Monocotyledons from the Early Campanian (Cretaceous) of Gr??nbach, Lower Austria

Abstract

Four monocotyledons, Araciphyllites austriacus sp. nov., Pandanites trinervis (Ettingshausen) comb. nov., Sabalites longirhachis (Unger) comb. nov. and one unclassified monocotyledon leaf type are described from the Campanian of Grünbach in Austria mostly on the basis of leaf morphology. A new species of Araciphyllites is described and compared to recent representatives of the family Araceae. An emended diagnosis of the genus Pandanites is proposed and species Pandanites trinervis is revised on the basis of the detailed morphology and anatomy of its leaves. The unique construction of leaves of the Pandanaceae is discussed and used for comparison with recent and fossil representatives of the family. Sabalites longirhachis is characterised by costapalmate leaves, which are not dissected into segments. Relationship to similar species of the genus Sabalites is discussed. A fragmentary preserved leaf of an unnamed monocotyledon is included to show the diversity of monocotyledons in the Grünbach Campanian wetlands.

Full text

Monocotyledons from the Early Campanian (Cretaceous) of
Gru
«nbach, Lower Austria
Jir
›|
¤Kvac
›ek a;, Alexei B. Herman b
National Museum, Prague, Va
¤clavske
¤na
¤m., 68, 115 79 Praha 1, Czech Republic
bGeological Institute, Russian Academy of Sciences, 7 Pyzhevskii Pereulok, 119017 Moscow, Russia
Received 13 May 2003; received in revised form 18 August 2003; accepted 19 September 2003
Abstract
Four monocotyledons, Araciphyllites austriacus sp. nov., Pandanites trinervis (Ettingshausen) comb. nov.,
Sabalites longirhachis (Unger) comb. nov. and one unclassified monocotyledon leaf type are described from the
Campanian of Gru
«nbach in Austria mostly on the basis of leaf morphology. A new species of Araciphyllites is
described and compared to recent representatives of the family Araceae. An emended diagnosis of the genus
Pandanites is proposed and species Pandanites trinervis is revised on the basis of the detailed morphology and
anatomy of its leaves. The unique construction of leaves of the Pandanaceae is discussed and used for comparison
with recent and fossil representatives of the family. Sabalites longirhachis is characterised by costapalmate leaves,
which are not dissected into segments. Relationship to similar species of the genus Sabalites is discussed. A
fragmentary preserved leaf of an unnamed monocotyledon is included to show the diversity of monocotyledons in the
Gru
«nbach Campanian wetlands.
 2003 Elsevier B.V. All rights reserved.
Keywords: Monocotyledons ; foliage; Araceae ; Araciphyllites; Pandanaceae ; Pandanites; Arecaceae ; Sabalites; Cretaceous ; Cam-
panian; Austria
1. Introduction
Monocots have been recognised in recent phy-
logenetic analyses based on morphological and
molecular data as one of the earliest diverging
groups in angiosperm evolution (Stevenson and
Laconte, 1995, Chase et al., 2000; Savolainen et
al., 2000; Soltis et al., 2000). They have a long
fossil history, which is thought to start in the
Early Cretaceous (Daghlian, 1981; Herendeen
and Crane, 1995, but see also Gandolfo et al.,
2000). Despite this, fossil records of Cretaceous
monocotyledons are not common if compared
with the large amount of data on Cretaceous di-
cotyledons. The monocotyledons typically occur
more frequently in palyno£oras than in the mega-
fossil record (compare Hotton et al., 1994 ; Cox et
al., 1995). Their mostly herbaceous nature and
ecological biases were probably responsible for
0034-6667 / 03 / $ ^ see front matter  2003 Elsevier B.V. All rights reserved.
doi: 10.1016/S0034-6667(03)00154-4
* Corresponding author.
E-mail addresses: jiri.kvacek@nm.cz (J. Kvac
›ek),
herman@geo.tv-sign.ru (A.B. Herman).
PALBO 2577 5-1-04
Review of Palaeobotany and Palynology 128 (2004) 323^353
R
Available online at www.sciencedirect.com
www.elsevier.com/locate/revpalbo

their under-representation in the fossil record
(Herendeen and Crane, 1995; Gandolfo et al.,
2002). During the Late Cretaceous the ¢rst occur-
rences of some recent families are con¢rmed
(Herendeen and Crane, 1995; Gandolfo et al.,
2000, 2002).
In the present paper we report on unequivocal
monocotyledon species based on leaf fossils from
the Lower Campanian Gru
«nbach Formation of
the Gosau Group in the Gru
«nbach^Neue Welt
Basin in the Eastern Calcareous Alps, Lower Aus-
tria. Those are among the earliest records for
Araceae, Pandanaceae and Arecaceae. Although
collected since the nineteenth century, the Gru
«n-
bach £ora had never been studied monographi-
cally. The ¢rst plant described from the Gru
«nbach
Formation was Flabellaria longirhachis Unger
(1850). Two years later Unger (1852) published
illustrations of this plant. Specimens studied and
depicted by Unger are kept in the Austrian Geo-
logical Survey and from this type collection we
have selected the lectotype. Later von Ettingshau-
sen (1852) published a monograph on fossil rep-
resentatives of the genus Pandanus in which he
included pandanaceous leaves from Gru
«nbach:
Pandanus austriacus Ettingshausen, P. trinervis
Ettingshausen and P. pseudoinermis Ettingshau-
sen. Most of his type material is also stored in
the above mentioned collection. In 1879 Ettings-
hausen (not published), in 1904^1906 Krasser
(1906) and in 1934 Kerner-Marilaun (1934) pre-
liminarily identi¢ed and labelled some fossil
plants in the same collection, but all these identi-
¢cations needed a revision. This was initiated in
1999 by the present authors (Herman and Kva-
c
›ek, 2002a,b).
2. Materials and methods
Numerous well-preserved plant fossils from
beds accompanying the coal seams of the Gru
«n-
bach Formation are housed at geological muse-
ums in Austria, the Czech Republic and Great
Britain, but the most important collection is that
of the Natural History Museum in Vienna, Aus-
tria.
The material described in this paper is housed
in the Department of Geology and Palaeontology
of the Naturhistorisches Museum (Natural Histo-
ry Museum, Vienna; NHMW), Geologische Bun-
desanstalt (the Austrian Geological Survey, Vien-
na; GBA), the Institute of Palaeontology
(University of Vienna; UV), Na
¤rodn|
¤Museum
(National Museum, Prague; NMP), and the Brit-
ish Museum of Natural History (London ;
BMNH).
The monocotyledon remains of the Gru
«nbach
plant assemblage are represented by leaf impres-
sions and compressions. The leaf fossils preserve
¢ne details of venation and morphology and show
little evidence of having undergone long distance
transport or decay prior to burial. Some of the
material yields fragments of cuticles.
The specimens were photographed under low
angle incident light using 35 mm Ilford FP4 ¢lm
in a Contax 167MM SLR camera with a Zeiss S-
Planar T* 2.8/60 macro lens. Besides normal
prints also large black-and-white prints (20U25
cm) were made and leaf outlines and venation
were drawn directly on them. Afterwards, the
photographic image was bleached away using an
aqueous solution of potassium iodide and iodine.
The bleached image was then conventionally ¢xed
leaving an ink drawing that was used for the sub-
sequent study.
Although leaf compressions appeared to be
well-preserved, cuticles obtained at maceration
were poor. Fragments of coali¢ed material ob-
tained by sampling were ¢rst treated with Schul-
ze’s reagent followed by treatment in low concen-
tration of KOH. Due to the high grade of
coali¢cation and chemical change, probably
caused by high calcium carbonate content in the
surrounding rock, it was only possible to obtain
small pieces of cuticle. The time in Schulze’s re-
agent was varied from 3 to 48 h, in order to de-
termine the optimal degree of oxidation. Best re-
sults were obtained when samples were treated for
10^24 h in Schulze’s reagent. The cuticles were
examined using a light microscope Olympus BX
50.
The descriptive terminology of monocotyledon
leaf morphology and venation used here follows
that published by Hickey and Peterson (1978) and
Mayo et al. (1997).
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353324

3. Geological setting and age of the Gru
«nbach £ora
The Gru
«nbach £ora derives from the Gru
«nbach
Formation (‘Coal-bearing Series’ according to
Ploechinger, 1961) of the Gosau Group in the
Gru
«nbach^Neue Welt Basin in the Eastern Cal-
careous Alps, Lower Austria (Fig. 1). The coal
seams of the Gru
«nbach Formation were exploited
from the second half of the nineteenth century
until the 1960s. Mining was extremely di⁄cult in
the highly tectonised basin and was ¢nally aban-
doned as uneconomic.
The Gru
«nbach^Neue Welt Basin of the Eastern
Alps represents a syncline with an overturned
limb (Ploechinger, 1961). The predominantly ter-
rigenous clastic ¢llings of the basin (Gosau
Group) consist of ¢ve lithostratigraphic units of
Late Santonian to Eocene age. The three lower
units, several hundred metres thick, represent
the Cretaceous part of the Gosau Group (Sum-
mesberger, 1997; Summesberger et al., 2000,
2002)(Fig. 2).
The Maiersdorf Formation consists of con-
glomerates, breccias, sandstones with rudist bio-
stromes, gastropod accumulations (Nerinea,Tro-
chactaeon) and brachiopod limestones. A Late
Santonian age of the Maiersdorf Formation has
been demonstrated by occurrences of Hippurites
(rudist biostrome near Gru
«nbach) as well as Pla-
centiceras polyopsis (Dujardin) and Cordiceramus
muelleri Petrascheck (Ploechinger, 1961; Summes-
berger et al., 2000). This dating is corroborated by
isotope study (probably Late Santonian according
to Sr86/Sr87 data : Kollmann et al., 2000).
The Gru
«nbach Formation, overlying the
Maiersdorf Formation, is composed of conglom-
erates, sandstones, siltstones, coaly siltstones and
coal seams. Plant fossils are the most common
fossils in the Gru
«nbach Formation. Foraminifera
from the Gru
«nbach Formation at Maiersdorf be-
long to the Globotruncana elevata Zone (Lower
Campanian) and the nanofossil association has
been assigned to the Campanian UC 15 Zone
(Hradecka
¤et al., 2000). Palynological samples
from a measured section of the Gru
«nbach Forma-
tion in the Segen Gottes coal mine of Gru
«nbach
were studied by Draxler (in Summesberger, 1997).
She identi¢ed 29 species of moss, lycopod, fern,
gymnosperm and angiosperm spores and pollen,
and emphasised that the most characteristic ele-
ments of the palyno£ora is pollen of the Norma-
polles group. A rich reptile fauna is also known
from this formation at Muthmannsdorf, where
carnosaur, iguanodon, pterosaur, scelidosaur
and crocodile remains were found (Bunzel,
1871^1873; Summesberger, 1997). The predomi-
nating palaeogeography during the accumulation
of the Gru
«nbach Formation is that of a large
island with unknown relief, at least temporarily
connected to the continent. Terrestrial freshwater
swamps and shallow water sediments indicate a
relatively large deltaic plain under warm and hu-
mid climate conditions.
The overlying Piesting Formation consists pre-
dominantly of sandstones and siltstones with ma-
rine fossils of Late Campanian and Maastrichtian
age. Foraminifera from the Piesting Formation at
Gru
«nbach can be attributed to the Globotruncana
ventricosa Zone (lower part of the Upper Campa-
nian). However, the major part of the Piesting
Formation yielded Late Campanian^Maastricht-
ian foraminifera (Hradecka
¤et al., 2000). Large
Orbitoides foraminifera are very common in the
so-called Orbitoides Sandstone in the lower part
of the formation. Nanofossil associations from
the Piesting Formation belong to UC 16, UC 17
and UC 18^?UC 19 zones which characterise the
Campanian^Lower Maastrichtian stratigraphic
interval (Hradecka
¤et al., 2000). The Late Cam-
panian ammonite Pseudokossmaticeras brandti
Fig. 1. Position of the Gru
«nbach^Neue Welt Basin in the
Eastern Alps, Austria (simpli¢ed from Summesberger, 1997
and Herman and Kvac
›ek, 2002a).
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 325

(Redtenbachter) occurs at Gru
«nbach, the Early
Maastrichtian Pachydiscus epiplectus (Redten-
bachter) at Muthmannsdorf (Summesberger,
1997), and the Late Campanian belemnite Belem-
nitella hoeferi (Schloenbach) at Gru
«nbach (Chris-
tensen, 1998). Triporate angiosperm pollen of
the Normapolles group allows Hradecka
¤et al.
(2000) to assign some samples from the Piesting
Fig. 2. Geological section of the Gosau Group in the Gru
«nbach^Neue Welt Basin in the Eastern Alps, Austria (simpli¢ed from
Summesberger et al., 2000 and Herman and Kvac
›ek, 2002a), and the stratigraphic position of the plant-bearing beds of the
Gru
«nbach £ora.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353326

Formation near Gru
«nbach to the Late Campa-
nian.
A rich fauna recently discovered in the Piesting
Formation at the Piesting sports ¢eld yielded di-
verse Maastrichtian inoceramids; the ammonite
Pachydiscus neubergicus indicates Early to early
Late Maastrichtian age (Tro
«ger et al., 2000,
2001; Kollmann et al., 2000). Palaeomagnetic
study of the section demonstrates that it probably
correlates with Chron 31R just above the Campa-
nian^Maastrichtian boundary (Tro
«ger et al., 2000,
2001). The neighbouring outcrop at roadcut S of
Piesting yielded Trochoceramus cf. morgani and T.
cf. dobrovi of latest Campanian^Early Maastricht-
ian age (Tro
«ger et al., 2000, 2001).
The Early Campanian age of the Gru
«nbach
£ora is therefore based on foraminifera, spores
and pollen from the Gru
«nbach Formation and
on the correlation of the plant-bearing deposits
with the underlying Maiersdorf and the overlying
Piesting formations dated by stratigraphically im-
portant marine fossils (Fig. 2).
The Gru
«nbach £ora consists of approximately
60 taxa assigned to algae, possible liverworts,
horsetails, ferns, cycadophytes, conifers, monoco-
tyledons and dicotyledons. With regard to the
number of species it is dominated by angiosperms
(about 70% of species) followed by ferns, conifers
and other groups of plants. The Gru
«nbach vege-
tation experienced a humid sub-tropical to warm-
temperate climate with warm/hot summers and
short relatively dry seasons (Herman and Kvac
›ek,
2000, 2002a,b).
4. Systematics
Monocotyledons
Alismatales
Araceae
Araciphyllites Wilde, Z. Kvac
›ek et Bogner 2003
Type:Araciphyllites tertiarius (Engelhardt) Wilde
et al., 2003
According to Wilde et al. (2003), the morphoge-
nus Araciphyllites is characterised by the eucamp-
todromous parallel-pinnate venation with a multi-
stranded midvein, laterals loosely di¡erentiated
into three orders, transversal veins oblique and
irregular forming polygonal-quadrangularly elon-
gate areoles and 1^3 submarginal veins formed by
terminal fusion of laterals, characters which were
used to accommodate the genus in the Araceae.
Araciphyllites shows a venation pattern similar
to Lysichiton and Orontium, the primitive mem-
bers of the Orontioideae subfamily of the Ara-
ceae. The genus Lysichiton (Plate IV, 4,5), in par-
ticular, is remarkably similar to A. austriacus
showing an incomplete wide multistranded mid-
rib. A. austriacus di¡ers from Lysichiton in having
a higher number of lateral vein orders and two
orders of transversal veins. The similarity of Ara-
ciphyllites to other recent monocotyledon families
is less pronounced. Members of the Alismataceae,
Hydrocharitaceae and Limnocharitaceae di¡er
from A. austriacus in lacking acrodromous venat-
ion with reticulate secondaries and simple midrib.
The genus Araciphyllites di¡ers from fossil taxa
Caladiosoma Berry (1925),Nitophyllites Iljinskaja
(1963) and Mussopsis Boyd (1992) in having
poorly distinguished primary laterals (ribs) and
simple patterns of the venation between them.
For more comparisons see Wilde et al. (2003).
Araciphyllites austriacus Kvac
›ek et Herman, sp.
nov. Fig. 3;Plate I, 1,2,4
Holotype designated here: NHMW 1999B0057/
0183 (Plate I, 1^4).
Locus typicus: Gru
«nbach am Schneeberg, Segen
Gottes Quarry (47‡47P52QN, 15‡59P17QE).
Stratum typicum:Gru
«nbach Formation, Lower
Campanian.
Etymology: Named after the country of origin.
Diagnosis: Leaves large, simple ; elongate-lanceo-
late, entire-margined. Apex and base unknown.
Venation eucamptodromous, parallel-pinnate.
Midrib massive, deeply impressed, probably
reach- ing the apex, consisting of numerous vas-
cular bundles. Laterals emerging at low angle
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 327

PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353328

from midrib. Parallel venation consisting of three
weekly di¡erentiated vein subsets arranged ac-
cording to the formula BdCdB. Transverse vein
sets of two categories, connecting adjacent paral-
lel veins or veins of the same subset (C^C, d^d)
being oriented obliquely or perpendicularly. Are-
oles polygonal-quadrangular, elongate.
Description: The only specimen available is the
holotype (Fig. 3A;Plate I, 3). It shows thin, lan-
ceolate elongate leaf lamina 260 mm long, nar-
rowing from more than 100 mm in the terminal
part to 60 mm in the basal part. The leaf margin
is entire, slightly bent to undulate in the basal
part. The midrib narrows from the base (12 mm
wide) to the apex (3 mm in width). The laterals in
three poorly di¡erentiated orders emerge at an
angle of about 10‡ from a multistranded midvein
(Fig. 3B;Plate I, 1). Each order has veins of
speci¢c width: B, 0.12^0.10 mm; C, 0.08^0.06
mm; d, 0.05^0.03 mm. Spacing of the parallel
veins varies from 0.8 to 1.2 mm (Plate I,2).
Transverse veins vary in width (0.10^0.03 mm) ;
they seem to form two width orders 0.08^0.06
mm and 0.05^0.02 mm. The ¢rst set frequently
connects two adjacent veins ‘C’, the second con-
nects veins ‘d’. Both transverse vein orders are
straight or bent, sigmoidal or s-shaped, spaced
Plate I. Araciphyllites austriacus J. Kvac
›ek et Herman sp. nov., Gru
«nbach, NHMW 1999B0057/0183
1. Detail of venation, note multistrained costa on right and sharp angle of outgoing veins, detail of ¢g. 2, U3.
2. Holotype, U0.6.
4. Detail of ¢g. 1, venation with alveoles, U4.5.
Lysichiton americanus Hulte
¤n et H. St. John
3. Veins near costa, U3.
5. Detail of ¢g. 3 venation with alveoles, U4.5.
Fig. 3. Araciphyllus austriacus J. Kvac
›ek et Herman, sp. nov., Gru
«nbach (scale bars represent 1 cm). (A) NHMW 1999B0057/
0183, leaf outline and venation (line drawing of Plate I, 1). (B) NHMW 1999B0057/0183, detailed venation (line drawing of Plate
I, 1).
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 329

PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353330

0.5^2.8 mm apart. They are perpendicularly or
obliquely oriented to the parallel veins forming
polygonal-quadrangularly elongate areoles paral-
lel to laterals (Fig. 3B ;Plate I, 2).
Discussion:Araciphyllites austriacus shows a mul-
tistranded midrib, poorly di¡erentiated laterals
and perpendicular veins forming areoles ^ charac-
ters, which ¢t well with the diagnosis of the mor-
phogenus Araciphyllites (Wilde et al., 2003). Ara-
ciphyllites austriacus di¡ers from the type of the
morphogenus in two respects : (1) in having two
orders of transverse veins, and (2) lacking prom-
inent submarginal vein anastomoses along the leaf
margin. However, the absence of anastomosing is
probably due to poor preservation of the margin-
al venation of the specimen at hand.
There are several poorly preserved monocotyle-
don leaves described from the Late Cretaceous of
Germany recalling in some characters Araciphyl-
lites austriacus.Zingiberopsis riggauensis Kno-
bloch (1979) from the Santonian of Oberpfalz
shows ¢ve orders of lateral veins, accompanied
by transverse veins connecting more than two ad-
jacent parallel veins. Unfortunatelly, the leaf frag-
ment of Z. rigauensis does not show a midrib.
Another similar leaf was described by Ru
«¥e and
Trostheide (2000) from the Santonian of Quedlin-
burg under a name cf. Pontederia montensis Fritel.
The material is poorly documented and needs
more detailed study.
Pandanales
Pandanaceae
Pandanites Tuzson 1913, p. 219
(non Pandanites Dorf 1942, p. 46, nom. illegit.)
Syn:Pandanophyllum Kryshtofovich 1929, p.
120(1364), nom. illegit. (non Pandanophyllum
Hasskarl ex Steudel 1855, p. 134)
Type:Pandanites acutidens Tuzson 1913, p. 219,
pl. 15, ¢g. 6
The holotype of P. acutidens was not available for
the present study. It is not located in the Natural
History Museum in Budapest (L. Hably, pers.
commun., 2001), nor in the University of Cluj
Napoca (I. Petrescu, pers. commun., 2001). We
used for comparison the drawing by Tuzson
(1913, 1914) and also the material of Pandanites
trinervis, which is considered here to be conspe-
ci¢c with the type of the morphogenus.
Emended diagnosis: Strap-like leaves, M-shaped in
transverse section. Parallel veins of one order (of
the same width), besides the midrib, transverse
veins thin, usually obliquely oriented. Leaves
with marginal spines, which also occur on keel
on the abaxial side of the leaf. Leaf apex attenu-
ate, base auriculate.
Discussion: The morphogenus Pandanites is
emended herein in order to accommodate all fos-
sil leaves with a⁄nities to the Pandanaceae. Pan-
danites is assigned to the Pandanaceae on the fol-
lowing macroscopical characters : strap-like,
parallel veined, typically armed leaves, M-shaped
in transverse section bearing one vein order. For
the Pandanaceae, the combination of leaf charac-
ters is unique leading, Stone et al. (1998) to state
explicitly: ‘Although very variable, the Pandana-
ceae can be readily recognised by their leaves
(shape, anatomy and phyllotaxis)’. This state-
ment, if accepted, gives us a powerful tool for
recognition of the Pandanaceae in the fossil rec-
ord. Also comparison of cuticle of Pandanites and
recent representatives of the Pandanaceae reveals
numerous similarities: (1) similar arrangements of
stomata in stomatal belts; (2) tetracytic type of
stomata with two subsidiary cells in lateral and
Plate II. Pandanites trinervis (Ettingshausen) J. Kvac
›ek et Herman comb. nov., Gru
«nbach
1. Holotype, von Ettingshausen, 1852, pl. 26 (IV), ¢g. 1, GBA 1852/02/01, U1.
2. Syntype of Pandanus pseudo-inermis von Ettingshausen, 1852, pl. 25(II), ¢g. 1, GBA 1852/02/02, U0.9.
3. Syntype of Pandanus pseudo-inermis von Ettingshausen, 1852, pl. 24(II), ¢g. 2, GBA 1852/02/03 U0.9.
4. Holotype, detail of venation, GBA 1852/02/01, U3.7.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 331

PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353332

two in polar position; (3) shape and arrangement
of ordinary cells.
The morphogenus Pandanites was described by
Tuzson in 1913 together with a single species P.
acutidens (diagnosis generico^speciphica), but it
did not become widely used. Instead, new names
have been introduced for later recorded fossil fo-
liage compared with the recent Pandanaceae.
Dorf (1942) may not have been aware of the Tuz-
son paper and therefore created a homonym Pan-
danites Dorf for fossil leaves with a⁄nity to the
Pandanaceae (based on the species P. corsonii
from the Upper Cretaceous of Wyoming).
The genus Pandanophyllum Kryshtofovich
(1929) from the mid-Cretaceous of Siberia is
based on specimens showing characters of the
Panadanaceae. The M-shaped transverse section
and marginal spines are clearly pronounced and
also mentioned in the text by Kryshtofovich
(1929, pl. 59, ¢gs. 1^3). Unfortunately, the name
Pandanophyllum Kryshtofovich is homonymous
with the name of a recent plant genus Pandano-
phyllum Hasskarl ex Steudel 1855 and is therefore
illegitimate.
Similar to Pandanites is a monotypic genus
Bactrites described by Berry (1924) from the
Upper Eocene of southeastern North America.
It is frequently assigned to palms (Berry, 1924 ;
Read and Hickey, 1972). However, arguments
for palm a⁄nity of Bactrites are not robust. Berry
(1924) states in the beginning of his diagnosis,
that Bactrites has pinnate leaves, yet all the type
material consists only of separate leaf fragments.
Read and Hickey (1972), although keeping Bac-
trites as a palm morphotaxon, declare that true
fossil palm leaves should have at least two orders
of veins. The leaves of Bactrites show only one
order of veins. This character and strap-like
armed leaves of Bactrites argue more for its a⁄n-
ity to the family Pandanaceae.
There are several recent genera displaying var-
ious characters similar to Pandanites. Some mem-
bers of the Hydrocharitaceae (e.g. Stratiotes, see
Ancibor, 1979) and Bromeliaceae show armed
leaves. The genus Puya Molina (Bromeliaceae)
has spinose-serrate leaves and auriculate bases.
Other monocotyledons with serrate or spiny
leaves (Blyxa,Nechamandra^Hydrocharitaceae,
Najas^Najadaceae, Thurnia^Thurniaceae) have
only minute marginal teeth and transversally £at
or V-shaped leaves. All the above-mentioned
monocotyledons di¡er from Pandanites in the ab-
sence of M-shaped transversal sections and armed
keels of leaves.
Pandanites trinervis (Ettingshausen) J. Kvac
›ek et
Herman, comb. nov. Fig. 4 ;Plate II, 1^4; Plate
III, 1^5; Plate IV, 1^9 ; Plate V 1^5
Basionym:Pandanus trinervis Ettingshausen 1852,
U
ºber fossile Pandaneen. Sitzungberichte der
math.-naturw. Classe der kais. Akademie der Wis-
senschaften VIII Bd., p. 494, pl. 26, ¢g. 1
Pandanus austriacus Ettingshausen 1852, p. 492,
pl. 23(I), ¢g. 1
Pandanus pseudo-inermis Ettingshausen 1852, p.
493, pl. 24(II), ¢gs. 1 and 2; pl. 25(II), ¢g. 1
Pandanus trinervis Ettingshausen; Schimper,
1870^72, p. 476
Pandanus austriacus Ettingshausen; Knobloch,
1977, p. 417
Pandanites acutidens Tuzson 1913, p. 219, pl. 15,
¢g. 6.
Pandanites acutidens Tuzson 1914, p. 245, pl. 15,
¢g. 6.
Pandanus spinatissimus Petrescu et Dusa 1980, pp.
149, 151, ¢g. 3, nom. nud.
Pandanus barburi Petrescu et Dusa 1980, pp. 149,
151, nom. nud.
Plate III. Pandanites trinervis (Ettingshausen) J. Kvac
›ek et Herman comb. nov., Gru
«nbach
1. The longest leaf fragment armed with unfrequent prickles, NHMW 1999B0057/0239, U0.5.
2. Attenuate leaf apex, NHMW 1999B0057/0584, U0.6.
3. M-shaped transversal section of leaf, NHMW 1999B0057/0102, U3.
4. M-shaped transversal section of leaf base, GBA 2001/06/02, U1.3.
5. Auriculate leaf base, NHMW 1999B0057/0285, U1.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 333

PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353334

Pandanus tenuissimus Petrescu et Dusa 1980, pp.
149, 151, nom. nud.
Holotype: GBA 1852/02/01, Ettingshausen 1852,
pl. 26(IV), ¢g. 1, (Plate II,1)
Stratum typicum:Gru
«nbach Formation, Lower
Campanian
Locus typicus: Dreista
«tten, Lower Austria,
47‡51P28QN, 16‡06P05QE.
Occurrence: Austria: Gru
«nbach am Schneeberg,
Muthmannsdorf; Gru
«nbach Formation, Gosau
Group, Early Campanian; Geistthal : Gosau
Group, ?Campanian ; Romania : Rusca Montana ;
Maastrichtian.
Material: GBA 1852/02/02, GBA 1852/02/03,
GBA 1852/03/05b, GBA 2001/06/02 and 2 un-
numbered specimens in GBA, NHMW 1970/
1396/625, NHMW 1970/1396/614, NHMW
1999B0057/0031, NHMW 1999B0057/0041,
NHMW 1999B0057/0061, NHMW 1999B0057/
0102, NHMW 1999B0057/0151, NHMW
1999B0057/0154 ^ 0156, NHMW 1999B0057/
0158, NHMW 1999B0057/0167 ^ 0169, NHMW
1999B0057/0172, NHMW 1999B0057/0174,
NHMW 1999B0057/0176 ^ 0181, NHMW
1999B0057/0184, NHMW 1999B0057/0186,
NHMW 1999B0057/0188, NHMW 1999B0057/
0190, NHMW 1999B0057/0192, NHMW
1999B0057/0196, NHMW 1999B0057/0197,
NHMW 1999B0057/0209, NHMW 1999B0057/
Plate IV. Pandanites trinervis (Ettingshausen) J. Kvac
›ek et Herman comb. nov., Gru
«nbach
1. Leafy branch with three or more leaves attached, NHMW1999B0057/176, U1.
2. Prickle on abaxial side of midrib ^ longitudinally broken leaf, NHMW1999B0057/549, U5.
3. Large marginal prickles NHMW1999B0057/534, U3.
4. Delicate marginal prickles, NHMW1999B0057/549, U2.2.
5. Adaxial cuticle, stoma arrowed, NHMW1999B0057/0541b, U200.
6. Abaxial cuticle, stomata arrowed, NHMW1999B0057/0541c, U200.
9. Stoma, NHMW1999B0057/0541c, U400.
Pandanus edulis Thou.
7. Abaxial cuticle, U200.
8. Stoma, U400.
Fig. 4. Pandanites trinervis (Ettingshausen) J. Kvac
›ek et Her-
man, comb. nov., Gru
«nbach (scale bars represent 1 cm). (A)
NHMW 1999B0057/0041, leaf margin and ¢ne venation near
margin. (B) NHMW 1999B0057/0154, leaf venation. (C)
NHMW 1999B0057/0285, leaf base, venation shown sche-
matically (line drawing of Plate III, 5).
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 335

PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353336

0213, NHMW 1999B0057/0239, NHMW
1999B0057/0259, NHMW 1999B0057/0260,
NHMW 1999B0057/0280, NHMW 1999B0057/
0284, NHMW 1999B0057/0285, NHMW
1999B0057/0289, NHMW 1999B0057/0331,
NHMW 1999B0057/0364, NHMW 1999B0057/
0435, NHMW 1999B0057/0474, NHMW
1999B0057/0512, NHMW 1999B0057/0528,
NHMW 1999B0057/0529, NHMW 1999B0057/
0534 ^ 0546, NHMW 1999B0057/0549, NHMW
1999B0057/0584, K 418, K 419, several unnum-
bered specimens in UV.
Emended diagnosis: Leaves spirally arranged,
3-ranked, linear-ensiform, with spines on margins
and abaxial face of the midrib. Apex attenuate;
base sheathing, auriculate. Midrib not always visi-
ble, parallel vein set built of one order of thick
veins, interconnected by thin transverse veins ori-
ented obliquely or perpendicularly to the margins.
Adaxial cuticle showing elongate ordinary cells
with rare stomata, abaxial cuticle with numerous
stomata arranged in wide bands, stomata on both
cuticles tetracytic, each guard cell pair surrounded
by 2 polar and 2 lateral cells.
Description: The holotype of Pandanites trinervis
(Plate II, 1), consisting of three leaf fragments in a
position, suggests their original emergence from a
shoot. Each of the fragments is linear (23^25 mm
in width), with marginal pricles 1 mm long, rela-
tively regularly distributed at distances of 3^5 mm
apart on both leaf margins. There is only one vein
order parallel to the margin (distance between
veins 0.3^0.5 mm), thin transversal veins connect
primaries at oblique, or rarely right angles (Plate
II, 4).
There are numerous specimens in the collection
of NHMW showing various parts of leafy shoots,
leaf apices, bases and leaf margins of Pandanites
trinervis. The whole terminal parts of branches
are preserved in Nos. NHMW 1999B0057/0179,
NHMW 1999B0057/0176, NHMW 1999B0057/
0192 showing a spiral orientation of leaves (Plate
IV, 1). Variously sized fragments of leaves are the
most common type of leaf fossils in the Gru
«nbach
locality. The longest leaf is 396 mm long and 28^
40 mm wide (Plate III, 1). The maximum width of
the lamina among the studied leaves is 94 mm
(No. NHMW 1999B0057/0280). Leaves are usu-
ally preserved compressed or impressed on the
bedding plane. Transversal M-shape sections of
leaves are occasionally preserved (Plate III, 3,4).
The sheathing leaf base of P. trinervis is open
(NHMW 1999B0057/0181, NHMW 1999B0057/
0543 and GBA 2001/06/02) or nearly auriculate
(NHMW 1999B0057/0285; Fig. 5C;Plate III,
5). The apex is attenuate showing a drip-tip (Plate
III, 2). Nearly all leaves are to some degree armed
with prickles, but leaves without prickles were
also recorded. Prickles occur on margins (Plate
IV, 3,4) and abaxial face of the midrib (Plate
IV, 2). Considerable variation in the size of mar-
ginal prickles (0.5^3 mm) and their spacing (2^20
mm) are documented in Fig. 5 and Plate IV, 3,4.
The midrib is well pronounced possibly being
scleri¢ed. Similar ¢bres occur in the middle of
each half of the leaf lamina (in transversal section
in a place of sinuses of ‘M’). The ¢bres are pre-
served, e.g. in the holotype (Plate II, 1,4) and in
the specimens NHMW 1999B0057/0154 (Fig. 4),
NHMW 1999B0057/0528, running parallel to
veins, and seemingly showing a second, thicker
order of veins (Plate II,4).
Plate V. Pandanites trinervis (Ettingshausen) J. Kvac
›ek et Herman comb. nov., Gru
«nbach
1. Stoma on adaxial cuticle showing long polar subsidiary cells, NHMW1999B0057/541b, Nomarski DIC, U750.
Pandanus weitchii Dall.
2. Stoma on abaxial cuticle showing short polar subsidiary cells, U1000.
3. Detail of venation showing perpendicular and rarely oblique (arrowed) orientation of secondary veins, U5.
4. Darker strips consisting of scleritic ¢bres, leaf after one year maceration, U10.
5. Half a leaf lamina showing ¢brous strip in central part, after one year maceration, U5.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 337

Venation of many leaves is not easily discerni-
ble due to thick coal matter. Only some specimens
(e.g. the holotype and No. NHMW 1999B0057/
0154), which were probably naturally macerated
before fossilisation, show the venation clearly
(Fig. 4B;Plate III, 4). The leaf lamina exhibits
only one order of veins (0.05^0.3 mm thick) and
obliquely or perpendicularly running transversal
veins (0.01 mm thick). The density of veins is
usually 16^30 per cm. Narrower parts of leaves
show a typically higher vein density than the
wider parts. The lowest density of eight veins
per cm was recorded in one leaf base (GBA
2001/06/02).
Although numerous specimens appear to be
leaf compressions, only a small number yield
good cuticles. They usually su¡er some degree
of corrosion. The abaxial cuticle is usually more
a¡ected. The specimen NHMW 1999B0057/0541
provides an example of an exceptionally well-pre-
served leaf compression, from which excellent cu-
ticles of both leaf sides were obtained. The adaxial
cuticle shows elongate ordinary cells (5^12U25^
65 Wm) with straight or bent anticlinal walls 2^5
Wm thick (Plate IV, 5) and isodiametric cells re-
sembling crystal cavities (10U10 Wm). Isolated
tetracytic stomata which sometimes occur in short
rows (Plate V, 1) have two later subsidiary cells
and two unspecialised polar cells (guard cells 3^
5U20^25 Wm; subsidiary cells 10^15U15^34 Wm).
The abaxial cuticle shows wide intercostal bands
and narrow costal bands of 5^20 cells (Plate IV,
6). Intercostal bands consist of isodiametric or
elongate ordinary cells (5^12U16^55 Wm) and nu-
merous longitudinally oriented tetracytic stomata.
Each stoma typically consists of two guard cells
(2^3U12^15(25) Wm) and four unspecialised sub-
sidiary cells (Plate IV, 9), two of which are in
lateral (7^12U25^40 Wm) and two in polar (10^
20U15^25(45) Wm) positions. In some stomata
one lateral cell may be replaced by two smaller
ones (Plate IV, 9). Hypodermis consists of isodia-
metric polygonal cells.
Discussion: Schenk (in Zittel, 1890) and Berry
(1924) were sceptical about the a⁄nity of the fos-
Fig. 5. Sabalites longirhachis (Unger) J. Kvac
›ek et Herman, comb. nov., detail of venation, veins ‘d’ and some ‘C’ shown sche-
matically, detail of Plate VII, 3, Gru
«nbach, NHMW 1999B0057/0528 (scale bar represents 1 cm) U6.7.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353338

sils described by von Ettingshausen (1852) to the
Pandanaceae. Schenk apparently did not observe
transversal veins, which are in our opinion quite
well pronounced, particularly in Pandanites triner-
vis. He also pointed out that densely spaced thin
veins might imitate thicker veins. Our new inves-
tigations of the type material of Ettingshausen
and numerous subsequently collected specimens
of Gru
«nbach Pandanaceae brought new light to
the problem. We have observed and described
additional characters ^ leaves spirally attached
on an axis forming growing apex, auriculate
bases, attenuate apices, spines on the abaxial
part of the midrib, tetracytic stomata, which un-
doubtedly con¢rm the a⁄nity of P. trinervis to
the Pandanaceae.
Pandanites trinervis, which is emended here, is
the most typical among the three species described
initially by von Ettingshausen (1852) from the
Campanian of Gru
«nbach. Two remaining species,
described by Ettingshausen in the same publica-
tion (von Ettingshausen, 1852), Pandanus austria-
cus, the holotype of which is currently of un-
known repository and P. pseudo-inermis are
poorly preserved showing small di¡erences to
Pandanites trinervis. Considering the ¢gure and
Ettingshausen’s description, it can be seen that
the di¡erences between Pandanus austriacus and
Pandanites trinervis are in size of spines and thick-
ness of veins, which we consider within the scope
of species variation. Leaves of recent representa-
tives of the Pandanaceae show even higher vari-
ability in size of spines at one species. The thick-
ness of veins and their number seem to be very
variable within the leaf of P. trinervis. The thick-
ness varies from 0.6 to 0.2 mm, and the density
ranges from 8 to 20 veins per cm, depending on
the position in the leaf. According to von Etting-
shausen (1852),P. trinervis di¡ers from P. pseudo-
inermis in the presence of only one vein order.
However, the thicker veins described in P. pseu-
do-inermis and also three thick parallel strands in
P. trinervis do not appear to be true veins. We
have studied more than 70 fossil specimens and
determined that preservation controls the appear-
ance of fossil leaves of the Pandanaceae. Addi-
tionally, we decided to run a one-year lasting ex-
periment of maceration of recent Pandanus leaves
in KOH. The leaf parenchyma gradually degraded
and the leaf became progressively more translu-
cent. Its sclerenchymatic ¢bres were not a¡ected,
becoming more clearly visible (Plate V, 4), partic-
ularly in a place of sinuses of ‘M’ (Plate V, 5).
Therefore, we suggest that the three veins of P.
trinervis and also the thicker veins of P. pseudo-
inermis represent actually ¢bres, which became
more pronounced in certain stages of natural
maceration in water (compare Plate II, 4 and
Plate V, 5). They are interpreted as ¢brous tissues
supporting the leaf mechanically in its most ex-
posed parts.
There are several occurrences of the Pandana-
ceae in the Cretaceous of the Gosau Group in
Austria (Summesberger, pers. commun., 2002).
Leaves from the vicinity of the village of Geistthal
(Gosau Group) were described by Knobloch
(1977). He assigned fragments of fossil leaves
(housed in the NMP, Nos K 118, K 119) to ‘Pan-
danus austriacus’. The specimens were revised and
compared to Pandanites trinervis. Although leaves
are poorly preserved, showing no good marginal
prickles, their slightly M-shaped transversal sec-
tions allow to interpret them as poorly preserved
leaves of P. trinervis.
Pandanites acutidens Tuzson (1913) is treated
here as synonymous to P. trinervis. It shows a
similarly M-shaped transversal section, the same
venation pattern and well-preserved marginal
prickles. In addition to Pandanites acutidens (Tuz-
son, 1913), several pandan leaves were described
from the Maastrichtian of Rusca Montana in Ro-
mania by Petrescu and Dusa (1980). All of them :
Pandanus spinatissimus Petrescu and Dusa 1980,
P. barburi Petrescu and Dusa 1980, P. tenuissimus
Petrescu and Dusa 1980 were invalidly published
as nomina nuda. They probably also represent var-
ious fragments of P. trinervis. The illustration of
P. spinatissimus in Petrescu and Dusa (1980, ¢g.
3) seems to show the apical part of P. trinervis
leaf showing denser marginal prickles.
Pandanophyllum ahnertii Kryshtofovich from
the Aptian of the Far East (Kryshtofovich,
1929, pl. 59, ¢gs. 1^3; age con¢rmed by Krassi-
lov, 1967) exhibits linear armed leaves with mar-
ginal prickles and M-shaped transversal section of
lamina. It is considered to be very close to Pan-
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 339

danites trinervis. Nevertheless, more study of the
material is needed for comparison.
Pandanites corsonii Dorf from the latest Creta-
ceous of the Rocky Mountains (Dorf, 1942, pl. 3,
¢g. 1) shows characters similar to Pandanites tri-
nervis, including the M-shaped cross section of
leaf lamina and venation pattern. It is di⁄cult
to ¢nd signi¢cant di¡erences between these two
fossils. A minor di¡erence is in the size of margin-
al prickles. P. trinervis has marginal spines 0.5^3
mm long, while P. corsonii has spines 5^6 mm
long. The latter also lacks spines on the abaxial
surface of the midrib, although it is di⁄cult to
clarify this character in the fossil state, a problem
already experienced in our studies of Pandanites
trinervis.
A probable new species of Pandanites is re-
corded by Upchurch and Mack (1998) from the
Maastrichtian of New Mexico. The authors men-
tion, without any illustration, the presence of
‘three longitudinal folds’ within an armed leaf.
This combination of characters is typical of trans-
versely M-shaped foliage of the Pandanaceae.
Bactrites pandanifoliolus Berry (1924, pl. 7, ¢gs.
1^6) is suggested here as possibly comparable
with Pandanites. It di¡ers from P. trinervis in hav-
ing slender marginal prickles, which are 1^3.5 mm
long, in the lack of a M-shaped transversal sec-
tion of the lamina and in the lack of spines on the
abaxial surface of the midrib. On the other hand,
it is very similar to P. trinervis in its venation
pattern, showing only one venation order (com-
pare Berry, 1924 pl. 7, ¢gs. 5, 6; Fig. 4 ;Plate II,4
herein). The lamina of B. pandanifoliolus is divid-
ed into irregular strips showing only one margin
without teeth. The important di¡erential charac-
ter, the M-shaped transverse section, may be lack-
ing due to taphonomical conditions : long trans-
portation and long natural maceration before
fossilisation. The delicate nature and preservation
of B. pandanifoliolus recalls some of the specimens
of P. trinervis recorded in the Gru
«nbach For-
mation. The leaves after maceration were prob-
ably much softer and could be squashed and £at-
tened. However, more complete material of
B. pandanifoliolus is needed to establish de¢nite
conclusions.
Most Tertiary leaves with a⁄nity to the Panda-
naceae were described in the 19th century from
southern Europe. Pandanus ettingshausenii Squi-
nabol (1891, pl. 16, ¢gs. 4, 4a, 4b) and P. inter-
medius Laurent (1899, pl. 4, ¢gs. 2, 2a) show
armed leaves with one order of veins recalling
the Pandanaceae, but do not exhibit the M-
shaped transversal section. P. lutetianus Bureau
(1888, pl. 22, ¢gs. 1^3) is poorly preserved. All
three species were already considered to be dubi-
ous by Schenk (in Zittel, 1890) and Berry (1924).
Their type material needs to be revised.
Pandanus rhenanus Kra
«usel et Weyland from
the European Miocene (Kra
«usel and Weyland,
1950; Peters, 1963; Palamarev and Usunova,
1969) is based on leaf fragments lacking diagnos-
tic characters of the foliage of Pandanaceae (M-
shaped transversal section), however its cuticle
and mesophyll structure do recall the leaves of
the members of the family. On the other hand,
Freycinetia rhenana Weyland, from the same stra-
ta (Weyland, 1957; Palamarev and Usunova,
1969), based on dispersed cuticles, di¡ers in the
type of stomatal apparatus from that of the Pan-
danaceae (Tomlinson, 1965, 1974). Overall leaf
morphology is needed to clarify both these enig-
matic plant fossils.
The only unequivocal representative of the Ter-
tiary Pandanaceae was described by Kra
«usel
(1929) under the name of a recent species Panda-
nus helicopus Kurz from the Neogene of Sumatra.
It shows a typical venation and M-shaped trans-
versal section of the leaf, which allows it to be
assigned to the family Pandanaceae. The generic
determination remains questionable.
Comparison of Pandanites trinervis with recent
leaves of Pandanus shows large similarities in both
macromorphology and anatomy. Similarity of
marginal prickles and venation are clear from
the illustrations (Plate V, 3). Recent pandans
(e.g. Pandanus veitchii Dall.) di¡er slightly in hav-
ing transversal veins, which are predominantly
near right-angled, but oblique transversals also
occur (Plate V, 3, arrowed).
Leaves of the recent Pandanaceae are extremely
£exible and robust at the same time. This charac-
ter is caused by the special construction of Pan-
danus leaf consisting of three layers of hypoder-
mis, having a transversely M-shaped construction
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353340

of the leaf and sclerenchymatic ¢bres supporting
the most mechanically exposed parts : the midrib,
folding and leaf base (e.g. Plate V, 5). These scler-
enchymatic ¢bres were probably also present in
fossil representatives of the family. The occur-
rence of three longitudinal bands in leaves of P.
trinervis and the presence of a false extra vein set
in the Ettingshausen’s P. pseudo-inermis are inter-
preted here as remains of sclerenchymatic ¢bres
(see above, compare Plate V, 4,5). It is clear from
our observations that this unique leaf construc-
tion of the Pandanaceae was already present in
the Cretaceous members of the family.
Stomata of most recent representatives of the
genus, e.g. Pandanus veitchii Dall. and P. edulis
Thou., are surrounded by more specialised subsid-
iary cells with papillae. Subsidiary polar cells of
Pandanites trinervis (Plate IV,9;Plate V, 1) are
not usually specialised ^ di¡erentiated into small,
more isodiametric thickly cutinised cells (compare
Tomlinson, 1965), in contrast with polar cells of
P. veitchii and P. edulis (Plate IV,8;Plate V, 2).
Unspecialised polar cells are con¢ned to some re-
cent species of Pandanus and to the genus Sara-
ranga (Tomlinson, 1965).
Interpretations and remarks: The family Pandana-
ceae shows numerous ancestral characters and is
considered to be quite ancient (Cox et al., 1995).
It consists of three recent genera: Pandanus L.
(700 species in Indo^Paci¢c region), Freycinetia
Gaudichaud-Beaupre
¤(200 species in Austral^
Asian region) and Sararanga Hemsley (2 species
in the Philippines and Melanesia) (Stone et al.,
1998). The genus Sararanga is considered to be
the most basal in the Pandanaceae clade (Dahl-
gren et al., 1985; Hotton et al., 1994). Within the
genus Pandanus there are about 60 sections occur-
ring mostly in Old world tropics from West Africa
to Polynesia. The subgenus Martellidendron from
Madagascar is considered as the most basal with-
in the genus (Callmander, 2001). Both Martelli-
dendron and Sararanga show reticulate pollen
grains with small lumina, characters, which are
regarded as ancestral not only within the family
but for all monocotyledons and angiosperms (e.g.
Walker and Walker, 1987).
Occurrence of fossil members of the Pandana-
ceae in the Late Cretaceous together with primi-
tive characters recorded in the recent representa-
tives (as absence of style, incompletely sealed
carpels; Stone et al., 1998), and molecular data
(Chase et al., 2000), support the assumption of
great antiquity of the family.
Arecales
Arecaceae
Sabalites Saporta 1865
Type: Sabalites major (Unger) Saporta 1865: 83,
pl. 2
Syn.
Geonomites Lesquereux 1878, p. 115, pro parte,
(non Geonomites de Visiani 1864, p. 456)
Juranyia Tuzson 1908, p. 1
Sabalites was emended by Read and Hickey
(1972), who also selected a new lectotype (con-
trary to Andrews, 1955) and whose classi¢cation
is followed here in. The genus is assigned to the
Arecaceae based on the costapalmate leaves. We
use the generic name Sabalites instead of Pal-
mophyllum Conwentz (1886), which was used by
Baikovskaya (1965) for similar Romanian materi-
al from the Maastrichtian of Transylvania. The
holotype of the genus Palmophyllum is a small
fragment with parallel veins, probably represent-
ing a monocotyledon leaf without diagnostic fea-
tures of palms and therefore excluded from the
Arecaceae (Read and Hickey, 1972). The genus
Juranyia Tuzson (1908) from the Maastrichtian
of Romania shows costaplmate lamina, the diag-
nostic characters of Sabalites and is therefore
synonymised here in. For other discussion as-
sociated with the genus, see Read and Hickey
(1972).
Sabalites longirhachis (Unger) J. Kvac
›ek et Her-
man, comb. nov. Fig. 5;Plate VI;Plate VII, 1^4;
Plate VIII, 1^4.
Basionym:Flabellaria longirhachis Unger 1850,
Genera et species plantarum fossilium, p. 332.
Syn.
Flabellaria longirhachis Unger ; Unger 1852, p. 91,
pl. 31, 32, ¢g. 1.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 341

Flabellaria longirhachis Unger; Schimper 1870^
1872, p. 492 (‘longirachis’).
Flabellaria longirhachis Unger; Saporta 1890, p. 3
(‘longirachis’).
Juranyia hemi£abellata Tuzson 1908, p. 1, pl. 1,
¢gs. 1, 2 (non pl. 2, ¢g. 3).
Juranyia hemi£abellata Tuzson ; Tuzson 1913, p.
222, pl. 17, ¢g. 1.
Juranyia hemi£abellata Tuzson ; Tuzson 1914, p.
248, pl. 17, ¢g. 1.
Palmophyllum longirhachis (Unger) Baikovskaya
1965, p. 373, pl. 1, ¢g. 2 (‘longirachis’).
?Sabal imperialis Dawson 1883, p. 26, pl. 6, ¢gs.
23, 23a
?Geonomites imperialis (Dawson) Bell 1957, p. 37,
pl. 22, ¢g. 5, pl. 23, ¢g. 2, pl. 24, ¢g. 3
Lectotype designated here: No. GBA 1852/03/05a ;
¢gured by Unger, 1852 pl. 31; re¢gured here in
Plate VI.
Stratum typicum:Gru
«nbach Formation, Lower
Campanian.
Locus typicus: Muthmannsdorf, Lower Austria,
47‡50P07QN, 16‡06P05QE.
Material: BMNH 844, GBA 1852/03/05a, GBA
1852/03/11, GBA 2001/06/01 and two unnum-
bered specimens, NHMW 591970, NHMW
4191970, NHMW 19781991, NHMW
1999B0057/0051, NHMW 1999B0057/0149,
NHMW 1999B0057/0150, NHMW 1999B0057/
0159, NHMW 1999B0057/0235, NHMW
1999B0057/0370, NHMW 1999B0057/0524,
NHMW 1999B0057/0525, NHMW 1999B0057/
0526, NHMW 1999B0057/0527, NHMW
1999B0057/0528, NHMW 1999B0057/0530,
NHMW 1999B0057/0531, NHMW 1999B0057/
0532, NHMW 1999B0057/0533, NHMW
1999B0057/0534, NHMW 1999B0057/0536, UV
8 unnumbered specimens.
Occurrence: Austria: Muthmannsdorf, Gru
«nbach
am Schneeberg and Meiersdorf; Gru
«nbach For-
mation, Lower Campanian; Romania : Rusca
Montana near Caransebes (Banat); Maastrichtian.
Emended diagnosis: Leaves costapalmate, indupli-
cate, lanceolate, with thick long costa. Leaf
segments fused, distally sometimes shortly dis-
sected. Each segment V-shaped in transversal sec-
tion, with midvein and four parallel vein orders
arranged according to the formula AddCddBdd-
Cdd(BddCdd)A. Veins of lower orders varying in
number. Transversal veins very delicate, usually
hardly visible. Leaf segments nearly linear, slightly
wider towards the apex, emerging at an acute an-
gle from costa; decurrently attached in apical part
of the costa. Both abaxial and adaxial cuticles
consisting of elongate cells, abaxial cuticle show-
ing tetracytic stomata.
Description: The lectotype (Plate VI) is an impres-
sion showing almost the complete leaf. It was
broken in two parts of which the basal part has
not been available for the present study and is of
unknown repository. Its preserved apical part
shows a portion of a leaf 320 mm long (original
leaf length 405 mm) and 170 mm wide. The costa
is 12^14 mm wide (maximum width of the missing
portion was measured from a drawing in Unger,
1852). The leaf segments vary from 5 to 10 mm in
width. The second type specimen (No. 1852/03/11)
shows a fragment of a leaf base with an 18 mm
wide costa (Plate VII, 2).
A complete leaf lamina is not preserved in any
of the specimens studied, but numerous fragments
of leaf margins (e.g. NHMW 1999B0057/0149)
indicate that the leaf lamina was undivided lan-
ceolate (Plate VII, 1,4; Plate VIII, 3). The largest
specimen studied (Plate VII, 3), shows the lamina
380 mm long and 200 mm wide. The maximum
width of the lamina, 240 mm, has been recorded
in the specimen NHMW 1999B0057/0150. A leaf
base with the petiole and a leaf apex is not
preserved in any specimen studied. Specimen
NHMW 1999B0057/0149 (Plate VII, 4) shows a
leaf lamina near its base. The lamina was prob-
ably gradually narrowing and consisted of fused
leaf segments. Specimen GBA 2001/06/01 shows
several marginal segments partly free (Plate
VIII, 1) and may represent a fragment close to
the leaf apex.
A long and massive costa is characteristic of
this species. It narrows gradually to the apex
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353342

and is typically 19^3 mm wide. In the apical part
of the leaf it is narrow and adaxially covered by
decurrent bases of leaf segments (Plate VIII, 3).
This specimen shows the adaxial surface of the
lamina. A clear induplicate splitting is possible
to observe in marginal parts of the specimen.
Specimen NHMW 1999B0057/0149 (Plate VII,
4) represents an abaxial part of the lamina show-
ing induplicately split segments from the opposite
side. Each segment is V-shaped in transversal sec-
tion showing four parallel vein subsets and the
midvein. Segment width is typically about 10
Plate VI. Sabalites longirhachis (Unger) J. Kvac
›ek et Herman, comb. nov., lectotype, Muthmannsdorf, GBA 1852/03/05a, U0.5.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 343

PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353344

mm. In some cases the width can vary within one
leaf lamina, e.g. from 6 to 16 mm (NHMW
1999B0057/0530). Some leaf fragments from the
collection of the University of Vienna (without
numbers) have segments at least 19 mm wide.
Segments are typically decurrent by their bases
and adaxially cover the costa (Plate VIII, 3).
The parallel veins of di¡erent orders follow the
formula AddCddBddCdd(BddCdd)A. Veins of
lower orders (C, d) can vary in number (Fig. 5;
Plate VII, 3), particularly ‘d’ veins, which,
although typically two in number, can vary
from one to three. The width of veins of di¡er-
ent orders is as follows: A : 0.4^0.2 mm, B:
about 0.1 mm, C: about 0.08 mm, d: about
0.02 mm. Spacing of the parallel veins varies
from 0.2 to 0.02 mm. Transverse veins are di⁄-
cult to observe, since they are irregularly spaced,
extremely ¢ne, about 0.01 mm wide, and per-
pendicularly or obliquely oriented to the parallel
venation.
The cuticle is very thin and di⁄cult to prepare.
Cuticle preparations were extracted from the
specimen NHMW 4191970 from Meiersdorf
(Plate VIII, 1). The cell structure shows elongated
ordinary cells 8^10U30^50 Wm with straight anti-
clinal walls and circular tetracytic stomata (Plate
VII, 2) ^ subsidiary cells 7^17U5^7 Wm ; guard
cells 8^10U2^5 Wm.
Discussion: The specimen described as Sabalites
longirhachis from the Campanian^Maastrichtian
of France, locality of Fuveau (Saporta, 1890),
was not available for the present study, but shows
the same gross-morphology including costapal-
mate leaf (Saporta, 1890, ¢g. 1) as the lectotype
and is considered as correctly identi¢ed.
The closest or even conspeci¢c fossil leaf to
Sabalites longirhachis is Juranyia hemi£abellata
Tuzson (1908, pl. 1, ¢gs. 1, 2) from the Maas-
trichtian (Popa, pers. commun., 2001) of Rusca
Montana, Transylvania in Romania. Considering
that we are dealing with a palm leaf morphotaxon
of limited systematic value, we suggest keeping
both leaf-types in one species. The type material
of Juranyia hemi£abellata has not been available
for the present study, being currently of unknown
repository (L. Hably, I. Petrescu pers. commun.,
2001). As it is possible to judge from the illustra-
tions by Tuzson (1908, 1913, 1914), it shows the
same leaf morphology as S. longirhachis having a
long thick costa and undivided lamina.
Geonomites imperialis (Dawson) Bell from the
Campanian of the Nanaimo Group in British Co-
lumbia, Canada (Bell, 1957, pl. 22, ¢g. 5, pl. 23,
¢g. 2, pl. 24, ¢g. 3) is very similar or even identical
to Sabalites longirhachis. It shows a similarly long
costa with the undivided lamina, but slightly dif-
fers from S. longirhachis in subtle characters such
as long decurrent bases of leaf segments. Since the
holotype and additional material of G. imperialis
has not carefully been revised, we question the
synonymy (?) of G. imperialis to S. longirhachis.
In order to conduct a further revision of Geono-
mites imperialis, it is particularly important to re-
vise the venation and number of vein orders, as
well as the cuticle.
The oldest unequivocal record of a palm leaf is
Sabalites carolinensis Berry (1914, pl. 5, 6) from
the Santonian of South Carolina, USA. It di¡ers
from S. longirhachis in having a shorter costa and
narrow wedge-shaped segments, which ultimately
split. All other Cretaceous megafossil records of
palms have not been revised and many of them
are doubtful including the ¢rst European records
(Go
«ppert, 1842; Crie
¤, 1892; De Lima, 1900). Par-
ticularly interesting are Palmoxylon andegavense
Crie
¤,P. guillieri Crie
¤,P. ligerianum Crie
¤(Crie
¤,
1892) from the Turonian of France (Berry,
1911). After stratigraphical and anatomical revi-
Plate VII. Sabalites longirhachis (Unger) J. Kvac
›ek et Herman, comb. nov.
1. Leaf margin, Gru
«nbach, GBA 2001/06/01, U0.5.
2. Abaxial cuticle showing stomata, Meiersdorf, NHMW 4191970b, U250.
3. Detail of venation, Gru
«nbach, NHMW 1999B0057/0528, U3.
4. Basal part of leaf, Gru
«nbach, NHMW 1999B0057/0149, U0.25.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 345

PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353346

sions they could appear as the oldest palm re-
mains.
The palm leaves described by Berry (1905) from
the Santonian of the Maghoty Formation as Fla-
bellaria maghotiensis, and later re-described as Sa-
balites maghotiensis (Berry) Berry are poorly pre-
served and equivocal. They do not show a clear
costa or segments of lamina. The foliage was
combined by Berry (1911) with fossil wood of
Palmoxylon cli¡woodensis Berry (Berry, 1916).
Palm seed of the recent genus Borassus from the
Upper Santonian of Senegal (Monteillet and Lap-
partient, 1981) is considered to be another ques-
tionable palm remain (Uhl and Dran¢eld, 1987).
Similarly, palm seeds from the Asuwa Flora in
Japan (Matsuo, 1962) of Late Turonian (Matsuo,
1970) or younger (Tanai, 1979) age, also need a
revision.
Among the other species of the genus Sabalites,
the present taxon is most similar to Sabalites te-
nuirachis (Lesquereux) Read et Hickey from the
Palaeocene of the Raton Formation in New Mex-
ico (Lesquereux, 1878, pl. 11, ¢g. 1). Besides the
stratigraphic and geographic di¡erence ‘the rays
seem to become free or cut from each other to-
wards their points’ (Lesquereux, 1878, p. 117). In
this character S. tenuirachis di¡ers from most
specimens of S. longirhachis that we studied. Sa-
balites ungeri (Lesquereux) Dorf (Lesquereux,
1878, pl. 11, ¢g. 2) from the Palaeocene of the
Raton Formation, New Mexico, di¡ers from Sa-
balites longirhachis in having a robust striated cos-
ta. All other palm leaves of the genus Sabalites,as
far as we know, have a very short costa in com-
parison with the material in hand. We have
mainly compared leaves of S. longirhachis to the
species accepted by Read and Hickey (1972):Sa-
balites rugosa (Knowlton) Read and Hickey
(Knowlton, 1917, pl. 58) from the Palaeocene of
Raton Formation, New Mexico, S. powellii (New-
berry) Berry (Newberry, 1898 pl. 63, ¢g. 6, pl. 64,
¢g. 1) from the Eocene of the Green River For-
mation, Wyoming, S. leei (Knowlton) Read and
Hickey (Knowlton, 1917, pl. 60) from the Palaeo-
cene of the Raton Formation, Colorado, S. in-
querenda (Knowlton) Read and Hickey (Knowl-
ton, 1917, pl. 56) from the Palaeocene of the
Raton Formation, New Mexico, S. grandifolia
(Newberry) Read and Hickey (Newberry, 1898,
pl. 25, pl. 63, ¢g. 5, pl. 64, ¢g. 2) from the Eocene
of Yellowstone, Montana, S. apalachicolensis
Berry (1917, pls 8, 9) from the Miocene of
Alum Blu¡, Florida, S. grayanus Lesquereux
(1878, pl. 12, ¢gs. 1, 2) and from the Eocene of
Montana (Knowlton, 1900, pl. 6, ¢g. 5). Some
Tertiary species of costapalmate leaves were as-
signed directly to the recent genus Sabal. One of
them, S. dortchii Daghlian from the Eocene of
Tennessee (Daghlian, 1978, pl. 3, ¢gs. 11, 12, pl.
4, ¢gs. 13, 14, pl. 5, ¢gs. 15^17, 20, 21, text-¢g.
12), shows remarkable similarity to the genus Sa-
balites. It shows a long costa, but Daghlian (1978)
argued for its assignment to Sabal on basis of
similarities of cuticle structure. Most other fossil
representatives of the genus Sabal di¡er from Sa-
balites longirhachis in having deeply dissected leaf
lamina: e.g. Sabal lamanonis (Brongniart) Heer
(Brongniart, 1822, pl. 3, ¢g. 1; Knobloch et al.,
1996, pl. 47, ¢gs. 1, 2, 5, pl. 48, ¢g. 1), S. raphi-
folia (Sternberg) Knobloch et Z. Kvac
›ek (von
Sternberg, 1821, pl. 21; Knobloch et al., 1996,
pl. 45, ¢g. 4, pl. 46, ¢gs. 2, 3, pl. 47, ¢g. 4) both
from the Eocene of the Czech Republic, S. vo-
ronkovii from the Eocene of western Kamchatka,
Russia (Budantsev, 1979, 1983, pl. 63, ¢gs. 1^3),
S. ptuchjanii Akhmetiev in Akhmetiev and Zapo-
rozhets (1989;Akhmetiev, 1993, pl. 24, ¢gs. 1, 5)
from the Oligocene of the Caucasus region.
Remarks: Together with Sabalites longirhachis,
Plate VIII. Sabalites longirhachis (Unger) J. Kvac
›ek et Herman, comb. nov.
1. Leaf compression showing costa, Meiersdorf, NHMW 4191970, U0.5.
2. Basal part of leaf (Unger, 1852, pl. 32, ¢g. 1), Muthmannsdorf, GBA 1852/03/11, U0.5.
3. Apical part of leaf, Gru
«nbach, NHMW 1999B0057/0235, U0.3.
4. Basal part of leaf with wide costa, Gru
«nbach, NHMW 1999B0057/533, U0.75.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 347

pollen of monocotyledon a⁄nity probably be-
longing to palms Arecipites sp. has been reported
from Gru
«nbach (Draxler in Summesberger, 1997).
Since there is only one species of palm foliage in
this locality we can assume that the fossil leaves
and pollen belong to the same plant.
Sabalites longirhachis provides evidence that
palms were established in Europe in the Late Cre-
taceous. In general it corroborates the concept of
Asmussen et al. (2000), who resolve the family
Coryfoideae contating palms with costapalmate
leaves as one of the basal members of the Areca-
ceae clade. The costapalmate foliage occurs in
tribes Corypheae and Borasseae of the subfamily
Coryphoideae (Drans¢eld and Uhl, 1998). It is
interesting that this subfamily also includes palms
with simple leaves, e.g. Johannesteismannia H.E.
Moore and Licuala Thunberg, although they are
considered to be derived within the subfamily
(Uhl and Dran¢eld, 1987). Induplicate splitting
of segments is another character which S. longi-
rhachis shares with the subfamily Coryophoideae,
although this character is typical also of the basal
tribe Caryoteae of the Arecoideae (Uhl and Dran-
¢eld, 1987). The leaves of S. longirhachis have a
unique laminar shape, which is costapalmate, in-
duplicate, simple and, only in the apical part,
shortly dissected. It represents a basic form,
from which both pinnate and palmate leaves
may have evolved.
Monocotyledon gen. et sp. indet. Plate IX, 1,2
Locality:Gru
«nbach am Schneeberg, Segen Gottes
Quarry (47‡47P52QN, 15‡59P17QE).
Material: NHMW 1999B0057/0537.
The only specimen available shows several leaf
fragments or fragments of one large entire-mar-
gined leaf without apex and base. The largest
fragment (250U95 mm) with enrolled margins
and without any costa (Plate IX, 1) shows parallel
venation with four vein orders arranged according
to the formula AdCdBdCdA (Plate IX, 2). Trans-
versal veins are conspicuously thick (Plate IX, 2),
of approximately the same thickness as the vein
subset B. They connect at right angles and regular
distances with veins of order A, crossing veins B,
C and d. The distance between the transversal
veins is 2^7 mm; veins of order A are approxi-
mately 2 mm apart. Thickness of veins : A : 0.3^
0.35 mm; B : 0.1 mm; C: 0.07 mm ; d: 0.05 mm.
Discussion: This monocotyledon leaf is similar to
the specimen described by Dorf (1942) from the
uppermost Cretaceous of the Rocky Mountains
(USA) as Typha sp. The Dorf specimen shows
similar venation pattern with three or four vein
subsets. It is also similar in the arrangement of
transversal veins connecting with veins of higher
subsets (A or B). A similar specimen was depicted
by Hickey (1977, pl. 11, ¢gs. 2, 3) from the Early
Eocene of North Dakota, USA. Typha sp. di¡ers
from the specimen from Gru
«nbach in having thin-
ner transversal veins with irregular distribution.
Our fossil monocotyledon leaf also bears a sim-
ilarity to Zingiberopsis riggauensis Knobloch
(1979) which is synonimised by Ru
«¥e and Trost-
heide (2000) with cf. Pontederia montensis Fritel
(1910) from the Santonian of Germany. It di¡ers
from the Gru
«nbach monocotyledon in having a
higher number of vein subsets (5) and thinner
transversal veins.
5. Conclusions
The Early Campanian Gru
«nbach £ora includes
four representatives of monocotyledons, which
probably belong to three di¡erent families. It con-
tains the oldest unequivocal megafossil record of
the Pandanaceae family. Representatives of the
Araceae and Arecaceae are amongst the oldest
in Europe. Interestingly, Pandanites trinervis is
the most common plant fossil in the Gru
«nbach
coal measures and it seems likely that it formed
nearly pure stands in the Gru
«nbach wetlands.
This fact, together with the comparatively high
variability of monocotyledons in the Gru
«nbach
coal measures shows that the monocotyledons
played an important role in the Cretaceous coal-
forming wetlands and supports the suggestion of
wetlands as a place of monocotyledons origin
(Les and Schneider, 1995). Their high specialisa-
tion to freshwater swampy habitats explains their
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353348

rare occurrences in river and salt marsh sediments
of most Cretaceous localities in the Northern
Hemisphere.
It is interesting to note that recent habitats sim-
ilar to those reconstructed for the Gru
«nbach For-
mation occur nowadays in the peat-swamp forests
of Sumatra. Although more tropical, they show a
similar £oral composition, particularly including
palms and pandans. A palm Pholidocarpus suma-
tranus of the subfamily Coryphoideae grows there
together with Pandanus helicopus (Uhl and Dran-
¢eld, 1987, Pl. 34D).
Acknowledgements
We are very grateful to Dr. H. Kollmann
(NHMW) for the logistical support which made
this research possible. Our warm thanks go to Dr.
H. Summesberger (NHMW) who discussed with
us the Cretaceous geology and stratigraphy of the
Neue Welt Basin, provided publications and or-
ganised a ¢eld excursion to the vicinity of Gru
«n-
bach. We thank Dr. J. Eder-Kovar (previously
NHMW, now Natural History Museum, Stutt-
gart) for her help with museum and laboratory
Plate IX. Monocotyledon leaf gen. et sp. indet., Gru
«nbach, NHMW 1999B0057/0537
1. Complete fragment of leaf, U1.
2. Detail of venation, U4.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 349

work and kind hospitality, Dr. R. Zetter from the
University of Vienna, Dr. F. Stojaspal from the
Austrian Geological Survey, Dr. Lila Hably from
the Natural History Museum, Budapest, Dr. Ius-
tinian Petrescu from the University of Cluj Napo-
ca for their help with the collections. We also
thank professor Z. Kvac
›ek for a valuable discus-
sion on fossil plant taxonomy and providing us
with his unpublished data on the Araceae. This
project was supported by the Oskar and Friede-
rike Ermann Fund for Earth Sciences at the
NHMW. The manuscript bene¢ted from reviews
by Professors Else Marie Friis and Steven Man-
chester.
References
Akhmetiev, M.A., 1993. Fitostratigra¢ya kontinental’nyh ot-
lozhenii paleogena i miotsena Vnetropicheskoi Azii (Phyto-
stratigraphy of Paleogene and Miocene continental deposits
of Boreal Asia). Geol. Inst., Russ. Acad. Sci., Trans. 475,
Nauka, Moscow, 143 pp.
Akhmetiev, M.A., Zaporozhets, N.I., 1989. New ¢nds of
palms from the Oligocene of Darrydag (Nakhichevan’
ASSR) (in Russian). Bulleten’ Moskovskogo Obschestva Is-
pytateley Prirody (Bull. MOIP), otd. geol. 64, 57^67.
Ancibor, E., 1979. Systematic anatomy of vegetative organs of
the Hydrocharitaceae. Bot. J. Linn. Soc. 78, 237^266.
Andrews, H.N., 1955. Index of generic names of fossil plants,
1820^1965. Geol. Surv. Bull. 1300, 1^354.
Asmussen, C.B., Baker, W.J., Drans¢eld, J., 2000. Phylogeny
of the palm family (Arecaceae) based on rps16 intron
and trnL-trnF plastid DNA sequences. In: Wilson, K.,
Morrison, D. (Eds.), Proceedings of the Second Inter-
national Conference on the Comparative Biology of the
Monocotyledons, CSIRO, Sydney, Australia, pp. 525^
535.
Baikovskaya, T.N., 1965. O verchnemelovych rasteniach Tran-
silvanii (Rumznia) (On the Upper Cretaceous plants of
Transsylvania (Romania)). Bot. Zhurnal 50, 371^374.
Bell, W.A., 1957. Flora of the Upper Cretaceous Nanaimo
Group of Vancouver Island, British Columbia. Mem.
Geol. Surv. Can. 293, 1^84.
Berry, E.W., 1905. A palm leaf from the mid-Cretaceous. Tor-
reya 5, 30^33.
Berry, E.W., 1911. The Upper Cretaceous. Maryland Geol.
Surv., The Johns Hopkins Press, Baltimore, 1022 pp.
Berry, E.W., 1914. The Upper Cretaceous and Eocene £oras of
South Carolina and Georgia. US Geol. Surv. Prof. Paper 84,
5^200.
Berry, E.W., 1916. A petri¢ed palm from the Cretaceous of
New Jersey. Am. J. Sci. (IV) 41, 193^197.
Berry, E.W., 1917. The physical conditions and age indicated
by the £ora of the Alum Blu¡ Formation. In : White, D.
(Ed.), Shorter Contributions to General Geology, 1916. US
Geol. Surv. Prof. Pap. 98, pp. 41^53.
Berry, E.W., 1924. The Middle and Upper Eocene £oras of
southeastern North America. US Geol. Surv. Prof. Pap. 92,
1^206.
Berry, E.W., 1925. Miocene Araceae related to Caladium from
Trinidad. Pan-Am. Geol. 44, 38^42.
Boyd, A., 1992. Musopsis n. gen. a banana-like leaf genus from
the Early Tertiary of eastren north Greenland. Am. J. Bot.
79, 1359^1367.
Brongniart, A., 1822. Sur la classi¢cation et la distribution des
ve
¤ge
¤taux fossil en general, et seux des terrains de sediment
supe
¤rieur en particulier. Me
¤m. mus. nat. hist. nat. (Paris) 8,
203^240.
Budantsev, L.Yu., 1979. Nahodka pal’my Sabal v eocenovykh
otlozheniyakh Severo-Zapadnoi Kamchatki (Palm Sabal
¢nds in Eocene of northwestern Kamchatka). Bot. Zhurnal
64, 1777^1784.
Budantsev, L.Yu., 1983. Istoriya arkticheskoy £ory epohi ran-
nego kaino¢ta (History of the Early Cenophytic arctic £ora).
Nauka Publ. House, Leningrad, 156 pp.
Bunzel, E., 1871^1873. Die Reptilfauna der Gosauformation in
der neuen Welt bei Wiener-Neustadt. Abh. k-k. geol.
Reichsanst. 5, 1^20.
Bureau, E.M., 1888. E
Łtudes sur la £ore fossile du calcaire
grossier parisien. Me
¤m. publ. Soc. Philomath. (Paris) occas.
centen. found., Gauthier-Villars, Paris, pp. 235^264.
Callmander, M.W., 2001. Pandanus subg. Martellidendron
(Pandanaceae), part II. Revision of sect. Martellidendron
Pic. Serm. in Madagascar. Bot. J. Linn. Soc. 137, 353^374.
Chase, M.W., Soltis, D.E., Soltis, P.S., Raudall, P., Fay, M.F.,
Hahn, W.H., Sullivan, S., Je¡rey, J., Molvray, M., Kores,
P.J., Givnish, T.J., Sytsma, K.J., Pires, J.C., 2000. Higher-
level systematics of the monocotyledons: An assessment of
current knowledge and a new classi¢cation. In: Wilson, K.,
Morrison, D. (Eds.), Proceedings of the Second Internation-
al Conference on the Comparative Biology of the Monoco-
tyledons, CSIRO, Sydney, Australia, pp. 3^16.
Christensen, W.K., 1998. Upper Campanian Belemnitella from
Austria. Beitr. Pala
«ontol. 22, 13^21.
Conwentz, H., 1886. Die Flora des Bernsteins, Vol. 2. Die
Angiospermen des Bernsteins. Wilhelm Engelmann, Danzig,
140 pp.
Cox, P.A., Huynh, K.-L., Stone, B.C., 1995. Evolution and
systematics of Pandanaceae. In: Rudall, P.J., Cribb, P.J.,
Cutler, D.F., Humphries, C.J. (Eds.), Monocotyledons: Sys-
tematics and Evolution. Royal Botanical Gardens, Kew, pp.
663^684.
Crie
¤, L., 1892. Recherches sur les Palmiers silici¢e
¤s des terrains
Cre
¤tace
¤s de l’Anjou. Bull. Soc. Etudes Sci. Angers 21, 97^
103.
Daghlian, C.P., 1981. A review of the fossil record of mono-
cotyledons. Bot. Rev. 47, 517^555.
Daghlian, C.P., 1978. Coryphoid palms from the lower and
middle Eocene of southeastern North America. Palaeonto-
graphica B 166, 44^82.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353350

Dahlgren, R.M., Cli¡ord, H.T., Yeo, P.F., 1985. The Families
of the Monocotyledons. Springer, Berlin.
Dawson, J.W., 1883. On the Cretaceous and Tertiary £oras of
British Columbia and the Northwest Territory. Trans. R.
Soc. Can., Rept. Prog. 1882^1883, 1, 4, pp. 15^34.
De Lima, W., 1900. Noticia sobre algum vegetaes fosseis da
£ora sononiana (sensu lato) do solo. Port. commun. Dir.
serv. geol. Port. 4, 1^12.
de Visiani, R., 1864. Palmae pinnatae tertiariae agri veneti.
Mem. R. Ist. Venet. Sci. 11, 435^460.
Dorf, E., 1942. Upper Cretaceous Floras of the Rocky Moun-
tain Region, I. Stratigraphy and Palaeontology of the Fox
Hills and Lower Medicine Bow Formations of Southern
Wyoming and Northwestern Colorado. Carnegie Inst.
Washington Publ. 508, Washington, DC, pp. 1^78.
Drans¢eld, J., Uhl, N.W., 1998. Palmae. In: Kubitzki, K.
(Ed.), Families and genera of vascular plants, vol. 4, Flower-
ing plants: Monocotyledons. Springer, Berlin, pp. 306-389.
Fritel, P.H., 1910. Etude sur les ve
¤ge
¤taux fossiles de l’e
¤tage
Sparnacien du basin de paris. Me
¤m. Soc. Geol. Fr., Pale
¤on-
tol. 16, 1^37.
Gandolfo, M.A., Nixon, K.C., Crepet, W.L., 2000. Monoco-
tyledons: A review of their Early Cretaceous record. In :
Wilson, K., Morrison, D. (Eds.), Proceedings of the Second
International Conference on the Comparative Biology of the
Monocotyledons, CSIRO, Sydney, Australia, pp. 44^52.
Gandolfo, M.A., Nixon, K.C., Crepet, W.L., 2002. Triurida-
ceae fossil £owers from the Upper Cretaceous of New Jer-
sey. Am. J. Bot. 89, 1940^1957.
Go
«ppert, H.R., 1842. Ueber die fossile Flora des Quadersand-
satins von Schlesien. Nova Acta Leopoldina 19, 97^134.
Herendeen, P.S., Crane, P.R., 1995. The fossil history of the
monocotyledons. In: Rudall, P.J., Cribb, P.J., Cutler, D.F.,
Humphries, C.J. (Eds.), Monocotyledons: systematics and
evolution. Royal Botanical Gardens, Kew, pp. 1^21.
Herman, A.B., Kvac
›ek, J., 2000. Campanian Gru
«nbach £ora
of Austria: A preliminary report. 6th International Creta-
ceous Symposium, August 27^September 4, 2000, Vienna,
Abstr., p. 46.
Herman, A.B., Kvac
›ek, J., 2002a. Campanian Gru
«nbach £ora
of Lower Austria: Preliminary £oristics and palaeoclimatol-
ogy. Ann. Naturhist. Mus. Wien 103, 1^21.
Herman, A.B., Kvac
›ek, J., 2002b. Campanian Gru
«nbach £ora
of Lower Austria, its composition and phytogeographic sig-
ni¢cance. In: Akhmetiev, M.A., Herman, A.B., Doludenko,
M.P., Ignatiev, I.A. (Eds.), Spcial volume dedicated to the
Memory of the Corresponding Member of the USSR Acad-
emy of Sciences, Professor Vsevolod Andreevich Vakhra-
meev (to the 90th Anniversary of his Birth), GEOS, Mos-
cow, pp. 269^275.
Hickey, L.J., 1977. Stratigraphy and palaeobotany of the
Golden Valley Formation (Early Tertiary) of western North
Dakota. Geol. Soc. Am. Mem. 150, 1^181.
Hickey, L.J., Peterson, R.K., 1978. Zingiberopsis, a fossil ge-
nus of the ginger family from Late Cretaceous to early Eo-
cene sediments of Western Interior North America. Can. J.
Bot. 56, 1136^1152.
Hotton, C.L., Le⁄ngwell, H.A., Skvarla, J.J., 1994. Pollen
ultrastructure of Pandanaceae and the fossil genus Pandani-
dites. In: Kurmann, M.H., Doyle, J.A. (Eds.), Ultrastruc-
ture of Fossil Spores and Pollen. Royal Botanical Gardens,
Kew, pp. 173^191.
Hradecka
¤, L., Lobitzer, H., Svobodova
¤, M., S
va
¤benicka
¤, L.,
2000. Biostratigraphy of selected exposures in the Gru
«n-
bach^Neue Welt Gosau Group (Late Cretaceous). In : 6th
International Cretaceous Symposium, August 27 to Septem-
ber 4, 2000, Vienna, Austria, Abstr., p. 51.
Iljinskaja, I.A., 1963. Iskopaemaya £ora gory Kiin-Kerish
Zaissanskogo basseina. Chast’ II. (Fossil £ora of the Moun-
tain Kiin-Kerish of the Zaisan Basin, Part II). Acta Inst.
Bot. Komarov 8, Paleobot. 4, 141^187.
Kerner-Marilaun, F., 1934. Das Klimazeugnis der Gosaufor-
mation. Sitz.-Ber. Akad. Wiss., math-naturw. Kl., 2a 143,
267^284.
Knobloch, E., 1977. Fossile P£anzenreste aus der Kreide und
dem Tertia
«r von O
ºsterreich. Verh. Geol. B.-A. 3, 415^426.
Knobloch, E., 1979. Zingiberopsis riggauensis sp. nov. ^ eine
interessante Monokotyledone aus der Kreide Bayerns.
Ve
flstn|
¤kU
Łstr
fledn|
¤ho u
¤stavu geol. 54, 297^300.
Knobloch, E., Konzalova
¤, M., Kvac
›ek, Z., 1996. Die obereo-
za
«ne Flora der Stare
¤Sedlo-Schichtenfolge in Bo
«hmen (Mit-
teleuropa). Rozpravy C
GU
Ł49, 1^260.
Knowlton, F.H., 1900. Flora of the Montana Formation. US
Geol. Surv. Bull. 163, 1^118.
Knowlton, F.H., 1917. Fossil £oras of the Vermejo and Raton
Formations of Colorado and New Mexico. In: Lee, W.T.,
Knowlton, F.H., Geology and Paleontology of the Raton
Mesa and other Regions in Colorado and New Mexico.
US Geol. Surv. Prof. Pap. 101, 323^455.
Kollmann, H.A., Summesberger, H., Wagreich, M., 2000.
Field Trip C ‘Cretaceous of eastern Austria’. Sixth Interna-
tional Cretaceous Symposium, Vienna, 56 pp.
Krasser, F., 1906. U
ºber die fossile Kreide£ora von Gru
«nbach
in Niedero
«sterreich. Anz. K. Ak. Wiss., math.-naturw. Cl.
1906, 1^3.
Krassilov, V.A., 1967. Rannemelovaya Flora yuzhnogo pri-
mor’ya i ee znachenie dlya stratigra¢i (Early Cretaceous
Flora of South Primorye and its Stratigraphic Signi¢cance).
Moscow, Nauka, 264 pp.
Kra
«usel, R., 1929. Fossile P£anzen aus dem Tertia
«r von Su
«d-
Sumatra. Beitr. Geolog. Pala
«ontol. Sumatra 11, 335^378.
Kra
«usel, R., Weyland, H., 1950. Kritische Untersuchungen zur
Kutikularanalyse tertia
«rer Bla
«tter I. Palaeontographica B 91,
1^92.
Kryshtofovich, A.N., 1929. Otkrytie drevneishikh dvudol’nykh
pokrytosemennykh i ekvivalentov potomakskich cloev na
Sutchane v Ussuriyskom kraye. Izv. Geol. Kom. 48, 9,
113^145.
Laurent, M.L., 1899. Flore des calcaires de Ce
¤las. The
¤ses pre
¤-
sente
¤es a la Faculte
¤des Sciences de Marseille pour obtenir le
grade de docteur es sciences naturelles, Moullot Fils Aine
¤,
Marseille.
Les, D.H., Schneider, E.L., 1995. The Nymphaeales, Alisma-
tidae, and the theory of an aquatic monocotyledon origin.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 351

In: Rudall, P.J., Cribb, P.J., Cutler, D.F., Humphries, C.J.
(Eds.), Monocotyledons: Systematics and Evolution. Royal
Botanical Gardens, Kew, pp. 23^42.
Lesquereux, L., 1878. Contribution to the fossil £ora of the
Western Territories, Part II. The Tertiary Flora. US Geol.
Surv. Rep. VII, 1^366.
Matsuo, H., 1962. A study on the Asuwa £ora (Late Creta-
ceous age) in the Hokuriku District, Central Japan. Sci.
Rep. Kanazawa Univ. 8, 177^250.
Matsuo, H., 1970. On the Omichidani £ora (Upper Creta-
ceous) inner side of Central Japan. Trans. Proc. Palaeontol.
Soc. Jpn. N.S. 80, 369^371.
Mayo, S.J., Bogner, J., Boyce, P.C., 1997. The genera of Ara-
ceae. Royal Botanic Gardens, Kew, 370 pp.
Monteillet, J., Lappartient, J.R., 1981. Fruit et graines du
Cre
¤taceae superieur des carrieres de Paki (Senegal). Rev.
Palaeobot. Palynol. 34, 331^344.
Newberry, J.S., 1898. The later extinct £oras of North Amer-
ica. US Geol. Surv. Monogr. 35, 1^151.
Palamarev, E., Usunova, K., 1969. Monokotylen aus dem
Plioza
«nen Braunkohlen Su
«dbulgariens. Mitt. Bot. Inst. 19,
127^135.
Peters, I., 1963. Die Flora der Oberpfa
«lzer Braunkohlen und
ihre o
«kologische und stratigraphische Bedeutung. Palaeon-
tographica B 112, 1^50.
Petrescu, I., Dusa, A., 1980. Flora din cretacicul superior de la
Rusca Montana ^ o raritate in patrimoniul paleobotanic
national (Upper Cretaceous £ora of the Rusca Montana ^
a rarity of the national paleobotanical patrimony). Ocrot.
nat. med. inconj. 24, 147^155.
Ploechinger, B., 1961. Die Gosaumulde von Gru
«nbach und der
Neuen Welt (Niedero
«sterreich). Jb. Geol. B. A. 104, 359^
441.
Read, R.W., Hickey, L.J., 1972. A revised classi¢cation of
fossil palm and palm-like leaves. Taxon 21, 129^137.
Ru
«¥e, L., Trostheide, F., 2000. Erga
«nzungen zur Oberkreide-
Flora von Quedlinburg (Deutschland) und einiger weiterer
Kreide-Fundsta
«tten (geographische Beziehungen). Feddes
Repert. 7^8, 433^444.
Saporta, G., 1865. E
Łtudes sur la vegetation du sud-est de la
France a l e
¤poque tertiaire. Ann. Sci. nat. botan. 5 3, 1^152.
Saporta, G., 1890. Le Nelumbium provinciale des lignites cre
¤-
tace
¤s de Fuveau en Provence. Me
¤m. Soc. Ge
¤ol. Fr., Pale
¤on-
tol. 5, 1^10.
Savolainen, V., Chase, M.W., Hoot, S.B., Morton, C.M., Sol-
tis, D.E., Bayer, C., Fay, M.F., De Brujin, A., Sullivan, S.,
Qiu, Y.-L., 2000. Phylogenetics of £owering plants based
upon a combined analysis of plastid atpB and rbcL gene
sequences. Syst. Biol. 49, 306^362.
Schimper, W.P., 1870^1872. Traite
¤de pale
¤ontologie ve
¤ge
¤tale
ou la £ore du monde primitif dans ses rapports avec les
formations ge
¤ologiques et la £ore du monde actuel. II T.
J.B. Baillie
¤re et ¢ls, Paris, 968 pp.
Soltis, D.E., Soltis, P.S., Chase, M.W., Mort, M.E., Albach,
D.C., Zanis, M., Savolainen, V., Hahn, W.H., Hoot, S.B.,
Fay, M.F., Axtell, M., Swensen, S.M., Prince, L.M., Kress,
J., Nixon, K.C., Farris, J.S., 2000. Angiosperm phylogeny
inferred from 18S rDNA, rbcL, and atpB sequences. Bot. J.
Linn. Soc. 133, 381^461.
Squinabol, S., 1891. Note sur quelques types de monocotyle-
donyle
¤done
¤es de Saint-Justine et de Sassello. Soc. ge
¤ol. Fr.
Bull. 19, 771^782.
Stevenson, D.W., Laconte, H., 1995. Cladistics analysis of
monocot families. In: Rudall, P.J., Cribb, P.J., Cutler,
D.F., Humphries, C.J. (Eds.), Monocotyledons: Systematics
and Evolution. Royal Botanical Gardens, Kew, pp. 543^578.
Steudel, E.G., 1855. Synopsis Plantarum Glumacearum, vol. 2.
J.B. Metzler, Stuttgart, 348 pp.
Stone, B.C., Huynh, K.-L., Poppendieck, H.-H., 1998. Panda-
naceae. In: Kubitzki, K. (Ed.), The Families and Genera of
Vascular Plants, vol. III, Flowering Plants, Monocotyledons
Lilianae (except Orchidaceae), Springer, Berlin, pp. 397^404.
Summesberger, H., 1997. The Cretaceous of the Gru
«nbach^
Neue Welt Basin. In: Kollmann, H.A., Hubmann, B.
(Eds.), Climates: Past, Present and Future. Excursion
Guide, Second European Palaeontological Congress, Vien-
na, 1997, pp. 77^89.
Summesberger, H., Wagreich, M., Tro
«ger, K.-A., Scholger, R.,
2000. Piesting-Formation, Gru
«nbach-Formation und
Maiersdorf-Formation ^ drei neue lithostratigraphische Ter-
mini in der Gosau Gruppe (Oberkreide) von Gru
«nbach und
der Neue Welt (Niedero
«sterreich). Ber. Inst. Geol. Pala
«on-
tol., K.-F.-Univ. Graz 2, 23.
Summesberger, H., Wagreich, M., Tro
«ger, K.-A., Scholger, R.,
2002. The Upper Cretaceous of Piesting (Austria): Inte-
grated stratigraphy of the Piesting Formation (Gosau
Group). In: Wagreich, M. (Ed.), Aspects of Cretaceous
Stratigraphy and Palaeobiogeography. Proceedings Sixth In-
ternational Cretaceous Symposium, Vienna. 2000, O
ºsterrei-
chische Akademie der Wissenschaften, Schriften Reihe der
Erdwissenschaftlichen Kommisionen 15, pp. 373^400.
Tanai, T., 1979. Late Cretaceous £oras from the Kuji District,
Northeastern Honshu, Japan. J. Fac. Hokkaido Univ. IV
19, 75^136.
Tomlinson, P.B., 1965. A study of stomatal structure in Pan-
danaceae. Pacif. Sci. 19, 38^54.
Tomlinson, P.B., 1974. Development of the stomatal complex
as a taxanomic character in the monocotyledons. Taxon 23,
109^128.
Tro
«ger, K.-A., Summesberger, H., Wagreich, M., Scholger, R.,
2000. Early Maastrichtian inoceramids of the Gosau Group
(Austria). In: Abstr. Sixth International Cretaceous Sympo-
sium, August 27^September 4, 2000, Vienna, p. 138.
Tro
«ger, K.-A., Summesberger, H., Wagreich, M., 2001. Early
Maastrichtian (Late Cretaceous) Inoceramids from the Pist-
ing Formation (Gosau Group, Austria). Beitr. Pala
«ontol. 26,
145^167.
Tuzson, J., 1908. Adatok Magyarorsza
¤g fosszilis £o
¤ra
¤ja
¤ho (II).
Additamenta ad £oram fossilem Hungariae. No
«ve
¤nytani
Ko
«zleme
¤nyek 7, 1^4.
Tuzson, J., 1913. Adatok Magyarorsza
¤g fosszilis £o
¤ra
¤ja
¤hoz
Magyarorza
¤g fosszilis £o
¤ra
¤ja
¤hoz (Additamenta ad £oram
fossilem Hungariae III). A magyar kira
¤lyi fo
«ldtani inte
¤zet
e
¤vko
«nyve 21, 209^233.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353352

Tuzson, J., 1914. Beitra
«ge zur fossilen Flora Ungarns (Addi-
tamenta ad £oram fossilem Hungariae III). Mitt. Jahrb. k.
ung. geol. Reichsanst. 21, 233^261.
Uhl, N.W., Dran¢eld, J., 1987. Genera Palmarum, A Classi-
¢cation of Palms Based on the Work of Harold E. Moore,
Jr. The L.H. Bailey Hortorium and The International Palm
Society, Allen Press, Lawrence, Kansas, 610 pp.
Unger, F., 1850. Genera et species plantarum fossilium. Aca-
demia Caesareae Scientiarum, Wilhelm Braumueller, Vindo-
bonae, 627 pp.
Unger, F., 1852. Iconographia plantarum fossilium. Abbildun-
gen Beschreib. foss. P£anzen. Denkschr. Akad. Wiss. 4, 73^
118.
Upchurch, G.R., Mack, G.H., 1998. Latest Cretaceous
leaf mega£oras from the Jose Creek Member, McRare
Formation of New Mexico. New Mexico Geol. Soc. Guide,
49th Field Conference, Las Cruces Country, pp. 209^
222.
von Ettingshausen, C., 1852. U
ºber fossile Pandaneen. Sit-
zungsber. math.-naturwiss. Klasse der kais. Akad. Wiss. 8,
489^495.
von Sternberg, K.M., 1821. Versuch einer geognostisch-bota-
nischen Darstellung der Flora der Vorwelt, Vol. 1, 2. F.
Fleischer, Leipzig, 33 pp.
Walker, J.W., Walker, A.G., 1987. Ultrastructure of Lower
Cretaceous angiosperm pollen and the origin and early evo-
lution of £owering plants. Ann. Mo. Bot. Gard. 71, 464^
521.
Weyland, H., 1957. Kritische Untersuchungen zur Kutikular-
analyse tertia
«rer Bla
«tter III. Palaeontographica B 103, 34^
74.
Wilde. V., Kvac
›ek, Z., Bogner, J., 2003. Fossil leaves of the
Araceae from the Eocene of Europe and notes to other
fossils of this family. Int. J. Plant. Sci.
Zittel, K.A., 1890. Handbuch der Palaeontologie. II Abt. Pa-
laeophytologie. R. Oldenbourg, Mu
«nchen/Leipzig, 958 pp.
PALBO 2577 5-1-04
J. Kvac
›ek, A.B. Herman / Review of Palaeobotany and Palynology 128 (2004) 323^353 353