Ancient Maya subsistence diversity: root and tuber remains from Cuello, Belize.

Hather, Jon G. ; Hammond, Norman

A first notice of the mushy element in the subsistence base of the Maya realms.

Introduction

The question of prehispanic subsistence in the Maya lowlands has been vigorously debated, since the assessment of Classic Maya (AD 250-900) structural development, particularly in terms of urbanism, has hinged, somewhat precariously, on the nature of their subsistence economy. A model based on retrodiction of the historic regime of swidden or milpa agriculture of predominantly maize, beans and squashes (e.g.. Morley 1946: 441-2) did not easily account for the population densities and proportion of non-farmers projected for Classic cities such as Tikal once survey revealed their unexpected size (Carr & Hazard 1961).

Bronson (1966) therefore reconsidered the subsistence economy of Classic Maya lowland society, suggesting the supplementation of maize with root crops: in this model agricultural yield is sufficiently increased and nutritionally enhanced to meet the needs of a population of greater density than would be possible under a standard milpa regime. Bronson considered that documentary and botanical evidence for four crops -- the sweet potato or camote (Ipomoea batatas); yam bean or jicama (Pachyrrhizus erosus); manioc (cassava, yuca) (Manihot esculenta); and malanga (yautia) (Xanthosoma spp.) -- suggested that the Maya region was a centre for either domestication or diversification. He argued that the Maya had a variety of root crops available to them before the colonial age, and arguably in the Classic Period. Marcus (1982) argued instead for a much later, post-Columbian introduction of three of these root crops, while admitting that a Classic Maya knowledge of wild roots and tubers, particularly for use as famine foods, may have been comprehensive. Pachyrrhizus, she suggests, was likely to have been a Postclassic (AD 900-1500) introduction and possibly a famine food.

Pachyrrhizus erosus is a very fleshy leguminous root crop and by virtue of its anatomy -- large thin walled parenchyma cells, sparsely distributed xylem vessels, high water content -- may be eaten without preparation or cooking; modern consumption is raw, as a snack or salad. These characters are not conducive to good preservation by charring: cellular characters will be largely destroyed by the expansion and release under pressure of water contained in the tissues; the resulting carbonized remains will be extremely fragile, almost certainly unable to survive most post-de-positional environments. Ipomoea batatas has been recovered from archaeological contexts (Rosendahl & Yen 1972; Hather & Kirch 1992), though not in the Maya lowlands; although variable in its morphology and anatomy, it is often not difficult to identify even from highly fragmented remains. The whole of the tuber is eaten, however, usually without preparation before cooking by boiling, steaming or roasting: the likelihood of its presence in the archaeological record is low. Different varieties of Manihot esculenta are boiled, or grated and dried into flour. Detecting its prehispanic utilization is inextricably bound up with the preservation of identifiable remains from such culinary practices; we report here the results of recent research at the Preclassic Maya site of Cuello, Belize, which advance our knowledge of ancient Maya root crop utilization.

Cuello is a small site located on the limestone ridge between the Rio Hondo and Rio Nuevo in northern Belize at 18 [degrees] 05' N, 88 [degrees] 35' W. Excavations between 1975 and 1993 have documented occupation from c. 1200 BC onwards, the earliest settlement known in the Maya lowlands, and a cultural development leading to the emergence of a complex Late Preclassic society after 400 BC (Hammond 1991). Floatation of more than 10 tonnes of soil recovered numerous plant macrofossils, which have been used to reconstruct the Preclassic environment and economy (Miksicek 1991). More than 1100 maize cupule and kernel fragments were recovered (Miksicek et al. 1981), together with carbonized fruits and/or wood from trees yielding edible crops such as avocado. Explanations for the virtual absence of root crops from the flot samples have been discussed by Miksicek (1991: 80).

Recovery and identification of carbonized root crop remains from humid tropical soils elsewhere (Hather 1991) led to a renewed attempt to resolve the problem of whether the Maya had cultivated them. Plant remains, preserved by charring, were recovered from midden deposits at Cuello in 1992 by on-site flotation and laboratory-based deflocculation and by wet sieving of five-litre bulk samples. Both resulted in the recovery of similar assemblages, but the materials recovered in the laboratory were less fragmented and eroded, allowing greater chance of identification. A combination of flotation and more careful laboratory-based separation is necessary to recover a complete assemblage of charred plant remains from archaeological contexts.

Analysis concentrated on the recovery and identification of soft tissues or parenchyma. Most root crops, certainly the ones under discussion here, are composed largely of parenchyma, a tissue comprising often large, thin-walled cells whose function is the storage of starch and related compounds. Possible fragments of such tissues were pressure fractured to obtain a clean, uneroded surface and examined under scanning electron microscopy. Comparative material from modern Maya root-crops was experimentally charred and examined in the same way, as well as being thin-sectioned and stained for transmitted light microscopy. Identifications were based upon anatomical and morphological characters of the type outlined by Hather (1991; 1993).

Parenchymatous tissues at Cuello

Ten samples were examined for charred plant remains. Six, from samples associated with the Early Middle Preclassic period (Swasey (1200-900 BC) and Bladen Xe (900-600 BC) ceramic complexes), and the Late Middle Preclassic period (Lopez Mamom ceramic complex (650-400 BC)) yielded the remains of three distinct classes of parenchymatous tissues:

Unidentifiable fragments of parenchymatous tissues, possibly derived from root or tuber tissues, but which could equally be derived from large seeds (for example the ramon or breadnut, Brosimum alicastrum), large fruits or other fleshy plant organs. These will not be considered further.

sample categories of parenchymatous tissue
no. 1 2 3

4415 + - -
4440 + + -
4448 + - +
4457 + + +
4458 + - +
4490 + + -

1 = unidentifiable remains
2 = cf. Xanthosoma sp.
3 = Manihot esculenta

TABLE 1. Occurrence of different categories of parenchymatous plant tissues
recovered from samples at Cuello.

Partially vesicularized parenchymatous tissue with approximately one-third recognizable cellular tissue. Vascular tissue is in small amphivasal concentric to possible collateral bundles with breakdown of phloem to solid carbon. Tissue preservation and its organization is very similar to Araceous tissues recovered from the Pacific (Hather & Hunt, forthcoming) and to modern reference material of Xanthosoma (FIGURES 1 (archaeological) & 2 (modern)).

Well preserved narrow lengths of root tissue with epidermis. The anatomical structure of these fragments indicated a root structure with characters of both the secondary xylem (largely parenchymatous) and phloem well preserved. Morphological and anatomical characters identify them as parts of narrow roots attaching the root tubers of Manihot esculenta to the main axis of the plant. Interestingly, this part would naturally be discarded in preparing the tuber for cooking (FIGURES 3 (archaeological) & 4 (modern)).

Of these identifications that of Xanthosoma is the less certain. Characters of its anatomy suggest this identification, but whether it was cultivated (and inferentially domesticated?) or collected cannot be determined. As other Araceous taxa that exist in the wild in the Cuello area have not been examined as reference material, identification stands as a possibility rather than a certainty. The identification of Manihot esculenta is of greater certainty, because the tissues are better preserved and more complete. Marcus (1982, citing Rogers 1963) states that Manihot esculenta does not occur in its wild form in the Maya area. If that is correct, this identification must perforce document cultivation of manioc by the Preclassic Maya.

Six samples included unidentifiable tissues; three yielded the remains of the possible Xanthosoma and three those of Manihot esculenta; one sample yielded unidentifiable tissues only, and one sample yielded both of the identified categories; maize was recovered at Cuello at an 82% ubiquity, though from a much larger range of samples (Miksicek 1991). The remains under discussion here were derived from a relatively small sample size and therefore the comparison of ubiquity is not statistically viable, but indications are that root and tuber ubiquity is approximately two thirds that of maize remains.

One sample of Manioc was submitted to the Oxford University Accelerator Unit for AMS dating with the following result: 2450[+ or -]70 b.p. (OXA-4452) which at 0.77 probability lies between 700 BC and 531 BC at 1[Sigma] level (0.23 probability for 800-753 BC); and with 0.94 probability between 808 and 475 BC at 2[Sigma] level (R. Housley pers. comm). These determinations are compatible with other AMS dates from occupation material and burials from the Middle Preclassic period at Cuello.

Conclusions

This analysis of plant remains from Cuello was directed at the recovery and identification of root and tuber tissues with direct reference to the question of whether Classic Maya subsistence economy could have been, at least partially, based on root crops. We have found Cuello assemblages to contain tissues derived from roots and tubers including one and perhaps two of the four root crops identified by Bronson (1966) in his model of Classic Maya subsistence economy. The relative importance of root and tuber foods and of maize cannot easily be determined from such archaeological data. Use of foods in the past will have altered from year to year, depending on crop success and minor variations in cultural preference. The accuracy of the archaeological record in representing the past depends on factors of preservation and post-depositional taphonomy. The comparison of such data quantified in terms other than that of ubiquity poses an enormous problem to archaeobotanists.

The nature of the involvement of roots and tubers in past subsistence patterns in the lowland Maya area depends on the particular species grown. Pachyrrhizus erosus, Ipomoea batatas and Xanthosoma spp. may be grown in gardens with other crops such as maize, and the yield of either would only increase proportionately to the area of land devoted to their cultivation if grown as a monocrop. Manihot esculenta, a large bushy crop commonly grown on relatively poorly prepared ground, is unlikely to have been grown in a multi-cropping system. In this respect manioc is likely to have been grown in rotation with other crops either following a fallow period and prior to a maize crop, or following maize and prior to fallow. A subsistence system relying on a variety of root crops and maize, beans and squash would probably be based on both relatively permanent house gardens growing a wide variety of crops possibly including Xanthosoma spp., Ipomoea batatas and Pachyrrhizus erosus, as well as manioc and maize rotating in milpa swidden fields.

Acknowledgements. Funding for this study was provided by the Science and Engineering Research Council, Institute of Archaeology, University College London, and Boston University. The Cuello excavations were funded principally by the National Geographic Society and Boston University. We thank Amanda Clarke, site director at Cuello in 1992, and David R. Harris, Director of the Institute of Archaeolog}, for practical and fiscal assistance, and Rupert Housley and Robert Hedges of the Oxford University Radiocarbon Accelerator Unit for AMS dating.

References

BRONSON, B. 1966. Roots and the subsistence of the ancient Maya, Southwestern Journal of Anthropology 22: 251-79.

CARR, R.F. and J.E. HAZARD. 1961. Map of the ruins of Tikal, El Peten, Guatemala. Philadelphia (PA): University of Pennsylvania Museum. Tikal Report 11.

HAMMOND, N. (ed.). 1991. Cuello: an early Maya community in Belize. Cambridge: Cambridge University Press.

HATHER, J.G. 1991. The identification of charred archaeological remains of vegetative parenchymatous tissues, Journal of Archaeological Science 18: 661-75.

1993. An archaeobotanical guide to root and tuber identification 1: Europe and South West Asia. Oxford: Oxbow. Monographs in Archaeology 28.

HATHER, J.G. & T. HUNT. Forthcoming. Prehistoric taro (Colocasia esculenta) from Hawaii.

HATHER, J.G. & P.V. KIRCH 1992. Prehistoric sweet potato (Ipomoea batatas) from Mangaia Island, Central Polynesia, Antiquity 65: 887-93.

MARCUS, J. 1982. The plant world of the 16th- and 17th-century Maya, in K.V. Flannery (ed.), Maya subsistence: studies in memory of Dennis E. Puleston: 239-67. New York (NY): Academic Press.

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

MIKSICEK, C.H., R.M BIRD, B. PICKERSGILL, S. DONAGHEY, J. CARTWRIGHT & N. HAMMOND. 1981. Preclassic lowland maize from Cuello, Belize, Nature 289: 56-9.

MORLEY, S.G. 1946. The ancient Maya. Stanford (CA): Stanford University Press.

ROGERS, D.J. 1963. Studies of Manihot esculenti Crantz, and related species, Bulletin of the Torrey Botanical Club 90: 43-54.

ROSENDAHL, P & D.E. YEN. 1972. Fossil sweet potato remains from Hawaii, Journal of the Polynesian Society 80: 379-85.