P1: GKW/RKP P2: GKW

Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

Journal of World Prehistory, Vol. 15, No. 2, 2001

Reassessing the Developmental and Chronological

Relationships of the Formative of Coastal Ecuador
John Edward Staller1

The Ecuadorian Formative was initially interpreted as a result of long-distance

diffusion, and migration, more recent research has changed our perception
of the Valdivia, Machalilla, and Chorrera culture complexes. In this analy-
sis, these archaeological investigations are reassessed in order to reevaluate
our understanding regarding the nature of these prehistoric cultures. Forma-
tive chronologies, distributions, interrelationships are reviewed in the light of
Pre-Hispanic occupations in El Oro Province. Previous investigations on the
Ecuadorian littoral are examined. Results indicate (1) Valdivia and Machalilla
are part of the same ceramic tradition and (2) the interrelationships of these
complexes to Late Formative Period ceramic sequences are more complex
than previously assumed. (3) A major adaptive shift related to an increased
dependence upon agriculture occurred during Valdivia VI–VIII, and contin-
ued into the Late Formative with the widespread distribution of Chorrera
related material culture. (4) Long-distance interaction with the Ecuadorian
highlands resulted in the spread of what may be termed a Chorrera horizon
and formed the basis of subsequent Andean cosmology where it was symbol-
ically represented according to local and regional artistic canons and over a
vast area of Andean South America.
KEY WORDS: Ecuadorian prehistory; Formative ceramics; radiocarbon dating.

INTRODUCTION

Before the mid-1960s most of what was known about pre-Hispanic

Ecuador centered around Inca expansion, and more recent coastal and
1 Department of Anthropology, The Field Museum, 1400 S. Lake Shore Dr., Chicago, Illinois
60605.

193

0892-7537/01/0600-0193$19.50/0 °
C 2001 Plenum Publishing Corporation
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

194 Staller

highland culture complexes (Bennett, 1946a,b; Bushnell, 1951; Collier, 1947,

1948; Collier and Murra, 1943; Dorsey, 1901; Ferdon, 1940a,b, 1941a,b, 1942;
Jijon y Caamaño, 1919, 1920, 1941, 1952; Kroeber, 1948; Murra, 1946; Rowe,
1946; Saville, 1909, 1910; Stirling, 1963; Uhle, 1920a,b, 1922a,b,c, 1923, 1924,
1937). It was the pioneering research of Estrada, Evans, and Meggers that
greatly revised our understanding of the prehistory of coastal Ecuador
(Estrada, 1956, 1957, 1958, 1962a, 1962b; Evans et al., 1959; Evans and
Meggers, 1958; Evans and Meggers, 1982; Meggers, 1987; Meggers et al.,
1965; Meggers and Evans, 1962). In their investigation of archaeological
sites in coastal Guayas Province, these archaeologists recorded a previously
unsuspected chronological depth along the littoral, documenting the exis-
tence of precocious ceramic innovation associated with the Early Formative
Period Valdivia Phase. Much has been written about the origins and devel-
opmental associations of the Valdivia Phase and of the Formative sequence
of Ecuador in general. Recent research in various subregions of the coast has
greatly revised our understanding of these developmental relationships and
it is for this reason that this synthesis has been generated. With the goal of
facilitating dialogue among specialists in the field, discussion of the Valdivia
Phase in this study will be in terms of the eight phase ceramic sequence of
Hill (1972/74) since it is the most widely cited chronology by Ecuadorian
specialists. The Hill seriation is based upon ceramic diagnostics from a num-
ber of sites in the Santa Elena Peninsula and the coastal regions of Guayas
Province (Fig. 1, Table I).
It should be stated at the outset, that the geology and geomorphol-
ogy of coastal Ecuador are tectonic and at times highly active and have
over the course of time affected sociocultural development along the lit-
toral, sometimes in dramatic and highly significant ways (Damp et al., 1990;
Issacson, 1994; Staller, 1994); however, the extent to which such geomor-
phological evidence has been integrated into archaeological reconstruction
is limited compared to coastal Peru (see Moseley, 1983; Moseley et al., 1981).
These differences in research foci are perhaps due in part to the extensive
prehistoric irrigation canals in coastal Peru. Detailed analysis of irrigation
networks has called attention to geomorphological changes through their
effects upon such systems. Irrigation features are absent in coastal Ecuador
except for the lowlands of the Gulf of Guayaquil, where they appear with
Milagro Quevedo culture as extensive networks of raised fields (Denevan
and Mathewson, 1985). Despite the fact that tectonic events, including ig-
neous volcanic eruptions, coastal uplift as well as subsidence have dramati-
cally altered the coastal landscape during different times in prehistory, the
timing and nature of such geological events and their consequences for so-
ciocultural development are only now being more closely examined.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 195

Fig. 1. Coastal provinces and major streams of the Ecuadorian littoral.

P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

196 Staller

Table I. Chronological Table of the Early Formative Period Valdivia Phase

Note. Based primarily upon the Hill, 1972/74 seriation.

P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 197

ORIGINS, DIFFUSIONS, INVASIONS, AND MIGRATIONS

All archaeological phases in coastal Ecuador are based upon changing

settlement patterns, diagnostic ceramic frequencies, and associated radio-
carbon dates. However, the origins of these ceramic phases were initially
interpreted before regional-temporal variations in settlement patterns, ce-
ramic attributes, and the associated cultural chronologies were firmly estab-
lished. Consequently, the origins of the Ecuadorian Formative were initially
perceived as a result of diffusion, migration, or invasion (Uhle, 1922c, 1923,
1924). Meggers, Evans, and Estrada (1965, pp. 157, 167–168) interpreted ce-
ramic innovation in the Valdivia Phase as a result of diffusion, in this case
transpacific contact from the Jomon culture of coastal Japan. In support of
this hypothesis, Meggers, Evans, and Estrada (1965) illustrated striking sim-
ilarities between Valdivia and Jomon diagnostics (also see Estrada et al.,
1962; Evans and Meggers, 1966, p. 63; Meggers, 1987). The Valdivia ceramic
tradition was subsequently credited with having influenced similar techno-
logical processes elsewhere in the hemisphere through stimulus diffusion
(Ford, 1969). The Middle Formative Period Machalilla Phase was similarly
perceived as the result of migrations of ceramic bearing cultures from re-
gions outside of the coast (Meggers et al., 1965). On the basis of excavations
at La Chorrera, the elaborately crafted zoomorphic and anthropomorphic
vessels and whistling bottles of the Late Formative Period were similarly in-
terpreted as having Mesoamerican origins (Evans and Meggers, 1957, 1966,
1982; Meggers, 1966, pp. 61–62; Meggers and Evans, 1962, see also Uhle,
1920b, 1922c, 1924). Other scholars projected a variety of different routes
and historical factors surrounding the spread of Formative pottery tech-
nology to and from the Ecuadorian coast (Bischof, 1967, 1972, 1975, 1980;
Coe, 1960; Collier, 1968; Grove, 1982; Lathrap, 1963, 1967, 1970, 1973, 1974;
Lathrap et al., 1975; Uhle, 1923). On the basis of these various interpretations,
the Formative of coastal Ecuador was initially conceived of as transplanted
from other regions and environments that supported more complex forms of
social organization. Remnants of such diffusionary explanations of the past
remain. Table II summarizes the most recent archaeological evidence regard-
ing the developmental interrelationships of the various cultural sequences.

POLITICAL AND ARCHAEOLOGICAL SUBREGIONS

OF COASTAL ECUADOR

In the following analysis, coastal Ecuador has been divided into various
archaeological subregions (Fig. 2). Archaeological research on Ecuadorian
prehistory has generally been discussed in reference to provincial boundaries.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

198 Staller

Table II. Chronological Table of Archaeological Phases of Coastal Ecuador

However, in this study the various coastal regions are also differentiated in
terms of sociocultural development and the distributions of the various For-
mative culture phases. I have identified a total of six subregions. They include
(1) SW coastal Ecuador, which incorporates the Santa Elena Peninsula and
SW coastal Guayas Province, especially the area between the Rı́o Verde
and Rı́o Valdivia, and coastal regions to the SE. It is this subregion that
has received the most attention by archaeologists and contains late prece-
ramic and the earliest Formative occupations. In fact, the late Preceramic
Period Las Vegas complex is peculiar to this subregion (Stothert, 1976, 1983,
1985). The Formative ceramic sequences derived from this area of the coast
constitute the most detailed database of what has been published regard-
ing the Ecuadorian Formative. (2) Southern Manabı́ Province refers to the
area around the towns of Manta and Machalilla as well as La Plata Island
and the coastline and coastal hills extending to Bahia de Caraquez and the
area around the Rı́o Chone. Southern Manabı́ has also been extensively
investigated and the settlement patterns and ceramic diagnostics indicate
an intense post Early Formative Period occupation. (3) Northern Manabı́
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 199

Fig. 2. Various archaeological subregions of Ecuador discussed and defined in this study.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

200 Staller

Province refers to the regions between the Rı́o Chone to the provincial
boundary with Esmeraldas Province. Archaeological research in this sub-
region has been for the most part concentrated along the Rı́o Jama. (4)
Esmeraldas Province has been the subject of extensive archaeological re-
search by various scholars but is in a general sense environmentally and cul-
turally distinct from the rest of coastal Ecuador. (5) The Gulf of Guayaquil,
which refers to those regions, river valleys, inlets, islands and coastlines that
empty into and surround the gulf to the regions north of the Straits of Jambelı́.
The Gulf of Guayaquil includes portions of Guayas Province and in this
analysis incorporates the Guayas Basin as well. In the later periods this
subregion is climatically, developmentally, and culturally distinct from the
other subregions within Guayas Province. Archaeological research on the
Formative has been rather sporadic, primarily focused on Puná Island and
in and near the cities of Milagro and Guayaquil. Our understanding of the
prehistory of this region has generally lagged behind other areas of the coast.
Survey and excavations on Formative Period sites have been largely focused
within the immediate vicinity of the Rı́o Daule and/or Rı́o Babahoya. How-
ever, the Punta Arena Peninsula was the focus of concentrated research in
the early 1960s to the mid-1970s (Fig. 1). Nevertheless, the area north and
east of Guayaquil, especially the regions between the eastern side of the
Colonche Hills and the yungas along western corridor of the Andes Moun-
tains have received only limited archaeological research. (6) Coastal El Oro
Province refers to the barrier island estuary of the Straits of Jambelı́, specif-
ically the area between the Rı́o Jubones near the city of Machala, and the
border of Peru at the Rı́o Zarumilla. Most archaeological research in this
subregion has been relatively recent and centered in the vicinity of the low-
lands and coastal hills between the Rı́o Arenillas and Rı́o Buenavista. These
various subregions directly correspond to areas of archaeological, environ-
mental, and climatic variability.
All the aforementioned subregions except Esmeraldas Province are di-
rectly pertinent to this synthesis of the Ecuadorian Formative. Archaeolog-
ical research in the Santiago–Cayapas Region of Esmeraldas Province ap-
pears to reflect a prehistory largely distinct from the rest of coastal Ecuador
(DeBoer, 1996; Ferdon, 1941a; Ferdon and Corbett, 1941; Saville, 1909; Uhle,
1923; Valdez, 1987, 1992). These distinctions are evident in the material
culture, which in later periods appear to reflect stylistic influences from
Mesoamerica and Colombia and such affiliations are also implied by var-
ious technological innovations as well. Complex sociocultural development
appears to have been for the most part post-Formative (DeBoer, 1996).
However, when evidence of complexity does appear in the archaeologi-
cal record it is in the form of earthen mound constructions, finely crafted
ceramic vessels and figurines, and personal adornments made of precious
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 201

metals (Ferdon and Corbett, 1941; Jijon y Caamaño, 1941; Saville, 1909,
1924; Valdez, 1992). As a result, archaeological sites in this subregion have
been extensively looted and constitute major collections in museum exhibits.
This is particularly the case with ancient objects associated with the La Tolita
culture, which is renown for the finely crafted objects of precious metal as-
sociated with funerary and ceremonial contexts (Ferdon and Corbett, 1941;
Saville, 1909, 1924; Valdez, 1992). Although such technological innovations
have much in common with pre-Hispanic cultures of Colombia, the finely
crafted anthropomorphic and zoomorphic vessels of the La Tolita culture
also have clear stylistic affinities to assemblages from other subregions of the
Ecuadorian coast (Ferdon and Corbett, 1941; Valdez, 1992). Valdez (1987)
also carried out excavations at the type-site and the chronology developed
from this research is presented in Table II. Although the Formative base
in Esmeraldas Province has been designated as the Chorrera Phase, it is as
yet not well defined in terms of diagnostic attributes. In a synthesis of the
most recent archeological evidence from the region, DeBoer, (1996, p. 68)
has stated that the evidence for Early Formative occupation consists of two
sherds which are “vintage Valdivia” from a mixed deposit. DeBoer (1996,
pp. 68–70; Fig. 4.2) states that one of the sherds was found near a site with a
3580 B.P. date, but the contextual association and the dated deposit are prob-
lematic. The earliest established ceramic horizon in Esmeraldas Province
dates to the latter half of the third millennium B.P.
Most of what is known about the Formative of coastal Ecuador is de-
rived from archaeological investigations centered in SW coastal Ecuador and
S Manabı́. As research has spread to other subregions our understanding of
the pre-Hispanic sequences has become increasingly complex in recent years.
Early interpretations of Formative prehistory were perceived as a result of
migrations and diffusion. The more recent ceramic and archaeological ev-
idence from coastal El Oro and northern Manabı́ have challenged most of
these interpretations. Since the initial perceptions of these Formative cul-
tures to one another and to prehistoric cultures in other regions were based
upon diffusion and migration as primary causal factors in culture change,
few syntheses have appeared that incorporate more recent results. The pri-
mary objective of this analysis of the Ecuadorian Formative is to reassess
the pre-Hispanic sequence in light of the most recent research.

CLIMATE, ENVIRONMENT, AND ECOLOGY

IN COASTAL ECUADOR

The primary factors effecting the climate and environmental zones of

coastal Ecuador are the cold Humboldt Current which runs along the eastern
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

202 Staller

Pacific and variations in the geographic distance of the Andean Cordillera

to the coast lowlands.
Annual precipitation patterns over most of SW and S coastal Ecuador
depend upon the changing ocean currents that push moisture in the form
of clouds along the western edges of the Andean Cordillera. When the cold
water of the Humboldt Current comes into contact with the warm tropical
air temperatures near the equator, it condenses into fog (garúa) and forms
low lying clouds that move to the east toward the coastal hills and foothills
of the cordillera. In SW coastal Ecuador, the clouds usually dissipate east of
the coastal hills but reform along the western edge of the Andean Cordillera
where they build up and release moisture in the form of precipitation in the
high sierra. The precipitation is carried as fresh water by stream channels or
absorbed into the groundwater and underground aquifers that feed into the
lowland savanna (pampa) and eventually empty into the sea.
The Humboldt Current generally runs parallel to the western coast of
South America until it reaches the Santa Elena Peninsula at about 2◦ South
Latitude in SW coastal Ecuador. The ocean currents then flow westward
toward the Galapagos Islands, and beyond along the equatorial South Pacific.
During periods of El Niño Southern Oscillation (ENSO) phenomenon, the
cold water of the Humboldt Current is initially replaced by the warm water of
the El Niño countercurrent along western South America in the area of the
Santa Elena Peninsula. ENSO has a history of greatly affecting the climate,
ecology (both marine and terrestrial) and the human carrying capacity over a
vast area of western South America (Fagan, 1999; Holmgren et al., 2001; Kerr,
1999; McPhaden, 1999; Zuta et al., 1976). Although all of coastal Ecuador is
to some extent affected by ENSO, the subregions most directly affected are
S Manabı́, SW coastal Ecuador, and coastal El Oro Province (Rollins et al.,
1987; Sandweiss, 1986, p. 18; Zuta et al., 1976 ) (Fig. 2).
Since the archaeological subregions along the Ecuadorian coast de-
scribed above also roughly correspond to regions of climatic and environ-
mental variation, they are discussed below in terms of the distribution of
Formative archaeological remains, and in terms of their carrying capacity
and potential for agricultural intensification.
Esmeraldas Province, Northern Manabı́, and the Gulf of Guayaquil
are the wettest subregions of coastal Ecuador, while S Manabı́, SW coastal
Ecuador, and coastal El Oro are the most arid (Jijon y Caamaño, 1952,
Fig. 2, map 1). The climate in S Manabı́, SW coastal Ecuador, and coastal El
Oro is characterized as semiarid, distinguished by marked annual variation
of rainy and dry seasons (Cañadas Cruz, 1983; Ferdon, 1950). The natural
vegetation has been classified as dry tropical forest, with hygrophytic or dry
premontane forest in higher elevations (>100 masl), and concentrations of
diverse tree species and shrubs along the banks of the major coastal streams
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 203

(Cañadas Cruz, 1983; Jijon y Caamaño, 1952, Fig. 2 map 1). The vegetation
regimes in these regions of coastal Ecuador consist of a dry tropical xero-
phytic formation, made up of a variety of succulents, trees (Jacuinia and
Capparis spp.), shrubs, and a diverse array of columnar cacti (Cañadas Cruz,
1983; Ferdon, 1950; Parker and Carr, 1992). The coastal ecology over most of
Ecuador is characterized by a high incidence of endemic plants and animals
(Parker and Carr, 1992, p. 39; Svenson, 1946, p. 415).
In SW Ecuador, the coastal streams are for the most part seasonal.
The shoreline in SW coastal Ecuador, and N and S Manabı́, are separated
from the Andes by a range of coastal hills referred to as the Cordillera
de Colonche and ranging between 100 to 300 masl (Fig. 3A). In the Gulf
of Guayaquil, coastal El Oro Province and the regions north of the Rı́o
Chone in N Manabı́, the various stream channels flow year-round. Differ-
ences in stream flow in N Manabı́ and coastal Esmeraldas Province are
related to a higher (>1000 mm) average annual precipitation in those re-
gions (Delavaud, 1982; Zeidler and Kennedy, 1994). Consequently, Northern
Manabı́ and the Gulf of Guayaquil are much more adapted to year-round
plant cultivation.
Differences in annual stream flow in these various subregions are also
related to the geographic proximity of the Andean Cordillera to the coast-
line. In areas of coastal Ecuador where the coastal lowlands are separated
from the cordillera by coastal hills there is greater variability in precipita-
tion (Fig. 3A). In coastal El Oro and the lowlands to the north in the Gulf of
Guayaquil the foothills of the cordillera begin their ascent only 15 km from
the seashore (Fig. 3B). The topography and geography of these regions of
coastal Ecuador more closely approximate the conditions in far north coastal
Peru (see Burger, 1992, Fig. 10).
The proximity of the cordillera to the coast and differences in over-
all oceanic depth in these subregions has significance for the distributions
and densities of maritime species at different periods of the annual cycle.
They also have important effects upon the lowland vegetation, since many
endemic species are adapted to the moisture provided by the coastal fog.
These geographic differences also have variable effects at the top of the
food chain especially upon human adaptation and population density.
The Andean Cordillera represents the dominant physiographic feature
in western South America and is generally a formidable dispersal barrier
for plants and animals (Fig. 4). However, just south of El Oro Province, be-
tween 4◦ and 6◦ S Latitude, the spine of the Andean Cordillera shifts from a
northeast–southwest to a northwest–southeast direction. Botanists and ge-
ologists refer to this area of the cordillera as the Huancabamba Deflection or
Depression (Sagástegui et al., 1999). The mountain valleys east and south of
coastal El Oro are the narrowest and lowest in altitude of the entire Andean
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

204 Staller

Fig. 3. Topographic profiles of the western Andean Cordillera in Ecuador. (A) Topographic
characteristics of S Manabı́ and SW coastal Ecuador, (B) Topographic characteristics of El Oro
Province at Santa Rosa. The dotted line at 3200 masl represents the maximum altitude for the
cultivation of most South American varieties of maize.

Cordillera (see Figs. 3B and 4). The local geography, ecology, and environ-
ment of coastal El Oro and the eastern portion of the Gulf of Guayaquil
subregion provide an ideal setting for the movement of human populations
and cultural plants between the coast, western highlands (yungas), and the
Upper Amazon (ceja de montaña).
Species diversity and a high incidence of endemic plants and animals
(Parker and Carr, 1992, p. 39–40) characterize the ecology in the southern re-
gion of the Ecuadorian cordillera. Such ecological complexity is believed to
 P1: GKW/RKP
P2: GKW
Journal of World Prehistory [jowo]

The Formative of Coastal Ecuador

PP270-346283
October 27, 2001
20:2

Fig. 4. Topographic profile of the Andes Cordillera in Ecuador. The profile illustrated represents the western spine of the Andes. The cordillera
east of the Gulf of Guayaquil and again in the south in El Oro is the part of the Andes that has the lowest overall altitude. The Andes in El Oro
Province essentially represents the northern portion of the Huancabamba Deflection.
205
Style file version Nov. 19th, 1999
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

206 Staller

be due to habitat heterogeneity and fragmentation (Sagástegui et al., 1999).

The Huancabamba Deflection is the only area of the Andes where related
Amazonian and Pacific versants have been reported on both the western and
eastern sides of the cordillera. The environmental zones in coastal El Oro
Province and the eastern portion of the Gulf of Guayaquil are more “tightly
packed” than in SW coastal Ecuador or S Manabı́. Such environmental and
ecological differences have important implications for settlement patterning
and population densities, as well as long-distance exchange with inland so-
cieties. Various lines of botanical and archaeological evidence suggest that
the environmental and ecological conditions that characterize the Huan-
cabamba Deflection appear to have facilitated the long distance exchange,
plant cultivation, and movement of plants and subsistence resources on both
sides of the cordillera.
Coastal El Oro Province, SW coastal Ecuador, and S Manabı́ are arid
coastlines distinguished by a marked variation of dry and rainy seasons.
Coastal El Oro is somewhat distinct since it has a true transitional envi-
ronment or ecotone (Staller, 2001). Rainfall significantly increases north
of Machala into the tropical lowlands of the Guayas Basin, and becomes
increasingly arid towards Peruvian border (Table III, c–e). A rainfall gradi-
ent also exists from east to west exemplified by a 30% reduction in annual
precipitation between Machala and Puerto Bolivar despite an overland dis-
tance of only 5 km, and a 75% annual reduction between Puerto Bolivar and
Zorritos, Peru, a N–S distance of only 35 km (Staller, 1994). Over 90% of the
annual precipitation along the coast occur during the 3-month long rainy sea-
son. The growing cycle in coastal El Oro, SW coastal Ecuador, and S Manabı́
is limited by the dry season, which extends from early April to late December
(Table III).
The precipitation cycles in coastal El Oro, SW coastal Ecuador, and
S Manabı́ are normally insufficient to sustain year-round cultivation with-
out irrigation and detailed analyses of aerial photos and systematic regional

Table III. Average Monthly Precipitation in S. Manabı́, SW Coastal Ecuador, and Coastal El
Oro Province
Weather
station Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec
(a) Manta 69 64 64 21 4.0 2.0 3.0 0.0 2.0 0.0 1.0 5.0
(b) La Libertad 35.3 92.9 84.5 17 2.3 0.8 0.5 1.2 1.0 1.0 2.5 6.3
(c) Machala 109.9 156.9 177 105.9 46.9 19 17 20 16 18 9.9 11.9
(d) P. Bolivar 66 134.8 160 87.9 11.9 13.9 9.9 8.8 9.9 13.9 7.1 10.9
(e) Zorritos, Peru 18 56.1 33 17 0.0 1.0 1.0 1.0 0.0 0.0 1.0 1.0
Note. All Measurements are in mm. (after Wernstedt 1972). Average annual precipitation:
Manta (213.2 mm.); La Libertad (245.4 mm.); Machala (708.9 mm.); Puerto Bolivar (505.9 mm.);
Zorritos, Peru (129 mm.).
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 207

Table IV. Average Monthly Temperatures in S. Manabı́, SW Coastal Ecuador, and Coastal El
Oro Province
Weather
station Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec
(a) Manta 25.8 24.7 24.6 24.6 24.1 23.9 23.9 24 24.4 25.2 25.2 25
(b) La Libertad 25.9 26.2 26.3 26.3 26.1 25.7 25.7 25.7 25.7 25.8 23.7 25.7
(c) P. Bolivar 25.8 26.5 26.6 26.5 25.5 23.8 23.2 22.7 23 23.2 23.7 25
(d) Zorritos, Peru 26.6 26.8 27.1 26.1 25.6 24.2 23 22.7 22.7 23.2 23.3 25.1
Note. All Measurements are in ◦ C. (after Wernstedt 1972). Average annual temperatures: Manta
(24.6◦ ); La Libertad (25.9◦ ); Machala (24.7◦ ); Puerto Bolivar (24.7◦ ); Zorritos, Peru (24.7◦ ).

surveys indicate an absence of pre-Hispanic irrigation technology in these

various subregions. However, artificial earthen retention ponds referred to as
albarradas have been identified in SW coastal Ecuador and appear to date to
the Late Formative Period (Stothert, 1995, pp. 148–149). Such features prob-
ably provided year-round water for consumption, but were apparently not
designed for agriculture. In the wetter, more semitropical Gulf of Guayaquil
subregion extensive systems of pre-Hispanic raised fields have been identi-
fied but are believed to be post-Formative (Denevan and Mathewson, 1985;
Marcos, 1987).
The average monthly temperatures throughout coastal Ecuador are sta-
ble ranging from 22.7 to 27.1◦ C and average 24.7◦ C (Table IV), and the rel-
ative humidity in these latitudes generally ranges from 70 to 90% (Cañadas
Cruz, 1983; Delavaud, 1982, p. 19). The general stability of the air temper-
atures is directly related to the cold Humboldt Current, which produces a
dense fog most of the dry season throughout all of coastal Ecuador. Even
during the driest months, cloud forests along the lower western slopes of the
Andes trap and recycle huge quantities of moisture (Lippi, 1998, p. 52). In
coastal El Oro, such conditions maintain a high water table, which also plays
a role in the year-round discharge of the local streams. These geographic,
climatic, environmental, and ecological conditions select for a high degree of
economic flexibility and played an important role in the early development
of sedentism, complex social organization, and long-term subsistence adap-
tation. This detailed discussion of the coastal climate, environment, ecology,
geography and geomorphology is in response to much of what has been writ-
ten in the archeological literature regarding pre-Hispanic Formative adap-
tation, diet, and settlement patterns. These factors have historically shown
a selection for adaptive flexibility rather than intensification and specializa-
tion and such long-term adaptive patterns are supported by the ethnographic
literature.
Coastal and inland societies in the various subregions of the Ecuado-
rian littoral show a clear pattern of economic interdependence (Lathrap
et al., 1977; Staller, 1994, 2001). Seashore fishing communities are primarily
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

208 Staller

adapted to maritime and aquatic resources, and carry out reciprocal ex-
change and resource sharing with agricultural communities dispersed fur-
ther inland along the coastal streams. Such cultural and economic interde-
pendence has been recorded ethnographically over most all of coast and
various lines of archaeological evidence suggest that such a pattern may be
of great antiquity (Damp, 1984a; Lathrap et al., 1977; Raymond, 1988; Staller,
1994, 2000a, 2001). The seasonal availability of maritime and terrestrial plant
and animal resources are complimentary in various ways. Such adaptive in-
terrelationships are evident by the cyclical variations in annual subsistence
strategies in coastal El Oro reconstructed on the basis of ethnographic ev-
idence generated in the field by the author and illustrated in Fig. 5. The
ethnographic evidence suggests that the availability of seasonally specific
resources is in various ways complimentary, and is adjusted by either so-
lar and/or lunar cycles. The near-shore mangrove and estuary environments
along S Manabı́ and SW coastal Ecuador are not of the scale and density as
those of the other subregions, and consequently appear to have maintained
lower population densities.
Archaeological evidence and stable carbon isotope data derived from
pre-Hispanic skeletons at numerous coastal sites indicate a distinct pat-
tern. They suggest that the archaic Las Vegas culture and the early Valdivia
I–III adaptation was primarily focused upon hunting and seasonal gathering
(van der Merwe et al., 1993). During the later portion of the Early Forma-
tive Period, sedentary pottery-bearing societies along the coast had vari-
able adaptive patterns, but were more focused upon maritime resources and
small-scale plant cultivation. Furthermore, once the overall adaptive pattern
shifts to sedentism along the coast, the economic and cultural interdepen-
dence between maritime societies and inland farmers appears to increase and
then become stable. This pattern of interdependence exists in the present,
and it may extend back to the Early Formative Period (Damp, 1984a, 1988;
Lathrap et al., 1977; Meggers et al., 1965; Raymond, 1988; Staller, 1994). In
fact, various lines of evidence suggest that the economy that formed the ba-
sis for sedentism along the Ecuadorian coast involved the year-round avail-
ability of maritime resources, seasonal hunting, and small-scale cultivation
(Estrada, 1956, 1958; Lanning, 1967; Meggers, 1966; Meggers et al., 1965;
Staller, 1994, 2001).

CRITICAL RESOURCES AND PATTERNS OF EXCHANGE IN

COASTAL ECUADOR

The economic terrestrial and aquatic resources along the coast are
distinct from those in the highlands and such differences appear to have
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 209

Fig. 5. Annual subsistence cycles in southern coastal El Oro Province.

promoted long-distance interaction early in the pre-Hispanic sequence. Ex-

change in cotton, salt, fish, and certain species of marine bivalves provided
an early stimulus to interaction with the inter-montane and highland valleys
(Staller, 1994, p. 186). Domesticated cotton has ancient origins in coastal
Ecuador and was no doubt essential in the manufacture of fishing nets and
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

210 Staller

textiles (Damp and Pearsall, 1994; Marcos, 1979, p. 18; White, 1985, p. 50).
Consumption of maritime and aquatic resources provides the necessary io-
dine lacking in the bland high carbohydrate diets of agricultural populations
living inland from the coast or in the Andean highlands (Burger, 1992, p. 32;
Knapp, 1988). Salt is another important source of iodine and was often used
as a preservative for meat and fish. Salt occurs naturally on the salt flats (sal-
itrál or pampa) and inter-tidal zones throughout SW Ecuador and coastal El
Oro. Maritime commodities and salt appear to have been exchanged with
highland populations for a variety of resources (both economic and exotic
sumptuary goods) adapted to or derived from higher elevations as well as the
tropical forest in the later periods. The earliest direct evidence of lowland–
highland interaction is in the form of obsidian flakes derived from outcrops
near the valley of Quito (Asaro et al., 1994; Burger et al., 1994). Highland
obsidian has been found in various Valdivia VI–VIII sites in coastal El Oro,
SW coastal Ecuador, and N Manabı́ (Burger et al., 1994; Staller, 1994, 2001;
Zeidler et al., 1994). In the highlands, evidence of exchange appears in the
form of marine shell and ceramic diagnostics with clear coastal affinities
at sites such as Cotocollao in the valley of Quito, and at Cerro Narrio and
Pirincay in the southern near Cuenca (Bruhns et al., 1990; Collier and Murra,
1943; Villalba, 1988).
Certain indigenous species of marine shellfish such as the Thorny Oys-
ter (Spondylus princeps), and the Strombus Conch (Strombus galeatus) were
central to Andean cosmology and religion (Marcos, 1978; Marcos and
Norton, 1981; Paulsen, 1974). Such species were probably exchanged along
with salt as sumptuary items in long-distance interaction between the coast,
highlands, and ceja de montaña. Ethnohistorians and archaeologists (Marcos
and Norton, 1981; Norton, 1981; Salomon, 1977/78, 1986) have documented
the existence of an institution of status traders (mindalá) who carried out
long-distance interaction between the coast and the highlands in the later
periods, although the antiquity this Andean institution has yet to be deter-
mined (see also Norton 1981).
Salomon (1985, Fig. 23.1) introduced a spatial and hierarchical model
of uses of goods and exchange for highland societies involving direct access
to commodities in terms of their material constituents and intrinsic and eco-
nomic value. The framework for analysis involves the extent to which goods
are locally available, or arrive into the center from distant peripheries. The
patterns of exchange are then categorized as an indirect reflection of the de-
gree of complexity of the societies involved in the long-distance interaction.
Commodities from the center included locally available consumables, maize,
wood, etc., at one end of the spectrum to exotic sumptuary goods that con-
note rank and “generosity” usually brought in from different environmental
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 211

zones in distant peripheries. Sumptuary goods from remote regions are gen-
erally associated with the elite and closely involved in religious cults and
the legitimization of social rank. According to Salomon (1977/78, 1985) the
necessity for direct access to such commodities created a vested interest
for extra cordillera linkages with the tropical forest (selva tropical) and the
littoral.

ARCHAEOLOGICAL THEORY AND CULTURE PROCESS IN

COASTAL ECUADOR

Some archaeologists held that the early origins of pottery technology

in coastal Ecuador the result of stimulus diffusion. Ceramic innovation was
initially perceived as spreading from regions of greater social complexity,
such as the Andean highlands, to areas thought to be peripheral sociocultural
complexity, such as the tropical forest and coastal lowlands (Estrada, 1958;
Estrada et al., 1962; Meggers, 1971, 1987; Meggers et al., 1965; Meggers and
Evans, 1966a,b; Uhle, 1922a,b, 1923, 1924).
Ceramic technology has generally been perceived as a marker of seden-
tary village life and an agricultural economy (Ford, 1969; Lathrap, 1970;
Willey, 1971). The great antiquity of Valdivia ceramics therefore conflicted
with the preconceptions of many in the field regarding the necessary precon-
ditions for Formative development, since it was apparently associated with
a nomadic fishing and shellfish collecting culture (Estrada, 1956, 1957; Ford,
1969; Meggers et al., 1965).
Andean archaeologists generally perceived complexity as synonymous
with urbanization. Their focus was usually based upon environmental dif-
ferences seen as the key to understanding sociocultural development and
interpreting culture change. The general consensus had been that only the
wetter, more densely vegetated lowland regions of Esmeraldas Province and
the Gulf of Guayaquil were amenable to intensive agriculture, and there-
fore social stratification and cultural innovation (Evans and Meggers, 1957;
Meggers, 1966; Raymond et al., 1980; Saville, 1909). Early sedentism along
the Ecuadorian littoral was perceived as directly related to the year-round
availability of maritime resources (Meggers, 1966, pp. 20–21, 1971).
After these pioneering studies, a large body of archaeological data was
reported establishing the existence of early sedentary populations associ-
ated with the Valdivia culture (Damp, 1979, pp. 10–11, 13–14, 1984a, p. 106,
1984b, pp. 574–576; Damp and Vargas, 1995, Fig. 13.2; Lathrap et al., 1977,
pp. 6–7; Raymond, 1989, pp. 9–12, 1993; Stahl, 1991; Stahl and Zeidler, 1990;
Zeidler, 1977, 1984, 1987). Archaeologists also recorded direct evidence
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

212 Staller

of the remnants of Valdivia domestic structures and reported in detail on

various related household activities (Damp, 1979, 1984a, 1988; Stahl and
Zeidler, 1990; Zeidler, 1984). A number of archaeological investigations at
sites such as Real Alto, San Isidro, San Lorenzo del Mate, and La
Emerenciana also presented evidence of artificial earthen mounds and long-
term continuity in associated ceremonial and ritual practices (Cruz and
Holm, 1982; Damp, 1979, 1984a, pp. 574–576; Damp and Vargas, 1995,
Fig. 13.2; Lathrap et al., 1977, pp. 6–7; Marcos, 1988a,b; Marcos et al., 1976;
Marcos et al., 1999; Raymond, 1989, pp. 9–12, 1993; Stahl, 1984, 1985, 1986,
1991; Zeidler, 1984, 1987; Staller, 1992/93, 1994; Zevallos-Menendez and
Holm, 1960). Archaeologists digging at Real Alto, recorded a “U” shaped
site plan consisting of domestic habitations surrounding a central plaza
(Lathrap et al., 1977; Marcos, 1988a,b; Marcos et al., 1976). Similar site plans
were later identified in Upper Upano drainage of eastern Ecuador (Salazar,
1998). Linear riverine distributions were reported for Valdivia settlement
patterns in various subregions of the coast (Damp, 1984b; Raymond, 1989;
Stahl, 1981; Zeidler, 1977). These lines of evidence supported the assertion
of an agricultural Valdivia economy based upon the cultivation of maize
(Lathrap et al., 1975, 1977; Marcos et al., 1976; Marcos et al., 1999; Raymond,
1988; Zevallos, 1971; Zevallos et al., 1977). Maize kernels were found to have
been used to create stylistic patterns on Valdivia VI pottery (Lathrap et al.,
1975; Zevallos et al., 1977) and was implied by the numerous milling stones
(manos and metates) found in excavation and on the surfaces of various
Valdivia sites (Lathrap et al., 1977; Norton, 1982). However, such imple-
ments have never been examined as to what kinds of plants were being
processed. Many in the field challenged evidence of a Valdivia subsistence
economy primarily based upon the cultivation of domestic plants. More re-
cent, stable isotope analysis of Valdivia skeletons and ethnobotanic analysis
of carbon residues on Valdivia VII–VIII sherds indicate that the integration
of maize in the subsistence economy was during the final part of the Early
Formative Period and that it was a secondary plant central to rituals of ex-
change (Staller, 2001; Staller et al., 2000; Staller and Thompson, in press;
Thompson and Staller, 2001).

PLANT CULTIVATION IN COASTAL ECUADOR

Later archaeological evidence for the precocious development of plant

domestication, particularly the early introduction of domesticated maize
into South America, were derived from the analysis of plant microfossils
from archaeological soils (Pearsall, 1978, 1979, 1992, 1993, 1994; Pearsall
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 213

and Piperno, 1990; Piperno, 1991, 1994). These data suggested maize was in-
troduced into coastal Ecuador as early as 8000 B.P. and that it was present
in the Ecuadorian Amazon by 6000 B.P. with at least two species exist-
ing on the coast by 5500 B.P. (Bush et al., 1989; Pearsall, 1978, 1979, 1991,
1992, 1993; Piperno, 1991, 1994). Subsequent isotope and ethnobotanical
research has challenged the contextual and methodological evidence for an
early introduction of maize in coastal Ecuador (Staller et al., 2000; Staller
and Thompson, in press). Early Ecuadorian dates for maize have also been
challenged by data from other regions. Although AMS dates (dendrocali-
brated) and morphological analysis of maize cobs from the Coxcatlan levels
at Tehuacan suggest it was fully domesticated before about 3540 ± 40 B.C.
(Benz, 1998; Benz and Iltis, 1990; Benz and Long, 2000, Table 1, p. 461;
Long et al., 1989; Smith, 1998). Rates of change slowed considerably during
the period 4770–1850 B.P. (Benz and Long, 2000, Fig. 1, G-C, C-D). Un-
der conditions of incipient cultivation increased grain number would only
enhance fitness when there was deliberate human intervention (Benz and
Long, 2000). An implication of this evidence is that there was not deliberate
selection for certain varieties in these early time periods. The contextual,
chronological, and methodological issues surrounding the introduction of
maize into South America remain unresolved, but it is generally accepted
(because of the great antiquity of ceramic innovation) that plant cultivation
has relatively ancient origins in coastal Ecuador (Lathrap, 1970; Zevallos,
1971).
The most recent interpretation regarding early plant domestication sug-
gests the Valdivia economy was mixed, based upon plant cultivation, hunting
of terrestrial mammals and birds, and the intense exploitation of maritime
resources (Stahl and Zeidler, 1990; Pearsall and Piperno, 1990; van der
Merwe et al., 1993). Early sedentism appears to be largely related to a
rich, stable maritime resource base, hunting of terrestrial fauna, and some
small-scale cultivation, supporting earlier interpretations of the subsistence
economy (Lanning, 1968, pp. 77–79; Meggers, 1966, pp. 34–37; Meggers et al.,
1965).

THE FORMATIVE OF COASTAL ECUADOR

Most research on the Formative of coastal Ecuador has been focused

upon the Early Formative Period Valdivia complex. Despite the intensity of
research on Valdivia, the chronology and range of material culture diagnostic
of the final portion of the sequence were established only recently (Marcos,
1989; Marcos et al., 1999; Staller, 1992/93, 1994, 2001; Zeidler and Sutliff,
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

214 Staller

1994). Moreover, the developmental relationships of the various Formative

cultures remain unclear, as did their origins. It is the Early Formative Period
settlement patterns and ceramic diagnostics that are the most extensively
documented and the Valdivia chronology is now firmly established.
Although various ceramic seriations have been reported for the Valdivia
Phase (Damp, 1984a; Lanning, 1968; Meggers et al., 1965), the most widely
accepted analysis is the eight-phase sequence developed by Hill (1972/74).
This seriation was based upon the analysis of surface and excavated col-
lections from SW coastal Ecuador (Table I). However, the final portions
of the Valdivia sequence, Phases VI–VIII, have not been well established
in this subregion. The most recent ceramic evidence suggests considerable
regional variability in stylistic and formal attributes during the final portion
of the sequence (Marcos et al., 1999; Staller, 2001; Zeidler and Sutliff, 1994).
Such regional variability continues into the later periods and has important
ramifications for interpreting the developmental relationships of the various
Formative culture complexes and their chronological extent in the different
subregions of the coast.
Victor Emilio Estrada (1956, pp. 9–10, 1958, pp. 7, 12, 21, 24–29) first
identified the Valdivia Phase from investigations at the type-site and cor-
rectly surmised that the ceramic tradition represented the Early Formative
Period. Subsequent 14 C dates supported his conclusions, and but further
revealed Valdivia as one of the earliest ceramic complexes of the pre-
Hispanic New World (Bischof, 1972, pp. 269–273, Fig. A, 1980, p. 335; Bischof
and Viteri, 1972, p. 549; Damp, 1979, pp. 15–26, 89–109, 1984a, p. 106, 1984b,
pp. 573–574; Damp and Vargas, 1995, pp. 157, 159–160, Figs. 13.3, 13.4,
Table 13.1; Estrada, 1956, pp. 9–10, 1958 pp. 21–42; Evans et al., 1959, pp. 7,
15–17; Evans and Meggers, 1962, p. 182; Ford, 1969; Hill, 1972/74, p. 7;
Lanning, 1967, pp. 76–77, 81, 85, 1968, pp. 10–11, 37–42; Lathrap et al., 1975,
p. 16; Lathrap et al., 1977, pp. 6–7; Marcos, 1978, 1988a,b; Marcos et al., 1999;
Meggers, 1966, Fig. 4; Meggers et al., 1965, p. 149; Norton, 1972, 1977, 1982;
Raymond et al., 1980, p. 701; Raymond et al., 1991, p. 9; Stahl, 1984, pp. 190–
226, Figs. 61, 62). An analysis of the Valdivia diagnostics stored at the Museo
Antropológico in Guayaquil by the author indicate that the initial exca-
vations carried out by Estrada at the type-site were primarily taken from
a late (Phase VI–VIII) occupation layer. Estrada (1958, pp. 7, 12, 1962b)
initially considered the diagnostics to be contemporaneous with Initial Pe-
riod cultures at Ancón and Guañape in the Virú valley of coastal Peru (see,
for example, Willey and Corbett, 1954). Most of the subsequent research
on Valdivia has until recently been primarily focused upon determining the
origins of the culture sequence and the earliest pottery diagnostics ( Damp,
1979, 1984a, 1988; Damp and Vargas, 1995; Lathrap et al., 1977; Marcos, 1978;
Meggers et al., 1965; Raymond et al., 1991; Stahl, 1984).
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 215

THE VALDIVIA TRADITION: DIAGNOSTIC FEATURES

THROUGH TIME AND SPACE

The earliest Valdivia pottery was well made, aesthetically pleasing, and
exhibits clear stylistic and formal standardization among distinct vessel shape
categories (Lathrap et al., 1975; Meggers et al., 1965). It was in part due
to these characteristics that Meggers et al. (1965) interpreted the origins
of the ceramic tradition as being derived from outside of the Ecuadorian
coast. Even the so-called “pre-Valdivia” pottery excavated by Bischof and
Viteri (1972) at San Pedro is not technologically simple. Although the San
Pedro assemblage consisted of only 27 sherds, this pottery, like other early
diagnostics, shows no indication of any experimentation, but rather gives the
impression of a fully developed ceramic technology. However, Damp and
Vargas (1994, Fig. 13.4) have identified Phase II diagnostics in excavations at
Real Alto that are identical to the San Pedro diagnostics and they concluded
that these sherds were intrusive.
Phase I pottery from the Valdivia sites of Real Alto and Loma Alta con-
sists of untreated or red slipped open bowls and jars with constricted or con-
cave necks, and fine line incising in geometric motifs (see Damp and Vargas,
1995, Fig. 13.3). Four distinct bowl forms have been identified and they are
differentiated on the basis of the angle of the vessel walls and rims. Fire
clouding on some of the basal sherds of such bowls indicates they were used
in food preparation and do not only function as serving vessels. Perhaps
the most detailed description of the stylistic characteristics of the Valdivia
ceramic tradition was presented by Lathrap et al., (1975, pp. 27–33), who
emphasized a conceptual distinction between open bowls and jars with con-
stricted necks (see also Staller, 1996b). Such forms occur throughout the
entire ceramic sequence but are temporally differentiated by their associ-
ated stylistic attributes.
Phase II is characterized by the presence of large engraved open bowls
with low tetrapod feet, found at various sites in SW coastal Ecuador (see
Lathrap et al., 1975, Fig. 19; Meggers et al., 1965, Plate 103). Other Phase II
diagnostics include cooking pots with piecrust rims and shoulder bosses
and elaborately decorated jars with castellated rims (Meggers et al., 1965,
Plate 103). Phases I and II diagnostics represents a long-term stability in
the technological and stylistic development of the ceramic tradition relative
to the later portion of the ceramic sequence. These early phases comprise
a period of over a thousand years, essentially constituting over half of the
ceramic sequence and are all derived from sites in SW coastal Ecuador.
The middle portion (Phases III–V) of the Valdivia ceramic sequence was
also primarily defined on the basis of diagnostics from SW coastal Ecuador.
There is a proliferation of Valdivia settlements inland in SW coastal Ecuador
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

216 Staller

and also evidence of expansion into other subregions, particularly S Manabı́,

the Gulf of Guayaquil, and coastal El Oro during this portion of the culture
sequence.
Valdivia III–V pottery is characterized by a conspicuous elaboration of
stylistic motifs and a dramatic increase in vessel shape variability in both bowl
and jar forms (see Hill, 1972/74, Plate V and Plate VI, Figs. 40–47; Marcos
et al., 1999, Figs. 2c, and 3a,b). Distinctions include an increase in open bowl
forms, some with cut and beveled as well as undulating rims, and concave
bottoms and the appearance of anthropomorphic motifs (see Lathrap et al.,
1975, p. 28, Fig. 20). Anthropomorphic motifs on Valdivia Phase III–V open
bowls often mimic stylistic patterns found on the facial treatment of the fig-
urines. Some bowl forms are also clear imitations of halves of carved bottle
gourds (Lathrap et al., 1975, p. 29). Stylistic motifs primarily consist of geo-
metric designs. However, the design motifs on bowls were usually made when
the clay was leather hard, while those on jars were made when the clay was
still moist (Staller, 1996b). Color differences are also evident. Open bowls
are generally brown slipped while jars are red-or maroon-slipped (Staller,
1996b). Such technical and stylistic attributes are particularly characteris-
tic of the decorative treatment of ritual and ceremonial bowls and jars and
may have metaphorical reference to the Spondylus/Strombus dyad (Paulsen,
1974; Staller, 1996b). The contrasting use of red slip zones and intentional
smudging also begins to appear during the middle of the Valdivia sequence
and continues with later Chorrera ceramics. Cooking pots (ollas) and storage
vessels of Phase V of the Valdivia sequence introduce tall neck forms with
geometric motifs and externally thickened rims (Marcos et al., 1999, Fig. 3b).
Frequently, both sides of the rim are treated with a red or maroon slip while
the rest of the vessel is the natural color of the fired clay (Lathrap et al., 1975,
Fig. 21). Jars with constricted necks continue from the earlier phases, but the
associated stylistic trend changes to incorporate patterns found on bowls.
During the final portion of the culture sequence Valdivia settlements
expanded inland in the subregions of SW coastal Ecuador, the Gulf of
Guayaquil, coastal El Oro, and S. Manabı́ (Fig. 6). In N Manabı́, there is an
occupational expansion along the riverine environments during Phase VIII.
Ceramic diagnostics pertaining to the final portion of the Valdivia sequence
indicate considerable stylistic and formal variability among subregions.
Valdivia VI–VIII diagnostics indicate an even greater variability in ves-
sel forms and stylistic attributes than the middle portion of the sequence
(Fig. 7). Jars with cambered rims and deeply concave necks and jars with
loop handles appear during Phase VI and continue to Phase VIII over
most of coastal Ecuador. However, after Phase VI striking technological
changes occur, including a proliferation of composite forms, and a shift from
a strictly coiling based technology to forming techniques that include the
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 217

Fig. 6. Distribution of Early (Phases I–III) and Late (Phases IV–VIII) Valdivia Phase sites in
coastal Ecuador.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

218 Staller

Fig. 7. Diagnostic vessel forms of the Valdivia VII–VIII Jelı́ Phase complex identified at La
Emerenciana.

use of paddles and hand modeling. Such technological changes are most
apparent in coastal El Oro and the Gulf of Guayaquil. It is also in these
subregions that the earliest bottle forms appear in association with Valdivia
VII–VIII diagnostics at sites such as La Emerenciana, La Cadena, Milagro
I, and San Lorenzo del Mate (Fig. 8). The three basic shapes of Valdivia
VII–VIII bottles are long neck single spouts, carinated spouts, and stirrup-
spouts (González de Merino, 1984, p. 97; Marcos, 1989, p. 16; Marcos et al.,
1999, Fig. 4; Reindel and Guillaume-Gentil, 1995; Staller, 1994, Figs. 45–53,
2001, Figs. 22–28). Single spout and carinated spouts were identified at La
Cadena, Milagro I and at San Lorenzo del Mate, while all three kinds of bot-
tle forms are present in the Jelı́ Phase complex from La Emerenciana. From
a developmental standpoint, single spouts are diagnostic of the later Chor-
rera and related ceramic complexes, however stirrup-spouts were initially
thought to be solely restricted to the Machalilla tradition (Lathrap et al., 1975,
Fig. 31; Meggers et al., 1965, Plate 155). Phase VII–VIII diagnostics also
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 219

Fig. 8. Diagnostic vessel forms from the Valdivia VII–VIII component at

San Lorenzo del Mate located in the Rı́o Mate Valley, within the Gulf
of Guayaquil subregion and the Phase VII component from Real Alto
located along the Rı́o Verde in SW coastal Ecuador. (a) Phase VII diag-
nostics (b) Phase VIII diagnostics (after Marcos et al., 1999, Fig. 4).

include bowls and jars with externally everted rims and hollow stirrup han-
dles, bowls with vertical walls and flat bottoms, as well as constricted bowls,
bowls with pedestal bases and plates with out-flaring walls (Lathrap et al.,
1975, p. 31, Figs. 38, 43, 44; Staller, 2001, Figs. 16b, 17, 19–21, 27c, 29; Zeidler
and Sutliff, 1994, Fig. 7.1a,b). Pedestal bowls continue to be manufactured
with the later Formative assemblages, as do the externally everted jars and
open bowls with vertical or slightly incurving walls.
Neckless cooking pots (ollas) or tecomate forms also appear in SW
coastal Ecuador, the Gulf regions, and coastal El Oro for the first time during
Valdivia VI–VIII (Marcos, 1989, p. 16; Marcos et al., 1999, Fig. 4; Meggers
et al., 1965, Plate 99c; Staller, 2001, Fig. 18). Such ollas are found with the
Machalilla Phase (Lathrap et al., 1975, Fig. 29) and are diagnostic of early
Initial Period pottery from the North and Central coast of Peru. Phase VII–
VIII stylistic and decorative trends include zoned punctation and incision,
step motifs, shell scraping, engraving, brushing, combing and the earliest
evidence of polychrome decoration on bowls, and shoulder bosses, nubbin
appliques, and appliqué fillet strips on jar forms. All of these attributes ap-
pear during Phase VI and continue to the end of the sequence in coastal
El Oro and the Gulf of Guayaquil. Other distinctive Valdivia VII–VIII
attributes in these subregions include the use of post fired paint, red slipped
bands, and anthropomorphic and zoomorphic effigy vessels. Effigy vessels
are the hallmarks of the later Machalilla and Chorrera ceramic complexes,
and the use of post fired paint and red slip bands are particularly diagnos-
tic of Machalilla. The stylistic and technological attributes that characterize
Valdivia pottery from the subregions of SW coastal Ecuador, coastal El Oro
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

220 Staller

and the Gulf of Guayaquil strongly suggest that the various Formative ce-
ramic traditions are developmentally related.
Valdivia VIII Piquigua Phase (3700–3570 B.P.) pottery from the Rı́o Jama
drainage in N Manabı́ has developmental similarities to late Valdivia pottery
from coastal El Oro and the Gulf of Guayaquil, but is also distinct in various
ways (Zeidler, 1992, 1994, Table 6.2; Zeidler and Sutliff, 1994, Fig. 7.1). It is
clear that the Piquigua Phase complex is directly derived from the earlier,
more well known Valdivia VIII diagnostics from S Manabı́ and SW coastal
Ecuador. Zeidler and Sutliff (1994, pp. 112–114, Fig. 7.1) have identified six
vessel shapes associated with this component (Fig. 9). All but the curvilinear
ollas with nubbin appliques on the shoulders and neck are present in the Jelı́
Phase ceramic complex from coastal El Oro. However, the elaboration of

Fig. 9. Diagnostic vessel forms of the Valdivia VIII Piquigua Phase complex at
San Isidro and related sites of the Rı́o Jama Valley in N Manabı́ (after Zeidler
and Sutliff, 1994, Fig. 7.1).
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 221

decorative techniques and the predominance of motifs consisting of zoned

incision and punctation (especially on bowls) and the apparent absence of
any bottle forms or neckless ollas distinguish the northern tradition. Other
distinguishing features include elaborately incised cooking pots with un-
dulating rims and decorated ceramic lids such as those from San Lorenzo
del Mate (Fig. 8). Stylistic attributes associated with Piquigua Phase pot-
tery include, certain bowl forms, and decorative attributes such as post fired
paint, shell scraping, red banding, and fillet appliques. However, associating
Piquigua Phase diagnostics to the later Tabuchila Phase ceramic complexes
is more problematic than with the Valdivia VI–VIII traditions from the Gulf
of Guayaquil and coastal El Oro. Developmental discontinuity is related
to a presumed 500-year hiatus resulting from a volcanic eruption (Issacson,
1994). Furthermore, Machalilla sites have not as yet been identified in this
subregion.
Valdivia diagnostics pertaining to Phase VI–VIII indicate greater re-
gional variation in formal and stylistic attributes in the subregions (Cruz and
Holm, 1982; Jadán, 1986; Marcos, 1989; Porras-Garces, 1973; Spath, 1980;
Staller, 1994; Zeidler and Sutliff, 1994). When attempting to explain this
some archaeologists have cited environmental differences, culture drift, or
diffusion and migration (Lathrap, 1970; Meggers et al., 1965; Spath, 1980).
On the Punta Arena Peninsula, and at El Encanto on Puná Island Phase
VI–VIII Valdivia sherds were examined by Spath (1980, pp. 71–74, 167)
and interpreted as a regional variant of Valdivia. Technological and stylis-
tic differences from the better-known Valdivia diagnostics of SW coastal
Ecuador and S Manabı́ were initially perceived as related to economic dif-
ferences. Societies adapted to the Gulf of Guayaquil were “more dependent”
upon maritime resources than Valdivia societies in SW coastal Ecuador and
S Manabı́ who were perceived as almost totally dependent upon plant cul-
tivation (Lathrap et al., 1975; Marcos et al., 1976; Marcos et al., 1999; Spath,
1980; Zevallos et al., 1977).
Valdivia occupation in the Gulf of Guayaquil was interpreted as an
example of cultural drift, a regional variant affected in various ways by
independent historical developments (Porras-Garces, 1973; Spath, 1980).
Rather than focus upon sociocultural development, chronology, or culture
process, most archaeologists implicitly assumed that the total range of vari-
ability for the Valdivia tradition had been already documented (Ledergerber,
1983). However, the notion that regional differences in the Phase VI–VIII
pottery assemblages from the Gulf of Guayaquil were a result of greater
maritime resource exploitation was contradicted by faunal evidence, which
shows a continued maritime focus (at least by coastal communities) in SW
coastal Ecuador and S Manabı́ (Byrd, 1976; Lanning, 1968; Meggers et al.,
1965; Staller, 1994). The most recent evidence from faunal analysis, analysis
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

222 Staller

of shellfish, and site distributions in the various subregions, during differ-

ent periods suggest the Valdivia economy was always mixed, involving the
exploitation of a variety of distinct environmental zones, during different
seasons.
In a study of the development of pottery technology in the New World,
Hoopes (1994) maintains that with increasing information regarding specific
regional sequences, the ability to make broad comparisons among pottery
sequences significantly declines. Such regional diversity is already apparent
by the final portion of the Valdivia Phase and unequivocal by Late Forma-
tive times (Beckwith, 1996). The most recent analyses of Valdivia VI–VIII
pottery from outside SW coastal Ecuador indicate that ceramic innovation
was influencing and influenced by a variety of diverse historical and cultural
processes beyond the Ecuadorian coast (Staller, 1994, pp. 391–399). Some
time between 4200 and 3650 B.P. ceramic innovation spread throughout the
lowland and highland regions of western South America. It is also at this
time that the first clear evidence of complexity appears in the archaeological
record of the Andes.
Most specialists on the Ecuadorian Formative recognize a significant
technological and stylistic divergence in Valdivia ceramics during and after
Phase III (Lathrap et al., 1977; Marcos and Manrique, 1988; Staller, 1994;
Zeidler, 1984) featuring increased variability in vessel shape and in the
degree of elaboration of decoration and stylistic attributes. In southern El
Oro and the Gulf of Guayaquil there is also a dramatic proliferation of
composite forms after Phase VI characterized by a shift from a coil based
ceramic technology to a variety of different forming techniques during the
final portion of the ceramic sequence throughout coastal Ecuador (Staller,
1994, 2001). However, such technological changes were not as apparent
in the Phase VII–VIII diagnostics from SW coastal Ecuador and N and
S Manabı́. The differences between Phase VII–VIII diagnostics from differ-
ent subregions in part explain why a developmental transition between Val-
divia and the later Formative ceramic traditions was so elusive for so many
years.
Another important factor influencing the regional variability evident
during Valdivia VI–VIII is related to changes in the subsistence adaptation.
Geographic expansion by Valdivia populations into the Gulf of Guayaquil
and coastal El Oro after Phase III, and into N Manabı́ after Phase VII im-
plies a greater focus upon plant cultivation by riverine communities located
inland. These subregions have a higher average annual precipitation than
SW coastal Ecuador and are generally more amenable to agriculture. An-
other important factor affecting the regional variability and is related to
long-distance interaction and the development of more complex forms of
social organization (Staller, 1992/93).
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 223

The early settlement surveys in SW coastal Ecuador and southern

Manabı́ indicated that Valdivia VI–VIII sites were fewer and smaller than
sites with earlier occupations (Damp, 1984a; Estrada, 1956; Lanning, 1968;
Lathrap et al., 1977; Schwarz and Raymond, 1996). For example, Meggers
et al. (1965) emphasized pottery diagnostics from late Valdivia sites in the
Punta Arena Peninsula to define their Period D. They explained these occu-
pations as a result Valdivia migrations from SW coastal Ecuador into the Gulf
of Guayaquil. Later, Columbia University surveys in SW coastal Ecuador un-
covered similar chronological patterns in Valdivia settlement and indicated
that Phase VI–VIII sites were rare, and restricted to small coastal localities
(Lanning, 1968). Although Lanning (1968) isolated Phase VI diagnostics he
found them to be rare in the Santa Elena Peninsula. Later investigations in
this subregion (Damp, 1984a; Damp et al., 1990; Ferdon, 1981; Schwarz and
Raymond, 1996) imply that the rarity of post Phase V occupations may be
reflecting a reduction of mangrove habitats due to either ENSO, geomor-
phological change and/or human overexploitation (Staller, 1994). Today, the
mangrove formation that once existed in SW coastal Ecuador has all but
disappeared. The apparent ancient degradation of this habitat supports the
premise that coastal populations continued to exploit maritime resources
even after inland populations had shifted to a mixed economy partly based
upon plant cultivation.

GEOGRAPHIC DISTRIBUTIONS AND SUBREGIONAL

CHRONOLOGIES

The Valdivia Phase was originally divided into four subphases desig-
nated from early to late as Periods A–D (Meggers et al., 1965). Subsequently
a number of other chronologies were presented (Table V). The most recent
research indicates that these various chronologies may be reflecting dif-
ferences in the intensity of archaeological research as well as pre-Hispanic
occupation in the various subregions (Staller, 1994). On the basis of regional
survey and laboratory analysis of artifact collections of Formative material
culture from SW coastal Ecuador, S Manabı́, and the Gulf of Guayaquil,
Lanning (1968) proposed a preliminary ceramic sequence of nine phases.
Lanning (1960, 1963, 1968) had a regional perspective unique among Andean
scholars of his generation. He was the only North American archaeolo-
gist to study the early Formative pottery of coastal Ecuador, as well as the
earliest Initial Period ceramics from far north and the North and Central
Coasts of Peru. On the basis of this analytical background, he correctly
surmised that Valdivia pottery technology was a precocious coastal phe-
nomenon that evolved in situ and, furthermore, despite an absence of direct
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

224 Staller

Table V. Valdivia and Machalilla Ceramic Chronologies

evidence, that the Machalilla tradition developed out of Valdivia (Lanning,

1968).
Later, Hill (1972/74) reanalyzed the Columbia University collections
and presented a revised eight phase (Phase I–VIII) ceramic seriation. This
analysis was based upon stylistic attributes, vessel forms, themes, and decora-
tive patterns (see Rowe, 1960, 1961). Hill found a discontinuous frequency
curve between fillet appliqué and brushed attributes when attempting to
merge Period C diagnostics from Buena Vista, with those excavated at the
Valdivia type-site (Hill, 1972/74, pp. 19, 24). She attributed the discontinu-
ity to mixed layers in the Cut 1 excavations at the site of Buena Vista by
Meggers et al. (1965). The Cut 1 excavation was in arbitrary increments on
a sloping river terrace at the base of a steep hill (Meggers et al., 1965, p. 18).
Hill (1972/74, pp. 19–20, 25) further contended that inverted stratigraphy or
secondary deposits indicated that late Period C diagnostics were in fact early
Period C or Phase VI.
The preference for the eight-phase sequence over the Period A–D se-
riation is related in part to the recognition that Period C at Buena Vista
actually corresponds to Phase VI (Fig. 10). On the other hand, weaknesses
with the Hill seriation, include small sample sizes, and an absence of strati-
graphic evidence and dates to establish phase distinctions due to shallow,
disturbed archaeological deposits, which Hill (1972/74, p. 15) mentions in
the published report. Phase VIII diagnostics were solely based upon surface
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 225

Fig. 10. SW and S Coastal Ecuador: Showing the approximate locations of Formative sites
mentioned in the text.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

226 Staller

finds from OGSE-46B a small coastal locality a few meters from the sea in
La Libertad on the Santa Elena Peninsula (see also Ferdon, 1940a,b, 1941b,
1942). The middle portion of the sequence (Phases III–V) was based upon
surface finds from four disturbed middens west of La Libertad (Hill, 1972/74,
pp. 20–21). Phase divisions were arbitrary since reliable temporal differences
were stylistically based primarily from surface collections (Hill, 1972/74,
pp. 13, 15).
A series of 21 uncalibrated 14 C dates ranging between 4950 and 2950
B.P. initially established the Valdivia cultural chronology (Evans et al., 1959;
Meggers, 1966, pp. 34–42; Meggers et al., 1965, pp. 149, 151). Columbia Uni-
versity excavations at two sites (G-42A and G-172) reported dates ranging
from 4750 to 3250 B.P. (Lanning, 1967, p. 85; Willey, 1971, p. 270). However,
later dates from Loma Alta placed the beginning of the culture sequence to
between 5450 and 5250 B.P. (Damp, 1979, 1984a; Norton, 1972, 1977, 1982;
Stahl, 1984). Early estimates for the end of the ceramic sequence were based
upon several Machalilla 14 C dates, as well as some Chorrera dates (Meggers,
1966; Meggers et al., 1965).
Phase VII and VIII of the Valdivia culture sequence was tentatively
postulated by Hill from two corrected dates from site OGSE-46B, a 2870–
1940 B.C. [L-1232H; 3900 B.P. ± 150 years] date, and another 2570–1740 B.C.
[L-1232I; 3750 B.P. ± 150 years] date associated with Phase VII pottery (Hill,
1972/74, p. 21). These dates were rather imprecise, suggesting the culture
sequence ended sometime after 4250 B.P. (Hill, 1972/74). A shell sample
from Buena Vista with a date of 1890–1620 B.C. [SI-69; 3450 B.P. ± 50 years]
still placed the end date at ca. 3650 B.P. (Meggers et al., 1965, p. 149). Despite a
paucity of 14 C dates for the end of the culture sequence, most archaeologists
have contended Machalilla was later than and probably a developmental
outgrowth of the Valdivia tradition (Bischof, 1967, p. 217, 1975; Lanning,
1967, p. 9, 1968, pp. 39, 41–42; Lathrap, 1971, pp. 84–85; Lathrap et al., 1975,
pp. 16, 33–34; Lathrap et al., 1977, p. 6; Pearsall, 1979, p. 6). The dividing line
between Valdivia and Machalilla Phase cultures was initially estimated at
ca. 3450 ± 200 B.P., with Machalilla terminating at ca. 2950 B.P. (Willey, 1971,
p. 353).
Geographic, cultural and developmental factors explain the difficulty
in documenting the end of the Valdivia culture sequence. Most of the early
research was confined to the SW coast where settlement surveys and excava-
tions recorded the presence of the earliest Valdivia occupations (see Fig. 6).
It is now possible to see the relative scarcity of late Valdivia occupation in SW
coastal Ecuador as related in part to a reduction of the mangrove formation,
and to settlement expansion after Phase III to the N, S, and SE, and finally
into N Manabı́ (Staller, 1994, 2000a). Changes in the geographic distribution
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 227

of Valdivia occupation is suggested to be related to an increased dependence

upon plant cultivation by inland populations especially after Phase VI, when
maize appears in the various subregions (Staller and Thompson, in press).
The 14 C measurements of samples from the site of San Isidro in
N Manabı́ Province give an end date for the Valdivia Phase of between
3500 and 3630 B.P. (Marcos, 1988a, p. 79; Zeidler, 1994, p. 107). These dates
are consistent with those reported from Valdivia VIII contexts at site of
Ayalán and the Anllulla shell mound in the Gulf of Guayaquil (Lubensky,
1980, 1995). Similar results were also derived from 14 C assays and AMS
dates from Phase VII–VIII layers at La Emerenciana in coastal El Oro. The
calibrated 14 C measurements and AMS dates from La Emerenciana suggest
an end date of ca. 1450 B.C. for the Valdivia culture sequence (Staller, 2001,
Tables 1 and 2; Staller and Thompson, in press). All of these calibrated 14 C
measurements taken together suggest an end date ranging between 1650–
1450 B.C. (Table VI).

Table VI. Radiocarbon Dates of Valdivia VII–VIII Occupations From Various

Subregions of Coastal Ecuador
14 C
No. laboratory Uncalibrated (B.P.) Calibrated
(material) provenience 14 C Age (years) 1–δ Age B.C.
14 C No. SMU-2225 (charcoal) 3707 ± 148 2288 ± 2245
(La Emerenciana)
14 C No. SMU-2226 (charcoal) 3400 ± 220 1941 ± 1428
(La Emerenciana)
14 C No. SMU-2241 (charcoal) 3361 ± 246 1935 ± 1323
(La Emerenciana)
14 C No. SMU-2563 (charcoal) 3775 ± 165 2459 ± 1922
(La Emerenciana)
14 C No. N-2908 3665 ± 95 2148 ± 1920
(charcoal) (Ayalán)
14 C No. N-2909 3630 ± 105 2115 ± 1871
(charcoal) (Ayalán)
14 C No. ISGS-1220 3500 ± 70 1869 ± 1754
(charcoal) (San Isidro)
14 C No. ISGS-1221 3630 ± 70 2093 ± 1893
(charcoal) (San Isidro)
14 C No. ISGS-1222 3520 ± 70 2075 ± 1875
(charcoal) (San Isidro)
14 C No. ISGS-1223 3560 ± 70 1949 ± 1837
(charcoal) (San Isidro)
Note. All SMU dates are standard assays and are calibrated using Calib 4.1.2 (Stuiver
et al., 1998) with a minus 24-years Southern Hemisphere atmosphere sample adjust-
ment and are at the one sigma range. Staller (1994, Table 15, pp. 393–394, 396) provides
more specific information on the 14 C that is adjusted for a 5568 year half-life prior to
calibration (after Macros, 1988a, p. 79 and Staller, 2001, Table 1).
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

228 Staller

VALDIVIA–MACHALILLA: A TRANSITION OR A TRADITION?

Most of the initial hypotheses regarding sociocultural development of

the Formative cultures of coastal Ecuador were ensconced in a
cultural-historical diffusionary framework. Technological innovations were
assumed to be based upon a Formative substrate reflected by the degree of
stylistic and formal similarities among the various Formative ceramic com-
plexes (Hoopes, 1994). Such reasoning led to many differing interpretations.
For example, Lathrap (1970, p. 67) suggested that the Machalilla culture
was of tropical origin, citing stylistic affinities between Machalilla and late
Tutishcainyo and Sanidine pottery in the Ucayali valley, and sherds from
the earliest levels at the southern highland site of Cerro Narrı́o (Collier
and Murra, 1943, Plates 16–17). However, the stylistic trends and distinct
vessel forms that differentiated Machalilla from Valdivia, such as red slip
bands, engraving, post fired paint, ceramic bottle forms, and a proliferation
of composite forms already begin to appear in the Valdivia VI–VIII diag-
nostics from the Gulf of Guayaquil and coastal El Oro subregions. In other
publications, Lathrap has stated that Valdivia and Machalilla were both a
result of a diffusion of floodplain farmers from the tropical forests east of
the Andes (Lathrap, 1971, 1973, 1974). Linking the Valdivia and Machalilla
complexes to the early ceramic cultures of the tropical forests was at the time
a provocative alternative to traditional ways of thinking about diffusion and
migrations, but little was known about the antiquity of pottery bearing cul-
tures in eastern Ecuador.
Recent research has indicated that Machalilla related sites are solely re-
stricted to the subregions of S Manabı́, SW coastal Ecuador, and the Punta
Arenas Peninsula in the Gulf of Guayaquil, and coastal El Oro (Fig. 11). It
is precisely in these subregions that Meggers et al., (1965) concentrated their
research, and it has been assumed that Machalilla and Valdivia related sites
were similarly distributed. The research has now indicated that the chronol-
ogy and distribution of Machalilla related sites are much more complex than
had been thought. Moreover, there is strong stylistic evidence that the early
pottery at Cerro Narrio and Cotocollao in the Ecuadorian Andes has strong
affinities to Machalilla (Lathrap, 1970, 1973; Lathrap et al., 1975; Villalba,
1988).
The Machalilla Phase was initially identified by Bushnell (1951,
pp. 17–21) from excavations at two sites near La Libertad on the Santa Elena
Peninsula. Machalilla ceramic diagnostics were later described by Estrada
(1958, pp. 13, 53) from excavations at the type-site (M-24). Numerous
Machalilla shell middens have been identified in SW coastal Ecuador and
S Manabı́ Province (Estrada, 1958; Meggers et al., 1965, Fig. 2). He (Estrada,
1958, pp. 55, 94) assigned Machalilla an intermediate chronological position
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 229

Fig. 11. Distribution of Machalilla Phase sites in coastal Ecuador.

P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

230 Staller

between Valdivia and Chorrera, asserting that red banded and engraved
sherds represented a single ceramic complex. Estrada (1958, pp. 55, 111) ini-
tially interpreted Machalilla as a developmental outgrowth of the Valdivia
tradition, and later speculated that the corrugated wares were stylistically
similar to those from the mouth of the Rı́o Amazon. Machalilla was later
seen as a result of the diffusion of pre-Hispanic societies from Mesoamerica
into coastal Ecuador at approximately the same time that Valdivia mate-
rial culture was disappearing from the region (Estrada et al., 1962, p. 174).
Therefore, Machalilla was perceived as having Mesoamerican origins, and
several archaeologists took an identical position for the succeeding Chor-
rera Phase (Bushnell, 1982; Coe, 1960, pp. 368–369; Evans and Meggers,
1957, p. 243, 1966, pp. 203–204; 1982, p. 125; Lathrap, 1960, p. 126; Meggers,
1972, pp. 74–75).
The most recent evidence from coastal El Oro and the Gulf of Guayaquil
indicates Machalilla and Chorrera developed out of the Valdivia tradition.
Such a developmental scenario is also consistent with evidence from coastal
El Oro and the Gulf of Guayaquil, but is more problematic in N Manabı́,
where Machalilla is apparently absent and there is a hiatus between early
Chorrera and Valdivia VIII. Although most Ecuadorian specialists now per-
ceive Valdivia and Machalilla as part of the same ceramic tradition, a devel-
opmental transition has yet to be identified in SW coastal Ecuador.
A transition was elusive because early Machalilla ceramic diagnostics
were absent at the type-site and at La Cabuya (G-159) (Meggers and Evans,
1962). Meggers et al. (1965) defined the Machalilla tradition on the basis of
type frequencies, dividing the sequence into three subphases A to C, ranging
between 3950 to 3000 B.P. (Table VI).
A 14 C assay [W-630; 4050 B.P. ± 200 years] taken from late Valdivia lev-
els thought to contain Machalilla trade sherds indicated a date for the pre-
sumed arrival of Machalilla migrants into coastal Ecuador (Meggers et al.,
1965, p. 172; Meggers and Evans, 1962, p. 191). The presence of presumed
Machalilla “trade” pottery in late Valdivia stratigraphic layers at both La
Cabuya and Buena Vista led to the conclusion that the two earliest Machalilla
subphases (A and B) were coeval with Valdivia C and D (Meggers et al., 1965,
p. 87). Meggers et al. (1965, pp. 173–178) further suggested that Machalilla
was coeval with and chronologically later than Valdivia, a result of a cultural
diffusion from an undetermined region beyond the coast (see also Meggers,
1966, pp. 47–51). Despite a presumed overlap of some 600–700 years, stylistic
and technological similarities were surprisingly absent, and Machalilla cul-
ture was interpreted as completely distinct from Valdivia (Estrada, 1962a,
p. 64; Meggers, 1966, pp. 47–51, 62; Meggers et al., 1965, pp. 18–21, 110–146;
Meggers and Evans, 1962, p. 191). Stirrup-spout fragments and sherds with
red slip bands from the surface and uppermost layers of the Valdivia type-site
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 231

also supported their hypotheses regarding the origins and developmental re-
lationships of these culture phases (Estrada, 1958, Cuadro 2; Meggers et al.,
1965, pp. 110–146). Machalilla sherds in the uppermost levels of Period D
Valdivia occupation layers were considered to represent the early portion
of the ceramic sequence (Estrada, 1958; Hill, 1972/74; Meggers et al., 1965).
The chronological range for Machalilla was therefore initially established
by 14 C measurements from samples taken from Valdivia C and D deposits.
The presumed early Machalilla diagnostics were likewise sherds taken from
late Valdivia occupation layers. On the basis of the ceramic evidence from
coastal El Oro and the Gulf of Guayaquil, it is now apparent that what were
initially assumed to be early Machalilla diagnostics could just as well have
been late Valdivia.
Many archaeologists challenged the assertion that Valdivia and
Machalilla were coeval, maintaining instead that the presumed overlap was
rather a result of mixing or inverted stratigraphy (Bischof, 1967, p. 219, 1975,
p. 50; Collier, 1968, p. 271; Hill, 1972/74, p. 19; Lanning, 1968, p. 47; Lathrap,
1967, p. 98; Paulsen and McDougle, 1974, pp. 4–5). Some cited the excava-
tions at Buena Vista, which was dug on a sloping river terrace at the base of
a hill, and pointed out that only 21 of the 9800 sherds recovered from below
80 cm were Machalilla diagnostics (Hill, 1972/74, pp. 19–20; Lathrap, 1967,
p. 98). Rather than representing an intrusion, the evidence could also infer
that Machalilla succeeded Valdivia chronologically (Bischof, 1967, p. 217;
Hill, 1972/74, p. 19; Lanning, 1967, p. 9, 1968, pp. 47–50; Lathrap, 1967, p. 98,
1971, pp. 84–85; Lathrap et al., 1975, p. 33). Moreover, 14 C measurements at
La Cabuya are inverted, also suggesting the possibility of mixing in those ex-
cavations (Meggers et al., 1965, pp. 149–152). Furthermore, Machalilla layers
at many sites in SW coastal Ecuador were unusually thin, ranging between
only 5 and 25 cm in thickness and they were often separated from underlying
Valdivia deposits by a thin sterile layer (Lippi, 1983). The presumed mixing
of Machalilla and Valdivia layers was attributed to these typically shallow
occupation layers (Estrada, 1958, p. 55; Lanning, 1968). Early estimates for
the end of the Valdivia sequence were also indirect, based upon a series 14 C
measurements of samples taken from Machalilla layers at the sites of La
Libertad and La Cabuya (Meggers et al., 1965; Meggers, 1966). A dividing
line was arbitrarily placed at 3450 ± 200 years, with Machalilla terminating
at ca. 2950 B.P. (Meggers et al., 1965, p. 149; Willey, 1971, p. 353).
Some scholars were concerned with the possibility of a developmental
link between Valdivia and Machalilla. One of the first and perhaps most
successful tests of the hypothesis was by Bischof (1975, p. 51). The focus
of his investigations at Palmar 3 (G-88) a locality near Valdivia (G-31) was
defining early Machalilla diagnostics. However, this goal was complicated by
a stratigraphic separation between the Valdivia and Machalilla occupation
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

232 Staller

layers at that site. The total absence of early Machalilla diagnostics made the
results of this research inconclusive (Bischof, 1975, pp. 54–55). Using ceramic
diagnostics and radiocarbon dates, he divided Machalilla into five 150–200
year periods, with an initial date at 3750–3550 B.P. for Phase II (Bischof, 1975,
p. 51). A hypothetical transition consisting of stylistic attributes such as fil-
let appliqué, brushing, red slip banding, and shell scraping was presented.
Significantly, the Jelı́ Phase diagnostic attributes and 14 C measurements from
La Emerenciana are to a large degree consistent with such a transition
(Staller, 2001), and the single and stirrup-spout bottles, carinated jars, and
constricted bowls identified with the Jelı́ Phase complex appear to provide
a developmental link between these ceramic components (Staller, 1994,
2001).
Paulsen and McDougle (1974, 1981) also reported a stratigraphic sepa-
ration between late Valdivia and early Machalilla at two sites in SW coastal
Ecuador. On the basis of their excavations they subdivided Machalilla into
five subphases (Table VI) ranging between 3250 and 2850 B.P. (Paulsen and
McDougle, 1974, pp. 7–14). Supporting Lanning (1968) and others (Bischof
and Viteri, 1972, p. 549; Lathrap, 1967, p. 97) Paulsen and McDougle (1981)
contended they were not coeval, but rather chronologically distinct cultures.
Archaeological investigations at two sites in SW coastal Ecuador by
Lippi (1982, 1983) produced radiocarbon measurements ranging between
3150 and 2750 B.P. However, the excavations at Rı́o Perdido and La Ponga
also recorded a stratigraphic separation between Machalilla and Valdivia
layers (Fig. 10). On the basis of this inconclusive evidence Lippi (1983,
p. 39) concluded that Machalilla was an in situ development representing
the Middle Formative Period, and presented a cultural sequence ranging
between 3350 and 2750 B.P. for SW coastal Ecuador. In the context of this
research he introduced a revised pottery seriation subdividing Machalilla
ceramics into eight phases averaging 50 years each (Lippi, 1983, p. 354).
The difficulty in documenting a developmental transition between the
Valdivia and Machalilla ceramic complexes was related to a number of fac-
tors: (1) One of the critical factors was where the early research was concen-
trated. There appears to have been a clear disjunction between late Valdivia–
early Machalilla site locations in S Manabı́ and SW coastal Ecuador and most
all of the early research on the Ecuadorian Formative was restricted to these
subregions. Changes in late Valdivia and early Machalilla settlement patterns
in these subregions appear to coincide with the disappearance of Valdivia
diagnostics and predominance Machalilla material culture particularly along
the S Manabı́ coast. Stratigraphic and faunal evidence from La Emerenciana
suggest that such widespread changes in site location may be related in part
to climatic and environmental factors, perhaps to ENSO. Stratigraphically,
the uppermost layer at La Emerenciana represents a sudden transgressive
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 233

event in what is an overall progradational stratigraphic sequence. A dramatic

reduction in the distribution and size of various marine shell species associ-
ated with this uppermost layer infer climatic and environmental changes to a
calibrated date of approximately 1450 B.C. (Staller, 2001, Table 1; Staller and
Thompson, in press, Tables 9A and 9B). (2) Changes in coastal geomorphol-
ogy may be another primary factor affecting the changes in site location.
SW coastal Ecuador and S Manabı́ have undergone significant geological
changes since the beginning of the Holocene (Damp et al., 1990), related to
periods of intense flooding, short-term changes in sea levels brought on by
ENSO cycles, as well as tectonic uplift and subsidence. Such changes could
have a profound effect on the hydrology and consequently on pre-Hispanic
site location. It is apparent from the Machalilla occupational debris found on
summits of steep cliffs along S Manabı́ Province that such geomorphological
changes coincided with this time period. Such geological changes may be in
part responsible for the disappearance of the mangrove formations (Damp
et al., 1990; Ferdon, 1981; Staller, 1994; Willey, 1971, p. 247). (3) There was
a gradual, but nevertheless widespread adaptive change to a greater depen-
dence upon agriculture throughout the coast between 4150 and 3500 B.P.
(4) Finally, the prevailing theoretical assumption that culture change and
the spread and early development of technological innovation was directly
related to migrations, diffusions, and invasions. Another assumption was
that the spread of technological innovation was from regions of greater so-
ciocultural complexity and richer more highly productive environments to
areas with more limited carrying capacity and less complex forms of social
organization (Hoopes, 1994).
Archaeological research in the subregions of coastal El Oro and the
Gulf of Guayaquil has greatly revised our perceptions regarding ceramic
development during the final epochs of the Valdivia tradition, particularly
the investigations by Cruz and Holm (1982) at the Valdivia VII–VIII cere-
monial center at San Lorenzo del Mate. This important site is on an ancient
beach terrace on the eastern banks of the Rı́o Mate (Cruz and Holm, 1982;
Marcos, 1989; Staller, 1998, 2000b). An artificial earthen mound measuring
8 m in height and 60 m in diameter called attention to the late Valdivia
component and the Phase VII–VIII diagnostics have affinities to the Jelı́
Phase at La Emerenciana (Marcos 1989; Marcos et al., 1999; Staller 1998,
2000b). When these are compared to the ceramic diagnostics from San Isidro
it is obvious that the final portion of the ceramic sequence represents a dra-
matic technological and stylistic departure from diagnostics pertaining to
the earlier part of the sequence. A significant developmental trend is the
appearance of composite forms including a variety of ceramic bottles and a
shift from a strictly coil based technology to a variety of forming techniques.
Such technological changes also appear in the Valdivia anthropomorphic
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

234 Staller

figurines (Evans et al., 1959; Lathrap et al., 1975; Marcos and Manrique, 1988;
Meggers et al., 1965). Valdivia figurines were initially manufactured using a
double coiling technique, while the torsos and heads of Valdivia VII–VIII Fe-
line Style figurines at San Lorenzo del Mate are made of composite parts and
like later Machalilla figurines have eyes made from clay coffee bean appliqué
pellets (Staller, 1998, 2000b, Figs. 8–10). Such technological characteristics
and stylistic attributes indicate a clear developmental link to later Machalilla
and Chorrera anthropomorphic figurines, further supporting an in situ cul-
tural development for the Ecuadorian Formative (Staller, 1998, 2000b).

THE LATE FORMATIVE PERIOD

Chorrera Phase represents the Late Formative Period of coastal

Ecuador. Many stylistic and technological elements found in Chorrera and
Chorrera related ceramics have their basis in the earlier Machalilla and
Valdivia Phases (Bischof, 1982; Evans and Meggers, 1982, p. 122; Lathrap
et al., 1975, p. 34). Most scholars perceive Chorrera as the artistic pinnacle
of pre-Columbian Ecuadorian ceramic art (Estrada, 1958; Lathrap et al.,
1975; Meggers, 1966). The tradition incorporates a wide variety of sophisti-
cated and finely crafted bowls, whistling bottles, and elaborately constructed
zoomorphic and anthropomorphic effigy forms (Lathrap et al., 1975, p. 34).
Because of the aesthetics of the pottery and figurines much of what is known
about Chorrera Phase comes from unprovenienced private and museum col-
lections. Relatively little is known regarding regional settlement patterns,
subsistence, or sociocultural development (Beckwith, 1996, p. 1; Engwall,
1995; Weinstein, 1998). This also led scholars of pre-Columbian art to form
distinct opinions with regard to the appearance of Chorrera material culture.
More recent research has shown that Chorrera Phase, as currently defined,
is much more variable in terms of certain vessel forms and associated deco-
rative motifs (Beckwith, 1996). Such regional differences in Late Formative
pottery may be seen as a direct outgrowth of the regional variability initially
manifest during Valdivia VII–VIII times, indicating the developmental in-
terrelationships of the Formative Period cultures of coastal Ecuador are
extremely complex and highly regional in nature. These regional differences
are primarily present in the utilitarian vessels, cooking pots and particularly
bowls (Beckwith, 1996). It is however the stylistic and symbolic similari-
ties in the whistling bottles, effigy vessels, and anthropomorphic figurines
that appears to unite the Late Formative regional phases ideologically and
stylistically.
Bushnell (1951) initially reported on the Late Formative Period from
excavations at the Engoroy cemetery in the Santa Elena Peninsula, and
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 235

classified the ceramic complex as the Engoroy Phase (see, also, Zevallos,
1965/66). Later, Evans and Meggers (1957) defined the Chorrera Phase from
excavations at the Hacienda La Chorrera located along the Babahoya River.
Estrada (1957, 1958, 1962a) carried out extensive research on Late Formative
Period sites in various subregions classifying the ceramic complexes in the N
and S Manabı́ coasts, as well as on the western cordillera (yumbos) near Santo
Domingo de los Colorados as, Olón, Véliz, and Tabuchila. On the basis of this
research, Estrada (1958, p. 69) concluded that regional ceramic complexes
such as Engoroy, Olón, and Tabuchila shared broad stylistic characteristics,
which were subsumed under the single rubric of Chorrera Phase (also see
Meggers, 1966, pp. 55–61; Evans and Meggers, 1954, 1957, 1982, p. 124).
Although this was meant to be a heuristic device, it has become reified over
time into a concept of a Chorrera style (Cummins, 1992, pp. 67–78; Engwall,
1995).

ORIGINS AND DISTRIBUTION OF THE CHORRERA PHASE

Initially, Estrada (1958) perceived Chorrera as the quintessential

Ecuadorian pottery complex but later, along with Evans and Meggers in-
terpreted Chorrera culture as a migration from Mesoamerica on the basis
of stylistic similarities to Ochos Phase pottery from coastal Guatemala (see
Coe and Flannery, 1967). Subsequently, archaeological investigations on the
Chorrera Phase have focused upon the recovery of artifacts from single sites
throughout coastal Ecuador and the Guayas Basin. Principal contributions
have consisted of descriptive studies of Late Formative Period material cul-
ture and the establishment of local chronologies (Aleto, 1988; Beckwith,
1996; Bischof, 1975; Cummins, 1992; Engwall, 1992, 1995; Marcos, 1982;
Parducci and Parducci, 1975; Paulsen and McDougle, 1981; Simmons, 1970;
Zedeño, 1985, 1987; Zeidler and Sutliff, 1994; Zevallos, 1965/66).
Late Formative Period sites classified as the Tachina Phase have also
been identified in the north in coastal Esmeraldas Province (López y
Sebastián and Caillavet, 1979). However, inland sites have yet to be identi-
fied in that region (Engwall, 1996). With the exception of the recent research
in northern Manabı́, the relative paucity of systematic large-scale research
on the Late Formative Period is surprising given the elaborate Chorrera
material culture and the fact that it is the most widespread complex in pre-
Hispanic Ecuador (Cummins, 1992; Engwall, 1995; Estrada, 1958). Véliz
Mendoza (1990, p. 31) and Staller (2000a) indicate that Late Formative oc-
cupations are distributed throughout the river drainages in the Guayas Basin
extending south to the Peruvian border (Fig. 12). In Far North Peru, Izumi
and Terada (1966) have identified the Pechiche Phase ceramic complex,
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

236 Staller

Fig. 12. Distribution of Chorrera Phase sites and some of the related ceramic complexes
in coastal Ecuador.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 237

which has clear stylistic affinities to complexes on the Ecuadorian side of

the frontier of the same time period (Staller, 2000a). Although the Late For-
mative Period Arenillas Phase, and Jubones Phase are undefined in terms
of stylistic and technological attributes, these finely crafted ceramics have
similarities to what has been reported from La Chorrera (Fig. 10) and may
be seen as regional expression of a Chorrera style (Staller, 1994, 2000a).
The early diagnostics from Cerro Narrio and the nearby site of Pirincay also
have clear affinities to late Valdivia–early Machalilla and Chorrera related
assemblages on the coast (Bruhns et al., 1990; Collier and Murra, 1943).
When archeological investigations were being carried out at Cerro Narrio
the antiquity of the early occupations at that site was unknown (Collier
and Murra, 1943, pp. 11–15). More recent comparative analysis by the au-
thor of the Cerro Narrio collections has indicated that ceramic innovation
in highland Cañar and Azuay Province shows stylistic and technological
influence from the Valdivia VII–VIII Jelı́ Phase complex from coastal El
Oro and that such stylistic affinities continued throughout the Late For-
mative Period. In the valley of Quito to the north, Formative pottery at
Cotocollao also indicates strong stylistic influences with coastal assemblages
from various subregions (see Villalba, 1988, p. 241, Figs. 92–93, 95, 114–
115, 134–135). The Chorrera style, thus, is a widespread phenomenon and
that it may be properly defined as a horizon (Meggers, 1966, p. 65; Willey,
1962).
Having established Chorrera as a widespread entity (Estrada, 1957,
1958, 1962a; Evans and Meggers, 1954; Meggers, 1966), Meggers et al. (1965)
explained the dispersal in terms of a bifurcated wave: one extending along
the coast and another wave extending inland along rivers with rich alluvial
soils and ample precipitation. Chorrera migration was assumed to have been
rapid, constituting a horizon style (also see Meggers, 1966, p. 65). Another
inference drawn for a “horizon” is that the spread of material culture is in
some way related to religious or cult activity which is culturally integrated
by disparate populations (Burger, 1988; Cummins, 1992; Willey, 1962). Chor-
rera style is a horizon to the extent that the ritual and ceremonial vessels
and figurines appear to have considerable symbolic homogeneity among
different regional phases, a pattern consistent with an interregional integra-
tion of a cult or set of religious beliefs (Cummins, 1992, pp. 15–18). Signifi-
cantly, many Chorrera effigy vessels and polished stone mortars suggest the
consumption of a variety of psychotropic plants consistent with shamanic
practices (Cummins, 1992, Illustrations 33, 35, 37, 46, pp. 68–73; Weinstein,
1998). However, it is at this point premature to make any general state-
ments about cosmology before a regional synthesis of the ceramic style has
been published. If Chorrera is a horizon style, then it begs the question of
what makes up its material constituents. Bowls with annular bases, and thin,
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

238 Staller

highly polished, red slipped vessels are widespread, but attributes and fea-
tures which can be designated as widely distributed chronological markers
are still unidentified (Engwall, 1996). One characteristic that distinguishes
Late Formative pottery from the late Valdivia and early Machalilla wares is
the thinness of the vessel walls and their overall fine craftsmanship, even of
the utilitarian vessels. Such technological distinctions point to the develop-
ment of full-time craft specialists.
Lanning (1968) was among the first to object to the notion of a Late For-
mative horizon, classifying Engoroy as a distinct regional entity. Lathrap and
others (1975, p. 15) concurred maintaining Engoroy was unrepresentative
and a less elaborate, variant of Chorrera. Subsequently, a number of schol-
ars have come to prefer to discuss the Late Formative pottery complexes in
terms of regional manifestations rather than under the rubric of Chorrera
(Aleto, 1988; Beckwith, 1996; Bischof, 1975; Cummins, 1992; Engwall, 1995,
1996; Lippi, 1983; Paulsen and McDougle, 1981; Simmons, 1970; Stothert,
1995). However, given the previously outlined regional distributions and
developmental affinities, the regional variability of the pottery style may be
explained to chronological and developmental differences. Estrada (1958)
excavated more Late Formative Period sites than any other investigator
before or since, and therefore had a breadth of knowledge regarding the
regional assemblages that was unique among archaeologists. Estrada chose
to label these regional phases under the rubric of Chorrera. As with ce-
ramic complexes of the earlier periods, the most complete descriptions of
the Late Formative pottery come from sites in SW coastal Ecuador and
S Manabı́ (Beckwith, 1996; Bischof, 1975; Bushnell, 1951, 1982; Paulsen and
McDougle, 1981) biasing the archaeological record.

CHORRERA CHRONOLOGY

The chronology of the Late Formative Period of coastal Ecuador was

initially based upon forty-four 14 C and obsidian hydration dates from var-
ious subregions (Meggers, 1966, Fig. 4). Meggers derived a chronological
range on this basis (Meggers, 1966, pp. 55–66). However, the 1840 ± 540 B.P.
date was derived from obsidian hydration, and the earliest 14 C measurement
2800 ± 115 B.P. has been questioned on contextual grounds (Meggers, 1966,
Fig. 4; Meggers et al., 1965, Table H). Subsequently, the chronological range
for the Late Formative was challenged. Archaeological investigations at San
Isidro produced a [2845 ± 95 years AA-4140] date from the earliest Chorrera
securely excavated contexts (Zeidler and Sutliff, 1994, p. 115). Bischof (1982,
p. 162) also places the end of the Machalilla Phase and beginning of the Late
Formative at approximately 2850 B.P., pointing out that the 2800 ± 115 B.P.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 239

date from Véliz that may actually be derived from late Machalilla occupa-
tions at La Cabuya. Paulsen and McDougle (1974, 1981) also report similar
early dates for the Engoroy occupation at OGSE-46D and suggest a chrono-
logical range of 2850 to 2250 B.P. for this subregion. Along the Jama River in
N Manabı́, Engwall (1995) obtained a [2500 B.P. ± 160 years; ISGS-2377] date
from a layer of volcanic tephra to mark the end of the Tabuchila Phase. This
date agrees with the terminal date for the early estimates. Excavations at
Dos Caminos (M3D2-008) produced three of the earliest dates for any Late
Formative Period complex. The uncalibrated dates [ISGS-3308; 2930 B.P. ±
80 years], [ISGS-3309; 2930 B.P. ± 80], [ISGS-3310; 2880 B.P. ± 70 all point
to an initial date of ca., 2900 B.P. (Engwall, 1995). Zeidler (1994) concludes
that the Late Formative Period Tabuchila Phase also begins at around this
time and ends at about 2300 B.P. This estimate is based upon the presence of
volcanic tephra overlying Tabuchila Phase occupation layers. In N Manabı́,
Valdivia VIII and early Late Formative Tabuchila Phase layers are strati-
graphically separated by a tectonic event (Issacson, 1994; Zeidler, 1994).
In SW coastal Ecuador and southern Manabı́, Late Formative occupa-
tions most often directly overlay Machalilla Phase occupation layers
(Bischof, 1975, 1982; Estrada, 1957, 1958, 1962a; Evans and Meggers, 1957;
Lanning, 1968; Paulsen and McDougle, 1974, 1981). However, in coastal El
Oro and the Guayas Basin Late Formative Period occupations tend to over-
lay late Valdivia layers (González de Merino, 1984; Reindel and Guillaume–
Gentil, 1995; Staller, 1994, 2000a).
In coastal El Oro, Late Formative Period Arenillas Phase occupations
are reported and associated with multicomponent sites with Valdivia VII–
VIII occupations (Staller, 1994, 2000a). At La Florida, Late Formative
Arenillas Phase deposits are directly superimposed upon Valdivia VII–VIII
Jelı́ Phase and early Machalilla occupations (Staller, 2000a, pp. 250–251).
Moreover, stylistic similarities in a number of vessel forms between these var-
ious components appear to reflect developmental continuity (Staller, 1994).
In the Guayas Basin at Peñon del Rı́o, a diverse ceramic assemblage has
been reported, including evidence of mound building and domestic habita-
tions (Muse, 1989; Zedeño, 1985, 1987). Despite an extensive Late Formative
occupation in this very large site, there is no evidence of Machalilla material
culture. Similarly, Evans and Meggers (1957, 1982) did not report any evi-
dence of Machalilla occupation at La Chorrera, and Late Formative occupa-
tions are directly over a Valdivia VIII occupation at Milagro I (González de
Merino, 1984). On the basis of this archaeological evidence we may infer that
Machalilla style pottery does not appear in this subregion and is marginally
represented in areas to the south. The evidence from the eastern portion
of the Gulf of Guayaquil and coastal El Oro suggests a direct development
from Valdivia VIII to Chorrera.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

240 Staller

In conclusion, early estimates suggest a chronological range of 3500–

2500 B.P. for the Late Formative Period (Meggers, 1966, Fig. 4), while the
chronological range for the Engoroy complex is placed at around
2750–2050 B.P. (Paulsen and McDougle, 1981). On the basis of radiocar-
bon dates Currie (1985, 1992) estimates the end of the Late Formative at
around 2250 B.P. in coastal El Oro, and Zeidler (1994) reports a compara-
ble estimate for Tabuchila Phase occupation in N Manabı́. The continuous
Formative occupation in coastal El Oro and the Gulf of Guayaquil supports
an estimated range of ca. 3400–3250 B.P. for the beginning of the Late For-
mative Period in these regions. Over the rest of coastal Ecuador, specifically
in areas of Machalilla related material culture, or in N Manabı́ where there
is a chronological hiatus, the Late Formative appears to begin at ca. 2950–
2750 B.P.
Some have dismissed the obsidian hydration date from La Chorrera
as due to a fault in dating technique (Beckwith, 1996, Table 2.1, pp. 15–
18). However, given the apparent absence of Machalilla in the Jama River
drainage, and the brief duration of occupation of the complex to the south
in El Oro (Staller, 1994, 2000a), it is possible to explain such seeming dis-
crepancies in the archaeological record in other terms.

DISCUSSION AND CONCLUSIONS

Previous analyses of Formative chronologies, ceramic seriations, and

settlement patterns suggest the earliest ceramic occupations are coastal and
concentrated between the Verde and Valdivia Rivers in SW coastal Ecuador.
This subregion represents the nucleus of Valdivia sociocultural development
(Fig. 6). Valdivia regional surveys further indicate a decline in settlements
related in part to an adaptive shift during the final portion of the Valdivia
culture sequence. This change in site locations strongly suggests a greater
reliance upon agriculture, and may be related in part to climatic, environ-
mental, and geomorphologic changes after Phase V times (Staller, 1994).
The only Valdivia VI–VIII ceremonial center in SW coastal Ecuador is La
Centenela. However, the diagnostics and stratigraphic occupations of this in-
land site have not been reported in any detail. A seeming absence of Valdivia
VIII ceremonial centers in SW coastal Ecuador and S Manabı́ suggests on-
going changes in population density reflecting a reduction in the mangrove
formation, and changes in the environment and in the subsistence economy
(Ferdon, 1981; Staller, 1994). In contrast, major Phase VII–VIII ceremo-
nial centers are present in coastal El Oro, the western side of the Gulf
of Guayaquil, and N Manabı́, supporting the hypothesis of fundamental
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 241

changes in the geographic locus of sociopolitical development. Changes in

settlement to a more inland riverine pattern, and expansion into N Manabı́,
the Guayas Basin, and southern coastal El Oro in part also explains why
documenting a developmental transition between Valdivia and the subse-
quent Machalilla culture complex was initially so elusive, because most of
the early research was concentrated in SW coastal Ecuador.
During Phase VII–VIII there appears to have been a relative decline
of previously established long-distance maritime interaction along the coast,
and a concomitant intensification and elaboration of long-distance exchange
with the northern and southern highlands of Ecuador and northern highlands
of Peru (Staller, 1992/93, 1994, 1996a, 1996b; Villalba, 1988; Zeidler, 1988,
1991). Complicating the issue of a transition were thin layers of Machalilla
refuse with almost all late Valdivia occupations south of the Rı́o Valdivia
and north of the Rı́o Verde (Bischof, 1975; Cliff et al., 1987; Lanning, 1968;
Lippi, 1982, 1983; Paulsen and McDougle, 1981; Schwarz and Raymond,
1996; Zeidler, 1977). The archaeological patterning in SW coastal Ecuador
had a profound effect upon hypotheses regarding a developmental link be-
tween the Valdivia and Machalilla Phases. The ceramic evidence presented
here indicates substantial regional diversity by late Valdivia VI–VIII times
and dramatic technological changes during Phase VII–VIII. Current ceramic
data suggest that the cultural history of late Valdivia societies in coastal El
Oro, the Gulf of Guayaquil, and N Manabı́ were developmentally, and to
some extent, chronologically distinct from those of SW coastal Ecuador and
S Manabı́. However, during the final portion (Phases VII–VIII) of the se-
quence, it is the continuity in stirrup-spouts, carinated vessels, composite
forms, and stylistic attributes such as fillet appliqué, zoned punctation, fine
line incision, and excision, as well as the figurines from Chacras, San Lorenzo
del Mate, and other sites, that link the Valdivia and Machalilla traditions
(Staller, 1994, 1996b, 1998, 2000b, 2001). These Formative components can
therefore be seen as derived from a single tradition.
The regional Late Formative complexes appear to have developed di-
rectly from a Valdivia VIII base in some regions of the Ecuadorian lowlands
and from Machalilla in other subregions (Staller, 1994). Pedestal bowls, ev-
erted jars, and single spout bottles of the Jelı́ Phase complex suggest a contin-
uous tradition associating Valdivia with the Late Formative components in
southern coastal El Oro and in the Gulf of Guayaquil (Staller, 1994, 2000a,
2001; Weinstein, 1998). On the other hand, the developmental association is
with Machalilla in SW coastal Ecuador and S. Manabı́. These regional dif-
ferences beg the question of how Machalilla should be understood in terms
of Ecuadorian prehistory both from a historical as well as a cultural stand-
point. I suggest that Machalilla may be seen as a chronological and cultural
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

242 Staller

extension of Valdivia (see also Lathrap et al., 1975). In N Manabı́ and over
most of the Guayas Basin, the Valdivia–Machalilla traditions develop di-
rectly into what may be properly termed a Chorrera style. The transition in
these subregions and in areas to the south, as far as the Piura coast in Peru,
appears to be much earlier dating to ca. 3400–3250 B.P. although this estimate
is not directly based upon dates from Late Formative contexts. Machalilla
occupation is most intense and of longest duration in the areas around the
Rı́o Valdivia and further to the north in S Manabı́ especially in the imme-
diate vicinity of the coastal village of Machalilla (Estrada, 1958; Evans and
Meggers, 1958; Meggers et al., 1965). It is in precisely these areas that the
Late Formative regional phases begin to appear at about ca. 2950–1700 B.P.
Similar early dates for the Late Formative Period appear to characterize N
Manabı́, but here the reasons are not so much developmental as they are
environmental and geological.
With regards to Valdivia, the regional chronological patterns suggest a
spread into N and S Manabı́ Province but this radiation must be understood
in terms of factors related to the decline of mangrove and geomorphologic
change in the coastal areas of SW Ecuador and S Manabı́. The areas of most
intense Machalilla occupation coincide with those portions of the coast in
which tectonic uplift has occurred, but where mangrove formations per-
sist along the coastal lowlands, and where aquatic and maritime resources
are most accessible. The adaptive pattern of coastal populations appears to
have been focused upon maritime resources and small-scale farming and
this pattern is essentially a continuation of the earlier Valdivia VII–VIII
adaptation.
Future investigations should more closely analyze the development and
integration of ceramic innovation in other regions of the Ecuadorian low-
lands as being derived from the Valdivia–Machalilla tradition as Lathrap and
others (1975) had suggested. Such phasing will then encompass the greater
portion of the Ecuadorian lowlands (south of Esmeraldas Province) more
precisely, since all but the Tabuchila Phase appear to be directly derived
from this ceramic tradition. The extent to which Tabuchila is derived from
Piquigua Phase Valdivia VIII components is at this time problematic given
the hiatus in occupation in this region. If the Tabuchila Phase is not derived
from a local Early Formative Period tradition, we must consider the pos-
sibility that it may have originated in this region from the spread of Late
Formative Period agricultural societies from the eastern side of the coastal
hills and ultimately from the Guayas Basin. Since regional variation in the
Late Formative ceramic complexes is only now beginning to be understood,
such developmental connections need further consideration as more system-
atic research is undertaken.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 243

Although the role of the Valdivia ceramic tradition has generally been
overlooked by scholars who focused upon the later periods, the impor-
tance of other aspects of the ritual complex which included Spondylus and
Strombus shell, psychotropic (San Pedro Cactus) and narcotic plants (coca,
Datura spp.), feline and avian symbolism, ritual consumables (red pepper,
maize, and salt), and various ceramic bottles is apparent (Staller, 1994,
Tables 9 and 10). It was initially assumed that Valdivia ceramics played a ma-
jor role in the development of ceramic innovation outside of the Ecuadorian
lowlands. As more research on the final portion of the Valdivia sequence be-
comes available, it will be increasingly apparent that the Valdivia–Machalilla
ceramic tradition was central to the development of ceramic innovations and
incipient complexity in other regions of western South America.
In conclusion, the synthesis of Formative Ecuador presented here sug-
gests the following: (1) A regional divergence in Valdivia material culture
during Phases VII–VIII directly supports the premise of a ceramic transition
linking Valdivia to the subsequent Machalilla Phase and demonstrates these
complexes are part of the same ceramic tradition. It is now apparent that the
Valdivia–Machalilla transition is characterized by a proliferation of distinct
vessel forms and stylistic attributes, as well as changes in the pottery technol-
ogy. (2) A major adaptive shift occurred during Phase VI–VIII times. The
changes in subsistence are related to an increased dependence upon agricul-
ture by inland Valdivia populations. During this adaptive shift there is a con-
comitant reduction in hunting and gathering of wild plants and animals and
an increased dependence upon estuarine and maritime resources by coastal
populations. (3) In association with changes in the subsistence adaptation
is a reduction of Valdivia settlements in SW coastal Ecuador between the
Rı́o Verde and Valdivia, related in part to geomorphologic changes and a re-
duction or possible extinction of the mangrove forest around Phase V times.
Changes in the distributions and densities of Valdivia settlement correspond
to chronological and regional differences in the locus of sociocultural devel-
opment. (4) There is a developmental link between the Valdivia–Machalilla
ceramic tradition and the Late Formative Period sequences in far north Peru,
coastal El Oro, the Gulf of Guayaquil as well as the northern and southern
highlands of Ecuador. (5) These developmental associations are related to
the establishment of long-distance interaction with the Andean highlands
and later the ceja de montaña of eastern Ecuador. A central feature of this
interaction was the spread of a ritual and religious cult expressed archaeo-
logically by Chorrera related material culture over a vast geographic area.
The ritual complex associated with the spread of this cult also involved a cos-
mology associated with the Strombus galeatus and Spondylus princeps dyad,
and included various consumable and sumptuary plants and resources. The
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

244 Staller

sudden appearance of finely crafted polished zoomorphic stone mortars con-

taining feline and avian symbolism indicates that some of these plants were
associated with Andean ritual ceremony. Similarly, the ideology and cos-
mology of this early cult is also symbolically expressed in the finely crafted
zoomorphic and anthropomorphic ceramic bottles and serving vessels. The
early spread of ceramic innovation associated with Valdivia VII–VIII and
Chorrera related material culture is evident archaeologically by the inte-
gration of composite forms such as constricted jars, elaborately decorated
open bowls with vertical walls, and ceramic bottles (especially stirrup spout
and single spout forms) and elaborate mold made effigy vessels over a large
geographic area (Fig. 13). It is suggested that the ideology associated with
this ritual complex formed the basis of Andean cosmology over a vast area
of western South America in subsequent periods where it was symbolically

Fig. 13. Diagnostic vessel forms of the Valdivia VII–VIII Jelı́ Phase complex from La
Emerenciana. These diagnostics appear to have significance to the spread of ceramic inno-
vation, (a) Single spout bottle, (b) Neckless olla, (c) Constricted Olla or Jar, (d) Two-tiered
stirrup-spout bottle, (e.) Stirrup-spout bottle (globular), (f) Open Bowl with vertical wall and
flat bottom, (g) Constricted bowl with decorated bottom, and (h) Everted bowl with decorated
bottom.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 245

represented according to local and regional artistic canons in a mosaic of

cultural expressions.

ACKNOWLEDGMENTS

I extend my deepest appreciation to Fulbright-Hays, the Department of

Anthropology at Southern Methodist University, Dallas, and to the Museo
Antropológico del Banco Central del Ecuador, Guayaquil for making the
archaeological research I carried out in El Oro province possible. I also
express sincerest thanks to the anonymous reviewers, their comments and
insights made this a better synthesis of the Formative of coastal Ecuador.

REFERENCES

Aleto, T. F. (1988). The Guayaquil Phase Complex: The Late Formative Period in the Gulf of
Guayaquil, Ecuador, PhD dissertation, University of Illinois, Urbana.
Asaro, F., Salazar, E., Michel, H. V., Burger, R. L., and Stross, F. H., (1994). Ecuadorian obsid-
ian sources used for artifact production and methods for provenience assignments. Latin
American Antiquity 5(3): 256–277.
Beckwith, L. A. (1996). Late Formative Phase Ceramics From Southwestern Ecuador, PhD
dissertation, Department of Archaeology, University of Calgary, Calgary, Alberta.
Bennett, W. (1946a). The archaeology of the central Andes, In Steward, J. H. (ed.), Handbook of
South American Indians, Vol. 2, Bulletin 143, Bureau of American Ethnology, Smithsonian
Institution, Washington, DC, pp. 61–148.
Bennett, W. (1946b). Excavations in the Cuenca Region, Ecuador, Yale Publications in Anthro-
pology, Vol. 35, Yale University Press, New Haven.
Benz, B. F. (1998). On the origin, evolution, and dispersal of maize. In Blake, M. (ed.)., Pacific
Latin America in Prehistory: The Evolution of Archaic and Formative Cultures., Washington
State University Press, Pullman, pp. 25–38.
Benz, B. F., and Iltis, H. H. (1990). Studies in archaeological maize. I. The “wild” maize from
San Marcos Cave reexamined. American Antiquity 55: 500–511.
Benz, B. F., and Long, A. (2000). Prehistoric maize evolution in Tehuacan Valley. Current
Anthropology 41(3): 459–465.
Bischof, H. (1967). Review of “Early Formative Cultures of Coastal Ecuador” by Meggers,
Evans, and Estrada, 1965. American Journal of Archaeology 71(2): 216–219.
Bischof, H. (1972). The origins of pottery in South America—Recent radiocarbon dates from
Southwestern Ecuador. Proceedings of the 40th International Congress of Americanists,
Vol. 1, Rome/Genoa, pp. 269–281.
Bischof, H. (1975). El Machalilla temprano y algunas sitios cercanos a Valdivia. In von Oberem,
U. (ed.), Estudios sobre la arqueologı́a del Ecuador, BAS 3 Estudios Americanistas de
Bonn, pp. 43–61.
Bischof, H. (1980). San Pedro und Valdivia—frühe keramik Komplexe an der Kuste sudwest
Ekuadors. Beiträge zur Allgemeinen und Vergleichenden Archäologie, Vol. 1, Deutsches
Archaologisches Institut, München, pp. 335–389.
Bischof, H. (1982). La Fase Engoroy—periodos, cronologı́a y relaciones. In Marcos, J. G.,
and Norton, P. (eds.), Primer Simposio de Correlaciones Antropológicas Andino-
Mesoamericano, Salinas, Ecuador, ESPOL, Guayaquil, Ecuador, pp. 135–176, Junio, 1971.
Bischof, H., and Viteri-Gamboa, J. (1972). Pre-Valdivia occupations on the southwest coast of
Ecuador. American Antiquity 37(4): 548–551.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

246 Staller

Bruhns, K. O., Burton, J. H., and Miller, G. R. (1990). Excavations at Pirincay in the Paute
valley of southern Ecuador, 1985–1988. Antiquity 64(3): 221–233.
Burger, R. L. (1988). Unity and heterogeneity within the Chavı́n horizon. In Keatinge, R. W.
(ed.), Peruvian Prehistory, Cambridge University Press, Cambridge, pp. 99–144.
Burger, R. L. (1992). Chavı́n and the Origins of Andean Civilization, Thames and Hudson,
London.
Burger, R. L., Asaro, F., Michel, H. V., Stross, F. H., and Salazar, E. (1994). An initial consid-
eration of obsidian procurement and exchange in pre-Hispanic Ecuador. Latin American
Antiquity 5(3): 228–255.
Bushnell, G. H. S. (1951). The Archaeology of the Santa Elena Peninsula in South-West Ecuador,
Cambridge Museum Occasional Papers, No. 1, Cambridge, England.
Bushnell, G. H. S. (1982). Semejanzas y relaciones entre Mesoamerica, el ecuador y el perú:
Unas observaciones en la cultura. In Marcos, J. G., and Norton, P. (eds.), Primer Simpo-
sio de Correlaciones Antropológicas Andino-Mesoamericano, Salinas, Ecuador, ESPOL,
Guayaquil, Ecuador, pp. 37–40, Junio, 1971.
Bush, M. B., Piperno, D. R., and Colinvaux, P. A. (1989). A 6000 year history of Amazonian
maize cultivation. Nature 340(6231): 303–305.
Byrd, K. M. (1976). Changing Animal Utilization Patterns and Their Implications: S.W., Ecuador
6500BC–AD1400. PhD dissertation, Department of Anthropology, University of Florida,
Gainesville, Florida.
Cañadas Cruz, L. (1983). El mapa bioclimático y ecológico del Ecuador, MAG-PRONAREG,
Quito.
Cliff, M. B., Staller, J. E., and Marks, A. (1987). A Preliminary Report on Early Occupation North
of Real Alto, Ecuador, Unpublished manuscript, Department of Anthropology, Southern
Methodist University, Dallas Texas.
Coe, M. D. (1960). Archaeological linkages with North and South America at La Victoria,
Guatemala. American Anthropologist 62: 363–393.
Coe, M. D., and Flannery, K. V. (1967). Early Cultures and Human Ecology of South Coastal
Guatemala, Smithsonian Contributions to Anthropology, Vol. 3, Smithsonian Institution
Press, Washington, DC.
Collier, D. (1947). The archaeology of Ecuador. In Steward, J. H. (ed.), Handbook of South
American Indians, Vol. 2, Bulletin 143, Bureau of American Ethnology, Smithsonian In-
stitution Press, Washington, DC, pp. 767–784.
Collier, D. (1948). Peruvian influences in Ecuador. Memoirs of the Society for American Ar-
chaeology 13(4): 80–86.
Collier, D. (1968). Review of “Ecuador” by Meggers 1966. American Antiquity 33(3): 269–
271.
Collier, D., and Murra, J. (1943). Survey and Excavations in Southern Ecuador, Field Museum
of Natural History Publication 528, Chicago.
Cruz, F., and Holm, O. (1982). Un informe muy preliminar: San Lorenzo del Mate, Provincia del
Guayas, Unpublished manuscript, Museo Antropológico del Banco Central del Ecuador,
Guayaquil, Ecuador.
Cummins, T. (1992). Tradition in Ecuadorian Pre-Hispanic Art, the ceramics of Chorrera and
Jama-Coaque. In Valdez, F., and Veintimilla, D. (eds.), Amerindian Signs 5000 years of
Precolumbian Art in Ecuador, Ediciones Colibri, Paris, pp. 63–133.
Currie, E. J. (1985). La cultura Jambelı́ con referencia particular al conchero Guarumal. In
Moreno Yanez, S. E. (ed.), Memoria del Primer Simposio Europeo Sobre Antropologı́a del
Ecuador, Instituto de Antropologı́a cultural de la Universidad de Bonn, Ediciones Abya
Yala, Quito, Ecuador, pp. 31–46.
Currie, E. J. (1989). Cultural Relationships in Southern Ecuador 300 BC-AD 300: Excavations at
the Guarumal and Punta Brava Sites, PhD dissertation, Institute of Archaeology, University
College, London.
Currie, E. J. (1992). New Evidence on the Late Formative in El Oro Province, Southwestern
Ecuador, Unpublished manuscript, Department of Archaeology, University of York,
England.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 247

Damp, J. E. (1979). Better Homes and Gardens: The Life and Death of the Early Valdivia
Community, PhD dissertation, Department of Archaeology, University of Calgary, Calgary,
Alberta.
Damp, J. E. (1984a). Environmental variability, agriculture, and settlement processes of coastal
Ecuador (3,300–1,500 B.C.). Current Anthropology 25: 106–111.
Damp, J. E. (1984b). Architecture of the Early Valdivia village. American Antiquity 49: 573–585.
Damp, J. E. (1988). La primera ocupación Valdivia de Real Alto: Patrones económicos, arqui-
tectónmicos e ideológicos, Biblioteca Ecuatoriana de Arqueologı́a, Corporción Editora,
Nacional, Quito.
Damp, J. E., Jackson, D., Vargas S. P., and Zambrano, P. (1990). On the waterfront: Quaternary
environments and the Formative occupation of southwestern Ecuador. Geoarchaeology 5:
171–185.
Damp, J. E., and Pearsall, D. M. (1994). Early cotton from coastal Ecuador. Economic Botany
48(2): 163–165.
Damp, J. E., and Vargas S. P. (1995). The many contexts of early Valdivia ceramics. In Barnett,
W. K., and Hoopes, J. W. (eds.), The Emergence of Pottery, Smithsonian Institution Press,
Washington, DC, pp. 157–168.
DeBoer, W. R. (1996). Traces Behind the Esmeraldas Shore. Prehistory of the Santiago-Cayapas
Region, Ecuador, The University of Alabama Press, Tuscaloosa and London.
Delavaud, A. C. (ed.). (1982). Atlas del Ecuador, Les Editions J. A., Paris and the Banco Central
del Ecuador, Quito.
Denevan, W. M., and Mathewson, K. (1985). Preliminary results of the Samborondón Raised-
Field Project. In Darch, J. P. (ed.), Drained Field Agriculture in Central and South America,
British Archaeological Reports, International Series 189, Oxford, pp. 167–203.
Dorsey, G. A. (1901). Archaeological Investigations on the Island of La Plata, Field Museum of
Natural History, Fieldiana, Vol. 2(5), Chicago.
Engwall, E. C. (1992). Archaeological Investigations of Chorrera culture: Tabuchila Phase Ce-
ramics from Northern Manabı́, Ecuador, Paper presented at the 57th Annual Meeting of
the Society for American Archaeology, Pittsburgh, Pennsylvania.
Engwall, E. C. (1995). The Tabuchila Phase: Late Formative Chorrera Culture in the Jama River
Region, Manabı́. Ecuador, Paper presented at the 60th Annual Meeting of the Society for
American Archaeology, Minneapolis, Minnesota.
Engwall, E. C. (1996). Coalescence and Chaos: Toward a Model of Chorrera Style(s), Paper
presented at the 61st Annual Meeting of the Society for American Archaeology, New
Orleans, Louisianna.
Estrada-Ycaza, V. E. (1956). Valdivia un sitio Arqueológico Formativo en la Costa de la Provincia
del Guayas, Ecuador, Publicación del Museo Victor Emilio Estrada Publicación No. 1,
Guayaquil, Ecuador.
Estrada-Ycaza, V. E. (1957). Prehistorı́a de Manabı́, Publicación del Museo Victor Emilio
Estrada, Publicación No. 4, Guayaquil, Ecuador.
Estrada-Ycaza, V. E. (1958). Las Culturas Pre-Clasicas, Formativas o Arcaicas del Ecuador,
Publicación del Museo Victor Emilio Estrada, Publicación No. 5, Guayaquil, Ecuador.
Estrada-Ycaza, V. E. (1962a). Arqueologı́a de Manabı́ Central, Publicación del Museo Victor
Emilio Estrada, Publicación No. 7, Guayaquil, Ecuador.
Estrada-Ycaza, V. E. (1962b). Correlaciones entre la arqueologı́a de la costa del Ecuador y
Perú. Humanitas 2(2): 31–61.
Estrada, V. E., Meggers, B. J., and Evans Jr., C. (1962). Possible transpacific contact on the coast
of Ecuador. Science 135: 371–372.
Evans, C., Jr., and Meggers, B. J. (1954). Informe preliminar sobre las investigaciones Ar-
quelógicas Realizadas Engoroy la Cuenca del Guayas, Ecuador. Cuadernos de Historia y
Arqueologı́a 4(12): 1–30.
Evans, C., Jr., and Meggers, B. J. (1957). Formative period cultures in the Guayas Basin, coastal
Ecuador. American Antiquity 22: 235–247.
Evans, C., Jr., and Meggers, B. J. (1958). Valdivia—An Early Formative culture of Ecuador.
Archaeology 11: 175–182.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

248 Staller

Evans, C., Jr., and Meggers, B. J. (1966). Mesoamerica and Ecuador. In Wuachope, R., Ekholm,
G. F., and Willey, G. R. (eds.), Handbook of Middle American Indians, Vol. 4, University
of Texas Press, Austin, pp. 243–264.
Evans, C., Jr., and Meggers, B. J. (1982). Técnicas decorativas diagneosticas y variantes re-
gionales chorrera, un anealisis preliminar. In Marcos, J. G., and Norton, P. (eds.), Primer
Simposio de Correlaciones Antropológicas Andino-Mesoamericano, Salinas, Ecuador, ES-
POL, Guayaquil, Ecuador, pp. 121–133, Junio, 1971.
Evans, C., Jr., Meggers, B. J., and Estrada E. (1959). Cultura Valdivia, Publicación del Museo
Victor Emilio Estrada No. 6, Guayaquil, Ecuador.
Fagan, B. (1999). Floods, Famines, and Emperors, El Niño and the Fate of Civilizations, Basic
Books, New York.
Ferdon, E. N., Jr. (1940a). Investigación arqueológica en el Ecuador. Boletı́n de la Academia
Nacional de Historia 29(55): 103–108.
Ferdon, E. N., Jr. (1940b). The archaeological survey of Ecuador. El Palacio 47(6): 137–144.
Ferdon, E. N., Jr. (1941a). Reconaissance in Esmeraldas. El Palacio 47(12): 257–274.
Ferdon, E. N., Jr. (1941b). The excavations at La Libertad. El Palacio 48(2): 38–42.
Ferdon, E. N., Jr. (1942). Ecavaciones arqueológicas cerca de La Libertad, provincia de Guayas.
Boletı́n de la Academia Nacional de Historia 22(59): 75–81.
Ferdon, E. N., Jr. (1950). The climates of Ecuador. School of American Research and Museum
of New Mexico, Bulletin 15: pp. 35–63.
Ferdon, E. N., Jr. (1981). Holocene mangrove formations on the Santa Elena Peninsula,
Ecuador. Pluvial indicators or ecological response to physiographic changes. American
Antiquity 46(3): 619–626.
Ferdon, E. N., Jr., and Corbett, J. M. (1941). Depósitos arqueológicos de La Tolita. Boletı́n de
la Academia Nacional de Historia 21(57): 5–15.
Ford, J. A. (1969). A Comparison of Formative Cultures in the Americas, Smithsonian Contri-
butions to Anthropology. Vol. 11, Smithsonian Institution Press, Washington, DC.
González de Merino, J. (1984). Investigación Cientifica de Arqueologica en el Sitio Milagro 1,
Canton Milagro, Provincia del Guayas, Primera Parte: Informe Preliminar; Segunda Parte:
Informe Final, Universidad de Guayaquil.
Grove, D. (1982). The Mesoamerican Formative and South America. In Marcos, J. G.,
and Norton, P. (eds.), Primer Simposio de Correlaciones Antropológicas Andino-
Mesoamericano, Salinas, Ecuador, Junio, 1971, ESPOL, Guayaquil, Ecuador, pp. 279–301.
Hill, B. (1972/74). The new chronology of the Valdivia ceramic complex from the coastal zone
of Guayas Province, Ecuador. In Nawpa Pacha, Vol. 10–12, Institute of Andean Studies,
Berkeley, 1–32.
Holmgren, M., Scheffer, M., Ezcurra, E., Gutiérrez, J. R., and Mohren, G. M. J. (2001). El Niño
effects on the dynamics of terrestrial ecosystems. Trends on Ecology & Evolution 16(2):
89–94.
Hoopes, J. W. (1994). Ford Revisited: A Critical Review of the Chronology and Relationship
of the Earliest Ceramic Complexes in the New World, 6000–1500 B.C. Journal of World
Prehistory 8(1): 1–49.
Issacson, J. S. (1994). Volcanic Sediments in Archaeological Contexts from Western Ecuador.
In Zeidler, J. A., and Pearsall, D. M. (eds.), Regional Archaeology in Northern Ecuador,
Volume 1: Environment, Cultural Chronology and Prehistoric Subsistence in the Jama
River Valley, University of Pittsburgh, Memoirs in Latin American Archaeology No. 8,
Pittsburgh, pp. 131–140.
Izumi, S., and Terada, K. (1966). Andes 3 Excavations at Pechiche and Garbanzal, Tumbes
Valley, Peru, 1960, Kadokawa, Tokyo.
Jadán, M. (1986). La cerámica del Complejo Piquiqua (fase 8) de la cultura Valdivia en San
Isidro, norte de Manabı́: Un analisis modal, Thesis, Escuela Superior Politecnica del Litoral,
ESPOL, Guayaquil, Ecuador.
Jijon y Caamaño, J. (1919). Contribución al conocimiento de las lenguas indigenes que se
hablamos en el Ecuador interandino y occidental, con anteriormente a la conquista español.
Boletı́n de la Sociedad Ecuatoriana de Estudios Históricos 2(6): 340–413.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 249

Jijon y Caamaño, J. (1920). Contribución al conocimiento de los aborı́genes de la provincia

de Imbabura. Boletı́n de la Sociedad Ecuatoriana de Estudios Históricos 4(10): 1–120, and
4(11): 183–248.
Jijon y Caamaño, J. (1941). El Ecuador Interandino y occidental, Vol. 1, Quito.
Jijon y Caamaño, J. (1952). Antropogia Prehispanica del Ecuador, Embajada de España Agencia
Española de Cooperación International, Museo Jacinto Jijon y Caamaño, Quito. Reprinted
1997.
Kerr, R. A. (1999). El Niño grew strong as cultures were born. Science 283: 467–468.
Knapp, G. (1988). Ecologı́a cultural prehispánico del Ecuador, Biblioteca de Geografia Ecua-
toriana, Vol. 3., Banco Central del Ecuador, Quito, Ecuador.
Kroeber, A. (1948). Review of “Archéologie de la province d’ Esmeraldas, I’ I’Equateur by
D’Harcourt 1947. American Antiquity 14(2): 139–140.
Lanning, E. P. (1960). Chronological and Cultural Relationships of Early Pottery Styles in Ancient
Peru, PhD dissertation, Department of Anthropology, University of California, Berkeley.
Lanning, E. P. (1963). A Ceramic Sequence for the Piura and Chira Coast, North Peru, Uni-
versity of California Publications in American Archaeology and Ethnology, University of
California Press, Berkeley and Los Angeles, California.
Lanning, E. P. (1967). Peru Before the Incas, Prentice-Hall, Englewood Cliffs.
Lanning, E. P. (1968). Informe preliminar de la Peninsula de Santa Elena, Unpublished
manuscript, Casa de la Cultura Ecuatoriana, Nucleo de Guayas, Guayaquil, Ecuador.
Lathrap, D. W. (1960). Review of: Valdivia: by Evans, Meggers, and Estrada, 1959. American
Antiquity 26(1): 125–127.
Lathrap, D. W. (1963). Possible affiliations of the Machalilla complex of coastal Ecuador. Amer-
ican Antiquity 29(2): 239–241.
Lathrap, D. W. (1967). Review of “Early Formative Cultures of Coastal Ecuador” by Meggers,
Evans, and Estrada 1965. American Anthropologist 69(1): 96–98.
Lathrap, D. W. (1970). Upper Amazon, Thames and Hudson, London.
Lathrap, D. W. (1971). The tropical forest and the cultural context of Chavı́n. In Benson, E. P.
(ed.), Dumbarton Oaks Conference on Chavı́n, Dumbarton Oaks Research Library and
Collection, Washington, DC, pp. 73–100.
Lathrap, D. W. (1973). The antiquity and importance of long-distance trade relationships
in the moist tropics of Pre-Columbian South America. World Archaeology 5(2): 170–
186.
Lathrap, D. W. (1974). The moist tropics, the arid lands, and the appearance of the great art
styles of the New World. In King, M. E., and Traylor, I., Jr. (eds.), Art and Environment in
Native America, Texas Tech University Museum, Lubbock, pp. 115–158.
Lathrap, D. W., Collier, D., and Chandra, H. (1975). Ancient Ecuador Culture, Clay, and Cre-
ativity 3000—300 B.C., Field Museum of Natural History, Chicago.
Lathrap, D. W., Marcos, J. G., and Zeidler, J. E. (1977). Real Alto: An ancient ceremonial center.
Archaeology 30(1): 2–13.
Ledergerber, P. (1983). El origen de más de un cuarto siglo de investigaciones sobre la cultura
Valdivia. Boletı́n de la Academia Nacional de Historia LXV: 25–40.
Lippi, R. D. (1982). A New Machalilla Ceramic Sequence, Paper presented at the 47th Annual
Meeting of the Society for American Archaeology, Minnesota.
Lippi, R. D. (1983). La Ponga and the Machalilla Phase of Coastal Ecuador, PhD dissertation,
Department of Anthropology, University of Wisconsin, Madison.
Lippi, R. D. (1998). Una Exploración Arqueológica del Pichincha Occidental Ecuador, Museo
Jacinto Jijón Caamaño, Pontificia Universidad Católica del Ecuador, Quito, Ecuador.
Long, A., Benz, B. F., Donahue, J., Jull, A., and Toolin, L. (1989). First direct AMS dates on
early maize from Tehuacan, Mexico. Radiocarbon 31: 1035–1040.
López y Sebastián, L. E., and Caillavet, C. (1979). La fase Tachina en el contexto cultural de
horizonte Chorrera. Actes du XLII Congress International des Americanistes, Paris, 1976,
Vol. IX-A, pp. 199–215.
Lubensky, E. (1980). Shell mound excavation at Ayalán, Unpublished manuscript, Museo
Antropológico del Banco Central del Ecuador, Guayaquil, Ecuador.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

250 Staller

Lubensky, E. (1995). The Anllulla Shell Mound in Coastal Ecuador, Manuscript distributed
at the Precolumbian Symposium on the Archaeology of Formative Ecuador, Dumbarton
Oaks, Washington, DC, October 7–8, 1995.
Marcos, J. G. (1978). The Ceremonial Precinct at Real Alto: Organization of Time and Space
in Valdivia Society, PhD dissertation, Department of Anthropology, University of Illinois,
Urbana.
Marcos, J. G. (1979). Woven textiles in a late Valdivia context (Ecuador). In Rowe, A. P., Benson,
E. P., and Schaffer, A. L. (eds.), The Junius B. Bird Pre-Columbian Textile Conference,
Dumbarton Oaks, Washington, DC, pp. 18–26.
Marcos, J. G. (1982). Los Morros. In Marcos, J. G., and Norton, P. (eds.), Primer Simposio
de Correlaciones Antropológicas Andino-Mesoamericano, Salinas, Ecuador, Junio, 1971,
ESPOL, Guayaquil, Ecuador, pp. 177–201.
Marcos, J. G. (1987). Los campos elevados de la cuenca del Guayas, Ecuador. In Denevan,
W., Mathewson, K., and Knapp, G. (eds.), Pre-Hispanic Agricultural Fields in the Andean
Region, British Archaeological Reports, International Series 359, Oxford, pp. 217–224.
Marcos, J. G. (1988a). Real Alto: la historia de un centro ceremonial Valdivia, Biblioteca Ecuato-
riana de Arqueologı́a, Primera Parte ESPOL Guayaquil, Corporación, Editora Nacional.
Marcos, J. G. (1988b). Real Alto: la historia de un centro ceremonial Valdivia, Biblioteca
Ecuatoriana de Arqueologı́a, Segunda Parte ESPOL Guayaquil, Corporación Editora
Nacional.
Marcos, J. G. (1989). Informe arqueológico. In Marcos, J. G., and Alvarez, S. (eds.), Proyecto
San Lorenzo del Mate, Fundación Pedro Vincente Maldonado, Guayaquil, pp. 1–20.
Marcos, J. G., and Garcia de Manrique, M. (1988). De la Dualidad Fertilidad-Virilidad a lo
Explicitmente Femenino o Masculino: La Relacion de las Figurinas con los Cambios en la
Organización Social Valdivia, Real Alto, Ecuador. In Miller, V. (ed.), The Role of Gender
in Pre-Columbian Art and Architecture, University Press of America, Lanham, MD, pp. 35–
51.
Marcos, J. G., Lathrap, D. W., and Zeidler, J. E. (1976). Ancient Ecuador revisited. In Field
Museum of Natural History Bulletin 47(6), The Field Museum, Chicago, pp. 3–8.
Marcos, J. G., and Norton, P. (1981). From the Yungas of Chincay Suyo to Cuzco: The role
of La Plata Island in Spondylus trade. In Browman, D., Burger, R. L., and Rivera, M. A.
(eds.), Social and Economic Organization in the Prehispanic Andes, British Archaeological
Papers, International Series no. 194, Oxford, pp. 7–20.
Marcos P., J. G., Pérez, A., and Spinolo, G. (1999). La producción durante el Formativo tem-
prano: El desarrollo agrı́cola, artesanal y el intercambio de exóticos en Real Alto. In
Ledergerber-Crespo, P. (ed.), Formativo Sudamericano Una Revaluación., Producciones
Digitales UPS, Abya-Yala Editing, Quito Ecuador, pp. 97–113.
McPhaden, M. J. (1999). Genesis and evolution of the 1997–1998 El Niño. Science 283: 950–954.
Meggers, B. J. (1966). In Daniel, G. (ed.), Ecuador, Ancient peoples and places series, Vol. 49,
Praeger, New York.
Meggers, B. J. (1971). Amazonia: Man and Culture in a Counterfeit Paradise, Aldine-Atherton,
Chicago.
Meggers, B. J. (1972). Prehistoric America, Aldine-Atherton, Chicago.
Meggers, B. J. (1987). El origen transpacifico de la cerámica Valdivia: Una revaluación. Boletı́n
del Museo Chileño de Arte Precolombino 2: 9–31.
Meggers, B. J., and Evans, C., Jr. (1962). The Machalilla culture: An Early Formative complex
on the Ecuadorian coast. American Antiquity 28(1): 186–192.
Meggers, B. J., and Evans, C., Jr. (1966a). A transpacific contact in 3000 B.C. Scientific American
214: 28–35.
Meggers, B. J., and Evans, C., Jr. (1966b). Transpacific origin of Valdivia Phase pottery on coastal
Ecuador, 36th Congreso Internacional de Americanistas Actas y Memorias, Vol. 1, Sevilla,
Spain, pp. 63-67.
Meggers, B. J., Evans, C., Jr., and Estrada, V. E. (1965). Early Formative of Coastal Ecuador:
The Valdivia and Machalilla Phases, Smithsonian Contributions in Anthropology, Vol. 1,
Smithsonian Institution Press, Washington, DC.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 251

Moseley, M. E. (1983). The good old days were better: Agrarian collapse and tectonics. American
Anthropologist 85: 773–779.
Moseley, M. E., Feldman, R. A., and Orcliff, C. R. (1981). Living with crisis: Human perception
of process and time. In Biotic Crises in Ecological and Evolutionary Time, Academic Press,
New York, pp. 231–267.
Murra, J. (1946). The tribes of Ecuador. In Steward, J. H. (ed.), Handbook of South American
Indians, Vol. 2, Bulletin 143, Bureau of American Ethnology, Smithsonian Institution Press,
Washington, DC, pp. 785–822.
Muse, M. C. (1989). Cronologı́a regional y relaciones de territorialidad en al región oeste del
área septentrı́onal andina, siglo XVI. In Bouchard, J. F., and Guinea, M. (eds.), Relaciones
interculturales en el área ecuatorial del pacifico durante la época precolumbino, British
Archaeological Reports, International Series 503, Oxford, pp. 187–197.
Norton, P. (1972). Early Valdivia middens at Loma Alta, Ecuador, Paper presented at
the 37th Annual Meetings of the Society for American Archaeology, Bal Harbour,
ME.
Norton, P. (1977). The Loma Alta connection, Paper presented at the 42nd Annual Meeting of
the Society for American Archaeology, New Orleans, LA.
Norton, P. (1981). The Chiefdom of Canlangone and the League of Merchants: The Spondylus—
Balsa Wood Cartel, Paper presented at the 80th Annual Meeting of the American Anthro-
pological Association, Los Angeles, CA.
Norton, P. (1982). Preliminary observations on Loma Alta, an early Valdivia midden in Guayas
Province, Ecuador. In Marcos, J. G., and Norton, P. (eds.), Primer Simposio de Correlaciones
Antropológicas Andino-Mesoamericano, Salinas, Ecuador, ESPOL, Guayaquil, Ecuador,
pp. 101–120, Junio, 1971.
Parducci, R., and Parducci, I. (1975). Vasijas y elementos diagnosticos, Fase Guayaquil. In
Cuadernos de Historia y Arqueologı́a, Vol. 42, Casa la Cultura, Guayaquil, Ecuador,
pp. 155–284.
Parker, T. A., III., and Carr, J. L. (1992). In Parker, T. A., and Carr, J. L. (eds.), Status of
forest remnants in the Cordillera de la Costa and adjacent areas of Southwestern Ecuador,
Rapid assessment program (RAP working papers 2) Conservation International, October
1992.
Paulsen, A. C. (1974). The Thorny Oyster and the voice of God: Spondylus and Strombus in
Andean Prehistory. American Antiquity 39: 597–607.
Paulsen, A. C., and McDougle, E. J. (1974). The Machalilla and Engoroy Occupations of the
Santa Elena Peninsula in South Coastal Ecuador, Paper presented at the 39th Annual
Meeting of the Society for American Archaeology, Washington, DC.
Paulsen, A. C., and McDougle, E. J. (1981). A Chronology of Machalilla and Engoroy Ceramics
of the South Coastal Ecuador, Paper presented at the 9th Annual Midwest Conference on
Andean and Amazonian Archaeology and Ethnohistory, Columbia, Missouri.
Pearsall, D. M. (1978). Phytolith analyses of archaeological soils: Evidence of maize cultivation
in Formative Ecuador. Science 199: 177–178.
Pearsall, D. M. (1979). The Application of Ethnobotanical Techniques to the Problem of Sub-
sistence in the Ecuadorian Formative, PhD dissertation, Department of Anthropology,
University of Illinois, Urbana.
Pearsall, D. M. (1991). Prehistoric Subsistence and Agricultural Evolution in the Jama River
Valley Manabı́ Province, Ecuador, Paper presented at the 56th Annual Meeting of the
Society for American Archaeology, New Orleans, LA.
Pearsall, D. M. (1992). The origins of plant cultivation in South America. In Wesley Cowan,
C., and Watson, P. J. (eds.), with the assistance of Benco, N. L., The Origins of Agriculture:
An International Perspective, Smithsonian Institution Press, Washington, DC and London,
pp. 173–205.
Pearsall, D. M. (1993). Contributions of phytolith analysis for reconstructing subsistence: Ex-
amples from research in coastal Ecuador. In Pearsall, D. M., and Piperno, D. R. (eds.),
Current Research in Phytolith Analysis: Applications in Archaeology and Paleoecology,
MASCA, University of Pennsylvania Museum, pp. 109–122.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

252 Staller

Pearsall, D. M. (1994). Issues in the analysis and interpretation of archaeological maize in South
America. In Johannesson, S., and Hastorf, C. A. (eds.), Corn and Culture in the Prehistoric
New World, Westview Press, Boulder, pp. 245–272.
Pearsall, D. M., and Piperno, D. R. (1990). Antiquity of maize cultivation in Ecuador: Summary
and reevaluation of the evidence. American Antiquity 55(2): 324–337.
Piperno, D. R. (1991). The status of phytolith analysis in the American tropics. Journal of World
Prehistory 5(2): 155–191.
Piperno, D. R. (1994). On the Emergence of Agriculture in the New World. Current Anthro-
pology 35(5): 637–643.
Porras-Garces, P. I. (1973). El Encanto-La Puná: Un sitio insular de la fase Valdivia asociado
con una conchero anular, Quito., Ediciones Huancavilca 5, Serie La Puná.
Raymond, J. S. (1988). Subsistence patterns during the Early Formative in the Valdivia Val-
ley, Ecuador. In Kennedy, B. V., and LeMoine, G. M. (eds.), Diet and Subsistence: Cur-
rent Archeological Perspectives, Archaeological Association of The University of Calgary,
Calgary, pp. 159–164.
Raymond, J. S. (1989). Early Formative Societies in the Tropical Lowlands of Western Ecuador:
A View From the Valdivia Valley, Paper presented at the Circum-Pacific Prehistory Con-
ference, Seattle, Washington, August 1989.
Raymond, J. S. (1993). Ceremonialism in the Early Formative of Ecuador. In Millones, L.,
and Onuki, Y. (eds.), El Mundo Ceremonial Andino, Senri Ethnological Studies, Vol. 37,
National Museum of Ethnology, Osaka, Japan, pp. 25–43.
Raymond, J. S., Marcos, J. G., and Lathrap, D. W. (1980). Evidence of formative settlement in
Guayas Basin, Ecuador. Current Anthropology 21: 700–701.
Raymond, J. S., Oyuela-Caycedo, A., and Carmichael, P. (1991). A Comparison of the Earliest
Ceramic Technologies of Ecuador and Columbia, Paper presented at the 56th Annual
Meeting of the Society for American Archaeology, New Orleans, LA.
Reindel, M., and Guillaume-Gentil, N. (1995). El proyecto arqueologico La Cadena: Estudios
sobre la secuencia cultural de la cuenca del Rı́o Guayas. In Alvarez, A., Alvarez, S., Fauria,
C., and Marcos, J. G. (eds.), Primer encuentro de investigadores de la costa Ecuatoriana
en Europa, Arqueologı́a, Ethnohistorı́a, Antropologı́a Sociocultural, Abya-Yala, Quito,
pp. 143–177.
Rollins, H. B., Sandweiss, D. H., Richardson, J. B., III. (1987). Growth increment and stable
isotope analysis of marine bivalves: Implications for the geoarchaeological record of El
Niño. Geoarchaeology 2(3): 81–197.
Rowe, J. H. (1946). Inca culture at the time of the Spanish Conquest. In Steward, J. (ed.),
Handbook of South American Indians, Vol. 2, Bulletin 143. Bureau of American Ethnology,
Smithsonian Institution, Washington D.C., pp. 183–330.
Rowe, J. H. (1960). The classification of artifacts in archaeology. American Antiquity 25(3):
313–323.
Rowe, J. H. (1961). Stratigraphy and seriation. American Antiquity 26(3): 324–330.
Sagastegui-Alva, A., Dillon, M. O., Sánchez-Vega, I., Leiva-González, S., Lezama-Asencio, P.
(1999). Diversidad Floristica del Norte de Perú, Tomo I, World Wildlife Fund, Peru.
Salazar, E. (1998). De vuelta al Sangay investigaciones arqueológicas en el Alto Upano
Amazonia Ecuatoriana. Bulletı́n de Instituto Frances Etudes Andines 27(2): 213–240.
Salomon, F. L. (Fall 1977/Spring 78). Pochteca and Mindalá: A Comparison of Long-Distance
Traders in Ecuador and Mesoamerica. Journal of Steward Anthropological Society 9(1/2):
231–248.
Salomon, F. L. (1985). The dynamic potential of the Complimentary Concept. In Masuda, S.,
Shimada I., and Morris, C. (eds.), Andean Ecology and Civilization: An interdisciplinary
Perspective on Andean Ecological Complimentarity, University of Tokyo Press, Tokyo,
pp. 511–531.
Salomon, F. L. (1986). Native Lords of Quito in the Age of the Incas: The Political Economy of
North Andean Chiefdoms, Cambridge University Press, Cambridge.
Sandweiss, D. H. (1986). The beach ridges at Santa Peru: El Niño, uplift, and prehistory. Geoar-
chaeology 1(1): 17–26.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 253

Saville, M. H. (1909). Archaeological researches in the coast of Esmeraldas. Sixteenth Congress

of the Americanists, Vienna, pp. 337. ff.
Saville, M. H. (1910). Antiquities of Manabı́, Ecuador Volume II, Contributions to South Amer-
ican Archaeology, Heye Foundation, New York, NY.
Saville, M. H. (1924). The gold treasure of Sigsig, Ecuador. Leaflets of the Museum of the
American Indian 3.
Schwarz, F. A., and Raymond, J. S. (1996). Formative settlement patterns in the Valdivia Valley,
SW coastal Ecuador. Journal of Field Archaeology 23: 205–224.
Simmons, M. P. (1970). The Ceramic Sequence From La Carolina, Santa Elena Peninsula,
Ecuador, PhD dissertation, Department of Anthropology, University of Arizona, Tucson.
Smith, B. D. (1998). Emergence of Agriculture, Scientific American Library Publication,
Freedman, New York.
Spath, C. D. (1980). The El Encanto Focus: A Post-Pleistocene Maritime Adaptation to Ex-
panding Littoral Resources, PhD dissertation, Department of Anthropology, University of
Illinois, Urbana.
Stahl, P. W. (1984). Tropical Forest Cosmology: The Cultural Context of the Early Valdivia
Occupations at Loma Alta, PhD dissertation, Department of Anthropology, University of
Illinois, Urbana.
Stahl, P. W. (1985). Native American cosmology in archaeological interpretation: Tropical forest
cosmology and the Early Valdivia phase at Loma Alta. In Thompson, M., Garcia, M. T.,
and Kense, F. J. (eds.), Status, Structure and Stratification: Current Archaeological Recon-
structions, Proceedings of the 16th Annual Conference, The Archaeological Association
of the University of Calgary, Calgary, Alberta., pp. 31–37.
Stahl, P. W. (1986). Hallucinatory imagery and the origin of early South American figurine art.
World Archaeology 18(1): 134–150.
Stahl, P. W. (1991). Arid landscapes and environmental transformations in ancient southwestern
Ecuador. World Archaeology 22: 346–359.
Stahl, P. W., and Zeidler, J. A. (1990). Differential bone refuse accumulation in food preparation
and traffic areas on an early Ecuadorian house floor. Latin American Antiquity 1(2): 150–
169.
Staller, J. E. (1992/93). El sitio Valdivia tardio de la Emerenciana en la Costa sur del Ecuador y
su significación del desarrollo de complejidad en la costa oeste de Sudamerica. Cuadernos
de Historia y Arqueologia (Parte 1) 46–47: 14–37.
Staller, J. E. (1994). Late Valdivia Occupation in El Oro Province Ecuador: Excavations at
the Early Formative Period (3500–1500 B.C.) site of La Emerenciana, PhD dissertation,
Department of Anthropology, Southern Methodist University Dallas, Texas.
Staller, J. E. (1996a). El sitio Valdivia tardio de la Emerenciana en la Costa Sur del Ecuador y
su significación del desarrollo de complejidad en la costa oeste de Sudamerica. Cuadernos
de Historia y Arqueologia (Parte 2) 48–50: 65–118.
Staller, J. E. (1996b). An Analysis of Cosmological and Gender Related Stylistic Attributes
in the Jelı́ Phase: A Phase VII–VIII Valdivia Ceramic Complex From La Emerenciana,
Paper presented at the 51st Annual Northeast Conference on Andean Ethnohistory
and Archaeology, Rainey Auditorium, University of Pennsylvania Museum, Philadelphia,
Pennsylvania.
Staller, J. E. (1998). Feline Symbolism in Valdivia Phase VIII Figurines From the Site of San
Lorenzo del Mate, Paper presented at The Institute of Andean Studies 38th Annual Meet-
ing, The University of California, Berkeley.
Staller, J. E. (2000a). Political and prehistoric frontiers: How history influences our understand-
ing of the past. In Boyd, M., Erwin, J. C., and Hendrickson, M. (eds.), The Entangled Past:
Integrating History and Archaeology, Proceedings of the 30th Annual Chacmool Con-
ference Calgary, Alberta, The Archaeological Association of The University of Calgary,
Alberta, pp. 242–258.
Staller, J. E. (2000b). Figurinas Valdivia VII–VIII del sitio San Lorenzo del Mate, provincia
del Guayas, y la transición Valdivia–Machalilla. Miscelánea Antropológica Ecuatoriana 9:
99–133.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

254 Staller

Staller, J. E. (2001). The Jelı́ Phase Complex at La Emerenciana, a late Valdivia site in southern
El Oro Province, Ecuador. In Andean Past, Vol. 6, Cornell University, Latin American
Studies Program, pp. 117–174.
Staller, J. E., and Thompson, R. G. (in press). A multidisciplinary approach to understanding
the initial introduction of maize into coastal Ecuador. Journal of Archaeological Science.
Staller, J. E., Tykot, R. H., and Thompson, R. G. (2000). A Multidisciplinary Approach to Explor-
ing the Early Introduction and Role of Maize in Western South America, Paper presented
at the 27th Annual Midwestern Conference on Andean Ethnohistory and Archaeology,
University of Central Michigan, Mt. Pleasant.
Stirling, M. W. (1963). A new culture in Ecuador. Archaeology 16(3): 170–175.
Stothert, K. E. (1976). The early prehistory of the Santa Elena Peninsula: Continuities between
the preceramic and ceramic cultures, Actas del XLI Congreso Internacional de American-
istas, Vol. 2, Mexico, pp. 88–98.
Stothert, K. E. (1983). Review of the Early Preceramic Complexes of the Santa Elena Peninsula,
Ecuador. American Antiquity 48(1): 122–127.
Stothert, K. E. (1985). The Preceramic Las Vegas Culture of Coastal Ecuador. American
Antiquity 50(3): 613–637.
Stothert, K. E. (1995). Las Albarradas tradicionales y el manejo de aguas en la Peninsula de
Santa Elena. Miscelánea Antropológica Ecuatoriana 8: 131–160.
Svenson, H. K. (1946). Vegetation of the Coast of Ecuador and Peru and its Relationship to That
of the Galapagos Islands, Contribution No. 104 Brooklyn Botanic Garden, New York.
Thompson, R. G., and Staller, J. E. (2001). An analysis of opal phytoliths from food residues of
selected sherds and dental calculus from excavations at the site of La Emerenciana, El Oro
Province, Ecuador. Phytolitharien, Bulletin of the Society for Phytolith Research 13(2/3):
8–16.
Uhle, M. (1920a). Los principios de las antiguas civilizaciónes peruanas. Boletı́n de la Sociedad
Ecuatoriana de Estudios Históricos 4(12): 448–458.
Uhle, M. (1920b). Los principios de la civilización en la sierra peruana. Boletı́n de la Sociedad
Ecuatoriana de Estudios Históricos 4(13/14): 44–56.
Uhle, M. (1922a). Sepulturas ricas de oro en la provincı́a del Azuay. Boletı́n de la Academia
Nacional de Historia 4(9): 108–114.
Uhle, M. (1922b). Las Huacas de Cañar. Journal de la Société des Américaniste de Paris 14:
242–244.
Uhle, M. (1922c). Influencias mayas en el alto Ecuador. Boletı́n de la Academia Nacional de
Historia 5(10–11): 205–240 and 5(12, 13, and 14): 1–3.
Uhle, M. (1923). Civilizaciones mayiodes de la costa pacifica de Sudamérica. Boletı́n de la
Academia Nacional de Historia 6(15, 16, and 17): 87–92.
Uhle, M. (1924). Cronolologı́a y relaciones de las antiguas civilizaciones suramericas. Boletı́n
de la Academia Nacional de Historia 9(24, 25, and 26): 190–207.
Uhle, M. (1937). Las ruinas de Cochasquı́. Revista del Museo Nacional de Lima 6: 86–91.
Valdez, F. (1987). Proyecto Arqueologico “La Tolita” 1983–1986, Fondo Arqueológico del
Museo del Banco Central “Guillermo Pérez Chiriboga,” Quito, Ecuador.
Valdez, F. (1992). The Tolita ceramic tradition. In Valdez, F., and Veintimilla, D. (eds.),
Amerindian Signs 5000 years of Precolumbian Art in Ecuador, Ediciones Colibri, Paris,
pp. 135–189.
van der Merwe, N., Lee-Thorp, J. A., and Raymond, J. S. (1993). Light, Stable Isotopes and the
subsistence base of formative cultures at Valdivia, Ecuador. In Lambert, J. B., and Grupe,
G. (eds.), Prehistoric Human Bone Archaeology at the Molecular Level, Springer-Verlag,
New York, pp. 63–97.
Véliz-Mendoza, A. (1990). El Cacique Guayaquile: El Pueblo y el Rı́o de su nombre. Culturas de
la época a la llegada de los conquisdadores españoles, Fundacion Ecuatoriana de Estudios
Sociales Feso, Quito, Ecuador.
Villalba, M. (1988). Cotocollao: Una Aldea Formativa del Valle de Quito, Miscelanea
Antropológica Ecuatoriana, Serie Monográfica, Vol. 2, Museo del Banco Central del
Ecuador, Quito, Ecuador.
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

The Formative of Coastal Ecuador 255

Weinstein, E. (1998). Images of Women in Ancient Chorrera Ceramics: Mythological Continuity

Across 2 millennia in the Tropical Forest of South America, Paper presented at the 26th
Annual Midwest Conference on Andean and Amazonian Archaeology and Ethnohistory,
University of Illinois, Urbana.
Wernstedt, F. L. (1972). World Climatic Data, Climatic Data Press.
White, A. (1985). Herbs of Ecuador: Medicinal Plants, Edicions Libri Mundi, Quito, Ecuador.
Willey, G. R. (1962). The great art styles and the rise of Pre-Columbian civilization. American
Anthropologist 64(1): 1–14.
Willey, G. R. (1971). An Introduction to South American Archaeology, Vol. 2: South America,
Prentice-Hall, New York.
Willey, G. R., and Corbett, J. M. (1954). Early Ancón and Early Supe Culture. Chavin Horizon
Sites of the Central Peruvian Coast, Studies in Archaeology and Ethnology, Vol. 3, Columbia
University Press, New work.
Zedeño, M. N. (1985). Análisis de cerámica Chorrera del sitio Peñon del Rı́o, Thesis, Centro de
Estudios Arqueológicos y Antropólgicos, ESPOL, Guayaquil.
Zedeño, M. N. (1987). Análisis término diferencial de la cerámica Chorrera. Gaceta Ar-
queológica Andina 13: 8–9.
Zeidler, J. A. (1977). Early Formative Settlement in the Chanduy Valley, Southwestern Ecuador,
Paper presented at the 42nd Annual Meeting of the Society for American Archaeology,
New Orleans, LA.
Zeidler, J. A. (1984). Social Space in Valdivian Society: Community Patterning and Domestic
Structure At Real Alto, 3000–2000 B.C., PhD dissertation, Department of Anthropology,
University of Illinois, Urbana.
Zeidler, J. A. (1987). La evolución local de asentamientos Formativos en el litoral Ecuatoriano:
El caso de Real Alto. In Marcos, J. G. (ed.), La Arqueologı́a de la Costa Ecuatoriana:
Nuevos Enfoques, Corporación Editoria Nacional, Quito, Ecuador.
Zeidler, J. A. (1988). Feline imagery, stone mortars, and Formative Period interaction spheres
in the Northern Andean Area. Journal of Latin American Folklore 14(2): 243–283.
Zeidler, J. A. (1991). Maritime exchange in the Early Formative period of coastal Ecuador;
geopolitical origins of uneven development. Research in Economic Anthropology 13: 247–
268.
Zeidler, J. A. (1992). The Piquigua Phase: A Terminal Valdivia Occupation in Northern Manabı́,
Paper presented at the 57th Annual Meeting of the Society for American Archaeology,
Pittsburgh, PA.
Zeidler, J. A. (1994). Archaeological testing in the Lower Jama Valley. In Zeidler, J. A., and
Pearsall, D. M. (eds.), Regional Archaeology in Northern Ecuador, Vol. 1: Environment,
Cultural Chronology and Prehistoric Subsistence in the Jama River Valley, University of
Pittsburgh, Memoirs in Latin American Archaeology No. 8., Pittsburgh, pp. 1–12.
Zeidler, J. A., and Kennedy, R. C. (1994). Environmental Settings. In Zeidler, J. A., and Pearsall,
D. M. (eds.), Regional Archaeology in Northern Ecuador, Vol. 1: Environment, Cultural
Chronology and Prehistoric Subsistence in the Jama River Valley, University of Pittsburgh,
Memoirs in Latin American Archaeology No. 8., Pittsburgh, pp. 13–42.
Zeidler, J. A., and Pearsall, D. M. (1994). The Jama Valley Archaeological/Paleoethnobotanical
Project. In Zeidler, J. A., and Pearsall, D. M. (eds.), Regional Archaeology in Northern
Ecuador, Vol. 1: Environment, Cultural Chronology and Prehistoric Subsistence in the Jama
River Valley, University of Pittsburgh, Memoirs in Latin American Archaeology No. 8.,
Pittsburgh, pp. 1–12.
Zeidler, J. A., and Sutliff, M. J. (1994). Definition of ceramic complexes and cultural occupation
in the Jama Valley. In Zeidler, J. A., and Pearsall, D. M. (eds.), Regional Archaeology in
Northern Ecuador, Vol. 1: Environment, Cultural Chronology and Prehistoric Subsistence in
the Jama River Valley, University of Pittsburgh, Memoirs in Latin American Archaeology
No. 8., Pittsburgh, pp. 111–130.
Zeidler, J. A., Giauque, R. L., Asaro, F., and Stross, F. H. (1994). Trace element analysis of
obsidian artifacts from the San Isidro site. In Zeidler, J. A., and Pearsall, D. (eds.), Regional
Archaeology in Northern Manabı́, Ecuador, Vol. 1: Environment, Cultural Chronology and
P1: GKW/RKP P2: GKW
Journal of World Prehistory [jowo] PP270-346283 October 27, 2001 20:2 Style file version Nov. 19th, 1999

256 Staller

Prehistoric Subsistence in the Jama River Valley, University of Pittsburgh, Memoirs in Latin
American Archaeology No. 8, Pittsburgh.
Zevallos, M. C. (1965/66). Informe Preliminar sobre el Cementerio Chorrera, Bahı́a de Santa
Elena, Ecuador. In Revista del Museo Nacional, Vol. 34, Mexico City, pp. 20–27.
Zevallos, M. C. (1971). La Agrı́cola en el formativo temprano del Ecuador (Cultura Valdivia).
Talleres gráficos de la Casa la Cultura Ecuatoriana, Nucleo del Guayas, Guayaquil.
Zevallos, M. C., Galinat, W., Lathrap, D. W., Leng, E. R., Marcos, J. G., and Klumpp, K. M.,
(1977). The San Pablo Corn Kernel and its friends. Science 196: 385–389.
Zevallos-Menendez, C., and Holm, O. (1960). Excavaciones Arqueológicas en San Pablo, In-
forme Preliminar, Guayaquil.
Zuta, S., Einfield, D., Valdivia-Lagos, J. P., and Blandin, C. (1976). Aspectos fisicos del fenomeno
“El Niño” 1972–1973. United Nations Food and Agriculture Organization, Fisheries Reports
185: 3–61.