| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Paleobotany |
2Department of Biological Sciences, LSCB 124, University of South Alabama, Mobile, Alabama 36688 USA; 3Abt. Paläobotanik am Geologisch-Paläontologischen Institut, Westfälische Wilhelms-Universität Münster, Hindenburgplatz 57, D-48143 Münster, Germany 4Department of Ecology and Evolutionary Biology and Natural History Museum and Biodiversity Research Center, University of Kansas, Lawrence, Kansas 66045 USA
Received for publication August 29, 2000. Accepted for publication February 23, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: filmy ferns • Filicales • Hopetedia • Hymenophyllaceae • North Carolina, USA • Pekin Formation • Triassic
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Pryer, Smith, and Skog, 1995
; Stevenson and Loconte, 1996
; Rothwell, 1999
). Historically, detailed comparative morphological and anatomical investigations, as well as consideration of the fossil record, had also led several precladistic workers (e.g., Bower, 1923–1928
) to correctly identify these families as basal. Based on the presence of numerous putatively primitive characters, including marginal sporangia, short sporangial stalks, oblique annuli, and a protostelic anatomy, it has repeatedly been suggested that the Hymenophyllaceae (the "filmy ferns") may also have been among the early filicaleans, even though the fossil evidence for this conclusion is ambiguous. Interestingly, the basal or near basal phylogenetic position of the Hymenophyllaceae is supported by all the modern cladistic analyses cited above.
The assumed "primitive" status of the Hymenophyllaceae is reflected in the numerous Paleozoic and early Mesozoic fossils that were assigned to the genera Hymenophyllites, Trichomanides, or Trichomanites, particularly by 19th century workers (e.g., references compiled in Jongmans and Dijkstra, 1960
, pp. 1366–1384 and 1965, pp. 3512–3516). However, most of these fossils represent indeterminate fronds with narrow laminae and display no unequivocal evidence for a membranaceous ("filmy") habit or a hymenophyllaceous soral morphology. By the early 20th century, relatively few fossils were attributed to the family, due at least in part to Seward's (1910, p. 365) critical review of earlier claims of Paleozoic and Mesozoic Hymenophyllaceae, in which he concluded, "The facts bearing on the geological history of this family are singularly meager. There is no evidence which can be adduced in favor of regarding the Hymenophyllaceae as ferns of great antiquity, which played a prominent part in the floras of the past." Discussion of a possible fossil record of the Hymenophyllaceae is now completely ignored in most reviews of the filicalean record (e.g., Tidwell and Ash, 1994
; Collinson, 1996
). Nevertheless, the basal or near basal position of the Hymenophyllaceae in all of the recent phylogenetic analyses cited above indicates that the family must have been a distinct lineage, at least by the late Paleozoic or early Mesozoic, despite the dearth of conclusive fossil representatives. The membranaceous laminae of the fronds of the Hymenophyllaceae, which in most species are only a single cell thick, is often considered the primary reason for the poor fossil record of the family (e.g., Iwatsuki, 1990
). Although this is undoubtedly a contributing factor, we suggest that even when such plants are preserved, they may frequently have been overlooked due to their inconspicuous appearance. Sterile remains of fossil Hymenophyllaceae lacking clear evidence for a membranaceous habit can easily be confounded with other fern (or even with seed fern) foliage and, thus, difficult to recognize. In addition, some fossil hymenophyllaceous ferns were misidentified due to the general similarity of some forms to bryophytes (Deng, 1997
).
In this paper, we describe and figure a new genus and species of fossil fern from the Late Triassic Pekin Formation of North Carolina. We interpret Hopetedia praetermissa gen. et sp. nov. as being the oldest unequivocal representative of the Hymenophyllaceae. Our conclusions are based primarily on general frond morphology, soral position and morphology, and indirect evidence for a filmy habit. In addition, the features of this fossil are considered with regard to several earlier relevant claims of fossil Hymenophyllaceae. The implications of this fossil taxon relative to current hypotheses of the phylogeny and soral evolution of the Hymenophyllaceae are also considered.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Litwin and Ash, 1993
) and is the source of a well-known Upper Triassic flora (reviewed by Gensel, 1986
). Although the Boren flora includes numerous cycadeoid and conifer remains, the newly described fern is commonly associated with other ferns including Cladophlebis sp. (Osmundaceae), Cynepteris lashiophora (Cynepteridaceae), Dictyophyllum sp. (Dipteridaceae), and Phlebopteris smithii (Matoniaceae). The plant fossils occur in a buff-colored siltstone and are represented by light to medium brown stained impressions. Only fragmentary organic material is preserved. Occasionally small amounts of carbonaceous material may be present on axes, major veins, or annuli of the sporangia; lamina material, however, is always preserved only as stained impressions. Although most specimens represent frond fragments, a few specimens (e.g., numbers T5340, T5357, T5352) are substantially articulated providing some insight into the overall frond morphology, with specimen T5352 showing attachment to a flexuous rhizome fragment. The terminal segments of nearly every specimen are fertile making it difficult to discern the morphology of the indusia, as they are obscured by the carbonaceous annuli of the enclosed sporangia. Only specimen T5340 shows empty sori sufficiently well preserved to reveal the funneliform structure. Due to the probable delicate, membranaceous structure of the fronds, the laminae produced a very faint, inconspicuous stain on the matrix making study and photography difficult. The contrast of the lamina against the matrix was greatly improved by placing polarizing filters in front of the light source and the camera lens. The filters were rotated until maximum contrast was achieved. In addition, some specimens were carefully wetted with water prior to photography in order to further increase the contrast. A color plate (Figs. 12–20) is included to accurately document the appearance of the lamina.
|
The holotype of "Trichomanides laxum" (Tenison-Woods, 1883)
was borrowed from the Australian Museum in Sydney, Australia. This species is sometimes cited as the oldest currently accepted fossil representative of the Hymenophyllaceae (e.g., Tryon and Tryon, 1982
). The specimen (#68460) was examined under the dissecting microscope in order to evaluate Tenison-Woods's (1883)
claims that Trichomanes-like sori were visible, but no preparation was required.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Order
Filicales.
Family
Hymenophyllaceae.
Genus
Hopetedia Axsmith, Krings et Taylor, gen. nov.
Generic diagnosis
Fronds delicate, arising from a creeping rhizome, stipe ascending, blade tripinnate, pinnae bearing alternately positioned pinnules; distal frond and pinna portions less differentiated, often with individual parts fused; pinnules membranaceous, segmented, individual segments characterized by one to several lobes, lobes usually fertile; indusia terminal on lobes, funnel-shaped, each containing a single sorus with sporangia on a short receptacle or sessile; annulus completely surrounding sporangium, oblique to nearly vertical.
Type species
Hopetedia praetermissa Axsmith, Krings et Taylor. Figs. 1–21 in this paper.
|
Species
Hopetedia praetermissa Axsmith, Krings et Taylor, sp. nov., Figs. 1–21.
Holotype
The specimen occurring on slabs T 5340a and T5340c is here designated as the type specimen (represented by Figs. 1–2, 7, 12, 16, 20). Each specimen number represents one portion of one original slab.
Paratypes
T 5357 (Fig. 8); T5352a (Fig. 3); T 5347 (Fig. 5).
|
Specific diagnosis
Creeping, flexuous rhizome; fronds delicate, up to 
20 cm long, loosely organized, stipe ascending, naked, blade deltoid in outline, with pinnae on both sides, rachis naked, pinnae widely positioned, irregularly alternately arranged, up to 4.5 cm long, costae mostly naked, but a narrow lamina (laminar rim) may be present in distal portions, pinnules widely positioned, membranaceous, up to 1.6 cm long, distinctly segmented, segments characterized by one to several entire-margined, rounded to slightly elongate lobes each of which is vascularized by a single vein, individual lobes up to 1 mm long and almost always fertile; indusia (involucres) terminal on lobes, funnel-shaped, their openings oriented toward the pinnule apex, each bearing a single sorus on a short receptacle, formed distal to the end of the veinlet, sporangia per sorus few (5–8), all of almost equal size, with short and thick sporangial stalks or sessile, arranged more or less radially (or helically) on the receptacle; individual sporangia obovate to circular in outline, between 0.2 and 0.5 mm in diameter, annulus complete, oblique to nearly vertical.
Stratum typicum
Middle Pekin-Formation, late Carnian, Upper Triassic of North Carolina.
Locus typicus
Boren Clay Company Pit near Gulf, North Carolina, USA.
Etymology
"praetermissus" (Latin) = overlooked, neglected.
Description
Hopetedia praetermissa was apparently a delicate fern with fronds arising from creeping, slightly flexuous rhizomes that may have been up to 2.5 mm in diameter (Figs. 1–5). Isolated rhizome fragments are found together with remains of H. praetermissa (Fig. 4). The arrangement of the fronds on the rhizomes is uncertain because only a single frond was found attached to a rhizome (Fig. 3, arrow). Numerous frond portions were discovered in the matrix, of which the largest specimen is 12.2 cm long (Fig. 1). The frond laminae are represented only by very faint stains on the matrix that can be clearly seen only after wetting and illumination under cross-polarized light. Although the lamina thickness could not be definitively determined, these facts are consistent with a membranaceous (i.e., frond laminae one to a few cells thick) habit like that of the extant members of the Hymenophyllaceae. Fully differentiated fronds range from 15 to 20 cm long, consisting of a rather long stipe, and a tripinnate leafy portion (blade) that is more or less triangular (deltoid) in general outline. The stipe and rachis are naked. Several specimens demonstrate the arrangement of the pinnae on the rachis (Fig. 1). Pinnae are irregularly and alternately positioned, and rather loosely arranged giving the entire frond a somewhat open (lax) appearance. Fully differentiated pinnae (Figs. 2, 5) are elongate and slightly triangular in outline, at least 4.5 cm long and basally up to 3.5 cm wide; their costae are usually naked (e.g., Figs. 2, 5) but a narrow lamina (laminar rim) may be present in distal portions (e.g., Fig. 8). Pinnae bear widely arranged, alternately positioned, membranaceous pinnules, which may be up to 1.6 cm long in basal pinna portions, but gradually become smaller toward the pinna apex. Pinnae terminate in small apical pinnules (Figs. 6–11). The pinnules are distinctly segmented (Figs. 13–15) with individual segments characterized by one to several entire-margined, rounded to elongate (up to 1.5 mm long and up to 0.6 mm wide) lobes; each lobe is vascularized by a single vein (Figs. 6, 10–15). In the distal frond portions individual pinnae and pinnules may be less well differentiated (Figs. 7, 12, 13), often with individual parts fused. Individual lobes of the segments are almost always fertile (Figs. 12–19). A suggested reconstruction of the frond appears as Fig. 21a. Each ultimate lobe bears a marginal, funnel-shaped (up to 0.5 mm long) indusium (involucre sensu Iwatsuki, 1977
) with its opening usually directed toward the pinnule apex (Fig. 20). Each indusium contains a single sorus with sporangia on a short, somewhat expanded receptacle (Fig. 18 arrows), formed at the end of the vein. Sori (e.g., Figs. 15–19) consist of only a few (5–8) sporangia, each almost equal in size. Sporangia are attached to the receptacle by a short and thick sporangial stalk; some appear to be sessile. It appears for some sori that the sporangia may be arranged more or less radially or helically on the receptacle (Figs. 17, 18). Generally, however, fossilization and diagenesis have compressed the sori (e.g., Figs. 13–16, 19), and thus the arrangement of the sporangia is no longer clearly discernible. Individual sporangia (Figs. 16–18) are obovate to more or less circular in outline, rather large (up to 0.5 mm in diameter), and are characterized by a complete, oblique to nearly vertical annulus (Fig. 16 arrow). A suggested reconstruction of the fertile lobes appears in Fig. 21c.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), are similar to many members of the Hymenophyllaceae generally and Hopetedia praetermissa specifically. Although Coniopteris is not common until the Jurassic, there are rare Triassic records (Kilpper, 1964
). Furthermore, studies by Bower (1908)
and Iwatsuki (1977)
have documented a general similarity in indusial morphology and development between the extant Dicksoniaceae and Hymenophyllaceae. Nevertheless, we can provide clear justification for not assigning Hopetedia to Coniopteris. Most importantly, there is no evidence for a membranaceous lamina in the coniopterids. Moreover, coniopterids typically have much larger fronds (e.g., up to 12 cm wide in Coniopteris hymeniophylloides) and are characterized by considerable sterile/fertile frond dimorphism, as well as dimorphic (apheleboid) basal pinnules based on Harris's (1961)
observations. In addition, individual fertile fronds of Coniopteris have regions containing numerous sterile pinnules. In contrast, the fronds of Hopetedia are smaller and there is no evidence for aphleboid pinnules. Apparently, all of the fronds of Hopetedia were typically fertile (e.g., Figs. 1–3), and sterile lobes (e.g., Fig. 20) are very rare. In fact, most specimens contain no fertile lobes whatsoever. Coniopterid sporangia are typically smaller and wedge shaped, rather than large and more circular as in Hopetedia (Fig. 16). Finally, the indusium of most coniopterids is cup-like and often possesses two flaps similar to those found in most extant members of the Hymenophyllum s.l. clade of the Hymenophyllaceae, rather than the Trichomanes-like, funneliform indusia of Hopetedia (Fig. 20). We suggest that these differences, as well as the hymenophyllaceous features detailed below, preclude assignment of Hopetedia praetermissa to Coniopteris or any other genus of the Dicksoniaceae, despite the superficial similarity. Several lines of evidence clearly support the affinity of Hopetedia praetermissa with the Hymenophyllaceae. The general frond morphology, consisting of three orders of segments with the ultimate order distinctly lobed (Figs. 2, 5), is common among extant forms. The narrow lamina and a single vein entering each ultimate lobe (e.g., Figs. 6–16, 19) are also common features. Considering the widespread occurrence of this basic morphology in several fern and Paleozoic and Mesozoic seed plant groups, however, a more critical point centers on the evidence for a membranaceous lamina, because this is one of the most distinctive features of the extant Hymenophyllaceae. Because the lamina of the Hopetedia frond is represented only by stains narrowly surrounding the rachises and veins, the thickness of the lamina cannot be conclusively determined. However, the extreme faintness of the laminar stain on the rock matrix, especially compared with other nonmembranaceous ferns from the same locality (Fig. 1), suggests that the lamina of Hopetedia was very thin. When viewing the figures in this paper, it must be kept in mind that the laminae are obvious only because the specimens were wetted and photographed with high contrast film. Also, polarizers were placed in front of the light sources and camera lens and rotated until maximum contrast was achieved. Without such treatment, the only plant parts relatively easily observed are the rachises, veins, and annuli, since these are usually highlighted by fragments of dark, organic residue (e.g., Figs. 6–11).
The rather lax organization of the Hopetedia frond is common among epiphytic members of extant Hymenophyllaceae. Many epiphytic forms also possess filiform rhizomes and are often rootless. Although the rhizomes of Hopetedia are known only from one small fragment that shows attachment (Fig. 3, arrow), and a few associated possible rhizome fragments (Fig. 4), they appear fairly robust and possess roots (Fig. 4). These facts, together with the nearly exclusive association with other ferns assumed to be terrestrial, indicate a probable terrestrial habit for Hopetedia.
Additional evidence for the hymenophyllaceous affinity of Hopetedia comes from the soral and sporangial structures. The marginal sori are shared with all extant members of the family. Particularly compelling is the funneliform structure of the indusium (Fig. 20, arrows), which is very similar to that found in many extant members of the Trichomanes s.l. clade, and a few members of the Hymenophyllum s.l. clade (Pryer et al., in press
). The indusia of Hopetedia occur singly at the apex of nearly every ultimate pinnule lobe and are apparently immersed in the lamina. Each sorus extends terminally from the single vein entering the lobe (epitactic) (Figs. 15–19). This combination of soral characters is widespread throughout the Trichomanes s.l. clade (Dubuisson, 1997
). The rather large, ovoid sporangia (Figs. 16–18) are close to the upper size limit seen in extant forms (0.5 mm). The short sporangial stalks and complete annuli (Figs. 16, arrow, 17) are typical for the Hymenophyllaceae. A more problematic issue is the nature of the receptacle in Hopetedia (Fig. 18, arrows), which is shorter than that of most modern forms. Undoubtedly as a consequence of the short receptacle length, the sporangial number per sorus is low (typically 5–8), and there is no evidence of basipetal development, as all sporangia appear to be at the same stage of development. Although this could be considered evidence against hymenophyllaceous affinity for Hopetedia, it must be emphasized that the morphology of the sori has been somewhat obscured by preservational factors. Furthermore, fairly short receptacles with small numbers of sporangia are known within the family, most commonly within members of the Hymenophyllum s.l. clade. Spores in situ would be most helpful in further supporting a hymenophyllaceous affinity, but these have not been discovered to date.
Taken together, we suggest that the evidence presented here is sufficient to confidently place Hopetedia praetermissa in the Hymenophyllaceae. However, since nearly all of the observable characters are widely distributed within the family, it remains impossible to definitively place Hopetedia with any particular extant subgroup. As mentioned above, the funneliform indusium is very similar to that seen in most members of the Trichomanes s.l. clade. Furthermore, the probable terrestrial growth habit of Hopetedia is also consistent with placement in the Trichomanes s.l. subclade, because extant members of the Hymenophyllum s.l. subclade are exclusively epiphytes. However, the very short receptacle is more consistent with placement in the Hymenophyllum s.l. clade.
Although relative stratigraphic position is considered suspect as direct evidence for phylogenetic position by most systematists, the status of Hopetedia as the oldest definitive fossil representative of the family by a considerable margin may indicate that the mosaic of Hymenophyllum s.l. and Trichomanes s.l. soral characters occurring in this fossil represents an ancestral condition. Perhaps the relatively short receptacle and tubular indusium characters later became sequestered within separate Hymenophyllum s.l. and Trichomanes s.l. descendant subclades, respectively. Although this scenario is possible, a recent rbcL phylogeny by Pryer et al. (in press)
has demonstrated that a tubular Trichomanes-type indusium does occur in some extant members of the Hymenophyllum s.l. clade. It is notable in this context that the New Zealand endemic form Cardiomanes reniforme, which has a Trichomanes-like indusium and receptacle and has historically been included in the genus Trichomanes, resolved as the basal taxon of the Hymenophyllum s.l. clade in the Pryer et al. (in press)
rbcL analysis. Although the bootstrap support for this phylogenetic placement is relatively weak (BS 
67), this result may indicate that the Trichomanes-like tubular indusium, which occurs in the Triassic form Hopetedia described here, is plesiomorphic for the family. However, Cardiomanes also possesses a long receptacle. Therefore, short receptacles, like those of Hopetedia, are apparently not the primitive character state for the family despite the antiquity of this fossil. This underscores the fact that the mosaic soral morphology of Hopetedia, or any of its other morphological features, cannot be definitively considered as primitive within the Hymenophyllaceae until this fossil is well enough understood to be included along with extant forms in an explicit phylogenetic analysis.
Suspected Paleozoic and Mesozoic Hymenophyllaceae
Numerous fossil specimens from the Paleozoic and Mesozoic have historically been placed in the genera Hymenophyllites, Trichomanides, and Trichomanites; however, a detailed evaluation of each is not possible here. Many of these have actually been rejected earlier (e.g., Seward, 1910
), but there are a few remaining and subsequent reports of possible fossil Hymenophyllaceae. In the following discussion we have concentrated on those reports claiming the presence of hymenophyllaceous fertile structures—a requirement we consider essential for assignment to the family.
Paleozoic records
Despite the large number of Paleozoic leaves historically assigned to Hymenophyllites, we believe that the only potential hymenophyllaceous taxon is Hymenophyllites quadridactylites from the Upper Carboniferous of France. Based on figures in Zeiller (1886
, pl. VIII, 1–3), this species is represented by rather robust fronds of at least four orders of division. More significant are the accompanying drawings (pl. VIII, 1a, 2a, 3a–c) of fertile pinnae showing clusters of sporangia just beyond the tip of the ultimate pinnule lobes. In some cases, these clusters are arranged in two elongate rows surrounding an area devoid of sporangia. Based on his personal communications with Zeiller, Kidston (1923
, p. 279) notes that Zeiller claimed the space was occupied by traces of a "columella" to which the sporangia were apparently attached; however, the presence of such a structure has never actually been documented. Furthermore, there is no convincing evidence for a tubular or bivalved indusium, and some of the sporangia appear more wedge-shaped, similar to those of Coniopteris. The robust fronds show no evidence of a membranaceous characteristic. We therefore concur with Kidston's (1923
, p. 279) assessment that "It will therefore be seen that though the fructification of Sphenopteris quadridactylites Gutbier has some resemblance to that of Hymenophyllum Sm., there are several points on which additional information is necessary before it can be definitely referred to the Hymenophyllaceae."
Among permineralized Paleozoic ferns, the family Psalixochlaenaceae (Holmes, 1977) has sometimes been allied with the Hymenophyllaceae based on gross frond morphology, anatomical features including a protostele, and marginal sporangia. However, based on figures in Holmes (1981, pl. III, 16, 17, pl. V, 30, 31, pl. VI, 32–36), the sporangia are born in slightly concave regions at the tips of the fertile lobes that are not similar to those found in any extant members of the Hymenophyllaceae. Furthermore, a recent and extensive morphology-based phylogenetic analysis of fossil and recent ferns has not supported the hypothesis of close link between the Psalixochlaenaceae and Hymenophyllaceae (Rothwell, 1999
).
Mesozoic records
Some workers have claimed that the oldest definitive fossil record of the Hymenophyllaceae is Trichomanides laxum (Tenison-Woods, 1883)
supposedly from the Jurassic of Queensland Australia (e.g., Tryon and Tryon, 1982
, p. 98). This assessment is based on the acceptance of Tenison-Woods's (1883)
suggestion that this fossil exhibits a creeping rhizome and a Trichomanes-like indusium and receptacle. We have examined the type material of "Trichomanides laxum" (Fig. 2 in Tenison-Woods, 1883)
and have determined that this fossil is actually a poorly preserved specimen of the corystospermalean seed fern leaf Dicroidium and is therefore Triassic in age. Based on the narrow lamina, this fossil is particularly close to the species Dicroidium elongatum. The structure interpreted as a rhizome by Tenison-Woods (1883)
is the frond rachis. We were unable to find any structure resembling an indusium. We suspect that parts of the leaf from which the lamina was lost, leaving only the narrow midrib behind, were mistaken for an elongate Trichomanes receptacle. Anderson and Anderson (1983)
have shown that Trichomanides spinifolium (Tenison-Woods, 1883)
is a Dicroidium elongatum leaf. A more detailed account of this material is under preparation for publication elsewhere.
With the exception of Hopetedia praetermissa described here, plausible records of the Hymenophyllaceae do not appear until the Jurassic. The most promising example is Hymenophyllites macrosporangiatus from the Jurassic of Russia (Vakhrameev, 1952
). The leaves of this form are somewhat linear and narrow, and the sporangia occur in large, dense clusters on the margins of the terminal segments (Vakhrameev, 1952
, pl. I, 1–9, pl. II, 1). These clusters extend a considerable distance out from the margins indicating an elongate receptacle, although none is visible. The sporangia are large and contain in situ spores (Fig. 20 in Vakhrameev, 1952
). Although this material does not clearly reveal the indusial morphology, the leaf, sporangia, and spore morphologies are consistent with placement in the Hymenophyllaceae.
Two reports of possible Cretaceous Hymenophyllaceae deserve mention. Eogonocormus cretaceum and E. linearifolium, from the early Cretaceous of northeastern China, are small, thalloid forms with creeping rhizomes (Deng, 1997
). These features, along with the marginal sori with in situ spores, borne on fan-like pinnule lobes make these fossils convincing candidates for inclusion in the Hymenophyllaceae. A less convincing example is Hymenophyllum priscum from the Upper Cretaceous of Chile (Menendez, 1972
). The illustrated material shows frond fragments with thickenings at the margin that are interpreted as possible bivalved indusia. However, we believe a hymenophyllaceous soral morphology has not been convincingly demonstrated.
Conclusions
Based on the evidence presented here, we suggest that the new Upper Triassic fossil Hopetedia praetermissa is the oldest convincing representative of the Hymenophyllaceae. It is, in fact, one of the few convincing fossils of the family from any time period. Despite the antiquity of Hopetedia, however, it cannot be assumed that the features of this fossil are necessarily primitive for the Hymenophyllaceae for several reasons. Most importantly, the Hymenophyllaceae should predate the Triassic considerably, possibly as early as the Permian or Carboniferous based on those phylogenetic analyses that resolve the family as the most basal filicalean taxon (e.g., Pryer, Smith, and Skog, 1995
, analysis 1B) or branching immediately subsequent to the Osmundaceae (e.g., Pryer, Smith, and Skog, 1995
, analyses 1A and 1C; Stevenson and Loconte, 1996
). Also, although the tubular Trichomanes-like indusial morphology of Hopetedia agrees with that determined to be plesiomorphic for the family based on the Pryer et al. (in press)
rbcL phylogeny, the short receptacle is more like that of most members of the Hymenophyllum clade and is therefore presumably derived. Finally, the homoplastic acquisition of a tubular, Trichomanes-like indusium within the family, as demonstrated by Pryer et al. (in press)
, warns against a naive acceptance of the Hopetedia morphology as ancestral to all of the modern forms based only on the age of this fossil and the mosaic condition of the soral morphology. This is especially true considering that our present understanding of Hopetedia is insufficient to justify including it in a rigorous phylogenetic analysis. In any event, Hopetedia praetermissa provides some paleobotanical support for the relatively basal phylogenetic position of the Hymenophyllaceae within the filicalean ferns and offers new hope and impetus for uncovering the fossil history of this fascinating family.
|
FOOTNOTES |
|---|
5 Author for reprint requests (phone: 334-460-7528; FAX: 334-414-8220; baxsmith{at}jaguar1.usouthal.edu)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Bower F. O. 1908 The origin of a land flora. McMillan, London, UK
———. 1923–1928 The ferns (Filicales). Cambridge University Press, Cambridge, UK
Collinson M. E. 1996 "What use are fossil ferns?"—20 years on: with a review of the fossil history of extant pteridophyte families and genera. In J. M. Camus, M. Gibby, and R. J. Johns [eds.], Pteridology in perspective, 349–394. Royal Botanic Gardens, Kew, UK
Deng S. 1997 Eogonocormus—a new early Cretaceous fern of Hymenophyllaceae from China. Australian Systematic Botany 10: 59-67
Dubuisson J.-Y. 1997 Systematic relationships within the genus Trichomanes sensu lato (Hymenophyllaceae, Filicopsida): cladistic analysis based on anatomical and morphological data. Botanical Journal of the Linnean Society 123: 265-296[CrossRef]
Gensel P. 1986 Plant fossils of the Upper Triassic Deep River Basin. In P. J. W. Gore [ed.], Depositional framework of a Triassic rift basin: the Durham and Sanford Sub-Basins of the Deep River Basin, North Carolina, 82–86, Field Trip, 3. Society of Economic Paleontologists and Mineralogists, Tulsa, Oklahoma, USA
Harris T. M. 1961 The Yorkshire Jurassic flora, vol. I, Thallophyta—Pteridophyta. Trustees of the British Museum (Natural History), London, UK
Hasebe M. P. G. Wolf K. M. Pryer K. Ueda M. Ito R. Sano G. J. Gastony J. Yokayama J. R. Manhart N. Murakami E. H. Crane C. H. Haufler W. D. Hauk 1995 A global analysis of fern phylogeny based on rbcL nucleotide sequences. American Fern Journal 85: 134-181[CrossRef]
Holms J. C. 1977 The Carboniferous fern Psalixochlaena cylindrica as found in the Westphalian A coal balls from England, Part I, Structure and development of the cauline system. Palaeontographica 164B: 33-75
———. 1981 The Carboniferous fern Psalixochlaena cylindrica as found in Westphalian A coal balls from England, Part II, The frond and fertile parts. Palaeontographica 176B: 147-173
Huber P. S. G. Lucas A. P. Hunt 1993 Revised age and correlation of the Upper Triassic Chatham Group (Deep River Basin, Newark Supergroup), North Carolina. Southeastern Geology 33: 171-193
Iwatsuki K. 1977 Studies in the systematics of filmy ferns. III. An observation on the involucres. Botanical Magazine 90: 259-267[CrossRef]
———. 1990 Hymenophyllaceae. In K. U. Kramer and P. S. Green [eds.], The families and genera of vascular plants, vol. I, Pteridophytes and gymnosperms, 157–163. Springer Verlag, Berlin, Germany
Jongmans W. S. J. Dijkstra 1960 Fossilium Catalogus. II. Plantae. Filicales, Pteridospermae, Cycadales, part 43. Uitgeverj Dr. W Junk, ‘s-Gravenhage, The Netherlands
———, and ———. 1965 Fossilium Catalogus. II. Plantae. Filicales, Pteridospermae, Cycadales, part 63. Uitgeverj Dr. W Junk, ‘s-Gravenhage, The Netherlands
Kidston R. 1923 Fossil plants from the Carboniferous rocks of Great Britain, part 4. Memoirs of the Geological Survey of Great Britain. Palaeontology 2: 275-376
Kilpper K. 1964 Über eine Rät/Lias-Flora aus dem Nördlichen Abfall des Alburs-Gebirges in Nordiran, Teil 1, Bryophyta und Pteridophyta. Palaeontographica 114B: 1-78
Litwin R. J. S. R. Ash 1993 Revision of the biostratigraphy of the Chatham Group (Upper Triassic), Deep River Basin, North Carolina, USA. Review of Palaeobotany and Palynology 77: 75-95
Menendez C. A. 1972 Paleophytologia Kurtziana III. 8, La flora del Cretacico superior de Cerro Guido, Chile (1–2). Ameghiniana 9: 209-219
Pryer K. M. A. R. Smith J. E. Skog 1995 Phylogenetic relationships of extant pteridophytes based on evidence from morphology and rbcL sequences. American Fern Journal 85: 205-282[CrossRef][ISI]
———, ———,J. S. Hunt J.-Y. Dubuisson In press rbcL data reveal two monophyletic groups of filmy ferns (Filicopsida: Hymenophyllaceae). American Journal of Botany.
Rothwell G. W. 1999 Fossils and ferns in the resolution of land plant phylogeny. Botanical Review 65: 188-218
Seward A. C. 1910 Fossil plants, a text-book for students of botany and geology, vol. II. Cambridge Biological Series, Cambridge University Press, Cambridge, UK
Stevenson D. W. H. Loconte 1996 Ordinal and familial relationships of pteridophyte genera. In J. M. Camus, M. Gibby, and R. J. Johns [eds.], Pteridology in perspective, 435–467. Royal Botanic Gardens, Kew, UK
Tenison-Woods J. E. 1883 On the fossil flora of the coal deposits of Australia. Proceedings of the Linnean Society of New South Wales 8: 1-131
Tidwell W. D. S. R. Ash 1994 A review of selected Triassic to Early Cretaceous ferns. Journal of Plant Research 107: 417-442[CrossRef][ISI]
Tryon R. M. A. F. Tryon 1982 Ferns and allied plants with special reference to tropical America. Springer Verlag, Berlin, Germany
Vakhrameev V. A. 1952 Stratigraphy and fossil flora of Jurassic and Cretaceous deposits of Vilyui Trough and adjacent part of near Verkhoyansk Foredeep. Regionalnaya stratigrafya SSSR (in Russian), vol. 1. Academy of Sciences of the USSR. Order of the Red Banner of Labour Geological Institute, Moscow, Russia
Zeiller R. 1886 Flore fossile du bassin houiller de Valenciennes. Quantain, Paris, France
This article has been cited by other articles:
|
|
 |
|
  K. M. Pryer, E. Schuettpelz, P. G. Wolf, H. Schneider, A. R. Smith, and R. Cranfill Phylogeny and evolution of ferns (monilophytes) with a focus on the early leptosporangiate divergences Am. J. Botany, September 1, 2004; 91(10): 1582 - 1598. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 MATONIACEOUS FERNS (GLEICHENIALES) FROM THE MIDDLE TRIASSIC OF ANTARCTICA Journal of Paleontology, January 1, 2004; 78(1): 211 - 217.
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
