86

Evolution of floral symmetry

Peter K Endress
Floral monosymmetry, which is conspicuous and prominent
in many angiosperms, has attracted much attention from
both developmental geneticists and pollination biologists.
A combined evolutionary biological approach to studying floral
monosymmetry in the Lamiales, the order that contains the
model plant Antirrhinum, is just beginning to take shape. In
contrast, floral leftright asymmetry has largely been neglected,
although it is much in discussion in animal biology, probably
because in flowers (unlike in animals) leftright asymmetry is
not predominant. Nevertheless, there are patterns in the
evolution of floral leftright asymmetry that are interesting
enough to be addressed by developmental genetics. These are
the direction of contortion in flowers with contort petal
aestivation, and the direction of deflection of pollination organs
in groups with enantiostylous flowers or in some groups with
enclosed pollination organs, such as beans (Phaseolinae,
Fabaceae) or louseworts (Pedicularis, Orobanchaceae).
Addresses
Institute of Systematic Botany, University of Zurich, Zollikerstrasse 107,
8008 Zurich, Switzerland; e-mail: pendress@systbot.unizh.ch
Current Opinion in Plant Biology 2001, 4:8691
1369-5266/01/$ see front matter
2001 Elsevier Science Ltd. All rights reserved.

Introduction
Floral symmetry is not only aesthetically attractive but is of
eminent biological significance. Research into symmetry in
plants has attracted much attention over the past few years
and is being approached from different directions [1,2,3].
In this review, I distinguish three kinds of symmetry: polysymmetry (i.e. actinomorphy, with several symmetry
planes), monosymmetry (i.e. zygomorphy, with one symmetry plane), and asymmetry (i.e. without any symmetry
plane) (Figure 1). In most of the recent publications, only
polysymmetry and monosymmetry are discussed, and are
referred to as symmetry and asymmetry, respectively [4,5].
Thus, the term asymmetry has not always been used in the
same sense. The development of monosymmetry is being
investigated using molecular developmental genetics, starting with the model plant Antirrhinum [1,68]. The
significance of monosymmetric flowers is also being studied from the point of view of pollination biology
[911,12,13]. Another kind of symmetry in flowers is
leftright asymmetry (i.e. enantiomorphy: with left and
right morphs). The developmental biology of leftright
asymmetry is being intensively studied in animals [14] but,
as yet, has attracted little attention in plants apart from
some studies on pollination biology. Some evolutionary
aspects of floral leftright asymmetry were, however, discussed recently [15,16]. In Arabidopsis, tortifolia mutants
produce leftright asymmetry in petioles by twisted growth
[17]. This seems to be the only example in which leftright

Figure 1

Polysymmetric

Monosymmetric

Asymmetric
Current Opinion in Plant Biology

The three forms of floral symmetry addressed in this review. Planes of

symmetry are indicated by broken lines.

asymmetry in plants has been addressed from a molecular

developmental point of view.

Evolution of floral monosymmetry

From our knowledge of angiosperm fossil history and phylogeny, we know that polysymmetric flowers predated
monosymmetric flowers; the first clearly monosymmetric
flowers are known from the Upper Cretaceous (Turonian)
period, that is 3040 million years after the first known
floral fossils, which were polysymmetric [1820]. Floral
monosymmetry has originated many times from
polysymmetry and several large, successful clades with
predominantly or entirely monosymmetric flowers have
evolved, such as orchids, legumes, Dipsacales, Lamiales,
and Asteraceae. Monosymmetric flowers are commonly
presented to the side in such a way that they have an upper
(i.e. adaxial) and a lower (i.e. abaxial) half that are different
in shape. Floral monosymmetry is a successful system
because of its potential for efficient precision mechanisms
in pollination biology. It has also been shown that bees
have a preference for symmetrical (i.e. monosymmetric
and polysymmetric) patterns, which is either innate or
learned [11,12,13]. It has been argued that the evolution
of monosymmetry in the mentioned clades was mediated
by bee pollination, at least in the initial stages.
The molecular developmental genetics of monosymmetry
was first elucidated in the model species Antirrhinum majus,
that is snapdragon (Antirrhinaceae). Antirrhinum flowers
consist of three five-merous whorls of sepals, petals and stamens, and a two-merous whorl of carpels. Monosymmetry
is expressed by the weaker early development of the upper
half of the flower relative to the lower half, which results in
the suppression of the upper (odd) stamen; in the mature
flower it is only present as a tiny staminode. It was found
that two genes, cycloidea and dichotoma, play a major role in
this monosymmetry: if both of these genes fail to function,
the flowers become polysymmetric [1,4,8,21,22]. Such

Evolution of floral symmetry Endress

secondarily polysymmetric flowers in Antirrhinaceae and

other families are called peloriae because Linnaeus, 250
years ago, created the generic name Peloria for a population
of Linaria vulgaris (also Antirrhinaceae) with monstrous,
polysymmetric flowers. The two genes, cycloidea and
dichotoma, differ slightly in the period during which they are
active and in the extent of the floral sector that they influence [21]. The two genes are similar in structure and,
together with some other genes, may have originated by
gene duplication [21,23,24].
Additional genes, radialis and divaricata, are also involved
in producing floral monosymmetry [25]. Studies on the
development of floral monosymmetry have been expanded
to other Antirrhinaceae, such as Linaria [26] and Misopates
[23], and to some Gesneriaceae [25,27,28,29]. Genes
of the cycloidea and dichotoma families also occur in these
closely related groups. In Linaria vulgaris, the mutation is
not based on a change in the nucleotide sequence but on
extensive methylation [26]. The evolution of the cycloidea
gene within a group of genera in Gesneriaceae was studied
in a combined phylogenetic and developmental study
[28], but its role in the formation of monosymmetric and
secondarily polysymmetric flowers in Gesneriaceae is not
yet clear [29]. As in the study on Linaria [26], the
nucleotide sequence of cycloidea was not altered in the polysymmetric forms of these Gesneriaceae. A variant of
monosymmetry that occurs repeatedly in various families of
Lamiales is reduction not only in the upper half of the
flower but also in the lower half. Such flowers may have
only two functional stamens, the upper pair, whereas the
odd stamen and the lower pair are sterile or lacking. This
implies the action of at least one additional gene, which
affects the lower half of the flower. The divaricata gene is
known to affect the lower half of the flower; however, it
does not cause reduction of stamens [22]. In comparative
developmental studies, it appears that floral monosymmetry is expressed at different stages depending on the
systematic group [30], and that even in polysymmetric
flowers there may be transient monosymmetric stages
[15,31]. It would be interesting to know whether this is a
genetic or epigenetic phenomenon.
Although monosymmetric flowers appeared later in
angiosperm evolution than polysymmetric ones, it is clear
that in the Antirrhinaceae, polysymmetric flowers are
evolutionarily derived from monosymmetric flowers. This
can be deduced from two sources. First, phylogenetic
analyses show that Antirrhinum and Linaria are nested in a
large group with monosymmetric flowers [32,33,34].
Second, the morphology of the Antirrhinaceae: polysymmetric Linaria often have five nectar spurs, which
indicates that this is an unnatural form; flowers with nectar spurs in monosymmetric clades always have a single
spur [35]. Within the Lamiales floral symmetry has also
changed by the loss of certain floral organs, for example in
Plantago, in which the odd sepal and the odd stamen have
been lost [15,36,37].

87

Figure 2
Orobanchaceae
Lamiaceae

Highly monosymmetric
lip flowers, odd staminode
mostly lacking

Acanthaceae
Bignoniaceae
Floral monosymmetry
expressed to various
degrees, odd staminode
Scrophulariaceae present or lacking
Verbenaceae

Antirrhinaceae
Gesneriaceae

Floral monosymmetry
relatively weakly expressed,
odd staminode present

Calceolariaceae
(Flowers tetramerous)
Oleaceae
Current Opinion in Plant Biology

Cladogram of Lamiales, simplified after Olmstead et al. [33]

(only the larger families are included), and trend of expression of
floral monosymmetry.

The phylogeny of Lamiales is relatively well resolved in

comparison with those of other angiosperm groups [33]
(Figure 2). The most highly nested families (i.e.
Acanthaceae, Lamiaceae, Orobanchaceae) have the most
extremely monosymmetric flowers with the odd staminode predominantly lacking and the two adaxial petals
close together, sometimes even appearing as a single organ
[15,36]. It may be predicted that the genetic complement responsible for monosymmetry is more complex in
these most highly nested families than in the more basal
family Gesneriaceae. Antirrhinaceae (with Antirrhinum) is
in between these two groups on the phylogenetic grade of
the Lamiales. It may therefore have intermediate complexity in its genetics of floral monosymmetry. The
relatively basal family Gesneriaceae has the most pronounced incidence of polysymmetric flowers in Lamiales.
From their systematic distribution and structure, these
polysymmetric flowers were interpreted as reversals,
which were based on a relatively weakly rooted genetic
constitution for monosymmetry [36]. This evolutionary
direction was later supported by molecular studies [29].
The two basal-most families in Lamiales [33] have
unusual flowers with four sepals and petals, and two stamens and carpels. These flowers are disymmetric in the
Oleaceae and monosymmetric in the small and poorly
studied family Calceolariaceae, which presents a difficulty
to the interpretation of the early evolution of floral form in
the Lamiales [15]. Extensive studies on the rate of evolutionary directions from polysymmetry to monosymmetry
and vice versa show that changes take place frequently in
both directions. The change from polysymmetry to monosymmetry seems to be easier extrinsically (i.e. ecologically)
but more difficult intrinsically (i.e. developmentally) than

88

Growth and development

Figure 3

stamen, such as in the Cannaceae, the Marantaceae, some

Zingiberaceae, a few orchids, some Vochysiaceae, and
Centranthus of Valerianaceae. The most complicated asymmetric flowers are those in which the pollination organs are
enclosed in a keel or the upper lip; the first condition is present in several genera of beans (Phaseolinae of legumes),
the second in Pedicularis (Orobanchaceae) [15].

(a) Enantiostyly

Left

Right

Left

Right

(b) Contort petal

aestivation

Current opinion in Plant Biology

The two forms of leftright asymmetry in flowers addressed in this review.

change in the opposite direction [34,38]. In Gesneriaceae,

an instance of reversion back to monosymmetry from a
clade with secondarily polysymmetric flowers has been
identified [28]; thereby suggesting that evolution back
and forth occurs more often than previously believed [35].

Evolution of floral leftright asymmetry

Leftright asymmetry is prominent in animals, and the
development of leftright asymmetry from a bilateral
(monosymmetric) stage is under discussion. Interest in
leftright asymmetry was initiated because of the medical
importance of disturbed leftright asymmetry in humans;
it then expanded to vertebrates and animals in general. In
the past three years the molecular developmental genetic
literature on leftright asymmetry in animals has exploded
[14]. Leftright asymmetry may occur in two morphs (i.e.
left and right), also called enantiomorphy. In animals,
species and larger taxa are commonly characterised by the
presence of only a single morph. In contrast, in plants,
leftright asymmetry is much less prominent and has
scarcely been tackled, neither from an evolutionary nor
from a developmental genetic point of view. Two kinds of
leftright asymmetry affect flower structure.
Flowers with an asymmetric shape at anthesis

This type of asymmetry has evident repercussions for pollination and has been discussed mainly by floral biologists.
The occurrence of this asymmetry is scattered in several
families in which otherwise monosymmetric flowers are predominant. In one asymmetrical form, that is enantiostyly,
the style and stigma are not in the middle of the flowers but
curved to one side (Figure 3). Enantiostyly is known to
occur in at least 14 genera from 10 or more families [16,39].
In another asymmetrical form, the flowers have only one

Enantiostyly occurs mainly in pollen flowers that are buzzpollinated (i.e. pollinated by vibration) [40]. It has been
hypothesised that an advantage of enantiostyly is the
removal of the style from the median plane, which avoids
damage to the style and stigma from large buzzing bees
[41,42]. In most cases of enantiostyly, both floral morphs are
known to occur on the same individual (monomorphic
enantiostyly). Only in a few species of three monocot families are they separated on different individuals (dimorphic
enantiostyly) [16]. Because the families in which enantiostyly is present are only distantly related, it can be
concluded that enantiostyly has evolved convergently many
times. In the Streptocarpus/Saintpaulia clade (Gesneriaceae),
enantiostyly had one or two origins [27]. Because dimorphic enantiostyly is much rarer than monomorphic
enantiostyly, and because species with dimorphic enantiostyly always have close relatives with monomorphic
enantiostyly, it can be concluded that dimorphic enantiostyly has evolved from monomorphic enantiostyly [16].
This evolutionary direction can also be derived from phylogenetic analyses of the families with dimorphic enantiostyly
[16]. For the same reasons, it can be stated that enantiostylous flowers have evolved from monosymmetric
flowers. The reproductive biological reasons that explain
why dimorphic enantiostyly is so rare have also been
discussed [16]. Studies so far indicate that in the unistaminate asymmetric flowers mentioned, both floral morphs
are present in the same inflorescence [15,43].
In contrast, in both (phylogenetically unrelated) forms of
flowers with enclosed pollination organs, as found in the
Phaseolinae (Fabaceae) and Pedicularis (Orobanchaceae),
only one morph of asymmetry has been identified in preliminary observations [15]. Moreover, a single morph is
found not only within individual plants, but also within
species and at higher levels. In the Phaseolinae, the keel is
curved anticlockwise (if viewed from the front). In
Pedicularis, the flowers (especially the upper lip) are distorted
clockwise. Why this contrast? It may be that flowers with
enclosed pollination organs are ergonomically more difficult
to work by pollinators than those with unenclosed pollination organs. If flowers with enclosed pollination organs can
be consistently worked from the same side, however, their
ergonomic disadvantage may be reduced pollination takes
less time. Therefore, it could be expected that there is
selective pressure for plants to produce only one floral
morph, not only at the level of an individual but at the population level. During speciation, this single morph may have
been retained so that it is now uniform in a genus or larger
group. In contrast, in more open and simpler flowers, the

Evolution of floral symmetry Endress

presence of two morphs in an individual has no effect or

only a minor effect on pollinator visiting time per flower.

89

Figure 4
Taxa with contort corolla always:

Flowers with a contort petal aestivation

Asterids

This kind of asymmetry is most pronounced in floral buds,

whereas in open flowers it may be less prominent. In such
flowers two morphs are possible: sinistrorse (i.e. contorted
to the left) and dextrorse (i.e. contorted to the right)
(Figure 3). It has long been known that some plant families show unfixed (flexible) behaviour in which both
morphs occur in the same species and individual (enantiomorphy); whereas others show fixed behaviour in which
only one morph occurs within a species, genus or even
family [44]. Spiral phyllotaxis, such as that present in flowers of many basal angiosperms and in the calyx of many
eudicots, is also leftright asymmetric. As far as is known,
both morphs may always occur in one individual [15,44].
The distribution of the two petal contortion behaviours is
also significant at higher systematic levels ([15];
Figure 4). First, within the eudicots, the largest group of
the angiosperms, contort flowers are represented mainly in
the rosids and asterids; in other angiosperms contort aestivation is infrequent or lacking. Second, most rosids with
contort aestivation (except some Myrtales and a few
Brassicales and Malpighiales) show the unfixed pattern.
Third, in contrast, all asterids with contort aestivation have
a fixed behaviour. In groups with a fixed behaviour, the
morph can change within the phylogeny of a family. In the
Apocynaceae, it has long been known that some genera are
consistently sinistrose, whereas others are consistently
dextrorse. Phylogenetic studies show that the switch from
one to the other morph has occurred only once or a few
times within the family [15,45,46].
In the unfixed pattern, the expression of the two morphs is
dependent on chance or is governed by the symmetry of
the entire inflorescence. Thus, the expression of the two
morphs seems to be epigenetic. In monochasial partial
inflorescences (e.g. in Oxalis), dextrorse and sinistrorse
morphs regularly alternate in the manner of pendulum
symmetry. In a racemose inflorescence (e.g. in Abutilon), the
distribution (i.e. the sequence of sinistrorse and dextrorse
morphs) is irregular. In contrast, in the fixed pattern, it has
to be assumed that the morph is genetically determined, as
it is in the body plan of animals. The developmental mechanism that determines the morph is, however, unknown in
flowers. It would also be interesting to know why the fixed
pattern is found so consistently in contort flowers of asterids but is uncommon in rosids. Is it a mere coincidence that
the fixed pattern evolved in asterids or did the ancestral
asterids have flowers of a degree of complexity that caused
selective pressure towards the maintenance of a single
morph for pollination biological reasons?

Left

or

at higher
taxonomic
levels!

Right

Taxa with contort corolla mostly:

Rosids

Left

Right

in each
individual!

Basal eudicots
Contort corolla
rarely present
Monocots

Basal
angiosperms

Contort corolla
lacking
Current Opinion in Plant Biology

Cladogram of angiosperms, simplified after Soltis et al. [20] (only the major
groups are included), and evolution of patterns of contort petal aestivation.

being elucidated by molecular systematics and molecular

developmental genetics. Leftright asymmetry in flowers
has been much less studied. A particularly interesting example of leftright asymmetry is contort petal aestivation,
which shows both morphs in each individual plant in most
rosids but only one morph even at higher systematic levels
in asterids. The study of flower evolution and especially of
floral symmetry is a prominent example of a field in which
an integrative approach of different aspects, intrinsic (i.e.
developmental) and extrinsic (i.e. ecological) is necessary
[3,24,38,47]. Phylogenetics, fossil history, evaluation of
evolutionary changes in the diversity of floral forms, molecular developmental genetics, and pollination biology from
the side of the plant and the pollinators, are all required to
develop an integrated view of flower evolution. It is important to know how and when a new kind of symmetry arose
for the first time, but also how easily one kind can be
transformed into another kind, once they have evolved.

Acknowledgements
I thank Michael Donoghue, Rivka Dulberger, Valentin Grob, Richard
Olmstead, Peter Rusert and Kay Schneitz for discussion and information. I am
grateful to Alex Bernhard for assistance with the illustrations.

References and recommended reading

Papers of particular interest, published within the annual period of review,
have been highlighted as:

of special interest
of outstanding interest

Conclusions

1.

Coen ES: Floral symmetry. EMBO J 1996, 15:6777-6788.

The evolution of monosymmetry and secondary polysymmetry in flowers of the Antirrhinaceae and Gesneriaceae is

2.

Jean RV, D Barab (Eds): Symmetry in Plants. Singapore: World

Scientific, 1998.

90

Growth and development

3. Diggle PK, Endress PK (Eds): Development, function and evolution

of symmetry in plants. Int J Plant Sci 1999, 160:S1-S166.
A special issue with articles that elucidate floral symmetry from different
view points: comparative developmental morphology, pollination biology,
and perception of symmetry by bees.
4.

Luo D, Carpenter R, Vincent C, Copsey L, Coen E: Origin of floral

asymmetry in Antirrhinum. Nature 1996, 383:794-799.

5. Hudson A: Development of symmetry in plants. Annu Rev Plant

Physiol Plant Mol Biol 2000, 51:349-370.

A review of symmetry studies in plants, mainly of the vegetative region.
6.

Coen ES: The role of homeotic genes in flower development

and evolution. Annu Rev Plant Physiol Plant Mol Biol 1991,
42:241-280.

7.

Coen ES, Nugent JM: Evolution of flowers and inflorescences.

Development 1994 (suppl):107-116.

8.

Coen ES, Nugent JM, Luo D, Cubas P, Chadwick M, Copsey L,

Carpenter R: Evolution of floral symmetry. Philos Trans R Soc Lond
Biol 1995, 350:35-38.

9.

Dafni A, Kevan PG: Floral symmetry and nectar guides:

ontogenetic constraints from floral development, colour pattern
rules, and functional significance. Bot J Linn Soc 1996,
120:371-377.

10. Neal PR, Dafni A, Giurfa M: Floral symmetry and its role in
plantpollinator systems: terminology, distribution, and
hypotheses. Annu Rev Ecol Syst 1998, 29:345-373.
11. West EL, Laverty TM: Effect of floral symmetry on flower choice
and foraging behaviour of bumble bees. Can J Zool 1998,
76:730-739.
12. Lehrer M: Shape perception in the honeybee: symmetry as a
global framework. Int J Plant Sci 1999, 160:S51-S65.
It is shown that in honeybees some types of response to visual stimuli are
based on innate behavioural programs. Polysymmetric and perpendicular
monosymmetric patterns (i.e. those most common in flowers) are preferred
to other patterns.
13. Giurfa M, Dafni A, Neal PR: Floral symmetry and its role in
plantpollinator systems. Int J Plant Sci 1999, 160:S41-S50.
Floral symmetry is discussed from the perspective of insect pollinator perception. The relationships between innate preferences of shapes and
learning are discussed.
14. Capdevila J, Vogan KJ, Tabin CJ, Izpisa Belmonte JC: Mechanisms
of leftright determination in vertebrates. Cell 2000, 101:9-21.
15. Endress PK: Symmetry in flowers: diversity and evolution. Int J
Plant Sci 1999, 160:S3-S23.
The biological and evolutionary significance of different kinds of floral symmetry is discussed, especially from the point of view of comparative development and diversity of forms. The author discusses new insights into
monosymmetry in Lamiales, and especially into the evolution of leftright
asymmetry in contort flowers throughout the angiosperms.
16. Barrett SCH, Jesson LK, Baker AM: The evolution and function of
stylar polymorphisms in flowering plants. Ann Botany 2000,
85 (suppl A):253-265.
This paper includes a discussion on the evolution of enantiostyly and on the
question of why dimorphic enantiostyly is so rare in angiosperms.
17.

Buschmann H, Fabri CO, Hauptmann M, Schaeffner AR: Tortifolias

lead the way to straight growth. Abstract 487 of the 11th
International Conference on Arabidopsis Research, 2000 June
2428, Madison, USA.

18. Crane PR, Friis EM, Pedersen KR: The origin and early
diversification of angiosperms. Nature 1995, 374:27-33.
19. Crepet WL: Timing in the evolution of derived floral characters.
Upper Cretaceous (Turonian) taxa with tricolpate and tricolpatederived pollen. Rev Palaeobot Palynol 1996, 90:339-359.
20. Soltis PS, Soltis DE, Chase MW: Angiosperm phylogeny inferred
from multiple genes as a tool for comparative biology. Nature
1999, 402:402-404.
21. Luo D, Carpenter R, Copsey L, Vincent C, Clark J, Coen E: Control of
organ asymmetry in flowers of Antirrhinum. Cell 1999,
99:367-376.
The authors show how dichotoma and cycloidea affect the symmetry or
asymmetry of individual petals in Antirrhinum by their expression at different
sites of activity during development.

22. Almeida J, Rocheta M, Galego L: Genetic control of flower shape in

Antirrhinum majus. Development 1997, 124:1387-1392.
23. Vieira CP, Vieira J, Charlesworth D: Evolution of the cycloidea gene
family in Antirrhinum and Misopates. Mol Biol Evol 1999,
16:1474-1483.
Studies of cycloidea in seven Antirrhinum, two Misopates, one Linaria, one
Cymbalaria, and one Digitalis species show that the cycloidea gene family
comprises at least five apparently functional genes.
24. Theissen G: Evolutionary developmental genetics of floral
symmetry: the revealing power of Linnaeus monstrous flower.
Bioessays 2000, 22:209-213.
The plea of a molecular developmental geneticist for integrative studies in
plant evolutionary biology.
25. Cronk Q, Mller M: Genetics of floral symmetry revealed.
Trends Ecol Evol 1997, 12:85-86.
26. Cubas P, Vincent C, Coen E: An epigenetic mutation responsible
for natural variation in floral symmetry. Nature 1999,
401:157-161.
A naturally occurring peloric mutant of Linaria vulgaris is characterised. Its
cycloidea homologue gene does not show changes in the nucelotide
sequence but it is extensively methylated and transcriptionally silent. This
modification is heritable. Occasionally, the mutant reverts to normal by
demethylation of the cycloidea homologue and restoration of gene expression. This indicates that epigenetic mutations may play a more significant
role in plant evolution than previously assumed.
27.

Harrison CJ, Mller M, Cronk QCB: Evolution and development of

floral diversity in Streptocarpus and Saintpaulia. Ann Botany 1999,
84:49-60.
A study of the diversity, development and evolution of floral form involving
various degrees of monosymmetry or polysymmetry in two genera of
Gesneriaceae.

28. Mller M, Clokie M, Cubas P, Cronk QCB: Integrating molecular

phylogenies and developmental genetics: a Gesneriaceae case
study. In Molecular Systematics and Plant Evolution. Edited by
Hollingsworth PM, Bateman RM, Gornall RJ. London: Taylor & Francis;
1999:375-402.
A combination of a phylogenetic study of Gesneriaceae (based on the trnL
intron and the spacer between trnL and trnF, and on ITS1 and ITS2) and of
the cycloidea gene in the same taxa. In three species that have reverted from
monosymmetry to polysymmetry, cycloidea appears to be intact. In one of
these clades, there has been a secondary reversion back to monosymmetry
from secondary polysymmetry, which also implies that the gene is intact.
29. Citerne HL, Mller M, Cronk QCB: Diversity of cycloidea-like genes
in Gesneriaceae in relation to floral symmetry. Ann Botany 2000,
86:167-176.
Phylogenetic analysis of the cycloidea gene in Gesneriaceae indicates that
independent duplication and gene-loss events have occurred during the evolution of Gesneriaceae. Apparently, reversals to polysymmetry were not
caused by loss of function of cycloidea genes.
30. Tucker SC: Evolutionary lability of symmetry in early floral
development. Int J Plant Sci 1999, 160:S25-S39.
An extensive review of the early development of polysymmetric, monosymmetric and asymmetric flowers in angiosperms. The review shows that the
time of symmetry expression during development varies widely but is often
similar in larger groups.
31. Buzgo M, Endress PK: Floral structure and development of
Acoraceae and its systematic relationships with basal
angiosperms. Int J Plant Sci 2000, 161:23-41.
32. Olmstead RG, Reeves PA: Evidence for the polyphyly of the
Scrophulariaceae based on chloroplast rbcL and ndhF sequences.
Ann Missouri Bot Gard 1995, 82:176-193.
33. Olmstead RG, dePamphilis CW, Wolfe AD, Young ND, Elisens WJ,
Reeves PA: Disintegration of the Scrophulariaceae. Am J Botany
2000, 87:in press.
An extensive phylogenetic study of the Scrophulariaceae and related families, which shows that the classical Scrophulariaceae are polyphyletic and
their subgroups need to be considerably rearranged. This work is an important basis for evolutionary studies surrounding the model plant Antirrhinum
and elsewhere.
34. Donoghue MJ, Ree RH, Baum DA: Phylogeny and the evolution of
flower symmetry in the Asteridae. Trends Plant Sci 1998,
3:311-317.
35. Endress PK: Diversity and Evolutionary Biology of Tropical Flowers.
Cambridge, UK: Cambridge University Press; 1994.

Evolution of floral symmetry Endress

91

36. Endress PK: Antirrhinum and Asteridae - evolutionary changes of

floral symmetry. Symp Ser Soc Exp Biol 1998, 53:133-140.

41. Dulberger R: The floral biology of Cassia didymobotrya and

C. auriculata (Caesalpiniaceae). Am J Botany 1981, 68:1350-1360.

37.

42. Fenster CB: Mirror image flowers and their effect on outcrossing
rate in Chamaecrista fasciculata (Leguminosae). Am J Bot 1995,
82:46-50.

Reeves PA, Olmstead RG: Evolution of novel morphological,

ecological, and reproductive traits in a clade containing
Antirrhinum. Am J Botany 1998, 85:1047-1056.

38. Ree RH, Donoghue MJ: Inferring rates of change in flower

symmetry in asterid angiosperms. Syst Biol 1999, 48:633-641.
To date, this is the most detailed discussion on rates of evolutionary changes
between polysymmetric and monosymmetric flowers in a large group of
angiosperms. Aspects of phylogenetics, developmental genetics and ecology
are discussed.
39. Wang G, Miura R, Kusanagi T: The enantiostyly and the pollination
biology of Monochoria korsakowii (Pontederiaceae). Acta Phytotax
Geobot 1995, 46:55-65.
40. Buchmann SL: Buzz pollination in angiosperms. In Handbook of
Experimental Pollination Biology. Edited by Jones CE, Little RJ.
New York: Scientific and Academic Editions; 1983: 73-113.

43. Stoutamire W: Pollination of Tipularia discolor, an orchid with

modified symmetry. Am Orchid Bull 1978, 47:413-415.
44. Schoute JC: On corolla aestivation and phyllotaxis of floral
phyllomes. Verh K Akad Wet Amsterdam, Afd Natuurkunde, Sec 2,
1935, 34(4):1-77.
45. Endress ME, Albert VA: A morphological cladistic study of
Apocynaceae: trends in character evolution within a broadened
familial circumscription. Am J Botany 1995, 82 (suppl):127.
46. Endress ME, Bruyns PV: A revised classification of the
Apocynaceae s.l. Bot Rev 2000, 66:1-56.
47.

Baum DA: The evolution of plant development. Curr Opin Plant

Biol 1998, 1:79-86.