ANTARCTIC CRETACEOUS BIODIVERSITY CHANGE 149
conditions. It could well be that this extreme
thermal and moisture gradient acted as an effec-
tive barrier to migration into the Antarctic
Peninsula region, and that the initial radiation of
angiosperms across Gondwana was not via the
Antarctic Peninsula but by some other route.
Peak of Cretaceous warmth as a forcing
mechanism
It is probably no coincidence that the arrival of
angiosperms and their subsequent rise to domi-
nance in the high southern latitudes correspond
to a global climatic warming trend. The peak of
the Cretaceous greenhouse was in the Turonian
(Huber 1998) and this coincides with the peak of
angiosperm diversity in the Antarctic Peninsula
(Fig. 4E). Certainly if the thermal regime that
existed in the Early Cretaceous was a barrier to
migration, then global warming would have
resulted in latitudinal range expansion as cli-
matic zones widened. This would have pushed
floristic boundaries southwards, and ultimately
have led to angiosperm arrival in the Antarctic
Peninsula region.
Following the peak of Cretaceous warmth in
the Turonian, the high southern latitudes
remained warm until late Santonian to early
Campanian times when cooling started to occur
(Huber 1998). This latest Cretaceous period saw
the origin of a number of groups that today
characterize temperate southern floras (e.g.
Nothofagus, certain groups in the Proteaceae
and Myrtaceae; Dettmann 1989). Furthermore,
events such as the uplift of the Andes (c. 80 Ma
BP) and a general cooling trend through the latest
Cretaceous provided opportunities for these taxa
to expand northwards. As the globe cooled
the tropical biomes contracted equatorwards
creating space for temperate biota to expand. A
good example of this is the fern family Lopho-
soriaceae, a group that arose in the Antarctic
Peninsula regions during the earliest Cretaceous
(Berriasian) (Dettmann 1986a; Cantrill 1998),
appeared later in southern South America
(Aptian-early Albian), and later still in Colombia
(Pleistocene). Although today it occupies an
extreme latitudinal range along the Andean
chain, it is clear that the closer to the tropics it
occurs, then the higher (cooler) the altitudinal
setting (Cantrill 1998). Clearly for this group,
migration is incumbent upon a suitable habitat
either provided by cooling climates or the
creation of high-altitude sites. Similar patterns of
later northward inceptions are seen in the earliest
appearance of Nothofagus in the Antarctic Penin-
sula (early Campanian) and subsequent spread to
southern South America (Maastrichtian).
Summary
Patterns of floristic replacement through the
Cretaceous in the high southern latitudes occur
amongst the understorey and colonizer elements
of the vegetation, suggesting that this was the
niche that the angiosperms initially occupied.
This is a pattern similar to that seen elsewhere in
the world (Lidgard & Crane 1988), and is sup-
ported by the lack of angiosperm wood in these
early deposits, despite the presence of pollen.
Later stage invasion occurred in the overstorey
with the appearance of tree species as attested to
by abundant angiosperm wood in the Late
Cretaceous sequences (Chapman & Smellie
1992; Poole & Francis 1999, 2000; Poole et al
2000a,b,c; Poole & Cantrill 2001; Poole &
Gottwald 2001).
The timing of the radiation into the Antarc-
tica Peninsula implies that this region was not
the initial gateway for invasion of Gondwana by
angiosperms. It seems most likely that the initial
invasion occurred through Africa or India but
further work is needed to refine not only the
plate tectonic reconstructions but also the floris-
tic history of these areas. The macro- and
microfloral record is not well known for south-
ern Africa, and is poorly age-constrained for
India, making it difficult to discriminate between
these hypotheses. Despite the terrestrial con-
nection between South America and the
Antarctic Peninsula, it is suggested that a steep
climatic gradient was probably responsible for
blocking the early radiation through this region.
However, mid-Cretaceous (Turonian) warming
resulted in latitudinal range expansion of cli-
matic belts and appears to have been the forcing
mechanism for the radiation of angiosperms into
the Peninsula.
References
ARCHANGFLSKY, A., ANDREIS, R. R., ARCHANGELSKY,
S. & ARIA BE, A. 1995. Cuticular characters
adapted to volcanic stress in a new Cretaceous
cycad leaf from Patagonia, Argentina. Consider-
ations on the stratigraphy and depositional
history of the Baquero Formation. Review of
Palaeobotany and Palynology, 89, 213-233.
ASKIN, R. A. 1983. Tithonian (uppermost Jurassic) -
Barremian (Lower Cretaceous) spore, pollen and
microplankton from the South Shetland Islands,
Antarctica. In: OLIVER, R. L., JAMES, P. R. &
JAGO, J. B. (eds) Antarctic Earth Science. Aus-
tralian Academy of Sciences, Canberra, 295-297.
ASKIN, R. A. 1990. Cryptogam spores from the upper
Campanian and Maastrichtian of Seymour Island,
Antarctica. Micropaleontology, 36, 141-156.
ASKIN, R. A. 1994. Monosulcate angiosperm pollen
from the Lopez de Bertodano Formation (upper
