| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Paleobotany |
2Department of Environmental and Plant Biology, Ohio University, Athens, Ohio 45701 USA; 3Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9
Received for publication January 3, 2002. Accepted for publication April 18, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
Key Words: Bennettitales • Cretaceous • Cycadeoidea • seed cones • Williamsonia
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
). For nearly a century numerous characters have implicated cycadeoids in the origin of flowering plants (Arber and Parkin, 1907
). Cladistic analyses of morphological characters have consistently supported this hypothesis by placing Bennettitales within the anthophytes, a clade that also includes flowering plants (Crane, 1985
; Doyle and Donoghue, 1986
, 1992
; Nixon et al., 1994
; Rothwell and Serbet, 1994
; Doyle, 1996
; Pryer et al., 2001
). However, results of these analyses are not fully concordant, and the absence of a complete character set for any of the extinct taxa weakens support values for all of the resulting hypotheses of cladistic relationships.
By contrast, several strongly supported but distinctly dissimilar hypotheses of relationships for living seed plants have been developed from analyses of molecular data (discussed by Donoghue and Doyle, 2000
; and Sanderson et al., 2000
). Unfortunately, the cumulative results of these studies further cloud, rather than clarify, our understanding of spermatophyte phylogeny (Donoghue and Doyle, 2000
). Because only five of the major seed plant clades have living representatives, there is a strong probability that the results of the molecular analyses reflect high levels of error and bias (Felsenstein, 1978
), and it is becoming increasingly likely that resolving the overall pattern of spermatophyte relationships will not be achieved by a analysis of extant groups alone (Sanderson et al., 2000
). Therefore, the discovery and characterization of fossils that provide missing characters for extinct spermatophyte clades are important priorities for systematic studies that focus on the resolution of seed plant phylogeny.
Over the past several years Late Cretaceous deposits of Vancouver and Hornby Islands along the west coast of southern British Columbia, Canada have yielded large numbers of plant fossils (Ludvigsen and Beard, 1997
; Smith and Stockey, 2002
) with superb anatomical preservation. These include the most recent permineralized cones of the Bennettitales yet discovered. The specimens are assignable to Cycadeoidea Buckland and Williamsonia Carruthers, and they display several previously unknown or incompletely understood characters (Rothwell and Stockey, 2001
).
The current investigation focuses on ovulate cones assignable to Cycadeoidea that display features of the sporophylls, seeds, interseminal scales, and receptacle. These cones differ from all previously described species of Cycadeoidea in overall size and shape, features of the interseminal scales and seeds, and unique features of the seed integument and sporophyll vascularization. Therefore, they are described as Cycadeoidea maccafferyi sp. nov. A new species of Williamsonia that displays structural features of the cone, as well as anatomical details of seeds, megagametophytes, pollen tubes, and immature embryos, is being published separately.
Many features of C. maccafferyi compare favorably with those that have been described thoroughly for other species of Cycadeoidea (including those assigned to Bennettites Carruthers by previous authors). Therefore, the description in this paper focuses primarily on specific characters of C. maccafferyi and on new information for the genus. This material helps clarify several systematically informative characters of the Bennettitales and allows for the development of a contrasting set of characters to differentiate confidently seed cones of Cycadeoidaceae from those of Williamsoniaceae.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
). These are from the Upper Cretaceous Haslam Formation of the Nanaimo Basin, now dated as early Campanian based on the occurrence of the ammonite Sphenoceramus schmidti (J. Haggart, Geological Survey of Canada, personal communication, 2001). A third specimen (EMB052) is from Piggot Creek, an abandoned quarry pit along a logging road east of Courtenay (49°46'30'' N, 125°22'30'' W; UTM coordinates 10U CA 162291) on central Vancouver Island, and comes from the Comox Formation believed to be Turonian/Coniacian (J. Haggart, personal communication, 2001). The fourth specimen (VIPM 118) was collected on the beach at Collishaw Point (49°33'04'' N, 124°41'03'' W; UTM coordinates 10U CK 777897) along the north coast of Hornby Island, just west of Vancouver Island (Ludvigsen and Beard, 1997
). Specimens at this locality come from the Upper Cretaceous (late Campanian) Lambert Formation of the Comox Basin, the date based on ammonite succession (Ludvigsen and Beard, 1997
; J. Haggart, personal communication, 2001).
Specimens were photographed and then serially sectioned by the well-known cellulose acetate peel technique (Joy, Willis, and Lacey, 1956
). Peels for microscopic examination and image capture were mounted on microscope slides with Eukitt mounting medium (O. Kindler, Freiburg, Germany). Images were captured with a Microlumina digital scanning camera (Leaf Systems, Bedford, Massachusetts, USA), and processed with Adobe Photoshop (Adobe Systems, Inc., San Jose, California, USA). Specimens, peels, and microscope slides are housed at the Vancouver Island Paleontology Museum (VIPM; 118 and 1884) and the Courtenay and District Museum, Courtenay, British Columbia (EMB 052 and BO 211).
Cycadeoid systematics and nomenclature have a long, confusing history of inconsistent application. Names used in this paper follow the rationale of Seward (1917)
and Watson and Sincock (1992)
. Specifically, we use Cycadeoidea Buckland and Cycadeoidaceae Buckland and consider Bennettitites Carruthers to be a junior synonym of Cycadeoidea. However, we use Bennettitales Engler, because it is the first and most distinctive name proposed for the order.
|
SYSTEMATICS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
Family
Cycadeoidaceae Buckland
Genus
Cycadeoidea Buckland
Species
Cycadeoidea maccafferyi sp. nov.
Specific diagnosis
Ovulate cones, up to 7.8 cm long and 3.7–7.6 cm wide (exclusive of bracts); displaying small dome-shaped receptacle with anastomosing vascular bundles. Receptacle with resinous cells; interseminal scales with 1–2 resin canals; resinous tissues absent from ovulate sporophylls. Sporophylls and interseminal scales 2.0–5.3 cm long. Seeds 12–19 mm long, 2.1–4.1 mm wide, with micropylar canal terminating just below apices of 5–6 encircling interseminal scales; round at apex, becoming distinctly 4–5 winged at base of micropylar tube. Wings produced both by sclerotesta and sarcotesta; greatly accentuated by sarcotesta apically, diminishing basally, absent from chalaza. Sporophyll with highly convoluted vascular strand immediately below seed; straight basally.
Holotype hic designatus
Cone specimen BO 211 from Brannen Lake, showing receptacle, interseminal scales, sporophylls, and seeds with megagametophytes and dicotyledonous embryos is here designated the holotype. This specimen is deposited in the Courtenay and District Museum, Courtenay, British Columbia, Canada.
Stratigraphy
Haslam Formation, Upper Cretaceous, early Campanian.
Paratypes
Specimen VIPM1884 in the Vancouver Island Paleontology Museum, Qualicum Beach, British Columbia, from Brannen Lake, EMB 052 (Courtenay and District Museum) from Piggot Creek, and specimen VIPM118 from Hornby Island are designated paratypes.
Etymology
The specific epithet maccafferyi is proposed for Dane McCaffery, Nanaimo, British Columbia, who generously provided the holotype specimen for study.
|
DESCRIPTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
).
|
|
|
The receptacle is more or less round in cross sections (Fig. 9) and ranges from 1.5 to 2.5 cm in diameter (Figs. 3, 4). Except for traces that extend into the interseminal scales and sporophylls (Fig. 9, at T), the receptacle is separated from these organs by calcite filling (Fig. 9). Ground tissues of the receptacle are compact, consisting of cells that appear resinous and that are interspersed with cells that have prominent walls and no internal contents (Figs. 10, 11). Numerous small, terete vascular bundles are preserved near the periphery of the receptacle (Fig. 10). Although the available material is insufficient to interpret vascular architecture in the receptacle, the bundles anastomose (Fig. 11) in a fashion that appears to be similar to trace divergence described for specimens of Williamsonia (Sharma, 1970
).
Individual bundles are small and oval, and are represented by tracheids (Fig. 11) with scalariform secondary wall thickenings. In cross sections most of the tracheids are randomly arranged, but there is some indication of radial alignment to one side of some bundles. This suggests that there was limited secondary vascular tissue production in the cones. Interseminal scale traces and sporophyll traces appear to diverge in a common phyllotactic helix (Fig. 9), and each trace consists of a terete xylem strand that is surrounded by a thin zone of dark, incompletely preserved cells (Fig. 10). A crescent-shaped patch of poorly preserved tissue occurs to the inside of each trace at the periphery of the receptacle (Fig. 10). This tissue accompanies the traces as they diverge from the receptacle (Fig. 10, at right), truncates rapidly (Figs. 9, 10), and is absent distally.
Sections in the midregion of the cone reveal cross sections of numerous polygonal interseminal scales (Fig. 8) and interspersed sporophyll stalks (Fig. 8, arrows). Toward the periphery of the cone at these levels, seeds of the shortest sporophylls are cut obliquely (Fig. 8 at S). The latter orientation is the result of peripheral ovules bending toward the cone surface (Figs. 3, 5). More distal levels of the cone reveal cross sections of seeds on the centrally located sporophylls (Fig. 6).
At their base the interseminal scales are arranged around the much larger sporophylls and are flattened with their greatest dimension paralleling the periphery of the sporophyll (Fig. 17). At all levels each interseminal scale displays a terete xylem strand that is surrounded by a narrow cylinder where no cells are preserved (Figs. 8, 16, 17). As is also characteristic of some previously described Cycadeoidea species (Lignier, 1894
; Stopes, 1918
), a small percentage of the interseminal scales display two traces (Fig. 17, at arrows). Cortical cells have prominent walls and no internal contents, and there is a distinct epidermis of darkly stained cells on the surfaces where interseminal scales abut each other (Fig. 17). Epidermis is less prominent on surfaces that are adjacent to a sporophyll (Fig. 17). Interseminal scales typically display one or sometimes two large resin canals. These canals are roughly round in cross sections and have dark contents (Fig. 17).
More distally, the interseminal scales increase in size and become polygonal in cross sections (Figs. 8, 12, 16). Cortical cells consist of thick-walled sclerenchyma (Fig. 16), and both the resin canals and epidermis are much less distinct (Figs. 8, 12, 16) than in basal sections. In longitudinal sections the cortical sclerenchyma cells are elongated (Fig. 26). Many cortical cells show scalariform or oval simple pits, but the dark walls of the epidermal cells are smooth (Fig. 26). At the level of the seeds, the interseminal scales become flattened parallel to the periphery of the seed body (Figs. 6, 13), but at their tips (i.e., at the cone periphery) they are again polygonal and surround the seed micropyles (Fig. 7). At distal levels the cortical sclerenchyma cells become isodiametric (Fig. 20), developing extremely thick, dark walls (Figs. 5, 20, 21) and very small lumina (Figs. 12, 15, 20), and producing a highly sclerotic tissue that encloses the seeds.
|
Somewhat more distally the sporophyll cortex diminishes to four to nine cell layers thick, and the cortical cells increase in size dramatically, measuring 40–50 µm in diameter (Fig. 16). At still more distal levels the thickness of this zone continues to decrease, being only about two cell layers thick at the juncture of the sporophyll apex and seed base (Figs. 14, 23, 24). In the apical-most regions, these cells form a single cell layer of sarcotesta that incompletely surrounds the sclerotesta of the seeds (Figs. 12, 15). From just above the base of the sporophyll to the apex of the seeds there are obvious intercellular spaces that separate and emphasize the tubular nature of these cortical/sarcotestal cells (Figs. 12, 14, 15, 26).
For most of the length of the sporophylls the course of the sporophyll trace is straight, but just below the base of the seed it becomes highly convoluted (Figs. 19, 23, 24). These convolutions extend for a distance of 1–1.5 mm immediately below the base of each seed (Fig. 19) and follow a roughly helical course in three dimensions (Figs. 23, 24). By measuring the length of the vascular strand along the length of the convolutions in a single section and then doubling this length to account for the convolutions occurring in three dimensional space, it is estimated that the trace is approximately 5–7 mm longer than the sporophyll over this small distance. Convolutions are absent from the adjacent tubular cells, which remain straight (Figs. 23, 24) over the entire length of the sporophyll and seed. Vascular trace convolutions of this type are not known for any other species of Cycadeoidea and are therefore a diagnostic character of C. maccafferyi. Perhaps more importantly, this distinctive structure is characteristic of contractile roots in monocotyledonous flowering plants (Davey, 1946
) and is the basis for suggesting that the seeds of this species were initially emergent and then retracted below the tips of the interseminal scales following pollination (Rothwell and Stockey, 2001
).
Each sporophyll terminates as a single erect seed (Figs. 1, 3, 5) 12–19 mm long and 2.1–4.1 mm wide. Seeds have a round micropylar tip (Fig. 7) and chalaza (Fig. 8), but there are four or five integumentary ribs in the midregion (Figs. 6, 12, 13, 15). The length of each rib increases distally, giving the seeds a stellate appearance at the base of the micropylar canal (Figs. 6, 12, 15), and then the rib truncates below the tip of the micropyle. Seeds show a multilayered integument (Figs. 12, 13, 15, 20–22) that surrounds a nucellus that is attached only at the chalaza (Fig. 22).
The seed integument is basically similar to that of many other nonangiospermous seed plants, consisting of endotesta, sclerotesta, and outer sarcotesta. However, unlike many major gymnospermous clades, there is no evidence of vascular tissue in the integument of any bennettitalean species. The endotesta and sclerotesta are comparable to those that have been previously recognized and thoroughly characterized for several species of Cycadeoidea, particularly species described as Bennettites gibsonianus (Carruthers, 1870
; Solms-Laubach, 1890
), B. morierei (Lignier, 1894
), and B. albianus (Stopes, 1918
). Endotestal cells are not always easily recognized, but where preserved they are narrow and longitudinally elongated, thin walled, and often contain dark contents (Fig. 22). Cells at the inner margin of the sclerotesta are also longitudinally elongated, but are wider than the endotestal cells and have thicker cell walls (Figs. 20, 22). Toward the outer margin of the sclerotesta the cells intergrade with progressively darker, thicker walled, and isodiametric stone cells (Figs. 20–22) and commonly display prominent lumina (Figs. 22, 26).
As is elaborated in the discussion section of this paper, cells of the sarcotesta have not always been recognized as such, but their features are more or less similar in all species of Cycadeoidea for which this tissue is well known (Lignier, 1894
; Stopes, 1918
). Most sarcotestal cells are tubular and are an apical continuation of the cortex of the sporophyll (Figs. 21, 22). As in the region of the sporophyll, these tubular cells are usually not in organic attachment to more centrally located tissues (cf. Figs. 16 and 20, at left). Small, isodiametric cells with thin walls connect the tubular sarcotestal cells to the sclerotesta in B. albianus (Stopes, 1918
), but similar cells are not preserved in C. maccafferyi. In C. maccafferyi, tubular sarcotestal cells completely surround the sclerotesta at the chalaza (Fig. 14), but are restricted to radii that alternate with ribs at more distal levels (Fig. 15).
Distal to the chalaza, large thin-walled sarcotestal cells are attached to the integumentary ribs (Figs. 13 at left and bottom, 15) and are comparable to tissue described for the best preserved specimens of previously described species (Table 1). These cells extend away from the sclerotesta on the radius of each integumentary rib and produce the stellate outline of the seed below the level of the micropylar tip (Figs. 6, 12, 15). Sarcotestal cells have not been observed at the most distal levels of the micropylar tube.
|
; Stockey and Rothwell, 2001
), no pollen chamber is produced. Rather, the slender tip of the nucellus is made up of tightly packed, thin-walled cells with amber contents (Figs. 20, 21).
The seed cavity in many seeds is occupied by embryos that are surrounded by incompletely preserved tissues of the megagametophyte (Figs. 18, 20, 21). The embryos of C. maccafferyi are comparable to those that have been thoroughly described for several other cycadeoid species (Carruthers, 1870
; Solms-Laubach, 1890
; Lignier, 1894
; Wieland, 1906
, 1916
). They have two straight cotyledons (Fig. 18) in which the provascular strands are prominent. The radicle is tightly pressed into the apical region of the seed cavity (Figs. 20, 21) and connected to the cotyledons by a short hypocotyl. The epicotyl is quite small and is not evident in most sections (Fig. 18).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
; Carruthers, 1870
; Ward, 1900
; Wieland, 1906
). Therefore, many species are characterized by a combination of morphological, cuticular, and internal anatomical features of vegetative and fertile organs. A large number of permineralized specimens of the Cycadeoidaceae have been described either as species of Cycadeoidea Buckland or Bennettites Carruthers, but all are now widely regarded as belonging to a single genus, wherein Cycadeoidea has nomenclatural priority (Seward, 1917
; Watson and Sincock, 1992
). The majority of Cycadeoidea species are described from stems bearing leaf bases and cones (e.g., Wieland, 1906
, 1916
). A few additional species, including C. maccafferyi, are represented by isolated seed cones for which the other organs of the sporophyte are unknown. Nevertheless, these isolated cones display a large enough number of diagnostic characters that they can be confidently identified as either belonging to well-known species of Cycadeoidea or to new species (Table 1).
Cycadeoidea maccafferyi is among the geologically youngest species of permineralized Cycadeoidaceae thus far described (Nishida, 1994
; Saiki and Yoshida, 1999
; Rothwell and Stockey, 2001
). Cones are larger and have distinctly larger seeds than any of the previously described species (Table 1). Cycadeoidea maccafferyi is also the only species to display distinctive contractile tissue below the seed at the distal end of the sporophyll (Figs. 19, 23, 24; Rothwell and Stockey, 2001
). Additional characters that set C. maccafferyi apart from previously described species include the combination of (1) monosporangiate seed cones with small dome-shaped receptacles, (2) length of sporophylls and interseminal scales, (3) distribution of tubular and attached sarcotestal cells, and (4) histological features of the various tissues (Table 1).
Sarcotesta or cupule?
In previous studies of Bennettitales, some authors have recognized the presence of a cupule surrounding the seeds, whereas others have not. Bennettitalean cupules have been interpreted from studies of both permineralized specimens and from cuticular studies of compressed remains (Harris, 1932
; Pederson, Crane, and Friis, 1989
). From permineralized specimens, the extremely elongated tubular cells that surround the vascular bundle in the sporophyll and the sclerotesta of Cycadeoidea seeds are interpreted quite differently by different authors. Lignier (1894)
provides the first detailed description of this tissue for B. morierei. He considers the tissue to be a multilayered epidermis at the level of the sporophyll, an interpretation that is accepted by Stopes for B. albianus (Stopes, 1918
). More distally, at the level of the seeds, Lignier interprets the tubular cells and the large, thin-walled cells that accentuate the integumentary ridges to be outer fleshy layers of the integument (i.e., "envelope tubuleuse" and "l'assise plissée," respectively, of Lignier, 1894
).
Wieland interprets the tubular cells to be outer cortex at the level of the sporophyll that "instead of passing over into an outer integumentary layer, forms a cup-shaped supporting basal husk..." (Wieland, 1906
, p. 120). We concur with Wieland (1906)
that the tubular cells of the sporophyll probably represent cortex. Stopes (1918)
interprets the tubular cells surrounding the seeds of B. albianus to be a multicellular aril-like cup that is an apical extension of the multilayered epidermis (sensu Lignier, 1894
) of the seed stalks (= cortex of Wieland [1906]
and as interpreted in this study). Interestingly, Stopes (1918)
also describes a zone of delicate cells that are in organic continuity with both the seed sclerotesta (to the inside) and the tubular cells (to the outside; e.g., Text-Figs. 7–9 of Stopes, 1918
), but considers these delicate cells to represent a "deliquescent layer of surrounding stalk tissues" (Stopes, 1918
, p. 404). To our knowledge, this is the only specimen of Cycadeoidea in which these delicate cells are preserved.
The interpretation of a cupule surrounding the seeds in the Bennettitales derives from the descriptions by Wieland (1906)
and Stopes (1918)
, whereas the alternative interpretation that the same tissues represent seed integument derives from the studies of Lignier (1894)
. We agree with Lignier (1894)
that both the tubular cells and the large thin-walled cells that accentuate the seed wings in Cycadeoidea are parts of the seed sarcotesta. This is for several reasons. First, the tubular cells are an apical extension of the outer tissues of the sporophyll, as is the sarcotesta in other groups of seed plants with seeds placed apically on sporophylls (i.e., cordaites—Rothwell, 1981
; walchian conifers—Mapes and Rothwell, 1984
; vojnovskyaleans—Rothwell, Mapes, and Mapes, 1996
) or on ovulate stalks (i.e., Ginkgo biloba L.—Chamberlain, 1935
; Bierhorst, 1971
). This indicates that the tubular cells are homologous to the mesophyll and epidermis of a leaf, not a separate organ, such as a cupule. Second, the tubular cells of B. albianus are in organic continuity with the sclerotesta (through the "deliquescent layer" of Stopes, 1918
). Third, a sarcotesta is differentiated in the integument of Williamsonia (Rothwell and Stockey, 2001
; Stockey and Rothwell, 2001
), but a cupule is clearly absent. Fourth, no cuticle has been recognized either to the outside of the tubular cells or between the tubular cells and the sclerotesta in the permineralized specimens, as would be expected if there were a cupule surrounding the seed.
If the tubular cells of Cycadeoidea do represent a cupule, then this genus is cupulate but the seeds lack a sarcotesta, whereas Williamsonia has sarcotesta but lacks a cupule. We consider the tubular cells of Cycadeoidea to be homologous to the peg-like sarcotestal cells of Williamsonia (Seward, 1912
; Stockey and Rothwell, 2001
), and for this to be, by far, the more parsimonious explanation of homology for this tissue. Interpreting the outer tubular cells of Cycadeoidea as part of the integument also establishes the existence of sarcotesta in this genus, as is characteristic of ovules and seeds in nearly all other gymnospermous spermatophyte clades (Chamberlain, 1935
; Rothwell and Serbet, 1994
). If this reasoning is correct, then seed cones of both Cycadeoidea and Williamsonia lack a cupule.
Compression remains from which a bennettitalean cupule has been interpreted are assignable to the Triassic genus Vardekloeftia Harris. In his original study of Vardekloeftia (Harris, 1932
), and in a more recent investigation using both light and scanning electron microscopy, Pedersen, Crane, and Friis (1989)
interpret the cuticles of V. sulcata Harris as being derived from the inside of the integument, from the outside of the integument, and from a cupule. However, the much more complete and easily interpreted permineralized cycadeoid specimens now allow us to explain the cuticular envelopes of Vardekloeftia in terms of what is preserved in Cycadeoidea and Williamsonia. Although some authors interpret bennettitalean seeds to have a pollen chamber (i.e., Sharma, 1976
; Ohana, Kimura, and Chitaley, 1998
), figures of permineralized specimens of Bennettites, Cycadeoidea, Williamsonia and Bennetticarpus all show nucellar plugs like those that we have found in Cycadeoidea maccafferyi and the new species of Williamsonia from Vancouver Island (Lignier, 1894
; Stopes, 1918
; Wieland, 1906
, 1916
; Sharma, 1970
, 1976; Delevoryas and Gould, 1973
; Ohana, Kimura, and Chitaley, 1998
). Seeds in these reproductive structures all lack a pollen chamber. Lignier (1894)
and Stopes (1918)
originally showed that Cycadeoidea (= Bennettites of these authors) has a plug of nucellar tissue in the micropyle and a nucellus that is free to the base of the integument. Despite some interpretations to the contrary, figures of all subsequently described, permineralized bennettitalean cones support Lignier and Stopes. With this improved understanding of bennettitalean seed anatomy, the elongated structure with a thick cuticle at the apex of Vardekloeftia (e.g., Plate V, Fig. 3 of Pedersen, Crane, and Friis, 1989
) is much more reasonably interpreted to be a nucellar plug like that of the other bennettitaleans. If this interpretation is correct, then the two cuticles surrounding the plug probably represent the inner and outer cuticles of the integument, rather than a cupule. It therefore follows that there is no evidence from fossils to support the conclusion that a cupule was produced by any bennettitalean.
Familial characters of cycadeoid seed cones
Bennettitalean species with massive, sparsely branched trunks traditionally have been assigned to the Cycadeoidaceae Buckland (i.e., Cycadeoidea Buckland and Monanthesia Wieland ex Delevoryas), whereas those with more slender, highly branched trunks are most often grouped within the Williamsoniaceae (e.g., Williamsonia Carruthers; Alvin et al., 1967
; Watson and Sincock, 1992
). Cones of the Cycadeoidaceae traditionally have been regarded as both monosporangiate and bisporangiate and are borne among the leaf bases on stout trunks (Delevoryas, 1968
). However, there is some evidence that the monosporangiate cones of some species may actually represent bisporangiate cones for which the microsporangiate structures simply have not been recognized (Crepet, 1972
). In contrast, cones of the Williamsonia are apparently all monosporangiate and are borne at the tips of lateral branches (e.g., Sahni, 1932
). While some other genera of the Williamsoniaceae appear to bear bisporangiate cones (Watson and Sincock, 1992
), cones of the family all appear to be borne at the tip of either main or axillary branches.
As now confirmed by the excellently preserved specimens from western Canada and elsewhere (Rothwell and Stockey, 2001
; Stockey and Rothwell, 2001
), seed cones of the two families display several other contrasting characters that allow for their familial assignment (Table 2). Whereas the interseminal scales and sporophylls of seed cones in the Cycadeoidaceae nearly parallel one another, those of the Williamsoniaceae do not. Instead, they radiate from the receptacle around an arc of up to 300°. Some bisporangiate cones of Cycadeoidea also have interseminal scales and sporophylls that radiate in many directions (e.g., Wieland, 1906
, 1916
; Delevoryas, 1968
; Crepet, 1974
), but those of the putative seed cones do not. Seeds of the Cycadeoidaceae have distinct longitudinal ribs below the apex of the micropylar canal, while those of the Williamsoniaceae are round or oval in cross sections at all levels (Seward, 1912
; Rothwell and Stockey, 2001
; Stockey and Rothwell, 2001
). The sarcotesta is also distinctly different in seeds of the two families. Sarcotestal cells of Williamsoniaceae form small, radially protruding pegs (Stockey and Rothwell, 2001
), whereas those of the Cycadeoidaceae are of two distinct types. Large, loosely arranged tubular cells extend from the base of the seeds to near the apex. Other large, thin-walled sarcotestal cells are attached to the sclerotesta and extend radially to accentuate the longitudinal ribs of the integument.
|
), all too often the systematic importance of these data are not widely appreciated because of the combination of relatively low support values for many nodes, and the much more easily developed data used in molecular analyses. For seed plants, the anthophyte hypothesis was first developed using morphological data from both living and extinct taxa (Crane, 1985
), and it has been supported by virtually every subsequent broad-based analysis of morphological characters (Doyle and Donoghue, 1986
, 1992
; Loconte and Stevenson, 1990
; Nixon et al., 1994
; Rothwell and Serbet, 1994
; Doyle, 1996
; Pryer et al., 2001
). Even as the homologies, definitions, and scorings of some characters have remained in dispute (cf. Nixon et al., 1994
; Rothwell and Serbet, 1994
; Doyle, 1996
), there is a concordance of results with respect to the existence of an anthophyte clade. We consider it to be unlikely that the results of all of these studies would include an anthophyte clade, unless they were detecting a true phylogenetic signal.
If this rationale is correct, then the degree of concordance among the results of analyses using different data sets and methodologies (not merely support values for the results of a given analysis) is the more valid test for hypotheses of phylogenetic relationships. Most current attempts to achieve congruence among cladistic studies of seed plants begin with the premise that the framework for spermatophyte relationships can be developed using molecular characters of living taxa, to which extinct taxa can be added later (e.g., Frohlich and Parker, 2000
). Given the small number of extant seed plant groups (i.e., only five), the probability of ever achieving success using this approach may be extremely low (Felsenstein, 1978
; Sanderson et al., 2000
). As a result, the alternative approach of developing the framework for seed plant relationships from morphological characters of living and extinct taxa, and then using this framework to test which results of the more restricted analyses (i.e., of living taxa only) reflect a true phylogenetic signal, may prove more fruitful. Although the latter approach necessitates the extremely time-consuming tasks associated with developing and scoring morphological characters (particularly for extinct clades), it may be our current best option for resolving the overall pattern of land plant phylogeny. In this regard, the continued focus on identifying, developing, and clarifying cladistically informative morphological characters for extinct clades is an important priority for systematic paleontology.
|
FOOTNOTES |
|---|
4 Author for reprint requests (Tel: 740-593-1129; FAX: 740-593-1130; rothwell{at}ohio.edu
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICSDESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
Arber E. A. N. J. Parkin 1907 On the origin of angiosperms. Journal of the Linnean Society of London, Botany 38: 29-80
Bierhorst D. W. 1971 Morphology of vascular plants. Macmillian, New York, New York, USA
Buckland W. 1828 On the Cycadeoidaceae, a family of fossil plants found in the Oolite quarries of the Isle of Portland. Transactions of the Geological Society of London, 2nd Series 2: 395-401
Carruthers W. 1870 On fossil cycadean stems from the secondary rocks of Britain. Transactions of the Linnean Society of London 26: 675-708
Chamberlain C. J. 1935 Gymnosperms, structure and evolution. University of Chicago Press, Chicago, Illinois, USA
Crane P. R. 1985 Phylogenetic analysis of seed plants and the origin of angiosperms. Annals of the Missouri Botanical Garden 72: 716-793[CrossRef][ISI]
Crepet W. L. 1972 Investigations of North American cycadeoids: pollination mechanisms in Cycadeoidea. American Journal of Botany 59: 1048-1056[CrossRef][ISI]
Crepet W. L. 1974 Investigations of North American cycadeoids: the reproductive biology of Cycadeoidea. Palaeontographica 148B: 144-169
Davey A. J. 1946 On the seedling of Oxalis hirta L. Annals of Botany, new series 10: 237-256
Delevoryas T. 1968 Investigations of North American cycadeoids: structure, ontogeny and phylogenetic considerations of cones of Cycadeoidea. Palaeontographica 121B: 122-133
Delevoryas T. R. E. Gould 1973 Investigations of North American cycadeoids: Williamsonian cones from the Jurassic of Oaxaca, Mexico. Review of Palaeobotany and Palynology 15: 27-42
Donoghue M. J. J. A. Doyle 2000 Seed plant phylogeny: demise of the anthophyte hypothesis?. Current Biology 10: R106-R109[CrossRef][ISI][Medline]
Doyle J. A. 1996 Seed plant phylogeny and the relationships of Gnetales. International Journal of Plant Sciences 157: S3-S39[CrossRef]
Doyle J. A. M. J. Donoghue 1986 Seed plant phylogeny and the origin of angiosperms: an experimental cladistic approach. Botanical Review 52: 321-431
Doyle J. A. M. J. Donoghue 1992 Fossils and seed plant phylogeny reanalyzed. Brittonia 44: 89-106[CrossRef][ISI]
Felsenstein J. 1978 Cases in which parsimony and compatibility methods will be positively misleading. Systematic Zoology 27: 401-410[CrossRef][ISI]
Frohlich M. W. D. S. Parker 2000 The mostly male theory of flower evolutionary origins: from genes to fossils. Systematic Botany 25: 155-170
Harris T. M. 1932 The fossil flora of Scoresby Sound, east Greenland—Part 3, Caytoniales and Bennettitales. Meddelelser Om Grønland 112: 1-114
Joy K. W. A. J. Willis W. S. Lacey 1956 A rapid cellulose peel technique in paleobotany. Annals of Botany, new series 20: 635-637
Kenrick P. P. R. Crane 1997 The origin and early diversification of land plants, a cladistic study. Smithsonian Institution Press, Washington, D.C., USA
Lignier O. 1894 Végétaux fossiles de Normandie, structure et affinités du Bennettites morierei Sap. & Mar. (Sp). Imprimerie E. Lanier, Caen, France
Lignier O. 1910 Le Bennettites morierei (Sap. Et Mar.) Lignier ne serait-il pas d'Origine infra-crétacé?. Bulletin de la Societie Linnéenne de Normandie 6 série, 2: 214-220
Loconte H. D. W. Stevenson 1990 Cladistics of the Spermatophyta. Brittonia 42: 197-211[CrossRef][ISI]
Ludvigsen R. G. Beard 1997 West coast fossils. Harbour Publishing, Madeira Park, British Columbia, Canada
Mapes G. G. W. Rothwell 1984 Permineralized ovulate cones of Lebachia from Late Paleozoic limestones of Kansas. Palaeontology 27: 69-94
Nishida H. 1994 Morphology and evolution of Cycadeoidales. Journal of Plant Research 107: 479-492[CrossRef][ISI]
Nixon K. C. W. L. Crepet D.W. Stevenson E. M. Friis 1994 A reevaluation of seed plant phylogeny. Annals of the Missouri Botanical Garden 81: 484-533[CrossRef][ISI]
Ohana T. T. Kimura S. Chitaley 1998 Bennetticarpus yezoites sp. nov. (Bennettitales) from the Upper Cretaceous of Hokkaido, Japan. Paleontological Research 2: 108-119
Pedersen K. R. P. R. Crane E. M. Friis 1989 The morphology and phylogenetic significance of Vardekloeftia Harris (Bennettitales). Review of Palaeobotany and Palynology 60: 7-24[CrossRef][ISI]
Pryer K. M. H. Schneider A. R. Smith R. Cranfill P. G. Wolf J. S. Hunt S. D. Sipes 2001 Horsetails and ferns are a monophyletic group and the closest living relatives to seed plants. Nature 409: 618-622[CrossRef][Medline]
Rothwell G. W. 1981 Cordaianthus duquesnensis sp. nov., anatomically preserved ovulate cones from the Upper Pennsylvanian of Ohio. American Journal of Botany 68: 239-247
Rothwell G. W. G. Mapes R. H. Mapes 1996 Anatomically preserved vojnovskyalean seed plants in Upper Pennsylvanian (Stephanian) marine shales of North America. Journal of Paleontology 70: 1067-1079[Abstract]
Rothwell G. W. R. Serbet 1994 Lignophyte phylogeny and the evolution of spermatophytes. Systematic Botany 19: 443-482[CrossRef][ISI]
Rothwell G. W. R. A. Stockey 2001 Anatomically preserved cycadeoid cones from the Cretaceous of Vancouver and Hornby Islands. Abstracts of the Fourth British Columbia Paleontological Symposium. University College of the Cariboo, Kamloops, British Columbia, Canada
Sahni B. 1932 A petrified Williamsonia (W. sewardiana. sp. nov.) from the Rajmahal Hills, India. Memoirs of The Geological Survey of India (Palaeontologica Indica), new series. 20: 1-19
Saiki K. Y. Yoshida 1999 A new Bennettitalean trunk with unilacunar five-trace nodal structure from the Upper Cretaceous of Hokkaido, Japan. American Journal of Botany 86: 326-332
Sanderson M. J. M. F. Wojciechowski J.-M. Hu T. Sher Khan S. G. Brady 2000 Error, bias and long-branch attraction in data for two chloroplast photosystem genes in seed plants. Molecular Biology and Evolution 17: 782-797
Seward A. C. 1912 A petrified Williamsonia from Scotland. Philosophical Transactions of the Royal Society of London 203B: 101-126
Seward A. C. 1917 Fossil plants, vol. 3. Cambridge University Press, Cambridge, UK
Sharma B. D. 1970 On the vascular organization of the receptacles of seed-bearing williamsonias from the Middle Jurassic rocks of Amarjola in the Rajmahal Hills, India. Annals of Botany 34: 1036-1070
Sharma B. D. 1976 Fruit development in Williamsonia Carr. (Bennettitales). Géobios 9: 503-507[CrossRef]
Smith S. Y. R. A. Stockey 2002 Permineralized pine cones from the Cretaceous of Vancouver Island, British Columbia. International Journal of Plant Sciences 163: 185-196[CrossRef]
Solms-Laubach H. G. 1890 Über die fructification von Bennettites gibsonianus Carr. Botanische Zeitung 49: 789-798, 50: 805-816, 51: 821-834
Stockey R. A. G. W. Rothwell 2001 An ovulate Williamsonia cone from the Upper Cretaceous of western North America. Botany 2001: Plants and people. Botanical Society of America, Albuquerque, New Mexico, USA
Stopes M. C. 1918 New bennettitean cones from the British Cretaceous. Philosophical Transactions of the Royal Society of London 208B: 389-440[CrossRef]
Ward L. F. 1900 Description of a new genus and twenty new species of fossil cycadean trunks from the Jurassic of Wyoming. Proceedings of the Washington (D.C.) Academy of Science 1: 253-300
Watson J. C. A. Sincock 1992 Bennettitales of the English Wealden. Monograph of the Palaeontographical Society, London, Publication Number 588: 1-228
Wieland G. R. 1906 American fossil cycads. Publication 34, Carnegie Institution of Washington, Washington, D.C., USA
Wieland G. R. 1916 American fossil cycads. Vol. II, Taxonomy. Publication 34. Carnegie Institution of Washington, Washington, D.C., USA
This article has been cited by other articles:
|
|
 |
|
  C. Pott, C. C. Labandeira, M. Krings, and H. Kerp Fossil Insect Eggs and Ovipositional Damage on Bennettitalean Leaf Cuticles from the Carnian (Upper Triassic) of Austria Journal of Paleontology, July 1, 2008; 82(4): 778 - 789. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. E. Ryberg, E. L. Taylor, and T. N. Taylor Secondary phloem anatomy of Cycadeoidea (Bennettitales) Am. J. Botany, May 1, 2007; 94(5): 791 - 798. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
