| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Anatomy and Morphology |
Institute of Systematic Botany, University of Zurich, Zollikerstrasse 107, 8008 Zurich, Switzerland
Received for publication 23 May 2007. Accepted for publication 8 November 2007.
ABSTRACT
The buzz-pollinated genus Senna (Leguminosae) is outstanding for including species with monosymmetric flowers and species with diverse asymmetric, enantiomorphic (enantiostylous) flowers. To recognize patterns of homology, we dissected the floral symmetry character complex and explored corolla morphology in 60 Senna species and studied floral development of four enantiomorphic species. The asymmetry morph of a flower is correlated with the direction of spiral calyx aestivation. We recognized five patterns of floral asymmetry, resulting from different combinations of six structural elements: deflection of the carpel, deflection of the median abaxial stamen, deflection or modification in size of one lateral abaxial stamen, and modification in shape and size of one or both lower petals. Prominent corolla asymmetry begins in the earl-stage bud (unequal development of lower petals). Androecium asymmetry begins either in the midstage bud (unequal development of thecae in median abaxial stamen; twisting of androecium) or at anthesis (stamen deflection). Gynoecium asymmetry begins in early bud (primordium off the median plane, ventral slit laterally oriented) or midstage to late bud (carpel deflection). In enantiostylous flowers, pronouncedly concave and robust petals of both monosymmetric and asymmetric corollas likely function to ricochet and direct pollen flow during buzz pollination. Occurrence of particular combinations of structural elements of floral symmetry in the subclades is shown.
Key Words: buzz pollination • enantiomorphy • enantiostyly • floral asymmetry • floral development • functional morphology • homology • petal venation
Asymmetric flowers are rare in angiosperms and are known to occur mostly within large families or orders with predominantly monosymmetric (zygomorphic) flowers (e.g., Leguminosae, Lamiales, Orchidaceae, Zingiberales) and only exceptionally in basal angiosperms (e.g., Winteraceae;Endress, 1999
). Enantiomorphy is a special kind of floral asymmetry in which flowers have two mirror-image morphs. Commonly in enantiomorphic flowers, the style is deflected to the left or to the right of the median plane, a condition known as enantiostyly, which occurs in at least ten angiosperm families of both monocots and dicots (Jesson, 2002) and seems to have evolved from monosymmetry multiple times (Jesson and Barrett, 2003). Left- vs. right-styled flowers may occur on different individuals (i.e., dimorphic enantiostyly) or on the same plant (i.e., monomorphic enantiostyly; see Jesson and Barrett, 2003, for an overview). The development of enantiostylous flowers has been explored in few taxa (Tucker, 1996, 1999; Jesson et al., 2003), whereas the genetics (Jesson and Barrett, 2002a, b) or role of enantiostyly in pollination biology (e.g., Delgado Salinas and Sousa Sánchez, 1977;Dulberger, 1981; Gottsberger and Silberbauer-Gottsberger, 1988) has been the focus of others. The large genus Senna (Cassiinae, Leguminosae; ca. 350 species,Randell and Barlow, 1998) is exceptional for displaying both species with monosymmetric flowers and species with enantiomorphic flowers (monomorphic), in which the gynoecium, androecium, and corolla contribute to the floral asymmetry (Marazzi et al., 2006,2007). Senna thus represents an ideal example to study floral asymmetry.
Floral whorls in Senna affect floral structure and symmetry in various ways (Marazzi et al., 2006,2007). The gynoecium is formed by a single carpel, as is typical for legumes, and is usually long, arcuate, and point-tipped with a chambered (enclosed receptive surface) or a craterlike stigma (receptive surface not enclosed;Owens and Lewis, 1989;Dulberger et al., 1994;Endress, 1994;Tucker, 1996;Marazzi et al., 2007). Unlike most other enantiostylous taxa, in Senna, not only the style, but the entire carpel is deflected to the side. The androecium consists of two five-merous whorls, is highly diverse, and has fascinated researchers for a long time (e.g.,Müller, 1883; Venkatesh, 1957;Lasseigne, 1979;Tucker, 1991;Marazzi et al., 2007). Of the mostly seven fertile stamens, only the three abaxial ones appear to be involved in floral asymmetry. The corolla is yellow and more or less differentiated into three upper (i.e., standard petal and wing petals) and two lower petals (i.e., keel petals in papilionoids). In several enantiostylous species, the upper petals are more or less reduced and the lower ones concave and/or modified in shape and size (Irwin and Barneby, 1982;Marazzi et al., 2006).
Expression of floral asymmetry during development in Senna species with highly asymmetric flowers has not been investigated before. Floral development has been studied in detail only in one species, S. didymobotrya (Tucker, 1996), which has enantiostylous flowers with monosymmetric androecium and corolla. This species was part of a comparative study of Senna, Cassia sensu stricto (s.s), and Chamaecrista (Tucker, 1996), the three genera of subtribe Cassiinae (Irwin and Barneby, 1981,1982). Flowers of Cassia s.s. are monosymmetric, whereas Chamaecrista species have asymmetric, enantiostylous flowers (Irwin and Barneby, 1982;Tucker, 1996). Cassiinae have superficially similar flowers at anthesis due to the same pollination syndrome, but differ in early floral development (e.g., sequence of petal initiation, asymmetric initiation, overlap between whorls, time of carpel initiation, etc.;Tucker, 1996,1997).Dulberger, 1981 observed that in S. didymobotrya deflection of the carpel occurs 6–12 h before anthesis. In other enantiostylous species, style deflection occurs either in the bud or at the beginning of anthesis (Jesson et al., 2003).
Enantiostyly has usually been correlated with buzz pollination by pollen-collecting bees, which vibrate the anthers to extract and collect the pollen for larval provision (e.g.,Buchmann, 1974,1983). Rarely, enantiostylous flowers offer nectar and are not buzz-pollinated (e.g., species of Wachendorfia, Haemodoraceae; see Vogel, 1998). Although enantiostyly was commonly interpreted as promoting cross-pollination, its functional significance has long been debated (e.g.,Todd, 1882;Ornduff and Dulberger, 1978;Dulberger, 1981;Fenster, 1995;Jesson and Barrett, 2002b). In buzz-pollinated flowers, enantiostyly is correlated with other floral features that likely have evolved in relation to the unusual pollination mode, including poricidal anthers (i.e., dehiscence restricted to apical pores) and heteranthery (i.e., different kinds of stamens in a flower), which are also found in Senna (e.g.,Buchmann, 1974; Delgado Salinas and Sousa Sánchez, 1977;Dulberger, 1981;Gottsberger and Silberbauer-Gottsberger, 1988;Owens and Lewis, 1989;Dulberger et al., 1994). The role of these features and of enantiostyly in the pollination biology of Senna has been investigated only in a few species (Buchmann, 1974; Delgado Salinas and Sousa Sanchéz, 1977;Fontanelle, 1979;Dulberger, 1981;Gottsberger and Silberbauer-Gottsberger, 1988;Carvalho and Oliveira, 2003;Laporta, 2003;Westerkamp, 2004). Implications for pollination biology of specialized anther dehiscence patterns are discussed by Marazzi et al. (2007).
In Senna, various kinds of enantiostyly, often with the androecium and corolla also affecting the floral asymmetry, appear to occur and characterize the major clades II–VI recognized by Marazzi et al. (2006) (Fig. 4A). The existence of different kinds of asymmetric flowers and the independent switches to these kinds inferred from the molecular phylogeny suggest that asymmetric flowers may have originated many times and may not be strictly homologous (Marazzi et al., 2006). Floral (a)symmetry in Senna involves several potentially independent structural elements. To understand the evolution of asymmetry, we must first obtain a clear understanding of the various elements. Therefore, in the current study, we investigated the diversity and patterns of floral (a)symmetry in the genus, and in particular, the patterns in corolla morphology and development of floral asymmetry. We addressed the following specific questions: (1) What patterns of floral asymmetry can be identified in Senna (2) What patterns can be recognized in the diversity of petal form? (3) Do the investigated features provide any synapomorphies congruent with the new infrageneric relationships supported by the molecular phylogeny of Senna (Marazzi et al., 2006)? (4) How do species with different patterns of floral asymmetry differ in development? (5) What are the implications of our results with regard to our current understanding of pollination biology?
|
Species studied
We studied 60 Senna species (one or more individuals per species) and two Cassia species, representing the sister genus of Senna (Marazzi et al., 2006). The Senna species studied represent all major clades and subclades of the molecular phylogeny of the genus of Marazzi et al. 2006 based on 81 Senna species and the diversity of morphological patterns observed during field collection (including color photographs of flowering individuals by the first author) and in subsequent morphological investigations (Marazzi et al., 2007). Most samples were collected in the field in Argentina, Australia, Bolivia, Brazil, Mexico, Panama, Paraguay, South Africa, and the United States, and a few were received from European and Australian botanic gardens.
Fifty-seven Senna species were investigated with stereomicroscopy (SM), and three species, S. martiana (Benth.) H. S. Irwin & Barneby, S. subulata (Griseb.) H. S. Irwin & Barneby and S. cf. velutina (Vogel) H. S. Irwin & Barneby, were studied from color photographs of their flowers. In addition 26 species of Senna and the two Cassia species were selected for detailed investigations on petal shape and venation of anthetic flowers. Four Senna species with asymmetric flowers were selected for developmental studies with scanning electron microscopy (SEM): S. aciphylla (clade IVa), S. mucronifera (clade IVb), S. tonduzii (clade VI), and S. wislizeni (clade III). A list of the specimens studied and voucher information are given in the Appendix.
Morphological investigation
Flowers at anthesis and buds of different stages were fixed and stored in 70% ethanol. For investigating petal shape and venation, petals of 1–2 selected anthetic flowers were flattened between two glass slides with the ventral side downward. The slides were then immersed in 70% ethanol and photographed with an Axiocam HRc digital camera (Carl Zeiss AG, Oberkochen, Germany) mounted on a Stemi SV11 stereomicroscope (Carl Zeiss AG). For the developmental study of each selected species, 2–10 buds per developmental stage were studied with SEM (early bud = from organ initiation to beginning of organ differentiation; midstage bud = at organ differentiation; late bud = from end of organ differentiation to growth of differentiated organ). Standard specimen preparation procedures were used for osmium tetroxide-impregnated SEM samples. Several other midstage to late buds were examined with SM. In addition, 2–3 floral buds at midstage of each species were selected for serial sectioning. They were embedded in Kulzer’s Technovit 2-hydroethyl methacrylate (Igersheim, 1993) and sectioned with a Microm HM 335 rotary microtome (Microm International GmbH, Walldorf, Germany) and conventional microtome knife (grade D); transverse section (TS) series were cut at 7 µm, stained with ruthenium red and toluidine blue (Weber and Igersheim, 1994), and mounted in Histomount (National Diagnostics, Atlanta, Georgia, USA) on glass slides. Fixed floral material and slides are deposited at the Institute of Systematic Botany of the University of Zurich (Z), Switzerland.
RESULTS
Diversity of floral symmetry and petals
In all studied Senna species, the sepals are arranged either in a clockwise or counterclockwise spiral, which means that monosymmetric flowers of Senna also have an enantiomorphic calyx. In the two Cassia species studied, the corolla is monosymmetric; the petals all have nearly the same size and form (but the lower petals are slightly concave) and a single main vein, which extends up to the petal tip (Fig. 1A). In the following paragraphs we describe floral symmetry and petal diversity in the major clades of Senna (Marazzi et al., 2006; see also Table 1 and Fig. 4A–D).
|
|
|
Clade III
Flowers are asymmetric, with the gynoecium deflected to the side in all species investigated. The androecium is nearly monosymmetric (species of subclade IIIb,S. spectabilis) or the median and one lateral abaxial stamen are deflected to the opposite side of the gynoecium (species of subclade IIIa). The corolla is asymmetric whereby the upper petals are not reduced (S. spectabilis, S. unijuga, Fig. 1F;S. wislizeni, Fig. 1G), or moderately reduced (S. atomaria, Fig. 2F;S. mollissima, Fig. 1H), and one lower petal (S. atomaria, S. mollissima, Fig. 1H; S. spectabilis, not shown) or both lower petals (S. unijuga, Figs. 1F, 2G; S. wislizeni, Figs. 1G, 2H) are modified in shape and size, i.e., the blade is highly asymmetric, concave and foot-shaped, and in addition, these two petals differ from each other (Figs. 1F–H). The standard petal is stalkless in S. mollissima (Fig. 1H). Upper petals have three main veins (S. unijuga, Fig. 1F; in S. wislizeni, Fig. 1G, the median vein is more conspicuous than the lateral ones), and the lower petals have apparently only two main veins (S. unijuga, Fig. 1F;S. wislizeni, Fig. 1G), or all petals have a single main vein, except the standard petal, in which two main veins form a double strand (S. mollissima, Figs. 1H, H'). In strongly modified petals, venation is particularly robust, and the basal part of the main veins is united (Figs. 1F, G), or many robust secondary veins extend from the basal part of the single main vein (Fig. 1H).
Clade IV
Flowers are asymmetric with the gynoecium deflected to the side in all species investigated. This clade includes S. skinneri plus a clade of two subclades IVa and IVb (Marazzi et al., 2006). The androecium of subclade IVa is asymmetric: all stamens are fertile and are arranged slightly irregularly (S. aciphylla, Fig. 2I;S. artemisioides, Fig. 2J), and one lateral abaxial stamen may be larger than the others (S. artemisioides, Fig. 2J; S. odorata, not shown). Senna skinneri and species of subclade IVb have seven fertile stamens, except for S. hayesiana with only the four middle stamens fertile (see Fig. 4E). The androecium is asymmetric, with only the median abaxial stamens deflected to the opposite side of the deflected gynoecium (S. skinneri, Fig. 2K; species of subclade IVb,Figs. 2L–N) or it is nearly monosymmetric (S. dariensis var. hypoglauca, S. hayesiana, Fig. 2O;S. quinquangulata, Fig. 2P). The corolla is asymmetric in most species (subclade IVa,Figs. 2I, J; species of subclade IVb and S. skinneri, Figs. 2K–N), or, rarely, it is nearly monosymmetric (S. dariensis var. hypoglauca, S. hayesiana, Fig. 2O;S. quinquangulata, Fig. 2P; S. rizzinii). In asymmetric corollas, the upper petals are not reduced, and one or both lower petals are concave, but not modified in shape and size (species of subclade IVa,Figs. 1I, 2I–J; species of subclade IVb,Fig. 2K, N), or one or both are concave and foot-shaped (species of subclade IVb,Figs. 1J, 2L, M). The standard petal may be emarginate or bilobed (species of subclade IVb*,Fig. 1J). Petals have three main veins (Figs. 1I, J). In strongly modified petals, venation is particularly robust, and the basal part of the main veins is united (Fig. 1J).
Clade V
Flowers are asymmetric, with the gynoecium deflected to the side in all species investigated. The androecium is nearly monosymmetric in species with the median abaxial stamen smaller than the lateral abaxial ones (S. cf. velutina, Fig. 3A), or it is asymmetric in species with three similar abaxial stamens, and the median abaxial stamen is deflected to the opposite side of the gynoecium (S. cana var. calva, S. uniflora, Fig. 3B). The corolla is nearly monosymmetric (S. cf. velutina, Fig. 3A; S. uniflora, Figs. 1K, 3B), or slightly asymmetric, with one lower petal concave and slightly modified in shape and size (S. cana var. calva, Fig. 1L). The standard petal may be emarginate (Fig. 1L). Petals have three main veins (Fig. 1L), or in S. uniflora, they apparently have only a single main vein (Fig. 1K). In the lower petal that is modified, venation is slightly more robust than in the other petals, but the basal part of the main veins appears to be free (Fig. 1L).
|
Clade VII
Flowers are monosymmetric in both subclades VIIa and VIIb. In some flowers of subclade VIIa, the gynoecium may be slightly deflected to the side, but because the deflection is only slight and inconsistent in these species, their flowers are not considered enantiostylous. Upper and lower petals have similar shapes (S. apiculata, S. armata, S. birostris, S. villosa, Figs. 1Q, 3P), or the lower petals are slightly longer and thinner than the upper ones and are concave (S. hirsuta, Figs. 1R, 3M; S. septemtrionalis, Fig. 3N; S. subulata, Fig. 3O). In S. villosa, the petals are short-stalked (Fig. 1Q). The standard petal is emarginate (Figs. 1Q, R, 3M–P; not in S. apiculata and S. armata, both of subclade VIIb). The blade of the upper lateral petals is monosymmetric (e.g.,S. hirsuta, Fig. 1R) or slightly asymmetric, forming two mirror-image petals (e.g.,S. pendula). Petals have three main veins (Fig. 1Q, R).
Floral development
We focus on the developmental stages at which the floral asymmetry becomes apparent especially in the corolla and androecium, and additionally consider calyx and corolla aestivation (i.e., overlapping of flanks of perianth organs in bud). We studied young floral stages with the SEM and midstage floral buds with transverse microtome sections of representatives of four clades (III, IVa, IVb, and VI) that are characterized by asymmetric flowers: S. wislizeni, clade III (Figs. 4F, 5); S. aciphylla, subclade IVa (Figs. 4G, 6); S. mucronifera, subclade IVb (Fig. 7; TS not illustrated but similar to Fig. 4F); and S. tonduzii, clade VI (Figs. 4H, 8).
|
|
|
|
Senna wislizeni (clade III)
In anthetic flowers of S. wislizeni, the carpel is deflected to the side, and the median and one lateral abaxial stamen are deflected to the opposite side; the corolla is asymmetric, with the upper petals not reduced and both lower petals concave and foot-shaped. Blade modification is more conspicuous in the lower petal opposite the deflected carpel (Figs. 1G, 2H).
Lower and upper petals develop unequally; they thus differ in size early in development, and the corolla is asymmetric (Fig. 5B). Organs of the outer androecial whorl are initiated when petals start to develop (Fig. 5C), while organs of the inner androecial whorl are initiated after the organs of the outer whorl start to develop (compare Fig. 5E, F). The carpel is initiated as a bulge in the center of the bud during initiation of the organs of the outer androecial whorl (Fig. 5C, E).
Shortly after petal initiation, one lower petal begins to differentiate into a footlike shape (Fig. 5D). Corolla aestivation is quincuncial (see Fig. 4C), with the lateral upper petals covering the standard petal and also the lower petals (Fig. 5D). The androecium is still monosymmetric after all organs of the outer androecial whorl have been initiated (Fig. 5E). With subsequent anther development, the androecium becomes asymmetric because the stamens and staminodes of the outer androecial whorl appear to be arranged in a twisted pattern, i.e., with the anther tips of the middle and median abaxial stamens touching (and slightly overlapping) the side of one neighboring anther tip (Fig. 5F). This twisted pattern disappears with development of the abaxial stamens of the inner whorl (Fig. 5G). Differentiation of the thecae in the median abaxial stamen is unequal, one theca becoming larger than the other one (Fig. 5G–I). In the midstage bud, stamens of the inner and outer whorls are of different size (Fig. 5H, K). In the late bud, all middle stamens are of similar size, while the median abaxial stamen is the largest of the abaxial stamens (Fig. 5I, L). During anther differentiation, the carpel becomes arcuate and is still in the plane of floral monosymmetry up to midstage bud (Fig. 5G, H, J, K), while it appears slightly deflected in late bud (Fig. 5I, L).
Senna aciphylla (clade IVa)
In anthetic flowers of S. aciphylla, the carpel is deflected to the side, all stamens are fertile and deflected in an irregular manner, and the corolla is asymmetric, with the upper petals not reduced and one lower petal concave, but not modified in shape and size (Fig. 1I, 2I).
Sepals are initiated in spiral sequence, which is reflected in a quincuncial calyx aestivation (Fig. 6A, B). The two lower petals and one lateral upper petal are initiated before the two remaining upper petals (Fig. 6C). Initiation of these upper petals nearly overlaps with the initiation of the stamens of the outer androecial whorl, except that one of the two adaxial stamens is initiated later in front of the fifth sepal, between the two last initiated upper petals (Fig. 6C). Abaxial stamens of the inner androecial whorl are initiated before the middle and adaxial stamens (Fig. 6D). The carpel is initiated after the stamens of the outer androecial whorl and abaxial stamens of the inner whorl but before the middle and adaxial stamens of the inner whorl (Fig. 6D).
Shortly after petal initiation, the lower petals are already larger than the upper petals, and, although the upper petals are of different size, the corolla appears nearly monosymmetric (Fig. 6D). Subsequently, the upper petals reach a similar size, and corolla aestivation becomes quincuncial (nearly quincuncial in Fig. 6E, but see TS in Fig. 4G and aestivation diagram in Fig. 4E). During anther development, one middle and the two adaxial stamens of the outer androecial whorl appear to be arranged in a twisted pattern (Fig. 6F, G, J). This twisted pattern disappears with development of the abaxial stamens of the inner whorl in midstage bud (Fig. 6H). At this stage, anther differentiation (and anther tip elongation) first begins in the stamens of the outer whorl and in the abaxial stamens of the inner whorl (Fig. 6H, I, K). Stamens of the outer whorl are larger than those of the inner whorl also in late bud (Fig. 6I, L). Although differentiation of the thecae in the median abaxial stamen is slightly unequal, and one theca thus becomes larger than the other one, the androecium still appears to be nearly monosymmetric, but in late bud becomes irregularly asymmetric (Fig. 6H, I), likely because of space constraints and small differences in the sizes of the stamens (see Fig. 6K, L). The gynoecium appears asymmetric in early bud because the ventral slit is oriented slightly laterally and appears either to the left in buds with clockwise calyx aestivation or to the right in buds with counterclockwise calyx aestivation (Fig. 6E, F). During anther differentiation, the carpel becomes arcuate and appears slightly deflected in midstage bud (Fig. 6H, I).
Senna mucronifera (clade IVb)
In anthetic flowers of S. mucronifera, the carpel is deflected to the side; the androecium is asymmetric, with the median abaxial stamen deflected to the opposite side; and the corolla is asymmetric, with the upper petals not reduced and both lower petals concave (one is only slightly asymmetric and the other is foot-shaped) (Fig. 1J, 2M).
The median abaxial sepal is first initiated (Fig. 7A). The lower petals are initiated slightly before the upper petals (Fig. 7B) and grow to different sizes (Fig. 7C); the corolla is thus early asymmetric. The organs of the outer androecial whorl are initiated when the petals begin to develop (Fig. 7C), whereas the organs of the inner androecial whorl are initiated after the organs of the outer whorl begin to develop (compare Fig. 7C, D). Carpels are initiated after initiation of the organs of the outer androecial whorl (not illustrated, but compare Fig. 7C, D).
Shortly after petal initiation, one lower petal becomes foot-shaped (Fig. 7C, D). Aestivation of the corolla is cochlear ascending (Fig. 7E). The androecium is nearly monosymmetric after the organs of both outer and inner androecial whorls have been initiated (Fig. 7F). With subsequent anther development, the androecium becomes asymmetric because differentiation of the thecae in the median abaxial stamen is unequal; one theca becomes larger than the other (Fig. 7F–I). Stamens are all of nearly the same size in early bud, only the median abaxial stamen is much larger (Fig. 7F). In midstage bud, the three abaxial stamens become much larger than the four middle stamens and the three adaxial staminodes (Fig. 7G, H, J, K). However, at this stage, the stamens of the inner whorl are still slightly smaller than those of the outer whorl and become similar in size in late bud (Fig. 7I, L). In late bud, the anther tips of the abaxial stamens begin to elongate. During anther differentiation in midstage bud, the carpel becomes arcuate and appears only slightly deflected (see adaxial view of style,Fig. 7J, K), but it becomes more deflected in late bud (Fig. 7I, L).
Senna tonduzii (clade VI)
In anthetic flowers of S. tonduzii, the carpel is deflected to the side; the androecium is asymmetric, with all the abaxial stamens deflected to the opposite side of the carpel; and the corolla is asymmetric, with the upper petals highly reduced and one lower petal concave and foot-shaped, the other petal with a strongly asymmetric blade (Figs. 1N, 3F). Stamen union occurs in S. tonduzii but in no other Senna species studied. The filaments of the three adaxial staminodes and four middle stamens are united (see Marazzi et al., 2007).
The median abaxial sepal is initiated first (Fig. 8A). The two lower and one of the upper petals appear to be initiated before the two upper petals adjacent to the fifth sepal (Fig. 8B). The lower petals are of slightly different size; the corolla is thus early asymmetric (Fig. 8B). The organs of the outer androecial whorl are initiated before the last two upper petal initials begin to develop (Fig. 8C), whereas the organs of the inner androecial whorl are initiated when organs of the outer whorl begin to develop (compare Figs. 8B, E). The carpel is initiated after initiation of the organs of the inner androecial whorl, and, interestingly, it appears to be displaced from the median plane of floral symmetry (Fig. 8E).
Shortly after petal initiation, one of the lower petals, which are conspicuously larger than the upper ones, begins to become foot-shaped, and the other lower petal also becomes asymmetric (Fig. 8C, D). Corolla aestivation seems not to be cochlear ascending, but the upper petals appear in a partial contort pattern (Fig. 8H). The androecium is nearly monosymmetric after all its organs have been initiated (Fig. 8E, F). Later, the androecium becomes asymmetric because unequal differentiation of the thecae in the median abaxial stamen results in one become larger than the other (Fig. 8G). The stamens are all approximately similar in size in early bud, except for the adaxial staminodes, which remain much smaller (Fig. 8G). In midstage bud, the three abaxial stamens become larger than the middle stamens (Fig. 8I). However, at this stage, the stamens of the inner whorl, especially the middle ones, are still smaller than those of the other whorl, but they become almost equal in size in late bud (Fig. 8K). During anther differentiation, the androecium becomes twisted and is thus asymmetric (compare Fig. 8G, I, J). Elongation of the anther tips begins in late bud (Fig. 8J–L). Filament union begins in late midstage bud (Fig. 8K, L). During anther differentiation in midstage bud, the carpel becomes arcuate and deflected (Fig. 8I). The deflection becomes stronger in late bud (Fig. 8J, K).
DISCUSSION
Patterns of floral (a)symmetry
About half of the approximately 350 species of Senna have asymmetric, enantiostylous flowers, with both left and right morphs on the same plant (Irwin and Barneby, 1982). In our study, the morph of a flower is correlated with the direction of spiral calyx aestivation: a clockwise spiral corresponds to a right morph and a counterclockwise spiral to a left morph (Fig. 4D). Interestingly, in S. aciphylla the spiral calyx aestivation appears to be correlated with the lateral orientation of the ventral slit in early gynoecium development, which means that the floral morph is also correlated with the ventral slit orientation: slit oriented to the left = right morph; slit oriented to the right = left morph (see Fig. 6E, F). The morph can thus be predicted by observing the calyx aestivation in the floral bud. Whether such a correlation occurs in other monomorphic enantiostylous genera with pentamerous flowers has, to our knowledge, not been investigated. Similarly, the two androecium asymmetry morphs in Convolvulus species (Convolvulaceae) are related to the direction of calyx aestivation (Endress, 1999). An enantiomorphic calyx is, however, present in all Senna flowers, including the monosymmetric ones. Species with asymmetric flowers occur in clades II–VI, while those with monosymmetric flowers characterize clades I and VII.
The first sepal is always in median abaxial position, although a subtending bract is well developed, while bracteoles appear to be absent (Tucker, 1996), except in S. paradictyon (Irwin and Barneby, 1982). In fact, bracteoles may be initiated and then suppressed, as was found in several papilionoids (Prenner, 2004a) and in our preliminary SEM studies of Senna species (unpublished data); bracteoles can also be completely absent, as in many other legumes (Sokoloff et al., 2007).
We recognized six major patterns of floral (a)symmetry, which we describe from the simplest to the most complex (Table 1;Fig. 4A, patterns 1–6). Pattern 1 corresponds to floral monosymmetry (although a slight, inconsistent gynoecium deflection can be observed in some flowers). The petals are usually flat (Senna clades I and VII,Cassia), or rarely, the lower petals may be concave (species of subclade VIIa). Pattern 2, the simplest pattern of floral asymmetry, involves the lateral deflection of the gynoecium only, while the androecium and corolla are monosymmetric (clades II, few species of subclade IVb). In pattern 3, floral asymmetry additionally involves the corolla, whereas the androecium is nearly monosymmetric (subclade IIIa) or slightly irregular (species of clade IVa). This androecial irregularity does not pertain to lateral deflection of stamens nor to a conspicuous size modification of one of the lateral abaxial stamens (discussed later), but rather to small differences in size among all stamens and space constraints during development. Corolla asymmetry is due to concavity of the lower petal opposite the deflected carpel (e.g.,S. aciphylla, clade IVa;Fig. 2I). In pattern 4, floral asymmetry additionally involves the deflection of the median abaxial stamen (S. skinneri, species of clades IVb, V, VI), and in pattern 5 the modification in size of one lateral abaxial stamen either opposite the deflected carpel (species of subclades IVa and VIa) or on the same side (e.g.,S. multijuga var. multijuga, clade VI;Fig. 3H). The corolla has an array of modifications especially of the lower petals, from one lower petal concave to both lower petals highly concave and modified in size and shape, i.e., foot-shaped. In addition, in pattern 5 the upper petals may be reduced (e.g.,S. acuruensis var. acuruensis, clade VI;Figs. 1M, 3G). In pattern 6, the most complex pattern of asymmetry, the deflection of the carpel and both the median and one lateral abaxial stamens and the modification of the size and shape of the lower petals contribute to the floral asymmetry (species of clades III, VI). The asymmetric corollas are diverse: the upper petals are not reduced and one lower petal (e.g., S. aversiflora of clade VI;Fig. 3D) or both lower petals (e.g., S. wislizeni of subclade IIIa;Figs. 1G, 2H) are foot-shaped, or, alternatively, the upper petals are reduced, one lower petal is foot-shaped and the other lower petal is either concave and almost monosymmetric (e.g.,S. pallida, clade VI;Figs. 1O, 3I), or flat and asymmetric (e.g., S. tonduzii, clade VI;Figs. 1N, 3F). Also, in flowers in which both lower petals are foot-shaped, these two differ from each other in size.
Finally, dissection of floral asymmetry allowed us to recognize at least six structural elements, involving five organs from three different floral whorls (Table 1) that in diverse combinations form the five patterns previously described: (1) deflection of the carpel (patterns 2–6); (3) deflection of the median abaxial stamen (patterns 4, 6); (3) deflection (pattern 6) or, rarely, (4) modification in size (pattern 5) of one lateral abaxial stamen; and modification in shape and size of (5) one or (6) both lower petals (patterns 3–6).
Patterns of petal diversity
Diversity in the corolla is particularly interesting in Senna because petals have undergone diverse morphological modifications (Table 1;Fig. 4B). They include: (1) reduction of upper petals (clades IIIb, VII, VI, excluding VIa), (2) emarginate or bilobed shape of standard petal (clades V, VII and species of IV), (3) concavity of the standard petal (clade II), (4) concavity of lower petals (species of clades III-VII), and (5) enlargement and modification into asymmetric shape of lower petals (clades III, VI, and species of clade IVb). Our results do not support Tucker’s 1997, p. 160) observation that the enlarged petal in Senna is a lateral upper ("wing") petal. The latter two petal modifications contribute to most of the diversity observed in corolla asymmetry. In the first kind, the upper petals may be unequally reduced, i.e., the lateral upper petals are of slightly different size, but this unequal reduction only weakly affects floral asymmetry. Moreover, they are often partially hidden by the enlarged lower petals. Strong petal modification causing asymmetric corollas is found also in the lower petals of Chamaecrista (Tucker, 1996,1997). However, in some species of this genus, one upper lateral petal may be strongly modified to enclose the stamens (Okpon, 1969).
Petal venation in Senna is unusually diverse (Fig. 1). In most species, petals have three veins, but in a few species of the basal lineages of Senna and in Cassia javanica, petals have a single main vein. Also, petals of the small-flowered S. uniflora (clade V,Fig. 1K) and the highly reduced upper petals in a few species of subclade VI (Fig. 1M) have a single main vein, probably because of the reduction in petal size. Rarely, there are two main veins in highly modified, asymmetric lower petals (Fig. 1F, G) or in the standard petal (Fig. 1H, H'). The veins are particularly strong in highly concave and asymmetric lower petals (Figs. 1F–H, J, M–P), and rarely also in the standard petal (Fig. 1D, E).
Implications for systematics of Senna
Patterns of floral (a)symmetry and petal morphology are constant within a few clades suggested by molecular analyses (Marazzi et al., 2006). Pattern 1 (i.e., floral monosymmetry) is constant in clades I and VII, pattern 2 in clade II, pattern 4 in clade V, pattern 5 in subclade VIa (equivalent to series Aphyllae), and emarginate standard petal in subclades IVb* (equivalent to series Trigonelloideae) and VIIa (see Table 1 and Fig. 4A). In addition, petal venation is constant in clade I (one main vein) and in clades IV–VII (three main veins). Compared to floral features not affecting floral (a)symmetry (Marazzi et al., 2007), the structural elements involved in floral (a)symmetry studied here are more evolutionarily flexible and provide less support for the major clades and subclades of Senna (Marazzi et al., 2006) and the current series of the genus (Irwin and Barneby, 1982).
Floral development and expression of floral asymmetry
The high diversity in patterns of floral asymmetry observed in Senna, in addition to monosymmetry, likely reflects diverse patterns of floral development. Floral development was previously known in detail from only one species,S. didymobotrya (Tucker, 1996), a species of clade II with deflected carpel but monosymmetric androecium and corolla. We investigated the floral development of another species with moderately asymmetric flowers (S. aciphylla, Fig. 6, clade IVa) and, for the first time, species with strongly asymmetric flowers (S. wislizeni, clade III,Figs. 2H, 5; S. mucronifera, clade IVb,Figs. 2M, 7; and S. tonduzii, clade VI,Figs. 3F, 8).
Organogenesis
The first sepal initiated is always median abaxial, and the others follow in spiral sequence (this study;Tucker, 1996). Petals are also initiated in spiral sequence but almost simultaneously. Petals are initially equal (S. didymobotrya, Tucker, 1996), or unequal with the lower petals larger than the upper (S. mucronifera, S. wislizeni, and S. tonduzii, this study), or the upper petals are initially unequal but later become of similar size (S. aciphylla, this study). The organs of the outer androecial whorl are initiated after initiation of all petals (S. wislizeni, this study; S. didymobotrya, Tucker, 1996), or their initiation overlaps with petal initiation (S. aciphylla, this study). In both androecial whorls, the organs are initiated in unidirectional order, from abaxial to adaxial. The carpel is initiated before any stamens (S. didymobotrya, Tucker, 1996), or after the outer androecial organs (S. aciphylla, S. mucronifera, and S. wislizeni, this study), or after inner abaxial androecial organs (S. tonduzii, this study).Tucker, 1996 cursorily mentions an organogenesis similar to that of S. didymobotrya in S. artemisioides (clade IVa),S. bicapsularis (clade VIIa),S. obtusifolia (clade IVb),S. occidentalis (clade VIIa), and S. surattensis (clade uncertain).
In contrast to Senna and most caesalpinioids, the first initiated sepal is lateral abaxial in Cassia javanica (Tucker, 1996) and some other caesalpinioids (Tucker, 1992,1998,2001,2002; Kantz, 1996). Spiral sequence of sepal and petal initiation combined with unidirectional stamen initiation, as in Senna, corresponds to one of the common patterns found in caesalpinioids, whereas another common pattern is spiral sequence of sepal initiation combined with unidirectional petal and stamen initiation, which occurs in Cassia and Chamaecrista, for example (Tucker, 1996). A similar range of variation in timing of carpel initiation as observed in the Senna species studied here appears to be present in Cassia and Chamaecrista (Tucker, 1996).
Organ development
The sepals enlarge in the sequence of their spiral initiation (this study;Tucker, 1996). The petals reach an equal size early (S. aciphylla, this study; S. didymobotrya, Tucker, 1996), or they remain unequal until the onset of petal overlapping (S. auriculata, S. bicapsularis, S.x floribunda, and S. obtusifolia, Tucker, 1996) or up to anthesis (S. mucronifera, S. tonduzii, and S. wislizeni, this study). Two patterns of corolla aestivation occur (Fig. 4C): (1) cochlear ascending, as typical of most caesalpinioids (S. auriculata, S. bicapsularis, S. corymbosa, S. lindheimeriana, S. multijuga, S. pallida, S. pendula, and S. surattensis, Tucker, 1996), and (2) quincuncial (S. aciphylla, S. mucronifera, S. wislizeni, and S. tonduzii, this study; and S. alata, S. artemisioides, S. didymobotrya, S. polyphylla, S. quinquangulata, and S. racemosa, Tucker, 1996). The two patterns are probably dependent on the speed of petal growth: early enlarging petals maintain the pattern of their spiral initiation sequence resulting in quincuncial aestivation, whereas late enlarging petals are influenced by the developing floral monosymmetry, which results in a cochlear ascending aestivation. In early development, the median abaxial stamen enlarges faster than the other organs of the outer androecial whorl (S. mucronifera, S. wislizeni, and S. tonduzii, this study; S. didymobotrya, Tucker, 1996) or almost as fast as the other organs of the same whorl (S. aciphylla, this study). In all species studied, the middle and adaxial organs of the inner androecial whorl remain smaller than their counterparts of the outer whorl up to late-stage bud. In midstage bud, anthers begin to differentiate, and anther tips of abaxial stamens begin to elongate. The three abaxial stamens become similar in size and larger than the middle ones (S. mucronifera and S. tonduzii, this study), or the median abaxial one remains smaller than the lateral ones but is larger than the middle ones (S. didymobotrya, Tucker, 1996). Size difference between the abaxial and middle stamens may appear only in late bud (S. wislizeni, this study). The carpel becomes arcuate and covered by hairs during midstage bud, and its stigmatic chamber is formed. The hairs fringing the stigmatic orifice develop in late bud.
Expression of floral asymmetry
The time at which floral asymmetry becomes evident during development differs according to the pattern of floral asymmetry at anthesis (patterns 1–6,Table 1,Fig. 4A). In enantiostylous Senna species, floral asymmetry involves either carpel deflection only or also stamen deflection or modification in size, and petal modification in size and shape (Table 1). Interestingly, flowers of all studied species with quincuncial corolla aestivation are enantiostylous. In contrast, those of most species with cochlear ascending corolla aestivation are monosymmetric. In general, ascending or descending cochlear patterns prevail in monosymmetric flowers of large clades of core eudicots, such as Leguminosae and Lamiales (Endress, 1994). However, both patterns, cochlear and quincuncial, appear to occur also in Chamaecrista (Okpon, 1969;Tucker, 1996). More species should be studied in Senna, including also species of the nonrepresented clades I and V, to test whether a quincuncial corolla aestivation is restricted to asymmetric, enantiostylous flowers.
Floral asymmetry appears at different developmental stages in the different floral whorls of Senna. Prominent corolla asymmetry is expressed in early bud; the lower petals become modified in shape and size when they begin to develop (Figs. 5C, D, 7C, D, 8C, D). Asymmetry in the androecium is expressed in early midstage bud, when the size of the thecae in the median abaxial stamen becomes unequal (Figs. 5F, 7F, 8G). In addition, the androecium in S. tonduzii becomes conspicuously twisted during anther differentiation (Figs. 8G, I, J). A twisted pattern is also observed during development of the outer androecial whorl in S. aciphylla (Fig. 6F, G, J) and S. wislizeni (Fig. 5F), but it later disappears. Deflection of abaxial stamens to the side takes place only at anthesis. Asymmetry in the gynoecium of S. tonduzii appears to be expressed early in development; the carpel primordium appears to be displaced from the median plane of the flower (Fig. 8E). In S. aciphylla, the ventral slit of the carpel is oriented slightly obliquely in early bud (Fig. 6E, F), as found in some other caesalpinioids with monosymmetric flowers (e.g.,Bauhinia malabarica, Tucker, 1988; Ceratonia siliqua, Tucker, 1992;Cassia javanica, Tucker, 1996). It would be interesting to know whether the early asymmetries in the gynoecium of Senna species are related to the deflection of the carpel (enantiostyly), which appears to occur in midstage bud (S. aciphylla, Fig. 6H, I; S. mucronifera, Fig. 7I, K, L; S. tonduzii, Fig. 8H–L) or late bud (S. didymobotrya, Tucker, 1997; S. wislizeni, Fig. 5I, L). In other enantiostylous families (mostly monocots), enantiostyly is expressed late in development: the style becomes deflected in late bud (Wachendorfia paniculata, Dilatris corymbosa, and Philydrum lanuginosum) or only at anthesis (Cyanella lutea, Monochoria australasica, species of Heteranthera, and Solanum rostratum) (Jesson et al., 2003). This is also the case in enantiostylous genera of Gesneriaceae (Saintpaulia and Streptocarpus, Harrison et al., 1999; Q. C. B. Cronk, University of British Columbia, Vancouver, Canada, personal communication).
In caesalpinioids, floral symmetry varies from nearly polysymmetric to moderately or pronounced monosymmetric (Tucker, 2003) and asymmetric (Senna, Chamaecrista, and Labichea;Tucker, 1996,1997,1998;Marazzi et al., 2006). Whereas in Senna most of the floral asymmetry is expressed in the midstage bud or later, in Chamaecrista fasciculata the entire organogenesis is asymmetric: the floral asymmetry in the androecium and corolla is expressed at early stages by precocious organ initiation on one side (left or right;Tucker, 1996,1999). A similar asymmetrical initiation was found in Schotia afra (Tucker, 2001) but in contrast to Chamaecrista, asymmetry does not persist in the flower at anthesis; it disappears during organ development. Also in Labichea lanceolata, which is distantly related to Senna (Bruneau et al., 2001) and characterized by reduced number of floral organs and dissimilar stamens, floral asymmetry is expressed very early in development, including an asymmetric floral apex and an asymmetric order of organ initiation (Tucker, 1998). The androecium of several monosymmetric papilionoids is also asymmetric in early developmental stages because the adaxial antesepalous stamen is formed to the left or right of the median plane (Prenner, 2004b). The asymmetrically curved or coiled keel, enclosing the asymmetric androecium and gynoecium of some papilionoids (Phaseolinae, Vicieae) is usually the result of late ontogeny (Tucker, 1999), as shown for Lathyrus latifolius (Prenner, 2003) and Vigna caracalla (Troll, 1951;Prenner, 2003).
Implications for pollination biology
Enantiostyly is restricted to buzz-pollinated flowers, in which other features (such as poricidal anther dehiscence, heteranthery, and point-tipped stigmas) evolved in relation to the unusual pollination biology (e.g.,Buchmann, 1983). In Senna, pollen-collecting bees extract pollen by vibrating the middle "feeding" stamens, which they clasp with their legs (e.g.,Buchmann, 1974; Delgado Salinas and Sousa Sánchez, 1977;Dulberger, 1981;Gottsberger and Silberbauer-Gottsberger, 1988;Westerkamp, 2004). The abaxial "pollinating" stamens are usually longer and display a higher diversity in anther dehiscence specialization than feeding stamens (Marazzi et al., 2007). Most species of Senna have a long and arcuate carpel with an extremely small, chambered or crater-like stigma (Marazzi et al., 2007).Dulberger et al. 1994 suggested that the diversity in inflection of the style tip and stigma position and orientation evolved in relation to the specific size and positioning of the bees visiting the flowers. However, different lengths and inflections of the entire carpel may also be associated with different sizes of bees. The morphology of the carpel (i.e., straight or arcuate, median or deflected) may also be related to which body part of the bee approaches or touches the stigma, thus suggesting a functional significance for enantiostyly (e.g.,Jesson et al., 2003;Jesson and Barrett, 2005).
Enantiostyly has commonly been regarded as a device to promote outcrossing (Todd, 1882) because pollen of a left floral morph is deposited on a place on the body of bees that corresponds to the position of the stigma in a right floral morph, and vice versa. In monomorphic enantiostylous taxa, such as Senna, the presence of both left and right floral morphs on the same plant and the observation that many of these plants are self-compatible indicate that geitonogamous self-pollination between different morphs is possible. However, compared to monosymmetry, enantiostyly appears to function to reduce geitonogamous pollen transfer (Jesson and Barrett, 2003,2005). Specifically,Jesson and Barrett 2003 suggested that enantiostyly associated with heteranthery and the deflection of a pollinating anther to the opposite side of the style may function to increase the precision of cross-pollen transfer and to reduce interference of stigmas and anthers within or between flowers on the same plant. Other authors have suggested that enantiostyly may only facilitate the access of the pollen-collecting bees to the middle anthers, which the bees clasp and vibrate, forcing the bees to adopt a position that results in a greater pollen removal (Westerkamp, 2004) and also protecting the gynoecium from damage by buzzing bees (Dulberger, 1981;Dulberger et al., 1994).
Diverse floral (a)symmetry patterns in Senna, including complex enantiostyly with highly asymmetric androecium and corolla, seem to be especially involved in pollen dispersal. Different pointing directions of anther pores appear to be related with different directions of pollen release (Marazzi et al., 2007). Most anthers release pollen directly toward the floral center and thus toward the bees. During buzzing, vibrational energy is transmitted either from the thorax of the bee (clasping the middle stamens) to the flower and thus to other floral parts, such as the abaxial pollinating stamens, the carpel, and petals (Buchmann and Hurley, 1978;Westerkamp, 2004) or by the body of the bee touching the pollinating stamens (Endress, 1997). In highly asymmetric flowers, anther pores of the abaxial pollinating stamens are usually directed toward the lower petals (Marazzi et al., 2007). These petals are positioned in such a way that pollen loss is avoided: they are highly concave, foot-shaped, and/or asymmetric, and they partially surround the pollinating stamens (Figs. 2F–H, L, M, 3D–L). Therefore, when bees buzz these flowers, pollen from the pollinating stamens is released toward the lower petals; vibration of these lower petals helps the pollen ricochet and finally adhere to the bees (e.g., Delgado Salinas and Sousa Sánchez, 1977;Westerkamp, 2004). We observed that in species of clade II, with enantiostylous flowers and monosymmetric corolla, the standard petal is pronouncedly concave and partially encloses the floral reproductive organs (Fig. 2C–E). In these flowers, the two large pollinating stamens are curved upward, and their anther pores are directed toward the standard petal (Marazzi et al., 2007;Fig. 2C–E). Therefore, the released pollen flow likely ricochets off the standard petal in a similar way, as described for flowers with concave and asymmetric lower petals. The particularly robust venation typical of both the highly concave standard petal and the asymmetric lower petals (Fig. 1D, E, and Figs. 1F–H, J, M–P, respectively), and their almost sessile shape, may be adaptations for efficiently transmitting the vibrations produced by the buzzing bees from the middle stamens (on which the bees clasp) to the concave petal blade.
Conclusion
Many kinds of floral asymmetry exist in angiosperms, which are expressed at different times during development and which may affect one or more floral whorls in diverse ways, such as asymmetric position of organ initiation, unequal organ differentiation, and deflection to one side. Floral symmetry is composed of several structural elements and is thus best interpreted as a character complex rather than a single character. Because patterns of floral symmetry may not be strictly homologous, the elements involved in the floral symmetry should be treated individually if used for phylogenetic hypotheses (Bruneau, 1997;Herendeen et al., 2003) or for optimization studies, such as ancestral character state reconstruction (Marazzi et al., 2006). In Senna, complexity of floral symmetry is particularly evident. The genus includes not only monosymmetric flowers, but also several kinds of asymmetric flowers with different organs of different floral whorls contributing to the floral asymmetry, and unrelated species have superficially similar asymmetric flowers. Detailed knowledge on the diverse floral morphology is a precondition for hypotheses on the evolution of floral asymmetry in Senna. In addition to reconstructing the evolution of each floral structural element involved in the floral (a)symmetry, testing for correlations among structural elements would allow us to recognize to what degree they evolved together or independently from each other.
Taxa used in this study, source, and voucher information. Clades or subclades of Senna by Marazzi et al. (2006) and sections by Irwin and Barneby (1982) (AS, Astroites; CH, Chamaefistula; PA, Paradictyon; PE, Peiranisia; PS, Psilorhegma; SE, Senna). Acronyms of herbaria or botanical gardens: CBG = ANBG = Australian National Botanic Gardens; BGB = Botanic Garden of the University of Basel; BGM = Botanic Garden of the University of Munich; BGZ = Botanic Garden of the University of Zurich; CTES = Instituto de Botánica del Nordeste, Corrientes; G = Conservatoire et Jardin Botaniques de la Ville de Genève; HUEFS = Universidad Estadual de Feira de Santana; KPBG = Kings Park and Botanic Garden, Perth; MEXU = Universidad Nacional Autónoma de México; BGCT = Parco Botanic Garden of Canton Ticino (Isole di Brissago); PMA = Universidad de Panamá; PY = Museo Nacional de Historia Natural de Paraguay; SI = Instituto de Botánica Darwinion, San Isidro; STRI = Smithsonian Tropical Research Institute, Balboa; Z = Herbarium of the University of Zurich and Botanic Garden. Cult. = specimen from cultivated plants.
|
FOOTNOTES
1 The authors thank the curators of the cited herbaria; I. Alvárez, A. Conceição, G. Flores, G. López, L. Paganucci de Queiroz, M. Quintana, and R. Vanni for valuable help in the field; the Australian National Botanic Gardens (ANBG), the Botanic Garden of the Canton Ticino (Switzerland), and the Botanic Gardens of the Universities of Munich, Basel, and Zurich for providing plant material for this study; U. Jauch for assistance with the SEM; people of the Endress laboratory for assistance with microtome sectioning; the ANBG and M. Belgrano for providing photographs of flowers of three Senna species; A. Bruneau and two anonymous reviewers for comments on the manuscript; Q.C.B. Cronk for information on Gesneriaceae; and the Georges-und-Antoine-Claraz-Schenkung (University of Zurich) and the Kommission für Reisestipendien der Schweizerischen Akademie der Naturwissenschaften for financial support of the collecting trips. 
2 Author for correspondence (e-mail: brigitte.marazzi{at}systbot.uzh.ch)
LITERATURE CITED
Bruneau, A. 1997. Evolution and homology of bird pollination syndromes in Erythrina (Leguminosae). American Journal of Botany 84: 54–71.[Abstract]
Bruneau, A., F. Forest, P. S. Herendeen, B. B. Klitgaard, AND G. P. Lewis. 2001. Phylogenetic relationships in the Caesalpinioideae (Leguminosae) as inferred from chloroplast trnL intron sequences. Systematic Botany 26: 487–514.[Web of Science]
Buchmann, S. L. 1974. Buzz pollination of Cassia quiedondilla (Leguminosae) by bees of the genera Centris and Melipona. Bulletin of the Southern California Academy of Sciences 73: 171–173.
Buchmann, S. L. 1983. Buzz pollination in angiosperm. In C. E. Jones, and R. J. Little [eds.], Handbook of experimental pollination biology, 73–113. Scientific & Academic Editions, New York, New York, USA.
Buchmann, S. L., AND J. P. Hurley. 1978. A biophysical model for buzz pollination in angiosperms. Journal of Theoretical Biology 72: 639–657.[CrossRef][Web of Science][Medline]
Carvalho, D. A., AND P. E. Oliveira. 2003. Biología reprodutiva e polinização de Senna sylvestris (Vell.) H.S. Irwin & Barneby (Leguminosae, Caesalpinioideae). Revista Brasileira de Botânica 26: 319–328.
Delgado Salinas, A. O., AND M. Sousa Sánchez. 1977. Biología floral del genero Cassia en la región de Los Tuxtlas, Veracruz. Boletín de la Sociedad Botánica de México 37: 5–52.
Dulberger, R. 1981. The floral biology of Cassia didymobotrya and C. auriculata (Caesalpiniaceae). American Journal of Botany 68: 1350–1360.[CrossRef][Web of Science]
Dulberger, R., M. B. Smith, AND K. S. Bawa. 1994. The stigmatic orifice in Cassia, Senna, and Chamaecrista (Caesalpiniaceae): morphological variation, function during pollination, and possible adaptive significance. American Journal of Botany 81: 1390–1396.[CrossRef][Web of Science]
Endress, P. K. 1994. Diversity and evolutionary biology of tropical flowers. Cambridge University Press, Cambridge, UK.
Endress, P. K. 1997. Relationships between floral organization, architecture, and pollination mode in Dillenia (Dilleniaceae). Plant Systematics and Evolution 206: 99–118.[CrossRef][Web of Science]
Endress, P. K. 1999. Symmetry in flowers: diversity and evolution. International Journal of Plant Sciences 160 ( 6 Supplement): S3–S23.[CrossRef][Web of Science][Medline]
Fenster, C. B. 1995. Mirror image flowers and their effect on outcrossing rate in Chamaecrista fasciculata (Leguminosae). American Journal of Botany 82: 46–50.[CrossRef][Web of Science]
Fontanelle, G. B. 1979. Contribuição ao estudo de biologia floral de Cassia silvestris Vell.: consideraçoes anatômicas. Leandra 8-9: 49–84.
Gottsberger, G., AND I. Silberbauer-Gottsberger. 1988. Evolution of flower structures and pollination in Neotropical Cassiinae (Caesalpiniaceae) species. Phyton 28: 293–320.[Web of Science]
Harrison, J. C., M. Möller, AND Q. C. B. Cronk. 1999. Evolution and development of floral diversity in Streptocarpus and Saintpaulia. Annals of Botany 84: 49–60.
Herendeen, P. S., A. Bruneau, AND G. P. Lewis. 2003. Floral morphology in caesalpinioid legumes: testing the monophyly of the " Umtiza clade." International Journal of Plant Sciences 164: S393–S407.[CrossRef][Web of Science]
Igersheim, A. 1993. The character states of the Caribbean monotypic endemic Strumpfia (Rubiaceae). Nordic Journal of Botany 13: 545–559.[CrossRef][Web of Science]
Irwin, H. S., AND R. C. Barneby. 1981. Tribe 2. Cassieae Bronn. In R. M. Pohlhill, and P. H. Raven [eds.], Advances in legume systematics, part 1, 97–106. Royal Botanic Gardens, Kew, UK.
Irwin, H. S., AND R. C. Barneby. 1982. The American Cassiinae. Memoirs of the New York Botanical Garden 35: 1–918.
Jesson, L. K. 2002. The evolution and functional significance of enantiostyly in flowering plants. Ph.D. dissertation, University of Toronto, Toronto, Ontario, Canada.
Jesson, L. K., AND S. C. H. Barrett. 2002a. Solving the puzzle of mirror-image flowers. Nature 417: 707.[CrossRef][Medline]
Jesson, L. K., AND S. C. H. Barrett. 2002b. The genetics of mirror-image flowers. Proceeding of the Royal Society of London, B, Biological Sciences 269: 1835–1839.
Jesson, L. K., AND S. C. H. Barrett. 2003. The comparative biology of mirror-image flowers. International Journal of Plant Sciences 164 ( 6 Supplement): S237–S249.[CrossRef][Web of Science]
Jesson, L. K., AND S. C. H. Barrett. 2005. Experimental tests of the function of mirror-image flowers. Biological Journal of the Linnean Society 85: 167–179.[CrossRef][Web of Science]
Jesson, L. K., J. Kang, S. L. Wagner, S. C. H. Barrett, AND N. G. Dengler. 2003. The development of enantiostyly. American Journal of Botany 90: 183–195.
Kantz, K. E. 1996. Floral development in the caesalpinioid tribe Caesalpinieae (Fabaceae). Ph.D. dissertation, Louisiana State University, Baton Rouge, Louisiana, USA.
Laporta, C. 2003. Floral biology and reproductive system of enantiostylous Senna corymbosa (Caesalpiniaceae). Revista de Biología Tropical 53: 49–61.
Lasseigne, A. 1979. Studies in Cassia (Leguminosae-Caesalpinioideae). III. Anther morphology. Iselya 1: 141–160.
Marazzi, B., E. Conti, AND P. K. Endress. 2007. Diversity in anthers and stigmas in the buzz-pollinated genus Senna (Leguminosae, Cassiinae). International Journal of Plant Sciences 168: 371–391.[CrossRef][Web of Science]
Marazzi, B., P. K. Endress, L. Paganucci de Queiroz, AND E. Conti. 2006. Phylogenetic relationships within Senna (Leguminosae, Cassiinae) based on three chloroplast regions: patterns in the evolution of floral symmetry and extrafloral nectaries. American Journal of Botany 93: 288–303.
Müller, F. 1883. Two kinds of stamens with different functions in the same flower. Nature 27: 364–365.
Okpon, E. N. U. 1969. Morphological notes on the genus Cassia. III. Floral ontogeny and aestivation. Notes from the Royal Botanic Garden Edinburgh 29: 339–342.
Ornduff, R., AND R. Dulberger. 1978. Floral enantiomorphy and the reproductive system of Wachendorfia paniculata (Haemodoraceae). The New Phytologist 80: 427–434.[CrossRef]
Owens, S. J., AND G. P. Lewis. 1989. Taxonomic and functional implications of stigma morphology in species of Cassia, Chamaecrista, and Senna (Leguminosae: Caesalpinioideae). Plant Systematics and Evolution 163: 93–105.[CrossRef][Web of Science]
Prenner, G. 2003. Floral ontogeny in Lathyrus latifolius (Fabaceae-Vicieae). Phyton 43: 392–400.[Web of Science]
Prenner, G. 2004a. New aspects in floral development of Papilionoideae: initiated but suppressed bracteoles and variable initiation of sepals. Annals of Botany 93: 537–545.
Prenner, G. 2004b. The asymmetric androecium in Papilionoideae (Leguminosae): definition, occurrence, and possible systematic value. International Journal of Plant Sciences 165: 499–510.[CrossRef][Web of Science]
Randell, B. R., AND B. A. Barlow. 1998. Senna. Flora of Australia 12: 89–138.
Sokoloff, D. D., G. V. Degtjareva, P. K. Endress, M. V. Remizova, T. H. Samigullin, AND C. M. Valiejo-Roman. 2007. Inflorescence and early floral development in Lotae (Leguminosae) in a phylogenetic and taxonomic context. International Journal of Plant Sciences 168: 801–833.[CrossRef][Web of Science]
Todd, J. E. 1882. On the flowers of Solanum rostratum and Cassia chamaecrista. The American Naturalist 16: 281–287.[CrossRef]
Troll, W. 1951. Botanische Notizen II. Abhandlungen der Akademie der Wissenschaften und der Literatur Mainz Mathematisch-Naturwissenschaftlichen Klasse 1951: 25–80.
Tucker, S. C. 1988. Dioecy in Bauhinia resulting from organ suppression. American Journal of Botany 75: 1584–1597.[CrossRef][Web of Science]
Tucker, S. C. 1991. Stamen structure and development in legumes, with emphasis on poricidal stamens of caesalpinioid tribe Cassieae. In W. G. D’Arcy, and R. C. Keating [eds.], The anther form, function and phylogeny, 236–254. Cambridge University Press, Cambridge, UK.
Tucker, S. C. 1992. The developmental basis for sexual expression in Ceratonia siliqua (Leguminosae: Caesalpinioideae: Cassieae). American Journal of Botany 79: 318–327.[CrossRef][Web of Science]
Tucker, S. C. 1996. Trends in evolution of floral ontogeny in Cassia sensu stricto, Senna, and Chamaecrista. American Journal of Botany 83: 687–711.[CrossRef][Web of Science]
Tucker, S. C. 1997. Floral evolution, development, and convergence: the hierarchical-significance hypothesis. International Journal of Plant Sciences 158 ( 6 Supplement): S143–S161.[CrossRef][Web of Science]
Tucker, S. C. 1998. Floral ontogeny of Petalostylis, Labichea, and Dialium (Caesalpinioideae: Cassieae), a series in floral reduction. American Journal of Botany 85: 184–208.[Abstract]
Tucker, S. C. 1999. Evolutionary lability of symmetry in early floral development. International Journal of Plant Sciences 160 ( 6 Supplement): S25–S39.[CrossRef][Web of Science][Medline]
Tucker, S. C. 2001. Floral development in Schotia and Cynometra (Leguminosae: Caesalpinioideae: Detarieae). American Journal of Botany 88: 1164–1180.
Tucker, S. C. 2002. Floral development in Detarieae (Leguminosae: Caesalpinioideae). 1. Radially symmetrical taxa lacking organ suppression. American Journal of Botany 89: 875–887.
Tucker, S. C. 2003. Floral development in legumes. Plant Physiology 131: 911–926.
Venkatesh, C. S. 1957. The form, structure and special ways of dehiscence in anthers of Cassia. III. Subgenus Senna. Phytomorphology 7: 253–273.
Vogel, S. 1998. Remarkable nectaries: structure, ecology, organophyletic perspectives. III. Nectar ducts. Flora 193: 113–131.
Weber, M., AND A. Igersheim. 1994. Pollen buds in Ophiorrhiza (Rubiaceae) and their role in pollenkit release. Botanica Acta 107: 257–262.[Web of Science]
Westerkamp, C. 2004. Ricochet pollination in cassias—and how bees explain enantiostyly. Preliminary communication. In B. M. Freitas, and J. O. P. Pereira [eds.], Solitary bees. Conservation, rearing and management for pollination, 225–230. Imprensa Universitária, Fortaleza, Brazil.
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  I. Ojeda, J. Francisco-Ortega, and Q. C. B. Cronk Evolution of petal epidermal micromorphology in Leguminosae and its use as a marker of petal identity Ann. Bot., November 1, 2009; 104(6): 1099 - 1110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Compiled by, F. Tooke, T. Chiurugwi, and N. Battey Flowering Newsletter bibliography for 2008 J. Exp. Bot., June 23, 2009; (2009) erp154v1. [Full Text] [PDF]
|
|
|
|
 |
|
 G. Prenner and B. B. Klitgaard Towards unlocking the deep nodes of Leguminosae: Floral development and morphology of the enigmatic Duparquetia orchidacea (Leguminosae, Caesalpinioideae) Am. J. Botany, November 1, 2008; 95(11): 1349 - 1365. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
