Abstract

Metapelitic rock samples from the NE Shackleton Range, Antarctica, include garnet with contrasting zonation patterns and two age spectra. Garnet porphyroblasts in K-rich kyanite–sillimanite– staurolite–garnet–muscovite–biotite schists from Lord Nunatak show prograde growth zonation, and give Sm–Nd garnet, U–Pb monazite and Rb–Sr muscovite ages of 518 ± 5, 514 ± 1 and 499 ± 12 Ma, respectively. Geothermobarometry and PT pseudo-section calculations in the model system CaO–Na2O–K2O– TiO2–MnO–FeO–MgO–Al2O3–SiO2–H2O are consistent with garnet growth during prograde heating from 540°C/7 kbar to 650°C/7·5 kbar, and partial resorption during a subsequent PT decrease to <650°C at <6 kbar. All data indicate that rocks from Lord Nunatak were affected by a single orogenic cycle. In contrast, garnet porphyroblasts in K-poor kyanite–sillimanite– staurolite–garnet–cordierite–biotite-schists from Meade Nunatak show two growth stages and diffusion-controlled zonation. Two distinct age groups were obtained. Laser ablation plasma ionization multicollector mass spectrometry in situ analyses of monazite, completely enclosed by a first garnet generation, yield ages of c. 1700 Ma, whereas monazite grains in open garnet fractures and in most matrix domains give c. 500 Ma. Both age groups are also obtained by U–Pb thermal ionization mass spectrometry analyses of matrix monazite and zircon, which fall on a discordia with lower and upper intercepts at 502 ± 1 and 1686 ± 2 Ma, respectively. Sm–Nd garnet dating yields an age of 1571 ± 40 Ma and Rb–Sr biotite analyses give an age of 504 ± 1 Ma. Integrated geochronological and petrological data provide evidence that rocks from Meade Nunatak underwent a polymetamorphic Barrovian-type metamorphism: (1) garnet 1 growth and subsequent diffusive garnet annealing between 1700 and 1570 Ma; (2) garnet 2 growth during the Ross Orogeny at c. 500 Ma. During the final orogenic event the rocks experienced peak PT conditions of about 650°C/7·0 kbar and a retrograde stage at c. 575°C/4·0 kbar.

INTRODUCTION

The combination of detailed petrological investigations with geochronological methods is necessary to set tight constraints on the pressure–temperature–time (PTt) evolution of metamorphic terranes. For terranes that underwent a single orogenic cycle, reconstruction of PTt paths is commonly straightforward, because all observed mineral assemblages and rock textures were formed along a single PT loop. The geochronological age data obtained from rock-forming and minor minerals commonly reflect closure of the measured isotopic system. For systems with high closure temperatures this can effectively correspond to the time of mineral growth (e.g. garnet and monazite) and for those with low closure temperatures it often corresponds to cooling at the end of the PT history (e.g. Rb–Sr in mica). Nevertheless, problems with the interpretation of age data can occur if inherited isotopic components are involved. For example, detrital zircons can survive a medium- to high-grade metamorphic event without any serious disturbance to the isotopic system (e.g. Rubatto et al., 2001). Furthermore, some minerals can be affected by different resetting mechanisms, causing post-crystallization disturbance of their isotopic systems below the respective closure temperature (e.g. Zeh et al., 2003). A further problem involves the correlation of geochronological with petrological information, as the minerals that we commonly date (e.g. monazite or zircon) are not necessarily diagnostic of PT conditions (e.g. Foster et al., 2002).

In polymetamorphic gneiss terranes, the situation is more difficult, because the observed mineral assemblages and rock textures result from several distinct tectono-metamorphic cycles. Petrological and geochronological information about the early PT evolution can be partly or even completely obliterated. This, in particular, is the case if the final metamorphic event occurred under PT conditions and/or strain similar to or even higher than the earlier metamorphic event(s). In general, it is difficult, if not impossible, to prove a polymetamorphic history by petrological methods alone. Petrological indicators for a polymetamorphic evolution could be reflected by an abrupt change of the garnet growth zonation, garnet resorption patterns prior to a new garnet overgrowth, or the orientation of mineral inclusions. However, interpretations of all these indicators are not unique (see Zeh & Millar, 2001). Furthermore, the isotope system of older minerals can be partly or completely disturbed as a result of the effect of reheating and tectonic processes (e.g. Rb–Sr mica, Sm–Nd garnet, U–Pb zircon or monazite; e.g. Copeland et al., 1988; Pidgeon, 1992; Zhu et al., 1997; Vavra & Schaltegger, 1999; Zhu & O'Nions, 1999; Foster et al., 2000). Furthermore, Sm–Nd garnet isochrons, which are commonly used to estimate the timing of the prograde evolution or the metamorphic peak (e.g. Vance & O'Nions, 1990), may not necessarily reflect the time of garnet formation. They can be significantly biased by the effect of rare earth element (REE)-rich mineral inclusions (e.g. Prince et al., 2000) or reflect the isotopic signature of the minerals that reacted to form the garnet (e.g. Thöni & Jagoutz, 1992). Thus, to prove a polymetamorphic history for a certain gneiss terrane it is necessary to combine detailed petrology with a variety of geochronological methods.

In this paper, we present a detailed petrological study combined with geochronological results obtained from metapelitic rocks from the northeastern part of the Shackleton Range (Fig. 1). PT information is derived using conventional geothermobarometry and PT pseudosections in the model system CaO–Na2O–K2O– TiO2–MnO–FeO–MgO–Al2O3–SiO2–H2O. Geochronological data were obtained by four methods: (1) high-precision isotope dilution U–Pb thermal ionization mass spectrometry (TIMS) analysis; (2) in situ by laser ablation plasma ionization multicollector mass spectrometry (LA-PIMMS); (3) Sm–Nd isochrons; (4) Rb–Sr isochrons.

Fig. 1.

(a) Map of the Shackleton Range. Inset: position of the Shackleton Range in Antarctica. Stars indicate sample localities. (b) Block diagrams showing the geological situations at Lord and Meade nunataks.

Fig. 1.

(a) Map of the Shackleton Range. Inset: position of the Shackleton Range in Antarctica. Stars indicate sample localities. (b) Block diagrams showing the geological situations at Lord and Meade nunataks.

GEOLOGICAL BACKGROUND

The Shackleton Range is formed from two east-trending mountain belts, situated at 80–81°S (Fig. 1). The southern belt is exposed in the Read Mountains and at Stephenson Bastion, and forms part of the East Antarctic Craton. The basement of the southern belt predominantly consists of Proterozoic ortho- and paragneisses of the Read Group, which underwent high-grade amphibolite- to granulite-facies metamorphism (Schubert & Will, 1994; Talarico & Kroner, 1999). Proterozoic metamorphism and cooling is indicated by a Rb–Sr whole-rock isochron age of 1763 ± 32 Ma from a granitic orthogneiss (Pankhurst et al., 1983) and numerous Rb–Sr and K–Ar mineral cooling ages in the range 1550–1650 Ma (Hofmann et al., 1980; Pankhurst et al., 1983). During the Late Precambrian the basement of the southern belt was unconformably overlain by marine sediments of the Watts Needles Formation, the age of which is constrained by stromatolites and acritarchs (Golanov et al., 1980; Weber, 1991).

The history of the northern belt of the Shackleton Range, which is exposed from east to west in the Pioneers Escarpment, the Herbert Mountains, La Grange Nunataks and the Haskard Highlands (Fig. 1), is more complex than that of the southern belt. It consists of at least three units, which were tectonically intermingled during the Ross Orogeny. The Stratton Group (Clarkson, 1995) comprises upper amphibolite- to granulite-facies ortho- and paragneisses, which are lithologically similar to those of the Read Group. U–Pb zircon data from Stratton Group orthogneisses provide evidence for granite emplacement at 2328 ± 47 Ma (Mathys Gneiss: La Grange Nunataks, Brommer et al., 1999), and 1810 ± 2 Ma (Mount Weston Gneiss, Haskard Highlands, Zeh et al., 1999). At c. 1700 Ma the Stratton Group orthogneisses underwent high-grade metamorphism. This event is reflected by a 1715 ± 6 Ma U–Pb zircon age from a leucosome in the Mathys Gneiss (Brommer et al., 1999), and U–Pb monazite and Sm–Nd garnet–whole-rock ages of 1737 ± 3 Ma and 1665 ± 60 Ma, respectively, obtained from the migmatitic Mount Weston Gneiss (Zeh et al., 1999). This metamorphic–magmatic event can be correlated to the Nimrod/Kimban orogeny in the Central Transantarctic Mountains (e.g. Goodge et al., 2001).

The Pioneers Group (Clarkson, 1995) comprises a succession of medium- to high-grade metapelites and metapsammites, locally interbedded with marbles, metabasic rocks and very rare ultrabasic rocks. PT data from the Haskard Highlands, La Grange Nunataks and the Herbert Mountains provide evidence that rocks of the Pioneers Group underwent a Barrovian-type metamorphism with peak conditions of T = 630–750°C and P = 7–11 kbar (Schubert & Will 1994; Brommer et al., 1999; Zeh et al., 1999; Zeh, 2001). Sm–Nd garnet– whole-rock, U–Pb monazite, Rb–Sr and K–Ar mineral data indicate that rocks of the Pioneers Group underwent metamorphism and final cooling between 530 and 500 Ma (Brommer & Henjes-Kunst, 1999; Brommer et al., 1999; Zeh et al., 1999).

The third unit found in the northern Shackleton Range is an ophiolite complex (Talarico et al., 1999). This unit is interpreted as a relic of the Mozambique ocean that once separated the East Antarctic and Kalahari cratons and was closed during the Ross Orogeny at about 500 Ma (Tessensohn et al., 1999).

PETROGRAPHY AND MINERAL CHEMISTRY

Samples used in this study were collected from Lord and Meade nunataks of the Pioneers Escarpment (Fig. 1). Both nunataks comprise metapelitic schists, intercalated with metabasitic rocks and few marbles, and have been ascribed to the Pioneers Group. Rocks from both localities show the same structural style, which is characterized by east- to NE-trending open folds, flat 3–25° east- to south-plunging stretching lineations, and a flat to steep 5–85° SSE- or NNW-dipping foliation. Two representative metapelitic samples were investigated in detail; one sample from Lord Nunatak (sample Lo1: 80°22′S, 24°5′W) and a second from Meade Nunatak (sample Me3/3: 22°01′S, 80°20′W) (Fig. 1). A third sample, from Meade Nunatak (sample Me1/1: c. 1 m beside Me3/3), provides additional petrological and geochronological constraints.

Samples Lo1 and Me3/3 both contain kyanite, sillimanite, garnet, staurolite, plagioclase, quartz, ilmenite, rutile, zircon, monazite and tourmaline. Sample Lo1 additionally contains abundant muscovite and sparse biotite (Figs 2 and 3), whereas sample Me3/3 contains abundant cordierite and biotite (Figs 24). Muscovite relics in Me3/3 are restricted to being inclusions in the cores of zoned plagioclase (Fig. 5a), and in quartz aggregates overgrown by kyanite (Fig. 5b). The evolution of the mineral assemblages of the investigated samples is illustrated in Fig. 2. Minerals were analysed by electron microprobe (Appendix A) in several domains of the samples (Table 1).

Fig. 2.

Evolution of mineral assemblages in samples Lo1 and Me3/3, observed in single thin sections. Polyphase: minerals formed during two orogenic events (for further explanation, see text). D1, D2, deformation events constrained by microtextures.

Fig. 2.

Evolution of mineral assemblages in samples Lo1 and Me3/3, observed in single thin sections. Polyphase: minerals formed during two orogenic events (for further explanation, see text). D1, D2, deformation events constrained by microtextures.

Fig. 3.

(a, c) Photomicrographs and (b) element distribution maps of garnet from sample Lo1. (a) Resorbed garnet porphyroblast with ilmenite inclusions (Ilm) is surrounded by kyanite (Ky), muscovite (Ms), staurolite (St), plagioclase (Pl), and biotite (Bt). The zonation profile xy is shown in Fig. 6a. Inset (upper left): polygonally recrystallized quartz is overgrown by garnet. (b) Element maps of Mg, Fe, Ca and Mn reveal a euhedral core with threefold symmetry, surrounded by a radial growth zone. It should be noted that, at lower right, the threefold zonation is truncated and overgrown by a later garnet generation, which also overgrows statically recrystallized quartz. At the outermost rim, garnet displays a diffusion zonation. (c) Garnet and kyanite are replaced by sillimanite (Sil). Kyanite contains ilmenite and rutile (Rt) inclusions.

Fig. 3.

(a, c) Photomicrographs and (b) element distribution maps of garnet from sample Lo1. (a) Resorbed garnet porphyroblast with ilmenite inclusions (Ilm) is surrounded by kyanite (Ky), muscovite (Ms), staurolite (St), plagioclase (Pl), and biotite (Bt). The zonation profile xy is shown in Fig. 6a. Inset (upper left): polygonally recrystallized quartz is overgrown by garnet. (b) Element maps of Mg, Fe, Ca and Mn reveal a euhedral core with threefold symmetry, surrounded by a radial growth zone. It should be noted that, at lower right, the threefold zonation is truncated and overgrown by a later garnet generation, which also overgrows statically recrystallized quartz. At the outermost rim, garnet displays a diffusion zonation. (c) Garnet and kyanite are replaced by sillimanite (Sil). Kyanite contains ilmenite and rutile (Rt) inclusions.

Fig. 4.

(a, b) Photomicrographs and (c, d) element distribution maps of garnet from sample Me3/3. (a) Garnet with primary biotite (Bt), rutile (Rt) and quartz inclusions is resorbed by matrix biotite, which traces the foliation. Along garnet fractures monazite (Mnz) occurs. The profile xy is shown in Fig. 6a. (b) Euhedral garnet grain (Grt) with apatite (Ap) inclusions is overgrown by kyanite (Ky), which contains rutile (Rt) inclusions. It should be noted that the rounded garnet core is overgrown by a second euhedral garnet generation, which contains abundant rutile and quartz inclusions, which both appear as black spray. (c) Element maps (Mn, Ca, Fe and Mg) of the garnet grain shown in (a). The garnet shows a pronounced diffusion zoning at the rim and along fractures. (d) Section of a garnet grain showing a spiderweb texture formed by Ca-rich veins. The web texture is truncated by open fractures filled with chlorite. It should be noted that the web textures can be observed only in Ca, but not in the Mg, Mn and Fe maps (further explanation is given in the text).

Fig. 4.

(a, b) Photomicrographs and (c, d) element distribution maps of garnet from sample Me3/3. (a) Garnet with primary biotite (Bt), rutile (Rt) and quartz inclusions is resorbed by matrix biotite, which traces the foliation. Along garnet fractures monazite (Mnz) occurs. The profile xy is shown in Fig. 6a. (b) Euhedral garnet grain (Grt) with apatite (Ap) inclusions is overgrown by kyanite (Ky), which contains rutile (Rt) inclusions. It should be noted that the rounded garnet core is overgrown by a second euhedral garnet generation, which contains abundant rutile and quartz inclusions, which both appear as black spray. (c) Element maps (Mn, Ca, Fe and Mg) of the garnet grain shown in (a). The garnet shows a pronounced diffusion zoning at the rim and along fractures. (d) Section of a garnet grain showing a spiderweb texture formed by Ca-rich veins. The web texture is truncated by open fractures filled with chlorite. It should be noted that the web textures can be observed only in Ca, but not in the Mg, Mn and Fe maps (further explanation is given in the text).

Fig. 5.

Photomicrographs of sample Me3/3. (a) Muscovite (Ms) flakes and rutile (Rt) grains are enclosed by plagioclase (Pl). (b) Muscovite and biotite flakes enclosed by quartz (Qtz) are overgrown by kyanite (Ky), which also contains rutile (Rt). (c) Cordierite–quartz (Crd + Qtz) symplectite associated with biotite (Bt). (d) Cordierite–quartz symplectite in contact with kyanite, staurolite (St) and biotite. (e) Association of staurolite, cordierite and sillimanite (Sil). (f) Staurolite core with a staurolite–quartz symplectite overgrowth replaces kyanite and occurs in contact with cordierite, biotite and ilmenite (black). The profile xy is shown in Fig. 6c.

Fig. 5.

Photomicrographs of sample Me3/3. (a) Muscovite (Ms) flakes and rutile (Rt) grains are enclosed by plagioclase (Pl). (b) Muscovite and biotite flakes enclosed by quartz (Qtz) are overgrown by kyanite (Ky), which also contains rutile (Rt). (c) Cordierite–quartz (Crd + Qtz) symplectite associated with biotite (Bt). (d) Cordierite–quartz symplectite in contact with kyanite, staurolite (St) and biotite. (e) Association of staurolite, cordierite and sillimanite (Sil). (f) Staurolite core with a staurolite–quartz symplectite overgrowth replaces kyanite and occurs in contact with cordierite, biotite and ilmenite (black). The profile xy is shown in Fig. 6c.

Table 1:

Mineral analyses of samples Lo1 and Me3/3; F/FM = FeO/(FeO + MgO)

 garnet
 
   biotite
 
 staurolite
 
  chlorite
 
 cordierite
 
 plagioclase
 
   ilmenite
 
 
Sample: Me3/3 Me3/3 Lo1 Lo1 Me3/3 Lo1 Lo1 Lo1 Me3/3 Me3/3 Me3/3 Me3/3 Me3/3 Me3/3 Me3/3 Lo1 Lo1 Me3/3 Lo1 
 core
 
rim
 
core
 
rim
 

 

 
core
 
rim
 
core
 
rim
 

 

 

 
core
 
rim
 
core
 
rim
 

 

 
SiO2 37·56 37·03 36·80 36·51 35·63 35·26 27·48 27·90 27·90 27·61 24·37 48·03 48·03 62·15 58·28 66·35 64·97 
TiO2 0·01 0·02 0·13 0·04 2·18 2·59 0·76 0·8 0·63 0·02 0·10 0·03 0·03 53·87 53·28 
Al2O3 21·36 20·98 20·78 21·04 18·68 19·56 54·28 54·22 53·29 54·55 23·13 32·91 33·21 24·32 25·99 20·75 22·03 
Cr2O3 0·04 0·03 0·01 0·05 0·34 0·10 0·06 
Fe2O3 0·63 0·03 0·64 0·63 0·84 1·71 1·7 0·02 0·05 
FeO 32·46 36·68 33·74 36·98 18·05 21·37 14·16 15·02 13·81 13·43 22·49 6·16 6·14 45·44 46·44 
MnO 0·41 0·93 0·79 0·05 0·11 0·10 0·07 0·07 0·13 0·54 0·48 
MgO 6·26 3·17 0·48 2·73 10·65 7·83 1·14 1·65 2·29 2·28 16·61 8·94 8·89 0·06 0·02 
ZnO — — — — — — 0·16 0·21 0·71 0·55 — — — — — — —   
CaO 1·38 1·48 7·17 2·28 — — — — 5·27 7·68 1·13 2·39 
Na20·25 0·34 — — — — 0·19 0·23 8·63 7·57 11·03 10·26 
K29·16 8·45 — — — — 0·14 0·07 0·08 0·06 
Total 100·11 100·35 100·54 100·32 94·95 95·51 98·09 99·80 98·63 98·54 87·63 98·04 98·36 100·51 99·59 99·36 99·78 99·97 100·21 
Oxygens 12 12 12 12 11 11 46 46 46 46 14 18 18 
Si 2·966 2·982 2·979 2·949 2·702 2·686 7·623 7·638 7·711 7·615 2·539 4·942 4·927 2·741 2·617 2·927 2·864 
Ti 0·001 0·001 0·008 0·002 0·124 0·148 0·158 0·165 0·131 0·003 0·008 0·002 0·002 1·016 1·007 
Al 1·989 1·992 1·981 2·004 1·670 1·757 17·750 17·491 17·362 17·731 2·841 3·992 4·016 1·264 1·376 1·079 1·144 
Cr 0·002 0·002 0·001 0·003 0·020 0·006 0·001 
Fe3+ 0·037 0·002 0·038 0·038 — — — — 0·066 0·133 0·132 0·001 0·002 
Fe2+ 2·143 2·470 2·300 2·498 1·145 1·362 3·286 3·439 3·192 3·098 1·959 0·530 0·526 0·953 0·976 
Mn 0·027 0·063 0·005 0·003 0·025 0·006 0·006 0·011 0·011 0·01 
Mg 0·736 0·38 0·070 0·329 1·204 0·889 0·469 0·672 0·943 0·936 2·581 1·371 1·359 0·002 0·001 
Zn — — — — — — 0·016 0·042 0·144 0·111 — — — — — — —   
Ca 0·117 0·128 0·630 0·197 — — — — 0·249 0·370 0·053 0·113 
Na 0·037 0·05 — — — — 0·038 0·046 0·738 0·659 0·943 0·877 
0·887 0·822 — — — — 0·008 0·004 0·005 0·003 
Sum 8·019 8·02 8·0124 8·025 7·790 7·720 29·327 29·450 29·480 29·494 10·000 11·013 11·020 5·000 5·026 5·008 5·004 1·984 1·993 
F/FM 0·74 0·87 0·97 0·88 0·49 0·61 0·88 0·84 0·77 0·77 0·43 0·28 0·28 — — — — — — 
Xalm 0·709 0·812 0·765 0·825 — — — — — — — — — — — — — — — 
Xpy 0·243 0·125 0·023 0·109 — — — — — — — — — — — — — — — 
Xgrs 0·039 0·042 0·210 0·065 — — — — — — — — — — — — — — — 
Xspss 0·009 0·021 0·002 0·001 — — — — — — — — — — — — — — — 
Xan — — — — — — — — — — — — — 0·25 0·36 0·05 0·11 — — 
 garnet
 
   biotite
 
 staurolite
 
  chlorite
 
 cordierite
 
 plagioclase
 
   ilmenite
 
 
Sample: Me3/3 Me3/3 Lo1 Lo1 Me3/3 Lo1 Lo1 Lo1 Me3/3 Me3/3 Me3/3 Me3/3 Me3/3 Me3/3 Me3/3 Lo1 Lo1 Me3/3 Lo1 
 core
 
rim
 
core
 
rim
 

 

 
core
 
rim
 
core
 
rim
 

 

 

 
core
 
rim
 
core
 
rim
 

 

 
SiO2 37·56 37·03 36·80 36·51 35·63 35·26 27·48 27·90 27·90 27·61 24·37 48·03 48·03 62·15 58·28 66·35 64·97 
TiO2 0·01 0·02 0·13 0·04 2·18 2·59 0·76 0·8 0·63 0·02 0·10 0·03 0·03 53·87 53·28 
Al2O3 21·36 20·98 20·78 21·04 18·68 19·56 54·28 54·22 53·29 54·55 23·13 32·91 33·21 24·32 25·99 20·75 22·03 
Cr2O3 0·04 0·03 0·01 0·05 0·34 0·10 0·06 
Fe2O3 0·63 0·03 0·64 0·63 0·84 1·71 1·7 0·02 0·05 
FeO 32·46 36·68 33·74 36·98 18·05 21·37 14·16 15·02 13·81 13·43 22·49 6·16 6·14 45·44 46·44 
MnO 0·41 0·93 0·79 0·05 0·11 0·10 0·07 0·07 0·13 0·54 0·48 
MgO 6·26 3·17 0·48 2·73 10·65 7·83 1·14 1·65 2·29 2·28 16·61 8·94 8·89 0·06 0·02 
ZnO — — — — — — 0·16 0·21 0·71 0·55 — — — — — — —   
CaO 1·38 1·48 7·17 2·28 — — — — 5·27 7·68 1·13 2·39 
Na20·25 0·34 — — — — 0·19 0·23 8·63 7·57 11·03 10·26 
K29·16 8·45 — — — — 0·14 0·07 0·08 0·06 
Total 100·11 100·35 100·54 100·32 94·95 95·51 98·09 99·80 98·63 98·54 87·63 98·04 98·36 100·51 99·59 99·36 99·78 99·97 100·21 
Oxygens 12 12 12 12 11 11 46 46 46 46 14 18 18 
Si 2·966 2·982 2·979 2·949 2·702 2·686 7·623 7·638 7·711 7·615 2·539 4·942 4·927 2·741 2·617 2·927 2·864 
Ti 0·001 0·001 0·008 0·002 0·124 0·148 0·158 0·165 0·131 0·003 0·008 0·002 0·002 1·016 1·007 
Al 1·989 1·992 1·981 2·004 1·670 1·757 17·750 17·491 17·362 17·731 2·841 3·992 4·016 1·264 1·376 1·079 1·144 
Cr 0·002 0·002 0·001 0·003 0·020 0·006 0·001 
Fe3+ 0·037 0·002 0·038 0·038 — — — — 0·066 0·133 0·132 0·001 0·002 
Fe2+ 2·143 2·470 2·300 2·498 1·145 1·362 3·286 3·439 3·192 3·098 1·959 0·530 0·526 0·953 0·976 
Mn 0·027 0·063 0·005 0·003 0·025 0·006 0·006 0·011 0·011 0·01 
Mg 0·736 0·38 0·070 0·329 1·204 0·889 0·469 0·672 0·943 0·936 2·581 1·371 1·359 0·002 0·001 
Zn — — — — — — 0·016 0·042 0·144 0·111 — — — — — — —   
Ca 0·117 0·128 0·630 0·197 — — — — 0·249 0·370 0·053 0·113 
Na 0·037 0·05 — — — — 0·038 0·046 0·738 0·659 0·943 0·877 
0·887 0·822 — — — — 0·008 0·004 0·005 0·003 
Sum 8·019 8·02 8·0124 8·025 7·790 7·720 29·327 29·450 29·480 29·494 10·000 11·013 11·020 5·000 5·026 5·008 5·004 1·984 1·993 
F/FM 0·74 0·87 0·97 0·88 0·49 0·61 0·88 0·84 0·77 0·77 0·43 0·28 0·28 — — — — — — 
Xalm 0·709 0·812 0·765 0·825 — — — — — — — — — — — — — — — 
Xpy 0·243 0·125 0·023 0·109 — — — — — — — — — — — — — — — 
Xgrs 0·039 0·042 0·210 0·065 — — — — — — — — — — — — — — — 
Xspss 0·009 0·021 0·002 0·001 — — — — — — — — — — — — — — — 
Xan — — — — — — — — — — — — — 0·25 0·36 0·05 0·11 — — 

Garnet

Garnet porphyroblasts in sample Lo1 have diameters between 4 and 6 mm, and are invariably resorbed and mechanically eroded at their edges (Fig. 3). Garnet shows a typical prograde growth zonation, characterized by a decrease of XFe = Fe/(Fe + Mg) = 0·97 → 0·87, spessartine [Xspss = Mn/(Fe + Mg + Ca + Mn) = 0·035 → 0·005] and grossular [Xgrs = Ca/(Fe + Mg + Ca + Mn) = 0·205 → 0·05] from core to rim, whereas the pyrope component [Xpy = Mg/(Fe + Mg + Ca + Mn) = 0·02 → 0·10] increases (Fig. 6a). The almandine zonation shows a significant kink [Xalm = Fe/(Fe + Mg + Ca + Mn) = 0·76 → 0·85 → 0·82], which divides garnet into a core (zone 1) and an outermost zone 2, which can also be seen in the Xgrs and Xspss zonation patterns (Fig. 6a). Garnet zones 1 and 2 contain inclusions of ilmenite, muscovite, and rare plagioclase and tourmaline. Rutile inclusions are rare at the garnet rim. Element distribution maps of garnet Lo1 reveal a triangular zonation pattern for garnet zone 1, whereas garnet zone 2 shows concentric zonation (Fig. 3b). The different zonation indicates that the growth mechanism changed during garnet formation. As shown in Fig. 3a and b, the triangular pattern of garnet zone 1 is cut off at one edge, and is unconformably overgrown by garnet with a zone 2 composition. These features provide evidence that garnet Lo1 was sheared after zone 1 formation. The deformation event is designated here as D1. After garnet zone 2 growth ceased, garnet Lo1 was resorbed at the expense of biotite and sillimanite, and mechanically eroded during D2 (Fig. 3a and c). Finally, a tiny diffusion zone (c. 50 µm) was established at the garnet edges (Fig. 3b).

Fig. 6.

Chemical compositions and zonations of (a) garnet, (b) biotite, (c) staurolite, and (d) plagioclase. (a) Garnet of sample Lo1 commonly shows a pronounced growth zonation, which can be divided into two zones (zone 1 and zone 2), whereas garnet from samples Me3/3 and Me1/1 show diffusion zonations. Mx, garnet surrounded by different matrix minerals; (Ky), garnet completely enclosed by kyanite; Xan, anorthite content of plagioclase inclusions. (For further explanation, see the text).

Fig. 6.

Chemical compositions and zonations of (a) garnet, (b) biotite, (c) staurolite, and (d) plagioclase. (a) Garnet of sample Lo1 commonly shows a pronounced growth zonation, which can be divided into two zones (zone 1 and zone 2), whereas garnet from samples Me3/3 and Me1/1 show diffusion zonations. Mx, garnet surrounded by different matrix minerals; (Ky), garnet completely enclosed by kyanite; Xan, anorthite content of plagioclase inclusions. (For further explanation, see the text).

Garnet porphyroblasts in sample Me3/3 have diameters between 0·5 and 6 mm. Matrix garnets commonly form angular and rarely round or elliptical relics, which are mechanically eroded and subsequently resorbed by biotite, with or without cordierite (Fig. 4a). Garnet erosion took place after formation of the kyanite-bearing peak metamorphic assemblage, and thus is correlated with D2, as defined above for sample Lo1. Euhedral garnet was only observed enclosed by kyanite (Fig. 4b). This garnet shows a rounded core, overgrown by a narrow rim, which contains abundant tiny rutile and quartz inclusions. The transition between the core and rim is marked texturally by the accumulation of fine rutile grains (Fig. 4b), and chemically by an abrupt increase of Xgrs (Fig. 6a). Primary inclusions in the core are quartz, biotite, rutile, zircon, apatite, monazite and tourmaline. Along garnet fractures quartz, biotite, chlorite, ilmenite and monazite occur. In contrast to sample Lo1, garnet in sample Me3/3, and in all other investigated samples from Meade Nunatak (e.g. garnet Me1/1 in Fig. 6a) invariably shows a pronounced retrograde diffusion zonation, characterized by an increase of XFe (from 0·74 to 0·87) and Xspss (from 0·01 to 0·04) toward garnet rims and open fractures, whereas Xgrs is nearly constant (Fig. 6a). Exceptions are garnet grains that show two growth generations (e.g. garnet Ky3/3; Fig. 4b). In these grains, Xgrs increases steeply at the transition between the inclusion-free core and the inclusion-rich euhedral overgrowth (Fig. 6a), indicating that the Xgrs zonation has not been modified by subsequent diffusion. In contrast, XFe and Xspss of garnet Ky3/3 show the same retrograde diffusion patterns as the matrix garnet, even through garnet Ky3/3 is completely enclosed by kyanite (Fig. 6a). This might indicate that kyanite formation took place after the diffusive garnet alteration. Alternatively, the diffusion patterns could result from Fe, Mg and Mn being transported by a fluid phase along kyanite microfractures after kyanite overgrowth, or from the diffusive modification of initially steep XFe and Xspss compositional gradients, established during garnet 2 overgrowth. The Xgrs profile was unaffected, as a result of the much slower diffusion rate of Ca in garnet, in contrast to Fe, Mg and Mn (Chakraborty & Ganguly, 1992).

Some matrix garnets show a characteristic web of 5–15 µm wide veins, in which Xgrs is 10–20% higher than in the surrounding garnet (Fig. 4d). These veins indicate brecciation and subsequent annealing of the garnet. The Xgrs-rich veins are cut by open fractures filled with retrograde biotite and chlorite (Fig. 4d), indicating that formation of the Xgrs-rich veins and the open fractures result from two distinct processes. This conclusion is also supported by the fact that Xgrs does not increase toward the open fractures (Fig. 4c and d), whereas Xalm and Xspss show a pronounced diffusion zonation. Furthermore, the web texture can only be recognized in the Ca but not in the Fe, Mg and Mn maps (Fig. 4d). This feature perhaps results from the much slower volume diffusion of Ca in garnet, in contrast to Fe, Mg and Mn (e.g. Chakraborty & Ganguly, 1992). This may have resulted in smoothing out of an initial XFe and Xspss zonation after veining by volume diffusion, whereas Xgrs was unaffected. The elevated Xgrs in the veins could indicate that garnet annealing and formation of the Xgrs-rich garnet overgrowth occurred during the same event.

Staurolite

Staurolite in sample Lo1 forms angular grains, which may be intergrown and overgrown by kyanite and locally replaced by chlorite. In sample Me3/3 staurolite is commonly sheared and/or surrounded by a rim of cordierite and biotite ± chlorite ± sillimanite (Fig. 5d–f). In both samples, staurolite contains rounded inclusions of quartz as well as euhedral ilmenite, rutile, and sparse biotite, tourmaline and zircon. Generally, XFe of staurolite in sample Me3/3 is lower (0·78–0·80) than that of the muscovite-bearing sample Lo1 (0·83–0·88; Table 1). Most matrix staurolite in both samples is chemically unzoned. However, XFe of individual staurolite grains is slightly different. In sample Lo1, a staurolite grain with a prograde growth zonation was found (Fig. 6c). In this grain XFe (0·88–0·84) and the Al content (8·9–8·75 p.f.u.) decrease from core to rim, whereas the Ti content increases slightly (0·06–0·08 p.f.u.).

Zoned staurolite grains were also observed in sample Me3/3, showing cores surrounded by symplectitic staurolite–quartz rims in contact with kyanite (Fig. 5f). These symplectites occur in domains with the assemblage kyanite, staurolite, cordierite, biotite and garnet. Symplectitic staurolite shows a pronounced chemical zonation from core to rim (Fig. 6c), which is characterized by a significant decrease of Si (3·85–3·75 p.f.u.) and Ti (0·08–0·00 p.f.u.), and an increase of Al (8·65–8·90 p.f.u.). XFe (0·78–0·77) and the Zn content (0·045–0·06 p.f.u.) are nearly constant. It should be noted that the Al, Si and Ti zonation is opposite to the prograde zonation observed in sample Lo1. From the intergrowth relationships and the domain assemblage we conclude that the staurolite symplectites in sample Me3/3 were formed during the retrograde evolution, via the FMASH net-reaction  

(1)
\[\mathrm{Ky}\ +\ \mathrm{Grt}\ +\ \mathrm{Qtz}\ +\ \mathrm{H}_{2}\mathrm{O}\ =\ \mathrm{St}\ +\ \mathrm{Crd}.\]

Cordierite

Cordierite was observed only in sample Me3/3, in which it commonly forms rims around kyanite, staurolite and locally around garnet, and is associated with biotite and in some places with sillimanite and chlorite (Fig. 5c–e). In some domains, cordierite–quartz symplectites were observed (Fig. 5c and d). Cordierite, including the symplectites, is commonly pinitized. Unaltered cordierite shows XFe ratios between 0·32 and 0·33, and Na contents between 0·04 and 0·05 p.f.u.

Plagioclase

Matrix plagioclase in all samples shows increasing anorthite contents [Xan = Ca/(Ca + Na)] from core to rim (Fig. 6d). Xan of plagioclase in the K-poor samples is generally higher (Me3/3: 0·25–0·36) than in the K-rich sample Lo1 (0·05–0·11). Xan of plagioclase inclusions in garnet Lo1 decreases from 0·15 to 0·08 from garnet core to rim (Fig. 6a).

Biotite

Biotite is abundant in sample Me3/3, but occurs rarely in sample Lo1, where it has generally formed at the expense of garnet (Fig. 3a). In general, XFe of biotite from sample Lo1 (0·59–0·62) is much higher than that of sample Me3/3 (0·38–0·49). The same holds true for Tivi (Lo1: 0·11–0·15 p.f.u.; Me3/3: 0·03–0·13 p.f.u.), whereas Alvi contents scatter in the same range (Lo1: 0·37–0·46 p.f.u.; Me3/3: 0·35–0·48 p.f.u.). Throughout the thin sections of the individual samples, biotite has a similar composition. An exception is biotite located on garnet fractures in sample Me3/3. These have higher Alvi contents, but lower Tivi and XFe (Fig. 6b). Lower Tivi contents were also observed in retrograde biotite formed at the expense of garnet Lo1 (Fig. 6b).

Chlorite

Chlorite in samples Lo1 and Me3/3 occurs along fractures in garnet and around resorbed garnet grains, commonly associated with biotite. Chlorite flakes in sample Lo1 have higher XFe (0·48–0·49) than those from sample Me3/3 (0·42–0·43). The Alvi scatters between 2·68 and 2·89 p.f.u. in sample Me3/3, and between 2·73 and 2·81 p.f.u. in sample Lo1.

Muscovite

Muscovite is abundant in sample Lo1 and was rarely observed in plagioclase cores and in quartz aggregates enclosed by kyanite in sample Me3/3 (Fig. 5a and b). Muscovite in both samples has Si contents between 3·01 and 3·08 p.f.u. The K/(K + Na) ratios range between 0·78 and 0·85 in sample Lo1, and between 0·75 and 0·80 in sample Me3/3.

Ilmenite and rutile

In sample Lo1 ilmenite was found enclosed in garnet (Fig. 3a), together with rutile in the matrix, and enclosed by staurolite and kyanite (Fig. 3c). Locally, ilmenite is overgrown by rutile. Ilmenite in garnet zones 1 and 2 shows little compositional variation. The ilmenite component (Xilm) scatters between 0·98 and 0·99, the pyrophanite component (Xpph) between 0·01 and 0·02, and the hematite component (Xhem) between 0 and 0·01. Matrix ilmenite grains and those enclosed by staurolite and kyanite show similar compositions. Rutile is nearly pure TiO2. In sample Me3/3, rutile occurs enclosed in plagioclase, and in garnet (Figs 4a, b and 5a, b), whereas ilmenite occurs in open garnet fractures, in cordierite domains, and together with rutile in staurolite. Ilmenite in all domains has the composition ilm94–98pph4–5hem1–2.

Zircon

Zircon is abundant in sample Me3/3, but was rarely observed in heavy mineral concentrates of sample Lo1. In this study only zircon from sample Me3/3 was investigated. The zircon grains are commonly smaller than 100 µm and occur as inclusions in nearly all minerals. They are round, never show facets, and invariably have corroded surfaces indicating that the zircon grains are either of detrital origin and/or were corroded after deposition by fluid interactions. Cathodoluminescence (CL) images of all investigated zircon grains show wide bands and diffuse domains with a characteristic bright luminescence (Fig. 7). These bands and domains truncate the former growth zonation, which is preserved as relics in a few grains, but in most grains is completely obliterated. This indicates extensive zircon alteration after their growth (see Vavra et al., 1999).

Fig. 7.

Cathodoluminescence images of zircon grains of sample Me3/3. Maximal zircon length 50 µm.

Fig. 7.

Cathodoluminescence images of zircon grains of sample Me3/3. Maximal zircon length 50 µm.

Monazite

Monazite was observed in the matrix of sample Lo1. In sample Me3/3 monazite is abundant in the matrix, where it is associated with biotite, cordierite, quartz and locally with sillimanite (Fig. 8e and f). By means of back-scattered scanning electron microscopy (SEM) coupled with energy dispersive spectrometry (EDX), a few tiny monazite inclusions (10–20 µm) completely enclosed by garnet could also be identified (Fig. 8a). Additionally, abundant monazite grains were observed within retrograde garnet fractures and patches filled with biotite (Fig. 8b). Monazite was also observed completely enclosed in kyanite (Fig. 8d), and in quartz–tourmaline aggregates partly surrounded by staurolite (Fig. 8c).

Fig. 8.

Back-scattered electron image (a), and photomicrographs (b–f) of monazite from sample Me3/3. (a, b) Monazites are either completely enclosed by garnet (Mnz36) or occur along retrograde fractures or in patches filled with biotite (Mnz33, 32 and 13). (c) Monazite, tourmaline (Trl) and quartz (Qtz) are partly enclosed by staurolite (St). (d) Monazite is completely overgrown by kyanite. (e) Monazite enclosed by symplectitic cordierite. (f) Sillimanite is overgrown by monazite.

Fig. 8.

Back-scattered electron image (a), and photomicrographs (b–f) of monazite from sample Me3/3. (a, b) Monazites are either completely enclosed by garnet (Mnz36) or occur along retrograde fractures or in patches filled with biotite (Mnz33, 32 and 13). (c) Monazite, tourmaline (Trl) and quartz (Qtz) are partly enclosed by staurolite (St). (d) Monazite is completely overgrown by kyanite. (e) Monazite enclosed by symplectitic cordierite. (f) Sillimanite is overgrown by monazite.

GEOTHERMOBAROMETRY

PT calculations were carried out with the ‘average pressure–temperature’ feature of the software package THERMOCALC v.3.1 (Holland & Powell, 1998), using the composition of minerals inferred to belong to the peak metamorphic assemblage. For sample Lo1, the compositions of garnet rims unaffected by retrograde diffusion, staurolite with the lowest XFe, biotite enclosed by kyanite, matrix muscovite, and plagioclase cores were used. For sample Me3/3 the compositions of garnet cores unaffected by retrograde diffusion, matrix biotite, staurolite cores, and plagioclase cores were used. In all samples kyanite, rutile, ilmenite, quartz and H2O are additional phases. The use of plagioclase cores is justified for the following reasons. Xan of plagioclase inclusions in garnet Lo1 decreases from garnet core to rim (Fig. 6a), whereas Xan in matrix plagioclase increases toward plagioclase rims (Fig. 6c). Thus, matrix plagioclase rims must be formed during the retrograde evolution. Activities of mineral end-members were calculated with the AX program of Tim Holland (available at http://www.esc.cam.ac.uk/astaff/holland/index.html). Mineral analyses and end-members used are shown in Tables 1 and 2, and the results of PT calculations in Table 2 and Fig. 9.

Table 2:

Results of P–T calculations

Sample End-members
 
XH2O
 
T (°C)
 
SD
 
P (kbar)
 
SD
 
corr.
 
sigfit
 
Me3/3-peak py, gr, alm, phl, ann, east, mst, fst, an, ilm, ru, ky, q 1·00 663 22 6·8 1·0 0·269 0·96 
  0·75 633 24 6·5 1·1 0·251 1·17 
  0·50 603 26 6·2 1·3 0·256 1·39 
Lo1-peak alm, py phl, ann, east, mst, fst, ab, mu, pa, cel, ilm, ru, ky, q 1·00 665 17 8·7 1·2 0·686 0·78 
  0·75 635 16 8·5 1·2 0·674 0·90 
  0·50 606 15 8·3 1·2 0·665 1·03 
Sample End-members
 
XH2O
 
T (°C)
 
SD
 
P (kbar)
 
SD
 
corr.
 
sigfit
 
Me3/3-peak py, gr, alm, phl, ann, east, mst, fst, an, ilm, ru, ky, q 1·00 663 22 6·8 1·0 0·269 0·96 
  0·75 633 24 6·5 1·1 0·251 1·17 
  0·50 603 26 6·2 1·3 0·256 1·39 
Lo1-peak alm, py phl, ann, east, mst, fst, ab, mu, pa, cel, ilm, ru, ky, q 1·00 665 17 8·7 1·2 0·686 0·78 
  0·75 635 16 8·5 1·2 0·674 0·90 
  0·50 606 15 8·3 1·2 0·665 1·03 

XH2O, mole fraction of H2O of a mixed H2O–CO2 fluid; SD, 2σ standard deviation; corr. and sigfit. obtained with THERMOCALC; end-member abbreviations according to Holland & Powell (1998).

Fig. 9.

PT pseudosection of sample Lo1 (a) and Me3/3 (b) in the model system CNKTiMnFMASH calculated with THERMOCALC (Holland & Powell 1998), using the dataset HP98. L1–L3 and M1–M3 are mineral assemblages observed in different domains of sample Lo1 and Me3/3, respectively (see Fig. 2). Black line is PT path constrained by the domain assemblages in the respective samples; ellipses show peak PT conditions estimated with THERMOCALC (see Table 2).

Fig. 9.

PT pseudosection of sample Lo1 (a) and Me3/3 (b) in the model system CNKTiMnFMASH calculated with THERMOCALC (Holland & Powell 1998), using the dataset HP98. L1–L3 and M1–M3 are mineral assemblages observed in different domains of sample Lo1 and Me3/3, respectively (see Fig. 2). Black line is PT path constrained by the domain assemblages in the respective samples; ellipses show peak PT conditions estimated with THERMOCALC (see Table 2).

Peak PT conditions of 665 ± 17°C at 8·7 ± 1·2 kbar were calculated for sample Lo1, and peak conditions of 663 ± 22°C at 6·8 ± 1·0 kbar for sample Me3/3, assuming XH2O= 1·0 (Table 2). These pressures and temperatures are identical, within error. Peak PT conditions estimated for sample Lo1 are regarded as more reliable because garnet in that sample was minimally affected by retrograde diffusion and the mode of the peak assemblage appears to have changed only slightly during the retrograde evolution. In contrast, the peak PT conditions derived from sample Me3/3 must be considered with more caution, because garnet in that sample underwent extensive resorption and diffusive re-equilibration (see above), and abundant cordierite, staurolite and sillimanite were formed during the retrograde evolution (Fig. 5). This may have caused a significant change of the matrix biotite composition in sample Me3/3 during the retrograde PT evolution, even though the amount of biotite has not changed, because it is the only K-bearing mineral in the rock.

PT PSEUDOSECTIONS

To obtain additional information about the metamorphic evolution, PT pseudosections in the model system CaO–Na2O–K2O–TiO2–MnO–FeO–MgO–Al2O3– SiO2–H2O (CNKTiMnFMASH) were constructed using THERMOCALC v.3.1 (Holland & Powell, 1998), with activity models described by White et al. (2000) and Zeh & Holness (2003), and H2O and quartz in excess. Additional components, e.g. Fe2O3, were not considered, because no ferric iron phases such as magnetite were observed in our samples and the hematite component in ilmenite is very small. As ferric iron is not strongly fractionated by any of the considered minerals, it will have little influence on the phase diagram topologies presented below. Bulk compositions of samples Lo1 and Me3/3 were obtained from X-ray fluorescence (XRF) data of representative sub-samples of about 10 mm × 20 mm × 30 mm size (Table 3), taken in proximity of the thin sections investigated. In the following section, mineral assemblages and composition trends observed in samples Lo1 and Me3/3 will be compared with those predicted in PT pseudosections.

Table 3:

XRF analyses of the investigated samples

Sample: Me3/3
 
Lo1
 
SiO2 53·49 60·79 
TiO2 0·84 0·93 
Al2O3 24·52 21·14 
FeOtot 10·74 8·02 
MnO 0·11 0·13 
MgO 3·91 0·94 
CaO 0·49 1·02 
Na20·58 1·81 
K22·79 2·62 
P2O5 0·06 0·16 
Total 97·53 97·56 
Sample: Me3/3
 
Lo1
 
SiO2 53·49 60·79 
TiO2 0·84 0·93 
Al2O3 24·52 21·14 
FeOtot 10·74 8·02 
MnO 0·11 0·13 
MgO 3·91 0·94 
CaO 0·49 1·02 
Na20·58 1·81 
K22·79 2·62 
P2O5 0·06 0·16 
Total 97·53 97·56 

Sample Lo1

A general feature of the phase diagram in Fig. 9a is that ilmenite is stable at pressures below c. 7·5 kbar, whereas rutile is stable at higher P. Thus, the presence of ilmenite in garnet zones 1 and 2 is consistent with prograde garnet formation at P < 7·5 kbar. Diagrams contoured with isopleths for the garnet mode, Xgrs, Xalm and Xspss (Fig. 10a–f) indicate that the observed garnet growth zonation could be explained by a nearly isobaric temperature rise from about 540 to 650°C at P = 7–7·5 kbar. Along this path (L1–L2, Fig. 9a), garnet will be formed, Xgrs will decrease from 18 to 4 mol % and Xspss from 5 to 1·7 mol %, whereas Xalm will first increase and subsequently decrease (71 → 80 → 77 mol %; Fig. 10a, d, e and f). Furthermore, XFe of staurolite will decrease (from 95 to 85 mol %) and Xan in plagioclase will decrease (Fig. 10c). These trends are in good agreement with those observed in sample Lo1. In the case of a prograde pressure increase the Xgrs zonation would be much less pronounced, and the Xspss contents in the garnet core much higher. The phase diagram in Fig. 9a can also be used to explain the observed peak assemblage (L2): ky–grt–st–bt–ms– pl–qtz–ilm–rt at P ≈ 7·5 kbar and T ≈ 650°C, which are within error of the PT conditions estimated above. The phase diagram predicts that biotite mode will go to zero at conditions near the metamorphic peak. This agrees with thin-section observations, showing that only a little biotite occurs in the matrix of sample Lo1, whereas the majority of biotite was formed during garnet resorption. This observation and the formation of matrix sillimanite is consistent with a subsequent P decrease and fall in T, leading to the formation of assemblage L3: sil–grt–st–bt–ms–pl–qtz. The phase diagram predicts that along the retrograde path (L2–L3; Fig. 9a) garnet will be resorbed, whereas biotite, sillimanite and plagioclase will be formed (Fig. 10a and b). The Tivi content in biotite will decrease and Xan will increase toward the plagioclase rim. All these features are in agreement with those observed in sample Lo1 (Fig. 3).

Fig. 10.

Contoured PT pseudosections for sample Lo1 (a–f) and sample Me3/3 (g–l). (a, g) Garnet mode; (b, h) plagioclase mode (mole proportions with Qtz and H2O in excess); (c, i) Xan; (d, j) Xgrs; (e, k) Xalm; (f, l) Xspss.

Fig. 10.

Contoured PT pseudosections for sample Lo1 (a–f) and sample Me3/3 (g–l). (a, g) Garnet mode; (b, h) plagioclase mode (mole proportions with Qtz and H2O in excess); (c, i) Xan; (d, j) Xgrs; (e, k) Xalm; (f, l) Xspss.

Sample Me3/3

Muscovite, biotite and rutile inclusions in plagioclase cores (Fig. 5a and b) and primary biotite and rutile inclusions in garnet (Fig. 4a) indicate that the garnet in sample Me3/3 was progressively formed in the assemblage (M1): bt–ms–grt–st–pl–rt, which requires P > 7 kbar and T = 600–640°C (Fig. 9b). The observations that kyanite in sample Me3/3 contains only rutile but no ilmenite, and that garnet is overgrown by kyanite (Fig. 4b), indicate that final garnet growth took place in the peak metamorphic assemblage (M2): ky–bt–(ms)–grt–st–pl–rt. Assemblage (M2) requires PT conditions of c. 7 kbar and 650°C (Fig. 9b), which are in agreement with the PT estimates made above. The phase diagram in Fig. 9b also predicts the retrograde assemblage (M3): sil–bt–grt–st–crd– pl–ilm (Fig. 5c–f), which is stable at P = 3–5 kbar and T = 560–610°C. During a change from (M2) to (M3), garnet will be resorbed and cordierite and plagioclase with higher Xan formed (Fig. 10g–i). Furthermore, Xalm and Xspss of garnet and Alvi of biotite will increase, whereas XFe and Tivi of biotite decrease (Fig. 10k and l). These predictions are consistent with the observed garnet resorption and diffusion patterns and with the composition of biotite measured in open garnet fractures (Fig. 6). However, the phase diagrams are unable to explain the steep Xgrs increase observed at the transition between zones 1 and 2 of garnet Ky3/3 (Fig. 6a). According to the predictions, garnet will be formed during a temperature rise from assemblage (M1) into (M2), whereas Xgrs should decrease gently from core to rim (Fig. 10g and j). It seems likely that the steep Xgrs increase indicates a significant gap in the garnet growth history (see Discussion).

In summary, we infer that rocks from Lord Nunatak underwent isobaric heating from T = 540 to 650°C at P = 7–7·5 kbar, followed by a decompressive retrograde PT path to T < 640°C at P < 6 kbar. Rocks from Meade Nunatak were metamorphosed at similar peak PT conditions (assemblage M2) and underwent a retrograde metamorphic stage at c. 550–610°C at 3–5 kbar (assemblage M3). From the agreement between the observed and calculated assemblages and the PT estimates it would be reasonable to conclude that rocks from Meade and Lord nunataks underwent similar PT evolution during a single metamorphic cycle (Fig. 9a and b). Problems with this interpretation include garnet in sample Lo1 preserving a prograde growth zonation, whereas garnet in sample Me3/3 shows evidence for a strong diffusive modification. If both samples underwent the same prograde history, garnet of sample Me3/3 should also show a growth zonation characterized by an increase of Xalm and decrease of Xgrs from core to rim as predicted by the phase diagrams (Fig. 10g–l). Furthermore, the steep Xgrs increase at the transition between zones 1 and 2 of garnet in sample Me3/3 provides evidence for polyphase garnet growth (Fig. 6a). A possible explanation for the observed differences could be that sample Me3/3 was affected by at least two distinct metamorphic events.

GEOCHRONOLOGY

To clarify the timing of metamorphism within the individual samples from Lord and Meade nunataks, and to prove or disprove the speculation made at the end of the last section, geochronological data were obtained employing four different methods. Monazite and zircon grains were dated by conventional dilution U–Pb TIMS analyses, and laser-ablation plasma ionization multicollector mass spectrometry (LA-PIMMS). Sm–Nd and Rb–Sr isochrons were also obtained. The analytical techniques are described in Appendices B–D and the results are presented in Tables 47 and Figs 1113.

Table 4:

Results of U–Pb TIMS analysis of monazite and zircon

Mineral (no. of grains) Weight (µg) U (ppm) Pb (ppm) Total PbC 206Pb/204Pb (±2σ%) 207Pb/235U (±2σ%) 206Pb/238U (±2σ%) 207Pb/206Pb (±2σ%) Age (Ma)
 
  
 
 

 

 

 

 

 

 

 
206Pb/238U
 
207Pb/235U
 
207Pb/206Pb
 
Sample Me3/3 (Meade Nunatak)            
Monazite (6) 5·8 3460 501 11 16849·9 (0·25) 0·71136 (0·21) 0·08529 (0·202) 0·06049 (0·056) 527·6 ± 1·0 545·6 ± 0·9 621·1 ± 1·2 
Monazite (4) 1·7 5547 1057 10 9506·0 (0·27) 0·64631 (0·19) 0·08143 (0·171) 0·05756 (0·071) 504·7 ± 0·8 506·2 ± 0·7 513·2 ± 1·6 
Monazite (3) 2·3 4298 888 15936·7 (0·98) 0·64323 (0·25) 0·08130 (0·222) 0·05738 (0·106) 503·9 ± 1·1 504·3 ± 1·0 506·2 ± 2·3 
Zircon (1) 1·7 1617 388 42 1058·1 (0·55) 3·25519 (0·54) 0·23851 (0·506) 0·09899 (0·179) 1378·9 ± 6·3 1470·4 ± 4·2 1605·1 ± 3·3 
Zircon (1) 1·1 449 118 1980·6 (0·22) 3·75340 (0·40) 0·26831 (0·391) 0·10146 (0·082) 1532·2 ± 5·3 1582·8 ± 3·2 1651·0 ± 1·5 
Zircon (1) 1·1 448 117 27 309·5 (0·72) 3·09003 (0·62) 0·22787 (0·362) 0·09835 (0·468) 1323·3 ± 4·3 1430·2 ± 4·8 1593·1 ± 8·7 
Sample Lo1 (Lord Nunatak)            
Monazite (2) 1·5 15405 6383 27 5483·7 (0·13) 0·65977 (0·18) 0·08308 (0·174) 0·05760 (0·058) 514·5 ± 0·9 514·5 ± 0·7 514·6 ± 1·3 
Monazite (6) 7·2 8485 3837 62 5561·9 (0·16) 0·65764 (0·18) 0·08284 (0·172) 0·05757 (0·055) 513·1 ± 0·8 513·2 ± 0·7 513·6 ± 1·2 
Monazite (8) 9·0 6731 2980 74 4531·5 (0·23) 0·65840 (0·18) 0·08289 (0·175) 0·05761 (0·058) 513·4 ± 0·9 513·6 ± 0·7 514·8 ± 1·3 
Mineral (no. of grains) Weight (µg) U (ppm) Pb (ppm) Total PbC 206Pb/204Pb (±2σ%) 207Pb/235U (±2σ%) 206Pb/238U (±2σ%) 207Pb/206Pb (±2σ%) Age (Ma)
 
  
 
 

 

 

 

 

 

 

 
206Pb/238U
 
207Pb/235U
 
207Pb/206Pb
 
Sample Me3/3 (Meade Nunatak)            
Monazite (6) 5·8 3460 501 11 16849·9 (0·25) 0·71136 (0·21) 0·08529 (0·202) 0·06049 (0·056) 527·6 ± 1·0 545·6 ± 0·9 621·1 ± 1·2 
Monazite (4) 1·7 5547 1057 10 9506·0 (0·27) 0·64631 (0·19) 0·08143 (0·171) 0·05756 (0·071) 504·7 ± 0·8 506·2 ± 0·7 513·2 ± 1·6 
Monazite (3) 2·3 4298 888 15936·7 (0·98) 0·64323 (0·25) 0·08130 (0·222) 0·05738 (0·106) 503·9 ± 1·1 504·3 ± 1·0 506·2 ± 2·3 
Zircon (1) 1·7 1617 388 42 1058·1 (0·55) 3·25519 (0·54) 0·23851 (0·506) 0·09899 (0·179) 1378·9 ± 6·3 1470·4 ± 4·2 1605·1 ± 3·3 
Zircon (1) 1·1 449 118 1980·6 (0·22) 3·75340 (0·40) 0·26831 (0·391) 0·10146 (0·082) 1532·2 ± 5·3 1582·8 ± 3·2 1651·0 ± 1·5 
Zircon (1) 1·1 448 117 27 309·5 (0·72) 3·09003 (0·62) 0·22787 (0·362) 0·09835 (0·468) 1323·3 ± 4·3 1430·2 ± 4·8 1593·1 ± 8·7 
Sample Lo1 (Lord Nunatak)            
Monazite (2) 1·5 15405 6383 27 5483·7 (0·13) 0·65977 (0·18) 0·08308 (0·174) 0·05760 (0·058) 514·5 ± 0·9 514·5 ± 0·7 514·6 ± 1·3 
Monazite (6) 7·2 8485 3837 62 5561·9 (0·16) 0·65764 (0·18) 0·08284 (0·172) 0·05757 (0·055) 513·1 ± 0·8 513·2 ± 0·7 513·6 ± 1·2 
Monazite (8) 9·0 6731 2980 74 4531·5 (0·23) 0·65840 (0·18) 0·08289 (0·175) 0·05761 (0·058) 513·4 ± 0·9 513·6 ± 0·7 514·8 ± 1·3 
Table 5:

Results of U–Pb LA-PIMMS analysis of zircon and monazite from sample Me3/3

Grain 206Pb*/238U (±2σ%) 207Pb*/235U (±2σ%) 207Pb*/206Pb* (±2σ%) Age (Ma) (±2σ%)
 
  
 
 

 

 
206Pb/238U
 
207Pb/235U
 
207Pb/206Pb
 
Domain DM1: matrix monazite       
mnz1-bt 0·0970 (4·08) 0·7540 (4·29) 0·0564 (1·31) 597 ± 24 571 ± 24 468 ± 24 
mnz5-sil 0·0904 (4·11) 0·7226 (4·28) 0·0580 (1·21) 558 ± 22 552 ± 25 529 ± 7 
mnz12-bt 0·3224 (4·24) 4·6708 (4·40) 0·1051 (1·20) 1801 ± 76 1762 ± 77 1716 ± 6 
mnz14-cd 0·0847 (4·10) 0·6733 (4·28) 0·0577 (1·23) 524 ± 22 523 ± 22 517 ± 14 
mnz15-bt 0·0927 (4·74) 0·7298 (4·90) 0·0571 (1·22) 571 ± 28 556 ± 28 496 ± 12 
Domain DM2: monazite completely enclosed by kyanite       
mnz3 0·0896 (3·77) 0·7106 (3·96) 0·0575 (1·21) 553 ± 20 545 ± 22 512 ± 10 
mnz7 0·0947 (4·44) 0·7519 (4·61) 0·0576 (1·23) 584 ± 26 569 ± 26 513 ± 13 
mnz8 0·0942 (5·39) 0·7399 (5·54) 0·0569 (1·30) 581 ± 32 562 ± 32 489 ± 22 
Domain DM3: monazite partly enclosed by staurolite       
mnz4 0·0917 (3·81) 0·7281 (4·00) 0·0576 (1·21) 565 ± 22 555 ± 22 514 ± 9 
Domain DM4: monazite completely enclosed by garnet       
mnz17 0·2863 (4·18) 4·1331 (4·38) 0·1047 (1·29) 1623 ± 68 1661 ± 73 1709 ± 18 
mnz36 0·2508 (3·71) 3·5769 (3·91) 0·1034 (1·22) 1443 ± 54 1544 ± 60 1687 ± 10 
Domain DM5: monazite associated with retrograde biotite along garnet fractures       
mnz32 0·0837 (4·23) 0·6677 (4·45) 0·0578 (1·37) 518 ± 22 519 ± 24 524 ± 29 
mnz33 0·0730 (4·24) 0·5677 (4·60) 0·0564 (1·79) 454 ± 20 456 ± 20 467 ± 57 
mnz13 0·0908 (3·97) 0·7185 (4·15) 0·0574 (1·21) 560 ± 22 550 ± 22 507 ± 10 
Zircon (in thin section)       
zrc3-ky 0·2914 (2·29) 4·1582 (2·39) 0·1035 (1·37) 1649 ± 75 1666 ± 80 1688 ± 28 
zrc5-ky 0·2307 (2·31) 3·2356 (2·41) 0·1017 (1·35) 1338 ± 63 1466 ± 76 1656 ± 26 
zrc6-grt 0·1856 (2·35) 2·5213 (2·43) 0·0985 (1·27) 1098 ± 52 1278 ± 62 1596 ± 20 
zrc7-cd 0·2781 (1·86) 3·9386 (1·96) 0·1027 (1·23) 1582 ± 58 1622 ± 63 1674 ± 18 
zrc8-bt 0·2273 (2·00) 3·1255 (2·10) 0·0997 (1·28) 1320 ± 52 1439 ± 60 1619 ± 22 
zrc9-bt 0·2700 (2·38) 3·8670 (2·48) 0·1039 (1·39) 1541 ± 74 1607 ± 80 1695 ± 30 
Zircon (mount)       
zrc12 0·2016 (1·81) 2·7731 (1·94) 0·0998 (1·37) 1184 ± 43 1348 ± 51 1620 ± 28 
zrc13 0·3013 (2·15) 4·2769 (2·27) 0·1029 (1·48) 1698 ± 75 1689 ± 76 1678 ± 35 
zrc15 0·2494 (2·70) 3·4959 (2·77) 0·1017 (1·24) 1435 ± 78 1526 ± 84 1655 ± 19 
zrc14 0·2963 (2·20) 4·2192 (2·32) 0·1033 (1·46) 1673 ± 75 1678 ± 78 1684 ± 34 
zrc16 0·2722 (1·97) 3·8585 (2·07) 0·1028 (1·24) 1552 ± 62 1605 ± 67 1675 ± 18 
zrc17 0·2856 (1·86) 4·0589 (1·97) 0·1031 (1·28) 1619 ± 60 1646 ± 63 1681 ± 21 
zrc19 0·2655 (2·10) 3·7652 (2·19) 0·1029 (1·26) 1518 ± 64 1585 ± 70 1676 ± 21 
zrc20 0·2677 (2·05) 3·7802 (2·15) 0·1024 (1·29) 1529 ± 63 1589 ± 68 1668 ± 22 
Grain 206Pb*/238U (±2σ%) 207Pb*/235U (±2σ%) 207Pb*/206Pb* (±2σ%) Age (Ma) (±2σ%)
 
  
 
 

 

 
206Pb/238U
 
207Pb/235U
 
207Pb/206Pb
 
Domain DM1: matrix monazite       
mnz1-bt 0·0970 (4·08) 0·7540 (4·29) 0·0564 (1·31) 597 ± 24 571 ± 24 468 ± 24 
mnz5-sil 0·0904 (4·11) 0·7226 (4·28) 0·0580 (1·21) 558 ± 22 552 ± 25 529 ± 7 
mnz12-bt 0·3224 (4·24) 4·6708 (4·40) 0·1051 (1·20) 1801 ± 76 1762 ± 77 1716 ± 6 
mnz14-cd 0·0847 (4·10) 0·6733 (4·28) 0·0577 (1·23) 524 ± 22 523 ± 22 517 ± 14 
mnz15-bt 0·0927 (4·74) 0·7298 (4·90) 0·0571 (1·22) 571 ± 28 556 ± 28 496 ± 12 
Domain DM2: monazite completely enclosed by kyanite       
mnz3 0·0896 (3·77) 0·7106 (3·96) 0·0575 (1·21) 553 ± 20 545 ± 22 512 ± 10 
mnz7 0·0947 (4·44) 0·7519 (4·61) 0·0576 (1·23) 584 ± 26 569 ± 26 513 ± 13 
mnz8 0·0942 (5·39) 0·7399 (5·54) 0·0569 (1·30) 581 ± 32 562 ± 32 489 ± 22 
Domain DM3: monazite partly enclosed by staurolite       
mnz4 0·0917 (3·81) 0·7281 (4·00) 0·0576 (1·21) 565 ± 22 555 ± 22 514 ± 9 
Domain DM4: monazite completely enclosed by garnet       
mnz17 0·2863 (4·18) 4·1331 (4·38) 0·1047 (1·29) 1623 ± 68 1661 ± 73 1709 ± 18 
mnz36 0·2508 (3·71) 3·5769 (3·91) 0·1034 (1·22) 1443 ± 54 1544 ± 60 1687 ± 10 
Domain DM5: monazite associated with retrograde biotite along garnet fractures       
mnz32 0·0837 (4·23) 0·6677 (4·45) 0·0578 (1·37) 518 ± 22 519 ± 24 524 ± 29 
mnz33 0·0730 (4·24) 0·5677 (4·60) 0·0564 (1·79) 454 ± 20 456 ± 20 467 ± 57 
mnz13 0·0908 (3·97) 0·7185 (4·15) 0·0574 (1·21) 560 ± 22 550 ± 22 507 ± 10 
Zircon (in thin section)       
zrc3-ky 0·2914 (2·29) 4·1582 (2·39) 0·1035 (1·37) 1649 ± 75 1666 ± 80 1688 ± 28 
zrc5-ky 0·2307 (2·31) 3·2356 (2·41) 0·1017 (1·35) 1338 ± 63 1466 ± 76 1656 ± 26 
zrc6-grt 0·1856 (2·35) 2·5213 (2·43) 0·0985 (1·27) 1098 ± 52 1278 ± 62 1596 ± 20 
zrc7-cd 0·2781 (1·86) 3·9386 (1·96) 0·1027 (1·23) 1582 ± 58 1622 ± 63 1674 ± 18 
zrc8-bt 0·2273 (2·00) 3·1255 (2·10) 0·0997 (1·28) 1320 ± 52 1439 ± 60 1619 ± 22 
zrc9-bt 0·2700 (2·38) 3·8670 (2·48) 0·1039 (1·39) 1541 ± 74 1607 ± 80 1695 ± 30 
Zircon (mount)       
zrc12 0·2016 (1·81) 2·7731 (1·94) 0·0998 (1·37) 1184 ± 43 1348 ± 51 1620 ± 28 
zrc13 0·3013 (2·15) 4·2769 (2·27) 0·1029 (1·48) 1698 ± 75 1689 ± 76 1678 ± 35 
zrc15 0·2494 (2·70) 3·4959 (2·77) 0·1017 (1·24) 1435 ± 78 1526 ± 84 1655 ± 19 
zrc14 0·2963 (2·20) 4·2192 (2·32) 0·1033 (1·46) 1673 ± 75 1678 ± 78 1684 ± 34 
zrc16 0·2722 (1·97) 3·8585 (2·07) 0·1028 (1·24) 1552 ± 62 1605 ± 67 1675 ± 18 
zrc17 0·2856 (1·86) 4·0589 (1·97) 0·1031 (1·28) 1619 ± 60 1646 ± 63 1681 ± 21 
zrc19 0·2655 (2·10) 3·7652 (2·19) 0·1029 (1·26) 1518 ± 64 1585 ± 70 1676 ± 21 
zrc20 0·2677 (2·05) 3·7802 (2·15) 0·1024 (1·29) 1529 ± 63 1589 ± 68 1668 ± 22 

(mnz, zrc)-bt, sil, cd, grt, ky indicates monazite or zircon enclosed by biotite, sillimanite, cordierite, garnet or kyanite.

Table 6:

Results of Sm–Nd analyses

Sample Mineral
 
Sm (ppm)
 
Nd (ppm)
 
147Sm/144Nd
 
143Nd/144Nd
 
TDM
 
Me1/1 staurolite 1·84 9·72 0·11462 0·51128  
 biotite 1·87 9·95 0·11360 0·51122  
 WR 5·29 24·87 0·12844 0·51132 2909 
 garnet 1·44 1·54 0·56516 0·51589  
Me3/3 garnet 1·23 2·66 0·28048 0·51297  
 WR 7·44 40·44 0·11112 0·51116 2667 
 biotite 1·27 7·28 0·10532 0·51121  
 monazite 16117·4 90892·8 0·10717 0·51118  
Lo1 garnet 2·78 0·99 1·70507 0·51712  
 WR 6·58 33·37 0·11924 0·51174 2037 
 staurolite 4·39 28·71 0·09238 0·51168  
 muscovite 1·15 7·51 0·09276 0·51174  
Sample Mineral
 
Sm (ppm)
 
Nd (ppm)
 
147Sm/144Nd
 
143Nd/144Nd
 
TDM
 
Me1/1 staurolite 1·84 9·72 0·11462 0·51128  
 biotite 1·87 9·95 0·11360 0·51122  
 WR 5·29 24·87 0·12844 0·51132 2909 
 garnet 1·44 1·54 0·56516 0·51589  
Me3/3 garnet 1·23 2·66 0·28048 0·51297  
 WR 7·44 40·44 0·11112 0·51116 2667 
 biotite 1·27 7·28 0·10532 0·51121  
 monazite 16117·4 90892·8 0·10717 0·51118  
Lo1 garnet 2·78 0·99 1·70507 0·51712  
 WR 6·58 33·37 0·11924 0·51174 2037 
 staurolite 4·39 28·71 0·09238 0·51168  
 muscovite 1·15 7·51 0·09276 0·51174  

WR, whole rock.

Table 7:

Results of Rb–Sr analyses

Mineral Rb
 
Sr
 
Rb/Sr
 
87Rb/86Sr
 
87Sr/86Sr
 
±
 
Age (Ma)
 
Sample Me3/3        
wr 55·81 33·52 1·6649 4·8848 0·8517 10 503·5 ± 0·5 
bt 268·72 3·46 77·5822 269·2853 2·7488 50·8  
Sample Lo1        
wr 185·67 341·43 0·5438 1·5746 0·7155 499 ± 12 
ms 316·70 352·95 0·8973 2·6000 0·7228  
Mineral Rb
 
Sr
 
Rb/Sr
 
87Rb/86Sr
 
87Sr/86Sr
 
±
 
Age (Ma)
 
Sample Me3/3        
wr 55·81 33·52 1·6649 4·8848 0·8517 10 503·5 ± 0·5 
bt 268·72 3·46 77·5822 269·2853 2·7488 50·8  
Sample Lo1        
wr 185·67 341·43 0·5438 1·5746 0·7155 499 ± 12 
ms 316·70 352·95 0·8973 2·6000 0·7228  
Fig. 11.

Concordia diagrams showing the results of U–Pb TIMS dating of monazite (a, b) and zircon (c) from sample Lo1 (a), and sample Me3/3 (b, c). (d) Discordia of all zircon and monazite analyses from sample Me3/3.

Fig. 11.

Concordia diagrams showing the results of U–Pb TIMS dating of monazite (a, b) and zircon (c) from sample Lo1 (a), and sample Me3/3 (b, c). (d) Discordia of all zircon and monazite analyses from sample Me3/3.

Fig. 12.

Concordia diagrams showing the results of U–Pb LA-PIMMS dating of monazite (a, b) and zircon (c) from sample Me3/3. (a) Results of monazite analyses measured in matrix biotite, cordierite, sillimanite, and of monazite completely enclosed by kyanite and in open garnet fractures. (b) Results of monazite analyses measured in a matrix biotite, and completely enclosed by garnet.

Fig. 12.

Concordia diagrams showing the results of U–Pb LA-PIMMS dating of monazite (a, b) and zircon (c) from sample Me3/3. (a) Results of monazite analyses measured in matrix biotite, cordierite, sillimanite, and of monazite completely enclosed by kyanite and in open garnet fractures. (b) Results of monazite analyses measured in a matrix biotite, and completely enclosed by garnet.

Fig. 13.

Sm–Nd results for samples Lo1, Me3/3 and Me1/1.

Fig. 13.

Sm–Nd results for samples Lo1, Me3/3 and Me1/1.

Isotope dilution U–Pb TIMS monazite and zircon dating

Three monazite fractions analysed from sample Lo1 yielded a concordant age of 514 ± 1 Ma (Fig. 11a). In contrast, three monazite fractions from sample Me3/3 fall on a discordia with a lower intercept at 503 ± 1 Ma and an upper intercept at 1722 ± 54 Ma [mean square weighted deviation (MSWD) = 0·59; Fig. 11b]. The discordance of the analyses indicates that the three monazite fractions contained different amounts of inherited radiogenic Pb, which probably results from a mixture between two distinct monazite generations, an older one formed at c. 1700 Ma and a younger formed at c. 500 Ma.

Analyses of three single zircon grains from sample Me3/3 yield a similar discordia, with a lower intercept at 479 ± 450 Ma and an upper intercept at 1683 ± 96 Ma (MSWD = 2·2; Fig. 11c). The discordance could be interpreted to result from radiogenic Pb loss at c. 500 Ma. This is supported by the fact that all monazite and zircon analyses from sample Me3/3 plot on a discordia with a lower intercept at 502·5 ± 1 Ma, and an upper intercept at 1686 ± 2 Ma (MSWD = 0·67; Fig. 11d). The meaning of the c. 1700 Ma zircon age is discussed below.

In situ LA-PIMMS monazite and zircon dating

Monazite

Monazite grains of sample Me3/3 were analysed in situ with LA-PIMMS from the following domains:

  • DM1, monazite enclosed by or associated with matrix biotite, cordierite and sillimanite (Fig. 8e and f);

  • DM2, monazite completely enclosed by kyanite (Fig. 8d);

  • DM3, monazite partly enclosed by staurolite (Fig. 8c);

  • DM4, monazite completely enclosed by garnet 1 (Fig. 8a);

  • DM5, monazite in garnet fractures and patches filled with retrograde biotite (Fig. 8a and b).

Four monazite analyses from the matrix (domain DM1) yielded 207Pb/206Pb ages between 468 ± 24 and 529 ± 8 Ma, whereas one grain (mnz 12, Table 5) gave a much older age of 1716 ± 6 Ma. Three monazite grains enclosed in kyanite (domain DM2) yielded 207Pb/206Pb ages between 489 ± 22 and 513 ± 14 Ma, and a monazite grain partly enclosed in staurolite (domain DM3) gave an age of 514 ± 9 Ma. Similar 207Pb/206Pb ages between 467 ± 58 and 524 ± 30 Ma were obtained from monazites, which occur together with biotite in retrograde garnet fractures and patches (domain DM5). Two monazite grains of c. 20 µm size, completely enclosed by garnet 1 (domain DM4), gave 207Pb/206Pb ages between 1687 ± 10 and 1709 ± 18 Ma (Table 5, Fig. 12b). A regression forced through all younger monazite analyses yields a discordia with an upper intercept age of 501 ± 28 Ma (MSWD = 2·1; Fig. 12a), within error of the U–Pb TIMS monazite age presented above. In general, the age data obtained by in situ LA-PIMMS analyses indicate monazite formation at c. 500 Ma and c. 1700 Ma, and confirm the results of U–Pb TIMS dating. The data also show that older monazite grains not only occur armoured in garnet (grains mnz 32, 36), but also enclosed by matrix biotite (grain mnz 12). Furthermore, the data indicate that garnet resorption, as well as kyanite, cordierite and sillimanite formation, took place during a metamorphic event at c. 500 Ma.

Zircon

Eight zircon grains from sample Me3/3 were analysed from a prepared grain mount, and six occluded in different minerals of one thin section (Table 5). Some zircon analyses are concordant (e.g. grains zrc13, 14; Table 5) and yielded ages of c. 1690 Ma; some analyses are discordant, and fall on a discordia, with an upper intercept at 1689 ± 11 Ma and a lower intercept at 261 ± 59 Ma (MSWD = 0·67; Fig. 12c). The upper and lower intercepts are within error of those obtained by zircon TIMS dating (Fig. 11c). The upper intercept age is also within error of that obtained from the monazite TIMS and LA-PIMMS analyses (Figs 11b and 12b). The geological significance of these age data is discussed below.

Sm–Nd and Rb–Sr results

Sm–Nd isotope analyses of garnet, whole-rock, staurolite and muscovite from sample Lo1 scatter about a reference line indicating an age of 514 ± 21 Ma (MSWD = 34; Fig. 13), and the whole-rock analysis gives an average crustal residence age (TDM) of 2037 Ma (Table 6). The high MSWD of the isochron results from staurolite and muscovite analyses, which plot above the regression defined by garnet and whole rock. This indicates that the matrix minerals were not in a perfect isotopic equilibrium with garnet. If only garnet and whole-rock analyses are used an isochron age of 518 ± 5 Ma is obtained. This age is identical, within error, to the 514 ± 1 Ma U–Pb monazite TIMS age, consistent with garnet and monazite formation in sample Lo1 during the same stage of the metamorphic history. As the prograde growth zonation of garnet in sample Lo1 is nearly unaffected by retrograde diffusion (Fig. 3b), we interpret the Sm–Nd age of 518 ± 5 Ma to represent the integrated time of prograde garnet formation. Rb–Sr muscovite– whole-rock analyses yielded an age of 499 ± 12 Ma (Table 7), which is slightly younger than the Sm–Nd garnet and the U–Pb monazite ages. The Rb–Sr age is interpreted as the final cooling of sample Lo1 below c. 500°C, the suggested Rb–Sr muscovite closure temperature (Jäger, 1979).

From samples Me3/3 and Me1/1, Sm–Nd analyses were obtained from garnet, whole-rock, biotite, staurolite and monazite fractions (Table 6). The Sm–Nd analyses indicate ages of 1578 ± 230 Ma (MSWD = 11·3) for sample Me3/3 and 1574 ± 90 Ma (MSWD = 11·5) for sample Me1/1, which are within error identical to or barely younger than the c. 1700 Ma zircon and monazite age data presented above. A single regression defined by all analyses yields an age of 1571 ± 40 Ma (MSWD = 8·3) for all analysed minerals, or 1590 ± 66 Ma (MSWD = 4·9) if only garnet and whole-rock analyses are used (Fig. 13). Biotite, staurolite and monazite plot slightly above the regression line (Fig. 13) defined by garnet and whole rock, indicating that the matrix minerals are not in perfect isotopic equilibrium.

The fact that the analyses of both samples (Me1/1 and Me3/3) gave identical age results gives a clue that the 1571 ± 40 Ma age could have geological significance. This age is consistent with the average crustal residence ages (TDM) of 2909 and 2667 Ma obtained from whole rock of samples Me1/1 and Me3/3 (Table 6), respectively, with the presence of c. 1700 Ma monazite and zircon inclusions enclosed by garnet, and with an Rb–Sr biotite–whole-rock cooling age of 503·5 ± 0·5 Ma obtained from sample Me3/3 (Table 7). However, the garnet analyses of both samples have relatively low 147Sm/144Nd ratios (0·28 and 0·57), and relatively high Nd contents, between 0·99 and 2·7 ppm (Table 6). This might indicate that the analysed garnet fractions contained tiny REE-rich inclusions, which could have influenced the isochron slope and thus the age information. Indeed, monazite inclusions were detected by SEM in garnet Me3/3 (Fig. 8a), but not in garnet Me1/1. Additionally, zircon and apatite inclusions were observed (Fig. 4b). As the analysed monazite fraction of sample Me3/3 has a 147Sm/144Nd ratio similar to that of whole rock (Table 6), we conclude that monazite inclusions in garnet can have only a minor effect on the slope of the isochron, but might be responsible for the low 147Sm/144Nd ratio of the garnet fraction. A serious bias of the Sm–Nd isochron slope by zircon inclusions in garnet also seems to be less likely. In fact, zircon inclusions are abundantly enclosed by staurolite and biotite, as well as by garnet (observed by microscopy and EDX). As both the staurolite and biotite fractions gave low 147Sm/144Nd ratios, similar to the whole rock (Fig. 13), potential zircon inclusions in garnet cannot have caused a significant shift of the isochron slope; neither can they account for the observed 147Sm/144Nd ratios of the garnet fraction. Finally, apatite inclusions are a potential source for the measured 147Sm/144Nd ratios of the garnet fractions (Table 6). However, if this is the case, then these inclusions must have been overgrown by garnet during the metamorphic event (M1), which took place at T ≈ 600–640°C. This temperature is higher than the Sm–Nd closure temperature for tiny apatite grains (<10 µm, Cherniak, 2000). This implies that matrix apatite, not armoured by garnet, would have been reset prior to a garnet overgrowth event at 500 Ma. This is not seen.

In summary, the obtained Sm–Nd ages of 1571 ± 40 Ma from samples Me1/1 and Me3/3 might reflect the time of garnet formation and/or cooling below the Sm–Nd garnet blocking temperature during the Proterozoic. The absence of a prograde growth zonation pattern in garnet samples Me1/1 and Me3/3 testifies to cooling below the isotopic closure temperature, rather than to new garnet formation at 1571 ± 40 Ma. Nevertheless, a slight age rejuvenation associated with new garnet formation along microfractures at about 500 Ma (spiderweb texture, Fig. 4d) cannot be completely excluded. Thus, it seems most likely that the 1571 ± 40 Ma age represents a minimum age for garnet zone 1 formation.

DISCUSSION AND CONCLUSIONS

Lord Nunatak (sample Lo1)

Petrological results indicate that metapelitic rocks from Lord Nunatak underwent a prograde evolution from T ≈ 540°C, P ≈ 7 kbar to T ≈ 650°C, P ≈ 7·5 kbar, followed by decompression and cooling to T < 650°C, P < 6·0 kbar. Textural relationships indicate deformation and metamorphism during prograde garnet growth (D1), and subsequently during the retrograde evolution (D2) leading to garnet erosion and the observed foliation. The well-preserved prograde garnet growth zonation and the very tiny diffusion zone at the garnet rim indicate that sample Lo1 remained at the metamorphic peak for only a short time and underwent fast cooling. This interpretation is supported by the Sm–Nd garnet–whole-rock, U–Pb monazite, and Rb–Sr muscovite–whole-rock ages of 518 ± 5 Ma, 514 ± 1 Ma, and 499 ± 12 Ma, respectively. The Sm–Nd age is interpreted as the integral time of prograde garnet growth, the U–Pb monazite age as the peak of metamorphism, and the Rb–Sr muscovite–whole-rock age as the final cooling below 500°C. On the basis of these ages, the linear cooling rate from the metamorphic peak at 650°C to below 500°C was c. 10°C/Ma. The combined geochronological and petrological data from sample Lo1 indicate that rocks from Lord Nunatak were metamorphosed during a single orogenic cycle between 525 and 500 Ma, during the Ross Orogeny. The sedimentary protolith of the metapelites was derived from a provenance with an average crustal residence age of TDM = 2037 Ma, possibly within the East Antarctic Craton.

Meade Nunatak (samples Me1/1 and Me3/3)

The interpretation of the petrological and geochronological data obtained from rocks of Meade Nunatak is more complex than for rocks from the Lord Nunatak. Petrological information indicates that Meade Nunatak samples underwent an early metamorphic event at T = 600–640°C at P ≈ 7 kbar (M1) followed by a peak metamorphic event at T ≈ 650°C, P ≈ 7 kbar (M2), and a final stage at T = 560–610°C, P = 3–5 kbar (M3). Between M2 and M3 the rocks were affected by deformation, which caused garnet erosion and led to the formation of a foliation, correlated with D2 in sample Lo1.

The geochronological results provide evidence that the sedimentary protolith of the Meade Nunatak metapelites was derived from rocks with average crustal residence ages (TDM) between 2909 and 2667 Ma. Furthermore, they point to events at 1700 Ma, >1570 Ma and c. 500 Ma. For the interpretation of these ages the following key observations must be taken into account: On the basis of points (1)–(4), (6) and (7), it can be argued that the first garnet generation and the older monazite grains were formed together during a Barrovian-type metamorphic event at c. 1700 Ga, leading to the formation of assemblage M1. Subsequently, the garnet growth zonation was smoothed out by volume diffusion and garnet rims were altered by diffusive Fe, Mg, Mn exchange with surrounding biotite. In this context the 1571 ± 40 Ma Sm–Nd garnet age can be interpreted as the minimum time of garnet cooling below the Sm–Nd isotopic closure temperature, which is thought to be c. 600°C for garnet that underwent moderate cooling of about 10°C/Ma (Mezger et al., 1992). The time gap between garnet formation and cooling may be the product of very slow cooling, possibly the result of samples Me3/3 and Me1/1 remaining at great depths and thus above the Sm–Nd closure temperature after the metamorphic peak. Slow cooling after 1700 Ma is supported by the fact that a potential prograde garnet growth zonation is completely smoothed out [point (2)].

  1. there is evidence for a polyphase garnet growth in sample Me3/3 (Figs 4b and 6a);

  2. garnet grains show a diffusion zonation;

  3. there are two generations of garnet fractures; older annealed Xgrs-rich veins and younger open fractures filled with retrograde biotite and chlorite;

  4. monazite grains enclosed by garnet and some matrix monazites yielded ages of c. 1700 Ma;

  5. round, facet-free altered zircon grains also yield ages of c. 1700 Ma;

  6. monazite inclusions in kyanite, cordierite, and biotite and along garnet fractures yield ages of c. 500 Ma (Fig. 8);

  7. Sm–Nd analyses of two distinct samples both gave an age of c. 1570 (±90–270) Ma.

During a second orogenic cycle, tiny euhedral rims of a second garnet generation with higher Xgrs were formed around garnet 1. This second garnet generation is overgrown by kyanite in the peak metamorphic assemblage M2. Formation of assemblage M2 occurred during the Ross Orogeny, as well defined by the c. 500 Ma ages obtained from monazite grains completely enclosed by kyanite (Table 5). It is likely that the Xgrs-rich veins (spiderweb texture) observed in some garnet 1 grains result from garnet annealing at the metamorphic peak during the second metamorphic cycle, when the Xgrs-rich garnet 2 rims were formed. Previous garnet brecciation probably was induced by deformation, either after the first metamorphic event (M1), or during prograde metamorphism related to the Ross Orogeny. Following the metamorphic peak (M2), garnet was again fractured and resorbed, and finally altered by diffusive Fe, Mg, Mn exchange with surrounding biotite.

Metamorphism and deformation during the Ross Orogeny resulted in recrystallization of older monazite grains in the matrix of sample Me3/3 together with formation of new monazite grains. Only some of the older monazite grains survived the second metamorphic event unaffected, either armoured by garnet or enclosed by matrix biotite. The newly formed or recrystallized monazite grains were overgrown by kyanite at the metamorphic climax (assemblage M2), and by sillimanite, cordierite and biotite during the retrograde evolution (assemblage M3). Additionally, new monazite grains formed along garnet fractures at c. 500 Ma.

An apparent problem with the evolutionary model presented above is the presence of c. 1700 Ma zircon grains [point (5)], which seem to be of detrital origin, as indicated by their round, facet-free, and eroded surfaces. In situ formation of zircon in sample Me3/3 at c. 1700 Ma seems to be unlikely, as the metamorphic peak temperature of the Meade Nunatak rocks did not exceed 650°C and thus melting did not occur. However, melt formation in metapelitic rocks is a common prerequisite for abundant zircon formation (e.g. Rubatto et al., 2001; Williams, 2001), even though there are rare cases that show zircon formation under sub-solidus, hydrothermal conditions (e.g. Claoué-Long et al., 1990). Hence, the crucial question is which event is dated by the 1700 Ma zircons? The most straightforward explanation would be that the 1700 Ma ages reflect the time of zircon formation in a metamorphic or magmatic rock, prior to deposition of the metasedimentary protolith. Such an explanation has been given by several workers, who investigated detrital zircons (e.g. Williams & Claesson, 1987; Zeh et al., 2001). However, if that interpretation holds true, metamorphism of sample Me3/3 must have taken place after 1700 Ma, which is in contradiction to the model presented above. The obtained zircon ages in combination with the other observations allow two distinct interpretations.

First, all zircon (and monazite) grains are detrital and were deposited after 1700 Ma from a homogeneous magmatic–metamorphic source, and metamorphism (M1) took place at 1571 ± 40 Ma. In that case, the 1571 ± 40 Ma Sm–Nd garnet age would not reflect the end of a protracted cooling history between 1700 and 1570 Ma as suggested above, but point to a Barrovian-type metamorphism after 1700 Ma but prior to 1571 ± 40 Ma. That could be explained by fast recycling of crustal material followed by metamorphism, perhaps along an active continental margin.

The second, more simple explanation is that the 1700 Ma zircon ages do not reflect the time of zircon formation but rather that of complete zircon alteration. This is supported by the CL images obtained from zircon grains of sample Me3/3 (Fig. 7). These images show zircons with wide bands and patches with bright luminescence, which truncate or in most cases completely obliterate former zircon growth zonation patterns. These features are similar to those described for altered zircons in other metamorphic rocks (e.g. Vavra & Schaltegger, 1999; Vavra et al., 1999), and for zircons affected by hydrothermal experiments (Pidgeon et al., 1973; Sinha et al., 1992; Rizvanova et al., 2000; Geisler et al., 2001). In the experiments, hydrothermal fluids of different compositions (NaCl, HCl, Na2CO3, CaCl2) lead, within a very short time span (hours or days), to the formation of reaction fronts in metamict zircons, which truncate and obliterate the zircon growth zoning and have characteristic bright CL images. The experiments also show that radiogenic Pb will be quantitatively released from metamict zones during fluid alteration, and that the metamict zones will be ‘hardened’ as a result of zircon recrystallization. As a consequence, the U–Pb zircon ages obtained will reflect the time of zircon alteration. Taking the experimental results and our CL images into account, the concordant U–Pb ages obtained from the detrital zircon grains of sample Me3/3 can be interpreted to reflect zircon alteration at c. 1700 Ma rather than zircon formation. Zircon alteration at 1700 Ma may have been achieved by H2O + ionic species, released during prograde dehydration reactions of chlorite, which caused formation of the first garnet generation in sample Me3/3.

The discordant zircon analyses could be explained by the fact that domains in some zircon grains were not ‘hardened’ at 1700 Ma, but recrystallized and/or lost Pb during a later event, either during the Ross Orogeny at c. 500 Ma and/or later; for example, during the break-up of Gondwana at c. 180 Ma (e.g. Heimann et al., 1994), and/or by recent Pb loss (Figs 11c and 12c).

The polymetamorphic evolution, suggested above for the rocks from Meade and Lord nunataks, is consistent with age data derived from other units in the Shackleton Range. Metamorphism at c. 1700 Ga is confirmed by U–Pb zircon and monazite, as well as Sm–Nd garnet ages obtained from migmatitic gneisses from the La Grange Nunataks and the Haskard Highlands (Brommer et al., 1999; Zeh et al., 1999). Furthermore, there is evidence for a cooling event between 1550 and 1650 Ma, as indicated by numerous K–Ar and Rb–Sr mineral– whole-rock data obtained from rocks of the southern belt of the Shackleton Range (Hofmann et al., 1980; Pankhurst et al., 1983), which was relatively unaffected during the Ross Orogeny (Talarico & Kroner, 1999). The latter data are identical to the 1571 ± 40 Ma Sm–Nd garnet–whole-rock age obtained from samples Me1/1 and Me3/3.

The Ross orogenic metamorphism completely reset the K–Ar and Rb–Sr systems of mica in all rocks from the northern belt of the Shackleton Range, including gneisses and schists of the Statton and Pioneers Group. Ar–Ar, K–Ar and Rb–Sr mica ages scatter between 510 and 495 Ma (Pankhurst et al., 1983; Brommer & Henjes-Kunst, 1999; Zeh et al., 1999) and are identical to the Rb–Sr ages of 499 ± 12 Ma and 504 ± 1 Ma obtained from samples Lo1 and Me3/3, respectively. These data provide evidence that all rocks of the northern belt of the Shackleton Range were heated or reheated above 300–500°C during the Ross Orogeny.

The age data obtained from Lord Nunatak provide evidence that the entire tectono-metamorphic evolution during the Ross Orogeny in the Shackleton Range took place within c. 30 Myr. This conclusion is consistent with results obtained from rocks of the Haskard Highlands, situated c. 150 km west of the Lord Nunatak (Fig. 1). These rocks yielded ages of c. 532 ± 18 Ma (U–Pb zircon, upper intercept), 506 ± 6 Ma (Sm–Nd garnet) and 500 ± 10 (Rb–Sr biotite; Pankhurst et al., 1983; Zeh et al., 1999). The petrological and geochronological information obtained from sample Me3/3 supports the conclusion that rocks from Meade Nunatak underwent a similarly fast PT evolution during the Ross Orogeny. This is consistent with the lack of resetting of Sm–Nd systematics of garnet from samples Me1/1 and Me3/3 at c. 500 Ma, despite the relatively high peak PT conditions of c. 650°C at 7 kbar.

In summary, the combination of detailed petrological and geochronological results presented in the paper indicates that some metapelites from the northern belt of the Shackleton Range underwent a single Barrovian-type tectono-metamorphic evolution between 530 and 500 Ma, whereas others were formed during two distinct metamorphic cycles between 1700 and 1570 Ma and at 500 Ma. Our example shows that polymetamorphism in the investigated rocks can be unravelled only by a combination of detailed petrological and geochronological investigations.

APPENDIX A: MICROPROBE AND XRF ANALYSIS

Electron microprobe analyses were carried out with a CAMECA SX-50 microprobe in the Mineralogical Institute, University of Würzburg, with three independent wavelength-dispersive crystal channels. Instrument conditions were 15 kV acceleration voltage, 15 nA specimen current, and 20 s integration time for all elements except for Fe (30 s). Natural and synthetic silicates and oxides were used for reference, and matrix corrections were carried out by the PAP program supplied by CAMECA. Point analyses were performed with a 5 µm beam diameter for plagioclase and muscovite, and a 1 µm beam diameter for all other minerals. Detection limits (1σ) for a typical silicate analysis were ∼1 wt % relative for each element. Qualitative element maps of garnet were produced with a JEOL microprobe at the University of Frankfurt (Department of Mineralogy) using instrument conditions of 15 kV acceleration voltage, 3 nA specimen current, and 100 ms integration time for all elements. Bulk compositions used for phase diagram calculations were analysed by conventional XRF using a Phillips PW 1480 spectrometer, and lithium tetraborate fusion discs.

APPENDIX B: ISOTOPE DILUTION U–Pb ANALYSIS OF ZIRCON AND MONAZITE

Zircon and monazite were prepared at the NERC Isotope Geosciences Laboratory (NIGL) in Keyworth, Nottingham, UK, using standard crushing and heavy mineral separation techniques. Selected grains were separated under alcohol. All zircon fractions were abraded (Krogh, 1982), and then washed in 4N HNO3 and H2O to remove all traces of pyrite. Monazite grains were only washed in 1N HNO3 for 20 min and then in H2O. U and Pb were extracted using the methods of Krogh (1973) and Corfu & Ayres (1984). Fractions were spiked with a mixed 205Pb + 235U isotopic tracer prior to digestion and chemistry (Krogh & Davis, 1985). U and Pb were loaded together onto outgassed single Re filaments with silica gel. Isotope analyses were performed at the NERC Isotope Geosciences Laboratory, Keyworth, on a VG 354 mass spectrometer using a Daly photomultiplier ion counting system, supplemented in part by multiple Faraday cup collection of the larger ion beams. Pb isotope ratios were corrected for initial common Pb in excess of laboratory blank using the common Pb evolution model of Stacey & Kramers (1975). The laboratory blank during the period of analysis was 3 pg. Ages were calculated using the decay constants recommended by Steiger & Jäger (1977). Data reduction was carried out using PBDAT (Ludwig, 1989). The data were plotted using ISOPLOT (Ludwig, 1990).

APPENDIX C: IN SITU U–Pb LA-PIMMS DATING

In situ U–Pb laser ablation plasma ionization multi-collector mass spectrometry (LA-PIMMS) analyses were conducted at NIGL using a VG Elemental Axiom double-focusing PIMMS instrument, coupled to a VG Elemental/New Wave Research Microprobe II 266 nm Nd:YAG laser ablation system with a computer-controlled xyz stage. Analyses were carried out according to the protocol described by Horstwood et al. (2003); a brief summary follows. The samples were separated and mounted in resin blocks or analysed in situ in polished thin sections. A 554 Ma monazite, previously well characterized by ID-TIMS at NIGL, was used as an ablation standard to characterize the procedural Pb/U fractionation. Before analysis, each grain or analysis area was pre-cleaned using a reduced power density (larger spot size and lower laser power). Analysis was then conducted using a raster approach with ablation pits of the order of 60 µm × 60 µm × 15 µm in size, where possible. A Cetac Technologies MCN6000 desolvating nebulizer was used to simultaneously aspirate a 2% HNO3 solution containing thallium via the ablation cell and into the plasma. Data collection of Hg, Tl, Pb and U isotopes included measurement of 204Pb/Hg on the axial ion-counter with all other peaks on Faraday detectors. Data were processed according to Horstwood et al. (2003), and included a common-Pb correction.

APPENDIX D: Sm–Nd AND Rb–Sr DATING

Sm–Nd and Rb–Sr analyses were carried out at the NERC, Keyworth following standard chemical procedures described by Millar & Pankhurst (1987) and Pankhurst & Rapela (1995). The analyses were carried out using a Finnigan MAT 262 mass spectrometer using static multicollection. Fourteen determinations of 143Nd/144Nd in the La Jolla standard solution during analytical work yielded a mean value of 0·511890 ± 0·000012 (1σ). 143Nd/144Nd ratios were normalized to 146Nd/144Nd = 0·7219. The average crustal residence ages (TDM) were calculated according to Arndt & Goldstein (1987). Nine determinations of 87Sr/86Sr in the NBS987 standard solution during analytical work yielded a mean value of 0·710196 ± 0·000009 (1σ). 87Sr/86Sr was normalized during run time to 86Sr/88Sr = 0·1194. Blanks for Nd and Sr were less than 400 pg. For regression calculations, errors of 0·2% and 0·003% (1σ) were used for 147Sm/144Nd and 143Nd/144Nd, respectively. Errors of 0·5% and 0·005% (1σ) were used for Rb/Sr and 87Sr/86Sr, respectively, based on long-term reproducibility of rock standards. All data were plotted with ISOPLOT (Ludwig, 1990).

This study benefited from fruitful discussions with Wolfgang Schubert, Sönke Brandt and Gavin Foster. Furthermore, we thank Karine David, Jane Evans and Steve Noble (NIGL Keyworth) for their support with the isotope analyses, and Heidi Höfer (University of Frankfurt) for producing the garnet element maps. We are also indebted for the very constructive reviews of Simon Harley, Julie Hollis and Richard White, which contributed to an improvement of the manuscript.

REFERENCES

Arndt, N. T. & Goldstein, S. L. (
1987
). Use and abuse of crust-formation ages.
Geology
 
15
,
893
–895.
Brommer, A. & Henjes-Kunst, F. (
1999
). Panafrican reactivation of Proterozoic basement in the northern Shackleton Range (Antarctica) during the Neoproterozoic assembly of Gondwana.
Terra Antartica
 
6
,
279
–292.
Brommer, A., Millar, I. L. & Zeh, A. (
1999
). Geochronology, structural geology and petrology of the northwestern Lagrange Nunataks, Shackleton Range, Antarctica.
Terra Antartica
 
6
,
269
–278.
Chakraborty, S. & Ganguly, J. (
1992
). Cation diffusion in aluminosilicate garnets: experimental determination in spessartine–almandine diffusion couples, evaluation of effective binary diffusion coefficients, and applications.
Contributions to Mineralogy and Petrology
 
111
,
74
–86.
Cherniak, D. J. (
2000
). Rare earth element diffusion in apatite.
Geochimica et Cosmochimica Acta
 
64
,
3871
–3885.
Claoué-Long, J. C., King, R. W. & Kerrich, R. (
1990
). Archean hydrothermal zircon in the Albibi greenstone belt: constraints on the timing of gold mineralisation.
Earth and Planetary Science Letters
 
98
,
109
–128.
Clarkson, P. D. (
1995
). Introduction. In: British Antarctic Survey (ed.) Geological Map of the Shackleton Range, GEOMAP Series, Sheet 4, pp. 1–7.
Copeland, P., Parrish, P. R. & Harrison, T.M. (
1988
). Identification of inherited radiogenic lead in monazite and implications for U–Pb systematics.
Nature
 
333
,
760
–763.
Corfu, F. & Ayres, L. D. (
1984
). U–Pb ages and genetic significance of heterogeneous zircon populations in rocks from the Favourable Lake area, northwestern Ontario.
Contributions to Mineralogy and Petrology
 
88
,
86
–101.
Foster, G., Kinny, P., Vance, D., Prince, C. & Harris, N. (
2000
). The significance of monazite U–Th–Pb age data in metamorphic assemblages; a combined study of monazite and garnet chronometry.
Earth and Planetary Science Letters
 
181
,
327
–340.
Foster, G., Gibson, H. D., Parrish, R. R., Horstwood, M., Fraser, J. & Tindle, A. (
2002
). Textural, chemical and isotopic insights into the nature and behaviour of metamorphic monazite.
Chemical Geology
 
191
,
183
–207.
Geisler, T., Ulonska, M., Schleicher, H., Pidgeon, R. T. & van Bronswijk, W. (
2001
). Leaching and differential recrystallization of metamict zircon under experimental hydrothermal conditions.
Contributions to Mineralogy and Petrology
 
141
,
53
–65.
Golanov, N. P., Michajlov, V. M. & Shuljatin, O. G. (
1980
). Pervie diagnostiruemie stromatoliti Antarktidi i ich biostratigraficheskoe znachenie.
Antarktika
 
19
,
152
–159.
Goodge, J., Fanning, C. & Bennett, V. (
2001
). U–Pb evidence of ∼1·7 Ga crustal tectonism during the Nimrod Orogeny in the Transantarctic Mountains, Antarctica: implications for Proterozoic plate reconstructions.
Precambrian Research
 
112
,
261
–288.
Heimann, A., Fleming, T. H., Elliot, D. H. & Foland, K. A. (
1994
). A short interval of Jurassic flood basalt volcanism in Antarctica as demonstrated by 40Ar/39Ar geochronology.
Earth and Planetary Science Letters
 
121
,
19
–41.
Hofmann, J., Kaiser, G., Klemm, W. & Paech, H.-J. (
1980
). K/Ar Alter von Doleriten und Metamorphiten der Shackleton Range und der Whichaway Nunataks, Ost- und Nordostumrandung des Filchner-Eisschelfs (Antarktis).
Zeitschrift für Geologische Wissenschaften
 
8
,
1227
–1232.
Holland, T. J. B. & Powell, R. (
1998
). An internally consistent thermodynamic data set for phases of petrological interest.
Journal of Metamorphic Geology
 
16
,
309
–343.
Horstwood, M. S. A., Foster, G. L., Parrish, R. R., Noble, S. R. & Nowell, G. M. (
2003
). Common-Pb corrected in-situ U–Pb accessory mineral geochronology by LA-MC-ICPMS.
Journal of Analytical Atomic Spectrometry
 
18
,
837
–846.
Jäger, E. (
1979
). The Rb–Sr method. In: Jäger, E. & Hunziker, A. (eds) Lectures in Isotope Geology. Heidelberg: Springer, pp. 13–26.
Krogh, T. E. (
1973
). A low contamination method for the hydrothermal decomposition of zircon and extraction of U and Pb for isotopic age determinations.
Geochimica et Cosmochimica Acta
 
37
,
485
–494.
Krogh, T. E. (
1982
). Improved accuracy of U–Pb zircon ages by the creation of more concordant systems using an air abrasion technique.
Geochimica et Cosmochimica Acta
 
46
,
637
–649.
Krogh, T. E. & Davis, G. L. (
1985
). The production and preparation of 205Pb for use as a tracer for isotope dilution analysis.
Carnegie Institution of Washington, Yearbook
 
74
,
416
–417.
Ludwig, K. R. (
1989
). PBDAT: a computer program for processing Pb–U–Th isotope data, version 1.20. US Geological Survey, Open-File Report 88–542.
Ludwig, K. R. (
1990
). Isoplot: a plotting and regression program for radiogenic isotope data, version 2.03. US Geological Survey, Open-File Report 88–557.
Mezger, K., Essene, E. J. & Halliday, A. N. (
1992
). Closure temperatures of the Sm–Nd system in metamorphic garnets.
Earth and Planetary Science Letters
 
113
,
397
–409.
Millar, I. L. & Pankhurst, R. J. (
1987
). Rb–Sr geochronology of the region between the Antarctic Peninsula and the Transantarctic Mountains: Haag Nunataks and Mesozoic granitoids. In: McKenzie, G. D. (ed.) Gondwana Six: Structure, Tectonics, and Geophysics. Geophysical Monograph, American Geophysical Union 40, 151–160.
Pankhurst, R. J. & Rapela, C. R. (
1995
). Production of Jurassic rhyolite by anatexis of the lower crust of Patagonia.
Earth and Planetary Science Letters
 
134
,
23
–36.
Pankhurst, R. J., Marsh, P. D. & Clarkson, P. D. (
1983
). A geochronological investigation of the Shackleton Range. In: Oliver, P. R. J. & Jago, J. B. (eds) Antarctic Earth Science. Canberra: Australian Academy of Sciences, pp. 176–182.
Pidgeon, R. T. (
1992
). Recrystallisation of oscillatory zoned zircon: some geochronological and petrological implications.
Contributions to Mineralogy and Petrology
 
110
,
463
–473.
Pidgeon, R. T., O'Neil, R. J. & Silver, R. T. (
1973
). Observations on the crystallinity and the U–Pb system of a metamict Ceylon zircon under experimental hydrothermal conditions.
Fortschritte in Mineralogie
 
50
,
118
.
Prince, C. I., Kosler, J., Vance, D. & Günther, D. (
2000
). Comparison of laser ablation ICP-MS and isotope dilution REE analyses— implications for Sm–Nd garnet geochronology.
Chemical Geology
 
168
,
255
–274.
Rizvanova, N. G., Levchenkov, O. A., Belous, A. E., Bezmen, N. I., Maslenikov, A. N., Komarov, A. N., Makeev, A. F. & Levskii, L. K. (
2000
). Zircon reaction and stability of the U–Pb isotope system during interaction with carbonate fluid: experimental hydrothermal study.
Contributions to Mineralogy and Petrology
 
139
,
101
–114.
Rubatto, D., Williams, I. S. & Buick, I. S. (
2001
). Zircon and monazite response to prograde metamorphism in the Reynolds Range, Central Australia.
Contributions to Mineralogy and Petrology
 
140
,
458
–468.
Schubert, W. & Will, T. (
1994
). Granulite-facies rocks of the Shackleton Range, Antarctica. Conditions of formation and preliminary petrogenetic implications.
Chemie der Erde
 
54
,
355
–371.
Sinha, A. K., Wayne, D. M. & Hewitt, D. A. (
1992
). The hydrothermal stability of zircon: preliminary experimental and isotope studies.
Geochimica et Cosmochimica Acta
 
56
,
3551
–3560.
Stacey, J. S. & Kramers, J. D. (
1975
). Approximation of terrestrial lead isotope evolution by a two-stage model.
Earth and Planetary Science Letters
 
26
,
207
–221.
Steiger, R. H. & Jäger, E. (
1977
). Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology.
Earth and Planetary Science Letters
 
36
,
359
–362.
Talarico, F. & Kroner, U. (
1999
). Geology and tectono-metamorphic evolution of the Read Group, Shackleton Range: a part of the Antarctic Craton.
Terra Antartica
 
6
,
183
–202.
Talarico, F., Kleinschmidt, G. & Henjes-Kunst, F. (
1999
). First report of an ophiolitic complex in the northern Shackleton Range, Antarctica.
Terra Antartica
 
6
,
293
–316.
Tessensohn, F., Kleinschmidt, G., Talarico, F., Buggish, W., Brommer, A., Henjes-Kunst, F., Kroner, U., Millar, I. L. & Zeh, A. (
1999
). Ross-age amalgamation of East and West Gondwana: evidence from the Shackleton Range, East Antarctica.
Terra Antartica
 
6
,
317
–326.
Thöni, M. & Jagoutz, E. (
1992
). Some new aspects of dating eclogites in orogenic belts: Sm–Nd, Rb–Sr, and Pb–Pb isotopic results from the Austroalpine Saualpe and Koralpe type-locality (Carinthia/Styria, southeastern Austria).
Geochimica et Cosmochimica Acta
 
56
,
347
–368.
Vance, D. & O'Nions, R. K. (
1990
). Isotopic chronometry of zoned garnets: growth kinetics and metamorphic histories.
Earth and Planetary Science Letters
 
97
,
227
–240.
Vavra, G. & Schaltegger, U. (
1999
). Post-granulite facies monazite growth and rejuvenation during Permian to Lower Jurassic thermal and fluid events in the Ivrea Zone (Southern Alps).
Contributions to Mineralogy and Petrology
 
134
,
405
–414.
Vavra, G., Schmid, R. & Gebauer, D. (
1999
). Internal morphology, habit and U–Th–Pb microanalysis of amphibolite-to-granulite facies zircons: geochronology of the Ivrea Zone (Southern Alps).
Contributions to Mineralogy and Petrology
 
134
,
380
–404.
Weber, B. (
1991
). Microfossils in Proterozoic sediments from the southern Shackleton Range: a preliminary report.
Zeitschrift für Geologische Wissenschaften
 
19
,
185
–197.
White, R. W., Powell, R., Holland, T. J. B. & Worley, B. A. (
2000
). The effect of TiO2 and Fe2O3 on metapelitic assemblages at greenschist and amphibolite facies conditions: mineral equilibria calculations in the system K2O–FeO–MgO–Al2O3–SiO2–H2O– TiO2–Fe2O3.
Journal of Metamorphic Geology
 
18
,
497
–511.
Williams, I. S. (
2001
). Response of detrital zircon and monazite, and their U–Pb isotopic systems, to regional metamorphism and host-rock partial melting, Cooma Complex, southeastern Australia.
Australian Journal of Earth Sciences
 
48
,
557
–580.
Williams, I. S. & Claesson, S. (
1987
). Isotopic evidence for the Precambrian provenance and Caledonian metamorphism of the high-grade paragneisses from the Seve Nappes, Scandinavian Caledonides, II. Ion microprobe zircon U–Pb-Th.
Contributions to Mineralogy and Petrology
 
97
,
205
–217.
Zeh, A. (
2001
). Inference of a detailed PT path from PT pseudosections using metapelitic rocks of variable composition from a single outcrop, Shackleton Range, Antarctica.
Journal of Metamorphic Geology
 
19
,
329
–350.
Zeh, A. & Holness, M.B. (
2003
). The effect of reaction overstep on garnet microtextures in metapelitic rocks of the Ilesha Schist Belt, SW Nigeria.
Journal of Petrology
 
44
,
967
–994.
Zeh, A. & Millar, I. L. (
2001
). Metamorphic evolution of garnet– epidote–biotite gneiss from the Moine Supergroup, Scotland, and geotectonic implications.
Journal of Petrology
 
42
,
529
–554.
Zeh, A., Millar, I. L., Kroner, U. & Görz, I. (
1999
). The structural and metamorphic evolution of the Northern Haskard Highlands, Shackleton Range, Antarctica.
Terra Antartica
 
6
,
249
–268.
Zeh, A., Brätz, H., Millar, I. L. & Williams, I. S. (
2001
). A combined zircon SHRIMP and Sm–Nd isotope study on high-grade paragneisses from the Mid-German Crystalline Rise: evidence for northern Gondwanan and Grenvillian provenance.
Journal of the Geological Society, London
 
158
,
983
–994.
Zeh, A., Williams, I. S., Brätz, U. & Millar, I. L. (
2003
). Different age response of zircon and monazite during the tectono-metamorphic evolution of a high-grade paragneiss from the Ruhla Crystalline Complex, Central Germany.
Contributions to Mineralogy and Petrology
 
145
,
691
–706.
Zhu, X. K., O'Nions, R. K., Belshaw, N. S. & Gibbs, A. J. (
1997
). Lewisian crustal history from in situ SIMS mineral chronometry and related metamorphic textures.
Chemical Geology
 
136
,
205
–218.
Zhu, X. K. & O'Nions, R. K. (
1999
). Zonation of monazite in metamorphic rocks and its implications for high temperature thermochronology: a case study from the Lewisian terrain.
Earth and Planetary Science Letters
 
171
,
209
–220.