Elsevier

Journal of Human Evolution

Volume 50, Issue 3, March 2006, Pages 347-358
Journal of Human Evolution

ESR and U-series analyses of enamel and dentine fragments of the Banyoles mandible

https://doi.org/10.1016/j.jhevol.2005.10.001 Get rights and content

Abstract

The Banyoles mandible presents a puzzle. Its anatomy has been described as pre-Neandertal, but the travertine in which it was found has been dated to 45,000 ± 4000 years. By this time, any pre-Neandertals had supposedly been absent from the European fossil record for more than 100,000 years. It was therefore proposed that the age of the travertine may represent a minimum age estimate, with the mandible possibly having been reworked from older deposits.

We carried out a non-destructive ESR analysis of an enamel fragment removed from a molar and performed a series of in situ laser ablation U-series analyses on dentine fragments adjacent to the enamel piece. The analyses resulted in an apparent combined ESR-U-series age of 66,000 ± 7000 years. The encasing travertine matrix was also analyzed for U-series isotopes and showed signs of U-mobilization. It cannot be excluded that the travertine matrix is older than the previously determined age. If the mandible was not reworked, then the combined ESR-U-series result on the tooth enamel would give its best age estimate. If, on the other hand, the mandible was reworked from another deposit, the actual ESR-U-series age will depend on the external dose rate from the previous matrix and the depth of its burial, which controls the degree of the attenuation of the cosmic dose rate over time. Considering a range of possible burial histories, the mean age of the mandible would lie somewhere between the combined ESR-U-series age and the previously determined age of the travertine matrix.

Regarding the morphology of the mandible, a review of its features in the context of larger Neandertal samples indicates that the anatomy of the specimen is not incompatible with such a young age determination, although it further highlights morphological variation in the late Neandertal sample.

Introduction

In April 1887, a human mandible was found by the stonemason Llorenç Roura during quarrying activities near the town of Banyoles (Bañolas), Catalonia, Spain. The mandible was passed on to Pere Alsius, the well-known local pharmacist and naturalist, the pioneer of prehistory of Catalonia, who published its first description (Alsius, 1915). The descendants of Pere Alsius are still the owners and custodians of this specimen (and are, indeed, also pharmacists). Figure 1 shows a map of the location, along with a photo of the travertine section and the approximate depth (4 m below the surface) at which the mandible was found (Alsius, 1915). This travertine section was completely removed a long time ago and, at present, it is not possible to correlate this site to any of the accessible travertine formations.

The mandible (Fig. 2) is nearly complete but is cracked along the left corpus and ascending ramus and damaged in the superior parts of the ascending ramus on both sides. It is most likely that the mandible was initially complete and that the missing parts were lost during the extraction of the mandible from the site of discovery. All the fractures appear to be post-mortem (Hernández-Pacheco and Obermaier, 1915, Maroto and Soler, 1993). The crowns or roots of all the teeth are in place, but the dentition shows peculiar spacing and extraordinary occlusal wear, especially marked on the buccal surfaces of the first and second molars. The mandible is certainly non-modern in morphology and large in terms of its corpus thickness and dental size (as far as can be inferred given the extreme wear). Its taxonomic assignment remains controversial. Originally, it was considered as a fairly typical Neandertal (Hernández-Pacheco and Obermaier, 1915, MacCurdy, 1915, Sergi, 1917/8). However, others emphasized its large size or the absence of derived Neandertal features, such as a retromolar space and posterior position of the mental foramina, implying the lack of midfacial projection. These researchers concluded that its morphology was more likely “pre-Neandertal” (Bonarelli, 1916, de Lumley, 1971/2, de Lumley, 1973, de Lumley, 1982, Stringer et al., 1984, Turbón, 1987, Roth, 1989, Roth and Simon, 1993) or, alternatively, demonstrated the continuity of primitive characters with those of later Neandertals (Hublin, 1988). After the travertine matrix adhering to the mandible was dated to 45,000 ± 4000 years using U-series dating (Julià and Bischoff, 1991; see below), suggestions of a pre-Neandertal morphology caused some unease because no other pre-Neandertal sites in Europe have yielded age estimates anywhere near as young. Based on its age, some workers have grouped the Banyoles mandible with Neandertals (e.g., Stringer and Grün, 1991, Sánchez, 1993, Strauss et al., 1993), while, alternatively, it has been suggested that the date of the travertine should be regarded as a minimum age because the mandible could be significantly older than the enclosing matrix due to reworking processes (Lalueza et al., 1993).

The travertine matrix of the mandible was removed in 1956 (Alcobé, 1993) and was first dated by radiocarbon to 17,600 ± 1000 BP (UCLA-930; Berger and Libby, 1966). This date includes a 2400 year correction for the dead carbonate fraction. Berger and Libby (1966) regarded the ages as too young for a Neandertal and suggested that either more modern travertine had contaminated older travertine or that the mandible was reworked from older deposits. Yokoyama et al. (1987) reported U-series dates on one travertine and one bone sample from a supposedly equivalent position in a nearby quarry, yielding closed system U-series ages of 73,000 ± 4000 years for the travertine (by α-spectrometry) and for the bone, a Th/U age of 16,200 ± 3200 years and a Pa/U age of 17,100 ± 900 years (γ-spectrometry). Using the above-mentioned radiocarbon age, Yokoyama et al. (1987) engaged in some open system modeling, which resulted in a best age estimate of 110,000 20,000 + 55,000 years for the travertine. Julià and Bischoff (1991) carried out a systematic U-series dating study on the travertine matrix, as well as samples from quarries in the near vicinity of the original site of the mandible, yielding an isochron age of 45,000 ± 4000 years for the matrix of the mandible, and isochron ages between about 50,000 (near the surface) to 140,000 years (base at about 2.8 m depth below the surface) for the Les Prederes quarry (about 200 m SW of the original location of the mandible) and 45,000 (surface) to 70,000 years (base at 4.7 m) for the Mata quarry (about 700 m W of the mandible location). The exact locality of the Yokoyama travertine sample is not known, but the result is compatible with those from the Les Prederes quarry. Brusi (1993) studied the complex travertine formations of the Banyoles basin and placed the site of the mandible in a travertine overspill called Estunes-Formiga, which contains the Les Prederes quarry. Maroto (1993) published a comprehensive review on all aspects of scientific research on the mandible up to that date including, for example, the notes of Alcobé (written in 1956 and published in 1993) on how the matrix was removed from the mandible.

In order to address the question of whether the mandible could have been reworked from older units, we carried out an ESR analysis on an enamel fragment removed from the right M3 (Fig. 3A). The analysis of the fragment followed procedures similar to those that were used to establish the antiquity of the BC5 mandible from Border Cave (Grün et al., 2003). On the dentine, laser ablation in situ U-series analysis was carried out. A small piece of the original travertine matrix was collected (Fig. 3B) for the estimation of the external dose rate (K, U, Th, and porosity), as well as for additional mass spectrometric U-series analysis. To make sure that the travertine sample was indeed from the encasing matrix, the sample was re-fitted to a cast of the mandible (Fig. 3C).

Section snippets

Methods

The enamel fragment was mounted in a Bruker ER 218PG1 programmable goniometer and measured at each dose step at 10° angle intervals for 400° (the spectra past 360° were used to check for short-term fading effects). ESR measurements were carried out on a Bruker Elexys spectrometer with a 15 kG magnet and a rectangular 4102 ST cavity. The samples were recorded with the measurement parameters routinely applied at the ANU ESR laboratory: accumulation of between 500 (natural sample) and 200 scans

Results

Figure 5A shows the U and Th scans of a cross section of the enamel piece. The measured average concentrations in the enamel were 0.19 ± 0.18 ppm U and 0.015 ± 0.034 ppm Th. Initially, we were unable to obtain a dentine sample for further analysis, so we decided to scan the interior surface of the enamel (Fig. 5B). The surface contains significantly higher U-concentrations than the enamel cross section, fluctuating between about 3 and 10 ppm, and reaching up to 21.8 ± 1.8 ppm in one 100 μm diameter

Discussion

Although it might be surprising that the dentine shows more than a hundredfold higher U-concentration than the matrix, a similarly high U-concentration (43 ppm) was measured on a bone by Yokoyama et al. (1987). This result is most likely due to the high porosity of the travertine, which allows groundwater to circulate and mobilize uranium quite freely. The open-system nature of the travertine is clearly evident in the U-series data (see Fig. 7B). The 232Th-normalized U-series data do not fall on

Conclusions

The travertine matrix of the Banyoles mandible shows signs of open-system behavior, and thus it cannot be ruled out that its true age is older than 50,000 years. If this is the case, the best age estimate for the mandible is given by the combined U-series/ESR result of 66,000 ± 7000 years. If the mandible was reworked from another deposit, its mean age would not exceed 60,000 years and would most likely be only a few thousand years older than its matrix. However, there is no taphonomic evidence

Acknowledgments

We thank the Alsius family for providing access to the mandible and allowing us to carry out our analyses. We thank Àngel Galobart, Institut de Paleontologia “M. Crusafont” of Sabadell, for having the courage to remove the enamel and dentine fragments from the mandible, and Joaquim Soler, Universitat de Girona, for helping us to overcome any language problems in our collaboration. We thank Daniel Turbón, Secció d'Antropologia of Universitat de Barcelona, for providing the travertine matrix

References (52)

  • C. Lalueza et al.

    Microscopic study of the Banyoles mandible (Girona, Spain): diet, cultural activity and toothpick use

    J. Hum. Evol.

    (1993)
  • X. Luo et al.

    High precision 230Th/238U and 234U/238U measurements using energy-filtered ICP magnetic sector multiple collector mass spectrometry

    International Journal of Mass Spectrometry and Ion Processes

    (1997)
  • J.R. Prescott et al.

    Cosmic ray and gamma ray dosimetry for TL and ESR

    Nuclear Tracks and Radiation Measurement

    (1988)
  • J.R. Prescott et al.

    Cosmic ray contributions to dose rates for luminescence and ESR dating: large depths and long-term time variations

    Radiation Measurements

    (1994)
  • R.M. Quam et al.

    Human remains from Valdegoba cave (Huérmeces, Burgos, Spain)

    J. Hum. Evol.

    (2001)
  • C.H. Stirling et al.

    High-precision in situ 238U-234U-230Th isotopic analysis using laser ablation multiple-collector ICPMS

    Geochimica et Cosmochimica Acta

    (2000)
  • S. Alcobé

    Mandíbula de Bañolas. Diario de trabajo

  • P. Alsius

    De la barra humana descoberta anys enrera a Banyoles

    Bulletí de la Institució Catalana d'Història Natural, 2a època

    (1915)
  • G. Bonarelli

    La mandíbula humana de Bañolas

    Physis 2 (Buenos Aires)

    (1916)
  • R. Berger et al.

    UCLA radiocarbon dates V

    Radiocarbon

    (1966)
  • Brusi, D., 1993. Les formacions travertíniques de la depressió de Banyoles. Ph.D. Dissertation, Universitat Autònoma de...
  • V. Cherdyntsev et al.

    Die Altersbestimmung der Travertine von Weimar-Ehringsdorf

    Abhandlungen des zentralen Geologischen Instituts

    (1975)
  • T. Cortada

    Estudi comparatiu del desgast dentari de la mandíbula de Banyoles des la perspectiva de l'anàlisi macroscòpica de les abrasions alimentàries

  • H. de Lumley

    Origine et Évolution de l'Homme

    (1982)
  • M.A. de Lumley

    La mandíbula de Bañolas

    Ampurias

    (1971/2)
  • M.-A. de Lumley

    Anténéandertaliens et Néandertaliens du bassin Méditerranéen occidental Européen

    Études Quaternaires

    (1973)
  • Cited by (47)

    • Reassessment of the human mandible from Banyoles (Girona, Spain)

      2023, Journal of Human Evolution
      Citation Excerpt :

      The travertine samples yielded an age estimate of 42.5 ± 4.1 ka, in good agreement with the results of Julià and Bischoff (1991), whereas the enamel/dentine samples yielded a somewhat older age estimate of 66.0 ± 7.0 ka. This discordance between the enamel/dentine age and the age of the adhering travertine may indicate the mandible was reworked, although Grün et al. (2006) do not favor this hypothesis. Even if it were reworked from the surface of an older travertine, it is unlikely to be more than a few thousand years older than the age of the adhering travertine (Grün et al., 2006).

    • The first direct ESR dating of a hominin tooth from Atapuerca Gran Dolina TD-6 (Spain) supports the antiquity of Homo antecessor

      2018, Quaternary Geochronology
      Citation Excerpt :

      This has been made possible by the combination of ESR measurements of enamel fragments with high resolution laser ablation (LA) ICP-MS U-series analyses. This “semi non-destructive” procedure was first tested to a human tooth from the Middle Pleistocene site of Florisbad (Grun et al., 1996), and has since been applied to a significant number of human remains, including the Late Pleistocene remains from Mungo (Thorne et al., 1999), Border Cave (Grün et al., 2003), Banyoles (Grün et al., 2006) and El Sidron (Torres et al., 2010), the last interglacial fossils from Skhul (Grün et al., 2005) and Tabun (Grün and Stringer, 2000; Coppa et al., 2005), and hominin teeth from the Middle Pleistocene site of Thomas Quarry (Raynal et al., 2010) and from the Early Pleistocene site of Swartkrans (Curnoe et al., 2001). More recently, ESR dating was the only way to obtain a direct age for the early ancestor of modern humans at Irhoud (Richter et al., 2017) and Misliya (Hershkovitz et al., 2018), and for Homo naledi (Dirks et al., 2017) that are all beyond the radiocarbon time range.

    • U-series dating and classification of the Apidima 2 hominin from Mani Peninsula, Southern Greece

      2017, Journal of Human Evolution
      Citation Excerpt :

      One has to keep in mind that U-series estimates are generally minimum ages (Grün et al., 2014), as it is often possible that faunal material went through a phase with little or no U-uptake. Without further independent chronological evidence, such as ESR measurements (e.g., Grün et al., 2005, 2006), it is difficult to establish an upper age bracket. Thus, our minimum age of 160 ka narrows down the geological age estimate provided by Harvati et al. (2011) and is consistent with the maximum cultural age provided by Andreikos (1997).

    View all citing articles on Scopus
    View full text