Abstract
It is commonly accepted that some of the latest dates for Neanderthal fossils and Mousterian industries are found south of the Ebro valley in Iberia at ca. 36 ka calBP (calibrated radiocarbon date ranges). In contrast, to the north of the valley the Mousterian disappears shortly before the Proto-Aurignacian appears at ca. 42 ka calBP. The latter is most likely produced by anatomically modern humans. However, two-thirds of dates from the south are radiocarbon dates, a technique that is particularly sensitive to carbon contaminants of a younger age that can be difficult to remove using routine pretreatment protocols. We have attempted to test the reliability of chronologies of 11 southern Iberian Middle and early Upper Paleolithic sites. Only two, Jarama VI and Zafarraya, were found to contain material that could be reliably dated. In both sites, Middle Paleolithic contexts were previously dated by radiocarbon to less than 42 ka calBP. Using ultrafiltration to purify faunal bone collagen before radiocarbon dating, we obtain ages at least 10 ka 14C years older, close to or beyond the limit of the radiocarbon method for the Mousterian at Jarama VI and Neanderthal fossils at Zafarraya. Unless rigorous pretreatment protocols have been used, radiocarbon dates should be assumed to be inaccurate until proven otherwise in this region. Evidence for the late survival of Neanderthals in southern Iberia is limited to one possible site, Cueva Antón, and alternative models of human occupation of the region should be considered.
Since the early 1990s, it has been widely acknowledged that the region south of the Ebro River and Cantabrian Cordillera in Iberia, here defined as southern Iberia, provided a refugium for the final Neanderthals (1–5). The earliest stages of the Aurignacian, an Upper Paleolithic lithic industry widely linked with an anatomically modern human (AMH) authorship (6–8), are absent from the region (5). In contrast, in northern Iberia the Aurignacian appeared around 42 ka calBP (calibrated radiocarbon date ranges, years before 1950), shortly after the disappearance of the Mousterian, a Middle Paleolithic lithic industry usually associated with Neanderthals (5). If the relationship between industry and species holds, this finding suggests a pause in AMH dispersal, providing a vacuum in which Neanderthals in southern Iberia could have survived (5, 9). Numerous Mousterian and Neanderthal assemblages in this region have yielded post-42 ka calBP dates supporting such a scenario, most famously at Carihuela (10), Gorham’s Cave (11, 12), and Zafarraya (13). A period of overlap between Neanderthals and AMHs in neighboring regions has implications for the inevitability of acculturation (1), interbreeding (14), and understanding the role of the environment in the spatial distribution of human populations (15). It is therefore crucial that the chronologies on which the pattern of occupation in Iberia is based are accurate.
Unfortunately, the chronology of the transition between the Middle and Upper Paleolithic has been clouded by doubts over the reliability of the chronological methods used, the extent to which taphonomic influences have blurred the association between the objects dated and the archaeological evidence, and the fact that the latest lithic assemblages are small and often undiagnostic, making them difficult to assign to the Mousterian (5, 16, 17). Because of this doubt, there is considerable debate regarding the extinction date of the “late” Neanderthals in Iberia and thus the length of any overlap with AMHs. Zilhão (5, 18, 19), for example, has suggested that Neanderthals survived across the southern region through Heinrich Event 4 (H4) until ca. 32–30 ka BP (ca. 36.7–34.5 ka calBP), and has termed this pattern the Ebro Frontier model. Zilhão suggests that this pattern can be explained, at least in part, by the cultural adaptation of AMHs to the open steppe-like environments found north of the Cantabro-Pyrenean chain. Only when woodland was replaced during the climatic deterioration of GI7 (ca. 35.5 ka) did they spread south. Others (2, 12, 20) have envisaged a gradual retreat of Neanderthals into the extreme southwestern region around Gibraltar and southern Portugal, with an extinction date of less than 30 ka BP (34.9–34.5 ka calBP). A minority (16, 17, 21) regard the evidence as too weak, and have entirely rejected the late survival of Neanderthals.
Here we show that many of these studies have underestimated the problematic nature of the radiocarbon dates in southern Iberia, which comprise two-thirds of the post-42 ka calBP dates for the Mousterian and Neanderthals. The significant effect that young contaminants have on the accuracy of Paleolithic radiocarbon chronologies has become apparent with the development and application of rigorous pretreatment protocols designed to more effectively remove exogenous sources of carbon, such as ABOx-SC of charcoal and ultrafiltration of bone collagen (22). With 1% modern carbon contamination, a sample of 50 ka BP will appear ca. 37 ka BP. Using improved pretreatment protocols, the existing Paleolithic chronologies at sites such as Fumane, Italy (23), Abri Pataud, France (24), Geissenklosterle, Germany (25), and Kent’s Cavern, United Kingdom (8) have lengthened by several millennia. Moreover, upon application of rigorous pretreatment protocols, the radiocarbon dates for these sites yielded results consistent with stratigraphy and in agreement with nonradiocarbon chronologies.
To test the reliability of the radiocarbon evidence for late survival of Neanderthals in southern Iberia, our study focused on radiocarbon dating several purportedly late Neanderthal, Middle, and early Upper Paleolithic assemblages using a rigorous pretreatment method for cleaning bone collagen involving ultrafiltration (22, 26). Because of the poor preservation of organic material in southern Iberia, the nitrogen content (%N) of bone was measured to identify which bones were most likely to contain enough collagen for radiocarbon dating (27). Of 215 bones screened, only 27 contained enough nitrogen to attempt collagen extraction (SI Text).
Site Description
Two sites have been dated: Jarama VI, which provides the primary evidence for a late Mousterian in central Iberia, and the Cueva del Boquete de Zafarraya, containing the latest Neanderthal fossils in Europe. A further nine sites were examined, but no suitable materials for radiocarbon dating were found (see Description of Sites and Samples That Could Not Be Radiocarbon Dated and %N Screening Results in SI Text, Figs. S1 and S2, and Table S1).
The rock shelter of Jarama VI (Fig. 1A) is located in a steep-sided limestone gorge in the Jarama valley on the southern flank of the Sistema Central. An area of 16 m2 was excavated between 1989 and 1993 (28). Unit I, formed of sands, silts, and products of gelifraction of the cave walls, was found in one corner of the excavation (Fig. 1B) and contained a rich Upper Paleolithic assemblage with blade and bladelet blanks. This unit is separated from the underlying fluvial deposit by an erosive event. Levels 2.1, 2.2, and 3 contain Mousterian assemblages with points, sidescrapers, burins, notches, and denticulates. The Levallois technique is present, and the complete reduction sequence is represented with cores and debitage products dominating the assemblages. Some of the sparse assemblage in level 2.2 is arranged around a hearth feature. A human metatarsal, tentatively identified as Neanderthal, was recovered from this level (29). Units 2.2 and 3 are separated by a thick flood deposit (level 2.3) and a second erosive event (28).
Fig. 1.
The site of Jarama VI (40°56′50,56′′N 3°19′1,06′′W, Valdesotos, Castilla-la-Mancha, 822 m above sea level). (A) The location of the cave in the Jarama canyon is marked with a white arrow. (B) Plan of the cave. (C and D) Stratigraphy and location of samples radiocarbon dated. Open circles represent published dates and closed circles represent dates from this work.
Charcoal from level 2.1 (29500 ± 2,700 BP Beta-56638, 44090–29830 calBP), the hearth feature in 2.2 (32600 ± 1,800 BP Beta-56639, 42260–34510 calBP), and a burrow, thought to relate to level 1 (23380 ± 500 BP Beta-56640, 29370–26920 calBP), has been conventionally radiocarbon dated (28) (Table S2). The site features prominently in discussions of the chronology of the final Mousterian because of the statistical agreement between the two Mousterian dates and the secure association of one date with human activity (3, 5, 19, 30).
Excavations at the montane site of Cueva del Boquete de Zafarraya (Fig. 2A) in 1981–1983 and 1990–1994 (31) produced a remarkable collection of cutmarked, fragmented, and burned Neanderthal remains. The site has been, and is still, regarded by some as providing the latest secure evidence for Neanderthals (4, 13), but others (19), although more cautious, still consider it late.
Fig. 2.
The site of Zafarraya (36°57′05 ′′N, 4°07′36′′W, Ventas de Zafarraya, Andalucia, 1,100 m above sea level) (A) located in the Boquete de Zafarraya. (B) Plan and location of samples (filled circles) and Neanderthal remains (large open circles) [modified from Barroso-Ruiz et al. (34)]. (C) Stratigraphic relationship between Neanderthal remains in P-Q 17-18 and radiocarbon dates. Note that Z6 (scapular) found in Q18 has not been plotted in B and C because coordinates are not available, the mandible Z18 was found in three fragments split between squares Q18 and P17, and four of the five Neanderthal remains in the Sala del Fondo were found in the level of disturbed deposits that capped the site. The location of samples dated in this study are denoted with filled circles and the location of Neanderthal fossils with open circles. Mousterian lithics were found 1 m above these radiocarbon dates, but no suitable samples for radiocarbon dating were found at a similar depth.
The cave is divided into the Sala de Entrada and the Sala del Fondo (Fig. 2B). Clandestine excavations in the early 20th century left a deep (5–6 m) hole in the Sala del Fondo and contributed to a level of disturbed deposits, removed before excavation. The in situ sediments are split into three units, the middle of which is formed of a silty-sand matrix with limestone fragments and contains a lithic assemblage assigned to the Typical Mousterian alongside the Neanderthal fossils (32). Upper Paleolithic bladelets have been recovered in this unit in the Sala del Fondo to a depth of 118 cm, but were found in an area where sediment was loosely packed, and are thought to be intrusive (33).
Of the 55 Neanderthal and AMH fossils from Zafarraya, a cluster of 10 Neanderthal remains, including two highly diagnostic mandibles, was recovered from the Sala de Entrada between 190 and 240 cm depth (Fig. 2 B and C) (34). Three of these remains, two femora and one tibia, are associated with a hearth [context D(Sm)] and are calcined and highly fragmented.
Two attempts have been made at directly dating the Neanderthal bones. U-Series dating by γ-spectroscopy was unsuccesful (35, 36) and no collagen could be recovered from two bones (Z1 and Z2) (Fig. 1C) for radiocarbon dating. The chronology is therefore based on bones of the most abundant fauna, Capra pyrenaica (37). Hublin et al. (13) dated three bone samples from the Sala del Fondo with Uranium Series (U-Series), and conventional radiocarbon after extracting collagen with the Longin (SI Methods) protocol. The deepest sample was found at a similar depth to the Neanderthals in the Sala de Entrada, and results ranged between 33.4 ± 2 ka and 28.9 ± 4.2 ka (see Published Radiometric Dates from Jarama VI and Zafarraya in SI Text and Table S2). These results were accepted (13) as the two methods agreed, the dates were in stratigraphic order, and collagen yields were high (>1%). However, this late chronology has been cast into doubt (35, 36, 38) by a series of 42 radiocarbon, U-Series, electron spin resonance, and amino acid racemisation dates on bone and teeth that are inconsistent within and between methods, and compared with depth (see Published Radiometric Dates from Jarama VI and Zafarraya in SI Text and Table S2). This result is unsurprising given that, in addition to the uncertainties regarding radiocarbon dates in this period, U-Series, electron spin resonance, and amino acid racemisation can be unreliable when dating the open systems of osseous remains (39–41).
Results
Of 30 anthropogenically modified bones screened for %N at Jarama VI, two contained >0.8%N, suggesting enough collagen was present to obtain a date, and a further five were considered marginal (%N Screening Results in SI Text). Three samples were dated: both samples containing >0.8%N (levels 1 and 3) and the only sample with a marginal content of nitrogen from unit 2 (Fig. 1C).
Two bones produced infinite radiocarbon ages and one (OxA-X-2310-22), a finite age with an error of 3,700 14C years (Table 1). This date should be regarded as a minimum age as the bone contained little collagen (<1%) and has an error approaching 4,000 14C years, the point at which radiocarbon dates become infinite (SI Methods). The bone from level 1 was found in the uppermost spit and may result from local disturbance or bioturbation (28).
Table 1.
Radiocarbon dates of bones from Jarama VI and Zafarraya
Sample reference | Treatment | OxA- | Date (BP) | Yield (mg) | Yield (%) | %C | δ13C (‰) | δ15N (‰) | C:N | Comment or source |
Jarama VI | ||||||||||
JA-6 | AF | 21,714 | >50,200 | 14.0 | 1.4 | 46.9 | −19.3 | 9.5 | 3.4 | |
JA-15 | AF | X-2310–22 | 49,400 ± 3,700 | 5.4 | 0.5 | 43.2 | −18.2 | 7 | 3.2 | Low % yield. Close to background. |
JA-18 | AF | X-2,290–56 | >47,000 | 7.4 | 0.7 | 45.3 | −18.8 | 10.3 | 3.3 | Low % yield |
Zafarraya | ||||||||||
Z8os | AG | 8,999 | 33,300 ± 1,200 | 3.9 | 1 | 32.5 | −19.1 | 5.3 | 3.3 | Michel et al. (36, 38) |
AF | 23,198 | >46,700 | 62.1 | 5.4 | 44.5 | −18.9 | 4.6 | 3.3 | ||
AF* | 26,440 | >46,700 | 15.3 | 1.9 | 41.2 | −19.1 | 5.2 | 3.2 | ||
ZAF2 | AF | 21,810 | 46,300 ± 2,500 | 21.3 | 2 | 44.6 | −19.7 | 7 | 3.3 | |
ZAF3 | AF* | 2.2 | 0.3 | Failed on low yield | ||||||
ZAF7 | AF* | 21,813 | >49,300 | 13.8 | 1.4 | 44.3 | −18.9 | 5 | 3.4 | |
ZAF8 | AF | 0.7 | 0.1 | Failed on low yield |
AF denotes the ultrafiltration pretreatment protocol, AG dontes a protocol identical except for the omission of ultrafiltration and, * denotes an additional solvent treatment. Z8os was dated with and without a solvent treatment to assess whether ultrafiltration concentrated hydrocarbons affecting the date. To obtain a reliable radiocarbon date, bone should contain >1% collagen, δ13C between −22 to −18‰, δ15N between 2‰ and 12‰, C:N 3.9–3.4 and % C >30% (42). All errors are given at 1SD. Typical 1SD errors for stable isotope values and %N measurement are ± 0.2‰ and ± 0.1%, respectively.
Together, the three new dates show that the radiocarbon dates on charcoal are severe underestimations. No bones with an adequate content of nitrogen from level 2.1 exist to date the final Mousterian occupation of the site. However, with the removal of the charcoal dates from discussion, there is no evidence for a late Mousterian occupation at Jarama VI.
At the Cueva del Boquete de Zafarraya, sampling was undertaken in two stages. To test the accuracy of the published radiocarbon dates, bones previously dated (35, 36, 38) were resampled for %N analysis (Table S3). All but one sample contained insufficient nitrogen for the ultrafiltration protocol to be attempted. In addition, previously undated Capra pyrenaica bones were examined. Reflecting the punctuated human occupation of the cave (37), cut-marks and evidence of smashing was not observed, and so unmodified bones were selected for screening. Although all eight samples from the Sala del Fondo failed the screening test (%N Screening Results in SI Text), 4 of 11 samples from the Sala del Entrada contained sufficient nitrogen to proceed with the ultrafiltration treatment (%N Screening Results in SI Text).
The %N analysis of bones previously dated correlates well with the quality assurance data obtained at the time of dating (see Published Radiometric Dates from Jarama VI and Zafarraya in SI Text). All except one bone, Z8os, failed the %N test, and all but Z8os produced collagen with unrealistic δ13C, %C, and %N values (42) (SI Methods), suggesting that degradation and or contamination of the protein was significant. Collagen was originally extracted from Z8os with a protocol that, except for the ultrafiltration step, was identical to the ultrafiltration protocol (laboratory code AG) (SI Methods) (36, 38), and a result of 33300 ± 1,200 BP (OxA-8999) was obtained. When redated with the ultrafiltration protocol, its age increased to >46700 BP (OxA-23198 and OxA-26440) (Table 1). All radiocarbon dates on bone from the site, including those from the apparently well-preserved bones in Hublin et al.’s (13) study, should therefore be viewed with extreme caution.
Two of the four bones treated for the first time here contained >1% collagen, and add further weight to this conclusion. ZAF7 gave an infinite date (OxA-21813) and ZAF2 a date of 46300 ± 2,500 BP (OxA-21810) (Table 1) close to the limit of the radiocarbon method. Z8os and ZAF2 were found stratigraphically above the Neanderthal remains in the Sala de Entrada, and ZAF7 was found within the hearth feature (Fig. 2 B and C). The dated bones were not cut-marked, and only date the context of the Neanderthal fossils and Mousterian lithics. Although these results can only be used to tentatively suggest that these remains date close to or beyond the limit of radiocarbon, they cast into doubt the previous post-42 ka calBP chronology. This collection of Neanderthals should no longer be cited as providing evidence for the southern Iberian Neanderthal late refugia.
Discussion
The warm temperatures across much of southern Iberia today (43) are not favorable for the preservation of the organic remains required for radiocarbon analyses, as reflected in the scarcity of datable bone (%N Screening Results in SI Text). The absence of well-preserved organic remains has lead to the dating of inappropriate sample types [e.g., sediment in the case of Carihuela (10)], the application of particularly gentle pretreatment protocols, and the acceptance of dates on material that would otherwise be considered too degraded (e.g., Zafarraya). The significance of these concerns is well illustrated by the difference of more than 10 ka radiocarbon years between radiocarbon dates on material pretreated with routine protocols and the more rigorous ultrafiltration bone pretreatment method at Jarama VI and Zafarraya. Unless radiocarbon dating has been undertaken on samples treated with a protocol demonstrated to give accurate dates on equivalent material, dates must be assumed to be inaccurate unless proven otherwise.
Bearing this finding in mind, alongside taphonomic factors, a review of sites thought to postdate 42 ka calBP is given in Table S4. None of these sites provide strong evidence for the late survival of Neanderthals or the Mousterian. For example, in support of the Ebro frontier hypothesis, Zilhão et al. (30) claim the most robust evidence for the late survival of Neanderthals comes from Sima de las Palomas and Cueva Antón (Murcia), Gruta da Oliveira (Portugal), Gorham’s Cave (Gibraltar, Natural History Museum Excavations), and Jarama VI. Of these sites, we can now remove Jarama VI. The other four sites have significant shortcomings that limit their reliability. Gorham’s Cave (11) has been dated by radiocarbon using charcoal pretreated with a deliberately gentle pretreatment (RR) (SI Methods) designed to ensure that some carbon survives the procedure and can be dated. Sadly, this method is even less likely (26) to remove contaminants than the routine protocol (acid-base-acid or ZR) (SI Methods) found to be problematic at other sites of a similar age (8). The chronologies of Sima de las Palomas (44) and Gruta da Oliveira (45) are based on radiocarbon dating of burnt bone and U-Series dating of bone. Burnt bone of Paleolithic age can sometimes produce unreliable radiocarbon dates because the origin of the carbon extracted is unknown (22). To obtain a reliable U-Series date episodes of uranium uptake and leaching must be identifiable (40). At Sima de las Palomas, two radiocarbon dates on burnt bone, associated with disarticulated Neanderthal fossils and Mousterian lithics, postdate 42 ka calBP (Table S4). From the same stratigraphic unit, there are two significantly older results: a U-Series date on bone where the uranium uptake history could not be established (43.8 ± 0.75 ka APSLP4) and an optically stimulated luminescence date (54.7 ± 4.7 ka X2509) (44). Given the problematic nature of these contradictory dates, the age of this unit and its Neanderthal fossils remains uncertain. Oliveira level 8 contains a Mousterian assemblage with consistent radiocarbon and U-Series dates, suggesting an age of less than 42 ka calBP (45). Despite this consistency, the dates suffer from the same methodological problems as those seen in Sima de las Palomas, and must again be used with care.
The only site to contain a date obtained using a method known to give reliable results is Cueva Antón, level Ik (46). A radiocarbon date on charcoal pretreated with the rigorous ABOx-SC method (32890 ± 200 BP OxA-21244, 38440–36810 ka calBP) (26) suggests that the unit probably formed shortly after 38.5 ka calBP. However, the dated charcoal is not functionally associated with the archaeology and the “handful” (46) of lithics is not yet fully published. No typical Upper Paleolithic artifacts have been uncovered (46), but the small size and largely undiagnostic nature of the published lithic assemblage means that it can only be tentatively assigned to the Middle Paleolithic. Although the ongoing excavation (46) of Cueva Antón may uncover a larger and more diagnostic lithic assemblage, in our view the assemblage cannot currently be used as the only evidence for the late survival of Neanderthals.
At Pego do Diabo, a single cutmarked bone has been dated, using ultrafiltered collagen, to 38750 ± 750 BP (44300–42100 calBP, OxA-15004) (30) and provides the last evidence of human activity before H4. This bone is not associated with a lithic industry, so it is not known whether it indicates a Neanderthal or AMH presence. The next reliable date is from Cueva Antón, level Ik (46), leaving a gap of at least 4,000 y during which no accurate dates exist (Fig. 3). Crucially, this period spans H4, the period during which the Ebro Frontier model (5, 18, 19) predicts Neanderthals were living in southern Iberia while AMHs were to the north. The first evidence of modern humans in southern Iberia is from the Evolved Aurignacian (5, 9). This industry is poorly dated in southern Iberia (Table S5), but in comparison with sites further north, is expected to fall after H4 (5, 24).
Fig. 3.
The interval between reliable radiometric age estimates in the south of Iberia. OxA-15004 (30) from Pego do Diabo is a cut-marked bone treated with the ultrafiltration protocol, and OxA-21244 (46) is a fragment of charcoal from a level containing lithics reminiscent of the Middle Paleolithic at Cueva Antón. Although the date’s relationship to human activity may be questioned, it is currently the only accurate date near H4 in southern Iberia. H4 is shown in gray and the Greenland Interstadials are numbered against the NGRIP δ18O record (47, 48). Radiocarbon dates are calibrated against IntCal09 (49) in OxCal v4.1 (50).
Three scenarios should be considered when examining the Middle to Upper Paleolithic transition in Iberia. These are: (i) the southern region of Iberia was abandoned by both Neanderthals and AMHs; (ii) Neanderthals existed in southern Iberia; and (iii) AMHs were present, spreading into the southern regions soon after they arrived in northern Iberia.
In our view, too few sites have been well dated to test these scenarios, and all may be possible. Given the scarcity of accurately dated sites, it possible that both AMHs and Neanderthals were present in the region during H4, and that the region was entirely abandoned. The latter theory is difficult to assess because an unconformity or hiatus between Middle and Upper Paleolithic exists in many sites (5, 51, 52). Given these uncertainties, when studying this period in Iberia, assemblages dated to <42 ka calBP should not be assumed to be produced solely by Neanderthals unless diagnostic lithic assemblages, or preferably fossils, are present.
The late survival of Neanderthals has played a role in discussion regarding the inevitability of acculturation, the duration over which interbreeding may have occurred and the role of the environment in the Middle to Upper Paleolithic transition. With doubt cast over the late survival of Neanderthals, the place of southern Iberia in these arguments must be viewed cautiously. It is crucial that efforts are directed toward testing existing chronologies, and the construction of new chronological datasets must continue before further conclusions can be drawn.
Methods
Bone samples were screened for %N before radiocarbon dating (27). If a bone contain more than 0.8%N, in around 70% of cases more than the 1% collagen required for dating (42) can be extracted with the ultrafiltration protocol (27). Bone with 0.5–0.7%N was considered marginal and only selected for dating if better preserved material was not available.
All samples were dated at the Oxford Radiocarbon Accelerator Unit. Radiocarbon dating followed the methods described in Brock et al. (26). Bones were treated with the ultrafiltration protocol, involving a series of HCl and NaOH washes to remove bone mineral, exogenous carbonates, and humic acids, followed by gelatinization, filtration (45–90 μm Ezee filter), and ultrafiltration (Vivaspin 15 30-kDa MWCO ultrafilter) to further clean the protein. As a cautionary measure, samples from Zafarraya were given a series of solvent washes before collagen extraction as glues were applied to some bones during excavation. The purified collagen was freeze-dried and combusted in an elemental analyzer (e.g., Carlo Erba NA 2000) linked to a continuous flow-isotope ratio mass spectrometer (CF-IRMS; e.g., Sercon 20–20). A portion of gas was fed into the CF-IRMS and the remaining CO2 was collected, converted to graphite, and dated by accelerator mass spectrometry (53) (SI Methods). Other pretreatment protocols discussed in the text are described in Table S6. Radiocarbon dates have been calibrated using the IntCal09 calibration curve (49) in OxCalv4.1 (50). IntCal09 extends to 50 ka calBP, and calibration is impossible for samples older than ca. 45 ka BP. Throughout the text conventional radiocarbon (BP) and nonradiocarbon dates are given at 1 SD, and calibrated radiocarbon date ranges (calBP) at 95% probability.
Supplementary Material
Acknowledgments
We thank C. Díez and R. Jacobi for identifying cut-marked bones; F. Brock for analysis of OxA-26440; V. Michel for comments regarding the published dates from Zafarraya; and V. Villaverde, C. Finlayson, D. Fa, C. Hernández, C. Mallol, M. Walker, E. Domenech, M. Cortes Sanchez, E. Aura, A. Marín, and A. Garcia Moreno for supplying samples and commenting on an earlier draft. Samples were taken from the Museu de Prehistòria de València, Museu Arqueològic Municipal de Alcoi and the Gibraltar Museum. This work was funded by Natural Environment Research Council Grant NE/D014077/1; work at El Salt was funded by a Spanish I+D Project under the direction of B. Galván, Universidad de La Laguna (Proyecto HAR2008-06117, Ministerio de Educación y Ciencia - Fondo Europeo de Desarrollo Regional); and the re-analysis of El Niño was funded by Instituto de Estudios Albacetenses “Don Juan Manuel” de la Excma Diputación de Albacete.
Footnotes
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1207656110/-/DCSupplemental.
References
- 1.D’Errico F, Zilhão J, Julien M, Baffler D, Pelegrin J. Neanderthal acculturation in western Europe? A critical review of the evidence and its interpretation. Curr Anthropol. 1998;39(Suppl 1):S1–S44. [Google Scholar]
- 2.Vega Toscano LG, Hoyos M, Ruiz-Bustos A, Laville H. In: L'homme de Neandertal, vol. 2, L'environnement [Neanderthal Man, vol.2, The Environment] Otte M, editor. Liège: Études et Recherches Archéologiquesde l'Université de Liège; 1988. pp. 169–180. French. [Google Scholar]
- 3.Straus LG. A mosaic of change: The Middle-Upper Paleolithic transition as viewed from New Mexico and Iberia. Quat Int. 2005;137(1):47–67. [Google Scholar]
- 4.Hublin J-J, Bailey SE. In: When Neanderthals and Modern Humans Met. Conard NJ, editor. Tuebingen: Kerns; 2006. pp. 105–128. [Google Scholar]
- 5.Zilhão J. Chronostratigraphy of the Middle-to-Upper Paleolithic transition in the Iberian Peninsula. Pyrenae. 2006;37(1):7–84. [Google Scholar]
- 6.Trinkaus E. Early modern humans. Annu Rev Anthropol. 2005;34:207–230. [Google Scholar]
- 7.Bailey SE, Weaver TD, Hublin J-J. Who made the Aurignacian and other early Upper Paleolithic industries? J Hum Evol. 2009;56(1):11–26. doi: 10.1016/j.jhevol.2009.02.003. [DOI] [PubMed] [Google Scholar]
- 8.Higham T, et al. The earliest evidence for anatomically modern humans in northwestern Europe. Nature. 2011;479(7374):521–524. doi: 10.1038/nature10484. [DOI] [PubMed] [Google Scholar]
- 9.Villaverde V, Aura JE, Barton CM. The Upper Paleolithic in Mediterranean Spain: A review of current evidence. J World Prehist. 1998;12(2):121–198. [Google Scholar]
- 10.Fernández S, et al. The Holocene and Upper Pleistocene pollen sequence of Carihuela Cave, southern Spain. Geobios. 2007;40(1):75–90. [Google Scholar]
- 11.Pettitt PB, Bailey RM. In: Neanderthals on the Edge. Stringer CB, Barton RNE, Finlayson JC, editors. Oxford: Oxbow Books; 2000. pp. 155–162. [Google Scholar]
- 12.Finlayson C, et al. Late survival of Neanderthals at the southernmost extreme of Europe. Nature. 2006;443(7113):850–853. doi: 10.1038/nature05195. [DOI] [PubMed] [Google Scholar]
- 13.Hublin J-J, Barroso-Ruiz C, Lara PM, Fontugne M, Reyss J. The Mousterian site of Zafarraya (Andalucia, Spain): Dating and implications on the Palaeolithic peopling processes of Western Europe. C R Acad Sci II. 1995;32(série IIa)1:931–937. [Google Scholar]
- 14.Trinkaus E. European early modern humans and the fate of the Neandertals. Proc Natl Acad Sci USA. 2007;104(18):7367–7372. doi: 10.1073/pnas.0702214104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.D'Errico F, Sánchez-Goni M. Neandertal extinction and the millennial scale climatic variability of OIS 3. Quat Sci Rev. 2003;22(8-9):769–788. [Google Scholar]
- 16.Jöris O, Álvarez Fernández A, Weninger B. Radiocarbon evidence of the Middle to Upper Palaeolithic transition in Southwestern Europe. Trabajos de Prehistoria. 2003;60(2):15–38. [Google Scholar]
- 17.Vaquero M. El tránsito Paleolítico Medio/Superior en la Península Ibérica y la Frontera del Ebro. Comentario a Zilhão. Pyrenae. 2006;37(2):107–129. [Google Scholar]
- 18.Zilhão J. In: El Origen Del hombre Moderno En El Suroeste De Europa [The Origin of Modern Humans in the Southeast of Europe] Cabrera Valdés V, editor. Madrid: UNED; 1993. pp. 127–145. Spanish. [Google Scholar]
- 19.Zilhão J. In: The Mediterranean From 50 000 to 25 000 BP: Turning Points and New Directions. Camps M, Szmidt C, editors. Oxford: Oxbow Book; 2009. pp. 293–312. [Google Scholar]
- 20.Finlayson C, et al. Gorham’s Cave, Gibraltar—The persistence of a Neanderthal population. Quat Int. 2008;181(1):64–71. [Google Scholar]
- 21.Jöris O, Street M. At the end of the 14C time scale—The Middle to Upper Paleolithic record of western Eurasia. J Hum Evol. 2008;55(5):782–802. doi: 10.1016/j.jhevol.2008.04.002. [DOI] [PubMed] [Google Scholar]
- 22.Higham T. European Middle and Upper Palaeolithic radiocarbon dates are often older than they look: Problems with previous dates and some remedies. Antiquity. 2011;85(327):235–249. [Google Scholar]
- 23.Higham T, et al. Problems with radiocarbon dating the Middle to Upper Palaeolithic transition in Italy. Quat Sci Rev. 2009;28(13-14):1256–1267. [Google Scholar]
- 24.Higham T, et al. Precision dating of the Palaeolithic: A new radiocarbon chronology for the Abri Pataud (France), a key Aurignacian sequence. J Hum Evol. 2011;61(5):549–563. doi: 10.1016/j.jhevol.2011.06.005. [DOI] [PubMed] [Google Scholar]
- 25.Higham T, et al. Τesting models for the beginnings of the Aurignacian and the advent of figurative art and music: The radiocarbon chronology of Geißenklösterle. J Hum Evol. 2012;62(6):664–676. doi: 10.1016/j.jhevol.2012.03.003. [DOI] [PubMed] [Google Scholar]
- 26.Brock F, Higham T, Ditchfield P, Ramsey CB. Current pretreatment methods for AMS radiocarbon dating at the Oxford Radiocarbon Accelerator Unit (ORAU) Radiocarbon. 2010;52(1):103–112. [Google Scholar]
- 27.Brock F, Higham T, Ramsey CB. Pre-screening techniques for identification of samples suitable for radiocarbon dating of poorly preserved bones. J Archaeol Sci. 2010;37(4):855–865. [Google Scholar]
- 28.Jordá Pardo JF. The wild river and the last Neanderthals: A palaeoflood in the geoarchaeological record of the Jarama Canyon (Central Range, Guadalajara province, Spain) Geodin Acta. 2007;20(4):209–217. [Google Scholar]
- 29.Lorenzo C, et al. New human fossil to the last Neanderthals in central Spain (Jarama VI, Valdesotos, Guadalajara, Spain) J Hum Evol. 2012;62(6):720–725. doi: 10.1016/j.jhevol.2012.03.006. [DOI] [PubMed] [Google Scholar]
- 30.Zilhão J, et al. Pego do Diabo (Loures, Portugal): Dating the emergence of anatomical modernity in westernmost Eurasia. PLoS ONE. 2010;5(1):e8880. doi: 10.1371/journal.pone.0008880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Barroso-Ruiz C, Medina Lara F, Caparros M. In: El Pleistoceno Superior de la Cueva del Boquete de Zafarraya [The Late Pleistocene of the Cave of the Boquete de Zafarraya] Barroso-Ruiz C, editor. Seville: Junta de Andalucia; 2003. p. 21. Spanish. [Google Scholar]
- 32.Barroso-Ruiz C, Medina Lara F, Boutie P, Barsky D. In: El Pleistoceno Superior de la Cueva del Boquete de Zafarraya [The Late Pleistocene of the Cave of the Boquete de Zafarraya] Barroso-Ruiz C, editor. Seville: Junta de Andalucia; 2003. pp. 431–468. Spanish. [Google Scholar]
- 33.Barroso-Ruiz C, Medina Lara F, Onoratini G, Joris C. In: El Pleistoceno Superior de la Cueva del Boquete de Zafarraya [The Late Pleistocene of the Cave of the Boquete de Zafarraya] Barroso-Ruiz C, editor. Seville: Junta de Andalucia; 2003. pp. 469–495. Spanish. [Google Scholar]
- 34.Barroso-Ruiz C, De Lumley MA, Caparrós M, Verdú L. In: El Pleistoceno Superior de la Cueva del Boquete de Zafarraya [The Late Pleistocene of the Cave of the Boquete de Zafarraya] Barroso-Ruiz C, editor. Seville: Junta de Andalucia; 2003. pp. 327–388. Spanish. [Google Scholar]
- 35.Michel V, et al. In: El Pleistoceno Superior de la Cueva del Boquete de Zafarraya [The Late Pleistocene of the Cave of the Boquete de Zafarraya] Barroso-Ruiz C, editor. Seville: Junta de Andalucia; 2003. pp. 113–133. Spanish. [Google Scholar]
- 36.Michel V, et al. In: La grotte du Boquete de Zafarraya, Málaga, Andalousie [The Cave of the Boquete de Zafarraya, Malaga, Andalucia] Barroso-Ruiz C, editor. Seville: Junta de Andalucia; 2006. pp. 487–518. French. [Google Scholar]
- 37.Barroso-Ruiz C, et al. In: El Pleistoceno Superior de la Cueva del Boquete de Zafarraya [The Late Pleistocene of the Cave of the Boquete de Zafarraya] Barroso-Ruiz C, editor. Seville: Junta de Andalucia; 2003. pp. 509–520. Spanish. [Google Scholar]
- 38.Michel V, et al. Les derniers Homo heidelbergensis et leurs descendants les néandertaliens: Datation des sites d’Orgnac 3, du Lazaret et de Zafarraya. C R Palevol. 2011;10(7):567–587. [Google Scholar]
- 39.Grün R, Schwarcz HP, Chadam JMM. ESR dating of tooth enamel: Coupled correction for U-uptake and U-series disequilibrium. Nucl Tracks Radiat Meas. 1988;14(1-2):237–241. [Google Scholar]
- 40.Millard AR, Hedges REM. A diffusion-adsorption model of uranium uptake by archaeological bone. Geochim Cosmochim Acta. 1996;60(12):2139–2152. [Google Scholar]
- 41.Penkman KEH, Kaufman DS, Maddy D, Collins MJ. Closed-system behaviour of the intra-crystalline fraction of amino acids in mollusc shells. Quat Geochronol. 2008;3(1–2):2–25. doi: 10.1016/j.quageo.2007.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Van Klinken GJ. Bone collagen quality indicators for palaeodietary and radiocarbon measurements. J Archaeol Sci. 1999;26(6):687–695. [Google Scholar]
- 43.Lines Escardó A. In: Climates of Northern and Western Europe. Wallén CC, editor. Amsterdam: Elsevier; 1970. pp. 192–239. [Google Scholar]
- 44.Walker MJ, et al. The excavation of buried articulated Neanderthal skeletons at Sima de las Palomas (Murcia, SE Spain) Quat Int. 2011;259(9):7–21. [Google Scholar]
- 45.Hoffmann DL, Pike AWG, Wainer K, Zilhão J. New U-series results for the speleogenesis and the Palaeolithic archaeology of the Almonda karstic system (Torres Novas, Portugal) Quat Int. 2012 doi: 10.1016/j.quaint.2012.05.027. [DOI] [Google Scholar]
- 46.Zilhão J, et al. Symbolic use of marine shells and mineral pigments by Iberian Neandertals. Proc Natl Acad Sci USA. 2010;107(3):1023–1028. doi: 10.1073/pnas.0914088107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Andersen KK, et al. The Greenland ice core chronology 2005, 15-42 ka. Part 1: Constructing the time scale. Quat Sci Rev. 2006;25(23-24):3246–3256. [Google Scholar]
- 48.Svensson A, et al. The Greenland ice core chronology 2005, 15-42 ka. Part 2: Comparison to other records. Quat Sci Rev. 2006;25(23-24):3258–3267. [Google Scholar]
- 49.Reimer PJ, et al. IntCal09 and Marine09 radiocarbon age calibration curves, 0-50,000 years calBP. Radiocarbon. 2009;51(4):1111–1150. [Google Scholar]
- 50.Bronk Ramsey C. Bayesian analysis of radiocarbon dates. Radiocarbon. 2009;51(1):337–360. [Google Scholar]
- 51.Mallol C, Hernández CM, Machado J. The significance of stratigraphic discontinuities in Iberian Middle-to-Upper Palaeolithic transitional sites. Quat Int. 2011;275:4–13. [Google Scholar]
- 52.Aubry T, et al. Palaeoenvironmental forcing during the Middle-Upper Palaeolithic transition in central-western Portugal. Quat Res. 2011;75(1):66–79. [Google Scholar]
- 53.Bronk Ramsey CB, Higham TFG, Leach P. Towards high-precision AMS: Progress and limitations. Radiocarbon. 2004;46(1):17–24. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.