EARLY AGRICULTURAL PERIOD SETTLEMENTSTRATEGIES IN THE SOUTHERN SOUTHWEST

David A. Gregory, Desert Archaeology, Inc.Fred L. Nials, Desert Archaeology, Inc.J. Brett Hill, Center for Desert Archaeology

ABSTRACT

In this chapter, Gregory, Nials, and Hill apply Geographic Information Sys-tem technologies to geomorphological, topographic, and archaeological data toexamine the settlement strategies of the Southwest’s first farmers between 2100B.C. and A.D. 200. By applying the concept of naturally defined stream reaches,they identify reach boundaries as places with floodplain and hydrological charac-teristics that would have been particularly desirable for farmers. Their examina-tion of settlement data from Early Agricultural sites in the Tucson Basin and south-east Arizona, and of data from more visible late prehistoric sites, shows preferentialsettlement of reach boundaries by farming communities throughout prehistory.Settlement data also suggests that Early Agricultural people had a secondary fo-cus on upper piedmont slopes, perhaps in areas with productive springs and di-verse wild plant resources. Furthermore, the stream reach method for examina-tion of land use strategies is one that can be applied in arid lands settings globally.

The settlement strategies of the Early Agricul-tural period, that interval between the arrival ofmaize and the appearance of a fully developed ce-ramic container technology in the Southwest (circa2100 B.C. and A.D. 200), are reviewed here. The taskis daunting for a variety of reasons, not the least ofwhich is the tremendous increase in data and newrevelations about Early Agricultural period lifewaysthat have occurred over the last two decades.

The current, earliest dates for the arrival of maizehave been continually pushed back in time to ap-proximately 2100 B.C. (Figure 1; the data on whichthis figure are based are likely already incomplete),making the Early Agricultural period considerablylonger than the subsequent and more completelyunderstood Early Ceramic period (A.D. 50-500). Ru-dimentary irrigation was practiced by about 1500B.C, and large floodplain settlements consisting ofhundreds of pit structures and/or other features arecharacteristic of that interval in some areas. Finally,a remarkable uniformity in subsistence strategiesover much of the interval is indicated by consistentlyhigh ubiquities in macrobotanical assemblages of notonly maize but several wild plant species, as well asabundant and varied faunal assemblages in manycases.

Several interpretive frameworks have been pro-posed regarding the still expanding body of EarlyAgricultural period data. New theoretical develop-

ments have been focused on the complex set of prob-lems relating to the spread of plant and animal do-mestication (and languages) on a global scale, andthe domestication and spread of maize (and lan-guages) in the New World, and the Southwest spe-cifically (Bellwood 2005; Bellwood and Renfrew2002; Benz 1999, 2006; Doolittle and Mabry 2006;Hard et al. 2006; Harris 1996; Hill 2001, 2002; Huckell1995; Mabry 2005a; Minnis 1992; Renfrew 1998;Staller et al. 2006; Wills 1988, 1995). Models basedon necessity, risk management, and opportunity oroptimization, as well as interpretations emphasiz-ing the role of migration, warfare, the biogeogra-phy of transmission, and environmental variationhave all been offered (Carpenter et al. 2002; Diehl2005a, 2005b, 2005c; Doolittle and Mabry 2006; Gre-gory and Nials 2005; Huckell et al. 2002; Hunter-Anderson 1986; LeBlanc 2002; Mabry 2002, 2005b;Matson 2002). Empirically and theoretically, thereis better information than ever before about this criti-cal and lengthy interval in Southwest prehistory.

Nonetheless, all interpretive frameworks sufferto varying degrees from several basic problems. Thefirst of these is the paradoxically happy circumstancenoted above: the rate of data accumulation has re-quired constant reevaluation of interpretations toaccommodate a rapidly expanding database. Thisis particularly true with respect to earlier and morewidely distributed dates on maize, which demand

2 The Latest Research on the Earliest Farmers

Figure 1. Direct dates on maize. (Data from Clark 2000; Damp et al. 2002; Ezzo and Deaver 1998; Ford 1981;Freeman 1998; Geib 1996; Geib and Davidson 1994; Geib and Huckell 1994; Gilpin 1994; Gregory 1996, 1999,2001; Gregory, ed. 2001; Gregory and Baar 1999; Gregory et al. 2007; Hard and Roney 1998; Huber and Miljour2004; Huber and Van West 2006; Huckell 1990; Huckell and Huckell 1984; Huckell, et al. 1995; Kearns 1996;Mabry 2006; Mabry et al. 1997; Moreno 2000; Parry et al. 1994; Roth 1992; Shackley 2005; Simmons 1986; Smiley1994; Tagg 1996; Upham et al. 1987; Whittlesey et al. 2007; Wills 1988.)

Early Agriculture Period Settlement Strategies in the Southern Southwest 3

reassessment of the temporal dimension of particu-lar interpretations. Second, the current spatial distri-bution of relevant data represents a non-systematicsample of sites produced largely by cultural resourcemanagement research and geographically spottyexcavations in rockshelters and caves. Unfortunately,many of the latter were excavated before the adventof AMS dating (or even before conventional radio-carbon dating), and before flotation sampling becamestandard practice. Finally, many interpretive frame-works have been developed primarily based on in-vestigations at a single site or a closely spaced set ofsites, and original research specifically designed toaddress these issues at a macro-scale has been largelylacking.

THE ENVIRONMENT AS A CONTEXTFOR SETTLEMENT STRATEGIES

A variety of factors may influence the set of de-cisions that culminate in settlement strategies. Per-haps the most important of these for arid lands farm-ers is the availability and reliability of water incombination with arable land. However, such fac-tors may also include the distribution and densityof natural plant and animal resources, technology,population size, and social relations, including war-fare. This list could certainly be expanded, but theconcentration here is on water and arable land.

Dean (1988) defines stable elements of the envi-ronment as: “climate type, topography, bedrock ge-ology, elevational zonation of plant communities,and the distribution of mineral resources and rawmaterials” and suggests that “their present states arevalid indicators of past conditions, and they neednot be reconstructed” (Dean 1988:121). While recon-struction is not necessary, the spatial parameters ofthese environmental features must at least be iden-tified and measured for purposes of settlementanalysis.

The general frame of reference here for quantifi-cation is a division of the landscape into floodplains,piedmont slopes, and bedrock uplands (Gregory andNials 2005). These divisions apply directly to all ofthe southern Basin and Range, and, with appropri-ate modification, to the Transition Zone and theColorado Plateau as well. The focus here is on flood-plains, particularly on stream reaches and reachboundaries. These aspects of the floodplain environ-ment have been specifically identified and measuredfor large areas of the southern Southwest (Nials etal. 2007, 2011). The three major environmental divi-sions for the Tucson Basin, as well as the distribu-tion of reach boundaries and springs, are illustratedin Figure 2.

The Stream Reach Concept

All arid lands streams are punctuated by loca-tions where specific geological structures, hydrologiccharacteristics, and geomorphic processes combineto alter the nature of sediment load, characteristicsurface flow, and/or groundwater availability. Theinfluence of these variables has long been recognizedand is well understood (Hendrickson and Minckley1984; Hinderlider 1913; Lee 1905; Leopold et al. 1964;Meinzer 1942). Such locations are of two principaltypes: (1) those created by the presence of bedrockor relatively impermeable sediments in, beneath, oradjacent to the stream channel; and, (2) those cre-ated by tributary confluences. These locations arereferred to as reach boundaries, and stream segmentsbetween such boundaries are defined as streamreaches, with the downstream boundary includedas part of each stream reach for purposes of analy-sis (Gregory and Nials 2005; Nials et al. 2007).

Figure 2. Environmental divisions in the Tucson Basin.

4 The Latest Research on the Earliest Farmers

Even casual perusal of topographic maps oraerial photographs reveals obvious indicators ofreach boundaries such as (1) where a floodplain lo-cally narrows or widens; (2) marked changes thatoccur in channel pattern and/or sinuosity; (3) ero-sion or deposition locally prevails; (4) local areas ofhigh water tables are indicated by vegetation orsprings; or, (5) streams become emergent or submer-gent. Thus, reach boundary locations and associatedreaches may be readily identified using a variety ofgeologic and geomorphic criteria recognizable byremote means, primarily through examination oftopographic and geologic maps, aerial photography,and satellite imagery. After reach boundaries areidentified for any given stream, variables, includ-ing reach length, floodplain area, floodplain area perunit length, and floodplain gradient may be obtainedvia measurements from topographic maps. Addi-tional quantification of reach properties is gainedthrough the use of GIS analyses of digital elevation

data and remote sensing imagery, including drain-age area, number of tributaries, total flow of tribu-taries, average slope of drainage area, and so on.More than 2,500 linear kilometers of floodplainshave been mapped and analyzed to date; some 300previously identified and measured reaches withina large study area are shown in Figure 3. Measure-ment of remaining study area reaches is ongoing.

Effects of Reach Boundaries

Reach boundaries resulting from outcrops of bed-rock or impermeable sediments create a barrier togroundwater flow, causing the water table to riseupstream from the barrier, and influence channelerosion (Figure 4) (Hinderlider 1913; Lee 1905;Leopold et al. 1964; Meinzer 1942). If the rise is suf-ficient, cienegas and springs form (Hendrickson andMinckley 1984), and the aquifer discharges into thesurface channel, resulting in either a new segment

Figure 3. Identified and measured stream reaches and reach boundaries in the study area.

Early Agriculture Period Settlement Strategies in the Southern Southwest 5

Figure 4. Idealized stream reach, showing reach boundaries and boundary effects.

of emergent stream or increased perennial flow. Theconsequent flow may be continuously emergentalong the entire length of a reach, or it may eventu-ally sink into valley-bottom sediments and the aqui-fer below. High water tables result in increased veg-etation that enhances sediment deposition and leadsto a wider, lower-gradient floodplain that allowsflood waters to spread over a larger area and reducefloodwater depth and velocity. A short, oversteep-ened segment typically occurs immediately belowthe boundary, characterized by straighter, deeper,and more constricted channels. Excess sedimentsmay be deposited downstream from the steeper seg-ment, and the channel may shift positions more fre-quently (Graf 1982, 1983b, 1983c, 1988), causing theactive floodplain to be wider below the boundary.

Tributaries have steeper gradients than main-stem streams and transport proportionally largeramounts of coarser sediment, much of which is ulti-mately deposited on the mainstem floodplain at andnear their confluence. These deposits locally elevateand narrow the floodplain surface on the masterstream, resulting in a flatter gradient above and anoversteepened segment immediately below the con-fluence. Groundwater contributions from largetributaries may also cause the water table in themainstem alluvial aquifer to rise. The combination

of additional groundwater and increased depositionof sediment on the mainstem floodplain at tributaryconfluences thus duplicates the effects of bedrockboundaries elsewhere. During the summer monthsin the Southwest, tributary floods are not necessar-ily coincident with mainstem floods, often leadingto a prolonged runoff period below the confluence.

The effects exerted on groundwater and surfacestreams by reach boundaries depend, in part, onvariables specific to a particular location and to adrainage basin. These include the type of boundary,nature and amount of runoff in the mainstem andtributaries, nature of valley bottom sediments, depthto water table, aquifer geometry, and rainfall distri-bution, among others. The particular combinationof these variables at any given boundary may pro-duce dramatically different conditions of surfacerunoff availability and water table depth.

Temporal Variation in Boundary Effects

Regional environmental variation is primarilyrelated to climate, and floodplain conditions withina region at any given time are best understood interms of repeated, climate-driven cycles of alluvialerosion and deposition. Although controversy over

Water table

6 The Latest Research on the Earliest Farmers

specific triggering mechanisms for cycles remains(Bull 1991, 1997; Cooke and Reeves 1976; Graf 1982,1983a, 1988; Gregory and Nials 2005; Leopold et al.1966; Martin 1964; Sayles and Antevs 1941; Watersand Haynes 2001; Winn 1926), each may be concep-tualized as consisting of two overlapping phases:(1) a relatively short period of erosion and valleybottom degradation leading to incision of arroyos,ultimately followed by subsequent filling and heal-ing of arroyos; and, (2) a more prolonged period ofrelative floodplain stability and soil formation (Bull1997; Cooke and Reeves 1976; Waters and Haynes2001). Environmental conditions during each phaseof a cycle are summarized in Table 1.

Past alluvial cycles have been roughly synchro-nous across much of the Southwest, and stabilityconditions have dominated late Holocene valleybottom environments for most of the past 6,000 years(Bryan 1925, 1941; Eddy and Cooley 1983; Freeman2000; Gregory and Nials 2005; Haynes 1968, 1987;Haynes and Huckell 1986; Waters 1988; Waters andHaynes 2001). Local geomorphic factors may alsoinitiate cycles, however, and individual phases donot necessarily occur simultaneously in differentdrainages, or even in different reaches within asingle stream (Gregory and Nials 2005; Schumm andHadley 1957; Waters and Haynes 2001). Five suchcycles have been documented for the Tucson Basin,and it is of particular interest that the Early Agricul-tural period is largely isomorphic with Cycle 3 inthis basin and probably elsewhere (see Gregory andNials 2005; Gregory et al. 2007:52-53).

It has been suggested that the spread of maizeagriculture into and across the Southwest was stimu-lated by stability phase conditions in valley bottomsduring Cycle 3 (Mabry 1998). More recently obtainedradiocarbon dates and stratigraphic evidence fromsites in the Tucson Basin, however, show that maizefarming was being conducted along the Santa Cruzfloodplain during the initial erosion-depositionphase of Cycle 3 (Figure 5), at a time when recentlydeeper arroyos were only partially filled, and fre-quent floods still swept a rapidly aggrading flood-plain (Gregory and Nials 2005; Gregory et al.2007:52-53; Mabry 2006; Nials 2008; Whittlesey etal. 2007).

Advantages of the Reach Concept

In addition to ease of identification and measure-ment, the stream reach concept offers several otheradvantages. The concept is universally applicable inall arid lands environments. Stream reaches andreach boundaries represent basic structural elementsin the landscape, and may be considered as stableor constant aspects of the environment for at least

the late Holocene (see Dean 1988), and thus, not onlyfor the period of concern but for earlier and laterintervals as well. For a given discharge regime andlocal variables, the nature of change associated withparticular types of boundaries is predictable in termsof consequences for geomorphic processes andstream character. This fact allows modeling of reachcharacteristics for different climate patterns, and itprovides a basis for assessing agricultural potentialand the suitability of irrigation and other agricul-tural technologies under variable conditions.

Reach boundaries also have predictable effectson floodplain environments vís a vís aboriginal ag-riculture. Areas immediately upstream and down-stream from boundaries typically represent the mostrisk-free settings for farming: emergent stream seg-ments in such locations may be the only places alonga drainage where surface water is present in timesof drought or low flow, where runoff is concentrated,and/or where the water table is closest to the sur-face (see Doolittle and Mabry 2006; Evenari et al.1971; Mabry 2002; Nabhan 1979; Nials et al. 2011).These conditions, in conjunction with engineeringadvantages created by oversteepened downstreamsegments, make reach boundaries some of the bestplaces to construct intake structures for canal irri-gation systems (Gregory and Nials 1985; Nials et al.2007, 2011). Closely spaced reach boundaries createparticularly attractive valley bottom conditions, be-cause these features favor continuous perennialstreamflow, foster subirrigation of relatively largeareas, and allow irrigation of the maximum amountof floodplain via short, relatively efficient canals.

The significance of reach boundaries, in particu-lar, the importance of emergent stream segments toirrigation agriculture, was identified long ago in theGreater Southwest (Hinderlider 1913; Lee 1905;Southworth 1919). Subsequent investigators havenoted these connections and have described theirrelationships to prehistoric agriculture in specific lo-cations (Gregory and Huckleberry 1994; Gregoryand Nials 1985, 2005; Nials 2008; Nials and Gregory1989; Waters 1988). With respect to current concerns,it is clear that reach boundaries and their character-istic effects were recognized and were consistentlyexploited by Early Agricultural period populations.

A Case Study

A recent study demonstrates the general utilityof the reach concept beyond the anecdotal settingsthat inspired it (Nials et al. 2007, 2011). Analysesdrawn from that study illustrate the broader signifi-cance of the concept. Local density analysis (LDA)(Johnson 1984; see Kintigh 1992) was used to evalu-ate the spatial correlation among reach boundaries

Early Agriculture Period Settlement Strategies in the Southern Southwest 7

and archaeological sites. The goal of LDA is to evalu-ate the degree of spatial association among points.The local density coefficient (LDC) is the mean den-sity of points of a given type found within a speci-fied radius of a second point type, divided by theglobal density of the first type of points. The den-sity of points is calculated as the number of pointsdivided by: (1) the area of the circle defined by aspecified radius; and, (2) the area being analyzed.The LDC provides a measure of the association ofpoint types at a designated scale fixed by the radiusused, with values around 1.0 or less indicating ran-dom distribution and values greater than 1.0 indi-cating relative spatial association.

This statistic provides an easily interpreted indi-cation of the degree to which prehistoric land usewas actually focused in the vicinity of reach bound-aries, as opposed to any other location on the land-scape. Further, it does so at a behaviorally mean-ingful scale indicated by the specified radius. A keyto understanding the significance of reach bound-aries is in relation to other locations where agricul-

ture might be practiced. Thus, the LDC was com-pared for sites and reach boundaries with LDCs forsites and two randomly distributed sets of the samenumber of points. One set was drawn from the uni-verse of all points with a larger than 50 ha water-shed, and one was from the universe of all pointshaving a larger than 50 km² watershed.

The 50-ha figure corresponds conservatively withboth ethnographic (see Bradfield 1971; Hack 1942)and archaeological (see Sandor 1992) indications ofthe watershed required for upland runoff agricul-ture. The 50-km² figure represents locations with asubstantial stream course that might be suitable forirrigation, and it is similar to the watershed used byPremo (2001) in his predictive model of Early Agri-cultural sites in the Tucson Basin. Analyses were con-ducted at radii of 1, 5, and 10 km to assess spatialassociation within distances that would be mean-ingful to farmers tending fields at varying distancesfrom a given settlement, allowing for a comparisonof the degree of association at increasing distancefrom a critical resource.

Table 1. Comparison of valley bottom processes and characteristics during erosional and stabilization phases of

alluvial cycles.

Erosional Phase Stability Phase

Flood characteristics Frequent, large peak discharges, increased

flood depths (stage), reduced flood durations

as arroyo networks integrate, progressively

smaller amounts of floodplain inundation;

high erosive power discharges; minimum

effectiveness of precipitation.

Flood frequency, peak discharges, stages, and

energy decreased as arroyos fill; amount of

floodplain inundation increases, along with

flood effectiveness of precipitation, duration;

relatively low erosive power.

Stream channel Initial stages: discontinuous channel (may

have braided segments), vertical erosion

dominates, continuous arroyo formed. Later

stages: lateral erosion dominates, expanding

arroyo gradually consumes former floodplain.

Initial stages: continuous, deep, single channel,

progressive deposition and infilling of channel

dominates. Later stages: arroyo is filled,

deposition is concentrated on floodplain.

Infiltration Minimal. Maximum.

Sediment load Progressively larger and coarser sediment

load until near end of phase.

Sediment load becomes finer, reduced in

quantity.

Floodplain processes Degradation, floodplain desiccates, secondary

arroyos may form, piping may be locally

important in some drainages.

Aggradation, floodplain becomes wetter,

arroyos heal, piping not important; floodplain

level may rise above preerosion level in some

valleys, not in others; geomorphic floodplain

forms in some valleys; deposition of relatively

fine-grained sediments predominates except

locally; soil formation occurs.

Water table Progressively lowers as channel is incised;

most existing cienegas destroyed.

Progressively rises as channel is filled; cienega

frequency at a maximum as cienegas reform.

Surface flow

(baseflow)

May initially increase as groundwater drains,

but eventually becomes minimal; baseflow

reduced, some perennial streams become

discontinuous.

Surface discharge at a maximum, some

streams become perennial, discontinuous

perennial streams may become continuous.

Vegetation Significantly decreased in quantity, more

xeric in nature.

Significantly increased in quantity;

phreatophytes; riparian communites, mesquite

bosques, and sacaton flats all may form locally.

8 The Latest Research on the Earliest Farmers

Site data consist of two sets of points, those dat-ing to the Early Agricultural period and those dat-ing to late prehistoric times. Early Agricultural (n =89) sites in the study area are from a database of allknown Paleoindian and Archaic sites in Arizona(Mabry 1998; Mabry and Stevens 2000). This sampledoes not include contiguous areas in New Mexico,Sonora, or Chihuahua. The late prehistoric data (n =275) sites in the study area are from the CoalescentCommunities database (Wilcox et al. 2003), and in-clude all known sites dating from A.D. 1200-1700,having more than 12 rooms. Within the Early Agri-cultural period data set, there is a bias toward thewell-investigated Tucson Basin. For this reason, spa-tial analyses outside the Tucson Basin will not fullyreveal the true relationship among Early Agricul-tural sites and reach boundaries. The late prehistoricdata set probably represents a more complete recordof settlement distribution for this time period in theTucson Basin and beyond. Five analyses were per-formed using various subsets of the site data: Tuc-son Basin Early Agricultural sites (n = 34), south-eastern Arizona Early Agricultural sites (n = 89),Tucson Basin late prehistoric sites (n = 32), south-eastern Arizona late prehistoric sites (n = 200), andall study area late prehistoric sites (n = 275).

The overwhelming indication from these analy-ses is that reach boundaries have a higher density

of archaeological sites within close proximity thando either set of random drainage locations. In con-trast to the mean coefficient for reach boundaries of3.84, the mean coefficients for 50-ha and 50-km² wa-tersheds are 1.33 and 1.94, respectively. Thus, sitesgenerally appear to be focused around locations withsubstantial watersheds, but they show the strongestspatial association with reach boundaries. These dif-ferences are shown clearly in Figure 6, which illus-trates coefficients for the five site groups obtainedat the 1-km radius.

The density of archaeological sites within 1 kmof reach boundaries is greater at every level of analy-sis than their density around other watershed loca-tions. Notably, the weakest differences are foundwith the Early Agricultural data outside the TucsonBasin. It remains to be demonstrated if this is due todifferences in Early Agricultural land use in neigh-boring valleys or to incomplete knowledge of bur-ied deposits or other sampling biases. The fact thatbetter-known late prehistoric sites maintain theirstrong associations with reach boundaries at all lev-els of analysis suggests the potential for similarlystrong associations with undocumented Early Agri-cultural sites and those of other intervals as well. Thestatistical trends demonstrated by these analysesstrongly support the validity of the stream reach con-cept as a powerful structuring influence on ancient

Figure 5. Comparative stratigraphic sequences from several areas of the Southwest, showing similarities and differ-ences in the timing of erosional and depositional phases of alluvial cycles.

Erosion Deposition

SANTA CRUZ

RIVER

(at Los Pozos)

Gregory and Nials

(2005)

SANTA CRUZ

RIVERWaters and Haynes

(2001)

SANTA PEDRO

VALLEY

(Curry Draw)

Waters and Haynes

(2001)

SOUTHERN

COLORADO

PLATEAUHack (1942)

BLACK MESAKarlstrom (2005)

RADIO-

CARBON

YEARS(Years B.P.)

0

1000

2000

3000

4000

5000

Maize present

in the southern

Southwest

Cycle

2C

ycle

3C

ycle

4

Early Agriculture Period Settlement Strategies in the Southern Southwest 9

land use in the southern Southwest, and as persis-tent places in the occupation and use of the environ-ment. Further profitable research along these lines issuggested.

Strategies for the Use of Piedmont Slopesand Bedrock Uplands

Largely because measurement of these areas andsubsequent analyses of site locations and types us-ing GIS techniques have not yet been accomplished,less quantitative information about land-use strategiesin piedmont slopes and bedrock upland regions iscurrently available (but see Fish et al 1992; Roth 1992,1996). Regardless, several potentially important ob-servations for future consideration are in order.

Next to the floodplains themselves, the mostabundant and reliable water sources are springs thatoccur consistently at, or near, the piedmont slope-bedrock outcrop interface. Thus, the upper portionsof the piedmont slope generally have a greater abun-dance and variety of wild plant resources, such asdense stands of saguaro cacti on south-facing slopes,than the lower portions. The upper portions of thepiedmont slope are virtually everywhere fartherfrom floodplain settlements than could be reachedand returned in a single day, requiring at least oneovernight stay.

Except a few sites at the toe of the piedmont, im-mediately adjacent to the floodplain, there is littleevidence on the piedmont slope for sites of the sizeand complexity found in the floodplain. Most of thesites are relatively small scatters with little apparentdepth, and visual inspection of available site distri-

Figure 6. Local density coefficients at 1 km radius for five site groups.

Tucson Basin

Early Agricultural

(n = 34)

Southeast

Arizona

Early Agricultural

(n = 89)

Tucson Basin

Coalescent

Communities

(n = 32)

Southeast

Arizona

Coalescent

Communities

(n = 200)

All

Coalescent

Communities

(n = 275)

Local D

ensity A

naly

sis

Coeffic

ients

Reach boundaries Random 50-ha locations Random 50-km2 locations

2.0

4.0

6.0

8.0

10.0

bution maps suggests a greaterdensity of sites toward the up-per portions of the piedmont.An abundance of saguaroseeds recovered from flood-plain sites suggests saguarocollecting camps may havebeen one kind of piedmontslope exploitation (see Dart1986). The only direct evidencefor use of bedrock uplandscomes in the form of bighornsheep remains regularly recov-ered from floodplain sites.Thus, the primary strategy offloodplain use may have beencomplimented by logisticalsettlements located to accessupper piedmont resources.

CONCLUSIONS

One far-reaching result of the last two decadesof work has been recognition that Early Agriculturalperiod populations made intensive and extensiveuse of floodplains for settlement, as well as for farm-ing and other subsistence activities. Extant surveydata indicate no other portion of the environmentwas exploited to this degree or in these particularways. Unfortunately, empirical data and the inher-ently dynamic characteristics of arid lands floodplainenvironments indicate that sites of this interval mayhave been wholly or partially destroyed by erosionalprocesses, and/or may be deeply buried and diffi-cult to discover and sample. However, there is littlequestion that the primary strategy of Early Agricul-tural period settlement and other activities focusedon the floodplains of major streams and their tribu-taries, especially at or near reach boundaries.

One other aspect of floodplain use deserves men-tion. Although a wide variety of site size and com-position is represented in investigated sites, the larg-est sites appear to be of two types. In the first case,the ratio of pit structures to extramural pits and otherfeatures is quite high (see Gregory, ed. 2001; Gregoryet al. 2007; Mabry et al. 1997); in the other, the re-verse is true, with many more pits and other extra-mural features than pit structures (see Ezzo andDeaver 1998; Freeman 1998). While the latter kindsof sites are earlier in some cases, dates overlap in oth-ers. This difference in storage capacity and its impli-cations for land-use strategies deserves attention infuture research.

Although additional research is necessary, an-other strategy appears to have been the use of

10 The Latest Research on the Earliest Farmers

logistical settlements to exploit resources of the up-per piedmont slope and perhaps of bedrock uplands.For the moment, a basic duality in settlement strat-egies may be hypothesized, with a primary focuson floodplains and a secondary focus on upper pied-

mont slopes. As GIS techniques and other analysesare applied to additional floodplain areas, as well asto piedmont slope and bedrock uplands, it may beanticipated that a much clearer picture of Early Ag-ricultural period settlement strategies will emerge.

REFERENCES CITED

Bellwood, Peter 2005 First Farmers: The Origins of Agricultural So-

cieties. Blackwell Publishing, Oxford, En-gland.

Bellwood, Peter, and Colin Renfrew (editors) 2002 Examining the Language/Farming Dispersal

Hypothesis. McDonald Institute for Archaeo-logical Research, Cambridge, England.

Benz, Bruce F. 1999 On the Origin, Evolution, and Dispersal of

Maize. In Pacific Latin America in Prehistory:The Evolution of Archaic and Formative Cul-tures, edited by M. M. Blake, pp. 25-38.Washington State University Press, Pull-man.

2006 Maize in the Americas. In Histories of Maize:Multidisciplinary Approaches to the Prehistory,Linguistics, Biogeography, Domestication, andEvolution of Maize, edited by J. E. Staller, R.H. Tykot, and B. F. Benz, pp. 9-21. ElsevierAcademic Press, Amsterdam, Boston.

Bradfield, Maitland 1971 The Changing Pattern of Hopi Agriculture.

Occasional Paper No. 30. Royal Anthropo-logical Institute, London.

Bryan, Kirk 1925 Date of Channel Trenching (Arroyo Cut-

ting) in the Arid Southwest. Science 62:338-344.

1941 PreColumbian Agriculture in the Southwestas Conditioned by Periods of Alluviation.Annals of the Association of American Geogra-phers 31:219-242.

Bull, William B. 1991 Geomorphic Responses to Climatic Change.

Oxford University Press, Oxford, England.

1997 Discontinuous Ephemeral Streams. Geomor-phology: An International Journal 19:227-276.

Carpenter, John P., Guadalupe Sanchez, and M. ElisaVillalpando

2002 Of Maize and Migration: Mode and Tempoin the Diffusion of Zea mays in NorthwestMexico and the American Southwest. InTraditions, Transitions, and Technologies:Themes in Southwestern Archaeology, editedby S. H. Schlanger, pp. 245-258. UniversityPress of Colorado, Boulder.

Clark, Caven P. (compiler) 2000 Archaeological Investigations at AZ V:13:201,

Town of Kearny, Pinal County, Arizona. Cul-tural Resources Report No. 114. Archaeo-logical Consulting Services, Ltd., Tempe,Arizona.

Cooke, Ronald U., and Richard W. Reeves 1976 Arroyos and Environmental Change in the

American Southwest. Oxford Research Stud-ies in Geography. Clarendon Press, Oxford,England.

Damp, Jonathan E., Stephen A. Hall, and Susan J.Smith

2002 Early Irrigation on the Colorado Plateaunear Zuni Pueblo. American Antiquity67:665-676.

Dart, Allen 1986 Archaeological Investigations at La Paloma:

Archaic and Hohokam Occupations at ThreeSites in the Northeastern Tucson Basin, Ari-zona. Anthropological Papers No. 4. Insti-tute for American Research, Tucson.

Dean, Jeffrey S. 1988 Dendrochronology and Paleoenvironment-

al Reconstruction on the Colorado Plateaus.In The Anasazi in a Changing Environment,edited by G. J. Gumerman, pp. 119-167.Cambridge University Press, Cambridge,England.

Diehl, Michael W. 2005a Early Agricultural Period Foraging and

Horticulture in Southern Arizona: Implica-tions from Plant Remains. In Subsistence andResource Use Strategies of Early AgriculturalCommunities in Southern Arizona, edited byM. W. Diehl, pp. 73-90. AnthropologicalPapers No. 34. Center for Desert Archaeol-ogy, Tucson.

12 The Latest Research on the Earliest Farmers

Diehl, Michael W. 2005b Epilogue: “Farmaging” During the Early

Agricultural Period. In Subsistence and Re-source Use Strategies of Early AgriculturalCommunities in Southern Arizona, edited byM. W. Diehl, pp. 181-184. AnthropologicalPapers No. 34. Center for Desert Archaeol-ogy, Tucson.

2005c When Corn Was Not Yet King. In Subsis-tence and Resource Use Strategies of Early Ag-ricultural Communities in Southern Arizona,edited by M. W. Diehl, pp. 1-18. Anthropo-logical Papers No. 34. Center for DesertArchaeology, Tucson.

Doolittle, William E., and Jonathan B. Mabry 2006 Environmental Mosaics, Agricultural Di-

versity, and the Evolutionary Adoption ofMaize in the American Southwest. In His-tories of Maize: Multidisciplinary Approachesto the Prehistory, Linguistics, Biogeography,Domestication, and Evolution of Maize, editedby J. E. Staller, R. H. Tykot, and B. F. Benz,pp. 109-121. Elsevier Academic Press,Amsterdam, Boston.

Eddy, Franklin W., and Maurice E. Cooley 1983 Cultural and Environmental History of Cienega

Valley, Southeastern Arizona. Anthropologi-cal Papers No. 43. University of ArizonaPress, Tucson.

Evenari, Michael, Leslie Sanan, and NaphtaliTadmor

1971 The Negev: The Challenge of a Desert. HarvardUniversity Press, Cambridge, Massachu-setts.

Ezzo, Joseph A., and William L. Deaver 1998 Watering the Desert: Late Archaic Farming at

the Costello-King Site. Technical Series No.68. Statistical Research, Inc., Tucson.

Fish, Suzanne K., Paul R. Fish, and John H. Madsen 1992 Parameters of Agricultural Production in

the Northern Tucson Basin. In The MaranaCommunity in the Hohokam World, edited byS. K. Fish, P. R. Fish, and J. H. Madsen, pp.41-52. Anthropological Papers No. 56. Uni-versity of Arizona Press, Tucson.

Ford, Richard I. 1981 Gardening and Farming Before A.D. 1000:

Patterns of Prehistoric Cultivation North ofMexico. Journal of Ethnobiology 1:6-27.

Freeman, Andrea K. L. 2000 Application of High-Resolution Alluvial

Stratigraphy in Assessing the Hunter-Gath-erer/Agricultural Transition in the SantaCruz River Valley, Southeastern Arizona.Geoarchaeology: An International Journal15:559-589.

Freeman, Andrea K. L. (editor) 1998 Archaeological Investigations at the Wetlands

Site, AZ AA:12:90 (ASM). Technical ReportNo. 975. Center for Desert Archaeology,Tucson.

Geib, Phil R. 1996 Archaic Occupancy of the Glen Canyon

Region. In Glen Canyon Revisited, by P. R.Geib, pp. 15-39. Anthropological Papers No.119. University of Utah Press, Salt LakeCity.

Geib, Phil R., and Dale Davidson 1994 Anasazi Origins: A Perspective from Pre-

liminary Work at Old Man Cave. Kiva60:191-202.

Geib, Phil R., and Bruce B. Huckell 1994 Evidence of Late Preceramic Agriculture at

Cibecue, East-Central Arizona. Kiva 59:433-454.

Gilpin, Dennis 1994 Lukachukai and Salina Springs: Late Ar-

chaic/Early Basketmaker Habitation Sitesin the Chinle Valley, Northeastern Arizona.Kiva 60:203-218.

Graf, William L. 1982 Spatial Variations of Fluvial Processes in

Semi-arid Lands. In Space and Time in Geo-morphology. Proceedings of the 12th AnnualBinghamton Symposium, edited by C. E.Thorn, pp. 193-217. Allen and Unwin, Bos-ton.

1983a The Arroyo Problem—Palaeohydrologyand Palaeohydraulics in the Short Term. InBackground to Palaeohydrology: A Perspective,edited by K. J. Gregory, pp. 262-303. JohnWiley, New York.

1983b Flood-related Channel Change in an Arid-regionRiver. Earth Surface Processes andLandforms 8:125-139.

Early Agricultural Period Settlement Strategies in the Southern Southwest 13

Graf, William L. 1983c Variability of Sediment Removal in a Semi-

arid Watershed. Water Resources Research19:643-652.

1988 Fluvial Processes in Dryland Rivers. SpringerPress, New York.

Gregory, David A. 1996 New Issues in the Interpretation of Archaic

Period Radiocarbon Dates. Paper presentedat the Conference on the Archaic Prehistoryof the North American Southwest, MaxwellMuseum, University of New Mexico, Albu-querque.

2001 An Evaluation of Early Agricultural PeriodChronology in the Tucson Basin. In Excava-tions in the Santa Cruz River Floodplain: TheEarly Agricultural Period Component at LosPozos, edited by D. A. Gregory, pp. 237-254.Anthropological Papers No. 21. Center forDesert Archaeology, Tucson.

Gregory, David A. (editor) 1999 Excavations in the Santa Cruz River Floodplain:

The Middle Archaic Component at Los Pozos.Anthropological Papers No. 20. Center forDesert Archaeology, Tucson.

2001 Excavations in the Santa Cruz River Floodplain:The Early Agricultural Period Component atLos Pozos. Anthropological Papers No. 21.Center for Desert Archaeology, Tucson.

Gregory, David A., and Sam W. Baar, IV 1999 Stratigraphy, Chronology, and Character-

istics of the Natural and Cultural Deposits.In Excavations in the Santa Cruz River Flood-plain: The Middle Archaic Component at LosPozos, edited by D. A. Gregory, pp. 13-32.Anthropological Papers No. 20. Center forDesert Archaeology, Tucson.

Gregory, David A., and Gary A. Huckleberry 1994 An Archaeological Survey in the Blackwater

Area: Vol. 1. The History of Human Settlementin the Blackwater Area. Cultural ResourcesReport No. 86. Archaeological ConsultingServices, Ltd., Tempe, Arizona.

Gregory, David A., and Fred L. Nials 1985 Observations Concerning the Distribution

of Classic Period Hohokam PlatformMounds. In Proceedings of the 1983 HohokamSymposium, part I, edited by A. E. Dittert Jr.and D. E. Dove, pp. 373-388. OccasionalPapers No. 2. Arizona Archaeological Soci-ety, Phoenix.

2005 The Environmental Context of Early Agri-cultural Period Occupations in the TucsonBasin. In Subsistence and Resource Use Strat-egies of Early Agricultural Communities inSouthern Arizona, edited by M. W. Diehl, pp.19-71. Anthropological Papers No. 34. Cen-ter for Desert Archaeology, Tucson.

Gregory, David A., Michelle N. Stevens, Fred L.Nials, Mark R. Schurr, and Michael W.Diehl

2007 Excavations in the Santa Cruz Floodplain: Fur-ther Investigations at Los Pozos. Anthropo-logical Papers No. 27. Center for DesertArchaeology, Tucson.

Hack, John T. 1942 The Changing Physical Environment of the

Hopi Indians of Arizona. Papers of the Pea-body Museum of American Archaeologyand Ethnography No. 35(1). Harvard Uni-versity, Cambridge, Massachusetts.

Hard, Robert J., and John R. Roney 1998 A Massive Terraced Village Complex in

Chihuahua, Mexico, 3000 Years BeforePresent. Science 279:1661-1664.

Hard, Robert, Arthur C. MacWilliams, John R.Roney, Karen R. Adams, and William L.Merrill

2006 Early Agriculture in Chihuahua, Mexico. InHistories of Maize: Multidisciplinary Ap-proaches to the Prehistory, Linguistics, Bioge-ography, Domestication, and Evolution ofMaize, edited by J. E. Staller, R. H. Tykot,and B. F. Benz, pp. 471-486. Elsevier Aca-demic Press, Amsterdam, Boston.

Harris, David R. (editor) 1996 The Origins and Spread of Agriculture and

Pastoralism in Eurasia. Smithsonian Institu-tion Press, Washington, D.C.

14 The Latest Research on the Earliest Farmers

Haynes, C. Vance Jr. 1968 Geochronology of Late Quaternary Allu-

vium. In Means of Correlation of QuaternarySuccessions, vol. 8, edited by R. B. Morrisonand H. E. Wright Jr., pp. 591-631. Proceed-ings of the VII Congress, International As-sociation of Quaternary Research, Univer-sity of Utah Press, Salt Lake City.

1987 Curry Draw, Cochise County, Arizona: ALate Quaternary Stratigraphic Record ofPleistocene Extinction and PaleoIndian Ac-tivities. In Geological Society of America Cen-tennial Field Guide–Cordilleran Section, pp.23-28. Geological Society of America, Boul-der.

Haynes, C. Vance Jr., and Bruce B. Huckell 1986 Sedimentary Successions on the Prehistoric

Santa Cruz River, Tucson, Arizona. Open FileReport No. 86-15. Arizona Geological Soci-ety, Tucson.

Hendrickson, Dean A., and W. L. Minckley 1984 Cienegas. Vanishing Climax Communities

of the American Southwest. Desert Plants6:131-175.

Hill, Jane H. 2001 Proto-Uto-Aztecan: A Community of Cul-

tivators in Central Mexico? American An-thropologist 103:913-934.

2002 Toward a Linguistic History of the South-west: “Azteco-Tanoan” and the Arrival ofMaize Cultivation. Journal of Anthropologi-cal Research 58:457-476.

Hinderlider, M. C. 1913 Irrigation of the Santa Cruz Valley, parts I

and II. Engineering Record 68:200-201.

Huber, Edgar K., and Heather J. Miljour 2004 Early Maize on the Colorado Plateau: New

Dates from West-central New Mexico. Pa-per presented at the Ninth Southwest Sym-posium, Ciudad de Chihuahua, Mexico, 9-10 January.

Huber, K. Edgar, and Carla R. Van West (editors) 2006 Fence Lake Project: Archaeological Data Recov-

ery in the New Mexico Transportation Corri-dor and First Five-Year Permit Area, Fence LakeCoal Mine Project, Catron County, NewMexico. Technical Series No. 84. StatisticalResearch, Inc., Tucson.

Huckell, Bruce B. 1990 Late Preceramic Farmer-Foragers in South-

ern Arizona: A Cultural and EcologicalConsideration of the Spread of Agricultureinto the Arid Southwestern United States.Unpublished Ph.D. dissertation, Depart-ment of Arid Lands Resource Sciences, Uni-versity of Arizona, Tucson.

1995 Of Marshes and Maize: Preceramic Agricul-tural Settlements in the Cienega Valley, South-eastern Arizona. Anthropological Papers No.59. University of Arizona Press, Tucson.

Huckell, Bruce B., and Lisa W. Huckell 1984 Excavations at Milagro, a Late Archaic Site

in the Eastern Tucson Basin. Ms. on file,Cultural Resource Management Division,Arizona State Museum, University of Ari-zona, Tucson.

Huckell, Bruce B., Lisa W. Huckell, and Karl K.Benedict

2002 Maize Agriculture and the Rise of MixedFarming-Foraging Economies in Southeast-ern Arizona During the Second MillenniumB.C. In Traditions, Transitions, and Technolo-gies: Themes in Southwestern Archaeology,edited by S. H. Schlanger, pp. 137-159. Uni-versity of Colorado Press, Boulder.

Huckell, Bruce B., Lisa W. Huckell, and Suzanne K.Fish

1995 Investigations at Milagro, a Late PreceramicSite in the Eastern Tucson Basin. TechnicalReport No. 94-5. Center for Desert Archae-ology, Tucson.

Hunter-Anderson, Rosalind L. 1986 Prehistoric Adaptations in the American South-

west. Cambridge University Press, Cam-bridge, England and New York.

Johnson, Ian 1984 Cell Frequency Recording and Analyis of

Artifact Distributions. In Intrasite SpatialAnalysis in Archaeology, edited by H. J.Hietala, pp. 75-96. Cambridge UniversityPress, Cambridge, England.

Early Agricultural Period Settlement Strategies in the Southern Southwest 15

Karlstrom, E. T. 2005 Late Quaternary Landscape History and

Geoarchaeology of Two Drainages on BlackMesa, Northeastern Arizona, USA. Geoar-chaeology 20:1-28.

Kearns, Timothy M. 1996 A proposed Phase Sequence for Tohachi

Flats and the Southern Chuska Valley,Northwest New Mexico. In Time, Place, andSociety: Project Synthesis, by T. M. Kearnsand J. L. McVickar, pp. 3-1 to 3-31. PipelineArchaeology 1990-1993: The El Paso Natu-ral Gas North System Expansion Project,New Mexico and Arizona, Vol. XIII. West-ern Cultural Resource Management, Inc.,Farmington, New Mexico.

Kintigh, Keith W. 1992 Tools for Quantitative Archaeology: Programs

for Quantitative Analysis in Archaeology. Ms.in possession of the author, Tempe, Ari-zona.

LeBlanc, Steven A. 2002 Conflict and Language Dispersal: Issues

and a New World Example. In Examiningthe Language/Farming Dispersal Hypothesis,edited by P. Bellwood and C. Renfrew, pp.357-367. McDonald Institute for Archaeo-logical Research, Cambridge, England.

Lee, Willis T. 1905 Underground Waters of Salt River Valley, Ari-

zona. Water-Supply and Irrigation PaperNo. 136. U.S. Geological Survey. U.S. Gov-ernment Printing Office, Washington, D.C.

Leopold, L. B., W. W. Emmett, and R. M. Myrick 1966 Channel and Hillslope Processes in a Semi-

arid Area, New Mexico. U.S. Geological Sur-vey Professional Paper 352-G:193-253.

Leopold, Luna B., M. Gordon Wolman, and John P.Miller

1964 Fluvial Processes in Geomorphology. W. H.Freeman, San Francisco.

Mabry, Jonathan B. 1998 Paleoindian and Archaic Sites in Arizona.

Technical Report No. 97-7. Center for De-sert Archaeology, Tucson.

2002 The Role of Irrigation in the Transition toAgriculture and Sedentism in the South-west: A Risk Management Model. In Tradi-tions, Transition, and Technologies, Themes inSouthwestern Archaeology, edited by S. H.Schlanger, pp. 178-199. University Press ofColorado, Boulder.

2005a Changing Knowledge and Ideas About theFirst Farmers in Southeastern Arizona. InThe Late Archaic Across the Borderlands: FromForaging to Farming, edited by B. J. Vierra,pp. 41-83. University of Texas Press, Aus-tin.

2005b Diversity in Early Southwestern Farmingand Optimization Models of Transitions toAgriculture. In Subsistence and Resource UseStrategies of Early Agricultural Communitiesin Southern Arizona, edited by M. W. Diehl,pp. 113-152. Anthropological Papers No. 34.Center for Desert Archaeology, Tucson.

2006 Radiocarbon Dating of the Early Occupa-tions. In Rio Nuevo Archaeology, 2000-2003:Investigations at the San Agustín Mission andMission Gardens, Tucson Presidio, TucsonPressed Brick Company, and Clearwater Site,edited by J. H. Thiel and J. B. Mabry, pp.19.1-19.5. Technical Report No. 2004-11.Desert Archaeology, Inc., Tucson. <http:// w w w . c d a r c . o r g / p d f / r n /rio_nuevo_ch19.pdf>

Mabry, Jonathan B., and Michelle N. Stevens 2000 The Archaeology of Archaic Hunter-gath-

erers and Early Farmers in SoutheasternArizona. In The Archaeology of the Land Be-tween: Regional Dynamics and Prehistory ofSoutheastern Arizona, edited by H. D.Wallace and A. I. Woosley. Amerind Foun-dation, Dragoon, Arizona, and Universityof New Mexico Press, Albuquerque, inpress.

Mabry, Jonathan B., Deborah L. Swartz, HelgaWöcherl, Jeffery J. Clark, Gavin H. Archer,and Michael W. Lindeman

1997 Archaeological Investigations of Early VillageSites in the Middle Santa Cruz Valley: Descrip-tions of the Santa Cruz Bend, Square Hearth,Stone Pipe, and Canal Sites. AnthropologicalPapers No. 18. Center for Desert Archaeol-ogy, Tucson.

16 The Latest Research on the Earliest Farmers

Martin, Paul S. 1963 The Last 10,000 Years: A Fossil Pollen Record

of the American Southwest. University of Ari-zona Press, Tucson.

Matson, R. G. 2002 The Spread of Maize Agriculture in the U.S.

Southwest. In Examining the Language/Farm-ing Dispersal Hypothesis , edited by P.Bellwood and C. Renfrew, pp. 341-356.McDonald Institute for Archaeological Re-search, Cambridge, England.

Meinzer, O. E. (editor) 1942 Hydrology. Dover Press, New York.

Minnis, Paul E. 1992 Earliest Plant Cultivation in the Desert Bor-

derlands of North America. In The Originsof Agriculture, An International Perspective,edited by C. W. Cowan and P. J. Watson,pp. 121-141. Series in Archaeological In-quiry, Smithsonian Institution Press, Wash-ington, D.C.

Moreno, Teresa K. 2000 Accelerator Mass Spectrometry Dates from

McEuen Cave. Kiva 65:341-360.

Nabhan, Gary P. 1979 The Ecology of Floodwater Farming in Arid

Southwestern North America. AgroEcosys-tems 5:235-255.

Nials, Fred L. 2008 Geomorphology and Stratigraphy. In Las

Capas: Early Irrigation and Sedentism in aSouthwestern Floodplain, edited by J. B.Mabry, pp. 35-53. Anthropological PapersNo. 28. Center for Desert Archaeology, Tuc-son.

Nials, Fred L., and David A. Gregory 1989 Irrigation Systems in the Lower Salt River

Valley. In The 1982-1984 Excavations at LasColinas: Vol. 5. Environment and Subsistence,by D. A. Graybill, D. A. Gregory, F. L. Nials,S. K. Fish, R. E. Gasser, C. H. Miksicek, andC. R. Szuter, pp. 39-58. Archaeological Se-ries No. 162. Arizona State Museum, Uni-versity of Arizona, Tucson.

Nials, Fred. L., David A. Gregory, and J. Brett Hill 2007 The Stream Reach Concept: Prehistoric Ag-

riculture in Southwestern Riverine Environ-ments. Paper presented at the 72nd AnnualMeeting of the Society for American Ar-chaeology, Austin.

2011 The Stream Reach Concept and the Macro-scale Study of Riverine Agriculture in Aridand Semiarid Environments Geoarchaeology26(5):724-761.

Parry, W. J., F. E. Smiley, and G. R. Burgett 1994 The Archaic Occupation of Black Mesa,

Arizona. In Archaic Hunter-gatherer Archae-ology in the American Southwest, edited byB. J. Vierra, pp. 185-230. Contributions inAnthropology No. 13(1). Eastern NewMexico University, Portales.

Premo, Lucas S. 2001 Predictive Model of Late Archaic Period Site

Locations in the Tucson Basin. UnpublishedMaster’s thesis, University of Arizona, De-partment of Anthropology, Tucson.

Roth, Barbara J. 1992 Sedentary Agriculturalists or Mobile

Hunter-Gatherers? Recent Evidence on theLate Archaic Occupation of the NorthernTucson Basin. Kiva 57:291-314.

1996 Regional Land Use in the Late Archaic ofthe Tucson Basin: A View from the UpperBajada. In Early Formative Adaptations in theSouthern Southwest, edited by B. J. Roth, pp.37-48. Monographs in World ArchaeologyNo. 5. Prehistory Press, Madison, Wiscon-sin.

Sandor, Jon A. 1992 Long-Term Effects of Prehistoric Agricul-

ture on Soils: Examples from New Mexicoand Peru. In Soils in Archaeology: LandscapeEvolution and Human Occupation, edited byV. T. Holliday, pp. 217-246. SmithsonianInstitution Press, Washington, D.C.

Sayles, E. B., and Ernst Antevs 1941 The Cochise Culture. Medallion Papers No.

29. Gila Pueblo, Globe, Arizona.

Schumm, S. A., and R. F. Hadley 1957 Arroyos and the Semi-arid Cycle of Erosion.

American Journal of Science 225:161-174.

Early Agricultural Period Settlement Strategies in the Southern Southwest 17

Shackley, M. Steven 2005 Chronometry and Geochemistry at McEuen

Cave: The Radiocarbon and ObsidianGeochemical Data. Paper presented at the70th Annual Meeting of the Society forAmerican Archaeology, Salt Lake City,Utah, 30 March-3 April.

Simmons, Alan H. 1986 New Evidence for the Early Use of Culti-

gens in the American Southwest. AmericanAntiquity 51:73-89.

Smiley, Francis E. 1994 The Agricultural Transition in the North-

ern Southwest: Patterns in the CurrentChronometric Data. Kiva 60:165-189.

Southworth, C. H. 1919 History of Irrigation on the Gila River. In

Hearings Before the Committee on Indian Af-fairs, House of Representatives, 66th Congress,First Session Vol. 2:103-223. Washington,D.C.

Staller, John E., Robert H. Tykot, and Bruce F. Benz(editors)

2006 Histories of Maize: Multidisciplinary Ap-proaches to the Prehistory, Linguistics, Bioge-ography, Domestication, and Evolution ofMaize. Elsevier Academic Press, Amster-dam, Boston.

Tagg, Martyn D. 1996 Early Cultigens from Fresnal Shelter, South-

eastern New Mexico. American Antiquity61:311-324.

Upham, Steadman., Richard S. MacNeish, WaltonC. Galinat, and Christopher M. Stevenson

1987 Evidence Concerning the Origin of Maiz deOcho. American Anthropologist 89:410-419.

Waters, Michael R. 1988 The Impact of Fluvial Processes and Land-

scape Evolution on the Archaeological Sitesand Settlement Patterns along the SanXavier Reach of the Santa Cruz River, Ari-zona. Geoarchaeology: An International Jour-nal 3:205-219.

Waters, Michael R., and C. Vance Haynes Jr. 2001 Late Quaternary Arroyo Formation and

Climate Change in the American South-west. Geology 29:399-402.

Whittlesey, Stephanie M., S. Jerome Hesse, andMichael S. Foster (editors)

2007 Recurrent Sedentism and the Making of Place:Archaeological Investigations at Las Capas, aPreceramic Period Farming Community in theTucson Basin Southern Arizona (Draft). Cul-tural Resources Report No. 07-556. SWCA,Inc., Tucson.

Wilcox, David R., William H. Doelle, J. Brett Hill,and Jim P. Holmlund

2003 Coalescent Communities GIS Database:Museum of Northern Arizona, Center forDesert Archaeology, Geo-Map Inc. Ms. onfile, Center for Desert Archaeology, Tucson.

Wills, W. H. 1988 Early Prehistoric Agriculture in the American

Southwest. School of American ResearchPress, Santa Fe.

1995 Archaic Foraging and the Beginning of FoodProduction in the American Southwest. InLast Hunters—First Farmers: New Perspectiveson the Prehistoric Transition to Agriculture,edited by T. D. Price and A. B. Gebauer, pp.215242. School of American Research Press,Santa Fe.

Winn, Fred 1926 The West Fork of the Gila River. Science

64:16-17.