PREVIOUS ARCHAEOLOGICAL WORK

4.2Previously recorded Aboriginal archaeological sites in Melbourne’s west

The earliest evidence for human occupation in the Melbourne area was found in the Maribyrnong River valley at Keilor, at the confluence of the Maribyrnong River and Dry Creek. This area, now known as the Keilor Archaeological Area, has been a focus for archaeological investigation since 1940, when the ‘Keilor cranium’ was discovered (Gill 1966: 581). Excavations since then have resulted in a large number of stone and bone artefacts crafted by humans, ochre, charcoal and hearth stones. The stone material commonly found at the site is silcrete, chert, quartz, basalt and chalcedony. These stone types were manufactured into a small number of formal tool types such as microliths, thumbnail scrapers, larger scrapers and backed blades.

More recently, other evidence has come to light about early occupation of the Maribyrnong River valley. In 1965, some bones were uncovered in the Keilor Terrace of the Maribyrnong River. These bones were found to have come from two shallow graves which provided a radiocarbon date of 6500 years old (Mulvaney 1970a: 1). Subsequent excavation exposed stone and bone artefacts and bone fragments. Some artefacts have been dated to over 17,000 years old (Mulvaney 1970b: 63). Stone artefacts were mostly flakes, small blades and cores made of quartzite, through silcrete became more frequent during more recent occupation (Mulvaney 1970b: 71).

Regional studies of the distribution of Aboriginal archaeological sites have been conducted within the region of the Maribyrnong River. Such studies include Presland’s study of metropolitan Melbourne (1983), du Cros’ study of Melbourne’s western region (1989) and the Sydenham Corridor (1990), Rhodes’ preliminary investigations of the Upper Maribyrnong River valley (1989) and the City of Keilor (1990), and Webb’s identification and documentation of silcrete quarries, which includes a discussion of the Maribyrnong River valley. The data from these studies can be used to formulate a predictive model for Aboriginal archaeological sites within the City of Maribyrnong, which identifies where sites are likely to occur, in relation to landform, association with watercourses, and levels of disturbance.

Hilary du Cros’ model for the western region of Melbourne suggests that sites will most likely occur within 100 metres of a major watercourse, on a river, creek, flood plain, flats, alluvial terrace or hill slope. Site types are usually surface artefact scatters, isolated artefacts, fresh water middens, scarred trees or burials. Stratified sites might occur on flood plains and alluvial terraces. Artefact scatters and isolated artefacts also occur on high ground (du Cros 1989: 67-68).

During her study of the Sydenham corridor, du Cros’ model of the western region was expanded upon and details on the sensitivity of the Maribyrnong River clarified. She found that ‘major rivers and creeks landscape’ including the Maribyrnong River featured the largest number of sites. Surface artefact scatters were the most common type of site found in this landscape, most of which were recorded, in the Maribyrnong valley, at the edge of the escarpment (du Cros 1990: 23). Sites were most likely found within 100 metres of major watercourses and stone sources were found to have been suitable for quarrying stone artefacts. However, it was noted that sites may extend up to 400 metres back from the edge of the escarpment (du Cros 1990: 29). du Cros also noted that in the Sydenham corridor there was an absence of mature eucalypts, as these had been removed during the 1800s. Such trees may originally have borne cultural scars (du Cros 1990: 24). Sites with extensive sub surface archaeological deposits such as hearths, burials or faunal material would most likely occur in areas with the best preservation.

Rhodes conducted a preliminary survey of the Upper Maribyrnong valley, including the Maribyrnong River, Deep Creek and Jacksons Creek. His survey resulted in the recording of 50 sites, all of which were open sites on exposed ground surfaces. Surface artefact scatters were most common, and within the Maribyrnong valley, most occurred on flood plains and alluvial terraces adjacent to the river, where ploughing had exposed them (Rhodes 1989: 12). In unploughed areas, sites tended to be located in eroded ground surfaces on hill slopes, exposed on tracks or eroded terraces. Scarred trees were found in very small numbers on the river channel. The lack of scarred trees along the major waterways was attributed to European land clearance (Rhodes 1989: 13).

Rhodes (1989: 12) speculated that the prevalence of sites on flood plains along the Maribyrnong River is partially a factor of ground surface visibility. The high number of sites found on alluvial terraces may partly due to their frequent use as market gardens, the practice of which exposes artefacts on the ground surface. The low number of sites on the escarpment may reflect the poor ground visibility in these areas. Site distribution may also be influenced by selective preference of alluvial terraces by Aboriginal people when choosing campsites. In the Upper Maribyrnong valley it was common to find sites on raised terraces above the flood plain and on hill slopes. Rhodes stated that “This may indicate a preference for use of higher areas above the flood plain, but in close proximity to campsites; however, there is also evidence of more intensive use of the flood plain (in some areas)” (Rhodes 1989: 13).

Another study by Rhodes which included the Maribyrnong River valley is the City of Keilor Archaeological Survey (1990). This comprised the boundaries of the City of Keilor and included a section of the Maribyrnong River. At the time of writing, 101 sites were present within the City, the majority of which were located in incised valleys. Rhodes found that within the major river and creek valleys, most sites tended to occur on the flood plain and escarpment, and that those sites found on undulating plains usually occurred within 100 metres of water. There were a greater range of site types in the incised valleys, which led Rhodes to suggest that these areas featured more intensive Aboriginal land use, whilst the more limited range of sites on the plains probably indicated a series of transient, short-term campsites (Rhodes 1990: 43).

A specific site type found in the Maribyrnong River valley are silcrete quarries, or stone outcrops. Webb (1995: 73) conducted an investigation of the quarries in this valley as part of a study of metropolitan Melbourne, and found that 11 confirmed silcrete quarries were recorded there. Seven of the stone sources occurred on the banks of the Maribyrnong River while the remaining sites are located on Jacksons Creek and a tributary. The silcrete varied in colour and quality, was present at most sites as boulders, and was extensively quarried. Most of the sites showed severe disturbance (Webb 1995: 74).