| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biology, Indiana University Southeast, New Albany, Indiana 47150
Received for publication July 16, 1998. Accepted for publication March 25, 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSSYSTEMATICS AND DESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
Key Words: gigantopterids • Gigantopteris; • hooks • lianas • paleoecology • Permian • stems • Vasovinea; • vessels
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICS AND DESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
). The distinctive characteristics of gigantopterids are their large compound or simple leaves, which are variable in morphology, ranging from oblong to roughly round shape, entire to toothed margin, and simple to complex reticulate venation. Conventionally the group has been classified as ferns (Schenk, 1883) or seed ferns (White, 1912
; Asama, 1959
; X. Li and Yao, 1983
). In terms of relationships to other plants, Asama (1974, 1982, 1988)
proposed that the simple-leafed gigantopterids could have given rise to the angiosperms, while X. Li and Yao (1983)
interpreted their reconstruction of the reproductive organs of Gigantonoclea fukienensis as being "parti-angiosperms" in nature. Despite the distinctive leaves, anatomically preserved reproductive organs have not been found so that the systematics of the group has been uncertain.
Recently, some permineralized gigantopterid leaves and axes have been reported with many anatomical characteristics that can help in determining the relationships of gigantopterids to other vascular plants. The permineralized leaves of the gigantopterid Delnortea abbottiae, reported from Texas, USA, are suggested as being structurally similar to gnetophytes (Mamay et al., 1988
). In contrast, the permineralized leaves of Gigantonoclea guizhouensis, from the Upper Permian in Guizhou, China, exhibit many features similar to those of angiosperms (H. Li and Tian, 1990
; H. Li et al., 1994
). Two types of permineralized gigantopterid stems also are found associated with the Guizhou gigantopterid leaves (H. Li, Taylor, and Taylor, 1992, 1993
). One type is the prickly Aculeovinea yunguiensis (H. Li and Taylor, 1998
), which is considered a seed plant. Another type containing vessels has been reported preliminarily (H. Li, Taylor, and Taylor, 1996
).
Before the report by H. Li, Taylor, and Taylor (1996)
, the fossil record of vessels could be traced back to the Early Cretaceous. In extant plants, vessels consist of vertically linked tracheary elements with perforate end walls so that they conduct water much more efficiently than tracheids with imperforate end walls. Vessels are commonly found in angiosperms and gnetophytes, with rarer occurrences in non-seed plants (see Discussion). Vessels in living seed plants have been classified in two types, i.e., foraminate vessels in gnetophytes and scalariform ones (and/or their derived simple form) in angiosperms (Bailey, 1944
; Carlquist, 1992, 1994, 1996a)
. Vasovinea tianii is different from the above two types, possessing unique, foraminate-like vessels in the secondary xylem and a possible scalariform-reticulate vessel in the metaxylem. Therefore, the discovery of vessels in gigantopterids is important not only as the earliest fossil record of vessels, but also in analyzing the systematic relationships and the ecological aspects of the group.
To complement the preliminary study (H. Li, Taylor, and Taylor, 1996
), we now provide a comprehensive description of the permineralized, vessel-bearing gigantopterid stems and establish them as a new taxon, Vasovinea tianii Li et Taylor gen. et sp. nov. With several lines of evidence from both permineralized and compression specimens, we also demonstrate that the new taxon is a member of the gigantopterids, reconstruct it together with Gigantopteris-type leaves, analyze its possible habit, and briefly discuss its systematic relationships to seed plants.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICS AND DESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
and H. Li and Taylor (1998)
.
Permineralized specimens from five limestone samples (L9407, PLY02, PLY03, PLY04, and L9414), and compressed specimens from three mudstone samples (L9426, L9448, and L9449) were used. All samples contain gigantopterid foliage and stem(s) in each, and some have additional compound hooks as well as other structures, although not all of these organs from each of the samples are figured in this paper. Samples PLY03 and PLY04 are small, but each contains a Vasovinea tianii stem and a piece of Gigantopteris-type leaf. Sample L9407 has numerous pieces of Gigantopteris-type leaves and several pieces of V. tianii stems. PLY02 is a large sample, weighing more than 8 kg, and contains hundreds of pieces of permineralized (usually gigantopterid) plant organs, including at least two types of gigantopterid leaves (Gigantopteris and Gigantonoclea), two types of gigantopterid stems (V. tianii and Aculeovinea yunguiensis; H. Li and Taylor, 1998
), and some undescribed reproductive organs. Specimens of Vasovinea and Gigantopteris in this sample are relatively fewer in number compared to those of Aculeovinea and Gigantonoclea.
The permineralized stems were sectioned and prepared using the well-known cellulose acetate peel technique (Phillips, 1976
). Areas with interesting structures from the peels were trimmed, cleaned, and mounted on microscope slides for further observations. Occasionally specimens were peeled in both the transverse and longitudinal sections to show the structural correlation of both sections (Figs. 1, 6). Specimens were photographed with an MP-4 camera using 4 x 5 film or with an Olympus 35-mm camera mounted on an Olympus Steroscan dissecting microscope. To prepare scanning electron microscope (SEM) samples, several permineralized wood pieces with vessels were etched with a 1% HCl solution for 
20 min so that the diluted solution could dissolve the calcium carbonate slowly to expose cellular structures, e.g., the perforation plates (Figs. 16–19). Then they were mounted and sputter coated with gold before being examined with SEM. The compressed specimens were directly photographed with a Polaroid MP-4 camera.
|
|
|
SYSTEMATICS AND DESCRIPTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICS AND DESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
Family
Gigantopteridaceae Koidzumi (1936)
Species
Vasovinea tianii gen. et sp. nov.
Generic and specific diagnosis
Stems slender, 1 cm in diameter, bearing compound hooks, prickles, and glandular and tendril-like trichomes. Cortex with paired vascular traces and sparganum structure in the outer part. Eustele with parenchymatous pith and mesarch protoxylem, consisting of tracheids with annular, helical, or helical-scalariform thickenings. Centripetal metaxylem tracheids commonly consisting of one to two layers, while centrifugal metaxylem usually consisting of two to three layers of tracheary elements; metaxylem tracheary elements with scalariform and/or transversely elongated bordered pits on their lateral walls. Outer portion of the secondary xylem consisting of radial files of tracheids and smaller vessels. Lateral walls of tracheary elements in secondary xylem exhibiting multiseriate, alternate, and transversely elongate to circular bordered pits. The inner portion of the secondary xylem consisting of large vessels with diameter increasing from 150 to 250 (up to 500) µm towards the primary xylem; vessel elements vertically connected by foraminate-like perforation plates on the long, inclined (usually) to short, almost horizontal (occasionally) end walls, each plate with multiseriate, alternate, obliquely elliptical to circular pores without borders. Homocellular, heteroseriate rays occurring every one to three radial tracheary files, consisting of uni- to bi-seriate xylic rays and multiseriate medullar rays between xylem segments.
Holotype
Slides L9407-C-B2, L9407-C-B16, and L9407-D-T2. Figs. 1–4, 6.
Paratypes
Slides PLY02-C10-1-1, PLY02-E-1, PLY03–01, PLY03–06, PLY03–07, PLY03–11, PLY03–34, and PLY04-B; Specimens PLY02 and PLY03. Figs. 5, 7–20, 29–30.
The slides and specimens of both the holotype and paratypes have been deposited in the National Museum of Plant History of China at the Institute of Botany, Chinese Academy of Sciences, Beijing, China.
Etymology
The generic name is composed of vaso- ([L] = vessel) and vinea ([L] = vine) indicating a liana stem with vessels. The specific name is proposed in honor of Professor Baolin Tian, the Beijing Graduate School, China University of Mining and Technology, for his contribution to the study of the Gigantopteris flora of western Guizhou, China.
Locality
Yueliangtian Coal Mine, Panxian County, Guizhou Province, China.
Stratigraphic occurrence
Lower and Upper Xuanwei Formations, Upper Permian.
Age
Longtanian-Changxingian, Late Permian.
Description
The stems are usually <1 cm in diameter and up to 5 cm in length. One stem measured 4 x 6 mm in transverse section (Figs. 1–3, 6), while others were compressed and calculated to be 1 cm in diameter (Figs. 5, 29–30). Well-preserved stems have an epidermis and hypodermis. The epidermis consists of one layer of small cells that have relatively thick walls. The layer is 10 µm thick and its cells are 10 µm wide and 8 to 20 µm long. Beneath the epidermis are 2–4 layers of hypodermal cells that are commonly 40–50 µm long and 15–40 µm in diameter, and densely arranged with the larger cells occurring to the outside and smaller ones to the inside. Attached to the stems are compound hooks (see below) and a variety of appendages including prickles and glandular and tendril-like trichomes.
Appendages
The appendages are epidermal-cortical outgrowths and lack any vascular tissue. They are commonly found on stems, but some smaller appendages can be found on the compound hooks (Fig. 1, right arrowhead). The prickles on the stems are usually 450 µm (up to 1000 µm) long and 450 µm (up to 550 µm) wide at their bases, but a narrower one found near the base of a compound hook is only 250 µm wide at the base (Fig. 2, right arrow). Each prickle is covered by the epidermis and internally composed of vertically elongate parenchyma cells, 20–30 µm in diameter and 40–50 µm long. A few parenchyma cells beneath the epidermal cells are narrower and have thickened cell walls. Some hypodermal cells (dark colored) beneath the base separate the prickle from the inner part, suggesting that the prickle is epidermal and cortical in origin. Structurally, these prickles are nearly identical to those of Aculeovinea yunguiensis (H. Li and Taylor, 1998
), but smaller in size.
Glandular trichomes can be found on both the stems (Figs. 26, 28) and the compound hooks. Generally, they are 500–2000 µm long and 500–1000 µm in diameter at their bases. Their basal parts are similar to the prickles in terms of their structure and the epidermal and cortical origin. However, their apices consist of either several large cells (Fig. 28) or a single larger oval cell, 200 µm in length and 100 µm in diameter, at the tip (Fig. 26).
|
) are long and curved in shape. These trichomes are 300 µm in diameter and can be several millimetres long. They contain similar-shaped parenchyma cells, with the inner cells slightly enlarged. They may be preserved as tubes (Fig. 30, arrowhead) when the inner parenchyma cells are no longer preserved. In other sections, these trichomes often appear as wart-like structures extending from the stems (Fig. 5, center) or circular rings (Fig. 5, arrowheads) embedded in the surrounding sediments.
Cortical histology
Beneath the hypodermis, in the well-preserved stems, the cortex consists of a sparganum structure (i.e., vertically parallel sclerenchymatous strands alternating with vertical parenchymatous tiers; see Taylor and Taylor, 1993
) towards the outer side (Fig. 3, bottom) and a wider parenchymatous zone towards the inside (Figs. 2, 3). The sclerenchyma cells are 20 µm in diameter and 1000–3000 µm long (Fig. 6, right). The cortical parenchyma cells are roughly cuboidal, 45–70 (up to 95) µm in dimension, and vertically linked (Fig. 6, arrow). They are often damaged or crushed, leaving empty spaces in the cortex (Figs. 3, 5, 6).
|
Along the periphery of the pith, primary xylem bundles and a few solitary tracheids are embedded among the parenchyma cells. Mesarch protoxylem strands can be found in the centers of some of the large bundles (Fig. 4, arrowhead at the bottom), but they are frequently difficult to distinguish. The protoxylem tracheids can be narrower than 20 µm in diameter with annular to helical or helical-scalariform thickenings (Fig. 9). The centripetal metaxylem usually is one to two cells wide (Fig. 4), and the tracheids commonly are less than 50 µm in diameter and have scalariform (Fig. 9, left lower part) or 2–3 rows of transversely elongated, bordered pits (Fig. 9, arrowhead). The centrifugal metaxylem commonly is 2–3 cells wide (Figs. 4, 8 [left]), with the inner tracheary elements being narrow and having scalariform bordered pits. The outer elements usually are large, up to 100 µm in diameter, and have up to five rows of transversely elongated, bordered pits on the lateral walls. The elongated pits are arranged into obliquely opposite rows when two or three rows are present. In some of those with five rows, the pit openings are only 2.5 µm high but can be up to 16 µm wide. However, as the row number increases, the pits appear to be alternately arranged and pit openings become narrower.
One centrifugal metaxylem element (Figs. 8 [arrow], 10), 50 µm in diameter, has two types of wall structures. The tip of the tracheary element forms a long, highly inclined end wall that has a scalariform-reticulate structure (Fig. 10, upper arrow). The lower part shows a small broken piece of its lateral wall with oval bordered pits (Fig. 10, lower arrow). Tracing downwards on the series of peels, scalariform bordered pits are found on the lateral wall of the same element. Therefore, this tracheary element has scalariform and laterally elongated bordered pits on the lateral walls. However, because there are no borders around the scalariform-reticulate structure and no primary walls appeared within the structure, the oblique end wall appears to be perforated. In other words, this element possibly could be a vessel member.
The secondary xylem has tracheary elements arranged into radial files, each with an inner portion of large vessels (1–5 vessels wide) and an outer portion of smaller tracheary elements. Several stems show an inner portion that is about 3–5 vessels wide. These vessels increase in diameter from less than 150 µm to 250 µm towards primary xylem (Figs. 2–5). The vessel outlines are round (Figs. 3, 5, 14) or squarish (Figs. 2–5) in transverse section. In transverse section, the tracheary elements may appear as tangential rows of cells of similar size and shape (Fig. 4). In addition, in tangential section (Figs. 6 [left], 12) the end walls appear at similar levels. One stem has large vessels arranged into radial files that are tangentially only 1–2 vessels wide, with each vessel up to 500 µm in diameter, but outer elements with a much smaller diameter (Figs. 15–16). This latter specimen may represent a different species, but there is insufficient material to establish a separate species at this time.
Overall, vessel elements can be up to 4500–5000 µm long. Their planar end walls vary from short (Fig. 12, right lower corner) to very long (up to 1200 µm long; Fig. 12, left), from oblique (Figs. 12, 14 [upper two vessels], 16 [right arrow]) to almost horizontal (Fig. 14, lower left vessel). Most of the end wall is perforated with multiseriate (up to 12 rows) pores (Figs. 7 [upper right], 8 [lower right], 11). In the central area, most pores are large and roughly circular, 8 x 10 µm (Figs. 7, 11) and up to 13 x 15 µm (Figs. 18 [middle left], 19 [right]), but some pores are much larger and elliptical, up to 13 x 21 µm (Fig. 19, lower left). Although those pores in the central area are completely perforated without remains of border membranes, the pores in peripheral regions are smaller and exhibit an incompletely dissolved primary cell wall (Fig. 18, arrow) that is smooth and shows no trace of margo threads or tori.
In transverse section, the outer zone, up to 20 cells wide (6–14 cells wide in Figs. 2–4), contains tracheary elements that are usually 40 x 50 µm (a few up to 120 µm) in dimension and arranged in radial files alternating with rays at every one to three files. One of the smaller tracheary elements, 120 µm in diameter, has a planar end wall (Fig. 16, left arrow) with incomplete pores (Fig. 17). The plate exhibits obliquely elongated pits, which have smooth rims (Fig. 17, left arrow), and a planar, partially dissolved primary wall (Fig. 17, upper arrow). Thus, it has the same type of perforation pattern as in the large vessels, though the dissolution is not as complete. In other words, some tracheary elements in the outer zone might be vessel elements, while others could be tracheids with imperforate end walls. The lateral walls of both large and small tracheary elements exhibit the same type of alternately arranged, multiseriate, bordered pits (Figs. 7 [lower right], 11 [lower right], 13, 20). These bordered pits are obliquely elliptical, 3 x 8 µm (Fig. 20, right upper), to nearly circular, 6–8 µm in diameter (Fig. 13).
Rays consist of homocellular parenchymatous cells, which are usually 8–15 µm wide (Fig. 4), 15–25 µm long, 25–50 µm high (Figs. 7 [at left], 13 [at right], 20 [at left]), and each cross-field has 
3–9 obliquely arranged pits (Fig. 7, at left). The xylic rays are well preserved, usually one to two cells wide (Figs. 4, 14) and as tall as 60 cells. Between the cauline xylem segments, there are wedge-shaped gaps (Figs. 2–6, 15 [arrows]) sometimes containing cellular remains that are over three cells wide and could be medullary rays.
The region of the cambium and inner phloem appears as a zone of amorphous cellular remains or as an empty gap, due to the poor preservation. In several transverse sections, there is a narrow, dense cellular zone, surrounding the secondary xylem (Figs. 2–3), composed of 2–4 small cells, which are usually rounded rectangular in shape, 15 µm in diameter, and have lumens commonly 6–8 µm in diameter. In tangential section, they are elongate, but their lengths are unknown because of their poor preservation. These cells might be the remains of the phloem fibers.
A pair of vascular traces (Fig. 2, "T"), each 750 µm in diameter, are found in the cortex, outside a gap (Fig. 2, arrow) between two cauline segments. The two traces are of similar size and appear to originate from within the gap, resulting in a unilacunar two-trace pattern. The trace on the left appears to have been just pinched off from the small-celled, side part of a main cauline xylem segment. The right trace is bilaterally symmetrical and ring-shaped with its abaxial side wider than the adaxial side (only 2–3 cells wide; Fig. 4, upper). The abaxial side has seven or more tracheids, each up to 40 µm in diameter, radially arranged into files. These tracheids represent the primary and, possibly, secondary xylem. The central area of the vascular ring is usually empty, occasionally containing the remains of crushed parenchyma cells. This type of trace structure (Figs. 2, 4) may represent the vascular traces of a leaf.
Compound hook structure
On the right side of Fig. 1 is a narrow branching axis, or a "compound hook," which is 4 cm long and has two pairs of opposite branches. Note that this figure includes both transverse and longitudinal sections of a stem (left) and the compound hook (right). The attachment and branching pattern of the compound hook were determined by examining the part and counterpart and a series of peels. The main axis of the compound hook, 3 mm in diameter, extends 1 cm from the stem (in the box) through the matrix and then branches the first time to form a pair of lateral hook tips. The lower one was obliquely cut through and left an oblique section, while the upper one was preserved in the counterpart (Fig. 1, large arrow). Then, the main axis becomes thinner (1.6 mm in diameter) and extends 1.6 cm and branches again, resulting in a second pair of lateral hook tips and a terminal tip, each 0.8 mm in diameter at their bases. With additional peels, the longitudinal section of the main axis of the compound hook and an oblique section of the upper hook tip of the second pair were exposed. These lateral and terminal hook tips curve backwards and are arranged in roughly the same plane. More details of the attachment of the compound hook to the main axis (Fig. 3, lower right) are shown in a transverse section which is 5 mm lower from the section in Fig. 2. Here, one large vascular trace (Fig. 3, "H") is relatively well preserved in an oblique section of the compound hook axis. This vascular trace is crescent shaped, bilaterally symmetric, and opens towards its adaxial side where the parenchyma cells occur.
As in the stem, the main axis of the compound hook also has sparganum cortex. The tracheids in this main axis are usually 20–45 µm in diameter and have helical thickening, but a few tracheids are 15–23 µm in diameter and have smaller, scalariform to transversely elongated, bordered pits. Distally, the vascular cells gradually become reduced in size and number and eventually vanish and leave only a hollow center surrounded by densely packed, thick-walled parenchyma cells, as seen in better preserved, dispersed hooks, which are 0.8 mm in diameter at their basal parts (Figs. 23, 24).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICS AND DESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
). We have changed the term because they lack vascular tissues, and their basal parts appear to be similar to the glandular trichomes in terms of their structure and epidermal and cortical origin.
The vessels in living plants are widely believed to have evolved from vertically linked tracheids by the dissolution of the primary cell walls between pit pairs on their end walls. One type of vessel has foraminate perforation plates, which develop by the loss of the pit membranes between circular bordered pit pairs. This type is characteristic of gnetophytes, whose circular bordered pits usually have pit membranes with tori and margo threads (Carlquist, 1996a, b, c
), and pores with the remains of the circular borders that may be slightly raised. Another type of vessel has scalariform perforation plates, which develop by the loss of the pit membranes (that have a uniform construction with micropores) between scalariform bordered pit pairs on the highly inclined, elongate, planar end walls (Carlquist, 1996a, b, c
). This type of vessel commonly occurs in some ferns (e.g., Carlquist and Schneider, 1997
; Schneider and Carlquist, 1998
) and basal angiosperms (e.g., Bailey, 1944
; Carlquist, 1992, 1994, 1996a
), but derived angiosperms may have vessels with simple perforations that evolved from the scalariform type.
The vessels of Vasovinea tianii in the secondary xylem also appear to develop by the dissolution of the pit membranes between pit pairs on the end walls, and the roughly round pores make the vessels superficially similar to the foraminate type rather than the scalariform type. However, on a closer examination, we found that the vessels of V. tianii are different from the foraminate type. First, the perforation plate is planar (e.g., Figs. 11, 16) and the number of pores is high (e.g., Figs. 7, 11, 14, 18, 19). Second, some pores are actually transversely elongate (not circular, e.g., Fig. 19) and lack borders (compared pores in Fig. 11 to bordered pits in Fig. 13). Third, within some pores the incompletely dissolved pit membranes are planar and lack tori and margo threads (Figs. 17, 18). Therefore we call these vessels foraminate-like to distinguish them from those with typical foraminate perforation plates. On the other hand, the possible vessel in the metaxylem appears to be similar to the scalariform type in general structure.
Another interesting and important structure is the compound hook. The hook tips are anatomically identical with the isolated, permineralized ones (Figs. 23, 24) associated with Vasovinea stems and Gigantopteris-type leaves. Also they are morphologically well matched with the compressed compound hooks (Figs. 21, 22) from the same locality. For example, the compressed compound hook shown in Fig. 21 has a 1.2 cm long basal axial part, a 1.4 cm long middle axial section, and five hook tips (the terminal tip is broken), with 2.6, 1.6 and 0.8 mm diameters, respectively. As with thorns and tendrils, the compound hooks could be modified from stems or leaves. We suggest that the compound hook of Vasovinea may be modified from a leaf, based on the following evidence: (1) it has a bilaterally symmetrical, crescent-shaped vascular trace (Fig. 3, "H"), (2) the hook tips are arranged in the same plane, and (3) the branching pattern is similar to pinnate leaf venation.
Affinity
Several lines of evidence show that Vasovinea tianii is a member of the gigantopterids. The most important evidence comes from the presence of a permineralized compound hook (Fig. 1), which is attached to the permineralized stem (Fig. 3, at right). This type of compound hook structure is unique to gigantopterids and has not been reported from other fossil or extant plants (see Menninger, 1970
). Significantly, the structure of the compound hook is identical with those found from the early Late Permian Gigantopteris floras of Shanxi (= Shansi in Halle, 1929
) Province, northern China (Halle, 1929
), and Fujian Province, southern China (Yao, 1983
). Halle (1929)
interpreted the hook-like structures as modified leaves of Gigantopteris nicotianaefolia. In 1995, one of us (Hongqi Li) borrowed and reexamined some of the Shanxi specimens from the Swedish Museum of Natural History. He confirmed that one of the gigantopterid leaves had thickened pinnate secondary veins that distally curved backwards to form hooks, but the basal part of each secondary vein still bore bilaterally a small amount of lamina. Thus, the Shanxi specimen clearly demonstrates a transitional type from a pinnately veined gigantopterid leaf to a compound hook with opposite hook tips. Our analysis of the permineralized compound hooks is in agreement.
Additional features showing the gigantopterid affinity include the presence of sparganum cortex, prickles, trichomes, and the configuration of the vascular trace. A similar sparganum cortex is commonly known in many Paleozoic seed ferns, such as the Carboniferous lyginopterids, medullosans, and callistophytes. Some species, such as Heterangium kentuckyensis (Pigg, Taylor, and Stockey, 1987
), Microspermopteris aphyllum (Pigg, Stockey, and Taylor, 1986
), and Callistophyton boyssetii (Rothwell, 1975, 1981
) even have similar prickle/trichome-like structures (called "cortical wings" in the first two references). However, among the plants reported from the Upper Permian flora of western Guizhou, a sparganum cortex and pickles/trichomes are only found in gigantopterids (see Tian et al., 1996
). These structures are characteristic of the leaf midribs of Gigantonoclea guizhouensis (H. Li et al., 1994
), the stems of Aculeovinea yunguiensis (H. Li and Taylor, 1998
), and Vasovinea tianii. The prickles were described as aculei (H. Li and Tian, 1990
) and spines (H. Li et al., 1994
) and have been redefined as prickles (H. Li and Taylor, 1998
) because of their epidermal and cortical origin, and lack of vascular tissue. Although the prickles in V. tianii are fewer in number, smaller, and contain fewer thick-walled parenchyma cells, they are similar to prickles of G. guizhouensis and A. yunguiensis in structure and origin. Finally, a sparganum cortex also is found around the vascular bundles in Gigantopteris-type leaves (Fig. 25).
The leaf traces (750 µm in diameter; Figs. 2 ["T"]. Fig. 4 [upper]) in the stem are identical to the vasculature (350 µm in diameter; Fig. 31, central part) in a section of a narrow, Gigantopteris-type leaf midrib, in terms of the radial rows of tracheids surrounding the triangle-shaped primary xylem. The reduction in the diameter of the vasculatures is reasonable, considering they grow out from the stem cortex through to the midrib. Some thicker, secondary veins of a large-sized, Gigantopteris-type leaf (possibly a different species from that of Fig. 31), have a heart-shaped vasculature (Fig. 25), and the tracheids (each up to 40 µm in diameter) are also arranged in radial rows. All these radially arranged tracheids are separated by parenchyma cells, so they appear to belong to secondary xylem. This distinguishes them from those U- or V-shaped vasculatures in Gigantonoclea guizhouensis leaf veins, which consist only of primary xylem. Therefore, these similar vasculatures suggest that Vasovinea and Gigantopteris-type leaves belong to the same kind plant.
The biological relationship of Vasovinea tianii stems, Gigantopteris-type leaves, and the compound hooks also is supported by their intimate association. More than a dozen compression leaf species of three gigantopterid genera have been reported from the Permian flora of western Guizhou (Gu and Zhi, 1974
; Tian and Zhang, 1980
; Zhao et al., 1980
). Except for the rare Linophyllum xuanweiensis, the other species belong to Gigantonoclea (with simple net veins) or Gigantopteris (with complex reticulate veins). These three genera all have a midrib (primary vein) and pinnate secondary veins. Based on our collections from the Guizhou Flora, we have recently recognized an actinodromous type of gigantopterid leaf that has either three or five primary veins (H. Li and Taylor, 1997a
). Both Gigantopteris and the actinodromous leaves have complex reticulate venation, and both are so large that it is difficult to tell from which type a leaf fragment with complex netted veins comes (e.g., Figs. 22, 27). Therefore, we prefer to use the term Gigantopteris-type for all gigantopterids with complex reticulate venation.
A gigantopterid leaf similar to Gigantonoclea guizhouensis has been reconstructed together with prickly Aculeovinea yunguiensis stems (H. Li and Taylor, 1998
). Gigantopteris-type leaves are frequently associated with compressed stems that are similar to compressed stems of A. yunguiensis in having tiny ribs (Fig. 27), but with fewer black dots than the latter. The tiny ribs and the dots appear to be the remains of a sparganum cortex and the broken bases of trichomes or prickles of Vasovinea, respectively. Gigantopteris-type leaves also are frequently associated with the compressed hooks, e.g., the compound hook in Fig. 21. Another specimen has two hooks (Fig. 22, arrows) and is associated with a leaf that has complex reticulate venation and compound rounded teeth, similar to the leaf in Figs. 29–30. In permineralized materials, Gigantopteris-type leaves are frequently associated with Vasovinea stems as demonstrated in Figs. 29–30. In one of our specimens (L9407), all of the well-preserved leaves have complex reticulate venation, and all of six or more anatomically preserved stems have vessels.
Although associational evidence strongly supports the inference that Gigantopteris-type leaves, compound hooks, and Vasovinea tianii stems could belong to the same plant species, whether the reticulate leaves are actinodromous or pinnately veined is still uncertain. We reconstructed the V. tianii stems with the actinodromous Gigantopteris-type leaves (Fig. 32) at this time, because in sample PLY04 the only leaf that is associated with the single V. tianii stem is of the actinodromous type. We also included the tendril-like trichomes, prickles, and compound hooks, since these structures are attached to the stems. We placed the hooks in a subopposite arrangement with the leaves because the location of the transverse section with the hook trace (Fig. 3) is 5 mm below the transverse section with the possible leaf traces (Fig. 2). However, this reconstruction may need to be further refined in the future in terms of whether the Vasovinea stems may actually have actinodromous or pinnately veined leaves, or both.
|
; Asama, 1959
). The latter view is supported by X. Li and Yao (1983)
who reported an impression specimen with both seed-bearing taeniopteroid and Gigantonoclea fukienensis leaves. However, whether the axes of those leaves were organically connected or just overlapped together appears unclear.
Although there are no permineralized reproductive organs to clarify whether gigantopterids belong to seed ferns or other groups, the anatomy of Aculeovinea yunguiensis suggests that the gigantopterids were seed plants (H. Li and Taylor, 1998
). The combination of characters in Vasovinea tianii, such as vessels, a eustele, sparganum cortex, and secondary xylem, separates it from the pteridophytes. These characteristics are typical of Paleozoic seed ferns, such as Heterangium, Microspermopteris, and Callistophyton (Taylor and Taylor, 1993
; Pigg, Stockey, and Taylor, 1986
; Pigg, Taylor, and Stockey, 1987
), although none of these genera share the full suite of characters of Vasovinea. In particular, none of them have reticulate-veined leaves and vessel-bearing stems. Although vessels also are found in some extant ferns and fern allies, such as some ferns (Jeffrey, 1917
; Bliss, 1939
; White, 1961
; Carlquist and Schneider, 1997
; Schneider and Carlquist, 1998
), Selaginella (Duerden, 1934
), and Equisetum (Bierhorst, 1958
), these vessels commonly have only scalariform perforations and none of these plants have both a eustele and secondary xylem.
In extant seed plants only gnetophytes and angiosperms typically have vessels, and their vessels are of the foraminate and scalariform/simple types, respectively (see above). There is considerable controversy over whether the two types of vessels represent two different evolutionary lineages or had a single origin in gnetophytes and angiosperms (Bailey, 1944
; Young, 1981
; Muhammad and Sattler, 1982
; Carlquist, 1992, 1994, 1996a
). However, the foraminate-like vessels of Vasovinea tianii differ from both those of gnetophytes and angiosperms. Although superficially similar to those of the gnetophytes, the perforation plate of Vasovinea is planar, has many more pores (which are borderless and can be transversely elongated), and has pit membranes (when they exist) without tori or margo threads. This type of foraminate-like perforation plate is not known from angiosperms, although the scalariform bordered pits and the possible vessels in the metaxylem of Vasovinea resemble those of angiosperms. Therefore, although these vessels are very important in tracing their origin, we cannot draw a comprehensive conclusion of the systematic relationships of the gigantopterids based on vessel features alone. Preliminarily results, based on phylogenetic analyses of a broad suite of characters (H. Li and Taylor, 1997b
) and molecular fossil data (Taylor et al., 1998
), suggest that gigantopterids are embedded well within the seed plants and may be another member of the anthophyte clade.
Paleoecology
Halle (1929)
reported some hook-bearing gigantopterids from central Shanxi and suggested that the gigantopterids grew as lianas in a tropical habitat. Yao (1983) reported additional hooks and also considered the gigantopterids as lianas growing in a lowland tropical forest. Geographically, Guizhou has been suggested to have been in a tropical forest biome during the Late Permian (Lin, Fuller, and Zhang, 1985
; Tian et al., 1990
; Nie, Rowley, and Ziegler, 1990
; Isozaki, 1997
). Our material strongly supports these paleoecological predictions and provides some additional ecophysiological interpretations.
The morphology, anatomy, and reconstruction of Vasovinea tianii support the interpretation that it grew as a vine or liana. In extant lianas, hooks, thorns, tendrils, and spines/prickles function as attachment organs so the slender plants can climb. In particular, hooks only have been found in vines, where they are considered very efficient climbing organs (Menninger, 1970
), and define a group called the hook climbers (Putz and Holbrook, 1991
). The tendril-like trichomes and prickles of V. tianii might also have provided an additional climbing mechanism.
Anatomical features such as a segmented xylem and the distribution and structure of its vessels also support the inference that Vasovinea tianii was a liana. Just like many living lianas, V. tianii contained xylem segments dispersed among parenchyma cells. The inflexible xylem segments within soft parenchyma tissues function like "multistranded cables" so that the liana stems could have withstood considerable deformation while maintaining the conductive function (see Carlquist, 1991
; Putz and Holbrook, 1991
). Vasovinea tianii stems have large vessels abruptly occurring at the inner portion of the secondary xylem and small vessels in the outer portion (and possibly in the metaxylem), an arrangement resembling that found in certain extant lianas (Carlquist, 1991
). The pores of Vasovinea have completely lost their borders and are elongate (8 x 10 to 13 x 15 µm; Figs. 7, 11, 18–19) and larger than bordered pits on lateral walls (Figs. 11 [lower right], 20). In extant lianas, vessel end walls tend to be more thoroughly perforated than those of tree and shrubby genera in the same family. The vines may have simple perforations with reduced borders, rather than scalariform perforations, or have large vessels with simple perforations and small vessels with scalariform perforations, instead of all scalariform vessels, or may have scalariform perforations with fewer bars than in trees (Carlquist, 1991
). Presumably, to overcome the large resistance to water flow in tracheary elements and to complement the rapid water loss from the relatively large leaves, the plants need an extremely efficient water-conducting tissue in their stems (see Ewers, Fisher, and Chiu, 1989
).
Other vessel parameters can also be correlated to habit. Lianas typically have larger vessel diameters (from an average of 157 µm to 558 µm in diameter; Carlquist, 1975
) than closely related woody trees in both Gnetum and dicots (Ewers, 1985
; Ewers and Fisher, 1989
; Ewers, Fisher, and Chiu, 1990
; Fisher and Ewers, 1995
). Bamber and Welle (1994)
reported that many liana species from the Queensland rain forest have vessels 14–57% larger than tree species of the same genera, and the largest vessel was 610 µm in diameter. In living lianas, the widest vessel members tend to be the longest (Ewers and Fisher, 1989
; Bamber and Welle, 1994
). Similarly, the vessel members of Vasovinea, up to 500 µm in diameter and 4500–5000 µm in length, are in the range of the largest and longest among all plants, both living and extinct.
The reconstruction of the thin Vasovinea stems with the large Gigantopteris-type leaves also resembles many extant lianas that usually possess high leaf-area to stem-diameter ratios (Carlquist, 1975, 1991
). Many cordate, actinodromous gigantopterid leaves are larger than 400 cm2, with some observed in the field up to 1600 cm2, but these large leaves are associated with slender stems, commonly no more than 1 cm in diameter.
It is also possible to predict the placement of Vasovinea within the forest structure. Extant lianas and vines with large, cordate leaves and long petioles usually grow in sunny areas, while those with small, narrow leaves and short petioles normally grow in more shady environments (Givnish and Vermeij, 1976
). Some Gigantopteris-type leaves have been found with well-differentiated palisade and spongy cells (Fig. 25, left) and guard cells sunken in the stomata (Guo, Tian, and Chang, 1993
). The mesophyll differentiation is related to plant species and to habitat with increased palisade differentiation related to increased exposure to light (Esau, 1965
). This is in contrast to the undifferentiated mesophyll as in Gigantonoclea guizhouensis (H. Li et al., 1994
). We suggest that both Gigantonoclea and Gigantopteris lived in a similar habitat based on the co-occurrence in the sediment, but that the Gigantopteris-Vasovinea plant grew in the sunny canopy, while the Gigantonoclea-Aculeovinea plant grew in the shady understory.
In summary, Vasovinea tianii is mainly characterized by its compound hooks, sparganum cortex, eustelic primary vascular architecture, foraminate-like vessels in the secondary xylem, and possible vessel elements with scalariform-reticulate structures and tracheids with scalariform or reticulate bordered pits in the metaxylem. These morphological features suggest that V. tianii represents a unique seed plant taxon, but its phylogenetic relationships to seed ferns, gnetophytes, and angiosperms remain to be further analyzed. The presence of compound hooks and other anatomical features clearly shows that V. tianii belongs to the gigantopterids, while additional evidence from the associated permineralized and/or compressed materials indicates that it should be reconstructed together with Gigantopteris-type leaves as lianas that grew in the Permian tropical rain forest of western Guizhou, China.
|
FOOTNOTES |
|---|
2
Author for correspondence (dwtaylo2{at}ius.edu)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSSYSTEMATICS AND DESCRIPTIONDISCUSSIONLITERATURE CITED |
|---|
———. 1974 Origin of angiosperms inferred from the evolution of leaf forms. In R. N. Lakhanpal [ed.], Symposium on origin and phytogeography of angiosperms, 1–4. Birbal Sahni Institute of Paleobotany, Lucknow, India.
———. 1982 Evolution and phylogeny of vascular plants based on the principles of growth retardation, Part 5, Origin of angiosperms inferred from the evolution of leaf forms. Bulletin of the National Science Museum (Tokyo) 8(2): 43–58.
———. 1988 The origin of the angiosperms. Ocean Press, Beijing (in Chinese). Translated by Z.-G. Gu and L. Shan from K. Asama, 1975, The origin of the angiosperms. Sanseido, Tokyo (in Japanese).
Bailey, I. W. 1944 The development of vessels in angiosperms and its significance in morphological research. American Journal of Botany 31: 421–428.[CrossRef][ISI]
Bamber, R. K., and B. J. H. Ter Welle. 1994 Adaptive trends in the wood anatomy of lianas. In M. Iqbal [ed.], Growth patterns in vascular plants. Dioscorides Press, Portland, OR.
Bierhorst, D. W. 1958 Vessels in Equisetum. American Journal of Botany 45: 534–537.[CrossRef]
Bliss, M. C. 1939 The tracheal elements in the ferns. American Journal of Botany 26: 620–624.[CrossRef][ISI]
Carlquist, S. 1975 Ecological strategies of xylem. University of California Press, Berkeley, CA.
———. 1991 Anatomy of vine and liana stems: a review and synthesis. In F. E. Putz and H. A. Mooney [eds.], The biology of vines. Cambridge University Press, Cambridge.
———. 1992 Pit membrane remnants perforation plates of primitive dicotyledons and their significance. American Journal of Botany 79: 660–672.[CrossRef][ISI]
———. 1994 Wood and bark anatomy of Gnetum gnemon L. Botanical Journal of the Linnaean Society 116: 203–221.[CrossRef]
———. 1996a Wood anatomy of primitive angiosperms: new perspectives and synthesis. In D. W. Taylor and L. J. Hickey [eds.], Flowering plant origin, evolution and phylogeny. Chapman and Hall, New York, NY.
———. 1996b Wood, bark, and stem anatomy of Gnetales: a summary. International Journal of Plant Science 157(6 Supplement): S58-S76.
———. 1996c Wood, bark, and pith anatomy of Old World species of Ephedra and summary for the genus. Aliso 13: 255–295.
———, and E. L. Schneider. 1997 SEM studies on vessels in ferns. 2. Pteridium. American Journal of Botany 84: 581–587.
Duerden, H. 1934 On the occurrence of vessels in Selaginella. Annals of Botany 48: 461–465.
Esau, K. 1965 Plant anatomy, 2nd ed. John Wiley & Sons, New York, NY.
Ewers, F. W. 1985 Xylem structure and water conduction in conifer trees, dicot trees, and lianas. IAWA Bulletin new series 6:309–317.
———, and J. B. Fisher. 1989 Variation in vessel length and diameter in stems of six tropical and subtropical lianas. American Journal of Botany 76:1452–1459.
———, ———, and S.-T. Chiu, 1989 Water transport in the liana Bauhinia fassoglensis (Fabaceae). Plant Physiology 91: 1625–1631.
———, ———, and ———. 1990 A survey of vessel dimensions in stems of tropical lianas and other growth forms. Oeecoloa 84:544–552.
Fisher, J. B., and F. W. Ewers. 1995 Vessel dimensions in liana and tree species of Gnetum (Gnetales). American Journal of Botany 82: 1350–1357.[CrossRef][ISI]
Givnish, T. J., and G. J. Vermeij. 1976 Sizes and shapes of liana leaves. American Naturalist 110: 743–778.[CrossRef][ISI]
Gu and Zhi [eds.] (= The Institute of Geology and Paleontology and the Institute of Botany, Academia Sinica). 1974 Paleozoic plants from China, Fossil plants of China, vol. 1. Science Press, Beijing (in Chinese).
Guo, Y., B. Tian, and J. Chang. 1993 The epidermal and anatomical study of gigantopterids leaves. Chinese Journal of Botany 5: 92–96.
Halle, T. G. 1929 On the habit of Gigantopteris. Geologiska Föreningens I Stockholm Förhandlingar 51: 236–242, pls. 2–3.
Isozaki, Y. 1997 Permo-Triassic boundary superanoxia and stratified superocean: records from lost deep sea. Science 276: 235–238.
Jeffrey, E. C. 1917 The anatomy of woody plants, University of Chicago Press, Chicago, IL.
Koidzumi, G. 1936 On the Gigantopteris flora. Acta Phytotaxonomica et Geobotanica 5(2): 130–139 (in Japanese and English).
Li, H., and D. W. Taylor. 1997a An actinodromous venation pattern in gigantopterids and its implication for a new structural classification of the group. American Journal of Botany (Supplement) 84: 136.
———, and ———. 1997b Phylogenetic relationships of gigantopterids and their affinities to seed plants. American Journal of Botany (Supplement) 84: 142–143.
———, and ———. 1998 Aculeovinea yunguiensis gen. et sp. nov., a new taxon of gigantopterid axis from the Upper Permian of Guizhou province, China. International Journal of Plant Sciences 159: 1023–1033.
———, E. L. Taylor, and T. N. Taylor. 1992 The anatomy of gigantopterid stems from the Upper Permian of China. Abstracts of the IVth IOP, p. 99 (Paris, French).
———, ———, and ———. 1993 The axial anatomy of gigantopterids from the Upper Permian of Guizhou Province, China. American Journal of Botany (Supplement) 80: 90–91.
———, ———, and ———. 1996 Permian vessel elements. Science 271: 188–189.[Abstract]
———, and B. Tian. 1990 Anatomic study of foliage leaf of Gigantonoclea guizhouensis Gu et Zhi. Acta Palaeontologica Sinica 29(2): 216–227 (in Chinese, with English summary).
———, B. Tian, E. L. Taylor, and T. N. Taylor. 1994 Foliage anatomy of Gigantonoclea guizhouensis Gu et Zhi (Gigantopteridales) from the Upper Permian of Guizhou Province, China. American Journal of Botany 81: 678–689.
Li, X., and Z. Yao. 1983 Fructifications of gigantopterids from South China. Palaeontographica 185 B: 11–26, pls. 1–6.
Lin, J., M. Fuller, and W. Zhang. 1985 Preliminary Phanerozoic polar wander paths for North and South China blocks. Nature 313: 444–449.
Mamay, S. H., J. M. Miller, D. M. Rohr, and W. E. Stein, Jr. 1988 Foliar morphology and anatomy of the gigantopterid plant Delnortea abbottiae, from the Lower Permian of West Texas. American Journal of Botany 75: 1409–1433.[CrossRef][ISI]
Menninger, E. A. 1970 Flowering vines of the world. Hearthside Press, New York, NY.
Muhammad, A. F., and R. Sattler. 1982 Vessel structure of Gnetum and the origin of angiosperms. American Journal of Botany 69: 1004–1021.[CrossRef][ISI]
Nie, S., D. B. Rowley, and A. M. Ziegler. 1990 Constraints on the locations of Asian microcontinents in Paleo-Tethys during the Late Paleozoic. In W. S. McKerrow and C. R. Scotese [eds.], Paleozoic palaeogeography and biogeography. Geological Society Memoir No. 12: 397–409.
Phillips, T. L. 1976 Fossil peat of the Illinois Basin, a guide to the study of coal balls of Pennsylvanian age. Illinois State Geological Survey, Champaign, IL.
Pigg, K. B., R. A. Stocky, and T. N. Taylor, 1986 Studies of Paleozoic seed ferns: additional studies of Microspermopteris aphyllum Baxter. Botanical Gazette 147: 126–136.[CrossRef][ISI]
———, T. N. Taylor, and R. A. Stocky, 1987 Paleozoic seed ferns: Heterangium kentuckyensis sp. nov., from the Upper Carboniferous of North America. American Journal of Botany 74: 1184–1204.[CrossRef][ISI]
Putz, F. E., and N. M. Holbrook. 1991 Biomechanical studies of vines. In F. E. Putz and H. A. Mooney [eds.], The biology of vines. Cambridge University Press, Cambridge.
Rothwell, G. W. 1975 The Callistophytaceae (Pteridospermopsida), I. Vegetative features. Palaeontographica, B, 151:171–196.
———. 1981 The Callistophytales (Pteridospermopsida): reproductively sophisticated Paleozoic gymnosperms. Review of Paleobotany and Palynology 32:103–121.
Schenk, A. 1883 Zweite Abhteilung: Pflanzliche Versteinerungen. In F. v. Richthofen [ed.], China, IV, 211–288. Verlag von Dietrich Reimer, Berlin.
Schneider, E. L., and S. Carlquist. 1998 SEM studies on vessels in ferns. 9. Dicranopteris (Gleicheniaceae) and vessel patterns in leptosporangiae ferns. American Journal of Botany 85: 1028–1032.[Abstract]
Taylor, D. W., H. Li, J. Dahl, F. J. Fago, G. W. Rothwell, and L. J. Hickey. 1998 The molecular fossil oleanane and preliminary data on its occurrence in gigantopterids, anthophytes and other seed plants. American Journal of Botany (Supplement) 85: 81–82.
Taylor, T. N., and E. L. Taylor. 1993 The biology and evolution of fossil plants. Prentice Hall, Englewood Cliffs, NJ.
Tian, B., and L. Zhang, 1980 Fossil atlas of Wangjiazhai Mine Region. Coal Ministry Press, Beijing (in Chinese).
———, M. Mei, P. Zhang, L. Shao, D. Ren, and K. Jin. 1990 The distribution and sedimentological environment of the Late Perm
