JOURNAL AND PROCEEDINGS OF THE ROYAL SOCIETY | OF NEW SOUTH WALES Volume 140 Parts 3 and 4 (Nos 425-426) 2007 ISSN 0035-9173 PUBLISHED BY THE SOCIETY BUILDING H47 UNIVERSITY OF SYDNEY, NSW 2006 Issued December 2007 THE ROYAL SOCIETY OF NEW SOUTH WALES OFFICE BEARERS FOR 2007-2008 Patrons His Excellency, Major General Michael Jeffery AC CVO MC, Governor General of the Commonwealth of Australia. Her Excellency Professor Marie Bashir AC CVO, Governor of New South Wales. President Mr J.R. Hardie, BSc Syd, FGS, MACE Vice Presidents Prof. J. Kelly, BSc Syd, PhD Reading, DSc NSW, FAIP, FInstP Mr C.M. Wilmot one vacancy Hon. Secretary (Gen.) . Mr A.J. Buttenshaw (from Nov. 2007) Hon. Secretary (Ed.) Prof. P.A. Williams, BA (Hons), PhD Macq. Hon. Treasurer Ms M. Haire BSc, Dip Ed. Hon. Librarian Ms C. van der Leeuw Councillors Mr A.J. Buttenshaw Mr. T. Danos Mr J. Franklin Em. Prof. H. Hora Dr M. Lake, PhD Syd Ms Jill Rowling BE UTS, MSc Syd A/Prof. W.A. Sewell, MB, BS, BSc Syd, PhD Melb FRCPA Ms R. Stutchbury three vacancies Southern Highlands Rep. vacant The Society originated in the year 1821 as the Philosophical Society of Australasia. Its main function is the promotion of Science by: publishing results of scientific investigations in its Journal and Proceedings; conducting monthly meetings; awarding prizes and medals; and by liason with other scientific societies. Special meetings are held for: the Pollock Memorial Lecture in Physics and Mathematics, the Liversidge Research Lecture in Chemistry, the Clarke Memorial Lecture in Geology, Zoology and Botany, and the Poggendorf Lecture in Agricultural Science. Membership is open to any person whose application is acceptable to the Society. Subscriptions for the Journal are also accepted. The Society welcomes, from members and non-members, manuscripts of research and review articles in all branches of science, art, literature and philosophy for publication in the Journal and Proceedings. ISSN 0035-9173 Copyright The Royal Society of New South Wales does not require authors to transfer their copyright. Authors are free to re-use their paper in any of their future printed work and can post a copy of the published paper on their own web site. Enquiries relating to copyright or reproduction of an article should be directed to the author. Journal & Proceedings of the Royal Society of New South Wales, Vol. 140, p. 27-46, 2007 ISSN 0035-9173/07/020027—20 $4.00/1 Abstract: Drilling near Mount Oxley and Mullagalah in the Bourke-Byrock area, NSW, intersected basaltic breccia pipes. ‘The Mount Oxley basalt is an unusual hybrid rock involving intimate veining and intermingling between a slightly evolved basanite and a strongly evolved, late-stage, baryte-bearing trachyte. ‘The basanite consists of abundant phenocrysts of altered olivine and diopside-augite, and rarer phenocrysts of nepheline, anorthoclase and Ti-rich magnetite, in a groundmass of plagioclase laths and ‘Ti-rich magnetite grains. ‘The trachytic component is dominated by alkali feldspar, largely sanidine, with calcic amphibole (Ti-Mg-rich hastingsite), baryte (with up to 2% Sr in coarser crystals) and secondary carbonates. Olivine-microgabbro and microsyenite xenoliths in the basalt suggest that the cumulates were formed from both the basanitic and trachytic magmas prior to emplacement. Xenoliths of, and xenocrysts from, high pressure ultramafic metamorphic assemblages (spinel harzburgite and spinel websterite) indicate a mantle source for the basanitic magma. ‘I'wo- pyroxene temperatures based on Wells thermometry suggest these ultramafic assemblages were re-equilibrated under an ambient paleogeotherm between 990—1035°C. Similar basalt appears in the Mullagalah breccia pipe, but lacks the phenocrystic nepheline and the hybrid baryte-bearing trachyitic component found in the Mount Oxley basalt. Xenoliths in the Mullagalah breccia include a cumulate-like olivine-bearing diopside-amphibole (K-'Ti-rich ferroan pargasite) assemblage. The Mount Oxley and Mullagalah intrusions are not well dated, but were probably formed during Late Mesozoic-Late Cenozoic intraplate basaltic activity that occurred in eastern Australia, from magmas generated at mantle depths exceeding 38 km. Keywords: Breccia pipe, baryte, basalt, trachyte, xenocrysts, xenoliths INTRODUCTION Cenozoic to Mesozoic intraplate basaltic erup- tives are a feature in eastern Australia (John- son 1989, Sutherland 2003). In New South Wales, the most westerly Cenozoic basalts give way to Miocene potassic, leucite-bearing lavas and minor trachytes, which form a lin- ear, apparently age-progressive trail extend- ing from Byrock, NSW, in the north to Cos- grove, Victoria, in the south (Cundari 1973, Byrnes 1993, Zhang and O’Reilly 1997, Paul et al. 2005, McQueen et al. 2007). Mesozoic intraplate basalts and silicic derivatives also outcrop among the western Cenozoic basalts and are plentiful in the Gunnedah and Dubbo areas (Tadros 1993, Meakin and Morgan 1999). This paper describes an unusual hybrid basaltic intrusion found in this western zone, which was encountered in drill cores near Mount Oxley, east of Bourke (Byrnes 1993). Although this intrusion received petrographic examination, neither nepheline nor baryte were reported. The significance of these minerals is discussed in this study. Another basaltic diatreme drilled nearby at Mullagalah is also described and is compared with the Mount Oxley intrusion. The geological setting in the Bourke-Byrock area (Figure 1) forms part of the Palaeozoic Lachlan Orogen, with exposures of folded Or- dovician to Devonian sedimentary and volcanic rocks, emplacements of serpentinite and a range of mineralisation (Byrnes 1993). Late Cenozoic leucititic bodies outcrop at Byrock were dated at ~17Ma (K-Ar dating, Sutherland 1985). Leucititic lavas of similar age outcrop farther south at El Capitan, near Cobar (Ar-Ar dating, McQueen et al. 2007). Minor fresh tholeiitic basalt was also drilled 25km west of Byrock, but is so far undated (Byrnes 1993). Surficial Cenozoic sedimentary beds, wind blown sands and alluvium from the present Darling, Bogan and Warrego drainages partly obscure bedrock, particularly along the boundary of the Palaeo- zoic basement rocks and the Mesozoic Surat Basin beds to the north (Figure 1). 28 SUTHERLAND et al. 146° 30°'E A \ JMULLIGALAH ™ \ bt O BYROCK NEW SOUTH WALES ® Cenozoic Leucitite | | | ae Basaltic Breccia \ @ ‘ k @ Mz Mesozoic Basin EL CAPITAN aS Boundary | Pz Palaeozoic Basement | F-.—F Fault st wr Drainage Barrier Highway —s a Major Roads 10 20 30 40 50 Figure 1. Locality map Bourke-Byrock area, New South Wales, showing locations of the Mt Oxley and Mullagalah basaltic diatremes (drill cores) in relation to the Paleozoic basement rocks, Mesozoic basin boundary, outcrops of Cenozoic leucitites and the main drainages and major roads. Inset shows area within New South Wales. UNUSUAL BARYTE-BEARING HYBRID BASALT 29 Samples of the Mount Oxley and Mullagalah basalts and associated intrusive breccias were made available to the authors as fragments, thin sections and polished sections, together with some initial petrographic reports and elec- tron microprobe (EMP) analyses of mineral phases. These materials were donated to the Australian Museum, Sydney, for further study by Dr Ian Plimer, then at North Broken Hill Ltd, and Ian Matthais, CRA Cobar office, 1983. The Mount Oxley samples (~ 419500 mE, 6671100mN) came from a hole (T16) drilled by North Broken Hill-Preussag Australia while investigating geochemical prospects and aero- magnetic anomalies at ~ 146.16°E and 30.08°S (Preussag Australia Pty Ltd & North Broken Hill Ltd 1981a,b, North Broken Hill & Preussag Australia Pty Ltd 1982). The Mullagalah drill core sample (MUI 647) came from a drilling depth of 647 feet within a magnetic anomaly formed by an inclined diatreme west of Mullagalah, at ~146.17°E and 30.44°S (North Broken Hill Ltd 1971 a,b). Initial mineral analyses for these company in- vestigations were undertaken by the Geology Department of the University of Melbourne. Further microscopic studies and mineral analyses were made on the donated samples using facilities at the School of Earth Sciences, Macquarie University and CSIRO Division of Exploration Geoscience, North Ryde, in 1989, and at the School of Natural Sciences, Univer- sity of Western Sydney, Parramatta, in 2006 to complete the study. Both the Mount Ox- ley and Mullagalah intrusive basalt breccias contain numerous xenoliths and xenocrysts of the local country rocks and deeper crustal and mantle lithologies which are also described here. The breccias lie southwest of the Bundabulla- Bokhara basalt fields which includes buried alkaline intrusives and rare outcrops (Madden 1999, Jaques 2006 and pers. comm. 2007). MATERIALS Mount Oxley Thin sections cut from the Mount Oxley core are basalt and basalt-intruded breccias (T16, 1-3; S89; 4C3/5). In the breccias, fragments of quartzite and arkose commonly reach up to 9mm across. ‘The quartz grains in these rocks are highly strained, partly recrystallised and show sutured grain boundaries. Some quartzite fragments show the effects of partial melting where altered brown glassy material is inter- spersed through the matrix, possibly as mini- mum melts of K-feldspar. The contact between the breccia and invading basalt is commonly marked by a coarser crystallised feldspathic selvage. The fine grained basalt is porphyrytic, and contains rare sub-euhedral nepheline phe- nocrysts up to 3mm long (Figure 2). Abundant phenocrysts and glomerocrysts of altered olivine and fresh clinopyroxene, up to 2mm across, occur with sporadic microphenocrysts of opaque iron oxide and alkali feldspar (Figure 3.1). The olivine is replaced by carbonate and clay. Eu- hedral clinopyroxenes are sometimes partially resorbed along their margins and exhibit strong core to rim and sector growth zoning. The basalt groundmass ranges from chilled glassy material near its outer contacts with the breccia and grades into a fine grained matrix containing small plagioclase laths and opaque iron oxide grains. eure 2: Noein here (2, in Tete 1.7mm wide) in basanite, with squarish in- clusion of anorthoclase, with reacted margins, within the phenocryst bottom right. Back scattered SEM image. Note bright crystals in the basanite host represent opaque Fe-oxides (Ti-rich magnetite). Photomicrograph: Adam McKinnon. 30 SUTHERLAND et al. Trachytic pools, veins and_ segregations inter-finger with the basalt and represent a late phase of crystallisation (Figure 3.2). Some veins form mosaics of alkali feldspar and in places have broken up and become incorporated into the host basalt. These trachytic veins can reach several centmetres in length and up to a cen- timetre in width. They grade into amphibole- baryte-carbonate-bearing trachyte that occu- pies pockets throughout the host basalt, and sometimes invades xenoliths within the host rock. This produces baryte-replacement of pyroxenes in pyroxenites (Figure 4), with baryte surrounding small cores of secondary quartz in some replacements. The last formed veins cut across all other structures in the rocks and represent fracture fillings containing secondary carbonates, clays and iron hydroxides. Figure 3. Mount Oxley intrusion petrology (T16). 3.1: Anorthoclase microphenocryst (centre) in basanite. Note altered olivine, rare clinopyroxene grains amid plagioclase laths and opaque Fe-oxide grains; 3.2: Pool of baryte-bearing trachyte (lighter coloured, coarser material, centre- right) interacting with basanite host (darker, finer-grained material, left); 3.3: Close up of trachyte pod, showing clinopyroxene prisms (strong relief), scattered amphibole (irregular grayish grains) and rare opaque iron oxide grains in an abundant alkali feldspar-baryte-bearing matrix; 3.4: Reaction zone between baryte-bearing trachyte (lighter material, centre-right) and microgabbro xenolith (grayish material, top left), marked by bladed ilmenite (opaque crystals). Amphibole forms scattered irregular intergrowths. Photomicrographs (PPL): Ian Graham. UNUSUAL BARYTE-BEARING HYBRID BASALT 31 Medium to coarse grained xenoliths up to 3cm across and associated disaggregated xenocrysts are scattered liberally throughout the host basalt. In addition to local crustal materials, there are olivine microgabbro (Fig- ure 5.1) and microsyenite assemblages. ‘These textures suggest that they represent cumulates related to the phenocrystic basalt and its late- stage trachytic component. The microgabbro xenoliths show reaction margins against the tra- chytic component, which partly replaces olivine (Figure 5.2). Ultramafic metamorphic assem- blages among the xenoliths are spinel meta- harzburgite (Figure 5.3), in which olivine is altered to mesh-textured carbonates, talc and clays (Figure 5.4), and spinel meta-websterite. The xenocrysts and composites (derived from the xenolith assemblages) typically show reac- tions with the host magmas, such as incipient melting and initial crystallisation of secondary minerals (Figure 6.1), feldspathic segregation mantles (Figure 6.2), opaque Fe-oxide reaction rims (Figure 6.3), and strong resorption (Fig- ure 6.4). Mullagalah The Mullagalah drill core sample (MU1 647) from a depth of 647 ft (190 m) is an unsorted vol- canic breccia (Figure 7.1). It contains volcanic lithic fragments of diverse, but related, basalts that show various degrees of crystallinity and degradation due to oxidation, as well as abun- dant xenoliths and xenocrysts of crustal and probable mantle origin. Fragments up to 2cm across make up 55% of the rock embedded in a pale yellow smectite clay matrix. Basalt forms about 60% of the fragments. The largest of these contains 15% of subhedral altered olivine (up to 1.5mm across) and sporadic clinopyrox- ene microphenocrysts (up to 0.2mm across), some showing hour glass and core to rim zon- ing (Figure 7.2). The groundmass shows flow banding and strongly chilled broken margins containing up to 40% of flow-aligned prismatic clinopyroxene microlites (Figure 7.3), set in a clay-degraded, in part carbonated and once vesicular volcanic glass. The matrix is charged with minute opaque oxide grains (< 0.02mm). Olivine phenocrysts are altered to patchy car- bonate, talc, and a yellow to brown smectite clay, which also pervades the glassy matrix. Some basaltic fragments contain prominent, eu- hedral diopside crystals, with simple and knee- shaped twins and partly resorbed margins and interiors. Sparse, altered olivine euhedra reach 1.5mm. The groundmass contains flow-aligned, wispy plagioclase microlites up to 0.05 mm long, and some cases sections or fragments contain abundant vesicular remnants up to 0.04mm in yellow brown iron-stained glass. Figure 4. Enstatite-rich meta-websterite, showing invasion of Ba-rich trachyte melt into xenolith. Note replacement of pyrox- enes with baryte (bright white patches). The less intense white grains in the host basanite represent ‘Ti-rich magnetite- ulvospinel grains in the groundmass. Large scale bar = 100 um. Back scattered SEM image: K. Kinealy. 32 SUTHERLAND et al. The breccia contains a range of xenocrysts (20%) mostly less than 1.5mm in size. These are cleavage fragments of diopside, pale yellow to orange brown, pleochroic amphibole, yellow- brown Cr-bearing spinel, green spinel; rare opaque Fe-oxides, carbonate-altered olivine, quartz, rare degraded plagioclase and oxidised biotite flakes up to 1mm long. Some quartz exhibits reactions rims of prismatic diopside. A diopside-amphibole composite 2.5mm across encloses a small altered olivine grain and may represents a cognate source for xeno-crystal material (Figure 7.4). Subrounded crustal xeno- liths include partly recrystallised microsyenite, foliated feldspathic microbreccia, foliated sandy and silty claystone, deformed and finely recrys- tallised mircoporphyritic felsic volcanic mate- rial, and recrystallised quartzite or vein quartz. The Mullagalah basalts differ from that at Mount Oxley in lacking significant groundmass plagioclase, phenocrystic nepheline or a late- stage baryte-bearing trachytic phase. Figure 5. Mount Oxley xenolith petrology (T16). 5.1: Cumulate-textured micrograbbo, showing euhedral diopside intergrown with large plagioclase crystals, with altered olivine in interstitial grains and inclusions in diopside; 5.2: Altered, reacted olivine grain (centre) in cumulate-textured microgabbro. Note original grain is marked by opaque iron oxide rims, a prominent reaction zone of alkali feldspar laths, intersertal amphibole, partly surrounding a smectitic clay altered core, with coarser trachytic material, bottom right; 5.3: Spinel meta-harzburgite xenolith in basanite host (dark material, left), showing spinel (grey crystal, centre) intergrown with orthopyroxene (lighter crystals, centre top and bottom) and altered olivine (mesh-textured material, centre bottom and right side); 5.4: Meta-harzburgite showing altered olivine, with magnesite replacements (mesh- textured diagonal zone, with mosaic cores), intergrown with orthopyroxene grains (left and right sides). Photomicrographs (PPL): Ian Graham. UNUSUAL BARYTE-BEARING HYBRID BASALT 33 ANALYTICAL TECHNIQUES Because the Mount Oxley basalt is invaded by ubiquitous late-stage baryte-bearing trachytic patches and veins and is riddled with xenocrysts and xenoliths, no bulk chemical analysis of the material was attempted. The mineralogy of the basalts, and a range of xenocryst and xenolith minerals from Mount Oxley and Mul- lagalah were investigated using a Leica DMLP polarising microscope. Electron microprobe (EMP) analysis, supplemented with back scat- tered electron (BSE) imaging confirmed the mineralogy. Because several EMP facilities were used over a protracted time, the analytical results come from several sources and operators. These include a JEOL probe fitted with wave- length dispersive spectrometers (WDS) at the University of Melbourne (I.R. Plimer and D. Sewell analysts), an automated ETEC probe at Macquarie University, North Ryde (B.J. Barron analyst), a CAMECA system in the Division of Exploration Geoscience, CSIRO, North Ryde (F.L. Sutherland and K. Kinealy analysts)and a JXA super probe with 3 WDS detectors at the School of Natural Sciences, University of Western Sydney, North Parramatta (A.R. McKinnon and F.L. Sutherland analysts). Figure 6. Mount Oxley xenocryst minerals (T16). 6.1: Plagioclase composite xenocryst (centre) in basanite (PPL), showing incipient melting, marked by patchy opaque Fe oxide dustings and invasions along grain boundaries. Note altered olivine phenocryst (bottom right); 6.2: Enstatite xenocryst (dark coloured core) with plagioclase-rich reaction zone (lighter rim) in basanite (XPL); 6.3: Spinel xenocryst (light coloured core) with prominent Fe-oxide reaction zone (opaque rim) in basanite (PPL); 6.4: Resorbed Ti-rich magnetite micro xenocryst (crystal is 1.1mm long) with more T-rich reaction rims, in basanite. Photomicrographs: Ian Graham; Back scattered SEM image: A. McKinnon 34 SUTHERLAND et al. Operating procedures and conditions for the EMP analyses mostly utilised a 15kV acceler- ation voltage, beam current of 20nA, Bence- Albee corrections and natural mineral stan- dards. Analytical precision was usually within 1% for elements above 10 wt% as oxides, within + 5% at 1-10 wt% oxides and within + 10% at < 1 wt% oxides. Because several different instruments provided EMP analyses incorpo- rating a range of variations in base line drift during the runs, as well as potential differences related to surface polish in the samples and operator factors, the results are presented here as normalised 100% totals. Most results repre- sent anhydrous minerals, while the few hydrous minerals were normalised by assigning a volatile content equivalent to the difference from 100% in the total. The accuracy of the representative analyses listed in Tables 1 to 5 can be judged by the closeness of the calculated cation totals to the theoretical cation total of the unit cell for the analysed mineral. In most cases, cation totals for the anhydrous minerals fell within 0.0-0.5% of the theoretical cation total, while cation totals for the hydrous minerals were usually within 1% of the theoretical totals. Figure 7. Mullagalah basalt breccia petrology (MUI 647). 7.1: Basalt fragments in breccia. Note microphenocrysts and plagioclase microlites in ground mass (left fragment) and dark glassy matrix (right fragments), with qaurtzitic fragments (top and bottom); 7.2: Contact of basalt fragment with breccia (left). Note large altered olivine phenocryst in the basalt (centre); 7.3: Flow-textured basalt fragment. Note alignment of olivine (altered grain, top right), clinopyroxene, microlitic plagioclase crystals and glassy bands; 7.4: Cumulate-like olivine-bearing amphibole (grey zones) and clinopyroxene (light zones) xenolith within breccia. Photomicrographs (PPL): Ian Graham. UNUSUAL BARYTE-BEARING HYBRID BASALT 39 End members for the pyroxene group min- erals were calculated after the method of Yoder and Tilley (1962) and the computer program of Cebeira (1990). Amphibole compositional names were determined using the computer program of Yavuz (1996). Spinel group end members were calculated from a computer program written by Ross Pogson, Australian Museum. Two-pyroxene temperatures for the co-existing pyroxenes in the ultramafic meta- assemblages used Wells (1977) thermometry which provides the most reliable results for such assemblages (generally within + 50°C at tem- peratures <1300°C; Taylor 1998, Trebaudino and Bruno, 1993). For comparison, Wood and Banno (1973) two-pyroxene temperatures were also calculated and generally yielded results 15— 95°C higher. RESULTS Representative photomicrographs and mineral analyses of the Mount Oxley and Mullagalah basalt breccias are presented in Figures 2—7 and Tables 1-5. Mount Oxley Basalt The phenocryst assemblage is dominated by al- tered olivine and clinopyroxene (Figure 3.1 and Table 1). The clinopyroxene is zoned Al-rich diopside (diso_56 hdig—22 tschzo jdo_s aco_5), with more calcic, Fe- and Ti-rich rims (digo hdag tschog jdg). Sparse opaque oxide phenocrysts belong to a Cr-bearing, Ti-rich magnetite- ulvospinel series (uspg2 mti7 mfg mc7 sp7). Rare large nepheline phenocrysts (Figure 2) are K-bearing (neg ksj4). They include co- existing zoned alkali feldspar, largely anortho- clase (abgg oreg ang), which shows a reaction rim marking the alkali feldspar / nepheline boundary. Alkali feldspar microphenocrysts (Figure 3.1) and twinned, zoned anorthoclase crystals (~abgo_75 Org9—32 aNs5—g) may also represent this phenocrystic phase. The ground- mass contains plagioclase laths of intermediate composition (~ anyg—52 ab4g—49 Ore—3), small ulvospinel grains and rare calcic amphibole grains (alumino-edenite) and variable amounts of glassy matrix. Overall, the mineralogy repre- sents a porphyritic, evolved ‘nepheline’-bearing basanite. The invasive late-stage trachytic crystalli- sations that infiltrate the basalt (Figures 3.2 and 3.3) are dominated by mosaics, tablets and laths of K-rich alkali feldspar, largely sanidine in composition (~ ors; aby ans). In places the al- kali feldspar is inter-grown with clinopyroxene, calcic amphibole (titano-magnesio-hastingsite), interstitial and clustered baryte grains and sec- ondary carbonate fillings. The baryte is Sr- bearing (up to 2%) in coarser crystals (Table 2). Rare isolated green sodic clinopyroxene grains may also belong to this assemblage and classify as hedenbergite, after Yoder and Tilly (1962) (hdgo diez jdio tscho), or as aegirine-augite under IMA nomenclature (Cebeira 1990).The high jd value favours aegirine-augite over typical hedenergite. The last-formed veins that cut across all the basalt-trachyte features, incorporated xenoliths and host breccia are dominated by carbonate minerals including magnesite, Al, Fe-rich ser- pentinitic clay (Wicks and O’Hanley 1988) and nontronitic clay (Gtiven 1988) (Table 2). Mount Oxley Xenoliths / Xenocrysts Xenoliths and xenocrysts in the basalt and breccias represent various lithologies and min- eral compositions. Those of interest to the basalt genesis and mantle source include the cumulate-textured magmatic equivalents and high pressure ultramafic metamorphic-textured xenolithic assemblages. Coarse olivine-clinopyroxene-plagioclase and medium grained olivine-plagioclase cumu- lates (Figure 4.1) contain similar mineralogy to phenocryst and matrix phases in the basalt and probably represent near-cognate early crystallisations formed prior to basalt eruption. Such crystallisations presumably also provided xenocrysts such as a strongly resorbed ulvospinel which shows increased Ti contents in the rim zone (Figure 5.4). Other igneous- textured alkali feldspar-rich xenoliths may represent early dismembered K-rich feldspathic cumulates related to the late-stage feldspathic 36 SUTHERLAND et al. veins which were introduced into and broken up in the host basalt. Ultramafic meta-assemblages in the xeno- liths include spinel meta-websterites (Figure 4) and spinel meta-harzburgite (Figures 5.3 and 5.4). Two types of meta-websterite differ in the range of Mg contents within the phases (Table 3). One type contains Al-rich enstatite (engg—g9 fS9_10 WO1_2), Al-rich augite (digg —67 tsch}g_17 jdg_7 acg—7 hd3_4) and Cr-bearing spinel (sp74_75 Cm 3—14 hcg-9 mt2—3 uspo—}). Meg numbers for the phases range from 0.90 to 0.72. This type is more magnesian and Cr-rich than in the second type, which contains Al-rich enstatite (en7g_77 fs21-22 wo,), Al-rich augite (dis9_60 hdjg_19 tschj2_13 jdg_9 acy_2) and Cr- bearing spinel (spgi_g2 hcg_19 cm7_g mtj_2 uspo—1). Mg numbers for the phases range from 0.81 to 0.60. Despite the different ranges in Mg numbers, the two types give similar re- equilibrated temperature ranges, based on Wells 2-Pyroxene thermometry (990° + 5°C, using Fe?t recalculated T). The Mg-rich spinel harzburgite (Table 4) contains Al-rich enstatite (engg fs9_19 wo,_2), Al-rich diopside (dig5—66 tschj4_15 jJdg_i9 hds_¢ ac4_5) and Cr-bearing spinel (splgi_g2 heg_10 cmg_—7 mti;—2 uspo_1). Mg numbers for these phases range from 0.91 to 0.78. The Wells 2- Pyroxene T estimates (1005 to 1035°C, Fe3+ calculation) are higher than those for the spinel meta-websterite assemblages (985-1000°C). Phenocryst Assemblage, Oxide wt % Mineral Cpx Cpx(core) Cpx (rim) Spl Nep Alk S1iO2 48.71 48.90 46.18 0.72 43.39 68.16 TiO» 0.93 1.65 240 20.88 - - AlpOx 6.92 8.36 7.95 oO sons 17.99 Cr203 0.83 O13 0.01 7 Nr A - - Fe203 2.60 - - 48.02. - - FeO 0.19 6.86 8.95 Li0S 0.31 - MnO 0.18 0.23 0.22 0.65 - - MgO a2 13.02 £1.02 Doo - - CaO 18.94 19.70 227 0.34 0.90 0.63 NagO Ora 1.15 0.76 0.20 17.39 8.09 K20 - 0.01 0.05 0.02 4.68 4.95 NiO 0.14 - - OCA - - Cation Sum 3.999 4.009 4.034 23.912 24.096 19.950 (Oxyge is) 6 6 6 32 a2 32 Cation Ratios% Ca 44.0 Ca 45.6 Ca 49.8 Fe 74.4 Na 83.9 Na 69.2 Mg 42.0 Mg41.9 Mg 34.5 Ti 19.5 K 14.2 K 27.7 Fe 14.0 Fe 12.4 Fe 15.7 Mg 6.0 Ca 1.8 Ca 3.0 Name Augite Diopside Diopside Ulvospinel Nepheline Anorthoclase Mg/(Mg+Fe?* ) 0.815 0.771 0.687 0.254 - - Table 1. Representative normalised EMP analyses, phenocryst phases, basanite host (T16, 4C 3/5). Cpx Clinopyroxene, Spl Spinel, Nep Nepheline, Alk Alkali feldspar. Fe2O3 and FeO by stoichiometry. Oxygen nos. are based on respective unit cells. Cpx (dis6.1, tsch2o.9, hdig.5, ac 5.1); Cpx core (dis0.2, hdai.g, tschig.7, jdg.2) and Cpx rim ( dizg.9, hdgg.7, tscha5.6, jds.7); Spl (usper.c, mti6.7, mfg.o, mc7.1, Sp6.g); Nep (meg3.9, ksj4.2); Alk (abeg9.2, oFa7.7, aNn~3.0). concentrations below the level of detection. Dashes indicate 37 UNUSUAL BARYTE-BEARING HYBRID BASALT ‘(oytuoOIuOU) stg :(,eyUTyUEd Jes, YOU-9J-[Y) Jog :(eyAreq SuLIeaq-1G) Jeg *(@0yosy ‘@Olpf ‘te 4zrp 64°6Spy) xdy ‘(oqtssurysey -olsousett-oueyy) dury $(SSue ‘Porqe ‘S'S1o) ypy ‘syjeo yun aATyoodsal WOT, ‘SOU UBSAXYC) °%OOT Wor, odUEIEyIp Aq Q*@H{ ‘AIjouOTYyOIO4s Aq Qoy pue &QO%aq ‘dnois oytqoourg ourg ‘dnods outyuoddog Jog ‘ojAIeg Ivegq ‘ouexoi1Adoury xdpy ‘efoqrydury dury ‘redsppoy TexrTy ATV (QT) oWueseq ‘SUTOA ‘SUITI UOT}VIOITY “(68S) oWUeseq AIXQ IV ‘SUTOA SUTIvoq-oyATeq ‘SosATeuRe GINA posl[eullou oaAtyeyuosoldoy *Z% o[qey, OFT'0 €19'0 - 62F'0 GER'0 ; (4-94 +31N) /31V JO JTUOTJUO NT ouTJuIdIaG OJAIC OUI IoquepeH OPSSUTSL TT JUTPIUeS JUIC NT 19 3IN VSLiv : V6L SIV JLT of Gg 8D GIp of 19% of EEIG ETE aw 18% ®O L'0OP &N LUCY 67S SIV 196"%Q €9P eD 9°§¢ SIN GPS M %soney uoryey ZG 6 v 9 &Z oF (suoSAxQ)) LIZ OL O82'S 066'0 916° GOS OT C66 wing WOTPD : : PS TE : : €OS . : LPT : . OLS = = 69'€9 , : Ord (1292) (9L°€1) : : (16'0) - (O%H) = = - = 60°0 = OZ 1Z'0 ST'0 : ZOO 9¢°T 18°8 O° 6r'0 v2'0 = 88'S G6'S 607 OPN 790 610 : 69°81 slain 90°T Or%D EST 18ST : £9°¢ Grrl Ir'0 OSV 10'0 r0'0 = 110 CIO 00°0 OU 2 69'ET : EET SP'8 VOT O24 GOST . I8'P ‘ . ®O%o4 GO'0 €0'0 = . 1V'0 00°0 OUD 96°IT 69°9 . 86% SL'PT 2961 OAV 80°0 E10 = Or'0 EG El'0 COLL L0°PP 09°8P - 9T1S ZO'CP CG'P9 COIS OUIG IOS leg xdy dury AV [eIOUIPY % IM OPIXO (QT) SuoTyeroqY 99e"7] YIM AIPIXO ‘(GQS) ose[quIessY UlsA 94e'T 38 SUTHERLAND et al. Xenocrysts in the basalt such as Mg-Al- rich pyroxenes and spinels show reaction rims (Figures 5.2 and 5.3) and resemble phases in the ultramafic meta-assemblages (Table 4). Com- positions are Al-rich enstatite (engg_91 fSs_9 wo;_2), Al-rich diopside (diss—s5g tscha4—25 hdjo_11 jd7 ac7) and Cr-bearing spinel (spgi—g2 he9_19 CMg—7 mt;—2 uspo_1). Mg numbers for these phases range from 0.96—0.71 and match values for these minerals in the ultramafic xeno- liths. Mullagalah Basalt Breccia The basalt fragments are dominated by phe- nocrysts of altered olivine and clinopyroxene, in Pyroxenite Xenolith Assemblage (T16), Oxide wt% a fine grained partly glassy groundmass. Only the fresh pyroxene was analysed. The zoned clinopyroxenes have Al-rich augitic cores (disg_59 tschg,; hdjg_11 acg—g jdg_3) and Al-rich diopsidic outer margins (dis5—60 tscha3_25 hd7_19 acg_19) and Mg num- bers range from 0.87 to 0.92 (Table 5). In composition (Cagg_47 Mgao—47 Feio_13), they have less Mg and more Fe than clinopyrox- ene phenocrysts in the Mount Oxley basalt (Ca4g4—s0 Mg35—42 Fej2-16), which have lower - Mg numbers (0.69-0.82). The compositions are also more aluminous (Alj2O3 8.7-9.0 at%) and sodic (Na2O 1.3-1.4 at%) than for Mount Oxley clinopyroxene phenocrysts (AlpO3 6.9-8.4 wt%; Na2O 0.7-1.2 wt%). Pyroxenite Xenolith Assemblage (S89), Oxide wt% Mineral Opx Cpx Spl Opx Cpx Spl SiO» 94.64 50.98 - 03.99 01.19 - TiO» 0.26 O73 0.34 0.21 0.84 0.21 AloO3 4.89 7.78 93.08 4.30 6.48 60.04 Cro03 0.52 0.62 E2306 0.16 0.28 7.48 Fe203 0.00 2.a2 0.00 0.00 0.44 0.00 FeO 6.70 IE abd 14.53 13.45 5.86 10.63 MnO 0.14 OL 0.26 0.26 0.18 0.19 MgO 32.08 14.99 18.64 27.14 13.95 ZA1t CaO 0.85 19.46 0.01 0.97 19.46 - Na 2O 0.07 1.89 0.07 a 1.32 - KO 0.02 - 0.01 - ~ - NiO 0.09 O.11 0.29 - - 0.45 Cation Sum 3.998 3.999 2a) 3.998 4.000 24.134 (Oxygens) 6 6 32 6 6 g2 Cation Ratios% Mg 88.4 Mg 45.8 Mg 56.9 Me 76.7 . \iG@aaass Mg 53.0 Fe 9.9 Ca 43.7 Fe 22.4 Fe 21.3 Mg 44.4 Fe 38.2 Can Fe 10.4 Cre20.7 Ca 2.0 Fe 11.1 Cr sil Name Enstatite Augite Spinel Enstatite Augite Spinel Mg /(Mg+Fe?+) 0.895 0.707 0.718 0.782 0.809 0.569 Table 3. Representative, normalised EMP analyses. Pyroxenite xenoliths, Mount Oxley basanite. Opx Orthopyroxene, Cpx Clinopyroxene, Spl Spinel. Oxygen nos. are based on respective unit cells. FegO3 and FeO by stoichiometry. Wells (1977) 2-pyroxene temperatures: T16 (993°C); S89 (988°C). Pyroxenite (T16) phases: Opx (engg 4, fS9.9, W0}1,7); Cpx (dig6.9, tschi¢.1, jde.9, ace.3, hd3.7); Spl (sp7a.2, cmi3.7, hes.g, mt29, Uspo.6). Pyroxenite (S89) phases: Opx (en7e6.7, fs21.3, w02.0); Cpx (dis9.1, hdis.5, tschi2.¢, jdg.2, aci.2); Spl (spgi.2, hcg.4, cm7.¢, mti.3, USpo.4). UNUSUAL BARYTE-BEARING HYBRID BASALT Xenoliths in the breccia include a cumulate- like, olivine-bearing clinopyroxene-amphibole- composite (Figure 7.4). Clinopyroxene, amphi- bole and spinel xenocrysts are probably derived from this association. The clinopyroxene is Mg- and Al-rich diopside (~ digg tschy2 jdio hd7 aca; Mg number 0.87—0.94), the spinel is a zoned Cr- bearing member of the spinel-hercynite series Xenolith Assemblage 39 (sp7o—74 hei7~21 Mt4.6 CMe uspy—2; Mg number 0.70-0.72) and the amphibole is K-Ti-enriched ferroan pargasite (Table 5). These phases may represent an ultramafic cumulate association, as the clinopyroxene has a higher Mg number (0.93) than for the clinopyroxene phenocrysts in the basalt (0.87—-0.92). Xenocryst Assemblage Oxide wt% Oxide wt % Mineral Opx Cpx Spl Opx Cpx Spl SiO» 54.15 51.39 0.02 54.42 51.02 0.00 TOs 0.13 0.70 0.27 0.14 0.78 O27 AlgO3 5.3 7.76 59.28 4.97 (Ae 61.24 Cro03 0.32 0.71 TAQ 0.36 0.49 6.62 Fe2O3 0.94 1.68 0.00 0.86 2,50 0.00 FeO ee 1.66 10.79 eo 0.99 9.48 MnO 0.15 0.15 0.41 0.18 O12 0.14 MgO 31.96 14.90 ZANo B242 14.78 21.59 CaO 0.90 19.03 - 0.83 19.47 0.03 Na2O 0.14 1.99 - 0.10 1.98 = KO 0.01 0.02 0.12 - 0.01 0.01 NiO 0.21 0.04 0.54 0.19 0.08 0.49 BaO 0.02 - - - - - Cation Sum 4.000 4.001 23.994 4.001 3.998 24.085 (Oxygens) 6 6 By) 6 6 32 Cation Ratios% Mg 89.0 Mg50.3 Mg 70.8 Mg90.7 Mg50.5 Meg 70.8 Fe 9.2 Ca46.2 Fel7.7 Fe 8.8 Ca47.6 Fe 17.7 Ca 1.8 Fe 3.4 Crii.d Ca 1.6 Fez.) -OCr ii Name Enstatite Diopside Spinel Enstatite Diopside Spinel Mg/(Mg-+Fe?* ) 0.909 0.941 0.784 0.913 0.964 0.708 Table 4. Normalised representative EMP analyses, harzburgite xenolith and xenocryst phases (T16). Opx Orthopyroxene, Cpx Clinopyroxene, Spl Spinel. Oxygen nos. are based on respective unit cells. Fe2O3 and FeO by stoichiometry. Xenolith: Opx (engg.o, fS9.2, wo1.6); Cpx (digs.7, tschi4.8, jdg.4, hds.5, aca.¢); Spl (spgi.6, heg.1, cm7.5, mt2.2, uspo.4). Xenocrysts: Opx (engo.7, fSs.8, w01.6); Cpx (diss.2, tschea.1, hdio.5, jdz7.o, acg.9); Spl (spgi.7, heo.s, cm6.6, Mt1.4, USpo.6). Wells (1977) 2-pyroxene temperature: xenolith (1005°C ). SUTHERLAND et al. AO ‘(opsesied uevodiey ‘uetueyy ‘uerssejod) dury :(ttdsn ‘téuro ‘8'rqur $6 0Z@oy ‘8°04ds) unta [dg {(V Tdsn ‘8'Tuto ‘T9qur ‘S2Toy ‘@€Lds) a100 [dg ‘(€Goe ‘9 9py ‘6 6pfl ‘6 TIyosq ‘6'891p) xdQ :yyyouex / ysA1D0OUeX *(94py ‘PF 6oe ‘6'€%yosy ‘4°6S1p) TA xdg pue (Tt ep ‘898 ‘POIpy ‘0 Téyosg ‘€°8S1p) a100 xdg ‘(0 Oloe ‘SOI py ST PZyosy ‘2°SS1p) xd_Q :sysA1o0UeY “(SesATeue Z Jo asvioAR) soqiydury “Yoo, Woy soUeIeyIp Aq poyeuitysa (Q%H) ‘Aseurorpstoys Aq Qeaq pue &O%aq ‘s[[eo yun aATyedsel UO poseq oie ‘sou UBSAXG ‘efoqrydury dury ‘foutds [dg ‘auexorhdoulyy xdd (279 INI) Booemq yeseq Yyeyeseynpy ‘seseyd yyyouex pue ysXr190u0yd ‘sosATeue GINA PoslyjeunloN “G oIqQeyT, 000'T 8690 9TL°0 666 0 9160 S880 OL8°O (4 -24+31N)/SIN oysedieg youtds joutds apisdoiq optsdoiq oySny aptsdoiq OUIVN] G16 8 Of 66 4D O'S eH GGL 8H LOT 4 GGL 24 SEP — Ve 6 EE COLVIN QVCPsN Screed 0'OF SIN 997 SIN 619 6'Z9 SIV OLV2O OLVRO Gor SIN SL4P XO = Bsorjyey woryey FG as Ze 9 9 9 9 (suoSAxQ) 916 OL 800 1G 910 VG LOOP 1Z6°€ 666'€ 666 '€ wing worye’) (96°0) : ; : : : ; (07H) 90°0 1G 0 ZT 0 IG'0 < . 800 OIN E : ‘ 5 4 = F Ord EV G 100 GO 0 100 100 VO'0 GOO O°M IV ; : ag oe T eV 8E'T O°®8N OF-TT 5 60 0 €G 0G LG°0¢G Sos 69°06 OX) 68 TT GEST 1681 GS Tl OV el co Tl 66°GL Os GOO GG O 1G 0 9T 0 60°0 LO 800 OUN 2 IZ Vl 6o ET 10°C OGG OG & GEG Oe CoO'GlL I8t 86S G80 807 86'G G8 & &O%oq r0'0 EO O8'T 080 €0°0 GO'0 ®OCID GOST £009 9G°6S 99°9 18°8 19°8 £0°6 ©O°C1IV Lies 69 0 Vo 0 GG 0 Ze 4 Gal 6V' I COLL 8G 6E 90°0 100 L0°GS VO'SP Il 6V VL LV COIS dury (wir) foutdg (e109) joutrdg xdg (uit) xdy (a100)xdgQ xdg [RIOUTP % IM 9PIXO % Ms OPIxO dsequIsse Y}W[OUIK—JSAIIOUDYX (ysoJ) ose[quIosse ySAIDOUOY J UNUSUAL BARYTE-BEARING HYBRID BASALT 4] T=? Taylor (1998) geotherm SEA O'Reilly & Griffen (1985) geotherm SEA(+80) Sutherland (2005) geotherm SP Spinel to Garnet Pyroxenite ‘GP experimental transition SE Spinel to Garnet Lherzolite GL experimental transition SW Mt Oxley Spinel Websterite 2-Pyroxene T SH Mt Oxley Spinel Harzburgite 2-Pyroxene T 900 1000 1100 1200 i om Figure 8. Pressure(GPa)—Temperature (°C) diagram showing relationships of two-pyroxene temperature estimates for Mount Oxley ultramafic xenolith meta-assemblages, with intersections of selected eastern Australian geotherms and experimental mineral assemblage transitions. A2 SUTHERLAND etal: DISCUSSION The porphyritic mafic intrusions described here were related to lamprophyres in some previous reports (Byrnes 1993), and were compared with camptonites and monchiq- uites. Camptonites contain amphibole to- gether with olivine and clinopyroxene as phe- nocrysts, while monchquites also contain micas and feldspathoids (Rock 1991). Amphibole is absent from the Mount Oxley phenocrysts, but does occur as a groundmass phase in the hybrid late-stage trachytic crystallisations. This hybrid occurrence is insufficient to jus- tify a camptonite nomenclature and a basaltic terminology is preferred for the mafic host. Amphibole occurs in the Mullagalah basalt, but is xenocrystic and probably derived from amphibole-clinopyroxenite xenoliths. For the ensuing discussion the Mount Oxley and Mul- lagalah mafic bodies are related to basanitic magmas. Mount Oxley Basalt Petrogenesis This basaltic intrusion displays a dual pet- rogenetic identity. Emplaced as a slightly evolved undersaturated magma, it become closely intermingled with a late-stage barium- and sulfur-bearing, evolved alkaline trachytic melt. The high temperature, high pressure ultramafic xenoliths within the basanite suggest initial evolution took place within the sub- continental mantle lithosphere, although the trachytic material could mark a higher level crustal fractionate. Such mantle-derived lava mingled with evolved felsic lava is rare in eastern Australian intraplate basalts. One example, however, is the meta-lherzolite and granulite xenolith-bearing nepheline hawaiite co-mingled with anorthoclase-bearing trachyte in the 3 Ma Mount St Martin volcanic centre in the Nebo Province, Queensland (Sutherland et al. 1977, Griffin et al. 1987). The hybrid lava there, how- ever, lacked phenocrystic nepheline in the mafic component and baryte in the felsic component, which feature in the Mount Oxley association. The Mount Oxley basanitic magma began crystallising its phenocrystic phases (olivine and diopside) at mantle depths prior to its eruption, while some crystallisation probably took place at higher levels to produce cumulate olivine microgabbros, as plagioclase joined the liquidus phases. These gabbros would crys- tallise at depths shallower than pressures re- lated to the olivine + plagioclase/two pyroxene- spinel transition zone (0.6—0.9 GPa, see O’Reilly et al. 1989). Nepheline, anorthoclase and ulvospinel would form accessory crystallising phases as the magma fractionated. A further injection of basanitic magma carrying mantle fragments probably broke up the higher level cumulates and intermingled with evolved tra- chytic melts, which had concentrated K, Ba, S and volatiles before erupting in explosive breccias. Such melt compositions possibly link into syenitic/carbonatitic melts. The age of the explosive emplacement is uncertain on present field relationships. Miocene leucitite emplace- ments are known to occur to the south at Byrock and El Capitan (McQueen 2007) and are inferred to the north east for a volcanic pipe exposed in the Bokhara area (Jaques 2006 and pers. comm. 2007). Buried plugs giving magnetic anomalies are also known to the north east in the Bundabulla area (Madden 1999), where alkali basalt and mantle xenolith-bearing ankaramite were recovered in drill cores recov- ered below the Early Cretaceous Surat Basin sequence. ‘These bodies suggest Mesozoic to mid-Cenozoic age limits for the Mount Oxley intrusions. The ultramafic xenoliths in the Mount Oxley intrusives represent mantle lithologies and their thermal conditions at the time of basanite eruption. The spinel meta-harzburgite was re-equilibrated at temperatures between 1000— 1035°C, while spinel meta-websterites were re-equilibrated at 985-1000°C, suggesting a slightly higher level origin within an ambient lithospheric geotherm (Figure 8). The absence of garnet in the spinel-bearing metapyroxenites and peridotites favours sampling from above the spinel-garnet pyroxenite and spinel-garnet peri- dotite transitions respectively (Figure 8). The lack of garnet, however, precludes precise pres- sure estimates and hence depths of origin, based on garnet-two pyroxene thermobarometry (see Sutherland et al. 2005). Potential pressures can UNUSUAL BARYTE-BEARING HYBRID BASALT 43 be assigned from intersections of two-pyroxene re-equilibration temperatures with a known geotherm, but this geotherm dependency pro- duces wide variations from projected models. Several xenolith-derived geotherms proposed for eastern Australian volcanic regions include a Cenozoic South East Australia (SEA) geotherm (O’Reilly and Griffin 1985), a lower temperature variant based on different thermobarometry (Taylor 1998), a more perturbed, higher tem- perature variant (Sutherland 2003) and an even cooler geotherm (Gaul et al. 2003). Intersec- tions of such geotherms with the Mount Oxley mantle xenolith re-equilibration temperatures are shown in Figure 8. ‘These range in pressure estimate differences by up to 0.5 GPa, or about 15 km in depth for the different models. Using an estimated Moho depth of 38 km (c. 1.05 GPa) for the Bourke-Byrock section (Collins et al. 2003) allows for mantle pressures within the spinel pyroxenite and spinel-peridotite stability fields respectively, compatible with the ultra- mafic lithological temperatures recorded for the xenolith suite. Mullagalah Basalt Petrogenesis The Mullagalah basalt fragments are incom- pletely crystallised, but their main phenocrysts and incipient crystallites suggests they repre- sent a similar but slightly less evolved magma than for the Mount Oxley basanite. The high pressure clinopyroxene and Cr-rich spinel xenocrysts suggest it also incorporated mantle fragments in its evolution. The common amphi- bole xenocrysts and amphibole-clinopyroxene cumulate assemblages indicate that a hydrous crystallisation was involved at depth, though not necessarily directly from the erupted Mul- lagalah magma. Mount Oxley — Mullagalah Magmatic Links The basanitic magmas that produced the Mount Oxley and Mallagalah intrusions are probably quasi-contemporaneous, but their age relationships to the other Mesozoic-Cenozoic mafic bodies in the area remain untested. They lie along a northern extension of the Miocene leucitite migratory line that trends south from Byrock. If their basanitic magmas formed as part of this trail, although tapping a less potas- sic mantle source, they would be about 18 Ma in age on extrapolation from the 17.2 Ma age at El Capitan (McQueen et al. 2007). Alternatively, they may be linked to the pre-Early Cretaceous buried alkali basaltic plugs further north at Bundabulla (Madden 1999). The relatively fresh nature of the Mount Oxley and Mullagalah basanites and their inclusion suites, apart from post-eruptive alteration of the olivines, would suggest Cenozoic links. A fresh appearance, however, is unreliable, as similarly fresh al- kali basalts and mantle xenolith suites around Dubbo, NSW, extend from Miocene to Jurassic or older ages (Meakin and Morgan, 1999). The Mount Oxley and Mullagalah breccia pipes were related to kimberlitic bodies in some explo- ration reports. The basanitic petrogenesis and lack of obvious garnets within the xenocryst and ultramafic xenoliths, however, weakens such ori- gins and renders the breccias unlikely diamond prospects. Mount Oxley Features Distinctive features are nepheline phenocrysts in mantle-origin basanite combined with abun- dant baryte in hybrid trachyte. The depth of nepheline crystallisation is uncertain, but the mineral can crystallise under mantle conditions (Edgar 1984). Nepheline stability is limited by reactions with SiO» in the system, where albite + nepheline gives liquid and nepheline + albite gives jadeite. This gives a maximum stability around 2.4 GPa at 1250°C. Baryte is rather rare in primary magmatic rocks, but is known in late-stage vesicles in porphyritic rhyolite, and in diorite and is common in hydrothermal veins (Zussman 1998). It is also recorded in syenite- carbonatite and lamproite associations (Fitton and Upton 1987) and in lamprophyres (Rock 1991). Secondary baryte crystals are known in Jurassic alkali basaltic diatremes (Dundas) and alkali dolerite intrusions (Prospect) that intrude the Sydney Basin Triassic beds. At Dun- das, baryte is found with calcite and pyrite in vughs in the diatreme and also in veins 44 SUTHERLAND et al. within metasedimentary fragments and wall rocks (Australian Museum specimens D35753, D41757). In the Prospect intrusion, baryte is rare and the last mineral to crystallise in the paragenetic sequence (England 1994), where it develops on albite, prehnite, calcite, siderite and pyrite (Australian Museum specimens D35330, 193535 1y)1D35686) D38535,/ D5 1168 D51170) iit would originate from late meteoric hydrother- mal fluids, rather than the earlier magmatic deuteric fluids based on oxygen isotope evidence (Williams and Carr 2005). In both the Dundas and Prospect intrusions, baryte was probably derived from barium released from the intruded Triassic shales, whereas the baryte in the Mount Oxley intrusion probably derived from Ba en- richment in fractionated trachytic magma. ACKNOWLEDGEMENTS Professor Ian Plimer of the University of Ade- laide initiated this study through donations of samples, reports and analytical results in 1983 from the original exploration of the Mount Oxley and Mullagalah intrusions. Ian Matthias, CRA Exploration PL, Cobar, NSW, supplied additional samples and information. Niels Munksgaard, School of Earth Sciences, Mac- quarie University, facilitated electron micro- probe analyses by Jane Barron, while Ken Kinealy, CSIRO Division of Exploration Geo- science, North Ryde, assisted analyses by Lin Sutherland, in 1989. Professor Peter Williams, School of Natural Sciences, University of West- ern Sydney, facilitated electron microprobe analyses by Adam McKinnon and Lin Suther- land in 2006. Dr Ian Graham, Australian Museum, provided photomicrography of thin sections and Ross Pogson, Australian Museum, assisted through provision of computer pro- grams and mineralogical calculations. Dr Larry Barron, Australian Museum, provided petrolog- ical discussion and read the script. Manuscript preparation was aided by Ms Jaqueline Timms, School of Natural Science, University of West- ern Sydney and Ms Francesca Kelly, St Peters, Sydney. Monique Ferguson kindly helped with the Figures. The paper is dedicated to Drs Edmund Potter and Maren Krysko von Tryst for their staunch services to the Royal Society of New South Wales. REFERENCES Byrnes, J.G., 1993. Bourke 1:250 000 Metal- logenic Map SH/55-10: Metallgenic Study of Mineral Deposit Data Sheets, 127 pp. Geo- logical Survey of New South Wales, Sydney. Cebeira, J.M., 1990. PX: A program for pyrox- ene classification and calculation. American Mineralogist 75, 1426-1427. Collins, C.D.N., Drummond, B.J. and Nicoll, M.G., 2003. Crustal thickness patterns on the Australian continent. Evolution and Dynam- ics of the Australian Plate (Hillis, R.R. and Mueller, R.D., eds), pp. 121-128. Geological Society of Australia Special Publication 22 and Geological Society of America Special Paper’ 372. Cundari, A., 1973. Petrology of the leucite- bearing lavas in New South Wales. Journal of the Geological Society of Australia 20, 465- 492. Edgar, A.D. 1984. Chemistry, occurrence and paragenesis of feldspathoids: a review. In: Feldspars and Feldspathoids, W.L. Brown (ed.) pp. 501-532. D. Reidel Publishing Company, Dordrechet. England, B.M., 1994. Minerals of the Prospect intrusion, New South Wales, Australia. The Mineralogical Record 25, 185-194. Fitton; J.G. and Upton, B.G.do ds. 937 Alkaline Igneous Rocks. Geological Society Special Publication No. 30. Oxford, 568 pp. Gaul, O.F., O’Reilly, S.Y. and Griffin, W.L., 2003. Lithosphere structure and evolution in Southeastern Australia. Evolution and Dynamics of the Australian Plate (Hillis, R.R. and Mueller, R.D., eds), pp. 185-202. Geological Society of Australia Special Publi- cation 22 and Geological Society of America Special Paper 372. Griffin, W.L., Sutherland, F.L. and Hollis, J.D., 1987. Geothermal profile and crust mantle transition beneath east-central Queensland: volcanology, xenolith petrology and seismic UNUSUAL BARYTE-BEARING HYBRID BASALT A5 data. Journal of Volcanology and Geothermal Research 31, 177-203. Giiven, N., 1988. Smectites. Hydrous Phyl- losilicates (exclusive of micas). S.W. Bailey (ed.), pp. 497-559. Progress in Mineralogy 19. Mineralogical Society of America. Jaques, A.L., 2006. Australian diamond deposits, kimberlites, and _ related rocks, 1:5000 000 scale map. Geoscience Australia, Canberra. Johnson, R.W. (ed.), 1989. Intraplate Volcan- ism in Eastern Australia and New Zealand. Cambridge University Press, Cambridge. McQueen, K.G., Gonzalez, O.R., Roach, IC., Pillans, B.J., Dunlap, W.J. and Smith, M.L., 2007. Landscape and regolith features related to Miocene leucitite lava flows, El Capitan northeast of Cobar, New South Wales. Aus- tralian Journal of Earth Sciences 54, 1-17. Madden, J., 1999. EL 5171 Bundabulla First and Final Report Angledool SH 55-07 1:250000 New South Wales Australia. Ge- ological Survey of New South Wales Report 1999 /259. Meakin, N.S. and Morgan, E.J., 1999. Dubbo 1:250000 Geological Sheet SI/55-4 (2nd edi- tion). Geological Survey of New South Wales Explanatory Notes. North Broken Hill Ltd, 197la. Assessment report, EL 292. Bourke-Byrock area. Geo- logical Survey of New South Wales, File GS 1971/346 (unpubl.). North Broken Hill Ltd, 1971b. Drilling aid, El 292, Byrock area. Geological Survey of New South Wales, File GS 1971/ 378 (unpubl.). North Broken Hill Ltd & Pressaug Australia Pty Ltd 1982. Final report, ELs 1208 and 1209, Mt Oxley, Bourke area. Geological Survey of New South Wales, File GS 1982- 339 (unpublished). O’Reilly, S.Y. and Griffin, W.L., 1985. A xenolith-derived geotherm for southeastern Australia and its geological implications. Tectonophysics 111, 41-63. O’Reilly, S.Y., Nicholls, ].A. and Griffin, W., 1989. Xenoliths and megacrysts of Eastern Australia. Intraplate volcanism in Eastern Australia and New Zealand. Johnson, R.W. (ed.), pp. 249-288. Press. Paul, B., Hergt, J.M. and Woodhead, J.D.., 2005. Mantle heterogeneity beneath the Cenozoic volcanic provinces of central Victo- ria inferred from trace-element and Sr, Nd. Pb and Hf isotope data. Australian Journal of Earth Sciences 52, 243-260. Preussag Australia Pty Ltd & North Broken Hill Ltd, 198la. Exploration reports, 3 and 4, ELs 1208 and 1209, Mount Oxley, Bourke area. Geological Survey of New South Wales, File GS 1981/ 095. Preussag Australia Pty Ltd & North Broken Hill Ltd, 1981b. Exploration reports, ELs 1208 and 1209, Mount Oxley, Bourke area. Geological Survey of New South Wales, File GS 1981/ 565. Rock, N.M.S., 1991. Lamprophyres. Blackie, Glasgow & London, 285 pp. Sutherland, F.L., 1985. Regional controls in eastern Australian volcanism. Geological So- ciety of Australia, New South Wales Division, Proceedings 1, 13-32. Sutherland, F.L., 2003. Boomerang migra- tory intraplate Cenozoic volcanism, eastern Australia rift margins and the Indo-Pacific mantle boundary. Evolution and Dynamics of the Australian Plate (Hillis, R.R., and Miller, R.D., eds), pp. 203-221. Geological Society of Australia Special Publication 22 and Geological Society of America Special Paper 372. Sutherland, F.L., Raynor, L.R. and Pogson, R.E., 2005. Table Cape vent xenolith suite, northwest Tasmania: Mineralogy and implications for crust-mantle lithology and Miocene geotherms in Tasmania. Papers and Proceedings of the Royal Society of Tasmania, 139, 7-22. Sutherland, F.L., Stubbs, D. and Green, D.C., 1977. K-Ar ages of Cainozoic volcanic suites, Bowen-St Laurence hinterland, north Queensland (with some implications for ge- netic models). Journal of the Geological Society of Australia 24, 447-460. Tadros, N. (Victor) (Compil. and ed.), 1993. The Gunnedah Basin New South Wales. Ge- Cambridge University 46 SUTHERLAND et al. ological Survey of New South Wales Memoir Geology 12. Taylor, W.R. 1998. An experimental test of some geothermometer and geobarometer formulations for upper mantle peridotites with applications to the thermometry of fer- tile lherzolite and garnet websterite. Neues Jahrbuch fur Mineralogie, Abhandlungen 172, 381-408. Trebaudino, M. and Bruno, E., 1993. Ef- fects of Al on enstatite solubility in CMAS clinopyroxenes: 1-Experimental results in the clinopyroxene-othopyroxene two phase field at P=18kbar. European Journal of Miner- alogy 5, 123-131. Wells, P.R.A. (1977). Pryoxene thermometry in simple and complex systems. Contributions to Mineralogy and Petrology 62, 129-139. Wicks, F.J. and O’Hanly, D.S., 1988. Ser- pentine minerals: structures and petrology. Hydrous Phyllosilicates (exclusive of micas), S.W. Bailey, (ed.), pp. 91-167. Reviews in Mineralogy 19, Mineralogical Society of America. Williams, M.L. and Carr, P.F., 2005. Isotope systematics of secondary minerals from the Prospect intrusion, New South Wales. Aus- tralian Journal of Earth Sciences 52, 799- 806. Wood, B.J..:and. Banno, 5\,,)aiG¥ee Gar- net orthopyroxene and _ orthopyroxene- clinopyroxene relationships in simple and complex systems. Contributions to Miner- alogy and Petrology 42, 109-121. Yavuz, F., 1996. Amphcal: A quickbasic program for determining the amphibole name from electron microprobe analysis using the IMA rules. Computers and Geosciences 22. (2) kode LOT: Yoder, H.S., and Tilly, C.E. 1962. Origin.of basalt magmas: an experimental study of natural and synthetic rock systems. Journal of Petrology 3, 342-532. Zhang, M. and O’Reilly, S.Y., 1997. Multi- ple sources for basaltic rocks from Dubbo, eastern Australia: geochemical evidence for plume-lithosphere mantle interaction. Chem- ical Geology 136, 33-54. Zussman, J. 1998. Baryte In Chang L.L.Y, Howie, R.A. and Zussman, J. eds. Rock Form- ing Minerals. Volume 5B Second Edition. Non-silicates: Sulphates, Carbonates, Phos- phates, Halides, pp. 3-29. The Geological Society, London. * School of Natural Sciences BCRI Campus, University of Western Sydney Locked Bag 1797, Penrith South DC NSW 2197 Australia + B.J. Barron 7 Fairview Avenue, St. Ives NSW 2075 Australia Author for correspondence. Dr Lin Sutherland email: L.Sutherland@uws.edu.au (Manuscript received 27.06.2007, accepted 12.11.2007) Journal & Proceedings of the Royal Society of New South Wales, Vol. 140, p. 47-54, 2007 ISSN 0035-9173/07/020047-8 $4.00/1 Applications and Limitations of Independent Component Analysis for Facial and Hand Gesture Surface Electromyograms GANESH R. NAIK AND DINESH K. KUMAR Abstract: In the recent past, there has been an increasing trend to use blind source separation (BSS) or independent component analysis (ICA) algorithms for biomedical data. This paper reviews the concept of ICA and demonstrates its usefulness and limitations in the context of surface electromyograms (SEMG) related to hand movements and facial muscles. In the first experiment ICA has been used to separate the electrical activity from different hand gestures. The second part of the study considers separating electrical activity from facial muscles. In both instances the surface electromyogram has been used as an indicator of muscle activity. The theoretical analysis and experimental results demonstrate that ICA is suitable for identification of different hand gestures using sEMG signals. The results identify the unsuitability of ICA when a similar technique is used for facial muscles with respect to different vowel classifications. | Keywords: Independent component analysis, surface electromyogram, motor unit action potentials, human-computer interaction, blind source separation INTRODUCTION Independent component analysis (ICA) has recently received a lot of attention both in biomedical signal processing and statistical sig- nal processing. Independent component analy- sis is a useful method for blind source separation (BSS) and unsupervised learning, where the observation vectors are assumed to be linear mixing of independent components. Efficient new ICA algorithms have been introduced to solve the blind source separation problems. ICA algorithms are successfully utilised for removing artefact and noise from recorded _biosignals, especially sEMG. Research that isolates motor unit action potential (MUAP) originating from different muscles and motor units has been reported in 2004 (Nakamura et al. 2004), where success is reported in the isolation of the differ- ent MUAP with applications for decomposing the sEMG at low levels of muscle activation. ICA has also been proposed for unsupervised cross talk removal from sEMG recordings of the muscles of the hand (Greco et al. 2003). Recently ICA had been utilised to identify different hand gestures (Naik et al. 2006). From the literature, ICA appears to be the emerging technology with solutions to most of the sEMG applications. Myo-electric activity originating from different muscles can be considered to be independent, making the use of ICA a suitable method for the separation of muscle activity originating from different muscles. Surface Electromyography (sEMG) is a sur- face recording of muscle activity. It is a result of the spatial and temporal integration of the MUAP originating from different motor units. Being non-invasive and an important indicator of muscle activity, sEMG is useful, but the presence of multiple muscle activity and the random nature of the transmission path makes the signal difficult to use reliably when muscle activity is small, and actions are complex. It is difficult to separate muscle activity originating from different muscles due to similarity in the signals. Earlier work by the authors has used sEMG to identify unspoken vowels with suc- cess (Kumar et al. 2004), but reliability issues exist. This paper reports research conducted to identify the applications of ICA in hand gesture identification, and to determine some of the limitations of using ICA for separation of sEMG from facial muscle activity originating from other muscles (cross talk). The paper reports theoretical analysis and experimental results and discusses the apparent discrepancies between the two. 48 NAIK AND KUMAR FACIAL MOVEMENT AND MUSCLES RELATED TO SPEECH The face can communicate a variety of informa- tion including subjective emotion, communitive intent, and cognitive appraisal. The facial musculature is a three-dimensional assembly of small pseudo-independently controlled muscles performing a variety of complex or facial func- tions such as speech, mastication, swallowing and mediation of motion (Lapatki et al. 2003, Parsons 1986). When using facial sEMG to determine the shape of the lips and the mouth, there arises the issue of the proper choice of muscles and their corresponding best location of electrodes. Face structure is more complex than that of the limbs due to a large number of overlapping muscles. It is thus difficult to identify the specific muscles that are responsible for specific facial actions and shapes. There is also the difficulty of cross talk due to the overlap between different muscles. This is made more complex because of the temporal variation in the activation and deactivation of the different muscles. It is impractical to consider the entire facial mus- culature and record its electrical activity. In this study, only four facial muscles have been selected. The Zygomaticus major arises from the front surface of the zygomatic bone and merges with the muscles at the corner of the mouth. The Depressor anguli oris originates from the mandible and inserts into the skin at an angle to the mouth and pulls the corner of mouth downward. The Masseter originates from the maxilla and zygomatic arch and inserts to the ramus of the mandible to elevate and protrude; it assists in side-to-side movements of the mandible. The Mentalis originates from the mandible and inserts into the skin of the chin to elevate and protrude the lower lip, to pull skin into a pout (Fridlund and Cacioppo 1996). SURFACE EMG AND INDEPENDENT COMPONENT ANALYSIS The EMG signal is widely used as a suitable means to have access to physiological processes involved in producing joint movements. The information extracted from the EMG signals can be exploited in several different applica- tions. sEMG is a non-invasive and painless procedure in which EMG signals are measured from electrodes on the skin. This technique has clear advantages over needle EMG. Most importantly it is painless for the patient and avoids health hazards for patient and doctor. Furthermore, sEMG is a quick and easy process that facilitates sampling of a large number of. MUPs (Fujimoto and Nishizono 1993). One major barrier is that, due to the wide pickup area of surface electrodes, sEMG waveforms exhibit significant interference. Surface EMG recordings provide a practical means to record from several muscles simultaneously but tend to be unreliable, i.e., recordings from a subject per- forming the same movement repetitively tend to have considerable trial-to-trial variability. sEMG recordings are also affected by cross- talk whereby several muscles may contribute to the recording of a given electrode, making the source of the signal difficult to identify. Recently, ICA has been proposed as a method to analyze sEMG recordings, and this addresses many of these concerns. One property of the sEMG is that the signal originating from one muscle can generally be considered to be inde- pendent of other bioelectric signals such as an electrocardiogram (ECG), electro-oculargram (EOG), and signals from neighbouring muscles. This opens an opportunity for the use of ICA in this application (Hyvarinen et al. 2001). BSS aims at recovering the sources from a set of observations. Applications include separating individual voices at a cocktail party. In the BSS problem, two processes are involved (Hyvarinen 1997, Comon 2001). These are the mixing and un-mixing processes. First, we observe a set of multivariate signals [x1(t), r2(t)...2,(t)| that are assumed to be linearly mixed with a set of source signals [si(t), So(t)...Sn(t)|. The mixing process is hidden so we can only observe the mixed signals. The task is to recover the original source signals from the observations through an un-mixing process. Equation (1) and (2) describe the mix- ing and un-mixing processes mathematically FACIAL AND HAND GESTURE SURFACE ELECTROMYOGRAMS A9 (Bell and Sejnowski 1995, Hyverinen and Oja 1997). GSAS (1) Wa = WAs (2) Mixing Unmixing For solving the BSS it is assumed that the number of observations is equal to the number of source signals. Matrix s contains the original source signals driving the observations, whereas the separated signals are stored in matrix u. They are both [n x t] matrices. A and W are both {nx n] matrices, the mixing and un-mixing matrix, respectively. If the separated signals are the same as the original sources, the mixing matrix is the inverse of un-mixing matrix, i.e., Aa Wy *. METHODOLOGY Experiments were conducted to evaluate the performance of the hand gesture recognition and facial muscle activity using surface EMG. Recording and Processing of Hand Gesture sEMG For the hand gesture experiments 5 subjects whose ages ranged from 21 to 32 years (4 males and 1 female) were chosen. The experiments were conducted on two different days on all five subjects. For the data acquisition a propri- etary surface EMG acquisition system by Delsys (Boston, MA, USA) was used. Four electrode channels were placed over four different muscles as indicated in Table 1 and Figure 1. A ref- erence electrode was placed at the Epicondylus Medialis. The experiments were repeated on two dif- ferent days. Subjects were asked to keep the forearm resting on the table with the elbow at an angle of 90 degree in a comfortable position. Three hand actions were performed and _ re- peated 12 times in each instance. Each time the raw signal sampled at 1024 samples/second was recorded. ‘The gestures used for the experiments are listed below and details are provided in Table 1. e Wrist flexion (without flexing the fingers) e Finger flexion e Finger and wrist flexion together but normal along centre line The hand actions and gestures represented low level muscle activity. The hand actions were selected based on small variations between the muscle activities of the different digitas muscles situated in the forearm. Figure 1. Hand gesture experimental set up with four electrodes. 50 NAIK AND KUMAR Channel Muscle 1 Brachioradialis 2 Flexor carpi radialis (FCR) 3 Flexor carpi ulnaris (FCU) 4 Flexor digitorum superficialis (FDS) Function Flexion of forearm Abduction and flexion of wrist Adduction and flexion of wrist Finger flexion while avoiding wrist Flexion Table 1. Placement of electrodes over the skin of the forearm. Recording and Processing of Facial sEMG Experiments were conducted on a single subject on two different days to test inter-day vari- ations. A male subject participated in the experiment. The experiment used 4 channel EMG configurations as per the recommended recording guidelines (Fridlund and Cacioppo 1996).