| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Anatomy and Morphology |
2Jodrell Laboratory, Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3AB, UK; 3Seed Conservation Department, Royal Botanic Gardens, Kew, Wakehurst Place, Ardingly, West Sussex RH17 6TN, UK; 4Department of Biology, University of Missouri– St. Louis, One University Boulevard, St Louis, Missouri 63121 USA; 5Royal Botanic Gardens, Sydney, NSW 2000, Australia
Received for publication February 18, 2005. Accepted for publication June 15, 2005.
ABSTRACT
Despite much recent activity in the phylogeny and developmental genetics of grasses, the enigmatic homologies of their reproductive structures remain largely unresolved, partly because their highly derived morphology has resulted in a unique associated terminology. Outstanding questions include whether grass lodicules and stamens are derived from a single perianth or stamen whorl, respectively, whether the grass caryopsis is homologous with a nut, and how the scutellum evolved. We investigated the reproductive structures of the putative sister group of grasses, the southwestern Australian family Ecdeiocoleaceae, which includes two genera, Ecdeiocolea and Georgeantha. The zygomorphic arrangement of the four (rather than six) stamens in male flowers of Ecdeiocolea indicates that they may represent three outer stamens plus the adaxial inner stamen. Within Ecdeiocoleaceae, characters such as the highly unusual nucellus structure of Ecdeiocolea are autapomorphic. Sister-group comparisons indicate that some characteristic grass features, notably the scutellum, do not occur in their putative closest relatives and that more data are needed on early-diverging grass genera to resolve these issues. The grass caryopsis could represent one end of a transformation series embodied by the reduced gynoecial structure and indehiscent fruit of other Poales such as Flagellaria, Joinvillea, and Ecdeiocolea.
Key Words: caryopsis • Ecdeiocolea • flower • Georgeantha • grasses • monocots • Poaceae • scutellum
Poaceae (grasses) is one of the most species-rich flowering plant families and includes many economically important crops. Parallel evolution of such features as the annual habit, C4 photosynthesis and several highly characteristic reproductive structures has facilitated a series of major radiations within Poaceae, culminating in the existing global distribution of about 10 000 species and 700 genera (Clayton and Renvoize, 1986
; Linder and Rudall, 2005
). A phylogeny of Poaceae was recently established using a combination of multiple data sets from both molecules and morphology (GPWG, 2001
), enabling improved understanding of relationships between basal and derived grasses. Furthermore, the sequencing of the genome of rice (Oryza sativa L.) has accelerated the already widespread use of rice and other crop grasses as model organisms for a range of studies, including investigations of the grass flower and inflorescence (e.g., Kellogg, 2000
, 2002
).
Evolutionary changes in reproductive structure are of fundamental importance to theories about the early evolution and subsequent diversification of grasses. Grasses possess several unusual or unique reproductive features of often doubtful homology. For example, many grasses possess only three stamens (rather than six in two whorls of three, which is the most common monocot condition); these have been interpreted by morphologists as representing either the outer stamen whorl (e.g., Clifford, 1986
; Hackel, 1896
) (Fig. 1D) or elements of both inner and outer stamen whorls (Coccuci and Anton, 1988
), the latter condition (Fig. 1E) requiring a zygomorphic floral interpretation. Ovary and fruit homologies are also problematic in grasses because it is not clear whether the ovary is monocarpellary or possesses an underlying tricarpellary condition (Philipson, 1985
), and whether the grass caryopsis (Fig. 1A, B) is homologous with other single-seeded, dry and indehiscent fruits within the Poales (e.g., Cyperaceae, Ecdeiocolea, Restionaceae). Even the embryos of grasses are considered unique, because they are highly complex, possessing a leaflike coleoptile, a coleorhiza, up to six leaf primordia, and a characteristic prominent outgrowth termed a scutellum that has not been recorded in other families and has been hypothesized to represent a modified cotyledon (Hackel, 1896
; Sargant and Robertson, 1905
; Reeder, 1953
; Shah and Sreekumari, 1980
; Negbi, 1984
).
|
). Recently, three combined molecular analyses have identified Ecdeiocoleaceae as the sister family to Poaceae (Briggs et al., 2000
; Bremer, 2002
; Michelangeli et al., 2003
), thereby increasing the potential significance of this hitherto relatively obscure family in assessing the homologies of grass structures.
Ecdeiocoleaceae consist of two monotypic genera, Ecdeiocolea monostachya F. Muell. and Georgeantha hexandra B. G. Briggs & L. A. S. Johnson, both restricted to nutrient-poor soils in southwestern Australia (Linder et al., 1998a
). Morphologically, Ecdeiocolea and Georgeantha are alike in many respects but differ in inflorescence structure and some aspects of flower and fruit morphology (Briggs and Johnson, 1998a
; Meney and Pate, 1999
). Georgeantha possesses the "normal" complement of six stamens compared with four in Ecdeiocolea, and the fruit is a loculicidally dehiscent capsule in Georgeantha and an indehiscent, dry fruit in Ecdeiocolea. A molecular analysis using combined data from rbcL and trnL-trnF (Briggs et al., 2000
) grouped Ecdeiocolea and Georgeantha as a sister pair in a well-supported topology as sister to Poaceae, though the tree was poorly sampled for other Poales. Two subsequent phylogenetic analyses were more widely sampled for species of Poales; one used combined data from plastid rbcL and atpB (Bremer, 2002
), the other from morphology, inversions in the plastid genome, and sequence data from plastid rbcL and mitochondrial atpA (Michelangeli et al., 2003
). Both analyses found weak support for a sister-group relationship between Poaceae and Ecdeiocoleaceae, and Bremer (2002)
also grouped Ecdeiocolea and Georgeantha as a sister pair. Thus, Ecdeiocoleaceae belong in the graminid-restiid clade (nine families) within the order Poales (20 families). Other graminid-restiids include two large families, Poaceae and Restionaceae; a small family, Centrolepidaceae (three genera); and three monogeneric families, Anarthriaceae sensu stricto, Flagellariaceae, and Joinvilleaceae. Two other genera, Hopkinsia and Lyginia, also belong in this clade, but have been various placed either in Restionaceae (Linder et al., 1998b
) or in their own families Hopkinsiaceae and Lyginiaceae (Briggs and Johnson, 1998b
); there is now robust support for their affinity with Anarthria (Briggs et al., 2000
). The graminid-restiid clade is present in all analyses and classifications since Dahlgren et al. (1985)
(e.g., Bremer, 2002
; Michelangeli et al., 2003
; Davis et al., 2004
; reviewed by Linder and Rudall, 2005
).
The phylogeny of Poaceae by the Grass Phylogeny Working Group (GPWG, 2001
) allows comparison of reproductive structures of Ecdeiocoleaceae with those of putatively early-diverging grass genera such as Anomochloa, Pharus, and Streptochaeta (e.g., Arber, 1929
; Judziewicz and Soderstrom, 1989
). Detailed micromorphological studies of the flower and fruit were not hitherto available for Ecdeiocoleaceae. We have examined flowers and fruits of Ecdeiocolea and Georgeantha to explore the enigmatic, yet crucial, homologies of the highly derived reproductive structures of grasses. Homologies of unique grass features are unlikely to be directly resolved by a broad sister-group comparison, because comparative observations of early-diverging grass lineages are required. Such observations will be the subject of future work. Thus, in this paper we aim to (1) reconstruct changes in reproductive characteristics in order to suggest the condition in ancestral graminids from which modern grasses have evolved and (2) establish which grass characters are truly synapomorphic of grasses and which are higher-level synapomorphies within graminid Poales.
MATERIALS AND METHODS
Flowers and fruits of Ecdeiocolea monostachya and Georgeantha hexandra were collected from a locality c. 35 km east of Jurien and from several sites c. 25 km north of Eneabba, Western Australia, in August and November 2003 and September 2004. Both species occur at both localities. Field visits in 2003 coincided respectively with the start of flowering and with fruiting of Ecdeiocolea, but with the end of flowering and after fruit dispersal in Georgeantha. Voucher collections have been placed in the National Herbarium of New South Wales (NSW).
Material was fixed in the field in formalin acetic alcohol (FAA: 70% alcohol, formaldehyde and glacial acetic acid, 85 : 10 : 5), and transferred to 70% ethanol. For scanning electron microscopy (SEM), some buds and flowers were dissected in 70% ethanol, then dehydrated through absolute ethanol, and critical-point-dried using a Balzers CPD 020. They were dissected and mounted onto specimen stubs using double-sided sellotape, coated with platinum using an Emitech K550 sputter coater, and examined using a Hitachi cold field emission SEM S-4700-II at the Royal Botanic Gardens, Kew, UK. Other specimens were rehydrated and fixed using the OTOTO method of Murphy (1978)
. This treatment maximizes electron conductivity of the tissues without gold sputter-coating. The treatment consisted of overnight fixation in 1% aqueous OsO4, six washes with deionized water, 30 min in fresh 1% thiocarbohydrazide (TCH), six deionized water washes for 1 h in 1% aqueous OsO4, six deionized water washes for 30 min in fresh 1% TCH, six deionized water washes for 1 h in 1% aqueous OsO4, and six final deionized water washes. The specimens were then dehydrated using an alcohol series (15, 30, 50, 70, 80, 90, 95, 100, 100%), critical-point dried in an SPI Jumbo critical point drier (Structure Probe, West Chester, Pennsylvania, USA) or Balzers CPD 020, and imaged in a Hitachi S450 SEM at University of Missouri, St. Louis, USA at 20 kV or a Philips XL20 SEM at Murdoch University, Australia.
For light microscope (LM) observations, material was embedded either in wax or in resin prior to sectioning. Fixed flowers and buds were dehydrated in an ethanol series to absolute ethanol. For wax sectioning, material was embedded using standard methods. For resin embedding the sample was transferred through an absolute ethanol-LR white resin series to absolute resin, and kept at 4°C for about a week, with daily changes of resin. Specimens were then moved to gelatin capsules and polymerized between 58–62°C at 600 mbar pressure for about 21 h. Specimens were sectioned using a Leica microtome. Sections were stained in toluidine blue and mounted in DPX mountant (a mixture of distyrene, a plasticizer, and xylene; Agar Scientific, UK). Light micrographs were taken at the Royal Botanic Gardens, Kew, UK, using a Leitz Diaplan photomicroscope fitted with a digital camera.
RESULTS
Flower
Flowers of Ecdeiocolea monostachya possess six tepals in two whorls of three. During development, the adaxial tepal of the inner whorl overlaps the abaxial tepal of the outer whorl at their margins. There are four stamens in male flowers (Figs. 2–5, 23, 28) and four staminodes in female flowers (Figs. 6–11, 19–21), each served by a single vascular bundle. Each of the four stamens or staminodes is positioned in the same floral sectors as one of the three outer tepals and the adaxial inner tepal (Figs. 8–10, 20), indicating that the two inner abaxial stamens or staminodes are absent, though no vestige of the "missing" organs is evident in mature flowers. In male flowers, a minute central pistillode is sometimes present (Fig. 29), served by two vascular traces. In female flowers, a central gynoecium consisting of two carpels (Figs. 12–18), each with a single elongated stigma that extends along the innermost edge of the stylar branch, and a single apically attached ovule (Fig. 30). There is no vestige of a third carpel, even during early development (Figs. 19, 20). Each of the two stylar branches possesses a single vascular bundle and bears a feathery stigma (Fig. 14). The fused part of the style is very short; it contains two vascular bundles and a single central stylar canal with a ring of epithelial cells that are tightly pressed together, giving the impression of a closed style (Fig. 15).
|
|
|
|
|
In Georgeantha hexandra, male flowers (Figs. 31–33) possess six stamens and a central small pistillode. Female flowers (Figs. 34–37) possess six staminodes in two distinct whorls and a central gynoecium consisting of three carpels, each with a single elongated and highly branched stigmatic surface that extends along the innermost edge of the stylar branch. There is a groove running down the middle region of each carpel, from the base of the ovary to the middle of the style (Figs. 36, 37); this groove does not represent the carpel margin as it lies directly above the main carpellary vascular bundle (and therefore is not a septal groove), but corresponds to the line of dehiscence of the mature capsular fruit. Each of the three stylar branches possesses a single vascular bundle and bears a feathery stigma. The fused part of the style contains two vascular bundles and a single central tripartite stylar canal.
|
). We term this structure nucellar transfusion tissue, because it is clearly distinct from the large region of nucellar tissue closer to the chalaza. The semi-mature seeds that were available for this study possess a highly unusual spherical mass of tissue at the micropylar end surrounding the three-celled proembryo (Figs. 41, 43, 44). Mature seeds contain a small, undifferentiated, depressed-ovoid embryo that lies flat, like a cap, against the starchy storage tissue (Fig. 42).
|
). However, in Ecdeiocolea the mature seed possesses a well-developed seed coat, which is perfectly free from the pericarp, rendering the fruit an achene rather than a caryopsis. Only one of the two ovules develops into a mature seed; the other normally aborts. The fruit of Georgeantha is a loculicidal capsule that develops from a tricarpellate gynoecium (Fig. 47). It usually produces a single seed, though capsules with two developed, dehiscent carpels and two seeds are not uncommon. Seeds of both genera of Ecdeiocoleaceae are globose (Figs. 48, 50), slightly pointed at the micropylar end (in Ecdeiocolea also at the chalazal end), and develop from bitegmic ovules. Both integuments are two layers thick and produce an exotestal seed coat.
|
, referring to the convex outer wall of the exotesta cells with their conspicuous secondary microsculpture of reticulate-undulate folds on the outer wall, most probably formed by the cuticle (Fig. 51). DISCUSSION
Flower organization
The grass flower is typically subtended by a bract-like structure, the palea, that surrounds three (or more commonly two) lodicules. The lodicules are normally interpreted as homologous with a single perianth whorl (probably the inner tepals) of other monocots, based on morphological, ontogenetic, and developmental genetic evidence (Arber, 1934
; Clifford, 1986
; Cocucci and Anton, 1988
; Schmidt and Ambrose, 1998
). Among early-diverging grasses, flowers of Pharus possess three lodicules (present only in male flowers), and flowers of Anomochloa lack lodicules entirely; instead they possess a hairy zone that Arber (1929)
interpreted as modified lodicules. Some other grasses, including Streptochaeta, possess structures that have been interpreted as six tepals by some authors (e.g., Arber, 1929
); possession of six tepals, as in Georgeantha and Ecdeiocolea, is the most common monocot condition and probably the plesiomorphic state in Poaceae. Loss of the outer tepal whorl presumably occurred at a relatively early stage in grass evolution, followed by loss of a further tepal associated with floral zygomorphy.
Within Ecdeiocoleaceae, Georgeantha possesses six stamens in male flowers (or staminodes in female flowers), but Ecdeiocolea has only four, in a bilaterally symmetric (zygomorphic) arrangement. In stamen number, Ecdeiocolea therefore resembles the early-diverging grass Anomochloa, which also possesses four stamens. However, there are two contrasting hypotheses of floral morphology in Anomochloa (summarized by Rudall and Bateman, 2004
): either (1) three outer stamens and the adaxial inner stamen are present (Arber, 1929
; Judziewicz and Soderstrom, 1989
) or (2) three inner stamens and the abaxial outer stamen are present (Cocucci and Anton, 1988
). Both of these models deviate from the two most iterative patterns of floral zygomorphy in monocots, in which either the abaxial or adaxial stamens are present rather than elements of both. Early floral development of Ecdeiocolea is like other monocots in that three adaxial stamens are present; the fourth (abaxial) one develops noticeably later. Thus, the morphology of male flowers of Ecdeiocolea fits the first model for Anomochloa more closely than the second. Further studies will address the developmental morphology of flowers of some early-diverging grasses.
Ovary
As in other monocots, a tricarpellary ovary is the plesiomorphic condition in Poales. However, reduction in carpel number has occurred several times, either by congenital loss or pseudomonomery. In the pseudomonomerous condition, all but one of the carpels in a syncarpous gynoecium abort at an early stage of development, though the sterile carpels normally remain discernible even at anthesis. Pseudomonomery is common in commelinids (Eames, 1961
; Uhl, 1972
; Uhl and Dransfield, 1987
; Lewis and Doyle, 2002
; Strange et al., 2004
), but less common within Poales.
In Poaceae, the gynoecium normally consists of a single carpel containing a single ovule that is located adaxially. Thus, the homologies of grass ovaries are problematic because it is not clear whether the two "missing" carpels have been entirely suppressed to the extent that only one carpel primordium is initiated, or whether they are entirely merged with the single fertile carpel. There is a single gynoecial primordium, but many grasses possess two styles, and some possess three styles, including some bamboos and also Puelioideae, which are sister to most of the grasses. There is typically a single style and single vascular trace per carpel; this led several authors to infer an underlying tricarpellary condition for grasses (Arber, 1934
; Philipson, 1985
; Cocucci and Anton, 1988
), termed pseudomonocarpellary by Philipson (1985)
. This condition, arguably an extreme type of pseudomonomery with no remnants of the aborted carpels, is unusual in monocots other than Poales, though a monocarpellary condition occurs in most Stemonaceae (Tomlinson and Ayensu, 1968
), a family not closely related to grasses. Within Restionaceae, which is closely related to Poaceae, the ovary is clearly tricarpellary. However, in most Restionaceae one or two of the carpels are aborted during development, though in many species their vasculature is retained and two or three styles are present (Linder, 1992
; Ronse DeCraene et al., 2002
). The unilocular gynoecia of some other Poales, such as Sparganium and Cyperaceae, have also sometimes been interpreted as pseudomonomerous (Eames, 1961
; Ronse DeCraene et al., 2002
). In Ecdeiocolea, the slight gynoecial zygomorphy evident in transverse sections of flowers, together with the position of the stamens, indicate that the abaxial carpel has been entirely suppressed, because there is no trace of residual vasculature.
Many grasses possess solid styles with a dense, often extensive, central transmitting tissue through which pollen tubes grow intracellularly (e.g., Arber, 1934
; Li and You, 1991
), though the phylogenetic distribution of this character requires review. This organization contrasts with the hollow stylar condition that occurs in most monocots, including both species of Ecdeiocoleaceae, which possess a stylar canal surrounded by a distinct layer of epidermal cells. In hollow styles, pollen tubes grow from the stigma to the ovary along the surface of the canal, normally through a thin layer of mucilage. Solid styles are rare in monocots though they represent the most common condition in eudicots; among other monocots, they occur in a few families that are not closely related to grasses, such as Alliaceae and Orchidaceae (reviewed by Rudall et al., 2002
). Thus, this condition presumably evolved de novo in Poaceae, though it also occurs in a few other graminid Poales, such as Flagellaria (P. Rudall, personal observation). Solid styles may in some cases be correlated with apomixis, which is common in grasses, because apomictic development that is triggered by pollination relies on rapid movement of chemical signals through a cellular matrix.
Fruit
With very few exceptions, the fruit in grasses is a caryopsis (an indehiscent one-seeded fruit in which the testa and pericarp are fused). This structure is clearly derived from the monocarpellary uniovulate ovary. The grass pericarp is very thin, consisting of only a few cell layers: a thick-walled outer epidermis, a few parenchymatous cell layers with vascular bundles, and an inconspicuous inner epidermis (Hackel, 1896
).
The two genera that constitute the family Ecdeiocoleaceae present strongly contrasting ovary/fruit conditions. Georgeantha has a tricarpellary ovary and a dehiscent capsular fruit, (the most common monocot condition), whereas Ecdeiocolea invariably possesses only two carpels, of which only one develops a seed within an indehiscent dry, nut-like fruit (Briggs and Johnson, 1998a
). The pericarp of this fruit is contiguous to the seed coat; thus it conforms to the definition of an "achene" (Spjut, 1994
). The pronounced groove in each carpel of the ovary of Georgeantha (Figs. 36, 37) marks the line of fruit dehiscence, similar to that of other Poales such as the Australian genus Anarthria (six species), which also has a tricarpellary ovary and capsular fruit. Such a dehiscence line is not present in ovaries of Ecdeiocolea or grasses, which produce an indehiscent fruit.
In their morphological analysis, Michelangeli et al. (2003)
scored the grass caryopsis as a nutlet (a dry indehiscent fruit with a woody pericarp), and thus as homologous with the nut of Ecdeiocolea, but they acknowledged that this homology required further testing. The term "nutlet" is strictly applied to a fruitlet (i.e., only part of an apocarpous or schizocarpous gynoecium); thus, we here prefer to use the term "achene" for this structure. If Ecdeiocoleaceae are sister to Poaceae, the possession of an indehiscent, single-seeded fruit could be a synapomorphy for Ecdeiocolea and Poaceae, as suggested by Michelangeli et al., (2003)
, though this would require an atavistic reversal to a capsule in Georgeantha, or perhaps a polymorphic common ancestor. (The difference between an achene and a caryopsis is not morphologically robust, since the caryopsis of the grasses could be interpreted as an achene in which the seed coat has undergone further reduction). Two other putatively closely related graminids, Joinvillea and Flagellaria, also possess indehiscent fruits, so this scenario appears most plausible (Fig. 52). In Joinvillea, the fruits are indehiscent with no carpel reduction, resulting in fleshy, three-seeded berries with dried fruit surface ornamentation rather like that of Ecdeiocolea (Fig. 49) (B. Briggs, unpublished data). In Flagellaria, the fruit is a fleshy drupe in which the seed coat is firmly fused with the fruit wall (Tillich, 1996
; W. Stuppy, personal observation).
|
). Furthermore, indehiscent unilocular fruits occur in Hopkinsia, the only indehiscent genus of Anarthriaceae, and in various Restionaceae, though the unilocular, one-seeded fruits of Centrolepidaceae are usually dehiscent. In some Restionaceae, the perianth persists around the nut, often forming a wing that aids wind dispersal (Linder, 1991
). We recommend that in a morphological analysis, fruit type should be represented by at least two characters: (1) dehiscent vs. indehiscent (the latter possibly a graminid feature, with the exception of Georgeantha: Fig. 52) and (2) fleshy vs. not fleshy. Fusion of the seed coat and fruit wall may also be a significant feature of the graminid lineage, but more comparative data are required; this condition characterizes Hopkinsia. Differentiation of a hard, indehiscent pericarp or endocarp is often correlated with reduction in ovule (seed) number per ovary, usually down to a single ovule, as in many graminids.
In indehiscent fruits, the role of mechanical protection is generally transferred from the seed coat to the pericarp or endocarp, so that the seed coat itself is usually relatively undifferentiated, lacking mechanical elements, and eventually becomes obliterated, as in Poaceae. In the grass caryopsis (e.g., in Triticum), the inner layer of the inner integument becomes compressed during development; the outer layer of the inner integument is crushed into a hyaline membrane covered with a cuticle, and the outer integument disintegrates completely (Esau, 1965
). Corner (1976)
believed that the degree of differentiation of a seed coat is an indicator of how recently the character transformation from "fruit dehiscent" to "fruit indehiscent" has occurred (a fundamentally gradualistic perspective). If so, the presence in Ecdeiocolea of a well-differentiated seed coat with a strong mechanical layer would indicate that the change from dehiscent to indehiscent fruits is a phylogenetically recent event. Conversely, the rudimentary nature of the grass seed coat could equally indicate that the change from capsular fruits (the probable plesiomorphic condition in Poales) to indehiscent fruits could have occurred deep in the phylogenetic history of the order. The most likely conclusion is that this transition has occurred multiple times.
Ovule
The ovule in Ecdeiocoleaceae is tenuinucellate, as in many other Poales, including the xyrids Eriocaulaceae, Mayacaceae, and Xyridaceae, and the graminids Anarthriaceae, Centrolepidaceae, and most Restionaceae (reviewed by Rudall, 1997
); both tenuinucellate and crassinucellate types are reported in Poaceae (Aulbach-Smith and Herr, 1984
), though the crassinucellate records require further review. In other respects, the structure of both the ovule and megagametophyte in Ecdeiocolea monostachya are highly unusual (Rudall, 1990
). Embryo sac development is apparently of a tetrasporic type, with a 1 + 3 arrangement of the four nuclei of the coenomegaspore, possibly reflecting a more general trend of suppression of cell division during megasporogenesis in Poales (Rudall and Linder, 1988
; Linder and Rudall, 1993
). The role and ultimate fate of highly unusual nucellar transfusion tissue of Ecdeiocolea remain equivocal; the spherical mass of tissue at the micropylar end of the developing seed could well represent a later stage, since it resembles it in shape and position. If so, it presumably gives rise to the storage tissue of the mature seed, which would then be a perisperm. Perisperm has infrequently been conclusively demonstrated in monocots, but may well be more common than reported, especially since a massive nucellus is common (Rudall, 1997
). Alternatively, the spherical mass in the seed of Ecdeiocolea could be developing endosperm; it is sharply delimited against the surrounding nucellar tissue and even appears to resorb it (Fig. 43). Moreover, it lies immediately adjacent to the embryo with no tissue separating the two that could be interpreted as a rudimentary endosperm (Fig. 44), as can be observed in developing seeds of other commelinids, such as Hydatellaceae (Hamann, 1975
), in which the persistent nucellus forms the sole storage tissue (perisperm) in the mature seed. However, the latter possibility appears less likely; if this tissue is endosperm it is already fully cellularized at a remarkably early stage, assuming that endosperm formation in Ecdeiocoleaceae is nuclear as in most monocots (Hydatellaceae being a rare exception). In other Poales cellularization of the endosperm commences only when the embryo is already at the globular stage, i.e., consisting of eight or more cells, as in Centrolepidaceae (Hamann, 1975
) and Mayacaceae (Venturelli and Bouman, 1986
). More detailed studies of seed development in Ecdeiocolea are necessary to resolve this conundrum.
Embryo
Poaceae invariably possess highly differentiated embryos with a prominent outgrowth of the embryo termed the scutellum. The scutellum is a characteristic structure that occurs in all grasses, including early-diverging taxa such as Anomochloa (Judziewicz and Soderstrom, 1989
) and Streptochaeta (Reeder, 1953
). The scutellum is apparently unique to grasses; it has been homologized with various structures (reviewed by Reeder, 1953
), but is normally considered a modified cotyledon (Hackel, 1896
; Negbi, 1984
). Both scutellum and coleoptile are sometimes regarded as modified foliar organs, though this inference was tentatively contradicted by Scanlon and Freeling (1998)
based on investigation of narrow sheath mutants of maize. Neither Ecdeiocoleaceae nor any other graminids possess a structure that could be interpreted as the morphological precursor to a scutellum, so our investigation does not elucidate its evolutionary origin. Seeds of Ecdeiocoleaceae contain a small, undifferentiated, depressed-ovoid embryo that lies flat, like a cap, against the starchy storage tissue (probably endosperm; discussed earlier) at the micropylar end. Such small, undifferentiated embryos also occur in several other Poales, including Rapateaceae (Venturelli and Bouman, 1988
), some of the xyrid families (Eriocaulaceae, Hydatellaceae, Mayacaceae, and Xyridaceae), and in the graminids Centrolepidaceae, Flagellariaceae, Joinvilleaceae, and Restionaceae (Prakash, 1969
; Hamann, 1975
; Venturelli and Bouman, 1986
; Campbell and Kellogg, 1987
; Takhtajan, 1988
; Johri et al., 1992
). In these species, the mature embryo is not differentiated into cotyledon, hypocotyl, radicle, or shoot apex. Endospermous seeds with small (relative to the endosperm), underdeveloped embryos are generally considered plesiomorphic within angiosperms, whereas seeds with little or no endosperm and a large embryo relative to the endosperm are generally considered to be derived (e.g., Bessey, 1915
; Takhtajan, 1991
; Taylor and Hickey, 1996
). It is therefore plausible that the minute, discoid type of embryo is the plesiomorphic condition in Poales. By contrast, larger, more elongate and more differentiated embryo types occur in Bromeliaceae and Typhaceae (including both Typha and Sparganium), the cyperids Cyperaceae and Thurniaceae, and above all in Poaceae.
Conclusions
This investigation underlines the unique nature of some characteristic morphological features of grasses, especially the scutellum, and demonstrates the necessity for more detailed investigations of ovular and seed development in Poales. Future work will focus on related comparative studies in early-diverging grasses. Intensified studies of the evolutionary-developmental genetics of Poaceae should focus on carefully selected, phylogenetically relevant "model" taxa. Sister-group comparison helps to clarify the homologies of the grass caryopsis, which could represent one end of a transformation series embodied by the reduced gynoecial structure and indehiscent fruit of other Poales such as Flagellaria and Ecdeiocolea; flower and fruit morphology are integrally linked.
Similarly, more work is required on the developmental morphology of flowers of early-diverging grasses with reduced stamen number, such as Anomochloa and some species of Pharus, to conclusively determine their homologies and assess whether similar factors have operated in Ecdeiocolea. Heterochrony may have played a critical role in the evolution of reproductive structures, leading in some cases to complete organ suppression, as in several other monocot groups, particularly those with zygomorphic flowers such as Orchidaceae and Zingiberales (Rudall and Bateman, 2002
, 2004
). The genetic control of zygomorphy is not well understood. Zygomorphy in flowers of the asterid eudicot Antirrhinum relies on differential expression of two pairs of transcription factors, CYCLOIDEA (CYC) and DICHOTOMA (DICH), and RADIALIS (RAD) and DIVARICATA (DIV) in the young floral meristem (Coen and Nugent, 1994
; Corley et al., 2005
). The former two proteins are members of the TCP family, which also includes TEOSINTE BRANCHED1 (TB1) in Zea (Doebley et al., 1997
). RAD and DIV each contain a MYB domain, with RAD acting in a non-autonomous fashion (Corley et al., 2005
). Recent work suggests that CYC at least may be responding to a dorsiventral pattern that is established earlier in development (Clark and Coen, 2002
), and thus that there may be even more genes involved in the regulatory network than those identified to date. As the network of regulators becomes better defined and their interactions characterized in more detail, it will be of considerable interest to determine if similar regulatory networks influence zygomorphy in monocots. Reduced stamen and carpel number in Ecdeiocolea relative to its sister genus Georgeantha, together with their contrasting fruit morphologies, provide potentially useful models for critical evaluation of floral and fruit evolution in Poales.
FOOTNOTES
1 The authors thank Carolyn Porter for supplying SEMs of the seeds, Philip Ladd for access to microscopy and laboratory facilities and help in locating plant material, Allan and Lorraine Tinker for assisting BGB with fieldwork in Western Australia, and Richard Bateman and two anonymous reviewers for critically reading the manuscript. The Millennium Seed Bank Project at Wakehurst Place is funded by the UK Millennium Commission, the Wellcome Trust, and Orange PLC. 
6 Author for correspondence (e-mail: p.rudall{at}rbgkew.org.uk
)
LITERATURE CITED
Arber A. 1929 Studies in the Gramineae. VI. 1. Streptochaeta. 2. Anomochloa. 3. Ichnanthus. Annals of Botany 43: 35-53
Arber A. 1934 The Gramineae. Cambridge University Press, Cambridge, UK
Aulbach-Smith C. A. J. M. Herr 1984 Development of the ovule and female gametophyte in Eustachys petraea and E. glauca (Poaceae). American Journal of Botany 71: 427-438[CrossRef][Web of Science]
Bessey C. E. 1915 The phylogenetic taxonomy of flowering plants. Annals of the Missouri Botanical Garden 2: 108-164
Bremer K. 2002 Gondwanan evolution of the grass alliance of families (Poales). Evolution 56: 1374-1387[CrossRef][Web of Science][Medline]
Briggs B. G. L. A. S. Johnson 1998a Georgeantha hexandra, a new genus and species of Ecdeiocoleaceae (Poales) from Western Australia. Telopea 7: 307-312
Briggs B. G. L. A. S. Johnson 1998b A guide to a new classification of Restionaceae and allied families. In K. A. Meney and J. S. Pate [eds.], Australian rushes: biology, identification and conservation of Restionaceae and allied families, 25–56. University of Western Australia Press, Nedlands, Australia
Briggs B. G. A. D. Marchant S. Gilmore C. L. Porter 2000 A molecular phylogeny of Restionaceae and allies. In K. L. Wilson and D. A. Morrison [eds.], Monocots, systematics and evolution, 661–671. CSIRO, Melbourne, Australia
Campbell C. S. E. A. Kellogg 1987 Sister group relationships in the Poaceae. In T. R. Soderstrom, K. W. Hilu, C. S. Campbell, and M. E. Barkworth [eds.], Grass systematics and evolution, 217–224. Smithsonian Institution Press, Washington, D.C., USA
Chase M. W. M. F. Fay D. S. Devey O. Maurin N. Ronsted J. Davies Y. Pillon G. Petersen O. Seberg M. N. Tamura C. B. Asmussen K. Hilu T. Borsch J. I. Davis D. W. Stevenson J. C. Pires T. J. Givnish K. J. Sytsma M. M. McPherson S. W. Graham H. S. Rai 2005 Multigene analyses of monocot relationships: a summary. In J. T. Columbus, E. A. Friar, J. M. Porter, L. M. Prince, and M. G. Simpson [eds.], Monocots: comparative biology and evolution, 2 vols., in press. Rancho Santa Ana Botanic Garden, Claremont, California, USA
Clark J. I. E. S. Coen 2002 The CYCLOIDEA gene can respond to a common dorsoventral prepattern in Antirrhinum. Plant Journal 30: 639-648[CrossRef][Web of Science][Medline]
Clayton W. D. S. A. Renvoize 1986 Genera graminum: grasses of the world. Her Majesty's Stationery Office, London, UK
Clifford H. T. 1986 Spikelet and floral morphology. In T. R. Soderstrom, K. Hilu, C. S. Campbell, and M. E. Barkworth [eds.], Grass systematics and evolution, 21–30. Smithsonian Institution Press, Washington, D.C., USA
Cocucci A. E. A. M. Anton 1988 The grass flower: suggestions on its origin and evolution. Flora 181: 353-362[Web of Science]
Coen E. S. J. M. Nugent 1994 Evolution of flowers and inflorescences. Development (Supplement): 107–116
Corley S. B. R. Carpenter L. Copsey E. S. Coen 2005 Floral asymmetry involves an interplay between TCP and MYB transcription factors in Antirrhinum. Proceedings of the National Academy of Sciences, USA 102: 5068-5073
Corner E. J. H. 1976 The seeds of dicotyledons. Cambridge University Press, Cambridge, UK
Dahlgren R. M. T. H. T. Clifford P. F. Yeo 1985 The families of the monocotyledons. Springer-Verlag, Berlin, Germany
Davis J. I. D. W. Stevenson G. Petersen O. Seberg L. M. Campbell J. V. Freudenstein D. H. Goldman C. R. Hardy F. A. Michelangeli M. P. Simmons C. D. Specht F. Vergara-Silva M. A. Gandolfo 2004 A phylogeny of the monocots, as inferred from rbcL and atpA sequence variation, and a comparison of methods for calculating jackknife and bootstrap values. Systematic Botany 29: 460-510[CrossRef]
Doebley J. A. Stec L. Hubbard 1997 The evolution of apical dominance in maize. Nature 386: 485-488[CrossRef][Medline]
Eames A. J. 1961 Morphology of the angiosperms. McGraw-Hill, New York, New York, USA
Esau K. 1965 Plant anatomy, 2nd ed. John Wiley, New York, New York, USA
GPWG (Grass Phylogeny Working Group). 2001 Phylogeny and subfamilial classification of the grasses (Poaceae). Annals of the Missouri Botanical Garden 88: 373-457[CrossRef][Web of Science]
Hackel E. 1896 True grasses. F. Lamson-Scribner. Constable, London, UK
Hamann U. 1975 Neue Untersuchungen zur Embryologie und Systematik der Centrolepidaceae. Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie 96: 154-191
Johri B. M. K. B. Ambegaokar P. S. Srivastava 1992 Comparative embryology of angiosperms, 2 vols. Springer-Verlag, Berlin, Germany
Judziewicz E. J. T. R. Soderstrom 1989 Morphological, anatomical, and taxonomic studies in Anomochloa and Streptochaeta (Poaceae: Bambusoideae). Smithsonian Contributions to Botany 68: 1-52
Kellogg E. A. 2000 The grasses: a case-study in macroevolution. Annual Review of Ecology and Systematics 31: 217-238[CrossRef][Web of Science]
Kellogg E. A. 2002 Are macroevolution and microevolution qualitatively different? Evidence from Poaceae and other families. In Q. C. B. Cronk, R. M. Bateman, and J. A. Hawkins [eds.], Developmental genetics and plant evolution, 70–84. Taylor and Francis, London, UK
Lewis C. E. J. J. Doyle 2002 A phylogenetic analysis of tribe Areceae (Arecaceae) using two low-copy nuclear genes. Plant Systematics and Evolution 236: 1-17[CrossRef][Web of Science]
Li B. L. R. L. You 1991 Structure and development of stigmatic branches and style and their relation to pollen tube growth in wheat. Acta Botanica Sinica 33: 712-717
Linder H. P. 1991 A review of the southern African Restionaceae. Contributions from the Bolus Herbarium 13: 209-264
Linder H. P. 1992 The gynoecia of Australian Restionaceae: morphology, anatomy and systematic implications. Australian Systematic Botany 5: 227-245
Linder H. P. B. G. Briggs L. A. S. Johnson 1998a Ecdeiocoleaceae. In K. Kubitzki [ed.], Families and genera of vascular plants, vol. IV, Monocotyledons, Alismatanae and Commelinanae (except Gramineae), 195–197. Springer-Verlag, Berlin, Germany
Linder H. P. B. G. Briggs L. A. S. Johnson 1998b Restionaceae. In K. Kubitzki [ed.], Families and genera of vascular plants, vol. IV, Monocotyledons, Alismatanae and Commelinanae (except Gramineae), 425–445. Springer-Verlag, Berlin, Germany
Linder H. P. P. J. Rudall 1993 The megagametophyte in Anarthria and its relatives: systematic implications. American Journal of Botany 80: 1455-1464[CrossRef][Web of Science]
Linder H. P. P. J. Rudall 2005 The evolutionary history of Poales. Annual Reviews in Ecology and Systematics 36: in press
Meney K. A. J. S. Pate [eds.] 1999 Australian rushes. Biology, identification and conservation of Restionaceae and allied families. University of Western Australia Press, Nedlands, Australia
Michelangeli F. A. J. I. Davis D. W. Stevenson 2003 Phylogenetic relationships among Poaceae and related families as inferred from morphology, inversions in the plastid genome, and sequence data from the mitochondrial and plastid genomes. American Journal of Botany 90: 93-106
Murphy J. A. 1978 Non-coating techniques to render biological specimens conductive. Scanning Electron Microscopy 2: 115-194
Negbi M. 1984 The structure and function of the scutellum of the Gramineae. Botanical Journal of the Linnean Society 88: 205-222
Philipson W. R. 1985 Is the grass gynoecium monocarpellary?. American Journal of Botany 72: 1954-1961[CrossRef][Web of Science]
Prakash N. 1969 The floral development and embryology of Centrolepis fascicularis. Phytomorphology 19: 285-291[Web of Science]
Reeder J. R. 1953 The embryo of Streptochaeta and its bearing on the homology of the coleoptile. American Journal of Botany 40: 77-80[CrossRef][Web of Science]
Ronse Decraene L. P. H. P. Linder E. F. Smets 2002 Ontogeny and evolution of the flowers of South African Restionaceae with special emphasis on the gynoecium. Plant Systematics and Evolution 231: 225-258[CrossRef][Web of Science]
Rudall P. J. 1990 Development of the ovule and megagametophyte in Ecdeiocolea monostachya. Australian Systematic Botany 3: 265-274[CrossRef]
Rudall P. J. 1997 The nucellus and chalaza in monocotyledons: structure and systematics. Botanical Review 63: 140-184
Rudall P. J. R. M. Bateman 2002 Roles of synorganisation, zygomorphy and heterotopy in floral evolution: the gynostemium and labellum of orchids and other lilioid monocots. Biological Reviews 77: 403-441[Medline]
Rudall P. J. R. M. Bateman 2004 Evolution of zygomorphy in monocot flowers: iterative patterns and developmental constraints (Tansley Review). New Phytologist 162: 25-44[CrossRef][Web of Science]
Rudall P. J. R. M. Bateman M. F. Fay A. Eastman 2002 Floral anatomy and systematics of Alliaceae with particular reference to Gilliesia, a presumed insect mimic with strongly zygomorphic flowers. American Journal of Botany 89: 1867-1883
Rudall P. J. H. P. Linder 1988 The megagametophyte and nucellus in Restionaceae and Flagellariaceae. American Journal of Botany 75: 1777-1786[CrossRef][Web of Science]
Rudall P. J. K. L. Stobart W. P. Hong J. G. Conran C. A. Furness G. Kite M. W. Chase 2000 Consider the lilies: systematics of Liliales. In K. L. Wilson and D. A. Morrison [eds.], Monocots: systematics and evolution, 347–359. CSIRO, Melbourne, Australia
Sargant E. A. Robertson 1905 The anatomy of the scutellum in Zea maïs. Annals of Botany 19: 115-122
Scanlon M. J. M. Freeling 1998 The narrow sheath leaf domain deletion: a genetic tool used to reveal developmental homologies among modified maize organs. Plant Journal 13: 547-561[Web of Science]
Schmidt R. R. B. A. Ambrose 1998 The blooming of grass flower development. Current Opinion in Plant Biology 1: 60-67[CrossRef][Web of Science][Medline]
Shah C. K. S. B. Sreekumari 1980 Developmental morphology and homologies in grass embryo. Current Science 49: 284-285[Web of Science]
Spjut R. W. 1994 A systematic treatment of fruit types. Memoirs of the New York Botanical Garden 70: 1-182
Strange A. P. J. Rudall C. J. Prychid 2004 Comparative floral anatomy of Pontederiaceae. Botanical Journal of the Linnean Society 144: 395-408[CrossRef]
Taylor D. W. L. J. Hickey 1996 Flowering plant origin, evolution and phylogeny. Chapman and Hall, New York, New York, USA
Takhtajan A. [ed.] 1988 Anatomia seminum comparativav. 2. Nauka, Leningrad, Russia
Takhtajan A. 1991 Evolutionary trends in flowering plants. Columbia University Press, New York, New York, USA
Tillich H. J. 1996 Seeds and seedlings in Hanguanaceae and Flagellariaceae (Monocotyledons). Sendtnera 3: 187-197
Tomlinson P. B. E. S. Ayensu 1968 Morphology and anatomy of Croomia pauciflora (Stemonaceae). Journal of the Arnold Arboretum 49: 260-277[Web of Science]
Uhl N. W. 1972 Inflorescence and flower structure in Nypa fruticans (Palmae). American Journal of Botany 59: 729-743[CrossRef][Web of Science]
Uhl N. W. J. Dransfield 1987 Genera palmarum. Allen Press, Lawrence, Kansas, USA
Venturelli M. F. Bouman 1986 Embryology and seed development in Mayaca fluviatilis (Mayacaceae). Acta Botanica Neerlandica 35: 497-516[Web of Science]
Venturelli M. F. Bouman 1988 Development of ovule and seed in Rapateaceae. Botanical Journal of the Linnean Society 97: 267-294
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  F. A. Gonzalez and P. J. Rudall Flower and fruit characters in the early-divergent lamiid family Metteniusaceae, with particular reference to the evolution of pseudomonomery Am. J. Botany, February 1, 2010; 97(2): 191 - 206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Yuan, S. Gao, D.-W. Xue, D. Luo, L.-T. Li, S.-Y. Ding, X. Yao, Z. A. Wilson, Q. Qian, and D.-B. Zhang RETARDED PALEA1 Controls Palea Development and Floral Zygomorphy in Rice Plant Physiology, January 1, 2009; 149(1): 235 - 244. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Kinney, J. T. Columbus, and E. A. Friar Unisexual flower, spikelet, and inflorescence development in monoecious/dioecious Bouteloua dimorpha (Poaceae, Chloridoideae) Am. J. Botany, February 1, 2008; 95(2): 123 - 132. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Tillich Seedling Diversity and the Homologies of Seedling Organs in the Order Poales (Monocotyledons) Ann. Bot., December 1, 2007; 100(7): 1413 - 1429. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Kellogg Progress and challenges in studies of the evolution of development J. Exp. Bot., October 1, 2006; 57(13): 3505 - 3516. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
