Despite new impetus for Late Pleistocene research in West Africa, little is known about the range of Middle Stone Age behaviours in this region. Yet, the multiplicity of Middle Stone Age lithic technologies testifies to significant behavioural and demographic dynamics, marked by innovation and adaptability. Here, we present the first indepth analysis of ochre remains in West Africa. New data from Toumboura III site, eastern Senegal, dated between 40 +/- 3 and 30 +/- 3 ka, point towards the use of ochre pieces as part of an occasional and specialized ochre crushing activity, probably dedicated to the production of powders, as well as the use of ochre sticks. Ochre pieces were studied at both macro and microscopic levels and while some of this iron-rich material likely accumulated in the deposits without anthropogenic intervention, another significant set of ochre pieces was found that was likely processed in situ. The impact scars on the pieces are not as striking as grinding traces for evidencing human exploitation. Nonetheless, they cannot be explained by natural phenomena. These remains could represent the earliest known evidence of ochre exploitation in Senegal. They potentially open new perspectives on symbolic behaviours in the Middle Stone Age of West Africa. They show that the full range of human behaviours in this region is yet far from being captured.

Abstract. The paleolithic site at Roc-en-Pail was discovered by chance between 1870 and 1871, in the context of quarrying activities when Prehistoric Science was at its youth. Nevertheless the lithic industries were immediate linked to the Moustier type with reference to the pioneer work of É. Lartet and H. Christy. The site was only latter largely excavated by Dr Michel Gruet, in the 1940s and 1950s then in 1969. Excavations archives from Dr Gruet together with the large collection he recovered allowed us to emphasize the scientific value of the site and to design a new field project. We aim to establish the history of researches carried at the site and to draw a knowledge review as a starting point to renewed fieldwork. We will try to determine why this site is so weakly visible in the literature whereas a five meters thick sequence with rich faunal and lithic assemblages was described by Dr Gruet. We first emphasize the confusion occurring in the literature between the Roc-en-Pail paleolithic site itself and close paleontological sites, from karstic structures. Secondly, we discuss the presence of a rockshelter, now destroyed, at Roc-en-Pail as proposed by Dr Gruet. Finally, we propose a critical review of the stratigraphy based on the preliminary results from our 2014 to 2018 excavations to highlight the interest of the site for our knowledge of Middle Paleolithic period in Western-Central France.

Resume. Le site paleolithique de Roc-en-Pail a ete decouvert fortuitement entre 1870 et 1871, en marge de l’exploitation d’une carriere alors que la science prehistorique n’en est encore qu’a ses debuts. Neanmoins, des sa decouverte, les industries lithiques sont rapportees au type du Moustier, en reference aux travaux pionniers d’E. Lartet et H. Christy. Il faudra attendre les recherches du Dr Michel Gruet dans les annees 1940-1950 puis en 1969, pour que de veritables fouilles soient entreprises sur le site. La documentation qu’il a laissee et l’imposante collection qu’il a recueillie nous ont permis de mesurer l’interet du site et de mettre en place un nouveau projet de recherches de terrain. Prealablement a ces nouvelles fouilles, l’objectif de cette contribution est de dresser l’historique des recherches et d’etablir un etat des connaissances. Nous tenterons de cerner les raisons pour lesquelles la visibilite de ce site dans la litterature est si faible alors que le Dr Gruet avait decrit une sequence de pres de cinq metres d’epaisseur riche en vestiges lithiques et fauniques. Nous mettrons d’abord en evidence les confusions faites jusqu’a present dans la litterature entre les occupations paleolithiques de Roc-en-Pail et des sites paleontologiques a indices d’occupation en contexte karstique. Ensuite, nous discuterons l’existence a Roc-en-Pail d’un abri-sous-roche, desormais detruit ainsi que propose par le Dr Gruet. Enfin, les resultats preliminaires des nouvelles fouilles nous amenent a reviser les interpretations du Dr Gruet quant a l’organisation des depots et a la sequence archeologique afin de souligner son potentiel pour la connaissance du Paleolithique moyen dans le centre-ouest de la France.

The Ravin Blanc I archaeological occurrence, dated to MIS 5, provides unprecedented data on the Middle Stone Age (MSA) of West Africa since well-contextualized archaeological sites pre-dating MIS 4/3 are extremely rare for this region. The combined approach on geomorphology, phytolith analysis, and OSL date estimations offers a solid framework for the MSA industry comprised in the Ravin Blanc I sedimentary sequence. The paleoenvironmental reconstruction further emphasizes on the local effects of the global increase in moisture characterizing the beginning of the Upper Pleistocene as well as the later shift to more arid conditions. The lithic industry, comprised in the lower part of the sequence and dated to MIS 5e, shows core reduction sequences among which Levallois methods are minor, as well as an original tool-kit composition, among which pieces with single wide abrupt notches, side-scrapers made by inverse retouch, and a few large crudely shaped bifacial tools. The Ravin Blanc I assemblage has neither a chronologically equivalent site to serve comparisons nor a clear techno-typological correspondent in West Africa. However, the industry represents an early MSA technology that could either retain influences from the southern West African ‘Sangoan’ or show reminiscences of the preceding local Acheulean. A larger-scale assessment of behavioral dynamics at work at the transition period between the Middle to Upper Pleistocene is discussed in view of integrating this new site to the global perception of this important period in the MSA evolutionary trajectories.

Les etudes menees le long de la Faleme, affluent du Senegal, dans le secteur Sansande‑Missira, permettent de se faire une idee assez precise de la stratigraphie des formations sedimentaires qui se devoilent a la faveur de l’incision recente du cours d’eau. Les datations par le radiocarbone et par luminescence optiquement stimulee calent l’essentiel de ces depots du dernier cycle climatique, avec une tres bonne representation de la periode 45-5 ka pour les sediments fins deposes lors des crues regulieres. Dans ces depots, la recurrence des sites archeologiques montre qu’il y avait, meme pendant la phase du dernier maximum aride, un ecoulement sud-nord suffisant pour attirer les populations.

Studies conducted along the Faleme River (East Senegal) since 2011, in Sansande-Missira sector, inform our understanding of the stratigraphy of Pleistocene alluvial formations. Radiocarbon and optically stimulated luminescence dating set a robust chronostratigraphical framework, of the last climate cycle and especially of the 45-5 ka period. The fine formations, punctuated by coarse formations, document the permanence of the flows while the recurrence of archaeological sites shows that there was sufficient flow to attract Palaeolithic populations, even during the last dry period.

The end of the Palaeolithic represents one of the least-known periods in the history of western Africa, both in terms of its chronology and the identification of cultural assemblages entities based on the typo-technical analyses of its industries. In this context, the site of Fatandi V offers new data to discuss the cultural pattern during the Late Stone Age in western Africa. Stratigraphic, taphonomical and sedimentological analyses show the succession of three sedimentary units. Several concentrations with rich lithic material were recognized. An in situ occupation, composed of bladelets, segments, and bladelet and flake cores, is confirmed while others concentrations of lithic materials have been more or less disturbed by erosion and pedogenic post-depositional processes. The sequence is well-dated from 12 convergent OSL dates. Thanks to the dating of the stratigraphic units and an OSL date from the layer (11,300-9,200 BCE [13.3-11.2 ka at 68%, 14.3-10.3 ka at 95%]), the artefacts are dated to the end of Pleistocene or Early Holocene. Palaeoenvironmental data suggest that the settlement took place within a mosaic environment and more precisely at the transition between the open landscape of savanna on the glacis and the plateau, and the increasingly densely-wooded alluvial corridor. These humid areas must have been particularly attractive during the dry season by virtue of their rich resources (raw materials, water, trees, and bushes). The Fatandi V site constitutes the first stratified site of the Pleistocene/Holocene boundary in Senegal with both precise geochronological and palaeoenvironmental data. It complements perfectly the data already obtained in Mali and in the rest of western Africa, and thus constitutes a reference point for this period. In any case, the assemblage of Fatandi V, with its bladelets and segments and in the absence of ceramics and grinding material, fits with a cultural group using exclusively geometric armatures which strongly differs from another group characterized by the production of bifacial armatures, accompanied in its initial phase by ceramics (or stoneware) and grinding material.

For more than twenty years, Iran has focused a great deal of research on the emergence and development of Upper Palaeolithic industries. South of the Caspian Sea, the site of Garm Roud 2, dated to 33878 ± 3300 Cal BP, is one of the few sites, with Mirak, to have yielded stratified evidence of occupation during this period. This paper presents the results of the typo-technological analysis of the lithic assemblage collected during the 2006 to 2008 excavation campaigns. The focus here is on a production of straight and twisted bladelets. Very few similarities have appeared between this assemblage and those from sites in the Zagros and the rest of Iran, which are generally older (about 35 to 40 000 Cal BP). Garm Roud 2 yields original evidence and may be seen as a key site for understanding Upper Palaeolithic technocomplexes and the development of bladelet production in this area.

Eurasian bifacial series present some common features, but are chiefly characterized by wideranging diversity in terms of bifacial technology and blank types, whereas the heavy duty component presents a limited number of types or categories. Heavy-duty tools are present throughout the geographical areas but the frequency of handaxes and/or cleavers is generally low, except in some regions, where they are made on flakes and/or cobbles/pebbles, on siliceous stones or other rock types. So far, it appears to be generally accepted that bifacial technology became widespread from 800 to 700 ka onwards, both for Europe and Asia, except for some earlier occurrences in the Levant and India. It would thus be reasonable to infer that bifacial technology first reached the Levant from Africa before moving toward Asia, then Europe. However, the existing data point to a much more complex reality, suggesting contemporaneous technological worlds, with or without links between them. In the state of current knowledge, and based on the methodology used for analysing lithic series, it is impossible to clearly argue in favour of either a unique phenomenon with movements of hominins or/and ideas from an African source, or to point to evidence of several onsets of bifacial technology over time on a local substratum. The palaeoanthropological background shows the difficulties involved in characterizing the few available hominin fossils and clearly relating them to bifacial technology. The current context suggests that each area should be analysed independently. Accumulative technological processes in some areas due to successive arrivals and the influence of the local substratum, and local onsets must be considered, as these contribute to the diversity of the strategies encountered and the varied forms of bifacial technology. (C) 2016 Academie des sciences. Published by Elsevier Masson SAS.

This second paper is part of a wider review of lithic complexes with bifacial technology, and is devoted to the Asian sector, from India to the south-eastern mainland and the archipelagos and China. For India, sites such as Attirampakham, Isampur, Morgaon and Singi-Talav are described in detail. For China, sites in the Bose Basin, but also Liangshan, Longgangsi and Houfang are included in discussions of technological strategies that are found a long way from East African roots. For the Southeast, discoveries from Thailand and Cambodia are presented, as are some major Indonesian sites (for instance Nebung and the Sangiran dome). (C) 2015 Academie des sciences. Published by Elsevier Masson SAS.

Assemblages with bifacial tools cover large areas of the Eurasian continent from the Middle Pleistocene to the Upper Pleistocene. These encompass varied technological strategies although the morphological results are often similar. Consequently, bifacial technology is diversified in Eurasia and cannot be correlated to "Acheulean-type behaviour" everywhere, or for all time periods. While early lithic assemblages with bifacial technology appeared in Africa as early as 1.8 Ma, this technology is observed in Eurasia from 1.5 Ma for the earliest evidence, and particularly from 1 Ma onwards. Our purpose here is to focus on Eurasian patterns. This presentation will be divided into three parts in order to describe, clarify and compare the bifacial phenomenon regarding its chronological and geographical extension, based on examples of key sites in diverse Eurasian areas. This first paper presents a review of detailed data from western and southern Europe and the Levant. The second focuses on the East, India, south-eastern Asia and eastern Asia (China). The third and final paper is a comparative analysis of Eurasian areas with occurrences of bifacial technology in relation to paleoanthropological remains in order to clarify the diversity of the bifacial technological phenomenon over time and space. Major European and Levantine sites are described in this first paper. For Europe, the bifacial component is divided into two main currents; before and after the major glacial event of MIS 12. Before MIS 12, sites such as la Noira, Notarchirico, Caune de l'Arago, Galeria II, Cagny-la-Garenne I-II or Boxgrove are taken into consideration. After MIS 12, sites are more numerous and have been widely reviewed (for instance Terra Amata, Guado San Nicolas, Cagny l'Epinette, Soucy...). For the Levant, sites such as Ubeidiya, GBY, Nadaouiyeh and El Meirah are presented. Data from the Arabian Peninsula are also taken into account. (C) 2015 Academie des sciences. Published by Elsevier Masson SAS.

Since 2011, international research program Peuplement Humain et Paleoenvironnement en Afrique (dir. Pr. E. Huysecom) focus on the Faleme Valley, eastern Senegal. In this area, several lithic industries identified as Early, Middle and Latter Stone Age, are known. This paper introduces the first elements of chronology based on Optically Stimulated Luminescence dating. Obtained dates cover the entire upper Pleistocene, which underline the importance of the Faleme Valley in the study of settlement dynamics and cultural evolution of western African Prehistoric Man. (C) 2017 Published by Elsevier Masson SAS.