See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/249550919

Timing of the prehistoric eruption of Xitle Volcano and the abandonment of

Cuicuilco Pyramid, Southern Basin of Mexico

Article in Geological Society London Special Publications · June 2000

DOI: 10.1144/GSL.SP.2000.171.01.17

CITATIONS READS

36 612

4 authors:

Silvia Gonzalez Alejandro Pastrana

Liverpool John Moores University Instituto Nacional de Antropología e Historia
90 PUBLICATIONS 1,847 CITATIONS 31 PUBLICATIONS 260 CITATIONS

SEE PROFILE SEE PROFILE

Claus Siebe Geoff Duller

Universidad Nacional Autónoma de México Aberystwyth University
186 PUBLICATIONS 5,275 CITATIONS 261 PUBLICATIONS 17,453 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Claus Siebe on 25 March 2014.

The user has requested enhancement of the downloaded file.

 Timing of the prehistoric eruption of Xitle Volcano and the
abandonment of Cuicuilco Pyramid, Southern Basin of Mexico

SILVIA GONZALEZ', ALEJANDRO PASTRANA',

CLAUS SIEBE~& GEOFF DULLER~

' School of Biological and Earth Sciences, Liverpool John Moores University, Byrom Street.
Liverpool L3 3AF, UK (e-mail: [email protected])
DICPA, Instituto Nacional de Antropologia e Historia, Seminario 8 Centro Histdrico,
Mexico City, C.P. 06060, Mexico
3~nstitutode Geofisica, Universidad Nacional Autdnoma de Mkxico, Coyoaccin,
Mexico City, C.P. 04510, Mexico
Aberystwyth Luminescence Laboratory, University of Wales, Aberystkvyth,
Dyfed, SY23 3DB, UK

Abstract: The Cuicuilco pyramid was one of the first true urban centres in the Basin of
Mexico. Its construction started a few centuries BC, during the Late Preclassic period. The
pyramid is partially covered by a basaltic lava flow produced by the Xitle monogenetic
volcano. New stratigraphic work around the pyramid and the volcano together with new
radiocarbon dates indicate that the pyramid and nearby settlements were abandoned as a
direct consequence of the volcanic activity of Xitle. The new dates, obtained from material
which clearly is contemporaneous with the volcanic activity, suggest that the eruption took
place around 1670 years BP, some 300 years later than previously thought.

The Cuicuilw circular pyramid was the most by one of the lava flows with a thickness of
important ceremonial centre in the Basin of 4-6 m (Fig. 2b).
Mexico around 2000 years ago (Heizer & Benny- The pyramid was rediscovered in 1922 (Cum-
hoff 1958, 1972). Situated close to a freshwater mings 1933). Excavations started with the aid of
lake (Xochimilco Lake), it represents one of the explosives, in order to break the basaltic layer
largest and oldest Preclassic centres in Meso- and uncover the structure. Further investigations
america. It should be evaluated in terms of the showed that the pyramid consists of at least three
beginnings of urbanization and the origin of superimposed structures, indicating occupa-
the city-state in the Basin of Mexico (Piiia-Chan tion of the site for several hundreds of years.
1967; Parsons 1989). Cuicuilco was suddenly Additional excavations have shown that the
destroyed by the appearance and volcanic activ- Cuicuilco pyramid was in the centre of a large
ity of Xitle volcano, 7 km SW of the pyramid settlement, with several minor pyramids, mounds
(Figs 1 and 2a). and habitation structures, pointing to the exis-
This small volcano is one of more than 200 tence of a well-organized culture and the begin-
scoria and cinder cones that constitute the nings of the city-state in Mesoamerica.
Quaternary Chichinautzin monogenetic volcanic New excavations were started in summer 1996
field, to the south of the Basin of Mexico. Prod- by the Mexican National Institute of Anthro-
ucts of the cones are andesitic to basaltic in com- pology and History (INAH) (PBrez-Campa el al.
position and mostly calcalkaline in character 1995). Several trenches around the main struc-
(Gondlez-Huesca, 1992). ture were dug with the purpose of clarifying the
The eruption of Xitle volcano started with the relationships between the volcanic stratigraphy
emission of a basaltic tephra, which covered and the associated archaeological horizons, so as
the Cuicuilco circular pyramid with a 10-15cm to understand how the population reacted to the
thick layer of ash. Initial ash fall was followed activity of Xitle.
by several basaltic lava flows, which flowed Because of the volcanic activity the pyramid is
downslope in a northerly direction, covering an well preserved. One of the important questions
area of c. 80 km2. The pyramid was surrounded is: was the pyramid in use at the moment of the

From: MCGUIRE,W . G., GRIFFITHS, D. R., HANCOCK, P. L. & STEWART, I. S. (eds) The Archaeology of
Geological Cafasrrophes. Geological Society, London, Special Publications, 171, 205-224. 1-86239-062-2/00/
$15.00 0The Geological Society of London 2000.
S. GONZALEZ ET AL.
 T H E C U I C U I L O PYRAMID

Fig. 2. (a) The circular pyramid of Cuicuilco was surrounded and partially buried by lava flows originated by
Xitle. Today, this archaeological site lies within the urban area of Mexico City. Photograph taken April 1997.
(b) Trench excavated with the help of explosives at Cuicuilco several decades ago by Cummings. The people are
standing at point BI where the lava is in direct contact with archaeological remains. Photograph taken in
April 1996.
 S. GONZALEZ ET A t .

eruption or was it already abandoned long NE of the crater, which include archaeological
before? Here, we present stratigraphic evidence soils and the lava and tephra produced by the
from the new excavations, which indicates that Xitle volcano. Composition, grain-size distribu-
the pyramid was still in use when the eruption tion and magnetic susceptibility of the various
started. For example, the sections excavated in layers were determined and new radiocarbon
the eastern ramp (see Fig. 13, below) show that dates from charcoal within the tephra layer are
it was mainly built of clay, which was intensively presented.
baked by the heat of the lava flow. The ramp The new data reported in this paper strongly
presents a clean, flat surface, without evidence of suggest that the abandonment of Cuicuilco was
charred vegetation. the direct result of the birth and activity of
In addition we present stratigraphic sections Xitle. This volcano is one of the youngest
located at 1.5 km to the SW and 12km to the monogenetic cones in the vicinity of the Basin

Fig. 3. Xitle pillow lavas, exposed during the construction of the Inbuna Tower in May-July 1997. The
construction site is 365m south of the Cuicuilco pyramid. Person is standing in a trench exposing the lacustrine
sediments of a formerly sxisting lake.
 THE CUICUILO PYRAMID

of Mexico. The burial of Cuicuilco by young with flow units ranging in thickness from 0.2 to
lava flows and ash has important implications 12.0m. The lavas display a young morphology
with regard to volcanic hazards that could affect with well-preserved flow structures and little or
Mexico City, with a population of more than no vegetation cover. Vesicles are very common,
20 million inhabitants: volcanic hazards that particularly towards the top of the flow units.
could affect Mexico City are not restricted to Within the flow, structures such as explosion
the large composite volcanoes such as Popoca- tubes, lava channels, pressure crests and tumuli
tbpetl, but also include the activity of relatively are visible (Martin del Pozzo el al. 19976).
small monogenetic volcanoes. Emplacement of the lava was controlled by
topography: the Xitle cinder and scoria cone
grew on the slope of the extinct Ajusco volcano
Xitle volcano and tephra and lava flowed down the slope to the N-NE
until it reached the floor of the Basin of Mexico
Xitle volcano is located on the lower slopes of (Fig. l), where it spread out over the settlements
Ajusco stratovolcano (3950 m) near the southern in the Cuicuilco and Copilco areas. During
edge of Mexico City (Fig. 1). This area is part of excavations in May-July 1997 for the construc-
the Chichinautzin Volcanic Field. Lavas erupted tion of the Inbursa Tower located 365m south
from Xitle produced the Pedregal de San Angel, of the Cuicuilco pyramid, pillow lavas were
a rocky terrain on which the southern suburbs exposed on top of lake sediments, indicating
of Mexiw City are constructed (Caii6n-Tapia that Xitle flows had reached the shore of a
ef al. 1995). Xitle may be described as a small former lake (Fig. 3).
cinder cone with a diameter of c. 500m and a The composition of the lava flows is basaltic
height of 140m above the surrounding ground. and the lavas can be described as olivine basalts
The Xitle lava around the Cuicuilco pyramid (Schmitter 1953; Badilla-Cruz 1977). Major ele-
and the National University of Mexico (UNAM) ment geochemical analysis of a lava sample near
comprises a complex series of pahoehoe flows the Olympic Stadium (S-9), (Gonzilez-Huesca

A Xitle: Lava
Xitle 1 ~ ; p ~ d
T--L--

Fig. 4. Plot of K20v. SiOl (Peccerillo &Taylor 1976), indicating that the composition of the Xitle lava (A) and
tephras (e)are of basaltic composition with a calcalkaline trend. The lava sample S-9 was taken close to the
Olympic Stadium on the University campus. (a) Tephra close to the crater; @) tephra at Cuicuilco pyramid;
(c) tephra at the Centro de Estudios para Extranjeros (CEPE-UNAM).
 S. GONZALEZ ET AL.

Table 1 . Major element analyses of Xitle lava (sample S-9) and tephras (samples A B and C)

Locality SiOz Al2O3 Fe203 MnO MgO CaO NazO K20 TiOz P205 Total

Fig. 5. Map showing the distribution of the Xitle lavas, as well as the location of the three studied sections.
Locality A, 1.5 km from the crater; locality B, Cuicuilco pyramid; locality C, CEPE-UNAM.
 THE CUICUILO PYRAMID

Fig. 6. Histograms showing the grain-size distribution of tephra samples from the three studied localities.

1992) indicates that the Xitle lavas are calcalka- Cuicuilco is one of the oldest and most impor-
line in character (Fig. 4 and Table 1). tant ceremonial centres of the Preclassic period,
Birth of the volcano started with the emplace- situated on the shores of the Xochimilco fresh-
ment of a widespread sandy to silty, grey ash-fall water lake (Fig. 1). It represents the initial phase
deposit, distributed in a NE direction from the of urbanism and the city-state complex in the
vent. This ash blanketed the original land sur- Basin of Mexico. The pyramidal structures of
face as well as the pyramid. Chemical analysis of Cuicuilco, like the main conical pyramid (named
the tephra at different distances of the crater Temple 1 of Cuicuilco A), were built consecu-
(Localities A B and C) indicates a basaltic com- tively with stratified infills, which were covered
position and a calcalkaline trend (Fig. 4 and with blocks or boulders of basaltic or andesitic
Table 1). lavas from lava flows of older adjacent volcanoes.
Grain-size analyses were performed on the Between the archaeological materials recov-
same tephra samples. These analyses document ered from the excavations below the Xitle lava,
a decrease in grain size from sand to silt with the following elements were found: floors, infills,
distance from the source. At locality B a marked ramps, walls, drainage channels and a megalithic
depletion in the coarse fraction can be inter- sculpture with remains of pigments. Other impor-
preted as reworking of the tephra (Fig. 6). tant cultural elements found include burials with
The volcanic activity of Xitle led to the different ritual practices, in groups of both sexes
abandonment of Cuicuilco and other nearby and ages, usually with offerings of pottery and
human settlements. However, the volcanic pro- ornamental objects made of jade, obsidian,
ducts (tephra and lava flow) protected the basalt, flint, onyx and bone. The objects are evi-
archaeological evidence from later destruction dence of contact and exchange with different
and looting, and provide a unique opportu- regions and cultures of Mesoamerica.
nity for obtaining a more complete picture of The earliest occupation in the Cuicuilco area
human life in the Basin of Mexico more than dates back to 2100 BC, when people lived in small
1600 years ago. villages with incipient agriculture. Between 800
and 600 BC the main structure of the pyramid was
constructed, in the form of a truncated cone
Cuicuilco pyramid archaeological sequence (Heizer & Bennyof 1972); it is at this time that
Cuicuilco became a city.
The chronology of cultural evolution in Pre- The time of the abandonment of Cuicuilco
hispanic Mexico has been divided into four main has been a point of discussion, with two main
periods. According to Coe (1994), the archae- hypotheses being proposed: (1) Cuicuilco was
ological time scale can be divided as follows: abandoned before the eruption of Xitle, as a
consequence of the rise of the city of Teotihua-
Period Important sites c6n in the NE part of the Basin of Mexico;
(2) Cuicuilco was abandoned because of the
Preclassic (200 BC-AD 250) Cuicuilco and volcanic activity of Xitle.
Chalcatzingo The first hypothesis (Lbpez-Camacho 1991;
Classic (AD250-750) TeotihuacLn C6rdoba et ai. 1994) is supported by evidence
Epiclassic ( ~ ~ 7 5 0 - 1 1 5 0 ) Tula, Cholula and found directly at the mounds and buildings of
Xochicalco
Postclassic (AD 1 150-1521) Tenochtitlhn-Tlatelolco, Cuicuilco B in the Olympic Villa area. Appar-
Texcoco and Tacuba ently, these mounds had been abandoned and
undergone decay and erosion for some time
 S. GONZALEZ ET AL.

before the eruption of Xitle. Lopez-Camacho several natural outcrops at different distances
and Cordoba et al. interpreted two 'bodies of from the Xitle crater (Fig. 5). At the same time,
debris' exposed during the first excavations and taking advantage of the excavations made
of 1922-1924, at the southern base of the main by INAH, we studied new sections around
pyramid in Cuicuilco, as evidence for the aban- the main conical pyramid (Fig. 7). We were
donment and decay of the structure before particularly interested in the contacts between
Xitle's eruption. archaeological structures, and covering ash fall
We believe, however, that the second hypoth- and lava flows.
esis is the correct one, and support the idea Location A is a geological outcrop located
of the decline of Cuicuilco as a direct result of 1.5 km SW of Xitle's crater (Fig. 8). At the base
the eruption. Stratigraphic information com- of the section are 120cm of a dark grey, sandy,
ing from the new trenches excavated around well-bedded air-fall ash (Xitle tephra). The ash is
Cuicuilco pyramid and from other outcrops bedded in layers that are 2-5cm thick. A well-
strongly supports this version, and is presented defined erosional unconformity indicates tem-
in the next section. porary cessation of the ash fall during the early
stages of the eruption. A sample of the ash was
taken 30cm below the contact with the lava flow
Stratigraphy and archaeology of the for chemical and granulometric analysis. These
Xitle eruption analyses revealed a basaltic composition and a
dominance of sand-sized fraction (Figs 4 and 6).
To better understand the relationships between Charcoal is abundant within the ash layers
the volcanic stratigraphy and the archaeology of and was radiocarbon dated at 1665f65 years Bp
the area around Cuicuilco we studied in detail (see also Table 2). The ash is topped by a thin

Fig. 7. Plan of the circular pyramid of Cuicuilco, after Haury (1925; in Schavelzon 1993). The position of the five
trenches excavated by INAH during the 1996 season is also shown.
 A. Xltle Latitude: 19' 14' 59'
Longlude: 99' 13' 07'
Alfie: 2,890 m.a.s.1.

Upper, vesicle-rich pan of lava

flow. Elongated megavesicles
llOcm up to 10 m in $ are frequent.

1 =an 0
0
I M i * . vesicle-poor pan m.

Vesicular-rich paR of flow.

Elongated vesicles are up
to 4 cm in $.

F->:--_ =---: --=-:I

--
I
-c

Scoriaceous basal flow breccia.

35 m _ - d - C -

Dark grey, sandy airfall ash, well

bedded, normally graded with
scoria fragments up to 1 cm in
$. Unconformity points towards
temporary cessation of ash fall
durlng the early stages of the
eruption.

Fig. 8. Stratigraphy at locality A 1.5 km south of Xitle volcano, showing the Xitle tephra and lava flow. Charcoal samples within the ash-fall sequence were dated at
1665 f65 years BP in this study. W
 S. GONZALEZ ET AL.

Table 2. New radiocarbon dares for Xille's volcanic activity

Locality Locality I4cage Calibrated results (2cr) 6°C

(years BP)
A A9587 1665f 65 250-548 cal AD -23.6
B1 A8985 1675f 40 252-296 a 1AD -23.3
3 18-450 MI AD
482-500 AD
514-53o~a AD
B2 A9586 1995 f60 380 cal BC-350 cal AD -25.0
360-380 AD
C A9585 2295f 115 760-620 cal BC -25.0
600-60 MI BC

scoriaceous lava flow (35 cm) from Xitle. This is (over the main corridor). We believe that a
followed by 220cm of a vesicular lava flow. low platform was in this position. During
Location B is at the Cuicuilco pyramid, the excavations by Cummings in 1922-
situated 7 km SE of Xitle's crater. A total of 1924, the lava flow was broken using dyna-
five trenches were studied by us. The main pyra- mite. This must have partially destroyed
mid and associated structures were first covered the low platform.
by a layer of basaltic ash fall (Xitle tephra) with (b) Trench B1 shows the interior filling of the
a thickness ranging between 20 and 30cm platform, which can also be observed along
(Fig. 4). Grain-size analysis of an ash sample 35% of the circumference of the excavated
taken in trench B3 (Fig. 6) indicates a depletion pyramid. The fill consists of basaltic and
in the coarse fraction, if compared with ash from rhyolitic blocks of a homogeneous size set
localities A and C. We interpret this as a result in a clay-rich matrix. In some places it is
of reworking by wind and rain. possible to observe that the platform was
The trenches BI, B2 and B3 are those that best built in two stages.
display the stratigraphic relationships between (c) The platform contains burials with offer-
the archaeological buildings, the ash fall and the ings (pottery and various stone objects).
lava flow (Figs 9 and 10). Trench B4 was dug For this reason it is impossible that it repre-
around the decorated andesitic column dis- sents a collapsed sector of the pyramid.
covered in 1996 and trench B5 on top of the We believe that the interior of the platform and
pyramid.
the ash layer on top have been misinterpreted
Trench BI (Fig. 9) is particularly important
in the past as material coming from the destruc-
because it shows the constructional fill at the tion and erosion of the pyramid. This interpreta-
base of the section (120cm thick), which has
tion has led to the conclusion that the main
been interpreted in the past as evidence of the
pyramid was abandoned before the eruption.
state of decay of the pyramid before the eruption
of Xitle (Lopez-Camacho 1991; Cordoba et al. Our observations do not support this hypothesis.
Trenches B2 and B3 were excavated in the SE
1994). The fill is overlain by an occupational soil ramp of the pyramid (Fig. 10). They show
horizon, which is 32cm thick, and contains frag- basically the same stratigraphy: constructional
ments of pottery and charcoal. The top 12 cm of fills rich in pottery and charcoal fragments at the
the occupational horizon display a light brown
base of the sections followed by the Xitle tephra.
colour produced by a baking effect from the hot The tephra is 8.5cm thick in trench B2 and
lava flow above. The Xitle tephra covers the
4cm thick in trench B3. The sequence is topped
occupational layer, with a thickness of 22cm. by the Xitle lava flow, which varies in thickness
A charred maize plant was found at the contact, between 3.20 and 3.70 m.
and yielded a 14Cage of 1675 f40 years BP. The In trench B3 a very strong baking effect of the
Xitle lava is 3 m thick in this area. The following
observations were made: fill on the ramp can be observed. The baking
reaches a depth of 30cm and can be recognized
(a) The contact of the lava with what we by its brick-red colour. The contact between the
believe is the last body of the pyramid is at ramp, tephra and lava flow is very clear and
a very consistent mean height of 3.50m planar, following the shape of the ramp. The
 83. Cuicuilco
82. Cuicuilco Latiude: 19' 18' 03'
Longitude: 99'10'51'
Anaude: 2,290 m.a.s.1.

UPper. vesicle-rich part of lava flow

Upper, vesicle-MI pan of lava fkw.

M i . veddepoor pan of lava flow.

---

Scoriaoews basal flow unit.

Middle, veskiepoor part d lava Row.
Darkgreyairfall ash, partly reworked.

Construcllonal fill. Brown, clayey-sandy

materlal with occasional boulders. Scoriaceous besal How breccia.
Rich In charcoal fragmenls, pottery, and
other arlifacs.
'C-14=1,995f60~.B.P.
Oreydre, sandy airfallash,panly mworked.

Const~dmnalfill. Brown. clayey-sandy material with

occasional boulders, rich In charcoal fragments, and
archaeological remains. Upper pan is brick-red in
wlour, intensely baked.
Constructional fill. Brown, clayey-sandy
material wiVl ocrasional boulders.
R i h In archaeological material.

Consbudknal WII, mostty boulders.

Fig. 10. Stratigraphic sections at localities B2 and B3 at Cuicuilco. Charcoal dated at 1995 f60 years BP was recovered from the constructional fill below the air-fall ash.
 THE CUICUILO PYRAMID

lack of charred vegetation, reworked material, study we obtained the following I4C dates
erosional features, and animal holes indicates (Table 2 and Fig. 12): 1665f 65 years BP
that the ramp was still in use at the time of (location A burnt branches of wood within the
Xitle's eruption. tephra sequence); 1675f 40 years BP (location
Location C is situated 12 km NE of the Xitle B1, on maize plant charred by the Xitle lava);
crater, in a garden within the grounds of the 1995 f 60 years BP (location B2, charcoal frag-
Centro de Estudios para Extranjeros (CEPE) on ment in constructional fill in contact with the
UNAM campus (Fig. 11). The section shows at tephra layer; 2295f 115 years BP (location C,
the base at least lOOcm of a dark brown soil. charcoal within dark soil below lava flow).
The soil is sandy-clayey, with abundant pottery In general, the I4C dates seem to cluster
and charcoal fragments. The upper part of the around 2000 years BP However, the dates directly
soil (10-20cm) is brick-red in colour and associated with the Cuicuilco archaeological
indurated as a result of the baking effect of the zone tend to be younger, together with the date
lava flow on top. The Xitle tephra is of silt grade obtained in this study near the crater (loca-
and is only 2.5 cm thick, whereas the lava flow is tion A). In previous studies young ages were
7.80 m thick. obtained on a root in a midden burned by
The outcrop is very extensive and measures at the lava at Cuicuilco A (pyramid) (1536i65
least 40m in length. Archaeological remains in years BP; Fergusson & Libby 1963), and for a
the soils below the Xitle tephra and lava flow midden in Cuicuilco B (1430f 200 years BP;
include floors, walls, pottery and bone frag- Crane & Griffin 1958). The young ages obtained
ments. This shows proof of the extension of the during this study are both on material that was
inhabited area at the time of the Xitle eruption, defmitely burnt by the volcanic activity, which
which was clearly much larger than the area would suggest that they represent the true age of
officially delimited and protected as an archae- the eruption. However, an explanation for the
ological site. clustering of ages around 2000 years BP still
eludes us.
The younger I4C dates support the archae-
Age of the eruption ologicai seqience for ~uicujicoproposed by
Heizer & Bennyof (1958, 1972), who suggested
The Xitle lava flow is certainly the most fre- that the eruption of Xitle occurred around
quently dated volcanic feature of the Chichi- AD 400.
nautzin volcanic field. None the less, there are As an independent way to assess the age of
still uncertainties with regard to the age of the the eruption, we tried to date the sediments
eruption. First radiocarbon determinations were baked by the lava at localities B3 (SE ramp at
carried out during the early development of the Cuicuilco) and C (CEPE-UNAM) using opti-
radiocarbon method, by Arnold & Libby (195 1) cally stimulated luminescence (OSL) (Duller
and Libby (1955). They reported an age of 1996). At each site a block of baked sediment
2422f 25 years BP on charcoal associated with 20 cm x 20 cm was carved from the section and
the eruption of Xitle volcano. wrapped in foil to avoid additional exposure to
More than 28 radiocarbon dates have been sunlight. In the laboratory the external surface
reported in the past decades from charcoal asso- of the blocks of sediment was removed, because
ciated with Xitle lavas (Urrutia-Fucugauchi it would have been exposed to light during
1996a, b), with dates ranging from 1536f65 sampling. The material from within the block
years BP to almost 4000 years BP being reported. was then disaggregated in sodium oxalate and
The main problem seems to reside in the poor hydrogen peroxide, and settled in distilled water
control on the depth within the soil with respect to obtain grains of 4-1 1 pm.
to the lava flow at which samples have been col- Initial luminescense measurements showed no
lected. For example, at the Cuicuilco pyramid, significant OSL signal above the level of the
evidence for human settlement spans several background on the Risoe automated TL-OSL
centuries in the anthropogenic fills of the archae- reader that was used. The fundamental problem
ological buildings. These often contain frag- was the absence of quartz or feldspar with a suit-
ments of charcoal that are considerably older able grain size (>4pm) in the samples. Although
than the volcanic activity of Xitle. some material was separated, we suspect that it
The most recently reported I4C dates include may have been clay pellets rather than true silt-
the following: 2030 f 60 and 2090 f70 years BP sized material. This yielded signal levels that
(Martin del Pozzo et al. 1997a); 1945 f55 and were at about the noise level of the instrument.
2025 f55 years BP (Cewantes-Laing & Moli- One possible problem was that the material was
nero 1995; Delgado et al. 1998). During our so young that it was not possible to obtain a large
 C. CEPE-UNAM
Latitude: 19'20'07'
Longitude. 99-11'15'
Altitude: 2,280 m.a.s.1.

PeRly vesicular, partly dense upper part of lava flow.

Scorieceous basal breccia.

Dark grey, silty alrfall ash.

Dark brown soil, sandy-clayey with abundant

pottery and charcoal fragments. Upper few cm
are bM-red in colour and indurated.

Fig. 11. Locality C near CEPE on UNAM campus. Charcoal within dark soil below the lava flow was dated at 2295f 115 years BP in this study.
 B g s
1
I
1
B I
1 t
1
w

I
I
t;
5: -
?i 8
I
0
I

Middle Late Terminal Post-Classic

Classic
Pre-Classic Pre-Classic Pre-Classic

- - -
-
Small villages and few Cuicuilco Teotihuacan w Toltec -Aztec r -Colonial- -Modern
larger settlements based Copilco
on cultivation of maize Tlatilco Cholula c

9 CI
w =!X
nl
::
s3
h
0

(O
35
E 5-
0 ra D
w 2
a
:
Y 1

0
83 7
tB
-Il
m
1 a.=
cr
5-
M
g
n CI'O
w w
A
G) a
w C
8 D
220 S. GONZALEZ ET AL.

signal from it. To test this, separate aliquots of much higher (413 x SI), indicating a differ-
the samples were irradiated in the laboratory ent magnetic mineralogy. It is possible that the
using a 90SrpY beta source. The doses were presence of additional magnetic minerals resul-
chosen such that they were equivalent to c. 5000- ted from the baking of the tephra produced by
10000 years of natural radiation. However, even the Xitle lava at this site.
after this irradiation the samples still yielded Curie temperature experiments were per-
a negligible OSL signal. These results suggest that formed on the same tephra samples, which
the 4-1 1 pm material that was separated from the were heated to 700°C in a strong magnetic field
samples did not have a suitable mineralogy and then allowed to cool. The results are
for the application of luminescence dating. presented in a graph of temperature v. magne-
tisation (Fig. 14). The curves from localities A
and C are both dominated by paramagnetic
Magnetic properties of anthropogenic soils minerals and show an increase in magnetization
and tephras after heating. Locality B3 shows a thermomag-
netic curve that is typical of magnetite, with a
During this project, magnetic properties of Curie temperature of c. 590°C.
different materials were determined, to discrimi- The values found for the magnetic suscept-
nate and correlate the Xitle tephra and the ibility and Curie temperatures for localities A
anthropogenic soils around the Cuicuilco pyra- and C are very similar, whereas values near
mid. The magnetic properties determined in- the pyramid (locality B3) are very different. The
cluded magnetic susceptibility of tephras and tephra collected at the pyramid is visibly red-
soils and Curie temperatures of the tephras. dened, indicating that baking occurred during
At some archaeological sites with long occu- the emplacement of the lava. The baking effect
pational sequences it is difficult to distinguish of the lava is locally very variable. Around
between an occupational surface (or floor) and locality C there are zones where the underlying
a fill, and magnetic susceptibility measurements sediments are considerably baked and others
have been useful in differentiating them (Tite nearby seem unaffected. It would seem that this
& Mullins 1971; Tite 1972). For this reason, baking has produced magnetite in the tephra,
we took a series of samples in each soil layer possibly through oxidation. Another possible
in trenches B1, B2, B3, B4 and B5 around the explanation is that during the reworking of
main pyramid. the tephra at the Cuicuilco site magnetic min-
The samples were air dried in the laboratory erals have been concentrated. These changes
and were used to fill cylindrical plastic pots are noticeable only through the measurement
2.5 cm in diameter. The samples were measured of the magnetic properties, which shows the
using a Bartington Susceptibility Meter (MS-2). usefulness of this approach in understanding
An example of the results obtained in the SE the processes of site formation in archaeological
ramp (trench B2) is shown in Fig. 13 together contexts.
with the stratigraphy.
The results show that the magnetic suscept-
ibility measurements represent a very useful tool Conclusions
in distinguishing occupational floors from fills.
Generally, the floors have higher values with New stratigraphic and archaeological evidence
distinct peaks of magnetic susceptibility, sug- found during the excavations of 1996, when
gesting high concentrations of magnetite, prob- five trenches were excavated around the Cui-
ably produced by fire. The fills, on the other cuilco pyramid, reinforces the view that the
hand, have lower susceptibilities. The maximum abandonment of this important archaeological
value was found at the contact between the zone resulted primarily from the eruption of the
original land surface (a deposit of fluvial origin Xitle volcano, 7 km SW of the pyramid.
containing rounded pebbles of andesite) and the The clean contact, with no evidence of charred
constructional fill of the pyramid. This suggests vegetation or animal holes, between the archae-
that human activity probably occurred in this ological buildings and the overlying tephra and
area for some time before the commencement of lava is strong evidence in support of this hypo-
the construction of the pyramid. thesis. Furthermore, the material found in front
Magnetic susceptibility measurements were of the decorated andesite column at the southern
also carried out on tephra samples at localities base of the pyramid, which had been previously
A, B3 and C. The results show similar values interpreted as rubble derived from the erosion of
for localities A (260 x 10-6SI) and C (233 x the pyramid and advocated as one of the most
10-6SI), whereas at locality B3 the value was important pieces of evidence in support of the
 THE CUICUILO PYRAMID

Legend

Basaltic lava flow

Air fall ash
Sandy clay
Tepetate
Xitle lava flow Lava blocks and sand
Conglomerate
Gravel
Xitle ash fall
reworked
Ramp floor - Burnt clay (brick effect)
Gravelly sand
Sand
Clay
Layer full of charcoal

Ramp floor

Actual corridor floor

Floor

Floor

Oxidation level

Floor

Conglomerate (original a Low frequency

land surface) High frequency

Fig. 13. Stratigraphy and magnetic susceptibility measurements in ramp SE of Cuicuilco (locality B2). The depth
is with respect to the top of the pyramid.

abandonment and degradation of the pyramid cluster around an age of c. 2000 years BP. How-
long before the Xitle eruption (Lopez-Camacho ever, we have obtained two new dates with excel-
1991; Cordoba et at. 1994), is in fact part of a lent stratigraphic control that suggest an age for
low platform containing undisturbed burials and the eruption of about 1670 years BP. The dis-
offerings. The shape of this platform has been crepancy of about 300 years cannot be explained
cast in the lava flow on top, with a steep angle of at present with the available information, but
c. 45" away from the pyramid. we believe that the most reliable dates for the
Despite the large number of determined radio- eruption are the younger ones.
carbon dates (more than 35) the age of the Xitle Efforts to obtain independent dates through
eruption is still a matter of debate. The majority luminescence dating of the baked soils were
of the samples with a good stratigraphic control unsuccessful because the mineralogy of the
 Near Crater Cuicuilco Pyramid Copilco (12 km from crater)

..
... Curie Temp= 380°C - Curie T e m p 390°C

.-
d
a
Mag. Sur= 260 Mag. Sua= 233
-
4

C
rn
DI x0 z0
z
6

110 210 1 410 SIB 101 101 0 110 101 300 (10 SDI 100 710 IS# 1 I00 210 SO0 (00 ID# 110 701 (11
0
Tenparalure C Tanparaiurs C Tamperalure C

Fig. 14. Curie temperature experiments on tephra samples at localities A, B and C.

 THE CUICUILO PYRAMID

material proved to be inappropriate for lumi- Volcano, Mexiw. Journal of Volcanology and
nescence dating. Geothermal Research, 65, 249-263.
Knowledge of the exact age of the Xitle erup- CERVANTES-LAING, P. & MOLINERO, R. J. 1995.
tion is important to better understand the fall of Eventos volctinicos a1 sur de la ciudad de Mexico.
BSc thesis, Facultad de Ingenieria, UNAM, Mex-
Cuicuilco, the most important ceremonial centre
ico City.
in the Basin of Mexico a t the end of the Late COE,M. 1994. Mexico, from the Olmecs to the Aztecs,
PreclassioEarly Classic period, and the rise 4th edn. Thames and Hudson, London.
of Teotihuadn in the NE of the Basin. This CORDOBA, C., MARTIN, A. L. &LOPEZ,J. 1994. Palaeo-
is especially so because around that time Popo- landforms and volcanic impact on the environ-
catkpetl stratovolcano, located 40km SE of ment of Prehistoric Cuicuilco, Southern Basin of
Cuicuilco, experienced a major eruption (Siebe Mexico. Journal of Archaeological Science, 21,
el al. 1996). Deposits of this eruption have been 585-596.
dated a t c . 2 0 0 ~ .Maybe a combination of CRANE, H. R. & GRIFFIN, J. B. 1958. University of
Michigan radiocarbon dates 111. Science, 128,
these two important volcanic events in the south
1117-1 123.
of the Basin of Mexico could explain why there CUMMINGS, B. 1933. Cuicuilco and the Archaic culture
was a major population shift to the north. of Mexico. University of Arizona Social Science
The favourable geographical location of the Bulletin, 4(8).
Basin of Mexico, with temperate climate, fertile DELGAW,H., MOLINERO, R., CERVANTES, P. et al.
volcanic soils and abundance of water, has been 1998. Geology of the Xitle Volcano in Southern
responsible for early human settlement since Mexico City - a 2000 year old monogenetic
Pleistocene-Early Holocene time. Volcanic activ- volcano in an urban area. Revista Mexicana de
ity in the central part of Mexico must have played Crencias Geol6gicas, 15(2), 115-1 3 1.
DULLER,G. A. T. 1996. Recent developments in
a n important role in determining the timing of luminescence dating of Quaternary sediments.
large migrations in Prehispanic times as well as Progress in Physical Geography, 20,133-1 51.
the rise and fall of cities such as Cuicuilco, Teo- FERGUSON, G. J. & LIBBY, W. F. 1963. 1963 UCLA
tihuacin and Cholula (Siebe et al. 1996, 1997). radiocarbon dates 11. Radiocarbon, 5, 1-22.
Xitle tephra and lava flows covered the GON~LEZ-HUESCA, I. S. 1992. La variacidn secular en
settlement of Cuicuilco, the first known urban Mixico central durante 10s tiltimos 30000 anbs
centre in the Mexican Altiplano, sealing in this por medio del estudio magnitico de lavas. PhD,
way a complete Prehispanic landscape and cul- National University of MCxico, Geophysics Insti-
tural remains. Archaeological investigations are tute, Mexico City.
HEIZER,R. & BENNYHOFF, J. 1958. Archaeological
difficult today, not only because they involve the investigations of Cuicuilco, Valley of Mexico
breaking of the lava layer but also because 1956. Science, 127, 232-237.
the area is now heavily populated. Only 5% - & -1972. Archaeological excavations at
of the archaeological zone has been investigated Cuicuilco, Mexiw 1957. National Geographic
so far, and it is to be hoped that much more Reports, 1955-1960,93-104.
information will be recovered in the future. LIBBY,W. F. 1955. Radiocarbon Dating, 2nd edn.
Chicago University Press, Chicago, IL.
We thank the team of archaeologists from INAH MPEZ-CAMACHO, J. 1991. Estratigrafia de la pirimide
involved in the excavation of Cuicuilco during the de Cuicuilco en retrospectiva. Cuicuilco,27,35-46.
1996 season, particularly M. PCrez-Campa, H. Gomez- MARTIN DEL POUO, A. L., CORDOBA, C. & MPEZ,J.
Rueda and S. Peiia. Work by C. Siebe was supported 1997a. Volcanic impact on the southern Basin of
by UNAM-DGAPA (IN107196) and CONACyT Mexico during the Holocene. Quaternary Inter-
(0264P-T9506). Radiocarbon dates were determined national, 43/44, 181-1 90.
by A. Long and C. Eastoe at the University of Arizona -, ESPINASA, R., LUGO,J. et al. 19976. Volcanic
in Tucson. Impact in Central Mexico. Excursion Guide.
IAVCEI General Assembly, Puerto Vallarta,
Mexico, 15-19 January.
PARSONS, J. R. 1989.Arqueologia regionalen la cuenca
References de MCxico: una estrategia para la investigacibn
futura. Anales de Antropologia, 26, 157-257.
ARNOLD, J. T. & LIBBY,W. F. 1951. Radiocarbon PECCERILLO, A. S. & TAYLOR,S. R. 1976. Geo-
dates. Science, 113, 1 1 1-1 20. chemistry of Eocene calcalkaline volcanic rocks
BADILLA-CRUZ, R. R. 1977. Estudio petrologiw de la from the Kastamonu area, northern Turkey.
lava de la parte noreste del Pedregal de San Contributions to Mineralogy and Petrology, 58,
Angel. ~ o l e t &de la ~ociedad~ e o l d ~ i Mexicana,
ca 63-81.
38,40-57. PBRU-CAMPA, M., PASTRANA,A. &GOMEZ-RUEDA, H.
CANON-TAPIA, E., WALKER, G. P. L. & HERRERO- 1995. Proyecto Arqueoldgico Cuicuilco. Arch-
BERVERA, E. 1995. Magnetic fabric and flow ivo del Consejo de Arqueologia, INAH, Mexico
direction in basaltic pahoehoe lava of Xitle City.
 S. GONZALEZ ET AL.

PINA-CHAN,R. 1967. Oficial Guide of Copilco- TI=,M. S. 1972. The influence of geology on the
Cuicuilco. Instituto Nacional de Antropologia e magnetic susceptibility of soils on archaeological
Historia, Mexico City. sites. Archaeometry, 14, 229-236.
SCHAVELZON, D. 1993. La Pircimide de Cuicuilco. - & MULLINS,C. 1971. Enhancement of the
Fondo de Cultura Emnomica, Mexico City. magnetic susceptibility of soils on archaeological
SIEBE,C., ABRAMS,M., MACIAS,J. L. & OBENHOLZ- sites. Archaeometry, 13(2), 209-219.
NER, J. 1996. Repeated volcanic disasters in Pre- URRUTIA-FUCUGAUCHI, J. 1996~. Palaeomagnetic
hispanic time at Popocatepetl, central Mexico: study of the Xitle-Pedregal de San Angel lava
past key to the future? Geology, 24, 3 9 M 2 . flow, southern Basin of Mexico. Physics of the
-, MACIAS,J. L., ABRAMS,M., RODR~GUEZ, S. Earth and P1anetar.y Interiors, 97, 177-196.
& CASTRO, R. 1997. Catastrophic Prehistoric -19966. Comentarios sobre la edad del campo
Eruptions at Popocatiped and Quaternary Explo- volclnico Pedregal de San Angel, Cuenca de
sive Volcanism in the Serb-Oriental Basin, Mexico - fechamientos por radiocarbono. GEOS,
East-Central Mexico. Excursion Guide. IAVCEI Union Geofisica Mexicans, June 96-98.
General Assembly, Puerto Vallarta, Mexico, Pre-
meeting Excursion 4, 12-18 January.
S C H M I ~ EE.
R ,1953. Investigation petrologica en las
lavas del Pedregal de San Angel. Memorias del
Congreso Cientifico Mexicano, 3, 2 18-237.

View publication stats