Search

While the use of fire has long been recognised as a crucial innovation in the cultural evolution of humankind, much research has focused on the (debated) chronology of its earliest use and control, and less on the ways in which fire was used in the deep past. At its latest by the Upper Palaeolithic, hunter-gatherers routinely used fire to heat a wide range of materials, adjusting parameters like temperature, exposure time and fuel type to the specific requirements of the treated materials, for instance in food preparation or tool production. Comparing analyses of the chemical and physical properties of modern materials, heated under a range of controlled conditions in a laboratory, to archaeological ones might allow the reconstruction of the “heating history” of excavated materials and hence to infer the function of particular fires in the past - provided changes affecting the properties of the heated archaeological material during burial time are taken into consideration. To investigate the feasibility of such an approach, heated materials sampled from ~40,000 to 25,000 year old fireplaces (hearths) and their sedimentary matrices from the Upper Palaeolithic Abri Pataud rock shelter in South-Western France are used here to study (1) the fuel type(s) used by the site’s occupants, (2) the temperatures reached in fireplaces and (3) the potential changes in human activities related to fireplaces over time, with the influence of post-depositional processes taken into explicit consideration throughout. For this purpose, we used a range of methods to analyse macroscopically visible as well as “invisible” (microscopic and molecular) heat-altered materials. The results suggest that charred organic materials (COM) encountered in the samples predominantly result from the fuel used in fireplaces, including the earliest reported use of dung as fuel. Earlier suggestions about the use of bone as fuel at the Abri Pataud are not supported by this study. The heating temperature of COM increased gradually from 350 °C in the Aurignacian to 450 °C in Gravettian levels. Py-GC–MS studies identified a range of organic compounds, biomolecules derived from plant as well as animal sources, still preserved in the sediments after exposure to heat and burial in the rock shelter more than 20,000 years ago. Mammalian mtDNA was identified in sediment samples retrieved from the fireplaces, including ancient mtDNA fragments that originated from one or more modern human-like mitochondrial genome(s). This makes the Abri Pataud the first archaeological site for which ancient modern human mtDNA has been retrieved from sediment samples. The absence of specific organic compounds (furans) in the Aurignacian levels and their presence in the Gravettian ones, the changes in temperatures reached through the Aurignacian-Gravettian sequence as well as changes in the character of the fireplaces (presence/absence of lining river pebbles) suggest that the functions of hearths changed through time. These results highlight the potential of multi-proxy analyses of macro- and microscopic traces of ancient fireplaces, and especially of a shift in focus towards molecular traces of such activities. Systematic sampling of fireplaces and their sedimentary matrix should become a standard part of the excavation protocol of such features, to improve our understanding of the activities of humans in the deep past., The genetic work was funded by the Max Planck Society and the European Research Council (grant no. 694707 to Svante Pääbo). We thank B. Nickel, J. Richter, B. Schellbach and A. Weihmann for work in the ancient DNA laboratory of the Max Planck Institute for Evolutionary Anthropology at Leipzig.

This article presents a set of Late Pleistocene marine mollusk radiocarbon (AMS) age estimates of 30–5014C kyr BP, whereas a MIS5 age (>75 ka) is indicated by quartz and feldspar OSL dating, biostratigraphy, U-Th dating, and age-depth relationships with sea level. These results indicate that the14C dates represent minimum ages. The age discrepancy suggests that the shells are contaminated by younger carbon following shell death. The enigmatic14C dates cannot be “solved” by removing part of the shell by stepwise dissolution. SEM analysis of the Late Pleistocene shells within a context of geologically younger (recent/modern, Holocene) and older (Pliocene) shells shows the presence of considerable amounts of an intracrystalline secondary carbonate precipitate. The presence of this precipitate is not visible using XRD since it is of the same (aragonitic) polymorph as the original shell carbonate. The combination of nanospherulitic-shaped carbonate crystals, typical cavities, and the presence of fatty acids leads to the conclusion that the secondary carbonate, and hence the addition of younger carbon, has a bacterial origin. As shell material was studied, this study recommends an assessment of possible bacterial imprints in other materials like bone collagen as well.

Until now little is known about the role vegetation plays in the organic matter formation, particularly at the molecular level. Most ecosystems have a long history, which is unknown or too complex to find distinct relations between vegetation and the chemical composition of soil organic matter. To gain a better insight in such relationships, the relative simple soil-vegetation system of the Hulshorster Sands and Leuvenum Forest is used. The complete vegetation succession has been investigated, making it possible to monitor the organic matter development. During the succession, thus with time, the pH decreases, the ectorganic layer becomes thicker and the soil develops towards a micro-podzol.Chapter 2 describes the organic matter composition and relation to the first stages of vegetation development in the pre-podzol phase. In the Ah horizon under grass ( Corynephorus canescens ), the organic matter has virtually no contribution of the covering vegetation. Algae, from an earlier stage, and mosses ( Polytrichum piliferum and Bryum sp.), from adjacent spots, are the main suppliers. Under mosses, the organic matter is very similar to that of the original species. An aliphatic biopolymer, derived from mosses, that produces a series of branched alkene/alkane doublets (C 20 -C 32 ) together with a homologous series of n -alkenes and n -alkanes (C 10 up to C 34 ) upon pyrolysis, appears to be very recalcitrant in the first stages of succession.In the vegetation stages following mosses, these aliphatic macromolecules are still an important source of the aliphatic part of organic matter. Apart from the aliphatic moss contribution, the organic matter under Callunavulgaris and Pinus sylvestris has a close relationship with the contributing plant parts: flowers, leaves and branches dominate the organic layer under Calluna , needles predominate the ectorganic horizon under pines, and the soil organic matter in the Ah horizons in both cases is almost completely determined by roots.In Chapter 3 the water soluble organic matter (WSOM) of the Ah horizons is compared with the bulk horizons. The WSOM contains little lignin, also when this compound was present in significant amounts in the Ah horizons. In contrast, the moss-derived aliphatics are prominently present in the WSOM. Their presence in water is probably due to the formation of micelle-like arrangements in which the hydrophobic parts are inside a hydrophilic mantle resulting into 'soluble' aliphatics.Chapter 4 deals with the fractionation of three H horizons under different vegetation in the forest with incipient podzolization: a water soluble fraction, a NaOH soluble fraction and the residue, the insoluble fraction (humin). The water soluble fractions are very similar to each other with regard to their chemical composition, and so are the NaOH soluble factions. The insoluble fractions differ in particular in their aliphatic part. During beech forest development, an increasing amount of suberin, mainly derived from beech roots, is observed. The bulk of the organic matter in the H horizons, however, appears to have a pine origin as concluded from the chemical data. Pollen analysis supports that beech litter did not contribute to the H horizons.Chapter 5 addresses the analysis (composition of polysaccharides, NMR, pyrolysis and thermochemolysis) and the unraveling of the origin of the ectorganic layers in the succession from pine to beech forest. Under pine, the bulk organic matter shows only little effect of the ground vegetation Deschampsia and Empetrum respectively. The chemical composition of the L and F1 horizons are very similar to each other, but towards the H horizons an increase of aliphatics, a decrease of polysaccharides and a more or less constant fraction of aromatic compounds is noticed. The lignin composition is dominated by guaiacol building blocks, and hardly any angiosperm-lignin is present in the ectorganic layers implying no significant contributions of the undergrowth. This contrasts with macromorphological analyses, because abundant above- and below-ground parts of Deschampsia and Empetrum are present in the organic horizons suggesting an important input of the understorey. The composition of the aliphatic fraction also shows an effect of the undergrowth.The distribution of cutin and suberin building blocks as measured upon thermochemolysis reveals the clear impact of Deschampsia and Empetrum in addition to pine, especially in the F1 and F2 horizons. The transition towards the beech forest is characterized by a more distinct influence of beech litter. From L to H horizons polysaccharide contents decrease clearly, aromatics remain constant and the amounts of aliphatic compounds increase relatively.Furthermore, with depth a clear decrease of syringyl lignin compared with guaiacyl lignin is observed. The latter is mainly due to a decrease of beech litter towards the H horizon: the L consists only of beech litter, the F is also dominated by beech litter in addition to beech roots, and the H horizon is virtually only derived from pine litter. Nevertheless, the composition of the H horizons is affected by beech through their roots, which is clearly marked by the presence of beech-suberin, and the contents of that biopolymer increases from younger to older beech stands.In Chapter 6 the WSOM fractions from L, F, and H horizons are compared on their chemical composition with those of the bulk of B horizons and fibers of incipient podzols. The WSOM shows hardly any resemblance with the B horizons, whereas it appears to be very similar to the fibers. This strongly suggests that in the initial phases of podzolization almost all WSOM moves through the B horizons and accumulates when the water movement stops resulting into the formation of humus fibers. In Chapter 7 it is elucidated that the organic matter in the B horizons is mainly composed of root-derived material, whereas hardly any organic matter is complexed with metals or clay. The organic matter in B horizons is highly aliphatic, consisting partly of suberan-like polymers and partly of suberin, whereas lignin and polysaccharides are virtually absent. In Chapter 8, the chemical implications of the negligible organic matter sources of both illuviation as mixing of above-ground litter is described. The clear chemical evidence of root-derived organic matter could be established.