This is a comprehensive geological report about the origins and current structure "anomalies" of Murphys Haystacks compiled by Dr. Rowl Twidale [see below for full details]. For a much simplified "laypersons" view of the Haystacks see The Geology: Simply Speaking

This "web" version may be a little difficult to read without it's included pictorial content and we understand that. Linking in the multitude of thumbnails for easier "in context" viewing is difficult for speed reasons. However, all of the hyperlinks used for images and diagrams in the reading of the report open into separate image windows to facilitate better viewing.

Murphys Haystacks consists of two separate though clearly related groups of large granite pillars and boulders standing near the crest of a broad domical hill, here called Oakfront Hill, just to the west of the Streaky Bay - Port Kenny Road, some 30km northwest of Port Kenny [Figure 1: a & b]

The major landforms present at Murphys Haystacks are pillars and boulders. Pillars are tall columns of rock that are in physical continuity with the granite of the hill [Fig 3a]. Boulders are more-or-less spherical masses that are detached from the underlying bedrock [Fig 3b]. Both pillars and boulders are thought to result from the break down of thick slabs or sheets of rock into which the granite is sub-divided and which have given Oakfront and other hills in the district their characteristic domical or convex-upward shape.

Such sheet structure is exposed in the flanks of nearby Freeman Hill, and even more clearly on Ucontitchie Hill and Mt Wudinna where the granite is not masked by a cover of limestone. The hills are also sub-divided by sets of fractures arranged in a fan pattern when seen in section. They intersect the thick slabs, resulting in the formation of essentially cubic or quadrangular blocks [Fig 3c].

The pillars and boulders at Murphys Haystacks typically have side-walls that are smoothly concave: they are said to be flared. Such flared slopes result from weathering, by soil moisture, beneath the land surface, and especially along fractures [Fig 4]. Water penetrates along a fracture altering the rock with which it comes into contact. The mica and feldspar in particular are readily altered on contact with moisture. Fresh granite is difficult for water to infiltrate, but the more the rock is altered the more permeable it becomes, and the more readily can more water enter the mass. Thus a zone of weathered rock is formed along a fracture. There is an abrupt transition between the altered rock and the still fresh rock. This boundary or junction is called the weathering front. It represents the limit of significant weathering and at several sites on Eyre Peninsula it can be seen to be concave-upward or outward where the slopes are steep [Plate 5].

Because the near-surface soil tends to dry out at times, weathering is slower there than deeper down, where soil moisture is ever present. For this reason the weathered zone in fractures is wider at depth than near the surface, or put another way the weathering front becomes curved, and penetrates deeper into the host block a few metres below the surface than just below it or at great depths beyond the limit of soil moisture. Thus the weathering front is curved, and concave with respect to the host block.

The main part of the block remains fresh and cohesive and so massive that it cannot be moved by rain and wash; but the weathered granite [known as grus, from the German word for fine gravel] is washed away, so exposing the weathering front as a smooth concavity or flare, one on each side of the fracture. This process of subsurface weathering has been followed by the erosion or removal of the weathered material. Thus flared slopes are due to two processes, weathering and erosion. Where such weathering and erosion has taken place along all six of the fractures that delimit each of the blocks, each resultant pillar or boulder comes to be flared on all sides. Hence the haystack forms that are so well and so abundantly developed near the crest of Oakfront Hill. Penetration of water beyond the weathering front may also be the cause of the flakes and shells [Plate 3] of rock that can be seen on some boulders and pillars.

In some instances the flared slope has advanced toward the centre of the block from all sides to such an extent that the platforms are wider than the remaining core [which has also been lowered by weathering], so that the residual looks like a boss standing above a shield [Plate 7]. Carried to its logical extension such soil weathering will presumably cause the elimination of the boss leaving simply a flat or only broadly domed surface a platform or a low whaleback [Plate 8]. At the base of some boulders at Murphy Haystacks a more or less narrow platform is found. This is due to a recent lowering of the soil surface [Plate 6].

Some of the pillars and boulders are not only flared but have developed in their sides huge hollows known as tafoni [singular tafone], an Italian word meaning windows or perforations, and apt in the sense first that these hollows interrupt or breach the smooth cut lines of the forms [Plate 9], and second that in extreme cases [as seen for instance on Granite Island near Victor Harbour, and on Mt Wudinna] a tafone can develop to such an extent that the hollow breaches the outer shell so that a window or opening is formed - a site much favoured by photographers who record companions with just their heads sticking out from the rock!

There is considerable argument about how tafoni form; they were first described from Corsica about a century ago and the argument has gone on unabated ever since. But as they commonly form alongside flared slopes [and not only Murphys Haystacks but at Ayers Rock and other places also] the two may have a common subsurface origin: tafoni may develop where the moisture attack is especially effective. Second, it is obvious that the outer skin of the granite, that which forms the projecting lips or visors, is somehow more resistant than the rest of the rock. It may be that a slight accumulation of iron and silica, formed either at the weathering from beneath the soil surface, or concentrated by the hyphae [roots] of lichens for instance [see projecting ribs in the tafone on block [Fig 2] are covered by lichens [Pl 4]. The most difficult problem connected with the tafoni however concerns how and why the inside walls of the tafoni have been and still are being worn back. The rock there is loose and fragmented, and can be loosened at the merest touch at some sites [as e.g. at Ucontitchie Hill where the rock is breaking down in layer upon layer of thin flakes]. Some workers have suggested that the crystallization of halite [common salt] and gypsum, The effectiveness of salt crystallisation in rupturing rocks - even those as cohesive as granite - is now beyond dispute for it can be replicated in the laboratory. The real problem is how saline solutions come to be within the rock.

At least one other feature seen at Murphys Haystacks is due to moisture attack at the weathering front. It will be noticed that the lower few centimeters [about a foot] of the sidewalls of many of the pillars have a rough surface which stands in marked contrast with the smooth higher surfaces [Plate 10]. Close examination shows that this roughness is due to quartz crystals [glassy or grey looking] standing out in relief, and this is because the feldspar and mica that was originally between the quartz have been rotted by water, and when the weathered rock was stripped away the tiny spaces previously occupied by the feldspar and mica are left empty. This feature is called pitting and it provides a measure of how much the plain surface has been lowered recently [in this case, probably since the clearance of trees from the hill by Europeans and the erosion of the topsoil, the same erosion that caused the solid calcrete later to be exposed over wide areas on the hill]. Similar soil erosion is implied by the steep basal slopes of some pillars [Plate 6].

Other features worthy of note are the vertical grooves [called Rille or Karren] which are due to water trickling down steep sidewalls, and even overhanging faces, and gradually weathering and eroding shallow grooves [at "G" Figure 2b and "Y", Figure 2a - Plate 11]. There are good examples in both the eastern and western groups at the Haystacks. There is a complication which is well illustrated at Yarwondutta Rock, near Minnipa. There on one of the northern flared slopes there are grooves that now stand out as ribs and it is suggested that the trickles initially erode a groove, but that this becomes a preferred site for lichen growth. The lichen protects the surface [as at x in the western group of Murphys see Figure 2b] and it is the adjacent areas that are next weathered and eroded leaving the original grooves higher and looking like ribs [Plate 12]. The same sort of thing can be seen on the northern slope of the Turtle Rock, near Mt Wudinna.

Some linear depressions, grooves and gutters are due to weathering along fractures or other weaknesses in the rock, and there are many examples. One on the crest of x [Figure 2b] in the western group is largely developed along a vein of fine granite rock [Plate 13]. Also on this crest is a shallow depression known as a rock basin or gnamma, again the result of water rotting the rock [Plate 14].

When did the Haystacks form? We have to distinguish between the granite on which the various landforms are developed and the period or periods of landform development.

The granite is the Hiltaba Granite and is part of a large mass that underlies much of the northwestern Eyre Peninsula, the southwestern Gawler Ranges and adjacent areas: hence its name, from Hiltaba Station, in the southwestern Gawler Ranges. The rock is pink and coarsely equigranular with crystals of grey glassy quartz and pink feldspar prominent. The granite was emplaced deep in the Earth’s crust, probably some 7-10 km below the then land surface. This age has been determined using radioactive decay methods (in particular lead-uranium series dating). That the granites are now exposed implies an enormous amount of erosion, for some 7-10 km of rock has been stripped away. The overlying rocks have been worn away, transported and deposited as sedimentary rocks on the continental shelf and inland basins.

The landforms we see at Murphys Haystacks are much younger than the rocks of which they are formed. The various pillars and boulders that constitute the Haystacks were already in existence in essentially their present form some 34,000 years ago. They are almost certainly much older but how much older is not known. At various times during the last 700,000 years (during the Middle and Late Pleistocene) huge dunes of calcareous sand spread far inland from the coast. Some of the granite hills of this coastal zone where buried by the dunes. Freeman Hill is an example for calcareous rocks are preserved on its crest. At Murphy Haystacks however, the evidence suggests that though the calcareous sand spread on the lower slopes the two groups of boulders and pillars granite hills stood above the dunes, like rocky islands in a sea of sand. The reason for suggesting this is that the upper zones of the calcareous dunes became cemented by lime, forming a solid limestone called calcrete as a B-horizon in the soil. Calcrete now forms a crust over the slopes of Oakfront Hill, and is also preserved in cracks, crevices and hollows at the base of the rocks [as at H in Figure 2a, and Plate 2] but no higher.

The calcrete at Murphys Haystacks contains is fossiliferous with fragments of marine organisms (forminifera) blown in with the dune sand. However such fossils would provide a date for the sea floor sediments from which the dunes were derived, whereas the calcrete provides a minimum (albeit a rather unreliable date - an approximation) age for the landforms over and against which the limestone has been plastered. The calcrete was dated using radioactive decay, by the Carbon 14 method. The calcrete at Murphys Haystacks is at least 34,000 years old (and some from the crest of Freeman Hill is about 26,000 years old). As the dune sand on which the calcrete was formed was blown into cracks and hollows in the boulders, the latter clearly predate the calcrete so that we can say Murphys Haystacks are at least 34,000 years old and, given the possible complications of 14C dating, probably much older.

Review of development

1: Emplacement of mass of granite at depth of 7-10 km about 1590 millions of years ago.

2: Deep erosion and removal of rocks that formerly covered the granite, which had been stressed resulting in the development of sheet structure as the mass came to be close to the land surface.

3: More than 34,000 years ago, the upper shell was broken down into cubic or quadrangular blocks, and flared slopes were developed on the sides of the blocks. Because the crestal area was in tension, the blocks there were weathered quickly. Likewise those on lower slopes were rotted rapidly and also eroded, so that only the blocks on upper slopes persisted.

4: During glacial periods of low Pleistocene sea level (the last 2 million years) huge coastal dunes, consisting mainly of shell fragments made of calcium carbonate developed and spread inland, covering many nearby hills and certainly lapping around the bases of the Haystacks.

5: In recent times a soil developed on the lime sand and since European settlement the topsoil has been eroded [to a depth of about 25cm] exposing the solid limestone or calcrete that we see on the hill crest and slopes, and above which the Haystacks stand.

The report in this geological report was produced by Dr Rowl Twidale [D.Sc. (Bristol), Ph.D. (McGill) and D.Hon.,Causa (Madrid)] and Dr Elizabeth M. Campbell [Ph.D. Adelaide].

The original report is being used with permission and is strictly copyright material. Any replication using any medium is strictly prohibited without the express consent of Dr. Twidale. Please respect copyright law at best and common professional courtesy at the very least.

He can be reached via

Department of Geology & Geophysics
University of Adelaide
ADELAIDE   5005
South Australia

phone  :  Int + (618) STD (08) 8303 5392
fax        :  Int + (618) STD (08) 8303 4347
e-mail  : rowl.twidale@adelaide.edu.au

See Eyre would like to offer a personal thank you to Dr. Twidale for his assistance while creating this report in a "web format".  He was most generous in his efforts to ensure accuracy and for this we are most appreciative. The time that was provided by Dr. Twidale must have been significant. Thank you.

Continue...