Content uploaded by Maria von Balthazar
Author content
All content in this area was uploaded by Maria von Balthazar on Nov 28, 2016
Content may be subject to copyright.
POTOMACANTHUS LOBATUS GEN.ET SP.NOV., A NEW FLOWER OF
PROBABLE LAURACEAE FROM THE EARLY CRETACEOUS (EARLY
TO MIDDLE ALBIAN)OF EASTERN NORTH AMERICA
1
MARIA VON BALTHAZAR,
2,6
KAJ RAUNSGAARD PEDERSEN,
3
PETER R. CRANE,
4
MARCO STAMPANONI,
5
AND ELSE
MARIE FRIIS
2
2
Department of Palaeobotany, Swedish Museum of Natural History, Box 50007, 10405 Stockholm, Sweden;
3
Department of
Geology, University of Aarhus, 8000 Aarhus, Denmark;
4
Department of the Geophysical Sciences, University of Chicago,
5734 S. Ellis Avenue, Chicago, Illinois 60637 USA; and
5
Swiss Light Source, Paul Scherrer Institut, CH-5232 Villigen,
Switzerland
A charcoalified fossil flower, Potomacanthus lobatus gen. et sp. nov., is described from the Early Cretaceous (Early to Middle
Albian) Puddledock locality, Virginia, USA. Internal floral structure was studied using nondestructive synchrotron-radiation x-ray
tomographic microscopy (SRXTM). The flower is bisexual and trimerous. The perianth consists of two whorls of tepals. The
androecium has two whorls of fertile stamens. Anthers open by two distally hinged valves. The gynoecium consists of a single
carpel that is plicate in the style and ascidiate in the ovary and contains a single pendant ovule. The fossil flower shares many
similarities with flowers of extant Lauraceae and is unlike flowers of other families of Laurales. However, the fossil flower also
differs in detail from all extant or fossil Lauraceae, particularly in configuration of the androecium. The new taxon, together with
previously described but more fragmentary material from the Puddledock locality, provides the earliest fossil record of plants
more closely related to Lauraceae than to any other extant family. It reveals several derived morphological characters that are
potential synapomorphies among extant representatives of the family Lauraceae and contributes to the growing evidence for an
early diversification of Laurales before the end of the Early Cretaceous.
Key words: androecium; Early Cretaceous; floral structure and reconstruction; fossil; Lauraceae; Laurales; Potomac Group;
x-ray tomography.
The angiosperm order Laurales comprises seven families
(Calycanthaceae, Siparunaceae, Gomortegaceae, Atherosper-
mataceae, Hernandiaceae, Monimiaceae, Lauraceae, and Caly-
canthaceae; Renner, 1999), of which Lauraceae are by far the
most species rich (2500–3500 spp.; Rohwer, 1993; Renner,
2004). Phylogenetic analyses based on molecular markers,
combined with inferences about the timing of phylogenetic
events in the group based on branch lengths calibrated from the
fossil record, suggest that the bulk of the species diversity has
accumulated since the Cenozoic for which the family is well
represented in many fossil floras by flowers, fruits, leaves, and
wood (Eklund, 1999; Chanderbali et al., 2001). These analyses
further suggest that the basal lineages of Lauraceae, which
include few extant species, were already established in the Late
Cretaceous (Chanderbali et al., 2001).
The fossil record, in particular from the Cretaceous, plays a
significant role in understanding the early history of angio-
sperms (e.g., Friis et al., 2006). Direct fossil evidence of
Lauraceae in the Cretaceous is sparse compared to that in the
Cenozoic, but there is nevertheless a growing record of leaves,
wood, and reproductive structures that clearly establishes that
the family was both present and widespread at an early stage in
angiosperm evolution. Interestingly, the fossil record of flowers
is more extensive from the Cretaceous than from the Cenozoic.
Especially important are several well-preserved fossil flowers
that have been described over the past two decades. Cretaceous
floral structures assigned to Lauraceae are now known based
on 12 taxa from several localities in North America, Europe,
and Asia (Drinnan et al., 1990; Crane et al., 1994; Herendeen et
al., 1994; Eklund and Kvac
ˇek, 1998; Eklund, 2000; Takahashi
et al., 2001; Frumin et al., 2004; A. Viehofen, Bedburg,
Germany; C. Hartkopf-Fro¨der, Geological Survey of North
Rhine-Westphalia, Krefeld, Germany, unpublished manu-
script). These fossils provide insights into the floral structure
of Cretaceous representatives of the family. Because of the
importance of floral structure in the systematics of extant
Lauraceae, the fossil flowers allow us to compare the fossils
with living representatives and draw preliminary conclusions
about their phylogenetic relationships. They also provide new
data for understanding the biogeographic history of the family
as well as its possible phylogenetic and functional differenti-
ation (e.g., floral biology and pollination).
We report here a new Early Cretaceous (Early to Middle
Albian) fossil flower of lauraceous affinity from the Pud-
dledock locality, Virginia, USA. The flower has several typical
lauraceous structural features but is remarkable for its simple
androecium. Nondestructive optical sectioning of the single
specimen using synchrotron-radiation x-ray tomographic
microscopy (SRXTM) provided crucial information on floral
organization and internal structure. This flower, along with
other more fragmentary floral material assigned to Lauraceae
that has already been described from the Puddledock locality
1
Manuscript received 5 June 2007; revision accepted 18 October 2007.
The authors thank P. von Knorring for drawing the reconstructions; S.
Bengtson, P. C. J. Donoghue, T. Huldtgren, and N. J. Gostling for help
with the SRXTM and AMIRA analyses; this work was performed at the
Swiss Light Source, Paul Scherrer Institut, Villigen, Switzerland. Further,
J. Scho¨ nenberger, P. K. Endress, and an anonymous reviewer are thanked
for helpful comments on the manuscript. Financial support was received
from the Swedish Research Council (EMF) and the U.S. National Science
Foundation (PRC). SRXTM was funded by the Swiss Light Source,
European Union FP6, project no. 20060902 (to P. C. J. Donoghue and S.
Bengtson).
6
Author for correspondence (e-mail: maria.vonbalthazar@nrm.se)
2041
American Journal of Botany 94(12): 2041–2053. 2007.
(Crane et al., 1994), provides the earliest evidence of the family
and securely establishes its presence in the Early Cretaceous.
The material further clearly demonstrates the phylogenetic and
functional diversification of Laurales and probable Lauraceae
at a very early phase of angiosperm evolution.
MATERIALS AND METHODS
Geology and age—The material was collected from the Tarmac Lone Star
Industries sand and gravel pit (Puddledock locality), located south of Richmond
and east of the Appomattox River in Prince George County, Virginia, USA (see
Friis et al., 1994, 1995). Based on palynological analyses (Dischinger, 1987),
Fig. 1. Flower and perianth morphology of Potomacanthus lobatus (PP44882). SEM and SRXTM micrographs. (A–C) Lateral views of flower
showing two perianth whorls, two stamen whorls, and the top of the gynoecium. (D) Adaxial surface of outer tepal showing papillate epidermal surface.
(E) Abaxial surface of inner tepal showing long trichomes. (F) Digital transverse section of perianth organ showing bulging epidermal cells. Scale bars in
Fig. 1A, C ¼1 mm; in Fig. 1B, D ¼500 lm; in Fig. 1E, F ¼100 lm. Abbreviations: g, gynoecium; is, inner stamen; it, inner tepal; ot, outer tepal; os, outer
stamen.
2042 AMERICAN JOURNAL OF BOTANY [Vol. 94
the sediments at this locality have been assigned to the basal part of subzone
IIB in the palynological zonation established for the Potomac Group by
Brenner (1963) and others (Doyle, 1969; Doyle and Hickey, 1976; Doyle and
Robbins, 1977; Hickey and Doyle, 1977). Subzone IIB is of Middle Albian age
but may extend down into the Early Albian (Doyle, 1992).
The fossil material comprises a single bisexual flower (PP44882) isolated
from sediment sample 82 collected by P. R. Crane, A. N. Drinnan, E. M. Friis,
and K. R. Pedersen. This is one of several samples collected in 1988 from
organic rich clay exposed in the lower part of the Puddledock pit. Sample 82
was collected in the northern wall of the clay pit from the uppermost part of the
horizon, which at this location was about 10 m above the bottom of the pit. This
part of the outcrop has now been lost through further quarrying.
Fossil preparation—Plant fossils were extracted from bulk sediment
samples by sieving in water followed by treatment with 40% HF and 10% HCl
and thorough rinsing in water. The fossils are typically small, charcoalified or
lignitized, and comprise fragments of wood and twigs as well as detached
reproductive organs. The Puddledock flora is extremely rich and contains a
diverse and abundant assemblage of angiosperm reproductive structures. Most
of the reproductive organs are fruits and seeds, but there are also well-preserved
flowers, often with their three-dimensional form intact (Crane et al., 1994; Friis
et al., 1994, 1995, 1997).
For scanning electron microscopy (SEM), the specimen was mounted on an
aluminium stub, sputter coated with Au for 60 s, and studied using a Hitachi
(Tokyo, Japan) S-4300 field emission scanning electron microscope at 1–5 kV.
Fig. 2. Digital tomographic reconstruction of flower morphology of Potomacanthus lobatus (PP44882). Colors indicate different floral organs: dark
green ¼outer perianth whorl, light green ¼inner perianth whorl, orange ¼outer stamen whorl, yellow ¼inner stamen whorl, blue ¼gynoecium. (A) Lateral
view of flower showing perianth organs in two whorls, stamens in two whorls, and gynoecium. (B) Lateral view of flower showing perianth, androecium,
and top of gynoecium; compare with Fig. 1A. (C) Lateral view of flower showing the outer and inner perianth organs. (D) Same as A, top view. (E) Lateral
view of flower with outer perianth whorl not shown. (F) Lateral view of flower with perianth not shown. Scale bars in Fig. 2A–2F ¼1 mm.
December 2007] VON BALTHAZAR ET AL.—LAURACEAE FLOWER FROM THE EARLY CRETACEOUS 2043
After SEM, the specimen was further investigated using synchrotron-radiation
x-ray tomographic microscopy (SRXTM) at the Tomography Station of the
Materials Science Beamline of the Swiss Light Source, the Paul Scherrer
Institute, Switzerland. For these x-ray studies, the specimen was remounted
without further treatment on a brass stub with a diameter of 3 mm and
examined using the technique outlined by Donoghue et al. (2006) and Friis et
al. (2007). Slice data derived from the scans were then analyzed and
manipulated using AMIRA (Mercury Computer Systems, Merignac Cedex,
France http://www.tgs.com/products/amira.asp) software for computed tomog-
raphy on a 2.5-GHz G5 Power Macintosh (Apple, Cupertino, California, USA)
Character state reconstruction—Androecial characters were mapped onto
a phylogenetic tree based on the results of Bayesian analyses of trnK intron data
by Rohwer and Rudolph (2005). We simplified the tree by reducing the number
of terminals to a single representative for each genus. Character state
reconstructions were implemented using Fitch optimization with state changes
equally weighted in MacClade 4.0 (Maddison and Maddison, 2000). Character
state data were mainly retrieved from Rohwer (1993). Androecial characters
were scored as having the following unordered states: fertile stamens ¼one
whorl, two whorls, three whorls, numerous, or polymorphic; staminodes ¼
absent, present, or polymorphic; staminal appendages ¼absent, present on third
whorl, present on another whorl, or polymorphic; sporangia number ¼
bisporangiate, tetrasporangiate, or polymorphic. Flowers of extant Potomeia
are consistently dimerous, and its four stamens were scored as two whorls.
Repository—The specimen is deposited in the palaeobotanical collections
of the Field Museum of Natural History, Chicago (PP).
RESULTS
Systematics—
Order—Laurales
Family—Lauraceae sensu lato
Genus—Potomacanthus gen. nov.
Fig. 3. Stamen and pollen morphology of Potomacanthus lobatus (PP44882). SEM and SRXTM micrographs. (A) Abaxial view of stamen from inner
whorl. Upper part of anther partly destroyed. (B) Lateral view of stamen from outer whorl showing one opened theca; valve is missing. Asterisk indicates
connective protrusion. (C) Abaxial view of the stamen showing anther epidermis. (D) Abaxial view of lower part of opened theca showing differentiated
endothecium cells. (E) Pollen grain found in situ in stamen from 3A. (F) Close-up of finely reticulate pollen surface bearing prominent clavae. (G–H)
Digital transverse sections of anther showing the two pollen sacs. Scale bars in Fig. 3F ¼1 mm; 3A–C, G, H ¼100 lm; 3D ¼50 lm; 3E ¼10 lm.
2044 AMERICAN JOURNAL OF BOTANY [Vol. 94
Derivation of generic name—From the Potomac Group
where the fossil was collected.
Generic diagnosis—Flower small, bisexual, actinomorphic,
and trimerous. Perianth consisting of six tepals in two
alternating whorls. Androecium of six stamens in two
alternating whorls: the outer whorl opposite the outer tepals.
No staminodes present. Stamens differentiated into anther and
filament but without a distinct joint at the base of the anther.
Anther dehiscence by two valves (interpreted as distally
hinged). Gynoecium of one superior, unilocular carpel
containing a single anatropous, pendant ovule. Ovary ascidiate,
style plicate. Ventral slit of carpel fused; canal present in the
innermost part of the ventral slit.
Type species—Potomacanthus lobatus sp. nov. (Figs. 1–6)
Derivation of species name—From Latin lobus (lobe),
referring to the lobed gynoecium.
Fig. 4. Gynoecium morphology of Potomacanthus lobatus (PP44882). SEM and SRXTM micrographs. (A) Lateral view of gynoecium. Arrow heads
point to length of ventral slit. (B) Close-up of stylar region showing ventral slit. (C) Close-up of stigmatic region with ventral slit (arrowhead). (D) Close-
up of ovary showing large trichomes. (E) Tomographic reconstruction of gynoecium showing ventral slit and sinuous ovary. (F) Transparent tomographic
image of E showing internal ventral canal and ovarial cavity. (G–K) Digital transverse sections of gynoecium. (G) At level of upper style. (H) At level of
lower style. (I) At level of ovule insertion. (J) At level of lower part of ovule. (K) At level below ovarial cavity. Scale bars in Fig. 4E, F¼1 mm; 4A ¼500
lm; 4B–D, I–K ¼100 lm; 4G, H ¼50 lm.
December 2007] VON BALTHAZAR ET AL.—LAURACEAE FLOWER FROM THE EARLY CRETACEOUS 2045
Specific diagnosis—As for the genus with the following
additions: tepals of outer perianth whorl more or less the same
length as those of the inner whorl, but broader and proximally
overlapping the bases of the inner tepals. Long simple trichomes
mainly on the abaxial surfaces of the tepals. Anther dehiscence
by latrorse valves. Style round in cross section, becoming
triangular toward its base. Ovary four-lobed in cross section.
Dimensions—Flower c. 2 mm long, c. 1.5 mm wide; outer
perianth organs c. 1.8 mm long, 0.8–1.2 mm wide; inner
perianth organs c. 1.6 mm long, c. 0.6 mm wide; filament 0.5
Fig. 5. Line drawings of a series of digital transverse sections, beginning at the apex and progressing to the base of the flower, of Potomacanthus
lobatus (PP44882). Colors correspond to those in Fig. 2 and indicate different floral organs: dark green ¼outer perianth whorl, light green ¼inner perianth
whorl, orange ¼outer stamen whorl, yellow ¼inner stamen whorl, blue ¼gynoecium. In Fig. 5I, the directions of the views of Fig. 2A–C are indicated.
Scale bar in Fig. 5A–K ¼500 lm.
2046 AMERICAN JOURNAL OF BOTANY [Vol. 94
mm long; anther 0.9 mm long; carpel 1.6 mm long; style 1.2
mm long; ovary 0.4 mm long.
Holotype—PP44882
Type locality—Puddledock locality, Tarmac Lone Star
Industries sand and gravel pit, located south of Richmond
and east of the Appomattox River in Prince George County,
Virginia, USA.
Type stratum—Basal part of Subzone IIB, Potomac Group.
Age—Early Cretaceous (Early to Middle Albian)
Description—
Flower morphology—The flower is bisexual, c. 2 mm long
and 1.5 mm broad (Fig. 1A–C). The floral organs are organized
in alternating trimerous whorls although the arrangement is
somewhat obscured by slight compression of the specimen
(Figs. 2D, 5). The flower appears to be anthetic or slightly
postanthetic, which is indicated by the presence of opened
anthers.
Perianth—The perianth consists of six tepals in two
alternate whorls; one of the outer tepals and two of the inner
tepals are partly missing (Figs. 1A–C, 2A–C, 5, 6). Tepals of
both whorls are oblong to slightly obovate (Fig. 2A–C), those
of the outer whorl (c. 1.8 mm long, 0.8–1.2 mm wide) are
slightly longer and distinctly broader than the tepals of the
inner whorl (c. 1.6 mm long, c. 0.6 mm wide; Fig. 2C). The
bases of the outer tepals partly overlap the inner tepals (Fig.
1A, 2C). Simple, unicellular trichomes, up to 0.3 mm long,
occur mainly on the abaxial surface of the inner tepals but also
more sparsely on the adaxial surface of both the inner and outer
tepals (Fig. 1E). The adaxial and abaxial surfaces of both the
inner and outer tepals are characterized by slightly bulging
epidermal cells (Fig. 1D, F).
Androecium—The androecium is composed of six stamens
in two alternating whorls with stamens of the outer whorl
opposite to the outer tepals (Figs. 2D, 5, 6). All stamens are
free from each other and free from the perianth (Fig. 2E, F).
Stamens are differentiated into anther and filament without a
distinct joint at the base of the anther (Fig. 3A). The filament is
broad and c. 0.7 mm long. Anthers are basifixed (Fig. 3A) and
c. 0.9 mm long including the rounded and dorsiventrally
flattened apical connective protrusion (Fig. 3B, C). Number of
sporangia per thecae appears to be one, and anthers are thus
bisporangiate (Fig. 3G, H). Opening of the thecae is latrorse
(Fig. 3B, H). Stomium shape indicates dehiscence by a distally
hinged valve (Fig. 3B, D). Stamens are apparently glabrous.
Pollen—A single pollen grain was found inside one of the
open anthers (Fig. 3E). It is c. 35 lm long and infolded. This
infolded area has the same ornamentation as the rest of the
pollen grain, which suggests that the pollen is probably
inaperturate. The pollen wall surface is finely reticulate with an
ornamentation of evenly spaced, irregularly shaped elements
(gemmae, Fig. 3F) up to 1 lm long. The surface of the pollen
and part of the sporangium are further covered with a fine
granular material of probable tapetal origin.
Gynoecium—The gynoecium consists of a single carpel, c.
1.6 mm long (Fig. 4A, E, F). The stigmatic region is not
completely preserved (Fig. 4C). The external surface of the
basal part of the style and the ovary are covered with simple
trichomes (Fig. 4D). The ovary is 0.4 mm long and four-lobed
in cross section. The style is 1.2 mm long with a cross section
that is circular for most of its length but triangular toward the
base (Figs. 4G–K, 5A–K). The ovary is superior and
Fig. 6. Line drawing reconstruction of Potomacanthus lobatus (PP44882). (A) Lateral view showing outer and inner tepals. (B) Lateral view with one
of the inner tepals removed showing the arrangement of the stamens in two whorls. (C) Apical view of open flower showing flower organ arrangement and
ventral slit of gynoecium. Scale bar ¼1 mm.
Fig. 7. Phylogenetic relationships of Laurales, redrawn from Renner
(2004). Branches in bold are well supported, the others have moderate
support.
December 2007] VON BALTHAZAR ET AL.—LAURACEAE FLOWER FROM THE EARLY CRETACEOUS 2047
Fig. 8. Character states for androecia of Lauraceae optimized onto the Bayesian estimate of phylogeny for the family (Rohwer and Rudolph, 2005).
Interrupted lines indicate equivocal reconstructions. (A) Fertile stamens. (B) Staminodes. (C) Staminal appendages. (D) Sporangia number.
2048 AMERICAN JOURNAL OF BOTANY [Vol. 94
completely ascidiate, whereas the style is plicate (Fig. 4B, D).
In the style, the ventral slit appears postgenitally fused (Figs.
4F, H, 5B–E). A conspicuous stylar canal is present for most of
the length of the style (Figs. 4F, H, 5B–E, 6). Whether this
canal was open at anthesis or filled with secretion is unclear. A
single ovule is present in the ovary (Figs. 4I, J, 5F–I). Insertion
of the pendant ovule is median and ventral-apical.
DISCUSSION
Interpretation of the fossil flower and its systematic
affinity—The use of synchrotron-radiation x-ray tomographic
microscopy (SRXTM) was essential to obtain information on
the floral organization of the single specimen available,
especially to reveal anatomical information in a nondestructive
way. Its use was particularly important for clarifying the
internal structure of the gynoecium. In contrast to SEM and
serial sectioning, the SRXTM-method provided information on
both external and internal structure in three dimensions (see
also Donoghue et al., 2006; Friis et al., 2007).
The fossil flower is characterized by a perianth comprising
two whorls of three tepals each, an androecium in two whorls
of three stamens each, bisporangiate stamens with valvate
dehiscence, a gynoecium of a single carpel that is plicate in the
style but ascidiate in the ovary, and a single pendant ovule (see
reconstruction of the flower and floral diagram, Figs. 6, 9).
Among extant angiosperms, this combination of flower
characters is only present in representatives of Lauraceae that
have two of the normally three or four stamen whorls
remaining. Flowers of other families in the order Laurales
(Calycanthaceae, Siparunaceae, Gomortegaceae, Atherosper-
mataceae, Hernandiaceae, and Monimiaceae; Renner, 1999)
are distinct from the fossil (Table 1, Fig. 7) (e.g., Schodde,
1969; Endress, 1980; Forman, 1983; Endress and Hufford,
1989; Kubitzki, 1993a–c; Philipson, 1993; Rohwer, 1993;
Endress and Igersheim, 1997; Renner, 1999, 2004; Stevens,
2001; Endress and Lorence, 2004; Staedler et al., 2007). A
number of floral characters in particular separate the fossil from
Monimiaceae and Hernandiaceae, the closest related families of
Lauraceae. While we cannot unequivocally place the fossil in
either the crown group Lauraceae or along the stem of the
lauraceous clade because of uncertain optimization of key
characters in phylogenetic analyses of extant Laurales, based
on the available evidence, we think it highly likely that the
fossil is more closely related to Lauraceae than to any other
extant family.
Flowers of Calycanthaceae differ from the fossil in the
presence of many spirally arranged tepals and stamens, anther
dehiscence by longitudinal slits (valves in Sinocalycanthus),
and the presence of often two ovules in the carpels. Flowers of
Siparunaceae, Atherospermataceae, as well as Gomortegaceae
differ in having more than one carpel, and flowers of
Siparunaceae and Atherospermataceae differ further by having
ovules in a basal position in the carpels. The merism and
phyllotaxis of the perianth and androecium in these families is
variable. In Gomortegaceae, for example, neither perianth nor
androecium are clearly spiral or clearly trimerously whorled.
Flowers of Gomortegaceae are further distinguished by the
presence of intermediate structures that create transitional
series from tepals through stamens to staminodes as well as two
or more united carpels. Flowers of Siparunaceae further differ
in their closed flowers and bisporangiate anthers with a single
valve. Flowers of Hernandiaceae differ from the fossil in
having an inferior ovary and few stamens in one whorl that all
bear staminal appendages. Flowers of Monimiaceae differ in
being unisexual (except Hortonia), usually in having many
carpels, and in having a perianth and androecium that is mostly
tetramerous or in decussate pairs (e.g., Schodde, 1969;
Endress, 1980; Endress and Hufford, 1989; Kubitzki, 1993a–
c; Philipson, 1993; Rohwer, 1993; Endress and Igersheim,
1997; Renner et al., 1997; Stevens, 2001; Staedler et al., 2007).
The fairly stable organization of the flower in Lauraceae has
long allowed clear delimitation of the family. However,
relationships among the genera remained unsettled (e.g., see
van der Werff and Richter, 1996) until the use of molecular
data for phylogenetic analyses, and a solid framework for
intrafamiliar relationships was established (Rohwer, 2000;
Chanderbali et al., 2001; Rohwer and Rudolph, 2005). Recent
analyses place the monotypic genus Hypodaphnis as sister to
all other Lauraceae (Chanderbali et al., 2001; Rohwer and
Rudolph, 2005). The next split in the phylogeny is between a
well-supported Cryptocaryeae clade and the remainder of the
family, which consists of a well-supported clade in which
Cassytha,Neocinnamomum/Caryodaphnopsis, and the Mezi-
laurus group are successive sister groups in a pectinate
arrangement to the species-rich core Lauraceae (Chanderbali
et al., 2001; Rohwer and Rudolph, 2005).
In the context of the current phylogenetic framework for the
family, characters of the androecium have some degree of
Fig. 9. Androecium diversity of Lauraceae from the Puddledock
locality. Floral diagrams (A–B). (A) Potomacanthus lobatus floral
diagram (PP44882; sample 82). (B) Reconstructed floral diagram based
on two flower fragments (PP43735, PP43751; sample 82; shading
indicates staminal appendages; redrawn from Figs. 3a, 4 in Crane et al.
[2004]). Diagrams of anthers in lateral view showing dehiscence (C–D).
(C) Potomacanthus lobatus (PP44882; sample 82). (D) Stamen fragment
(PP43768; sample 73; redrawn from Figs. 3c–3e in Crane et al. [2004]).
December 2007] VON BALTHAZAR ET AL.—LAURACEAE FLOWER FROM THE EARLY CRETACEOUS 2049
plasticity (Fig. 8A–D). In most taxa, there are three whorls of
stamens with staminal appendages inserted on the third (inner)
whorl. This condition appears also to be the ancestral one for
the family (Fig. 8A, C). However, in several lineages, and
especially in the Cryptocaryeae and Mezilaurus group
(including Williamodendron and Sextonia), there is a reduction
to one or two whorls of stamens. This reduction results in the
loss of staminal appendages in several representatives of the
Mezilaurus group (Rohwer, 1993). A similar pattern of
reduction may account for the morphology of the fossil flower.
The presence of only six fertile stamens in two whorls and the
absence of staminal appendages may reflect loss of the
innermost stamen whorl. In this case, the fossil would be
placed within extant Lauraceae. Notwithstanding this variabil-
ity in extant Lauraceae, however, the androecium of the fossil
(with six fertile stamens in two whorls, anthers with two
valves, no staminodes and no filament appendages) is not
found in any extant Lauraceae. In this context, it is possible
that the configuration of the androecium in the fossil represents
the plesiomorphic condition in the Lauraceae stem group, and
we are therefore cautious about its precise phylogenetic
placement.
The gynoecium of the fossil consists of a single carpel with a
single ovule, as is invariably the case in extant taxa of
Lauraceae (Rohwer, 1993). In the cross sections of the fossil
gynoecium, however, an open canal in the style is clearly
present. It is possible that this canal was filled by a secretion at
anthesis. No open canals have been reported for extant
representatives of Lauraceae, but the innermost tissues of the
ventral slit are composed of presumed secretory cells with
swollen cell walls (Endress and Igersheim, 1997). This tissue
could have collapsed during fossilization. However, open stylar
canals are known from the closely related families Calycan-
thaceae and Gomortegaceae and from some Monimiaceae
(Endress and Igersheim, 1997).
Pollen grains of extant Lauraceae are generally characterized
by an extremely thin exine, which reduces their preservation
potential and thus contributes to their sparse fossil record (e.g.,
Erdtman, 1952; Herendeen et al., 1994; Eklund, 2000;
Sampson, 2000). It is not completely certain that the single
pollen grain found inside one of the anthers of the fossil was
produced by the flower. Some of the pollen characters suggest
a lauraceous relationship, but details of pollen ornamentation
differ from that of any fossil or extant member of the family.
Most extant Lauraceae have the pollen exine covered with
spines (Raj and van der Werff, 1988; van der Merwe et al.,
1990; Shang and Tang, 1995). The exine is strongly verrucate
only in Eusideroxylon,Potoxylon,Crypotcarya, and Raven-
sara.InCassytha, additional scattered papillate processes are
found on the verrucate exine (van der Merwe et al., 1990;
Shang and Tang, 1995). The pollen exine of Dahlgrenoden-
dron is distinct from that of all other Lauraceae in being striate
(van der Merwe et al., 1990). Among other Laurales, the exine
structure of the fossil pollen grain is somewhat similar to that
of extant Gomortegaceae or some Atherospermataceae, in
which the exine is ornamented with punctae or clavae
(Foreman, 1983; Hesse and Kubitzki, 1983).
Pollination biology—In most Lauraceae, the staminal
appendages and staminodes are nectariferous. Pollinators
comprise a relatively broad spectrum of bees, flies, and beetles
but also other insects (Kubitzki and Kurz, 1984; Endress, 1990;
Forfang and Olesen, 1998). However, some genera have
staminal appendages that lack a nectariferous function (e.g.,
Aniba,Licaria; Kubitzki and Kurz, 1984), and a few genera
have lost their staminal appendages altogether (e.g., Anaueria,
Mezilaurus,Williamodendron; Rohwer, 1993) as is probably
also the case for the fossil flower. Pollen seems to be the only
compensation offered to visiting insects in these nectarless
representatives (Kubitzki and Kurz, 1984). Small pollen-
collecting Trigona bees (Meliponinae) are reported to be the
main flower visitor of the nectarless Aniba and Licaria as well
TABLE 1. Comparison of floral characters of Potomacanthus lobatus with extant families of Laurales.
a
Character
Taxon
Potomacanthus Calycanthaceae Siparunaceae Gomortegaceae
Flower Bisexual Bisexual Unisexual Unisexual
Perianth no. 6 5–50 4–6(7) plus calyptra 7–10, intergrading with stamens
Perianth phyllotaxis Trimerous whorls Spiral Whorls Intermediate between spiral and
di- or trimerous whorls
Stamen no. 6 5–30 (1–)2–70 7–13, outer petaloid
Androecium phyllotaxis Trimerous whorls Spiral Irregular, or when low
numbers tetramerous
Intermediate between spiral and
di- or trimerous whorls
Sporangia no. 2 4 2 with one valve 2
Anther dehiscence Valves Longitudinal slits (valves
in Sinocalycanthus)
Valves Valves
Staminal appendages Absent Absent Absent Present
Staminodes Absent Present Absent Present
Pollen Presumably inaperturate,
exine finely reticulate with gemmae
Disulcate, exine psilate Inaperturate Inaperturate, ornamented with punctae
Carpel no. 1 1–43 3–30 (2–)3(5)
Ovary position Superior Inferior Inferior Inferior
Ovule no./carpel 1 1 (2) 1 1
Ovule insertion Apical Basal Basal Apical
a
Notes: Data from Schodde (1969), Endress (1980), Forman (1983), Endress and Hufford (1989), Kubitzki (1993a–c), Philipson (1993), Rohwer
(1993), Endress and Igersheim (1997), Renner et al. (1997), Stevens (2001 onwards), Endress and Lorence (2004), Staedler et al. (2007).
2050 AMERICAN JOURNAL OF BOTANY [Vol. 94
as of Mezilaurus at the time of anther dehiscence (Kubitzki and
Kurz, 1984; van der Werff, 1987). In closely related families
where no nectar is present in the flowers, other pollination
systems with pollen-feeding beetles or gall midges are present
(Monimiaceae, Lorence, 1985; Siparunaceae, Renner et al.,
1997).
Pollen presentation at anthesis in extant Lauraceae is on the
opened flaplike anther valves (Endress, 1990). A similar mode
of pollen presentation may have occurred in the fossil. In extant
Lauraceae, the pollen is sticky because of small drops of
adhesive material on the exine (Kubitzki, 1981; Hesse and
Kubitzki, 1983). It has been suggested that these droplets help
pollen stick together and thus increase the efficiency of pollen
collection by insect visitors (Kubitzki and Kurz, 1984). Sticky
pollen, perhaps facilitated by the pollenkitt and the exine
structure, may also be important to allow the pollen to adhere
to the anther flaps. The same kind of pollen presentation
mechanism, and similar spiny exine structure, is known for the
closely related Hernandiaceae (Kubitzki, 1981, 1993c; Endress,
1990; Endress and Lorence, 2004).
Comparison with other fossil material—The fossil record
of flowers with lauraceous affinity from the Cretaceous has
grown rapidly in recent years, and 12 taxa have been reported
since 1990 (Table 2; for an overview, see also Eklund, 1999,
2000; Frumin et al., 2004).
The oldest records are from the Early Cretaceous Pud-
dledock locality in the United States. They include the fossil
flower described here, as well as two fragments of another
flower taxon and a single tetrasporangiate stamen (Crane et al.,
1994). The flower fragments come from the same sample
(sample 82) as the flower described here but differ in the fusion
of the lower part of the tepals to the outermost stamen whorl
and the presence of paired staminal appendages associated with
each stamen (Fig. 9). The center of these fragmentary flowers
is not preserved, and thus the exact number of stamen whorls is
unknown (Crane et al., 1994).
Other lauraceous flowers from younger sediments show that
the family was already widespread in the northern hemisphere
by the mid-Cretaceous (North America, Drinnan et al., 1990;
Herendeen et al., 1994; Eklund, 2000; Central Europe, Eklund
TABLE 2. Summary of androecium characters from Cretaceous fossil flowers and flower fragments with lauraceous affinity.
Taxon No. fertile stamen whorls Staminode whorl Staminal appendages No. sporangia Reference
Late Cretaceous
Mauldinia mirabilis 3 Present Present Disporangiate Drinnan et al., 1990
M. bohemica 3 Present Present Disporangiate Eklund and Kvacek, 1998
M. sp.? ? ? Disporangiate Herendeen et al., 1999
M. hirsuta ? Present Present Disporangiate Frumin et al., 2004
Perseanthus crossmanensis 3 Present Present ? Herendeen et al., 1994
Neusenia tetrasporangiate 3 Present Present Tetrasporangiate Eklund, 2000
Taxon A ? ? ? Disporangiate Eklund, 2000
Taxon B ? ? Present ? Eklund, 2000
Lauranthus futabensis 3 Not present Not present Tetrasporangiate Takahashi et al., 2001
Hypogynous flower type 2 More than 2 ? ? Tetrasporangiate Takahashi et al., 1999
Early Cretaceous
Flower fragments ? ? Present ? Crane et al., 1994
Stamen ? ? ? Tetrasporangiate Crane et al., 1994
Potomacanthus lobatus 2 Not present Not present Disporangiate This work
TABLE 1. Extended.
Taxon
Atherospermataceae Hernandiaceae Monimiaceae Lauraceae
Unisexual/bisexual Unisexual/bisexual Unisexual (except Hortonia) Unisexual/bisexual
6–20 6–8(12) 3 to many Mostly 6
Mostly di- or tetramerous whorls;
some trimerous or irregular
Two whorls of 3–4(6)
or one whorl of 4–8
Spiral or whorls Mostly trimerous whorls
(4–)6–100 3–5(7) 4 to many Mostly 3–12, some up to 32
When few mostly di- or tetramerous
whorls, otherwise spiral
1 whorl When few whorls Mostly trimerous whorls, some
dimerous or irregular
224(Monimia 2) 2 or 4
Valves Valves Longitudinal slits Valves
Present Present (except
Sparattanthelium)
Present in Hortonia,Peumus, Monimia Present (absent when androecia
reduced to two whorls)
Present Absent or present Present in Hortonia and Peumus (female) Mostly present
Di- or meridionally sulcate, exine
reticulate, some ornamented
with clavae
Inaperturate, exine echinate
or with globules
Inaperturate (Macrotorus monosulcate), exine Inaperturate, mostly echinate
3 to many 1 (1–few) to many 1
Superior Inferior Superior to inferior Mostly superior (some inferior)
111 1
Basal Apical Apical Apical
December 2007] VON BALTHAZAR ET AL.—LAURACEAE FLOWER FROM THE EARLY CRETACEOUS 2051
and Kvac
ˇek, 1998, A. Viehofen, Bedburg, Germany; C.
Hartkopf-Fro¨der, Geological Survey of North Rhine-West-
phalia, Krefeld, Germany unpublished manuscript; Central
Asia, Frumin et al., 2004; Japan, Takahashi et al., 1999, 2001).
The extinct genus Mauldinia Drinnan, Crane, Friis & Pedersen
is the most widely distributed of these Cretaceous fossils and is
known from six species described from North America, Central
Asia, and Europe (Drinnan et al., 1990; Herendeen et al., 1994;
Eklund and Kvac
ˇek, 1998; Frumin et al., 2004; A. Viehofen,
Bedburg, Germany; C. Hartkopf-Fro¨der, Geological Survey of
North Rhine-Westphalia, Krefeld, Germany, unpublished
manuscript). Mauldinia is characterized by bisexual flowers
that are arranged in distinctive compound inflorescences with
distinctly flattened bilobed lateral inflorescence units (e.g.,
Drinnan et al., 1990). Flowers are trimerous with two whorls of
tepals in which the outer whorl of tepals is shorter than the
inner one. The androecium consists of three whorls of fertile
stamens and an inner fourth whorl of staminodes. Paired
appendages are present on the third stamen whorl. Anthers
open by two apically hinged valves (for overview, see Frumin
et al., 2004). Mauldinia thus differs from the fossil flower
described here in the unequal sized of the tepals, an additional
whorl of fertile stamens, the presence of a staminode whorl,
and the presence of staminal appendages.
Other fossil flowers of Lauraceae from the Late Cretaceous,
such as Lauranthus (Takahashi et al., 2001), Perseanthus
(Herendeen et al., 1994), and Neusenia (Eklund, 2000), also
have three whorls of fertile stamens. Perseanthus and Neusenia
similarly have staminal appendages and an additional inner
whorl of staminodes. Neither staminal appendages nor an
additional inner whorl of staminodes were reported for
Lauranthus (Takahashi et al., 2001), but a reinvestigation of
the specimen by x-ray analyses indicates the presence of both
these characters also in this flower (M. Takahashi, Niigata
University, Japan, personal communication).
To summarize, it is evident that the androecia of fossil
Lauraceae from the Late Cretaceous typically have three
whorls of stamens, an additional whorl of staminodes, and
staminal appendages on the third stamen whorl. Variation in
androecial structure occurs primarily in the presence of
representatives with both bi- and tetrasporangiate anthers
(e.g., bisporangiate in Mauldinia, Frumin et al., 2004;
tetrasporangiate in Neusenia, Lauranthus, Eklund, 2000;
Takahashi et al., 2001). However, the fossil flower described
here is the only fossil flower of Lauraceae described so far that
lacks staminal appendages and has only two staminal whorls. It
contributes to the accumulating evidence from the Early and
mid-Cretaceous showing that the androecial organization was
already diverse at an early stage in the evolution of the
lauraceous clade.
Conclusions—The new fossil taxon Potomacanthus lobatus
gen. et sp. nov. from the Early Cretaceous (Early to Middle
Albian) Puddledock locality, Virginia, USA shares many
similarities with flowers of extant Lauraceae, indicating that
the fossil is probably more closely related to that family than to
any other extant taxon. Among extant Lauraceae, the
morphology of the androecium is more variable than that of
the perianth and gynoecium. Potomacanthus lobatus, together
with other material described previously from the Puddledock
locality, shows that variation in the form of the androecium
was also a feature of early members of the lauraceous clade.
Both tetrasporangiate and bisporangiate stamens, as well as
androecia with and without staminal appendages (Fig. 9), were
already present by the end of the Early Cretaceous.
Among extant taxa, where androecial structure can be
considered along with vegetative and other features, its
androecial variation is useful systematically. The reconstruc-
tions of androecial characters states among extant representa-
tives show that ancestral androecia are composed of three
whorls and possess staminodes. However, based on current
knowledge, the otherwise great flexibility in androecial
characters among extant taxa makes phylogenetic signals
based solely on androecial morphology weak, and it precludes
more precise systematic placement of the fossil material.
Nevertheless, it is of interest that Potomacanthus differs from
all other extant or fossil Lauraceae in the details of the
androecium. It most likely represents a separate and now
extinct taxon within the family Lauraceae or on the stem
lineage to the extant group.
Potomacanthus lobatus, together with previously described
material of Lauraceae (Crane et al., 1994) and the calycantha-
ceous flower Virginianthus (Friis et al., 1994), all from the
Puddledock locality, conclusively establish the presence of
Laurales before the end of the Early Cretaceous. The
Puddledock fossils also demonstrate that differentiation within
Laurales had already produced lineages that today are
represented by extant families, and that further structural
diversification was probably already underway in the Laur-
aceae. The early fossil record of plants similar to extant
Lauraceae further shows that the first major differentiation of
many angiosperm groups took place over a relatively short
period of time already during the Early Cretaceous.
LITERATURE CITED
BRENNER, G. J. 1963. The spores and pollen of the Potomac Group of
Maryland. Maryland Department of Geology, Mines & Water
Resources Bulletin 27: 1–215.
CHANDERBALI, A. S., H. VAN DER WERFF,AND S. S. RENNER. 2001.
Phylogeny and historical biogeography of Lauraceae: evidence from
the chloroplast and nuclear genomes. Annals of the Missouri
Botanical Garden 88: 104–134.
CRANE, P. R., E. M. FRIIS,AND K. R. PEDERSEN. 1994. Palaeobotanical
evidence on the early radiation of magnoliid angiosperms. Plant
Systematics and Evolution 8 (Supplement): 51–72.
DISCHINGER, J. B. 1987. Late Mesozoic and Cenozoic stratigraphic and
structural framework near Hopewell, Virginia. US Geological Survey
Bulletin 1567: 1–48.
DONOGHUE, P. C. J., S. BENGTSON, X.-P. DONG, N.-J. GOSTLING,T.
HULDTGREN,J.A.CUNNINGHAM,C.YIN,Z.YUE,F.PENG,AND M.
STAMPANONI. 2006. Synchrotron x-ray tomographic microscopy of
fossil embryos. Nature 442: 680–683.
DOYLE, J. A. 1969. Cretaceous angiosperms pollen of the Atlantic Coastal
Plain and its evolutionary significance. Journal of the Arnold
Arboretum 50: 1–35.
DOYLE, J. A. 1992. Revised palynological correlations of the lower
Potomac Group (USA) and the Cocobeach sequence of Gabon
(Barremian-Aptian). Cretaceous Research 13: 337–349.
DOYLE, J. A., AND L. J. HICKEY. 1976. Pollen and leaves from the mid-
Cretaceous Potomac Group and their bearing on early angiosperm
evolution. In C. B. Beck [ed.], Origin and early evolution of
angiosperms, 139–206. Columbia University Press, New York, New
York, USA.
DOYLE, J. A., AND E. I. ROBBINS. 1977. Angiosperm pollen zonation of the
continental Cretaceous of the Atlantic Coastal Plain and its
application to deep wells in the Salisbury Embayment. Palynology
1: 43–78.
DRINNAN, A. N., P. R. CRANE,E.M.FRIIS,AND K. R. PEDERSEN. 1990.
2052 AMERICAN JOURNAL OF BOTANY [Vol. 94
Lauraceous flowers from the Potomac group (mid-Cretaceous) of
eastern North America. Botanical Gazette 151: 370–384.
EKLUND, H. 1999. Big survivors with small flowers—fossil history and
evolution of Laurales and Chloranthaceae. Ph.D. dissertation,
Uppsala University, Uppsala, Sweden.
EKLUND, H. 2000. Lauraceous flowers from the Late Cretaceous of North
Carolina, U.S.A. Botanical Journal of the Linnean Society 132: 397–
428.
EKLUND, H., AND J. KVACEK. 1998. Lauraceous inflorescences and flowers
from the Cenomanian of Bohemia (Czech Republic, Central Europe).
International Journal of Plant Sciences 159: 668–686.
ENDRESS, P. K. 1980. Ontogeny, function and evolution of extreme floral
construction in Monimiaceae. Plant Systematics and Evolution 134:
79–120.
ENDRESS, P. K. 1990. Evolution of reproductive structures and function in
primitive angiosperms (Magnoliidae). Memoirs of the New York
Botanical Garden 55: 5–34.
ENDRESS, P. K., AND L. HUFFORD. 1989. The diversity of stamen structures
and dehiscence patterns among Magnoliidae. Botanical Journal of the
Linnean Society 100: 45–85.
ENDRESS, P. K., AND A. IGERSHEIM. 1997. Gynoecium diversity and
systematics of the Laurales. Botanical Journal of the Linnean Society
125: 93–168.
ENDRESS, P. K., AND D. H. LORENCE. 2004. Heterodichogamy of a novel
type in Hernandia (Hernandiaceae) and its structural basis.
International Journal of Plant Sciences 165: 753–763.
ERDTMAN, G. 1952. Pollen morphology and plant taxonomy. Angiosperms.
Almqvist & Wiksell, Stockholm, Sweden.
FOREMAN, D. B. 1983. The morphology and phylogeny of the
Monimiaceae (sensu lato) in Australia. Ph.D. dissertation, University
of New England, Armidale, Australia.
FORFANG, A.-S., AND J. M. OLESEN. 1998. Male-biased sex ratio and
promiscuous pollination in the dioecious island tree Laurus azorica
(Lauraceae). Plant Systematics and Evolution 212: 143–157.
FRIIS, E. M., P. R. CRANE,AND K. R. PEDERSEN. 1997. Anacostia, a new basal
angiosperm from the Early Cretaceous of North America and Portugal
with monocolpate/trichotomocolpate pollen. Grana 36: 225–244.
FRIIS, E. M., P. R. CRANE,K.R.PEDERSEN,S.BENGTSON,P.C.J.
DONOGHUE,G.W.GRIMM,AND M. STAMPANONI. 2007. Phase-contrast
x-ray microtomography links Cretaceous seeds with Gnetales and
Bennettitales. Nature 450: in press.
FRIIS, E. M., H. EKLUND,K.RPEDERSEN,AND P. R. CRANE. 1994.
Virginianthus calycanthoides gen. et sp. nov.—a calycanthaceous
flower from the Potomac Group (Early Cretaceous) of eastern North
America. International Journal of Plant Sciences 155: 772–785.
FRIIS, E. M., K. R. PEDERSEN,AND P. R. CRANE. 1995. Appomattoxia
ancistrophora gen. et sp. nov., a new Early Cretaceous plant with
similarities to Circaeaster and extant Magnoliidae. American Journal
of Botany 82: 933–943.
FRIIS, E. M., K. R. PEDERSEN,AND P. R. CRANE. 2006. Cretaceous
angiosperm flowers: Innovation and evolution in plant reproduction.
Palaeogeography, Palaeoclimatology, Palaeoecology 232: 251–293.
FRUMIN, S., H. EKLUND,AND E. M. FRIIS. 2004. Mauldinia hirsuta sp. nov.,
a new member of the extinct genus Mauldinia (Lauraceae) form the
Late Cretaceous (Cenomanian-Turonian) of Kazakhstan. Internation-
al Journal of Plant Sciences 165: 883–895.
HERENDEEN, P. S., W. L. CREPET,AND K. C. NIXON. 1994. Fossil flowers
and pollen of Lauraceae from the Upper Cretaceous of New Jersey.
Plant Systematics and Evolution 189: 29–40.
HESSE, M., AND K. KUBITZKI. 1983. The sporoderm ultrastructure in Persea,
Nectandra,Hernandia,Gomortega and some other lauralean genera.
Plant Systematics and Evolution 141: 299–311.
HICKEY, L. J., AND J. A. DOYLE. 1977. Early Cretaceous fossil evidence for
angiosperm evolution. Botanical Review 43: 3–104.
KUBITZKI, K. 1981. The tubular exine of Lauraceae and Hernandiaceae, a
novel type of exine structure in seed plants. Plant Systematics and
Evolution 138: 139–146.
KUBITZKI, K. 1993a. Calycanthaceae. In K. Kubitzki [ed.], The families
and genera of vascular plants, vol. 2, 197–200. Springer Verlag,
Berlin, Germany.
KUBITZKI, K. 1993b. Gomortegaceae. In K. Kubitzki [ed.], The families
and genera of vascular plants, vol. 2, 318–320. Springer Verlag,
Berlin, Germany.
KUBITZKI, K. 1993c. Hernandiaceae. In K. Kubitzki [ed.], The families and
genera of vascular plants, vol. 2, 334–338. Springer Verlag, Berlin,
Germany.
KUBITZKI, K., AND H. KURZ. 1984. Synchronized dichogamy and dioecy in
neotropical Lauraceae. Plant Systematics and Evolution 147: 253–
266.
LORENCE, D. H. 1985. A monograph of the Monimiaceae (Laurales) in the
Malagasy region (S. W. Indian Ocean). Annals of the Missouri
Botanical Garden 72: 1–165.
MADDISON, D. R., AND W. P. MADDISON. 2000. MacClade 4.0. Sinauer,
Sunderland, Massachusetts, USA.
PHILIPSON, W. R. 1993. Monimiaceae. In K. Kubitzki [ed.], The families
and genera of vascular plants, vol. 2, 426–437. Springer Verlag,
Berlin, Germany.
RAJ, B., AND H. VAN DER WERFF. 1988. A contribution to the pollen
morphology of neotropical Lauraceae. Annals of the Missouri
Botanical Garden 75: 130–167.
RENNER, S. S. 1999. Circumscription and phylogeny of the Laurales:
evidence from molecular and morphological data. American Journal
of Botany 86: 1301–1315.
RENNER, S. S. 2004. Variation in diversity among Laurales, Early
Cretaceous to Present. Biologiska Skrifter 55: 441–458.
RENNER, S. S., A. E. SCHWARZBACH,AND L. LOHMAN. 1997. Phylogenetic
position and floral function of Siparuna (Siparunaceae: Laurales).
International Journal of Plant Sciences 158 (6 Supplement): S89–
S98.
ROHWER, J. G. 1993. Lauraceae. In K. Kubitzki [ed.], The families and
genera of vascular plants, vol. 2, 366–391. Springer Verlag, Berlin,
Germany.
ROHWER, J. G. 2000. Toward a phylogenetic classification of the
Lauraceae: evidence from matK sequences. Systematic Botany 25:
60–71.
ROHWER, J. G., AND B. RUDOLPH. 2005. Jumping genera: the phylogenetic
positions of Cassytha,Hypodaphnis, and Neocinnamomum (Laur-
aceae) based on different analyses of trnK intron sequences. Annals of
the Missouri Botanical Garden 92: 153–178.
SAMPSON, F. B. 2000. Pollen diversity in some modern Magnoliids.
International Journal of Plant Sciences 161 (Supplement): S193–
S210.
SCHODDE, R. 1969. A monograph of the family Atherospermataceae R. Br.
Ph.D. dissertation, University of Adelaide, Adelaide, Australia.
SHANG, C., AND G. TANG. 1995. Pollen morphology of the family
Lauraceae. Cathaya 7: 53–62.
STAEDLER, Y. M., P. H. WESTON,AND P. K. ENDRESS. 2007. Floral
phyllotaxis and floral architecture in Calycanthaceae (Laurales).
International Journal of Plant Sciences 168: 285–306.
STEVENS, P. F. 2001 [onwards]. Angiosperm phylogeny website, version 7,
May 2006 [and more or less continuously updated since]. Available at
http://www.mobot.org/MOBOT/research/APweb/.
TAKAHASHI, M., P. R. CRANE,AND H. ANDO. 1999. Fossil flowers and
associated plant fossils from the Kamikitaba locality (Ashizawa
Formation, Futaba Group, Lower Coniacian, Upper Cretaceous) of
Northeast Japan. Journal of Plant Research 112: 187–206.
TAKAHASHI, M., P. S. HERENDEEN,AND P. R. CRANE. 2001. Lauraceous fossil
flower from the Kamikitaba locality (Lower Coniacian; Upper
Cretaceous) in northeastern Japan. Journal of Plant Research 114:
429–434.
VAN DER MERWE, J. J. M., A. E. VAN WYK,AND P. D. F. KOK. 1990. Pollen
types in the Lauraceae. Grana 29: 185–196.
VAN DER WERFF, H. 1987. A revision of Mezilaurus (Lauraceae). Annals of
the Missouri Botanical Garden 74: 153–182.
VAN DER WERFF, H., AND H. G. RICHTER. 1996. Toward an improved
classification of Lauraceae. Annals of the Missouri Botanical Garden
83: 409–418.
December 2007] VON BALTHAZAR ET AL.—LAURACEAE FLOWER FROM THE EARLY CRETACEOUS 2053
