et al., 1953).
Subsequently, Stebbins (e.g. 1950) became one of the architects of the Modern Evolutionary
Synthesis, together with Dobzhansky (e.g. 1937, 1970), Mayr (e.g. 1942, 1997), and
Simpson (e.g. 1944).

Hennig, a German entomologist, is regarded the founder of modern evolutionary studies with
the publication of his “Phylogenetic Systematics” (Hennig, 1966). He proposed a cladistic
approach to infer the evolutionary history of a taxon. Species are classified hierarchically
based on evolutionary ancestry. Classical cladistic analyses use mainly morphological and
anatomical characters, but also cytological and biogeographical data.

Since the advent of DNA sequencing species relationships in evolutionary contexts are
mainly inferred by molecular data. Phylogenies based on molecular data have several
advantages: In very high or very low taxonomic ranks where morphological variation is often
too low or too high to reflect evolutionary patterns, a molecular approach is advantageous.
Furthermore, molecular data can be generated in relatively short time; whereas it is often
time consuming to assess morphological characters (Sudhaus, 2007). The virtually infinite
abundance of molecular characters allows for statistical approaches to infer the reliability of a
given reconstruction (Knoop & Müller, 2006). Still, other factors influence the reliability of a
phylogenetic reconstruction as well: First of all, a phylogenetic gene tree reflects the history
of the applied gene or marker and not necessarily that of the species (Maddison, 1997).
Furthermore, DNA from different cell organelles can reflect different evolutionary histories,
equivalent to the different evolutionary histories of the complete organelles. And marker loci
can vary in their mutational rates; coding gene regions are in general more conserved than
non coding regions.

One approach to obtain reliable phylogenetic histories for species groups is to use multiple
marker loci (e.g. Qiu et al., 1999; Graham & Olmstead, 2000; Soltis et al., 2000; Zanis et al.,
2002). Another important factor is the choice of the employed marker: chloroplast markers in
Angiosperms are passed on mostly in the maternal lineage, while nuclear markers show
biparental inheritance patterns. In plant phylogenetic studies chloroplast regions are widely
amplified and analysed because they exhibit a suitable rate of evolution for phylogenetic
reconstructions and are easily accessible. To trace the biparental history of taxa nuclear
markers are used. Inconsistencies of chloroplast and nuclear based reconstructions can be
used to identify potential reticulate evolution events (e.g. Rieseberg & Soltis, 1991). One of
the most widely used nuclear markers in angiosperm phylogeny is ITS (internal transcribed

3
 1. INTRODUCTION: OBJECTIVE AND OBJECT

spacer) (e.g. Baldwin et al., 1995). ITS is located within the highly conserved 18S-26S
nuclear ribosomal gene cluster and arranged in numerous tandem repeats. The manifold
applications using ITS are due to the high number of copies and the conserved flanking
regions which makes the spacer region easily accessible even with universal primers
(Baldwin et al., 1995); in some plant groups different paralogues are present, which reflect
complex inheritance patterns and therefore, complicate phylogenetic analysis (e.g. Álvarez &
Wendel, 2003). In most plant groups, however, the ITS paralogues show high uniformity,
attributed to rapid concerted evolution (e.g. Arnheim et al., 1980; Arnheim, 1983; Hillis et al.,
1991). At present the use of single copy nuclear genes is time and cost consuming, so they
are not yet widely used for routine analysis. Primer binding is often impeded by secondary
structures of the large genome and primer binding sites must be very specific so as to not
accidentally target orthologues. Furthermore, for the identification of suitable loci for
phylogenetic reconstructions of certain taxa and for primer design, sequence information of
closely related taxa are needed (Álvarez et al., 2008). The high potential of nuclear single
copy genes to reveal phylogenetic coherences makes their application increasingly common
(e.g. Strand et al., 1997; Sang, 2002; Zhang & Hewitt, 2003; Small et al., 2004; Álvarez et al.,
2008).

With the “molecular revolution” and the use of DNA sequences for phylogenetic analysis the
focus of karyological and morphological contributions to evolutionary studies has changed
(Endress, 2002): Structural similarities can illuminate cryptic molecular groups as well as
groups with a conflicting molecular and morphological background, and character
progressions can be reconstructed on molecular phylogenies (Endress, 2002).

1.1.3 The Objective

The present study investigates aspects of phylogeny, character evolution and systematic
classification in the genus Crepis. It aims to contribute towards the understanding of
character evolution and speciation mechanisms in higher plants. Based on the reconstruction
of a molecular phylogeny, the evolution of karyotypic changes, e.g. in chromosome number
or genome size, and its influence on diversification in diploid plant genera are of special
interest. Hypotheses on character evolution are postulated for the model group Crepis, which
might prove to be significant for character evolution in higher plants.

4
 1. INTRODUCTION: OBJECTIVE AND OBJECT

FIG. 1: Crepis and related genera. a) Crepis aurea, b) C. blattarioides c) C. tectorum d) C. jacquini, e) C. mollis,
f) C. chrysantha, g) C. sibirica, h) C. multicaulis, i) Youngia tenuifolia, k) Askellia nana.

5
 1. INTRODUCTION: OBJECTIVE AND OBJECT

1.2 THE OBJECT: CREPIS AS MODEL GROUP

Crepis is a genus within the Compositae (Syn: Asteraceae) tribe Cichorieae Lam. & DC.
(formerly Lactuceae Cass.) and subtribe Crepidinae. Following the traditional classification of
the Crepidinae (Ixeris-Youngia-Line (Stebbins, 1953)), Crepis series (Jeffrey, 1966) and
Crepidinae (Bremer, 1994)) the most closely related genera to Crepis were considered to be
Lapsana (sensu Pak & Bremer,
1995), Youngia (monographed by
Babcock & Stebbins, 1937,
Figs.1,3-4), Ixeris and Dubyaea
(treated by Stebbins, 1940). A
recent revision of the Cichorieae by
Kilian et al. (2008) taking molecular
sequence data into account largely
confirmed these relations. Due to
the lack of discriminating
characters for many genera within FIG. 2: Worldwide distribution of Crepis.
the Crepidinae some species have
been classified in the past under different generic names; e.g. the species of the genus
Askellia (Figs.1, 3-4). Babcock (1947a,b) already perceived the intermediate position of
Askellia between Crepis and Ixeris and Youngia (Fig.3), even though most of the species
(e.g. A. nana, A. flexuosa) were treated under Youngia to which they were recently
reassigned (Adylov & Zuckerwanik, 1993). Other genera (e.g. Ixeris by Pak & Kawano,
1990a,b,c, 1992; Youngia by Sennikov & Illarionova, 2007) have been subdivided into new
genera due to e.g. carpological and karyological characters.

Crepis is with over 200 species (Bremer, 1994) one the largest genera of the Crepidinae and
even of the Cichorieae. Species of the genus are distributed throughout the northern
hemisphere with single species occurring in South East Asia (Fig.2). Some species also
occur in tropical east Africa, South Africa and West Africa, as well as the Canary Islands and
Madeira. The origin of Crepis is thought to be in the Altai/Tien Shan region in Central Asia
(Babcock, 1947a). From there the genus spread north-eastward into North America, south-
westward into southern Europe and northern Africa and westward across the southern end of
the Ural Mountains into north-eastern Europe (Babcock, 1947a). The genus presently has its
highest species diversity in the circum-Mediterranean area.

6
 1. INTRODUCTION: OBJECTIVE AND OBJECT

FIG. 3: Growth form and habit of Crepis and related genera. a) Crepis pygmaea, b) C. multicaulis, c) C.
jacquini, d) Askellia nana, e) C. sibirica.

Crepis, as all Cichorieae, is notorious for its lack of discriminating characters. Many
characters vary more within a species than between closely related species. In the past this
often led to unclear specific and generic boundaries. The taxonomic history of the genus
Crepis is a long tale of lumping and splitting natural groups on various taxonomic levels.

7
 1. INTRODUCTION: OBJECTIVE AND OBJECT

FIG. 4: Floral characters of Crepis and related genera. a) ligules, style branches (Crepis sibirica), b) achenes,
pappus (C. sibirica), c) involucrum in two distinct rows, capitula (C. sibirica), d) achenes, pappus (C. tectorum), e)
achenes, pappus (Youngia tenuicaulis), f) glabrous and cylindrical involucrum, capitula (Y. tenuifolia), g) glabrous
and cylindrical (Askellia nana).

Linné (1753) assigned 13 species to Crepis (type species: C. biennis) of which 10 are still
part of the genus today. Moench (1794) accepted only 3 genera (Crepis, Catonia and
Barkhausia); where Cassini (1830) recognised 14 genera (Crepis, Anisoderis, Barkhausia,
Brachyderia, Catonia, Gatyona, Intybellia, Nemauchenes, Omalocline, Paleya, Phaecasium,
Pterotheca, and Zacintha). 60 years later Hoffmann (1889) assigned about 170 species to
Crepis comprising those genera listed in Cassini (1830) and several of the 130 species
mentioned in Bentham & Hooker (1873). The most recent revision of the genus Crepis has
been published by Babcock (1947a,b). Babcock’s monograph of the genus not only includes
detailed descriptions of 196 species, but extensive hypotheses on origin, phylogeny,
character evolution and speciation within the genus.

Crepis species occur in different types of habitats (Fig.3) ranging from alpine zones,
swamps, low grasslands, forests to beaches. Size ranges from only a few centimetres in
height (e.g. C. pygmaea) to nearly two meters in C. sibirica (Fig.3a,e). The capitula of Crepis

8
 1. INTRODUCTION: OBJECTIVE AND OBJECT

possess two distinct rows of involucral bracts (Fig.4c). Florets are ligulate (Fig.4a). Even
though the prevailing flower colour is the typical bright yellow of composites (Fig.1), some
species show other shades, like yellowy orange in C. aurea (Fig.1a) or whitish yellow in C.
albiflora. C. incarnata, C. rubra and C. incana have purple flowers. Some of the yellow
flowered species have a red dorsal stripe on the ligule. Corolla tubes are either pubescent or
glabrous with 5 ligule teeth; style branches are usually cylindrical and attenuate at the apex
(Fig.1a). The receptacle is either areolate or alveolate. Areoles can be separated by a
membranous ridge which is occasionally fibrillate. The fibrillae are replaced in rare cases by
palae. Colour ranges from dark green to light yellow. The achenes are narrowly terete to
fusiform, more or less attenuate, and sometimes beaked (Fig.4b). Some species (e.g. C.
sancta) have biform achenes. The simple pappus bristles are never plumose, scabrid
barbellulate and often pure white, seldom dusky or yellowish (Fig.4b,d). Except for the 15
North American species of section Psilochaenia and very few other species, the majority of
species is diploid. The basic chromosome number ranges from x=3 to x=6, respectively x=11
in the polyploid species of section Psilochaenia.

In his monograph, Babcock (1947a,b) defined 27 sections based on morphology,

geographical distribution, chromosome number and karyotype composition and assigned the
196 Crepis species accordingly. He assumed that the sectional classification reflected
phylogenetic coherences within the genus. He formulated several progressive character
changes; e.g. that chromosome number and size would decrease during evolution, while
chromosome asymmetry would increase. He furthermore postulated that tap rooted species
are derived and rhizomatous species are basal. He considered large species with few, big
heads, a high chromosome number, and a rhizome to be primitive (e.g. C. sibirica, Fig.33);
advanced species are generally more fragile in habit with many small flower heads, have a
lower chromosome number and are tap rooted (e.g. C. multicaulis, Fig.3b). In an often
overlooked publication he later withdrew his theory that rhizomes are basal, as
ontogenetically the taproot develops first (Babcock, 1949). This resulted in a minor
rearrangement in the phylogenetic order of the sections (Babcock, 1949). Babcock (1947a)
described karyotype evolution as driving force of speciation in Crepis L.. Following his
hypotheses, it is mainly a change in chromosome number that leads to interspecific sterility
and therefore to speciation. According to Babcock (1947a) hybridisation plays a minor role
for explaining speciation processes in the genus.

An essay on karyotype evolution in the Cichorieae was published in 1953 by Stebbins and
co-workers. The karyological studies on Crepis by Babcock & Jenkins (1943) contributed
considerably to a phylogenetic understanding of the tribe Cichorieae. Crepis traditionally is
important for karyological studies in higher plants. Even before Babcock (1947a,b) published
his monograph on the genus, Crepis was often the subject of studies on plant genetics (e.g.
Hollingshead, 1930a,b; Tobgy, 1943; Sherman, 1946), evolution, and speciation. Babcock

9
 1. INTRODUCTION: OBJECTIVE AND OBJECT

and Jenkins’ (1943) work on Crepis karyotypes was extensive but limited by the methods
available at the time (Fig.5). Since then many studies on different aspects in the field of
genetics and karyotype evolution in Crepis have been published (B chromosomes e.g. by
Maluzynska & Schweizer (1989), Maluzynska (1990), Jamilena et al. (1994); banding
patterns e.g. by Siljak-Yakovlev & Cartier (1982), Dimitrova & Greilhuber (2001);
chromosomal aberrations e.g. by
Dimitrov (1994); or the evolution of
quantitative karyotype characteristics
e.g. by Jones & Brown (1976),
Dimitrova & Greilhuber (2000)). With
the extensive systematic treatment of
FIG. 5: Karyotypes of C. neglecta and C. kotschyana the genus by Babcock (1947a,b)
reproduced from Babcock & Jenkins, 1943. based on morphological, karyological
and biogeographical characters a
solid foundation has been laid on which additional studies on the genus Crepis can build.
The low chromosome number, the predominant diploidy of species and the excellent
response of the chromosomes to staining procedures, make chromosomal features easily
accessible and comparison between species is facilitated without much effort. These
features make Crepis an excellent model group to study speciation due to karyotype
changes in (diploid) higher plants. Furthermore, Crepis species represent a large number of
growth forms and habitat adaptations and are broadly distributed.

10
 1. INTRODUCTION: OBJECTIVE AND OBJECT

1.3 COMMENTS ON THE STRUCTURE OF THE PRESENTED STUDY

The present study covers aspects of phylogeny, speciation, character evolution, and
systematic classification in the genus Crepis L. (Compositae, Cichorieae). Crepis is used as
model system representing the angiosperms; it is believed that the gained insights on
character as well as karyotype evolution can illuminate evolutionary processes in other
flowering plant genera.

The study is divided into two parts dealing with different aspects of evolution, speciation and
phylogeny in the genus Crepis. The first part discusses character evolution - especially
considering karyological traits as used by Babcock (1947a,b) to constitute the systematic
treatment of the genus - and phylogeny in a molecular framework. In the second part the
systematic classification of Crepis s.l. is reconsidered including molecular as well as
morphological characters.

The first part with the main focus on phylogenetic investigations and its implications for
karyological and morphological character evolution comprises two chapters (2,3). Chapter 2
illustrates the basic phylogenetic coherences within the genus. Molecular phylogenies based
on the nuclear marker ITS (internal transcribed spacer) and the chloroplast region matK
(maturase K) of 40% of accepted Crepis species are presented. General trends in character
evolution and the incongruence between the natural molecular groups and the current
taxonomic classification are discussed. Babcock’s hypotheses on character evolution and
speciation are revaluated. Chapter 3 includes a more detailed discussion of character
evolution and speciation and investigates genome size evolution within Crepis. Genome size
of 21 species is measured and the correlation to several factors (e.g. life form and
geographic distribution) is statistically tested. The direction of genome size variation during
evolution is reconstructed on a molecular phylogeny inferred from the nuclear marker ITS.

The second part evaluates in chapter 4 new characters for their applicability for generic and
infrageneric delimitation and comparative morphological studies. These characters include
achene morphology and anatomy, pappus ultrastructure and pollen grain morphology.
Chapter 5 discusses the taxonomic implications of the molecular phylogeny presented in
chapter 2 with a critical reassessment of the current systematic classification. Chapter 6
deals with a newly found congener of Crepis, former genus Dianthoseris, which in the past
was of unclear position within the Cichorieae.

Chapters 2,3 and 6 are published or submitted papers (see appendix), so chapters 4 and 5,
even though not published yet, follow in their internal structure publication requirements, viz.
the chapters are divided into abstract, introduction, material/methods, results, discussion,
acknowledgements, literature cited and appendix.

Chapter 7 presents perspectives on how results and data accumulated in the present study
could further be interpreted. It also discusses aspects and questions of species evolution

11
 1. INTRODUCTION: OBJECTIVE AND OBJECT

which were brought up during the present investigation and which await further treatment.
Chapter 8 summarises the insights gained from the different approaches to understand
character evolution, speciation, and species relations in the genus Crepis as model group for
higher plants.

12
 2. PHYLOGENY AND CHARACTER EVOLUTION

Chapter 2

Babcock revisited: New insights into generic delimitation and

character evolution in Crepis L. (Compositae: Cichorieae) from ITS
and matK sequence data

Neela Enke & Birgit Gemeinholzer

Botanic Garden and Botanical Museum Berlin-Dahlem, Freie Universität Berlin, Königin-Luise-Str. 6-8,
14195 Berlin, Germany. n.enke@bgbm.org

Published 2008, Taxon 57 (3): 756-768.

ABSTRACT
In 1947 Babcock published his widely acknowledged monograph of the genus Crepis L.
including a sectional classification of the species as well as extensive hypotheses about
character evolution. To reinvestigate Babcock’s evolutionary hypotheses and the generic
delimitation of Crepis L. a phylogenetic analysis was conducted using ITS and chloroplast
matK sequence data. The results revealed Crepis L. to be polyphyletic. A monophyletic clade
including Central Asian and North American species of Crepis section Ixeridopsis is clearly
isolated from Crepis sensu stricto and, as also supported by additional morphological
evidence, needs to be transferred to the genus Askellia Weber (1984).

A second clade comprising the genera Lapsana L. and Rhagadiolus Juss. as well as a
statistically strongly supported clade of several Crepis species is sister to a third clade: the
monophyletic Crepis s.str.. Within Crepis s.str. the molecular data do not support Babcock’s
sectional delimitation which is mainly based on his hypotheses about karyotype evolution.
Hence, morphological and karyological characters are re-assessed with regard to the
molecular phylogeny.

KEYWORDS: Askellia, Crepis, ITS, matK, phylogeny.

13
2. PHYLOGENY AND CHARACTER EVOLUTION

2.1 INTRODUCTION
The genus Crepis L. (Cichorieae: Crepidinae) comprises about 200 species and is distributed
throughout the northern hemisphere and Africa. The genus presumably originated in the
Pamir/Altai region in Central Asia (Babcock, 1947a). Presently the centre of diversity is the
circum-Mediterranean area.

In the first edition of Species Plantarum, Linné (1753) assigned 13 species to Crepis (Type:
Crepis biennis L.) of which 10 are still valid today. Hoffmann (1889) included about 170
species comprising those genera listed in Cassini (1830) (Crepis, Anisoderis, Barkhausia,
Brachyderia, Catonia, Gatyona, Intybellia, Nemauchenes, Omalocline, Paleya, Phaecasium,
Pterotheca, Zacintha) and several of the 130 species mentioned in Bentham & Hooker
(1873).

Babcock (1947a, b) was the last to revise the genus and assigned 196 species to 27
sections on the basis of morphological and karyological similarities. This included a priori
assumptions about character evolution. According to Babcock, primitive characters are being
perennial, exhibiting a rhizome, and a chromosome number of x = 6; derived features are
being annual, having a taproot, and a chromosome number of less than x = 6. In a rarely
cited publication Babcock (1949) withdrew the rhizome as basic character because seedlings
of rhizomatous species first develop a taproot which later is lost. Therefore a taproot is to be
considered the basic character state.

Babcock's interpretation of karyotype evolution is of special importance, since it tended to be

cited as exemplary (e.g. Stebbins, 1950; Briggs & Walters, 1984). Babcock defined several
karyomorphotypes depending on number and size of chromosomes and their symmetry, and
arranged the species into sections due to these chromosomal characteristics (Babcock &
Jenkins, 1943; Babcock, 1947a). He postulated karyotype rearrangements to be the driving
force of speciation while hybridisation between members of different karyomorphotypes is
largely inhibited (Babcock, 1947a).

To reinvestigate Babcock’s evolutionary hypotheses as well as the generic and infrageneric

delimitation of Crepis a molecular phylogeny based on ITS and matK sequence data has
been established. The applicability of the molecular markers ITS and matK to reveal intra-
and intergeneric relationships in the Cichorieae (Compositae) has already been
demonstrated in several studies (e.g. Baldwin, 1993; Goertzen et al.; 2003, Samuel et al.,
2003). Pre-studies on several chloroplast regions (trnS-trnR, trnS-trnFM, atpB-rbcL, trnS-
trnG, psbA-trnH, trnK-trnQ) have been carried out of which matK proved to be informative
and universally amplifiable.

14
 2. PHYLOGENY AND CHARACTER EVOLUTION

FIG. 1: Phylogram derived from the Bayesian Inference (ITS dataset). A, overview of the tree, the dashed line
indicates the two parts shown in B and C. B, upper part of the tree, all clades belong to Crepis s.str.; C, lower part
of tree, showing Askellia and Lagoseris groups as well as other Crepidinae. Bayesian posterior probabilities >
0.70 and bootstrap values >50% given above branches. Arrows with numbers indicate nodes discussed in the
text; a and b denote subclades of Lagoseris. Clades within Crepis s.str. are numbered as not to reflect any
infrageneric taxonomic system. Shaded clades are discussed in detail in the text.

15
2. PHYLOGENY AND CHARACTER EVOLUTION

FIG.1 (continued)

2.2 MATERIAL AND METHODS

Sampling – 200 samples from 102 Crepis species of 26 sections were obtained. Two
separate individuals (from different collections) per species were sampled whenever
possible. As outgroup closely related taxa of the Cichorieae such as Taraxacum, Ixeris,
Tolpis and Youngia were chosen. Samples were taken from dried herbarium specimens of
various herbarium collections (B, M, MSB, E, UPS and US). The sequences of the outgroup
taxa and some Crepis species were downloaded from NCBI (GenBank, EMBL).

A list of samples, voucher location and GenBank numbers is provided in the appendix.

DNA isolation, amplification and sequencing – Total genomic DNA was isolated from
dried herbarium specimen. 24 mg per sample of mostly leaf material was taken. The samples
were crushed and DNA was then extracted using Quiagen DNeasy Mini Kit and following
standard procedure.

The ITS and matK regions were amplified in two overlapping parts using the primers ITS-A
and ITS-C (Blattner, 1999) for ITS 1, ITS2-D, ITS-B (Blattner, 1999) for ITS 2 and trnK-710f

16
 2. PHYLOGENY AND CHARACTER EVOLUTION

(Johnson & Soltis, 1995) and matK-iR (Fehrer et al., 2007) for matK 1. For matK 2 specific
primers were designed: matK-ifN (5’-CATTCRAYATTTTCTTTTT-3’) and matK-rN (‘5-
TTATATAAATCCTTCCTG-3). For ITS the following protocol during the Polymerase Chain
Reaction (PCR) was used: initial denaturation 2 min at 94°C, denaturation 20 sec at 94°C,
annealing 45 sec at 52°C, elongation 1 min at 72°C (40 cycles) and final extension 10 min at
72°C. The matK region was amplified in 40 cycles under following conditions: denaturation
40 sec at 94°C, annealing 1 min 30 sec at 50°C, elongation 2 min at 72°C, preceded by initial
denaturation 2 min at 94°C and followed by a final extension 15 min at 72°C. PCR was
carried out with a reaction volume of 11.5µl core mix plus 1 µl DNA (1:10, 1:50 or 1:100
dilution depending on usability). The reaction volume contained 8.05 µl ddH2O, 1.25 µl 10x
buffer (Biotherm) and dNTP’s (Fermentas), 0.25 µl BSA (BioLabs), 0.25 µl of each primer
(10pmol/µl) and 0.04 µl 1u Taq-Polymerase (Biotherm). The PCR products were purified with
Milipore DNA purification Kit (Roth) and then cycle sequencing was carried out using CEQ
DCTS Quick Start Kit (Beckmann-Coulter) following the standard procedure. As sequencer a
CEQ8000 (Beckmann-Coulter) was used.

Sequence Alignment – The sequences were edited in ChromasLite2000 (Technelysium

Pty. Ltd., Helensvale, Australia) and aligned by hand using BioEdit (Hall, 1999) following
Goertzen et al. (2003) for ITS gap-coding. ITS alignment was sometimes ambiguous. To
avoid alignment ascendancies some positions were excluded (1--19, 118,222--223, 242,288-
-289, 372--376, 516--519, 555--556, 706--727). For matK indels at positions 118--122, 353--
366, 799--807 and 813--817 have been coded as one mutational step each. The alignments
are available from the first author upon request.

Phylogenetic Analyses – The trees were reconstructed using MrBayes 3.1.2 (Ronquist &
Huelsenbeck, 2003) and PAUP 4.0b10 (Swofford, 2002).

A Bayesian Analysis was performed on both datasets, using gamma distribution rate
variation among sites and 10 million generations of the MCMC chains in two independent
runs, trees saved every 100 generations. The first 30 000 trees were discarded as burn-in for
the analysis then reached stationarity. All other trees sampled were used to calculate a strict
consensus tree.

ITS and matK datasets were analysed using Maximum Parsimony. All heuristic searches
were conducted in Paup 4.0b10 with equal weights, 1000 closest sequence additions and
tree bisection-reconnection (TBR) branch swapping, permitting 10 trees to be held at each
step. An evaluation of the trees was performed by using bootstrap analysis with 1000
replicates, equal weights, TBR swapping, MulTrees option in effect and 10 trees held at each
step.

17
2. PHYLOGENY AND CHARACTER EVOLUTION

The strict consensus tree of Bayesian inference was compared to the bootstrap 50% majority
rule consensus tree.
Trees were drawn using TreeView (Page, 1996) and Adobe Illustrator (Adobe Systems, Inc.,
San Jose, California, USA).

FIG.2: Phylogram derived from the Bayesian Inference (matK dataset). Numbering of clades according
to ITS phylogeny (Fig. 1). For further explanation see Fig. 1.

18
 2. PHYLOGENY AND CHARACTER EVOLUTION

2.3 Results
Phylogeny – 123/73 (ITS/matK) sequences could be obtained from 78/52 Crepis taxa
representing 24/20 of the 27 sections. 22/13 accessions including the genera Ixeris
(ITS/matK), Taraxacum (ITS), Heteracia (ITS), Youngia (ITS/matK), Lapsana (ITS/matK),
and Rhagadiolus (ITS/matK) represent the Crepidinae. Sonchus (ITS/matK), and Hyoseris
(ITS) (Hyoseridinae) were chosen as outgroup taxa. Different accessions for ITS and matK
are due to sequence availability.

For the ITS Maximum Parsimony analysis both spacer regions as well as the 5.8rDNA
sequence have been analysed. In total 671 characters were included in the analysis, of
which 327 were parsimony informative (48.7%). For matK in total 943 characters were
included of which 191 were phylogenetically informative (20.3%).

The trees from Bayesian and Maximum Parsimony analyses were congruent, but the basal
nodes in the ITS Maximum Parsimony bootstrap analysis could not be resolved.

Fig.1 (ITS) and Fig.2 (matK) depict strict consensus trees from the Bayesian Analyses.
Concerning the distribution of Crepis both trees are congruent in their overall topology but
differ in their terminal branching patterns; with the exception of the Askellia clade, which is
sister to Lagoseris in the ITS tree but forms a monophyletic and unresolved group with Ixeris,
Youngia and Garhadiolus in the chloroplast phylogeny (Figs. 1, 2). The relations of the
outgroup taxa will be dealt with elsewhere.

The trees support three main clades comprising species of Crepis s.l. (Bayesian posterior
probabilities (PP) above 70% and bootstrap values (BS) above 50% are given separately for
each marker): The first clade is the above-mentioned one of unresolved relations among the
outgroup taxa. It includes only species of Babcock's section Ixeridopsis (ITS: PP1.00/BS100;
matK: PP1.00/BS99) and was described as genus Askellia by Weber (1984).

The second main clade comprises the genera Lapsana and Rhagadiolus, as well as a clade
which comprises Crepis species from Babcock’s sections Intybellia, Lagoseris, Phaecasium,
Microcephalum, and Pterotheca (ITS: PP1.00; matK: PP1.00 ). This clade preliminary was
named Lagoseris. The name, however, does not refer to any generic circumscription as the
group needs further validation. The clade is sister to Crepis s.str. (ITS: PP0.70; matK:
PP1.00/BS74 ; Node 2, Figs. 1, 2). In the ITS phylogeny the Crepis species of clade
Lagoseris fall into two groups a and b (ITS: PP0.97; Fig. 1); subclade a comprises the
species of sections Lagoseris, Microcephalum, and Pterotheca, subclade b those of sections
Intybellia and Phaecasium. The latter is polyphyletic in the matK phylogeny.

All other species sampled belong to Crepis s.str., which is monophyletic (Node 3; Figs. 1, 2)
in both the ITS and the matK analyses (ITS: PP1.00; matK: PP1.00/BS79). In the matK
analysis several clades lack resolution within Crepis s.str. due to low variation of the
chloroplast marker, thus clades are defined and named following the ITS based phylogeny.

19
2. PHYLOGENY AND CHARACTER EVOLUTION

The clades are numbered as not to suggest any infrageneric taxonomic system. Within
Crepis s.str. six out of eleven clades are supported with PP > 0.90. The clades do not
coincide with Babcock’s sections except for Clade I. Clades II, IV, V, VII, IX, X and XI are
highly heterogeneous in respect to Babcock’s sections, distribution and habit. At present,
only Clades I, III, VI and VIII provide feasible results for detailed discussion, while for the
other more detailed analyses are necessary.

Clade I is monophyletic in both analyses and comprises only members of section Mesomeris
(ITS: PP1.00/BS73; matK: PP1.00/BS60).

Clade III (ITS: PP1.00) forms a monophyletic group in the ITS based phylogeny, but lacks a
monophyletic origin in the matK analysis (Figs. 1, 2). It comprises mainly species of section
Hostia joined by taxa from other sections. Within Clade III C. zacintha and C. pusilla (section
Zacintha) form a very close relationship (ITS: PP1.00/BS100; matK: PP1.00).

Clade VI (ITS: PP0.83/BS73; matK: PP1.00/BS92) includes C. neglecta, C. corymbosa, C.

fuliginosa, and C. cretica, only members of section Phytodesia, but other species of
Babcock's section Phytodesia are also found elsewhere, e.g. C. capillaris within Clade XI and
C. nicaeënsis in Clade II.

Clade VIII comprises C. leontodontoides (section Gephyroides) and C. aurea (section

Brachypodes). Their association is statistically well supported (ITS: PP1.00/BS84; matK:
PP1.00/BS99) by both markers.

Character Evolution – To investigate Babcock’s overall hypotheses on progressive

character evolution in Crepis L and the suitability of these characters for infrageneric
delimitation three morphological characters (basic chromosome number/life history/root type)
upon which Babcock's phylogenetic hypotheses are based were mapped on the molecular
phylogeny of the ITS sequence data. According to Varela et al. (2004) it can be assumed
that the nuclear marker reflects the evolution of morphological traits. Fig. 3 shows the
distribution of all three applied characters to be variable in and between clades of the tree.
Babcock (1947a) implied a direction of evolution in his hypotheses: chromosome number
should always decrease and annuals should derive from perennials. Concerning the root
type he withdrew his initial hypothesis that taproots are derived but still implied a directional
character development (Babcock, 1949). Our molecular data do not support a unidirectional
character development.

20
 2. PHYLOGENY AND CHARACTER EVOLUTION

FIG.3: Characters (1, 2, 3) mapped on Bayesian strict consensus tree (ITS). Babcock's sections are
given in parentheses behind each taxon. Numbers designate clades within Crepis s.str.

21
2. PHYLOGENY AND CHARACTER EVOLUTION

2.4 DISCUSSION
Phylogeny – The Askellia clade comprises only species belonging to Babcock’s section
Ixeridopsis. The section comprises seven species of which only C. nana and C. flexuosa
have been investigated on a molecular base. However, all members of section Ixeridopsis
are morphologically and cytologically very similar to each other and differ distinctly from
Crepis s.str. in the tuft-like growth, the absence of hairs, the nearly always entire leaves, the
smaller number of florets, and the chromosome number (x = 7) that is unique in Crepis s. l. In
closely related genera x = 7 is only known for Ixeridium (Pak & Kawano, 1992) which is
molecularly isolated from Askellia in the ITS phylogeny and differs in featuring compressed
achenes; Ixeris and Youngia have a chromosome number of x = 8 (Pak & Kawano, 1992,
Babcock & Stebbins, 1937). MatK sequence data also suggests a close relationship of
Askellia to Ixeris, Youngia, and Garhadiolus. The members of section Ixeridopsis resemble
Ixeris in habit but differ in achene morphology (Babcock, 1947b). Achenes are terete in
Askellia and fusiform and flattened in Ixeris and Ixeridium. Youngia differs in chromosome
number (x = 8) and by forming compressed achenes (Babcock & Stebbins, 1937).
Garhadiolus is characterised amongst other features by minutely hispid or scabrid achenes
(Jaubert & Spach, 1847-50), which are not present in Askellia.

According to the molecular data Askellia indeed is transitional between Ixeris/Ixeridium and
Crepis s.str. as Babcock suggested (1947a,b), but not part of Crepis s.str.. As Lapsana and
Rhagadiolus differ in their generic circumscription substantially from all of these genera (see
discussion below), it is not reasonable to include them into Crepis s.l.. So following the
molecular, karyological, and morphological evidence, it is necessary to reassign the seven
species of Crepis section Ixeridopsis to Askellia, a genus which was defined by Weber
(1984) based on the unique number of chromosomes, including all species of section
Ixeridopsis.

The second main clade comprises Lapsana and Rhagadiolus as well as Crepis species from
sections Intybellia, Lagoseris, Phaecasium, Microcephalum, and Pterotheca. In both
analyses all sampled species of these five sections appear in this clade, indicating a
common monophyletic origin.

Lapsana is traditionally considered to be closely related to Crepis (Youngia-Ixeris-Line in

Stebbins (1953), Crepis series by Jeffrey (1966) and Crepidinae in Bremer (1994)).
Presently, European L. communis is considered to be the only species of Lapsana, while the
East Asian species have been separated as Lapsanastrum by Pak & Bremer (1995).
Lapsana is far more closely related to Crepis than to Lapsanastrum (Pak & Bremer, 1995).

Rhagadiolus has been treated as member of the Leontodontinae (Stebbins, 1953) or

Hypochaerinidae (Bremer, 1994), different terms for the same generic combination
(comprising e.g. Hypochaeris, Picris, Leontodon). Due to molecular evidence Rhagadiolus

22
 2. PHYLOGENY AND CHARACTER EVOLUTION

was included in the Crepidinae close to Lapsana as sister clade to Crepis (Whitton et al.,
1995; Gemeinholzer et al., 2006). The basic chromosome number of Lapsana communis is x
= 7 (Pak & Bremer, 1995), that of Rhagadiolus is x = 5 (Meikle, 1979).

The group of Crepis species in the Lagoseris clade fall into two subgroups (a and b; Figs.1,
2).

Subclade a comprises sections Lagoseris, Microcephalum, and Pterotheca. These species

have a chromosome number of x = 5. The karyotype of C. sancta shows strong
resemblances to the one of C. multicaulis even though Babcock (1947b) placed them into
two different sections: Pterotheca and Microcephalum. In the Flora of the USSR (Bobrov &
Tzelelev, 2000) Lagoseris and Pterotheca are treated as subgenera of genus Lagoseris
M.Bieb., amongst other features for their peculiar bristles on the receptacle which can
exceed the achenes.

Subclade b comprises the species of the rhizomatous section Intybellia and the tap rooted
section Phaecasium. All of these species possess a chromosome number of x = 4 (Babcock,
1947b) and their karyotypes are highly similar. The karyotype of these species is distinct
from those of other species of Crepis s.l., e.g. in the A chromosome which has a proximal
arm which almost equals the distal one, a feature that does not occur in any other Crepis
species. Babcock considered combining these two sections were it not for their different root
habit (Babcock & Jenkins, 1943).

The putative close relation of Lapsana and Rhagadiolus to the five sections of Crepis is
unexpected as they are both distinct in fruit morphology. The most obvious difference
between the genera is the absence of a pappus in Lapsana and Rhagadiolus. In Crepis s.l.
the pappus might be much reduced (e.g. C. patula) but is always present. Traditionally this
has been an important trait in genus delimitation (Cassini, 1830; Hoffmann, 1889).
Nevertheless, recent studies show that this character can vary within genera or even within
species (e.g. Lasthenia; Chan et al., 2002). Furthermore, the achene of Lapsana is strongly
compressed and the fruits of Rhagadiolus are only negligibly ribbed and densely hispidulous
(Meikle, 1979), all features not known from Crepis s.l.. The chromosome number of Lapsana
(x = 7) is unique within this clade, whereas Rhagadiolus features the same number as some
of the Crepis taxa (x = 5).

Morphologically the whole Lapsana/Rhagadiolus/Lagoseris clade is far more variable and

less easy to define than the Askellia clade and a formal taxonomic revision will require
additional studies. At the present stage it seems appropriate to conserve the genera Lapsana
and Rhagadiolus due to their morphological distinctness and combine the Crepis sections
Intybellia, Lagoseris, Phaecasium, Microcephalum, and Pterotheca as new genus Lagoseris
M.Bieb..

23
2. PHYLOGENY AND CHARACTER EVOLUTION

The third main clade, Crepis s.str., comprises all other Crepis species sampled. Out of
eleven monophyletic clades of the ITS analysis, only one coincides with the sections defined
by Babcock (1947a,b). This is Clade I, recognized both in the ITS and the matK trees, that
corresponds to Babcock's section Mesomeris. The species of section Mesomeris are
rhizomatous, have a basic chromosome number of x = 6, and are distributed in the east
Mediterranean, except C. mollis, that occurs throughout middle Europe, C. willementioides in
the Far East, and C. lyrata in Southern Siberia.

Clade III is monophyletic in the ITS phylogeny but forms two clades in the chloroplast based
tree. It comprises mainly species from section Hostia joined by taxa from sections Zacintha
(C. zacintha, C. pusilla) and Berinia (C. taygetica, C. triasii). The species of section Hostia
are annual and monocarp as are C. zacintha and C. pusilla. The typically carinate involucral
bracts of section Hostia can also be found in the other species. The close relationship
between C. zacintha and C. pusilla, as suggested by Merxmueller (1968) due to
morphological similarity, is corroborated by the molecular results. These two species not only
form a molecular and morphological subgroup within Clade III, but are sister to Clade VI in
the chloroplast phylogeny, exhibiting a genetic as well as morphological isolation from their
presumed closest relatives (Figs.1, 2).

Even though other members of section Phytodesia are found elsewhere in the tree (Clades II
(C. nicaeënsis), XI (C. capillaris)), Clade VI partly reflects Babcock’s assumptions about the
relationships within section Phytodesia. He placed C. neglecta, C. corymbosa, C. fuliginosa,
and C. cretica in one of four subgroups of section Phytodesia due to their peculiar narrow
chromosomes which are distinct from those of the other members of this section (Babcock,
1947b).

The close molecular relationship between C. leontodontoides and C. aurea of Clade VIII
mirrors their similarity in karyotype and morphology. Babcock already recognised these
similarities but assigned these two species to different sections (Gephyroides and
Brachypodes) due to their different root types.

Character Evolution – Fig. 3 depicts the three main morphological characters on which
Babcock’s classification (Babcock, 1947a) is based in relation to the molecular phylogenetic
reconstruction (ITS). Babcock’s hypotheses allow for multiple evolutionary character state
emergences which, once developed, only change in a directional manner such as
chromosome numbers only increase, tap rooted plants derive from rhizomatous ones, or
changes occur only from annual to perennial life history. Babcock (1947a) rejected
hybridisation as main cause of speciation in the genus whereas in his view karyotypic
changes were the driving force of evolution. His understanding of phylogenetic coherence

24
 2. PHYLOGENY AND CHARACTER EVOLUTION

was strongly influenced by his time and predated the publication of Hennig’s pioneering
Phylogenetic Systematics (Hennig, 1966).

The basic chromosome number, upon which all of Babcock’s hypotheses about character
evolution are based, is highly variable in and between the molecular clades and thus
contradicts Babcock’s hypotheses on progressive character evolution and speciation in the
genus (Fig.3). An increase of chromosome number occurred several times during evolution
(e.g. from x = 4 (C. nicaeënsis, C. tectorum) to x = 6 (e.g. C. jaquinii, C. paludosa) in Clade II
and from x = 4 (e.g. C. sonchifolia) to x = 5 (C. guioliana) in Clade V; Fig.3). Both reduction
and rise in chromosome number throughout evolution are known from other genera of the
Compositae, e.g. Hypochaeris (Cerbah et al., 1998). The chromosome number can also vary
between closely related species like C. pusilla (x = 5) and C. zacintha (x = 3) (Merxmueller,
1968). The only clade being characterised by its chromosome number is Askellia (Weber,
1984). The fact that chromosome number is highly variable within Crepis s.str., even among
closely related species, suggests a broad capacity of this genus to undergo karyotype
changes.

Chromosome number alone does not seem to indicate phylogenetic relationships. The
suitability of chromosome number as delimiting character for infrageneric groups has been
rejected for other Compositae as well, e.g. Artemisia (Torrell et al., 1999) and even for
closely related genera, e.g. the genus Hypochaeris (Parker, 1975). In general, variation in
chromosome number is known from several groups throughout herbaceous Angiosperms
(Levin, 2002; e.g. Brachyscome Cass. (Compositae, Watanabe et al., 1999) Lopezia Cav.
(Onagraceae, Plitmann et al. 1975; O’Kane et Schaal, 1998), Clarkia Pursh (Onagraceae,
Lewis, 1952; Gottlieb & Ford, 1996,), as well as in the centrospermous (Ehrendorfer, 1976)
and basal angiosperms (Grant, 1982)). Karyotypes are characterised not only by the number
of chromosomes but genome size and the distribution of gene loci on the chromosomes.
Chromosomal rearrangements generally exhibit a low level of homoplasy and it is thought
unlikely that the same chromosomal patterns appear in different phylogenetic lineages
(Lysak et al. 2006). Furthermore, it is known that different karyotypic traits are not
mandatorily linked to each other; e.g. DNA content, which depends on the genome size, and
chromosome number (Levin, 2002). According to Jones & Brown (1976) Crepis species with
divergent chromosome numbers can have a similar DNA content (e.g. C. neglecta and C.
rubra), while other species exhibit the same chromosome number but differ in DNA content
(e.g. C. capillaris and C. zacintha). Crepis species with identical chromosome numbers might
exhibit a different distributional pattern of gene loci. Therefore, an ostensible homoplasy of
chromosome number should be doubted. Given the complexity of karyotypic traits more
detailed analyses of chromosome morphology and structure are necessary to explain the
evolutionary pattern of the development of karyotypes within Crepis.

25
2. PHYLOGENY AND CHARACTER EVOLUTION

The varying chromosome numbers within the Crepis s.str. clades and the different
phylogenetic relationships of taxa in the nuclear and plastid phylogeny suggest hybridisation
between species, maybe even species with different chromosome numbers. Presumably
similar chromosome morphology and structure could allow for hybridisation across taxa and
might therefore be more indicative of phylogenetic relations than chromosome number alone.
For a detailed analysis of reticulate evolution in the genus, however, more data on
chloroplast relationships is needed.

Rhizomes are most common within Clades I and II, but can also be found in other clades (IV,
V, VII, VIII, and XI). Rhizomes most likely evolved as environmental adaptation to humid
habitats and have arisen several times within Crepis s.str. (Fig.3). Annual life history has
been developed mainly in two clades (Clade III and Clade VI) but can also sporadically be
found elsewhere (Clades II, VII, XI) (Fig.3). According to the molecular reconstruction
multiple origins for life history and root type as well as changes from rhizome to tap root (e.g.
rhizomatous C. jaquinii to tap rooted C. lacera, Clade II) and from annual to perennial life
history (e.g. annual C. capillaris to perennial C. pygmaea, Clade XI) can be demonstrated
(Fig.3).

Studies showed that edaphic heterogeneity has been crucial for the evolution of growth form
and life history in grass species (Verboom et al., 2004). The close association of root type
and life history to environmental influences could imply that these characters only exhibit a
weak phylogenetic signal and might therefore be inapt for the analysis of phylogenetic
relationships. Summarising the above, based on our analysis the examined character
variability within the clades does not allow for hypotheses about progressive character
changes.

2.5 SUMMARY AND CONCLUSIONS

The ITS phylogeny revealed Crepis L. sensu Babcock to be polyphyletic, which is also
confirmed by the chloroplast marker matK. Three main groups comprising species of Crepis
s.l. can be deduced from the molecular data. The first one includes the species of Crepis
section Ixeridopsis and corresponds to the genus Askellia (Weber, 1984); the second one
comprises the genera Lapsana and Rhagadiolus as well as Crepis species from sections
Intybellia, Lagoseris, Phaecasium, Microcephalum, and Pterotheca. In case further
morphological studies confirm the outcomes of the presented molecular data it seems most
reasonable to transfer the Crepis species of this clade to the genus Lagoseris, rather than to
fuse them with Lapsana and Rhagadiolus, which feature differences in achene and karyotype
morphology. The third and largest group comprises the majority of sampled Crepis species

26
 2. PHYLOGENY AND CHARACTER EVOLUTION

as Crepis s.str.. Within this group the composition of clades deviates from Babcock’s
sectional concept.

The results of the molecular analyses confirm only parts of Babcock’s sectional arrangement
but contradict his hypotheses about character evolution in the genus. His assumption of
absent infrageneric hybridisation cannot be maintained as the differences between the
nuclear and chloroplast marker hint on crossings between taxa of different sections, even
between taxa featuring distinct chromosome numbers.

According to the here presented data karyotype evolution in the genus is far more complex
than Babcock assumed. Yet, chromosomal similarities in karyomorphotypes, which could
explain hybridisation between taxa of different base chromosome numbers, might provide
promising evidence for subsequent analyses. To shed more light on both infrageneric
delimitation and karyotype evolution further investigations into karyotype morphology and
hybridisation are essential, potentially leading to a refined concept of infrageneric delimitation
within Crepis.

2.6 ACKNOWLEDGEMENTS
We thank the curators of B, M, MSB, E, UPS and US for providing plant material. We are
also obliged to Konrad Bachmann and two anonymous reviewers for their helpful
suggestions on the manuscript and Norbert Kilian for supporting us with plant material as
well as valuable advice throughout the project. This research was funded by the DFG GE
1242/3-1.

2.7. LITERATURE CITED

Babcock, E.B. 1947a. The Genus Crepis. Part One: The Taxonomy, Phylogeny, Distribution and
Evolution of Crepis. University of California Publications 21. University of California Press,
Berkeley & Los Angeles.
Babcock, E.B. 1947b. The Genus Crepis. Part Two: Systematic Treatment. University of California
Publications 22.University of California Press, Berkeley & Los Angeles.
Babcock, E.B. 1949. Supplementary notes on Crepis. II. Phylogeny, distribution and Matthew’s
principle. Evolution 3: 374--376.
Babcock, E.B. & Jenkins, J.A. 1943. Chromosomes and phylogeny in Crepis, III: The relationships of
one hundred and thirteen species. Univ. Calif. Publ. Agric. Sci. 18: 241--292.
Babcock, E.B. & Stebbins, G.L. 1937. The Genus Youngia. Carnegie Institution of Washington
Publication 484. Carnegie Institution of Washington, Washington, DC.

27
2. PHYLOGENY AND CHARACTER EVOLUTION

Baldwin, B.G. 1993. Molecular phylogenetics of Calycadenia (Compositae) based on ITS sequences
of nuclear ribosomal DNA: chromosomal and morphological evolution reexamined. Amer. J.
Bot.. 80: 222--238.
Bentham, G. & Hooker, J.D. 1873. Genera Plantarum 2. Lowell Reeve, London.
Blattner F.R. 1999. Direct amplification of the entire ITS region from poorly preserved plant material
using recombinant PCR. BioTechniques 27: 1180--1186.
Bobrov, E.G. & Tzevelev, N.N. 2000. Compositae. Tribe Cichorieae. 691-707. In: Komarov, V.L.
Flora of the USSR vol. 29. Science Publishers, Inc., Enfield.
Bremer, K. 1994. Asteraceae. Cladistics and Classification. Timber Press, Portland, Oregon.
Briggs, D. & Walters, S.M. 1984. Plant Variation and Evolution. 2nd edition. Cambridge University
Press, Cambridge.
Cassini, H. 1830. Tableau synoptique des Synanthérées. In: Cuvier, F. (ed.), Dict. Sci. Nat. 60: 566--
587. Paris.
Cerbah, M., Souza-Chies, T., Jubier, M.F., Lejeune, B. & Siljak-Yakovlev, S. 1998. Molecular
phylogeny of the genus Hypochaeris using internal transcribed spacer of nuclear rDNA:
Inference for chromosomal evolution. Molec. Biol. Evol. 15: 345--354.
Chan, R., Baldwin, B.G. & Ornduff, R. 2002. Cryptic goldfields: A molecular phylogenetic
reinvestigation of Lasthenia californica sensu lato and close relatives (Compositae: Helianthae
sensu lato). Amer. J. Bot. 89: 1103--1112.
Ehrendorfer, F. 1976. Chromosome numbers and differentiation of centrospermous families. Pl. Syst.
Evol. 126: 27--30.
Fehrer, J., Gemeinholzer, B., Chrtek, J. Jr. & Bräutigam, S. 2007. Incongruent plastid and nuclear
DNA phylogenies reveal ancient intergeneric hybridization in Pilosella hawkweeds (Hieracium,
Cichorieae, Asteraceae). Molec. Phylog. Evol. 42: 347--361.
Gemeinholzer, B., Kilian, N. & Lack, W.H. 2006. Molecular phylogenetics and systematics of
Cichorieae (Asteraceae). In: Abstracts. The International Compositae Alliance (TICA - Deep
Achene, Barcelona, July 2006). Barcelona.
Goertzen, L.R., Cannone, J.J., Gutell, R.R. & Jansen, R.K.. 2003. ITS secondary structure derived
from comparative analysis: implications for sequence alignment and phylogeny of the
Asteraceae. Molec. Phylog. Evol. 29: 216--234.
Gottlieb, L.D. & Ford, V.S. 1996. Phylogenetic relationships among the sections of Clarkia
(Onagraceae) inferred from the nucleotide sequences of PgiC. Syst. Bot. 21: 45--62.
Grant, V. 1982. Chromosome number patterns in primitive angiosperms. Bot. Gaz. 143: 390--394.
Hall, T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for
Windows 95/98/NT. Nucleic Acids Symp. Ser. 41: 95--98.
Hennig, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana.
Hoffmann, O. 1889. Compositae. 373-374. In: Engler, A. & Prantl, K. (eds). Die natürlichen
Pflanzenfamilien, IV. vol. 5. Engelmann. Leipzig.
Jaubert, H.P. & Spach, E. 1847-50. Illustrationes Plantarum Orientalium 3. Paris.
Jeffrey, C. 1966. Notes on Compositae I. The Cichorieae in East Tropical Africa. Kew Bull. 18: 427--
486.
Johnson, L.A. & Soltis, D.E. 1995. Phylogenetic inference in Saxifragaceae sensu stricto and Gilia
(Polemoniaceae) using matK sequences. Ann. Missouri Bot. Gard. 82: 149--175.
Jones, R.N. & Brown, L.M. 1976. Chromosome evolution and DNA variation in Crepis. Heredity 36:
91--104.

28
 2. PHYLOGENY AND CHARACTER EVOLUTION

Levin, D.A. 2002. The Role of Chromosomal Change in Plant Evolution. Oxford University Press, New
York.
Lewis, H. 1952. Chromosome phylogeny and habitat preference in Clarkia. Evolution 7: 102--109.
Linné, C. 1753. Species Plantarum. Stockholm.
Lysak, M.A., Berr, A., Pecinka, A., Schmidt, R., McBreen, K. & Schubert, I. 2006. Mechanisms of
chromosome number reduction in Arabidopsis thaliana and related Brassicaceae species.
PNAS 103: 5224--5229.
Meikle, R.D. 1979. Rhagadiolus: A misunderstood genus. Taxon 28: 133--144.
Merxmueller, H. 1968. Melitella (Cichoriaceae) - Über ein Vorkommen in Australien und die
taxonomische Einreihung. Mitteilungen der botanischen Staatsammlung München 7: 271--
275.
O’Kane Jr., S.L. & Schaal, B.A. 1998. Phyolgenetics of Lopezia (Onagraceae): Evidence from
chloroplast DNA restriction sites. Syst. Bot. 23: 5--20.
Page, R.D.M. 1996. TREEVIEW: An application to display phylogenetic trees on personal computers.
Comput. Appl. Biosci. 12: 357--358.
Pak, J. & Bremer, K. 1995. Phylogeny and reclassification of the genus Lapsana (Asteraceae:
Lactuceae). Taxon 44: 13--21.
Pak, J. & Kawano, S. 1992. Biosystematic Studies on the Genus Ixeris and its Allied Genera
(Compositae-Lactuceae). Mem. Fac. Sci. Kyoto. Univ. (Ser. Biol.) 15: 29--61.
Parker, J.S. 1975. Aneuploidy and isolation in two Hypochaeris species. Chromosoma 52: 89--101.
Plitmann, U., Raven, P.H., Tai, W. & Breedlove, D.E. 1975. Cytological studies in Lopezieae
(Onagraceae). Bot. Gaz. 136: 322--332.
Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed
models. Bioinformatics 19: 1572--1574.
Samuel, R., Stuessy, T.F., Tremetsberger, K., Baeza, C.M. & Siljak-Yakovlev, S. 2003.
Phylogenetic Relationships among species of Hypochaeris (Asteraceae, Cichorieae) based on
ITS, plastid trnL intron, trnL-F spacer, and matK sequences. Amer. J. Bot. 90: 496--507.
Stebbins, G.L. 1950. Variation and Evolution in Plants. Columbia University Press, New Yorck.
Stebbins, G.L. 1953. A new classification of the tribe Cichorieae, family Compositae. Madroño 12: 65-
-81.
Swofford, D.L. 2002. PAUP*: Phylogenetic analyses using parsimony (* and other methods). 4.0
Beta. Sinauer Associates, Sunderland, MA.
Torrell, M., Garcia-Jacas, N., Susanna, A. & Vallès, J. 1999. Phylogeny in Artemisia (Asteraceae,
Anthemideae) inferred from nuclear ribosomal DNA (ITS) sequences. Taxon 48: 721--736.
Varela, E.S., Lima, J.P.M.S., Galdino, A.S., Pinto, L.S., Bezerra, W.M., Nunes, E.P., Alves, M.A.O.
& Grangeiro, T.W. 2004. Relationships in the subtribe Diocleinae (Leguminosae;
Papilionoideae) inferred from internal transcribed spacer sequences from nuclear ribosomal
DNA. Phytochemistry 65: 59-69.
Verboom, G.A., Linder, H.P., & Stock, W.D. 2004. Testing the adaptive nature of radiation: growth
form and life history divergence in the African grass genus Ehrharta (Poaceae:
Ehrhartoideae). Amer. J. Bot 91: 1364-1370.
Watanabe, K., Yahara, T., Denda, T. & Kosuge, K. 1999. Chromsomal evolution in the genus
Brachyscome (Asteracea, Asterae): Statistical tests regarding correlation between changes in
karyotype and habit using phylogenetic information. J. Pl. Res. 112: 145--161.

29
2. PHYLOGENY AND CHARACTER EVOLUTION

Weber, W.A. 1984. New names and combinations, principally in the Rocky Mountain Flora IV.
Phytolgia 55: 1--11.
Whitton, J., Wallace, R.S. & Jansen, R.K. 1995. Phylogenetic relationships and patterns of character
change in the tribe Lactuceae (Asteraceae) based on chloroplast DNA restriction site variation.
Canad. J. Bot. 73: 1058--1073.

2.8 APPENDIX
Taxa sampled for DNA.

GENUS, Section, Taxon, Voucher, Origin, GeneBank Accesion No. (ITS/matK)

CREPIS, Desiphylion, C. paludosa Moench, P. Brückner 478/76 (B), Germany: Brandenburg, ITS EU366428; C.
viscidula Froel. ssp. geracioides (Hausskn.) Kamari; K.H. Rechinger 20933 (B), Greece: Mt.Smolika, ITS
EU363629; Omalocline, C. pygmaea L. ssp. anachoretica Babc, B. Valdes et al. It1968/88 (B), Spain: Granada,
ITS EU363638/matK EU363569; Brachypodes, C. aurea (L.) Cass. ssp. aurea, K.Faber AU87 (B), Austria:
Tauern, ITS EU363627/matK EU363564; C. chrysantha (Ledeb.) Turcz. ssp. chrysantha, E. Raab-Straube
020201 (B), Russia: Altay Rep., ITS EU363622/matK EU363560; C. dioritica Schott et Kotschy ex Boiss., H.Kehl
8/II-C (B), Turkey: Antalya, ITS EU363620; C. jaquini Tausch spp. kerneri Merxm., R. Vogt et al. RV 15954 (B),
Italy: Friaul, matK EU363568; C. jaquini Tausch, ITS AJ633378; C. kerneri Rech. f., T. Wraber 9748/4 (B),
Slowenia, ITS EU363636; C. kerneri Rech. f., matk AJ633152; C. rhaetica Hegetschw., ITS AJ633379;C. rhaetica
Hegetschw., A. Kuhns & Ch. Zidorn 98-00204 (priv.), Austria, matK EU363537; Mesomeris, C. hierosolymitana
Boiss., A. Liston 7-82-30/2 (M), Palaestina: Nahal Bezet, matK EU363543; C. hierosolymitana Boiss., H. Roessler
5317 (M), Libanon: Jabal Barouk, ITS EU363602; C. lapsanoides (Gouan) Tausch, H. Kalheber 88-2748 (M),
France: Dpt.Hts.Pyr., ITS EU363599; C. mollis (Jacq.) Asch., ITS AJ633380; C. mollis (Jacq.) Asch., A. Zidorn,
Ch. Zidorn 99-00383 (priv.), Austria: Aschland, matK EU363538; C. smyrnaea DC., K. P. Buttler 23431 (M),
Turkey: Bursa, ITS EU363598/matk EU363546; Soyeria, C. blattarioides (L.) Vill., T. Eckhardt 1555 (B),
Switzerland: Valais, ITS EU363624/ matk EU363561; C. bocconi P.D. Sell, ITS AJ633375/ matk AJ633146;
Paleya, C. albida Vill. ssp. albida, O. Angerer M-0088570 (M), France: Dep. Lozère, ITS EU363606/matk
EU363550; C. albida Vill. ssp. grossi (Pau) Babc., E. Bayer & J. Grau B.G.79 (M), Spain: Andalucia, ITS
EU363594; C. albida Vill. ssp. scorzoneroides (Rouy) Babc., A. Segua 16.113 (M), Spain: Soria, ITS EU363595;
Anisoramphus, C. alpestris (Jacq.) Tausch, ITS AJ633373/ matk AJ633153; C. hypochaeridea (DC.) Thell., N.J.
Devenish 915 (B), S. Africa: Transvaal, ITS EU363617; Gephryoides, C. leontodontoides All., Vogt 10/1992 (B),
France: Corsica, ITS EU363592/ matk EU363542; C. tingitana Ball., Ch. Zidorn 970526a (priv.), Spain: Andalusia,
ITS EU363586; C. tingitana Ball., matk AJ633149; Berinia, C. auriculaefolia Sieber ex Spreng., R. Jahn s.n. (B),
Greece: Crete, ITS EU363626/ matk EU363563; C. baldaccii Halácsy, Franzén et al. 669 (B), Greece: Ioannina,
ITS EU363625; C. baldaccii Halácsy, K.H. Rechinger 21313 (B), Greece: Mt. Timphi, matk EU363562; C. biennis
L. ITS AJ633355; C. chondrilloides Jacq., J. Schimmitat M-0088551 (M), Jugoslavia: Biokovo, ITS EU363593; C.
chondrilloides Jacq., O. Angerer M-0088547 (M), Italy: Triest, matk EU363545; C. darvazica Krasch.; V.
Goloskokov s.n. (B), Kazakhstan: Alatau, ITS EU363600/ matk EU363558; C. guioliana Babc., W. Greuter 14533

30
 2. PHYLOGENY AND CHARACTER EVOLUTION

(B), Greece: Epirus, ITS EU363618; C. incana Sibth. et Sm., W. Greuter et al. 14821 (B), Greece: Parnassos,
matk EU363554; C. lacera Ten., C. Ricceri 8741 (B), Italy: Umbria, ITS EU363634; C. macropus Boiss. et Heldr.,
J. & F. Bornmüller 14397 (B), Turkey: Bursa, ITS EU363589; C. macropus Boiss. et Heldr., Davis & Coode
D37063 (E), Turkey: Bolu, matk EU363577; C. merxmuerlleri Kamari et Hartvig, Hartvig & Seberg 5059 (B),
Macedonia: Mt Smolikas, ITS EU363644; C. oporinoides Boiss. ex Froel., B. Valdés et al. It916/88 (B), Spain:
Sierra Nevada, ITS EU363633/matk EU363567; C. pannonica (Jacq.) K. Koch, E. Vitek 99-393 (B), Austria:
Vienna, matk EU363571; C. sibthorpiana Boiss. et Heldr., P.H. Davis 18140 (E), Crete: Mt Svowitchii, ITS
EU363648/matk EU363574; C. sonchifolia C.A. Mey., Nazarova (B), Armenia, ITS EU363637; C. taygetica Babc.,
W. Lippert 21366 (M), Greece: Peloponnes, ITS EU363603/matk EU363548; C. triasii (Camb.) Fries, H.
Merxmüller 148/57 (M), Spain: Mallorca, ITS EU363597; C. turcica Degen et Baldacci, matk AJ633360; C.
turcomanica H. Krasch., J.R. Edmondson 1177 (E), Iran: Khorasan, ITS EU363652; Macropodes, C. crocea
(Lam.) Babc., Sukaczev et Poplavskaja (B), Russia: Czita, ITS EU363590; C. hookeriana J. Ball., D. Podlech
47476 (MSB), Marocco: Atlas, ITS EU363605/matk EU363549; C. oreades Schrenk,Schevireva et Kojuvapova
619/1000 (B), Tajikistan: Pamir Alay, ITS EU363640/matk EU363572; Ixeridopsis, C. flexuosa (DC.) Benth. Et
Hook. F., Timokhina et Djukov M-0088593 (M), Russia: Rep. Tuva, ITS EU363596/matk EU363544; C. nana
Richards, Jonsell et Urbanska 6456 (UPS), USA: Alaska, ITS EU363591; C. nana Richards, O. Martensson U27
(UPS), USA: Alaska, matk EU363541; Intybellia, C. incarnata (Wulf.) Tausch, H.&H. Doppelbaur 14684 (M),
Italy: Südtirol, ITS EU363608; C. incarnata (Wulf.) Tausch, matk AJ633151; C. praemorsa (L.) Tausch , G. van
Buggenhout 13637 (B), Italy: Bolzano, ITS EU363654/matk EU363578; Mesophylion, C. bungei Ledeb., ITS
AJ633374/matk AJ633147; C. nigrescens Pohle, O. Rebristaja US-329530(US), Russia: Tiumen, ITS EU363609;
C. tectorum L., E. Willing 4.502 D (B), Germany: Berlin, ITS EU363643/matk EU363536; Psilochenia, C.
acuminata Nutt., L.S. Rose 55156 (B), USA: California, ITS EU363616; Lagoseris, C. frigida (Boiss.) Babc., P.
Hein 74 (B), Turkey: Bolkar Daglari, IST EU363612/matk EU363555; C. purpurea (Willd.) M. Bieb, Ivanov (B),
Russia: Tauria, ITS EU363653; C. sahendi Boiss. et Buhse, Davis & Polunin D24074 (E), Turkey: Hakkari, ITS
EU363651; Phaecasium, C. palaestina (Boiss.) Bornm., A. Danin et al. 53.019 (B), Israel: Galilee, ITS
EU363639; C. pterothecoides Boiss., A. Danin (B), Israel: Negev, matk EU363570; C. pulchra L., ITS
AJ633369/matk AJ633145; Hostia, C. alpina L., J. Trelawny 1423 (E), Turkey: Hakkari, ITS EU363649/matk
EU363575; C. foetida L. ssp. foetida, J. Lambinon 00/F/214bis (B), France: Ardéche, ITS EU363619/matk
EU363556; C. foetida L. ssp. afghanistanica Babc., D. Podlech 18361 (MSB), Afghanistan: Baghlan, ITS
EU363604; C. foetida L. ssp. rhoeadifolia (Bieb.) Celak, J. Lambinon 00/214 (B), France: Ardéche, ITS
EU363613; C. kotschyana Boiss., K.H. Rechinger 51631 (B), Persia: Khorasan, ITS EU363635; C. rubra L., ITS
AJ633350/matk AJ633141; C. thomsonii Babc., M. Jabcobs 6374 (E), Iran: Lorestan-Sheshom, ITS EU363647;
C. tybakiensis Vierh., N. Böhling 5452b (B), Greece: Lasithiou, matk EU363566; C. tybakiensis Vierh., Greuter et
Matthäs 19649 (B), Greece: Crete, ITS EU363631; Microcephala, C. multicaulis Ledeb. ssp. congesta (Regel)
Babc., Nüsser 422 (B), Pakistan: Nanga Parbat, ITS EU363642/matk EU363573; Pterotheca, C. sancta (L.)
Babc. ssp. nemauensis (Gouan) Babc., J. Lambinon 84/F/373 (B), France: dep. Gard, ITS EU363632; C. sancta
(L.) Babc., matk AJ633150; Zacintha, C. pusilla (Sommier) Merxm., Th. Raus 7715 (E), Greece: Kasos, ITS
EU363650/matk EU363576; C. zacintha (L.) Babc., Ralf Hand 5323 (priv.), Greece: Rhodes, ITS EU363655/matk
EU363579; Phytodesia, C. capillaris (L.) Wallr., ITS AJ633381/matk AJ633142; C. corymbosa Ten., P. Hiepko

31
2. PHYLOGENY AND CHARACTER EVOLUTION

065 (B), Italy: Apulia, matk EU363559; C. cretica Boiss., R. Jahn 4.4.1996/02 (B), Greece: Selinou, ITS
EU363614; C. fuliginosa Sibth.& Sm., D. Philos 4032 (B), Greece: Euboea, ITS EU363601/matk EU363547; C.
nicaeënsis Balb., v.Ooststroom& Hennipan 23519 (B), Jugoslavia: Macedonia, ITS EU363641; C. neglecta L. ssp.
neglecta, Eisenblätter & Willing 82.085 (B), Greece: Kardhitsa, ITS EU363610/matk EU363553; C. neglecta L.
ssp. corymbosa (Ten.) Nyman, Eisenblätter & Willing 81.622 (B), Greece: Trikala, ITS EU363611; C. parviflora
Desf., R.D. Meikle M-0088516 (M), Cyprus: Yerosa, ITS EU363607/matk EU363551; Lepidoseris, C. vesicaria L.
ssp. vesicaria, R.&E. Willing 88.781 (B), Greece: Ilias, matk EU363565; C. vesicaria L. ssp. stellata (Ball) Babc.,
R. Vogt 11757 (B), Marocco: Anti-Atlas, ITS EU363630; C. vesicaria L. ssp. haensleri (Boiss.ex DC.) P.D. Sell,
ITS AJ633371; Nemauchenes, C. aspera L., R. Hand 3708 (B), Cyprus: Polystipos, ITS EU363628; C. setosa
Hall.f., N. Enke 0123a (B), Italy: Pisa, ITS EU363585; Psammoseris, C. bellidifolia Loisel., J. Lambinon
84/Co/433 (B), France: Corsica, ITS EU363615; C. bursifolia L., Molina & Gavilán 15645 (B), Spain: Madrid, ITS
EU363623; Species without sectional assignment, C. chupantha, T. Elias, Shetler & Murray US 498520 (US),
Russia: Republ. Tuva, matk EU363552; C. cytherea Kamari, R. Jahn (B), Greece: Kythira, ITS EU363646;
OTHER GENERA: Cephalorynchus tuberosus (Steven) Schchian, Cho 10189 (B), matk EU363582*; Chondrilla
canescens Kar. & Kir., matk AJ633349; Garhadiolus hedypnois Jaub. et Spach, ITS AJ633307; Garhadiolus
hedypnois Jaub. et Spach, Fayed & El Garf DB0305 (B), Egypt: Alexandria, matk EU363540; Heteracia szovitzii
Fischer & C.A. Meyer, ITS AJ633283; Hololeion maximoviczii Kitam., ITS AJ633425; Hypochoeris maculata L.,
ITS AJ633311; Hyoseris radicata L., ITS AJ633299; Ixeridium laevigatum (Blume) Pak & S. Kawano,
Bartholomew & Boufford 6192 (US), Taiwan: Chiayi Hysien, ITS EU363588; Ixeris chinensis (Thunb.) Nakai,
Bartholomew, Boufford, Li 749 (US), ITS EU363587/matk EU363539; Ixeris stolonifera A. Gray, ITS
AJ633284/matk AJ633156; Lactuca dissecta D. Don., Kilian 2547 (B), Tadshikistan: Warob, matk EU363580*;
Lapsana communis L., ITS AJ633285/matk AJ633138; Phitosia crocifolia (Boiss. et Heldr.) Kamari & Greuter,
Strid et al. 15261(B), Greece: Messinias, ITS EU363621/matk EU363557; Picris echioides L. ITS AJ633321,
Prenanthes purpurea L., ITS AJ633342; Rhagadiolus edulis Gaertner, ITS AJ633297; Rhagadiolus stellatus (L.)
Gaertner, ITS AJ633296/matk AJ633224; Scorzonera troodea Boiss, R. Hand 3271 (priv.), Greece: Cyprus, matk
EU363584*; Sonchus oleraceus L., ITS AY862581/matk DQ840449; Soroseris glomerata (Decne) Stebbins, T.N.
Ho 1692 (CAS), China: Quinghai, ITS EU363656*; Steptoramphus czerepanovii Kirp., Cho 10155 (B), matk
EU363581*; Taraxacum bessarabicum Fisch., ITS ZJ633287; Taraxacum erythrospermum Andrz. ex Bess., ITS
AJ633291; Taraxacum laevigatum DC, ITS AJ633288; Youngia denticulata,ITS AJ633293; Youngia japonica (L.)
DC, ITS AJ633294; Youngia tenuifolia (Willd.) Babc. & Stebbins, Beljaeva et al. (B), Russia: Primorje, ITS
EU363645/matk EU363583.

All sequences bearing numbers starting with EU produced in the present study. * Samples
provided by N. Kilian.

32
 3. GENOME SIZE EVOLUTION

Chapter 3

Shrinking Genomes? Evidence from Genome Size Variation in

Crepis L. (Compositae)

N. Enke1,3, J. Fuchs2 & B. Gemeinholzer1

1
Botanic Garden and Botanical Museum Berlin-Dahlem, Freie Universiät Berlin, Königin-Luise-Str. 6-8,
14195 Berlin, Germany. 2 IPK-Gatersleben , Correnstr. 3, 06466 Gatersleben, Germany. 3 Author for
Correspondence: n.enke@bgbm.org.

Submitted Molecular Biology and Evolution, 2008

ABSTRACT

Large scale surveys of genome size evolution in angiosperms showed that the ancestral
genome most likely was small with a tendency towards an increase in DNA content during
evolution. Due to polyploidisation and self replicating DNA elements, angiosperm genomes
were considered to have a “one way ticket to obesity” (Bennetzen & Kellogg, 1997). New
findings on how organisms can lose DNA challenged the hypotheses of unidirectional
evolution of genome size. The present study shows that within 30 diploid species of the
genus Crepis there is a striking trend towards genome contraction. Genome size of 21
species has been estimated by flow cytometry. Findings were combined with additional data
from literature. The directionality of genome size evolution was analysed by reconstructing
ancestral character states on a molecular phylogeny based on ITS sequence data. DNA
content is shown to be correlated to distributional aspects as well as life form. Genome size
is significantly higher in perennials than in annuals. Within sampled species very small
genomes are only present in Mediterranean or European species, whereas their Central and
East Asian relatives show small or intermediate 1C-values. The only cases of well supported
1C-value increase within the genus are due to polyploidisation.

KEYWORDS: ancestral character state reconstruction, Crepis, Compositae, flow cytometry,

genome evolution, ITS.

33
3. GENOME SIZE EVOLUTION

3.1 INTRODUCTION

Two of the most obvious and probably the most discussed attributes of karyotypes are
chromosome number and size. Total chromosome size is directly linked to DNA content
(given as C-value) while chromosome number is not (Levin, 2002). Both characters vary
enormously throughout angiosperms. Chromosome number ranges from n=2 to n= ca.300
(Grant, 1982; Masterson, 1994). In diploid Actinidia species the basic chromosome number
is as high as x=29 (Yan et al., 1997). Genome size varies nearly 2000fold within the
angiosperms (Greilhuber et al., 2006); the smallest genome is not found in Arabidopsis as
previously assumed but in the carnivorous genus Genlisea (Greilhuber et al., 2006). As
within the genome the genetic constituency of each organism is encrypted and because of
the heterogeneity of these traits there is a lot of interest in understanding chromosome
evolution, its direction as well as its underlying mechanisms (e.g. Leitch et al., 1998; Soltis et
al., 2003; Bennett & Leitch, 2005).

For the sake of unambiguousness and comparability in the discussions of genome size
variation it is inevitable to distinguish between genome size and 1C-value. According to
Bennett et al. (1998) genome size is the 2C value divided by ploidy level, so the genome size
corresponds to the 1C-value in diploid but can be lower than the 1C-value in polyploids. Or,
as Greilhuber et al. (2005) put it, 1C-value and 1Cx-value should be discriminated: 1Cx-value
is the “monoploid” genome size (Greilhuber et al., 2005) whereas 1C is the unreplicated
reduced chromosome complement (Bennett & Smith, 1976). In this study the 1C-value is
used.

The “C-value paradox” (Thomas, 1971) refers to the phenomenon that the amount of DNA is
not reflected by the complexity of an organism. More complex organisms do not necessarily
feature more DNA, in fact the basic set of genetic information required for normal
development is similar in most plants (Flavell, 1980). Thus, most variation in genome size is
due to non-coding or repetitive DNA elements (Flavell, 1986; Kubis et al., 1998; Schmidt &
Haslop-Harrison, 1998). The “C-value enigma” (Gregory, 2001) addresses the question of
why there is such a considerable variation in non-coding DNA, how it is distributed among
taxa and how it developed. Within a species genome size is almost always constant,
whereas it can vary strongly between species (Greilhuber, 1998, 2005). As a consequence,
the use of genome size to infer phylogenetic relations is questionable, but it still can be
useful for species delimitation (Murray, 2005).

The remarkable variation in genome size throughout the angiosperms as well as within more
confined taxa has many causes. In the context of size variation the correlation between
genome size and life history often has been discussed (e.g. Bennett, 1972; Sims & Price,
1985; Bennett & Leitch, 1995; Watanabe et al., 1999). Following the nucleotype theory
(Bennett, 1972) annual plants mostly have small genomes while perennials can have larger

34
 3. GENOME SIZE EVOLUTION

genomes. In some cases it appears that genome size is also correlated to various
environmental and physiological factors (Knight & Ackerley, 2002; Jakob et al., 2004; Ohri,
2005).

In the basic angiosperms the overall direction of genome size evolution is unambiguous: it
increases (Soltis et al., 2003). It therefore complies with the hypothesis of Bennetzen &
Kellog (1999) who postulated “a one way ticket to genomic obesity”: Due to the accumulation
of retroelements genome size generally increases. Repeated circles of polyploidy also lead
to an increase in genome size (Wendel, 2000). But within plant families or genera decreases
as well as increases have been observed (e.g. Watanabe et al., 1999; Wendel et al., 2002;
Jakob et al., 2004). Various explanations have been proposed about how DNA is lost (e.g.
Vicient et. al., 1999; Kirik et al., 2000; Hancock, 2002; Petrov, 2002; Zuckerkandl, 2002).
Differences in DNA content could be caused by the loss of whole chromosomes or parts of it
(Dart et al., 2004). Vicient et al. (1999) showed that in barley the removal of BARE-1
elements through intrachromosomal homologous recombination between LTR’s (long
terminal repeats) can be responsible for a reduction of DNA content. Similar mechanisms
have been found in rice (Ma et al., 2004).

So far studies on genome size variation were mostly either concerned about infraspecific
variation (e.g. Greilhuber & Ebert, 1994; Turpeinen et al., 1999; Temsch & Greilhuber, 2001)
or variation within families or even higher taxa (e.g. Kellogg, 1998; Wendel et al., 2002; Soltis
et al., 2003; Greilhuber et al., 2006). Only few studies investigate genome size variation
between species belonging to clearly confined groups (genera or subgeneric ranks) to
discuss its role in the evolution of the genera, which is presented here for the Compositae
genus Crepis.

The genus Crepis comprises over 200 species in the holarctic region and Africa. Over 80%
of the species are diploid. Polyploidy is virtually constrained to one section being distributed
in North America. The basic chromosome number varies from x = 3--6 respectively x = 11 in
the 15 polyploid North American species. These rare features make it ideal to explore
changes of genome size and basic chromosome number that are not connected to changes
of ploidy level.

Even before Babcock published his extensive work on karyotype evolution in the genus
(Babcock, 1947a,b), Crepis has been a popular object of cytological studies (e.g.
Hollingshead, 1930; Tobgy, 1943; Sherman, 1946). Babcock scrutinized the karyotypes of
over one hundred Crepis species and summed up the results in several hypotheses: First,
chromosome number decreases during evolution, so that species with small chromosome
numbers are derived. Second, short living annuals have undergone a reduction in the
quantity of nuclear DNA. Furthermore, he stated chromosomal rearrangements to be the
driving force of evolution in the genus. Since Babcock’s hypotheses, new methods in
karyotype studies (e.g. genome size estimation) brought additional insights into the karyology

35
 3. GENOME SIZE EVOLUTION

of Crepis. (e.g. Siljak-Yakovlev & Cartier, 1982; Kamari, 1992; Dimitrova & Greilhuber, 2000).
With the establishment of a molecular phylogeny (Enke & Gemeinholzer, 2008) the findings
of Babcock and his successors can be further interpreted in an evolutionary context of
phylogenetic relations.
Table 1: 1C value, used standards and tested characters of the presently measured Crepis species. Ind =
individuals, S = samples. Distr = distribution, A = North Africa, Middle East, Mediterranean or Southeast Europe, B =
Europe, C = Eurasia, D = Central and East Asia. Alt = altitude, alp = alpine, mont = montane, low = lowland. end =
endemic. p = perennial, a= annual.

Basic
1C Number chrom.
value SD of number Ploidy Life
Species (pg) (+/-) Ind S Standard (1x) level Section Clade Distr Alt Range History
Crepis aurea 1.75 0.075 5 6 G. max 5 D Brachypodes VIII A alp wide p
C. biennis 9.51 0.492 6 8 P. sativum 5 O Berinia VII B - - -
C. blattarioides 3.58 0.051 3 5 S. cereale 4 D Soyeria XI B mont wide p
C. bungei 4.05 0.053 5 5 S. cereale 4 D Mesophylion VII D low wide p
C. chrysantha 4.64 0.053 5 5 S. cereale 4 D Brachypodes VII D alp wide p
C. conyzaefolia 5.90 0,069 7 10 P. sativum 4 D Soyeria - B mont wide p
C. crocea 9.61 0.135 8 10 P. sativum 4 T Macropodes II D - - p
C. foetida 1.46 0.034 3 4 G. max 5 D Hostia III A low wide a
C. kerneri 5.56 0.046 3 4 S. cereale 6 D Brachypodes II A alp wide p
C.
1.14 0.012 5 5 R. sativus 5 D Gephyroides VIII A lowl end p
leontodontoides
C. lyrata 2.79 0.037 8 7 P. sativum 6 D Mesomeris - D mont wide p
C. multicaulis 1.78 0.018 5 5 G. max 5 D Microcephalum Lagoseris D mont wide p
C. paludosa 4.53 0.074 4 10 S. cereale 6 D Desiphylion II B mont wide p
C. polytricha 9.62 0.300 6 8 P. sativum 4 T Brachypodes - D - - p
C. pusilla 1.11 0.025 3 4 R. sativus 5 D Zacintha III A mont wide a
C. sibirica 6.90 0.086 4 5 P. sativum 5 D Desiphylion - C mont wide p
C. turcica 6.41 0.100 3 6 P. sativum 4 D Berinia - A mont end p
C. vesicaria 4.18 0.050 5 5 S. cereale 4 D Lepidoseris X B mont wide -
Askellia
6.78 0.072 3 5 P. sativum 7 D Ixeridopsis - D - - p
karelinii
Youngia
2.08 0.066 4 5 G. max 5 D - Youngia D - - p
tenuicaulis
Youngia
4.10 0.060 5 5 S. cereale 5 T - Youngia D - - p
tenuifolia

The aim of the present study is to investigate variation of genome size in the genus Crepis
and its relations to speciation and phylogeny. Therefore, it discusses genome size variation
in a molecular phylogenetic framework. Estimations of the ancestral character state for the
1C-value were performed based on a phylogenetic tree inferred from the nuclear marker ITS.
The chloroplast marker matK has also been considered for reconstruction but been
discarded for two reasons: First, the tree inferred by Enke & Gemeinholzer (2008) shows
very low resolution and second, sequence availability is very low. It would, however, be of
interest for future analyses to use a better resolving chloroplast marker to detect possible
reticulate evolution within the genus, also in correlation to genome size evolution.

36
 3. GENOME SIZE EVOLUTION

3.2 MATERIAL AND METHODS

Some of the 1C-values used in the statistical analysis and the ancestral character state
reconstruction have been taken from literature (table 2). These have been inferred by
different methods; so the values should not be directly compared. Therefore the 1C-values
were assigned to different character classes following Leitch et al. (1998) and Soltis et al.
(2003), defining 1C-values ≤1.4pg as “very small”, >1.4pg but ≤ 3.5pg as “small” and 1C-
values >3.5pg as “intermediate”. In cases where the 1C-Values of individual species differed
between studies, these differences, however, never influenced the character class the
species were assigned to. Whenever 1C-values are directly compared, they are always from
within one study if not stated differently.

Two datasets were used throughout the study: data set (1) comprising all species measured
for DNA content in the present study and data set (2) comprising all Crepis species with
known DNA content. The composition of the data sets can vary slightly between the
ancestral character state reconstruction and statistical analysis due to the availability of
sequence information for a given species.

TABLE 2: 1C value and tested characters of Crepis species from published data. Sources of published
data are given in table. Ind = individuals, S = samples. Distr = distribution, A = North Africa, Middle
East, Mediterranean or Southeast Europe, B = Europe, C = Eurasia, D = Central and East Asia. Alt =
altitude, alp = alpine, mont = montane, low = lowland. end = endemic. p = perennial, a= annual.

Basic
1C chrom.
value number Ploidy Life
Species (pg) (1x) level Section Clade Distr Alt Range History
1
Crepis alpina 3.0 5 D Hostia III A mont wide a
5
C. bithynica 2.8 5 D Macropodes - A mont wide p
1
C. fuliginosa 0.9 3 D Phytodesia VI A mont end a
2
C. incarnata 6.0 4 D Intybellia Lagoseris A mont wide p
3
C. lapsanoides 5.6 6 D Mesomeris I B mont end p
1
C. neglecta 1.8 4 D Phytodesia VI A mont wide a
3
C. palaestina 6.1 4 D Phaecasium Lagoseris A low end a
5
C. paludosa 4.5 6 D Desiphylion II B mont wide p
3
C. praemorsa 5.3 4 D Intybellia Lagoseris C mont wide p
3
C. pontana 6.9 5 D Soyeria - B mont end p
5
C. pulchra 5.5 4 D Phaecasium Lagoseris A mont wide a
3
C. rubra 2.9 5 D Hostia III A mont wide a
5
C. sancta 2.2 5 D Pterotheca Lagoseris A mont wide a
5
C. schachtii 2.8 5 D Macropodes - A mont end p
5
C. setosa 1.7 4 D Nemauchenes - B low wide a
3
C. tectorum 3.4 4 D Mesophylion II C low wide a
5
C. viscidula 4.9 6 D Desiphylion II A alp end p
5
C. zacintha 1.1 3 D Zacintha III A low wide a
1
Wallace, 1972; 2Marie & Brown, 1993; 3Bennett & Smith 1976; 4 Dimitrova et al., 1999; 5Dimitrova & Greilhuber, 2000.

37
3. GENOME SIZE EVOLUTION

Genome Size Estimation – Seed from 22 accessions of 21 Crepis species and two species
of closely related Youngia (table 1) were germinated and cultivated in pots (vouchers in B).
Species from several sections (Babcock, 1947a,b) as well as different clades derived from
molecular data (Enke & Gemeinholzer, 2008) were chosen to represent a variety of
taxonomic and molecular groups. Material was taken from leaves of seedlings except for C.
aurea, C. pannonica and C. turcica where leaf material of adult plants grown the previous
year was used. For the internal standards leaves of adult plants potted and grown in a
greenhouse were used (Glycine max (L.) Merr. convar. max var. max ‘Cina 5202’ (Genbank
Gatersleben, accession number: SOJA 392 (2.23 pg/2C)), Pisum sativum L. subsp. sativum
convar. sativum var. ponderosum Alef. ‘Viktoria, Kifejtö Borsó’ (Genbank Gatersleben,
accession number: PIS 630 (9.07 pg/2C)), Secale cereale subsp. cereale (Genbank
Gatersleben, accession number: R 737 (16.01 pg/2C)) and Raphanus sativus L. convar.
sativus Small radish group ‘Voran’ (Genbank Gatersleben, accession number: RA 34 (1.11
pg/2C)).

Leaf fragments of the sample TABLE 3: Correlation of genome size and karyological, systematical
and distributional factors. Correlation coefficient (rs) and significance
plant and the respective standard from Spearman rank-order correlation.
plant (see table 1) were chopped
data set (1) data set (2)
in 1ml of modified WPB (Loureiro rs p rs p
et al., 2007; 0.2 M Tris HCl, 4 Chromosome number -0,253 0,363 -0,005 0,975
Section -0,447 0,095 -0,310 0,074
mM MgCl2·6H2O, 2 mM EDTA
Clade -0,243 0,471 -0,007 0,972
Na2·2H2O, 86 mM NaCl, 10 mM Life History 0,557* 0,039 0,519** 0,001
potassium metabisulfite, 1 % Distribtuion 0,229 0,411 0,331 *
0,046
PVP-30, 1 % (v/v) Triton X-100,
pH 7.5) supplemented with 50µg/ml propidium iodide and 50µg/ml DNAse-free RNAse,
filtered through a 35µM mesh and stored on ice until measurement. 4 to 10 samples of 3-8
individuals per taxon (see table 1) were measured on two consecutive days using a
FACStarPLUS flow sorter (BD Biosciences) equipped with an argon ion laser INNOVA 90C
(Coherent). Usually, 10,000 nuclei per sample were analysed.

Ancestral Character State Reconstruction – The ancestral character state reconstruction

was carried out for both datasets on most parsimonious trees inferred from ITS sequence
data. Sequences from Enke & Gemeinholzer (2008) were analysed by addition of new
accessions (see appendix). Phylogenetic histories were reconstructed using Maximum
Parsimony and Bayesian inference. Maximum Parsimony analyses were conducted in Paup
4.10b* (Swofford, 2002) with equal weights, 1000 closest sequence additions and tree
bisection-reconnection (TBR) branch swapping, permitting 10 trees to be held at each step.
Maxtrees was set to unlimited. An evaluation of the trees was performed using bootstrap
analysis with 10000 replicates, equal weights, TBR swapping, MulTrees option in effect and

38
 3. GENOME SIZE EVOLUTION

10 trees held at each step. Bayesian analyses were conducted using MrBayes (Ronquist &
Huelsenbeck, 2003) assuming gamma distribution rate variation among sites and applying
10 million generations of the MCMC chains in two independent runs, trees saved every 100
generations. The first 27 000 trees were discarded as burn-in for the analysis then reached
stationarity. All other trees sampled were used to calculate a strict consensus tree. Trees
were rooted with Youngia as outgroup (Enke & Gemeinholzer, 2008).

The character history was traced independently for data sets (1) and (2) on rooted trees with
the Ancestral Character State Reconstruction package of Mesquite 2.5 (Maddison &
Maddison, 2008). When using character classes, character states were treated as unordered
categorical and a most parsimonious approach for character state reconstruction was used
(Figs.1,3). Continuous character states were assumed for data set (1) when using 1C-values.
In this case the character history was reconstructed with a squared parsimony model (Fig.2).

Statistical Analysis – Descriptive statistics were conducted with SPSS 16.0 including all
diploid species. The correlation between genome size and chromosome number, life form,
distribution (divided into geographic region, altitudinal rank and distributional range) and
phylogenetic relations (clade, section) was tested on both data sets. For data set (1) direct
1C-values were used, for data set (2) character classes. Tables 1-2 show all species with
1C-values and the tested attributes for each species. Sources for 1C-values taken from
literature are given in the table. Chromosome numbers were taken from Babcock (1947a,b)
and Kamari (1992). Information on life form was taken also from Babcock (1947a,b) as well
as distributional and habitat information. Species were assigned to be distributed in four
geographic areas (following the categories used by Kilian et al. (2008)): Group A is
distributed in North Africa, Middle East, Mediterranean or Southeast Europe. Group B
comprises species from Europe, group C from Eurasia, and group D from Central and East
Asia. The species were assigned to one of three altitudinal ranks (lowland, mountainous,
alpine) and one of two distributional patterns: widespread or endemic. Sections used in
tables and figures refer to the infrageneric taxonomic classification sensu Babcock (1947b)
and clades to the molecular groupings found by Enke & Gemeinholzer (2008).

Correlation tests were carried out using Spearman’s rank order correlation as the data were
mainly categorical and/or showed no normal distribution.

To test the significance of difference in 1C-value between annuals and perennials (both data
sets) a Mann-Whitney-U test was applied, appropriate to test not normally distributed data.

39
3. GENOME SIZE EVOLUTION

FIG.1: Ancestral character state reconstruction for data set (1) using character classes on 1 of 2 most
parsimonious trees. Bayesian posterior probabilities (PP) > 90 and bootstrap values (BS) > 75 are given
above branches. Arrows indicate decreases in genome size. A and B denote nodes discussed in the text.
Clades named following Enke & Gemeinholzer (2008).

3.3 RESULTS
Ancestral character state reconstruction – For data set (1) two most parsimonious trees,
for data set (2) three most parsimonious trees were inferred. Trees inferred by Maximum
Parsimony and Bayesian analyses showed congruent topologies. One of the most
parsimonious trees of each data set was used for the ancestral character state
reconstruction. The trees are shown in Figs.1-3. Posterior probabilities and bootstrap values
above 75 are given above branches. The clades are named following Enke & Gemeinholzer
(2008). Species relationships are similar to those found by Enke & Gemeinholzer (2008),

40
 3. GENOME SIZE EVOLUTION

FIG.2: Ancestral character state reconstruction for data set (1) using 1C-value on 1 of 2 most parsimonious
trees. Bayesian posterior probabilities (PP) > 90 and bootstrap values (BS) > 75 are given above branches.
Arrows indicate decreases, squares increases in genome size. Clades named following Enke &
Gemeinholzer (2008).

differences (mainly the position of C. setosa) are due to the considerably smaller sample size
in the present study.

The ancestral character state reconstruction for data set (1) (only data obtained by the
presented study and using character classes) is ambiguous (Fig.1). From node A to B the
ancestral character state could be either intermediate or small. In only two cases (Fig. 1,
arrows) unambiguous character state changes can be observed. In both cases a decrease
from small to very small can be observed.

If 1C-values are used for ancestral state reconstruction on data set (1) the ancestral 1C-
value is estimated between 2.8 and 3.7pg (Fig.2). Only in clades II and VII the overall trend
towards a decline in 1C-value is reversed (Fig.2, squares). In clade VII the increases are due
to the high 1C-values of the two tetraploid species C. crocea, C. polytricha and octoploid C.
biennis.

In data set (2) representing character classes and all genome sizes analyzed here or
elsewhere published (Fig.3) the basic character state for Crepis s.str. and the Lagoseris
group is an intermediate 1C-value equivalent to the closest relatives to Crepis. Decrease of

41
3. GENOME SIZE EVOLUTION

genome size to either a small or very small 1C-value occurred seven times during evolution
of the species within the genus (Fig.1, arrows). In three cases (Fig.1, stars) uncertain
character states (small/very small) are found. If ambiguous states were treated as small the
number of decreases raises to ten. Increases can only be found if uncertain states are
treated as very small (3 times, Fig.1, squares).

The decreases found by the analysis of data set (1) using character classes (Fig.1, arrows)
reoccur in the analysis of data set (2) (Fig.3, arrows, clades III, VIII). In the analysis of data
set (1) using 1C-values these decreases are located at a deeper node (Fig.2, arrows) due to
the fact that C. leontodontoides (1.14pg/1C) and C. aurea (1.75pg/1C) respectively C. pusilla
(1.11pg/1C) and C. foetida (1.46g/1C) fall into different character classes (Fig.1, very small,
small) but into the same categorical group (Fig.2, 1.12-1.97pg).

Fig.3 (arrows, Lagoseris group, Clade I) shows that the small genome sizes of C. multicaulis
and C. lyrata are derived. Consequently, the ancestral character states small/2.8-3.7pg for
data set (1) (Figs.1-2) might be underestimated in both analyses, as C. multicaulis (Lagoseris
group) as well as C. lyrata (Clade I) have considerably lower 1C-values than their relatives
included in data set (2).

Genome Size Variation/Statistical Analysis – The 1C-values within the screened Crepis-
species show a 8.5fold difference between the highest (C. crocea 9.62 pg/1C) and lowest
value (table 1). This difference, however, includes the tetraploid C. crocea. The difference
between highest (C. sibirica 6.90pg/1C) and lowest diploid (C. pusilla 1.11 pg/1C) is a 6fold
difference.

The correlation between the genome size and other factors (chromosome number, life form,
distribution, and phylogenetic relations (clade, section)) is shown in table 3. In data set (1) life
history and 1C-value show a significant positive correlation implying that perennial plants
tend to have larger genomes than annuals. In data set (2) both life history and distribution
are significantly positive correlated to genome size class. Fig.4 visualises the positive
correlation between genome size classes and distribution. Very small genomes are only
present in plants from the circum Mediterranean region and Europe, even though plants from
these two regions are present in all three genome classes. The three analyzed species with
widespread Eurasian distributions feature only intermediate genomes. Analyzed species
from Central Asia have either small or intermediate genomes.

Genome sizes within annuals (data set (1) mean 1C = 1.3pg) are significantly lower than in
perennials (data set (1) mean 1C = 4.1pg) (Mann-Whitney-U, data set (1) P < 0.05; data set
(2) P < 0.005).

42
 3. GENOME SIZE EVOLUTION

FIG.3: Ancestral character state reconstruction for data set (2) using character classes on 1 of 3 most
parsimonious trees. Bayesian posterior probabilities (PP) > 90 and bootstrap values (BS) > 75 are given
above branches. Arrows indicate decreases, squares increases in genome size. Stars mark ambiguous
branches. Clades named following Enke & Gemeinholzer (2008).

3.4 DISCUSSION
Ancestral Character State Reconstruction – The results of the different analyses based on
character classes and direct 1C-value are comparable and lead to the same conclusion that
1C-value in diploid Crepis species decreases. However, the ancestral character state
analysis using 1C-values is more sensitive (viz. the only analysis detecting the increases

43
3. GENOME SIZE EVOLUTION

caused by polyploidisation, Fig.2, squares), but relies on the assumption that 1C-values
inferred by different methods are directly comparable.

The character state reconstructions within the evolution of Crepis (Figs.1-3) show an overall
trend towards decrease in genome size. In the present study, the only reliable increases in
1C-value within Crepis are due to polyploidisation (Fig.2, clade VII, squares). The rise of 1C-
value in clade II (Fig.2, C. kerneri) in the analysis of data set (1) using 1C-value could be a
result of the possible underestimation of ancestral genome size in data (1) and therefore has
to be treated with caution.

The observed decrease in genome size confirms findings of earlier studies in Crepis (Jones
& Brown, 1976; Dimitrova & Greilhuber, 2001). Only few studies so far used molecular
phylogenetic data and direct genome size estimations such as flow cytometry or Feulgen
densitometry to infer directionality of genome size evolution on a generic level. Jakob et al.
(2004) analysed genome size evolution on a molecular phylogenetic background within
Hordeum and discovered increases as well as decreases among diploid species. Wendel et
al. (2002) also observed a “bidirectional” genome size evolution in diploid Gossypium
species. Based on total chromosome length which, according to e.g. Raina & Rees (1983) or
Jones & Brown (1976), is positively correlated to genome size, Watanabe et al. (1999) found
decreases as well as increases in the evolutionary history of Brachyscome (Asteraceae).
Regarding this context, the here presented trend in Crepis towards a decrease in genome
size is remarkably striking. However, only about a fifth of all known Crepis species could be
included in this study and the addition of further species might somewhat change the
observed trend.

Mechanistically not much is known about how DNA content decreases in Crepis. Tobgy
(1943) reported unequal reciprocal translocations to be the cause of a reduction from x=4 in
C. neglecta to x = 3 in C. fuliginosa. The decrease in chromosome number was accompanied
by a decrease in total chromosome length (Tobgy, 1943) and therefore a reduction in nuclear
DNA content. Kamari (1976) found C. fuliginosa more closely related to C. cretica than to C.
neglecta. C. fuliginosa also features x = 4 but its total chromosome length is not known.
Furthermore, a similar decrease in chromosome number from x = 5 in C. alpina to x = 4 in C.
kotschyana did not lead to a change in total chromosome length (Sherman, 1947). Even
though losses of whole chromosomes can lead to a decrease of genome size (Dart et al.,
2004) it can only be hypothesised at the present stage if it plays a role within Crepis.

As increases in 1C value within Crepis are mainly due to polyploidisation, the highest amount
of DNA of all species is found in two tetraploid species (C. crocea, C. polytricha) and the
octoploid C. biennis. One might expect the 1C-value of the tetraploid to be the sum of the
1C-values of its diploid ancestors, respectively to vary proportionally with the ploidy level, but
this is rarely the case. Frequently genome downsizing after polyploidisation has been
observed (Leitch & Bennett, 2004). C. crocea (9.61pg/1C), however, has nearly 15% more

44
 3. GENOME SIZE EVOLUTION

DNA than twice the amount of its close relative C. bungei (4.05pg/1C). C. crocea (2n = 4x =
16) is thought to result from hybridisation between C. bungei (2n = 2x = 8) and C. oreades
(2n = 2x = 8) (Babcock, 1947b). Unfortunately C. oreades was not available for genome size
estimation. For C. polytricha (9.62pg/1C) the amount of DNA is slightly (5%) higher than
twice the amount of its presumed closest relative C. chrysantha (4.64pg/1C). C. polytricha
(2n = 4x = 16) might have originated as an amphidiploid hybrid between C. chrysantha (2n =
2x = 8) and some other species (Babcock, 1947b). The relation of 1C-value of polyploid
species to their diploid relatives is similar in Magnolia (Soltis et al., 2003); however, the lack
of data on one of the putative parents, in both C. crocea and C. polytricha, should be taken
into account.

The observable contraction in genome size within Crepis contradicts the “one way ticket to
obesity”-hypothesis of Bennetzen & Kellogg (1997) and the general trends within the
angiosperms (Soltis et al., 2003). Increase of genome size within Crepis seems to be caused
largely by polyploidisation. The genus Crepis clearly demonstrates that the DNA amount of a
species reflects a dynamic balance between expansion and contraction, in terms of both
mechanisms and selective consequences (Bennett & Leitch, 2005).

Genome Size Variation and

Correlation – The 6fold variation of
genome size found in the diploid
species measured in this study is
within the 7fold differences found by
Jones & Brown (1976) and the
5.15fold variation found by
Dimitrova & Greilhuber (2000) for
diploid Crepis species. The variation
is similar to those found in other
genera of the Asteraceae. In
Artemisia the genome size varies
7fold (Torrell & Vallès, 2001). In the
Echinops group there is a 13fold
variation (Garnatje et al., 2004). So
FIG.4: Geographic distribution of species frequency sorted for far only little is known about why DNA
genome size class. content varies greatly between even
closely related species; e.g. C. lyrata
(2.8pg/1C) and C .lapsanoides (5.6pg/1C, Bennett & Smith, 1976). In genera of other
families genome size can vary more widely (e.g. 24fold in Genlisea, Lentibulariacea
(Greilhuber et al., 2006).

45
3. GENOME SIZE EVOLUTION

As already reported by Babcock (1947a), a positive correlation between genome size and life
form in Crepis is indicated by present results. Annual species have a significantly lower
genome size than perennials. But still variation within perennials is high: Both C.
leontodontoides (1.14pg/1C) with one of the lowest and C. sibirica (6.90pg/1C) with the
highest 1C-value are perennials. Even though this correlation is common throughout
angiosperms (e.g. Bennett et al., 1998; Bennett & Leitch, 2005), some authors give an
alternative explanation for the often observed correlation between life history and genome
size: selfing species tend to have smaller genomes than outcrossing ones, and because
selfing species are often annuals, a correlation between annuality and small genomes is
apparent (Govindaraju & Cullis, 1991; Albach & Greilhuber, 2004). Not many Crepis species
have been tested for self compatibility, and so far no obligate selfer is known. Known self
compatible species are C. tectorum, C. pulchra, C. multicaulis, and C. alpina (Babcock,
1947a). Except for C. pulchra with an intermediate 1C-value, all of these species have small
1C-values. Except for C. multicaulis all of the species are also annuals. Only two species are
known with no or low self compatibility, C. foetida (Babcock, 1947a; Hughes & Babcock,
1950) and C. sancta (Cheptou et al., 2000). Both have small 1C-values and are annual. Due
to the small sample size of the species included in this study, it is hard to estimate how
representative these species are for the whole genus, but the annual species have very
small, small and only in rare cases intermediate 1C-values, whereas perennials are more
variable in genome size in general.

A correlation between the geographic distribution of species and genome size has been
found before: genomes were generally larger in species of temperate regions than in tropical
species (Avdulov, 1932; Levin & Funderburg, 1979; Ohri, 2005). Even though this has been
shown to be especially true in Asteraceae (Ohri, 2005), it is not applicable to Crepis as this
genus is mainly distributed in temperate regions. A correlation between geographic
distribution and genome size still could be found: Most species featuring very small genomes
are found in the circum Mediterranean region. In Central and East Asia, the easternmost
areas of the distribution range of Crepis, no species with very small genomes are found
(Fig.4). The continental climate of Central and East Asia is characterised by hot summers
and cold winters whereas in the Mediterranean climate fluctuations of temperature are
considerably lower. In contrast to their continental relatives, species in the Mediterranean are
not frequently subjected to frost. So, one factor to explain the absence of plants with very
small genomes from Central and East Asia could be freeze tolerance: A positive correlation
between freeze tolerance and genome size could be demonstrated for maize populations
(McMurphy & Rayburn, 2006). That temperature can influence genome size has also been
shown on population level by Turpeinen et al. (1999). Crepis is thought to have originated in
the Central Asian Altai/Tien Shan region (Babcock, 1947a), and to have spread from there
towards Europe and the Mediterranean. This means that many of the more derived species
are found in the mild climate of the Mediterranean (where species are also more likely to

46
 3. GENOME SIZE EVOLUTION

have a very small genome), whereas species with very small genomes are absent from
Central and East Asia, where a harsher climate prevails.

The altitudinal rank of species (lowland, mountainous or alpine) and endemical status
(widespread or restricted) showed no correlation to genome size. To further explore the
connection between the distribution of a species and its genome size, geographic and
ecological data is needed on a much finer scale (as has been used to explain infraspecific
(e.g. Kalendar et al., 2000) or infrageneric (Jakob et al., 2004) genome size variation).

Genome size is often correlated to phylogeny within the Asteraceae (e.g. Cerbah et al.,
1998; Torrell & Vallès, 2001; Garnatje et al., 2004), but in the present case no significant
correlation between genome size and clade (respectively taxonomic section sensu Babcock,
1947(a,b)) could be found. This can be in account of the comparatively small sample size of
the large genus and the heterogeneity of samples. Other chromosomal characters proved to
be more informative for phylogenetic purposes even in unclear species groups, such as
comparative C-band analysis in the C. praemorsa complex (Siljak-Yakovlev & Cartier, 1982).
Genome size is fairly constant within species but varies between species, which make it
useful for taxonomic purposes (Ohri, 1998; Murray, 2005). Genome size per se is not
reflecting the phylogeny within Crepis.

How genome size is influenced by selective pressure, is largely unknown. One approach is
the “large genome constraint hypotheses”: genera with large genomes are less likely to be
highly specious as accumulation and replication of “junk” DNA is associated with evolutionary
costs (Knight et al., 2005). Furthermore, it is now accepted that selection works on non
coding DNA as well as on coding regions (Bennett et al., 2000).

3.5 SUMMARY AND CONCLUSION

Within Crepis there is a remarkably unidirectional trend towards a decrease in genome size
of diploid species. The rare cases of increases in genome size within Crepis are always
resulting from polyploidisation. Life form and genome size are correlated. Annuals have
significantly lower 1C-values than perennials. This, however, could also be related to the
reproductive mode (selfing/outcrossing) of species. Crepis species in the milder climates of
the Mediterranean are also more likely to have smaller genomes than their Central Asian
relatives.

The present study should be a starting point for subsequent investigations into mechanisms
of genome size variation, especially mechanisms of decrease, in diploids. Comparative
studies of coding/noncoding DNA ratios in Crepis could shed light on the questions of the C-
value enigma. Analysis of speciation rates could improve insights on the “large genome

47
3. GENOME SIZE EVOLUTION

constraint hypothesis”: If species with large genomes are not likely to speciate fast, is the
reverse also true? Are plant groups with small genomes more prone to fast diversification?
The comparison of chromosomal maps or chromosomal banding patterns between close
relatives could elucidate how differences in chromosomal composition could be connected to
genome size.

Furthermore, it would be interesting to test environmental influences on genome size; e.g. if

the smaller genomes in the Mediterranean are connected to the mild climate or what other
factors (such as aridity, elevation, or physiological adaption) could be linked to the observed
patterns.

In conclusion this study emphasises the need to further investigate mechanisms and factors
which influence the genome size of a species, how selection acts upon genome size, and to
define the role genome size variation plays in speciation.

3.6 ACKNOWLEDGEMENTS
We thank T. Borsch and J.F. Wendel for stimulating advice. We also thank the curators of B,
E and M for providing plant material for DNA sequence studies. The present study is funded
by the DFG (GE 1242/3-2).

3.7 LITERATURE CITED

Albach, D.C. & Greilhuber, J. 2004. Genome Size Variation and Evolution in Veronica. Annals of
Botany. 94: 897-911.
Avdulov, N.P. 1931. Karyo-Systematische Untersuchungen der Familie Gramineen. Bulletin Applied
Botany Genetics and Plant Breeding 44: 1-428.
Babcock, E.B. 1947a. The Genus Crepis I. The Taxonomy, Phylogeny, Distribution and Evolution of
Crepis. University of California Publications 21. University of California Press, Berkeley & Los
Angeles.
Babcock, E.B. 1947b. The Genus Crepis II. Systematic Treatment. University of California
Publications 22. University of California Press, Berkeley & Los Angeles.
Bennett, M.D. 1972. Nuclear DNA Content and Minimum Generation Time in Herbaceaeous Plants.
Proceedings of the Royal Society of London B 181: 109-135.
Bennett, M.D., Johnston, S., Hodnett, G.L., & Price, H.J. 2000. Allium cepa L. Cultivars from Four
Continents Compared by Flow Cytometry Show Nuclear DNA Constancy. Annals of Botany
85: 351-357.
Bennett, M.D. & Leitch, I.J. 1995. Nuclear DNA Amounts in Angiosperms. Annals of Botany 76:113-
176.

48
 3. GENOME SIZE EVOLUTION

Bennett, M.D. & Leitch, I.J. 2005. Genome Size Evolution in Plants. Pp. 89-162. In: Gregory, T.R.
(ed.) The Evolution of the Genome. Elsevier Academic Press. Amsterdam, New York.
Bennett, M.D., Leitch, I.J., & Hanson, L. 1998. DNA Amounts in Two Samples of Angiosperm
Weeds. Annals of Botany 82 (Supplement A): 121-234.
Bennett, M.D. & Smith, J.B. 1976. Nuclear DNA Amounts in Angiosperms. Philosophical
Transactions of the Royal Society of London B 274: 227-274.
Bennetzen, J.L. & Kellogg, E.A. 1997. Do Plants Have a One-Way Ticket to Genomic Obesity? The
Plant Cell 9: 1509-1514.
Cerbah, M., Souza-Chies, T., Jubier, M.F., Lejeune, B., & Siljak-Yakovlev, S. 1998. Molecular
Phylogeny of the Genus Hypochaeris Using Internal Transcribed Spacers of Nuclear rDNA:
Inference for Chromosomal Evolution. Molecular Biology and Evolution 15(3): 1998.
Cheptou, P.O., Berger, A., Blanchard, A., Collin, C., & Escarre, J. 2000. The Effect of Drought
Stress on Inbreeding Depression in Four Populations of The Mediterranean Outcrossing Plant
Crepis sancta (Asteraceae). Heredity 85: 294-302.
Dart, S., Kron, P., & Mable, B. K. 2004. Characterizing Polyploidy in Arabidopsis lyrata Using
Chromosome counts and Flow Cytometry. Canadian Journal of Botany 82: 185-197.
Dimitrova, D., Ebert, I., Greilhuber, J., & Kozhuharov, S. 1999. Karyotype Constancy and Genome
Size Variation in Bulgarian Crepis foetida s.l. (Asteraceae). Plant Systematics and Evolution
217: 245-257.
Dimitrova, D. & Greilhuber, J. 2000. Karyotype And DNA-Content Evolution In Ten Species Of
Crepis (Asteraceae) Distributed In Bulgaria. Botanical Journal of the Linnean Society 132:
281-297.
Dimitrova, D. & Greilhuber, J. 2001. C-Banding Patterns and Quantitative Karyotype Characteristics
of Bulgarian Species of Crepis (Asteraceae). Plant Biology 3:88-97.
Enke, N. & Gemeinholzer, B. 2008. Babcock Revisited: New Insights Into Generic Delimitation And
Character Evolution in Crepis L. (Compositae: Cichorieae) From ITS And MatK Sequence
Data. Taxon 57(3): 756-768.
Flavell, R.B. 1980. The Molecular Characterization and Organization of Plant Chromosomal DNA
Sequences. Annual Review of Plant Physiology 31: 569-596.
Flavell, R.B. 1986. Repetitive DNA and Chromosome Evolution in Plants. Philosophical Transactions
of the Royal Society of London B 312: 227-242.
Garnatje, T., Vallès, J., Garcia, S., Hidalgo, O., Sanz, M., Canela, M.A., & Siljak-Yakovlev, S.
2004. Genome Size in Echinops L. and Related Genera (Asteraceae, Cardueae):
Karyological, Ecological and Phylogenetic Implications. Biology of the Cell 96: 117-124.
Govindaraju, D.R. & Cullis, C.A. 1991. Modulation of Genome Size in Plants: The Influence of
Breeding Systems and Neighbourhood Size. Evolutionary Trends in Plants 5: 43-51.
Grant, V. 1982. Periodicities in The Chromosome Numbers of The Angiosperm. Botanical Gazette
143:379-389.
Gregory, T.R. 2001. Coincidence, Coevolution, or Causation? DNA Content, Cell Size, and the C-
value Enigma. Biological Reviews 76: 65–101.
Greilhuber, J. 1998. Intraspecific Variation in Genome Size in Angiosperms: A Critical Reassessment.
Annals of Botany. 82 (Supplement A): 27-35.
Greilhuber, J. 2005. Intraspecific Variation in Genome Size in Angiosperms: Identifying its Existence.
Annals of Botany. 95: 91-98.

49
3. GENOME SIZE EVOLUTION

Greilhuber, J., Borsch, T., Müller, K., Worberg, A., Porembski, S., & Barthlott, W. 2006. Smallest
Angiosperm Genomes Found in Lentibulariaceae, with Chromosomes of Bacterial Size. Plant
Biology. 8: 770-777.
Greilhuber, J., Doležel, J., Lysák, M., & Bennet, M.D. 2005. The Origin, Evolution, and Proposed
Stabilisation of the Terms “Genome Size” and “C-value” to Describe Nuclear DNA Contents.
Annals of Botany. 95: 255-260.
Greilhuber, J. & Ebert, I. 1994. Genome Size Variation In Pisum sativum. Genome 37: 646-655.
Hancock, J.M. 2002. Genome Size And The Accumulation of Simple Sequence Repeats: Implications
of New Data From Genome Sequencing Projects. Genetica. 115: 93-103.
Hollingshead, L. 1930. Cytological Investigations of Hybrids and Hybrid Derivative Crepis capillaris
and Crepis tectorum. University of California Publications in Agricultural Sciences. 6: 55-94.
Hughes, M.B. & Babcock. E.B. 1950. Self-Incompatibility in Crepis foetida (L.) subsp. rhoeadifolia
(Bieb.) Schinz et Keller. Genetics. 35: 570-588.
Jakob, S.S., Meister, A., & Blattner, F.R. 2004. The Considerable Genome Size Variation of
Hordeum Species (Poaceae) Is Linked to Phylogeny, Life Form, Ecology, and Speciation
Rates. Molecular Biology and Evolution. 21(5): 860-869.
Jones, R.N. & Brown, L.M. 1976. Chromsome Evolution and DNA Variation in Crepis. Heredity.
36(1): 91-104.
Kalendar, R., Tanskanen, J., Immonen, S., Nevo, E., & Schulman, A.H. 2000. Genome Evolution of
Wild Barley (Hordeum spontaneum) by BARE-1 Retrotransposon Dynamics in Response to
Sharp Microclimatic Divergence. Proceedings of the National Academy of Science, USA. 97:
6603-6607.
Kamari, G. 1976 Cytotaxonomic Study of the Crepis neglecta complex in Greece. Publication of the
Botanical Institute, University of Patras.
Kamari, G. 1992. Karyosystematic Studies On Three Crepis Species (Asteraceae) Endemic To
Greece. Plant Systematics and Evolution. 182: 1-19.
Kellogg, E.A. 1998. Relationships of Cereals, Crops And Other Grasses. Proceeding of the National
Academy of Science, USA. 95: 2005-2010.
Kilian, N., Gemeinholzer, B. & Lack, H. W. 2008 (in press): Tribe Cichorieae. In: Funk, V., Susanna,
A., Stuessy, T. & Bayer, R. (ed.), Systematics and evolution of the Compositae. Vienna.
Kirik, A., Salomon, S. & Puchta, H. 2000. Species-specific Doublestrand Break Repair And Genome
Evolution In Plants. EMBO Journal. 19: 5562-5566.
Knight, C.A. & Ackerley, D.D. 2002. Variation in Nuclear Content Across Environmental Gradients: A
Quantile Regression Analysis. Ecology Letters. 5: 66-76.
Knight, C.A., Molinari, N.A., & Petrov, D.A. 2005. The Large Genome Constraint Hypothesis:
Evolution, Ecology and Phenotype. Annals of Botany 95(1): 177-190.
Kubis, K., Schmidt, T., & Heslop-Harrison J.S. 1998. Repetitive DNA Elements As A Major
Component of Plant Genomes. Annals of Botany. 82 (Supplement A): 45-55.
Leitch, I.J. & Bennett, M.D. 2004. Genome Downsizing in Polyploid Plants. Biological Journal of the
Linnean Society. 82: 651-663.
Leitch, I.J., Chase, M.W., & Bennett, M.D. 1998. Phylogenetic Analysis of C-DNA Values Provides
Evidence for a Small Ancestral Genome Size in Flowering Plants. Annals of Botany. 82: 85-
94.
Levin, D.A. 2002. The Role of Chromosomal Change in Plant Evolution. Oxford University Press, New
York.

50
 3. GENOME SIZE EVOLUTION

Levin, D.A. & Funderburg, S.W. 1979. Genome Size in Angiosperms: Temperate versus Tropical
Species. American Naturalist. 114: 784-795.
Loureiro, J., Rodriguez, E., Doležel, J., & Conceição, S. 2007. Two New Nuclear Isolation Buffers
for Plant DNA Flow Cytometry: A Test with 37 Species. Annals of Botany. 100(4): 875-888.
Ma, J., Devos, K.M., & Bennetzen, J.L. 2004. Analysis of LTR-Retrotransposon Structures Reveal
Recent and Rapid Genomic DNA Loss in Rice. Genome Research 14:860-869.
Maddison, W.P. & Maddison, D.R. 2008. Mesquite: A Modular System For Evolutionary Analysis.
Version 2.5. http://mesquiteproject.org.
Marie, D. & Brown, S.C. 1993. A Cytometric Exercise in Plant DNA Histograms, with 2C Values for 70
Species. Biology of the Cell. 78: 41-51.
Masterson, J. 1994. Stomatal Size in Fossil Plants: Evidence for Polyploidy In Majority of
Angiosperm. Science. 264:421-424.
McMurphy, L.M., & Rayburn, A.L. 2006. Genome Size Variation in Maize Populations Selected for
Cold Tolerance. Plant Breeding. 106(3): 190 – 195
Murray, B.G. 2005. When Does Intraspecific C-value Variation Become Taxonomically Significant?
Annals of Botany. 95: 119-125.
Ohri, D. 1998. Genome Size Variation and Plant Systematics. Annals of Botany 82 (Supplement A):
75-83.
Ohri, D. 2005. Climate and Growth Form: The Consequences for Genome Size in Plants. Plant
Biology. 7: 449-458.
Petrov, D.A. 2002. DNA Loss And Evolution of Genome Size in Drosophila. Genetica 115: 81-91.
Raina, S.N. & Rees, H. 1983. DNA Variation Between and Within Chromosome Complements of Vicia
Species. Heredity 51: 335-346.
Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed
models. Bioinformatics 19: 1572--1574.
Schmidt, T. & Heslop-Harrison, J.S. 1998. Genomes, Genes, and Junk: The Large Scale
Organization of Plant Chromosomes. Trends in Plant Science. 3: 195-199.
Sherman, M. 1946. Karyotype Evolution: A Cytogenetic Study of Seven Species And Six Intraspecific
Hybrids Of Crepis. University of California Publications in Botany. 18: 369-408.
Siljak-Yakovlev, S. & Cartier, D. 1982. Comparative Analysis of C-Band Karyotypes in Crepis
praemorsa subsp. praemorsa and subsp. dinarica. Plant Systematics and Evolution. 141(2):
85-90.
Sims, L.E. & Price, H.J. 1985. Nuclear DNA Content Variation in Helianthus (Asteraceae). American
Journal of Botany. 72: 1213-1219.
Soltis, D.E., Soltis, P.S., Bennett, M.D., & Leitch, I.J. 2003. Evolution of Genome Size in the
Angiosperms. American Journal of Botany 90(1): 1596-1603.
Swofford, D.L. 2002. PAUP*: Phylogenetic Analyses Using Parsimony (* and Other Methods), version
4.0 Beta. Sinauer, Sunderland.
Temsch, E.M. & Greilhuber, J. 2001. Genome Size In Arachis duranensis: A Critical Study. Genome.
44: 826-830.
Thomas, C.A. 1971. The Genetic Organizations of Chromosomes. Annual Review of Genetics. 5:
237-256.
Tobgy, H.A. 1943. A Cytological Study of Crepis fuliginosa, C. neglecta And Their F1 Hybrid, And Its
Bearing On The Mechanism of Phylogenetic Reduction In Chromosome Number. Journal of
Genetics. 45:67-111.

51
3. GENOME SIZE EVOLUTION

Torrell, M. & Vallès, J. 2001. Genome Size in 21 Artemisia Species (Asteraceae, Anthemideae):
Systematic, Evolutionary, and Ecological Implications. Genome. 44(2):231-238.
Turpeinen, T., Kulmala. J., & Nevo, E. 1999. Genome Size Variation In Hordeum spontaneum
Populations. Genome. 42(6):1094-1099.
Vicient, C.N., Suoniemi, A., Anamthamat-Jonsson, K., Tanskanene, J., Beharav, A., Nevo, E., &
Schulman, A.H. 1999. Retrotransposon BARE-1 And Its Role In Genome Evolution In The
Genus Hordeum. Plant Cell 11:1769-1784.
Wallace, H., Sparkes, C.A. & Maden, M. 1972. Nuclear DNA Content of Three Crepis Species.
Heredity 29: 267-373.
Watanabe, K., Yahara, T., Denda, T., & Kosuge, K. 1999. Chromosomal Evolution in the Genus
Brachyscome (Asteraceae, Asterae): Statistical Tests Regarding Correlation Between
Changes in Karyotype and Habit Using Phylogenetic Information. Journal of Plant Research.
112: 145-161.
Wendel, J.F. 2000. Genome Evolution In Polyploids. Plant Molecular Biology. 42: 225-249.
Wendel, J.F., Cronn, R.C., Johnston, J.S., & Price H.J. 2002. Feast and famine in plant genomes.
Genetica. 115: 37–47.
Yan, G., Yao, J., Ferguson, A.R., McNeilage, M.A., Seal, A.G., & Murray, B.G. 1997. New Reports
of Chromosome Numbers in Actinidia (Actinidiaceae). New Zealand Journal of Botany. 35:
181-186.
Zuckerkandl, E. 2002. Why So Many Noncoding Nucleotides? The Eukaryote Genome As An
Epigenetic Machine. Genetica 115: 105-129.

3.8 APPENDIX:
Taxa Sampled for DNA Content

GENUS, Taxon, Voucher, Location.

ASKELLIA: Askellia karelinii (Popov & Schischk.) W.A.Weber, N. Enke, T. Dürbye & B. Gemeinholzer NE0163
(B) Russia. CREPIS: Crepis aurea (L.) Cass., N. Enke NS0051 (B), Austria; C. biennis L., Roy et al. 844 (B),
Austria; C. blattarioides (L.) Vill., T. Dürbye 3195 (B), France; C. bungei Ledeb., N. Enke, T. Dürbye & B.
Gemeinholzer NE0178 (B), Russia; C. chrysantha Froel., N. Enke, T. Dürbye & B. Gemeinholzer NE0196 (B),
Russia; C. conyzifolia Dalla Torre, R. Vogt 15877 (B), Austria; C. crocea (Lam.) Babc., N. Enke, T. Dürbye & B.
Gemeinholzer NE0177 (B), Russia; C. foetida L., R. Hand 5263 (B), Greece: Cyprus; C. kerneri Rech. f., N. Enke
NS0047 (B), Italy; C. leontodontoides All., BG Liege 126-25-91-14 (B), France; C. lyrata (L.) Froel., N. Enke, T.
Dürbye & B. Gemeinholzer NE0149 (B), Russia; C. multicaulis Ledeb., N. Enke, T. Dürbye & B. Gemeinholzer
NE0164 (B), Russia; C. multicaulis Ledeb., E. Raab-Straube 020302 (B), Russia; C. paludosa Moench, BG
Bratislava 136-01-06-10 (B), Slovakia; C. polytricha (Ledeb.) Turcz., N. Enke, T. Dürbye & B. Gemeinholzer
NE0185 (B), Russia; C. pusilla (Sommier) Merxm., R. Hand 5130 (B), Greece: Cyprus; C. sibirica L., BG Halle
092-22-97-14 (B), Russia; C. turcica Dega & Bald, BG Paris 138-01-06-10 (B), Turkey; C. vesicaria T. Dürbye
1244 (B), Spain; YOUNGIA: Youngia tenuicaulis L., N. Enke, T. Dürbye & B. Gemeinholzer NE0149 (B), Russia;
Y. tenuifolia (Willd.) Babc. & Stebbins, N. Enke, T. Dürbye & B. Gemeinholzer NE0165 (B), Russia.

52
 3. GENOME SIZE EVOLUTION

Taxa Sampled for DNA Sequence

GENUS, Taxon, Voucher, Origin, GeneBank Accesion No. (ITS)

ASKELLIA: Askellia karelinii ( Popov & Schischk. ) W.A.Weber, N. Enke, T. Dürbye & B. Gemeinholzer NE0163
(B), Russia: Kosh-Agatsh FJ424075. CREPIS: Crepis alpina L., J. Trelawny 1423 (E), Turkey: Hakkari
EU363649; C. aurea (L.) Cass., K. Faber AU87 (B), Austria: Tauern, EU363627; C. biennis L., BG Frankfurt 217
109 34/03 (GAT) AJ633355; C. blattarioides (L.) Vill., T. Eckhardt 1555 (B), Switzerland: Valais EU363624; C.
bungei Ledeb., BG Uppsala/Sweden 31/113/03 (GAT), Mongolia: Tovaimag AJ633374; C. chrysantha Froel., N.
Enke, T. Dürbye & B. Gemeinholzer, NE0170 (B), Russia: Kosh-Agatsh FJ424077; C. crocea (Lam.) Babc., N.
Enke, T. Dürbye & B. Gemeinholzer, NE0177 (B), Russia: Kosh-Agatsh FJ424078; C. foetida L., J. Lambinon
00/F/214bis (B), France: Ardéche EU363619; C. fuliginosa Sibth. & Sm., D. Philos 4032 (B), Greece: Euboea
EU363601; C. incarnata (Wulf.) Tausch, H. & H. Doppelbaur 14684 (M), Italy: Südtirol EU363608; C. kerneri
Rech. f., T. Wraber, 9748/4 (B), Slowenia EU363636; C. lapsanoides (Gouan) Tausch, H. Kalheber 88-2748 (M),
France: Dpt. Hts. Pyr. EU363599; C. leontodontoides All.,Vogt 10/1992 (B), France: Corsica EU363592; C. lyrata
(L.) Froel., N. Enke, T. Dürbye & B. Gemeinholzer, NE0198 (B), Russia: Ongudai FJ424081; C. multicaulis Ldb.,
N. Enke, T. Dürbye & B. Gemeinholzer, NE0164 (B), Russia: Kosh-Agatsh FJ424076; C. neglecta L., Eisenblätter
& Willing 82.085 (B), Greece: Kardhitsa EU363610; C. palaestina (Boiss.) Bornm., A. Danin & al. 53.019 (B),
Israel: Galilee EU363639; C. paludosa Moench, P. Brückner 478/76 (B) EU366428; C. polytricha (Ldb.) Turcz., N.
Enke, T. Dürbye & B. Gemeinholzer, NE0185 (B), Russia: Ulagan FJ424080; C. praemorsa (L.) Tausch, G. van
Buggenhout 13637 (B), Italy: Bolzano EU363654; C. pusilla (Sommier) Merxm., Th. Raus 7715 (E), Greece:
Kasos EU363650; C. rubra L., Hort. Bot. Haunensis, 301-S1948-2634*A130 (GAT) AJ633350; C. sancta (L.)
Babc., J. Lambinon 84/F/373 (B), France: dep. Gard EU363632; C. setosa Hall. f., N. Enke 0123a (B), Italy: Pisa
EU363585; C. tectorum L., E. Willing 4.502 D (B), Germany: Berlin EU363643; C. vesicaria L., R. Vogt 11757 (B),
Marocco: Anti-Atlas EU363630; C. viscidula Froel., K.H. Rechinger 20933 (B), Greece: Mt. Smolika EU363629; C.
zacintha (L.) Babc., R. Hand 5323 (priv.), Greece: Rhodes EU363655. YOUNGIA: Youngia tenuicaulis (Babc. &
Stebbins) Czerep., N. Enke, T. Dürbye & B. Gemeinholzer, NE0181 (B) , Russia: Kosh-Agatsh FJ424079; Y.
tenuifolia (Willd.) Babc. & Stebbins, Beljaeva & al. s.n. (B), Russia: Primorje EU363645.

53
3. GENOME SIZE EVOLUTION

54
 4. ANATOMY AND MORPHOLOGY

Chapter 4

In the Search of Additional Characters Supporting Systematic

Delimitation in Crepis L. and Related Genera in the Subtribe
Crepidinae (Cichorieae/Compositae)

ABSTRACT
Recent molecular work on Crepis and its allied genera (e.g. Askellia, Lapsana, Rhagadiolus,
and Youngia) raised several questions on generic and infrageneric classification within the
Crepidinae. Lapsana and Rhagadiolus as well as Crepis species from sections Intybellia,
Phaecasium, Lagoseris, Microcephalum, and Pterotheca were found to be closely related. To
resolve the incongruence between clades inferred by molecular approaches and current
taxonomic classification additional morphological characters need to be assessed.

Morphological characters within the genus have been intensively studied; however, extensive
parallel evolution of traits complicates the recognition of natural groups. Here, a pilot study
was carried out to test additional characters (fruit morphology and anatomy, pappus
ultrastructure, pollen morphology, and style branch papillae) for their applicability in generic
delimitation and infrageneric classification.

KEYWORDS: Crepis, achene morphology, achene anatomy, pappus bristles, pollen, SEM,
style branch papillae.

55
 4. ANATOMY AND MORPHOLOGY

4.1 INTRODUCTION
The uniform character combination of milky latex and capitula with 5-dentate, ligulate flowers
makes the members of the predominantly holarctic Cichorieae easy to identify. However, the
classification within the tribe is obstructed by the notorious lack of characters suitable for
delimitation. Since Tournefort (1694) first recognised the Cichorieae until the most recent
classification by Kilian and co-workers (2008), the circumscription of the Cichorieae did not
change much, whereas the generic and suprageneric classification was subject to various
changes. This is also true for the subtribe Crepidinae. The subtribe Crepidinae gained
importance in the first half of the 20th century through the work of two American botanists,
E.B. Babcock and G.L. Stebbins who studied the genera of the Crepidinae not only
morphologically but also cytologically and used the results to establish new classifications
and generic circumscriptions (e.g. Babcock et al., 1937; Babcock & Stebbins, 1937; Babcock
& Jenkins, 1943; Babcock, 1947a,b). Among the most notable works is Babcock’s (1947a,b)
monograph of the genus Crepis, that was used as basis for recent molecular work (Enke &
Gemeinholzer, 2008). The availability of molecular phylogenies (e.g. Enke & Gemeinholzer,
2008) led to new insights into generic interrelations within the Crepidinae, but also raised
new questions: discrepancies between relations found by molecular inference and
taxonomically recognised groups demonstrate the need to further evaluate the current
taxonomic classification. Of special importance here is the delimitation of Lapsana L. and
Rhagadiolus Juss. as these two genera were found to be nested within Crepis L. by
molecular data (Enke & Gemeinholzer, 2008), even though they are easily distinguished on
morphological grounds (e.g. fruit morphology). The infrageneric delimitation within Crepis
also needs revision: The molecular analyses by Enke & Gemeinholzer (2008) could not
corroborate the current taxonomic sections (Babcock, 1947b). A thorough morphological re-
evaluation of species and genera, and the identification of additional discriminating
characters are needed to achieve a revised classification. Jeffrey (1966) stated that
microcharacters (e.g. the shape of hairs on the stigmatic surface) provide good criteria for
taxonomy in the Cichorieae (Compositae). The microcharacteristic investigations by e.g. Pak
& Kawano (1990), Sennikov & Illarionova (2007) and Torres & Galetto (2007) confirm the
usability of a micromorphological approach. Here, a pilot study is carried out to screen
possible micromorphological and anatomical traits for their applicability in generic and
infrageneric classification within Crepidinae. These traits include fruit morphology and
anatomy, pappus ultrastructure, pollen morphology and style branch papillae.

56
 4. ANATOMY AND MORPHOLOGY

4.2 MATERIAL AND METHODS

Achene Habit and Surface - Dry achenes of 23 species (Youngia tenuicaulis, Y. tenuifolia,
A. elegans, A. flexuosa, A. karelinii, A. nana, Lapsana communis, Rhagadiolus spec., Crepis
bocconi, C. bungei, C. chrysantha, C. crocea, C. foetida, C. lapsanoides, C. lyrata, C.
multicaulis, C. paludosa, C. purpurea, C. pusilla, C. pyrenaica, C. tectorum, and C. zacintha)
were documented at the Wild M5A stereomicroscope equipped with digital camera Leica
DFC 290 and Leica Application Suite software Version 2.5.0. The fruits were taken from
either herbarium specimen or living plants (appendix).

Ultra Thin Sections for Light Microscopy - Achenes of 21 species (Rhagadiolus spec.,
Askellia flexuosa, A. nana, Crepis acuminata, C. albida, C. biennis, C. capillaris, C.
chondrilloides, C. foetida, C. kerneri, C. lapsanoides, C. leontodontoides, C. mollis, C.
multicaulis, C. neglecta, C. paludosa, C. praemorsa, C. purpurea, C. sancta, C. tectorum,
and C. zacintha) for light microscopy were taken from herbarium sheets or living plant
material (Appendix). Dried achenes were stored in 96% ethanol. Fruit from living material
was progressively dehydrated by ascending ethanol solutions (30%, 50%, 70%, 90%, and
96%). The achenes remained 24h in each dilution. For infiltration with resin (Unicryl, BBI
International) the objects were first transferred into a mix of 1:2 Unicryl and Ethanol (96%),
then stepwise into 1:1, 2:1 and last into 100% Unicryl. The samples remained 3-6 days
(depending on size) at each step. The objects were embedded in gelatine capsules filled with
resin and dried for 3-5 days in a heating cabinet at 40°C. The ultra thin sections through the
middle part of the achene were cut at a rotation microtome (Supercut 2065, Reichert/Jung).
The transverse sections (3- 4 µm) were stained with toluidine blue (Serva, 0.5%, 20-35 s),
mounted in corbit-balm (Kobe) and dried for 2 days at 40°C. Micrographs were taken at Zeiss
microscope Standard 14 with the digital documentation system Zeiss Axio Cam MRc and
Axio Vision software (release 4.4, Zeiss).

Preparation of Pollen for SEM – Pollen samples were taken from 12 species (Crepis
albida, C. biennis, C. dioscoridis, C. foetida, C. hypochaeridea, C. lapsanoides, C.
leontodontoides, C. paludosa, C. pulchra, C. sancta, C. tectorum, and C. vesicaria). Prior to
coating samples were treated by acetolysis following Erdtman (1960) to avoid artefacts by
the protoplast. After acetolysis pollen grains were suspended in ethanol and the pollen
surface was cleaned from debris in an ultrasonic bath. The pollen suspension was
transferred onto a 14mm cover slip mounted on an SEM stub, and left to dry. The samples
were coated with gold and studied with a LEO Supra 55VP.

57
4. ANATOMY AND MORPHOLOGY

FIG.1 (continued p.60)

58
 4. ANATOMY AND MORPHOLOGY

Preparation of Other Samples for SEM – Samples (10 species, Askellia nana, Crepis
bungei, C. capillaris, C. kerneri, C. lapsanoides, C. leontodontoides, C. pulchra, C. sancta, C.
tectorum, C. zacintha) for SEM were fixed in FAA (5 ml formaldehyde solution (min. 35%), 15
ml glacial acetic acid, 20 ml Ethanol (96%), and 60 ml Aqua dest.) for 2x 24h. Dehydration of
the object was facilitated by a subsequent treatment of ascending Ethanol dilutions (70%,
80%, 90%, 96% and 2x 100%). The samples remained in each step for at least 1h. Then the
samples were treated with acetone (100%) twice for 1h. The samples were transferred to the
Critical Point Drier K 850 (EMITECH) for final desiccation. Then the objects were mounted on
aluminium stubs and coated with gold/palladium (layer thickness 20nm) in a Low Voltage
Cool Sputter Coater K 550 (EMITECH). The specimen stubs were studied with a Philips SEM
515. The objects studied were pappus bristles, papillae on the inside of style branches.

4.3 RESULTS
Results are summarised in Table 1. Missing data was complemented by values from
literature as far as possible. Sources are given in Table 1.

Achene Morphology

Achenes of the 23 sampled Crepidinae species are either terete or fusiform, unbeaked,
attenuate, or beaked and in most cases straight with no or 8-20 ribs. Few achenes are
striate. Size ranges between 1.3 and 7mm. Colour varies from light gold brown to black
(Fig.1).

Youngia – Achenes are black with lighter spicules, terete and have up to 12 ribs. Size varies
between 2.6 mm (Y. tenuifolia) and 3.7 mm in Y. tenuicaulis.

Askellia – Achenes are fairly uniform in size (3.5-3.7 mm) and shape (terete). The fruits are
smooth, and unbeaked with 8-10 ribs, except for A. flexuosa, where achenes are fusiform,
attenuate and spicate. The prevailing achene colour is light brown, the only exception to that
is A. elegans, which has black achenes.

Lagoseris group, Lapsana, and Rhagadiolus – Within sampled genera of the Crepidinae
the achenes of Lapsana and Rhagadiolus are distinct from all others by gross morphology:
Lapsana and Rhagadiolus both possess no pappus in contrast to the barbellulate bristles of
all Crepis species. The fruits of Lapsana are 2.6 mm in average, flattened, broadly fusiform,
and light brown with twenty prominent lighter ribs with a smooth surface. The achenes of
Rhagadiolus are fusiform and between 4.7mm and 7mm long. The outer achenes are

59
 4. ANATOMY AND MORPHOLOGY

enclosed in the involucral bracts, stellately outspread and smooth. The inner achenes are
strongly curved, yellowish white, with few shallow but strongly spiculate ribs. The fruits of the
five Crepis species of the Lagoseris clade are heterogeneous: size ranges between 2.5-5
mm, the shape is terete, attenuate or biform, with colours being light brown or brown with a
smooth, rugulose, or spiculate surface. Number of ribs varies between 10 and 12, except for
C. sancta where achenes are striate. C. multicaulis and C. sancta both have conspicuously
soft and bendable pappus bristles.

Crepis s.st. – C. lapsanoides of Clade I has a broadly fusiform, unbeaked and light brown
achene with ca. 17 smooth ribs. Size is 3.4-3.7mm. The other analysed species of Clade I,
C. lyrata, has terete, slightly curved, light brown achenes of 2.6-2.8mm length with ca. 18
smooth ribs. In Clade II the achenes of C. tectorum and C. paludosa share no similarity
except size (2.6-2.9mm). The first species has dark brown, fusiform, and spiculate achenes.
The latter possesses strongly terete and brown achenes, with few prominent, lighter and
smooth ribs. In clade III C. pusilla and C. zacintha both have comparatively small (1.3-1.8
mm), slightly curved and pubescent achenes. C. foetida differs from these two species in the
biform achenes, which can be strongly beaked. Size is ca. 7mm. The only sampled species
of clade IV, C. bocconi, has terete, attenuate, and brown achenes of 5.6mm length with 18
smooth ribs. The Central Asian species of Clade VII (C. chrysantha, C. bungei, and C.

FIG. 1 (continued): Achene morphology and surface structure. a) Youngia

tenuicaulis, b) Y. tenuifolia, c) Askellia flexuosa, d) A. karelinii, e) A .nana, f) A.
elegans, g) Rhagadiolus spec., h) Crepis sancta, i) Lapsana communis, k) C.
multicaulis, l) C. purpurea, m) C. lapsanoides, n) C. lyrata, o) C. paludosa, p) C.
tectorum, q) C. pusilla, r) C. zacintha, s) C. foetida, t) C. bocconi, u) C. bungei,
v) C. crocea, w) C. chrysantha, x) C. blattarioides.

60
 4. ANATOMY AND MORPHOLOGY

crocea) have slender achenes of (dark) brown colour with prominent spiculate or rugulose
ribs. The fruits are terete or fusiform, unbeaked to attenuate and between 3.1mm and 6.9mm
long. C. blattarioides of Clade XI has fusiform, attenuate, and brown achenes of 3.9-4.3mm
length with 18 smooth ribs.

Achene Anatomy

In anatomical contexts ribs are referred to as costae. The achenes are normally of rounded
outline with (8) 10-12 (20) costae made of sclerenchymatous cell bundles. The exocarp is
one-layered with a thick outer cell wall, but the cells can be collapsed. Parenchymatic
regions might or might not be present in the mesocarp between costae or between costae
and testa. Endocarp is two layered and collapsed (Fig.2b). Four different achene types could
be found in Crepis s.l. (Fig.2).

Type Ia – Achenes are of a rounded outline. The cells of the exocarp have thick outer cell
walls but are (partly) collapsed. The 10-12 costae are far apart with distinct intercostal areas
where parenchyma cells are partly collapsed. No intercostal sclerenchymatous cells are
present.

Type Ib – Similar to Ia, except that intercostal parenchyma cells are well developed. 3-6
layers of protoplastic parenchyma cells present in the mesocarp between the testa and the
costae.

Type Ic – This type has no distinct costae. Sclerenchymatous islands are embedded in the
parenchymatous cells of the mesocarp. No intercostal sclerenchyma is found. The outer
walls of the one layered exocarp are only slightly thickened.

Type II - The achenes are of round outline, with 8-12 pointed costae. Parenchyma is well
developed beneath the sclerenchyma of the costae, but often collapsed in the intercostal
regions. The sclerenchyma builds a band in the intercostal areas.

Type III - Achenes are +/- circular in outline. The costae are (weakly) prominent, intercostal
regions are mostly made up of 1-6 cell layers. Intercostal sclerenchymatous cells are
present.

Type IV – Exocarp can be collapsed. Costae are very prominent with deep or no intercostal
furrows. Sometimes 3-6 layers of protoplastic parenchymatous cells are found between testa
and sclerenchyma but never between the sclerenchymatous islands of the costae. Even
though the costae seem to merge in some cases, they are always separated by a layer of
collapsed parenchyma cells or intercostal furrows.

61
 4. ANATOMY AND MORPHOLOGY

Askellia – Both sampled species (A. flexuosa and A. nana) have Type Ia achenes.

Lagoseris group and Rhagadiolus – Except for C. purpurea (Type VI) and C. praemorsa
(Type Ia) with deep intercostal furrows, the other 4 sampled species of the Lagoseris group
as well as Rhagadiolus are of Type III. The exocarp is one layered with thick outer cell walls
(intermediate in C. sancta and C. multicaulis) and bearing protruding elements, except for C.
incarnata. Rhagadiolus spec. and C. sancta bear long cellular appendages.

Crepis s.str. – Both species of Clade I feature achenes of Type I. C. mollis has achenes of
Type Ic, whereas C. lapsanoides of Type Ib, the most frequent type. In Clade II C. tectorum
is Type Ib and C. paludosa a Type II. C. kerneri, however, has to be considered intermediate
between Type Ib and II, as it has slightly pointed costae but no intercostal sclerenchyma. In
Clade III C. zacintha shows a distinct achene anatomy: it is of angular outline and a Type IV
where costae are nearly merged. C. foetida is Type Ib. C. acuminata and C. chondrilloides of
Clade V are Type II respectively Type Ia. In Clade VI only C. neglecta as Type Ia has been
sampled. C. biennis (Clade VII) is Type Ib as is C. albida of Clade IX. C. leontodontoides of
Clade VII is Type IV; costae are nearly merged as in C. zacintha. C. capillaris (Clade XI) is
Type Ib and achene anatomy shows high similarity to C. tectorum.

FIG.2: Achene cross sections stained with toluidine blue. LM. a) Typa Ia, A. nana b) Type Ib, C. foetida c) Type
Ic, C. mollis d) Type II, C. acuminata e) Type III, C. multicaulis f) Type IV C. zacintha. cM collapsed, C costa, En
endocarp, Ex exocarp, Ic intercostae, IS intercostal sclerenchyma, M mesocarp, SB sclerenchymatic band.

62
 4. ANATOMY AND MORPHOLOGY

Pappus Bristles

Pappus bristles vary in diameter, number of cells they are composed of, and prominence and
frequency of spikes. The different pappus bristles are shown in Fig.3.

Askellia – A. nana has pappus bristles made up of 6-8 cells in diameter. The 1-4 spikes per
100µm are adherent to the bristle, thus the pappus bristles are smooth. The diameter is 30-
32µm.

Lagoseris group – The pappus bristles of the sampled species of the Lagoseris clade are
fine (only 10-15µm in diameter, especially in C. sancta) and made up of 2-3 cells. The spikes
are very far apart (0-2 per 100µm) but sticking out prominently.

Crepis s.str. – C. lapsanoides of Clade I has very spiky pappus bristles (6-7 spikes/100µ)
and the bristles are comparatively strong (19-20µm, 4-6 cells). C. kerneri (Clade II) has very
coarse pappus bristles (35-37µm, 6-7 cells) with prominent spikes. In C. tectorum (Clade III)
the spikes are pointed, sharp and long, the bristles are slender (14-15µm, 3-4 cells). C.
zacintha, also of Clade III, has shorter spikes but similarly slender bristles (15-16µm, 4-5
cells). In Clade VII C. bungei has the strongest bristle of all sampled species (37-40µm, 6-
7cells), but only comparatively few adherent spikes (3-4 spikes/100µm). In C.
leontodontoides (Clade VII) the spiky pappus measures 20-22µm and is made up of 3-5
cells. The pappus of C. capillaris (Clade XI) is one of the finer ones within Crepis s.str. (16-
17µm, 4-5 cells). It has 3-4 spikes per 100µm.

Pollen Morphology

Terminology is according to Blackmore (1984). All sampled species have echinolophate

pollen of the Cichorium intybus type. 5 species fall into the Cichorium intybus subgroup
(Crepis albida, C. dioscoridis, C. lapsanoides, C. paludosa, C. pulchra), 7 species into the
Taraxacum officinalis subgroup (C. biennis, C. foetida, C. hypochaeridea, C.
leontodontoides, C. sancta, C. tectorum) (Fig.4). Grain size ranges from 25-42µm.

Style Branch Papillae

The papillae showed some variation and three types could be identified (Fig.5). The first
type, Type A, is a slender (7.5-10µm) papilla, the sides are nearly parallel and gradually
narrow into a tip. Type B is narrow at the base, widens in the middle (11-20µm) and narrows
from the widest point in a straight line to the tip. The median widening in Type C is more
distinct (19-25µm) and the narrowing towards the tip is incurved rather than straight. Type B
is the most frequent type (5 species: Askellia nana, Crepis bungei, C. capillaris, C. kerneri,
and C. pulchra). Type A (C. leontodontoides, C. zacintha) and Type C (C. lapsanoides, C.
tectorum) are found in 2 species each.

63
 4. ANATOMY AND MORPHOLOGY

4.4 DISCUSSION
In achene morphology Askellia and both sampled Youngia species differ mainly in colour and
outline, as the fruits of the Youngia species are more rounded and black, whereas the
Askellia achenes are gold brown and columnar; except for A. elegans, where the achene is
very similar to Youngia in their dark colour with spicules. A. elegans is restricted to North
America, so no distributional overlap with the strictly Asian Youngia species is given. Achene
anatomy in Youngia (sec. Desiphylum = Crepidifolium) differs from the presently studied
Askellia species (Type Ia, Fig.2): Youngia features no clear furrows between costae
(Sennikov & Illarionova, 2008), and therefore resembles Type Ic. The achene cross section
of A. elegans published by Pak (1993) is of Type Ia. However, according to Pak (1993)
variation within the genus Askellia is high and as the cross section given there for A. nana
differs from the present findings. Character continuity within and between species should be
investigated in further analyses on additional samples.

Of the here studied morphological aspects the only character uniquely found in Askellia is the
thick pappus (Ø30-32µ) with a smooth surface (Fig.3).

Achene and pollen features of Lapsana, Rhagadiolus, and the five Crepis sections (Intybellia,
Phaecasium, Lagoseris, Microcephalum, and Pterotheca) are by far more heterogeneous
than in Askellia.

In fruit anatomy Type III achenes are uniquely present in the Lagoseris clade. It is not the
only type found in the Lagoseris group though; Type Ia (C. praemorsa) and Type IV (C.
purpurea) are found as well (table 1, Fig.2). Regarding the pollen types both Cichorium
intybus and Taraxacum officinale subtypes are found in the Lagoseris group. The structure of
the pappus bristles differ to Crepis s.str. Both sampled species (C. pulchra, C. sancta) of the
Lagoseris group have few celled, fine (10-15µm) and very soft pappus bristles, which in its
smoothness (0-2 spikes/100µ) are unique in all sampled species (Fig.3).

Within Crepis s.str. the significance of presently tested microcharacters varies strongly.
General trends of the variability of microcharacters are discussed on exemplary groups.
Species affiliation to clades and sections is given in table 1.

First are the species of section Zacintha, namely C. pusilla and C. zacintha (Clade III); their
peculiar morphology is reflected not only in a close molecular relation (Enke & Gemeinholzer,
2008) but also in the presently studied characters: Fruit morphology, fruit anatomy and style
branch papillae, which show unique characters among tested taxa of Crepis s.str., except the
similarity to C. leontodontoides (table 1). C. zacintha and C. leontodontoides show
resemblance in style branch papillae; both species have Type A papillae. C. pusilla and C.
zacintha have by far the smallest achenes in the genus (Fig.1). In the case of section
Zacintha, molecular results, morphology and microcharacters agree.

64
 4. ANATOMY AND MORPHOLOGY

FIG. 3: Pappus bristles. SEM. a) A. nana. b) C. sancta. c) C. pulchra. d) C. lapsanoides. e) C. kerneri. f) C.

tectorum. g) C. zacintha. h) C. bungei. i) C. leontodontoides. k) C. capillaris.

Second are the Central Asian species of Clade VII: C. chrysantha, C. crocea, and C. bungei;
even though the species in clade VII are from various sections (Brachypodes, Macropodes,
and Mesophylion, respectively) they show morphological similarity. For the three sampled
species this is also reflected in their achene morphology (Fig.1). C. bungei and C. oreades
are thought to be the putative diploid parents of tetraploid C. crocea (Babcock, 1947b). C.
chrysantha is thought to be one of the putative parents of C. polytricha (Babcock, 1947b).
The “hybrid” complex is obviously closely related, and needs a new systematic treatment.

65
 4. ANATOMY AND MORPHOLOGY

Two species from different clades as well as different sections show resemblance in their
fruit anatomy C. tectorum (Clade II, sec. Mesophylion) and C. capillaris (Clade XI, sec.
Phytodesia), both species have highly similar Type Ib achenes.

The distinctiveness of the North American species of section Psilochaenia is mirrored in fruit
morphology and anatomy. The ribs are very prominent and distinctly pointed in C. acuminata
(Fig.2d), which has not been observed in any of the other species.

The presently studied C. vesicaria (sec. Lepidoseris) specimen has pollen of the Cichorium
intybus subtype. According to Blackmore (1984) C. vesicaria has pollen of the Taraxacum
officinale subtype (Table 1). As C. vesicaria with at least eight subspecies (Babcock, 1947b)
is a highly polymorphic species, pollen type constancy throughout the (sub)species might not
be given and should be further investigated.

FIG.4: Pollen grain types. SEM. a) Cichorium intybus group, C. foetida b)

Taraxacum officinale group, C. tectorum. Polar view. Scale 4 µm.

Systematic Usability of Tested Characters

The limitation of the presently tested characters is the sample size, which is in almost all
cases too small to provide conclusive evidence for the delimitation of groups within the large
and heterogeneous genus Crepis. However, some characters proved to be more indicative of
relationships between species than others.

As studies in the Cichorieae have shown before (e.g. Pak & Kawano, 1990; Pak, 1993; Pak
et al., 2001; Zhu et al., 2006; Sennikov & Illarionova, 2008) fruit anatomy is helpful to
investigate infrageneric relationships. With a wider sampling of Crepis species a possible
developmental linkage of achene types to phylogeny might provide new insights (viz. did one
type of achenes derive from another). Especially for the different subtypes of Type I achenes
(Fig.2) a progressive development of types could be hypothesised. Furthermore, not only
cross sections of the achenes might be useful to support systematic problems but also
longitudinal sections could provide additional information. Longitudinal sections of testa
epidermis cells by Tegel (2002) suggested a high systematic value of this feature at generic
level. The testa epidermis is less exposed to selective pressure than the pericarp and
therefore more likely to carry independent characters (Grau, 1980; Tegel, 2002). However,
variation within tested Crepis species is low (Tegel, 2002). Zarembo & Boyko (2008) came to

66
 4. ANATOMY AND MORPHOLOGY

a similar conclusion in Cardueae where variation of testa structures is low within but helpful
to distinguish between genera.

The pappus has always been an important feature to discriminate groups on all taxonomic
levels in the Cichorieae (for a recent review see e.g. Kilian et al, 2008). The here presented
data suggests pappus ultrastructure as possibly being discriminative at least on the generic
level. Again, a larger sample size could provide evidence for a revised infrageneric
classification of Crepis s.str.. In addition to presently tested ultrastructural characters of the
pappus, other features should be included; e.g. the initiation of the pappus on the achene
and the number of pappus parts per achene; the latter is genetically determined (Bachmann
et al., 1981)

FIG. 5: Style branch papillae. SEM. a) Type A, C. zacintha. b) Type B, C. pulchra. c) Type C, C. bungei.

Micromorphological investigations of Torres & Galletto (2007) found stigmatic papillae within
the Cichorieae to be useful characters to support systematics. Even though continuity of
characters could be observed within studied species, style branch variation should be
interpreted with caution. The species might not be restricted to one type of papillae. Types
might vary both with location on style branches and with life stage of the plant.

Pollen analyses of the present study partly differ from those specified by Blackmore (1984)
(Table 1), which formerly has been carried out with different technology and different
resolution. Pollen was not informative for intergeneric delimitation within Crepis, at least not
at the morphological level which has been assessed in the present study. Pollen
ultrastructure and anatomy could provide additional information.

In conclusion, some characters feature promising variation for delimitation on sectional level
but a far bigger sample size is needed in regard to species number as well as sampled
individuals per species. The first is needed to test the interspecific constancy, whereas the
second will give evidence on infraspecific variation respectively constancy of characters.

67
 4. ANATOMY AND MORPHOLOGY

4.5 SUMMARY AND CONCLUSION

The microcharacters, especially the pappus, vary more between the sampled genera
(Askellia, Crepis, Lapsana, Rhagadiolus and Youngia) than within. Within Crepis the
variation of characters reflects either the sectional classification (Babcock, 1947b), namely in
the section Zacintha, the groups based on molecular marker (e.g. Clade VII), or
morphological and physiological similarity as in C. tectorum and C. capillaris.

The use of several achene features is considered promising for genus delimitation as well as
infrageneric classification; namely anatomy and pappus ultra structure. Style branch micro
morphology shows some potential as diagnostic character, but needs further investigation as
does pollen morphology.

The applicability of the tested characters to the delimitation of genera and subgeneric groups
is mainly impeded by sample size.

4.6 AKNOWLEDGEMENTS
The author thanks H.C. Weber, K. Dörr and M. Rath (Philipps University, Marburg), S.
Blackmore, M. Watson and F. Christie (Royal Botanic Garden Edinburgh) and M. Lüchow
(Botanic Garden Berlin) for technical advice, and the curators of B and E, R. Hand and B.
Zimmer for providing plant material. This study is funded by the DFG (GE 1242/3-2).
Synthesys provided the funding for the visit at the RBGE.

68
 4. ANATOMY AND MORPHOLOGY

TABLE 1. Summary of Results. Species ordered for affiliation to molecular clades.

Papilla
1 2
Species Clade Section Achene Characters Pappus Pollen Type
length No of Ø spikes/ No of Ø
(mm) color Ribs Ø (µm) 100µm cells Type (µm)
Youngia
Youngia - 3.4-3.7 black 10-12 - - - - - - -
tenuicaulis
Y. tenuifolia Youngia - 2.6 black 10-12 - - - - - - -
3.5 light
Askellia flexuosa Askellia Ixeridopsis 10 Ia - - - - - -
(4-5.52) brown
3.7 light
A. nana Askellia Ixeridopsis 10 Ia 30-32 1-4 6-7 - - B
(4-62) brown
light
A. karelinii Askellia Ixeridopsis 3.4 10 - - - - - - -
brown
3.6-3.7
A. elegans Askellia Ixeridopsis black 10 - - - - - - -
(52)
Lapsana light
Lapsana - 2.6 20 - - - - - - -
communis brown
light
Rhagadiolus spec. Rhagadiolus - 4.8-7.0 - III - - - - - -
brown
3.2-3.4
C. multicaulis Lagoseris Microcephalum brown 10 III - - - - - -
(42)
2 2
C. pulchra Lagoseris Phaecasium 4-6 - striate - 14-15 0-2 2-3 C 28-34 B
3.4 light
C. purpurea Lagoseris Lagoseris 12 IV - - - - - -
(4-52) brown
2 light 2 3 3
C. praemorsa Lagoseris Intybellia 4.5-5 20 Ia - - - T 38-46 -
brown2
light 3-
C. sancta Lagoseris Pterotheca 2.8-4.6 III 10-11 0-2 2-3 T 28-34 -
brown striate2
3.4-3.7 dark
C. lapsanoides Clade I Mesomeris 20 Ib 19-20 6-7 4-6 C 31-37 C
(5-62) brown
2.6-2.8 light
C. lyrata Clade I Mesomeris 20 - - - - - - -
(3.5-42) brown
2 2 2 3 3
C. mollis Clade I Mesomeris 3-4.5 brown 20 Ic - - - C 34-38 -

C. kerneri Clade II Brachypodes - - - Ib 35-37 4-6 6-7 - - B

2.9 (4.5- light
C. paludosa Clade II Desiphylion 10 II - - - C 36-42 -
5.52) brown
2.6-2.8 dark 3
C. tectorum Clade II Mesophylion 10 Ib 14-15 2-3 3-4 T(C ) 26-32 B
(3-42) brown
light
C. foetida Clade III Hostia 6.9 10 Ib - - - T 26-32 -
brown
dark
C. pusilla Clade III Zacintha 1.3 6 - - - - - - -
brown
1.8
C. zacintha Clade III Zacintha brown 10 IV 16-17 3-4 4-5 - - A
(2.52)
5.6
C. bocconi Clade IV Soyeria brown 18 - - - - - - -
(10-12*)
light
C. acuminata Clade V Psilochaenia 5.5-9* brown 122 II - - - - - -
*
C. chondrilloides Clade V Berinia 5-7* - 14-182 Ia - - - - - -
2 dark Striate
C. dioscorides - Brachypodes 5.0 - - - - C 26-32 -
brown2 2

2 dark 2
C. hypochaeridea Clade V Anisoramphus 6.5-10 10-13 - - - - T 31-37 -
brown2
2 light 2
C. neglecta Clade VI Phytodesia 2-2.5 10 Ia - - - - - -
brown
2 light 2 3
C. biennis Clade VII Berinia 4.0-7.5 10-20 Ib - - - T(C ) 33-39 -
brown2
3.1 dark
C. bungei Clade VII Mesophylion 12 - 37-40 3-4 6-7 - - C
(4-52) brown
dark
C. chrysantha Clade VII Brachypodes 4.2-6.3 14 - - - - - - -
brown
4.0-4.9 16-
C. crocea Clade VII (II) Macropus brown - - - - - - -
(5-62) (182)
2 2 2
C. leontodontoides Clade VIII Gephyroides 3.5-5 brown 10 IV 13-14 2-4 2-3 T 25-31 A
2 light 2
C. albida Clade IX Paleya 10-17 15 Ib - - - C 26-32 -
brown2
2 light 2 3
C. vesicaria Clade X Lepidoseris 5-9 10-12 - - - - C(T ) 27-33 -
brown2
light/d
C. capillaris Clade XI Phytodesia - ark 102 Ib 16-17 3-4 4-5 T3 34-413 B
brown2
3.9-4.3
C. blattarioides Clade XI Soyeria brown 18 - - - - - - -
(5.8-82)
1 2 3
Enke & Gemeinholzer (2008); Babcock (1947b); Blackmore (1984)

69
 4. ANATOMY AND MORPHOLOGY

4.7 LITERATURE
Babcock, E.B. 1947a. The Genus Crepis I. The Taxonomy, Phylogeny, Distribution and Evolution of
Crepis. University of California Publications 21. University of California Press, Berkeley & Los
Angeles.
Babcock, E.B. 1947b. The Genus Crepis II. Systematic Treatment. University of California
Publications 22. University of California Press, Berkeley & Los Angeles.
Babcock, E.B. & Jenkins, J.A. 1943. Chromosomes and phylogeny in Crepis, III: The relationships of
one hundred and thirteen species. University of California Publications in Agricultural Science
Sci. 18: 241-292.
Babcock, E.B. & Stebbins, G.L. 1937. The Genus Youngia. Carnegie Institution of Washington
Publication 484. Carnegie Institution of Washington, Washington, DC.
Babcock, E.G., Stebbins, G.L., & Jenkins, J.A. 1937. Chromosomes and Phylogeny in Some
Genera of the Crepidinae. Cytologia Fuji Jubilee Volume 188-210.
Bachmann, K., Chambers, K.L., & Price, H.J. 1981. Genetic Determination of Pappus Part Number
in the Annual Hybrid Microseris B87 (Asteraceae – Lactuceae). Plant Systematics and
Evolution 138(3-4): 235-246.
Blackmore, S. 1984. Compositae – Lactuceae. In: Punt, W. & Clarke, G.C.S. (eds.) The Northwest
European Pollen Flora IV. pp. 45-85. Elsevier, Amsterdam.
Enke, N. & Gemeinholzer, B. 2008. Babcock Revisited: New Insights into Generic Delimitation and
Character Evolution in Crepis L. (Compositae: Cichorieae) From ITS and MatK Sequence
Data. Taxon. 57(3): 756-768.
Erdtman, G. 1960. The acetolysis method. A revised description. Svensk Botanisk Tidskrift 54:561-
564.
Grau, J. 1980. Die Testa der Mutisieae und ihre systematische Bedeutung. Mitteilungen der
botanischen Staatsammlung München 16: 269-332.
Jeffrey, C. 1966. Notes on Compositae I. The Cichorieae in East Tropical Africa. Kew Bull. 18: 427--
486.
Kilian, N., Gemeinholzer, B. & Lack, H. W. 2008 (in press): Tribe Cichorieae. In: Funk, V., Susanna,
A., Stuessy, T. & Bayer, R. (ed.), Systematics and evolution of the Compositae. Vienna.
Pak, J.H. 1993. Taxonomic Implications of Fruit Wall Anatomy and Karyology of Crepis sect.
Ixeridopsis (Compositae; Lactuceae). Korean Journal of Plant Taxonomy 23(1): 11-20.
Pak, J.H. & Kawano, S. 1990. Biosystematic Studies on the Genus Ixeris (Compositae – Lactuceae)
I. Fruit Wall Anatomy and its Taxonomic Implications. Acta Phytotaxonomica et Geographica
41: 43-60.
Pak, J.H., Park, J.K., & Whang, S.S. 2001. Systematic Implications of Fruit Wall Anatomy and
Surface Sculpturing of Microseris (Asteraceae, Lactuceae) and Relatives. International Journal
of Plant Sciences 162(1): 209-220.
Sennikov, A.N. & Illarionova, I.D. 2007. Generic Delimitation of the Subtribe Ixeridinae newly
Segregated from Crepidiinae (Asteraceae – Lactuceae). Komarovia 5(2): 57-115.
Tegel, F. 2002. Die Testaepidermis der Lactuceae (Asteraceae) – ihre Diversität und systematische
Bedeutung. Doctorate Thesis, Ludwigs-Maximillians-University of Munich.
http://edoc.ub.unimuenchen.de/archive/00000104/

70
 4. ANATOMY AND MORPHOLOGY

Torres, C. & Galetto, L. 2007. Style Morphological Diversity of Some Asteraceae Species from
Argentina: Systemtic and Functional Implications. Journal of Plant Research 120: 359-364.
Tournefort, J. P. de 1694. Élemens de botanique 1–3. Imprimerie Royale, Paris.
Zarembo, E.V. & Boyko, E.V. 2008. Carpology of Some East Asian Cardueae (Asteraceae). Anales
del Jardin Botanico de Madrid 65(1):129-134.
Zhu, S.X., Qin, H.N., & Shih, C. 2006. Achene Wall Anatomy and Surface Sculpturing of Lactuca L.
and Related Genera (Compositae: Lactuceae) with Notes on Their Systematic Significance.
Journal of Integrative Plant Biology. 48(4): 390-399.

71
 4. ANATOMY AND MORPHOLOGY

4.8 APPENDIX

Taxa Sampled for Achene Morphology

GENUS, Taxon, Voucher, Location.

ASKELLIA: Askellia elegans (Hook.) W.A. Weber, W.J. Cody, Gutteridge s.n. (B), Canada; A. flexuosa (Ledeb.)
W.A. Weber, S. Smirnov s.n. (B), Mongolia, Altai Mountains; A. karelinii (M.Pop. et Schischk. ex Czer.) W.A.
Weber, N. Enke, T. Dürbye & B. Gemeinholzer NE0163 (B), Russia, Kosh-Agatsh; A. nana (Rich.) W.A. Weber,
N. Enke, T. Dürbye & B. Gemeinholzer NE0186 (B), Russia, Ulagan. CREPIS: Crepis blattarioides Vill., N. Enke
NE0070 (B), France, Pyrenees; C. bocconi P.D. Sell, B. Zimmer 2644 (B), Austria; C. bungei Ledeb., N. Enke, T.
Dürbye & B. Gemeinholzer NE0178 (B), Russia, Kosh-Agatsh; C. chrysantha (Ledeb.) Froel., N. Enke, T. Dürbye
& B. Gemeinholzer NE0168 (B), Russia, Kosh-Agatsh; C. crocea (Lamk.) Babc., N. Enke, T. Dürbye & B.
Gemeinholzer NE0167 (B), Russia, Kosh-Agatsh; C. foetida L., R. Hand 5263 (priv), Greece, Cyprus; C.
lapsanoides (Gouan) Tausch, N. Enke NE0020 (B), France, Pyrenees; C. lyrata Froel., N. Enke, T. Dürbye, B.
Gemeinholzer NE0147 (B), Russia, Ongudai; C. multicaulis Ledeb., N. Enke, T. Dürbye & B. Gemeinholzer
NE0164 (B), Russia, Kosh-Agatsh; C. paludosa Moench, N. Enke NE0019 (B), France, Pyrenees; C. purpurea
(Willd.) M. Bieb.; Steven s.n. (B), Russia; C. pusilla (Sommier) Merxm., R. Hand 5310 (priv), Greece, Cyprus;C.
sancta (L.) Babc., K.H. Rechinger s.n. (B), Persia; C. tectorum L., N. Enke NE0076 (B), Switzerland, Vallis; C.
zacintha (L.) Babc., R. Hand 5323 (priv), Greece, Cyprus; LAPSANA: Lapsana communis L., Willing 18929 (B),
Germany; RHAGADIOLUS: Rhagadiolus spec. Juss., BG St. Gallen s.n. (B), France; YOUNGIA: Youngia
tenuicaulis (Babc. & Stebbins) Czerep., N. Enke, T. Dürbye & B. Gemeinholzer NE0180(B), Russia, Kosh-Agatsh;
Y. tenuifolia (Willd.) Babc. & Stebbins, N. Enke, T. Dürbye & B. Gemeinholzer NE0148 (B), Russia, Ulagan.

Taxa Sampled for Achene Ultra Thin Sections

GENUS, Taxon, Voucher, Location.

ASKELLIA: Askellia flexuosa (Ledeb.) W.A.Weber, Kürschner Sonnentag 01-203 (B) China, Gansu; A. nana
(Rich.) W.A. Weber, J.A. Calder s.n. (B), Canada, Fairy Lake; CREPIS: Crepis acuminata Nutt., L.S. Rose s.n.
(B), USA, California; C. albida Vill., R. Valdes s.n. (B), Spain, Almeria; C. aurea Rchb., N. Enke NE0142 (B),
Austria; C. biennis L., N. Enke NE0143 (B), Austria; C. capillaris (L.) Wallr., N. Enke NE0043 (B), Spain,
Pyrenees; C. chondrilloides Jacq., Bornmüller s.n. (B); C. foetida L., R. Hand 5263 (priv), Greece, Cyprus; C.
kerneri Rech. f., N. Enke NE104 (B), Italy, Dolomites; C. lapsanoides (Gouan) Tausch, N. Enke NE0020 (B),
France, Pyrenees; C. leontodontoides All., Greuter & Agababian 24457 (B), Italy, Sicily; C. mollis (Jacq.) Asch.,
Koziol s.n. (B), Poland; C. multicaulis Ledeb., Raab-Straube 020302 (B), Russia, Altay; C. multicaulis Ledeb.,
Dürbye s.n. (B), Kirghizia, Tien Shan; C. neglecta (Sm) Vierh., Nielssen s.n. (B) Greece, Etolias; C. paludosa
Moench, N. Enke NE0019 (B), France, Pyrenees; C. praemorsa (L.) Tausch, van Bouggenhout s.n. (B), Italy,
Bolzano; C. purpurea (Willd.) M. Bieb., Steven s.n. (B), Russia; C. rubra L., E.Willing 13378 (B), Greece, Etolia;
C. sancta (L.) Babc., K.H. Rechinger s.n. (B), Persia; C. sancta (L.) Babc., J. Lambinon s.n. (B), France, Gard; C.
tectorum L., N. Enke NE0076 (B), Switzerland, Vallis; C. zacintha (L.) Babc., R. Hand 5323 (priv), Greece,

72
 4. ANATOMY AND MORPHOLOGY

Cyprus; RHAGADIOLUS: Rhagadiolus stellatus (L.) Gaertn., R. Hand 2265 (B), Cyprus; R. stellatus.(L.) Gaertn.,
W. Lang s.n. (B) Cyprus, Salamis.

Taxa Sampled for SEM (Pollen)

GENUS, Taxon, Voucher, Location.

CREPIS: Crepis albida Vill., P.F.Cannon, P.R. Crane, S.L. Jury, D.M. Moore 1023 (E), Spain, Almeria; C. biennis
L., N.Enke NE0146 (B), Austria; C. dioscorides L., Raus et al. s.n. (B), Greece, Pelopones; C. foetida ssp.
Commutata (Spreng) Babc., J.R. Edmondoson, M.A.S. McClintock E 2513 (E), Greece, Tokmakia; C.
hypochaeridea (DC.) Thell., N.J. Devenish 1657 (E), South Africa; C. lapsanoides (Gouan.)Tausch, D.W. Dresser
1256a (E), Spain, Oviedo; C. leontodontoides All., BG Liege (B), France, Corse; C. paludosa (L.) Moench, M.F.
Gardner, S.G. Gardner s.n. (E), Germany; C. pulchra L., 137-02-06-70 (B), BG Konstanz; C. sancta (L.) Babc.,
Romi s.n. (B), Italy, Siena; C. tectorum L. Coll., Aune Haakana s.n. (E), Finland, Nylandia; C. vesicaria L., N.Enke
NE0016 (B), Frankreich, Pyrenees.

Taxa Sampled for SEM (Pappus Bristles, Style Branch Papillae)

GENUS, Taxon, Voucher, Location.

CREPIS: Crepis albida Vill., P.F.Cannon, P.R. Crane, S.L. Jury, D.M. Moore 1023 (E), Spain, Almeria; C. biennis
L., N.Enke NE0146 (B), Austria; C. dioscorides L., Raus et al. s.n. (B), Greece, Pelopones; C. foetida ssp.
Commutata (Spreng) Babc., J.R. Edmondoson, M.A.S. McClintock E 2513 (E), Greece, Tokmakia; C.
hypochaeridea (DC.) Thell., N.J. Devenish 1657 (E), South Africa; C. lapsanoides (Gouan.)Tausch, D.W. Dresser
1256a (E), Spain, Oviedo; C. leontodontoides All., BG Liege (B), France, Corse; C. paludosa (L.) Moench, M.F.
Gardner, S.G. Gardner s.n. (E), Germany; C. pulchra L., 137-02-06-70 (B), BG Konstanz; C. sancta (L.) Babc.,
Romi s.n. (B), Italy, Siena; C. tectorum L. Coll., Aune Haakana s.n. (E), Finland, Nylandia; C. vesicaria L., N.Enke
NE0016 (B), Frankreich, Pyrenees.

73
4. ANATOMY AND MORPHOLOGY

74
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

Chapter 5

Guideline to a New Infrageneric System in Crepis L.

(Compositae/Cichorieae)

ABSTRACT
The molecular analysis of the genus Crepis by Enke & Gemeinholzer (2008) revealed
several problems concerning the systematic subdivision of the genus. The genus Crepis
proved to be polyphyletic and split into three clades statistically well supported by molecular
markers ITS and matK. The first group comprises most sampled species as Crepis s.str., the
second clade species of five Crepis sections (Intybellia, Lagoseris, Phaecasium,
Microcephalum, and Pterotheca) as well as the genera Lapsana and Rhagadiolus; the third
group includes all species of Crepis section Ixeridopsis.

Suggestions concerning a revised infrageneric classification are made, considering both

molecular and morphological evidence: the genera Lapsana and Rhagadiolus are preserved
in their current generic circumscription and Crepis is treated as paraphyletic taxon. Crepis
sections Desiphylion, Omalocline, Mesomeris, Psilochaenia, Lagoseris, Hostia,
Microcephalum, Pterotheca, Zacintha, Lepidoseris, Nemauchenes, and Psammoseris are
retained in their current sectional delimitation.

KEYWORDS: Crepis, infrageneric classification, Lagoseris, Phaecasium

75
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

5.1 INTRODUCTION
The genus Crepis L. with over 200 species (Bremer, 1994) is widely distributed throughout
the northern hemisphere and Africa. The last revision of the genus comprises detailed
descriptions of 196 species in 27 sections (Babcock, 1947a,b). Babcock was one of the first
to use not only morphology and biogeography but karyotypic similarity as criterion for
infrageneric relations. Babcock (1947a,b) also assumed that the sectional system reflected
phylogenetic relations within the genus.

Cichorieae, including Crepis, are

notorious for their lack of
discriminative morphological
characters. Many characters vary
more within a species than
between closely related species
and this is also true for some
characters on the generic level. In
the past this often led to unclear
specific and generic boundaries.
Molecular data can provide
further insights on species
relationships. But to draw
taxonomic consequences from
phylogenies inferred from
molecular sequence data,
support is needed from additional
morphological, anatomical or
karyological characters as names
of plant species are linked to type
specimen and taxa are defined by
morphological character
complexes.
FIG.1: Distribution of Babcock’s sections (right) on molecular
clades (left) named following Enke & Gemeinholzer, 2008. Line The recent establishment of a
width corresponds to number of species. Sections held in grey molecular phylogeny of the
have not been sampled for molecular data.
genus (Enke & Gemeinholzer,
2008) based on ITS and matK sequence data revealed Crepis L. to be diphyletic. Two clades
comprising Crepis species are statistically well supported by both nuclear and chloroplast
marker (Enke & Gemeinholzer, 2008): one group comprises the (monotypic) genera Lapsana
and Rhagadiolus as well as Crepis species from sections Intybellia, Lagoseris, Phaecasium,
Microcephalum, and Pterotheca. The other, larger group includes the main part (ca. 80%) of

76
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

sampled Crepis species as monophyletic clade Crepis s.str. (see Enke & Gemeinholzer,
2008: Fig.1-2). The infrageneric classification of Babock (1947b) is in many cases not
congruent with molecular clades, indicating that the present sections do not represent natural
groups. In fact, many molecular clades comprise more than one section, whereas most
sections emerge into more than one clade (Fig. 1; and Enke & Gemeinholzer, 2008: Fig.3).
Some suggestions are made towards the taxonomic treatment of the Lagoseris clade
(including the genera Lapsana and Rhagadiolus) as well as the delimitation of infrageneric
groups within Crepis s.str. (sensu Enke & Gemeinholzer, 2008). For this Babcock's
morphological, karyological and biogeographic findings are critically reassessed, and
recently presented evidence) from molecular and morphological analyses are taken into
account (Enke & Gemeinholzer, 2008; Enke et al., subm.; chapter 4).

5.2 DISCUSSION
A comparison of all species sampled for DNA sequence data (Enke & Gemeinholzer, 2008)
in regard to Babcock’s (1947b) sectional classification, the clades inferred by a molecular
phylogenetic approach (Enke & Gemeinholzer, 2008), and the proposed revised sectional
system are shown in table 1.

Sections Desiphylion, Omalocline, Mesomeris, Psilochaenia, Lagoseris, Hostia,

Microcephalum, Pterotheca, Zacintha, Lepidoseris, Nemauchenes, and Psammoseris are
temporarily preserved in their current sectional circumscription as neither molecular nor
morphological data indicate otherwise.

Lagoseris Clade

The close relation of Crepis sections Intybellia, Phaecasium, Lagoseris, Microcephalum, and
Pterotheca of the Lagoseris clade, and the genera Lapsana and Rhagadiolus, is statistically
well supported by both nuclear and chloroplast data (Enke & Gemeinholzer, 2008). All
sampled species of the five Crepis sections (Intybellia, Phaecasium, Lagoseris,
Microcephalum, and Pterotheca) appear in the Lagoseris clade, no sectional overlap with
Crepis s.str. could be observed (Fig.1. and Enke & Gemeinholzer, 2008). However, not only
the distinct characteristics of Lapsana and Rhagadiolus but also the variation of characters
within the Crepis species render this clade problematic.

Two of the five Crepis sections within the Lagoseris clade, namely Lagoseris and Pterotheca,
are treated as genus Lagoseris in the Flora of the USSR (Bobrov & Tzevelev, 2000). The
exclusion of both species from Crepis is mainly based on the presence of conspicuously long
bristle-like palae on the receptacle, which can sometimes exceed the achenes (Bobrov &
Tzevelev, 2000). Babcock (1947a) reported natural occurrences of individuals of C. sancta
(sec. Pterotheca) lacking palae. Collins (1924) discovered that the presence and absence of
palae on the receptacle is due to a very simple genetic mechanism. Furthermore, palae are

77
5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

TABLE 1 (CONTINUED ON NEXT PAGE) All species sampled by Enke & Gemeinholzer (2008) ordered for
sectional assignment (after Babcock, 1947b) with molecular relations and revised sectional assignment.
Sectional
Species 2 Molecular Clades3 Revised Classification
Classification
C. paludosa Moench
C. viscidula Froel.
Desiphylion Clade II Desiphylion
C. pygmaea L. Omalocline Clade XI Omalocline
C. jaquini Tausch
C. kerneri Rech. f.
Clade II Brachypodes
C. dioritica Schott et Kotschy Clade V
C. chrysantha (Ledeb.) Turcz Brachypodes Macropodes
C. polytricha Turcz.1 Clade VII
C. rhaetica Hegetschw.
C. aurea (L.) Cass Clade VIII New Section II
C. hierosolymitana Boiss.
C. lapsanoides (Gouan) Tausch
C. lyrata Froel.1 Mesomeris Clade I Mesomeris
C. mollis (Jacq.) Asch.
C. smyrnaea DC.
C. blattarioides (L.) Vill. Clade XI
Soyeria
C. bocconi P.D. Sell Clade IV
C. albida Vill. Paleya Clade IX Gephyorides
C. alpestris (Jacq.) Tausch Clade XI
Anisoramphus
C. hypochaeridea (DC.) Thell. Clade V
C. leontodontoides All. Clade VIII New Section II
Gephyroides
C. tingitana Ball. Clade IX Gephyroides
C. taygetica Babc.
Clade III
C. triasii (Camb.) Fries
C. lacera Ten. Clade II
C. baldaccii Halácsy
Clade IV
C. darvazica Krasch.
C. chondrilloides Jacq.
C. guioliana Babc.
C. macropus Boiss. et Heldr.
C. merxmuerlleri Kamari et Hartvig Clade V Berinia
C. sibthorpiana Boiss. et Heldr. Berinia
C. sonchifolia C.A. Mey.
C. turcica Degen et Baldacci
C. turcomanica H. Krasch.
C. biennis L. Clade VII
C. oporinoides Boiss. Clade XI
C. auriculaefolia Sieber ?
C. incana Sibth. et Sm. ?
C. pannonica (Jacq.) K. Koch ?
1
C. crocea (Lam.) Babc.
Clade VII Macropodes
C. oreades Schrenk Macropodes
C. hookeriana J. Ball. Clade X
A. flexuosa (Ledeb.) W.A. Weber
A. karelinii (Popov. & Schischk.)
W.A. Weber Ixeridopsis Askellia Askellia
A. nana (Richards.) W.A. Weber
C. incarnata (Wulf.) Tausch Phaecasium
C. praemorsa (L.) Tausch
Intybellia Lagoseris
C. bungei Ledeb. Mesophylion
Clade VII
C. nigrescens Pohle Mesophylion ?
C. tectorum L. Clade II Phytodesia
C. acuminata Nutt. Psilochaenia Clade V Psilochaenia
C. frigida (Boiss.) Babc.
C. purpurea (Willd.) M. Bieb Lagoseris Lagoseris Lagoseris
C. sahendi Boiss. et Buhse
C. palaestina (Boiss.) Bornm.
C. pterothecoides Boiss. Phaecasium Lagoseris Phaecasium
C. pulchra L.
C. alpina L.
C. foetida L.
C. kotschyana Boiss. Clade III
C. rubra L.
Hostia Hostia
C. thomsonii Babc.
C. tybakiensis Vierh.
C. multicaulis Ledeb Microcephalum Microcephalum
Lagoseris
C. sancta (L.) Babc Pterotheca Pterotheca
C. pusilla (Sommier) Merxm.
C. zacintha (L.) Babc.
Zacintha Clade III Zacintha
C. corymbosa Ten.
C. cretica Boiss.
C. fuliginosa Sibth.& Sm.
Clade VI New Section
C. neglecta L. Phytodesia
C. capillaris (L.) Wallr. Clade XI
C. nicaeënsis Balb. Clade II Phytodesia
C. parviflora Desf. ?

78
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

TABLE 1 (CONTINUED)

Sectional
Species 2 Molecular Clades3 Revised Classification
Classification
C. vesicaria L. Lepidoseris Clade X Lepidoseris
C. aspera L. ?
Nemauchenes Nemauchenes
C. setosa Hall.f. ?
C. bellidifolia Loisel.
C. bursifolia L.
Psammoseris Clade X Psammoseris

1 2 3
Sequence published in Enke et al. (subm.); Babcock, 1947a,b; Enke & Gemeinholzer, 2008

not unique to sections Lagoseris and Pterotheca, they are also present in C. commutata
(Crepis s.str., sec. Hostia, syn. C. foetida ssp. commutata; Babcock & Cave,1938). These
findings led Babcock (1947a) to the inclusion of Pterotheca and Lagoseris into Crepis.

C. multicaulis (as the only representative of sec. Microcephalum sampled in the molecular
analysis of Enke & Gemeinholzer (2008)) resembles C. sancta (sec. Pteroctheca) in some
aspects of morphology and karyotype (Babcock & Jenkins, 1943), but not in the receptacular
palae as they are lacking in C. multicaulis. C. multicaulis and C. sancta also show a close
molecular relation (Enke & Gemeinholzer, 2008). However, the morphological resemblance
of C. multicaulis to other species (e.g. the other species of sec. Microcephalum) is more
poignant than that to C. sancta (Babcock, 1947b). So, to include the whole section
Microcephalum into a putative new genus Lagoseris would be rather artificial, especially
because all the species of Microcephalum lack the bristle like palae, which are one of the
main delimiting characteristics of the genus Lagoseris sensu Bobrov & Tzevelev (2000).

C. praemorsa and C. incarnata of section Intybellia are very closely related and sometimes
treated as subspecies of C. praemorsa (Sell, 1976). The section Intybellia is based on the
genus Intybellia Monn. (non Cass.; Intybellia Cass. corresponds to Crepis section Pterotheca
respectively the genus Pterotheca Cass.), and has been recombined on several taxonomic
levels. There is some taxonomic overlap between the sections Intybellia and Phaecasium.
Monnier (1929) included C. pulchra (presently in section Phaecasium) into Intybellia along
with C. praemorsa and C. incarnata. Before that C. pulchra was included in genus
Phaecasium as P. lampsanoides Cass. (Cassini, 1826) and later returned to the genus as P.
pulchrum Benth. et Hook. by Bentham & Hooker (1873). Babcock & Jenkins (1943) would
have merged the two sections into one section Intybellia due to their identical karyotype
features if it were not for differences in root morphology. As root morphology is influenced by
ecological factors (Verboom et al., 2004), it is inapt as systematically discriminating factor.
The karyotypic resemblance between C. praemorsa and C. pulchra is likewise supported, if
banding patterns given by Siljak-Yakovlev & Cartier (1982) for C. praemorsa and Dimitrova &
Greilhuber (2001) for C. pulchra are compared. Both molecular markers (ITS, matK) support
the close relationship of these two sections (Enke & Gemeinholzer, 2008). This would
provide sufficient evidence to merge sections Intybellia and Phaecasium into Phaecasium
Cass.

The Crepis species of the Lagoseris group differ in pappus ultrastructure from Crepis s.str.
(chapter 4). The pollen of Lapsana is similar to the pollen of Crepis (Cichorium intybus

79
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

pollen, subtype Taraxcum officinale, chapter 4, Blackmore, 1984), but according to Osman
(2006) Rhagadiolus has a distinct pollen type with 21 lacunae compared to only 15 in Crepis
and Lapsana.

The here discussed data is ambiguous with respect to the question whether to exclude the
five sections clustering within the Lagoseris clade from Crepis, or to treat Crepis as
paraphyletic genus. The variation of characters within Crepis species of the Lagoseris clade
is mostly within the range known for species of Crepis s.str. Chemosystematical evidence
shows that C. multicaulis and C. pulchra (as representatives of the Lagoseris group) are very
similar in the composition of their phytochemical compounds to the other 21 sampled Crepis
species (all belonging to Crepis s.str.), whereas Lapsana differs (Zidorn, 2008). For all
characters which have been used to accept the species of sections Intybellia, Phaecasium,
Lagoseris, Microcephalum, and Pterotheca as separate genera Lagoseris (including
Lagoseris, Pterotheca and possibly Microcephalum) and Phaecasium (including Intybellia
and Phaecasium) equivalent character states could be found within Crepis s.str.. Lapsana
and Rhagadiolus differ from all Crepis species in achene features and the latter distinctly in
pollen type. As has been shown by Tegel (2002), the cell wall structure of the testa epidermis
in the achenes is fenestrate in Lapsana and Rhagadiolus; whereas it is unstructured in all
sampled Crepis species except for C. biennis (Tegel (2002) sampled C. sancta and C.
pulchra of the Lagoseris group). Conclusively, no argument could be found to encourage an
exclusion of the species of the Lagoseris group from Crepis; neither could any convincing
argument be found to merge Lapsana and Rhagadiolus into Crepis.

As the discussed characters allow no palpable decision whether to exclude the relevant
sections from Crepis, it is proposed to preserve the current generic circumscription of Crepis,
even though it would be paraphyletic from a molecular point of view, until further evidence
emerges. To expand the generic description of Crepis to include Lapsana and Rhagadiolus
seems inappropriate given the morphological distinctness of both genera. Furthermore, the
phylogeny of Crepis s.l. (Enke & Gemeinholzer, 2008) might reflect a more complex
evolutionary history than can be drawn from dichotomous branching patterns of phylogenetic
trees, so further analyses and investigations are still needed.

Crepis s.str.

Some of Babcock’s sections are retained in their present circumscription (e.g. section
Mesomeris), while others are partly reorganised (e.g. Phytodesia). Several species are
relocated to other sections or left for further consideration.

The Central Asian species of Clade VII (C. bungei, C. chrysantha, C. crocea, and C.
polytricha) are obviously related as the species are similar in morphological, karyological and
molecular characteristics (Enke & Gemeinholzer, 2008, Enke et al., subm.; chapter 4). Two

80
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

additional species in Clade VII which share morphological and molecular similarities are C.
oreades and C. rhaetica (Enke & Gemeinholzer, 2008). C. bungei and C. oreades are
considered to be the putative diploid parents of tetraploid C. crocea, whereas C. chrysantha
is suspected to be one of the parents of tetraploid C. polytricha (Babcock, 1947b). Popov
(1957-59), on the other hand considered C. bungei, C. crocea and C. polytricha to have
arisen from a hybridisation between Youngia tenuifolia and C. chrysantha. The species of
this “hybrid complex”, however, belong to different sections: C. chrysantha, C. polytricha and
C. rhaetica belong to section Brachypodes, C. bungei to section Mesophylion, and C.
oreades and C. crocea to section Macropodes. C. bungei and C. oreades are the type
species for their sections.

Section Brachypodes is heterogeneous. Seven out of ten species in this section have been
sampled for DNA sequence data; three species (namely C. chrysantha, C. polytricha and C.
rhaetica) cluster in Clade VII, whereas the four others clustered in three different clades: C.
jacquini and C. kerneri both in Clade II, C. dioritica in Clade V and C. aurea in Clade VIII
(Enke & Gemeinholzer, 2008). C. chrysantha and C. polytricha have a basic chromosome
number of x = 4, C. jacquini and C. aurea have x = 5. The chromosome numbers of C.
rhaetica and C. dioritica are unknown.

Section Mesophylion includes four species, C. bungei, C. ircutensis, C. nigrescens and C.

tectorum. In the Flora of the USSR (Bobrov & Tzevelev, 2000) C. bungei and C. ircutensis
are both treated as C. bungei. C. bungei and C. tectorum show a distant relation in the matK
analysis, but cluster in completely different clades in the nuclear phylogeny; C. bungei in
Clade VII and C. tectorum in Clade II (Enke & Gemeinholzer, 2008). Morphologically C.
nigrescens is very similar to C. tectorum but differs from it in the type of pubescence on the
stem and involucrum as well as a larger and darker corolla (Bobrov & Tzevelev, 2000). In the
molecular based phylogeny it is, however, sister to C. bungei (Enke & Gemeinholzer, 2008).
C. bungei and C. tectorum share a karyotypic similarity (Babcock, 1947b). Microcharacters,
however, indicate some relation of C. tectorum to C. capillaris (sec. Phytodesia, Clade XI;
Chapter 4).

C. oreades and C. crocea from section Macropodes are the only species in the section with a
Central Asian distribution, whereas all other species are of Mediterranean or African
distribution. The only other member of section Macropodes sampled for DNA sequence data
is C. hookeriana, a Northwest African species, found in Clade X (Enke & Gemeinholzer,
2008) and shows alliances to species centred in NW African/SE Spain, e.g. C. dianthoseris,
C. albida, C. tingitana and C. oporinoides, and beyond in the Mediterranean and S Europe to
species such as C. vesicaria and C. alpestris (Enke et al., 2008).

For the species of the “hybrid complex” it is proposed, with all reservations in respect to
further morphological and karyological studies, to transfer C. chrysantha, C. polytricha and C.
rhaetica to section Macropodes and to leave C. bungei in section Mesophylion. Section

81
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

Brachypodes needs critical reassessment, as molecular data (Enke & Gemeinholzer, 2008)
indicates that section Brachypodes is polyphyletic (table 1).

As the relations between the species within the “hybrid complex” and the relation of the
“hybrid complex” to other species such as e.g. C. tectorum and C. nigrescens are very
complex they need further careful investigation. Interestingly, the tetra-/octoploid species C.
biennis (sec. Berinia) clusters within Clade VII as well (Enke & Gemeinholzer, 2008), so all
polyploid species sampled for ITS sequence data (C. biennis, C. crocea and C. polytricha)
are found in clade VII the “hybrid complex”. Genetical and cytological characterisation of the
species would be the method of choice as morphology in this case is hard to interpret in
regard to the hybridogeneous origin of at least some of the species.

As mentioned above C. tectorum shows similarity to C. capillaris. The two species feature
gross morphological similarities, but have been placed in different sections as the seed of C.
capillaris show a longer viability (Babcock, 1947b). The genetical investigations of
Hollingshead (1930a,b) suggest some relation between these two species. Fruit anatomical
similarities could, however, reflect convergent evolution. These two polymorphic species
share some similarity, such as their annuality, their wide distribution and some gross
morphological congruencies with two other species, namely C. nicaeënsis and C. parviflora
(both sec. Phytodesia). C. tectorum is sister to C. nicaeënsis in the ITS phylogeny (Clade II),
whereas C. tectorum and C. parviflora cluster together in the chloroplast based phylogeny
(Enke & Gemeinholzer, 2008). After further careful morphological consideration and the
discussion of the above mentioned objections, it might appear reasonable to transfer C.
tectorum to sec. Phytodesia, of which C. nicaeënsis is the type species.

Clade VI partly reflects the relations Babcock (1947b) assumed for section Phytodesia. The
C. neglecta complex comprises in addition to C. neglecta (including subspecies) the species
C. fuliginosa and C. cretica. The closer relation within these species than to the others of
same section is also reflected in their karyotypes (Babcock & Jenkins 1943). Cytological
studies by Tobgy (1943) and Kamari (1976) demonstrated the close relation within this
complex. Clade VI can be considered to be equivalent to the C. neglecta complex. The type
species of Phytodesia, however, is C. nicaeënsis of Clade II, so a new section for this group
is necessary.

Most species of section Berinia sampled for DNA sequence data cluster in Clades IV and V
(table 1 and Enke & Gemeinholzer, 2008). Section Berinia, the biggest section in Babcock’s
(1947) monograph of the genus Crepis, is divided into four subsections: Corymbiformae,
Subcorymbiformae, Divaricatae and Strictae, which are however not reflected by molecular
data (Fig.3 in Enke & Gemeinholzer, 2008). The species of this large and heterogeneous
section need careful further consideration. The type species of section Berinia is C.
chondrilloides that clusters within Clade V, so it is assumed that Clade V and IV correspond
to section Berinia.

82
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

Clade VII includes only C. aurea (section Brachypodes) and C. leontodontoides (section
Gephyroides). Both species have similar and fairly small karyotypes (Babcock & Jenkins,
1943). The two species have been considered to be closely related before, but have been
placed into different sections due to differing root morphology (Babcock & Jenkins, 1943;
Babcock 1947b). Avery (1930) reported viable hybrids between these two species. Their
close relation is likewise well supported by nuclear as well as chloroplast data (Enke &
Gemeinholzer, 2008). C. aurea and C. leontodontoides should be placed into one section,
not necessarily in section Brachypodes, as both species show considerable differences in
the karyotype to other species of this section (Babcock & Jenkins, 1943). Section
Gephyroides, however, is not suited to include C. aurea and C. leontodontoides, as C.
tingitana, whose sister taxon in the molecular phylogeny is C. albida (section Paleya), is the
type species of section Gephyroides and differs morphologically as well as cytologically from
both C. aurea and C. leontodontoides.

Because C. elymaitica (type species of section Paleya) has to be reconsidered as member of

the genus Crepis (unpublished molecular data) the morphologically heterogeneous section
Paleya should not be maintained. The close association of C. albida (section Palyea) to C.
tingitana (sec. Gephyroides) based on molecular data (Enke & Gemeinholzer, 2008) could,
after further morphological investigation, legitimate a transfer of C. albida to section
Gephyroides.

The species not discussed (no revised sectional assignment in table 1) are of unclear
relations within the genus and need further consideration.

5.3 SUMMARY AND CONCLUSIONS

In the present study and with regard to data published in various other studies the integrity of
the genera Lapsana and Rhagadiolus from Crepis s.l. has been demonstrated. No
characters could be found for the species of the five sections Intybellia, Phaecasium,
Lagoseris, Microcephalum, and Pterotheca which would justify an exclusion from Crepis (as
either one or several genera) at this stage. Therefore, the temporary acceptance of Crepis as
paraphyletic taxon and the conservation of Lapsana and Rhagadiolus as separate genera
are suggested.

Some reclassifications of infrageneric groups within Crepis s.str. are proposed: the fusion of
sections Intybellia and Phaecasium and the renaming of the “core group” of section
Phytodesia. Clades I, III, IV and V are found to be largely congruent with taxonomic sections
Mesomeris (Clade I), Hostia and Zacintha (Clade III), and Berinia (Clades IV, V).
Furthermore, it is proposed, to assign the species of Clade VII (the “hybrid complex”) to two
sections Mesophylion and Macropodes. Further genetical and cytological analyses of the

83
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

“hybrid complex” are needed to resolve the relations within the “hybrid complex” as well as to
other species.

The proposed revision of the sectional classification of the genus Crepis should be treated as
provisional, because it almost exclusively considers species included into the molecular
analyses by Enke & Gemeinholzer (2008). Taxonomic consequences can be drawn in future
projects after careful morphological and karyological studies taking more species into
account, to infer the applicability of the here presented suggestions

5.4 LITERATURE
Avery, P. 1930. Cytological Studies of Five Interspecific Hybrids of C. leontodontoides. University of
California Publications in Agricultural Science 6: 135-167.
Babcock, E.B. 1947a. The Genus Crepis I. The Taxonomy, Phylogeny, Distribution and Evolution of
Crepis. University of California Publications 21. University of California Press, Berkeley & Los
Angeles.
Babcock, E.B. 1947b. The Genus Crepis II. Systematic Treatment. University of California
Publications 22. University of California Press, Berkeley & Los Angeles.
Babcock, E.B. & Cave, M.S. 1938. A Study of Intra- and Interspecific Relations of Crepis foetida L.
Zeitschrift für Induktive Abstammungs- und Vererbungslehre 75: 124-160.
Babcock, E.B. & Jenkins, J.A. 1943. Chromosomes and phylogeny in Crepis, III: The relationships of
one hundred and thirteen species. University of California Publications in Agricultural Science
Sci. 18: 241-292.
Bentham, G. & Hooker, J.D. 1873. Genera Plantarum 2. Lowell Reeve, London.
Blackmore, S. 1984. Compositae – Lactuceae. In: Punt, W. & Clarke, G.C.S. (eds.) The Northwest
European Pollen Flora IV. pp. 45-85. Elsevier, Amsterdam.
Bobrov, E.G. & Tzevelev, N.N. 2000. Compositae. Tribe Cichorieae. 691-707. In: Komarov, V.L.
Flora of the USSR vol. 29. Science Publishers, Inc., Enfield.
Bremer, K. 1994. Asteraceae. Cladistics and Classification. Timber Press, Portland, Oregon.
Cassini, H. 1830. Tableau synoptique des Synanthérées. In: Cuvier, F. (ed.), Dictionnaire des
Sciences Naturelles 60: 566-587. Paris.
Collins, J.L. 1924. Inheritance in Crepis capillaris (L.) Wallr., III: Nineteen Morphological and Three
Physiological Characters. University of California Publications in Agricultural Science 2: 305-
320.
Dimitrova, D. & Greilhuber, J. 2001. C-Banding Patterns and Quantitative Karyotype Characteristics
of Bulgarian Species of Crepis (Asteraceae). Plant Biology 3:88-97.
Enke, N., Fuchs, J. & Gemeinholzer, B. subm. Shrinking Genomes? Evidence from Genome Size
Variation in Crepis L. (Compositae).
Enke, N. & Gemeinholzer, B. 2008. Babcock Revisited: New Insights into Generic Delimitation and
Character Evolution in Crepis L. (Compositae: Cichorieae) From ITS and MatK Sequence
Data. Taxon. 57(3): 756-768.

84
 5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

Enke, N., Kilian, N., Nemomissa, S. & Gemeinholzer, B. 2008. Afro-alpine Dianthoseris actually a
congener of Crepis s.str. (Compositae, Cichorieae, Crepidinae). Botanische Jahrbücher 127
(3): 389–405.
Hollingshead, L. 1930a. A Lethal Factor in Crepis Effective Only in an Interspecific Hybrid. Genetics
15: 114-140.
Hollingshead, L. 1930b. Cytological Investigations of Hybrids and Hybrid Derivative of Crepis
capillaris and Crepis tectorum. University of California Publications in Agricultural Science 6:
55-96.
Kamari, G. 1976 Cytotaxonomic Study of the Crepis neglecta complex in Greece. Publication of the
Botanical Institute, University of Patras.

Monnier, A. 1929. Essai Monographique sur les Hieracium 78. Nancy.

Osman, A.K.E. 2006. Pollen Types of the Egyptian Species of Tribe Lactuceae (Subfamily
Cichorioidea-Compositae). Acta Botanica Croatica 65(2): 161-180.
Popov, M.G. 1957-1959. Flora srednej Sibiri. Izdatel'stvo AN SSSR, Moskva.
Sell, P.D. 1976. Crepis. In: Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H.,
Walters, S.M, & Webb, D.A. (eds.) Flora Europaea IV. Pp 344-357.
Siljak-Yakovlev, S. & Cartier, D. 1982. Comparative Analysis of C-Band Karyotypes in Crepis
praemorsa subsp. praemorsa and subsp. dinarica. Plant Systematics and Evolution. 141(2):
85-90.
Tegel, F. 2002. Die Testaepidermis der Lactuceae (Asteraceae) – ihre Diversität und systematische
Bedeutung. Doctorate Thesis, Ludwigs-Maximillians-University of Munich. http://edoc.ub.
unimuenchen.de/archive/00000104/
Tobgy, H.A. 1943. A Cytological Study of Crepis fuliginosa, C. neglecta And Their F1 Hybrid, And Its
Bearing On The Mechanism of Phylogenetic Reduction In Chromosome Number. Journal of
Genetics. 45:67-111.
Verboom, G.A., Linder, H.P., & Stock, W.D. 2004. Testing the adaptive nature of radiation: growth
form and life history divergence in the African grass genus Ehrharta (Poaceae:
Ehrhartoideae). American Journal of Botany 91: 1364-1370.
Zidorn, C. 2008. Sesquiterpene Lactones and Their Precursors as Chemosystematic Markers in the
Tribe Cichorieae of the Asteraceae. Phytochemistry. doi:10.1016/j.phytochem.2008.06.013

85
5. GUIDELINE TO AN INFRAGENERIC CLASSIFICATION

86
 6. DIANTHOSERIS CONGENER OF CREPIS

Chapter 6

Afroalpine Dianthoseris actually a congener of Crepis s.str.

(Compositae,
Cichorieae, Crepidinae)

Neela Enke, Norbert Kilian, Sileshi Nemomissa & Birgit Gemeinholzer

N. Enke, B. Gemeinholzer, Kilian, N., Botanic Garden and Botanical Museum Berlin-Dahlem, Königin-
Luise-Str. 6-8, D-14195 Berlin, Germany; e-mail: n.enke@bgbm.org, b.gemeinholzer@bgbm.org,
n.kilian@bgbm.org;S. Nemomissa Duguma, National Herbarium, Biology Department, Science
Faculty, Addis Ababa University, P.O. Box 3434, Addis Ababa, Ethiopia; e-mail:
nemomssa@bio.aau.edu.et

Published 2008, Botanische Jahrbücher 127(3): 389-405.

ABSTRACT
Enke, N., Kilian, N., Nemomissa, S. & Gemeinholzer, B.: Afro-alpine Dianthoseris actually a
congener of Crepis s.str. (Compositae, Cichorieae, Crepidinae).—Bot. Jahrb. Syst. 127:
389–405. 2008. — ISSN 0006-8152.

The phylogenetic relationship of Dianthoseris is reconstructed with Maximum Parsimony and

Bayesian Inference based on both nrDNA ITS and cpDNA matK datasets. The analyses
revealed that the East African afro-alpine D. schimperi is deeply nested within Crepis s.str.,
forming a clade with the more widespread afro-alpine C. newii. This clade is sister-group to a
clade of species centred in NW Africa/SE Spain, viz. C. hookeriana, C. albida, C. tingitana
and C. oporinoides, and beyond in the Mediterranean and S Europe, viz. C. vesicaria and C.
alpestris. An emended, illustrated description of D. schimperi, for the first time considering
mature achenes, is given, its delimitation from the afro-alpine Ethiopian local endemic
Nannoseris inopinata is re-evaluated, its morphological and karyological affinities within
Crepis are discussed and its distribution is mapped. Two new nomenclatural combinations in
Crepis, C. dianthoseris for D. schimperi and C. inopinata for N. inopinata, are validated.
KEYWORDS: Dianthoseris schimperi, Nannoseris inopinata, Crepis dianthoseris, ITS, matK,
molecular phylogeny, morphology, taxonomy.

87
 6. DIANTHOSERIS CONGENER OF CREPIS

6.1 INTRODUCTION
The acaulescent East African Dianthoseris schimperi A. Rich. was first collected in the alpine
zone of the Ethiopian Simen Mts (Puff & Nemomissa, 2001, 2005) in 1840 by the German
plant collector Wilhelm Georg Schimper (Gillett, 1972) and the name validly published in
1848 for a species in a genus of its own. Since then, it was treated as a distinct genus of the
Compositae tribe Cichorieae. Although Schultz Bipontinus (1866) and Chiovenda (in Pirotta,
1904) suggested its inclusion in the genera Omalocline (currently part of Crepis) and
Sonchus, respectively, this was not followed by others. Besides, Launaea rueppellii (Oliv.
&Hiern) Boulos, an acaulescent rosette herb confined to the upper montane zone of Ethiopia
(Kilian, 1997: 107), was sometimes considered as a close relative of Dianthoseris (Oliver &
Hiern in Oliver, 1877: 456, Hoffmann ,1890–94: 373, Chiovenda in Pirotta, 1903-04: 200). As
a result the taxonomy of D. schimperi remained relatively stable (Fries, 1928: 162; Hedberg,
1957: 252; Jeffrey, 1966: 433; Jeffrey & Beentje, 2000: 92; Tadesse, 2004: 52).

In the 20th century Dianthoseris was associated with the Launaea-Sonchus alliance of
subtribe Crepidinae s.l. by Stebbins (1953: 77), with the Sino-Himalayan Dubyaea-Soroseris
alliance of Crepidinae s.str. by Jeffrey (1966: 430, 433–434, “but the link is by no means a
close one and it remains as an isolated and specialized plant with no close affinities”) and
with Crepidinae s.str. by Bremer (1994: 180) and Lack (2006: 184).

Dianthoseris schimperi was stated by Jeffrey (1988: 434) to have “the very low somatic
chromosome number 4”, citing an unpublished result communicated to him by O. Hedberg.
This number, the lowest within the entire Cichorieae, was subsequently repeated by Jeffrey
& Beentje (2000: 92 as “2n = 4”) and Lack (2006: 185 as “x = 2”), but is obviously erroneous,
because Hedberg’s own publication (Hedberg & Hedberg, 1977: 24), hitherto being the only
report of the chromosome number of D. schimperi, instead gives 2n = 8, counted in two
provenances, from the Simen Mts in Ethiopia and Mt Kenya in Kenya.

In the course of preparing a molecular phylogeny of the tribe Cichorieae, Kilian et al. (2008,
in press) have also considered the position of Dianthoseris. Since the analyses revealed that
Dianthoseris is nested within Crepis s.str., the first author, working on a molecular phylogeny
of Crepis (Enke & Gemeinholzer, 2008, in press), elucidated its affinities within Crepis. The
present study has aimed at (1) clarifying the relationships of D. schimperi within Crepis
based on nrDNA ITS and cpDNA matK datasets, new (micro)morphological and the
published karyological data, (2) re-evaluating the status of Nannoseris inopinata, described
as a related local endemic Ethiopian species but recently considered as conspecific with D.
schimperi, and (3) drawing from these analyses the necessary taxonomic and nomenclatural
conclusions.

88
 6. DIANTHOSERIS CONGENER OF CREPIS

6.2 MATERIAL AND METHODS

Plant material - The investigation is based on field studies by the third author in Ethiopia, on
herbarium material from the herbaria B, CAS, E, ETH, MSB, STU, UPS and US
(abbreviations according to Holmgren & Holmgren, 1998) and high resolution images of
types and other specimens accessed online via Aluka (2007).

Molecular phylogenetic analyses - 27 taxa from 6 genera of subtribe Crepidinae (Crepis,

Taraxacum, Ixeris, Youngia, Lapsana and Soroseris) and one species, Hyoseris radiata, of
subtribe Hyoseridinae were included in the analysis. Sequences for 10 taxa were obtained
from herbarium specimens, the others were downloaded from NCBI (GenBank, EMBL). The
relevant data for the source of the sequences are presented in Table 1.
TABLE 1 (continued next page)
Taxon ITS matK
Crepis albida Vill. subsp. albida EU363606 EU363550
Crepis alpestris Rchb. AJ633373 AJ633153
Iran, Prov. Mazanderan, Distr. Nur,
inter Kamarband et jugum Naftab,
Crepis asadbarensis Bornm. –
3200 m, 8.8.1948, Rechinger 6445
(B), ne025
Crepis chondrilloides Jacq. EU363593 EU363545
Ethiopia, Wollo, Amhara, Mt
Abune Yosef, Peak Guli
Crepis dianthoseris N. Kilian & al. ≡ Dianthoseris id.
Bamba (4284 m), 4000-4200
schimperi Sch. Bip.
m, 27.11.2001, Ortiz & Vivero
26 (ETH 74116), NK155
Afghanistan, Prov. Baghlan, Surkhab-
C. foetida subsp. afghanistanica Babc. ≡ Crepis Tal, 6 km NO Dahane Eshpushta,
EU363604
trichocephala (Krash.) V.V. Nikitin 1300 m, Podlech 18361 (MSB
01615), ne122
Crepis foetida L. subsp. foetida EU363619 EU363556
Crepis hierosolymitana Boiss. EU363602 EU363543
Crepis hookeriana Ball EU363605 EU363549
Crepis hypochaeridea Thell. EU363617 –
Crepis incarnata Tausch EU363608 AJ633151
Crepis macropus Boiss. & Heldr. EU363589 EU363577
Crepis mollis Asch. AJ633380 EU363538
Crepis neglecta L. subsp. neglecta EU363610 EU363553
id., ne032;
Tanzania, Arusha region, Mt Cameroon, Mt. Cameroon, above
Crepis newii subsp. oliveriana (Kuntze) C. Jeffrey & Meru, c. 3450 m, 4.9.1966, Buea, near upper forest limit above
Beentje ≡ C. oliveriana (Kuntze) C. Jeffrey Stein W. Bie (UPS) [2n = 8] , Mann's spring; Lat/long; 04:10N,
ne032 09:13E, Alt: 2200-2300m, Hedberg
(UPS), ne036
Crepis oporinoides Boiss. EU363633 EU363567
West slope of Lalaat Musa. 7800-
Crepis robertioides Boiss. – 8000ft, Davies 9761 (E 00228145),
ne259
Crepis sahendi Boiss. & Buhse EU363651 –

89
 6. DIANTHOSERIS CONGENER OF CREPIS

TABLE 1 (CONTINUED). List of the taxa and the source of the sequences used in the molecular phylogenetic
analyses. Accession numbers only are given for the sequences downloaded from GenBank/EMBL, the source
of the plant material and the accession number are cited for original sequences of the present study.

Taxon ITS matK

Crepis sancta (L.) Babc. AJ633372 AJ633150
Crepis sibthorpiana Boiss. & Heldr. EU363648 EU363574
Crepis tectorum L. EU363643 EU363536
Crepis tingitana Ball EU363586 AJ633149
Crepis vesicaria subsp. stellata (Ball) Babc. EU363630 –
Crepis vesicaria L. subsp. vesicaria – EU363565
Crepis zacintha (L.) Babc. EU363655 EU363579
Hyoseris radiata L. AJ633299 AJ633215
Ixeris chinensis (Thunb.) Nakai EU363587 EU363539

China,Yunnan Province, Dali Xian, E.

side of Diancang Shan mountain
range. Viciinity of Butterfly Springs.
Ixeris repens A. Gray – Badly disturbed hillsides &
abandoned paddy fields 2550 m
25°55' N 100°05' E , Bartholomew
6192 (US 3043724), DB243
Ixeris stolonifera A. Gray AJ633284 AJ633156
Lapsana communis L. AJ633285 AJ633138, AJ633138
China, Gansu, Qilian Mt, 3400- China, Qinghai, Qilian Mt, 3400-3600
Soroseris erysimoides (Hand.-Mazz.) C. Shih 3800 m, 27.7.1991, T.N. Ho m, 4.7.1991, T.N. Ho 1641 (CAS
2812 (CAS 795834), NK160 791443), NK159
China, Qinghai, Bayan Har pass,
Soroseris glomerata (Decne.) Stebbins EU363656 4700 m, 12.8.1996, T. N. Ho 1692
(CAS 939054), NK162
China, Qinghai, 3850 m,
Soroseris trichocarpa (Franch.) C. Shih 14.8.1996, T.N. Ho & al. 1787 –
(CAS 939119), NK175
Taraxacum bessarabicum Fisch. AJ633287 –
Armenia, Mt Teghenis, 40 32'N, 44
Taraxacum crepidiforme DC. – 40'E, 2300 m, 16.6.2002, C.
Oberprieler CHO 10074 (B), NK084
Taraxacum erythrospermum Besser AJ633291 –
Youngia denticulata (Houtt.) DC. AJ633293 AJ633139
Youngia japonica (L.) DC. AJ633294 –

Total genomic DNA was isolated from 24 mg per sample of herbarium leaf material, which
was ground down. DNA was then extracted according to standard protocol using Quiagen
DNeasy Plant Mini Kit. Both ITS and matK fragments were amplified in two overlapping parts
using the primers ITS-A and ITS-C (Blattner, 1999) for ITS 1, ITS2-D, ITS-B (Blattner, 1999)
for ITS 2 and trnK–710f (Johnson & Soltis, 1995), matK-iR (Fehrer et al., 2007), matK-ifN
and matK-rN (Enke & Gemeinholzer, 2008) for matK. During the Polymerase Chain Reaction
(PCR) the following protocol was used (ITS/matK): initial denaturation 2 min/2 min at 94 °C,
denaturation 1 min/1 min 30 sec at 94 °C, annealing 1 min at 66 °C/ 2 min at 62 °C,

90
 6. DIANTHOSERIS CONGENER OF CREPIS

FIG. 1 : Phylogram derived from ITS sequence data by Bayesian Analysis (50% majority rule). Bayesian
Posterior Probabilities (>0.90, PP) and bootstrap values (>80, BS) are given above branches. Former
Dianthoseris schimperi ≡ Crepis dianthoseris highlighted. A and B indicating nodes of interest.

elongation 2 min/ 3 min at 72 °C (38 cycles) and final extension 10 min /10 min at 72°C. PCR
was carried out with a reaction volume of 11.5 µl core mix plus 1 µlDNAsolution (1:10, 1:50
or 1:100 dilution depending on usability). The reaction volume contained 6.73 µl ddH2O, 2.5
µl 5x buffer (Finnzymes), 1.25 µl dNTP’s (Fermentas), 0.4 µl BSA (Finnzymes), 0.25 µl of
each primer (10 pmol/µl) and 0.12 µl 1u Phusion High FidelityDNAPolymerase (Finnzymes).
The PCR products were purified with MSB Spin PCRapace (Invitek) and subsequently sent
to StarSEQ (Mainz, Germany) for sequencing.

91
 6. DIANTHOSERIS CONGENER OF CREPIS

The sequences were edited in ChromasLite2000 (Technelysium Pty. Ltd., Helensvale,

Australia) and aligned by hand using BioEdit (Hall, 1999) following Goertzen et al. (2003) for
ITS gap-coding. For matK an indel at position 118–122 has been coded as one mutational
step. The alignments are available from the first author upon request.

The phylogenetic relationships from both ITS and matK datasets were reconstructed with
Maximum Parsimony (MP), using PAUP* 4.0b10 (Swofford, 2002), and with Bayesian
Inference (BI), using Mr Bayes 3.1.2. (Ronquist & Huelsenbeck, 2003). For the BI analyses a
gamma distribution rate variation among sites and 10 million generations of the MCMC
chains in two independent runs were used, trees being saved every 100 generations. The
first 25 000 trees were discarded as burn-in for the analysis then reached stationarity. All
remaining trees sampled were used to calculate a 50 % majority rule consensus tree. For the
MP analyses heuristic searches were conducted in PAUP 4.0b10 with equal weights, 1000
closest sequence additions and tree bisection-reconnection (TBR) branch swapping,
permitting 10 trees to be held at each step. An evaluation of the trees was performed by
using bootstrap analysis with 1000 replicates, equal weights,TBR swapping, MulTrees option
in effect and 10 trees held at each step. Trees were drawn using TreeView (Page, 1996) and
Adobe Illustrator (Adobe Systems, Inc., San Jose, California, USA).

Micromorphology - Flowers taken from herbarium material were shortly boiled in water,
then spread on a microscope slide and photographed with aLEICADFC290 digital camera
mounted on a WILDM5 optical reflected-light microscope (up to × 40 magnification). Achenes
and pappus were mounted onto SEM stubs on double-sided sticky tape, coated with 20 nm
Au-Pd using an Emitech K550 sputter-coater, examined using a Philips SEM 515 scanning
electron microscope and documented with the Point Electronic WinDISS III digital imaging
device (hard- and software).

6.3 RESULTS
Molecular Phylogeny

A total of 33/33 (ITS/matK) sequences were obtained from 22/22 Crepis species and 10/8
other taxa of the Crepidinae. Hyoseris (Hyoseridinae) was chosen as outgroup taxon. Minor
variation in taxon sampling for ITS and matK are caused by differences in sequence
availability. In total 683/974 (ITS/matK) characters were included in the Maximum Parsimony
analysis, of which 198/75 were parsimony informative (29/7.7 %).

The trees inferred from the nuclear (Fig. 1) and plastid (Fig. 2) markers are similar in their
topology, but the ITS tree is better resolved in the apical branches.

92
 6. DIANTHOSERIS CONGENER OF CREPIS

In both trees Dianthoseris schimperi is nested deeply within Crepis s.str. and placed far away
from Soroseris, to which an affinity had been hypothesized by Jeffrey (1966). Furthermore,
Dianthoseris schimperi forms a clade with Crepis newii subsp. oliveriana (Kuntze) C. Jeffrey
& Beentje (Fig. 1–2, clade A), which is only in the matK tree statistically supported by
Bayesian posterior probabilities (1.00) and bootstrap values (98). This clade A is part of the
larger clade B (Fig. 1–2, B) in both the nuclear and the chloroplast trees, which comprises
the species C. vesicaria L., C. hookeriana Ball, C. tingitana Ball (ITS tree only), C. alpestris
Rchb., C. oporinoides Boiss. and C. albida Vill. Clade B is statistically supported only in the
ITS tree by Bayesian posterior probability (0.96). In the ITS tree the clade comprising C.
hookeriana and C. vesicaria is sister-group to Dianthoseris and C. newii subsp. oliveriana
(Bayesian posterior probability 1.00), while in the matK tree the relations between these taxa
are unresolved.

Morphology

Description - Acaulescent perennial rosette herb with thick, deepseated, lignified taproot
and usually unbranched caudex. Leaves in a basal rosette appressed to the ground,
glabrous, (narrowly) lanceolate to oblanceolate, 2–4(–7) × 0.2–1.2(–2) cm, with a
conspicuously broad and pale midrib, tapering to a narrow base, apex obtuse and
mucronulate, margin entire to remotely sinuatedentate. Capitula usually solitary, sessile, with
up to approximately 100 flowers. Involucre 11–16(–18) mm long, involucral bracts in several
rows, abaxially and adaxially entirely glabrous; inner involucral bracts subequal, linear-
lanceolate, 10–12 × 2.5–3 mm, margins distally shortly ciliolate; outer involucral bracts
similar to the inner in shape and size to even larger and grading in size and shape into the
innermost rosette leaves. Flowers (fully developed marginal ones) with a 4.5–7 mm long
corolla tube and a ligule varying considerably in size between different plants, of 1.5–9 × 0.5–
1.8 mm, therefore being either distinctly longer (Fig. 3E) to shorter than the tube; anthertube
without appendages 1.2–2.5 mm long; basal appendages up to 0.5 mm, apical appendages
up to 0.3 mm long. Achenes (fully mature ones, Fig. 3A, C-D) 3.2–3.8 mm long, 0.8–1 mm in
diameter, subspherical in cross-shape and apically less attenuate than basally, glabrous,
with c. 20 longitudinal ribs, 5 of which being more prominent than the others, achene surface
otherwise smooth. Pappus 6–10 mm long, white, of 2(–3) distinct, basally connected series
of smooth to weakly barbellulate setae of equal length; setae of the outer 1(–2) series clearly
thicker than those of the inner one and curved outwards at maturity, the setae of the inner
series remaining straight (Fig. 3B).

93
 6. DIANTHOSERIS CONGENER OF CREPIS

Fig. 2. Phylogram derived from matK sequence data by Bayesian Analysis (50% majority rule). Bayesian
Posterior Probabilities (>0.90, PP) and bootstrap values (>80, BS) are given above branches. Former
Dianthoseris schimperi ≡ Crepis dianthoseris highlighted. A and B indicating nodes of interest.

Notes - The taproot of Dianthoseris schimperi seems to bear adventitious buds, which lead
to a “branching” of the taproot and the growth of new rosette shoots. Mature achenes have
not been available to the authors of previously published descriptions, which explains the
lower achene size of 2.5 mm given by them (Jeffrey & Beentje, 2000; Tadesse, 2004). We
found that the achene length is almost 4 mm in the mature fruits of Ortiz & Vivero 26 (ETH).

94
 6. DIANTHOSERIS CONGENER OF CREPIS

Delimitation - An acaulescent species habitually very similar to Dianthoseris schimperi was

collected in the late 1960s from one locality in the Simen Mts (Sebald, 1968) and described
as Nannoseris inopinata Cufod. (Cufodontis, 1968). However, Tadesse (1999: 29, 2004: 52)
treated N. inopinata as conspecific with D. schimperi. An inspection of the holotype (see
below, Taxonomy, Note 2) as well as of a second specimen referable to this taxon with
almost mature fruits from the same locality (Begemdir, prov. Simien, 13°15–16’N, 38°3–13’E,
Geech, 3600 m, thin soil with short grass, 15.10.1973, O. Hedberg & G. Aweke 5360,
ETH37263) revealed, however, that N. inopinata should be maintained as a separate species
allied to D. schimperi. The differences between the two species in the leaves (for N.
inopinata see Cufodontis (1968: fig. 5 [habit; b/w photograph of herbarium specimen]) and
the indumentum of the involucre are conspicuous, as Dianthoseris schimperi is very uniform
in appearance across its disjunct distribution area, as already noted by Fries (1928: 162).
This, however, does not hold true for the corolla size in D. schimperi. Whereas Sebald
1247,STU(Fig. 3E) has large corollas with a tube of 6–7 mm and ligule of 8.5–9 mm, and
thus with a ligule longer than the tube, Hedberg 5543, ETH, has a tube of c. 4.5 mm and a
ligule of 3.5–4 mm, thus a ligule shorter than the tube. Still smaller corollas with a tube of c.
4.5 and a ligule of only 1.5 mm length are reported from the Kenya/Uganda/Tanzania
subarea (Jeffrey& Beentje, 2000: 92). In N. inopinata (Fig. 3F) the corolla tube is always
longer than in D. schimperi, while the ligule length is within its wide range of variation in D.
schimperi. The involucre (apart from the indumentum), the achenes and the pappus are
rather similar in both species, the latter with 10–12 mm, however, a little longer.

6.4 DISCUSSION
Molecular phylogeny - According to our molecular phylogenetic analyses, Dianthoseris
schimperi has to be considered as a member of Crepis, having its closest affinities to the
widespread and polymorphic, afro-montane to afro-alpine C. newii Oliv. & Hiern, of which
subsp. oliveriana was included in our analyses. Treated by Jeffrey & Beentje (2000: 70) as a
subspecies, the taxon was considered by Babcock (1947) as a separate species, C.
oliveriana, and like C. newii as a member of C. sect. Anisorhamphus. This is one of the two
largest sections of the genus. Except for its sole non-African member C. alpestris, the section
comprises the vast majority of the tropical African species. The inclusion of C. alpestris in the
larger clade B in both the nuclear and the chloroplast trees confirms its moderate affinity to
C. newii subsp. oliveriana and D. schimperi. In contrast, the third and only other member of
C. sect. Anisorhamphus included in our analyses, C. hypochaeridea (ITS tree only, see Fig.
1), is a sister-group to a clade consisting of three Mediterranean species of Babcock’s C.
sect. Berinia, i.e. C. macropus Boiss. & Heldr., C. sibthorpiana Boiss. & Heldr. and C.
chondrilloides Jacq. This indicates that Babcock’s section Anisorhamphus is polyphyletic,

95
 6. DIANTHOSERIS CONGENER OF CREPIS

and a similar result for other sections of Crepis was reported by Enke & Gemeinholzer
(2008). The remaining six members of clade B are heterogenous in terms of their previous
sectional classification, belonging to six different sections in Babcock’s system of Crepis: C.
oporinoiodes Boiss. from SE Spain belongs to the aforementioned section Berinia and C.
hookeriana Ball from Morocco belongs to section Macropodes, which, besides others,
comprises two endemics (C. xylorrhiza Sch. Bip. and C. tenerrima (Sch. Bip.) R. E. Fr.) and
one more widespread species (C. rueppellii Sch. Bip. [Syn.: C. abyssinica Sch. Bip., C.
forsskalii Babc.]) of the Abyssinian Highlands not available for our analyses. C. tingitana Ball
(ITS tree only, see Fig. 1) from N Morocco/SE Spain belongs to section Gephyroides and C.
albida Vill. from E Spain to section Paleya, which also includes C. achyrophoroides from the
Abyssinian Highlands also not available for our analyses.

Finally the W European-W Mediterranean C. vesicaria L. belongs to section Lepidoseris.

Since the relationships within Crepis are insufficiently known, according to the available
molecular phylogenetic results, and since a complete revision of the taxonomy of Crepis is
needed, some limitations to the analysis of the systematic position of Dianthoseris within
Crepis exist. Our molecular analyses nevertheless give two straightforward hints: (1)
Dianthoseris schimperi has close affinities to the afromontane-afroalpine C. newii group and
(2) the C. newii group seems to be allied to species centred in NW African/SE Spain, such as
C. hookeriana, C. albida, C. tingitana and C. oporinoides, and beyond in the Mediterranean
and S Europe to species such as C. vesicaria and C. alpestris.

Chromosome numbers - Dianthoseris schimperi shares its chromosome number of n=4

with all species of clade B except C. albida and C. tingitana. These latter two species have n
= 5 (for references see Watanabe 2008). The latter taxa are the sister-group to the remainder
of clade B in the ITS tree (Fig. 1). A decrease in chromosome number is considered an
evolutionary trend in Crepis (Babcock 1947). This has been confirmed by the molecular
phylogenetic analyses of Crepis (Enke in prep.) and the Cichorieae in general (Kilian et al.
2008).

Morphology - The glabrous achenes with c. 20 ribs, the 2–3-seriate pappus and the
capitulum size of Dianthoseris schimperi perfectly match Crepis. The involucre, however,
appears rather odd. The complete absence of an indumentum on the involucral bracts is a
rare feature in Crepis but, e.g., also found in C. pulchra. More peculiar is, however, the fact
that the outer bracts are not distinctly smaller than the inner series of bracts as is usual in
Crepis, but are of equal length even at maturity instead. This has, however, to be considered
in the context of the fully acaulescent habit of the species. Whereas in caulescent species
the outer involucral bracts usually grade in shape und size into the usually small bracts of the
capituliferous axis, the capitulum in D. schimperi immediately terminates the rosette shoot
and the rosette leaves next to the involucre are intermediate in size and shape between the
outer involucral bracts and the rosette leaves further outside. The peculiarities of the

96
 6. DIANTHOSERIS CONGENER OF CREPIS

FIG. 3. A-E: Crepis dianthoseris (≡ Dianthoseris schimperi) – A: achene (pappus removed), overview; B:
pappus with the 1(-2) outer series of setae curved outwards and the inner, somewhat thinner, series of
setae being erect; C: achene base; D: achene apex (pappus removed); E: marginal flower; A-D from Ortiz &
Vivero 26, ETH, E from Sebald 1247, STU. – F: Crepis inopinata (≡ Nannoseris inopinata), marginal flower
at the same stage of anthesis as the one in E; from the holotype, Sebald 1046, STU. – Scale bars: A = 0.9
mm, B = 0.5 mm, C-D = 0.2 mm, E-F = 2 mm.

involucre of D. schimperi may thus be interpreted as a consequence of the acaulescent

growth form. The acaulescent growth form and the involucre of Nannoseris inopinata are
essentially similar, apart from minor differences in the shape of the involucral bracts and the

97
 6. DIANTHOSERIS CONGENER OF CREPIS

fact that the inner bracts are densely glandular-hairy. The acaulescent growth form as
present in both species is otherwise without parallel in Crepis, which supports the hypothesis
that the two species are closely related.

According to Blackmore & Persson (1996) Dianthoseris has subechinolophate pollen grains
with rudimentary paraporal lacunae and rounded abporal lacunae, which is more similar to
Dubyaea and Soroseris than to Crepis, where echinolophate pollen grains with tricolporate,
ectocolpi divided into three lacunae, and somewhat angular, large abporal lacunae
predominate. However, missing congruence of pollen features with taxonomic entities
recognized by molecular and other morphological characters are rather frequent and make
pollen features often difficult to interpret.

Table 2. Differential characters between Nannoseris Regarding the systematic position of

inopinata and Dianthoseris schimperi. Further Dianthoseris schimperi within Crepis, the
explanations are given in the text.
morphological characters are inconclusive,
Characters N. D. shimperi even considering the species indicated by
inopiniata
margin entire to the molecular results as next allies. This is,
distinctly
Leaves remotely
runcinate
sinuate-dentate however, not surprising, if we take into
abaxially
densely
consideration (1) the poverty of features in
covered with our habitually much reduced species, and
Inner involucral bracts long, pale
entirely glabrous
glandular (2) that among the closest allies of D.
trichomes
Corolla of fully schimperi are members of no less than
developed marginal
flowers
seven of Babcock’s morphology-based
tube length [mm] 7.5-9 4.5-7 sections.
Ligule size [mm] 5-5.5 x c.1.2 1.5-9 x 0.5-1.8
Pappus length [mm] 10-12 6-10

6.5 TAXONOMY
Crepis L. = Dianthoseris [Sch. Bip. in Flora 25: 439. 1842, nom. inval., ex] A. Rich., Tent. Fl.
Abyss. 1: 468. 1848 = Omalocline subg. Dianthoseris (Sch. Bip.) Sch. Bip. in Jahresber.
Pollichia 22–24: 321. 1866 = Nannoseris Hedberg in Symb. Bot. Upsal. 15: 251. 1957, nom.
illeg. — Typus: Dianthoseris schimperi A. Rich.

Crepis dianthoseris N. Kilian, Enke, Sileshi & Gemeinholzer, nom. nov. = Dianthoseris
schimperi [Sch. Bip. in Flora 25: 439. 1842, nom. inval., ex] A. Rich., Tent. Fl. Abyss. 1: 468.
1848 [non Crepis schimperi (A. Rich.) Schweinf., Beitr. Fl. Aethiop.: 144. 1867, i.e. Crepis
foetida L.] = Omalocline schimperi (A. Rich.) Sch. Bip. in Jahresber. Pollichia 22–24: 321.
1866 [& Schweinf., Beitr. Fl. Aethiop.: 285, 307. 1867 as “Homalocline schimperi”, orth. var.]
= Sonchus dianthoseris [“Sonchus dianthoseris var. schimperi”] Chiov. in Annuario R. Ist.
Bot. Roma 8 [Pirotta, Fl. Eritrea]: 200. 1904, nom. illeg. = Nannoseris schimperi (A. Rich.)
Hedberg in Symb. Bot. Upsal. 15: 251. 1957.—Holotype: [Ethiopia], “in regione superiori

98
 6. DIANTHOSERIS CONGENER OF CREPIS

montis Bachit [Bwahit, 13°13’N, 38°13’E] 12000–13000 pedes supra”, 19.8.1840, Schimper
755 [erroneously quoted as “775” in the protologue] (P015825!; isotypes: BM000924929!,
BR0000008361790!, K000251829!, K000251831!, K000251832!, M0105528!, NY00167853!,
P015826!, P015827!, see Aluka 2007).

Crepis nivalis Schweinf. & Asch. in Schweinfurth, Fl. Aethiop.: 284. 1867, nom. nud. (fide
Babcock 1947: 924).

Notes - (1) For the invalid publication of the name Dianthoseris by Schultz-Bipontinus, its
later validation by Richards and the resulting illegitimacy of Hedberg’s generic name
Nannoseris see Jeffrey (1966: 433).

(2) The epithet of Dianthoseris schimperi is not available for the corresponding combination
in Crepis, because it would result in a later homonym of Crepis schimperi (A. Rich.)
Schweinf., which is treated as a taxonomic synonym of C. foetida L. by Jeffrey (1966: 462)
and as a species closely related to the latter by Babcock (1947: 705). If Nannoseris inopinata
were actually conspecific with Dianthoseris schimperi, as is treated by Tadesse (1999: 29),
the epithet inopinata would have to be taken up for the Dianthoseris species in Crepis.
However, as shown above, N. inopinata is a different, although closely related species:
Crepis inopinata (Cufod.) N. Kilian, Enke, Sileshi & Gemeinholzer, comb. nov. = Nannoseris
inopinata Cufod. in Stuttgarter Beitr. Naturk. 195: 7. 1968.—Holotype: [Ethiopia],
“Hochsemyen, beim Lagerplatz Kurbät-Mätaya zwischen Amba Ras und Buahit, 13°15.5’N,
38°11.4’E, 3600 m, in beweidetem Grasland zwischen Erica arborea-Büschen”, 9.11.1966,
O. Sebald 1046 (STU 000506, see Aluka 2007).

The binomial Sonchus dianthoseris Chiov. of 1904, to which that author designated “var. a
Schimperi = Dianthoseris schimperi Schultz Bip. ....” as “la forma typica” (constituting definite
indication of a type of that binomial according Art. 7.5 of the Code, McNeill et al., 2006) and
in which he included as a second variety Dianthoseris rueppellii (see 3, below) was correctly
qualified as illegitimate by Jeffrey (1966: 433): Chiovenda should have adopted the epithet
rueppellii of the latter included species after the epithet schimperi was not available in
Sonchus because of S. schimperi A. Braun & Bouche of 1857 (a synonym of S. oleraceus,
see Boulos 1973: 155). The use of the epithet dianthoseris in Crepis is, however, sanctioned
by Art 58.1 of the Code (McNeill et al., 2006).

(3) The second similarly acaulescent species formerly placed in Dianthoseris is actually a
member of Launaea (subtribe Hyoseridinae): D. rueppellii [Sch. Bip. in Flora 25: 440. 1842,
nom. inval. ex] Oliv. & Hiern in Oliver, Fl. Trop. Afr. 3: 456. 1877 = Sonchus dianthoseris var.
rueppellii Chiov. in Annuario Ist. Bot. Roma 8 [Pirotta, Fl. Eritrea]: 200. 1904, nom. illeg. =
Sonchus rueppellii (Oliv.&Hiern) R. E. Fr. in Acta Horti Berg. 8: 112. 1925 = Launaea
rueppellii (Oliv. & Hiern) Boulos (see Jeffrey 1966: 467, Kilian 1997: 104).

99
 6. DIANTHOSERIS CONGENER OF CREPIS

Ic. - Fig. 3A-E; Fries 1928: t. 10,4 (habit; b/w photograph of herbarium specimen); Hedberg
1957: 252, fig. 18 = Tadesse 2004: 52, fig. 36 (habit and single flower; drawings); Jeffrey &
Beentje 2000: 91, fig. 24 (habit & details; drawings); Puff & Nemomissa 2005: 155, fig. D55A-
C (habit, flowering, dug out plant showing taproot; colour photographs);

FIG. 4. Distribution of Crepis dianthoseris (≡ Dianthoseris schimperi) (circle). – Georeferenced map based on the
specimens seen and literature data and generated with DIVA-GIS (Hijmans & al. 2005) using an adaptation of
the SRTM 90 m digital elevation data (CGIAR-CSI 2004).

Distribution and ecology - Crepis dianthoseris occurs in the Eastern African countries
Ethiopia (Gonder, Gojam & Bale region; Tadesse, 1999: 30, 2004: 52), Kenya, Uganda,
Tanzania (Jeffrey, 1966: 433;Jeffrey & Beentje, 2000: 92), where it is restricted to elevations
between 3600 and 4500m (–5800 m, on Mt Kilimanjaro, see Hedberg, 1970; up to c.
4200min Ethiopia, see Sebald, 1968: 31), see Fig. 4. It grows on moist ground in open
sparse vegetation, such as afro-alpine grasland, Helichrysum heath and open Erica arborea

100
 6. DIANTHOSERIS CONGENER OF CREPIS

scrub, especially on solifluction soils and in rock crevices (Jeffrey & Beentje, 2000: 92;
Tadesse, 2004: 52; Puff & Nemomissa, 2005: 154; Sebald, 1968: 31; label data).

Flowering plants are found almost throughout the year; fruiting specimens are, however, very
rare in herbaria.

Additional specimens seen - Ethiopia: Bale: Bale Mts, National Park, 6°51’–7°10’N, 39°41–
48’E, at Garba Goracha camp, 3950 m, on ground disturbed by Giant Mole Rat, 30.10.1973,
O. Hedberg 5543 (ETH 36765); Bale Mts, Wasama, 6°55’N, 39°46’E, 4120 m, afro-alpine
Helichrysum heath, 12.1.1990, G. & S. Miehe 958 (ETH 36753); Bale Mts, Chuofo Hadji
Biftu, 6°53’N, 39°46’E, 4200 m, afro-alpine Helichrysum heath, open substratum, 18.1.1990,
G. & H. Miehe 1175 (ETH 36760). — Gonder: Begemdir, prov. Simien, 13°15–16’N, 38°3–
13’E, on the crest, 4225 m, on solifluction terraces between boulders, 18.10.1973, O.
Hedberg & G. Aweke 5436 (ETH 36752); Begemder, Hochsemyen, am Südwesthang des
Kiddis Ared im alpinen Grasland, 4200 m, 14.11.1966, O. Sebald 1247 (STU); c. 42 km from
Debra Tabor, eastern ascent of Mt Guna, 11°45’N, 38°15’E, 3800 m, alpine meadow, moist
area, 6.10.1981, C. Puff, D. Mantelli & E. Kelbessa 8110006–1/5 (ETH 36755); Amhara,
Peak Guli Bamba (4284 m), afro-alpine belt close to the peak, boulder, rocky slopes,
montane grassland with Lobelia rhynchopetalum, 4000–4200 m, 27.11.2001, S. Ortiz & J. L.
Vivero 26 (ETH 74116).

6.6 ACKNOWLEDGEMENTS
We thank the herbaria of CAS, E, MSB, STU, UPS and US for the loan of specimens,
Prof.Werner Greuter for nomenclatural advice, Mrs Monika Lüchow (light and scanning
electron microscopy) and Mrs Christa Menz (image processing) for excellent technical
assistance, and the referces Henk Beentje (Kew) and Tod Stuessy (Vienna) for valuable
comments on an earlier version of this paper.

6.7 REFERENCES
Aluka 2007-continued: African plants. Published on the Internet: http://www.aluka.org/
Babcock, E. B. 1947: The genus Crepis 1-2. University of California Publications in Botany 21, 22.
Blackmore, S. & Persson, V. 1996: Palynology and systematics of the Crepidinae (Compositae:
Lactuceae). – Pp. 111–122 in: Hind, D. J. N. & Beenthe, H. J. (ed.), Compositae: Systematics.
Proceedings of the International Compositae Conference, Kew, 1994, 1, Kew.
Blattner, F.R. 1999: Direct amplification of the entire ITS region from poorly preserved plant material
using recombinant PCR. BioTechniques 27: 1180-1186.
Boulos, L. 1973: Révision systématique du genre Sonchus L. s.l. IV. Sous-genre 1. Sonchus.
Botanical Notes 126: 155-196.
Bremer, K. 1994: Asteraceae. Cladistics & classification. Portland.

101
 6. DIANTHOSERIS CONGENER OF CREPIS

CGIAR-CSI [Consortium of Spatial Information] 2004: NASA Shuttle Radar Topographic Mission
(SRTM) 90 m digital elevation data (DEMs). Published on the internet http://srtm.sci.cgiar.org.
Cufodontis, G. 1968: Crassulaceae, Celastraceae, Thymelaeaceae und Compositae aus dem
Tanasee-Gebiet und dem Semyen-Gebirge (Äthiopien). Ergebnisse der botanischen Reise
Oskar Sebald im Jahre 1966 nach Äthiopien, Nr. 2. Stuttgarter Beiträge zur Naturkunde 195.
Enke, N. & Gemeinholzer, B. 2008: Babcock revisited: New insights into generic delimitation and
character evolution in Crepis L. (Compositae: Cichorieae) from ITS and matK sequence data.
Taxon (in press).
Fries, R. E. 1928: Zur Kenntnis der Compositae des tropischen Ostafrikas. Acta Horti Bergiana 9:
109-164, t. 1-10.
Gillett, J. B. 1972: W. G. Schimper's botanical collections localities in Ethiopia. Kew Bulletin 27: 115-
128.
Goertzen, L.R., Cannone, J.J., Gutell, R.R. & Jansen, R.K.. 2003: ITS secondary structure derived
from comparative analysis: implications for sequence alignment and phylogeny of the
Asteraceae. Molecular Phylogeny and Evolution 29: 216-234.
Hall, T.A. 1999: BioEdit: a user-friendly biological sequence alignment editor and analysis program for
Windows 95/98/NT. Nucleic Acids Symposium Series. 41: 95-98.
Hedberg, I. & Hedberg, O. 1977: Chromosome numbers of afroalpine and afromontane angiosperms.
Botanical Notes 130: 1-24.
Hedberg, O. 1957: Afroalpine vascular plants. A taxonomic revision. Symbolae Botanicae
Upsalienses 15.
Hedberg, O. 1970: Evolution of the afroalpine flora. Biotropica 2(1): 16-23.
Hijmans, R., Guarino, L., Mathur, P. & Jarvis, A. 2005: DIVA-GIS Version 5.2. – Published on the
Internet http://diva-gis.org.
Hoffmann, O. 1890-94. Compositae. Pp. 87–391 in: Engler, A. & Prantl, K. (ed.), Die natürlichen
Pflanzenfamilien 4(5). Leipzig.
Holmgren, P. K. & Holmgren, N. H. 1998-continued: Index herbariorum: a global directory of public
herbaria and associated staff. – Published on the internet at http://sweetgum.nybg.org/ih/
Jeffrey, C. 1966: Notes on Compositae: I. The Cichorieae in East Tropical Africa. Kew Bulletin 18:
427-486.
Jeffrey, C. & Beentje, H. J. 2000: Cichorieae. Pp. 63-108 in: Beentje, H. J. (ed.), Flora of tropical East
Africa. Compositae (Part 1). Rotterdam.
Johnson, L.A. & Soltis, D.E. 1995: Phylogenetic inference in Saxifragaceae sensu stricto and Gilia
(Polemoniaceae) using matK sequences. Annals of the Missouri Botanical Garden 82: 149-175.
Kilian, N. 1997. Revision of Launaea Cass. (Compositae, Lactuceae, Sonchinae). Englera 17.
Kilian, N., Gemeinholzer, B. & Lack, H. W. 2008 (in press): Tribe Cichorieae. In: Funk, V., Susanna,
A., Stuessy, T. & Bayer, R. (ed.), Systematics and evolution of the Compositae. Vienna.
Lack, H. W. 2006. Tribe Cichorieae Lam. & DC. Pp. 180-199 in: Kadereit, J. W. & Jeffrey, C. (ed.),
The families and genera of vascular plants 8. Flowering plants. Eudicots. Asterales. Berlin, etc.
McNeill, J. , Barrie, F. R., Burdet, H. M., Demoulin, V., Hawksworth, D. L, Marhold, K., Nicolson,
D. H., Prado, J., Silva, P. C., Skog, J. E., Wiersema, J. H., Turland, N. J. (ed. & compilers)
2006: International Code of Botanical Nomenclature (Vienna Code) adopted by the Seventeenth
International Botanical Congress Vienna, Austria, July 2005. Regnum Vegetabile 146.
Page, R.D.M. 1996: TREEVIEW: an application to display phylogenetic trees on personal computers.
Computer Applications in the Biosciences 12: 357-358.

102
 6. DIANTHOSERIS CONGENER OF CREPIS

Pirotta, P. P. 1903-04: Flora della colonia Eritrea. Roma.

Puff, C. & Nemomissa, S. 2001: The Simen Mountains (Ethiopia): comments on plant biodiversity,
endemism, phytogeographical affinities and historical aspects. Systematics and Geography of
Plants 71: 975-991.
Puff, C. & Nemomissa, S. 2005: Plants of the Simen: a flora of the Simen Mountains and
surroundings, northern Ethiopia. – Scripta Botanica Belgique. 37.
Ronquist, F. & Huelsenbeck, J.P. 2003: MrBayes 3: Bayesian phylogenetic inference under mixed
models. Bioinformatics 19: 1572-1574.
Schulz Bipontinus, C. H. 1866: Beitrag zum System der Cichorieen. Jahresberichte Pollichia 22-24:
296-322.
Sebald, O. 1968: Bericht über botanische Studien und Sammlungen am Tana-See und im Semyen-
Gebirge (Äthiopien). Ergebnisse der botanischen Reise Oskar Sebald im Jahre 1966 nach
Äthiopien, Nr. 2. Stuttgarter Beiträge zur Naturkunde 194.
Stebbins, G. L. 1953. A new classification of the tribe Cichorieae, family Compositae. Madroño 12:
65–81.
Swofford, D.L. 2002: PAUP*: Phylogenetic analyses using parsimony (* and other methods). 4.0
Beta. Sunderland, MA.
Tadesse, M. 1999: New combinations, varieties and synonyms in African Compositae. Compositae
Newsletter 33: 23-32.
Tadesse, M. 2004: Asteraceae (Compositae). – In: Hedberg, I., Friis, I. & Edwards, S. (ed.), Flora of
Ethiopia and Eritrea 4(2). Addis Ababa & Uppsala.
Watanabe, K. 2008: Index to chromosome numbers in Asteraceae. Published on the internet:
http://www-asteraceae.cla.kobe-u.ac.jp/index.html. Accessed 31 January 2008.

103
6. DIANTHOSERIS CONGENER OF CREPIS

104
 7. OUTLOOK

Chapter 7

Outlook

The morphological, karyological and molecular data acquired and interpreted in the course of
the present analysis on Crepis yielded new insights into phylogeny, evolution and
systematics of the genus; phylogenetic reconstructions revealed the genus Crepis to be
polyphyletic, the measurement and reconstruction of character progression demonstrated
genome size to shrink during evolution, and a survey of morphological and anatomical data
helped to resolve systematic ambiguities. But with the applied methods not all questions
towards an understanding of the evolutionary history of the genus could be answered: such
as explaining the phylo- and biogeography of the genus, and clarifying the role of karyotype
progression and hybridisation for speciation processes within the genus.

Further analyses might combine fossil records and molecular sequence data by applying
molecular clock hypotheses on a phylogenetic tree. Supporting data is already available:
Babcock (1947a) dated the origin of the genus into the late Tertiary (Miocene, 24-5 mya) in
the Central Asian Altai/Tien Shan region. The fossil record of Crepis reaches back into the
Middle Pliocene (Piacenzian, 3.6-2.6 mya): Reid & Reid (1916) report fossil achenes of two
Crepis species (C. spec., C. blattarioides) for the Reuverian Beds in southern Holland; later
Babcock (1947a) identified the achenes to belong to C. tergluoensis and C. conyzaefolia.
The fossil pollen record is not genus specific but well documented for Compositae in general
(e.g. Blackmore et al., 1986). A molecular clock analyses (r8s, Sanderson, 2002), partly
based on the recent analysis of the biogeographic history of the Cichorieae (Tremetsberger
et al., subm.), could further illuminate the time of origin of the genus Crepis. Combined with
the analysis of distributional data via a dispersal-vicariance approach (Ronquist, 1997) using
the program DIVA, the biogeographic history of the genus could be reconstructed. So
Babcock’s hypotheses that the genus originated in Central Asian Mountains and from there
spread south-westward into the Mediterranean regions and Africa, westward into Europe and
north-eastward into North America via the Bering land Bridge, which joined Siberia and
Alaska repeatedly during the Pleistocene (1.8 my - 10.000 BP), can be re-evaluated. The
evolution of the 15 polyploid Crepis species of sect. Psilochaenia also can be explored based
on the phylogeographic analysis of the history of the genus Crepis.

The historical importance of Crepis as model group for karyological studies has been
perpetuated by the present results (chapter 3) providing promising starting points to further

105
 7. OUTLOOK

illuminate karyotype evolution within the genus. Babcock (Babcock & Cameron, 1934;
Babcock et al., 1937; Babcock & Jenkins, 1943; Babcock, 1947a) postulated hypotheses on
karyological evolution in the genus: chromosome size and number decrease while
chromosomal asymmetry increases during evolution. But the methodological potential has
increased since Babcock’s time and neither present molecular results (chapter 2) nor studies
by e.g. Jones & Brown (1976) or Dimitrova & Greilhuber (2001) could confirm Babcock’s
postulations. New hypotheses should be tested: Can monophyletic clades (ITS, chapter 2)
be characterised by karyotypic similarities (karyomorphotypes)? Is hybridisation possible
between species of different karyomorphotypes? To identify and characterise different
karyomorphotypes the following methods should be applied as they proved to yield promising
results in earlier studies on Crepis (Dimitrova & Greilhuber, 2000, 2001): Giemsa C-banding
(Schwarzacher et al., 1980), Ag-Nor staining (Bloom & Goodpasture, 1976; Kodama et al.,
1980), measurements of chromosomes (total length of chromosomes, length of short and
long arms and the satellite), measurements of band positions and the calculation of
short/long arm ratio, and the calculation of centromeric and asymmetry index. The data
gathered could be mapped on the molecular phylogenetic reconstruction to trace character
changes and might possibly be used for cladistic analyses. The distribution of certain
karyomorphotypes within species could indicate possible hybridisation events within the
genus, as the role of chromosomal rearrangements and hybridisation within the genus up to
now remains largely obscure. Babcock (1947a) considered hybridisation as minor agent of
speciation, but the incongruence of nuclear and chloroplast phylogenies (see chapter 2) hint
on possible reticulation in the phylogenetic history of the genus Crepis. To identify signatures
of hybridisation in molecular data it is important to eliminate other causes of phylogenetic
incongruence (McBreen & Lockhart, 2006), so the incongruence of nuclear and chloroplast
phylogenies has to be analysed, statistically assessed and taxa which are possibly
responsible for the incongruence have to be identified (for methodological approaches see
e.g. Johnson & Soltis, 2001). Network reconstructions (e.g. SplitsTree, Huson & Bryant,
2006) account for more complex evolutionary events than dichotomous divergence and can
therefore be used to analyse and visualise reticulate evolution.

It is also recommended to further validate the phylogenetic relations within the genus by
using other chloroplast markers to resolve apical branching patterns in the chloroplast
phylogeny (e.g. psbA-trnH) and nuclear single-copy genes to sustain basal divergences and
further illuminate ambiguous molecular groupings, e.g. Clade II (chapters 2,4). Possible
candidates (e.g. DHS, QG8140) were recently proposed for Compositae (Álvarez et al.,
2008).

As the incongruence of molecular and current taxonomic groups within Crepis s.str. could so
far not completely be resolved, an expansion of morphological analyses (as indicated in
chapter 4) is of high systematic and taxonomical value.

106
 7. OUTLOOK

The here presented suggestions are all directly based on or connected to data accumulated
within the present study, but other approaches are also of interest: e.g. to investigate the
correlation of genome size to distribution area (chapter 3), or extending the karyotype studies
using FISH (Fluorescence In-Situ Hybridisation) methodology (e.g. to clarify relations within
the “hybrid complex” (Clade VII, chapters 3,4,5)). A fusion of current and future results would
not only further enlighten the evolutionary history of Crepis and karyotype evolution in higher
plants, but also contribute towards a revision and new infrageneric classification of the
genus.

107
7. OUTLOOK

108
 8. SUMMARY

Chapter 8

Summary

8.1 SUMMARY
Karyotype alterations play an active role in plant speciation processes. N. Enke’s dissertation
“Phylogeny and Character Evolution in the genus Crepis L. (Cichorieae, Compositae)”
contributes towards the understanding of the influence of karyotype changes on
diversification in diploid plant genera using Crepis as model group.

The genus Crepis is distributed in the Holarctic and Africa with the highest diversity in the
Mediterranean. Most of the species within the genus are diploid; except for the 15 species of
section Psilochaenia and approximately five additional species. The basic chromosome
number in the genus ranges from x = 3 to 6 (7), respectively x = 11 in section Psilochaenia.

The genus Crepis was revised by Babcock in 1947 (The genus Crepis I&II, University of
California Press). He assigned 196 of the over 200 species known today to 27 sections
mainly due to karyological characters, such as chromosome number and shape. Babcock
also postulated hypotheses on evolution and speciation within Crepis: karyotype
rearrangements are the driving force of speciation, while hybridisation only plays a minor role
in species formation.

Based on Babcock’s monograph the present study (1) postulates phylogenetic hypotheses
for Crepis, (2) reassesses existing hypotheses on karyotype evolution and speciation
mechanisms within the genus, (3) identifies morphological and anatomical characters
reflecting infrageneric groups, and (4) revaluates the current infrageneric classification.

The phylogenetic relations within the genus are inferred from both nuclear (ITS) and
chloroplast (matK) markers. Genome size is measured by flow cytometry and evaluated on a
molecular phylogenetic background. Achene anatomy and morphology, pollen morphology
and structure of style branch papillae are investigated via SEM and LM for their applicability
to delimitate infrageneric groups.

The phylogenetic reconstructions based on 123 ITS sequences of 78 species and 73 matK
sequences of 52 species differ in apical branching pattern but support three main clades: the
first comprises approximately 80% of sampled species as Crepis s.str., the second includes
the genera Lapsana and Rhagadiolus and all sampled taxa of five Crepis sections, the third
corresponds to former Crepis section Ixeridopsis; now genus Askellia.

109
 8. SUMMARY

Combined karyological and molecular analyses show a complex pattern of karyotype

evolution within the genus. Chromosome number is highly variable in and between clades,
and de- and also increased during evolution. A trend toward a decrease in genome size
within Crepis is observed. Annuals predominantly feature small genomes while in perennials
genome size is variable. Species from the Mediterranean in general feature smaller genomes
than species from N-Europe, Eurasia and Central/E-Asia.

Of the tested microcharacters pappus structure differs between the three clades inferred by
molecular analyses. Achene anatomy, pollen morphology and style branch papillae provide
no evidence for infrageneric classification, mostly due to low sample size. Achene anatomy
and style branch papillae show sufficient variation for systematic use if sample size is
broadened.

As taxonomic consequences of the presented study the genera Lapsana and Rhagadiolus
are preserved and the genus Crepis is treated as paraphyletic. The former genus
Dianthoseris is included into Crepis. Comments on a revised sectional classification are
given.

110
 8. SUMMARY

8.2 ZUSAMMENFASSUNG
Artbildungsprozesse bei Pflanzen können auf Karyotypveränderungen zurückgehen. Die
Dissertation von N. Enke “Phylogeny and Character Evolution in the Genus Crepis L.
(Cichorieae, Compositae)” analysiert den Einfluß von chromosomalen Veränderungen auf
die Artneubildung in diploiden Pflanzengruppen. Die Gattung Crepis wird modellhaft in
diesem Rahmen untersucht.

Crepis L. ist in der gesamten Holarktis und Afrika verbreitet mit der höchsten Diversität im
Mediterranraum. Die meisten Arten der Gattung sind diploid, bis auf 15 Arten der Sektion
Psilochaenia und ca. 5 andere Arten. Die Chromosomengrundzahl variiert zwischen x=3 und
x=6 (7), bzw. x=11 in der Sektion Psilochaenia.

Die letzte Revision der Gattung geht auf Babcock (1947, The Genus Crepis I&II, University of
California Press) zurück. Er gliederte 196 der über 200 heute bekannten Arten auf Grund
hauptsächlich karyologischer Ähnlichkeiten, wie z.B. Chromsomenform und – zahl, in 27
Sektionen. Babcock postulierte, daß die Hauptursache von Artbildungsereignissen
Änderungen des Karyotyps seien, und Hybridisierung nur eine kleine Rolle spiele.

Aufbauend auf Babcocks Monografie postuliert die vorliegende Arbeit (1) phylogenetische
Hypothesen für Crepis, überprüft (2) bereits bestehende Hypothesen zu Karyotypevolution
und Artbildungsmechanismen in der Gattung, identifiziert (3) morphologische und
anatomische Merkmale zur Definition infragenerischer Gruppen, und bewertet (4) die
bestehende Gliederung in Sektionen neu.

Die phylogenetischen Zusammenhänge der Gattung Crepis wurden mit Hilfe des
Kernmarkers ITS und des Chloroplastengens matK rekonstruiert. Genomgrößen wurden
mittels Flow Cytometry gemessen und in einem phylogenetischen Zusammenhang
interpretiert. Frucht- und Pollenmerkmale, sowie die Papillae der Narbenäste wurden mittels
LM und REM auf eine Eignung zur Abgrenzung infragenerischer Gruppen untersucht.

Die Phylogenierekonstruktionen (ITS/matK: 123/73 Sequenzen von 78/52 Arten),

unterscheiden sich in der Anordnung der apikalen Gruppen, aber unterstützen beide drei
Hauptkladen: Die erste besteht aus ca. 80% der untersuchten Arten als Crepis s.str., die
zweite umfasst neben den Gattungen Lapsana und Rhagadiolus alle untersuchten Arten aus
fünf Crepis Sektionen, die dritte entspricht der früheren Crepis Sektion Ixeridopsis,
mittlerweile Gattung Askellia.

Die gemeinsame Interpretation karyologischer und molekularer Ergebnisse ließ ein

komplexes Muster der Karyotypevolution erkennen. Die Chromosomengrundzahl variiert
stark sowohl innerhalb als auch zwischen den Kladen. Sowohl eine Ab- als auch eine
Zunahme der Chromosomengrundzahl während der Aufspaltung rezenter Arten konnte
beobachtet werden, sowie eine Abnahme der Genomgröße. Annuelle haben tendenziell
kleine Genomgrößen, während ausdauernde Arten eine höhere Variation zeigen. Des

111
 8. SUMMARY

Weiteren besitzen Arten der Mediterranregion im Allgemeinen kleinere Genome als Arten
aus N- und Mitteleuropa, Eurasien sowie Zentral- und O-Asien.

Von den untersuchten Mikromerkmalen unterschied die Struktur der Pappusborsten

zwischen den drei Kladen. Frucht-, Pollen- und Narbenastmerkmale zeigten auf Grund der
geringen Stichprobenmenge keine interpretierbaren Muster. Die behandelten Merkmale
könnten jedoch bei Einbeziehung weiterer Arten eine Eignung als Unterscheidungsmerkmal
infragenerischer Gruppen aufweisen.

Taxonomische Konsequenzen aus den vorliegenden Ergebnissen sind die Aufrechterhaltung

der Gattungen Lapsana und Rhagadiolus, die Behandlung von Crepis als paraphyletisches
Taxon und die Eingliederung der ehemaligen Gattung Dianthoseris. Anmerkungen zu einer
revidierten sektionalen Gliederung der Gattung werden gemacht.

112
 9. LITERATURE CITED

Literature Cited

Adylov, T.A. & Zuckerwanik, T.I. 1993. Conspectus Florae Asiae Mediae. Editio Academiae
scientiarum Reipublicae Uzbekistan, Tashkent, Uzbekistan.
Albach, D.C. & Greilhuber, J. 2004. Genome Size Variation and Evolution in Veronica. Annals of
Botany 94: 897-911.
Aluka 2007-continued: African plants. - Published on the Internet: http://www.aluka.org/
Álvarez, I. & Wendel, J.F. 2003. Ribosomal ITS Sequences and Plant Phylogenetic Inference.
Molecular Phylogenetics and Evolution 29:417-434.
Álvarez, I., Costa, A., & Feliner, G.N. 2008. Selecting Single-Copy Nuclear Genes for Plant
Phylogenetics: A Preliminary Analysis for Senecioneae (Asteraceae). Journal of Molecular
Evolution 66:276-291.
Arnheim, N. 1983. Concerted Evolution in Multigene Families. Pp. 38-61. In: Nei, M. & Koehn, R. (ed.)
Evolution of Genes and Proteins. Sinauer Associates, Sunderland.
Arnheim, N., Krystal, M., Schmickel, R., Wilson, G., Ryder. O., & Zimmer, E. 1980. Molecular
Evidence for Genetic Exchanges Among Ribosomal Genes on Nonhomologous
Chromosomes in Men and Apes. Proceedings of the National Academy of Science 77:7323-
7327.
Avdulov, N.P. 1931. Karyo-Systematische Untersuchungen der Familie Gramineen. Bulletin Applied
Botany Genetics and Plant Breeding 44: 1-428.
Avery, P. 1930. Cytological Studies of Five Interspecific Hybrids of C. leontodontoides. University of
California Publications in Agricultural Science 6: 135-167.
Babcock, E.B. 1947a. The Genus Crepis I. The Taxonomy, Phylogeny, Distribution and Evolution of
Crepis. University of California Publications 21. University of California Press, Berkeley & Los
Angeles.
Babcock, E.B. 1947b. The Genus Crepis II. Systematic Treatment. University of California
Publications 22. University of California Press, Berkeley & Los Angeles.
Babcock, E.B. 1949. Supplementary notes on Crepis. II. Phylogeny, distribution and Matthew's
principle. Evolution 3: 374-376.
Babcock, E.B. & Cameron, D.R. 1934. Chromosomes and Phylogeny in Crepis, II.: The
Relationships of One Hundred Eight Species. University of California Publication in
Agricultural Science 6:287-324.
Babcock, E.B. & Cave, M.S. 1938. A Study of Intra- and Interspecific Relations of Crepis foetida L.
Zeitschrift für Induktive Abstammungs- und Vererbungslehre 75: 124-160.
Babcock, E.B. & Jenkins, J.A. 1943. Chromosomes and phylogeny in Crepis, III: The relationships of
one hundred and thirteen species. University of California Publications in Agricultural Science
18: 241-292.
Babcock, E.B. & Stebbins, G.L. 1937. The Genus Youngia. Carnegie Institution of Washington
Publication 484. Carnegie Institution of Washington, Washington, DC.
Babcock, E.G., Stebbins, G.L., & Jenkins, J.A. 1937. Chromosomes and Phylogeny in Some
Genera of the Crepidinae. Cytologia Fuji Jubilee Volume 188-210.

113
 9. LITERATURE CITED

Bachmann, K., Chambers, K.L., & Price, H.J. 1981. Genetic Determination of Pappus Part Number
in the Annual Hybrid Microseris B87 (Asteraceae - Lactuceae). Plant Systematics and
Evolution 138(3-4): 235-246.
Baldwin, B.G. 1993. Molecular phylogenetics of Calycadenia (Compositae) based on ITS sequences
of nuclear ribosomal DNA: chromosomal and morphological evolution reexamined. American
Journal of Botany 80: 222-238.
Baldwin, B.G., Sanderson, M.J., Porter, M.J., Woiciechowski, M.F., Campbell, C.S., & Donoghue,
M.J. 1995. The ITS Region of Nuclear Ribosomal DNA: A Valuable Source of Evidence on
Angiosperm Phylogeny. Annals of the Missouri Botanical Garden 82: 247-277.
Barakat, A., Carels, N., & Bernardi, G. 1997. The Distribution of Genes in the Genomes of
Gramineae. Proceedings of the National Academy of Sciences 94: 6857-6861.
Bennett, M.D. 1972. Nuclear DNA Content and Minimum Generation Time in Herbaceaeous Plants.
Proceedings of the Royal Society of London B 181: 109-135.
Bennett, M.D., Johnston, S., Hodnett, G.L., & Price, H.J. 2000. Allium cepa L. Cultivars from Four
Continents Compared by Flow Cytometry Show Nuclear DNA Constancy. Annals of Botany
85: 351-357.
Bennett, M.D. & Leitch, I.J. 1995. Nuclear DNA Amounts in Angiosperms. Annals of Botany 76: 113-
176.
Bennett, M.D. & Leitch, I.J. 2005. Genome Size Evolution in Plants. Pp. 89-162. In: Gregory, T.R.
(ed.) The Evolution of the Genome. Elsevier Academic Press. Amsterdam, New York.
Bennett, M.D., Leitch, I.J., & Hanson, L. 1998. DNA Amounts in Two Samples of Angiosperm
Weeds. Annals of Botany 82 (Supplement A):1 21-234.
Bennet, M.D. & Smith, J.B. 1976. Nuclear DNA Amounts in Angiosperms. Philosophical
Transactions of the Royal Society of London B 274: 227-274.
Bennetzen, J.L. 1996. The Contributions of Retroelements to Plant Genome Organisations, Function
and Evolution. Trends in Microbiology 4: 347-353.
Bennetzen, J.L. 2000. Transposable Element Contribution to Plant Gene and Genome Evolution.
Plant Molecular Biology 42:251-269.
Bennetzen, J.L. & Kellogg, E.A. 1997. Do Plants Have a One-Way Ticket to Genomic Obesity? The
Plant Cell 9: 1509-1514.
Bentham, G. & Hooker, J.D. 1873. Genera Plantarum 2. Lowell Reeve, London.
Blackmore, S. 1984. Compositae - Lactuceae. In: Punt, W. & Clarke, G.C.S. (eds.) The Northwest
European Pollen Flora IV. pp. 45-85. Elsevier, Amsterdam.
Blackmore, S. & Persson, V. 1996: Palynology and systematics of the Crepidinae (Compositae:
Lactuceae). - Pp. 111-122 in: Hind, D. J. N. & Beenthe, H. J. (ed.), Compositae: Systematics.
Proceedings of the International Compositae Conference, Kew, 1994, 1. Kew.
Blackmore, S., van Campo, E., & Crane, P.R. 1986. Lophate Compositae Pollen from the Miocene
and Pliocene oft he Mediterranean Region. Canadian Journal of Botany 64:3101-3112.
Blattner, F.R. 1999. Direct amplification of the entire ITS region from poorly preserved plant material
using recombinant PCR. BioTechniques 27: 1180-1186.
Bloom, S.E. & Goodpasture, C. 1976. An Improved Technique for Selective Silver Staining of
Nucleolar Organizer Regions in Human Chromosomes. Human Genetics 34:199-206.
Bobrov, E.G. & Tzevelev, N.N.. 1964. Compositae. Tribe Cichorieae. In: Komarov, V.L. Flora of the
USSR 29. Moscow.

114
 9. LITERATURE CITED

Bobrov, E.G. & Tzevelev, N.N. 2000. Compositae. Tribe Cichorieae. 691-707. In: Komarov, V.L.
Flora of the USSR vol. 29. Science Publishers, Inc., Enfield.
Boulos, L. 1973: Révision systématique du genre Sonchus L. s.l. IV. Sous-genre 1. Sonchus. -
Botanical Notes 126: 155-196.
Bremer, K. 1994. Asteraceae. Cladistics and Classification. Timber Press, Portland, Oregon.
Briggs, D. & Walters, S.M. 1984. Plant Variation and Evolution. 2nd edition. Cambridge University
Press, Cambridge.
CGIAR-CSI [Consortium of Spatial Information] 2004: NASA Shuttle Radar Topographic Mission
(SRTM) 90 m digital elevation data (DEMs). Published on the internet http://srtm.sci.cgiar.org.
Cassini, H. 1830. Tableau synoptique des Synanthérées. In: Cuvier, F. (ed.), Dictionnaire des
Sciences Naturelles 60: 566-587. Paris.
Cerbah, M., Souza-Chies, T., Jubier, M.F., Lejeune, B., & Siljak-Yakovlev, S. 1998. Molecular
Phylogeny of the Genus Hypochaeris Using Internal Transcribed Spacers of Nuclear rDNA:
Inference for Chromosomal Evolution. Molecular Biology and Evolution 15(3): 1998.
Chan, R., Baldwin, B.G. & Ornduff, R. 2002. Cryptic goldfields: A molecular phylogenetic
reinvestigation of Lasthenia californica sensu lato and close relatives (Compositae: Helianthae
sensu lato). American Journal of Botany 89: 1103-1112.
Cheptou, P.O., Berger, A., Blanchard, A., Collin, C., & Escarre, J. 2000. The Effect of Drought
Stress on Inbreeding Depression in Four Populations of The Mediterranean Outcrossing Plant
Crepis sancta (Asteraceae). Heredity 85: 294-302.
Collins, J.L. 1924. Inheritance in Crepis capillaris (L.) Wallr., III: Nineteen Morphological and Three
Physiological Characters. University of California Publications in Agricultural Science 2: 305-
320.
Cufodontis, G. 1968: Crassulaceae, Celastraceae, Thymelaeaceae und Compositae aus dem
Tanasee-Gebiet und dem Semyen-Gebirge (Äthiopien). Ergebnisse der botanischen Reise
Oskar Sebald im Jahre 1966 nach Äthiopien, Nr. 2. Stuttgarter Beiträge zur Naturkunde 195.
Dart, S., Kron, P., & Mable, B. K. 2004. Characterizing Polyploidy in Arabidopsis lyrata Using
Chromosome counts and Flow Cytometry. Canadian Journal of Botany 82: 185-197.
Dimitrov, B.D. 1994. Types of Chromosomal Aberrations Induced by Seed Aging and Crepis
capillaris. Environmental and Molecular Mutagenesis 23:318-322.
Dimitrova, D., Ebert, I., Greilhuber, J., & Kozhuharov, S. 1999. Karyotype Constancy and Genome
Size Variation in Bulgarian Crepis foetida s.l. (Asteraceae). Plant Systematics and Evolution
217: 245-257.
Dimitrova, D. & Greilhuber, J. 2000. Karyotype And DNA-Content Evolution In Ten Species Of
Crepis (Asteraceae) Distributed in Bulgaria. Botanical Journal of the Linnean Society 132:
281-297.
Dimitrova, D. & Greilhuber, J. 2001. C-Banding Patterns and Quantitative Karyotype Characteristics
of Bulgarian Species of Crepis (Asteraceae). Plant Biology 3:88-97.
Dobzhansky, T.G. 1937. Genetics and the Origin of Species. Columbia University Press, New York.
Dobzhansky, T.G. 1970. Genetics of the Evolutionary Process. Columbia University Press, New York.
Ehrendorfer, F. 1976. Chromosome numbers and differentiation of centrospermous families. Plant
Systematics and Evolution 126: 27-30.
Endress, P.K. 2002. Morphology and Angiosperm Systematics in the Molecular Era. The Botanical
Review 68(4):545-570.

115
 9. LITERATURE CITED

Enke, N. & Gemeinholzer, B. 2008. Babcock Revisited: New Insights into Generic Delimitation and
Character Evolution in Crepis L. (Compositae: Cichorieae) From ITS and MatK Sequence
Data. Taxon. 57(3): 756-768.
Enke, N., Fuchs, J. & Gemeinholzer, B. subm. Shrinking Genomes? Evidence from Genome Size
Variation in Crepis L. (Compositae).
Enke, N., Kilian, N., Nemomissa, S. & Gemeinholzer, B. 2008. Afro-alpine Dianthoseris actually a
congener of Crepis s.str. (Compositae, Cichorieae, Crepidinae). Botanische Jahrbücher 127
(3): 389–405.
Erdtman, G. 1960. The acetolysis method. A revised description. Svensk Botanisk Tidskrift 54:561-
564.
Fehrer, J., Gemeinholzer, B., Chrtek, J. Jr. & Bräutigam, S. 2007. Incongruent plastid and nuclear
DNA phylogenies reveal ancient intergeneric hybridization in Pilosella hawkweeds (Hieracium,
Cichorieae, Asteraceae). Molecular Phylogeny and Evolution 42: 347-361.
Flavell, R.B. 1980. The Molecular Characterization and Organization of Plant Chromosomal DNA
Sequences. Annual Review of Plant Physiology 31: 569-596.
Flavell, R.B. 1986. Repetitive DNA and Chromosome Evolution in Plants. Philosophical Transactions
of the Royal Society of London B 312: 227-242.
Flavell, R. B., Rimpau, J., & Smith. D.B. 1977. Repeated Sequence DNA Relationships in Four
Cereal Genomes. Chromosoma 63: 205-222.
Fries, R. E. 1928: Zur Kenntnis der Compositae des tropischen Ostafrikas. Acta Horti Bergiana 9:
109-164.
Gall, J.G. 1981. Chromosome Structure and the C-value Paradox. Journal of Cell Biology 91: 3s-14s.
Garnatje, T., Vallès, J., Garcia, S., Hidalgo, O., Sanz, M., Canela, M.A., & Siljak-Yakovlev, S.
2004. Genome Size in Echinops L. and Related Genera (Asteraceae, Cardueae):
Karyological, Ecological and Phylogenetic Implications. Biology of the Cell 96: 117-124.
Gemeinholzer, B., Kilian, N. & Lack, W.H. 2006. Molecular phylogenetics and systematics of
Cichorieae (Asteraceae). In: Abstracts. The International Compositae Alliance (TICA - Deep
Achene, Barcelona, July 2006). Barcelona.
Gillett, J. B. 1972: W. G. Schimper's botanical collections localities in Ethiopia. Kew Bulletin 27: 115-
128.
Goertzen, L.R., Cannone, J.J., Gutell, R.R. & Jansen, R.K.. 2003. ITS secondary structure derived
from comparative analysis: implications for sequence alignment and phylogeny of the
Asteraceae. Molecular Phylogeny and Evolution 29: 216-234.
Gokhman, V.E. 2007. Karyosystematics of Parasitic Hymenoptera: Taxonomic Decisions a the
Species Level. Comparative Cytogenetics 1(1): 85-88.
Gottlieb, L.D. & Ford, V.S. 1996. Phylogenetic relationships among the sections of Clarkia
(Onagraceae) inferred from the nucleotide sequences of PgiC. Systematic Botany 21: 45-62.
Govindaraju, D.R. & Cullis, C.A. 1991. Modulation of Genome Size in Plants: The Influence of
Breeding Systems and Neighbourhood Size. Evolutionary Trends in Plants 5: 43-51.
Graham, S. W. & Olmstead, R.G. 2000. Utility of 17 Chloroplast Genes for Inferring the Phylogeny of
the Basal Angiosperms. American Journal of Botany 87:1712-1730.
Grant, V. 1982. Periodicities in the Chromosome Numbers of the Angiosperm. Botanical Gazette
143:379-389.
Grant, V. 1982. Chromosome Number Patterns in Primitive Angiosperms. Botanical Gazette 143: 390-
394.

116
 9. LITERATURE CITED

Grau, J. 1980. Die Testa der Mutisieae und ihre systematische Bedeutung. Mitteilungen der
botanischen Staatsammlung München 16: 269-332.
Gregory, T.R. 2001. Coincidence, Coevolution, or Causation? DNA Content, Cell Size, And The C-
value Enigma. Biological Reviews 76: 65-101.
Gregory, T.R. 2005. Macroevolution and the Genome. Pp. 679-729. In: Gregory, T.R. (ed.) The
Evolution of the Genome. Elsevier, California.
Greilhuber, J. 1998. Intraspecific Variation in Genome Size in Angiosperms: A Critical Reassessment.
Annals of Botany. 82 (Supplement A): 27-35.
Greilhuber, J. 2005. Intraspecific Variation in Genome Size in Angiosperms: Identifying its Existence.
Annals of Botany. 95: 91-98.
Greilhuber, J., Borsch, T., Müller, K., Worberg, A, Porembski, S., & Barthlott, W. 2006. Smallest
Angiosperm Genomes Found in Lentibulariaceae, with Chromosomes of Bacterial Size. Plant
Biology 8: 770-777.
Greilhuber, J., Dolezel, J., Lysák, M., & Bennet, M.D. 2005. The Origin, Evolution, and Proposed
Stabilisation of the Terms "Genome Size" and "C-value" to Describe Nuclear DNA Contents.
Annals of Botany. 95: 255-260.
Greilhuber, J. & Ebert, I. 1994. Genome Size Variation in Pisum sativum. Genome 37: 646-655.
Greilhuber, J. & Ehrendorfer, F. 1988. Karyological Approaches to Plant Taxonomy. ISI Atlas of
Science: Animal and Plant Sciences 1: 289-297.
Guerra, M. 2008. Chromosome Numbers in Plant Cytotaxonomy: Concepts and Implications.
Cytogenetic and Genome Research 120:339-350.
Hall, T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for
Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95-98.
Hancock, J.M. 2002. Genome Size And The Accumulation of Simple Sequence Repeats: Implications
of New Data From Genome Sequencing Projects. Genetica 115: 93-103.
Hedberg, I. & Hedberg, O. 1977: Chromosome numbers of afroalpine and afromontane angiosperms.
Botanical Notes 130: 1-24.
Hedberg, O. 1957: Afroalpine vascular plants. A Taxonomic Revision. Symbolae Botanicae
Upsalienses 15.
Hedberg, O. 1970: Evolution of the afroalpine flora. Biotropica 2(1): 16-23.
Hennig, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana.
Hijmans, R., Guarino, L., Mathur, P. & Jarvis, A. 2005: DIVA-GIS Version 5.2. Published on the
Internet http://diva-gis.org.
Hillis, D.M., Moritz, D., Porter, C.A., & Baker, R.J. 1991. Evidence for Biased Gene Conversion in
Concerted Evolution of Ribosomal DNA. Science 251:308-310.
Hoffmann, O. 1889. Compositae. 373-374. In: Engler, A. & Prantl, K. (eds). Die natürlichen
Pflanzenfamilien, IV. vol. 5. Engelmann. Leipzig.
Hoffmann, O. 1890-94. Compositae. Pp. 87-391 in: Engler, A. & Prantl, K. (ed.), Die natürlichen
Pflanzenfamilien 4(5). Leipzig.
Hollingshead, L. 1930a. A Lethal Factor in Crepis Effective Only in an Interspecific Hybrid. Genetics
15: 114-140.
Hollingshead, L. 1930b. Cytological Investigations of Hybrids and Hybrid Derivative of Crepis
capillaris and Crepis tectorum. University of California Publications in Agricultural Science 6:
55-96.

117
 9. LITERATURE CITED

Holmgren, P. K. & Holmgren, N. H. 1998-continued: Index herbariorum: a global directory of public

herbaria and associated staff. Published on the internet at http://sweetgum.nybg.org/ih/
Hughes, M.B. & Babcock. E.B. 1950. Self-Incompatibility in Crepis foetida (L.) subsp. rhoeadifolia
(Bieb.) Schinz et Keller. Genetics 35: 570-588.
Huson, D.H. & Bryant, D. 2006. Application of Phylogenetic Networks in Evolutionary Studies.
Molecular Biology and Evolution 23(2):254-267.
Jakob, S.S., Meister, A., & Blattner, F.R. 2004. The Considerable Genome Size Variation of
Hordeum Species (Poaceae) Is Linked to Phylogeny, Life Form, Ecology, and Speciation
Rates. Molecular Biology and Evolution 21(5): 860-869.
Jamilena, M., Ruiz Rejon, C., & Ruiz Rejon, M. 1994. A molecular analysis of the origin of the Crepis
capillaris B chromosome. Journal of Cell Science 107(3): 703-708.
Jaubert, H.P. & Spach, E. 1847-50. Illustrationes Plantarum Orientalium 3. Paris.
Jeffrey, C. 1966. Notes on Compositae I. The Cichorieae in East Tropical Africa. Kew Bulletin 18:
427-486.
Jeffrey, C. & Beentje, H. J. 2000: Cichorieae. - Pp. 63-108 in: Beentje, H. J. (ed.), Flora of tropical
East Africa. Compositae (Part 1). Rotterdam.
Johnson, L.A. & Soltis, D.E. 1995. Phylogenetic inference in Saxifragaceae sensu stricto and Gilia
(Polemoniaceae) using matK sequences. Annals of the Missouri Botanical Garden 82: 149-
175.
Johnson, L.A. & Soltis, D.E. 2001. Assessing Congruence: Empirical Examples from Molecular Data.
Pp. 297-343. In: Soltis, D.E., Soltis, P.S., & Doyle, J.J. (eds). Molecular Systematics of Plants
II: DNA Sequencing. Kluwer Academic Publishers, Boston.
Jones, R.N. & Brown, L.M. 1976. Chromosome evolution and DNA variation in Crepis. Heredity 36:
91-104.
Jonson, K. 1998. Robertsonian Fusion and Centric Fission in Karyotype Evolution of Higher Plants.
Botanical Review 64: 273-289.
Kalendar, R., Tanskanen, J., Immonen, S., Nevo, E., & Schulman, A.H. 2000. Genome Evolution of
Wild Barley (Hordeum spontaneum) by BARE-1 Retrotransposon Dynamics in Response to
Sharp Microclimatic Divergence. Proceedings of the National Academy of Science, USA. 97:
6603-6607.
Kamari, G. 1976. Cytotaxonomic Study of the Crepis neglecta Complex in Greece. Publication of the
Botanical Institute, University of Patras.
Kamari, G. 1992. Karyosystematic Studies On Three Crepis Species (Asteraceae) Endemic To
Greece. Plant Systematics and Evolution. 182: 1-19.
Kellogg, E.A. 1998. Relationships of Cereals, Crops and Other Grasses. Proceeding of the National
Academy of Science, USA. 95: 2005-2010.
Kihara, H. & Ono, T. 1926. Chromosomenzahlen und systematische Gruppierung der Rumex-Arten.
Zeitschrift für Zellforschung und mikroskopische Anatomie 4: 475-481.
Kilian, N. 1997. Revision of Launaea Cass. (Compositae, Lactuceae, Sonchinae). Englera 17.
Kilian, N., Gemeinholzer, B. & Lack, H. W. 2008 (in press): Tribe Cichorieae. In: Funk, V., Susanna,
A., Stuessy, T. & Bayer, R. (ed.), Systematics and evolution of the Compositae. Vienna.
Kirik, A., Salomon, S. & Puchta, H. 2000. Species-specific Doublestrand Break Repair and Genome
Evolution In Plants. EMBO Journal. 19: 5562-5566.
Knight, C.A. & Ackerley, D.D. 2002. Variation in Nuclear Content Across Environmental Gradients: A
Quantile Regression Analysis. Ecology Letters. 5: 66-76.

118
 9. LITERATURE CITED

Knight, C.A., Molinari, N.A., & Petrov, D.A. 2005. The Large Genome Constraint Hypothesis:
Evolution, Ecology and Phenotype. Annals of Botany 95(1): 177-190.
Knoop, V. & Müller, K. 2006. Gene und Stammbäume. Elsevier, Munich.
Kodama, Y., Yoshida, MC. & Sasaki, M. 1980. An Improved Silver Staining Technique for Nucleolus
Organizer Regions by Using Nylon Cloth. Japanese Journal for Human Genetics 25:229-233.
Kubis, K., Schmidt, T., & Heslop-Harrison J.S. 1998. Repetitive DNA Elements as a Major
Component of Plant Genomes. Annals of Botany. 82 (Supplement A): 45-55.
Lack, H. W. 2006. Tribe Cichorieae Lam. & DC. - Pp. 180-199 in: Kadereit, J. W. & Jeffrey, C. (ed.),
The families and genera of vascular plants 8. Flowering plants. Eudicots. Asterales. Berlin,
etc.
Leitch, I.J. & Bennett, M.D. 2004. Genome Downsizing in Polyploid Plants. Biological Journal of the
Linnean Society. 82: 651-663.
Leitch, I.J., Chase, M.W., & Bennett, M.D. 1998. Phylogenetic Analysis of C-DNA Values Provides
Evidence for a Small Ancestral Genome Size in Flowering Plants. Annals of Botany 82: 85-94.
Levin, D.A. 2002. The Role of Chromosomal Change in Plant Evolution. Oxford University Press, New
York.
Levin, D.A. & Funderburg, S.W. 1979. Genome Size in Angiosperms: Temperate versus Tropical
Species. American Naturalist 114: 784-795.
Levitsky, G.A. 1924. The material basis of heredity. State Publication Office of the Ukraine, Kiev. [in
Russian]
Levitsky G.A. 1931. The morphology of chromosomes. Bulletin of Applied Botany, of Genetics and
Plant Breeding 27: 19-174.
Lewis, H. 1952. Chromosome phylogeny and habitat preference in Clarkia. Evolution 7: 102-109.
Linné, C. 1753. Species Plantarum. Stockholm.
Loureiro, J., Rodriguez, E., Dolezel, J., & Conceição, S. 2007. Two New Nuclear Isolation Buffers
for Plant DNA Flow Cytometry: A Test with 37 Species. Annals of Botany 100(4): 875-888.
Lysak, M.A., Berr, A., Pecinka, A., Schmidt, R., McBreen, K. & Schubert, I. 2006. Mechanisms of
chromosome number reduction in Arabidopsis thaliana and related Brassicaceae species.
Proceedings of the National Academy of Sciences 103: 5224-5229.
Ma, J., Devos, K.M., & Bennetzen, J.L. 2004. Analysis of LTR-Retrotransposon Structures Reveal
Recent and Rapid Genomic DNA Loss in Rice. Genome Research 14:860-869.
MacLean, N. 1973. Suggested Mechanisms for Increase in Size of the Genome. Nature New Biology
246: 205-206.
Maddison, W.P. 1997. Gene Tree in Species Trees. Systematic Biology 46:523-536.
Maddison, W.P. & Maddison, D.R. 2008. Mesquite: A Modular System for Evolutionary Analysis.
Version 2.5. http://mesquiteproject.org.
Maluzynska, J. 1990. B Chromosomes of Crepis capillaris (L.) Waller. In Vivo and In Vitro.
Uniwersytet Slaski, Katowice.
Maluzynska, J. & Schweizer, D. 1989. Ribosomal RNA Genes in B Chromosomes of Crepis capillaris
Detected By Non-Radioactive In Situ Hybridisation. Heredity 62:59-65.
Marie, D. & Brown, S.C. 1993. A Cytometric Exercise in Plant DNA Histograms, with 2C Values for 70
Species. Biology of the Cell 78: 41-51.
Masterson, J. 1994. Stomatal Size in Fossil Plants: Evidence for Polyploidy In Majority of
Angiosperm. Science 264:421-424.

119
 9. LITERATURE CITED

Mayr, E. 1942. Systematics and the Origin of Species. Columbia University Press, New York.
Mayr, E. 1997. Evolution and the Diversity of Life: Selected Essays. Belknap Press, Cambridge,
Massachussets.
McBreen, K. & Lockhart, P.J. 2006. Reconstructing Reticulate Evolutionary Histories of Plants.
Trends in Plant Science 11(8):398-404.
McMurphy, L.M., & Rayburn, A.L. 2006. Genome Size Variation in Maize Populations Selected for
Cold Tolerance. Plant Breeding 106(3): 190 - 195
McNeill, J. , Barrie, F. R., Burdet, H. M., Demoulin, V., Hawksworth, D. L, Marhold, K., Nicolson,
D. H., Prado, J., Silva, P. C., Skog, J. E., Wiersema, J. H., Turland, N. J. (ed. & compilers)
2006: International Code of Botanical Nomenclature (Vienna Code) adopted by the
Seventeenth International Botanical Congress Vienna, Austria, July 2005. Regnum Vegetabile
146.
Meikle, R.D. 1979. Rhagadiolus: A misunderstood genus. Taxon 28: 133-144.
Merxmueller, H. 1968. Melitella (Cichoriaceae) - Über ein Vorkommen in Australien und die
taxonomische Einreihung. Mitteilungen der botanischen Staatsammlung München 7: 271-275.
Moench,C. 1794. Methodus plantas horti botanici et agri Marburgensis. Officina Nova Libraria
Academiae, Marburg.
Monnier, A. 1929. Essai Monographique sur les Hieracium 78. Nancy.
Murray, B.G. 2005. When Does Intraspecific C-value Variation Become Taxonomically Significant?
Annals of Botany 95: 119-125.
Naranjo, T. & Corredor, E. 2004. Clustering of Centromeres Precedes Bivalent Chromosome Pairing
of Polyploid Wheats. Trends in Plant Science 9(5): 214-217 .
O'Kane Jr., S.L. & Schaal, B.A. 1998. Phyolgenetics of Lopezia (Onagraceae): Evidence from
chloroplast DNA restriction sites. Systematic Botany 23: 5-20.
Ohri, D. 1998. Genome Size Variation and Plant Systematics. Annals of Botany 82 (Supplement A):
75-83.
Ohri, D. 2005. Climate and Growth Form: The Consequences for Genome Size in Plants. Plant
Biology 7: 449-458.
Osman, A.K.E. 2006. Pollen Types of the Egyptian Species of Tribe Lactuceae (Subfamily
Cichorioidea-Compositae). Acta Botanica Croatica 65(2): 161-180.
Otto, S.P. & Whitton, J. 2000. Polyploid Incidence and Evolution. Annual Review of Genetics 34:401-
437.
Page, R.D.M. 1996. TREEVIEW: An application to display phylogenetic trees on personal computers.
Computer Applications in the Biosciences 12: 357-358.
Pak, J.H. 1993. Taxonomic Implications of Fruit Wall Anatomy and Karyology of Crepis sect.
Ixeridopsis (Compositae; Lactuceae). Korean Journal of Plant Taxonomy 23(1): 11-20.
Pak, J. & Bremer, K. 1995. Phylogeny and reclassification of the genus Lapsana (Asteraceae:
Lactuceae). Taxon 44: 13-21.
Pak, J.-H. & S. Kawano. 1990a. Biosystematic Studies on the Genus Ixeris and its Allied Genera
(Compositae-Lactuceae I. Fruit Wall Anatomy and Its Taxonomic Implications. Acta
Phytotaxonomica et Geobotanica 41: 43-60.
Pak, J.-H. & S. Kawano. 1990b. Biosystematic Studies on the Genus Ixeris and its Allied Genera
(Compositae-Lactuceae II Karyological Analyses. Cytologia 55:533-570.

120
 9. LITERATURE CITED

Pak, J.-H. & S. Kawano. 1990c. Biosystematic Studies on the Genus Ixeris and its Allied Genera
(Compositae-Lactuceae III Fruit Wall Anatomy and Karyology of Crepidiastrum and Paraixeris,
and their Taxonomic Implications. Acta Phytotaxonomica et Geobotanica 41: 109-128.
Pak, J.-H. & S. Kawano. 1992. Biosystematic Studies on the Genus Ixeris and its Allied Genera
(Compositae-Lactuceae). IV. Taxonomic Treatments and Nomenclature. Memoires of the
Faculty of Sciences of the Kyoto University (Series Biology) 15: 29-61.
Pak, J.H., Park, J.K., & Whang, S.S. 2001. Systematic Implications of Fruit Wall Anatomy and
Surface Sculpturing of Microseris (Asteraceae, Lactuceae) and Relatives. International Journal
of Plant Sciences 162(1): 209-220.
Parker, J.S. 1975. Aneuploidy and isolation in two Hypochaeris species. Chromosoma 52: 89-101.
Petrov, D.A. 2002. DNA Loss and Evolution of Genome Size in Drosophila. Genetica 115: 81-91.
Pirotta, P.P. 1903-04: Flora della colonia Eritrea. Roma.
Plitmann, U., Raven, P.H., Tai, W. & Breedlove, D.E. 1975. Cytological studies in Lopezieae
(Onagraceae). Botanical Gazette 136: 322-332.
Popov, M.G. 1957-1959. Flora srednej Sibiri. Izdatel'stvo AN SSSR, Moskva.
Puff, C. & Nemomissa, S. 2001: The Simen Mountains (Ethiopia): comments on plant biodiversity,
endemism, phytogeographical affinities and historical aspects. Systematics and Geography of
Plants 71: 975-991.
Puff, C. & Nemomissa, S. 2005: Plants of the Simen: a flora of the Simen Mountains and
surroundings, northern Ethiopia. Scripta Botanica Belgique 37.
Qiu, Y.L., Lee, J.L., Bernasconi-Quadroni, F., Soltis, D.E., Soltis, P.S., Zanis, M., Zimmer, E.A.,
Chen, Z., Savolainen, V., & Chase, M.W. 1999. The Earliest Angiosperms: Evidence from
Mitochondrial, Plastid and Nuclear Genomes. Nature 402:404-407.
Raina, S.N. & Rees, H. 1983. DNA Variation Between and Within Chromosome Complements of Vicia
Species. Heredity 51: 335-346.
Reid, C. & Reid, E.M. 1916. The Pliocene Floras of the Dutch-Prussian Border. Mededelingen van de
Rijksopsporingsdiensit voor Delftstoffen 6: 1-178.
Rieseberg, L.H. & Soltis, D.E. 1991. Phylogenetic Consequences of Cytoplasmatic Gene Flow in
Plants. Evolutionary Trends in Plants 5:65-84.
Riley, R., Chapman, V., & Kimber, G. 1959. Genetic Control of Chromosome Pairing in Intergeneric
Hybrids with Wheat. Nature 183: 1244-1246.
Robertson, W.R. B. 1916. Chromosome studies. I. Taxonomic relationships shown in the
chromosomes of Tettigidae and Acrididae. V-shaped chromosomes and their significance in
Acrididae, Locustidae and Gryllidae: chromosome and variation. Journal of Morphology
27:179-331.
Ronquist, F. 1997. Dispersal-Vicariance Analysis: A New Approach to the Quantification of Historical
Biogeography. Systematic Biology 46(1): 195-203.
Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed
models. Bioinformatics 19: 1572-1574.
Samuel, R., Stuessy, T.F., Tremetsberger, K., Baeza, C.M. & Siljak-Yakovlev, S. 2003.
Phylogenetic Relationships among species of Hypochaeris (Asteraceae, Cichorieae) based on
ITS, plastid trnL intron, trnL-F spacer, and matK sequences. American Journal of Botany 90:
496-507.
SanMiguel, P. & Bennetzen, J.L. 1998. Evidence That a Recent Increase in Maize Genome Size was
Caused by the Massive Amplification of Intergene Retrotransposons. Annals of Botany
(Supplement A) 82: 37-44.

121
 9. LITERATURE CITED

Sanderson, M. J. 2002. Estimating absolute rates of molecular evolution and divergence times: a
penalized likelihood approach. Molecular Biology and Evolution 19:101-109.
Sang, T. 2002. Utility of Low-Copy Nuclear Gene Sequences in Plant Phylogenetics. Critical Reviews
in Biochemistry and Molecular Biology 37:121-147.
Schifino, M.T. & Moraes Fernandes, M.I.B. 1987. Cytological Evidence of Genetic Control of
Chromosome Pairing in Artificial Autotetraploids of Trifolium riograndense Burkart. Genetica
74(3): 225-230.
Schmidt, T. & Heslop-Harrison, J.S. 1998. Genomes, Genes, and Junk: The Large Scale
Organization of Plant Chromosomes. Trends in Plant Science. 3: 195-199.
Schulz Bipontinus, C. H. 1866: Beitrag zum System der Cichorieen. Jahresberichte Pollichia 22-24:
296-322.
Schwarzacher, T., Ambros, P., & Schweizer, D. 1980. Application of Giemsa Banding to Orchid
Karyotype Analysis. Plant Systematics and Evolution 134:293-297.
Sebald, O. 1968: Bericht über botanische Studien und Sammlungen am Tana-See und im Semyen-
Gebirge (Äthiopien). Ergebnisse der botanischen Reise Oskar Sebald im Jahre 1966 nach
Äthiopien, Nr. 2. - Stuttgarter Beitr. Naturk. 194.
Sell, P.D. 1976. Crepis. In: Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H.,
Walters, S.M, & Webb, D.A. (eds.) Flora Europaea IV. Pp 344-357.
Sennikov, A.N. & Illarionova, I.D. 2007. Generic Delimitation of the Subtribe Ixeridinae newly
Segregated from Crepidiinae (Asteraceae - Lactuceae). Komarovia 5(2): 57-115.
Sherman, M. 1946. Karyotype Evolution: A Cytogenetic Study of Seven Species and Six Intraspecific
Hybrids of Crepis. University of California Publications in Botany. 18: 369-408.
Siljak-Yakovlev, S. & Cartier, D. 1982. Comparative Analysis of C-Band Karyotypes in Crepis
praemorsa subsp. praemorsa and subsp. dinarica. Plant Systematics and Evolution. 141(2):
85-90.
Simpson, G.G. 1944. Tempo and Mode in Evolution. Columbia University Press , New York.
Sims, L.E. & Price, H.J. 1985. Nuclear DNA Content Variation in Helianthus (Asteraceae). American
Journal of Botany. 72: 1213-1219.
Small, R.L., Cronn, R., & Wendel, J.F. 2004. Use of Nuclear Genes for Phylogeny Reconstructions in
Plants. Australian Systematic Botany17:145-170.
Soltis, D. E., Soltis, P.S., Chase, M.W., Mort, M.E., Albach, D.C., Zanis, M., Savolainen, V., Hahn,
W.H., Hoop, S.B., Fay, M.F., Axtell, M., Swensen, S.M., Prince, L.M., Kress, W.J., Nison,
K.C., & Farris, J.S. 2000. Angiosperm Phylogeny Inferred from 18S rDNA, rbcL, and atpB
Sequences. Botanical Journal of the Linnean Sociecy 133:381-461.
Soltis, D.E. & Soltis, P.S. 1999. Polyploidy: Origins of Species and Genome Evolution. Trends in
Ecology and Evolution 9:348-352.
Soltis, D.E., Soltis, P.S., Bennett, M.D., & Leitch, I.J. 2003. Evolution of Genome Size in the
Angiosperms. American Journal of Botany 90(1): 1596-1603.
Stebbins, G. L. 1940. Studies in Cichorieae: Dubyaea and Soroseris. Endemics of the Sino-
Himalayan region. Memoirs of the Torrey Botanical Club 19(3).
Stebbins, G.L. 1950. Variation and Evolution in Plants. Columbia University Press, New Yorck.
Stebbins, G.L. 1953. A new classification of the tribe Cichorieae, family Compositae. Madroño 12: 65-
81.
Stebbins, G.L., Jenkins, J.A. & Walters, M.S. 1953. Chromosomes and Phylogeny in the
Compositae, Tribe Cichorieae. University of California Publications in Botany 26(6): 401-430.

122
 9. LITERATURE CITED

Strand, A.E., LeebensMack, J., & Milligan, B.G. 1997. Nuclear DNA-based Markers for Plant
Evolutionary Biology. Molecular Ecology 6:113-118.
Sudhaus, W. 2007. Die Notwendigkeit morphologischer Analysen zur Rekonstruktion der
Stammesgeschichte. Species, Phylogeny and Evolution 1:17-32.
Swofford, D.L. 2002. PAUP*: Phylogenetic Analyses Using Parsimony (* and Other Methods), version
4.0 Beta. Sinauer, Sunderland.
Tadesse, M. 1999: New combinations, varieties and synonyms in African Compositae. Compositae
Newsletter 33: 23-32.
Tadesse, M. 2004: Asteraceae (Compositae). In: Hedberg, I., Friis, I. & Edwards, S. (ed.), Flora of
Ethiopia and Eritrea 4(2). Addis Ababa & Uppsala.
Tegel, F. 2002. Die Testaepidermis der Lactuceae (Asteraceae) - ihre Diversität und systematische
Bedeutung. Doctorate Thesis, Ludwigs-Maximillians-University of Munich. http://edoc.ub.
unimuenchen.de/archive/00000104/
Temsch, E.M. & Greilhuber, J. 2001. Genome Size In Arachis duranensis: A Critical Study. Genome
44: 826-830.
Thomas, C.A. 1971. The Genetic Organisation of Chromosomes. Annual Review of Genetics 5: 237-
256.
Tobgy, H.A. 1943. A Cytological Study of Crepis fuliginosa, C. neglecta And Their F1 Hybrid, And its
Bearing on the Mechanism of Phylogenetic Reduction in Chromosome Number. Journal of
Genetics 45:67-111.
Torrell, M. & Vallès, J. 2001. Genome Size in 21 Artemisia Species (Asteraceae, Anthemideae):
Systematic, Evolutionary, and Ecological Implications. Genome 44(2):231-238.
Torrell, M., Garcia-Jacas, N., Susanna, A. & Vallès, J. 1999. Phylogeny in Artemisia (Asteraceae,
Anthemideae) inferred from nuclear ribosomal DNA (ITS) sequences. Taxon 48: 721-736.
Torres, C. & Galetto, L. 2007. Style Morphological Diversity of Some Asteraceae Species from
Argentina: Systemtic and Functional Implications. Journal of Plant Research 120: 359-364.
Tournefort, J. P. de 1694. Élemens de botanique 1–3. Imprimerie Royale, Paris.
Tremetsberger, K. , Gemeinholzer, B., Zetzsche, H., Blackmore, S., Kilian, N., Talavera, S. &
Stuessy, T.F. subm. Fossil-calibrated Relaxed Molecular Clock Reveals Biogeographic
History of Cichorieae (Asteraceae) in Spite of Topological Uncertainties.
Turpeinen, T., Kulmala. J., & Nevo, E. 1999. Genome Size Variation In Hordeum spontaneum
Populations. Genome 42(6):1094-1099.
Varela, E.S., Lima, J.P.M.S., Galdino, A.S., Pinto, L.S., Bezerra, W.M., Nunes, E.P., Alves, M.A.O.
& Grangeiro, T.W. 2004. Relationships in the subtribe Diocleinae (Leguminosae;
Papilionoideae) inferred from internal transcribed spacer sequences from nuclear ribosomal
DNA. Phytochemistry 65: 59-69.
Vendrely, R. 1955. The Deoxyribonucleic Acid Content of the Nucleus. Pp. 155-180. In: Chargaff, E.
& Davidson, J.N. (ed.) The Nucleic Acids II. Academic Press, New York.
Verboom, G.A., Linder, H.P., & Stock, W.D. 2004. Testing the adaptive nature of radiation: growth
form and life history divergence in the African grass genus Ehrharta (Poaceae:
Ehrhartoideae). American Journal of Botany 91: 1364-1370.
Vicient, C.N., Suoniemi, A., Anamthamat-Jonsson, K., Tanskanene, J., Beharav, A., Nevo, E., &
Schulman, A.H. 1999. Retrotransposon BARE-1 And Its Role In Genome Evolution In The
Genus Hordeum. Plant Cell 11:1769-1784.
Wallace, H., Sparkes, C.A. & Maden, M. 1972. Nuclear DNA Content of Three Crepis Species.
Heredity 29: 267-373.

123
 9. LITERATURE CITED

Watanabe, K. 2008: Index to chromosome numbers in Asteraceae. Published on the internet:

http://www-asteraceae.cla.kobe-u.ac.jp/index.html.
Watanabe, K., Yahara, T., Denda, T. & Kosuge, K. 1999. Chromsomal evolution in the genus
Brachyscome (Asteracea, Asterae): Statistical tests regarding correlation between changes in
karyotype and habit using phylogenetic information. Journal of Plant Research 112: 145-161.
Weber, W.A. 1984. New names and combinations, principally in the Rocky Mountain Flora IV.
Phytologia 55: 1-11.
Wendel, J.F. 2000. Genome Evolution in Polyploids. Plant Molecular Biology. 42: 225-249.
Wendel, J.F., Cronn, R.C., Johnston, J.S., & Price H.J. 2002. Feast and famine in plant genomes.
Genetica 115: 37-47.
Whitton, J., Wallace, R.S. & Jansen, R.K. 1995. Phylogenetic relationships and patterns of character
change in the tribe Lactuceae (Asteraceae) based on chloroplast DNA restriction site variation.
Canadian Journal of Botany 73: 1058-1073.
Yan, G., Yao, J., Ferguson, A.R., McNeilage, M.A., Seal, A.G., & Murray, B.G. 1997. New Reports
of Chromosome Numbers in Actinidia (Actinidiaceae). New Zealand Journal of Botany. 35:
181-186.
Zanis, M., Soltis,D.E., Soltis,P.S., Mathews,S. & Donoghue, M.J. 2002. The Root of the
Angiosperms Revisited. Proceedings of the National Academy of Science 99:6848-6853.
Zarembo, E.V. & Boyko, E.V. 2008. Carpology of Some East Asian Cardueae (Asteraceae). Anales
del Jardin Botanico de Madrid 65(1):129-134.
Zhang, D.X. & Hewitt, G.M. 2003. Nuclear DNA Analysis in Genetic Studies of Populations: Practice,
Problems and Prospects. Molecular Evolution 12:563-584.
Zhu, S.X., Qin, H.N., & Shih, C. 2006. Achene Wall Anatomy and Surface Sculpturing of Lactuca L.
and Related Genera (Compositae: Lactuceae) with Notes on Their Systematic Significance.
Journal of Integrative Plant Biology 48(4): 390-399.
Zidorn, C. 2008. Sesquiterpene Lactones and Their Precursors as Chemosystematic Markers in the
Tribe Cichorieae of the Asteraceae. Phytochemistry doi:10.1016/j.phytochem.2008.06.013
Zuckerkandl, E. 2002. Why So Many Noncoding Nucleotides? The Eukaryote Genome as an
Epigenetic Machine. Genetica 115: 105-129.

124
 10. APPENDIX

APPENDIX

10.1 PUBLICATION LIST

10.1.1 Publications in Journals

Enke, N., Kilian, N., Nemomissa, S., & Gemeinholzer, B. 2008. Afro-alpine Dianthoseris
actually a congener of Crepis s.str. (Compositae, Cichorieae, Crepidinae). Botanische
Jahrbücher 127(3): 389-405.

Enke, N. & Gemeinholzer, B. 2008. Babcock revisited: new insights into generic delimitation
and character evolution in Crepis L. (Compositae: Cichorieae) from ITS and matK sequence
data. Taxon 57(3): 756-768.

Enke, N. Fuchs, J. & Gemeinholzer, B. Shrinking Genomes? Evidence from Genome Size
Evolution in Crepis L. (Cichorieae, Compositae). Molecular Biology and Evolution (subm).

10.1.2 Talks and Posters at National and International Conferences

10.1.2.1 Talks

Enke, N. 2006. Character Evolution within Crepis L. (Compositae) – First Insights. Royal
Botanic Garden Edinburgh, Scotland.

Enke, N. & Gemeinholzer, B. 2007. First Insights into Speciation Processes in Crepis L.
(Compositae). GfBS Jahrestagung 2007, Vienna, Austria. (Award for Best Student
Presentation).

Enke, N. & B. Gemeinholzer, B. 2007. Contributions Towards a New Generic and

Infrageneric Classification Delimitation of Crepis L. (Compositae). Young Systematists
Forum, London, England.

Enke, N. 2008. Molekulare Phylogenie und Morphologie: Der lange Weg zu einer
Neugliederung der Gattung Crepis L. (Cichorieae, Compositae). Invited Talk at the
Colloquium of Organismic Biology, Phillips-University, Marburg, Germany.

Enke, N. & Gemeinholzer, B. 2008. Mechanisms of Speciation: Hybridisation and Karyotype

Evolution in Crepis L. (Cichorieae, Compositae). Systematics 2008, Göttingen, Germany.

Enke, N. & Gemeinholzer, B. 2008. Karyotype Evolution and Speciation in Higher Plants:
The Genus Crepis L. (Cichorieae, Compositae). Botany 2008, Vancouver, Canada.

125
 10. APPENDIX

10.1.2.2 Posters

Gemeinholzer, B., Enke, N. & Jahn, R. 2007. Establishing DNA-Barcoding Methods in

Diatoms for Diversity Assessments. Biodiversity Infomatics and the Barcode of Life, Aarhus.

Enke, N. & Gemeinholzer, B. 2008. Possible hybrid origin of Asian/American alpine genus
Askellia W. A. Weber (Cichorieae, Compositae). Xth Symposium of the International
Organisation of Plant Biosystematists, Strbske Pleso, Slovakia.

126
 10. APPENDIX

10.2 ERKLÄRUNG ÜBER DEN PERSÖNLICHEN ANTEIL AN DEN PUBLIKATIONEN

Kapitel 2 - Babcock revisited: New Insights into Generic Delimitation and Character
Evolution in Crepis L. (Compositae: Cichorieae) from ITS and matK Sequence Data.

Datenerfassung:
- Eigenanteil der Autorin: Recherche nach geeigneten Herbarbelegen in Herbaren B,
E, C und CGE, Nachbestimmen der Herbarbelege, Recherche für geeignete
Außengruppen Sequenzen in der GeneBank, Extraktion der DNA, PCR, Aufreinigen
der PCR Produkte, Sequenzierung, Editierung und Alignierung der Sequenzen.
- Bereitstellung von Herbarbelegen aus M und W, Beratung beim Primerdesign und
Anpassen der PCR Protokolle durch B. Gemeinholzer.

Datenauswertung:
- Eigenanteil der Autorin: Rekonstruktion der Phylogenien basierend auf ITS und matK
mittels Maximum Parsimony und Bayesischer Statistik, Interpretation der Daten.

Manuskript:
- Eigenanteil der Autorin: Erstellung des englischen Manuskripts.
- Korrekturen durch B. Gemeinholzer.

Kapitel 3 – Shrinking Genomes? Evidence from Genome Size Variation in Crepis L.

(Compositae).

Datenerfassung:
- Eigenanteil der Autorin: Samenaufsammlungen in den Gebirgen Südeuropas, DNA
Extraktion zusätzlicher Proben, PCR, Aufreinigen der PCR-Produkte, Sequenzierung,
Editierung und Alignierung der Sequenzen, Recherche nach bereits veröffentlichten
Genomgrößen in der C-Value Database, Auswahl geeigneter Pflanzen und
Vorbereitung der Proben für die Genomgrößenmessung.
- Samenaufsammlung in Sibirien gemeinsam mit B. Gemeinholzer.
- Messung der Genomgrößen am FacStarPLUS gemeinsam mit J. Fuchs.

Datenauswertung:
- Eigenanteil der Autorin: Phylogenierekonstruktion mittels Maximum Parsimony
basierend auf ITS, Rekonstruktion der Ancestral Character States in Mesquite,
Interpretation der Daten.

Manuskript:
- Eigenanteil der Autorin: Erstellung des englischen Manuskripts.
- Korrekturen durch B. Gemeinholzer.

127
 10. APPENDIX

Kapitel 6 - Afroalpine Dianthoseris actually a congener of Crepis s.str. (Compositae,

Cichorieae, Crepidinae)

Datenerfassung:
- Eigenanteil der Autorin: Auswahl geeigneter DNA Sequenzen, DNA Extraktion
zusätzlicher Proben, PCR, Aufreinigen der PCR-Produkte, Sequenzierung, Editierung
und Alignierung der Sequenzen.
- Erfassen morphologischer, geographischer und karyologischer Daten durch N. Kilian,
S. Nemomissa und B. Gemeinholzer

Datenauswertung:
- Eigenanteil der Autorin: Rekonstruktion der Phylogenien basierend auf ITS und matK
mittels Maximum Parsimony und Bayesischer Statistik, Interpretation der molekularen
Daten.
- Taxonomische Umkombinierung gemeinsam mit N. Kilian.
- Auswertung morphologischer, geographischer und karyologischer Daten durch N.
Kilian, S. Nemomissa und B. Gemeinholzer.

Manuskript:
- Eigenanteil der Autorin: Erstellung und Endfassung des englischen Manuskripts
(Methoden/Ergebnisse/Diskussion) für den molekularen Bereich.
- Endfassung des englischen Gesamtmanuskripts durch N. Kilian.

128
 10. APPENDIX

10.3 ACKNOWLEDGEMENTS

I thank

- Dr. B. Gemeinholzer for the initiation of the project and her encouragement, the
fruitful discussions and the excellent supervision during the past three years,
- Prof. Dr. W. Greuter and Prof. Dr. Th. Borsch for mentoring the presented doctoral
thesis,
- J. Zimmermann, M. Krummel and A. Bergfeld for practical help with the labwork,
- M. Meyer, H. Kanda, and C. Ludwig for raising and looking after my plants,
- M. Lüchow, J. Bansemer and M. Beer for technical support,
- Th. Dürbye for advice and help on seed collection,
- J. Baumeister and M. Enke for helping with the seed collection,
- S. Smirnov for guidance through the Siberian flora,
- H. Enke for extensive advise on computer eccentricities,
- D. Lauterbach for advice on statistics,
- N. Kilian for fruitful discussions on Compositae,
- W.-H. Kusber for advice on taxonomic and nomenclatural problems,
- R. Hand, Ch. Zidorn, E. v. Raab-Straube and numerous other collectors for providing
plant material,
- Prof. S. Blackmore, Dr. M. Watson, F. Christie, and M. Hollingsworth for their support
during my stay at the Royal Botanic Garden Edinburgh,
- Prof. Dr. H.C. Weber, M. Rath and K. Dörr for their support during my work at the
Philipps-University Marburg,
- Dr. J. Fuchs for the technical support and realisation of the genome size
measurements at the IPK Gatersleben,
- my collegues N. Abarca, Th. Dürbye, L. Ferrufino, R. Jahn, M. Krummel, W.-H.
Kusber, D. Lauterbach, N. Mandel, A. Schories, H. Zetzsche, and J. Zimmermann for
their moral support,
- my parents ME and HE, my family SP, AS and JZ as well as my friends NA, JB, MB,
LF, PL, MR and AS for bearing with me and keeping me sane and healthy. I thank
them for their trust in me and their never ending encouragement.

129