Pollen and Phytoliths in Stone Mounds at Pouerua, Northland, New Zealand: Implications for the Study of Polynesian Farming
Horrocks, M., Jones, M. D., Carter, J. A., Sutton, D. G., Antiquity
Current archaeological understanding of pre-European Maori farming is based largely on inferences drawn from anecdotal accounts of the late 18th and early 19th centuries (Best 1925; Leach 1979), and indirect evidence such as landscape architecture or soil structure (e.g. Law 1975; Challis 1976). While this has allowed identification of some Maori gardening practices, the ability directly to identify these in archaeological landscapes is limited. An example of how this has affected archaeological research of farming in New Zealand is in the research of terrace gardening -- this practice is relatively unstudied because of the difficulty of distinguishing house and garden terraces (Leach 1979).
Forms of direct evidence of farming would include the preserved remains of in situ cultivated plants. Early Maori successfully introduced six plants to New Zealand for cultivation: sweet potato Ipomaea batatas, yam Dioscorea spp., taro Colocasia esculenta, ti pore Cordyline fruiticosa, gourd Lagenaria siceraria and paper mulberry Broussenetia papyrifera (Leach 1979).
Pollen and opal phytoliths, both microscopic, are two of the most likely types of plant parts to be preserved in sediments. The outer wall of pollen grains (and spores of ferns and other non-polleniferous plants) is composed of sporopollenin, an organic substance highly resistant to decay. Opal phytoliths are silica bodies deposited in plant tissue during the life of the plant and are highly durable, generally being preserved for longer periods in sediments than pollen. Both pollen and phytoliths are often specific to particular plant families, genera or species.
Pollen preserved in sediments potentially provides a high level of spatial discrimination. Wind-pollinated plants generally produce abundant pollen which may be dispersed long distances (up to hundreds of kilometres), whereas animal- (mainly insect) pollinated plants produce much smaller amounts of pollen, most of which is deposited on the ground within a few metres of the parent plant. Spores from nonpolleniferous plants also vary in dispersal distance. The difference between species' pollen production and dispersal results in pollen representations which may change considerably over just a few metres. In a recent criminal case in New Zealand, Horrocks & Walsh (1999) were able to differentiate a crime scene and alibi scene only 7 m apart by demonstrating that the pollen assemblages in soil samples from these two directly adjacent `localized areas' were significantly different. As most phytoliths are presumably deposited immediately beneath parent plants, they should provide a level of spatial discrimination similar to that of pollen.
Pollen in sediments also potentially provides a fine level of temporal discrimination. From pollen analysis of swamp sediments in southeastern Australia, Green et al. (1988) showed that fine sampling of sediments at intervals of 1 cm or less can yield a temporal resolution of 5-20 years per sample. Phytoliths should provide a level of temporal discrimination similar to that of pollen.
The use of pollen and phytoliths in archaeology is described in detail in Pearsall (1989) and Piperno (1988), respectively. Fossil pollen has provided evidence for prehistoric farm activity in, for example, England (Dimbleby 1985) and Lowland Central America (Wiseman 1983) (identifying farm fields). Phytoliths have also provided evidence for farm activity in Europe and Asia (Rovner 1983) (several domesticated cereals), Ecuador (Pearsall 1978) and Panama (Piperno 1985) (maize Zea mays), and Papua New Guinea (Wilson 1985) (banana Musa spp.). In the Polynesian region, however, palynological/ phytolith analyses of prehistoric farm sites are limited to Pearsall & Trimble's (1984) Hawaiian study. Pearsall & Trimble were able to identify deforestation, subsequent farm activity and the eventual abandonment of the fields through changes in tree:grass phytolith ratios. …