The oldest flora of the South China Block, and the stratigraphic bearingsof the plant remains from the Ngoc Vung Series, northern Vietnam Paul Goneza,⇑, Hung Nguyên Huub, Phuong Ta Hoac,
Trang 1The oldest flora of the South China Block, and the stratigraphic bearings
of the plant remains from the Ngoc Vung Series, northern Vietnam
Paul Goneza,⇑, Hung Nguyên Huub, Phuong Ta Hoac, Gặl Clémentd, Philippe Janvierd
a
Palaeobiogeology, Palaeobotany, Palaeopalynology, Department of Geology (B18), University of Liège, allée du 6 aout, Sart Tilman, 4000 Liège 1, Belgium
b
Vietnam National Museum of Nature, 18 Hoang Quoc Viet Str., Hanoi, Viet Nam
c
Vietnam National University, Department of Geology, 334 Nguyên Trai Street, Thanh Xuan, Hanoi, Viet Nam
d
Muséum National d’Histoire Naturelle, UMR-CNRS 7207 CR2P, Bâtiment de Paléontologie, CP 38, 47 rue Cuvier 75231 Paris cedex 05, France
a r t i c l e i n f o
Article history:
Received 14 October 2010
Received in revised form 12 July 2011
Accepted 25 August 2011
Available online 10 September 2011
Keywords:
Early land plants
Zosterophylls
Late Silurian
Basal euphyllophyte
Early Devonian
Phytogeography
Vietnam
a b s t r a c t
Several outcrops of the Late Silurian and Devonian of the Ngoc Vung Series, northern Vietnam, yielded plant remains The Late Silurian localities delivered the earliest known flora of the South China block Although the fossils are fragmentary, they complement our knowledge about the global composition
of the flora The major components of the flora are plants with dichotomous habit and terminal bivalvate sporangia, which are close relatives to zosterophylls, and zosterophylls Plants with possible euphyllo-phyte affinities and bryoeuphyllo-phytes are occasionally present This floral composition is similar to that of the rich, younger South China block assemblages from the Posongchong and Xujiachong Formations of China, considered Pragian in age The South China block flora is therefore likely to have been dominated
by zosterophylls and pre-zosterophylls at least from the Late Silurian to the Pragian (i.e a 20 million years long period) It also strengthens the hypothesis that more derived plants were present on eastern Gondwana earlier that elsewhere, in the first steps of tracheophyte evolution The Devonian localities of the Ngoc Vung Series delivered a thick fibrous stem fragment and a basal euphyllophyte These latter plant remains provide some stratigraphic data The large stem fragment is consistent with an Eifelian age for the Duong Dong Formation (part of the Ngoc Vung Series), as suggested by the brachiopod fauna The accompanying basal euphyllophyte displays a combination of characters (axes 3–4 mm wide and lat-eral branchings) that is also consistent with an Eifelian age, but possibly more characteristic of the Emsian flora It is therefore suggested that the stratigraphic range of the Duong Dong Formation might
be extended down to the Emsian
Ĩ 2011 Elsevier Ltd All rights reserved
1 Introduction
Early land plants of Vietnam have hitherto barely been studied:
only three accounts have been published In central Vietnam,
(Zosterophyllum sp and Gosslingia sp.) in the Tay Trang Formation,
Lower Devonian Concerning northern Vietnam,Tran et al (1964)
mentioned the presence of Eogaspesia sp in the Lower Devonian
Si Ka Formation, but specimens are not figured Janvier et al
sp in Dơ Son peninsula, in an outcrop of the Ngoc Vung Series
(according to the latest stratigraphic zonation by Nguyên et al
(2007)) This Dơ Son plant locality is currently regarded as Late
Silurian (probably Pridolian) in age (Janvier et al., 1987, 2003;
Braddy et al., 2003) and belongs to the South China block The fossil
plants from Dơ Son thus represent the oldest known flora of the South China block
The Ngoc Vung Series do not however only include Upper Silurian sediments Its stratigraphic extension is likely to encompass Upper Silurian to Eifelian strata (Hung et al., 2007) The stratigraphic scheme of the series is still debated and commented in several re-cent publications (Tong et al., 2006, in press; Nguyên et al., 2007) More precise stratigraphic data are still badly needed for these plant localities of northeastern Vietnam Here we present a detail study of new material sampled in several fossil plant localities (including Dơ Son) of the Ngoc Vung Series, northeastern Vietnam, along with a re-investigation of the material figured byJanvier et al (1987) The work on the Dơ Son locality aims to assess the floral composition
of the oldest flora of the South China block We discuss its affinities with the younger South China block floral assemblages, such as those of the Early Devonian Posongchong and Xujiachong Forma-tions, and with the coeval Khazakhstanian assemblage from the Wutubulake Formation of China The second goal of this work is to consider the bearings of the other palaeobotanical samples of the
1367-9120/$ - see front matter Ĩ 2011 Elsevier Ltd All rights reserved.
⇑ Corresponding author Tel.: +32 43665260.
E-mail address: paul.gonez@hotmail.fr (P Gonez).
Contents lists available atSciVerse ScienceDirect
Journal of Asian Earth Sciences
j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / j s e a e s
Trang 2Ngoc Vung Series on the stratigraphy of the Middle Palaeozoic of the
Quang Ninh area
2 Geology
2.1 Main features of the Ngoc Vung Series
The Ngoc Vung Series mainly yield exposures in two areas: the
Haiphong area and Bai Tu Long Bay (Fig 1) In Bai Tu Long Bay,
exposures are restricted to the easternmost islands, including Tra
Ban, Quan Lan and Ngoc Vung (Fig 1A) In the Haiphong area, the main exposure is the hill of Ngoc Xuyen townlet, on the Dô Son peninsula (Fig 1B)
The Ngoc Vung Series are comprised of two formations (from base to top) The Van Canh Formation is about 400 m thick Its
low-er boundary has been obslow-erved during recent fieldwork (Decemblow-er 2010) in Kien an The Van Canh Formation is comformably under-lain by the Kien an Formation, which is Ludlow-Pridoli in age (Tong
et al., 2006) The lithology of the Van Canh Formation consists of coarse-grained quartzitic sandstones, interbedded with grey clay and siltstone lenses Fossils are found in clay and siltstone lenses,
Fig 1 Location of the investigated localities from the Ngoc Vung Series: (A) in Bai Tu Long Bay; (B) on the Dô Son peninsula, Haiphong area (redrawn from Chu et al., 2001;
Trang 3mostly at Dô Son They are mainly eurypterids (Janvier et al., 1989;
Braddy et al., 2003), but placoderms, lingulids and early land plants
fragments are also common (Janvier et al., 1987) Sedimentology
suggests that the Van Canh Formation represents alluvial channels
in a deltaic system, with a slight marine influence in Dô Son, as
suggested by the presence of eurypterids, acritarchs and
sco-lecodonts (Braddy et al., 2003)
The Van Canh Formation is comformably overlain by the Duong
Dong Formation The latter is 130 m thick in the type locality but
lateral variations of thickness exist (Fig 2) Although its lithology
is roughly similar to that of the Van Canh formation, its
palaeonto-logical facies is different The siltstone lenses mostly contain
vari-ous brachiopods, which suggest a plainly marine environment
uncomform-ably overlain by the Middle-Upper Devonian Dô Son Formation
On the Dô Son peninsula, the Duong Dong Formation does not
ap-pear There is most probably a lack of sedimentation between the
Van Canh Formation and the Dô Son Formation, which are only
separated by an erosion surface at Xom Chê quarry (Janvier et al.,
2003)
2.2 Age of the Ngoc Vung Series
Palynological samples were collected in each locality of the
Ngoc Vung Series we prospected, but did not yield any spore
(con-traBraddy et al., 2003)
Fossils are relatively rare in the Van Canh Formation They are
mostly recovered from the Dô Son peninsula A Late Silurian age is
suggested for the Van Canh Formation (at any rate on the Dô Son
peninsula) on the basis of several macrofossils, namely the
euryp-terid Rhinocarcinosoma (Braddy et al., 2003), the inarticulate
bra-chiopod Laima (Janvier et al., 1987) and several placoderm fish
remains, one of which can be referred to the nomen nudum
‘Wangol-epis’ hitherto recorded exclusively from the Silurian of China
(Nguyên et al., 2007) However scarce are these macrofossils, fishes
are clearly different from the now classical Lochkovian-Pragian taxa
of northern Vietnam and South China, but closer to those found in
association with undoubtedly Ludlow-Pridoli marine invertebrates
at My Duc, Quang Binh Province (Janvier et al., 2003) Moreover,
Braddy et al (2003)reports that palynological samples from locality
2 (Fig 1B) studied by Marshall yielded spores that indicate a Late Silurian age for this locality Yet we failed to find any evidence for palynomorphs from the same locality
The overlying Duong Dong formation is regarded as Eifelian in age, at least in its uppermost beds, according to the brachiopods fauna However, it is strongly suspected to be diachronic, as some lower beds yield fossils that are characteristic of the Pragian Eury-spirifer tonkinensis assemblage (Janvier et al., 2003; Tong et al.,
likely to encompass Pragian to Eifelian sediments Therefore, the Ngoc Vung series are comprised of deltaic deposits that may range
in age from the Upper Silurian to the Lower Middle Devonian 2.3 Palaeogeographic location of the Ngoc Vung Series
The Ngoc Vung Series are part of the South China block It is considered as a perigondwanan terrane It was located on the equator throughout the Devonian times, probably very close to the Gondwana, on the eastern side of the continent (Scotese, 2010)
3 Materials and methods
We have investigated five outcrops of the Ngoc Vung Series Three of them are located in the Bai Tu Long Bay area (Fig 1A) (1) The road cut of the path that runs along the western coast of Tra Ban Island is part of the Van Canh Fm It exhibits thin bedded silts and clays, rather crumbly and light brown coloured, in which occur plant fragments (4) The natural exposure on the eastern beach of Quan Lan Island north of Minh Chau belongs to the Duong Dong Fm It consists vertical strata of grey bluish silts scarcely pop-ping out from the sand of the beach They only released one fossil
of vegetal origin (5) The quarry along the road heading from the Ngoc Vung Island harbour is also part of the Duong Dong Fm It displays a vertical outcrop in which grey bluish silt lenses are
Fig 2 Stratigraphic sketch of the investigated localities and position of the studied samples Tk: thickness Dotted layers: sandstones; dashed layers: siltstones and clays;
Trang 4interbedded in thick horizontal sandstones layers The silt lenses
yield large plant fragments, but they are difficult to access, so only
small rock fragments have been sampled The other two exposures
are on the Dô Son peninsula (Fig 1B): (2) an old quarry along the
northern slope of the hill of Ngoc Xuyen and (3) the Xom Che
quar-ry Both of those outcrops are part of the Van Canh formation and
consist of grey bluish clay silt lenses interbedded in coase grained
sandstone The silt lenses deliver some small plant fragments and
abundant eurypterid remains Localities 1, 2 and 3 are part of the
Van Canh Formation; localities 4 and 5 are part of the Duong Dong
Formation (Fig 2) The material is housed in the Geology Museum
(Bao Tang Dia Chat), Hanoi, Vietnam We also re-observed the
material figured byJanvier et al (1987), which was sampled in
the locality 3 The material belongs to Geology Museum, Hanoi,
Vietnam, some specimens being still provisionally deposited in
the Museum National d’Histoire Naturelle, Paris
All the fossil plants we collected are adpressions sensuShute
and Cleal (1987), e.g ‘‘plant fossil specimen showing a mixture
of compression and impression states’’ They were prepared and
studied according to standard palaeobotany techniques All
speci-mens were freed from the sediment that partially embeds them
by means of the technique called ‘‘dégagement’’: sediment
cover-ing the fossil surface is removed by sharpened needles under a
dis-section microscope (Fairon-Demaret et al., 1999) The specimens
were measured from digital photographs taken with a Zeiss Stemi
2000C stereomicroscope equipped with a CCD camera and the
soft-ware AxioVision digital image processing We attempted scanning
electron microscopy observations on two three-dimensionally
pre-served specimens, but no cellular detail was available
4 Results
4.1 Localities from the Van Canh Formation
4.1.1 Locality 1: vegetative features
Locality 1 only yielded tiny, unbranched axes portions, the
width of which ranges from 0.25 to 0.5 mm
4.1.2 Localities 2 and 3
4.1.2.1 Vegetative features The localities, close to each other, are
here described and discussed together
Locality 2 delivered the best preserved fossils Slender axis
frag-ments were recovered They are about 1–1.5 mm wide and exhibit
a distinct central strand, 0.1–0.12 mm wide, the cells of which are
not preserved (Fig 3A) A bulbous structure sticking out from an
unbranched axis was also released by ‘‘dégagement’’ technique
(Fig 3B) The vascular trace displayed by the axis does not
pene-trate the bulged structure This can either be a preservation
arte-fact or the indication of an unvascularized structure The bulbous
structure looks like a slightly vertically stretched bladder It is
0.65 mm wide and 0.8 mm high
Sixty-five samples were collected at Xom Che quarry, locality 3
Vegetative remains are very abundant Most of the vegetative
frag-ments are unbranched axes, sometimes of a great length: the
lon-gest is 14 cm They occasionally exhibit a coalified central strand,
0.1–0.15 mm wide Sometimes, the central strand preserves the
outline of elongated cells (Fig 3C and D) The cells are about
400lm long and 20lm wide
All axes are smooth and of constant diameter Many instances of
branching axes are recovered, and commonly display K-type
branchings (Fig 3E) Those axes are 0.25–0.7 mm wide Some
tri-chotomies (Fig 3F) are also encountered, but they most probably
result from the folding of a K-type branching They display the
same width range The presence of lateral branchings that
dichot-omize (Fig 3G) is noteworthy The main axis of this system is
0.55 mm wide The lateral branch is 0.15 mm wide and its two daughter-axes are 0.1 mm wide Isotomous branchings are rare Two minute, delicate fossils display peculiar features Specimen ST390-7 exposes a slender axis, 0.2 mm wide and 4 mm long This axis bears a lateral pointed structure on each of its sides The pointed structures are 1.5–2.2 mm long and are roughly banana-shaped (Fig 3H) There are two possible interpretations for those unbranched lateral structures (1) Those banana-shaped emer-gences could be sporangia Rebskia musaeformis Schweiter 2000, from the Emsian of Germany, also displays banana-shaped sporan-gia However, those are much bigger (5–7 mm), terminal, and exhi-bit a rounded structure at their apex (Schweitzer, 2000) Thus our specimens are clearly different from Rebskia Moreover, we are reluctant to favour the sporangia hypothesis, because the only known plants with lateral sporangia in the Late Silurian are lyco-phytes or lycolyco-phytes-like plants, i.e with reniform and dehiscent sporangia (2) The second hypothesis is that those structures are microphylls The size and shape are consistent with this interpre-tation This hypothesis is more parsimonious, because microphylls are the characteristic vegetative lateral organs of lycophytes; and early lycophytes had already evolved by the Late Silurian (Kotyk
et al., 2002) The oldest occurrence of microphylls is gedinnian (i.e the equivalent of the Lochkovian) (Schweitzer, 1982) Specimen ST390-8 also consists of a slender axis, 0.2 mm wide and 4.9 mm long It shows a 0.8 mm long, bifurcating lateral struc-ture The basal part is 0.25 mm long; the two furcas are longer They have different lengths: 0.25 and 0.55 mm They both demon-strate a pointed end (Fig 3I) Once again two interpretations are possible for this structure: either it can be interpreted as a sporan-gium or as a lycophyte lateral vegetative organ The sporansporan-gium hypothesis is not favoured, according to the same arguments Though bifurcating sporangia are reported in the Late Silurian (Fanning et al., 1991), their shape is rather different and they are borne terminally We consider that the vegetative organ hypothe-sis is more parsimonious This structure will therefore be tenta-tively interpreted as a bifurcating microphyll, yet multifurcate microphylls only appear by the Emsian in the fossil record (Gensel and Kasper, 2005)
4.1.2.2 Sporangia Eleven complete sporangia were recovered in locality 3 They demonstrate a significant variation in size and shape (Fig 4), thus testifying of the diversity of the original flora The biggest sporangium (Fig 5A) is 3.9 mm high and 2.35 mm wide It is borne by a 7.9 mm long portion of unbranched slender axis, 0.5 mm wide The lower portion of the sporangium consists
in an abrupt widening of the subtending axis, at an angle of 60° The upper part of the sporangium has a more irregular and curved outline This structure resembles the body illustrated byEdwards
et al (2001, Fig 36) The authors compare this structure to Tarran-tia Fanning et al 1992 We disagree with this comparison TarranTarran-tia exhibits a more regular oval or ovoid outline, and the dimensions
of its sporangium are smaller (around 1 mm high) We would rather attribute our specimen to the genus Sporogonites Halle,
1916 Most of the features match Halle’s diagnosis (1916): ‘‘spore producing body consisting of a simple stalk and a terminal capsule Stalk 0.5 mm in diameter and up to at least 50 mm long, faintly striated longitudinally Capsule elongated or clavate, 6–9 mm long and 2–4 mm in diameter in the thicker upper part, with a rounded apex and a tapering base [ .]’’.The only absent characters are the striation of the stalk and the spores The smaller dimensions of our specimen are characteristic of S yunnanense Li 1966 Yet our specimen does not show any pointed apex, and the elongated epi-dermal cells of the wall of the capsule mentioned in Li’s diagnosis are not perceptible Therefore, we provisionally refer to the speci-men as cf Sporogonites yunnanense
Trang 5Fig 3 Remarkable vegetative features of the plant remains from the Dô Son peninsula, Van Canh Formation: (A) specimen ST390-1, axis exhibiting a central strand; (B) specimen ST390-2, bulbous structure emerging from an unbranched axis; (C) specimen ST390-3, axis preserved as the outlines of the cells are perceptible, general view; (D) specimen ST390-3, detailed view; (E) specimen ST390-4, K-type branching; (F) specimen ST390-5, the central daughter internodes is dichotomous; (G) specimen ST390-6, the main axis bears a lateral branch that is dichotomous, only the coalified central strand of the whole branching system is preserved; (H) specimen ST390-7, axis portion that bears two structures tentatively interpreted as microphylls; (I) specimen ST390-8, axis portion that bears a lateral original structure tentatively interpreted as a bifurcating microphyll.
Trang 6Another terminal body is clearly vertically stretched It is
0.42 mm wide and 0.67 mm high, and is tentatively interpreted
as an elongated sporangium of unknown affinity (Fig 5B)
The remaining nine sporangia are all rounded to reniform in
shape, but exhibit different characteristics Their width ranges
from 1.17 to 3 mm, and their height from 1.78 to 2.57 mm The
limit between the sporangia and the subtending axis could not
be defined clearly in these specimens, due to preservation
Two specimens terminate isotomous axes (Fig 5C–E) The
ter-minal branches of the largest specimen (Fig 5C and D) are
5.18 mm and 4.27 mm long The sporangia are rounded,
respec-tively 2.75 mm high–3.0 mm wide, and 2.57 mm high–3.23 mm
wide They are the largest ones found in this locality Specimen
ST390-12 (Fig 5E) is more slender: the only complete terminal
branch is 3.15 mm and the single preserved sporangium is
1.7 mm high Sporangium width cannot be measured with
cer-tainty due to deformation Nevertheless, this sporangium shows
a conspicuous rim parallel to the sporangial outline This feature
is interpreted as a sub-distal dehiscence line It is one of the
small-est sporangia of the flora
The other specimens are borne by unbranched axes Sporangial
dimensions do show significant variation: their width ranges from
1.17 mm to 2.22 mm, and their height from 1.78 mm to 2.21 mm
(see alsoFig 4) Among the latter, three exhibit a longer
subtend-ing axis (Fig 5F–H): respectively 5.18 mm, 4.0 mm and 2.9 mm A
sub-distal dehiscence line is perceptible on specimens ST390-14
and ST390-13 (Fig 5F and G) The latter (specimen ST390-15,
Fig 5H) shows that its two valves are separated and are clearly
un-equal in size
The last three specimens are borne by short portions of
sub-tending axis, whose length ranges from 1.75 to 2.17 mm These
three sporangia consist of oval to reniform bodies (Fig 5I–K)
One of them exhibits a clear sub-distal dehiscence line (Fig 5I)
Sporangial wall of another is ornate with a reticulum (Fig 5K–L)
We also re-observed the material figured byJanvier et al (1987)
who mentioned the presence of Cooksonia Lang 1937 sporangia in
the flora These sporangia are weak impressions of rounded bodies
with a slightly blurred outline, due to poor preservation They do
not match with the updated definition of Cooksonia, which states
that the sporangium is characteristically trumpet-shaped (Gonez
oval to reniform sporangia similar to those described above, i.e probably bivalved and dehiscent The lack of morphological detail prevents from a more precise determination
4.1.2.3 cf Filiformorama Wang et al., 2006 The specimen (Fig 5M and N), found in locality 3, is 26 mm long Axes are constant in diameter and their curves hint that they were flexible in life The main axis is 0.3 mm wide Our specimen displays one single anisot-omous branching at the base of the main axis The lateral axis is 5.6 mm long for 0.2 mm wide and unbranched (Fig 5M) The ter-minal body is 0.45 mm high for 0.5 mm wide, poorly and incom-pletely preserved (Fig 5N) It is interpreted as the base of a terminal structure, most probably a sporangium Those features are reminiscent of FiliformoramaWang et al., 2006 However, Fili-formorama sporangia are bigger and tongue-shaped or reniform (Wang et al., 2006) Here poor preservation prevents from a secure three-dimensional reconstruction of the terminal body Therefore,
we refer this specimen to cf Filiformorama sp with great reservation
4.2 Localities from the Duong Dong Formation 4.2.1 Locality 4
Locality 4 yielded larger axes, 1–6 mm wide, one of which could possibly be trifurcate, but bad preservation prevents from assess-ing it with certainty A poorly preserved specimen is of much larger size The fragment is 16 mm wide for 14 mm high The texture of the preserved organic matter is porous and the outline of the fibres
is clearly perceptible (Fig 6A) This fossil could be a wood frag-ment However, the preservation prevents from assessing that with certainty: neither aligned tracheids, nor vascular rays can be dem-onstrated We will therefore consider that this fragment is part of a large plant, stems at least 1.5 cm large, whose stem anatomy is fibrous
4.2.2 Locality 5 Locality 5 yielded plant fragments of different sizes The larger remains are 3.3–4 mm wide and display anisotomous branchings, the lateral emissions are 1.7–2 mm wide They are inserted at 120° (Fig 6B) The width of the main axis is large and constant The fragment is thus interpreted as a proximal or median part of the stem In one fragmentary specimen, the main axis (3.2 mm wide), tapers to 2 mm after two successive lateral emissions that are 1.7 mm wide Here the width of the main axis is less important and tapers distally The fossil is thus interpreted as a distal part of a main axis, close to the apex The 1.7 mm wide lateral emissions of this specimen are branched isotomously (Fig 6C) The smaller re-mains are dichotomous systems that sometimes terminate in pointed recurvations (Fig 6D) The latter can also be found as dis-persed organs in the sediment (Fig 6C)
5 Discussion 5.1 Floral composition of the localities 5.1.1 Locality 1
The fragmentary preservation of the vegetal remains prevents from providing a precise floral composition However, the fact that all, very abundant, axis fragments are all very short and narrow suggests that the flora was likely to be comprised of early embry-ophytes, probably with a rhyniophytoid architecture, i.e with dichotomous axes and terminal sporangia ‘‘Rhyniophytes’’ have been demonstrated to be polyphyletic (Kenrick and Crane, 1997),
Fig 4 Outlines of the sporangia population from Xom Che quarry, Van Canh
Formation (A) ST390-11; (B) ST390-12; (C) ST390-13; (D) ST390-14; (E) ST390-15;
F: ST390-17; G: ST390-16; (H) ST390-6; (I) ST390-10; (J) ST390-9.
Trang 7the preferred term ‘‘rhyniophytoid’’ (Pratt et al., 1978) only refers
to a morphology, and does not convey any taxonomic bearing
5.1.2 Localities 2 and 3
Our fossils demonstrate a rather remarkable diversity in the
flora of the Van Canh Formation on the Dô Son peninsula
Bulbous structures similar to those we found in Dô Son have
hitherto only been demonstrated in two taxa (1) In the
tracheo-phyte Bitelaria dubjanskii Istchenko & Istchenko 1979, similar
bul-bous structure are present, but they occur as clusters (Johnson and
Gensel, 1992) (2) Stockmansella remyiSchultka and Hass, 1997is a
Rhyniopsida sensu Kenrick and Crane, 1997 (Paratracheophyta
Gerrienne et al., 2006) It consists in dichotomous creeping axes
with lateral hazelnut-shaped sporangia, and rhizoid-bearing
bulges (Schultka and Hass, 1997) The rhizoid-bearing bulges look
similar in shape to that we illustrate onfigure 3B However, they are much smaller: 0.2–0.3 mm wide for 0.3–0.5 mm long (Schultka
and weakest clue for the presence of this taxon, we are reluctant
to postulate that the flora might have contained Rhyniopsida/ Paratracheophyta In brief, this bulged structure cannot be referred
to any existing taxon
Sporogonites is believed to be a plant at a bryophytic grade of organization (Halle, 1916; Andrews, 1960; Kenrick and Crane,
1997)
The rounded to reniform sporangia terminating dichotomous axes are relevant to plants with rhyniophytoid architecture At least two taxa were present in this locality Firstly, a taxon repre-sented by plants with large, rounded sporangia; secondly a taxon with sporangia showing a sub-distal dehiscence line Compression
Fig 5 Fertile remains from Xom Che quarry, Van Canh Formation: (A) specimen ST390-9, cf Sporogonites yunnanense; (B) specimen ST390-10, terminal structure interpreted
as a sporangium of previously unknown morphology; (C) specimen ST390-11, rounded sporangia borne on isotomous axes; (D) specimen ST390-11, counter-part; (E) specimen ST390-12, reniform dehiscent sporangium borne by isotomous axes; (F) specimen ST390-13, rounded dehiscent sporangium; (G) specimen ST390-14, reniform dehiscent sporangium; (H) specimen ST390-15, sporangium with unequal valves, the larger upper valve is broken, and thus reveals part of the lower smaller one; I, specimen ST390-16, rounded dehiscent sporangium; (J) specimen ST390-6, oval sporangium; (K) specimen ST390-17, sporangium with a reticulate wall; (L) specimen ST390-17, detail
of the reticulum; (M) specimen ST390-4, cf Filiformorama sp., general view, the arrow points toward the anisotomous branching; N, specimen ST390-17, cf Filiformorama sp, detail of the terminal body interpreted as the base of a sporangium.
Trang 8fossils of plant with rhyniophytoid architecture and rounded to
reniform sporangia are represented by three genera: Uskiella Shute
& Edwards 1989, Renalia Gensel 1976, Hsüa Li 1982 (Gonez and
cannot be attributed to any of those taxa, because of the lack of
morphological details on its sporangia Our second specimen
shares a feature in common with Aberlemnia Gonez & Gerrienne
2010: it possesses a sub-distal dehiscence line (Gonez and
Gerri-enne, 2010a) Yet, other sporangial characters defining the genus
are not displayed: the features of the sporangium/subtending axis
limit are not preserved We will therefore refer to the specimen as
cf Aberlemnia sp Aberlemnia, and possibly other plants of
rhynio-phytoid architecture with reniform dehiscent sporangia, are most
probably a stem-group of early lycophytes, i.e zosterophylls (
Go-nez and Gerrienne, 2010b) Other incertae sedis plants with
differ-ent or unknown architecture (such as cf Filiformorama or the
putative elongated sporangia) are present in this locality
Other rounded to reniform sporangia are borne by unbranched
portions of axis Three specimens each demonstrate a distinctive
feature (reticulate sporangial wall, subdistally dehiscent
sporan-gium, unequal valves), which suggests that there were at least
three different taxa of parent plants Two interpretations are
possi-ble for those sporangia borne by unbranched short portions of axis
(1) The fossils are broken parts of terminal dichotomous axis In
this case, the fossils represent other taxa of plant with
rhyniophy-toid architecture and reniform sporangia, i.e stem-group of
zoste-rophylls (2) The fossils that possess the shortest subtending axes
may represent detached laterally borne dehiscent and reniform
sporangia, a diagnostic feature of zosterophylls It is not possible
to produce a more precise determination of these specimens, due
to the fragmentary nature of the material
The presence of a central strand in some axes, interpreted as
vascular on the basis of size and shape of the observable in situ cell
outlines can either be part of plants with rhyniophytoid
architec-ture and dehiscent sporangia or of zosterophylls Zosterophylls are known to possess G-type tracheids and some of the plants with rhyniophytoid architecture and dehiscent sporangia have been demonstrated to be vascular as well (Li, 1992)
On the contrary, the seven occurrences of K-type branchings shed little doubt on the actual presence of zosterophylls Even though those branchings are not an ‘‘official’’ diagnostic character
of zosterophylls, they have hitherto only been recorded as part of zosterophylls, when not dispersed Therefore the flora includes two major components: (1) plants with rhyniophytoid architecture and reniform and/or dehiscent sporangia, which are likely to be a stem-group of zosterophylls; and (2) zosterophylls
The plants bearing the lateral structures resembling micro-phylls could also share affinities with lycophytes sensu lato More derived plants, i.e plants with euphyllophytes affinities, have possibly been present in the flora, as suggested by the lateral branching systems Minute plants with lateral fertile systems have recently been discovered in the Late Silurian of North China (Wang
et al., 2007)
5.1.3 Locality 4
It is also impossible to deduce a floral composition because of the fragmentary preservation The flora includes large plants with fibrous stem
5.1.4 Locality 5 Plant remains, all considered as belonging to the same taxon, found in locality 5 are comprised of thick axes that branch anisot-omously at 120°, and more slender dichotomous systems The slender vegetative dichotomous systems are most likely the con-tinuation of the lateral branches of the thicker axes Indeed, the thick axes lateral emissions width is comparable to that of the ba-sal internodes of the dichotomous systems On the specimen ST390-20, the larger axes and the slender isotomous systems are
Fig 6 Remarkable vegetative features of the plant remains of the Duong Dong Formation: (A) specimen ST390-18, large fibrous stem fragment from locality 4; (B) specimen ST390-19 from locality 5, main axis showing lateral branches inserted at 120°; (C) specimen ST390-20 from locality 5, main axis that tapers after two successive lateral emissions, inserted at 120°, the lateral emissions are dichotomous systems, the tapered axis branch near the edge of the fossiliferous block, arrow points to at an isolated recurved tip; (D) specimen ST390-21 from locality 5, vegetative ramified system, the terminal branches are recurved tips, some of them are indicated by arrows.
Trang 9connected The plant can thus be understood as main axes that
bear lateral, helicoidally-arranged, ramified vegetative systems
This fits the definition of euphyllophytes (Kenrick and Crane,
1997) The gross morphology of the plant indeed strongly recalls
that of some basal euphyllophytes, especially Psilophyton Dawson
1859 Similar branchings, axes width and terminal vegetative
recurvations are present in Psilophyton dawsonii (Banks et al.,
1975) Yet we prefer not to name that plant, and refer it to a ‘‘basal
euphyllophyte’’, since no sporangia can allow its precise taxonomic
attribution, nor warrant the erection of a new taxon This basal
euphyllophyte is dominant in the assemblage of Ngoc Vung Island,
which may even be monospecific
5.2 Stratigraphic bearings of the plant findings
Plants are generally, but unjustly, regarded as poor
biostrati-graphic tools for the Siluro-Devonian period The fossil record is
scarce and characterized by provinciality (Raymond et al., 2006),
which may generate biases Nonetheless, the morphological and
anatomical characters of the early land plants show remarkably
ra-pid changes throughout the Devonian, so that their
biostrati-graphic potential has been pointed out (Edwards and Davies,
1990) A biostratigraphic method, based on characters, rather than
taxa, was developed byGerrienne and Streel (1994) Unfortunately
our fossils display too few anatomical characters to apply here
However, since the stratigraphy of the Siluro-Devonian of the
Quang Ninh area needs to be refined, we consider that their
strati-graphic bearings are worth to be mentioned
The very tiny plants fragments from locality 1 (Van Canh Fm)
suggest a Late Silurian age Early Devonian axes are usually larger
indeed Size is not a widely used criterion for stratigraphy, but
con-cerning plants, a relationship between axes size and sediment age
is clearly established for the Silurian and Devonian (Edwards,
from localities 2 and 3 (Van Canh Fm) are mostly zosterophylls
and plants with terminal reniform sporangia, which are
character-istic of Upper Silurian–Lowermost Devonian proximal deposits
This is consistent with the faunal data, which suggest a Late Silu-rian age for the Dô Son outcrops
The large fibrous stem recovered at locality 4 is younger The oldest plant displaying axis around 1.5 cm thick is PerticaKasper and Andrews, 1972 Pertica is a basal euphyllophyte from the Trout Valley Formation, Maine, USA, which is LateEmsian–Early Eifelian
in age (Kasper and Andrews, 1972) This is consistent with
supposed to be Eifelian The basal euphyllophyte remain found at locality 5 is also consistent with an Eifelian age However, basal euphyllophytes (i.e with lateral vegetative systems) with such axes width (3–4 mm) are more characteristic of the Emsian ( Gerri-enne, 1990; Gerrienne and Streel, 1994) Therefore, the exposure of the Duong Dong Fm on Ngoc Vung island is possibly of Emsian age, all the more that the Duong Dong Fm is likely to be diachronic ( Jan-vier et al., 2003) This strengthens the hypothesis stating that the stratigraphic extension of the Duong Dong Formation may there-fore be more important than previously thought (Nguyên et al., 2007; Tong et al., Pragian), encompassing accepted for publication
to Eifelian sediments
5.3 Comparisons 5.3.1 Comparison with other Late Silurian floras
anal-ysis of Late Silurian macrofloras The authors led two analyses: one based on taxa and another based on morphological traits Both analyses yielded the same four phytogeographic units, ensuring that the definition of areas is not biased by regional taxonomic usage The phytogeographic units are (1) Bathurst Island, (2) a South Laurussian–Northwest Gondwanan assemblage, (3) an East Gondwanan assemblage (Australia), and (4) a Khazakhstanian assemblage We will now briefly compare our South Chinese block assemblage to each Late Silurian phytogeographic unit (Fig 7) More importance will be given to the Khazakhstanian flora, as it
is the closest assemblage to South China, in terms of palaeogeography
Fig 7 Global palaeogeography of the Late Silurian Squares: location of floras from the Khazakhstanian unit; circles: floras from the South Laurussian–Northwest Gondwanan unit; triangle: flora from Bathurst Island; star: flora from the Australian unit; rhomb: flora from the Dô Son peninsula (modified from Raymond et al., 2006 ).
Trang 105.3.1.1 Flora from the Pridoli of Khazakhstania: data from the
Wutubulake Formation of China The Wutubulake flora is coeval to
that of the Van Canh Formation at Dô Son However Wutubulake
was part of the Khazakhstanian block (Edwards, 1990; Edwards
et al., 2001) Khazakhstania was located on the equatorial belt of
the Gondwana (Scotese, 2010) The South China block, to which be-longs the Dô Son plant remains outcrop, is supposed to be a peri-gondwanan terrane, i.e it was possibly closer to the Gondwana
in Silurian-Devonian times Plants showing rhyniophytoid archi-tecture with reniform sporangia were present in both terranes, as
Table 1
Composition of Late Silurian–Early Devonian plant assemblages from southeastern Asia.
Van Canh Fm flora Watabulake Fm flora Xujiachong Fm flora Posongchong Fm flora Duong Dong
Fm flora (Late Silurian) (Late Silurian) (Pragian) (Pragian) (? Emsian–?
Givetian)
Plants at a bryophyte grade
of organization
Sporogonites cf.
S.yunnanense
Sporogonites yunnanense ( Hsü,
1966 ) Plants with a rhyniophytoid
architecture
At least three taxa, possibly including Aberlemnia sp.
Junggaria spinosa ( Cai
et al., 1993 )
Hsüa robusta ( Li, 1992 )
Salopella xinjiangensis ( Cai et al., 1993 )
Hsüa deflexa ( Wang
et al., 2003 ) Sciadophyton sp ( Cai
et al., 1993 ) Lycophytina sensu Kenrick and
Crane (1997) or plants with
lycophytes affinities
Zosterophylls indet.
Zosterophyllum sp.
( Cai et al., 1993 )
Zosterophyllum australianum ( Hao,
1992 )
Baragwanathia sp ( Hao and Gensel, 1998 )
Zosterophyllum yunnanicum ( Wang,
2007 )
Zosterophyllum australianum ( Hao and Gensel, 1998 ) Zosterophyllum longa
( Wang, 2007 )
Zosterophyllum sp.1 ( Hao and Gensel, 1998 )
Huia gracilis ( Wang
et al., 2002 )
Huia recurvata Drepanophycus
qujingensis ( Li and Edwards, 1995 )
Gumuia zyzzata ( Hao, 1989b )
Bracteophyton variatum ( Wang and Hao, 2004 )
Adoketophyton subverticillatum ( Li and Edwards, 1992; Hao
et al., 2003 ) Guangnania cuneata
( Wang and Hao, 2002 )
Discalis longistipa ( Hao, 1989a ) Stachyophyton yunnanense ( Geng, 1983; Wang and Cai,
1996 ) Wenshania zhichangensis ( Zhu and Kenrick, 1999 ) Guangnania cuneata ( Wang and Hao, 2002 )
Hueberia zhichangensis ( Yang
et al., 2009 ) Zenghlia radiata ( Hao et al.,
2006 ) Halleophyton zhichangense ( Li and Edwards, 1997 ) Demersatheca contigua ( Li and Edwards, 1996 )
Basal euphyllophytes sensu Kenrick and
Crane (1997) or plants with
euphyllophyte affinities
Probably at least one taxon
Watabulaka multidichotoma ( Wang et al., 2007 )
Psilophyton primitivum ( Hao and Gensel, 1998 )
Basal euphyllophyte Eophyllophyton bellum ( Hao and
Beck, 1993 ) Polythecophyton demissum ( Hao
et al., 2001 ) Plant with sphenopsid affinities Estinnophyton yunnanense ( Hao
et al., 2004 ) Plant of unknown affinities cf Filiformorama
sp.
Filiformorama simplexa ( Wang et al.,
2006 )
Hedeia sinica ( Wang
et al., 2002 )
Hedeia sinica ( Hao and Gensel,
1998 ) Celatheca beckii ( Hao and Gensel,
1995 ) Catenalis digitata ( Hao and Beck, 1991b )
Yunia dichotoma ( Hao and Beck, 1991a )