Bats of Southern and Central Africa. Ara Monadjem

2004, Cotterill 2005). There are steep escarpments in this region, most notably Zambia’s Muchinga Escarpment, which bounds the western edge of the Luangwa valley. The significance of the Muchinga Escarpment as a biogeographical boundary was originally emphasised by Neave (1907, 1910a, b) and Ansell (1978). The enhancement of the Muchinga Escarpment by the East African Rift created a prominent eastern bound to the Kalahari Plateau; this rugged scarp extends south, bordering the Gwembe trough, markedly eroded by the Middle Zambezi (today artificially impounded in Lake Kariba at Kariba Gorge).

      Equally impressive escarpments form the steep flanks of the Middle Zambezi valley in Zambia and Zimbabwe (Cotterill 2003, Moore et al. 2009, 2020) (Figure 28). These escarpments bound formerly extensive landscapes, and have witnessed a prolonged history of uplift and erosion that has persistently lifted and warped the Kalahari Plateau, faulting its margins in parts (de Wit 2007).

Drainage systems

      The modern hydrology of southern Africa is dominated by several major drainage basins, with the Congo and Cuanza basins in the extreme north, the Zambezi and Limpopo rivers, which drain eastwards into the Indian Ocean, and the Kunene and Orange–Vaal systems that drain west into the Atlantic (Figure 24). River topology strongly influences the distributions of bat species that depend on riparian habitats for food and daylight roosts. This can be seen very clearly in the Namib Desert, where the Kuiseb River supports populations of tree-roosting bats.

      The drainage systems that characterise southern Africa today are very different from their precursors. Repeated uplift and erosion of the Kalahari Plateau and the propagation of the East African Rift System into southern Africa caused radical rearrangements of the region’s drainage systems.

      Following the final stages in the break up of Gondwana (∼127 Ma), the Palaeo-Limpopo River drained a large part of the interior of central southern Africa, with most of the rivers of today’s lower DRC, Angola, Botswana and Zimbabwe flowing into it. About 80 Ma, epeirogenic uplift created vast depressions across the Kalahari Plateau Basin in the interior of southern Africa. What are today the headwaters of the Kunene, Orange, Limpopo and Zambezi rivers originally flowed into the Kalahari Basin (Moore et al. 2009). These endorheic rivers maintained vast lakes in the interior, notably in the Etosha basin and northeastern Botswana; one of the largest of these lakes was Palaeo-Lake Makgadikgadi, which had an area of ∼67,000 km² (Cotterill 2006, Moore et al. 2012).

      Figure 23. The view north along the western escarpment of the Tanganyika graben, northeastern Zambia, illustrates the dramatic topography expressed along the East African Rift System. The vegetation is dominated by miombo woodland, with gallery and scarp forest, and also includes Itigi thicket on the valley floor (© F. P. D. Cotterill).

      The southwest propagation of the East African Rift System into southern Africa had a marked influence on the development of the Zambezi River System. Concomitant with this rifting, the Zambezi and its tributaries incised deeply westwards across the eastern rim of the Kalahari Plateau, capturing several of the rivers that were feeding the palaeo-lakes. This prolonged episode began with the Zambezi’s piracy of the Palaeo-Luangwa, and its most recent piracies of the Upper Zambezi and Upper Kafue rivers (Moore et al. 2009, 2012, 2020). The Proto-Congo and Middle Zambezi rivers expanded their catchments significantly through the piracy of most of the large endorheic rivers. Continued tectonism on the Kalahari Plateau, associated with tectonism along the East African Rift System, also led to the diversion of the Kwando (= Chobe) River into the Zambezi River. Today, the only large endorheic rivers still draining into the Kalahari Basin are the Cubango and Quito, which form the Okavango Delta in northern Botswana (Goudie 2005, Cotterill 2006, Moore et al. 2009, 2012) (Figure 24).

Figure 24. The extent of the elevated expanse of the Kalahari Plateau is depicted in this computer-generated Digital Elevation Model (DEM). Situated south of the Congo basin, the Kalahari Plateau is bounded by flanking coastal escarpments, and has been greatly altered by the East African Rift System (notably the Gwembe-Luangwa valleys) (Generated from the SRTM-3 (Shuttle Radar Topographic Mapping 3) dataset compiled by NASA, with annotations, including locations of principal cities; © F. P. D. Cotterill, AEON).

      These dynamic drainage systems are landforms of central significance to palaeobiological and geological research, because their history provides invaluable keys to decipher the complexities of landscape evolution. Although comparatively subtle, these changes have profoundly influenced the distribution and evolution of Africa’s fauna, including bats (Cotterill 2003, 2004, 2005, Moore et al. 2009, 2012, 2020).

CLIMATES PAST AND PRESENT

      The climate of southern Africa is dominated by a characteristic seasonality, with a cool-dry season interfacing with a hot-dry season, and then a hot-wet season, through the annual cycle. The southwestern Cape region experiences a winter-rainfall regime with hot, windy summers. A gradient of increasing rainfall trends from west to east; precipitation is both erratic and lowest in the deserts of the Namib, Namaqua, and the southwestern Kalahari. Adiabatic rainfall is not infrequent along the windward margins of the coastal escarpments and mountains. Prevailing mean annual temperatures and mean annual rainfall also exhibit an overall increase from south to north, with precipitation under increasing influence of the Inter-Tropical Convergence Zone and the Congo Air Boundary. The Monsoon rainfall system also influences the climate of northeastern Mozambique (Tyson and Preston-Whyte 2000).

      The evolution of southern Africa’s landscapes and, especially, its vegetation has been influenced strongly by palaeoclimatic dynamics. Empirical evidence for past conditions across the hinterland is very poor, especially for the time before the Late Pleistocene (120,000 years ago). Nevertheless, the widespread fossil dune systems across the central and northern Kalahari Plateau testify to widespread arid conditions that prevailed across a vast portion of the subregion, presumably during the Late Cenozoic (de Menocal 2004, Goudie 2005). Such extreme periods of aridity were interspersed with equally humid phases, when mesic vegetation expanded south and west of the Congo basin (Benson and Irwin 1965).

      Several lines of evidence suggest these expansions were focused along drainage lines in the Upper Congo and Zambezi systems. The significance of this interplay of both geomorphological evolution and palaeoclimatic dynamics in exercising dominant controls over palaeo-environments, is illustrated by the modern habitat mosaic characterising the Great Equatorial Divide. Here, Late Cenozoic drainage evolution has seen the overall migration of the Congo’s headwaters southward, by their repeated captures of Zambezi tributaries. These piracy events have culminated in the Great Equatorial Divide shifting south by ∼400 km through the Late Cenozoic (F. P. D. Cotterill, unpublished data). This has contributed significantly to a ‘shuffling’ juxtaposition of forest species and savanna species across these landscapes. It explains interesting biogeographical patterns in fishes (Schwarzer et al. 2012), mammals, birds (Cotterill 2002a, b, 2006) and snakes (Broadley et al. 2003), which are congruent with the high endemism and species richness of plants documented across this region (Linder 2001). This history is invoked to explain vicariant speciation of forest-adapted vertebrate species with their ranges centred on mushitu (gallery forest) and fringes of mesic miombo savannas in Katanga and northern Zambia. Alongside vegetation shifts that accompanied major changes in the palaeoclimate through the Plio-Pleistocene, this geomorphological evolution plausibly explains why the Ikelenge Pedicle (northwestern Zambia) and Katanga in the DRC are zones of high endemism. At least one horseshoe bat species, Rhinolophus sakejiensis, is known only from the Ikelenge Pedicle (Cotterill 2002a, b, 2006, Broadley et al. 2003). It is this forest–savanna mosaic that dominates the vegetation of the southern Congo basin and northern Angola, and justifies inclusion of such a vast