1 Introduction

1Palaeo-anthropology has developed strongly within the period since 1924, when the skull of the “Taung Child” (Figure 1) was discovered at the lime-mining site of Taung (meaning “the place of the lion”) in the area that is now the North West province of South Africa, approximately 200 km north-west of the city of Kimberley (Figure 2). In the pages of Nature, Raymond Dart (1925) described it as a new genus and species, Australopithecus africanus (the “ape-like creature from South Africa”), recognising it as a Plio-Pleistocene hominin despite criticism from palaeontologists who had accepted the fraudulent “Piltdown Man”, prior to its exposure in 1953 (Dart and Craig, 1959). The “Taung Child” was ape-like in the sense of having a small brain, but human-like in the sense of having small canines with no diastema in the mandibular and maxillary dental arcades. The date of this partial cranium including jaw, face and brain endocast, is uncertain, but it is considered to be at least 2.5 million years.

2Since 1925, there has been an exponential increase in the discovery of Plio-Pleistocene fossils from South Africa, associated with notable palaeontologists including not only Raymond Dart but also Robert Broom, John Robinson, Phillip Tobias, Ron Clarke, Bob Brain and Elisabeth Vrba. Subsequently (generally after 1990) a new generation of palaeontologists in South Africa included Lee Berger, Colin Menter, José Braga, Dominique Gommery, Frank Sénégas, Sandrine Prat, Francis Thackeray, Kris Carlson, Dominic Stratford, Travis Pickering, Jason Heaton, Morris Sutton, Justin Adams, Stephany Potze, Lazarus Kgasi, Christine Steininger, Brian Kuhn, Job Kibii and a growing number of young individuals who ardently study new hominin fossils as fast as they are discovered by speleologists such as Pedro Boshoff and his colleagues.

3Hominin specimens from South Africa have been attributed to at least eight species (Table I), including not only Australopithecus africanus (from Taung, Sterkfontein and Makapansgat) but also Australopithecus prometheus (Sterkfontein and Makapansgat), Australopithecus sediba (Malapa), Paranthropus robustus (Kromdraai, Swartkrans, Sterkfontein, Coopers Cave, Drimolen, Gondolin), early Homo (e.g. Kromdraai) similar if not identical to H. habilis from Olduvai Gorge in Tanzania, Homo naledi (Rising Star), Homo erectus (Swartkrans), Homo heidelbergensis (Elandsfontein), early "Homo sapiens" (here referred to as H. sapiens helmei from Florisbad, similar to Homo sapiens idaltu from Herto in Ethiopia) and Homo sapiens sapiens (Klasies River and Border Cave).

4Plio-Pleistocene caves from which hominin species have been recovered are within the Cradle of Humankind World Heritage Site (Figure 2) which was recognised by UNESCO in 1999. South African Late Pleistocene hominin sites associated with the behavioural and anatomical appearance of Homo sapiens (including Blombos and Klasies River) are currently being proposed for World Heritage status.

2 The South African palaeontological and archaeological sites

2.1 Makapansgat

5Makapansgat is situated in the Limpopo Province, north of the Gauteng Province where Sterkfontein is located (Figure 2). As at Taung, it includes several localities which were mined for lime in the 1920’s. In addition to the “Makapansgat Limeworks” area, other fossiliferous sites include the Cave of Heaths, Buffalo Cave and Historic Cave (otherwise known as Gwasa’s Cave).

6In 1925, William Eitzman gave Raymond Dart (University of the Witwatersrand) a number of blocks of breccia from the Makapansgat Limeworks Deposits (MLD). Hominids from such breccia were initially described by Dart (1948) as Australopithecus prometheus, circa 3 million years old. The species nomen “prometheus” was derived from the Greek god of fire. Dart had assumed that bones coated with black manganese dioxide related to the use of fire, but this was found to be incorrect. Specimens attributed to A. prometheus were later named A. africanus (also found at Sterkfontein and Taung), but Ron Clarke has resurrected the name “prometheus” to refer not only to hominin specimens from Makapansgat, but also to the “Little Foot” cranium from Sterkfontein (Stw 573) and other crania (Sts 71, Stw 252 and Stw 505), representing a second species of Australopithecus found at both Makapansgat and Sterkfontein.

7Dart examined postcranial bones, horns and teeth of various antelopes, and suggested that they had been used as tools or weapons. He adopted the name “osteodontokeratic” (ODK) for a presumed culture. Brain discredited this in a taphonomic study which indicated that natural breakage processes accounted for the patterns that were misinterpreted by Dart as indicators of hominin behaviour. However, at the younger site of Swartkrans, Brain (1981) demonstrated that bone and horn tools had in fact supplemented the use of stone tools in the early Pleistocene in South Africa, between one and two million years ago. As mentioned below, Brain and Sillen (1982) also demonstrated the controlled use of fire at Swartkrans, circa 1 million years ago.

2.2 The Cradle of Humankind

8In 1999, UNESCO declared an extensive area (“Sterkfontein, Swartkrans, Kromdraai and environs”) as a World Heritage Site, including at least 12 cave localities. Several of these are described below.

2.2.1 Sterkfontein

9As recognised by Thackeray (in press), three historical periods of exploration at Sterkfontein can be documented. The first episode dates between 1895 and 1935, when the limestone caves were explored by prospectors interested in the mining of calcium carbonate that was used partly for the processing of gold. G. Martignalia was one of the prospectors who used dynamite to blast dolomite and limestone at Sterkfontein in about 1895. This mining activity led to the discovery of fossils in calcified sands called breccia (Draper, 1896). Some of the fossil samples were sent to the British Museum (Natural History). It was not until seven decades later (after the exposure of the “Piltdown Man” hoax or joke) that Kenneth Oakley reported on the occurrence of fossils of a baboon, small carnivore, large equid, porcupine, rodents, small lizard and birds (Brain, 1981).

10An extinct baboon (Parapapio) was discovered at Sterkfontein in 1935 by Trevor Jones, who was at that time a student of Professor Raymond Dart at the University of the Witwatersrand. This fossil was described as Parapapio broomi, in honour of Robert Broom. More fossil primates were found at Sterkfontein in 1936 by G.W.H. Schepers and H. le Riche. Broom was taken to the site, and on this initial visit he picked up yet another fossil baboon, turned to Trevor Jones and said “I will name this one in honour of you”, whereupon it was described as Parapapio jonesi. Amélie Beaudet has recently undertaken micro-CT scans of the endocasts of fossil baboons (Beaudet et al., 2016), including the type specimens of P. broomi and P. jonesi, raising questions as to whether these two specimens represent the same species.

11On 17 August 1936, the Sterkfontein mine manager (George Barlow) showed Broom an australopithecine endocranial cast (Sts 60). After a short period, Broom (1936) found associated elements of the same cranium (TM 1511), at one time called Plesianthropus but now referred to as Australopithecus africanus. This marks the beginning of the second historical episode of fieldwork and research (between 1936 and 1966), when the Sterkfontein caves were the subject of scientific exploration and research by Robert Broom, his young assistant John Robinson, and Bob Brain who were all based at the Transvaal Museum (now the Ditsong National Museum of Natural History). In this period more than 100 Plio-Pleistocene hominin fossils were discovered, including an almost complete skull nicknamed “Mrs Ples” (Sts 5), and a partial skeleton (Sts 14) (Broom, 1947, 1950; Robinson, 1972). In addition, stone tools were discovered. These fossils are curated at the Transvaal Museum where Broom was based from 1934 until his death in 1951. Broom’s palaeo-anthropological activities were continued at the museum by John Robinson, Bob Brain and Elisabeth Vrba.

12The third historical episode at Sterkfontein (1966 to the present) includes a period when many palaeontologists from the University of the Witwatersrand worked at the site, notably Phillip Tobias, Alun Hughes, Tim Partridge, Ron Clarke, Kathy Kuman and Dominic Stratford. Systematic fieldwork and research (e.g. Tobias and Hughes (1969); Tobias (1973); Hughes and Tobias (1977); Clarke and P.V. Tobias (1995); Clarke, R.J. (1998, 2008)) led to remarkable discoveries of fossils and artefacts, initially under the aegis of the Department of Anatomy and later the School of Anatomical Sciences where Phillip Tobias was based for more than 60 years.

13Having renewed excavation in 1966, Tobias established a Sterkfontein Research Institute, with Alun Hughes supervising fieldwork. Ian Watt undertook detailed surveys with Tim Partridge (1978) who recognised six litho-stratigraphic Members in the Sterkfontein sequence. “Mrs Ples” (Sts 5) (Figure 3) and a large male australopithecine cranium (Stw 505) were associated with Member 4, whereas a specimen initially attributed to Homo habilis (Stw 53, now regarded by Ron Clarke as Australopithecus) was considered to be associated with Member 5. Stone artefacts were analysed by Kathy Kuman and her colleagues, who have recognized both Oldowan and Acheulian industries.

14An almost complete australopithecine skeleton and skull (StW 573), nicknamed “Little Foot” (Clarke and Tobias, 1995; Clarke, 1998; Partridge et al., 1999, 2003) and dated at 3.67 million years ago (Granger et al., 2015), was discovered by Ron Clarke, Stephen Motsumi and Nkwane Molefe from Member 2 in the Silberberg Grotto. Clarke (2008) believes that this specimen represents a “second species” of Australopithecus (A. prometheus) at Sterkfontein, distinct from A. africanus but having affinities with australopithecine fossils from Makapansgat (dated circa 3 million years), first described by Dart as A. prometheus.

15Initial attempts to estimate the age of Sterkfontein cave deposits were made using biostratigraphy, based on primates, suids, bovids and carnivores, compared to fossils from radio-metrically lake and river deposits in Kenya and Tanzania. Vrba (1976) made major contributions to an understanding of chronology and palaeo-ecology, based on a study of bovids from the Sterkfontein valley. White and Harris (1977) used suids for biostratigraphic purposes. Uranium-lead (U/Pb) and palaeomagnetic dates, in addition to faunal data, have contributed to an understanding of chronology. Member 4 is certainly older than 2.1 million years. Member 5 is thought to be around 1.8 million years old. Efforts continue to be made, by Laurent Bruxelles and others, to obtain accurate dates for the Sterkfontein deposits and associated fossils.

16The Institute for Human Evolution (IHE) was established in 2007 with Trefor Jenkins as Interim Director, succeeded by Francis Thackeray as Director (2009-2013). The Evolutionary Studies Institute (ESI) was subsequently formed under the direction of Bruce Rubidge (2013 to 2017). The IHE and ESI facilitated the continuation of research and excavations at Sterkfontein, now being undertaken by Dominic Stratford and colleagues.

2.2.2 Kromdraai

17Kromdraai is the site where the holotype specimen of Paranthropus robustus (Figure 4) was discovered in 1938. The original discovery was made by Gert Terblanche, a schoolboy who had known of Broom’s interest in hominins. The boy had served as a guide in the tourist caves at Sterkfontein, and had given a single molar hominid tooth from Kromdraai to George Barlow, the manager of the lime mine at Sterkfontein. Barlow in turn passed the tooth on to Broom, who recognized that it was not a “gracile” hominin specimen of Australopithecus africanus of the kind that he had discovered with miners at Sterkfontein since 1936. For one thing, the breccia appeared different. Barlow eventually revealed to Broom that the new hominin tooth came from another cave deposit. Broom immediately tried to track down the schoolboy. The school was situated not far from the Kromdraai cave site, and Broom was delighted to find the boy who had in a trouser pocket “four of the most wonderful teeth ever seen in the world’s history”. Later that afternoon, Terblanche took Broom to the Kromdraai site where he had hidden a partial cranium and mandible. Broom (1938) described this material as a new genus and species, Paranthropus robustus.

18The type specimen of P. robustus (TM 1517) is believed to be older than the Olduvai Event recognized from a palaeomagnetic reversal beginning at 1.95 million years ago, as recognized in the Olduvai deposits in Tanzania (Thackeray et al., 2002). TM 1517 is smaller than the “Zinjanthropus” cranium (OH5, based at 1.8 million years), discovered by Mary Leakey in Bed I at Olduvai Gorge and described by Phillip Tobias (1967) as Australopithecus boisei, otherwise attributed to Paranthropus boisei.

19When Louis Leakey first saw the original specimens of robust australopithecines from South Africa, he admitted that they were so similar to the East African form that they could be considered to belong to the same species (Brain, pers. comm. to Thackeray). Morphometric analyses of TM 1517 from Kromdraai and OH 5 from Olduvai Gorge have been used to infer that these two specimens are associated with a high probability of conspecificity (Thackeray, 1997), despite differences in size.

20The type locality of Paranthropus robustus was referred to as “Kromdraai B” (KB), also known as the “hominin site”, initially considered to be distinct from “Kromdraai A” (KA), the so-called “faunal site” situated about 30 metres to the west of KB. KA has a much higher concentration of bone in calcified deposits, but as yet has not yielded hominin fossils.

21Broom spent some time working at Kromdraai B in 1941, when he discovered a juvenile mandible of P. robustus with deciduous dentition (Broom and Schepers, 1946). Between January and April 1947 (with the encouragement of the Prime Minister, Jan Smuts), Broom worked at Kromdraai A and discovered a diversity of fauna, including the type specimen of Dinofelis (a saber-tooth cat), baboons and antelope.

22Brain (1981) of the Transvaal Museum worked at Kromdraai B in 1955 and 1956, increasing the sample size of robust australopithecine hominins and non-hominin fauna. Bovid fauna was described by Elisabeth Vrba (1976) who also undertook excavations at Kromdraai B and discovered additional hominins associated with ungulates, primates and carnivores. She used the non-hominin fauna to draw inferences about palaeo-ecology, chronology and agents of accumulation (Vrba, 1976; Vrba and Panagos, 1982).

23Partridge (1982) divided the Kromdraai B deposits into five Members. The Kromdraai hominins were thought to be confined to Member 3. More recent excavations at Kromdraai and analyses of fauna as well as sediments have been conducted since 1993 (e.g. Thackeray et al., 1995; Thackeray, 1997; Thackeray et al., 2001; Braga and Thackeray, 2003). Recent fieldwork by Braga and Thackeray (2016), as part of a joint French-South African research effort, has greatly supplemented the number of hominin fossils, including Paranthropus and early Homo.

24As yet no hominins have been discovered at Kromdraai A, but stone artefacts from that site clearly indicate a hominin presence, associated with the early Acheulean or Oldowan technology (Kuman et al., 1997). A polyhedral core from Kromdraai B (KB5501) is likely to be Oldowan. It is not certain whether the Kromdraai B artefacts were manufactured by P. robustus or early Homo since both taxa are known to be represented in the KB deposits (Thackeray et al., 2001; Braga and Thackeray, 2003).

25On the basis of faunal comparisons and seriation (McKee et al., 1995), it appears that Kromdraai A may be generally younger than Kromdraai B. In terms of palaeomagnetic analyses of breccia and calcite, Kromdraai B is at least 1.95 million years old (Thackeray et al., 2002), but Kromdraai A deposits have normal polarity throughout (J. Kirschvink, pers. comm. to F. Thackeray), suggesting that the KA deposits are younger than 1.8 million years old.

26Recent fieldwork at Kromdraai indicate that the site is much more extensive than originally perceived (Braga et al., 2016) (Figure 5), and that KA and KB are probably part of a single cave system (Bruxelles et al., 2016).

2.2.3 Swartkrans

27The site is situated one kilometre west of Sterkfontein. Hominins from this site were first discovered in 1948 at a time when Robert Broom and John Robinson of the Transvaal Museum were extending their search for Plio-Pleistocene fossils from the dumps of breccia discarded by lime-miners (Brain, 1981).

28Bob Brain undertook long-term fieldwork at Swartkrans, begining in 1967, continuing for more than 20 years, firstly to recover fossils from the dumps, secondly to clear the area to expose in situ fossil deposits, and thirdly to undertake systematic excavations. Five Members have been identified at the site, and fossils of at least two hominin species have been found: Paranthropus robustus (Figure 6) and early Homo from Members 1, 2 and 3 (spanning a period between about 1 and 2 million years ago); Member 4 was thought (from Middle Stone Age stone artefacts) to date to the Late Pleistocene; Member 5 is Holocene.

29A notable discovery by Bob Brain and Andrew Sillen (1988) included burnt bones from Member 3, representing the controlled use of fire about 1 million years ago.

30Systematic excavations at Swartkrans have been undertaken more recently by Travis Pickering and Morris Sutton together with Bob Brain and Kathy Kuman, supplementing the assemblages of hominin fossils (primarily Paranthropus robustus, comprising about 90% of the Early Pleistocene samples, contrasting with a small sample (10%) of early Homo). Paranthropus robustus was evidently a victim of leopard predation, whereas early Homo (with the controlled use of fire) may have been able to ward off predators.

2.2.4 Cooper’s Caves

31The Cooper’s Caves are situated 1 km east of Sterkfontein (Broom and Schepers, 1946). Specimens attributed to Paranthropus robustus and Homo sp. have been recovered, from deposits dated within the period 1-2 million years ago (Steininger et al., 2008).

2.2.5 Drimolen

32The history of fieldwork and research at the Plio-Pleistocene of Drimolen begins with the discovery of the site by an explorer, Pedro Boshoff, who was employed for a period of time at the Transvaal Museum (now Ditsong National Museum of Natural History). Formal fieldwork at the site was undertaken by André Keyser who was based at the Geological Survey, with a background in the study of Permian fossils from the Karoo. He was assisted initially by enthusiastic French palaeontologists, José Braga and Dominique Gommery. Juvenile mandibles of Paranthropus robustus attracted great interest especially after a press conference that was held in Paris. Substantial funding for the fieldwork and research was obtained through the French Embassy in Pretoria. Subsequent work was undertaken by Colin Menter with support from Lee Berger, both associated with the University of the Witwatersrand (Keyser et al., 2000). An essentially complete cranium of a female Paranthropus robustus was a prized specimen (Keyser, 2000), together with the hominins which were discovered earlier by Braga and Gommery (Schwartz and Tattersall, 2005), and a fragmentary pelvis of P. robustus (Gommery et al., 2002). The fossil material has been substantially supplemented and continues to be studied by Colin Menter and his team (Adams et al., 2016).

2.2.6 Bolt’s Farm Cave System

33Bolt’s Farm Cave System is the name for an area of more than 1.3 square kilometres, on which there are more than 30 fossiliferous sites. The area was explored in 1936 by Robert Broom (1937) soon after he was appointed as a palaeontologist at the Transvaal Museum (now called the Ditsong National Museum of Natural History). The area was explored again by a “California Expedition” led by Camp (1948) with the assistance of F. Peabody in 1947 and 1948. Several sites were explored. Freedman (1957, 1965) described non-hominin primates. Basil Cooke (1991) published a map of almost 20 pit localities, and described suids (including a cranium of Phacochoerus modestus), carnivores and bovids. In the 1980’s Ugo Ripamonti collected breccia blocks from a quarry, yielding a specimen of Theropithecus (Gilbert, 2007).

34Between 1996 and 1999, a French team associated with the Palaeontology Expedition to South Africa (PESA) renewed exploration on Bolt’s Farm. In 1996, Brigitte Senut, Martin Pickford and Jacques Michaux discovered a site which they named Waypoint 160 (Figure 7), and which yielded new rodent species, including Euryotomys bolti (Sénégas and Avery, 1998) and Boltimys broomi (Sénégas and Michaux, 2000), indicating an age that might be as old as 4-4.5 million years.

35Within the period 2001-2003, the Human Origins and Past Environments (HOPE) team explored additional sites, including Femur Dump and Alcelaphine Site. A revised map of pits explored by the California Expedition was prepared (Thackeray et al., 2008). Femur Dump was recognised as the dump of Pit 23 and from which new specimens of Dinofelis, Cercopithecoidea and rodents were obtained (Gommery et al., 2008). The California Expedition discovered a mandible associated with a well preserved skull of Dinofelis barlowi at Pit 23 (Figure 8). Of particular importance was the discovery of Parapapio at Waypoint 160, dated within the period 4-4.5 million years ago. Although no hominins have as yet been found on sites at Bolt’s Farm Cave System, the potential for finding them is indicated by the discovery of Parapapio which has been found elsewhere in association with australopithecines in South Africa and East Africa.

36Since 2006, two projects developed: (1) The PAI Protea French-South African “Palaeo-anthropology and palaeoenvironmental studies in South Africa” programme, funded by the National Research Foundation (NRF, South Africa) and the French Ministries for Research and Foreign Affairs (2006-2007); (2) The PICS CNRS-NRF “Plio-Pleistocene hominids and palaeoenvironments of the Cradle of Humankind (South Africa) (2008-2010).

37The HRU (HOPE Research Unit), established in 2007 by Dominique Gommery, Frank Sénégas, Stephany Potze and Lazarus Kgasi, has been operating at Bolt’s Farm Cave System for almost ten years. It has led to the discovery of additional sites on Bolt’s Farm Cave System, including Milo’s Pit A & B as well as Brad Pit A, B & C. The discovery of Metridiochoerus andrewsi (Stage 1) at Milo’s Pit A suggests a date of between 2.58 and 3.04 million years (Gommery et al., 2012a).

38Taken together, the Bolt’s Farm Cave System sites cover a period within 0.9 and 4.5 million years. The fossil assemblages are exceptionally important since they have the potential to document palaeo-environmental changes within a period of transition between Australopithecus and Homo.

2.2.7 Minnaar’s Cave

39Minnaar’s Cave was first reported by Broom and Schepers (1946) as the Plio-Pleistocene site from which a fossil jackal (Canis mesomela) was discovered. The exact location has been uncertain until Gommery et al. (2012b) rediscovered the site, and collected an assemblage of material that included Dinofelis barlowi, hyaenids (probably Chasmoporthetes), jackals, Papio and bovid fossils (lechwe, tsessebe, sitatunga, sable as well as alcelaphines such as wildebeest and hartebeest), indicating an age of more than 2 million years, associated with a mosaic palaeo-environment.

2.2.8 Plover’s Lake

40Plover’s Lake (situated within 10 km north-east of Sterkfontein) includes deposits dating to circa 1 million years ago, with non-hominin fauna that is likely to have been accumulated by leopards as agents of accumulation (Thackeray and Watson, 1994). Lee Berger excavated Late Pleistocene deposits with Middle Stone Age artefacts of the kind associated with “anatomically modern” Homo sapiens (de Ruiter et al., 2008).

2.2.9 Gondolin

41Gondolin is situated 24 kilometres north-west of Sterkfontein in the Cradle of Humankind World Heritage Site. It was first excavated by Elisabeth Vrba with David Panagos in 1979. The Plio-Pleistocene faunal assemblages were described by Watson (1993), Adams and Conroy (2005) and Adams (2010), indicating an age of about 1.8 million years. The fauna includes ungulates that depend on grassland and woodland vegetation, reflecting a mosaic habitat.

42The first hominin alleged to have been discovered at Gondolin was an isolated premolar of Homo sapiens, curiously reported by David Panagos (personal communication to Elisabeth Vrba, and to Francis Thackeray in 1990) as having come from the calcified Plio-Pleistocene breccia. An investigation by Thackeray and Suzanne Ubick led to the recognition of an antimere (another isolated premolar, representing the same individual) in the collection of modern Homo sapiens at the Transvaal Museum. An allegation is that someone attempted to play a joke on Elisabeth Vrba, who had accepted Panagos’ claim that the hominin tooth was derived from the breccia. Fortunately (unlike the infamous Piltdown hoax), the joke was exposed by Thackeray before Vrba could publish on this “discovery” of a hominin allegedly from the calcified breccia of deposits from Gondolin. Among the potential suspects is Panagos who was known to have had a sense of humour, who prepared the Gondolin fossils from breccia, and who also fell into disagreement with Vrba before his death.

43Genuine hominin fossils from Gondolin include a large molar of Paranthropus robustus (Kuykendall and Conroy, 1999), comparable in size to molars of Paranthropus boisei from Tanzania and Kenya. It is interesting to note that the type specimen of Paranthropus boisei from Olduvai Gorge (OH 5, described by Phillip Tobias (1967)) is dated at 1.8 million years, coeval with the large molar attributed to Paranthropus robustus from Gondolin. The question arises as to whether specimens attributed to P. robustus and others attributed to P. boisei represent the same species, consistent with a view expressed by Thackeray (1997) who compared the type specimen of P. robustus from Kromdraai with the type specimen of P. boisei from Olduvai Gorge.

2.2.10 Haasgat

44Haasgat is a Plio-Pleistocene cave site situated near the northern limit of the Cradle of Humankind in the Gauteng Province (Keyser, 1991; Herries et al., 2014). It has yielded a hominin (Leece et al., 2016) in addition to a rich assemblage of other mammals, especially baboons (Adams, 2012).

2.2.11 Gladysvale

45Gladysvale, situated within 30 kilometres north-east of Sterkfontein, was explored by Robert Broom in 1936. André Keyser conducted exploratory work in 1998, extended by Lee Berger who discovered specimens attributed to Australopithecus africanus (from breccia that may date to 2.4 million years) and Homo sp. of uncertain age (Berger and Tobias, 1994). Remarkably, human hair was found in a hyaena coprolite, dated at circa 200,000 years before present.

2.2.12 Malapa

46Malapa is a site situated near Gladysvale, less than 30 km north east of Sterkfontein, and has yielded two skeletons attributed to Australopithecus sediba (Berger et al., 2010). MH 1 is a young adolescent male (Figure 9), while MH 2 is an adult female. The two individuals probably fell into the solution cavity, dropping more than 10 metres.

47The site has been dated at 1.97 million years (Dirks et al., 2010). The two specimens attributed to A. sediba are interesting in that the cranial capacity (circa 500 cc) is consistent with the brain size of other specimens attributed to Australopithecus, whereas the dentition is reminiscent of early Homo, and the relative length of the arms is similar to that of chimpanzees. The mosaic nature of the anatomy of A. sediba supports the view that there is no clear boundary between Australopithecus and Homo (Thackeray, 2016).

2.2.13 Rising Star

48Rising Star is situated within 3 kilometres south of Sterkfontein. It is the site of discovery of an extraordinary hominin species, Homo naledi, reported by Lee Berger et al. (2015) and Paul Dirks et al. (2015). An assemblage of more than 1500 fossils, representing 15 individuals, has been attributed to this species.

49The cave Rising Star was explored by a group of speleologists who ventured into an underground dolomitic solution cavity, through sinuous, narrow, tortuous passages leading to the Dinaledi Chamber where the hominin fossils were found, including fragments of skulls and postcranial bones. There were no carnivore or antelope fossils of the kind that are often found in cave assemblages such as at Sterkfontein, Swartkrans and Kromdraai.

50Among the questions related to Homo naledi are the following: 1) How old are the fossils attributed to Homo naledi?, 2) Is it true that the fossils represent a new species, distinct from all others that are recognised in Plio-Pleistocene contexts?, 3) What are the affinities of this species?, 4) How did the fossils get into the cave, if not by carnivores or the result of individuals falling into the solution cavities?, 5) Were carcasses really brought into the back of the cave, and deliberately deposited there as suggested by Dirks et al. (2015)?, 6) Was there really only one opening to the Dinaledi Chamber?

51Clear answers to these questions were not immediately available. It has been suggested that the specimens are most similar (in terms of cranial anatomy) to early Homo. Some of the postcrania suggest similarities with modern Homo sapiens, but H. naledi had a cranial capacity of about 500 cc or less (similar to cranial capacities of australopithecines). An implication of this amazing discovery is that it strengthens the view that there is no clear boundary between Australopithecus and Homo (Thackeray, 2016). Morphometric analyses of crania support the view that Homo naledi is warranted as a distinct species (Thackeray, 2015a, 2015b).

2.3 Other Middle and Late Pleistocene South African sites

52Other sites are of interest for purposes of understanding the evolution of the genus Homo in South Africa.

2.3.1 Elandsfontein (Hopefield or Saldanha)

53Elandsfontein (also known as the Hopefield or Saldanda site) is situated in the Western Cape province, about 200 km north of Cape Town. It has yielded part of a Middle Pleistocene cranium discovered by Keith Jolly and Ronald Singer in 1953, and attributed to Homo heidelbergensis (Rightmire, 1996).

2.3.2 Florisbad

54Florisbad is situated near Bloemfontein in the Free State province. It is an open (spring) site dated at about 260,000 BP, and is most famous for a partial cranium that was originally classified as Homo helmei (Dreyer, 1935), but which has more recently been attributed to Homo sapiens helmei (Thackeray, 2010), similar to Homo sapiens idaltu from Ethiopia dated circa 200,000 BP.

2.3.3 Klasies River and Border Cave

55Klasies River (situated within 300 km east of Cape Town, near Humansdorp in the Western Cape province), and Border Cave (situated on the border of Swaziland and Kwazulu-Natal, South Africa), have yielded Late Pleistocene specimens of “anatomically modern” Homo sapiens dated at circa 100,000 years before present, in association with Middle Stone Age artefacts (Singer and Wymer, 1982; Deacon, 1995).

3 Conclusion

56Since 1924 with the discovery of the Taung Child and its description by Raymond Dart in 1925, South Africa has played an important role in contributing towards an understanding of human evolution, supplementing in a substantial way the data obtained from Tanzania, Kenya, Ethiopia, Malawi and Chad. The recent discoveries of fossils from sites such as Sterkfontein, Swartkrans, Kromdraai, Bolt's Farm, Malapa and Rising Star are contributing remarkably to an appreciation of the complexity and diversity of hominin evolution in the Plio-Pleistocene. Excavations and research will certainly continue to progress in South Africa, focussing not only on studies of anatomy but also on palaeo-environments and the behaviour of hominins. There is great potential in extending palaeo-anthropological research from the Pleistocene into the Miocene, as reflected for example by exciting fieldwork on Bolt's Farm.

Acknowledgements

57The following organisations are thanked for support in the context of palaeo-anthropology in South Africa: The French Embassy in South Africa which has supported co-operation between France and South Africa since 1995; the AESOP and AESOP+ programmes coordinated by José Braga (University of Toulouse, France) and Lorna Holtman (University of the Western Cape, South Africa); the French Ministry of Foreign Affairs; the CNRS; the National Research Foundation and Ministry of Science and Technology (South Africa), and the DST/NRF Centre of Excellence for the Palaeosciences; the PAST Trust (Palaeontological Scientific Trust); the Andrew W. Mellon Foundation; the Ford Foundation; the University of the Witwatersrand; the Ditsong National Museum of Natural History (formerly Transvaal Museum); National Geographic and the Institut des Déserts et des Steppes (Paris). SAHRA (South African Heritage Resources Agency) is thanked for the excavation and export permits. I thank Dominique Gommery for the kind invitation to present this paper, and José Braga for his enthusiastic efforts at Kromdraai.