Journal of Petrology Advance Access originally published online on August 27, 2004
Journal of Petrology 2004 45(10):1959-1981; doi:10.1093/petrology/egh044
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology 45(10) © Oxford University Press 2004; all rights reserved
Petrology of Whiteschists and Associated Rocks at Mautia Hill (Tanzania): Fluid Infiltration during High-Grade Metamorphism?
INSTITUT FÜR GEOWISSENSCHAFTEN, UNIVERSITÄT KIEL, 24098 KIEL, GERMANY
RECEIVED JULY 28, 2003; ACCEPTED APRIL 27, 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Talc–kyanite schists (whiteschists), magnesiohornblende–kyanite–talc–quartz schists and enstatite–sapphirine–chlorite schists occur at Mautia Hill in the East African Orogen of Tanzania. They are associated with metapelites and garnet–clinopyroxene–quartz metabasites. Geobarometry (GASP/GADS equilibria) applied to the latter two rock types indicates a peak pressure of P = 10–11 kbar. These results are confirmed by the high fO2 assemblage hollandite–kyanite–quartz and late-stage manganian andalusite that contains up to 19·5 mol. % Mn2SiO5. Maximum temperatures of T = 720°C are inferred from late-stage yoderite + quartz. A clockwise P–T evolution is constrained by prograde kyanite inclusions in metapelitic garnet and late-stage reaction rims of cordierite between green yoderite and talc that reflect conditions at least 3–4 kbar below the peak pressure. Oxidizing conditions are recorded throughout the metamorphic history of the whiteschists and chlorite schists, as indicated by the presence of haematite coexisting with pseudobrookite and/or rutile. Increasing water activity near peak pressures is thought to have led to the breakdown of the high-pressure assemblages (Tlc–Ky–Hem and Mg-Hbl–Ky–Hem) and the subsequent formation of certain uncommon minerals, e.g. yellow sapphirine, Mn–andalusite, green and purple yoderite, piemontite and boron-free kornerupine. The proposed increase in water activity is attributed to fluid infiltration resulting from the devolatilization of underlying sediments during metamorphism.
KEY WORDS: fluid infiltration; high-pressure amphibolite facies; East African Orogen; Pan-African; whiteschist
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
It has been shown experimentally that the critical mineral assemblage of whiteschists, talc and kyanite, is stable only under relatively low geothermal gradients that are realized in the high-pressure part of the amphibolite facies and in the eclogite facies (Schreyer & Seifert, 1969
; Massonne, 1989). In agreement with the experimental data, it has been found worldwide in different orogenic belts that whiteschists are formed during either crustal thickening or subduction, i.e. in geodynamic regimes that are characterized by clockwise P–T paths. Whiteschists occur in the Alps (Abraham et al., 1974; Chopin & Monié, 1984; Chopin et al., 1991; Pawlig & Baumgartner, 2001), the Hindukush (Kulke & Schreyer, 1973; Schreyer, 1977; Faryad, 1999), Tasmania (Råheim & Green, 1974) and the Dabie Shan in China (Rolfo et al., 2000). They occur also in Precambrian belts, such as the Zambezi Belt of Zambia and Zimbabwe (Vrána & Barr, 1972; Johnson & Oliver, 2002; John et al., 2004) and the Mozambique Belt of Tanzania (McKie, 1959). Therefore, it is likely, even in cases where eclogites and blueschists are lacking, that whiteschists mark the sites of palaeo-collision zones. In contrast to the geodynamic settings of whiteschist metamorphism, the formation of whiteschists as a rock type, which generally displays a simple MgO–Al2O3–SiO2–H2O (MASH)-silicate mineralogy, coexisting with iron oxides or sulfides, is more diverse and controversial. In several cases, it has been demonstrated that whiteschist formation resulted from metasomatic processes, the timing of which is mostly unknown. Thus, whiteschists may be used to study fluid–rock interaction processes at high-pressure amphibolite and eclogite facies conditions during crustal thickening. In many studied examples, the precursor to whiteschists is granitic (Rolfo et al., 2000; Pawlig & Baumgartner, 2001). However, metabasic rocks have also been proposed (Johnson & Oliver, 2002). Only a few occurrences are known where metasedimentary protoliths are likely (McKie, 1959; Schreyer & Abraham, 1976). In these cases, whiteschists occur in close association with other metasedimentary rocks (e.g. metapelites, marbles, quartzites, metaevaporites). In several whiteschist occurrences, evidence for high oxygen fugacity has been found (Vrána & Barr, 1972; Grew et al., 1998; Johnson & Oliver, 2002), resulting in iron occurring predominantly in the trivalent state and being nearly restricted to haematite. It occurs only in small amounts in the coexisting silicates, which are near-Mg end-members. In other cases, iron is nearly confined to sulfides (Kulke & Schreyer, 1973; Pawlig & Baumgartner, 2001).
The MASH–Fe2O3 schists of Mautia Hill (Tanzania, Fig. 1) became famous in the 1950 s because they feature an unusually large mineralogical variety compared with more simple whiteschist mineralogies that occur elsewhere. They therefore provide a good opportunity to investigate the processes and metamorphic boundary conditions leading to whiteschist formation. Up to now, Mautia Hill is the only known occurrence of purple yoderite (McKie, 1959) and yellow sapphirine (McKie, 1963b). Other minerals, e.g. enstatite, magnesiochlorite, högbomite (McKie, 1963a), pseudobrookite (McKie, 1963b), boron-bearing kornerupine (McKie, 1965), green yoderite (McKie & Bradshaw, 1966), piemontite, and manganian andalusite (Basu & Mruma, 1985) have also been found. In the course of this study, additional minerals were discovered, such as yellow spinel, cerianite, cordierite, anthophyllite, boron-free kornerupine, geikielite and hollandite. Some of these were found to occur not only in whiteschists, but also in Ca-rich amphibole–chlorite schists. Further rock types occurring at Mautia are metabasites, metapelites, marbles, quartzites and pegmatites. The great variety of rock types allows the application of conventional geothermobarometers in addition to equilibria of the uncommon mineral assemblages to characterize the P–T–fluid conditions that prevailed during whiteschist metamorphism. Phase petrological considerations, thermobarometry and mineral-chemical data were used to estimate the fO2 and aH2O conditions and to determine the time of fluid infiltration, which seems to have been essential for the formation of the uncommon whiteschist mineralogy at Mautia Hill.
|
|
PREVIOUS WORK AND GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Mautia Hill is located approximately 10 km NNE of Kongwa (Mpwapwa District, central Tanzania, Fig. 1). The hill is about 2·0 km long, 0·5 km wide, east-trending and rising some tens of meters above the surrounding peneplain (Fig. 2).
|
The geology of Mautia Hill was first mentioned in 1938, when Temperley mapped the sheet ‘Mpwapwa’ (Temperley, 1938
). He described the yoderite-bearing schists as glaucophane–kyanite gneisses. Yoderite and other minerals were examined in detail in the 1950 s and 1960 s (McKie, 1959, 1963a, 1963b, 1965; McKie & Bradshaw, 1966). Part of the P–T history of Mautia Hill has been deduced on the basis of fluid inclusion studies (Basu & Mruma, 1985; Mruma, 1986; Mruma & Basu, 1987). Phase petrological studies of Mautia Hill rocks are only preliminary and have not been published (Möller, 1995; Rickers, 1996). Mautia Hill was mapped in detail by T. Heinrichs in 1981 (Heinrichs, unpublished data; Fig. 2). Several rock types (metabasites, metapelites, marbles, whiteschists, chlorite schists, pegmatites, quartzites, migmatitic gneisses) occur at Mautia Hill as thin layers, parallel to the elongation of the hill, but are poorly exposed and most occur only as boulders. The best-exposed rocks are a marble and a Ky–Bt quartzite (mineral abbreviations after Kretz, 1983) that form the crest of the hill. The great lithological variety cannot be tracked into the surrounding plain, which consists of uniform migmatitic granite gneisses and is mainly covered by alluvial sediments. The earliest structure observed is a schistosity (S1) that is isoclinally refolded by a later deformational event (D2). The resulting S2 foliation trends WNW. A ubiquitous L2 lineation is parallel to F2 fold axes, dipping in an ESE direction. It is defined by the alignment of large grains of kyanite or amphibole in different rock types. The F2 folding is associated with the formation of the peak assemblage (En–Spr–Chl) in chlorite schists (Fig. 3). The close association and common deformation history of the different rock types makes it most likely that all lithological units at Mautia Hill experienced the same metamorphic evolution.
|
Mautia Hill is situated near the western border of the Pan-African Mozambique Belt, the formation of which is attributed to the collision of East and West Gondwana (McWilliams, 1981
; Stern, 1994). To the west of Mautia Hill is the Palaeoproterozoic (2·0–1·8 Ga) Usagaran Belt, which itself adjoins the Archean (2·6–2·5 Ga) Tanzania Craton (Fig. 1). The Mozambique Belt consists mainly of uniform granulite-facies orthogneisses (Appel et al., 1998), which contrasts with the great variety of rock types occurring at Mautia Hill. In addition, the metamorphic evolution of rocks at Mautia Hill is distinct from that of the granulites further east. The P–T path for the whiteschists is inferred to be clockwise and the age of metamorphism seems to be at least 50 Myr younger (A. Möller, personal communication) than that in rocks of the Pan-African Mozambique Belt, further east (610–655 Ma; Möller et al., 2000), where an anticlockwise P–T evolution has been proposed (Appel et al., 1998). |
PETROGRAPHY AND MINERAL CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Analytical procedure
The chemical compositions of minerals were determined with a ‘JEOL Superprobe JXA-8900R’ electron microprobe, at the University of Kiel, equipped with five wavelength-dispersive spectrometers (WDS). The accelerating potential was 15 kV or 20 kV for a beam current of 20 nA. Sample spot sizes were 1–5 µm in diameter. Both synthetic and natural mineral standards were used. The raw data were corrected using the CITZAF method (Armstrong, 1995
).
The boron content of kornerupine was also measured with the electron microprobe. The accelerating voltage was 10 kV for a beam current of 40 nA and a spot size of 5 µm. We used LDE2 (2D = 98 Å) and LDEB (2D = 145 Å) crystals. Natural danburite (Dyar et al., 2002) was used as a calibration sample. Analysis was performed with 100 s counting time on peak and 50 s on each lower and upper background. For comparison, we also performed measurements in peak area integration mode (15, 000 s total per scan, step size 50 µm).
Metapelites
This rock type occurs within the widespread Bt–Ky quartzites (Fig. 2). It is porphyroblastic, medium- to coarse-grained and composed of quartz (30–40 vol. %), biotite (25–30 vol. %), plagioclase (15–20 vol. %) and garnet (5–15 vol. %). Accessory minerals are haematite, rutile, monazite, zircon and apatite. The rocks show a well-developed schistosity, except for samples with a lower biotite content, which are more massive. Because they do not contain alkali-feldspar, they are more precisely characterized as K-poor semipelites. Representative microprobe analyses are given in Table 1.
|
Ellipsoidal garnet forms porphyroblasts that contain inclusions of Pl, Qtz and Ky. Apart from a narrow (80 µm) re-equilibrated rim, it is chemically unzoned. It has an XMg [=Mg/(Mg + Fe2+)] of 0·4, XAlm = 0·55, XPrp = 0·3–0·4, XGrs = 0·05–0·1 and XSps < 0·05. No significant amount of Fe3+ was detected. The matrix plagioclase shows inverse zonation with an anorthite content of 0·26–0·32 mol. % in the core and 0·32–0·38 mol. % at the rim. Some inclusions in garnet have a slightly lower XAn [Ca/(Ca + Na + K)]. Biotite occurs as large laths within the matrix of the rock. Assuming FeO = Fetot, the XMg is 0·7–0·9. Locally, Bt is intergrown with haematite. Kyanite crystals are about 2·5 mm long and show conspicuous green colour. Microprobe analyses point to a formula
.
Metabasites
The metabasites crop out at the southwestern slope of the hill. They can be subdivided into two groups. The first one is a medium-grained, clearly foliated rock, which consists of plagioclase (15–25 vol. %), garnet (20–30 vol. %), clinopyroxene (20–25 vol. %) and hornblende (20–45 vol. %). The metabasites of the second group have a more massive appearance, and they contain about 15 vol. % clinozoisite. Other constituents are plagioclase (40 vol. %), clinopyroxene (15 vol. %) and garnet (15 vol. %). Accessory minerals are titanite, rutile, apatite and quartz. Some microprobe analyses are given in Table 1.
Garnet is homogeneous, except for a thin rim (40 µm) that re-equilibrated during cooling. The XMg ratio is in the range from 0·1 to 0·2, with a small increase towards the rim. The range of garnet compositions is XAlm = 0·4–0·5, XPrp = 0·10–0·15, XSps < 0·05 and XGrs 
0·35. In Czo-bearing rocks, the XGrs is slightly higher (0·40). Plagioclase mainly occurs as intergranular grains. Its anorthite content is in the range from 40 to 43% in the core and up to 50% towards the rim. In Czo-rich rocks, these values are slightly higher. Clinopyroxene is characterized by low Na and XEn = 0·35, XFs = 0·17, XWol = 0·48, and an XMg ratio of ca. 0·7. It is unzoned and locally replaced by ferrotschermakitic hornblende. The XMg ratio of the amphibole is about 0·5. According to the nomenclature of Leake et al. (1997), most amphiboles are classified as magnesiohornblende and some as edenite (K + Na = 0·47–0·53 p.f.u., on the basis of 23 oxygens). Clinozoisite occurs in the matrix of the unfoliated metabasites and has a Fe3+/(Fe3+ + Al) ratio of 0·18. In places, it forms symplectitic Czo–Qtz intergrowths around garnet. In this case, the Fe3+ content is slightly lower.
Other rock types
The following rock types occur in close association at Mautia Hill. Because their mineralogy is not very informative for reconstructing the P–T–fluid history, they are not described in detail.
Biotite–kyanite quartzites
Bt–Ky quartzites locally form the crest of the hill (Fig. 2). They are coarse-grained, show a distinct schistosity and contain conspicuous green kyanite crystals, which are up to 15 cm long and form 15–25 vol.% of the rock. Further constituents are quartz (40–50 vol. %), biotite (20–35 vol. %) and accessory haematite, rutile, apatite and zircon. Local ore mineralization (chrysocolla, galena, chalcopyrite) is obvious. Biotite is Mg-rich, with an XMg of 0·8–0·9. Kyanite contains 0·85–1·1 wt. % Fe2O3 and traces of Cr2O3. Haematite and rutile may be intergrown.
Amphibole–piemontite/epidote quartzites
Both amphibole–piemontite quartzites and amphibole–epidote quartzites occur at Mautia Hill (Fig. 2). They consist of quartz (60–70 vol. %), amphibole (10–20 vol. %) and piemontite/epidote (15–20 vol. %). Rutile, apatite, haematite and zircon occur as accessory phases. The amphibole is a tremolite with XMg = 0·96–0·99. Piemontite has a Mn2O3 content of 5–7 wt. %. Epidote contains just 0·3 wt. % Mn2O3, but up to 9 wt. % Fe2O3.
Marbles
Nearly pure dolomite marbles form the western crest of the hill (Fig. 2). Locally, they contain minerals like piemontite, margarite, amphibole, högbomite [general formula (Fe,Mg)1·8(Al,Ti)3·7(O,OH)8] and late-stage sudoite [Mg2Al3(Si3Al)O10(OH)8]. Piemontite is relatively Mn-poor (1·4–2·3 wt. % Mn2O3). Högbomite contains 4·5–5·0 wt. % TiO2.
Pegmatites
Pegmatites occur in some areas of the eastern part of the hill (Fig. 2). They are undeformed and consist of millimetre to centimetre crystals of quartz, plagioclase, biotite, apatite, green tourmaline (dravite), haematite and green kyanite (up to 5 cm). The presence of kyanite points to exceptional high pressures for the formation of pegmatites.
Altered impure marbles
A thin band of impure marble occurs south of the crest of the hill. Beside dolomite, also calcite, phlogopite, rutile and tourmaline occur. Cavities are filled with spherulitic aggregates of chalcedony, pointing to late low-temperature alteration. Tourmaline is pleochroic from orange to pinkish and is classified as an Al-poor member of the alkali group of Hawthorne & Henry (1999).
Migmatitic gneisses
Migmatitic gneisses have granitic compositions and consist of alkali-feldspar, plagioclase, biotite and quartz. They are common in the eastern part of Mautia Hill (Fig. 2).
Whiteschists and chlorite schists
These unusual rock types contain rare silicate minerals, for which Mautia Hill is famous (McKie, 1959, 1963a, 1963b, 1965; McKie & Bradshaw, 1966; Basu & Mruma, 1985; Mruma, 1986; Mruma & Basu, 1987; Fockenberg & Schreyer, 1994). The bulk chemistry of some of these rocks is well approximated by the system MgO–Al2O3–SiO2–H2O–Fe2O3 (MASH–Fe2O3). Other rocks additionally contain CaO and CO2 (CMASH–CO2–Fe2O3). Further components, such as FeO, MnO, Mn2O3, MnO2, BaO, PbO and TiO2, are present in minor amounts, but some of them are important for the mineralogical variety. The CMASH–CO2–Fe2O3 and MASH–Fe2O3 rocks form bands between dolomite marble and the Amph–Piem/Ep quartzite (Fig. 2).
Rocks of the CMASH–CO2–Fe2O3 system
On the basis of their primary mineral assemblages, the rocks of the CMASH–CO2–Fe2O3 chemical system have been subdivided into three subgroups: amphibole–dolomite–chlorite schists, amphibole–spinel–dolomite schists and hornblende–kyanite–talc–quartz schists. Representative microprobe analyses of selected minerals are given in Table 2. Phase relations and the compositional space of the different subgroups are shown in Fig. 4a.
|
|
Amphibole–dolomite–chlorite schist. This Qtz-free rock type is medium- to coarse-grained, greyish-yellow in hand specimen and features a distinct foliation. It consists of magnesiochlorite (
45 vol. %), amphibole (35 vol. %) and late-stage enstatite (20 vol. %). The amount of dolomite is highly variable, resulting in different carbonate/silicate ratios: the rock group includes nearly pure marbles but also nearly carbonate-free chlorite schists. Accessory minerals are haematite, rutile, talc and zircon. Amphiboles are euhedral and lie in a finer-grained matrix of magnesiochlorite. Amphibole is partially replaced and surrounded by Chl-, Dol- and En-bearing reaction rims. Haematite contains lamellae of rutile and/or geikielite and coexists with rutile. Amphibole–spinel–dolomite schist. This rock looks similar to the previously described one. Dolomite is subordinate in volume and, in some patches, the rock is made up purely of amphibole. Spinel is everywhere completely rimmed by yellow sapphirine, which itself is separated from the matrix amphibole by enstatite, magnesiochlorite and dolomite (Fig. 5a). Zircon, haematite, talc and baryte are accessories. A small amount of cerianite is included within sapphirine.
|
Hornblende–kyanite–talc–quartz schist. These eye-catching schists consist of greenish and reddish portions. The matrix consists mostly of quartz and talc, but local clusters of orange-coloured biotite occur. Biotite is locally intergrown with talc. In the greenish portions of the rock, kyanite is rimmed by symplectites of Mn-andalusite, piemontite and yoderite (Fig. 5c). A common feature of the schists is colourless (in hand specimen reddish) magnesiohornblende, which is partially replaced and locally pseudomorphed by intergrowths of purple yoderite, talc and piemontite (Fig. 5d). Intergrowths of talc + Mn-andalusite or yoderite + Mn-andalusite + talc (Fig. 5f) also occur within the pseudomorphs. These intergrowths form the greenish patches seen in hand specimen. The products of breakdown are not everywhere homogeneously distributed within the pseudomorphs: for example, piemontite is missing in some pseudomorphs, whereas yoderite is missing in others (Fig. 5e and f). Large prograde grains of piemontite occur in the matrix and as inclusions in kyanite and Mg-hornblende. Late-stage piemontite formed at the expense of Mg-Hbl. Accessory minerals include apatite, hollandite (BaMn8O16), cerianite, manganian haematite (Fe1·8Mn0·2O3) and rutile. The latter is typically intergrown with Mn-haematite. Locally, hollandite forms a corona between Mn-haematite and biotite/talc and seems to be a late-stage formation.
Mineral chemistry of the CMASH–CO2–Fe2O3 rocks
Magnesiochlorite from amphibole–dolomite–chlorite schists and amphibole–spinel–dolomite schists has a uniform composition near the clinochlore end-member. The small chemical variability is due to-similar Tschermak (2Al = SiMg) and dioctahedral substitutions (2Al = 3Mg) in all analysed samples (Fig. 4b and Table 2). In amphibole–dolomite–chlorite schists, the cores of amphiboles (first generation) are tremolitic, whereas the rims are classified as magnesiohornblende (second generation; Fig. 6a). The XMg of tremolite and Mg-hornblende is in the range from 0·96 to 1·00. Both amphibole generations show the effect of increasing Tschermak substitution with decreasing XMg values (Fig. 6a). Zoned amphibole crystals with tremolite cores and Mg-hornblende rims show a sharp boundary between these amphibole generations, pointing to a sudden change in metamorphic conditions (P, T or fluid; Fig. 6b). Hornblende–kyanite–talc–quartz schists contain only magnesiohornblende with XMg = 0·93–0·99, but no tremolite. Magnesiohornblende is slightly enriched in alumina compared with amphiboles of the Amph–Dol–Chl schists (Fig. 6a). Enstatite has an XMg = 0·98–1·00. In Amph–Dol–Chl schists, it contains up to 2 wt. % Al2O3, in spinel- and sapphirine-bearing rocks up to 4 wt. % Al2O3 (Table 2). Calculating the amount of Fe3+ assuming stoichiometry (20 oxygens, 14 cations), yellow sapphirine has an XMg of 0·98 to 1·00. It is optically positive and commonly polysynthetically twinned. The composition is near the 7:9:3 [MgO:(Al, Fe)2O3: SiO2] end-member, but subtly poorer in alumina [Fe3+/(Fe3+ + Al) = 0·05; Fig. 4c; Table 2]. The yellow colour may be related to the high oxidation ratio of iron (McKie, 1963b). Spinel is pale yellow and, in some cases, slightly birefringent. It has a composition near the Mg end-member, but contains small components of magnesioferrite (3%) and gahnite (2%; Table 2). Spinel contains lamellae of haematite and högbomite, possibly as a result of oxidation–exsolution or a former Ti content in spinel (Fig. 5b). Högbomite has a yellow colour, like sapphirine and spinel, contains up to 2 wt. % TiO2 and is Fe-poor (Table 2). Piemontite in hornblende–kyanite–talc–quartz schists shows a weak compositional variability [Mn3+/(Mn3+ + Fe3+ + Al) = 0·10–0·17; Table 2]. Early piemontite, e.g. inclusions in hornblende, contain less Mn than the late-stage reaction products formed by breakdown of hornblende. In hornblende–kyanite–talc–quartz schists, prograde talc contains 0·2–0·3 Al p.f.u. (formula calculated on the basis of 11 oxygens; Table 2). Other CMASH–CO2–Fe2O3 rocks contain only small amounts of late-stage talc, which is nearly Al-free (Fig. 4d). Kyanite contains <0·4 wt. % Cr2O3 and up to 2·5 wt. % Fe2O3 (Table 2). Though Chinner et al. (1969) pointed to 1·6 wt. % as the maximum Fe2O3 content in kyanite, conditions during whiteschist metamorphism seem to have enabled higher values (e.g. 
). This is true not only for Mautia Hill, but also for yoderite-bearing whiteschists in Zimbabwe (2 wt. % Fe2O3; Johnson & Oliver, 2002). Manganian andalusite features a great variability in the amount of Mn2SiO5 component (3·0–19·5 mol. %; Fig. 4e). So far, purple yoderite has only been known to occur in whiteschists sensu stricto (McKie, 1959), but, here, it has been found as a late-stage mineral in hornblende–kyanite–talc–quartz schists (Fig. 5c–f). It contains up to 2·2 wt. % Mn2O3, up to 5·5 wt. % Fe2O3 and c. 0·3 wt. % P2O5 (Table 2). Comparing the compositions of the three coexisting Fe–Mn–Al-bearing silicates, Piem, Yod, and Mn–And, which have been formed during the late-stage metamorphic evolution, it appears that the Mn3+/Fe3+ ratio decreases in the order Mn–And, Piem, purple Yod (Fig. 7). Biotite has a conspicuous orange–reddish colour. Microprobe analyses show 2·2 wt. % Mn2O3 (Mntot = Mn3+) and up to 5·5 wt. % BaO (Table 2). Except for the Spl-bearing rocks, haematite is common in all CMASH–CO2–Fe2O3 samples and coexists with rutile (Fig. 8b). It is Mn3+-bearing and contains lamellae of rutile and/or geikielite (Fig. 8c). In hornblende–kyanite–talc–quartz schists, haematite is locally rimmed by hollandite (c. 5 wt. % PbO, nearly Sr-free). Further accessories are apatite (ca. 3 wt. % fluorine) and cerianite.
|
|
|
Rocks of the MASH–Fe2O3 system
Rocks that can be described in this chemical system are whiteschists sensu stricto and kornerupine-bearing sapphirine–enstatite–chlorite schists. Some types of these rocks have been formed as a result of different quartz–chlorite modal ratios in the low-grade precursor rocks (Chl–Qtz schists). Other types have been distinguished on the basis of different late-stage reaction histories. Table 3 shows representative microprobe data of some of the constituent minerals. Minerals of rocks of the MASH–Fe2O3 system found at Mautia Hill are plotted in Fig. 9a.
|
|
Whiteschists. In some quartz-bearing whiteschists, no late-stage reaction rims between talc and kyanite have been developed. The content of talc and quartz in these schists is relatively low, resulting in locally monomineralic domains, consisting purely of greenish kyanite. Furthermore, they contain rutile and haematite (Fig. 8b).
Other talc–kyanite schists contain large amounts of late-stage Al-bearing anthophyllite, which, together with quartz, forms rims between talc and kyanite (Fig. 9e and Fig. 10a). Rutile and haematite are common; the latter contains lamellae of rutile and/or geikielite (Fig. 8c).
|
A third type of whiteschist has already been described by McKie (1959)
. These rocks are Mn-bearing and contain kyanite and talc in a quartz-rich matrix. Around kyanite, there are the well-known reaction rims of purple yoderite and quartz (Figs 9b and 10b). The fourth type shows large kyanite porphyroblasts lying in a matrix of talc. Kyanite is surrounded by pale purple to greenish yoderite, which, itself, is separated from talc by a narrow rim of pinitized cordierite (Fig. 10c). Quartz occurs only as small inclusions in kyanite. The mineral assemblage prior to yoderite formation in this rock type was just Tlc + Ky and the rock was poorer in SiO2 than those described above (Fig. 9c, d). In addition, both yoderite-bearing types of whiteschist contain Rt–Hem symplectites, formed after pseudobrookite coexisting with haematite (Fig. 8a).
Sapphirine–enstatite–chlorite schist. In SiO2-poor MASH–Fe2O3 rocks, neither talc nor kyanite is present. Instead, they consist of magnesiochlorite, enstatite and sapphirine (Figs 9f and 10d). Yellow sapphirine is typically surrounded by kornerupine (Fig. 10d), whereas kornerupine also forms large euhedral crystals, lying within the matrix of the rock. The Spr–En–Chl schists contain haematite and pseudobrookite, which, in most cases, is replaced by Hem–Rt intergrowths (Fig. 8a).
Mineral chemistry of the MASH–Fe2O3 rocks
Talc has a similar composition to talc in the CMASH–CO2–Fe2O3 rocks (Fig. 4d; Table 3). Its Al content ranges from 0·1 to 0·3 p.f.u. (formula based on 11 oxygens). The talc–kyanite schists contain Fe-rich kyanite (up to 2·3 wt. % Fe2O3). Yoderite occurs in two different specifications. The purple yoderite of the Mn-bearing whiteschists is similar in composition to that of the Ca-bearing rocks, but contains only 0·6 wt. % Mn2O3 (Table 3). In Qtz-free whiteschists, pale-purple to green yoderite contains c. 0·2 wt. % Mn2O3, but c. 6 wt. % Fe2O3. Except for the small amount of manganese, it is similar to the green yoderite described by McKie & Bradshaw (1966) from Mautia Hill and Johnson & Oliver (1998) from northern Zimbabwe. Cordierite rimming the green yoderite is strongly pinitized and analyses of the pinitized material resulted in low totals with XMg 0·86 (Table 3). Anthophyllite has XMg = 0·97–0·99 and shows decreasing alumina content with increasing distance from kyanite (1·5–0·5 Al p.f.u., based on 23 oxygens). Magnesiochlorite, which occurs in the kornerupine-bearing schists, is nearly a pure clinochlore, but contains more Al2O3 than chlorite from CMASH–CO2–Fe2O3 rocks (Fig. 4b; Table 3). Enstatite has near end-member composition, but contains up to 5·5 wt. % Al2O3. Sapphirine of the MASH–Fe2O3 rocks is yellow and, thus, similar in appearance to that of Ca-bearing rock types. In the Yod–Crd-bearing whiteschists, sapphirine occurs as lamellar inclusion in Rt–Hem intergrowths and has peraluminous composition (Fig. 4c). Kornerupine, which seems to form at the expense of sapphirine, is pale violet in thin section. Electron microprobe analyses indicate that kornerupine is boron-free (Table 3). This is in agreement with the assumption that Al + Fe3+ + Cr + B = 6·9 p.f.u. (based on 21·5 oxygens; Grew et al., 1990a; Fig. 4f) and the boron-poor chemistry is also approved by ion microprobe analyses (0·04 wt. % B2O3, H. Marschall, personal communication). Our observation contrasts that of McKie (1965) and Grew et al. (1990a), who found 1·92 and 1·46 wt. % B2O3 to occur in kornerupine from Mautia Hill (Fig. 4f). However, both investigated different mineral assemblages from those studied here. McKie (1965) found corundum to occur together with kornerupine. Grew et al. (1990a) described kyanite and tourmaline in close association with kornerupine. In the samples described here, this is not the case. Apparently, both boron-free and boron-bearing kornerupine occur at Mautia Hill. Relics of pseudobrookite (Fe1·46Al0·10Ti1·22Mg0·20Mn0·02O5) are rarely preserved. In most cases, it is replaced by symplectitic intergrowths of rutile and haematite (Fig. 8a). The intergrowths frequently contain small amounts of corundum; in Yod–Crd-bearing whiteschists, Spr is also included. In addition, haematite containing lamellae of rutile forms separate grains, which are included in almost all other minerals, e.g. in kyanite, anthophyllite, yoderite and kornerupine. In Ath-bearing rocks, haematite contains large lamellae of rutile and small exsolution lamellae of near-Mg end-member geikielite (Fig. 8c).
|
THERMOBAROMETRY |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Like all rock types metamorphosed under low geothermal gradients, whiteschists generally feature a clockwise pressure–temperature evolution (e.g. Schreyer & Abraham, 1976
; Johnson & Oliver, 2002; John et al., 2004). In the case of Mautia Hill, where several different rock types have experienced the same metamorphic evolution, commonly used geothermometers and geobarometers can be applied to metabasites and metapelites to obtain the P–T framework for the evolution of the highly oxidized whiteschists and chlorite schists, to which conventional geothermobarometers are not applicable.
First indications of the P–T path are obtained from kyanite inclusions in metapelitic garnet, showing that garnet growth initiated in the stability field of kyanite. Maximum temperatures during uplift are restricted to <720°C by the upper thermal stability of the assemblage yoderite + quartz (Fockenberg & Schreyer, 1994). Our observation of late-stage cordierite reaction rims between yoderite and talc confirms earlier suggestions of a clockwise P–T path (Schreyer, 1977). However, this earlier interpretation was based on the evidence of yoderite reaction rims between talc and kyanite. These rims were assumed to have formed during decompression, but can also form at constant pressures, as a result of oxidation or increase in water activity in the coexisting fluid (see later discussion).
Metapelites
Temperature estimates in metapelites are based on Fe–Mg partitioning between coexisting garnet and biotite. Three different calibrations have been applied (Ferry & Spear, 1978; Hodges & Spear, 1982; Kleemann & Reinhardt, 1994). The estimates were made using mineral formulae, calculated without correcting for ferric iron. Even though the presence of haematite makes it most likely that some ferric iron is present in garnet and biotite, for comparability reasons, it has not been taken into consideration. In any case, the maximum discrepancy between calculated temperatures from assumed Fe3+-free minerals and those obtained with calculated Fe3+ contents amounts only to c. 30°C. Peak temperatures were calculated using analyses of garnet cores and matrix biotite from three samples. The resulting temperatures range from 720 to 740°C at P = 10 kbar (Fig. 11 and Table 4) using the calibration of Kleemann & Reinhardt (1994). This lies slightly outside the stability field of yoderite + quartz, whereas the calculated temperatures using Ferry & Spear (1978) or Hodges & Spear (1982) appear to be too high (T = 835–870°C; Table 4).
|
|
The garnet–aluminosilicate–quartz–plagioclase (GASP) geobarometer has been applied for pressure calculations. The calibration of Koziol & Newton (1989)
gives estimates within the kyanite stability field in agreement with petrographic observations. Peak-metamorphic pressures were obtained using garnet core and plagioclase core compositions. Rim compositions of both minerals were used to determine retrograde pressures. Calculated pressures (at T = 700°C) are in the range of P = 9·5–10·1 kbar (peak) and P = 8·8–9·5 kbar (retrograde; Fig. 11 and Table 4). Microprobe analyses for thermobarometry are given in the Electronic Appendix (http://www.petrology.oupjournals.org).
Metabasites
The garnet–plagioclase–clinopyroxene–quartz (GADS) barometer (Eckert et al., 1991) has been applied to metabasites. Fe–Mg partitioning between coexisting garnet and clinopyroxene (Ellis & Green, 1979; Powell, 1985), and hornblende (Graham & Powell, 1984) have been used for temperature estimates. Peak-metamorphic conditions were calculated using core compositions of minerals; for retrograde conditions, the rim compositions have been used. Clinozoisite-bearing metabasites do not contain much hornblende. In these rocks, temperature estimates are restricted to garnet–clinopyroxene thermometry. For the sake of comparability, all calculations were performed without consideration of ferric iron. The results are listed in Table 5 and shown in Fig. 11. Resulting temperatures are unrealistically high (T > 850°C). Temperatures in clinozoisite-free metabasites are also overestimated (T = 770–800°C). Even results of the garnet–hornblende thermometer lie outside the yoderite + quartz stability field (T 770°C; Fig. 11).
|
Using core compositions of garnet, plagioclase and clinopyroxene, the peak-metamorphic pressures of 10·1–10·7 kbar for both types of metabasites are in good agreement with estimates from GASP equilibria. Retrograde pressures (rim compositions of garnet, plagioclase and clinopyroxene) are in the range of P = 8·3–10·0 kbar. Microprobe analyses for thermobarometry are given in the Electronic Appendix (http://www.petrology.oupjournals.org).
Hornblende–kyanite–talc–quartz schists
Manganian andalusite, the formation of which is assigned to the breakdown of kyanite, quartz and Mn-oxides, shows a variation in the amount of Mn2SiO5 component, from 3 to 19·5%. Under high oxygen fugacities (e.g. MnO2–Mn2O3 buffer), which are supported by the occurrence of oxide minerals such as hollandite, the amount of Mn2SiO5 in manganian andalusite coexisting with hollandite, kyanite and quartz can be used as a geobarometer (Abs-Wurmbach et al., 1983). The application of this barometer yields peak pressures of P = 10–12 kbar for temperatures in the stability field of yoderite + quartz. Mn-andalusites with lower Mn contents probably were-formed during later stages of the decompression path. Alternatively, the low Mn contents may be as a result of local deficiency of manganese on the thin-section scale.
|
REACTION HISTORY |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The schists of Mautia Hill exhibit a number of spectacular reaction textures. In the following paragraphs, these reactions are qualitatively deduced and stoichiometrically balanced using the idealized formulae given in Table 6.
|
Intergrowths of rutile and haematite are common in all types of MASH–Fe2O3 rocks at Mautia Hill. These symplectitic grains mostly show a volumetric ratio of 1/3 rutile and 2/3 haematite (equivalent to c. 45 mol % TiO2 and c. 55 mol % Fe2O3). Therefore, these grains are interpreted as products of decomposition of pseudobrookite during cooling:
 | (1) |
McKie (1963b) reported pseudobrookite from the sapphirine-bearing rocks of Mautia Hill, and we also observed some relics in a kornerupine-bearing schist. Pseudobrookite is a common mineral in, for example, basaltic rocks and miarolitic andesites (Frost & Lindsley, 1991), but, to our knowledge, its occurrence in regional-metamorphic rocks is uncommon. Haematite typically contains lamellae of rutile, which have been described by Haggerty (1991) to be the result of an ‘exsolution’-like process related to oxidation. The exsolution of geikielite (instead of the usual ilmenite) from haematite (Fig. 8c) is in agreement with the high-fO2 conditions, where ilmenite is thought to be unstable. An oxygen fugacity in the range of the MnO2–Mn2O3 buffer or higher is in agreement with the Fe–Mn oxide and silicate mineralogy of the MASH–Fe2O3 and CMASH–CO2–Fe2O3 rocks (Abs-Wurmbach & Peters, 1999).
CMASH–CO2–Fe2O3 system
Textural evidence shows that Mg-Hbl + Spl + Dol + Hem formed the prograde metamorphic assemblage in hornblende–spinel–dolomite schists. Under oxidizing conditions, this assemblage broke down to form Chl, Spr, En and additional Dol:
 | (2) |
 | (3) |
It seems that both reactions are typically coupled in the studied rocks, although it should be possible that they occur independently from each other. The lamellae of haematite and högbomite in spinel can be explained by oxidation exsolution (Buddington & Lindsley, 1964; Grew et al., 1990b) according to the reaction Splss + H2O = Hem + Hög. Yet another indicator for extremely oxidizing conditions is cerianite (Ce4+O2), which forms inclusions in sapphirine or talc and has to be ascribed to metamorphism, because of the textural evidence and its Pan-African U–Th–Pb age (CHIME dating; P. Appel, personal communication). Previously, this mineral has been described as forming at the Earth's surface during weathering processes (Frondel & Marvin, 1959; Braun et al., 1990). Tremolite is a prograde, lower-grade metamorphic relic, forming the cores of zoned amphiboles.
Hornblende–kyanite–talc–quartz schists show the decomposition of the peak-metamorphic assemblage Mg-hornblende + kyanite + haematite. Symplectitic intergrowths of Tlc, Piem, purple Yod and Mn-And partially pseudomorph Mg-Hbl (Fig. 5d–f). Kyanite is rimmed by manganian andalusite-bearing symplectites (Fig. 5c). These textures may be related to the reactions
 | (4) |
 | (5) |
Mass transport via a fluid phase must have taken place at least on the thin-section scale. This is indicated by allochemical replacement of Mg-hornblende. For example, Piem is the only Ca-bearing product phase, but locally missing inside the pseudomorphs. The absence of Mn-andalusite in the prograde mineral assemblage can be explained in two ways: either the pressures were in excess of 13 kbar or the high-fO2 mineral hollandite was missing, which, in our case, was a necessary reactant for the formation of Mn-And at lower pressures. The described reaction texture, showing late-stage formation of hollandite, supports the second possibility. The late-stage minerals contain trivalent and even tetravalent cations, pointing to a minimum fO2 near that of the MnO2–Mn2O3 buffer (Abs-Wurmbach & Peters, 1999).
The occurrence of yoderite in Ca-rich rocks demands special attention: Fockenberg & Schreyer (1994) argued that the tie-line amphibole + kyanite forms a chemical barrier that restricts the formation of yoderite to a Ca-poor bulk chemistry. The coexistence of yoderite and piemontite, formed as late-stage reaction products at the expense of Mg-hornblende + kyanite, is inferred from textural relationships (Fig. 5d and e). This indicates that Mg-hornblende + kyanite + haematite cannot be stable in the whole yoderite stability field. This assemblage appears to be restricted to the high-pressure part of this stability field, because it becomes unstable in the low-pressure part of that stability field. The assumption can be made that Mg-Hbl + Ky + Hem is a high-pressure assemblage in analogy to the whiteschist paragenesis Tlc + Ky + Hem that also reacted to yoderite-bearing assemblages during the late-stage metamorphic evolution.
MASH–Fe2O3 system
Most rock types at Mautia Hill that can be described within the MASH–Fe2O3 system are whiteschists. They are characterized by the mineral assemblage Tlc + Ky, which represents at high pressures a high-temperature equivalent of Mg-Chl + Qtz (Schreyer, 1968; Massonne, 1989). During the prograde burial and heating of the rocks of Mautia Hill, prograde Mg-Chl + Qtz broke down at c. 600°C, forming Tlc + Ky.
McKie (1959) described the Qtz-rich whiteschists that contain purple yoderite and deduced a mineral reaction for the Yod–Qtz rims around kyanite. Later, this reaction was modified by Fockenberg & Schreyer (1991), who recognized that yoderite is not a MASH mineral, but contains essential Fe2O3 and/or Mn2O3. A possible Yod + Qtz-forming reaction is
 | (6) |
 | (7) |
as a high-temperature low-pressure breakdown of Tlc + Ky. At Mautia Hill, the late-stage decompression may be related to this reaction.
The quartz-free Tlc–Ky schists show rims of green yoderite and cordierite between talc and kyanite. The reaction
 | (8) |
). However, breakdown of Tlc + Ky + Hem + H2O takes place at much higher pressures (at c. 13 kbar, high H2O activity). Thus, this Crd-forming reaction may be metastable. Another possible way to form the rims involves a two-stage reaction sequence:
 | (9) |
 | (10) |
The kornerupine-bearing sapphirine–enstatite–chlorite schists are not whiteschists sensu stricto, but their mineralogy and metamorphic evolution are similar to those of whiteschists. In rocks with SiO2-poor bulk chemistry (below the tie-line Tlc–Ky; Fig. 9), the prograde assemblage containing chlorite and very little quartz also breaks down at temperatures of about 600°C, but, here, no Tlc + Ky would form. Instead, the following reaction:
 | (11) |
 | (12) |
determined this mineral reaction experimentally to lie at 5–8 kbar and 750–800°C. Because estimated peak temperatures are lower than 720°C, this reaction must have taken place at still lower temperatures, possibly as a result of incorporation of Fe3+ in kornerupine or a reduced water activity during cooling. |
DISCUSSION AND CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
P–T evolution
The origin of whiteschists is uncertain and two competing models have been proposed. The first suggests metamorphism of an oxidized and metasomatized precursor rock. The second proposes synmetamorphic oxidation and metasomatism. The main factors controlling the formation and subsequent breakdown of the whiteschist assemblage Tlc + Ky are pressure, temperature, oxygen fugacity and water activity. For the rocks of Mautia Hill, maximum pressure conditions of 10–11 kbar are thought to be well constrained by conventional barometry. Retrograde decompression is documented by late-stage formation of cordierite, by mineral growth zoning (Ca increase in plagioclase and Ca decrease in garnet towards the rims) and the GASP and GADS barometric results obtained with mineral rim compositions (Tables 4 and 5). These net-transfer reactions used for barometry are thought to be insensitive to re-equilibration. Furthermore, significant Ca–Al diffusion at temperatures near 700°C is unlikely. In contrast, the results of geothermometers based on Fe–Mg exchange are easily influenced by late re-equilibration and the high oxidation state of the rocks. Therefore, experimentally determined mineral stabilities provide better opportunities for estimation of peak temperatures. The upper thermal stability of Yod + Qtz has been determined in the MASH–Fe2O3 system and estimated for different aH2O (Fockenberg & Schreyer, 1994
). Presumably, a change as a result of the incorporation of Mn2O3 into yoderite is relatively small compared with uncertainties in using the stability fields of sapphirine + H2O (Ackermand et al., 1974) and boron-free kornerupine (Seifert, 1975; Wegge & Schreyer, 1994). The effect of the incorporation of Fe3+ into these minerals is not well known. In addition, the occurrence of anthophyllite (stable below 800°C) and talc (stable below 760°C at 10 kbar; Greenwood, 1963) further constrains maximum peak temperatures and makes a temperature of 700–720°C likely (Fig. 11).
Influence of water activity and oxygen fugacity
To explain the variety of unusual late-stage minerals in the MASH–Fe2O3 and CMASH–CO2–Fe2O3 rocks, the influence of oxygen fugacity (fO2) and water activity (aH2O) has to be considered. Inclusions of haematite and piemontite in kyanite, as well as the Fe2O3 content of peak-metamorphic kyanite and spinel, indicate an oxidized protolith or oxidation during prograde metamorphism. However, breakdown of Mn-haematite + biotite to reaction rims of hollandite around Mn-Hem and formation of rutile lamellae in haematite indicate progressive further oxidation. Assuming maximum pressures of P = 10–11 kbar, it is reasonable to expect that the prograde assemblage Chl + Qtz + Hem breaks down to form directly a yoderite-bearing assemblage. Because Tlc + Ky is formed from Chl + Qtz, yoderite cannot have been stable at this time. Low water activities restrict the stability field of Yod + Qtz to lower temperatures and pressures (<9 kbar; Fig. 11; Fockenberg & Schreyer, 1994). Therefore, our petrographic observations and geobarometric data are in agreement with the hypothesis that the metamorphic conditions involved increasing water activity, leading to the formation of yoderite in the whiteschists of Mautia Hill (P–T path 1; Fig. 11)—a possibility already discussed by Fockenberg & Schreyer (1994). Most former workers, without the benefit of conventional barometry on associated metapelites and metabasites from Mautia Hill, only considered decompression as the cause of the formation of late-stage yoderite rims around kyanite (Schreyer & Yoder Jr, 1968; Schreyer, 1977; Mruma & Basu, 1987; Möller, 1995). If decompression really was the reason for yoderite formation (P–T path 2; Fig. 11), the results of our conventional barometry (GASP, GADS) must be influenced by re-equilibration during decompression, leading to inappropriate pressure estimates: P = 10–11 kbar instead of 13–15 kbar. The absence of prograde Mn-andalusite in the highly oxidized whiteschists, but its late-stage formation, may support the higher-pressure interpretation. However, formation of Mn-And can also be explained by late-stage occurrence of hollandite (as discussed above). The infiltration scenario at maximum pressure conditions of 10–11 kbar, which is proposed here, is in agreement with the results of fluid inclusion studies (Basu & Mruma, 1985; Mruma, 1986; Mruma & Basu, 1987), which point to an early CO2-rich fluid and a late H2O-dominated fluid.
Metasomatism and precursor rock of whiteschists
McKie (1959) speculated on the nature of the precursor rocks and the origin of the special chemistry of whiteschists at Mautia Hill. Although he saw the possibility of their formation through Mg-metasomatism of an argillaceous sandstone, he favoured an interpretation involving isochemical metamorphism of a saponitic bentonite. Later workers demonstrated, on the basis of geochemistry and stable isotope data, that Mg-metasomatism is generally responsible for the formation of whiteschists and leucophyllites (Demény et al., 1997; Pawlig & Baumgartner, 2001; Johnson & Oliver, 2002). We prefer this interpretation for Mautia Hill. The close association and common deformation history of all the rock types at Mautia Hill point to a tectonic origin for the layering, and a subsequent common metamorphic history. We consider it unlikely that the great lithological variety (metabasites, metapelites, metagranites, marbles and various kinds of schists) reflects an original sedimentary succession. The precursor rock of the whiteschists and chlorite schists at Mautia is not obvious, because no transition from these schists into weakly metasomatized lithologies is exposed. Apart from boron-bearing kornerupine (McKie, 1965; Grew et al., 1990a) in chlorite schists and tourmaline in impure marbles and pegmatites, no B2O3-bearing minerals are present, which could provide indications for former evaporites (Moine et al., 1981). The fact that whiteschists and chlorite schists occur only in a thin band between Mn-bearing quartzites and dolomite marbles allows for the interpretation that the protolith was of sedimentary origin, e.g. a karstbauxite (Yalçin et al., 1993). The intense metasomatic overprint does not allow a more detailed interpretation from our petrographic observations.
Geodynamic setting
As discussed above, metamorphism was accompanied by the infiltration of an aqueous fluid (increase in aH2O) at high pressures (P = 10–11 kbar), which led to the development of the oxidized mineralogy of Mn-bearing whiteschists at Mautia Hill. The most likely geodynamic scenario that could account for this observation is that water-rich rocks, buried below the Mautia Hill unit, released their water as a result of progressive metamorphic dehydration. Channelized fluid flow may be responsible for the localized metasomatism and oxidation seen in rocks at Mautia Hill, absent in adjoining rock units. The burial of water-rich rocks below the Mautia Hill unit, which is part of the Pan-African East African Orogen, might be related to subduction of oceanic lithosphere (‘Mozambique Ocean’) in the course of the amalgamation of Gondwana. Alternatively, an intracrustal nappe stacking in the boundary region between the East African Orogen and the adjoining Palaeoproterozoic Usagaran Belt and Archean Tanzania Craton may have initiated prograde metamorphism of buried rocks and the release of fluid phases, causing the metasomatism at Mautia Hill. So far, a Pan-African suture zone between the East African Orogen and the Usagaran Belt, marked by eclogites or ophiolites, has not been documented. Mn-bearing schists, similar to those described here, have been reported to occur further to the southwest in the Konse Series deposited along the border of the Tanzania Craton (Meinhold & Frisch, 1970). Such sediments could have acted as a fluid source if they were buried by Pan-African crustal-thickening processes like those documented at Mautia Hill. In any case, the Sm–Nd garnet age (536 ± 2 Ma; A. Möller, personal communication) suggests that the crustal thickening at Mautia Hill occurred in the late stages of the Pan-African cycle, contrasting with the age of metamorphism in the Mozambique Belt, further east (610–655 Ma; Möller et al., 2000). The significance of the different isotopic ages and the different P–T evolutions for Mautia Hill and the rest of the Tanzanian East African Orogen is not fully understood. The c. 540 Ma event may have succeeded the magmatic underplating process, which is thought to have caused the peak metamorphism, further to the east in the same belt (Appel et al., 1998). Alternatively, the different mineral ages might reflect the different behaviour of the applied isotopic systems: at Mautia Hill, garnet has been dated with the Sm–Nd method, whereas U–Pb monazite ages have been obtained in the rest of the belt. Only further dating with different isotopic methods will resolve the problem. John et al. (2004) obtained similar dating results for the Lufilian Arc–Zambezi Belt orogen of Zambia, where whiteschist metamorphism occurred at the late stages of continental collision at 531–532 ± 2 Ma (U–Pb monazite), following subduction and eclogite formation at about 600 Ma (Sm–Nd Grt–wr).
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data for this paper are available on Journal of Petrology online.
|
ACKNOWLEDGEMENTS |
|---|
We thank T. Heinrichs for supplying important samples and the unpublished map of Mautia Hill. Further thanks are due to P. Appel and B. Mader for their help in conducting the microprobe analyses, to A. Möller for the geochronological information, to A. Fehler for producing the thin sections, to H. Marschall for providing ion microprobe data on the kornerupine chemistry and to T. John for helpful discussions. We appreciate the discussion with D. Lattard concerning Fe–Ti oxide mineralogy and oxygen fugacity. We are grateful for the editorial work of G. Clarke, as well as comments and criticism expressed by W. Schreyer and an anonymous reviewer. We especially thank W. Schreyer for constructive ideas, which helped much to improve the manuscript.
|
FOOTNOTES |
|---|
* Corresponding author. Telephone +49 431 880 3489. Fax: +49 431 880 4457. E-mail: nj{at}min.uni-kiel.de
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPREVIOUS WORK AND GEOLOGICAL...PETROGRAPHY AND MINERAL...THERMOBAROMETRYREACTION HISTORYDISCUSSION AND CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Abraham, K., Hörmann, P. & Raith, M. (1974). Progressive metamorphism of basic rocks from the southern Hohe Tauern area, Tyrol (Austria). Neues Jahrbuch für Mineralogie, Abhandlungen 122, 1–35.
Abs-Wurmbach, I. & Peters, T. (1999). The Mn–Al–Si–O system: an experimental study of phase relations applied to parageneses in manganese-rich ores and rocks. European Journal of Mineralogy 11, 45–68.
Abs-Wurmbach, I., Langer, K. & Schreyer, W. (1983). The influence of Mn3+ on the stability relations of the Al2SiO5 polymorphs with special emphasis on manganian andalusites (viridines), 
: an experimental investigation. Journal of Petrology 24, 48–75.
Ackermand, D., Seifert, F. & Schreyer, W. (1974). Instability of sapphirine at high pressures. Contributions to Mineralogy and Petrology 50, 79–92.
Appel, P., Möller, A. & Schenk, V. (1998). High-pressure granulite facies metamorphism in the Pan-African belt of eastern Tanzania: P–T–T evidence against granulite formation by continent collision. Journal of metamorphic Geology 16, 491–509.[CrossRef][Web of Science]
Armstrong, J. T. (1995). CITZAF: a package of correction programs for the quantitative electron microbeam X-ray analysis of thick polished materials, thin films, and particles. Microbeam Analysis 4, 177–200.
Basu, N. K. & Mruma, A. H. (1985). Mineral chemistry and stability relationship of talc–piemontite–viridine bearing quartzite of Mautia Hill, Mpwapwa District, Tanzania. Indian Journal of Earth Sciences 12, 223–230.
Braun, J. J., Pagel, M., Muller, J. P., Bilong, P., Michard, A. & Guillet, B. (1990). Cerium anomalies in lateritic profiles. Geochimica et Cosmochimica Acta 54, 781–795.[CrossRef][Web of Science]
Buddington, A. F. & Lindsley, D. H. (1964). Iron–titanium oxide minerals and synthetic equivalents. Journal of Petrology 5, 310–357.
Chinner, G. A., Smith, J. V. & Knowles, C. R. (1969). Transition-metal contents of Al2SiO5 polymorphs. American Journal of Science 267A, 96–113.
Chopin, C. & Monié, P. (1984). A unique magnesiochloritoid-bearing, high-pressure assemblage from the Monte Rosa, Western Alps: petrologic and 40Ar–39Ar radiometric study. Contributions to Mineralogy and Petrology 87, 388–398.[CrossRef][Web of Science]
Chopin, C., Henry, C. & Michard, A. (1991). Geology and petrology of the coesite-bearing terrain, Dora Maira massif, Western Alps. European Journal of Mineralogy 3, 263–291.[Web of Science]
Demény, A., Sharp, Z. D. & Pfeifer, H.-R. (1997). Mg-metasomatism and formation conditions of Mg-chlorite–muscovite–quartzphyllites (leucophyllites) of the Eastern Alps (W. Hungary) and their relations to Alpine whiteschists. Contributions to Mineralogy and Petrology 128, 247–260.[CrossRef][Web of Science]
Droop, G. T. R. (1987). A general equation for estimating Fe3+ concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. Mineralogical Magazine 51, 431–435.[Web of Science]
Dyar, M. D., Wiedenbeck, M., Robertson, D., Cross, L. R., Delaney, J. S., Ferguson, K., Francis, C. A., Grew, E. S., Guidotti, C. V., Hervig, R. L., Hughes, J. M., Husler, J., Leeman, W., McGuire, A. V., Rhede, D., Rothe, H., Paul, R. L., Richards, I. & Yates, M. (2002). Reference minerals for the microanalysis of light elements. Geostandards Newsletter 25, 441–463.
Eckert, J. O., Newton, R. C. & Kleppa, O. J. (1991). The 
H of reaction and recalibration of garnet–pyroxene–plagioclase–quartz geobarometers in the CMAS system by solution calorimetry. American Mineralogist 76, 148–160.[Abstract]
Ellis, D. J. & Green, D. H. (1979). An experimental study of the effect of Ca upon garnet–clinopyroxene Fe–Mg exchange equilibria. Contributions to Mineralogy and Petrology 71, 13–22.[CrossRef][Web of Science]
Faryad, S. W. (1999). Metamorphic evolution of the Precambrian South Badakhshan block, based on mineral reactions in metapelites and metabasites associated with whiteschists from Sare Sang (Western Hindu Kush, Afghanistan). Precambrian Research 98, 223–241.[CrossRef][Web of Science]
Ferry, J. M. & Spear, F. S. (1978). Experimental calibration of the partitioning of Fe and Mg between biotite and garnet. Contributions to Mineralogy and Petrology 66, 113–117.[CrossRef][Web of Science]
Fockenberg, T. & Schreyer, W. (1991). Yoderite, a mineral with essential ferric iron: its lack of occurrence in the system MgO–Al2O3–SiO2–H2O. American Mineralogist 76, 1052–1060.[Abstract]
Fockenberg, T. & Schreyer, W. (1994). Stability of yoderite in the absence and in the presence of quartz: an experimental study in the system MgO–Al2O3–Fe2O3–SiO2–H2O. Journal of Petrology 35, 1341–1375.
Frondel, C. & Marvin, U. B. (1959). Cerianite, CeO2, from Pocos de Caldas, Brazil. American Mineralogist 44, 882–884.[Web of Science]
Frost, B. R. & Lindsley, D. H. (1991). Occurrence of iron–titanium oxides in igneous rocks. In: Lindsley, D. H. (ed.) Oxide Minerals: Petrologic and Magnetic Significance. Mineralogical Society of America, Reviews in Mineralogy. 433–468.
Graham, C. M. & Powell, R. P. (1984). A garnet–hornblende geothermometer: calibration, testing, and application to the Pelona schist, southern California. Journal of Metamorphic Geology 2, 13–31.[Web of Science]
Greenwood, H. J. (1963). The synthesis and stability of anthophyllite. Journal of Petrology 4, 317–351.
Grew, E. S., Chernosky, J. V., Werding, G., Abraham, K., Marquez, N. & Hinthorne, J. R. (1990a). Chemistry of kornerupine and associated minerals, a wet chemical, ion microprobe, and X-ray study emphasizing Li, Be, B and F contents. Journal of Petrology 31, 1025–1070.
Grew, E. S., Hiroi, Y. & Shiraishi, K. (1990b). Högbomite from the Prince Olav Coast, East Antarctica: an example of oxidation–exsolution of a complex magnetite solid solution? American Mineralogist 75, 589–600.[Abstract]
Grew, E. S., Pertsev, N. N., Vrána, S., Yates, M. G., Shearer, C. K. & Wiedenbeck, M. (1998). Kornerupine parageneses in whiteschists and other magnesian rocks: is kornerupine + talc a high-pressure assemblage equivalent to tourmaline + orthoamphibole. Contributions to Mineralogy and Petrology 131, 22–38.[CrossRef][Web of Science]
Haggerty, S. E. (1991). Oxide textures: a mini-atlas. In: Lindsley, D. H. (ed.) Oxide Minerals: Petrologic and Magnetic Significance Mineralogical Society of America Reviews in Mineralogy. 129–219.
Hawthorne, F. C. & Henry, D. J. (1999). Classification of the minerals of the tourmaline group. European Journal of Mineralogy 11, 201–215.
Hodges, K. V. & Spear, F. S. (1982). Geothermometry, geobarometry and the Al2SiO5 triple point at Mt. Moosilauke, New Hampshire. American Mineralogist 67, 1118–1134.[Abstract]
Holdaway, M. J. (1971). Stability of andalusite and the aluminium silicate phase diagram. American Journal of Science 271, 97–131.[Abstract]
John, T., Schenk, V., Mezger, K. & Tembo, F. (2004). Timing and P–T evolution of whiteschist metamorphism in the Lufilian Arc–Zambesi Belt orogen (Zambia): implications for the assembly of Gondwana. Journal of Geology 112, 71–90.[CrossRef][Web of Science]
Johnson, S. P. & Oliver, G. J. H. (1998). A second natural occurrence of yoderite. Journal of Metamorphic Geology 16, 809–818.[CrossRef][Web of Science]
Johnson, S. P. & Oliver, G. J. H. (2002). High fO2 metasomatism during whiteschist metamorphism, Zambezi Belt, Northern Zimbabwe. Journal of Petrology 43, 271–290.
Kleemann, U. & Reinhardt, J. (1994). Garnet–biotite thermometry revisited: the effect of AlVI and Ti in biotite. European Journal of Mineralogy 6, 925–941.
Koziol, A. M. & Newton, R. C. (1989). Grossular activity–composition relationships in ternary garnets determined by reversed displaced-equilibrium experiments. Contributions to Mineralogy and Petrology 103, 423–433.[CrossRef][Web of Science]
Kretz, R. (1983). Symbols for rock-forming minerals. American Mineralogist 68, 277–279.[Abstract]
Kulke, H. & Schreyer, W. (1973). Kyanite–talc schist from Sar E Sang, Afghanistan. Earth and Planetary Science Letters 18, 324–328.[CrossRef][Web of Science]
Leake, B. E., Woolley, A. R., Arps, C. E. S., Birch, W. D., Gilbert, M. C., Grice, J. D., Hawthorne, F. C., Kato, A., Kisch, H. J., Krivovichev, V. G., Linthout, K., Laird, J., Mandarino, J., Maresch, W. V., Nickel, E. H., Rock, N. M. S., Schumacher, J. C., Smith, D. C., Stephenson, N. C. N., Ungaretti, L., Whittaker, E. J. W. & Youzhi, G. (1997). Nomenclature of amphiboles: Report of the Subcommittee on Amphiboles of the International Mineralogical Association Commission on New Minerals and Mineral Names. European Journal of Mineralogy 9, 623–651.
Massonne, H.-J. (1989). The upper thermal stability of chlorite + quartz: an experimental study in the system MgO–Al2O3–SiO2–H2O. Journal of Metamorphic Geology 7, 567–581.[Web of Science]
McKie, D. (1959). Yoderite, a new hydrous magnesium iron aluminosilicate from Mautia Hill, Tanganyika. Mineralogical Magazine 32, 282–307.
McKie, D. (1963a). The högbomite polytypes. Mineralogical Magazine 33, 563–580.
McKie, D. (1963b). Order–disorder in sapphirine. Mineralogical Magazine 33, 635–645.
McKie, D. (1965). The magnesium aluminium borosilicates: kornerupine and grandidierite. Mineralogical Magazine 34, 346–357.[Web of Science]
McKie, D. & Bradshaw, N. (1966). A green variety of yoderite. Nature 210, 1148.
McWilliams, M. O. (1981). Palaeomagnetism and Precambrian tectonic evolution of Gondwana. In: Kröner, A. (ed.) Precambrian Plate Tectonics; Developments in Precambrian Geology, Amsterdam: Elsevier, pp. 649–687.
Meinhold, K. D. & Frisch, T. (1970). Manganese–silicate-bearing metamorphic rocks from Central Tanzania. Schweizerische Mineralogische und Petrographische Mitteilungen 50, 493–507.
Moine, B., Sauvan, P. & Jarousse, J. (1981). Geochemistry of evaporite-bearing series: a tentative guide for the identification of metaevaporites. Contributions to Mineralogy and Petrology 76, 401–412.[CrossRef][Web of Science]
Möller, A. (1995). Pan-African granulites and Early Proterozoic eclogites in the basement of eastern Tanzania: P–T–T history and crustal evolution of the complex Mozambique Belt. Doctoral thesis, Christian-Albrechts-Universität, Kiel (Germany).
Möller, A., Mezger, K. & Schenk, V. (2000). U–Pb dating of metamorphic minerals: Pan-African metamorphism and prolonged slow cooling of high pressure granulites in Tanzania, East Africa. Precambrian Research 104, 123–146.[CrossRef][Web of Science]
Mruma, A. H. (1986). P–T regime of the white schist of Mautia Hill, Tanzania. Tanzanian Journal of Science 12, 27–34.
Mruma, A. H. & Basu, N. K. (1987). Petrology of the talc–kyanite–yoderite–quartz schist and associated rocks of Mautia Hill, Mpwapwa District, Tanzania. Journal of African Earth Sciences 6(3), 301–311.[CrossRef]
Pawlig, S. & Baumgartner, L. P. (2001). Geochemistry of a talc–kyanite–chloritoid shear zone within the Monte Rosa granite, Val d'Ayas, Italy. Schweizerische Mineralogische und Petrographische Mitteilungen 81, 329–346.
Powell, R. (1985). Regression diagnostics and robust regression in geothermometer/geobarometer calibration: the garnet–clinopyroxene geothermometer revisited. Journal of Metamorphic Geology 3, 231–243.[Web of Science]
Råheim, A. & Green, D. H. (1974). Talc–garnet–kyanite–quartz schist from eclogite-bearing terrane, Western Tasmania. Contributions to Mineralogy and Petrology 43, 223–231.[CrossRef][Web of Science]
Rickers, K. (1996). Metamorphose am Südostrand des Tansania–Kratons. Diploma thesis, Christian-Albrechts-Universität, Kiel (Germany).
Rolfo, F., Compagnoni, R., Xu, S. & Jiang, L. (2000). First report of felsic whiteschists in the ultrahigh-pressure metamorphic belt of Dabie Shan, China. European Journal of Mineralogy 12, 883–898.
Sandiford, M., Neall, F. B. & Powell, R. (1987). Metamorphic evolution of aluminous granulites from Labwor Hills, Uganda. Contributions to Mineralogy and Petrology 95, 217–225.[CrossRef][Web of Science]
Schreyer, W. (1968). A reconnaissance study of the system MgO–Al2O3–SiO2–H2O at pressures between 10 and 25 kb. Carnegie Institution of Washington Yearbook 66, 380–392.
Schreyer, W. (1977). Whiteschists: their compositions and pressure–temperature regimes based on experimental, field, and petrographic evidence. Tectonophysics 43, 127–144.[CrossRef][Web of Science]
Schreyer, W. (1986). The mineral cordierite: structure and reactions in the presence of fluid phases. Berichte der Bunsengesellschaft für Physikalische Chemie 90, 748–755.
Schreyer, W. (1988). Experimental studies on metamorphism of crustal rocks under mantle pressures. Mineralogical Magazine 52, 1–26.
Schreyer, W. & Abraham, K. (1975). Peraluminous sapphirine as a metastable reaction product in kyanite–gedrite–talc schists from Sar e Sang, Afghanistan. Mineralogical Magazine 40, 171–180.[Web of Science]
Schreyer, W. & Abraham, K. (1976). Three-stage metamorphic history of a whiteschist from Sar e Sang, Afghanistan, as part of a former evaporite deposit. Contributions to Mineralogy and Petrology 59, 111–130.[CrossRef][Web of Science]
Schreyer, W. & Seifert, F. (1969). Compatibility relations of the aluminium silicates in the systems MgO–Al2O3–SiO2–H2O and K2O–MgO–Al2O3–SiO2–H2O at high pressures. American Journal of Science 267, 371–388.[Abstract]
Schreyer, W. & Yoder, H. S., Jr (1968). Yoderite: synthesis, stability, and interpretation of its natural occurrence. Carnegie Institution of Washington Yearbook 66, 376–393.
Seifert, F. (1975). Boron-free kornerupine: a high-pressure phase. American Journal of Science 275, 57–87.
Simon, G. & Chopin, C. (2001). Enstatite–sapphirine crack-related assemblages in ultrahigh-pressure pyrope megablasts, Dora-Maira massif, western Alps. Contributions to Mineralogy and Petrology 140, 422–440.[CrossRef][Web of Science]
Stern, R. J. (1994). Arc assembly and continental collision in the Neoproterozoic East African Orogen: implications for the consolidation of Gondwanaland. Annual Reviews of Earth and Planetary Sciences. 319–351.
Temperley, B. N. (1938). The geology of the country around Mpwapwa, Tanganyika Territory, Department of Lands and Mines, Geological Division, Dar es Salaam.
Vrána, S. & Barr, M. W. C. (1972). Talc–kyanite–quartz schist and other high-pressure assemblages from Zambia. Mineralogical Magazine 38, 837–846.[Web of Science]
Wegge, S. & Schreyer, W. (1994). Boron-free kornerupine: its upper pressure stability limit in the system MgO–Al2O3–SiO2–H2O (MASH). European Journal of Mineralogy 6, 67–75.
Windley, B. F., Ackermand, D. & Herd, R. K. (1984). Sapphirine/kornerupine-bearing rocks and crustal uplift history of the Limpopo belt, Southern Africa. Contributions to Mineralogy and Petrology 86, 342–358.[CrossRef][Web of Science]
Yalçin, Ü., Schreyer, W. & Medenbach, O. (1993). Zn-rich högbomite from gahnite in the metabauxites of the Menderes Massif, SW Turkey. Contributions to Mineralogy and Petrology 113, 314–324.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||