The coexistence of multiple species competing for a finite set of resources is a widely debated topic in community ecology. Species with strongly overlapping niches are expected to drive each other towards exclusion, but such species may also coexist if they have similar competitive abilities. This compromise can lead to a peculiar pattern of clumped coexistence, where multiple species share similar niches, leaving gaps open in the theoretically available niche space. Large mammal herbivores may be a good example of this, where species' dietary niches clump as either grazers, browsers, or intermediate-feeders, rather than being continuously distributed over the resource gradient. Here, we develop a model of such emergent neutrality amongst species competing for a set of predefined resources each distributed along a finite niche axis. The model is commensurate with stable isotope niches, thus allowing us to compare its predictions with empirical evidence for changes in community niche structure over evolutionary time. We present stable carbon and oxygen isotope evidence for six discontinuous Late Quaternary assemblages from the central interior of South Africa, demonstrating the emergence of a clumped niche structure from an initial pattern of strongly overlapping diets. We show that species tend to cluster on parts of the niche axis where resource availability is highest, mirroring the proliferation of grazer species as landscapes became more grass-dominated. However, the presence of competition means that species’ niches continue to differentiate, explaining the persistence of browser and intermediate-feeder species even in these open, predominantly treeless landscapes. These results highlight that species interactions are a necessary factor for robust inferences about the evolutionary dynamics of palaeocommunities.

The blue antelope ( Hippotragus leucophaeus ) is the only large African mammal species to have become extinct in historical times, yet no nuclear genomic information is available for this species. A recent study showed that many alleged blue antelope museum specimens are either roan ( H. equinus ) or sable ( H. niger ) antelopes, further reducing the possibilities for obtaining genomic information for this extinct species. While the blue antelope has a rich fossil record from South Africa, climatic conditions in the region are unfavourable to the preservation of ancient DNA. Nevertheless, we recovered two blue antelope draft genomes, one at 3.4x mean coverage from a historical specimen (~200 years old) and one at 2.1x mean coverage from a fossil specimen dating to 9,800–9,300 cal BP, making it currently the oldest palaeogenome from Africa. Phylogenomics show that blue and sable antelope are sister species, confirming previous mitogenomic results, and demonstrate ancient gene flow from roan into blue antelope. We show that blue antelope genomic diversity was much lower than in roan and sable antelopes, indicative of a low population size since at least the early Holocene. This supports observations from the fossil record documenting major decreases in the abundance of blue antelope after the Pleistocene-Holocene transition. Finally, the persistence of this species throughout the Holocene despite low population size suggests that colonial-era human impact was likely a decisive factor in the blue antelope’s extinction.

Native to southern Africa, the blue antelope ( Hippotragus leucophaeus ) is the only large African mammal species known to have become extinct in historical times. However, it was poorly documented prior to its extinction ~ 1800 AD, and many of the small number of museum specimens attributed to it are taxonomically contentious. This places limitations on our understanding of its morphology, ecology, and the mechanisms responsible for its demise. We retrieved genetic information from ten of the sixteen putative blue antelope museum specimens using both shotgun sequencing and mitochondrial genome target capture in an attempt to resolve the uncertainty surrounding the identification of these specimens. We found that only four of the ten investigated specimens, and not a single skull, represent the blue antelope. This indicates that the true number of historical museum specimens of the blue antelope is even smaller than previously thought, and therefore hardly any reference material is available for morphometric, comparative and genetic studies. Our study highlights how genetics can be used to identify rare species in natural history collections where other methods may fail or when records are scarce. Additionally, we present an improved mitochondrial reference genome for the blue antelope as well as one complete and two partial mitochondrial genomes. A first analysis of these mitochondrial genomes indicates low levels of maternal genetic diversity in the ‘museum population’, possibly confirming previous results that blue antelope population size was already low at the time of the European colonization of South Africa.

We report a newly extended stratigraphic sequence with associated Palaeolithic sites from the area of the extinct Kilombe volcano in central Kenya. The extended archaeological sequence runs from Oldowan finds, through the Acheulean, and up to the Middle Stone Age. The sedimentary sequences within the Kilombe caldera and south flanks of the mountain have been dated through 40Ar/39Ar measurements and palaeomagnetic studies. A series of 40Ar/39Ar values date the geological sequence from 2.493 +/- 0.095 Ma, near the beginning of the Lower Pleistocene, through to 0.118 +/- 0.030 Ma near the Middle to Upper Pleistocene transition. It includes the first entirely new area of Oldowan localities in East Africa south of Ethiopia for thirty years, and the first in a rugged mountainous setting. Trachyte lavas of Kilombe mountain were extruded during and after c. 2.5 Ma, followed by formation of a caldera and subsequent caldera lake, and sedimentation of a sequence of tuffs, diamictites, sandstones, and claystones. Sections in the mid-part of this intra-caldera fill-sequence have produced dates of 1.8-1.7 Ma, associated with an Oldowan industry and fauna dated precisely at 1.814 +/- 0.004, and overlain by Acheulean finds at higher level. On the southern outward flanks of Kilombe mountain, a second major sequence is bounded at the base by trachyphonolite and a tuff yielding dates in the range 1.58-1.50 Ma. The main Acheulean archaeological site (GqJh1) falls within the overlying sedimentary sequence and has an age of c. 1.0 Ma, on the basis of a new 40Ar/39Ar date for the Three-Banded Tuff and palaeomagnetic reversal stratigraphy. Further 40Ar/39Ar dates indicate an age of c. 0.48-0.46 Ma for a marker ashflow tuff (AFT), prominent across the area. At Moricho, west of Kilombe, sediments above the AFT have been dated in the range 270,000-120,000 years and are associated with Middle Stone Age assemblages. In total, these sites attest to hominin activity from an Oldowan horizon dated to 1.8 Ma up to Later Stone Age stone scatters within the last 100,000 years.

The cranium from Broken Hill (Kabwe) was recovered from cave deposits in 1921, during metal ore mining in what is now Zambia1. It is one of the best-preserved skulls of a fossil hominin, and was initially designated as the type specimen of Homo rhodesiensis, but recently it has often been included in the taxon Homo heidelbergensis2–4. However, the original site has since been completely quarried away, and—although the cranium is often estimated to be around 500 thousand years old5–7—its unsystematic recovery impedes its accurate dating and placement in human evolution. Here we carried out analyses directly on the skull and found a best age estimate of 299 ± 25 thousand years (mean ± 2σ). The result suggests that later Middle Pleistocene Africa contained multiple contemporaneous hominin lineages (that is, Homo sapiens8,9, H. heidelbergensis/H. rhodesiensis and Homo naledi10,11), similar to Eurasia, where Homo neanderthalensis, the Denisovans, Homo floresiensis, Homo luzonensis and perhaps also Homo heidelbergensis and Homo erectus12 were found contemporaneously. The age estimate also raises further questions about the mode of evolution of H. sapiens in Africa and whether H. heidelbergensis/H. rhodesiensis was a direct ancestor of our species13,14. Analyses of the hominin skull from Broken Hill, Zambia, place it at an earlier date than previously thought, confirming that later Middle Pleistocene Africa was home to at least three lineages of hominin.

Although the Middle Stone Age (MSA) of southern Africa, associated with major cultural innovation including aspects of symbolic behavior and the development of complex hunting tools, has been the focus of intensive research, well-documented contexts for the early Middle Stone Age (EMSA) are rare. Here, we present archeological and ecological data on the EMSA occupation of Wonderwerk Cave excavated by Peter Beaumont, along with the results of luminescence dating of associated sediments to ca. 240–150 kyr, overlapping with the timing of the first known modern humans. The lithic assemblage shows a shift to prepared core flake production but lacks complex hunting equipment characteristic of the later MSA. Although ocher is present, there is no evidence of ornaments or incised objects. Multiproxy paleoclimate data from Wonderwerk Cave demonstrate that the EMSA occupation occurred under significantly wetter environmental conditions than the current semiarid regime. The Wonderwerk Cave EMSA provides strong support for the argument that critical aspects of the MSA archeological record developed long after the first appearance of modern humans.

The Cornelia-Uitzoek fossil site has produced a large collection of bones, Acheulean artifacts and a Homo sp. tooth dated to ~1 million years ago. The faunal assemblage defines the Cornelian Land Mammal Age and is characterized by a number of extinct species of large mammals that reflect an open grassland environment. Bones were accumulated by hyenas, whereas artifacts appear to be chance inclusions, although they suggest human presence in the immediate surroundings. Previous studies established the absolute chronology of the site and a broad stratigraphic sequence. However, the sedimentary units identified in the field were not linked to specific formation processes, thus limiting the understanding of the depositional history of the site and its potential in reconstructing Pleistocene environments. Using a microgeoarchaeological approach based on infrared spectroscopy and micromorphology of sediments, we were able to determine the formation and post-depositional processes of the entire stratigraphic sequence, and to show that the site is characterized by alluvial sediments accumulated under different river flow regimes. Our results provide context to the faunal and lithic assemblages and are in agreement with existing paleoenvironmental data for the site.

Klasies River is arguably one of the most informative Middle Stone Age sites in Africa. Its extensive occupational sequence and the large quantity of faunal material, lithic artefacts, and Homo sapiens remains makes this a key site to explore modern human evolution. Two major excavations of the Klasies sequence - the first by Singer and Wymer in 1967/68, and later by Hilary Deacon in the 1980s and 1990s - yielded a vast quantity of faunal material. However, interpretation of the Singer and Wymer fauna has been marred by , flawed collection techniques. In contrast, all faunal remains from the Deacon collection were recovered and analysed with the potential to provide less biased results. Here, we report on the fauna from the Deacon assemblage, placing particular emphasis on the description of the material. We find that the 'Klasies pattern' - or an ungulate skeletal-part profile dominated by skulls and feet for larger bovids, with a more equitable skeletal-part distribution for smaller bovids - is not represented in the Deacon collection. The Klasies pattern was therefore likely caused by sampling bias. Generally, bovid remains dominate the assemblage, while seal and hyrax are the most common identified taxa. The palaeoenvironmental implication of the material is briefly discussed, which confirms the results from Klein's (1976) analysis of the Singer and Wymer collection.

Given the steep present-day climatic gradients in southern South Africa, comparative studies of its coastal and inland paleoclimate provide important insight into the region's spatial climate dynamics. We present a comparative study of new speleothem stable oxygen (δ18O) and carbon (δ13C) isotopic composition from Efflux Cave (EC; 113–19 ka, hiatus between 105 and 94 ka) in the Little Karoo inland basin and Herolds Bay Cave (HBC; 93–62 ka) at the boundary between the Cape coastal lowlands and the Palaeo-Agulhas Plain. Gaussian kernel based cross-correlation analyses against palaeoclimate proxy records show that rainfall seasonality at EC is related to rainfall amount in the South African interior, whereas HBC is similar to the South African east coast. Our new proxy records, corroborated by archaeological evidence from the Cape coastal lowlands and the Little Karoo, suggest that in both regions, surface water was readily available and grassland ecosystems with variable amounts of C4 and C3 grasses as well as fynbos communities could be found nearby. Cross-correlation analyses of the new records from EC and HBC compared to previously published speleothem records from southern South Africa show only limited statistical correlation. However, there are some similarities in overall trends of palaeoclimate change in the Cape coastal lowlands and the Little Karoo whereby Marine Isotope Stage (MIS) 5b (mainly Greenland Interstadial 22) and 4 generally have more summer rainfall and higher abundances of C4 vegetation than MIS 5a and MIS 3.

Objective: To document and describe the occurrence of an enamel pearl on the distal root surface of the maxillary M3 of the fossil hominin specimen from Florisbad, South Africa that is dated to ca. 259,000 years B.P., and is an early representative of Homo sapiens or as a member of the evolutionary line that was directly ancestral to modern humans. Design: The molar was examined macroscopically and by micro-computed tomography (mu CT) to enable accurate measurement and visualization of the structure of the enamel pearl. Results: The single pearl has a diameter of 0.97 mm; it is a Type 2 "composite" pearl comprising an enamel cap and dentine core without pulp chamber involvement. The size of the Florisbad pearl falls within or just below the size ranges of this anomaly in modern human samples. Type 2 pearls are most commonly encountered in recent human populations, and the location of the pearl on the distal root surface of the Florisbad M-3 is consistent with its most frequent location in recent humans. Pearls in recent human populations affect between 0.2-4.8% of individuals, and 1.7-6.8% of permanent molars. Pearls have been documented in several prehistoric human dentitions, and all examples are less than 4000 years old. Conclusions: Enamel pearls have been associated with periodontal disease, but it is not possible to relate its presence to the advanced periodontal inflammation and alveolar bone loss in the Florisbad fossil. Florisbad presents the earliest evidence of this anomaly in the fossil record pertaining to modern humans.

New pollen evidence and a review of past palynological research at the Pleistocene Florisbad thermal spring mound, which produced faunal, hominin and archaeological remains, allow a reconstruction of past conditions in the central Free State grassland, South Africa. Pollen sequences at the spring consist of alternating pollen rich organic peaty horizons interrupted by hiatuses or pollen-poor layers of sandy and clayey deposits. The paleoclimatic interpretation of pollen sequences in springs like Florisbad, which contain a mixture of local and regional pollen and are influenced by spring mound cycles and changes in swamp configuration, requires consideration of taphonomy and site geomorphology to separate local from regional influences. By correlating the pollen stratigraphy in different parts of the site we suggest that the lower layers containing the Florisbad hominin (dated to 259 ka) and its associated Middle Pleistocene fauna, experienced cool moist and grassy conditions. Dating of some of the overlying Middle Stone Age layers suggest that they encompass the Last Interglacial period (MIS 5e; ca. 124-119 ka). In these levels the pollen evidence of upland fynbos shrubs unexpectedly suggests that cool conditions prevailed. This raises questions about the taphonomy, precision of the age estimate of occupation or the way we interpret pollen spectra. (C) 2019 Elsevier B.V. All rights reserved.

The influence of climatic and environmental change on human evolution in the Pleistocene epoch is understood largely from extensive East African stable isotope records. These records show increasing proportions of C4 plants in the Early Pleistocene. We know far less about the expansion of C4 grasses at higher latitudes, which were also occupied by early Homo but are more marginal for C4 plants. Here we show that both C3 and C4 grasses and prolonged wetlands remained major components of Early Pleistocene environments in the central interior of southern Africa, based on enamel stable carbon and oxygen isotope data and associated faunal abundance and phytolith evidence from the site of Wonderwerk Cave. Vegetation contexts associated with Oldowan and early Acheulean lithic industries, in which climate is driven by an interplay of regional rainfall seasonality together with global CO2 levels, develop along a regional distinct trajectory compared to eastern South Africa and East Africa.

A fossilised large mammal bonebed was discovered eroding out of a gully in the Free State of South Africa. The bonebed is similar to 1.5 m below the modern land surface, and extends over an area 35 x 13 m. Surface scatters of stone tools occur in a 1 km radius of the site, and a large fire place associated with spirally fractured burnt bone is preserved to one side. The purpose of this research was to excavate and taphonomically analyse the faunal sample to elicit the cause of death, and radiocarbon date it to establish when it happened. The bonebed is represented by black wildebeest, including juvenile and adult individuals. Faunal remains are randomly oriented and many are complete. Weathering stage 1 on most of the bones together with the articulation pattern suggest that the carcasses were exposed for more than a year and less than three before being buried by hillslope sediment. Two-thirds of those fractured record a spiral breakage pattern. There are a few trample marks on bones and evidence of some termite activity. No stone tools were found in the section of bonebed we excavated, and there is no evidence of manmade or carnivore damage on the fauna. Calcrete nodules in the underlying deposits and phytoliths representative of desertification throughout the sedimentary sequence suggest that the animals died under drought conditions between 3840 +/- 40 and 3500 +/- 40 cal BP, and that human activity at the site was marginal. (C) 2017 Elsevier Ltd and INQUA. All rights reserved.

Here we report on the taphonomy of 29 lower jaws recovered from 25 recent cheetah kill assemblages collected in the Mountain Zebra National Park (MZNP, Eastern Cape Province, South Africa). Cheetah kills commonly attract other carnivores and so serve as "feeding hubs". As only a limited number of carnivores were present in MZNP at the time of this study, we could constrain the potential sources of taphonomic damage to cheetahs, black-backed jackals and brown hyaenas. The collected jaws derive from eight different prey species that differ in body weight as well as age(adult versus juvenile). Patterns of bone damage were examined for three different portions of the mandible: the anterior portion(incisor-diastema region), the mandible corpus and the posterior mandible(gonion, ramus, mandible condyle, coronoid process). Surface micro-damage was also scored. Statistically significant differences in the extent of taphonomic damage was found in the posterior portion of the mandible as well as the mandible corpus between large-sized prey species and medium to small-sized prey, but no differences were found in damage intensity in the anterior portion of the mandible. Surprisingly no significant age-related differences were found between adults and juveniles in the extent of mandibular damage, but sample sizes are small. Dimensions of pits, punctures and scores on all jaws studied(irrespective of species) indicate the primary activity of a small-sized canid (jackal) and/or medium-sized felid (cheetah), with secondary input of a large-sized carnivore(hyaena). Though sample sizes examined are small, this offers a promising approach for the investigation of inter-specific taphonomic signatures on prey carcasses from carnivore "feeding hubs".

This paper provides an interpretation of the chronology of the Early Later Stone Age (Early LSA) in South Africa and Southern Namibia. The Early LSA tradition first appears in Africa at Border Cave between ∼46 and 44ka, and in Southern Africa this tradition slowly appears to the west, replacing groups producing Middle Stone Age (MSA) technologies. In this region, the latest transformation from MSA to Early LSA occurs in southern Namibia between 25 and 23ka and perhaps in secluded areas in the interior as late as 21ka. The timing and causes of this dispersion are poorly understood. The transition from Early LSA technologies to the Robberg Industry appears to be a rapid sub-continental wide cultural transformation.

Within the past 80 years, the Kromdraai site in South Africa has provided a diverse Early Pleistocene fauna (notably bovids, carnivores, primates, large rodents, birds, proboscidea). Since 2014, the Kromdraai bone accumulation has been the focus of intensive fieldwork that demonstrated that the site is much larger than previously recognised. In the present taphonomic study of a new and large faunal sample including more than 2400 remains, stratigraphically controlled, we aim to test previous interpretations of the Kromdraai assemblage as representing a death trap or a carnivore lair. In particular, we aim to discuss the relationships between faunal communities in the context of carnivores, either predator-prey or predator-predator interactions. We investigate the relative abundance of anatomical elements and their fragmentation, the mortality profiles and tooth-mark frequency. We conclude that carnivores (particularly felids and hyenids) played a major role in the accumulation of fauna from Member 2, which (thus far) is the oldest depositional period investigated at Kromdraai. However, the high species diversity suggests that the secondary predators (scavengers) could have modified the bone deposit produced by the primary predators. The presence of hominin remains is also questioned. Our results shed new light on the palaeoecology of the Kromdraai Member 2 hominins, in terms of opportunistic predators and/or prey of large carnivores.