The Merensky Reef and UG2 layers at De Wildt, western limb of
the Bushveld Complex

Zamani Addmore DUBE
(0303958G)

Thesis submitted to the Faculty of Science, University of the Witwatersrand Johannesburg in
partial fulfilment of the requirements for the degree of Master of Science.

02 September 2010

2
DECLARATION

I declare that this thesis is my own unaided work. It is being submitted for the Degree of Master of
Science in the University of the Witwatersrand, Johannesburg. It has not been submitted before for
any degree or examination in any other University.

Zamani A. Dube ? 02/09/ 2010

3
TABLE OF CONTENTS

Acknowledgments
List of Figures
List of Tables
Abstract

5
6
11
12
CHAPTER 1: INTRODUCTION

1.1. Preamble
1.2. Location and physiography
1.3. Aims of the research
1.4. Exploration Background
1.5. Acronyms and Abbreviations

13
13
15
15
16
CHAPTER 2: REGIONAL GEOLOGY

2.1. Preamble
2.2. Stratigraphy of the Bushveld Complex
2.3. Metallogenesis of the Bushveld Complex
2.4. Intrusive and discordant bodies
2.4.1. Iron Rich Ultramafic Pegmatoids (IRUPs)
2.4.2. Potholes
2.5. Structure and Metamorphism
2.5.1. Structure of the Bushveld Igneous Complex
2.5.2. Contact metamorphism
2.6. Regional Geophysics
2.7. Floor Rocks to the Bushveld Complex
2.8. Local Geology ? the Brits area

20
20
23
26
27
27
28
28
29
30
31
32
CHAPTER 3: METHODOLOGY

3.1. Research approach
3.2. Photogrammetry
3.3. Aeromagnetic data
3.4. Mapping and logging
3.5. Petrography and mineralogy

35
35
37
38
39
CHAPTER 4: REMOTE SENSING & FIELD INVESTIGATION RESULTS

4.1. Surface Mapping
4.2. Photogrammetry
4.3. Aeromagnetic Interpretation
4.4. Radiometric Survey

43
43
46
48
4

CHAPTER 5: STRATIGRAPHY OF DE WILDT AREA

5.1. Character of the Main Zone
5.2. Character of the Critical Zone
5.3. Character of the Marginal Zone
5.4. Merensky Reef at De Wildt
5.5. UG2 Chromitite layer at De Wildt
5.6. UG2 Chromitite facies
5.7. UG2 chromitite isopach
5.8. UG2 hangingwall units
5.9. UG2 footwall units
5.10. Dykes, sills and replacement rocks at De Wildt
5.11. Summary

50
50
52
56
61
61
70
74
74
74
76

CHAPTER 6: PETROGRAPHY

6.1. Quartzite xenoliths
6.2. Hornfels
6.3. Chromite-rich feldspathic pyroxenite
6.4. Replacement pegmatites
6.5. Summary

77
77
78
78
78

CHAPTER 7: MINERALIZATION

7.1. Merensky Reef PGE content
7.2. UG2 chromitite Isochon
7.3. PGM value distribution along the UG2 chromitite
7.4. Mineralized UG2 footwall at De Wildt
7.4.1. UG2 footwall isopach
7.4.2. UG2 footwall isochon
7.4.3. Combined UG2 and UG2 footwall isopach

81
81
82
86
89
90
91

CHAPTER 8: DISCUSSION

8.1. Merensky Reef
8.2. UG2 Chromitite layer
8.3. Meta-sedimentary xenoliths
8.4. Metamorphism
8.5. Structure and deformation pattern

92
94
100
102
103
CHAPTER 9: CONCLUSIONS

REFERENCES
APPENDIX
105
112

5
ACKNOWLEDGMENTS

I would like to extend my gratitude to Prof. Kim A.A. Ncube-Hein for her mentorship skills,
unwavering patience and time (especially during the field visits) in supervising this work. Also I
would like to express my sincere thanks to Nkwe Platinum as they provided funding and data for this
work. This thesis was improved by the careful review of Prof. Grant Cawthorn

6
LIST OF FIGURES

Figure 1: General geological map of the Bushveld Complex (adapted from Van der Merwe and
Cawthorn 2005). The main limbs of the complex are shown; Western Limb, Eastern Limb and
Northern Limb. The location of the De Wildt project, east of the Brits Graben is shown.

Figure 2: Mineral permits of farm portions making the De Wildt prospect are shaded in blue (as at
January 2005).

Figure 3: Diamond drill holes in and adjacent to the de Wildt prospect.

Figure 4: Stratigraphy of the Rustenburg Layered Suite of the Bushveld Complex (after Sch?rmann et
al., 1988).

Figure 5: Regional geological variations in the nature of Merensky Reef (after Viljoen, 1994). The
Swartklip facies is located due south of the Pilanesberg Complex while the Rustenburg facies is
centred on the town of Rustenburg. The De Wildt project area lies due east of the town of Brits.

Figure 6: Floor folds in the Magaliesberg Quartzite Formation in the south-western Bushveld
Complex, west of the De Wildt project area (modified after Davey, 1992). Floor folds include
the Kookfontein and Spruitfontein upfolds.

Figure 7: Major gravity anomalies in the western BC (modified after Viljoen, 1999). Gravity highs
occur north of Pilanesburg, north of the town of Rustenburg and due west of the De Wildt
project area. The gravity highs correspond with thickened mafic units that may represent feeder
sites.

Figure 8: The second vertical magnetic derivative map across the De Wildt project (Antonie, 2004).
Structures such as dykes and faults, which disrupt the mafic layers of the BC, where delineated
from this image.

Figure 9: Schematic representations of borehole logs drilled in the De Wildt project area by Nkwe
Platinum. The boreholes were used in the study to characterize the distribution and continuance
of the MR and UG2 layers.

Figure 10: Stratigraphic logs drawn from historical diamond boreholes that were re-logged and used
in the study (MR=Merensky Reef, UG2=Upper Group 2 chromitite and UG1=Upper Group 1
chromitite layer).

Figure 11: Schematic representation of logs of boreholes drilled by Nkwe Platinum in the western
portion of the De Wildt project area (MR=Merensky Reef, UG2=Upper Group 2 chromitite
layer).

Figure 12: Schematic map that combines interpretations of black and white aerial photographs and
Google Earth images of the De Wildt project area (blue) with those of topography and 1st and
2nd vertical derivative aeromagnetic data. West-trending lineaments are evident in the BC mafic
units and these are crosscut by an array of NE and NW-trending faults and two dyke sets.
Displacement on the Schietfontein, Zilkaatsnek and Lindiwe fault is sinistral (apparent) but the
faults are most likely scissor faults with normal displacement and west block down. The BC
mafic units dip north.

7
Figure 13: Photographic mosaic of De Wildt project area (orange box). NE trending lineaments (faint
blue) are apparent across the prospect. Quartzite hills occur to the east and south of the project
area and are crosscut and displaced by NW and NE-trending fractures. Small streams are
marked in green.

Figure 15: Interpretation map of Potassium and Thorium radiometrics across the project area. Zones
of high Th-K radiation occur in the central and western parts of the De Wildt Project area
(interpreted from aeromagnetic data from Antoine, 2004).

Figure 16: Provisional local stratigraphy of De Wildt prospect. The base of the GMA is interpreted as
the boundary between the Main Zone and the Critical Zone. The vertical separation between the
Merensky Reef and UG2 chromitite layer ranged from 115 m to 170 m.

Figure 17: Schematic borehole cross section depicting of the separation between Merensky Reef and
UG2 across the De Wildt project area. In the intersections in which the UG2 reef was multi-
layered, the top reef contact of the first UG2 chromitite unit was used for correlation. Pothole
interpretations were based on transgressed/missing footwall layers, thickened hangingwall
layers of MR, steep dips and overall stratigraphic attenuation of the MR-UG2 interval without
any discernable evidence of faults.

Figure 18: A calc-silicate xenolith that is rich in epidote (shown in red line) within a BC norite
sequence in borehole PWD4. Part of the disturbed UG2 footwall with flow structures as defined
by chromitite and magnetite stringers is delineated by the dashed green lines.

Figure 19: Diamond drill core of sedimentary xenoliths (in yellow) in core of the BC within Marginal
Zone of the BC at De Wildt project. Below the xenoliths is part of Transvaal Supergroup
quartzite intersected in borehole DWD11.

Figure 20: Poorly developed UG2 chromitite layer intersected in borehole DWD10. The UG2 is
underlain by meta-sedimentary xenoliths (borehole DWD10 intersection).

Figure 21: Typical Merensky Reef intersection of greater than 20 m thickness at De Wildt together
with schematic of the borehole (PDW01). The MR is altered with chlorite and serpentinite filled
joints.

Figure 22: Merensky Reef pyroxenite and the different footwall units intersected in borehole. Image a
is of a medium crystalline feldspathic pyroxenite layer making up the MR. Image b, is of a
poorly defined basal chromitite stringer that defines the bottom contact of MR. The MR
pyroxenite is weathered and chloritized. Image c, shows a well developed basal chromitite
stringer of MR, the undulating contact between the chromitite stringer and the footwall
anorthosite is defined. Image d, shows a half NQ core section of an irregular basal chromitite
stringer of MR in contact with mottled anorthosite footwall.

Figure 23: A schematic of the different Merensky Reef facies intersected at De Wildt. In (a) the MR is
characterized by a medium crystalline feldspathic pyroxenite that has a basal chromitite stringer.
In (b) the MR is wide but devoid of a basal chromitite stringer. In the potholed facies (c), the
MR pyroxenite is significantly thinned (in this case down to ~5.0 m).

8
Figure 24: (a) UG2 seam intersected in DWD07, the seam is cut by quartz-biotite veins and marked
by a generally high content of interstitial plagioclase. (b) Altered UG2 seam intersected at a
depth of > 604 m, the top reef contact is poorly developed, marked by vague plagioclase lenses
and anorthosite patches. The footwall pyroxenite has been heavily altered to chlorite giving the
rock a dark green appearance.

Figure 25: Schematic depiction boreholes in the De Wildt project area showing the UG2 facies types
in each borehole. Normal and split-reef facies occur in the area but the split-reef facies can be
further subdivided into B, C, D, and E types.

Figure 26: UG2 facies distribution in the De Wildt project area.

Figure 27: (a) UG2 normal facies. (b) The narrow B type sub-facies of UG2 split-reef facies.

Figure 28: Upper chromitite layer of the UG2 package, and the chromite-rich feldspathic pyroxenite
that separates the two chromitite layers.

Figure 29: Part of the UG2 package showing the chromite bearing norite unit. The chromite occurs in
the form of blebs and irregular stringers. The lower chromitite layer in is poorly packed
(intersected in borehole DWD12).

Figure 31: Borehole log showing the UG2 Split reef D-type sub-facies.

Figure 32: Narrow UG2 chromitite layer (a) and the upper UG2 chromitite layer with feldspar
inclusions giving it a spotted texture (b).

Figure 33: Chromite-rich feldspathic pyroxenite with chromite blebs immediately above the pencil
(borehole DWD13).

Figure 34: UG2 chromitite isopach map for De Wildt. Areas in which the footwall mineralization is
not developed have thicknesses are less than 1.0 m.

Figure 35: Photomicrographs of polished thin sections of selected core samples from various
boreholes at De Wildt. All images are under crossed polars. Images A and B are quartzite
characterized by veins composed of secondary oxide minerals. Sericite defines a zone around an
(unidentified) opaque in image B. Image C shows a highly altered rock in which saussuritized
plagioclase gives the rock a dusty appearance. Image D characterize contacts between
recrystallized quartzite and dolerite: secondary mineral such biotite have formed between the
contact and define contact metamorphism to hornblende-hornfels facies. Image E was derived
from a metasedimentary xenolith that underwent low-grade metamorphism (borehole PDW04).

Figure 35: Photomicrographs of polished thin sections of selected core samples from various
boreholes at De Wildt. All images are under crossed polars. Images F and G are hornfelsed
rocks marked by the presence of fractured quartz, plagioclase, biotite and vague pyroxene
crystals. The deformation in image F is interpreted to be localized and a result of magmatic
fluids associated with intrusion of the dolerite dykes prominent within the De Wildt project
area. Images H and I are of feldspathic pyroxenite hosting disseminated chromite, intersected
where the UG2 seam was anticipated. The chromitite grains do not exhibit a preferred
orientation but are scattered throughout the crystal mass. Image J is of pegmatite consisting of
plagioclase, clino and othopyroxene and biotite crystals with oxides minerals in interstitices.

9
Figure 36: Average PGE values for Merensky Reef. In most of the boreholes drilled by Nkwe
Platinum the grades were less than 0.1 g/t over widths of less than one metre near the top and
bottom contacts.

Figure 37: UG2 chromitite isochon map over the De Wildt project area.

Figure 38a: PGM value distribution within the UG2 chromitite layer (normal facies type) in borehole
SK02.

Figure 38b: Plots of Pt, Pd, Rh, Ru and Pt/Pd ratios of UG2 chromitite layer (from borehole PDW01).
The Pt/Pd ratio value display an overall upward increase within the chromitite layer.

Figure 39: UG2 chromitite isochon map showing the plot of centimetre grams per tonne across the De
Wildt project area.

Figure 40: Schematic depiction of the mineralized zone below the UG2 layer. The mineralized
footwall units (feldspathic pyroxenite/norite) host disseminated chromite, chromite blebs and
chromite slivers extending up to 4 m below the UG2 chromitite layer.

Figure 41: (a) UG2 feldspathic pyroxenite footwall characterized by irregular zones with
disseminated chromite, intersected in borehole DWD07. (b) UG2 feldspathic pyroxenite
footwall with irregular and vague chromite stringers, intersected in borehole DWD06.

Figure 42: Chromitite slivers and chromite blebs in a coarse crystalline-pegmatoidal meosonorite
underlying the UG2 chromitite layer in one borehole DWD08 intersection. Chromite
patches/blebs and slivers are shown by yellow lines.

Figure 43: UG2 footwall isopach across the De Wildt project area. The term ?footwall? in this case
refers to the chromitiferous UG2 footwall units consisting of feldspathic pyroxenite/norite
hosting disseminated chromite, chromite stringers and chromite blebs

Figure 44: UG2 footwall isochon showing the variation of grades within the mineralized footwall
units across the De Wildt project area. The term ?footwall? in this case refers to the
chromitiferous UG2 footwall units consisting of feldspathic pyroxenite/norite hosting
disseminated chromite, chromite stringers and chromite blebs

Figure 45: PGE value distribution along the mineralized UG2 footwall unit as represented in borehole
2098. The footwall unit is made up of a feldspathic pyroxenite with disseminated chromite
(~30%) and chromite slivers.

Figure 46: UG2 footwall isochon plot of centimetre grams per ton within the mineralized units across
the De Wildt project area.

Figure 47: Combined UG2 layer and mineralized footwall isopach. This plot gives total length
variation of the mineralized horizon (UG2 seam and footwall) across the De Wildt project area.

Figure 48: Schematic representation (not to scale) of the UG2 chromitite seam across the De Wildt
project area showing the relationship between the UG2 normal reef facies, UG2 split-reef facies
and feldspathic pyroxenite (fpx). The normal UG2 facies is confined to the west of the prospect
while the split-facies occurs predominantly in the eastern zone on Schietfontein and
Krelingspost farms.

10
Figure 49: Schematic representation of the stratigraphic succession on the eastern zone of the project
area (at Krelingspost Farm and eastward) toward the margin of the BC. In (A) the UG2
chromitite layer is interpreted to abut the floor rocks of the Transvaal Supergroup at depth. In
(B) the wedging out of the Upper Critical Zone occurs i.e. the lower stratigraphic units are cut
out against the floor rocks towards the east. This also explains why there was no intersection of
the UG1 layer in the eastern portions of the prospect and the absence of UG2 layer out-crop in
the vicinity of the quartzite hills.

Figure 50: Potassic quartzite hornfels intersected in borehole at Krelingspost Farm. It is not known
whether this unit is connected to the floor or comprises a Transvaal Supergroup outlier.
11
LIST OF TABLES

Table 1: The location and collar position of diamond boreholes drilled by different companies in De
Wildt project area. Coordinates in WGS 84, Cape datum.

Table 2: Simplified stratigraphy of Transvaal Supergroup and age data, modified after Eriksson et al.
(1993) and Dankert and Hein (2010).
(1) Burger and Coertze (1973-1974; Rb-Sr whole rocks age);
(2) Martin et al. (1998; single zircon U-Pb SHRIMP);
(3) Walveren and Martini (1995; single zircon Pb-evaporation);
(4) Unpublished; vvvvv = volcanic or pyroclastic unit.

Table 3: Specifications of airborne magnetic survey conducted over the De Wildt project area (after
Antoine, 2004).

Table 4: The UG2-Merensky Reef interval in the De Wildt project area. The data records the
disturbed Merensky ReefUG2 interval from the boreholes where layers were disrupted by
potholing. Faults were omitted.

Table 5: Thickness of the Marginal Zone intersected in several boreholes. The boundary between the
Critical Zone and Marginal Zone is tentatively taken at the level of the first appearance of the
fine-grained ?chill-margin? contact-type norite.

Table 6: Merensky Reef intersections within the prospect.

Table 7: UG2 chromitite reef intersections within the De Wildt project area.

Table 8: Immediate hanging and footwall lithologies to the UG2 chromitite layer.

Table 9: Prill Splits for the UG2 chromitite and UG2 footwall at De Wildt are shown on the left while
prill splits of the UG2 and MR reefs from other BC localities are given on the right.

Table 10: A summary table of the different mineralized UG2 footwall lithologies intersected in
various boreholes within the project area.
12
ABSTRACT

Geological and geophysical investigations involving drill-core studies, aeromagnetic data and
surface mapping at the De Wildt property in the south-eastern part of the western limb of the Bushveld
Complex have revealed a highly variable UG2 chromitite layer and a wide Merensky Reef (Merensky
Reef) layer. A full stratigraphic succession of the Rustenburg Layered Suite of the Bushveld Complex
is not developed at De Wildt. Instead, the Lower Zone and Lower Critical Zone are not developed and
the Upper Critical Zone is developed in close proximity to the Transvaal floor rocks. Numerous
deformed metasedimentary xenoliths and hornfels were intersected that are interpreted to be the result
of contact metamorphism during cooling Bushveld magmas. The rock succession at De Wildt has been
disturbed by post-Bushveld faulting and dolerite dykes whose emplacement is interpreted to have been
influenced by the adjacent Brits graben to the west.
The Merensky Reef intersected at De Wildt prospect, while exhibiting its regional
characteristics of a non-pegmatoidal feldspathic pyroxenite and thick distal facies, has undergone
further thickening up to about 20 m, and is marked by sub-economic PGE values. The UG2 chromitite
layer exhibited lateral variation from east to west, exhibiting variation in being a single massive
chromitite seam (averaging 1.10 m in thickness) into a multi-layered unit and/or consisting of only
disseminated chromite and chromitite stringers within a feldspathic pyroxenite. Also, the UG2 layer is
characterized by a PGE-enriched immediate footwall feldspathic pyroxenite and norite that is 4 metres
in thickness. Such a thick UG2 footwall (consisting of chromitite lenses and blebs) may indicate the
presence of higher than usual amounts of late-stage fluids during the fluidization processes. Prill splits
of the UG2 chromitite show that the reef has a high Pt:Pd ratio of 3:1 in contrast to the common
western Bushveld UG2 Pt:Pd ratio of 2:1. It is interpreted that the UG2 chromitite layer was exposed
to late stage magmatic and/or meteoric fluid metasomatism with the subsequent remobilization of Pd
as evidenced by the presence of replacement magnetite and quartz-biotite veins within the UG2 layer.
13
CHAPTER 1
INTRODUCTION

1.1. Preamble

The Bushveld Complex of South Africa is one of the most intensively studied layered igneous
intrusions on Earth. Historically, most research has been conducted on open-pits and under-ground
mine workings, but the geology away from these mines is not well documented, in particular, the
setting of the platinum-rich Upper Group 2 (UG2) chromitite and Merensky Reef layers. Maier and
Bowen (1996) reported that in most parts of the western Bushveld Complex, the UG2-Merensky Reef
interval is situated about 100 m below the Critical Zone-Main Zone boundary. However, as a result of
poor exposures and a relatively high degree of faulting, the field relationships (of UG2-Merensky Reef
interval) to the east of Brits are unclear.
The De Wildt area (including the farms, Zilkaatsnek 439 JQ, Schietfontein 437 JQ and
Krelingspost 425 JQ), lies on the western limb of the Bushveld Complex, northwest of Pretoria, about
15 km to the east of Brits. An exploration program carried out by Nkwe Platinum (SA) (Pty) Limited
(hereafter referred to as Nkwe Platinum), which commenced in November 2003, provided the
opportunity to investigate the nature of the UG2-Merensky Reef interval in that region. The program
included surface mapping, aeromagnetic survey and diamond drilling (20 boreholes), and was
supplemented by geochemical sampling of core and a GIS investigation of the De Wildt prospect
region.
The strato-structural setting of UG2 chromitite and Merensky Reef in the De Wildt area was
correlated and compared with the strato-structural setting of the UG2 chromitite and Merensky Reef in
the adjacent Rustenburg facies to the west of the De Wildt project area in order to establish similarities
and differences in the nature of the reefs, and to develop an overall model for reef variation.

1.2. Location and physiography

The De Wildt project is located on the topographic-cadastral Sheet 2527DB Brits (Chief
Directorate: Survey and Mapping, 1996). The co-ordinates central to the De Wildt prospect are 25?
37?S and 027? 56?N. The property is located approximately 15 km east of the town of Brits, and
approximately 40 km north of Pretoria (Tshwane) in the western limb of the BC (Fig. 1). The
operating Maroelabult section of Barplats?s (BPL) Crocodile River mine and the Elandsfontein
platinum project for Elands Platinum are situated west of the De Wildt project. The ferrochrome
operation of Hernic Ferrochrome is situated to the southwest.
14

Figure 1. General geological map of the Bushveld Complex (adapted from Van der Merwe and Cawthorn 2005). The main limbs of the complex are shown; Western Limb,
Eastern Limb and Northern Limb. The location of the De Wildt project, east of the Brits Graben is shown.
15
The De Wildt project area covers three farms namely, Zilkaatsnek 439 JQ, Schietfontein 437 JQ and
Krelingspost 425 JQ, and covers a total area of ~1200 ha for which Nkwe Platinum has the mineral
prospecting rights (Fig. 2). A railway line from Pretoria to Brits is situated along the northern
boundary of the property while the N4 toll road is situated on the southern property boundary. The
townships of Makhau and Ga-Rankuwa are located to the north of the De Wildt area: R556 road
crosses the De Wildt and provides access to the two townships and the town of Brits. The area has
relatively low summer rainfall of approximately 600-650 mm per annum and an average annual
temperature of approximately 19?C. The area is situated within a savannah biome, comprising
scattered trees and shrubs (Acocks, 1988). The topography of the area attains 1200 m with prominent
hills of the Magaliesberg Quartzite and the Bushveld Complex Main Zone norite on the southern and
northern boundary, respectively. The area is marked by generally warm to hot summers and mild to
cold winters.

1.3. Aims of the research

? To investigate and document a comprehensive structural and geological setting of the
Merensky Reef and UG2 Chromitite layer in the De Wildt area, based on geophysical and
geological datasets.
? To attempt a revision of the stratigraphy of the project area.
? To evaluate the variable nature of the UG2 Chromitite layer in the De Wildt project area in
relationship to surrounding operating mines.
? To establish a model for variations in the UG2 Chromitite layer in the De Wildt project area
that will aid in evaluating the platinum potential.

1.4. Exploration Background

An exploration program by Nkwe Platinum was initiated in November 2003. The initial stages
of the program involved surface mapping, aeromagnetic survey and diamond drilling. A total of 20
boreholes were drilled in 2005/6. Additional borehole data became available to Nkwe Platinum after
its acquisition of the Impala Platinum old order mining rights in 2004. These boreholes are shown in
Fig. 3.
Historically, five companies have conducted exploration studies in the project area and some on
farm portions adjacent to the project area (Table 1). They included Phelps Dodge and Gencor-Impala
during the late 1980?s, Century Minerals (Pty) Ltd, the African Development Trust (SADT) and
Meteorex in the 1990s. Phelps Dodge drilled 18 vertical-diamond-drill holes (SK1-18) on
Schietfontein Farm (Figs. 2 and 3). A total of 11 holes intersected the UG2 at depth of 49-50m (SK2-
10, 13 and 14) while the percussion hole ZK05 intersected the UG2 at 15 m. Gencor-Impala drilled
more than seven diamond-drill holes (boreholes 1960, 1977, 1990, 2055, 2092 and 2098) which
16
intersected the UG2 at depths of 190-1170m. The boreholes were drilled on Zilkaatsnek and
Schietfontein Farms (Figs. 2 and 3). The African Development Trust (SADT) drilled two holes on
Hoekfontin 432JQ Farm, which is situated north of the De Wildt project area, and SADT drilled two
diamond-drill holes (HFK1and HFK2), which intersected the UG2 at depths of 800-1100 m. In
2000/2001 Century Minerals (Pty) Limited drilled three diamond boreholes (W1, R1 and R2) and
excavated three trenches on Zilkaatsnek Farm. The boreholes did not intersect the UG2 reef as they
were collared in the UG2 footwall; trenches were excavated in an attempt to expose the UG2. Lastly,
Metorex drilled one borehole on Zilkaatsnek (ZK1) which intersected the UG2 Chromitite layer at
about 160 m.

1.5. Acronyms and Abbreviations

The stratigraphic nomenclature of the Bushveld Complex used in this research accords with the
South African Committee for Stratigraphy (SACS) of 1980. The following abbreviations are used
herein:-

BC - Bushveld Complex
MR - Merensky Reef
UG2 - Upper Group 2
GMA - Giant Mottled Anorthosite
RLS - Rustenburg Layered Suite/Sequence
PGEs/PGMs - Platinum Group Elements/ Platinum Group Metals
MZ-Main Zone
CZ ? Critical Zone
MgZ-Marginal Zone
17

Table 1: The location and collar positions of diamond boreholes drilled by different companies in De Wildt
project area. Coordinates in WGS 84, Cape datum.
BHID X COLLAR Y COLLAR Z COLLAR

BHID y x z

GENCOR/IMPALA
DH1960 98950 -2834605 1248
DH1977 94783 -2835129 1216
DH1990 90813 -2834376 1186
DH1995 92301 -2833692 1128
DH2046 88513 -2834081 1177
DH2055 91736 -2836535 1213
DH2063 90161 -2837257 1208
DH2092 91816 -2837219 1221
DH2098 91083 -2836830 1216
DHZ05 90860 -2837780 1222
DHZN5 91950 -2838080 1237

NKWE PLATINUM
DWD01 94353 -2836954 1232
DWD02 93875 -2836791 1225
DWD02A 93861 -2836790 1225
DWD03 95295 -2836642 1233
DWD03A 95272 -2836561 1232
DWD04 96545 -2836114 1249
DWD05 93875 -2837289 1233
DWD06 94500 -2836297 1216
DWD07 96130 -2835651 1232
DWD08 96144 -2835954 1238
DWD09 93855 -2836405 1218
DWD10 97917 -2835324 1249
DWD11 95120 -2835673 1221
DWD12 97820 -2834924 1242
DWD13 95689 -2836067 1234
DWD14 95653 -2835593 1225
PDW01 91989 -2837374 1222
PDW02 92894 -2836886 1220
PDW03 93328 -2836889 1220
PDW04 93324 -2837090 1222

PHELPS DODGE
SK01 91460 -2836630 1208
SK02 91470 -2836980 1214
SK03 91610 -2837280 1218
SK04 91850 -2837830 1228
SK05 91710 -2837900 1229
SK06 91725 -2837550 1223
SK07 92456 -2837599 1232
SK08 92730 -2837420 1228
SK09 91665 -2837321 1225
SK10 93010 -2837280 1222
SK11 93610 -2837010 1228
SK12 91450 -2837090 1228
SK13 92150 -2837700 1218
SK14 92764 -2836781 1218
SK15 93360 -2836450 1222
SK16 93380 -2836520 1225
SK17 92720 -2838200 1242
SK18 92850 -2837930 1238
ZK01 90237 -2837694 1209
20
CHAPTER 2

REGIONAL GEOLOGY

2.1. Preamble

The Bushveld Complex (BC) is the world?s largest layered igneous intrusion, exhibiting
layering over millimetres to hundreds of metres (Eales and Cawthorn, 1996; Eales, 2001). It hosts over
half the world?s platinum, chromium, vanadium and refractory minerals, and arguably has ore reserves
capable of lasting for a hundred to a thousand years (Vermaak, 1995). The complex was emplaced at
2061 ? 27 Ma based on single zircon age from felsic volcanic rock from the Rooiberg Volcanics
(Walraven et al., 1990). The BC has an areal extent of approximately 65 000 km2 (Tankard et al., 1982
and Cawthorn et al., 2006).

2.2. Stratigraphy of the Bushveld Complex

The Bushveld Complex consists of four main suites of igneous rocks namely, the Rooiberg
Group, the Rustenburg Layered Suite, the Rashoop Granophyre Suite and the Lebowa Granite Suite
(SACS, 1980; Cawthorn et al., 2006).
The BC intruded metasedimentary rocks of the Transvaal Sequence, largely along an
unconformity between Magaliesberg Quartzite of the Pretoria Group (Walraven, 1997; Eriksson et al.,
2007). The floor rocks of the BC exhibit considerable lateral variation and consist of meta-volcanic
rocks (lavas), quartzite, shale and dolomite (Viljoen and Sch?rmann, 1998). The Bushveld Granite
Suite consists of the Rashoop Granophyre and Lebowa Granites, and forms the bulk of the roof-rock
assemblage, together with quartzite, hornfels and dolomite (Kruger, 2005).
The RLS crops out in five limbs/lobes (Fig. 1) including the Far Western Limb, Western Limb,
the Eastern limb, the Northern Limb (also known as the Potgietersrus limb) and the Bethal Limb
(Eales, 2001). The Bethal Limb does not crop out, but is known to exist from deep borehole data and
positive gravity anomalies.
The lithological units of the northern, western and eastern limbs approach 8km in stratigraphic
thickness and describe a saucer-shaped ellipse that extends over an east-west distance of
approximately 370 km and a north-south distance of approximately 200 km (Cawthorn et al., 2006).
The Rustenburg Layered Suite (RLS) of the Western Limb (which host the Brits prospect)
extends for 200 km along an arc from Thabazimbi in the north to Pretoria in the south (Eales and
Cawthorn, 1996; Cawthorn et al., 2006). It is divided into northern and southern sectors that are
separated by the intrusive Pilanesberg Alkaline Complex.
21
The RLS is subdivided into five zones (Fig. 4), the Marginal Zone overlain by Lower Zone,
Critical Zone, Main Zone and the Upper Zone (Cousins, 1959 citing Hall, 1932, van der Merwe,
2008). The Critical Zone occurs only in the Western and Eastern Limbs of the BC where it is enriched
in Platinum Group Elements (PGEs) in the Merensky Reef and UG2 reefs, relative to marginal
cumulates (Cousins, 1959 citing Hall, 1932).
The Marginal Zone is the lowermost unit of the BC and ranges in thickness from 0 to 250 m in
the western Bushveld. The zone generally comprises marginal contact rocks or arguably, the chill zone
at the base of the BC (Kruger, 2005), i.e., it is mixed and contaminated by metasedimentary xenoliths.
The zone consists of heterogeneous gabbronorite, norite and hybrid metasediments (Vermaak and
Hendricks, 1976; Kruger, 2005).
The Lower Zone overlies the Marginal Zone and is characterized by cumulate pyroxene and
olivine-rich layers including pyroxenites and harzburgite (Viljoen and Shurmann, 1998). The
thickness of the Lower Zone has been influenced by the topography and structure of the sedimentary
floor such that local attenuation of the ultramafic rocks occurs where floor-rock structural complexities
exist (Cawthorn et al., 2006).
The Critical Zone overlies the Lower Zone. The base of the Critical Zone coincides with the
first appearance of cumulus chromite in the sequence, while the top is defined by poikilitic anorthosite
known as the Giant Mottled Anorthosite (GMA); it is situated a few metres above the Merensky Reef
(De Klerk, 1995). The Critical Zone is subdivided into the Lower and Upper Critical Zones on the
basis of cumulus plagioclase. The Lower Critical Zone consists entirely of ultramafic layers such as
orthopyroxenite, harzburgite and chromite layers (Kruger, 2005). The Upper Critical Zone contains
both ultramafic and mafic layers such as chromite, feldspathic orthopyroxenite, norite, and anorthosite
layers with some olivine bearing layers (Kruger, 2005).
The Lower and Upper Critical Zones are well layered and contain extensive deposits of
chromitite (Vermaak and von Gruenewaldt; 1986, Bristow et al., 1993; Viljoen and Sch?rmann, 1998).
The chromitite layers have been subdivided into three groups (Viljoen and Sch?rmann, 1998). From
the base, these are the Lower Group (LG), the Middle Group (MG) and the Upper Group (UG). A total
of thirteen (13) chromitite seams of varying thickness are recognized (Cousins and Feringa, 1964;
Viljoen and Sch?rmann, 1998) including:

? Seven LG seams with a total thickness of 3.5 m (LG1, LG2 and LG2A, LG3, LG4, LG5, LG6
and LG6A and LG7).
? Four MG seams with a total thickness of 4 m (MG1, MG2, MG3 and MG4).
? Two UG seams with a total thickness of 2 m (UG1 and UG2).
22

Figure 4. Stratigraphy of the Rustenburg Layered Suite of the BIC (after Sch?rmann et al., 1988).

23
The Lower Critical Zone hosts the LG, MG1 and MG2 seams (Fig. 4), while the upper Critical Zone
hosts the MG3, MG4, UG seams, and the Merensky Reef. The UG2 chromitite seam and Merensky
Reef are the main platiniferous units/horizons of the BC and are situated towards the top of the Upper
Critical Zone (Viljoen and Sch?rmann, 1998). The Bastard Reef cyclic unit overlies the Merensky
Reef cyclic unit and consists of thin layers of chromitite, pyroxenite, norite and anorthosite (Viljoen
and Hieber 1986; Cawthorn et al., 2006). Near the township of Marikana (situated west of De Wildt
project), the Bastard Reef occurs as a pyroxenite (~2 m) or norite (~12 m) (Davey, 1992). The Bastard
Reef is located ~10 m above the Merensky Reef in some localities in the western BC (Farquhar, 1986,
Viljoen and Hieber, 1986).
The Main Zone overlies the Critical Zone. The base of the Main Zone occurs at the top of Giant
Mottled Anorthosite (GMA) and is marked by the appearance of cumulus clinopyroxene and a loss of
chromitite. The Main Zone makes up almost half the thickness of the BC intrusion. It is characterized
by a relatively monotonous sequence of norite that passes upward into gabbronorite, with occasional
anorthosite and pyroxenite bands. Several mottled anorthosite layers occur in the lower portion of the
Main Zone and a distinctive pyroxenite layer (Pyroxenite Marker) occurs two thirds of the way up.
The middle to upper portions of the Main Zone are resistant to erosion and give rise to the distinctive
hills that are evident throughout the Western Limb of the BC (Bristow et al., 1993). However, it
should be borne in mind that some authors (Kruger, 2005) put the Main Zone-Critical Zone boundary
at the Merensky Reef level based on different initial Sr-isotope compositions and the existence of an
unconformity. Kruger (2005) explained the unconformity to be the result of major magma influxes
concomitant to erosion of the existing cumulates. As such the Merensky Reef is regarded as the first
layer at the base of the Main Zone. The first appearance of cumulus magnetite defines the top of the
Main Zone (Bristow et al., 1993) or, the bottom of the Upper Zone. The Main Zone is one of the
world?s major sources of dimension stone.
The Upper Zone overlies the Main Zone and consists of gabbro, magnetite-gabbro, anorthosite
and troctolite layers. The upper contact of the Upper Zone is non-conformable with the Bushveld
Granite Suite or Transvaal Sequence. The Upper Zone hosts twenty-five layers of magnetite that make
up approximately 8% by volume of Upper Zone rocks. It is thus relatively well layered (Eales and
Cawthorn, 1996).

2.3. Metallogenesis of the Bushveld Complex

PGM mineralization occurs in 13 of the chromitite seams of the Critical Zone (LG1, LG2, LG3,
LG4, LG5, LG6, LG7, MG1, MG2, MG3, MG4, UG1 and UG2) and increases progressively upward
from the lowermost layer to the uppermost UG2 chromitite layer (Viljoen and Sch?rmann, 1998). The
Merensky Reef, UG2 Chromitite and the Platreef are the main sources of platinum (PGEs) in the
Bushveld Complex. In the western limb of the Bushveld Complex only economic PGE mineralization
24
occurs in the Merensky Reef and UG2 chromitite layers. The platinum to palladium (Pt: Pd) ratio of
UG2 in the western Bushveld averages 2.5, while in the eastern limb the ration is 0.94. The grade
(Pt+Pd+Rh+Au) of the UG2 at Rustenburg Platinum Mines Ltd is reported to be 5.49 g/t. The Pt:Pd
ratio of the Merensky Reef in the western limb averages 2.6 (Viljoen and Sch?rmann, 1998).
At its base, the Merensky Reef consists of a dark pyroxenite unit (the Merensky pyroxenite)
with the mineralized portion of the reef being referred to as the Merensky Reef Layer, i.e., that part of
the Merensky pyroxenite that is economic to mine. The Merensky Reef pyroxenite is associated with
thin chromitite stringers of variable thickness (Vermaak and Van Der Merwe, 2000). The Merensky
Reef can be traced for considerable distance laterally: approximately 140 km in both western and
eastern limbs. Seismic surveys suggest that the reef can be traced as far as 50 km down-dip and to a
vertical depth of 6 km (Cawthorn and Eales. 1996). The reef dips 9? to 27? toward the centre of the
complex.
Viljoen and Sch?rmann (1998) recorded that the Merensky Reef typically forms a
heterogeneous pegmatoidal, feldspathic pyroxenite with scattered sulphide blebs. It is generally bound
by narrow (? 1 cm) chromitite stringers. The footwall to the Merensky is typically noritic or
anorthositic, though in some places, feldspathic pyroxenite or harzburgite are associated with pothole-
type Merensky reef. The upper contact is gradational from feldspathic pyroxenite and norite, to
anorthosite.
In the western lobe of BC, the Merensky Reef has been divided into the Swartklip facies due
south of Pilanesburg, and the Rustenburg facies centred on the town of Rustenburg (Fig. 5), owing to
regional variations in thickness and the abundance, size and type of potholes. The Swartklip facies
consists of narrow pegmatoidal and potholed Merensky Reef of high PGE grade, while the Rustenburg
facies consists of wider, generally non-pegmatoidal Merensky Reef of lower PGE grade. The
Rustenburg facies extends from the eastern side of Rustenburg Section to the Brits area (close to the
De Wildt project area); it widens from 1m in the Rustenburg Section to 14m in the Brits area where it
becomes sub-economic in PGEs (Viljoen, 1994). Naldrett et al. (2009) explained that the Merensky
Reef was the result of major magma influxes that gave rise to pre-Merensky and Merensky cyclic
units. They show an upward decrease in Pt tenor and upward increase in Cu/Pd ratios across the
pyroxenite of Merensky Reef and also laterally away from feeder zones (Naldrett et al., 2009).
The UG2 chromitite of the Upper Critical Zone is a massive continuous layer that hosts
mineable grades of chromite, with interstitial base metal sulphides and PGM minerals. It has been
traced for over 280 kilometres in strike, is 0.50 m to 1.50 m thick, and dips north from 10? at
Rustenburg to 26? in the northwestern limb (Viljoen and Sh?rmann, 1998). Up to three thin chromitite
stringers (leader seams) are situated a few tens of centimetres above the main chromitite layer in the
hangingwall pyroxenite. The leader chromitite seams are known to bifurcate and coalesce (Leeb-Du
Toit, 1986; Viljoen et al., 1986). The UG2 chromitite is generally underlain by a pegmatoidal
25
feldspathic pyroxenite that often displays a mottled appearance due to the presence of large bronzite
grains within the chromite crystals (Leeb-Du Toit, 1986 and Viljoen et al., 1986).
PGE mineralization is concentrated on the upper and lower contacts of the UG2 chromitite
(Maier and Barnes, 2008), with minor concentrations in the leader chromitites. Mineralization may
also extend a few tens of centimetres into the hangingwall and footwall rocks. The top and bottom
contacts of UG2 chromitite are characterized by high PGE contents that suggest the UG2 seam formed
by injection of chromitite slurries into semi-consolidated footwall units (Maier and Barnes, 2008).
However, Heimstra (1985) showed that the UG2 chromitite layer is made up of three layers
representing distinct cycles with discrete Pt/Pd ratios; the uppermost cycle has the highest Pt/Pd ratio
of about 3 while the lower two thirds or remainder has a Pt/Pd ratio close to 1.3. In the simple case,
there is no silicate between the three layers and they appear as one, while in places they are separated
by partings and are thus discrete (adapted by Cawthorn, 2004 from Hiemstra 1985).

Figure 5: Regional geological variations in the nature of Merensky Reef (after Viljoen, 1994). The Swartklip
facies is located due south of the Pilanesberg Complex while the Rustenburg facies is centred on the town of
Rustenburg. The De Wildt project area lies due east of the town of Brits.

26
Recent models for the formation of massive chromitite seams involve magma mixing of
relatively differentiated residual magma with a magnesium-rich primitive magma, in association with
variations in the pressure of the chamber (Lipin, 1993; Cawthorn, 2005). Maier and Barnes (2008)
suggested that the chromitite seams may have formed from flowing crystal mush that was injected
along broadly bedding-parallel fractures into semi-consolidated footwall cumulates, along an inclined
floor of the magma chamber. Mathez and May (2005) suggested that the chromitite may have acted as
a barrier to ascending late magmatic fluids causing fluidization of the footwall cumulates. Maier and
Barnes (2008) added that fluidization may have aided downward injection of the chromitite into the
footwall. Variations in thickness, lithology and composition within the BC are controlled by proximity
to the main feeder zone, such as at the Union Section (Maier and Eales, 1997). Scoon and Teigler
(2005) noted that the Upper Group chromitites (UG2 and UG1) tend to be thicker in the distal facies
(relative to the main feeder zone) while the Merensky Reef is well developed in the proximal facies
but thicker and poorly mineralized in the distal facies. Teigler et al. (1992) and Scoon and Teigler
(2005) observed that the Upper Critical Zone in the distal facies is thinner and dominated by
plagioclase cumulates, while in the proximal facies it is thick, well developed and dominated by
pyroxene cumulates.

2.4. Intrusive and discordant bodies

Rocks of the Rustenburg Layered Suite are intruded by lamprophyre and dolerite dykes and
sills, and the Pilanesberg Complex (Leeb-Du Toit, 1986). Carbonate-rich dykes are exposed in the
Amandelbult section of the Rustenburg Platinum Mines, and are brecciated and kimberlitic. In the
Impala Mines lease area, micaceous (lamprophyre) dykes that range in thickness from centimetres to a
few metres, are exposed in underground workings and are hosted by en echelon joint sets (Leeb-Du
Toit, 1986). Vermaak and Von Gruenewaldt (1986) reported the presence of abundant NW-trending
alkaline dykes of the Pilanesberg Complex that crosscut Transvaal metasedimentary rocks and the BC
in the region immediately west of Brits Township.
The Pilanesberg Complex is a large body of intrusive and extrusive alkaline rocks with aerial
extent of approximately 530 km2. The complex is situated about 50 km NNW of Rustenburg at the
contact of granitic and mafic rocks of the BC (Lurie, 1986) and has an east-west diameter of 28 km
and a north-south diameter of 24 km. It forms a well-defined ring structure. The complex intruded the
Western Limb of the BC ca. 1138 ? 45 Ma (Rb-Sr, in nepheline syenite; Harmer and Sharpe, 1985). It
was emplaced during intrusion of syenite along ring fractures and the extrusion of lava and deposition
of pyroclastite. The final volcanic events associated with emplacement were marked by the intrusion
of foyaitic cone sheets and ring dykes. Alkaline dykes (syenitic) are associated with the ring complex
(Lurie, 1986).

27
2.4.1. Iron Rich Ultramafic Pegmatoids (IRUPs)
IRUPs are discordant bodies in the BC that assume pipe-like/carrot-like, vertically elongate or
irregular bodies (Eales, 2001). They are composed of clinopyroxene, olivine, ilmenite, amphibole,
mica, and sulphide minerals together with Fe-Ti-Cr oxides. They are dark-brown to black in colour.
Their width varies from a few centimetres to several kilometres (e.g. the Middellaagte Pipe at
Amandelbult Section Mine is approximately two kilometres wide). The pipes are replacement bodies
that passively replaced existing layered rocks by downward magmatic fluid percolation, particularly
plagioclase-rich rocks such as norite and anorthosite (Eales and Cawthorn, 1996). In the western limb
of the Bushveld, at Amandelbult, IRUPs are known from their pronounced magnetic signature (Eales
and Cawthorn, 1996). They are spatially associated with dykes and faults but preferentially replace
anorthosite layers between the UG2 and Merensky Reef. Elements that are a product of replacement
include Fe, Ti, V and Ca. A marked excess of Fe is expressed by a high percentage of magnetite.

2.4.2. Potholes
Potholes are considered as disruptions of normal magmatic layering (Viljoen, 1994) and
significantly and/or deleteriously impact mining of the UG2 and Merensky Reef layers. A pothole is a
feature in which the unit truncates its footwall and forms a basin-shaped depression. The disrupted
magmatic layer occurs within the pothole in a modified form when compared to its regional character
(Viljoen, 1994, Lomberg et al.; 1999). Models for pothole formation have included:
? Injection of magma at specific feeder points followed by gravity sagging and chemo-thermal
activity (Campbell, 1986).
? Kinloch et al. (1990) and Boudreau (1992) postulated that pothole formation resulted from
development of significant intercumulus liquid and gas over-pressure in solidifying cumulates.
By escape of gas or fumarolic activity these regions of gas over-pressure eventually remove
cumulus material to form potholes, as suggested by the presence of hydrous minerals in
potholes.
? Potholes are rolls (Lomberg et al., 1999). A roll is a feature in which there is a small change in
strike and dip of a layer due to small-scale slumping. The affected layer only registers a small
drop in elevation with little discernable change compared to layers affected by potholes. Rolls
are common both to the UG2 and Merensky Reef but seldom disrupt mining operations.

Van der Merwe and Cawthorn (2005) documented the existence of irregular basal contacts on the UG2
and Merensky Reef. They summarized a number of processes that could account for small-scale
structural features at the base of UG2 and Merensky Reef including redissolution, interference ripples,
diapirism, gas-escape structures and impact generated dimples. Vermaak (1976) ascribed the
undulating contacts to the base of Merensky Reef to impact-generated dimpling. The dimpled surface
of the basal chromitite stringer of the Merensky Reef overlying an anorthosite footwall is the result of
28
impact of boulder-sized aggregates of pyroxenite onto unconsolidated chromitite and the underlying
anorthosite (Vermaak, 1976). Kinnaird et al. (2002) citing Carr et al. (1999) suggested that potholes
are the result of downward collapse of crystal mush into pull-apart sites, the pull-apart forming during
tensile stress and magma loading, and that the texture is similar to dish-shaped potholes on the base of
the chromitite units.

2.5. Structure and Metamorphism

2.5.1. Structure of the Bushveld Complex
Regionally, the emplacement of the BC magma into Kaapvaal Craton was facilitated by craton-
wide lateral extension following plume-related underplating and gravitational loading (Uken, 1998).
The subsequent isostatic adjustment in response to subsidence of the RLS is considered to have
resulted not only in the formation of the current lapolithic shape of the BC (Gerya et al., 2004), but
also its lobate character (Uken, 1998). The structures developed in response to magmatic loading;
large basement domes (~100 km in diameter) around the perimeter of BC, medium sized domes (~40
km in diameter) within the BC and extending to the base of the Transvaal Supergroup, and small-scale
domes and diapers (~8 km in diameter) occurring only within the contact metamorphosed floor rocks
in the metasedimentary units of the Pretoria Group in the north-eastern BC (Uken and Watkeys, 1997;
Uken, 1998). Kruger (2005) concluded that the BC is lobate, interconnected, wide but shallow, and a
sill-like intrusion with upturned margins. This is supported by Clarke et al. (2008) who showed that
the BC is a sill-shaped sequence of mafic-ultramafic rocks.
During the emplacement of the BC, contact aureole tectonism was expressed by sagging and up-
folding/doming of the floor with formation of local basins, and this exerted a significant influence on
the subsequent development of the Critical Zone (Hatton and Von Gruenewaldt, 1985). The up-
folds/domes are described by Clarke et al. (2008). The intrusion of the essentially sill-like RLS
resulted in the heating (thermal metamorphism) of the underlying floor rocks; the heated, plasticized
and molten rock rose diapirically into the overlying crystallizing magma (Uken and Watkeys, 1997;
Uken, 1998). Davey (1992) suggested that floor folds in the western BC within the Magaliesberg
Quartzite (e.g., the Kookfontein and Spruitfontein upfolds) could be the reason why the Lower Critical
Zone is absent from the Critical Zone (Fig. 6). Viljoen (1994) attributed lateral stratigraphic variations
in chromitite layers and the thickening of the Merensky Reef pyroxenite east of Rustenburg Section to
syn-emplacement folding/doming. Davey (1992) postulated that the Spruitfontein up-fold might
separate two magma sub-chambers in which similar, but not identical, processes took place during the
development of the Critical Zone.
29

Figure 6: Floor folds in the Magaliesberg Quartzite Formation in the south-western Bushveld Complex, west of
the De Wildt project area (modified after Davey, 1992). Floor folds include the Kookfontein and Spruitfontein
upfolds.

2.5.2. Contact metamorphism
Heat released from cooling of the BC magmas is associated with metamorphism of the country
rock at a lateral distance of up to 50 km from the contact (Cawthorn et al., 2006). Regionally, the
contact aureole in the floor rocks to the BC thickens from south of Pretoria toward the east, west and
north. The peak average thermal gradient estimated across the aureole was 80 ? 20?C/km with
pressures of 2.1 ? 0.4 kbar in the upper part of the Pretoria Group (Gerya et al., 2004).
Contact metamorphism of Transvaal Supergroup wall rocks was described by Hartzer (1995)
from the Crocodile River, Marble Hall and Dennilton domes: contact metamorphism is generally low-
to medium-grade. However, rocks in the inner contact aureole attained pyroxene-hornfels facies, as
characterized by the presence of augite and hypersthene (Hartzer, 1995). In contrast, xenoliths of
meta-argillite (of the Transvaal Supergroup) within the BC are characterized by the assemblage
sapphirine, mullite, spinel and corundum (Sharpe and Chadwick, 1982), which is common for high
grade hornfelses adjacent to mafic intrusions (Pattison and Tracy, 1991).
In argillitic units of the Crocodile River dome, contact metamorphism attained 410-510?C and
2.5 kbar pressure, as expressed by the development of the metamorphic assemblage andalusite, biotite,
muscovite and quartz (Hartzer, 1995). Carbonate rocks attained 530?C and <2.5 kbar pressure with
30
development of the calc-silicate mineral assemblage tremolite ? talc, calcite, quartz and dolomite.
Volcanic rocks are characterised by the metamorphic assemblage hornblende, plagioclase, quartz,
muscovite and biotite, which is characteristic of hornblende-hornfels facies metamorphism.
Sedimentary rocks at the contact with Bushveld granite in the Rooiberg fragment underwent contact
metamorphism to albite-epidote hornfels facies (Rozendaal et al., 1986).
Buick et al. (2000) citing Wallmach et al. (1989; 1995) reported that the RLS contains
numerous calc-silicate xenoliths that were initially metamorphosed at ~1200-1300?C and later
retrogressed by hydrous fluid infiltration at ~700?C. The occurrence of hydrothermal vein systems
(comprising calcite at low temperature, hornblende and plagioclase) throughout the RLS has been
attributed to the ingress of hydrous C-O-H bearing fluids at ~350-700?C (Schiffries and Rye, 1990).
Buick et al. (2000) showed through stable oxygen isotope work that the fluids were derived by
dehydrating the country rocks in the floor of the BC intrusion. They suggested that the retrogression
was a result of channelled infiltration of fluid into cooling RLS mafic rocks where it exchanged 18O
with low 18O RLS mafic rocks. Stevens et al. (1997) documented the occurrence of high grade mid-
crustal granulite facies metamorphism that may have resulted from the intraplating of Bushveld
Complex-related ultramafic magmas before the Vredefort impact event.

2.6. Regional Geophysics

Geophysical surveys including gravity, seismic and magnetic have been conducted by Davison
and Chunnett (1999) and Viljoen (1999) to investigate the structure and emplacement of the BC. The
complex exhibits a 30 km long major gravity high immediately northeast of the Pilanesberg and a
second 9 km long gravity high northeast of Rustenburg in the western Bushveld (Fig. 7). These gravity
highs correspond with thickening of mafic units and are interpreted as indications of feeder sites for
mafic magma in the western BC (Viljoen, 1999). Small gravity highs occur over the Brits graben and
in the area between Brits and Pretoria adjacent to the De Wildt project area.
Regional gravity interpretation in which there was a lack of positive gravity anomaly at the
central area of the BC led to a disregard of the possible connection of the western and eastern limbs at
depth (Cawthorn and Webb, 2001). However, based on petrographic and stratigraphic similarities in
the UG1, UG2, middle group chromitites, Merensky Reef, Upper Zone/Middle Zone transition it is
assumed that the two major limbs are connected and form part of one large magma chamber
(Cawthorn and Webb, 2001). Cawthorn and Webb (2001) suggested that the depression of the
continental crust under the BC by up to 6 kilometers was the result of isostatic readjustments due to
the enormous mass of overlying mafic rock.
31

Figure 7: Major gravity anomalies in the western BC (modified after Viljoen, 1999). Gravity highs occur north
of Pilanesberg, north of the town of Rustenburg and due west of the De Wildt project area. The gravity highs
correspond with thickened mafic units that may represent feeder sites.

2.7. Floor Rocks to the Bushveld Complex

The BC intrudes the Transvaal Supergroup which is composed of a 10-15 km thick sequence of
relatively undeformed sedimentary and volcanic rocks (Button, 1986; Eriksson et al., 1995, 1993;
2007). The complex intruded the Transvaal Supergroup along an unconformity above the
Magaliesberg Formation of the Pretoria Group (Walraven, 1997; Eriksson et al., 2007). The floor to
the Rustenburg Layered Suite of the BC consists of intercalated mudstone-sandstone units, andesite
lava and subordinate conglomerate beds, diamictite and minor carbonate units of the Pretoria Group
(Hatton and Schweiter, 1995; Eriksson et al., 2007).
The Transvaal Supergroup (Table 2) has been correlated to Griqualand West basins of South
Africa and the Kanye basin of Botswana by Eriksson et al. (1993; 2007). The Transvaal basin is dated
at ca. 2.5-2.2 Ga, but the age of the Pretoria Group is constrained to ca. 2.35-2.20 Ga based on a Rb-Sr
whole rock age of 2224 ? 21 Ma for the Hekpoort Formation of the Pretoria Group (Burger and
32
Coertze, 1973-1974), and an unpublished 2350 Ma age (Eriksson et al., 1995; Frimmel and Minter,
2002) for the Timeball Hill Formation of the Pretoria Group.

2.8. Local Geology ? the Brits area

The De Wildt area is situated approximately 5 km east of the Brits graben (Figs. 1 and 5). The
graben is the main structural lineament within the immediate surrounding area and extends from the
Crocodile River fragment in the NNW to the edge of the Johannesburg Dome in the SSE (Wilhelm et
al., 1997). The graben contains sub-graben blocks that are bounded by the steeply dipping, NNW-
trending Bokfontein and De Kroon faults (Judeel and Hartmann, 2008), with normal block
displacements in the order of 600 m. The trend and dip of igneous layers in the graben varies from one
block to the next. Several fault-controlled syenite and dolerite dykes crosscut the graben (Lomberg et
al., 2006). Iron-rich pegmatoids (IRUPs) and potholes have been mapped SE of the De Wildt prospect
in the graben in the Elandsfontein property (Tiegler et al., 1992; Scoon and Mitchell, 1994; Lomberg
et al., 2006) as indicated from aeromagnetic data and disturbed UG2 chromitite and Merensky reefs
intersected in boreholes.
Published data for the UG2 chromitite and Merensky Reef in the region of the De Wildt project
is limited and includes Maier and Bowen (1996) citing Moen (1989) who documented the UG2, and
Butcher and Merkle (1991) who studied with Upper Zone rocks. Maier and Bowen (1996) citing Moen
(1989) recorded that the UG2 chromitite layer on the Schietfontein farm overlies a large meta-
sedimentary footwall block, but did not investigate the cumulate sequence between the UG2 and the
base of the complex. It is thus not known if the sequence in the De Wildt region is different from that
present at Rustenburg and Amandelbult in western BC. Maier and Bowen (1996) reported that the
UG2 hanging-wall pyroxenite was separated from the UG2 chromitite by a 15 m norite layer in the
Schietfontein farm portion. Lomberg (1991) reported that the UG1 chromitite layer is poorly
developed to the east of the Zilkaatsnek 439-JQ farm portion (Fig. 2) and consists of numerous thin
chromitite lenses, with no development of the Lower Zone.
Maier and Bowen (1996) noted that in the western BC, the UG2-Merensky Reef interval is
situated about 100 m below the Critical Zone-Main Zone boundary. However, the UG2-Merensky
Reef relationship east of Brits is not known as a result of poor exposure and a large number of faults.
Roberts (1992) documented that in the area bounded by Rosslyn, Hartbeespoort Dam and Pretoria the
cumulate succession of BC, which includes the De Wildt project area, is broadly similar to other BC
localities, except for the absence of Lower Group chromitites, the appearance of cumulus plagioclase
below the Middle Group chromitites and an exceptional thick Merensky Reef (12 m). The western
limb of the BC has been shown to thin eastwards towards Pretoria (Cawthorn and Webb, 2001).

33
Further to this, the 2527 Brits Geological Series map records the presence of dykes and faults
throughout the Brits area and within the De Wildt prospect (Chief Directorate: Survey and Mapping,
1996). Faults are less pronounced away from the graben and towards the east. A NE-trending dyke
intrudes the Magaliesberg Quartzite hills about 5 km south of the De Wildt prospect. A NE-trending
sinistral (apparent) strike-slip fault also occurs south of the De Wildt prospect, with a calculated
sinistral displacement of approximately 250 m. Quartzite outcrops occur in the centre the prospect,
while two parallel ridges of the Magaliesberg Quartzite occur on the eastern margin of the prospect.
With respect to radiometric data over the De Wildt area, Antonie (2004) reported that the
spectrometer uranium channel is flat while those of thorium and potassium showed disparate
anomalies. Zones of increased gamma radiation in the K channel correlated with areas where floor
rocks were mapped and thus could be used to indicate potential inliers.
34

Table 2: Simplified stratigraphy of Transvaal Supergroup and age data (modified after Eriksson et al. (1993) and
Dankert and Hein (2010).
(1) Burger and Coertze (1973-1974; Rb-Sr whole rocks age).
(2) Martin et al. (1998; single zircon U-Pb SHRIMP).
(3) Walveren and Martini (1995; single zircon Pb-evaporation).
(4) Unpublished; vvvvv = volcanic or pyroclastic unit.
Group Formation Age (Ma) Lithology

Loskop 2060 (U-Pb) Mudrock sandstone
Rooiberg Selonsrivier and Damwal 2090 Felsites
Dullstroom,Leeuwpoort 2089+/-15 (Rb-Sr) Mafic & felsic lavas clastic sediments

Rayton/Woodlands
Mudrock sandstone pyroclastic
carbonate rock
Magaliesberg sandstone
Silverton Mudrocks volcanics carbonate rock
Dasport sandstone
Pretoria Strubenkop Mudrock sandstone
Dwaalheuwel Sandstones conglomerates
Hekpoort 2224 +/-21 (Rb-Sr)1

Basaltic andesites
Boshoek Conglomerate sandstone
Timeball Hill 2263 (Rb-Sr) Mudrocks sandstone tilloid
Rooihoogte Breccias conglomerates sandstone

Duitschland
Penge 2432+/-31 (U-Pb SHRIMP) Iron Formation
Frisco Dolomite chert
Chuniespoort Eccles Dolomite chert
Lyttelton Dolomite chert
Monte Christo Dolomite chert
Oaktree
2550? 3 Ma4(Single zircon Pb-evap)3
2583 ? 5 Ma2(U-Pb SHRIMP)2
2588 ? 7 Ma2(U-Pb SHRIMP)2

Black Reef Black Reef Sandstone/conglomerate

Buffelsfontein Sandstone mafic felsic lavas mudrock
Godwan Group Sandstone lavas
Pre-Transvaal Wachteenbeertje Fm. Mudrock carbonate sandstone
Woikberg Group Mudrock sandstone basalts
Upper Groblersal Group Sandstone lavas
Lower Groblersal Group Metamorphic rocks
35
CHAPTER 3

METHODOLOGY

3.1. Research approach

This research attempts to investigate and establish the structural setting of UG2 and Merensky
Reef;

? Photogrammetry: Black and white aerial photographs were combined into a photographic mosaic
for the De Wildt region and interpreted. This was overlain with Google Earth imagery. The
interpretation was validated with surface mapping to establish the position of faults and dykes.
? Aeromagnetic data (1st vertical magnetic derivative and total magnetic intensity, 2nd vertical
magnetic derivative, Fig. 8; see Appendix) yielded insight into the underlying geology by depicting
magnetic horizons and anomalies. A magnetic interpretation map was compiled.
? Geological data from diamond drill core and surface plans was incorporated to provide a
comprehensive model for the litho-stratigraphic and structural setting of the Merensky Reef and
UG2 chromitite layers.
? Petrographic and mineralogical studies of selected samples from boreholes were investigated to
establish lithologies (Fig. 3, borehole map).
? The strato-structural setting of UG2 chromitite and Merensky Reef in the De Wildt area was
correlated and compared with the strato-structural setting of the UG2 chromitite and Merensky
Reef in the adjacent Rustenburg facies to the west of the De Wildt project area to establish
similarities and differences in the nature of the reefs, and to develop an overall model for reef
variation.
? An interpretive map based on combined photogrammetric, magnetic, topography, borehole data
and historic geology for the De Wildt area was compiled.

3.2. Photogrammetry

Black and white digital aerial photographs (rasters) and a flight plan of the study area were
obtained from the photographic database at the University of the Witwatersrand Johannesburg. The
1:36 000 aerial photographs were captured on 14-25 July 1961 using a Wild 15 AG 39 No.296 camera
or lens, with a focal length of 152.37 mm and in a 9 x 9 inches format. From these images, a
photographic mosaic was created using Arc-GIS version 9.0. Geo-referencing of the raster images was
completed using georeferenced topo-cadastral sheet 2527DB Brits (Chief Directorate: Survey and
Mapping, South Africa, 1996) in which the study area is located.
36

Figure 8. The second vertical magnetic derivative map across the De Wildt project (Antoine, 2004). Structures such as dykes and faults, which disrupt the mafic layers of BC,
were discerned from this plan. The De Wildt project area is marked in blue.

3.3. Aeromagnetic data

A detailed aeromagnetic survey over the De Wildt area was flown on 13-14 November 2003 by
Fugro Airborne Surveys (Pty) Ltd on behalf of Nkwe Platinum (SA) (Pty) Limited, using a horizontal
gradient mode magnetometer mounted on a Bell 206 helicopter. A total of 1012 line kilometres were
acquired using a line spacing of 50 m. A sensor to ground clearance of 20 m was used for both the
magnetometer and gamma ray spectrometer data. A north-south line direction was used with
orthogonal control lines flown at an interval of 500 m. The magnetic data was sampled at 10 Hz, while
spectrometer and navigation (GPS) data were acquired at 1 Hz. Additional survey specifications are
given in Table 3. Total magnetic intensity data was enhanced using a second vertical derivative filter
that preferentially sharpened the magnetic signatures of shallow sources. The resulting 2nd vertical
magnetic derivative map was interpreted and found to be useful in delineating the extent of various
lithological units, and dykes and faults masked by overburden.
The data from the aeromagnetic plan was extracted using the methodology of Hammond (1998);
? ?when a number of essentially parallel units can be discerned, a relatively intact block of layered
rock might have been imaged??.
? ?where two magnetic units cross each other with little effect, an interpreter is limited to relatively
few geological relationships that have this pattern (e.g., joint sets, dykes)?.
? ?unless a given discontinuity is a major structure with a very large displacement, then a chance
exists that some correspondence between units on either side of the fault can be discerned?

38
Mitchell and Manthree (2002) documented the occurrence of a high iron-rich zones associated with
the transition from Critical Zone to Main Zone of the BC. Ashwal et al. (2005) detailed that
anorthositic rocks, particularly within the Main Zone, show higher magnetic susceptibility than any
other polyphase cumulates as a result of the presence of intercumulus magnetite and dust-like
inclusions of magnetite. Thus, it has been possible to separate the Giant Mottled Anorthosite (GMA)
from other mafic rocks in the region.

3.4. Mapping and logging

Geological surface mapping was conducted throughout the project area as delineated by the
farm boundaries. Sites of outcrop were measured and recorded using a GPS (Garmin) instrument
(UTM, WGS 84). The mapping exercise involved general mapping principles that included rock
description and identification, structural measurements such as dip and strike of bedding, layering and
banding. A geological surface plan was compiled using the collected data.
All diamond boreholes drilled by Nkwe Platinum were logged at the core-shed at De Wildt. The
logging involved recording borehole depth, lithological descriptions according to BC standards
(SACS, 1980), description of mineralization, alteration and structures such as faults, fractures and
joints. Also, photographs of reef intersections and other lithologies of interest were made. The data log
for each borehole was captured into several ExcelTM spreadsheets and subsequently exported into
Golden Software GrapherTM (version 5) from which stratigraphic columns were compiled (Figs. 9-11).
Grapher TM together with Corel DRAWTM (version 9), were used to compile diagrams and illustrations.
The construction of UG2 chromitite isopachs and isochons used Golden Software SurferTM. Drill core
data from historic logs (SK boreholes) was presented to Nkwe Platinum in the form of manuscripts
which were then captured into ExcelTM spreadsheets after relogging. The data was later incorporated
into the Nkwe Platinum database and used for analysis and interpretation.
The geological data obtained from more than 40 diamond holes (inclusive of historical
boreholes) drilled in the project area was used to investigate the structural setting as well as the
geology of the Merensky Reef and UG2 layers. Cross-sections of the project area were drawn to
interpret structures such as faults and dykes that displace the Merensky Reef and UG2 reefs, and also
to establish the lateral continuity of the Merensky Reef and UG2 layers across the project area.
The UG2 isopach and isochon maps of UG2 chromitite data from boreholes were drawn and
used to depict the variation of the UG2 chromitite layer across the project area in terms of width and
grade. The investigation of the PGE value distribution along the UG2 chromitite reef, based on assay
results, was also conducted. This helped delineate the portions of the UG2 chromitite reef that were
enriched in PGEs. In addition, prill splits, which are a percentage break down of the individual PGE
constitution within the UG2, was calculated from assays of the UG2 chromitite samples. Attention was
paid to establishing the Pt:Pd ratio of the UG2 chromitite reef at De Wildt prospect. The Pt:Pd ratios
39
were chosen as these are commonly calculated for the BC by mining and exploration companies in
South Africa and allowed comparisons to published Pt:Pd ratios from various localities within the BC.

3.5. Petrography and mineralogy

More than 35 samples of NQ drill-core (not orientated) were taken from boreholes that had features of
interest and not normally encountered in the BC mafic rocks. The core was split using a Costor core-
splitter at Nkwe Platinum?s core yard to obtain half core samples of between 10-20 cm lengths. The
samples were labelled and bagged, and details of the depths at which the sampling took place were
recorded. The samples were submitted to SGS Laboratory in Booysens Johannesburg for preparation
of polished thin sections and these were used for mineralogical and micro-structural analysis.

Figure 9: Schematic representations of borehole logs drilled in the De Wildt project area by Nkwe Platinum. The boreholes were used in the study to characterize the
distribution and continuance of the Merensky Reef (MR) and Upper Group 2 chromitite (UG2) layers.
41

Figure 10: Stratigraphic logs drawn from historical diamond boreholes that were re-logged and used in the study (MR=Merensky Reef, UG2=Upper Group 2 chromitite and
UG1=Upper Group 1 chromitite layer).
42

Figure 11: Schematic representation of logs of boreholes drilled by Nkwe Platinum in the western portion of the
De Wildt project area (MR=Merensky Reef, UG2=Upper Group 2 chromitite layer)
43

CHAPTER 4

REMOTE SENSING & FIELD INVESTIGATION RESULTS

4.1. Surface Mapping

The De Wildt project area is covered by black turf, with minimum exposure of the Rustenburg
Layered Suite. Limited outcrop of the Bushveld Complex rocks occur along the Magaliesberg
mountain ridge, which is situated at the southern boundary of Visserhoek, Schietfontein and
Zilkaatsnek farm portions (Figs. 2, 12). The Magaliesberg mountain ridge consists of sandstone (and
quartzite) of the Magaliesberg Formation of the Transvaal Supergroup. The southern part of project
area is generally marked by relatively fine-grained norite units that are tentatively correlated with the
Marginal Zone of the BC. Fine to medium-grained norite and gabbronorite boulders subcrop in the
northern part of the prospect and define the Main Zone in the De Wildt project area. Rugged hills to
the far north characterise Main Zone gabbronorites. Norite (leuco-norite) units (with a strong NW
trending mineral lineation defined by plagioclase crystals) and scattered pyroxenite subcrop were
confined to a narrow zone in the centre of the study area. This zone is interpreted to overlie the Critical
Zone of BC (Fig. 12). The rocks in the study area generally trend E-W and dip 18-28?N.
In the east of the study area, pyroxenite and melanorite units of the BC crop out and contact the
Magaliesberg quartzite floor rocks. The quartzite units trend E-W and dip approximately 28? N. The
contact between the BC and Transvaal metasedimentary floor rocks is marked by recrystallized and
contact metamorphosed quartz veins.

4.2. Photogrammetry

In the photographic mosaic, the east-west trending Magaliesberg Quartzite units are readily
defined in the south of the study area (Fig. 13). The Main Zone Hills, which are composed of gabbro-
norite, lie to the north. Settlements that make up the township of Ga-Rankuwa are situated north and
northeast of the study area. Two ridges of quartzite encroach into the eastern portion of the study area.
Northeast-trending lineaments crosscut the study area east and west of the Main Zone Hills. Numerous
NNE and NNW-trending master joints crosscut the Magaliesberg Quartzite in the south of the project
area. A railway line that traverses east-west along the northern boundary of the study area, and local
roads, are also evident in the mosaic. Farm portions are also partially discernable within project area
and throughout the surrounding areas. Open cast pits occur west and east of the De Wildt project area.
44

Figure 12: Schematic map that combines interpretations of black and white aerial photographs and Google Earth images of the De Wildt project area (blue) with those of
topography and 1st and 2nd vertical derivative aeromagnetic data. West-trending lineaments are evident in the BC mafic units, and these are crosscut by an array of NE and NW-
trending faults and two dyke sets. Displacement on the Schietfontein, Zilkaatsnek and Lindiwe fault is sinistral (apparent) but the faults are most likely scissor faults with
normal displacement and west block down. The BC mafic units dip north.
45

46
4.3. Aeromagnetic Interpretation

The 2nd vertical derivative aeromagnetic map for the De Wildt region (c.f. Fig. 8) was
interpreted using the methodology of Hammond (1998). North-south orientated lineaments were
ignored to avoid flight line stitch mismatch (flight lines were north-south). Three distinct zones (i.e.,
western, central and eastern zone) were defined on the basis of lithological character and fault/fracture
patterns. The zones are bounded by NNE-ENE trending faults that are hereafter termed the Zilkaatsnek
and Schietfontein faults (Fig. 12). The position of the Schietfontein fault coincides with the position of
NE-trending fault on the 2527 Brits Geological Series map of the De Wildt region and a NE-trending
lineament that passes east of the Main Zone Hills (Figs. 13, 12). A NE-trending fault, termed the
Lindiwe Fault, can be interpreted from topographic data.
NNW-trending magnetic highs (shown in dark green lines in Figs. 14, 12) are interpreted as
mafic dykes that crosscut all BC mafic units and the Zilkaatsnek and Schietfontein faults. Based on
crosscutting relationships the dykes were emplaced after the BC mafic units and after deformation
associated with fault formation.
The western zone covers the Zilkaatsnek farm portion and is marked by relatively undisturbed
E-W trending linear magnetic highs and lows that are interpreted as mafic units of the BC. Within
these units, the Giant Mottled Anorthosite (GMA) is resolved as two distinct parallel high-low
couplets, as marked by two yellow lines in Figure 14. The couplets are interpreted as the boundary
between the Main Zone and Critical Zone in the De Wildt area. An anorthosite unit is also interpreted
higher up (and northward) in the Main Zone stratigraphy (on the top edge of the magnetic image, Fig.
14).
The central zone is characterized by folded mafic units of the BC. NNW and NE-trending
lineaments that crosscut the mafic units are interpreted as fractures and joints. The Zilkaatsnek fault
(apparent) displaces the Giant Mottled Anorthosite unit by a maximum of approximately 150 m in an
(apparent) sinistral sense, but displacement is unequal along the length of the fault being lesser in the
northeast and greater in the southwest. The Zilkaatsnek fault is thereby interpreted as a scissor fault
that is hinged in the northeast. However, the kinematic of the Zilkaatsnek fault have yet to be
established.
The eastern zone underlies the Krelingspost Farm and part of the eastern portion of
Schietfontein farm. The zone is bounded by the NE-trending Schietfontein Fault and is crosscut by NE
and NNE-trending magnetic lineaments (low) that are interpreted as mafic dykes (termed the Older
Dykes). The Older Dykes are crosscut by NW-trending dykes and were thus emplaced before them.
The Schietfontein Fault has a calculated sinistral displacement of approximately 300 m, but again, is
hinged in the northeast and displacement is unequal along the length of the fault. The Schietfontein
fault is thereby interpreted as a scissor fault.
47

Figure 14: Geological interpretation of the magnetic signatures from the 1st and 2nd vertical aeromagnetic data. The Central Zone shaded in grey is highly disturbed and
characterized by faults and fractures. NW-trending dykes traverse the prospect (interpreted from aeromagnetic data from Antoine, 2004).
48
Furthermore, the average dip of BC mafic units across the Schietfontein and Zilkaatsnek faults
(i.e., from east to west) decreases from ~28?N to ~18?N suggesting that the western blocks were
displaced normally rather than laterally; the surface projection of shallowly north-dipping BC mafic
units in the western zone (relative to the eastern zone) resolves an apparent sinistral sense across the
faults.
The BC mafic units in the eastern zone are discontinuous, with numerous interpreted fractures
or joints. A series of subordinate NE and NW-trending magnetic lineaments transects the De Wildt
region and weakly displace the BC mafic units and NW-trending lineaments (dykes). They are
interpreted as mesoscopic faults.

4.4. Radiometric Survey

The company, Nkwe Platinum, commissioned a high-spatial resolution airborne magnetic and
spectrometer survey over the De Wildt project area in November 2003. Gamma-ray spectrometer data
for potassium (K), thorium (Th) and uranium (U) were made available to Nkwe Platinum. The
interpretation of the radiometric data is shown in Fig. 15. Very high K-Th levels were registered in the
south of the study area. The western and central zones were characterised by predominately high Th
levels, while the eastern zone was characterised by moderate K and Th: the latter approximately
coincides with ridges of the Magaliesberg Quartzite in the east of the De Wildt project area. The zone
of high K and Th to the south of the prospect does not coincide with the east-west trending
Magaliesberg Quartzite Mountain ridge of the Transvaal Supergroup: its meaning needs further
investigation.

Figure 15: Interpretation map of Potassium and Thorium radiometrics across the project area. Zones of high Th-K radiation occur in the central and western parts of the De
Wildt Project area (interpreted from aeromagnetic data from Antoine, 2004).

50
CHAPTER 5
STRATIGRAPHY OF DE WILDT AREA

Boreholes drilled by Nkwe Platinum were collared in the Main Zone and Upper Critical Zone of
the BC in the De Wildt project area. The Upper Critical Zone becomes increasingly disturbed towards
the eastern parts of the project. The Lower Zone is absent. Fine-grained norite with abundant
metasedimentary xenoliths forms the Marginal Zone in the De Wildt project area.

5.1. Character of the Main Zone

In the De Wildt project area, the Main Zone is characterized by sequences of gabbronorite,
norite, and anorthosite, with a maximum thickness of 640 m (Fig. 16). Two mottled anorthosite layers
occur about 250 m above the Giant Mottled Anorthosite (GMA) within a sequence of norite. In this
study the GMA has been used as stratigraphic marker delineating the Main Zone-Critical Zone
boundary. The base of GMA thus marks the top of the Critical Zone (Mitchell, 2002). The GMA could
be traced across the prospect from the west to the east and hence was used as a marker horizon.
However, it should be borne in mind that in some literature, the MZ/CZ boundary is taken at the base
of the Merensky Reef (Kruger, 2005).

5.2. Character of the Critical Zone

The Merensky Reef at De Wildt is located at between 40 m and 180 m below the GMA, i.e., 40-
180 m below the Main Zone. The (true) distance between the GMA and the Merensky Reef increases
from the west to east. To the west of the project area at Zilkaatsnek Farm and in the far-western
boundary of the Schietfontein Farm, the Merensky Reef pyroxenite is located approximately 40 m
below the GMA, while in the east of Schietfontein and Krelingspost farms, it is 105 m to 180 m below
the GMA. The Bastard Reef unit is not developed.
The Merensky Reef pyroxenite was correlated from borehole to borehole throughout the project
area and thus formed an important marker horizon in the CZ. The Merensky Reef-UG2 interval was
dominated by a sequence of leuconorite, anorthosite and mottled anorthosite with gradational contacts.
In general, the Merensky Reef-UG2 vertical separation ranged from 114 m to 188 m (Table 4, Fig.
17). However, in boreholes PDW02 and DWD09, the Merensky was intersected at greater depths than
expected such that the Merensky Reef-UG2 stratigraphic interval was significantly reduced in
thickness to 29.57 m and 16.38 m respectively. In borehole DWD13, the Merensky Reef pyroxenite
overlay a feldspathic pyroxenite with disseminated chromite and chromitite slivers instead of a UG2
chromitite layer, while in borehole DWD12 the Merensky Reef pyroxenite was not intersected.
Instead, the interval hosted IRUPs.
51

Figure 16: Provisional local stratigraphy of De Wildt prospect. The base of the GMA is interpreted as the
boundary between the Main Zone and the Critical Zone. The vertical separation between the Merensky Reef and
UG2 chromitite layer ranged from 114 m to 188 m.

52
With respect to the UG1, the layer was intersected about 128 m below UG2 layer in borehole
SK03 in the far west of Zilkaatnesk Farm, but drilling at Schietfontein and Krelingspost farms did not
intersect the UG1 below the UG2 reef, despite drilling to depths of 190 m and 250 m. Rather, a
sequence of fine-grained quartz-norite and quartzite (often intruded by dolerites) was intersected.
Furthermore, boreholes drilled at De Wildt did not intersect the Lower Critical Zone and Lower
Zone. Instead, they intersected a thick sequence of fine-grained melanorite which alternated with
quartzite and norite-quartzite hybrid rocks, and metasedimentary rocks. In borehole PDW4 (Fig. 18), a
large calc-silicate xenolith overlay a poorly developed UG2 layer. The UG2 reef had a very poor PGE
content.

Table 4: The UG2-Merensky Reef interval in the De Wildt project area. The data records the disturbed
Merensky Reef-UG2 interval from the boreholes where layers were disrupted by potholing. Faults were omitted.

5.3. Character of the Marginal Zone

A zone consisting of very fine norite, similar to the chill-margin norite at the base of the BC
rock succession, was intersected in several boreholes (Table 5). The norite hosts abundant
metasedimentary xenoliths (mostly quartzite) and hybrid rocks. The norite, together with abundant
metasedimentary xenoliths and hybrid rocks, is similar to the Marginal Zone documented by Vermaak
(1976) and Cawthorn et al. (2006) and is thus tentatively classified as the Marginal Zone (MgZ) in the
De Wildt project area. The metasedimentary units (Transvaal Supergroup metasedimentary sequences)
are interpreted as blocks that were lifted off the floor during the emplacement of the BC.
Furthermore, numerous fine-grained magnetic melanorite units in the MgZ were intersected in
borehole DWD10. The units are characterized by discontinuous magnetite stringers and blebs.
Intercalations of quartz dominate the MgZ in boreholes DWD10 and DWD11 (Figs. 19-20). In
borehole DWD12, the UG2 layer effectively overlies the MgZ, being separated from it by coarse-
grained meso-norite (4.16 m in thickness) that hosts chromitite blebs and slivers. The MgZ is crosscut
by numerous diabase dykes and sills, some of which are magnetic.

Borehole UG2-Merensky Reef interval (m)
DWD06 151.90
DWD07 164.22
PDW01 114.56
SK02 188.61
SK03 170.66
SK06 132.63
53

Figure 17: Schematic borehole cross section depicting of the separation between Merensky Reef and UG2 across the De Wildt project area. In the intersections in which the
UG2 reef was multi-layered, the top reef contact of the first UG2 chromitite unit was used for correlation. Pothole interpretations were based on transgressed/missing footwall
layers, thickened hangingwall layers of MR, steep dips and overall stratigraphic attenuation of the MR-UG2 interval without any discernable evidence of faults.

Figure 18: A calc-silicate xenolith that is rich in epidote (shown in red line) within a BC norite sequence in
borehole PWD4. Part of the disturbed UG2 footwall with flow structures as defined by chromite and magnetite
stringers is delineated by the dashed green lines.

BHID Marginal Zone thickness (m)
DWD09 110.68
DWD10 188.43
DWD11 221.44
DWD13 177.09
DWD12 16.61

Table 5: Thickness of the Marginal Zone intersected in several boreholes. The boundary between the Critical
Zone and Marginal Zone is tentatively taken at the level of the first appearance of the fine-grained ?chill-margin?
contact-type norite.
55

Figure 19: Diamond drill core of sedimentary xenoliths (in yellow) in core of the BC within Marginal Zone of
the BC at De Wildt project. Below the xenoliths is part of the Transvaal Supergroup quartzite intersected in
borehole DWD11.

Figure 20: Poorly developed UG2 chromitite layer intersected in borehole DWD10. The UG2 is underlain by
metasedimentary xenoliths (borehole DWD10 intersection).

Poorly developed UG2 seam

5.4. Merensky Reef at De Wildt

The Merensky Reef was intersected in boreholes drilled mainly on Zilkaatsnek 439 JQ and
Schietfontein Farm portions and includes boreholes PDW01, PDW02, PDW03, PDW04, DW06,
DWD07 and DWD09 drilled by Nkwe Platinum (Table 6). In most borehole intersections, the
Merensky feldspathic pyroxenite was jointed, moderately chloritized and serpentinized. The measured
width of normal Merensky Reef ranged from 11.6 m to 20.5 m, while that of possibly potholed
Merensky Reef was typically less than 5 m. The reef consists of a medium to coarse-grained
feldspathic pyroxenite with a bottom chromitite stringer/layer of variable thickness (Fig. 21). The
feldspathic pyroxenite is characterized by distinct euhedral to subhedral orthopyroxene crystals that
constitute over 80% of the unit (Fig. 22a). Intercumulus plagioclase makes up approximately 20% of
the unit. Biotite is present as an accessory mineral. The bottom chromitite stringer/layer ranged from
0.2 cm to 2.0 cm in thickness. The contact with the overlying Merensky pyroxenite was often very
irregular.
In all boreholes studied, the upper contact of the Merensky Reef was gradational through
feldspathic pyroxenite into melanorite and anorthosite. The gradational upper contact was marked by
disseminated sulphide, consisting predominantly of pyrrhotite and chalcopyrite. In contrast, the
immediate footwall to the Merensky Reef was composed of norite, anorthosite (mottled/poikilitic) or,
pyroxenite of variable thickness (Figs. 22b-d; 23).
In borehole PDW04, the Merensky Reef has a well-defined basal chromitite layer about 2.0 cm
in thickness, while the upper portion was marked by a relatively narrow zone of scattered sulphide
(blebs). In boreholes PDW02 and PDW01, the Merensky Reef basal chromitite stringer/layer was not
well developed. Instead, the basal contact was defined by irregular and diffuse chromitite stringers
with minor chromite blebs and disseminated chromite over a 1.0-2.5 cm interval. Disseminated
sulphide mineralization was apparent in the Merensky Reef intersected in borehole PDW01 near the
upper contact of the Merensky reef pyroxenite. Here, sulphide extended upward for 50 cm into the
hangingwall norite. Sulphides were not found in the footwall. However, in borehole SK06 the
anorthosite (mottled/poikilitic) in the immediate footwall to the Merensky Reef hosted disseminated
sulphides.
57
58

Figure 21: Typical Merensky Reef intersection of greater than 20 m thickness at De Wildt together with schematic of the borehole (PDW01). The MR is altered with
chlorite and serpentinite filled joints.
59

Figure 22: Merensky Reef pyroxenite and the different footwall units intersected in borehole. Image a is of a
medium crystalline feldspathic pyroxenite layer making up the MR. Image b, is of a poorly defined basal
chromitite stringer that defines the bottom contact of MR. The MR pyroxenite is weathered and chloritized.
Image c, shows a well developed basal chromitite stringer of MR, the undulating contact between the chromitite
stringer and the footwall anorthosite is defined. Image d, shows a half NQ core section of an irregular basal
chromitite stringer of MR in contact with mottled anorthosite footwall.

Hangingwall
Footwall
Chromitite stringer
Merensky
Reef
Zone of disseminated
sulphides
Norite
Anorthosite (mottled)
Norite
medium to coarse grained
crystalline feldspathic pyroxenite
Anorthosite (mottled)
(11.5 -20.5m)
(20mm)
(0.50 -1.10m)
(0.10 -2.0m)
(0.40 -18.0m)
(0.10 -2.5m)
(1.5 -6.0m)
Gradational top contact
MR facies with a basal Cr stringer
Hangingwall
Footwall
Merensky
Reef
Zone of disseminated
sulphides
Norite
Norite
medium to coarse grained
crystalline feldspathic pyroxenite
Anorthosite (mottled)
(11.5 -20.5m)
(0.50 -1.10m)
(0.10 -2.0m)
(0.40 -18.0m)
Gradational top contact
MR facies with no basal Cr stringer
Hangingwall
Footwall
Merensky
Reef
Zone of disseminated
sulphides
Norite
Norite
medium to coarse grained
crystalline feldspathic pyroxenite
Anorthosite (mottled)
(~5.0 m)
(0.50 -1.10m)
(0.10 -2.0m)
(0.40 -18.0m)
Gradational top contact
Potholed MR faciesa b c

Figure 23: A schematic of the different Merensky Reef facies intersected at De Wildt. In (a) the MR is characterized by a medium crystalline feldspathic pyroxenite that has a
basal chromitite stringer. In (b) the MR is wide but devoid of a basal chromitite stringer. In the potholed facies (c), the MR pyroxenite is significantly thinned (in this case
down to ~5.0 m).
61
5.5. UG2 Chromitite layer at De Wildt

The UG2 chromitite layer intersected in boreholes in the De Wildt project area is highly
variable in character; it is multi-layered, poorly developed (Table 7) and is not overlain by the triplets
or leader seams. The UG2 layer also displays a lack of lateral continuity. The chromitite underlies the
Merensky Reef at a (true) depth range of 114 to 188 m in boreholes and is composed of ~ 90 %
chromite by volume and 5-10% plagioclase. Pyroxene, magnetite, pyrrhotite, chalcopyrite and pyrite
form accessory minerals. Magnetite occurs in the form of discontinuous stringers and lenses.
The UG2 chromitite layer is also crosscut by quartz-biotite veins and fractures and is generally
altered, often with magnetite lenses and blebs in places (Figs. 24a-b). However, in boreholes collared
on the west of the project area, diffuse pyroxene crystals enclose chromite grains and give the UG2
reef a poikilitic texture. The thickness of the reef varies from 0.40 to 1.25 m.
The UG2 chromitite dips generally dips at 19-25? as measured from the core-bedding angle
using a clino-ruler, however, is steeply dipping (about 60?) at Krelingspost Farm. The relationship is
consistent with block rotation along the Schietfontein Fault (c.f. Section 4.2).

5.6. UG2 Chromitite facies

Two main facies were noted in the UG2 chromitite layer; the ?A type? normal facies and the
split-reef facies (Fig. 25). The split reef facies is characterized by a UG2 chromitite layer that has
partitioned into two or more chromitite units. The chromitite units are separated by an intervening
chromitiferous norite or feldspathic pyroxenite layer. The split reef facies can be further divided into
four sub-facies depending on the number of massive chromitite seams developed including;

?B type? - consists of two discrete chromitite units.
?C type? ? consists of three discrete chromitite units.
?D type? - consists of one chromitite unit underlain by a chromitiferous pyroxenite/norite.
?E type? ? consists of chromitiferous pyroxenite, devoid of massive chromitite units.

The ?A type? normal reef facies occurs only in the western zone of the De Wildt project area,
while split-reef facies occurs in the central and eastern zones (Fig. 26). The distribution of the split-
reef facies throughout the project area does not follow a well defined pattern and as such, a closely
spaced drill program was needed to constrain the UG2 chromitite reef setting.
62

Table 7: UG2 chromitite reef intersections within the De Wildt project area
63

Figure 24: (a) UG2 seam intersected in DWD07, the seam is cut by quartz-biotite veins and marked by a
generally high content of interstitial plagioclase. (b) Altered UG2 seam intersected at a depth of > 604 m, the top
reef contact is poorly developed, marked by vague plagioclase lenses and anorthosite patches. The footwall
pyroxenite has been heavily altered to chlorite giving the rock a dark green appearance.
a
b
64

Figure 25: Schematic depiction of boreholes in the De Wildt project area showing the UG2 facies types in each
borehole. Normal and split-reef facies occur in the area but the split-reef facies can be further subdivided into B,
C, D, and E types.
65
92000 93000 94000 95000 96000 97000
-2837500
-2837000
-2836500
-2836000
-2835500
-2835000
normal
B type
C type D type
E type
D type
B type
normal facies
C type
E type
split reef sub-facies
N

Figure 26: UG2 facies distribution in the De Wildt project area.

The normal facies ?A type? was intersected in boreholes SK02, SK04 and SK07 at the Zilkaatsnek
Farm and consists of a 1.10-1.25 m massive chromitite layer. The chromitite unit is fine-crystalline
with minor sulphide. The UG2 leader seams are absent or poorly developed in borehole SK04 (Fig.
27a). In borehole SK02, the massive chromitite layer is overlain by a 1.0 cm thick chromitite stringer
about 6 cm into the hangingwall norite. It is possible that this stringer may be part of the UG2 leader
seams. In boreholes SK04 and SK07, weakly disseminated chromite was present in the immediate
overlying hangingwall of the UG2 chromitite seam. As such there is a possibility that the disseminated
chromite may be part of the remnants of the triplets that are poorly developed. In the case where the
leader seams were not developed at all, it is suggested that the leader seams coalesced with the UG2
layer. Such coalescence may have contributed in the widening of the UG2 chromitite layer or package.
The ?B type? split reef sub-facies was intersected in boreholes SK05, SK06, SK08, DWD09,
DWD10, DWD12, PDW01, PDW04 and PDW02 and consists of an upper and lower chromitite layer,
that is separated by a chromite-rich pyroxenite or norite unit (Fig. 27b). The upper chromitite layer is
59-84 cm thick while the lower layer from 8-42 cm thick. However, in the east Krelingspost Farm, the
B? type facies consists of relatively narrow upper and lower chromitite layers; 11-15 cm thick for the
upper layer and between 14-32 cm for the lower layer as intersected in boreholes DWD10 and
DWD12.
The upper chromitite layer is poikilitic and characterised by pyroxenite lenses and irregular
feldspar veins, whereas the lower chromitite layer is fine-grained and hosts norite lenses (Figs. 28-29).
The poikilitic texture of the upper chromitite layer of the UG2 could be the result of crystallisation of
orthopyroxene (part of the UG2 pyroxenite) around and enclosing the pre-existing chromite crystals
during chromitite formation.
The chromite-rich feldspathic pyroxenite separating the chromitite layers ranges from 8 cm to
63 cm in thickness. However, in borehole DWD09, the chromitite layers are separated by a 1.3 m
thick, coarse-grained norite with irregular chromitite stringers. The norite hosts a 14 cm thick
anorthosite layer.
The ?C type? split reef sub-facies was intersected in boreholes DWD07 and SK09 and consists
of three chromitite layers/units separated by either a chromite-rich pyroxenite or norite (Fig. 30). The
chromitite layers are variable in thickness and host norite and/or pyroxenite as internal waste. The
middle chromitite layer is typically narrow (8-12 cm in thickness). In these boreholes the upper
chromitite layer is 32-137 cm thick, while the lower layer is 31-92 cm thick. In borehole DWD07, the
three layers are separated by pegmatoidal mesonorite that coarsens towards the lower contact. The
mesonorite hosts irregular chromitite stringers/slivers and disseminated chromite. In borehole SK09,
the chromitite layers are separated by fine-grained porphyritic feldspathic pyroxenite with blebs of
chromite.
67

Figure 27: (a) UG2 normal facies. (b) The narrow B type sub-facies of UG2 split-reef facies.

150
100
50
0
de
pt
h
(m
)
UG2 Chromitite layer
SK04
109.53 m
110.63 m
feldspathic pyroxenite
UG2 pyroxenite 109.36 m
109.42 m
top soil
norite
pyroxenite
anorthosite
GMA
spotted anorthosite
UG2 chromitite
dyke
IRUP
fault zone
xenolith
quartzite
Legend
a
1050
1000
950
900
850
800
750
700
650
600
550
500
450
400
350
300
250
200
150
100
50
0
de
pt
h(
m
)
DWD12
UG2 Chromitite package
chromite-rich norite
UC
LC
999.90 m
999.98 m
1000.29 m
1000.61 m
1004.47 m
UC-Upper Chromitite unit
LC-Lower Chromitite unit
top soil
norite
pyroxenite
anorthosite
GMA
spotted anorthosite
UG2 chromitite
dyke
IRUP
fault zone
xenolith
quartzite
Legend
b
68

Figure 28: Upper chromitite layer of the UG2 package, and the chromite-rich feldspathic pyroxenite that
separates the two chromitite layers.

Figure 29: Part of the UG2 package showing the chromite bearing norite unit. The chromite occurs in the form
of blebs and irregular stringers. The lower chromitite layer in is poorly packed (intersected in borehole DWD12).

Figure 30: Stratigraphic logs showing the different UG2 split-reef. Figure (a) presents the C type sub-facies
intersected in borehole DWD07.Figure (b) presents the thick B type sub-facies intersected in borehole PDW01.
UG2 Chromitite package
PDW01
300
200
100
0
de
pt
h
(m
)
218.97 m
219.80 m
220.43 m
220.66 m
upper
chromitite unit
chromite-rich
feldspathic pyroxenite
LC
LC- Lower Chromitite unit
UC
UC-Upper Chromitite unit
top soil
norite
pyroxenite
anorthosite
GMA
spotted anorthosite
UG2 chromitite
dyke
IRUP
fault zone
xenolith
quartzite
Legend
607.68 m
608.00 m
608.75 m
608.83 m
608.90 m
609.82 m
upper chromitite unit
chromite-rich norite
middle chromitite unit
lower chromitite unit
700
600
500
400
300
200
100
0
de
pt
h
(m
)

top soil
norite
pyroxenite
anorthosite
GMA
spotted anorthosite
UG2 chromitite
dyke
IRUP
fault zone
xenolith
quartzite
Legend
(a) (b)
70

The ?D type? split reef sub-facies was intersected in boreholes SK03, SK10, DWD2A and
DWD06 (Figs. 31-32), and comprises a massive, but relatively narrow chromitite layer that is 15-55
cm in thickness. It is underlain by chromite-rich feldspathic pyroxenite. The contact between the
chromitiferous footwall and the overlying chromitite layer is defined by irregular chromite lenses
and/or stringers at scales of centimetres and less. The chromitite layer of the D type? sub-facies is
interpreted as coalesced leader seams (and or with part of the upper portion of the UG2 chromitite
layer), while the underlying chromitiferous footwall is interpreted as the remnants of the rest of the
UG2 chromitite layer.
The ?E type? split reef sub-facies of the UG2 Chromitite was intersected in borehole DWD13
(Fig. 33) that was collared in the central part of Schietfontein Farm. The facies is devoid of any
massive chromitite layers/units. Instead, it consists of a chromite-bearing feldspathic pyroxenite ~150
cm in thickness) that hosts highly disseminated chromite and irregular chromitite blebs and stringers.
The chromitite slivers are arranged randomly or, in a disordered fashion. The E-type sub-facies is
tentatively interpreted as the remnants of the UG2 reef within a major pothole because stratigraphic
layers are missing and the Merensky Reef-UG2 interval (in borehole) is attenuated. The pothole
arguably led to the transgression of the mafic units within the Merensky pyroxenite-UG2 reef interval
such that the Merensky pyroxenite rests directly on this chromite-bearing unit.

5.7. UG2 chromitite isopach

An isopach map of thickness of the UG2 chromitite layer is presented in Fig. 34. In the case of
the split-reef facies in which the lower part of the seam consists of disseminated chromite within a
feldspathic pyroxenite/norite, only the upper chromitite layers were used to construct the isopach map.
The units with disseminated chromite were classified as part of the footwall. The intersections in
borehole DWD13 are devoid of massive chromitite units, but consist of feldspathic pyroxenite that
hosts disseminated chromite; it was thus was excluded from the map.
For the isopach map, a UG2 layer thicker than one metre is confined to the western zone of the
project area, primarily in the Zilkaatsnek Farm. The central and southern parts of the project areas are
characterized by UG2 layer that is 0.0-0.75 m in thickness. The north of the prospect area is
characterised by UG2 reef which is 0.75-1.0 m in thickness, with isolated regions of greater than a
metre in thickness. In summary, the UG2 in the De Wildt project area is laterally variable and thins
towards the east and south.
71
700
600
500
400
300
200
100
0
de
pt
h(
m
)
DWD06
591.89 m
592.09 m
596.59 m
UG2 chromitite package
massive chromitite unit
chromite-rich
feldspathic pyroxenite
top soil
norite
pyroxenite
anorthosite
GMA
spotted anorthosite
UG2 chromitite
dyke
IRUP
fault zone
xenolith
quartzite
Legend

Figure 31: Borehole log showing the UG2 Split reef D-type sub-facies.

Figure 32: Narrow UG2 chromitite layer (a) and the upper UG2 chromitite layer with feldspar inclusions giving
it a spotted texture (b).

Figure 33: Chromite-rich feldspathic pyroxenite with chromite blebs immediately above the pencil (borehole
DWD13).

DWD06 DWD07
a b
73
UG2 Isopach
N
UG2 isopach (m)
0.0 - 0.5
0.5 - 0.75
0.75 - 1.0
1.0 - 1.25
> 1.25
92000 93000 94000 95000 96000 97000
92000 93000 94000 95000 96000 97000
-2837000
-2836000
-2835000
SK01
SK2
SK3
SK4
SK5
SK6 SK7
SK8
SK9 SK10
SK13
SK14
DWD 3
DWD 7
DWD 9
DWD 8
DWD 10
DWD 11
DWD 12
DWD 13
PDW1
PDW2 PDW3
PDW4
2098
2055
2092
DWD01
DWD02DWD02A
DWD03A
DWD04
DWD05
DWD06
SK11
SK12
SK15
SK16
SK17
SK18
ZK01

Figure 34: UG2 chromitite isopach map for De Wildt. Areas in which the mineralized footwall is not developed
have thicknesses less than 1.0 m.

74
5.8. UG2 hangingwall units

The UG2 chromitite layers in the De Wildt project area are predominantly overlain by coarse-
grained mesonorite layers whose thickness varies from 0.5 to 51 m (Table 8). The contact between the
UG2 and the overlying mesonorite is sharp or irregular, and in some instances may be gradational
through mesonorite, with increasing proportion of disseminated chromite. In some boreholes from
Zilkaatsnek Farm, the UG2 is overlain by a feldspathic pyroxenite unit which ranges from 0.05 m to 3
m in thickness. The UG2 pyroxenite intersected is feldspathic with minor biotite, while in boreholes
SK04 and SK13 the pyroxenite hosts disseminated chromite minerals and minor sulphide
mineralization (in borehole DWD06).

5.9. UG2 footwall units

The UG2 Chromitite layer is predominantly underlain by a pegmatoidal feldspathic pyroxenite
that is less than 1.0 m in thickness. The pegmatoidal pyroxenite is also known as the UG2 pegmatoid
(c.f., Hiemstra, 1985, Penberthy and Merkle, 1999). The contact between UG2 and the UG2
pegmatoid is irregular. In the De Wildt project area the pegmatoidal feldspathic pyroxenite hosts
disseminated chromite, chromitite lenses and chromite blebs. The UG2 pegmatoid is gradational
downwards into a medium-grained pyroxenite that hosts disseminated chromite and diffuse chromite
lenses/blebs. PGE mineralization extends into the footwall pyroxenite unit. However, not all boreholes
intersected the UG2 pegmatoid.

5.10. Dykes, sills and replacement rocks at De Wildt project area

Dolerite dykes and sills were intersected in boreholes throughout the project area. The dykes
trend NW. The dykes are variable in thickness ranging from 0.2 m to 55.73 m. They are associated
with fracture zones and small-scale fault displacement. The dykes are associated with metasomatism
of the BC mafic rocks, with alteration extending only a few centimetres from the intrusive contacts.
A number of iron-rich ultramafic pegmatoids (IRUP) were intersected in boreholes across the
prospect. The IRUPs occur as centimetre to metre-scale, dark grey-black patches or irregular zones
within norite and anorthosite layers. The IRUP zones are composed of large crystals of clinopyroxene,
plagioclase and biotite with accessory replacement magnetite, sulphide and olivine. Of significance,
was the 100 m thick (apparent thickness) IRUP zone intersected in borehole DWD12, which replaced
the normal mafic rock succession of the Upper Critical Zone with an irregular iron-rich ultramafic
pegmatoid in the depth interval 790.34 - 911.37 m.

Table 8: Immediate hangingwall and footwall lithologies to the UG2 chromitite layer.

5.11. Summary

? The intersection of the Marginal Zone directly below the Upper Critical Zone implies the
absence of the Lower Critical Zone and Lower Zone stratigraphic horizons of the BC at De
Wildt.
? The Merensky Reef intersected ranges from 11.6 m to 20.5 m and may sometimes have a
basal chromitite developed. The reef was characterized by low PGE contents averaging less
1g/t with top and bottom PGE value peaks associated with the top and bottom contacts
respectively.
? The UG2 chromitite layer within the prospect exhibited lateral variation from the east to west.
The UG2 layer changes from being a single massive chromitite seam into a multi-layered and
sometimes only consisting of only disseminated chromite and chromite stringers within a
feldspathic pyroxenite. Also, the UG2 layer was characterized by a mineralized immediate
footwall extending down to 4 m in some portions of the prospect
.
77
CHAPTER 6
PETROGRAPHY

A summary of petrographic descriptions of the lithologies deemed unsual or not so common
within the RLS encountered (particularly in close proximity to the UG2 layer) in the study area is
presented, such as the numerous quartzite units and other xenoliths, and replacement magnetite. The
objective is to assess whether the UG2 layer at De Wildt may have formed or, been subjected to, any
modifying conditions arising from possible contact metamorphism, deformation and alteration related
to the emplacement of dyke swarms or, during cooling of the BC intrusion. Also, a petrographic
overview is presented of the wider than normal UG2 footwall layer made up of a feldspathic
pyroxenite with disseminated chromite and chromite blebs/slivers.

6.1. Quartzite xenoliths

Photomicrographs of quartzite intersected in borehole in the De Wildt project area are presented
in Figures 35a-b. The polished sections are characterized by sub-rounded to anhedral quartz crystals
and are weakly sericitised. Micro-faults crosscut the quartz grains with micro-displacement of the
crystals. They are in-filled by oxide minerals.
The quartzite units are interpreted as xenoliths into the BC mafic rocks and have suffered brittle
deformation and contact metamorphism to form feldspathic quartzite. Some of the rocks intersected at
De Wildt are highly deformed to quartzite hornfels (Fig. 35c). Plagioclase is saussuritized giving a
dusty appearance to the thin sections.

6.2. Hornfels

Hornfelsed metasedimentary rocks where intersected in several boreholes in contact with
dolerite. In general, the hornfels are composed of recrystallized feldspathic quartzite. In borehole
DWD3A/D (Fig. 35d), a zone of biotite (yellowish-reddish tint) characterised the contact between
dolerite and hornfels, and granoblastic clinopyroxene characterised the chilled margins of dolerite.
Suturing of quartz along grain boundaries occurred.
In borehole PDW4 (drilled in Schietfontein Farm), hornfelsed metasediments was composed of
relatively equal proportions of chlorite (35%), epidote (30%) and actinolite (30 %) (Fig. 35e). The
anhedral crystals are diffuse and distorted. Elongate needles of accessory sillimanite (5%) were
randomly orientated throughout the matrix and together with chlorite, actinolite and epidote, suggest
that an aluminosilicate-potassic protolith was subjected to hornblende-hornfels contact metamorphism
during emplacement of the BC.
78
In one or two samples, the protolith is decidedly quartzo-feldpathic and composed of coarse
plagioclase, medium-grained quartz and biotite (Fig. 35f). Quartz occurs in anhedral-subhedral and
interconnected aggregates. Plagioclase crystals and their ghosts (remnants) are randomly orientated.
Diffuse intergrowths of quartz-plagioclase occur in some samples. Biotite is formed within
interstitices. Anhedral quartz crystals sometimes present together with highly deformed plagioclase
and fractured orthopyroxene suggesting a higher grade of contact metamorphism and crystal
deformation (Fig. 35g).

6.3. Chromite-rich feldspathic pyroxenite

A photomicrograph of the poorly developed UG2 seam composed of feldspathic pyroxenite and
hosting disseminated chromite is presented Figures 35h-i. It consists of euhedral grains of chromite
(opaque crystals) embedded within intercumulus plagioclase (white) and orthopyroxene crystals. The
crystal margins are crosscut by chromite crystals. Orthopyroxene crystals exhibit small-scale fractures.
Clusters of chromite grains occur in intercumulus plagioclase within feldspathic (plagioclase-rich)
pyroxenite i.e., coarse-grained feldspathic pyroxenite hosting disseminated chromite (Fig. 35i).

6.4. Replacement pegmatites

Replacement pegmatites predominantly consist of large (~6 mm in diameter) prismatic,
twinned plagioclase crystals (45%), together with subhedral to anhedral clinopyroxene crystals (35%),
othopyroxene (10%), oxides (7%) and biotite (3%) (Fig. 35j). The random plagioclase laths define a
sub-ophitic texture. Opaque minerals occur at existing minerals boundaries and are interpreted as
replacement magnetite and/or ilmenite.

6.5. Summary

? Chromite grains are disseminated within pyroxenes and intercumulus plagioclase.
? Deformed feldspathic quartzite and hornfels rocks are micro-fractured and micro-faulted.
They are characterised by recrystallized quartz crystals. The rocks have undergone brittle
deformation.
? Two alteration types occur at De Wildt and include sericitisation and saussuritization of
feldspars. Significant alteration is associated with contact metamorphism. Metamorphic
assemblages consisting of random crystals of chlorite, epidote and actinolite are consistent
hornblende-hornfels facies. Hornfels assemblages are interpreted as metasedimentary
xenoliths of the Transvaal Supergroup.

Figure 35: Photomicrographs of polished thin sections of selected core samples from various boreholes at De
Wildt. All images are under crossed polars. Images A and B are quartzite characterized by veins composed of
secondary oxide minerals. Sericite defines a zone around an (unidentified) opaque in image B. Image C shows a
highly altered rock in which saussuritized plagioclase gives the rock a dusty appearance. Image D characterize
contacts between recrystallized quartzite and dolerite: secondary mineral such biotite have formed between the
contact and define contact metamorphism to hornblende-hornfels facies. Image E was derived from a
metasedimentary xenolith that underwent low-grade metamorphism (borehole PDW04).

Figure 35: Photomicrographs of polished thin sections of selected core samples from various boreholes at De
Wildt. All images are under crossed polars. Images F and G are hornfelses marked by the presence of fractured
quartz, plagioclase, biotite and vague pyroxene crystals. The deformation in image F is interpreted to be
localized and a result of magmatic fluids associated with intrusion of the dolerite dykes prominent within the De
Wildt project area. Images H and I are of feldspathic pyroxenite hosting disseminated chromite, intersected
where the UG2 seam was anticipated. The chromitite grains do not exhibit a preferred orientation but are
scattered throughout the crystal mass. Image J is of pegmatite consisting of plagioclase, clino and othopyroxene
and biotite crystals with oxides minerals in interstitices.

81
CHAPTER 7
MINERALIZATION

A summary of the PGE mineralization within the Merensky Reef and UG2 chromitite layers is
presented. Also the mineralized UG2 footwall layers consisting of a feldspathic pyroxenites/norite
hosting disseminated chromite and chromite blebs/slivers are outlined.

7.1. Merensky Reef PGE content

In the Dewildt project area, two major zones of PGE?s occur along the Merensky Reef; an upper
zone near the top gradational contact and a lower zone associated with the basal chromitite
stringer/layer. However, the average PGE values for the Merensky Reef (or Merensky pyroxenite) in
boreholes drilled by Nkwe Platinum were less than 0.1 g/t for Pt, Pd, Rh and Au (also known as 3E +
Au, in South African platinum industry). The Merensky Reef intersections from historical drill core
averaged close to 1g/t (3E+Au) over widths of less than 1.0 m near the top and bottom contacts (Fig.
36).

Figure 36: Average PGE values for Merensky Reef over 1 m from the bottom contact. In most of the boreholes
drilled by Nkwe Platinum the grades were less than 0.1 g/t over widths of less than one metre near the top and
bottom contacts

7.2. UG2 chromitite Isochon

The UG2 isochon map (Fig. 37) presents the five PGE?s and gold distribution or grades (5E +
Au, grams per tonne) of the UG2 layer across the De Wildt project area; there is a decrease in grade
from the west to the east. The relatively higher grades of 3.0 g/t in the western zone coincide with the
distribution of normal facies across the De Wildt project area. The central and eastern zones are
characterized by grades less than 2.0 grams per tonne (g/t), coinciding with the zone in which the split-
reef facies is dominant.
82
92000 93000 94000 95000 96000 97000
92000 93000 94000 95000 96000 97000
-2837000
-2836000
-2835000
-2837000
-2836000
-2835000
SK01
SK2
SK3
SK4
SK5
SK6 SK7
SK8
SK9 SK10
SK13
SK14
DWD 3
DWD 7
DWD 9
DWD 8
DWD 10
DWD 11
DWD 12
DWD 13
PDW1
PDW2 PDW3
PDW4
2098
2055
2092
DWD01
DWD02DWD02A
DWD03A
DWD04
DWD05
DWD06
SK11
SK12
SK15
SK16
SK17
SK18
ZK01
UG2 isochon (5E + Au), g/t
0.0 - 1.0
1.0 - 2.0
2.0 - 3.0
> 3.0
N
7.3. PGM value distribution along the UG2 chromitite

The top and bottom contacts, and the middle part of the normal facies UG2 chromitite layer,
display elevated PGE values, while the immediate hangingwall and footwall units have relatively low
PGE values (Fig. 38a). The highest PGE peaks are exhibited by the bottom contact and middle part of
the UG2 layer. Prill splits of the UG2 chromitite show that the reef has a high Pt:Pd ratio of 3:1, i.e.,
the platinum content of the UG2 chromitite is more than three times that of palladium (Fig 38b, Table
9) and the Pt:Pd ratio show a general upward increase within the UG2 chromitite layer. Pt, Rh and Ru
show similar variations with peak values in the middle of the chromitite layer. The average Pt/Pd ratio
for the mineralized UG2 footwall is 1.7 (Table 9). In contrast, prill splits of the UG2 from the western
Bushveld show that the platinum content is twice that of the palladium (Pt:Pd ratio of 2:1). This may
indicate that the UG2 chromitite layer at De Wildt may have been subjected to late stage fluid activity
with subsequent remobilization of Pd.
A plot of the centimetre grams per tonne of the PGEs distribution within the UG2 chromitite
layer across the prospect is presented in Fig. 39. The western zone is characterized by relatively high
PGE value distribution, with values greater than 400 cmg/t (5E + Au). Some locations within the
western zone have values that exceed 600 cmg/t. The eastern zone and part of the central zone record
values less than 200 cmg/t PGE distribution. Generally, the PGEs distribution decreases from west to
east such that the eastern zone of the project area record relatively poor PGE values. The poor values
in the eastern zone coincide with the presence of narrow ?B type? split reef sub-facies, in which the
UG2 chromitite seams are poorly packaged (intersections in boreholes DWD10 and DWD12), while
the western zone hosts predominantly normal facies and ?B type? split reef sub-facies which are
massive and relatively well packed.

Figure 37: UG2 chromitite isochon map over the De Wildt project area.

Figure 38a: PGM value distribution within the UG2 chromitite layer (normal facies type) in borehole SK02.

Figure 38b: Plots of Pt, Pd, Rh, Ru and Pt/Pd ratios of UG2 chromitite layer (from borehole PDW01). The Pt/Pd
ratio values display an overall upward increase within the chromitite layer.

84
UG2 isochon (cmg/t)
N
0.0 - 200
200 - 400
400 - 600
> 600
91500 92000 92500 93000 93500 94000 94500 95000 95500 96000
91500 92000 92500 93000 93500 94000 94500 95000 95500 96000
-2837500
-2837000
-2836500
-2836000
-2837500
-2837000
-2836500
-2836000
SK01
SK2
SK3
SK4
SK5
SK6 SK7
SK8
SK9 SK10
SK13
SK14
DWD 3
DWD 7
DWD 9
DWD 8
DWD 10
DWD 11
DWD 12
DWD 13
PDW1
PDW2 PDW3
PDW4
2098
2055
2092
DWD01
DWD02DWD02A
DWD03A
DWD04
DWD05
DWD06
SK11
SK12
SK15
SK16
SK18
ZK01

Figure 39: UG2 chromitite isochon map showing the plot of centimetre grams per tonne (5E + Au) across the
De Wildt project area.
85
Table 9: Prill Splits for the UG2 chromitite layer and UG2 footwall at De Wildt are shown on the left, while the prill splits of the UG2 and MR reefs from other BC localities
are given on the right.
De Wildt Project - Prill Splits
(5PGE)
Prill Splits in other localities in the Western Bushveld
Complex
UG2 Chromitite
Layer Crocodile River Rustenburg Section
BHID Farm %Pt %Pd %Rh %Ru %Ir %Au 5PGE Splits % 3PGE Splits %
PDW1D1 Schietfontein 53.27 13.93 10.14 16.82 5.18 0.67 UG2 MR UG2
PDW2D1 Schietfontein 57.66 8.76 10.46 18.73 3.89 0.49 Pt 49.7 Pt 63.7 55
PDW3D1 Schietfontein 51.32 13.67 11.02 18.69 4.32 0.98 Pd 21.6 Pd 27.7 34
2092 D3 Zilkaatsnek 42.22 34.73 7.26 11.92 2.72 1.14 Rh 8.1 Rh 4.9 10
2098 D1 Zilkaatsnek 59.34 8.60 9.26 16.69 5.62 0.5 Ru 16.6 Au 3.7 1
1977 D4 Krelingspost 62.75 7.84 10.46 13.07 4.25 1.63 Ir 3.5
Average 54.43 14.59 9.77 15.99 4.33 0.90 Au 0.5
Pt/Pd 3.73
Pt/Pd 2.30 Pt/Pd 2.30 1.62
Lower UG2 Chromitite Layer (of the split-
reef facies)
BHID Farm %Pt %Pd %Rh %Ru %Ir %Au
2098 D1 Zilkaatsnek 56.86 10.62 9.96 15.93 4.42 2.21 Marikana Amandelbult
1977 D4 Krelingspost 51.54 22.03 7.93 12.78 4.85 0.88 3PGE Splits 3PGE Splits
Average 54.20 16.33 8.95 14.36 4.64 1.55 UG2 MR UG2
Pt/Pd 3.32 Pt 61.8 Pt 61.9 58
Pd 27.7 Pd 29.5 29
Chromite-rich UG2 footwall
layers Rh 9.7 Rh 4.9 12
BHID Farm %Pt %Pd %Rh %Ru %Ir %Au Au 0.9 Au 3.7 1
PDW1D1 Schietfontein 55.17 17.24 10.34 10.34 3.45 3.45
2092 D3 Zilkaatsnek 33.88 40.24 7.06 13.18 4.71 0.94 Pt/Pd 2.23 Pt/Pd 2.10 2.0
2098 D1 Zilkaatsnek 53.11 15.77 9.96 14.11 5.39 1.66
1977 D4 Krelingspost 43.48 32.61 8.70 10.87 2.17 2.17
Average 46.41 26.47 9.02 12.13 3.93 2.06
Pt/Pd 1.75
Platinum Map of South Africa (2003), Barker and Associates
publishers
86
H/W
F/W
UG2 seam
FPX-CR
0.29 -4.0 m
mineralized zone
Mineralized UG2 footwall
7.4. Mineralized UG2 footwall at De Wildt

A mineralized zone below the UG2 chromitite is present in more than 60% of boreholes from
the project area (Fig. 40). Chromite is discontinuous over approximately 4.0 m; the zone consists of a
chromite-bearing feldspathic pyroxenite or mesonorite mixed with chromitite at millimetre to
centimetre scale. In contrast, at Zilkaatsnek (western zone) and Schietfontein farms, the mineralized
UG2 footwall is dominantly composed of feldspathic pyroxenite (Table 10), whereas on the
Krelingspost Farm, the footwall is composed of a coarse-grained mesonorite.
The mineralized feldspathic pyroxenite footwall layers are brown-grey in colour. The
feldspathic pyroxenite is closely mixed at a centimetre scale with chromitite (Fig. 41). It consists of
~5% chromite in the form of disseminations, chromite stringers/blebs and or chromite patches. The
feldspathic pyroxenite is commonly medium-grained, with pegmatoidal segregations. In comparison,
the mineralized UG2 footwall mesonorite is coarse-grained with rare zones of pegmatite. It is
characterized by randomly distributed slivers/stringers of chromite as well as chromite blebs (Fig. 42)
and ranges in thickness from 0.29 m to 4.0 m. In borehole SK09 it is approximately 10 m thick.

Figure 40: Schematic depiction of the mineralized zone below the UG2 layer. The mineralized footwall units
(feldspathic pyroxenite/norite) host disseminated chromite, chromite blebs and chromitite slivers extending up to
4 m below the UG2 chromitite layer.

Table 10: A summary table of the different mineralized UG2 footwall lithologies intersected in various
boreholes within the project area.

BHID Nature of footwall lithology
SK02 Pegmatoidal feldspathic pyroxenite characterized by
disseminated chromite and sulphides (pyrrhotite and
chalcopyrite)

SK04 Fine to medium grained, pseudoporphyritic feldspathic
pyroxenite with disseminated chromite.

SK06 Feldspathic pyroxenite with chromite stringers and chromite
inclusions

SK08 Feldspathic pyroxenite which grades into a pegmatoidal norite
with minor specs of sulphides

SK09 Feldspathic pyroxenite (with large clinopyroxene phenocrysts
in places) and blebs of chromitite, chromite forms in small
and large blebs randomly scattered within the feldspathic
pyroxenite, discontinuous melanorite layers/patches occur in
places.

DWD08 Coarse grained norite with oikocrysts, slivers and blebs of
chromite together with some stringers of magnetite sits ~
13.25 m below the UG2 chromitite layer.

DWD06 Coarse grained to pegmatoidal feldspathic pyroxenite with
disseminated chromite and chromite blebs, occurring
immediately below the UG2 layer (or UG2 remnant)

DWD09 Immediate footwall is characterized by a medium to coarse
grained norite with irregular chromite stringers and blebs.

Figure 41: (a) UG2 feldspathic pyroxenite footwall characterized by irregular zones with disseminated
chromite, intersected in borehole DWD07. (b) UG2 feldspathic pyroxenite footwall with irregular and vague
chromite stringers, intersected in borehole DWD06.

Figure 42: Chromite slivers and chromite blebs in a coarse crystalline-pegmatoidal norite underlying the UG2
chromitite layer in one borehole DWD08 intersection. Chromite patches/blebs and slivers are shown by yellow
lines.
a b
(a) (b)
89
UG2 footwall isopach (m)
N
0.0 - 1.0 1.0 - 2.0 2.0 - 3.0 3.0 - 4.0 4.0 - 5.0 > 5.0
92000 93000 94000 95000 96000 97000
92000 93000 94000 95000 96000 97000
-2837000
-2836000
-2835000
-2837000
-2836000
-2835000
SK01
SK2
SK3
SK4
SK5
SK6 SK7
SK8
SK9 SK10
SK13
SK14
DWD 3
DWD 7
DWD 9
DWD 8
DWD 10
DWD 11
DWD 12
DWD 13
PDW1
PDW2 PDW3
PDW4
2098
2055
2092
DWD01
DWD02DWD02A
DWD03A
DWD04
DWD05
DWD06
SK11
SK12
SK15
SK16
SK17
SK18
ZK01

7.4.1. UG2 footwall isopach
An isopach map (Fig. 43) of the UG2 footwall isopach was compiled using the thickness of the
chromitiferous UG2 footwall (i.e., inclusive of feldspathic pyroxenite and/or mesonorite layers and the
lower UG2 chromitite unit where present). In the western and eastern zones of the project area, the
thickness of the mineralized UG2 footwall averages 4.0 m. However, several boreholes intersected
mineralization extending up to 5.0 m into the UG2 footwall. The central zone is characterised by
relatively narrow mineralized footwall layers less than 2.0 m in thickness. Mineralized UG2 footwalls
of less than 1.0 m in thickness occur both in the west and central zone.

Figure 43: UG2 footwall isopach across the De Wildt project area. The term ?footwall? in this case refers to the
chromitiferous UG2 footwall units consisting of feldspathic pyroxenite/norite hosting disseminated chromite,
chromitite stringers and chromitite blebs.

90
UG2 foot wall isochon (g/t)
0.5 - 1.0
1.0 - 1.5
1.5 - 2.0
2.0 - 2.5
2.5 - 3.0
> 3.0
N
91500 92000 92500 93000 93500 94000 94500 95000 95500 96000
91500 92000 92500 93000 93500 94000 94500 95000 95500 96000
-2837500
-2837000
-2836500
-2836000
-2837500
-2837000
-2836500
-2836000
SK01
SK2
SK3
SK4
SK5
SK6 SK7
SK8
SK9 SK10
SK13
SK14
DWD 3
DWD 7
DWD 9
DWD 8
DWD 10
DWD 11
DWD 12
DWD 13
PDW1
PDW2 PDW3
PDW4
2098
2055
2092
DWD01
DWD02DWD02A
DWD03A
DWD04
DWD05
DWD06
SK11
SK12
SK15
SK16
SK18
ZK01

7.4.2. UG2 footwall isochon
A plot of the PGE grades variation of the mineralized UG2 footwall layers across the prospect is
given in Figure 44. The PGE grades are higher in the west and eastern portions of the prospect and
range from 2.0-3.0 g/t (5E + Au). Grades exceeding 3.0 g/t (5E + Au) are scattered throughout the
western zone at Zilkaatsnek Farm. The mineralized UG2 footwall in the central zone hosts PGE grades
less than 2.0 g/t (5E + Au). A clear-cut grade distribution of PGEs along the UG2 footwall could not
be established, which could reflect the highly irregular nature of the mineralized units throughout the
project area. However, there was a direct correlation between the PGE values and the presence of
chromite (Fig. 45).

Figure 44: UG2 footwall isochon showing the variation of grades within the mineralized footwall units across
the De Wildt project area. The term ?footwall? in this case refers to the chromitiferous UG2 footwall units
consisting of feldspathic pyroxenite/norite hosting disseminated chromite, chromitite stringers and chromitite
blebs.

91
0 1 2 3 4
PGM value distribution (3E+Au) 3g/t
194
192
190
188
186
de
pt
h
(m
)
UG2
H/W
F/W
2098

Figure 45: PGE value distribution along the mineralized UG2 footwall unit as represented in borehole 2098. The
footwall unit is made up of a feldspathic pyroxenite with disseminated chromite (~30%) and chromite slivers.

The isochon map in Fig. 46 presents centimetre grams per tonne of the 5 PGE?s plus Au in the
mineralized UG2 footwall. The western zone of the prospect hosts relatively high values with PGE
values that range between 400-600 cmg/t. The far western margins of the prospect host PGE values
exceeding 600 cmg/t. In the rest of the prospect (central and eastern parts) the mineralized UG2
footwall layers contain PGEs values less than 400 cmg/t.

7.4.3. Combined UG2 and UG2 footwall isopach
The isopach map in Figure 47 presents the combined thickness distribution of UG2 reef and
mineralized UG2 footwall across the prospect. The eastern zone of the project area has combined
thicknesses ranging between 1.0 m to 3.0 m. Patches/pockets of combined UG2 and UG2 footwall
thickness ranging from 3.0 to 5.0 m (and sometimes exceeding 5.0 m) only occur on the western zone.
The central zone is characterized by combined thickness of less than 1.0 m and represents a UG2
footwall of insignificant PGE value and/or where the UG2 reef itself is poorly developed.
92
91500 92000 92500 93000 93500 94000 94500 95000 95500 96000
-2837500
-2837000
-2836500
-2836000
SK01
SK2
SK3
SK4
SK5
SK6 SK7
SK8
SK9 SK10
SK13
SK14
DWD 3
DWD 7
DWD 9
DWD 8
DWD 10
DWD 11
DWD 12
DWD 13
PDW1
PDW2 PDW3
PDW4
2098
2055
2092
DWD01
DWD02DWD02A
DWD03A
DWD04
DWD05
DWD06
SK11
SK12
SK15
SK16
SK18
ZK01
UG2 foot wall isochon (cmg/t)
N
0.0 - 200
400 - 600
> 600
200 - 400
92000 93000 94000 95000 96000 97000
92000 93000 94000 95000 96000 97000
-2837000
-2836000
-2835000
-2837000
-2836000
-2835000
SK01
SK2
SK3
SK4
SK5
SK6 SK7
SK8
SK9 SK10
SK13
SK14
DWD 3
DWD 7
DWD 9
DWD 8
DWD 10
DWD 11
DWD 13
PDW1
PDW2 PDW3
PDW4
2055
2092
DWD01
DWD02DWD02A
DWD03A
DWD04
DWD05
DWD06
SK11
SK12
SK15
SK16
SK18
Combined UG2 and foot wall isopach (m)
3.0 - 5.0
> 5.0
1.0 - 3.0
< 1.0
N

Figure 46: UG2 footwall isochon plot of centimetre grams per tonne (5E+ Au) within the mineralized units
across the De Wildt project area.

Figure 47: Combined UG2 layer and mineralized footwall isopach. This plot gives total length variation of the
mineralized horizon (UG2 seam and footwall) across the De Wildt project area.
93
CHAPTER 8
DISCUSSION

8.1. Merensky Reef

The Merensky Reef in the De Wildt project area is 11.6 m to 20.5 m in thickness which is
greater than intersections at Amandelbult and the Rustenburg area where the reef averages 1.6 m and
1.0 m, respectively (Viljoen, 1994). In addition, the Merensky Reef at De Wildt is medium
grained/crystalline but at Amandelbult and Rustenburg area is pegmatoidal (Viljoen, 1994).
Furthermore, the PGE grade at De Wildt is sub-economic averaging less than 1.0g/t (3E+Au). The
style of the Merensky Reef at De Wildt correlates well with the distal facies of the Merensky Reef that
extends from the Western Platinum area to Brits, as documented by Viljoen (1999). He described the
distal facies as wide, non-pegmatoidal and low-grade to sub-economic, with few major potholes. A
wide and non-pegmatoidal Merensky Reef facies was also reported by Davey (1992) west of De Wildt
and east of the Spruitfontein upfold (Fig. 5).
The variable immediate footwall units to the Merensky Reef that were composed of anorthosite
or, norite and pyroxenite, could suggest the occurrence of rolling of the Merensky Reef. The irregular
bottom contact of the Merensky Reef defined by a chromitite stringer at De Wildt compares
favourably with the small-scale undulations at the base of Merensky Reef reported elsewhere in the
BC by Van der Merwe and Cawthorn (2005). While there is no definitive model to explain all the
features of such structures, Vermaak (1976) attributed these undulations to impact generated dimpling
of the floor by boulder-sized or aggregates of chromitite and pyroxenes onto unconsolidated
chromitite and underlying anorthosite. However, in boreholes such as PDW02, where the Merensky
Reef is situated close to the UG2 layer, it is likely that potholing may have occurred as indicated by
attenuated and/or missing stratigraphic layers, with no evidence of faults.
The PGE value distribution within the Merensky Reef at De Wildt, though comparatively lower
than in other parts of the BC, is concentrated on the upper portion of the reef and at the bottom, in
association with basal chromitite stringers. As such, the Merensky Reef at De Wildt may also be
tentatively regarded as ?top loaded? and ?bottom loaded? with respect to PGE value distribution.
The Bastard Reef (BR) was not intersected in the De Wildt project area. The Bastard Reef to the
west of De Wildt was reported by Davey (1992) as noritic in character and about 12 m wide, and could
be identified by the presence of an erratic 1-2 mm basal chromitite stringer. However at De Wildt, a
predominantly norite sequence that is devoid of any chromitite stringers, made identification of the
Bastard Reef impossible; alternatively, it could simply be that this unit was not deposited.
94

8.2. UG2 Chromitite layer

The two characteristic facies types of the UG2 chromitite layer encountered at De Wildt give
some indication of the physical conditions during formation of the chromitite layer. The poikilitic
texture observed at De Wildt is common in other parts of the BC (Viljoen et al., 1986, Leeb-Du Toit,
1986) and has been attributed to ortho-pyroxenite crystals forming around and enclosing the pre-
existing chromite grains (Leeb-Du Toit, 1986). However, the finely crystalline UG2 chromitite unit
that is dominant in the eastern zone of the project area, and which is devoid of any poikilitic texture,
indicates rapid cooling of the seam.
Furthermore, the UG2 chromitite layer intersected in boreholes in the De Wildt project area is
different from the typical UG2 encountered elsewhere in the western limb of BC in a number of ways.
These differences include; (1) the absence of UG2 leader seams/triplets, (2) variation in thickness and
the occurrence of a split-UG2 reef facies, (3) the occurrence of a norite in the immediate hangingwall
to the UG2 mean seam, (4) the relatively lower PGE content, and (5) the existence of a mineralized
UG2 footwall layer. Consequently, the interpretation or distinction of the UG2 layer from any other
chromitite layer within the Bushveld stratigraphy was ascertained by considering PGE content. This
was despite the relatively lower PGE content when compared to other BC localities. In summary:

? The UG2 layer intersected in all boreholes at De Wildt does not have the usual leader seams
or triplets which occur a few centimetres into the immediate overlying UG2 hangingwall
pyroxenites (Viljoen and Sch?rmann, 1998). The absence of the leader seams is attributed to the
coalescence of the leader seams with the main UG2 chromitite seam during
formation/emplacement. However, in some intersections where disseminated chromite was present
a few centimetres into the UG2 hangingwall, the disseminated chromite is interpreted to be the
remnant of poorly developed or disturbed leader seams/triplets. The UG2 is usually overlain by
?leader? chromitite layers whose thickness ranges from 10 to 12 cms and occurs within the
immediate hanging-wall feldspathic pyroxenite (Viljoen and Sch?rmann, 1998).

? The thickness variations and splitting of the UG2 chromitite layer across the De Wildt project
area (Fig. 48). A localized lateral variation in the development of the UG2 layer from west to the
east across the De Wildt prospect was observed. A normal ?A? type facies, consisting of a massive
chromitite seam is confined to the western portion at Zilkaatsnek Farm. The UG2 layer separates
to form split reef facies towards the east.

Variations in thickness, lithology and composition within the BC are interpreted with respect to
proximity to the main feeder zone located at Union Section (Teigler et al., 1992; Maier and Eales,
95
1997; Scoon and Teigler (2005). The UG2 and UG1 chromitite layers tend to be thicker in the
distal facies, while the Merensky Reef is well developed in the proximal facies but thicker and
poorly mineralized in the distal facies (Scoon and Teigler, 2005). Teigler et al. (1992) and Scoon
and Teigler (2005) observed that the Upper Critical Zone in the distal facies is thinner and
dominated by plagioclase cumulates while in the proximal facies, it is thick, well developed and
dominated by pyroxene cumulates. The normal UG2 chromitite (A-type facies) is interpreted to
have crystallised and formed far from the main feeder source in the distal facies because of its
greater width of up to 1.25 m.

The highly variable UG2 layer at De Wildt that varies in thickness, localized lateral
discontinuities, and splitting may be an overprint of syn- and post-Bushveld factors such as
faulting, formation of basement highs due to doming, and potholes;

Presence of basement highs ? Thinning and/or complete absence of UG2 layer, together with the
split-reef facies, is dominantly confined to the eastern zone of the prospect. The eastern zone is
considerably disturbed as revealed in aeromagnetic maps that show the zone is crosscut by
numerous faults. Also, two quartzite hills that are interpreted to be part of the Transvaal
Supergroup, crop out in the eastern zone. It is not known whether these hills are large floor rock
bodies or xenoliths rafted away from the Magaliesberg Quartzite Ridge. Regardless, boreholes to
the immediate west of these hills intersected only a thin zone of BC mafic rock and these were
characterized by a steep layer-dip of up to 60?. The boreholes then intersected the floor rocks
(feldspathic quartzite) i.e., the floor rocks were intersected at shallower depth than anticipated
(boreholes DWD03, DWD09 and DWD14). This suggests some continuity of the floor rocks at
depth to the west and thus the possibility that these rocks form Magaliesberg Quartzite outliers in
the BC.

The occurrence of floor upfolds or domes, such as Kookfontein and Spruitfontein in the western
BC (Fig. 5, Chapter 2), has been shown to affect the development of the Critical Zone. Uken and
Whitey (1997), Scoon (2002) and Gerya et al. (2004) observed thinning and draping of magmatic
units above diapers. The Critical Zone is uplifted and attenuated against the flanks of domes such
that individual layers or units are significantly thinned and may completely disappear or ?pinch-
out? (Scoon, 2002). The affected rocks display a strong lineation and foliation. Such basement
highs have been explained by Scoon (2002) as the result of diapirism triggered by gravitational
loading and heating of floor rocks (Transvaal Supergroup) in response to intrusion by the BC. The
variable widths of the UG2 chromitite layer encountered at De Wildt may have also been
influenced by formation of these basement structures.
96
normal UG2 facies Split-reef facies 'chromitiferous'
type
massive UG2 chromitite seam
fpx
UG2 Footwall
UG2 Hangingwall
fpx with diss chromite
fpx with diss chromite
UG2 Chromitite layer
chromitite slivers
0
1.25m
(Zilkaatsnek Farm) (Schietfontein Farm) (Krelingspost Farm)
W
E
(towards margin of BC)
fpx

Figure 48: Schematic representation (not to scale) of the UG2 chromitite seam across the De Wildt project area showing the relationship between the UG2 normal reef facies,
UG2 split-reef facies and feldspathic pyroxenite (fpx). The normal UG2 facies is confined to the west of the prospect while the split-facies occurs predominantly in the eastern
zone on Schietfontein and Krelingspost farms.
97

Faults and potholes - The Schietfontein, Zilkaatsnek and Lindiwe faults are tentatively interpreted
as antithetic faults to synthetic faults that form the Brits Graben; the Brits Graben influences the
orientation of regional structural lineaments. Fault formation and the associated fracturing are
possible modifying factors to the occurrence of normal UG2 reef. While the occurrence of a UG2
seam that is devoid of massive chromitite layer (but consisting of only disseminated chromite
hosted within a feldspathic pyroxenite) is puzzling, it may be associated with slumping and
faulting of stratigraphic layers. Drillcore from pothole-edge intersections are difficult to interpret
and range from pinch-outs of the reef to UG2 reef remnants in the form of chromitite slivers or
disseminated chromite in pyroxenite or norite. In contrast, the splitting of the UG2 seam into
multi-layers suggests that the formation of the UG2 chromitite was not a single event. Instead,
multiple magma injections into the chamber occurred during the crystallization processes to form
the UG2 seam at De Wildt.

? The UG2 layer at De Widlt is predominantly overlain by a norite unit of variable thickness
(0.5 ? 51 m) instead of the usual UG2 pyroxenite common in other BC localities. The norite
varies from medium to coarse crystalline and is developed in over ninety percent of the boreholes
drilled at De Wildt. The absence of the UG2 pyroxenite is interpreted as the result of no deposition
or, complete erosion during emplacement of the norite layer. The occurrence of thin and sparse
UG2 pyroxenite in some borehole intersections is attributed to partial erosion of the earlier formed
cumulate; irregular contacts between the UG2 pyroxenite and the overlying norite support this
interpretation. The presence of a norite unit suggests that the Critical Zone rocks at De Wildt
crystallized from a highly evolved parental magma. This interpretation concurs with Scoon and
Tiegler (1995) who suggested that the Critical Zone in the distal facies of the BC is dominantly
characterized by plagioclase cumulates whereas in the proximal facies, pyroxenite cumulates are
abundant.

? The PGE content of the UG2 layer at De Wildt is lower than that from other sites in the
western BC. The grades for the UG2 layer at De Wildt ranged from 1.5 to 3.0 g/t (3PGE + Au)
over 1.3 metres, while the mineralized UG2 footwall layer (where developed) averaged 2.0 g/t
(3PGE +Au) over 2.35 metres. The lower grades are a result of the greater vertical spread of the
primary chromite accumulations and associated PGEs within the relative thick UG2 layer at De
Widlt. The PGE values along the UG2 chromitite (A-facies) were preferentially concentrated close
to the upper and lower contacts as well in the middle of the massive chromitite layer. The
occurrence of a middle PGE peak along the UG2 chromitite layer is a deviation from that
encountered in other parts of the BC where the peaks usually occur close to the upper and lower
contacts. The three peaked UG2 layer at De Wildt thus contrasts with the ?top and bottom loaded?
98
UG2 layer of Maier and Barnes (2008) who suggested that the chromitite seams formed by the
injection of chromitite slurries into semi-consolidated footwall cumulates. Instead the observations
at De Wildt compare well with Hiemstra (1986) and Cawthorn (2004) who concluded that the UG2
layer is made up of three cyclic units of distinct PGE composition.

However, the prill splits of UG2 chromitite reef intersected at De Wildt are greater than 3:1 for
Pt:Pd ratios while the average ratios for the mineralized UG2 footwall is 1.7 (Table 9, Chapter 5). The
ratio is higher than in UG2 chromitite layers at other localities in the western BC such as at
Amandelbult, Crocodile River and Marikana whose UG2 chromitite Pt:Pd ratio averages 2:1. The
relatively high Pt:Pd ratios obtained from the UG2 chromitite layer at De Wildt may imply that;

(a) The leader seams coalesced and became part of the upper chromitite unit; the leader seams
have been shown to have higher Pt:Pd ratios similar to those of the UG2 layer at De Wildt
(Bleeker pers. Comm., 2007). The UG2 seam at De Wildt was devoid of any leader in addition
to being relatively wide. UG2 Leader seams have been shown to bifurcate and coalesce at local
scale in other parts of the BC (Viljoen et al., 1986).

(b) The UG2 chromitite layer at De Wildt has been subjected to some degree of fluid activity as
evidenced by the presence of replacement magnetite, amphibole, chlorite and epidote within the
rocks. Also the UG2 seam at De Wildt is in close proximity to the Transvaal floor rocks. The
high fluid activity at De Wildt may have resulted in the remobilization of the highly mobile Pd
leaving behind a Pt enriched UG2 chromitite seam. This could account for the relatively lower
overall PGE content of the De Wildt UG2 seam and concur with observations of Maier and
Bowen (1996) of a Pd depleted UG2 seam at Schietfontein Farm. Hey (1999) demonstrated
depletion of Pd at the top of weathered UG2 seam (and its enrichment at the base of the
weathered horizon) at Union Section in the northwestern Bushveld Complex, indicating that
depletion may be important when interpreting PGE content. Buick et al. (2000) demonstrated
that retrogression due to channelled infiltration of fluids from dehydrating country rocks could
affect cooling of the RLS mafic rocks.

The presence of replacement magnetite in the form of stringers and lenses within the UG2
chromitite, and the presence of quartz-biotite filled fractures/veins with the UG2 layer at De Wildt
suggests that the chromitite was also exposed to late stage magmatic fluids. The very coarse-grained
and often pegmatoidal character of the UG2 footwall feldspathic pyroxenite at De Wildt is arguably
the result of recrystallization in response to secondary fluid activity. Such textures are common on the
UG2 footwall feldspathic pyroxenites in other parts of the BC. For example, Davey (1992) reported an
occurrence of a UG2 footwall at Marikana in the western limb of the BC (west of De Wildt prospect)
99
that consisted of coarse-grained feldspathic pyroxenite and which hosted chromitite stringers
averaging ? 1 m in thickness. Regardless, a 4 m wide UG2 footwall pegmatoidal feldspathic
pyroxenite hosting lenses and blebs of chromitite may be unique to De Wildt. The formation of the
UG2 pegmatoid has been explained in terms of fluidization processes, in which the chromitite may
have acted as a barrier to ascending late magmatic fluids (Mathez and May, 2005). Maier and Barnes
(2008) suggested that fluidization may have aided downward injection of the chromitite into their
footwall. Similarly, in the De Wildt project area, the chromitite lenses and blebs that characterize the
mineralized UG2 footwall up to 5 m below the UG2 seam may indicate that the amount of late stage
fluid infiltration was higher than usual. As such a greater portion of the basal parts of UG2 seam was
physically eroded, resulting in a wide UG2 footwall horizon with chromite lenses and blebs. In
concurrence, Penberthy and Merkle (1999) suggested that the unusual elevated PGE content within the
UG2 chromitite footwall elsewhere in the BC was caused by the physical erosion of the lowermost
part of UG2 by the underlying pegmatoid. The formation of the UG2 pegmatoid has been shown to
postdate the formation of UG2 (Hiemstra, 1985).
The occurrence of a UG2 footwall layer consisting of a coarse-grained norite unit hosting blebs
and lenses of chromite, as opposed to the usual feldspathic pyroxenite layer, tentatively suggests the
presence of a UG2 pothole. The contact between the UG2 and its footwall pyroxenite layer has been
shown to be erosional and in UG2 potholes, the footwall units are truncated, transgressed and eroded
(Viljoen, 1994, Lomberg et al., 1999). As such, the absence of the usual UG2 footwall pyroxenite
layer, coupled with lack of evidence of faults and other structural disturbances, is attributed to
potholing of the UG2 chromitite layer.
The Lower Zone and the Lower Critical Zone were not intersected at De Wildt. The fact that the
Lower and Lower Critical Zone were not intersected compares well with studies by Kruger (2005)
who documented that the Lower and Lower Critical Zones of the BC are only restricted to a belt
between Rustenburg and Northam in the western limb of the Bushveld. Similarly, Roberts (1992)
noted that the maximum thickness of BC (including the Ruighhoek pyroxenite and now known to
constitute the LCZ of the RLS) lies to the west of the De Wildt project area but is absent to the east of
the De Wildt project area.
The Merensky Reef-UG2 separation at De Wildt generally ranges between 114 m and 188 m
across the project area. However, in some boreholes such as PDW02 and DWD09, the Merensky
Reef-UG2 separation is significantly attenuated to less than 30 m. The cause of this is ascribed
possibly to Merensky potholes in which the underlying footwall layers were transgressed and eroded.
In borehole DWD13, the Merensky Reef pyroxenite overlies a disturbed UG2 chromitite layer. This
borehole is interpreted to have intersected a major pothole in which both the Merensky Reef
pyroxenite and UG2 chromitite layer have been significantly affected or disturbed.
IRUP bodies/zones were scattered throughout the Merensky Reef-UG2 interval at De Wildt. In
borehole DWD12 from Krelingspost Farm, the Merensky Reef together with surrounding norite units
was completely replaced by IRUPs. Eales and Cawthorn (1996) attributed the formation of IRUPs to
100
the passive replacement of existing plagioclase-rich layered rocks by downward magmatic fluid
percolation. As such it is likely that the Merensky Reef and other cumulates lying in this interval were
subjected to late stage fluids associated with Upper Zone magmas that completely metasomatized and
replaced them to the Iron-Rich Pegmatoid. Boreholes collared in the far eastern portion of
Krelingspost Farm indicate that the UG2 layer lies close to the Marginal Zone (Figs. 49a-b) which
implies thinning and wedging of the Upper Critical Zone towards the east across the project area. On
this farm, the UG2 layer may overlie the Marginal Zone rocks (and probably the Transvaal floor rocks
as well) as the Critical Zone thins against the floor rocks.

8.3. Metasedimentary xenoliths

Xenoliths of metasedimentary rocks derived from the underlying Transvaal Supergroup occur in
the eastern zone of the De Wildt project area but become rare towards the west. The most dominant
xenolith rock type is feldspathic quartzite (Fig. 50). The discontinuous quartzite hills on the
Krelingspost Farm are interpreted as large floor rock upfolds/highs associated with the main
Magaliesberg Mountain Ridge south of the prospect. Hartzer (1995) attributed the occurrence of
deformed Transvaal Supergroup country rocks within the BC to various mechanisms including (1)
pre-intrusive folding and doming, (2) syn-intrusive xenolith development or doming, and (3) post-
intrusive thrusting.
A relatively high percentage of metasedimentary units of variable composition were intersected
in the De Wildt project area. The presence of these rocks indicates that the Critical Zone lies in close
proximity to the Transvaal Supergroup floor rocks. As such, the xenoliths encountered are interpreted
to have rafted-off the Transvaal floor rock and were subsequently incorporated into the magma during
the emplacement of the BC. Antonie (2004) noted that zones of increased gamma radiation in the K
channel within the De Wildt prospect from radiometric data appeared to delineate basement inliers
although this cannot be conclusively established from the limited outcrop and radiometric-dataset
available. The De Wildt geology is comparable to the Kliprivier area (Dube, 2006) in the eastern limb
of BC in terms of the abundance of floor rock xenoliths within mafic rocks, and the occurrence of
wide reefs with diluted PGE grades. The two prospects are situated at the margins of the BC.
The small gravity anomaly to the east of the Brits graben is situated a few kilometres east of the
De Wildt prospect and was interpreted by Viljoen (1999) as a feeder zone or magma conduit. This
additional point of magma injection may be responsible for the existence of the highly thickened and
norite-dominant Upper Critical Zone at De Wildt, and also for the relatively structurally complex
stratigraphy prevailing. Other localities in the BC where main feeder zones are interpreted (such as the
Amandelbult-Northam areas) are characterized by a relatively complicated stratigraphy (Kruger,
2005). However, Kruger (2005) cautioned interpretation of positive gravity anomalies as representing
feeder zones arguing that the anomalies are likely to be enhanced by the inward dip and surface
exposure of enormous volumes of dense mafic rocks.
101
A B
N S W E
UG2 layer rests directly on the floor rocks at depth Thinning of the Upper Critical Zone,
Lower stratigraphic units of this zone progressively wedge out towards the east
Main Zone
Critical Zone
Marginal Zone
Merensky Reef pyroxenite
UG2 Chromitite layer
Legend
Transvaal Supergroup
Main Zone
Critical Zone
Marginal Zone
Merensky Reef pyroxenite
UG2 Chromitite layer
Legend
Transvaal Supergroup

Figure 49: Schematic representation of the stratigraphic succession on the eastern zone of the project area (at Krelingspost Farm and eastward) toward the margin of the BC. In
(A) the UG2 chromitite layer is interpreted to abut the floor rocks of the Transvaal Supergroup at depth. In (B) the wedging out of the Upper Critical Zone occurs i.e. the lower
stratigraphic units are cut out against the floor rocks towards the east. This also explains why there was no intersection of the UG1 layer in the eastern portions of the prospect
and the absence of UG2 layer out-crop in the vicinity of the quartzite hills.

102

Figure 50: Potassic quartzite hornfels intersected in borehole at Krelingspost Farm. It is not known whether this
unit is connected to the floor or comprises a Transvaal Supergroup outlier.

8.4. Metamorphism

Metasedimentary xenoliths that are interpreted as enclaves of the floor rocks were observed in
boreholes throughout the De Wildt project area. The xenoliths are ubiquitously deformed and
characterized by micro-structures. Brittle deformation features include micro-fissures/fractures and
micro-faults in crosscut to quartz crystals in quartzite. They probably formed during retrograde contact
metamorphism associated with cooling of the BC magma or cooling of post-BC dolerite intrusions.
Late stage or secondary minerals such as magnetite and chlorite fill these microstructures. Other
secondary minerals including biotite occur on the contact of hornfelsed quartzite and dolerite and may
be a result of fluid activity (metasomatism) after intrusion of dolerite dykes. The replacement of
plagioclase by sericite may indicate interaction of the BC rocks with magmatic fluids associated with
dyke intrusion and/or fluids from the dehydration of floor rocks. The occurrence of contact
metamorphosed xenoliths within the RLS has been reported elsewhere in the BC (Wallmach et al.,
1989; 1995).
Mbonambi (2007) noted the presence of an interchange between quartz and tridymite in
microstructures in samples of deformed quartzite xenoliths from drill-core at De Wildt and suggested
the interchange indicated prograde metamorphism and subsequently retrograde metamorphism. The
presence of feldspathic (potassium-rich) quartzite xenoliths at De Wildt (Fig. 50) may be explained in
terms of metasomatism; the interaction of potassium-rich fluids with quartzite xenoliths may have
resulted in chemical reactions between quartz (SiO2) and potassium anions to form a relatively stable
assemblage such as the K-feldspar-rich quartzite that is abundant in the prospect.
Overall, the presence of quartz-norite and quartzite xenoliths within the Upper Critical Zone at
De Wildt indicates a close spatial relationship of BC mafic units to the Transvaal Supergroup
103
metasedimentary rocks and the absence of the Lower Zone and Lower Critical Zone, which normally
underlie the Upper Critical Zone in a complete succession of the Rustenburg Layered Suite.
Hornblende-hornfels facies metamorphism of floor rock xenoliths is characteristic for the De Wildt
prospect. The minerals epidote, actinolite and chlorite are characteristic of greenschist facies in the
temperature range 300-500?C (Lapidus et al., 1990) and low pressure. The type and extent of
metamorphism may have been controlled by the composition of the protolith, i.e., the Pretoria Group
of the Transvaal Supergroup.

8.5. Structure and deformation pattern

The NW-trending mineral lineation defined by feldspars within a norite outcrop at Zilkaatsnek
Farm is parallel to the regional trend of lineaments such as dykes and faults. The regional trend of the
lineaments is likely to have been influenced by the adjacent Brits graben. The interpreted sequence of
deformational events at De Wildt is;

1. Formation of NE- and NW-trending scissor faults with west block down.
2. Intrusion of the NE-trending dolerite dykes (Old Dyke) along NE-trending joints and faults
(Fig. 14).
3. Intrusion of NW-trending dolerite dykes into pre-existing fracture-fault systems.

The formation of NE- and NW-trending scissor faults may be associated with the formation of
drag and/or rollover folds, which would account for the ?chaotic? magnetic signature in the central
zone giving the impression of a highly disturbed and/or folded zone. Certainly, the Merensky Reef and
UG2 layers are intricately disrupted by a series of NE-trending faults at Zilkaatsnek and Schietfontein
Farms. In these portions, blocks of mafic rocks have been down-thrown or uplifted (horst and graben)
with the implication that the UG2 layer was intersected at confusingly shallow or deeper depths than
expected.
Furthermore there is a strong possibility that some distortion or enhancement of magnetic
signatures may have occurred as a result of the presence of the highly magnetic IRUP bodies scattered
throughout the prospect, as evidenced by their intersections in a number of boreholes. An extensive
east-west trending layered sill consisting of meta-quartz dolerites was reported by Roberts (1992)
south of the De Wildt project area.
.
104
CHAPTER 8

CONCLUSIONS

Airborne magnetic and drill core data revealed that the three farms making up the De Wildt
project area are highly disturbed by faults, potholes, intruded by dyke swarms and modified by the
presence of floor rock up-folds and xenoliths. A full stratigraphic succession of the Rustenburg
Layered Suite of the Bushveld Complex is not developed at De Wildt; the Lower Zone (LZ) and
Lower Critical Zone (LCZ) are not developed. As such the Upper Critical Zone was developed in
close proximity to the floor rocks.
The Merensky Reef in the De Wildt project area, while exhibiting its regional character (non-
pegmatoidal, wide-type of the distal facies) has undergone further thickening up to about 20 m.
Thinning or wedging-out of the Upper Critical Zone stratigraphic succession towards the east, was
observed in the vicinity of basement highs/upfolds. Both the Merensky Reef and UG2 layers were
generally correlated along strike within the farm portions. However, the UG2 layer exhibited an extra-
ordinary lateral variation from west to east.
On Zilkaatsnek Farm, the UG2 layer was massive and well packaged while at Schietfontein and
Krelingspost farms, it became split and multi-layered, and in some cases only consisted of
disseminated chromite and chromite lenses within a feldspathic pyroxenite. The thickness variations of
the UG2 layer and its localized lack of lateral continuity is attributed to; (1) the presence of large
potholes, (2) the presence of basement/floor-rock up folds and xenoliths and (3) post-Bushveld
faulting. While the splitting of the UG2 seam may not be fully understood, it does suggest that
multiple magma injections took place during the crystallization processes to form the chromitites. The
relatively high Pt:Pd ratios of the UG2 chromitite at De Wildt indicates that the UG2 layer has been
exposed to late stage magmatic and/or meteoric fluids with the subsequent remobilization of Pd.
The NW trend of dykes at De Wildt is a regional feature and their emplacement is interpreted to
have been determined/controlled by the pre-existing structural lineaments (fractures and faults)
associated with formation of the Brits graben. Fault formation at De Wildt is interpreted as antithetic
to that associated with the formation of the Brits graben. Dyke-related metamorphism of the BC mafic
rocks at De Wildt is insignificant. Instead, contact metamorphism of Transvaal Supergroup
sedimentary xenoliths is intense.

105
REFERENCES

Acocks, J.P.H., 1988. Veld types of South Africa. Memoirs of the Botanical Survey of South Africa
57: 1-146.
Antoine, L., 2004. Memorandum to Nkwe Platinum (Pty) Limited (07 January 2004): Technical note
on the results of the airborne geological survey by Fugro Services, De Wildt project, western
Bushveld. Internal company report to Nkwe Platinum (Pty) Limited (unpublished).
Ashwal, L.D, Webb, S.J., Knoper, M.W., 2005. Magmatic stratigraphy in the Bushveld Northern
Lobe: continuous geophysical and mineralogical data from the 2950 m Bellevue drillcore. South
African Journal of Geology 8(2), 199-232.
Boudreau, A.E., 1992. Volatile fluid overpressure in layered intrusions and the formation of potholes.
Australian Journal of Earth Sciences 39, 277? 287.
Bristow, D.G., Cawthorn, R.G., Harmer, J., Lee, C.A., Tegner, C., Viljoen, M.J., Walraven, F.,
Wilson, J.R., 1993. Symposium on layering in igneous complexes. Field excursion to the
Bushveld Complex, 11th-17th September, 1993, 60p.
Buick, I.S., Gibson, R.L., Cartwright, I., Maas, R., Wallmach, T., Uken, R., 2000. Fluid flow in
metacarbonates associated with emplacement of the Bushveld Complex, South Africa. Journal
of Geochemical Exploration 60-70, 391-395.
Burger, A.J., Coertze, F.J., 1973-1974. Age determinations - April 1972 to March 1974. Annals of the
Geological Survey of South Africa 10, 135-141.
Butcher, A.R., Merkle, R.K.W., 1991. Unsual textures and structures associated with a magnetite layer
in the Bushveld Complex: a contribution to the adcumulate debate. Mineralogical Magazine 55,
465-477.
Butcher, P. J., Merkle, M., 1991. Structure and rock sequences of the Critical Zone of the eastern
Bushveld Complex. Special Paper, Mineralogical Society of America 1, 93-107.
Button, A., 1986. The Transvaal sub-basin of the Transvaal Sequence. In: Anhaeusser, C.R., Maske,
S., (Eds.), Mineral deposits of Southern Africa, Volume I. Geological Society of South Africa,
Johannesburg, pp. 811?817.
Carr, H.W., Kruger, F.J., Groves, D.I., Cawthorn, R.G. 1999. The petrogenesis of Merensky Reef
potholes at the western platinum mine, Bushveld Complex: Sr-isotopic evidence for
synmagmatic deformation. Mineralium Deposita 34, 335-347.
Campbell, I.H., 1986. A fluid dynamic model for the potholes of the Merensky Reef. Economic
Geology 81, 1118?1125.
Cawthorn, R.G., Webb, S.J., 2001. Connectivity between the western and eastern limbs of the
Bushveld Complex. Tectonophysics 330, 195-209.
Cawthorn, R.G. 2004 Origin of variation of Pt/Pd values in the UG2 chromitite layer, Bushveld
Complex. Geoscience Africa, University of the Witwatersrand, July 2004, pp. 110-113.
Cawthorn, R.G., 2005. Pressure fluctuations and the formation of PGE-rich Merensky and chromitite
reefs. Mineralium Deposita 40, 231-235.
106
Cawthorn, R.G., Eales, H.V., Walraven, F., Uken, R., Watkeys M.K., 2006. The Bushveld Complex.
In: Johnson, M.R., Anhaeusser, C.R., Thomas, R.J., (Eds.), The Geology of South Africa.
Geological Society of South Africa. Johannesburg/Council for Geoscience, Pretoria, pp. 261-
281.
Chief Directorate: Survey and Mapping, 1996. 2527DB 1:50 000 geological map sheet Brits. South
Africa (Fifth Edition, 1996), Republic of South Africa.
Clarke, B., Uken, R., Reinhardt, J., 2008. Structural and compositional constraints on the emplacement
of the Bushveld Complex, South Africa. Lithos 11, 21-36.
Clarke, B.M., Uken, R., Watkeys, M.K., Reinhardt., 2005. Folding of the Rustenburg layered suite
adjacent to the Steelpoort pericline: implications for syn-Bushveld tectonism in the eastern
Bushveld Complex. South Africa Journal of Geology 108, 397-412.
Cousins, C.A., 1959a. The Bushveld Igneous Complex, The geology of South Africa?s Platinum
resources. Platinum Metals Review 3, 94-99.
Cousins, C.A., Feringa, G., 1964. The chromite deposits of the western belt of the Bushveld Complex.
In: Haughton S.H., (Ed.) The geology of some ore deposits in southern Africa. Special
Publications of the Geological Society of South Africa 2, 183-202.
Dankert, B.T., Hein, K.A.A., 2010. Evaluating the structural character and tectonic history of the
Witwatersrand Basin. Precambrian Research 177, 1-22.
Davey S.R., 1992. Lateral variations within the Upper Critical Zone of the Bushveld Complex on the
farm Rooikoppies 297 JQ, Marikana, South Africa. South African Journal of Geology 95, 141-
149.
Davison, G.E., Chunnett, G.K., 1999. Seismic exploration for Merensky Reef: The way ahead. South
African Journal of Geology 102, 261-267.
De Klerk, W.J., 1995. Textures exhibited by feldspars in the Giant Mottled Anorthosite (GMA) of the
Bastard unit in the Upper Critical Zone, Western Bushveld Complex. Mineralogy and Petrology
54, 25-34.
Eales, H.V., 2001. A first Introduction to the geology of the Bushveld Igneous Complex and those
aspects of South African geology that relate to it. The Council of Geosciences 1, pp. 7.
Eales, H.V., Cawthorn, R.G., 1996. The Bushveld Complex. In: Cawthorn, R.G. (Ed.) Layered
Intrusions. Developments in Petrology 15, 103-145 and 181-229.
Eriksson, P.G., Altermann, W., Hartzer, F.J., 2007. The Transvaal Supergroup and its Precursors. In:
Johnson, M.R., Anhaeusser, C.R., Thomas, R.J., (Eds.), The Geology of South Africa.
Geological Society of South Africa, Johannesburg/Council for Geoscience, Pretoria, pp. 237-
260.
Eriksson, P.G., Hattingh, P.J., Altermann, W., 1995. An overview of the geology of the Transvaal
Sequence and Bushveld Complex, South Africa. Mineralium Deposita 30, 98-111.
Eriksson, P.G., Schweitzer, J.K., Bosch, P.J.A., Schreiber, U.M., Van Deventer, J.L., Hatton, C.J.,
1993. The Transvaal Sequence: an Overview. Journal of African Earth Sciences 16, 25-51.
Farquhar, J., 1986. The Western Platinum Mine. In: C.R., Maske, S., (Eds.), Mineral Deposits of
Southern Africa, Volume II. Geological Society of South Africa. Johannesburg, pp. 1135-1142.
107
Frimmel, H.E., Minter, W.E.L., 2002. Recent Developments Concerning the Geological History and
Genesis of the Witwatersrand Gold Deposits, South Africa. In: R. Goldfarb and R.L. Nielsen
(Eds.), Global Exploration 2002 - Integrated Methods for Discovery. Special Publication.
Society of Economic Geologists, Littleton, pp. 17-45.
Gerya, T.V., Uken, R., Reinhardt, J., Watkeys, M.K., Maresch, W.V., Clarke, B.M.., 2004. Cold
diapirs triggered by intrusion of the Bushveld Complex: Insight from two dimensional
numerical modelling. In: Whitey, D.L., Teyssier, C., and Siddoway, C.S., (Eds.) Gneiss domes
in orogeny: Boulder, Colorado, Geological Society of America Special Paper 380, 117-127.
Hall, A.L., 1932. The Bushveld Igneous Complex of Central Transvaal. Memoir 28, Union of South
Africa Geological Survey, Government Printer, Pretoria, 1932.
Hammond, H., 1998. Structural Interpretation of Aeromagnetic data; a rudimentary guide.
Unpublished technical memo to North Limited, Australia, pp. 1-46.
Harmer R.E., Sharpe M.R., 1985. Field relations and strontium isotope systematics of the marginal
rocks of the Eastern Bushveld Complex. Economic Geology 80: 813?837.
Harmer, R.E., 1985. Rb-Sr isotopic study of units of the Penaars River Alkaline Complex, north of
Pretoria, South Africa. Transactions of the Geological Society of South Africa 88, 215-223.
Hartzer, F.J., 1995. Transvaal Supergroup inliers: geology, tectonic development and relationship with
Bushveld Complex, South Africa. Journal African Earth Sciences 21, 521-547.
Hatton, C.J., Schweitzer, J.K., 1995. Evidence for synchronous extrusive and intrusive Bushveld
magmatism. Journal of African Earth Sciences 21, 579-594.
Hatton, C.J., Von Gruenewaldt, G., 1985. The geological setting and petrogenesis of the Bushveld
chromitite layers. Research. Report., Institute of Geological Research on the Bushveld
Complex, University of Pretoria 57, 1-48.
Hey, P.V., 1999. The effects of weathering on the UG2 Chromitite reef of the Bushveld Complex,
with special reference to platinum group minerals. South African Journal of Geology 102, 251-
260.
Hiemstra, S.A., 1985. The distribution of some platinum-group elements in the UG2 Chromitite layer
of the Bushveld Complex. Economic Geology 80, 944-957.
Judeel, G. du T., Hartmann, A.W., 2008. Ground support at Crocodile River Mine located in the Brits
graben of the western limb of the Bushveld Complex. Abstracts, 6th International Symposium on
Ground Support in mining and Civil Engineering Construction, Capetown 30 March ? 3 April
2008, pp. 525-544.
Kinnaird, J.A., Kruger, F.J., Nex, P.A.M., Cawthorn, R.G., 2002. Chromite formation ? a key to
understanding processes of platinum enrichment. Transactions of the Institute of Mining and
Metallurgy 111, B23-B35.
Kinloch, E.D., Peyerl, W., 1990. Platinum Group minerals in various rock types of the Merensky
Reef: Genetic implications. Economic Geology 85, 537-555.
Kruger, F.J., 2005. Filling the Bushveld Complex magma chamber: lateral expansion, roof and floor
interaction, magmatic unconformities, and the formation of giant chromitite, PGE and Ti-V-
magnetite deposits. Mineralium Deposita 40, 451-472.
Lapidus, D.F., 1990. Collins dictionary of geology. Harper Collins publishers, UK, pp. 265.
108
Leeb-Du Toit, A., 1986. The Impala Platinum Mines. In: Anhaeusser, C.R., Maske, S., (Eds.),
Minerals Deposits of Southern Africa, Volume II. Geological Society of South Africa.
Johannesburg, pp. 1091-1106.
Lipin, B.R., 1993. Pressure increases, the formation of chromite seams, and the development of the
Ultramafic series in the Stillwater Complex, Montana. Journal of Petrology 34, 955-976.
Lomberg, K., Rupprecht, S., Cunningham, G., 2006. Crocodile River Mine, South Africa. Independent
Technical Report prepared for RSG Global on behalf of Barplats Investment Limited, 122p.
Lomberg, K.G., 1991. Motivation to relinquish the options to the mineral rights to portions of
Zilkaatsnek 439 JQ, Geological Memorandum No. 188, Platinum and Bushveld Exploration
Mineral Resource Division-Rotund Project, 22 October 1991.
Lomberg, K.G., Martin, E.S., Patterson, M.A., Venter, J.E., 1999. The morphology of potholes in the
UG2 Chromitite layer and Merensky Reef (pothole reef facies) at Union Section, Rustenburg
Platinum Mines. South African Journal of Geology102, 209-220.
Lurie, J., 1986. Mineralization of the Pilanesberg Alkaline Complex. In: Anhaeusser, C.R., Maske, S.,
(Eds.), Mineral Deposits of Southern Africa, Volume II. Geological Society of South Africa,
Johannesburg, pp. 2215-2228.
Maier, W., Barnes, S.J., 2008. Platinum-group elements in the UG1 and UG2 chromitites and the
Bastard reef, at Impala platinum mine, western Bushveld Complex, South Africa: Evidence for
late magmatic cumulate instability and reef constitution. South African Journal of Geology 111,
159-176.
Maier, W.D., 2005. Platinum-group element (PGE) deposits and occurrences: mineralization styles,
genetic concepts, and exploration criteria. Invited presidential review, Journal of Earth Sciences
41, 165-191.
Maier, W.D., Eales, H.V., 1997. Correlation within the UG2-Merensky Reef interval of the Western
Bushveld Complex, based on geochemical, mineralogical and petrological data. Geological
Survey South Africa Bulletin, 120, 56.
Maier, W.D., Bowen, M.P., 1996. The UG2-Merensky Reef interval of the Bushveld Complex,
northwest of Pretoria. Mineralium Deposita 31, 386-393.
Martin, D. McB., Clendenin, C.W., Krape?, B., McNaughton, N.J., 1998. Tectonic and
geochronological constraints on late Archaean and Palaeoproterozoic stratigraphic correlation
within and between the Kaapvaal and Pilbara Cratons. Journal of the Geological Society 155,
311-322.
Mathez, E.A., Mey, J.l., 2005. Character of the UG2 chromitite and host rocks and petrogenesis of its
pegmatoidal footwall, northeastern Bushveld Complex. Economic Geology, 100, 1617-1630.
Mbonambi, H.M., 2006. Relative Chronology of Intrusive, Structural and Metamorphic Events
through Petrographic and Microstructural Investigation of Diamond Drill Core, De Wildt area,
Brits. Unpublished B.Sc. (Hon) research draft, University of the Witwatersrand Johannesburg.
Mitchell, A.A., Manthree, R., 2002. The Giant Mottled Anorthosite: a transitional sequence at the top
of the Upper Critical Zone of the Bushveld Complex. South African Journal of Geology 105,
15-24.
Moen, F.H.G., 1989. A petrological study of Bushveld-type rocks from boreholes on the Moloto
geophysical anomaly, northeast of Pretoria. South African Journal of Geology 92, 84-94.
109
Moen, R.T., 1989. The geochemistry of pyroxenites at 12#, Impala platinum mine. Technical report.
Naldrett, A.J., Wilson, A., Kinnaird, J., Chunnett, G., 2009. PGE Tenor and Metal Ratios within and
below the Merensky Reef, Bushveld Complex: Implications for its Genesis. Journal of
Petrology 50, 625-659.
Pattison, D.R.M., Tracy, R.J., 1991. Phase equilibria and thermobarometry of calcareous, ultramafic
and mafic rocks, and iron formations. In Kerrick, D.M., (Ed.) Contact metamorphism. Reviews
in Mineralogy 26, 105-206. (Mineralogical Society of America.)
Penberthy, C.J., Merkle, R.K.W., 1999. Lateral variations in the platinum-group element content and
mineralogy of the UG2 Chromitite layer, Bushveld Complex. South African Journal of Geology
102, 240-250.
Roberts, K., 1992. An investigation of the floor and marginal rock-layered rock relationships of the
Bushveld Complex, northwest of Pretoria. Unpublished Master of Science thesis, Geological
Survey (Pretoria), 131p. ISBN: 0621144177.
Rozendaal, A., Toros, M., Anderson, J.R., 1986. The Rooiberg tin-deposits, West-central Transvaal.
In: Anhaeusser, C.R., Maske, S., (Eds.), Mineral Deposits of Southern Africa, Volume II.
Geological Society of South Africa, Johannesburg, 1307-1328.
SACS (South Africa Committee for Stratigraphy), 1980. Stratigraphy of South Africa. Part 1: (Kent,
L.E, Compiler), Lithostratigraphy of the Republic of South Africa, South West Africa/Namibia,
and the Republics of Bophuthatswana, Transkei and Venda. Geological Survey of South Africa
Handbook 8, 690p.
Schiffries, C.M., Rye, D.M., 1990. Stable isotope systems of the Bushveld Complex: Constraints on
hydrothermal processes. American Journal of Science 290, 209-245.
Sch?rmann LW, Grabe PJ, Steenkamp CJ (1998) Chromium. In: Wilson M.G.C., Anhaeusser, C.R.
(Eds.) Mineral Resources of South Africa, 6th Edition. Council Geoscience, Handbook, Volume
16, pp 90?105.
Scoon, R.G., Teigler, B., 1995. A new LG-6 chromitite reserve at Eerste Geluk in the boundary
between the central and southern sectors of the eastern Bushveld Complex. Economic Geology
90, 969-982.
Scoon, R.N., 2002. A new occurrence of Merensky Reef on the flanks of the Zaaikloof Dome,
northeastern Bushveld Complex: Relationship between diapirism and magma replenishment.
Economic Geology 97, 1037-1049.
Scoon, R.N., Mitchell, A.A., 1994. Discordant iron-rich ultramafic pegmatites in the Bushveld
Complex and their relationship to iron-rich intercumulus and residual liquids. Journal of
Petrology 35, 881-917.
Sharpe, M. R., Chadwick, B., 1982. Structures in Transvaal sequence rocks within and adjacent to the
eastern Bushveld Complex. Transactions of the Geological Society of South Africa 85, 29?41.
Stevens, G., Gibson, R.L., Droop, G.T.R., 1997. Mid-crutal granulite facies metamorphism in the
Central Kaapvaal craton. The Bushveld Complex connection. Precambrian Research 82, 113-
132.
Tankard, A.J., Jackson, M.P.A., Eriksson, K.A., Hobday, D.K., Hunter, D.R., Minter, W.E.L., 1982.
Crustal Evolution of Southern Africa, 3.8 Billion Years of Earth History. Published by
Springer-Verlag (New York, N.Y.), pp.6 and 174.
110
Tiegler, B., Eales, H.V., Scoon, R.N., 1992. The cumulate succession in the Critical Zone of the
Rustenburg Layered Suite at Brits, western Bushveld Complex. South African Journal of
Geology 95, 17-28.
Uken, R., 1998. The Geology and Structure of the Bushveld Complex metamorphic aureole in the
Olifants River area. Doctor of Philosophy thesis, University of Natal Durban, South Africa,
277p.
Uken, R., Watkeys, M.K., 1997. Diapirism initiated by the Bushveld Complex, South Africa. Geology
25, 723-726.
Van der Merwe, J., Cawthorn, R.G., 2005. Structures at the base of the Upper Group 2 chromitite
layer, Bushveld Complex, South Africa, on Karee Mine (Lonmin Platinum). Lithos 83, 214-
228.
Van der Merwe, M.J., 2008. The geology and structure of the Rustenburg Layered Suite in
Potgietersrus/Mokopane area of the Bushveld Complex, South Africa. Mineralium Deposita 43,
405-420.
Vermaak, CF., 1976. The Merensky Reef- thoughts on its environment and genesis. Economic
Geology 71, 1270-1298.
Vermaak, C.F., 1995. The Platinum Group Metals; a global perspective. Published by MINTEK
(South Africa), ISBN-10: 0869999265, 247p.
Vermaak, C.F., Hendricks, L.P., 1976. A review of the mineralogy of the Merensky Reef, with
specific reference to new data on the precious metal mineralogy. Economic Geology 77, 1244-
1269.
Vermaak, C.F., Van Der Merwe, M.J., 2000. The platinum mines and deposits of the Bushveld
Complex, South Africa. Council for Mineral Technology, Randburg, South Africa, 118 pp.
Vermaak, C.F., Von Greunewaldt, G., 1986. Introduction to the Bushveld. In: Anhaeusser, C.R.,
Maske, S., (Eds.), Minerals Deposits of Southern Africa, Volume II. Geological Society of
South Africa, Johannesburg, pp. 1021-1029.
Viljoen, M.J., 1994. A review of regional variations in facies and grade distribution of the Merensky
Reef, Western Bushveld Complex with some mining implications. Journal of the South African
Institute of Mining and Metallurgy 3, 183-194.
Viljoen, M.J., 1999. The nature and origin of the Merensky Reef of the western Bushveld Complex
based on geological facies and geophysical data. South African Journal of Geology 102, 221-
239.
Viljoen, M.J., Hieber, R., 1986. The Rustenburg Section of Rustenburg Platinum Mines Limited with
reference to the Merensky Reef. In: Anhaeusser, C.R., Maske, S., (Eds.), Mineral deposits of
Southern Africa, Volume II. Geological Society of South Africa, Johannesburg, pp. 1107-1134.
Viljoen, M.J., Sch?rmann, L.W., 1998. Platinum Group Metals. In: Wilson, M.G.C., Anhaeusser,
C.R., (Eds.) Mineral Resources of South Africa, 6th edition. Handbook Council for Geosciences
16, pp. 532-568.
Viljoen, M.J., Theron, J., Underwood, B., Walters, B.M., Weaver, J., Peyerl, W., 1986. The
Amandelbult Section of Rustenburg Platinum Mines Limited, with reference to the Merensky
Reef. In: Anhaeusser, C.R., Maske, S., (Eds.), Minerals Deposits of Southern Africa, Volume II.
Geological Society of South Africa, Johannesburg, pp. 1041-1060.
111
Wallmach, T., Hatton, C.J., Droop, G.T.R., 1989. Extreme facies of contact metamorphism developed
in calc-silicate xenoliths in eastern Bushveld Complex. Canadian Mineralogist 27, 509-523.
Wallmach, T., Hatton, C.J., De Waal, S.A., Gibson, R.L., 1995. Retrogressive hydration of calc-
silicate xenoliths in the eastern Bushveld Complex: evidence for late magmatic fluid movement.
Journal of African Earth Sciences 21, 633-646.
Walraven, F., 1997. Geochronology of the Rooiberg Group, Transvaal Supergroup, South Africa.
Information Circular, 316. Economic Geology Research Unit, University of the Witwatersrand,
21 p.
Walraven, F., Armstrong, R.A., Kruger, F.J., 1990. A chronostratigraphic framework for the north?
central Kaapvaal Craton. Tectonophysics, 171, 23-48.
Wilhelm, H.J., Zhang, H., Chen, F.L., Elsenbroek, J.H., Lombard, M., de Bruin, D., 1997.
Geochemical exploration for platinum-group elements in the Bushveld Complex, South Africa.
Mineralium Deposita 32, 349-361.
112

APPENDIX
113

1.0. 1st vertical magnetic derivative map of the De Widlt project area.
114

2.0. Total magnetic intensity map of the De Widlt project area.
115

3.0. Ungeoreferenced raster images over the De Wildt project area and surrounding areas.

116

4.0. Flight plan over the De Widlt project area
117

5.1. Stratigraphic legend used in all boreholes

5.2. Stratigraphic log for borehole DWD01

118

5.3. Stratigraphic log for borehole DWD02

5.4. Stratigraphic log for borehole DWD2A

119

5.5. Stratigraphic log for borehole DWD3A

5.6. Stratigraphic log for borehole DWD04

120

5.7. Stratigraphic log for DWD05

5.8. Stratigraphic log for borehole DWD06
121

5.9. Stratigraphic log for DWD07

122

5.10. Stratigraphic log for borehole DWD08

123

5.11. Stratigraphic log for borehole DWD09

124

5.12. Stratigraphic log for borehole DWD10

125

5.13. Stratigraphic log for borehole DWD11

126

5.14. Stratigraphic log for borehole DWD12

127

5.15. Stratigraphic log for borehole DWD13

128

5.16. Stratigraphic log for borehole DWD14.

129

5.17. Stratigraphic log for borehole PDW01

130

5.18. Stratigraphic log for borehole PDW02

131

5.19. Stratigraphic log for borehole PDW03.

132

5.20. Stratigraphic log for borehole PDW04

133

5.21. Stratigraphic log for borehole SK02

134

5.22. Stratigraphic log for borehole SK03

135

5.23. Stratigraphic log for borehole SK04.

136

5.24. Stratigraphic log for borehole SK05.

137

5.25. Stratigraphic log for borehole SK06.

138

5.26. Stratigraphic log for borehole SK07.

139

5.27. Stratigraphic log for borehole SK08.

140

5.28. Stratigraphic log for borehole SK09.

141

5.29. Stratigraphic log for borehole SK10.

142

5.30. Stratigraphic log for borehole SK13.
143

6.0. UG2 chromitite and UG2 footwall grades at De Wildt project.
144

7.0. Prill split results for boreholes PDW01 and 2092.