New isotope data from a neoproterozoic porphyritic garnitoid-charnockite suite from Natal, South Africa

Precam&qan Research, 62 (1993) 83-101

83

E sevier Science Publishers B.V., Amsterdam

New isotope data from a Neoproterozoic porphyritic granitoidcharnockite suite from Natal, South Africa R.J. Thomas a, B.M. Eglington b, S.A. Bowring c, E.A. R e t i e f ° a n d

F. W a l r a v e n d

aGeological Survey, P.O. Box 900, Pietermaritzburg, 3200, South Africa bEMATEK, CSIR, P. 0. Box 395, Pretoria, 0001, South Africa CDepartment of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA dIsotope Laboratory, Geological Survey, P. Bag X112, Pretoria, O001, South Africa (Received April 2, 1992; accepted after revision August 24, 1992 )

ABSTRACT Thomas, R.J., Eglington, B.M., Retief, E.A. and Walraven, F., 1993. New isotope data from a Neoproterozoic porphyritic granitoid-charnockite suite from Natal, South Africa. Precambrian Res., 62: 83-101. The tectonic setting and age of emplacement of the Oribi Gorge Suite, voluminous rapakivi-textured granites and charnockites ("megacrystic granites") from the ~ 1 Ga (Kibaran) Natal Metamorphic Province has been controversial. Isotopic dates of ~ 1.0 Ga have been obtained from five plutons (various isotopic systems ), but two of these (Fafa and Oribi Gorge plutons) have also given ~ 0.89 Ga Rb-Sr whole-rock isochrons. The age problem has been exacerbated by apparently equivocal field and structural relationships, in which both syn- and post-tectonic settings have been argued. New U Pb isotopic analyses for zircon fractions from a sample of the Fafa pluton give a concordia date of 1029 +_15 Ma. Single zircon evaporation analyses of selected grains from the Oribi Gorge pluton (Bomela locality) give a data of 1068 + 2 Ma. Rb-Sr model dates of biotite separates from three plutons range between 970 and 880 Ma, generally similar to the Rb-Sr whole-rock dates. It is significant that no Pan-African ( ~ 500 Ma ) dates were obtained. The zircon data strongly support a > 1.0 Ga age for the emplacement of the Oribi Gorge Suite, an age similar to both the main regional tectogenesis and syntectonic granites of the Margate Suite. The Oribi Gorge Suite can thus be regarded as late (syn)tectonic and not partly post-tectonic as previously envisaged. Consequently, the need to invoke two periods of identical A-type magmatism is removed. The younger Rb-Sr whole-rock and biotite dates may represent the time when the various plutons or the entire terrane cooled through the relevant blocking temperatures, though this was not a simple, homogeneous process.

Introduction - - the "megacrystic granite" problem The correlation, age and tectonic setting of the voluminous, "megacrystic granites" and charnockites (both sensu lato ) of the ~ 1.0 Ga Natal Metamorphic Province in central and southern Natal has proven controversial (e.g. Du Toit, 1946; Cain, 1975; Kerr, 1985; Evans et al., 1987; Eglington et al., 1986). This has been due to poor outcrop, deep weathering and Correspondence to: Dr. R.J. Thomas, Geological Survey, P.O. Box 900, Pietermaritzburg, 3200, South Africa.

0301-9268/93/$06.00

the extremely variable field aspect of the rocks, which range from almost black, foliated enderbites, through dark-green, fayalite-bearing charnockites to pink and grey, biotite + hornblende-bearing rapakivi-textured granites and gneisses. In the absence of regional geological maps this diversity precluded any obvious correlation between outcrops and led to a proliferation of conflicting nomenclature. The diversity is further reflected in the observed fabrics such that, whereas some rocks are totally unfoliated, there is a continuous gradation to augen gneisses through to mylonites. As a consequence, both post- and syntectonic set-

© 1993 Elsevier Science Publishers B.V. All rights reserved.

84 tings have been argued for various outcrops and an Archaean age has even been suggested (Cain, 1975). However, Thomas (1988a,b, 1989, 1991 ), in the first regional syntheses, noted that all the major "megacrystic granite" outcrops in Natal exhibit the same spectrum of lithological, geochemical and structural features. He accordingly grouped all the very coarse grained, foliated and unfoliated megacrystic granites and charnockites south of the Lilani-Matigulu Shear Zone into a single Oribi Gorge Suite (Fig. 1 ). The suite, comprising ten major plutons, is restricted to (and comprises ~ 40% of) the Margate and Mzumbe terranes of the Natal Metamorphic Province (Thomas, 1989, 1991 ). The only unequivocal

R.J. THOMAS ET AL.

field relationships show that the Oribi Gorge Suite is younger than syntectonic granites of the Margate Suite which have given Rb-Sr wholerock dates of ~ 1050 Ma (Eglington et al., 1986; Thomas et al., 1990). However, the major stumbling block to the general acceptance of this unified Oribi Gorge Suite has been the inconsistency of published isotopic data (Fig. 2; Table 1 ). Dates of ~ 1.0 Ga have been obtained, by various isotopic systems, from the five plutons for which data are available (Mgeni, Fafa, Mvenyane, Oribi Gorge and Port Edward plutons). However, two intrusions (the Fafa pluton and the Bomela locality from within the Oribi Gorge pluton) have also given younger ( ~ 0.89 Ga ) Rb-

20 km I ~

Cover rocks

O

UND 215 (Zircon evaporation)

O

UNO 189, 221, B1, B9 (Rb:Sr b~Stitel

~]

Mzumbe Terrane

~

Shear Zones

Oribi Gorge Suite

Fig. 1. Simplifiedgeologicalmap of the Natal Metamorphic Province (modified after Thomas, 1989), showing the distribution ofthe Oribi Gorge Suite, with the ten, informallynamed, plutons.

Margate Terrane

Fig. 2. Map showing the distribution of previously isotopic data for four plutons of the Oribi Gorge Suite and location of samples from the present study.

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

85

TABLE 1 S u m m a r y o f all previously published isotope data from the Oribi Gorge Suite Pluton

Isotope system

Age ( M a )

Ro

n

MSWD

Ref.

Mgeni Mgeni Mgeni Mgeni Mgeni a Mgeni b Mgeni a Fafa a Fafa a Mvenyane a Oribi Gorge b Oribi Gorge c Oribi Gorge c Port Edward b Port Edward b

R b - S r (biotite) Rb-Sr U - P b (sphene, apatite ) U - P b (zircon) Rb-Sr Rb-Sr Rb-Sr U - P b (zircon) Rb-Sr Rb-Sr Rb-Sr Rb-Sr U - P b (zircon) R b - S r (biotite)

940 1111 940 ± 7 1030 ± 20 1128 ± 38 1001 ± 35 1029 ± 57 999 _+20 878 ± 22 992 :+ 39 891 ± 56 1003 :+ 29 1037 ± 18 990

0.7077 0.7053 0.7087 0.7056 0.7091 0.7054 -

7 13 9 18 7 5 7 5 6 -

0.25 1.7 18.9 2.22 0.52 1,82 0.75 0,01 -

Id 1° 2 2 2°,3 3 3 4 5 6 7 6 6 1d

10

1.72

Rb-Sr

987 +_ 19

_+ 12 _+8 ± ± ± ±

8 6 11 8

0.7045 ± 1

7

References: 1 = N i c o l a y s e n and Burger (1965); 2 = B u r g e r and Walraven (1980); 3 = E g l i n g t o n et al. (1989); 4 = B u r g e r and Coertze ( 1976 ); 5 = Eglington and Kerr ( 1989 ); 6 = T h o m a s ( 1988a ); and 7 = Eglington et al. ( 1986 ). n = n u m b e r of samples. aRapakivi granite and non-charnockite. bCharnockite. cCharnockite and non-charnockite. dRecalculated from original references using decay constant of Steiger and J~iger ( 1977 ).

Sr whole-rock isochrons (Eglington et al., 1986; Eglington and Kerr, 1989). The two distinct sets of ages given by the isotope data appeared to be supported by the fabrics observed at the various sample sites. The lack of a regional fabric at the sample site in the core of the Fafa pluton (and the alleged lack thereof at Bomela) led Eglington et al. (1986), Evans et al. (1987) and Eglington and Kerr (1989) to interpret the ~0.89 Ga Rb-Sr isochrons obtained from these localities as young, post-tectonic, intrusive dates. Conversely, the outcrops which provided the older ~ 1 Ga isochrons tended to be strongly foliated (e.g. the Port Edward pluton ), supporting a syntectonic setting for these bodies (Eglington et al., 1986 ). Despite the problem of ages however, there was general consensus that in petrogenetic terms, the "megacrystic granites" represent relatively high temperature, H20-poor melts with distinctive A-type, "within-plate" granite

characteristics (e.g. Eglington et al., 1986; Thomas, 1988a). The general characteristics of the Oribi Gorge Suite are set out in Table 2. Nevertheless, it proved impossible to reconcile the two distinct isotopic ages, with the field relationships and the geochemistry, which suggested that the rocks form part of a single, possibly coeval, igneous suite (Thomas, 1988a). If the ~ 1.0 and ~ 0.89 Ga isotopic dates both represent crystallisation ages (Eglington et al., 1986 ), there must have been two discrete magmatic events which each generated coarsegrained rapakivi-textured granites and charnokites and two major co-axial orogenic events which deformed them. This, then is the central problem. On the one hand, Eglington et al., (1986) and Eglington and Kerr (1989) invoked two periods of A-type granite emplacement (one at ~ 1.0 Ga, the other at 0.89 Ga), with apparent support from the field relationships. On the other hand, the regional map-

86

R,J. THOMAS ET AL,

TABLE 2 Petrographic and geochemical characteristics of the Oribi Gorge Suite Rapakivi-textured granite

Porphyritic charnockite

Pink or grey, porphyritic Quartz+K-feldspar+ plagioclase+ hornblende+ biotite+ accesory opaque, apatite, zircon

Dark green or brownish green, black, porphyritic Quartz+ K-feldspar+ plagioclase+ orthopyroxene~ clinopyroxene + fayalite _ garnet _+ accessory opaque ( including sulphides ), apatite, zircon, graphite

Geochemistry: Tholeiitic trend on AFM diagram; plots within A-type granite field of Whalen et al. ( 1987); WPG field of Pearce et al. (1984); F e O / ( F e O + M g O ) = ~0.9; high KEO/Na:O, and ( K 2 0 + N a 2 0 ) / C a O , elevated FeO, Zr, Nb, Y, Ba. St, typically ~0.705.

ping of Thomas ( 1988, 1991 ) suggested that all the A-type porphyritic granitoids and charnockites of the Mzumbe and Margate terranes were the result of a single magmatic event at ,-- 1.0 Ga. Following this brief review of the Natal "megacrystic granite" problem, the aim of this communication is to present new U - P b and Pb-Pb (zircon) and Rb-Sr (biotite) isotopic data for certain crucial rocks, in an attempt to solve the enigma presented by the Oribi Gorge Suite. In the light of the new data a revised, unified interpretation of the age and petrogenetic setting of the suite is proposed and the implications for Proterozoic crustal evolution in Natal are discussed. In a wider context, the cause of the variation isotopic ages obtained and the Rb-Sr isotope systematics of the rocks are critically evaluated. From a more global point of view, voluminous rapakivi granites and megacrystic charnockites are characteristic features of many Meso- to Neolaroterozoic mobile belts of the Northern Hemisphere (e.g. Vorma, 1971; Emslie, 1978; Anderson, 1981; Harrison et al., 1990; Emslie, 1991; Vaasjoki et al., 1991). Coarse-grained Kibaran-aged rapakivi-textured granites and charuockites have similarly been recorded from the probable continuation of the Natal Metamorphic Province in Namaqualand (Jackson, 1979; Andreoli and Hart, 1990) and Antarctica (Arndt et al., 1991 ). In view of the volumetric importance of these granitoids and their similarity to typical Pro-

terozoic granite-charnockite suites worldwide, it is clearly important to establish whether the Natal rocks do indeed form a single coeval suite and, if so, when it was emplaced.

The fabric problem One of the main areas of discussion about the Oribi Gorge Suite, relating to its age relative to tectonic events, has centred around the fabrics seen in the rocks. Most plutons have a strong tectonic foliation, particularly at the margins (including the Bomela locality, where a pervasive regional fabric is developed; cf. Eglington et al., 1986). In Oribi Gorge however, near to the core of the Oribi Gorge pluton, there is no tectonic foliation, but a subparallel alignment of undeformed, tabular, feldspar megacrysts is developed which Thomas (1988a) interpreted as an igneous fabric. Similarly, the rapakivi-textured granites exposed in the Fafa and Mtwalume rivers, close to the core of the Fafa pluton, are totally unfoliated (Fig. 3a). This led Evans et al. (1987 ) to conclude that the Fafa granite represents a late intrusion which postdated the main regional tectogenesis. However, according to Thomas (1988a), other parts of the Fafa pluton, particularly away from the core areas, exhibit a strong regional tectonic fabric which is locally mylonitic and folded. In general, the plutons have a strong, but variable, fabric which parallels their margins. The plutons themselves are typically ellipsoidal in shape,

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

87

Fig. 3. Fabrics and textures in the Oribi Gorge Suite (Fafa pluton). (a) Undeformed rapakivi-textured granite from near the core of the pluton (Manono River). (b) Mylonite in a sinistral transcurrent shear zone [Manono River, same locality as (a) ]. with long axes oriented west-southwest, subparallel to the fabric in the enveloping gneisses (Fig. 1). T h o m a s (1988a), in attempting to reconcile the great variation in development o f fabrics seen within the plutons, appealed to similar features reported from Proterozoic rapakivi-textured granitoid plutons o f the north Atlantic region. These have been interpreted as

mushroom-shaped bodies, with u n d e f o r m e d cores and highly deformed marginal flukes (e.g. Bridgewater et al., 1974; Wikstrrm, 1984). More recently, however, such fabric variations in large rapakivi-type granitoid plutons have been ascribed to processes o f syn-intrusion extensional shear zones (e.g. Hutton, et al., 1990).

88

In the Oribi Gorge Suite, the fabric problem has been compounded by a widespread ductile sinistral transeurrent shearing event which produced extensive, WSW-trending, sub-vertical augen gneiss belts and mylonites (Fig. 3b). This transcurrent deformation is considered to be a late manifestation of the main N E SW-directed accretionary orogenic phase in Natal (Jacobs et al., 1993). It must therefore be emphasised that the variable intensity of tectonic fabrics developed within the relatively competent plutons of the suite can be ascribed to syn-plutonic extensional strain which was heterogeneously distributed, giving rise to discrete high- and low-strain zones.

Isotopic constraints The problems with the interpretation of the dates which were obtained from the Oribi Gorge Suite arose because a considerable quantity of isotopic work was done before regional geological mapping was completed. Thus, it was impossible to place the locations from which isotopic dates were procured into their regional context. Isotopic dates from five of the ten major bodies of the Oribi Gorge Suite have been published. The majority of these have been determined by the Rb-Sr method (mainly at the CSIR, Pretoria), with a small number of U - P b zircon analyses (CSIR and Univ. Witwatersrand). The available data are summarised in Table 1 and Fig. 2. Clearly, the majority of the dates cluster around ~ 1.0 Ga, but two Rb-Sr whole-rock isochrons, from the Bomela charnockite locality and the Fafa granite, are significantly younger at 891 + 56 and 878 +-22 Ma, respectively (Eglington et al., 1986; Eglington and Kerr, 1989 ). In interpreting these young dates, Eglington and Kerr (1989) considered that these two bodies represented granitoid intrusions which postdated the main regional tectonogenesis, the age of which had been constrained to ~ 1.0 Ga (Eglington et al., 1986; Eglington and Kerr, 1989). However, as

R.J. THOMAS ET A L

pointed out in Thomas (1988a), a U-Pb zircon model age of 999___20 Ma has been reported from granite at Mpongozini (Burger and Coertze, 1976) which regional mapping showed to be part of the Fafa pluton (Thomas, 1988a). In addition, Thomas (1988b) demonstrated that the "Bomela charnockite" is but a single locality within the large Oribi Gorge pluton, which has provided Rb-Sr t wholerock), and U - P b (zircon concordia) dates of 1003+29 and 1037___18 Ma, respectively ( Thomas, 1988a). Clearly a critical re-evaluation of the isotope data is necessary. Fortunately, most of the analyses have been made at the same laboratory, generally eliminating any question of inter-laboratory bias. The relative wealth of isotope data from the Mgeni pluton was evaluated by Eglington et aL (1989b), who concluded that the pluton was emplaced syntectonically 1030 +- 20 Ma. The dates from the Mvenyane and Port Edward plutons are based on single, reasonably precise, Rb-Sr whole-rock isochrons. Samples taken from a 5 km 2 area within the Mvenyane pluton provided a good 5-point isochron of 992 +_39 Ma, with a wide range of Rb/Sr ratios (4-12). Similarly, the Port Edward date is based on a statistically good, ten-point isochron of 987+__19 Ma, with a considerable spread of Rb/Sr ratios (0.3-1.4). The Port Edward date is supported by an identical RbSr biotite date of 990 Ma (Nicolaysen and Burger, 1965 ), for which the original analytical data is regrettably unavailable. The Oribi Gorge pluton date of Thomas (1988a) is based on samples taken from a 80 km 2 area within the pluton (excluding the Bomela locality). These samples gave an isochron of 1003_ 19 Ma with a large spread of Rb/Sr ratios ( 1-4 ). The Rb-Sr date was supported by a statistically good U - P b zircon concordia date of 1037 + _ 18 Ma, from the same samples as the Rb-Sr data. The young date obtained by Eglington et al. (1986 ), from samples collected at Bomela quarry, near to the faulted southeastern margin of the Oribi Gorge

NEW ISOTOPEDATAFROM A NEOPROTEROZOICPORPHYRITIC SUITE

pluton, was based on a 7-point isochron of 891 +_56 Ma, but with a very small spread of R b / S r ratios (1.1-1.6). Eglington and Kerr (1989) obtained the young Fafa pluton date from a 7-point Rb-Sr isochron (878 + 56 Ma), with a good range of R b / S r ratios (2-5.7), although one sample had a much higher R b / S r ratio than the others. Burger and Coertze (1976) procured a 999 + 20 Ma model 2°7pb/ 2°6pb date (recalculated) for zircon from Mpongozini, a locality within the Fafa pluton. This date, made from the bulk zircon population is ~ 20% discordant and, as such, is probably an underestimate of the true age of the zircon.

New isotope data Clearly the controversy concerning the validity and age of the Oribi Gorge Suite could best be ascertained by testing the same samples which gave the young Rb-Sr isochrons by other techniques. It was therefore decided to obtain U - P b data from small, hand-picked zircon populations from one of the original Fafa pluton samples, along with P b - P b data from individual zircon grains from Bomela. In addition, Rb-Sr analyses were performed on biotite separates from five localities within three plutons (Fafa, Oribi, George and Mvenyane).

89

U-Pb isotope resultsfor the Fafa pluton In order to test the Rb-Sr isochron date of 878 + 22 Ma obtained by Eglington and Kerr ( 1989 ) for the Fafa pluton, one sample ( U N D 199) was selected for U - P b zircon isotopic analysis. Zircons were extracted from the 150250 a m and < 150 a m heavy mineral fractions using standard techniques and hand-picked. Morphological descriptions are provided in the Appendix. The zircons were prepared at M.I.T. using the Parrish (1987) method and spiked with a mixed 2°8pb/235U tracer. Isotopic measurements were performed on a VG 354 mass spectrometer; procedural blanks ranged from 25 to 50 pg total Pb and from 10 to 20 pg U. The data were reduced following Ludwig (1988). The isotope data (Table 3) was regressed using GEODATE version 2.2 (Eglington and Harmer, 1991 ), with the analytical precisions shown and the decay constant recommended by Steiger and J~iger (1977). Isochron-errorchron systematics were assessed on the basis of F parameters (5% level of significance) assuming an infinite number of replicates (cf. Brooks et al., 1972; Harmer and Eglington, 1990). The zircon fractions provide a concordia intercept date of 1029+- ~1oMa, with a lower intercept at 42+_6o Ma (Fig. 4).

TABLE 3 U - P b zircon isotopic data for U N D 199 from the Fafa pluton Fraction

Sample wt (mg)

U (ppm)

Pb (ppm)

2°6pb 2°4pb

2°Tpb 2°6pb

2°6pB 238U

1o% error

2°7pb 23~U

1o% error

CC

IA 1B lCaa 1D

0.116 0.134 0.008 0.022

384.1 313.3 397.4 421.5

53.1 52.8 34.1 59.6

3353 362 196 678

0.073277 0.073347 0.072557 0.073438

0.13982 0.14662 0.08499 0.14424

0.1575 0.1610 0.4400 0.2370

1.4127 1.4828 0.8503 1.4606

0.1660 0.3495 1.095 0.3510

0.994 0.869 0.674 0.848

Atomic ratios corrected for fractionation, blank and common lead; common lead correction based on the model of Stacey and Kramers (1975). aa=Air abraded fraction (1C), after Krogh (1982). CC = Error correlation coefficient. Note that morphological descriptions of the zircons with illustrations are provided in the Appendix.

90

R.J. THOMASET AL

°° .a g,ao,,o o.o ,991

i;]

I

//• 0.15

-

~OlO-

005-

/ ~ t e r c e p t = 71¢/j//

Lower intercept =

~/~ 06O

oo

1029 42

+11 ~60

/-10 Ma !

-61 Ma

MSWD = 037 !

02 0', 0'6 0'6

,!o

i

,,2 ,,

,'6 ,16

2orPblZS~U

Fig. 4. Concordia diagram for zircon separates from U N D 199 from the Fafa pluton.

Single-grain zircon evaporation results (Bomela locality) To follow up the "Bomela charnockite" dispute and the young 891 + 56 Ma Rb-Sr date, zircons were separated from sample U N D 215 from the Bomela Halt locality in the Oribi Gorge pluton, and analysed by means of the single-zircon evaporation technique at the Geological Survey Laboratories, South Africa. Six hand-picked zircon grains were selected for analysis--four in which cores and overgrowths

could be recognized (grains 1 to 4) and two consisting entirely of the clear phase (grains 5 and 6) (see Appendix). Pb-isotopic measurements were made using a Finnigan MAT 261 thermal ionisation mass spectrometer, following Kober ( 1986, 1987 ). In order to minimise common Pb corrections, only data with 2°4pb/ 2°6pb ratios smaller than 0.0001 were collected, and a common Pb correction was made using the Stacey and K_ramers (1975 ) Pb-isotopic composition corresponding to the age indicated by the 2°7pb/2°6pb ratio. Corrections are made for non-linearity of the secondary electron multiplier. Age, error and weighted mean were calculated using GEODATE version 2.2 (Eglington and Harmer, 1991 ). All errors are reported using 95% confidence limits, assuming an infinite number of replicates. The analytical data are given in Table 4 and Fig. 5. Relatively invariant z°7pb/2°6pb ratios were found for grain 1 over the entire temperature range analysed, indicating a date of 1068 + 5 Ma. However, an almost two-fold increase in the 2°8pb/2°6pb ratio is apparent over temperature range 1050-1080°C. This confirms that Pb was emitted from different zones in the zircon which, although having indistinguishable dates, have different U / T h ratios. A

TABLE 4 Single-grain zircon evaporation data for the Oribi Gorge pluton (Bomela Halt locality )--sample UND 215 Grain

Temperature ( °C )

Number of ratios

zO7pb/ 206pb

208Pb / 2°6Pb

1

1050 1060 1070 1080 1040 1050 1060 1070 1050 1070 1050 1070 1080 1100

40 45 18 19 86 86 45 91 17 8 25 5! 37 72

0.07495 _+0.00022 0.07491 _+0.00013 0.07513+0.00014 0.07432 _+0.00053 0.07274 _+0.00005 0.07362 _+0.00008 0.07461 _+0.00008 0.07484 _+0.00007 0.07424 + 0.00020 0.07472 + 0.00042 0.07397 + 0.00016 0.07524 + 0.00007 0.07522 + 0.00013 0.07516 + 0.00007

0.0544 _+0.0005 0.0658 _+0:0003 0.0821 +_0.0017 0.0833 _+0.0005 0.0337 + 0.0002 0.0368 + 0.0009 0.0391 +_0.0013 0.0418 + 0.0002 0.0331 + 0.0004 0.0605 + 0.0004 0.0487 _+0.0001 0.0438 __+0.0010 0.0568 _+0.0003 0.0479 + 0.0024

2

3 4 5 6

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

10

05 £3

a~l

J3 .Q 0 O_

0 0 072

0.073

, i i i , 0074

0075

0076

1.0 bean = 007499 _+87 (95% coof) Mean datbas;g68-+ 2 (95% c°n' ) t

rr

/

~ \

Uk

05

0 0.072

0.073 207

0.074 206

Pb/

0075

0076

Pb

Fig. 5. Results of single-grain zircon evaporation dating of UND 215 from the Oribi Gorge pluton (Bomela Halt locality). (a) Results from all analysed zircon grains at all temperature steps, (b) Results excluding grains 3 and 4 and the lower-temperature data from grain 2. 1200 m / / ~

1000

J

/

~

//~ Oribi Gorge / pluton

800

~- 600

400 Port Edward pluton 200

0 1.0

1.4

1.8

2.2

21.6

3.0

Watson and Harrison M

Fig. 6. Plot of Watson and Harrison (1984) M value against Zr, showing isotherms for temperatures in the range ~ 750-900 ° C for several plutons of the Oribi Gorge Suite. (M=cationic [ N a + K + 2 Ca]/[Si.A1] ).

range of Pb-isotopic compositions was emitted from grain 2 at temperatures between 1040 and 1070°C, corresponding to dates from 1003 _+6 to 1076 _+6 Ma. Since only two growth zones were recognized in this grain, the range of dates

91

is interpreted to reflect varying admixtures of Pb from each zone. One or both of the zones may be discordant and Pb-loss must have taken place at some stage in the geological history of this grain. Grains 3 and 4, both slightly smaller than the other grains, emitted only limited quantities of Pb at a relatively low temperature of 1050°C. The dates indicated are 1050_+25 and 1041_+23 Ma, respectively. These are lower than the high-temperature dates obtained from grains 1 and 2, although that of grain 3 is not statistically different at the 95% confidence limit. Nevertheless, these grains may have lost Pb at some time in the geological past and are probably moderately discordant. Grains 5 and 6 emitted little or no Pb at temperatures below 1070 ° C and at higher temperatures the Pb was found to be isotopically homogeneous. Across the whole temperature range--1070°C to 1100°C - - the emitted Pb indicated dates of 1075 _+3 and 1073 _+5 Ma for grains 5 and 6, respectively. The dates yielded by grains 5 and 6 is within statistical confidence limits of the dates indicated by the high-temperature components of grains 1 and 2. Combination of these data give a date of 1068 _+2 Ma (Fig. 5 ). From the morphological information (see Appendix) it would appear that the overgrowths on grains 1 and 2 and the entire grains 5 and 6 belong to the same growth phase of the U N D 215 zircon population. The internal consistency of the dates from grains 5 and 6 and the overgrowths on grains 1 and 2 indicates negligible Pb-loss from these domains during their geological history and argues against a xenocrystic origin. As a result, the dates can be taken to reflect the crystallization age of the zircons and the Oribi Gorge pluton. Chemical data for the various plutons comprising the Oribi Gorge Suite provide additional evidence that the cores evident in certain zircon grains are unlikely to be derived from the source region of the magmas. In all cases the M(M= cation ration [ Na + K + 2Ca ] / [Si.A1]) factor of Watson and Harrison

92

R.J THOMASET AL.

TABLE 5 Rb-Sr isotope results for five biotite separates from three plutons of the Oribi Gorge Suite, and one biotite separate from the Sezela pluton ( U N D 221 ) Sample

Rb

Sr

87Rb/S6Sr

t 6(%)

87Sr/86Sr

I 0(% )

B1 UND199 UND189 B9 UND215 UND221

770.4 449.5 837.3 728.4 740.4 283.2

7.46 28.14 10.26 8.18 13.09 32.59

473.2 49.16 342.9 395.7 208.3 25.95

1.0 0.8 0.8 1.0 0.8 0.8

6.673 t.3601 5.3298 6.186 3.4948 1.0253

0.05 0.02 0.02 0.05 0.02 0.02

TABLE 6 Summary of available U - P b and Pb-Pb zircon and Rb--Sr isotopic data for five plutons of the Oribi Gorge Suite, and the Sezela pluton. The Rb-Sr biotite (bi) data are regressed with the corresponding whole-rock data (resulting errorchrons are denoted by an E). Rb--Sr biotite/whole-rock sample pair model ages are also given Pluton

Po~ Edward Ofibi Gorge

Mvenyane Fafa

Samples

Rb-Sr whole rock

Rb-Sr whole rock and biotite

Date

Ri

Date

0.70453_+ 13

_+9905

987-+ 192 BI, B2, B3, B4, B5, B6 BI,B3,B4,B5,B6 (B1 bi) B1,B1 bi UND189, 190, 191, 215,216,232,233 UND215 B7, B8, B9, Bt0, B11, (B9 bi) U N D 199, 200, 201, 202, 203, 204, 205 ( U N D 199 bi) U N D 199, U N D 199 bi

Mgeni

Sezela

U - P b zircon

UND 131,211,212, 213,214,217,218,219, 220,221,222 UND221, UND221 bi

R~

1038_+18' 1003_+29 ~

0.7054_+9

0.7075± 18E 0.7104+ 1~ 0 7076 ± 2

891_+ 562

0.7091± II

924_+49 882± 18 938_+ l0

991_+39 ~

0.7057_+31

973+17

0.707 ~ 17

11 1029_+]-6

878-+223

0.7087±8

915-+ 19

0.7074 J: 7E

1030_+204 940 -+ 74 (apatite)

1001 _+354

1068 _+2 (Pb-Pb)

952± 163

0.7077-+ 124

0.7032 ±1

925-+ 15 ~9405

0 7073 L 7

921 ± 19

0.7033 ~C2E

895__+ 14

0.7033± 2

Data sources: t=Thomas (1988a); 2=Eglington et al. (1986); 3=Eglington and Kerr (1989); 4=Eglington et al. (1989bh 5= Nicolaysen and Burger ( 1965 ).

(1984) is high ( > 1 ) and consistent with complete solubility of zircon in the melts at rnagmatic temperatures (Fig. 6). The varying morphology of zircon grains in the various plutons is thus probably related to different periods of crystallization during cooling of the melt.

R b - S r isotope data on biotite separates

For the present study, biotite separates were extracted from five samples from three plutons of the Oribi Gorge Suite, for which isotope data were previously available (Oribi Gorge, Mvenyane and Fafa pluton--see Table

93

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

1 ). One biotite from the late-tectonic Sezela quartz monzonite pluton was analysed for comparison, whilst Nicolaysen and Burger ( 1965 ) and Eglington et al. (1989b) have previously provided biotite data from the Port Edward and Mgeni plutons, respectively (Table 1 ). Hand-picked biotite crystals were leached for 10 minutes in an ultrasonic bath using 1M HC1 to remove any possible contaminants. Rb and Sr concentrations and 878r/86Sr measurements were determined by isotope dilution mass spectrometry at the CSIR. All 878r/ 86Sr ratios are normalised to 86Sr/88Sr = 0.1194. The value obtained for strontium standard NBS 987 was 0.71027+_2 (95% confidence). Rb and Sr blanks are less than 1 ng and are thus negligible relative to the concentrations in the samples. The analytical data and precisions are given in Table 5. Error correlation between the 87Rb/86Sr and 87Sr/86Sr errors is assumed to be 0.76, following Brooks et al. (1972). The data were processed using the GEODATE version 2.2 (Harmer and Eglington, 1990; Eglington and Harmer, 1991 ). Errors are calculated on the basis of 60 replicates at the 5% level of significance. Uncertainty limits quoted for dates and initial 875r/86Sr (Sri) are thus for 95% confidence ( = 2or in this case). The decay constant used is that recommended by Steiger and J~iger (1977). All biotite analyses were also regressed with the relevant whole-rock data and these results are shown in Table 6. In cases where errorchrons resulted, the biotite separate data were regressed with the corresponding whole-rock data and specified separately. The Rb-Sr biotite dates obtained are within the range 970880 Ma. The Rb-Sr biotite dates do not vary consistently with the corresponding whole-rock dates. The biotite data from the Fafa pluton (958 + 15 Ma) is older than the equivalent whole-rock date ( 878 + 22 Ma). Similarly, the two biotite dates from Bomela (934 + 15 Ma) are older than the relevant whole-rock date (891 ___56 Ma), but the biotite date of sample B 1 ( 883 + 18 Ma), from elsewhere in the Oribi

Gorge pluton is younger than its corresponding whole-rock date ( 1003 _+29 Ma). The biotite from the Mvenyane pluton (969 + 16 Ma) is similar to the corresponding whole-rock date (992+39 Ma). In all cases the available zircon dates are older than the corresponding RbSr whole-rock and mineral dates. Discussion

Available mineral and whole-rock dates from the Oribi Gorge Suite and other intrusive lithologies in southern Natal are illustrated in Fig. 7. The zircon data presented in this paper demonstrate that the Oribi Gorge Suite was intruded between 1070 and 1030 Ma. This gives strong support to the conclusion of Thomas (1988a,b) that the rocks represent a single porphyritic granite-charnockite suite which was emplaced prior to 1.0 Ga, rather than having formed as a result of two episodes of intrusion at ~ 1.0 and 0.89 Ga, respectively. The minor differences in age emerging from this study suggest that the granite-charnockite suite south of the Melville Shear Zone, i.e., from the Margate Terrane, may be older than those from the Mzumbe Terrane to the north. A similar age relationship has been reported from the granulite and amphibolite terranes of the HeimefrontOella, Western Dronning Maud Land, Antarctica (Arndt et al., 1991 ). The original U - P b zircon date of 1037 ___18 Ma of Thomas (1988a) for the Oribi Gorge pluton was based on the analysis of zircon fractions comprising several hundred grains and consequently may represent a mixing of domains formed during the period from 1070 to 1000 Ma. Clearly, more detailed single-grain zircon work would be required to test whether this apparent range in ages between the two terranes in Natal is real. The dates now available for the Oribi Gorge Suite are similar to the age of the main Kibaran tectonothermal event in Natal (Eglington et al., 1989a), and to that of the synthetic Margate Granite Suite (Eglington et al., 1986; Thomas et al., 1990). In the light off the syn-

94

R.J. T H O M A S E T AL.

End of cooling ~ metamorphism

SW - NE shearing ?

Oribi Gorge Suite intrusion

:o ,~mon Legend Rb~Srwt~ole*rock P~pS~blot~ U-Pbapetite

J -

' ~o

i

!

i

N,~o~,~p~. Uad01e Delta

oetarsheet

----

!1

,

!

[

i

¢"

_

i

i

~r-'~ :: :

J-r

e

Mzumbe

=

2]

............

880

:

900

g20

940

..... :

Port E~

g60 980 D a t e (Ma)

ard

1000

'.. . . . . . . . . . . . . . . . . . . .

1020

1040

1060

Lgi~

1080

Fig. 7. Isotopic dates obtained from the Oribi Gorge Suite compared with those from other intrusive units and major tectono-metamorphicevents in the Mzumbeand Margate terranes.

to late-tectonic setting now proposed for the Oribi Gorge Suite, it is clear that these plutons are not rapakivi granites (sensu stricto ), as this term is generally restricted in use for post-tectonic granitoids. Similarly, the new constraints on the age of formation of the Oribi Gorge Suite requires a slight modification of the interpretation of the S m - N d data presented by Eglington et al. (1989a). Whilst there is no change to the given Nd/TDM dates, the initial end values require recalculation for the Fafa and Oribi Gorge (Bomela locality) localities because these are now known to be 170-130 Ma older, end values for the entire Oribi Gorge Suite now range from - 0 . 3 to +2.8, with exclusively positive values in the Mzumbe Terrane and both positive and negative values in the Margate Terrane. Whereas field evidence shows unequivocally that the Oribi Gorge Suite postdates the syntectonic granites of the Margate Suite, the isotopic ages of the two suites are statistically similar ( ~ 1070 to 1030 Ma). In view of the difficulty of discriminating between the relative ages of intrusion of the Port Edward en-

derbite (Oribi Gorge Suite) and the Nicholson's Point granite (Margate Suite) in the field, it has been suggested that both granitoids were approximately coeval (Grantham, 1983; Eglington et al., 1986). Furthermore, it has been suggested, that the intrusion of the hot, essentially anhydrous enderbite magmaat Port Edward, was in part responsible for the in-situ charnockitisation of the garnet-biotite granite of the Margate Suite as Nicholson's Point (Grantham, 1983; Eglington et al., 1986 ). This dehydration process is manifest as thin aureoles (up to 2 m wide) of charnockitised Nicholson's Point granite immediately adjacent to the contact with the enderbite. On a larger scale, charnockitic aureoles in Margate garnet leucogranite, up to 2 km wide, are developed around large charnockite bodies such as the Oribi Gorge pluton (Thomas, 1988a). The new isotope data thus support a close temporal relationship between the intrusion of the Oribi Gorge and Margate Granite Suites between 1070 and 1030 Ma. The Rb-Sr biotite data show that the micas reflect a younger event(s) between 970 and

95

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

880 Ma, though this varies from pluton to pluton. This is true, not only for the three plutons of the Oribi Gorge Suite for which data are available, but also for the younger, late-tectonic Sezela quartz monzonite pluton (Table 5). This intimates that the younger isotopic disturbance is not confined to the Oribi Gorge Suite alone, but represents a more widespread, regional event. From the field relationships, it is known that the youngest igneous rocks of the southern part of the Natal Metamorphic Province are certain post-tectonic microgranite dykes from the Margate area which have given a Rb-Sr date of ~ 9 5 0 Ma (Thomas et al., 1990) and a U - P b zircon date of 1026 + 3 Ma (Thomas et al., 1993 ). What then is the significance of the 970-880 Ma Rb-Sr whole-rock and biotite dates, in terms of the tectonothermal evolution of the Natal Metamorphic Province? This young event was quite widespread throughout southern Natal, as not only have rocks of the Oribi Gorge Suite from two widely spaced localities given young Rb-Sr whole-rock dates (Fafa and Bomela), but a coarse-grained granite sheet from Marble Delta has given a similar Rb-Sr whole-rock date of 899 + 11 Ma (Naidoo et al., 1989). This granite has a penetrative foliation, so this date also probably represents post-crystallization resetting. The young whole-rock and biotite dates could be interpreted as the time at which the plutons cooled and/or the entire terrane cooled through the Rb-Sr biotite and K-feldspar blocking temperatures ( ~ 300 and 350°C, respectively; Dodson, 1979). However, the considerable range of dates obtained suggests that this was not a simple, homogeneous cooling process and that the entire terrane may have been affected by localised post-tectonic thermal events at different times. If the latter is true, then this thermal event has left little or no petrographic evidence, though some chloritisation of biotite and/or kaolinisation of feldspar is locally apparent. The late transcurrent shearing event, noted earlier, might have resulted in localized open-

system isotopic behaviour and could be the cause of at least some of the resetting. However, in a recent detailed study, Bingen et al. (1990) address the problem of discrepant U Pb, Rb-Sr whole-rock and Rb-Sr mineral dates from the ~ 1.0 Ga orogenic belt of southern Norway. This area is remarkably similar to southern Natal in terms of lithologies, tectonomagmatic environments and isotopic problems. Bingen et al. (1990) demonstrated that the Rb-Sr system remained open at a mineralogical scale (particularly with regard to Kfeldspar) until at least 170 Ma after intrusion of the granitic plutons. It is possible that the very high K-feldspar content of the Natal samples has facilitated this resetting, since diffusion of Sr in K-feldspar is known to be relatively fast (Misra and Venkatsubramanian, 1977 ) and a diffusion distance of ~ 30 m m in 10 Ma at 600°C has been estimated (Bingen et al., 1990). It is significant that no Pan-African ages are reflected in the biotite separates. This contrasts with the Kibaran-aged granitoids of the eastern continuation of the Natal belt in the Falklands Plateau and Dronning Maud Land, Antarctica which consistently give Kibaran ( ~ 1.0 Ga) whole-rock dates, but Pan African ( ~ 500 Ma) mica dates (e.g. Beckinsale et al., 1977; Jacobs, 1991 ). This gives strong support for the conclusions of Thomas et al. (1990, 1993) that there was no Pan-African magmatic activity or thermal reworking in southern Natal, unlike the adjacent crust of Gondwana to the east. Conclusions

In the light of the above review of the problems, the suggested structural re-interpretation and the new isotope data, a revised model for the petrogenetic setting and age of emplacement of the Oribi Gorge Suite is proposed: ( 1 ) The Oribi Gorge Suite is a voluminous, rapakivi-textured granite-charnockite suite. The suite encompasses ten large intrusions

96

which together make up ~ 40% o f the surface areas o f the Margate and M z u m b e Terranes o f the Neoproterozoic Natal Metamorphic Province. (2) The suite was emplaced at a late-(syn ) tectonic stage o f the m a i n regional Kibaran tectogenesis, immediately after the intrusion o f the tabular, syntectonic granites o f the Margate Suite, and was locally responsible for dehydrating and "charnockitising" the latter. (3) The variable, but locally intense, pervasive foliation seen in the rocks o f the Oribi Gorge Suite (albeit, with local fabric-free, lowstrain zones) indicates that the rocks were emplaced in a syn-to late-kinematic regime and solidified prior to the cessation o f the main regional deformation. Its emplacement m a y be related to a widespread episode o f t r a n s c u r r e n t shearing seen throughout the M z u m b e and Margate terranes, some o f which have locally been observed to have a transtensional character (Jacobs et al., 1993). (4) The A-type, within-plate geochemical signature o f the suite m a y be a function o f the composition o f the lower crustal source region, c o m b i n e d with the local prevailing thermodynamic and fluid phase conditions at the time o f m e l t i n ~ In this regard the use o f discriminant geochemical diagrams for the Oribi Gorge Suite, such as those in Pearce et al. (1984), may be misleading in terms o f tectonic setting. (5) The R b - S r biotite data isotope data, suggest that major tectono-metamorphic events in Natal had ceased by ~ 950 Ma. This is in agreement with published ages o f 1000-950 Ma for demonstrably late, post-tectonic intrusions. The younger whole-rock a n d biotite ages recorded from Natal reflect a relatively widespread period o f open-system behaviour, Furthermore, the R b - S r biotite data confirm recent suggestions that there was no significant thermal reworking in Natal during Pan-African times.

R.J. THOMAS ET AL,

R J T and F W t h a n k the Chief Director, Geological Survey, for permission to publish this article. We are grateful to John Tarney, Johan Kruger and Alfred Kr~iner for constructive critisisms o f the paper.

Appendix--zircon descriptions UND 199 (Fafa pluton) (Fig 8a) Elongate euhedral crystals (length:breadth= ~ 7) predominate over stubby, equant grains. Strongly zoned grains occur together with near-structurelesscrystalscontaining numerous inclusions. / 110} and {11 t } crystal forms predominate, although higher-ordered pyramids may also be developed, with sharp to chisel-shaped terminations. Some grains contain well-developedelongate cores with rounded terminations. These cores are typically metamictised and altered but some translucent and transparent zones and cores are present. The transparent domains exhibit microzonal structures which are mirrored in the adjacent mantle. This feature suggests that the cores are not xenocrystic,but reflect early crystallisation of high-U domains, followedby an hiatus in growth before the development of lower-U domains. Opaque mineral inclusions which are present in both core and mantle, greatly predominate over fluid inclusions. The bulk population was hand-picked to exclude zircons with obvious cores. Zircons from this sample are illustrated in Fig. 8a.

B1 and B4 (Oribi Gorge ptuton)(Fig. 8b) These samples were from the set that provided a concordia age of 1037+ 18 Ma (Thomas, 1988a). Moderate to elongatecrystals (length:breadth= 2-7 ) show a variety of terminations ranging from sharp to multifaceted, and sometimes modified by extension growth. A small number of grains exhibitslightsignsof resorption. Three phases of growth are evident: ( 1) The earliest phase is brown, sub-translucent to intensely metamietised and characteristicallymicrozoned. (2) Althoughthe early phase may be the sole or dominant component of some grains, it is usually mantled by phase 2 which is generallystructureless and transparent. Phase 2 is, however, the overall dominant phase in the two samples. Microzoning in phase 1 may be mirrored. albeit in an attenuated form, in the second phase. The interface between phases 1 and 2 varies from sharp to diffuse and radial fractures may begin at the contact. Opaque mineral inclusionsare common, rangingfrom semi-trans-

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

97

Fig. 8. Photomicrographs of zircon populations described in the Appendix. (a) UND 199 from the Fafa pluton (this study. (b) B4 from the Oribi Gorge Pluton (Thomas, 1988a). The bar scale represents 150/tm. parent to opaque blebs, thread microlites and stubby rods. (3) The very subordinate third phase invariably occurs as bulbous extension growths on phases 1 and 2. It never occurs as discrete grains. Both regular and irregular microfractures characterise this phase. The three phases are thought to have grown in the same crystallisation cycle, separate by only minor hiati. Figure 8b illustrates a selection of zircon grains from sample B4.

UND 215 (Oribi Gorge pluton, Bomela locality) (Fig. 9) This mixed population comprises two, possibly three,

different phases exhibiting a variety of structural interrelationships within the composite grains which they form. Discrete single phase crystals are also present, varying from ~ 100 to 250/~m in width and displaying a wide range of very characteristic morphological features (Fig. 9). The grains range from subhedral to anhedral with stubby types predominating over elongated (L/B ~ 3.5:1 ) and ovoidal grains; round and chisel-shaped terminations are observed. Although the outlines are mostly straight, both positive and negative curvilinear shapes, are also developed. The tract varies from simple prism-pyramid combinations to multifaceted. The bulk of the assemblage comprises a transparent to translucent phase (P2) which, although usually associ-

98

R.J. THOMAS ET AL

99

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

ated in a core-mantle relationships with phase P3, also forms discrete crystals (Fig. 9, # 5 and # 6 ). The core domains are formed by one or more disjointed structural units, which are characteristically microfractured and set in an irregularly and incompletely developed P3 mantle (Fig. 9, # 1-4 and # 7 ) . The interface between these phases appears to be fractured enhanced and therefore typically sharply demarcated, but it may also diffuse. Many P2 domains show faint multiple growth zoning ( M G Z ) and sector zoning. In general, the P3 phase is slightly less transparent than P2; MGZ is not developed. Very few of the microfractures from the P2 cores extend into an enveloping P3 mantle (Fig. 9, # 7). A variety of euhedral and globular opaque inclusions; stubby to elongated microlites and fluid inclusions are characteristics of the P2 domains. A sharp interface between rare, sub-round domains in P2, may indicate a possible P 1 phase, but grains with these features have not been analysed. Despite the obvious structural interrelationship between P2 and P3 domains; the restriction of microfractures and inclusions to P2 domains, and resorbed outlines, all of which suggest two generations of zircon, the age data dictate careful interpretation of these features. The average 1074+_2 Ma age of grains 5 and 6 represents the age of P2 formation. The composite P2-P3 grains 14, however, present equivocal results. Grain 1 is a very good example of a composite grain, yet the constant 2°7pb/ 2°6pb ratio (1068+-5 Ma), suggests that P2 and P3 are equivalent. The two-fold change in 2°SpbF°6pb and hence in U / T h ratio, however, indicates different conditions of formation for the two phases. Although the interface between P2 and P3 domains in grain 2 is sharply demarcated, the phases are optically fairly similar, yet a range in dates from 1003 + 6 to 1076 + 6 Ma is shown. In contrast, the composite grains 3 and 4 both provide single dates of 1051 + 25 and 1041 + 22 Ma, respectively. These data suggest that, except for grain 2, P2 and P3 formed during the same crystallization cycle, albeit separated only by a small hiatus during which some P2 grains were mechanically fractured. The massive P2 grains probably attest to armouring a n d / o r favourable orientation during deformation. More data are required to confirm the 1003+-6 Ma date of P3 in grain 2, and to fully evaluate this younger phase.

References Anderson, J.L., 1981. Proterozoic anorogenic plutonism of North America. Mem. Geol. Soc. Am., 161: 133154. Andreoli, M.A.G. and Hart, R.J., 1990. Metasomatized granulites and eclogites of the Mozambique belt: implications for mantle devolatilization. In: H.K. Herbert and S.E. Ho (Editors), Stable Isotopes and Fluid Processes in Mineralization. Univ. West. Aust. Publ., 23: 121-140. Arndt, N.T., Todt, W., Chauvel, C., Tapfer, M. and Weber, K., 1991. U - P b zircon age and Nd isotopic compositions of granitoids, charnockites and supracrustal rocks from Heimefrontfjella, Antarctica. Geol. Rundsch. 80: 759-777. Beckinsale, R.D., Tarney, J., Darbyshire, D.P.F. and Humm, M.J., 1977. Rb-Sr and K - A r age determinations on samples of the Falkland Plateau basement at site 330 DSDP. Init. Rep. D Sea D P, 36: 923-927. Bingen, B., Demaiffe, D. and Hertogen, J., 1990. Evolution of feldspars at the amphibolite-granulite-facies transition in augen gneisses (SW Norway): geochemistry and Sr isotopes. Contrib. Mineral. Petrol., 105: 275-288. Bridgwater, D., Sutton, J. and Watterson, J., 1974. Crustal downfolding associated with igneous activity. Tectonophysics, 21: 57-77. Brooks, C., Hart, S.R. and Wendt, I., 1972. Realistic use of two-error regression treatments applied to rubidium-strontium data. Rev. Geophys. Space Phys., 10: 551-577. Burger, A.J. and Coertze, F.J., 1976. Summary of age determinations carried out during the period April 1974 to March 1975. Ann. Geol. Surv. S. Afr., 11: 317-321. Burger, A.J. and Walraven, F., 1980. Summary of age determinations carried out during the period April 1978 to March 1979. Ann. Geol. Surv. S. Afr., 14:109-118. Cain, A.C., 1975. A preliminary review of the stratigraphic relationships and distribution of metamorphism in the northern part of the Natal-Namaquarides, South Africa. Geol. Rundschau, 64:192-216. Dodson, M.H., 1979. Theory of cooling ages, 194-202. In: E. J~iger and J.C. Hunziker (Editors), Lectures in Isotope Geology, Springer, Berlin, 329 pp.

Fig. 9. Photomicrographs of the zircon crystals ( # 1-6), analysed by the evaporation technique. The bar scale represents 100 ~tm. # 1 Composite grain comprising a disjointed core domain (P2) with acicular microlites, incompletely mantled by P3. Note the iron oxide-filled circular microfractures. # 2. The internal structure is similar to that of grain # 1. The mantle has been damaged during crushing. # 3. A composite grain; the elongated P2 core, which exhibits faint MGZ and inclusions, is surrounded by a narrow P3 mantle. # 4 The internal structure is similar to that of grain # 3. # 5. A subhedral P2 crystal with irregularly developed microfractures. # 6. An ovoidal P2 grain. # 7. An elongated, MGZ, P2 central domain is transected by iron-oxide-filled, curved microfractures and partly mantled by structureless P3. The P2-P3 interface is fracture enhanced; also note that the microfractures do not extend into the mantle. Not analysed.

100

Du Toit, A.L., 1946. The geology of parts of Pondotand, East Griqualand and Natal. Explanation of Sheet 119, Geol. Surv. S. Afr., 32 pp. Eglington, B.M. and Harmer, R.E., 1991. GEODATE: a program for the processing and regression of isotope data using IBM-compatible microcomputers. CSIR manual EMA-H 9102, 70 pp. Eglington, B.M. and Kerr, A., 1989. Rb-Sr and Pb-Pb geochronology of Proterozoic intrusions from the Scottburgh area of southern Natal. S. Afr. J. Geol., 92: 400-409. Eglington, B.M., Harmer, R.E. and Kerr, A., 1986. Petrographic, Rb-Sr isotope and geochemical characteristics of intrusive granitoids from the Port EdwardPort Sbepstone area, Natal. Trans. Geol. Soc. S. Afr., 86: 199-213. Eglington, B.M., Harmer, R.E. and Kerr, A., 1989a. Isotope and geochemical constraints on Proterozoic crustal evolution in south-eastern Africa. Precambrian Res., 45: 159-174. Eglington, B,M., Harmer, R.E. and Kerr, A., 1989b. RbSr isotopic constraints on the ages of the Mgeni and Nqwadolo Granites, Valley of a Thousand Hills, Natal. S. Afr. J. Geol., 92: 393-399. Emslie, R.F., 1978. Anorthosite massifs, rapakivi granites and late Proterozoic rifting of North America. Precambrian Res., 7: 61-98. Emslie, R.F., 1991. Granitoids of rapakivi granite-anorthosite and related associations. In: I. Haapala and K.C. Condie (Editors), Precambrian Granitoids--Petrogenesis, Geochemistry and Metallogeny. Precambrian Res., 51: 173-192. Evans, M.J., Eglington, Kerr, A. and Saggerson, E.P., 1987. The geology of the Proterozoic rocks around Umzinto, southern Natal, South Africa. S. Air. J. Geol., 90:471 488. Grantham, G.H., 1983. The tectonic, metamorphic and intrusive history of the Natal Mobile Belt between Glenmore and Port Edward, Natal. M.Sc. thesis (unpubl. ), Univ. Natal, Pietermaritzburg. 176 pp. Harmer, R.E. and Eglington, B.M., 1990. A review of the statistical principles of geochronometry: Towards a more consistent approach for reporting geochronological data. S. Aft. J. Geol., 93: 845-856. Harrison, T.N., Brown, P.E., Dempster, T.J., Hutton, D.H.W. and Becker, S.M., 1990. Granite magmatism and extensional tectonics in southern Greenland. Geol. J., 25: 287-293. Hutton, D.H.W., Dempster, T.J., Brown, P.E. and Becket, S.D., 1990. A new mechanism of granite emplacement: intrusion in active extensional shear zones. Nature, 343: 452-455. Jackson, M.P.A., 1979. A major charnockite-granolite province in southwestern Africa. Geology, 7: 22-26. Jacobs, J., 199 I. Structural evolution and cooling history of the Heimefrontfjella Mountains (Western Dron-

R.J. THOMAS ET AL,

ning Maud Land). Ber: Polarforsch. 97, 141 pp. Jacobs, J. Thomas, R.J. and Weber, K., 1993. Accretion and indentation tectonics at the southern edge of the Kaapvaal Craton during the Kibaran (Grenville) Orogeny. Geology, in press. Kerr, A., 1985. Characterization of the granitic rocks of the Valley of a Thousand Hills area, Natal. S. Afr. J. Sci., 81: 475-479. Kober, B., 1986. Whole-grain evaporation for 2°Tpb/2°rpbage investigations on single zircons using a double-filament thermal ion source. Contrib. Mineral: Petrol.. 93: 482-490. Kober, B., 1987. Single-zircon evaporation combined with Pb + emitter bedding for 2°Tpb/2°SPb-age investigations using thermal ion source mass spectrometry, and implications to zirconology. Contrib. Mineral. Petrol., 96:63-71. Krogh, T.E., 1982. Improved accuracy of U-Pb zircon ages by the creation of more concordant systems, using an air abrasion technique. Geochim. Cosmochim. Acta, 46: 637-649. Ludwig, K.R., 1988. PBDAT for MSDOS: a computer program for IBM-PC compatibles for processing raw P h - U - T h isotope data. U.S.G.S. Open-File Rep. 88542, 35 pp. Misra, N.K. and Venkatsubramanian, V.S., 1977. Stron' tium diffusion in feldsapr: a laboratory study. Geochim. Cosmochim. Acta, 41: 837-838. Naidoo, D.D., Eglington, B.M. and Harmer, R.E., 1989. A Rb-Sr isotope study of a young granite sheet at Marble Delta, southern Natal. S. Aft. J. Geol.. 92: 389392. Nicolaysen, L.O. and Burger, A.J., 1965. Note on an extensive zone of 1000 million-year old metamorphic and igneous rocks in southern Africa. Sci. Terre, 10: 497518. Parrish, R.R., 1987. An improved micro-capsule for zircon dissolution in U - P b geochronology. Chem. Geol. (Isotope Geosci. Sect.), 66: 99-102. Pearce, J.A., Harris, N.B.W. and Tindle, A.G., 1984~ Trace element discrimination diagrams for the tectonic interpretation of granite rocks. J. Petrol., 25: 956-983. Stacey, J.S. and Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth Planet. Sci. Lett., 26:207-221. Steiger, R.H. and J~ger, E., 1977. Subeommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth Planet. Sci. Lett., 36: 359-362. Thomas, R.J., 1988a. The petrology of the Oribi Gorge Granitoid Suite: Kibaran charnockitic granitoids from southern Natal. S. Afr. J. Geol., 91:275-291. Thomas, R.J., 1988b. The geology of the Port Shepstone area. Explanation of Sheet 3030 (Port Shepstone), Geol. Surv. S. Afr., 136 pp. Thomas, R.J., 1989. A tale of two tectonic terranes. S. Aft. J. Geol., 92: 306-321.

NEW ISOTOPE DATA FROM A NEOPROTEROZOIC PORPHYRITIC SUITE

Thomas, R.J., 1991. Oribi Gorge Suite. In: M.R. Johnson (Editor), Catalogue of South African Lithostratigraphic Units. S. Afr. Comm. Strat., 3: 37-40. Thomas, R.J., Eglington, B.M. and Kerr, A., 1990. The geology and geochronology of the Belmont pluton and microgranite dykes from the Margate area--the search for Pan-African magmatism in Natal. S. Afr. J. Geol., 93: 766-775. Thomas, R.J., Eglington, B.M. and Bowring, S.A., 1993. Dating the cessation of Kibaran magmatism in Natal, South Africa. J. Afr. Earth Sci., 16, in press. Vaasjoki, M., Ramo, T. and Sakko, M., 1991. New U-Pb ages from the Wiborg rapakivi area: constrainst on the temporal evolution of the rapakivi granite-anorthosite-diabase dyke association of southeastern Finland. In: I. Haapala and K.C. Condie (Editors), Precam-

101 brian Granitoids--Petrogenesis, Geochemistry and Metallogeny. Precambrian Res., 51: 227-244. Vorma, A., 1971. Alkali feldspars of the Wiborg rapakivi massif in southeastern Finland. Bull. Comm. Geol. Finl., 246: 1-72. Watson, E.B. and Harrison, M., 1984. Accessory minerals and the geochemical evolution of crustal magmatic systems: A summary and prospectus of experimental approaches. Physics Earth Planet. Inter., 35:19-30. Whalen, J.B., Currie, K.L. and Chappel, B.W., 1987. Atype granites: geochemical characteristics, discrimination and petrogenesis. Contrib. Mineral. Petrol., 95: 407-419. Wikstr6m, A., 1984. A possible relationship between augen gneisses and postorogenic granites in S.E. Sweden. J. Struct. Geol., 6: 409-415.