See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/313212681

The evolutionary history of flowering plants

Article in Journal and proceedings of the Royal Society of New South Wales · December 2016
DOI: 10.5962/p.361755

CITATIONS READS

6 12,700

1 author:

Charles Foster
The University of Sydney
56 PUBLICATIONS 1,551 CITATIONS

SEE PROFILE

All content following this page was uploaded by Charles Foster on 02 February 2017.

The user has requested enhancement of the downloaded file.

Journal & Proceedings of the Royal Society of New South Wales, vol. 149, parts 1 & 2, 2016,
pp. 65-82. ISSN 0035-9173/16/010065-18
The evolutionary history of flowering plants
Charles S.P. Foster1
1 School of Life and Environmental Sciences, University of Sydney, New South Wales 2006, Australia
This paper was an RSNSW Scholarship Winner in 2015

Email: charles.foster@sydney.edu.au

Abstract
In terms of species richness and important ecological roles, there are few biological groups that
rival the success of flowering plants (Angiospermae). Angiosperm evolution has long been a
topic of interest, with many attempts to clarify their phylogenetic relationships and timescale of
evolution. However, despite this attention there remain many unsolved questions surrounding
how and when flowers first appeared, and much of the angiosperm diversity remains to be
quantified. Here, I review the evolutionary history of angiosperms, and how our
understanding of this has changed over time. I begin by summarising the incredible
morphological and genetic diversity of flowering plants, and the ways in which this can be
studied using phylogenetic inference. I continue by discussing both the relationships between
angiosperms and the other major lineages of seed plants, and the relationships between the
main groups within angiosperms. In both cases, I outline how our knowledge has changed
over time based on factors such as the different conclusions drawn from morphological and
genetic data. I then discuss attempts to estimate the timescale of angiosperm evolution and
the difficulties of doing so, including the apparent conflict between ages derived from fossil
and molecular evidence. Finally, I propose future directions for angiosperm research to help
clarify the evolutionary history of one of the most important groups of organisms on the
planet.

atmosphere, and act as primary producers of

Introduction
food for many animal groups, with their
The diversity and interactions of life on spread and appearance shaping habitat
Earth have long been of scientific interest. structure globally (Brodribb and Feild 2010;
Quantifying biodiversity and the timescale Magallón 2014). In addition, angiosperms
over which it arose allows inferences about have developed important mutualistic
the biological history of the planet to be relationships with many groups of
made, and can provide insight into how organisms, such as pollination interactions
ecosystems might change in response to with insects, birds, and small mammals (van
events such as climate change (Thuiller et al. der Niet and Johnson 2012; Rosas-Guerrero
2011; Bellard et al. 2012). Flowering plants et al. 2014).
(angiosperms) have been of particular focus However, to properly quantify the
because of their important economic and extent and impact of groups such as
cultural roles within society, as well as their angiosperms, biological entities must first be
ubiquity and importance within natural recognised and described into distinct groups
ecosystems. Specifically, angiosperms such as species, and, ideally, placed into
sequester large amounts of carbon from the higher-order classifications. The goal is to

65
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
recognise groups that contain only the diversity of angiosperms. Angiosperms are
descendants of a common evolutionary among the most species-rich groups of
ancestor (monophyletic groups), which organisms on the planet, and are by far the
represent natural evolutionary groups. largest group of plants. The exact number of
For most of history, biological groups species is difficult to determine because of
and the relationships between them have high amounts of taxonomic synonymy, and
been recognised through observations of the the fact that many species potentially remain
form and structure of organisms. When to be discovered (Bebber et al. 2010; Pimm
these data are shared between two or more and Joppa 2015). Despite this, we can be
taxa after being inherited from their most fairly certain that there are at least 350,000
recent common ancestor, they are known as species of angiosperms, and probably c.
synapomorphies. In addition to aiding the 400,000 in total (Pimm and Joppa 2015). As
classification of extant taxa, these expected in a group of this size, there is
morphological data are also able to link extreme variation in morphology, life history
extant and extinct diversity through characteristics, and growth form.
comparison with the fossil record, which can Angiosperms variously exist as herbaceous
suggest a timescale of evolution. However, annuals, vines, lianas, shrubs or trees, and
morphological data often cannot reliably can be found growing in aquatic or terrestrial
distinguish between competing taxonomic environments, or even growing on and/or
hypotheses because of a lack of informative parasitising other plants.
characters, or can be misled by the Similarly, there is large variation in
independent evolution of similar traits in genome size and content within
organisms that are not closely related angiosperms. For example, it is estimated
(convergent evolution). Morphological data that throughout their evolutionary history
have been supplemented by molecular data over 70% of angiosperms have had an
since the inception of molecular increase in the number of copies of
phylogenetics in the mid-20th Century. chromosomes contained within each cell
Molecular data typically comprise (ploidy level) from the typical diploid state
sequences of the nucleotides of DNA, or the (Levin 2002). Most of the functions essential
amino acids that they encode. Each for growth and development are controlled
nucleotide or amino acid within a sequence by genes located within the cell nucleus,
represents a character that can be used for which are collectively known as the nuclear
phylogenetic analysis. Therefore, molecular genome. Paris japonica Franch., a small
data sets can contain millions of characters herbaceous plant native to Japan, has the
for phylogenetic reconstruction, which largest accurately measured genome known
makes such data sets especially useful for to science (Pellicer et al. 2010). At nearly 150
evaluating the taxonomic hypotheses that billion nucleotides, its octoploid genome is
have been suggested by morphology. more than 50 times larger than the human
Analysis of molecular data is also useful for genome, and nearly 2500 times larger than
estimating the evolutionary timescale of the smallest known plant nuclear genome of
organisms using molecular clocks (Lee and Genlisea tuberosa Rivadavia, Gonella &
Ho 2016), especially for groups with poor A.Fleischm., a carnivorous angiosperm from
fossil records. Brazil (Fleischmann et al. 2014).
Both morphological and molecular data Plant cells also contain specialised
have been used extensively to evaluate the organelles known as chloroplasts and

66
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
mitochondria, which are responsible for the diverse group arose over a supposedly short
essential processes of photosynthesis and period of time. Indeed, the traditional view
cellular respiration, respectively. Both of is that angiosperms originated in the early
these organelles are predominantly Cretaceous. The subsequent appearance of
uniparentally inherited and contain their own fossils with highly diverse morphologies,
independent genomes, which is thought to over what was apparently an extremely rapid
be because of their origins as free-living timescale, was famously described by Darwin
organisms that were engulfed by early as an “abominable mystery” in a letter to
eukaryotic cells in separate endosymbiotic Joseph Hooker in 1879 (first published in
events (Sagan 1967; Schwartz and Dayhoff Darwin and Seward 1903).
1978). The chloroplast genome varies To understand fully the evolutionary
substantially among angiosperms, with the history of angiosperms, their diversity needs
order of genes differing between groups, and to be characterised in a phylogenetic context.
with some genes being lost completely. For This approach indicates whether key traits
example, the chloroplast genome is for success are clade-specific, or have
drastically reduced in many parasitic plants, evolved multiple times in parallel.
with many genes important for Additionally, incorporating temporal
photosynthesis having been lost (Bungard information into these analyses can allow
2004). inferences to be made about the
The mitochondrial genome of plants is environmental conditions that might have
more enigmatic, and is disproportionally less driven angiosperm diversification.
studied than the nuclear and chloroplast In this review, I begin by discussing our
genomes. Plant mitochondrial genomes are understanding of the relationships among
large compared with animal mitochondrial the major seed plant lineages, and the
genomes, and their content is highly importance of this for reconstructing the
dynamic, with many gene gains, losses, origin of flowers. I then discuss the
transfers, duplications and rearrangements, relationships of the major lineages within
as well as a large proportion of repeated Angiospermae, and examine estimates of the
elements and introns (Kitazaki and Kubo evolutionary timescale of angiosperms. I
2010; Galtier 2011). Of direct importance propose a number of the future directions
for reconstructing the evolutionary history of that are likely to improve our understanding
plants is that the three genomes evolve at of the evolutionary history of angiosperms.
very different rates. The nuclear genome
evolves at the highest rate, the chloroplast
Higher relationships of angiosperms
genome evolves at an intermediate rate, and,
in contrast to its dynamic nature, the
and the origin of flowers
mitochondrial genome has by far the lowest Angiosperms are recognised as members of
evolutionary rate (Wolfe et al. 1987). the superdivision Spermatophyta along with
The global dominance of angiosperms cycads, conifers, gnetophytes, and Ginkgo.
indicates that they are ideally adapted to exist The last four extant cone-bearing lineages are
within many different habitats, and their known as acrogymnosperms, whereas extant
great morphological and genomic variation and extinct cone-bearing lineages combined
suggests a history of varied selective are known as gymnosperms (Cantino et al.
pressures. This has long challenged those 2007). The five extant spermatophyte
who have sought to quantify how such a lineages are linked by the production of

67
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
seeds. Estimates of the number of seed evaluating the seed plant phylogeny,
plant species vary, but are consistently in the including determining the sister lineage to
region of many hundred thousand species angiosperms, using comparative morphology
(Govaerts 2001; Scotland and Wortley 2003). to assess homology of the reproductive and
Among other potential factors, the success vegetative structures of the seed plant
of these lineages is perhaps due to the lineages (e.g., Doyle and Donoghue 1986).
diversification of regulatory genes important One major hope was that determining
for seed and floral development following the sister lineage to angiosperms might prove
ancient whole-genome duplication events especially useful for inferring the origin and
along the lineages leading to seed plants and structure of the first flowers. Throughout
angiosperms (Jiao et al. 2011). the 20th century, the two main hypotheses
Angiosperms can be readily for the origin of flowers were that they
distinguished from gymnosperms through a evolved from branched, unisexual
suite of synapomorphies. These include the reproductive structures found in most
presence of flowers with at least one carpel, gymnosperms ("pseudanthial" theory,
which develop into fruit (cf. the “naked” Wettstein 1907), or that flowers evolved
seeds of gymnosperms); stamens with two from bisexual, flower-like structures, such as
pairs of pollen sacs (cf. the larger, heavier in the extinct group Bennettitales
corresponding organs of gymnosperms); a ("euanthial" theory, Arber and Parkin 1907).
range of features of gametophyte structure The inferred homology of morphological
and development, including drastically structures consistently suggested that
reduced male and female gametophytes gnetophytes were the extant sister lineage to
compared with gymnosperms; and phloem angiosperms, with several potential close
tissue with sieve tubes and companion cells (non-angiosperm) fossil relatives.
(cf. sieve cells without companion cells in Specifically, various features of wood
gymnosperms) (Doyle and Donoghue 1986; anatomy and flower-like structures seemed
Soltis and Soltis 2004). The production of to suggest a close relationship between
endosperm through double fertilisation was angiosperms, gnetophytes, and the extinct
previously considered to be a further order Bennettitales, with this group being the
synapomorphy of angiosperms, but this sister lineage to the rest of the gymnosperms
phenomenon has also been observed in (Crane 1985; Doyle and Donoghue 1986).
some gnetophyte lineages (Friedman 1992; Therefore, based on the strength of
Carmichael and Friedman 1996). morphological evidence, the euanthial theory
Collectively, the synapomorphies of was the most popular view in the 20th
angiosperms are thought to be responsible Century.
for providing the evolutionary advantages The acceptance of the euanthial theory,
that led to their global dominance, which coupled with the predominance of
coincided with a decline in gymnosperm Cretaceous Magnolia-like fossils at the time,
diversity (Bond 1989). However, to led to suggestions that the ancestral flowers
reconstruct the evolution of these characters were similar to present-day magnolias. This
and evaluate their importance for implies that magnolias and their close
angiosperm evolution, it is necessary to relatives were some of the earliest-diverging
determine which lineage of seed plants is angiosperm lineages (Endress 1987).
most closely related to angiosperms. The However, most molecular phylogenetic
majority of earlier studies focused on studies from the 1990s onwards have

68
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
recovered different relationships between the followed Marcello Malpighi in referring to
extant seed plant lineages. The dominant these leaves as ‘cotyledons’. Accordingly,
theme in these modern studies is that all flowering plants with one cotyledon have
extant gymnosperm lineages form a subsequently been referred to as
monophyletic sister group to angiosperms monocotyledons or ‘monocots’, and those
(Chaw et al. 1997; Bowe et al. 2000; Chaw et with two cotyledons have been called
al. 2000; Ruhfel et al. 2014; Wickett et al. dicotyledons or ‘dicots’.
2014) (Figure 1). Particularly strong evidence Although the most widely known early
has emerged for a close relationship between classification scheme by Linnaeus was based
gnetophytes and conifers (Qiu et al. 1999; solely on floral reproductive characters, the
Winter et al. 1999). Indeed, the evidence division into monocots and dicots has since
seems to suggest that gnetophytes might been recognised as an important diagnostic
even be nested within conifers and the sister feature to inform classification, with varying
group to Pinaceae (Bowe et al. 2000; Chaw et implications for the angiosperm phylogeny.
al. 2000; Zhong et al. 2010). A minority of early authors argued that some
Overall, because none of the extant key morphological differences between
gymnosperm lineages is more closely related monocots and dicots, such as vascular
to angiosperms than to other gymnosperms, bundle anatomy, were irreconcilable with a
they cannot directly inform hypotheses on monophyletic origin of angiosperms.
the homologies of angiosperm characters, or Instead, these authors argued that
on the sequence of development of these angiosperms should be recognised as a
characters (Doyle 2012). Therefore, while polyphyletic group (= derived from more
the relationships among the major seed plant than one common evolutionary ancestor)
lineages have been largely resolved, the (e.g., Meeuse 1972; Krassilov 1977).
structural origin of flowers, and the affinity However, the predominant view was that
of the earliest flowers to modern species, angiosperms are monophyletic, and the
remains controversial. Progress in this area division into monocots and dicots
is likely to be achieved through improved constitutes a natural split within flowering
understanding of the relationships among plants. This was echoed in many angiosperm
the major angiosperm groups. classification systems developed in the 20th
century, including the highly influential
Takhtajan (1980) and Cronquist (1981)
Major relationships within
systems.
Angiospermae To infer the evolutionary relationships
The major relationships within angiosperms within monocots and dicots, many cladistic
have historically proved difficult to analyses were undertaken in the latter half of
determine, and have long been in a state of the 20th century using pollen, floral, and
flux. This has largely been due to differing vegetative characters. This approach led to
ideas of the characters, initially many informal subgroups being proposed.
morphological but later molecular, needed to For example, Donoghue and Doyle (1989b)
reconstruct the angiosperm phylogeny. An recognised five major groups of
early discovery was that flowering plants angiosperms, corresponding to Magnoliales,
have either one or two embryonic leaves Laurales, Winteraceae-like plants,
(Ray 1686–1704). While John Ray was the ‘paleoherbs’ (‘primitive’ herbaceous lineages
first to observe this dichotomy, he later including water lilies and Amborella), and

69
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants

Figure 1: The relationships among seed plant lineages, scaled to geological time based on fossil
ages. Numbers in green circles refer to the following: (1) oldest Ginkgo fossil (Yang et al. 2008);
(2) oldest cycad fossil (Gao and Thomas 1989); (3) oldest gnetophyte fossil (Rydin et al. 2006);
(4) oldest conifer fossils (Wieland 1935); (5) oldest angiosperm fossils (discussed in Doyle
2012); (6) oldest acrogymnosperm fossil ; (7) an estimated maximum age for crown-group seed
plants (discussed in Magallón and Castillo 2009; Foster et al. 2016).

plants with tricolpate pollen. Although the monophyletic group (Doyle 1969).
constituent members of the subgroups Consequently, Doyle and Hotton (1991)
varied across studies, the recognition of chose to recognise tricolpates as distinct
tricolpates as a monophyletic group was a from the rest of the dicots, coining the term
consistent finding (e.g., Donoghue and ‘eudicots’ for this group.
Doyle 1989b; Donoghue and Doyle 1989a), Taxonomic concepts for the major
leading to suggestions that dicots had angiosperm groups have changed over time,
multiple evolutionary origins (Endress et al. which makes it difficult to chronicle
2000; Endress 2002). Indeed, stratigraphical concisely the changing opinions about the
studies in which triaperturate pollen earliest-diverging angiosperms. For example,
(tricolpate) fossils were consistently found to the group Magnoliidae now has a very
originate in younger sediments than both different circumscription compared with the
monocots and non-tricolpate dicots had past, so statements in earlier studies
already hinted that dicots did not form a regarding the relationships between

70
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
magnoliids and other groups might no choice of molecular markers. An influential
longer be applicable. Nevertheless, it is clear early attempt with molecular data to resolve
that the most common view historically was the seed plant phylogeny and, necessarily, to
that Magnolia-like flowers probably occupied determine the earliest-diverging angiosperm
a position at or near the root of the lineage, analysed sequences for the
angiosperm phylogeny. However, there chloroplast rbcL gene from nearly 500 seed
were other suggestions for the earliest- plant taxa using maximum parsimony (Chase
diverging angiosperm lineages, including et al. 1993). In this case, the widespread
Piperales+Chloranthales, several of the aquatic genus Ceratophyllum was found to be
lineages in the formerly recognised paleoherb the sister lineage to all other flowering plants.
group, or even monocots (Burger 1977, However, this has subsequently been found
1981). to be an anomalous result seemingly unique
Attempts to clarify the relationships to single-gene parsimony analyses of rbcL. A
within the angiosperm phylogeny have since series of studies in 1999 found that the
been greatly strengthened by the inclusion of monotypic genus Amborella is strongly
molecular data. Some aspects of early supported as being the sister lineage to all
classification schemes based on morphology other flowering plants (Mathews and
have been strongly supported by molecular Donoghue 1999; Parkinson et al. 1999; Qiu
data (reviewed by Endress et al. 2000; et al. 1999; Soltis et al. 1999), and this finding
Endress 2002). For example, the key has subsequently been supported by nearly
concepts of the monophyly of angiosperms, all large multigene analyses (Moore et al.
monocots and eudicots, the polyphyly of 2007; Soltis et al. 2011; but see Goremykin et
dicots, and the position of magnoliids as an al. 2013; Ruhfel et al. 2014; Wickett et al.
early diverging angiosperm lineage, were all 2014; Xi et al. 2014; Goremykin et al. 2015).
further supported by molecular data These studies have also revealed that the
(Endress et al. 2000). However, many base of the angiosperm phylogeny
molecular estimates of angiosperm constitutes a grade of several successive
evolutionary relationships have contradicted lineages, originally referred to as the ANITA
estimates based on morphological data. For (Amborella/Nymphaeales/Illiciaceae-
example, molecular data have firmly resolved Trimeniaceae-Austrobaileya) grade, but now
the family Hydatellaceae within known as the ANA
Nymphaeales, rather than within Poales as (Amborella/Nymphaeales/Austobaileyales)
former morphology-based studies had grade.
concluded (Saarela et al. 2007). Molecular The remaining ~99.95% of angiosperms
data have also helped to clarify the extent of are collectively referred to as
convergent evolution within angiosperms, Mesangiospermae. Within this group, five
such as C4 photosynthesis evolving major lineages are recognised: Chloranthales,
independently at least 60 times (Sage et al. Magnoliidae, Ceratophyllales, monocots, and
2011). eudicots (clade names here are standardised
Arguably the most important finding to Cantino et al. 2007). Unfortunately,
from analyses of molecular data has been the despite large increases in the amount of
rooting of the angiosperm phylogeny. available genetic data and improved analytical
Success was not immediate, with techniques, the relationships among these
disagreements being found among the results mesangiosperm groups have remained
of molecular analyses, depending on the uncertain (Figure 2). When analysing

71
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
chloroplast genome sequences, the most developmental biology (evo-devo) studies
common finding is that eudicots (Preston and Hileman 2009). However, to
+Ceratophyllum form the sister group to gain a fuller understanding of the
monocots, with these three lineages being evolutionary history of angiosperms, it is
the sister group to magnoliids necessary to know more than just the
+Chloranthales. Large nuclear DNA data relationships among the major flowering
sets, which have only become available in plant groups; a reliable estimate of the
recent years, tend to resolve different angiosperm evolutionary timescale is also
relationships. For example, they have needed.
supported a sister relationship between
eudicots and magnoliids+Chloranthales, with
Evolutionary timescale of
monocots being the sister group to these
three lineages (Wickett et al. 2014).
angiosperms
However, the number and choice of nuclear To understand how angiosperms came to
DNA markers can affect inferred dominance, including how the crucial
relationships within Mesangiospermae. For morphological traits that led to their success
example, analysis of a selection of 59 low- first evolved, it is necessary to have some
copy nuclear genes inferred a grouping of idea of the timescale of angiosperm
Ceratophyllum+Chloranthales and eudicots, evolution. Traditionally, the evolutionary
with successive sister relationships to timescale of organisms has been elucidated
magnoliids and monocots (Zeng et al. 2014). through study of the fossil record. In this
Additionally, the choice of phylogeny approach, the first appearance of each taxon
reconstruction method can lead to the in the fossil record, as determined by
estimation of different topologies (Xi et al. morphology, provides an indication of when
2014). it first evolved. When considering the fossil
Nevertheless, despite conflicting record, it is important to distinguish between
topologies sometimes being inferred, we “crown” and “stem” groups. A crown group
currently have an understanding of the is the least inclusive monophyletic group that
angiosperm phylogeny that is greater than at contains all extant members of a clade, as
any other time in history. The power of well as any extinct lineages that diverged after
molecular data to resolve the historically the most recent common ancestor of the
challenging relationships among flowering clade (Magallón and Sanderson, 2001). In
plants is now well established. In response contrast, a stem group is the most inclusive
to the rapid advances in the field, a monophyletic group that contains all extant
cosmopolitan consortium of researchers members of a clade, as well as any extinct
regularly collaborate to release timely lineages that diverged from the lineage
summaries of the state of knowledge of the leading to the crown group (Magallón and
angiosperm phylogeny (see Angiosperm Sanderson, 2001).
Phylogeny Group 1998, 2003, 2009, 2016). The fossil record of seed plants is
We now have a viable framework to allow ancient, with the oldest fossils of
fields related to phylogenetics to flourish and progymnosperms occurring in sediments
provide a greater understanding of the from the Late Devonian, ~365 million years
important evolutionary steps that have ago (Ma) (Fairon-Demaret and Scheckler
contributed to the overwhelming success of 1987; Rothwell et al. 1989; Fairon-Demaret
angiosperms, such as through evolutionary 1996).The fossil record of gymnosperms

72
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants

Figure 2: A comparison of several different estimates of the relationships among eudicots,

magnoliids, monocots, Ceratophyllum, Chloranthales, and ANA-grade angiosperms, based
on the comparison presented in Zeng et al. (2014). The different topologies represent
findings from studies using nuclear DNA (nrDNA), chloroplast DNA (cpDNA),
mitochondrial DNA (mtDNA), and a combination of morphological and molecular data.
A sample of suitable references for the topologies are as follows: (a) Zhang et al. (2012); (b)
Moore et al. (2011); Zeng et al. (2014); (c) Moore et al. (2007); Moore et al. (2010); Foster et
al. (2016); (d) Qiu et al. (2010); (e) Endress and Doyle (2009).

is rich, with fossils becoming common from Unfortunately,, the fossil record of
the Late Carboniferous to Early Triassic angiosperms is not as extensive or
(Magallón 2014), and revealing an extinct informative.
diversity far greater than the extant diversity.
The oldest known fossil that can be appeared in the Valanginian to early
confidently assigned to the stem group of Hauterivian (early Cretaceous, ~139.8–129.4
angiosperms has suggested that angiosperms Ma), albeit in sparse amounts, with vast
arose as early as 247.2–242.0 Ma (million amounts of angiospermous microfossils
years ago) (Hochuli and Feist-Burkhardt occurring by the Barremian (~129.4–125
2013). Accepted pollen fossils (microfossils) Ma) (Doyle 2012). There is a noticeable
suggest that crown-group angiosperms first disparity in the number and presence of

73
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
fossils between lineages, particularly at the biogeographic events and rate estimates from
family level and below, with many excellent other groups can be used as calibrations, but
fossils being present for some groups but these are subject to a wide range of errors
none for others (Magallón 2014). (Ho et al. 2015).
While fossil data have traditionally Collectively, molecular dating studies
provided the only source of information have yielded remarkably disparate estimates
about the evolutionary timescale of major for the age of crown-group angiosperms
groups, molecular dating techniques provide (summarised in Bell et al. 2010; Magallón
a compelling alternative, especially for groups 2014; Foster et al. 2016). Inferred ages have
that lack fossils. In these approaches, ranged from the extreme values of 86 Ma
evolutionary timescales can be estimated (when considering only the 3rd codon
using phylogenetic methods based on positions of rbcL; Sanderson and Doyle
molecular clocks. When the concept of the 2001) to 332.6 Ma (Soltis et al. 2002). Most
molecular clock was first proposed, age estimates fall between 140 and 240 Ma,
evolutionary change was assumed to but this still represents a substantial amount
correlate linearly with time and to remain of variation. Additionally, the earliest
constant across lineages (“strict” molecular analyses found that crown-group
clock) (Zuckerkandl and Pauling 1962). angiosperms were considerably older than
However, it has since become clear that implied by the fossil record, in some cases by
strictly clocklike evolution is the exception, more than 100 million years (e.g. Martin et al.
rather than the rule (Welch and Bromham 1989). Smaller disparities between molecular
2005). and fossil estimates were obtained in later
Rates of molecular evolution vary studies (e.g. Sanderson and Doyle 2001).
substantially across vascular plant lineages However, some more recent estimates have
(Soltis et al. 2002), and are often strongly tended to support a more protracted
correlated with life history strategies. For timescale for angiosperm evolution (e.g.
example, substitution rates in herbaceous Smith et al. 2010), echoing the results of the
annual lineages of angiosperms are known to earliest molecular studies.
be substantially higher than in woody Progress in molecular dating can be
perennial plants (Smith and Donoghue 2008; characterised in terms of increasing
Lanfear et al. 2013). Consequently, a variety methodological complexity and improving
of molecular clock models have been sampling of taxa and genes (Ho 2014). A
developed to account for evolutionary rate persistent problem, however, has been the
variation among lineages (Ho and Duchêne need for a trade-off between taxon sampling
2014). Fossil data are still intricately linked and gene sampling. Low gene sampling has
with these methods, because fossils are used been typical of studies of angiosperm
to provide temporal information to calibrate evolution, albeit with some other exceptions,
the molecular clock, thereby providing including the 12 mitochondrial genes
absolute rather than relative ages of nodes. analysed by Laroche et al. (1995), 58
For example, in Bayesian analyses, temporal chloroplast genes analysed by Goremykin et
information is incorporated through al. (1997), 61 chloroplast genes analysed by
calibrations priors, which can take the form Moore et al. (2007), and the 83 chloroplast
of a variety of probability distributions (Ho genes analysed by Moore et al. (2010).
and Phillips 2009). In the absence of fossils However, most of these studies had sparse
for a particular group being studied, angiosperm taxon sampling. Among the few

74
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
other studies that have included more than sampling of angiosperms and life history-
50 taxa, the largest number of genes sampled associated rate heterogeneity (Beaulieu et al.
was five. The largest taxon samples have 2015). However, comprehensive
been those of Zanne et al. (2014), which investigations of the impact of models,
used a staggering 32,223 species, and priors, and gene sampling on Bayesian
Magallón et al. (2015), which included 792 estimates of the angiosperm evolutionary
angiosperm taxa and one of the largest timescale, using a genome-scale data set and
samples of fossil calibration points ever used. numerous, widely distributed fossil
An exception to the above trade-off between calibrations, have still yielded remarkably
taxon and gene sampling is the study by robust estimates of a Triassic origin of
Foster et al. (2016), which analysed 76 angiosperms (Foster et al. 2016). This
chloroplast genes from 193 angiosperm taxa. implies a long period of no angiosperm
The most controversial aspect of fossilisation, or that fossils of this age simply
angiosperm molecular dating studies has remain to be discovered (but see Wang et
been an apparent incongruence between al2007; Gang et al. 2016).
molecular estimates and those extrapolated Despite the disparate estimates for
purely from fossil occurrence data. Many the origin of crown-group angiosperms, the
modern molecular dating estimates without timescale of evolution within this group is
strongly informative temporal calibrations beginning to be understood with increased
tend to suggest that angiosperms arose in the precision. Of particular note is that estimates
early to mid-Triassic (Figure 3) (Foster et al. for the origin of most modern angiosperm
2016), which implies a considerable gap in orders seem to be consistent regardless of
the fossil record (Doyle 2012). This the age inferred for the angiosperm crown
contradicts the claim that the evolutionary group (Magallón et al. 2015; Foster et al.
history of crown-group angiosperms is well 2016). Ordinal diversification is most
represented in the fossil record (Magallón commonly estimated to have begun in the
2014), despite several lines of evidence early Cretaceous, and is concentrated
supporting this suggestion: the gradual predominantly from this time through to the
increase in abundance, diversity, and mid-Cretaceous (Magallón et al. 2015; Foster
distribution of fossil angiosperms; the et al. 2016). Modern angiosperm families are
ordered progression of both morphological estimated to have originated steadily from
and functional diversification; and the the early Cretaceous, with the peak of family
agreement between the stratigraphic record genesis occurring from the late Cretaceous to
and molecular data in the sequential the early Paleogene (Magallón et al. 2015)
appearance of angiosperm lineages. During this time, the supercontinent Pangaea
If the fault lies instead with the molecular largely completed its breakup into the
estimates, then it has been suggested that the continents of the present day. Concurrently,
substantial disparity between molecular and there were dramatic shifts in climate, with
fossil-based estimates of the age of crown global temperatures and CO2 levels far
angiosperms might be a result of the choices higher than in the present day (Hay and
of molecular markers, taxa, calibrations, or Floegel 2012). These changes, particularly in
models of rate variation (Magallón 2014). temperature, would have had significant
Particular blame has been placed on the impacts on the levels and efficiency of.
inability of molecular dating methods to
account properly for non-representative

75
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants

Figure 3: A recent estimate of the angiosperm evolutionary timescale, modified from Foster et
al. (2016). Numbers in parentheses after taxon names refer to the number of taxa included
from those groups in the study. Green circles indicate estimates of the crown age for lineages
when more than one taxon has been included, and the blue star indicates the inferred age for
the origin of crown-group angiosperms. The dashed line indicates the time by which all
modern orders were inferred to have arisen.

photosynthesis (Ellis 2010; Hay and Floegel history, the relationship of angiosperms to
2012). Selective pressures would have been other seed plants, the relationships within
high, ultimately influencing the evolution of angiosperms, the timescale of angiosperm
angiosperms and, presumably, other taxa that evolution, and the reasons for the relative
interacted with them. success of angiosperms compared to
gymnosperms were all largely unknown or
Concluding remarks and future
not understood.
directions Thankfully, we have now made great
The substantial diversity and global progress in the quest to answer these
dominance of flowering plants have puzzled questions. Work remains to identify
and intrigued many researchers throughout potential stem-group relatives of seed plants,
history. The classification of angiosperms but we now have reliable estimates of the
has long proved difficult because of the phylogeny of extant seed plants. However,
monumental size and such varied the most widely accepted seed plant
morphologies within this group. phylogeny suggests that no extant
Subsequently, the key evolutionary gymnosperm lineage preserves the
innovations that first occurred to produce evolutionary steps that led to the origin of
flowers, as well as the reasons for the the first flowers. Therefore, in some respects
overwhelming success of angiosperms, have the resolution of the seed plant phylogeny
historically been obscured. Therefore, it is has been somewhat of a disappointment for
reasonable to surmise that for most of those wanting to reconstruct the

76
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
development of the flower (Doyle 2012). simultaneous analysis of extant and extinct
While this might be considered a setback, taxa (Ronquist et al. 2012; Gavryushkina et
our greatly improved knowledge of the al. 2014; Heath et al. 2014). Overall, it is
angiosperm phylogeny, including a strongly clear that our understanding of the
supported position for the root, allows evolutionary history of angiosperms has
increasingly sophisticated questions to be changed considerably over time, and we are
asked about angiosperm macroevolution now in an exciting new era of angiosperm
(e.g., Turcotte et al. 2014; Zanne et al. 2014). research.
Similarly, our modern estimates for the
timescale of angiosperm evolution allow us
Acknowledgements
to explore further the selective pressures that
might have shaped the present-day I would like to thank Simon Ho for helpful
distribution and diversity of flowering plants. feedback on this manuscript.
Despite our significant improvements in
understanding the patterns and timescale of References
angiosperm evolution, the field is far from
Angiosperm Phylogeny Group, A. P. G. (1998)
settled. The celebrated consistent, strongly
An ordinal classification for the families of
supported phylogeny based on chloroplast flowering plants, Annals of the Missouri Botanical
markers is increasingly being recognised as Garden, 85, 531–553.
only one estimate of the angiosperm Angiosperm Phylogeny Group, A. P. G. (2003)
phylogeny. The alternative phylogenies An update of the Angiosperm Phylogeny
inferred through analysis of nuclear markers, Group classification for the orders and families
and through the choice of phylogeny of flowering plants, Botanical Journal of the Linnean
reconstruction methods, suggests that more Society, 141, 399–436.
work is needed to reconcile potentially Angiosperm Phylogeny Group, A. P. G. (2009)
conflicting evolutionary histories. An update of the Angiosperm Phylogeny
Group classification for the orders and families
Additionally, the controversy surrounding
of flowering plants: APG III, Botanical Journal of
the age of flowering plants shows no signs of the Linnean Society, 165, 105–121.
abating. Modern knowledge of the fossil Angiosperm Phylogeny Group, A. P. G. (2016)
record suggests that the rapid radiation of An update of the Angiosperm Phylogeny
angiosperm lineages was not quite as Group classification for the orders and families
explosive as implied by Darwin’s of flowering plants: APG IV, Botanical Journal of
“abominable mystery” proclamation, yet a the Linnean Society, 181, 1, 1–20.
new mystery is why molecular date estimates Arber, E. A. & Parkin, J. (1907) On the origin of
still generally far pre-date the oldest angiosperms, Journal of the Linnean Society of
angiosperm fossils. It is unlikely that London, Botany, 38, 263, 29–80.
Beaulieu, J. M., O’Meara, B., Crane, P. &
increasing the amount of genetic data will
Donoghue, M. J. (2015) Heterogeneous rates of
solve this problem (Foster et al. 2016); molecular evolution and diversification could
instead, increased sampling from explain the Triassic age estimate for
underrepresented groups and angiosperms, Systematic Biology, 64, 5, 869–878.
methodological improvements in Bebber, D. P., Carine, M. A., Wood, J. R. I.,
incorporating fossil data appear to be the Wortley, A. H., Harris, D. J., Prance, G. T.,
way forward. The last point appears to be an Davidse, G., Paige, J., Pennington, T. D. &
especially promising avenue of research, with Robson, N. K. B. (2010) Herbaria are a major
new methods being developed for the frontier for species discovery, Proceedings of the

77
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
National Academy of Sciences, 107, 51, 22169– rbcL, Annals of the Missouri Botanical Garden, 80, 3,
22171. 528–580.
Bell, C. D., Soltis, D. E. & Soltis, P. S. (2010) The Chaw, S.-M., Parkinson, C. L., Cheng, Y.,
age and diversification of the angiosperms re- Vincent, T. M. & Palmer, J. D. (2000) Seed
revisited, American Journal of Botany, 97, 8, 1296– plant phylogeny inferred from all three plant
1303. genomes: monophyly of extant gymnosperms
Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, and origin of Gnetales from conifers, Proceedings
W. & Courchamp, F. (2012) Impacts of climate of the National Academy of Sciences, 97, 8, 4086–
change on the future of biodiversity, Ecology 4091.
Letters, 15, 4, 365–377. Chaw, S.-M., Zharkikh, A., Sung, H.-M., Lau, T.-
Bond, W. J. (1989) The tortoise and the hare: C. & Li, W.-H. (1997) Molecular phylogeny of
ecology of angiosperm dominance and extant gymnosperms and seed plant evolution:
gymnosperm persistence, Biological Journal of the analysis of nuclear 18S rRNA sequences,
Linnean Society, 36, 3, 227–249. Molecular Biology and Evolution, 14, 1, 56–68.
Bowe, L. M., Coat, G. & dePamphilis, C. W. Crane, P. R. (1985) Phylogenetic analysis of seed
(2000) Phylogeny of seed plants based on all plants and the origin of angiosperms, Annals of
three genomic compartments: Extant the Missouri Botanical Garden, 72, 4, 716–793.
gymnosperms are monophyletic and Gnetales' Cronquist, A. (1981) An integrated system of
closest relatives are conifers, Proceedings of the classification of flowering plants, Columbia University
National Academy of Sciences, 97, 8, 4092–4097. Press, New York City, USA.
Brodribb, T. J. & Feild, T. S. (2010) Leaf Darwin, F. & Seward, A. C. (1903) More letters of
hydraulic evolution led a surge in leaf Charles Darwin: a record of his work in a series of
photosynthetic capacity during early hitherto unpublished letters, John Murray, London,
angiosperm diversification, Ecology Letters, 13, 2, UK.
175–183. Donoghue, M. J. & Doyle, J. A. (1989a)
Bungard, R. A. (2004) Photosynthetic evolution "Phylogenetic analysis of angiosperms and the
in parasitic plants: insight from the chloroplast relationships of Hamamelidae", in Crane, P. R.
genome, Bioessays, 26, 3, 235-247. & Blackmore, S. (eds.) Evolution, systematics, and
Burger, W. C. (1977) The Piperales and the fossil history of the Hamamelidae. Vol. 1. Introduction
monocots, The Botanical Review, 43, 3, 345–393. and "Lower" Hamamelidae., Clarendon Press,
Burger, W. C. (1981) Heresy revived: the Oxford.
monocot theory of angiosperm origin, Donoghue, M. J. & Doyle, J. A. (1989b)
Evolutionary Theory, 5, 40, 189-226. "Phylogenetic studies of seed plants and
Cantino, P. D., Doyle, J. A., Graham, S. W., Judd, angiosperms based on morphological
W. S., Olmstead, R. G., Soltis, D. E., Soltis, P. S. characters", in Fernholm, B., Bremer, K. &
& Donoghue, M. J. (2007) Towards a Jörnvall, H. (eds.) The hierarchy of life: molecules and
phylogenetic nomenclature of Tracheophyta, morphology in phylogenetic analysis, Elsevier Science,
Taxon, 56, 3, 1E–44E. Amsterdam, 181–193.
Carmichael, J. S. & Friedman, W. E. (1996) Doyle, J. A. (1969) Cretaceous angiosperm pollen
Double fertilization in Gnetum gnemon of the Atlantic Coastal Plain and its
(Gnetaceae): its bearing on the evolution of evolutionary significance, Journal of the Arnold
sexual reproduction within the Gnetales and the Arboretum, 50, 1–35.
anthophyte clade, American Journal of Botany, 83, Doyle, J. A. (2012) Molecular and fossil evidence
6, 767–780. on the origin of angiosperms, Annual Review of
Chase, M. W., Soltis, D. E., Olmstead, R. G., Earth and Planetary Sciences, 40, 301–326.
Morgan, D., Les, D. H., Mishler, B. D., Duvall, Doyle, J. A. & Donoghue, M. J. (1986) Seed plant
M. R., Price, R. A., Hills, H. G. & Qiu, Y.-L. phylogeny and the origin of angiosperms: an
(1993) Phylogenetics of seed plants: an analysis experimental cladistic approach, The Botanical
of nucleotide sequences from the plastid gene Review, 52, 4, 321–431.

78
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
Doyle, J. A. & Hotton, C. L. (1991) (2016) A Whole Plant Herbaceous Angiosperm
"Diversification of early angiosperm pollen in a from the Middle Jurassic of China, Acta
cladistic context", in Blackmore, S. & Barnes, S. Geologica Sinica (English Edition), 90, 1, 19–29.
H. (eds.) Pollen and spores: patterns of diversification, Gavryushkina, A., Welch, D., Stadler, T. &
Clarendon Press, Oxford, 169–195. Drummond, A. J. (2014) Bayesian inference of
Ellis, R. J. (2010) Biochemistry: Tackling sampled ancestor trees for epidemiology and
unintelligent design, Nature, 463, 7278, 164–165. fossil calibration, PLoS Computational Biology, 10,
Endress, P. K. (1987) The early evolution of the 12, e1003919.
angiosperm flower, Trends in Ecology & Evolution, Goremykin, V. V., Hansmann, S. & Martin, W. F.
2, 10, 300–304. (1997) Evolutionary analysis of 58 proteins
Endress, P. K. (2002) Morphology and encoded in six completely sequenced
angiosperm systematics in the molecular era, chloroplast genomes: revised molecular
The Botanical Review, 68, 4, 545–570. estimates of two seed plant divergence times,
Endress, P. K., Baas, P. & Gregory, M. (2000) Plant Systematics and Evolution, 206, 1–4, 337–351.
Systematic plant morphology and anatomy: 50 Goremykin, V. V., Nikiforova, S. V., Biggs, P. J.,
Years of progress, Taxon, 49, 3, 401–434. Zhong, B., Delange, P., Martin, W., Woetzel, S.,
Fairon-Demaret, M. (1996) Dorinnotheca streelii Atherton, R. A., McLenachan, P. A. &
Fairon-Demaret, gen. et sp. nov., a new early seed Lockhart, P. J. (2013) The evolutionary root of
plant from the upper Famennian of Belgium, flowering plants, Systematic Biology, 62, 1, 50–61.
Review of Palaeobotany and Palynology, 93, 1, 217- Goremykin, V. V., Nikiforova, S. V., Cavalieri,
233. D., Pindo, M. & Lockhart, P. (2015) The Root
Fairon-Demaret, M. & Scheckler, S. E. (1987) of Flowering Plants and Total Evidence,
Typification and redescription of Moresnetia Systematic Biology, 64, 5, 879–891.
zalesskyi Stockmans, 1948, an early seed plant Govaerts, R. (2001) How many species of seed
from the Upper Famennian of Belgium, Bulletin- plants are there?, Taxon, 50, 4, 1085–1090.
Institut Royal des Sciences Naturelles de Belgique. Hay, W. W. & Floegel, S. (2012) New thoughts
Sciences de la Terre, 57, 183-199. about the Cretaceous climate and oceans, Earth-
Fleischmann, A., Michael, T. P., Rivadavia, F., Science Reviews, 115, 4, 262–272.
Sousa, A., Wang, W., Temsch, E. M., Heath, T. A., Huelsenbeck, J. P. & Stadler, T.
Greilhuber, J., Müller, K. F. & Heubl, G. (2014) (2014) The fossilized birth–death process for
Evolution of genome size and chromosome coherent calibration of divergence-time
number in the carnivorous plant genus Genlisea estimates, Proceedings of the National Academy of
(Lentibulariaceae), with a new estimate of the Sciences, 111, 29, E2957–E2966.
minimum genome size in angiosperms, Annals Ho, S. Y. W. (2014) The changing face of the
of Botany, 114, 8, 1651–1663. molecular evolutionary clock, Trends in Ecology
Foster, C. S. P., Sauquet, H., van der Merwe, M., & Evolution, 29, 9, 496–503.
McPherson, H., Rossetto, M. & Ho, S. Y. W. Ho, S. Y. W. & Duchêne, S. (2014) Molecular-
(2016) Evaluating the Impact of Genomic Data clock methods for estimating evolutionary rates
and Priors on Bayesian Estimates of the and timescales, Molecular Ecology, 23, 5947–5965.
Angiosperm Evolutionary Timescale, Systematic Ho, S. Y. W. & Phillips, M. J. (2009) Accounting
Biology, in press. for calibration uncertainty in phylogenetic
Friedman, W. E. (1992) Evidence of a pre- estimation of evolutionary divergence times,
angiosperm origin of endosperm: implications Systematic Biology, 58, 3, 367–380.
for the evolution of flowering plants, Science, Ho, S. Y. W., Tong, K. J., Foster, C. S. P., Ritchie,
255, 5042, 336–339. A. M., Lo, N. & Crisp, M. D. (2015)
Galtier, N. (2011) The intriguing evolutionary Biogeographic calibrations for the molecular
dynamics of plant mitochondrial DNA, BMC clock, Biology Letters, 11, 9, 20150194.
Biology, 9, 1, 61. Hochuli, P. A. & Feist-Burkhardt, S. (2013)
Gang, H. A. N., Zhongjian, L. I. U., Xueling, L. I. Angiosperm-like pollen and Afropollis from the
U., Limi, M. A. O., Jacques, F. & Xin, W. Middle Triassic (Anisian) of the Germanic

79
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
Basin (Northern Switzerland), Frontiers in Plant Polycarpicae 1. A general survey, Acta Botanica
Science, 4, 344. Neerlandica, 21, 2, 113–127.
Jiao, Y., Wickett, N. J., Ayyampalayam, S., Moore, M. J., Bell, C. D., Soltis, P. S. & Soltis, D.
Chanderbali, A. S., Landherr, L., Ralph, P. E., E. (2007) Using plastid genome-scale data to
Tomsho, L. P., Hu, Y., Liang, H. & Soltis, P. S. resolve enigmatic relationships among basal
(2011) Ancestral polyploidy in seed plants and angiosperms, Proceedings of the National Academy of
angiosperms, Nature, 473, 7345, 97–100. Sciences, 104, 49, 19363–19368.
Kitazaki, K. & Kubo, T. (2010) Cost of having Moore, M. J., Soltis, P. S., Bell, C. D., Burleigh, J.
the largest mitochondrial genome: evolutionary G. & Soltis, D. E. (2010) Phylogenetic analysis
mechanism of plant mitochondrial genome, of 83 plastid genes further resolves the early
Journal of Botany, 2010, 620137. diversification of eudicots, Proceedings of the
Krassilov, V. A. (1977) The origin of National Academy of Sciences, 107, 10, 4623–4628.
angiosperms, The Botanical Review, 43, 1, 143– Parkinson, C. L., Adams, K. L. & Palmer, J. D.
176. (1999) Multigene analyses identify the three
Lanfear, R., Ho, S. Y. W., Davies, T. J., Moles, A. earliest lineages of extant flowering plants,
T., Aarssen, L., Swenson, N. G., Warman, L., Current Biology, 9, 24, 1485–1491.
Zanne, A. E. & Allen, A. P. (2013) Taller plants Pellicer, J., Fay, M. F. & Leitch, I. J. (2010) The
have lower rates of molecular evolution, Nature largest eukaryotic genome of them all?, Botanical
Communications, 4, 1879. Journal of the Linnean Society, 164, 1, 10–15.
Laroche, J., Li, P. & Bousquet, J. (1995) Pimm, S. L. & Joppa, L. N. (2015) How many
Mitochondrial DNA and monocot-dicot plant species are there, where are they, and at
divergence time, Molecular Biology and Evolution, what rate are they going extinct?, Annals of the
12, 6, 1151–1156. Missouri Botanical Garden, 100, 3, 170–176.
Lee, M. S. Y. & Ho, S. Y. W. (2016) Molecular Preston, J. C. & Hileman, L. C. (2009)
clocks, Current Biology, 26, 10, R399–R402. Developmental genetics of floral symmetry
Levin, D. A. (2002) The role of chromosomal change in evolution, Trends in Plant Science, 14, 3, 147–154.
plant evolution, Oxford University Press, New Qiu, Y.-L., Lee, J., Bernasconi-Quadroni, F.,
York. Soltis, D. E., Soltis, P. S., Zanis, M., Zimmer, E.
Magallón, S. (2014) A review of the effect of A., Chen, Z., Savolainen, V. & Chase, M. W.
relaxed clock method, long branches, genes, (1999) The earliest angiosperms: evidence from
and calibrations in the estimation of angiosperm mitochondrial, plastid and nuclear genomes,
age, Botanical Sciences, 92, 1, 1–22. Nature, 402, 6760, 404–407.
Magallón, S. & Sanderson, M. J. (2001) Absolute Ray, J. (1686–1704) Historia Plantarum, H.
diversification rates in angiosperm clades, Faithorne, London.
Evolution, 55, 9, 1762–1780. Ronquist, F., Klopfstein, S., Vilhelmsen, L.,
Magallón, S., Gómez-Acevedo, S., Sánchez- Schulmeister, S., Murray, D. L. & Rasnitsyn, A.
Reyes, L. L. & Hernández-Hernández, T. P. (2012) A total-evidence approach to dating
(2015) A metacalibrated time-tree documents with fossils, applied to the early radiation of the
the early rise of flowering plant phylogenetic Hymenoptera, Systematic Biology, 61, 6, 973–999.
diversity, New Phytologist, 207, 2, 437–453. Rosas-Guerrero, V., Aguilar, R., Martén-
Martin, W., Gierl, A. & Saedler, H. (1989) Rodríguez, S., Ashworth, L., Lopezaraiza-Mikel,
Molecular evidence for pre-Cretaceous M., Bastida, J. M. & Quesada, M. (2014) A
angiosperm origins, Nature, 339, 46–48. quantitative review of pollination syndromes:
Mathews, S. & Donoghue, M. J. (1999) The root do floral traits predict effective pollinators?,
of angiosperm phylogeny inferred from Ecology Letters, 17, 3, 388–400.
duplicate phytochrome genes, Science, 286, 5441, Rothwell, G. W., Scheckler, S. E. & Gillespie, W.
947–950. H. (1989) Elkinsia gen. nov., a Late Devonian
Meeuse, A. D. J. (1972) Facts and fiction in floral gymnosperm with cupulate ovules, Botanical
morphology with special reference to the Gazette, 150, 2, 170-189.

80
JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
Ruhfel, B. R., Gitzendanner, M. A., Soltis, P. S., heterogeneity among lineages of tracheophytes:
Soltis, D. E. & Burleigh, J. G. (2014) From integration of molecular and fossil data and
algae to angiosperms–inferring the phylogeny evidence for molecular living fossils, Proceedings
of green plants (Viridiplantae) from 360 plastid of the National Academy of Sciences, 99, 7, 4430–
genomes, BMC Evolutionary Biology, 14, 1, 23. 4435.
Saarela, J. M., Rai, H. S., Doyle, J. A., Endress, P. Takhtajan, A. L. (1980) Outline of the
K., Mathews, S., Marchant, A. D., Briggs, B. G. classification of flowering plants
& Graham, S. W. (2007) Hydatellaceae (Magnoliophyta), The Botanical Review, 46, 3,
identified as a new branch near the base of the 225–359.
angiosperm phylogenetic tree, Nature, 446, Thuiller, W., Lavergne, S., Roquet, C.,
7133, 312–315. Boulangeat, I., Lafourcade, B. & Araujo, M. B.
Sagan, L. (1967) On the origin of mitosing cells, (2011) Consequences of climate change on the
Journal of Theoretical Biology, 14, 3, 225–274. tree of life in Europe, Nature, 470, 7335, 531–
Sage, R. F., Christin, P.-A. & Edwards, E. J. 534.
(2011) The C4 plant lineages of planet Earth, Turcotte, M. M., Davies, T. J., Thomsen, C. J. M.
Journal of Experimental Botany, 62, 9, 3155–3169. & Johnson, M. T. J. (2014) Macroecological and
Sanderson, M. J. & Doyle, J. A. (2001) Sources of macroevolutionary patterns of leaf herbivory
error and confidence intervals in estimating the across vascular plants, Proceedings of the Royal
age of angiosperms from rbcL and 18S rDNA Society of London B: Biological Sciences, 281, 1787,
data, American Journal of Botany, 88, 8, 1499– 20140555.
1516. van der Niet, T. & Johnson, S. D. (2012)
Schwartz, R. M. & Dayhoff, M. O. (1978) Origins Phylogenetic evidence for pollinator-driven
of prokaryotes, eukaryotes, mitochondria, and diversification of angiosperms, Trends in Ecology
chloroplasts, 199, 4327, 395–403. & Evolution, 27, 6, 353–361.
Scotland, R. W. & Wortley, A. H. (2003) How Wang, X., Duan, S., Geng, B., Cui, J. & Yang, Y.
many species of seed plants are there?, Taxon, (2007) Is Jurassic Schmeissneria an angiosperm?,
52, 1, 101–104. Acta Palaeontologica Sinica, 46, 4, 490.
Smith, S. A., Beaulieu, J. M. & Donoghue, M. J. Welch, J. J. & Bromham, L. (2005) Molecular
(2010) An uncorrelated relaxed-clock analysis dating when rates vary, Trends in Ecology &
suggests an earlier origin for flowering plants, Evolution, 20, 6, 320–327.
Proceedings of the National Academy of Sciences, 107, Wettstein, R. (1907) Handbuch der systematischen
13, 5897–5902. Botanik, 2 edn., Franz Deuticke, Leipzig.
Smith, S. A. & Donoghue, M. J. (2008) Rates of Wickett, N. J., Mirarab, S., Nguyen, N., Warnow,
molecular evolution are linked to life history in T., Carpenter, E., Matasci, N., Ayyampalayam,
flowering plants, Science, 322, 5898, 86–89. S., Barker, M. S., Burleigh, J. G. &
Soltis, D. E., Smith, S. A., Cellinese, N., Wurdack, Gitzendanner, M. A. (2014)
K. J., Tank, D. C., Brockington, S. F., Refulio- Phylotranscriptomic analysis of the origin and
Rodriguez, N. F., Walker, J. B., Moore, M. J. & early diversification of land plants, Proceedings of
Carlsward, B. S. (2011) Angiosperm phylogeny: the National Academy of Sciences, 111, 45, E4859–
17 genes, 640 taxa, American Journal of Botany, 98, E4868.
4, 704–730. Winter, K.-U., Becker, A., Münster, T., Kim, J. T.,
Soltis, P. S. & Soltis, D. E. (2004) The origin and Saedler, H. & Theissen, G. (1999) MADS-box
diversification of angiosperms, American Journal genes reveal that gnetophytes are more closely
of Botany, 91, 10, 1614–1626. related to conifers than to flowering plants,
Soltis, P. S., Soltis, D. E. & Chase, M. W. (1999) Proceedings of the National Academy of Sciences, 96,
Angiosperm phylogeny inferred from multiple 13, 7342–7347.
genes as a tool for comparative biology, Nature, Wolfe, K. H., Li, W.-H. & Sharp, P. M. (1987)
402, 6760, 402–404. Rates of nucleotide substitution vary greatly
Soltis, P. S., Soltis, D. E., Savolainen, V., Crane, P. among plant mitochondrial, chloroplast, and
R. & Barraclough, T. G. (2002) Rate

81
 JOURNAL & PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES
Foster – History of Flowering Plants
nuclear DNAs, Proceedings of the National Academy angiosperm phylogeny using conserved nuclear
of Sciences, 84, 24, 9054–9058. genes and estimates of early divergence times,
Xi, Z., Liu, L., Rest, J. S. & Davis, C. C. (2014) Nature Communications, 5, 4956.
Coalescent versus concatenation methods and Zhong, B., Yonezawa, T., Zhong, Y. &
the placement of Amborella as sister to water Hasegawa, M. (2010) The position of Gnetales
lilies, Systematic Biology, 63, 6, 919–932. among seed plants: overcoming pitfalls of
Zanne, A. E., Tank, D. C., Cornwell, W. K., chloroplast phylogenomics, Molecular Biology and
Eastman, J. M., Smith, S. A., FitzJohn, R. G., Evolution, 27, 12, 2855–2863.
McGlinn, D. J., O’Meara, B. C., Moles, A. T., Zuckerkandl, E. & Pauling, L. (1962) "Molecular
Reich, P. B., Doyer, D. L., Soltis, D., Stevens, disease, evolution and genic heterogeneity.", in
P., Westoby, M., Wright, I., Aarssen, L., Bertin, Kasha, M. & Pullman, B. (eds.) Horizons in
R., Calaminus, A., Govaerts, R., Hemmings, F., Biochemistry., Academic Press., New York, 189–
Leishman, J., Soltis, P. S., Swenson, N. G., 225.
Warman, L. & Beaulieu, J. M. (2014) Three keys
to the radiation of angiosperms into freezing
environments, Nature, 506, 7486, 89–92.
Zeng, L., Zhang, Q., Sun, R., Kong, H., Zhang,
N. & Ma, H. (2014) Resolution of deep

82

View publication stats