South African Cave Research Articles

The earliest report of bone-bearing breccia from a Monte Christo Formation cave (South Africa)

A series of historical documents by the 19th-century amateur naturalist and palaeontologist Alfred Brown report the earliest discovery of bone-bearing cave breccia in the former Transvaal (South Africa). The oldest of these reports dates from 1890 and predates the first mention of the existence of bone-bearing breccia at the famous Sterkfontein-Kromdraai caves by five years. The breccia fragment was kept by Brown in his collection, where it was noted to come from the Monte Christo gold mine near Ventersdorp (North-West Province). Brown's specimen is lost, but we could relocate the mine and confirm the presence of breccia deposits. Alfred Brown's notes are the earliest scientific report about the bone-bearing nature of Plio-Pleistocene South African caves and the first to acknowledge them as a potential source of hominin fossils, some 30 years before the discovery of the Taung Child, Australopithecus africanus. The finding strengthens the chronology of the discovery of the South African fossil hominin sites. It also significantly shortens the gap between the gold rush and the first discovery of bone-bearing breccia in the Witwatersrand.

Early Pleistocene hominin subsistence behaviors in South Africa: Evidence from the hominin-bearing deposit of Cooper's D (Bloubank Valley, South Africa)

Evidence of the consumption of meat through hunting or scavenging by Early Pleistocene hominins is scarce, particularly in South Africa. Moreover, the interpretations of taphonomic evidence are subject to an important discussion commonly called the ‘hunting-vs-scavenging debate.’ Until today, only the Swartkrans Members 1–3 site has yielded a butchered bone assemblage large enough to permit reconstruction of carcass acquisition strategies by Early Pleistocene hominins in South Africa. This leaves an information gap between 1.4 and 1.0 Ma. Here, we provide the first evidence of meat consumption by hominins during this gap, based on the zooarchaeological study of the large mammal bone assemblage recovered from the Cooper's D site, South Africa. Based on skeletal part representation, our results show density-mediated attrition of bovid bones due to predepositional and postdepositional destruction. We argue that this attrition is the result of both abiotic (i.e., decalcification) and biotic (i.e., carnivore ravaging) processes. Bovid mortality profiles point out the involvement of ambush predators such as large felids. Bone surface modifications also indicate that the assemblage has been accumulated mostly by carnivores but with some hominin involvement as well. We observe all the stages of animal carcass processing (skinning, disarticulation, defleshing, marrow extraction) as well as the exploitation of a diversity of prey size classes at both Swartkrans Members 1–3 and Cooper's D. Thus, our study shows the importance of the Cooper's D bone assemblage for understanding Early Pleistocene hominin subsistence behaviors. Moreover, this article highlights the need for including long bone flake specimens in the analysis of large bone assemblages from South African caves to better understand the Early Pleistocene hominin bone damage record.

One hominin taxon or two at Malapa Cave? Implications for the origins of Homo

A report on the skeletons of two individuals from the Malapa cave site in South Africa attributes them both to a new hominin species, Australopithecus sediba. However, our analysis of the specimens’ mandibles indicates that Australopithecus sediba is not a ‘Homo-like australopith’, a transitional species between Australopithecus africanus and Homo. According to our results, the specimens represent two separate genera: Australopithecus and Homo. These genera are known to have jointly occupied sites, as seen in several early South African caves, so one cannot rule out the possibility that Malapa also contains remains of the two taxa. Our results lead us to additionally conclude that all the Australopithecus species on which the relevant mandibular anatomy is preserved (not only the ‘robust’ australopiths but also the ‘gracile’ – more generalised – ones) are too specialised to constitute an evolutionary ancestor of Homo sapiens. Furthermore, given that the Malapa site contains representatives of two hominin branches, one of which appears to be Homo, we must seek evidence of our origins much earlier than the date assigned to Malapa, approximately 2 million years before present. Support for this claim can be found in Ethiopian fossils attributed to the genus Homo and dated at 2.4 and 2.8 million years before present.Significance:
 
 The proposed hominin species Australopithecus sediba, from the Malapa Cave in South Africa, seems to actually consist of two species, each of which represents a different hominin genus: Homo and Australopithecus. If, indeed, this is the case, Homo must have originated prior to the Malapa remains, contrary to the scenario suggested in the original report on Au. sediba.

Threat analysis of modelled potential migratory routes forMiniopterus natalensisin South Africa

AbstractMigrant cave‐dwelling insectivores that rely on specific caves for maternity and hibernation, like the Natal long‐fingered batsMiniopterus natalensisin South Africa, may be at particular risk of population decline in an urbanising world. As a step towards the conservation of caves and cave‐dwelling bats in South Africa, this study aimed to (i) broadly identify the number of caves used by bats (any species) and specificallyM. natalensis, throughout South Africa, (ii) investigate the number of maternity and hibernacula roosts currently known forM. natalensis, (iii) assess the number of caves located in formal protected areas, (iv) determine potential migration paths/corridors between hibernacula and maternity sites and (v) evaluate the potential threats (like onshore wind facilities) along potential migratory routes. A meta‐analysis of scientific literature and websites was conducted to identify caves throughout South Africa and locations of maternity and hibernacula roosts forM. natalensis. Roosts were assessed to determine whether (i) they were located in protected areas, (ii) they were used for eco‐tourism and (iii) the distance to primary roads and onshore wind energy facilities. Next, likely migratory paths were modelled between maternity and hibernacula sites using least‐cost path analysis and the threats along potential routes were investigated. A total of 92 caves were identified, 50 were reported to contain bats.M. natalensiswere recorded in 37 caves, and of those, only 9% are currently located inside protected areas. A total of 12 least‐cost paths were modelled, and various paths intersected potential threat risk elements. Our analysis provides the first description of the potential migration corridors forM. natalensisin South Africa, as well as the current conservation status of bat‐inhabited caves. For a developing country set to experience increased urbanisation pressures, this study highlights the need for conservation measures for South African caves and the dependent bats.

World’s oldest camp bedding found in South African cave

World’s oldest camp bedding found in South African cave

Who ate OH80 (Olduvai Gorge, Tanzania)? A geometric-morphometric analysis of surface bone modifications of a Paranthropus boisei skeleton

Olduvai Gorge (Tanzania) is one of the key areas for the study of human origins, given the sheer abundance of archaeological and paleontological sites discovered. Several of these archaeological sites have yielded numerous hominin fossils and traces of their activities, thus offering invaluable insights into the nature and origins of human behavior. Nevertheless, certain taphonomic discussions that have been of great importance for the study of the South African cave sites have remained unnoticed in East Africa. One of these issues revolves around the interpretation of Paranthropus as a common prey of predators. In this paper, we analyze the postcranial remains of OH80, a partial skeleton of a Paranthropus boisei discovered at the BK site (Bell Korongo, Bed II of Olduvai Gorge, Tanzania) in 2010. Some of the specimens of this skeleton have been reported to show surface modifications tentatively associated to the action of carnivores. Here, several pits observed on OH80-12, the Paranthropus boisei femur, were digitally reconstructed and analyzed through geometric morphometrics to determine the nature of the marks. The OH80-12 marks are compared with a modern sample of percussion marks and several carnivore tooth pits, since the action of both agents has previously been identified at the site and can sometimes result in equifinality problems. Alongside already published geometric morphometric tests, here we present a very newly developed approach based on the combination of geometric morphometric data and the training of machine learning algorithms. Both methodologies highlight that OH80 was consumed by carnivores, being a felid the most likely modifying agent.

Speleothem evidence for C3 dominated vegetation during the Late Miocene (Messinian) of South Africa

During the Late Miocene, Africa experienced a number of ecological transitions including the spread of C4 grasslands, the expansion of the Sahara Desert, the Messinian Salinity Crisis and a number of mammalian migrations and expansions, including the origin of the hominin clade. A detailed understanding of the relationship between environmental change and hominin evolution is hampered by the paucity of data available from terrestrial localities, especially in southern Africa. Here, we present a stable isotope and trace element record from a speleothem from the South African cave site of Hoogland. Uranium-lead dating and magnetostratigraphy places the speleothem within the Messinian Age (7.25–5.33 Ma) of the Late Miocene, making it the oldest known cave deposit from the region near the UNESCO Fossil Hominids of South Africa World Heritage Site (locally known as the “Cradle of Humankind”). Low carbon isotope values indicate a predominantly C3 vegetation in the vicinity of the cave throughout the period of speleothem growth. It is not possible to determine if this represents a C3 grassland or a C3 woodland, but it is clear that an equivalent C3-rich environment has yet to be found during the Messinian of east Africa. We conclude that the C4 grass expansion occurred millions of years later in South Africa than it did in eastern Africa, and that this vegetation shift should be considered when comparing African vegetation change with the late Miocene hominin fossil record.

Ages-at-death distribution of the early Pleistocene hominin fossil assemblage from Drimolen (South Africa).

A prevailing hypothesis in paleoanthropology is that early Pleistocene hominin bones were accumulated in South African caves by carnivores, which used those shelters, and the trees surrounding them, as refuge and feeding sites. We tested this hypothesis at the site of Drimolen, by comparing its hominin age-at-death distribution to that of the nearby and roughly contemporaneous site of Swartkrans. We employed standard dental aging systems in order to categorize the Drimolen hominin teeth into age classes of 5 years each. We then compared the age-at-death distribution for Drimolen with the published data available for the Swartkrans hominins. Age-at-death distributions indicate that the age category "young adults" is the best represented age category at Swartkrans and the most poorly represented one at Drimolen. Moreover, Drimolen has a preponderance of infant specimens. Both sites have a low frequency of old adult specimens. Differences observed in frequencies of the age-at-death categories suggest different mechanisms of hominin skeletal accumulation at Drimolen and Swartkrans. Swartkrans' frequency curve reflects mortality in a population subjected to predation and is thus consistent with the carnivore-accumulating hypothesis. In contrast, the Drimolen curve is similar to that of wild populations of living apes. Living primates have been observed exploiting caves as sleeping shelters, for nutritional, security, drinking, and thermoregulatory purposes. We suggest that similar cave use by Pleistocene hominins can explain, in large part, the accumulation of hominin bones at Drimolen. Such a conclusion is another illustration of the growing awareness that a "one-size-fits-all" taphonomic model for South African early Pleistocene hominin sites is probably insufficient.

World’s oldest drawing is Stone Age crayon doodle

‘Hashtag’ pattern drawn on rock in South African cave is 73,000 years old. ‘Hashtag’ pattern drawn on rock in South African cave is 73,000 years old.

Terrestrial laser scanning and photogrammetry techniques for documenting fossil‐bearing palaeokarst with an example from the Drimolen Palaeocave System, South Africa

AbstractThis paper presents the results of a recent three‐dimensional (3D) survey at the Drimolen Makondo palaeontological deposits in the Hominid Sites of South Africa UNESCO World Heritage site. The Drimolen Makondo is a palaeokarstic feature that consists of a heavily eroded 2.6‐2.0 Ma fossil‐bearing palaeocave remnant. With photogrammetry and a laser scan survey, two 3D site models were created, georectified, and imported into geographical information system software. This paper outlines both of these survey techniques and provides an assessment of the relevant merits of each method and their applicability for detailed recording and archival documentation of palaeokarstic palaeontological and archaeological sites. Given the complex depositional context of many of the fossil‐bearing South African cave systems and their importance for understanding our evolutionary history, new methods are critical to visualising and analysing 3D spatial data. The utility of 3D models lies in their ability to integrate with total station survey techniques to accurately record and control excavations and provide a means of visualising stratigraphic, sedimentary, and spatial contexts in various geographical information system platforms. The use of low‐cost and time‐efficient digital photographic surveys to create accurate 3D models, if completed accurately, can provide researchers with a means of contextualising excavation data without the need for expensive and highly specialised equipment. The development of this method combined with differential global positioning systems provides a solution in more remote locations to recording highly accurate fossil and 3D site contexts with increasing ease. It also allows the sites to be recorded as part of an evolving landscape rather than as single isolated localities. This technique should be a standard technique implemented when working on irreplaceable UNESCO World Heritage sites such as the hominin‐bearing caves of South Africa.

Stable carbon isotope ecology of small mammals from the Sterkfontein Valley: Implications for habitat reconstruction

Carbon isotope analyses of tooth enamel have been widely employed by paleoecologists to understand past habitats. Most such studies use large- to medium-bodied mammals and exclude small taxa. However, analysis of fossil small mammals holds promise for resolving questions about past environments because these animals are often present in the fossil record (especially in South African cave sites), are diverse in dietary and habitat preference, yet have limited lifespans and home range sizes. Thus, the carbon isotope compositions of small mammal communities and/or species might reflect the composition of vegetation in local environments at fine scales.In this study, we assessed the degree to which carbon isotope compositions of small mammal tooth enamel record spatial changes in habitat in a southern African savanna environment. Modern small mammal specimens were collected from the pellet accumulations of three barn owl (Tyto alba africanus) roosts located within microhabitats varying from open grassland to mixed woodland. We examined rodent carbon isotope compositions within taxa, between taxa, and between tooth types in an effort to characterize variation within this group. We also compared rodent community δ13Cenamel composition between microhabitat types to evaluate how well small mammal isotope data reflect vegetation composition local to the roosts. Our analyses suggest that the relationship between small mammal community carbon isotopic means and vegetation composition is complex, but that with appropriate taxonomic control and consideration of relative abundance, small mammals have potential as a proxy for reconstructing past habitats.To complement this modern study, we performed isotopic analysis of the enamel of small mammals from three hominin-bearing sites in the Cradle of Humankind World Heritage Site, South Africa. Our results suggest a greater contribution of C4 resources to the diets of small mammals and likely more C4 grass in the past than occurs today.

How We Came to Be: Evidence from Primate Relatives and Fossil Ancestors Offer Clues to How Human Capabilities for Birth, Posture, and Walking Evolved

How We Came to Be: Evidence from Primate Relatives and Fossil Ancestors Offer Clues to How Human Capabilities for Birth, Posture, and Walking Evolved

New early Pleistocene hominin teeth from the Swartkrans Formation, South Africa

We describe 14 hominin teeth and tooth fragments excavated recently from Swartkrans Cave (South Africa). The fossils derive from Members 1 (Lower Bank) and 3, from the Member 2/3 interface and from two deposits not yet assigned to member (the “Talus Cone Deposit” and the “Underground North Excavation” [UNE]) of the Swartkrans Formation, and include the first hominin fossil from the UNE, the two smallest Paranthropus robustus deciduous maxillary second molars in the entire hominin fossil record, and one of the smallest P. robustus permanent maxillary second molars from Swartkrans. The small permanent molar is accompanied by another tooth from a different individual but from the same stratigraphic level of the Swartkrans Formation; this second tooth is among, if not, the largest P. robustus permanent maxillary first molars known from anywhere—lending credence to assertions that degrees of body size sexual dimorphism previously ascribed to this species may be underestimated. It is more equivocal whether this evidence also supports hypotheses proposing that P. robustus assemblages from Swartkrans (as well as those from other South African cave sites) formed through the taphonomically biasing actions of large carnivores.

New human species discovered

Bizarre skeletons emerge from South African cave.

ARCHAEOLOGY. Development threatens home of early humans.

A rare record of ancient culture, left by early Homo sapiens in a South African cave, is at risk.

Ice Age Tools Hint at 40,000 Years of Bushman Culture

Archaeologists studying a South African cave say they have found 44,000-year-old artifacts—including bone tools and poisoned arrowheads—nearly identical to those still in use by hunter-gatherers.

Homo 'incendius'

An analysis of microscopic and spectroscopic features of sediments deposited in a South African cave one million years ago suggests that human ancestors were using fire much earlier than had been thought.

South African Cave Slowly Shares Secrets of Human Culture

In the hands of a skilled archaeologist, a South African site serves as a laboratory for testing ideas about early human culture and cognition.

Morphometric variation in the papionin muzzle and the biochronology of the South African Plio‐Pleistocene karst cave deposits

Papionin monkeys are widespread, relatively common members of Plio-Pleistocene faunal assemblages across Africa. For these reasons, papionin taxa have been used as biochronological indicators by which to infer the ages of the South African karst cave deposits. A recent morphometric study of South African fossil papionin muzzle shape concluded that its variation attests to a substantial and greater time depth for these sites than is generally estimated. This inference is significant, because accurate dating of the South African cave sites is critical to our knowledge of hominin evolution and mammalian biogeographic history. We here report the results of a comparative analysis of extant papionin monkeys by which variability of the South African fossil papionins may be assessed. The muzzles of 106 specimens representing six extant papionin genera were digitized and interlandmark distances were calculated. Results demonstrate that the overall amount of morphological variation present within the fossil assemblage fits comfortably within the range exhibited by the extant sample. We also performed a statistical experiment to assess the limitations imposed by small sample sizes, such as typically encountered in the fossil record. Results suggest that 15 specimens are sufficient to accurately represent the population mean for a given phenotype, but small sample sizes are insufficient to permit the accurate estimation of the population standard deviation, variance, and range. The suggestion that the muzzle morphology of fossil papionins attests to a considerable and previously unrecognized temporal depth of the South African karst cave sites is unwarranted.

Stratigraphy, U-Th chronology, and paleoenvironments at Gladysvale Cave: insights into the climatic control of South African hominin-bearing cave deposits

Gladysvale Cave is one of the few Plio-Pleistocene hominin-bearing cave sites in South Africa that contains a well-stratified cave fill with clastic sediments interspersed with flowstones. The clastic sediments can be divided into units based on the presence of intercalated flowstones, forming flowstone bounded units (FBU). Ten MC-ICP-MS uranium-series dates on several flowstone horizons in the Gladysvale Internal Deposit fan indicate deposition from the late mid-Pleistocene (∼570 ka) to Holocene (∼7 ka) during limited periods of higher effective moisture. Clastic sedimentation occurred during the interceding, presumably more arid, periods. This sequence is not consistent with earlier models for South African caves that simply assumed interglacial sedimentation and glacial erosion. 13C/ 12C data suggest that flowstone tended to form during periods with higher proportions of C 3 plants in the local vegetation, while clastic sediments reflect higher proportions of C 4 grasses, although this is not always the case. We argue that flowstones are precipitated during periods of higher effective precipitation and restricted cave entrances, while clastic sediments accumulated during periods with more open vegetation. The sedimentary fill of the fossiliferous deposits are, therefore, highly episodic in nature, with large periods of time unlikely to be represented. This has serious implications for the other hominin-bearing caves close by, as these deposits are likely to be similarly episodic. This is especially pertinent when addressing extinction events and reconstructions of paleoenvironments, as large periods of time may be unrecorded. The Gladysvale Cave fill sediments may serve as a climatically forced chronostratigraphic model for these less well-stratified and well-dated Plio-Pleistocene sites.