CH.!:i';IIIG AItIAL UTII.IZATION
PATTERilS AID THEIR IMPLICATIONS:
SOUTIlWEST ECUADOR (6500 B.C.-A.D. 1-:00)
KATIFLEEI N.'ARY YRIPJ
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
INI PARTIAL FULFILIlIElIT OF THiE REQUITE:1EPTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
In the course of this study numerous individuals have provided
guidance in the identification, analysis, and interpretation of the
material considered here. I would especially like to thank Elizabeth
Wing, my major professor, for always being available for consultation
and for reading several drafts of this report; laxine Margolis, Gerald
Milanich, William Maples, and David Webb, my committee members, for
reading ny dissertation and offering many helpful suggestions; Clifford
Evans, Donald Lathrap, Ronald Liptak, Gene NM-Dougle, Presley Norton,
Allison Paulsen, and Karen Stothert for making faunal collections from
coastal Ecudor available for analysis; Donald Lathrap and Jorge Mlarcos
for making it possible for me to visit sites in the area and collect
fishes; 'alter Auffenberg, Pierce Brodkorb, Carter Gilbert, and Fred
Thompson for aiding in the identification of some of The faunal remains
and suggesting useful readings; and my fellow students and the staff of
the Florida State Museum for their suggestions and encouragement.
TABLE OF CONTENTS
ACi OW EDG ENTS . . . . . . . . . . . ii
LIST OF TABLES . . . . . . . .. . . . . v
LIST OF FIGURES . . . . . . . . . . . .. . vii
ABSTRACT . . . . . . . . . . . . . . viil
I. INTRODUCTION . . . . . . . . .. .. . . 1
II. NIUTRiTIONAL IEEDS AiD CALORIC REQUIREMENTS . . . . 10
III. IETHODOLOGY . . . . . . . . . . . 1S
IV. CHANGING CLIMATE AND THE ECOLOGICAL SETTING . . . .. 27
The Santa Elena Peninsula. . .. . . . . . 27
Climatic Change and Prehistoric Occupation of the
Santa Elena Peninsula . . . . . . . .. 31
Su imary . . . . . . . . .. . . 33
V. F'Ui.AL /Il.LYSIS AiD RECOliSTrUCTIOI . . . . . . 39
Pre-Valdivia . . . . . . . . . . . 40
Valdiv.ia . . . . . . . . . . . . 50
Post-Valdivia . . . . . . . . .. . . 72
VI. DISCUSSION . . . . . . . .. . . . 81
Protein Scarcity and Protein Acquisition . . . . 8]
Changes in Protein Exploitation and Subsistence
Orientation . . . . . . . . . ... 83
Hunting and Fishing Methods . . . . . . .. 91
Human Behavioral Patterns . . . . . . . 93
Internreal Comparison . . . . . . . . .. 95
Sum: ary . . . ...... .. . . .. .... . 97
AFPPi.II : A ... .. . .. . . . .. ... .. . . 99
B . . . . . . . .. . . . . 132
C . . . . . . . .. .. ... . . . . 141
D ............... .............. 143
E . . . . . . . . . . . . . . 144
F . . . . . . . . . . . . . . 146
IIBLIOGJiJl'ilY . . 149
BIOG\RP;IICAI SKETCH . . . . . . 155
LIST OF TABLES
1. COMPARISON OF METHODS USED IN; ESTI'MATING LIVE EIGHT . . .
2. PERCENTAGE OF FOODS FROM AQUATIC AND TERRESTRIAL HABITATS I1MI.
3. PERCENTAGE OF FOODS FROM AQUATIC AID TERRESTRIAL -AkBITATS -
B IO; lASS
. . . . . . . . . . . 86
FAUNIAL LIST OGSE-80
FAULIAL LIST OGSE-33
FAULiAL LIST OGSE-63
FAUIIAL LIST OGSE-42
FAUIIAL LIST OGSE-62
FAUIIAL LIST OGSE-62C
FAUIL LIST OGSE-174
FAUNAL LIST VALDIVIA
FAUIIAL LIST LOMA ALTA
FAUNAL LIST LO;L\ ALTA
FAULil.L LIST REAL ALTO
FAUIA.L L. IST REAL ALTO
FAUNAL LIST REAL ALTO
FAUNAL LIST REAL ALTO
FAUNAL LIST REAL ALT':
FAUiIAL LIST REAL ALTO
FAUAI:L LIST REAL ALTO
FAUIIAL LIST F.C'dL ALTO
FAUNAL LIST REAL :.LTO
JIII . .
BURIAL LI .
STRUCTURE 8 -
23. FAUNAL LIST REAL ALTO 1JOn,-FF.A'IUPJ; flATERIAL . . . . . 123
24. FAUNAL LIST OGSE-46D I4iACHALII.LA
FAUNAL LIST OGSL-46D
FALUTAL LIST OGSE-46D)
FAUNAL LIST OGCH-20
FAUilh LIST CGSE-46U
FAUl!IAL LIST OGSE-41E
FOOL) VALUES OGSE-80
FOOD VALUES OGSE-63
FOOD VALUES OGSE-63
FOOD VALUES OGSE-62C
FOOD VALUES VALDIVIA
FOOD VALUES OCSE-46D
FOOD VALUES OCSE-46U
* . .
. . .
* . .
. . .
* . .
LIST OF FIGURES
SITE LOCATIONS WITHIN THE STUDY AREA . . . . . ... .
CLIMATE CHANGE: SiATA, ELENA PENINSULA . . . . ... 32
RELATIVE PERCENT OF PRINCIPAL VERTEBRATES OGSE-80 ..... 44
DISTRIBUTION OF HEIGHTS OF CAPTUFIED FISHES OGSE-80 ..... 45
RELATIVE PERCEIIT OF PRINCIPAL VERTEBRATES OCSE-63 . . .. 49
RELATIVE PERCENT OF PRINCIPAL VERTEBRATES OCSE-62 . . .. 55
DISTRIBUTION OF HEIGHTS OF CAPTURED FISHES OGSE-62 .... 56
RELATIVE PERCENT OF PRINCIPAL VERTEBRATES OGSE-62C ..... 57
DISTRIBUTION OF WEIGHTS OF CAPTURED FISHES OGSE-62C ..... 58
RELATIVE PERCENT OF PRINCIPAL VERTEBRATES VALDIVIA ... 61
DISTRIBUTION OF WEIGHTS OF CAPTJPRED FISHES VALDIVIA .... 62
P.ELATIVE PERCENT OF PRINCIPAL VERTEBRATES LO:-A ALTA .... 66
RELATIVE PERCENT OF PRINCIPAL VERTEBRAILS REAL ALTO (Mliddle). 69
RELATIVE PERCENT OF PRINCIPAL VERTEBRATES REAL ALTO (Late). 71
RELATIVE PERCENT OF PRINCIPAL VERTEBRATES OCSE-46D. . . 74
DISTRIBUTION OF WEIGHTS OF CAPTURED FISHES OCSE-46D . ... 75
RELATIVE PERCENT OF PRINCIPAL VERTEBRATES OGSE-46U . . 79
DISTRIBUTION OF WEIGHTS OF CAPTURED FISHES OGSE-46U ..... 80
AGRICULTURAL AND DEER HUNTING SEASONALITY . . . . ... 89
Abstract of Dissertation Presented to t:le Graduace Council
of the University of Florida in Partial Fulfillment of the
Fequircimnts for the Degree of Doctor of Philooopliy
CIlL'lGIi;G A11I-LUL UTILIZ.ATION
PATTERIJS A.TD THEIR IMPLICATIO:iS:
SOUTHWEST ECUADOR (6500 B.C.-A.D. 1400)
Kathleen Iary Byrd
Chairperson: Elizabeth S. Using
llajor Department: Anthropology
The purpose of this study is to determine subsistence practices
and related huri;n behavioral patterns for Valdivia Phase (3000 B.C.-
1500 B.C.) in:ihbitants of southwest Ecuador. This goal is accomplished
by an analysis of vertebrate, faunal remains and the application of
cultural, ecological research methods. A total of fifteen samples is
considered, including three pre-Valdivia, eight Valdivia, and four post-
Valdivia since. (6500 B.C.-A.D. 1400). For most of these sites faunal
lists with minimum numbers of individuals, nuriber of bone fragir.ents and
bone heights are included. In addition, biomi, s, edible meat, calories,
and protein estimates are computed for the principal sites. Based on
these analyses, questions concerning protein scarcity and protein acqui-
sition, changes in protein exploitation and subsistence orientation,
hunting aiid fishing methods, and human behavioral patterns for the
varcojs groups aru considered.
I I RODUCTIOU
1Ihatever is one's theoretical viewpoint, few. anthronologiscs today
would argue against a particular stage in cultural development in
which one or more cultural groups. shifted from a hunting-gathering
subsistence system to a sedentary one baseJ on agriculture. This
change in subsistence emphasis is an important shift because it laid
the foundations for later cultural evolution. With the domestication
of plants and an increasing reliance on cultigens, people become
more sedentary, populations increased, more complex: social, political,
and economic systems developed and in some areas urban centers and
civilizations arose. In the flew Uorld this initial shift from a
hunting-gathering econo:;,y to a sedentary agricultural one is referred
to as the Formative Stage.
The e:-act definition of the Formative Stage and chose traits that
are most diagnostic of it are the subjects of some debate. Gordon 2..
l1illey and Philip Phillips (195S:1.44) stress the presence of maize and/
or manioc agriculture an-' "... the successful socioeconomic integration
of such an agriculture into well-established sedentary village life" in
their definition of the Formative. James A. Ford indicates several
po.-sible o.cr-'.implifications in llilley and Phillip's definition and
offers a definition based more on certain artifact types. Ford (1969:5)
views the Formative
... as the 3000 years (or less in some regions)
during which the elements of ceramics, ground
stone tools, handmade figurines, and manioc and
maize agriculture were being diffused and welded
in the region extending from Peru to the eastern
1lhichever position is held, the salient points in each appear to be
the incorporation of agriculture and sedentism anJ their related cultural
characteristics into a new way of life. This new form sets the stage
for subsequent cultural evolution.
One of the earliest manifestations of the Formative Stage is the
Valdivia Phase of coastal Ecuador. This phase has received considerable
study in the last 25 years (Bischof and Gamboa 1972; Bushnell 1951;
Estrada 1956 and 1961; Hill 1966; Lanning 1976a; Lathrap and Ilarcos 1975;
rlegger.-, Evans, and Estrada 1965; Norton 1971; Paulsen 1971; Porras 1973;
Stothert 1974; Zvalnlos ;ieneendez 1970; Zevallos Menendez and Holm 3960).
For the most part, these studies have addressed themselves to the estab-
lishe2nt of the ceramic chronologies and they provide basic site
information as well as development of hypotheses concerning the origins
of the Formative of this region. Recently there has been a growing
interest in obtaining evidence of agriculture from these siLts. This line
of research has met with some success (Zevallos lienandez 1970) even
though nost plant remains--and thus direct evidence of agriculture--are
not wel] preserved in sites of this region.
.'ith few exceptions (Sar-ma 1974; Ileggers, Evans, and Estrada 1965)
little attention has been paid to the non-agricultural segment of
subsistence. The relatively high survival rate of animal bone provides
ample opportunities to study at least this aspect of the quest for food.
Detailed analysis of food bone refuse can provide information not only
on past dietary patterns, but also on the technology used to obtain the
a-iirals. Mihen bone analysis is coupled with human, ecological, research
methods, additional information on subsistence related, human behavioral
patterns may be revealed. This study attempts to arrive at a better
understanding of some Formative patterns through the analysis of the
vertebrate remains associated with eight Early ForTaitive (Valdivia)
sites in Guayas Province, Ecuador. To understand subsistence patterns
prior to the Early Formative, the vertebrate remains from three pre-
Veldivia sites are., analyzed. Past-Valdivia developments are indicated
by the remains from four additional sites (Fig. 1).
To achieve the aim of this study, i.e. to analyze the subsistence
practices and related human behavioral patterns of the Formative Valdivia
culture of the area, a modification of Julian II. Ste:rard's (1955)
cultural ecological procedures is applied to the archaeological material.
The natural environme-nt of the area and technology used to exploit the
food resources are examined.
Environments are not static, but are subject to change. Some
cnvironnen;al areas are relatively stable while others, due to their
location at the edge of two different and unstable climatic zones, are
particularly susceptible to environmental fluctuation. In all areas
changes in the climatic conditions can profoundly affect their animal
populations. Therefore, before any real understanding of resource use
within an area can be achieved, some attempt to reconstruct previous
envLronmental conditions is necessary.
Two lines of evidence can be used in attempting to deter.:ine the
technology employed to obtain the animals. First, the artifactual
remains themselves can be considered. Secondly, the animals fund in the
midden can he analyzed and, by using ecological and ethnological studies,
SITE LOCATIONS WITHIN THE STUDY AREA
FJ Pre-Voldivia Site
0 Valdivia Site
/' Post-Valdivia Site
r 1- 20 km
0 10 20 km
the '.uniing and fishing methods effective in catching these animals
suggested. Coupling the environmental reconstruction with an analysis
of subsistence-related technology indicates which of the available
rcsourcze were ui.sed and ho'.i the people might have obtained them.
Having reconstructed the enviornment and the subsistence technology
of the culture, the second step of Steward's procedure can be attempted.
In this step, the human behavioral patterns connected with particular
technologies that are effective in catching certain animals are
examined. One method of viewing this is by utilizing game theory.
Game theory studies on ethnographic populations (Davenport 1971;
Gould 1972) have revealed people's attempt to maximize returns while
minimizing the time, the energy, and the risk involved in obtaining them.
On any given day the subsistence strategies adopted take into account
the :mount of time arid energy that will be expended in attempting to
achieve a certain economic goal. For e:-ample, will it take more time
and energy to track, kill, butcher, and carry back to camp a large
mammal or will more of the desired food be obtained and less time and
energy expended by spending the day fishing? Is the risk of not
obtaining food greater if the person hunts or if he or she fishes? Will
more proteins and calories be obtained by hunting or by fishing? The
abundance, ease of capture and nutritional and energy values of the
various resources will determine which strategies or combination of
strategies are most effective in obtaining the needed foods. Rodents
might be abundant and by using traps they may be easy to capture, but
they are snail in size and provide little meat. Deer are less abundant,
harder -o capture, but for each successful hunt a greater volume of meat
is obtained. Fishl might be very abundant and easy to capture, but are
relatively small, and when compared with mammals, have a lo'. caloric and
pr':,t,:in content. The strategy chosen by a people on a particular day
takes these factors into consideration.
Subsistence strategies are closely related to the hunan behavior
patterns of a people. Some procurement : tecchniques require relatively
large numbers of people, while others are more successful if carried out
individually. For example, in the South American tropical forest where
species densities are lot:, the maximum terrestrial hunting returns for
a population, as a whole, occur if the hunters, individually or in groups
of two or three, exploit several different areas. Using this method the
hunters maximize the possibilities that at least one of the areas hunted
will provide some game.
In areas with large gregarious herds, as in the North American
Creat Plains, communal hunting provided maximum returns for the time and
energy expended. In this region individual hunters could kill only a
relatively few animals before the herd scattered. If, on the other
hand, a coma.unal drive and jump is practiced, a larger number of animals
can be obtained. The same principle is applicable to fishing. letting
and poisoning of waters, in which large densities of fishes occur, result
in greater returns if a number of people cooperate in the operation. Hook
and line fishing, because of the relatively lo:' densities of carnivorous
fishes, provides greater return if the fishermen distribute themselves
individually or in small groups over a wider area.
The third step in Steward's procedure involves determining the
extent to which subsistence-related, human behavioral patterns effect
other aspects of culture. There are m.ny methods of studying this
relationship. One technique of reconstructing past cultural patterns is
through ethnographic analogy. Ideally, by comparing archaeologically
reconstructed exploitative patterns with a series of well research
ethnographic samples, it would be possible to suggest certain subsistence-
related, prehistoric cultural patterns. Unfortunately, good ecologically-
oriented ethnographic studies of the subsistence patterns of a large
series of groups, relying on different subsistence bases, have yet to be
undertaken. Until this is done, detailed correlations between certain
susbistence patterns and other aspects of a culture cannot be attempted.
Nevertheless, some generalizations can be made. A comparison of two
groups of people--one which relies primarily on aquatic fish resources,
and the other which depends on terrestrial forms for their animal
protein, suggests some of these general correlations.
Yolanda Ilurphy and Robert F. 11urphy (1974) have worked among tw:o
groups of ilundurucu Indians in Brazil, one savanna dwellers and the
other riverbank inhabitants. Both groups rely primarily on slash and
burn agriculture for their caloric and carbohydrate needs. ;Although
the savanna group fish, most of their animal food comes from hunting.
Both individual andr comm;nnal hunting are practiced. Yields obtained by
individual hunters vary and when a large animal is caught the ilundurucu
share it with other families in the village. One of the central focuses
of the savanna Nundurucu is the men's house and all its social roles,
duties, and functions. murphy and Murphy (1974:223) believe that
"... the need for cooperation in hunting utilizes general human fears is
shaping the institution of the man's house."
The OLTler MI.r'idurucu group moved to the ri.'ers primarily to exploit
the rubber trees. liere they rely on aquatic protein resources,
principally fishes. These Mlundurucu do not share animal protein.
The very nature of fishinlg--the in.lividualitv, of
the activity, tile ease of the catch, the time
available and necessary, the size of tle fisl
itself--all militate against collectivization of
the catch. And, hunting, does little to promote
broader social cohesion (Murphy and Murphy 1974.
If the presence of institutions like mens' houses are causally
related to hunting conditions, i.e. low species density, high individual
risk, and large animal size and rapid spoilage, then, all other factors
being equal, groups living under these same conditions would be expected
to have similar functionally related institutions. In addition, certain
radisLribution channels would be anticipated. On the other hand, in
groups experiencing none of these pressures, i.e. fishing groups, an
institution of the men's house type would not be expected to occur, nor
would the same form of redistribution channels appear.
Elaborate and time consuming procurement or food production methods
also r.sggest certain social elements. Specialization requires an
exchange of materials in which the specialists are able to trade their
gcods for those that they are not able to obtain directly through their
i;n efforts. This leads to the developr.Ient of exchange systems and
Lthirr social and cultural ramifications.
In recent years, anthropologists have become concerned with various
points that Ste;iard did not treat, points that are revalent to the
present study (Vayda and Rappaport 1968). Particularly important among
these is the ecological dimension of hur.ian populations. ;iuman beings
do not li.'ve separated from all other living things, but are an active
component in the ecosystem and, as such, their very presence alters it.
People both effect and are affected by changes in the ecology of an
area. In subsistence-related Lerms, climatic changes can radically Lodi-
fy tile rypes and abundances of fod availability. A ricultural crops
thLit are gro.::t on rarginally productive lands are particularly sujcepjibl..
to linus l S freezes, diru.ghts, or floods. But hunting, fishing, and
a.tricultral tec.liCques c.so can alter an area. The plot cl.earin. and
peric-dic. shifts ini gardens practiced by slash and burn agriculturclists
contribute co the modification of the T-.viroimrnt. Tne :.echnizu-as
involved in slash end b-rn agriculture result in increasing forent-edge
conditions and therefore, species that prefer this type of habitat. At
the same tije,;e the area available fo- species that favor deep, undi.trubed
forests is decre-ised. Fish poiseni'.g 'OE pDc.!sI and acti'itiei such as
fire drives are other e:-:xmples of ecoloFical m.od!ifications by human
Sove l atiorns.
Recenc human ecological studie.; (lHarris 1965 Rappaport 196,-) have
sup- .orted Ste,'ird's basic assumption, i,'. tha;: there exists a causal
r;eaiio.;nhip betueer, basic suLibsitance sti-rategies and other aspects of
c-lture. A: successful subsistence str-ite-y is mandatory of a people are
t. survive. The pnrticuiar set. of stranegies adopted apiear to be
cauaally related to other a-pcects of a culituCe.
The remainder of this study applies these cultural, ecological
,tl:i.ods of analysis to the vertebra;e-related, subsiste-ce strategies
of the Valdivi'. Cultul.:, and Early For-aative manifestation of coastal
Ucuadcr. J'h- study attcm~tts to de-rive infonli ,ticn on certain animal-
rclated, sbsiU:-ter.e: techniqoues and Lhe corresponding human behavioral
pattei:.-s and to determiinc hoe' and why these techniques and patterns
c;iacgJd Lhrough timre.
NUTRITIONAL NELDS AID CALORIC REQUIREMIENITS
In any complete study of people and their relationship to their
environment, simply listing the resources utilized does not provide an
adequate description of the importance of th12 various foods in the diet.
ifiether a people relied primarily on cooperative net fishing or on
solitary hunting or some combination of these strategies, the relative
importance of the animals and of the methods adopted to obtain them
greatly effects subsistence-related, cultural i.anifestations. Therefo.:e,
in studying subsistence systems some idea of the relative quantity and
quality of the various foods in the diet of the people, and of the
strategies employed to acquire these foods, is needed. This necessitates
consideration of both nutritional requirements and caloric need,.
Without the right (in the nutritional sense) kinds of foods, a
people will cease to function and die. ilu-nans have learned, probably
through trial and error, that the combinations of certain foods and ihe
adoption of certain methods of obtaining such enabled them to be healthy
and to reproduce. Th2 foods eaten by a people and the methods or
strategies adopted to obtain these foods differ greatly from area to
area, but to remain healthy all populations must fulfill their basic
Uith respect to human growth and metabolism, food serves two basic
purposes. It provides the structural material used in growth and main-
tenance of the body and it furnishes the energy that is needed in
metabolism. The first of these is referred to as the quality of the
food, i.e. the foods' chemical ingredients; the second as the quantity
of the food, i.e. its energy content (Sebrell and Haggerty 1967).
For proper growth and development certain elements and compounds
must be available to an organism from its food supply. Humans need the
organic compounds of carbohydrates, fats, proteins, and vitamins and
certain inorgainc minerals.
Carbohydrates, composed of the elements carbon, hydrogen, and
oxygen, vary in coriplexity from simple three-carbon sugars to complex :
polymers (Pike and Brown 1967). Carbohydrates are divided into simple
compounds, the monosaccharides and disaccharides of the sugars, and
complex compounds, the polysaccharides of cellulose and starches (Arlin
1972). Since the human digestive system is only able to digest a limited
amount of cellulose and most is passed out of the body largely unchanged,
cellulose is largely unimportant in human nutrition. The polysaccharide
plant starches provide the principal energy source for most human
populations. So important, in fact, that peoples are often categorized
according to their starch food, e.g. rice growers or maize horticultur-
alists (Arlin 1972).
Lipids (or fats) are made up primarily of carbon and hydrogen.
The triglycerides, compounds of glycerol and three fatty acids, are
important in terms of nutrition. It is in these forms that energy is
stored by animals and, to a lesser extent by plants (Arlin 1972).
Certain lipids furnish an energy source for cells, others function as
structural compounds, and still others as hormones (Pike and Brown 1967).
Proteins are composed of the basic organic elements carbon,
hydrogen, and oxygen, but in addition also contain nitrogen and sulfur.
The basic structural units of proteins are the amino acids. The amino
acids contain an amino group (-llii) and an acid group and have side
chains which are responsible for Lthe various chemical properties of the
acids (Arlin 1972). Although there are only about 20 amino acids, the
combination of amino acids present in any particular compound, its
position in the molecule, and the spatial arrangement of the molecules,
result in thousands of different kinds of proteins (Pike and Bro;;n 1967).
All foods contain some protein, but both the amount present and the
proportion of the various amino acids vary from one protein source to
another. In human nutrition, protein foods are required in order for
the body to obtain the amino acids necessary for its own protein
synthesis ;which, in turn, is needed for growJth and maintenance. Only
when all the necessary amino acids are available will the synthesis of
a particular protein occur. The lac: of one of che needed amrLino acids
vill result in the termination of the construction of that particular
protein. The human body can manufacture most of the amino acids if
enough nitrogen is present, but since the only available source of
nitrogen is protein, protein is therefore a necessary food constituent.
In addition, there are eight amino acids, the essential amino acids,
that cannot be synthesized. These must be supplied in the diet if
normal protein manufacture is to occur (Arlin 1972).
The other two classes of nutrients necessary for humans are vitamins
and minerals. The human body requires vitamins in trace amounts for
health and growth. Vitamins are all organic chemicals, but are
ot:heruise unrelated. Some vitamins cannot be synthesized in adequate
amounts by cells and must be ingested. Minerals, inorganic chemicals,
are also essential in small quantities for normal body development
(Arlin 1972). Of the 16 essential mineral elements, calcium, phosphor's,
sodium, iron and potassium are required in greatest quantities.
In addition to furnishing the building material for the body, food
also provides the energy that is needed in metabolism. This energy
requirement is measured in kilogrumn calories (Kcals. or Cals.) and is
defined as the amount of heat required to elevate the temperature of one
kilogram of water one degree centigrade. lihen oxidized within the cell,
one gram of protein provides four calories; one gram of carbohydrates four
calories; and one gram of fat nine calories.
In general, plants manufacture carbohydrates, store excess energy
as starch, and rely on cellulose for structure. Animals store energy as
fat, snthasize very little carbohydrates and often depend on a cal-
careous skeleton for support. They also require large amounts of protein
in the form of muscles for locomotion (Arlin 1972). These muscles
consist primarily of protein and fat with a high proportion of water.
Heat also functions as a source of vitamins and minerals.
With the exceptions of mill: and liver, only plants provide carbo-
hydraLes. Animal sources of foods are usually high in fats since
animals store their energy in this form. The actual amounts of fat
vary according to the organism and its condition. Poultry, for example,
provides less fat by weight than beef. Host species of fish are also
relatively low in fats. Certain invertebrates, e.g. oysters, crabs,
shrimp, clams, and lobster, are essentially fat-free. Fats occur in sig-
nificant ainouunts in plant foods only in seeds, nuts and fruits (Arlin
1972). Protein occurs in all foods whatever their origins. Some protein
foods, however, have a higher quality or biological value based on the
efficiency in which their proteins are digested and absorbed, and the
proportions in which the essential amino acids are present. Although
animal protein is both more abundant per unit weight (e.g. per 100
grams) and has a higher quality or biological value than almost all
plan12 protein, 301 of the world's protein comes from cereal grains and
/i40 from other plant sources. Since cereal grains are structured to provide
a complete food source for the sprouting plants, they contain starch,
protein, vitamins and minerals needed for growth of the plant. The primary
purpose of tubers and roots, on the other hand, is to store energy and
they do this in their starchy underground structures. This is an
important distinction when comparing the relative value of these two
food sources (e.g. wheat is about 12, protein, rice 8.7 and the potatoes
and nanioc contain only 2% or less (Arlin 1972).
Uith respect to protein, nutritional needs can be met in three ways.
First, a person can consume large amounts of food. This is the method
adopted by many rice-eating peoples. By consuming up to one pound of
raw rice per day a person can obtain 30 to 35 grams of protein.
In addition, rice is fairly adequate vith respect to amino acids.
Maize, on the other hand, is so deficient in sone essential amino acids
that no matter what volume is consumed it alone can never furnish the
protein necessary for hunan growth and maintenance. Most types of maize
Jack the vitamin niacin and the amino acids lysine and tryptophan (Arlin
Another method for obtaining an adequate amount of protein and
amino acids consists of adding a small amount of animal protein to the
diet. A small amount of meat or fish added to rice, beans or corn will
supplement the cereal protein to such a degree that "... it will
adequately sustain an individual of small stature" (Arlin 1972:242).
The third method that can be used to meet minimum. protein require-
ments involves the use of complementary vegetable proteins. For example,
the amino acids in cereals and legumes supplement each other and together
provide the essential amino acids needed. The presence of large pop-
ulations in Latin America, who live principally on a corn-bean diet
illustrates this third method.
Although three alternative methods of obtaining adequate proteins
are theoretically possible, not all of these methods are possible
alternatives for a given people living in a particular setting. Environ-
mental or ecological factors, combined with the level of technological
developments, favor the utilization of certain methods and preclude
others. In most cases, the most efficient way to fulfill nutritional
needs is through a combination of carbohydrate-rich plant foods and
protein-rich animal sources with both animals and certain plant parts
providing the fats needed.
IIct all the nutritional and caloric parameters of a prehistoric
diet car, be quantified. Some dimensions are more amenable to this
type of analysis than others. iiethods are no,. being developed to
ascertain the relative importance of plant and animal foods in the diet
of prehistoric populations (Crown 1973). The analysis of trace elements
in human bone provide the data base for this type of study. The techniques
used in trace elemental analysis are still in the process of being refined
and, unfortunately, could not be applied to the material from the
sites considered in this study.
Data from the animal remains from archaeological middens are more
readily available for a quantitative approach. The presumed nutritional
value of these remains can be viewed in two principal ways: the degree
to which they furnish the necessary calories or energy units for human
populations, and the degree to which they provide the required proteins.
An energetic or caloric view of an ecosystem furnishes the oppor-
tunity to see the system as a wholc. Since energy functions as a common
denominator for all trophic levels, a caloric approach to an ecosystem
provides an opportunity to view the net gains and losses for each element
of the entire system and is ideal for studying all parameters of the food
web. This method has a wide range of application, including politics,
economics, and religion (Odum 1971). Energy, however, is difficult to
measure. Even for a numerically mrall segment of the system--human pop-
ulations-many difficulties arise in attempLing to obtain adequate caloric
measuremencs. In addition, for a complete ecosystemic study, information
on all trophic levels is needed. This approach is not suitable when
remains from only one part of the food jweb is available. It is
important, however, not to lose sight of this energetic aspect of
subsistence and its ramifications.
Protein functions as one of the basic nutrients and as such can
act as a limiting factor in population growth and culture development
(Carneiro 1961; Gross 1975). Protein can be measured and is amenable to
study based oni zooarchaeological data. It, like the caloric approach,
possesses some inherent drawbacks. ilost studies on the importance of
protein, have been carried out on United State populations under optimum
conditions. The requirements of prehistoric peoples conceivably could
have been different. Also, protein quality can deteriorate uith cooking,
but the rate is not constant. Considerable error could be introduced
if the zooarchaeological remains are viewed as raw, baked, or boiled
meat. The protein approach in analysing arcliaeological food bone does
furnish daca on the relative importance of various foods utilized to
pr-ovi e this basic nutrient.
Inhen calories and protein values of animals are considered together,
certain differences appear. In some cases certain animals \:ill provide
pr-oportionately more protein, but fe:jer calories, than another group of
animals. Ihen vietving archaeological food rei.ains quantitatively, it
should be remembered that calories and protein serve two, very different,
functions in the body. Both are required.
Since protein is an essential nutritional requirement for growth
and development, its consumption and the methods used to obtain it are
an important human activity. Therefore, a study of protein foods
utilized provides significant data about a culture. To study this
aspect of Formative cultural manifestations in southwestern Ecuador, bone
refuse from eight sites of the Valdivia Phase is considered (Fig. 1).
Four of these sites are located on the Santa Elena Peninsula (OGSE-174,
OGSE-62, OGSE-62C, OGSE-42). Two other sites are situated farther north
along the Valdivia River, one at its mouth (Valdivia) and the other
about 15km upstream (Loma Alta). The seventh site, which because of
its two cultural divisions is considered as two sites, is east of the
Santa Elena Peninsula and five k:. upstream from Chanduy on the Rio
Verde (Real Alto). All sites are located in Cuayas Province, Ecuador.
These eight sites form the data base for the following reconstruction.
Seven additional sites are also treated here. These sites provide
a longer time frame and are included to indicate changing exploitation
patterns from pre-Valdivia through post-Valdivia times. Three of these
sites are pre-Valdivia (OGSE-80, OGSE-3S, OGSE-63) and four post-Valdivia
(OGSE-460, OGSE-46U, OGSE-41E, OGC11-20). Three are located on the
Santa Elena Peninsula, with the fourth being eastward along the Rio
The faunal bone samples fror most of these sites are small and,
unless the field archaeologists indicated some anomaly, all the material
from a site is treated as one unit. In ti.o cases the excavations
revealed heterogeneous distributions of artifactual materials. For this
reason the Loma Alta sample is treated as two units, JII and JIII.
The presence of wall-trenches, pits, and burials at Real Alto necessitated
the analysis of this site in a number of discrete units.
Certain types of error are inherent in any method of analysis used.
Although many of these errors can be minimized by careful processing of
the materials, some sources of error remain. A cognizance of these
possible sources of inaccuracy is necessary to avoid misinterpretation.
In zooarcliaeological analysis, the error sources can be divided into
four types: initial deposition practices, post-depositional and pre-
excavational factors, excavation techniques, and analytical inaccuracies.
The way a people butchered their meat, cooked and served the food
and disposed of the refuse all effect the bone remains found in the site.
The practice of butchering in specialized areas or butchering large
animals at the kill site and returning to camp only parts of the carcass,
bias the sample. The location of the test pits in butchering areas
can result in a very different faunal reconstruction than the analysis of
other refuse materials. Food preparation techniques can also affect the
bone remains. Long periods of roasting or boiling of entire carcasses
or joints of great can weaken the structure of the organic constituents
of the bone and decrease its survival time (Chaplin 1971). Also dietary
practices such as consuming small animals whole, e.g. sardines or an-
chovies, or the grinding of Lhe bone into meal may eliminate material
from analysis. The disposal of the bone after cons-umption can result
in a further uneven distribution of the material. Large bones may have
been eliminated from the refuse areas by their use as raw materials in
t;'e manufacture of utilitarian, ritual, or decorative objects.
Secondly, even after deposition, the bones are still susceptible
to destruction through the activity of rodents and carnivores and by
weathering factors and soil conditions. The o.'erall effect of these
various factors and conditions vary from modifying the faunal composition
of the sample radically to causing very little change in the midden bones.
The third source of error, a controllable one, concerns the recovery
techniques. All too often excavators keep only certain bones, or, if
they use a screen, use one with so large a mesh size that it results
in the loss of many otherwise recoverable bones. These small, seemingly
insignificant bones often supply very detailed climatic information,
and through analysis of habit and habitat of the species represented,
may provide informative dati on procurement patterns and practices.
Finally, once back in the lab the level of identification depends
on the comparative material available for consultation. Especially in
areas here the taxonomy of the animals concerned has not been fully
refined, this level of analysis can result in inaccurate identification.
W\ichout adequate comparative materials many otherwise identifiable bones
can only be assigned to relatively high taxons, such as orders or
families. This is unfortunate, since identification to species level
for animals whose habits and habitats are :ell known can provide
detailed information on various aspects of a people's e:.ploitative
In addition to identification, the analytical methods used can
result in erroneous reconstructions. Especially susceptible to error
are the methods used to determine the relative numbers and importance
of the species represented in the sample. Three methods are widely used
in the determination of the relative numbers of species and their
importance in the diet (ChlaplirL 1971); the minimum number of individuals
(I. il) method, the fragment method, and the icei.ht method. All three
methods contain some inherent problems, but, for a number of reasons,
the IllI method is used here (Appendix A). This method simply tabulates
for each tocal sample the most often recurring bone of a species, i.e.
four, distal, right humerii of deer represent four deer. Nevertheless,
the number of fragments and the wJeights represented by the various
species in all the samples, except Valdivia, are tabulated in the
Appendi::. These are included to provide the data needed for those
wishing to use a different method. The bone from the Valdivia
site is mineralized and for this reason \jas not weighed.
Once the -U1I is determined, the biomass, the total live weight
represented by each species, genus, family or order, is calculated for
each of the seven principal sites (Appendix: B). Samples from the other
sites were either too small or the nature of the samples such that
Liomass estincmtes would be misleading.
Several methods hr.-e been developed to estimate the size of animals
represented by the archaeological remains- (Irhite 1953; Reed 1963; Casteel
1974; S3:!ith 1975; Wing 1976a). Each method has its limitations (Uing
1976a), but because Casteel's method appears to be the most accurate for
the typ-s of animals considered here it was used. Casteel's method is
based on the relationship between skeletal weight or certain bone
measure;:.e-nts of an animal to the animal's live weight. By weighing the
i.lelo:0:ons of a series of animals of known live weight or measuring a
certain -.s:.letal element, it became possible to generate least square
regression curves. These curves utilize the formula
log y = m(log x) +b
where: y=live weight
m=slope of the line
b=y-intercept of the log-
and indicate the reliability or the correlation coefficient (r) of the
To generate the formulae employed .n this study, live weights and
hseletal weights from specimens at the Florida State Museun were used.
Least square regression curves were computed using skeletal weight and
live t:eights for mammals, birds, turtles, catfish, and perciform fishes
and on cervical centrun width on the perciform fishes and the largest
centrun :.idth on the sharks and snakes. These formulas are listed in
Appendi:: C. It should be noted that the number of specimens used in the
calculations of these curves in some cases are very small. For this
reason considerable error could be introduced into the live WeigIit
Because of the variability in the types of zooarchaeological remains
it wa.- necessary to use different methods to arrive at live weight
estimates for different animals. The first method, the one that is
probably most accurate, uses the centrum width measurement and the
relevnnt formula (in Appendix C) to estimate live weight for the animal
The second method is somewhat more complex due to the fact that
all the comparative specimens do not have accurate live weight data.
In order to arrive at the live weight estimates a series of comparative
skeletons of the species to be esLimated were weighed. The live weights
for these comparative skeletons ~:ere calculated using the formiulae
(Appendix C). The archaeological bone was then compared wiith the com-
parative skeletal series to dete-rmine which specimen it most resembled
in size. The live weight estimate, computed for the comparative specimen
closest to the archaeological bone in size, was used as an estimate for
hlhe archaeological bone as well. In some cases the archaeological bone
fell, in size, halfway between the two comparative specimens. In these
cases the average, estimated, live eight of the two specimens was used
for the estimated live weight of the archaeological bone.
A third r.ethod for estimating live weight was used in certain
cases where a series of comparative specimens were not available. In
these instances a proportion was set up relating some particular
measurement to skeletal weight for a comparative specimen and the
archaeological bciie. By this method the skeletal weight of the archaeological
material could be estimated. This calculated, skeletal w iicht was
then useCd in the relevant skeletal eight to live weight formula and
the live weight estimated.
In a very few instances nc mreasurabl. elements were present in the
archaeological sample and another method for estimating live weight had
to be used. This fourth method used the bone weight of the archaeological
material to represent the skeletal weight of the animal and employed the
formula for that class of animal. These estimates resulted in very low
estimates and are probably only slightly better than no estimates at all.
In several cases none of the above methods could be used. For
these animals, an estimate of average live weight from biological
stuides, wis used.
V.ie above methods were enploye.'d to determine the live weights for
the siLes discussed in this study. Another method for estimating live
:eight was alro attempted. This last method used tie archaeological bone
s:eight as skeletal weight and calculated live weight using tie relevant
formulae. This was done to determine if the live weights estimated by
this means differed significantly fiom the more complicated and tine
consuming method used in this study. The results of this comparison are
included in Table 1. As can be seen, when the live weight are calculated
by these two methods very different estimates are obtained. The different
values resulting from the two methods and magnitude of their inaccuracies
should be remembered when considering the estimates in Chapter 5. For
this study the first method is used for the principal calculations.
The live weight or biomass estimates provide the basis for sub-
sequent calculations. For seven sites where biomass calculations were
taken, bar graphs illustrate the relative importance in terms of IE1I and
biomass of the various species from the sites. Edible meat weights were
also conouted from the bionass figures. Data on file at the Florida
State museum m were utilized for these percentage of edible neat estimates.
Th-e entire animals, except the bones, was considered edible for all the
species with the exception of mammals. The weight of skin, in addition
to the bone of the marinals, is assumed to be inedible and, as such, was
subtracted from the calculated live weight to determine the edible portion.
Percentngea used in these edible meat calculations appear in Appendix D.
Having, in this manner, determined the edible meat weight, the calories
and protein values of the foods were computed based on the figures
published by -htt. and Merrill (1975) and Leung (1961) (Appendi:.: E).
When vie'.ing the resulting charts and graphs certain points should
be considered. The number of the individu3as (-D'!) of each species
in-.icate the abundances of that species. Large animals nay provide
considerable' food, but unless portions are distributed among members of
C(.i'PArpITSO'I OF METHODS USED III ESTIIL'.TING LIVE l-EIGHT
Method I (used in this study)
live ut. live ut.
OGSE-SO 52494 61 33813 39
OGSE-63 129055 95 6530 5
OGSE-46D 8324 15 47005 85
Method II (live eight calc. from bone ut.)
I amma s Fishes
bone Calc. bone Calc.
__t. live wt.t. wt. live ut. ,
Formulas used in Method II
Log (live /t.) = 1.0133 (log bone w:t.)
Log (live Ut.) = 0.7775 (log bone -..t.)
the community or some type of preservation is attempted, the neat not
consumed will spoil. Smaller animals often supply a more reliable flow
of food than the occasional large kill. Diomass indicates, in ove-all
terms, the relative importance of a part Tular specie? in the diet.
\ie:-ing minimum numbers of individual and biorrass estimates together,
provides a more balanced picture of tha day-to-day e::ploitation of
animal protein foods.
Live weight or biomnass estimates do not necessarily indicate the
real value of a food source. Certain animals have a relatively high
biomass figure, but actually contain little edible meat, e. g. turtles.
'he edible meat weight, although balanced by the fact that it introduces
yet another estimate, is still a better indication of the actual food
consuLine. Assuming the estimates are accurate and that all the
potentially edible parts were actually eaten, nutritional compilations,
based on edible meat ;eights, suggest ho; efficient a particular food
was in fulfilling basic protein and caloric requirements. A close
consideration of the sizes of the animals, their feeding habits, and
the habitats tiry occupy, provide information on exploitation patterns
and procurement techniques.
The methods of identification anj quantified described above
furnish the bases for subsistence reconstruction of the sites considered
CllNiGIllG CLI;'1ATE AND THE ECOLOGICAL SETTING
lluman populations do not live in a vacuum and, as many researchers
have pointed out, in order to arrive at an adequate understanding of a
people's culture it is necessary to view a society in its ecological
setting (Vayda and Rappaport 1968). This is particularly important if
the research centers around the causal relationships or interrelation-
ships between subsistence activities and the other aspects of culture.
Without an appreciation of the resource availabilities, densities, and
ecological patterns, subsistence strategies may appear incomprehensible.
In ethnological studies an analysis of the area, with respect to resource
availabilities and densities, seasonal abundances, productivity, and the
nutritional value of the various foods, can provide the needed informa-
tion to investigate the subsistence patterns. In archaeological
research the problem is not as easily resolved. In some instances data
exist which indicate that in the past the area of concern exhibited a
different faunal and, probably, floral composition than is present today.
The Santa Elena Peninsula is one such example.
The Santa Elena Peninsula
Today the Santa Llena Peninsula area is characterized by semiarid
steppes with the extreme wesLern part of the peninsula being an arid
desert (Tretartha 1962; Sheppard 1930). Further north and east, tropical
wet and dry savannas occur (Trewartha .1962). The vegetation is classi-
fied as ::erophvtic (Acosta-Solis 1970: Sverlson 1946). Annual grasses
thinly cover the sandy, soil of the area and small groves of d:arf trees
and rounded shrubs are present along the arro'os (Svenson 1946). The
average temperature at Ancon, on the peninsula, is 23.90C With a high
in March of 26.9C and a low in August of 21.4"C (Acosta-Solis 1970).
The average rainfall is 325 mr., i;ith 97': of the precipitation occurring
from January through April with March a particularly wet mornth (Acosta-
Solis 1970). The other months are virtually. rainless. These combinations
of conditions result in a cold, rainless, foggy season (May through
December) and a w arm, rainy season (January through April).
The relative positions of the equatorial counter-current and the
Peruvian, or IHumbolt, current appear responsible for these seasonal
conditions. The currents and their related winds also account for the
estL to east change from colder, drier coastal climates to the warmer,
x:etter area inland.
The PeCLr.'ian current originates in the South Pacific and flous
north along the Chilean and Peruvian coasts. The i.main current veers
west and aw-ay from the contingent at about 5S latitude, although one
branch e::tends northward almost to thE. equator (Tre'.wacrthn 1962; Schott
1932). Tne. current experiences an offshore movement which is
corpensated by up-.-elling of colder waters s from a greater depth (Trewartha
1962). This cold upuelled water appears to be one of the factors
responsible for the arid conditions, the relatively low air temperature,
and the fog or "girua" of the coast (Treuortha 1962).
The northern eaiiatorial counter-current and the smalll E Nino
cicrrr.t are "... genuinely tropical in origin and have a r,uch low.'er
salt content and much higher temperature... than those froi the south'
(Trc;.:arthia 196'2:24). This northern current extends in diminishing force
to 2S latitude, but generally heads west near the equator. Along the
Santa Elena Peninsula, its main force is felt from January through April
and brings with it the warmer temperatures, increasing rainfall and
storms of the rainy season (Acosta-Solis 1970; Sclott 1932).
Periodically the equatorial counter-current extends southward as
far as Callan, Peru, dislocating the cooler Peruvian current. This
results in torrential rains and massive flooding and erosion in the
normally desert coastal Peru area (Murphy 1926). These rare occurrences
seem to be due to the displacement of the "... equatorial convergence
zones, together with its disturbances .well to the south of its usual
position to the north of the equator" (Treuartha 1962:32). These
periodic shifts in currents appear also to have been important factors
during prehistoric times.
Evidence of Climatic Change
Although the climatic history of South America is known in broad
terms, information on more restricted locales is not always as readily
available. In certain areas, however, information on past climatic
conditions can be obtained. Reconstruction of the Recent climatic
conditions for the Santa Elena Peninsula and the ecological settings
of the area is possible, based on studies of an ocean core sample
(Hough 1953) and archaeological research (Lanning 1967).
Akkaraju V. N. Sarnma (1974) has attempted such a reconstruction.
He baser, his analysis on the relative percentage of "wet" or pluvial
indications, i.e. mollusks who inhabit mangrove habitats which require
alluvium from active rivers to grow, to the percentage of intertidal
species which he believes '... were used for foodstuffs more frequently
when the mangrove mollusks were absent" (Sarma 1974:129). lie further
correlates his findings with paleoccological evidence from other areas
of South Amaer:ica. He concludes
Therefore, by analyzing the distribution of
pluvial indicators, as opposed to dry-habitat
indicators in shell-midden contends, the broad
outlines of the climatic records \'ere obtained.
These pluvial periods seem to have occurred during
the following periods: Vegas (6500-5000 B.C.);
Valdivia (2650-1800 B.C. and 1700-1600 B.C.);
Engoroy Guangala (1850 B.C. (350? B.C.) A.D. 50)...
When all the breaks in the seriations of the Penin-
sula are compared with climatic evidences, it is
strongly suggested that periods when the Peninsula
showed no archaeological records at all :ere
periods of aridity. The reason seems to be that
the availability of water was a critical factor
and during arid phases people migrated to better
and inure hospitable regions (Sarma 1974:129-130).
Sarma assumes that the relative amounts of shells in the sites reflect
their abundance in the area. Although this might have been the case,
Sarma's reconstruction fails to take into account possible changing
resource utilization patterns or food taboos.
Allison C. Paulsen uses Sar ca's climatic shift model to explain
obvious gaps in the ceramic chronology and changes in the Santa Elena
Peninsula settlement pattern during the period from 500 B.C. to contact
(Paulsen 1971). Although Sarma's and Pauls2n's data appear mutually
supportive, a nore detailed review of an ocean core sample and zoo-
archaeological analysis discussed in this study suggests a somewhat
different model of climatic change.
The reconstruction represented here includes information gained
from an ocean core sample analyzed by Jack L. Hough (1953). Tihe core
sample was taken at 8056.2' latitude and 29005.2'W longitude, an area
roughly due ve;st of Chir.ibote, Peru. This core contained material dating
back. to 9)0,000 years ago, but of inr.erest here is the segment dating
from about 11,000 to the presen,-. An.-lysis of this sample showed the
presence of globigerina ooze, which according to !hough (1953) indicates
warmer waters than the area e::hibits today. In addition, there are
medium, dark brown, strata characteristic of conditions not much
different than those at present and a dark brown zone composed of
clays which were deposited during colder times. The core signifies wIrmer
periods at about 7000 to 5000 years ngo (5000 to 3000 B.C.) and during
two shorter inter-als, one at around 1900 years ago (A.D. 50) and the
other at 100 (A.D. 350). These zones of oo:-e indicate that the northern
tropical currents shifted radically southward at least as far as 9S
latitude with a corresponding dislocation of the Peruvian current. The
Santa Elena Peninsula then assumed a rrore tropical configuration due to
the south.'ard movement of the equatorial currents during these times.
Two areas of dark brown strata, suggesting colder conditions, arc
present in the segment of the core of interest here. One colder cycle
appears at about 3200 years ago (1250 B.C.) and another 2800 years ago
(850 B.C.). These periods represent a north:-ard shift in the position
of the convergence of the two currents. Ifnether this shift extended
far enough north to directly effect the Santa Elena Peninsula is
unkno-rn, but if it did, a colder and dryer climate would be expected.
-Uhen the core sample is correlated with animal habitat information and
human subsistence and settlement data the following reconstruction
results (Fig. 2).
Climatic Change and Prehistoric Occupation
of the Santa Elena Peninsula
6500 B.C.-5000 ;.C. Vegas Occupation
The earliest fatinal remains considered in this study were left by
the people defined in the Vegas Comple:. This preceranic group exploited
SANTA ELEIIA PENINSULA
550 B,C.-A.D. 50
Achallan and Valdivia
the -aingroves and coastal waters and also the savannas and forests of
the area as the faunal remains from sites of this period telsify. As
Sarma (1974) points out, the presence of mangrove-specific mollusks
indicates a moister environments during Vegas time than today. Uater must
have flowed nearly year round to provide the alluvium needed for the
growth of the mangrove swamps (lest 1956).
Date from the core sample suggest a relative cool climate during
this period. Although cool, the area was warmer and wetter than it is
today. Mangrove forests extended along the coast and savannas probably
covered the inland areas. The river valleys and other areas might have
supported some forest growth.
5000 B.C.-3000 I.C. Uninhabited
Although an extensive survey was conducted on the Santa Elena Pen-
in;sula (Lanning 1967), no evidence of occupation of the locale had bzen
found during this period. Sarma (1974) believes that this abandonment
of the peninsula was due to increasingly arid conditions. Hough's (1953)
ocean core sample, however, suggests that this period was a time of
increasing warmth, probably resulting from a southward shift of the
equatorial counter-current during this period. If this were the case,
the Santa Elena Peninsula would have experienced increased rainfall
instead of arid conditions and probably would have resembled parts of the
present humid tropical forest of Columbia.
Studies of human subsistence in tropical rain forests suggest that
for foragers and horticult.'ralists the quest for meat (i.e. protein) is
of primary concern (Carneiro 1961; Gross 1975; Lathrap 1973; Holmberg
1969). In the tropical rain forests, species densities are low and,
with the exception of a few terrestrial animals, most inhabit the high
forest canopy and are, therefore:, lird to obtain. Only along major
rivers, rich in aquatic protein resources, did large concentrations of
people occur (Meggers 1971). On the interfluvial areas and along rivers
with low nutrienit levels only small population aggregates were supported.
Gross (1975) has suggested that this is due to the very loi. carrying
capacity of tlese areas with respect to protein sources. Due to this,
many agriculturally productive, but protein-poor areas, experience lo:
population densities. The Vegas people, faced with the encroachment of
the tropical forest, presumably also experienced the pressures of protein
Presented wiLh this problem, the Vegas people had three alterna-
tives: (1) they could leave the area; (2) th-y could try to get along on
decreasing amounts of protein by radically reducing their numbers and
extending their ran:e; or, (3) they could move to the rivers and shores
to attempt to exploit the aquatic protein sources there. The rivers of
the vcstern -Andean coast support relatively few, riverine, fish species
(Eigennann 1921). The coastal area experiences a greater range and
abundance of fishes or sea food, but the Vegas groups did not appear to
have a technology adequate to exploit these marine resources to their
fullesc extent. They seem to have adopted the first alternative and
left the area.
3000 D.C.-1600 B.C. Achallan and Valdivia Occuoation
Betw:cen 3000 B.C. and 1600 B.C. the Santa Elena Peninsula w:as once
again inhabited. Early members of this Tligration brought with them the
Achallan Cultural Complex (Stothert 197'.). Another uave of people,
entering the area at Llih same time oi. a little later than the Ach.illans,
was the Valdivians. This latter group is reputed to have introduced
agriculture into the general area (Lathrap 1975). These new people
exploited many of the same habitats that the Vegas groups had found
productive. Both tli faunal rains and the ocean core sample clharacter-
ize a climnatic shiift back to the mangrove wooded coast and probably tile
inland savannas and forest of the Vegas period.
1600 B.C.-1000 B.C. Uninhabited
Based on the lack of archaeological evidence, this period is
believed to represent another time of abandonment of the Santa Elena
Peninsula (Sarma 1974). The ocean core indicates nuch colder waters
around 1250 D.C., presumably resulting from the northward shift of the
Peruvian current. The movement of this cold southern current into the
peninsula area could have brought about colder, drier conditions. The
increasing aridity rendered agricultural and hunting subsistence methods
increasingly inefficient, and evidently lead to the abandonr-ent of the
1003 B.C.-850 B.C. Machalilla Occupation
Sarma (1974:117) suggests that the Hachalilla "... occupation of
the peninsula took place in an arid time and was brief." .nen correlated
with the core sample this cultural manifestation does fall betuilen two,
short, cold periods giving some support to Sarna's position. It should
he noted, however, that the area was sufficiently moist to support
mangroves (Sarma 1974).
850 B.C.-550 B.C. Uninhabited
Sarma interprets no break in occupation during this tine. The
core sample, ho.,ver, reflects another cold period aro'.nd 850 B.C.,
t.'lic'h as Fprobarbly of approximately the same intensity as the 1230 B.C.
episode. 'ihe Santa Elena Peninsula was abandoned during the former dry
period, and, although not conclusive in itself, a gap in Sarma' radio-
carbon dates support this as a possible third period of abandornment.
Paulson al:,o notes a gap of around 200 years in the ceramic occupation
of tle Santa Elena Peninsula Letween the Mac.halilla and En3oroy times,
but she dates this break at 1100 B.C.-900 B.L. (Paulsen 1971).
550 E.C.-A.D. 800 Engoroy and Guangala Occupations
Both Sarma (1974) and Paulsen (1971) see a continuous occupation
of the peninsula during this 1350 year span. Again during this period
mangrove species are found in the middens. The presence of a fox
(Ousicyon cf. sechurae) from a ridden of the early part of this span
(E.ugoroy) suggests that dry savannas or semi-deserts could have existed
inlanJ. Information on the terrestrial vertebrates of the later Guangala
period is not available. The only site analyzed from this time period
contained no terrestrial forms of food value.
For this general time range Hough's core sample indicates both a
coal reri.d formerly correlated on the Santa Elena Peninsula with
suvvannas and a 'a.riier span presumably similar to, but of shorter duration
than the 5000 B.C.-3000 B.C. episode. This warmer period would have
occurred around A.D. 100. This should have resulted in a return of
forest conditions. Neither Sarma nor Paulsen note any human displace-
ment. at this stage. Paulsen does indicate a move of people during
Guangla P-riod VI times. At A.D. 600 this resulted in the abandonment
of the inland sites located near man-made catch basins and the movement
of tlhe populations to thl, shore areas (Paulsen 1971).
'As indicated in the Pacific core sai-.ple, increasingly farmerr
conditions '.pre felt again around A.D. 800. Possibly the encroaching
hu-.iid tropical forest, which wouldl d be experienced ii1 the more nortlherly
Santa Elena Peninsula area earlier than A.D. 800, could have resulted,
as in the case of other tropical forest aILe;s, in more coimetition over
the increasingly scarse protein foods. This pressure miuht explain the
initial riovement of the Guangala people from the no-. protein-poor basin
areas to the r.ore protein-rich shore. This nay also account for the
ultimate Guangala abandonment of the peninsula.
A.D, 800-A.D. 1000 Uninhabited
Contrary to Sarma's reconstruction, 1Hough's analysis of the Pacific
core sample indicates that in this area. the equatorial counter-current
had again shifted south during this period, bringing :.ith it a return
of w.ar.m, moist conditions. Tropical forest vegetation preswuiably once
raorc covered the Santa Elena Peninsula and during this period the
Guang.la people appear to have abandoned the peninsula.
A.D. 1I00-A.D. 1400 Libcrtad Occupation
During this time period the Santa Elena Peninsula again supported
a human population, this time members of the Libert-d Culture. Although
cooler than the proceeding 200 years, the areas was still moist enough to
support mangrove stands. The only vertebrate, zooarchaeological col-
lection available contained no terrestrial species that could provide
cli.iatic indications. Based on the evidence of a ureceeding moist
climate nd the subsequent semii-arid and arid conditions ex:tant today,
presumably the Santa 'lcnl Peninsula was passing through a transitional
S u mmary V
As Iloush's (1953) core sample indicates, the Santa Elena Penin-
sula %arS apparentlyy subjected to periods of high aridity resulting in
semii-d.cc-Lr,- cotditons 'ind to cpi-odes of ir.ceasing i.armth anJ moisture
lea'.ina to the development of humid, tropical forests. Th:z- arcltaeulogi.cal
evidence sugge.st. that both climatic c::tremnas generally resulted in the
ob,-ndonrjent of the area. Jitliout substantial change in the t'_chnoiogical
e:ploit.atiun base and the related modification in certain cultu-al
institutioLns, the various peoples were unable. to adequately utilize the
area'; available resources.
In aJdition to the radical climate changes described abu.ve, the
Sance hL .ena Pe'ninLsula, during its 7500 years of .sporadic human settlement,
undoubtedly e:spe-'ienced both seasonal and, at irrct-ular yerly inter .al:.,
minor climatic fluctuations as it does today. These would result from
relntivaly slight shifts in tlir po:;itiucns of itie ocean currents. The
overall effect of these minor episudes on the prAhistoric peoples need
not. have been great.
Although the Santa Elena Peninsula furnishes the main focus for the
reconstruction above, the focus of interest in this study encompasses a
soIme-..hat larger region. Areas a little further north would be affected
slightly earlier by any soutlharJ movement of the currents and late by
the northward displacement. Inland areas, both to the north and east,
:o.uid he slightly less influenced by either shifts, a result of their
co.itior.s relative to the currents and coastal winds.
FAUIAL ANALYSIS AID RECONSTF.UCTIONI
Thei previous chapters presented the theoretical frarme'.-orl and
methodology for this study. This section concerns the identification
and analysis of the faunal material from the various sites. It is an
attempt to arrive at an understanding of past subsistence patterns and
exploitaLion techniques. The following chapter suLiaarizes these findings
for the cultural phases and identifies the subsistence-related, human
behavioral patterns practiced by the various populations.
For ease of presentaCion, the archaeological sites are divided
into three main groups, pre-Valdivia, Valdivia, and post-Valdivia. The
'Vldivia sectio;l is further divided into coastal and inland sites. For
each of the groups considered in this study, some general remarks on
th-. cultures as a whole, and particularly on other subsistence aspects
of the people, are included.
Each site is then considered. Appendix A contains the detailed
inFo;rm-tion on species present and their ri I, their number of tragmen::s,
and their bone weights. Appendix B tabulates the biomass, edible ieat,
calories and proccins estimated for certain of the sites. The text of
this chapter provides summary observations on the species present and
their relative ir,-;ortance in fulfilling the people's nutritional needs.
Thi:3S sugg,'st- tlih value of the various vertebrate resources in the diet
of the people. No::t, the technology used to obtain Lhe food resources
are e:;iamined. hiere available, artifactual evidence, i.e. projectile
pFoints, fishhookls, etc., is considered, but the discussion of exploitat-
tion techniques largely depends on the habits and habitats of the
princional species present. Information on the relative importance of the
various .ninal. and on their habits and !habitat preferenrcs provide the
basis for observation of possible subsistence-related, 'human behavioral
llaterials from three sites on the Santa Elena Peninsula provide the
data for the pre-Valdivia group. T-o of these sites exhibit Vegas
affiliations while one has been assigned to the newly defined Achallan
Complex (Stotlhect 1974).
The prececraric, Vegas Complec:. represents some of the earliest
archaeological material yet discovered on the Santa Elana Peninsula.
Dated at be ;ueen 6500 B.C. and 5000 3.C., this cultural manifestation
consists primarily of shell riddens located along the western section of
the peninsula (Stothart 1974). Excavators have uncovered several types
of stone tools fraro Vegas period middens including side- and end-scrapers,
flake knives, gravers, denticulates and spokeshaves (Willey 1971).
Heav.%- ducy choppers, grindstones, and hammerstones also have baen found
(Scothere t974). No bifaces or stone projectile points have as yet
In his sumr-nry of the \'egas Complex, Gordon R.. \illey states that
... the shoreline sites are shell middens which
offer our only direct evidence of marine subsistence-
and LInnLng suggests that the former midden dwellers
might have followed a seasonal round of winter
shellfishing at the beach and summer plant gathering
and fishing along the streams (Uilley 1971:262).
As is seen belo-., this does not appear to be entirely the case. From
the faunal sample from the two Vegas sites analyzed, both sea turtle and
marine fishes :.:e; identified in considerable number's. Although these
samples provide no evidence to either support or refute seasonal occu-
pation of the Vegas sites, a seasonal Fhift in subsistence emphasis is
cot.aon for manj, largely Iunting and gathering peoples. It should be
noted, however, that today and presumably for some time in the past, the
western slopes of the Andes and the coastal strip have been characterized
by a relatively impoverished, freshwater, fish fauna (Eigenmann 1921).
AIL' certaii climatic factors characteristic of the area cause p-.riodic
desiccation of the Santa Elena Peninsula. This results in the destruction
of the meager, freshwater, fish fauna. It is interesting to note that,
with the exception of two, fresh.iater catfish, no fre3hwater fonrs were
found in any of the middens considered in this study. Although this
could s.I;rply reflect cultural selection of marine forms, it might
indicate tlh exceptionally lo densities of freshwater fish populations
and, therefore, little or no advantage in fishing t.he freshwater streams.
The cemairns from OGSE-SO represent the largest, pre-Valdivia, bone
sample available for analysis. It is, therefore, particularly important.
OGSE-SD is located about 3.5 km from the sea and is a shell midden site
with a Vogas occupation overlaid by a shallo:.- Valdivia deposit. Three
different types of burials ere found in the midden. Kare- Stothert,
the excavator of the site, believed two of these were of Vegas affiliation.
The third she identified as possibly an intrusive "'aldivian burial type.
Faunal remains. This Vegas ridden contained the bones of a wide
variety of vertebrates, including humans. The human remains slho' no
in!dicarions of cannibalism. The human bones crre. sc-atcered in the midden, a
pactern ch!aracterisitic of food refuse, but it is ransoned that, since
che midcden functioned as a burial area, these human bones could '.cll
represent burials that were dislodged by later internents. For these reasons,
the human bones are not viewed as food remains and have been deleted from
consideration. The only other evidence of the use of vertebrates for
other than or in .addition to food purposes was the presence and location
of a large number of fox teeth (Stothert 1974). The teeth were included
in a burial as grave goods. Other vertebrate remains found at the sit.e-
presumably represent Vegas food items.
In terms of numbers of individuals, rodents are the most abundant
ri:nmmal collected for food. This large number of rodent remains is
surprising due to the nearly total lack of these animals in other midden
samples considered in this study. The rodents from OSSE-SO are a small,
rat-like animal and, although numerous contributed relatively litLle
i'.eat to the aboriginal diet if they were in fact used as food and not
merely incidental to the site.
The fox remains constituted both more individuals and more meat
than the rodents, but not all the remains of the fox can be assumed to
be food refuse. Stothert, while excavating the Vegas type III burials
nio;:ed tlihee, conical, piles of fox teeth distributed around one human
skeleton. Analysis o[ fox teeth from the total CGSE-80 sample and the
calculaticn of the minii.um number of individuals (;.UI) they represent
indicates that the excavated portion of this site contained the remains
from at least 27 fox:es. 'Then the nI is calculated on skeletal parts,
other than teeth, at lea3t four individuals are represented. The !GII of
fouLr undoubtedly represents a more reliable estimate of the importance
of fc:.: as a food source at this site. For this reason subsequent cal-
culations of the dietary inportcnce of tile fo:; is based on a 1GII of four.
Although less numerous than either rodents or foxes, the deer
provided by, f3r more edible meat (/1't) than an,' other vertebrate source.
Thie size of the deer bones arc rather small and are probably from the
relatively small, brocket deer (Maza3-.a). Other n-~Lals found in the
midden that are both less numerous and presumably of less nutritional
importance include rabbit, weasel, and an unidentified fox-like mamiial.
These animals functioned as additional meat sources for the Vegas
inh bitb L ant .
Although na:rmals represented tle most important vertebrate food
(c. 59" of the edible meat), fish also contributed considerable meat to
the diet (40?). Catfish and drur.i were particularly abundant, although
the less common shark, snook, jack, and tuna actually supplied more meat
providing more proteins and calories. Additional fish sources were ray,
snJ'ap etr, grouper, and mullet.
The Vegas material contained several] other vertebrates including
fro,. snakEe, sea turtle, an-1 parrot. These remains represent a minor
food component of the diet. Figure 3 suimnarizes the IOII, biomass, edible
ime 2t, calories, and protein estimar.es for this site.
Reconstructed huntigc and fishing untteris. Although there exists
little artifactual evidence of stone or bone hunting and/or fishing
equipmen-lt, somn ob-ecrv' nations on procurement techniques can be made using
the principal faunal reinains themselves. The Vegas people apparently
huI-itcd fox for two principal reasons, for food and for the teeth that ::ere
"SO I i i ......- l ...., .1 --- ..-i ......
LL- L- c/)
x -- LLC
, '0 \\- \
'Io o ,
_r'___ [., ~I
L -o "
o -0 o U o o
c C C
o 0 0
inter-red ithl one of the burials at the site. This suggests some type
or specialized hunting or at least the retention of the teeth of this
animal. The em-phasis on fox: hunting becomes particularly apparent when
tlie OCSE-30 remains are compared with those Froi.i other sites where fo:.
bones are uncommon or absent.
No direct evidence exists to indicate lhoiw the Vegas people
obtained either these foxes or the single deer fo-ind at the site. Several
possible methods could have been effective in capturing these animals.
These methods include the use of so-m type of projectiles or traps of
cither the snare or death fall types.
The rodents are basically nocturnal animals and forage from the
late afternoon to early, morni.n-3. This time span generally represents a
period of low hunting activity by the human predators. The rodents are
also of relatively small size (about 80 grams live weight), and, as such,
direct hunting methods, e.g. single stalking of the animals, uould result
in little return for the energy expended. Trapping would probably
represent the most productive method for obtaining these small, nocturnal
animals. They might have been attracted to the rubbish around the camp
and trapped there.
Most of the fishes found in the midden are indiscriminate carni-
vores and readily take a baited hook. This method could have been
utilized to catch them. The mullet, a fish that cannot be easily caught
with a baited hook, must have been ta:e-n by another method, possible by
spearing or hand-catching or, as the .southern United States blacks do,
by wrapping filamentous algae around hookls (Wing 1976b). Although some
of the fishes are fairly large, about 9500 grams, nost are smaller,
under 700 grams, with the majority in the 100 to 200 grains range (Fig. 4).
lo evidence, such as large numbers of herds of terrestrial animals or
schooling fish species, which night incjcnte that the people practiced
cooperative hunt ing and/or fishing was found in the sanole. In fact,
the fish- remains suggest that cooperative fishing ?:os not practice..
Fisherman working singly or in small groups probably were the most
efficient way of obtaining the fishes represented ini tih ldilden.
The shell midden OCSE-38 is the other Vegas Complex site consLdered
in this study. Like OGSE-80, this site is located on the Santa Elena
Peninsu-la. it is situated, however, nearer to the shoreline. Originally
it w:as expected that this site could demonstrate either close similarities
with OGSEI-30, indicating a general subsistence exploitation pattern
despiLte slight differences in ecological setting for V.'as sites, or a
different resource foc'i sugge;-tive of a modiEication of subsistence
base to take advantage of the more readily available resources. UnfortunatCly,
the suall size of ti-h OGSE-33 s.anple makes it impossible to test either
Faunal remains. The siall sample from OGSE-33 does indicate that
these Vegas people utilized nmany of the sarne resources as the CGSE-SO
group including fox and rodent and the marine forms, catfish, jack,
mullet, and sea Lurcle. This site did include the pufEer, a fish absent
fron tlhe other Vegas' midden. Due to the extremely sm-all size of this
sample no biomass, edible meat weigh:, calories or protein estimates
were attempted for this site. It would seem, however, that terrestrial
forms or the sea turtle were probably the most important food sources.
Thir: Acha;lla:. represents a recently defined cultural complc;:.
Stother: believes, based on her world at OCSE-63, that the Achallan
"... sbo:jed some technological impoverishment" (Stotlert 1974:14) when
co.Fpared to the Vegas group, but had added rather crude ceramics to
their cultural inventory. To date only one ;ite, OGSE-63, has been
assigned to this comple:-. Its assignation is based on detailed lithic
analysis (Stothert 1974) and it has been carbon dated at around 2700 .. C.
Stothert believes this date too recent and suggests the middle of the
4th millennium B.C. as more accurate.
This site is located along the Rio Achallan on the Santa Elena
Peninsula. Stothert suggests that originally the site consisted of a
ring of small middens containing shell.
As in the case of OGSE-33, the zooacchaeological sample from this
site is small. Nevertheless, since it represents the only material from
this period, and could be viewed as intermediate between Vegas and Valdivia,
biomass, edible meat, calories, and protein estimates are computed
(Fig. 5). Because of the small sample size these estimates could include
Faunal remains. Ifhen this Achallan material is compared with
earlier Vegas samples certain dissimilarities appear. This is
particularly evident he en e:-:amining the mammal remains. LThile tie two
Vegas sites exhibited a wide subsistence base, the Achallan people were
more selective. The fox and rodent, that are well represented in the Vegas
faunal samples, are lacking in the OCSE-63 material. Deer is the only
mo.mal pre.-ent at this site. By a comparison of the sites of the bones
two specie.- of deer appear in the midden, the larger uhlite-tailed deer
and the smaller bracket.
The amrph.i.bcns, snakes, and bird bone pr-een'- in the Vegas sites
are missing from the Achallan sample. This mihtl: be the result of the
small size of the sample. The fish.s e-:ploited are nut unarkedly
different in type than those from Vegas sites and .res'r..bly represent
a continuation of Vegas-like fishing patterns.
The Valdivia Phase represents the early' Formative manifestation in
Lcu.dor and one of the earliest Formative phases in the geu UJorld. As
such, it provides an excellent opportunity to study the gradual shift
from nomadic hunting and gathering to sedentary horticulture. How this
shift came about and under what conditions it occurred is the interest
of many researchers.
Some scientists believe that early sedentisL.u ias possible on the
Ecuadorian coast because of the abundant and- reliable food found in the
coastal area, i.e. shellfish and fish (r'eggers 1966). These researchers
note that, at the same tine as early Valdivia, coastal Peruvian groups
were already cultivating beans, squash, bottle gourd:. and cotton (Lanning
1.976b). They suggest that rhe early, Valdivians h:,d more or less permanent
settlements on 'he coast where they c;ploited the local food resources
andc possibly, like their Peruvian neighbors to the south, engaged in
incipient agriculture (!'egsers 1966).. The Vallivian people theoretically
suuolemeinted lheir macina protein resources by occasional huntLin" trips
Other researchers (Lathrap 1975) believe that the Valdivia people
were engaged in substantial agricultural practices as early as Valdivia
I ti-mes (c, 3400 B.C.) and were well established by Valdivia III (c. 2500
B.C.). These archaeologists find support for this hypothesis in the
location of the sites, the presence of storage pits and grinding
The presence oF agriculture and its importance in the diet is an
interesting question, but one that is difficult to study. Conditions on
the periodically wet Guayas coast result in poor preservation of plant
remains and, to date, no direct evidence of substantial agriculture for
this period has heen unearthed. Indirect evidence (Lathrap and Marcos
1975) and analogies with other areas of the sane time period (Ileggers
1966) suggest thlt the early Valdivians could have practiced agriculture
at least in its incipient forms. Presumably, increasing reliance was
placed on agricultural crops through time. Although plant foods including
agricultural products are important components in a diet, animal protein
sources are as important, if not more important, than plant foods in
supplying needed nutrients, especially protein. .What protein sources
the Valdivians used and how' they obtained the animals is the question
that is considered here.
Certain similarities are found in all the Valdivia sites studies
below', but it wouldd be erroneous to speak of a "Valdivia hunting and
fishing pattern". Considerable differences are evident among the sites,
especially .whien comparing the Santa Elena Penirnsula sites with those
either inland or farther north. Some of these differences are undoubted-
ly attributable to local biological and ecological factors, such as the
presence of lhabitats particularly favored by certain species, while
other differences are inore easily explained as resulting from cultural
patte-rns nd practices.
Zooarchaeological materials weree available from five coastal
san:ples, four of which ereue located on the Santa Elena Peninsula and
the fifth is to the north at the mouth of the Valdivia River. Two of
these samples came from one site, OSSE-62 (numbered OGSE-62 and OCSE-
62C). OGSE-62C was assigned to the middle Valdivia (Stothert 1975)
subphase, while OCSE-62 is simply listed as Valdiv.ia. These two
samples could have been regarded as one unic, but they are considered
individually here. It was felt thac cre'acing these samples individually
provided the opportunity to study tie variability within a site. For
this reason biomass and food value estimates, as well as relative
number charts and fish size graphs, are constructed for each sample.
The other cto Santa Elena Peninsula sites were too small for any kind
of reliable estimates. Faunal lists and a short description of the
remains are included for these sites. The fifth 'aldivia site, the one
from which this cultural phose takes its nam.ie, is considerably different
from the Santa Elena Peninsula sites both in species present and the
size of the individuals. Material from this site suggest a slightly
different subsistence emphasis.
OCSE-42 represents the Valdivia Phase I occupation on the Santa
Elena Peninqula. Based on ceramic similarities and date:; from the Phase
I, Lama Alta site, OGSE-42 was occupied around 3400 B.C. (Bischof 1972),
although the actual dates, based on shell, were late- (Stothert 1974).
LKncen Stothert, the excavator of the ridden, believes that this site
waz occupied for a fairly short period of time and that the middec
deposit itself nay have been in the form of a ring, as in the case
of the Achallan Complex site CGGSE-63 (Stothert 1974).
Lh.en the animal bones from this sire were first submitted for
analysis it was thought that this material might slhow a subtle shift in
exploitation emphasis. If this shift in protein utilization had occurred
it mriht be correlated with the introduction of a neu culture type whhich
possessed at least incipient agriculture. Although thE nacerial fro.i
CS:E--42 contains sone elements suggesting a shift, i. e. decreasing
amounts of terrestrial forms, the sample \:as so snall that this apparent
change may be the result of the sampling itself.
Fauna1. re-mains. The species found in the midden material include
brocket deer and marine catfish, snook, drum, and sea turtle.
CGSE-62 is situated abcut 100 meters south of the Achallan ComLplex:
site, OGSE-63, discussed above. Occupation at this site began during
ValdiviaL Phase III times (c. 2500 B.C. Lanning 1968) and continued
through Valdivia V (Stc.therL 1974). .s in the case of OCSE-63 and OGSE-
:2, the miidden at OGSE--62 ',as distributed in a form suggestive of a ring
of sxall deposits (Stothert 1974).
Although similar to Achlallan in location and settlement, here
sinmilaricies stop. OGSE-62 (OGSE-62C) had both a much more elaborate.
ceracr.ic inventory and, more important for this study, a very different
protein base. This latter difference is not explainable on site location
along ;ince both sites are situated relatively close to each other.
Although a sll.;.c .n the climate, in the highly variable Santa Elena
Penin!sula area, could explain the different numbers of species present,
the difference could also be due to a shift of c::ploitarion emphasis.
Faunal remains. Ihe vertebrate faunial remains from OGSE-62 are
alrmst entirely mac-ine (93.9' HIJI). Principal abundant species include
the catfish (41.9%) and the prunts (31.4%). Other important fishes
represented here were the grouper, jacks, snappers, and porgies (Fig. 6).
Due to the larger sizes (Fig. 7) of these numerically fewer fish, the
groupers, jacks, snappers, and porgies contributed more meat than the
catfish and the grunts. Both numerically and nutritionally minor
components of the diet are the sea turtle and a mammal. The true
nutritional importance of these last two animals are considerably
underestimated due to the method of calculating their live ',eight, but
the relative importance of these resources, versus fish, is probably
not too accurate.
Faunal remains. The material from OGSE-62C resembles in many
respects che proceeding sample. nevertheless, these remains include
proportionally more catfish (63.6?') and fewer grunts (25%) than at
OGSE-62 (Fig. S). Also fewer species are represented. Due to the overall
size distribution (Fig. 9) third fishes from OGSE-62C actually contributed
more meat in ;weight than those in the 0GSE-62 sample. i:uinerically minor
species in this sample included the snook, grouper, jack, and porgy.
o o -
-- a - -
_ _- I. 1__. .
O.,SE--62 and OGS.-62C
Reconstructed hunting and fishing patterns. eihen CG3.-62 and
OGS.;-G.C are compared, several differences become evident. Particularly
noticeable is the wider range of species present in the OGSE-62 sa-.ple
which ;rLe absent from OGSE-62C. These species account for 9.4, of the
IC'I from CGSE-62. This need not be a- im-portant as it first appears.
All the species represented in rhe samples (with the exception of mullet)
are inshore carnivores and can he taken with a baited boo!: and line.
Shell fishhooks have been found in Valdivia midJens and these hooks range
in size from sbcut 1.8 cm. by 2.0 cm. to 2.5 cm. by 2.8 cm. (Ieggers,
Evans, and Estrada 1965). The seemingly disproportionate reprasentatioii
of certain forms could simply represent fish caught on.a day or season of
Tlhe year when "the snappers v'ere running" or result from a particular
fisherman's atLachment to a particularly good fishing area.
Ancthcr method had to be used to catch the mullet. Bullets are
herbivores and as such ace not readily taken by a baited hook and line.
One effective method for catching this fish is using nets either of the
gill net or seine type. This could result in the capture of large
numbers of this schooling species. Weirs and traps could also result in
numbers of mullets as well. In any event, only two individuals of this
species were identified from the samples. If nets or some other collective
techniques had been regularly used on this abunJant species more
individuals would be expected. This suggests that none of these methods
were employed. The Valdivian fisherman could have resorted to spearing,
or ihand catching or algae-baited hook to catch these herbivorous animals.
OGSE-174 represents nnotl--r Santa ELer.a Peninsula midden. No
sub'hhase association is available for this Valdivia site. If, ho-wever
OGSE--42 and OGSE-62 are representative of Santa Elena Peninsula coastal
middens, the relatively large representation of mammal remains suggests
Ln early Valdivia affiliation.
Faunal remains. Although three deer were identified from this site
only one is assignable to a species, i.e. Odocoileus, the other deer
material was either too fragmentary or of a size that made species
correlations impossible. The marine resources represented in this midden
re3e,nble those front other sites discussed. Again the catfish and the
grunt are parLicularly abundant. Snook, jack, and sea turtle also occur.
The Valdivia site material is the only Valdivia coqstel site outside
the Santa Elena Peninsula area considered in this study. This site is
located north of the peninsula and at the mouth of the Valdivia
River (Fig. 1). The ceramic and lithic material from this site formed
the bases for the original description of the Valdivia Phase (leggers,
Evans and Estrada 1965).
Faunal remains. The vertebrate fauna from the Valdivia site differs
somewliat from those discussed above. At this site terrestrial species
include the peccary and two types of deer, the ;fliite-tai]ed, and the
brocket. Based on comparisons of the size of the deer bones, the larger
white-tailed deer appears more co:rmmon in the i uidden. Catfishi are still
a common fish in the midden, but at Valdivia snook is also abundantly
represented (Fig. 10). :More neat 'was available to thl people from
snook and deer sources than all other sources combined (Fig. 11).
*2-*-- -- : L -- IA -
I- :: :
0 0 0 0
Reconstructed hunting and fishing patterns. The technique used to
o.btni.n the terrestrial forms is difficult to determLine. As yet no stone
or bone projectile points have been uncovered in Valdivia middens.
Perishable wooden projectiles or traps and deathfz;lls are techniques that
could have been used.
All the fish species from the Valdivia midden sample are inshore
carnivores and will take a hook. 11o herbivorous forms are included in
the remains. If herbivorous forms had been present in numbers, another
or additional method of fishing, e.g. cooperative netting, might have
been suggested. The fish bones from this site, then, present a picture
of individual fishermen exploiting the inshore or shoreline waters .ith
baited hooks and lines. Further evidence for this is the shell fishliooks
that have been found at Valdivia. Although boats could have been used,
there is no evidence to suggest that they would have been needed to
obtain the fishes present.
One difference in this site, compared with other sites, is the
proportionally greater representation of large fish, particularly snook,
and the low numbers of smaller animals-. The lack of smaller species is
pL-obably in part due to the excavation techniques used. The relatively '
large representation of snook, undoubtedly, results from fishing the
estuary that is believed to have existed near the site during Valdivia
times (Ileggers, Evans, and Estrada 1965). This sort of ecological
setting would have been particularly attractive to the snook.
Zooarclih ological materials from two, inland Valdivia sites were
available for analysis, the Loma Alta site, located about 15 kim. upstream
from the Valdivia site, and the Real Alto site situation about 5 km.
from the rea along the Rio Verde. Loma Alta represents an ea:!.y
Valdivia site (Ihase I) while the principal occuJpntion at Real Altu
occurred from Valdivia I1T through VIII times (Harcos 1975). In an
attempt to study, chlanIe through time this latter site w.ns divided into
two groups, a middle Valdi'.ia component represented by ITIi-V ceramics and
a later Valdivia indicated by VT-VIII ceramic types.
Excavations at Loma Alta and Real Al-o revealed a discontinuous
distribution of faunal r..aterial in pits, burials and house structures,
which necessitated a different approach to the material. Since there
was no w'ay of knowing hwJ many pits were associated with any particular
house floor, it was impossible to provide ,n-, kind of meaningful
co-bination of features and, therefore, the units were treated individually.
nTe Loma /Alta site contains a uijer variety of vertebrates than
any other site considered in this study:. Representatives of all the
vertebrate classes are present. This indicates a wide subsistence base.
Ho dcubt sone of this results from the location of the site in a presumed
foriested area, although other factors also are evident. Tle types and
i.u.'bers of the remains oere not evenly distributed throughout the site.
The JII unit has more catfish, while JIII contained more deer (Fig. 12)
Faunal ria'ins. Not all the remains from the Loma Alta site
coiistituted food items. 'he large number of human bones in the sample
suggest that at least some of these bones represent burials rather
[lirt food remains. The dog also may represent something other than, or
in addition to, a food item. The dogs could have served a variety of function.
either as guard dogs, hunting dogs, caiap scavengers or pets. The
burned condition of .some of the bones suggest that the dog also ended
up as food.
At Lo;:a '-.lta the t'ro samples analyzed contained the sane principal
ania.rls, but in very different proportions, Since there was no v'ay of
c!eteriniinng which sample -..as rore representative of the site as a
whole, it was felt that no biormass estimates should be attempted.
c:.-ever-, the size of the species present, and their relative numbers,
suggest that terrestrial forms, especially the deer, were the principal
protein sources. The Valdivia hunters obtained both principally
forest-edge and/or savanna animals such as the white-tailed deer, agouti,
and rabbit and rrerumibl: deep forest inhabitants, the brocket deer and
the tanir. Peccary, opossum, an:! the carnivores, the riountain lion
and the fox vere also hunted. Several small rodents, and an armadillo
represent other mnammals captuLred. The Valdivians- at Lor a Alta; also caught
birds, as the considerable numbe- of bird remains froi the site indicates.
Additional animals include snakc, land turtle, and frog.
Of parLicular interest, though of seemingly minor imp-ortance, are
the fish remains from Lo-im Alta. All the fish species represented in
the riidden are of marine forms. It is estimated that Loma Alta is 15 kLm.
(9.3 miles) fro.-n the sea and, therefore, from the marine habitat where
these fishes are found.
r.aconstruction of hunting and fishing patterns. There is good:
evidence for he seasonal e:.pl.oitation of deer. The fauna.-l sample from
Lo.in Alta contained nine mandible fragments representing a minimum of si::
deer. Lased on tooth eruption (Scv.erin'..hauz 1949), one individual wJas
betwo-in fnor and seven months old, t'.:. others were around 15 i.onths,
two 17 months and one was an adult. The deer that could be aged in
r.onthsL indicate n deer hunting season confined to about a five i'ionth
period. This could be interrupted as a seasonal occupation of the site
or a restricted ie-riod of exploitation of deer possibly revolving
around other seasenally-regulatcd activities. The latter possible seems
Pore likely. A people who seasonally moved into the area -'ould be
unlikely to engage in trade with coastal areas for their fish or to
send om.e people out into the hills to hunt while others travel the
15 Ikm. down to the shore to fish. Loma Alta is more likely to represent
at leyat a semi-sedentary village engaged in an exchange system w'ith
coastal groups for marine resources, perhaps with Llhe Valdivia since
Exchange systems are complex sets of social obligations and
reciprocal arrangements which need some impetus for develomenLt. They
generally develop because of the need for scarce and desired resources.
Could protein scarcity have acted as the impetus for the establishment
of trade with the coastal areas? Although it is not possible to say
with any certain.ty whether this was the case, possible evidence for
this exists. Among the faunal remains from the Lma Alta site there were
a large proportion of human bones, especially of fetal or infant
individuals. A total sample of four fetuses or infants, representing
in age seven months, seven and a half rionths and mid-ninth month for
the fetuses and a baby within its first year (Miaples 1974'), were found
in the fainal samples. Could this large and disproportionate number
of fetus'.?s and .infant remains indicate low. nutritional levels re.-ultin;
.in sr.ontLLLeous abortions (liulinski 1976) or infant icjle e ;which is corLrion
in many protein poor areas suLch as the Aimazon? Could c:'arJ for fish
with the coastal sites be an attempt o obtain more animal protein?
Rl:2a- tIco it. the last Vnldivin si.tu considered here. Donald
.aticrap, director of the excavation at the site, states that its location
indicates a river valley rather than marine focus. ile beli.'eve that
Real Alto represents an a5riculttural village based on n.,ize cultivation
Because of the long occupation at Peal Alto and tne cultural
shifts that are evident between middle and later Valdivia times, the
material from this site .las separated into t:o groups. Not all the
rat'erial front Pl.a) Alco was analyzed but only a sample from certain
features lhese u;its :ere chosen because of their phase affiliations,
size of tho somples and lac'- of obvious evidence of intrusion. iate-rials
from five different units =ere considered for the earlier cultural
division. Two of these .were materials from the floors of structures
(St.ructu-'es VII and X), two were from features (Features 10 and 171),
and one ;as fror. thle fill of a burial pit (Burial LI).
Faunal remains. The bone from the vario-us units w'.ere identified
arnd the relative numbers of the various species computed. The results
of th.e ;lI comi.ilat-ions are indicated on Fig. 13 and Fig. 14. The
s:uipl s exhibit considerable variation. In most instances, in Mliddle
valdivi a times, catfish represented the main component of the sample in
tero-s of l.r;I with the drunis thle most numerous species represented in one
IIZ__IJ~ -- -
sample. Drum is also of considerable importance in several of the other.
samples as uell. Deer are prose.it in all. the units an d birds and sLa
turtles and an occasional snake also occur.
Lecausc of the uneven distribution of the fauna 1 re7-ins in tn-
nuii2?rous pits., occupation floors, ;all trenches and burials :tt the site,
biomass estimates were not attempted. There was no :,ay of determining
which h and how many pits riere conitemmporaaeous with esch house floor or
even thi initial volumes of the various features. Since there %was not
any means of correlating pits and floors and other features, it was felt
that any biomass estimates would result in an erroneous representation
and indicate a greater accuracy than the material warranted. Although
bio:nass estimates were nor undertaken for this site, deer undoubtedly
represents the most important, single food source in terms of the amount
of meat provided. Fish, ho:-.ever, especially the raarine catfish, provide
a more constant food source.
.Then the sr.terial from Mliddle Valdivia samples is compared with later
Valdivia material a shift in emphasis become.z apparent. Material from
thei Structure VIIT wall trench, Features 101 and 10G and general non-
feature ridden material indicates a greater exploitation of catfish
in later Valdivia times (Fig. 14).
Reconstruction of hunting and fishing patterns. As in the casE of
Loma i.lta the question of trade or specialized hunting and/or fishing
arises. It could be argued that if Real Alto like Lona Alta was
primarily an agricultural village most of its daily activities would d
be related to agricultural pursuits. Fish could be obtained by e::change.
It should be noted, lihn;ever, that 5 km is not a very great distance to go
co obtain food. This is at the outer limits, but within the range
indicated l.y E. S. iiggs and C. Vita-Fitnzi (1972) as the c:-:ploitable
distance For hort[culturalists. Also, it would be possible for the village
to have specialized fishermen who could exchange parts of cheir catch
for agricultural goods. Real Alto's nearness to the coast does not
seem to result in the same sorts of pressures discernable at Loma -lta.
h.Trether exchanged between villages, obtained by specialists, or caught
by the Valdivia horticulturalist, fishing techniques similar to tliose
practiced by the coastal fishermen were presumable used. These would
include baited hooks and lines for the carnivorous, inshore fishes found
at this site. There are again some herbivorous remains ini the sample
(mullet and sea chub), but their n'lmbers are veiy low and co not
necessarily suggest nets or traps.
r!aterial from only four post-Valdivia samples was available for
analysis. These samples included remains from four cultural phases;
Haclhalilla and Engoro, bones, Guangala phase material and the late
Libertad remains. Three of the samples are from the Santa Elena
Peninsula, the fourth is from the Rio Verde Valley.
:lachaliila and Engorov Phases
This site contains both HIachalilla and Engoroy phase material, but
vr-ry srmll samples of each. In addition, a larger sa.nple from this site
was also included, but had a mixed Mlachallla and Engoroy association.
The material from this site was, therefore, treated in three units;
IMachalilla material, Engoroy remains, and all the bone from the site
together Biompas estimates, fisl species size and nutritional values
were calculated for the site as a whhole (Fig. 15). The faunal lists for
each unit are in Appendix A.
The hiachalilla Phase represents an introduction of some net traits
into the study area. Some researchers see the introduction of a new
people, in part, contemporaneous with the Valdivia Phase inhabitants.
They believe that Machalilla represents a site-unit intrusion into the
area and that the Ilachalillans lived more or less harmoniously with their
Valdivia neighbors (Ileggers, Evans, and Estrada 1965). Other archaeo-
3ogists feel that the Ilachalilla people lived later than the Valdivians
and in fact developed out of the earlier Valdivia Phase (Lathrap 1967;
Faunal remains. The IHachalilla fish remains from OCSE-46D do not
differ in type from those of other Santa Elena Peninsula sites,
although some different species are represented. Catfish are still the
most abundant species. Other species, not found in previously
discussed Santa Elena Peninsula sites, include the dog and the agouti.
The Engoroy material, sometimes included under the Chorrera Phase
heading, represents the only late Formative material available for
iaalyt.is. Tie Chorrera Phase is stated to have evolved out of the
Ear'li.er Macbhlilla and to represent a subsistence shift from sea food
resources to agricultural crops (Willey 1971). Fairly good evidence for
contact w:itlh lesoaimericn is also available for this phase (HIeggers 1966).
Although some sites of Engoroy -Chorrera affiliation might indicate
a decreasing importance in marine foods, this does not seem to be the
case with the Engoroy inhabitants at OGSE-46D. At this site marine
1-- 'N Qo
--'--- \ -J J -
\ I 1 LL
L-- 0-- -- 0
S-.I I I I-
vertebrates continued to be widely exploited. The remains indicate that
a vider variety of species iere exploited by the Engoroy Lilan by the
earlier Hachalilla people. Terrestrial deer and fox were also taken.
/gain there is evidence of doY.
Because of the small size of the Machalilla and Engoroy samples
and their basic similarities with respect to the types of species present
and their relative numbers, the material from these t.'o samples were
combined with the other faunal remains from OCSE-46D. -igures 15 and .16
represent these combinations. The results of the calculations indicate
that marine resources were of primary importance both numerically and
nutritioially. Although the deer estimates in this set of calculations
was based on bone weight and is, therefore very lo;-, the smali number
of deer bones and the abundance of aquatic resources suggest that deer
\as probably not overly important in the diet. Most of the primary
species exploited by these peoples were those also utilized by previous
groups. In general, the size of the fish captured appears to have
increased compared with previously discussed Santa Elena Peninsula sites
(Fig. 16). This might have resulted from the introduction of larger
fishlihooks during "achalilla times.
The other Machalilla site considered in this study is OGCH-20. It
Is located on the Rio Verde, about five kilometers upstream from Chanduy.
This site is of interest since it represents an inland Machalilla
occupation, one that is in the same vicinity as Real Alto, the middle
and late Inland Valdivia site discussed above. The people at the site
presumably utilized the Rio Verde valley in the same manner as the
Valdivian, i.e. by cultivating crops.
Faunal remains. The faunal sample from the site differs from the
;eal Alto material in having an additional mammal, the fox. This
exploitation of masiaials other than dear %as also evidenL at the coastal
OG.SE--46D site. This differs from the general Valdivia exo:-loitation pa3.tera.
Tie Valdivians, when they exploited any mammals at all, relied on deer.
Thil other faunal remlair.s included large numribers of fish. As in most
cases consider-ed in this study, marine catfish was by far the most
abundant fish. Also of importance were the drums and to a lesser ex.:tern
The Guangala Phase represents Lhe local iianifectation of the
P.egiunal Developmental Peri'od. C:arateristics of his:ii period include
"... differentiation in sociopolitical organization, florescence in art
style and elaboration in technology" (Mecger3 1966:67)
At Guailnala sites maize agriculture appears to be -:idespre~ a as the
presence of mane and metate frangents indicate. Interior incised pottery
bo.71,e suggest that manioc or peppers %.ere also grown (Meggers 1966).
:1arine resources continued to be utilized by coastal groups, whilee deer
-.~'2-e hunted in the inland sices (Meggers 1966). Shell fishhooks and
atlacl nooks have been found in the nidderns.
Faunal remains. Tha Guangala faunal material from this site shof.s
a pronounced marine focus. All the idcncifiable remains are of marine
fishes. Particjulrly abundant in the milden are the marina catfish,
al though grunts and puffers are also nurierous. In terrni of actual
jl:;:ritio0ial and caloric values, the single shiL-k from the midden appears
of primary ir,portance. There is no doubt that these Guangalans made
inch Iore use of the available marine vertebrates than the earlier
I-.achalilla and Engoroy peoples who had settled in about the sane area.
The last site representative of the post-Valdivia vertebrate
exploitation on the Santa Elena Peninsula is a very small sample from
a Libertad Phase site, OGSE-41E.
Faunal remains. This site is located near the sea and represents
a Finale phase occupation. The small test conducted into this midden
resulted in very few bones. The sample is again mainly fishes, with
grurita the nost abundant form. With the exception of a snall barracuda
all the species present ihad been found in other Santa Elena Peninsula
sites. Of the very snall sample only one bone could be tentatively
identified as mammal, all other identifiable bones were of fish (Fig. 17).
Over a period of seven or eight thousand years the southwestern
coast of Ecuador has served as the home for numerous peoples. Some of
these groups were simple hunters and gatherers, while e others were
advanced agriculturalists with well developed ceramics and art styles.
Although these populations differed radically in many aspects of their
cultures, they all faced the same basic problem hou to obtain adequate
amounts of protein, fats, carbohydrates, vitar.ins, minerals, and calories.
To fulfill these basic needs these peoples relied upon a variety of
different foods. The relative importance of the various foods differed
frcn group to group.
Protein Scarcity and Protein Acquisition
A variety of different subsistence strategies can be practiced by
peoples to obtain adequate protein for healthy growth and development
(See Chapter II). These strategies include the consumption of large
quantities of protein-rich plant foods, the consumption of complimentary
plant foods, or the consumption of a combination of plant and animal
Although these strategies are all possible alternatives in most
areas, in any given area one strategy is more efficient in fulfilling
nutritional and caloric needs. Animal bone was present at all the sites
examined in this study. This indicates that the subsistence strategy
chosen by these groups included the consumption of animal foods.
In areas whe:e protein is abundant, protein acquisition may have
little direct effect on the culture, but uwhbre iL is limited the obtaining
of protein can greatly influence other aspects of culture. In such
cases tlih need for protein may result in the dispersal of the population
(Holmberg 1969, Carneiro 1973), increase hostilities between groups
cor.petiiig over the same limited resource, development of reciprocal
relations between members within the villages (Gross 1975), or lead
to the formation of long distance exchange networks.
Archaeologically, the degree to which a people were obtaining
adequate amounts of protein, can be reflected in the human osteological
remains. Protein deficiency can limit population growth by increasing
the incidence of miscarriages, spontaneous abortions, and high infant
mortality (llulinski J976). It may further result in the adoption of
population-regulating mechanisms such as female infanticide. Even for
those individuals who survive to maturity, protein deficiency can effect
the overall growth and stature of the individual. By e.an'ining the
human bone remains from a site some indication of how successful a
people were in obtaining protein can be determined.
Thu indicators of protein deficiency may not be readily recognized
by the field excavator. Studies on human growth and stature require
knowledge of human osteology and a familiarity with different growth
patterns. Also, human fetal and infant bones are very different from
adult bone and can be easily missed by excavators not trained in human
osteology. Unless a physical anthropologist is present during excavation
and removes the fetal and infant bones from the sample, the lack of fetal
and infant remains in a faunal sample probably means there were few at
the site. Their presence in large numbers in the faunal remains suggests
At the Valdivia period, Loan Alta site a proportionally large
nuriber of human fetal and infant remains were uncovered (Chapter V).
This is interpreted as an indication th.t these inland people were living
under conditions of protein stress. In an apparent attempt to overcome
Lhis deficiency the inland Loma Alta people tried to obtain additional
r-roteiin by utilizing marine resources. Theyl did this by acquiring f isl
from the coast, probably by means of exchange.
This tyne of exchange between inland and coastal peoples has been
noted in other areas. The exchange between valley and coastal sites was
evident as early as Period 7 (2500-1700 B.C.) in coastal Peru (ilasi:ish
et al 1975). During this period Richard MIacileish believes that the
coastal fishermen "... sent marine protein foods into the valley and
received cultivated plant foods in return" (Mlaclleish et al 1975:33).
Hone of the other sites considered in this study exhibited the same
high proprotion of human fetal and infant remains as was present at Lora
Alta. This could indicate that protein deficiency was not a problem at
these sites. Their location near the protein-ricl coastal waters is
probably responsible for this.
Chaaigees in Protein E:xploitation and Subsistence Orientation
The relative importance of terrestrial, as opposed to aquatic
resources, used b- the prehistoric peoples of the study area varied.
During sore cultural phases, terrestrial resources were more important,
while at others aquatic animals were irore significant in the diet.
Tncse changing patterns of protein exploitation and subsistence
are summarized in Tables 2 and 3. The relative IiniimLum number of
intdi. ideals of terrestrial and aquatic animals represented in samples from
PERCENTAGE OF FOODS F.O:I AQUATIC AIJ'
TE[RI:STRIAL. "A-'dITATS n1!I
Santa Elcna Peninsula
OGSF- OCSE- OGSE- OGSE- OGSE- OGSE- OGSE-
80 63 174 62 62C 46D 46U
aquatic % 55 75 82 99 99 87 98
terrestrial 7 45 25 18 1 1 13 2
sample size 56 12 17 86 SS 48 66
Ilorth of the
Santa Elena P.
Valdivia Loma Alta
aquatic % 88 ,.69 26
terrestrial % 12 _31 74
sample siz- 89 133 23
East of the
Santa Elena P.
Real Alto OCCH20
St. VII St. VIII
95 97 96.8
5 3 3.2
133 116 281.0
the principal sites considered in this study are presented in Table 2.
This first set of calculations indicates how intensively the vertebrates
from the two exploitation areas were utilized by the various cultural
groups. 'The second table (Table 3) illustrates the amount of meat
obtained from these two sources.
Estimates of tlhe weight of the aquatic and terrestrial forms were
based on the weight of the archaeological bone. The bone weight .as
used in the formulas in place of the skeletal weight and the formulas
for skeletal weight to live weight from Appendi:-: C was used. The
perciform fish formula was employed to estimate aquatic resources, the
ma;rrial formula was used for the terrestrial animals. The only site
for which estimates of live weight were not computed in this manner
was the Valdivia site. The bones from this site were partially mineralized,
so weight calculations would be extremely inaccurate. For this site, the
estimates calculated in Chapter V were used.
The data presented in Tables 2 and 3 illustrates a steady shift in
subsistence orientation on the Santa Elena Peninsula. Early cultural
groups relied more on terrestrial animals, while the Valdivia peoples
depended more on aquatic forms. The early post-Valdivia people again
hunter terrestrial animals, but later groups shifted back to almost
e:cclusive auatic oe.-ploitation.
The inland and northern sites sho,' a somew.:hat different orienta-
tion. At these Valdivit sites terrestrial resources are more numerous
and provide far :.lore m2at than aquatic vertebrates. Even at these
sites aquatic resources were widely e::ploited.
Several factors could be responsible for the chan.ning subsistence
patterns described here. These include the introduction of agriculture
PERCEi'AGCI OF FOODS FROM AQUATIC AIL; TR'I'r'.STRJ.AL
Santa Elana Peninsula
Pre-Valdivia Valdivia Pocst-Vald'ivia
OGSE- OGSE- OGSE- OCSE- OCSE- OGSE- OGSE-
80 63 174 62 62C 46D 46U
aquatic % 5 ]0 24 99 99.7 95 99.7
terresLrial % 95 90 76 1 0.3 5 0.3
bior.:a3s (g) 21300 5435 9007 7955 9124 12372 5273
North of the
S :nta Elena P.
Valdivia Loma Alta
East of the
Santa Elena P.
Post-Va d ivia
Real Alto OCCH-20
29 9 16
71 91 84
538951 68915 54633 36567
and the changing climatic conditions. I propose that the change in
emphasis between pre-Valdivia and Valdivia tines is linked to another
subsistence shift, probably the introduction of agriculture.
Although there is little direct evidence of agriculture, indirect
evidence such as storage pits, grinding iiiplements and the locations of
che sites indicates that agriculture had been introduced into the area
and was being practiced during Valdivia times. Crops,e.g. corn, require
periods of fairly intensive cultivacion and care. During these periods
there is less time to devote to other subsistence activities such as
hunting and fishing. ';iis often results in the scheduling of subsistence
activities :.ith certain periods of time devoted to one strategy, i.e.
the cultivation of agricultural crops, and other tines devoted to other
pursuits. This scheduling can ta':e the form of gardening at one time
during the day and hunting and/or fishing at another or a seasonal cycle
of cultivation supplemented by limited hunting and/or fishing in areas
near the village at one period of the year with extensive Punting and
fishing more prevalent in another season.
There is evidence of scheduling from one Valdivia site. At Loma
Alta all the deer that could be aged indicate that they probably were
killed during a particular time of year. This suggests that deer hunting
wns more or less restricted to a particular season.
Although there has been little research conducted on the breeding
cycles of deer in coastal Ecuador, studies on the white-tailed deer from
Venezuala irdicaLe that these species breed year round with a peak in
rating during the dry season (Grokx 1972). The gestation period for
Iorth American Ihite-tailed deer is between 195 and 212 days, with an
average around 202 days (Lowery 197"). If the Ecuadorian deer follow
the sar.,e cyclef, the peak rating would be between Hay and December, the
dry season of present-day coastal Ecuador. This would result in peak
births beL'ue.2; October and July. Since the survival rate of the fa:..ns
xyould be greatest for those born during rainy season, when the lactating
does have abundant food resources, Lho peak in birth and survival of
most fawns- would be expected from January through April. If Janaury is
assumed to be the birth month of the deer at the Loma Alta site, the
individuals that could be aged were killed anywhere from April to
September (Fig. 19). At the other extreme, if they were born at the
end of the rainy season (April) the range would be from July to December.
The onset of the rainy season is also a time of peak agricultural
activities. During this period presumably little time would be available
to engage in substantial hunting endeavors. Instead, more time would be
spent in the planting, cultivating and harvesting of the crops. The
length of this period of agricultural activity is largely dependent on
the crops planted. For corn, anywhere from three to four months would
be devoted to the cultivation and harvesting of this crop. If the crop
Was planted at the onset of the rainy season, January through April
would be devoted to its cultivation.
L.hen the deer hunting season, as indicated from the Loma Alta
material, is compared with the period of corn cultivation a yearly
scheduling of economic activities is suggested. The data on Fig. 19
illustrate this. Agricultural activities, i.e. corn cultivation, uould
be restricted to the rainy season of JanLuary through April with deer
hunting practiced anywhere from April through December.
In addition to th" scheduling of econo.-ic activities, the incro-
duction of agriculture can also result in tl'e expansion of the possible
1 G C C
'-4 1- C1 I
'i-1 J1 C
o .I I
3~f -cr::c Q~
s- '3 '3
sul'sistence base with the introduction of newi cops and an increase in
population size resulting in the shrinkage of the territory exploited by
th-_ inhabitants of a site. This has bean seen ethnographically when
the radius of the ex:p]oitation spheres of the hunters and gatherers
(10 mni.) is compared to the area exploited by horticulturalists (radius
5 I.k.) (OLiggs and Vita-Finzi 1972). This decrease in exploitation area
results in proportionately less terrestrial animals available to hunt.
It would encourage a shift to other protein foods. In the study area
T believe this to have taken the form of increased use of the previously,
underexploited, aquatic vertebrates.
Another possible factor which may explain the shift in subsistence
patterns is climatic change. TheI Santa Elena Peninsula is and has been
an area that is particularly susceptible to climatic fluctuation (Chaitar
IV). In the past there have been several periods much wetter or dryer
than today. Because of the close relationship between climate, flora,
and. faunra, a shift in climatic conditions can profoundly alter the
resources available for exploitation. This necessitates the use of
alternative methods of exploitation, a change to different resources,
or an abandonment of the area.
During periods of extreme wetness or dryness the Santa Elena Penin-
sula was uninhabited (Chapter IV). Presumably under these environmental
conditions, the people were either unw-illing or unable to make the
adjustments necessary in order to continue living in the area. During
tines of minor climatic change extreme action such as the abandonment of
the area might not be necessary. During these periods, by exploiting a
slightly different subsistence base, a people could continue to live
in an area.
As noted in Chapter IV, the ilachalilla and Engoroy peoples occupied
thne peninsula during times of fluctuating cl.iTatic conditions. Immediately
previous to ilachalilla occupation the climate of the Santa Elena Penin--
sula 'w.as very cold and dry. During i'achalilla times it became somewhat
warme:-- but still remained cool and relatively dry. Cold and very dry
conditions existed between the Machalilla and Engoroy occupations and
the climate was cool and dry again in Engoroy times.
As illustrated above (Chapter V) the Iachalilla and Engoroy groups,
considered here, exploited a slightly different subsistence base than
previous groups. I would suggest that this exploitation pattern results
from slightly different climatic conditions and, therefore, different
species availability and densities.
It might be argued that the introduction of a ne-I people, who
utilized a different resource base, is responsible for the different
subsistence pattern reflected in the bone remains. I believe that, since
the sare resource configuration is present during both the aclhalilla and
the later Engoroy times, the migration of new people with a different
e::ploitation pattern cannot be used to explain these difference.
Hunting and Fishing Methods
liow a people obtained their animal protein is another important
aspect of subsistence studies. The faunal composition of a site often
indicates what methods were used to obtain the fishes present. For
example, if the habit and habitats of the fishes that are available in
the area are known, inferencea about the ncLhoaIs used to catch them can
T.he coastal waters of Ecuador today contain many different types
of ilarine fishes. Estuaries and inshore waters are inhabited by large