Australia: The Land Where Time Began

A biography of the Australian continent 

Lake Mungo – Evidence for Seed Grinding in the Pleistocene

Archaeological sites at Lake Mungo, southwestern New South Wales, have often produced grinding stones and fragments, the function of which has been inferred mostly on the basis of grindstone morphology. The antiquity of the grinding of grass seed has been of particular interest, usually being associated with large sandstone dishes that are deeply grooved. Previously no compelling evidence of seed grinding before the Pleistocene/Holocene boundary has been found in studies of grinding stones from this region. One of the problems associated with previous studies of grindstones in this area has been that the grinding stones have been found on deflated surfaces making it difficult to determine their provenance and date. In this paper Fullagar et al. report a functional study of 17 sandstone artefacts that had been collected from the central part of the Mungo lunette and a suite of OSL ages have provided bracketing age estimates for the stratigraphic units. Of the artefacts tested 10 were attributed to Unit E that was deposited between about 25 ka and 14 ka, and 4 were attributed to Unit F, which was deposited about 8 ka. There are 3 artefacts from the Golgol lag for which the ages are not known. Usewear indicates it is likely seeds were ground on 14 of the artefacts; the use-related residues include starch, cellulose and collagen. Fullagar et al. suggest the results of this study provide additional evidence for plant processing and seed grinding activities in Sahul during the Pleistocene.

Lake Mungo is a dry lake in semiarid southeastern Australia (Fullagar et al., 2015: Fig.1) where archaeological investigations have identified some of the oldest burials, faunal remains, hearths, ochre, flaked artefacts and grinding stones that have been found in Sahul (Pleistocene Australia-New Guinea) (Bowdler et al., 2003; Mulvaney & Bowler, 1981). In Sahul grinding stones were used for many functions, such as the production of ground-edge axes and hatchets; as well as for the processing of bone, shell, ochre, small animals, medicines, drugs, poisons and many other materials, as has been witnessed ethnographically (see Gott, 2002). The antiquity of the grinding technology, the emergence of seed grinding activities, and correlations with climate change with the consequent shifts in the availability of resources, have been key problems in the prehistory of Sahul.  Cuddie Springs, in semiarid southeastern Sahul is the only site from which secure evidence for the processing and grinding of specific plant foods, such as tubers and seeds, in a Pleistocene context has been reported (Fullagar et al., 2008).

Fullagar et al. present in this paper the results of a functional study of 17 sandstone artefacts, which included 8 refitted fragments, LMGS2-9, all of which had been collected from the central part of the Mungo lunette. Bracketing age estimates for the strata in this part of the lunette (Fitzsimmons et al., 2014) has been provided by OSL (optically stimulated luminescence), thereby providing approximate ages for the associated artefacts from the late Pleistocene, which included fragments of grinding stones which have been documented within the mapped region (Fullagar et al., 2015: Figs. 1 & 2; Fitzsimmons et al., 2014; Stern et al., 2013).

The ethnographic evidence and the archaeological significance of grinding stones in the local and regional Aboriginal economies has been described (Allen, 1974). According to Fullagar et al. there is a great deal of ethnographic evidence for the processing of grass seeds, which have been identified as a major subsistence base in arid regions of mainland Australia (Tindale, 1977). The archaeological significance and distinct morphologies of implements for grinding seeds in Central Australia, particularly the large dished millstones and mullers that were used for the processing of seeds have also been documented (Smith, 1985, 1989). Though when dealing with small fragments of grinding stones, which are the most common form that are recovered in archaeological contexts, it is not always possible to extrapolate the original shape or size of a grinding stone on the basis of morphology.

Also, Fullagar et al. say there is no evidence that suggests the grinding stones that were used in the Pleistocene are necessarily of the same morphology as those that have been ethnographically documented in Central Australia. Usewear traces associated with residues are the best indicators of the function of an artefact.

Fullagar et al. say the aim of the study presented in this paper was to reassess seed grinding at Lake Mungo by adopting an integrated approach to the analysis of usewear, optically and biochemically visible residues and microscopic technological features of sandstone artefacts that are associated with dated contexts.

Conclusion

A multidisciplinary study of Lake Mungo grindstones presented here has demonstrated that interpretable wear traces and residues are preserved on small ground stone fragments that were recovered from exposed surfaces of stratified, eroding landforms in semiarid Australia. The original morphologies of the grindstones are difficult to reconstruct from small fragments, but include concave, convex, facetted and flat surfaces, with many bearing similarities with ethnographic implements for seed grinding from Central Australia. The Usewear of 14 of the 17 artefacts recovered is consistent with the results obtained from experiments and ethnographic data indicating seed grinding (Fullagar et al., 2015: Fig. 5). According to Fullagar et al. all the pieces of sandstone that were examined almost certainly derive from larger broken grinding stones that must have been carried into this landscape. Of the specimens studied 1 has wear that is poorly developed with uncertain function (LMGS12); 1 specimen (LMGS13) has no distinct Usewear; and 1 specimen (LMGS11) has evidence of plant processing, though not necessarily including seeds. Residues could be derived from use, though they were recovered in low abundance. Fullagar et al. suggest the poor residue recovery is probably the result of poor preservation after erosion from the depositional matrix. Fullagar et al. suggest variation in the observations of residue may in part be attributable to the difference between worked surface areas sampled by the researchers and methodical approaches. The majority of plant and animal tissues that were recovered have not been identified with taxonomic precision.

Compelling evidence from the results of this integrated approach to functional analysis has been provided that 14 of the 17 grinding stone fragments that were used in the processing of seeds. Multiple functions for at least 3 artefacts is suggested by the presence of other plant tissue and animal residues that include collagen. A precise provenance has been determined for only 1 of the grinding stones, though all can be linked directly with particular strata in the central Mungo lunette. Combined with the chronological sequences that have been established for units within the lunette (Fitzsimmons et al., 2014; Stern et al., 2013) the recent geomorphological study has provided strong evidence of a Pleistocene context (25-14 ka) for the fragments of grinding stones from Lake Mungo. Further support for the theoretical argument that the development of seed grinding may be linked with environmental stress that was associated with the LGM is provided by the Usewear and the residue traces indicating that seed grinding occurred during the Pleistocene at Lake Mungo (Edwards & O’Connell, 1995; Fullagar & Field, 1997: 302).

Sources & Further reading

1. Fullagar, R., E. Hayes, B. Stephenson, J. Field, C. Matheson, N. Stern and K. Fitzsimmons (2015). "Evidence for Pleistocene seed grinding at Lake Mungo, south-eastern Australia." Archaeology in Oceania 50: 3-19.