| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Anatomy and Morphology |
2 Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville, Tennessee 37996 USA 3 Division of Science, Truman State University, Kirksville, Missouri 63501 USA 4 School of Science, The College of New Jersey, Ewing, New Jersey 08628 USA
Received for publication 17 November 2007. Accepted for publication 10 January 2008.
ABSTRACT
Cabomba is a small water lily genus that is native to the New World. Studies of pollen development and associated changes in the anther yield valuable characters for considering the evolution of reproductive biology in seed plants. Here we characterized the complete ontogenetic sequence for pollen in Cabomba caroliniana. Anthers at the microspore mother cell, tetrad, free microspore, and mature pollen grain stages were studied using scanning electron, transmission electron, and light microscopy. Tetragonal and decussate tetrads both occur in C. caroliniana, indicating successive microsporogenesis. The exine is tectate-columellate, and the infratectal columellae are the first exine elements to form, followed by a continuous tectum and a thin foot layer. A lamellate endexine initiates in the early free microspore stage, but becomes compressed in mature grains. Tectal microchannels and sculptural rods also initiate during the early free microspore stage, and significant pollenkitt deposition follows, supporting the hypothesis that these elements function in entomophily. The tapetum is morphologically amoeboid, with migratory tapetal cells directly contacting developing free microspores within the anther locule. Results from this study illustrate the importance of including ontogenetic data in analyzing pollen characters and in developing evolutionary and ecological hypotheses. The new palynological data also emphasize the character plasticity that occurs in basal angiosperms.
Key Words: basal angiosperms • Cabomba • Cabombaceae • development • Nymphaeales • ontogeny • pollen • ultrastructure
Cabomba is a small aquatic genus in the order Nymphaeales (water lilies). The Nymphaeales have traditionally encompassed two families: Cabombaceae (Brasenia, Cabomba) and Nymphaeaceae (Victoria, Euryale, Nymphaea, Ondinea, Barclaya, Nuphar). Phylogenetic relationships among these genera are fairly well resolved, with Brasenia and Cabomba consistently placed as sister taxa within a separate family (Les et al., 1999
; Podoplelova and Ryzhakov, 2005; Lohne et al., 2007). Recently, molecular evidence has indicated that the small family Hydatellaceae (Hydatella, Trithuria) is sister to Nymphaeaceae and Cabombaceae and may be included within Nymphaeales (Saarela et al., 2007; Friis and Crane, 2007).
Encompassing five to seven species, Cabomba is distributed throughout tropical and temperate regions of the American continents and the West Indies (Orgaard, 1991; Williamson and Schneider, 1993). Cabomba, especially C. caroliniana Gray, is a popular aquarium plant, and the genus has been introduced globally through discarded aquaria. Furthermore, C. caroliniana can be highly invasive and is regulated in several states in the United States and Australia (Mackey, 1996; Wilson et al., 2007).
Cabomba is characterized by the presence of two types of leaves: (1) highly dissected, fan-shaped, submerged leaves that give the genus its common name fanwort; and (2) minute, floating, peltate leaves associated with axillary or extraaxillary flowers (Orgaard, 1991). Cabomba is protogynous with a 2-day flowering period. On the first day of anthesis, flowers are functionally pistillate. Sepals and petals reflex, and the separate carpels radiate slightly, extending over the nondehiscent anthers. Stigmata are receptive in first-day flowers. Flowers close in the evening, submerge, and reemerge as functionally staminate flowers in the morning of the second day of anthesis. In second-day flowers, carpels are aggregated toward the flower center, and stigmata are no longer receptive. Filaments have elongated, bringing anthers to the same height as the stigmata. Anther dehiscence occurs at midday. Cabomba is pollinated by small insects, primarily flies (Diptera), and many of its floral and pollen characters are adapted for this pollination syndrome (Schneider and Jeter, 1982; Osborn et al., 1991).
The Nymphaeales are of considerable interest because an array of phylogenetic studies has indicated that the water lilies are among the earliest-diverging lineages of flowering plants. These analyses consistently place either Amborella, or Amborella plus Nymphaeales, as sister to the remaining angiosperms (e.g., Qiu et al., 1999; Barkman et al., 2000; Doyle and Endress, 2000; Graham and Olmstead, 2000; Soltis et al., 2000; Borsch et al., 2003; Hilu et al., 2003; Aoki et al., 2004; Lohne and Borsch, 2005; Qiu et al., 2006; Saarela et al., 2007; see also Soltis and Soltis, 2004 and references therein). Studies of pollen development and associated changes in the anther yield valuable characters for investigating critical evolutionary changes in the reproductive biology of seed plants, as well as for assessing evolutionary relationships (Blackmore and Crane, 1988; Blackmore and Barnes, 1990; Kreunen and Osborn, 1999; Taylor and Osborn, 2006; Blackmore, 2007; Blackmore et al., 2007). Recent investigations of pollen morphology and development in Nymphaeales are adding to our understanding of critical pollen characters in these basal angiosperms. However, there are still considerable gaps and conflicting reports about pollen characters in the water lilies. Pollen ontogeny has been studied in detail in only three (Brasenia, Cabomba, Nymphaea) of the 10 genera comprising Nymphaeaceae, Cabombaceae, and Hydatellaceae (Gabarayeva and Rowley, 1994; Gabarayeva and El-Ghazaly, 1997; Gabarayeva et al., 2001; Gabarayeva et al., 2003; Taylor and Osborn, 2006).
Several studies of mature pollen morphology and anther histology have been undertaken for Cabomba, with the majority of these investigations utilizing either transmitted light microscopy (LM) (Wodehouse, 1932; Snigerevskaya, 1955) or both LM and scanning electron microscopy (SEM) (Galati, 1985). Mature pollen of Cabomba has been described as bicellular, with grains elliptical and monosulcate. Parallel sculptural rods characterize the nonapertural pollen surface in all species except C. palaeformis Fassett, which has been described as verrucate (Orgaard, 1991). The presence of parallel rods makes Cabomba one of only two water lily genera with major sculptural elements on the pollen surface; Nuphar pollen is characterized by vertical spines (Takahashi, 1992). Ultrastructure of the mature pollen wall in Cabomba has been studied with SEM and transmission electron microscopy (TEM) (Ueno and Kitaguchi, 1961; Roland, 1968; Osborn et al., 1991; Gabarayeva et al., 2003).
Fewer studies have addressed pollen development in Cabomba. The majority of these used either LM (Padmanabhan and Ramji, 1966; Yakovlev, 1981) or LM and SEM (Galati, 1985), and these studies presented most or all of their results exclusively in the form of pen-and-ink line drawings. The ontogeny of the pollen wall was not examined in any of these published studies. Only Gabarayeva et al. (2003) have investigated pollen wall development in the genus, and their study concentrated specifically on exine substructure after the onset of meiosis in C. aquatica Aubl.
The objective of the current study was to comprehensively characterize pollen ontogeny in Cabomba caroliniana using LM, SEM, and TEM. In contrast to previous investigations, this report focuses on both ontogeny of pollen wall ultrastructure and associated developmental characters in the microspore and tapetum throughout the entire ontogenetic sequence. Developmental and morphological characters are then discussed with regard to characters known from Nymphaeales and other basal angiosperms.
MATERIALS AND METHODS
Plant material
Floral buds of Cabomba caroliniana are located in submerged clusters associated with floating leaves. Over 200 submerged buds at various stages of development were collected from Aquarena Springs in San Marcos, Texas, USA. Fresh buds were sorted into eight size classes based on bud length and associated developmental characters. In addition to buds, emergent, pistillate first-day flowers and staminate second-day flowers were also collected.
The ends of buds in the two smallest size classes (<1.5 mm and 1.5–3.0 mm) were sliced open using a double-edge razor blade to facilitate fixation. These buds were chemically fixed for 24 h in 3% glutaraldehyde (in 0.2 M phosphate buffer; pH 7.4) and then washed in the buffer at least three times. The buds were postfixed for 3–6 h in buffered 2% osmium tetroxide and then buffer-washed. All larger buds and the emergent, open flowers were dissected to remove the perianth, then fixed using the same protocol. All specimens were dehydrated in a graded ethanol series and stored in 75% ethanol.
Pollinators, primarily Notiphila flies, were also collected from open flowers using standard insect collection jars that had been saturated with ethyl acetate. The flies were stored dry in glass vials to prevent any loss of pollen grains or pollen surface materials.
Microscopy
For LM and TEM, several buds from each size class were further dehydrated in a graded ethanol series, infiltrated and embedded in Spurr's epoxy resin (Electron Microscopy Sciences, Hatfield, Pennsylvania, USA) using standard protocols. For observing gross pollen and anther histology, embedded specimens were thick-sectioned with either an RMC MT-6000XL (Tucson, Arizona, USA) or a Leica Ultracut UCT (Bannonkburn, Illinois, USA) ultramicrotome using glass or diamond knives. Thick sections (850 nm) were collected on glass slides, stained with Richardson's stain (methylene blue and azure II) and imaged with an Olympus BHS (Lake Success, New York, USA) compound microscope using bright-field or differential interference contrast optics. Well-preserved specimens at a variety of developmental stages were thin-sectioned using a diamond knife. Thin sections (70 nm) were collected and dried on formvar-coated copper slot grids. Specimens were stained with uranyl acetate (8 min) and lead citrate (6 min) and imaged with a JEOL JEM-100SX (Peabody, Massachusetts, USA) transmission electron microscope at 80 kV.
For SEM, anthers and buds were further dehydrated and critical point dried following standard protocols. Anthers were either cut into cross sections with a double-edged razor blade or macerated and mounted onto aluminum stubs with double-sided tape, with the edges of the tape coated with colloidal graphite. Stubs were sputter-coated with gold-palladium and imaged with a JEOL JSM-6100 scanning electron microscope at 5 kV. In addition, some thick sections of interest were adhered to glass coverslips using poly L-lysine, and the sections were then de-embedded using a Cured Epoxy Remover kit (Electron Microscopy Sciences). The glass coverslips bearing the de-embedded thick sections were then adhered to aluminum stubs using colloidal graphite, sputter-coated, and imaged as described above. For the insect pollinators, air-dried flies were mounted directly onto stubs using colloidal graphite, sputter-coated, and imaged with a Hitachi S-500 (Pleasanton, California, USA) scanning electron microscope at 20 kV.
Size ranges of pollen characters were determined by measuring characters on micrographs of at least five microspores or pollen grains.
RESULTS
Timing of pollen development is synchronous within each anther and is consistent among anthers of the same flower. Results are presented next for key ontogenetic stages, including sporogenous tissue, microspore mother cell, tetrad, free microspore, and mature pollen grain stages.
Sporogenous tissue stage
Sporogenous tissue is found in buds <1.5 mm in length. In sporogenous tissue, the cells are slightly polygonal and tightly appressed, completely filling the locular space (Fig. 1). The cytoplasm stains densely, and the nuclei are large. During the early sporogenous stage, the tapetum is undifferentiated (Fig. 1).
|
The tapetum also differentiates in the microspore mother cell stage. Tapetal cells have one to two nuclei, are tightly adjoined to each other, and remain in contact with the middle layers of the anther wall throughout the microspore mother cell stage (Fig. 2).
Tetrad stage
Tetrads are also located in buds <1.5 mm long. Each microspore mother cell undergoes meiosis to form four microspores that are held together in a tetrad. A thick layer of callose surrounds the entire tetrad, while thinner callosic septae extend between the individual microspores (Figs. 6–8). The majority of tetrads occur in tetragonal or decussate configurations, with both types co-occurring in the same anther locule (Figs. 6–8). A small number of tetrads with tetrahedral geometries were also documented. When these occurred, they were found in the same locule as tetragonal and decussate types.
|
Middle tetrad stage
In the middle tetrad stage, a slightly electron-dense layer is laid down at the base of the primexine, adjacent to the plasma membrane. The plasma membrane forms slight projections that extend through the primexine toward the callose (Fig. 10). These projections develop into procolumellae, and as the tetrad stage progresses, sporopollenin is deposited on the procolumellae, and these elements form true columellae. These infratectal columellae are the first elements of the exine to develop (Fig. 11). After the columellae are initiated, the tectum begins to develop. The tectum is laid down in continuous or nearly continuous layers, such that there are few gaps in the horizontal span of the tectum (Fig. 11). As the tectum develops, it is also homogenous in ultrastructure and is consistent in thickness in all regions of the microspore.
Late tetrad stage
By the late tetrad stage, the thickness of the tectum ranges from 45 to 90 nm. The columellae have become increasingly robust, with widths ranging from 140 to 170 nm and heights ranging from 210 to 320 nm by the end of the tetrad stage. Columellae tend to be widest in the center of the infratectal space, tapering slightly where they contact the adjacent wall layers (Figs. 11, 12). Although single columellae are the predominate infratectal element, several branched columellae were also observed, with multiple branches joined near the base of the columellae (Fig. 11). After significant tectum and infratectum deposition, a thin foot layer is laid down at the base of the columellae (Fig. 11). As the foot layer is deposited, the microspore mother cell coat surrounding the callose begins to disintegrate. Formation of the aperture also initiates during the late tetrad stage (Fig. 12). The region of the exine where the aperture will develop invaginates slightly, forming a shallow groove in the distal wall of the microspore (Figs. 7, 12). The tectum in this apertural region (Fig. 12) is more discontinuous than the tectum in nonapertural regions of the wall (Fig. 11).
The tapetum is attached to the anther wall throughout the tetrad stage. In early tetrads, tapetal cells remain tightly appressed to each other and to the underlying middle layers. As development continues, the middle lamellae joining the anticlinal walls of the tapetal cells begin to disassociate, allowing the tapetal cells to separate from one another (Figs. 6, 13). As they separate, tapetal cells become irregularly shaped and protrude into the anther locule (Fig. 13).
Free microspore stage
Free microspores are found predominately in buds <3.0 mm long, with the majority of these buds ranging from 1.5 to 3.0 mm long. A few larger buds (3.0–7.0 mm) also contain free microspores. At the end of the tetrad stage, the callose disassociates, and the microspores become separated from one another, initiating the free microspore stage.
Early free microspore stage
In the early free microspore stage (Figs. 14–18), all major components of the exine continue to develop. The columellae elongate to their maximum height, reaching heights that range from 470 to 565 nm. As columellae elongate, they become approximately two times thinner than those in the late tetrad stage. The tectum also continues to thicken, reaching thicknesses near 100 nm. As the tectum thickens, it becomes dissected with thin, vertically oriented microchannels that extend through the entire tectum (Figs. 16–18).
|
|
|
Middle free microspore stage
During the middle free microspore stage (Figs. 19–22), additional sporopollenin is deposited on the columellae, so that the width of the columellae ranges from 250 to 360 nm by the time of intine initiation (Fig. 21). This range is approximately double the width of the columellae at the end of the tetrad stage. Similarly, the thickness of the tectum increases nearly threefold. The foot layer also becomes thicker, from 100 to 230 nm by initiation of the intine (Fig. 21).
|
During the middle free microspore stage, tapetal cells migrate into the anther locule, coming into close proximity to, and/or in direct contact with developing microspores (Fig. 19). The distance into the locule that tapetal cells migrate differs by anther. In some anthers, tapetal cells were present in the center of the locule, while in others the tapetal cells were clearly detached from the middle layers, but remained near the anther wall. Tapetal cells continue to have densely staining cytoplasm and remain distinctly multinucleate (Fig. 19).
Late free microspore stage
In the late free microspore stage (Figs. 23–26), the innermost layer of the pollen wall, the intine, begins to develop. The plasmalemma pulls away in places from the endexine, leaving gaps where the intine forms (Fig. 25). As more of the plasmalemma separates from the endexine, the intine becomes continuous (Fig. 26). The intine increases in thickness throughout the late free microspore stage, becoming thickest beneath the aperture. As the intine develops, the endexine becomes compressed in nonapertural regions, and as a result, the individual lamellae are no longer discernible (Fig. 26). The endexine, however, remains uncompressed in the apertural regions.
Following initiation of the intine, the tapetal cells begin to undergo programmed cell death, and prior to tapetum disintegration, orbicules are released into the anther locule (Fig. 26). Numerous orbicules can be observed near the developing microspores, as well as on the surface of late free microspores (Figs. 24, 26). Pollenkitt deposition begins during the late free microspore stage (Fig. 24), with substantial pollenkitt accumulating later in development between the first and second day of anthesis. By the end of the late free microspore stage, radial cell wall thickenings, or bands, develop in the endothecial layer of the anther wall (Fig. 23).
Mature grain stage
Mature pollen grains are exclusively found in buds >1.5 mm in length, with the majority of mature grains occurring in buds >5.0 mm long and in open flowers. Mature pollen grains are prolate, averaging 81 µm in length and 61 µm in width (Figs. 27, 28).
The mature pollen wall is characterized by robust infratectal columellae, averaging 540 nm in height and 460 nm in width; a thick tectum averaging 380 nm in thickness and dissected by vertically oriented microchannels that average 65 nm in diameter; and a distinct foot layer averaging 140 nm in thickness (Figs. 29, 35, 36). The lamellate endexine is compressed in the nonapertural regions so that the individual lamellae are not discernible, but the endexine remains uncompressed beneath the aperture. The intine is bilayered, distinctly channeled, and forms a thick inner layer, reaching an average thickness of 1.88 µm in nonapertural regions and 1.37 µm beneath the aperture (Figs. 29, 38).
|
DISCUSSION
This study is the first on pollen wall ontogeny in Cabomba caroliniana and the first to comprehensively report the entire pollen developmental sequence in the genus using multiple and coordinated microscopical approaches. Several key developmental characters are discussed next and are summarized in Fig. 40. These characters are considered in light of what is known from studies of C. aquatica (Gabarayeva et al., 2003) and other Nymphaelaean genera and in the context of current phylogenetic hypotheses for water lilies and other basal angiosperms.
|
Tetrad geometry and microsporogenesis
Tetrads in Cabomba are predominately found in tetragonal and decussate configurations. A few tetrads with tetrahedral geometries were also observed, but these were rare. These data are consistent with Yakovlev (1981) and Galati (1985), but conflict with the report of tetrahedral geometries as the primary tetrad type by Padmanabhan and Ramjii (1966). This character was not reported by Gabareyeva et al. (2003) for C. aquatica, so interspecific comparisons cannot be made. Tetragonal and decussate tetrad geometry indicates that microsporogenesis in Cabomba is successive.
Microsporogenesis can be described as simultaneous if cytokinesis and callose deposition both occur after meiotic divisions are complete or as successive if callose is deposited between the first and the second meiotic divisions. Simultaneous microsporogenesis results in tetrads that are held in a tetrahedral configuration, while successive microsporogenesis results in tetrads that have an array of configurations, including tetragonal and decussate (Furness et al., 2002). Furness et al. (2002) considered this trait to be labile in basal angiosperms with simultaneous microsporogenesis as the plesiomorphic character state, based on the predominance of this type in basal angiosperms, and indeed, all land plants. However, other authors have considered successive microsporogenesis to be the plesiomorphic state (Sampson, 1970; Blackmore and Crane, 1988; Tobe et al., 2000).
This character is of significant interest because it is highly variable within water lilies and potentially among basal angiosperms. Brasenia undergoes successive microsporogenesis (Taylor and Osborn, 2006), as do Amborella (Furness et al., 2002) and Trimenia (Endress and Sampson, 1983). In contrast, Nymphaea (Gabarayeva and Rowley, 1994), Nuphar (Takahashi, 1992), and Victoria (Roland, 1965; Osborn et al., 2004), as well as Austrobaileya (Zavada, 1984), Kadsura, and Schisandra (Saunders, 2000) undergo simultaneous microsporogenesis. However, in Schisandra grandiflora, both tetragonal and decussate tetrads are reported (Saunders, 2000), indicating successive microsporogenesis in this species. Investigations of Hydatellaceae and a reevaluation of understudied genera of Nymphaeaceae will likely add clarity to our understanding of the early evolution of microsporogenesis in angiosperms.
Tectum ultrastructure
The exine in Cabomba is relatively thick, with a well-defined tectum, infratectum, and foot layer. The tectum initiates in the middle tetrad stage, soon after the columellae are formed. In C. caroliniana, the tectum is laid down in a nearly continuous layer. In C. aquatica, the tectum also initiates during the middle tetrad stage, but Gabarayeva et al. (2003) report that it develops discontinuously through the late tetrad stage. Continuous tectum development in C. caroliniana is also in contrast to tectum formation in Brasenia, which occurs through the deposition of small, globular tectal elements (Taylor and Osborn, 2006). In Brasenia, the tectum never becomes continuous and microchannels result from the discontinuities (Taylor and Osborn, 2006), whereas in C. caroliniana, the tectum forms as a nearly continuous layer and later becomes dissected with microchannels. It is possible that the tectum develops discontinuously during the earliest stage of development and that this tectal character was not observed in the current study because of difficulty in finding early tetrads. However, no indication of discontinuous development was observed in later tetrads, as reported for Brasenia and C. aquatica, and microchannels in C. caroliniana are clearly the result of secondary dissection and not artifacts of discontinuous development.
Interestingly, in the developing apertural region of C. caroliniana pollen, the tectum is less continuous than in the nonapertural wall. Initially, this tectal structure within the apertural region is due to the invagination of the pollen wall. After the apertural groove forms, however, sporopollenin deposition is modified in this region such that large supratectal rods do not form (discussed in greater depth later). The tectum in the apertural region remains discontinuous for the duration of the free microspore stage, resulting in a less organized tectal layer over the aperture, thereby facilitating pollen tube emission at germination.
Infratectum ultrastructure
The infratectal columellae in Cabomba are the first elements of the pollen wall to form, and these remain distinct throughout ontogeny. In mature pollen grains, the columellae are robust and in this way are similar to the infratectal columellae in Brasenia (Taylor and Osborn, 2006). Gabarayeva et al. (2003) reported an abundance of complex, or branching, columellae in C. aquatica. However, in the current study of C. caroliniana, the majority of columellae were simple and unbranching.
In Cabomba, the exine has been reported as granular (Wodehouse, 1932; Walker, 1976), clavate-intectate (Ueno, 1962), and baculate (Galati, 1985). However, the most recent investigations have shown that Cabomba is clearly tectate-columellate (Osborn et al., 1991; Gabarayeva et al., 2003), and the current study supports this interpretation and augments it with ontogenetic data. Granular infratectal elements have historically been considered the plesiomorphic state in angiosperms (Doyle et al., 1994), and this type of infratectum was a character uniting early angiosperms with putative gymnospermous relatives under the anthophyte hypothesis (for discussion, see Osborn et al., 1993; Osborn and Taylor, 1995; Osborn, 2000). However, this assessment was based, in part, on reports of granular infratectal elements in Nymphaeales, which have since been shown to be incorrect (e.g., Osborn et al., 1991, Osborn et al., 2004; Rowley et al., 1992; Gabarayeva and Rowley, 1994; Gabarayeva and El Ghazaly, 1997; Gabarayeva et al., 2001; Taylor and Osborn, 2006). Brasenia and Cabomba, as well as all six genera of Nymphaeaceae have columellar infratectal elements, although the columellae in Nymphaeaceae are less robust than those in Cabombaceae (Rowley et al, 1992; Gabarayeva and Rowley, 1994; Gabarayeva and El Ghazaly, 1997; Gabarayeva et al., 2001; Osborn et al., 2004; Taylor and Osborn, 2006). In addition, columellar infratectal elements have been described in Austrobaileyales (Zavada, 1984; also see Doyle and Endress, 2000), Illicium religiosum (Takahashi, 1994), and Schisandra chinensis (Gabarayeva and Grigorjeva, 2003). Not all basal angiosperms have infratectal columellae, however. In Amborella, the exine lacks columellae and is instead composed of tectal cupules that rest directly on the foot layer (Hesse, 2001). Schisandra pollen is described as having a bacular infratectum (Saunders, 2000). Moreover, the infractal elements of both Hydatella and Trithuria are described as round bacula (Bortenschlager et al., 1966), although pollen structure in these genera is in critical need of reassessment.
Gabarayeva et al. (2003) suggested that the interpretation of the infratectum in Cabomba as granular by Walker (1976) is due to lateral stretching of the exine after columellae formation, resulting in fragmented columellae that resemble granules. However, in the current study no evidence of stretching was observed, and the vast majority of columellae remained intact and distinct throughout exine development. We suggest that fragmented columellae as described for C. aquatica (Gabarayeva et al., 2003) are due to diagonal orientation, branching, and tapering of columellae. Columellae that are either oriented slightly diagonally or are branched in section view potentially appear only near the tectum or the foot layer. Tapered columellae, if sectioned peripherally, would appear only in the middle of infratectum. This interpretation is supported by the presence of diagonal, branching, and tapering columellae in SEM sections of C. caroliniana. No columellae that had actually separated into fragments were observed. In addition, columellae did appear as granules in oblique sections of the exine. This appearance is also a potential reason for the misinterpretation by Walker (1976).
Endexine
The endexine is initiated in the early free microspore stage with individual layers deposited in discontinuous segments. As the free microspore stage progresses, adjacent endexine segments join, become continuous, and the overall endexine layer thickens. Endexine is present in both apertural and nonapertural regions of the pollen wall. However, in nonapertural regions, the endexine layer as a whole becomes compressed during the late free microspore stage such that individual lamellations are not discernible in mature grains. Individual endexine lamellations often remain identifiable below the aperture. Endexine formation occurs in a similar manner in C. aquatica (Gabarayeva et al., 2003) and Brasenia (Taylor and Osborn, 2006), although lamellations are more distinct throughout endexine development in Brasenia.
The nature of the endexine in Cabomba has also been debated. Roland (1968) reported a multilamellate endexine present in the apertural region in C. aquatica, and Gabarayeva et al. (2003) confirmed the presence of distinct lamellations in the apertural region of this species. Using Walker's (1974) micrographs, Orgaard (1991) considered the endexine in C. caroliniana as thickened under the aperture and much reduced in the nonapertural regions. Osborn et al. (1991), however, reported the absence of a lamellate endexine in C. caroliniana. Orgaard (1991) and Osborn et al.'s (1991) interpretations were likely due to the compressed nature of individual lamellae in mature pollen, which contribute to the appearance of the endexine as nonlamellate. The development of individual endexine lamellae and their subsequent ontogenetic compression were elucidated in the current study only through an investigation of the complete developmental sequence in Cabomba. These new data underscore the importance of including ontogenetic data in character analysis.
Sculptural rod formation
The nonapertural surface of mature Cabomba pollen is characterized by the presence of both large and small rods that extend the length of the pollen grain. Gabarayeva et al. (2003) termed these macro- and microstriae, respectively. The large sculptural rods initiate during the early free microspore stage, while the smaller rods are not apparent until the middle free microspore stage after substantial development of the large rods. Whereas the large rods physically disrupt the tectum, causing a shift in the orientation of microchannels, the smaller rods form between the microchannels and have no apparent impact on tectum ultrastructure.
Cabomba and Nuphar are the only two Nymphaealean genera with pollen that has major sculptural elements, and the morphological and developmental nature of the sculptural elements in these genera differs dramatically. Nuphar pollen has large vertical spines, whereas the rods of Cabomba are longitudinally oriented. Moreover, in Cabomba the rods form during the early free microspore stage, after all other wall layers and elements of the ektexine have already developed. The sculptural rods in Cabomba do not extend into the infratectal layer, although the bases of the rods extend into the tectum. In contrast, sculptural spines in Nuphar pollen are the first elements of the sporoderm to form, initiating in the tetrad stage and extending to the base of the ektexine (Takahashi, 1992). Despite the differences between Cabomba and Nuphar, the presence of major sculptural elements in these two genera is interesting in light of current phylogenies that place Nuphar as basal in Nymphaeaceae, with close affinities to Cabombaceae (Les et al., 1999; Lohne et al., 2007).
The surface of the apertural wall of Cabomba is covered with short, randomly oriented rodlets. These rodlets develop during the early to middle free microspore stage and are similar in morphology and development to the rodlets that cover the nonapertural surface of Brasenia pollen grains. In Brasenia, the rodlets form as part of the outer layer of the tectum, initiating during the middle free microspore stage. Rodlets in Brasenia are 1.0 µm long (Taylor and Osborn, 2006), compared to 1.8 µm in Cabomba.
Osborn et al. (1991) hypothesized that the large sculptural rods in Cabomba were an adaptation for insect pollination. These authors suggested that the large rods increase the surface area available for contact with insect hairs and that the parallel orientation of the rods creates large troughs in which pollenkitt can accumulate. Brasenia, in contrast, is wind pollinated and does not produce pollenkitt. Instead, the discontinuous tectum, and subsequent formation of rodlets, potentially reduces the density of the exine allowing for more efficient wind transport (Osborn et al., 1991; Taylor and Osborn, 2006). The similarities in development between the surface rodlets in Brasenia and the nonapertural rods and apertural rodlets in Cabomba provide insight into how exine development has been altered in response to selection pressure for ecological parameters (i.e., pollination syndrome).
Gabarayeva et al. (2003) suggested that the sculptural rods in C. aquatica are structurally supported by the infratectal columellae, reporting that "every distal columella tip is aligned with a stria." The current study, however, provided no evidence that the development of columellae and sculptural rods is linked. Columellae were not necessarily aligned with large sculptural rods and, more importantly, both large and small sculptural rods were not necessarily supported by columellae. It is possible, however, that development of the infratectal columellae somehow influences the later initiation of sculptural rods.
Microchannels
The tectum, infratectal columellae, and sculptural rods in mature Cabomba pollen are dissected with well-defined microchannels. Development of microchannels in the tectum initiates during the early free microspore stage. When the tectal microchannels are first formed, they are perpendicular to the plasmalemma. As sporopollenin accumulates in and on the tectum to form the bases of the sculptural rods, the original tectal material is displaced, "pushing" the microchannels at the base of each sculptural rod into diagonal orientations. As the free microspore stage progresses, smaller sculptural rods develop on the tectum in locations between the tectal microchannels. In the late free microspore stage, the columellae and sculptural rods also become dissected with microchannels. These channels are more randomly oriented than those that perforate the tectum.
In C. aquatica, microchannels also initiate during the early to middle free microspore stage, and these channels ultimately dissect the entire exine, including the foot layer (Gabarayeva et al., 2003). Microchannels have not been identified in the foot layer of C. caroliniana, neither in the current study nor that of Osborn et al. (1991). Gabarayeva et al. (2003) also reported a radial orientation of microchannels within the tectum compared to a more random orientation of microchannels in other layers of the exine.
Osborn et al. (1991) hypothesized that the microchannels, in combination with the orientation of the sculptural rods, play a role in pollenkitt storage. These authors showed that copious pollenkitt is present within the tectal microchannels and covering the entire pollen surface (between and on top of the sculptural rods) at pollen dispersal. To fulfill a role in pollenkitt storage, microchannels must be fully developed before pollenkitt is exocytosed from the tapetum. Ontogenetic data from the current study confirm this hypothesis. Microchannels develop in the tectum during the early free microspore stage and later in the columellae and supratectal rods during the middle free microspore stage. Thus, microchannel development in Cabomba coincides with sculptural rod formation, entry of the migratory tapetal cells into the anther locule, and occurs before pollenkitt deposition.
Tapetum type
The tapetum in Cabomba caroliniana can be morphologically characterized as amoeboid, while retaining some characteristics of secretory tapeta. The tapetal layer differentiates in the microspore mother cell stage, and tapetal cells remain attached to the anther middle layers throughout the microspore mother cell and tetrad stages. In the early and middle free microspore stage, the binucleate tapetal cells become detached from the anther wall and migrate to the center of the locule, where they come into direct contact with developing microspores. The tapetal cells directly contact the surface of the microspores, allowing for the release of transported material adjacent to sites of absorption and polymerization on the microspore. This tapetum type is in contrast to a secretory tapetum, in which tapetal cells remain attached to the anther middle layers for the entirety of pollen development and secrete their contents into the anther locule, much farther from the sites of absorption and polymerization (Pacini et al., 1985; Pacini and Franchi, 1994). It is important to note, however, that amoeboid and secretory tapeta represent the ends of a spectrum of different types of tapetal form and function, and several intermediate types have been identified (Pacini et al., 1985; Rowley et al., 1992; Furness and Rudall, 2001; Furness, 2008).
Although tapetal cells in C. caroliniana enter the locule and contact the developing microspores, they also apparently retain the ability to act as secretory tapetal cells. In the current study, the degree to which tapetal cells entered the locule was variable. In some anthers, the tapetal cells reached the middle of the locule, while in others the cells remained near the middle layers of the anther wall. In addition, orbicules (or Ubisch bodies) are present in C. caroliniana. These small bodies are composed of sporopollenin and are of tapetal origin. Orbicules are almost always associated with secretory tapeta and have been hypothesized to transport sporopollenin from the tapetum to the microspore or to aid in pollen dispersal (Keijzer, 1987; Furness and Rudall, 2001). Orbicules have been reported in a few eudicot taxa with amoeboid tapeta, specifically in the euastrids. This condition has not been reported in early-divergent eudicots or in monocots (Furness, 2008). The current study is the first report of orbicules associated with an amoeboid tapetum in early-divergent angiosperms (see Furness and Rudall, 2001; Furness, 2008).
New ontogenetic data from the current study on the presence of an amoeboid tapetum in C. caroliniana is in conflict with previous studies. Wunderlich (1954) and Padmanabhan and Ramji (1966) describe the tapetum in C. caroliniana as secretory, and line drawings by Yakovlev (1981) also illustrate a secretory tapetum in this species. Line drawings by Galati (1985), however, indicate that the tapetal cells in C. australis (= C. caroliniana) are detached from the anther middle layers, but remain near the anther wall. This situation was observed in the current study as well, so it is probable that the variable amount of migration of tapetal cells in C. caroliniana has resulted in the previous characterization of the tapetum as secretory.
A secretory tapetum, considered the plesiomorphic type in angiosperms, is reported in most gymnosperms and in the vast majority of basal angiosperms, including Amborella, Austrobaileya, and Trimenia (Pacini, 1997; Furness and Rudall, 2001), as well as Brasenia (Taylor and Osborn, 2006), Victoria, Euryale, Nuphar, and Nymphaea (Clausen, 1927; Schnarf, 1931; Wunderlich, 1954; see Furness and Rudall, 2001 for a review). However, transitional and amoeboid tapeta have been reported in Nymphaeales and other basal angiosperms. Rowley et al. (1992) describe Nymphaea colorata as having a cyclic invasive tapetum, a type of amoeboid tapetum in which cells migrate to the center of the locule and then back to the edge of the anther walls. This type of tapetum is considered by Furness (2008) to be "invasive nonsyncytial," a transitional tapetum type in which cells invade the anther locule but function as secretory cells. In addition, the tapetum of Kadsura has been described as "invasive nonsyncytial" (Furness and Rudall, 2001). Although Gabarayeva et al. (2003) report that the tapetum in C. aquatica is secretory, they described the cells as invading the anther locule during the tetrad stage and contacting each microspore, with tapetal inclusions penetrating the exine channels. The differences in timing of locule invasion between C. aquatica and C. caroliniana is especially interesting, because invasion during the tetrad stage is a common feature of monocots, while invasion during the free microspore stage is a general feature of dicots (Pacini et al., 1985). These observations in Cabomba highlight the plasticity present in basal angiosperms and support the conclusion, put forth by Furness and Rudall (2001), that characters present in these taxa potentially represent evolutionary experimentation in early angiosperm lineages.
Conclusion
Cabomba is an important genus in the basal angiosperm lineage Nymphaeales. Pollen characters, including developmental characters, are invaluable for evolutionary analyses and are critical for understanding the evolution of reproductive biology in seed plants. Although several reproductive characters have been increasingly studied in basal angiosperms, there are still conflicting reports about salient pollen characters and considerable gaps in our understanding of pollen development in these taxa. Many of the previous investigations provide only line drawings, making it difficult to accurately interpret characters and, in most cases, impossible to examine traits not explicitly described by the authors. In the current study, we have comprehensively described the complete pollen ontogenetic sequence in C. caroliniana, including concomitant changes in the anther and reexamined the ultrastructure of the mature pollen wall. In particular, the pattern of microsporogenesis, infratectal ultrastructure, and endexine development were evaluated to confirm or reject conflicting descriptions of these traits. Moreover, the presence of a morphologically amoeboid tapetum with intermediate function has been described in Cabomba, and orbicules produced by an amoeboid tapetum have been identified for the first time in basal angiosperms. Developmental timing of sculptural rods and tectal microchannels, along with invasion of the locule by migratory tapetal cells, supports the hypothesis that these elements of the exine function in pollenkitt storage. Continuous tectum development with microchannel formation and sculptural rod development also illuminate ontogenetic differences between Cabomba and its sister genus Brasenia, which reflect reproductive adaptations for entomophilous and anemophilous pollination syndromes in these taxa. The majority of the pollen characters discussed in this paper can only be interpreted in light of knowledge about the entire developmental sequence. This elucidation emphasizes the crucial role that ontogenetic data play in understanding character evolution and phylogenetic reconstruction.
FOOTNOTES
1 The authors thank P. Williamson and D. Dodson (Texas State University, San Marcos) for assistance with field collection. This study was supported in part by the National Science Foundation (grants IBN-0212521, MRI-0216391), Truman State University (Undergraduate Student Research Grants), and The College of New Jersey. 
5 Author for correspondence (e-mail: josborn{at}tcnj.edu)
LITERATURE CITED
Aoki, S., K. Uehara, M. Imafuku, M. Hasebe, AND M. Ito. 2004. Phylogeny and divergence of basal angiosperms inferred from APETALA3- and PISTILLATA-like MADS-box genes. Journal of Plant Research 117: 229–244.[Web of Science][Medline]
Barkman, T. J., G. Chenery, J. R. McNeal, J. Lyons-Weiler, W. J. Ellisens, G. Moore, A. D. Wolfe, AND C. W. dePamphilis. 2000. Independent and combined analyses of sequences from all three genomic compartments converge on the root of flowering plant phylogeny. Proceedings of the National Academy of Sciences, USA 97: 13166–13171.
Blackmore, S. 2007. Pollen and spores: Microscopic keys to understanding the earth's biodiversity. Plant Systematics and Evolution 263: 3–12.[CrossRef][Web of Science]
Blackmore, S., AND S. H. Barnes. 1990. Comparative studies of mature and developing pollen grains. In D. Claugher [ed.], Scanning electron microscopy in taxonomy and functional morphology, 1–21. Clarendon Press, Oxford, UK.
Blackmore, S., AND P. R. Crane. 1988. The systematic implications of pollen and spore ontogeny. In C. J. Humphries [ed.], Ontogeny and systematics, 83–115. Columbia University Press, New York, New York, USA.
Blackmore, S., A. H. Wortley, J. J. Skvarla, AND J. R. Rowley. 2007. Pollen wall development in flowering plants. New Phytologist 174: 483–498.[CrossRef][Web of Science][Medline]
Borsch, T., K. W. Hilu, D. Quandt, V. Wilde, C. Neinhuis, AND W. Barthlott. 2003. Non-coding plastid trnT-trnF sequences reveal a well resolved phylogeny of basal angiosperms. Journal of Evolutionary Biology 16: 558–576.[CrossRef][Web of Science][Medline]
Bortenschlager, V. S., G. Erdtman, AND J. Praglowski. 1966. Pollenmophologische Notizen über einige Blütenpflanzen incertae sedis. Botaniska Notiser 119: 160–171.
Clausen, P. 1927. Ueber das Verhalten des Antheren-Tapetums bei einigen Monokotylen und Ranales. Botanisches Archiv 18: 1–27.
Doyle, J. A., M. J. Donoghue, AND E. A. Zimmer. 1994. Integration of morphological and ribosomal-RNA data on the origin of angiosperms. Annals of the Missouri Botanical Garden 81: 419–450.[CrossRef][Web of Science]
Doyle, J. A., AND P. K. Endress. 2000. Morphological phylogenetic analysis of basal angiosperms: Comparison and combination with molecular data. International Journal of Plant Sciences 161 ( 6 Supplement): S121–S153.[CrossRef][Web of Science]
Endress, P. K., AND F. B. Sampson. 1983. Floral structure and relationships of the Trimeniaceae (Laurales). Journal of the Arnold Arboretum 64: 447–473.[Web of Science]
Friis, E. M., AND P. Crane. 2007. New home for tiny aquatics. Nature 446: 269–270.[CrossRef][Medline]
Furness, C. A. 2008. A review of the distribution of plasmodial and invasive tapeta in eudicots. International Journal of Plant Sciences 169: 207–223.[CrossRef][Web of Science]
Furness, C. A., AND P. J. Rudall. 2001. The tapetum in basal angiosperms: Early diversity. International Journal of Plant Sciences 162: 375–392.[CrossRef][Web of Science]
Furness, C. A., P. J. Rudall, AND F. B. Sampson. 2002. Evolution of microsporogenesis in angiosperms. International Journal of Plant Sciences 163: 235–260.[CrossRef][Web of Science]
Gabarayeva, N. I., AND G. El-Ghazaly. 1997. Sporoderm development in Nymphaea mexicana (Nymphaeaceae). Plant Systematics and Evolution 204: 1–19.[CrossRef][Web of Science]
Gabarayeva, N. I., AND V. V. Grigorjeva. 2003. Comparative study of the pollen wall development in Illicium floridanum (Illiciaceae) and Schisandra chinensis (Schisandraceae). Taiwania 48: 147–167.
Gabarayeva, N. I., V. V. Grigorjeva, AND J. R. Rowley. 2003. Sporoderm ontogeny in Cabomba aquatica (Cabombaceae). Review of Palaeobotany and Palynology 127: 147–173.[CrossRef][Web of Science]
Gabarayeva, N. I., AND J. R. Rowley. 1994. Exine development in Nymphaea colorata (Nymphaeaceae). Nordic Journal of Botany 14: 671–691.[CrossRef][Web of Science]
Gabarayeva, N. I., B. Walles, G. El-Ghazaly, AND J. R. Rowley. 2001. Exine and tapetum development in Nymphaea capensis (Nymphaeaceae): A comparative study. Nordic Journal of Botany 21: 529–548.[CrossRef][Web of Science]
Galati, B. G. 1985. Estudios embriológicos en Cabomba australis (Nymphaeaceae). I. La esporogénesis y las generaciones sexuadas. Boletín de la Sociedad Argentina de Botánica 24: 29–47.
Graham, S. W., AND R. G. Olmstead. 2000. Utility of 17 chloroplast genes for inferring the phylogeny of the basal angiosperms. American Journal of Botany 87: 1712–1730.
Hesse, M. 2001. Pollen characters of Amborella trichopoda (Amborellaceae): A reinvestigation. International Journal of Plant Sciences 162: 201–208.[CrossRef][Web of Science]
Hilu, K. W., T. Borsch, K. Muller, D. E. Soltis, P. S. Soltis, V. Savolainen, M. W. Chase, M. P. Powell, L. A. Alice, R. Evans, H. Sauquet, C. Neinhuis, T. A. B. Slotta, J. G. Rohwer, C. S. Campbell, AND L. W. Chatrou. 2003. Angiosperm phylogeny based on matK sequence information. American Journal of Botany 90: 1758–1776.
Keijzer, C. J. 1987. The processes of anther dehiscence and pollen dispersal. II. The formation and the transfer mechanism of pollenkitt, cell-wall development of the loculus tissues and a function of orbicules in pollen dispersal. New Phytologist 105: 499–507.[CrossRef][Web of Science]
Kreunen, S. S., AND J. M. Osborn. 1999. Pollen and anther development in Nelumbo (Nelumbonaceae). American Journal of Botany 86: 1662–1676.
Les, D. H., E. L. Schneider, D. J. Padgett, P. S. Soltis, D. E. Soltis, AND M. Zanis. 1999. Phylogeny, classification, and floral evolution of water lilies (Nymphaeaceae; Nymphaeales): A synthesis of non-molecular, rbcL, matK, and 18S rDNA data. Systematic Botany 24: 28–46.[CrossRef][Web of Science]
Lohne, C., AND T. Borsch. 2005. Molecular evolution and phylogenetic utility of the petD group II intron: A case study in basal angiosperms. Molecular Biology and Evolution 22: 317–332.
Lohne, C., T. Borsch, AND J. H. Wiersema. 2007. Phylogenetic analysis of Nymphaeales using fast-evolving and noncoding chloroplast markers. Botanical Journal of the Linnean Society 154: 141–163.[CrossRef][Web of Science]
Mackey, A. P. 1996. Cabomba (Cabomba spp. ) in Queensland. Pet Status Review Series, Land Protection Branch, Department of Natural Resources and Mines, Coorparoo, Queensland, Australia.
Orgaard, M. 1991. The genus Cabomba (Cabombaceae): A taxonomic study. Nordic Journal of Botany 11: 179–203.[CrossRef][Web of Science]
Osborn, J. M. 2000. Pollen morphology and ultrastructure of gymnospermous anthophytes. In M. M. Harley, C. M. Morton, and S. Blackmore [eds.], Pollen and spores: Morphology and biology, 163–185. Royal Botanic Gardens, Kew, UK.
Osborn, J. M., J. A. Schwartz, B. L. Gutman, N. A. Melrose, A. M. Ingraham, M. L. Taylor, J. N. Strandquist, P. J. Hudson, J. M. Miesner, A. N. Schoenekase, T. Thiemann, AND A. S. Doores. 2004. Pollen ontogeny in the Nymphaeales. Polen 14: 20–21 (Abstract).
Osborn, J. M., AND T. N. Taylor. 1995. Pollen morphology and ultrastructure of the Bennettitales: In situ pollen of Cycadeoidea. American Journal of Botany 82: 1074–1081.[CrossRef][Web of Science]
Osborn, J. M., T. N. Taylor, AND M. R. de Lima. 1993. The ultrastructure of fossil ephedroid pollen with gnetalean affinities from the Lower Cretaceous of Brazil. Review of Palaeobotany and Palynology 77: 171–184.[CrossRef][Web of Science]
Osborn, J. M., T. N. Taylor, AND E. L. Schneider. 1991. Pollen morphology and ultrastructure of the Cabombaceae: Correlations with pollination biology. American Journal of Botany 78: 1367–1378.[CrossRef][Web of Science]
Pacini, E. 1997. Tapetum character states: Analytical keys for tapetum types and activities. Canadian Journal of Botany 75: 1448–1459.[CrossRef]
Pacini, E., AND G. G. Franchi. 1994. Role of the tapetum in pollen and spore dispersal. Plant Systematics and Evolution 7 ( Supplement): 1–11.
Pacini, E., G. G. Franchi, AND M. Hesse. 1985. The tapetum: Its form, function, and possible phylogeny in Embryophyta. Plant Systematics and Evolution 149: 155–185.[CrossRef][Web of Science]
Padmanabhan, D., AND M. V. Ramji. 1966. Developmental studies on Cabomba caroliniana Gray. II. Floral anatomy and microsporogenesis. Proceedings of the Indian Academy of Sciences, section B 64: 216–223.[Web of Science]
Podoplelova, Y., AND G. Ryzhakov. 2005. Phylogenetic analysis of the order Nymphaeales based on the nucleotide sequences of the chloroplast ITS2-4 region. Plant Science 169: 606–611.[CrossRef][Web of Science]
Qiu, Y.-L., J. Lee, F. Bernasconi-Quadroni, D. E. Soltis, P. S. Soltis, M. Zanis, E. A. Zimmer, Z. Chen, V. Savolainen, AND M. W. Chase. 1999. The earliest angiosperms: Evidence from mitochondrial, plastid, and nuclear genomes. Nature 402: 404–407.[CrossRef]
Qiu, Y.-L., L. B. Li, T. A. Hendry, R. Q. Li, D. W. Taylor, M. J. Issa, A. J. Ronen, M. L. Vekaria, AND A. M. White. 2006. Reconstructing the basal angiosperm phylogeny: Evaluating information content of mitochondrial genes. Taxon 55: 837–856.[Web of Science]
Roland, F. 1965. Précisions sur la structure et l'ultrastructure d'une tétrad calymée. Pollen et Spores 7: 5–8.
Roland, F. 1968. Etude de l'ultrastructure des apertures. II. Pollens a sillons. Pollen et Spores 10: 479–519.
Rowley, J. R., N. I. Gabarayeva, AND B. Walles. 1992. Cyclic invasion of tapetal cells into loculi during microspore development in Nymphaea colorata (Nymphaeaceae). American Journal of Botany 79: 801–808.[CrossRef][Web of Science]
Saarela, J. M., H. S. Rai, J. A. Doyle, P. K. Endress, S. Mathews, A. D. Marchant, B. G. Briggs, AND S. W. Graham. 2007. Hydatellaceae identified as a new branch near the base of the angiosperm phylogenetic tree. Nature 446: 312–315.[CrossRef][Medline]
Sampson, F. B. 1970. Unusual features of cytokinesis in meiosis of pollen mother cells of Pseudowintera traversii (Buchan.) Dandy (Winteraceae). Beitrage zur Biologie der Pflanzen 47: 71–77.
Saunders, R. M. K. 2000. Monograph of Schisandra (Schisandraceae). Systematic Botany Monographs 58: 1–118.
Schnarf, K. 1931. Vergleichende Embryologie der Angiospermen. Borntraeger, Berlin, Germany.
Schneider, E. L., AND J. M. Jeter. 1982. Morphological studies of the Nymphaeaceae. 12. The floral biology of Cabomba caroliniana. American Journal of Botany 69: 1410–1419.[CrossRef][Web of Science]
Snigerevskaya, N. C. 1955. The morphology of the pollen "Nymphaeales." Botanicheskii Zhurnal 40: 108–115.
Soltis, D. E., P. S. Soltis, M. W. Chase, M. E. Mort, D. C. Albach, M. Zanis, V. Savolainen, W. H. Hahn, S. B. Hoot, M. F. Fay, M. Axtell, S. M. Swenson, L. M. Prince, W. J. Kress, K. C. Nixon, AND J. S. Farris. 2000. Angiosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequences. Botanical Journal of the Linnean Society 133: 381–461.[CrossRef][Web of Science]
Soltis, P. S., AND D. E. Soltis. 2004. The origin and diversification of angiosperms. American Journal of Botany 91: 1614–1626.
Takahashi, M. 1992. Development of spinous exine in Nuphar japonicum De Candolle (Nymphaeaceae). Review of Palaeobotany and Palynology 75: 317–322.[CrossRef][Web of Science]
Takahashi, M. 1994. Exine development in Illicium religiosum Sieb. et Zucc. (Illiciaceae). Grana 33: 309–312.[CrossRef][Web of Science]
Taylor, M. L., AND J. M. Osborn. 2006. Pollen ontogeny in Brasenia (Cabombaceae, Nymphaeales). American Journal of Botany 93: 344–356.
Tobe, H., T. Jaffré, AND P. H. Raven. 2000. Embryology of Amborella (Amborellaceae): Descriptions and polarity of character states. Journal of Plant Research 113: 271–280.[CrossRef][Web of Science]
Ueno, J. 1962. On the fine structure of the pollen walls of angiospermae. II. Victoria. Journal of Biology. Osaka City University 13: 99–104.
Ueno, J., AND S. Kitaguchi. 1961. On the fine structure of the pollen walls of angiospermae. I. Nymphaeaceae. Journal of Biology. Osaka City University 12: 83–90.
Walker, J. W. 1974. Aperture evolution in the pollen of primitive angiosperms. American Journal of Botany 61: 1112–1137.[CrossRef][Web of Science]
Walker, J. W. 1976. Evolutionary significance of the exine in the pollen of primitive angiosperms. In I. K. Ferguson, and J. Muller [eds.], The evolutionary significance of the exine, 251–308. Academic Press, London, UK.
Williamson, P. S., AND E. L. Schneider. 1993. Cabombaceae. In. K. Kubitzki, J. G. Rohwer, and V. Bittrich [eds.], The families and genera of vascular plants, vol 2. Flowering plants: Dicotyledons. Magnoliid, hamamelid and caryophyllid families, 157–161. Springer-Verlag, Berlin, Germany.
Wilson, C. E., S. J. Darbyshire, AND R. Jones. 2007. The biology of invasive alien plants in Canada. 7. Cabomba caroliniana A. Gray. Canadian Journal of Plant Science 87: 615–638.[Web of Science]
Wodehouse, R. P. 1932. Tertiary pollen. I. Pollen of the living representatives of the Green River flora. Bulletin of the Torrey Botanical Club 59: 313–340.[CrossRef]
Wunderlich, R. 1954. Uber das Antherentapetum mit besonderer Berucksichtigung seiner Kernzahl. Österreichische Botanische Zeitschrift 101: 1–63.[CrossRef]
Yakovlev, M. S. 1981. Evolution of sexual structures in higher plants (main types of meiosis). Acta Societatis Botanicorum Poloniae 50: 333–338.[Web of Science]
Zavada, M. 1984. Pollen wall development of Austrobaileya maculata. Botanical Gazette (Chicago, Ill.) 145: 11–21.[CrossRef]
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
