Climatic history and forest distribution in eastern Africa - Biogeography and Ecology of the Rain Forests of Eastern Africa - page 24

From the mid-Cretaceous onwards Gond-
wanaland began to drift apart so that by the begin-
ning ofthe Teriary (65 Myr BP) oceanic currents
could flow around Africa. At the end ofthe Creta-
ceous, Africa was about 15-18° south of its
present posiion puting the ITCZ over what is
now the Sahara. Moist air would have been com-
ing from the Tethys Sea to the north and the
newly ormed seas (Atlanic and Indian) on the
westen and easten edge of the continent. Fossil
evidence from the late Cretaceous and early Ter-
iay indicates the presence of moist forest in
present-day North Africa and it has been sug-
gested that a lowland equatorial rain forest
covered the area in the Palaeocene 60 Myr BP
(xelrod & Raven, 1978).
At the ime of the breakup of Gondwanaland
there was rifing taking place in east and southeast
Africa (Sowerbutts, 1972). This suggests that the
progenitors of the Easten Arc and easten rift
mountains were formed about 100 Myr BP. They
were probably in eistence before northward
movement of Africa closed the Tethys sea. This
northward movement limited northern oceanic
currents, reducing rainfall and beginning frag-
mentaion of the pan-African rain forest. From
the mid-Cretaceous to the mid-Teriary there is
no evidence of major tectonic disturbance in
Africa and most of southern and cenral Africa
was eroded in the 'great African planaion' (King,
1978). Progenitors of the Eastern Arc mountains
would have been eroded during this period, and it
is unlikely that they were major mountains prior to
the mid-Teriary uplift.
For 30 million years following the development
of the pan-African rain forest, Africa moved
northwards. The Sahara sill possessed forest or
woodland in the Eocene about 40 Myr BP
(Aubreville, 1970) but aridity was spreading dur-
ing the Oligocene. By the ime Africa completed
its union with Eurasia in the mid-Miocene 17-18
Myr BP, the pan-African forest would have begun
to be fragmented into a westen area under the
Atlanic Ocean climate and an easten area under
the Indian Ocean climate. The Easten Arc
mountains would have been attracing orographic
precipitaion with a corresponding rain shadow to
their west, though its extent would have depended

on how high they were at this ime following the
long period of erosion.
Closure of the Tethys sea stopped -the flow of
equatorial oceanic currents through it, and the
northern high pressure areas of the ITCZ which
brought rain into northen Africa during the early
Teriary were now largely over land and so carried
less moisture into the equatorial region. Lower
Miocene fossil evidence (24 Myr BP) rom
Bugishu near Mount Elgon (Chaney, 1933) has
an abundance of small leaves suggesing a
seasonal climate supporing dry woodland, as do
the Ethiopian sites of the same period (Axelrod &
Raven, 1978). The mid-Miocene flora of
Rusinga Island in Lake Victoria (Chesters, 1957)
has been interpreted as riverine thicket in savanna
woodland (Axelrod & Raven, 1978; Kortlandt,
1983), and a comparison of mammal remains in
five East African fossil sites (Andrews, Lord &
Nesbit Evans, 1979, reviewed by Kortlandt, 1983)
suggests a mixed woodland savanna in the area.
However, the Ruisenga Island fossils are those of
evergreen forest and semi-evergreen species (A.
C. Hamilton, personal communicaion) and so
Guineo-Congolian moist forests sill eisted in
central Africa at that ime.
The easten African coast, and what mountains
there were, would have received increasing
amounts of moisture from the Indian Ocean as
the breakup of Gondwanaland proceeded,
becoming forested in the Miocene if not before.
At this ime it has been suggested that connection
between the western Guineo-Congolian forests
and easten African coastal and Eastern Arc for-
ests would have been by gallery forests along
rivers extending to easten Africa from westen
Africa across a flat, eroded landscape (Axelrod &
Raven, 1978). However, fossil evidence from the
early Miocene indicates that there was easterly
drainage from the area around Lake Rudolf in
Kenya and westerly drainage from Lake Victoria
(Andrews & van Couvering, 1975). This suggests
that the early Miocene east-west watershed was
slightly east of its present posiion, but sill some
distance from the Eastern Arc mountains. Direct
contact by gallery forest would have been limited
at that ime, though the watershed may have been
low enough to permit migraion of orest organ-