The prehistory of the Arabian peninsula: Deserts, dispersals, and
Transcription
The prehistory of the Arabian peninsula: Deserts, dispersals, and
Evolutionary Anthropology 21:113–125 (2012) ARTICLE The Prehistory of the Arabian Peninsula: Deserts, Dispersals, and Demography HUW S. GROUCUTT AND MICHAEL D. PETRAGLIA As a geographic connection between Africa and the rest of Eurasia, the Arabian Peninsula occupies a central position in elucidating hominin evolution and dispersals. Arabia has been characterized by extreme environmental fluctuation in the Quaternary, with profound evolutionary and demographic consequences. Despite the importance of the region, Arabia remains understudied. Recent years, however, have seen major developments in environmental studies and archeology, revealing that the region contains important records that should play a significant role in future paleoanthropological narratives.1–3 The emerging picture of Arabia suggests that numerous dispersals of hominin populations into the region occurred. Populations subsequently followed autochthonous trajectories, creating a distinctive regional archeological record. Debates continue on the respective roles of regional hominin extinctions and population continuity, with the latter suggesting adaptation to arid conditions. There exists a striking imbalance between the clear importance of Arabia and what is actually known about the region. The incorrect perception of Arabia as an unchanging Huw Groucutt is a doctoral candidate at the School of Archaeology, University of Oxford. His research focuses on the Middle Paleolithic of Arabia and Late Pleistocene hominin dispersals. He has recently conducted excavations in Saudi Arabia and is undertaking a comparative analysis of lithic technology in Africa and Southwestern Asia. Email: huw.groucutt@ rlaha.ox.ac.uk Michael Petraglia is Professor of Human Evolution and Prehistory, Senior Research Fellow, and Co-Director of Centre for Asian Archaeology, Art & Culture, School of Archaeology, University of Oxford. He is also a Senior Research Fellow, Linacre College (Oxford), and a member of the Human Origins Program, Smithsonian Institution. Email: michael.petraglia@rla ha.ox.ac.uk Key words: Arabia; paleoenvironments; Paleolithic; lithic technology; hominins C 2012 Wiley Periodicals, Inc. V DOI 10.1002/evan.21308 Published online in Wiley Online Library (wileyonlinelibrary.com). desert has certainly played a role in the lack of serious research. In some cases, cultural and political issues have hampered research, as, for example, in contemporary Yemen. Numerous archeological surveys have been conducted but, until recently, these have been unsystematic. Most prehistoric sites consist of lithic scatters from surface contexts. As a result, knowledge of Arabian prehistory often comes from sites potentially, but not necessarily, representing multiple phases of occupation, which lack absolute dates and paleoenvironmental information. Attribution to cultural phases has typically reflected typological analyses of biased collections. The absence of a pre-Holocene hominin fossil record in Arabia precludes definitive identification of the manufacturers of the widespread lithic industries. These problems must be recognized and used to guide future research. Nevertheless, the Arabian record constitutes an important and understudied dataset. The recent discovery of stratified archeological sites in Saudi Arabia, Yemen, Oman, and the United Arab Emirates demonstrates the research potential of the region, as do increasingly detailed environmental records. Early western explorers of Arabia recognized stone tools, often in association with ancient lake beds.4 These early findings were followed by the discoveries of archeologists,5 which significantly advanced from the 1970s onward, with large-scale surveys such as the Comprehensive Survey of the Kingdom (1976-1981) in Saudi Arabia.6,7 Analogous developments took place in southern Arabia.8–11 Over the last decade, systematic surveys and interdisciplinary excavations have greatly increased our understanding of prehistoric Arabia,2,3,12–14 as have developments in paleoenvironmental studies.15–17 Research is now extending to the coastal waters around Arabia18 and to the use of remote sensing techniques.19 In the paleoanthropological literature, the Arabian Peninsula often serves as a useful blank on the map in which to draw hypothetical and rather abstract dispersal arrows. This has particularly reflected patterns of global genetic variation.20,21 In such models, the specific paleoenvironmental and biogeographical characteristics of Arabia have little role, yet it is precisely such contexts that are critical to defining patterns of hominin dispersal and adaptation. As genetic studies of Arabian populations have increased in scale, they reveal a complex pattern.22–25 Such studies show that modern Arabian populations are mostly derived from Western Asia, reflecting dispersals since the Last Glacial Maximum (LGM). In some areas, however, there are relatively high levels of ‘‘African’’ lineages, which have generally been attributed to historical 114 Groucutt and Petraglia ARTICLE Figure 1. Geography of the Arabian Peninsula. The locations of Pleistocene fossil localities and key sites for paleoenvironmental reconstruction are shown; note the southern and coastal concentration of the latter. Black lines depict major paleorivers.28 (Basemap courtesy of NASA’s Earth Observatory.) [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] processes such as slavery.26,27 Genetic evidence, then, is poorly placed to elucidate the position of Arabia in the dispersal of hominin populations. Likewise, the extinction of regional populations means they will not be represented in contemporary genetic structure. In this situation, archeology, in the context of paleoenvironmental fluctuation, offers a key way to elucidate the dispersal of hominin populations into Arabia and their subsequent evolutionary and cultural trajectories. GEOGRAPHY AND CLIMATE The vast size of Arabia, at more than 3 million km2, suggests that hominin evolutionary processes have been complex and regionally variable. The land mass consists of a range of topographical and environmental settings, including highlands, lowlands, a coastline some 7,000 km long, and an interior with numerous paleorivers and paleolakes (Fig. 1). Understanding the variable topography, ecology, and geography of Arabia is therefore a necessary background to elucidate changes in hominin demography.28,29 To the north of Arabia there are no fundamental barriers to hominin dispersal, given favorable paleoenvironmental conditions. The Sinai Peninsula offered a connection to Africa.30 On the eastern edge of Arabia, the Persian Gulf today is a shallow sea but, for most of the Pleistocene, was a large river valley, leading some to suggest that it formed an important population refugia.31 The possibility of hominins crossing the Bab al Mandab has been much debated.32 Today the Red Sea here is less than 30 km wide with small islands, meaning that the maximum single water crossing is less than 18 km. Isotopic studies suggest that there has not been a land bridge since the Miocene,33 but during times of low sea level the gap would have been significantly reduced. However, Arabia was concordantly arid during these times. Consequently, some have stressed the temporal lag between environmental amelioration and sea level rise2; others emphasize the hypothesis of a significant role for ‘‘coastal oases.’’34 Modern observers are familiar with the dry side of Arabia, most iconically demonstrated by the Rub’ al Khali, which, at 600,000 km2, forms the world’s largest sandy desert. In the east, the Wahiba Sands area is of importance, having been well studied in paleoenvironmental terms.35,36 Contemporary Arabia, however, has some areas, such as the Asir-Yemen highlands and Oman’s Dhofar, that receive considerably more precipitation, The landscape of Arabia reveals numerous signs of more humid conditions in the past. Groucutt and Petraglia 115 ARTICLE Vast paleoriver systems cut across the Peninsula, most starting in the western highlands and extending east to the Persian Gulf (Fig. 1). For instance, the now dry Wadi al Batin drained a large area of northern Saudi Arabia and left a huge alluvial fan around modern Kuwait, some 300 m deep and covering 190 by 130 km.28 Such paleorivers can clearly be seen in satellite images.37 Paleolakes of various sizes are found across Arabia, from the relatively small to the vast, such as the 2,000 km2 Mudawwara paleolake on the border of Saudi Arabia and Jordan.13,38,39 The environmental evolution of Arabia primarily reflects the varying contributions of precipitation from two key weather systems: winter rains from the Mediterranean and, most importantly, the summer rains of the Indian Ocean monsoon system. The latitudinal movement of the Inter-Tropical Convergence Zone (ITCZ), and with it the monsoonal rains, is the single most important environmental variable in Arabian prehistory. As the ITCZ moved north, Arabia sprang to life as lakes refilled and rivers flowed. Paleoenvironmental scientists have made considerable progress in elucidating this and related processes.15–17,40 Environmental oscillation between wet and dry periods can be studied at two levels, first at a broad ‘‘glacial’’ (generally dry) and ‘‘interglacial’’ (with wet phases) level and second in terms of more short-term change. Figure 2 gives an overview of environmental change in Arabia and related archeological evidence. At a broad level, evidence indicates significant increases in precipitation in each interglacial period back to MIS 11.41 However, most of our knowledge covers the last 350 kyr, and primarily the younger half of this, as shown in Figure 2.15,17 Glacial periods were often characterized by arid conditions. There are, however, complications to this dichotomy. In MIS 6, for instance, recent evidence demonstrates at least short phases of increased precipitation,15 which contrast with traditional views of this as a hyperarid period.42 MIS 5, particularly MIS 5e, 5c, and 5a, has been shown by various studies to have been particularly wet, with extensive speleothem, calcrete, and lake formation.13,17,43 A pronounced deterioration occurred with MIS 4; however, few records are known for this period, so we are dealing primarily with an absence of evidence (for example, of speleothem formation). MIS 3 has been labeled the ‘‘debated pluvial.’’15 Many radiocarbon dates had suggested a prolonged humid period 30-25 ka,44,45 but this is not The latitudinal movement of the InterTropical Convergence Zone (ITCZ), and with it the monsoonal rains, is the single most important environmental variable in Arabian prehistory. As the ITCZ moved north, Arabia sprang to life as lakes refilled and rivers flowed. reflected in records such as speleothems and probably reflects problems with the original radiocarbon dates.15,46 A recent study of paleolakes demonstrated that some key examples previously dated to MIS 3 actually date to MIS 5.16 MIS 2, encompassing the LGM, seems to have been generally extremely arid.35,47 The most recent major humid period dates to 10-6 ka. This ‘‘Holocene wet phase’’48–51 is best known from southern Arabia, where most studies have been conducted, but precipitation in northern Arabia was more than 300% greater than at present.52 A period of aridification followed, with evidence of abrupt drying.53 Several recent studies have demonstrated, within this broad pattern of glacial-interglacial oscillation, the existence of short-term wet phases. Presumably, some of the longer wet periods also included short dry phases. Recent studies in Saudi Arabia and the United Arab Emirates demonstrate a short-lived wet phase at 55 ka.39,54 Another short pluvial period occurred 14 ka, not long after the LGM.55,56 Much progress has been made in elucidating paleoenvironmental change in Arabia, but many questions remain. A key question concerns the relative contributions of the two weather systems. This has implications with regard to the opening of dispersal routes. Within interglacial periods, what were the dry periods, such as MIS 5b and 5d, like? And what was the impact of the short, wet phases in terms of demography and dispersals? Only recently has the chronology of these shortterm fluctuations begun to be understood. While records such as speleothems provide excellent overall records, they seem to not be sensitive enough to detect short-term pluvial episodes. FLORA AND FAUNA Plants and animals provide information that is relevant to understanding hominin occupations. The contemporary biome of Arabia reflects its position as an interface between biogeographical zones, particularly the Palearctic and Afrotropical. Changing environmental conditions increased the influence of one or the other of these, with the resulting mix then filtered through arid phases that gave a distinctly Arabian character to many taxa. One particularly important example are the baboons, Papio hamadryas, which are found in southwestern Arabia. These are the only wild baboons found outside sub-Saharan Africa. The most recent genetic study to examine the timing of baboon dispersal into Arabia reported coalescence ages of 333 ka, 216 ka, and 105 ka.57 While such dates come with rather wide error margins, they significantly coincide with interglacial periods (Fig. 2). It has been claimed that the northerly Arabian populations are more genetically similar to African populations,57 perhaps suggesting a Sinai dispersal 116 Groucutt and Petraglia ARTICLE Figure 2. Arabian environmental, faunal, and archeological records over the last 150 kyr. Thick vertical blue lines represent key humid periods, thin grey lines represent short humid episodes. Note the correlation of archeological assemblages with periods of increased rainfall, with the possible exception of Assemblage A at Jebel Faya. Environmental information and figure structure adapted from Rosenberg and coworkers16; chronology of baboon dispersal after Fernandes57; key archeological sites.2,3,13,14,92 [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] route. However, another study observed the opposite pattern,58 although issues with sample size are notable. Through such research, the wider floral and faunal context of hominin dispersals into and occupation of Arabia can be elucidated. Arabia has a rich Miocene fossil record,59 but few younger fossils are known. Thomas and colleagues60 discovered Pleistocene fossils (n ¼ 139) at three paleolake contexts in the An Nafud desert. They suggested that the fauna had an Afro-tropical character. Taxa represented include fish, tortoise, carnivores (Crocuta crocuta, Panthera cf. gombaszoegensis, Vulpes cf. vulpes), elephant, horse, hippo (Hexaprotodon sp.), and giant buffalo (Pelorovis cf. oldowayensis). The represented taxa and isotopic values from teeth indicate a savannah-like environment around the lakes.60 The oryx present at the An-Nafud fossil localities has been described as a specifically arid-adapted form,61 indi- cating the potential survival of certain mammals in dry periods. The discoverers’ suggestion of an Early Pleistocene date must be taken with caution, as it was not based on absolute dates; moreover, the three localities need not be synchronous. McClure62 reported fossils of a variety of species, including gazelle, hippopotamus, oryx, ostrich, and buffalo from the area of the Mundafan paleolake, which was recently redated to MIS 5.16 Many sites preserve Holocene faunal remains.63 THE LOWER PALEOLITHIC Arabia has a rich Lower Paleolithic record (Figs. 3 and 4).7 Southwestern Saudi Arabia seems to have a particularly large number of Lower Paleolithic sites, although it is unclear to what extent this reflects survey and publication bias. In comparison to younger Paleolithic sites, those of the Lower Paleolithic appear to extend further into the western and northwestern fringes of the Rub’ al Khali desert, perhaps indicating the occurrence of more extended wet phases in the earlier Pleistocene. The paleoenvironmental conditions of the Early Pleistocene are unclear, but presumably followed a pattern of oscillation similar to that seen more recently, but this was probably less extreme in its fluctuation. Unfortunately, few excavations have taken place, so that the chronology and internal variability of the Lower Paleolithic is essentially unknown. Localities appear to be correlated with hilltops, ridges, and terraces, and are often located in close proximity to raw material sources. Several claims have been made for ‘‘Oldowan-like’’, or ‘‘Mode 1’’ assemblages in Arabia.64–69 For instance, at 201-49 near Shuwayhitiyah in northern Arabia, Whalen and colleagues collected lithics from 16 localities.66 Their collection was dominated by heavy-duty tools (choppers, polyhedrons, and such), with a fairly small bifacial component (Fig. 3). Likewise, claims for ‘‘pre-Acheulean’’ sites have been made in Yemen. From AlGuza Cave, Amirkhanov68 reports an Oldowan ‘‘pebble industry’’ lacking a bifacial component. However, as with other purportedly Oldowan assemblages, care needs to be taken to differentiate artifacts from geofacts. Similarly, Jagher reports that intensive efforts to relocate the dozens of purportedly Oldowan sites around Huqf, Oman,70 did not result in the discovery of any artifacts.71 Chauhan67 reports the discovery of Oldowan-like lithics on Perim Island, a small island on the Yemeni side of the Bab al Mandab. If confirmed, this may suggest the crossing of the Bab al Mandab in the Early Pleistocene. The presence of hominins using Oldowan-like, or Mode 1, technology in Arabia is certainly possible, given the findings at sites such as Dmanisi to the north. However, problems pervade the Arabian finds. To take the example of the Shuwayhitiyah localities, the assemblages include lithics that are clearly recent (probably Holocene); in addition, the collection strategy of the discoverers is unclear, and the use of quartz as a raw material may be a significant factor Groucutt and Petraglia 117 ARTICLE Figure 3. Above: view from the top of the jebel at Dawādmi, looking west over Acheulean site 206-76 (where vehicles are located). This is the location of one of the first Paleolithic site excavations in Arabia. Acheulean sites and artifacts are distributed at the base of the jebel (over a distance of ca. 10 km) and below the andesite dykes, which served as raw material sources for stone tool manufacture. Below: selected handaxes from Dawādmi, manufactured from andesite. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] in determining artifact form. More generally, arguments based solely on typology carry many inherent limitations. Care needs to be taken to distinguish geofacts from artifacts. Also, excavations are required to recover chronometrically datable assemblages that can be placed in a paleoenvironmental context. If they are taken at face value, it is significant that more Oldowan-like sites are known in Arabia than in North Africa, perhaps indicating a significant role for Arabia in the early hominin colonization of Eurasia.72 While Whalen and colleagues believed that the Shuwayhitiyah collection is strongly reminiscent of the Developed Oldowan of East Africa,66 comparative technological analysis is required to elucidate the relationship of the Arabian Lower Paleolithic to that of surrounding regions. Also, excavations need to be done to clarify the demographic processes of populations in the Peninsula. We are on firmer ground with the Acheulean, as sites have been identified across Arabia. Spectacular ‘‘Acheulean landscapes’’ are found at Dawādmi and Wadi Fatimah, in central and western Saudi Arabia, respectively. At Dawādmi, various Acheulean sites were identified in association with an andesite and rhyolite dyke.73 Two sites here, 206-76 and 207-68, were subsequently excavated.74,75 The discovers rather speculatively suggested distinct ‘‘functional activity areas,’’ supposedly demonstrating eight different activities. Six uranium-thorium dates for calcareous matter adhering to the artifacts produced minimal ages of 201-61 ka.75:22 The surface collection and excavation of these sites produced thousands of lithics, but there have been no detailed technological comparisons of these lithics to assemblages from surrounding regions. An important factor at the Dawādmi sites is the presence of abundant evidence of early stage reduction. Giant cores demonstrate procurement and reduction of raw material along the dykes.7,76 The 32 Acheulean sites identified near the Red Sea in Wadi Fatimah appear,77,78 based on selective surface collections, to have a typological structure similar to those excavated at Dawādmi and also are located close to sources of raw material. In both cases, flakes are the dominant type, with characteristic bifaces and cleavers making up around 1% of the assemblages (Fig. 3). Uranium-thorium dating of calcareous nodules at site 210-351 produced an age ‘‘in the range of 200,000 years.’’78:78 The Dawādmi and Wadi Fatimah sites share similarities in technology and landscape position, being close to raw material sources and associated with riverine systems, which presumably functioned as dispersal corridors.76 The distinctive Acheulean bifaces and cleavers at these sites are generally characterized by deep flake scars; only rarely are they highly symmetrical (Fig. 3). Measurements of biface elongation (length to width ratio) comfortably place Arabian assemblages within the Acheulean, their mean elongations falling between those of assemblages in Africa and India.79 Whalen’s excavations revealed the presence of buried Lower Paleolithic sites in Arabia. Future excavations at these and other localities are likely to be highly informative. At present, the only absolute dates available reflect minimum ages.75,78 Comparative designations, typically classifying the Arabian material as ‘‘Middle Acheulean,’’74,75,78 are premature. Detailed statements on the spatial and temporal relationships of the Arabian assemblages to those of surrounding regions require considerably more research. 118 Groucutt and Petraglia ARTICLE Figure 4. Lower and Middle Paleolithic sites of Arabia. Note that some points represent groups of sites. (DEM data courtesy of the International Centre for Tropical Agriculture37; basemap image courtesy of Nick Drake and Paul Breeze, Kings College, London.) [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] THE MIDDLE PALEOLITHIC This is the most well-represented Paleolithic phase in Arabia (Fig. 4), and its study has seen significant recent advances. As in the Lower Paleolithic, sites are often associated with raw material sources and lacustrine and riverine systems, and are located at strategic positions in the landscape. Sites are found across Arabia except within modern sandy desert areas. The widespread surface record guided initial conceptions of Middle Paleolithic variability6 and can now be anchored to stratified sites excavated in the last five years. Rose and colleagues’ systematic survey in Oman’s Dhofar reveals the rich density of Middle Paleolithic sites found when modern survey techniques are applied.3 Traditionally, the Middle Paleolithic of Arabia has been viewed from a Levantine and European perspective, with sites often described as ‘‘Mousterian.’’ This inappropriate framework, combined with the traditional lack of stratified sites, made it difficult to address spatial and temporal variability and hampered the factoring of the Arabian record into debates such as that on Out-of-Africa dispersals. Nevertheless, important information was gathered at the hundreds of surface sites identified. These sites demonstrate the range of typo-technological variability found in the Arabian Middle Paleolithic and give some insight into the way hominins related to the landscape, while stratified sites have the benefits of a stronger chance of assemblage contemporaneity and allow both absolute dating and paleoenvironmental contextualization. The Arabian record demonstrates particular characteristics, including the often somewhat irregular, multiplatform character of cores, which is associated with features such as a general paucity of platform preparation. A significant feature of the Arabian Middle Paleolithic is the widespread presence of a bifacial component, which was confirmed by the excavation of Jebel Faya (Fig. 5).2 Rose80 has stressed the significance of bifacial technology in Oman, suggesting that it implies Late Pleistocene population connections with sub-Saharan Africa. Surveys have shown significant regional variation in Arabia. The production of points, for instance, seems to be variable, being common in Dhofar, among other sites,3 but seemingly rare in other areas. Zarins and coworkers,73 for example, perhaps reflecting their Levantine-oriented expectations, found the lack of points between Riyadh and Jeddah ‘‘disconcerting.’’ Similarly, the representation of Levallois technology appears to be variable in Arabia, although this, in part, reflects evolution from typological to volumetric conceptions. For instance, as Crassard points out, the ‘‘polyhedrons’’ and ‘‘discoids’’ illustrated by Whalen and Schatte81,82 are in fact recurrent centripetal Levallois cores. Various ‘‘industrial’’ designations have been proposed,6 but these are problematic. For instance, the possibility that at the single site of Bani Groucutt and Petraglia 119 ARTICLE Figure 5. Selected Arabian Middle Paleolithic artifacts. 1) Levallois core from RASA 2004149-1, Yemen81; 2) and 3) Levallois cores from Jubbah, Saudi Arabia13; 4) unifacially retouched point from Jubbah, Saudi Arabia13; 5) Levallois flake from assemblage C, FAYNE1, UAE2; 6) Levallois point from Aybyt Thania, Oman3; 7) Levallois point from Aybut Auwal, Oman3; 8) bifacial foliate from FAY-NE1, UAE.2 Khatmah, an assemblage dominated by tanged points and scrapers from the western fringe of the Rub’ al Khali relates to the otherwise North African Aterian needs to be tested by excavations and comparative technological analysis.83 A reanalysis of the Bani Khatmah lithics suggests that the Arabian ‘‘Aterian’’ may date to the Holocene.84 A report on the Middle Paleolithic site of Shi’bat Dihya 1 and others in the Wadi Surdud of western Yemen is to be published shortly.14 A key development is that this site is now dated to 55 ka, in contrast to initial estimates of 80-70 ka.12,85 More than 5,000 lithics having lengths greater than 2-cm, as well as faunal remains, were excavated from a single thin layer covering 21 m2. Isotopic and paleobotanical data indicate that the site was occupied during relatively arid conditions. The timing of occupation correlates with the short wet phase previously mentioned at 55 ka (Fig. 2), which occurred between more arid periods on either side. The lithic technology reflects a variety of reduction schemes for the production of flakes, blades, and points. Levallois technology is present, but not common. No immediate parallels are found with contemporary African or Levantine assemblages. Other sites in Wadi Surdud are dated to younger periods of MIS 3. The overall picture is of a distinctive regional variant, an autochthonous development from a presently unknown ancestral population. Elsewhere in Yemen, more Levallois-oriented point production perhaps hints at earlier connections with the Levant or may reflect convergent evolution in a similar habitat.11,81,86 At the site of Jebel Faya (FAYNE1), excavations uncovered stratified Paleolithic assemblages.2 Published information addresses the top three Paleolithic assemblages (labeled A to C with increasing depth). Assemblage C (500 lithics, see Fig. 5) is dated by optically stimulated luminescence (OSL) to 95 6 13 ka, 123 6 10 ka, and 127 6 16 ka, although the authors indicate that the latter date may be problematic.2 This last interglacial assemblage demonstrates a variety of reduction strategies. These include the production of volumetric blades and Levallois debitage, as well as bifaces, and a variety of retouched forms. Similar assemblages have been discovered nearby in surface contexts,87,88 showing that this technological package characterizes the Middle Paleolithic of the Arabian side of the eastern Persian Gulf, at least during MIS 5. While the closest technological parallels are found with northeast Africa, it is not impossible that assemblage C represents an autochthonous development, or indeed a dispersal from, for instance, the Indian subcontinent.89 In the overlying assemblages (A and B), bifacial and Levallois reduction is apparently absent except for a few convergent flakes in B, which are similar to Levallois points.2 Neither backed nor microlithic technologies are represented. Assemblage B is undated, but on stratigraphic grounds was deposited between 90 and 40 ka, while A is dated to 40.2 6 4.0 ka and 38.6 6 3.1 ka, confirming hominin occupation of eastern Arabia in MIS 3. Both assemblages B and A are characterized by diverse reduction strategies, centered on the production of flakes. In assemblage B, core types include Kombewa and radial, which are also found in assemblage A, which is dominated by orthogonal multiplatform cores. Neither A nor B has a substantial retouched component. A third spatially and temporally specific Middle Paleolithic industry is found in Dhofar, Oman. Rose and colleagues3 identified 110 sites with technology very similar to that of the MIS 5 Nubian Complex in northeast- 120 Groucutt and Petraglia ern Africa. Previously rare examples of Nubian-like cores had been identified,86 but the new discoveries constitute a widespread and seemingly homogenous industry. The sites are found exclusively on the Nejd Plateau and have not been identified near the coast. All of the sites take the form of surface assemblages, with the exception of Aybut Auwal. Here a small number of artifacts, including a diagnostic Type 1 Nubian core, were found in situ, while many other lithics appear to have eroded from the same sediments onto the surface. Two OSL samples from the same stratigraphic unit as the excavated lithics produced age estimates of 106 6 9 ka and 107 6 9 ka, correlating with MIS 5c. Lithics from Aybut Auwal and three other sites were analyzed in detail. At all of the sites, Nubian Type 1 cores are dominant. The core technology of the Dhofar Nubian is remarkably similar to the diagnostic cores of the Nubian Complex in North Africa. Debitage is dominated by flakes, although blades are also common. Aside from points, sidescrapers are the most numerous tool. A bifacial component was not identified at any site. In Oman, another industry, which Rose10 labeled the ‘‘Nejd Leptolithic,’’ is widespread. This is characterized by the production of large blades from unidirectional cores with little platform preparation. Blanks were rarely retouched. An interesting feature of this industry is the frequent presence of lipped platforms, which are often taken as an indication of the use of a soft hammer. Rose and colleagues3 suggest, on geomorphological grounds, that the Nejd Leptolithic is younger than at least some of the MIS 5 Nubian sites. These assemblages appear similar to some from Oman described by Jagher.71 We might see these large-blade assemblages as a distinctive regional post-MIS 5 development in Oman. The situation in Saudi Arabia is less clear than that in southern Arabia, where most recent research has focused. The authors of this paper are conducting research at the Jubbah paleolake in northern Saudi Arabia,13 an area highlighted by an initial reconnaissance survey in the ARTICLE 1970s.90 At the site of Jebel Qattar 1, we identified a lithic assemblage characterized by centripetal Levallois and discoidal reduction, in association with a calcrete dated by OSL to 75 6 5.0 ka. As well as notches, denticulates, and scrapers, we found an invasively retouched unifacial point (Fig. 5). Field work in 2011 led to the identification of several stratified Middle Paleolithic sites around the paleolake. Our analysis is ongoing and promises to cast light on the character of the Middle Paleolithic in interior northern Arabia. seen most clearly at Jebel Faya and Wadi Surdud, and perhaps with other examples such as the large blade assemblages of Oman, the outlines of post-MIS 5 regional autochthonous trends are coming into focus. So far, evidence of assemblages similar to the Howiesons Poort and other precocious facies of the MSA is absent in Arabia, while the predominantly unidirectionalconvergent production of Levallois points similar to that in the early and late phases of the Levantine Middle Paleolithic is known only from Yemen. THE LATE PALEOLITHIC As research gathers pace in Arabia, we are now beginning to be able to correlate paleoenvironmental and archeological records for the Middle Paleolithic and move toward comparisons with surrounding regions. As research gathers pace in Arabia, we are now beginning to be able to correlate paleoenvironmental and archeological records for the Middle Paleolithic and move toward comparisons with surrounding regions. At a broad level, features such as the presence of a bifacial component at many Arabian Middle Paleolithic sites arguably orients the Peninsula toward Africa, where a façonnage component was frequent in the Middle Stone Age (MSA).80 In contrast, the Levantine Middle Paleolithic seems to lack any evidence of façonnage technology. This regional divergence took place more than 200 kyr ago. The increasing evidence of variability in the Arabian Middle Paleolithic suggests that multiple dispersals into the area occurred, followed by regional autochthonous trajectories. Where enough data are available, the Arabian assemblages suggest connections to Africa in MIS 5.2,3 Subsequently, as Pre-Holocene evolutionary and cultural trajectories are poorly understood in Arabia. The character of the Late Paleolithic, by which we mean the period 40-10 ka, which elsewhere is typically associated with blade-dominated or microlithic assemblages and frequent evidence of symbolism and other complex behaviors, is one of the great mysteries of Arabian archeology. As discussed in a recent review,91 such evidence is ephemeral in Arabia. With the usual caveats in mind, such as a general reliance on surface sites, a generous interpretation suggests a possible Late Paleolithic component at a small number of sites focused in northwestern Arabia (Fig. 6). It must be stressed that such attributions are problematic. The only excavated and dated Late Paleolithic site, Al Hatab, has lithics unlike those of the Epipaleolithic/Upper Paleolithic elsewhere, and again appears to reflect an autochthonous southern Arabian development (Fig. 7).92 ‘‘Upper Paleolithic’’ sites have been defined by some on primarily typological grounds, emphasizing the presence of types such as burins and blades64,73; others stress ‘‘intermediary’’ levels of ‘‘patination, materials, and workmanship.’’ In Yemen, Amirkhanov9 used both of these perspectives to define a number of ‘‘Upper Paleolithic’’ sites. These sites need more study, particularly absolute dating and technological analysis. Superficially, many appear actually to be compatible with a Middle Paleolithic designation. One interesting assemblage is the material from Faw Well Groucutt and Petraglia 121 ARTICLE an autochthonous development of laminar technologies extending back to perhaps MIS 4.92 The assemblage includes hard hammer blades, bifacial foliates, burins, and endscrapers. The Arabian record seems to lack specific features relating to lithic technology and other features of the African Late Stone Age. There are possible hints at occasional contacts with the laminar industries of the Levantine Upper Paleolithic but, in the absence of detailed technological analyses, these are broad-scale comparisons. These features, such as the assemblage from Faw Well, perhaps suggest that dispersals from the Levant occurred during the short wet phases of MIS 3. It is possible that such populations survived through the LGM in southern Arabia. This is perhaps supported by the distinctively Arabian character of assemblages at sites like Al-Hatab, indicating regional autochthonous developments.92 It is possible that the sites in southern Arabia represent a continuation of occupation from MIS 5. Surface sites in the north and west are poorly studied, and indeed may even date to the Holocene. INTO THE HOLOCENE Figure 6. Late Paleolithic and Early/Middle Holocene sites of Arabia. Note that some points represent groups of sites. (DEM data courtesy of the International Centre for Tropical Agriculture37; basemap image courtesy of Nick Drake and Paul Breeze, Kings College, London.) [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] in southwestern Saudi Arabia. This assemblage, which seems to be unlike any other in Arabia, but lacks any absolute dates, is characterized by blade and microblade production. Edens93 studied part of the assemblage and suggests possible similarities with the Upper Paleolithic Ahmarian industry in the Levant. More classically ‘‘Epipaleolithic-like’’ sites have a position similar to those of the ‘‘Upper Paleolithic.’’ The more convincing examples are located in the far north of the Arabian Peninsula, perhaps indicating brief incursions from the Levant.64,94 The excavations at Al-Hatab in Oman uncovered a lithic assemblage dating to 13 ka and possibly extending into the earliest Holocene. That assemblage has typo-technological similarities to material from the sites of Ras Aı̈n Noor and Dhanaqr, which the discoverers suggest reflects The early to middle Holocene of Arabia has implications for general patterns of occupation and dispersals in Arabia, as well as adaptation to arid environments. Considerably more detailed archeological and environmental records are available for the Holocene than are for the Pleistocene. Many absolute dates are available and at least 28 sites preserve faunal remains.63 Widespread rock art provides new insights into human behavior, although its study would be helped by an evolution from the stylistic to more systematic and geographic information system (GIS)-based analyses. In addition, many of the key themes we have discussed in relation to the Paleolithic also characterize debates such as that about the origins of the ‘‘Neolithic’’ in Arabia. The large number of early to middle Holocene sites probably demonstrates a substantial demographic increase, associated in particular with 122 Groucutt and Petraglia ARTICLE Figure 7. Selected Arabian Late Paleolithic artifacts. 1) backed blade; 2) double platform core; 3) single platform core, Faw Well, Saudi Arabia93; 4) blade; 5) blade core, Al Hatab, Oman.92 the Holocene wet phase (Fig. 6). A number of autochthonous developments can clearly be seen in Holocene Arabia (Fig. 8). These include the development of new reduction strategies such as the ‘‘Wa’sha method’’ in Yemen11,95 and the development of other forms of projectile points in the Neolithic,96 including ‘‘Fasad points.’’97 Nonlithic developments include the domestication of animals and the development of seafaring.98 Debate continues on whether the Holocene occupation of Arabia reflects population dispersal into the area,63,97 probably from the Levant, or was a primarily indigenous development.31 In terms of lithic technology, for instance, the distinctive Fasad points of southern Arabia have been seen as a derivation from Levantine early Neolithic technology97 or as an autochthonous development from Arabian terminal Pleistocene traditions.92 The discovery of a single Fasad point dating to 13 ka at Al Hatab can be taken as supporting the latter position,92 but it is debatable how much a single, unique example of anything can show. The presence of apparently closely pre-Neolithic sites in Yemen is intriguing and again may also support the latter perspective.99 This may be a rather abstract dichotomy, as numerous demographic scenarios are possible, including both population continuity and population dispersals into the area. In fact, genetic evidence seems now to demonstrate clearly some population continuity from the terminal Pleistocene into the Holocene, with additional dispersals from the Levant at 10 ka.22 The Neolithic in Arabia has a distinctive character that reflects adaptations to the arid environment. The use of domesticated animals seems to have been of considerably greater importance than domesticated plants. Sometime later, perhaps around the third millennium BCE, the domestication of camels occurred in Arabia. Through such adaptations to arid environments, people were able to survive in extreme areas and began to transcend the ancient correlation between demography and environmental amelioration. DISCUSSION AND CONCLUSION In recent years there has been a pronounced acceleration in the pace of archeological research and discovery in Arabia, yet the region remains remarkably poorly understood. However, newly emerging information correlates the human and environmental stories (Fig. 2). We hypothesize that much of Arabian population history is cyclical; that is, with increased precipitation, hominin populations repeatedly expanded into the area. They subsequently followed autochthonous trajectories before becoming extinct, at least until the Late Pleistocene/Holocene, when adaptations to arid conditions developed. Given the severity of environmental fluctuation, which, as Figure 2 shows, is increasingly well understood, it is reasonable to assume that there has not been long-term population continuity in Arabia at an evolutionary scale. This emphasizes the need to understand the Arabian record in an interregional context. For the Lower Paleolithic it is too early to make definitive statements in this regard, but Homo heidelbergensis and Homo erectus probably played a role, as possibly did early Homo. There has been little focus on this phase in recent years, as attention has focused on modern humans, but there is clearly great potential to elucidate the Lower Paleolithic in Arabia. The key development in recent years has been the demonstration, particularly in Dhofar and at Jebel Faya,2,3 that Arabia contains Africanlike Middle Paleolithic assemblages dating to MIS 5. This strongly suggests that dispersals took place at this time, although fossil evidence would help to substantiate this hypothesis. For example, the archeological evidence from Dhofar suggests population dispersal from northeast Africa by MIS 5c,3 a time when baboons also seem to have dispersed into Arabia,57 and when paleoenvironmental evidence demonstrates a wet phase. This shows how an interdisciplinary perspective can elucidate the Paleolithic occupation of Arabia (Fig. 2). Further excavations are needed to distinguish between different hypotheses. For instance, does the occupation at Jebel Faya represent a continuous occupation of the Persian Gulf area between 125 and 40 ka, as the discoverers suggest,2 or does it rather indicate repeated dispersals into the area? Groucutt and Petraglia 123 ARTICLE Figure 8. Selected Arabian Early/Middle Holocene artifacts. 1) Neolithic arrowhead from Jubbah, Saudi Arabia; 2) fluted point from Manayzah, Yemen11; 3) Fasad point from Nad al-Thamam, UAE97; 4) Fasad point from FAY-NE1, UAE97; 5) Wa’shah core from HDOR 538, Yemen95; 6) bifacial foliate from HDOR 538, Yemen11; 7-10) Neolithic points from Khuzmum, Yemen11; 11) Wa’shah point from HDOR 538, Yemen.95 In general, indications of possible population connections to surrounding regions remain rather speculative, and this is compounded by the absence of fossil evidence. We suggest, however, that the emerging picture suggests a general lack of connections between Africa and Arabia after MIS 5, the last interglacial. Instead, there are perhaps indications that Arabia sometimes saw connections to the Levant. The modern Arabian genomic structure is dominated by lineages reflecting post-LGM population movements from the north.22,23 These dispersing populations mixed with existing populations in a manner that remains to be firmly understood. The Middle Paleolithic record is increasingly demonstrating, probably in common with other prehistoric phases, that after dispersals into the area, environmental deterioration divided populations into refugia. These include the Yemeni highlands and the Persian Gulf. The combination of regionalization, an arid environmental setting, and small population sizes will have all played a significant role in determining the character of human adaptations. To take the lithic evidence, variability will express demographic, raw material, functional, and cultural factors. The balance between these remains to be understood. Hence, relations to surrounding areas and between regions within Arabia remain obscure. The key point is that advances in empirical data need to be complemented by theoretical developments such as those of behavioral ecology, rather than by a simplistic or ‘‘empiricist’’ perspective that merely compares the basic morphology of lithic artifacts, for instance, while ignoring their context. To conclude, we highlight the significance of the Arabian record in casting light on the development of adaptations to extreme environments. The Arabian record is, in fact, uniquely positioned to elucidate such developments, as populations dispersed into the Arabian Peninsula from neighboring ‘‘hot spots’’ such as East Africa and were trapped by the desiccation of the routes they had originally followed. Given widespread public and scholarly interest in and concern about contemporary climate change, the long-term records of human occupations and environmental change in Arabia are of great importance. With recent advances, archeologists and our colleagues in related disciplines have begun to outline the chapters of the Arabian story. Now it is time to fill in the pages. ACKNOWLEDGMENTS We thank HRH Prince Sultan bin Salman bin Abdul-Aziz Al Saud, President of the Saudi Commission for Tourism and Antiquities (SCTA), and Professor Ali Al-Ghabban, Vice President of SCTA, for permission to 124 Groucutt and Petraglia conduct research in the Kingdom of Saudi Arabia. We acknowledge Abdullah Alsharekh for his unwavering support of our research, as well as Rémy Crassard and Adrian Parker for discussions on the Arabian evidence, and James Blinkhorn for commenting on a draft of this paper. We acknowledge the financial support of the Leakey Foundation, the National Geographic Society, and the Arts and Humanities Research Council (UK, doctoral studentship to H.G.). We appreciate the constructive comments of John Fleagle, John Shea, Frank Preusser, and three anonymous reviewers. REFERENCES 1 Petraglia MD, Rose JI, editors. 2009. The evolution of human populations in Arabia: paleoenvironments prehistory and genetics. Netherlands: Springer. 2 Armitage SJ, Jasim SA, Marks AE, et al. 2011. The southern route ‘‘out of Africa’’: evidence for an early expansion of modern humans into Arabia. Science 331:453–456. 3 Rose JI, Usik VI, Marks AE, et al. 2011. The Nubian complex of Dhofar, Oman: an African Middle Stone Age industry in southern Arabia. PLoS ONE 6:e28239. 4 Philby HSJ. 1933. Rub’ Al Khali: an account of exploration in the Great South Desert of Arabia under the auspices and patronage of his Majesty ‘Abdul’Aziz ibn Sa’ud, King of the Hejaz and Nejd and its dependencies. Geogr J 81:1–21. 5 Caton-Thompson G, Gardner EW. 1939. Climate, irrigation, and early man in the Hadhramaut. Geogr J 93:18–35. 6 Petraglia MD, Alsharekh AM. 2003. The Middle Palaeolithic of Arabia: implications for modern human origins, behaviour and dispersals. Antiquity 77:671–684. 7 Petraglia MD. 2003. The Lower Paleolithic of the Arabian Peninsula: occupations, adaptations, and dispersals. J World Prehist 17:141–179. 8 de Maigret A. 1986. Archaeological activities in the Yemen Arab Republic, 1986, East West 36:376–470. 9 Amirkhanov H. 2006. Stone age of south Arabia. Moscow: Nauka. 10 Rose JI. 2006. Among Arabian sands: defining the Paleolithic of southern Arabia. Ph.D. dissertation. Dallas:. Southern Methodist University. 11 Crassard R. 2008. La préhistoire du Yémen: diffusions et diversités locales à travers l’étude d’industries lithiques du Hadramawt. BAR, 1842. Oxford: Archaeopress. 12 Delagnes A, Macchiarelli R, Jaubert J. 2008. Middle Paleolithic settlement in Arabia: first evidence from a complex of stratified archaeological sites in western Yemen. Presented at the Annual Meeting of the Palaeoanthropology, Vancouver. 13 Petraglia MD, Alsharekh AM, Crassard R, et al. 2011. Middle Paleolithic occupation on a Marine Isotope Stage 5 lakeshore in the Nefud Desert, Saudi Arabia. Quaternary Sci Rev 30:1555–1559. ARTICLE 14 Delagnes A, Brenet M, Crassard R, et al. n.d. The Middle Paleolithic assemblage of Shi-bat Dihya 1 (Wadi Surdud site complex, Yemen). J Hum Evol. In review. 15 Parker AG. 2009. Pleistocene climate change in Arabia: developing a framework for hominin dispersal over the last 350 ka. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 39–49. 16 Rosenberg TM, Preusser F, Fleitmann D, et al. 2011. Humid periods in southern Arabia: windows of opportunity for modern human dispersal. Geology 39:1115–1118. 17 Fleitmann D, Burns SJ, Pekala M, et al. 2011. Holocene and Pleistocene pluvial periods in Yemen, southern Arabia. Quaternary Sci Rev 30:783–787. 18 Bailey GN, Flemming NC, King GCP, et al. 2007. Coastlines, submerged landscapes and human evolution: the Red Sea Basin and the Farasan Islands. J Island Coastal Archeol 2:127–160. 19 Kenney D. 2011. The ‘‘works of the old men’’ in Arabia: remote sensing in interior Arabia. J Archaeol Sci 38:3185–3203. 20 Quintana-Murci L, Semino O, Bandelt, HJ, et al. 1999. Genetic evidence of an early exit of Homo sapiens sapiens from Africa through eastern Africa. Nat Genet 23:437–44. 21 Macaulay V, Hill C, Achilli A, et al. 2005. Single, rapid coastal settlement of Asia revealed by analysis of complete mitochondrial genomes. Science 308:1034–1036. 22 Černý V, Mulligan CJ, Fernandes V, et al. 2011. Internal diversification of Mitochondrial Haplogroup R0a reveals post-Last Glacial Maximum demographic expansions in south Arabia. Mol Biol Evol 28:71–78. 23 Abu-Amero K, Ali H, Gonzalez A, et al. 2009. Saudi Arabian Y-chromosome diversity and its relationship with nearby regions. BMC Genet 10:59. 24 Abu-Amero KK, Larruga JM, Cabrera VM, et al. 2008. Mitochondrial DNA structure in the Arabian Peninsula. BMC Evol Biol 8:45. 25 Rı́dl J, Edens CM, Černý V. 2009. Mitochondrial DNA structure of Yemeni population: regional differences and the implications for different migratory contributions. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 69–78. 26 Abu-Amero KK, Gonzalez AM, Larruga JM, et al. 2007. Eurasian and African mitochondrial DNA influences in the Saudi Arabian population. BMC Evol Biol 7:32. 27 Richards M, Rengo C, Cruciani F, et al. 2003. Extensive female-mediated gene flow from sub-Saharan Africa into near eastern Arab populations. Am J Hum Genet 72:1058–1064. 28 Edgell HS. 2006. Arabian deserts: nature origin and evolution. Netherlands: Springer. 29 Vincent P. 2008. Saudi Arabia: an environmental overview. London: Taylor and Francis. 30 Waldmann N, Torfstein A, Stein M. 2010. Northward intrusions of low- and mid-latitude storms across the Saharo-Arabian belt during past interglacials. Geology 38:567–570. 31 Rose JI. 2010. New light on human prehistory in the Arabo-Persian Gulf Oasis. Curr Anthropol 51:849–883. 32 Bailey G. 2009. The Red Sea, coastal landscapes, and hominin dispersals. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 15–138. 33 Fernandes CA, Rohling EJ, Siddall M. 2006. Absence of post-Miocene Red Sea land bridges: biogeographic implications. J Biogeogr 33:961– 966. 34 Faure H, Walter RC, Grant DR. 2002. The coastal oasis: Ice Age springs on emerged continental shelves. Global Planet Change 33:47– 56. 35 Preusser F, Radies D, Matter AA. 2002. 160,000-year record of dune development and atmospheric circulation in southern Arabia. Science 296:2018–2020. 36 Radies D, Preusser F, Matter A, et al. 2004. Eustatic and climatic controls on the development of the Wahiba Sand Sea, Sultanate of Oman. Sedimentology 51:1359–1385. 37 Jarvis A, Reuter HI, Nelson A, et al. 2008. Hole-filled seamless SRTM data V4, International Centre for Tropical Agriculture (CIAT). Available online at http://srtm.csi.cgiar.org. 38 Petit-Maire N, Carbonel P, Reyss JL, et al. 2010. A vast Eemian palaeolake in southern Jordan (298N). Global Planet Change 72:368–373. 39 Parton A, Parker AG, Farrant AR, et al. 2010. An early MIS3 wet phase at palaeolake Aqabah: preliminary interpretation of the multi-proxy record. Proc Sem Arabian Studies 40:267–276. 40 Preusser F. 2009. Chronology of the impact of Quaternary climate change on continental environments in the Arabian Peninsula. C R Geosci 341:621–632. 41 Blechschmidt I, Matter A, Preusser F, et al. 2009. Monsoon triggered formation of Quaternary alluvial megafans in the interior of Oman. Geomorphology 110:128–139. 42 Anton D. 1984 Aspects of geomorphological evolution: paleosols and dunes in Saudi Arabia. In: Jado AR, Zötl JG, editors. Quaternary Period in Saudi Arabia. II. Sedimentological, hydrogeological, hydrochemical, geomorphological, and climatological investigations of western Saudi Arabia. Vienna: Springer Verlag. p 275– 296. 43 Rosenberg TM, Preusser F, Blechschmidt I, et al. 2012. Late Pleistocene palaeolake in the interior of Oman: a potential key area for the dispersal of anatomically modern humans outof-Africa? J Quaternary Sci 27:13–16. 44 McClure HA. 1976. Radiocarbon chronology of late Quaternary lakes in the Arabian Desert. Nature 263:755–756 45 Schulz E, Whitney JW. 1986. Upper Pleistocene and Holocene lakes in the An Nafud, Saudi Arabia. Hydrobiologia 3:175–190. 46 Immenhauser A, Dublyansky YV, Verwer K, et al. 2007. Textural, elemental, and isotopic characteristics of Pleistocene phreatic cave deposits (Jabal Madar, Oman). J Sediment Res 77:68–88. 47 Clark ID, Fontes JC. 1990. Paleoclimatic reconstruction in northern Oman based on carbonates from hyperalkaline groundwaters. Quaternary Res 33:320–336. 48 Radies D, Hasiotis ST, Preusser F, et al. 2005. Paleoclimatic significance of Early Holocene faunal assemblages in wet interdune deposits of the Wahiba Sand Sea, Sultanate of Oman. J Arid Environ 62:109–125. 49 Neff U, Burns SJ, Mangini A, et al. 2001. Strong coherence between solar variability and the monsoon in Oman between 9 and 6 kyr ago. Nature 411:290–293. 50 Fleitmann D, Burns SJ, Mudelsee M, et al. 2003. Holocene forcing of the Indian monsoon recorded in a stalagmite from southern Oman. Science 300:1737–1739. Groucutt and Petraglia 125 ARTICLE 51 Parker AG, Preston G, Walkington H, et al. 2006. Developing a framework of Holocene climatic change and landscape archaeology for southeastern Arabia. Arab Archaeol Epigr 17:125–130. 52 Engel M, Brückner H, Pint A, et al. n.d. The early Holocene humid period in NW Saudi Arabia: sediments, microfossils and palaeo-hydrological modelling. Quatern Int. In press. 53 Arz HW, Lamy F, Patzold J. 2006. A pronounced dry event recorded around 4.2 ka in brine sediments from the northern Red Sea. Quaternary Res 66:432–441. 54 McLaren SJ, Al-Juaidi F, Bateman MD, et al. 2009. First evidence for episodic flooding events in the arid interior of central Saudi Arabia over the last 60 ka. J Quaternary Sci 24:198–207. 55 Goudie AS, Colls A, Stokes S, et al. 2000. Latest Pleistocene and Holocene dune construction at the north-eastern edge of the Rub Al Khali, United Arab Emirates. Sedimentology 47:1011–1021. 56 Ivanochko TS, Ganeshram RS, Brummer GJA, et al. 2005. Variations in tropical convection as an amplifier of global climate change at the millennial scale. Earth Planet Sci Lett 235:302–314. 57 Fernandes CA. 2009. Bayesian coalescent inference from mitochondrial DNA variation of the colonization time of Arabia by the hamadryas baboon (Papio hamadryas hamadryas). In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 89–102. 58 Winney BJ, Hammond RL, Macasero W, et al. 2004. Crossing the Red Sea: phylogeography of the hamadryas baboon, Papio hamadryas hamadryas. Mol Ecol 13:2819–2827. 59 Whybrow PJ, Hill A, editors. 1999. Fossil vertebrates of Arabia. New Haven: Yale University Press. 60 Thomas H, Geraads D, Janjou D, et al. 1998. First Pleistocene faunas from the Arabian Peninsula: An Nafud Desert, Saudi Arabia. CR Acad Sci IIA 326:145–152. 61 O’Regan HJ, Turner A, Bishop LC, et al. 2011. Hominins without fellow travellers? First appearances and inferred dispersals of AfroEurasian large-mammals in the Plio-Pleistocene. Quaternary Sci Rev 30:1343–1352. 62 McClure HA. 1984. Late Quaternary paleoenvironments of the Rub al Khali. Ph.D. dissertation. London: University College London. 63 Drechsler P. 2009. The dispersal of the Neolithic over the Arabian Peninsula. BAR 1969. Oxford: Archaeopress. 64 Parr PJ, Zarins J, Ibrahim M, et al. 1978. Preliminary report on the second phase of the Northern Province survey 1397/1977. Atlal 2:29–30. 65 Zarins J, Murad AJ, Al-Yish KS. 1981. The comprehensive archaeological survey program. The second preliminary report on the southwestern province. Atlal 5:9–42. 66 Whalen N, Ali J, Sindhi H, et al. 1986. A Lower Pleistocene site near Shuwayhitiyah in northern Saudi Arabia. Atlal 10:94–101. 67 Chauhan PR. 2009. Early Homo occupation near the Gate of Tears: examining the paleoanthropological records of Djibouti and Yemen. In: Hovers E, Braun DR, editors. Interdisciplinary approaches to the Oldowan. Netherlands: Springer. p 49–59. 68 Amirkhanov H. 2008. Cave al-Guza the multilayer site of Oldowan in South Arabia. Moscow: TAUS. 69 Whalen N, Davis P, Pease D. 1989. Early Pleistocene migrations into Saudi Arabia. Atlal 12:59–75. 86 Inizan ML, Ortlieb L. 1987. Préhistoire dans la région de Shabwa au Yemen du Sud (R.D.P. Yemen). Paléorient 13:5–22. 70 Whalen N, Zoboroski M, Schubert K. 2002. The Lower Palaeolithic in southwestern Oman. Adumatu 5:27–34. 87 McBrearty S. 1999. Earliest stone tools from the Emirate of Abu Dhabi, United Arab Emirates. In: Whybrow PJ, Hill A, editors. Fossil vertebrates of Arabia. New Haven: Yale University Press. p 373–388. 71 Jagher R. 2009. The Central Oman Paleolithic Survey: recent research in southern Arabia and reflection on the prehistoric evidence. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 139–150. 72 Lahr MM. 2010. Saharan corridors and their role in the evolutionary geography of ‘‘Out of Africa 1.’’ In: Fleagle JJ, Shea JJ, Grine FE, Baden AL, Leakey RE, editors. Out of Africa 1: the first hominin colonization of Eurasia. New York: Springer. p 27–46. 73 Zarins J, Whalen N, Ibrahim M, et al. 1980. Comprehensive archaeological survey program: preliminary report on the central and southwestern provinces survey, 1979 Atlal 4:9–36. 74 Whalen N, Sindhi H, Wahida G, et al. 1983. Excavation of Acheulean sites near Saffaqah in al-Dawadami (1402/1982). Atlal 7:9–21. 75 Whalen N, Siraj- Ali J, Davies W. 1984. Excavations of Acheulean sites near Saffaqah, Saudi Arabia. Atlal 8:43–58. 76 Petraglia MD, Drake N, Alsharekh A. 2009. Acheulean landscapes and large cutting tool assemblages in the Arabian Peninsula. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 103–116. 77 Whalen N, Killick A, James N, et al. 1981. Saudi Arabian archaeological reconnaissance 1980: preliminary report on the western province survey. Atlal 5:43–58. 78 Whalen N, Siraj-Ali J, Sindhi H, et al. 1988. A complex of sites in the Jeddah-Wadi Fatima Area. Atlal 11:77–87. 79 Shipton C, Petraglia MD. 2010 Inter-continental variation in Acheulean bifaces. In: Norton C, Braun R, editors. Asian palaeoanthropology. London: Springer. p 49–56. 80 Rose JI. 2004. The question of Upper Pleistocene connections between East Africa and South Arabia. Curr Anthropol 45:551–555. 81 Crassard R. 2009. The Middle Paleolithic of Arabia: the view from the Hadramawt region, Yemen. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 151–168. 82 Whalen N, Schatte K. 1997. Pleistocene sites in southern Yemen. Arab Archaeol Epigr 8:1–10. 83 McClure HA. 1994. A new Arabian stone tool assemblage and notes on the Aterian industry of North Africa. Arab Archaeol Epigr 5:1–16. 84 Scerri E. n.d. The Aterian in Arabia: a reanalysis of the evidence. Proc Sem Arabian Studies. In press. 85 Macchiarelli R. 2008. From Africa to Asia through Arabia: models, predictions and witnesses of first phases of human settlement. Presented at the Conference on the First Great Migrations of Peoples, UNESCO, Paris. 88 Rose JI, Petraglia MD. 2009. Tracking the origin and evolution of human populations in Arabia. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 1–12. 89 Petraglia MD. 2011. Trailblazers across Arabia. Nature 470:50–51. 90 Garrard A, Harvey CPD, Switsur VR. 1981. Environment and settlement during the Upper Pleistocene and Holocene at Jubba in the Great Nefud, Northern Arabia. Atlal 5:137– 148. 91 Maher LA. 2009. The Late Pleistocene of Arabia in relation to the Levant. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 187–202. 92 Rose JI, Usik VI. 2009. The ‘‘Upper Paleolithic’’ of south Arabia. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 169–185. 93 Edens CA. 2001. A bladelet industry in southwestern Saudi Arabia. Arab Archaeol Epigr 12:137–142 94 Adams RM, Parr PJ, Ibrahim M, et al. 1977. Saudi Arabian archaeological reconnaissance, 1976: preliminary report on the first phase of the comprehensive archaeological survey program. Atlal 1:21–40. 95 Crassard R. 2008. The ‘‘Wa’shah method’’: An original laminar debitage from Hadramawt, Yemen. Proc Sem Arabian Studies 38:3–14. 96 Charpentier V, Inizan ML. 2002. Fluting in the Old World: the Neolithic projectile points of Arabia. Lithic Tech 27:39–46. 97 Uerpmann HP, Potts DT, Uerpmann M. 2009. Holocene (re-)occupation of eastern Arabia. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 205–214. 98 Boivin N, Blench R, Fuller DQ. 2009. Archaeological, linguistic and historical sources on ancient seafaring: a multidisciplinary approach to the study of early maritime contact and exchange in the Arabian Peninsula. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 251–278. 99 Fedele FG. 2009. Early Holocene in the highlands: data on the peopling of the Eastern Yemen plateau, with a note on the Pleistocene evidence. In: Petraglia MD, Rose JI, editors. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. Netherlands: Springer. p 215–236. C 2012 Wiley Periodicals, Inc. V