Академический Документы
Профессиональный Документы
Культура Документы
Calixtlahuaca, a MiddleLate Postclassic site in the Toluca Valley of central Mexico, was occupied ca. A.D.
11001530. Our excavations reveal some of the processes involved in the creation, functions, and decay of
a large hilltop urban center. At its height, the majority of the sites surface (264 ha) was covered with
residential-agricultural terraces supported by a complex water management system. House construction
techniques included the use of adobe brick, wattle-and-daub, and stone pavements. Our fieldwork
contributes to a growing body of research on hilltop political capitals in Mesoamerica. Using a refined
chronology, we illuminate the processes by which people constructed the residential zones of this ancient
hilltop city.
Keywords: agricultural terraces, Aztec, hilltop centers, Postclassic, urbanism, Toluca Valley
Previous fieldwork
Figure 1 Location of Calixtlahuaca in central Mexico. Jose Garca Payon excavated at Calixtlahuaca be-
tween 1930 and 1938 and, as was common at that time,
The site extends onto the plain and the slopes of a
he concentrated his efforts on the sites monumental
small adjacent hill, Cerro San Marcos. Most of the
architecture. Structure 3a large four-stage circular
site is above the limit of 2700 masl that Sanders
pyramid dedicated to the wind god Ehecatlis the
(1976) set for the Prehispanic cultivation of maize in
best known building. Garca Payon excavated other
the Basin of Mexico. However, in view of the higher
religious structures at the site, including Structure 4a
precipitation in the Toluca Valley (ca. 800 mm per
rectangular temple that yielded offerings of vessels
year) and the advantages of cold air drainage that the
with the likeness of the deity Tlalocand a cross-
slopes offer with respect to the plain, modern farmers
shaped building decorated with tenoned stone cones
are able to grow maize, maguey, and other crops.
and skulls. He also excavated a large architectural
Ethnohistorical background complex on the plain at the base of the hill known
Few historical data exist for social conditions be- as Structure 17. Although Garca Payon called this
fore the conquest of the Toluca Valley by the Aztec structure a calmecac (school), it is almost certainly a
Empire around 1476 under the Mexica king Axaya- royal palace (Smith 2008: 115119). His excavation of
catl. Home to speakers of the Nahuatl, Matlatzinca, several smaller structures in a patio group (Group C in
Otomi, and Mazahua languages (Garca Castro 2000), our nomenclature) revealed burials with rich offerings
the valley was a complex ethnic and political mosaic. of ceramic vessels, bronze objects, greenstone jewelry,
The most powerful capital prior to Axayacatls con- obsidian, rock crystal, and other items, giving rise to
quest was known as Matlatzinco, and several lines the name of Panteon (cemetery), by which the complex
of independent documentary evidence confirm that the is known today. Similar burials with offerings were
Figure 2 View of Cerro Tenismo from the northwest. The entire hill was covered with terraces and houses. Horizontal lines of
vegetation mark modern terrace risers.
Table 1 Excavation units; lot is another level of physical unit of excavation and provenience control.
Phases{
303 54 18.0 Testing near royal palace Fan and gully with artifacts x x
304 17 11.0 Testing near royal palace Fan and gully with artifacts x
305 42 8.0 Testing near royal palace Alluvial fan with artifacts x x
306 13 10.0 Domestic deposits Nothing
307 73 77.0 Domestic deposits House and midden x x x
308 30 14.0 Terraces and drains House fragment, terrace, drain x x
309 73 84.0 Domestic deposits House with exterior pavements x
310 29 14.0 Domestic deposits Midden x
311 75 51.5 Terraces House, midden, terrace x x x
312 22 12.0 Terraces Colluvium with artifacts ? ? ?
313 34 9.0 Drain Gully filled with Postclassic refuse x x
314 32 52.0 Domestic deposits Modern terrace x
315 169 101.0 Domestic deposits Burned house and pavement x
316 133 91.8 Domestic deposits House with exterior pavement x x x
317 224 127.0 Domestic deposits House, pavements, burial x
318 9 47.0 Infilled gully Gully with artifacts
319 54 12.8 Terraces Terraces x
320 102 48.0 Terraces and house House floor, midden x x
321 81 42.2 Terraces Terraces and burials x
322 12 4.0 Test pit Ancient terrace x
323 120 36.5 Domestic deposits Midden, burned house fragment x x
324 103 28.0 Terraces and house Terraces, pavement, pit x x x
325 7 3.5 Water control feature Colluvium, wall debris x
326 26 11.3 Terraces Terraces, gully with artifacts x
327 13 7.0 Terraces Modern terraces x
328 7 2.0 Terraces Modern terraces ? ? ?
329 18 12.0 Terraces Terraces, buried ditch x
* Square meters.
{ D (Dongu); N (Ninupi); Y (Yata).
Figure 4 Twenty radiocarbon dates from Calixtlahuaca. A) Plot of uncalibrated radiocarbon ages; B) Calibrated ages (from
Bronk Ramsey 2008).
present state, however, the drains hardly serve either discover houses or certain artifact assemblages. In
purpose. In a few places along the courses of the the end, terrace fills were excavated in practically all
master drains we observed partially destroyed and units (TABLE 1), with the exception of those situated at
buried features that may have acted as plunge pools the very foot of the hill. Units at the lower elevations
that took time to fill before they overflowed, thereby yielded information on how sediment delivery from
reducing the velocity of flow. Cerro Tenismo responded to changing land use. We
Simpler field morphologies dominate outside the were less fortunate in the case of the master drains,
central area of the site. There are treads that slope at as landowners prevented access. Segments of the sys-
a gradient just slightly less than that of the natural tem that channeled runoff from the hill were none-
slope and are separated only by ditches and berms theless excavated in several units. Our dataset is the
planted in maguey, an arrangement known in much largest sample of stratigraphically explored terraces in
of central Mexico as metepantle (West 1971; Wilken Mexico.
1987: 105113). Other fields are lined on all four sides Three excavations in the center of the siteUnits
by dry-laid stone walls (tecorrales). These walls do 308, 311, and 321produced a similar stratigraphic
little to modify slope gradient and their main purpose sequence, in which a natural soil profile was buried by
seems to be to keep out livestock. Finally, there are two generations of wedge-shaped terrace fills (FIGS. 6,
also sectors of the hill that have a gradient equal to 7). The soil profile in question had developed in sandy
the natural one, but where the former presence of yellowish sediments of pyroclastic origin. Though
terracing is indicated by half-buried stone alignments, unconsolidated, some are indurated enough to merit
subtle breaks in slope gradient, or changes in the the folk designation of tepetate, a subsoil horizon too
vigor of grass growth. Elsewhere, similar patterns hard to be plowed (Etchevers et al. 2003). Their upper
point to the purposeful elimination of a former riser parts have been modified by the accumulation of
and the merging of two narrow treads into a single, organic matter and illuviation of clay to such an
wider, more inclined tread. extent that they have a texture of clay or silty clay
The obvious problem that this agricultural land- and a bluish black color. They also have a strongly
scape poses for the study of Aztec urbanism is developed structure of prisms that swell and shrink
distinguishing the layout of the Postclassic city, in with the alternation of wet and dry seasons. The
which the function of terraces was presumably both colors, mottling patterns, and occasional slickensides
residential and agricultural (Evans 1985; Smith 2012: displayed by these horizons point to seasonal
182185), and understanding the modifications that saturation in water. We are certain that this soil
have accrued over over the half millennium since the profile predates the human occupation of the site
citys abandonment. Terraces are inherently unstable because the attributes described take millennia to
landforms, and in those parts of the highlands where develop, and because it contains no artifacts, except
their maintenance was discontinued in the Colonial for those that fall into the cracks that separate the
period, they have in most cases completely disin- prisms. The soil in question is very difficult to till even
tegrated (Borejsza 2006; Cordova and Parsons 1997; with the use of a plow because of the high clay
Fisher 2005). Therefore, the relatively good state of content, and farmers say it is infertile.
preservation at Calixtlahuaca is in itself an indication The overlying terrace fills document the progressive
that they have been managed and modified to a modification of the natural soil. The fills become more
significant degree since the Postclassic. sandy, more friable, and their color passes from bluish
black, through gray, to yellowish brown. The structure
Terrace excavations becomes much less pronounced and takes on angular,
As surface observation could not completely disen- then subangular blocky forms. These differences are
tangle the contributions of different generations of most pronounced at the front of the terrace treads and
farmers, we implemented an excavation program that least at the back where the fills pinch out and the
would uncover terraces, associated water management clayey soil may appear at the modern ground surface.
features, and the products of terrace degradation. We There is, in most places, a clear boundary between the
targeted places where observations of the modern two generations of terrace fills, pointing to a break in
ground surface and cuts revealed by recent gullying or terrace use or a significant change in the techniques of
construction activity led us to expect the preservation their management. The older terrace fills contain a
of long and complex stratigraphic sequences. We higher concentration of charcoal, sherds, and other
usually laid out trenches perpendicular to the features artifacts, suggesting more regular additions of refuse.
we targeted. Thus, trenches were slope-parallel on In Unit 321 they also contain vestiges of bedding that
terrace treads and slope-perpendicular in the case of may indicate purposeful flooding of the tread in order
drains. Other relevant contexts were explored as to raise its surface and replenish nutrients. The
opportunities arose in units originally opened to cumulative result of terracing in all three units (308,
Figure 6 100N section of excavation Unit 321. Dates are radiocarbon ages calibrated at two sigma.
311, and 321) is a composite soil profile better suited 326, and the terrace fills rest directly on top of the
for agriculture: the friable and sandy surface horizons natural substrate. It is possible that the clay-rich
are easy to till and allow the percolation of rainwater, horizons never formed here, but it seems more likely
which is then stored in the underlying clay-rich that they were removed by natural or anthropogenic
horizons where it can be tapped by the roots of the erosion prior to the Postclassic occupation of the site
crop. or by the cut-and-fill operations involved in con-
The geometry of the two generations of fills is also structing terraces. The stratigraphic sequences also
interesting. In each, the slope was cut into and the record two stages of terrace construction separated
spoil was redeposited as terrace fill; but the introduc- by an erosional unconformity. The older terrace fills
tion of extra fill from farther afield cannot be ruled are present in the form of isolated remnants buried by
out. In Units 311 and 321, the length of the colluvial deposits generated by terrace collapse and
stratigraphic sections was sufficient to reveal that, dissected by gullies. The younger generation of
between one generation and the next, some risers terraces was created after the gullies were backfilled
were eliminated and the treads that they separated and the colluvium was subjected to a second round of
were merged (FIG. 7). We thus have stratigraphic cut-and-fill terrace construction. In Units 320 and
confirmation of the process of merging treads that we 324, Postclassic house foundations and floors were
had hypothesized from surface observations, and we found, with associated ceramics dating to the Dongu
have some prospects for dating the process. In Units phase. In both cases, they were situated at the back of
308 and 321, differences in stratigraphy between the the tread (the portion that is cut), rather than at the
center and the sides of the treads reveal that in the front (the portion that is filled). This is the preferred
past they occupied, respectively, convex and concave position for houses in terraced landscapes because the
sectors of the hillside. The latter required the addition weight of the construction is borne by the compact
of much larger volumes of fill in order to raise them natural substrate.
to the level of the rest of the tread and to create a Units 329 and 319 were situated at the periphery of
single bench terrace. the site, the former near the summit of Cerro
Higher up the hill, on a steeper slope, the clay- Tenismo and the latter on its western footslope
enriched horizons are absent in Units 320, 324, and (FIG. 3). They have in common a loamy, pale brown
Toluca Valley began to rebound (Ouweneel 1991), or houses that were completely excavated, and three
in the middle of the 19th century when the others were partially excavated. The majority of the
inhabitants of San Francisco Calixtlahuaca were excavated houses contained components assigned to
dispossessed of land on the valley floor (Garca more than one chronological period. Excavation
Payon 1979: 216). The intervening stage of terrace units were selected to provide a sample for different
abandonment and decay probably began between portions of the site, from the valley floor to the upper
1520 and 1600 when the effects of epidemics and slopes of the hill. Specific excavation locations
congregation efforts in the Toluca Valley took hold were selected based on a combination of landowner
(Garca Castro 1999). permissions, geomorphological probability of pre-
Our preliminary summary of local land use history served remains, and the presence of visible architec-
is as follows. Middle Postclassic settlers encountered ture at the ground surface.
a hill of irregular topography with radially alternat-
ing concave and convex sectors. They placed walls Dongu phase
and other obstacles across the concave (i.e., water- Thirteen of the 27 excavation units encountered
gathering) sectors, allowing the accumulation of deposits dating to the Middle Postclassic Dongu phase
sediment in cross-channel terraces. In the convex (TABLE 1). These units are scattered over the entirety of
sectors, they created bench terraces by cut-and-fill the portion of the site sampled by excavation, except
techniques. Over time most concave sectors were for the eastern settlement on Cerro San Marcos
raised to the same elevation as the convex sectors. As (FIG. 3), suggesting a rapid colonization of Cerro
a result, rectangular bench terrace plots covered most Tenismo. Portions of houses were encountered in
of the slope, while surface runoff was confined to the Dongu phase deposits in five units (307, 315, 320, 323,
narrow drains between them. Terraces in the center of and 324).
the site supported public structures and private Unit 307 contained stratigraphic deposits from all
residences. The residences shared terrace treads with three ceramic phases. This unit was excavated in a
gardens in which domestic refuse was discarded. plowed field located between the palace and the
Terraces on the periphery of the site were devoted circular temple. A trench placed on a low topo-
primarily to staple crops. Colonial depopulation graphic rise encountered the remains of a partially
and the relocation of settlement away from Cerro destroyed house, a terrace wall, and several midden
Tenismo resulted in the destruction of this mixed deposits (FIG. 8). The house appears to be a narrow
urban-rural landscape. Breached walls of neglected structure of which only the central portion has
terraces and former drains concentrated runoff, survived. The southern (upslope) wall extends several
initiating the development of gullies, which removed meters to the east beyond the north-south wall, which
terrace fills and construction debris, channeling them was either an interior dividing wall or the eastern wall
towards the foot of the hill. The hill was probably of the house. Several large square pavement stones
turned over to grazing, which may have contributed along the north wall were likely part of an exterior
to land degradation and prevented the regrowth of pavement, most of which have eroded. The house
natural vegetation. When farmers began to reclaim probably had an earth floor. The remains of a terrace
the slopes in the 18th or 19th century, gullies were wall were located 5 m north of the house. A pit
filled in, walls rebuilt, and new terraces superimposed excavated into tepetate just south of the structure
on Prehispanic ones. Since animal and later tractor contained a very dense, stratified midden.
drawn plows required wider treads, older treads Occupation of this terrace began in the Dongu
were often merged, resulting in a pattern of slope- phase with the construction of the house, which was
perpendicular strips (FIG. 7). In each strip in situ used continuously for the remainder of the Postclassic
remains of the Postclassic settlement were either sequence. The lowest fill in the midden dates to the
obliterated or preserved by burial. The periphery of Ninupi phase; there are two Ninupi strata and two
the site was left to grazing and there the damaged Yata strata. The upper Ninupi deposit was extremely
vestiges of the Postclassic settlement survive at the dense with artifacts; the sherd density was 15,800
modern ground surface in a more random pattern. sherds per cu m. Unit 307 yielded 18 of the 38 pieces of
Without extensive excavation, the maze of stone walls copper/bronze recovered by the project, including one
visible on the slopes of Cerro Tenismo today cannot item from a Dongu phase level. The 10 metal items
be used to reconstruct the layout of the house lots of from Dongu phase deposits at Calixtlahuaca are the
Postclassic Matlatzinco. largest collection of Middle Postclassic bronze objects
yet excavated in central Mexico.
Houses and Domestic Deposits A burned wattle-and-daub structure excavated in
Approximately half of our excavations at the site Unit 315 also dates to the Dongu phase. This
targeted domestic structures. Six units contained excavation was initiated to test a small line of stones
located 200 m east of Structure 3, noted during the from the line of stones. Initial clearing revealed an
2006 survey. The slope was actively eroding in this amorphous layer of stones with several short seg-
area, exposing the tepetate several meters downslope ments of aligned stones both parallel and transverse
to the hillslope. The stones appear to have been survey), which we investigated to determine if
transported from upslope, perhaps from features obsidian tool production took place in this area.
associated with the house in Unit 316 located We did not obtain permission to excavate in the
immediately upslope from Unit 315 (FIG. 3). Exca- obsidian filled field, but secured permission to work
vation below the stones revealed a layer of burned on the terrace directly upslope (FIG. 3). The upslope
daub fragments encompassing a 463 m area, which (south) wall of the house in Unit 311 was well
in turn rested on a layer of heated clay. These features preserved and the quality of the masonry exceeded
represent the partially preserved floor of a burned that of other domestic structures. The east and west
wattle-and-daub house. Below these house remains walls were only partially preserved and the north
we located a canal cut into tepetate running downhill (downslope) wall had been destroyed by erosion.
to the east. The entire sequence in Unit 315 dates to Although the quantities of obsidian in this unit were
the Dongu phase. The amorphous stone layer that high, the kinds of obsidian tools and debitage
covered these deposits contains ceramics from all recovered did not differ greatly from those in other
three Postclassic phases, most likely eroded down excavations. A piece of the cloud wing from a
from Unit 316. massive Templo Mayor-style brazier, provisionally
Three other units yielded small portions of Post- interpreted as an offering, was recovered from terrace
classic houses associated with Dongu ceramics. Unit fill within the unit.
320, excavated in an area of narrow modern terraces, Unit 316 was excavated in a vacant lot immediately
revealed the poorly defined back wall of a Postclassic upslope from Unit 315. The landowner had used a
house built on a tread cut deep into the slope and a bulldozer to level the property in preparation for
patch of a pumice surfaced floor. Most of the house house construction, revealing a portion of a Postclassic
was destroyed by erosion and buried by the second stone pavement. This turned out to be one of two
stage of terracing. Unit 323 was placed on a high exterior pavements associated with a house with a
terrace that had also been farmed in recent years. packed earth floor. Like Unit 309 (described below),
Three parts of the terrace were tested, resulting in the the earth floor was covered with a layer of jumbled
identification of two areas of Postclassic occupation. stones, probably from the collapsed wall foundations.
Excavation in the central zone encountered dense Most architecture in Unit 316 was built during the
refuse deposits of the Dongu phase including a Ninupi phase. Occupation began, however, during the
horizontal lens of burned refuse and debris, but Dongu phase, as evidenced by poorly represented
without an associated structure. All of the ceramics structures and terraces immediately to the north and
from these deposits date to the Dongu phase. The a shallow pit with Dongu ceramics. The Ninupi
deposits were subsequently buried, most likely in the occupation began with the construction of a structure
Colonial period, by colluvium containing Yata cera- or structures represented by several fragmentary walls
mics. The eastern portion of Unit 324 contained an under the southern pavement. These were evidently
ancient gully or drain running perpendicularly across a dismantled before the visible pavements were laid over
modern field that exposed a cross section of three fill that contained Ninupi phase ceramics. A patch of
Postclassic terraces. The middle terrace included a burned earth north of the house also dates to this
Dongu phase pavement that covered a midden phase, as does an extensive midden north of the visible
intruding into tepetate. The midden contained an architecture that covered the burned earth. These
unusually large deposit of mostly intact small ceramic middens were then covered by Yata refuse.
vessels of the type we call Crude unfinished. We Many other excavations encountered Ninupi
cannot yet identify the function of these vessels; deposits, but without associated domestic architec-
however, their size, finish, and shapes suggest that ture. Three units close to the royal palace (303, 304,
they may have been used in some kind of craft activity. and 305) contained high densities of unabraded sherds
from this phase. In Unit 303, immediately north and
Ninupi phase downslope from the palace, Ninupi sherds were
Eleven excavation units included deposits of the
recovered from what appeared to be a Colonial gully;
Ninupi phase and two others may have been occupied
the materials in its fill likely eroded from the palace,
at this time. Two of the excavated housesin Units
which supports the notion that the palace itself dates
311 and 316were built during the Ninupi phase in
to the Ninupi phase.
areas with prior Dongu occupations. Unit 311 was
excavated in a modern field above a terrace wall Yata phase
where we suspected that modern terracing had buried Only 10 units yielded deposits securely dated to the
(and preserved) Postclassic deposits. The field imme- Yata phase; two of these units, 309 and 317, contain
diately downslope from this location contained dense houses. Unit 309 was located by testing a level area at
concentrations of obsidian (documented in the 2006 the bottom of a large sloping field below Unit 307.
Figure 9 Photograph of the house in Unit 309, looking south Figure 10 Photograph of the architecture in Unit 317,
(upslope). looking east.
The house located in Unit 309 is a rectangular features, a significant portion was recovered in levels
building with an earth floor and exterior pavements associated with the architecture of Unit 317. It is
on two sides (FIG. 9). Three superimposed (exterior) likely that there was an additional small wattle-and-
stone pavements were found on the north side and a daub structure on the main platform, perhaps lack-
pavement on the east side was at least partially ing a stone foundation. The identification of terrace
enclosed inside a room. These pavements were risers above and below the group suggests that the
carefully constructed using large stones with flattened architecture occupied the entire width of the terrace
upper surfaces. Only the western house wall is fully tread. The back of the terrace containing the archi-
preserved. The earth floor in the center of the house tecture is heavily cut into the underlying tepetate, a
was covered with an irregular pile of stones and other smaller scale example of the technique used to create
construction debris, similar to Unit 316; these were level areas for several of the monumental structures
removed prior to taking the photograph in Figure 9. at the site.
The deposit of apparent architectural collapse in- Unit 317 dates exclusively to the Yata phase. The
cluded several stones carved with geometric and stone wall foundations included several fragments of
glyph-like designs. All of the construction phases in sculpture reused as building stones. One burial was
Unit 309 date to the Yata phase and demonstrate that recovered on what would have been the terrace below
frequent remodeling occurred over a relatively short the one that contains the house, but it cannot be
time period. In Unit 309, like in Unit 307, we had associated definitively with the structure. Like most
only limited time to explore outside of the structure Yata phase occupations, the ceramics include a wide
and did not find any dense midden deposits. range of imported types from the Basin of Mexico,
Nevertheless, the unit did include two reconstruct- among them Aztec III Black-on-orange, Aztec III/IV,
able, locally produced, imitation Aztec III Orange- Aztec III spinning bowls, comales (distinctive shallow
ware tripod bowls. vessels used to toast tortillas), Guinda redware,
Unit 317 was placed to follow the edge of pave- globular copas (small cups), Texcoco molded censers,
ment visible at the ground surface near the site and Texcoco fabric marked salt vessels (Parsons
museum. This excavation revealed a series of stone 1966). One fragment of turquoise was found in the
features (FIG. 10), the northern (downslope) portions unit (the only such item in our excavations), but no
of which had been destroyed by erosion. A partial copper items.
house structure was present in the center of the unit, Several of the units described for the Dongu and/or
associated with a low platform revealing two phases Ninupi phases also contained Yata phase deposits.
of construction. There are two additional unattached The houses described in Units 307 (Dongu construc-
rooms or outbuildings with stone floors to the tion) and 316 (Ninupi construction) continued to be
southwest of the main house. Several abutting used through the Yata phase. One interesting feature
pavements cover the area between the main house of several of the Yata deposits is the presence of
and the additional rooms. Part of the earth floor of ceramic figurines depicting Spaniards (with Spanish-
the house showed traces of burning. Abundant style hats and clothing). Most of these figurines were
fragments of burned daub were recovered from this recovered from Unit 307, and several from Unit 317,
unit. Although most of the daub was recovered from in the uppermost levels of the Yata phase. These
the overburden from upslope that had covered the deposits lacked other markers of Colonial occupation
such as iron objects, glazed ceramics, or cow and reconstructing ancient settlement layouts from the
horse bones. The presence of these Spanish influenced modern layout of terrace walls and other field
figurines suggests that the Yata phase occupation of boundaries. Minimal agricultural activity between site
Calixtlahuaca continued after the Spanish conquest, abandonment and the present is often assumed by
probably for a short interval. survey archaeologists, particularly if the abandonment
coincided with the epidemics of the Colonial period.
House form and construction
Although conditions obviously vary from site to site,
Six of the excavated houses (in Units 307, 309, 311,
we suspect that most hilltop centers have suffered
315, 316, and 317) were sufficiently complete to make
significant damage from agricultural activities and
some inferences about house form and construction.
geomorphic processes.
Four had stone wall foundations that resemble the
What little we can reconstruct of the terracing at
bases for adobe walls of Aztec period houses in
Calixtlahuaca suggests a diffuse settlement boundary
Morelos (Smith 1992). Such walls are around 30 cm
and a lack of neat separation between urban and
in width and consist of two parallel rows of stones
rural spaces, long known to be characteristic of
laid several courses high. The two houses without
intensive agriculture cross culturally (Netting 1993),
wall foundations (in Units 315 and 316) contained
and of Postclassic central Mexico (Smith 2012: 189)
large quantities of burned daub, most likely repre-
and highland Oaxaca (e.g., Kowalewski et al. 2009) in
senting wattle-and-daub rather than masonry con-
particular. The sophisticated management of water
struction. One house (Unit 317) had both stone wall
flow down the terraced hillside at Calixtlahuaca has
foundations and abundant burned daub concentrated
few parallels in Mesoamerica, though the drains
in an area without walls, suggesting a combination
require better dating. The amount of labor invested in
of construction methods within the same house
the moving of stone and earth, and the modification
complex.
of soil properties is far greater than has been
All houses apparently had earth interior floors,
documented at other Aztec settlements (Smith 2008).
some covered with an irregular layer of pumice.
Smaller rooms in Units 309 and 317 had paved stone How this labor was organized is difficult to ascertain
floors. In contrast to the palace and temples ex- from the available archaeological data. Modern farm-
cavated by Garca Payon, lime plaster was not iden- ing practices are poor analogues for the Postclassic.
tified on floors or walls of the excavated houses. Due to the prevalence of absentee landholders and a
Three houses were associated with extensive, well general lack of interest in agricultural pursuits, the few
made, exterior stone pavements akin to those in the farmers active today liberally borrow stone from
royal palace. Although the erosion of the downslope neighbors and show little concern for the state of the
portion of most of the houses makes the identification risers and drains beyond their own plots. At the height
of their overall layout difficult, most appear to have of the Postclassic occupation, when the hillside was
one large room with one or two smaller rooms either covered with residences and public buildings, and
attached or freestanding. when commoners presumably derived most of their
Although similar in size to the small, one room, livelihood from their terraced house lots, we would
Aztec period houses of Morelos (Smith 1992; Smith imagine the existence and strict enforcement of rules
et al. 1999), these six houses vary far more in form regarding the maintenance of risers and drains, both
and materials than the Morelos examples. We because of the threats posed to neighbors by stretches
recovered over 500 kg of burned daub, most of it that fell into disuse and because of the importance of
from four houses: Unit 317 (376 kg), Unit 315 (90 maintaining property boundaries (Stone 1994). Many
kg), Unit 323 (61 kg), and Unit 316 (14 kg). Some of modern plots are aligned on the master drains whose
the daub bears impressions of maguey stalks. Three layout appears to be ancient. However, we do not
houses exhibited evidence of burning and destruction. interpret this as an indication of large scale planning
The destroyed houses date to different phases and or land allocation on the part of Calixtlahuacas rulers.
thus cannot be linked to a single event, such as the Comparative research shows that sophisticated hydr-
Aztec conquest or the abandonment of the site during aulic agricultural systems can develop gradually from
the Colonial period. Unlike the Morelos houses, minimally coordinated and incremental inputs by
those excavated at Calixtlahuaca cannot be divided multiple generations of farmers (Billman 2002: 373
easily into elite and commoner residences based on 374; Perez Rodrguez 2006: 3) and that preindustrial
size and form. elites concentrate on extracting surpluses rather than
on managing technical aspects of farming (Tilly 1985;
Discussion Mann 1986).
A number of features of the terraces and houses Perez Rodrguez (2006) commented on the possible
at Calixtlahuaca are worth discussing. Methodolo- role of social units (siqui) similar to the Aztec calpolli
gically, our excavations underscore the difficulty of in organizing terrace farming and other economic
activities in the Mixteca, but found it challenging to Calixtlahuaca is the dispersal and isolation of
identify them archaeologically. She focused instead public spaces. Even though it does not display the
on the household level and argued that, in keeping typical Aztec palace plan, Tizatlan is reminiscent of
with Nettings (1993) descriptions of smallholders Calixtlahuacas royal palace in its position on the
engaged in intensive agriculture, it was the stability periphery of the city. The multiple small plazas of
of usufruct rights that motivated people to invest Tlaxcallan that Fargher and colleagues (2010, 2011a,
in the physical infrastructure around their houses. 2011b) interpret as belonging to different residential
Archaeologically, such stability could be visible in the wards, however, are not identifiable at Calixtlahuaca.
repeated rebuilding and expansion of the same house Though covering more surface area than Calix-
and a certain degree of affluence. The latter is tlahuaca, Tlaxcallan lacks comparable monumental
presently being evaluated by artifact analyses. The architecture. Other currently studied centers with
former is not pronounced in the sample of houses we Postclassic terracing of comparable extent and sophis-
have excavated, but the fragmentary preservation of tication are in the Mixteca Alta and include Cerro
adobe and wattle-and-daub walls is a major issue. In Jazmn (Perez Rodrguez et al. 2011), Yucundaa
several cases, the damage to houses from repeated (Spores and Robles Garca 2007), and Coixtlahuaca
cycles of terracing may have destroyed all but the (Kowalewski et al. 2010). The concept of agrarian
earliest stages of construction. urbanism, as employed by archaeologists working in
The labor manifest in the terraces of Calixtlahuaca the Mixteca Alta (Kowalewski et al. 2009: 346349),
is in itself proof of agricultural intensification, may prove useful for interpreting Aztec centers such as
especially if we stick to traditional economic defi- Calixtlahuaca.
nitions that gauge intensification by inputs rather Calixtlahuaca also promises to bring new evidence
than outputs (yields) per unit area (Boserup 1965; to bear in the debate over the timing and causes of
Brookfield 1972; Turner and Doolittle 1978). We are land degradation in Mexico (Endfield and OHara
currently performing laboratory assays modeled on 1999; Fisher 2005; Hunter 2009). The episode of
work on the fertility of terrace soils in the Andes and terrace destruction and gullying at the site is
the American Southwest (Homburg and Sandor 2011; stratigraphically better constrained than in many
Sandor 1992) to address the subject of yields. We other cases. The site-specific focus of our research
hypothesize that the site represented prime agricultural allows us to relate the depositional environments
land suitable for growing maize and other demanding directly with erosional features a little higher upslope.
crops. This would set it apart from other terraced sites However, we defer any definitive statements on the
of the Aztec period (Borejsza et al. 2008; Cordova and subject of land degradation until more radiocarbon
Parsons 1997; Evans 1988; Smith and Price 1994), dates have been analyzed.
which were mostly rural backwaters founded on Although the site as a whole shows considerable
previously eroded and agro-ecologically marginal land continuity in its Postclassic occupation, locally there
often by socially marginalized groups such as the was much change and fluidity. Many house lots were
Otomi in the Basin of Mexico and Tlaxcala. The strong abandoned and others occupied across the transitions
association of Aztec households with maguey exploita- between chronological phases, and in several cases,
tion, cloth production, and other craft activities (Evans multiple stages of major architectural change occurr-
2005, 1990; Parsons 2010) independent of the quality of ed during an individual phase. Another interesting
farmland may not hold true for Calixtlahuaca. We plan feature of Calixtlahuaca is the prevalence of wattle-
to compare geoarchaeological analyses with ceramic and-daub construction. This type of construction
and lithic artifact studies to shed light on this issue. dominated Formative period architecture throughout
In general, terraced sites that are also large urban Mesoamerica (Flannery 1976). In later Prehispanic
centers may be a more direct analogue for Calix- periods and the ethnographic record (Prieto 1994),
tlahuaca, particularly in terms of the value of land however, it is more prevalent in the lowlands than in
and the economic activities of its households. In the highlands. We are unaware of other Postclassic
central Mexico, the only comparable and contem- sites in highland western Mesoamerica with a similar
poraneous center that has been explored with modern abundance of burned daub. There are few modern
field techniques is Prehispanic Tlaxcallan, the name examples of wattle-and-daub houses in the Toluca
that Fargher and colleagues (2010, 2011a, 2011b) give Valley, which suggests a disjunction in the forms of
to the remains spread over several hills on the peasant housing from the Late Postclassic period to
outskirts of the Colonial city of Tlaxcala, including the present. This situation contrasts with that in the
Tepeticpac and Ocotelulco. They consider all to be Basin of Mexico and Morelos, where the dominant
part of the same urban center, with nearby Tizatlan a Aztec period house typebuilt of adobe bricks on a
contemporaneous and related, but spatially isolated stone foundationsurvived as the traditional rural
rural seat of authority. One possible parallel with house form into the 20th century.
Finally, Calixtlahuaca was built on a hilltop, yet Charles Frederick (Ph.D. 1995, University of Texas)
defense and security do not seem to have been of is a consulting geoarchaeologist based in Dublin,
primary concern. Preexisting settlement patterns in the Texas. His research interests include ancient agricul-
valley or agricultural concerns may have been more tural terracing, irrigation, dynamic human landscapes,
important. The royal palace sits unprotected at the and geoarchaeology in Mesoamerica and Texas.
base of the hill and we did not locate any defensive
Isabel Rodrguez Lopez (Lic. 2005, Escuela Nacional
walls or ditches, although the terraces themselves
de Antropologa e Historia) has worked on a variety of
could have served a defensive purpose. We cannot yet
field projects in central Mexico and specializes in the
say why the city was built on a hill, but our ongoing
study of archaeological ceramics.
analyses of the data are revealing how the city was
built and how the use of the hilltop changed through Cynthia Heath-Smith (B.A. 1977, University of
time. Florida) is an adjunct faculty member in the School of
Human Evolution and Social Change at Arizona State
Acknowledgments
University. Her interests include Mesoamerican archae-
Permission for fieldwork at Calixtlahuaca was granted
ology, early state societies, subsistence patterns, and
by the Consejo de Arqueologa, INAH. Funding was
settlement patterns.
provided by two grants from the National Science
Foundation (nos. 0618462 and 0150482) and sup-
port from Arizona State University. The Colegio References
Mexiquense, which houses our laboratory in Toluca, Billman, B. R. 2002. Irrigation and the Origins of the Southern
Moche State on the North Coast of Peru, Latin American
has aided the research in many ways. We thank our Antiquity 13: 371400.
crew of student supervisors (from the United States, Borejsza, A. 2006. Agricultural Slope Management and Soil Erosion
in Tlaxcala, Mexico. Ph.D. dissertation, University of
Mexico, Germany, France, and Denmark), and we California, Los Angeles. Ann Arbor: UMI Microfilms.
acknowledge the important contributions of Juliana Borejsza, A., I. Rodrguez Lopez, C. D. Frederick, and M. D.
Bateman. 2008. Agricultural Slope Management and Soil
Novic and Amy Karabowicz to the research described Erosion at La Laguna, Tlaxcala, Mexico, Journal of
here. We also thank our field crew and laboratory staff Archaeological Science 35: 18541866.
Boserup, E. 1965. The Conditions of Agricultural Growth: The
from San Francisco Calixtlahuaca for their help. Economics of Agrarian Change Under Population Pressure.
Many other people from the community of San Chicago: Aldine.
Francisco Calixtlahuaca were very helpful, particu- Bronk Ramsey, C. 2008. Oxcal Program, v. 4.0. Oxford: Oxford
Radiocarbon Accelerator Unit.
larly Lic. Diosdado Millan Arellano, who was Pre- Brookfield, H. C. 1972. Intensification and Disintensification in
sidente del Comisariado de Bienes Comunales. We Pacific Agriculture: A Theoretical Approach, Pacific Viewpoint
13: 3048.
also thank two anonymous reviewers for helpful Carrasco, P. 1999. The Tenochca Empire of Ancient Mexico: The
comments on an earlier draft. Triple Alliance of Tenochtitlan, Tetzcoco, and Tlacopan.
Norman: University of Oklahoma Press.
Michael E. Smith (Ph.D. 1983, University of Illinois) Cordova, C. E., and J. R. Parsons. 1997. Geoarchaeology of an
Aztec Dispersed Village on the Texcoco Piedmont of Central
is Professor of Anthropology in the School of Human Mexico, Geoarchaeology 12: 177210.
Evolution and Social Change at Arizona State Doolittle, W. E. 1990. Canal Irrigation in Prehistoric Mexico: The
Sequence of Technological Change. Austin: University of Texas
University. His research interests include Aztec and Press.
Mesoamerican archaeology, households, economic and Endfield, G. H., and S. L. OHara. 1999. Degradation, Drought,
and Dissent: An Environmental History of Colonial
political organization, comparative urbanism, and Michoacan, West Central Mexico, Annals of the Association
archaeological epistemology. of American Geographers 89: 402419.
Etchevers, J. D., C. Hidalgo, C. Prat, and P. Quantin. 2003.
Aleksander Borejsza (Ph.D. 2006, University of Tepetates of Mexico, in R. Lal, ed., Encyclopedia of Soil
Science. New York: Marcel Dekker, 17451748.
California at Los Angeles) is Associate Professor at Evans, S. T. 1985. The Cerro Gordo Site: A Rural Settlement of
the Universidad Autonoma de San Luis Potos, the Aztec Period in the Basin of Mexico, Journal of Field
Archaeology 12: 118.
Mexico. He uses archaeology and earth science to Evans, S. T. 1988. Excavations at Cihuatecpan, An Aztec Village in
study past agriculture and rural life, the Preceramic the Teotihuacan Valley. Vanderbilt University Publications in
Anthropology 36. Nashville: Department of Anthropology,
and Formative periods of Mesoamerica, and late Vanderbilt University.
Quaternary fluvial environments. Evans, S. T. 1990. The Productivity of Maguey Terrace
Agriculture in Central Mexico During the Aztec Period,
Angela Huster (M.A. 2009, Arizona State Uni- Latin American Antiquity 1: 117132.
Evans, S. T. 2005. Men, Women, and Maguey: The Household
versity) is a Ph.D. candidate in Anthropology and Division of Labor Among Aztec Farmers, in R. E. Blanton,
Teaching Associate in the School of Human Evolution ed., Settlement, Subsistence and Social Complexity: Essays
Honoring the Legacy of Jeffrey Parsons. Los Angeles: Cotsen
and Social Change at Arizona State University. Her Institute of Archaeology, U.C.L.A., 198228.
research interests include Mesoamerican archaeology, Fargher, L. F., R. E. Blanton, and V. Y. Heredia Expinoza. 2010.
Egalitarian Ideology and Political Power in Prehispanic
the role of provincial peoples in empires, the organiza- Central Mexico: The Case of Tlaxcallan, Latin American
tion of market networks, and textile production. Antiquity 21: 227251.
Fargher, L. F., R. E. Blanton, V. Y. Heredia Espinoza, J. State University and Instituto Nacional de Antropologa e
Millhauser, N. Xiuhtecutli, and L. Overholtzer. 2011a. Historia.
Tlaxcallan: The Archaeology of an Ancient Republic in the Netting, R. M. 1993. Smallholders, Householders: Farm Families
New World, Antiquity 85: 172186. and the Ecology of Intensive, Sustainable Agriculture. Stanford:
Fargher, L. F., V. Y. Heredia Espinoza, and R. E. Blanton. 2011b. Stanford University Press.
Alternative Pathways to Power in Late Postclassic Highland Ouweneel, A. 1991. Growth, Stagnation, and Migration: An
Mesoamerica, Journal of Anthropological Archaeology 30: Explorative Analysis of the Tributario Series of Anahuac
306326. (17201800), Hispanic American Historical Review 71: 531
Fisher, C. T. 2005. Demographic and Landscape Change in the 577.
Lake Patzcuaro Basin, Mexico: Abandoning the Garden, Parsons, J. R. 1966. The Aztec Ceramic Sequence in the Teotihuacan
American Anthropologist 107: 8795. Valley, Mexico. 2 vols. Ph.D. dissertation, University of
Flannery, K. V., ed. 1976. The Early Mesoamerican Village. New Michigan. Ann Arbor: University Microfilms.
York: Academic Press. Parsons, J. R. 2010. The Pastoral Niche in Pre-Hispanic
Fowler, W. R. Jr. 1996. Special Section: Recent Chronological Mesoamerica, in J. E. Staller and M. D. Carrasco, eds.,
Research in Central Mexico, Ancient Mesoamerica 7: 215 Pre-Columbian Foodways: Interdisciplinary Approaches to
331. Food, Culture, and Markets in Ancient Mesoamerica. New
Frederick, C. D., and A. Krahtopoulou. 2000. Deconstructing York: Springer, 109136.
Agricultural Terraces: Examining the Influence of Construction Perez Rodrguez, V. 2006. States and Households: The Social
Method on Stratigraphy, Dating and Archaeological Visibility, Organization of Terrace Agriculture in Postclassic Mixteca
in P. Halstead and C. D. Frederick, eds., Landscape and Land Alta, Oaxaca, Mexico, Latin American Antiquity 17: 322.
Use in Postglacial Greece. Sheffield: Sheffield Academic Press, Perez Rodrguez, V., K. C. Anderson, and M. K. Neff. 2011. The
7994. Cerro Jazmn Archaeological Project: Investigating Pre-
Garca Castro, R. 1999. Indios, territorio y poder en la provincia hispanic Urbanism and Its Environmental Impact in the
matlatzinca: la negociacion del espacio poltico de los pueblos Mixteca Alta, Oaxaca, Mexico, Journal of Field Archaeology
otomianos, siglos XVXVII. Mexico City and Toluca: CIESAS, 36: 8399.
Instituto Nacional de Antropologa e Historia, and El Colegio Prieto, V. 1994. Vivienda campesina en Mexico. Mexico City:
Mexiquense. Secretara de Turismo and Secretara de Desarrollo Social.
Garca Castro, R. 2000. Los grupos indgenas del valle de Sanders, W. T. 1976. The Natural Environment of the Basin of
Toluca, Arqueologa Mexicana 43: 5055. Mexico, in E. R. Wolf, ed., The Valley of Mexico: Studies in
Garca Castro, R. 2004. Pobladores y gobierno indgena en Pre-Hispanic Ecology and Society. Albuquerque: University of
Metepec, siglos XVIXVIII, in M. T. Jarqun Ortega, ed., New Mexico Press, 5967.
Metepec: de aldea a ciudad. Toluca and Metepec: El Colegio Sanders, W. T., A. G. Mastache, and R. H. Cobean, eds. 2003. El
Mexiquense and H. Ayuntamiento Constitucional de Metepec, urbanismo en mesoamerica/Urbanism in Mesoamerica.
6984. University Park and Mexico City: Pennsylvania State
Garca Payon, J. 1936. La zona arqueologica de Tecaxic- University and Instituto Nacional de Antropologa e Historia.
Calixtlahuaca y los matlatzincas: etnologa y arqueologa Sandor, J. A. 1992. Long-Term Effects of Prehistoric Agriculture
(primera parte). Mexico City: Talleres Graficas de la Nacion. in Soils: Examples from New Mexico and Peru, in V. T.
Garca Payon, J. 1941a. La ceramica del Valle de Toluca, Revista Holliday, ed., Soils in Archaeology. Washington, D.C.:
Mexicana de Estudios Antropologicos 5: 209238. Smithsonian Institution Press, 217245.
Garca Payon, J. 1941b. Manera de disponer de los muertos entre Smith, M. E. 1992. Archaeological Research at Aztec-Period Rural
los matlatzincas del Valle de Toluca, Revista Mexicana de Sites in Morelos, Mexico. Volume 1, Excavations and
Estudios Antropologicos 5: 6478. Architecture/Investigaciones arqueologicas en sitios rurales de
Garca Payon, J. 1979. La zona arqueologica de Tecaxic- la epoca Azteca en Morelos, Tomo 1, excavaciones y arquitec-
Calixtlahuaca y los matlatzincas: etnologa y arqueologa tura. Memoirs in Latin American Archaeology 4. Pittsburgh:
(textos de la segunda parte), editado por W. Tommasi de University of Pittsburgh.
Magrelli y L. Manrique Castaneda. Biblioteca Enciclopedica del Smith, M. E. 2008. Aztec City-State Capitals. Gainesville:
Estado de Mexico 30. Toluca: Estado de Mexico. University Press of Florida.
Garca Payon, J. 1981. La zona arqueologica de Tecaxic- Smith, M. E. 2012. The Aztecs. Oxford: Blackwell Publishers.
Calixtlahuaca y los matlatzincas: Etnologa y arqueologa Smith, M. E., C. Heath-Smith, and L. Montiel. 1999. Excavations
(tablas, planos e ilustraciones de la segunda parte), editado of Aztec Urban Houses at Yautepec, Mexico, Latin American
por L. Manrique Castaneda. Biblioteca Enciclopedica del Estado Antiquity 10: 133150.
de Mexico 31. Toluca: Estado de Mexico. Smith, M. E., and M. Lind. 2005. Xoo-Phase Ceramics from
Hare, T. S., and M. E. Smith. 1996. A New Postclassic Oaxaca Found at Calixtlahuaca in Central Mexico, Ancient
Chronology for Yautepec, Morelos, Ancient Mesoamerica 7: Mesoamerica 16: 169177.
281297. Smith, M. E., J. Novic, A. Huster, and P. C. Kroefges. 2009.
Hernandez Rodrguez, R. 1950. Documentos relacionados con Reconocimiento superficial y mapeo en Calixtlahuaca,
San Bartolome Tlatelolco, Revista Mexicana de Estudios Expresion Antropologica 36: 3955.
Antropologicos 9: 233251. Smith, M. E., and T. J. Price. 1994. Aztec-Period Agricultural
Homburg, J. A., and J. A. Sandor. 2011. Anthropogenic Effects Terraces in Morelos, Mexico: Evidence for Household-Level
on Soil Quality of Ancient Agricultural Systems of the Agricultural Intensification, Journal of Field Archaeology 21:
American Southwest, Catena 85: 144154. 169179.
Hunter, R. 2009. Positionality, Perception, and Possibility in Spores, R., and N. M. Robles Garca. 2007. A Prehispanic
Mexicos Valle del Mezquital, Journal of Latin American (Postclassic) Capital Center in Colonial Transition:
Geography 8: 4969. Excavations at Yucandaa Pueblo Viejo de Teposcolula,
Kowalewski, S. A., A. K. Balkansky, L. R. S. Walsh, T. J. Oaxaca, Mexico, Latin American Antiquity 18: 333353.
Pluckhahn, J. F. Chamblee, V. Perez Rodrguez, V. Y. Heredia Stone, G. D. 1994. Agricultural Intensification and Perimetrics:
Espinoza, and C. A. Smith. 2009. Origins of the Nuu: Ethnoarchaeological Evidence from Nigeria, Current Anth-
Archaeology in the Mixteca Alta, Mexico. Boulder: University ropology 35: 317324.
Press of Colorado. Tilly, C. 1985. War Making and State Making as Organized
Kowalewski, S. A., G. G. A. Luis Barba Pingarron, B. A. Steere, Crime, in P. Evans, D. Rueschmeyer, and T. Skocpol, eds.
J. Blancas Vazquez, M. Yadira Cortes Vilchis, L. Lopez Bringing the State Back In. New York: Cambridge University
Zarate, A. Ortiz Butron, T. J. Pluckhahn, and B. Vilchis Press, 169186.
Flores. 2010. La presencia azteca en Oaxaca: La provincia de Tomaszewski, B. M. 2006. A Cost-Effective Approach to GPS/
Coixtlahuaca, Anales de Antropologa 44: 77103. ArcGIS Integration for Archaeological Surveying, ArcUser
Mann, M. 1986. The Sources of Social Power, Volume 1: A History Fall 2006: (http://www.esri.com/news/arcnews/fall06articles/a-
of Power from the Beginning to A.D. 1760. New York: cost-effective.html).
Cambridge University Press. Tomaszewski, B. M., and M. E. Smith. 2011. Politics, Territory,
Mastache, A. G., R. H. Cobean, A. Garca Cook, and K. G. Hirth, and Historical Change in Postclassic Matlatzinco (Toluca
eds. 2008. El urbanismo en mesoamerica/Urbanism in Valley, Central Mexico), Journal of Historical Geography 37:
Mesoamerica. University Park and Mexico City: Pennsylvania 2239.
Tommasi de Magrelli, W. 1978. La ceramica funeraria de Vargas Pacheco, E. 1975. La ceramica, in R. Pina Chan, ed.,
Teotenango. Biblioteca Enciclopedica del Estado de Mexico Teotenango: el antiguo lugar de la muralla. Mexico City:
61. Toluca: Estado de Mexico. Gobierno del Estado de Mexico, 189264.
Turner, B. L. II, and W. E. Doolittle. 1978. The Concept and West, R. C. 1971. Population Densities and Agricultural Practices
Measure of Agricultural Intensity, Professional Geographer in Precolumbian Mexico, with Emphasis on Semi-Terracing, in
30: 297301. Verhandlungen des XXXVIII Internationalen Amerikanisten-
Umberger, E. 1996. Aztec Presence and Material Remains in the kongresses. Munchen: Klaus Renner, 361369.
Outer Provinces, in F. F. Berdan, R. E. Blanton, E. H. Boone, Whitmore, T. M., and B. L. Turner II. 2001. Cultivated Landscapes
M. G. Hodge, M. E. Smith, and E. Umberger, eds., Aztec Imperial of Middle America on the Eve of Conquest. New York: Oxford
Strategies. Washington, D.C.: Dumbarton Oaks, 151180. University Press.
Umberger, E. 2007. Historia del arte e Imperio Azteca: la Wilken, G. C. 1987. Good Farmers: Traditional Agricultural Resource
evidencia de las esculturas, Revista Espanola de Antropologa Management in Mexico and Central America. Berkeley:
Americana 37: 165202. University of California Press.