| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Anatomy and Morphology |
2Department of Biological Sciences, Florida International University, Miami, FL 33199 USA; 3Botany—Centre for Ecology, Evolution and Systematics, The University of New England, Armidale, NSW 2351, Australia; 4Royal Botanic Gardens Sydney, Sydney, NSW 2000, Australia
Received for publication October 1, 2005. Accepted for publication June 21, 2006.
ABSTRACT
Fundamental questions of floral morphology remain unresolved in the grasslike monocots in order Poales, including what constitutes a flower and what constitutes a spikelet. The mapaniid sedges have particularly complex spikeletlike structures, variously interpreted as clusters of flowers or spikelets. Recent phylogenetic studies of Cyperaceae have identified the mapaniid clade as sister to the rest of the family, but the homology of mapaniid reproductive units (RUs) and spikeletlike units (SLUs) to other sedge flowers and spikelets is unclear. We examined reproductive development in the mapaniid Exocarya sclerioides. Inflorescence branches terminated in a SLU with bracts and 1–4 RUs. RUs had four small leaflike structures (LLSs): two lateral LLSs, each associated with a stamen, an abaxial LLS associated with a stamen, and an adaxial LLS. The gynoecium terminated the RU. All RUs were axillary to bracts, and unexpanded bracts and RUs were produced beyond expanded RUs, so SLUs were racemose. RUs developed from a single primordium that initiated two lateral LLSs, then two lateral stamens, then the gynoecium. Initiation of the abaxial LLS and stamen and the adaxial LLS followed. We hypothesize that the RU is a sympodial branch that terminates in a hermaphroditic flower with two stamens and a gynoecium; the two lateral LLSs are halves of a deeply divided prophyll.
Key Words: Australia • Cyperaceae • Mapanioideae • New Guinea • pseudanthium • sedge flower • spikelet
Fundamental questions of floral morphology remain unresolved in the grasslike monocots in order Poales, including what comprises a flower and what comprises a spikelet. This is not surprising, given the reduced size of the structures involved. The grasses (family Poaceae) and sedges (family Cyperaceae) have generally been considered to have spikelets as the basic reproductive unit, although spikelet structure differs in the two families. Historically, the related rushes (family Juncaceae) were generally considered to have flowers as the basic reproductive unit, but researchers (e.g., Köbele and Tillich, 2001
) recently have suggested that the basic unit in the rushes may also be a spikelet (often one-flowered). Our study focused on the mapaniid group of sedges, which has particularly complex spikeletlike structures. There has been much debate about the interpretation of these structures with consequent disagreement on how the mapaniids relate to the rest of the family.
The Cyperaceae is the third largest monocotyledon family and has c. 115 genera that are divided into 2–4 subfamilies and 12–17 tribes (Goetghebeur, 1986
, 1998
; Bruhl, 1995
; Muasya et al., 2000
). Subfamily and tribal delimitations have been problematic in part because simplification of the flowers—which frequently are reduced to stamens and/or carpels—is combined with ramification of the flower-bearing branches or spikelets. Recent cladistic and molecular research is leading to a new understanding of phylogeny and tribal limits within the family (Muasya et al., 2000
; Simpson et al., in press
).
The mapaniid sedges, which include 13 genera, are sister to the rest of the family (Goetghebeur, 1985
, 1998
; Bruhl, 1995
; Muasya et al., 2000
; Simpson et al., 2003
, in press). Previously the mapaniids were placed in one tribe (Hypolytreae, Bruhl, 1995
) or in two tribes, Hypolytreae or Chrysitricheae, in the Mapanioideae (Goetghebeur, 1998
). Most recently, Simpson et al. (2003)
supported recognition of Hypolytreae or Chrysitricheae but with a new circumscription (Table 1). In particular, Simpson et al. (2003)
transferred Capitularina and Exocarya from Goetghebeur's (1986, 1998) Hypolytreae to Chrysitricheae. This result was derived from a combined phylogenetic tree that used pollen structure and chloroplast sequence data. Pollen type was consistent within Simpson et al.'s (2003)
tribes—the genera grouped into the Hypolytreae have individual pollen grains (monads), while the Chrysitricheae have pseudomonads, as do the rest of the Cyperaceae. While this study provided strong bootstrap support for the mapaniid clade, support for subdivision into the tribes was relatively weak.
|
) LLSs interspersed among the stamens and/or surrounding the gynoecium, instead of the three or multiples of three found in most monocots. Each stamen is usually opposite a leaflike structure, but some of the leaflike structures can be interspersed among stamens or between stamens and the gynoecium (Clarke, 1909
; Goetghebeur, 1998
). Eiten (1976)
described the mapaniid reproductive units (her group II) as "pseudospikelets" with many "pseudanthia (a reduced axis with flowers, each looking like a single bisexual flower)," while Kukkonen (1984)
and Simpson (1992)
referred to Eiten's pseudanthia as "spicoids." Comparative studies of mature flowers have been useful in characterizing the diversity of floral structure across the family (Zhang et al., 2004
) but limited in their ability to determine even primary homology of reproductive parts between mapaniid reproductive units and the rest of the sedges. To understand the positional relationships of androecia, gynoecia, and leaflike structures in the mapaniids and whether or how these change ontogenetically, as well as phylogenetically, we need developmental data of reproductive structures for species in the Mapanioideae (cf. Bruhl, 1991
).
Exocarya is a monotypic genus placed in the Chrysitricheae by Simpson et al. (2003)
but in the Hypolytreae s.s. by Goetghebeur (1998)
and Hypolytreae s.l. by Bruhl (1995)
. Exocarya sclerioides (F. Muell.) Benth. is native from northeastern New South Wales and Queensland in Australia to Papua New Guinea. The purpose of this study was to describe the mature structure of the reproductive units in E. sclerioides and to examine development of these units in order to understand spikelet and flower structure in mapaniid sedges, as well as homologies of these reproductive units to spikelets and flowers in other members of the Cyperaceae.
Terminology
Simpson (1992
, p. 14, table 3) reviews the terminology applied by recent researchers to mapanioid reproductive units. We have applied neutral terms to the different reproductive structures in E. sclerioides. The flowerlike structure, which has been called a "pseudanthium" (Eiten, 1976
), a "spikelet" (Koyama, 1967), a "cymule" (Koyama, 1985), a "flower" (Kern, 1974
), and a "spicoid" (Kukkonen, 1984
; Simpson, 1992
), we refer to as a reproductive unit (RU). The small leaflike structures borne on these units, which could be considered leaves, prophylls, bracts, tepals, sepals, petals, or some combination of these leaf homologues, we refer to as leaflike structures (LLSs). The discrete structure that consists of a branch bearing first sterile bracts and later fertile bracts that subtend the RUs, we call a spikeletlike unit (SLU) because of its gross similarity to what are called spikelets in other members of the Cyperaceae.
MATERIALS AND METHODS
Material was collected in 2002 and 2004 from individuals in a single population on the North Coast of New South Wales near Cascade (2002 collections are K.L. Wilson 10167 & J.J. Bruhl; K.L. Wilson 10168 & J.J. Bruhl; and K.L. Wilson 10169 & J.J. Bruhl; 2004 collections are K.L Wilson 10240 and 10241; all lodged at NE and NSW). Several inflorescence branches per individual were fixed whole in formalin : propionic acid : 70% ethanol (FPA, 5 : 5 : 90) and stored in 70% ethanol with 1% glycerol. The 2002 collections were cultivated in a glasshouse at UNE in plastic pots with a medium of peat moss, perlite, and sand; fertilized with slow release fertilizer supplemented by periodic soluble fertilizer; and watered regularly. Additional inflorescences were sampled from this glasshouse-grown material.
Individual SLUs were dissected under a dissecting microscope and photographed for mature structure. SLUs at different developmental stages were embedded in paraplast for sectioning (Ruzin, 1999
) or partially dissected and processed for viewing with scanning electron microscopy (SEM). Paraplast-embedded SLUs were sectioned at 5, 8, or 10 µm, stained in safranin-haematoxylin, examined with a compound light microscope and photographed with a digital camera. SEM material was critical point dried with CO2, coated with gold-palladium, and examined on a JOEL (Peabody, Massachusetts, USA) JSM-5900-LV low vacuum scanning electron microscope at the Florida Center for Analytical Electron Microscopy at Florida International University (FIU). Specimens were examined at 15 or 20 kV in either backscatter or secondary electron emission mode and photographed digitally. Some specimens were further dissected after initial SEM examination, recoated, and viewed again.
RESULTS
Mature structure
Inflorescences overall are anthelate and bear spikeletlike units (SLUs) singly at branch tips (Fig. 1). Each SLU branch has a two-keeled prophyll at its base; this prophyll has a deep notch between the pointed keels (Fig. 6). Lateral axes have an elongated internode, the epipodium, distal to the prophyll (Fig. 1). Additional bracts, which are borne along the axis at successively shorter internodes, have a single keel (Fig. 2). Each of these bracts subtends an axillary bud that can grow out as an additional reproductive branch (Fig. 1). At the distal end of the SLU branch, however, the bract buds do not grow out, and the internodes between bracts are shorter than 1 mm. This segment of the branch forms the SLU. The SLUs are ovate and c. 3 mm long at maturity (mean ± SD = 2.8 ± 0.3, N = 30; Fig. 3); if fruits begin to develop, the SLUs become gibbous. Each SLU has spirally arranged bracts that increase in size distally up to the part of the axis that bears the reproductive units (RUs). Thus, the first-formed bracts on the SLU are sterile, although these bracts can subtend undeveloped buds. The first RU is usually borne in the axil of the sixth bract on the SLU (range = 4–10, N = 30). Gender and number of mature RUs in SLUs varies within inflorescences and among individuals. SLUs matured 1–4 RUs in the material we examined, with the most common condition being two or three (Table 2). RUs are either male or bisexual. Male RUs are not produced distal to female RUs (Table 2), and if only a single RU is matured by a SLU, it is bisexual. We did not find any completely male SLUs, although we did find SLUs that produced only bisexual RUs (Table 2).
|
|
The vernation of the LLSs varies among RUs, with LLS position in the RU and with developmental stage. In male RUs the adaxial LLS is relatively narrow, but in hermaphroditic RUs it is broader and wraps around the gynoecium (Figs. 4, 5); in both cases the adaxial LLS is interior to the two lateral LLSs (Figs. 9, 10). The abaxial LLS extends its margins inside the two lateral LLSs, while the adaxial LLS is found inside the margins of one or both of the lateral LLSs (Figs. 9, 10).
|
Development of the RUs
RUs are initiated in the axils of bracts as oval mounds that are flattened adaxially and rounded abaxially (Figs. 11, 12). These axillary meristems expand parallel to the axil and simultaneously produce two lateral lobes that develop into the two lateral LLSs, leaving a broad meristematic mound between them (Fig. 13). The two LLSs are positioned somewhat adaxially on the meristem (Figs. 13–15). The RU primordium does not initiate the abaxial or adaxial LLSs at this stage (Figs. 13–17). The outer portion of the meristem between the two LLSs develops into the two lateral stamens, one opposite each lateral LLS (Figs. 16–18). The central mound left by segmentation of the original meristem into LLSs and stamens then segments into two additional primordia, which become the stylar branches of the gynoecium (Figs. 17–19). The RU has not initiated the abaxial and adaxial LLSs when the gynoecium is defined (Figs. 16–18). The abaxial LLS is initiated after the two lateral LLSs and stamens and the bilobed gynoecium differentiate (Figs. 18–20). This LLS and its associated stamen lag behind development of the two lateral LLSs and stamens (Figs. 19–22, 24–27). Like the abaxial LLS, the adaxial LLS is initiated after development of the two lateral LLSs (Figs. 14, 16) and is smaller than the LLSs in early development (Fig. 23).
|
|
Structure of the SLU
Our model for development of the SLU of E. sclerioides is given in Fig. 28. The SLU of E. sclerioides superficially appears to terminate in a hermaphroditic RU, but the SLU develops racemosely, with unexpanded bracts and RUs produced beyond the RUs that mature. The positional relationships of bracts and axillary buds is especially clear in immature SLUs (Figs. 7, 8, 11). Our observations and interpretation of racemose SLU development is concordant with Eiten's (1976) conclusion that pseudanthia (RUs) of mapaniids are racemosely arranged on a rachis (i.e., the SLU). Evidence for racemose SLU development is seen in the axillary position of all of the RUs and the continuation of the SLU tip beyond the last-matured RU; this SLU tip bears unexpanded bracts with axillary buds that are unexpanded RUs.
|
; Snyder and Richards, 2005
); this temporal separation presumably promotes outcrossing in this wind-pollinated species. The variation in RU number per SLU in an E. sclerioides inflorescence would make such simultaneity less likely and may allow for geitonogamy.
Structure of the reproductive unit (RU)
Our model for RU structure of E. sclerioides is given in Fig. 29; its general structure agrees with that of Clarke (1909)
, Goetghebeur (1986)
, Simpson (1992)
, and Wilson (1993)
. We propose that this structure is a spikelet rather than a single flower. As compared to non-mapaniid Cyperaceae, the mature structure of RUs of E. sclerioides is unusual in the number (four instead of three or a multiple of three), arrangement (position), and vernation of the LLSs (Figs. 19--21). The two lateral LLSs are exterior, while the abaxial and adaxial LLSs are inside these structures and are initiated on the meristem after differentiation of the lateral LLSs, the two lateral stamens, and the gynoecium.
|
We hypothesize that there is a common genetic and developmental basis for prophylls in the Cyperaceae, which are addorsed, frequently bilobed, and often bikeeled (Blaser, 1944
; Holttum, 1948
). Under this hypothesis, interpretation of the two LLSs in E. sclerioides as a deeply divided prophyll is consistent with the presence of the single adaxial prophyll found throughout the family (cf. Bruhl, 1991
). The second flower in the cymose spikelet of Cladium jamaicense, which is produced laterally after the terminal flower differentiates, arises as a primordium in the axil of a subtending bract; the primordium produces a single prophyll, then two lateral stamens that arise simultaneously, then a terminal gynoecium (Richards, 2002
). Thus the timing and pattern of initiation of structures in the sawgrass flower are similar to initial patterns of differentiation for the RU of E. sclerioides except that E. sclerioides produces two lateral LLSs instead of a single prophyll.
Support for interpreting the two LLSs as homologous to a single prophyll is seen in the patterns of variation in RU structure in other genera of the Mapanioideae. The structure of the RU in Exocarya resembles that of RUs in the more speciose Hypolytreae, especially Hypolytrum and Mapania (Clarke, 1909
; Goetghebeur, 1986
). The outermost structures on the RU of Hypolytrum and Mapania can be either two lateral LLSs, each with a single keel or a single bikeeled LLS (Clarke, 1909
, e.g., tab. CIV; Eiten, 1976
). The presence of two lateral or a single bikeeled LLS can vary within a species; different specimens of H. supervacuum C.B. Clarke manifest both cases (Clarke, 1909
, tab. CV), which provides additional evidence for homology of the two LLSs with a single prophyll. In monotypic Principina grandis, a single structure in the position of our two lateral LLSs surrounds the stamens and gynoecium (Uittien, 1935; presented again in Goetghebeur, 1986
, fig. 8.1.5). Future floral developmental work should sample representative mapaniid species that vary in the presence of a single addorsed LLS vs. two lateral LLSs.
In Simpson et al.'s (2003)
reorganization of tribes in the Mapanioideae, Exocarya is grouped with four small genera in the Chrysitricheae (Table 1). RU structure in these genera is very diverse and variable (Clarke, 1909
; Goetghebeur, 1986
), ranging from Chrysitrix, which has several outer sterile LLSs, >150 stamens and associated LLSs, inner sterile LLSs, and a central gynoecium, through Lepironia, which has two outer LLS, a more limited number of inner fertile, then sterile LLSs around a central gynoecium, to Exocarya, which has two outer fertile LLSs, one inner fertile and one inner sterile LLS, and a central gynoecium. Thus, the common theme among these variable RU structures is the presence of the two outer LLSs, and these are generally in the same lateral position.
After producing the lateral LLSs, the RU primordium initiates the two lateral stamens and the gynoecium from the central meristematic region. We propose that these two stamens and the gynoecium are a single bisexual flower that terminates the RU axis and that the abaxial LLS (with its stamen) and adaxial LLS are bracts on the branch that bears this flower. This interpretation is consistent with the cymose structure of spikelets described for Cladium jamaicense (Richards, 2002
) and sympodial spikelets in several other members of the Schoeneae (Zhang et al., 2004
). An alternative interpretation is that the entire RU is a branch that bears bracts acropetally, each of which potentially subtends a stamen; this branch then terminates in a carpel. Our developmental data do not support this interpretation because initiation of the abaxial LLS and stamen occurs after initiation of the gynoecium, as does initiation of the adaxial LLS. Under the first hypothesis, the RU is a cymose branch with a terminal bisexual flower, and male flowers develop sympodially on this axis. Developmental studies of other mapaniids, especially members of the Chrysitricheae, will help to distinguish between these hypotheses and to understand variations on mapaniid RU structure, and we have embarked on these studies. There is also need to extend the sample of these critical species in molecular phylogenies and ultimately conduct evolutionary–developmental studies to test our hypotheses of organ interpretation in the mapaniids and more generally in the Cyperaceae.
Recent phylogenetic studies separate Cyperaceae into two subfamilies, the Mapanioideae and the Cyperoideae (Simpson et al., in press
). Floral developmental studies in the Cyperoideae, where there is consensus on which reproductive unit is homologous to a flower, show a somewhat different pattern of development (Richards, 2002
; Vrijdaghs et al., in press
). A primordium in the axil of a glume forms two lateral stamen primorida; an abaxial stamen primordium arises simultaneously or slightly later in some species (Vrijdaghs et al., in press
). The remaining floral apex becomes the gynoecium primordium and develops an annular ovary primordium around an ovule primordium. Perianth parts, if present, are produced relatively late in development; they arise in trimerous whorls below the stamens when the ovary primordium is formed. These flowers differ from the RUs of E. sclerioides in that they lack LLSs; they have perianth structures that arise later in development in trimerous whorls below the fertile parts, rather than interspersed with them as the LLSs are in E. sclerioides; and the gynoecial structure differs in formation of the annular ovary primordium, then differentiation of the stigma/style lobes. The greatest similarity between RUs in E. sclerioides and flowers in the Cyperoideae appears to be in the development of the two lateral stamens and gynoecium of the RU of E. sclerioides, which we hypothesize is homologous to a flower, and the stamens and gynoecium of the Cyperoideae. Studies of reproductive development in other mapaniids will show how common this developmental pattern is in the subfamily.
FOOTNOTES
1 The authors thank R. Willis (UNE) and I. Simpson (UNE) for assistance with the maintenance of glasshouse plants and the Florida International University FCAEM Laboratory for support of scanning electron microscopy. 
5 Author for correspondence (Richards{at}fiu.edu
)
LITERATURE CITED
Blaser H. W.. 1944. Morphology of the Cyperaceae. II. The prophyll. American Journal of Botany 31: 53-64.[CrossRef][ISI]
Bruhl J. J.. 1991. Comparative development of some taxonomically critical floral/inflorescence features in Cyperaceae. Australian Journal of Botany 39: 119-127.[CrossRef]
Bruhl J. J.. 1995. Sedge genera of the world: relationships and a new classification of the Cyperaceae. Australian Systematic Botany 8: 125-305.[CrossRef][ISI]
Bruhl J. J. Watson L. Dallwitz M. J.. 1992. Genera of Cyperaceae: interactive identification and information retrieval. Taxon 41: 225-234.
Clarke C. B.. 1909. Illustrations of Cyperaceae Williams and Norgate, London, England.
Eiten L. T.. 1976. Inflorescence units in the Cyperaceae. Annals of the Missouri Botanical Garden 63: 81-112.[CrossRef][ISI]
Goetghebeur P.. 1985. Studies in the Cyperaceae. 6. Nomenclature of the suprageneric taxa in the Cyperaceae. Taxon 34: 617-632.[CrossRef][ISI]
Goetghebeur P.. 1986. Genera Cyperacearum. Ph.D dissertation, University of Ghent, Belgium, Ghent.
Goetghebeur P.. 1998. Cyperaceae. In K. Kubitzki, H. Huber, P. J. Rudall, P. S. Stevens, and T. Stuetzel [eds.] Flowering plants, monocotyledons: Alismatanae and Commelinanae (except Gramineae) 141-190 Springer-Verlag, Berlin, Germany.
Holttum R. E.. 1948. The spikelet in Cyperaceae. Botanical Review 14: 525-541.[ISI]
Kern J. H.. 1974. Cyperaceae. In C. G. G. J. van Steenis [ed.] Flora Malesiana, series I, Spermatophyta 435-753 Noordhoff International Publishing, Leyden, Netherlands.
Köbele C.P. Tillich H.-J.. 2001. Die Infloreszenzen der Juncaceae. Sendtnera 7: 137-161.
Kukkonen I.. 1984. On the inflorescence structure in the family Cyperaceae. Annales Botanici Fennici 21: 257-264.[ISI]
Muasya A. M. Bruhl J. J. Simpson D. A. Culham A. Chase M. W.. 2000. Suprageneric phylogeny of Cyperaceae: a combined analysis. In K. L. Wilson and D. A. Morrison [eds.] Monocots: systematics and evolution 593-601 CSIRO Publishing, Collingwood VIC, Australia.
Richards J. H.. 2002. Flower and spikelet morphology in sawgrass, Cladium jamaicense Crantz (Cyperaceae). Annals of Botany 90: 361-367.
Ruzin S. E.. 1999. Plant microtechniques and microscopy Oxford University Press, New York, New York, USA.
Simpson D. A.. 1992. A revision of the genus Mapania (Cyperaceae) Royal Botanic Gardens, Kew, UK.
Simpson D. A. Furness C. A. Hodkinson T. R. Muasya A. M. Chase M. W.. 2003. Phylogenetic relationships in Cyperaceae subfamily Mapanioideae inferred from pollen and plastid DNA sequence data. American Journal of Botany 90: 1071-1086.
Simpson D. A. Muasya A. M. Alves M. Bruhl J. J. Dhooge S. Chase M. W. Furess C. A. Ghamkhar K. Goetghebeur P. Hodkinson T. R. Marchant A. D. Nieuwborg R. Reznicek A. A. Roalson E. H. Smets E. Starr J. R. Thomas W. W. Wilson K. L. Zhang X. In press Phylogeny of Cyperaceae based on DNA sequence data—a new rbcL analysis. In J. T. Columbus, E. A. Friar, C. W. Hamilton, J. M. Porter, L. M. Prince, and M. G. Simpson [eds.] Monocots: comparative biology and evolution Rancho Santa Ana Botanic Garden, Claremont, California, USA.
Snyder J. M. Richards J. H.. 2005. Floral phenology and compatibility of sawgrass, Cladium jamaicense (Cyperaceae). American Journal of Botany 92: 738-745.
Vrijdaghs A. Muasya A. M. Goetghebeur P. Caris P. Nagels A. Smets E. In press A floral ontogenetic approach to homology questions within the Cyperoideae (Cyperaceae). Botanical Review..
Wilson K. L.. 1993. Cyperaceae. In G. J. Harden [ed.] Flora of New South Wales, vol. 4 293-396 New South Wales University Press, Kensington, Australia.
Zhang X. Wilson K. L. Bruhl J. J.. 2004. Sympodial structure of spikelets in the tribe Schoeneae (Cyperaceae). American Journal of Botany 91: 24-36.
This article has been cited by other articles:
|
|
 |
|
  Flowering Newsletter bibliography for 2006 J. Exp. Bot., April 20, 2007; (2007) erm028v2. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
abaxial LLS and stamen have been initiated. The adaxial LLS, if present, is not well developed (black arrow). 21. RU primordium from abaxial side with lateral LLSs, lateral stamens, and bilobed gynoecium well developed. The abaxial LLS (black arrow) and stamen primordium are visible. Scale bars = 20 µm, 
