A reconstruction of Middle Preclassic Maya subsistence economy at Cahal Pech, Belize.

Powis, Terry G. ; Stanchly, Norbert ; White, Christine D. 等

Introduction

Recent research has revealed that many of the traditional cultural hallmarks of Classic Maya civilization (AD 250-900) seem to have had their origins in the preceding Preclassic or Formative period (2000 BC-AD 250). In particular, there are strong indications that the Maya made the transition from a relatively egalitarian to ranked and stratified society during this early period of cultural development (Adams & Culbert 1977; Awe 1992; Hammond 1992; Healy & Awe 1995b). It is now generally accepted that it was during the latter half of the Middle Preclassic (Mamom phase 650-300 BC) that several of the diagnostic traits of complex culture were established (Hammond 1986: 403; Sharer 1992: 131).

An integral component in studies of increasing cultural complexity is the subsistence economy which supported that development. At present, knowledge of late Middle Formative resource utilization and the extent of organized subsistence economies across the Maya lowlands remains limited. With the notable exceptions of Cuello (Miksicek 1991; Wing & Scudder 1991), Cerros (Carr 1986; Cliff & Crane 1989) and Colha (Carr 1985; Shaw 1991), few sites have yielded substantial subsistence data dating to this period.

In the upper Belize River Valley region, located in west-central Belize [ILLUSTRATION FOR FIGURE 1 OMITTED], investigations into Preclassic subsistence economies have only recently begun. Settlement and midden densities collected from the region suggest that an extensive swidden agricultural system (slash-and-burn) was being practiced in restricted areas during the Middle Formative period (Fedick 1989: 240). Relatively little research has focused on the actual cultigens exploited by the Formative inhabitants of the Belize Valley, and there remain significant questions about the degree of importance of maize (versus other plant and animal foodstuffs) in the diet at this early date (Healy & Awe 1995b: 7). The non-plant component of the Preclassic diet also needs to be more completely researched.

This study represents an attempt to provide more substantive data regarding animals and plants used by the Middle Formative Maya of the Belize Valley. Combining the animal and plant remains with analysed isotopic data from Middle Preclassic contexts at Cahal Pech provides an invaluable opportunity to reconstruct and illuminate patterns of procurement, consumption, trade and social and ritual use of specific types of foods. The recovery of these remains helps to address not only questions about intra- and inter-site resource utilization within the valley itself, but also adds to the understanding of subsistence strategies throughout the Maya lowlands for this early time period.

Archaeological investigations at the site of Cahal Pech have focused considerable attention on Formative period occupation levels (Awe 1992; Healy & Awe 1995b; 1996; see also Ball & Taschek 1991). One of the long-term objectives of this research was reconstructing subsistence practices throughout the Formative period. The exploitation of animal and plant resources by the Preclassic Maya was of specific concern to the present study. Excavations in the Cahal Pech site core and its peripheral settlement clusters (Cas Pek, Tolok and Zotz groups) have yielded over 20,000 faunal remains, constituting one of the largest assemblages yet recovered from a lowland Maya site (Stanchly 1995: 125). Several recognized Mesoamerican cultigens and economically valuable tree species have been identified in the Cahal Pech archaeobotanical collection (Wiesen & Lentz 1997). Carbonized remains from the site core have also yielded some of the earliest securely dated maize cupules and textile-impressed plaster fragments, probably produced from cotton, in the Maya lowlands (Lawlor et al. 1995: 157-62).

Background

The medium-sized Maya site of Cahal Pech is situated on an imposing acropolis which overlooks the Belize River [ILLUSTRATION FOR FIGURE 2 OMITTED]. The site core consists of 34 structures arranged around seven plazas. Investigations in Plaza B of the site core have documented evidence that suggests the site was initially settled by the end of the Early Preclassic (1200-900 BC) period (Awe 1992:113). Occupation in the site's periphery began during the late Middle Preclassic (650-300 BC) with a number of small, dispersed communities being established on hilltops within a 2-km radius of the site core. Some of the best evidence of animal and plant remains from these early contexts at Cahal Pech comes from the peripheral settlement cluster known as the Tolok group [ILLUSTRATION FOR FIGURE 3 OMITTED]. The Tolok group is a large settlement group located approximately 500 m to the southeast of the site core. It consists of 16 mounds which are situated on top of a long, asymmetrically shaped ridge (Powis 1996: 34). Investigations into the Formative occupation levels at Tolok have revealed an abundance of faunal and floral material, especially from a deeply buried midden deposit inside Structure 1.

Structure 1 Investigations at Tolok

Excavations at the interior base of Structure 1 revealed the presence of an undisturbed midden deposit. The midden was found inside an abandoned lateral-shaped chultun (Chultun D), a subterranean chamber cut into bedrock and used by the Maya for storage and/or water collection (Powis 1996: 39). Middle Preclassic deposits such as this have also been found at the site of Cuello (Hammond et al. 1995: 124-5). The overall dimensions of Chultun D were approximately 3-1 m long x 2.5 m wide x 1.1 m deep. The Structure 1 midden was partly filled in with a 60-cm layer of charcoal-rich rubbish that included over 2000 large and unabraded sherds of vessels belonging to the Chunhinta, Jocote, Joventud, Savana and Sayab groups (see Gifford 1976). On the basis of associated ceramics from other valley sites, the midden deposit has been firmly dated to the latter half of the late Middle Formative period (450-300 BC). One charcoal sample was collected from the midden, and sample produced a radiocarbon date of (Beta-77199) 2220[+ or -]150 BP, which has a one-sigma calibrated age range of 390-150 cal BC (Healy & Awe 1995c: 199-200).

TABLE 1. List of identified taxa recovered from the Structure 1
midden deposit at the Tolok group, Cabal Pech, Belize.

scientific name common name

invertebrates

Pelecypoda bivalves
Nephronaias sp. freshwater pearly oyster
Gastropoda univalves
Pomacea flagellata apple snail (freshwater)
Pachychilus glaphyrus jute snail (freshwater)
Pachychilus indiorum jute snail (freshwater)
Strombidae conchs (marine)
Prunum sp. marginellas (marine)
Oliva sp. olive shells (marine)
Scaphopoda tusk shells
Dentalium sp. dentalium (marine)
Brachyura true crabs

vertebrates

Osteichthyes bony fish
Scaridae parrotfish (marine)
Sparisoma sp. parrotfish (marine)
Lachnolaimus sp. hogfish (marine)
Epinephalus sp. grouper (marine)
Lutjanidae snapper (marine)
Siluriformes catfish (freshwater)

Reptilia reptiles

Chelonia turtles (freshwater)
Iguanidae iguanas

Aves birds

Meleagridae turkeys
Crax rubra great currasow

Mammalia mammals

Didelphis marsupialis common opossum
Dasypus novemcinctus nine-banded armadillo
Sylvilagus sp. forest rabbit
Rodentia rodents
Agouti paca paca
Canis familiaris domestic dog
Odocoileus virginianus white-tailed deer
Mazama americana red brocket deer

Faunal remains

A total of 5674 bone and shell remains were recovered from the midden deposit, representing local terrestrial, freshwater, avian and domestic species, as well as marine species (TABLE 1). Mammal, fish, bird, reptile, univalve and bivalve species have been identified (TABLES 2 & 3). To date, 2871 (or 50.6% of the total assemblage) have been identified to the level of zoological order or lower taxon. All of the invertebrate remains have been identified to order or lower taxon while only 74 bones (or 2.3% of the vertebrate sample) remains have been identified below the class level.

Although only a small proportion of the assemblage has been identified, preliminary analysis of the material indicates that the Middle Formative Maya of the Tolok group exploited various niches within their local environment and enjoyed access to non-local animal resources, including reef fishes and shellfish species. The exploitation of a wide variety of resources is a familiar occurrence in Formative Maya faunal assemblages and is a reflection of the inhabitants' familiarity with the diversity present in tropical ecosystems (Carr 1985; Hamblin 1984; Pohl 1985a; Shaw 1991; Stanchly 1995; Wing & Scudder 1991).

Terrestrial animals

Local terrestrial species are represented by both white-tailed deer (Odocoileus virginian us) and the smaller red brocket deer (Mazama americana), agouti (Agouti paca), armadillo (Dasypus novemcinctus), rabbit (Sylvilagus sp.), domestic dog (Canis familiaris), opossum (Didelphis marsupialis), and small rodents (Family Rodentia) (TABLE 2). With the possible exception of the rodents, all of the above species would have been consumed by the Maya. Avian species identified include turkey (Family Meleagrididae) and currasow (Crax rubra). The majority of the reptile remains belong to an as yet unidentified species of turtle, while some lizards (cf. iguana) are present.

Marine fish and shellfish

Fish remains include both local freshwater fish such as catfish (Family Siluriformes) and several marine reef fishes including parrot fish (Family Scaridae), hog fish (Family Labridae), grouper (Family Serrenidae), and snapper (Family Lutjanidae). The presence of marine fish-skull elements, some of which exhibited evidence of heat alteration, suggests that they were imported whole, processed and consumed on site. This also suggests that many of the postcranial elements present could represent marine species as well. However, the majority of these are considered to be non-diagnostic elements (e.g. vertebrae, spines) and therefore identification to the species level will not be possible. The presence of marine fishes can only be attributed to trade or direct exploitation of the Caribbean coastal environment by the Cahal Pech Maya, as the site lies some 110 km from the Caribbean coast.

Marine shell remains identified include queen conch (Strombus gigas), tusk shells (Dentalia sp.), olive shells (Oliva sp.) and marginella shells (Prunum sp.). Two crab claws (Brachyura) were also noted (TABLE 3). All parts of the conch shell were found in the midden assemblage. The recovery of lip, shoulder, spine and columella fragments indicates that some specimens were imported whole. This marine shell species may have been consumed and/or used by the Maya in the manufacture of shell ornaments. The presence of Middle Preclassic shell ornament production has been identified in the Cas Pek group at Cahal Pech (Lee & Awe 1995) as well as at Pacbitun, located approximately 12 km south of Cahal Pech (Hohmann & Powis 1996).

Fresh water shellfish

Freshwater snails and clams account for half (49.3%) of the faunal assemblage and include at least two species of the jute snail (P. glaphyrus and P. indiorum), the apple snail (Pomacea flagellata) and local river clam (Nephronaias sp.) (see TABLE 3). The recovery of these broken, discarded shellfish remains in the midden may indicate that they were an important food source for the site's inhabitants during the Formative period. Indeed, excavations at Pacbitun have showed high concentrations of both freshwater snails and clams. Investigations in Plaza B of the site core have resulted in the recovery of more than 230,000 freshwater mollusca dating to the Middle Formative period (Hohmann & Powis 1996: 121). Large quantities of freshwater shell species, particularly Nephronaias sp., have also been recovered from Preclassic contexts at the sites of Blackman Eddy (n = 21,000) (Garber et al. 1997: 15) and Barton Ramie (n = 765) (Willey et al. 1965: 504-7). It has been suggested by researchers that these freshwater shellfish served a dual role in Formative Maya society as both a dietary supplement and ritual item (Healy et al. 1990: 180; Moholy-Nagy 1978; Pohl 1985b: 109; Powis 1997).

Archaeobotanical remains

Five soil samples (8-10 litres each) were collected from the Structure I midden. These samples were subjected to water flotation analysis. The carbonized plant remains recovered (total weight = 6.0 g) were delivered to the archaeobotany lab at the New York Botanical Garden for analysis. The identification of charred seeds, fruit rinds, corn-cob fragments, wood charcoal fragments and other non-woody plant parts suggests that the late Middle Formative residents of Cahal Pech, including the Tolok group, were agriculturalists who cultivated and harvested a number of local plant communities for both fuel and construction purposes (Wiesen & Lentz 1997). The diversity of cultigens and timber products represented in the Cahal Pech assemblage is reflected, with few exceptions, in the Structure I midden at the Tolok group (TABLE 4).

TABLE 2. List of vertebrate frequencies identified in the Structure
1 midden at the Tolok group.

taxon NISP MNI %

Osteichthyes 2072 13 36.5
Sparisoma sp. 16 9
Lachnolaimus sp. 1 1
Epinephalus sp. 1 1
Lutjanidae 1 1
Siluriformes 3 1
unidentified fish 2050 -
Reptilia 19 3 0.3
Chelonia 3 1
Iguanidae 2 1
Sauria 1 1
unidentified reptile 13 -
Aves 37 2 0.7
Meleagridae 1 1
Crax rubra 1 1
unidentified bird 35 -
Mammalia 423 13 7.5
Odocoileus virginianus 15 2
Mazama americana 1 1
Agouti paca 6 2
Didelphis marsupialis 8 2
Didelphidae 1 1
Dasypus novemcinctus 3 1
Rodentia 5 2
Canis familiaris 3 1
Sylvilagus sp. 2 1
unidentified mammal 379 -
class unknown 326 - 5.7

totals 2877 62 50.7
TABLE 3. List of invertebrate frequencies identified in the
Structure 1 midden at Tolok.

taxon NISP MNI %

Pelecypoda 2143 923 37.8
(bbb)
Nephronaias sp. 2143 923
Gastropoda 631 570 11.1
Pachychilus glaphyrus 300 298
Pomacea flagellata 159 120
Pachychilus indiorum 117 114
Pachychilus sp. 29 29
Strombidae 22 5
Scaphopoda 19 - 0.3
Dentalium sp. 19 -
Prunum sp. 3 3
Oliva sp. 1 1
Brachyura 4 2 0.1

totals 2797 1495 49.3
TABLE 4. List of palaeobotanical remains identified in the Structure
1 midden at Tolok.

sample no. taxon/part weight (g)

CP-9450-004 Pinus sp. charcoal .23
 hardwood charcoal .14
 spermatophyte tissue .21
 unknown rind .02

CP-9450-007 Astronium sp. charcoal .44
 Cucurbita sp. rind 1.15
 Pinus sp. charcoal .65
 cf. Psidium guajava rind .08
 hardwood charcoal [less than].01
 unknown bark [less than].01

CP-9450-006 Aspidosperma sp. charcoal .16
 cf. Nectrandra sp. charcoal .01
 Pinus sp. charcoal 2.31
 Zea mays kernels .05
 hardwood charcoal .26

CP-9450-009 Pinus sp. charcoal .30
 hardwood charcoal .06

total 6.00

Economic species

Two traditionally recognized Mesoamerican cultigens, maize (Zea mays) and squash (Cucurbita sp.), were identified in the Tolok midden. Despite their low occurrence, particularly maize ([less than]0.01 g), they have been recovered from other Middle Preclassic contexts in the group. The early presence of these remains at Tolok and elsewhere in the Cahal Pech area is not surprising as they are known to have been important staples of the Maya diet during the Classic and Postclassic periods. Another Maya staple, the common bean (Phaseolus sp.), was not recovered from the Tolok midden. However, it has been found in Structure B-4 3-Sub, an early Late Formative structure (300-100 BC) located in the site core of Cahal Pech (Wiesen & Lentz 1997: 13).

Other edible or otherwise utilized domestic and wild species of plant remains recovered from the Tolok midden include wild fig (Ficus sp.) charcoal fragments, a guava fruit cast (cf. Psidium guajava), and a number of identifiable wood charcoals such as pine (Pinus sp.), glassy wood (Astronium graveolens), malady (Aspidosperma sp.) and aguacatillo (Nectandra sp.) (see TABLE 4). Pine charcoal represents the most frequently identified species of the entire carbonized plant remains collection. This pattern is repeated at other Maya sites such as Cuello (Miksicek 1991: 75-6) and Yarumela (Lentz et al. 1997: 70), suggesting a widely distributed reliance on pine. Other macro-remains possibly used as timber sources were the three large deciduous forest species called glassy wood, aguacatillo and malady. Importantly, the only other previous recording of the use of malady in the Maya area was at the site of Ceren, El Salvador, where it was determined to have been part of a collapsed roof structure (Wiesen & Lentz 1997: 11; Lentz 1996: 257).

Missing species

Although a variety of carbonized plant remains have been identified in the Tolok midden, there is an absence of certain plant species (e.g. ramon, coyol palm, cotton) and root crops (e.g. manioc, malanga) that have been recovered from other Maya sites (see Hather & Hammond 1994; Webster et al. 1997: 58). This absence may be the result of various taphonomic processes and/or recovery techniques. Missing species such as ramon (Brosimum alicastrum), coyol palm (Acrocomia aculeata) and cotton (Gossypium sp.) have been identified elsewhere in Middle Formative occupation levels at Cahal Pech (Lawlor et al. 1995: 157; Wiesen & Lentz 1997: 14-16).

Isotopic evidence from human bone

Animals and plants recovered from archaeological sites denote possible ancient menus, but by themselves they cannot be used to reconstruct actual food consumption. It is important to know which available resources were selected for consumption by the Formative Maya and which were consumed consistently enough, and in quantities large enough, to have been nutritionally significant. Isotopic analysis of human bone provides a direct means of bridging the distance from menu to meal.

The isotopic composition of foods expected to be of greatest importance at Tolok, and to the Maya in general, is shown in FIGURE 4. Most of the cultigens and wild plants in the Maya area use a C3 photosynthetic pathway and have [Delta]13C values close to -27[per thousandth]. Significantly, maize uses a C4 pathway and has a [Delta]13C value closer to -9.5[per thousandth] (Schwarcz et al. 1985). The only other C4 plants possibly available to the Maya which might confound the maize signature would be amaranth (amaranthus spp.), epazote (Chenopodium ambrosoides) and the CAM plants; nopal cactus (Opuntia), and pinuela (Bromelia karatas) (Wright & White 1996). None of these, however, are expected to have been consumed in any significant quantity by the Maya. Archaeological plants will have [Delta]13C values which are 1.5[per thousandths] higher than the modern plants on which these averages are based, because photosynthetic carbon is derived completely from atmospheric C[O.sub.2], and fossil fuel burning has produced a decrease in the [Delta]13C of atmospheric carbon (Marino & McElroy 1991).

The plant values at the base of the food chain are passed on with a systematic fractionation of about 5[per thousandth] to their consumers. Thus, the [Delta]13C of human bone collagen will also include the [Delta]13C of consumed animals. Collagen is thought to represent the total protein component of the diet (i.e. plant plus animal protein) (Ambrose & Norr 1993). Most terrestrial animals in the Maya area consume C3 plants, but deer, peccary and dog, may have consumed significant quantities of maize from raiding plots, scavenging or purposeful feeding. There is intersite variability in the degree to which these animals consumed maize (Reed 1994; Tykot et al. 1996; van der Merwe et al. 1994; White & Schwarcz 1989; White et al. 1993; White et al. 1997). Marine fish and marine molluscs also have signatures that are more C4-like than their freshwater counterparts. Stable nitrogen isotope ratios can be used to distinguish the consumption of marine fish from C4 plants, and freshwater fish from C3 plants (e.g. root crops). [TABULAR DATA FOR TABLE 5 OMITTED] However, shellfish consumption is more difficult to identify specifically. Freshwater shellfish fall within the range of C3 plants, and reef shellfish overlap considerably with C4 plants. This problem is further complicated by the fact that the marine food web for Belize and the Caribbean has markedly different isotopic values from marine webs in other parts of the world (Keegan & DeNiro 1988; Tykot et al. 1996; van der Merwe et al. 1994).

With the exception of nitrogen-fixing plants such as beans, plant types are not distinguishable by their nitrogen isotope ratios. Nitrogen isotopes are, however, useful in determining food-chain levels of resources because 15N is fractionated at each trophic level. Particular to the Maya area, freshwater fish are the most enriched in 15N, followed in order by marine fish and terrestrial animals.

Sample and method

Stable carbon and nitrogen isotope ratios of bone collagen were analysed for four individuals (Burials 7-10) recovered from the Tolok group (TABLE 5). The burials were intrusive into Structure 14, a late Middle Preclassic round building which is contemporaneous in date with the Structure 1 midden (Powis 1996: 54-7, 77-84). 'Collagen' was extracted from crushed bone using an adaptation of the Longin (1971) procedure to maximize yield (White et al. 1993). Reproductability on the mass spectrometer is generally better than [+ or -]0.1[per thousandth] for duplicate samples.

Sample integrity was assessed using 'collagen' yields and C/N ratios (see TABLE 5). Due to poor preservation, collagen yield is expectedly low (mean = 2.88% [+ or -] 1.8). Although there is no statistically significant correlation between yield and either [Delta]13C values or [Delta]15N values, the sample size is small. C/N ratios falling within the range of 2.9 to 3.7 (DeNiro 1985) generally indicate preservation of the primary values. The mean for Tolok C/N ratios (3.5[+ or -]0.3) is within that range, but the low yield, high C/N ratio and high [Delta]13C value for Burial 9 suggest that this individual may have been diagenetically altered and it is, therefore, excluded from the analysis.

Results

The mean [Delta]13C value for the Tolok group is - 12.8 [+ or -] 1.3[per thousandth], which is very close to the general Preclassic values at the Belizean sites of Lamanai (-12.7[per thousandth]) and Cuello (-13.0[per thousandth]) (Tykot et al. 1996; van der Merwe et al. 1994; White & Schwarcz 1989). Preclassic samples from Seibal (-9.6[per thousandth]) and Altar de Sacrificios (-10.4[per thousandth]) (Wright 1994) have higher values suggesting significantly greater maize consumption at other contemporaneous sites (TABLE 6 & [ILLUSTRATION FOR FIGURE 4 OMITTED]). The [Delta]13C results therefore seem to support archaeological evidence modelling a broad subsistence base where maize is an important component of the diet, but does not dominate it. Within the Tolok sample, maize consumption in the earlier time period (late Middle Formative period) appears to be markedly lower (-13.8[per thousandth]) than in the later (Protoclassic period) time period (-12.1[per thousandth]). This might denote an increase of maize production after 400 BC and a corresponding shift to a more limited subsistence base (White et al. 1996: 8-9). Notably, however, these temporal patterns [TABULAR DATA FOR TABLE 6 OMITTED] should be further tested with more samples dating to this time period.

Previous isotopic analysis on the Cahal Pech skeletal assemblage has also documented spatial differences in maize consumption (White et al. 1996). The peripheral Tolok settlement group appears to be consuming less maize than the more central and higher status Cas Pek and Zotz groups, which have [Delta]13C values of -10.6 [+ or -]1.0[per thousandth], and -10.6 [+ or -] 0-6[per thousandth] respectively [ILLUSTRATION FOR FIGURE 5 OMITTED] (White et al. 1996: 6-9). As craft specialists, the Cas Pek residents may have experienced higher status. If so, the dietary distinction might indicate that maize was socially valued during the Formative period. The [Delta]15N values (mean = 9.5[per thousandth]) of the Cas Pek group suggest the consumption of reef fishes here was even greater than at the Tolok group (White et al. 1996).

The combined [Delta]15N and [Delta]13C values in the Tolok sample indicate that meat consumption was a mixture of terrestrial herbivores, reef fish and possibly some freshwater fish. The isotopic data seem to support a model derived from the faunal analysis of a mixed subsistence base in which terrestrial herbivores and reef fish are important; however, they do not indicate that either freshwater shellfish or fish were consistently consumed in quantities large enough to register an isotopic distinction in bone. Compared to other Preclassic samples, protein resources consumed by the Tolok group were similar to those at Cuello (8.9 [+ or -] 0.6[per thousandth]) (van der Merwe et al. 1994) and Altar de Sacrificios (8-4 [+ or -] 0.7[per thousandth]) (Wright 1994), but more freshwater fish were consumed at Lamanai (10.2[per thousandth]) (White & Schwarcz 1989) and Seibal (9.7 [+ or -] 0.8[per thousandth]) (Wright 1994). Domesticated dog remains have also been found in greater quantities at other Preclassic sites. For example, dog remains at Tolok constituted less than 1% of the total Preclassic faunal assemblage compared to 7% at Cuello (Wing & Scudder 1991: 88-95) and 34% at Dzibilchaltun (Wing 1975: 381). Isotopically, it is difficult to determine whether dogs, specifically, were consumed. It is possible to talk about the general source of protein (from [Delta]15N) which may have included dogs, but there is no way of knowing whether it did or not. And if dogs were consumed and they were fed maize, they could also be contributing to the maize consumption indicated by the [Delta]13C values.

Although the sample in this study is small (3 burials), when compared with other sites these data increase knowledge about geographical similarity and variability in diet. In general, the isotopic values support faunal, botanical and archaeological evidence of a broad subsistence base which includes maize and significant quantities of long-distance reef resources provided on a regular, long-term basis (White et al. 1996: 10). The temporal difference in [Delta]13C within the Tolok group suggests a possible subsistence shift during the Formative period. In addition, the comparison of the Tolok group diet with that of other settlement groups from the Middle to the Late Formative period may indicate differentiation by social status within the Cahal Pech community.

Conclusion

The information gleaned from the Structure 1 midden in the Tolok group provides significant new data on Middle Formative Maya diet and subsistence practices. Preliminary results of the analysis of the animal and plant remains mirror a fairly common pattern seen in Maya archaeofaunas and archaeobotanical remains; that is, the utilization of a wide variety of resources reflective of the diversity seen in tropical ecosystems.

This analysis of animal and plant remains from Cahal Pech has documented that the Middle Formative Maya were agriculturalists whose diet consisted, in part, of such cultivated plants as maize, beans and squash. The products of the coyol palm, as well as the ramon and fig trees, also contributed to Formative Maya food preparations. Their diet was further supplemented by a mixture of terrestrial herbivores (e.g. deer, agouti), marine reef fish (e.g. parrotfish, grouper) and small quantities of freshwater fish (e.g. catfish) and shellfish (e.g. river snail and clam).

The identification of marine fishes and other marine resources in Preclassic Maya faunal assemblages is not in itself rare (Carr 1985; Shaw 1991; Wing & Scudder 1991); however, most of these assemblages have been recovered from sites on or near the Caribbean coast where procurement of marine resources would have been more easily facilitated. The presence of Caribbean reef fishes found so far inland at this early date suggests that the Maya were able to preserve these fishes so that they would not spoil during the 110-km trek up river. We suggest that the Preclassic Maya may have been salting or smoking marine fishes for inland transport. Salt-making communities dating to the Late to Terminal Classic have been identified along the Caribbean coast of Belize (MacKinnon & Kepecs 1989; McKillop 1995: 225). The presence of marine fishes at several inland sites found in Formative contexts may indicate that salt-making technology was known by the Preclassic Maya. Although no evidence for Preclassic salt-making communities have been found along the Belizean coast, this may be because such communities may now be inundated by rising sea levels (Dunn & Mazzullo 1993:123). An alternative method of inland transport may have been by keeping live fish in canoes partly filled with saltwater from the sea (Norman Hammond pers. comm.).

Regardless of the method of transport, the presence of marine fishes at Tolok has far-reaching implications for an understanding of not only Formative period subsistence patterns, but also of trade networks and coastal exploitation by the Maya. Moreover, pertinent information regarding ancient Maya technology, that is marine fish procurement and processing techniques (e.g. salt drying or smoking), will also be gained from analysis of the presence of reef fishes in the Cahal Pech assemblage. What is not clear from the analysis of the midden material is the exact nature of access to marine resources by the Preclassic Tolok inhabitants. However, it is certain that these inhabitants had sustained access to coastal resources during both the Middle and Late Formative periods by direct exploitation or through trading networks.

Further analysis and interpretation of this material will allow archaeologists to attain a better understanding of the role both fauna and flora played in Formative period Maya society. It can also provide insights into the development of certain trends, such as the differential access to resources by certain privileged members of Maya society which existed by the Classic period.

Acknowledgements. Investigations at Cahal Pech were licensed by the government of Belize through the Department of Archaeology. Funding for research was provided by the Social Sciences and Humanities Research Council (SSHRC) of Canada. Particular thanks are extended to Bobbi Hohmann and Sam Wilson for editorial suggestions and to Andrew Allan for producing the computer graphics. Identification of the faunal assemblage has been made with the aid of the skeletal reference collections housed in the Department of Paleobiology, Royal Ontario Museum. We thank Kevin Seymour for allowing us access to this collection. David Cruz, Carolyn Doherty and Karen Kisiel are also thanked for their assistance with the zooarchaeological analysis. Kim Law is thanked for processing and collection of the isotopic data. We want to express our special appreciation to David Lentz and Anne Wiesen of the New York Botanical Garden for their analysis of the palaeobotanical samples.

References

ADAMS, R.E.W. & T.P. CULBERT. 1977. The origins of civilization in the Maya Lowlands, in R.E.W. Adams (ed.), The origins of Maya civilization: 3-24. Albuquerque (NM): University of New Mexico Press.

AMBROSE, S.H. & L. NORR. 1993. Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate, in J. Lambert & G. Grupe (ed.), Prehistoric human bone: Archaeology at the molecular level: 1-37. Berlin: Springer-Verlag.

AWE, J.J. 1992. Dawn in the land between the rivers: Formative occupation at Cahal Pech, Belize and its implications for Preclassic development in the central Maya Lowlands. Unpublished Ph.D dissertation, University of London.

BALL, J.W. & J.T. TASCHEK. 1991. Late Classic Lowland Maya political organization and central-place analysis: new insights from the upper Belize Valley, Ancient Mesoamerica 2(2): 149-65.

CARR, H.S. 1985. Subsistence and ceremony: Faunal utilization in a Late Preclassic community at Cerros, Belize, in Pohl (1985a): 115-32.

1986. Preliminary results of analysis of fauna, in R.A. Robertson & D.A. Freidel (ed.), Archaeology at Cerros, Belize, Central America, volume 1: An interim report: 127-46. Dallas (TX): Southern Methodist University Press.

CLIFF, M.B. & C.J. CRANE. 1989. Changing subsistence economy at a Late Preclassic Maya community, in P. McAnany & B. Isaac (ed.), Research in economic anthropology: Prehistoric Maya economies of Belize: 295-324. Greenwich (CT): JAI Press Inc.

DENIRO, M.J. 1985. Post-mortem preservation in in-vivo bone collagen interpretations in relation to paleodietary reconstruction, Nature 317: 806-9.

DUNN, R.K. & S.J. MAZZULLO. 1993. Holocene paleocoastal reconstruction and its relationship to Marco Gonzalez, Ambergris Caye, Belize, Journal of Field Archaeology 20: 121-31.

FEDICK, S.L. 1989. The economies of agricultural land use and settlement in the upper Belize Valley, in P. McAnany & B. Isaac (ed.), Research in economic anthropology: Pre-historic Maya economies of Belize: 215-53. Greenwich (CT): JAI Press Inc.

GARBER, J.F., M.K. BROWN & C.J. HARTMAN. 1997. Excavations on Structure B1 at Blackman Eddy: results of the 1996 field season, in J.F. Garber & D.M. Glassman (ed.), The Belize Valley archaeology project: Results of the 1996 field season: 6-33. San Marcos (TX): Southwest Texas State University.

GIFFORD, J.C. 1976. Prehistoric pottery analysis and the ceramics of Barton Ramie in the Belize Valley. Cambridge (MA): Harvard University, Peabody Museum of Archaeology and Ethnology. Memoir 18.

HAMBLIN, N.L. 1984. Animal use by the Cozumel Maya. Tucson (AZ): University of Arizona Press.

HAMMOND, N. 1986. New light on the most ancient Maya, Man (n.s.) 21: 399-413.

(Ed.). 1991. Cuello: An early Maya community in Belize. Cambridge: Cambridge University Press.

1992. Preclassic Maya civilization, in E.C. Danien & R.J. Sharer (ed.), New theories on the ancient Maya: 137-44. Philadelphia (PA): University of Pennsylvania, University Museum. Monograph 77.

HAMMOND, N., A. CLARKE & S. DONAGHEY. 1995. The long goodbye: Middle Preclassic Maya archaeology at Cuello, Belize, Latin American Antiquity 6(2): 120-28.

HATHER, J.G. & N. HAMMOND. 1994. Ancient Maya subsistence diversity: root and tuber remains from Cuello, Belize, Antiquity 68: 330-35.

HEALY, P.F. & J.J. AWE (ed.). 1995a. Belize Valley Preclassic Maya project: Report on the 1994 field season. Peterborough (ONT): Trent University. Occasional Paper in Anthropology 10.

1995b. Preclassic Maya of the Belize Valley: 1994-1995 project research objectives, in Healy & Awe (1995a): 1-17.

1995c. Radiocarbon dates from Cahal Pech, Belize: Results from the 1994 field season, in Healy & Awe (1995a): 198-215.

1996. Belize Valley Preclassic Maya project: Report on the 1995 field season. Peterborough (ONT): Trent University. Occasional Papers in Anthropology 12.

HEALY, P.F., K. EMERY & L. WRIGHT. 1990. Ancient and modern Maya exploitation of the jute snail (Pachychilus), Latin American Antiquity 1(2): 170-83.

HOHMANN, B. & T. POWIS. 1996. The 1995 excavations at Pacbitun, Belize: Investigations of the Middle Formative occupation in Plaza B, in Healy & Awe (1996): 98-117.

KEEGAN, W.F. & M.J. DENIRO. 1988. Stable carbon and nitrogen isotope ratios of bone collagen used to study coral-reef and terrestrial components of prehistoric Bahamian diet, American Antiquity 53: 320-36.

LAWLOR, E.J., A.J. GRAHAM & S.L. FEDICK. 1995. Preclassic floral remains from Cahal Pech, Belize, in Healy & Awe (1995a): 150-72.

LEE, D. & J.J. AWE. 1995. Middle Formative architecture, burials, and craft specialization: Report on the 1994 investigations at the Cas Pek group, Cahal Pech, Belize, in Healy & Awe (1995a): 95-115.

LENTZ, D. 1996. Foodstuffs, forests, fields, and shelter: A paleoethnobotanical analysis of vessel contents from the Ceren site, El Salvador, Latin American Antiquity 7(3): 247-62.

LENTZ, D., C.R. RAMIREZ & B.W. GRISCOM. 1997. Formative-period subsistence and forest-product extraction at the Yarumela site, Honduras, Ancient Mesoamerica 8(1): 63-74.

LONGIN, R. 1971. A new method of collagen extraction for radiocarbon dating, Science 230: 241-2.

MCKILLOP, H. 1995. Underwater archaeology, salt production, and coastal Maya trade at Stingray Lagoon, Belize, Latin American Antiquity 6: 214-28.

MACKINNON, J.J. & S.M. KEPECS. 1989. Prehispanic saltmaking in Belize: New evidence, American Antiquity 54: 522-33.

MARINO, B. & M.B. MCELROY. 1991. Isotopic composition of atmospheric C[O.sub.2] inferred from carbon in [C.sub.4] plant cellulose, Nature 349: 127-31.

MIKSICEK, C.H. 1991. The ecology and economy of Cuello I: The natural and cultural landscape of Preclassic Cuello, in Hammond (1991): 70-84.

MOHOLY-NAGY, H. 1978. The utilisation of pomacea snails at Tikal, Guatemala, American Antiquity 43(1): 65-73.

POHL, M. 1985a. The privileges of Maya elites: Prehistoric vertebrate fauna from Seibal, in M. Pohl (ed.), Prehistoric Lowland Maya environment and subsistence economy: 133-45. Cambridge (MA): Harvard University, Peabody Museum of Archaeology & Ethnology. Paper 77.

1985b. Osteological evidence for subsistence and status, in Pohl (1985a): 107-13.

POWIS, T. 1996. Excavations of Middle Formative period round structures at the Tolok Group, Cahal Pech, Belize. Unpublished M.A. thesis, Trent University, Peterborough (ONT).

1997. Ancient Lowland Maya utilisation of the freshwater clam (Nephronaias sp.). Paper presented at the 62nd annual meeting of the Society for American Archaeology, Nashville.

REED, D.M. 1994. Ancient Maya diet at Copan, Honduras, as determined through the analysis of stable carbon and nitrogen isotopes, in K.D. Sobolik (ed.), Paleonutrition: The diet and health of prehistoric americans: 210-21. Carbondale (IL): Center for Archaeological Investigations, Southern Illinois University. Occasional Paper 22.

SCHWARCZ, H.P., J. MELBYE, M.A. KATENBERG & M. KNYF. 1985. Stable isotopes in human skeletons of southern Ontario: Reconstructing paleodiet, Journal of Archaeological Science 12: 187-206.

SHARER, R.J. 1992. The Preclassic origin of Lowland Maya states, in E.C. Danien & R.J. Sharer (ed.), New Theories on the ancient Maya: 131-6. Philadelphia (PA): University of Pennsylvania, University Museum. Monograph 77.

SHAW, L. 1991. The articulation of social inequality and faunal resource use in the Preclassic community of Colha, Northern Belize. Unpublished Ph.D dissertation, University of Massachusetts, Amherst (MA).

STANCHLY, N. 1995. Formative period Maya faunal utilisation at Cahal Pech, Belize: Preliminary analysis of the animal remains from the 1994 field season, in Healy & Awe (1995a): 124-49.

TYKOT, R.H., N.J. VAN DER MERWE & N. HAMMOND. 1996. Stable isotope analysis of bone collagen, bone apatite and tooth enamel in the reconstruction of human diet: A case study from Cuello, Belize, in M.V. Orna (ed.), Archaeological chemistry: Organic, inorganic, and biochemical analysis: 355-65. Washington (DC): American Chemical Society.

VAN DER MERWE, N.J., R.H. TYKOT & N. HAMMOND. 1994. Diet and animal husbandry of the Preclassic Maya at Cuello, Belize. Paper presented at the 59th Annual Meeting of the Society for American Archaeology, New Orleans.

WEBSTER, D., N. GONLIN & P. SHEETS. 1997. Copan and Ceren: Two perspectives on ancient Mesoamerican households, Ancient Mesoamerica 8: 43-61.

WHITE, C.D., P.F. HEALY & H.P. SCHWARCZ. 1993. Intensive agriculture, social status, and Maya diet at Pacbitun, Belize, Journal of Anthropological Research 49: 347-75.

WHITE, C.D., F. LONGSTAFFE & R. SONG. 1996. Preclassic Maya diet at Cahal Pech, Belize: The isotopic evidence from human bone collagen. Paper presented at the 61st Annual Meeting of the Society for American Archaeology, New Orleans.

WHITE, C.D., M.D. POHL & H.P. SCHWARCZ. 1997. Maya husbandry of deer and dog. Paper presented at the 62nd annual meeting of the Society for American Archaeology, Nashville.

WHITE, C.D. & H.P. SCHWARCZ. 1989. Ancient Maya diet: As inferred from isotopic and chemical analysis of human bone, Journal of Archaeological Science 16: 451-74.

WIESEN, A. & D. LENTZ. 1997. Floral remains at Cahal Pech and surrounding sites: Summary report. New York (NY): New York Botanical Garden.

WILLEY, G.R., W.R. BULLARD, JR., J.B. GLASS & J.C. GIFFORD. 1965. Prehistoric Maya settlement in the Belize Valley. Cambridge (MA): Harvard University, Peabody Museum of Archaeology and Ethnology. Paper 54.

WING, E. 1975. Animal remains at Lubaantun, in N. Hammond (ed.), Lubaantun: A Classic Maya realm: 379-83. Cambridge (MA): Harvard University, Peabody Museum of Archaeology & Ethnology. Monograph 2.

WING, E. & S. SCUDDER. 1991. The ecology and economy of Cuello II: The exploitation of animals, in Hammond (1991): 84-97.

WRIGHT, L. 1994. The Sacrifice of the earth? diet, health and inequality in the Pasion Maya lowlands. Unpublished Ph.D dissertation, University of Chicago.

WRIGHT, L. & C.D. WHITE. 1995. Human biology in the Classic Maya collapse: Evidence from paleopathology and paleodiet, Journal of World Prehistory 10: 147-98.