Content uploaded by Howard Falcon-Lang
Author content
All content in this area was uploaded by Howard Falcon-Lang on Jan 21, 2015
Content may be subject to copyright.
1. Introduction
Fossil forests occur at multiple horizons in the mid-
Cretaceous (late Albian) Triton Point Formation,
Fossil Bluff Group of southeast Alexander Island,
Antarctic Peninsula (Jefferson, 1982; Moncrieff &
Kelly, 1993; Nichols & Cantrill, unpub. data) (Fig. 1).
This formation accumulated within a fore-arc basin
and is divided in two members; the lower Citadel
Bastion Member consists of braided alluvial plain
deposits and the upper Coal Nunatak Member consists
of coastal plain deposits characterized by more later-
ally restricted (probably meandering) river channels
(Cantrill & Nichols, 1996; Nichols & Cantrill, unpub.
data). Upright, coniferous tree trunks (up to seven
metres high) occur in both facies associations where
they are rooted in carbonaceous palaeosols and buried
by coarse-grained sandstone units interpreted as major
flood deposits (A. C. M. Moncrieff, unpub. British
Antarctic Survey Field Report AD6/2R/1988/G4,
1989). Transported logs and wood fragments occur in
siltstone and sandstone units variously interpreted as
river channel deposits, flood deposits and crevasse
splay deposits. Abundant foliage remains occur on
palaeosols (e.g. Cantrill & Nichols, 1996) and in fine-
grained overbank deposits (Jefferson, 1982).
Quantitative analysis of 68 wood fragments from this
formation has revealed the presence of araucarian
(Araucarioxylon Kraus and Araucariopitys Jeffrey; 13 %
of specimens), podocarp (Podocarpoxylon Gothan sp. 1
and sp. 2; 85% of specimens) and taxodioid conifers
(Taxodioxylon Hartig; 2% of specimens). This conifer
diversity is mirrored in the foliage record that includes
Araucaria Juss. and Araucarites Presl. (Araucariaceae);
Podocarpites Unger (Podocarpaceae); Athrotaxites
Andra (Taxodiaceae); Brachyphyllum Brongniart,
Elatocladus Halle, Pagiophyllum Heer, and Podozamites
(Brongniart) Braun (Incertae Familae) (Cantrill &
Falcon-Lang, unpub. data). Estimates of tree height
based on stump diameter suggest that these conifers were
tall trees up to 29 m high (Falcon-Lang & Cantrill, 2000).
Other plant fossils present include liverworts, lycopsids,
sphenopsids, ferns, ginkgophytes, taeniopterids, bennet-
titaleans and angiosperms (Cantrill & Nichols, 1996).
Preliminary palaeoecological reconstructions based
on the distribution of plant remains on exposed
palaeosol surfaces indicate that the braidplain was
dominated by shrubby taeniopterids interspersed with
stands of podocarp and taxodioid conifers (Cantrill &
Nichols, 1996). On the coastal plain, open canopy
forests dominated by podocarp and araucarian
conifers occurred in some regions (Jefferson, 1982;
Chaloner & Creber, 1989) and graded into broken
conifer woodlands and fern thickets on more imma-
ture substrates (Cantrill, 1995a,b; Cantrill & Nichols,
1996; Falcon-Lang & Cantrill, 2000). This temperate
rainforest vegetation grew well within the Southern
Hemisphere polar circle (palaeolatitude of 75° S)
during the mid-Cretaceous thermal optimum (Creber
& Chaloner, 1985; Spicer, Rees & Chapman, 1993;
Geol. Mag. 138 (1), 2001, pp. 39–52. Printed in the United Kingdom © 2001 Cambridge University Press 39
Leaf phenology of some mid-Cretaceous polar forests,
Alexander Island, Antarctica
H. J. FALCON-LANG* & D. J. CANTRILL
British Antarctic Survey, High Cross, Madingley Road, Cambridge CB3 0ET, UK
(Received 30 March 2000; accepted 10 October 2000)
Abstract – The leaf longevity and seasonal timing of leaf abscission within a plant community is
closely related to climate, a phenomenon referred to as leaf phenology. In this paper the leaf phenol-
ogy of some mid-Cretaceous (late Albian) forests which grew at latitude of 75° S on Alexander Island,
Antarctica, is analysed. Five independent techniques for determining leaf longevity are applied to the
fossil remains of each of the canopy-forming trees. These techniques utilize: (1) the anatomical char-
acter of growth rings in trunk woods, (2) leaf trace persistence in juvenile branch and stem woods, (3)
leaf physiognomy, (4) comparison with nearest living relatives, and (5) leaf taphonomy. The applica-
tion of techniques 1–5 suggests that the araucarian and podocarp conifers, which comprised more
than 90 % of the canopy-forming vegetation, were evergreen with leaf retention times in excess of 5–13
years. The application of techniques 3–5 to rare taxodioid conifers indicates the existence of both ever-
green and deciduous habits in this group, whilst both ginkgos and taeniopterids, which are locally
abundant, are interpreted as possessing a deciduous habit. The polar forests of Alexander Island were
therefore dominantly evergreen. Preliminary analysis of five other mid-Cretaceous polar forests sug-
gests the presence of dominantly evergreen vegetation in Australia and Antarctica, and mixed ever-
green–deciduous vegetation in Alaska, northern Russia and New Zealand. Cold month mean
temperature probably exerted the largest influence on the leaf phenology at each of these forest sites.
* Author for correspondence: falconlang@hotmail.com
Clarke & Jenkyns, 1999). In this environment, plant
growth was subject to conditions of extreme light sea-
sonality characterized by up to 70 days of unbroken
darkness each winter and a humid climate (Parrish,
Ziegler & Scotese, 1982; Read & Francis, 1992).
Despite this broad understanding of the nature of the
Alexander Island ecosystem, it is still not known
whether these forests were dominantly evergreen or
deciduous. It is particularly important to understand
this palaeoecological dimension because the leaf
longevity and seasonal timing of leaf abscission in a
plant community is closely related to climate, a phenom-
enon referred to as leaf phenology (Woodward, 1987).
Analyses of the leaf phenology of fossil plant communi-
ties could yield valuable palaeoclimatic data needed to
test and refine numerical models of past global climates,
vegetation and environments (Wolfe & Upchurch, 1987;
Otto-Bliesner & Upchurch, 1997; Upchurch, Otto-
Bliesner & Scotese, 1999; Falcon-Lang, 2000a). In addi-
tion, the presence or absence of winter vegetation has
important implications for the ecology of terrestrial
faunas described from the Cretaceous polar circle, such
as theropod, sauropod, and ornithopod dinosaurs
(Dettman et al. 1992). In this paper we make a system-
atic attempt to analyse the leaf longevity of the most
common trees in the mid-Cretaceous fossil forests of
Alexander Island. These data are then discussed in rela-
tion to palaeoclimate models to interpret the leaf phe-
nology of the Alexander Island forests, and compared
with preliminary analyses of leaf phenology from other
mid-Cretaceous polar forest sites in Alaska, Australia,
New Zealand and northern Russia.
2. Analysis of the leaf longevity of the Alexander
Island trees
There is no single, simple way of determining the leaf
longevity of a fossil plant. In an earlier paper one of us
(Falcon-Lang, 2000a) reviewed some of the tech-
niques available for such analysis but stressed the need
to utilize as many techniques as possible in order to
test and refine the interpretation. Here we critically
describe five independent techniques for analysing fossil
leaf longevity and apply each of them to the remains
of the canopy-forming vegetation on Alexander
Island, the conifers, ginkgos and taeniopterids.
2.a. Growth rings in trunk woods
2.a.1. Method
The first technique, developed by Falcon-Lang
(2000a,b), utilizes the anatomical characteristics of
growth rings in coniferopsid trunk woods. In trans-
verse section, growth rings are composed of parallel
rows of tracheid cells which decrease in diameter
across the ring. The larger cells are produced in the
early part of the growing season (earlywood) and the
smaller cells are formed towards the end of the grow-
40 H.J.FALCON-LANG &D.J.CANTRILL
Figure 1. Location details of study area. (a) Map of Antarctic
Peninsula showing the position of the Alexander Island fossil
forests; (b) detailed outcrop map of southeast Alexander
Island showing the outcrop of the Albian Triton Point
Formation, a stratigraphic unit within the Fossil Bluff Group.
Dashed line delimits inferred extent of Triton Point Formation
(Moncrieff & Kelly, 1993). Principal localities mentioned in
the text where the fossil wood was collected are shown.
ing season (latewood). In this technique the radial
diameters of individual tracheid cells (viewed using
standard transverse thin sections) are measured across
each ring increment with the aid of a calibrated scale
mounted in the eye-piece of the transmitted light
microscope. Five adjacent rows of cells are measured
for each ring increment and averaged (cf. Falcon-
Lang, 1999a). The cumulative sum of each cell’s devia-
tion from the mean of this data is then calculated
(Creber & Chaloner, 1984) and plotted as a zero-
trending curve (the CSDM curve). For each growth
ring increment the percentage skew of the zenith of
the CSDM curve relative to the centre of the plot is
ascertained (Falcon-Lang, 2000a) (Fig. 2). Only
growth rings lacking subtle false rings, growth inter-
ruptions or any other growth abnormalities can be
used in this analysis because the presence of these fea-
tures gives rise to CSDM curves with multiple zeniths.
Falcon-Lang (2000a,b) showed that the percentage
skew of the zenith of the CSDM curve is closely related
to the leaf longevity of the parent tree in modern conif-
eropsids. Only deciduous species possess CSDM skews
of less than –10 % and only evergreen species have
skews of greater than +10 %. Of the woods which
yielded skew values between –10 % and 0 %, 81% were
from deciduous conifers and 19% from evergreen
conifers, whilst between 0 % and +10% skew, 46 % were
from deciduous conifers and 54% from evergreen
species. Woods that possess growth rings with CSDM
skews in the range from –10 % to +10 % therefore can-
not be unambiguously assigned evergreen or deciduous
status, although the majority with a negative percent-
age are likely to be deciduous. Nevertheless, in many
cases the woods produced by deciduous and evergreen
conifers may be distinguished from one another on the
basis of percentage skew values.
2.a.2. Data and interpretation
Poor preservation meant that the only Araucarioxylon
wood specimen could not be analysed by the present
technique. However, two specimens of Araucariopitys,
eight specimens of Podocarpoxylon sp. 1, four speci-
mens of P. sp. 2 and one specimen of Taxodioxylon
possessed sufficiently well-preserved growth rings for
the study. The former two genera possessed growth
rings with subtle ring boundaries continuous around
the stem circuit (Fig. 3a,b); the latter genus possessed
Cretaceous leaf phenology 41
Figure 2. Diagram illustrating how percentage skew is
calculated (see text for explanation; after Falcon-Lang, 2000a).
Figure 3. Transverse sections of conifer woods, (a)
Podocarpoxylon sp. 1 exhibiting subtle growth rings, mag.
×25, KG. 4710.1; (b) Araucariopitys exhibiting subtle
growth rings, mag. ×25, KG. 4740.11; (c) Taxodioxylon
exhibiting marked growth rings, mag. ×25, KG. 4626.1.
much more marked ring boundaries continuous
around the stem circuit (Fig. 3c). Five ring increments
were measured for Araucariopitys,Podocarpoxylon sp.
2 and Taxodioxylon, and ten ring increments were
measured for P. sp. 1. The results were as follows:
Araucariopitys gave percentage skew values ranging
from +34.6% to +61.8 %, for Podocarpoxylon sp. 1
values ranged from +38.1% to +76.1 %, and for P. sp.
2 values ranged from +51.5% to +83.1 % (Table 1;
Figs 4, 5). Compared with the modern data in Falcon-
Lang (2000a) and summarized above, these skew val-
ues are consistent with woods produced by evergreen
conifers with long leaf retention times, probably more
than five years. For example, modern specimens of
Araucaria araucana (Molina) K. Koch, which retain
leaves for 3–15 years, give a mean skew value of
+66.7 % (Falcon-Lang, 2000a). Taxodioxylon percent-
age skew values range from –11.7% to +6.38 % (Table
1; Figs 4, 5). Compared with modern data, these val-
ues straddle the deciduous and the evergreen fields
with short leaf retention times. Therefore the leaf
longevity of the parent tree cannot be interpreted with
certainty; it may have been deciduous or evergreen
with a short leaf retention time.
2.b. Leaf traces in juvenile branch and stem woods
2.b.1. Method
The second technique utilizes the leaf traces in juve-
nile stem or branch coniferopsid wood. Leaf traces are
bundles of vascular tissue which join the living leaf to
the primary vascular cylinder at the centre of the
young stem. In deciduous plants, the onset of sec-
ondary growth and the centripetal addition of a layer
of secondary wood by the vascular cambium, buries
the proximal part of the leaf trace and ruptures away
the distal portion: leaf traces thus are only encoun-
tered embedded in the first ring increment of such
stems (Eames & McDaniels, 1947; Esau, 1977). In
evergreen plants, leaf traces are extended by a type of
secondary growth which increases them in length by
addition of new tissue. As the woody plant stem
increases in diameter by secondary growth, the
phloem is progressively stripped away from the leaf
trace, so that the buried portion of the trace consists
only of a xylem strand (Eames & McDaniels, 1947;
Esau, 1977). Thus, where a tree has a high leaf reten-
tion time, as in Araucaria araucana, the leaf trace
may be very long, passing through many growth ring
42 H.J.FALCON-LANG &D.J.CANTRILL
Table 1. Results of the application of Falcon-Lang’s (2000a) ‘percentage skew technique’ to the Alexander Island coniferous woods
Wood taxon Number of Percentage Interpretation of
(Specimen number) Locality Facies association cells per ring skew leaf phenology
Araucariopitys
KG. 4740.11 Coal Nunatak Coastal plain 135 +51.1 Evergreen
KG. 4740.11 Coal Nunatak Coastal plain 52 +34.6 Evergreen
KG. 4740.11 Coal Nunatak Coastal plain 136 +61.8 Evergreen
KG. 4702.17 Coal Nunatak Coastal plain 30 +40.0 Evergreen
KG. 4702.17 Coal Nunatak Coastal plain 46 +52.2 Evergreen
Mean 79.8 +47.9 Evergreen
Podocarpoxylon sp. 1
KG. 4657.9 Titan Nunataks Braided alluvial plain 30 +66.7 Evergreen
KG. 4657.9 Titan Nunataks Braided alluvial plain 24 +58.3 Evergreen
KG. 4660.4 Citadel Bastion Braided alluvial plain 39 +43.6 Evergreen
KG. 4710.2 Triton Point Braided alluvial plain 21 +38.1 Evergreen
KG. 4717.42 Titan Nunataks Braided alluvial plain 51 +52.9 Evergreen
KG. 4717.44 Titan Nunataks Braided alluvial plain 20 +40.0 Evergreen
KG. 4717.45 Titan Nunataks Braided alluvial plain 67 +76.1 Evergreen
KG. 4717.45 Titan Nunataks Braided alluvial plain 41 +41.4 Evergreen
KG. 4717.46 Titan Nunataks Braided alluvial plain 51 +69.8 Evergreen
KG. 4719.4 Titan Nunataks Braided alluvial plain 45 +64.4 Evergreen
Mean 38.9 +55.1 Evergreen
Podocarpoxylon sp. 2
KG. 2814.256 Coal Nunatak Coastal plain 83 +83.1 Evergreen
KG. 2814.256 Coal Nunatak Coastal plain 84 +54.2 Evergreen
KG. 4660.1 Citadel Bastion Braided alluvial plain 41 +61.0 Evergreen
KG. 4710.3 Triton Point Braided alluvial plain 44 +59.1 Evergreen
KG. 4747.13 Coal Nunatak Coastal plain 33 +51.5 Evergreen
Mean 57 +61.8 Evergreen
Taxodioxylon
KG. 4626.1 Hyperion Nunataks Braided alluvial plain 128 –11.7 Deciduous
KG. 4626.1 Hyperion Nunataks Braided alluvial plain 74 +4.1 Evgrn/Decid.
KG. 4626.1 Hyperion Nunataks Braided alluvial plain 97 +4.6 Evgrn/Decid.
KG. 4626.1 Hyperion Nunataks Braided alluvial plain 134 –6.0 Evgrn/Decid.
KG. 4626.1 Hyperion Nunataks Braided alluvial plain 47 +6.4 Evgrn/Decid.
Mean 96 –0.5 Deciduous?
increments before being occluded by parenchyma
(Eames & McDaniels, 1947). By counting the number
of growth ring increments through which a leaf trace
passes before being occluded, a very precise estimate
of the leaf longevity of the parent plant can be made
(Fig. 6a; W. G. Chaloner, pers. comm., 1997; Falcon-
Lang, 1999a). However, this technique is difficult to
apply to fossil specimens because leaf traces, due to
their small size and widely spaced distribution around
the stem circumference, are rarely intersected in the
plane of thin sections; only a very few examples have
Cretaceous leaf phenology 43
Figure 4. Example graphs showing the Cumulative Sum of the Deviations from the Mean (CSDM) for four measured growth
rings in the Alexander Island woods. Note that Araucariopitys and the two Podocarpoxylon species possess CSDM curves with
high positive skew values whilst the CSDM curve of Taxodioxylon is almost symmetrical.
Figure 5. Results of analysis of Alexander woods using
Falcon-Lang’s (2000a) new technique. The grey area repre-
sents the zone of overlap between the deciduous and ever-
green fields based on data in Falcon-Lang (2000a). The
results imply that Araucariopitys and the two
Podocarpoxylon species were evergreen with long leaf reten-
tion times whilst Taxodioxylon may have been deciduous or
evergreen with a short leaf retention time.
been described in the literature with sufficient detail to
apply the technique (e.g. Gordon, 1935; Galtier &
Scott, 1994; Falcon-Lang & Cantrill, 2000). In addi-
tion, the technique is dependent on the ring incre-
ments in young woods representing annual periods of
growth, a questionable assumption for some fossil
wood assemblages (Falcon-Lang, 1999a).
2.b.2. Data and interpretation
In two specimens of Araucariopitys, vascular traces
290–415 µm in diameter (up to 0.140 mm2in cross-
sectional area; n=8), and composed of scalariform or
reticulate thickened tracheids of 6–8 µm diameter
were found. The first specimen (KG. 4702.28) con-
sisted of a 45 mm diameter woody cylinder surround-
ing a parenchymatous pith and was interpreted as part
of a small vertical stem because reaction wood was
absent (Creber, 1975). This fossil specimen possessed
two vascular traces which passed through five ring
increments before disappearing out of the plane of
section. The second specimen (KG. 4740.11) consisted
of a 65 mm diameter, strongly asymmetric woody
cylinder surrounding a parenchymatous pith. Marked
asymmetry in stem cross-section is most commonly
found in branches (where it is termed reaction wood)
and is due to the influence of gravity on wood forma-
tion. In gymnosperms the widest stem radius occurs
on the underside of the branch (Creber, 1975). On the
basis of this observation, specimen KG. 4740.11 is
interpreted as a lateral branch. Six vascular traces
extended from the lower side of this fossil branch and
passed through up to 13 ring increments before disap-
pearing from the plane of section; one such trace is
31 mm in length (Fig. 6b). Vascular traces were also
noted in one specimen of Podocarpoxylon sp. 1. This
specimen (KG. 4719.4) consisted of a woody cylinder
(90 mm in diameter), surrounding a poorly preserved
pith and was interpreted as a small vertical stem
because reaction wood was absent. Four large leaf
traces, up to 475 µm in diameter (up to 0.178 mm2
cross-sectional area) occurred close to the pith region
and extended through the first five growth increments
(Fig. 6c).
There are three pieces of evidence which suggest
that all these persistent vascular traces are leaf traces
and not small branch traces. First, the vascular traces
are almost identical in cross-sectional area (Fig. 6c,d)
to the leaf traces of modern araucarian conifers such
as Agathis australis (D. Don) Laudon (0.134 mm2:
Lacey, 1953) and Araucaria araucana (0.095 mm2:H.J.
Falcon-Lang, pers. obs., 1997). Second, the vascular
traces are composed entirely of cells with scalariform
to reticulate wall ornamentation, uncharacteristic of
the more mature tissue one would expect to encounter
in branches (Esau, 1977). Third, small branch traces
are also present in the Araucariopitys specimens; they
are more than 2 mm in diameter (an order of magni-
tude larger than the leaf traces), and exhibit a concen-
tric outer layer of secondary xylem, a feature not seen
in leaf traces.
The Alexander Island conifers grew in a polar
environment. In this setting, the onset of the cool,
dark winter would be the most likely trigger of cam-
bial dormancy and growth ring formation, and indi-
vidual ring increments in the woods therefore probably
44 H.J.FALCON-LANG &D.J.CANTRILL
Figure 6. Leaf traces. (a) Longitudinal section through mod-
ern specimen of branch of Araucaria araucana (19 mm diam-
eter) showing three persistent leaf traces. The traces, which
pass through 11 growth ring increments, were attached to liv-
ing leaves when the specimen was collected; they indicate that
leaf longevity was > 11 years in this specimen. (b) Leaf trace
departing through juvenile stem of Araucariopitys, transverse
section. Arrows mark ring boundaries intersected by leaf
trace, mag. ×25 , KG. 4740.11. (c) Cross-section of circular
leaf trace in latewood of Podocarpoxylon sp. 1. Arrow marks
strand of secondary xylem on the underside of the trace,
oblique tangential section, mag. ×25, KG. 4719.4. (d) Cross-
section of oval leaf trace in Araucarioxylon, tangential
section, mag. ×100, KG. 4702.4.
represent an annual growth period (Francis, 1986).
Furthermore, in the closest modern analogues to the
vegetation of the Triton Point Formation, the cool
temperate podocarp forest of South Island, New
Zealand, conifers usually only make one growth ring
per annum (Franklin, 1969). Assuming that growth
rings were formed annually, the fossil leaf trace data
indicate that Araucariopitys and Podocarpoxylon sp. 1
were both evergreen with long leaf retention times, at
least five years in Podocarpoxylon sp. 1 and at least 13
years in the case of Araucariopitys. However, there is
limited evidence that the leaf traces of present-day
araucarian conifers may continue to extend in length
even after leaf fall (Eames & McDaniels, 1947); if this
is confirmed, then the estimates of leaf longevity for
the Araucariopitys plant may be too high.
2.c. Leaf physiognomy
2.c.1. Method
In the third technique, the form and structure of fossil
leaves (leaf physiognomy) are examined for clues that
may suggest leaf habit. In angiosperms, evergreen
leaves are usually thicker with a smaller surface area
than their deciduous counterparts and possess a thick
cuticle, sunken stomata and an array of sclerophyllic
characteristics such as spines or hairs to avoid winter
herbivore predation (Chabot & Hicks, 1982; Wolfe
& Upchurch, 1987; McElwain & Chaloner, 1996).
However, interpretation of the leaf physiognomy of
non-angiosperms is more difficult and has only been
attempted to a limited degree (Chaloner & Creber,
1990). For example, Pigg & Taylor (1993) interpreted
the presence of an abscission scar on the leaf petiole of
a glossopterid to indicate a deciduous habit.
2.c.2. Data and interpretation
The coniferous foliage of Alexander Island can be bro-
ken down into three classes; there are (1) large broad
leaves, (2) small appressed scale leaves, and (3) linear
dorsiventrally flattened leaves that are often twisted to
lie in one plane (Cantrill & Falcon-Lang, unpub. data).
The large, broad leaves are either broadly attached at
the base and strongly imbricate (e.g. Araucaria; Fig. 7a),
or are constricted basally and almost petiolate (e.g.
Podozamites; Fig. 7b). The leaves of Araucaria alexan-
derensis Cantrill show evidence of seasonal variation
in leaf size; a transition from large leaves (40 mm long)
to small leaves (7 mm long) and then back to larger
leaves (40 mm) again occurs in KG. 4745.15. This type
of growth pattern is seen in many extant conifers
including araucarian species such as A. bidwillii Hook.
(Stockey, 1982, plate III, 7) and Wollemia nobilis W. G.
Jones et al. (Chambers, Drinnan & McLoughlin, 1998,
fig. 1) where the variation in size relates to the growing
season. The fluctuation in leaf size seen in KG.
Cretaceous leaf phenology 45
Figure 7. (a) Broad-leafed Araucaria alexanderensis
Cantrill, mag. ×1, KG. 2815.165, and (b) broad-leafed
Podozamites binatus Cantrill with scaly taxodiaceous conifer
leafy shoot (Brachyphyllum) in top-left corner, mag. ×1,
KG. 1704.5a.
4745.15 indicates that this foliage is evergreen and
represents two seasons of growth.
Podozamites leaves are always found attached to
stems and never as isolated leaves (Fig. 7b). Stems
range from those with two to three leaves to larger
stems with up to 20 leaves. This suggests that the
foliage preserved does not represent seasonally deter-
minate shoots that abscised in the autumn, but rather
portions of the plant that have been broken off.
Further support for this interpretation is seen in the
stem bases that lack an abscission zone or expanded
base, a feature often seen in deciduous shoots. In
addition, deciduous foliage units typically have small
leaves at the base that grade into mature leaves more
distally. All the leaves seen in Podozamites are of a
uniform size. Thus the combined evidence supports an
evergreen interpretation.
Both these broad foliage types occur in the upper
part of the Triton Point Formation associated with
the woody stems, Araucariopitys and Podocarpoxylon.
Lacey (1953) noted that the cross-sectional area of a
leaf trace shows a positive correlation with leaf area.
For example Agathis australis possesses leaf traces
with a large cross-sectional area (0.134 mm2) and pos-
sesses large leaves of dimensions 7 ×4 cm (Lacey,
1953). The Alexander Island woody conifer stems
Araucariopitys and Podocarpoxylon sp. 1, which pos-
sess leaf traces of a similar size to Agathis australis,
therefore may have borne equally large broad leaves.
On this basis, the two Alexander Island foliage types
attributed to Araucaria and Podozamites (maximum
leaf dimensions: 4 ×1.2 cm and 8.2 ×1.1 cm respec-
tively) are the most likely taxa borne by these two
wood genera (Cantrill & Falcon-Lang, unpub. data).
The small appressed scale-like leaves are represented
by the foliage of Athrotaxites titanensis Cantrill and
Brachyphyllum (Fig. 7b). It is suggested that most of
this foliage is also evergreen because foliage units are
not determinate and do not appear to be shed as
foliage units. However, there is limited evidence that
one species of Brachyphyllum may have been decidu-
ous; these shoots are of similar dimensions suggesting
a initially determinate habit, and that the plant may
have shed foliage units like extant deciduous taxodia-
ceous conifers. These Brachyphyllum foliage units
occur in the braided river systems at a similar strati-
graphic level to the Taxodioxylon wood, which may
also have been derived from a deciduous taxodiaceous
conifer (cf. Section 2.a.2).
The linear lanceolate leaves in the Alexander Island
foliage assemblage possess a single midvein and occur
attached in a helical fashion with the leaves oriented
into two planes (e.g. Elatocladus). Comparison with
the foliage of the deciduous conifers Taxodium Rich
and Metasequoia Hu & W.C. Cheng, allows the leaf
habit of Elatocladus to be interpreted. In these two
extant genera a resting bud surrounded by scale leaves
overwinters, and in the spring produces a determinate
shoot with linear lanceolate, helically inserted leaves.
In the autumn these foliage units are shed, except
where the long shoots become indeterminate; resting
buds form in the leaf axils of this latter shoot and the
leaves are shed. Consequently deciduous conifers have
a distinctive morphology when compared to conifers
with longer leaf retention times. The foliage shoots are
characterized by distinct basal buds and scale leaves.
Non-deciduous forms lack these features. The linear
lanceolate foliage seen in the Alexander Island mater-
ial lacks this morphology and therefore is considered
to be evergreen also.
Other foliage of linear lanceolate leaves such as
Podocarpites wileyii Cantrill possesses resting buds
with scale leaves and leafy shoots with longer leaves.
However, these are rarely found as isolated foliage
units but rather as isolated branches with short lateral
shoots. This suggests that these small shrubby
podocarps were not deciduous but produced resting
buds that presumably overwintered prior to leaf and
shoot expansion in the spring (Cantrill & Falcon-
Lang, unpub. data).
Finally, specimens of Ginkgoites Seward are charac-
terized by the long shoot–short shoot syndrome seen
in fossil Ginkgo (Zhou & Zhang, 1989) and other
extant (e.g. Larix Mill., Malus) and fossil (e.g.
Glossopteris Brongniart and Nilsoniocladus Kimura &
Sekido) deciduous plants. On this basis Ginkgoites was
almost certainly deciduous. A similar shoot morphol-
ogy is observed in Taeniopteris Brongniart and it is
also considered to be deciduous.
2.d. Nearest Living Relatives
2.d.1. Method
In the fourth method, the Nearest Living Relative
technique, the assumption is made that related mod-
ern and fossil groups have similar leaf longevity.
Clearly, the closer the taxonomic relationship between
the two groups in question, the more accurate the
interpretation of leaf life span possible. This method
becomes increasingly difficult to apply for pre-Tertiary
floras where only family-level taxonomic affinity may
be assumed (Chaloner & Creber, 1990). In addition,
axiomatic to the method is that the ecological and cli-
matological requirements of taxa were constant
throughout geological time and were not affected by
evolutionary changes; this assumption is questionable
for floras of Cretaceous age (A. B. Herman, pers.
comm., 2000)
2.d.2. Data and interpretation
Three conifer families, the Araucariaceae, Podocar-
paceae and Taxodiaceae, occur in the Alexander
Island forest assemblage together with ginkgopsids
(Falcon-Lang & Cantrill, 2000; Cantrill & Falcon-
46 H.J.FALCON-LANG &D.J.CANTRILL
Lang, unpub. data). Today, all members of the
Podocarpaceae and Araucariaceae are evergreen,
usually with long leaf retention times. For example,
amongst the Araucariaceae, Araucaria araucana grow-
ing in Chile retains leaves for 15 years (Vidakovic,
1982) whilst the podocarp conifers Halocarpus
biformis (Hook) Quinn and Halocarpus bidwillii
(Hook. F. ex Kirk) Quinn growing in New Zealand
may hold leaves for 10 and 20 years respectively
(Wardle, 1963; Odgen & Stewart, 1995). Alexander
Island Araucariaceae and Podocarpaceae may there-
fore also have been evergreen with long leaf retention
times. In contrast, present-day Taxodiaceae possess
the greatest number of deciduous genera of any
conifer family; of the eight extant genera, Taxodium,
Glyptostrobus Endl. and Metasequoia are all decidu-
ous (Vidakovic, 1982), whilst the remaining genera are
evergreen with short leaf retention times. The only tax-
odiaceous foliage found on Alexander Island was
assigned to Athrotaxites, a form genus that bears some
similarity to the extant evergreen conifer genus
Athrotaxis D. Don, endemic to Tasmania (Cantrill &
Falcon-Lang, unpub. data). The interpretation of
Athrotaxites as evergreen seems plausible given its
shoot morphology discussed above. Finally, the only
living relative of the foliage genus Ginkgoites is Ginkgo
biloba, which is deciduous.
2.e. Leaf taphonomy
2.e.1. Method
The fifth technique relies on the interpretation of the
transport and depositional history of plant parts
(plant taphonomy), based on the distribution of fossil
plant remains in sedimentary facies. One might expect
that deciduous leaves which are synchronously shed at
a particular time in the year would tend to accumulate
in discrete layers (leaf mats), whilst evergreen leaves
shed throughout the year would occur randomly
within a sequence (Spicer & Parrish, 1986). This idea
has been developed to good effect by Retallack (1980),
who correlated the occurrence of glossopterid leaves
with the autumn–winter component of Permian lacus-
trine varved units, thus demonstrating autumnal
abscission and a deciduous habit. Interpretation of
leaf habit is more problematic, however, where leaves
occur in sequences with no intra-annual temporal
framework such as floodbasin mudstone units. For
example, Parrish et al. (1998) recorded discrete leaf
mats of Agathis Salsib. from the Cretaceous of New
Zealand; all leaves were well preserved which sug-
gested to the authors that they were synchronously
shed in the autumn by a deciduous plant. However,
similar well-preserved leaf mats of Cordaites Unger
are found in the Upper Carboniferous of Nova Scotia
(Falcon-Lang, 1999b; Falcon-Lang & Scott, 2000); it
is known that cordaitaleans were evergreen so these
leaf accumulations must merely represent the product
of random sedimentary processes. As a consequence it
is important not to place too much emphasis on
taphonomic data.
2.e.2. Data and interpretation
Only a few observations pertinent to this paper can be
made with regard to the taphonomy of the Alexander
Island floras. First, Ginkgoites leaves, which are wide-
spread within the sequence, often form prominent leaf
mats. These leaf mats are characterized by leaves of a
uniform size class and preservational state suggesting
they were synchronously shed. Taeniopteris also forms
similar leaf mats at a number of localities (e.g. KG.
4737). These data indicate that both these taxa may
have been deciduous.
3. Discussion
Data acquired using the five independent techniques
above strongly suggest that the canopy-forming vege-
tation of Alexander Island that included conifers,
ginkgos and taeniopterids was dominantly evergreen.
The araucarian and podocarp conifers that composed
more than 90 % of the forest canopy were evergreen
with leaf retention times of at least 5–13 years. Some
rare taxodiaceous conifers were also evergreen but
with much shorter leaf retention times; other taxodia-
ceous conifers may have been deciduous as were all the
ginkgos and taeniopterids. The understorey vegetation,
dominated by ferns, was probably mainly deciduous.
For example, pteridophytes such as Phyllopteroides
Medwell (Osmundaceae) frequently occur as isolated
pinnules; this may imply that this fern died down
during the winter months as do extant Osmunda L.
Some of the understorey angiosperms may have been
evergreen, however (Cantrill & Nichols, 1996).
3.a. Southern Hemisphere polar vegetation
Forest vegetation is also known to have grown in the
Southern Hemisphere polar region at two other sites
during the mid-Cretaceous (Fig. 8a). In New Zealand,
late Albian–Cenomanian vegetation grew at a latitude
of 70° S. One of the dominant elements of this flora
was the foliage of Agathis, a canopy-forming araucar-
ian conifer. On the basis of cuticle thickness and
taphonomic data, Parrish et al. (1998) interpreted one
Agathis species as being deciduous whilst another was
assigned evergreen status. Other plants present were
liverworts, ferns, sphenopsids, taeniopterids, cycads,
bennettites, pentoxylaleans, ginkgophytes, podocarp
conifers and angiosperms which probably included a
mixture of evergreen and deciduous plants (Stopes,
1914, 1916; Parrish et al. 1998).
In southeastern Australia, abundant Albian fossil
floras have been studied in the Otway Basin; these
Cretaceous leaf phenology 47
represent vegetation which grew at a latitude of 66° S.
Floras are dominated by the wood and foliage of ever-
green araucarian and podocarp conifers which would
have formed the forest canopy; taxodiaceous conifers
which may also have been evergreen occur rarely
(Cantrill & Webb, 1987; Cantrill & Douglas, 1988;
Cantrill, 1991, 1992; Dettman et al. 1992). The under-
storey comprised a mixed evergreen–deciduous vege-
tation which included cycads, gymnosperms, ferns,
ginkgophytes and angiosperms (Douglas & Williams,
1982).
In summary, the Southern polar forests of
Alexander Island and southeastern Australia appear
to have been composed of an almost exclusively ever-
green coniferous canopy with a mixed evergreen–
deciduous understorey, but New Zealand forests from
a similar latitude appear to have contained a larger
deciduous component.
3.b. Northern Hemisphere polar vegetation
Vegetation patterns have also been reconstructed for
several mid- to Late Cretaceous sites in the northern
polar region (Fig. 8b). Spicer & Parrish (1986)
described a number of floras from the Albian–
Cenomanian Nanushuk Group of northern Alaska
which grew at a latitude of more than 75° N. These
were largely composed of deciduous plants including
angiosperms, ginkgophytes, ferns, vine-like cycads and
conifers (Axelrod, 1984; Spicer & Parrish, 1990; Spicer
& Herman, 1996). The dominant conifers were proba-
bly deciduous because they possessed foliage morpho-
types comparable to that of modern taxodiaceous
shoot-dropping species. Some other conifers such as
the broad-leafed Podozamites which grew in Alaskan
peat mire environments have also been assigned decid-
uous status but this interpretation is based on equivo-
cal taphonomic grounds; microphyllous conifers such
as Brachyphyllum were almost certainly evergreen
(Spicer & Parrish, 1986). In the same vicinity, Coniacian
and Campanian–Maastrichtian plant assemblages
have been described from the Colville Group and were
also dominantly deciduous being composed of
angiosperms, ginkgophytes, ferns and sphenopsids.
Some of these conifers, particularly the cuppressa-
ceous forms, were probably evergreen whilst other
more dominant taxodiaceous forms were deciduous
(Parrish & Spicer, 1988).
Finally, mid- to Late Cretaceous forest vegetation is
also known from many sites in northern and north-
eastern Russia where it flourished at latitudes of
65–82° N (Spicer & Herman, 1998). These forests were
composed of deciduous caytonealeans, ginkgos,
czekanowskialeans, vine-like cycads, and many
conifers and angiosperms. Some of the conifers were
of the taxodiaceous shoot-dropping species, but many
others were of araucarian or cuppressacean affinity,
and possessed evergreen scale-like leaves (Herman,
1999; Kelley, Spicer & Herman, 1999). Considered in
totality, the Northern Hemisphere polar forests of
Alaska and northern Russia appear to have contained
a much larger deciduous component than the
Southern forests of Antarctica, Australia and New
Zealand (Wolfe & Upchurch, 1987). This conclusion
conflicts with early studies which suggested that both
Northern and Southern Hemisphere polar forests were
equally deciduous (Spicer, Rees & Chapman, 1993).
48 H.J.FALCON-LANG &D.J.CANTRILL
Figure 8. Mid-Cretaceous polar-centred geographies show-
ing the position of fossil forests mentioned in the text. (a)
Southern Hemisphere, late Albian stage, 105 Ma, and (b)
Northern Hemisphere, Albian–Cenomanian stages, 95 Ma
(Palaeo-coastlines from Smith, Smith & Funnel, 1994, and
R. Livermore, BAS). Light grey – deep ocean; medium grey
– continental marine shelf; dark grey – land.
3.c. Mid-Cretaceous polar environments
Why did Northern Hemisphere polar forests contain a
much larger number of deciduous trees than the
Southern Hemisphere forests? Growing within the
polar circles (palaeolatitude >66°), all these floras
would have had to adapt to conditions of extreme light
seasonality including a protracted period of perma-
nent darkness during the winter months (Read &
Francis, 1992). The employment of different ecological
strategies in these two coeval polar communities sug-
gests that environmental conditions differed between
the Northern and Southern Hemisphere poles. In this
section we explore the nature of the Cretaceous polar
environments and attempt to relate these environmental
conditions to the observed patterns of leaf phenology.
In the Northern Hemisphere, the Alaskan and
Russian polar forests both grew on maritime peninsu-
las almost entirely surrounded by ocean or on islands.
Multivariate analyses of foliar physiognomic data
from these regions suggest that the cold month mean
temperature never fell very low, being in the region of
+6.2 °C at 75° N in Alaska, –0.6 °C at 82° N in north-
ern Russia, and +5.5°C at 70°N in northeast Russia
during the late Albian–Coniacian stages, and that
Arctic Ocean temperature was warm year-round,
never falling much below 5.7 °C (Herman & Spicer,
1996; Spicer & Herman, 1998). These figures compare
reasonably well with the results of climate computer
models run by Valdes, Sellwood & Price (1996).
Although dropstone evidence suggests that seasonal
sea-ice may have formed adjacent to continental high
northern latitude regions such as Spitsbergen in the
Albian (Frakes & Francis, 1988; Francis & Frakes,
1993), there is no widespread evidence for sea-ice in
the Arctic Circle; permanent ice was probably only
present at high altitude in Alaska and Russia (Spicer &
Parrish, 1990). Collectively these data suggest a mild,
maritime climate during the Arctic winter months.
In contrast, the polar floras of the Southern
Hemisphere grew along the palaeo-Pacific coast of the
Gondwanan supercontinent in New Zealand and the
Antarctic Peninsula, and in a more continental setting
in southeast Australia (Fig. 8a). Numerical computer
climate models predict cold month mean temperatures
to have been around +4°C for maritime New Zealand
but much lower (down to –8 °C) for Antarctica and
southeast Australia (Valdes, Sellwood & Price, 1996).
More primitive terrestrial climate models have pre-
dicted that Australia–Antarctica would have experi-
enced cold month mean temperatures as low as –18 °C
(Barron & Washington, 1984). Freezing winter condi-
tions are implied by the presence of Aptian ikaite
pseudomorphs (glendonites nodules) in central
Australia which develop under cold bottom water con-
ditions (Dettman et al. 1992; Francis & Frakes, 1993),
and Aptian–Albian dropstones in northern Australia
suggest that seasonal sea-ice periodically extended
as far north as 55° S (Frakes & Krassay, 1992). The
presence of a small permanent Early Cretaceous
Antarctic ice sheet has also been inferred by Stoll &
Schrag (1996) to explain rapid sea-level changes
during this interval. All these data suggest that
Southern Hemisphere polar vegetation in Australia
and Antarctica grew under a climate characterized
by cold, dark winter conditions whilst New Zealand
floras existed in a maritime setting where winter
temperatures were milder.
3.d. Mid-Cretaceous polar leaf phenology
As noted in the introduction, the leaf phenology of
plants is ‘finely tuned’ to the growing environment in
order to achieve the highest possible annual carbon
gain. The largest problem facing plants growing well
within the polar circle relates to the high levels of dark
respiration incurred during the winter darkness. This
may not have been such a severe problem during the
Cretaceous greenhouse phase as it would be today,
because it is known that whole-canopy dark respira-
tion is reduced under conditions of elevated carbon
dioxide (Barker et al. 2000). Nevertheless, physiologi-
cal adaptation to prolonged periods of dark respira-
tion may have been one of the main factors influencing
leaf phenology.
For example, in polar regions with warm dark
winters like those of the Cretaceous Northern
Hemisphere (e.g. +6°C cold month mean tempera-
ture), a deciduous habit would probably be favoured
because the presence of winter leaves would permit
high rates of dark respiration and quickly lead to etio-
lation of seedlings (Spicer & Corfield, 1992). Creber &
Chaloner (1985) have suggested that some evergreen
plants may have been able to survive in such warm
winter dark conditions by altering their physiology
so that optimum respiration temperature was set a few
degrees higher than ambient temperature. Another
option would be to evolve scale-like, xeromorphic
foliage which could effectively close stomata and mini-
mize dark respiration carbon-loss; this strategy
appears to have been utilized by some araucarian and
cuppressacean conifers from the northern Russian
sites (Herman, 1999).
In contrast, in polar regions with cold dark winters
like those of mid-Cretaceous Australia and Antarctica
(perhaps as low as –8°C), an evergreen habit would
probably have been favoured. Here, temperatures
would be sufficiently low to suppress metabolic activ-
ity during the winter darkness, therefore minimizing
respiration losses (Gower & Richards, 1990). Under
such circumstances, an evergreen habit would be
advantageous as it would permit photosynthesis to
begin as soon as light and temperature levels exceeded
the necessary threshold, whilst at the same time reduc-
ing the carbon-expenditure in making a new leaf
canopy (Chabot & Hicks, 1982). In contrast New
Cretaceous leaf phenology 49
Zealand floras existed under a milder maritime cli-
mate, and as in the Northern Hemisphere, plants with
a deciduous habit were more competitive.
This thesis for explaining mid-Cretaceous phenologi-
cal patterns is supported by the work of Read & Francis
(1992). They examined the response of some seedlings
to ten weeks of permanent darkness at both 4°C and
15 °C. In general evergreen conifer species such as
Dacrycarpus cinctus (Pilg.) de Laub., Lagarostrobus
franklinii (Hook.) Quinn, Microstrobus niphophilus J.
Garden & L.A.S. Johnson and Phyllocladus aspleni-
ifolius Hook. F. survived 70 days of darkness without
long-term damage at the lower temperature but were
more severely affected at the higher temperature. In
contrast, deciduous taxa such as Ginkgo biloba L. and
Nothofagus antarctica were apparently unaffected by
either the warm or cold dark treatment.
4. Conclusions
(1) A systematic study of the leaf phenology of a fossil
forest is presented for the first time. Five independent
lines of evidence are drawn upon to strongly suggest
that the mid-Cretaceous (late Albian) forest vegetation
of Alexander Island consisted of a dominantly ever-
green canopy with a mixed evergreen–deciduous
understorey.
(2) Preliminary analysis of all known polar forests
implies the existence of a rather complex ecophysio-
logical mosaic during mid- to Late Cretaceous times;
Antarctic and Australian forests were dominantly
evergreen whilst Alaskan, Russian and New Zealand
forests possessed a rather larger deciduous component.
(3) These two ecological strategies represent
responses to different polar environments. In Alaska,
Russia and New Zealand, forests experienced mild
dark winter climates; these conditions favoured a
deciduous strategy which acted to minimize dark-
respiration carbon loss. In Antarctica and Australia,
forests experienced cold, dark winter conditions; these
forests developed an evergreen strategy to efficiently
exploit the short summer growing season whilst low
winter temperatures thermally depressed metabolic
rates and minimized dark-respiration losses.
Acknowledgements. This paper was prepared as part of a
post-doctoral project examining Cretaceous polar forests
and was funded by the British Antarctic Survey (BAS).
Chris Gilbert and Pete Bucktrout (BAS) are thanked for
preparing the photographs. Discussion with Chris Page
improved our grasp of conifer ecology, whilst thorough and
thoughtful reviews by W. G. Chaloner and A. B. Herman
greatly tightened up our arguments.
References
AXELROD, D. I. 1984. An interpretion of Cretaceous and
Tertiary biota in polar regions. Palaeogeography,
Palaeoclimatology, Palaeoecology 45, 105–47.
BARKER,J.T.,ALLEN,L.H.,BOOTE,K.J.& PICKERING,N.B.
2000. Direct effects of atmospheric carbon dioxide
concentration on whole canopy dark respiration.
Global Change Biology 6, 275–86.
BARRON,E.J.& WASHINGTON, W. M. 1984. The role of geo-
graphic variables in explaining palaeoclimates: results
from Cretaceous climate model sensitivity studies.
Journal of Geophysical Research 89, 1267–79.
CANTRILL, D. J. 1991. Broad leafed coniferous foliage from
the Lower Cretaceous Otway Group, southeastern
Australia. Alcheringa 15, 177–90.
CANTRILL, D. J. 1992. Araucarian foliage from the Lower
Cretaceous of southern Victorian, Australia.
International Journal of Plant Sciences 153, 622–45.
CANTRILL, D. J. 1995a. The occurrence of the fern
Hausmannia Dunker (Dipteridaceae) in the Cretaceous
of Alexander Island, Antarctica. Alcheringa 19, 243–54.
CANTRILL, D. J. 1995b. Fern thickets from the Cretaceous
of Alexander Island, Antarctica containing Alamatus
bifarius Douglas and Aculea acicularis sp. nov.
Cretaceous Research 17, 169–82.
CANTRILL,D.J.& DOUGLAS, J. G. 1988. Mycorrhizal conifer
roots from the Lower Cretaceous of the Otway Basin,
Victoria. Australian Journal of Botany 36, 257–72.
CANTRILL,D.J.& NICHOLS, G. J. 1996. Taxonomy and
palaeoecology of Early Cretaceous (Late Albian)
angiosperm leaves from Alexander Island, Antarctica.
Review of Palaeobotany and Palynology 92, 1–28.
CANTRILL,D.J.& WEBB, J. A. 1987. A reappraisal of
Phyllopteroides Medwell (Osmundaceae) and its strati-
graphic significance in the Lower Cretaceous of eastern
Australia. Alcheringa 11, 59–85.
CHABOT,B.F.& HICKS, D. J. 1982. The ecology of leaf life
spans. Annual Review of Ecology and Systematics 13,
229–59.
CHALONER,W.G.& CREBER, G. T. 1989. The phenomenon
of forest growth in Antarctica: a review. In Origins
and evolution of the Antarctic biota (ed. J. A. Crame),
pp. 85–8. Geological Society of London Special
Publication no. 47.
CHALONER,W.G.& CREBER, G. T. 1990. Do fossil plants
give a climatic signal? Journal of the Geological Society,
London 147, 343–50.
CHAMBERS,T.C.,DRINNAN,A.N.& MCLOUGHLIN, S. 1998.
Some morphological features of Wollemi Pine
(Wollemia nobilis: Araucariaceae) and their comparison
to Cretaceous plant fossils. International Journal of
Plant Sciences 159, 160–71.
CLARKE,L.J.& JENKYNS, H. C. 1999. New oxygen isotope
evidence for long-term Cretaceous climatic change in
the Southern Hemisphere. Geology 27, 699–702.
CREBER, G. T. 1975. The effects of gravity and the Earth’s
rotation on the growth of wood. In Growth Rhythms
and the History of the Earth’s Rotation (eds G. D.
Rosenberg and S. K. Runcorn), pp. 75–87. London:
Wiley-Interscience.
CREBER,G.T.& CHALONER, W. G. 1984. Influence of envi-
ronmental factors on the wood structure of living and
fossil trees. Botanical Review 50, 357–448.
CREBER,G.T.& CHALONER, W. G. 1985. Tree growth in the
Mesozoic and early Tertiary and the reconstruction of
palaeoclimates. Palaeogeography, Palaeoclimatology,
Palaeoecology 52, 35–50.
DETTMAN, M. E., MOLNAR, R. E., DOUGLAS,J.G.,BURGER,
D., FIELDING, C., CLIFFORD,H.T.,FRANCIS, J., JELL,
P. , R ICH, T., WADE, M., RICH,P.V.,PLEDGE, N., KEMP,
A. & ROZEFELDS, A. 1992. Australian Cretaceous ter-
50 H.J.FALCON-LANG &D.J.CANTRILL
restrial faunas and floras: biostratigraphic and biogeo-
graphic implications. Cretaceous Research 13, 207–62.
DOUGLAS,J.G.& WILLIAMS, G. E. 1982. Southern polar
forests: The Early Cretaceous floras of Victoria and
their palaeoclimatic significance. Palaeogeography,
Palaeoclimatology, Palaeoecology 39, 171–85.
EAMES,A.J.& MCDANIELS, L. H. 1947. An Introduction
to Plant Anatomy, 2nd ed. New York: McGraw-Hill,
427 pp.
ESAU, K. 1977. The Anatomy of Seed Plants, 2nd ed. New
York: John Wiley and Sons, 550 pp
FALCON-LANG, H. J. 1999a. The Early Carboniferous
(Courceyan-Arundian) monsoonal climate of the
British Isles: evidence from growth rings in fossil woods.
Geological Magazine 136, 177–87.
FALCON-LANG, H. J. 1999b. Fire ecology of a Late
Carboniferous floodplain, Joggins, Nova Scotia.
Journal of the Geological Society, London 156, 137–48.
FALCON-LANG, H. J. 2000a. A method to distinguish
between woods produced by evergreen and deciduous
coniferopsids on the basis of growth ring anatomy: a
new palaeoecological tool. Palaeontology 43, 785–93.
FALCON-LANG, H. J. 2000b. The relationship between leaf
longevity and growth ring markedness in modern
conifer woods and its implications for palaeo-
climatic studies. Palaeogeography, Palaeoclimatology,
Palaeoecology 160, 317–28.
FALCON-LANG,H.J.& CANTRILL, D. J. 2000. Cretaceous
(Late Albian) Coniferales of Alexander Island,
Antarctica. 1: Wood taxonomy: a quantitative approach.
Review of Palaeobotany and Palynology 111, 1–17.
FALCON-LANG,H.J.& SCOTT, A. C. 2000. Upland ecology of
some Late Carboniferous cordaitalean trees from Nova
Scotia and England. Palaeogeography, Palaeoclimatology,
Palaeoecology 156, 225–42.
FRAKES,L.A.& FRANCIS, J. E. 1988. A guide to
Phanerozoic cold polar climates from high latitude
ice-rafting in the Cretaceous. Nature 333, 547–9.
FRAKES,L.A.& KRASSAY, A. A. 1992. Discovery of proba-
ble ice-rafting in the Late Mesozic of the Northern
Territory and Queensland. Australian Journal of Earth
Sciences 39, 115–19.
FRANCIS, J. E. 1986. Growth rings in Cretaceous and
Tertiary wood from Antarctica and their palaeoclimatic
implications. Palaeontology 29, 665–84.
FRANCIS,J.E.& FRAKES, L. A. 1993. Cretaceous climates.
Sedimentology Review 1, 17–30.
FRANKLIN, D. A. 1969. Growth rings in rimu from South
Westland terrace forest. New Zealand Journal of Botany
7, 177–88.
GALTIER,J.& SCOTT, A. C. 1994. Stanwoodia, a new genus of
probable early gymnosperms from the Dinantian of
East Kirkton, Scotland. Transactions of the Royal
Society of Edinburgh: Earth Sciences 82, 113–23.
GORDON, W. T. 1935. The genus Pitys Witham emend.
Transactions of the Royal Society of Edinburgh 58,
279–311.
GOWER ,S.T.& RICHARDS, J. H. 1990. Larches: deciduous
conifers in an evergreen world. BioScience 40, 819–26.
HERMAN, A. B. 1999. Composition and age of the Grebenka
flora from the Anadyr’ River area (the Middle
Cretaceous, Northeastern Russia). Stratigraphy and
Geological Correlation 7, 265–78.
HERMAN,A.B.& SPICER, R. A. 1996. Palaeobotanical evi-
dence for a warm Cretaceous Arctic Ocean. Nature 380,
330–3.
JEFFERSON, T. H. 1982. Fossil forests from the Lower
Cretaceous of Alexander Island, Antarctica.
Palaeontology 25, 681–708.
KELLEY, S. R., SPICER,R.A.& HERMAN, A. B. 1999. New
40Ar/39Ar dates for Cretaceous Chauna Group tephra,
north-eastern Russia, and their implications for the
geologic history and floral evolution of the North
Pacific region. Cretaceous Research 20, 97–106.
LACEY, W. S. 1953. Scottish Lower Carboniferous plants:
Eristophyton waltoni sp. nov. and Endoxylon zonatum
(Kidston) Scott from Dunbartonshire. Annals of
Botany N.S. 17, 579–96.
MCELWAIN,J.C.& CHALONER, W. G. 1996. The fossil cuticle
as a record of environmental change. Palaios 11, 376–88.
MONCRIEFF,A.C.M.& KELLY, S. R. A. 1993. Litho-
stratigraphy of the uppermost Fossil Bluff Group (Early
Cretaceous) of Alexander Island, Antarctica: history of
an Albian regression. Cretaceous Research 14, 1–15.
ODGEN,J.& STEWART, G. H. 1995. Community dynamics of
New Zealand conifers. In Ecology of the Southern
Conifers (eds N. J. Enright and R. S. Hill), pp. 81–119.
Australia: Melbourne University Press.
OTTO-BLIESNER,B.L.& UPCHURCH, G. R. 1997.
Vegetation-induced warming of high-latitude regions
during the Late Cretaceous period. Nature 385, 804–7.
PARRISH,J.T.,DANIEL,I.L.,KENNEDY,E.M.& SPICER,R.A.
1998. Palaeoclimatic significance of mid-Cretaceous
floras from the Middle Clarence Valley, New Zealand.
Palaios 13, 149–59.
PARRISH,J.T.& SPICER, R. A. 1988. Late Cretaceous terres-
trial vegetation: a near-polar temperature curve. Geology
16, 22–5.
PARRISH,J.T.,ZIEGLER,A.M.& SCOTESE, C. R. 1982.
Rainfall patterns and the distribution of coals and evap-
orites in the Mesozoic and Cenozoic. Palaeogeography,
Palaeoclimatology, Palaeoecology 40, 67–101.
PIGG,K.& TAY L O R , T. N. 1993. Anatomically preserved
Glossopteris stems with attached leaves from the central
Transantarctic Mountains, Antarctica. American Journal
of Botany 80, 500–16.
READ,J.& FRANCIS, J. E. 1992. Responses of some southern
hemisphere tree species to a prolonged dark period
and their implications for high-latitude Cretaceous and
Tertiary floras. Palaeogeography, Palaeoclimatology,
Palaeoecology 99, 271–90.
RETALLACK, G. J. 1980. Late Carboniferous to middle
Triassic megafossil floras from the Sydney Basin.
Bulletin of the Geological Survey of New South Wales
26, 384–430.
SMITH,A.G.,SMITH,D.G.& FUNNEL, B. M. 1994. Atlas
of Mesozoic and Cenozoic coastlines. Cambridge:
Cambridge University Press, 99 pp.
SPICER,R.A.& CORFIELD, R. M. 1992. A review of terres-
trial and marine climates in the Cretaceous with
implications for modelling the ‘Greenhouse Earth’.
Geological Magazine 129, 169–80.
SPICER,R.A.& HERMAN, A. B. 1996. Nilssoniocladus in the
Cretaceous Arctic: new species and biological insights.
Review of Palaeobotany and Palynology 92, 229–43.
SPICER,R.A.& HERMAN, A. B. 1998. Cretaceous climate of
Asia and Alaska: a comparison of palaeobotanical evi-
dence with a climate computer model. Palaeontological
Journal 32, 105–18.
SPICER,R.A.& PARRISH, J. T. 1986. Palaeobotanical evi-
dence for cool north polar climates in middle Cretaceous
(Albian–Cenomanian) time. Geology 14, 703–6.
Cretaceous leaf phenology 51
SPICER,R.A.& PARRISH, J. T. 1990. Late Cretaceous–early
Tertiary palaeoclimates of northern high latitudes: a
quantitative view. Journal of the Geological Society,
London 147, 329–41.
SPICER, R. A., REES,P.M.& CHAPMAN, J. L. 1993.
Cretaceous phytogeography and climate signals.
Philosophical Transactions of the Royal Society, London
Series B 341, 277–86.
STOCKEY, R. A. 1982. The Araucariaceae: An Evolutionary
Perspective. Review of Palaeobotany and Palynology 37,
133–54.
STOLL,H.M.& SCHRAG, D. P. 1996. Evidence for glacial
control of rapid sea level changes in the Early
Cretaceous. Science 272, 1771–4.
STOPES, M. C. 1914. A new Araucarioxylon from New
Zealand. Annals of Botany 28, 341–50.
STOPES, M. C. 1916. An early type of the Abietineae from
the Cretaceous of New Zealand. Annals of Botany 30,
111–25.
UPCHURCH, G. R., OTTO-BLIESNER,B.L.& SCOTESE,C.R.
1999. Terrestrial vegetation and its effects on climate
during the latest Cretaceous. In Evolution of the
Cretaceous Ocean-Climate System (eds E. Barrera and
C. C. Johnson), pp. 407–22. Boulder, Colorado:
Geological Society of America Special Paper no. 332.
VALDES,P.J.,SELLWOOD,B.W.& PRICE, G. D. 1996.
Evaluating concepts of Cretaceous equability.
Palaeoclimates 2, 139–58.
VIDAKOVIC, M. 1982. Conifers: morphology and variation.
Craficki Zavod Hrvarske, Croatia, 755 pp.
WARDLE, P. 1963. Growth habits of New Zealand sub-alpine
shrubs and trees. New Zealand Journal of Botany 1,
18–47.
WOLFE,J.A.& UPCHURCH, G. R. 1987. North American
non-marine climates and vegetation during the Late
Cretaceous. Palaeogeography, Palaeoclimatology,
Palaeoecology 61, 33–77.
WOODWARD, F. I. 1987. Climate and Plant Distribution,
Cambridge University Press, 174 pp.
ZHOU,Z.& ZHANG, B. 1989. A Middle Jurassic Ginkgo
with ovule-bearing organs from Henan, China.
Palaeontographica 211A, 113–33.
52 Cretaceous leaf phenology
