| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Development and Morphogenesis |
2Morfología Vegetal, Facultad de Ciencias Agracias, National University of Litoral, Kreder 2805, S3080HOF Esperanza, Santa Fe, Argentina; 3Instituto de Botánica Darwinion, C.C. 22, B1642HYD San Isidro, Buenos Aires, Argentina
Received for publication April 23, 2004. Accepted for publication November 30, 2004.
ABSTRACT
Inflorescence development in Panicum maximum and Urochloa plantaginea was comparatively studied with scanning electron and light microscopy to test the transfer of P. maximum to Urochloa and to look for developmental features applicable to future cladistic studies of the phosphoenol pyruvate carboxykinase (PCK) subtype of C4 photosynthesis clade (P. maximum and some species of Brachiaria, Chaetium, Eriochloa, Melinis, and Urochloa). Eleven developmental features not discernable in the mature inflorescence were found: direction of branch differentiation; origins of primary branches; apical vs. intercalary development of the main axis; direction of spikelet differentiation; direction of glume, lemma and palea differentiation; position of the lower glume (in some cases); size of the floret meristem; pattern of distal floret development; pattern of gynoecium abortion; differential pollen development between proximal and distal floret; and glume elongation. Inflorescence homologies between P. maximum and U. plantaginea are also clarified. Panicum maximum and U. plantaginea differ not only in their mature inflorescence structure but also in eight fundamental developmental features that exclude P. maximum from Urochloa. The following developmental events are related to sex expression: size of floret meristem, gynoecium abortion, pollen development delay in the proximal floret, glume elongation and basipetal floret maturation at anthesis.
Key Words: development • homology • inflorescence • Paniceae • Panicum maximum • Poaceae • sex expression • Urochloa plantaginea
The grass subfamily Panicoideae includes approximately 208 genera grouped in several tribes; among these, Paniceae, with more than 110 genera, and Andropogoneae, with 85 genera, are the largest and most important ones (Clayton and Renvoize, 1986
; Watson and Dallwitz, 1992
). Because the tribe Paniceae is highly diverse in morphological, physiological, anatomical, and karyological characters (Zuloaga et al., 2000
; Duvall et al., 2001
; Giussani et al., 2001
), different evolutionary schemes have been proposed for this tribe and its genera (Aliscioni et al., 2003
). According to recent findings and the increase of samples studied, the phylogeny of Paniceae is undergoing several changes, even though the taxonomical delimitation of some of its genera is still unclear (Zuloaga et al., 2000
; Duvall et al., 2001
; Giussani et al., 2001
; Aliscioni et al., 2003
).
Recent studies on the phylogeny of Paniceae (Zuloaga et al., 2000
; Duvall et al., 2001
; Giussani et al., 2001
) showed that Brachiaria eruciformis (Smith) Griseb., Chaetium bromoides (J. Presl.) Benth. ex Hemsl., Eriochloa punctata (L.) Desv. Ex Hamilton f. intermedia Parodi, Melinis repens (Willdenow) Zizka, Urochloa acuminata (Renvoize) Morrone & Zuloaga, U. plantaginea (Link) Webster, U. mutica (Forsskal) Nguyen, and Panicum maximum Jacq. form a monophyletic group with high bootstrap support. This was called "the PCK clade" because all the taxa use the phosphoenol pyruvate carboxykinase (PCK) subtype of the C4 photosynthetic pathway (Aliscioni et al., 2003
). In spite of strong support for the monophyly of the PCK clade, relationships among these taxa are still unclear. An example of this problem is the controversial taxonomic affiliation of P. maximum, which has been referred to as Urochloa (Webster, 1987
; Giussani et al., 2001
; Aliscioni et al., 2003
) as well as to the subgenus Megathyrsus Pilger of Panicum, recently upgraded to a new independent genus (Simon and Jacobs, 2003
). Except for the anatomy related to the photosynthetic pathway, no other morphological features distinguish the PCK clade. Among the different morphological features of the taxa involved in the PCK clade, the structure of the inflorescence is remarkably diverse. However, the morphology of mature inflorescences of Poaceae is not enough to understand their morphological diversity and relationships (LeRoux and Kellogg, 1999
; Kellogg, 2000a
, b
, 2003
, 2004
; Doust and Kellogg, 2002
). A comparative analysis of inflorescence development in Setaria, Pennisetum, and Cenchrus, also closely related members of the tribe Paniceae, showed that only a few changes in the pattern of development account for the considerable range of variation seen at maturity (Doust and Kellogg, 2002
).
Considering the potential value of the inflorescence in determining systematic relationships within Paniceae, a comparative study of inflorescence development in two members of the PCK clade, P. maximum and U. plantaginea, is carried out with two aims: (1) to test if inflorescence development supports inclusion of P. maximum in Urochloa or its segregation in a new, independent genus Megathyrsus and (2) to search for features in the development of inflorescences that could be used in future cladistic studies of the PCK clade.
Panicum maximum was selected for study because of its uncertain taxonomic affiliation. Among the species of Urochloa, U. plantaginea is one of the closest species to P. maximum in the analyses of Giussani et al. (2001)
and Aliscioni et al. (2003)
, but its mature inflorescence differs greatly from that of P. maximum. Urochloa plantaginea is an annual herb with bilateral inflorescences and spikelets on short pedicels (Morrone and Zuloaga, 1992
). Panicum maximum is a perennial herb with radiate, lax inflorescences and spikelets on long pedicels (Zuloaga, 1979
; Zuloaga and Morrone, 1995
). Both species have bifloral spikelets in which the distal floret is hermaphroditic and the proximal one is male in P. maximum and neutral (only a lemma and a palea are observed) in U. plantaginea (Zuloaga, 1979
; Morrone and Zuloaga, 1992
).
MATERIALS AND METHODS
Fresh inflorescences of Panicum maximum and Urochloa plantaginea were collected from natural populations in Santa Fe, Argentina between September 2001 and March 2002. Twenty-five plants were studied per accession. About 150 samples of inflorescences (in total) were fixed in FAA (formalin : acetic acid : 70% ethanol, 10 : 5 : 85, v/v) to be studied with a stereomicroscope. About 25 samples were selected from the original stock for scanning electron microscopy (SEM) and light microscopy studies.
For SEM observations, fixed inflorescences were dissected and classified with a stereomicroscope Zeiss DV4 (Jena, Germany), according to the different stages of development. After that, the samples were dehydrated in an alcohol series plus two final changes of 100% acetone. Dehydrated material was dried by critical point with CO2 as transitional fluid and coated with gold-palladium using a BAL-TEC SCD 050 (Balzers, Switzerland). All samples of inflorescences, spikelets, and florets were observed and photographed using a JEOL JSM-T 100 (Kent, UK) scanning electron microscope from the Electron Microscopy Service of La Plata Museum, Buenos Aires, Argentina. Measurements of the floral meristems were standardized following the instructions provided by the Electron Microscopy Service of La Plata Museum.
For studies with light microscopy, fixed samples were dehydrated with isobutyl alcohol, and infiltrated with and embedded in Histoplast (Ruzin, 1999
). Longitudinal and transverse sections 10 µm thick were stained with safranin, fast green, and Mayer's haematoxylin (Johansen, 1940
), and mounted with Eukitt (Hatfield, PA, USA) on a glass slide.
RESULTS
Morphology of the mature inflorescence
The structure of the mature inflorescence of both species has been previously described by Zuloaga (1989)
, Zuloaga and Morrone (1995)
, Morrone and Zuloaga (1992)
, Reinheimer and Vegetti (R. Reinheimer, unpublished data, National University of Litoral) and will be briefly mentioned here.
Panicum maximum has a lax and radiate inflorescence (Fig. 1) where the main axis ends in a terminal spikelet (Fig. 2). The inflorescence includes 18–56 primary branches, each one ending in a terminal spikelet (Figs. 1, 2). The highest branch degree observed is the fifth-order (Fig. 2). Primary branches are alternate; characteristically, they are pseudoverticillate in the proximal region of the inflorescence and sometimes subopposite in the middle region of the inflorescence (Fig. 2).
|
Branch system development
A comparison of inflorescence branch system development between P. maximum and U. plantaginea is presented in Table 1. During vegetative growth, the apical meristem of the shoot of P. maximum and U. plantaginea elongates intravaginally and produces leaf primordia in two ranks (distichous) (Fig. 5). The transition from the vegetative state to the flowering one is evident when the apical meristem elongates beyond the last formed leaf primordium to form the main axis of the inflorescence. In addition, the apical meristem of P. maximum (90.105 µm diam.) is larger than the apical meristems of U. plantaginea (68.586 µm diam.) (Figs. 6, 11).
|
|
|
In P. maximum, while the first-order branches are increasing in length and the apical meristem is still elongating and producing primary branches, second-order branches are initiated distichously at the base of first-order branches in the proximal region of the inflorescence (Figs. 7, 8). Initiation of the second-order branches is acropetal in the whole inflorescence and along first-order branches. Meanwhile, the main axis is still elongating, new second-order branches are being produced on distal primary branches, and third-order branches are originating distichously at the base of the first formed (proximal) secondary branches (Fig. 9). The inflorescence of P. maximum becomes more complex when new higher-order branches are produced. Among the samples observed, the maximum branch degree in P. maximum is up to the fifth-order. Third-, fourth- and fifth-order branches also are initiated in acropetal succession on both the whole inflorescence and on their subtending branches. In contrast to P. maximum, secondary branch primordia of U. plantaginea are amphipetaly initiated in two ranks on the abaxial side of every primary branch (Figs. 13– 15). All secondary branches in U. plantaginea differentiate in basipetal direction on the whole inflorescence, but in amphipetal succession on the primary subtending branch. The primary branches flatten as they elongate, and new secondary branches are initiated (Figs. 14, 15). The secondary branches arrest their development at the moment that glumes differentiate at their top. Less frequently, tertiary branches may be initiated on the proximal region of the basal primary branch. Hence, paired spikelets can be found at the proximal region of the last formed basal, primary branch in the inflorescence of U. plantaginea.
In both species, the inflorescences emerge from the sheath by elongation of internodes. As a result of the differential elongation of the internodes of the main axis, the inflorescence of P. maximum characteristically shows a subverticillate arrangement of primary branches at the proximal region of the inflorescence and, sometimes, opposite the middle of the inflorescence (Fig. 10).
Spikelet development
A comparison of spikelet development between P. maximum and U. plantaginea is presented in Table 2. During the development of spikelets, P. maximum and U. plantaginea differed in the order of branch primordia on which spikelets are differentiated and the size of meristems from which the floral organs were initiated. Spikelet differentiation on the whole inflorescence and on branches is basipetal. In both species, glumes and lemmas are initiated acropetally on the spikelet axis.
|
|
|
In P. maximum, differentiation and maturation of the floral organs within the spikelet is basipetal (Figs. 18, 19). Three stamen primordia develop first in the distal floret. Two of them are initiated on the lateral flanks of the meristem and one, abaxially (Fig. 18). Just after the inception of stamen primordia, one palea differentiates on the floret axis alternately with the lemma and surrounding the floret meristem (Figs. 18, 19). Meanwhile, three stamen primordia and a palea are initiated in the proximal floret following the same pattern as the distal one (Figs. 18, 19). Later, the stamen primordia expand to form thecae (Fig. 20). Before the two lodicules differentiate in a whorl outside the stamen primordia, the gynoecial primordium develops from the remaining floret meristem (Figs. 20, 21). The gynoecium of the distal floret develops a gynoecial ridge on the same side as the upper lemma, surrounding the ovule primordium (Figs. 20, 21). At the same time, the proximal floret is enveloped by the glumes, while the distal floret remains exposed and the anthers of both florets elongate above the gynoecium (Figs. 20, 22). The gynoecium of the lower floret arrests its development before the gynoecial ridge becomes evident (Figs. 30, 31). After that, filaments of the stamens gradually elongate and, in the distal floret, the branches of the style and stigmas develop. Thus, in P. maximum both florets arise as hermaphroditic primordium, and while the distal floret remains hermaphroditic up to anthesis, the proximal one develops as a male floret by abortion of the gynoecium primordium.
|
By anthesis, the floral organs of every spikelet are completely enveloped by glumes in both species.
Histological development of floral organs
Histological development of florets differs within and between each species. In P. maximum, microsporangial development starts at the same time in both florets and continues simultaneously up to the pollen mother cell stage (PMC, Figs. 30, 37a). Afterward, the proximal floret delays pollen development at the PMC stage, while the distal floret continues pollen development up to the 1-celled stage (Figs. 32–34, 37b). After that, pollen development in the distal floret is temporary delayed at 1-celled stage, while pollen development restarts in the proximal floret reaching the 1-celled stage (Figs. 35, 36, 37c). Finally, pollen in the distal floret restarts its development, reaching the 3-celled stage even before the pollen of the proximal floret. At anthesis, the distal floret opens before the proximal one (Fig. 37d, e).
When pollen in the proximal and distal floret of P. maximum is at the PMC stage, the gynoecium of the proximal floret ceases development (Fig. 30). A reconstruction of the arrested gynoecium based on longitudinal seriate sections (15 florets) shows cellular death, involving both epidermal and subepidermal cells, in a subapical, transversal plate one cell thick. Cells of this plate gradually lose their nuclei and cytoplasm (Fig. 31). Only the cell walls remain intact (Fig. 31). In contrast, during distal floret development, all cells of the gynoecium maintain the integrity of their cytoplasm and nucleus (Fig. 30).
A reconstruction based on transverse serial sections of the proximal sterile anthecium of U. plantaginea (15 florets) shows cells of the stamen and the gynoecium totally collapsed, without inner structure (Fig. 29).
DISCUSSION
Changes in the shoot apex related to flowering transition
The transition to flowering in shoots of Panicum maximum and Urochloa plantaginea involves the same meristem elongation observed in other members of the Poaceae family (Stür, 1986
; Fraser and Kokko, 1993
; Orr and Sundberg, 1994
; Sundberg et al., 1995
; Sundberg and Orr, 1996
; Doust and Kellogg, 2002
). The change of phyllotaxis in P. maximum correlates with the increase in diameter of the apical meristem and the number of orthostichies. (cf. Sundberg and Orr, 1996
).
Branch system development
Development of the branch system of P. maximum and U. plantaginea differs with respect to the growth of the main axis, number and disposition of primary branch primordia, direction of branch differentiation in the whole inflorescence and on branches from which they are originated, and the degree of ramification and disposition of secondary branch on the primary branch. These developmental differences produce the two different mature morphologies described by Zuloaga (1989)
, Zuloaga and Morrone (1995)
, Morrone and Zuloaga (1992)
, and Reinheimer and Vegetti (unpublished data, National University of Litoral) (Figs. 1–4). However, there are some additional differences in the development of the branching system of P. maximum and U. plantaginea that cannot be seen in mature inflorescences: (1) the origin of primary branches and development of the main axis, and (2) the direction of the differentiation of branch primordia.
Primary branches in the inflorescence of P. maximum develop from "regular" buds derived from the apical meristem. The main axis of the inflorescence is the result of contiguous, elongated internodes also produced by the apical meristem, which finally differentiates a terminal spikelet. In U. plantaginea the apical meristem of the main axis only produces the first primary branch. After the arising of the second primary branch, the apical meristem of the main axis is laterally displaced, remaining inactive and not producing any further structure. The main axis of U. plantaginea's inflorescence develops by intercalary growth of the internode below the first formed primary branch primordium and the last vegetative leaf. The main rachis never ends in a spikelet. We infer that in U. plantaginea the origin of the second and following primary branches are from meristems that form de novo along this elongating internode. These meristems are similar to the adventitious buds described by Rauh (1937)
. There are three features supporting the adventitious character of these kinds of meristems (cf. Rauh, 1937
): (1) they arise on an elongating internode, (2) they arise basipetaly, and (3) the correlation between the development of meristems that form de novo and the incomplete development of the main axis. Urochloa decumbens has a similar inflorescence structure as U. plantaginea, and its main axis also shows the same kind of intercalary growth with adventitious buds (Stür, 1986
). Although adventitious buds are well known in other angiosperm families (Rauh, 1937
), these two species of Urochloa are the only records of adventitious buds in inflorescences of Poaceae.
Concerning homologies, the main axis of U. plantaginea's inflorescence is homologous to the first basal internode in the main axis of P. maximum's inflorescence. The most distal primary branch of U. plantaginea is homologous to the first proximal primary branch in P. maximum. The second and following primary branches (in basipetal order) of U. plantaginea are not homologous to the first most distal primary branch in the same inflorescence, nor are they homologous to any structures in the inflorescence of P. maximum.
Regarding the direction of initiation of branch primordia, P. maximum shows acropetal initiation at every branch degree, both in the inflorescence as a whole as well as along each level of branching (for instance: secondary branches are acropetal on each primary branch and also along the whole inflorescence). Initiation of the primary branches of U. plantaginea is basipetal and amphipetal in the secondary ones along the primary branches, but the initiation of the secondary branches is basipetal when considering the entire inflorescence (the secondary branches appear first on the distal primary branch).
Spikelet development
The degree of the branches on which spikelets arise differs between both species: in P. maximum spikelets are initiated on first-, second-, third-, fourth- or fifth-order branches, while in U. plantaginea spikelets are differentiated on primary, secondary, or less frequently, tertiary branches. In both species, the direction of spikelet differentiation is basipetal, both along the whole inflorescence, and also in the branches on which they arise. This basipetal differentiation of spikelets implies that the branch maturity needed to form spikelets is not related to the timing of differentiation of branches. In P. maximum, the last formed branches (whatever degree of ramification) are the first to produce spikelets. In U. plantaginea, the amphipetal differentiation of secondary branches does not correlate with the basipetal differentiation of spikelets on the secondary branches of the same primary branch. Besides, the differentiation of spikelets is basipetal along the inflorescence (it begins in the older, apical primary branch). In contrast, Stür (1986)
reported an amphipetal differentiation of spikelets in Urochloa decumbens (sub Brachiaria decumbens), and also amphipetal differentiation of secondary branches.
Floral development
Although the distal floret of P. maximum and U. plantaginea has the same growing pattern, three related differences can be observed between both species in the development and sex expression of the proximal floret: (1) developmental changes determining sex expression, (2) the different size of the floret meristem, and (3) the timing of elongation of the glumes.
Panicum maximum has bifloral spikelets in which the proximal floret is male and the distal one is hermaphroditic (Zuloaga and Morrone, 1995
). Both florets start their development as hermaphroditic primordial, but only the distal ripens as hermaphroditic. The arrest of the gynoecium primordium in the proximal floret determines the formation of a male proximal floret. This phenomenon was also observed in some members of the tribe Andropogoneae, as in Zea mays L. (Sundberg and Orr, 1996
), Heteropogon contortus (L.) P. Beauv. ex Roem and Schult. (LeRoux and Kellogg, 1999
), Tripsacum dactyloides L. (Orr et al., 2001
), and in one species of the tribe Paniceae, Panicum repens L. (LeRoux and Kellogg, 1999
). Concerning the floral development of Andropogoneae, LeRoux and Kellogg (1999)
concluded that cell death in a subepidermal core of the gynoecium primordium leads to the arrest of gynoecium growth and the formation of a male floret; they hypothesized that this mechanism of sex expression may be shared among the subfamily Panicoideae (LeRoux and Kellogg, 1999
). The arrested gynoecium in P. maximum is new evidence for that hypothesis. However, there are differences in the pattern of cell death in the arrested gynoecium of the male floret between Panicum and members of Andropogoneae. While in Andropogoneae, cell death occurs in a core of subepidermal cells of the gynoecium primordium, in P. repens dead cells appear in an epidermal ring at the base of the gynoecium primordium, and in P. maximum, death of both epidermal and subepidermal cells occurs in a subapical, transverse plate one cell thick. In P. maximum, dead cells also retain their cell walls, as has been reported for the Andropogoneae and P. repens (LeRoux and Kellogg, 1999
), but cell death occurs earlier in P. maximum than in P. repens (before the gynoecial ridges appear) and the Andropogoneae. Therefore, not only does the pattern of cell death vary in location, but it also varies in timing, which is not as subtle as LeRoux and Kellogg (1999)
suggested.
Species of Urochloa have bifloral spikelets in which the distal floret is always hermaphroditic and the proximal one can be male or neutral (Morrone and Zuloaga, 1992
). Urochloa plantaginea is an example of the last case. Both florets develop as hermaphroditic primordia, but when the thecae are clearly differentiated and the gynoecium ridge is just arising, the proximal floret ceases its growth resulting in a sterile anthecium. This pattern of development of the proximal floret could be shared by other species of Urochloa, although it is not common to other members of the PCK clade (Eriochloa montevidensis, R. Reinheimer, unpublished data).
The proximal and distal floret in P. maximum clearly differ not only in gynoecium development (because the proximal gynoecium aborts), but they also differ subtly in pollen development. Six stages of compared pollen development between the proximal and distal floret can be distinguished: (1) anthers of both florets begin to develop at the same time until the pollen mother cell (PMC) forms; (2) anthers of the proximal floret arrest their development at the PMC, while anthers in the distal floret undergo meiosis and reach the 1-celled pollen stage; (3) anthers of the proximal floret restart development, undergo meiosis, and reach the 1-celled pollen stage, while the pollen in the distal floret is arrested at the 1-celled stage; (4) pollen in the distal floret undergoes mitosis and reaches the 3-cell stage earlier than the proximal floret; (5) anthesis takes place first in the distal floret of the spikelet; (6) finally, anthesis occurs in the proximal floret. The arrest of pollen development at the PMC in the proximal floret (stage 2) is simultaneous with the abortion of the gynoecium primordium, suggesting a relationship between these two developmental events. Perhaps the genetic control that aborts the gynoecium primordium is also involved in the general delay of floret development (particularly anther and pollen development) and is related to the basipetal maturation of the spikelet's florets.
In maize, Irish and Nelson (1993)
found that stamens and gynoecia with regular development are larger than those that will be aborted. Le Roux and Kellogg (1999)
did not find this size difference in floral organs of Andropogoneae. Although observations in P. maximum and U. plantaginea agree with those of Le Roux and Kellogg (1999)
, there is a relationship between floral meristem size and sex expression in florets in both species. In P. maximum, the proximal meristem (male floret) is about 50% smaller than the distal meristem (hermaphroditic floret). In U. plantaginea, the difference in size between the proximal and distal meristems is even larger than in P. maximum, the meristem of the aborted, proximal floret being less than the 30% the size of the distal one. Therefore, sex expression of florets seems to be already determined as early as the differentiation of floret meristems.
Irish and Nelson (1993)
and Irish et al. (1994)
, studying the floral development in Z. mays, related timing of the elongation of the glume with sex expression of the florets. In the maize tassel, glumes elongate and envelop florets when floral organ primordia differentiate, before abortion of the gynoecium. In the ear, florets are enclosed by glumes after abortion of the stamens. These authors suggested that sex determination genes in maize, and possibly in some Andropogoneae, as suggested by LeRoux and Kellogg (1999)
, influence elongation of the glumes. The elongation of glumes in P. maximum and U. plantaginea agrees with the hypothesis of Le Roux and Kellogg (1999)
: the fact that glumes cover the distal floret earlier in U. plantaginea than they do in P. maximum correlates with the earlier sex expression in U. plantaginea (particularly in the proximal floret).
Taxonomical consequences
Brown (1977)
excluded P. maximum from Panicum and mentioned the convenience of transferring this species to the genus Brachiaria because of the presence of the phosphoenol pyruvate carboxykinase (PCK) subtype of C4 photosyntesis and rugose upper anthecium. Later, Webster (1987)
suggested that P. maximum should be transferred to the genus Urochloa due to a similar upper anthecium orientation and photosynthetic subtype. Panicum maximum was related, in the phylogenetic analyses of Zuloaga et al. (2000)
, Giussani et al. (2001)
, and Aliscioni et al. (2003)
to the Urochloa clade. Consequently, these authors treated the species as Urochloa maxima, following Webster's concept (1987)
. Recently, Simon and Jacobs (2003)
questioned the transfer of P. maximum to Urochloa, mainly because of the difference in the degree of branching of the inflorescence and because the latest cladistic analyses of Paniceae (Zuloaga et al., 2000
; Giussani et al., 2001
; Aliscioni et al., 2003
) support more the segregation of P. maximum from Panicum than its inclusion in Urochloa. Therefore, these authors considered subgenus Megathyrsus Pilger at a generic level, including two species: M. maximus (= Panicum maximum) and M. infestus (= P. infestus). Our new findings on inflorescence development in P. maximum and U. plantaginea also support segregation of P. maximum from Urochloa, not only for the higher degree of branching, as Simon and Jacobs (2003)
stressed, but also due to the different pattern of initiation of primary inflorescence branches, direction of branch differentiation, and phyllotaxis (Table 1)—differences that establish a gap between both developmental patterns. We also suspect that the lack of monophyly of Urochloa (Guissani et al., 2001
) could be supported also by developmental features of the inflorescence. The structural differences in the inflorescence between the group U. mutica-U. plantaginea and U. acuminata (R. Reinheimer et al., unpublished data), could involve not only minor changes of phyllotaxis and number of orthostichies, but also a deeper change in the pattern of development of the main axis and primary branches, as the one observed between P. maximum and U. plantaginea.
Among the developmental features analyzed here the following ones cannot be discerned in mature inflorescences and could be potential sources of new morphological characteristics to be used in future cladistic analysis: (1) the direction of branch differentiation on both the entire inflorescence and each branching level; (2) development of adventitious buds; (3) primary branch initiation (apical vs. intercalary elongation); (4) direction of spikelet differentiation on both the entire inflorescence and each branching level; (5) direction of glume, lemma, and palea initiation; (6) position of the lower glume (in some cases); (7) size of the floret meristem; (8) pattern of distal floret development; (9) pattern of gynoecium abortion; and (10) differential pollen development between proximal and distal floret; (11) glume elongation. Panicum maximum and U. plantaginea share developmental features 4, 5, and 8, and differ by characters 1, 2, 3, 6, 7, 9, 10, and 11. Among the developmental events related to floret sex expression, some seem to precede sex expression (as size of floret meristem), some seem to be simultaneous with sex expression (as gynoecium abortion and pollen development delay), and some others seem to follow determination of sex (as glume elongation and basipetal floret maturation at anthesis).
FOOTNOTES
1 The authors thank Dr. Fernando Zuloaga and Dr. Elizabeth A. Kellogg for critical reading of the manuscript, the two anonymous reviewers for comments on the manuscript and the National Council for Scientific and Technological Research (CONICET) for financial support (PIP 02131). 
4 Author for correspondence (e-mail: renatarein{at}fca.unl.edu.ar
)
LITERATURE CITED
Aliscioni S. A. L. M. Giussani F. O. Zuloaga E. A. Kellogg 2003 A molecular phylogeny of Panicum (Poaceae: Paniceae): test of monophyly and phylogenetic placement within the Panicoideae. American Journal of Botany 90: 796-821
Brown W. V. 1977 The Kranz syndrome and its types in grass systematics. Memoirs of the Torrey Botanical Club 23: 1-97
Clayton W. D. S. A. Renvoize 1986 Genera Graminum. Grasses of the world. Kew Bulletin Additional Series XIII. H. M. S. O., London, UK
Doust A. N. E. A. Kellogg 2002 Integrating phylogeny, developmental morphology and genetics: a case study of inflorescence evolution in the "bristle grass" clade (Panicoideae: Poaceae). In Q. C. B. Cronk, R. M. Bateman, and J. A. Hawkins [eds.], Developmental genetics and plant evolution, 298–314. Taylor & Francis, London, UK
Duvall M. R. J. D. Noll A. H. Minn 2001 Phylogenetics of Paniceae (Poaceae). American Journal of Botany 88: 1988-1992
Fraser J. E. G. Kokko 1993 Panicle, spikelet, and floret development in orchard grass (Dactylis glomerata). Canadian Journal of Botany 71: 523-532[ISI]
Giussani L. M. J. H. Cota-Sánchez F. O. Zuloaga E. A. Kellogg 2001 A molecular phylogeny of the grass subfamily Panicoideae (Poaceae) shows multiple origins of C4 photosynthesis. American Journal of Botany 88: 1993-2012
Irish E. E. T. M. Nelson 1993 Development of tassel seed 2 inflorescences in maize. American Journal of Botany 80: 292-299[CrossRef][ISI]
Irish E. E. J. A. Langale T. M. Nelson 1994 Interactions between the tassel seed genes and other sex determining factors. Developmental Genetics 15: 155-171
Johansen D. A. 1940 Plant microtechnique. McGraw-Hill, New York, New York, USA
Kellogg E. A. 2000a A model of inflorescence development. In K. L. Wilson and D. A. Morrison [eds.], Monocots II. Systematics and evolution, 84–88. CSIRO, Collingwood, Australia
Kellogg E. A. 2000b Molecular and morphological evolution in the Andropogoneae. In S. W. L. Jacobs and J. Everett [eds.], Grasses: systematics and evolution, 149–158. CSIRO, Collingwood, Australia
Kellogg E. A. 2003 Inflorescence development in the grass family. Boletín de la Sociedad Argentina de Botánica 38 (Supplement): 19 (Abstract)
Kellogg E. A. 2004 Evolution of developmental traits. Current Opinion in Plant Biology 7: 92-98[CrossRef][ISI][Medline]
Leroux L. G. E. A. Kellogg 1999 Floral development and the formation of unisexual spikelets in the Andropogoneae (Poaceae). American Journal of Botany 86: 354-366
Morrone O. F. O. Zuloaga 1992 Revisión de las especies sudamericanas nativas e introducidas de los géneros Brachiaria y Urochloa (Poaceae: Panicoideae: Paniceae). Darwiniana 31: 43-109
Orr A. R. M. D. Sundberg 1994 Inflorescence development in a perennial Teosinte: Zea perennis (Poaceae). American Journal of Botany 81: 598-608[CrossRef][ISI]
Orr A. R. R. Kaparthi C. L. Dewald M. D. Sundberg 2001 Analysis of inflorescence organogenesis in eastern gamagrass, Tripsacum dactyloides (Poaceae): the wild type and the gynomonoecious gsf1 mutant. American Journal of Botany 88: 363-381
Rauh W. 1937 Die Bildung von Hypokotyl- und Wurzelsprossen und ihre Bedeutung für die Wuchsformen der Pflanzen. Nova Acta Leopoldina, Neue Folge 4: 395-553
Ruzin S. E. 1999 Plant microtechnique and microscopy. Oxford University Press, New York, New York, USA
Simon B. K. S. W. L. Jacobs 2003 Megathyrsus, a new generic name for Panicum subgenus Megathyrsus. Austrobaileya 6: 571-574
Stür W. W. 1986 Reproductive development of the apex of Brachiaria decumbens Stapf. Annals of Botany 58: 569-575
Sundberg M. D. A. R. Orr 1996 Early inflorescence and floral development in Zea mays land race Chapalote (Poaceae). American Journal of Botany 83: 1255-1265[CrossRef][ISI]
Sundberg M. D. C. Lafargue A. R. Orr 1995 Inflorescence development in the "standard exotic" maize, Argentine popcorn (Poaceae). American Journal of Botany 82: 64-74[CrossRef][ISI]
Watson L. M. J. Dallwitz 1992 The grass genera of the world. CAB International, Wallingford, UK
Webster R. D. 1987 The Australian Paniceae (Poaceae). J. Cramer, Berlin, Germany
Zuloaga F. O. 1979 El género Panicum (Gramineae) en la República Argentina. I. Darwiniana 22: 3-44
Zuloaga F. O. 1989 El género Panicum (Poaceae: Paniceae) en la República Argentina. III. Darwiniana 29: 289-370
Zuloaga F. O. O. Morrone 1995 Panicum L. In A. T. Hunziker [ed.], Flora Fanerogámica Argentina, fascículo 12, 59–88. Pugliese Siena, Córdoba, Argentina
Zuloaga F. O. O. Morrone L. M. Giussani 2000 A cladistic analysis of the Paniceae: a preliminary approach. In S. W. L. Jacobs and J. Everett [eds.], Grasses: systematics and evolution, 123–135. CSIRO, Collingwood, Australia
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
