The Coonawarra: a viticultural frontier? Or just a case of sour grapes?

This paper discusses the definition of the Coonawarra Viticultural Region by the Geographical Indications Committee of the Australian Wine and Brandy Corporation. The paper focuses on the geographical elements, critiques the extended process that resulted in the determination, and provides some alternative suggestions based on geography.

**********

By 1990 about 400 viticultural regions were recognised in Australia. In the late 1980s the need for formal definition arose as a result of international treaty obligations connected with trade and intellectual property rights. From 1994 a committee set up by the Australian Wine and Brandy Corporation (AWBC) became responsible for the official delineation and definition of Australian viticultural regions. This was the Geographical Indications Committee (GIC). The committee of three worked within a framework of Regulations which were to give "geographical" guidance.

In 1997, the GIC presented an Interim Determination for the Coonawarra Region. Whereas most previous determinations had proceeded with few problems, the Coonawarra Determination was to cause great discontent and conflict and, arguably, has still not been satisfactorily resolved. It had become a "frontier of dissent".

BACKGROUND TO THE REGION

This area of South Australia is part of the Murray-Darling Basin, which has experienced a series of marine incursions over geological time and the consequent accumulation of great thicknesses of marine sediments including limestone (O'Driscoll, 1960). The sea began its most recent retreat from this area about a million years ago, leaving a series of fossil coastal sand dunes roughly parallel to the present coast (Sprigg, 1952, Blackburn et al, 1965) (Figure 1). Though locally called "ranges" the highest and most easterly of these, the East and West Naracoorte Ranges are only 50 to 60 metres above the plains to the west and 30 to 40 metres above the plains to the east. The relative elevation is a result of movement along the Kanawinka fault line (marked "KF") (O'Mara, 1957). Some 20km to the west of the Naracoorte Ranges is the Cave Range ("CR") which rises only 10 to 20 metres above the surrounding plain. The West Naracoorte Range is estimated to be between 720,000 and 770,000 years old and the Cave Range between 650,000 and 680,000 years (Drexel and Preiss, 1995). Between and roughly parallel to the Naracoorte and Cave Ranges lies a subdued elongated surface feature in the form of a low platform (not shown in this figure). This platform extends for approximately 20 to 25 kilometres north-south and is about 2.5 kilometres wide at its maximum. It stands 1 to 2 metres above the surrounding countryside. Its origin is uncertain but may represent the degraded and weathered calcrete core of a fossil dune.

[FIGURE 1 OMITTED]

The predominant soil on this platform is Terra Rossa. This is the most extensive occurrence of this soil type in the area though it occurs frequently throughout the region in discrete locations (Blackburn, 1959, 1964). Because of its shape it is generally referred to as "the cigar" and is shown in Figure 2, extending northwards from Penola and labelled "P" in black. The fossil dunes interrupt the movement of surface run-off to the coast, and flooding of the inter-dune corridors was a major obstacle to settlement. Extensive drainage works over the last 100 years have alleviated this problem.

[FIGURE 2 OMITTED]

From the beginning of settlement, wool was the main agricultural product of the area. From the 1850s more intensive orchard production occurred on the Terra Rossa platform, attracted by the suitable soil but also by the fact that the slight elevation gave some protection from flooding. For the same reason, the main regional centre, Penola, and the small township of Coonawarra, have grown up on the platform, and the main north-south road and railway are strategically located on or close to this strip.

Viticulture and wine making also developed early and gradually replaced orchards as the main land use on the platform. John Riddoch had planted vines and in 1890-91 commenced wine making, using his own grapes and also some supplied by others who had planted vines rather than fruit trees. Others also attempted wine making, without much success, but William Redman established a winery in 1910, and achieved some commercial success, selling his wine to wineries outside the area, as did Riddoch. And so a wine region gradually evolved, and became widely known, associated with the name "Coonawarra", with an expansion of wine making in the years following the end of World War Two and the establishment of "Coonawarra" labels from the 1950s. Part of the image which has been promoted, rightly or wrongly, are the special and distinctive properties of the Terra Rossa soil for grape growing and wine making, and its association with the so-called "cigar".

The heart or core of the Coonawarra is undoubtedly the "cigar". But, especially since the 1940s, viticulture has expanded off the platform onto suitable well drained soils, including some Terra Rossa, but also onto related Rendzinas and other soils. Over the last 20 years or so the reputation of the Coonawarra for fine wines became firmly established. The ability to place the name "Coonawarra" on a wine label had become a very valuable asset in relation to market access, prices received for wines and grapes and thus also for land values. Growers and wine makers in the traditional core, on the Terra Rossa "cigar", came to resent the claims to their naming "rights" by the peripheral newcomers, despite the fact that some were in fact their suppliers and that some had used the name in good faith for decades. As yet the "Coonawarra Region" had no formal or legal definition. Figure 2 is based on Blackburn (1959, 1964) and shows his Soil Associations, together with Hundred boundaries and some other cultural landscape elements. Plotted and coded on this are some of the early proposals for Viticultural Regions which will be referred to later.

EARLY ATTEMPTS TO DEFINE THE COONAWARRA

The AWBC is the main controlling body of the industry, one of its functions being the recognition, in response to the wishes of the growers and winemakers in each region, of regional names and boundaries. Most grape growing regions have their own local industry organizations. At a meeting in October 1984 the Viticultural Council of the South East (VCSE) resolved that the term "Coonawarra" should be applied to the Hundreds of Penola and Comaum. In 1985 this was confirmed and the area described as the "Coonawarra Viticultural Area" (outlined dashed in black on Figure 2). The decision was based on the plantings shown on the 1984 Department of Lands topographic maps and which were largely confined to these two Hundreds and at that time mainly concentrated on the "cigar". This was the first formal, though not legal, definition. It was not unanimous and was made without the involvement of all interested parties. Nevertheless, almost 20 years later, it would assume legal significance.

In 1984 a local grower proposed a more extensive region, to be called "Penola", not Coonawarra, while in March 1992 an even more extensive proposal was submitted to the AWBC. This included the Hundreds of Killanoola, Monbulla, Nangwarry and Grey and also was to be known as "Penola", even though it did not include the Hundred of Penola nor the town. This is shown in Figure 2 in orange. The proposal was modified in May 1993 to include the town of Penola and an area to its south in the Hundred of Penola (shown in Figure 2 as a dotted orange line). This submission was withdrawn in August 1993 with the grape grower applying for inclusion within "Coonawarra".

In November 1993 the Coonawarra Grape Growers Association Inc. (CGGA) submitted an application to the AWBC for a regional ascription which was based on the "cigar" but excluded the town of Penola and included a number of detached "islands" of peripheral vineyards to the east and south. This was an attempt to define those with traditional affiliation with the name, and also to include those with specific commitments to viticulture. The areas are shown on Figure 2, shaded red. This "non-contiguity" was to cause problems.

FORMAL PROPOSALS: GIC

The Australian Wine and Brandy Corporation Amendment Act 1993 established the Geographical Indications Committee (GIC) as a separate statutory body that had the potential to avert threatened conflict. The committee comprises three members, an independent presiding member appointed for three years, a grape grower representative and one to represent the winemaking industry. There is no requirement that a member be a geographer! The non-presiding members have no specific term of office and can be replaced in any determination if conflict of interest arises, as happened in the case of Coonawarra. The Committee examines applications for regional definitions and then determines and formally defines the regional boundary.

An important feature of the legislation which set up the committee is that the GIC does not impose a regional boundary "from above", but arrives at a definition which reflects the wishes of the grape growers and wine makers involved. Related to this is the requirement that the process of defining regions should be "inclusive" not "exclusive". That is, the committee should look for reasons to include those who wish to be included, rather than reasons for keeping them out.

The Committee operates within a set of regulations and of particular relevance are Regulation 24 which attempts to define "region" and "sub-region" and Regulation 25 which sets out the variables which the Committee should consider when making a regional determination (Foale and Smith, 2001).

Regulation 24

A Sub-Region is an area of land that is "part of a region and is discrete and homogeneous in its grape growing attributes to a degree that is substantial". A Region is an area of land that "may comprise one or more sub-regions" and is a single tract of land that is "discrete and homogeneous in its grape growing attributes to a degree that is measurable and is less substantial than a sub-region".

Regulation 25

(1) Whether the region falls within the definition of a subregion, a region, a zone or any other area

(2) The history of the founding and development of the area, ascertained from local government records, newspaper archives, books, maps or other relevant material

(3) The existence in relation to the area of natural features, including rivers, contour lines and other topographical features;

(4) The existence in relation to the area of constructed features, including roads, railways, towns and buildings;

(5) The boundary of the area suggested in the application to the Committee under section 40R;

(6) Ordnance survey map grid references in relation to the area;

(7) Local government boundary in relation to the area;

(8) The existence in relation to the area of a word or expression to indicate that area including:

(i) Any history relating to the word or expression; and

(ii) Whether, and to what extent, the word or expression is known to wine retailers beyond the boundaries of the area; and

(iii) Whether and to what extent the word or expression has been traditionally used in the area or elsewhere, and

(iv) The appropriateness of the word or expression;

(9) The degree of discreteness and homogeneity of the proposed geographical indication in respect of the following attributes:

(i) The geological formation of the area;

(ii) The degree to which the climate of the area is uniform, having regard to the temperature, atmospheric pressure, humidity, rainfall, number of hours of sunshine and any other weather conditions experienced in the area throughout the year;

(iii) Whether the date on which harvesting a particular variety of wine grapes is expected to begin in the area is the same as the date on which harvesting grapes of the same variety is expected to begin in neighbouring areas;

(iv) Whether part or all of the area is within a single drainage basin;

(v) The availability of water from an irrigation scheme;

(vi) The elevation of the area;

(vii) Any plans for the development of the area proposed by Commonwealth, State or Municipal authorities;

(viii) Any other relevant traditional divisions within the area;

(ix) The history of grape and wine production in the area.

(10) The GIC was also not prohibited from having regard to any other relevant matters.

It can be seen that some of these regulations have dubious geographical value. For example, the use of map grid references, normally 6 digits, which can define locations to within only 100 metres. In any case, grid references are not surveyed points, so that their use in defining legal boundaries is of doubtful validity. The surveyed and legal framework of the cadastre is ignored. Atmospheric pressure is not spatially constant and so cannot be used to define a static boundary. But perhaps most important, to operationalise concepts such as "measurable", "discreetness", "homogeneity" and "less substantial" in grape growing attributes is difficult to say the least and no criteria for measurement are supplied.

The GIC proceeded on a programme to determine formally Australia's viticultural regions, within the framework of its regulations. For most of the determinations there was minimal controversy or conflict. However, the previous application of these Geographical Indications in determining some regions is puzzling. For example the McLaren Vale Region is hardly "homogeneous in its grape growing characteristics" for it includes dramatic variation in geology, topography, soils, land use and microclimate, apparently ignoring most of the regulatory criteria.

In December 1995 the CGGA and the Coonawarra Vignerons Association (CVA) jointly applied to the GIC for the determination of a region within, but not including the whole of, the two Hundreds of Penola and Comaum. This was to embrace all existing vineyards in those Hundreds including those where financial commitment to establishment had been made as at June 30, 1995, and where viticultural approval had been given by the Penola Viticultural Council (PVC). This appears to have been based on the CGGA submission of 1993 in that it encompassed the detached vineyards referred to above. It has not been plotted separately on Figure 2 because it was to be identical with the Interim Determination of the GIC.

This was a boundary of expediency. It did not conform to the GIC criteria in some respects, and ignored accepted regionalisation principles. It was however more compact than previous proposals and embraced the principle of "inclusivity". But in doing so it incorporated a large area which was generally regarded as unsuitable for viticulture, laying the foundation for possible future conflict.

Also in December 1995, the Koppamurra Grape Growers Association (KGGA) applied for a Determination for a Koppamurra Region, later to be known as "Wrattonbully". This will be commented upon later.

An application was made by a grape grower in June 1996 for a Coonawarra incorporating the Hundreds of Penola, Comaum, Joanna, Robertson, Killanoola and Monbulla (outlined in Figure 2 in yellow). The use of Hundreds as defining units suggests a surprising lack of appreciation of the subtleties and complexities of the environment and the vignerons' utilisation of it.

COONAWARRA INTERIM DETERMINATION

In April 1997, the GIC presented an Interim Determination for the Coonawarra Viticultural Region (shown outlined in black in Figure 2 and numbered as "1"), giving interested parties the opportunity to present objections prior to the Final Determination. This is clearly based on the 1995 CGGA and CVA proposal. To assist it the committee had access to considerable background material relating to geology, climate, geomorphology, soils, and history. Yet it appeared to have ignored much of this with a determination which simply restricted the region to land within the two Hundreds and related to the viticultural status at that time.

The 60 metre contour at the foot of the Naracoorte Range was used as part of the eastern boundary on the assumption that it represented the limit of the last marine incursion and coincided with the Kanawinka fault, neither of which is correct and which are of doubtful relevance anyway! The 60 metre contour was then used as a guide in the southeast of the region cutting through an area of uniform geology, soil, climate and hydrology. To the west, apparently retreating to the 1984 and 1995 proposals, the Hundred boundary was used which also has no relationship with the biophysical features. Furthermore, the convoluted extension to the east, extending well beyond the 60 metre contour, was designed to include the detached vineyards to the east of the East Naracoorte Range with "historical rights". A group to the north who felt that their historical rights were at least as valid was excluded, while those around, and to the south of, Penola were included. There were 33 submissions (objections) following the Interim Determination.

The lack of certainty regarding the north-west extent of Terra Rossa beyond the Hundred boundary was causing a reluctance to extend the boundary in that direction. Terra Rossa was assuming a dominant, though questionable, role, the Regulations requiring consideration of many variables other than soil in determining a Geographic Indication. However, in September 1997, a seismic/aeromagnetic survey which had been carried out by the mining sector, was used by one of the excluded growers to the north as "evidence" that the Terra Rossa extended beyond the Hundred boundary. Expert opinion on the validity of this conclusion was divided.

A meeting in February 1998 of the GIC, CGGA and the CVA discussed the Koppamurra application and in October 1998 the Wrattonbully Wine Industries Association (WWIA), formerly the Koppamurra Grape Growers Association, reapplied for a Wrattonbully Region, extending north to Naracoorte and from the State border to the western margin of the Naracoorte Ranges. This however was held in abeyance, presumably because its boundaries would depend upon the outcome of the Coonawarra Determination.

As a result of the considerable agitation and pressure, but still feeling that the information available was inadequate for a Final Determination, in February 1999, the GIC issued a "Variation to the Coonawarra Interim Determination" (shown in Figure 2 as "2"). Penola and the vineyards to the south remained but the easterly extension into and beyond the Naracoorte Range was excluded. Some recognition of the possible future existence of a Wrattonbully Region may have influenced this deletion. "Wrattonbully" was replacing "Koppamurra" as the preferred name for this region. The Naracoorte Range system was regarded by the GIC as a logical part of Wrattonbully. In addition, the well-established group of vineyards to the north was now included. This was explained by the GIC as a consequence of its acceptance of the evidence, though disputed, of the seismic survey, referred to above, that the Terra Rossa did indeed extend into this area. The boundary of Coonawarra now extended beyond the Hundred of Comaum! In order to accommodate one grower, a small area in the southeast was also attached. This was because the GIC recognised that at the time there was no potential region with which that grower could identify. This consideration presumably caused the retention of the Penola area (1,2), but was not used for others to the north and west.

The result was a more compact region, but still using much of the "logic" of the earlier determination. It still omitted some vineyards with a strong claim, and eliminated some which had been included in the first determination. And the more remote producers who considered that they had some claim, were still excluded. The number of objections to the variation increased to 49.

In June 1999 the newly formed Coonawarra Penola Wine Industries Association (CPWIA) applied for an extensive region, which is shown as a solid red line on Figure 2. It included all of the Hundred of Comaum, most of Penola, and parts of, Spence, Joanna, Robertson, Killanoola, Monbulla, Grey and Nangwarry. This was a group who were mainly peripheral to the "core" area of the two Hundreds and thus excluded by the GIC. The alleged ancient shoreline of the Cave Range was included in the west and the southern margins of Bool Lagoon in the north. The plains north-east of the Naracoorte Range and the State border bounded the eastern extent. The vineyards in the extreme south were also included. It was an all-encompassing proposal, which would have satisfied the applicants in one blow, but also cause considerable agitation amongst the coherent host of Coonawarra proper. It showed significant logic in the placement of the northern, western and southeastern boundaries. In effect it did away with the need for more than one regional determination to accommodate all interested parties.

In June 1999 a revised radiometric-based boundary was proposed, but again was not regarded as sufficiently valid, was not pursued and is not plotted on Figure 2.

In July 1999 the Naracoorte Lucindale District Council (DC) applied for the Hundreds of Penola, Monbulla, Comaum, Killanoola, Joanna and parts of Naracoorte and Jessie to be "Coonawarra" (shown in Figure 2 in blue). Interestingly, this incorporated the northern two thirds of the Penola DC but only a small part of Lucindale. This was rejected because, under the Act, a District Council is not entitled to apply.

In late 1999 the GIC met separately with the CGGA, CVA, a joint committee of the CGGA and CVA, the Limestone Coast Wine Industry Council, the CPWIA, and the WWIA and considered a range of options for a Coonawarrra region. It is clear from the documentation accompanying the Final Determination, which will be described later, that during these consultations the GIC was becoming convinced that a Coonawarra Region should be centred on the "cigar" and confined spatially. To cater for all those growers who would thus be excluded, the creation of two other adjacent regions would be required, to be called "Penola" and "Wrattonbully". And so, in April 2000, the CPWIA formally applied for a Determination based on this "three region concept".

A GEOGRAPHIC APPROACH

One of the grape growers who had been included in the first interim determination and then excluded in the second, realising the value of a holistic approach, sought to enrol two geographers (the authors) in order to press his claim. Convinced that the grape grower had a reasonable case, not only because of historical association with the "Coonawarra", but also because of the biophysical environment of the vineyard, the geographers set about preparing their argument. This they did by bringing together what they considered to be the important spatial variables including the geology and geomorphological evolution of the region, soils, topography, hydrology, climate, community of interest and the cadastre, attempting throughout to conform to the GIC regulations. Climate constituted a vital element and its various components, obtained from meteorological stations across the area, have been plotted and compared (Figure 3a). Climate obviously varies little across the area. Topography is plotted along the line W-X-Y on Figure 1 and is shown as Figure 3b with vertical exaggerations of both 25:1 and 1:1. This demonstrates the relative topographical insignificance of the Naracoorte Range complex. Figure 3b also includes the local geology and the location of the Kanawinka Fault is shown. The codes will be referred to later.

[FIGURE 3 OMITTED]

An important source of data became available about this time. This was the mapping of Land Systems by the Department of Primary Industries and Resources South Australia (PIRSA) with supplementary data as "Soil Landscapes" (PIRSA, 1994). This provided further and more detailed information on the 21 relevant features which go to make up the Land System classification. They include soils, topography, hydrology and salinity. This mapping refined and to a large extent supplanted the work of Blackburn and others, and gave further strength and validity to the decision making process. Figure 4 shows a PIRSA Land System plot of the Coonawarra area, or district, and describes the Land System characteristics and their codes. The original is in colour.

[FIGURE 4 OMITTED]

The geographers concluded that the physical and cultural environment of this part of the South-East of South Australia is remarkably homogeneous. Regions which had already been defined by the GIC, for example the Barossa and McLaren Vale, contained far greater biophysical variation than was present in this area. This fact, in one sense, set a precedent, and opened the way for a wider interpretation of the GIC regulations. But the use of precedent is not logical if the precedent is flawed. On the other hand, perhaps it emphasised the need for a more rigorous application of the regulations. A problem in applying the regulations is the meaning of the terms "discrete" and "homogeneous". It is not unreasonable in these circumstances to assess their meaning against previous, noncontentious determinations. In fact the GIC proposal appeared to have the characteristics of a "sub-region" located within a wider "Coonawarra Region" yet to be defined, rather than a region in itself.

These factors presented problems. Where in a relatively uniform physical/social landscape can a boundary be drawn? Since viticulture is very much dependent upon the physical nature of the land and all the other relevant factors appeared either too homogeneous or poorly identified or difficult to identify, it seemed to the geographers that the PIRSA Land Systems should play a major role in defining the Coonawarra Region. The PIRSA Land Systems incorporated the most relevant physical determinants, apart from what many see as the most important, namely climate, which in any case varied little across the area. In addition, the land system boundaries had been mapped in some detail, though legal definition of a regional boundary would require that they be related to the cadastre. However, the spatial relationship between the Land Systems and the cadastre was clearly strong enough for reasonably detailed and accurate plotting. A further problem lies in the fact that the natural Land Systems continue into Victoria. A rational consequence would be a region which extended across State boundaries. But though the state border constitutes an illogical social, economic, administrative and political "frontier", it is probably too well fortified to penetrate.

Using the PIRSA Land Systems, together with other researched data, including climatic data, the geographers attempted a regional definition, and in July 1999 submitted a detailed preliminary report which included a proposed boundary. The purpose of the exercise was to define, in a logical and consistent manner, a Coonawarra Region which encompassed the areas most suitable for viticulture and at the same time remain compact and coherent and avoid potential future problems. The logical way to portray this was in cartographic form. This is shown on Figure 2 as a dotted yellow line and again in Figures 5 and 6. The fact that the GIC determination had produced a relatively compact and cohesive region was recognised and kept in mind.

[FIGURES 5-6 OMITTED]

Preliminary plotting, by hand, of the PIRSA Land Systems and Soil Landscapes on the 1:50 000 topo/cadastral maps provided some valuable insights into the physical and cultural relationships. Even though the cadastre, as a matter of historical expediency, tended to be oriented to the cardinal points, there was clear evidence that it had, in parts, accommodated existing biophysical and cultural elements in the landscape. Field reconnaissance reinforced the geographers' confidence in both PIRSA's and the Department of Lands' methodology and consequently, the validity of using their data as the basis for boundary plotting. Strong correlation was observed between the mapped Land Systems and the land use, especially the location of grape production, and the terrain and vegetation. Again, this reinforced confidence in the Land System methodology and in the ability of vignerons to make the correct land use decisions, at least in relation to the biophysical environment.

The base map for Figure 5 is a computer plot of PIRSA data, showing the Land Systems from Figure 4; topography, with the 60 metre contour emphasised; surface hydrology; the cadastre in magenta, with Hundred boundaries emphasised. The Kanawinka and Lucindale faults are portrayed in deep brown. The earlier proposals portrayed in Figure 2 have been omitted for the sake of clarity while other boundaries shown on the map will be referred to later.

The inclusion of the Naracoorte Land System (NRC) to the east was a major consideration, and was based on the fact that it was derived from the marine sediments of the plain to the west, giving it a direct genetic and generic relationship with the "Coonawarra" as defined by the GIC. It also included some soils well suited to viticulture including Terra Rossa and related soils. In addition, the escarpment, to the crest of the Range, has a greater locational association with the western plains that it faces than with the relatively higher plains to the east. The eastern margin of this System and the cadastre are largely co-incident with the eastern shoulder of the East Naracoorte Range. The geographers were also aware of the existence of some vineyards on the plains to the north-east, to which the name Wrattonbully is loosely applied, and considered that their "Coonawarra" should not extend too far in that direction for fear of undermining the integrity of that area which was almost certain to be formally defined in the near future.

The geographers were not formally required to provide a cartographic definition, but merely to present a case for the inclusion in a "Coonawarra", wherever the boundary was placed, of the grape grower who had sought their services. However, geographers, being what they are, felt impelled to express their arguments spatially and included a plot of the boundary of what they considered would be a logical Coonawarra.

Having proposed inclusion of the Naracoorte Range, the problem was to link it to the GIC determination or to follow it with such modifications as seemed appropriate in view of the above observations. The eastern margin of the NRC Land System was therefore chosen as the north-eastern boundary. To the north, a surveyed road cutting through a low point in the range, was selected as a logical border. This entered the western plain to the north of the GIC boundary but close to the common boundary of the Maaoope (MAO) and Glen Roy (GLR) Land Systems, the former being generally too prone to waterlogging for viticulture. This common boundary was followed briefly in the north, and although logic would have seen the boundary continue along the edge of the MAO, in the absence of PIRSA data for that area at that time, a straight line boundary was the best option, through the GLR, to meet up once more with the boundary between the GLR and MAO Land Systems. This was then followed southwards until meeting the boundary of the Krongart (KRG) Land System. In doing this, a significant area of GLR to the south-west, partially separated from the main body by a small tongue of MAO, was excluded, although logic and consistency would have it included. The perceived need for restraint influenced this decision. Since the KRG Land System was also considered generally unsuited for viticulture, though not to the same extent as the MAO Land System, its northern boundary, coincident with the GLR was followed until its intersection with the Nangwarry (NGW) Land System. This also is considered unsuitable for viticulture and is largely planted to pine. The boundary between the NGW and Kalangadoo (KLN) Land Systems was followed to its intersection with the Victorian border, which was then followed northward to link up with the point of origin. A plot and preliminary report were submitted to the GIC (Foale and Smith, 1999).

THE FINAL DETERMINATION

In its "Statement of Reasons" (GIC, 2000) the GIC explained it had considered a range of options in arriving at its Final Determination:

1. Region "Greater Coonawarra" extending from Naracoorte to south of Penola with subregions, "Coonawarra", "Wrattonbully", "Penola"

2. Regions "Wrattonbully" and "Penola" with subregion "Coonawarra".

3. Regions "Greater Coonawarra" and Wrattonbully" with subregion "Coonawarra".

4. Regions "Coonawarra" and "Wrattonbully" with subregion "Coonawarra Terra Rossa".

5. Regions "Coonawarra" and "Wrattonbully".

6. Regions "Coonawarra", "Penola" and "Wrattonbully".

Having considered all the evidence and all the proposals and all the conflicting interests, while one member of the committee favoured the two region concept (option 5), the GIC decided, by a majority of two to one, to adopt the "three region solution" (option 6), and in May 2000 issued their Final Determination for the Coonawarra, and at the same time Interim Determinations for a "Wrattonbully Region" and a "Penola Region". These boundaries have been plotted in green on Figure 2 to permit comparison with earlier proposals and on Figure 5 (though largely obscured by later boundaries), and the regions coded "C", "P" and "W". The decision involved a "Coonawarra" at the core and a large "Penola Region" surrounding it on all sides except the north-east where "Wrattonbully" completed the encirclement.

The "Coonawarra" itself was based on the GIC Variation to the Interim Determination, but with some amendments. Vineyards in the north (area labelled "2" in Figure 2) had been excluded in the Interim Determination, included in the Variation and excluded in the Final Determination. These now became part of "Penola"! To the south (area labelled "1,2" in Figure 2) vineyards included in the Interim Determination and its Variation were now ceded to Penola, which also annexed the small area in the south-east ("2" in Figure 2). The area in the Naracoorte Ranges in the north-east which had been included in the Interim determination and then excluded in the Variation (labelled "1" in Figure 2) was now won by "Wrattonbully".

The "Coonawarra" thus reverted to the core area marked "1,2,3" on Figure 2.

"Penola" followed, in essence, the June 1999 proposal, but surrendered the north-east sector, beyond the Naracoorte Ranges, to the extensive "Wrattonbully". The Penola Region extended to Bool Lagoon in the north and beyond the Cave Range in the west. The most remote vineyards in the south were also included as well as the town of Penola and also some of the forested areas to the southeast of Coonawarra. The State border again served as a boundary in the east.

The area to the north-east of "Coonawarra" was designated "Wrattonbully", reflecting the WWIA proposal of 1998, and extended ambitiously to the north of Naracoorte, encompassing the vineyards there, to the Victorian border in the east, and the western toe of the Naracoorte Range in the west. And so now all the vineyards were in the potential control and protection of a benevolent authority. Coonawarra had been surrounded!

The Determination was especially illogical since the town of Penola, which was excluded from "Coonawarra", is the main service centre for the whole of this part of the South-East. A Coonawarra Region which does not include the main service centre makes no sense. Thus a grape grower from around Bool Lagoon in the north of the proposed Penola Region, travelling south along the main road to Penola, would enter the Coonawarra Region, pass through the township of Coonawarra, leave the Coonawarra Region and re-enter the Penola region, arrive at the town of Penola and pick up his supplies or deliver his grapes!! In effect it made Coonawarra a sub-region of "Penola".

In the "Reasons for the Determination" a new criterion was introduced to justify the placement of the western boundary so as to include some, but not all, of the Glen Roy (GLR) Land System, and the eastern boundary to include some, not all, of the Kalangadoo (KLN) Land System, namely proximity to the "cigar"! Is it possible that winds could blow some of the grape growing goodness of the "cigar" onto adjacent areas? At the same time the area around and to the south of Penola, on the same land system as the "cigar"--the Penola Land System--and more "proximate" than large areas which had been included, had been excluded and allocated to the proposed Penola Region.

The result was a boundary which became a frontier of dissent and which did not meet the needs of either camp. Unlike the traditional frontier, dissidents occurred in both camps and the aggressors were not trying to storm the frontier with rape and pillage in mind, but on the contrary were trying to persuade those blessed by inclusion to accept them by extending their frontier. The aggressors were in fact asking to be conquered so that they could share the bounty that the name "Coonawarra" would bring. However, their desires were resented and resisted by those grape growers and wine makers who had been included.

There were some subtle and long-term problems hidden within the GIC determination. For example, areas fortuitously included within the new boundary which were not at the time used for grape production might in the future be so developed, thus reaping the benefit of the Coonawarra name, without the "historical" right or the expense and trauma of litigation. This would also have an obvious impact on land values. Some included landholders may abandon viticulture. Others, on land with identical grape growing characteristics as areas within the boundary, and with long term association with the region and the name, were excluded. Some of these were literally on the other side of a property boundary fence. Long-term users of the name who were excluded would look with anger at the newcomers over the fence. A decision based on the situation at an arbitrary date is not logical. Human history is a continuum, compared with which the physical environment is relatively stable.

The determination of the boundary, based in part on precedence and in part on dubious interpretation of the biophysical and cultural elements of the landscape, also posed a potential problem for the likely future establishment of adjacent viticultural regions. Their integrity should not be compromised by premature inclusions. Division based on factions, has shown over the centuries to breed dissent and conflict, often not surfacing for decades. And if significant areas incorporating existing and potential vineyards are excluded, where should their affiliation lie?

ADMINISTRATIVE APPEALS TRIBUNAL 

The result of this Determination was to be expected. Some 46 of the excluded grape growers appealed against the decision of the GIC, their appeals to be heard before the Administrative Appeals Tribunal (AAT). Initially, individuals or small groups of growers contracted solicitors to handle their claims. There appears to have been little coordination at first, with each Applicant seeking support from expert witnesses in a bewildering range of topics which reflected, in part, their own views of what issues were important and applied to them. All of this material was presented in a series of reports with little cohesion. One of the solicitors realised the problems inherent in the individual approach and sought to bring the Applicants together to mount a coordinated attack in the form of a Class Action.

In response, some of those who had been included in the GIC region, in particular those in the traditional nucleus of the Terra Rossa "cigar", continued to oppose those whom they saw as attempting to usurp the name "Coonawarra". These, the Respondents, amongst whom were some powerful wine making companies, enrolled their own expert witnesses, some even from overseas, and prepared their own set of reports, equally biased and lacking in understanding of the principles of regionalisation. This was arming for war!

What was at issue was a set of interrelated, biophysical and cultural, spatial variables. Yet none of the Applicants or Respondents thought to provide a solution to the problem either verbally or even in the form of a map, which is of course the ultimate graphic expression of these spatial variables. However, the solicitors for the Applicants realised the essential role of the map and set the same geographers another task. Assuming that all Applicants had their appeals upheld for whatever reasons, that is, the Appeals Court accepted that they all had the right to belong to a Coonawarra Region, where should the boundary be located? The bases for the claims were varied and included for example long association with wineries in the Coonawarra/Penola area, the presence on their properties of Terra Rossa soils, and a similar climate to the region defined in the Final Determination. Many other themes, of varied relevance, were developed.

By this time the services of a GIS consultant were made available to the Applicants which facilitated the mapping and spatial correlation of data. In particular the locations of the properties of the 46 applicants could be correlated with the other data--Land Systems, cadastre, GIC determinations and a range of biophysical variables. The holdings of these Applicants are plotted in black on Figure 6.

The boundary proposed by the geographers took account of the PIRSA Land Systems, paying particular attention to their suitability for viticulture, as well as climate, hydrology, community of interest and the cadastre, as they had done previously. The principle of "inclusivity" was a major consideration. The need to address the possible inclusion of the applicants whose properties fell within the guidelines resulted in an expanded boundary, extending to the Cave Range in the west, areas to the south and to the north-east on the Pinnaroo Block. To the north and north-west there is significant coincidence with the GIC's Interim Determination for a Penola Region, although the geographers' boundary was arrived at independently. Some fine tuning was employed in the south-east sector.

In fact the boundary cut through some properties at the edge of the region which contained areas of both favourable and unfavourable systems. In striking fashion, these applicants all had their vines on those parts of their holdings which were on the suitable Land System, and thus included.

The inevitable inclusion of non-viticultural land within the proposed region was in keeping with normal regional practice, but was a cause for concern among some of the legal fraternity. But few regions are monocultural. This was a new "frontier", which was relatively more extended, diffuse, and in some parts less viable, particularly in the west, extreme south and in the north-east. Elsewhere it made good sense.

The boundary was presented in map form to the Tribunal (shown as a thin, solid, yellow line on Figure 5), as well as a detailed report which included a summary of the geographers' researched data (Foale and Smith, 2001).

The geographers had earlier aired the possibility of a region and sub-region solution, which would conform to the regulations. The Barossa Region for example contains two subregions, the Barossa Valley and the Eden Valley. The suggestion was for the Terra Rossa "cigar" to form a compact sub-region within a larger and less compact region. "The Coonawarra" and "Greater Coonawarra" were suggested as regional names. But this idea pleased neither the besieged nor the besiegers!

The potential extension of the "frontier" only fuelled the resistance of the besieged still further. The class action, supported now by a battalion of legal reinforcements, in turn backed up with an army of expert witnesses, but confronted by a similar force, was heard before the AAT in March and April 2001.

The hearing before the AAT was held over 25 days with the final decision to be made by the presiding judge. The army of expert witnesses on both sides presented a bewildering array of evidence, much of which was irrelevant and served only to cloud the issues. The battalion of barristers subjected the witnesses to a constant and fierce barrage of crossfire.

It had become apparent during the proceedings that as information relating to the biophysical, social, economic and cultural environment crossed the divide into the legal domain there arose problems of interpretation and understanding, a conceptual-legal "frontier". Thus it was difficult to incorporate within the rigid legal process, the principle that a region may be defined differently depending upon the criteria used and the relative weights placed on those criteria.

Thus, when the hearing closed there was a feeling amongst many of the applicants that the essentials had not been fully understood. The battlefield was unfortunately confrontational rather than conferential and so counterproductive. The real legal and conceptual frontier had not been resolved. An uneasy peace lasted until the Tribunal's finding was handed down.

THE JUDGEMENT

It took five months for the Appeal Court to reach its decision, which was announced in October 2001. In spite of the weeks of legal argument, the volumes of evidence relating to topography, geology, climate, soils, hydrology, salinity, land systems, grape growing requirements, factors affecting grape quality, timing of harvest, marketing image and historical affiliation, and the processions of expert witnesses, the decision appears to have been based largely on the historical event that in 1984 there had been partial agreement amongst some growers in the area that "Coonawarra" should be confined to the two Hundreds of Penola and Comaum. The straight lines of the Hundred boundaries are easily understood!

The boundary was defined by text only, no map being presented. However the boundary has been plotted by the authors as a dotted line coloured red-brown on Figure 5. Again, a lack of understanding is evident, this time by the Appeals Tribunal. The legacy of the two Hundreds remained. In the north, the Hundred boundary was used in part, but with a diversion to include one lucky Applicant. A few of those included in the First Interim Determination but excluded in the Final were fortunate, though many with demonstrated rights were excluded. To the west, despite the general criticism of the use of the Hundreds as regional boundaries, even greater use was made, and the full western edge of the two Hundreds formed the boundary. Part of the southern boundary also followed the Hundred, and then linked with the GIC Interim Determinations and with the geographers' boundary, from where it deviated into the pine plantations of the Nangwarry Land System. The southerly extension included all of the Applicants in the Penola area who had been excluded in the Final Determination. The former extension to the north-east was resurrected and extended with a considerable length of the Victoria-South Australia border used as the regional boundary. This linked through the Naracoorte Ranges, with the former 60 metre contour, to rejoin the Hundred boundary to the north. The effect was to include some of the Applicants as well as other actual or potential grape growers, but exclude others (shown on Figure 6).

The AAT directed the GIC to implement its boundary but the GIC had difficulty in interpreting the AAT description. In the event, it set aside plotting because of the probability of further appeals.

The decision of the Appeals Tribunal not only replaced the Coonawarra Final Determination of the GIC but also made obsolete its interim proposals for Penola and Wrattonbully Regions. With 24 of the contestants satisfied, by inclusion within the new boundary, and the remainder exhausted, for a moment the battlefield appeared quiet. But five of the remaining 22 decided to fight a rear guard battle. The fragile frontier was challenged once again. There again remained "unoccupied" land, with no allegiance and a latent desire for membership; and the traditional tribe still harboured resentment towards the interlopers.

But, would the Wrattonbullies descend from their upland ramparts and invade the northern plains? Or would the Naracoortiers penetrate the northern swamps and lagoons with the same object in mind? Will the Cave Rangers cast envious eyes beyond the western swamps or the Gambier Mounted infiltrate the southern plains? Would the Naracoortiers resent the dominance of the Wrattonbullies and force them back beyond Mosquito Creek? Could the Coonawarriors resist these multiple pressures? The result was awaited with eager apprehension.

FEDERAL COURT APPEAL

Early in 2002, five Applicants who had been excluded from the AAT Determination, took their appeals to the Federal Court of Australia (FCA). The properties of three of these applicants are located very close to the AAT boundary in the north and two are close in the west. Following hard on their heels and opposing their inclusion, were some of the "core" grape growers and wine makers from inside the frontier, as well as the GIC, the CGGA and the CVA. The locations of these five Applicants are not identified on the map.

In September 2002 the FCA found in favour of the five Applicants. In doing so the FCA had overturned the AAT Determination, which in turn had set aside the GIC Interim and Final Determinations. The FCA instructed the legal representatives of the five and those of the Respondents, along with the GIC, to agree on a boundary that would include the five but not include the remaining Applicants whose appeal had been rejected by the Appeals Tribunal but had not pursued their case in the higher court. At the same time, those who were previously included should remain so. This boundary was then to be remitted to the GIC for formal definition.

In its "Reasons for Judgment" the Court was critical of the AAT in a number of areas and in particular for its use of the Hundred boundaries of Penola and Comaum as an "exclusionary device", as the major determinant of who should and should not be included in the region. This the FCA criticised as a misunderstanding of Regulation 25. The intent of a geographical indication is that it should:

   ... indicate a wine, the characteristics of
   which are essentially attributable to the
   region where the grapes, from which it is
   made are grown. The characteristics of wine
   essentially attributable to the region where
   the grapes are grown will not be influenced
   by the location within that region of local
   government or land survey boundaries
   administratively fixed for reasons unrelated
   to soil, climate or other conditions which bear
   on grapevine horticulture (Federal Court of
   Australia, 2002, paragraphs 58 and 59).

In using the Hundred boundaries the AAT had also ignored the requirements of Regulation 24.

   In the definition of "region" there is not the
   slightest hint that pre-existing administrative
   boundaries, such as those fixed for local
   government or land survey purposes for
   reasons unrelated to grape growing attributes
   of the land, could have any relevance to
   identifying the relevant single tract of land that
   is discrete and homogeneous in its grape
   growing attributes to the requisite degree
   (Federal Court of Australia, 2002, paragraph
   65).

The FCA downgraded the significance of the 1984 Viticultural Council resolution confining the Coonawarra to the two Hundreds of Penola and Comaum on which the AAT had placed such importance. The resolution had not been presented to members in advance of the meeting and

   ... it seems clear that the boundary proposed
   by the resolution was not a boundary based
   on any research or expert evidence. At the
   most, the resolution provides some evidence
   that some viticulturists knowledgeable in
   and about the area considered that the region
   which produced grapes from which wine
   recognised as "Coonawarra" was made,
   extended beyond--and well beyond--the
   cigar (Federal Court of Australia, 2002,
   paragraph 68).

In effect the Court was questioning the present relevance of the "cigar" in defining the boundary. For the AAT "proximity" to the cigar had been an important factor and in its "Reasons" had argued:

   Whether one characterises it as a "marketing
   tool" or even challenges the homogeneity of
   the cigar itself, it is historically and
   scientifically the signature of the Coonawarra
   Wine Region. Proximity to this strip of arable
   soil would be in our view, an important factor
   in the determination of the boundary. Because
   of this, we do not consider, at this time, unless
   an overwhelming countervailing reason was
   demonstrated that land outside the two
   Hundreds and not proximate to the topography
   of the cigar could justify inclusion in a
   Coonawarra Wine Region (Administrative
   Appeals Tribunal, 2001, paragraph 137).

Whatever the validity of this view, it had not been used consistently or logically by the AAT, the five Applicants being far more "proximate" than many whom the AAT had included. Thus the FCA pointed out that:

   The eastern boundary of the GIC
   Determination included land to the east of
   the cigar--at some points extending five to
   six and a half kilometres to the east. The
   GIC determination to the north barely
   extended beyond the cigar, and to the south
   did not include all of the cigar on the basis
   that the township of Penola should not be
   included in the Determination (Federal
   Court of Australia, 2002, paragraph 72).

The AAT boundary extended the eastern boundary even further, "as far as the Victorian border".

Finally, the AAT had excluded some of the Applicants on the ground that inclusion would require the incorporation of an intervening area, which would compromise the homogeneity of the region

   Inclusion would involve the addition of a
   tract of intervening unsuitable ground to the
   region which would affect its homogeneity.
   This factor outweighs, in our view, the
   particular historical evidence of connection
   to the Coonawarra Region presented to us
   (Administrative Appeals Tribunal, 2001,
   paragraph 166).

The FCA dismissed this proposition

   Having regard to the evidence that the Terra
   Rossa soil type typical of the cigar did not
   constitute more than 23 to 25 per cent of the
   area within the GIC determination, and
   apparently a lower percentage in the case of
   the area encompassed by the AAT
   Determination, the presence of some
   unsuitable land that might be included in the
   extension hardly seems a factor that could
   carry weight sufficient to overcome what the
   AAT has described as historical evidence of
   connection to the Coonawarra Region (Federal
   Court of Australia, 2002, paragraph 87).

Following the FCA instructions, a new boundary to include the five Applicants was eventually agreed upon and is shown in red on Figure 6. In spite of the major criticisms directed at the Tribunal, in effect the new boundary simply reproduces that of the AAT with small extensions to embrace the five Applicants, 3 in the north and 2 in the west. An extension into the plantation forest area in the south-east is also evident. It appears that the "proximity principle" has been applied yet again but this time not in relation to the cigar but to the AAT's discredited Hundred boundary. It seems that the FCA, following the lead of the GIC, the AAT and their corps of legal advisors, have used the Applicants' full property boundaries as their boundary, rather than follow the more logical course of including only those sections which are suitable for viticulture. Large areas to the north and west remain outside the region. Where will their allegiance lie? And how will a future Wrattonbully regard the enclave in the south?

An arbitrary boundary, based upon the inclusion of a group of vineyards claiming some historical/commercial rights at a particular arbitrary date and within a wider homogeneous environment has little logic, unless other factors are incorporated. The foundation of the viticultural structure is surely the biophysical environment. This has been ignored, totally. It is relatively permanent while vested interest is notoriously fickle, short sighted, self seeking and impermanent.

CONCLUSION

Despite Interim Determinations, a Final Determination, a class appeal before the Australian Appeals Tribunal, a Final Judgement from that court, an Appeal to the Australian Federal Court and a Judgment from that court, dissatisfaction and conflict persist. Arguably, there has been no satisfactory conclusion.

ADDENDUM 

Given the concerns raised above relating to both the AAT judgement and the instructions of the Federal Court, is it possible to propose a more logical region?

The principle of "homogeneity" is incorporated in the Regulations and was emphasised by both the GIC and AAT. However, as has been stated, it is extremely rare for any geographical region not to display variation of some sort, except in very limited and specialised locations. The GIC and AAT Determinations, as well as the other proposals made, all contain a variety of environments. The existence of non-homogeneous intervening land is thus not necessarily a valid reason for excluding peripheral vineyards. Furthermore, their exclusion leaves them in "geographical limbo", with the options of remaining so, or affiliating with another, and possibly less appropriate region or establishing their own region. A problem with the second policy is that the regulations require that a region must produce a minimum of 500 tonnes and comprise at least five vineyards.

The best approach is surely to draw the boundary on principles of regional geography, to its maximum reasonable and definable extent. This would then automatically include all those with existing, and potential viticultural/regional characteristics. Those excluded would logically belong to another region.

Two possibilities are suggested here. Both take account of the relatively uniform environment in the area and both place heavy reliance on the PIRSA Land Systems in placing the boundaries and, of course, the cadastre. They also attempt to incorporate the principle of compactness and cohesion.

The first, in Figure 6, shown in solid yellow, slightly heavier to distinguish it from previous boundaries, abandons the use of the Hundred boundaries, cutting as they do through areas of uniform climate and Land Systems. To the north-east, the eastern shoulder of the East Naracoorte Range is a logical boundary as was initially proposed by the geographers. In this area, five of the original Applicants would remain excluded, although they have strong claims for inclusion on historical grounds. To include them would represent a possible incursion into a region becoming known unofficially as "Wrattonbully" and would present problems in the future. They lie beyond the geographers' suggested eastern piedmont. History moves forward.

To the north, a logical way through the East Naracoorte Range via a surveyed road is suggested, which follows the geographers' second proposal, and then follows the southern margin of Bool Lagoon and its inlet and outlet channels, even though this would include an area of the MAO Land System which is generally not suited to viticulture. The adoption of this boundary would include a number of grape growers who were excluded by the GIC in its Final Determination and by the AAT, at least some of whom have historical claims. It would also be even more difficult and illogical in the future to define and fit these growers into a separate region to the north or east.

The western boundary follows and then cuts south through the MAO Land System to meet the interface of the MAO and GLR near its most westerly extension and follows this interface southwards. A small southerly extension of the MAO occurs to the south, which is virtually detached from the main body, and which isolates a significant body of GLR to its west. This is included by following the GLR/MAO interface to the south-west. The MAO extension is thus included as an acceptable anomaly, but the GLR pocket is thus saved from virtual isolation. The KRG System to the south is also regarded as not generally suited to viticulture so that its interface with the GLR is followed eastward until the boundary of the geographer's initial proposal is reached (shown on Figure 5). From here, this boundary is followed, with some fine-tuning, in the south and south-east sections.

This regional definition is a reasonable compromise, incorporating virtually all of the accepted or proposed viticultural terroir and complying with all relevant regulations. It is also reasonably compact. Those who are contiguous to the boundary and have "historical" rights, could be included, should no other options be available to them.

If the geographers' "Coonawarra logic" is applied to the future definition of a Wrattonbully, that region might have some claim to the east-facing escarpment and crest of the East Naracoorte Range. A claim to the whole of the range could not be substantiated due to the argument that the NRC Land System is derived from the sediments of the western plains. The placement of the Coonawarra eastern boundary at the eastern margin could be construed as an incursion. The reverse situation is evident in the Interim Wrattonbully proposal accompanying the GIC Final Determination (Figure 6), which seceded to Wrattonbully the whole of the East and West Naracoorte Ranges. A compromise may be necessary by locating the common boundary between Coonawarra and Wrattonbully within the swale separating the West and East Naracoorte Ranges (shown as a dashed yellow line on Figure 6). Thus the ranges are shared between the contestants. This represents what may be termed a geographical/viticultural region. Its validity is not compromised by ephemeral affiliations.

While this suggested region would include a number of Applicants who had been excluded from the ATT determination it would still exclude some, for example those near to the Cave Range in the west. If these Applicants are successful in having their rights to belong to the Coonawarra Region recognised, for whatever reasons, then a logical boundary for the region would be the one presented by the geographers to the AAT and previously described and plotted in Figure 5 and reproduced also in Figure 6. Those in the west, near to the Cave Range, would seem to have a reasonable case given the fact that the legislation establishing the GIC incorporated the principle of inclusivity. In addition, there is no viticulture immediately to the west of those Applicants and therefore no immediate prospect of a separate viticultural region in that area to which they might give their allegiance.

A problem arising from this hypothetical extension to the west is that, in order to include what at present appear to be merely two grape growers, the region will have to be increased in size by 160 square kilometres, because the Regulations do not allow detached "islands". As a by-product a number of viticulturalists who had not appealed, or had abandoned their appeal, would be included, as well as others who may, in the future, attempt viticulture and so gain the historical benefit.

The logical boundary in the south-west would then follow the interface between the KRG and GLR Land Systems to join up with the southern boundary of the compact region described above (Figure 6) which it would then follow eastwards and northwards.

Figure 7 has been prepared, as a summary, in order to "clarify" the situation, and incorporates all of the well-defined and/or formal regional proposals. A simple base of Hundred boundaries is used, shown as dashed lines except where they have been used as regional boundaries. Solid lines are used for the various proposals, thickened where boundaries coincide. The geographers' proposal for a two region solution involves the "compact" proposal shown in figure 6 in solid yellow as the "Greater Coonawarra" and the core region shown with short dashed lines and suggested as "The Coonawarra".

[FIGURE 7 OMITTED]

Is there logic entangled somewhere in all this?

ACKNOWLEDGEMENTS

Special thanks to Chris Deardon of Daedalus (SA) Pty Ltd for his patience and cooperation in the preparation and production of the computer maps and diagrams. His combination of geographical understanding and GIS skills was invaluable.

REFERENCES

Administrative Appeals Tribunal (2001), Nos. S200/182, 183, 186-227, 305 and 313, Decision and Reasons for Decision, Adelaide, October, 2001.

Blackburn, G. (1959), Soils of County Grey, South Australia, Soil and Land Use Series No. 33, CSIRO, Melbourne.

Blackburn, G. (1964), The Soils of Counties MacDonnell and Robe, South Australia, Soil and Land Use Series No. 45, CSIRO, Melbourne.

Blackburn, G, Bond, R.D. and Clarke, A.R.P. (1965), Soil Development Associated with Stranded Beach Ridges in South-East South Australia, Soil Publication No. 22, CSIRO, Melbourne.

Drexel, J.F. and Preiss, W.V. (eds) (1995), The Geology of South Australia, Bulletin No. 54, Volume 2, Department of Mines and Energy, South Australia.

Federal Court of Australia (2002), FCAF 295, Beringer Blass Wine Estates Limited v Geographical Indications Committee, Reasons for Judgment, Adelaide, September, 2002.

Foale, M.R. and Smith, D.L. (1999), The Coonawarra Region, Report to the Geographical Indications Committee.

Foale, M.R. and Smith, D.L. (2001), The Coonawarra Region, Report to the Administrative Appeals Tribunal, No. S2000/182, 183, 186-227, 305 and V2000/668, 676.

GIC (2000), Statement of Findings on Material questions of Fact and reasons for Decision: Final Determination of the Geographical Indication of the Coonawarra Region, Geographical Indications Committee, June, 2000.

O'Driscoll, E.P.D. (1960), The Hydrology of the Murray Basin Province in South Australia, Bulletin No. 35, 2vols., Department of Mines, South Australia.

O'Mara, J.H. (1957), Surface Structural Geology of Selected Areas in the Gambier-Robe-Naracoorte District, Internal Report, Department of Mines, South Australia.

PIRSA (1994), Land Systems, South East of South Australia, Primary Industries and Resources, South Australia.

Sprigg, R.C. (1952), The Geology of the South-East Province, South Australia, with Special Reference to Quaternary Coast-line Migrations and Modern Beach Developments, Bulletin No. 29, Geological Survey of South Australia, Department of Mines, South Australia.

Max Foale, former Map Librarian and Cartographer in the Geography Department, The University of Adelaide.

Derek Smith, Senior Lecturer in the Department of Geographical and Environmental Studies, The University of Adelaide. [e-mail]