Bajo Communities: A Case Study from the Central Peten

Bajo Communities: A Case Study from the Central Peten and theorists of state formation (Palerm and Wolf 1957; Steward 1955; Wittfogel 1957), who argued that the impoverished nature of tropical environments precluded the in situ development of social complexity. Thus, despite their artistic accomplishments, scientific achievements, and architectural monuments, the Maya were thought to lack certain attributes associated with great civilizations. These included large populations, dense urban centers, and elaborate civil bureaucracies. "True" Mesoamerican civilizations, such as those found in the more arid Basin of Mexico, were unquestionably supported by an intensive agricultural regime. Yet, ethnographic and ethnohistoric evidence for intensive systems of cultivation among the ancient Maya was largely ignored (Turner 1978). At the same time archaeological research, concentrated in the monumental precincts of Maya sites, focused upon questions of a different nature. The swidden thesis was ultimately challenged by the accumulation of new archaeological evidence. Beginning in the 1950s, several projects moved away from site centers and began to focus on regional patterns of settlement and population growth. Large-scale investigations in Peten, Guatemala, and Belize (Bullard 1960; Haviland 1969; Willey et al. 1965) yielded evidence of substantial urban and rural populations throughout the Maya lowlands. Drawing on these and similar results, Late Classic population figures of 500-800 people/ km2 in site centers, and more than 150 people/km 2 in rural areas, were suggested (Culbert and Rice 1990). Confronted with these new data, Mayanists concluded that at their peak, population densities would have far exceeded the carrying capacity of swidden agriculture, estimated at no more than 50 to 77 people/km2 (Cowgill 1962; Reina 1967). The vision of the Maya as developmentally hampered by their environment and by their agricultural base was called into question and prompted consideration of other subsistence options. Some Mayanists examined possible alternatives to maize as the staple crop. Bronson (1966) proposed a reliance on root crops, while Puleston (1982) argued for arboriculture, especially of the ramon (Brosimum alicastrum) tree. Lange (1971) suggested marine resources, and Wiseman (1978) proposed a model of a productive artificial rainforest. To date, archaeobotanical evidence supports neither the root crop nor the arboriculture argument (Turner and Miksicek 1984), while marine foods are not common in most archaeological contexts (McKillop 1994). The artificial rainforest hypothesis

Julie L. Kunen, T. Patrick Culbert, Vilma Fialko, Brian R. McKee, and Liwy Grazioso Julie L. Kunen is a doctoral candidate in anthropology at the University of Arizona, T. Patrick Culbert is professor emeritus of anthropology at the University of Arizona, Vilma Fialko is an archaeologist with the Instituto de Antropologia e Historia in Guatemala, Brian R. McKee is a doctoral candidate at the University of Arizona, and Liwy Crazioso is the Director of Museums with the Instituto de Antropologia e Historia in Guatemala.

A vital task facing scholars concerned with the structure of ancient Maya society is reconstructing the relationships among subsistence, population, and social organization. Particularly in the southern Maya lowlands, where extremely dense populations were found by Late Classic times, substantial modifications of the landscape through a variety of land management techniques were necessary. The nature of these modifications, the range of innovations, and the degree of intensification are still, however, poorly understood. This article explores the significance of one form of landscape modification entailed by the occupation and use of seasonal wetlands (bajos). Below, we examine the debate over the significance of intensive wetland agriculture in the Maya lowlands, then discuss the preliminary findings of a case study designed to investigate the duration and intensity of use of a major central Peten bajo.

The Intensive Agriculture Debate The "swidden thesis" dominated research on Maya agriculture and social organization until the middle of this century (Turner 1993). The image of the ancient Maya as slash-and-burn agriculturalists practicing long-fallow swiddening was based on knowledge of the farming practices of colonial and more recent Maya occupants of the lowlands. This knowledge was coupled with the belief that the region had never been densely populated. The swidden thesis was also bolstered by a narrow vision of tropical environments as ecologically homogeneous and therefore incapable of supporting large populations. Such a view also accorded with the ideas of environmental determinists (e.g., Meggers 1954) Culture & Agriculture

15

Vol. 22, No. 3

Fall 2000

has similarly fallen out of favor, since pollen evidence for widespread deforestation suggests forest management would not have been a viable strategy (Abrams and Rue 1988). Other scholars proposed combinations of several cultivation strategies, including infield-outfield systems (Netting 1977), multiple cropping (Culbert, Magers, and Spencer 1978), kitchen gardening (Wilken 1971), and silviculture (Gomez-Pompa 1987; McKillop 1994). Still other researchers (Gliessman et al. 1983; Harrison 1977; Healy et al. 1983; Siemens 1982; Turner 1983) began to focus on growing evidence for agricultural intensification in the form of hillslope terracing and utilization of wetlands through construction of raised or drained fields and canals. These scholars focused on the possibility of planting cycles involving farming in upland areas during the rainy season, supplemented by plantings in river floodplains or in wetlands that retained sufficient moisture during the dry season. The debate regarding the extent and significance of wetland agriculture among the Maya is particularly contentious. Many scholars currently believe that intensive agriculture involving the modification of wetlands was a central component of ancient Maya subsistence, allowing for the maintenance of dense populations over centuries of complex social development (Culbert et al. 1996,1997; Harrison and Turner 1978; Turner and Harrison 1983). In contrast, other researchers (Fedickand Ford 1990; Pohl 1985,1990; Pohl et al. 1996; Pope and Dahlin 1989; Pope, Pole, and Jacob 1996; Pope et al. 2000) have argued that wetland agriculture was neither as geographically nor as temporally widespread as its advocates believe, nor was it as significant a part of the subsistence base. This debate between proponents of what Turner characterizes as "the new orthodoxy" of intensive agriculture and its critics has become polemical (see Turner [1995] for a thorough discussion of the debate). Resolution of this conflict must be based in both a careful evaluation of existing evidence and the pursuit of new data. Investigations designed specifically to assess the existence of ancient wetland agricultural systems in various parts of the lowlands are called for. To fully understand the opposing interpretations of existing data on wetland agriculture, one must first distinguish among three different types of wetland ecosystems categorized by scholars today—riverine floodplains, permanent wetlands, and seasonally inundated wetlands, or bajos. It must be recognized, however, that these categories may not have been so fixed in the past, as evidence that today's seasonal and permanent swamps were in ancient times affected by processes of siltation, erosion, and seasonal fluctuations in the water table has slowly accumulated (e.g., Dunning et al. 2000). Research on the modification of river floodplains foragricultural purposes has yielded evidence of

Culture & Agriculture

field systems along the Candelaria River in Campeche (Siemens and Puleston 1972) and in the floodplains of the New River and Hondo River in northern Belize (Lambert and Arnason 1983; Pohl 1990; Puleston 1977; Siemens 1982). Channelized and possibly raised field systems have been identified in Pulltrouser Swamp, a permanent wetland in northern Belize (Berry and McAnany 2000; Turner 1993; Turner and Harrison 1983).1 Evidence of agricultural use of seasonal wetlands has been reported for the Bajo de Morocoy in southern Quintana Roo (Gliessman et al. 1983, Harrison and Turner 1978; Turner 1974), the Bajo el Laberinto in southeast Campeche (Folan and Gallego 1998), and the bajo near Rio Azul in northern Peten (Culbert 1990). The most recent evidence comes from McAnan/s continuing research at Pulltrouser Swamp in northern Belize (Berry and McAnany 2000; McAnany 1998), Pedicle's research on agricultural features in a zone of many permanent and seasonal wetlands in northern Yucatan (Fedick 1996; Fedick and Hovey 1995), and evidence of bajo use and modification in the central Peten (Culbert et al. 1996,1997; Fialko 1999). Understanding the differences among wetland types and recognizing that these differences have important implications for the types of landscape modifications that were possible has been a vital first step. While debate about the chronology, function, and extent of agricultural systems in riverine and permanent wetlands continues, there is now little controversy surrounding the idea that these areas were indeed utilized by the ancient Maya. The same cannot be said for the bajos of central and northern Peten, where questions persist about both the suitability of these seasonal wetlands for cultivation and the interpretation of what little archaeological evidence for their use exists. The paucity of research on bajos can be explained in part by the logistical difficulty of studying or even accessing these densely forested lands. The Bajo de Morocoy has been investigated primarily from the air, and the recognition of patterns of channelized fields was possible only after large parts of the bajo had been cleared of vegetation. Subsequently,excavation consisted of only a brief study (Gliessman et al. 1983). Study of the Bajo Pedernal at Rio Azul was similarly limited in scope, consisting ofa single, 100-meter long trench spanning the bajo (Culbert et al. 1990). Fedick's recent study in the arid northern Yucatan focused on a very different wetland zone whose hydrology and geology is not comparable to the more "typical" central Peten bajos (Fedick 1996; Fedick and Hovey 1995). Finally, while Folan's research in the Bajo El Laberinto identified agricultural fields within and on the edges of the bajo, the primary focus of that research was on the role of multi-cropping in the survival of Calakmul's large ancient population, and not on bajo

16

Vol. 11, No. 3

Fall 1000

modification and use per se (Folan and Gallegos 1998,1999). The lack of research in seasonal wetlands belies their importance; as much as 40 percent of the land in central Peten, the heart of the lowlands, can be classified as bajo (Rice and Culbert 1990: Table 1.1). Investigation of these lands is pivotal, therefore, in evaluating competing reconstructions of Maya subsistence practices and social organization.

The Bajo Communities Project The Bajo Communities Project (BCP) is designed to redress the paucity of research on bajo agricultural use through a multifaceted study of a major bajo in central Peten.2 The project is unique in two ways. First, it recognizes that one obstacle to understanding agricultural use of seasonal wetlands is the misconception that bajo environments are uniformly unproductive. Thus, a primary research goal is the investigation and documentation of variability among bajo microenvironments. Second, we realize that a thorough investigation of ancient bajo use requires both the study of settlements associated with bajos and the direct testing of bajos themselves for evidence of relict agricultural features. Our comprehensive approach thus involves the employment of a suite of investigative techniques and analyses including (1) use of satellite imagery and global positioning systems (GPS) to identify broad-scale vegetational and elevational variability; (2) ground survey, including mapping of vegetational associations, soil moisture measurements, archaeological survey, and precise topographic and architectural mapping; (3) excavation, featuring stratigraphic testing, chronological assessment of settlements, and testing of bajo agricultural features; and (4) post-excavation analyses, including artifact and ecofact analyses. The Bajo la Justa,a 150 km2 seasonal wetland separating the sites of Yaxha and Nakum, lies within the area of the Institute de Antropologia e Historia's (IDAEH, Institute of Anthropology and History) Proyecto Triangulo (Triangle Project). This bajo was selected as the locus of the BCP study (Figure 1). As part of the research of the related Subproyecto Intersitios (Intersite Subproject), directed by Lie. Vilma Fialko, a transect crossing 6 km of the bajo between Yaxha and Nakum was cut and staked at 25 m intervals, providing a measured cross section of the bajo. A 17 km jeep road between Yaxha and Nakum served as a second transect, roughly parallel to the shorter one, and allowed access to the bajo interior. These two transects served as baselines for the BCFs research. Initial inspection of the transect and the road indicated that a wide variety of bajo microenvironments was represented.

Culture & Agriculture

17

During two seasons of fieldwork (1995 and 1996), the BCP accomplished four objectives. First, we recorded and investigated a variety of microenvironments found within the bajo. Second, we surveyed areas of higher land, characterized by upland forest, within and on the edges of the bajo. Third, we tested a sample of archaeological sites associated with the bajo to establish a preliminary chronology of settlement and to collect samples for archaeobotanical analysis. Finally, we placed trenches in a selected area of the bajo in order to identify possible modifications of the landscape associated with ancient agriculture. Below, we summarize our findings in these four arenas.

Bajo Microenvironments It is becoming increasingly evident that changes in wetland hydrology over the last 1,000 years have altered the characteristics of what are today seasonal wetlands (Dunning et al. 2000; Pope et al. 2000). However, it is also clear that the substantial environmental heterogeneity that exists today within bajos is indicative of differences that would have prevailed in ancient times. Thus, study of modern microenvironments provides relevant clues to past land use patterns. Documentation of the microenvironmental variability within the bajo first requires the establishment of a classification of vegetational associations. These associations, mapped on the ground, are then correlated with Landsat Thematic Mapper (TM) satellite imagery.3 The correlation, when completed, will create a vegetation "signature" for different bajo types, so that these can be identified in satellite imagery and used as a guide for further archaeological investigations (Figure 2). Once vegetation signatures are established, quantification of the various types of vegetation across the landscape will allow for stratified sampling of bajos with reduced need for labor-intensive surveying. The use of satellite imagery in classifying vegetation categories also has important implications for the management of the lands of the Maya Biosphere Reserve, of which the Bajo la Justa is a part. Our classification of bajo vegetation is based on information provided by knowledgeable local consultants who are native to Peten (e.g., Lanza n.d.). To establish this classification, we surveyed the 6 km transect between Yaxha and Nakum and the 17 km jeep road between the two sites, recording the vegetational association at 25 m intervals. In surveying the road, we were careful to avoid recording the disturbance vegetation immediately alongside the track. We also recorded multiple readings from a hand-held global positioning system (GPS) along both the transect and the road to facilitate the correlation of the transects and the satellite imagery.

Vol. 22, No. 3

Fall 2000

Figure 1 Bajo Communities Project Area

-BAJO LA JUSTA

Bajo Sites kilometers

Culture & Agriculture

18

— •

%

Aguada

Vol. 11, No. 3 Fall2J000

Figure 2 Landsat TM Image of the Project Area. Dark areas of uniform texture are portions of the Bajo la Justa Lighter areas with rougher texture are upland ridges. Poza Maya, Cara Fea, and Yaxha are archaeological sites. Numbers 1-6 identify upland ridges or 'islands' surveyed during the project, and 7 is the Aguada Maya.

Culture & Agriculture

Vol. 22, ,\i.

;

Fd

Figure 3 Revised Typology of Bajo Vegetation Suites

Bajo vegetation

Scrub bajo

Palm bajo

Mixed scrub

Mixed palm

Escobal

Corozal

Botanal

Pucteal

Lundell (1933,1937), whose vegetation studies remain landmarks in the field, recognized a basic dichotomy in bajo environments and made a distinction between escoba bajo and tintal bajo. In Lundell's usage, escoba bajo is a term describing areas dominated by palms, with a higher canopy and a more open understory than is found in other bajo areas. In contrast, tintal bajo refers to thick, low-canopied, thorny scrub that is extremely difficult to walk through. Lundell thought that escoba bajo was limited in distribution to the fringes of the more extensive tintal bajo. We find that this dichotomy obscures the existence of a larger variety of bajo subtypes. Moreover, the use of the species names "escoba" (Crysophilia argentes Bartlett) and "tintal" (Haematoxylum campechianum L.) to refer to what are actually more generic categories of vegetation is misleading. Recognizing, as have other scholars (e.g., Wright et al. 1959), the occurrence of many varieties of bajo vegetation, we have refined Lundell's classification to distinguish two general categories of dominant vegetation: palm bajo and scrub bajo. Within each of these types a number of subtypes, or habitats, can exist in which a particular plant species dominates (Figure 3). For example, escobal, corozal, and botanal are all species-specific types of palm bajo in which escoba palm, corozo palm (Orbignya cohune), and botan palm (Sabal mayarum) dominate the assemblage, respectively. Similarly, tintal, pucteal, sapamucheal, and huechal are species-specific types of scrub bajo dominated by palo de tinto, pucte {Bucida buceras), sapamuche (Oreopanax guatemalense), and zacate de hueche {Scleria sp.), respectively. In certain habitats, however, no single species dominates, but instead a mix of several palm species or several types of scrub is found. In these

Culture & Agriculture

20

Tintal

Huechal

areas, then, a more accurate terminology is mixed palm bajo or mixed scrub bajo. Contrary to Lundell's belief, palm bajo is not confined to the fringes of larger expanses of scrub bajo. Instead, we found that significant portions of both the transect and the road can be characterized as palm bajo of various subtypes: 31 percent of the vegetation along the transect (1.85 km) and 59 percent of the vegetation along the road (10.3 km) is palm bajo, with 14 percent (2.5 km) of the palm bajo along the road classified specifically as corozal. Our local consultants inform us that palm bajo is quite desirable land for farming because it is relatively well-drained, yet retains sufficient moisture for crop planting towards the end of the rainy season and into the dry season. Our readings from soil moisture probe tests confirm this, indicating that palm bajo retains enough moisture in the dry season for crops to succeed. In tests along the transect during the dry season, 14 of 15 readings in palm bajo yielded moisture levels between 60 and 80 percent. In contrast, moisture readings in scrub bajo were much more variable, ranging from 25 to 90 percent. Readings from the high forest surrounding Yaxha were uniformly lower, between 33 and 65 percent—too low for dry season cultivation. It has been argued by some researchers that bajos are too wet in the rainy season and too parched in the dry season to sustain crops (Pope and Dahlin 1989). Our data suggest that for palm bajo, this is not the case (also see Folan and Gallegos 1998). Upland forest is found at the ends of the road and transect, where the sites of Yaxha and Nakum sit on the most prominent ridges in the area. A patch of upland forest also occurs at a point 4.2 km along the transect from Yaxha,

Vol. 22, No. 3 Fail 2m

about 40 structures arranged in nine courtyard groups. Many of the structures are very large, up to 20 m in height, and feature vaulted architecture. Test pits dug previously by the IDAEH archaeologists revealed as much as 7 m of fill and floors in several plaza areas. Five small structures approximately 200 meters east of the site center were not investigated by the IDAEH archaeologists and were mapped by us. Test excavations in this group (Group B) are discussed below. During the 1995 vegetation survey a new site, Cara Fea, was discovered on a forested ridge about 3 km west of Poza Maya (Figure 4). In contrast to the large monumental architecture at Poza Maya, Cara Fea consists primarily of much smaller structures, many of them residential. Twentythree mounds are arranged in eight patio groups. Five of these mounds are pyramidal, probably shrines, but the largest of these is only 4 m high. Poza Maya and Cara Fea were the focus of our program of test excavations, which are discussed below. In addition to the ridges on which these two sites are located, we surveyed six other "islands" of high land and found evidence of occupation in all of these areas (Table 1; also see Figure 2). "Island" 1, adjacent to Cara Fea in an area of corozal bajo, was only partially explored, but six mounds arranged in two patio groups were encountered. "Island" 2, noted briefly in the section on vegetation above, was encountered in an elevated patch of upland forest located along the transect between Yaxha and Nakum. This settlement consists of two definite and two possible structures, along with a chultun. Approximately 200 m southeast of this small group was "Island" 3, with a more formal group of structures. Here, we encountered a group of three rectilinear structures and a fourth pyramidal mound 4 m high. "Island" 4 was a large ridge consisting of two higher areas connected by a slightly lower saddle, a short distance southeast of Cara Fea. A fairly complete, although nonsystematic, investigation revealed 32 mounds, many of them scarred by looters' trenches. Most of these were arranged in eight plaza groups, although six isolated mounds were encountered. The largest mound on this ridge measures 20 m on a side and 6-7 m in height. It is situated on the edge of a plaza along with five smaller mounds. Seven chultuns were associated with the structures. One complete vessel dating to the Late Classic (Tepeu 2) and a nearly complete Late Classic vessel (probably Tepeu 1), were recovered from looters' trenches.4 On "Island" 5,2 km southwest of Poza Maya, we found the remains of six structures. Finally, investigation of a 6th "island" of high land, 500 m long and beginning about 1 km east of Poza Maya, revealed 15 mounds arranged in five patio groups, plus an isolated mound. Here, the terrain rises and falls gradually, with areas of palm bajo lacking evidence

where a group of four structures was encountered. The association between high forest and settlement is obviously not coincidental, and is discussed in more detail below. The vegetation along the transect and the road that is neither palm bajo nor upland forest is scrub bajo of various types. The results of our vegetation survey undermine the notion that bajos are homogenous and uniformly unproductive. Rather, a great variety in vegetation and soil moisture levels can be found within a single bajo. Was this variability exploited by the ancient occupants of the Bajo la Justa? Our archaeological investigations indicate that it almost certainly was. Bajo Communities Scholars have long recognized that many Maya sites are located on the edges of major bajos (Adams 1980; Folan and Gallegos 1998; Gliessman et al. 1983; Lundell 1937). Tikal, for example, is situated at the edge of the large Bajo de Santa Fe, while Calakmul is adjacent to the Bajo El Laberinto. In the Bajo la Justa, Yaxha and Nakum are major sites at the edge of either end of the bajo. We believe that the pattern of sites located on ridges within and on the edges of bajos is a very strong indication oibajo use. Therefore, the BCP's second objective was the testing of this pattern through exploration of the many natural rises, marked by upland forest vegetation, around and within the bajo. We identified these "islands" of higher land in two ways: first, through encounters during the vegetation survey, and second, by identifying patches of high forest in the satellite imagery and then locating these areas through ground checks (see Figure 2). We should note also that while settlements were consistently encountered on each island of higher land we explored, no sites were encountered in the low-lying areas of the bajo, which are susceptible to inundation. Prior to our first field season, one large site in the Bajo la Justa was already well-known and had been investigated by IDAEH archaeologists Claudia Molina and Bernard Hermes. The site, Poza Maya, occupies a large ridge within the bajo, midway between Yaxha and Nakum. A square aguada, 250 m on a side, is located less than 1 km to the southwest of Poza Maya at the juncture of the bajo and the upland ridge upon which the site center is located. This aguada was mapped in 1995 by Rafael Chang, and a canal running around the periphery of the feature was identified. A test trench across this canal confirms its artificial nature, while excavations on the aguada banks reveal these to be constructed as well. The function of the aguada is not clear, however. It may well have served a purpose in flood control or in providing irrigation water to dry season crops (Scarborough and Gallopin 1991). Poza Maya's center consists of

Culture & Agriculture

21

Vol. 22, No. 3

Fall 2000

Figure 4 Total Station Map of Cara Fea

other directions away from Poza Maya descended almost immediately into low-lying terrain, where no further mounds were encountered. Given the scale of architecture at Poza Maya, it is surprising that we did not encounter more evidence of small residential settlements in close proximity to the site center.

of structures alternating with slightly higher ground on which there is high forest mixed with palms. It is in these areas of higher ground that the structures were encountered. These mounds were uniformly small, measuring up to 3 m high. Test excavations in two of these groups, Groups P and R, are discussed in the next section. Informal survey in

Culture & Agriculture

22

Vol. 22, No. 3 Fall 2000

Table 1 Results of Survey of 'Islands' of High Land Associated with the Bajo la Justa 'Island' No. 1 2 3 4 5 6

Location Adjacent to Cara Fea Along Yaxha-Nakum transect Between transect and Cara Fea Southeast of Cara Fea 2 km southwest of Poza Maya 1 km east of Poza Maya

No. Structures

2 patio groups Chultun associated with structures Formal patio group Los Xateros site; mapped and tested in 1999 Recorded, but not mapped Groups P,Q, and R

6 4 4 32 6 15

We do not yet know how to interpret this finding, since Poza Maya was clearly an important site, and its residents presumably would have relied upon a surrounding sustaining area for basic support. It is likely, however, that due to the heavy vegetation some structures were not recorded during survey. In addition, we did not achieve 100 percent coverage, and our results, therefore, must be seen as preliminary. In summary, all of the "islands" of high land surveyed during our two seasons of fieldwork reveal evidence of occupation in the form of structure and chultuns. Importantly, the different bajo settlements encountered throughout our survey evince a diversity of scale, ranging from isolated mounds to small groups of structures, larger formal plaza arrangements, to the medium-sized center of Poza Maya. We believe that this pattern may indicate an organizational structure, perhaps hierarchical, among bajo communities (Lou 1996). This structure would surely include Yaxha and Nakum, the two major centers on the edges of the bajo. We have by no means investigated all of the ridges in and around the Bajo la Justa, however, and our survey must properly be characterized as opportunistic rather than systematic. Neither do we know all we would like about the linkages among the settlements in the Bajo la Justa. We have chosen, therefore, to evaluate each community individually, even as we attempt to address their articulation in a regional system. Understanding the meaning of an organizational structure within the bajo requires an assessment of the chronology, settlement patterns, and artifact assemblages of these communities. Are they the result of a sudden, brief surge of population, or do they represent a long-standing residential presence in the bajo? If of long duration, do these communities change significantly over time? Does each site present evidence for the same or different kinds of productive activities? Do bajo communities differ in their layouts and artifact contents from communities in other areas? Are specialized activities focused upon bajo use present in bajo communities?

Culture & Agriculture

Comments

23

Are indications of these same activities found elsewhere? These are questions that require extensive excavation and comparative work. Our program of test excavations in two very different bajo communities and on one of the raised "islands" has begun to address these issues.

Bajo Community Excavations To gain an understanding of the history and patterning of settlement in the Bajo la Justa, we undertook a program of test excavations at Poza Maya and Cara Fea and in the structures on 'Island' 6. From these excavations an initial reconstruction of the history of the two sites is possible. Ceramic analysis provides preliminary site chronologies, while limited analysis of lithic artifacts provides insight into domestic activities. Archaeobotanical samples are still undergoing analysis; therefore those data are not reported here. Poza Maya It is evident that Poza Maya was an important site, judging from the labor that was invested in the construction of its principal buildings and artificially raised plazas. Located on a ridge in the interior of the bajo midway between Yaxha and Nakum, we hypothesize that it was an administrative center for the utilization of bajo resources. Our excavations focused on groups of small structures on the periphery of the monumental center. Molina's unpublished map of Poza Maya identified two such groups. One group, which she identified as Group H, consists of three structures on a platform, arranged in a C-shape. This group is located immediately east of the site core, at the base of the ridge on which the monumental architecture sits. The second group, unnamed on Molina's map, consists of two structures on a platform, located below the ridge on the southwest side. In addition to these structures, we located another group (Group B), consisting of four small structures arranged in a patio group, and an additional mound separated from the patio group by a plaza (Figure 5). Group B is 65 m southeast of Molina's Group H. Vol. 11, No. 3

Fall 1000

Figure 5 Tape and Compass Maps of Small Residential Groups in the Vicinity of Poza Maya. Groups H and B are immediately east of the site center. Groups P, Q, and R on 'Island' 6 lie at a distance of approximately 1km east of the site center.

D

Group Q

-260 meters-

KEY Sho»«l taat

X

• ^ B

Structure m Op

Q

M /T\

25

0

1

Culture & Agriculture

Plat/arm

Subeparatlan

BA

Chullufi

Test excavations in Group B were placed adjacent to and behind mounds, and in the plaza separating the patio group and the single mound. In addition, we examined the looters' trench in the east mound of Group H. Each of the test units revealed a series of up to seven plastered floors and as much as 3.5 m of fill above bedrock. These excavations, and the looters' trench, produced predominantly Early Classic ceramics (Tzakol complex). Some Preclassic and a few Late

UoUn* trancb

ff.l

METERS

•" ^

(P

f^ORTH

Classic sherds were identified, but not enough to indicate substantial occupation during these time periods. The ceramic evidence of both early (sharp z-angle bowls and scutate lids) and late (basal flange bowls) Early Classic markers indicates lengthy Early Classic occupation of these residential groups on the fringe of the center of Poza Maya.

24

Vol. 21, No. 3 Fall 2000

scarce, found primarily in structure D-2 and in group F shovel tests. Tepeu 1 is very common in groups D and C, while Tepeu 2 material dominates in group A, and is found as well in group C. Insignificant amounts of Tepeu 3 or Postclassic ceramic material was recovered throughout the site, and it seems likely that Cara Fea was essentially abandoned at the end of Tepeu 2 (about A.D. 830).

'Island' 6

Test excavations were also carried out in Groups P and R on 'Island' 6, described earlier in the context of the settlement survey (Figure 5). Test units in Group P included excavations on top of and at the edges of the mounds. These revealed a long occupation of the patio group from possibly the Late Preclassic through the Terminal Classic. Each mound, however, appeared to have a slightly different history. Structure 1 dates to the Late Classic (Tepeu 1 and Tepeu 2), and includes a small amount of Preclassic material. Structure 2 ceramics were a mixture of Early and Late Classic (Tzakol, Tepeu 1, and Tepeu 2). Structure 3 has a clear Terminal Classic (Tepeu 3) occupation, along with Early Classic and Late Classic (Tepeu 1) ceramics. Investigations Group Rconsisted of examination of the back dirt from a looters' trench and the excavation of a test unit in the plaza. Two nearly complete vessels, most likely dating to the Early Classic (Tzakol), were recovered from the back dirt. Sherds from the test unit in the plaza were also entirely Early Classic (Tzakol).

Lithic Analysis

Our preliminary excavations do not allow us to make many statements about the activities carried out at different sites, or at different groups within the same site. Some limited information is available from analysis of the chert artifacts excavated at Poza Maya and Cara Fea. Excavations at Poza Maya yielded 141 pieces of chert debitage, as well as four formal tools. Excavations at Cara Fea yielded 390 pieces of chert debitage, representing the primary, secondary, and tertiary portions of the reduction sequence, and included several probable biface resharpening flakes. Nine tools and several cores and probable hammerstones were also observed. There were no clear qualitative differences in the assemblages over time or between sites, and the collection strategy does not allow quantitative comparison over time and space. The chipped stone assemblages from Cara Fea and Poza Maya probably represent local household production and consumption. Analysis of the lithic assemblages yielded no evidence of specialized tools suggestive of bajospecific activities.

Cara Fea

Test excavations at Cara Fea were carried out in plaza areas, behind mounds, on the edges of structures, and in chultuns (see Figure 4). In addition, we cleaned profiles in looters' trenches in three structures, and screened the back dirt for cultural material. Our excavations yielded Late Preclassic ceramic material in almost all of the groups. Some Early Classic ceramics were recovered, but there is inconclusive evidence of widespread occupation during this time period. Late Classic ceramics, both Tepeu 1 and Tepeu 2, were extremely common and were found in almost all groups tested. Almost no Terminal Classic or Postclassic material was encountered. The information for dating individual structures comes from our work in looters' trenches, since structure excavation was not part of the research design. The three structures that can be dated in this manner are all of different time periods. Structure D-2 was constructed no later than the Early Classic (Tzakol 1-2), structure D-l dates to the Late Classic (Tepeu 1), and structure A-l was constructed later in the Late Classic (Tepeu 2). In addition, the frequency of Late Preclassic ceramics from the looters' trenches in structures A1 and D-2 also suggests the occupation of these groups during this period. Despite the lack of many individual structure dates, shovel tests and test units provide substantial evidence for the timing and length or occupation of various parts of the site. Late Preclassic (Chuen and Cauac in the Tikal ceramic sequence) material was collected predominantly from groups D and F, as well as group C. Tzakol material was

Culture & Agriculture

Settlement Histories

The chronological information derived from excavations at Poza Maya and Cara Fea and on 'Island' 6 (which we consider part of the peripheral settlement around Poza Maya) allows us to make certain statements about the settlement history of these two bap communities. First, the residential groups tested in the vicinity of Poza Maya have very different settlement histories. Group B, a short distance from the site center, was occupied only in Early Classic times. In contrast, Group P, approximately 1 km east of the site center, has a much longer settlement history. Here, occupation ranges from possibly the Late Preclassic through the Late Classic and into the Terminal Classic. This is the only evidence for a Terminal Classic occupation at any of the sites tested. Group R, like Group B, was most likely occupied only in the Early Classic. Cara Fea has yet a different, also lengthy, history of occupation. There is a clear Late Preclassic component at the site, but the evidence of significant Early Classic occupation is less clear. A Late Classic component is unequivocal, but there is no apparent Terminal Classic occupation.

25

Vol. 22, No. 3

Fall 2000

From this chronological information, we conclude that the bajo communities we examined are not ephemeral settlements resulting from brier population explosions, but are a long-standing presence in the bap, lasting several centuries These communities do not remain static, but change over time. Some structures are abandoned, new ones are constructed, and still others are reoccupied. Furthermore, bap communities, while shartnga common pattern of location in relation to the ba , vary dramatically in their size and layout. Poza Maya is a significant regional center, replete with monumental stone architecture. Small residential communities affiliated with the site center are hovve\ er, apparently few in number. In contrast, Cara Fea is a predominantly residential site, much smaller in size Yet it too contains nonresidential structures in the form of small, pyramidal buildings. In turn Cara Fea is larger than most of the other sites which were surveyed, but not excavated. Los Xateros, however, features a pyramidal mound 20m wide and 7m high. Presumably these differences in scale and site design reflect

differences in the roles played by each site in Maya society, although much more research is required before this issue can be addressed in any detail.

Bajo Agriculture During the 1995 season, Dr. Laura Levi noted an area of scrub bap along the Yaxha-N.ikum transect with unusual topography. The area in question was only 1.75 km northot the center of Yaxha. Here, the ground surface undulated, with what appeared to be large, flat surfaces separated by lower areas with a channel-like appearance. We thought that the surface topography might indicate artificial features in the bap, such .is channels cut between planting platforms, and the following year explored this possibility with a series of trenches. We excavated tour trenches perpendicular to the surface features in order to examine their stratigraphy (Figure 6). Two of these trenches were placed parallel to

Figure 6 Photograph of a Bajo Excavation (Op 8A), Northeast Profile.

H Culture & Agriculture

Vol. 22, Na. 3

Fall "000

each other along the same feature in order to test the linear continuity of the features. The trenches revealed that channels had been cut into the underlying sascab. Sometime after this cutting, the channels filled in with water-carried, clay-rich deposits, yielding a very dramatic stratigraphic profile. Each trench featured a wide u-shaped depression, filled with clay varying in color

from light gray to almost black, cut into the decomposing white limestone bedrock (Figure 7). We attribute some of the color variation in the fill to post-depositional staining by organic material (i.e., Stratum IV). The profile of one excavation (Op 8C) indicates a second canal cutting episode (shown in the profile by a dashed line), although rodent disturbance has marred the line. After the BCPs departure

Figure 7 Drawings of Profiles Op 8A (above) and Op 8C (below).

Profile of Operation 8A Northeastern Wall

Profile of Operation 8C Northeastern Wall Strata I, IA,IB. Very dark clay loam. Very high bioturbation and organic content. A horizon. Strata II, HA,MB. Clay. Varies in color from very light gray to nearly black. Stratum III. Sascab. Decomposing limestone bedrock.

centimeters

Culture & Agriculture

Stratum IV. Sascab. Stained slightly darker than Stratum III through organic material carried downward by ground water.

150

17

Vol. 11, No. 3

Fall 1000

from the field in 1996, Vilma Fialko continued investigations and excavated what she characterizes as additional features, possibly canals, similar to the ones just described. We consider these excavations evidence that the Maya were purposefully modifying the bajo. We do not believe that these features can be explained by natural processes, such as the stream action of water. Measurements of the surface topography of the bap surrounding the excavations reveal that local relief is not sufficient to have permitted the cutting action responsible for these channels. Furthermore, the profiles are not consistent with patterned ground, such as gilgae, which yield a wedge-shaped, not a u-shaped, profile. Patterned ground also tends to form a polygonal pattern when viewed in plan, and no polygons were present in the area of the trenches. We cannot yet explain the function of these features, as the excavations we undertook were very limited in extent. These were accomplished by hand digging through very hard clay, and without a backhoe more extensive exploration is difficult. Nevertheless, we feel it is probable that the channels in the bap served as canals associated with agriculture.

that features that may be canals associated with agriculture were constructed within the bajo. Perhaps our most important finding was that the bajo communities we investigated do not fit a model in which population pressure during the Late Classic period forced some Maya to make use of marginal environments, such as bajos, for residential and subsistence needs. Instead, we found that occupation and use of bajos began much earlier, by the Late Preclassic, and extended into the Terminal Classic. We conclude, therefore, that the utilization of bajos was an important subsistence strategy relatively early, and was a significant component of ancient Maya adaptation throughout much of Maya prehistory. Notes

Conclusion The Bap Communities project is designed to address fundamental questions about the subsistence and settlement practices of the ancient Maya. Much debate has surrounded the suggestion that wetland agriculture was a significant component of lowland Maya adaptation. Progress in this debate requires systematic investigation of the utilization of baps in the central Peten. While our results are both preliminary and localized, they are part of a growing body of data indicating that use of seasonal wetlands was indeed a viable stra tegy in environments, such as the centra 1 Peten, in which much of the land is inundated during part of the year. Our project employed a combination of both traditional and innovative research techniques to contribute multiple strands of evidence to the discussion of ancient Maya adaptation. These investigative techniques included the use of satellite imagery and GPS, ecological studies of vegetational communities, archaeological survey and excavation, and artifact and ecofact analyses. We conclude from our research that the Maya did utilize baps. The evidence includes (1) the documentation of the microenvironmental variability within bajos, and the concomitant recognition that certain bap types are amenable to dry season agriculture; (2) the recognition that settlement on ridges of land around and within the bajo is ubiquitous; (3) the discovery that sites associated with the bajo have long histories of occupation, ranging from the Late Preclassic through the Terminal Classic; and (4) evidence

Culture & Agriculture

28

1. These fields were originally identified as raised fields, indicating that they were constructed by the application of soil from adjacent drainage canals and by importation of upland soils. Evidence of soil importation froma distance is lacking, however, indicating that field levels either were not raised or were only raised through the use of canal muck (Turner.1993). The construction of raised fields in the Maya Lowlands similar to the chinampas of central Mexico is open to question, and current terminological practice is to refer to wetland fields as channelized, ditched, or drained. In addition, Harrison (2000; Berry and McAnany 2000) now considers the western arm of Pulltrouser Swamp to be a seasonal swamp, while eastern and southern arms continue to be classified as perennial wetlands. 2. The Bajo Communities Project is co-directed by Dr. T Patrick Culbert of the University of Arizona and Lie. Vilma Fialko, of the Instituto de Antropologia e Historia (IDAEH), Guatemala. Research was conducted as part of the Subproyecto Intersitiosot the larger IDAEH Proyecto Triangulo, directed by Lie. Vilma Fialko. The authors would like to acknowledge IDAEH for its support. The project was funded in 1995 by grants from the Foundation for Advancement of Mesoamerican Studies and the Vice-President tor Research of the University of Arizona, and in 1996 by a grant from the Wenner-Gren Foundation. Project staff, in addition to the authors, included Dr. Laura Levi (University of Texas at San Antonio) in 1995. 3. We acknowledge Dr. Tom Sever of NASA for providing us with TM satellite imagery, and for collaborating on the creation of a correlation between bajo subtypes documented on the ground and the signatures of these subtypes in the remote imagery. In addition, we were joined in our study of the vegetational and topographic variability within the bajo by Sr. Felipe Lanza, a native Petenero and 20 year employee of the Forestry Division of Tikal National Park. Without Sr. Lanza's invaluable assistance, neither our revised bap classification nor our understanding of the agricultural potential of the bajo would have been possible.

Vol. 11, No. 3

Fall 2000

4. This site, later named Los Xateros, was systematically mapped and tested by the BCP in 1999 and 2000, These investigations revealed 115 structures in 27 groups. Test excavations reveal that the site dates from the Late Preclassic to the Early Classic, with evidence for a very small Late Classic occupation.

Culbert, T. Patrick, Laura J. Levi, and Luis Cruz 1990 Lowland Maya Wetland Agriculture: The Rio Azul Agronomy Program. In Vision and Revision in Maya Studies. Flora S. Clancy and Peter D. Harrison, eds. Pp. 115-124. Albuquerque: University of New Mexico Press.

References Cited

Culbert, T. Patrick, and Don S. Rice, eds. 1990 Precolumbian Population History in the Maya Lowlands. Albuquerque: University of New Mexico Press.

Abrams, E. M., and D. J. Rue 1988 The Causes and Consequences of Deforestation among the Prehistoric Maya. Human Ecology 16:377-395.

Dunning, Nicholas P., Vernon Scarborough, T. Patrick Culbert, John G. Jones, Sheryl Luzzader-Beach, and Timothy Beach 2000 The Rise of the Bajos and the Rise of Maya Civilization. Paper presented at the 65th Annual Meeting of the Society for American Archaeology, Philadelphia, PA.

Adams, R. E. W. 1980 Swamps, Canals, and the Locations of Ancient Maya Cities. Antiquity 54:206-214.

Fedick, Scott L. 1996 Settlement Ecology, Wetlands, and Politics in the Yalahau Region of Northern Quintana Roo, Mexico. Paper presented at the 61st Annual Meeting of the Society for American Archaeology, New Orleans.

Kimberly, Berry, and Patricia McAnany 2000 Feeding and Clothing K'axob: Evidence for Wetland Cultivation at a Maya Settlement. Paper presented at the 65th Annual Meetingof the Society for American Archaeology, Philadelphia, PA.

Fedick, Scott L., and Anabel Ford 1990 The Prehistoric Agricultural Landscape of the Central Maya Lowlands: an Examination of Local Variability in a Regional Context. World Archaeology 22(l):18-33.

Bronson, Bennet 1966 Roots and the Subsistence of the Ancient Maya. Southwestern Journal of Anthropology 22:251-279. Bullard, William R. 1960 Maya Settlement Pattern in Northeastern Guatemala. American Antiquity 25:355-372.

Fedick, Scott L., and Kevin Hovey 1995 Ancient Maya Settlement and Use of Wetlands at Naranjal and the Surrounding Yalahau Region. In The View from Yalahau: 1993 Investigations in Northern Quintana Roo. Scott Fedick and Karl Taube, eds. Pp. 7685. Field Report Series No. 2, Latin American Studies Program. Riverside: University of California-Riverside.

Peten,

Cowgill, Ursula M. 1962 An Agricultural Study of the Southern Maya Lowlands. American Anthropologist 64:273-286.

Fialko, Vilma 1999 Recursos hidraulicos en Tikal y sus periferias. Paper presented at the XIII simposio de investigaciones arqueologicas en Guatemala, Guatemala City.

Gilbert, T. Patrick, Vilma Fialko, Brian McKee, Liwy Grazioso, and Julie Kunen 1997 Investigation arqueologica en el Bajo la Justa: La temporada de 1996. In X simposio de investigaciones arqueologicas en Guatemala. Juan Pedro Laporte and Hector Escobedo, eds. Pp.367-372. Guatemala: Museo Nacional de Antropologia e Etnologia.

Folan, William J., and Silverio Gallegos Osuna 1998 Uso del suelo en el estado de Campeche, Mexico y alrededores. In Los investigadores de la cultura maya 5. Pp. 459-478. Universidad Autonoma de Campeche, Campeche, Mexico. 1999 Unas observaciones sobre el uso del suelo del sitio arqueologico deCalakmul, Campeche. In Loscamellones y chinampas tropicales. J. J. Jimenez-Osornio and V. M. Rorive, eds. Pp. 55-68. Merida, Yucatan, Mexico: Universidad Autonoma de Yucatan.

Culbert, T. Patrick, Laura Levi, Brian McKee, and Julie Kunen 1996 Investigaciones arqueol6gicas en el Bajo la Justa entre Yaxha y Nakum. In IX simposio de investigaciones arqueologicas en Guatemala. Juan Pedro Laporte and Hector Escobedo, eds. Pp. 51-57. Guatemala: Museo Nacional de Antropologia e Etnologia. Culbert, T. Patrick, Pamela Magers, and Mara Spencer 1978 Regional Variability in Lowland Maya Agriculture. In PreHispanic Maya Agriculture. Peter D. Harrison and B. L. Turner II, eds. Pp. 157-161. Albuquerque: University of New Mexico Press.

Culture & Agriculture

Gliessman, S. R., B. L. Turner II, F. J. Rosado May, and M. F. Amador 1983 Raised-Field Agriculture in the Maya Lowlands of Southeastern Mexico. In Drained Field Agriculture in Central and South America. J. P. Darch,ed. Pp. 91-110. BAR International Series, 189. Oxford: British Archaeological Series.

29

Vol. 22, No. 3

Fall 2000

G6mez-Pompa, A. 1987 On Maya Silviculture. Mexicanos 3:1-17.

Mexican

McKillop, Heather 1994 Ancient Maya Tree Cropping. Ancient Mesoamerica 5:129-140.

Studies/Estudios

Harrison, Peter D. 1977 The Rise of the Bajos and the Fall or" the Maya. In Social Process in Maya Prehistory: Studies in Memory of Sir Eric Thompson. Norman Hammond,ed. Pp. 470-507. London: Academic Press. 2000 In the Garden of Pleasure and Pain. Paper presented at the 65th Annual Meeting of the Society for American Archaeology, Philadelphia, PA.

Meggers, Betty J. 1954 Environmental Limitations on the Development of Culture. American Anthropologist 56:801-824. Netting, Robert McC. 1977 Maya Subsistence: Mythologies, Analogies, Possibilities. In The Origins of Maya Civilization. R. E. W. Adams, ed. Pp. 299-333. Albuquerque: University of New Mexico Press.

Harrison, Peter D., and B.L. Turner, II, eds. 1978 Pre-Hispanic Maya Agriculture. Albuquerque: University of New Mexico Press.

Palerm, Angel, and Eric Wolf 1957 Ecological Potential and Cultural Development in Mesoamerica. Social Science Monograph, 3. Washington, DC: Pan American Union.

Haviland, William A. 1969 A New Population Estimate for Tikal, Guatemala. American Antiquity 34:429-433.

Pohl, M. D., K. O. Pope, J. G. Jones, J. S. Jacob, R. Piperno, S. D. DeFrance, D. L. Lentz, J. A. Gifford, M. E. Danforth, and K. Josserand 1996 Early Agriculture in the Maya Lowlands. Latin American Antiquity 7:355-372.

Healy, P. L., J. D. H. Lambert, J. T. Arnason, and R. J. Hebda 1983 Caracol, Belize: Evidence of Ancient Agricultural Terraces. Journal of Field Archaeology 4:397-410. Lambert, J. D. H.,and J. T. Arnason 1983 Ancient Maya Drained Field Agriculture: Its Possible Agricultural Productivity at Lamanai, Belize. In Drained Field Agriculture in Central and South America. J. P. Darch, ed. Pp. 111-122. BAR International Series, 189. Oxford: British Archaeological Series.

Pohl, Mary, ed. 1985 Prehistoric Lowland Maya Environment and Subsistence Economy. Cambridge: Harvard University Press. 1990 Ancient Maya Wetland Agriculture: Excavations on Albion Island, Northern Belize. Boulder: Wesrview Press.

Lange, Frederick 1971 Marine Resources: A Viable Subsistence Alternative for the Prehistoric Maya Lowlands. American Anthropologist 73:619-639.

Pope, Kevin O., and Bruce H. Dahlin 1989 Ancient Maya Wetland Agriculture: New Insights from Ecologicaland Remote Se rising Research. Journal of Field Archaeology 16:87-106.

Lanza, Felipe n.d. Informe rrabajosclasificacion de bosques Yaxha-Nakum. Unpublished MS.

Pope, Kevin O., Mary Pohl, John Jacob, John Jones, and Eliska Rejmankova 2000 EcosystemDynamics: The Palaeoecological Imperative in Archaeological Research. Paper presented at the 65th AnnualMeeting of the Society for American Archaeology, Philadelphia, PA.

Lou, Brenda 1996 Chalpate, asentamiento y orientacion de un centro satelite de Tikal. In X Simposio de investigaciones arqueoldgicas en Guatemala. Juan Pedro Laporte and Hector Escobedo, eds. Pp. 373-380. Guatemala: Museo Nacional de Antropologia e Etnologfa,

Pope, Kevin O., Mary D. Pohl, and John S. Jacob 1996 Formation of Ancient Maya Wetland Fields: Natural and Anthropogenic Processes. In The Managed Mosaic: Ancient Maya Agriculture and Resource Use. S. Fedick, ed. Pp. 165-176. Salt Lake City: University of Utah Press.

Lundell, Cyrus L. 1933 The Agriculture of the Maya. Southwest Review 19:65-77. 1937 The Vegetation of Peten. Carnegie Institution, 478. Washington, DC: Carnegie Institution.

Puleston, Dennis E. 1977 The Art and Archaeology of Hydraulic Agriculture in the Maya Lowlands. In Social Process in Maya Prehistory: Studies in Memory of Sir Eric Thompson. Norman Hammond, ed. Pp. 449-467. London: Academic Press.

McAnany, Patricia A. 1998 Where the Water Meets the Land:1997 Excavations in Maya Residences and Wetland Fields, K'axob, Belize. Boston: Boston University, Department of Archaeology and International Programs. Culture & Agriculture

30

Vol. 22, No. 3

Fall 2000

1983 Once Beneath the Forest: Prehistoric Terracing in the Rio Bee Region of the Maya Lowlands. Boulder: Westview Press. 1993 Rethinking the "New Orthodoxy:" Interpreting Ancient Maya Agriculture. In Culture, Form and Place: Essays in Cultural and Historical Geography. Kent Mathewson, ed. Baton Rouge: Louisiana State University Press.

Puleston, Dennis E. 1982 The Role of Ramon in Maya Subsistence. In Maya Subsistence: Studies in Memory of Dennis E. Puleston. Kent V. Flannery, ed. Pp. 353-366. New York: Academic Press. Reina, Ruben E. 1967 Milpas and Milperos: Implications for Prehistoric Times. American Anthropologist 69:1-20.

Turner, B. L. II, and Peter D. Harrison, eds. 1983 Pulltrouser Swamp. Austin: University of Texas Press.

Scarborough, Vernon L., and Gary G. Gallopin 1991 A Water Storage Adaptation in the Maya Lowlands. Science 251:658-662.

Wilken, Gene C. 1971 Food-Producing Systems Available to the Ancient Maya. American Antiquity 36:432-448.

Siemens, Alfred H. 1982 Pre-Hispanic Use of Wetlands in the Tropical Lowlands of Mesoamerica. In Maya Subsistence: Studies in Memory of Dennis E. Puleston. Kent V. Flannery, ed. Pp. 205-225. New York: Academic Press.

Willey,G. R.Jr., W. R. BullardJ. B. Glass, and J. C. Gifford 1965 Prehistoric Maya Settlements in the Belize Valley. Papers of the Peabody Museum of Archaeology and Ethnology, 54. Cambridge: Harvard University Press.

Siemens, Alfred H.,and Dennis E. Puleston 1972 Ridged Fields and Associated Features in Southern Campeche: New Perspectives on the Lowland Maya. American Antiquity 37:228-239.

Wiseman, Frederick M. 1978 Agricultural and Historical Ecology of the Maya Lowlands. In Pre-Hispanic Maya Agriculture. Peter D. Harrison and B. L. Turner II, eds. Pp. 63-115. Albuquerque: University of New Mexico Press.

Steward, Julian, ed. 1955 Irrigation Civilizations: A Comparative Study. Social Science Monograph, 1. Washington, DC: Pan American Union.

Wittfogel, Karl 1957 Oriental Despotism: A Comparative Study of Total Power. New Haven, CT: Yale University Press.

Turner, B. L. II 1974 Prehistoric Intensive Agriculture in the Maya Lowlands. Science 185:118-124. 1978 The Development and Demise of the Swidden Thesis of Maya Agriculture. In Pre-Hispanic Maya Agriculture. Peter D. Harrison and B. L. Turner II, eds. Pp. 13-22. Albuquerque: University of New Mexico Press.

Culture & Agriculture

Wright, A. C. S., D. H. Romney, R. H. Arbuckle, and V. E. Vial 1959 Land Use in British Honduras. London: Her Majesty's Stationary Office.

31

Vol. 22, No. 3

Fall 2000