- Social Sciences
- African Studies
- American Studies
- Asian Studies
- Communication Sciences
- Ethnic Studies
- European Studies
- Gender Studies
- Physical Sciences
- Life Sciences
- Animal Communications
- Cell Biology
- Evolutionary Biology
- Food Science and Technology
- Human Anatomy
Landscape Change and the Cultural Evolution of the Hohokam along the Middle Gila River and Other River Valleys in South-Central Arizona
by John C. Ravesloot, Michael R. Waters
Landscape Change and the Cultural Evolution of the Hohokam along the Middle Gila River and Other River Valleys in South-Central Arizona
John C. Ravesloot, Michael R. Waters
Updated: November 4th, 2012
LANDSCAPE CHANGE AND THE CULTURAL EVOLUTION OF THE HOHOKAM ALONG THE MIDDLE GILA RIVER AND OTHER RIVER VALLEYS IN SOUTH-CENTRAL ARIZONA
Michael R. Waters and John C. Ravesloot
Changes in river floodplain morplzology can have devastating consequences for irrigation agriculturalists. Channel erosion occurred in the late nineteenth century, on tlzefloodplain of the middle Gila River, Arizona and severely impacted the native Akitnel O'odham (Pima)farmers. Prior to the Akimel O'odham, the prehistoric Hohokam also pursued irrigation agriculture along this river Geoarchaeological investigations of the Gila Riverfloodplain document a major period of channel cutting and widening sometime beM1eenA.D. 1020 to 1160. This channel erosion is coincident with thepartial abandonment of large Hohokam villages and significant population rearrangements. It also marks the beginning of a major social reorganization when ball-courts were replaced by platform mounds as the social integrative structure and the Hohokam sphere of influence contracted. Other rivers utilized by the Hohokanz-the Santa Cruz Rivei; San Pedro Rive]; and Tonto Creek--nlso experienced channel cutting between A.D. 1050 and 1150. Thus, a regional episode of channel erosion appears to have been a major factor that contributed to the reorganization seen in the Hohokam archaeological record. These synclzrotlous landscape changes ~)ouldhave severely impacted Hohokam irrigation systems and food production capabilities. This undoubtedly created stresses within Hohokam society which in turn may have accelerated social, political, economic, ideological, and demographic changes that were already underway.
Cambios morfoldgicos en el cauce de un riopueden tener cotlsequencias devastadoras para la agricultura de irrigacidtl.La erosidn del Rio Gila Medio, que ocurrid a1 final del siglo IXX impact6 severamente a 10s agricultores tlativos Akimel O'odam (Pima). Antes de e'stos, la poblacidn prehistdrica Hohokam practicd agricultura de irrigacidn en este rfo.ltlvestigaciotles geoarqueoldgicas en el cauce del Rio Gila documentan utl perfodo de gran entrenchamiento y ensanchanziento entre I020 y I160 d.C. Esta erosidn coincidid con el abandonoparcial de estensos asentamientos Hohokam en esta drea del rio y con un movinziento demogrrijico sigtl$cativo. Este proceso tambie'n marca el principio de utla reorganizacidtl social, cuando se reemplazaroil las canchas de pelota por 10s tnontfculosde platafornza comofornzas arquitectdnicas integrativas y se contrajo la esfera de influencia Hohokam. Otros rios utilizados intetlsametlte por la poblacidn Hohokam, incluyendo el Rio Santa Cruz, Rio San Pedro, Quebrada Tonto, y Ri'o Salt, tambie'tl esperimentaron entrenchanliento entre 1050 y 1150 d.C. Por lo tanto, un episodio regional de erosidn del cauce parece haber sido utl factor principal que cotltribuyd a la reorganizacidtl observada en el registro arqueoldgico Hohokam. Estos canzbios sincrdnicos en el paisaje habrian impactado severamente 10s sistemas de irrigacidn y capacidad de produccidn de alimentos. Esto sin duda cred presiones dentro de la sociedad Hohokam , la que entotlces habria acelerado 10s incipientes cambios sociales, politicos, ecotldmicos, ideoldgicos, y demogrrificos de ese tienzpo.
rehistoric agriculturalists known as the turies, the Hohokam sphere of influenceexpanded to Hohokarn intensively occupied the middleGila include almost all of south-central Arizona (Figure and SaltRiverValleys of Arizona for hundreds I). After this long period of sustained growth, expanof years (Gumeman 1991;Haury 1976). Character-sion of the size and complexity of villages, and culistics of the Hohokam cultural pattern (i.e., red-on-tural elaboration,the Hohokam experienced a period buff ceramics, distinctive iconography, pithouse of major reorganization. Between A.D. 1050 and architecture,villagelayout,ballcourts,mortuary prac-1150settlementsthat had been occupiedfor centuries tices, and irrigation agriculture)are identifiablein the were partially abandoned,the geographicrange of the archaeological record around A.D. 700-750 (Doyel Hohokam pattern decreased, and the regional system 1991; Wallace et al. 1995). Over the next three cen-of more than 125 ballcourts was gradually replaced
Michael R. Waters .Department of Anthropology, Texas A&M University, College Station, TX 778434352 John C. Ravesloot Cultural Resource Management Program, Department of Land and Water Resources, P.O. Box 2140, Gila River Indian Community, Sacaton, AZ 85247
American Antiquity, 66(2), 2001, pp. 285-299 Copyright@ 2001 by the Society for American Archaeology
AMERICAN ANTIQUITY [Vol. 66, No. 2, 20011
Figure 1. Map of the drainage basin of the Gila River which includes the Santa Cruz River, San Pedro River, Tonto Basin, and Salt River. The Gila River Indian Reservation and other localities mentioned in the text are shown on the map. The dash- dot line delineates the approximate maximum limits of Hohokam influence near the end of the Pre-Classic period.
by platform mounds as the predominant form of pub- lic architecture (Abbott 2000; Crown 1991; Doyel 199 1; Fish 1989; Gregory 1991). This reorganization occurred throughout the entire Hohokam region, but was most dramatic in the Hohokam core area of the Salt and Gila River Valleys.
By A.D. 1250, the Hohokam had developed a new organizational system, villages with platform mounds were established, people aggregated into fewer but larger settlements, canal systems were con- solidated, and many other ritual and material aspects of the culture changed (Abbott 2000; Doyel 1991; Fish 1989; Haury 1976). This new organization sus- tained itself until around A.D. 1400 to 1450, when the system collapsed. After the mid-fifteenth century, villages and canal systems were abandoned and it is unclear how the people living along the Salt and Gila Rivers, and elsewhere in the Hohokam homeland were organized. This was also aperiod of major reor- ganization throughout the entire American South- west (Cordell and Gumerman 1989; Fish et al. 1994).
Many hypotheses have been proposed to explain the two major periods of reorganization in Hohokam prehistory, ranging from changing environmental conditions to changing trade and alliance networks (e.g., Crown 1991; Doyel 1980, 1991; Fish 1989; Gumerman 1991; Rice 1998). However, a mecha- nism for culture change that has not received ade-
I \ I I Figure 2. Geomorphic map of the Gila River Indian Resenation, Arizona. Map shows the distribution of landforms, including the Gila River terraces. The locations of the four strati- graphic transects are indicated. Note that the Pleistocene terrace is largely covered by the eolian sand sheet.
AMERICAN ANTIQUITY [Vol. 66, No. 2,20011
quate attention is the impact of landscape change on Hohokamcultural evolution (e.g., Huckleberry 1995; Nials et al. 1989; Weaver 1972). The term "land- scape," as used in this paper, refers to the geomor- phic landscape-the platform on which all biological organisms (plants, humans, and other animals) inter- acted through time (Waters 1992). Physical land- scapes are dynamic and constantly changing (e.g., the channel on a floodplain may downcut, widen, and later backfill) and humans are adapting to these changes. Diachronic landscape reconstructions, cou- pled with archaeological data, provide insights into human-land interaction.
This paper presents the results of a multiyear study of the middle Gila River Valley that was designed to reconstruct the late Quaternary land- scape history of the river's floodplain. This recon- struction is pivotal as a context for understanding the evolution of Hohokam society through time. The GilaRiver was the lifeline for the Hohokam and pro- vided them with the only reliable source of water in an otherwise arid environment. It was along this river and the other perennial rivers in south-central Ari- zona that the Hohokarn culture emerged, flourished, and collapsed.
The Gila River
The Gila River, which drains about 150,000 km2,is a major tributary of the Colorado River (Thomsen and Eychaner 1991). This paper is concerned with the middle segment of the GilaRiver, specifically the segment that emerges from a bedrock gorge 26 km east of Florence and flows over a broad desert basin to its junction with the Salt River (Figure 1). This segment of the middle Gila River has a watershed (east of Sacaton) of more than 47,400 km2.
The middle Gila River is characterized by a wide streambed with a braided pattern of sandy and grav- elly bars and channels. The streambed is entrenched and flanked by a floodplain and three terraces (Fig- ure 2). Adjacent to the streambed is a low terrace, Terrace 1(T-1), which is still inundated during large floods. Above this is a prominent terrace, Terrace 2 (T-2), which is underlain by sediments dating from 18,000 B.P. to present. Still higher is a single Pleis- tocene terrace, Terrace 3 (T-3). Water is usually absent from the river, except following heavy rain- fall and in most cases this water infiltrates into the ground after flowing a short distance without ever reaching the confluence with the Salt River. Only dur- ing extreme floods, such as that of October 1983 and February 1993, does the river have enough flow to reach the Salt River.
Late Quaternary *lluvial History Of the
Middle Gila River
Previous research on the alluvial history of the Gila River is confined to that of Huckleberry (1993,1994, 1995). Expanding on this research, we investigated the alluvial stratigraphy of the middle Gila River along four transects across the floodplain (Figure 2). Backhoe trenches were excavated at 50 to 100 m intervals from the edge of the Pleistocene terrace (T- 3) across the tread of the main Holocene terrace (T- 2) and in some places onto the lower Holocene T-1 surface. The stratigraphy of each trench was exam- ined and described, and radiocarbon samples were collected for dating.
The late Pleistocene and Holocene sediments are
divided into six units, labeled I to VI, from oldest to
youngest. One paleosol is defined. A generalized
stratigraphic cross section of the late Quaternary
alluvial stratigraphy of the Gila River is presented in
Figure 3. Temporal control for this sequence is pro-
vided by 27 radiocarbon dates and diagnostic arti-
facts. All dates presented here are in calibrated years
before present (cal B.P.; Stuiver and Reimer 1993).
For a detailed discussion of the late Quaternary geol-
ogy of the middle Gila River see Waters and Raves-
The late Pleistocene alluvial history of the Gila
River begins after a period of erosion. Sometime
before 18,000 cal B.P., the Gila River abandoned its
floodplain and downcut into its alluvium creating T-
3 (Figure 4a). During channel downcutting, the Pleis-
tocene terrace alluvium was beveled and a deep
channel was eroded along the axis of the GilaValley.
By 18,000 cal B.P. and probably earlier, alluvium
began to fill the channel with sand and gravel (Fig-
ure 4a; Unit I). This continued until about 4000 cal
B.P. During this 14,000 year period, the channel of the Gila River narrowed. Starting around 5000 cal B.P., overbank silts (Unit 11) began to accumulate on the older channel sediments adjacent to the channel and on the beveled Pleistocene sediments. On the floodplain, overbank silts continued to accumulate until around 2000 cal B.P. (Figure 4b). This was fol- lowed by renewed overbank deposition of silts and clays (Unit III; Figure 4c) that continued to accu- mulate until about 500 cal B.P. Concurrent with the
175240 yr B.P.
400+70 yr B.P.
945+45 yr B.P.
Sand Sheet 965240 yr B.P.-
470+40 yr B.P. 475rt60 yr B.P. 555k60 yr B.P. 570240 yr B.P. 1255rt55yr B.P. 1630245yr B.P. 1765260 yr B P. 2070280 yr B.P.
c185 yr B.P.
2235+65 yr B.P. 2460+60 yr B.P. 2490245 yr B.P. 2660+50 yr B.P. 2665+50 yr B.P. 3430+50 yr B.P. 4200255 yr B.P. 4460+50 yr B.P. 4485+55 yr B.P. 4870+80 yr B.P.
3920250 yr B.P.
4580250 yr B P. 89152105 yr B.P. 14,7702200yr B.P.
Figure 3. Generalized diagram of the late Quaternary alluvial units of the middle Gila River. Also shown are the radiocar-
bon dates in uncalibrated 14C years B.P.from these units.
later phases of deposition of Unit I1Z (ca. 850-950 cal B.P.; A.D. 1050-1 150) was a significant period of channel downcutting and widening (Figure 4d). This wide channel eventually filled with coarse and fine sediments (Unit IV). A period of floodplain sta- bility occurred from about 500 cal B.P. until about 200 cal B.P. when the Orchard soil formed on Unit 111.Deposition continued in the channel until about 200 cal B.P. (ca. A.D. 1715-1885) and overbank silt buried the Orchard paleosol (Figure 4e; Unit V).
At the end of the late nineteenth century, the chan- nel of the Gila River downcut and widened, creating a new streambed covered with sandy bars and chan- nels (Figure 4f; Unit VI). During this process, it aban- doned its older floodplain creating T-2. The river also beveled parts of T-2, creating the T-1 erosional ter- race that is underlain in places by the same late Pleis- tocene and Holocene sediments that underlie T-2. The Gila River is now confined within a deep, wide channel that rarely overtops its banks. The T-1 sur- face is sometimes covered during large floods and the T-2 surface has only been covered during rare floods in the nineteenth century, prior to the devel- opment of upstream flood control features. This flooding left thin overbank deposits (Unit V).
The alluvial stratigraphy of the middle Gila River is characterized by several periods of sustained
aggradation punctuated by major sedimentological and depositional changes. These changes to the Gila River floodplain appear to be responses to fluctuat- ing sediment loads derived from the upstream reaches of the drainage basin, and climatically dri- ven changes in flood magnitude and frequency dur- ing the late Quaternary (Waters and Ravesloot 2000). Changes to the Gila River in the late nineteenth cen- tury are attributed to high-magnitude flooding that occurred in the late 1800s and early 1900s, coupled with the human impacts that were occuning along the stream course prior to this time (e.g., wood cut- ting, reductions of streamflow, reduction in vegeta- tion protection on the banks, and watershed destruction; Dobyns 1981; Huckleberry 1993; Rea
1997; Waters and Ravesloot 2000; Wilson 1999).
Hohokam Cultural Evolution and Landscape
History along the Middle Gila River
The evolution of Hohokam culture, outlined in four developmental periods, is placed within the context of documented landscape changes along the middle Gila River in order to assess how landscape change may have contributed to this evolution. The Hohokam chronology proposed by Dean (1991) is followed in this paper.
It has long been believed that an agricultural way
AMERICAN ANTIQUITY [Vol. 66, No. 2,20011
- ca. 18,000 to 5000 cal yr B.P. (b) ca. 5000 to 2000 cal yr B.P
- ca. 2000 to 800-950 cal yr B.P. (d) ca. 800-950 cal yr B.P.
- ca. 800-950 to 200 cal yr B.P. (f) < 150 cal yr B.P.
Figure 4. Series of diagrams outlining the sequential landscape history of the middle Gila River from 18,000 cal B.P. to present.
of life based on the cultivation of corn, beans, squash, and cotton was established along the middle Gila River at the beginning of the FormativeIPioneer Period (ca. A.D. 300; Doyel 1991; Haury 1976). However, it is likely, based on the presence of Mid- dle and Late Archaic Period settlements in the area (Bubemyre et al. 1998; Fish 1968) that yet-to-be identified preceramic horticultural settlements are present, but lie deeply buried in the Holocene allu- vium. The emergence and archaeological identifica- tion of the Hohokam Culture appears as early as
A.D. 700-750 during the Snaketown phase of the Pioneer Period (Dean 1991; Doyel1991; Wallace et al. 1995; Wilcox 1979; Wilcox and Sternberg 1983). The Hohokam cultural pattern, reflecting the begin- nings of an integrated, regional belief and ritual sys- tem, was characterized by irrigation agriculture, production of red-on-buff pottery, a distinctive iconography, exotic artifacts, a cremation mortuary complex, and adoption of a public architectural com- ponent, the ballcourt (Crown 199 1 ;Doyel199 1; Fish
1989; Gumerman 1991; Haury 1976).
During the emergence and establishment of the Hohokam pattern the Gila River was characterized by a narrow channel with a broad floodplain (Figure 4c). Terrace 2 as seen today, and T-0 and T-1 did not exist. Instead, the floodplain was lower and covered with silt (Unit 11) and later with silty clay (Unit ID) as the floodplain aggraded. Flow was perennial and confined to the channel except during floods when the water would overtop its banks and inundate the adjacent lowlands. This flooding resulted in overbank deposition and vertical aggradation of the floodplain. As during the historic period, the banks of the chan- nel were probably lined with cottonwood and wil- low trees, and mesquite forests and grasslands extended over the floodplain (Hoover 1929; Rea
1983, 1997). The geological evidence indicates that there was no widening of the channel during the Pio- neer period. The river channel was stable and had a predictable streamflow regime providing excellent conditions for the establishment of canal systems and the development of Hohokam culture.
Pioneer period canals appear to have brought water to crops planted on the floodplain. These for- mer fields lie buried beneath the tread of T-2. Although overbank flooding may have periodically damaged canal systems and other structures on the floodplain, these floods would have also renewed nutrients in the fields. In addition, Haury (1976) has shown that agricultural fields were also located on the Pleistocene terrace (T-3). On T-3, crops were probably planted in the eolian sand sheet. The tex- ture of the sediments on the floodplain and those cov- ering the Pleistocene terrace would have been very suitable for plant growth.
During the pre-Classic, extending from A.D. 750
to around A.D. 1150, the middle Gila River Valley
and the Salt River Valley were the primary focus of Hohokam regional development. The Colonial period (A.D. 750-950) was characterized by the establishment of numerous villages, hamlets, and farmsteads throughout the region, including areas peripheral to the middle Gila and Salt River Valleys where irrigation agriculture could not be practiced (Czaplicki and Ravesloot 1989; Doelle and Wallace 1991; Doyel 1991; Haury 1976). Habitation sites containing courtyard groups focused on extramural work areas became a common organizational pattern of settlements (Howard 1985; Wilcox et al. 1981). Ballcourtsincreasedinnumber,andtheirgeographic range extended throughout south-central Arizona (Wilcox 1991; Wilcox and Sternberg 1983). During this period, the Hohokam sphere of influence expanded to its maximum with Hohokam cultural traits found throughout most of south-centralArizona (Doyel 1991; Haury 1976). The Sedentary period
(A.D. 950-1 150) witnessed continued growth in the number, size, and extent of Hohokam settlements, including ballcourt villages, and irrigation networks in the Gila and Salt River Valleys and peripheral areas (Ciolek-Torrello and Wilcox 1988; Crown 1991; Doyel 1991; Fish et al. 1992; Haury 1976; Spoerl and Gumerman 1984; Wilcox and Stemberg 1983). Many large villages reached their maximum size and complexity at this time.
During most of the pre-Classic, the Gila River maintained a stable configuration similar to that of the preceding Pioneer period (Figure 4c). There is no evidence of channel widening and silty clay (Unit HI) continued to accumulate on the floodplain as it built vertically.
During all periods (Pioneer, Colonial, and Seden- tary), the largest and most permanent settlements in the Middle Gila River Valley were located on the edge of the Pleistocene terrace (T-3) (Brodbeck and Neily 1998; Gregory 1991; Gregory and Huckle- berry 1994; Haury 1976). The Hohokarn most likely determined that the highest terrace (Pleistocene ter- race, T-3) was a good place to settle because it had not flooded in recent memory. Furthermore, locations on the edge of the Pleistocene terrace offered views of the agricultural fields on the floodplain below and also maximized farmable acreage on that surface.
By the end of the Sedentary period, the Hohokarn cultural pattern that had been established during the previous centuries was significantly transformed. This reorganization may have begun as early as A.D.
1050 (Doyel1991,1993). During this time, villages in areas peripheral to the Gila and Salt River Valleys were abandoned and populations settled primarily along the major river drainages. In addition, some ancestral villages such as Snaketown were depopu- lated, and the populations moved and reorganized in nearby locations (Abbott 2000; Crown 1991; Doyel 1991; Gregory 1991; Haury 1976; Ravesloot and Lascaux 1993; Wilcox et al. 1981).
During the subsequent Classic period (A.D. 1150-1450), the changes initiated during the Seden- tary Period continued and resulted in the develop- ment of new architectural forms such as post-reinforced and adobe-wall structures and walled compounds, a shift in mortuary practices from pri- marily cremations to inhumations, areduction in the production of red-on-buff pottery and an increase in red ware pottery, and a major shift in regional exchange networks. The period also witnessed the continued decline and eventual collapse of the regional ballcourt system and the emergence of a new form of public architecture, the platform mound (Gregory 1987, 1991; Wilcox and Stemberg 1983).
The Classic period was also characterized by a hierarchy of settlement types: villages with only one or a small number of walled residential compounds, and those with one or more platform mounds as well as other compounds. Hohokam villages became con- nected by singular canal systems, consolidating the independent systems into longer central systems (Gregory 1991), forming almost a linear settlement pattern of irrigation communities. Settlements dis- tributed along a single canal or canal system have been postulated to represent irrigation communities (Doyel1980; Howard 1987). Irrigation communities were comprised of one or more platform mound vil- lages that served as administrative centers to regu- late the allocation of water and organize the
construction and maintenance of the canal system. The Classic period was also a time of agricultural diversification. The Hohokam enhanced food pro- duction capabilities by supplementing their irriga- tion agriculture with expanded dry and floodwater (Ak-chin) farming strategies on the bajadas (Crown 1987,1991; Fish et al. 1985; Fish et al. 1992; Masse 1991).
Coinciding with these cultural changes is a major change on the floodplain of the middle Gila River. Sometime between A.D. 1020 and 1160, the chan- nel of the Gila River downcut and significantly
AMERICAN ANTIQUITY [Vol. 66, No. 2, 20011
widened (Figure 4d; Hucklebeny 1995; Waters and Ravesloot 2000). When the two oldest radiocarbon dates obtained on charcoal (945 + 45 B.P. [AA- 271081 and 965 +40 B.P. [AA-271181) from the channel alluvium (Unit IV) are calibrated using CALIB 4.2 (Stuiver andReimer 1993), the calibrated age of the samples lies somewhere between A.D. 1020 and 1160, closely corresponding with the pre- Classic to Classic transition. The channel downcut- ting and widening along the middle Gila River corresponds to a period of intensified high-magni- tude flood activity in southem Arizona (Ely 1997). As during the historic period, the riparian zone must have been destroyed as the channel widened and cut into its floodplain. The channel created as a result of the flooding was comparable in width to the modem
channel of the GilaRiver (modem width ranges from .5 to 2.3 km, but is typically at least 1 km in most places) and may have been even wider. Channel widening disrupted nearly a millennium of floodplain stability. This wider channel had a braided streambed and the main flow channel shifted over the streambed with each large flow. Because the position of the average-flow channel changed relative to established canal headgates, it would have been difficult for the Hohokam to get water into their canals. Thus, every year Hohokam engineers coped with the problem of diverting streamflow across the porous streambed. At times the channel would have been close to the headgates and at other times on the opposite side of the streambed. Any temporary diversion dams con- structed to get water into the headgates were vul- nerable to being washed out within such a dynamic channel environment. Dealing with these problems required considerable labor and organization to keep
the canal systems operating.
The historic period of the late nineteenth and early twentieth centuries provides an analog to the Clas- sic period situation. As a result of the floods of the late nineteenth century, the channel of the Gila River widened and the Akimel O'odham had a very diffi- cult time trying to utilize the water that flowed down the channel. Lee (1904) reports that in one year intake dams built in the streambed of the Gila were washed out five times. Historic records and oral histories indicate that the Akimel O'odham had tremendous problems trying to divert water into the headgates of their canals as the low-flow channel shifted over the streambed and periodic high flows destroyed diver- sion structures (Figure 5; Hoover 1929; Rea 1997; Wilson 1999). In fact, the situation became so criti- cal that Akimel O'odham were malnourished as a consequence of repeated crop failures. This was not only because of the reduced flow down the Gila River, but also because of the problems related to get- ting water into the canals.
The channel widening in the eleventh century
A.D. along the middle Gila River is likely a major contributing factor leading to the cultural changes seen at the start of the Classic period. Such a change to the configuration of the Gila River would have threatened the Hohokam lifeway, forcing a major change in irrigation and agricultural strategies. The increased demands for labor to maintain and repair the canal systems may have led to consolidation of the canal systems and communities. This may have also led to increased interaction and development of more complex organizational strategies for dealing with water allocation and maintenance of canal sys- tems (Howard 1993). The Classic-period Hohokam responded to fluvial instability by pooling their resources and organizing their labor, reengineering their canals, placing canal heads in stable positions, and increasing and diversifying food production by pursuing dry and floodwater farming. We believe that the dramatic reorganization in Hohokam vil- lages and canal systems between the pre-Classic and Classic periods may have been, in part, a response to the widening of the middle Gila River channel.
During the remainder of the Classic period, the wide channel of the Gila River progressively nar- rowed as the channel became clogged with coarse alluvial sediments (Figure 4e). At the same time, the riparian zone was likely being restored. The flood- plain of the river continued to aggrade with silty clay until ab0utA.D. 1300 to 1400.At this time the flood- plain became stable and soil formation dominated. Little flooding appears to have taken place. This sta- ble floodplain surface was probably farmed and larger villages were established on the overlooking Pleistocene terrace.
The end of the Classic period was marked by the collapse of the regional platform mound system and depopulation of the Gila and Salt River Valleys. It has been suggested that the abandonment of late Classic-period communities coincided with a period of drought and large floods that substantially impacted and in some cases destroyed the canal sys- tems on which these irrigation agriculturalists depended (Nials et al. 1989). This period of abrupt
Figure 5. Photograph of the Gila River taken near Olberg in March 1915. View is looking downstream with Sacaton visible to the left in the distance. At least 50 Akimel O'odham men assisted by 21 two-horse powered wagons repair a brush intake weir on the streambed of the Gila River. They are trying to divert water into the historic Santan Canal. The Gila River chan- nel is about one kilometer wide in this view. (Homer Shantz Collection, Herbarium, College of Agriculture, University of
change in community organization has been tenta- tively defined and dated as the Polvoron phase or the terminal period of prehistoric occupation in the Gila and Salt River Valleys (Sires 1983). The Polvoron phase was characterized by dispersed rancheria-style villages, shallow pithouses, "degenerate" red ware, and a mixed subsistence strategy (Chenault 1993; Crown 1991; Doyel 1991; Hackbarth 1995; Sires
1983; Teague and Crown 1984). No major landscape change along the middle Gila River is documented for this time period (Figure 4e). Thus, the demise of the Classic-period Hohokam in the middle Gila River valley cannot be linked to landscape changes.
The Protohistoric period (ca. A.D. 1450 and 1700) marks the time between the end of the Hohokam and the beginning of the Spanish colonization of the Southwest. Historic descriptions of this area record the presence of Piman-speaking groups that pursued floodwater and small-scale irrigation farming on the active floodplain of the Gila River. Protohistoric
fields were located on the T-1 and T-2 surfaces. By this time, the Gila River had returned to its original narrow channel configuration with a fully developed riparian zone and a broad floodplain that was regu- larly flooded. Due to a multitude of factors (e.g., introduction of wheat, warfare between the Akimel O'odham and Apache, and subsequent aggregation of settlements), irrigation agriculture became more intensified after the beginning of the historic period (Wilson 1999). The period of channel downcutting and widening documented on the middle Gila River in the late nineteenth century decimated the tradi- tional farming lifeway of the Akirnel O'odham (Rea
Cultural and Landscape Changes along the
The Salt River is a major tributary of the Gila River that lies directly north of the middle Gila River. Like the Gila, it was a core area of Hohokam settlement
AMERICAN ANTIQUITY [Vol. 66, No. 2, 20011
(Figure 1). The sequence of Hohokam cultural evo- lution on the Salt River is similar to that along the Gila River. However, occupation may have been more concentrated along the Salt River because the river carried more water that could be used for irri- gation. The streamflow of the Salt River, delineating periods of average, below and above average flows, has been reconstructed for the period from the eighth century A.D. to the present by the analysis of tree rings in the upper watershed of the Salt River. (Gray- bill 1989). This streamflow chronology was used to infer how above-average streamflow and inferred rare high-magnitude floods, as well as periods of drought, may have impacted canal irrigation systems emanating from the Salt River. The streamflow chronology and its interpreted effects on the Hohokam canal systems is cited as the cause for some of the developmental changes seen in the Hohokam sequence (Nials et al. 1989). Streamflow reconstructions spanning the period from the pre- Classic to the Classic periods show that the largest annual discharge in the whole time series occurred at A.D. 899 and that the ninth century was marked by hydrologic instability with several large floods. Major flooding events would have destroyed or dis- rupted the Hohokam canal systems along the Salt River. Nials and others (1989:75) state that stream- flow from A.D. 1052 to 1196 was "apparently quite favorable for Hohokam irrigation agriculture." How- ever, high-magnitude annual discharges occurred in
A.D. 1052, 1087, and 1 129. Streamflow during the preceding period from A.D. 900 to 1051 was even more favorable for irrigation agriculture. Thus, based on streamflow data alone, there were no real hydro- logic changes at the juncture between the pre-Clas- sic and Classic periods. Regrettably, there is only limited alluvial stratigraphic data for the lower Salt River (Huckleberry 1999), and streamflow recon- struction (while useful) is not a substitute for a strati- graphic sequence. Even small periods of flooding can cause major landscape changes if the internal land- scape variables are favorable for erosion (Bull 1991; Schumm 1977). Such an event would not stand out in the streamflow chronology, but would be evident in the stratigraphic record.
We believe that the alluvial history of the Salt River may mimic that of the middle Gila River because the Salt River is a major tributary of the Gila River. The Gila River controls the base-level of the Salt River. As a consequence, the Salt River would respond to base-level changes on the Gila River; if the channel of the Gila River downcut and widened, it is likely that the channel of the Salt River would also downcut and widen. This event would have cre- ated tremendous problems for the irrigation-based agricultural systems along the Salt River as it did for the Hohokam along the Gila River. The possibility of channel cutting along the Salt River between A.D. 1050 and 1150, while geologically reasonable, requires testing.
Hohokam Reorganization and Landscape
Change in Other Areas of Southern and
The influence of the Salt and Gila River Hohokam extended to the south and east to include the Tucson Basin, San Pedro Valley, and Tonto Basin (Figure 1). In these areas, landscape changes are also coincident with cultural changes at the pre-Classic to Classic period transition. In the Tucson Basin and San Pedro Valley, there is also evidence of landscape change at the end of the Hohokam Classic period.
Prehistoric agriculturalists lived south of the Salt and
GilaValleys in the Tucson Basin (Figure 1; Czaplicki
and Ravesloot 1989; Doelle and Wallace 1991; Fish
1989; Fish et al. 1992; Kelly 1978; Wallace 1995,
1996; Wallace and Doelle 2001 ;Wallace et al. 1995).
By A.D. 1050, after a long period of in situ devel
opment, large villages that were occupied year-round
existed in the Tucson Basin. The occupants of these
villages farmed the floodplain of the Santa Cmz
River (the main stream traversing the basin) and
lands on the bajadas fringing the mountains. A sys-
tem of canals was established extending from the
Santa Cmz River, although it was less elaborate than
the canal systems along the Salt and Gila Rivers.
The inhabitants of the Tucson Basin went through
a major reorganization betweenA.D. 1050 and 1150.
Importation of buff-ware vessels from the Gila and
Salt River Valleys nearly ceased and locally pro-
duced brown wares became predominant. Ballcourts
stopped functioning as important integrative facili-
ties and were eventually abandoned. Much of the
associated Hohokam material culture was replaced,
and the populations of large sites became dispersed
into small hamlets (Doelle and Wallace 1991; Wal-
lace 1995, 1996; Wallace and Doelle 2001; Wallace
et al. 1995). Agricultural pursuits became more
diverse with an increased emphasis on dry and flood- water farming on the bajadas. Some canal systems were abandoned (e.g., at Los Morteros; Wallace 1995), and new canals were constructed. With these changes came shifts in architectural forms and mor- tuary practices. Further, settlements ceased to be occupied year-round indicating increased mobility. By the early Classic period, populations had aggre- gated and lived in large dispersed communities with associated platform mounds. This pattern continued until approximately the mid-fifteenth century when another reorganization occurred.
Extensive alluvial stratigraphic and geochrono- logic studies have been undertaken along the Santa Cruz River (Freeman 1997; Haynes and Huckell 1986; Waters 1988). These studies show that prior to A.D. 1050, the Santa Cruz River was aggrading. This was an ideal time for the experimentation, emer- gence, and development of agriculture (Freeman 1997; Huckell1990). Stratigraphic and radiocarbon evidence indicates that the dramatic cultural reorga- nization between A.D. 1050 and 1150 is coincident with the cutting of a deep channel into the floodplain of the Santa Cruz River. Arroyo cutting would have disrupted centuries of floodplain stability and would have created stress on the agriculturally dependent populationin the Tucson Basin. Entrenchment would have left canal headgates high above the channel
floor, making it virtually impossible to restore the
previously existing canal system. Channel cutting
and its resultant negative impacts on the Tucson
Basin Hohokam between A.D. 1050 and 1150 were
most likely a contributing factor that precipitated the
reorganization seen during this time. Subsequently,
the channel filled with sediment and floodplain sta-
bility returned to the Santa Cruz River. However,
arroyo cutting occurred again on the floodplain of
the Santa Cruz River sometime betweenA.D. 1410
and 1450 (Waters 1988). This arroyo-cutting episode
coincides with the reorganization and depopulation
of the Hohokam in the Tucson Basin at the end of
the Classic period.
Lower San Pedro Valley
Archaeological information on the late prehistoric period of the lower San Pedro Valley is limited (Wal- lace and Doelle 1997,2001). However, the available data indicate that the late-prehistoric inhabitants of the lower San Pedro Valley had a similar history of cultural evolution and reorganization as defined for
the Tucson Basin. By A.D. 1050, a number of large Hohokam villages were nucleated around ballcourts in the lower San Pedro Valley. Sites of this time period have imported buff ware from the Gila River Valley as well as other Hohokam artifact types and architectural styles. As in the Tucson Basin, the Hohokam went through a major reorganization between A.D. 1050 and 1150 when settlements became more dispersed, the ballcourt system col- lapsed, ritual paraphernalia was replaced, and the importation of buff wares ceased. After 1150, many new traits appear, such as adobe architecture, walled compounds, and platform mounds. Populations also appear to have concentrated into large settlements, and agricultural diversification intensified. The Clas- sic period pattern of platform mounds and associ- ated compound communities continued to flourish into the mid-fifteenth century.
The alluvial history of the San PedroValley is well documented (Haynes 1987). This stratigraphy shows that there was a period of erosion betweenA.D. 1050 and 1150 (synchronous with the downcutting of the Santa Cruz River), when the river channel downcut into its floodplain. This period of channel entrench- ment would have disrupted centuries of floodplain stability and agricultural pursuits on the floodplain. This stress may have been an important factor in the reorganization seen in the late eleventh and early twelfth centuries. Also, it is interesting to note that, like the Santa Cruz River, there was another period of channel entrenchment in the mid-fifteenth century, coincident with reorganization at the end of the Clas- sic period.
A major cultural change is noted in the archaeolog- ical record of the Tonto Basin ar0undA.D. 1050. The Tonto Basin is located 95 km northeast of the Salt- Gila Basin and is drained by both the middle Salt River and Tonto Creek (Figure 1). The Hohokam expanded into the basin around A.D. 700 (Dean 2000; Elson and Gregory 1995; Elson et al. 1995; Rice, ed. 1998) and pursued irrigation agriculture on the floodplain (Waters 1998), as well as floodwater farming on the alluvial piedmont. The material cul- ture, site layout, and other evidence shows that these emigrants were Hohokam with clear affinities to the people in the Salt and Gila River Valleys.
The alluvial stratigraphy of the Salt River and Tonto Creek in the Tonto Basin shows that just prior
AMERICAN ANTIQUITY [Vol. 66, No. 2, 20011
to A.D. 1050, both the Salt River and Tonto Creek cut into their respective floodplains at A.D. 1000 (Waters 1998). Filling of the channel began before
A.D. 1215-1280. This channel cutting coincides with the end of the pre-Classic and may have initiated a period during which irrigation agriculture on the floodplain of either river in the Tonto Basin was dif- ficult to practice. Furthermore, it is presumed that this event would have led to a reduction in popula- tion size in the Tonto Basin, with remaining settle- ments located on the bajada.
~~d~~~~~ changes along river floodplains canbe destabilizing events for irrigation agriculturalists. In the H~hokam core area along the middle Gila River, and most likely along the Salt River, and in the out- lying areas of the Tucson Basin, lower San PedroVal- ley, and Tonto Basin, landscape change in the form of channel entrenchment and widening coincided with the abrupt cultural changes that took place between the pre-Classic and Classic periods. This seems to be more than mere coincidence. Channel cutting and erosion would have resulted in crop fail- ures, loss of farmland, and the need to abandon older canal systems and construct new ones. These events would have caused serious problems for Hohokam agriculturalists. As shown, a regional episode of channel erosion appears to have contributed to social, political, economic, and demographic changes seen in the Hohokam culture area between ca. A.D. 1050
to 1150 by accelerating cultural changes that were already underway.
While landscape change coincides with cultural changes in the Hohokam area during the pre-Clas- sic to Classic transition, the landscape does not appear to have changed in most places in south-cen- tral Arizona at the end of the Classic period (ca. A.D. 1450). ~h~~~ is no stratigraphic evidence documenting changes to the floodplain of the Gila River, Salt River, or Tonto Creek at this time. Thus, land- scape change does not appear to be a factor in the
the Hohokam in these areas. The exceptions are the Santa Cruz and San Pedro Rivers where cutting occurred at the end of the Classic period. Thus, in places, landscape change was not a factor in the reorganization of the ~ohikamat the end of Classic
taking an human-ecological systems approach when investigating prehistory, one
can look at the environmental and cultural variables that may cause changes in prehistoric cultures. One of these variables is the landscape. Geoarchaeolog- ical studies can determine if landscape changes cor- relate with significant cultural changes. Thus, changes in the landscape should be considered for their possible impact on prehistoric people. Over a decade of geoarchaeological studies in south-central Arizona have shown some correlations and non-cor- relations between landscape change and culture change that help to explain major transitions in Hohokam culture history.
Acknowledgmeizts. The authors thank Jeffrey S. Dean, Mark L. Chenault, Paul Fish, Andrea Freeman, Gary Huckleberry, Tim Kohler, Amadeo Rea, Glen Rice, Patricia Spoerl and one anonymous reviewer for their many substantive comments and editorial suggestions on an earlier draft of this paper. The research for this paper was conducted by the Gila River Indian Community, Cultural Resource Management Program in con- junction with the Pima-Maricopa Irrigation Project (P-MIP). The Pima-Maricopa Irrigation Project was developed by the Gila River Indian Community, resulting from the community's administration of U.S. Bureau of Reclamation, Department of Interior funds under the Tribal Self-Governance Act of 1994
(P.L. 103-413), for the design and development of a water
delivery system utilizing Central Arizona Project water. The
authors would also like to thank Maria Nieves Zedefio for
preparing the Spanish abstract and Lynn Simon for preparing
Abbott, D. R.
2000 Ceramics aizd Community Ovganizatioiz Amorzg the
Hohokam. University of Arizona Press, Tucson.
Brodbeck, M., and R. B. Neily
1998 An Archaeological Survey of tlze Pima Canal andAlte~ native Caizal Alignmerzts in the Blachsatei Maizagemerzt Area, Pima-Maricopa Irrigation Project, Gila River Indiaiz Commuizity.P-MIP Report No. 7. Cultural Resource Man- agement Program, Gila River Indian Community, Sacaton.
Bubemyre, T., M. Brodbeck, and R. B. Neily
1998 A Cultural Resource Survey of the Borderlands Ai.ea,
Gila River Indian Community, Maricopa Courzty, Arizorza.
CRMPTecln~calReportNo.97-23.~~1tural~esourceManagement Progxam, Gila River Indian Community, Sacaton. Bull,W, B, 1991 Geomorphic Responses to Climatic Change. Oxford UniversiQ Press, Oxford. Chenault, M. L.
1993 The Hohokam Post-Classic Polvoron Phase. In Early Desert Farming and Irrigatiorz Settlements: Archaeological Iiwestigatioizs in the Phoenix Sky Harbor Center; Volume 4: Special Studies, Syrzthesis, arzd Conclusions, edited by D. H. Greenwald and J. H. Ballagh, pp. 117-140. SWCAArchae- ological Report No. 93-17. SWCA, Inc., Environmental Consultants, Flagstaff.
Ciolek-Torrello, R. ,and D. R. Wilcox (editors)
1988 Holzokatn Settlemerzt Along the Slopes of the Picaclzo
Mourztalns, Tucson Aqueduct Project. Syizthesis and Con-
elusions. Research Paper No. 35, Volume 6. Museum of Northern Arizona, Flagstaff.
Cordell, L. S., and G. J. Gumelman 1989 Cultural Interaction in the Prehistoric Southwest. In Dyrzanzics of Southwest Prehistory, edited by L. S. Cordell and G. J. Gumerman, pp. 1-18. Smithsonian Institution Press, Washington, DC. Crown, P. L. 1987 Classic Period Hohokam Settlement and Land Use in the Casa Grande Ruins Area, Arizona. Journal of Field Archaeology 14: 147-162. 1991 The Hohokam: Current Views of Prehistory and the Regional System. In Chaco & Hoiiokam: Prehistoric Regional Systems in the American Southwest, edited by P.
L. Crown and W. J. Judge, pp. 135-157. School of Ameri- can Research Press, Santa Fe, New Mexico. Czaplicki, J. S., and J. C. Ravesloot (editors)
1989 Hohokatn Archaeology Along Phase B of the Tucsoiz Aqueduct, CeiztralArizoiza Project, Volume I: Syrztlieses aizd Iizterpretatioizs. Archaeological Series No. 178. Arizona State Museum, University of Arizona, Tucson.
Dean, J. S.
1991 Thoughts on Hohokam Chronology. In Exploring tiie Holiokam: Prehistoric Desert Peoples of the American Southwest, edited by G. J. Gumerman, pp. 61-149. Univer- sity of New Mexico Press, Albuquerque.
2000 Salado. The University of New Mexico Press, Albu- querque. Dobyns, H. E
1981 From Fire to Flood: Historic Human Destructioiz of Soizoran Desert Riveriize Oases. Ballena Press Anthro- pological Papers No. 20. BallenaPress, Socorro, New Mex- ico.
Doelle, W. H., and H. D. Wallace
1991 The Changing Role of theTucsonBasinin the Hohokam Regional System. In Exploring the Hohokam: Prehistoric Desert Peoples of the American Southwest, edited by G. J. Gumerman, pp. 279-345. University of New Mexico Press, Albuquerque.
Doyel, D. E.
1980 Hohokam Social Organization and the Sedentary to Classic Transition. In Curlrizt Issues in Hohokam Prehis- tory: Proceedings of a Symposium, edited by D. E.Doyel and F. T. Plog, pp. 2340. Anthropological Research Papers No. 23. Arizona State University, Tempe.
1991 Hohokam Cultural Evolution in the Phoenix Basin. In
Exploring the Hohokam: Prehistoric Desert Peoples of the American Southwest, edited by G. J. Gumerman, pp. 23 1-278. University of New Mexico Press, Albuquerque.
1993 Interpreting Prehistoric Cultural Diversity in the Arizona Desert. In Culture and Contact: Charles E. Di Peso's Gran Cizichinzeca, edited by A.I. Woosley and J. C. Ravesloot, pp. 3944. University of New Mexico Press, Albuquerque.
Elson, M. D., and D. A. Gregory
1995 Tonto Basin Chronology and Phase Sequence. In The Roosevelt Comnzunity Developnzent Study, edited by M. D. Elson, M. T. Stark, and D. A. Gregory, pp. 61-77. Anthro- pological Papers No. 15, Center for Desert Archaeology, Tucson, Arizona.
Elson, M. D., D. A. Gregory, and M. T. Stark
1995 New perspectives on Tonto Basin Prehistory. In The Roosevelt Community Development Study, edited by M. D. Elson, M. T. Stark, and D. A. Gregory,pp. 441479. Anthro- pological Papers No. 15, Center for Desert Archaeology, Tucson, Arizona.
Ely, L. L. 1997 Response of Extreme Floods in the Southwestern United States to Climatic Variations in the Late Holocene. Geotnorphology 19: 175-201.
Fish, P. R. 1968 GilaDunes: A Chiricahua Stage SiteNearFlorence,Ari- zona. Ms. On file, Department of Anthropology, Arizona State University, Tempe. 1989 The Hohokam: 1,000Yeass of Prehistory in the Sono- ran Desert. In Dynamics of Southwest Prehistory, edited by
L. S. Cordell and G. J. Gumerman, pp. 1943. Smithsonian Institution Press, Washington, D.C. Fish, P. R., S. K. Fish, G. J. Gumerman, and J. J. Reid
1994 Towards an Explanation for Southwestern "Abandon- ment~.'' In Themes in Southwest Prehistory, edited by G. J. Gumerman, pp. 135-163. School of American Research Advanced Seminar Series, Santa Fe.
Fish, S. K., P. R. Fish, C. H. Miksicek, and J. H. Madsen 1985 Prehistoric Agave Cultivation in Southern Arizona. Desert Plants 7(2):107-112. Fish, S. K., P. R. Fish, and J. H. Madsen (editors)
1992 The Maluna Community iiz tiie Hoiiokanl World. Anthropological Papers No. 56. University of Arizona Press, Tuc- son.
Freeman, A. K. L.
1997 Middle to Late Holocene Stream Dynamics of the Santa Cruz River, Tucson, Arizona: Implications for Human Set- tlement, theTransition to Agriculture andArchaeologica1 Site Preservation. Unpublished Ph.D. dissertation, University of Arizona, Tucson.
Graybill, D. A.
1989 The Reconstruction of Prehistoric Salt River Stream- flow. In The 1982-1984 Excavatioizs at Las Colinas: Envi- ronment and Subsisteizce, edited by D. G. Graybill, D. A. Gregory,E L. Nials, S. K. Fish, C. H. M~ksicek, R. E. Gasser, and C. R. Szuter, pp. 25-38. Archaeological Series No. 162, Volume 5. Arizona State Museum, University of Arizona, Tucson.
Gregory, D. A.
1987 The Morphology of Platform Mounds and the Stmc- ture of Classic PeriodHohokam Sites. InThe Hohokatn Village: Site Structure and Orgarzizatioiz, edited by D.E.Doyel, pp. 183-210.AmericanAssociation for the Advancement of Science, Glenwood Springs, Colorado.
1991 Form and Variation in Hohokam Settlement Patterns. In Ciiaco and Hoiiokam: Prehistoric Regioizal Systems in the American Southwest, edited by P. L. Crown &W. J. Judge, pp. 159-193. School of American Research Press, Santa Fe, New Mexico.
Gregory, D. A,, and G. Hucklebeny
1994 An Arcizaeological Suivq of the BlackwaterArea, 1701- ume I: The History of Human Settlement in the Blackwater Area. Cultural Resources Report No. 86. Archaeological Consulting Services, Ltd., Tempe.
Gumerman, G. J.
1991 Understanding the Hohokam. In Exploring the Hohokam: Prel7istoric Desert Peoples of the American Southwest, edited by G. J. Gumerman, pp. 1-27. University of New Mexico Press, Albuquerque.
Hackbarth, M. R.
1995 Integration of the MTS Data with the Late Classic Hohokam. In Arcl7aeology at the Head of the Scottsdale Canal Systenz. lblunze 3:Canal and Syntl7etic Studies,edited by M. R. Hackbarth, T. K. Henderson, and D. B. Craig, pp. 173-195. Anthropological Papers No. 95-1. Northland Research, Inc.
Haury, E. W. 1976 The Hoizokanz: Desert Farnzers and Craftsmen. University of Arizona Press, Tucson.
AMERICAN ANTIQUITY [Vol. 66, No. 2,20011
Haynes, C. V. 1987 Curry Draw, Cochise County, Arizona: A Late Quater- nary Stratigraphic Record of Pleistocene Extinction and Paleo-Indian Activities. In Geological Socieiy of America Centennial Field Guide-Codilleran Section, edited by M.
L. Hill, pp. 23-28. Geological Society of America, Boulder, Colorado. Haynes, C. V., and B. B. Huckell
1986 Sedimentary Successions of the Prehistoric Santa Cruz River, Tucson, Arizona. Open file report, Arizona Bureau of Mines and Geology, University of Arizona, Tucson.
Hoover, J. W. 1929 The Indian Country of Southern Arizona. Geographical Review 19:3840. Howard, J. B.
1985 Courtyard Groups and Domestic Cycling: A Hypothet- ical Model of Growth. In Proceedings of the 1983 Hohokam Symposium, edited by A. E. Dittel-t, Jr, and D. E. Dove, pp. 31 1-326. Occasional Paper No. 2, Part 1. Arizona Archaeological Society, Phoenix.
1987 The Lehi Canal System: Organization of a Classic Period Community. In The Hohokam L'illage: Site Structure and Organization, edited by D. E. Doyel, pp. 211-221. American Association for the Advancement of Science, Glenwood Springs, Colorado.
1993 A ~aleoh~draulic
Approach to Examining Agricultural Intensification in Hohokam Irrigation Systems. In Research in EconomicAntl7ropology: ~cononzic~~pects
ofWater Man- agement in the Prehispanic New World, supplement 7, edited by V. L. Scarborough and B. L. Isaac, pp. 263-324. JAI Press, Greenwich, Connecticut.
Huckleberry, G. A.
1993 Late Holocene Stream Dynamics on the Middle Gila River, Pinal County, Arizona. Unpublished Ph.D. disserta- tion, University of Arizona, Tucson.
1994 Contrasting Channel Response to Floods on the Mid- dle Gila River, Arizona. Geology 22:1083-1086.
1995 Archaeological Implications of Late Holocene Chan- nel Changes on the Middle Gila River, Arizona. Geoarclzaeology: An International Journal 10: 159-182.
1999 Prehistoric Flooding and Its Effect on Indigenous Agri- culture in the Northern Sonoran Desert, U.S.A. Journal of Arid Land Studies 9:277-284.
Huckell, B. B.
1990 Late Preceramic Farmer-Foragers in Southeastern Ari- zona: A Cultural andEcologica1 Consideration of the Spread of Agriculture into the Arid Southwestern United States. Unpublished Ph.D. dissertation, University of Arizona, Tuc- son.
Kelly, I. T.
1978 The Hodges Ruin: A Holzokam Conzmuniry in the Tuc- son Basin. Anthropological Papers No. 30. University of Arizona, Tucson.
Lee, W. T.
1904 The Underground Waters of Gila lblley,Arizona. Water-Supply and Irrigation Paper no. 104. Series 0, Underground Waters, 25. United States Geological Survey, Department of the Interior, Washington, DC.
Masse, W. B.
1991 The Quest for Subsistence Sufficiency and Civilization in the Sonoran Desert. In Chaco & Hohokam: Prehistoric Regional Systems in the American Southwest,edited by P.
L. Crown and W. J. Judge, pp. 195-223. School of Ameri- can Research Press, Santa Fe, New Mexico.
Nials, E L., D. A. Gregory, and D. A. Graybill 1989 SaltRiver Streamflow and Hohokamhgation Systems.
In The 1982-1 984 Excavations at Las Colinas: Environment
and Subsistence, edited by D. G. Graybill, D. A. Gregory, F.
L. Nials, S. K. Fish, C. H. Miksicek, R. E. Gasser, and C. R. Szuter,pp. 59-76. Archaeological Series 162,Volume 5.Ari- zona State Museum, University of Arizona, Tucson.
Ravesloot, J., and A. Lascaux (editors) 1993 The Maricopa Road Site: A Pre-Classic Hohokam L'il- luge. Anthropological Field Studies No. 28. Office of Cul- tural Resource Management, Department of Anthropology, Arizona State University, Tempe.
Rea, A. M. 1983 Once a River: Birdlife andHabitat Changes on the Mid- dle Gila. University of Arizona Press, Tucson. 1997 At the Deserts Green Edge: An Ethnobotany of the Gila River Pinza. The University of Arizona Press, Tucson. Rice, G. E. 1998 War and Water: An Ecological Perspective on Hohokam Irrigation. Kiva63:263-301. Rice, G. E. (editor) 1998 A Synthesis of Tonto Basin Prehistory: The Roosevelt Archaeology Studies, 1989 to 1998. Roosevelt Monograph Series 12, Anthropological Field Studies No. 41. Office of Cultural Resource Management, Department of Anthropol- ogy, Arizona State University, Tempe. Schumrn, S. A. 1977 The Fluvial System. John Wiley and Sons, New York. Sires, E. W., Jr., 1983 Excavations at El Polvoron. In Hohokanz Archaeology Along the Salt-Gila Aqueduct, Central Arizona Project, Lbl- ume 4: Prehistoric Occupation of the Queen Creek Delta, edited by L. S. Teague and P. L. Crown, pp. 221-294. Archae- ological Series No. 150. Arizona State Museum, University of Arizona, Tucson. Spoerl, P., and G. Gumerman 1984 Prehistoric Cultural Development in Central Arizona: Archaeology of the Upper New River Region. Center for Archaeological Investigations Occasional Paper 5. Southern Illinois University, Carbondale. Stuiver, M., and P. Reimer 1993 Extended 14C Database and Revised CALIB Radio- carbon Calibration Program. Radiocarbon 35:51&520. Teague, L. ,and P. M. Crown (editors) 1984 Hohokam Archaeology Along the Salt-Gila Aqueduct, Central Arizona Project, lblume 9: Synthesis and Conclu- sions. Archaeological Series No. 150. Arizona State Museum, University of Arizona, Tucson. Thomsen, B. W., and J. H. Eychaner 1991 Predevelopnzent Hydrology of the Gila River Indian Reservation, South-Central Arizona. Waters-Resources Investigations Repol-t 894174. U.S. Geological Survey, Tucson. Wallace, H. D. !995 Arclzaeological Investigations at Los Morteros: A Pre- historic Settlement in the Northern Tucson Basin. Anthropological Papers No. 17. Center for Desert Archaeology, Tucson. 1996 Settlement Reconstructions and Population Dynanzics in the Northern Tucson Basin, Arizona: New Perspectives from Los Morteros. Paper presented at the 61St.4nnual meet- ing of the Society for American Archaeology, New Orleans. Wallace, H. D., and W. H. Doelle
1997 From Ballcourts to Platform Mounds to Ranclzerias: A Comparison of Three Organizational Strategies on the Lower Sun Pedro River. Paper Presented at the 62nd Annual Meet- ing of the Society for American Archaeology, Nashville, Tennessee.
2001 Classic Period Warfare in Southern Arizona. In Deadly Landscape: Case Studies in Prehistoric Southwestern War- fare, edited by G. Rice and S. LeBlanc. University of Utah Press, Salt Lake City.
Wallace, H. D., J. M. Heidke, and W. H. Doelle 1995 Hohokam Origins. Kiva 60:575-618. Waters, M. R.
1988 Holocene Alluvial Geology and Geoarchaeology of the San Xavier Reach of the Santa Cruz kver, Arizona. Geological Society ofAmerica Bulletin 100:479-491.
1992 Principles of Geoarchaeology. University of Arizona Press, Tucson.
1998 The Effect of Landscape and Hydrologic Variables on the Prehistoric Salado: Geoarchaeological Investigations in the Tonto Basin, Arizona. Geoarchaeology: an International Journal 13:105-160.
Waters, M. R., and J. C. Ravesloot
2000 Late Quaternary Geology of the Middle Gila River, Gila kver Indian Reservation, Arizona. Quaternary Research 54:49-57.
Weaver, D. E.,JI: 1972 A Cultural-Ecological Model for the Classic Hohokam Period in the Lower Salt River Valley. Kiva 38:43-52. Wilcox, D. R.
1979 The Hohokam Regional System. In An Arclzaeolonical Test ofthe Sites in the ~2la ~utte-Suntan Region, south-?en- tral Anzona, edited by G. Rice, D. Wilcox, K.Rafferty, and
J. Schoenwetter, pp. 77-116. Anthropological Research Papers No. 18. Arizona State University, Tempe.
1991 The Mesoamerican Ballgame in the American South- west. In The MesoamericanBallgame, edited by V. L. Scar- borough andD. R. Wilcox, pp. 101-28. University ofArizona Press, Tucson.
Wilcox, D. R. ,T. R. McGuiIe, and C. Stemberg
1981 Snaketown Revisited: A Partial Cultural Resource Sur- vey,Analysis of Site Structure, and an Etlznolzistoric Study of the Proposed Hohokanz-Pima National Monument.
Archaeological Series No. 155. Arizona State Museum, Uni- versity of Arizona, Tucson. Wilcox, D. R., and C. Stemberg
1983 Hohokam Ballcourts and Their Interpretation. Archaeological Series No. 160. Arizona State Museum, University of Arizona, Tucson.
Wilson, J. P.
1999 Peoples of the Middle Gila: A Documentary History of the Pimas and Maricopas, 1500s-1945. Manuscript on file with the Office of Cultural Resources, Department of Land and Water, Gila River Indian Community, Sacaton, Arizona.
Received February 8,2000; Revised September 13,2000;
Accepted September 20,2000