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

net/publication/259607637

Flower Diversity and Angiosperm Diversification

Article in Methods in molecular biology (Clifton, N.J.) · January 2014

DOI: 10.1007/978-1-4614-9408-9_4 · Source: PubMed

CITATIONS READS
43 11,379

2 authors:

Pamela S Soltis Douglas E Soltis

University of Florida University of Florida
770 PUBLICATIONS 74,466 CITATIONS 1,010 PUBLICATIONS 90,084 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Douglas E Soltis on 20 December 2014.

The user has requested enhancement of the downloaded file.

American Journal of Botany 96(1): 110–128. 2009.

FLORAL VARIATION AND FLORAL GENETICS IN BASAL

ANGIOSPERMS1

Pamela S. Soltis,2,5 Samuel F. Brockington,2,3 Mi-Jeong Yoo,2,3 Ana Piedrahita,2,3

Maribeth Latvis,2,3 Michael J. Moore,4 Andre S. Chanderbali,2,3 and
Douglas E. Soltis3
2Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611 USA; 3Department of Botany, University
of Florida, Gainesville, Florida 32611 USA; and 4Department of Biology, Oberlin College, Oberlin, Ohio 44074 USA

Recent advances in phylogeny reconstruction and floral genetics set the stage for new investigations of the origin and diversifi-
cation of the flower. We review the current state of angiosperm phylogeny, with an emphasis on basal lineages. With the surprising
inclusion of Hydatellaceae with Nymphaeales, recent studies support the topology of Amborella sister to all other extant angio-
sperms, with Nymphaeales and then Austrobaileyales as subsequent sisters to all remaining angiosperms. Notable modifications
from most recent analyses are the sister relationships of Chloranthaceae with the magnoliids and of Ceratophyllaceae with eu-
dicots. We review “trends” in floral morphology and contrast historical, intuitive interpretations with explicit character-state recon-
structions using molecular-based trees, focusing on (1) the size, number, and organization of floral organs; (2) the evolution of the
perianth; (3) floral symmetry; and (4) floral synorganization. We provide summaries of those genes known to affect floral features
that contribute to much of floral diversity. Although most floral genes have not been investigated outside of a few model systems,
sufficient information is emerging to identify candidate genes for testing specific hypotheses in nonmodel plants. We conclude
with a set of evo-devo case studies in which floral genetics have been linked to variation in floral morphology.

Key words: ABC model; basal angiosperms; evo-devo; perianth evolution; symmetry; synorganization.

The unifying feature of angiosperms is the flower, with its
associated aspects of reproduction. The fossil record clearly in-
dicates that many diverse floral forms were present early in an-
giosperm evolution (see papers in this issue), so the underlying
genetic architecture of the flower must have allowed for appar-
ently early, rapid, and extensive diversification in floral mor-
phology. The genetics of floral traits have long been of interest
to angiosperm horticulturalists and evolutionary biologists, and
much is known about the genetic control of specific features in
a specific study system. Despite this wealth of information,
only recently have we begun to understand the fundamental
genes that make a flower. Studies of Arabidopsis thaliana (L.)
Heynh indicate that thousands of genes are active throughout
the development of a flower, and perhaps as many as 20 000
genes are expressed at some point in pollen development alone
(Zhao et al., 2001). Despite the apparent genetic complexity of
the transitions from floral meristem to mature flower, these
thousands of genes are themselves governed by a much smaller
set of regulators known as transcription factors. Understanding
the roles of transcription factors is therefore key to deciphering
the genetics of floral diversity. Furthermore, such studies have
begun to extend beyond traditional genetic models and in the
process have started to provide clues to the genetic control of
floral organs early in angiosperm history. In this paper, we re-
view the early evolutionary history of angiosperms, review
general patterns of floral diversity and evolution, describe the
roles and effects of several key transcription factors, and sug-
1
Manuscript received 29 May 2008; revision accepted 18 November 2008.
This work was supported in part by NSF Plant Genome Grants
PGR-0115684 and PGR-0638595. The authors thank three anonymous
reviewers for helpful comments on a previous draft of this paper.
5 Author for correspondence (e-mail: psoltis@flmnh.ufl.edu)
doi:10.3732/ajb.0800182
110
gest testable hypotheses for linking morphological diversity
with gene expression.
PHYLOGENETIC OVERVIEW
Ultimately, the solution to Darwin’s “abominable mystery”
will come from the integration of paleobotany, systematics, and
related fields. Here, we consider historical views of the earliest
angiosperms and the most recent results from molecular phylo-
genetics, both based on analyses of extant diversity, and leave
detailed discussion of the fossil record to other authors in this
issue (see also Crane et al., 2004; Crepet et al., 2004; Crepet,
2008). We do not address the origin of the angiosperms, which
has been discussed in several excellent recent reviews (e.g.,
Doyle, 1994, 1996, 2006; Hilton and Bateman, 2006). Phyloge-
netic analyses (e.g., Crane, 1985; Doyle and Donoghue, 1986)
strongly support the monophyly of angiosperms by identifying
many synapomorphies for the angiosperm clade: double fertil-
ization resulting in endosperm formation, the carpel, phloem
tissue composed of sieve tubes and companion cells, stamens
with two pairs of pollen sacs, and aspects of gametophyte de-
velopment and structure. Throughout, we will refer to “basal
angiosperms” as a group, despite the fact that they do not form
a clade; however, this collective term is engrained in the botani-
cal literature and is more convenient than the more precise ref-
erence to “nonmonocot, noneudicot angiosperms.” With this
explanation, we provide an overview of the phylogeny of “basal
angiosperms.”
Hypotheses surrounding the “most primitive angiosperms”
have changed dramatically and repeatedly over the past century.
Indeed, the meaning of “most primitive angiosperms” has itself
changed. At various times, this label was applied to those gen-
era or families that had retained the largest number of putatively
plesiomorphic features, regardless of the actual antiquity of the
group as inferred from the fossil record. In other interpretations,
January 2009] Soltis et al.—Floral variation and floral genes
the “most primitive angiosperms” were in fact those considered
to represent the earliest lineages, regardless of the collective
“primitiveness” of their characteristics. Through the application
of molecular phylogenetics, the relative branching order and in-
ferences of plesiomorphic characters can be decoupled, permit-
ting reconstruction of both phylogeny and character evolution
based on that phylogeny.
Early views of the first angiosperms were based primarily on
hypotheses of what constituted the most “primitive” flowers.
As described in greater detail later, Cronquist (1968, 1981,
1988), Takhtajan (1991, 1997), and Stebbins (1974), for ex-
ample, considered many of the families of Magnoliales (sensu
AGP II, 2003) to be the vestiges of the earliest angiosperms, on
the basis of their floral morphology. All have some combina-
tion of features inferred to be “primitive”: an elongate floral
axis, a moderate or large number of floral organs, spiral phyl-
lotaxy, and lack of connation and adnation. This view of the
earliest angiosperms prevailed throughout most of the latter
half of the 20th century and received support from the fossil
record with the discovery of Archaeanthus (Dilcher and Crane,
1984), a large, Magnolia-like flower dating to the early
Cretaceous.
Despite the longstanding 20th-century view of Magnolia (or
something similar) as the prototypical angiosperm, this hypoth-
esis was by no means the only one of the earliest angiosperms.
Following the philosophy that classifications should be ar-
ranged from simple to complex forms (later interpreted as hav-
ing evolved from simple to complex), Adrien de Jussieu (1843)
placed the Amentiferae (i.e., species with wind-pollinated flow-
ers borne in catkins (aments)) as the first group of angiosperms,
and Engler (Melchior, 1964) largely followed suit. The Amen-
tiferae, comprising Platanaceae, Betulaceae, Fagaceae, Juglan-
daceae, and Salicaceae, are now recognized as being
polyphyletic, and their simple flowers are highly reduced rather
than ancestrally simple.
Some (e.g., Burger, 1977) have suggested that the monocots
were among the “original” angiosperms, but despite their long
fossil record (e.g., Araceae date to ~115 million years ago
[mya]; Friis et al., 2004) and estimated ages for nearly all major
lineages of more than 100 million years (Bremer, 2000; Wik-
ström et al., 2001), phylogenetic analyses clearly nest the
monocots well within the angiosperms, with strong support in
recent analyses for a placement as the sister to the eudicots +
Ceratophyllaceae (Moore et al., 2007; Jansen et al., 2007).
Other candidates proposed as early lineages of angiosperms
were the “paleoherbs” (i.e., ancient lineages of herbaceous
plants, such as Nymphaeales and Piperales; Taylor and Hickey,
1992), a result supported by some early molecular studies (e.g.,
Hamby and Zimmer, 1992).
Despite progress in understanding both phylogeny and char-
acter evolution, morphological data sets for basal angiosperms
have too few characters, many of which are plagued by ho-
moplasy, to establish relationships with confidence. Likewise,
however, single-gene molecular analyses may also be prone to
uncertainty and/or error. The landmark rbcL analysis of 499
species (Chase et al., 1993) identified Ceratophyllum as the sis-
ter to all other angiosperms (but without support; Chase and
Albert, 1998), which fell into either a “uniaperturate pollen”
clade or a “triaperturate pollen” clade. Although Ceratophyllum
has a long fossil record (Dilcher, 1989), extending back nearly
120–125 mya, few would have argued on any other basis that
Ceratophyllum was one of the earliest angiosperm lineages. As
additional genes were analyzed (e.g., 18S-26S rDNA, D. Soltis
111
et al., 1997; atpB, Savolainen et al., 2000), it became clear that
the placement of Ceratophyllum was an artifact of the initial
rbcL analysis; its placement has continued to vary, ranging
from sister to the monocots (e.g., Qiu et al., 1999; Zanis et al.,
2002) to sister to the eudicots (e.g., D. Soltis et al., 2000; Moore
et al., 2007), a position that is relatively strongly supported
based on analyses of whole plastid genomes (Jansen et al.,
2007; Moore et al., 2007).
During the past decade, Amborella trichopoda Baill. has
emerged as the sister to all other extant angiosperms, either
alone or with Nymphaeales, in nearly every study that includes
adequate taxon sampling (and generally multiple genes) (e.g.,
D. Soltis et al., 1997, 2000; Mathews and Donoghue, 1999,
2000; Parkinson et al., 1999; Qiu et al., 1999, 2000, 2005; P.
Soltis et al., 1999, 2000; Barkman et al., 2000; Graham and
Olmstead, 2000; Borsch et al., 2003; Hilu et al., 2003; Leebens-
Mack et al., 2005; Jansen et al., 2007; Moore et al., 2007). Until
recently, not much was known about this dioecious shrub from
the cloud forests of New Caledonia. In recent years, however,
that has changed dramatically, with reports of its ecology (Feild
et al., 2000, 2004), floral biology, and population structure
(Thien, 2003), pollen (Hesse, 2001), anatomy (Carlquist and
Schneider, 2001, 2002; Furness and Rudall, 2001), floral devel-
opment (Posluszny and Tomlinson, 2003; Buzgo et al., 2004),
floral genes (e.g., Kim et al., 2004, 2005; Yamada et al., 2004;
Albert et al., 2005; Fourquin et al., 2005, 2007; P. Soltis et al.,
2006; D. Soltis et al., 2007; Duarte et al., in press), and genome
(Soltis et al., 2008). Although Amborella sits at the end of a
long branch in most phylogenetic trees (see discussion by Gra-
ham et al., 2000; Qiu et al., 2001), its position as sister to all
other extant angiosperms seems quite secure based on analyses
of genes from all three plant genomes (e.g., Qiu et al., 2005)
and extensive analyses of protein-coding genes from the plastid
genome (Leebens-Mack et al., 2005; Jansen et al., 2007; Moore
et al., 2007). Although in a few analyses, Nymphaeales have
appeared as sister to Amborella or all other angiosperms (e.g.,
Barkman et al., 2000; Graham and Olmstead, 2000; P. Soltis
et al., 2000; Kim et al., 2004), these sister-group relationships
rarely received substantial support and were often dependent on
the method of analysis. Most studies have instead supported
Nymphaeales as the subsequent sister to all other extant angio-
sperms, although see Graham and Iles (2009, pp. 216–227
in this issue) for a note of caution. This early appearance of the
water lily lineage is consistent with the recent discovery of a
water lily fossil from 115 to 125 mya (Friis et al., 2001).
The surprising recent report of Hydatellaceae—formerly
classified in Poales—as sister to extant Nymphaeales (Saarela
et al., 2007) has not altered the relative positions of Amborella
and Nymphaeales (+ Hydatellaceae) as sisters to all remaining
extant angiosperms (e.g., Saarela et al., 2007; Piedrahita et al.,
unpublished data). To address this issue and to clarify addi-
tional relationships among basal angiosperms, we obtained se-
quences from the inverted repeat (IR) region of the plastid
genomes of species of Sarcandra, Austrobaileya, Aristolochia,
Canella, Saururus, Tasmannia, Cabomba, Cinnamomum, and
Saruma (A. Piedrahita et al., unpublished data) using the ampli-
fication, sequencing and annotation of plastomes (ASAP)
method (Dhingra and Folta, 2005), along with the IR regions
from the complete plastid genomes of Hydatella, Hedyosmum,
Musa, Yucca, and Elais (M. Moore et al., unpublished data).
The ASAP method is based on 27 overlapping primer pairs that
span the typical angiosperm IR, including noncoding regions.
We aligned these IR sequences with 25 published sequences
112 American Journal of Botany
using the program MAFFT (Katoh et al., 2002; Katoh and Toh,
2008), followed by manual adjustment, eliminated those inter-
genic regions that could not be confidently aligned, and ana-
lyzed the remaining data (24 463 positions) using the program
GARLI (Zwickl, 2006) with default parameters (Fig. 1A; A.
Piedrahita et al., unpublished data).
Amborella was sister to all other extant angiosperms; Hyda-
tella and Nymphaeales were sisters, and together Hydatellaceae
and Nymphaeales were sister to Austrobaileyales and all re-
maining angiosperms, as found by Saarela et al. (2007). Analy-
ses based on morphology (either with the rest of the tree
constrained to relationships found in previous molecular and
combined molecular-morphological analyses or not constrained)
likewise placed Hydatellaceae as sister to Nymphaeales (Saarela
et al., 2007; Endress and Doyle, 2009, pp. 22–66 in this issue).
The placement of Hydatellaceae as sister to Nymphaeales indi-
cates that this clade is more diverse morphologically than previ-
ously appreciated (Saarela et al., 2007; Rudall et al., 2007;
Friedman, 2008). Although 10 morphological synapomorphies
unite Hydatellaceae and Nymphaeales, the inflorescence and ex-
tremely reduced flowers of Hydatellaceae bear little resemblance
to the reproductive structures of other early angiosperm lin-
eages, with the possible exception of Archaefructus (see Saarela
et al., 2007). Hydatellaceae therefore seem to offer little new
insight into the early evolution of floral form, although aspects
of anatomy and embryology are consistent with its placement as
sister to Nymphaeales (Rudall et al., 2007, 2008; Friedman,
2008), and it remains a relatively understudied family.
Austrobaileyales, comprising Austrobaileyaceae, Trime-
niaceae, Illiciaceae, and Schisandraceae, have emerged as the
sister group to all remaining extant angiosperms. This clade is
strongly supported in all analyses, as is its placement relative to
other clades. It has a long fossil history, with reports of Trime-
niaceae and Illiciaceae extending back to the early to late Cre-
taceous (Frumin and Friis, 1999; Yamada et al., 2008).
All remaining angiosperms form a large clade in the IR tree,
the Mesangiospermae sensu Cantino et al. (2007), comprising
monocots, eudicots, Ceratophyllaceae, Chloranthaceae, and
magnoliids (Magnoliales, Laurales, Piperales, and Canellales).
Although the results of the IR analysis are generally congruent
with those recently reported by Jansen et al. (2007) and Moore
et al. (2007) for the basal lineages of angiosperms, two major
discrepancies exist among the Mesangiospermae. First, in the
IR analysis, Ceratophyllum is sister to the monocots, rather than
to the eudicots as in Moore et al. (2007). Second, Chloran-
thaceae are sister to the eudicots + monocots, rather than to the
magnoliids, as in Moore et al. (2007) and Jansen et al. (2007).
In both cases, the branches in the IR tree are extremely short,
and bootstrap support is less than 50%, suggesting a pentachot-
omy in the IR tree. Relationships among these five lineages
have been very difficult to discern, but recent analyses of whole-
plastid genome sequences may have resolved this pentachotomy
(Moore et al., 2007; Jansen et al., 2007), and see Saarela et al.
(2007) and Graham and Iles (2009, pp. 216–227 in this is-
sue). In the whole-plastid trees, Ceratophyllaceae are sister to
the eudicots, this clade (Ceratophyllaceae + eudicots) is sister to
the monocots, and Chloranthaceae are sister to the magnoliids
(although the latter is not strongly supported). The Chloran-
thaceae-magnoliid clade is sister to the monocots-Ceratophyl-
laceae-eudicots clade (Fig. 1B).
Chloranthaceae, with their ancient fossil record (see Couper,
1958; Walker and Walker, 1984; Friis et al., 1986, 2000; Doyle
et al., 2003; Eklund et al., 2004), have long been considered
[Vol. 96
basal angiosperms. Their position as sister to magnoliids
(Moore et al., 2007) differs from some previous analyses, in
which they appeared as sister to a clade of magnoliids + eu-
dicots (Zanis et al., 2002; Davies et al., 2004).
The magnoliids include most of those families typically con-
sidered “primitive” or “basal” angiosperms (e.g., Cronquist
[1981 and earlier] and Takhtajan [1997 and earlier] Magnolii-
dae). Despite substantial attention from the systematics commu-
nity, relationships among these families were not well understood
until molecular phylogenetic analyses. In fact, data sets of five
or more genes have generally been required to resolve relation-
ships with good support in magnoliids (e.g., Qiu et al., 1999,
2000, 2005; Zanis et al., 2002). Within magnoliids, Magnoliales
are sister to Laurales, and Piperales are sister to Canellales.
Magnoliales comprise Myristicaceae, Degeneriaceae, Him-
antandraceae, Magnoliaceae, Eupomatiaceae, and Annonaceae,
and relationships among these families are now clear (Sauquet
et al., 2003). Laurales (sensu APG II, 2003; see Renner, 1999)
comprise Calycanthaceae (including Idiospermaceae), Monimi-
aceae, Gomortegaceae, Atherospermataceae, Lauraceae, Sipu-
runaceae, and Hernandiaceae. Although Amborellaceae,
Trimeniaceae, and Chloranthaceae have occasionally been
placed in Laurales (e.g., Perkins, 1925; Thorne, 1974; Cron-
quist, 1981, 1988; Takhtajan, 1987, 1997), they clearly are not
part of this clade. Recent phylogenetic studies unite Aristolo-
chiaceae, Lactoridaceae, Piperaceae, Saururaceae, and Hydno-
raceae as Piperales (e.g., Qiu et al., 1999, 2005; P. Soltis et al.,
1999; Barkman et al., 2000; D. Soltis et al., 2000; Nickrent
et al., 2002; Zanis et al., 2002). While not considered closely
related on morphological grounds, Canellaceae and Winter-
aceae form a well-supported sister group in all multigene analy-
ses (e.g., Qiu et al., 1999; P. Soltis et al., 1999; D. Soltis et al.,
2000; Zanis et al., 2002, 2003), as well as in the nonmolecular
analysis of Doyle and Endress (2000). Canellales are strongly
supported as sister to Piperales, although neither Canellaceae
nor Winteraceae have been proposed as closely related to any
members of Piperales. Putative synapomorphies for all four
major clades of magnoliids are given in Doyle and Endress
(2000), Stevens (2001 onward), and Soltis and Soltis (2004).
The remaining ~97% of angiosperm species diversity (Drin-
nan et al., 1994) falls primarily into the monocot (22%) and eu-
dicot (75%) clades, with Ceratophyllaceae as sister to eudicots
(Moore et al., 2007). Relationships among most major lineages
of monocots and eudicots are now clear (Fig. 1; see Chase et al.,
2006, and D. Soltis et al., 2005, respectively, for summaries).
Because this paper focuses primarily on floral evolution in early
lineages of angiosperms, we will not address these relationships
further here, but refer the reader to Fig. 1 and the reviews cited.
However, early evolutionary splits within the monocots and eu-
dicots (e.g., Ranunculales) are potentially informative for recon-
structions of early angiosperm evolution (e.g., Endress and
Doyle, 2009, pp. 22–66 in this issue). Therefore, consideration
of floral morphology and floral genetics in eudicots in particular
may be relevant for understanding the early diversification of the
flower, and the discussion that follows consequently includes
information on eudicots, where appropriate.
PATTERNS OF FLORAL EVOLUTION
Historical intuitive interpretations—Throughout most of the
20th century, floral evolution was discussed as “trends”—for ex-
ample, Armen Takhtajan’s (1991) book Evolutionary Trends in
January 2009] Soltis et al.—Floral variation and floral genes 113

Fig. 1. (A) Phylogenetic tree based on maximum likelihood (ML) analysis of the inverted repeat (IR) region of the plastid genome for 39 angiosperms
and two gymnosperm outgroups (Pinus and Cycas) (Piedrahita et al., unpublished data). (B) Summary phylogenetic tree of major lineages of angiosperms,
based primarily on analyses of the Angiosperm Phylogeny Group (2003), Moore et al. (2007), and Jansen et al. (2007), as described in text.
114 American Journal of Botany
Fig. 1.
Flowering Plants. Perhaps the best-known “trends” are those re-
ferred to as “Bessey’s Dicta” (Bessey, 1915), which, either directly
or indirectly, had a profound impact on views of floral diversifica-
tion for much of the 20th century. Most prominent for considering
patterns of floral evolution are the following abbreviated state-
ments: polymery to oligomery, petaly to apetaly, actinomorphy to
zygomorphy, hypogyny to epigyny, apocarpy to syncarpy, poly-
carpy to oligocarpy, polystemony to oligostemony, apostemony to
synstemony, and monocliny to dicliny. Bessey also stressed that
evolution is not always “upward” and that evolutionary rates may
differ among organs/structures, such that a given plant may be a
mosaic of ancestral and derived features. Bessey’s views influ-
enced many generations of systematists, particularly in North
America. Although explicit reconstructions of specific features
across phylogenetic trees show generally complex patterns of flo-
[Vol. 96
Continued.
ral evolution (discussed later), the “trends” suggested by Bessey
are in many cases largely supported (see e.g., Ronse De Craene
et al., 2003; D. Soltis et al., 2005).
Although Bessey’s hypotheses are in one sense “phylogenetic”—
that is, they describe putative changes through time—they
followed neither an explicitly derived phylogenetic tree nor an
explicit method for inferring evolutionary transformations in
individual characters. Likewise, many other prevailing views
of floral evolution lack explicit methodology and empha-
size the particular interests of their authors. Despite these
shortcomings, the views of floral change on a macroevolu-
tionary scale put forward by Bessey in the early 20th century
and by Cronquist, Takhtajan, and Stebbins in the latter half of
the 1900s, have contributed many hypotheses that are test-
able given current understanding of angiosperm phylogeny.
January 2009] Soltis et al.—Floral variation and floral genes
Historical intuitive interpretations vs. phylogenetic recon-
struction— Using explicit hypotheses of phylogeny, several
recent studies have addressed the evolution of specific floral
characters in light of reconstructions of the ancestral flower.
Here, we review both the ancestral reconstruction of the flower
and patterns of evolution, based in large part on the phyloge-
netic trees of the Angiosperm Phylogeny Group (2003;
Mathews and Donoghue, 1999; Parkinson et al., 1999; Qiu
et al., 1999, 2000; P. Soltis et al., 1999; Barkman et al., 2000;
D. Soltis et al., 2000; Zanis et al., 2002; Hilu et al., 2003; re-
viewed in D. Soltis et al., 2005) and compare and contrast these
findings with the historical intuitive approach of Bessey, Cron-
quist, Tahktajan, Stebbins, and others. We limit this discussion
to “trends” in overall floral organization because it is floral or-
gan identity and orientation that are best understood from a
genetic perspective (see later). We focus on four central fea-
tures of the flower: (1) the size, number, and organization of
floral organs; (2) the evolution of the perianth; (3) floral sym-
metry; and (4) floral synorganization.
Trends in flower size, organ number, and phyllotaxy—De
Jussieu (1843) considered that wind-pollinated flowers borne in
catkins (aments) represented the first flowers, with more com-
plex forms arising later. This view of the Amentiferae as the
“model” for the earliest angiosperms was followed by Engler
(Melchior, 1964). Of course, we now recognize that wind pol-
lination has evolved multiple times in the angiosperms, each
time accompanied by a reduction in floral size and complexity
(e.g., Stebbins, 1970; Culley et al., 2002).
Takhtajan (1991) viewed the ancestral flower as being of
moderate size, with subsequent evolution toward both larger
and smaller sizes. The earliest flowers had an indefinite (but
moderate) number of separate organs, spirally arranged on a
somewhat elongate receptacle (e.g., like those of Degeneria
[Degeneriaceae], many Magnoliaceae, Galbulimima [Himan-
tandraceae], Eupomatia [Eupomatiaceae], and Winteraceae).
Per Takhtajan (1991), through time, the receptacle shortened,
bringing stamens and carpels into close proximity. “Cyclic”
flowers (i.e., those with whorled, rather than spirally arranged,
organs) arose early in angiosperm evolution in multiple lin-
eages. The transitions from spiral to cyclic flowers proceeded
from the perianth inward or from the carpels outward. Further-
more, Takhtajan (1991) viewed this transformation, especially
between spiral phyllotaxy and trimery, as reversible (see also
Endress, 1987), generating further diversity in combinations of
spiral vs. whorled organ arrangements. A further trend noted by
Takhtajan is the reduction in the number of homologous floral
organs through time—i.e., “oligomerization”—and a concomi-
tant fixation of the number of parts. Takhtajan viewed these
smaller flowers with fixed numbers of parts as more “integrated
than archaic, less oligomerized flowers”—a concept similar to
Endress’s (1990a) synorganization. On occasion, as in Magno-
liaceae, Cactaceae, and Aizoaceae, polymerization led to sec-
ondarily derived, large flowers, with many parts.
Consideration of these observations in light of the current
phylogeny suggests that the ancestral condition is rather more
uncertain (and see Endress and Doyle, 2009, in this issue). Al-
though both Amborella and Austrobaileyales have spiral phyl-
lotaxy, consistent with the model of Cronquist (1988), Takhtajan
(1991), and others, Nymphaeales have whorled phyllotaxy,
making the ancestral state equivocal; the fossil record is like-
wise equivocal, as both spiral and whorled phyllotaxy are pres-
ent in many early angiosperms. Likewise, reconstructions of
115
the number of parts of the ancestral perianth are equivocal; Am-
borella and Austrobaileyales again share a state—an indetermi-
nate number of parts. However, trimery, present in Nymphaeales
and reconstructed as the ancestral state for all angiosperms ex-
cept Amborella, Nymphaeales, and Austrobaileyales, evolved
early and characterizes many lineages of basal angiosperms (D.
Soltis et al., 2005).
Evolution of the perianth— The origin and diversification of
the perianth were central to Takhtajan’s views of floral evolu-
tion. He proposed that the first perianth originated from protec-
tive bracts (“bracteopetals”; Kozo-Poljanski, 1922). Bracteal
perianths are purportedly found in Degeneriaceae, Magnoli-
aceae, Annonaceae, Canellaceae, Winteraceae, Illiciaceae,
Schisandraceae, Austrobaileyaceae, Lauraceae, and Calycan-
thaceae (see also Smith, 1928; Eames, 1961; Hiepko, 1965).
Traditionally, petals of monocots and most “Magnoliopsida”
sensu Takhtajan (i.e., essentially all eudicots) have been con-
sidered “andropetals,” having originated through sterilization
and modification of stamens. A staminal origin of petals has
been recognized for over a century (e.g., Celakovsky, 1896;
Worsdell, 1903; Arber and Parkin, 1907; Kozo-Poljanski, 1922;
Troll, 1927, 1928; Eames, 1931, 1961; Hiepko, 1965). More
recently, Ronse De Craene (2007) has questioned the accepted
staminal origin of petals in eudicots and suggested that at least
some eudicots have petals of bracteal origin.
Takhtajan considered the evolutionary replacement of brac-
teopetals by andropetals unlikely and proposed a sequence by
which andropetals may have originated within flowers that bore
a perianth of bract-like sepals. Following the view that petals
are essentially stamens with arrested development (Goebel,
1933), Takhtajan proposed a two-step origin of the perianth in-
volving a bracteal origin of sepals and an andropetaloid origin
of petals.
Character reconstructions based on molecular phylogenetic
analyses have identified many examples of independent deriva-
tion of a differentiated perianth (e.g., Ronse De Craene et al.,
2003; D. Soltis et al., 2005). The ancestral perianth was likely
undifferentiated, without distinct sepals and petals, and both the
derivation of a differentiated perianth and the loss of perianth
appear to have arisen multiple times independently throughout
the course of angiosperm evolution (Ronse De Craene et al.,
2003; D. Soltis et al., 2005). Explicit models of perianth origin
have recently been proposed (e.g., Baum, 1998; Albert et al.,
1998, 2002; Frohlich and Parker, 2000; Frohlich, 2006).
An outstanding example of a clade in which multiple deriva-
tions of a differentiated perianth have occurred is the Caryo-
phyllales. Although a clade of core eudicots rather than an older
group, Caryophyllales can provide excellent information on
transitions of perianth form. Recent phylogenetic analysis of
the order, together with a parsimony reconstruction analysis of
perianth features (Brockington et al., in press), suggests that a
simple uniseriate perianth was the ancestral condition within
the Caryophyllales; subsequently, there have been a minimum
of six independent origins of a differentiated perianth.
Floral symmetry— Shifts between radially symmetric (acti-
nomorphic) and bilaterally symmetric (zygomorphic) flowers
have occurred multiple times throughout the history of angio-
sperms (e.g., Stebbins, 1974; Endress, 1996, 1999; Donoghue
et al., 1998; Ree and Donoghue, 1999; Endress and Matthews,
2006). Zygomorphy is evident in the fossil record in the early
Tertiary (Crepet et al., 1991; Dilcher, 2000; Crepet, 2008) and
116 American Journal of Botany
among extant plants is generally restricted to the monocots (Or-
chidaceae and some Zingiberales) and a few clades of core eu-
dicots (the rosid family Fabaceae and the asterids Lamiales and
Asteraceae) (Stevens, 2001 onward). However, zygomorphy
also occurs in the basal angiosperm Aristolochia (Aristolochi-
aceae). A general “trend” toward zygomorphy has been ob-
served within many clades of asterids, with shifts to zygomorphy
viewed as having important implications for pollination ecol-
ogy (e.g., Crepet et al., 1991; Crepet, 1996, 2008; Neal et al.,
1998; Dilcher, 2000). Multiple gains and losses of zygomorphy
in the asterids (e.g., Donoghue et al., 1998; Ree and Donoghue,
1999; Cubas, 2004) suggest a relatively simple genetic switch
between actinomorphy and zygomorphy. Furthermore, zygo-
morphic flowers across the asterids are not homologous, and
thus switches both from and to actinomorphy may occur by dif-
ferent genetic mechanisms (Donoghue et al., 1998). Finally,
changes in symmetry may occur during development, for ex-
ample, from zygomorphy early in development to actinomor-
phy in mature flowers (Endress, 1999; Tucker, 1999).
Floral synorganization— One of the few major trends in
floral evolution is that of increasing synorganization of the
structural components of the flower (Endress, 1990a). Synor-
ganization (Remane, 1956; Vogel, 1959, 1969; Endress, 1990a)
is the connection of two or more structural elements to form a
functional system. In flowers, synorganization results gener-
ally from congenital or postgenital fusion of parts (sepals, pet-
als, stamens, or carpels) or occasionally from an otherwise
close association of parts. Despite the widespread occurrence
of floral synorganization, particularly in eudicots, little or noth-
ing is known about the potential genetic control of synorgani-
zation, whether associations are among like or unlike floral
parts. A prerequisite for experiments designed to address the
genetic control of synorganizational associations is a thorough
understanding of morphological evolutionary patterns, from
both broad and narrow phylogenetic perspectives. Plant fami-
lies of relatively recent derivation (e.g., the asterids) generally
have more highly synorganized flowers than those families that
represent more ancient lineages (e.g., magnoliids and other
basal angiosperms) (e.g., Endress, 1990b). Although synorga-
nization takes many forms, it is most commonly expressed as
the fusion of carpels to form a syncarpous gynoecium, a condi-
tion that characterizes ~80% of the estimated 260 000-plus
(Takhtajan, 1997) species of angiosperms (Endress, 1982).
Syncarpy is considered evolutionarily successful (Endress,
1990b) because it provides for centralized selection of pollen
tubes (Endress, 1982; Endress et al., 1983).
Although more ancient groups of angiosperms exhibit less
synorganization than more derived groups, the evolutionary
origins of synorganization are unclear. Synorganized flowers
characterize most eudicots, but several basal lineages of eu-
dicots lack synorganized flowers (e.g., most Ranunculaceae,
Trochodendron, Nelumbo). Among magnoliids, Eupomatia
(Magnoliales) exhibits synorganization of stamens to form a
synandrium, demonstrating that synorganized flowers have
evolved repeatedly. Many groups of monocots also have synor-
ganized flowers (e.g., irises, orchids, gingers, bromeliads),
which likely represent parallels to conditions in the eudicots.
Apparent losses of synorganization have also occurred in some
genera (e.g., Paeonia, in Saxifragales). The phylogenetic distri-
bution of synorganized flowers is a key to understanding floral
diversity and the constraints under which floral evolution has
proceeded (Endress, 1999).
[Vol. 96
Synorganization is lacking or limited in those flowers with
spiral (or irregular) phyllotaxy (Endress, 1990a, b), in which
the structural elements of the flower are arranged along a spiral
on the floral apex. Increased synorganization is observed in
those flowers with whorled phyllotaxy, wherein identical struc-
tural elements are arranged in rings on the floral apex. Appar-
ently correlated with spiral phyllotaxy and a low potential for
synorganization are the variability in the number of structural
elements of the flower and a high frequency of deviations from
the most common floral pattern (Endress, 1990a). For example,
many magnoliids and other basal lineages have spiral phyl-
lotaxy and variable numbers of parts (e.g., Friis and Endress,
1990; Takhtajan, 1991), and little or no synorganization of
structural elements, although even phyllotaxy appears highly
labile in these plants (e.g., Endress, 1987, 1990b, 1994); most
fossil flowers from the early Cretaceous have spiral phyllotaxy,
but whorled phyllotaxy was also present at that time (see Friis
and Endress, 1990). More recent lineages, such as those of the
asterid clade, show whorled phyllotaxy, fixed (or nearly so)
merosity, and synorganization both among elements of a whorl
and between whorls (as in Apocynaceae; Endress, 1990a). In
fact, Endress (1990a) considered whorled arrangement, radial
symmetry, and a small, fixed number of floral organs prerequi-
sites for the evolution of complex synorganizations. The stabil-
ity of these associations and the possible preconditions needed
for the development of synorganized flowers is supported by
mapping several of these characters onto a phylogenetic tree for
angiosperms (e.g., Soltis et al., 2005).
Synorganization of stamens occurs less frequently than syn-
organization of carpels or petals, although synorganized sta-
mens do occur in some groups, including the basal angiosperm
Eupomatia (Eupomatiaceae) and the core eudicots Fabaceae,
Malvaceae, and Asteraceae. Associated with this more limited
synorganization is greater variability in both number and ar-
rangement of stamens. Patterns of stamen organization are as-
sociated with other floral traits and may depend, to some extent,
on the structure and organization of the flower as a whole. In
this respect, the degree of synorganization of the entire flower
may either constrain or allow the evolution of certain types of
stamen organization. Evolutionary and developmental con-
straints on basic floral structure are generally attributed to phys-
ical limitations, such as the size and shape of the receptacle
and/or floral apex (e.g., Endress, 1990a, b; Takhtajan, 1991).
Polymery (or true polyandry), the spiral arrangement and cen-
tripetal development of many stamens, is often associated with
spiral phyllotaxy, an undifferentiated perianth, the presence of
staminodes, complex floral vasculature, and typically trimery
or dimery as the basic merosity of the flower (Ronse De Craene
and Smets, 1994, 1995).
The reconstructed ancestral flower of crown angio-
sperms— In contrast to the paradigm of large, strobiloid flow-
ers as ancestral, both explicit character reconstructions (e.g.,
Doyle and Endress, 2000; Ronse De Craene et al., 2003; D.
Soltis et al., 2005) and the fossil record (e.g., Taylor and
Hickey, 1992; Friis et al., 2000) suggest that the earliest flow-
ers were instead of small to moderate size (see also Endress,
2001, 2006). As noted, however, both the phyllotaxy and
merosity of the ancestral angiosperm flower are equivocal. The
perianth was probably undifferentiated, and the stamens were
likely laminar, producing pollen characterized by irregular ra-
dial (columellar) elements mixed with granules—and “inter-
mediate” morphology (Doyle and Endress, 2000), in contrast
January 2009] Soltis et al.—Floral variation and floral genes 117

Fig. 2. (A) ABC(E) model of floral organ identity, illustrating combinatorial action of A-, B-, C-, and E-function genes. (B) Fading borders model,
showing gradients in levels of gene products in overlapping regions of gene action (based on gene expression studies of basal angiosperms, especially Kim
et al., 2005); note that the E-function has not yet been examined in detail, and the E box is therefore left open.
118 American Journal of Botany
to Cronquist’s (1988) view of granular infratectal pollen struc-
ture as ancestral. The earliest carpels appear to have been as-
cidiate and closed by secretion (Endress and Igersheim, 2000),
rather than plicate per Eames (1961) and Cronquist (1968,
1988), although postgenital fusion is also present in the earliest
lineages of extant angiosperms (Endress and Igersheim, 2000).
Morphological and developmental studies of extant species
continue to allow inferences of other features of early angio-
sperms (e.g., endosperm ploidy, Williams and Friedman
[2002], Friedman [2008]; endosperm formation, Floyd and
Friedman [2000]), as do new reports and interpretation of the
fossil record (e.g., Sun et al., 1998, 2002; Friis et al., 2001,
2004; Gandolfo et al., 2004).
DEVELOPMENTAL GENETICS OF THE FLOWER
Overview— Because of its central role in both agricultural
and evolutionary studies, the flower has received extensive at-
tention from morphological, developmental, and genetic per-
spectives, and syntheses of developmental genetics have
emerged, based on prominent models and crops (see, e.g., re-
views in Davies et al., 2006; Endress, 2006; Frohlich, 2006;
Irish, 2006; Rijpkema et al., 2006; D. Soltis et al., 2006; Teeri
et al., 2006; Zahn et al., 2006a; Theissen and Melzer, 2007). It
may be that small changes in the timing or location of gene
expression can lead to large changes in floral phenotype. Devel-
opmental genetics is providing a host of candidate genes with
which to test hypothesized effects and infer causation of floral
diversity. For example, the genetic control of floral organ iden-
tity—via the ABC model (Bowman et al., 1991; Coen and
Meyerowitz, 1991)—has been thoroughly investigated in Ara-
bidopsis and Antirrhinum and extended, in one form or another,
to most other angiosperms (see Irish, 2006; P. Soltis et al., 2006;
Theissen and Melzer, 2007, for reviews). Likewise, knowledge
of other gene systems, such as the transcription factors that
regulate floral symmetry, allows for testable hypotheses on the
shifts between radially and bilaterally symmetrical flowers.
Eventually, detailed analyses of the genetics of model organ-
isms will provide further candidates for linking genes with phe-
notypes. In the sections that follow, we present examples of
floral diversity that may be related to specific gene actions.
Such studies are in their infancy, in an evolutionary context,
particularly for basal angiosperms, and the results presented
here are clearly limited to the small sample of plant species in-
vestigated to date. Therefore, some of the examples we describe
involve eudicots rather than basal angiosperms, simply to il-
lustrate the ways in which developmental genetics and studies
of floral evolution can be integrated. However, before present-
ing these examples, we will review briefly some of the funda-
mental aspects of floral genetics.
The ABC model—The ABC model describes the activities of
transcription factors that regulate floral organ identity (Fig. 2).
This combinatorial model of gene activity posits that A function
specifies sepals, the combined activity of A and B functions spec-
ifies petals, B and C functions together specify stamens, and C
function alone specifies carpels. In Arabidopsis, APETALA1
(AP1) and APETALA2 (AP2) control the A function, APETALA3
(AP3) and PISTILLATA (PI) are B-function genes, and AGA-
MOUS (AG) is the C-function gene. Recognition of the role of
SEPALLATA (SEP) genes in the specification of floral organ
identities (E function; Pelaz et al., 2000; Theissen, 2001; Ditta
[Vol. 96
et al., 2004) has led to the revision of the “ABC” model as the
“ABCE” model. The D function controls ovule identity, and the
D-function gene in Arabidopsis is SEEDSTICK (STK) (Colombo
et al., 1995; Pinyopich et al., 2003). All of these genes, except
AP2, are MADS-box genes. A major question is, “What regu-
lates the regulators?” This topic has been addressed in detail for
the A, B, and C functions by Zahn et al. (2006a).
Most of these key regulators of floral organ identity are
MADS-box genes. Despite the occurrence of MADS-box genes
in Physcomitrella, Ceratopteris, and Ophioglossum, these
genes are not orthologs of the ABC genes (Krogan and Ashton,
2000; Theissen et al., 2000; Münster et al., 2002). In contrast,
phylogenetic analyses (e.g., Becker et al., 2003; Zahn et al.,
2005) have demonstrated clear orthologies between ABC genes
from angiosperms and MADS-box genes from gymnosperms.
However, the functions of these ABC orthologs in gymno-
sperms are uncertain. Analyses of gene expression in conifers
and gnetophytes suggest a role in reproduction, but direct func-
tional equivalencies between gymnosperm MADS genes and
their counterparts in angiosperms are not possible. B-gene or-
thologs are expressed only in male reproductive structures in
gymnosperms, suggesting a conserved role in specifying male
reproductive function; however, the C-gene orthologs are ex-
pressed in both male and female structures (Becker et al., 2003).
The common ancestor of angiosperms and gymnosperms likely
shared a complex set of MADS-box genes, with conserved
roles in reproduction (Ng and Yanofsky, 2001). Despite numer-
ous genetic-based models for the origin of the flower (e.g.,
Frohlich and Parker, 2000; Albert et al., 2002; Theissen et al.,
2002), how that ancestral reproductive program was modified
to yield a flower largely remains an abominable mystery itself.
Where divergent floral morphologies are due to alterations in
organ specification, the ABC genes represent excellent candi-
dates for the genetic control of floral differences, although many
aspects of floral morphology are likely to be controlled by other
means. Much of floral diversity is represented by changes in the
number and arrangement of floral organs. Because the activities
of the ABC regulators have little to do with expanding and de-
clining numbers of floral organs (with the exception of organ
absence, perhaps), other genes and gene systems are needed to
explain such differences.
CYCLOIDEA and other genes that control floral symme-
try— Floral symmetry in the model Antirrhinum majus L.
(snapdragon) is controlled by three transcription factors in two
gene families. CYCLOIDEA (CYC), of the TCP family, speci-
fies dorsal floral identity (Luo et al., 1996, 1999; Almeida et al.,
1997; reviewed in Howarth and Donoghue, 2006) and is the
best studied of the symmetry-controlling genes; DIVARICATA
and RADIALIS of the MYB family are also involved in floral
symmetry (Almeida et al., 1997; Galego and Almeida, 2002;
Corley et al., 2005; Costa et al., 2005). In Antirrhinum majus
(snapdragon), CYC and its close relative DICHOTOMA (DICH)
are partially redundant in function, differing slightly in the tim-
ing of expression, but both are required for production of a
“normal” bilaterally symmetrical flower with proper dorsiven-
trality. Mutations lead to radially symmetrical flowers and loss
of dorsiventrality. A complex pattern of gene duplication
throughout the history of eudicots has produced a set of paral-
ogs (Howarth and Donoghue, 2006), the functions of which are
still unknown, but which may contribute in combination with
other genes to fundamental shifts in floral morphology. Ho-
mologs of CYC are also involved in determining symmetry in
January 2009] Soltis et al.—Floral variation and floral genes
the asterid Gerbera (Broholm et al., 2008) and have been re-
cruited, apparently independently, to control floral symmetry in
legumes (Fabaceae) and Iberis (Brassicaceae), both rosids
(Feng et al., 2006; Busch and Zachgo, 2007; Wang et al., 2008).
The control of symmetry in other lineages, including basal an-
giosperms like Aristolochia and relatives (which have contrast-
ing types of symmetry), is as yet unknown but could involve
other genes in new and complex ways.
Other genes affecting floral morphology— The ABC model
applies only to organ identity and thus can be used only for test-
ing hypotheses of organ homology, ancestry, etc. Changes in
the numbers and arrangements of floral organs are controlled by
other genes, but data are generally only available for Arabidop-
sis. Whether the orthologs to the Arabidopsis genes maintain
similar functions in other species is not yet known; however,
the Arabidopsis genes provide possible candidates for evo-devo
study. In Arabidopsis, the homeodomain protein WUSCHEL
and the CLAVATA ligand–receptor system control the fate of
shoot apical meristems, thereby affecting the number of floral
organs. Both WUSCHEL and the CLAVATA genes act by regu-
lating hormone levels in the shoot apical meristem (Leibfried
et al., 2005). The CLV/WUS genetic module may cause changes
in meristem size and therefore shifts in the number of floral or-
gans. Other candidates are PERIANTHIA (PAN) and ETTIN
(ETT). PAN is required for floral organ patterning in Arabidop-
sis thaliana and acts downstream of genes affecting floral mer-
istem identity and independently of those controlling meristem
size; pan mutants show a switch to pentamery in sepals, petals,
and stamens in A. thaliana (Running and Meyerowitz, 1996).
Also in A. thaliana, ETT specifies regional floral meristem
identity and affects perianth organ number, stamen number,
and differentiation in stamens and gynoecia (Sessions et al.,
1997). Like PAN, ETT acts independently of CLV. More infor-
mation on the patterns and timing of gene expression is needed
for all of these genes in Arabidopsis and other models before
they can be fully considered as candidate genes to explain al-
terations in floral organ number on an evolutionary scale.
Carpel closure, a key event in early angiosperm evolution,
has been addressed thoroughly from a morphological perspec-
tive (Endress and Igersheim, 2000). The carpels of most of the
basal lineages of extant angiosperms (Amborella, Cabomba,
Brasenia, Austrobaileya, Trimenia, Kadsura, and Ascarina)
are closed by secretion (Endress and Igersheim, 2000; Endress,
2005), suggesting that this was the plesiomorphic state. At
least some fusion occurred independently in multiple lineages
of basal angiosperms, such as Nymphaea, Illicium, and the
magnoliids (Endress and Igersheim, 2000), and eventually car-
pels closed by fusion were established as the main state in the
remainder of the angiosperms. The genes CRABS CLAW and
TOUSLED control major facets of carpel development, includ-
ing carpel fusion in Arabidopsis, and have been shown to have
conserved roles in Amborella and Cabomba (Fourquin et al.,
2005, 2007). However, the fact that these genes have con-
served expression in Arabidopsis, Amborella, and Cabomba
indicates that they are not responsible for variation in the de-
gree of carpel fusion among angiosperm lineages, and the
gene(s) controlling the extent of fusion have not been identi-
fied. Another gene involved in carpel development is YAB-
BY-2, which in Arabidopsis is expressed in the abaxial domains
of the leaf, carpel, stamen, and perianth and contributes to dor-
siventrality (reviewed in Yamada et al., 2004). The Amborella
ortholog of YABBY-2, AmbF1, is expressed instead in the
119
adaxial tissues of carpel and leaf, suggesting a possible shift in
the polarity pathway between basal angiosperms and Arabi-
dopsis (Yamada et al., 2004).
USING FLORAL GENES TO TEST HYPOTHESES OF
FLORAL EVOLUTION
Correlating variation in organ identity with gene expres-
sion patterns— In many basal angiosperms, the perianth is
considered “undifferentiated” with no true sepals or petals.
However, developmental studies of Amborella suggest that in
fact the perianth is differentiated into outer and inner tepals
(Buzgo et al., 2004), but with a gradual transition from lateral
receptacular bracts to tepals, from outer tepals to inner tepals,
and from inner tepals to stamens in male flowers (and from te-
pals to staminodes and then carpels in female flowers). This
gradual transition from outer to inner floral organs also appears
to characterize other basal angiosperms such as Austrobailey-
ales and may therefore represent the ancestral condition (e.g.,
Ronse De Craene et al., 2003; Soltis et al., 2005).
This view of floral morphology is not consistent with the
discrete whorls of morphologically distinct floral organs pres-
ent in most eudicots and predicted by the ABC model. Neither
the ABC model nor the alternative sliding/shifting boundary
modification (Bowman, 1997; Kramer et al., 2003) can account
for such morphological intergradations. The fading borders
model (Buzgo et al., 2004) suggests that gradual transitions in
organ morphology result from a gradient in levels of expression
of floral regulators across the meristem (Fig. 2B). Overlapping
expression of floral regulators would impose some features of
adjacent organs onto each other and thus produce morphologi-
cally intergrading rather than distinct floral organs. Expression
of B, C, and E homologs in basal angiosperms is often broader
than specified by the ABC model, and expression of B-function
homologs in particular supports the fading borders model (Kim
et al., 2005).
Given the typically broad expression patterns of B and C ho-
mologs in most basal angiosperms studied to date, the ABCE
model of Arabidopsis and Antirrhinum appears to have been
derived from an ancestral genetic program that expressed floral
regulators broadly across the floral meristem (see Kim et al.,
2005). Although fundamental components of the floral genetic/
developmental program have been conserved across the angio-
sperms, evolutionary diversification of this ancestral program
may have occurred through localized expression of different
regulators to different regions of the meristem. For example,
restricted expression of floral organ regulators resulted in the
specification of discrete floral whorls (Kim et al., 2005), as seen
in most eudicots and explained by the combinatorial action of
the ABCE model of Arabidopsis. In the magnoliid Asimina
(Annonaceae) and in the core eudicots, restricted gene expres-
sion results in flowers with distinct and differentiated sepals,
petals, stamens, and carpels—an obviously independent deriva-
tion of the whorled floral structure in these distant lineages
(Kim et al., 2005). In the basal eudicots Ranunculus and Aqui-
legia (Ranunculaceae), shifts in location of B function, directed
by duplicate B-function genes, are correlated with changes in
perianth morphology (Kramer et al., 2003; Kramer and Zimmer,
2006; Rasmussen et al. [2009], pp. 96–109 in this issue).
In general, divergent floral morphologies may be controlled by
underlying shifts in the timing and/or location of expression of
a key set of floral regulators. Different “attempts” at fine-tuning
120 American Journal of Botany
of the ancestral program may be responsible for having gener-
ated much of the floral diversity that exists, or has existed.
Decoupling morphological variation and candidate gene
expression— In many cases, morphological variation is corre-
lated with spatiotemporal patterns of gene expression. How-
ever, there are also instances in which the evolution of
morphological variation does not appear to correlate with pat-
terns of candidate gene expression; these are of equal impor-
tance when attempting to understand the extent to which key
regulatory genes have influenced the evolution of morphologi-
cal novelty. An analysis of MADS-box gene expression in Aris-
tolochia (Aristolochiaceae; Jaramillo and Kramer, 2004)
revealed that orthologs of the B-function genes APETALA3 and
PISTILLATA are expressed in the showy perianth but that their
expression is restricted to the nonpetaloid tissue. These AP3
and PI orthologs do not seem to be involved in the production
of the petaloid parts of the perianth. Clearly, therefore, the evo-
lution of tissues characteristic of particular floral organs, e.g.,
petals, can occur independently of proposed candidate genes.
AP3 and PI have been shown to be expressed in the showy pe-
rianth parts of petaloid monocots such as Tulipa (Liliaceae) but
are also expressed in the morphologically distinct and nonpeta-
loid second-whorl lodicules of derived grasses such as Oryza
and Zea (Poaceae; Ambrose et al., 2000; Nagasawa et al., 2003).
Whipple et al. (2007) demonstrated that AP3 and PI expression
is maintained in the second perianth whorl of basal monocots
through derived grasses, irrespective of the degree of morpho-
logical change that has occurred in these organs. These results
support a conserved role for the B function in the second whorl
of perianth across angiosperms, but they argue that this conser-
vation is due to the role of these genes in allowing differentia-
tion of perianth organs rather than specifying petal identity (and
characteristics) per se. B-gene expression can perhaps be main-
tained as a molecular switch even when the downstream mor-
phological pathways that are turned on by this switch have
undergone considerable diversification. Additional studies of
instances in which floral morphology and patterns of gene ex-
pression are not correlated will provide further insight into the
evolution of floral development.
Bracteopetals vs. andropetals in Lauraceae— The underly-
ing genetic mechanism of floral organ identity provides a means
for testing among hypotheses of bracteal and andropetaloid ori-
gins of the perianth (e.g., Albert et al., 1998, 2002; Baum, 1998;
Chanderbali et al., 2006; discussed later). The perianth in most
basal angiosperms is typically composed of morphologically
similar tepals, presumably of bracteopetaloid origin (e.g.,
Takhtajan, 1991; see earlier). However, there may be a bias in
studies of bracteopetals vs. andropetals, with many apparent ex-
amples of bracteopetals yet to be thoroughly examined by in
situ hybridization of MADS-box gene orthologs. One of the few
studies to date, in addition to Amborella (Kim et al., 2005) and
Aristolochia (Jaramillo and Kramer, 2004), in which the spatial
and temporal aspects of gene expression have been examined, is
that of Persea americana Mill. (Lauraceae; Chanderbali et al.,
2006). The flowers of Lauraceae are trimerous, with three sta-
men whorls and an inner whorl of staminodes surrounding a
single carpel (Rohwer, 1993). The two perianth whorls are mor-
phologically similar and have been considered bracteopetals on
the basis of their generally “sepaloid” appearance (Ronse De
Craene et al., 2003; Albert et al., 1998). However, homologs of
the C-class gene AGAMOUS—typically only expressed in sta-
[Vol. 96
mens and carpels—are also expressed in the tepals of Persea
(avocado), perhaps the vestige of a staminal ancestry of Persea
tepals and suggesting andropetaly rather than bracteopetaly in
Persea and other Lauraceae (Chanderbali et al., 2006). Although
the expression of AG homologs in tepals of Persea may instead
reflect expansion of gene expression from the reproductive or-
gans into the perianth, both genetic and morphological data
make this interpretation less likely (Chanderbali et al., 2006).
The replacement of outer stamen whorls by tepaloid staminodes
in several members of Lauraceae suggests a close developmen-
tal pathway between stamens and tepals and further supports the
andropetaloid origin of tepals in this family. Microarray analy-
sis has identified over 1000 genes with elevated expression in
floral tissues (relative to leaves), including homologs of AP3,
PI, AG, and SEP3 and many other transcription factors; this ap-
proach further supports the interpretation of a staminal origin of
the perianth in Persea (Chanderbali et al., 2008). Further study
of gene expression and function in Persea, a transformable
(Cruz-Hernández et al., 1998; Litz et al., 2005) woody, basal
angiosperm, should help to address fundamental questions of
homology and the origin of the perianth.
In this case study in Lauraceae, the expression of floral genes
has been employed to question traditional concepts of organ
homology. To detect differences in gene expression between
bracteopetals and andropetals, we need a clear understanding of
the homology of the petaloid organ (either to stamens or bracts).
As argued by Ronse De Craene (2007), traditional concepts of
bracteopetals and andropetals, particularly in core eudicots,
need re-evaluation. Nonetheless, indisputable examples of
bracteopetals and andropetals within closely related lineages do
exist, for example in the core Caryophyllales (see later; Brock-
ington et al., in press).
Caryophyllales perianth reconstruction— Although a dis-
tinct lineage of eudicots rather than one of basal angiosperms,
Caryophyllales exhibit patterns of perianth evolution that allow
for untangling bracteopetals from andropetals. Molecular phy-
logenetic analyses have identified many examples of indepen-
dent derivation of a differentiated perianth (Ronse De Craene
et al., 2003; D. Soltis et al., 2005). From an ancestral simple
uniseriate perianth in Caryophyllales, there have been between
six and 10 independent origins of a differentiated perianth
(Brockington et al., in press). Although in Caryophyllales the
homology of the perianth parts to petals in other core eudicots
is uncertain, it is clear that many lineages (Sesuvioideae of
Aizoaceae, Nyctaginaceae, Portulacaceae, Cactaceae) possess
petaloid organs that are bracteal in origin. Furthermore, the oc-
currences of stamen-derived petals within Caryophyllales
(Caryophyllaceae, Aizoaceae, and Glinus and Corbichonia in
Molluginaceae) are independent events (Ronse De Craene,
2007; Brockington et al., in press). These independent occur-
rences are valuable as there are very few examples of petals
within core eudicots that are unquestionably stamen-derived
(Ronse De Craene, 2007). The pattern of perianth evolution in
Caryophyllales therefore presents an important opportunity to
address questions as to differences and/or similarities in the ge-
netic determination of bracteopetals and andropetals. The ge-
netic and developmental control of both andropetals and
bracteopetals within the predominantly andropetaloid core eu-
dicots merits further investigation.
Radial vs. bilateral symmetry— CYCLOIDEA and related
genes have been proposed as key elements in the many shifts
January 2009] Soltis et al.—Floral variation and floral genes
between actinomorphy and zygomorphy. However, the loss of
zygomorphy is much more complex than a simple loss of CYC
expression. Furthermore, CYC and its relatives play other roles
than simply controlling floral symmetry. In Antirrhinum majus
and other Antirrhineae, mutations in both CYC and DICH are
required to produce ventralized flowers with radial symmetry.
It would seem that CYC/DICH double mutants could arise fairly
regularly, resulting in independent losses of zygomorphy within
otherwise zygomorphic clades. However, mutations in CYC
seem not to be responsible for at least some of the switches
from zygomorphy to actinomorphy in the asterids (Donoghue
et al., 1998). Changes in symmetry through development also
suggest that the classic gene action of CYC cannot fully explain
all switches between zygomorphy and actinomorphy. Finally,
evidence of extensive gene duplication of the CYC lineage in
core eudicots (Howarth and Donoghue, 2006) suggests that the
genetic control of bilateral symmetry may be more complex
than typically envisioned. For example, in Lonicera morrowii
A. Gray (Caprifoliaceae), expression of duplicate copies varies
with development and may be partitioned among petals (How-
arth and Donoghue, 2006), but additional work is needed to
evaluate the generality of this result among species with multi-
ple CYC paralogs.
Mohavea confertiflora A. Heller is a close relative of Antir-
rhinum majus that differs in corolla symmetry (i.e., it has radial
symmetry) and pollination mode, possibly due to changes in
expression of CYC and DICH orthologs. Hileman et al. (2003)
demonstrated that shifts in the expression of CYC and DICH
orthologs are correlated with changes in floral morphology in
Mohavea: abortion of the lateral stamen is correlated with ex-
pansion of CYC and DICH expression, while a reduction in
DICH expression is associated with the symmetry of the adaxial
petals. Therefore, differences in levels and location of expres-
sion of these key regulators have had a profound effect on floral
morphology and consequently on pollination ecology of this
species.
Gene duplications and their role in floral diversifica-
tion— Gene duplications represent the ultimate source for the
origin of genetic novelty (Ohno, 1970), and it is reasonable to
predict that duplications in genes controlling aspects of floral
morphology may be responsible for morphological novelty.
This prediction has been best studied in Aquilegia vulgaris L.
(Kramer et al., 2007; see Rasmussen et al. 2009, pp. 96–109
in this issue). Duplications of the floral organ identity gene
AqvAP3 have produced three paralogs, and each has its own
function in specifying the petaloid sepals, petals, and stamin-
odes (Kramer et al., 2007). These paralogs have diverged in
expression in space and time. Early in floral development,
AqvAP3-1 is expressed in petal, stamen, and staminode primor-
dia but later in development its expression is restricted to sta-
minodes. AqvAP3-2 is expressed early in stamens and
staminodes, but then its expression shifts to stamens and petals
with some sepal expression very late in development. AqvAP3-3
is restricted to petals throughout most of development, with
very low expression in stamens at later stages. Functional anal-
yses of AqvPI, which is expressed in all floral organs except
carpels, demonstrates that the Aquilegia B-class genes control
petal, stamen, and staminode identity with little contribution to
sepal development, despite the petaloid appearance of the se-
pals (see Rasmussen et al., 2009, pp. 96–109 in this issue).
The partitioning of gene function among duplicate copies
(i.e., subfunctionalization, Lynch and Conery, 2000) allows for
121
possible functional divergence and the evolution of novelty. It
may be that the presence of multiple AqvAP3 paralogs, each
with its own function, allowed for the derivation of both peta-
loid sepals and staminodes in Aquilegia. Taking this line of rea-
soning further, other floral novelties may also be related to gene
duplications and partitioning of gene function among duplicates.
For example, many key floral regulators have undergone gene
duplication at various points in the phylogenetic history of an-
giosperms (see also paper by D. Soltis et al., 2009, pp. 336–348
in this issue). The AP3 and PI gene lineages are the products of
a gene duplication in the common ancestor of the angiosperms,
possibly more than 260 mya (Kim et al., 2004); a single ho-
molog to this gene pair is present in extant gymnosperms. Like-
wise, the SEP genes, which are required for forming the quartets
that carry the AP1, AP2, AP3, and PI protein subunits, were
duplicated in the common ancestor of the angiosperms (Zahn
et al., 2005). The coincidence of these duplicate genes may have
actually led to the development of the floral organ identity pro-
gram responsible for the origin of the flower (Zahn et al., 2005).
Additional duplication of the AP3 gene lineage, along with du-
plications in the AP1, AG, and SEP gene lineages (Kramer et al.,
1998, 2004; Litt and Irish, 2003; Kramer and Hall, 2005; Zahn
et al., 2005; reviewed in Kramer and Zimmer, 2006; Irish, 2006),
near the origin of the core eudicots suggests that these duplica-
tions may also be involved in the “new” floral morphology that
emerged in the core eudicots: a synorganized flower based on a
pentamerous groundplan (D. Soltis et al., 2003). Duplications
within the AP1 lineage (Litt and Irish, 2003) are associated with
shifts in expression patterns and possibly function. The paralogs
AP1 and FRUITFULL (FUL), of the euAP1 and euFUL gene
lineages, respectively, have different functions (reviewed in
Kramer and Hall, 2005). Early in flower development, AP1 con-
tributes to floral meristem identity and later helps to specify se-
pal and petal identity; FUL also contributes to floral meristem
identity early in development and to carpel and fruit develop-
ment later. However, despite similar expression patterns of
euAP1 orthologs across the angiosperms, the AP1 function of
Arabidopsis thaliana is not conserved (see Kramer and Hall,
2005); this expression pattern is not observed for euFUL or-
thologs. Meanwhile, the sister lineage to the euAP1 and euFUL
lineage, the FUL-like clade, occurs in basal angiosperms and
has different expression patterns and therefore likely different
functions from AP1, FUL, and their orthologs. Expression lev-
els of the FUL-like genes in the basal angiosperms Nuphar and
Magnolia were greater in leaves and carpels than in perianth or
stamens—more like those of euFUL orthologs than euAP1 or-
thologs, perhaps indicating the ancestral expression pattern of
the AP1 lineage (Kim et al., 2005; Yoo et al., unpublished data).
However, expression was high in leaves and all floral organs in
Cabomba (Yoo et al., unpublished data). An AP1 homolog has
not been detected in Amborella (Kim et al., 2005). Finally, du-
plications in the CYC gene lineage throughout the core eudicots
may also be related to changes in floral morphology (Howarth
and Donoghue, 2006). Whether these duplications in many flo-
ral gene lineages arose via independent duplication events or
through polyploidy remains unknown; regardless, however,
they supply a rich source of genetic material that may generate
floral novelty. Such genes deserve greater attention in compara-
tive studies.
Although the functions of ABC homologs in basal angio-
sperms have not yet been demonstrated, their expression pat-
terns may serve as proxies for initial inferences of functional
divergence of duplicate gene pairs. Duplications of ABC genes
122 American Journal of Botany [Vol. 96

have occurred on more local phylogenetic levels (e.g., within a
species or genus), and while some duplicates have similar or
identical expression patterns (e.g., EScaAG1 and EScaAG2 in
Eschscholzia californica Cham., Zahn et al., 2006b; Nyod.
AG1–1 and -2 in Nymphaea odorata Aiton, Yoo et al., unpub-
lished data), several duplicate genes have divergent expression
patterns and therefore possibly divergent functions (Table 1).
For example, whereas duplicate AP3 homologs in Nuphar ad-
vena (Aiton) W. T. Aiton have identical expression patterns
across floral organs late in floral development, when analyzed
with RT-PCR, early in development Nu.ad.AP3.2 is clearly ex-
pressed in inner tepals, and Nu.ad.AP3.1 is not (as detected by
in situ hybridization; Kim et al., 2005). Putative alternatively
spliced AP3-homolog variants in Nymphaea odorata (class I
and class VI) differ dramatically in their expression patterns,
with Nyod.AP3 class I expressed at moderate to high levels in
leaves, sepals, petals, inner stamens, and carpels, while Nyod.
AP3 class VI expression is restricted to inner stamens and car-
pels (Yoo et al., unpublished data). Likewise, differences in
expression occur among AGL2-homolog duplicates and AGL6-
homolog duplicates in Cabomba caroliniana A. Gray (Yoo
et al., unpublished data). In Illicium floridanum J. Ellis, the ex-
pression of AP3 homologs varies, with Il.fl.AP3.2 expressed in
outer tepals, inner tepals, and stamens, Il.fl.AP3.1 expression
limited to inner tepals and stamens, and Il.fl.AP3.3 expression
restricted to inner tepals (Kim et al., 2005). It is tempting to
speculate that differences in expression patterns of duplicate
AP3 homologs in these basal angiosperms may contribute to
morphological variation in sepaloidy vs. petaloidy in perianth
organs and stamens; however, functional studies have not yet
been conducted.
Variation in expression of AG homologs in Persea ameri-
cana (avocado) occurs among floral organs, with PamAG-1 and
-2 expressed in outer and inner tepals, stamens, and carpels, but
PamAG-3 is expressed only in reproductive structures (Chand-
erbali et al., 2006). Furthermore, expression of AG homologs in
the related P. borbonia (L.) Spreng., with a dimorphic perianth,
is absent from the outer tepals: PamAG-1 and -2 are expressed
in inner tepals, stamens, and carpels, and PamAG-3 is expressed
only in the reproductive organs, as in P. americana. Because
expression of AP3 and PI homologs occurs in both whorls, the
bipartite perianth cannot be attributed to expression of the “B-
function” genes. Instead, variation in expression patterns of AG
duplicates between closely related species may result in mor-
phological differences, such as the dimorphic perianth of P.
borbonia.
As with the major duplications of ABC genes described, it is
unclear whether the duplications found within the basal angio-
sperms arose via polyploidy, tandem duplications, or some
other mechanism. However, given that many basal angiosperms
are ancient polyploids (e.g., Stebbins, 1950; Soltis and Soltis,
1990; Cui et al., 2006; see D. Soltis et al., 2009, pp. 336–348
in this issue), many of these duplicate pairs may have arisen via
polyploidy.
PROSPECTUS
Floral variation, particularly in basal angiosperms, is exten-
sive, and its genetic control is only beginning to be understood.
Knowledge of the genes controlling floral development and most
floral morphology is still in its infancy but is expanding rapidly.
With the addition of new models, generalities can be tested, re-
sulting in the eventual identification of candidates to explain mor-
phological differences. For some hypotheses, such as the fading
borders model, predictions from morphology and development
have been upheld by genetics. For others, such as the homology
of the perianth of grass flowers, conservation of gene expression
patterns does not maintain a stable morphology. The decoupling
of morphology and genetics may be real, or it may reflect errors
in homology assessment at the morphological, developmental,
and/or genetic level. Opportunities to study this decoupling may
arise through identification of morphological convergence.
Convergence in morphological features is evident in several
of the characters described by Bessey (1915) as important trans-
formations from ancestral to derived states. In fact, Bessey’s
dicta often correspond to patterns of character-state change
identified through explicit reconstruction of character evolution,
at least as “trends” if not as absolutes. Whereas Bessey’s dicta
may suggest a simple transition from ancestral to derived state,
the features identified by Bessey have often undergone conver-
gent evolution, with multiple derivations of the “same” charac-
teristic. Comparison of the genetics of these states would indicate
whether there are multiple ways of arriving at the same point.

Table 1. Expression patterns of duplicate genes in basal angiosperms. Number of pluses indicates level of expression; – indicates no expression.
Superscript p indicates petals instead of tepals. See text for references.
Species Gene Leaves Sepals Outer tepalsp Inner tepalsp Stamens Carpels

Eschscholzia californica (Papaveraceae) EScaAG1 – – – – +++ +++

EScaAG2 – – – – +++ +++
Nymphaea odorata (Nymphaeaceae) Nyod.AG1–1 – +++ ++p – +++ ++
Nyod.AG1–2 – ++ ++p – ++ +++
Nyod.AP3 I +++ +++ ++p – ++ ++
Nyod.AP3 VI – – – – +++ +++
Nuphar advena (Nymphaeaceae) Nu.ad.AP3.1 – +++ +++ – +++ +++
Nu.ad.AP3.2 – +++ +++ +++ +++ +++
Cabomba caroliniana (Cabombaceae) Caca.AGL2–1 – +++ +++p – +++ +++
Caca.AGL2–2 – +++ +++p – +++ +++
Caca.AGL2–3 ++ +++ +++p – +++ +++
Caca.AGL6–1 – +++ +++p – + –
Caca.AGL6–2 – +++ +++p – +++ +++
Caca.AGL6–3 – – – – – +++
Illicium floridanum (Illiciaceae) Il.fl.AP3–1 – – – +++ +++ –
Il.fl.AP3–2 – – +++ +++ +++ –
Il.fl.AP3–3 – – – +++ – –
January 2009] Soltis et al.—Floral variation and floral genes
Synorganization is a major theme in discussions of floral evo-
lution, yet it has not been addressed from an evo-devo perspec-
tive. A complex set of morphological features is involved in any
synorganization—it typically involves more than one organ
type, and it may involve the number of organs, floral symmetry,
etc., of a flower. Future studies should look to characterize the
morphological and developmental features in such a way as to
identify key elements of a given synorganization—and then take
an evo-devo approach to understanding those elements. Ulti-
mately, an evo-devo perspective on synorganization should help
to explain both the evolution of specific floral organs as well as
the collective, integrated organ systems that arise via synorgani-
zation, one of the major trends/processes in the evolution of flo-
ral variation.
LITERATURE CITED
Albert, V. A., M. H. G. Gustafsson, and L. Di Laurenzio.
1998. Ontogenetic systematics, molecular developmental genetics,
and the angiosperm petal. In D. E. Soltis, P. S. Soltis, and J. J. Doyle
[eds.], Molecular systematics of plants II, 349–374. Kluwer, Boston,
Massachusetts, USA.
Albert, V. A., D. Oppenheimer, and C. Lindqvist. 2002. Pleiotropy, re-
dundancy and the evolution of flowers. Trends in Plant Science 7:
297–301.
Albert, V. A., D. E. Soltis, J. E. Carlson, W. G. Farmerie, P. K. Wall,
D. C. Ilut, L. A. Mueller, et al. 2005. Floral gene resources from
basal angiosperms for comparative genomics research. BMC Plant
Biology 5: 5.
Almeida, J., M. Rocheta, and L. Galego. 1997. Genetic control of flower
shape in Antirrhinum majus. Development 124: 1387–1392.
Ambrose, B. A., D. R. Lerner, P. Ciceri, C. M. Padilla, M. F.
Yanofsky, and R. J. Schmidt. 2000. Molecular and genetic analy-
ses of the Silky1 gene reveal conservation in floral organ specification
between eudicots and monocots. Molecular Cell 5: 569–579.
APG [Angiosperm Phylogeny Group] . 2003. An update of the Angio-
sperm Phylogeny Group classification for the orders and families
of flowering plants. Botanical Journal of the Linnean Society 141:
399–436.
Arber, E. A. N., and J. Parkin. 1907. On the origin of angiosperms.
Journal of the Linnean Society. Botany 38: 29–80.
Barkman, T. J., G. Chenery, J. R. McNeal, J. Lyons-Weiler, and C.
W. dePamphilis. 2000. Independent and combined analysis of se-
quences from all three genomic compartments converge to the root of
flowering plant phylogeny. Proceedings of the National Academy of
Sciences, USA 97: 13166–13171.
Baum, D. 1998. The evolution of plant development. Current Opinion in
Plant Biology 1: 79–86.
Becker, A., H. Saedler, and G. Theissen. 2003. Distinct MADS-box gene
expression patterns in the reproductive cones of the gymnosperm
Gnetum gnemon. Development Genes and Evolution 213: 567–572.
Bessey, C. E. 1915. The phylogenetic taxonomy of flowering plants.
Annals of the Missouri Botanical Garden 2: 109–164.
Borsch, T., K. W. Hilu, D. Quandt, V. Wilde, C. Neinhuis, and W.
Barthlott. 2003. Non-coding plastid trnT-trnF sequences re-
veal a well resolved phylogeny of basal angiosperms. Journal of
Evolutionary Biology 16: 558–576.
Bowman, J. L. 1997. Evolutionary conservation of angiosperm flower de-
velopment at the molecular and genetic levels. Journal of Biosciences
22: 515–527.
Bowman, J. L., D. R. Smyth, and E. Meyerowitz. 1991. Genetic in-
teractions among floral homeotic genes of Arabidopsis. Development
112: 1–20.
Bremer, K. 2000. Early Cretaceous lineages of monocot flowering
plants. Proceedings of the National Academy of Sciences, USA 97:
4707–4711.
Brockington, S. F., S. Crawley, K. Hilu, D. E. Soltis, and P. S. Soltis.
In press. Phylogeny of the Caryophyllales s.l. and the evolution of the
123
perianth in the core Caryophyllales. International Journal of Plant
Sciences.
Broholm, S. K., S. Tähtiharju, R. A. E. Laitinen, V. A. Albert, T. H.
Teeri, and P. Elomaa. 2008. A TCP domain transcription factor con-
trols flower type specification along the radial axis of the Gerbera
(Asteraceae) inflorescence. Proceedings of the National Academy of
Sciences, USA 105: 9117–9122.
Burger, W. 1977. The Piperales and monocots—Alternative hypotheses
for the origin of the monocotyledonous flowers. Botanical Review 43:
345–393.
Busch, A., and S. Zachgo. 2007. Control of corolla monosymmetry in the
Brassicaceae Iberis amara. Proceedings of the National Academy of
Sciences, USA 104: 16714–16719.
Buzgo, M., P. S. Soltis, and D. E. Soltis. 2004. Floral developmental
morphology of Amborella trichopoda (Amborellaceae). International
Journal of Plant Sciences 165: 925–947.
Cantino, P., J. A. Doyle, S. G. Graham, W. S. Judd, R. G. Olmstead, D.
E. Soltis, P. S. Soltis, and M. J. Donoghue. 2007. Towards a phylo-
genetic nomenclature of Tracheophyta. Taxon 56: 822–846.
Carlquist, S., and E. L. Schneider. 2001. Vegetative anatomy of the
New Caledonia endemic Amborella trichopoda: Relationships with
the Illiciales and implications for vessel origin. Pacific Science 55:
305–312.
Carlquist, S., and E. L. Schneider. 2002. The tracheid–vessel el-
ement transition in angiosperms involves multiple independent
features: Cladistic consequences. American Journal of Botany 89:
185–195.
Celakovsky, L. J. 1896. Uber den phylogenetischen Entwicklungsgang der
Bluthe und uber den Ursprung der Blumenkrone 1/2. Sitzungsberichte
der Koniglichen Bomischen Gesellschaft der Wissenschaften in Prag.
Mathematisch-naturwissenschaftliche Classe 1896: 1–91.
Chanderbali, A. S., V. A. Albert, V. E. T. M. Ashworth, M. T. Clegg,
R. E. Litz, D. E. Soltis, and P. S. Soltis. 2008. Persea americana
(avocado): Bringing ancient flowers to fruit in the genomics era.
BioEssays 30: 386–396.
Chanderbali, A., S. Kim, M. Buzgo, P. Zheng, D. G. Oppenheimer, D. E.
Soltis, and P. S. Soltis. 2006. Genetic footprints of stamen ancestors
guide perianth evolution in Persea (Lauraceae). International Journal
of Plant Sciences 167: 1075–1089.
Chase, M. W., and V. A. Albert. 1998. A perspective on the contribution
of plastid rbcL DNA sequences to angiosperm phylogenetics. In D.
E. Soltis, P. S. Soltis, and J. J. Doyle [eds.], Molecular systematics of
plants II, 488–507. Kluwer, Boston, Massachusetts, USA.
Chase, M. W., M. F. Fay, D. S. Deevey, O. Maurin, N. Ronstad, J. Davies,
Y. Pillon, et al. 2006. Multi-gene analyses of monocot relation-
ships: A summary. In J. T. Columbus, E. A. Friar, C. W. Hamilton,
J. M. Porter, L. M. Prince, and M. G. Simpson [eds.], Monocots:
Comparative biology and evolution, 62–74. Rancho Santa Ana
Botanic Garden, Claremont, California, USA.
Chase, M. W., D. E. Soltis, R. G. Olmstead, D. Morgan, D. H. Les,
B. D. Mishler, M. R. Duvall, et al. 1993. Phylogenetics of seed
plants: An analysis of nucleotide sequences from the plastid gene
rbcL. Annals of the Missouri Botanical Garden 80: 528–580.
Coen, E., and E. Meyerowitz. 1991. The war of the whorls: Genetic inter-
actions controlling flower development. Nature 353: 31–37.
Colombo, L., J. Franken, E. Koetje, J. Vanwent, H. J. M. Dons, G. C.
Angenent, and A. J. Van Tunen. 1995. The petunia MADS box
gene FBP11 determines ovule identity. Plant Cell 7: 1859–1868.
Corley, S. B., R. Carpenter, L. Copsey, and E. Coen. 2005. Floral asym-
metry involves an interplay between TCP and MYB transcription fac-
tors in Antirrhinum. Proceedings of the National Academy of Sciences,
USA 102: 5068–5073.
Costa, M. M. R., S. Fox, A. I. Hanna, C. Baxter, and E. Coen.
2005. Evolution of regulatory interactions controlling floral asymme-
try. Development 132: 5093–5101.
Couper, R. A. 1958. British Mesozoic microspores and pollen grains.
Palaeontographica Abteilung B, Palaeophytologie 103: 75–179.
Crane, P. 1985. Phylogenetic analysis of seed plants and the origin of
angiosperms. Annals of the Missouri Botanical Garden 72: 716–793.
124 American Journal of Botany
Crane, P. R., P. Herendeen, and E. M. Friis. 2004. Fossils and plant phy-
logeny. American Journal of Botany 91: 1683–1699.
Crepet, W. L. 1996. Timing in the evolution of derived floral charac-
ters: Upper Cretaceous (Turonian) taxa with tricolpate and tricol-
pate-derived pollen. Review of Palaeobotany and Palynology 90:
339–360.
Crepet, W. L. 2008. The fossil record of angiosperms: Requiem or re-
naissance? Annals of the Missouri Botanical Garden 95: 3–33.
Crepet, W. L., K. C. Nixon, and E. M. Friis. 1991. Fossil evidence for
the evolution of biotic pollination. Philosophical Transactions of the
Royal Society of London, B, Biological Sciences 333: 187–195.
Crepet, W. L., K. C. Nixon, and M. A. Gandolfo. 2004. Fossil evidence
and phylogeny: The age of major angiosperm clades based on meso-
fossil and macrofossil evidence from Cretaceous deposits. American
Journal of Botany 91: 1666–1682.
Cronquist, A. 1968. The evolution and classification of flowering plants.
Nelson, London, UK.
Cronquist, A. 1981. An integrated system of classification of flowering
plants. Columbia University Press, New York, New York, USA.
Cronquist, A. 1988. The evolution and classification of flowering plants.
New York Botanical Garden Press, Bronx, New York, New York,
USA.
Cruz-Hernández, A., Witjaksono, R. E. Litz, and M. A. Gomez-Lim,
1998. Agrobacterium tumefaciens-mediated transformation of em-
bryogenic avocado cultures and regeneration of somatic embryos.
Plant Cell Reports 17: 497–503.
Cubas, P. 2004. Floral zygomorphy, the recurring evolution of a success-
ful trait. BioEssays 26: 1175–1184.
Cui, L., P. K. Wall, J. Leebens-Mack, B. G. Lindsay, D. Soltis, J. J. Doyle,
P. Soltis, et al. 2006. Widespread genome duplications throughout
the history of flowering plants. Genome Research 16: 738–749.
Culley, T. M., S. G. Weller, and A. K. Sakai. 2002. The evolution of
wind pollination in angiosperms. Trends in Ecology & Evolution 17:
361–369.
Davies, B., M. Cartolano, and Z. Schwarz-Sommer. 2006. Flower de-
velopment: The Antirrhinum perspective. Advances in Botanical
Research 44: 279–321.
Davies, T. J., T. G. Barraclough, M. W. Chase, P. S. Soltis, D. E. Soltis,
and V. Savolainen. 2004. Darwin’s abominable mystery: Insights
from a supertree of the angiosperms. Proceedings of the National
Academy of Sciences, USA 101: 1904–1909.
De Jussieu, A. 1843. Monographie de la Famille des Malpighiaceìes. Archives
du Museìum d’Histoire Naturelle 3: [5]–151, [255]–616, pl. 1–23.
Dhingra, A., and K. M. Folta. 2005. ASAP: Amplification, sequencing
and annotation of plastomes. BMC Genomics 6: 176.
Dilcher, D. 2000. Toward a new synthesis: Major evolutionary trends in
the angiosperm fossil record. Proceedings of the National Academy of
Sciences, USA 97: 7030–7036.
Dilcher, D. L. 1989. The occurrence of fruits with affinities to
Ceratophyllaceae in lower and mid-Cretaceous sediments. American
Journal of Botany 76: 162.
Dilcher, D. L., and P. R. Crane. 1984. Archaeanthus: An early angio-
sperm from the Cenomanian of the western interior of North America.
Annals of the Missouri Botanical Garden 71: 351–383.
Ditta, G., A. Pinyopich, P. Robles, S. Pelaz, and M. F. Yanofsky.
2004. The SEP4 gene of Arabidopsis thaliana functions in floral or-
gan and meristem identity. Current Biology 14: 1935–1940.
Donoghue, M., R. Ree, and D. Baum. 1998. Phylogeny and the evolu-
tion of flower symmetry in the Asteridae. Trends in Plant Science 3:
311–317.
Doyle, J. A. 1994. Origin of the angiosperm flower: A phylogenetic per-
spective. Plant Systematics and Evolution (Supplement) 8: 7–29.
Doyle, J. A. 1996. Seed plant phylogeny and the relationships of Gnetales.
International Journal of Plant Sciences 157: S3–S39.
Doyle, J. A. 2006. Seed ferns and the origin of angiosperms. Journal of
the Torrey Botanical Society 133: 169–209.
Doyle, J. A., and M. J. Donoghue. 1986. Seed plant phylogeny and the
origin of the angiosperms: An experimental cladistic approach in the
angiosperms. Botanical Review 52: 321–431.
[Vol. 96
Doyle, J. A., and P. K. Endress. 2000. Morphological phylogenetic
analysis of basal angiosperms: Comparison and combination with mo-
lecular data. International Journal of Plant Sciences 161 (Supplement
6): S121–S153.
Doyle, J. A., H. Eklund, and P. S. Herendeen. 2003. Floral evolution
in Chloranthaceae: Implications of a morphological phylogenetic
analysis. International Journal of Plant Sciences 164 (Supplement 5):
S365–S382.
Drinnan, A. N., P. R. Crane, and S. B. Hoot. 1994. Patterns of flo-
ral evolution in the early diversification of non-magnoliid dicoty-
ledons (eudicots). Plant Systematics and Evolution (Supplement)
8: 93–122.
Duarte, J. M., P. K. Wall, L. M. Zahn, P. S. Soltis, D. E. Soltis, J.
Leebens-Mack, H. Ma, and C. W. dePamphilis. In press. Utility of
Amborella trichopoda and Nuphar advena ESTs for comparative se-
quence analysis. Taxon.
Eames, A. J. 1931. The vascular anatomy of the flower with refutation of
the theory of carpel polymorphism. American Journal of Botany 18:
147–188.
Eames, A. J. 1961. Morphology of the angiosperms. McGraw-Hill, New
York, New York, USA.
Eklund, H., J. A. Doyle, and P. S. Herendeen. 2004. Morphological phy-
logenetic analysis of living and fossil Chloranthaceae. International
Journal of Plant Sciences 165: 107–151.
Endress, P. K. 1982. Syncarpy and alternative modes of escaping disad-
vantages of apocarpy in primitve angiosperms. Taxon 31: 48–52.
Endress, P. K. 1987. Floral phyllotaxis and floral evolution. Botanische
Jahrbucher fur Systematik 108: 417–438.
Endress, P. K. 1990a. Patterns of floral construction in ontogeny and
phylogeny. Biological Journal of the Linnean Society 39: 153–175.
Endress, P. K. 1990b. Evolution of reproductive structures and functions
in primitive angiosperms (Magnoliidae). Memoirs of the New York
Botanical Garden 55: 5–34.
Endress, P. K. 1994. Floral structure and evolution of primitive angio-
sperms: Recent advances. Plant Systematics and Evolution 192:
79–97.
Endress, P. K. 1996. Diversity and evolutionary biology of tropical flow-
ers. Cambridge University Press, Cambridge, UK.
Endress, P. K. 1999. Symmetry in flowers: Diversity and evolution.
International Journal of Plant Sciences 160: S3–S23.
Endress, P. K. 2001. The flowers in extant basal angiosperms and infer-
ences on ancestral flowers. International Journal of Plant Sciences
162: 1111–1140.
Endress, P. K. 2005. Carpels in Brasenia (Cabombaceae) are com-
pletely ascidiate despite a long stigmatic crest. Annals of Botany 96:
209–215.
Endress, P. K. 2006. Angiosperm floral evolution: Morphological devel-
opmental framework. Advances in Botanical Research 44: 1–61.
Endress, P. K., and J. A. Doyle. 2009. Reconstructing the ancestral an-
giosperm flower and its initial specializations. American Journal of
Botany 96: 22–66.
Endress, P. K., and A. Igersheim. 2000. Gynoecium structure and evo-
lution in basal angiosperms. International Journal of Plant Sciences
161: S211–S223.
Endress, P. K., M. Jenny, and M. E. Fallen. 1983. Convergent elaboration
of apocarpous gynoecia in higher advanced dicotyledons (Sapindales,
Malvales, Gentianales). Nordic Journal of Botany 3: 293–300.
Endress, P. K., and M. L. Matthews. 2006. First steps towards a flo-
ral structural characterization of the major rosid subclades. Plant
Systematics and Evolution 260: 223–251.
Feild, T. S., N. C. Arens, J. A. Doyle, T. E. Dawson, and M. J.
Donoghue. 2004. Dark and disturbed: A new image of early angio-
sperm ecology. Paleobiology 30: 82–107.
Feild, T. S., M. A. Zweiniecki, T. Brodribb, T. Jaffré, M. J. Donoghue,
and N. M. Holbrook. 2000. Structure and function of tracheary
elements in Amborella trichopoda. International Journal of Plant
Sciences 161: 705–712.
Feng, X., Z. Zhao, Z. Tian, S. Xu, Y. Luo, Z. Cai, Y. Wang, et al. 2006.
Control of petal shape and floral zygomorphy in Lotus japonicus.
January 2009] Soltis et al.—Floral variation and floral genes
Proceedings of the National Academy of Sciences, USA 103:
4970–4975.
Floyd, S. K., and W. E. Friedman. 2000. Evolution of endosperm de-
velopmental patterns among basal flowering plants. International
Journal of Plant Sciences 161: S57–S81.
Fourquin,C.,M.Vinauger-Douard,P.Chambrier,A.Berne-Dedieu,and
C. P. Scutt. 2007. Functional conservation between CRABS CLAW
orthologues from widely diverged angiosperms. Annals of Botany
100: 651–657.
Fourquin, C., M. Vinauger-Douard, B. Fogliani, C. Dumas, and C. P.
Scutt. 2005. Evidence that CRABS CLAW and TOUSLED have con-
served their roles in carpel development since the ancestor of the ex-
tant angiosperms. Proceedings of the National Academy of Sciences,
USA 102: 4649–4654.
Friedman, W. E. 2008. Hydatellaceae are water lilies with gymnosper-
mous tendencies. Nature 453: 94–97.
Friis, E. M., P. R. Crane, and K. R. Pedersen. 1986. Floral evidence for
Cretaceous chloranthoid angiosperms. Nature 320: 163–164.
Friis, E. M., and P. K. Endress. 1990. Origin and evolution of angiosperm
flowers. Advances in Botanical Research 17: 99–162.
Friis, E. M., K. R. Pedersen, and P. R. Crane. 2000. Reproductive struc-
ture and organization of basal angiosperms from the Early Cretaceous
(Barremian or Aptian) of western Portugal. International Journal of
Plant Sciences 161: S169–S182.
Friis, E. M., K. R. Pedersen, and P. R. Crane. 2001. Fossil evidence
of waterlilies (Nymphaeales) in the Early Cretaceous. Nature 410:
357–360.
Friis, E. M., K. R. Pedersen, and P. R. Crane. 2004. Araceae from the
Early Cretaceous of Portugal: Evidence on the emergence of mono-
cotyledons. Proceedings of the National Academy of Sciences, USA
101: 16565–16570.
Frohlich, M. W. 2006. Recent developments regarding the evolutionary
origins of flowers. Advances in Botanical Research 44: 63–127.
Frohlich, M. W., and D. S. Parker. 2000. The mostly male theory of
flower evolutionary origins: From genes to fossils. Systematic Botany
25: 155–170.
Frumin, S., and E. M. Friis. 1999. Magnoliid reproductive organs from
the Cenomanian-Turonian of north-western Kazakhstan: Magnoliaceae
and Illiciaceae. Plant Systematics and Evolution 216: 265–288.
Furness, C. A., and P. J. Rudall. 2001. The tapetum in basal angio-
sperms: Early diversity. International Journal of Plant Sciences 162:
375–392.
Galego, L., and J. Almeida. 2002. Role of DIVARICATA in the con-
trol of dorsoventral asymmetry in Antirrhinum flowers. Genes &
Development 16: 880–891.
Gandolfo, M. A., K. C. Nixon, and W. L. Crepet. 2004. The oldest complete
fossil flowers of Nymphaeaceae and implications for the complex insect
entrapment pollination mechanisms in early angiosperms. Proceedings
of the National Academy of Sciences, USA 101: 8056–8060.
Goebel, K. 1933. Organographie der Pflanzen, vol. 3, A, III. G. Fischer,
Jena, Germany.
Graham, S. W., and W. J. Iles. 2009. Different gymnosperm outgroups
have (mostly) congruent signal regarding the root of flowering-plant
phylogeny. American Journal of Botany 96: 216–227.
Graham, S. W., and R. G. Olmstead. 2000. Utility of 17 chloroplast
genes for inferring the phylogeny of the basal angiosperms. American
Journal of Botany 87: 1712–1730.
Graham, S. W., P. A. Reeves, A. C. E. Burns, and R. G. Olmstead.
2000. Microstructural changes in noncoding chloroplast DNA:
Interpretation, evolution, and utility of indels and inversions in basal
angiosperm phylogenetic inference. International Journal of Plant
Sciences 161 (Supplement 6): S83–S96.
Hamby, R. K., and E. A. Zimmer. 1992. Ribosomal RNA as a phylogenetic
tool in plant systematics. In P. S. Soltis, D. E. Soltis, and J. J. Doyle
[eds.], Molecular systematics of plants, 50–91. Chapman and Hall,
New York, New York, USA.
Hesse, M. 2001. Pollen characters of Amborella trichopoda
(Amborellaceae): A reinvestigation. International Journal of Plant
Sciences 162: 201–208.
125
Hiepko, P. 1965. Vergleichend-morphologische und entwicklungs-ge-
schichtliche Untersuchungen uber das Perianth bei den Polycarpicae.
Botanishce Jahrbücher für Systematik 84: 359–508.
Hileman, L. C., E. M. Kramer, and D. A. Baum. 2003. Differential
regulation of symmetry genes and the evolution of floral morpholo-
gies. Proceedings of the National Academy of Sciences, USA 100:
12814–12819.
Hilton, J., and R. M. Bateman. 2006. Pteridosperms are the backbone of
seed-plant phylogeny. Journal of the Torrey Botanical Society 133:
119–168.
Hilu, K. W., T. Borsch, K. Muller, D. E. Soltis, P. S. Soltis, V.
Savolainen, M. Chase, et al. 2003. Inference of angiosperm phylog-
eny based on matK sequence information. American Journal of Botany
90: 1758–1776.
Howarth, D. G., and M. J. Donoghue. 2006. Phylogenetic analysis of the
‘‘ECE’’ (CYC_TB1) clade reveals duplications predating the core eu-
dicots. Proceedings of the National Academy of Sciences, USA 103:
9101–9106.
Irish, V. 2006. Duplication, diversification, and comparative genetics of
angiosperm MADS-box genes. Advances in Botanical Research 44:
127–159.
Jansen, R. K., Z. Cai, L. A. Raubeson, H. Daniell, C. W. dePamphilis, J.
Leebens-Mack, K. F. Muller, et al. 2007. Analysis of 81 genes from
64 plastid genomes resolves relationships in angiosperms and identi-
fies genome-scale evolutionary patterns. Proceedings of the National
Academy of Sciences, USA 104: 19369–19374.
Jaramillo, M. A., and E. M. Kramer. 2004. APETALA3 and PISTILLATA
homologs exhibit novel expression patterns in the unique perianth in
Aristolochia (Aristolochiaceae). Evolution & Development 6: 449–458.
Katoh, K., K. Misawa, K. Kuma, and T. Miyata. 2002. MAFFT: A
novel method for rapid multiple sequence alignment based on fast
Fourier transform. Nucleic Acids Research 30: 3059–3066.
Katoh, K., and H. Toh. 2008. Recent developments in the MAFFT multiple
sequence alignment program. Briefings in Bioinformatics 9: 286–298.
Kim, S., V. A. Albert, M.-J. Yoo, J. S. Farris, P. S. Soltis, and D. E.
Soltis. 2004. Phylogeny and diversification of B-function MADS-
box genes in angiosperms: Evolutionary and functional implications
of a 260-million-year-old duplication. American Journal of Botany 91:
2102–2118.
Kim, S., J. Koh, M.-J. Yoo, H. Kong, Y. Hu, H. Ma, P. S. Soltis, and
D. E. Soltis. 2005. Expression of floral MADS-box genes in basal
angiosperms: Implications for the evolution of floral regulators. Plant
Journal 43: 724–744.
Kozo-Poljanski, B. M. 1922. An introduction to phylogenetic system-
atics of the higher plants. Nature and Culture, Voronezh, USSR [in
Russian].
Kramer, E. M., R. L. Dorit, and V. F. Irish. 1998. Molecular evolution
of genes controlling petal and stamen development: Duplication and
divergence within the APETALA3 and PISTILLATA MADS-box gene
lineages. Genetics 149: 765–783.
Kramer, E. M., and J. C. Hall. 2005. Evolutionary dynamics of genes
controlling floral development. Current Opinion in Plant Biology 8:
13–18.
Kramer, E. M., L. Holappa, B. Gould, M. A. Jaramillo, D. Setnikov,
and P. M. Santiago. 2007. Elaboration of B gene function to in-
clude the identity of novel floral organs in the lower eudicot Aquilegia.
Plant Cell 19: 750–766.
Kramer, E. M., M. A. Jaramillo, and V. S. Di Stilio. 2004. Patterns of
gene duplication and functional evolution during the diversification
of the AGAMOUS subfamily of MADS box genes in angiosperms.
Genetics 166: 1011–1023.
Kramer, E. M., and E. A. Zimmer. 2006. Gene duplication and floral
developmental genetics of basal eudicots. Advances in Botanical
Research 44: 354–376.
Krogan, N. T., and N. W. Ashton. 2000. Ancestry of plant MADS-box
genes revealed by bryophyte (Physcomitrella patens) homologues.
New Phytologist 147: 505–517.
Leebens-Mack, J., L. A. Raubeson, L. Cui, J. V. Kuehl, M. H. Fourcade,
T. W. Chumley, J. L. Boore et al. 2005. Identifying the basal
126 American Journal of Botany
angiosperm node in chloroplast genome phylogenies: Sampling one’s
way out of the Felsenstein zone. Molecular Biology and Evolution
22: 1948–1963.
Leibfried, A., J. P. C. To, W. Busch, S. Stehling, A. Kehle, M. Demar,
J. J. Kieber, and J. U. Lohmann. 2005. WUSCHEL controls mer-
istem function by direct regulation of cytokinin-inducible response
regulators. Nature 438: 1172–1175.
Litt, A., and V. F. Irish. 2003. Duplication and diversification in the
APETALA1/FRUITFULL floral homeotic gene lineage: Implications
for the evolution of floral development. Genetics 165: 821–833.
Litz, R. E., R. S. Witjaksono, D. Efendi, F. Pliego Alfar, and A. Barceló-
Muñoz. 2005. Persea americana, avocado. In R. E. Litz [ed.],
Biotechnology of fruit and nut crops, 326–349. CAB International,
Wallingford, UK.
Luo, D., R. Carpenter, L. Copsey, C. Vincent, J. Clark, and E. Coen.
1999. Control of organ asymmetry in flowers of Antirrhinum. Cell
99: 367–376.
Luo, D., R. Carpenter, C. Vincent, L. Copsey, and E. Coen. 1996. Origin
of floral asymmetry in Antirrhinum. Nature 383: 794–799.
Lynch, M., and J. S. Conery. 2000. The evolutionary fate and conse-
quences of duplicate genes. Science 290: 1151–1155.
Mathews, S., and M. J. Donoghue. 1999. The root of angiosperms phylogeny
inferred from duplicate phytochrome genes. Science 286: 947–950.
Mathews, S., and M. J. Donoghue. 2000. Basal angiosperm phylogeny
inferred from duplicate phytochromes A and C. International Journal
of Plant Sciences 161: S41–S55.
Melchior, H. [ed.]. 1964. A. Engler’s Syllabus der Pflanzenfamilien, 12th
ed. Borntraeger, Berlin, Germany.
Moore, M. J., C. D. Bell, P. S. Soltis, and D. E. Soltis. 2007. Using
plastid genomic-scale data to resolve enigmatic relationships among
basal angiosperms. Proceedings of the National Academy of Sciences,
USA 104: 19363–19368.
Münster, T., W. Faigl, H. Saedler, and G. Theissen. 2002. Evolutionary
aspects of MADS-box genes in the eusporangiate fern Ophioglossum.
Plant Biology 4: 474–483.
Nagasawa, N., M. Miyoshi, Y. Sano, H. Satoh, H. Hirano, H. Sakai, and
Y. Nagato. 2003. SUPERWOMAN1 and DROOPING LEAF genes
control floral organ identity in rice. Development 130: 705–718.
Neal, P. R., A. Dafni, and M. Giurfa. 1998. Floral symmetry and its
role in plan–pollinator systems: Terminology, distribution, and hy-
potheses. Annual Review of Ecology and Systematics 29: 345–373.
Ng, M., and M. F. Yanofsky. 2001. Function and evolution of the plant
MADS-box gene family. Nature Reviews Genetics 2: 186–195.
Nickrent, D. L., A. Blarer, Y.-L. Qiu, D. E. Soltis, P. S. Soltis, and M.
Zanis. 2002. Molecular data place Hydnoraceae with Aristolochiaceae.
American Journal of Botany 89: 1809–1817.
Ohno, S. 1970. Evolution by gene duplication. Springer, Berlin,
Germany.
Parkinson, C. L., K. L. Adams, and J. D. Palmer. 1999. Multigene anal-
yses identify the three earliest lineages of extant flowering plants.
Current Biology 9: 1485–1488.
Pelaz, S., G. S. Ditta, E. Baumann, E. Wisman, and M. F. Yanofsky.
2000. B and C floral organ identity functions require SEPALLATA
MADS-box genes. Nature 405: 200–203.
Perkins, J. 1925. Ubersicht uber die Gattungen der Monimiaceae.
Engelmann, Leipzig, Germany.
Pinyopich, A., G. S. Ditta, B. Savidge, S. J. Liljegren, E. Baumann, E.
Wisman, and M. F. Yanofsky. 2003. Assessing the redundancy of
MADS-box genes during carpel and ovule development. Nature 424:
85–88.
Posluszny, U. B., and P. B. Tomlinson. 2003. Aspects of inflorescence
and floral development in the putative basal angiosperm Amborella
trichopoda (Amborellaceae). Canadian Journal of Botany 81: 28–39.
Qiu, Y., O. Dombrovska, J. Lee, L. Li, B. A. Whitlock, F. Bernasconi-
Quadroni, J. S. Rest, et al. 2005. Phylogenetic analysis of basal
angiosperms based on nine plastid, mitochondrial, and nuclear genes.
International Journal of Plant Sciences 166: 815–842.
Qiu, Y.-L., J. Lee, F. Bernasconi-Quadroni, D. E. Soltis, P. S. Soltis, M.
Zanis, E. A. Zimmer. et al. 1999. The earliest angiosperms: Evidence
[Vol. 96
from mitochondrial, plastid, and nuclear genomes. Nature 402:
404–407.
Qiu, Y. L., J. Lee, F. Bernasconi-Quadroni, D. E. Soltis, P. S. Soltis,
M. Zanis, E. A.Zimmer, et al. 2000. Phylogeny of basal angiosperms:
Analyses of five genes from three genomes. International Journal of
Plant Sciences 161: S3–S27.
Qiu, Y.-L., J. Lee, B. A. Whitlock, F. Bernasconi-Quadroni, and O.
Dombrovska. 2001. Was the ANITA rooting of the angiosperm phy-
logeny affected by long branch attraction? Molecular Biology and
Evolution 18: 1745–1753.
Rasmussen, D. A., E. M. Kramer, and E. A. Zimmer. 2009. One size
fits all? Molecular evidence for a commonly inherited petal iden-
tity program in the ranunculids. American Journal of Botany 96:
96–109.
Ree, R. H., and M. J. Donoghue. 1999. Inferring rates of change in
flower symmetry in asterid angiosperms. Systematic Biology 48:
633–641.
Remane, A. 1956. Die Grundlagen des natürlichen Systems, der
vergleichenden Anatomie und der Phylogenetik. Akademische
Verlagsbuchhandlung Geest & Portig, Leipzig, Germany.
Renner, S. 1999. Circumscription and phylogeny of the Laurales:
Evidence from molecular and morphological data. American Journal
of Botany 86: 1301–1315.
Rijpkema, A., T. Gerats, and M. Vandenbussche. 2006. Genetics of
floral development in Petunia. Advances in Botanical Research 44:
237–278.
Rohwer, J. G. 1993. Lauraceae. In K. Kubitzki [ed.], The families
and genera of vascular plants, 366–391. Springer-Verlag, Berlin,
Germany.
Ronse De Craene, L. P. 2007. Are petals sterile stamens or bracts? The
origin and evolution of petals in the core eudicots. Annals of Botany
100: 621–630.
Ronse De Craene, L. P., and E. Smets. 1994. Merosity in flowers:
Definition, origin, and taxonomic significance. Plant Systematics and
Evolution 191: 83–104.
Ronse De Craene, L. P., and E. Smets. 1995. The distribution and sys-
tematic relevance of the androecial character oligomery. Botanical
Journal of the Linnean Society 118: 193–247.
Ronse De Craene, L. P., P. S. Soltis, and D. E. Soltis. 2003. Evolution of
floral structures in basal angiosperms. International Journal of Plant
Sciences 164: S329–S363.
Rudall, P. J., M. V. Remizowa, A. S. Beer, E. Bradshaw, D. W. Stevenson,
T. D. Macfarlane, R. E. Tuckett, et al. 2008. Comparative ovule
and megagametophyte development in Hydatellaceae and water lilies
reveal a mosaic of features among the earliest angiosperms. Annals of
Botany 101: 941–956.
Rudall, P. J., D. D. Sokoloff, M. V. Remizowa, J. G. Conran, J. I. Davis,
T. D. Macfarlane, and D. W. Stevenson. 2007. Morphology of
Hydatellaceae, an anomalous aquatic family recently recognized as
an early-divergent angiosperm lineage. American Journal of Botany
94: 1073–1092.
Running, M. P., and E. M. Meyerowitz. 1996. Mutations in the
PERIANTHIA gene of Arabidopsis specifically alter floral organ num-
ber and initiation pattern. Development 122: 1261–1269.
Saarela, J. M., H. S. Rai, J. A. Doyle, P. K. Endress, S. Mathews, A. D.
Marchant, B. G. Briggs, and S. W. Graham. 2007. Hydatellaceae
identified as a new branch near the base of the angiosperm phyloge-
netic tree. Nature 446: 312–315.
Sauquet, H., J. A. Doyle, T. Scharaschkin, T. Borsch, K. W. Hilu, L. W.
Chatrou,andA.LeThomas.2003. PhylogeneticanalysisofMagnoliales
and Myristicaceae based on multiple data sets: Implications for character
evolution. Botanical Journal of the Linnean Society 142: 125–186.
Savolainen, V., M. Chase, C. M. Morton, D. E. Soltis, C. Bayer, M. F.
Fay, A. De Bruijn, S. Sullivan, and Y.-L. Qiu. 2000. Phylogenetics
of flowering plants based upon a combined analysis of plastid atpB and
rbcL gene sequences. Systematic Biology 49: 306–362.
Sessions, A., J. L. Nemhauser, A. McColl, J. L. Roe, K. A. Feldmann, and
P. C. Zmabryski. 1997. ETTIN patterns the Arabidopsis floral mer-
istem and reproductive organs. Development 124: 4481–4491.
January 2009] Soltis et al.—Floral variation and floral genes
Smith, G. H. 1928. Vascular anatomy of Ranalian flowers. II.
Ranunculaceae (continued), Menispermaceae, Calycanthaceae,
Annonaceae. Botanical Gazette (Chicago, Ill.) 85: 152–177.
Soltis, D. E., V. A. Albert, J. Leebens-Mack, C. D. Bell, A. Paterson,
C. Zheng, D. Sankoff, et al. 2009. Polyploidy and angiosperm diver-
sification. American Journal of Botany 96: 336–348.
Soltis, D. E., V. A. Albert, J. Leebens-Mack, J. D. Palmer, R. A.
Wing, C. W. dePamphilis, H. Ma, et al. 2008. The Amborella ge-
nome: An evolutionary reference for plant biology. Genome Biology
9: 402.
Soltis, D. E., A. Chanderbali, S. Kim, M. Buzgo, and P. S. Soltis.
2007. The ABC model and its applicability to basal angiosperms.
Annals of Botany 100: 155–163.
Soltis, D. E., J. Leebens-Mack, and P. S. Soltis [eds.]. 2006. Develop-
mental genetics of the flower. Advances in Botanical Research 44.
Academic Press/Elsevier, San Diego, California, USA.
Soltis, D. E., A. E. Senters, S. Kim, J. D. Thompson, P. S. Soltis, M. J. Zanis, L.
Ronse De Craene, et al. 2003. Gunnerales are sister to other core eudicots
and exhibit floral features of early-diverging eudicots. American Journal of
Botany 90: 461–470.
Soltis, D. E., and P. S. Soltis. 1990. Isozyme evidence for ancient poly-
ploidy in primitive angiosperms. Systematic Botany 15: 328–337.
Soltis, D. E., P. S. Soltis, M. W. Chase, M. E. Mort, D. C. Albach, M.
Zanis, V. Savolainen, et al. 2000. Angiosperm phylogeny inferred
from a combined data set of 18S rDNA, rbcL, and atpB sequences.
Botanical Journal of the Linnean Society 133: 381–461.
Soltis, D. E., P. S. Soltis, P. K. Endress, and M. W. Chase. 2005. Angiosperm
phylogeny and evolution. Sinauer, Sunderland, Massachusetts, USA.
Soltis, D. E., P. S. Soltis, D. L. Nickrent, L. A. Johnson, W. J. Hahn, S.
B. Hoot, J. A. Sweere, et al. 1997. Angiosperm phylogeny inferred
from 18S ribosomal DNA sequences. Annals of the Missouri Botanical
Garden 84: 1–49.
Soltis, P. S., and D. E. Soltis. 2004. The origin and diversification of
angiosperms. American Journal of Botany 91: 1614–1626.
Soltis, P. S., D. E. Soltis, and M. W. Chase. 1999. Angiosperm phylog-
eny inferred from multiple genes: A research tool for comparative bi-
ology. Nature 402: 402–404.
Soltis, P. S., D. E. Soltis, S. Kim, A. Chanderbali, and M. Buzgo.
2006. Expression of floral regulators in basal angiosperms and the ori-
gin and evolution of the ABC model. Advances in Botanical Research
44: 483–506.
Soltis, P. S., D. E. Soltis, M. J. Zanis, and S. Kim. 2000. Basal lineages
of angiosperms: Relationships and implications for floral evolution.
International Journal of Plant Sciences 161: S97–S107.
Stebbins, G. L. 1950. Variation and evolution in plants. Columbia
University Press, New York, New York, USA.
Stebbins, G. L. 1970. Adaptive radiation of reproductive characteristics in
angiosperms, I: Pollination mechanisms. Annual Review of Ecology
and Systematics 1: 307–326.
Stebbins, G. L. 1974. Flowering plants: Evolution above the species level.
Belknap Press, Cambridge, Massachusetts, USA.
Stevens, P. F. 2001 onward. Angiosperm phylogeny website, version 9,
June 1, 2008 [more or less continuously updated]. Website http://www.
mobot.org/MOBOT/research/APweb/.
Sun, G., D. Dilcher, S. Zheng, and Z. Zhou. 1998. In search of the first
flower: A Jurassic angiosperm, Archaefructus, from northeast China.
Science 282: 1692–1695.
Sun, G., Q. Li, D. L. Dilcher, S. Zheng, K. C. Nixon, and X. Wang.
2002. Archaefructaceae, a new basal angiosperm family. Science
296: 899–904.
Takhtajan, A. 1968. Classification and phylogeny with special reference
to the flowering plants. Proceedings of the Linnean Society of London
179: 221–227.
Takhtajan, A. 1987. System of Magnoliophyta. Academy of Sciences
USSR, Leningrad, USSR.
Takhtajan, A. 1991. Evolutionary trends in flowering plants. Columbia
University Press, New York, New York, USA.
Takhtajan, A. 1997. Diversity and classification of flowering plants.
Columbia University Press, New York, New York, USA.
127
Taylor, D. W., and L. J. Hickey. 1992. Phylogenetic evidence for the
herbaceous origin of angiosperms. Plant Systematics and Evolution
180: 137–156.
Teeri, T. H., M. Kotilainen, A. Uimari, S. Ruokolainen, Y. P. Ng,
U. Malm, E. Pollanen, et al. 2006. Floral developmental genet-
ics of Gerbera (Asteraceae). Advances in Botanical Research 44:
323–351.
Theissen, G. 2001. Development of floral organ identity: Stories from the
MADS house. Current Opinion in Plant Biology 4: 75–85.
Theissen, G., A. Becker, A. Di Rosa, A. Kanno, J. T. Kim, T. Münster,
K.-U. Winter, and H. Saedler. 2000. A short history of MADS-
box genes in plants. Plant Molecular Biology 42: 115–149.
Theissen, G., A. Becker, C. Kirchner, T. Münster, K.-U. Winter,
and H. Saedler. 2002. How land plants learned their floral ABCs:
The role of MADS-box genes in the evolutionary origin of flow-
ers. In Q. C. B. Cronk, R. M. Bateman, and J. A. Hawkins [eds.],
Developmental genetics and plant evolution, 173–205. Taylor &
Francis, London, UK.
Theissen, G., and R. Melzer. 2007. Molecular mechanisms underlying
origin and diversification of the angiosperm flower. Annals of Botany
100: 603–619.
Thien, L. 2003. The population structure and floral biology of Amborella
trichopoda Baillon (Amborellaceae). Annals of the Missouri Botanical
Garden 90: 466–490.
Thorne, R. F. 1974. A phylogenetic classification of the Annoniflorae.
Aliso 8: 147–209.
Troll, W. 1927. Zur Frage nach der Herkunft der Blumenblatter. Flora
122: 57–75.
Troll, W. 1928. Organisation und Gestalt im Bereich der Blute. Springer,
Berlin, Germany.
Tucker, S. C. 1999. Evolutionary lability of symmetry in early flo-
ral development. International Journal of Plant Sciences 160:
S25–S39.
Vogel, S. 1959. Organographie der Blüten kapländischer Ophrydeen
mit Bemerkungen zum Koaptations-Problem I/II. Abhandlungen der
Akademie der Wissenschaften und der Literatur Mainz Jahrbuch
1960: 81–94.
Vogel, S. 1969. Über synorganisierte Blütensporne bei einigen Orchideen.
Österreichische Botanische Zeitschrift 116: 244–262.
Walker, J. W., and A. G. Walker. 1984. Ultrastructure of Lower Cretaceous
angiosperm pollen and the origin and early evolution of flowering
palnts. Annals of the Missouri Botanical Garden 71: 464–521.
Wang, Z., Y. Luo, X. Li, L. Wang, S. Xu, J. Yang, L. Weng, et al.
2008. Genetic control of floral zygomorphy in pea (Pisum sativum
L.). Proceedings of the National Academy of Sciences, USA 105:
10414–10419.
Whipple,C.J.,M.J.Zanis,E.A.Kellogg,andR.J.Schmidt.2007. Conservation
of B class gene expression in the second whorl of a basal grass and out-
groups links the origin of lodicules and petals. Proceedings of the National
Academy of Sciences, USA 104: 1081–1086.
Wikström, N., V. Savolainen, and M. W. Chase. 2001. Evolution of the
angiosperms: Calibrating the family tree. Proceedings of the Royal
Society of London, B, Biological Sciences 268: 2211–2220.
Williams, J. H., and W. E. Friedman. 2002. Identification of diploid en-
dosperm in an early angiosperm lineage. Nature 415: 522–526.
Worsdell, W. C. 1903. The origin of the perianth of flowers with special
reference to the Ranunculaceae. New Phytologist 2: 42–48.
Yamada, T., M. Ito, and M. Kato. 2004. YABBY2-homologue expres-
sion in lateral organs of Amborella trichopoda (Amborellaceae).
International Journal of Plant Sciences 165: 917–924.
Yamada, T., H. Nishida, M. Umebayashi, K. Uemura, and M. Kato.
2008. Oldest record of Trimeniaceae from the Early Cretaceous of
northern Japan. BMC Evolutionary Biology 8: 135.
Zahn, L. M., B. Feng, and H. Ma. 2006a. Beyond the ABC model:
Regulation of floral homeotic genes. Advances in Botanical Research
44: 163–207.
Zahn, L. M., H. Kong, J. H. Leebens-Mack, S. Kim, P. S. Soltis, L.
L. Landherr, D. E. Soltis, and C. W. dePamphilis. 2005. The
evolution of the SEPALLATA subfamily of MADS-box genes: A pre-
128 American Journal of Botany
angiosperm origin with multiple duplications throughout angiosperm
history. Genetics 169: 2209–2223.
Zahn, L. M., J. H. Leebens-Mack, J. M. Arrington, Y. Hu, L. L.
Landherr, C. W. dePamphilis, A. Becker, et al. 2006b. Conservation
and divergence in the AGAMOUS subfamily of MADS-box genes:
Evidence of independent sub- and neofunctionalization events.
Evolution & Development 8: 30–45.
Zanis, M. J., D. E. Soltis, P. S. Soltis, S. Mathews, and M. J.
Donoghue. 2002. The root of the angiosperms revisited. Proceedings
of the National Academy of Sciences, USA 99: 6848–6853.
View publication stats
Zanis, M. J., D. E. Soltis, P. S. Soltis, Y.-L. Qiu, and E. A. Zimmer.
2003. Phylogenetic analyses and perianth evolution in basal angio-
sperms. Annals of the Missouri Botanical Garden 90: 129–150.
Zhao, D., Q. Yu, C. Chen, and H. Ma. 2001. Genetic control of re-
productive meristems. In M. T. McManus and B. Veit [eds.],
Meristematic tissues in plant growth and development, p. 89–142.
Sheffield Academic Press, Sheffield, UK.
Zwickl, D. J. 2006. GARLI. Genetic algorithm for rapid likelihood infer-
ence. Computer program distributed by author, website http://www.
bio.utexas.edu/grad/zwickl/web/garli.html.