STUDY AREA (Part 1)

In order to understand the technology of stone implements, in this case points, a multi-faceted approach is needed that includes an understanding of past environmental changes and the present environmental setting. This chapter will present a description of the geomorphological setting of archaeological sites analysed during this study, from which an understanding of the spatial positioning of the sites relative to each other and the primary resources within the region will be reached. These will include the stone resources, with an explanation of their suitability as a raw material and their availability.

The chapter also includes a detailed explanation of the environmental change and the associated impact of this on the prehistoric peoples of the region. This allows an understanding of one of the possible causes for the change in stone resource exploitation. A brief outline of the previous archaeological studies for each site will conclude this chapter.

Location

Smith & Cundy (1985) produced a revised distribution map for points (unifacial and bifacial) and backed blades in the Northern Territory. This distribution map was based on data collected from numerous controlled surface collections, isolated finds and archaeological excavations.

Stone points are well represented in the northern half of the Northern Territory, that is above the 18⁰ S parallel of latitude. There is a steep boundary gradient between those areas with points and those areas without points. Smith & Cundy (1985) see this as being indicative of a strong local factor limiting the distribution of points. Those sites north of approximately 10⁰ S latitude have much larger numbers of bifacial points. This study will be restricted to sites above the 10⁰ S latitude or that geographic area referred to as the Top End of the Northern Territory.

The study area extends over a large geographical area, encompassing several different local environments. These include the sparse sandstone plateau of Arnhem Land, the Adelaide River floodplains and the coastal plateau of Yarar (Figure 1).

Monsoonal temperature and precipitation regimes influence the climate of the area. Rainfall is negligible between April and October, building to a high monthly average in January and February. In contrast to precipitation, average daily temperatures remain relatively constant through the year.

Geologically the entire area is tectonically stable, and is predominantly composed of Lower Proterozoic rock types. Sedimentary rock types dominate with only limited supplies of volcanic or metamorphic materials (Table 1). The relationship and distance from the sites to the volcanics is significant for implement typology and the technology utilised. This is because cryptocrystalline volcanics are the most suitable raw materials for use in flaked tool technologies. These occur in proportionately small deposits throughout the region and are shown as italicised in Table 1.

The geological and climatic factors have produced a landscape with relatively low topographic relief with skeletal soils and stunted vegetation. The floodplain regions have deep sediment deposits covering flooded river valleys.

Stone Resources

Raw materials are one of the main influences on the morphology of an implement. They are seen as influential, by Crabtree (1972) and Allen & Barton (1989), but not deterministic as seen by Schrire (1982) and McCarthy, Brammell & Noone (1946).

Raw Material Properties

While the specific rock types used as raw materials in the manufacture of points vary, they share similar properties. The inherent properties of the raw material determine its level of suitability to the knapper. Crabtree (1972) has identified certain properties that are shared by "good quality raw materials".

The first of these properties is that the material should be hard and consolidated. Materials such as sandstone and siltstones, generally, are too soft and friable to flake well. Secondly, the material should be fine grained. These fine-grained materials are described as microcrystalline or, where the individual grains cannot be determined microscopically, cryptocrystalline in form. The coarser grained raw material produces more irregular flaked edges and surfaces. This in itself does not detract from the potential efficiency of the implement, but does make a fracture through the material that is hard to control. Finally, the material should be isotropic, that is having no preferred cleavage planes. These properties are most frequently found in the siliceous group of materials (Crabtree 1972; Schick & Toth 1993:122).

An isotropic, cryptocrystalline form then, is more suitable for the maintenance of the cutting edge, ensuring it remains sharp and robust. This is because, to maintain the edge, a knapper removes a small amount of material. A finer grained material, such as Gerowie Tuff, will allow small regular flakes to be removed from the margins and is more likely to be easily rejuvenated than the coarser grained quartzite raw material (Schick & Toth 1993; Cotterell & Kamminga 1990).

Raw Material properties influence the morphology of stone points. It is the effect of these properties that caused the conclusion by some authors, of stone point morphology being controlled or determined by the raw material.

There are only a limited number of raw materials within the study area suitable for stone implement manufacture. These include siltstone, greywacke, quartzite, silcrete, quartz and several varieties of chert and tuff. Of these it is a tuffaceous chert, commonly referred to as Gerowie Tuff, quartzite and to a lesser degree quartz, that dominates the archaeological record.

Quartzite & Silcrete

Quartzite is metamorphosed arenaceous sandstone. The constituent grains re-crystallise and develop an interlocking mosaic texture (Whitten & Brooks 1972:375). Silcrete forms from the silica replacement of the original constituent minerals by silica rich solutions. This produces a matrix of fine-grained quartz and amorphous silica (Langford Smith 1978:103).

Silcrete particles vary in size, resulting in a raw material that is variable in appearance and mechanical properties. When composed of very fine-grained matrixes it exhibits similar properties as chert, and with a larger sized matrix its properties resemble quartzite.

Quartz

Quartz is a crystalline form of silica, colourless to white in colour, with a vitreous lustre and hardness of seven. It exhibits a conchoidal fracture and is extremely resistant to weathering. Although having an internal trigonal crystallography, quartz crystals exhibit no recognised or predictable cleavage plane that would affect fracture path.

It forms in either tabular or sheet-like veins that intrude by lithostatic pressure into pre-existing joints or newly developed joints in the bedrock. These veins form from hydrothermal and magmatic fluids released during syn- or post metamorphic and igneous periods. The resultant veins may vary in width from a few millimetres to a metre or more (Thorpe & Brown 1990:16).

Deposited during metamorphic or igneous phases, veins of quartz are frequently subjected to changes in lithostatic pressure and stress. These changes produce a material that may have numerous macro and micro fractures, and that may influence the fracture path generated during knapping. Despite these problems, quartz is one the most frequently encountered raw materials for stone implements in the Australian archaeological record (Cotterell & Kamminga 1990; Cook & Kirk 1991).

Gerowie Tuff

The term Gerowie Tuff applies to a geological formation characterised by a predominance of rocks displaying evidence of volcanic activity. All of the material from the Gerowie Tuff formation appearing in the sample is the black crystal chert, tuffaceous non-albitic and tuffaceous albitic chert, these materials will be collectively referred to as "Gerowie Tuff" (Pietsch & Stuart-Smith 1987:23).

The formation lithology consists of siltstone, phyllite, occasional chert nodules, and thinly bedded argillite, devitrified and spotted black crystal chert and tuffaceous chert, all of which are suitable for the manufacture of implements. The latter beds are up to 10 metres thick throughout the sequence (Stuart-Smith et al. 1987:9). The black crystal chert, tuffaceous non-albitic chert and tuffaceous albitic chert is cryptocrystalline in nature, with no preferred cleavage plain and exhibits a conchoidal fracture (Stuart-Smith et al. 1987:9).

When freshly fractured, the material has a deep black vitreous surface. This, however, weathers through a number of observable intermediate steps to a white powdery cortex.

Weathering

The rock types of quartz, quartzite and silcrete visually remain relatively unaffected by the weathering process. Gerowie Tuff however undergoes a marked change in appearance when weathered. The change radically alters the appearance of Gerowie Tuff and may lead to a mistaken classification of the raw material.

Gerowie Tuff undergoes tripolitic weathering, which is an extreme form of patination where the fresh black surface transforms into a chalky, powdery white/grey outer cortex. This patination development moves through several readily identifiable stages. The sample contained examples of all stages of this weathering process.

Initial analysis divided the Gerowie Tuff material into four broad stages of weathering and patination development (Figure 2). These groupings were no patination, where the vitreous black surface is present. The first stage of patination development exhibits a dull black blue/grey colour with numerous white specks appearing in the surface layer similar in appearance to hornfels. The second stage of patination, the outer surface has taken on a grey/light brown outer surface with little or no powder formation and may appear as a chert like material. When fully patinated the surface is cream/grey in colour and has an earthy lustre. The outer surface exhibits a hardness of only one and scratches easily by a fingernail. It also exhibits an extremely fine-grained powdery surface. If the implement is broken the internal black vitreous material is visible and may also appear where recent flakes have been removed from the implement.

Those implements that exhibited full patination development could be considered older implements relative to those implements where there was no patination. If a significant difference existed between the stages of patination development, then it could be argued that a technological change had occurred over time within the Gerowie Tuff implements. Due to the lack of absolute dating techniques for open scatter sites, it would be difficult to establish chronologically a meaningful comparison between the sites. This then would impose a variable on the sample that is impossible to compensate for when comparing the raw materials.

To determine the existence of any technological change over time Anova analysis based on the attributes, Elongation and Robustness, was conducted. The next chapter provides full description of these attributes. This analyses indicated that at the 0.01% level of significance of Elongation (F 2.01, F crit 3.82, P-value 0.11) and Robustness (F 0.36, F crit 3.82, P-value 0.78) there was not a significant difference between these groupings. Therefore, the hypothesis of a technological change over time in the Gerowie Tuff implements is rejected. Because of this for the remainder of the study, the term Gerowie Tuff refers to all the stages of patination.

Having described the stone resources available and utilised in this region, the previous environmental changes of the region will now be examined.

Environmental Change

Environmental change has affected the entire region, but has produced differing geomorphological features that have affected resource distribution. Chappell and Grinrod (1984) and Woodroffe et al. (1985, 1986), have developed a detailed interpretation of the chronology and geomorphological changes in Mary River / Kakadu area (Table 2). The large fluvial induced features of the Mary River / Kakadu area are repeated on a much smaller scale in the Yarar area, there is, however, far fewer geomorphological studies of the Yarar area.

A period of eustatic sea-level rise and climatic change that commenced about 18,000 B.P. formed the coastal lowlands and floodplains of the South Alligator. Sea levels rose from 150 metres below current levels and achieved stabilisation approximately 6,500 B.P. Sedimentation infilling of the river valleys commenced at this time with a fluctuating rate until recent times. At approximately 8,500 to 8,000, B.P. the marine transgression reached the base of the old river valleys, which were 10-12 metres below Australian Height Datum (Woodroffe et al.1986, 1988).

Following the marine transgression, the Big Swamp phase from 6,800 to 5,300 B.P. occurred when the estuarine system shoaled, leading to the development of widespread mangrove forests dominated by Rhizophora that extended across the floodplain. In the lower reaches of all the coastal rivers, these stands of mangroves were up to 5km in width from the river channel. During this time, the riverbed was 6 – 8 metres below the present level of the plains. By 4,000 B.P. the vast mangrove forests were gone. While the exact date is yet to be determined, it appears that the forests were declining for several hundred years (Allen & Barton 1989; Thom & Chappell 1975; Woodroffe et.al. 1986).

The sinuous river phase, 4,000 to 2,500 B.P., followed the Big Swamp phase. This phase is characterised by a meandering stream-channel such as the South Alligator River does in present times. Channel width and rate of discharge was less than the present day, however sediment volumes allowed coastal progradation to accelerate during this phase (Woodroffe et.al. 1986).

With the demise of the big swamp mangrove forests, broad saline and hypersaline plains developed adjacent to the river channels at about 3,000 B.P. While no data is available for the Fitzmaurice River, which is south of Yarar, the Daly River north of Yarar has exhibited similar formation processes and stages. During the Sinuous Phase, numerous river meanders were cut off from the main river channel forming ox box lakes. The length of river exhibiting a meandering segment varied, the Daly River was the shortest while the Adelaide River had the longest segment. A longer meander stage results in a greater number of channel cut-offs and freshwater billabongs (Woodroffe et.al. 1986).

The final stage of river development is the Cuspate Phase. This is characterised by the presence of more acute angled meanders in the middle and upper reaches of the rivers. There is also a reduced frequency of salt-water intrusions into the upper reaches of the rivers. The presence of numerous freshwater billabongs and the ponding of freshwater against the floodplain and continued sedimentation of the lower sections of the rivers created freshwater wetlands within the last 2,000 years (Hope et al. 1985).

Wetland development in the Daly and Fitzmaurice Rivers did not form as extensive an area as the Mary, Adelaide and Alligator Rivers. This is in part due to the shorter meander channel and greater channel migration rate of the Daly River. It is also a product of more elevated land to maintain a separation of the two river floodplains, unlike the Mary, Adelaide and Alligator Rivers (Thom & Chappell 1975; Woodroffe et.al. 1986).

Wetland development is associated with the development of deep soil profiles, increased soil fertility and the spread of grasses. This region then has experienced a cyclical resource supply history. During the Big Swamp Phase, the large stands of mangroves provided a high level of resources. The Sinuous Phase followed the Big Swamp Phase and the resource supply fell due to the loss of the mangroves and development of hypersaline plans. In the final Cuspate Phase, the extensive development of freshwater wetlands has again provided a relatively high resource base.

Part 2 - Sites Included in This Study

Introduction

Typologies & Points

Author: Wayne Roddom, Dept. Archaeology and Anthropology
Feedback: peter.hiscock@anu.edu.au .
Date Last Modified: 5-June 1998
URL: http://artalpha.anu.edu.au/web/arc/aboutus/studs/studyare.htm