A structural metamorphic study of the Broken Hill Block...

A structural metamorphic study of the Broken Hill Block, NSW,Australia

C. J . FORBES,1 , 2 , * P . G. BETTS,2 R. WEINBERG1,2 AND I. S. BUICK2

1Predictive Mineral Discovery, Cooperative Research Centre ([email protected])2School of Geosciences, Australian Crustal Research Centre, Monash University, Clayton, Australia

ABSTRACT The prograde pressure–temperature (P–T) path for the complexly polydeformed Proterozoic BrokenHill Block (Australia) has been reconstructed through detailed structural analysis in conjunction withcalculation of compositionally specific P–T pseudosections of pelitic rock units within a high-temperature shear zone that formed early in the tectonic evolution of the terrane. Whilst the overall P–Tpath for the Broken Hill Block has been interpreted to be anticlockwise, the prograde portion of thispath has been unresolved. Our results have constrained part of this prograde path, showing an earlyheating event (M1) at P–T conditions of at least c. 600 �C and c. 2.8–4.2 kbar, associated with anelevated geothermal gradient (c. 41–61 �C km)1). This event is interpreted to be the result of rifting atc. 1.69–1.67 Ga, or at c. 1.64–1.61 in the Broken Hill Block. Early rifting was followed by an episode oflithospheric thermal relaxation and burial, during which time sag-phase sediments of the upper BrokenHill stratigraphy (Paragon Group) were deposited. Following sedimentation, a second tectonothermalevent (M2/D2) occurred. This event is associated with peak low-pressure granulite facies metamorphism(c. 1.6 Ga) and attained conditions of at least 740 �C at c. 5 kbar. A regionally pervasive, high-temperature fabric (S2) developed during the M2/D2 event, and deformation was accommodated alonglithology-parallel, high-temperature shear zones. The larger-scale deformation regime (extensional orshortening) of this event remains unresolved. The M2/D2 event was terminated by intense crustalshortening during the Olarian Orogeny, during which time the first mappable folds within the BrokenHill Block developed.

Key words: Broken Hill Block; geothermal gradients; high-temperature shear zones; metamorphism;prograde P–T path.

INTRODUCTION

Pressure–temperature (P–T) paths are extremely usefulin determining the orogenic history of a region, and awell-constrained P–T path will allow delineation of therelative timing of burial, heating, and uplift andexhumation of a terrane. Through recognition ofclockwise or anticlockwise P–T paths, characteristicP–T evolutions of burial and heating (clockwise) orheating and burial (anticlockwise) followed by exhu-mation can be constructed. In many cases, absoluteages can be incorporated into the history so thatpressure–temperature–time (P–T–t) paths can be con-structed, placing even tighter constraint on the oro-genic history (e.g. Kamber et al., 1998; Muller, 2003).

By combining the P–T(–t) path with field observa-tions, the P–T evolution of a terrane can be combinedwith its deformation history. In this way, structuresthat may have accommodated bulk strain and/ordeformation, or that may have influenced the large-scale geometry of a terrane may be more easilyrecognized and characterized, leading to the

development of more complete geological histories andbetter constraints on the tectonothermal architectureof an orogen.

However, difficulties are often encountered withincomplexly polydeformed terranes where evidence ofthe early P–T and deformational history has beenobscured by episodic deformation and/or metamor-phic events. In these cases, only partial construction ofa P–T path may be possible, and recognition of earlydeformation elements becomes difficult and oftensurrounded by controversy (e.g. Boger & White, 2003).

The Palaeoproterozoic Broken Hill Block, westernNew South Wales, Australia, is such a terrane wherethe prolonged and intense deformational and meta-morphic evolution of the terrane has hinderedreconstruction of its early history. The P–T pathsconstructed for this terrane (e.g. Stuwe & Ehlers,1997), although vital to partially establishing the ter-rane history, have focussed on the post-peak meta-morphic history associated with the Olarian Orogeny(c. 1.6–1.59 Ga; Page et al., 2000a), and the progradepath remains unconstrained. This paper focuses onunderstanding the relationships between deformationand metamorphism within an early-formed high-tem-perature shear zone in the southern Broken Hill Block,

*Present address: Geological Survey of Western Australia, 100 Plain

St, East Perth WA 6004, Australia.

J. metamorphic Geol., 2005, 23, 745–770 doi:10.1111/j.1525-1314.2005.00608.x

� 2005 Blackwell Publishing Ltd 745

with the aim of reconstructing the prograde P–T path.Detailed structural analysis was conducted in con-junction with metamorphic petrology and calculationof P–T pseudosections across the shear zone. Recon-struction of the prograde P–T path allows resolutionof previously unrecognized geothermal anomalies,which assist in modelling the early tectonothermalarchitecture of this complex region.

THE BROKEN HILL BLOCK

The Broken Hill Block (Fig. 1) comprises palaeopro-terozoic metasediments intercalated with mafic andfelsic intrusive and extrusive igneous rocks that arecollectively termed the Willyama Supergroup (Fig. 2;

Stevens et al., 1988). The sediments were depositedwithin an intraplate rift basin (Stevens et al., 1988)between c. 1.71–1.67 Ga (Page et al., 2000a). Sag-phase sediments of the upper Willyama Supergroup(Paragon Group, Fig. 2) were deposited between 1.66and 1.6 Ga (Page et al., 2000a; Raetz et al., 2002). TheWillyama Supergroup underwent intense shorteningand metamorphism during the Olarian Orogeny c. 1.6–1.59 Ga (Page et al., 2000a). Early stages of orogenesisinvolved thin-skinned deformation and the develop-ment of large-scale recumbent (e.g. Marjoribankset al., 1980; Hobbs et al., 1984) to shallowly inclined,highly non-cylindrical (sheath-like) folds (Forbeset al., 2004) associated with high-temperature, low-pressure amphibolite to granulite facies metamorphism

Fig. 1. Geological map of the Broken Hill Block showing the locality of the Round Hill Area (modified from Willis et al., 1983).

746 C . J . FORBES ET AL .

� 2005 Blackwell Publishing Ltd

(Phillips & Wall, 1981; Stuwe & Ehlers, 1997). Thenappes were overprinted by north–south trending,upright folds that developed during thick-skinneddeformation (Marjoribanks et al., 1980; Hobbs et al.,1984; Wilson & Powell, 2001; Forbes et al., 2004;Forbes & Betts, 2004). Post-orogenic granitic intru-sions of the Mundi Mundi Suite (c. 1.59 Ga; Pageet al., 2000a) and various pegmatites were subse-quently emplaced.

Neoproterozoic (c. 840–780 Ma; Preiss, 2000a)Adelaidean rift and passive margin sequences weredeposited unconformably onto the deformed remnantsof the Willyama Supergroup (Powell et al., 1994).Adelaidean sediments and Willyama Supergroupbasement were then deformed and metamorphosed togreenschist facies during the Delamerian Orogeny(Priess, 2000b) at c. 520–490 Ma (Harrison &McDougall, 1981).

The tectonic setting of deformation during, andprior to, the Olarian Orogeny is highly controversialbecause of the obscuring of early-formed structuresduring the intense, prolonged tectonic history of theBroken Hill Block involving multiple orogenic cycles.Until recently, tectonic models for the Broken HillBlock have focussed on the Olarian Orogeny, inter-preting the earliest deformation phase as a large-scalenappe-forming event that resulted in regional over-turning of the stratigraphy (e.g. Marjoribanks et al.,1980; Hobbs et al., 1984). The tectonic setting of theBroken Hill Block has since been re-interpreted in the

context of a mid-crustal fold and thrust belt (Whiteet al., 1995).

The Olarian Orogeny is characterized by a regionalhigh-temperature, low-pressure metamorphic regime;however, the origin of this thermal regime is highlyconjectural. Metamorphism in the Broken Hill Blockwas initially attributed to a single event of changinggrade across the terrane, resulting in high-grade am-phibolite to granulite facies metamorphism in thesouthern Broken Hill Block, and greenschist faciesmetamorphism in the north (Binns, 1964; Phillips,1980; Phillips & Wall, 1981). This interpretation hassince been challenged by Stuwe & Ehlers (1997), whoidentified two distinct metamorphic events within theblock. The first event occurred at 4–5 kbar and 650 �C,and is associated with early shortening at c. 1600 Ma.This event is termed M2 by Stuwe & Ehlers (1997), as itis suspected to have occurred post-peak metamorph-ism (M1). Their second metamorphic event (M3) was agarnet + chloritoid + staurolite + chlorite mineralassemblage that formed at 5 kbar and 480 �C. Thisassemblage statically overprints a foliation that isassociated with upright folding (Stuwe & Ehlers, 1997).The timing of this static overprint is suggested to be asyoung as the Grenvillian (c. 1.1 Ga) or Pan-African(550–500 Ma) (Stuwe & Ehlers, 1997).

An opposing view was recently put forward bySwapp & Frost (2003), who suggested that the BrokenHill Block may have undergone higher-pressuremetamorphism up to conditions above 9 kbar at775 �C, based on the presence of rutile in equilibriumwith garnet, aluminosilicate, and ilmenite in granulitefacies pelites preserved in the Round Hill Area (Fig. 1).

Recently, Noble (2000); Venn (2001) and Gibson &Nutman (2004) have suggested that the earliest phasesof deformation in the Broken Hill Block were the resultof mid-crustal extension. This type of extensionalmodel can account for regional high-temperature, low-pressure metamorphism and the development of apervasive lithology-parallel fabric defined by a high-temperature, low-pressure mineral assemblage that isnot associated with any folding in the area (e.g. Noble,2000; Venn, 2001; Gibson & Nutman, 2004). However,the timing of lithospheric thinning is conjectural, andmay be contemporaneous with basin evolution anddeposition of the Willyama Supergroup c. 1.69 Ga(Gibson & Nutman, 2004), or may have occurredimmediately prior to the Olarian Orogeny at c. 1.61 Ga(Venn, 2001).

THE ROUND HILL AREA

The Round Hill Area, located in the southern Broken Hill Block(Figs 1 & 3) has undergone multiple phases of deformation asso-ciated with amphibolite to granulite facies metamorphism, and pre-serves a well-exposed early high-temperature shear zone. The RoundHill Area has been subdivided in this study into the southern andnorthern areas (Fig. 3) based on differences in strain and degree ofleucosome development. The Southern Round Hill Area is domin-ated by partially melted metasediments of the Thackaringa Group

Paragon Group

Dalnit Bore Metasediments

Bijerkerno Metasediments

King Gunnia Calc-Silicate MemberCartwrights Creek Metasediments

Sundown Group

BrokenHill

Group

Purnamoota Subgroup

Hores Gneiss Silver KingFormation

Freyers Metasediments

Parnell Formation

Allendale MetasedimentsEttlewood Calc-Silicate Member

Thackaringa Group

Rasp Ridge Gneiss Himalaya Formation

Kyong Formation

Cues FormationAlders Tank Formation

Alma Gneiss Lady Brassey Formation

Thorndale Composite Gneiss

Clevedale Migmatite

MulculcaFormation

Ednas Gneiss

Redan Gneiss

Adelaideanunconformity

Will

yam

aS

uper

grou

p

Fig. 2. Stratigraphic column of the Willyama Supergroup(adapted after Stevens et al., 1988).

HIGH-TEMPERATURE SHEAR ZONES IN BROKEN HILL , AUSTRAL IA 747


and middle to upper parts of the Broken Hill Group (Fig. 2)(Bradley, 1980). The Northern Round Hill Area comprises inter-bedded psammite and pelite units of the Sundown Group (Fig. 2)(Bradley, 1980).

Southern Round Hill Area

The Southern Round Hill Area (Fig. 3) comprises partially meltedmetasedimentary rocks that preserve well-defined quartzofeldspathicleucosome layers (Fig. 4a,b). Leucosomes are 8–10 cm wide and arecommonly sub-parallel to compositional layering and the dominant,layer-parallel foliation preserved within the metasedimentary rocks.The leucosomes comprise medium- to coarse-grained (up to 5 mm)quartz + feldspar, and encompass garnet clusters that form small,

disjointed and highly fractured clots that aggregate to form large (upto 5 cm diameter), irregular-shaped masses (Fig. 4a,b).The host rock to the leucosomes is dominated by sillimanite-bio-

tite-garnet pelitic gneiss interlayered with lesser psammopelitic topsammitic layers. Foliations are best developed within the peliticlayers. Two generations of sillimanite are preserved within the pel-ites. The first occurs as fibrolitic, milky white wisps up to 3 cmlength, and are overprinted by coarse sillimanite prisms up to 5 cmlength. Alkali feldspar is generally the only feldspar in the rock, andis commonly partially to totally retrogressed to sericite and fine-grained quartz. Garnet porphyroblasts are up to 7 cm in length, andare wrapped by the dominant foliation in the area (S2). The garnetoften occurs as irregular-shaped clots, however, they occasionallyshow dextral, and more commonly, sinistral asymmetry.

L3

L4

L4

L3

L3

L4

L3

L4

L3L4

L4

L3

RHA51

L3

L4

100 m

RHA58

RHA55

RHA51

RHA50

RHA59

Lithologies Symbols

Fig. 3. Geological map of the Round Hill Area showing the high-temperature shear zone mapped in this study. The Northern andSouthern Round Hill Areas, and locality of selected samples used are shown.



Psammopelitic to psammitic layers are up to 30 cm thick and aredominated by quartz and feldspar with minor biotite, sillimanite andsubordinate fine-grained garnet. Foliations are not as well developedin these layers as in the pelitic layers. A well-developed quartz,

sillimanite and biotite mineral lineation (Lm) is preserved on theplane of compositional layering. The mineral lineations plungemoderately W–SW throughout the area (Fig. 5).Strain intensity gradually increases northwards through the

Southern Round Hill area towards a high-temperature shear zone(Fig. 3) that separates the Southern from the Northern Round HillArea. The high-temperature shear zone is defined by increaseddevelopment of kinematic shear features such as S-C fabrics andsheared porphyroclasts. The C-fabric is NE-trending and mostly dipsmoderately to the NW (Fig. 3). Aspect ratios of garnet porphyro-blasts increase from c. 1:1 in the south to c. 14:1 in the north of thearea (Noble, 2000). Within the shear zone, the Lm mineral lineationbecomes more pronounced on the S1 plane and the defining silli-manite, biotite and quartz minerals become more elongate. It isdifficult to determine an overall shear sense on the high-temperatureshear zone as it has most likely been reactivated at least once since itsformation. The original orientation of the shear zone is also difficultto determine due to refolding during later events.Small, retrograde shear zones up to c. 25 m wide anastamose

through the northern extent of the Southern Round Hill Area(Fig. 3), and can easily be distinguished from the high-temperatureshear zone as they are dominated by phyllitic, chlorite + sericite-rich metasedimentary rocks that preserve a well-developed musco-vite ± chlorite mineral lineation. The northern extent of theSouthern Round Hill Area is defined by a marked decrease inkinematic indicators (S-C fabric, sheared porphyroblasts).

Northern Round Hill Area

The Northern Round Hill Area (Fig. 3) comprises pelite-dominatedmetasedimentary rocks interbedded with lesser psammite, whichpreserve less evidence of leucosome development or strain parti-tioning into high-temperature shear zones than in the SouthernRound Hill Area (Fig. 4c). Pelitic units comprise sillimanite, biotite,quartz, feldspar (dominantly K-feldspar) and garnet. Garnet occursas irregularly shaped clots up to 3 cm in length. However, thesebecome less common and smaller in size towards the north of theNorthern Round Hill Area. Sillimanite occurs as aligned, elongate,milky white needles up to 3 cm in length. Psammitic layers are up to30 cm thick and comprise medium-grained quartz, feldspar (domi-nantly alkali feldspar), biotite and minor sillimanite. The Lm minerallineation is not as well developed in the Northern Round Hill Area asin the Southern Round Hill Area; however, where preserved, itplunges moderately SW (Fig. 5a).Numerous retrograde shear zones occur sub-parallel to composi-

tional layering and S1 throughout the Northern Round Hill Area.These retrograde shear zones are defined by phyllitic metasedimentspreserving a well-developed muscovite ± chlorite lineation.Pegmatites comprising coarse grained quartz, feldspar and platy

white mica are also preserved sub-parallel to compositional layeringand S1 throughout the Northern Round Hill Area. The pegmatitesare up to 20 m wide and are occasionally foliated by S3.

Deformation elements recognizable in the field

The earliest fabric recognizable in outcrop within the Round HillArea is a pervasive, well-developed foliation that occurs parallel tocompositional layering and is best developed within pelitic horizons.This fabric is termed S2 as it post-dates an inclusion fabric (S1) thatis defined by alignment of fine-grained biotite and sillimanite, andhas only been recognized in thin section. This assemblage is describedin the petrogenesis section. S2 is defined by sillimanite and alignedaggregates of feldspar + quartz. Sillimanite occurs as fibrolitic wispsthat anastamose between other grain aggregates within the SouthernRound Hill Area. In the Northern Round Hill Area, sillimanitewithin the S2 fabric tends to be more prismatic. S2 wraps aroundgarnet porphyroblasts and generally trends N–NE (020–045�), and ismoderately to steeply dipping (Figs 3 & 5b); however, it becomesmore variably oriented in the Northern Round Hill Area, where ithas been folded during later deformation events (Figs 3 & 5b). No

Fig. 4. Field photographs of the Round Hill Area. SouthernRoundHill Area pelites showingwell-developed leucosomes (lightlayers) (a and b). Garnet clots in leucosome and mesosome layers(b). Lighter is c. 5 cm long. (c) Interbedded psammite and pelitefrom the Northern Round Hill Area. Pencil is c. 15 cm length.



folding associated with S2 has been documented. Within theSouthern Round Hill Area, the S2 fabric defines a shear fabric pre-serving moderate- to well-developed S-C fabrics.

The third-generation foliation (S3) is c. 1 mm spaced and definedby biotite and coarse-grained sillimanite prisms that cut through andoverprint garnet porphyroblasts. S3 is commonly shallowly tomoderately dipping, and variably oriented throughout the area(Fig. 5c). The S3 fabric trend is often at a high-angle (60–90�) to S2(Fig. 3). Within the Northern Round Hill Area, S3 is axial planar toinclined to recumbent, tight to open folds. F3 folds are variablyplunging because of later refolding (Fig. 5d). Recumbent F3 foldshave been interpreted as regional F2 folds by Forbes & Betts (2004)and Forbes et al. (2004), and as F1 folds by Marjoribanks et al.(1980) and Hobbs et al. (1984).

The fourth-generation fabric (S4) is a well developed, c. 2–5 mmspaced foliation defined by sillimanite and lesser biotite that over-grows and crenulates pre-existing fabrics. S4 trends NE (020–030�), issteeply dipping (Fig. 5e), and is commonly sub-parallel or at a lowangle to S2. Within the Northern Round Hill Area, S4 is also definedby spaced (up to 2 cm) stylolites and fractures (up to 10 cm spacing).Within the Northern Round Hill Area, S4 is axial planar to upright,open to close, parallel folds that are mostly plunging shallowly to theSW, however occasionally plunge steeply to the NW (Fig. 5f). Thismay be attributed to gentle warping during later deformation.

METAMORPHIC PETROLOGY

Petrogenesis

The petrogenesis of pelites from the Southern andNorthern Round Hill Areas was determined from thin-section analysis. In general, a similar sequence ofoverprinting minerals was observed, with subtle butsignificant differences in samples from the differentareas. This section presents these mineral overprintingrelationships, followed by an interpretation of themineral petrogenesis and its relation to mappedstructural elements.

Observed mineral overprinting relationships

The earliest recognized mineral assemblage is repre-sented by fine-grained biotite + quartz + sillima-nite ± plagioclase ± muscovite inclusions preservedwithin coarse-grained minerals that comprise the

Fig. 5. Equal-area stereographic projections presenting data from the Round Hill Area. (a) Mineral lineations defined by sillima-nite + biotite + quartz; (b) Poles to S2; (c) Poles to S3; (d) L3 intersections (solid squares and circles); F3 plunge-plunge directions(open squares); (e) Poles to S4; (f) L4 intersections.



matrix of the pelites. Biotite, quartz and sillimaniteinclusions occur within K-feldspar and retrogressedcordierite porphyroblasts (Fig. 6a,b). Biotite has adeep red colour, and appears as euhedral, tabulargrains that are commonly equally spaced and aligned(Fig. 6a–c); however, locally it can show a more ran-dom orientation. Quartz inclusions occur as small,round grains, and the less common sillimanite inclu-sions occur as fine, prismatic grains that are oftenaligned in a similar orientation to biotite inclusions.Within the shear zone (sample RHA51; Fig. 3), rarefine-grained inclusions of plagioclase occur withincordierite porphyroblasts. Muscovite also occurswithin the cordierite inclusions as semi-continuousgrains with highly irregular grain boundaries as theresult of chlorite and sericite (pinite) retrogression ofthe hosting cordierite. The muscovite also has higherbirefringence than the surrounding pinitized cordierite.

Inclusions of sillimanite and, less commonly, biotiteare also observed within garnet porphyroblasts(Fig. 6d). Sillimanite inclusions are fine-grained nee-dles and are locally randomly oriented, but generallyhave an overall trend through the garnet that locallyhas a sigmoidal shape, possibly as the result of shear-ing associated with later fabric development. Whereobserved, biotite inclusions within garnet are euhedralto subhedral, and occasionally contain sillimaniteinclusions that are continuous into the host garnet.

Small plagioclase feldspar and biotite grains, inter-preted to be related to the inclusion assemblage, alsooccasionally occur within the matrix. Biotite has highlyirregular grain boundaries and a deep red-brown col-our. Plagioclase feldspar is commonly sericitized, hashighly irregular grain boundaries, and is highly frac-tured (Fig. 6e).

The matrix of the pelites comprises granoblasticgarnet, cordierite, quartz, feldspar and biotite (Fig. 6f).Cordierite is retrogressed to pinite (chlorite + seri-cite), and contains remnant, black radiation haloesattributed to monazite and zircon inclusions. Largefeldspar grains are orthoclase-rich alkali feldspar (seeMineral chemistry) with very fine exsolution lamellaeoccasionally observed. In general, the K-feldspar doesnot display twinning, and is easily distinguished frommatrix quartz by its poikioblastic texture. Partial tototal retrogression of K-feldspar to sericite and fine-grained quartz is evident. Matrix quartz is commonlyrecrystallized and inclusion-free. Biotite interpreted tobe part of the granoblastic matrix assemblage com-monly occurs at the grain boundaries of quartz,K-feldspar and cordierite, or as small, decussateaggregates (Fig. 6f). The biotite has a euhedral tosubhedral habit, and is of paler colour than the biotitein the inclusion assemblage, commonly having a browncolour in plane-polarized light. The biotite is alsoobserved partially within cordierite, where it occa-sionally contains fine-grained sillimanite and small,round quartz inclusions. Grain boundaries of biotitewithin cordierite are commonly irregular and diffuse,

possibly as the result of retrogression of cordierite.Garnet occurs as large porphyroblasts that commonlydisplay an elongate shape; however, in the NorthernRound Hill Area, garnet tends to be more rounded inshape. Garnet can be classified into two types; inclu-sion-poor and -rich, the latter being not common northof the high-temperature shear zone. Inclusion-poorgarnet (Fig. 6a,c) commonly has well-defined grainboundaries and contains globular-shaped inclusions ofquartz and possibly feldspar; however, the feldspar isretrogressed and difficult to identify. The garnet itselfis relatively clear. Inclusion-rich garnet is characterizedby the high density of included sillimanite (Fig. 6a,c).However, this type of garnet lacks the globular quartzand feldspar inclusions observed within the inclusion-poor garnet (Fig. 6b,d). The sillimanite inclusions arecontinuous outside the garnet and into the matrix,where sillimanite appears as randomly oriented fibro-litic mats.

The granoblastic matrix is preserved as lensoidalpods that are wrapped by the dominant fabric withinthe pelites. In the Southern Round Hill Area, fibroliticmats define the fabric. Pseudomorphed porphyro-blasts, possibly once andalusite, occasionally occurwithin the fabric (Fig. 6g). However, definite identifi-cation of the primary mineral remains uncertain as theporphyroblasts are uncommon, they are small in sizeand have been highly deformed during simple shearing.S-C foliations are preserved within the dominant fab-ric. In the Northern Round Hill Area, a mix of fibroliteand prismatic sillimanite, which are fractured andbroken because of later deformation, define the fabric.Evidence of shearing within the fabric is not asprominent. In all samples, the fabric cross-cuts thegranoblastic matrix; however, locally garnet porphy-roblasts show evidence of being dragged into the fab-ric, resulting in a slight sigmoidal shape. In thinsection, sillimanite inclusions within the garnet are alsodragged into the main fabric, occasionally giving an�S�-shape to the overall sillimanite inclusion fabrictrend. Microlithons of earlier grown red-brown andpale brown coloured biotite occur within the fabric,and are particularly common in samples in theSouthern Round Hill Area. The grain boundaries ofthe biotite are highly irregular where they are in con-tact with the main fabric.

Splays of fine sillimanite needles oriented sub-par-allel to the trend of the dominant fabric are focussedalong grain boundaries and overprint minerals withinthe coarse-grained matrix, particularly K-feldspar,cordierite and quartz (Fig. 6f). Late retrograde biotitelocally overgrows the garnet porphyroblasts. The bio-tite preserves sillimanite inclusions (where presentwithin pre-existing garnet), and has an anhedral shapethat mimics the grain boundaries of the garnet.Coarse-grained, euhedral sillimanite prisms cross-cutall pre-existing mineral assemblages (Fig. 6c). The sil-limanite prisms either occur with their c-axis both sub-parallel and orthogonal to the thin section surface.



Fig. 6. Photomicrographs from the Round Hill Area. (a) S1 inclusionfabric defined by fine-grained biotite preserved within K-feldspar (ksp)grains. The fabric has a similar orientation in multiple K-feldspar grainswithin the photo. The S2 fabric is defined by fibrolite (RHA51); (b) S1inclusion fabric defined by fine-grained biotite (bi) preserved withinK-feldspar (ksp). The S2 fabric cross cuts the coarse K-feldspar andgarnet (g) within the photograph. Inclusion-rich (top) and -poor (bottom)garnet can be seen in the photograph (RHA51); (c) rare S1 inclusionfabric defined by biotite from sample RHA58. Late coarse-grained silli-manite (S3) overprints the coarse-grained matrix assemblages and S2; (d)inclusion-rich garnet (g with sill) hosting fine sillimanite needles, andinclusion-poor garnet (g) (RHA59); (e) rare plagioclase (pl) preservedwithin cordierite (cd) (RHA59); (f) Example of coarse-grained matrix ofthe pelites comprising cordierite (cd), K-feldspar (ksp) and quartz (qtz).Fine-grained biotite inclusions define the S1 fabric; (g) sheared porphy-roblasts pseudomorphed by fibrolite within the S2 fabric. The oxide in thephotograph is ilmenite (ilm).



The most obvious difference between the mineralpetrogenesis of the Southern and Northern Round HillAreas is the development of the inclusion assemblage,which is very commonly preserved within the SouthernRound Hill Area but is much less common in theNorthern Round Hill Area. Where observed in theNorthern Round Hill Area, the inclusions are definedsolely by biotite (Fig. 6c). Plagioclase was not observedin the matrix of the sample from the Northern RoundHill Area. Garnet porphyroblasts within the NorthernRound Hill area also lack the numerous sillimanite andoccasional biotite inclusions that are common withingarnet in the Southern Round Hill Area (Fig. 6d).Within the Northern Round Hill Area, rare partialrims of younger garnet growth are observed. The lategarnet has partly euhedral grain boundaries andovergrows the dominant fabric. As this garnet growthoccurred after the earliest stages of orogenesis, which isthe focus of this study, the origin of the partial rims isnot further considered.

Interpreted mineral petrogenesis

The aligned inclusion assemblage represents the earli-est phase of recognizable mineral growth in the RoundHill pelites, and defines S1. It shows a general patternof segregation into sillimanite-rich zones within garnetporphyroblasts, and biotite + sillimanite ± plagio-clase ± muscovite-rich zones outside of garnet por-phyroblasts within cordierite and K-feldspar. This maybe a function of compositional differences in the ori-ginal sediment, or may be the result of local strainvariations within the rock at the time the metamorphicminerals defining the inclusion assemblage weregrowing. In the latter case, sillimanite-rich zones arerepresentative of locally high-strain domains (e.g.Cobbold, 1977; Bell & Rubenach, 1983; Vernon, 1996).Additionally, the S1 fabric may be associated withlarger-scale strain variation as it is better developedwithin higher-strain areas of the Southern Round HillArea. However, relating this potential strain parti-tioning to a tectonic event is highly ambiguous as thefabric has only been recognized as an inclusionassemblage in thin section.

The granoblastic minerals overgrowing the inclusionassemblage are generally separated into garnet-richzones and K-feldspar + cordierite + quartz-richzones. This is interpreted to be due to the ease ofmineral nucleation and growth within different partsof a pre-existing, partially developed fabric. Theremnants of the pre-existing fabric are represented bythe inclusions described as the earliest recognizablemineral assemblage. During prograde metamorphism,K-feldspar and cordierite would have preferentiallygrown within original quartz- and feldspar-rich zones,and garnet nucleation would have been preferable inzones rich in sillimanite and biotite (e.g. Daniel &Spear, 1998) (Fig. 7). Overall, the textural relation-ships between coarse matrix minerals and the occur-

rence of leucosomes suggest that the following reaction(e.g. Yardley, 1989) has occurred:

biotiteþ sillimanite þ plagioclaseþ quartz

¼ cordierite þ garnetþK-feldsparþ liquid:

Nucleation and rapid growth of multiple garnetresulted in them eventually coalescing to form largeporphyroblasts (Fig. 7). The original garnet nucleihave become unrecognizable because of compositionalhomogenization of the garnet at high temperatures,which would erase any compositional gradientsdeveloped during the original growth (Daniel & Spear,1998). Garnet that is free of sillimanite inclusions isinterpreted to have overgrown pods of originallyquartz and feldspar-rich material that occurred withinthe sillimanite + biotite rich zones of the originalfabric (Fig. 7). Therefore, although there appear to betwo texturally different types of garnet within the peakassemblage, they are interpreted to be of the samegeneration.

The dominant (fibrolitic) sillimanite fabric describedin the thin sections is the mapped S2 fabric. Themicrostructural relationship of the S2 fabric wrappingand truncating the granoblastic garnet + cordier-ite + K-feldspar + quartz mineral assemblage can beused to interpret the relative timing of growth of thetwo mineral assemblages in two ways: (1) as two sep-arate metamorphic events where the garnet + cordi-erite + K-feldspar + quartz assemblage grew duringan early event and was overprinted by a second eventassociated with growth of the dominant fabric and alsosynchronous with deformation (e.g. Hand et al., 1992);or (2) a single event whereby the coarse-grained min-eral assemblage grew in locally low-strain domainswithin the rock unit, and the fibrolitic sillimanite foli-ation grew within high-strain domains (e.g. Vernon,1987, 1996).

Distinguishing between these two scenarios for theRound Hill pelites is difficult, but may be done if onemineral assemblage is recognized as being unstableduring the growth of the other (see Vernon, 1996).However, the minerals of the two assemblages withinthe Round Hill pelites (cordierite + garnet + K-feldspar + quartz and fibrolitic sillimanite) are stablewith one another (see following section on pseudo-sections) at granulite facies conditions.

Careful geochronological analysis of datable min-erals (e.g. zircon, monazite) that can be demonstratedto have grown synchronously with the metamorphicmineral assemblages may be used to distinguishbetween the scenarios (see Vernon, 1996); however,no such isotopic data are available. Previous datingof the timing of peak metamorphism (c. 1.6 Ga; Page& Laing, 1992) and the Olarian Orogeny (c. 1.6–1.59 Ga; Page et al., 2000a) may be useful for con-straining the relative timing of growth of the peakmineral assemblage and the S2 fabric. The coarse-grained, blocky sillimanite prisms that overprint the



Fig.7.Schem

aticsketch

oftheinterpretedmineralpetrogenesisfortheRoundHillpelites

(see

textforexplanation);Bi,biotite;cd,cordierite;fsp,feldspar;g,garnet;ksp,

K-feldspar;mu,muscovite;

pl,plagioclase;q,quartz;

sill,sillim

anite.



high-temperature mineral assemblage (Fig. 6e) areinterpreted to define the S3 fabric that is axial planarto the first generation of folds mapped in the RoundHill Area, and are considered to have developedduring the earliest stages of the Olarian Orogenyfollowing peak metamorphism (Page & Laing, 1992;Page et al., 2000a). The combination of overprintingrelationships and published ages of peak metamor-phism and timing of the Olarian Orogeny implies theS2 fabric must have developed almost synchronouslywith the peak mineral assemblage. The closetemporal relationship between the peak assemblageand the S2 fabric implies the coarse grainedgarnet + cordierite + K-feldspar + quartz peakassemblage may have grown in locally low-straindomains, and the fibrolitic sillimanite foliation withinlocally high-strain domains during a single granulitefacies metamorphic event. Localized strain contrastswithin the pelites developed during deformation, anddifferent metamorphic minerals preferentially nucle-ated and grew within the high and low strain zones.The splays of fine sillimanite needles orientedsub-parallel to the trend of the dominant fabric andoverprinting the coarse-grained assemblage within thelow-strain zones are interpreted to have grown withthe S2 fabric. This synchronous growth of sillimanitein both high- and low-strain zones was also describedby Vernon (1996).

Mineral chemistry

Electron microprobe analyses were undertaken on four thin sec-tions of metapelite, one each from samples RHA58 and RHA59,and two from sample RHA51 (Fig. 3) to determine the com-positions of minerals and identify any zoning. Analyses wereconducted using a Cameca SX50 electron microprobe at theUniversity of Melbourne Joint Electron Microprobe (JEM)Facility, using an accelerating voltage of 15 kV, and a current of25.05 lA. PAP matrix corrections were used for matrix correctionand data reduction. Representative results for individual samplesare shown in Tables 1–3. The Fe3+ content in garnet, feldspar,biotite, sericite and ilmenite was estimated using the AX softwareof Holland and Powell (http://www.esc.cam.ac.uk/astaff/holland/ax.html).

Garnet

Garnet porphyroblasts were analysed in sections RHA59R,RHA51A and RHA51B. In all cases, garnet is almandine-rich(average XAlm: sample RHA59 ¼ 0.82; sample RHA51 ¼ 0.81),with minor pyrope, and negligible grossular and spessartine con-tents (Tables 1 & 2; Fig. 8). Garnet profiles mostly show localizedvariations in composition, with no obvious zonation patterns(Tables 1 & 2; Fig. 8), possibly reflecting homogenization of garnetbecause of intragranular diffusion during high-temperature meta-morphism (e.g. Bohlen, 1987; Marmo et al., 2002). However,localized rimward increases in almandine content and a corres-ponding decrease in pyrope content were observed in sampleRHA59 (Fig. 8b), and weakly within both thin sections fromsample RHA51 (e.g. Fig. 8d). This is interpreted to reflect minordiffusional Fe-Mg exchange during cooling. Minor to negligiblevariation in grossular and spessartine contents occur in theseprofiles.

Table 1. Representative electron microprobe mineral analyses for sample RHA59, representing pelitic rocks below thehigh-temperature shear zone.

Sample RHA59R

Spot: 113 122 126 137 158 162 165 149 151 154 155 136 159 160 161 166

to 12 oxygen to 8 oxygen to 11 oxygen to 5 oxygen to 3 oxygen

Mineral: gnt (rim) gnt (core) gnt (core) gnt (rim) fsp fsp fsp bi bi bi bi sill (gnt incl) sill sill ilm ilm

SiO2 39.24 39.08 38.99 36.27 65.70 65.85 65.85 36.20 36.44 35.94 36.66 38.52 38.48 37.91 0.00 0.01

TiO2 0.03 0.00 0.02 0.02 0.00 0.00 0.03 3.37 3.05 3.36 3.64 0.05 0.00 0.00 54.83 53.55

Al2O3 19.87 20.29 20.38 23.00 18.57 18.67 18.56 18.21 19.08 18.80 18.76 53.94 61.98 61.80 0.00 0.00

Cr2O3 0.00 0.00 0.00 0.00 0.00 0.03 0.00 0.00 0.00 0.11 0.02 0.00 0.02 0.04 0.00 0.00

FeO 35.83 35.78 36.09 33.60 0.00 0.02 0.32 21.97 21.47 21.11 20.36 6.84 0.12 0.14 44.62 44.75

MnO 0.99 0.96 0.94 0.45 0.00 0.00 0.01 0.03 0.03 0.00 0.02 0.23 0.00 0.00 0.45 0.55

MgO 3.00 2.94 3.30 1.65 0.00 0.01 0.01 7.73 8.21 7.83 8.51 0.36 0.01 0.00 0.24 0.11

ZnO 0.06 0.06 0.00 0.00 0.08 0.06 0.00 0.00 0.02 0.08 0.19 0.00 0.19 0.02 0.03 0.09

CaO 1.24 1.18 1.23 0.62 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.22 0.01 0.01 0.12 0.00

Na2O 0.04 0.01 0.00 0.05 2.31 1.60 1.69 0.15 0.16 0.06 0.15 0.01 0.01 0.03 0.01 0.05

K2O 0.00 0.01 0.00 0.18 13.43 14.51 14.03 9.46 9.42 9.58 9.44 0.00 0.02 0.01 0.00 0.02

Total 100.30 100.31 100.95 95.84 100.09 100.75 100.51 97.13 97.87 96.88 97.77 100.17 100.84 99.96 100.30 99.13

Si 3.13 3.12 3.04 3.00 3.00 3.00 3.01 2.73 2.72 2.71 2.73 1.27 1.22 1.21 0.00 0.00

Ti 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.19 0.17 0.19 0.20 0.00 0.00 0.00 1.03 1.02

Al 1.87 1.91 1.88 2.24 1.00 1.00 1.00 1.62 1.68 1.67 1.65 1.54 1.70 1.71 0.00 0.00

Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00

Fe3+ 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 – – – 0.00 0.00

Fe2+ 2.39 2.39 2.51 2.33 0.00 0.00 0.00 1.39 1.34 1.33 1.27 0.14 0.00 0.00 0.93 0.95

Mn 0.07 0.07 0.06 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.01 0.01

Mg 0.36 0.35 0.37 0.20 0.00 0.00 0.00 0.87 0.91 0.88 0.94 0.01 0.00 0.00 0.01 0.00

Zn 0.01 0.01 0.00 0.00 0.01 0.01 0.00 0.00 0.00 0.01 0.02 0.00 0.00 0.00 0.00 0.00

Ca 0.11 0.10 0.10 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00

Na 0.01 0.00 0.00 0.01 0.20 0.14 0.15 0.02 0.02 0.01 0.02 0.00 0.00 0.00 0.00 0.00

K 0.00 0.00 0.00 0.02 0.78 0.84 0.82 0.91 0.90 0.92 0.90 0.00 0.00 0.00 0.00 0.00

Total 7.94 7.94 8.00 7.89 5.00 5.00 4.98 7.73 7.74 7.73 7.72 2.97 2.93 2.93 1.98 1.99

Sample locality is shown in Fig. 3.



Table 2. Representative electron microprobe mineral analyses for sample RHA51, representing pelitic rocks within the high-tem-perature shear zone.

Sample RHA51

Sample (A/B): A A B B A A B B A B B A B B B B

Spot: 21 19 76 83 31 33 108 112 28 95 92 27 104 45 105 86

to 12 oxygen to 8 oxygen to 11 oxygen to 5 oxygen

Mineral: gnt (core) gnt (rim) gnt (core) gnt (rim) fsp fsp fsp fsp bi (incl.) bi (incl.) bi (early) bi (after gnt) bi (after gnt) sill (incl.) fib (fol.) sill (blade)

SiO2 38.17 38.85 38.79 38.51 65.55 65.70 66.59 65.81 35.90 35.69 35.54 36.11 36.43 38.25 38.57 37.50

TiO2 0.02 0.00 0.04 0.03 0.00 0.05 0.03 0.00 3.21 3.32 3.21 3.16 3.29 0.03 0.03 0.02

Al2O3 20.97 21.07 20.46 20.79 18.72 18.81 18.60 18.35 18.74 19.30 19.24 18.31 18.58 49.32 61.70 62.96

Cr2O3 0.00 0.00 0.00 0.05 0.02 0.09 0.00 0.01 0.06 0.03 0.05 0.00 0.04 0.02 0.05 0.02

FeO 35.34 35.84 35.92 36.29 0.00 0.05 0.00 0.02 21.06 21.06 21.23 20.19 19.22 11.52 0.15 0.34

MnO 0.91 0.94 0.86 0.89 0.00 0.01 0.01 0.01 0.06 0.02 0.00 0.02 0.00 0.25 0.01 0.00

MgO 3.43 3.25 3.46 2.95 0.00 0.01 0.00 0.01 7.54 7.38 7.72 8.46 8.90 0.92 0.01 0.00

ZnO 0.00 0.02 0.00 0.00 0.08 0.02 0.09 0.09 0.04 0.25 0.18 0.04 0.12 0.03 0.00 0.00

CaO 1.20 1.16 1.17 1.13 0.11 0.04 0.07 0.06 0.00 0.00 0.00 0.00 0.00 0.36 0.01 0.00

Na2O 0.00 0.06 0.00 0.01 2.27 1.93 2.16 1.80 0.15 0.12 0.12 0.19 0.16 0.03 0.00 0.00

K2O 0.01 0.01 0.01 0.00 13.56 13.90 13.70 14.26 9.37 9.41 9.35 9.50 9.28 0.02 0.03 0.00

Total 100.06 101.20 100.72 100.65 100.31 100.63 101.26 100.43 96.12 96.58 96.64 95.98 96.03 100.74 100.57 100.86

Si 3.05 3.07 3.08 3.07 2.99 2.99 3.01 3.01 2.73 2.70 2.69 2.74 2.74 1.28 1.24 1.19

Ti 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.18 0.19 0.18 0.18 0.19 0.00 0.00 0.00

Al 1.98 1.96 1.92 1.95 1.01 1.01 0.99 0.99 1.68 1.72 1.72 1.64 1.65 1.43 1.72 1.74

Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Fe3+ 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 – – –

Fe2+ 2.36 2.37 2.39 2.42 0.00 0.00 0.00 0.00 1.34 1.33 1.34 1.28 1.21 0.24 0.00 0.01

Mn 0.06 0.06 0.06 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00

Mg 0.41 0.38 0.41 0.35 0.00 0.00 0.00 0.00 0.85 0.83 0.87 0.96 1.00 0.03 0.00 0.00

Zn 0.00 0.00 0.00 0.00 0.01 0.00 0.01 0.01 0.00 0.03 0.02 0.00 0.01 0.00 0.00 0.00

Ca 0.10 0.10 0.10 0.10 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00

Na 0.00 0.01 0.00 0.00 0.20 0.17 0.19 0.16 0.02 0.02 0.02 0.03 0.02 0.00 0.00 0.00

K 0.00 0.00 0.00 0.00 0.79 0.81 0.79 0.83 0.91 0.91 0.90 0.92 0.89 0.00 0.00 0.00

Total 7.96 7.96 7.96 7.95 5.01 4.99 4.99 5.00 7.72 7.74 7.75 7.74 7.72 3.00 2.96 2.94


Table 3. Representative electron microprobe mineral analyses for sample RHA58, representing pelitic rocks above the high-tem-perature shear zone.

Sample RHA58L

Spot: 25 26 27 34 19 22 11 15 5 6 3 31 32 33 38 39

to 8 oxygen to 11 oxygen to 5 oxygen to 11 oxygen to 3 oxygen

Mineral: fsp fsp fsp fsp bi (incl.) bi (early) bi (clear) bi (hydro) sill (fol.) sill (fol.) sill (blade) ser ser ser ilm ilm

SiO2 65.51 66.66 64.63 65.93 35.79 35.09 34.35 35.08 36.81 36.87 36.77 47.20 48.20 47.59 0.00 0.00

TiO2 0.05 0.00 0.03 0.00 2.14 2.79 2.61 2.64 0.02 0.00 0.02 0.15 0.07 0.23 52.75 52.40

Al2O3 18.73 19.54 18.84 19.14 19.32 18.98 19.24 19.01 63.06 63.16 62.99 35.75 36.53 35.62 0.00 0.02

Cr2O3 0.00 0.01 0.01 0.00 0.02 0.02 0.03 0.03 0.05 0.11 0.08 0.01 0.01 0.00 0.00 0.05

FeO 0.01 0.01 0.04 0.00 21.12 22.01 22.37 21.48 0.08 0.08 0.15 1.18 1.10 1.19 45.77 45.55

MnO 0.01 0.00 0.01 0.02 0.07 0.02 0.06 0.08 0.01 0.00 0.00 0.02 0.00 0.02 0.79 0.75

MgO 0.00 0.01 0.00 0.00 7.53 7.25 7.39 7.46 0.02 0.00 0.02 0.74 0.71 0.78 0.09 0.11

ZnO 0.05 0.08 0.00 0.00 0.15 0.06 0.08 0.02 0.00 0.07 0.00 0.00 0.00 0.03 0.33 0.00

CaO 0.05 0.16 0.08 0.16 0.03 0.00 0.00 0.02 0.01 0.01 0.03 0.00 0.02 0.00 0.00 0.00

Na2O 1.83 6.15 3.00 3.50 0.10 0.07 0.12 0.12 0.02 0.01 0.01 0.49 0.49 0.44 0.03 0.01

K2O 13.90 8.15 12.23 11.77 8.83 9.10 9.07 9.07 0.01 0.00 0.00 10.55 10.54 10.62 0.00 0.00

Total 100.15 100.77 98.88 100.53 95.10 95.39 95.33 95.01 100.09 100.32 100.08 96.10 97.66 96.51 99.77 98.89

Si 3.00 2.98 2.98 2.98 2.74 2.70 2.65 2.70 1.18 1.18 1.18 3.11 3.11 3.12 0.00 0.00

Ti 0.00 0.00 0.00 0.00 0.12 0.16 0.15 0.15 0.00 0.00 0.00 0.01 0.00 0.01 1.00 1.00

Al 1.01 1.03 1.02 1.02 1.74 1.72 1.75 1.73 1.76 1.75 1.75 2.77 2.78 2.75 0.00 0.00

Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Fe3+ 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 – – – 0.00 0.00 0.00 0.00 0.00

Fe2+ 0.00 0.00 0.00 0.00 1.35 1.42 1.45 1.38 0.00 0.00 0.00 0.07 0.06 0.07 0.97 0.97

Mn 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.02

Mg 0.00 0.00 0.00 0.00 0.86 0.83 0.85 0.86 0.00 0.00 0.00 0.07 0.07 0.08 0.00 0.00

Zn 0.01 0.01 0.00 0.00 0.02 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Ca 0.00 0.01 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Na 0.16 0.53 0.27 0.31 0.02 0.01 0.02 0.02 0.00 0.00 0.00 0.06 0.06 0.06 0.00 0.00

K 0.81 0.47 0.72 0.68 0.86 0.89 0.90 0.89 0.00 0.00 0.00 0.89 0.87 0.89 0.00 0.00

Total 4.99 5.02 5.00 5.00 7.72 7.74 7.78 7.74 2.94 2.94 2.94 6.97 6.96 6.97 2.00 2.00




Feldspar

Feldspar defining the peak metamorphic mineral assemblage wasanalysed in all four samples, and has high orthoclase content(Tables 1–3; Fig. 9). Alkali feldspar in sample RHA59R is

dominantly orthoclase (Or0.79)0.87), with minor albite (Ab0.14)0.21)and no anorthite (An0), where Or ¼ K/(K + Na + Ca), Ab ¼ Na/(K + Na + Ca) and An ¼ Ca/(K + Na + Ca). Sample RHA51 isdominantly orthoclase (Or0.74)0.86), with minor albite (Ab0.14)0.26)and negligible anorthite (An0.002)0.005). Sample RHA58L shows thehighest variation in alkali feldspar composition of Or46)83-Ab17)53An0.2)0.8 (Fig. 9). Plagioclase feldspar was not analysed as itis extensively sericitized, and where preserved is very fine-grained andhighly fractured.

Biotite

Three generations of biotite were observed within the Round HillArea. The earliest biotite occurs as fine-grained inclusions withincordierite and feldspar. Biotite inclusions throughout all sampleshave similar XFe (Fe

2+/(Fe2+ + Mg)), but are slightly more mag-nesian in sample RHA59R (RHA59R: XFe ¼ 0.60–0.62; RHA51:XFe ¼ 0.61–0.62; RHA58: XFe ¼ 0.61–0.62). Biotite replacing garnethas a lower XFe than the earlier biotite inclusions (RHA59R: XFe ¼0.57; RHA51: XFe ¼ 0.55–0.58). The average Ti content of biotitedecreases from south to north (RHA59: Ti ¼ 0.17–0.20 cations per11 oxygen; RHA51: Ti ¼ 0.18–0.19 cations pfu; RHA58: Ti ¼ 0.12–0.19 cations pfu). The concentration of Ti within individual biotitegenerations did not show any consistent variation.

Sillimanite/fibrolite

Fibrolite was analysed from within the pervasive foliation preservedin the samples (Tables 1–3). The prisms cross-cutting the pervasivefabric and garnet porphyroblasts were determined to be sillimanite(Tables 1–3).

Oxides

All oxides analysed in the Round Hill Area are ilmenite (Tables 1–3).Ilmenite in sample RHA59R shows the highest variation, with XIlm

(Fe2+/(Fe2+ + Mn + Mg)) ranging from 0.97 to 0.98; XGeik (Mg/(Fe2+ + Mn + Mg)) ranging from 0.004 to 0.0023 and XPyro (Mn/(Fe2+ + Mn + Mg)) ranging from 0.003 to 0.010. SampleRHA58L showed lesser variation in oxide composition (XIlm ¼ 0.98;XGeik ¼ 0.003–0.008; XPyro ¼ 0.01–0.02).

Cordierite

No unaltered cordierite was recognized in any samples.

RHA59 Traverse 1(a)

(b)

(c)

(d)

RHA59 Traverse 2

RHA51 Traverse 2

RHA51 Traverse 3

Fig. 8. Compositional profiles through garnet. Sample localitiesshown in Fig. 3.

Fig. 9. Normalized orthoclase (Or):albite (Ab):anorthite (An)tri-plot for felspar analysed in the Round Hill pelites. Samplelocalities shown in Fig. 3.



P–T pseudosections

Pseudosections assist in visualization of changes inmultivariant mineral assemblages and univariantreactions for a set bulk-rock composition over arange of P–T conditions (e.g. Hensen, 1971; Guiraudet al., 1990; Powell & Downes, 1990; White et al.,2001; Boger & White, 2003), and were used in thisstudy to model the prograde P–T path of the RoundHill pelites. The pseudosections shown were calcula-ted using the individual bulk-rock compositions ofselected samples from across the Round Hill Area,which were analysed using the Geochemical Evolu-tion and Metallogeny of Continents (GEMOC) XRFfacilities at Macquarie University. Results of whole-rock XRF analysis are shown in Table 4. Threesamples from across the Round Hill Area wereselected for metamorphic analysis. Pseudosectionswere calculated in the model system Na2O-CaO-K2O-FeO-MgO-Al2O3-SiO2-H2O (NCKFMASH) usingTHERMOCALC (Powell & Holland, 1988; Powellet al., 1998). Pelites from the Southern Round HillArea are represented by pseudosections drawn forsamples RHA59 (Fig. 10a) and RHA51 (Fig. 10b).RHA51 was taken from within the high-strain zone ofthe Southern Round Hill Area. Figure 10(c) is takento represent pelites in the Northern Round Hill Area,and is drawn for sample RHA58. Sample localitiesare shown in Fig. 3. Each section includes the phasesandalusite (and), biotite (bi), chlorite (chl), clinozoi-site (cz), cordierite (cd), garnet (g), K-feldspar (ksp),kyanite (ky), liquid (liq), muscovite (mu), plagioclase(pl), sillimanite (sill), and staurolite (st). Quartz (qtz)is assumed to be in excess throughout the wholesystem. H2O is assumed to be the pure fluid phaseand in excess below the solidus. H2O was set indi-vidually for each section at the wet solidus so that therocks were just fluid-saturated (Fig. 10a–c).

Setting the effective bulk-rock composition used tocalculate a pseudosection is a highly critical step as thisdirectly influences the series of metamorphic reactions

seen within a rock. Mineral zonation within coarse-grained assemblages, particularly garnet, during pro-grade and retrograde events can significantly changethe effective bulk-rock composition as mineral coresmay become chemically isolated from the remainder ofthe rock during metamorphism (Stuwe, 1997; Marmoet al., 2002). Anhydrous minerals, such as garnet,commonly preserve extensive zonation resulting fromslow diffusional processes during prograde metamor-phism (e.g. Yardley, 1977; Vance & Holland, 1993;Clarke et al., 1997). However, at higher temperatures(c. >650 �C) the rate of volume diffusion increasesand garnet can become chemically homogenized, lim-iting the extent of chemical fractionation of the bulk-rock composition (e.g. Woodsworth, 1977; Bohlen,1987; Marmo et al., 2002). Electron microprobe resultsof the Round Hill samples show no significant chem-ical zonation within the minerals, particularly withrespect to garnet (Fig. 8). As it would be difficult toaccount for any potential fractionation of the bulk-rock composition during porphyroblasts growth, wehave assumed the analysed bulk-rock composition is arepresentative of the samples.There is also the issue of a generally limited under-

standing of the original bulk composition of a rockprior to granulite facies metamorphism, and that thecomposition prior to peak metamorphism (i.e. duringprograde metamorphism) would differ from that ana-lysed now due to melt loss or gain during heating. Thisproblem is manifest by the lack of knowledge of thevolume of melt lost from or accumulated into the rock,the composition of this melt, and the awareness thatthe composition of the melt can change progressivelythroughout the history of the rock, also changing thebulk-rock composition (e.g. White & Powell, 2002).White et al. (2004) have described complex melt pro-duction within the Round Hill pelites, whereby pro-duction of melt was spatially focused around garnetnuclei. In this model, it is possible that a significantvolume of melt may have migrated away from the siteof melt production (White et al., 2004).

Table 4. Normalized bulk-rock compositions (in wt%) for pelitic samples from the Round Hill Area (this study), and from across theBroken Hill Block (data taken from the Broken Hill GIS database compiled by the New South Wales Geological Survey).

Sample

Bulk-rock composition (wt% oxides)

SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5

RHA50 75.41 0.72 12.58 4.99 0.05 1.05 0.31 0.80 3.94 0.15

RHA51 60.45 0.87 24.45 5.41 0.03 1.67 0.12 0.92 6.00 0.08

RHA55 58.29 0.86 24.56 7.88 0.13 1.52 0.12 0.42 6.05 0.14

RHS58 58.61 1.06 25.96 6.74 0.05 1.67 0.10 0.53 5.20 0.07

RHS59 53.19 0.78 21.18 17.79 0.33 1.90 0.58 0.46 3.67 0.14

Average all BHG & TG pelites 61.56 0.91 21.03 8.47 0.21 1.69 0.83 0.63 4.55 0.12

Average northern BHG & TG pelites 60.49 0.98 22.25 8.19 0.11 1.68 0.42 0.78 4.98 0.11

Average all SG pelites 60.71 0.91 22.29 8.16 0.17 1.65 0.58 0.55 4.86 0.12

Average northern SG pelites 60.04 0.89 23.32 8.04 0.14 1.49 0.25 0.37 5.29 0.18

Data for samples RHA59, RHA51 and RHA58 was used to calculate pseudosections shown in Fig. 10(a, b and c) respectively. Sample localities are shown in Fig. 3. Average pelite

compositions from across the Broken Hill Block taken for samples with SiO2 ¼ 55–65 wt% and Na2O <3 wt%.

TG, Thackaringa Group; BHG, Broken Hill Group; SG, Sundown Group.



The whole-rock geochemistry of the Round Hillpelites used here have been compared with the whole-rock chemistry of stratigraphically equivalent units inlower grade (amphibolite facies) areas of the BrokenHill Block that have not undergone significant melt-

ing (Table 4; Fig. 11). Figure 11 shows that there islittle significant difference between the compositionsof amphibolite facies and partially melted granulitefacies pelites, apart from sample RHA59, which isslightly more Fe-rich and Al-depleted (Fig. 11). As

T

P

Fig. 10. NCKFMASH P–T pseudosections calculated for pelitic samples from the Round Hill Area. In all pseudosections, H2O istaken to be the fluid phase and is set in excess below the solidus. Bulk-rock compositions used below the solidus are shown in Table 4.Above the solidus, H2O is set for individual pseudosections so that the rocks are just saturated; see specific figure captions forbulk-rock compositions. Quartz is taken to be in excess throughout the pseudosections. P–T paths shown are drawn for the observedmineral overprinting relationships relevant to each sample. (a) P–T pseudosection calculated for RHA59, representative ofpelites below the high-temperature shear zone. Bulk-rock composition above the solidus in mol.% is Al2O3 ¼ 42.23, CaO ¼ 1.97,MgO ¼ 9.10, FeO ¼ 21.24, K2O ¼ 7.33, Na2O ¼ 1.44 and H2O ¼ 16.69. The locality of the RHA59 is shown in Fig. 3. (b) P–Tpseudosection calculated for RHA51, representative of pelites within the high-temperature shear zone. Bulk-rock compositionabove the solidus and below the muscovite outline is Al2O3 ¼ 43.09, CaO ¼ 0.38, MgO ¼ 7.35, FeO ¼ 6.08, K2O ¼ 11.41,Na2O ¼ 2.53 and H2O ¼ 29.15. Bulk-rock composition above the solidus and above the muscovite outline is Al2O3 ¼ 55.83, CaO ¼0.49, MgO ¼ 9.52, FeO ¼ 7.88, K2O ¼ 14.78, Na2O ¼ 3.28 and H2O ¼ 8.21. The locality of RHA51 is shown in Fig. 3. (c) P–Tpseudosection calculated for RHA58, representative of pelites above the high-temperature shear zone. Bulk-rock compositionabove the solidus and below the muscovite out line is Al2O3 ¼ 46.13, CaO ¼ 0.33, MgO ¼ 7.60, FeO ¼ 7.60, K2O ¼ 10.14,Na2O ¼ 1.57 and H2O ¼ 26.62. Bulk-rock composition above the solidus and above the muscovite out line is Al2O3 ¼ 57.32, CaO ¼0.41, MgO ¼ 9.45, FeO ¼ 9.45, K2O ¼ 12.60, Na2O ¼ 1.95 and H2O ¼ 8.82. The locality of RHA58 is shown in Fig. 3.



the main phases formed through melting reaction (i.e.garnet and cordierite) have relatively low SiO2 con-tent, melt extraction should leave a restite with lowSi, and high Mg, Fe, Al contents compared withnormal pelite (e.g. Mengel et al., 2001; White &Powell, 2002). Table 4 shows the SiO2 content ofsample RHA59 is less than that of other Round Hillpelites and the average Broken Hill Group/Thackar-inga Group and Sundown Group pelites, and that theFe2O3 and MgO content of RHA59 is higher. Thissuggests sample RHA59 may be slightly restitic andmay have lost some melt.

As sample RHA59 has the most restitic composi-tion, it was used to calculate a trial pseudosection totest the sensitivity of the pseudosection to the additionof 30% melt of a given composition to the analysedbulk-rock composition (Fig. 12). The volume of meltadded was chosen from the percentage of melt expec-

ted for an average pelitic composition at the approxi-mate conditions of the Round Hill pelites, ascalculated in the NCKFMASH system by White et al.(2001) (see also White & Powell, 2002). Field evidenceshows that melt has been retained within the RoundHill pelites in the form of leucosomes (Fig. 4a,b),therefore the addition of 30% melt back into thepseudosection will model the largest possible variationin bulk-rock chemistry because of loss of melt from thesystem. The melt composition used to test the pseu-dosections was taken from White et al. (2001). Thepseudosection topologies were very similar, and nosignificant variation in the position or slope of the wetsolidus and the muscovite and biotite breakdownreactions are apparent (Figs 10a & 12).As the only assumption used in the original pseudo-

sections calculated for this study is the addition of H2Oto the point where the rocks were just saturated at the

T

P

Fig. 10. (Cont’d)



wet solidus, and as the determination of the true com-position and volume of melt lost from the Round Hillmetapelites is beyond the scope of this study, we haveassumed that the bulk-rock compositions determinedusingXRFanalysis are the closest estimate there is to theeffective bulk-rock composition of the Round Hillpelites. The following description of P–T paths deter-mined for the RoundHill pelites is based on the originalpseudosections calculated without attempting to addmelt back into the bulk-rock composition.

P–T paths

This study focuses on the prograde P–T path of thepelites. Two metamorphic events were identified inthe Round Hill Area. The first event (M1) is definedby the S1 inclusion assemblage and was synchronouswith the D1 event. The second metamorphic event

(M2) is represented by the coarse-grained peak min-eral assemblage, and is considered to have beenapproximately synchronous with D2 deformation.

M1

The M1 event is characterized by the S1 mineralassemblage of biotite + sillimanite + plagioclase ±muscovite. This assemblage is stable over a wide fieldin P–T space, particularly for samples RHA51(Fig. 10b) and RHA58 (Fig. 10c). The conditions ofM1 metamorphism are best constrained using sampleRHA59, for which the garnet stability field is large(Fig. 10a). The lack of garnet within the M1 assem-blage places a pressure constraint on the conditions ofmetamorphism. M1 metamorphism therefore occurredat temperatures of at least c. 600 �C at pressure of c.2.8–4.2 kbar (Fig. 10a).

T

P

Fig. 10. (Cont’d)



M2

The M2 event is represented by the coarse-grained gar-net + cordierite + K-feldspar + sillimanite + quartzpeak mineral assemblage identified in all samples, con-straining peak metamorphic conditions to a singlequadrivariant field in each pseudosection (Fig. 10a–c).For sample RHA59, this constrains M2 to between 4.5and 6.5 kbar and 750 and 870 �C (Fig. 10a); to 5 to>8 kbar at greater than c. 750 �C within sampleRHA51 (Fig. 10b); and to c. 4.8–8.1 kbar and 740 togreater than 900 �C in sample RHA58 (Fig. 10c).Overall, the fields inwhich theM2assemblagewithin theRound Hill pelites fall indicate that high-grade condi-tions within the area attained pressures of at leastc. 5 kbar and temperatures of at least 740 �C. Thesevalues agree with previous peak metamorphic estimatesof Powell &Downes (1990), who reported conditions of750–800 �C at c. 5 kbar for rocks in the Round HillArea.

Overall P–T path

Two end-member prograde P–T paths from the M1/D1 to M2/D2 events have been drawn on Fig. 13. Thispath shows an increase in temperature of at least140 �C associated with a pressure increase of c. 0.8–

2.2 kbar, depending on the original pressure of theM1/D1 event (Fig. 13). This implies some burial be-tween the D1 and D2 events.It is noteworthy that if the petrogenesis of sample

RHA59 is used on the trial pseudosection to which30% melt was added (Fig. 12), no significant variationin the gradient of the P–T paths of the applied and trialpseudosections is apparent. This is considered sup-portive of using the applied pseudosection to deter-mine the general prograde P–T path followed by theRound Hill pelites. However, extreme care would needto be taken in delineation of a detailed P–T path forthese pelites (see Vernon, 1996), particularly in relationto addressing the issue of a changing bulk-rock com-position during prograde metamorphism as a result ofmelt production and removal (see White & Powell,2002).The P–T path of the Round Hill Area can be further

extended using the work of Stuwe & Ehlers (1997),who constructed a retrograde P–T path for the NineMile Area (Fig. 1), which is in the vicinity of theRound Hill Area. Stuwe & Ehlers (1997) recognizedtwo distinct metamorphic events; an older event asso-ciated with the Olarian Orogeny and a younger eventpossibly associated with the Delamerian Orogeny. Forclarity, we have only incorporated the first metamor-phic event described by Stuwe & Ehlers (1997) into our

Fig. 11. Comparative bulk-rock composition plots of RoundHill pelites (analysed in this study) and whole-rock chemistry ofselected stratigraphic units from the Willyama Supergroup.



study. The first metamorphic event of Stuwe & Ehlers(1997) is associated with shortening during the earlystages of the Olarian Orogeny at c. 1600 Ma, and oc-curred at conditions of 4–5 kbar and 650 �C. Meta-morphism was followed by near-isobaric cooling thatwas interrupted by younger episodes of deformation.The overall P–T path is anticlockwise (Fig. 13).

In contrast, Swapp & Frost (2003) suggested the P–T path of the Broken Hill Block is clockwise, and thatthe peak metamorphic assemblage in the Round Hillpelites is defined by garnet + rutile + aluminosili-cate + ilmenite. The presence of rutile in the assem-blage implies higher-pressure conditions (at least9 kbar) associated with temperatures of 775 �C. Swapp

& Frost (2003) suggested that this high-pressure eventwas followed by high-temperature decompression atleast down to conditions of 5.5 kbar at 750 �C(Fig. 13), which they suggest are conditions of the lastequilibrated assemblage preserved in the pelites. Elec-tron microprobe analysis revealed no rutile within theRound Hill pelites and all oxides analysed wereilmenite (Tables 1–3), giving no evidence of high-pressure metamorphism or a clockwise P–T path.

DISCUSSION

Within the Round Hill Area, two metamorphic events(M1 & M2), both associated with deformation, and

T

P

Fig. 12. NCKFMASH trial P–T pseudosection calculated for sample RHA59 with 30% melt added back into the bulk-rockcomposition used to calculate the section. Melt composition was taken from White et al. (2001). Quartz is taken to be in excessthroughout the section. H2O is taken to be the fluid phase and is set in excess below the solidus. Above the solidus, H2O is set sothat the rocks are just saturated. Bulk-rock composition above the solidus in mol.% is Al2O3 ¼ 49.67, CaO ¼ 1.65, MgO ¼ 7.17,FeO ¼ 16.85, K2O ¼ 9.78, Na2O ¼ 2.93 and H2O ¼ 21.74.



occurring prior to the onset of crustal shorteningduring the Olarian Orogeny have been identified.These events record burial, and the M2 event is asso-ciated with peak granulite facies metamorphism atc. 1.6 Ga (Page et al., 2000a). Whether these arecompletely separate tectonothermal events or part of acontinuous P–T path of heating and burial is difficultto distinguish. Perturbations in the geothermal gradi-ent can assist in constraining the origin of thesetectonothermal events, and can be inferred from cal-culating the geotherm at the conditions of the meta-morphic events, as determined on a P–T grid.Calculation of the approximate local geothermal gra-dients of the M1 and M2 events was done assuming1 kbar of pressure equals a rock pile 3.5 km thick. Thegeothermal gradient of the M1 event was calculated tobe between 41 and 61 �C km)1, depending on theoriginal pressure (c. 2.8–4.2 kbar at 600 �C; Fig. 13).The M2 event geothermal gradient was calculated tobe 42 �C km)1 (Fig. 13). Both the M1 and M2 eventsare therefore associated with elevated geothermalgradients (assuming a normal geotherm of25 �C km)1).

Prograde path A shown on Fig. 13 is drawn ori-ginating at higher pressure at the M1 event, and

approximately follows the M1 geotherm of41 �C km)1 to peak M2 metamorphic conditions(Figs 13 & 14a). This path implies burial of 0.8 kbar,equivalent to 2.8 km of sediment between the M1 andM2 events (Fig. 14a). Path B (Fig. 13) originates atlower pressures, and implies burial of 2.2 kbar(c. 7.7 km sediment) between M1 and M2 (Fig. 14a).This path also moves to a shallower geotherm,implying that although the temperature of the rockpackage was becoming hotter towards conditions ofpeak metamorphism, the thermal gradient wasdecreasing (Figs 13 & 14a).Understanding the origin of the elevated M1 geo-

thermal gradient and the extent to which it was carriedover into the M2 event is critical to modelling theprograde tectonic history of the Broken Hill Block.Elevated geothermal gradients may be the result ofa number of reasons, including: (a) burial ofanomalously high heat-producing rock packages (e.g.Chamberlain & Sonder, 1990; Sandiford & Hand,1998; Sandiford et al., 1998; McLaren et al., 1999); (b)emplacement of voluminous granitoid sheets (e.g. Luxet al., 1986; Barton & Hanson, 1989; De Yoreo et al.,1989; Collins & Vernon, 1991); (c) heat flow into thebase of the crust through magma underplating

T

P

Fig. 13. Prograde P–T path for the BrokenHill Block showing the M1/D1 and M2/D2events. Paths A and B correspond to the twopossible end-members of the prograde P–Tpath, originating at higher (path A) or lower(path B) pressure conditions of M1. Theactual prograde P–T path would lie betweenthese two possibilities. The retrograde pathdetermined by Stuwe & Ehlers (1997) hasbeen added (light grey) and shows that thecomplete P–T path is anticlockwise. Theclockwise interpretation of Swapp & Frost(2003) is also shown (dark grey) (see text fordiscussion). Geothermal gradients calculatedfor the M1/D1 (41 �C km)1 and 61 �C km)1

at 600 �C depending on the original pressureconditions) and M2/D2 (42 �C km)1) eventsare shown.



(convection) and/or emplacement of mantle or lower-crustal derived magmas into the lower crust (advec-tion) (e.g. Bohlen, 1987, 1991; Bohlen & Mezger,1989).

Elevation of the M1 geothermal gradient inresponse to burial of high heat-producing rockpackages (e.g. U- and Th-rich sediments: Chamberlain& Sonder, 1990; highly radiogenic granites: Sandiford& Hand, 1998; Sandiford et al., 1998; McLaren et al.,1999) is not considered applicable, as no such anom-alous sequence has yet been identified within the Bro-ken Hill Block. Additionally, this scenario requiresburial prior to heating, whereas the prograde path ofthe Round Hill pelites indicates heating prior to burial.Emplacement of voluminous granitoid sheets can alsobe eliminated as maintenance of high-temperatureconditions would require granitoid intrusions to com-prise up to 50% of the total rock package volume(Barton & Hanson, 1989; Collins & Vernon, 1991), andno such volume of granites intruded early in the evo-lution of the Broken Hill Block have been recognized.In a similar manner, attainment of high temperaturesduring metamorphism solely by intrusion of maficmagmas can be dismissed as voluminous mafics havenot been recognized within the Broken Hill Block.Additionally, amphibolite bodies within the terraneyield ages of 1691 ± 9 Ma (e.g. Gibson & Nutman,2004), which is much older than the timing of peakmetamorphism (c. 1.6 Ga; Page et al., 2000a).

Demonstrative evidence for a magmatic underplatemay be a long wavelength positive gravity anomaly,

melt products with mantle signatures (e.g. picrites,peridotites, lherzolites), or depleted mantle isotopicsignatures (e.g. Foden et al., 2002). Dismissal of ele-vation of the geothermal gradient as a result of amagmatic underplate because of lack of such evidencewithin the Broken Hill Block would require muchcaution as such signatures may have been obscuredduring the intense deformational and metamorphichistory of the terrane during the Olarian andDelamerian Orogenies.

As a magmatic underplate cannot be unequivocallydemonstrated within the Broken Hill terrane, we con-sider the setting in which magmatic underplatingoccurs, and whether the characteristics of such envi-ronments may be identified within the region. Mag-matic underplating occurs in response to partialmelting of the asthenosphere because of adiabaticdecompression of the mantle (e.g. Van der Pluijm &Marshak, 1997). Asthenospheric decompression ismost commonly considered a response of lithosphericthinning (e.g. Wyborn et al., 1988; Lister et al.,1991; Van der Pluijm & Marshak, 1997; Giles et al.,2002).

Lower units of the Broken Hill stratigraphy com-prise quartzofeldspathic sequences grading into turbi-dites that were interpreted to be deposited within anevolving rift-basin (e.g. Stevens et al., 1988) c. 1.71–1.67 Ga (Page et al., 2000a). Gibson & Nutman (2004)suggested the timing of early rifting can be constrainedto 1.69–1.67 Ga and ceased at the time of depositionof the uppermost rift sediments in the Broken Hill

Fig. 14. Schematic representation of tectonic histories and resultant geothermal gradients interpreted for the M1/D1 and M2/D2prograde P–T path of the Broken Hill Block; (a) M1/D1 rifting c. 1.69–1.67 Ga; (b) M1/D1 rifting c. 1.64–1.61 Ga (see text fordiscussion).



stratigraphy (Sundown Group; Fig. 2), which have amaximum depositional age of c. 1.67 Ga (Page et al.,2000a). An early rifting event c. 1.69 Ga provides amechanism by which the M1 geothermal gradient mayhave been raised. However, the timing of the M2/D2event has been constrained to c. 1.6 Ga from the dat-ing of peak metamorphic assemblages (Page et al.,2000a). Therefore, the mechanism of burial betweenthe M1 and M2 events needs to be addressed, and thecause of the high temperatures of peak granulite faciesmetamorphism accounted for.

The increase in pressure of 0.8–2.2 kbar in the pro-grade P–T path between the M1 and M2 events(Fig. 13) is interpreted to be in response to depositionof the Paragon Group sag-phase sediments (Fig. 2)from c. 1.66 to 1.6 Ga (Page et al., 2000a; Raetz et al.,2002). This pressure increase would requirec. 2.5–7.5 km of sediment to be deposited. A moreprecise estimate of the required sedimentary thicknessrequires knowledge of the rock composition and den-sity (e.g. Yardley, 1989). As neither the top of theParagon Group in the Broken Hill Block, nor that of

Fig. 14. (Cont’d)



the laterally equivalent rocks (Mount Howden Sub-group) in the neighbouring Olary Domain is exposed,no assessment of the thickness of the sedimentary pileduring late sedimentation can be made. However, ananalogy can be drawn from the Mount Isa Inlier,which has been suggested to have shared a similartectonic history to the Broken Hill Block untilc. 1.5 Ga (Giles et al., 2004), and within which equiv-alent units to the Paragon Group are the McNamaraGroup. The McNamara Group represents sag-phasesediments up to 8 km thick (Betts et al., 1998; South-gate et al., 2000), deposited between 1.65 and 1.59 Ga(Page & Sweet, 1998; Page et al., 2000b) into the IsaSuperbasin (Southgate et al., 2000). This possiblysupports the interpretation that the Paragon Groupmay have been of sufficient thickness to provideenough sediment overburden to increase the pressureof the middle-crust by 0.8–2.2 kbar.

From prograde path A (Fig. 13), the period betweenthe M1 and M2 events can be modelled as being burialin response to sag-phase sedimentation following earlyrifting c. 1.69 Ga. However, lithospheric thermalrelaxation commonly follows rifting, and results inlowering the geothermal gradient (e.g. Morgan &Ramberg, 1987), and it is unlikely that the crustal pilewould follow a constant geothermal gradient withincreasing depth following rifting. A second deforma-tion event at c. 1.6 Ga that would alter the thermalstate of the crust during peak M2 metamorphism isalso not considered in this scenario. As there is clearevidence of a second deformation event within theRound Hill Area, this model is not considered to fullyaccount for the observations made.

Figure 14(a) shows two possible scenarios drawnfrom path B (Fig. 13) depicting the link between a1.69 Ga M1/D1 rift event and the c. 1.6 Ga M2/D2event. The first scenario models the D2 event asextensional (Fig. 14ai), and the second models aD2-thrusting event (Fig. 14aii). In the first model(Fig. 14ai), the lithosphere would have begun to sub-side and thermally relax at the cessation of c. 1.69–1.67 Ga rifting. As a result, the geothermal gradient ofthe lithosphere would have been reduced. A conse-quence of extension is that the geothermal gradient israised, therefore, as the final M2/D2 geothermal gra-dient was c. 42 �C km)1, the geothermal gradientwithin the Round Hill Area would have cooled to<42 �C km)1 during the interval between the M1/D1and M2/D2 events (Fig. 14ai). Renewed extensionduring the M2/D2 event would have resulted in ele-vation of the cooled geothermal gradient back to42 �C km)1 (Fig. 14ai). Although the geothermal gra-dient was decreasing, the high temperatures of granu-lite facies metamorphism would have been attainedduring the M2 event as the temperature of the rockpackage would be maintained through burial, andwould also have increased because of any additionalheat input during D2 lithospheric thinning. Thisscenario can also be applied to path A (Fig. 13).

In the thrusting model (Fig. 14aii), burial and ther-mal relaxation following early rifting would have ini-tiated cooling of the geotherm. As crustal thickeningresults in a decrease in the geothermal gradient, thegeotherm during burial would have lowered to a valuebetween 61 and 42 �C km)1 (Fig. 14aii). Crustalthickening during thrusting at c. 1.6 Ga would havelowered the geothermal gradient further toc. 42 �C km)1 (Fig. 14aii). The high temperature of therock package would have been maintained because ofburial.

Although the two models described account forvariations in temperature and thermal gradient duringthe prograde path of the Round Hill pelites, it is dif-ficult to account for the pressure conditions of the M1rift event (2.8–4.2 kbar) at 1.69–1.67 Ga as this is thetime of deposition of the Sundown Group (Page et al.,2000a), and the Round Hill pelites are located acrossthe boundary of the underlying Broken Hill Group(RHA59, RHA51) and Sundown (RHA58) Groups.Other possible scenarios for the early evolution of theBroken Hill Block could be that the M1/D1 eventoccurred closer to the time of the M2/D2 eventc. 1.6 Ga as shown in Fig. 14(b). In these cases, theSundown Group and parts of the Paragon Groupwould have already been deposited at the time of theM1/D1 event. The M1/D1 event is again interpreted asbeing extensional to account for the high geothermalgradient (Fig. 14b). A rift event c. 1.69–1.67 Ga is alsoincluded to account for basin evolution duringdeposition of the Willyama Supergroup (Fig. 14b)(Stevens et al., 1988). The M2/D2 event is also againinterpreted as being either extensional (Fig. 14bi) orcontractional (Fig. 14bii), and both interpretations areable to account for the temperatures and thermalgradient variations of the prograde path in ways sim-ilar to those described for the scenarios shown inFig. 14(a). The implications of M2/D2 extension fol-lowing rifting just prior this event are that the earlyevolution of the Broken Hill Block involved threetransient episodes of extension separated by periods ofsag-phase sedimentation and lithospheric thermalrelaxation (Fig. 14bi). Only two episodes of transientextension are implied if the M2/D2 event involvedthrusting (Fig. 14bii). In both cases, sag-phase sedi-mentation is terminated by the M2/D2 event c. 1.6 Ga;however, the nature of this event is unconstrained.

The M2 event is linked with high-temperature shearzones and development of the pervasive S2 fabricwithin the Round Hill Area. Determining whetherthese shear zones accommodated crustal thinning orthickening during the D2 event is difficult, as theprolonged and intense tectonothermal history of theBroken Hill terrane following peak metamorphism hasobscured evidence of its kinematic significance. High-temperature, lithology-parallel shear zones have beendescribed in other areas of the Broken Hill Block, andhave been placed in extensional (e.g. Noble & Lister,2001; Gibson & Nutman, 2004) and shortening (e.g.



White et al., 1995; Forbes & Betts, 2004) regimes. Thedistinguishing characteristic between the shear zones isthat deformation related to shear zone activity in ashortening regime is associated with strain partitioningand recognizable folding attributed to the OlarianOrogeny. Shear zone activity at granulite facies con-ditions within the Round Hill Area is associated withdevelopment of the S2 fabric, which is folded by thefirst shortening event (D3) recognized in the area, andis interpreted to represent the early stage of the OlarianOrogeny. This lends towards the high-temperatureshear zone being similar to other shear zones of anextensional origin, however does not unequivocallydemonstrate it.

CONCLUSIONS

Two tectonothermal events, M1/D1 and M2/D2occurred in the Broken Hill Block prior to the OlarianOrogeny. The M1/D1 event attained conditions of atleast c. 600 �C at pressure of c. 2.8–4.2 kbar, andinvolved heating to attain a raised geothermal gradient(c. 41–61 �C km)1). The M2/D2 event occurred atconditions of at least 740 �C at c. 5 kbar, and wasassociated with peak granulite facies metamorphism.These events were separated by an episode of burial ofc. 0.8–2.2 kbar, depending on the original pressureconditions of the M1 event. The overall P–T path forthe Broken Hill Block is anticlockwise.

The M1/D1 event is interpreted to be associatedwith rifting at either c. 1.69–1.67 Ga, during whichtime the Willyama Supergroup was deposited into anevolving basin, or at c. 1.64–1.61 Ga just prior to theM2/D2 event. The M1/D1 event was followed by aperiod of lithospheric thermal relaxation and burialduring sag-phase sedimentation (Paragon Group). TheM2/D2 event occurred c. 1.6 Ga, and involved devel-opment of lithology-parallel high-temperature shearzones and a regionally pervasive S2 fabric. Whetherthese shear zones were active within a crustal thinningor thickening environment remains unresolved. TheD2 event was terminated by the onset of crustalshortening during the Olarian Orogeny.

ACKNOWLEDGEMENTS

This paper is published with permission of the CEO,pmd*CRC. B. Chappell at GEMOC is thanked for theXRF analysis, and A. Priymak at the University ofMelbourne for helping with electron microprobe ana-lysis. We are grateful to S. Boger, R. White andK. Evans for their assistance with THERMOCALCand in constructing the pseudosections and to D. Giles,M. Noble, R. Anczkiewicz and M. Raetz for theirhighly helpful discussion pertaining to the petrologyand tectonic interpretations. J. Clulow is thanked forhis assistance in the field. Constructive reviews byJ.-M. Huizenga and M. Rubenach, and the editorialhandling of D. Robinson are much appreciated.

REFERENCES

Barton, M. D. & Hanson, R. B., 1989. Magmatism and thedevelopment of low-pressure metamorphic belts: implicationsfrom the western United States and thermal modelling. Geo-logical Society of America Bulletin, 101, 1051–1065.

Bell, T. H. & Rubenach, M. J., 1983. Sequential porphyroblastgrowth and crenulation cleavage development during pro-gressive deformation. Tectonophysics, 92, 171–194.

Betts, P. G., Lister, G. S. & O’Dea, M. G., 1998. Asymmetricextension of the Middle Proterozoic lithosphere, Mount Isaterrane, Queensland, Australia. Tectonophysics, 296, 293–316.

Binns, R. A., 1964. Zones of progressive regional metamorphismin the Willyama Complex, Broken Hill District, New SouthWales. Journal of the Geological Society of Australia, 11, 283–330.

Boger, S. D. & White, R. W., 2003. The metamorphic evolutionof metapelitic granulites from Radok Lake, northern PrinceCharles Mountains, east Antarctica; evidence for an anti-clockwise P–T path. Journal of Metamorphic Geology, 21,285–298.

Bohlen, S. R., 1987. Pressure-temperature time paths and atectonic model for the evolution of granulites. Journal ofGeology, 95, 617–632.

Bohlen, S. R., 1991. On the formation of granulites. Journal ofMetamorphic Geology, 9, 223–229.

Bohlen, S. R. & Mezger, K., 1989. Origin of granulite terranesand the formation of the lowermost continental crust. Science,244, 326–329.

Bradley, G. M., 1980. Mount Gipps 1:25,000 Geological Sheet7234 III S. Geological Survey of New South Wales, Sydney.

Chamberlain, C. P. & Sonder, L. J., 1990. Heat-producing ele-ments and the thermal and baric patterns of metamorphicbelts. Science, 250, 763–769.

Clarke, G. L., Aitchison, J. C. & Cluzel, D., 1997. Eclogites andblueschists of the Pam Peninsula, NE New Caledonia: areappraisal. Journal of Petrology, 38, 843–876.

Cobbold, P. R., 1977. Description and origin of banded defor-mation structures. I. Regional strain, local perturbations, anddeformation bands. Canadian Journal of Earth Sciences, 14,1721–1731.

Collins, W. J. & Vernon, R. H., 1991. Orogeny associated withanticlockwise P–T-t paths: evidence from low-P, high-Tmetamorphic terranes in the Arunta inlier, central Australia.Geology, 19, 835–838.

Daniel, C. G. & Spear, F. S., 1998. Three-dimensional pat-terns of garnet nucleation and growth. Geology, 26,503–506.

De Yoreo, J. J., Lux, D. R., Guidotti, C. V., Decker, E. R. &Osberg, P. H., 1989. The Acadian thermal history of westernMaine. Journal of Metamorphic Geology, 7, 169–190.

Foden, J., Song, S. H., Turner, S. et al., 2002. Geochemicalevolution of lithospheric mantle beneath S.E. South Australia.Chemical Geology, 182, 663–695.

Forbes, C. J. & Betts, P. G., 2004. Development of type 2 foldinterference patterns in the Broken Hill Block: implicationsfor strain partitioning across a detachment during theOlarian Orogeny. Australian Journal of Earth Sciences, 51,173–188.

Forbes, C. J., Betts, P. G. & Lister, G. S., 2004. Synchronousdevelopment of type 2 and type 3 fold interference patterns:evidence for recumbent sheath folds in the Allendale Area,Broken Hill, NSW, Australia. Journal of Structural Geology,26, 113–126.

Gibson, G. M. & Nutman, A. P., 2004. Detachment faulting andbimodal magmatism in the Palaeoproterozoic WillyamaSupergroup, south-central Australia: keys to recognition of amultiply deformed Precambrian metamorphic core complex.Journal of the Geological Society, London, 161, 55–66.

Giles, D., Betts, P. G. & Lister, G. S., 2002. Far-field continentalbackarc setting for the 1.80–1.67 Ga basins of northeasternAustralia. Geology, 30, 823–826.



Giles, D., Betts, P. G. & Lister, G. S., 2004. 1.8–1.5-Ga linksbetween the North and South Australian Cratons and theEarly-Middle Proterozoic configuration of Australia. Tecton-ophysics, 380, 27–41.

Guiraud, M., Holland, T. & Powell, R., 1990. Calculated mi-neral equilibria in the greenschist-blueschist-eclogite facies inNa2O-FeO-MgO-Al2O3-SiO2-H2O. Contributions to Miner-alogy and Petrology, 104, 85–98.

Hand, M., Dirks, P. H. G. M., Powell, R. & Buick, I. S., 1992.How well established is isobaric cooling in Proterozoic oro-genic belts? An example from the Arunta inlier, central Aus-tralia. Geology, 20, 649–652.

Harrison, T. M. & McDougall, I., 1981. Excess 40Ar in meta-morphic rocks from Broken Hill, New South Wales:implications for 40Ar/39Ar age spectra and the thermal historyof the region. Earth and Planetary Science Letters, 55, 123–149.

Hensen, B. J., 1971. Theoretical phase relations involving cor-dierite and garnet in the system MgO-FeO-Al2O3-SiO2. Con-tributions to Mineralogy and Petrology, 33, 191–214.

Hobbs, B. E., Archibald, N. J., Etheridge, M. A. & Wall, V. J.,1984. Tectonic history of the Broken Hill Block, Australia. In:Precambrian Tectonics Illustrated (eds Kroner, A. & Greiling,R.), pp. 353–368. E. Schweizerbartsche Verlagbuchhandl,Stuttgart.

Kamber, B. S., Frei, R. & Gibb, A. J., 1998. Pitfalls and newapproaches in granulite chronometry. An example from theLimpopo Belt, Zimbabwe. Precambrian Research, 91, 269–285.

Lister, G. S., Etheridge, M. A. & Symonds, P. A., 1991.Detachment models for the formation of passive continentalmargins. Tectonics, 10, 1038–1064.

Lux, D. R., DeYoreo, J. J., Guldotti, C. V. & Decker, E. R.,1986. Role of plutonism in low-pressure metamorphic beltformation. Nature, 323, 794–797.

Marjoribanks, R. W., Rutland, R. W. R., Glen, R. A. & Laing,W. P., 1980. The structure and tectonic evolution of theBroken Hill Region, Australia. Precambrian Research, 13,209–240.

Marmo, B. A., Clarke, G. L. & Powell, R., 2002. Fractionationof bulk rock composition due to porphyroblast growth: effectson eclogite facies mineral equilibria, Pam Peninsula, NewCaledonia. Journal of Metamorphic Geology, 20, 151–165.

McLaren, S., Sandiford, M. & Hand, M., 1999. High radiogenicheat-producing granites and metamorphism – an examplefrom the western Mount Isa Inlier, Australia. Geology, 27,679–682.

Mengel, K., Richter, M. & Johannes, W., 2001. Leucosome-forming small-scale geochemical processes in the metapeliticmigmatites of the Turku area, Finland. Lithos, 56, 47–73.

Morgan, P. & Ramberg, I. B., 1987. Physical changes in thelithosphere associated with thermal relaxation after rifting.Tectonophysics, 143, 1–11.

Muller, W., 2003. Strengthening the link between geochronolo-gy, textures and petrology. Earth and Planetary Science Let-ters, 206, 237–251.

Noble, M., 2000. Geology of the Broken Hill Synform. MScthesis, Monash University, Melbourne.

Noble, M. P. & Lister, G. S., 2001. Discovery of a km-scale earlyextensional shear zone that deformed the Broken Hill Pb-Zn-Ag orebody, NSW, Australia. Geological Society of AustraliaAbstracts, 64, 127–128.

Page, R. W. & Laing, W. P., 1992. Felsic metavolcanic rocksrelated to the Broken Hill Pb-Zn-Ag orebody, Australia:geology, depositional age, and timing of high-grade meta-morphism. Economic Geology, 87, 2138–2168.

Page, R. W. & Sweet, I. P., 1998. Geochronology of basin phasesin the western Mt Isa Inlier, and correlation with the McAr-thur Basin. Australian Journal of Earth Sciences, 45, 219–232.

Page, R. W., Stevens, B. P. J., Gibson, G. M. & Conor, C. H. H.,2000a. Geochronology of Willyama Supergroup rocks be-tween Olary and Broken Hill, and comparison to Northern

Australia. Australian Geological Survey Organisation Record,10, 72–75.

Page, R. W., Jackson, M. J. & Krassay, A. A., 2000b. Con-straining sequence stratigraphy in north Australian basins:SHRIMP U-Pb zircon geochronology between Mt Isa andMcArthur River. Australian Journal of Earth Sciences, 47,431–459.

Phillips, G. N., 1980. Water activity changes across anamphibolite-granulite facies transition, Broken Hill, Australia.Contributions to Mineralogy and Petrology, 75, 377–386.

Phillips, G. N. & Wall, V. J., 1981. Evaluation of prograderegional metamorphic conditions; their implications for theheat source and water activity during metamorphism in theWillyama Complex, Broken Hill, Australia. Bulletin de Min-eralogie, 104, 801–810.

Powell, R. & Downes, J., 1990. Garnet porphyroblast-bearingleucosomes in metapelites: mechanisms, phase diagrams, andan example from Broken Hill, Australia. In: High-tempera-ture Metamorphism and Crustal Anatexis (eds Ashworth,J. R. & Brown, M.), pp. 105–123. Unwin Hyman Ltd,London.

Powell, R. & Holland, T. J. B., 1988. An internally consistentthermodynamic dataset with uncertainties and corrections: 3.Applications, methods, worked examples and a computerprogram. Journal of Metamorphic Geology, 6, 173–204.

Powell, C.M., Preiss,W.V.,Gatehouse, C.G.,Krapez, B.&Li, Z.X., 1994. South Australian record of a Rodinian epicontinentalbasin and its mid-Neoproterozoic break-up (c. 700 Ma) to formthe Palaeo-Pacific Ocean. Tectonophysics, 237, 113–140.

Powell, R., Holland, T. & Worley, B., 1998. Calculating phasediagrams involving solid solutions via non-linear equations,with examples using THERMOCALC. Journal of Meta-morphic Geology, 16, 577–588.

Preiss, W., 2000a. Overview of the Curnamona Province and itstectonic setting. Australian Geological Survey OrganisationRecord, 10, 79–81.

Priess, W. V., 2000b. The Adelaide Geosyncline of South Aus-tralia and its significance in Neoproterozoic continentalreconstruction. Precambrian Research, 100, 21–63.

Raetz, M., Krabbendam, M. & Donaghy, A. G., 2002. Compi-lation of U-Pb zircon data from the Willyama Supergroup,Broken Hill region, Australia: evidence for three tectonos-tratigraphic successions and four magmatic events? AustralianJournal of Earth Sciences, 49, 965–983.

Sandiford, M. & Hand, M., 1998. Australian Proterozoic high-temperature, low-pressure metamorphism in the conductivelimit. In: What Drives Metamorphism and Metamorphic Rela-tions? Special Publications 138, (eds Treloar, P. J. & O’Brien,P. J.), pp. 109–120, Geological Society, London.

Sandiford, M., Hand, M. & McLaren, S., 1998. High geothermalgradient metamorphism during thermal subsidence. Earth andPlanetary Science Letters, 163, 149–165.

Southgate, P. N., Bradshaw, B. E., Domagala, J., et al. 2000.Chornostratigraphic basin framework for Palaeoproterozoicrocks (1730–1575 Ma) in northern Australia and implicationsfor base-metal mineralisation. Australian Journal of EarthSciences, 47, 461–483.

Stevens, B. P. J., Barnes, R. G., Brown, R. E., Stroud, W. J. &Willis, I. L., 1988. The Willyama Supergroup in the BrokenHill and Euriowie Blocks, New South Wales. PrecambrianResearch, 40/41, 297–327.

Stuwe, K., 1997. Effective bulk composition changes due tocooling: a model predicting complexities in retrograde reactiontextures. Journal of Petrology, 129, 43–52.

Stuwe, K. & Ehlers, K., 1997. Multiple metamorphic events atBroken Hill, Australia. Evidence from Chloritoid-bearingparagenesis in the Nine-Mile Mine Region. Journal of Pet-rology, 38, 1167–1186.

Swapp, S. M. & Frost, B. R., 2003. Evidence for high-pressuremetamorphism in the granulites of the Broken Hill area.Geoscience Australia Record, 13, 174–175.



Van der Pluijm, B. A. & Marshak, S., 1997. Earth Structure: AnIntroduction to Structural Geology and Tectonics. McGraw-Hill, New York.

Vance, D. & Holland, T., 1993. A detailed isotopic and petro-logical study of a single garnet from the Gassetts Schist,Vermont. Contributions to Mineralogy and Petrology, 114,101–118.

Venn, C. J., 2001. The geodynamic evolution of the Mount Robeand Mount Franks region, Broken Hill Australia: Discoveryof crustal-scale extensional shear zones and giant sheath folds.PhD thesis, Monash University, Melbourne.

Vernon, R. H., 1987. Growth and concentration of fibrous sil-limanite related to heterogeneous deformation in K-feldspar-sillimanite metapelites. Journal of Metamorphic Geology, 5,51–68.

Vernon, R. H., 1996. Problems with inferring P–T-t paths inlow-P granulite facies rocks. Journal of Metamorphic Geology,14, 143–153.

White, R. W. & Powell, R., 2002. Melt loss and the preservationof granulite facies mineral assemblages. Journal of Meta-morphic Geology, 20, 621–632.

White, S. H., Rothery, E., Lips, A. L. W. & Barclay, T. J. R.,1995. Broken Hill area, Australia, as a Proterozoic fold andthrust belt: implications for the Broken Hill base-metaldeposit. Transactions of the Institution of Mining andMetallurgy. Section B, Applied Earth Sciences, 104, B1–B17.

White, R. W., Powell, R. & Holland, T. J. B., 2001. Calculationof partial melting equilibria in the system Na2O-CaO-K2O-FeO-MgO-Al2O3-SiO2-H2O (NCKFMASH). Journal ofMetamorphic Geology, 19, 139–153.

White, R. W., Powell, R. & Halpin, J. A., 2004. Spatially-focussed melt formation in aluminous metapelites fromBroken Hill, Australia. Journal of Metamorphic Geology, 22,825–845.

Willis, I. L., Brown, R. E., Stroud, W. J. & Stevens, B. P. J.,1983. The Early Proterozoic Willyama Supergroup: strati-graphic subdivision and interpretation of high to low-grademetamorphic rocks in the Broken Hill Block, New SouthWales. Journal of the Geological Society of Australia, 30, 195–224.

Wilson, C. J. L. & Powell, R., 2001. Strain localisation andhigh-grade metamorphism at Broken Hill, Australia: aview from the Southern Cross area. Tectonophysics, 335,193–210.

Woodsworth, G. J., 1977. Homogenization of zoned garnetsfrom pelitic schists. Canadian Mineralogist, 15, 230–242.

Wyborn, L. A. I., Page, R. W. & McCulloch, M. T., 1988.Petrology, geochronology and isotope geochemistry of thepost-1820 Ma granites of the Mount Isa Inlier: mechanismsfor the generation of Proterozoic anorogenic granites. Pre-cambrian Research, 40/41, 509–541.

Yardley, B. W. D., 1977. The nature and significance of themechanism of sillimanite growth in the Connemara Schists,Ireland. Contributions to Mineralogy and Petrology, 65,53–58.

Yardley, B. W. D., 1989. An Introduction to MetamorphicPetrology. Longman Group Limited, Essex, UK.

Received 4 April 2005; revision accepted 10 August 2005.



A structural metamorphic study of the Broken Hill Block...

Documents

Transcript of A structural metamorphic study of the Broken Hill Block...