Vous êtes sur la page 1sur 18

Valsequillo Pleistocene archaeology and

dating: ongoing controversy in Central


Mexico
Silvia Gonzalez, David Huddart and Matthew Bennett
Abstract
A review of the Valsequillo archaeological nds from the last century is given, such as the cranial
remains (Dorenburg and Ostrander skulls), the engraved bone fragments, butchering marks, green
bone fractures and int points and scrapers. However, most of these nds are now missing. Their
original dating is reviewed, along with the controversial Uranium Series and Fission Track dates
from Hueyatlaco. Further relative dating techniques, such as the use of diatoms and bone
assemblages, are discussed. Recently human and animal footprints from the Xalnene Ash at
Toluquilla quarry have been described, mapped and dated by optically stimulated luminescence, and
there has been new dating of the Valsequillo Gravels by AMS radiocarbon dating of molluscs
and electron spin resonance of mammoth bone. The Xalnene Ash dates have proved controversial
and the dating issues are reviewed. A suggested dating framework for the Valsequillo sequence in
which the archaeological artefacts and footprints are found is given and all the stratigraphy with
archaeological remains is considered to be Late Pleistocene in age.
Keywords
Archaeology; human footprints; dating; Xalnene Ash; Pleistocene stratigraphy.
Introduction
North American archaeologists generally have rejected any claims that the colonization of
the American continent occurred prior to a 11,500 BP date (Meltzer 1993). While many
much older sites have been claimed, the sceptics ask dicult questions, there is usually
erce debate and generally a rejection of the sites proposed antiquity. Such has been the
case at Valsequillo in Central Mexico where controversy originated in the 1960s, continued
through the 1980s, after the deposits were originally dated, and the debate has now
resurfaced after further claims for the sites antiquity (Renne et al. 2005; Gonzalez et al.
World Archaeology Vol. 38(4): 611627 Debates in World Archaeology
2006 Taylor & Francis ISSN 0043-8243 print/1470-1375 online
DOI: 10.1080/00438240600963155
2006; Largent 2006; Adler 2006). In this paper there is a review of this ongoing debate
over the age of the Valsequillo archaeology, with an initial discussion of the earlier
nds and dates, before a summary of the suggested geological events and their
archaeology. The paper concludes with a summary of the further dating needs for the
Valsequillo area.
Valsequillo is about 100km south east of Mexico City, south of the city of Puebla, where
a modern reservoir constructed in the 1940s mirrors the position of a Pleistocene lake in
the area. Around its shores and in quarries there is a sequence of uvial, lacustrine and
volcanic deposits in which mammalian bones and human artefacts have been found. Most
nds are on, or near, the Tetela peninsula in the Valsequillo Gravels (Irwin-Williams 1978:
9) (Fig. 1). However, reservoir, high water conditions have plagued the archaeological
excavations during the 1960s at, for example, the El Mirador site (Irwin-Williams 1978:
14) and more recently at the Hueyatlaco site (Ochoa-Castillo et al. 2003: 62).
Finds of cranial remains in the Valsequillo Basin
Two primitive looking human skulls have been reported to come from the area. The
Dorenburg skull was collected over a century ago south of Puebla (Reichelt 1899). It was
on display in a Leipzig Museum but unfortunately was destroyed during the Second
World War. A sample of diatomite taken from within the skull cavity, preserved as a
reference slide at the California Academy of Sciences (San Francisco), shows the same
Figure 1 Location of the Valsequillo Basin and important early archaeological sites mentioned in the
text.
612 Silvia Gonzalez et al.
extinct diatom taxa and diatom suite as in the sediments at Hueyatlaco, dating both by
biostratigraphy to be older than 80,000 years (VanLandingham 2004).
The Ostrander skull was reported to have been illegally collected by someone unknown
at Hueyatlaco in the 1960s or 1970s. It was preserved at Merced College, California, but
was probably re-buried by a local Native American tribe without any attempt to date it. It
was a partial skull with thick brow ridges and low cranium. These cranial remains perhaps
indicate the potential of the Valsequillo area for unravelling the antiquity of man in the
Americas, despite their unfortunate loss to the scientic community. The pattern of nding
important prehistoric archaeology that has later disappeared or been destroyed seems to
be a constant theme for the area. This situation makes impossible a re-evaluation of these
early nds.
Early work prior to the 1960s
In more than thirty years of eldwork in the area south of Puebla, Armenta Camacho
(1957, 1959, 1978) documented over 100 sites with mammoth, mastodon and other large,
extinct mammals, at times with evidence of human presence. There have also been reports
of animals such as camel, bison, horse, four-horn antelope, peccary, sloth, glyptodont,
short-faced bear, sabre-toothed cat and dire wolf, and in general the area is considered as
very important from the point of view of Late Pleistocene vertebrate palaeontology
(Osborn 1905; Gu enther 1968; Gu enther et al. 1973). The evidence for human presence
came from butchering marks, green-bone fractures, engravings and a mammoth jaw with
an embedded int point. The rst nd by Armenta Camacho came in June 1933, when a
mammoth leg bone was found in an eroding gully and in 1935 he discovered a mastodon
leg bone with an embedded int eroding from the left bank of the Rio Alseseca (Steen-
McIntyre 1998a: 32). In May 1959 he found a triangular fragment of mastodon pelvis with
what he interpreted as an array of engravings (Armentas sample Tetela 1, from
Hueyatlaco, Fig. 2), including a serpent head, sabre-toothed cat leaping upon, or over,
Figure 2 An example of engraved mastodon bone (Tetela 1) from the collection of Juan Armenta
Camacho (after Armenta Camacho 1978).
Valsequillo Pleistocene archaeology and dating 613
rhynchotheres and, in his sample Atepetzingo 1, a proboscidian longbone with a large
rhyncothere head (Rhyncotherium is one of the three commonest genera of the
Gomphotheriidae family of elephant ancestors recorded in the Mexican Pliocene and
Pleistocene (Alberdi and Corona-M. 2005); diering from elephants in their tooth
structure, most had four tusks and their retracted facial and nasal bones suggest that the
Gomphotheres had elephant-like trunks), trunk and tusks, crisscrossing, or criss-crossed
by a confusion of small gures. Armentas sample Tetela III was a longbone fragment
whose crisscrossing pattern shows a feline head and apparently a number of rhynchotheres
among other gures. However, rhynchotheres apparently did not survive to the
Pleistocene according to Lambert and Shoshani (1998: 612).
Unfortunately, most of this extensive butchered bone collection and the engravings
went missing under dubious circumstances after they were lodged at INAH (Instituto
Nacional de Antropologia e Historia) collections in 1966 and Armenta Camacho was
banned from collecting any more material (bones and lithics) from the gullies by the
Mexican authorities. However, there remains some doubt about the authenticity of these
carvings and some of the evidence put forward for kill sites. The problem is that it is
not known what happened to this important collection of archaeological material
(bones and lithics) and if they were destroyed. This is unfortunate as they could be
examined today with modern technology to try to solve the question of their authenticity
and age.
Archaeological excavations in the 1960s and 1970s
The initial survey and excavation began in 1962 in the lower part of a dissected alluvial
formation about 30m thick, known as the Valsequillo Gravels, and four sites were
reported and excavated (El Horno, El Mirador, Tecacaxco and Hueyatlaco; see Fig. 1).
Stratigraphic sections were sequenced (Irwin-Williams 1967a: 339), while the nal
excavation was concluded in 1973. Two distinct stone tool types were reported to occur
at the Hueyatlaco site separated by a layer with no artefacts. The lower, older El Horno
site contains only unifacial akes, or blade-points (Irwin-Williams 1967a, 1978; Szabo
et al. 1969), while the upper level contains bifacial tools, scrapers and tanged projectile
points (Fig. 3). All the artefacts are of non-local origin. Irwin-Williams (1978)
characterized the archaeological site as an area that contained a kill-site, with activities
indicating hunting, butchering and camping. She considered there to be three periods of
occupation.
However, controversy began in 1967 when Jose Luis Lorenzo (Director of Prehistory at
INAH) launched allegations regarding the integrity of the Valsequillo project and its
leader and other archaeologists, including the authenticity of the artefacts retrieved from
Hueyatlaco, accusations of planted artefacts by labourers working on the site and mixing
of artefacts (Lorenzo 1967). This was totally refuted by Irwin-Williams (1967b, 1969).
Independently INAH also excavated a trench in the site which was archaeologically barren
very close to the original excavations in 1966, but this perhaps is not surprising given that
the archaeological nds are all in reworked, channelized, river gravels and it is only in the
Hueyatlaco lapilli channel that the archaeological artefacts were found. There was an
614 Silvia Gonzalez et al.
apparent lack of understanding of the site formation processes at the time and of the
sedimentology, where a lot of volcanic ash and volcanic mudows (lahars) are found in
association with the archaeological nds in a uvial channel.
Malde (1966) and Malde and Irwin-Williams (1967) compared ages of dated sequences
based on molluscs in the barrancas (gullies) close to the Tetela peninsula with the
archaeological sequences and, although the inferred original continuity of the deposits is
interrupted by erosional gaps, because the gaps were short, and because of lithological
similarities, correlations were made. The gravelly base of the Valsequillo Gravels at El
Horno was suggested to be over 35,000 years old and the middle part of the sequence at
Hueyatlaco, with the bifacial artefacts, was thought to correspond with the middle
Barranca Caulapan gravels, where a shell date of 22,000 years old was associated with a
scraper. Furthermore the Hueyatlaco lapilli channel was correlated with the La Malinche
pyroclastic ow at Barranca Xotanacatla. Here the ow buried an organic soil dated to
25,920 +1,000 (W-1911) and at Barranca Angostura an equivalent soil was dated to
23,940 +1000 (W-1908). The Buena Vista lapilli was suggested to be equivalent to an
extensive blanket (6m thick) of volcanic sand which occurred between two phases of
glacial advances on La Malinche. The soil on the underlying glacial sediment was dated to
8,240 +300 (W1909) and, although two other dates were shown to be slightly younger, it
was considered that this eruption occurred around 8,000 years ago. It was correlated
petrographically with the supercial drape of ash and lapilli that mantles the northern end
of the Tetela peninsula. Unfortunately, Steen-McIntyre could never be convinced of the
exact match and correlations of the volcanic ashes on the Tetela peninsula with those
related to specic volcanic episodes on La Malinche and Popocatepetl (Steen-McIntyre
and Malde 1970: 77) and she believed that the equivalent, Hueyatlaco tephras must be
buried deep on the anks of La Malinche.
Figure 3 Technology and typology of artefacts from ve Paleoindian sites in the Valsequillo Basin,
Central Mexico (after Szabo et al. 1969).
Valsequillo Pleistocene archaeology and dating 615
One of the main problems of the Pleistocene vertebrate bones from the Valsequillo area
is the lack of collagen to allow direct radiocarbon dating and, for this reason, other dating
methods were used, like Uranium Series. Also some of this initial dating of the associated
animal bones (Szabo et al. 1969: 243) proved controversial and the original authors noted
that some of the dates were perhaps too old, stating that we cannot explain why some of
these dates are much older than expected from the archaeological evidence. For example,
Irwin-Williams believed that at least two of the Uranium Series dates could not be correct
(M-B-3 at Hueyatlaco and M-B-8 at El Horno) because the stone tools in association with
the dated bone could not have been in use at Valsequillo more than 200,000 years before
the date generally accepted for the development of analogous tools in the Old World
(Irwin-Williams 1978); she believed an age this old was impossible. After Steen-McIntyre
et al. (1981) claried some of the stratigraphic issues, Irwin-Williams became critical of the
dating results, and was able to accept only a much more conservative date of 23,000 years
BP for the archaeology, which was still controversially early. This was based on a single
radiocarbon date from molluscs in association with a single int scraper at the Barranca
Caulapan, 5km to the east of Hueyatlaco. The reaction from the scientic community to
all this anomalous dating evidence was to reject it (Webb and Clark 1999: 12) and the
Valsequillo area did not gure centrally in the debate related to the origin and dispersion
of humans into the Americas.
Dating of the deposits in the late 1970s
The archaeological site at Hueyatlaco was dated at approximately 250,000 to 275,000
years or older by radiometric dates (U/Th and ssion track dating) of fossil butchered
bone associated with bifacial stone tools and on overlying, stratigraphically younger,
tephra layers (Szabo et al. 1969; Steen-McIntyre et al. 1981; Steen-McIntyre 1998a: 34,
1998b: 50). Two Uranium-Series dates on a butchered mastodon tooth fragment from the
El Horno site, a few meters lower in the sequence agreed with this age (Szabo et al. 1969).
The stone artefacts at Hueyatlaco were recovered from sediments rich in diatoms
containing taxa that became extinct before the end of the Sangamon interglacial, 80,000
years ago (VanLandingham 2004). According to Steen-McIntyre et al. (1981) and Malde
and Steen-McIntyre (1981) the topographic, geological and pedological evidence
supported this age, including the depth of burial of the artefacts and the extent of
dissection of the sediments by the adjacent Rio Atoyac (over 50m). Steen McIntyre (1981:
365, 1985: 277) used relative dating techniques like the tephra hydration method, where
the extent of hydration of the volcanic glass shards included in the sequences is measured
(Steen-McIntyre 1975), the extent of etching of the glass and the weathering of the heavy-
mineral phenocrysts, to obtain an estimate age (250,000 years) for the Buena Vista lapilli
and the Hueyatlaco Ash to support the radiometric ages. However, the dates need to be
considered with caution because spuriously old Uranium Series dates are often
encountered in bone from situations where the more mobile uranium is leached,
increasing the apparent
230
Th/
234
U and a priori assumptions of uranium uptake, such as
the early uptake model employed at the time to date the Tetela peninsula bones, do not
identify, or account for, leaching or recent uptake. They have been shown to be unreliable
616 Silvia Gonzalez et al.
and potentially are likely to give Uranium Series dates grossly in error (Pike et al. 2002).
The large error ranges and the absence of other archaeological sites with similar antiquity
within the Americas has led to these very old dates being rejected by the majority of
archaeologists and palaeontologists. This included the original excavator, Irwin-Williams,
who declared that these dates were incorrect because they could not be correlated with the
lithic types found in the sites. The sites then have largely been ignored, or re-interpreted to
t the ruling paradigm (Irwin-Williams 1978, 1981: 258). The very old dates were
inconsistent with the tool typology (Irwin-Williams 1981) and some dates were in reverse
stratigraphic order (see Steen-McIntyre et al. 1981: Table 2). Nevertheless Irwin-Williams
insisted on a date of around 23,000 years for the site (Dincauze 1984: 88) and this has been
the maximum date believed, but even this has proved controversial, almost doubling the
pre-Clovis entry to the Americas. A dierent approach could be the analysis of the
vertebrate bone assemblages using biostratigraphy and Graham (1978), on the basis of a
bone sample curated at the Smithsonian Museum in Washington, considered the fauna to
be Wisconsin in age. However, Gu enther (1968, 1973) had noted the presence of species
extinct by the end of the Illinoian and Sangamon time and that the fauna was
heterogeneous, with redeposition of older fossils into younger levels at several sites. The
Caulapan dates in the middle of the sequence agree with Gu enthers date (1973: 14) for
Faunal Zone II near Totimehuacan at 26,000+530 (KI-266) and Pichardo (1999) suggests
that the presence of the short-horned, Bison antiquus, with a rst appearance date in North
America from the mid-Caulapan level, establishes that there are at least two distinct
chronofaunas. However, there is no suggestion that the dates from the biostratigraphy are
as old as the U-Series dates or the ssion track dates.
Current excavations at Hueyatlaco
This site was re-excavated again by a group from the Instituto Nacional de
Antropologia (INAH) Mexico in 2000 (Ochoa-Castillo et al. 2003) who reported a
stratigraphic column, with the basaltic Xalnene Ash at the bottom of the channel
sequences, but with no dates. They reported the presence of large amounts of bones,
including camel, horse, dwarf pronghorn and mammoth (2003: 62). Just one small
debitage piece is indisputably of human origin and was found at the same level as the
extinct animal bones but was not in actual contact with the bones. These samples were
located in a silty clay that underlies the Hueyatlaco Ash. In 2003 the group went back to
the site and took additional samples for diatoms, sedimentology and dating but the
results have not yet been published.
Our contribution to the dating and geological framework for Valsequillo
In the summer of 2003 and as part of a larger British research initiative dealing with the
environmental conditions for human evolution and dispersal of humans across the world,
we re-excavated Paleoindian sites in Mexico. This was an attempt to reconstruct paleo-
climatic conditions at the sites, together with radiocarbon and Uranium Series dating
Valsequillo Pleistocene archaeology and dating 617
of paleoindian human remains from the Preceramic human collection curated at the
Museo Nacional de Antropologia in Mexico City (Gonzalez et al. 2003).
We decided to work in the Valsequillo area to try to establish the Quaternary sequence
of the basin, together with a detailed programme of dating the deposits to try to construct
a robust chronology of the area in order to understand further the dating of the early
archaeological sites. During this work we discovered, in an abandoned quarry close to
Cerro Toluquilla, a series of traces preserved on top of a layer of volcanic ash (Xalnene
Ash) that we have interpreted as human footprints (Gonzalez et al. 2006).
Human footprints in Valsequillo and their age
The Toluquilla footprint layer contains human and animal footprint traces preserved on
the upper bedding planes of an indurated, coarse, basaltic volcanic ash (Xalnene Ash),
which was deposited in the shallow Pleistocene lake (Lake Valsequillo) (Fig. 1). The
Xalnene Ash was deposited from a subaqueous, monogenetic volcano (Cerro Toluquilla)
that erupted within Lake Valsequillo some time around 40 ka (Gonzalez et al. 2006). The
human and associated animal footprints occur on bedding plane surfaces exposed
throughout the eastern end of the Valsequillo Basin, but are best exposed on the oor of
an abandoned quarry close to Cerro Toluquilla (Fig. 1 and Plate 1A). The footprints were
made and preserved during the latest stages of deposition of the ash and are present in
several layers in the top 20cm of the ash succession, where they are interbedded with lake
sediments (Gonzalez et al. 2006). The Xalnene Ash was exposed on lake shorelines during
low stands in the water level, either associated with water displacement during the volcanic
eruption or due to climatically driven uctuations in the water balance. The traces are
associated with desiccation cracks formed during exposure of the lake oor. Humans and
animals traversing the lake shore and exposed lake oor left short trails and individual
prints in the coarse ash (Plate 1). Lake marginal transgressions preserved the traces
through burial by lake sediment and subsequent pulses of volcanic ash. The volcanic ash is
cut by a number of irregular, silt-lled hydrofractures formed by the loading of underlying
saturated lake silts by the ash during deposition. This indicates that the ash has not
undergone immediate post-depositional re-working and the footprints are also cross-cut
by these hydrofractures and desiccation cracks and were formed syn-depositionally with
the ash.
The stratigraphic context and geochronological control for the Xalnene Ash is shown in
Figure 4 and discussed in detail in Gonzalez et al. (2006). The Toluquilla Footprint Layer
has been dated to 38+8.57 ka (sample number: TW04-10) using optically stimulated
luminescence dating of baked, silty xenoliths within the ash which were interpreted as
being baked at the time of the eruption, which reset the time signal. The ash layer is older
than the stratigraphically younger, uvial sediments from the Valsequillo Gravels exposed
in the Barranca Caulapan (Fig. 1) where there are radiocarbon dates between
9.15 +0.5 ka (W1896) from the top of the sequence to 38.9 +0.8 ka (Oxa-14355) on a
mollusc shell at its base, as well as an electron spin resonance date on a mammoth molar
and U-Series dates on bones. All of the dates obtained with dierent methods are in
agreement. There are no weathering horizons, or soils, in these gravels, or above the
618 Silvia Gonzalez et al.
Xalnene Ash at the footprint quarry. It is on this stratigraphic basis that the human
footprints of the Toluquilla Footprint Layer are believed to provide evidence of early
human occupation of the Americas, due to their potential age of more than 40,000 years
ago.
Current controversy related to the age of the Xalnene Ash
However Renne et al. (2005) also dated the Xalnene Ash but with startlingly dierent
results. This unit is the one in which we have interpreted a potential suite of human and
animal footprints. Renne et al. (2005) document that this ash is much older, at 1.3 Ma
Plate 1 Examples of human footprints (BG) at the Toluquilla quarry (A) location.
Valsequillo Pleistocene archaeology and dating 619
years based on single grain Ar-Ar dating, and they suggest that the footprints cannot be
human, because of the age but are the result of quarrying operation. However, Gonzalez
et al. (2006) and Huddart et al. (in press) recognize also the presence of a much younger set
of anthropogenic traces superimposed and cross-cutting the older traces, which have been
interpreted as potential fossil footprints. In both papers a set of well-established criteria
have been used to establish that the older traces on the ash are human. Other footprints
have also been found within the Valsequillo Basin on naturally exposed Xalnene Ash
bedding planes where no quarrying has occurred.
There are further important dating issues. The ash was produced by very explosive
subaqueous eruptions from Cerro Toluquilla, incorporating fragments of the country
rock, giving as a result an extremely heterogeneous ash, not homogeneous as suggested by
Renne et al. (2005). They admit that if the lapilli they dated were derived by erosion of pre-
existing tephra, their age could substantially pre-date deposition. We consider this likely.
Our attempts to date the ash by Ar-Ar dating had no success (Gonzalez et al. 2006). Non-
atmospheric argon in the lapilli was detected yielding apparent ages in the range 2.2
4.6 Ma. The ages gave a saddle shape, commonly associated with extraneous argon
which might be dissolved in the glass, or present as older mineral fragments on the
outsides of the lapilli. A total fusion age for the overlying summit lava did not release
sucient radiogenic argon to determine an age, with insucient potassium concentra-
tions. Additional OSL dating on the palaeolake sediment sequence below the ash suggests
age estimates ranging between 35 and 200 ka, in clear disagreement with the 1.3 Ma date.
There are further geological problems with a 1.3 Ma ash date. If correct, there must be a
basin regional unconformity where a million years of the Pleistocene would be
unaccounted for. However, in the regional geological mapping of the Valsequillo basin
there has been noted a conformable sequence from lower lake sediments, through Xalnene
Ash and upper lake sediments with incorporated lahar units at the Cerro Tepalcayo
quarry, north-east of San Francisco Totimehuacan (Plate 2). Furthermore Sangamonian
Figure 4 Stratigraphic context and geochronological control for the Valsequillo Basin deposits.
620 Silvia Gonzalez et al.
(80220,000 yr BP) diatoms in sediments in the Valsequillo Basin have been reported by
VanLandingham (2004), which we consider to be derived from sediments underlying the
Xalnene Ash. Our view of the Late Pleistocene basin development is that there is no
regional unconformity. This is a conclusion reached after our detailed geological mapping
of the Quaternary lake, uvial and volcanic sequences which was initiated by the
pioneering work of Malde (1968).
The ashs reverse magnetization, suggesting a date older than 0.79 Ma (Renne et al.
2005), needs careful evaluation. There are other possible explanations: 1) self-reversal of
the magnetization common in subaqueous basalts, which contain inhomogeneous
titanomagnetite grains as remanence carriers; 2) the presence of the Laschamp
Geomagnetic Excursion, dated to around 40 ka. It is possible that the reversal noted by
Renne et al. (2005) is not the Brunhes/Matuyama reversal but a much younger one. No
properly orientated samples were reported by these authors and detailed palaeomagnetic
and magnetic mineralogy studies are required to assess the true status of the reverse
magnetization in the ash. Attempts to date the Hueyatlaco site using palaeomagnetism
were unsuccessful (Liddicoat et al. 1981).
Plate 2 Stratigraphic sequence in a quarry at Cerro Tepalcayo, north east of San Francisco
Totimehuacan in the Valsequillo Basin. There are no major unconformities in the sequences above
and below the Xalnene Ash. The lower lake sediments are followed conformably by the Xalnene Ash
and this is overlain, again conformably, by a sequence of marginal lahars and upper lake sediments.
There is no signicant time gap in deposition. There is also no geological evidence for a major
unconformity in the Valsequillo Basin, which is required by Renne et al.s dates. This is a conclusion
reached after detailed geological mapping of the Quaternary lake, uvial and volcanic sequences.
Valsequillo Pleistocene archaeology and dating 621
Thus there are some misgivings about the dates, the palaeomagnetism and the
discussion presented by Renne et al. (2005), but there is obviously the need for further
discussion of the geology, dates, archaeology and footprints in this basin. This is long
overdue given its importance in the understanding of when humans entered the Americas.
The suggested dating framework for the Valsequillo sequence
The following tentative revised dating framework for the Valsequillo archaeological
deposits is suggested (Table 1). It is considered that the older U-Series and ssion track
dates from the Tetela peninsula should be discounted and that the chronology should be
based on the AMS and conventional radiocarbon dates, ESR and U-Series dates from the
Barranca Caulapan and the OSL dates from the Toluquilla quarry area.
It is suggested that the radiometric dates on the volcanic sequence from the western
slopes of La Malinche can be used to date relatively some of the volcanic deposits on the
Tetela peninsula and to the south-east of Puebla. Our ongoing tephrochronological and
other geological work will in the future help to conrm, or to revise this chronology.
Conclusions
1. The Valsequillo Basin is a very important region in the study of early human
presence in the Americas, with dates of at least 22,000 years BP and should be
considered in the models of early peopling of the continent.
Table 1 Dating framework suggested for the Valsequillo Basin sequence
Stratigraphic unit and artefacts Dates
9. Tetela brown mud (lahar); Buena Vista lapilli c. 8k
8. Valsequillo uvial gravels, lahars, and white ash sequence
(Wisconsin)
c. 9kc. 39k
7. Lahars in the Atoyac and Atepizingo valleys and in the
Periferico Ecologico (from La Malinche Volcano)
c. 23k
6. Scraper found in the Valsequillo Gravels at Barranca
Caulapan site.
c. 23k
5. Hueyatlaco white ash, in the Atepinzingo valley and the
convoluted rhyolitic ash in the Periferico Ecologico
c. 25k
4. Hueyatlaco artefacts associated with megafaunal bones
(mammoth, horse and camel)
c. 2539k
3. Upper lake sediments succession c. 2539k
2. Xalnene basaltic ash c. 40k
1. Lower lake sediments succession (Sangamonian) 40200k
Note
The animal bones and artefacts found in the Valseqilllo uvial gravels sequence are all reworked and not in situ
The AMS radiocarbon dates are in situ from a freshwater molluscan fauna from ner grained, overbank sands/
silts found in the sediment sequence
622 Silvia Gonzalez et al.
The human footprints discussed here are an important addition to the story of
the colonization of the Americas. The initial colonization of this continent remains
a contentious issue, with dierent theories as to when it happened and by whom
(e.g. Dixon 1999; Chatters 2000; Fiedel 2000). Some researchers believe the
settlement was 30 ka or older (Bonnichsen and Turnmire 1999); but the most
accepted dates of occupation are in the latest Pleistocene, related to the Clovis
First model, while the oldest directly radiocarbon dated human remains found so
far are around 11.5 ka (Dillehay 2000; Gonzalez et al. 2003). This hypothesis of
human settlement is complicated, however, since the earliest accepted human
occupation date (12.5 ka) comes from the Monte Verde site in southern Chile
(Dillehay 1989, 1997), where lithic technologies are very dierent from the Clovis
sites in the south west of the USA.
2. The human footprints and their dating (Gonzalez et al. 2006) at Toluquilla help to
resolve the controversy related to the age of the archaeological (lithics and worked
animal bone) and associated megafaunal remains that were reported in the
Valsequillo Gravels in the 1960s and 1970s. The latter must be younger than the
footprint level in the revised chronological framework (see Table 1).
3. This work suggests that other sites in Central Mexico with the same type of
fossiliferous gravels associated with lithics, which have been claimed to be of great
antiquity in the past, should be studied, e.g. Tequixquiac in the north of the Basin
of Mexico (Ba rcena 1882). It is suggested by Lettow and Otto (20045) that in
Oaxaca similar human footprints on an ash boulder are as old as between 25,000
and 30,000 years. These potentially early Mexican sites need to be carefully re-
evaluated in a wider continental scale approach, rather than as isolated sites.
Evidence from South America from Pedra Furada (Guidon and Delibrias 1986;
Pessis 1996; Parenti 2001; Santos et al. 2003) with dates up to 50,000 years BP and
Monte Verde 1 with dates around 33,000 years BP (Dillehay and Collins 1988;
Dillehay 1989, 1997, 2000) have also suggested that there may have been earlier
occupation of the Americas than currently accepted.
4. Despite dating controversy in these locations, from the evidence presented herein
and in Gonzalez et al. (2006), the Clovis First model of human occupation can no
longer be accepted as the rst evidence of human presence in the Americas. A new
timing for the initial migrations, as well as new routes of migration need to be
considered.
Future dating needed
There is a clear need for further dating to be obtained on the age of the Xalnene Ash, both
in the Toluquilla Quarry and in the Valsequillo Basin to resolve the current dilemma for
the age of this ash; the underlying lake sediments need to be further dated using OSL and
U-Series dating on the stromatolitic limestones and the volcanic ashes in the Valsequillo
basin need to be correlated using the electron microprobe to obtain the geochemical
composition of the tephras. The need for additional research and renewed eorts to date
the deposits accurately using contemporary dating techniques was suggested by Dixon
Valsequillo Pleistocene archaeology and dating 623
(1999: 97) but, as has been demonstrated, the area remains controversial. Further work on
the palaeomagnetism of the lavas and ashes of the basin needs to be carried out and the
magnetic mineralogy of the sequences needs elaboration to help resolve the apparent
reversal of the Xalnene Ash. This further dating, palaeomagnetism and correlation is
currently ongoing by the authors.
Acknowledgements
We wish to acknowledge the nancial support from NERC (grant number: NER/T/S/
2002/00467). Virginia Steen-McIntyre has provided discussions and invaluable informa-
tion on the Valsequillo research from the 1960s and 1970s, especially an archive of
publications and photos.
Silvia Gonzalez,
School of Biological and Earth Sciences, Liverpool John Moores University, Liverpool
David Huddart,
School of Biological and Earth Sciences, Liverpool John Moores University, Liverpool
Matthew Bennett
School of Conservation Sciences, Bournemouth University, Dorset
References
Adler, R. 2006. The rst Americans. New Scientist, 190(2546): 426.
Alberdi, M. T. and Corona-M. E. 2005. Revisio n de los gonfoterios en el Cenozoico tard o de
Me xico. Revista Mexicana de Ciencias Geologicas, 22: 24660.
Armenta Camacho, J. 1957. Hallazgos Prehistoricos en el valle de Puebla. Centro de Estudios
Histo ricos de Puebla 2.
Armenta Camacho, J. 1959. Hallazgo de un artefacto asociado con mamut en la valle de Puebla,
Mexico Instituto Nacional de Antropologia e Historia, Direccion Prehistoria, 7: 725.
Armenta Camacho, J. 1978. Vestigios de labor humana en huesos de animales extintos de Valsequillo,
Puebla, Mexico. Puebla, Me xico: Consejo Editorial del Gobierno del Estado de Puebla.
Ba rcena, M. 1882. Descripcio n de un hueso labrado, de llama fo sil, encontrado en los terrenos
posterciarios de Tequixquiac. Anales del Museo Nacional de Mexico, 1a epoca, 2: 43944.
Bonnichsen, R. and Turnmire, K. L. 1999. An introduction to the peopling of the Americas. In Ice
Age Peoples of the Americas: Environments, Origins and Adaptations (eds R. Bonnichsen and
K. L. Turnmire). Corvallis, OR: Oregon State University Press, pp. 126.
Chatters, J. C. 2000. The recovery and rst analysis of an early Holocene human skeleton from
Kennewick. Washington. American Antiquity, 65: 291316.
Dillehay, T. D. 1989. Monte Verde: A Late Pleistocene Settlement in Chile, Vol. 1, Palaeoenvironment
and Site Context. Washington DC: Smithsonian Institution Press.
Dillehay, T. D. 1997. Monte Verde: A Late Pleistocene Settlement in Chile, Vol. 2, The
Archaeological Context and Interpretation. Washington DC: Smithsonian Institution Press.
Dillehay, T. D. 2000. The Settlement of the Americas. New York: Basic Books.
624 Silvia Gonzalez et al.
Dillehay, T. D. and Collins, M. 1988. Early cultural evidence from Monte Verde in Chile. Nature,
332: 1502.
Dincauze, D. 1984. An archaeo-logical evaluation of the case for pre-Clovis. In Advances in World
Archaeology, Vol. 3. Seattle, WA: Academic Press, pp. 275323.
Dixon, E. J. 1999. Bones, Boats and Bison: Archaeology and the First Colonization of Western North
America. Albuquerque, NM: University of New Mexico Press.
Fiedel, S. J. 2000. The peopling of the New World: present evidence, new theories, and future
directions. Journal of Archaelogical Research, 8: 39103.
Gonzalez, S., Jimenez-Lopez, J. C., Hedges, R., Huddart, D., Ohman, J. C., Turner, A. and Pompa y
Padilla, J. A. 2003. Earliest humans in the Americas: new evidence from Me xico. Journal of Human
Evolution, 44: 37987.
Gonzalez, S., Huddart, D., Bennett, M. R. and Gonzalez-Huesca, A. 2006. Human footprints in
Central Mexico older than 40,000 years ago. Quaternary Science Reviews, 25: 20122.
Graham, R. 1978. Biostratigraphy, zoogeography and palaeoecology of the late Pleistocene
Valsequillo fauna from archaeological sites in Puebla. Mexico. Abstract, Papers of the 144th
National Meeting of the AAA, 1217 February, 160.
Gu enther, E. W. 1968. Untersuchungen zur jungeiszeitlichen und nacheiszeitlichen geologischen und
pala ontologischen Geschichte. In Das Mexiko-project der deutchen Forschungsgemeinschaft 1:
Berichte uber begonnene und geplante Arbeiten (ed. F. Tichy). Wiesbaden: Franz Steiner Verlag,
pp. 327.
Gu enther, E. W. 1973. Einfuhrung in die Geologischen und palaontologischen untersuchungen im
Valsequillo. Das Mexiko-project der Deutschen Forschungsgemeinschaft 6. Wiesbaden: Franz Steiner
Verlag, pp. 120.
Gu enther, E. W., Bunde, H. and Nobis, G. 1973. Das Mexiko-project der Deutschen
Forschungsgemeinschaft 6: Geologische and Palaontologische Untersuchungen im Valsequillo bei
puebla (Mexiko). Wiesbaden: Franz Steiner Verlag.
Guidon, N. and Delibrias, G. 1986. Carbon-14 dates point to man in the Americas 32,000 years ago.
Nature, 321: 76971.
Huddart, D., Bennett, M. R., Gonzalez, S. and Velay, X. in press. Documentation and preservation
of Pleistocene human and animal footprints: an example from Toluquilla, Valsequillo Basin (Central
Mexico). Ichnos.
Irwin-Williams, C. 1967a. Associations of Early man with horse, camel and mastodon at
Hueyatlaco, Valsequillo (Puebla, Mexico). In Pleistocene Extinctions: The Search for a Cause (eds
P. S. Martin and H. E. Wright). New Haven, CT: Yale University Press, pp. 33750.
Irwin-Williams, C. 1967b. Comments on allegations by J. L. Lorenzo concerning archaeological
research at Valsequillo, Puebla. Paleo-Indian Institute, Eastern New Mexico University Miscellaneous
Publications, 1: 17.
Irwin-Williams, C. 1969. Comments on the association of archaeological materials and extinct fauna
in the Valsequillo region, Puebla, Mexico. American Antiquity, 34: 823.
Irwin-Williams, C. 1978. Summary of archaeological evidence from the Valsequillo region,
Puebla, Mexico. In Cultural Continuity in Mesoamerica (ed. D. L. Browman). London: Mouton,
pp. 722.
Irwin-Williams, C. 1981. Commentary on geological evidence for age of deposits at Hueyatlaco
archaeological site, Valsequillo, Mexico. Quaternary Research, 16: 258.
Lambert, D. W. and Shoshoni, J. 1998. Proboscidea. In Evolution of Tertiary Mammals of North
America (eds C. M. Janis, K. M. Smith and L. L. Jacobs). Cambridge: Cambridge University Press,
p. 612.
Valsequillo Pleistocene archaeology and dating 625
Largent, F. 2006. Toluquilla, Mexico: an American Laetoli? Mammoth Trumpet, 21(2): 1822.
Lettow, H. and Otto, J. 20045. The First Mexican, Stonewatch, 9: 2845. Available at:
5www.stonewatch.de4
Liddicoat, J. C., Coe, R. S., Lambert, P. W., Malde, H. E. and Steen-McIntyre, V. 1981.
Paleomagnetic investigation of Quaternary sediment at Tlapacoya, Mexico, and at Valsequillo,
Puebla, Mexico. Geosica Internacional, 20: 25062.
Lorenzo, J. L. 1967. Sobre me todo arqueolo gico. Boletin of the Instituto Nacional de Antropologia e
Historia, 28.
Malde, H. E. 1966. Valsequillo region, Puebla, Mexico. Unpublished radiometric date report, 12 April.
Malde, H. E. 1968. Preliminary draft on the stratigraphy of Valsequillo region (except Valsequillo
Gravel, volcanic ash stratigraphy, etc.), unpublished report.
Malde, H. E. and Irwin-Williams, C. 1967. Preliminary Report on Radiocarbon Dates from the
Valsequillo Area, Puebla, Mexico. Cambridge, MA: Peabody Museum of Archaeology and
Ethnology, Harvard University.
Malde, H. E. and Steen-McIntyre, V. 1981. Reply to comments by C. Irwin-Williams,
Archaeological Site, Valsequillo, Mexico. Quaternary Research, 16: 41825.
Meltzer, D. J. 1993. Search for the First Americans. Washington: Smithsonian Books.
Ochoa-Castillo, P., Pe rez-Campa, M., Mart n del Pozzo, A. L. and Arroyo-Cabrales, J. 2003. New
excavations in Valsequillo, Puebla, Me xico. Current Research in the Pleistocene, 20: 613.
Osborn, H. F. 1905. Recent vertebrate paleontology: fossil mammals of Mexico. Science, 21: 9312.
Parenti, F. 2001. Le Gisement Quaternaire de Pedra Furada (Piau, Bresil): Stratigraphie,
Chronologie, Evolution Culturelle. Paris: Editions Re cherche sur les Civilisations.
Pessis, A-M. (ed.) 1996. FUMDHAMentos: Proceedings of the International Meeting on the Peopling
of the Americas, Sao Raimundo Nonato, Piau, Brazil (1993). Revista da Fundac a o Museu do
Homen Americano 1, Sa o Raimundo Nonato: FUNDHAM.
Pichardo, M. 1999. Valsequillo biostratigraphy 11: bison, tools, correlate with Tequixquiac.
Anthropologischer Anzeiger, 57: 1324.
Pike, A. W. G., Hedges, R. E. M. and Van Calsteren, P. 2002. U-Series dating of bone using the
diusion-adsorption model. Geochemica et Cosmochimica Acta, 66: 427386.
Reichelt, H. 1899 (1900). Fossile Bacillariaceen eines Kalktues aus Mexico. Zeitschrift fur
angewandte Mikroskopie, 5: 310.
Renne, P., Feinberg, J. M. Waters, M. R., Arroyo-Cabrales, J., Ochoa-Castillo, P., Perez-Campa,
M. and Knight, K. B. 2005. Age of Mexican ash with alleged footprints. Nature, 438.
Santos, G. M., Bird, M. I., Parenti, F., Field, L. K., Guidon, N. and Hausladen, P. A. 2003. A
revised chronology of the lowest occupation layer of Pedra Furada Rock Shelter, Piau , Brazil: the
Pleistocene peopling of the Americas. Quaternary Science Reviews, 22: 230310.
Steen-McIntyre, V. 1975. Hydration and superhydration of tephra-glass a potential tool for
estimating the age of Holocene and Pleistocene ash beds. In Quaternary Studies (eds R. P. Suggate
and M. M. Cresswell). Wellington: Royal Society of New Zealand, pp. 2718.
Steen-McIntyre, V. 1981. Tephrochronology and its application to problems in NewWorld
archaeology. In Tephra Studies (eds S. Self and R. S. J. Sparks). NATO Advanced Study Institutes,
Series C, 75. Dordrecht: Reidel, pp. 35572.
Steen-McIntyre, V. 1985. Tephrochronology and its application to archaeology. In Archaeological
Geology (eds G. Raoo Jr. and J. Giord). New Haven, CT, and London: Yale University Press,
pp. 265302.
626 Silvia Gonzalez et al.
Steen-McIntyre, V. 1998a. Has man been in the New World for a quarter-million years? The Barnes
Review, JanuaryFebruary: 316.
Steen-McIntyre, V. 1998b. Suppressed evidence for ancient man in Mexico. Nexus, August
September: 4751.
Steen-McIntyre, V. and Malde, H. E. 1970. Status of volcanic-ash chronology at Valsequillo
archaeological sites, Puebla, Mexico. Society for American Archaeology, 35th Annual Meeting,
30 April2 May, Mexico City, abstract, 77.
Steen-McIntyre, V., Fryxell, R. and Malde, H. E. 1981. Geological evidence for age of deposits at
Hueyatlaco archeological site, Valsequillo, Mexico. Quaternary Research, 16: 117.
Szabo, B. J., Malde, H. E. and Irwin-Williams, C. 1969. Dilemma posed by Uranium-Series dates on
archaeologically signicant bones from Valsequillo, Puebla, Mexico. Earth and Planetry Science
Letters, 6: 23744.
VanLandingham, S. L. 2004. Corroboration of Sangamonian age of artefacts from the Valsequillo
region, Puebla, Mexico by means of diatom biostratigraphy. Micropalaeontology, 50: 31342.
Webb, M. and Clark, S. 1999. Anatomy of an anomaly. Disputatio, 6: 318.
Dr Silvia Gonzalez is Reader in Quaternary Geology and Geoarchaeology at Liverpool
John Moores University and has been researching Mexican Late Pleistocene-Early
Holocene geoarchaeology in the Basin of Mexico and Valsequillo Basin for the last eight
years. She graduated from UNAM with degrees in geology and geophysics. She has also
researched Mesolithic-Neolithic footprints and coastal geoarchaeology on the Merseyside
coasts.
Professor David Huddart is Professor of Quaternary Geology at Liverpool John Moores
University and Director of Research and Enterprise. He has researched the sedimentology
of Late Pleistocene-early Holocene Mexican geoarchaeological sites in the Basin of
Mexico and the Valsequillo Basin for the last six years and intertidal footprints and the
eects of sea-level change on coastal geoarchaeology on the Sefton coast (Merseyside).
Professor Matthew Bennett is Professor of Geographical and Environmental Sciences and
Head of Department in the School of Conservation Sciences at Bournemouth University.
He has researched Mexican geoarchaeology in the Valsequillo Basin for the last three
years, in particular the geological sequence and the laser scanning of the footprint traces.
Valsequillo Pleistocene archaeology and dating 627