Content uploaded by Wladyslaw Wlady Altermann
Author content
All content in this area was uploaded by Wladyslaw Wlady Altermann on Nov 28, 2018
Content may be subject to copyright.
ELSEVIER Sedimentary Geology 120 (1998) 225–256
Sedimentation rates, basin analysis and regional correlations of three
Neoarchaean and Palaeoproterozoic sub-basins of the Kaapvaal craton
as inferred from precise U–Pb zircon ages from volcaniclastic
sediments
Wladyslaw Altermanna,Ł, David R. Nelsonb
aInstitut fu¨r Allgemeine und Angewandte Geologie, Ludwig-Maximilians-Universita¨t, Luisenstraße 37, D-80333 Mu¨nchen, Germany
bGeological Survey of Western Australia, Department of Mines, 100 Plain Street, Perth, W.A., Australia
Received 29 April 1997; accepted 26 June 1997
Abstract
Calculation of sedimentation rates of Neoarchaean and Palaeoproterozoic siliciclastic and chemical sediments covering
the Kaapvaal craton imply sedimentation rates comparable to their modern facies equivalents. Zircons from tuff beds
in carbonate facies of the Campbellrand Subgroup in the Ghaap Plateau region of the Griqualand West basin, Transvaal
Supergroup, South Africa were dated using the Perth Consortium Sensitive High Resolution Ion Microprobe II (SHRIMP
II). Dates of 2588 š6 Ma and 2549 š7 Ma for the middle and the upper part of the Nauga Formation indicate that
the decompacted sedimentation rate for the peritidal flat to subtidal below-wave-base Stratifera and clastic carbonate
facies, southwest of the Ghaap Plateau at Prieska, was of up to 10 m=Ma, when not corrected for times of erosion and
non-deposition. Dates of 2516 š4 Ma for the upper Gamohaan Formation and 2555 š19 for the upper Monteville
Formation, indicate that some 2000 m of carbonate and subordinate shale sedimentation occurred during 16 Ma to
62 Ma on the Ghaap Plateau. For these predominantly peritidal stromatolitic carbonates, decompacted sedimentation
rates were of 40 m=Ma to over 150 m=Ma (Bubnoff units). The mixed siliciclastic and carbonate shelf facies of the
Schmidtsdrif Subgroup and Monteville Formation accumulated with decompacted sedimentation rates of around 20 B. For
the Kuruman Banded Iron Formation a decompacted sedimentation rate of up to 60 B can be calculated. Thus, for the
entire examined deep shelf to tidal facies range, Archaean and Phanerozoic chemical and clastic sedimentation rates are
comparable. Four major transgressive phases over the Kaapvaal craton, followed by shallowing-upward sedimentation,
can be recognized in the Prieska and Ghaap Plateau sub-basins, in Griqualand West, and partly also in the Transvaal
basin, and are attributed to second-order cycles of crustal evolution. First-order cycles of duration longer than 50 Ma
can also be identified. The calculated sedimentation rates reflect the rate of subsidence of a rift-related basin and can be
ascribed to tectonic and thermal subsidence. Comparison of the calculated sedimentation rates to published data from other
Archaean and Proterozoic basins allows discussion of general Precambrian basin development. Siliciclastic and carbonate
sedimentation rates of Archaean and Palaeoproterozoic basins equivalent to those of younger systems suggest that similar
mechanical, chemical and biological processes were active in the Precambrian as found for the Phanerozoic. Particularly
for stromatolitic carbonates, matching modern and Neoarchaean sedimentation rates are interpreted as a strong hint of a
similar evolutionary stage of stromatolite-building microbiota. The new data also allow for improved regional correlations
across the Griqualand West basin and with the Malmani Subgroup carbonates in the Transvaal basin. The Nauga Formation
ŁCorresponding author. E-mail: wlady.altermann@iaag.geo.uni-muenchen.de
0037-0738/98/$ – see front matter 1998 Elsevier Science B.V. All rights reserved.
PII S0037-0738(98)00034-7
226 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
carbonates in the southwest of the Griqualand West basin are significantly older than the Gamohaan Formation in the
Ghaap Plateau region of this basin, but are in part, correlatives of the Oaktree Formation in the Transvaal and of parts of
the Monteville Formation on the Ghaap Plateau. 1998 Elsevier Science B.V. All rights reserved.
Keywords: basin analysis; sedimentation rates; Archaean; Proterozoic; Kaapvaal craton; SHRIMP
1. Introduction
In the absence of biostratigraphic markers, high-
precision isotopic data on the age and duration of
sedimentation are essential aspects of the study of
Archaean and Proterozoic sedimentary basins. Pre-
cambrian siliciclastic basins containing thousands
of metres of sedimentary fill are often bracketed
by rare and imprecise stratigraphic data, and lat-
eral lithostratigraphic correlations lack arguments
other than similar facies development. As a conse-
quence, poorly constrained basin models and equiv-
ocal tectonic interpretations are commonly presented
for Precambrian sediments. Precambrian carbonate
basin-fills are equally vulnerable. More particularly,
the carbonate sedimentary processes and the mech-
anism of carbonate precipitation are generally not
well understood for the Archaean (see discussions
by Grotzinger, 1989, 1990; Sumner and Grotzinger,
1996). Although stromatolites and microbial remains
are known from older deposits, the earliest large car-
bonate platforms apparently developed in intracra-
tonic basins, following cratonic stabilization. This
was until recently ascribed to the Palaeoprotero-
zoic, around 2.5–2.0 Ma ago (Grotzinger, 1989).
With the development of new dating techniques,
it has now become apparent that the earliest large
carbonate platforms developed during the Neoar-
chaean, between 2700 Ma and 2500 Ma (Jahn et
al., 1990; Arndt et al., 1991; Hassler, 1993; Barton
et al., 1994). Consequently, the time span between
cratonization and subsequent carbonate basin devel-
opment is now believed to be shorter, with less than
1.0 billion years separating the formation of granite–
greenstone terranes at around 3.5 Ga to 3.0 Ga from
the formation of huge stromatolitic platforms in the
Neoarchaean (Beukes, 1986; Altermann and Her-
big, 1991; Jahn and Simonson, 1995; Altermann and
Siegfried, 1997). The rise of these platforms was
made possible by the widespread absence of clas-
tic input during periods of tectonic quiescence and
volcanic indolence. These two conditions are basic
prerequisites for chemical or bio-chemical precipi-
tation. In the presence of clastic detritus, microbial
organisms that facilitate carbonate precipitation can
be buried or swept away from the sediment surface
and from the water column, and inorganic precipita-
tion is hindered by the attachment of metal ions like
Ca and Fe to mineral grains. The scarcity of clastic
detritus thus also allows purely chemical precipi-
tates like Banded Iron Formations (BIF) to develop.
It is certainly not coincidental, that large Precam-
brian BIF provinces are often underlain by carbonate
platforms. Hence, the conspicuous carbonate (shale)
and BIF association must be explained not only in
terms of palaeoenvironmental atmospheric and hy-
drospheric evolution (Eriksson et al., 1998), but also
as a function of basin development (Simonson and
Hassler, 1997). Comparisons of sedimentation and
subsidence rates of clastic and chemical sedimen-
tary basins of the Precambrian and Phanerozoic, as
attempted here, may reveal important aspects of tec-
tonic history, rates of erosion and sediment transport,
genesis of mineral deposits and the evolution of
carbonate precipitating microbiota.
The Kaapvaal craton of southern Africa hosts
three major Archaean to Palaeoproterozoic sub-
basins, in which clastic and chemical sediments
and igneous rocks accumulated. The Transvaal basin
in the Transvaal geographic region, the Griqualand
West basin in the Northern Cape Province of South
Africa and the Kanye basin of Botswana share
lithostratigraphically similar deposits which uncon-
formably cover the 2.7 Ga old volcanic Ventersdorp
Supergroup (Armstrong et al., 1991). In this con-
tribution the Kanye basin is not discussed and the
Griqualand West basin is subdivided into the Prieska
sub-basin and Ghaap Plateau sub-basin, because of
their different development. Carbonates are volumet-
rically dominant rocks in the Prieska, Ghaap Plateau
and Transvaal sub-basins and, together with thin,
lowermost siliciclastic rocks, form the base of the
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 227
Transvaal Supergroup, being overlain by BIF de-
posits. The iron-rich chemical precipitatesare in turn
overlain by a thick sequence of predominantly clastic
sediments. Similar volcano-sedimentary basin devel-
opment can be deduced in other Archaean cratonic
terranes, but especially well on the Pilbara craton of
Western Australia, where the lithostratigraphic suc-
cession is strikingly similar to that of the Kaapvaal
craton (Cheney, 1996).
At first glance, the three sub-basins discussed
here host mainly chemical sediments, and thus might
appear unsuitable for a special volume on Precam-
brian clastic depositional systems. Nevertheless, we
feel that the sediments discussed herein impressively
demonstrate the interplay of clastic and chemical
sedimentation and its appearance in the geologic
record of the Precambrian. Moreover, the over-
whelming presence of the chemical sediments in
the discussed sections is misleading. As our calcula-
tions and age data demonstrate, clastic sedimentation
played a major role at different times in differ-
ent sub-basins. In some areas pelitic sedimentation
dominated the environment for periods longer than
50 m.y., with only short intervals occupied by car-
bonate sediments. Because of different compaction
behaviour, however, carbonates apparently dominate
the sedimentary record. Upon decompaction, silici-
clastic sediments would make up between one third
and half of the sedimentary section below the BIF.
The discussion of the development of the intracra-
tonic Griqualand West–Transvaal basin is based on
new age data presented herein, and on novel facies
interpretation of the sediments in question (Alter-
mann, 1997; Altermann and Siegfried, 1997). Subse-
quently, we argue the possible processes responsible
for the basin development and the widespread accu-
mulation of siliciclastic, biochemical and chemical
sediments. We also compare our data and inter-
pretation to other Precambrian examples from the
literature in an attempt to elaborate the principal
aspects of sediment accumulation for chemical and
clastic deposits during the Precambrian. Throughout
this contribution we use the detailed stratigraphic
subdivision of Beukes (1980a), but with some mod-
ifications for the Prieska sub-basin of Griqualand
West. A detailed discussion of various depositional
and stratigraphic models for the Griqualand West
and Transvaal carbonates is presented in Altermann
and Wotherspoon (1995) and in Altermann (1997).
General stratigraphyis shown in Figs. 1–3 and 7.
2. Regional geology and stratigraphy of
Griqualand West
The Vryburg Formation of the Schmidtsdrif Sub-
group (Beukes, 1979) of the Ghaap Group (Fig. 1)
is the lowest stratigraphic unit above the unconfor-
mity cutting into the 2709 Ma (Armstrong et al.,
1991) Ventersdorp Supergroup lavas in Griqualand
West. This formation consists of shales, quartzites,
siltstones and lava. According to the South African
Committee for Stratigraphy (SACS, 1980), it cor-
relates with the Black Reef Quartzite Formation in
Transvaal (Fig. 7). A lava in the Vryburg Formation
was dated by Walraven et al. (in press) at 2642 š3
Ma. Stromatolitic carbonates of the upper Schmidts-
drif and succeeding Campbellrand Subgroups con-
formably cover the Vryburg Formation. A tuff band
in the upper part of the Gamohaan Formation, at
the top of the Campbellrand Subgroup (Figs. 1, 3
and 7), was dated by Sumner and Bowring (1996)
at 2521 š3 Ma, giving a good approximation of
the minimum age of the Ghaap Plateau carbonates.
The carbonates are overlain by shales and subse-
quently by the Kuruman and Griquatown BIF of the
Asbestos Hills Subgroup (Fig. 1). The Griquatown
BIF has an age of 2432 š31 Ma (Trendall et al.,
1990). The Koegas Subgroup of mainly siliciclastic
deposits is conformably superimposed on the BIF
sediments, and is covered by the Makganyene glacial
deposits of the Postmasburg Group with a regional
unconformity (Figs. 1 and 7). Again unconformably,
the 2222 š13 Ma old (Cornell et al., 1996) Ongeluk
basaltic andesite formation covers the glacial tillite
(Altermann and Ha¨lbich, 1991).
The only continuous section through the
Schmidtsdrif and Campbellrand strata is preserved in
the Kathu drillcore. Altermann and Siegfried (1997)
give a detailed description and facies interpretation
of the sediments in the drillcore (Fig. 3). The entire
Archaean sediment pile, in the core, with a total
thickness of almost 3000 m, exceeds by far the 1900
m thickness deduced from outcrops (Beukes, 1980a).
This thickness increase is attributed to lateral facies
variation and to differing sedimentary conditions, but
also to a minor extent, to faulting and folding and
228 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
Fig. 1. Simplified geological map of the Griqualand West sub-basin and its relative geographic position with respect to the Transvaal
sub-basin. Sample location for the four analyzed samples and for the sample dated by Barton et al. (1994) are shown. The sample dated
by Sumner and Bowring (1996) was taken south of Kuruman. Note Prieska region southwest of Griquatown fault zone; Ghaap Plateau
region northeast of it.
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 229
Fig. 2. Composite stratigraphic section through the Schmidtsdrif and Campbellrand (Nauga Formation) Subgroups between Prieska and
Westerberg (Fig. 1). Lithology and facies interpretation for each member and formation are briefly summarized, and the position of the dated
samples and their ages are given. Other ages are from the literature or calculated using compacted sedimentation rates. The sedimentation
rates given in Bubnoff units are for decompacted sediments. Note that the section is disrupted in the middle to save space in the figure.
230 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 231
to the difficulty of thickness measurements in poorly
outcropping formations.
In a facies distribution model developed by
Beukes (1980a), for the Campbellrand Subgroup
carbonates, two different facies realms in the south-
western and northeastern part of the Griqualand
West basin are separated by a synsedimentary hinge,
the Griquatown growth fault. North of this fault,
the Reivilo to the Kogelbeen Formations form the
‘Ghaap Plateau Facies’ sequence of stromatolitic
carbonate platform sediments (Beukes, 1980a). The
Monteville and Gamohaan Formations, respectively,
at the base and at the top of the Campbellrand Sub-
group, north of the fault zone (Figs. 1, 3 and 7), were
interpreted as basinal, shelf, or endoclastic basinal
facies framing the platform. South of the Griquatown
fault zone, these formations pass into the basinal
Nauga Formation (compare Figs. 2, 3 and 7), which
includes the entire carbonate section of the Camp-
bellrand Subgroup accumulated south of the fault.
A thick sequence of shales (Naute Shale Member)
with some chert beds of great lateral continuity cov-
ers the Nauga Formation carbonates. The difference
in thickness between the basinal carbonates south
of the Griquatown fault (600 m) and the platform
north of the fault (1600 m on the Ghaap Plateau) is
striking. Together with Beukes’ (1980a) depositional
model, this difference tempted Grotzinger (1989) to
hypothesize a possible relief of 950 m between the
base and the top of the Campbellrand platform, at
the time of its terminal drowning.
Altermann and Herbig (1991) proposed an alter-
native model in which the intracratonic Griqualand
West basin experienced its highest subsidence rates
in its central parts, north of the Griquatown fault. The
subsidence was matched by stromatolitic growth and
carbonate accumulation (building the Ghaap Plateau)
and thus, shallow marine conditions prevailed. South
of the Griquatown fault, peritidal flats often exposed
to erosion prevented the accumulation of a thick
pile of carbonate strata. The decline in carbonate
sedimentation was accompanied by siliciclastic in-
Fig. 3. Brief lithological description and stratigraphic subdivision (Altermann and Siegfried, 1997) of the borehole drilled at Kathu,
Sishen (Fig. 1). The ages of the formations were dated on samples from outcrops remote from Kathu, and are thus tentatively correlated
here on lithostratigraphic grounds. The sedimentation rates given in Bubnoff units are for decompacted sediments. Note that the section
is disrupted in the upper part (thick dyke intrusion) to save space.
flux, evident from the increase in shale content. This
increase culminated in the deposition of the Naute
Shales, followed by precipitation of BIF of the As-
bestos Hills Subgroup, which date between around
2500 and 2432 Ma (Trendall et al., 1990, 1995;
Barton et al., 1994).
The detailed sedimentology, geochemistry and
petrography of tuffs from the Campbellrand Sub-
group are described by Altermann (1996a). Separa-
tion of fine and coarse grains in tholeiitictuffs of the
Nauga Formation carbonates suggests deposition in
shallow water, perhaps a few metres to 40 m depth.
A tuff layer close to the top of the Nauga Formation
carbonates was dated by the SHRIMP U–Pb method
on zircons, at 2552 š11 Ma (Barton et al., 1994).
Proximal tuffs were found within the Nauga Forma-
tion, near Prieska. The tuffs thin out and become
finer-grained towards the north and away from the
peritidal flats described by Altermann and Herbig
(1991). Altermann (1996a) suggested that the vol-
canic centres were located along the southwestern
margin of the Transvaal sea, to the south and south-
west of the present margin of the Kaapvaal craton.
Volcanoes might have formed islands and the craton
and the epeiric basin probably extended farther to
the southwest, into areas now occupied by younger
Proterozoic mobile belts (Altermann and Ha¨lbich,
1991). Zircons collected from these tuffs are the
source for the new age data presented herein.
New investigations of the Nauga Formation show
rapid lateral and vertical facies changes within the
lower part of this formation. Vertically, the Nauga
Formation can be subdivided into five informal mem-
bers, as illustrated in Fig. 2 (Kiefer et al., 1995;
Altermann, unpubl. data).
(1) A mixed siliciclastic and carbonate clastic
member at the base of the formation.
(2) A peritidal member, consisting of widespread
tufted Stratifera-like mats with abundant palisade
structures intercalated with loferite beds and tidal
channel carbonate sand bodies.
(3) A chert member follows, in which the tidal flat
232 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
features give way to deep lagoonal platy dolmicrites,
dolarenites and microbial laminites. Three laterally
persistent chert marker horizons are intercalated.
(4) An overlying proto-BIF member consists
mainly of carbonates with some coiled thin micro-
bial mats (Kiefer et al., 1995) resembling those in the
chert member. It comprises three laterally persistent
BIF-like horizons (proto-BIF of Button, 1976).
(5) The approximately 150 m thick Naute Shale
member covers the carbonates. These finely lami-
nated shales, intercalated with rare thin tuffites and
prominent chert beds, represent deposition on the
shelf, probably below the storm wave-base.
2.1. Correlation to Malmani Subgroup in the
Transvaal
No continuous outcrops exist between the sedi-
ments of the preserved Transvaal basin and of the
Griqualand West basin, although the two sub-basins
share the same basement of Ventersdorp Supergroup
volcanics (Fig. 1). The Black Reef Quartzite Forma-
tion (Fig. 7) is generally accepted as the Transvaal
basin equivalent of the 2642 Ma old Vryburg Forma-
tion (lower Schmidtsdrif Subgroup) in Griqualand
West, for both formations unconformably cover the
Ventersdorp Supergroup (compare Figs. 2, 3 and 7;
SACS, 1980). The upper Schmidtsdrif Subgroup is
commonly correlated with the Oaktree Formation
at the base of the Malmani Subgroup in Transvaal
(Altermann and Wotherspoon, 1995). Tuffs in the
upper Oaktree Formation were dated at 2550 š3Ma
(U–Pb on zircons) by Walraven and Martini (1995).
Like the Ghaap Plateau facies, the Malmani Sub-
group carbonates also consist of several formations.
These formations were grouped into genetic units
and attributed by Clendenin (1989) to transgression–
regression cycles. The first two transgressive cy-
cles are documented in the lower Monte Christo
Formations of the lower Malmani sediments. The
upper three formations of the Malmani carbonates
(Lyttelton, Eccles and Frisco Formations) reflect to-
gether the third major transgression, followed by the
deposition of the Penge Iron Formation (fourth trans-
gressive cycle) which correlates with the Kuruman
BIF in Griqualand West.
Beukes (1986) correlated the Campbellrand Sub-
group with the Malmani Subgroup in the Transvaal,
by defining stratigraphic units on the basis of stroma-
tolite morphology and on carbonate facies and pet-
rography. Such a lithostratigraphic approach is only
applicable if the cyclicity and hydrodynamic condi-
tions were uniform across the entire basin. In this cor-
relation, the Gamohaan Formation, at the top of the
Campbellrand Subgroup, passes northeastward into
the Frisco Formation, at the top of the Malmani Sub-
group carbonates in Transvaal. The Monteville and
Reivilo Formations of Griqualand West interfinger
with the Oaktree and Monte Christo Formations at the
base of the Malmani Subgroup (Beukes, 1986, fig. 7).
3. Sample localities and description
Four samples were processed for zircon dating.
The sampling sites are shown in Fig. 1.
(1) Sample WA92=4 was collected from the up-
permost tuff bed of the Gamohaan Formation, at
the Kuruman Kop peak, north of the town Kuru-
man (Fig. 1). This stratigraphic level was correlated
by Beukes (1980a) with the stratigraphic position of
the sample dated by Barton et al. (1994) and of the
sample WA93=12 described below. The stratigraphic
section through the Kuruman Kop was recorded and
depicted in detail by Ha¨lbich et al. (1992, fig. 10).
The sample is from the upper Gamohaan Formation,
from lithofacies ‘e’ (microbial laminites, grainstones
and shales) of Ha¨lbich et al. (1992), and lies ap-
proximately 40 m below the nearly 30 m thick Tsi-
neng member (Beukes, 1980b), which represents a
transition from carbonate to BIF sedimentation. The
stratigraphic thickness to the massive Kuruman IF
proper is around 75 m. It is probably the same tuff
bed as that dated by Sumner and Bowring (1996) at
2521š3 Ma. The horizon is 45 cm thick and consists
of three graded, fine lapilli to ash tuff intervals with
thin dolarenitic interlayers, and with some tuffaceous
admixture. The pure tuff beds are interpreted as fall-
out tuffs, as they are normally graded, lack Bouma in-
tervals and there is a general absence of layers resem-
bling turbidites within this facies (Altermann, 1996a).
Over 50 zircons were recovered from about 7
kg of rock. The zircons are morphologically ho-
mogeneous, short- to long-prismatic (100–150 µm),
idiomorphic, pink and clear. Rare inclusions are
present in some of the zircons.
(2) Sample WA93=41 was collected from an out-
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 233
crop along the road from Douglas to Niekerkshoop,
onthe farm Suiversfonteinand isfrom the upper Mon-
teville Formation, Ghaap Plateau,adjacent to the Gri-
quatown fault. The carbonate facies at the sampling
area is of cross-bedded and finely laminated dolaren-
ites. The volcaniclastic band in these dolarenites is 20
cm thick. A few thin shale beds are intercalated in the
lower part of the outcrop, together with three promi-
nent Fe-rich chert beds, that are brecciated in places
alonga minor fault(approximately 10mbelowthe tuff
band). The breccia exhibits weak Pb (galena) miner-
alization. Above the tuff band, cross-bedded dolaren-
ites pass upward into stromatolitic mats. The micro-
bial lamination builds lateral linkage between small
conical to sub-conical columns. Abundant, cm-large
fenestral cavities filled by calcite and rarelyby quartz,
are irregularly distributed in the columns and between
the laminae. The bioherms resemble thyssagetacean
stromatolites, as described by Hofmann and Masson
(1994).Theoverall faciesisinterpreted as shallowing-
upward, entirelysubtidal,but withupward decreasing
hydrodynamic energy. The volcaniclastic bed itself
shows no internal sedimentary structures apart from
a faint lamination. Because different zircon popula-
tions were found in this sample, it may represent a
reworked sediment, such as a tuffite. This interpreta-
tion is consistent with the nature of the cross-bedded
dolarenites directly above and below the tuffite.
About 25 zircons were recovered from 8.5 kg of
sample material. The sample was rich in pyrite. Two
of the zircons were well rounded and abraded and
of orange-brown colour. These were not analyzed.
Other zircons were broken, long- or short-prismatic,
xenomorphic, between 50 and 100 µm long, and
some of them were abraded (subangular to sub-
rounded). They exhibit common inclusions and all
were pink and turbid.
(3) Sample WA93=12 was collected on the farm
Kliphuis, at Prieska, from the uppermost tuff bed
in the carbonates, below the Naute Shales. It comes
from the top of the chert member, 10 m above the
three prominent chert marker horizons of the Nauga
Formation. In the measured section, it is located 48
m below the Naute Shale member and almost at the
same stratigraphic level as the sample dated by Bar-
ton et al. (1994). However, the sample dated by Bar-
ton et al. (1994) from the farm Nauga, approximately
30 km northwest of the farm Kliphuis and approx-
imately 60 m below the Naute Shale member, was
taken 6 m below the three prominent chert marker
horizons of the chert member (Figs. 1 and 2). These
three chert horizons are very uniformly distributed
between Nauga and Kliphuis. The tuff band dated by
Barton et al. (1994) pinches out and is not present
in the Kliphuis area. On the farm Klein Naute, mid-
way between Nauga and Kliphuis (Fig. 1), 28 m of
sedimentary section separate these two tuff horizons.
About 30 zircons were recovered from 7.35 kg
of rock. The zircons are morphologically similar,
equant to long-prismatic, idiomorphic, 100–200 µm
long, pink and dim. Rare inclusions are present in
some of the zircons.
(4) Sample WA93=15 was collected on the farm
Engelwildgeboomfontain, at Prieska, close to the
Kliphuis farm boundary. It is from the same sec-
tion as WA93=12 and stratigraphically about 230
m below it, within the peritidal member of the
Nauga Formation. The section (shown in Fig. 2)
does not outcrop continuously and has been as-
sembled from several shorter sections, measured by
‘Jacob’s staff’, and only a few tens to hundreds of
metres apart. This was necessitated by folds and
faults displacing the measured sections of strata rel-
ative to each other. Approximately 10 m of strata,
judged from detailed mapping, are missing between
the measured sections from the Engelwildgeboom-
fontain and Kliphuis farms, and are probably of shale
that makes no outcrops. The tuff bed sampled here is
only 5 cm thick and roughly correlative of Beukes’
(1980a) ‘tuff 4’ from the ‘Central Dolomite Zone’
(Beukes, 1980a, fig. 22). This zone is characterized
by microbial laminites with Stacked Hemispheroids-
Inverted (SH-I) structures and interpreted as peritidal
to supratidal Stratifera-like biostromes (compare Al-
termann and Herbig, 1991).
Over 100 zircons were recovered from 5.0 kg
of rock. The zircons are equant to long-prismatic,
idiomorphic, 100–150 µm long, pink and clear, with
some inclusions.
4. Analytical procedures
Samples were crushed in a jaw crusher and bro-
ken to <2 mm particle size in a cylindrical rolling
mill, and then passed through a 180 µmsieve.The
sieved fraction was processed using a Wilfley ta-
234 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
ble. The heavy mineral fraction was further purified
using a Frantz magnetic separator and methylene io-
dine. Zircons were hand picked from the resulting
mineral fraction, mounted in epoxy and sectioned
approximately in half, and the mount surface was
then polished to expose the grain interiors.
U–Th–Pb measurements were made using the
ion microprobe SHRIMP-II at Curtin University of
Technology, employing operating and data-process-
ing procedures similar to those described by Comp-
ston et al. (1984) and Williams et al. (1984). Pb=U
ratios were determined relative to that of the stan-
dard Sri Lanka zircon CZ3, which has been assigned
a206Pb=238U value of 0.0914 corresponding to an
age of 564 Ma. Reproducibility of the Pb=U ratio of
the standard was better than š1:6%; this uncertainty
is included in the quoted analytical errors. Errors
given on individual analyses are based on counting
statistics and are at the 1¦level; those given on
pooled analyses are at 2¦; or 95% confidence. Ages
cited are based on weighted mean 207Pb=206Pb ratios.
Features such as zircon morphology (size, shape,
zonation, etc.) and chemistry (U and Th contents,
Th=U ratios), degree of discordance of each analy-
sis and evidence of radiogenic Pb loss were taken into
account in the assessment of the validity of pooled
analyses. Dates were determined using the mean
207Pb=206Pb ratios determined from pooled analyses.
Individual analyses were weighted according to the
inverse square of the individual analytical error (based
on counting statistics) of the analysis, for the de-
termination of the weighted mean 207Pb=206Pb ratio
of pooled analyses. Analyses more than š2¦from
the weighted mean value were treated as outliers and
deleted from the pool, and the weighted mean value
then recalculated.This process was repeated until all
pooled analyses were within š2¦of the weighted
mean value and the remaining pooled data were nor-
mally distributed about the mean. Where there was
no obvious justification (based on zircon morpho-
logical or chemical differences) for deletion of out-
liers and their deletion did not significantly affect
the age and error obtained, the outliers were retained
within the pooled population used to determine the
weighted mean date and error. A chi-square test was
applied to grouped analyses in order to assess the rel-
ative effects of analytical sources of error, such as
counting statistics, and geological sources of error,
such as that arising from the inclusion of analyses of
slightly older xenocryst zircons or zircons that may
have lost small amounts of radiogenic Pb. Chi-square
values for grouped analyses of less than or equal to
unity indicate that scatter about the weighted mean
value determined for the grouped analyses can be ac-
counted for by analytical sources of error alone. A
chi-square valuesignificantly greater than unity indi-
cates that analyses are not normally distributed about
the weighted mean value and that other (geologi-
cal) sources of error are present within the grouped
population. In these cases, the 95% confidence error
is based on the observed scatter about the weighted
mean 207Pb=206Pb ratio of pooled analyses.
5. Analytical results
5.1. WA92=4
Analytical data are summarized in Table 1 and
shown on a conventional concordia plot in Fig. 4.
All analyses plot within the error of the concordia,
or are only slightly discordant. Sixteen analyses of
16 zircons gave a 207Pb=206Pb age of 2516 š4Ma
(95% confidence). This is regarded as the crystalliza-
tion age of the zircons and the age of the tuff layer.
One analysis (4.1) had a slightly lower 207Pb=206Pb
ratio corresponding to an age of 2476 š9(1¦)
Ma. This analysis is probably of a zone which has
experienced some post-crystallization loss of radio-
genic Pb. Cathodoluminescence imaging of the zir-
con growth zones reveals no abnormalities at the
analyzed site. If this analysis is included in the sta-
tistical calculations, the weighted mean 207Pb=206Pb
age is 2513 š4 Ma, and thus insignificantly differ-
ent from the calculated age of 2516 š4Ma.One
analysis (3.1), indicated an early Palaeozoic age and
is believed to be a contaminant introduced during
sample preparation, and is not discussed further.
5.2. WA93=41
A total of sixteen analyses were obtained on four-
teen zircons. The analyses fall into four statistically
distinguishable age groups (Fig. 4).
Group 1, consisting of seven spots on five zir-
cons (0.1, 3.1, 3.2, 4.1, 6.1, 6.2, 8.1), has a pooled
weighted mean 207Pb=206Pb age of 2637 š30 Ma.
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 235
Fig. 4. Conventional concordia plots for the four analyzed samples, showing isotopic composition of the zircons. The apparent over-concordance of the zircons in sample
WA93 =41, possible problems and analyzing technique are discussed in the text. The shaded box in WA92=4 is zircon 4.1, not included in the age calculation of 2516 š4
Ma (see text). Note that all error boxes shown in the plots are 1¦, while ages with 2¦errors are given and discussed in the text.
236 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
Table 1
Ion microprobe U–Th–Pb analyses of zircons from the four samples discussed
Grain U Th Th=URad.Pb 206 Pb=204Pb Calculated atomic ratios, 204Pb corrected Age (Ma)
nr. (ppm) (ppm) (ppm)
204Pb corr. 208Pb=232Th 206Pb=238U207Pb=235U207Pb=206Pb 208Pb=232Th 206Pb=238 U207Pb=235U207Pb=206Pb
(š1¦)(š1¦)(š1¦)(š1¦)(š1¦)(š1¦)(š1¦).š1¦/
WA92=4. Gamohaan Fm.
1.1 311.2 144.3 0.46 157.54 10785 0.1270 (0.0034) 0.4534 (0.0010) 10.4669 (0.2348) 0.1674 (0.0006) 2417 (60) 2410 (42) 2477 (21) 2532 (10)
2.1 211.3 97.9 0.46 106.05 89445 0.1248 (0.0034) 0.4501 (0.0096) 10.3594 (0.2384) 0.1669 (0.0010) 2378 (62) 2396 (43) 2467 (22) 2527 (11)
4.1 219.2 111.9 0.51 109.42 2817 0.1167 (0.0022) 0.4479 (0.0059) 10.0024 (0.1474) 0.1620 (0.0008) 2231 (40) 2386 (26) 2435 (14) 2476 (9)
5.1 66.5 30.1 0.45 35.23 2874 0.1324 (0.0044) 0.4773 (0.0069) 10.8595 (0.2050) 0.1650 (0.0017) 2513 (79) 2515 (30) 2511 (18) 2508 (18)
6.1 280.9 137.8 0.49 148.32 7937 0.1272 (0.0021) 0.4728 (0.0062) 10.8063 (0.1533) 0.1658 (0.0006) 2420 (38) 2496 (27) 2507 (13) 2516 (6)
7.1 292.5 140.9 0.48 152.27 8000 0.1285 (0.0021) 0.4655 (0.0061) 10.6937 (0.1507) 0.1666 (0.0006) 2443 (37) 2464 (27) 2497 (13) 2524 (6)
9.1 227.8 130.8 0.57 117.55 3401 0.1223 (0.0021) 0.4543 (0.0060) 10.3375 (0.1517) 0.1650 (0.0008) 2333 (39) 2414 (27) 2465 (14) 2508 (8)
10.1 256.2 125.1 0.49 136.61 6803 0.1306 (0.0022) 0.4769 (0.0063) 10.8822 (0.1554) 0.1655 (0.0007) 2480 (39) 2514 (28) 2513 (13) 2513 (7)
11.1 286.1 148.9 0.52 149.80 4505 0.1256 (0.0021) 0.4656 (0.0061) 10.6760 (0.1530) 0.1663 (0.0007) 2392 (38) 2464 (27) 2495 (13) 2521 (7)
12.1 294.0 144.2 0.49 153.90 7535 0.1289 (0.0021) 0.4671 (0.0062) 10.7310 (0.1520) 0.1666 (0.0006) 2450 (38) 2471 (27) 2500 (13) 2524 (6)
13.1 255.6 126.8 0.50 132.51 6849 0.1274 (0.0021) 0.4625 (0.0061) 10.5862 (0.1502) 0.1660 (0.0006) 2424 (38) 2450 (27) 2488 (13) 2518 (6)
14.1 253.0 127.5 0.50 132.38 9615 0.1265 (0.0021) 0.4669 (0.0062) 10.6644 (0.1516) 0.1657 (0.0006) 2407 (37) 2470 (27) 2494 (13) 2514 (6)
15.1 335.3 163.5 0.49 178.11 11236 0.1282 (0.0020) 0.4758 (0.0062) 10.8798 (0.1508) 0.1659 (0.0005) 2439 (36) 2509 (27) 2513 (13) 2516 (5)
16.1 318.1 177.1 0.56 164.71 6369 0.1244 (0.0020) 0.4567 (0.0060) 10.4201 (0.1465) 0.1655 (0.0006) 2370 (35) 2425 (27) 2473 (13) 2512 (6)
17.1 273.2 144.4 0.53 141.10 4367 0.1197 (0.0020) 0.4610 (0.0061) 10.4644 (0.1502) 0.1646 (0.0007) 2285 (37) 2444 (27) 2477 (13) 2504 (7)
20.1 290.1 144.7 0.50 151.36 7692 0.1285 (0.0021) 0.4654 (0.0061) 10.5837 (0.1492) 0.1649 (0.0006) 2443 (37) 2464 (27) 2487 (13) 2507 (6)
21.1 224.5 253.9 1.13 112.64 721 0.0555 (0.0014) 0.4464 (0.0059) 10.1628 (0.1651) 0.1651 (0.0013) 1092 (26) 2379 (26) 2450 (15) 2509 (13)
WA93=41. Monteville Fm.
0.1 97.0 83.53 0.86 59.19 518 0.1278 (0.0052) 0.5094 (0.0091) 12.3012 (0.3431) 0.1751 (0.0034) 2431 (93) 2654 (39) 2628 (27) 2607 (32)
1.1 43.9 23.66 0.54 26.30 501 0.1214 (0.1117) 0.5312 (0.0107) 13.8209 (0.4887) 0.1887 (0.0050) 2316 (201) 2747 (45) 2738 (34) 2731 (44)
2.1 117.9 73.5 0.62 75.52 2454 0.1428 (0.0042) 0.5508 (0.0093) 14.3453 (0.2915) 0.1889 (0.0018) 2697 (74) 2829 (39) 2773 (19) 2732 (16)
3.1 127.0 117.45 0.92 81.76 1478 0.1319 (0.0033) 0.5297 (0.0089) 13.2374 (0.2697) 0.1812 (0.0017) 2505 (58) 2740 (38) 2697 (19) 2664 (16)
3.2 71.7 44.44 0.62 43.05 1294 0.1258 (0.0050) 0.5275 (0.0093) 12.7370 (0.3024) 0.1751 (0.0024) 2395 (91) 2713 (39) 2660 (23) 2607 (23)
4.1 49.2 23.11 0.47 29.77 725 0.1223 (0.0105) 0.5479 (0.0108) 13.5093 (0.4285) 0.1788 (0.0040) 2332 (190) 2816 (45) 2716 (30) 2642 (38)
5.1 73.6 40.28 0.55 41.61 566 0.1158 (0.0077) 0.5110 (0.0092) 11.3450 (0.3414) 0.1610 (0.0035) 2214 (139) 2661 (39) 2552 (28) 2466 (38)
6.1 60.2 32.15 0.53 37.92 810 0.1460 (0.0086) 0.5537 (0.0102) 13.7946 (0.3941) 0.1807 (0.0035) 2755 (152) 2840 (43) 2736 (27) 2659 (33)
6.2 51.6 27.63 0.54 31.58 877 0.1390 (0.0097) 0.5400 (0.0105) 13.3333 (0.4287) 0.1791 (0.0042) 2630 (171) 2784 (44) 2704 (31) 2644 (39)
7.1 120.1 63.75 0.53 71.26 1696 0.1278 (0.0046) 0.5249 (0.0090) 13.3752 (0.2824) 0.1848 (0.0019) 2430 (82) 2720 (38) 2707 (20) 2696 (17)
8.1 93.6 86.18 0.92 60.35 274 0.1315 (0.0059) 0.5322 (0.0094) 12.6009 (0.3903) 0.1717 (0.0040) 2496 (105) 2751 (40) 2650 (30) 2574 (40)
9.1 92.2 52.57 0.57 55.44 1365 0.1367 (0.0052) 0.5230 (0.0091) 13.4908 (0.3030) 0.1871 (0.0023) 2590 (92) 2712 (39) 2715 (21) 2717 (20)
10.1 155.5 101.08 0.65 87.95 1762 0.1275 (0.0033) 0.4918 (0.0081) 11.3826 (0.2238) 0.1679 (0.0015) 2426 (59) 2578 (35) 2555 (19) 2536 (15)
11.1 104.6 68.92 0.66 64.84 3360 0.1391 (0.0038) 0.5364 (0.0091) 12.7075 (0.2594) 0.1718 (0.0016) 2632 (67) 2769 (38) 2658 (19) 2575 (16)
12.1 47.6 20.9 0.44 30.14 1288 0.1564 (0.0094) 0.5611 (0.0107) 14.6018 (0.3977) 0.1887 (0.0032) 2936 (165) 2871 (44) 2790 (26) 2731 (29)
13.1 70.2 60.41 0.86 39.67 225 0.1205 (0.0074) 0.4778 (0.0087) 10.3559 (0.4292) 0.1572 (0.0055) 2299 (134) 2518 (38) 2467 (39) 2426 (61)
WA93=12. Chert mb., Nauga Fm.
1.1 43.2 57.2 1.32 23.15 438 0.1120 (0.0037) 0.4057 (0.0062) 9.4237 (0.3004) 0.1685 (0.0044) 2146 (68) 2195 (29) 2380 (30) 2542 (43)
2.1 199.8 189.4 0.95 88.55 582 0.0859 (0.0019) 0.3701 (0.0049) 8.7287 (0.1538) 0.1711 (0.0017) 1665 (36) 2030 (23) 2310 (16) 2568 (17)
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 237
Table 1 (continued)
Grain U Th Th=URad.Pb 206Pb=204Pb Calculated atomic ratios, 204Pb corrected Age (Ma)
nr. (ppm) (ppm) (ppm)
204Pb corr. 208Pb=232Th 206Pb=238U207Pb=235U207Pb=206Pb 208Pb=232Th 206Pb=238U207Pb=235U207Pb=206Pb
(š1¦)(š1¦)(š1¦)(š1¦)(š1¦)(š1¦)(š1¦).š1¦/
3.1 23.1 87.8 3.80 20.58 450 0.1276 (0.0032) 0.4721 (0.0081) 10.2919 (0.4021) 0.1581 (0.0052) 2428 (57) 2493 (36) 2461 (37) 2435 (56)
4.1 62.1 39.3 0.63 34.44 1799 0.1307 (0.0037) 0.4804 (0.0068) 11.1832 (0.2119) 0.1688 (0.0018) 2482 (66) 2529 (30) 2539 (18) 2546 (18)
5.1 236.0 202.9 0.86 132.16 2188 0.1236 (0.0021) 0.4650 (0.0062) 10.9194 (0.1613) 0.1703 (0.0009) 2355 (36) 2462 (27) 2516 (14) 2561 (9)
6.1 58.7 86.8 1.48 34.97 812 0.1304 (0.0028) 0.4281 (0.0061) 9.7489 (0.2270) 0.1652 (0.0028) 2478 (51) 2297 (28) 2411 (21) 2509 (28)
9.1 99.7 146.3 1.47 63.01 2222 0.1275 (0.0022) 0.4662 (0.0064) 11.0080 (0.1867) 0.1712 (0.0015) 2426 (39) 2467 (28) 2524 (16) 2570 (14)
9.2 31.9 34.9 1.09 19.24 972 0.1281 (0.0039) 0.4794 (0.0072) 11.0291 (0.2844) 0.1669 (0.0032) 2437 (70) 2525 (31) 2526 (24) 2526 (33)
10.1 142.1 98.8 0.70 78.61 3378 0.1292 (0.0024) 0.4723 (0.0064) 11.0669 (0.1714) 0.1699 (0.0010) 2455 (43) 2494 (28) 2529 (15) 2557 (10)
11.1 25.9 31.3 1.21 16.57 991 0.1331 (0.0046) 0.4980 (0.0084) 11.4876 (0.3425) 0.1673 (0.0038) 2526 (82) 2605 (36) 2564 (28) 2531 (38)
12.1 16.2 18.1 1.12 10.48 948 0.1410 (0.0065) 0.5074 (0.0096) 11.6933 (0.4490) 0.1671 (0.0052) 2666 (115) 2646 (41) 2580 (36) 2529 (52)
12.2 24.3 49.2 2.02 16.88 820 0.1181 (0.0032) 0.4867 (0.0079) 10.9398 (0.3307) 0.1630 (0.0038) 2255 (58) 2556 (34) 2518 (29) 2487 (40)
12.3 22.1 176.0 7.97 31.24 618 0.1349 (0.0029) 0.4804 (0.0081) 11.0851 (0.4073) 0.1674 (0.0051) 2557 (52) 2529 (35) 2530 (35) 2531 (52)
13.1 44.7 88.7 1.98 32.05 1272 0.1328 (0.0029) 0.4861 (0.0074) 11.2800 (0.2614) 0.1683 (0.0026) 2520 (51) 2554 (32) 2547 (22) 2541 (27)
15.1 232.6 625.6 2.69 169.11 709 0.1181 (0.0017) 0.4494 (0.0060) 10.4310 (0.1719) 0.1685 (0.0014) 2256 (31) 2393 (27) 2475 (15) 2543 (14)
16.1 130.4 70.0 0.54 70.10 2525 0.1281 (0.0027) 0.4757 (0.0064) 11.0701 (0.1741) 0.1688 (0.0011) 2437 (49) 2509 (28) 2529 (15) 2546 (11)
16.2 148.0 97.5 0.66 83.38 4115 0.1325 (0.0024) 0.4852 (0.0065) 11.2939 (0.1706) 0.1688 (0.0009) 2515 (42) 2550 (28) 2548 (14) 2546 (9)
16.3 159.5 133.1 0.83 90.36 3205 0.1272 (0.0021) 0.4723 (0.0063) 10.9614 (0.1644) 0.1683 (0.0009) 2421 (38) 2439 (28) 2520 (14) 2541 (9)
16.4 39.9 19.1 0.48 21.26 915 0.1247 (0.0067) 0.4796 (0.0071) 11.0538 (0.2690) 0.1672 (0.0029) 2375 (121) 2525 (31) 2528 (23) 2529 (30)
17.1 40.6 42.0 1.03 15.99 471 0.0946 (0.0038) 0.3072 (0.0047) 7.2513 (0.2474) 0.1712 (0.0049) 1828 (70) 1727 (23) 2143 (30) 2569 (48)
18.1 138.5 243.6 1.76 74.99 816 0.1008 (0.0017) 0.3860 (0.0052) 8.9665 (0.1570) 0.1685 (0.0016) 1941 (31) 2104 (24) 2335 (16) 2543 (16)
WA93=15. Peritidal mb., Nauga Fm.
2.1 103.8 64.1 0.62 58.32 9024 0.1336 (0.0032) 0.4851 (0.0080) 11.7252 (0.2236) 0.1753 (0.0013) 2535 (58) 2549 (35) 2583 (18) 2609 (13)
4.1 67.1 39.0 0.58 38.38 3558 0.1352 (0.0042) 0.5000 (0.0086) 11.8581 (0.2527) 0.1720 (0.0018) 2567 (75) 2614 (37) 2593 (20) 2577 (18)
7.1 113.1 86.0 0.76 65.88 3814 0.1331 (0.0030) 0.4898 (0.0080) 11.7459 (0.2211) 0.1739 (0.0013) 2526 (53) 2570(35) 2584 (18) 2596 (12)
9.1 78.0 46.0 0.59 43.35 4450 0.1305 (0.0039) 0.4843 (0.0082) 11.6037 (0.2408) 0.1738 (0.0018) 2478 (70) 2546 (35) 2573 (20) 2594 (17)
14.1 83.4 48.2 0.58 47.93 2875 0.1344 (0.0040) 0.5035 (0.0085) 11.9568 (0.2462) 0.1722 (0.0017) 2548 (71) 2629 (37) 2601 (19) 2580 (17)
15.1 133.1 96.3 0.72 76.98 2795 0.1316 (0.0031) 0.4920 (0.0081) 11.6417 (0.2230) 0.1716 (0.0014) 2498 (56) 2579 (35) 2576 (18) 2573 (14)
16.1 89.3 68.1 0.76 52.23 3995 0.1305 (0.0032) 0.4949 (0.0083) 11.7120 (0.2315) 0.1716 (0.0014) 2479 (57) 2592 (36) 2582 (19) 2574 (14)
17.1 135.2 95.1 0.70 77.69 7427 0.1342 (0.0028) 0.4879 (0.0079) 11.7433 (0.2122) 0.1763 (0.0009) 2544 (50) 2561 (34) 2584 (17) 2602 (10)
20.1 116.1 82.0 0.71 68.02 5444 0.1383 (0.0032) 0.4974 (0.0081) 11.8112 (0.2233) 0.1722 (0.0013) 2618 (56) 2603 (35) 2590 (18) 2579 (13)
21.1 134.1 100.6 0.75 78.21 4437 0.1344 (0.0030) 0.4918 (0.0080) 11.6983 (0.2187) 0.1725 (0.0013) 2548 (53) 2578 (35) 2581 (18) 2582 (12)
24.1 128.5 82.4 0.64 74.30 6839 0.1356 (0.0031) 0.4994 (0.0081) 11.7918 (0.2178) 0.1713 (0.0012) 2571 (54) 2611 (35) 2588 (17) 2570 (12)
25.1 119.2 87.9 0.74 68.52 4730 0.1338 (0.0029) 0.4851 (0.0079) 11.5522 (0.2148) 0.1727 (0.0012) 2539 (52) 2550 (34) 2569 (18) 2584 (12)
26.1 58.9 36.6 0.62 33.40 7424 0.1388 (0.0042) 0.4874 (0.0085) 11.7166 (0.2544) 0.1743 (0.0019) 2627 (75) 2560 (37) 2582 (21) 2600 (19)
27.1 70.8 43.6 0.62 39.76 3411 0.1314 (0.0039) 0.4872 (0.0083) 11.6275 (0.2445) 0.1731 (0.0018) 2495 (70) 2559 (36) 2575 (20) 2588 (18)
28.1 205.0 157.8 0.77 118.04 25950 0.1331 (0.0024) 0.4824 (0.0077) 11.5732 (0.1970) 0.1740 (0.0008) 2525 (44) 2538 (33) 2571 (16) 2596 (7)
29.1 176.9 144.1 0.81 103.23 7732 0.1337 (0.0026) 0.4847 (0.0077) 11.6013 (0.2032) 0.1736 (0.0010) 2536 (46) 2548 (34) 2573 (17) 2592 (9)
30.1 66.0 33.4 0.51 38.11 3272 0.1408 (0.0046) 0.5112 (0.0088) 12.2081 (0.2588) 0.1732 (0.0018) 2662 (81) 2662 (38) 2621 (20) 2589 (17)
31.1 115.5 80.7 0.70 65.15 6534 0.1230 (0.0028) 0.4857 (0.0079) 11.5272 (0.2139) 0.1721 (0.0012) 2344 (50) 2552 (34) 2567 (17) 2578 (12)
For analytical procedure, see text description. Note that calculated atomic ratios and ages are within 1¦error margins.
238 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
Group 2, consisting of five spots on five zircons
(1.1, 2.1, 7.1, 9.1, 12.1), has a pooled weighted mean
207Pb=206Pb age of 2718 š26 Ma.
Group 3, two spots on two zircons (10.1, 11.1),
has a pooled weighted mean 207Pb=206Pb age of
2555 š19 Ma.
Group 4, consisting of analyses 5.1 and 13.1, has a
pooled weighted mean 207Pb=206Pb age of 2455 š32
Ma.
All analyses plot within error of the concordia or
are slightly reverse discordant (Fig. 4). Groups 1 and
2 are interpreted to provide the ages of older for-
mations eroded and redeposited within this reworked
volcanic layer. The age of 2555 š19 Ma (95% con-
fidence) is regarded as the best approximation of
the age of deposition of the tuffite layer. This age
has been reported as 2555 š11 Ma by Altermann
(1996b, 1997) and Altermann and Nelson (1996); re-
calculation of the pooled weighted mean 207Pb=206 Pb
age, however, results in 2555 š19 Ma. Analyses
belonging to Group 4 may reflect some post-crystal-
lization loss of radiogenic Pb in these two zircons.
Alternatively, these zircons may be contaminants.
5.3. WA93=12
Twenty-two analyses were obtained on sixteen
grains from this sample. The results are summarized
in Table 1 and shown on a concordia plot in Fig. 4.
Some analyses were discordant, indicating recent
loss of radiogenic Pb. Twenty-one analyses of fifteen
grains gave an age of 2549š7 Ma (95% confidence).
This is the best estimate of the crystallization age of
the zircons in the tuff and is equivalent to the deposi-
tional age of the tuff. One analysis (14.1) indicated a
Palaeozoic age and this zircon is interpreted to be a
contaminant.
The age of 2549š7 Ma determined for WA93=12
from the uppermost tuff of the Nauga Formation
at Prieska, is within error of the age of 2552 š11
Ma determined by Barton et al. (1994), using the
SHRIMP I in Canberra, for a tuff from a similar
stratigraphic level from Nauga Farm (see discus-
sion below). The two samples are taken about 30
km apart. They are vertically separated by approxi-
mately 30 m of chert and carbonate sediments, with
WA93=12 being the stratigraphically higher sample
and from the chert member of the Nauga Formation.
5.4. WA93=15
The analytical data obtained for this sample are
summarized in Table 1 and shown on a concordia
plot in Fig. 4. Eighteen analyses on eighteen grains
gave an age of 2588 š6 Ma (95% confidence). This
is interpreted as the deposition age of the tuff.
6. Regional interpretation of age-dated samples
Sample WA92=4, from the uppermost tuff layer
at the Kuruman Kop peak, was dated at 2516 š4
Ma. This tuff band is therefore at least 22 Ma, and
up to 44 Ma younger than the uppermost tuff layer
in the carbonates at Prieska, some 250 km south of
Kuruman (sample WA93=12; 2549š7 Ma). This age
difference indicates that the Gamohaan Formation is,
at least in its uppermost part, significantly younger
and therefore not correlative of the Nauga Formation
carbonates, but of the Naute Shale member, that
was deposited between 2549 š7 Ma and the 2500
Ma Kuruman BIF. From the discussion below, it
becomes clear that the stromatolitic formations of the
Ghaap Plateau faciesbelow the Gamohaan and above
the Monteville Formation (WA93=41, 2555š19 Ma)
must also largely fall into the time of the deposition
of Naute Shale member.
The age of 2516 š4MaonWA92=4 is within an-
alytical error of the 2521 š3 Ma date acquired for a
tuff band sampled south of Kuruman, probably from
the same stratigraphicposition within the Gamohaan
Formation (Sumner and Bowring, 1996). The strati-
graphic thickness between the WA92=4 sample and
the Kuruman BIF is around 75 m (Ha¨lbich et al.,
1992, fig. 10) and it can be speculated that, with
a bulk sedimentation rate of 2 m=Ma to 4 m=Ma
(Barton et al., 1994) for a carbonate, shale and BIF
succession, the Kuruman BIF sedimentation in this
area started about 2500 Ma to 2480 Ma ago. This
is consistent with the calculations by Barton et al.
(1994), for the onset of BIF sedimentation in the
Prieska area, and withthe zircon age data of Trendall
et al. (1995).
Sample WA93=41 yielded different morphologi-
cal and age populations of partly abraded and broken
zircons, consistent with its interpretation as a possi-
bly reworked tuffaceous sediment (i.e. tuffite). The
complex age structure is difficult to interpret. The
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 239
sample is from the upper Monteville Formation,
the lowest formation of the Campbellrand Subgroup
carbonates on the Ghaap Plateau, and according to
Altermann and Siegfried (1997) in their study of the
Kathu borehole, this unit is about 2000 m below
the top of the Gamohaan Formation. The Campbell-
rand Subgroup sediments total 2460 m thickness in
the borehole and the Monteville Formation is 540
m thick, while the upper Gamohaan Formation has
been removed by erosion (compare Fig. 3).
The oldest age group in this sample, 2718 š26
Ma, coincides with the age of the Ventersdorp Super-
group (2714 š8 and 2709 š4 Ma; Armstrong et al.,
1991). Zircons of this age group are therefore inter-
preted as sedimentary detritus from the Ventersdorp
volcanics. The Ventersdorp lavas underlie uncon-
formably the Schmidtsdrif Subgroup and were also
locally exposed to erosion during the time of deposi-
tion of the Monteville Formation. Alternatively, these
zircons may have been deposited in the Schmidts-
drif Subgroup and subsequently redeposited in the
Monteville Formation.
The age of 2637 š30 Ma is very close to the age
of 2642 š3 Ma determined for the Vryburg lavas
(Walraven et al., in press) and is, therefore, too old
to represent the Monteville Formation. Thus, most
likely, this age group also reflects the age of some
source area of siliciclastic debris. If Ventersdorp
Supergroup rocks were exposed during Monteville
times, then the Vryburg Formation may also have
been exposed in the vicinity. This age may thus
indicate the existence of a locally developed uncon-
formity between the Monteville Formation and the
Schmidtsdrif Subgroup. A possible source area for
this detritus can be inferred in the Vryburg rise,
northeast of the sampling site.
The age group of 2555 š19 Ma in the sample
WA93=41 is interpreted as providing the depositional
age for this upper Monteville Formation tuffite. It is,
however, based on two zircons only. It is younger
than the age of 2642 š3 Ma of the Schmidtsdrif
lavas below the carbonates, as dated by Walraven et
al. (in press), and older than sample WA92=4 from
the Gamohaan Formation. This interpretation is also
supported by a similar age obtained by Jahn et al.
(1990), for stromatolitic carbonates approximately at
the same stratigraphic level (2557 š49 Ma, Pb–Pb
on carbonate). The age of 2555 š19 Ma is within the
error margins of the Oaktree Formation age (2550 š3
Ma; Walraven and Martini, 1995), at the base of the
Malmani Subgroup, in an identical litho-stratigraphic
position, in the Transvaal basin (Beukes, 1986). This
age is also within the error margins of the 2549 š7
Ma age of sample WA93=12 from the top of the chert
member of the Nauga Formation (compare Figs. 2
and 7). On the basis of this result, the Campbellrand
Subgroup carbonates on the Ghaap Plateau above
the Monteville Formation accumulatedwithin a time
span of about 50 Ma, as did the Naute Shales in the
Prieska area.
Sample WA93=12 from the chert member of the
upper Nauga Formation was dated at 2549 š7Ma.
Sample WA93=15 from the middle Nauga Formation
was dated at 2588 š6 Ma. Prior to this time, almost
half of the peritidal Nauga Formation carbonates
had been accumulated (Fig. 2). As the upper Nauga
Formation is thus only slightly younger than the up-
per Monteville Formation (Fig. 3), the older Nauga
carbonates cannot be correlated with the Campbell-
rand Subgroup on the Ghaap Plateau. As the upper
Oaktree Formation in the Transvaal basin was dated
at 2550 š3 Ma (Walraven and Martini, 1995), the
carbonate formations there between the Oaktree and
the BIF units must also be younger than the Nauga
Formation carbonates, and are thus rather correlative
of the Naute Shales, assuming an age for the BIF in
all basins of 2500 Ma (Trendall et al., 1995).
7. Implications for depositional rates of Archaean
sediments
Various types of calculation of depositional rates
have been made by different authors for variable
Precambrian formations. Barton et al. (1994) defined
the sediment accumulation rate as the amount of sed-
iment vertically accumulated over a given period of
time, irrespective of possible unconformities. Gener-
ally, however, sediment accumulation rate (SAR) is
defined as:
SAR DWs
.AÐt/
where Wsis the weight of sediment deposited during
time t, over an area A.
Sedimentation rate (SR) is defined as:
SR Dh
t
240 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
where his the uncompacted thickness of the sed-
imentary section and tis the duration of its de-
position. Hence, sedimentation rate is calculated in
Bubnoff units (B Dmm ka1or m Ma1) and ac-
cumulation rate is expressed in g m2a1or t m2
Ma1(Einsele, 1992).
Workers such as Arndt et al. (1991), Barton et
al. (1994), Walraven and Martini (1995) and Barley
et al. (1997) did not correct the sedimentation rates
for compaction, but discussed the possible alteration
of the sedimentary record by intraformational ero-
sional gaps or by times of non-deposition, and the
possible effects of compaction. Archaean and Pro-
terozoic sedimentation rates may thus be calculated
for compacted sediment, defined here as compacted
sedimentation rate (cSR):
cSR Dh.x/
t
where h.x/equals the thickness of the sediment
column, irrespective of post- and syn-depositional
alteration, and tis the time period during which the
column formed.
Thickness correction for compaction and other
diagenetic influences is complex and can only be
estimated for maximum values in the present case.
Chemical crystallization in pore space and recrystal-
lization of sediment particles, but especially of car-
bonates, during diagenesis, can increase the sediment
thickness. This occurs, for instance, when aragonite
is transformed to calcite. On the other hand, pres-
sure dissolution may result in a thickness decrease.
The amount of such changes, however, is very dif-
ficult to quantify, and can be judged only from thin
sections, which cannot be examined for every part
of the sediment column. Such detailed information
does not exist yet for the rocks under discussion.
Stylolitization is, however, visible virtually in every
outcrop and thin section examined. Pressure dissolu-
tion can result in up to 20–35% thickness reduction
in carbonates and therefore must be assumed also
for the rocks under consideration. In the following
calculations we compensate, however, only for a
conservative estimate of 5% of thickness reduction
by pressure dissolution, as applicable to the Ghaap
Group carbonates in Griqualand West because of a
lack of any quantitative investigations.
Although mechanical compaction in stromatolitic
carbonates is probably negligible, as evidenced by
the excellent form preservation of stromatolites and
microfossils, carbonate muds and arenites undergo
considerable mechanical compaction, mainly in the
first 200 to 300 m of burial. Thickness reduction
in carbonate muds can exceed up to 50%, and in
carbonate sands, up to 30%, within this overburden
range (Goldhammer, 1997). Here, for the reason of
lack of quantitative data, we conservatively estimate
a thickness reduction of only 20% for carbonate
muds and sands. Our conservative estimate is sup-
ported by manifold signs of early lithification, found
by many authors (Klein et al., 1987; Altermann and
Herbig, 1991; Altermann and Wotherspoon, 1995;
Sumner and Grotzinger, 1996). In this estimate we
have also summarized carbonate muds and sands
into one category to facilitate the calculations. This
seems reasonable because the mud-to-sand ratio is
fairly high (probably >5: 1; Altermann and Herbig,
1991; Altermann and Siegfried, 1997) and because
carbonate sands tend to lithify more readily due to
their greater initial porosity.
Siliciclastic pelites may compact from >80% orig-
inal porosity to about 10%, arenites (and coarse
tuffs) compact from about 45% porosity to 20%, but
thesevaluescanincrease significantlywith increasing
amounts of pelitic matrix. Considering the high over-
burden of thousands of metres of sediment, of in part
very high density (average density of BIF approxi-
mates 3100 kg m3), we assume 70% compaction for
shales and 25% compaction for sandstones.
Silicified carbonates (cherts) can be treated as
carbonates sensu lato. Early silicification leads to
excellent preservation of the stromatolite morphol-
ogy and of microfossils, and therefore compaction
is probably negligible. Late diagenetic or post-dia-
genetic silicification usually does not alter the mor-
phology of bioherms significantly. Therefore, silici-
fied carbonates are treated here as uncompacted. The
amount of compaction in other silicified sediments
cannot be ascertained because the processes and tim-
ing of silicification were not investigated in detail.
Several periods of silicification are known, however,
for the sediments in question, most of them probably
of very late, post-diagenetic stage (Altermann and
Wotherspoon, 1995).
Compaction in BIF and primary cherts is most
difficult to substantiate. The literature does not offer
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 241
any standard figures for BIF compaction, although
compaction is clearly evident in structures such as
pods, pillows, billows and macules (Trendall and
Blockley, 1970; Trendall and Morris, 1983; Findlay,
1994). Siliceous oozes contain up to 80% water, but
lithify faster than muds, and therefore probably com-
pact less. The Red Sea siliceous, Fe-rich oozes, and,
in some respects, BIF-like silica-rich ore sludge can
have pore water contents in excess of 70–90% (We-
ber-Diefenbach, 1977). Therefore, for the Kuruman
BIF, we assume up to 90% compaction. This is in
accordance with Trendall and Blockley (1970), who
assumed compaction of up to 95% for generation
of genetic models based on deposition of seasonal
varves for the Hamersley BIF. This implies that 300
m of BIF represents a thickness of about 3000 m of
original sediment, but reflects a much lower subsi-
dence, assuming BIF deposition at roughly 100 m to
200 m water depth (Klein and Beukes, 1989).
In all calculations of basin subsidence, the amount
of compaction must be taken into consideration.
Hence, decompacted sedimentation rates do not di-
rectly reflect the rate of subsidence, but are rather a
function of compaction and subsidence. Compaction
as assumed above reflects the final stage of lithifica-
tion, disregarding gradual thickness decrease related
to a growing overburden, and concomitant dewater-
ing or dissolution. For a proper basin analysis, back-
stripping of the sedimentary pile, where the gradual
changes of sediment and water column over the layer
are restored step by step, is necessary. However, be-
cause of the lack of data on periods of exposure
and of age data within the sedimentary column, and
data on the burial and thermal history for these Ar-
chaean to Proterozoic sub-basins (Altermann, 1997),
our attempts to backstrip the sedimentary columns
were unreliable. The sedimentation and subsidence
rates given here are thus probably in the lower range
of the real figures, and should be regarded as mini-
mum calculations based on conservative estimatesof
compaction.
The preservation potential of sediment varieswith
its composition and with the depositional environ-
ment. Evidently, shallow water, peritidal environ-
ments are predisposed to frequent erosion and non-
deposition, while deep water sediments are usually
exposed only during pronounced sea-level low stands
or periods of tectonicuplift. In modern shallow water
carbonates, the rapid freshwater cementation, how-
ever, drastically increases the resistance of these sed-
iments to erosion and thus, shallow water carbonates
behave in this respect very differently to siliciclas-
tic rocks (Dravis, 1997). Cherts and BIF deposits
are resistant to erosion and their chance of expo-
sure is less due to the generally deeper depositional
environment, although their lithification is orders of
magnitudes slower than that of peritidalcarbonates.
In most depositional environments, the estimated
sedimentation rate, when calculated over a long pe-
riod of time (>100 ka), will only approximate the
actual sedimentation (SR) or sediment accumula-
tion (SAR) rate. In the Precambrian, because of the
‘poor’ time resolution of ¾5 Ma and greater, this
difference is of major importance, especially in tidal
flat or other marginal marine to fluvial deposits,
which are typically sites of discontinuous sedimen-
tation. In fossil tidal flats, for example, generally
only less than 50% of the actual sedimentation is
recorded. About 60% to 90% of the time covered
by a sediment column is characterized by erosion
and=or non-deposition (Drummond and Wilkinson,
1993; Osleger, 1994). Therefore, by implication, the
preserved sediments reflect only a fraction of the
observed time span, and corrections are necessary
for times of erosion and non deposition. In pre-
vegetational depositional systems, exposed horizons
and times of non-deposition are especially difficult
to recognize because of the lack of typical environ-
mental markers. The sedimentation rate, as defined
above, is rather a direct function of the average basin
subsidence rate and sediment supply within the given
time limits. Wider time limits covering broader fa-
cies variation result in an average rate that is remote
from the true rates of deposition for the particular
sedimentary units.
7.1. The Nauga Formation and Schmidtsdrif
Subgroup in the Prieska sub-basin
7.1.1. Sedimentation rates for the Naute Shale, the
proto-BIF and chert members of the Nauga
Formation
Barton et al. (1994) calculated sedimentation rates
of 2–4 m=Ma (cSR) for this carbonate, shale and
BIF succession. About 180 m of sedimentary rock,
consisting of 50 m of carbonate (mainly muds) and
242 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
Fig. 5. Comparison of decompacted sedimentation rates for Griqualand West and for modern sedimentary facies associations (from
Einsele, 1992). Note that the Griqualand West rates are calculated over extremely long periods of time and times of non-deposition and
erosion for the Nauga Formation, the Ghaap Plateau and the Schmidtsdrif Subgroup are neglected here (see discussion in text). However,
the deep subtidal to shelf sediments (BIF, Naute Shale) have comparable sedimentation rates to modern carbonate and siliciclastic shelf
sediments and black shales.
chert sediments of the proto-BIF member and 130
m of shales and cherts of the Naute Shale member
of the Nauga Formation, separate the 2549 š7Ma
sample WA93=12 from the base of the Kuruman
BIF (Fig. 2). The facies vary from below-wave-base
photic zone carbonates to below-storm-wave-base
shales. Somewhat farther to the southeast, the shales
reach their maximum thickness of 170 m. The cherts
in the Naute Shale vary between 15 m and 40 m in
thickness and exhibit at least two regional, and up
to seven local horizons of intraformational breccias
as well as disconformities (Altermann, 1990; Kiefer
et al., 1995), that mark erosional or non-depositional
time intervals. Correction for compaction of about
130 m of shale and 35 m of carbonate mud, on
average, and for carbonate dissolution, results in a
sediment column of over 500 m.
The deposition of the Kuruman Banded Iron For-
mation started at around 2500 Ma (Barton et al.,
1994; Trendall et al., 1995). Thus, sedimentation
rates of around 10 B for the proto-BIF member and
the Naute Shale member of the Nauga Formation can
be derived. This is comparable to modern black shale
accumulation rates (Fig. 5). The 2552 š11 Ma sam-
ple of Barton et al. (1994) is ¾30 m stratigraphically
below the site of WA93=12 (2549 š7 Ma) (Fig. 2).
These ages agree within their assigned analytical er-
rors, and indicate that the samples were deposited
within 21 Ma of each other, at maximum. During
this time, at least 30 m of sediment accumulated.
Consequently, sedimentation rates must have been at
least 1.5 B to 30 B (in the case of identical age), for
these below-wave-base, photic zone carbonates and
cherts, which also include intraformational breccias
and disconformities.
7.1.2. Sedimentation rates for the upper Nauga
Formation carbonates (peritidal member)
Sample WA93=15 was taken approximately 230
mbelowWA93=12, within the same measured strati-
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 243
graphic section. The age of 2588š6Maisonaverage
39 Ma older than the 2549 š7 Ma date determined
for WA93=12 (Fig. 2). The resulting sedimentation
rate (cSR) for the carbonatesseparating the two sam-
ples is between approximately 4 B and 9 B, with
an average of 6 B. When corrected for compaction
and dissolution and for the intercalated shales, which
together constitute less than 10% of the stratigraphic
section, the total decompacted sediment thickness
increases to around 300 m and the SR to about
8 Bubnoff, on average. This is extremely low for
carbonates. The lower facies are peritidal, passing
upward, within the uppermost 20 m, into subtidal,
below-wave-base carbonate deposits with condensed
sedimentation (Fig. 2). It can be assumed that, in
the peritidal member, up to 90% of the time repre-
sented by this sediment section is not recorded in the
beds, but in the contacts between the sedimentary
layers. Erosional surfaces and desiccation features
were described by Altermann and Herbig (1991) in
these deposits. However, even if corrected for 90%
of missing record (thus, assuming the extreme case
that the 300 m of sediment represent only 10% of
the time of 39 Ma, and multiplying the sediment
column by 10), a sedimentation rate of only 60 B to
around 115 B is achieved. This is at least ten times
lower than the growth rate of modern carbonate reefs
(Fig. 5) and about four times lower than the 400
B reported for Holocene stromatolites at Shark Bay
(Chivas et al., 1990). This discrepancy is probably
caused by the long time interval covered by the sec-
tion and by the presence of condensed sediments in
its upper part. The assumption that 90% of the time
is represented by layer boundaries is necessary for
comparison to modern growth rates of stromatolitic
carbonates, which are observed and calculated for
much shorter time intervals than dealt with in the
present case. The above example, when compared to
Phanerozoic deposits, for instance, represents the du-
ration of the entire Triassic system. The compacted
sedimentation rates of below 10 B are comparable to
the classic Jurassic carbonate sedimentation in Ger-
many, when calculated for the total thickness of the
entire system (compare with Bosscher and Schlager,
1993).
7.1.3. Sedimentation rates for the lower Nauga
Formation (basal to lower peritidal member)
There are no continuous outcrops from the mea-
sured section containing the samples WA93=15 and
WA93=12 down to the base of the Nauga Forma-
tion. The section measured through the Schmidtsdrif
Subgroup and the overlying lower Nauga Formation
(Fig. 2) was assembled from several shorter sections
northwest of Prieska and correlated with the help
of tuff horizons. It represents the average lithology
and sediment thickness, which may differ substan-
tially locally. The lavas encountered at the base of
this section (Vryburg Formation) are presumably, on
lithostratigraphic grounds, time equivalent to the lava
dated by Walraven et al. (in press), at 2642 š3Ma.
The base of the Nauga Formation carbonates is some
280 m below the 2588 š6 Ma tuff bed. Using the
above cSR of 6 B on average, calculated for peritidal
carbonates, this base must be around 2635 Ma old.
Because the facies and the lithologies are largely
similar below and above the dated tuff bed, such an
approach seems reasonable. The decompacted sedi-
mentation, including 10% of shale in the section, is
calculated to be 350 m of sediment and an SR of
about 8 B is indicated, or 80 B assuming 90% of the
time as representing non-deposition and erosion.
7.1.4. Sedimentation rates for the Schmidtsdrif
Subgroup at Prieska
The sediments of the Schmidtsdrif Subgroup
above the Vryburg lava consist of around 40 m
of carbonate, 80 m of shale and 40 m of coarser sili-
ciclastics (Fig. 2). When corrected for compaction,
this accounts for around 375 m to 400 m of sedi-
ment (depending on the locally varying proportions
of sediment type) deposited in roughly 7 Ma, on
average .2642 š3–2635), and gives a SR of 50–60
B. Again, correction for times of non-deposition and
erosion, which are common in this facies, should be
allowed, resulting in possible figures of up to 600
Bubnoff, in good agreement with modern tidal to
deltaic sediments (Fig. 5). The total thickness of the
sedimentary pile between the 2588 š6Masample
and the Vryburg lavas approximates 450 m (Fig. 2)
andcoversatimespanof65Mato51Ma.An
average cSR of 8 m=Ma can be calculated for this
section of peritidal carbonates and marginal marine
to fluvial siliciclastic rocks.
244 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
7.2. The Campbellrand and Schmidtsdrif Subgroups
in the Ghaap Plateau sub-basin
7.2.1. Sedimentation rates for the platformal
carbonates of the Campbellrand Subgroup
Less than 1600 m of predominantly stromatolitic
platform carbonates separate the Monteville Forma-
tion (2555š19 Ma) from the 2516š4 Ma uppermost
Gamohaan Formation in field outcrops on the Ghaap
Plateau. The minimum timefor deposition is thus 16
Ma and the maximum time available, 62 Ma. This
implies sedimentation rates between approximately
26 B and 100 B (cSR). However, in the Kathu bore-
hole, the Campbellrand Subgroup is 2460 m thick,
with the upper part of the Gamohaan Formation
removed during Palaeozoic erosion, and it is thus
significantly thicker than estimated from outcrops
(Fig. 2). The general facies association, nonethe-
less, does not differ significantly from that observed
in outcrops (compare Beukes, 1980a; SACS, 1980;
Altermann and Siegfried, 1997). The Reivilo, Fair-
field, Klipfontein Heuwel, Papkuil, Klippan, Kogel-
been and Gamohaan Formations of the Campbell-
rand Subgroup in the borehole consist of several
generally shallowing-upward cycles, of various stro-
matolitic carbonate facies and some shale, chert and
rare tuff intercalations (Fig. 3). These formations
(the age of 2555 š19 Ma is for the uppermost part of
the Monteville Formation; Fig. 3) total around 2000
m in thickness, only about 5% of this being shale
and an equally small portion of the carbonates being
non-stromatolitic calcareous mudstones and aren-
ites (Fig. 3). The decompacted thickness estimate is
about 2500 m, and implies sedimentation rates of
40 B to 156 B. Sedimentation rates in this range
are known from Phanerozoic tidal flats and, although
within the lower limits thereof (Fig. 5), are in good
agreement with sites of low subsidence rate (Scholle
et al., 1983; Einsele, 1992). The section is continu-
ous and no evidence for exposure or disconformities
was recognized in the drillcore; however, they should
be expected, at least, in the peritidal facies of this
section, and have been described in outcrop (Eriks-
son and Truswell, 1974; Beukes, 1986). Correction
for up to 90% of the time in the peritidal facies
being of non-deposition or erosion, results in sedi-
mentation rates from 400 B to more than 1500 B, in
agreement with modern growth rates of stromatolitic
carbonate platforms (Chivas et al., 1990) and reefs
(Fig. 5).
7.2.2. Sedimentation rates for the Schmidtsdrif
Subgroup in the Kathu borehole
The Vryburg Formation in the Kathu core is at
least 277 m thick (the base was not reached by
the drill) and consists of shales and quartzites, with
subordinate dolarenites and shaly dolomites (Fig. 3).
These are interpreted as shallow shelf to deep la-
goonal deposits. In outcrop, the Vryburg Formation
is at most 100 m thick and consists of wavy-lami-
nated, intertidal stromatolitic dolomites and calc- and
dolarenites, which interfinger with, and pass upward
into siliciclastic facies. The overlying Boomplaas
Formation is 185 m thick in the Kathu borehole.
It consists of black shales, transported oolite beds
and crypt-microbiallaminites, and is thus interpreted
as upper shelf facies, deeper than the platformal
carbonates and in situ oolites observed in surface
outcrops, where this formation is no more than 100
m thick. The overlying Lokammona Formation is
55 m thick in the borehole (Fig. 3) and comprises
black shales with minor tuff and dolomite interca-
lations. The thickness and lithology of the Lokam-
mona Formation in outcrop are very similar. In both
core and outcrop, the Lokammona is interpreted as
a transgressive phase over the Boomplaas platform
(Beukes, 1979; Altermann and Siegfried, 1997). The
overlying Campbellrand Subgroup starts with the
Monteville Formation, which in the Kathu borehole
is 540 m thick and contains domal stromatolites,
thick pyritic shale intercalations, a lava flow a few
metres thick and, in the upper part, small columnar
stromatolites, dolarenites and oolites. A shallowing-
upward platformal carbonate association was inter-
preted for this borehole section by Altermann and
Siegfried (1997). The Monteville Formation is sig-
nificantly thinner and of overall shallower platformal
character in surface outcrops (Beukes, 1980a).
Using the age of 2555 š19 Ma for the upper
Monteville Formation and the age of 2642 š3Ma
for the Vryburg lavas (Walraven et al., in press),
109 Ma to 65 Ma separated the deposition of the
lower Schmidtsdrif and the lower Campbellrand
Subgroups. Only 250 m of sedimentary rocks on
the Ghaap Plateau (SACS, 1980) separate the top of
the Vryburg Formation from the top of the Mon-
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 245
teville Formation. The same stratigraphic interval is
represented by a nearly 800 m thick sediment pile at
Kathu. For the borehole section, a sedimentation rate
(cSR) of 7 B to 12 B results and for the outcrops,
2 B to 4 B are calculated. When decompacted, the
shelf sediments in the Kathu borehole (340 m of
shales, only a few metres of quartzite, and 460 m of
carbonate with a high proportion of mudstones and
dolarenites; Altermann and Siegfried, 1997) reflect
approximately 1700 m of sediment and concomi-
tant sedimentation rates of 16 to 26 B; these are
comparable to modern black shale and carbonate
shelf deposits (Fig. 5). This section in the borehole
(Boomplaas, Lokammona and Monteville Forma-
tions) is interpreted as entirely subtidal, reflecting
mainly below-wave-base shelf faciesand, thus, times
of non-deposition were probably negligible. For the
277 m thick Vryburg Formation in the borehole,
decompacted thickness accounts for around 650 m,
including 5% quartzites, 50% shales and 45% car-
bonate muds. However, because the hole did not
reach the base of the formation, and as the time of
initiation of the sedimentation is not known, sedi-
mentation rates for the complete section cannot be
calculated.
7.3. The Malmani Subgroup and Black Reef
Formation in the Transvaal sub-basin:
sedimentation rates
Walraven and Martini (1995) dated the upper
Oaktree Formation, at the base of the Malmani Sub-
group in the Transvaal sub-basin, at 2550 š3Ma,
and calculated a cSR of 8 m=Ma for the Nauga For-
mation carbonates and of 17 m=Ma for the Ghaap
Plateau carbonates. They estimated the base of the
Malmani Subgroup in the central Transvaal sub-
basin to be 2556 Ma, and thus 86 Ma younger than
the correlated Vryburg and Black Reef Formations
(Fig. 7). The top of the Chuniespoort Group (carbon-
ates and BIFs) was estimated to be between 2472
and 2400 Ma, depending on the varying thickness
of the preserved sediments. Two possible explana-
tions were given by Walraven and Martini (1995) for
the very low sedimentation rates of the quartzites of
the Black Reef Formation: (a) the formation is not
a correlative of the Vryburg Formation, but signif-
icantly younger; (b) there is a significant period of
non-deposition between the Black Reef and the dated
Oaktree tuff. The suggestion, that a significant period
of non-deposition is partly responsible for the low
Bubnoff numbers, may also be true for the Nauga
Formation of Griqualand West, where brecciated
cherts are present within the Naute Shales, between
the carbonate and BIF sediments (Altermann, 1990).
For the five peritidal carbonate formations of the
1500 m to 1800 m thick sequence of Malmani Sub-
group carbonates (Button, 1972), a cSR of 26 B to 32
B results, when the base of the Malmani Subgroup is
assumed to be 2556 Ma and the base of the succeed-
ing BIF is taken as 2500 Ma (Trendall et al., 1995).
When decompacted, sedimentation rates comparable
to those calculated for the Ghaap Plateau, in the
order of <100 B result.
7.4. Banded iron formations in the Transvaal and
Griqualand West sub-basins: sedimentation rates
Sedimentation rates for the Kuruman and overly-
ing Griquatown BIF of the Asbestos Hills Subgroup
are difficult todetermine, because suitable age deter-
minations are not available. Additionally, folding and
thrusting complicates correlation of BIF units across
the Griqualand West basin (Altermann and Ha¨lbich,
1991). The base of the Kuruman BIF in Griqualand
West and of its Transvaal correlative, the Penge BIF,
is around 2500 Ma (Trendall et al., 1995), whereas
the base of the Griquatown BIF in Griqualand West
is 2432š31 Ma (Trendall et al., 1990). The thickness
of BIF between the two dated tuffaceous horizons is
estimated to be 210 m (Beukes, 1980b; Barton et al.,
1994). This implies sedimentation rates of 2 B to 6
B, as were calculatedby Arndt et al. (1991) and Bar-
ton et al. (1994) for a mixed lithological succession
of carbonates, shales, BIF and chert. Upon decom-
paction, the BIF sediments reflect a thickness of
2100 m and sedimentation rates of 20 B to 60 B, the
latter thus being comparable to uncompacted pelagic
sediments (Fig. 5) (Mu¨ller and Mangini, 1980).
More recently, Barley et al. (1997) published new
age data for volcanic rocks within the Hamersley
Range of the Pilbara craton in Western Australia,
and derived sedimentation rates (SR) of 30 B and
more, for the BIF and shales included in this suc-
cession. This is an order of magnitude higher than
previous calculations of sedimentation rates (3–4 B)
246 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
for a BIF, volcanic, shale and carbonate rock succes-
sion from the same region (Arndt et al., 1991), or
from South Africa (Barton et al., 1994), and in good
agreement with the results presented here for the
decompacted Kaapvaal craton BIF. However, if as
suggested by Barley et al. (1997), this sedimentation
was related to enormous volcanic and hydrother-
mal activity in the Hamersley basin, differences in
intensity of volcanism may explain divergent BIF
accumulation rates in other basins.
Earlier calculations of the sedimentation rate of
BIF resulted in higher figures than presented above.
Trendall and Blockley (1970, p. 298) arrived at
basin subsidence rates of 2000 to 6000 years per
one foot (50 m to 150 m in one million years) for
the Fortescue and Hamersley Group basins of the
Pilbara craton. Their calculations for the compacted
sedimentation rate for BIF were based on microband
counting (varve model) and the inferred quantities
of annual Fe deposition in the basin. For the Dales
Gorge member BIF, a cSR of 20 B to 70 B can be
deduced from Trendall and Blockley’s (1970, p. 262)
calculation of 2000 to 3000 years of deposition time
for one Knox or Calamina cyclothem (a common
cyclic sequence of banding types in the BIF, and
on average 7 cm and 14 cm thick, respectively,
in the above calculation). This cSR is an order of
magnitude higher than the calculations based on
isotopic age data, as presented here.
For the Kuruman BIF, a weighted average mi-
croband thickness of 0.58 mm was calculated by
Klein and Beukes (1989, p. 1772), and considered
to represent an annual varve. Under this assumption
a compacted sedimentation rate (cSR) of 570 B was
derived. However, with such a high cSR, the 210 m
of Kuruman BIF separating the two dated horizons
of 2500 Ma and 2432 š31 Ma would have been
deposited within less than 0.5 m.y. At Prieska, where
the Kuruman BIF approximates 750 m thickness, the
time of deposition would be less than 1.5 m.y., when
disregarding possible tectonic duplication suggested
by Altermann and Ha¨lbich (1991).
8. Basin historyand tectonic interpretation
The correlation of the Naute Shale member of the
Prieska sub-basin (Figs. 1 and 2) with the Reivilo
to Gamohaan Formations of the Ghaap Plateau sub-
basin (Fig. 3), and of the upper Nauga Formation
carbonates at Prieska with the Monteville Forma-
tion on the Ghaap Plateau and Oaktree Formation
in the Transvaal (Fig. 7), requires a new sedimen-
tation scheme and a new model for basin develop-
ment. The Boomplaas Formation above the Vryburg
Formation established the first carbonate platform
between 2642 and 2588 Ma ago. This was subse-
quently transgressed by the Lokammona Formation
shales (Beukes, 1979) and followed, on the Ghaap
Plateau, by the Monteville Formation when platfor-
mal conditions returned at around 2555 Ma. Before
2588 š6 Ma, almost half of the tidal flat carbonates
in the Prieska area had already accumulated. In the
Transvaal basin and in the Ghaap Plateau sub-basin,
the Oaktree and Monteville carbonate sedimenta-
tion commenced only prior to 2550 Ma and 2555
Ma, respectively. There is as yet no evidence for
continuous and stable carbonate basin development
in the Transvaal sub-basin prior to about 2556 Ma
(Walraven and Martini, 1995). This suggests a trans-
gression, progressing from the west or southwest
towards the east or northeast.
The major transgressive step at around 2550 Ma
drowned the tidal flats in the southwest and shifted
the main site of carbonate sedimentation to the north
and east. At the time of the subtidal and below-
wave-base carbonate sedimentation at Prieska, car-
bonate sediments of the Monteville and Oaktree
Formations had accumulated under platform con-
ditions. Continuous subsidence in the basin centre
was matched by stromatolitic growth (Altermann
and Herbig, 1991) under predominantly subtidal
conditions, locally passing into supratidal settings
(Eriksson and Truswell, 1974). An area of crustal
updoming like the Maremane or Ganyesa Domes
could have served as the source for clastic sediments
intercalated with the Ghaap Plateau carbonates, and
also may have been a barrier between the intracra-
tonic basins. Smith et al. (1990) investigated Sm–Nd
isotopes in shales at the base of the Griqualand West
and Transvaal basin sequences, and found profound
differences in the geochemistry and isotopic char-
acteristics of the sediments in the two sub-basins.
These differences probably reflect different source
areas for the Vryburg and Black Reef shales (Smith
et al., 1990; Barton and Hallbauer,1996).
Carbonate deposition on the Ghaap Plateau and
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 247
in the Transvaal lasted for at least 30 m.y. longer
than in the Prieska area, where the Naute Shales
were deposited between roughly 2550 Ma and 2500
Ma. The source area for the mud influx must have
been located farther to the south or west. Otherwise,
the large amount of pelitic detritus would have over-
whelmed the stromatolitic platform carbonates in the
north and east, if not derived from this SW direction.
As demonstrated by Altermann (1996a), the tuffs in
the Nauga Formation originated in the southwest.
However, it is now clear that the tuffs in the upper
Campbellrand Subgroup, on the Ghaap Plateau, are
younger and not correlative of tuffs in the periti-
dal member of the Nauga Formation. Volcanism in
the southwestern Griqualand West sub-basin could
have accompanied rifting and thermal uplift. Erosion
in these parts of the basin could have exposed the
source area required for derivation of the more than
100 m of Naute Shale member, and for the occa-
sional shale intercalations accumulated on the Ghaap
Plateau.
From about 2500 Ma to 2432 Ma, BIFs of the
Kuruman and Penge Formations were deposited at
>100 m depth, following a substantial deepening
of the basin, to below storm wave-base (Klein and
Beukes, 1989). During deposition of the Griquatown
BIF, a shallowing-upward cycle commenced (Beukes
and Klein, 1990). In southwestern Griqualand West,
the BIFs are conformably followed by the Koegas
Subgroup, a succession mainly comprising shales,
cherts and arkoses, with minor BIF and carbonate
deposits. These shallow marine to deltaic sedimen-
tary rocks are locally over 600 m thick and are cov-
ered, above a low angular unconformity (Altermann
and Ha¨lbich, 1991), by glaciogenic tillites in turn
disconformably followed by the 2222 Ma Ongeluk
lavas (Cornell et al., 1996). In the Transvaal, the
Penge BIF is unconformably covered by the Pretoria
Group siliciclastic rocks, with a hiatus of perhaps
more than 100 Ma. Thus, sedimentation rates for
deposits younger than the BIF cannot be calculated.
It seems certain, however, that between the 2350 Ma
old Bushy Bend lavas (F. Walraven, pers. commun.,
in Eriksson et al., 1995) at the base of the Preto-
ria Group of the Transvaal and the 2222 Ma old
Hekpoort–Ongeluk lavas, at least 400 to 800 m of
mudrock and subordinate sandstones were laid down
in a deep periglacial turbiditic basin, with some distal
deltaic intercalations (Eriksson and Reczko, 1998).
The cSR for these sediments thus ranges between 3
B and 7 B, but decompacted sedimentation rate is in
range of 20 B, assuming a shale to sandstone ratio
of 5: 1 (Eriksson et al., 1995). Despite including a
regional unconformity, these sedimentation rates are
comparable with those of modern deltas and delta-
front shales and turbidites (Fig. 5).
The decompacted sediment thickness of the var-
ious age-bracketed sections of the Schmidtsdrif and
Campbellrand Subgroups are plotted against their
upper age boundaries, in order to obtain a sedimen-
tation curve (Fig. 6). The diagrams in Fig. 6A and
Fig. 6B are based on time-level plots, as proposed
by Friend et al. (1989), but modified to suit our
purposes. These curves are not subsidence curves,
because back-stripping has not been carried out, but
when compared to estimated depth of deposition,
subsidence can be deduced easily from these dia-
grams. The Prieska area (Fig. 6A) and the Kathu
borehole (Fig. 6B) are treated separately. The es-
timated depth of deposition for the Naute Shales
and BIF is below storm wave-base, in water depth
greater than 100 m. The maximum depth is difficult
to determine but could be greater than the 200 m
assumed. The sedimentation rates (SR) for both ar-
eas are plotted against their upper age boundaries in
Fig. 6C. Although in both areas the density of age
data is different and our decompaction data are cer-
tainly imprecise, the different shapes of the curves
in Fig. 6A and Fig. 6B can only be interpreted as
reflecting differing depositional histories for the two
areas. This is even more apparent in Fig. 6C, where
the discrepancy in sedimentation rates between 2550
Ma and 2500 Ma is striking. This difference is due to
varying lithology of highly compactible, slowly ac-
cumulating shales versus poorly compactible, rapidly
growing carbonates, and due to different absolute
amounts of subsidence.
When evaluating the subsidence in both areas, a
similarity in timing becomes apparent. For the time
period between 2550 and 2500 Ma, subsidence ap-
pears to be greatest in both Fig. 6A and Fig. 6B.
Two concave-upward parts of the curve connect-
ing the bars of estimated depth of deposition are
recognizable in Fig. 6A. This curve shape can be
interpreted as typical of a rift basin or passive con-
tinental margin. For the Kathu borehole (Fig. 6B),
248 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
Fig. 6. Curves for decompacted thickness of sediment and estimated depth of its deposition for the Prieska area (A) and the Ghaap Plateau
(B). Differences in subsidence and compaction (due to varying sediment types) can be deduced from comparison of both curves. Greatest
subsidence is implied by the thickness of the less compactible carbonates of the Ghaap Plateau, which remain at shallow depositional
depth between 2555 Ma and 2516 Ma (B). Approximately at the same time, mainly highly compactible shales were accumulated in the
Prieska area and the depth of deposition for the shales increased to below storm wave-base (A). Since the overlying BIFs were deposited
at 100 m to 200 m depth (compare discussion in text), but are highly compactible (around 90%), subsidence must have been less
effective than compaction during BIF deposition. This is emphasized in (C), where the poorly compactible shallow-platform carbonates
of the Ghaap Plateau are shown to have the highest sedimentation rate. Note that the shaded and solid bars in (C) mark different areas of
deposition, while in (A) and (B) they indicate different scales of decompacted sediment thickness and depth of deposition.
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 249
a single rift-related curve shape extends for an ad-
ditional 100 m.y. backwards in time. The curve,
however, is based on less age-data, which may be
the reason for this divergent behaviour. Rift basin
or passive continental margin-related subsidence is
typically a slow and long-lasting activity (McCabe,
1991). Interpretation of a rift basin–passive margin
analogue for the Griqualand West basin is suggested
by the marked scarcity of clastic debris, other than
suspension deposits in the Campbellrand and As-
bestos Hills Subgroups, and also supported by the
presence of basaltic volcanism within the carbon-
ates (Altermann, 1996a). Passive margins are the
preferred site of formation of carbonate platforms.
The acceleration of subsidence during the period of
2550 Ma to 2500 Ma might be due to overlapping
mechanical and thermal subsidence. Usually, the ini-
tial mechanical subsidence is accompanied by higher
heat flow due to stretching, which is then followed
by thermal subsidence due to cooling of the stretched
asthenosphere (Einsele, 1992). Where both processes
overlap, faster subsidence would be expected.
9. Implications for sediment deposition in the
Neoarchaean
From the calculations above, it is difficult to un-
derstand why the growth of stromatolitic carbonate
platforms was incapable of continuing after ma-
jor transgressions. A paradox of drowned carbonate
platforms like that postulated for many modern and
Phanerozoic carbonates (Schlager, 1981) can be de-
tected also in the Neoarchaean. Modern carbonate
accumulation can match subsidence rates or relative
sea-level increases of up to 500 m=Ma (Aigner et al.,
1989; Bosellini, 1989; Chivas et al., 1990; Bosscher
and Schlager, 1993). Over short duration, the accu-
mulation rates of carbonates may be even higher,
in excess of 1000 m=Ma. The drowning of carbon-
ate platforms in the Phanerozoic requires subsidence
rates or sea-level rise in excess of 4000 m=Ma, at
least for a short period of time, until the basin floor
sinks below the zone of euphotic activity (Schlager,
1989). Sedimentation rates in the order of a few
metres per million years are typical of Phanerozoic
oceanic pelagic deposits that lack terrigenous sedi-
ment influx and which are fed only by planktonic
rain. Sediment deposition in the order of a few tens
of metres per million years, as calculated for the de-
posits of the Kaapvaal craton, are known from young
tidal flats and carbonate shelves of low subsidence
rate (Scholle et al., 1983; Einsele, 1992).
Because of the calculated sedimentation rates and
the conspicuous lack of coarse-clastic sedimentary
rocks in the entire Griqualand West and Transvaal
sub-basin successions, the subsidence rates reflected
in the major transgression episodes and follow-
ing sedimentation cycles must have been moder-
ate. It is evident that stromatolitic growth-rates un-
der favourable conditions should have been able
to match the subsidence rate. Archaean stroma-
tolites and carbonates, however, differ from their
Neoproterozoic and Phanerozoic counterparts in the
scarcity of coarse pelletal carbonate sands, due pre-
sumably to the lack of organisms producing such
pellets or carbonate skeletons, that could be worn
to produce carbonate arenites (Grotzinger, 1989).
Therefore, Archaean stromatolites are mainly finely
laminated and trap rare carbonate detritus between
the column branches, whereas younger stromatolites
tend to trap and bind carbonate sands also along
the microbial lamination. This may lead to the de-
velopment of finer lamination and slower growth
in Archaean stromatolites. Klein et al. (1987) and
Altermann and Schopf (1995) demonstrated the con-
spicuous lack of detrital grains in microfossiliferous
stromatolites of the Campbellrand Subgroup. Addi-
tionally, in the photomicrographs presented by Klein
et al. (1987), minute aragonite needles can be ob-
served in the Siphonophycus transvaalensis mats and
filaments, evidence that such microbiota, after decay,
could contribute only to the production of micrite.
Grotzinger (1989) proposed that carbonate precipi-
tation could also have been triggered indirectly by
photosynthetic decrease of the CO2content of the
seawater, in the presence of cyanobacterial bioherms.
Such ‘chemical’ precipitation might be the reason
for the slow sedimentation rates of Archaean car-
bonates compared to modern reefs (Fig. 5). Sumner
and Grotzinger (1996) proposed a purely chemical
subtidal precipitation of carbonate in a saturated ma-
rine environment for parts of the Griqualand West
and other Precambrian deposits, which would prob-
ably account for even lower sedimentation rates.
On the other hand, there is ample evidence, for
the cyanobacterial and bacterial communities, that
250 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
their metabolism was similar to that observed today
(Schopf and Klein, 1992; Schopf, 1993). In support
of this, Lanier (1986) calculated organic production
rates for microfossils from the Malmani Subgroup
stromatolites in the range of modern microbial mats,
and thus the growth rates of Precambrianand modern
stromatolites should be comparable, provided that a
similar sediment binding or precipitation mechanism
operated. The calculated decompacted sedimentation
rates of over 150 B for the stromatolitic carbonates of
the Ghaap Plateau also support comparable growth
rates for modern and Precambrian stromatolites.
The apparent slow drowning may be due to an
inability of the microbial organisms secreting or fa-
cilitating the precipitation of carbonate to cope with
possible climaticchanges or changes in the chemical
environment. The latter could have been influenced
by increased hydrothermal activity, as suggested by
BIF geochemistry (Klein and Beukes, 1989). For
the climate, it can be speculated that the fixation of
CO2in the first giant carbonate platforms of Africa,
Australia, northern America and India reduced the
greenhouse effect and led to lower temperatures, as
discussed for the Upper Precambrian by many in-
vestigators (e.g. Eriksson et al., 1998). As no higher
organisms directly secreting carbonate were present,
growth rates comparable to those of coral reefs, for
example, seem unlikely in the Early and Middle Pre-
cambrian. The drowning of the carbonate platforms
was thus probably facilitated by the development
of unfavourable conditions for stromatolitic growth.
The drowning of the carbonates of the Nauga For-
mation (2549 Ma) and of the Gamohaan carbonate
platform (2516 Ma) coincide with two ‘events’, re-
spectively: the introduction of pelitic sediment into
the basin from the southwest, and the increase of Si
and Fe content in the sediments and by inference in
the seawater, from volcanic and hydrothermal activ-
ity (Klein and Beukes, 1989; Barley et al., 1997).
As no thick shales separate the Gamohaan Forma-
tion from the Kuruman BIF, shale sedimentation
alone cannot be responsible for the drowning of the
carbonate platforms. Other factors, such as climatic
changes may thus have hastened the end of carbon-
ate deposition. The increasing Fe and Si content led
finally to precipitation of BIF when pelitic sediment
deposition was insignificant. During pelitic(and car-
bonate) sedimentation, the Fe dissolved in the water
must have been bound by clay or carbonate minerals
and no BIFs were precipitated. Ha¨lbich et al. (1992),
have demonstrated that the shales of the carbonate–
BIF transition have comparable Fe contents to the
BIF units.
10. Conclusions
Long periods of non-deposition and erosion have
been assumed for peritidal and associated silici-
clastic and carbonate facies deposits of the Kaap-
vaal craton. Decompacted sedimentation rates were
nonetheless calculated using conservative estimates
of compaction and dissolution, and no back-strip-
ping was performed. The results obtained probably
underestimate the true sedimentation rates. They are
generally comparable to Phanerozoic deposits, when
observed over similarly long periods of time, and,
with consideration of times of non-deposition, to
modern sedimentation rates.
Four long-term (millions of years) transgression–
regression cycles can be recognized in the Griqua-
land West sub-basin and can be partly correlated to
Transvaal sub-basin deposits. (1) The Vryburg For-
mation and the Boomplaas carbonate platform sedi-
ments represent, respectively, the first transgression
and a succeeding long-duration shallowing-upward
cycle in Griqualand West. (2) The second transgres-
sive step is marked by the Lokammona shales and the
overlying shallowing-upward (regressive) cycle rep-
resented by the lower Nauga Formation. (3) The up-
per Nauga Formation (chert member) together with
the Monteville and Oaktree Formations mark the
third transgressive phase, followed by a long-term
shallowing-upward regressive cycle with the devel-
opment of a stromatolitic carbonate platform that
formed the Campbellrand and Malmani Subgroups.
(4) The fourth transgressive step is the development
of the Kuruman and Penge BIF-pelagic basin in all
provinces. The Griquatown BIF and the overlying
Koegas siliciclastics mark the fourth regressive cy-
cle, not preserved in the Transvaal sub-basin, due to
a possibly 100 m.y. long period of erosion before the
Pretoria Group sediments were laid down (Eriksson
et al., 1995).
An attempt to correlate these four cycles in the
Prieska, Ghaap Plateau and Transvaal areas is sum-
marized in Fig. 7. In this figure, the transgression–
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 251
Fig. 7. Comparison of recognizable second-order transgression–regression cycles for the Prieska, Ghaap Plateau and Transvaal areas.
The most complete curve can be sketched for the Prieska area and the least complete for the Transvaal area, due to varying density
of age data and preservation of the sedimentary column. Nevertheless, similar development for the upper part of the curve can be seen
(2550 Ma to 2432 Ma approximately). Note the uncertainties in the shape of the curves and the differences in facies and lithology at
Prieska compared to the Ghaap Plateau and Transvaal basin areas, as discussed in the text. The cited ages are: 2222š13 Ma for Ongeluk
Formation (Cornell et al., 1996); 2350 Ma for Bushy Bend lavas in the Timeball Hill Formation (F. Walraven, pers. commun., in Eriksson
et al., 1995); 2432 š31 Ma for the lower Griquatown BIF (Trendall et al., 1990, and discussion by Barton et al., 1994); 2500 Ma for
the base of the Kuruman and Penge BIF (Trendall et al., 1995); 2516 š4 Ma for the upper Gamohaan Formation; 2521 š3 Ma for the
upper Gamohaan Formation (Sumner and Bowring, 1996) and 2555š19 Ma for the upper Monteville Formation on the Ghaap Plateau
(this work); 2549 š7 Ma for the chert member and 2588š6 Ma for the peritidal member of the Nauga Formation at Prieska (this work);
2550 š3 Ma for the upper Oaktree Formation in the Transvaal sub-basin (Walraven and Martini, 1995); 2641 š3MafortheVryburg
Formation on the Ghaap Plateau (Walraven et al., in press); 2709 š4 Ma for the upper Ventersdorp Supergroup (Makwassie quartz
porphyry, Armstrong et al., 1991). The other formations are not dated and their correlation across the different sub-basins is uncertain
(see text). The sedimentary section between the Vryburg (Black Reef) Formation and the BIF (including Koegas Subgroup at Prieska)
represents a first-order cycle as recognized by Cheney (1996). Cycles of third and fourth order, as described by Clendenin (1989), are not
shown in this figure. They may be included in the two transgressions of the Schmidtsdrif Subgroup or between the Monteville and the
Reivilo Formations. Because of lack of age data, however, their duration can only be calculated using compacted sediment thickness.
252 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
regression cycles 1 and 2, as described above, are not
recognized in the Transvaal sub-basin due to the lack
of age data, and due to the uncertain age and cor-
relation of the Black Reef Formation. For the same
reason, the third cycle (the Oaktree Formation trans-
gression) cannot be separated from the Black Reef.
The role of the Black Reef Formation in this scenario
and its possible correlation to the Vryburg Forma-
tion are equivocal and require further confirmation
(see discussion by Walraven and Martini, 1995). The
upper Black Reef encompasses a transgressive shale
cover, which grades into the overlying carbonates of
the Oaktree Formation (Clendenin, 1989). Shallow
platform carbonates are developed above the Oak-
tree and Monteville Formations. Only at Prieska,
did deposition of the Naute Shale member remain
predominantly below the wave-base, perhaps with
the exception of the silicified and brecciated inter-
vals, which may correlate with some regressions in
the Monte Christo Formation and below the Lyttel-
ton Formation in Transvaal (Clendenin, 1989). The
fourth cycle leadingto the deposition of banded iron
formations, is recognizable in all three sub-basins.
The shape of the transgression–regression curves in
Fig. 7 is inferred because the duration of these events
and their relative intensity (i.e. the amount of relative
sea-level rise or fall) are not known. Slow transgres-
sions followed by rapid regressions are, however,
typical of Phanerozoic sea-level fluctuations (Cloet-
ingh et al., 1985).
These four large-scale transgressive–regressive
(shallowing-upward) cycles are in the order of tens
of millions of years duration, and must therefore
be attributed to second-order sequences of crustal
evolution (Vail et al., 1991). They represent ma-
jor regional transgressions and regressions and build
sequence cycles that can be subdivided into se-
quences comprising third-order system tracts, and
into fourth-order parasequences. Cyclicity above the
second-order (duration of less than 3 m.y.) can be de-
duced from lithological columns, such as the Kathu
borehole, and differ regionally in their facies, as
comparisons to adjacent realms demonstrate (e.g. on
the Ghaap Plateau: Altermann, 1997, Altermann and
Siegfried, 1997; or in the Transvaal basin: Clen-
denin, 1989). More detailed facies work and precise
age data are needed for identificationand correlation
of system tracts, parasequences and Milankovitch
cycles in Archaean and Palaeoproterozoic basins.
For example, the conspicuous stromatolite cyclic-
ity observed in many formations of the Transvaal
Supergroup (Eriksson and Truswell, 1974) may be
attributed to fourth- and fifth-order cycles, of 0.1–
0.5 m.y. and 0.02–0.1 m.y. duration, respectively
(Mitchum and van Wagoner, 1991).
Comparison of the Transvaal Supergroup depos-
itories with other Archaean and Palaeoproterozoic
basins is hindered by the lack of sufficient data. The
only basin equally well investigated as the Kaap-
vaal craton sub-basins, is the Hamersley basin of
the Pilbara craton, Western Australia. Both basins
are of comparable age, contain comparable sedi-
mentary fills, and have even been considered to be
parts of the same original cratonic depositional sys-
tem (Trendall, 1968; Button, 1976; Cheney, 1996).
First-order cycles (>50 Ma), on the Kaapvaal cra-
ton, can be identified in the four sequences recog-
nized by Cheney (1996). In his interpretation, the
lower Transvaal Supergroup (>2432 Ma) is the sec-
ond first-order cycle, the first being represented by
sedimentary and volcanic rocks of the Ventersdorp
Supergroup. Based on our interpretation presented
here, we would rather extend this second first-order
tectonically driven cycle to include the Koegas Sub-
group of Griqualand West, and thus to have an upper
age limit of <2432 Ma (Griquatown BIF) for this
second cycle. However, as Cheney (1996) correctly
stated, with more data the sequences will have to be
redefined in the future, and the boundaries will shift
with the identification of other sequences. The last
two first-order cycles recognized in the Transvaal
Supergroup, the Pretoria and the Rooiberg Groups,
are not discussed here, but it should be emphasized
that Cheney (1996) was also able to identify three
of the four unconformity-bounded sequences on the
Pilbara craton of Western Australia. Thus, our sec-
ond-order cycles identified herein may be found also
within the Hamersley Group, if Cheney (1996) is
correct. From published data (Trendall et al., 1990,
1995; Arndt et al., 1991; Barley et al., 1997) it is,
however, evident that sedimentation rates and lithos-
tratigraphic sequences of both cratonic basins are
similar across the Archaean–Proterozoic boundary.
Analyses of subsidence and sedimentation rates
for Precambrian sedimentary basins are still an ex-
ception, compared to the much more regularly pub-
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 253
lished pure facies descriptions and lithostratigraphic
correlations. To our knowledge, the most recently
published attempt to draw decompacted sedimenta-
tion rate and subsidence curves for Precambrian sed-
imentary basins was by Maynard and Klein (1995).
These authors investigated the subsidence history of
the clastic Witwatersrand basin, underlying the Ven-
tersdorp Supergroup forming the basement to the
rocks discussed herein. The Witwatersrand deposits
are >2800 Ma and consist mainly of sandstones
and shales with subordinate conglomerates, and have
been interpreted as having been laid down within a
strike-slip modified retroarc foreland-basin. Because
the Witwatersrand sedimentscontain the largest gold
deposits known, it has attracted the most attention
from sedimentologists (e.g. Bickle and Eriksson,
1982; Burke et al., 1986; Winter, 1987; Stanistreet
and McCarthy, 1990) and a large amount of data and
manifold interpretations have been presented. The
calculations by Maynard and Klein (1995), using a
computer program to correct for compaction and the
load of the sediment fill, resulted in subsidence rates
of 40 m=Ma to 50 m=Ma, on average, for the Do-
minion and West Rand Groups of the Witwatersrand
Supergroup.
A similar exercise was performed for the ¾1100
Ma to 1050 Ma old White Pine Cu deposit of north-
ern Michigan, USA (Maynard and Klein, 1995).
Rates of subsidence for the sedimentary phase of
the basin were calculated to be greater than 80
m=Ma for shales, sandstones and conglomerates.
These rates were, however, calculated on a much
smaller observational scale, of less than ten million
years, separating the available age-data points. The
basin was interpreted as a rift-related basin with sev-
eral episodes of mechanical subsidence. Although a
direct comparison of our calculations with those per-
formed by Maynard and Klein (1995) is not possible,
because of the lack of back-stripping in our calcu-
lations, the results in all three Precambrian basins
published so far (and obviously from the Pilbara
sedimentary successions, as well as from the lower
Pretoria Group) are similar, and are also comparable
to Phanerozoic sedimentation and subsidence rates.
Precambrian siliciclastic and chemical sediments
accumulated at rates comparable to their younger
equivalents. From the calculated sedimentation rates,
a conclusion as to the subsidence rates of Archaean
and Palaeoproterozoic basins is also possible. The
evolution of Precambrian biochemical, chemical and
siliciclastic sedimentary basins imply long-term first-
and second-order cyclicity, as in the Phanerozoic,
and thus similar crustal processes and analogous
rates of denudation. Similarly, the biological and
chemical processes of carbonate sedimentation in
varying facies realms appear comparable to their
Phanerozoic and modern carbonate facies equiva-
lents, and thus a similar metabolism and evolution-
ary stage of carbonate-fixing stromatolitic microbial
organisms are inferred.
Acknowledgements
Zircon analyses were carried out on the Sensitive
High-Resolution Ion Microprobe mass spectrome-
ter located at Curtin University of Technology. The
SHRIMP II laboratory is supported by the Aus-
tralian Research Council. We thank especially J.R.
de Laeter, Allan Kennedy, Bob Pidgeon and Alec
Trendall for their kind support and the Geological
Survey of Western Australia for excellent sample
mounting. WA was supported by the German Re-
search Foundation (DFG) grant DFG Al 295=3-3
and by a stipendium to the Curtin University and
greatly enjoyed the organization and collaborative
spirit at this institution, but especially the com-
pany of Frank So¨llner (IAAG-LMU). Frank helped
at various stages of the investigations, with endless
discussions and contributed substantially to our suc-
cess. Pat Eriksson patiently waited for this paper.
Alec Trendall also corrected an early version of the
manuscript and helped with many critical remarks
and suggestions. Pat Eriksson and an anonymous
referee critically reviewed and improved contents
and style of this contribution.
References
Aigner, T., Doyle, M., Lawrence, D., Epting, M., von Vliet,
A., 1989. Quantitative modeling of carbonate platforms: some
examples. Soc. Econ. Paleontol. Mineral. Spec. Publ. 44, 28–
37.
Altermann, W., 1990. Facies development at the Archean–
Proterozoic boundary, along the southwestern rim of the Kaap-
vaal craton, South Africa. In: Glover, J.E., Ho, S.E. (Eds.), 3rd
Int. Archean Symp., Perth. Geoconferences (W.A.) Inc., Perth
ed. Vol. Extended Abstracts, pp. 299–301.
254 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
Altermann, W., 1996a. Sedimentology, geochemistry and palaeo-
geographic implications of volcanic rocks in the upper
Archean Campbell Group, western Kaapvaal Craton, South
Africa. Precambrian Res. 79, 73–100.
Altermann, W., 1996b. Discussion on ‘Zircon Pb–evaporation
age determinations of the Oak Tree Formation, Chuniespoort
Group, Transvaal Sequence: implications for the Transvaal–
Griqualand West basin correlations’. S. Afr. J. Geol. 99 (3),
337–338.
Altermann, W., 1997. Sedimentological evaluation of Pb–Zn
exploration potential of the Precambrian Griquatown Fault
Zone in the Northern Cape Province, South Africa. Mineral.
Deposita 32, 382–391.
Altermann, W., Ha¨lbich, I.W., 1991. Structural history of the
south western corner of the Kaapvaal Craton and the adjacent
Namaqua Realm: New observations and reappraisal. Precam-
brian Res. 52, 133–166.
Altermann, W., Herbig, H.G., 1991. Tidal flat deposits of the
Lower Proterozoic Campbell Group along the southwestern
margin of the Kaapvaal Craton, Northern Cape Province,
South Africa. J. Afr. Earth Sci. 13 (3/4), 415–435.
Altermann, W., Nelson, D.R., 1996. Neoarchean carbonate plat-
form development: An example from the Kaapvaal Craton,
South Africa. 17th Int. Assoc. Sedimentol. Reg. African–
European Meeting, March, 1996, Sfax, Tunisia, Abstr. Vol.,
pp. 8–9.
Altermann, W., Schopf, J.W., 1995. Microfossils from the
Neoarchean Campbell Group, Griqualand West Sequence of
the Transvaal Supergroup, and their paleoenvironmental and
evolutionary implications. Precambrian Res. 75, 65–90.
Altermann, W., Siegfried, H.P., 1997. Sedimentology, facies de-
velopment and a type-section of an Archean shelf-carbonate
platform transition, Kaapvaal craton, as deduced from a deep
borehole at Kathu, South Africa. J. Afr. Earth Sci. 24 (3),
391–410.
Altermann, W., Wotherspoon, J.McD., 1995. The carbonates of
the Transvaal and Griqualand West Sequences of the Kaapvaal
craton, with special reference to the Lime Acres limestone
deposit. Mineral. Deposita 30 (2), 124–134.
Armstrong, R.A., Compston, W., Retief, E.A., Williams, I.S.,
Welke, H.J., 1991. Zircon ion microprobe studies bearing on
the age and evolution of the Witwatersrand Triad. Precambrian
Res. 53, 243–266.
Arndt, N.T., Nelson, D.R., Compston, W., Trendall, A.F., Thorne,
A.M., 1991. The age of the Fortescue Group, Hamersley
Basin, Western Australia, from ion microprobe zircon U–Pb
results. Aust. J. Earth Sci. 38, 261–281.
Barley, M.E., Pickard, A.L., Sylvester, P.J., 1997. Emplacement
of large igneous province as a possible cause of banded iron
formation 2.45 billion years ago. Nature 385, 55–58.
Barton, E.S., Hallbauer, D.K., 1996. Trace-element and U–Pb
isotope composition of pyrite types in the Proterozoic Black
Reef, Transvaal Sequence, South Africa: Implications on gen-
esis and age. Chem. Geol. 133, 173–199.
Barton, E.S., Altermann, W., Williams, I.S., Smith, C.B., 1994.
U–Pb zircon age for a tuff in the Campbell Group, Griqualand
West Sequence, South Africa: Implications for Early Protero-
zoic rock accumulation rates. Geology 22 (4), 343–346.
Beukes, N.J., 1979. Litostratigrafiese onderverdeling van die
Schmidtsdrif-Subgroep van die Ghaap-Groep in Noord-Kaap-
land. Trans. Geol. Soc. S. Afr. 82, 313–327.
Beukes, N.J., 1980a. Stratigrafie en litofasies van die Camp-
bellrand-Subgroep van die Proterofitiese Ghaap-Groep, No-
ord-Kaapland. Trans. Geol. Soc. S. Afr. 83, 141–170.
Beukes, N.J., 1980b. Lithofacies and stratigraphy of theKuruman
and Griquatown Iron-Formations, Northern Cape Province,
South Africa. Trans. Geol. Soc. S. Afr. 83, 69–86.
Beukes, N.J., 1986. The Transvaal Sequence in Griqualand West.
In: Anhaeusser, C.R., Maske, S. (Eds.), Mineral Deposits of
Southern Africa, Vol. I. Geological Society of South Africa,
Johannesburg, pp. 819–828.
Beukes, N.J., Klein, C., 1990. Geochemistry and sedimentology
of a facies transition — from microbanded to granular iron-
formation — in the early Proterozoic Transvaal Supergroup,
South Africa. Precambrian Res. 47, 99–139.
Bickle, M.J., Eriksson, K.A., 1982. Evolution and subsidence of
early Precambrian sedimentary basins. R. Soc. Philos. Trans.
A 305, 225–247.
Bosellini, A., 1989. Dynamics of Tethyan carbonate platforms.
Soc. Econ. Paleontol. Mineral. Spec. Publ. 44, 3–13.
Bosscher, H., Schlager, W., 1993. Accumulation rates of carbon-
ate platforms. J. Geol. 101, 345–355.
Burke, K., Kidd, W.F.S., Kusky, T.M., 1986. Archean foreland
basin tectonics in the Witwatersrand, South Africa. Tectonics
5, 439–456.
Button, A., 1972. Algal stromatolites of the early Proterozoic
Wolkberg Group, Transvaal sequence. Economic Geology Re-
search Unit, Univ. Witwatersrand, Johannesburg, Info. Circ.
69, 9 pp.
Button, A., 1976. Transvaal and Hamersley basins — Review of
basin development and mineral deposits. Miner. Sci. Eng. 8
(4), 262–293.
Cheney, E.S., 1996. Sequence stratigraphy and plate tectonic
significance of the Transvaal succession of southern Africa
and its equivalent in Western Australia. Precambrian Res. 79,
3–24.
Chivas, A.R., Torgersen, T., Pollach, H.A., 1990. Growth rates
and Holocene development of stromatolites from Shark Bay,
Western Australia. Aust. J. Earth Sci. 37, 113–121.
Clendenin, C.W., 1989. Tectonic Influence on the Evolution of
the Early Proterozoic Transvaal Sea, Southern Africa. Ph.D.
Thesis, Univ. Witwatersrand, Johannesburg, 376 pp.
Cloetingh, S., McQueen, H., Lambeck, K., 1985. On a tectonic
mechanism for regional sealevel variations. Earth Planet. Sci.
Lett. 75, 157–166.
Compston, W., Williams, I.S., Meyer, C., 1984. U–Pb
geochronology of zircons from lunar breccia 73217 using a
sensitive high mass-resolution ion microprobe. J. Geophysical
Res. 89, B252–B534.
Cornell, D.H., Schu¨tte, S.S., Eglington, B.L., 1996. The On-
geluk Basaltic Andesite Formation in Griqualand West, South
Africa: submarine alteration in a 2222 Ma Proterozoic sea.
Precambrian Res. 79 (1/2), 101–124.
W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256 255
Dravis, J.J., 1997. Rapidity of freshwater calcite cementation —
implications for carbonate diagenesis and sequence stratigra-
phy. Sediment. Geol. 107 (1/2), 1–10.
Drummond, C.N., Wilkinson, B.H., 1993. Aperiodic accumula-
tion of cyclic peritidal carbonate. Geology 21, 1023–1026.
Einsele, G., 1992. Sedimentary Basins Evolution, Facies and
Sediment Budget. Springer, Berlin, 628 pp.
Eriksson, K.A., Truswell, J.F., 1974. Tidal flat associations from
the lower Proterozoic carbonate sequence in South Africa.
Sedimentology 21, 293–309.
Eriksson, P.G., Reczko, B.F.F., 1998. Contourites associated with
pelagic mudrocks and delta fed turbidites in the Early Protero-
zoic Timeball Hill Formation, Transvaal Supergroup, South
Africa. Sediment. Geol. 120, 319–335, this volume.
Eriksson, P.G., Hattingh, P.J., Altermann, W., 1995. An overview
of the geology of the Transvaal Sequence and Bushveld Com-
plex, South Africa. Mineral. Deposita 30 (2), 98–111.
Eriksson, P.G., Condie, K.C., Tirsgaard, H., Mueller,W.U., Alter-
mann, W., Miall, A.D., Aspler, L.B., Catuneanu, O., Chiaren-
zelli, J.R., 1998. Precambrian clastic sedimentation systems.
Sediment. Geol. 120, 5–53, this volume.
Findlay, D., 1994. Diagenetic boudinage, an analogue model for
the control on hematite enrichment in iron ores of the Hamers-
ley Iron Province of Western Australia, and a comparison with
Krivoy Rog of Ukraine, and Nimba Range, Liberia. Ore Geol.
Rev. 9, 311–324.
Friend, P.F., Johnson, N.M., McRae, L.E., 1989. Time-level plots
and accumulation patterns of sediment sequences. Geol. Mag.
126, 491–498.
Goldhammer, R.K., 1997. Compaction and decompaction algo-
rithms for sedimentary carbonates. J. Sediment. Petrol. 67,
26–35.
Grotzinger, J.P., 1989. Facies and evolution of Precambrian car-
bonate depositional systems: Emergence of the modern plat-
form archetype. Soc. Econ. Paleontol. Mineral. Spec. Publ. 44,
79–106.
Grotzinger, J.P., 1990. Geochemical model for Proterozoic stro-
matolite decline. Am. J. Sci. 290-A, 80–103.
Ha¨lbich, I.W., Lamprecht, D.F., Altermann, W., Horstmann, U.E.,
1992. A carbonate-banded iron formation transition in the
early Proterozoikum of South Africa. J. Afr. Earth Sci. 15 (2),
217–236.
Hassler, S.W., 1993. Depositional history of the Main Tuff Inter-
val of the Wittenoom Formation, late Archean–early Protero-
zoic Hamersley Group, Western Australia. Precambrian Res.
60, 337–359.
Hofmann, H.J., Masson, M., 1994. Archean stromatolites from
Abitibi greenstone belt, Quebec, Canada. Geol. Soc. Am. Bull.
106, 424–429.
Jahn, B-m., Simonson, B.M., 1995. Carbonate Pb–Pb ages of
the Wittenoom Formation and Carawine Dolomite, Hamersley
Basin, Western Australia (with implications for their correla-
tion with the Transvaal Dolomite of South Africa. Precambrian
Res., 72: 247–261.
Jahn, B.-m., Bertrand-Sarfati, J., Morin, N., Mace, J., 1990. Di-
rect dating of stromatolitic carbonate from the Schmidtsdrif
Formation (Transvaal Dolomite), South Africa, with implica-
tions on the age of the Ventersdorp Supergroup. Geology 18,
1211–1214.
Kiefer, R., Franck, C., Klemm, D.D., Altermann, W., 1995.
Stratigraphic subdivision and facies distribution of the Camp-
bellrand and Asbestos Hills Subgroups near Prieska, Griqua-
land West. Centennial Geocongress, Geol. Soc. S. Afr., April,
Johannesburg, Abstr. Vol. II, pp. 1154–1157.
Klein, C., Beukes, N.J., 1989. Geochemistry and sedimentology
of a facies transition from limestone to iron-formation depo-
sition in the early Proterozoic Transvaal Supergroup, South
Africa. Econ. Geol. 84, 1733–1774.
Klein, C., Beukes, N.J., Schopf, J.W., 1987. Filamentous micro-
fossils in the Early Proterozoic Transvaal Supergroup: Their
morphology, significance, and paleoenvironmental setting. Pre-
cambrian Res. 36, 81–94.
Lanier, W.P., 1986. Approximate growth rates of Early Protero-
zoic microstromatolites as deduced by biomass productivity.
Palaios 1, 525–542.
Maynard, J.B., Klein, G.D., 1995. Tectonic subsidence analysis
in the characterization of sedimentary ore deposits: Examples
from the Witwatersrand (Au), White Pine (Cu), and Molango
(Mn). Econ. Geol. 90, 37–50.
McCabe, P.J., 1991. Tectonic controls on coal accumulation. Soc.
Geol. Fr. Bull. 162, 277–282.
Mitchum, R.M., van Wagoner, J.C., 1991. High-frequency se-
quences and their stacking patterns: sequence-stratigraphic ev-
idence of high-frequency eustatic cycles. Sediment. Geol. 70,
131–160.
Mu¨ller, P.J., Mangini, A., 1980. Organic carbon decomposition
rates in sediments of the Pacific manganese nodule belt by
230Th and 231Pa. Earth Planet. Sci. Lett. 51, 94–114.
Osleger, D., 1994. Aperiodic accumulation of cyclic peritidal
carbonate: Comment and reply. Geology 22, 479–480.
South African Committee for Stratigraphy (SACS), 1980.
Stratigraphy of South Africa, Part 1 (Comp. L.E. Kent).
Lithostratigraphy of the Republic of South Africa, South
West Africa=Namibia, and the Republics of Bophuthatswana,
Transkei and Venda. Handbook Geol. Surv. S. Afr. 8, 690 pp.
Schlager, W., 1981. The paradox of drowned reefs and carbonate
platforms. Geol. Soc. Am. Bull. 92, 197–211.
Schlager, W., 1989. Drowning unconformities on carbonate plat-
forms. Soc. Econ. Paleontol. Mineral. Spec. Publ. 44, 15–
25.
Scholle, P.A., Bebout, D.G., Moore, C.H., 1983. Carbonate depo-
sitional environments. Am. Assoc. Pet. Geol. Mem. 33, 345–
440.
Schopf, J.W., 1993. Microfossils of the Early Archean Apex
Chert: New evidence of the antiquity of life. Science 260,
640–646.
Schopf, J.W., Klein, C., 1992. The Proterozoic Biosphere: A
Multidisciplinary Study. Cambridge University Press, New
York, 1348 pp.
Simonson, B.M., Hassler, S.W., 1997. Was the deposition of
large Precambrian Iron Formations linked to major marine
transgressions? J. Geol. 104, 665–676.
Smith, C.B., Duane, M.J., Clendenin, C.W., 1990. Sm–Nd iso-
topic systematics of shales from the Chuniespoort and Ghaap
256 W. Altermann, D.R. Nelson/Sedimentary Geology 120 (1998) 225–256
Groups, Transvaal basin. Geocongress ’90, Cape Town, pp.
518–520.
Stanistreet, I.G., McCarthy, T.S., 1990. Middle Ventersdorp
graben development north of Carletonville goldfield and its
relevance to the evolution of the Witwatersrand Basin. S. Afr.
J. Geol. 93, 272–288.
Sumner, D.Y., Bowring, S.A., 1996. U–Pb geochronologic con-
strains on deposition of the Campbellrand Subgroup, Transvaal
Supergroup, South Africa. Precambrian Res. 79 (1/2), 25–36.
Sumner, D.Y., Grotzinger, J.P., 1996. Herringbone calcite: Pet-
rography and environmental significance. J. Sediment. Res. A
66 (3), 419–429.
Trendall, A.F., 1968. Three great basins of Precambrian banded
iron formation deposition: a systematic comparison. Geol. Soc.
Am. Bull. 79, 1527–1544.
Trendall, A.F., Blockley, J.G., 1970. The iron formations of the
Precambrian Hamersley Group, Western Australia. Geol. Surv.
Bull. 119, 1–366.
Trendall, A.F., Morris, R.C., 1983. Iron-formation: Facts and
problems. Developments in Precambrian Geology 6. Elsevier,
Amsterdam, 558 pp.
Trendall, A.F., Compston, W., Williams, I.S., Armstrong, R.A.,
Arndt, N.T., McNaughton, N.J., Nelson, D.R., Barley, M.E.,
Beukes, N.J., de Laeter, J.R., Retief, E.A., Thorne, A.M.,
1990. Precise zircon U–Pb chronological comparison of the
volcano-sedimentary sequences of the Kaapvaal and Pilbara
Cratons between about 3.1 and 2.4 Ga. In: Glover, J.E., Ho,
S.E. (Eds.), 3rd Int. Archean Symp., Perth. Geoconferences
(W.A.) Inc., Perth ed. Vol. Extended Abstracts, pp. 81–83.
Trendall, A.F., de Laeter, J.R., Nelson, D.R., 1995. Chronology
of Gondwana BIFs: Progress report from recent zircon U–Pb
results. 3rd Aust. Conf. Geochronology. Curtin University ed.
Vol. Abstracts, Perth, pp. 1–3.
Vail, P.R., Audemard, F., Bowman, S.A., Eisner, P.N., Perez-
Cruz, C., 1991. The stratigraphic signatures of tectonics,
eustasy and sedimentology. In: Einsele, G., Ricken, W.,
Seilacher, A. (Eds.), Cycles and Events in Stratigraphy. Am.
Assoc. Pet. Geol. Mem. 26, 617–659.
Walraven, F., Martini, J., 1995. Zircon Pb-evaporation age de-
terminations of the Oak Tree Formation, Chuniespoort Group,
Transvaal Sequence: implications for Transvaal–Griqualand
West basin correlations. S. Afr. J. Geol. 98 (1), 58–67.
Walraven, F., Beukes, N.J., Retief, E.A., in press. 2.64 Ga zircons
from the Vryburg Formation, Griqualand West: implications
for the age of the Transvaal Supergroup and accumulation
rates in carbonate=iron formation successions. Precambrian
Res., in press.
Weber-Diefenbach, K., 1977. Geochemistry and diagenesis of
Recent heavy metal ore deposits at the Atlantis-II-deep (Red
Sea). In: Klemm, D.D., Schneider, H.-J. (Eds.), Time- and
Strata-Bound Ore Deposits (1st ed.), Springer, Berlin, pp.
419–436.
Williams, I.S., Compston, W., Black, L.P., Ireland, T.R., Foster,
J.J., 1984. Unsupported radiogenic Pb in zircon: a case of
anomalously high Pb–Pb, U–Pb and Th–Pb ages. Contrib.
Mineral. Petrol. 88, 322–327.
Winter, H. de la R., 1987. A cratonic foreland model for Wit-
watersrand Basin development in a continental back-arc plate-
tectonic setting. S. Afr. J. Geol. 90, 409–427.
