AN ANALYSIS OF LINEAR MARKINGS AS A FORM OF WRITING

Kevin J. Sharpe

 

Introduction

Across the world – at least firmly represented in Europe and Australia – is a tradition from Pleistocene times of marking objects in lines, often in sets of more-or-less parallel lines. The lines in this tradition are called meanders (sometimes also referred to as macaroni), and can be of engravings or scratches on cave walls, on rocks portable or permanent, on bones, or can be of finger tracings on suitable surfaces.

This paper centers on an Australian site of the meander tradition, that of Koonalda Cave in South Australia. Christine Whitehead and I first visited the Cave in 1973 as part of an archaeological expedition led by Alexander Gallus; our task was to draw and photograph the known wall finger markings and engravings at the rear of the upper chamber section of the Cave. But in the daily round of climbing and clambering to our workplace, we discovered that a considerable area of the floor of the upper chamber contained smooth and rounded boulders which were engraved, plus prehistoric torches, and animal bones. All these dating to around 19,000 years B.P

I have visited Koonalda only once since that time, at the beginning of 1976. But all along my problem has been how to understand and approach the meanders. At first, I saw the boulder engravings as defining pathways through the upper chamber, and being made as an activity within the rituals performed in it. Important also seemed to be the aesthetic or artistic element: they exhibit an interest in the holes, cracks and other imperfections in the rock surface, and an emphasizing of the sculptural form of the rocks.

On the 1976 visit, I realized the inadequacy of the pathways concept, for the area of ritual use was more of different activity areas or floors, with individual places of more specific interest, such as manholes in the floor. In fact, the whole floor of the upper chamber was found to be of engraved round and smooth boulders, even if some is covered with massive rockfalls.

We also found, after using the suggestions of Alexander Marshack – whose work will be discussed shortly – that the engravings tend to be made in sets of two, three or four parallel lines, each made with one tool, and each set overlying or underlying other sets in their entirety. The problem this paper sets out chiefly to face is what further can we say about the intention and meaning of the meanders to the meander-makers and their peoples. I wish to set out an approach which is structural: is there a consistency to the actual process of making meanders which might relate to a consistent story being told and represented by the meanders?

This is a working paper describing a plan of attack so that further work on the Koonalda meanders can be approached constructively. I have adopted a practice which may at first seem strange, that of enclosing statements within lines: these are assumptions being made in the attempt to work out a methodology and are open to investigation, or are matters to be looked at, avenues to be explored.

As a final point in this introduction, I assume that the meanders are made by human beings and not by animals, for instance by bats and owls in the Cave. While I have faced this objection before, and Gallus too for the wall markings at the rear of the Cave, it nevertheless keeps cropping up as a serious objection to the manner of investigation represented by our work, and warrants a more Text Box: Investigate the possibility of the Koonalda-type of engravings being made naturally and not by human beings.
 
careful investigation.

 

 

Marshack's Techniques

One of the most significant contributions of Marshack to the study of prehistoric meanders is a technique of analysis. Suppose we are looking at a number of lines engraved by some tool, perhaps a flake of flint. Marshack suggests a close examination

1.      of the engraved lines themselves, looking at their cross-sections – depth, width and shape – and

2.      of the points at which lines meet or overlie.

Different cross-sections of lines would imply a different tool was used perhaps by different people, and perhaps at different times. An examination of line junctions tells which lines overlie others, and hence the temporal sequence of their compilation. From both of these techniques one can give the story of the lines' construction, their structure: the order in which they were engraved with (presumably) differently intentioned lines corresponding to those made with different tools.

This analytic technique is an ideal; in practice, it is not so easy. For one thing, with the Koonalda engravings for instance, some sets contain no overlapping or meeting lines, and hence the temporal dimension of their construction is lost. Moreover, the rock surface has a deteriorated and "granular structure," which means analysis can be carried out only to a certain size dimension at which point the rock surface itself swallows up the lines, making them indistinguishable from "background" marks. This is especially so for worn lines. The rock also has surface cracks, sometimes short and of the same dimension as engraved lines. Not always can cross-sections and overlays be ascertained with certainty, involving the structural analysis of engravings with a degree of subjectivity and error. This will become more apparent as examples are analyzed.

As a further technique for analysis, the brightness of radiance under ultra-violet light can be used as an indicator of the relative ages of engravings. The brighter the light, the newer or more recent the engraving, for the oxidizing or crystallizing process decreases luminosity under ultra-violet light.

Marshack's investigations so far appear to be mainly of engravings on portable artifacts held in museum collections, and originating from a wide scattering of places over Europe. While from this he can describe the scope of this style of human expression and to some extent compare the various sub-styles within this tradition, he does not appear to examine a large number of examples from one site with an eye to seeing structural repetitions with its implications of similar intentions on the part of the engravers. Perhaps it is that in Koonalda Cave – and maybe in other sites in Australia too – we have a unique opportunity to compare myriad upon myriad of examples from the same site, looking for close similarities in structure. But this will be explained in a later section of this paper.

Marshack-Analysis of Koonalda Engravings

The accompanying figures are the results of carrying out a Marshack-type of analysis on some examples of Koonalda engravings.

One general conclusion can be drawn from this Marshack-analysis, and that is that engravings are made in sets of around three or four more-or-less parallel lines. Cross-sections confirm the visual supposition that each set of parallel lines, called a stream of lines, is made by one tool, and has a unity of itself. Different tools usually make overlying streams. This is an important and significant conclusion, for it implies at least some order to the chaos of the engravings. One should note that this tradition of engraving in streams of parallel lines – the meander tradition – is worldwide.

A further term is helpful too. We call a group of streams a cluster of streams if they exhibit a unity because of their physical inter-related overlying or separation from other groups of streams. Sometimes it is easy to separate out clusters and sometimes it is not.

It appears that in some cases the clusters of streams can be broken into three cycles of streams, a cycle being either the initial laying down of streams which then become related (this gives a little distortion in the idea of a cycle), or a self-contained temporally adjacent set of streams which are also physically related; in other words, the next (temporally) stream of the cluster beyond a particular cycle bears no physical connection to the members of the cycle. For instance, the clusters we have so far analyzed can be broken into the following cycles (some with a stretch of the imagination):

Cluster Number 1

Cycle 1: Streams 1,2.
Cycle
2: Streams 3,4,5.
Cycle
3: Streams 6,7,8,9,10,11.
Cycle
4: Streams 12,13,14,15,16.

Cycle 1 is the initial unrelated streams.
Cycle
2 builds on stream 1.
Cycle
3 builds from stream 2, bringing in cycle 2.
Cycle
4 builds on stream 2.

Cluster Number 3

Cycle 1: Stream 1, hole H.
Cycle
2: Streams 2,3,4,5,6.
Cycle
3: Streams 7,8.

Cycle 1 is the initial unrelated stream and hole.
Cycle
2 builds on stream 1.
Cycle
3 relates the hole H to cycle 2.

Cluster Number 4

Cycle 1: Streams 1,2,3.
Cycle
2: Streams 4,5,6,7.
Cycle
3: Streams 8,9,10,11,12,13,14.

Cycle 1 is the initial unrelated streams.
Cycle
2 builds from stream 1 through stream 2.
Cycle
3 builds from stream 3 relating it to cycle 2.

Cluster Number 5

Cycle 1: Holes Hl, H2, H3, H4, stream 1.
Cycle
2: Streams 2,3,4,5,6.
Cycle
3: Streams 7,8,9,10.

Cycle 1 is the initial unrelated streams and holes.
Cycle
2 builds from stream 1, bringing in hole Hl.
Cycle
3 builds from hole H4 to cycle 2, bringing in holes H3 and H2.

Cluster Number 6

Cycle 1: Hole H, shell S.
Cycle
2: Streams 1,2.
Cycle
3: Streams 3,4,5.

Cycle 1 is the initial unrelated shell and hole.
Cycle
2 builds from the hole H to the shell S.
Cycle
3 builds from stream 1, bringing in shell S.

Cluster Number 7

Cycle l: Stream 1, holes Hl, H2, H3, H4, crack C.
Cycle
2: Streams 2,3,4,5,6,7.
Cycle
3: Streams 8,9.

Cycle 1 is the initial unrelated stream, holes and crack.
Cycle
2 builds from hole H4, bringing in holes Hl, crack C, and stream 1.
Cycle
3 builds from hole H3 to crack C, bringing in holes H1 and H2, and cycle 2.

Cluster Number 8

Cycle 1: Shell S, streams 1,2.
Cycle
2: Streams 3,4.
Cycle
3: Streams 5,6,7,8.

Cycle 1 is the initial unrelated streams and shell.
Cycle
2 builds from stream 1.
Cycle
3 builds from shell S, bringing in streams 1 and 2, and cycle 2.

The following points should be noted:

1.      The order the streams are given in is the order – as far as is known – in which the streams were engraved. In the figures a more recent stream is drawn cutting across an older stream, as it would in actuality, breaking the earlier's continuity.

2.      The order in which holes, cracks, or shells are numbered is arbitrary.

3.      Each cycle follows the stream sequence through one particular intention, and usually it follows through the continuous relating of stream to stream via intersections or touchings. Thus, a new cycle is started when another intention is indicated with the next stream unrelated to those in the previous cycle.

4.      This is except for the first cycle in each cluster that comprises the natural holes, shells or cracks in the rock surface which will later be included, plus those initial unrelated streams that become related in the cluster. No temporal order is intended within this cycle, and in fact, it would be possible that some of its streams were engraved after those of later cycle (for instance, in cluster number 1, stream 2 could have been engraved after stream 5 and before stream 6).

5.      As emphasized elsewhere in this paper, this stream and cycle analysis is based on inadequate data and should not be taken as the unquestionable structural analysis of the clusters. It is sufficient, however, to pose a methodology for further investigation of the engravings. The hypothesis under investigation in 1976 when the information for the above was collected concerned whether or not the idea of stream in this context was a reasonable one to make.

Text Box: The structural information for the analyses of all the above clusters needs to be checked out in the Cave itself.

 

6.      No cycle analysis is given for cluster number 2; any relative order of streams here is guesswork. This is included as an extreme example; we have many other drawings of clusters that show an excessive lack of intersections. That is not to say that the streams in such clusters are unrelated; the "arrow" image of cluster number 2 implies a visual relationship between the streams.

7.      From the cycle analysis given there appears to be no repetition between the cycles of the various clusters, except that often they can be broken into three cycles.

Text Box: That clusters are of three cycles would therefore seem a reasonable topic for further empirical investigation.

Any further comparison – for instance, between the way streams are related within each cycle – should probably be left until further data are gathered and we are more sure of the structural breakdown of each cluster.

8.      Finally, the following are some of the ways streams (etc.) are related within each cycle: by cutting across, touching, paralleling, having a common direction focus, and by including.

That is about as far as Marshack's analytical techniques will carry us by themselves. Where do we go from here? There are many avenues open. For instance, one could subject them to an analysis in which the engravings are treated as art (which it is highly probable that they are, while at the same time probably something else as well), or a symbolic and psychological analysis in which the engravings are seen as the expression of archetypal, or developmental stages, or psychological conditions. Another approach is to consider the engravings as a form of writing, of cultural, stylized interpersonal communication. Of course, any of these can only be hypotheses for investigation, avenues to be explored and perhaps discarded. It is with this last approach that this paper is chiefly concerned.

Text Box: The Koonalda meanders are, amongst other things, both artistic and aesthetic expressions of the engravers and a form of writing.

A Further Structural Analysis

Meanders as a Form of Writing

Let us now look at the Koonalda engravings as a form of writing, as having a deep structural consistency, as being the representations of a set of myth stories in an interpersonally recognizable manner. One should question this approach if the consistency that one would expect with an interpersonal form of writing is not present. The existence of one element of consistency, the universality of streams, is encouragement for it. The next few sections are a further examination of this hypothesis.

The object of the exercise is to find a way of looking at the lines and their relative structures that allows them to present themselves in a unified and consistent manner. At present, we have one level of consistency, namely the streams, but beyond that, there is no so-far apparent structural consistency in the relations between the streams – except perhaps a three-cycle breakdown of clusters.

It should further be noted that this analysis would not give us automatically the meaning of the meanders for the people who made them. I am merely trying to see if this can be taken as a form of writing and, if so, how the structure of the engravings relates to what we see as the structure of writing. The translation is a further step, and one to which I shall return also.

Basic to our written language are words, grammar being the way words are related and thereby used. If we were to suppose that the Koonalda meanders are a form of writing, it would seem reasonable to look for words represented in them.

Amongst the variables we need to consider in deciding what is what, are:

1.      the physical nature of the stream (the number of lines in it, the spacing between the lines, the lines' cross-section, whether they have any branching, whether and how they curve, etc.);

2.      the relation between two streams (do they cross, or touch, with one or more lines of the stream; do they only meet; are they parallel, etc.?);

3.      the relation between streams and natural aspects of the rock (shells, holes, cracks, etc.); (iv) the visual and physical nature of the cluster (does it form a visual pattern? how many streams in it? Etc.).

The most obvious choice for word equivalents in the engravings is the streams, since they are the consistent ingredients of the engravings beyond the lines themselves (perhaps the "letters"?). Streams are physically related, spatially and temporally in clusters; a cluster could therefore be taken as a whole statement or perhaps even as a collection of statements forming an entity of itself such as a paragraph or story.

Text Box: Streams are words and clusters statements or stories.

 

It is possible to take a line (or a cluster) to be a single word, as another appoach, and interpret certain streams as, say, people or objects, and others as their relationships.

Suppose we accept a "stream = word", "cluster = statement" approach, then it would seem that the various characteristics of streams as in (1) above tell something of the qualities or nature of the referent of the word. The references involved we cannot at present say. It would also seem reasonable that the relations between the streams not only say smoething of the meaning of one word relative to another (for example, stream A crossing stream B at right angles with all its lines means character A is the son of character B), but in so doing depict the grammar and word use of the language so represented.

Text Box: The relationships between the streams within a cluster and the nature of the streams themselves respectively tell of the relationships between the people, places, etc., of the myth story being represented, and of the character of the people, places, etc., represented.

Since the cluster examples above are only structurally analyzed approximately – the data not being fully reliable at this stage – this relationship type of analysis should wait.

Consistency and Repeatedness

If all the clusters in Koonalda Cave – and there would be at least tens of thousands of them – were different with this sort of analysis, we could conclude that something is wrong with it or its underlying hypothesis that the engravings are a form of writing. We should expect at least some of the characters and their relations to be the same. But this raises a number of points and needs some justification.

One of our initial working assumptions after the 1973 visit to Koonalda Cave was that each Koonalda engraver had made a mark and later came back and added to the existing marks so as to participate, to express his or her place in the ritual tradition. Under this idea, streams within a cluster could have been made over many thousands of years and we would probably not expect to find a consistent structure within and between the clusters. We may then have to stretch the point a little to explain the sometimes (for instance, see cluster number 3) very careful relating of streams within a cluster: lines touching or paralleling each other, etc. This does, however, need further exploration in the field, perhaps using ultra-violet lights as mentioned earlier, to see if there is a difference in ages between the majority of streams in a cluster.

Text Box: Are the streams within a cluster of the same age?

As an addition to this point, one might suggest that only streams of the same age be included within a cluster; that is, that clusters may overlie clusters. The analysis suggested must be on only an individual cluster.

We also need to recognize that there are different styles and apparently different meander structures even within Koonalda Cave itself. We can compare, for instance, the fine engravings on many of the smooth and rounded floor boulders with the coarse ones on the boulders at the entrance to the "cavern," or with the wall finger-markings, or with the large-scale wall engravings above the "squeeze." If there are different traditions of meander marking within Koonalda Cave then we should not expect all the engravings to conform to one structural pattern, but perhaps just a few.

And thirdly, a collection of myth stories might be told accompanying the rituals being performed in the upper chamber that is, it might not always within one engraving tradition be the same myth being represented, but one from a presumably small repertoire of myths. Some consistency should therefore be expected among subgroups of clusters within any one engraving tradition.

Why should a story or character be represented the same way with different engravings? Obviously, this need not necessarily be so and different engravers may well have been at liberty to express the character of that written about in whatever way they so felt. This would mean that we might find no further consistency beyond the basic structures so far described from a Marshack-analysis, and have no hope of ever being able via this method to translate this writing.

It is important that I leave myself open to the possibility that perhaps the meanders are not writing and not culturally consistent, the meaning expressed by the engraver perhaps not comprehendible by someone else with the same cultural tradition. Perhaps a mode of expression is employed in which individuals have expressed themselves, and others within the same cultural tradition cannot understand that expression. The proof of this pudding is in its eating: there is no inter-personal communication at this level, given the correctness of the analysis so far, if no structural consistency is found.

Another fact potentially negative to my hypothesis is the high density of lines in some places. Could it be that lines with a deliberate structure were made over top of others?

However, this pessimism is perhaps counted against. One reason for optimism is that a consistent structure has so far been found and it would seem likely that consistency would continue: why should each engraver specifically stick to stream, cycle and cluster method and then divert into individual styles of expression, and not be individualistic right from the start?

Further reasons for supposing an inter-personal consistency with regard to representing characters and their relationships lie in the answers to another question concerning repeatedness. Should we expect the same myth to be repeated, the same statements to be made or the same characters referred to? If not, then we should not expect there to be any further repetitions in the actualized variables of the meanders. To answer this we need to consider the nature of the northwest passage of Koonalda Cave where the meanders are found and build a case for the ritual use of the Cave, and that the meanders in fact are the written representation of the myths told in the rituals.

Why do we think that the upper chamber was used for ritual purposes and not, for instance, for a night-shelter, or as a place for a pleasant Sunday afternoon picnic? There are many reasons. The finds themselves do not imply a utilitarian use: no implements specialized that way, only skulls and vertebrae of animals, and so on. Secondly, the upper chamber takes us nearly an hour to reach from the surface of the Plain using steel ladders, paths, lanterns, and other modern paraphernalia; it is a very difficult place to and one of the more inaccessible parts of Koonalda Cave. Thirdly, the Cave probably would have been around 180 kilometers away from the nearest habitable land; if the Koonalda people 19,000 years ago were anything like the Mirning, who in modern times inhabited the Nullarbor, the only part of the region on which they could live is the coastal belt of up to 20 kilometers; further inland virtually no one went and it was much feared. That long ago the sea-level was around 90 meters lower than today, and instead of Koonalda being just on the edge of the coastal belt it would have been 180 kilometers or so beyond it.

All these indicate that the going to Koonalda and to its upper chamber were extra-special for extra-ordinary purposes, probably to perform rituals. These may, or course, have been in association with some other activity such as the collecting of water in particularly dry periods (although in those circumstances the water in the Cave was probably too saline for consumption), or for the mining of flint. If it be the latter, then the Koonalda flint may be of a special (non-physical) worth since probably the same seams of flint were available from the cliffs or scarp which now are the sea cliffs south of Koonalda, and then would have been rather inland (although they may have been covered with sand dunes). Further to that, the Koonalda people probably used one place for one particular ritual, that place and its contents being closely associated with the ritual. This appears to be so for modern Aborigines. Thus, we should expect the meanders in Koonalda to be representations of one particular set of myths, those associated with one particular ritual, if they are representations of myth stories. We should, that is, expect the meanders to exhibit structures from out of a (small) range, those representing the (small) number of myth stories told and retold in the upper chamber.

As eons went by, it may be that the upper chamber was used for different ritual purposes, and that the tradition of making meanders changed. Thus, it is not surprising that there are also a number of different styles of engraving and different structures within the meanders.

Text Box: Look at, from ethnographic examples, the change, or lack of, in rituals and their myths over time, the correspondence between a ritual and the place(s) where it is performed, and the relation between myths and rituals.

Translating the Meanders

The most obvious question next to arise is whether the meanders, presuming they are a form of writing, can be translated. To so bridge a 19,000-year gap would be some feat. One may realistically fear this to be impossible because, in examining the stories such as of Cyrus Gordon, the great acts of translation depend on some connecting link between the people whose language is written down and us. The name of a king, for instance, often forms that link on the basis of which deciphering can begin. One may, of course, live in hope that such a link be found.

There is a more tenuous approach to translating which I will now outline. Suppose, first, that the engravings in Koonalda Cave have been structurally analyzed into a number of different story representations, each with its own peculiar internal structure and characterization, as defined by the nature of the streams and their relationships. The myth story that each depicts will have the same structure as its representation, the characters the same relationships and, maybe, qualities. One should therefore look for a myth story with that structure, for here may be the story so represented in Koonalda.

The problem now would seem to be compiling a collection of all the myths of the Koonalda people so that structural analyses can be made of each of them. It is at this point that more tenuous connections are made, for we would probably need to compile all the myths of the Mirning and perhaps of their near neighbors. Difficulties arise because this tribe is now extinct and almost the only records we have of their mythology are in the popular writings and notes of Daisy Bates; but at least there is some Text Box: A collection of all known myths of the Mirning and perhaps also of their neighbors needs to be made, the chief source for which will be the writings and notes of Daisy Bates. A structural analysis of each myth can then be made and compared with the structural analyses of the Koonalda engravings: clusters with the same structure as a myth may well be the depiction of the myth.
 
record.

 

 

 

Two preliminary investigations first need to be undertaken. The first is to make a case for the continuity of myths over vast periods within a people. I understand that this is so, especially as far as the structure of myths are concerned, representing as they do the statement and solution of the dichotomies a people find themselves in (see, in this regard, the work of Levi-Strauss and other structuralists).

The second investigation is whether the Mirning were the people in whose territory the Nullarbor lay 19,000 years ago, or whether they are more recent inhabitants, replacing the Koonalda people. For, if they are different people, it is perhaps expecting too much for their myths to be roughly the same. The archaeological investigations of Ljubomir Marun, whose doctoral dissertation was precisely centered on this question, provides an answer: there is cultural continuity over this time span and one can take the Mirning to be the direct descendants of the Koonalda people.

Other Matters

There are a host of other questions to be asked, both theoretical and practical. Let me outline two of the theoretical ones:

1.      One can rightfully ask why this tradition of meander engravings ceased and probably the representational art we know of the Aborigines developed. One can say that the latter too is a form of writing in the sense of it being in culturally standardized symbols, even if they often were representational symbols, and are usually the depiction of mythic stories within the context of rituals. Far from being a step backwards from a non-representational form of expression, contemporary Aboriginal art is a step forwards in that the mythic symbolism expressed in the representational form is more evocative and universally relatable.

It may well be that both traditions – the representational and the non-representational – co-existed; they appear to have in Europe, and given sand drawings and engraved portable artifacts from the ethnographic Australian scene this may not be too unreasonable. The meander tradition of Koonalda may not have ceased 19,000 years ago (the Cave may have been abandoned then because of a massive rockfall within the upper chamber), but continued in other caves, on portable or non-preserved materials.

The meander tradition could also be seen within a parallel to the development of children's drawings in the life cycle of an individual's maturation. If this be a valid parallel, the history of the human race, biological as well as mental, is written again in the development of each human being from conception to childhood. The work of Piaget suggests that children at an early stage draw in parallel lines, meanders, and then later can express themselves more representationally. Is this what happened also in the evolution of humankind? The question as to why the change from meanders to representationalism, then – if there is one – can be answered in terms of it being a developmental or evolutionary change in the mental history of the human species

Text Box: Investigate the parallels between children's art and the meander tradition and the extent of ethnographic meander/representational art in Australia.

 

2.      Why was this meander tradition worldwide? Why is it found throughout Europe and Australia, and probably elsewhere as well? One could answer this in terms of culture diffusion, the learning of good ideas from neighboring peoples in a chain around the globe, or, again, in terms of it being a developmental stage all humanity went through on the road to maturation.

Summary

As an attempt at a methodology for the study of meander engravings such as are found in Koonalda Cave, South Australia, I suggest the following:

1.      An examination of the lines themselves according to the techniques of Alexander Marshack, noting cross-sections and which lines overlie which at points of contact and intersection.

2.      A re-telling the temporal story of each cluster of streams of lines, breaking them into cycles.

3.      Extracting the repetitions or structural patterns in the stream-cycle – cluster analyses, noting also the character of each defined within the relevant variables. Each separable type of cluster is the written form of one particular myth story.

4.      From a structural analysis of the myths of the people whose meanders are so analyzed, or of their nearest kin whose myths are recorded, correlate each separable type of cluster with a myth having the same structure. This may be a way of translating the engravings.

Acknowledgments

Many people were involved in the 1976 expedition to Koonalda and the preparation of this report:

1.      For financial support, encouragement and advice, the National Geographic Society (especially Mary Griswold Smith and Victor Boswell), for supplying gear, sponsorship and other support, the South Australian Museum (especially Graeme Pretty).

2.      For permission to enter Koonalda Cave, the South Australian Aboriginal and Historical Relics Advisory Board.

3.      For accompanying us on this visit: Sandor Gallus (nominated by the South Australian Museum as the field investigator), Ian     Lewis, Kevin Mott, and Neil Chadwick.

4.      For hospitality while at Koonalda, Mr. and Mrs. Cyril Gurney.

5.      For stimulus and support, Alexander Marshack and Hallam Movius Jr. of Harvard University.

6.      For information and comments in the preparation of the report, Mike Smith, Betty Ross, C. Merewether and A. Cooper (and Sandor Gallus for permission to use Neil Chadwick’s and their notes on the upper chamber, the latter compiled ? January 1975), Graeme Pretty, J. N. Jennings of the Australian National University, and C. D. Ollier of the University of New England.

7.      For assistance in the preparation of the manuscript itself, the late Alf Armstrong, Sandra Myer my typist, and the Universities of Otago and Auckland.

8.      For financial support in the creation of this paper, the University of Auckland Research Fund.

9.      Most of all, I want to express my great debt to Sandor Gallus.

Copyright (c) Kevin J. Sharpe, 1982.