Int. J. Plant Sci. 172(5):725–734. 2011.
Ó 2011 by The University of Chicago. All rights reserved.
1058-5893/2011/17205-0010$15.00 DOI: 10.1086/659454
CYCAD WOOD FROM THE LOPINGIAN (LATE PERMIAN) OF SOUTHERN
CHINA: SHUICHENGOXYLON TIANII GEN. ET SP. NOV.
Shi-Jun Wang,1 ,* Xiao-Yuan He,y and Long-Yi Shaoz
*State Key Laboratory of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Beijing 100093, People’s
Republic of China; and State Key Laboratory of Palaeobiology and Stratigraphy, Nanjing Institute of Geology and Palaeontology,
Chinese Academy of Sciences, Nanjing 210008, People’s Republic of China; yYunnan University, Kunming 650031,
People’s Republic of China; and zState Key Laboratory of Coal Resources and Safe Mining, School of
Geosciences and Surveying Engineering, China University of Mining and Technology (Beijing),
Beijing, 100083, People’s Republic of China
Anatomically preserved wood collected from coal seam no. 1 of the Naluozhai coal mine in the Lopingian
(Late Permian) Wangjiazhai Formation in western Guizhou Province, China, is documented and interpreted.
The wood belonged to a trunk with an inferred diameter of 25 cm or more and consists of axial tracheids and
horizontal rays. Tracheids vary greatly in size, and small ones are usually irregularly arranged. Tracheid walls
are 5–10 mm thick and usually bilayered. There are usually biseriate araucarioides-type bordered pits on the
radial tracheid walls. Two to five oval bordered pits are in each cross-field. Ray cells are thin walled, although
thick-walled cells exist in some rays. Rays are quite variable in width and height. Uni- and partly biseriate rays
are usually 2–20 cells high, while bi- and partly tri- to tetraseriate ones are 5–79 cells high. The cell size of rays
is also quite variable and is usually larger than that of tracheids in the tangential direction. There are numerous
leaf traces in the wood, and each of them extends horizontally through the wood within a broader ray. The last
character is unique in known Paleozoic woods and leads to the establishment of Shuichengoxylon tianii gen. et
sp. nov. The affinity of the wood with cycads, its implication in the evolution of cycad xylotomy, and its
growth environment are discussed.
Keywords: coal ball, fossil wood, new genus, cycads.
Introduction
He et al. 1996; Liu and Yao 2002; Wang et al. 2006) or petioles and young stems (Li and Taylor 1998, 1999; Yao et al.
2000; Seyfullah et al. 2009). Discovery of and research on fossil
wood are important to complement this growing body of information on Permian seed plant diversity, and they constitute an
essential step in developing whole-plant concepts to include
these species in evolutionary and phylogenetic analyses.
In this article, a fourth kind of fossil wood from the Permian of southern China is described and interpreted. This wood
is very different from previously described wood genera and
species, leading to the erection of a new genus and species,
Shuichengoxylon tianii gen. et sp. nov. Its affinity with cycads
and environmental significance of the wood is discussed.
In the Late Paleozoic of China, anatomically preserved fossil
wood is known mainly from Permian-aged strata and is rare
in the Devonian and Carboniferous. The earliest research on
anatomically preserved wood from the Permian of China was
conducted by Sze (1934), and since that time 18 genera and
37 species have been recorded (Tian and Zhang 1980; Li 1986;
Zhang et al. 2006; Wang et al. 2009). Of these, only three species are from the South China phytogeographical province
(Guizhouoxylon dahebianense Tian et Li [Tian and Li 1992],
Walchiopremnon gaoi Tian, Hu et Zhao [Tian et al. 1996], and
Amyelon? sp. [Wang et al. 2009]), with the majority coming
from the North China province. This is unusual because fossil
wood is often found in Permian strata from South China, where
it is preserved mainly in coal balls and volcaniclastic sediments.
Permian and especially Lopingian (Late Permian) fossil wood
provides valuable information on the structure and organization
of seed plants during that time and key evidence for the habit
of extinct plant species. In South China, existing knowledge of
Permian seed plants comes mainly from studies of isolated seeds
and leaves (Yang and Chen 1979; Tian and Zhang 1980; Zhao
et al. 1980; Zhu et al. 1984; Yao and Crane 1986; Huang et al.
1989; Li and Tian 1990; Li 1991; Li et al. 1994; Shen 1995;
1
Material and Methods
A single, incomplete specimen of this species was collected
from coal seam no. 1, which is located in the upper part
of the Wangjiazhai Formation in the Naluozhai coal mine,
Shuicheng, western Guizhou Province, China. It is preserved
as a calcified permineralization and was not embedded in
a rock matrix. This location is well known for its Lopingianaged carbonate coal-ball flora, and we conclude that it is
highly likely that the wood was preserved within a coal ball
and subsequently became isolated. The Wangjiazhai Formation is believed to have been deposited during the Changsingian stage of the Permian (Shao et al. 1998; Wang et al.
Author for correspondence; e-mail: [email protected].
Manuscript received November 2010; revised manuscript received January
2011.
725
726
INTERNATIONAL JOURNAL OF PLANT SCIENCES
2011); this age determination is based on regional-scale correlation and plant megafossil biostratigraphy.
The wood was first cut into three slabs, with four surfaces
representing transverse sections of the wood and designated
NLZ-5 A, NLZ-5 B/Top, NLZ-5B/Bot, and NLZ-5 C. Then,
the slab with surfaces NLZ-5 B/Top and NLZ-5B/Bot was
cut along the radial and tangential directions, yielding four
surfaces designated NLZ-5 T1, NLZ-5 T2, NLZ-5 R1, and
NLZ-5 R2. Each surface was prepared by the peeling method
(Galtier and Phillips 1999), and peels were mounted on slides
with Eukitt. The slides were observed and photographed
with a Nikon microscope under transmitted light with a Nikon Coolpix 4500 digital camera. Images were adjusted with
Corel PHOTO-PAINT (ver. 12), and plates were constructed
with Corel-DRAW (ver. 12). The wood specimen and the
peels and slides containing the specimen studied in this article
are deposited in the Coal Ball Laboratory of the State Key
Laboratory of the Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Beijing.
Systematics
Division—Cycadophytes
Order—Cycadales
Genus—Shuichengoxylon gen. nov.
Generic diagnosis. Wood only; pith, cortex, and periderm
unknown. Secondary xylem consisting of axial tracheids and
horizontal rays. Radial tracheid walls possessing alternately
and contiguously arranged bordered pits. Pits in cross-field
bordered. Rays parenchymatous, with or without thickwalled cells. Rays uni- to multiseriate. Leaf traces numerous,
extending through the wood in a horizontal course. In tangential section, leaf traces spirally arranged, each associated
with a broader ray.
Type species. Shuichengoxylon tianii gen. et sp. nov.
Etymology. The generic name Shuichengoxylon is derived
from the fossil location of Shuicheng, Guizhou Province.
Species. Shuichengoxylon tianii gen. et sp. nov. (figs. 1–4).
Specific diagnosis. Wood with radial thickness up to 10
cm. Tracheids square to rectangular in shape in transverse
section and greatly variable in size. Larger tracheids arranged
regularly in radial rows, with radial diameters of 33–50 mm
and tangential diameters of 45–55 mm. Smaller tracheids arranged somewhat irregularly, with diameters of 20 mm or
more. Tracheid walls 5–10 mm thick. Pits confined to the radial tracheid walls, bordered, mainly biseriate, alternately
and contiguously arranged, with diameters of 15–19 mm.
Two to five oval bordered pits in each cross-field. Rays numerous, ;6 or 7 per square millimeter and 1, 2, or sometimes 3–4 cells wide. Uni- and partly biseriate rays 2–20 cells
high, bi- and partly tri- to tetraseriate rays 5–79 cells high.
Ray cells thin walled, walls somewhat undulate. Some rays
with thick-walled cells with bordered pits on horizontal and
radial walls. Ray cells with tangential diameter larger than
that of tracheids common. Leaf traces abundant and spirally
diverging, ;1.34 per cm2, with a vertically elongated elliptical shape in tangential section, up to 500–650 mm high,
270–330 mm wide.
Type specimen. Hand specimen NLZ-5 and peels and
slides made from it.
Type locality. Naluozhai coal mine, Shuicheng, Guizhou
Province.
Horizon. Coal seam no. 1, upper part of the Wangjiazhai
Formation.
Age. Changhsingian stage, Permian.
Etymology. Specific epithet is in honor of Bao-Lin Tian,
for his contributions to the studies of the coal geology and
palaeobotany in this region of China.
Description. The specimen comprises a section of trunk
with a preserved length of ;15 cm that represents approximately one-third or one-fourth of a complete trunk and lacks
a pith, cortex, phloem, and periderm. In cross-section, the
specimen is broadly oval in shape (fig. 1A), with a longer
axis (in radial direction of the trunk) of ;10 cm and a shorter
axis (in tangential direction of the trunk) of ;9 cm. It is deduced that the complete trunk with pith, cortex, phloem, and
periderm preserved could have attained a diameter not less
than 25 cm. The specimen, in general, is well preserved, although there are some fractures that mostly extend radially.
Some of these radial fractures run along the thin-walled rays.
Secondary xylem consists only of axial tracheids and horizontal parenchymatous rays. Tracheids vary conspicuously in
size. Larger tracheids are arranged regularly in radial rows
(fig. 1B), while smaller ones are arranged somewhat irregularly (fig. 1B). Larger tracheids are square to rectangular in
shape, with rounded angles, 33–50 mm in radial diameter
and 45–55 mm in tangential diameter. Smaller tracheids are
mostly slightly radially elongate in shape, with diameters of
20 mm or more. Tracheid walls are 5–10 (usually 7) mm thick
and consist of two layers: the outer layer is light colored and
loosely constructed, and the inner layer is dark colored and
densely constructed. The outer layer is usually thicker than
the inner one (fig. 1D). The inner layer is absent in some tracheids, interpreted here as a taphonomic artifact. Occasionally, much larger, radially elongate tracheids, 200 mm 3 70
mm in size, can be seen (fig. 1C). Ray cells are mostly radially
elongate and rectangular in shape and have straight or oblique
tangential walls, although infrequently, oblique, rounded, and
sharp tangential walls are also present. Cell walls are typically
thin and infrequently broken, but thick-walled cells are occasionally seen, some having contiguous bordered pits on the
horizontal walls (fig. 1D). Rays are 1–4 cells wide in the tangential direction. The tangential diameter of the ray cell is
quite variable and is often larger than that of tracheids (fig.
1E). There are ;3–7 (usually 4–6) rays per millimeter.
Radial tracheid walls typically possess biseriate and occasionally uni- or triseriate, alternately and contiguously arranged
bordered pits. Pits are circular and 15–17 mm in diameter, with
oblique and oval apertures (fig. 1F). No spiral thickenings have
been observed on the radial tracheid walls. Rays are procumbent, consisting of thin-walled parenchymatous cells and occasionally thick-walled cells, some of which bear contiguous
bordered pits on the radial walls (fig. 2A). Ray cells are mostly
bricklike (muriform) in shape, with a radial length of 110–290
mm. Horizontal and tangential walls are usually smooth and
infrequently slightly undulate. Usually, no pits are observed on
the radial walls of ray cells, although rarely large, elliptical,
simple pits can be seen (fig. 2B). In each cross-field, there usu-
Fig. 1 Cross section of the wood. A, Outline of the specimen, with the lower side toward the center of the trunk. Peel NLZ-5C. Scale bar ¼ 1
cm. B, 1, irregularly arranged smaller tracheids; 2, rays consisting of wide cells; 3, a triseriate ray consisting of narrow cells. Slide WP2-0116. Scale
bar ¼ 200 mm. C, Large, radially elongate tracheid (arrow). Slide WP2-0116. Scale bar ¼ 100 mm. D, 1, tracheid with two-layered wall; 2,
tracheids with one-layered wall; arrow indicates a ray cell with contiguous bordered pits on the horizontal wall. Slide WP2-0116. Scale bar ¼ 200
mm. E, Three rays consisting of wide cells. Slide WP2-0116. Scale bar ¼ 100 mm. F, Radial section of the wood showing biseriate araucarioidestype bordered pits on radial tracheid walls. Slide WP2-0117. Scale bar ¼ 50 mm.
Fig. 2 A-C, Radial section of the wood. Slide WP2–0117. A, Procumbent ray consisting of thin-walled parenchymatous cells; arrow indicates
a thick-walled cell bearing contiguous bordered pits on the radial wall. Scale bar ¼ 100 mm. B, Ray cells with large, elliptical, simple pits on the
walls (arrows). Scale bar ¼ 50 mm. C, Cross-fields with several small, bordered pits in each of them. Scale bar ¼ 30 mm. D–F, Tangential section of
the wood. Slide WP2-0118. D, Rays of differing shape, width, and height; arrows indicate tracheids bending toward rays. Scale bar ¼ 200 mm. E,
1, tracheid inserting into a ray by its tip; 2, tracheids bending toward rays. Scale bar ¼ 100 mm. F, Two rays showing the shape of ray cells; 1, cell
with undulate wall; 2, much smaller ray cells. Scale bar ¼ 100 mm.
WANG ET AL.—CYCAD WOOD SHUICHENGOXYLON TIANII GEN. ET SP. NOV.
ally are two to five oval and obliquely oriented bordered pits
arranged in two or three horizontal rows (fig. 2C).
Tracheids possess long and acute tips, and because of their
thick walls the boundaries between them are obscure (fig.
2D); thus, the actual length of the tracheids is difficult to
measure, but some tracheids are at least 4 mm long. Tracheids neighboring broad rays that contain large leaf traces
tend to bend toward the ray (figs. 2D, 2E, 3F). Furthermore,
some tracheids can have their tips inserted into rays (fig. 2E).
No pits or spiral thickenings are observed on the tangential
tracheid walls. Rays are numerous, 6 or 7 per square millimeter, and are uniseriate, partly biseriate, biseriate, partly triseriate, or tetraseriate in width. Uni- and partly biseriate rays
are nearly equal in number to bi- to tetraseriate ones. Uniseriate and partly biseriate rays are up to 20 (usually 3–10) cells
high, and bi- to tetraseriate rays are 5–79 (usually 11–40) cells
high. Rays are of various shapes. Linear rays are nearly uniform in width (fig. 3A). Fusiform rays are widest in the middle
and taper toward the tips (fig. 3A). Tadpole-shaped rays are
widest at or near one tip (fig. 2D), and dumbbell-shaped rays
are narrowest in the middle and broaden toward the tips.
There are also irregularly shaped rays. The width of the ray
depends on the number of cell files and the size of cells. Cells
at the tips of the rays are somewhat rounded triangular, and
others are square, rectangular, polygonal, or nearly round in
shape (fig. 3A, 3B, 3D). Ray cells vary greatly in size, from 20
to 62 mm in diameter, within rays (figs. 2E, 2F, 3B) and from
ray to ray. The horizontal or radial walls of some ray cells are
more or less undulate (fig. 2F) or even collapsed, possibly because of the thin and delicate nature of the cell walls. No pits
have been observed on the tangential walls. In some rays,
thick-walled cells can be seen (fig. 3B).
One of the most distinct features of this wood specimen is
that it possesses numerous leaf traces and that each of them is
associated with a broader ray. These leaf traces extend
through the wood in a horizontal course (fig. 3C). In a tangential section of the wood with dimensions of ;20 mm 3
40 mm, there are 11 leaf traces that diverge in an approximately spiral sequence (fig. 4). The leaf trace is usually located
in the middle part of a broader, i.e., bi- to tetraseriate, ray.
The ray tapers from the broadest part, where the leaf trace is
located, and forms taillike ends (fig. 3E). The leaf trace is vertically elliptical in the tangential section of the wood and measures 500–650 mm high and 270–330 mm wide (fig. 3E, 3F),
although some leaf traces are much smaller, only 200–220 mm
high and 120–150 mm wide (fig. 3D). Leaf traces seem to be
concentric, consisting of an outer xylem cylinder and a central
pithlike area. The xylem cylinder is several cell layers thick
and consists of nearly isodiametric tracheids 15–30 mm in diameter with spiral, uni- to multiseriate, contiguous and opposite scalariform pits on the longitudinal walls. The central area
consists of small, fiberlike cells that are also nearly round, with
a diameter of 10–13 mm and straight end walls (fig. 3F, 3G).
Comparison of Shuichengoxylon tianii with Other
Paleozoic Woods and Its Affinity
The key diagnostic feature of the new genus is its numerous leaf traces in the wood, along with the fact that each leaf
729
trace is associated with a broader ray. This feature, as far as
we are aware, has not been reported in any other Paleozoic
wood and easily distinguishes this genus from previously recognized genera and species.
Maheshwari (1972) described the wood Megaporoxylon
antarcticum Maheshwari from the Permian of Antarctica.
The species also has a secondary xylem at least 7.5 cm thick,
within which there are numerous leaf traces, ;5/cm2, organized in a spiral arrangement. However the leaf traces of M.
antarcticum run through the wood obliquely, and in that species there are only narrow, mainly uniseriate or rarely partly
biseriate rays; there are no broader rays within which the
leaf traces are located. Instead, the xylem of the leaf trace is
surrounded by parenchyma that does not extend upward and
downward into long and pointed tails; i.e., the leaf trace is
not located in the ray. In our specimen, however, the leaf
trace runs through the wood horizontally and is located
within a broader ray that extends upward and downward
into long and pointed tails (fig. 3D, 3E).
Our specimen represents a part of a trunk with an inferred
diameter of ;25 cm. Generally, in gymnosperms only small
branches that bear leaves produce leaf traces, while the trunk
only extremely infrequently produces leaf traces, as it does
not bear leaves. However cycad trunks are an exception. The
trunk of S. tianii is typically covered by petiole bases or leaf
scars left by shed leaves and have abundant leaf traces that
pass outward through the wood and the cortex and finally
enter the leaf base. In fossil and extant cycad trunks with
monoxylic wood, there usually are two kinds of rays, narrower and broader. The leaf trace typically extends outward
within the broader rays (Greguss 1968); thus, in tangential
section, a leaf trace in each broader ray can be observed. In
some fossil cycad trunks, the leaf trace is distributed within
broader rays. These broader rays containing leaf traces have
also been called leaf gaps by some authors (Archangelsky
and Brett 1963; Gould 1971; Taylor et al. 2009). Lyssoxylon
grigsbyi Daugherty, a cycad trunk from the Upper Triassic of
Arizona and New Mexico, has thick wood and possesses two
kinds of rays, a narrower one that is uniseriate, sometimes
biseriate, or rarely triseriate, and a broader one that may be
up to 28 cells wide, with a leaf trace located in its lower central part (Daugherty 1941; Gould 1971). Michelilloa waltonii
Archangelsky et Brett, a Triassic cycad trunk from Argentina,
has both narrower rays and broader rays containing leaf
traces. The broader rays are large and fusiform in the tangential section of the trunk, 2–5 mm high, and ;0.5 mm wide.
The leaf trace and a large duct are located at the lower and
upper parts of the ray, respectively (Archangelsky and Brett
1963). Furthermore, in the wood of Antarcticycas schopfii
Smoot, Taylor et Delevoryas, a cycad trunk from the Early–
Middle Triassic of the Antarctica, and Centricycas antarcticus Cantrill, a cycad trunk from the Late Cretaceous of the
Antarctica, multiseriate rays containing leaf traces have also
been documented (Smoot et al. 1985; Cantrill 2000). As the
genera Lyssoxylon, Michelilloa, and Centricycas are based
on fossil stems with wood, pith, and cortex preserved, it is
not possible to assign our wood to any of these genera, because features of its pith and cortex are unknown. Present
evidence shows that leaf traces in rays do not occur in
gymnospermous stems or trunks except in cycads. So the ray
Fig. 3 A, Tangential section of the wood showing several rays of differing shape and width. Slide WP2-0118. Scale bar ¼ 200 mm. B,
Tangential section of the wood showing thick-walled cells in the ray (arrows). Slide WP2-0118. Scale bar ¼ 100 mm. C, Radial section of the wood
showing a leaf trace (LT) extending horizontally in the ray; R, tailed part of the ray. Slide WP2-0117. Scale bar ¼ 500 mm. D, Tangential section of
WANG ET AL.—CYCAD WOOD SHUICHENGOXYLON TIANII GEN. ET SP. NOV.
containing the leaf trace can probably be considered a characteristic feature indicating an affinity with cycads.
Other characteristic features of our specimen are that many
ray cells are tangentially wider than the tracheids and that their
walls are very thin and often undulate. These features are present in many modern cycads (Greguss 1968, pl. 12, figs. 1–8)
and some fossil cycads, such as Cetricycas antarcticus (fig. 4D
in Cantrill 2000) and Bucklandia sahnii Bose (Bose 1953, pl. 3,
figs. 25, 26). However, in most gymnosperm wood, the ray cells
are tangentially narrower than the tracheids. In the rays of our
specimen, in addition to thin-walled parenchymatous cells there
are also thick-walled cells (fig. 3B). This feature is also present
in the fossil cycad L. grigsbyi (Daugherty 1941; Gould 1971).
As stated above, our specimen probably represents the
wood of a cycad. Because of the similarities to the wood of
L. grigsbyi, the new genus is assigned to the order Cycadales
(Taylor et al. 2009).
Discussion
Associated Cycads from the Permian
From the Wangjiazhai Formation in Shuicheng and its laterally equivalent strata from the Longtan, Xuanwei, and Leping
formations (see Wang et al. 2011), a number of mega- and
microfossils attributed to cycads have been reported. The gymnosperm root Amyelon xui Li, from coal balls in the Wangjiazhai Formation at Suicheng, is similar to our specimen in
possessing rays consisting of large, thin-walled cells in the secondary xylem (Li 1986; Wang et al. 2009). It is possible that
A. xui and our specimen belong to the same plant, and in this
regard A. xui resembles the roots of the Triassic cycad Antarcticycas schopfii in possessing a diarch stele and a similar extraxylem structure (Smoot et al. 1985). No other cycad
magafossils have been recorded from the Wangjiazhai Formation, although dispersed pollen attributed to cycads is relatively common. Zhang (1987) reported pollen grains of
Cycadopites follicularis Wilson et Webster, Cycadopites sp. A,
and Cycadopites sp. B from coal seam no. 1 in the Wangjiazhai
coal mine, Shuicheng, and from our own observations we conclude that Cycadopites-like pollen grains are common in acetate peels made from coal balls from Shuicheng.
Laterally equivalent geological formations have yielded
other evidence. Monocolporate pollen of Entylissa spp. from
the Longtan Formation in Changxing, Zhejiang Province,
may have been produced by a cycad, but this kind of pollen
may also have an affinity with Ginkgoales (Ouyang 1962).
Possible cycad leaves of Pterophyllum eratum Gu et Zhi (Gu
and Zhi 1974; Zhao et al. 1980) and Nilssonia sp. (Tian and
Zhang 1980) have been described from the lower part of
Xuanwei Formation in eastern Yunnan and western Guizhou,
as have those of Nilssonia undulata Stockmans et Mathieu,
Nilssonia xichaense He et al. (He et al. 1996), and Lepingia
731
Fig. 4 Diagram showing the arrangement of leaf traces in the tangential section of the wood: 1, larger leaf trace; 2, smaller leaf trace.
Slide WP2-0118. Scale bar ¼ 5 mm.
emarginata Liu et Yao (Liu and Yao 2002) from the Loping
Formation in Leping, Jiangxi Province. Collectively, this pollen and these leaves offer additional evidence to the existence
of cycads in the Lopingian of South China.
Guo et al. (1990) described the rachis morphology and
anatomy of Plagiozamites oblongifolius Halle from the Late
Permian Xuanwei Formation in Guizhou Province. It bears
a U- or inverted-V-shaped vascular bundle resembling that of
a cycad plant. However, there are differences between the rachis of P. oblongifolius and those of cycads; in the former the
vascular bundle is continuous, while in the latter the inverted-V-shaped vascular bundle is discontinuous or consists
of several small individual bundles. Guo et al. (1990) considered that P. oblongifolius probably represents a primitive cycad and suggested a possible evolutionary trend from the
continuous inverted-V-shaped vascular bundle of P. oblongifolius to the discontinuous one of the derived representatives.
However, we conclude that P. oblongifolius is not a cycad
but a noeggerathid, on the basis of its leaf organization and
architecture, with the noeggerathids known to have been heterosporous ferns (Wang 2008).
In the Permian Cathaysian flora of North China, cycads
are relatively common and are preserved mainly as impression-compressions of leaves and fertile organs. These include
vegetative shoots of Procycas densinervioides Zhang et Mo,
Cladotaeniopteris shaanxiensis Zhang et Mo (Zhang and
Liang 1985), and Cathaysiocycas rectanervis Yang and Phasmatocycas pinnata Yang (Yang 2006); the leaves Pterophyllum pruvostii Stockmans et Mathieu and Pterophyllum
cutelliforme Sze (Gu and Zhi 1974), Pterophyllum eratum
Gu et Zhi (Gu and Zhi 1974; Zhao et al. 1980; He et al.
1996; Yang 2006), Nilssonia nobilisoides Yang, Nilssonia
brevisegmenta Yang, and Nilssonia permica Yang (Yang
2006), Tianbaolinia circinalis Gao et Thomas (Gao and
Thomas 1989), and Ctenis densinervis Yang; microsporangiate cones Cycadostrobilus paleozoicus Zhu (Zhu et al. 1994)
the wood showing two rays with a small leaf trace (LT) within each of them; arrow indicates a tracheid inserting into a broad ray. Slide WP20118. Scale bar ¼ 200 mm. E, F, Tangential section of the wood showing the broad ray with a leaf trace in it. P, central pithlike part of the leaf
trace; R, tailed part of the ray; X, xylem cylinder. Arrows in F show tracheids bending toward the ray. Slide WP2-0118. Scale bars ¼ 150 (E) and
100 mm (F). G, Enlargement of C, showing the structure of the leaf trace; P and X as in F. Scale bar ¼ 50 mm.
732
INTERNATIONAL JOURNAL OF PLANT SCIENCES
and Liulinia lacinulata Wang (Wang 1986); megasporophylls
Eophyllogonium Mei, Dilcher et Wan (Mei et al. 1992),
Crossozamia chinensis (Zhu et Du) Gao et Thomas and
Crossozamia minor Gao et Thomas (Gao and Thomas
1989), and Liella mirabilis Yang et Zhao and Pania cycadina
Yang (Yang 2006). However, in each case these are disarticulated into organs, so that whole-plant concepts have yet to
be developed for these species (Bateman and Hilton 2009).
In North America and Europe, several Paleozoic cycad fertile organs have been described, and they are mainly from
the Permian, including Phasmatocycas bridwellii Axsmith
et al. (Axsmith et al. 2003), Phasmatocycas coriacea (Cridland et Morris) Axsmith et al. (Cridland and Morris 1960),
Phasmatocycas kansana Mamay (Mamay 1973), Sobernheimia jonkeri Kerp (Kerp 1983), and Archaeocycas whitei
Mamay (Mamay 1973). However, the situation is different
for anatomically preserved Paleozoic cycad fossils, because
few have been reported to date. Lasiostrobus polysaccii Taylor, an anatomically preserved pollen cone from coal balls in
the Upper Pennsylvanian of Illinois, has been suggested to
have affinities with the cycads. However its affinity is controversial, and it may be either a conifer or an early ginkgophyte (Taylor 1970). Two kinds of anatomically preserved
gymnospermous stems, Koleoxylon chinense Zhang et Zheng
and Koleoxylon xuetaiziensis Zhang et Zheng, have been described from coal balls of the early Early Permian Taiyuan
Formation in the Xuetaizi coal mine, Chaoyang, Liaoning
Province, northeastern China, and have been considered
cycads. However, the pith of the stems seems to be septated
in some specimens (Zhang et al. 2006, pl. 3–2D); thus, an
affinity with cordaitean coniferophytes may be more appropriate.
Evolutionary Implications
The living cycads possess two kinds of stele, one monoxylic and the other polyxylic. Monoxylic steles consist of
only a single vascular cylinder, while in polyxylic steles there
are additional vascular tissues in the form of narrow concentric cylinders (rings or partial rings) or apparently isolated
clumps in cross-sections (Norstog and Nicholls 1997).
Within the Cycadaceae, only Cycas is polyxylic, while both
the Stangeriaceae and Zamiaceae possess both monoxylic
and polyxylic steles. The polyxylic stele of some species of
Cycas can be composed of several vascular cylinders. Our
specimen obviously has a monoxylic stele and is very similar
to the living cycad Dioon spinulosum Dyer, which also possesses a thick monoxylic wood (Chamberlain 1911; Langdon
1920; Greguss 1968, pl. 67, fig. 1; Norstog and Nicholls
1997). Until now, the oldest cycad stems have been from five
Triassic genera, each of which is monoxylic. Polyxylic steles
have been interpreted as derived with respect to monoxylic
steles (Crane 1988; Stevenson 1990); this interpretation is
supported by the fossil record because the oldest fossil stems
from the Triassic (Antarcticycas, Lyssoxylon, and Michelilloa) and Jurassic (Fascivarioxylon) assigned to Cycadales are
monoxylic (Artable et al. 2004). The evidence of our specimen supports the statement that monoxylic wood is symplesiomorphic in the wood of cycads.
A comparatively ancient divergence of Cycas from the remainder of the cycad lineages was recently inferred from a
molecular sequence data set including the trnL intron and ITS2
region from species in 10 genera of extant Cycadales (Bogler
and Francisco-Ortega 2004). This appears to be concordant
with the finding that Cycas lineage may have diverged from
the other extant lineages no later than the Permian. On the basis of minimum-age mapping, girdling leaf traces and the
V-shaped pattern of the vascular bundles had both evolved no
later than the Permian, indicating that the Cycadales were well
differentiated from other seed plant groups by that time
(Hermsen et al. 2006). The Permian age of our specimen provides clear evidence for the supposition that the Cycadales differentiated from other seed plant groups no later than the
Permian.
Habit and Paleoecology of Shuichengoxylon tianii
Until now, the earliest reconstructed cycads, such as A.
schopfii (Hermsen et al. 2009) and Leptocycas (Delevoryas
1968, 1975, 1982; Delevoryas and Hope 1971, 1976), both
from the Triassic, have been slender plants. These authors
suggest that the earliest cycads were slender, branched plants
with small amounts of secondary xylem (Delevoryas 1982).
Our specimen, however, is much larger and was arborescent,
with abundant secondary xylem.
The Wangjiazhai Formation comprises paralic coal deposits
and consists of black-gray calcarenaceous silty sandstone, mudstone, gray fine sandstone, gray-brown clay, thin-layered mud
limestone, bioclastic limestone, and ;10 coal seams (Tian and
Zhang 1980; Wang et al. 2011). Impression-compression plant
fossils in the formation include mainly lycopsids, sphenopsids,
ferns, and pteridosperms, although there are also a small number
of cycads, ginkgopsids, and conifers (Tian and Zhang 1980).
The plant fossils from coal balls in coal seam no. 1 (C605)
also have the same composition as impression-compressions
(Tian and Wang 1995) and are interpreted as preserving the
same source flora. Using an analysis of the diversity of plant
fossils, leaf physiognomy, and growth rings in fossil woods,
Guo (1990) indicated that Shuicheng was located at a lowlatitude region near the equator in the latest Permian and was
in a climate of tropic-subtropic seasonal rain forest. Both Tian
and Zhang (1980) and Guo (1990) conclude that the environment was humid.
The secondary xylem of our specimen is constructed uniformly, and no conspicuous growth rings are formed, suggesting that there was no distinctively periodic change of the
environment in which the parent plant lived. There are no
growth rings in the secondary xylem of the associated cycadalean root A. xui (Li 1986; Wang et al. 2009), which also
indicates a uniform environmental condition. It is considered
that the parent plant of our specimen and A. xui was living
in the peat swamp of coal seam no. 1 (C605) and that the
environment was uniform, without distinctively periodic
change. This is different from the condition of living cycads,
which generally live in dry habitats such as hills. Evidence
from sedimentology and coexisting plant species demonstrate
that S. tianii grew in a lowland swamp setting, as is similarly
concluded for the Triassic cycad A. schopfii from Antarctica
(Smoot et al. 1985; Hermsen et al. 2009). We conclude that
WANG ET AL.—CYCAD WOOD SHUICHENGOXYLON TIANII GEN. ET SP. NOV.
the present distribution of cycads is due to competition of
geologically younger and more competitive plant groups, including ferns and angiosperms.
Acknowledgments
We are grateful to Dr. Jason Hilton (University of Birmingham)
and an anonymous reviewer for their comments on the manu-
733
script. We thank Xue-Jian Yang (Institute of Botany, Chinese
Academy of Sciences [CAS]) for help in photo processing.
This research was supported by CAS project KZCX2-EW120, the National Natural Science Foundation of China
(grants 41030213, 30670140), and the State Key Laboratory
of Palaeobiology and Stratigraphy, Nanjing Institute of Geology and Palaeontology, CAS (grant 073103), which are gratefully acknowledged.
Literature Cited
Archangelsky S, DW Brett 1963 Studies on Triassic fossil plants from
Argentina. II. Michelilloa waltonii nov. gen. et sp. from the
Iscigualasto Formation. Ann Bot 27:147–154.
Artable AE, AB Zamuner, DW Stevenson 2004 Two new petrified
cycad stems, Brunoa gen. nov. and Worsdellia gen. nov., from the
Cretaceous of Patagonia (Bajo de Santa Rosa, Rio Negro Province),
Argentina. Bot Rev 70:121–133.
Axsmith BJ, R Serbet, M Krings, TN Taylor, EL Taylor, SH
Mamay 2003 The enigmatic Paleozoic plants Spermopteris and
Phasmatocycas reconsidered. Am J Bot 90:1585–1595.
Bateman RM, J Hilton 2009 Palaeobotanical systematics for the
phylogenetic age: applying organ-species, form-species and phylogenetic species concepts in a framework of reconstructed fossil and
extant whole-plants. Taxon 58:1254–1280.
Bogler DJ, J Francisco-Ortega 2004 Molecular systematic studies in
cycads: evidence from trnL intron and ITS2 rDNA sequences. Bot
Rev 70:260–273.
Bose MN 1953 Bucklandia sahnii sp. nov. from the Jurassic of the
Rajmahal Hills, Bihar. Palaeobotanist 2:41–50.
Cantrill DJ 2000 A petrified cycad trunk from the Late Cretacous of
the Larsen Basin, Antarctica. Alcheringa 24:307–318.
Chamberlain CJ 1911 The adult cycad trunk. Bot Gaz 52:81–104.
Crane PR 1988 Major clades and relationships in the ‘‘higher’’
gymnosperms. Pages 218–272 in CB Beck, ed. Origin and evolution
of gymnosperms. Columbia University Press, New York.
Cridland AA, JE Morris 1960 Spermopteris, a new genus of
pteridosperms from the Upper Pennsylvanian series of Kansas.
Am J Bot 47:855–859.
Daugherty LH 1941 The Upper Triassic flora of Arizona. Carnegie
Institute of Washington Publications, no. 526. Carnegie Institute of
Washington, Washington, DC. 108 pp.
Delevoryas T 1968 Some aspects of cycadeoid evolution. J Linn Soc
Lond Bot 61:137–146.
——— 1975 Mesozoic cycadophytes. Pages 173–191 in SW Campbell, ed. Gondwana geology: papers from the Third International
Gondwana Symposium, Canberra, Australia, 1973. Australian
National University Press, Canberra.
——— 1982 Perspectives on the origin of cycads and cycadeoids.
Rev Palaeobot Palynol 37:115–132.
Delevoryas T, RC Hope 1971 A new Triassic cycad and its phyletic
implications. Postilla 150:1–21.
——— 1976 More evidence for a slender growth habit in Mesozoic
cycadophytes. Rev Palaeobot Palynol 21:93–100.
Galtier J, TL Phillips 1999 The acetate peel technique. Pages 67–71
in TP Jones, NP Rowe, eds. Fossil plants and spores: modern
techniques. Geological Society, London.
Gao ZF, BA Thomas 1989 A review of fossil cycad megasporophylls,
with new evidence of Crossozamia Pomel and its associated leaves
from the Lower Permian of Taiyuan, China. Rev Palaeobot Palynol
60:205–223.
Gould RE 1971 Lyssoxylon grigsbyi, a cycad trunk from the Upper
Triassic of Arizona and New Mexico. Am J Bot 58:239–248.
Greguss P 1968 Xylotomy of the living cycads. Akadémiai Kiadó,
Budapest.
Gu and Zhi (Nanjing Institute of Geology and Palaeontology, Institute
of Botany Writing Group) 1974 Fossil plants of China. Vol 1.
Palaeozoic plants from China. Science, Beijing. (In Chinese.)
Guo YT 1990 Plant palaeoecology of Late Permian coal-bearing
strata of western Guizhou Province, China. J China Coal Soc 15:
48–53.
Guo YT, BL Tian, DX Han 1990 The anatomical study of
Plagiozamites oblongifolius and the systematic position of Plagiozamites. Acta Bot Sin 32:799–804.
He XL, DS Liang, SZ Shen 1996 Research on the Permian flora
from Jiangxi Province, China. China University of Mining and
Technology Publishing House, Xuzhou. (In Chinese with English
summary.)
Hermsen EJ, TN Taylor, EL Taylor 2009 Morphology and ecology of
the Antarcticycas plant. Rev Palaeobot Palynol 153:108–123.
Hermsen EJ, TN Taylor, EL Taylor, DWM Stevenson 2006 Cataphylls of the Middle Triassic cycad Antarcticycas schopfii and new
insights into cycad evolution. Am J Bot 93:724–738.
Huang LM, YN Huang, MT Mei, SS Li 1989 The Early Permian
coal-bearing strata and flora from southwestern Fujian Province,
South China. China Coal Industry Publishing House, Beijing. (In
Chinese with English summary.)
Kerp JHF 1983 Aspects of Permian palaeobotany and palynology. I.
Sobernheimia jonkeri nov. gen., nov. sp., a new fossil plant of
cycadalean affinity from the Waderner Gruppe of Sobernheim. Rev
Palaeobot Palynol 38:173–183.
Langdon LM 1920 Stem anatomy of Dioon spinulosum. Bot Gaz 70:
110–125.
Li HQ, DW Taylor 1998 Aculeovinea yunguiensis gen. et sp.
(Gigantopteridales), a new taxon of gigantopterid stem from the
Upper Permian of Guizhou Province, China. Int J Plant Sci 159:
1023–1033.
——— 1999 Vessel-bearing stems of Vasovinea tianii gen. et sp. nov.
(Gigantopteridales) from the Upper Permian of Guizhou Province,
China. Am J Bot 86:1563–1575.
Li HQ, BL Tian 1990 Anatomic study of foliage leaf of Gigantonoclea guizhouensis Gu et Zhi. Acta Palaeontol Sin 29:216–227.
Li HQ, BL Tian, EL Taylor, TN Taylor 1994 Foliar anatomy of
Gigantonoclea guizhouensis (Gigantopteridales) from Upper Permian of Guizhou Province, China. Am J Bot 81:678–689.
Li ZM 1986 Amyelon radicans first known from China. J Integr
Plant Biol 28:209–212. (In Chinese with English abstract.)
——— 1991 Stephanospermum cf. akenioides Brongniart and Sarcospermum petiolulatum Z. M. Li sp. nov. in Chinese coal balls.
Cathaya 3:115–133.
Liu LJ, ZQ Yao 2002 Lepingia: a new genus of probable cycadalean
affinity with taeniopterid lamina from the Permian of South China.
Int J Plant Sci 163:175–183.
Maheshwari HK 1972 Permian wood from Antarctica and revision
of some Lower Gondwana wood taxa. Palaeontogr B 138:1–43.
734
INTERNATIONAL JOURNAL OF PLANT SCIENCES
Mamay SH 1973 Archaeocycas and Phasmatocycas: new genera of
Permian cycads. US Geol Surv J Res 1:687–689.
Mei MT, DL Dilcher, ZH Wan 1992 A new seed-bearing leaf from
the Permian of China. Palaeobotanist 41:98–109.
Norstog KJ, TJ Nicholls 1997 The biology of the cycads. Cornell
University Press, Ithaca, NY.
Ouyang S 1962 The microspore assemblage from the Lungtan Series
of Changhsing, Chekiang. Acta Palaeontol Sin 10:76–119.
Seyfullah L, J Hilton, SJ Wang, J Galtier 2009 Anatomically preserved pteridosperm stems and rachises from Permian floras of
China. Int J Plant Sci 170:814–828.
Shao L, P Zhang, P Ren, J Lei 1998 Late Permian coal-bearing
successions in southern China: coal accumulation on carbonate
platforms. Int J Coal Geol 37:235–256.
Shen GL 1995 Permian floras. Pages 127–223 in XX Li, ed. Fossil
floras of China through the geological ages. English ed. Guangdong
Science and Technology, Guangzhou.
Smoot EL, TN Taylor, T Delevoryas 1985 Structurally preserved
plants from Antarctica. I. Antarcticycas, gen. n., a Triassic cycad
stem from the Beardmore Glacier area. Am J Bot 72:1410–1423.
Stevenson DW 1990 The biology, structure, and systematics of the
Cycadales. Mem NY Bot Gard 57:1–210.
Sze HC 1934 On the occurrence of an interesting fossil wood from
Urumchi (Tihua) in Sinkiang. Bull Geol Soc China 13:581–592.
Taylor TN 1970 Lasiostrobus gen. nov., a staminate strobilus of
gymnosperm affinity from the Pennsylvanian of North America. Am
J Bot 57:670–690.
Taylor TN, EL Taylor, M Krings 2009 Paleobotany: the biology and
evolution of fossil plants. 2nd ed. Academic Press, Burlington, MA.
Tian BL, T Hu, H Zhao 1996 The first discovery of Walchiopremnon
gaoi sp. nov. in China. Pages 118–125 in ZL Li, ed. Selected papers
on coal geology: celebrating Prof. Gao Wentai’s eightieth birthday
and his sixty-year career in geology. China Coal Industry Publishing
House, Beijing. (In Chinese with English abstract.)
Tian BL, HQ Li 1992 A new special petrified stem, Guizhouoxylon
dahebianense gen. et sp. nov., from Upper Permian in Shuicheng
district, Guizhou, China. Acta Palaeontol Sin 31:336–345. (In
Chinese with English summary.)
Tian BL, SJ Wang 1995 Paleozoic coal ball floras in China. Pages
202–211 in XX Li, ed. Fossil floras of China through the geological
ages. English ed. Guangdong Science and Technology, Guangzhou.
Tian BL, LW Zhang 1980 Fossil atlas of Wangjiazhai Mine Region in
Shuicheng, Guizhou Province. China Coal Industry Publishing
House, Beijing. 110 pp. (In Chinese.)
Wang H, L Shao, LM Hao, PF Zhang, IJ Glasspool, JR Wheeley, PB
Wignall, TS Yi, MQ Zhang, J Hilton 2011 Sedimentology and
sequence stratigraphy of the Lopingian (Late Permian) coal
measures in southwestern China. Int J Coal Geol 85:168–183.
Wang J 2008 A note on fructification of Plagiozamites Zeiller
(Noeggerathiales) based on association. Global Geol 11:115–124.
Wang SJ, J Hilton, MM Liang, L Stevens 2006 Permineralized seed
plants from the Upper Permian of southern China: a new species of
Cardiocarpus. Int J Plant Sci 167:1247–1257.
Wang SJ, KQ Sun, JZ Cui, SM Ma 2009 Fossil plants from coal balls
in China. Higher Education, Beijing. 141 pp. (In Chinese with
English abstract.)
Wang ZQ 1986 Liulinia lacinulata, a new male cone of cycads from
latest Permian in Shanxi. Acta Palaeontol Sin 25:610–616.
Yang GX 2006 The Permian Cathaysian flora in western Henan
Province, China: Yuzhou flora. Geological Publishing House,
Beijing. (In Chinese with English abstract.)
Yang GX, F Chen 1979 Palaeobotany. Pages 104–139 in HF Hou, LP
Zhan, BW Chen, eds. The coal-bearing strata and fossils of Late
Permian from Guangdong. Geological Publishing House, Beijing.
(In Chinese.)
Yao ZQ, PR Crane 1986 Gigantopterid leaves with cuticles from the
Lower Permian of South China. Am J Bot 73:715–716.
Yao ZQ, LJ Liu, G Rothwell 2000 Szecladia new genus, a Late
Permian conifer with multiveined leaves from South China. J
Palaeontol 74:527–531.
Zhang GY 1987 Sporopollen assemblage of coal seam no. 1 in
Wangjiazhai coal mine, Shuicheng, Guizhou Province. J China Univ
Mining Tech 16:73–88.
Zhang SZ, XY Liang 1985 Note on the discovery of cycadophytes
with slender growth habit in the Permian of northern China with
special reference to the reconstruction of cycadophytes. J Northwest
Univ (Nat Sci) 2:76–80. (In Chinese with English abstract.)
Zhang W, Y Li, SL Zheng, N Li, YD Wang, XJ Yang, JJ Yang, TM Yi,
XP Fu 2006 Fossil woods from China. China Forestry Publishing
House, Beijing. 356 pp.
Zhao XH, ZG Mo, SZ Zhang, ZQ Yao 1980 Late Permian flora
from western Guizhou and eastern Yunnan. Pages 70–122 in
Nanjing Institute of Geology and Palaeontology, Academia Sinica,
ed. Late Permian coal-bearing strata and biota from western
Guizhou and eastern Yunnan. Science, Beijing. (In Chinese.)
Zhu JN, YF Hu, ZJ Li 1984 Late Permian strata and fossil plants
from Junlian area, South Sichuan. Prof Pap Stratigr Palaeontol 11:
133–147.
Zhu JN, XS Zhang, J Ma 1994 A new genus and species: Cycadostrobilus paleozoicus Zhu of Cycadaceae from the Permian of
China. Acta Phytotaxon 32:340–344. (In Chinese with English
abstract.)