Lake Mungo, an analysis of the surface collection

Abstract

Prehistoric artefacts are being eroded from lunettes bordering dry lakes of the Willandra system in western New South Wales. At Lake Mungo the artefacts are found on two principal surfaces which were laid down when the lake was active. These Mungo and Zanci deposits have been radio-carbon dated to about 40,000 years BP and 18,000 BP respectively. The results of an analysis of the characteristics of the artefacts are presented. Following a summary of the geomorphological and climatic studies relevant to the formation of the lunette, the effect of erosion on the provenance of the artefacts was examined and it was concluded that artefacts up to 2 cm maximum dimension may be moved by surface flow following heavy rain. This was confirmed when the conditions were simulated experimentally but no support was obtained when the displacement of marked artefacts on the lunette surface was recorded. Problems associated with the detection and recognition of artefacts from a surface of a similar colour were assessed as being unimportant provided it is not intended to recover items smaller than about 1 cm in size. Smaller stones are likely to remain undetected. An analysis of the surface distribution of the collected implements was inconclusive due to the small sample size and the effect of chance displacement. In particular, an apparently statistically significant result is shown to have little archaeological relevance. The distribution of the more numerous unretouched flakes showed some correspondence with the older Mungo surface but movement of smaller flakes down the surface of the lunette was indicated also. A detailed comparison of the dimensions and characteristics of the implements was unable to identify major differences between those recovered from the Mungo and Zanci surfaces. However, the mean dimensions of both groups were considerably smaller than those of three other Australian assemblages of comparable age, including · two samples collected previously from the Lake Mungo lunette. In the latter case, however, it is suggested that the first surface collection from a given area may be biassed by erosion which has insufficient time to affect subsequent annual collections. The thesis concludes by discussing areas for further research which could lead to a more complete interpretation of the surface collection. Show more