New Data on the Late Neandertals: Direct Dating
of the Belgian Spy Fossils
Patrick Semal,1* He´le`ne Rougier,1,2,3* Isabelle Crevecoeur,1,3 Ce´cile Jungels,1 Damien Flas,4
Anne Hauzeur,5 Bruno Maureille,3 Mietje Germonpre´,1 Herve´ Bocherens,6 Ste´phane Pirson,1
Laurence Cammaert,1 Nora De Clerck,7 Anne Hambucken,1 Thomas Higham,8
Michel Toussaint,9 and Johannes van der Plicht10
1Department of Paleontology, Royal Belgian Institute of Natural Sciences, 1000 Brussels, Belgium
2Department of Anthropology, California State University Northridge, Northridge, CA 91330
3Laboratoire d’Anthropologie des Populations du Passe´, PACEA–UMR 5199, Universite´ de Bordeaux,
33405 Talence, France
4Department of Prehistory, Royal Museums of Art and History, 1000 Brussels, Belgium
5Re´gion Centre, Institut National de Recherches Arche´ologiques Pre´ventives, 37100 Tours, France
6Biogeologie, Institut fu¨r Geowissenschaften, Universita¨t Tu¨bingen, 72076 Tu¨bingen, Germany
7Microtomography Research Group, Department of Biomedical Sciences, Universita¨t Antwerpen,
2610 Antwerpen, Belgium
8Oxford Radiocarbon Accelerator Unit, Research Lab for Archaeology and the History of Art,
University of Oxford, Oxford OX1 3QY, England
9Direction de l’Arche´ologie, Ministe`re de la Re´gion wallonne, 5100 Namur, Belgium
10Center for Isotope Research, Groningen University, 9747 AG Groningen, Netherlands
KEY WORDS northwest Europe; Middle to Upper Paleolithic transition; acculturation
ABSTRACT In Eurasia, the period between 40,000
and 30,000 BP saw the replacement of Neandertals by
anatomically modern humans (AMH) during and after
the Middle to Upper Paleolithic transition. The human
fossil record for this period is very poorly defined with
no overlap between Neandertals and AMH on the basis
of direct dates. Four new 14C dates were obtained on the
two adult Neandertals from Spy (Belgium). The results
show that Neandertals survived to at least
36,000 BP
in Belgium and that the Spy fossils may be associated to
the Lincombian–Ranisian–Jerzmanowician, a transi-
tional techno-complex defined in northwest Europe and
recognized in the Spy collections. The new data suggest
that hypotheses other than Neandertal acculturation by
AMH may be considered in this part of Europe. Am J
Phys Anthropol 138:421–428, 2009. VC 2008 Wiley-Liss, Inc.
The replacement of Neandertals by anatomically mod-
ern humans (AMH) in Eurasia is one of the major issues
in paleoanthropology (Gravina et al., 2005; Orlando
et al., 2006; Trinkaus, 2007). This population process
corresponds to the period between 40,000 and 30,000 14C
BP, during which the transition from the Middle to
Upper Paleolithic took place. Different scenarios have
been proposed for explaining these two major events,
from a certain degree of both biological and cultural con-
tinuity (e.g. Wolpoff et al., 1994) to complete replacement
of Neandertals by modern humans together with either
acculturation or independent technical evolution of the
last Neandertals (Mellars et al., 2007; Zilha˜o et al.,
2008a).
Understanding the biological and/or cultural relation-
ships of the two groups has been limited by the dearth
of well-dated and diagnostic human remains associated
to a secure archeological context. Numerous Neandertal
remains come from old excavations that yielded incom-
plete or no stratigraphic data or for which the dating
and/or associated archeological context have not been
revaluated by current methods. Between 40,000 and
30,000 BP, techno-complexes described as transitional
have been documented across Eurasia. Identifying the
manufacturers of these industries is one of the main
keys for answering the question of the biocultural rela-
tionships between Neandertals and AMH. The only tran-
sitional techno-complex to which diagnostic human
remains have been associated is the Chaˆtelperronian of
southwestern France and northern Spain. The Neander-
tal specimens discovered at Saint-Ce´saire and Arcy-
The first three authors contributed equally to this work.
Grant sponsor: FRS-FNRS; Grant number: 2005-2006/1.5.287;
Grant sponsor: NERC; Grant number: NE/D014077/1.
*Correspondence to: Patrick Semal, Department of Paleontology,
Royal Belgian Institute of Natural Sciences, Vautier Street 29, 1000
Brussels, Belgium. E-mail: patrick.semal@naturalsciences.be or
He´le`ne Rougier, Department of Anthropology, California State Uni-
versity Northridge, 18111 Nordhoff St., Northridge, CA 91330-8244,
USA. E-mail: helene.rougier@csun.edu.
Received 23 July 2008; accepted 12 September 2008
DOI 10.1002/ajpa.20954
Published online 10 November 2008 in Wiley InterScience
(www.interscience.wiley.com).
V
C 2008 WILEY-LISS, INC.
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 138:421–428 (2009)

Renne (France) suggest that at least some of the transi-
tional industries are related to Neandertals. Bordes
(2002) invalidated the Chaˆtelperronian/Aurignacian
interstratification at the two sites where it was proposed,
but this matter is debated again (Mellars et al., 2007,
2008; Zilha˜o et al., 2008a,b).
In the northwest of Europe, the Chaˆtelperronian is not
recognized, but sites located between Wales and south-
ern Poland have provided another transitional techno-
complex, the Lincombian–Ranisian–Jerzmanowician (LRJ;
Flas, 2006; Jacobi, 2007).
Neandertals from only a handful of sites have been
directly dated using AMS radiocarbon techniques.
Among them, those from Engis, El Sidro´n, Feldhofer,
Okladnikov, Spy, and Vindija may be attributed to the
period under examination (Schmitz et al., 2002; Higham
et al., 2006a; Rosas et al., 2006; Toussaint and Pirson,
2006; Krause et al., 2007). Besides dating methodology
problems (Smith et al., 1999; Higham et al., 2006a;
Rosas et al., 2006; Krause et al., 2007), inconsistencies
with the archeological context (Skinner et al., 2005;
Toussaint and Pirson, 2006) may also invalidate the
youngest dates. Finally, the most reliable results for
northwest Europe point to
40–39,000 BP for the young-
est directly dated Neandertal remains (Schmitz et al.,
2002). The oldest occurrence of European AMH (Pes
tera
cu Oase, Romania), at
35,000 BP (Trinkaus et al.,
2003), indicates that there is no clear overlap between
Neandertals and modern humans based on the direct
AMS dating of human fossils.
It is therefore essential to obtain new dates on human
fossils using rigorous protocols to evaluate the various
hypotheses regarding the replacement of Neandertals by
AMH in Europe. In this context, we have directly dated
several human bones and teeth from the Spy Cave col-
lections hosted at the Royal Belgian Institute of Natural
Sciences.
SPY CAVE
The discoveries of Spy Cave and their
importance in Paleoanthropology
Spy Cave is one of the richest prehistoric sites in Bel-
gium. The first human remains were discovered in 1885
during excavations by M. De Puydt and M. Lohest. They
hired a miner, A. Orban, who dug a gallery to reach the
lower levels. Orban worked by candlelight, and his finds
were inventoried outside (Lohest et al., 1925). The miner
was often alone in the field, and the few preserved ar-
chives mention quick excavations. The goal of De Puydt
and Lohest was to prove the contemporaneity of human
specimens with layers yielding remains of extinct spe-
cies. In early July 1886, a statement was recorded in a
hurry (because of the risk of collapse for the gallery) to
testify to the discovery of in situ human fossils. This was
presented in the first ever published monograph on
Neandertal remains, and their study was entrusted to
the paleontologist J. Fraipont (Fraipont and Lohest,
1887). Since then, many excavations (whether official or
not) have been carried out at Spy (Rougier et al., 2004).
The most extensive one, by F. Twiesselmann (Royal Bel-
gian Institute of Natural Sciences) between 1948 and
1956, encountered only archeologically rich sediments
from disturbed layers. Most of the material coming from
the various excavations has been only partially studied
and remains unpublished.
Reassessment of Spy Cave and its contents
The work of De Puydt and Lohest (1887) allows us to
reconstruct partially the context of the human remains.
The first skeleton discovered (Spy I) was quite incom-
plete and its position is difficult to assess, since the
bones were not in anatomical connection; the second
skeleton (Spy II) was also incomplete, but lying on the
right side, with one hand against the mandible. A.
Hrdlicˇka, who had the opportunity to study the Spy
human remains in the 1910s and 1920s, noted that ‘‘A
repeated critical examination of the specimens leaves a
serious doubt as to the accuracy of [Fraipont’s] distribu-
tion. No photographs or sketches were made on the spot;
the bones were not marked, and have evidently become
mixed up, their distribution being decided upon later.
The specimens indicate very strongly different relations.’’
(Hrdlicˇka, 1930, p 188).
Recent reassessment of all the collections from Spy
has allowed us to discover numerous unpublished
human remains from Neandertals and AMH mainly
among the unsorted fauna (Rougier et al., 2004). The
number of human remains from Spy has increased from
89 numbered bones and teeth to 1,816 (including at least
24 new Neandertal remains), representing all skeletal
elements. The minimum number of individuals (MNI) is
26, the great majority of which probably derives from a
collective Neolithic burial. The large number of recent
remains questions the homogeneity of the original Spy
collection. Beside the two Neandertal individuals recog-
nized in 1886, only one additional immature individual
had been mentioned (Twiesselmann, 1971), and Trinkaus
(1978) suggested the existence of at least three adult
Neandertal individuals based on the foot bones.
The morphometrical analysis of the newly recognized
human remains has allowed us to compare these new
specimens with the two adult Neandertals to further
consider issues of attribution. The following results have
emerged: 1) the original collection includes mature and
juvenile anatomically modern bones; 2) the attribution of
the remains to each adult Neandertal skeleton has been
reassessed; 3) among the numerous human remains rec-
ognized within the faunal collections, some show charac-
teristics that bring them closer to the range of Neander-
tal variability. For instance, only five of the thirteen
hand bones from the original human collection can be
attributed to Neandertals. IC discovered ten more Nean-
dertal hand bones, which can be associated with the five
from the original collection based on morphometry and
anatomical connections. The new human remains that
refit with the original collection are of particular inter-
est, since they were discovered among unwashed fauna,
which makes them available for isotopic analyses, unlike
the original, consolidated, and varnished specimens.
Stratigraphical and archeological context of the
Neandertal skeletons
The discoverers identified only three fauna-bearing
levels. The two Neandertal individuals were found in the
deepest level, above brown clay lying on the bedrock and
below thin yellow clay (Fraipont and Lohest, 1887; De
Puydt and Lohest, 1887). The hard reddish breccia of
the intermediate fauna-bearing level lying just above the
skeletons was an argument for unmixed deposits. The
possibility of a burial was rejected by the discoverers but
commonly inferred later (Rutot, 1909; Maureille and
422 P. SEMAL ET AL.
American Journal of Physical Anthropology