THE DUNCAN RANCH SITE 1:
EVIDENCE iFOR LATE WOODLAND-INCIPIENT ANTELOPE CREEK
Alice Ann Gustafson
B.A., University of Denver, 1965
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
This thesis for the Master of Arts
Alice Ann Gustafson
has been approved for the
/-g 7- Str
Gustafson, Alice Ann (M.A., Anthropology)
The Duncan Ranch Site 1: Evidence for Late Woodland-Incipient Antelope Creek
Thesis directed by Professor Ruben G. Mendoza
A prehistoric site in Hutchinson County, Texas, near White Deer Creek (a
southern tributary of the Canadian River) has yielded information concerning the
indigenous development of Antelope Creek village communities. Distinctive
ceramic sherds, projectile point types, the plastered floor of a prehistoric
structure, and an extensive lithic scatter suggest that the Duncan Ranch Site 1 may
be one of the early incipient communities linked to the Antelope Creek Phase of
the Texas Panhandle. From the examination of the human ecology, the surface
patterning of artifact distribution, and material recovered from test units, site data
support the hypothesis that Antelope Creek developed from local "Woodland"
groups. This development in turn is the result of interaction with other cultural
groups using the Canadian River as a corridor of intercultural exchange and
This abstract accurately represents the content of the candidates thesis,
recommend its publication.
Ruben G. Mendoza
I am deeply indebted to Kay Adams who was instrumental in coordinating the
initial survey and testing. Without her help and enthusiasm, I would not have
been able to do this thesis. I also want to acknowledge the help of Rebecca Watt,
Tom Hardy, Dan Snyder, Aemon Everett, Linda and Wayne Hughes, Patsy and
Robert Sims, and members of the Panhandle Archaeological Society who spent
hours in the hot Texas sun sifting sand. The kind visit of Dr. Jack Hughes to the
site which alerted me to important features and his sharing of volumes of
information gave this thesis much needed insight into Panhandle archaeology. I
would also like to thank Ed Day and Dr. Linda Curran-Everett for their visits to
the site and Duane Quaitt for reading this manuscript. In addition, this thesis
would not have been possible without the generous help and encouragement of Dr.
Ruben Mendoza and the insightful evaluations, much needed critiques, and support
from Dr. Tammy Stone. The recent suggestions, advice, and encouragement of
Dr. Chris Lintz helped me finalize my last draft. And, my mother, who
supported me with her positive outlooks throughout graduate school, gave me the
inspiration to finish. However, any errors which appear in this document are my
1. INTRODUCTION 1
Organization of the Study and Research Design 3
Previous Research on the Antelope Creek Origins 5
2. THE MIGRATION CONTROVERSY 17
Origin Theories of Antelope Creek Communities 17
Criteria for Determining Cultural Impact of Migration 18
Criteria for Identifying Incipient Antelope Creek Communities 26
3. PREHISTORIC BACKGROUND 32
Cultural Sequence of the Texas Panhandle 32
Late Prehistoric: Neo-Indian/Ceramic Phase 40
Early Ceramic/Woodland 40
Late Woodland/Incipient Village 42
Discussion of Late Woodland-Transitional Sites 51
Discussion of Antelope Creek Sites 62
4. CULTURAL ECOLOGY 72
The Human Ecological System 72
The Biophysical Environment 78
Project Location and Description 85
Soils and Local Deposition 88
Prehistoric Environmental Conditions 94
Lithic Material Resources 96
Fauna and Flora 98
Potential Plant Utilization 103
Plant Resource 104
5. METHODOLOGY 113
Field Methodology 115
Survey Methods 116
Shovel Test Methods 118
Excavation Methods 119
Trench 1 and 2 124
Field Results 126
Survey Results- 126
Shovel Probe Results 129
Test Units Results 129
Unit I Discussion 130
Unit II Discussion 130
; Unit III Discussion 152
6. DATA ANALYSIS AND HYPOTHESIS TESTING 158
Laboratory Analysis 160
Hypothesis Testing 167
Tool Assemblage and Technology 169
Ceramic Technology 177
Subsistence Economy and Technology 185
Ecological System 191
Lithic Resource Procurement 197
Trade and Exchange 198
Regional Developments and Mobility 198
7. CONCLUSION 205
APPENDIX 1 211
APPENDIX 2 214
APPENDIX 3 216
APPENDIX 4 226
1.1 Topographic Map of Site 41HC124 2
1.2 Cultural Regions Which Relate To The Research Area 9
4.1 Geological Formations in the Canadian River System 84
4.2 White Deer Creek and Terrace; Duncan Ranch Site 1 86
4.3 Site Map of the Duncan Ranch Site 1 109
5.1 West Wall Profile, Unit I 120
5.2 Plastered Floor, Unit III 122
5.3 Photograph of Plastered Floor 123
5.4 Tools from the Duncan Ranch Site 1 127
6.1 Map of Selected Sites in the Texas Panhandle 159
6.2 Projectile Points from Duncan Ranch Site 1 161
6.3 Projectile Points from Duncan Ranch Site 1 162
6.4 Ceramics from the Duncan Ranch Collection 180
3.1 Chronological Table for the Panhandle of Texas 33
3.2 Cultural Historical Sequence in the Texas Panhandle 35
5.1 Test Unit I 131
5.2 Test Unit II 134
5.3 Test Unit III 137
5.4 Trench B 140
5.5 Trench X 143
5.6 Trench Y 146
5.7 Trench M 149
6.1 Presence-Absence Traits for Antelope Creek, Transitional, and Early
Woodland Sites. 171
6.2 Ceramic Inventory/Survey Surface Collection 181
6.3 Groundstone Analysis 187
6.4 Metate Analysis 187
During the spring of 1992, a prehistoric site, the Duncan Ranch Site 1
(41HC124) in Hutchinson County, was focus of a preliminary archaeological
survey in the north Texas Panhandle by this author [Figure 1.1]. With the help of
Kay Adams, past president of the Colorado Archaeological Society Denver
Chapter, members of the Panhandle Archaeological Society, three Colorado
graduate student associates, and several family members, the survey yielded
approximately 1100 artifacts including material from the surface and seven shovel
tests. In September 1992 and March 1993, three 1 by 1 meter square units at the
site were excavated which extended through several levels of occupation. Test
units I and II exposed a hearth and related midden features consisting of a large
quantity of fire-cracked rocks, charcoal pieces, mano fragments, and small fire-
altered pebbles. In Test Unit III, a caliche and sand-based plastered feature,
representing a possible pit house floor, was exposed. Two test trenches, lateral
extensions of Test Unit III, were shovel excavated in October 1993. These test
units and trenches added over 1350 artifacts to the site collection.
As a result of the research, survey, and excavations, the recovery of a pit
house floor, the identification of a extensive lithic scatter, and the distinctive
Figure 1.1: Topographic Map of Site 41HC124
7.5 minute series, USGS Topographic Series
Adobe Creek S.E. Quadrangle
ceramic types, a possible developmental process in the cultural history of the
Texas Panhandle was postulated. This process represents the suggested continuity
between an earlier Woodland and Antelope Creek phase occupations.
Organization of the Study and Research Design
The Duncan Ranch Site 1 was examined for its relationship to indigenous
development of Antelope Creek adaptations because initially the site displayed
certain chronological features, notably both Antelope Creek and late Woodland
period diagnostics. By examining the attendant evidence on the site, this study
evaluates the hypothesis that the Antelope Creek phase manifestation resulted from
indigenous development. The alternative hypothesis is that the site does not
support indigenous development. The early component on the Duncan Ranch Site,
therefore, would be a separate occupation showing no continuity with later
Antelope Creek developments. Implications of this alternative hypothesis are that
if this was not an indigenous development, that any Antelope Creek manifestation
in the Texas Panhandle developed elsewhere either at other locations or spread by
migration after A.D. 1100 into the area from other separate and distinct cultural
In an attempt to isolate and define the problem, the author used the following
1. Previous Research on Antelope Creek Origins
2. The Migration Controversy
3. Prehistoric Overview
4. Human Ecology and Paleoenvironment Reconstruction
6. Data Analysis and Hypothesis.
The research design was formulated to address pertinent questions whose
answer augment and contribute to the available knowledge of the region (Dancey
1981). This problem must define the nature of the question and address cultural
change, adaptation, and settlement patterns. And, it should use information on a
regional scale with cross-cultural comparison of archaeological remains. Included
is a review of the literature concerning the natural and social environment relevant
to the research problem, specifically a discussion of previous research, cultural
history, geology, plant, animal, and human ecology. The present and what is
known of past environment systems are included in this section. It was important
to discern those environmental factors which would have affected the preservation,
condition, and visibility of the artifactual remains, and influenced cultural
processes. Lastly, the formulation of hypotheses to express possible solutions to
the thesis problem was a major focus. By this method, a number of possibilities
were conceived, including contradictory ideas. These hypotheses are tested with
archaeological ;data recovered from the site to provide an explanation of the
Previous Research on Antelope Creek Origins
Early research concerning the origin of Antelope Creek communities in the
Texas Panhandle concentrated on its larger Antelope Creek period sites (beginning
A.D. 1200) and compared them to other prominent regional manifestations to the
east and west. Lintz (1984:10-12) suggests that local variation was not a
consideration to early researchers. Early predictions concluded that such an
abrupt development of cultural complexity at known Antelope Creek sites could be
only the result of a migration from another culturally sophisticated area [Figure
1.2]. If not a migration, then local cultures were greatly influenced from ideas
from the east and the west. The attendant developments were not the logical
response to environmental and demographic variables present within the Canadian
River System. Such interpretations promoted the use of taxonomic systems from
other defined cultural areas. Thus, our knowledge and explanation of cultural
process has remained obscured by these factors (Lintz 1984:351-353).
Further clouding interpretations was the fact that early cultural classifications
in the Texas Panhandle were incomplete, especially where temporal organization
was concerned (Lintz 1984). Before the 1920s, only cursory excavations of
Arkansas, Oklahoma, and Texas complexes had been completed. As such, these
cultures were seen in relation to the regional systems to the east and west of the
southern plains; regional and local variation was largely unexplored. Bandelier
(1892:137) was possibly the first explorer to document the Canadian villages (c.f.
Lintz 1986:6). One of the earliest archaeological explorations of significant
Antelope Creek sites was by T.L. Eyerly at the Wolf Creek "Buried City" sites in
the northeastern Texas Panhandle (Eyerly 1907; c.f. Lintz 1986:6). In the 1920s,
Warren K. Moorehead (1921) and J. Alden Mason (1929) initiated a more
extensive expedition into this region, locating over 100 sites and documenting
mainly the prehistoric sites with cord-marked pottery, burials, and the extensive
evidence of prehistoric architectural structures (c.f. Lintz 1984:12-13).
Moorehead referred to the Canadian groups as the "Canadian Valley Cultures" and
considered them a manifestation of a Plains culture that changed as it spread up
the Canadian Valley (Moorehead 1921:10; c.f. Lintz 1984:13). Ronald Olsen and
J. Mason conducted a brief survey of the Canadian River Valley, testing the
Alibates 28 site and visiting other sites in the area. At this time, Mason suggested
that the architecture testified to Southwest Puebloan influence. Mason proposed
that "Texas may be found to hold the key to many important problems concerning
the relationship of Mexican, the Pueblos, the Plains, and the Southwest Indian
cultures (Mason 1929:318; c.f. Hughes and Hughes-Jones 1987:9).
Between 1929 and 1945, Ele and Jewel Baker (1939; 1941), W.C. Holden
(1931), E. B. Sayles (1935), and Floyd Studer (1931), funded by such projects as
the Federal Emergency Relief Administration and the Works Progress
Administration t focused their research on the larger architectural sites of the
Canadian River Valley (c.f. Lintz 1986:10-13). Baker and Baker (1939, 1941) in
their WPA excavations noted room size variation and layout. Based on their
research, they suggested a similarity between the late prehistoric sites on the
Canadian River and the Upper Republican River Valley of the Nebraska. Holden
(1931:43, 1933:39) expanded the excavations of Antelope Creek sites from 1929-
1933, working both in New Mexico and Texas (c.f. Lintz 1984:16-17). Though
Southwest ceramic sherds were recovered and provided chronological reference,
Holden resisted the temptation to use Pecos Classification (Kidder 1927) for this
Texas village manifestation (Holden 1932, 1933; c.f. Lintz 1984). He,
additionally, observed significant differences between Antelope Creek and the
Southwest puebloan expressions. Holden theorized that Antelope Creek cultures
were derived from migrations of Central Plains or Eastern-Prairie groups (Lintz
Floyd Studer, an avocational archaeologist, and later the Chair of Archaeology
at West Texas State University, documented and excavated several Antelope Creek
sites and the Alibates Quarry over the course of fifty years. Studer labeled these
manifestations "Post Basketmaker Culture" (Studer 1931) and "Texas Panhandle
Pueblo Culture" (Studer 1952, 1955; c.f. Lintz 1984). However, Studers (1931)
use of Southwestern terminology was heavily influenced by Southwest Puebloan
taxonomies (Lintz 1984:40). Paradoxically, though Studer attributed their mode
of life to a Southwest puebloan lifestyle, he insisted on a Central Plains or eastern
origin (Studer 1934:81, 1955:94; c.f. Lintz 1986:11-12).
E.B. Sayles (1935) with consent of Holden and Studer, excavated Antelope
Creek Ruin 22; and was involved in locating and excavating other ruins in the
region. Sayles indicated that the "Antelope Creek Focus" was an extension of a
Southwest expansion (Sayles 1935; c.f. Lintz 1984:17). Using a taxonomic
system derived from linguistics, Sayles applied a Southwestern-like "Panhandle
Phase" puebloan culture. This approach now appears flawed as presumed
similarities in languages are not consonant with similarities in culture (Lintz
1984:38; Trigger 1968).
Alex D. Krieger (1946) proposed a comprehensive synthesis of Texas
Panhandle cultures. As the first study to make regional comparisons, Krieger
adopted the term "Panhandle Aspect" for cultures spread across the Canadian
Valley portions; of Texas and Oklahoma. Using Wedels (1941) and
Bakers(1939) descriptions of the Central Plains cultures and the Midwest
Taxonomic system, Krieger defined the Antelope Creek focus (Krieger 1946:74;
Figure 1.2 Cultural Regions Which Relate to the Research Area
c.f. Lintz 1986:15). He specified the more explicit term "Antelope Creek Focus"
for the specific Canadian-based Texas cultures (Krieger 1946; c.f. Lintz 1984:24-
25, 326). Krieger described the Antelope Creek focus in this commentary:
Antelope Creek Focus represents a Plains bison-hunting and agricultural
population basically very similar to Upper Republican which wandered into
the Canadian Valley and there came into contact with Puebloan civilization
as it was expanding eastward toward the end of the classic or Pueblo III
period...[The] relationship seems to have been rather one of trade plus the
borrowing of an architectural style, without affecting the remainder of the
Antelope Creek culture. (Krieger 1946:141-148)
After Studer retired from his position at West Texas State University, Jack
Hughes became his replacement. Hughes early work concentrated on the Archaic
and Woodland cultures in the Texas Panhandle (Speer 1980). He assigned two
defined classifications for Late Woodland cultural manifestations within the
Panhandle of Texas region, designated as the Lake Creek Focus and the Palo
Duro complex (Hughes 1962). The time period of these two foci extends between
A.D. 100-1100 (Hughes 1991). Based on an earlier but similar cordmarked
ceramic tradition and lithic technology, Hughes suspected that the Antelope Creek
...may have developed gradually out of Woodland under Basketmaker
influences to the west and Hopewellian and Gibson influences to the east,
and endured long enough to experience Puebloan influences from the west
and Mississippian and Fulton influences from the east. (Hughes 1969:210;
c.f. Lintz 1984)
Supporting1 Hughes position, other researchers (Duffield 1970; Gunnerson
1987:126, Lintz 1984:366, and Peterson 1990) suggest that the Lake Creek
complex may be a likely candidate as an early developmental stage of the
Antelope Creek village phase. Hughes (1991:26-28) elaborated on other cultural
complexities of Woodland social interactions in the Panhandle. Hughes proposed
that the Palo Diiro cultures based primarily on the southern side of the Canadian
may have been intermediaries between the Mogollon of the Rio Grande and Pecos
rivers and Texas Woodland Lake Creek Focus. He suspected the Lake Creek and
Palo Duro complexes represent a significant interaction which spurred the
developmental stage from Woodland hunter-gatherer to villager-gardener in the
Panhandle (Hughes 1991:27). Lintz (personal communication, 1994) suggests
there is much overlapping between these two cultures.
In contrast to Hughes view, Briscoe (1987; 1989) does not support the Lake
Creek complex as the appropriate group responsible for the later Antelope Creek
phase (c.f. Lintz et al. 1993:41). Briscoes research (1987; 1989) on a Woodland
occupation in Roger Mills County, Oklahoma suggests that Lake Creek complex
was separated by more than 800 years from the Antelope Creek culture.
However, this interpretation was based on radiocarbon dates from only one site.
A series of dates from multiple sites is necessary to define Woodland chronologies
(c.f. Lintz et al. 1993:41). James Couzzourts (1985; 1988) Tascosa Creek,
Maintenance Bam, and Boys Ranch site investigations include several dates
placing Woodland periods in a similar time frame as the Swift Horse site (c.f.
Lintz et al. 19S>3:41).
In the late 1960s two distinct foci the Optima (Watson 1950:7-68) in
Oklahoma and Antelope Creek in the Texas Panhandle were combined as
variations of the same cultural manifestation (c.f. Lintz 1984:29). Green
(1986:186) and Schneider (1969:176; c.f. Lintz 1986) proposed these foci be
included under a regional "Panhandle Aspect" classification. Campbell (1969:504-
507) assigned the Apishapa foci to the Panhandle Aspect cultures as well,
suggesting that the Antelope Creek and Optima cultures were derived from
population movements of people from Southeast Colorado to the Texas Panhandle
where they met with the Custer/Washita people and established an occupation in
the Canadian Valley. He based his hypothesis on comparison with an earlier
known architectural tradition of Southeast Colorado, identified as Graneros
(Campbell 1969:504-507; c.f. Lintz 1984).
D.A. Baerreis and R.A. Bryson (1965; 1966) examined cultural adaptation in
the Central Plains and the Texas Panhandle. They postulated the existence of an
extreme climatic drying episode in the Central Plains around A.D. 1200. This
drought forced groups from the Upper Republican River Valley to abandon their
homeland in search of more suitable conditions in the Texas Panhandle. They
reasoned that the Antelope Creek people were previously derived from the Upper
Republican River Valley settlements. In contrast to this theory, other investigators
(Dillehay 1974; Duffield 1970; c.f. Lintz 1984) (Hall 1982; Hughes 1991; Lintz
1984; Lintz et al. 1993:22) support desiccation of the environment in the Texas
Panhandle, beginning at circa. A.D. 1000, supported by an increase in bison
remains on sites. Sites displaying the proposed Central Plains migration which
should have occurred between the two cultural areas have not been reported.
Baerreis and Brysons interpretations have been questioned based on this
discrepancy (Lintz 1984). In addition, Lintz (1978; 1984:29) argues for definite
stylistic differences between the Antelope Creek and the Central Plains
assemblages. No donor complex can be found for the specific architectural
techniques which exist in Antelope Creek sites (Lintz 1984:374-377; Gunnerson
1987:126). Data suggest that Antelope Creek architecture is Central Plains design
overlaid with Southwestern influences (Lintz 1984). Within the variation of
Antelope Creek architecture, a significant diversity exists within this Southern
plains village (Lintz et al. 1993:26-27).
If the Antelope Creek phase developed from the adaptation, diffusion, or
acculturation of a local Woodland group, insufficient radiocarbon dates and site
data in the Panhandle impede our understanding of the developmental changes.
David Hughes (1991:143) suggests that Woodland cultures in the area are
underestimated because of post depositional forces and layering of later village
sites. Recent excavations in the Palo Duro Creek watershed (Peterson 1988; Lintz
et al. 1993; Quigg et al. 1993) has located several important Woodland sites.
But, research documenting Woodland culture would augment our understanding of
this cultural phase, including radiocarbon dating, macro- and micro-botanical
studies, cultural1 organization, and settlement patterns.
Neighboring regions, such as Oklahoma, West Texas, and Colorado, have
yielded incipient village developments, logical outgrowths of local adaptations
(Baugh 1984; Campbell 1969; Collins 1966; 1971; Hofman 1984). Specifically,
Southeast Panhandle sites excavated by Cruse (1992) and Prewitt, Inc. (1993)
(Boyd, personal communication 1994) provide excellent evidence for Late
Woodland developments in the southeast Texas Panhandle. Stimulated by the
influence of Mogollon trade and exchange, a systemic village development
involved in networking relationships seems to have developed.
Lintz (1986:25-37) assigns the Antelope Creek, the Buried City, the Apishapa,
the Raton section of Antelope Creek and the Oklahoma branch cultures to a broad
classification called the Upper Canark variant. The Upper Canark cultures are
broadly defined while regional variations separate each manifestation; trait
similarities appear to be only generally based. According to Lintz research,
stylistic differences of the Antelope Creek phase, particularly in architecture and
material assemblage, negate an origin from Apishapa, the Central Plains, or the
Southwest cultures (Lintz 1984). Hughes and Hughes-Jones (1987) assign the
Buried City finds on Wolf Creek to a specific but separate complex. Lintz et al.
(1993:45) suggest these cultures are differentiated by organization at band-level
from other Antelope Creek phase groups on the Canadian River.
The focus of this study is to examine the indigenous development hypothesis
of Antelope Creek manifestation from the testing of the Duncan Ranch Site 1
(41HC124) in Hutchinson County, Texas. Six research problems, or domains, are
defined in this report. These domains will be addressed in this report and are: 1.
Previous research on Antelope Creek Origins (as addressed in the beginning of
this chapter); 2. The Migration Controversy; 3. Prehistoric Overview; 4.
Paleoenvironmental reconstruction and Cultural Ecology; 5. Research Design and
Methodology; and 6. Data Analysis and Hypothesis Testing.
In the 1920s, research in the North Texas Panhandle began documenting
prehistoric sites with ceramics, burial, and extensive structures. However, as a
result of concentrating on only the later larger sites, these researchers proposed
conflicting assessments of the origins of Antelope Creek communities. The
manifestation was often attributed to a migration from other more sophisticated
cultural areas. The exact origin remained obscured by the variability and melding
of different traits found at many Antelope Creek sites. Theories of climatic
conditions during this period have been used to support and negate theories of
migrations to this region. The most recent research on Antelope Creek sites
(Lintz 1978, 1984, 1986) has looked at architectural variability in these
communities and defined the Antelope Creek phase, in association with the
Apishapa, the Raton section, and the Oklahoma branches, under a broader
category called the Upper Canark variant. Lintz (1984, 1986) and Hughes (1962)
attributed Antelope Creek communities in the Panhandle of Texas to the
indigenous development of a local group. Despite Lintz and Hughes work the
migration theory is still important (Briscoe 1987; Baerreis and Bryson 1966;
Campbell 1969; Harrison personal communication, 1994; Keller 1975). The issue
has never been resolved and needs more in depth and comparative work on
specific assemblages of early occupations in the Texas Panhandle to clarify this
important issue (Hughes and Hughes-Jones 1987: 26; J. Hughes 1991; Lintz 1984;
Lintz et al. 1993:41-47).
THE MIGRATION CONTROVERSY
Origin Theories of Antelope Creek Communities
Origins of Antelope Creek are suggested to be the direct result of migrations
by unspecified Eastern cultures (Holden 1932, 1933; Studer 1934:81; 1955:94;
Kenner 1969), Central Plains groups (Moorehead 1921:21; Wendorf 1960;
Krieger 1946:74), and unspecified Southwestern groups (Sayles n.d.; Studer 1931)
(c.f. Lintz 1986:218). Lintz (1984:352) reports an array of confusing theories
surrounding Antelope Creek origins. These include cultural diffusion westward
from the Gibson (Caddoan) aspect (Bell 1961; Hobbs 1941; Keller 1975),
migrations of Upper Republican groups from Nebraska (Baerreis and Bryson
1966; Watson 1950; Wedel 1959), Apishapa migrations from Colorados
Chaquaqua Plateau (Campbell 1969), a coalescence of the Apishapa and Upper
Republican peoples (Bousman 1974), cultural affiliation with early Basketmaker
peoples (Krieger 1946; Studer 1931; 1952), or a combination of Upper Republican
and Southwestern groups (Krieger 1947; Spencer and Jennings 1965:80). These
ideas are still supported by some researchers; though, in general it is suspected
that an unspecified local Woodland group formed the cultural base for the
Antelope Creek phase.
Criteria For Determining the Cultural Impact of Migration
Evidence for determining migrations in prehistory is difficult to assess
archaeologically; however, there are certain patterns which are significant. Emil
Haury (1958:382) has outlined a set of criteria which he deems important for the
identification of migration in prehistory. These criteria include:
1. If there suddenly appears in a cultural continuum an aggregation of traits
readily identifiable as new, and without local prototypes, and
2. If the products of the immigrant group not only reflect borrowed
elements from the host group, but also, as a lingering effect, preserve
unmistakable elements from their own pattern.
The likelihood of migration is increased if the following criteria are met:
1. If identification of an area is possible in which this constellation of traits
was the normal pattern, and
2. If a rough time equivalency between the "at home" and the displaced
expressions of the similar complexes can be established.
More refined patterns to support migration into a new area (Rouse 1958:64;
c.f. Lintz 1984; Rouse 1986) include:
1. Specific locations in the study region which seem distinct from the local
2. The source or homeland must be identifiable.
3. Contemporaneity of the cultural units must be established.
4. There must have been favorable conditions for migration.
Betty Meggars (1971; c.f.Lintz 1984:353-354) has proposed additional tests
1. The "donor" culture must be temporally earlier than the "recipient"
2. A complex of traits, rather than a single trait or few traits should be
3. The specific traits should reveal a long period of development in the
homeland region and appear fully developed with little indication of
experimentation in the recipient region.
James Deetz (1967:96-97) argues that in the case of migration, we should see
"complete templates connected with both sexes present in both the culture of
origin and the culture which presumably was the end result of the migration." If
the migrant culture is assimilated into the indigenous culture, a combination of the
elements inherent in both cultures might be evident, but only as whole
assemblages (Deetz 1967:96-97).
According to Lintz (1984:351-382), cultural patterns of Antelope Creek
manifestation have not demonstrated sufficient similarities with any of the
proposed places of origin to justify the claim of a migration. Instead, the timing
of the proposed migrations appears to coincide with a general shift from
Woodland to Village phase developments throughout the Southern and High Plains
The migrations into the Canadian Valley suggested by many researchers
(Baerreis and Bryson 1966; Keller 1975; Krieger 1946; Watson 1950; Wedel
1959) are based on similar subsistence strategies, general architectural plans, and
the numerous related ceramic, bone, and stone tools (c.f. Lintz 1984a:336-337).
The Antelope Creek phase, though similar in material culture, differs from Upper
Republican construction in its specific masonry architecture: the development of
contiguous structures, the presence of floor channels to segregate work areas, and
the occasional use of raised platform structures. Other specific traits of the
Antelope Creek phase are bison tibia digging sticks and notched rib rasps, and
these appear in contrast to cultural traits of Upper Republican Valley, such as
bone beamers and gorgets, antler bow guard bracelets, and fish hooks. If present,
ceramic types can be differentiated by the Upper Republican incised "herringbone"
design and "x" motifs as opposed to the Antelope Creek vessels and their dot
punctate design on the rims or lips, diagonal punctation on the lips, or finger-
gouged dots along rims (Lintz 1984:356). Collared rim sherds with similarities
between the two. cultures are dated to A.D. 1300-1350 which places the trait later
than the proposed migration (Lintz 1984). Lintz (1984:356) suggests similarity of
particular traits between these two cultures is actually more generalized. Often,
these traits characteristics, such as side-notched projectile points and hunting and
horticulture subsistence, occur from North Dakota to North Central Texas and
Western Oklahoma (Lintz 1986:220).
To defend an Upper Republican cultural origin for the Canadian River Valley
cultures, Baerreis and Bryson (1966) suggest that, as the climate at A.D. 1200 in
the Central Plains changed drastically, spurring a quick and decisive migration
into areas of like environments (the Canadian Valley). Baerreis and Bryson
(1966) argue that sites located on the migration pathway may not be found as they
had too limited an impact on the landscape (c.f. Lintz 1986:16-17). Lintz disputes
the Central Plains explanation and posits that cultural similarities may actually
reflect diffusion related to trade, exchange, networking, and information sharing
(Lintz 1984a:337). Conversely, Harrison (personal communication, 1994)
suggests that the migration from the Upper Republican Valley to the Canadian
Valley may have occurred earlier than previously speculated, circa. A.D. 900.
However, this scenario does not conform to our present knowledge of Central
Plains cultural patterns. Gunnerson (1987) suggests that a pattern of fusion and
fission, not abandonment, in the Central Plains would better explain the Middle
Ceramic "shifts of population centers-of-gravity" (Gunnerson 1987:120).
Bell (1961) and Keller (1975) suggested that late Gibson Caddoan peoples
followed the buffalo and better climates up the Canadian River and settled in the
Antelope Creek I area. Population pressures in the Southeast and presence of large
herds of bison in the Canadian River Valley account for the proposed migration.
The proposed migration began in wave migrations around A.D. 1200 from the
Caddoan area. Scarcity of bison remains in Late Woodland sites in the Panhandle
of ,Texas, dated between A.D. 1000 or 1200-1300, and an increase in such
remains thereafter were given to support this hypothesis. Recent archaeological
research in Oklahoma (Baugh 1984; Hofman 1974, 1984) supports a Woodland to
Village cultural sequence on the Washita and Canadian Rivers of Oklahoma.
Here, village communities were establish earlier in areas postulated as migration
pathways. From this research, Baugh (1984) and Moore (1983) have questioned
using the culturally derived "Plains Woodland" categorization. The first evidence
of com agriculture on the plains now has been located on the periphery of the
Central Plains, (i.e. the Lodaiska site near Morrison, Colorado (Irwin and Irwin
1959). Village phase cultural developments may have spread eastward from the
Southwest into Colorado and the Texas and Oklahoma panhandles, instead of
westward from the Central Plains or the Eastern cultural centers into the Texas
Panhandle (Baugh 1984:23; Cruse 1992). And, radiocarbon dates from Caddoan
and Southern Plain-Prairie areas do not support the hypothesis of a westward
expansion (Lintz 1984:355).
Campbell (1976) proposed a migration of Apishapa peoples from the
Chaquaqua Plateau in Colorado (c.f. Lintz 1978; 1984:357-358). He based his
migration hypothesis on the similarity of a Graneros Focus circular structure from
southeast Colorado with what he determined were early Antelope Creek circular
structures. Though circular structures appear in both Antelope Creek and
Apishapa masonry construction, Lintz(1978; 1984; 1984a) research on
architectural variation and chronological radiocarbon determinations place these
circular structures later than the more elaborate rectangular structures of the
Antelope Creek phase. The circular dwellings within Antelope Creek
communities appear more related to variability in construction than demonstrated
early forms (Lintz 1978:327, 1984, 1986). Also, homestead and hamlet sites, an
important aspect of Antelope Creek community patterns, have not been found in
the Apishapa sites. Burials at Antelope Creek sites are often capped with piles of
stones, not associated with southeast Colorado burials, though subsistence,
projectile points, snub-nose scrapers, globular cord-marked pottery, slab metates,
oval manos, and flexed burials suggest relationships. The more specialized traits
of Antelope Creek cultures, such as diamond-shaped beveled knives, "guitar-pick"
scrapers, pin drills, and other pertinent chipped stone tools, are not associated
with Apishapa assemblages (Lintz 1984:357-358). Apsihapa cultures appear
related but their existence does not explain the appearance of Antelope Creek
cultures on the Canadian River.
The large contiguous structures in Antelope Creek sites suggested to early
researchers a Southwestern affiliation. In general, the large formalized ceremonial
structures, multiiroom contiguous dwellings with tunnel entrances, east to west
traffic channels, and casual field houses do display some similarities with
Puebloan designs. However, a more detailed description of room patterns, altars
designs, and central channels reveal important differences between the two cultural
approaches. Rather than a source, Lintz (1984:327) interprets a basic Central
Plains-type construction technique overlain by Southwest influences (Lintz 1984:
327). Lintz summarizes his research on Antelope Creek phase origins as:
...adjacent cultural manifestations are not of the proper age, or do not
display the appropriate architectural and artifactual trait complex to be the
potential source of groups immigrating on to the Plains. Instead, the Lake
Creek complex, an indigenous Plains Woodland adaptation in the Texas
Panhandle, is regarded as a likely antecedent of the Antelope Creek phase.
This conclusion is based primarily on the continuity of a cordmarking
ceramic tradition, and presumed parallel developments in Plains Village
transformations which occurred elsewhere along the eastern margin of the
Plains. Antelope Creek thus reflects the introduction of new architectural
ideas around the same time that a local Woodland group was being
transformed into a Plains Village Complex. (Lintz 1984:384)
This conclusion is further supported by a comparison with a Pecos Pueblo
population; present osteometric and dentition data from skeletal remains indicate a
closer phenotypic relationship between Antelope Creek people and those of
Central Plains groups (Paterson 1974). Also, more discrete data includes results
on cranial testing from Antelope Creek, Upper Republican, and Washita River.
Such materials display more similarity to Plains skeletal features (McWilliams and
The occurrence of early villages in the north Panhandle may represent the
resilient and efficient adaptation of a local community on the Canadian River.
This community, set between adjacent cultural areas, changed as travelers, goods,
information, and ideas passed through the Canadian breaks. An examination of
the cultural strategies and the prevalent environmental systems as conditions for
change will be presented in the following chapter. It is necessary to examine in
depth a Woodland community and its relationship to the following more
technologically complex cultural phase. Stimulation for this change may have
resulted from those societal relationships with non-local groups who came into the
area looking for Alibates lithic resources and game. Thus, cultural systems play
an important part in the process of change as well as environmental stress. In
contrast, the Woodland communities may have resisted change, leaned toward
homeostasis, and disappeared.
The ambiguity and longevity of the migration theory continues into the present
literature because the village community appears without precedence in the
archaeological record, sites are more abundant, developed, and infused with
Southwest and Central Plains traits than expected (Lintz 1986:2). The
concentration of archaeological activity on the larger Antelope Creek sites and a
dearth of written reports and excavations of late Woodland sites keeps the
migration theory in debate. Hughes (1978:208-210) suggests that there are many
unreported "Woodland" sites in the North Texas Panhandle which show local
developmental traits consonant with descriptions of Late Woodland cultural
change. The Duncan Ranch site 1 may represent one of these transitional sites.
Criteria For Identifying Incipient Antelope Creek Communities
To research and test the Duncan Ranch site and its socio-cultural relationship
to the Antelope Creek phase, criteria important for identifying the transformation
of an indigenous culture were compiled. By establishing such criteria, the
hypothesis of indigenous development is examined in this report for its positive or
negative relationship. The cultural dynamics of the research area, such as
technological, chronological, social, and paleo-environmental data, are important
to this analysis. The Duncan Ranch Site 1 appears to contain those elements
which are necessary to examine Woodland cultural development. Relevant criteria
to expore such developments were gathered from Plog (1974), Trigger (1968),
Lintz (1984), Lintz et al. (1993), and Quigg et al. (1993). They include the
1. Some loverlapping of technological traits between the local indigenous
culture and the later Antelope Creek developments should be evident.
This evidence includes similar traditions in the utilization of raw
materials, tools, and settlement patterns. For example, we should have
evidence of important continuity of a ceramics tradition, displayed in
the stylistic and technological continuity between earlier and later forms.
And, traits should be evident, as the appearance of a combined mix or
over-lapping of small basal notched points with small side-notched
2. There should be evidence or indications of incipient horticulture.
Hunter/gatherer lifestyles would be maintained equally in conjunction
with horticultural practices.
3. There should be evidence of a local ecological system which may be
critical to transitional community developments (e.g. arising
territoriality, exploitation of local fauna, experimentation with seed
harvesting, and population growth and density in the region).
4. Radiocarbon evidence would support an occupation dated to between
5. Sedentism in the form of permanent architectural features, however
rudimentary, would be observed on some transitional sites. These
structures would be the basis for later elaboration of more sophisticated
architecture (e.g., upright masonry walls, semi-subterranean rooms,
threshold collars and central channel, storage bins and platforms) (Lintz
1984:366). And, these features should be found to predate the onset of
the Antelope Creek phase (A.D. 1200).
6. Transitional cultures would be aware of and use locally obtained
resources for the production of tools, (i.e. lithic resources such as
employed by Antelope Creek peoples including specifically Alibates
Flint and also Tecovas Jasper, Ogallalah Quartzite, and Potter Chert).
7. As diffusion may represent reasons for change, the presence of non-
local materials including obsidian and ceramics on Woodland,
Transitional, or Antelope Creek sites would indicate limited trade or
contact with other cultural areas, such as with the Southwest Mogollon,
Central Plains Keith focus, Apishapa, and Eastern Prairies groups. It is
possible that the onset melding of these cultural forms would be present
on transitional sites.
8. Neighboring areal developments would support a regional shift in
Woodland to village phase cultural change.
9. No evidence of abrupt intrusion, rapid cultural change, or conflict
should be found in the stratigraphic record of the site in question.
Continuity between an earlier and later culture should result in a continuum of
processes and solutions. If a relationship between the older and newer culture
contains a such cultural processes and adaptive patterns, an evolutionary
continuum should be considered (Trigger 1968:356-358). If there is no
pronounced deviation, sudden gap in the stratigraphic record, and/or elements of
forced intrusion, the hypothesis of indigenous development will again be
supported. Lintz adds the need for "recognition of clear assemblage segregation
in stratified situations" (Lintz et al. 1993:41-47).
In the Texas Panhandle, relatively few intermediate or transitional assemblages
have been reported. One explanation may be a short duration of the transitional
phase or its lack of visibility, small sample size, or the insistence on the part of
many investigators to only excavate those sites with obvious architecture (Lintz
1984a:338). Additionally, the change may have been very brief, thus clouded by
later more intense cultures. And, these cultures continued at least a part of the
previous lifestyle. Discontinuities may only be a state of current archaeological
data and research.
Indigenous development of Antelope Creek phase on the Canadian River is
still disputed regardless of research in neighboring areas (e.g., Western
Oklahoma) (Baugh 1984) and noted Late Woodland sites and possible early
Village sites in the Texas Panhandle, e.g. Palo Duro and Lake Creek sites
(Hughes 1962,; 1991; Lintz 1984), including 41M05, 41PT29 (Green 1986), the
Sandy Ridge site (Quigg et al. 1993:117-214), the Middle Cheyenne, the Farley
Boys Ranch, the Maintenance Bam (Couzzourt 1982), and Tascosa Creek sites
(Couzzourt 1985; 1988), Site A16 (Speer 1986:7), the Kent Creek site (Cruse
1992), Sites Al242, A1266, A1275, A1286 (Area 1, Area 4) in the Red Deer
Creek Watershed (Hughes et al. 1978), the Blue Clay site (Willey et al. 1978), a
possible early Blackdog Village-Structure 5 (970170) (Keller 1975:63), and the
Deadmans Shelter site (Willey and Hughes 1978). Blackdog Village-F.17 (A.D.
880-860 +. 210) and Canyon Country Club Cave Nl/Wl, Level 3 (A.D. 690 +.
65) are MASCA corrected tree-ring early dates at known Antelope Creek sites
(Lintz 1984a:342-344). Few comparative assemblage studies have been completed
on Woodland assemblages from Texas Panhandle sites (Lintz et al. 1993).
Explanation for Antelope Creek variation involves possible intercultural
relationships which developed during previous periods. Itinerant craftsmen,
merchants, pilgrims, and ambassadors are potential carriers of information, traits,
and ideas, but these groups and individual contacts do not represent a mass
movement or takeover by any specific group. Importantly, incipient village
groups can display elements of change related to contact with other groups but the
indigenous cultural system will remain intact (Trigger 1968:357). Certainly,
control over such a valuable resource, such as Alibates chert, was an important
factor in the development of elaborated trade and exchange networks in the
region. Also, the spring-fed terraces of the Canadian tributaries provided
propitious environments and protection from the vagaries of the weather system.
And, as humans are potential mimics, successful adaptations (such as horticultural,
ceramic technology, and architecture) are often emulated but locally expressed in a
variety of ways! (Butzer 1984:28; Plog 1974). Lintz (1984:374) argues that the
concept of "unregulated diffusion" as defined by Schroeder (1966) may best
explain the mechanism providing the unique variability found at Antelope Creek
The origin of Antelope Creek communities in the Texas Panhandle has been
attributed to migration of Upper Republican, Caddoan, Southwestern, or Apishapa
groups or a combination of those cited. Several researchers have outlined criteria
for determining the influence or takeover of a migrant group in an archaeological
assemblage. Conversely, Lintz argues that most trait similarities between the
cultures in question are generic. Specific traits of Antelope Creek appear to
suggest that the community structure was a logical outgrowth of a local indigenous
community. Criteria to test this hypothesis at the Duncan Ranch Site 1 includes:
1. The overlapping of technological traits between the local indigenous
culture and the later Antelope Creek development.
2. Evidence for experimentation with horticultural practices.
3. Ecological compatibility.
4. Radiocarbon dates which support a pre-Antelope Creek dated to between
5. Sedentism in the form of architectural features on some transitional sites
which predate the Antelope Creek phase.
6. Use of local resources for the production of tools.
7. Evidence of possible diffusion from trade and contact with other
pertinent areas would support the flow of information.
8. Evidence of neighboring areas developing similarly from local
Woodland to fully established village adaptation.
9. No evidence of an abrupt intrusion, rapid cultural change, or conflict.
1 Cultural Sequence of the Texas Panhandle
Prehistoric! cultural developments in the Panhandle of Texas can be
summarized in terms of four distinct horizons: Paleo-Indian, Archaic, the Late
PrehistoricNeo-Indian or Ceramic, and Protohistoric [see Table 3.1]. These
subdivisions are widely recognized chronological periods in the Southern High
Plains. For the purpose of discussions that follow, I will briefly summarize the
primary characteristics of these phases, with the addition of a more defined Late
Woodland to Incipient Antelope Creek transition during the Neo-Indian/Ceramic
period in the study area.
Paleo-Indian exploitation is well documented in the Southern High Plains.
Paleo-Indian type-sites which supplied the taxonomic nomenclature for major
cultural groups are located in the region; these sites include Clovis, Folsom,
Miami, Midland, Plainview, Milnesand, San Jon, and Rex Rogers (Willey et al.
1978a; Bertram T989:11). From 10,000 to 5000 B.C., Paleo-Indian groups
traversed the region known as the Llano Estacado and the Panhandle Plains. It is
Table 3.1: CHRONOLOGICAL TABLE FOR THE PANHANDLE OF TEXAS
Paleo-Indian Clovis Folsom Plano 10,000 B.C. 5,000 B.C. 9,500 B.C. 8,400 B.C. 8,800 B.C. 8,200 B.C. 8,200 B.C. 5,500 B.C.
Archaic Early Middle Late Archaic 5.500 B.C. A.D. 200 5.500 B.C. 3,000 B.C. 3.000 B.C. 1,000 B.C. 1.000 B.C. A.D. 300
Neo-Indian/Ceramic Early Ceramic A.D. 200 A.D. 1600 A.D. 200 A.D. 700 (Early Palo Duro Complex components, Early Lake Creek components, Wolf Creek sites)
Late Woodland-Incipient Village c
A.D. 700 A.D. 1100/1200 (Late Lake Creek
Focus sites and Palo Duro Complex sites, possible Red Deer Creek Watershed sites)
Antelope Creek Phase Early Middle Late A.D. 1200- A.D. 1600 A.D. 1200 A.D. 1250 A.D. 1250 A.D. 1350 A.D. 1350 A.D. 1450/A.D. 1600
Protohistoric Tierra Blanca A.D. 1650 A.D. 1850 A.D. 1450 A.D. 1700 (Tierra Blanca complex sites, Garza complex sites)
likely these nomadic bands planned their movements in accord with the migration
patterns of the megafauna and other animals traversing the Southern Plains.
Hofman et al. (1989) assign these hunters to a "Early Specialized Hunters
Adaptation Type" (c.f. Lintz et al. 1993:24). The Paleo-Indian stage is separated
into three periods based on distinctive point types and associated extinct fauna.
These three periods include the Clovis, the Folsom, and the Plano. On the
southern High Plains, Clovis sites (9,500-9,000 B.C.) are usually recognized by
their distinctive lanceolate points and formal tools associated with the butchering
and processing pf animal remains (Lintz et al. 1993:24). Large herbivore
(mammoth) remains and other extinct animals (i.e., ancient horse) are associated
with these early scavenger/hunters. An occupation predating the Paleo-Indian
stage (before 9,500 B.C.) in the Texas Panhandle has yet to be confirmed.
Between 8,800-8,200 B.C., Folsom hunters, in small bands consisting of a
dozen individuals, exploited Bison antiquus with communally-planned, hunting
strategies (Lintz et al. 1993:24). However, as the environment and climate
changed, the large megafauna disappeared. Resulting from these conditions,
human adaptive strategies altered by the end of the Plano period and the advent of
the Holocene. New tool technologies and subsistence economies appear in the
archaeological record. The surviving human groups, a continuum of late Paleo-
Indian (Plano) and Early Archaic hunter/gatherers, exploited the smaller bison
HORIZON PERIOD CULTURE PHASE EXAMPLE SUBSISTENCE SETTLEMENT
Hiitoric A.D.1870 Horse nomads Euro-Amer. Comanche Bison/hunting Village/camps
A.D.1S60 Historic Aboriginal Historic Aboriginal Apache -
Protohistoric A.D. 1700 A.D. 1450 Village/Nomads Tierra Blanca Apache Bison hunting Limited Agriculture Village/Camps
Neo-Indian A.D.' 1500/1600 Complex village Antelope Creek Antelope Creek Agriculture Complex village
Village A.D.1150 Buried Cities Courson Site Hunting Farmsteads,
Neo-Indian A.D. 1150 Incipient village Lake Creek complex Lake Creek Intensified Seasonal to semi-permanent
Transitional A.D. 700 Palo Duro complex Undefined sites Kent Creek Duncan Ranch Hunting/gathering Riverine economy Possible masonry structures
Neo-Indian Early Ceramic A.D. 700 A.D. 100 Woodland Lake Creek complex Palo Duro complex Lake Creek Blue Clay Deadmans Shelter Intensified hunting/gathering Seasonal to semi- permanent villages Rock shelters
Archaic A.D. 200 5000 B.C. Late Archaic Middle Archaic Late Archaic Undefined Rex Rodgers Little Sunday Hunting/gathering Rock shelters Bison hunting Temporary camps Seasonal flora/fauna Mobile camps
Paleo-Indian 8000 B.C. Plano Proj. point type Rex Rogers Bison oxidentalis Rock shelters
8400 B.C. 10000 B.C. Folsom Clovis Ass.with megafauna Blackwater Draw Miami Bison Antiquus Temporary camps Kill sites
Table 3.2: Cultural Historical Sequence in the Texas Panhandle
(Bison occidentalism. The Plano tool kit and settlement patterns came to reflect
differences in specialized activities including expanded local foraging and
collecting. The distinct lanceolate blade, a horizon marker for this era,
incorporated parallel flaking patterns and basal thinning (Wormington 1957).
The Archaic stage, beginning approximately 8,000 years ago, ended sometime
after A.D. 1-200 in the Texas Panhandle. In general, this stage is poorly
understood and the assemblage characteristics lack definition (Lintz et al.
1993:24). Environmental changes encouraged herds of the smaller Bison bison
and the demise of the larger megafauna. Archaic cultures, loosely structured on
hunting and foraging strategies, seem to have planned activities to coincide with
seasonal availability of resources and necessities (Hughes and Hughes-Jones
1987:27; Story 1982:144). Carrasco (personal communication 1993) suggests that
Archaic hunter-gatherers were not adverse to limiting their mobility if food was
plentiful. In times of scarcity, groups would use less desirable foods and/or
disperse or fission; in favorable times local populations increased. J. Hughes
(1991:11) suggests ripening of seeds, fruits, and nuts, the spawning of fish, the
movement of deer and bison, and social interactions with other groups dictated
these movements. The Little Sunday Site in Randall County is thought typical of
this stage (Hughes 1974, 1991:11-12). Diagnostic traits indicate a more localized
subsistence pattern. There is increased documentation of mortuary remains.
Though pertinent areal distinctions are in evidence in the archaeological record,
most campsites are located near springs, creeks, lakes, and playas (Speer 1980:4).
Antevs (1968) suggests that a dry "Altithermal" climatic condition extended
across vast areas of the Plains during the early Archaic; eventually the climate
shifted to a more moderate (Medithermal) climatic regime. And, since the onset
of the Medithermal, there has been a general drying of the environment with
periodic cooler and wetter periods. Bison and other large game animals appear to
be affected by these climatic shifts and human groups depending on the bison are
in turn affected (Antevs 1968; Dillehay 1974; Huebner 1990). Because of this
absence of bison remains, Dillehay (1974:180-196) postulated that the bison were
present during the later Archaic periods (Presence Period II: 2500-500 B.C.) but
not in the earlier periods. From the scarcity of remains from Southern Plains
archaeological sites, he postulates another bison absence between A.D. 500-1000
and the return in sizable numbers about 700 years ago (Lintz et al. 1993:22). The
resulting effects of bison migratory changes on the Southern High Plains required
human groups to move or, if less mobile, acquire another form of subsistence.
The early Archaic lithic assemblage in the Texas Panhandle is characterized
by a scarcity of recognizable dart-points but includes gouges (Clear Fork),
hammers, choppers, and increased number of fired-scarred rock cobbles (boiling
stones) (Duffield 1964). Lintz et al. (1993:24) suggests that large, concave base
points and shallow side-notched points (Uvalde) represent Early Archaic diagnostic
points. The remains of these Archaic groups are so scarce that often they seem to
disappear from the landscape. The Rex Rodgers site in Tule Canyon is one of the
only known early Archaic sites in the upper Texas Panhandle (Lintz et al.
1993:24). Atlatl and dart points were usually manufactured from Alibates
agatized dolomite, local and exotic cherts, quartzites, Tecovas jaspers, petrified
wood, and other cryptocrystalline silicates (Hughes 1991:24).
Middle and Late Archaic sites in the Texas Panhandle are located on the
upland playa lakes, along canyon and valley rims, and on benches and terraces of
streams and tributaries of the Canadian and Red rivers. The scarcity of Middle
Archaic sites (B.C. 3000-1000) may indicate poor preservation or actual small
numbers of groups in the area (Lintz et al. 1993:24). Middle Archaic dart points
appear with weak to barbed shoulders and concave or indented bases (Lintz et al.
1993:24). Small groups in the Late Archaic period (1000 B.C-A.D. 300)
continued with seasonal hunting and foraging. Their diagnostic projectiles include
the Ellis, Marcos, and Shumla corner-notched and small side-notched Ensor or
Edgewood dart points (Lintz et al. 1993).
Basket technology is considered a likely technology of these late Archaic
peoples, including clay covered baskets (Cordell 1984; Speer 1980). Ovate and
triangular knives, thick ended scrapers, small manos, thin grinding slabs, and
numerous hearthstones and small "boiling pebbles", gouges, and diagnostic
comer-indented and comer-notched dartpoints are common in the Late Archaic.
Late Archaic bison kills have been reported in the Southeastern part of the
Panhandle (Speer 1980:4). Cited from Lintz et al. (1993:48), two Late Archaic
assemblages are defined: the Summers, Oklahoma and the Bitter Creek in Hall,
County, Texas (Hughes and Hood 1976). Kill sites at the Twilla, Bell, Collier,
and Strong sites (Lintz et al. 1993:25) are reported as representing Archaic-
Woodland transition period. Three Archaic comer-notched projectile points (Ellis-
type) have been identified from the Duncan Ranch site assemblage.
Though technologies as the bow and ceramics have not been documented with
Archaic cultures, it is possible that late period groups experimented with high
organic tempered ceramics (Cordell 1984). The high organic temper of more
mobile groups might have prevented preservation (Skibo et al. 1989). Even so,
the advent of ceramic technology did not appear to change daily life dramatically.
Story (1982) suggests that the effects of collective cultural experiences in
exploiting Archaic environments influenced cultural change. In turn, the resulting
social mechanisms provided impetus for growth and diversity in human
populations. Some of these adaptive mechanisms were perhaps:
Such things as knowledge of medicinally useful plants and in what
quantities plants can be consumed without toxic side effects, as well as the
development of cultural roles such as headmen, traders, and curing
shamans.... (Story 1982:144)
The Late Prehistoric: Neo-Indian/Ceramic Phase
Neo-Indian horizon or Ceramic phase originated in North Texas at
approximately A.D. 200 and ends with the Protohistoric Tierra Blanca stage.
This phase in cultural development has been referred to as Neo-American (Suhm
et al. 1954), Neo-Indian (Willey and Hughes 1978), Formative (Willey and
Phillips 1959), or Ceramic (Johnson 1987) due to the existence of ceramic
technology. In recent studies, the term Neo-Indian has been replaced by Ceramic,
Formative, or Late-Prehistoric as it less clearly defines cultural variability
(Johnson 1987:24; Lintz et al. 1993:26). However, since these terms are still
regionally specific and often used in the Panhandle, this report will use the Late
Prehistoric Neo-Indian and Ceramic classifications interchangeably.
Since onset of the Neo-Indian/Early Ceramic subphase, Woodland-like
adaptations in the North Texas Panhandle have been defined by Jack Hughes
(1991) as beginning around A.D. 100-400. A suggested A.D. 200 is noted as a
compromise between these two dates (Hughes 1991:26-27). Two notable regional
complexes, the Lake Creek and the Palo Duro, represent such Woodland
occupations in the Texas Panhandle. It has been argued that these early ceramic
complexes display some similarities to the Mogollon cultures of the Southwest,
Basketmaker cultures, the Graneros Focus of Colorado, the Central Plains Keith
focus of Kansas and Nebraska, and the Eastern Prairie Gibson aspect cultures
(Bell 1961; Campbell 1976; Couzzourt 1987b; Cruse 1992; Hughes 1991; Wedel
1959), hince the ambiquity of origin arises (c.f Lintz 1984, 1986; Lintz et al.
1993:46). An extension of Archaic type adaptations define these occupations,
including technologies as, comer-notched, basal-notched arrowpoints, and dart
points (Ellis). But, addition of a Plains-like cordmarked ceramics and brownware
separated these cultures from previous Archaic technologies.
Two middens were located in Deadmans Shelter, a Palo Duro complex site.
The earliest midden was radiocarbon dated to A.D. 12+_ 60 and A.D. 210 _ 40
(SI-1899 and SI-1900) (Baugh 1984:23-25). Lake Creek complex sites do not
have adequate numbers of radiocarbon dates. Briscoe (1987, 1989) examined the
Swift Horse site in Western Oklahoma, an early Lake Creek complex site (ca.
A.D. 450) with only small comer-notched points and early radiocarbon dates of
circa A.D. 500 (Lintz et al. 1989:45) (Thurmond 1991:250). The Twilla site
(Lintz et al. 1991) yielded two dates from bison bone fragments (1290+95) and
(1120+100) (Thurmond 1991:250) but is linked to a possible Archaic-Woodland
Hughes (1991:27) notes that the environment of the Panhandle at A.D. 1 was
notably more mesic and continued being moist until A.D. 1000. These climatic
conditions created increased biotic environments which could support foraging and
semi-sedentism. Maize has recently been documented in early ceramic contexts in
the Palo Duro Creek area (Lintz et al. 1993:26; Peterson 1988).
Late Woodland/Incipient Village
This reports defines the Late Woodland substage as including an incipient
village adaptation in the Texas Panhandle. This adaptation constitutes the
transition to village with the introduction of permanent architectural structures,
though rudimentary, e.g., Kent Creek site (Cruse 1992) and Sam Wahl site,
Prewitt and Associates, Inc. (Lintz, personal communication 1994). An increase
in the numbers of large manos and slab metates and other groundstone artifacts,
improved storage facilities, and ceramics around A.D. 700-800 are indirect
evidence of this increased sedentism in the Texas Panhandle and western
Oklahoma (Cruse 1992:22-23). Experimentation with feral and early horticultural
plants are likely given the presence of ceramics and a mix of one and two-hand
manos. These tools increased the efficiency of storing, cooking, and processing
of vegetal and faunal foods (Cruse 1992). These more efficient technologies
increased the possibility of lifestyles organized around local semi-permanent
Though presently lacking chronological parameters and architecture, Hughes
(1962) assigns the Lake Creek site to this period of development. Based on the
similarity of cord-marked ceramics, Hughes (1962; 1969:210) suggests the Lake
Creek complex may be the local culture from which developed the Antelope Creek
phase. Relevant traits include ceramics displaying Mogollon or Eastern Jornada
brownware ceramic characteristics, and a local Woodland cordmarked pottery
type. Diagnostic tools found in Woodland contexts are comer-notched (Scallom)
arrowpoint and corner-notched (Ellis) and side-notched (Reed) dart points (Hughes
1962). J. Hughes suggests that the scatter of Scallom, Deadman, and other
barbed arrow points "blanketing" the Southern Plains in conjunction with
Mogollon type brown ware and Woodland cordmarked pottery indicate
interactions of peoples from different cultures (Hughes 1991:28). Dart points may
indicate the continuation of late Archaic technologies.
Interactions with the Mogollon cultures, specifically the Jornada, has taken on
new importance in the Panhandle with the location of Mogollon influenced
structures in association with pertinent brownware at the Kent Creek (Cruse 1992)
and Sam Wahl sites (Lintz, personal communication 1994). Couzzourt (1987b)
also finds evidence of interaction with Central Plains groups with the presence of
Keith focus calcite tempered sherds on some Wolf Creek Woodland sites. Keith
focus spans the period from A.D. 200 to 1000 (c.f. Lintz et al. 1993:47).
In Northeast Texas, population increases were occurring at this time along
with interregional conflicts (Watt 1956; Lynott 1981). Caddoan intruders were
moving into this region and bringing their bow and arrow technology, a definable
ceramic style, and maize horticulture. New methods of storage and domesticated
dogs, which they used for hunting and food, were a part of their culture (Speer
1986). In the North Texas Panhandle, there has been no overwhelming evidence
of conflict involving Caddoan groups which would represent replacement or
abandonment of established Woodland communities (Lintz 1984; Hughes and
Around A.D. 1100, a pit house to above-ground masonry transition and
agricultural involvement developed within the Southern Plains (Hughes 1991).
Early Woodland sites (A.D. 100-600) (Quigg et al. 1993:175; Lintz et al. 1993)
are predominantly associated with small corner-notched (Lake Creek sites) or
basal-notched points (Palo Duro sites), where as side-notched forms indicate
cultures beginning after A.D. 1200 (Lintz et al. 1993). A period of transition
defines an intermediate stage on the Southern Plains, where comer to side-notched
forms dominate the material assemblages (Hofman 1978:6-35, 1984; c.f. Quigg et
al. 1993:175). Relevant to this, economic strategies have changed dramatically
with the documentation of horticulture, architecture, ceramic technology, and new
tools. What is known of the development of the Antelope Creek phase of this
change in cultural history is discussed in some detail in this chapter.
(Neo-Indian) Village Phase cultures in the study area are the Antelope Creek
phase and other Upper Canark variant cultures. Lintz (1984:44; Lintz et al.
1993:26) prefers to use the Upper Canark variant terminology due to its more
precise delineation of Late Prehistoric (A.D. 800-1450) village cultures in the
panhandles of Texas and Oklahoma, southeastern Colorado, and northeastern New
Mexico. "It is bounded by the sand dune belt of central/eastem Cimarron County
in Oklahoma, the divide between the North Canadian and Cimarron Rivers, and
could include part of the Rolling Redbed Plains in far western Oklahoma"
(Hughes and Hughes-Jones 1987:26). The sites of the Upper Canark region were
densely inhabited from A.D. 1000-1500 (Lintz 1984; 1986), with occupation
continuing in some areas on into A.D. 1600 (Hughes 1991).
Three distinct and culturally different complexes the Antelope Creek, the
Buried Cities, and the Apishapa phase define the Upper Canark variant; the
New Mexico foci seems radically different (Lintz 1984:44-45; 1986; Hughes
1991:30; Hughjes-Hughes-Jones 1991:107-148). Antelope Creek phase sites
included two other regional variants: the Optima focus in the Panhandle of
Oklahoma (Watson 1950) and the Washita River focus in the far East Texas
Panhandle and Western Oklahoma (Hofman 1975; 1984). The Zimms complex of
west central Oklahoma is delineated as a slight variant from the Antelope Creek
phase (Flynn 1984; Lintz et al. 1993:26). Antelope Creek phase sites along the
Canadian are dated from A.D. 1200-1450 (Lintz et al. 1993:26-27). Wolf Creek
Buried Cities sites (410C1A-Q) are dated from A.D. 1150 and appear to be a
different variation of the Panhandle Village phase (Hughes and Hughes-Jones
1987; Hughes 1991; Lintz et al. 1993:27).
Though the -Antelope Creek phase extends from the Landergin Mesa on the
west to the Oklahoma border on the east, the most intense occupation was at the
Alibates quarries and the Lake Meredith area, near the oil fields between Fritch
and Borger, Texas (J. Hughes 1991). The main concentration of these
communities was along the Canyons, valleys, mesas, and rims of the Canadian
River near the Sanford Reservoir. The use of Alibates chert from the quarry near
the Sanford Reservoir, rather than material gathered from Canadian River cobbles,
decorated cordmarked ceramics, larger contiguous dwellings, and subtle
architectural construction methods separate these Antelope Creek cultures from
Buried Cities variation (Hughes and Hughes-Jones 1987; Lintz et al. 1993:26).
However, all of these Panhandle occupations are village adaptations with people
residing in masonry architecture (isolated or contiguous) with dual economies
based on horticulture and hunting (Lintz et al. 1993:27). This report will use
Antelope Creekj community patterns for comparisons with earlier groups in the
Lintz (1991:93-94) recognizes two distinct subphases within the Antelope
Creek phase. The Early Subphase (A.D. 1200-1350) is comprised of a hierarchal
architectural organization: (1) hamlet (multi-family villages and contiguous
residence units),' (2) homesteads (single family residences), and (3) subhomesteads
(field houses or farmsteads). Hamlets are placed along lateral tributaries and
adjacent to springs. There is a rare occurrence of exotic tradewares and little
evidence of hostilities.
The Late Subphase (A.D. 1350-1500) coincides with the spread of
environmental degradation. Many sites in the Southern Plains at A.D. 1300 show
an increase in bison processing tools (Creel 1991). Subhomestead and hamlets
disappear at Antelope Creek area along the Canadian River. Abandoned fields
and agricultural "huts" (Lintz 1991) at Antelope Creek terraces are evident. The
large community hamlets shifted to single family dwellings on the tributaries;
then, these are too abandoned. A marked increase in Southwest tradewares on
some sites in the late period relates to practices which Lintz (1984) postulates are
buffering strategies during a pro-longed drought. The presence of conflicts at the
Footprint site and Handley ruin suggests a state of unrest in the late phase (Eyerly
1907; Green 1986; Lintz 1991; Moorehead 1921; Paterson 1974; c. f. Lintz
1991). Lintz et al. (1993:45) also comments on the differentially occupied,
nucleated, and dispersed settlement patterns of the Panhandle region during the
Antelope Creek phase. The lack of settlements in certain areas may represent
buffer zones or spacing between cultural groups.
As the Canadian region declined environmentally (A.D. 1450), Speth (1991)
suggest that horticulturalists came to a critical juncture. Crop failure and the
depletion of winter resources may have resulted in a period of sever stress. The
buffalo herds, returning to the increasing short-grass prairie environments of the
drying Canadian River Valley, provided a viable economic alternative (Speth
1991). This shift can be seen in relative densities of gardening tools (e.g.
presence of bison scapula hoes and tibia digging-stick tips) which decline at
Antelope Creek sites after A.D. 1300 while bisort processing tools increased
During the Protohistoric period, Texas Panhandle villages were abandoned.
Antelope Creek villagers may have migrated to the tributaries of the Canadian
River and returned to a hunting and gathering lifestyle (Spielmann 1983). Or,
they may have joined Caddoan groups to the east (Hughes 1991). The expedition
of Spanish explorer, Francisco Coronado of 1541, does not record the existence of
any settled group in the abandoned villages along the Canadian (Gunnerson 1987).
Some investigators have suggested that the historic Tejas nomads were Antelope
Creek peoples who returned to a hunting based economy (Schroeder 1959: c.f.
Lintz 1984a). Unfortunately Tejas occupation, associated with the Antelope Creek
abandonment, has not been defined in the Texas Panhandle (Lintz 1984a:340).
Two spatially distinct protohistoric complexes have been defined in the Texas
Panhandle: the Apachean-derived Tierra Blanc complex of the upper Panhandle
and the Garza complex of the lower Texas Panhandle (Lintz et al. 1993:27). A
comparable Wheeler complex lies in western Oklahoma (Baugh 1986, Habicht-
Mauche 1988). These people constructed temporary tippis or conical huts and had
a meager tool assemblage (Lintz et al. 1993:27). Assemblage traits were Tierra
Blanca ceramics (Habicht-Mauche 1988:175-189), Apachean Perdido Plain wares
(Gunnerson 1987), jacal-like structures, Talco-type arrowpoints, and Anasazi
tradeware (Hughes 1992). Type sites for the Tierra Blanca complex include those
located in Deaf Smith County (Holden 1931; Hughes 1991; Spielmann 1983) and
the Fifth Green site on the Palo Duro Creek in Randall County (J. Hughes 1991).
The descriptions of the bison-hunting Querechos encountered on the
Panhandle-Plains by the Coronado expedition in 1541 and of the corn-
growing Faraones discovered by the Onate expedition in 1601 probably
constitute eyewitness accounts of the lifestyle of the Tierra Blanca
complex. (J. Hughes 1991:35)
Exotic materials of Southwestern and Caddoan tradeware pottery, obsidian,
and exotic shells indicate broad trade relationship were established in the
Protohistoric period. Nocona Plain ceramics from the Henrietta focus of North
Central Texas, clay-tempered sherds from the Frankstone focus of East Texas, and
Toyah Focus Leon Plain and Doss Red Ware indicate these interactions (J.
Hughes 1991:34-36). However, presence of dismembered human remains
documents interregional conflicts. As noted in later Spanish manuscripts, local
groups may have raided to control resources within or near the Texas Panhandle
(Moorehead 1921; c.f. Lintz 1991:103). And, the change in the cultural system
of the area coincides with the appearance of the Apachean people around A.D.
1500 (Lintz et al. 1993:37). Plains Indian farmers were forced to abandon
agricultural settlements to survive direct confrontation with these intruders.
In contrast to this view, a more mutualistic relationship with the southwestern
Pueblos may have existed between the Late Prehistoric and Protohistoric nomadic
bison hunters from the Panhandle (Spielmann 1982; Lintz 1991; Lintz et al.
1993:34). Bison hunters traveling to Pueblo villages traded products from hunting
and bison processing tools (made of Alibates chert) in exchange for certain
agricultural produce (com), polychrome ceramics, and other materials (obsidian,
shells, etc.) (Kidder 1932; Habicht-Mauche 1991:53; Spielmann 1983).
Spielmann argues this trade was established well before the historic period and its
necessity was spawned by basic nutritional requirements, e.g. carbohydrates for
the bison hunting communities and protein for the agricultural communities.
Glazed-painted ceramics, a basic utilitarian commodity, are found on Tierra
Blanca sites and are linked with a variety of functions including storage,
processing, and transport of food (Habicht-Mauche 1991:53).
Discussion of Late Woodland-Transitional Sites
The predominant material in the Duncan Ranch Site 1 attests to an occupation
based on a riverine adaptation including both hunting-gathering and possible
horticultural activities. The site provides a contextual framework for examining
cultural similarities to Lake Creek and the Palo Duro complex sites, other Late
Woodland sites, and early Antelope Creek sites. The Lake Creek complex
(Hughes 1962) has been suggested as the possible Woodland origin of the
Antelope Creek phase (Gunnerson 1987; Hughes, 1962; Lintz 1984; 1986;
Peterson 1991). New information on archaeological sites now includes cultural
changes occurring in the Texas and Oklahoma panhandles prior to the Antelope
Creek phase (Baugh 1984; Cruse 1992; Hofman 1978; Hofman 1984; Hughes
1978; Lintz 1984; Lintz et al. 1993; Peterson 1988). Peterson (1988:23) and
Quigg et al. (1993) report several sites in the Palo Duro Creek area on the
Northern Canadian River, specifically sites 41HF7, 41HF5 and 41HF88 contain
Woodland components. Surveys and excavations have identified the Lake Creek
assemblages at several other sites: 41M05 and 41PT29 (Green 1986), Swift Horse
site in Western Oklahoma (Briscoe 1989), the Middle Cheyenne, the Boys Ranch,
the Maintenance Bam sites (Couzzourt 1982), 41HF5/6 and 41HF7/101 (Peterson
1991:103-126).; Thermoluminescent dates from the Maintenance Bam site have
been recovered and these compare with the Swift Horse radiocarbon dates,
ranging between 1,820 and 1,533 years ago (Lintz et al. 1993:26). And, the
presence of maize pollen in lacustrine sediments (41HF88), dated 1740.+60 TBP
(TX5803) suggest maize may have been practiced early in this area (Peterson
The Palo Duro and the Lake Creek complexes began as early Woodland
manifestations in the North Texas Panhandle (A.D. 100-200) but their occupations
may extend to A.D. 1070 (Hughes 1991:24; Quigg et al. 1993). Because of the
lack of testing and quantified, comparative studies, Lintz (1984:353) suggests an
ambiguity still exists in our knowledge of Panhandle Woodland cultures, especially
between A.D. 800 and 1150. Minimal absolute dates exist from cultural contexts
from this and earlier periods in the region. This scarcity makes it difficult to
critically examine the nature of cultures during the transition period (Lintz et al.
1993:41). The poorly defined late Woodland culture permits the resurrection of
the migration theory (Lintz 1984:353).
Many sites of this period are located on terraces within the Canadian River
system and on its tributaries. Hughes (1991:25-26) places Lake Creek sites north
of the Canadian and the Palo Duro sites, south of the Canadian River. Present
information confirms overlapping of cultural elements of both complexes on many
sites and the broadening range of the Palo Duro complex (Lintz personal
The type site for the Lake Creek complex is the Lake Creek site on the Haley
Ranch in the eastern part of Hutchinson County (Hughes 1962; J. Hughes
1991:25-26). This site is located on the western bank of Lake Creek, a northern
tributary of the Canadian River. The site lies in a focal area for settlements in the
northern Panhandle (Peterson 1991:30). The Lake Creek site occupies a terrace
bench between the Canadian River and a tributary draw to the south, above a
marshy area of the Canadian River (Hughes 1962; 1991; Peterson 1991:31).
Archaeological investigations collected over one hundred and fifty-four pieces of
material culture (lithic tool, ceramics, and fire-cracked rock) from surface surveys
and a test trench excavation. Forty-five Alibates flint tools, one tool each of
Tecovas jasper Edwards chert, and six miscellaneous non-sourced lithic tools were
recovered (Hughes 1962). Lake Creek projectile points include small comer-
notched arrowpoints (Scallom), with an occasional large comer-notched (Ellis)
and small side-notched (Reed) points. An important addition to the artifact
assemblage were forty-three cord-marked sherds representing tall, cord-roughened
Woodland ceramic vessels with conical bottoms. Hughes (1991:24) has identified
similarities between these Lake Creek cordmarked types and contemporary Eastern
Woodland pottery. The intrusion of cordmarked ceramics into the Southern Plains
can be traced to the Burial Mound II or the "Middle Woodland" period in the
Kansas City area, circa 100 B.C. to A.D. 300 (Quigg et al. 1993:210). On the
basis of the cord marking style, Hughes (1962) argues that the Lake Creek ceramic
tradition displays continuity with the later similar "Borger Cordmarked" Antelope
Creek ceramics. Temper matrix in Woodland ceramics includes a crushed rock or
scoria, a black/gray to red basalt from pebbles in Pliocene deposits. Sherd
exteriors range, in color from predominantly gray to light-brown and black
(Hughes 1962:69). Differences between Antelope Creek and Lake Creek
Woodland ware is described mainly in depth of cordmarking, wall thickness, wall
hardness, and temper. Present in the Lake Creek assemblage are a few Mogollon
brown ware sherds, noted from the Pecos Valley (Peterson 1991:31). Tempered
with angular crushed feldspar, these sherds were used to cross-date the Lake
Creek site to periods between A.D. 900-1300 (Hughes 1962:72; Hughes et al.
1978; Hughes 1991: 25-26). Additional evidence of Southwest contact at the site
is a single obsidian flake.
Much of the tool inventory of the Lake Creek site appears similar to tools of
the Late Archaic stage. Snub-nosed end scrapers, thin-butted end scrapers, side
scrapers, flake scrapers, gouges, gravers and awls are present but their
proportions in the overall inventory has not been determined. Lake Creek sites
are typically rich in materials and represent extensive open camp sites; however,
many of the sites and any architecture developments have not been fully
documented. Rock hearths are scarce but refuse deposits consist of fire-cracked
cobbles and pebbles, debitage, mollusk shells, and animal bones. A single
fragment of a diamond-shaped, alternate-beveled knife typical of the later
Antelope Creek phase also was recovered at the Lake Creek site. Thick grinding
slabs resembling those of Antelope Creek phase are present (Hughes 1962:72).
The few isolated burials from Lake Creek sites are flexed interments associated
with Scallom projectile points, and accompanied by personal tools and ornaments
Bison bones at most Lake Creek sites are not abundant; thus, Hughes (1991)
contends that this supports Dillehays (1974) postulated Bison Absence Period II
(A.D. 500 to A.D. 1200/1300). Data from studies by Duffield (1970) and
Briscoe (1989) also documents a lack of bison bones. However, Peterson
(1991:31-32) questions the use of this assumption on Panhandle sites in that
Dillehay did not report on sites from the Texas Panhandle. Bison bones have
been reported on chronologically similar sites to the east and the west (Peterson
1991:31-32; Baugh 1984).
The type site for the Palo Duro complex is the Deadmans Shelter site in Tule
Canyon, a major branch of the Palo Duro Canyon (Hughes 1991:26-27).
Deadmans Shelter, a rock shelter, contains middens dated to two different
Woodland periods. The site has yielded radiocarbon dates of A.D. 465 _ 70 and
A.D. 710 65 (SI-1897 and SI-1898)(Baugh 1984:23-25). Jornada Mogollon
plainware ceramic were recovered in association with barbed, comer-notched
projectile points (Deadmans) (Hughes 1962; 1991). According to Hughes
(1991:27), the Palo Duro complex may represent nomadic people who traded and
carried information and goods between the Southwest and Lake Creek
communities. Another site, the Blue Clay site (41BI42, Area I) (Willey et al.
1978) supports Hughes observation on the Palo Duro complex. Baugh (1984)
includes the Blue Clay site as an incipient village site. This site contains abundant
samples of Alma Plain brown ware and obsidian but has no firm temporal dates
(Baugh 1984:23-25). The association of these sites with Southwestern pottery and
obsidian indicates contact in some form with the Southwest pueblos (Hughes
1991:35-36). Lintz (1984:376) suggests a Sacramento Mountain-Canadian Valley
influence of circa A.D. 1200 formed from the Palo Duro complexes interaction
between Lake Creek communities and unspecified groups in the Sierra Blanca
District. Incorporating some architectural design into Panhandle Woodland
adaptation is a likely result of these connections (Cruse 1992).
Other sites in the Texas Panhandle are often Woodland overlain with Antelope
Creek phase sites. Hughes has yet to assign sites in the Red Deer Creek area to
either the Lake Creek, Palo Duro, or another complex. Along the Red Deer
Creek Watershed, Hughes (1978) found five sites with possible early village and
Woodland components based on the presence of Scallom points and ceramics.
Hughes (1978:103-110) reports that sherds with scoria and sand tempers from a
Red Deer Creek Watershed (A 1275) were seriated. Scoria appeared in earlier
contexts than those with sand and bone or calcite. Scoria temper, usually
associated with the earlier Woodland period, appears in association with later
Harrell points and other items more typical of Antelope Creek assemblages.
Hughes argues that the scoria temper is a likely carry-over from earlier Woodland
periods (Hughes et al. 1978:110). Quigg et al.(1993:331) report a basalt and
quartz sand tempered ceramic sherd of possible Woodland derivation in a
Antelope Creek or Buried City site (41HC86) in the Palo Duro Creek vicinity.
However, previous descriptions are based on macroscopic identifications and are
difficult to compare to one detailed petrographic analysis.
At the Fatheree site (A1286) (Hughes et al. 1978) a later village component
overlies a possible Woodland component. Indicated by the presence of scoria and
sand tempered cord-marked sherds, the dark midden zone in Area 1 also contained
Deadman, Scallom, and Young arrowpoints with Mogollon pottery and a discoidal
hammerstone. Bison remains are notably scarce; most bone fragments are deer,
pronghorn, rabbit, prairie dog, mole, and turtle. Mussel shells are in great
quantity. Area 4 contained Fresno, Washita and Deadman points, and Area 5
contained Deadman points and ceramic cordmarked Woodland-type sherds. Site
A1286 is a stratified site with unique artifactual materials ranging from Archaic,
Woodland, to Antelope Creek type assemblages (Hughes et al. 1978:166-168).
Site 41PT29, on the north side of the Canadian River below Devils Canyon,
appears to be a stratified pre-Antelope Creek and Antelope Creek phase
occupation. The sites assemblage includes two Scallom comer-notched arrow
points, one unnotched Fresno point, a small sub-triangular knife, three oval
knives, forty knife fragments, one heavy scraper-gouge, several cores, choppers,
utilized flakes, and debitage (Green 1986:28-29). Frederick et al. (1993:465)
report Blocks A and B at the Sandy Ridge Site (41HF5) yielded Woodland
materials. Block A contained cordmarked ceramics with associated radiocarbon
dates of 1090+60 B.P. (Tx-7037) and 1280+70 B.P. (Tx-7036). Block B at
41HF5 revealed two humate dates of 880+60 B.P. (Tx-7034) and 560+50 B.P
(Tx-7035). The,earlier date is within the span of what Lintz recognizes as the
terminal Woodland period (A.D. 1070) (Frederick etal. 1993:465).
Early architecture is represented at only a few sites in the Panhandle. The
Courson Site A on Wolf Creek may have a pit structure overlaid with later village
phase structures (Hughes and Hughes-Jones 1987; Hughes personal
communication: 1994). Kent Creek, a Palo Duro complex site on the more
southern Prairie Dog Fork of the Red River in Hall County, is the first studied
architectural site in the Panhandle (Cruse 1992). The Kent Creek site is dated to
A.D. 710j+120 and this date places the site in the same time period as the Lake
Creek and other known Palo Duro complex sites. The Kent Creek site is situated
on a ridge toe near a spring-fed tributary of Kent Creek adjacent to a flood-plain
which accommodated "sub-irrigational" horticulture (Cruse 1992). Until the
recovery of the architecture at Kent Creek, Late Woodland groups in the
Panhandle were considered mobile hunter-gatherers (Cruse 1992:7).
The pithouses at the Kent Creek site bear no evidence of plastered floors,
however, the existing earthen floor appears to have been smoothed and filled. A
clay-plastered entryway is located on the east wall and slopes downward to the
present structure level. The semi-subterranean pithouse incorporates rounded
comers and a shallow depth of 35 cm. Located within the pithouse are four
features and a single central post surrounded by smaller peripheral posts
reminiscent of Mogollon structural design (Cruse 1992:128). The four features
include a grouping of fire-altered rock, a stepped feature at the entryway, a burial,
and a hearth (Cruse 1992:53-54). A storage pit, hearths, sherds,and a cache of
manos were also recovered. As evidence of networking and interaction with the
Jornada Mogollon, Cruze uses the presence of obsidian tools and flakes, Mogollon
brownware sherds, subfloor burial, and a related pithouse architectural pattern.
Sam Wahl site, recently investigated by Prewitt and Associates, Inc. (Lintz,
personal communication 1994), contains a Palo Duro complex village with several
pit structures. The Blackdog Village (Keller 1975), and the Borger Bridge site
(Harrison 1983) have possible early architectural (pit structure) indications though
Lintz questions; the date of Blackdog village feature as it came from contexts
which also contain dates of later Antelope Creek occupations (Lintz 1984; 1994
personal communication). The Courson Site A has a possible early architectural
component (A.D. 1200) (Hughes and Hughes-Jones 1987:117). The Maljamar
phase located in southwest Texas Panhandle, A.D. 1000-1300, contains pit
structures (Corley, 1965; Corley and Leslie, 1960; Hughes 1991). With the
presence of a caliche plastered floor, the Duncan Ranch Site 1 also testifies that
occupations in the Woodland period in the North Texas Panhandle included more
Preliminary characteristics of these Late Woodland adaptations in the North
Texas Panhandle consists of the following diagnostic traits:
1. Predominantly, Woodland-like arrowpoints, such as the Scallom
and/or Deadman projectile points; a mix of some unnotched (Fresno)
points (Kent Creek) and side-notched forms (Hughes 1962; Hofman
1978; 1984; Quigg et al. 1993:175).
2. Examples of dart points such as the side-notched and comer-notched
dartpoints, i.e. the comer-notched Ellis.
3. Thick-walled, rough cordmarked pottery with scoria-crushed, sand,
bone, shell, or calcite temper, usually with thinner-walled, higher
fired cordmarked ceramics.
4. Mogollon brownware or other ceramic ware, obsidian, and shell
ornaments indicative of networking and contact including trade and
exchange with the Southwest between A.D. 900-1200.
5. Early horticultural indications including one-hand manos and limited
two-hand manos, slab or small basin-type metates, few thick basin or
trough metates, local seed gathering and processing, com pollen,
hoes, and fire-altered rock.
6. A riverine settlement pattern focused on tributaries and terraces near
7. Large scatters of utilized lithic flakes with differential useware
8. Fresh water mollusk shell fragments.
9. An absence or minimal presence of bison remains and bison bone
gardening tools, in particularly when compared to the increase of
these remains in later Antelope Creek and village assemblages.
10. Small quantities of later bison processing tools in contrast to an
increase of thick, steep angled end-scrapers and diamond beveled
knives as yielded on Antelope Creek sites (after A.D. 1300).
11. Limited reports of pit structures with pertinent features which may
represent later elaborated structures on Antelope Creek sites.
Lintz et al. (1993) argue that transitional sites would best be defined from
single occupation sites with distinct stratigraphic layers. Because of a temporal
gap of approximately 800 years, recent arguments from a Lake Creek complex
site (Swift Horse site) (Briscoe 1989:107), based on five radiocarbon dates
(Thurmond 1991:250), lays controversy to this complexs relationship to Antelope
Creek (Lintz et al. 1993:41). They are concerned that tight chronological controls
from a series of late Woodland sites need to be a prerequisite for understanding
these cultural developments" (Lintz et al. 1993:41).
Discussion of Antelope Creek Sites
To compare the postulated transitional phase with the later Antelope Creek
phase, it is important to define the Antelope Creek manifestation in more depth.
The Antelope Creek type-sites will be used in this report include the Alibates 28,
Antelope Creek 22 (Baker and Baker 1939, 1941), Antelope Peak, Footprint
(Green 1986), Turkey Creek, and South Ridge (Etchieson 1979). Those sites
range in size from single-room round structures (5 m square) to contiguous and
multi-room dwellings of 35 rooms or more (Lintz 1974: 327). The South Ridge
site includes a Palo Duro component, appearing in firm stratigraphic contexts
separate from the Antelope Creek occupation (Etchieson 1979).
The Antelope Creek phase is characterized by an economy that combines
foraging and horticulture (maize, squash, and beans). Its settlement patterns
revolve about a nucleus of farmsteads, hamlets, and villages located along rivers
and creeks, canyons and valleys, on mesas, buttes and ridges, and benches and
terraces (Lintz 1984:46). Non-structural Antelope Creek sites (e.g.camps,
quarries) are almost unknown (Lintz 1984); however, (Quigg et al. 1993: 330)
reports that "non-village sites exist". The Antelope Creek phase has been dated
with at least 58 radiocarbon samples and five paleo-magnetic dates, and 17
ceramic crossdates using Southwestern trade wares (Lintz 1984; 54). MASCA
tree ring procedure (Ralph, Michaels and Han 1974) has provided phase dates
extending between A.D. 1200 through 1500. These dates correlated with
magnetometer and ceramic cross-dates although Lintz (1978; 1984) suspects that
the magnetometer dates are less reliable.
The architectural style of Antelope Creek sites include foundations consisting
of vertical rows of stone slabs, single dwellings to multiple groupings, and
contiguous dwellings with thatched roofs (Lintz 1984:44). The Alibates 28 (Unit
II) village site contains a larger community of multiple free-standing to one-room
family dwellings. The Roy Smith, Footprint, Tarbox, Saddleback, Alibates 28
(Unit I), and Antelope Creek 22 sites incorporate contiguous multi-room
dwellings, including as many as twenty-five to thirty-five rooms (Lintz 1978:328).
Room patterns are variable but the distinct feature that separate these sites from
traditional High Plains Village patterns are the frequent use of upright dolomite
slabs for walls and foundations (Lintz 1978:329). Slab walls may be double or
single rows with alternating layers of adobe, daub, and horizontal masonry and
frequently are riot dressed. The arrangement may be haphazard. Some
constructions combine vertical post reinforced walls with slabs, adobe puddle wall,
or adobe "blocks" (Lintz 1978:330). With contours dictated by topography,
village layout typically appears random and unplanned. The contiguous room
layouts at the sites of Saddleback and Arrowhead Peak Ruins span the entire mesa
top (Lintz 1984a:330). Floors may be covered with caliche mud (Peterson 1991).
The Antelope Creek artifact assemblage, with its distinctive lithic knapping
style and specialized tool morphology, remains different from other regional
assemblages of the same time period. Antelope Creek phase lithics display wide
variation, including several triangular arrow point types, such as unnotched Fresno
and Young poirits and side-notched points of Washita, Shetley, and Harrell
varieties. Infrequent large corner-notched dart points appear at many sites.
Specialized tools occur as well, including "guitar pick" scraper/preforms, pin
drills, key or T-type drills, end-scrapers, and ovoid knives. Plentiful types
include the thick "keeled" end-scrapers, comer-tanged knives, and diamond-
shaped beveled Harahey knives (Lintz 1984:62-63; 1984a:333; Hughes 1991).
Many of the tools are manufactured from material (specifically, Alibates chert)
traced to the Alibates Quarry located near the Antelope Creek sites on the
Canadian River (Ed Day [National Park Service], personal communication 1992;
Lintz et al. 1993). An abundant assortment of pecked and groundstOne artifacts
are evident including manos, milling-stone fragments, faceted sandstone "biscuit-
shaped" hammerstones, awl sharpeners, abraders, and shaft smoothers (sometimes
occurring together in tool kits) (Lintz 1991).
Other notable tools at Antelope Creek villages are "cloud blowers" and elbow
pipes. Ceramics generally consist of cord-marked globular jars with wide mouths,
disk-shaped spindle whorls, and clay beads. The variation of bone tools range
from bison scapula hoes and knives to bison tibia digging sticks and awls of
various shapes and functions. Rib-edge, split-rib, splinter and split-deer
metatarsal forms, pins, pegs, beads, bone wedges, rib rasps, antler-tine billets,
pressure flaking tools, rare-eyed needles, shaft straighteners, and bison femur-head
hide grainers are more specialized bone-tool implements (Lintz 1991). Freshwater
mollusks were used as scrapers and spoons, or more rarely, cut into pendants
(Lintz 1984; Hughes 1991). Coiled baskets have been reported at several
Antelope Creek sites. Woven mats have been identified only at site Alibates 28
(Studer 1934:90; c.f.Lintz 1984). Items of wood are poorly preserved and, as
such, appear to be under-represented (Lintz 1984).
Storage cysts and pit or cached burials on most sites have been recovered
from house floors and on occasion in large multiple graves (Hughes 1991:32).
The graves of Antelope Creek individuals can be rock filled and often included a
few personal possessions. The burial position of the body is generally flexed.
Skulls are rounded with high foreheads. Bison tibia hoes or dibble sticks are
found with the bodies. Hughes (1991:32) suggests that mortuary implements may
not have been used prior to burial.
Trade at Antelope Creek sites is evidenced by the presence of olivella and
columella shell beads, conus tinklers, conch shell pendants and gorgets from the
Gulf of Mexico or Sea of Cortez, obsidian from New Mexico, mica, turquoise,
and painted and polychrome pottery from the southwestern Pueblos. Catlinite
pipes from Minnesota and small siltstone elbow pipes from the Landergin Mesa
also are present (Hughes, personal communication 1991). Niobrara jasper,
Edwards Plateau chert, and occasional cordmarked sherds with collared rims and
lip tabs denote trade with other Plains groups (Lintz 1984: 64). The most
important items of exchange for Antelope Creek people were the Alibates chert
resource and bison products. The Alibates material can be obtained from the
quarry sites on the Canadian River and in cobbles carried downstream in the
Canadian. The increase in large bifacially shaped quarry blanks and possibly
guitar pick preforms on many sites (Pecos) and in other regions during the Late
Prehistoric period suggests that quarrying and manufacture of such items in excess
of local need was undertaken for purposes of exchange (Hughes, personal
communication 1994; Lintz 1991). It has been suggested that Antelope Creek
people had control of the quarry and established workshops there (Hughes 1991).
On a interregional scale, Spielmann (1983; 1991) has suggested that
mutualistic trade; and exchange between the Southwest and the Plains peoples
occurred during the Antelope Creek phase. This interaction is indicated by the
presence of Alibates "flint" tools and bifaces and bison products in notable
Southwest sites, i.e. Pecos (Kidder 1932) appears as southwestern polychrome
ceramics, obsidian, pipes made of materials from the Langerdin Mesa and
Minnesota, olivella and other shells, and other non-local items are found on
Antelope Creek sites (Holden 1932; 1933; Lintz 1991:92-93). These items
suggest that possible buffering strategies were used during periods of stress.
These artifacts of Southwest origin increase incrementally in the late period (Lintz
1991:93). Spielmann (1983) postulates a mutually beneficial system between
cultures. Antelope Creek people traded bison products (protein, especially) for
needed carbohydrates (maize) during the long lasting drought. Perishable goods,
such as bison meat, fat, and hides, are recorded from the Plains during historic
periods (Creel 1991).
The nearest site to the Duncan Ranch is the Jack Allen site (A654) (Harrison
1983). This pit structure (House I) incorporates a small, square to rectangular
floor with a central trough and a raised platform located along the west wall and
contains a long entryway. The platform extends outside the house to give access
to an exterior hearth. In addition to a central hearth, an ovate depression was
located at the east entrance to the structure. Posts demarcate the wall outline.
Wall construction was of wattle and daub, however no evidence of stone slab
foundation. The lack of stone foundations may result from the scarcity of
dolomite outcrops in the vicinity. In addition, the artifactual inventory includes a
sand and bone-tempered cordmarked ceramic (Flynn 1984:284). The pit house on
the Jack Allen site resembles, in design, the Zimms site pit structure in western
Oklahoma (Lintz 1984:329; Flynn 1984:281). Presently, there are no dates for
The Antelope Creek phase represents a local adaptation to the short-grass
prairie (Lintz 1991:90). Lintz (1984; 1991) suggests the diagnostic aspects of this
adaptation include the following characteristics:
1. Subsistence equally on reliant on hunting (primarily bison), foraging,
2. Semi-permanent villages.
3. Villages located adjacent to floodplains.
4. Semi-subterranean rectangular lodge structures with east facing
entrances and four-post roof support systems surrounding a central
5. Cache pits within and near houses.
6. Cordmarked pottery tempered with sand and finely ground quartz.
Shell and bone are also used as temper on some sites. Ceramics are
often cordmarked with closely parallel and cross-hatched marking, and
7. Scrapers and small triangular unnotched and side-notched projectile
8. Bison scapula hoes and tibia digging-sticks.
9. Notched bison rib "rasps" (Hughes 1991; Lintz 1991: 90; see also
Lehmer 1954: 139-140).
10. Tradewares from the Southwest, Upper Republican, Caddoan, and
Landergin Mesa areas.
Clearly, thb Panhandle of Texas has a long sequence of prehistoric
occupation. The continuum begins with the Paleo-Indian era and ends with the
Protohistoric Tierra Blanca phase cultures. By A.D. 1200, the region around the
Canadian River region was densely occupied and a dual economy village phase,
based on agriculture and hunting, ensued. This phase, the Antelope Creek phase,
established its settlements on the Canadian River, its valleys, tributaries, and
mesas. There is ample evidence of cultural interaction with the Plains, the
Southwest, and other cultural areas. By A.D. 1450, climatic change precipitated a
return to bison hunting as the result of ensuing crop failure, and eventually these
villages were abandoned.
The origins of the Antelope Creek phase remains unclear and a matter of
some controversy. Some scholars argue it is the result of a new groups migrating
into the area while others argue it constitutes a local adaptation to changing
environmental and cultural conditions. Support for a local adaptation is suggested
in apparent archaeological continuity of Late Woodland to village adaptations
developments in the Panhandle of Texas. Analogs to this cultural development are
provided in local adaptations, such as the Lake Creek (Hughes 1962; 1991; Lintz
1984) and the Palo Duro complexes (Hughes 1962; 1991; Couzzourt 1982; 1987)
and including the Kent Creek site (Cruse 1992).
The Duncan Ranch Site 1 (date of 1000+60 B.P., calibrated AD 960-1180, 2
sigma) is dated to the possible transition period between the late Woodland and
Antelope Creek phases and will be used to test this theory. Expectations are that
this site will display those necessary characteristics and support indigenous
development of the Antelope Creek phase, but critical evaluations require looking
at data both positive and negative. In regards to this expectation, because the
indigenous hypothesis suggests these developments are the result of a local groups
adaptation to a changing environment, a better understanding of local ecology is
needed before the remains of the Duncan Ranch site 1 can be presented.
Therefore, the next chapter outlines the ecology of the area.
The Human Ecological System
In order to study the cultural system on the Duncan Ranch Site 1, past and
present environmental systems of the research area need to be examined. The
interaction and response between the environment and human cultural systems
rests on the basic assumption that cultural groups convene and organize their
lifeways to promote the reproductive success of each group member. And, groups
must provide a substantial subsistence from their habitat or cease to thrive.
Changes in the ecological system can affect the decision making processes of
particular groups which, in turn, affects the survival of individual members.
Actions, perceived and acted upon by individuals, have to received acceptance by
the community before adaptive cultural change results. Change, for example, in
subsistence requires community response (Butzer 1984). These decisions generate
positive or negative feedback that either supports the status quo or initiates change
in behavior. Some cultural systems are more sensitive to "noise" in the
environment, explaining some of the unique adaptations found in similar
environments (Butzer 1984:294). Thus, reconstructions of past environmental
systems, including local variations, are important to the understanding of cultural
change within the archaeological record.
If, as Lee (1969) suggests, hunting and gathering is a successful lifestyle, then
important processes between human culture, decision making, and the
environmental systems, in the present and the past, must condition any deviation
(Plog 1974:8). It is also assumed here that within every culture, pressures from
the natural environment are countered by "deviation-countering" cultural
responses. And, these responses are sellected to re-establish order and to return
the system to its original stable state and prevent collapse of the socio-cultural
system (Plog 1974:47). Responses stimulated by selective pressures are revealed
in "adjustments, oscillations, and modifications" and include "acceptance,
integration, or diffusion from external elements and/or a recombination of existing
cultural components" (Butzer 1984:284). However, under some conditions, the
environmental stimulus may exceed some important threshold and the homeostatic
responses will fail to establish equilibrium. In these situations, the cultural system
is placed under stress. When exceeding stress occurs or even before stress signals
critical threat to survival, the existing system will be either altered, replaced, or
eliminated (Minnis 1985:20; Plog 1974:48). Change will be expressed in a
product of shifts in social structure including behavior, subsistence, settlement
rearrangements, or demise of the population. And, changes will be revealed
archaeologically in patterned spatial distributions at a single locus or at multiple
loci or within the patterned variation which shifts through time (Plog 1974).
Cultures, while searching for homeostasis, must provide the means for
variation. Information sharing groups, ceremonial or occasional activities (e.g.,
eating of sacred foods), fortuitous accidents, experiments, and/or situational
horticulture, may be the future seeds of change. According to Plog (1974:53),
evidence shows "that under stress conditions, experimentation increases." The
probability of survival under environmental change is directly proportional to the
cultures density and diversity (Plog 1974:53). The more diverse technology and
behavior a group has, the greater is their chance of survival (Plog 1974:52; Salins
and Service 1960:97). A variety of innovations or expressions can be produced
by decisions and behaviors of human groups in reaction to stresses (Butzer
1984:281). Some responses may be accidental or fortuitous, others may be more
purposeful. Innovation seems to occur when oscillating or "searching" trajectories
lead to fortuitous possibilities, both latent and emergent, and these establish a net
gain which provides a signal to other trajectories or causes the destruction of past
information (Butzer 1982:281). Cultural change is established when variations are
accepted and increase the likelihood of survival (Butzer 1984: 281; Plog 1974:51).
And, as humans are innately imitators, those practices which establish a net gain
often will be emulated by others (Plog 1974:51).
Behavioral; responses in times of environmental stress may be due to
perception of environmental or cultural condition, real or perceived as real.
Trigger (1989:333) suggests that societies are not closed systems but react in
relation to the natural and social environments and that the development or change
of a culture may be in response to the broader social network of which it is a part.
These responses include reaction to climatic change, recognition or perception of
danger, mobility limitations, a newly accepted or recognized resource, ritual
shifts, and acceptance or avoidance of a consumable (e.g. the eating of a once
sacred food or the introduction of maize to the diet) (Minnis 1985).
As the spacio-temporal unpredictability of a resource increases (e.g., the
bison), large information-sharing groups increase the chance of survival.
However, these networks have a limit and this limit is determined by the resource
abundance, mobility, and benefit of the network to individuals in any one group
(Smith 1981:43-44). Binford (1983:205) proposes that the transition from a
mobile hunterfgatherer to sedentary subsistence would involve a set of conditions
in which information concerning resource availability is restricted or no longer
helpful and movements from one place to another are unrealistic. As population
increases and the main resource area decreases, strategies to reduce subsistence
risk are necessary, including the adoption of an intensive resource procurement
system. Intensified exploitation of local resources [such as riverine mussels,
grains, local fauna], situational horticulture, and/or agriculture would be necessary
strategies for survival (Huber 1984:152).
Other researchers note that additional factors may have conditioned a shift in
subsistence. Gilman (1987:530-564) argues that transitions from hunter-gatherer
to village sederitism was a combination of a growing population, an increasing
network of information, intensification of subsistence, and the amount of time and
space needed to store, process, and cook food. Areas of known "low-risk"
environments containing a broad spectrum of resources and cross cutting several
ecological zones were settled more frequently. Their success in supporting
sedentism was conditional on many factors, e.g. costs of moving from patch or
patch, regional' population density, threats, territoriality, and access to other vital
resources, such as bison, water, and fuel. But, change of this sort will not occur
unless the culture has accepted new perceptions, ideas that are perhaps contrary to
a more mobile!existence. According to J. Moore (1983:173-191) information,
socially organized, is the primary unit which transmits these gains or expressions.
"The very nature of information exchange creates pressures to organize and
structure the flow of information exchanged" (Moore 1983:187).
Hayden (1992) suggests that communal feasting activities between neighboring
or visiting groups may have been the social strategy which stimulated the
accumulation of surplus agriculture and labor investment, and allowed acceptance
of change. From net-working alliances and their associated trade and exchange
relationships, familial and fictive tie building, and group ceremonial participation,
access to resources and information was facilitated. Minnis "Model of
Opportunity" (1985:310) suggests the need to change in pre-agricultural
communities is encouraged by the necessity to have a predictable resource, to
offset environmental disturbances, to share during feasting, and to retain important
These theoretical aspects can be applied to more specific areal developments
in the Panhandle of Texas between A.D. 200-1200. During this mesic period,
riverine environments were exploited by Woodland semi-nomadic people. Though
exact demographic data are unavailable, the diffuse spread of small basal and
comer notched projectile points in association with a thick cordmarked ceramic,
sometimes exotic Mogollon brownware, pit structures, and increase in
groundstone and groundstone size attests to these occupations. And, conditions
appear to have been favorable for seasonal human settlements of some duration.
Change among Woodland populations was conditioned by the unique natural
and social environmental system of the Canadian River and its tributaries. One
important adaptive advantage may have been the predictable springs, and lakes,
such as on White Deer Creek, formed from seepage of the Ogallalah Aquifer
(Lintz 1991; Lintz et al. 1993: 37). These springs supplied and encouraged the
growth and variation of wild flora and fauna, and provided subirrigated terraces
on which horticulture could begin. However, though settlements began during a
moist climatic period, social conditions facilitated change. The knowledge of past
periodic droughts, the change in migratory patterns of the bison, knowledge and
acceptance of new foods to the diet, need for cooperation in defense, warfare, and
construction, and stress concerning food and resource availability are predicted to
have encouraged the maintaining of extra-regional ties. Interactions with other
cultural groups, in turn, supplied information, kinship alliances, and stimulated
innovation and/or acceptance of new ideas and rituals. During the Late
Woodland, the Canadian River and its tributaries had the necessary natural and
social environmental conditions for investment in later more elaborate and
permanent architectural and agricultural communities of the Antelope Creek phase.
Given the above discussion, knowledge of the past and present biophysical
environmental characteristics of the research area is integral to interpretations of
prehistoric behavioral strategies. Though the regional physical and biotic
characteristics of the Texas Panhandle may display more general patterns, local
environment systems may display their own patterns. These local systems have
contributed to the formation of larger regional patterns and prehistoric settlements
(Lintz 1984:66-67). It is assumed that idiosyncratic responses are produced by
local social and environmental conditions. Bayhan and Morris (1990:26) argue
that there "remains appreciable evidence to indicate that prehistoric people did
adapt to climatic conditions which were not static but changing throughout the
Holocene." Though we cannot assume present environmental conditions represent
the past, there are certain static features are relevant to interpretations (Lintz
1984). The cumulative effect of modem farming, ranching, and irrigation has
impacted, erodied, polluted, and altered the terrain. And, the accelerated loss of
animal and plant species conditions our perception of past environments. Past
environment would have been more vegetated and contained increased plant and
animals diversity. Also, alterations, including erosion and effects from human
impact, occurred in the past. Documentation of extinct fauna, including prairie
vole, curlew, soft shell turtle, ancient equus, giant ground sloth, mammoth, and
bison species attest to these different conditions. Extinct forms of plant and
animal life can be markers of certain periods (e.g. the prairie vole and soft shell
turtle as markers of pre-Antelope Creek periods (Speer 1984). Human settlements
and their agricultural practices have altered landscapes, conditioning the growth of
later plants and animals (Wilson and Briske 1979).
This paper will focus on five biophysical environmental variables which are
assumed important to ecological conditions and human response. These factors
include: (1) physiography of the region, (2) location, soils, and terrain of the site,
(3) climate, (4) the presence, use, appearance, or disappearance of a particular
plant or animal variety, and (5) raw material accessibility and use.
The study region lies in the north central part of the Panhandle of Texas. The
region is a part of the Kansan Biotic Province (Blair 1950) in the short and mixed
grass plains district of the Southern High Plains (Allred 1956, c.f. Lintz 1986:60;
Hughes 1991:4). The Canadian River flows through the Panhandle, beginning at
Lake Conchos in New Mexico, and meets with the Arkansas River in Oklahoma.
The confluence of these waters eventually flows on to the Mississippi.
Tributaries, both north and south, connect with the Canadian River at three to five
kilometer intervals and range from 20 to 35 kilometers in length (Lintz 1984).
Although the Canadian River often appears dry, it will flow intermittently through
underground channels of porous sand, forming notorious quicksand spots [Figure
Near the north eastern edge of the Llano Estacado, the Canadian River
entrenches, exposing an ancient Pleistocene valley called the Canadian Breaks
(Hughes 1991:6). These canyons display the characteristic Permian red sandstone
cliffs layered between formations of dolomite and gypsum. Runoff from the
"Breaks" forms the characteristic rich, sienna color of the Canadian River
(Hughes 1991:5-6). The formation of the Canadian Breaks began in the Cenozoic
Period. Geologic formations of the Canadian system are the Permian
Whitehorse, Alibates, and Quartermaster formation, the Triassic Trujillo and
Tecovas formations, the Tertiary Ogalallah formation, and the present Quaternary
wind blown deposits. Canyon rims, cliffs, the upland hills and prairies, slopes,
gallery forest, benches, terraces, floodplains, and stream channels expose these
formation layers. Complex ecozones were formed in the valley and a rich variety
of flora and fauna inhabits all biotic zones (Hughes 1991:6-7; Lintz 1986:39-84).
The first fprmation, the Permian "red beds", are concentrated in the eastern
two thirds of the Canadian valley. The Whitehorse layer "bed" is the earliest and
is composed of red mudstone and shale (Lintz 1984:78-79). The Alibates
dolomite layer Hies above the Whitehorse and consists of upper and lower dolomite
sediments intersected with the characteristic red mudstone and shale. Between the
faulted mudstone, replacement minerals from an ancient salt water sea created
formations from calcitization and chertification. From this process important
prehistoric lithic materials were formed, e.g. Alibates chert (Hughes 1991:7; Lintz
1984:78-79; Bertram 1989; Shaeffer 1958). Downstream from the quarry,
cobbles of Alibates chert have been carried by the Canadian River well into
Oklahoma. These corticated cobbles are sometimes "basketball sized" (Etchieson,
personal communication 1994). The third Permian formation, the Quartermaster,
lies above the Alibates dolomite deposits and is similar to the earlier White Horse
deposits, though it contains more clay (Bowers 1975:23; Lintz 1984:79).
Triassic age sediments, Tecovas and Trujillo, are exposed in the southeastern
part of the Llano Estacado and within the western third of the Canadian valley
(Lintz 1984:79). The Tecovas formation represents 85 percent of the Triassic
exposure. Its mottled red, orange, yellow, green and/or blue jasper (Tecovas
Jasper) has been used by prehistoric groups. The Trujillo formation consists of
gray-brown conglomerates, sandstone, shale, pebbles of quartz, limestone,
siltstone, thin cherts, arid petrified wood (Lintz 1984; 1986:48; Hughes 1991).
Above the Triassic sediments, the Pliocene Ogallalah formation exhibits light
brown to buff sandstone (Lintz 1984:80). Gravels, including Potter chert, a gray
sedentary metamorphic silicate, lie in this formation. These gravels also provided
lithic materials for the prehistoric inhabitants of the region (Bell 1984). A few
lithic remains in the Duncan Ranch collection are made of Potter chert, and this
material has been noted in other studies (Lintz et al. 1993:70). The Ogallalah
formation supports an important aquifer which supplies ample water for streams
and springs in the region (Lintz 1984:68-69). Prehistoric agriculture was given an
advantage because of the Ogallalah aquifer; its natural seepage provided
"subirrigation" for terraces and fields (Cronin 1969, c.f. Lintz 1984; 1986; 1991;
Lintz et al. 1993:37).
The most recent geological formation is the Quaternary wind blown sand and
silt which was deposited during the Pleistocene and Holocene epochs (Lintz
1984:80). The Duncan Ranch Site 1 lies within these Quaternary aeolian sand and
silt deposits. The local terrain has been altered somewhat by the flooding and
down-cutting of White Deer Creek. Ogallalah gravel beds are exposed in the road
cut on the site from this erosion (Martin, personal communication 1992). Since
the early 1950s the Ogallalah Aquifer has dropped in water level from irrigation,
creating downcutting of streams and drying springs (Peterson 1988:7). In the
past, a 100 year drought may have only affected the aquifer sightly (Lintz
1984:65) but an extended drought, such as experienced in the late Antelope Creek
period, would have appreciably lowered the water table (Lintz et al. 1993:27).
The Canadian River and its tributaries show signs of being intensely occupied
by sedentary and nomadic groups of humans throughout history (Bertram 1989).
The river system, with its lure of bison herds and quality lithic materials, provided
the necessary resources for ancient and modem communities, traders, and
explorers traveling from other regions. Remains of prehistoric peoples are present
on playas, springs, basins, mesas, benches, and near drainages in the valley. In
1965 the Sanford Dam project constructed a reservoir, Lake Meredith, near the
main aggregation of Antelope Creek sites, covering and critically damaging many
sites (Green 1986), and many sites have been vandalized.
Figure 4.1: Geological Formations in the Canadian River System
The study of prehistoric groups cannot be separated from the physiography
and of the region and its adjacent areas. Though prehistoric people had no means
of transportation except by walking or water transportation, they did not live in
isolation. Alliances and networking systems were intergal to their survival. The
broad, flat Canadian River with its sinewy streams is easily forded. Cutting
through the expansive Southern High Plains, it exposed mesas, canyons, and
terraces which contain valuable resources and provide a negotiable corridor
between the east to the west. Thus, in the past, important advantages derived
from its physiographic characteristics encouraged mobility, resource access,
human interaction, and settlements for prehistoric inhabitants, wanderers, traders,
Project Location and Description
The Duncan Ranch is located in Hutchinson County, approximately 33.0
kilometers from Skellytown and Borger, Texas and 12 kilometers south of the
Canadian River. White Deer Creek, a southern tributary of the Canadian, divides
the ranch and lies between Spring Creek and West Fork Creek [Figure 4.2].
Based on geologic and environmental information, there is substantial support that
White Deer Creek was running throughout prehistoric periods (Hughes, personal
communication 1992). Located above White Deer Creek, the Duncan Ranch Site
(41HC124) is located on a hill east of White Deer Creek [Figure 4.3]. It is
approximately 11.0 kilometers from the Jack Allen site, and approximately 66.0
kilometers from the main concentration of Antelope Creek sites on the Canadian
The Duncan Ranch exists as a partnership between Estelle Duncan, Patsy
Ward, Donna Weathers, and Alice Gustafson. Originally, it was purchased in
1940 by Drs. Frank B. and Robert A. Duncan to provide an escape from the
rigors of the medical profession. The ranch has been a cow and calf operation
since this date. To the northeast of the site, an old fort, Adobe Walls, stands as
the location of last U.S. Calvary battle with the Comanches (Peterson 1991).
The Duncan Ranch Site 1 is multi-component site, used perhaps seasonally
and strategically over many years of prehistoric occupation [Figure 4.2]. The site
covers approximately 17 acres. Approximately 24,000 square meters of the site
were surveyed for this project.
The location of the site has significant strategic, logistic, and aesthetic
attributes. The north to northwest facing aspect gently slopes down to a slight
ridge on the west, overlooking a riparian terrace formed by White Deer Creek.
On this terrace stands a large mound covered with hackberry trees of
approximately the same age; one ancient cottonwood tree grows at the north end.
The mound may only be the result of flooding, rather than a remnant of
prehistoric activity (Hughes, personal communication 1992). To the north of the
site, a man-made lake was formed from the damming of an adjacent marsh in the
1950s. However, part of the marsh still exists and contains cattails and willows.
The marsh stretches into the next pasture parallel to White Deer Creek, forming
small springs, ponds, and fertile "subirrigated" fields. The man-made lake is
presently inhabited by many varieties of water fowl, mammals, and reptiles.
Cottonwoods which border the creek and lake are being decimated by an over-
population of beavers. Still, immense thickets of cottonwoods, willows,
grapevines, wild flowers and native grasses exist, even during droughts. Thus,
beaver activity may have been, in the past as well as the present, beneficial for
maintaining ground water levels and creating deep pools of water in the creek
(Martin 1989). The site is close enough to the Canadian River valley to have
benefited from the broader Canadian ecological zones. Positioning the site
between various ecological zones of White Deer Creek area indicates this
settlement on the creek would have provided diverse kinds of environments
necessary to sustain community growth.
Soils and Local Deposition
Local soils are determined by the chemistry, topography, and geography
which influence its particular deposition. Soil characteristics are dependent upon
the parent material, climate, organisms, moisture, accumulation of decay, salts,
oxidation-reduction, shrinking-swelling of clays, eluviation, illuviation, and time
(Birkeland 1974; c.f. Ferring 1992:16-17). These conditions affect the local
vegetation, location of communities, horticulture, water, and preservation of
artifacts. Analysis of pertinent soil formations on sites contribute to our
understanding of paleoenvironmental reconstructions. Occurrences, such as the
translocation of salts, can help preserve shell and bone in calcic horizons, while,
in leached horizons, the translocation of carbonates leads to the deterioration of
bone (Ferring 1992:17). The presence of charcoal in stratigraphic levels may
indicate human occupation, destruction of the site by fire, or wild fire.
Calcareous soils are often associated with more arid conditions, though a period of
moister conditions has been identified in neighboring areas between 2000 and
1000 B.P. and postulated for the Woodland period (Ferring 1986; 1992:10). Soil
changes from prehistoric agriculture can be responsible for limited grass cover on
such terraces, especially blue grama. Wilson and Briske (1979)(c.f. Sandor
1992:225) from their recent research on the Great Plains report that blue grama
has a difficult time reestablishing itself on cultivated fields. Apparently, blue
grama (Bouteloua gracilisl needs moist conditions to establish communities.
However, once established, the grasses will be hardy and survive during drought
conditions (Sandor 1992:225).
The Soil Survey Staff 1975 taxonomy has differentiated five regional orders in
soil in the Canadian System. Alfisons, Inceptisols, Mollisols, Entisols and
Aridosols are found in warm subhumid to semiarid climatic regions. The first
three occur in the outer valley; Entisols generally occur within the inner valley.
Aridosols are found occasionally in the outer valleys, forming on late Pleistocene
or older surfaces (Lintz 1984:82).
On the Duncan Ranch Site 1 local soil series are represented by the Likes,
Lincoln, Tivoli, Mobeetie and Veal, and Mansker series (US Dept of Agriculture
1966). The surficial soils on the site are basically alkaline and calcic in
chemistry. Crests and uplands soils are moderately fertile and support mixed and
short grasses and various prairie fauna. In general, the uplands in the region are
not well suited for modem crop production, though some areas farm sorgum and
wheat without irrigation (Peterson 1988:7). Soils on the terraces and floodplains
support a variety of riverine fauna and flora, and could have been conducive to
prehistoric agriculture (Martin, personal communication 1993). Terraces were
often selected for agricultural/horticultural settlements because of soil fertility and
protection from flooding (Ferring 1992:25). In general, the upland areas are
dominated by Mansker soils; other soil types are evident on the floodplain and
terraces. Numerous calcareous outcroppings in the area from the Ogallalah
formation (Frederick 1993:80-81) provided dolomite, calcite, and gypsum for
architectural materials, ceramic temper, flooring, pigments, and grinding
Archaeologist recognize the importance of soils in delineating sites and
defining occupations in stratigraphic levels. Lintz (personal communication 1994)
suggests that the lack of defined and separate occupations on the Duncan Ranch
Site 1 may be because of the mixing of artifactual material during a period of slow
or little soil deposition. However, there are subtle changes in soil composition in
the stratigraphic levels during testing and the darkest charcoal laden stratum exists
from approximately 25cm to 65 cm below surface. At this point in the study,
more radiocarbon dates would help in the interpretation of the soil horizons.
A period of floodplain stability and soil formation between 2000 and 1000
B.P. was identified through pedogenic data in north central Oklahoma (Hall 1977;
Artz and Reid 1984); in southwestern Oklahoma (Pheasant 1982; Ferring 1986c;
Hall and Lintz 1984; Hall 1987; Ferring and Hall 1987; and in north central
Texas (Ferring 1986b, in press) (c.f. Ferring 1992:12-13). Termination of these
soils was characterized by alluviation and burial between A.D. 800-1200 (Ferring
White Deer Creek cuts a swath near the site through the upper alluvial soils
revealing the sands, silts, clays, and gravel beds of the Ogallalah deposition. Its
periodic flooding redeposits soils, downcuts and erodes the terrain. Possible