Modern analysis of Table Mountain Group sediments began with I. C. Rust's D.Sc. thesis "On the sedimentation of the Table Mountain Group in the western Cape Province" in 1967. Rust defined the stratigraphy of the Table Mountain Group, produced computer generated isopach and palaeocurrent maps for each formation and attempted palaeoenvironmental analyses based on what data he had available. For work dated prior to 1967 the reader is directed to Rust's excellent review in Chapter 2 of his thesis. The thesis served as a basis for Rust's later published work on the Cape Supergroup. Current published palaeoenvironmental models of the lower Table Mountain Group (the Piekenierskloof, Graafwater and Peninsula Formations) are based on a transgressive fluvial - littoral - shallow shelf model (Tankard et al., 1982) following earlier facies and palaeoenvironmental analyses (Tankard and Hobday, 1977: Rust, 1977; Hobday and Tankard, 1978: Vos and Tankard, 1981). The validity of this model has recently been questioned (Turner, 1986; 1987) although no comprehensive alternative has been proposed to date. The sedimentology of the upper Table Mountain Group i.e. the Pakhuis, Cedarberg, Rietvlei, Skurweberg and Goudini Formations (the latter three the newly named Nardouw Subgroup) has not been studied systematically. Good progress has recently been made on the fossil content of the Cedarberg Formation (Gray et al., 1986; Cocks and Fortey, 1986) and palaeoenvironmental analyses initiated in the Nardouw Formation. This thesis documents contributions to the geology of the Table Mountain Group. It is not the intention of the author to present an extensive overview and treatise on the lower Table Mountain Group, but rather to concentrate on three topics that can provide some insight into Table Mountain Group geology. The following three topics were selected 1) Petrology and Diagenesis of lower Palaeozoic sandstones in the s.w. Cape Sandveldt (Clanwilliam and Piketberg Discricts). 2) Palaeoenvironmental indicators in the Faroo Member, (Graafwater Formation) at Carstensberg Pass, R364. 3) Facies analysis of conglomerates and sandstones in the Piekenierskloof Formation: Processes and implications for pre-Devonian braid-plain sedimentology. These topics form the basis of the thesis.

The present investigation examines the relationship between the Proterozoic Richtersveld and Bushmanland Subprovinces in the westernmost part of the Namaqua Province, near Eksteenfontein, Republic of South Africa. There is a controversy about this relationship because isotopic data contrast with field evidence. On a regional scale the Richtersveld Subprovince is separated from the Bushmanland Subprovince by the northward-dipping Groothoek Thrust. North of the thrust the Richtersveld Subprovince is comprised of low grade volcano/ plutonic rocks of the Vioolsdrif Terrane and medium grade volcano sedimentary sequences of the Pella Terrane. Medium grade rocks of the Steinkopf Terrane (Bushmanland Subprovince) lie immediately south of the thrust. Late Proterozoic strata of the Stinkfontein Formation (Gariep Group) overlie the Namaqua Province in the west; Cambrian Nama Group outliers occur east of the Stinkfontein Formation. Isotopic data show that lithologies of the Richtersveld Subprovince formed between 2000 - 1730 Ma, whereas those of the Bushmanland Subprovince are younger. It is not clear whether the Namaqua metamorphic imprint (at 1200 - 1100 Ma), which is manifest in terranes south of the Groothoek Thrust, extended as far as the Vioolsdrif Terrane in the north. Early Proterozoic structural and metamorphic imprints are inferred to have been obliterated during this event. The westernmost part of the Namaqua Province was overprinted for a distance of 100 km from the coast, during the Pan-African event at 700 Ma and 500 Ma. An area measuring nearly 500 km2 , traversing the western extremity of the boundary between the Richtersveld and Bushmanland Subprovinces was mapped on a scale of 1:36,000. Field mapping was carried out with the aid of aerial photographs, whereas laboratory techniques included map compilation, structural analysis, X-ray diffractometry, geochemical (XRF) and electron microprobe analyses. Supracrustal units of the Richtersveld Subprovince are composed of quartzo-feldspathic gneisses, schists, and minor meta-pelites. Supracrustals of the Bushmanland Subprovince are less diverse than those of the Richtersveld Subprovince and have a disconformable relationship with them. Most intrusive rock-types are thick granitic sheets, except the Early Proterozoic Vioolsdrif Granodiorite which forms part of a batholithic pluton in the north. The Sabieboomrante adamellite gneiss, Kouefontein granite gneiss and Dabbieputs granite gneiss could not be correlated with lithologies commonly occurring in the Richtersveld and Bushmanland Subprovinces. They have been given the new rock names. Mafic and ultramafic rocks of the Klipbok complex occur along the strike of the Groothoek Thrust. They form part of the Richtersveld Subprovince.

M.Sc. / Please refer to full text to view abstract

The Platreef occurs at the base of the Northern Limb of the Bushveld Complex and is variably mineralised with PGE, Cu, and Ni. The Platreef varies in thickness from a few meters to a few hundred meters and rests on progressively older sediments of the Transvaal Supergroup and Archaean granite basement northwards. Recent studies have highlighted the importance of magmatic processes, contamination of the magma by footwall rocks and syn- and post metasomatic fluid activity on the observed mineralisation. Along the Platreef strike, the PGE grade profiles are generally top-loaded from Overysel to Tweefontein North and more variable and bottom loaded from Tweefontein Hill southwards emphasizing the importance of the change in mineralisation style at Tweefontein in relation to the whole Platreef. This study presents the first significant PGM data on the Tweefontein farm, including ten boreholes along strike, providing insight into the distinctly different PGE mineralisation styles observed. Samples were selected based on assay data, varying rock types, stratigraphic position and proximity to geological features. The selected samples were investigated using petrography, geochemistry and the automated SEM techniques of QEMSCAN and MLA. Over 9000 PGM were analysed forming one of the most comprehensive PGM studies on the Platreef to date. The lowermost footwall intersected along the Tweefontein strike is banded ironstone of the Penge Formation. This is overlain by a metasedimentary footwall package, of variable thickness, derived from the shales and dolomites of the Duitschland Formation. Iron-rich, recrystallised, noritic sills occur at the base of the Platreef and are thought to represent sills which intruded prior to the emplacement of the Platreef. A pre- and possibly syn-Bushveld structural control resulted in irregular floor topography defined by a topographic footwall high in the central Tweefontein area and topographic depressions at Tweefontein North and Tweefontein Hill. The depression areas at Tweefontein are similar to the footwall basins at Turfspruit to the south, in which the Platreef is more lithologically complex compared to the footwall high areas. The footwall basins at Tweefontein and Turfspruit contain basal massive and submassive sulphides, which may not necessarily carry significant PGE grade. The Platreef lithologies at Tweefontein are composed of pyroxenites and norites with minor harzburgitic lithologies and contain numerous cross-cutting granitic veins. Xenoliths/interlayers of metamorphosed Duitschland lithologies occur primarily near the base of the Platreef, but also in the middle and upper Platreef sequence reflecting roof pendants. Unlike the Platreef on the farms adjacent to Tweefontein, the Platreef and footwall lithologies are relatively unaltered, but localised serpentinisation and chloritisation occur within harzburgitic lithologies and metasedimentary interlayers. Based on the stratigraphy and geochemical characteristics, the Platreef at Tweefontein can be subdivided into the upper and lower Platreef. The upper Platreef subdivision occurs in the top 20-40 m of the sequence and is defined by higher Mg#, Cr, Cr (ppm)/MgO and Pt/Pd values compared to the lower Platreef. In addition, the majority of the grade and base metal sulphide (BMS) content is enriched in the upper versus the lower Platreef, particularly for the northern and central parts of Tweefontein. The upper and lower Platreef may have been derived from different magma sources based on the “R Factor” concept proposed by Campbell and Naldrett in 1979 whereby the abundance of the PGE relative to the BMS content is linked to the proportion of magma with which the sulphide ore equilibrated (Naldrett, 2005b). Previous detailed geochemical studies from Tweefontein Hill southwards highlighted compositional breaks in the Platreef sequence thought to represent distinct sill-like intrusions (Hutchinson and Kinnaird, 2005; Kinnaird, 2005; Manyeruke et al., 2005; Nyama et al., 2006). They reported a more primitive sill at the top of the Platreef, which correlates to the upper Platreef at Tweefontein. The lower Platreef is therefore likely to represent a different sill intrusion. A relatively homogenous pyroxenitic package characterises the upper Platreef, although a more heterogeneous package is observed close to and at Tweefontein Hill. At Tweefontein North, the base of the upper Platreef is often marked by a chromitiferous package comprising a pegmatoidal feldspathic pyroxenite unit, up to 6 m thick, capped by a chromitite layer. Due to similar stratigraphy and high PGE grades, this distinct horizon has been compared to the Merensky Reef found elsewhere in the Bushveld Complex. The predominant base metal sulphides (BMS) in the Platreef at Tweefontein are pyrrhotite, pentlandite, chalcopyrite with minor pyrite aligned with that found elsewhere along the Platreef strike. There is an increase in BMS content, primarily pyrrhotite, towards the base of the Platreef with massive and submassive sulphide development near the base and in the footwall, particularly at Tweefontein Hill. Sulphur isotopes and detailed mineralogical studies at Turfspruit have shown that the addition of S, As and Sb into the magma from the Duitschland footwall triggered the development of a PGE-poor sulphide liquid which was then able to mix, modify and dilute the magmatic sulphides (Hutchinson and McDonald, 2008). Due to the similarity in footwall between Turfspruit and Tweefontein, these proposed processes help to explain the increase in BMS towards the base and the development of basal massive and submassive sulphides, which are not necessarily associated with significant PGE grade. At Tweefontein North, the processes dominating the top-loaded PGE mineralisation were primarily magmatic. The PGM assemblage, hosted by base metal sulphides and magmatic silicates, is dominated by Pt-and Pd-bismuthides and -tellurides with minor PGE-sulphides and Pt-arsenides. PGE-sulphides occur in the Platreef where the chromitiferous horizon is developed, which may indicate an environment low in volatile activity and one of the most primary mineralisation styles along the Platreef strike. The footwall high, which separates the depressions at Tweefontein North and Tweefontein Hill may have kept the Platreef at Tweefontein North relatively protected from additional processes affecting Tweefontein Hill. In contrast, assimilation of the Duitschland footwall is thought to play a key role in the development of the variable but predominantly bottom-loaded PGE mineralisation at Tweefontein Hill. The PGM assemblage is Pd-dominant characterised by Sb-, As- and Bi-bearing PGM, reflecting the incorporation of Sb, As and Bi from the Duitschland footwall. The association of the PGE mineralisation with the extensive basal sulphide development implies that the mineralisation at Tweefontein Hill probably occurred due to the gravitational settling of a sulphide liquid containing a mix of sedimentary and PGE-hosting magmatic components. Due to a significant PGM-BMS association in the mineralised footwall and metasedimentary interlayers/xenoliths, a downward migrating sulphide melt is believed to be the main mechanism responsible for the redistribution of PGE, predominantly Pd, into the mineralised metasedimentary lithologies. Finally, the Platreef and footwall lithologies may be locally modified by late-stage felsic and hydrothermal fluids to form bismuthide- and arsenide-dominant PGM assemblages, primarily hosted in quartz and serpentine respectively. This study shows the PGM and sulphide mineralisation at Tweefontein to be multifaceted, involving magmatic processes, assimilation of the Duitchland footwall into the Platreef magma and late-stage hydrothermal and felsic fluid activity. Footwall composition and irregular floor topography, resulting in depression areas at Tweefontein North and Tweefontein Hill, are believed to play a key role in what processes become significant along the Tweefontein strike. This research represents a significant contribution to the understanding of the distinctly different PGE mineralisation styles at Tweefontein and allows for a complete comparison of the Platreef PGE mineralisation from Overysel to Turfspruit.

A dissertation submitted to the Faculty of Science, University of Witwatersrand in the fulfillment of the requirements for the degree of Master of Science (Geology) 2017 / The LG-6 chromitite layer is the thickest (0.90 to 1.20 m thick) chromitite layer in the Lower Group chromitites of the Bushveld Complex and is of economic significance owing to the relatively high Cr-content. It can be traced across the entirety of the western limb and is mined in both the western limb and the eastern limb. This study evaluates previously published models of chromitite formation using data from the LG-6 chromitite at Ruighoek Mine, western Bushveld Complex. Data includes petrographic studies of the reef and host rocks, whole rock analysis of the silicate host rocks and reef, and mineral chemistry for orthopyroxene, olivine and Cr-spinel using electron probe microanalysis (EPMA). In the Ruighoek region the LG-6 chromitite comprises up to 95 vol. % chromite and is typically hosted by orthopyroxenite. Borehole data indicated an area (about 250 m2 in size) where the LG-6 is entirely hosted by harzburgite (42% orthopyroxene, 11% Cr-spinel, 14% olivine, 32% serpentine and 1% other) rather than orthopyroxenite. The whole rock and mineral chemistry revealed that the LG-6 chromitite in this area has an exceptionally high Cr/Fe ratio, up to 2.1. The whole rock data also indicated several compositional reversals in terms of MgO, Al2O3, Cr2O3, FeO, and Mg# (Mg/ [Mg+Fe2+]) for the unenriched borehole 13R-3, and compositional reversals in Cr/Fe and Cr# (Cr/ [Cr+Al]) for the enriched borehole 13R-9 upwards through the chromitite layer. The hanging wall harzburgites are characterized by an increase in Mg# for the mineral chemistry of the Cr-spinel, orthopyroxene and olivine compared to those in the footwall harzburgite. Importantly, spatial 3D modelling of borehole data at Ruighoek mine (19 drill-cores) indicates that the elevated Cr/Fe ratio in LG-6 chromitite is coincident with a depression in the topography of the chamber floor at the time of formation of the LG-6 chromitite. These data are difficult to reconcile with existing models for chromitite formation in layered intrusions, such as the models for gravity settling, addition of a Cr-spinel crystal-laden magma, or a pressure increase. Thus, this work has developed a new model for formation of the LG-6 chromitite at Ruighoek Mine. The exceptionally high Cr/Fe ratio of LG-6 chromitite and its close association with harzburgite is attributed to multiple replenishments of the chamber by relatively primitive magmas. These are inferred to either be saturated in olivine and chromite, or chromite alone. The occurrence of relatively primitive rocks within the depression is suggested to be related to a local feeder situated within the depression. Injection of new, relatively dense magma pulses from the feeder are inferred to spread out across the chamber floor as basal flows owing to compositional stratification of the resident magma at the time of development of the LG-6 chromitite. The replenishing magmas contributed to the existing compositional stratification in the chamber, resulting in the most primitive composition within the depression of the chamber floor. Subsequent crystallisation of the most primitive magmas within the depression resulted in local development of LG-6 chromitite with exceptionally high Cr/Fe ratios together with the enclosing harzburgitic rocks. The thickness of the LG-6 chromitite is attributed to continuous replenishment by large volumes of new, chromite-saturated, magmas via the feeder channel located in the depression. This study suggests that magma stratification and the replenishment of the chamber by chromite-saturated magmas played an important role in the development of the chromitite layers of the Bushveld Complex. / MT 2018

The Sani Pass in the Natal Drakensberg is situated in the north-eastern sector of the Lesotho Highlands which forms a major Karoo-age basaltic massif in the Karoo Igneous Province. The volcanic section exposed in the pass is approximately 800m thick, and comprises a succession of regularly stratified, massive and amygdaloidallavas which were extruded mainly by fissure-type eruptions. Dolerite dykes, which now occupy thefissures,form a network ofpredominantly NE-SW and NW-SE trending topographic features. During post-eruption cooling hydrothermal solutions percolated through the volcanic succession and produced an amygdale zonation which was controlled predominantly by ambientpressure and temperature conditions. An original maximum thickness of 1 820m of the volcanic succession has therefore been estimated and an average fossil geothermal gradient of 111° C/km is conceived to have persisted during amygdale formation. New electron microprobe data are presented for the silicate phases in the Sani Pass basalts and dolerites. These data do not effectively separate the Sani Pass volcanic succession into different geochemical units. Microprobe analysesfor olivine, albeit limited, are in the forsterite range and indicate that a proportion of olivine in the high-MgO basalts is due to cumulus enrichment. The pyroxenes are predominantly augite and minor pigeonite, with some ofthe augites displaying a tholeiitic trend similar to that recognised at Skaergaard. Plagioclase is mainly in the labradorite to bytownite range, the phenocrysts being slightly enriched in anorthite compared to the groundmass. The use ofwhole-rock geochemistry for 67 basalts and 8 dolerites has permitted the recognition of five geochemically distinct magma types, namely, the Giant's Cup, Agate Vale, Sakeng, Mkhomazana and the Phinong. The Phinong basalts comprise the upper two-thirds ofthe volcanic succession and although are generally homogeneous, there is a slight tendency for the more evolved rocks to be found higher up in the stratigraphic sequence. The remaining magma types precede the Phinong succession and are generally enriched in silica and have higher Zr/Nb and lower PfZr ratios than the Phinong basalts. Within the pre-Phinong succession the Giant's Cup basalts are generally depleted in the compatible elements, while the overlying Agate Vale basalts are enriched in incompatible elements. Except for a marginally lower Na20 and Sr content, the chemistry ofthe Sakeng basalts is variable, generally overlapping with the other magma types. The Mkhomazana basalts are slightly enriched in MgO, Ni, Cr and Sc compared to all other pre-Phinong basalts. The dolerites in the area adjacent to the Sani Pass are geochemically similar to the Phinong basalts. The Phinong magma type is considered to be equivalent to the Lesotho magma type, based on their geochemical and stratigraphical similarities. In terms ofdiscriminant diagrams the Giant's Cup, Sakeng and Mkhomazana basalts generally show some compositional overlap with the Phinong, or plot in incoherentfields, but the Agate Vale basalts are distinctly different and might indicate a new magma type within the Karoo Central area. Broad compositional overlap between the Phinong basalts and those preserved at Kirwan and Heimefrontfjella, Antarctica, indicates juxtaposition of Antarctica along the southern African east coast in a reconstructed Gondwanaland. Petrogenesis of the Sani Pass basalts has been examined in terms of alteration, open and closed system fractional crystallization, partial melting procesess and a heterogeneous source. Although limited alteration and conduit contamination have occurred, the most feasible mechanism responsible for the geochemical variation lies in the existance ofinhomogeneities in the upper mantle at the time ofgeneration of the Sani Pass magmas. / Thesis (M.Sc.)-University of Durban-Westville, 1992.

The Central and Southern Domains of the Koras Group, situated on the Doornberg Lineament, are the structurally preserved remnants of a once more widespread late-syntectonic cover sequence. Detailed examination of the field relationships, lithology and petrography together with new geochemical data (30 analyses) has resulted in the proposal of a new geological succession consisting mainly of two cycles of bimodal basaltic-rhyolitic volcanics with interbedded, immature conglomerates and lithic greywackes. These two cycles, named the Boomrivier and Leeudraai Formations, are overlain by the immature, polymictic orthoconglomerates and red arkosic sandstones of the Kalkpunt Formation. The first volcanic cycle commenced with the Lambrechtsdrif basaltic andesites and was followed, after a short hiatus, by the Swartkopsleegte rhyodacites. The second cycle comprises the Rouxville basalts and basaltic andesites and the Swartkop and Kenilworth rhyolites. Field evidence suggests that eruption of the rhyolitic and basaltic volcanics in the second cycle was contemporaneous. Geochemically, the volcanics can be classified as an "average-K" to high-K, tholeiitic, subalkaline association which exhibits general similarities to other Southern African bimodal associations e.g., the tholeiitic lavas of the Wilgenhoutsdrif Group. The Koras Group is petrologically similar to the Sinclair Sequence which is presently considered to be its coeval equivalent, but the dominantly calc-alkaline character of the Sinclair rocks distinguishes them from the dominantly tholeiitic Koras lavas. In a short literature review, the four main hypotheses for the petrogenesis of bimodal associations: liquid immiscibility, crystal fractionation, two-stage partial melting and separate magma sources, are described and the most feasible explanation for the origin of the Koras lavas is thought to be a "separate magma source" hypothesis in which two cycles of mantle-derived basalts and crustal-derived rhyolites were produced in a zone of high heat flow and erupted in an area of crustal weakness. The middle- or late-Proterozoic Koras Group was formed during unstable tectonic conditions, in a depositional setting that was probably controlled by late folding of the underlying pre-Koras sequences as well as the major strike-slip movement and subordinate dip-slip faulting in the Doornberg Lineament.

A mapping project of the surface geology of the Lavino chrome mine and its surroundings was initiated in order to establish the surface geological relationships in the area. In so doing the chromitite layer presently being mined has been identified and potential exploration targets in the area have been outlined. The Lavino Chrome mine field area is situated within the eastern lobe of the Bushveld Igneous Complex. The area is bounded by in the north by the Steelpoort Lineament, in the west by the Dwars River fault and in the east by the contact with the Transvaal Sequence floor rocks. Layered igneous rocks (pyroxenites, norites and anorthosites) of the Rustenburg Layered Suite dominate the geological landscape at the Lavino mine. The fact that outcropping igneous rocks of the Critical Zone abut directly against the quartzite floor rocks on the mine property makes this area unique in the Bushveld Complex. The hills in the field area are capped by mafic/ultramafic iron-rich sheet - like bodies. Extensive strike-slip faulting is seen in outcrop in the area to the north/northwest of present mining operations. On the basis of field relationships, the main chromitite layer presently being mined at Lavino is identified as the Middle Group chromitite layer MG 1. Three other prominent chromitite layers stratigraphically associated with MG 1 are identified as the Middle Group chromitites MG 2, MG 3 and MG 4. Several other less prominent outcropping chromitite layers are tentatively identified as those belonging to the Lower and Upper group of chromitites. The disconformable nature of the contact between the layered igneous rocks and the Transvaal Sequence floor rocks has resulted in the development of a wedge of undifferentiated pyroxenites in the north of the field area. The economically important LG 6 chromitite layer may be developed in subcrop within this wedge.

Scheelite, wolfram Ite and cassIterIte mIneralizat ion is hosted wIthin numerous quartz-tourmaline-feldspar-fluorite veins at Van Rooi's Vley, N.W. Cape Province . MineralizatIon and hydrothermal alteration within, and around, these veins is hIghly complex and reflects the intricate interaction of hydrotherma l activity upon a structurally deformed sequence of ProterozoIc med ium to high-grade gneisses. Four distinct stages of alteration and mineralization occurred, including a l ate 'epithermal stage'. Although the location of mineralization was strongly controlled by st ructure, the concentration of mineralizati on was controlled by physicochemical variables, of which host-rock geochemistry was particularly important . Further W/Sn mineralization occurs on a local scale, some of which is spatia lly related to minor leucogranite dykes. Leucogranite bodies are not uncommon within the region and some are enriched in Wand Sn. By comparing FIB ra tio s,W/Sn ratios, the alteration mineralogy, the ore mineralogy and the Fe-content of tourmaline, the deposits within the Van Rooi's Vley area can be placed into a 'proximal' to 'distal' classification, with respect to a common source of mineralizing hydrothermal fluids. The Van Rooi's Vley deposit, whilst affiliated to greisen-style deposits, represents a ' distal' quartz-vein lode deposit. / KMBT_363 / Adobe Acrobat 9.53 Paper Capture Plug-in

Abstract The gold mineralization event within Archaean granite-greenstone terranes occurred during the late Archaean, and followed the intrusion of syn- to late-tectonic granitic plutons into previously deformed greenstone belts. An Archaean granite/gneiss-hosted gold deposit, in terms of this project, is classified as having a gold-assay cutoff of 1g/metric ton over widths of at least several metres, or higher grades over narrower widths and/or verbal descriptions that indicate such values. Fluid inclusion studies and isotopic data identify two possible origins for the auriferous fluids; namely magmatic and metamorphic. The exploration target according to the magmatic model, is a late-Archaean, hydrothermally altered, mineralized and fractured granitic intrusion preferably with a granodioritic or quartz-dioritic composition. The exploration target according to the metamorphic replacement model is a granitic stock that has intruded a zone of crustal weakness such as a shear zone, active during the late Archaean. Alternatively, the granitic intrusion should be affected by regionally extensive late-Archaean shearing. It should be hydrothermally altered, deformed and mineralized. Five areas within the Pietersburg granite-greenstone terrane were selected for the 'Regional Area Selection' phase of exploration for Archaean granite/gneiss-hosted gold deposits; namely Roodepoort, Waterval, Ramagoep, Moletsie and Matlala. Roodepoort contains a known granodiorite-hosted gold deposit; the Knight's Pluton, and served as an orientation survey for this project. The use and interpretation of LANDSAT images formed an integral part of exploration techniques; to assess their usefulness in the exploration of Archaean granite/gneiss-hosted gold deposits. Area selection criteria for granite/gneiss-hosted gold mineralization at Roodepoort are the major ENE-trending shear zone, the NNW-trending lineament and hydrothermal alteration, shearing, quartz-stockworks and sulphide mineralization within the Knight's Pluton. The origin of the gold within the Knight's Pluton is uncertain; both magmatic and metamorphic models are possibilities. Ongoing exploration is in progress at Roodepoort. The only area selection criterion for granite/gneiss-hosted gold mineralization at Waterval is the sericitized, subcropping granites located within trenches. Gold mineralization is insignificant. No area selection criteria for Archaean granite/gneiss-hosted gold mineralization were located at Ramagoep, Matlala and Moletsie. No further exploration is recommended for all these areas. The MES image interpretations were successful in identifying lineaments, granitic outcrops, greenstones, vegetation and soil cover. The Clay-iron images adequately differentiated betweeen iron-rich and clay-bearing areas. However, not all clay-bearing areas were associated with hydrothermal alteratian; field checks were necessary to discriminate between weathered granites and hydrothermally altered granites. The Wallis images served to locally enhance the contrasts of the MES and Clay-iron images.