Scholarly article on topic 'New constraints on the P–T path of HT/UHT metapelites from the Highland Complex of Sri Lanka'

New constraints on the P–T path of HT/UHT metapelites from the Highland Complex of Sri Lanka Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Geoscience Frontiers
Keywords
{"Prograde evolution" / "Melt re-integration" / "Highland Complex" / "Sri Lanka" / "UHT granulites"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — P.L. Dharmapriya, Sanjeewa P.K. Malaviarachchi, L.M. Kriegsman, Andrea Galli, K. Sajeev, et al.

Abstract We report here rare evidence for the early prograde P–T evolution of garnet–sillimanite–graphite gneiss (khondalite) from the central Highland Complex, Sri Lanka. Four types of garnet porphyroblasts (Grt1, Grt2, Grt3 and Grt4) are observed in the rock with specific types of inclusion features. Only Grt3 shows evidence for non-coaxial strain. Combining the information shows a sequence of main inclusion phases, from old to young: oriented quartz inclusions at core, staurolite and prismatic sillimanite at mantle, kyanite and kyanite pseudomorph, and biotite at rim in Grt1; fibrolitic sillimanite pseudomorphing kyanite ± corundum, kyanite, and spinel + sillimanite after garnet + corundum in Grt2; biotite, sillimanite, quartz ± spinel in Grt3; and ilmenite, rulite, quartz and sillimanite in Grt4. The pre-melting, original rock composition was calculated through stepwise re-integration of melt into the residual, XRF based composition, allowing the early prograde metamorphic evolution to be deduced from petrographical observations and pseudosections. The earliest recognizable stage occurred in the sillimanite field at around 575 °C at 4.5 kbar. Subsequent collision associated with Gondwana amalgamation led to crustal thickening along a P–T trajectory with an average dP/dT of ∼30 bar/°C in the kyanite field, up to ∼660 °C at 6.5 kbar, before crossing the wet-solidus at around 675 °C at 7.5 kbar. The highest pressure occurred at P > 10 kbar and T around 780 °C before prograde decompression associated with further heating. At 825 °C and 10.5 kbar, the rock re-entered into the sillimanite field. The temperature peaked at 900 °C at ca. 9–9.5 kbar. Subsequent near-isobaric cooling led to the growth of Grt4 and rutile at T ∼880 °C. Local pyrophyllite rims around sillimanite suggest a late stage of rehydration at T < 400 °C, which probably occurred after uplift to upper crustal levels. U–Pb dating of zircons by LA-ICPMS of the khondalite yielded two concordant 206Pb/238U age groups with mean values of 542 ± 2 Ma (MSWD = 0.24, Th/U = 0.01–0.03) and 514 ± 3 Ma (MSWD = 0.50, Th/U = 0.01–0.05) interpreted as peak metamorphism of the khondalite and subsequent melt crystallization during cooling.

Academic research paper on topic "New constraints on the P–T path of HT/UHT metapelites from the Highland Complex of Sri Lanka"

BIOSCIENCE FRONTIERS

Accepted Manuscript

New constraints on the P-T path of HT/UHT metapelites from the Highland Complex of Sri Lanka

P.L. Dharmapriya, Sanjeewa P.K. Malaviarachchi, L.M. Kriegsman, Andrea Galli, K. Sajeev, Chengli Zhang

PII: S1674-9871(17)30027-0

DOI: 10.1016/j.gsf.2016.12.005

Reference: GSF 534

To appear in: Geoscience Frontiers

Received Date: 4 September 2016

Revised Date: 5 December 2016

Accepted Date: 18 December 2016

Please cite this article as: Dharmapriya, P.L., Malaviarachchi, S.P.K., Kriegsman, L.M., Galli, A., Sajeev, K., Zhang, C., New constraints on the P-T path of HT/UHT metapelites from the Highland Complex of Sri Lanka, Geoscience Frontiers (2017), doi: 10.1016/j.gsf.2016.12.005.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 New constraints on the P—T path of HT/UHT metapelites

2 from the Highland Complex of Sri Lanka

3 P. L. Dharmapriyaa'b'c, Sanjeewa P. K. Malaviarachchiab*,

4 L. M. Kriegsmanc'd, Andrea Gallie, K. Sajeevf, Chengli Zhangg

5 a Postgraduate Institute of Science, University of Peradeniya, 20400, Sri Lanka

6 b Department of Geology, Faculty of Science, University of Peradeniya, 20400, Sri Lanka

7 c Naturalis Biodiversity Center, Darwinweg 2, NL-2333 CR Leiden, The Netherlands

8 d Department of Earth Sciences, Utrecht University, Budapestlaan 4, NL-3584 CD Utrecht,

9 The Netherlands

10 e Department of Earth Sciences, ETH Zurich, Sonnegstrasse 5, CH-8092 Zurich, Switzerland

11 Centre for Earth Sciences, Indian Institute of Science, Bangalore 560012, India

12 g Key Laboratory of Continental Dynamics of Northwest University, Department of Geology,

13 Northwest University, Xi'an 710069, China

18 * Corresponding Authors: Sanjeewa, P.K Malaviarachchi

19 E-mail addresses : malavi@pdn.ac.lk,

20 Postal Addresses : Department of Geology, Faculty of Science, University of

21 Peradeniya, 20400, Sri Lanka.

22 Phone No : +94 81 2394215

ABSTRACT

We report here rare evidence for the early prograde P-T evolution of garnet-sillimanite-graphite gneiss (khondalite) from the central Highland Complex, Sri Lanka. Four types of garnet porphyroblasts (Grti, Grt2, Grt3 and Grt4) are observed in the rock with specific types of inclusion features. Only Grt3 shows evidence for non-coaxial strain. Combining the information shows a sequence of main inclusion phases, from old to young: oriented quartz inclusions at core, staurolite and prismatic sillimanite at mantle, kyanite and kyanite pseudomorph, and biotite at rim in Grt1; fibrolitic sillimanite pseudomorphing kyanite ± corundum, kyanite, and spinel+sillimanite after garnet+corundum in Grt2; biotite, sillimanite, quartz ± spinel in Grt3 and: ilmenite, rulite, quartz and sillimanite in Grt4.

The pre-melting, original rock composition was calculated through stepwise reintegration of melt into the residual, XRF based composition, allowing the early prograde metamorphic evolution to be deduced from petrographical observations and pseudosections. The earliest recognizable stage occurred in the sillimanite field at around 575 °C at 4.5 kbar. Subsequent collision associated with Gondwana amalgamation led to crustal thickening along a P-T trajectory with an average dP/dTof ~30 bar/0C in the kyanite field, up to ~660 °C at 6.5 kbar, before crossing the wet-solidus at around 675 °C at 7.5 kbar. The highest pressure occurred at P> 10 kbar and T around 780 °C before prograde decompression associated with further heating. At 825 °C and 10.5 kbar, the rock re-entered into the sillimanite field. The temperature peaked at 900 °C at ca. 9-9.5 kbar. Subsequent near-isobaric cooling led to the growth of Grt4 and rutile at T ~880 °C. Local pyrophyllite rims around sillimanite suggest a late stage of rehydration at T< 400 °C, which probably occurred after uplift to upper crustal levels. U-Pb dating of zircons by LA-ICPMS of the khondalite yielded two concordant 206Pb/238U age groups with mean values of 542 ± 2 Ma (MSWD = 0.24, Th/U = 0.01-0.03) and 514 ± 3 Ma (MSWD = 0.50, Th/U = 0.01-0.05) interpreted as peak metamorphism of the khondalite and subsequent melt crystallization during cooling.

Keywords - Prograde evolution, melt re-integration, Highland Complex, Sri Lanka, UHT granulites

1. Introduction

Granulites give information on the petrological evolution of the Earth's middle to lower crust (e.g. Ouzegane et al., 2003). Petrologists have shown great interest in aluminous pelitic granulites, as these rock types may preserve peak mineral assemblages and a vast

58 range of reaction textures useful to reconstruct the near peak P-T evolution (e.g. Waters,

59 1986; Hensen, 1987; Droop, 1989). Nevertheless, in numerous high-grade terrains, most

60 evidence on the prograde evolution is obliterated due to strong ductile deformation at high

61 temperatures (Whitney and Dilek, 1997; Mathavan et al., 1999) or by peak metamorphic

62 equilibration. Thus, the preservation of prograde metamorphic stages is generally uncommon

63 (e.g., Hiroi et al., 1994; Raase and Schenk, 1994; Ouzegane et al., 2003). Exceptionally,

64 crucial information for the reconstruction of early stages of a pressure-temperature-time (P-

65 T-t) path for a specific tectonic domain may be preserved as relict reaction textures and/or

66 microstructures trapped as single or composite inclusions within mineral porphyroblasts,

67 commonly garnet, during the prograde metamorphic history.

68 In this study, we report inclusion microtextures associated with garnet evidencing

69 early prograde metamorphismof corundum-spinel-kyanite-staurolite bearing garnet-

70 sillimanite-graphite gneiss (khondalite) from the Highland Complex of Sri Lanka. Coupling

71 textural observations (textures associate with in garnet and in the matrix) and thermodynamic

72 modeling we reconstruct the prograde to retrograde P-T path followed by the khondalite.

73 Also we present related U-Pb zircon geochronology data of studied rock. After summarizing

74 results of this study and most recently published petrological and U-Pb zircon

75 geochonological data of HT/UHT metasediments of the Highland Complex, we attempted

76 reconstruction of P-T-t path followed by the Highland Complex metasediments.

78 2. General geology of Sri Lanka

79 Sri Lanka represents a small but important crustal fragment of eastern Gondwana. On

80 the basis of Nd model ages and zircon U-Pb dating (Milisenda et al., 1988, 1994; Kroner et

81 al., 1991; Liew et al., 1991), the Proterozoic basement of Sri Lanka has been subdivided into

82 four litho-tectonic units (Cooray, 1994; Fig. 1a), namely from west to east the Wanni

83 Complex (WC), Kadugannawa Complex (KC), Highland Complex (HC) and Vijayan

84 Complex (VC).

86 2.1. Overview of geology, petrology and geochronology of Sri Lankan basement

88 2.1.1. The Highland Complex (HC)

89 The HC contains granulite facies metasedimentary and metaigneous rocks including

90 quartzites, marbles, calcsilicates, pelitic gneisses, charnockites and orthogneisses (Cooray,

91 1962, 1984, 1994; Mathavan and Fernando, 2001). In the central and northern part of the HC,

hundreds of meters' thick marble and quartzite units are traceable for more than 40 km. In contrast, in the southwestern part of the HC, marble and quartzite are scarce (Mathavan et al., 1999) and cordierite-bearing metapelitic gneisses, orthogneisses and thin bands of wollastonite- and scapolite-bearing calcsilicates are the prominent rock types (Perera, 1984; Prame, 1991).

Using conventional thermobarometric calculations and petrogenetic grids, the HC has been classically interpreted as a tilted crustal section with a peak metamorphic gradient increasing from 4.5-6 kbar and 700-750 °C in the southwest up to 8-9 kbar and 800--900 °C in the east and southeast (Faulhaber and Raith, 1991; Raase and Schenk, 1994; Schumacher and Faulhaber, 1994; Kriegsman, 1996; Mathavan et al., 1999; Kriegsman and Schumacher, 1999; Braun and Kriegsman, 2003). In addition, rare UHT granulites that formed at extreme crustal conditions of 925-1150 °C and 9-12.5 kbar have been reported from a few localities (Fig. 1) in the central and southwestern HC (e.g. Osanai, 1989; Kriegsman and Schumacher, 1999; Bolder-Schrijver et al., 2000; Osanai et al., 2000; Sajeev and Osanai, 2004a, b; Osanai et al., 2006; Sajeev et al., 2007; Dharmapriya et al., 2015a). These UHT granulite conditions were suggested using conventional thermobarometric calculations and petrogenetic grids (Osanai, 1989; Osanai et al., 2000, Sajeev and Osanai, 2004a, 2004b; Osanai et al., 2006, 2016a, b) and pseudosection modeling (e.g. Sajeev et al., 2007; Dharmapriya et al., 2015a)

The HC yields Nd-model ages of 3400-2000 Ma (Milisenda et al., 1988, 1994). Detrital zircons separated from the HC granulite facies metasediments suggest provenance U-Pb ages between 3200 Ma and 2000 Ma (Kroner et al., 1987; Holzl et al., 1991, 1994), and a peak metamorphic age around 610-550 Ma (e.g Baur et al., 1991; Holzl et al., 1991, 1994; Kroner et al.,1994). Discordant zircons from the HC orthogneisses yields 2000-1850 Ma and 670 Ma as upper intercept ages, which have been interpreted as the timing of pluton emplacement, and 610-530 Ma as lower intercept ages, which have been interpreted as metamorphic ages (e.g., Baur et al., 1991; Holzl et al., 1991, 1994; Kroner and Williams, 1993). Similar metamorphic zircon ages of c. 570 Ma have been reported by Sajeev et al. (2007) from HP/UHT mafic granulites.

In contrast, Sajeev et al. (2010) reported U-Pb zircon and monazite ages from the HC UHT metapelits in the HC clustering around 2300 Ma, 1700 Ma and 1400-830 Ma, some of those having metamorphic overgrowths at c. 570 Ma. Using CHIME dating for monazite in metasediments, Malaviarachchi and Takasu (2011) reported a wide range of metamorphic ages from 728 to 460 Ma. Most recently, Dharmapriya et al. (2015b, 2016) reporteddetrital zircon U-Pb ages from ca. 2800 Ma to 720 Ma from UHT metapelites indicating that the

126 metasediments of the HC have been derived from Neoarchean to Neoproterozoic multiple

127 provenances during Neoproterozoic ear (after 700 Ma).Takamura et al. (2016) also have

128 provided evidence for Neoproterozoic sedimentation of the HC and authors have reported

129 detrital zircon age spectrum from ca. 3500 to < 700 Ma. Dharmapriya et al. (2016) reported

130 that evidence for multiple metamorphic events from ca. 665 Ma to 530 Ma with peak UHT

131 metamorphism around 580 Ma - 530 Ma. In addition, Santosh et al. (2014), He et al. (2015)

132 and Takamura et al.(2015) reported multiple late Neoproterozoic to early Cambrian U-Pb

133 spectra of zircons from metaigneous rocks in the HC. Santosh et al. (2014) reported Hf

134 crustal model ages of zircon from mafic and intermediate granulites and charnockites in the

135 range of 2800-1500 Ma.

136 The WC, VC and KC rocks (see Santosh et al., 2014; Dhatmapriya et al., 2015a, b; He et al.,

137 2015, 2016 for further details) have yielded Nd-model ages of 2000-1000 Ma, 1800-1100 Ma and

138 2000-1000 Ma (Milisenda et al., 1988; 1994) respectively and were metamorphosed under upper

139 amphibolite to granulite facies conditions (e.g. Cooray, 1994; Kehelpannala, 1997; Mathavan et al.,

140 1999; Kroner et al., 2013; He et al., 2015, 2016).

142 2.2. Overview of the deformation history of the Sri Lankan basement

144 Tthe Sri Lankan basement underwent a polyphase deformational evolution (e.g.,

145 Berger and Jayasinghe 1976; Kriegsman, 1991, 1994, 1995; Kleinschrodt, 1994;

146 Kehelpannala, 1997). Berger and Jayasinghe (1976) suggested that the basement was

147 subjected to at least three deformation phases in which D1 and D2 formed the major lineation

148 and a composite foliation (L-S fabric), whereas D3 resulted in the formation of large-scale

149 upright folds. The same authors, as well as Kriegsman (1991, 1994) and Kehelpannala (1997)

150 suggested that D2 was coeval with the peak of metamorphism during Gondwana assembly

151 (e.g., Kroner et al., 2003; Kehelpannala, 2004). Evidence for D1 is preserved only within

152 garnet porphyroblasts as aligned mineral inclusions defining a well-developed crenulation S1

153 (Kehelpannala, 1991; Kriegsman, 1991) oblique to the main matrix foliation S2 (Kriegsman,

154 1991, 1994; Kehelpannala, 1997).

155 A slightly different interpretation is given by Kehelpannala (1997), who suggested

156 that the HC underwent six phases of ductile deformation, where D1 to D3 are similar to those

157 described by earlier workers. However, the author argued that D4 produced large, gentle,

158 nearly E-W trending, upright folds, whereas D5 was responsible for the large-scale upright

159 folds and D6 caused local refolding of the D5 structures.

161 2.3. Suggested P-T trajectories for Sri Lankan basement rocks

163 As illustrated by Fig. 2, the reconstruction of the P-T path for the HC granulites is

164 still a matter of debate and no general consensus exists on the P-T trajectory followed by the

165 HC. According to many studies, which reported sillimanite both as inclusion and as matrix

166 mineral in garnet-bearing metapelites, the equilibration at peak T occurred in the sillimanite

167 field (e.g., Perera, 1987, 1994; Hiroi et al., 1994; Raase and Schenk, 1994, Fig. 2a-f). Since

168 relict kyanite and staurolite have been observed as inclusions in porphyroblastic garnet in

169 sillimanite-bearing metapelites (Hiroi et al., 1994; Raase and Schenk, 1994), most of the

170 workers suggested a clockwise P-T trajectory for the HC granulites.

171 Kriegsman (1993, Fig. 2c) and Raase and Schenk (1994, Fig. 2d) inferred a strong

172 pressure increase at amphibolite facies conditions from the sillimanite field to the kyanite

173 field, probably due to crustal thickening during prograde metamorphism. Kriegsman (1993)

174 inferred that P increased up to peak T, while Raase and Schenk (1994) suggested that peak P

175 was followed by near isobaric heating from the kyanite to the sillimanite field up to peak T

176 (Fig. 2d). Alternatively, several workers (e.g., Ogo et al., 1992; Hiroi et al., 1994; Osanai et

177 al., 2006; Dharmapriya et al., 2015a,b) argued that peak P was followed by prograde

178 decompression up to peak T from different peak P conditions.

179 After peak T (in the sillimanite field), many studies suggested a period of near

180 isobaric cooling (IBC), followed by rapid near isothermal decompression (ITD) within the

181 sillimanite field (Perera, 1987, 1994; Schumacher et al., 1990; Faulhaber and Raith, 1991,

182 Prame, 1991; Raase and Schenk, 1994; Mathavan and Fernando, 2001; Osanai et al., 2006).

183 In contrast, other works suggested an ITD stage directly after peak T (Ogo et al., 1992;

184 Kriegsman, 1993; Hiroi et al., 1994; Takamura et al., 2015).

185 Hiroi et al. (1994) and Raase and Schenk (1994) showed evidence for local growth of

186 andalusite indicating that the latest metamorphic stage of the HC occurred in the andalusite

187 field.

189 3. Field relations and sample descriptions

190 Highly aluminous garnet-sillimanite-graphite gneisses (khondalite) were collected

191 from a road exposure south of Gampola (Fig. 1a and b). This area mainly consists of

khondalites, quartzites, charnockitic and granitic gneisses, and marbles (Fig. 1b). The investigated khondalite occurs as slightly weathered rocks within the road cut. The sampling point is located close to the inferred ductile shear zone separating the HC from the KC (Kröneret al., 1991; Voll and Kleinschrodt, 1991; Cooray, 1994). Rocks of this area exhibit strong ductile deformation features such as ribbon quartz, recrystallized elongate feldspars and a well demarcated sillimanite lineation.

The khondalite, which is overlain by a ~30 cm thick quartzite layer, is characterized by coarse-grained, euhedral to subhedral garnet porphyroblasts, with diameters ranging from 0.25 to 3 cm (Fig. 3a-c). The rock displays a well-developed foliation defined by stretched quartz (up to 5 cm long) and coarse prismatic sillimanite needles (up to 1.3 cm long). Sillimanite is present within the main foliation between elongated quartz grains or wrapping around porphyroblastic garnet. Elongate feldspars and fine- to medium-grained graphite flakes can also be identified. The strike and dip of the foliation of the exposure is N300E and 250NW.

4. Petrography

4.1. Textures of garnet porphyroblasts

Four types of garnet (Grt1, Grt2, Grt3 and Grt4) can be distinguished on the basis of specific inclusion features, from the collected samples from the road exposure. Rarely, three types of garnet (Grt1, Grt3 and Grt4) can be identified in approximately 10 cm x 10 cm sized rock specimens. Grt2 are relatively less abundance and occurred only some domains where matrix quartz are relatively less abundance compare to rest of the portions of the rock .

Garnet type 1 (Grt1): these porphyroblasts display euhedral to subhedral crystal shapes and are 1 to 1.5 cm in diameter. The core is mainly inclusion-free while the mantle contains numerous oriented quartz inclusions with minor ilmenite, which define a discontinuous internal foliation (Si) oblique to the matrix foliation (Fig. 4a and b). Towards the mantle area, there is prismatic, up to 0.5 mm long sillimanite, isolated or in contact with quartz (Fig. 4a-d), and isolated staurolite grains (Fig. 4a). The orientation of staurolite (close to the top-right edge of the garnet in Fig.4a) is parallel to Si. Staurolite (Fig. 4d and e), isolated biotite flakes (Fig. 4a and e), up to 0.4 cm long kyanite and minor ilmenite inclusions are present within the rim (Fig. 4f). Rarely, euhedral kyanite inclusions are partially pseudomorposed by a second generation of sillimanite. In this texture (Fig. 4f), the kyanite core, with typical low order interference colours and oblique extinction, is still preserved,

while the rim has been converted into pseudomorphic sillimanite with higher order interference colours and parallel extinction.

Garnet type 2 (Grt2): the core and mantle area of these euhedral to subhedral porphyroblasts (up to 3 cm in diameter) contain numerous isolated inclusions of anhedral, medium- to fine-grained corundum (0.5 - 4 mm in size) coexisting with clusters of sillimanite (Fig. 5a and b). The subhedral and monoclinic (Fig. 5a) shape of these clusters suggests that sillimanite may have grown as pseudomorphs after kyanite. Medium- to coarsegrained prismatic sillimanite (up to 4 mm long) and medium-grained spinel also occur associated with corundum grains (Fig. 5d). Occasionally, prismatic sillimanite has numerous of spinel inclusions (Fig. 5e-g) and rare corundum inclusions (Fig. 5e-g). Isolated anhedral staurolite (up to 0.3 mm) is also present in the core and mantle (Fig. 5e). Towards the mantle to rim contact, garnet encloses isolated quartz grains, medium- to fine-grained clusters of alkali-feldspar (Fig. 5h) and patches of pyrophyllite, which often contain tiny, partially consumed sillimanite needles (Fig. 5i). Additionally, the core to rim area of garnet contains

oriented rutile needles and minor apatite rods with ilmenite (Fig. 5j and k). Some ilmenite

needles show orientations at an angle of approximately 45 - 60 (Fig. 5k). Similar to Ague and Eckert (2012), we interpret these as exsolution features.

Garnet type 3 (Grt3): the presence of medium- to fine-grained, elongated and locally curved quartz grains together with minor alkali-feldspar, rutile and ilmenite inclusions at the core and inner mantle areas of these coarse-grained (up to 2.5 cm in diameter), rounded porphyroblasts (Fig. 6a and b).Curved quartz inclusion could indicate that this particular area of Grt3 formed under strong non-coaxial stress. The outer mantle and rim areas lack evidence for rotation and contain fewer inclusions than core and inner mantle. Thus, the outer mantle probably formed under the absence of a strong non-coaxial stresses. The outer mantle to the rim parts of Grt3 contain relict biotite flakes, coexisting with sillimanite and fine-grained spinel (Fig. 6c). Relatively coarse-grained alkali-feldspars (~3 mm) are found throughout the garnet. Locally, in the mantle to rim areas, sillimanite is found rimmed by pyrophyllite (Fig. 6d).

Garnet type 4 (Grt4): is fine-grained (up to 0.25 - 0.60 cm in diameter), anhedral and encloses oriented sillimanite needles and minor quartz, alkali-feldspar, zircon and rutile (Fig 6e-g). Oriented sillimanite inclusions display the same orientations as sillimanite grains in the rock matrix (Fig. 6e). Theses garnets occur close to coarse-grained sillimanite, ilmenite, quartz (Fig. 6f-i) and alkali-feldspars in the matrix, which is described in the following section. In addition to these inclusions, euhedral rutile is present in the matrix close to Grt4

260 (Fig. 6g). Occasionally, partially broken-down sillimanite, quartz and ilmenite grains present

261 adjacent to garnet-rutile intergrowth (Fig. 6i).

262 As a summary, Grt1contains oriented quartz inclusions and minor ilmenite in the core,

263 prismatic sillimanite and staurolite in the mantle and, kyanite and biotite in the rim. Grt2

264 encloses isolated corundum grains (0.5 - 4 mm) associated with clusters of fibrolitic

265 sillimanite, medium-grained spinel and isolated staurolite in both core and mantle, and

266 isolated quartz grains towards the rim. Grt3 displays curved quartz, alkali-feldspar, rutile and

267 ilmenite inclusions in the core and inner mantle, whereas the outer mantle to rim area

268 contains relict biotite, sillimanite and spinel and does not show evidence for rotation. Also

269 only Grt3 shows evidence for non-coaxial strain. Garnet type 4 (Grt4) is finer-grained,

270 anhedral and encloses oriented sillimanite needles, minor quartz, alkali-feldspar, zircon and

271 rutile.

273 4.2. Matrix minerals

275 The matrix of the studied khondalite consists of garnet, prismatic, coarse-grained

276 sillimanite, stretched quartz and alkali-feldspars with variable size (from 0.3 up to 1 cm) as

277 major constituents, plagioclase, ilmenite and rutile as minor constituents and rounded zircon

278 grains and fine to medium graphite flakes as an accessory phase. Elongated quartz and

279 coarse-grained sillimanite needles show a preferred orientation and define the main lineation

280 (Fig. 7a and b). Rarely, garnet porphyroblasts are also elongated parallel to quartz (Fig. 7a).

281 Medium-grained quartz may contain many sillimanite inclusions (Fig. 7b), while coarse-

282 grained, elongated quartz may also bear tiny sillimanite inclusions in the rims (Fig.7a).

283 Sillimanite clusters (from 0.2 up to 1.5 cm long) are observed among quartz (Fig. 7a-e)

284 coexisting with garnet (Fig. 7e). Up to 2 mm long, anhedral ilmenite frequently occurs among

285 prismatic sillimanite (Fig. 7c). Coarse-grained, euhedral to subhedral alkali-feldspar is mainly

286 observed within the pressure shadows of some porphyroblastic Grt2 (Fig. 7d). Rarely, thin

287 pyrophyllite rims have formed around matrix sillimanite in the presence of quartz (Fig. 7f).

288 Tiny rutile grains only occur around Grt4 (Fig. 6g).

290 5. Whole rock chemistry, mineral chemistry and laser raman analysis

291 5.1. Whole rock chemistry

292 The bulk composition of the khondalite was obtained using a Panalytical Axios wave-

293 length dispersive XRF spectrometer (WDXRF, 2.4 kV) at ETH Zurich, Switzerland. A

representative restitic sample (10 cm x 10 cmx 5 cm; in which Grti, G1I3 and G1I4 was occurred) was reduced in grain size with a hydraulic press in a stainless-steel beaker and grinded to powder with an agate mill. Standard analyses were performed on fused glass-beads prepared from rock powder mixed with Lithium-Tetraborate (1:5 mixture) using a Claisse M4® fluxer. The calibration is based on -40 certified international standards. The bulk rock composition is given in Table 1 (Cxrf of Table 1). The investigated khondalite is rich in Si02 (-66.5 wt.%, S/SFM = 0.74), displays a moderate A1203 content (-19 wt.%, A/AFM = 0.47), is poor in MgO relative to FeO (-1.5 wt.% vs. 7.3 wt.%, XMg = 0.26) and shows CaO < Na20 < K20 contents (-0.9, 1.3 and 2.7 wt.%, respectively).

5.2. Mineral chemistry

Mineral compositions were analyzed using a JEOL JXA8530 Field Emission Electron Probe Microanalyzer (EPMA) owned by Naturalis Biodiversity Center and operated by a national consortium at the University of Utrecht, Netherlands. All analyses were carried out using an accelerating voltage of 15 kV and 20 nA beam current, 1-3 pm spot size. Standards used include natural diopside for Si and Ca, forsterite for Mg, hematite for Fe, corundum for Al, tephroite for Mn, synthetic KTiPOs for K, synthetic Ti02 for Ti, jadeite for Na, pure chromium for Cr and pure nickel for Ni. Matrix corrections were based on the PRZ (Armstrong method oxide) procedure (Armstrong, 1988).

Garnet. All garnets (types 1 to 4) are mainly almandine-pyrope solid solutions, with minor grossular and spessartine (Table 2). Grti and G1I4 show nearly constant core to rim X\im (-0.74), XPyr (0.21), Xqb (-0.04) and XSpe (-0.02) values. Grt3 shows XMm, slightly increasing from core (0.74) to rim (0.76) and lower Xp^, (0.19 in the core and 0.18 at the rim). Grt2 shows lower XPrp values increasing from core (0.13) to rim (0.15), at the same values as Grti and G1I4, which reflects its higher (0.06 in the core) and XSpe (0.07-0.08).

Staurolite: Staurolites (Table 2) enclosed by Grti and Grt2are Zn-bearing (ZnO =1.6 wt.%) and rich in Ti (Ti02= 1.6 wt.%). Staurolite inclusions in Grt2 show a slightly higher XMg value than in Grti (0.29 vs. 0.23).

Spinel: Spinel inclusions in Grt2 and Grt3are mainly a hercynite-gahnite-spinel solid solution solid solution with an average XMg value of 0.24 and 0.18, respectively (Table 3). Spinels in both garnet types are poor in Cr2C>3 (-0.4 wt.%) and rich in ZnO (5.3 wt.% in Grt2 and 10.3 wt.% in G1I3).

Biotite: Biotite inclusions in Grt1 and Grt3 are Fe-rich and display similar XMg values of 0.43 and 0.40, respectively. Biotite in both garnets is extremely rich in TiO2 (8.6 wt.% and 9.9 wt.%, respectively, Table 3).

5.3. Laser Raman analysis

For further confirmation, kyanite inclusions in Grt2 and staurolite inclusions in Grt1 and Grt2 were analyzed by Raman spectroscopy using a Thermo DXR Raman microscope installed at the Department of Geology, Naturalis Biodiversity Center, Leiden, Netherlands with 532 nm laser excitation. Raman spectra were collected at room temperature in the confocal mode, for analysis of individual grains on a micron scale (1-2 |im). An Olympus LD lens (x50) was used for all measurements. A grating of 1800 grooves/mm, and a pinhole size of 25 |im was used, which combined with the optical path length, yields a spectral resolution of 1.0 cm-1. Spectra were collected in the range of 100-1800 cm-1.

Spectra for kyanite in Grt1 (represent in Fig. 4a and b) are shown in Fig. 8a and a spectrum for kyanite in Grt2 (Fig. 5c) is shown in Fig. 8b. A spectrum for staurolite in Grt1 (Fig. 4e) is shown in Fig. 8c and for staurolite in Grt2 (represent in Fig. 5e) is shown in Fig. 8d.

6. Mineral reactions

The studied khondalite contains different garnet types of varying sizes, different microtextures and reaction textures which may have resulted independently in the silica-saturated and silica-deficient microdomains, in response to different effective bulk compositions (e.g. Dharmapriya et al., 2015b, 2016c).

In Grt1, quartz and sillimanite enclosed in the mantle area predate kyanite enclosed in the rims, while several isolated staurolite grains occur in the mantle to rim area. This suggests that Grt1 could have formed via the prograde reaction

St + Qtz = Grt + Sil + V (1) The presence of kyanite together with staurolite in garnet rims (Fig. 4a and e) indicates that reaction (1) may have progressed from the sillimanite stability field to the kyanite stability field as

St + Qtz = Grt + Ky + V (2) However, several reactions involving muscovite and biotite may also account for the formation of similar inclusion bearing garnets during the prograde evolution. The presence of

rare biotite inclusions close to staurolite inclusions towards the rim area of Grt1 may suggest the reaction

St + Qtz + Mus = Grt + Sil + Bt + V (3)

followed by

St + Qtz + Mus = Grt + Ky + Bt + V (4) Reactions (3) and (4) have been inferred during the prograde evolution of metapelites from elsewhere by Schreyer and Chinner (1966), Hounslow and Moore (1967) and Carmichael (1969). The lack of muscovite in the khondalite could be due to its complete consumption during the progress of reactions (3) and (4). When staurolite, muscovite or quartz are completely used up during reaction (3) and/or (4), the other two reactants will stay stable together with any or all of the products to some higher temperature (e.g. Carmichael, 1969).

In the core to mantle areas of Grt2, the occurrence of isolated corundum grains and corundum together with kyanite pseudomorphs (Fig. 5a) suggests the possible prograde dehydration reaction

St = Grt + Crn + Ky + V (5) as suggested by Hiroi et al.(1994).

The lack of quartz in the vicinity of corundum grains (Fig. 5 a and b) indicates that Grt2 may have formed within quartz-deficient micro-domains within the kyanite field via reaction (5). The presence of fibrolitic sillimanite pseudomorphing kyanite (Fig.5a and b) in Grt2 suggest the subsequent prograde reaction,

Ky = Sil (6)

The observed early garnet-corundum-kyanite ± staurolite assemblage has been previously reported from the HC (Hiroiet al.,1994; Raase and Schenk, 1994) and from the Palghat-Cauvery Shear Zone, southern India (Shimpo et al., 2006).

In the outer mantle area of Grt3, numerous quartz, minor alkali-feldspar and relict biotite inclusions indicate that the prograde dehydration reaction,

Ti-rich Bt + Sil + Qtz = Grt + Kfs + Ilm + Melt (7) took place in quartz-saturated domains.

The local occurrence of biotite, partially broken down sillimanite and tiny spinels (Fig. 7c) could indicate a second melting reaction,

Ti-rich Bt + Sil = Grt + Spl + Kfs + Ilm + Melt (8) in micro-domains where quartz had been fully consumed via reaction (7) .

The co-existence of medium- to coarse-grained prismatic sillimanite with tiny spinel inclusions in Grt2, the proximity of medium-grained spinel, corundum and garnet (Fig. 5d and

e), the presence of corundum grains enclosed by sillimanite, and the lack of isolated spinel inclusions within garnet, could all indicate the operation of the prograde reaction,

Grt + Crn = Spl + Sil (9)

in quartz-deficient domains.

In Grt4, the presence of sillimanite, quartz and rutile inclusions, the orientation of sillimanite inclusions parallel to matrix sillimanite, and the occurrence of matrix rutile only close to Grt4, are all consistent with the reaction

Sil + Ilm + Qtz = Grt + Rt (10)

The formation of pyrophyllite moats around matrix sillimanite (Fig. 7f) and sillimanite inclusions in the rim area of Grt2 (Fig.5i) and in the outer mantle area of Grt3 (Fig.6d and e), could indicate the retrograde reaction

Sil + Qtz + H2O = Prl (11) Since H2O is essential for this reaction, sillimanite inclusions within some garnet porphyroblasts have also interacted with H2O during the reaction. A possible mechanism for such interaction is presented in the discussion section.

Textural observations indicate that the peak assemblage of most prominent Grt1, Grt2 and Grt3 bearing domains of the rock probably was garnet (Grt1 and Grt2)+sillimanite+K-feldspar+quartz+plagioclase+ilmenite.

7. P-T estimates

7.1. Peak metamorphic conditions

Peak-metamorphic conditions were calculated using pseudosections in the chemical system NCKFMASHTO with the software Perplex_X_07 (Connolly, 2005), using the 2004 update of the Holland and Powell (1998) internally-consistent thermodynamic database and mineral solution models as listed in the caption to Fig. 9. The modeled bulk rock composition which represents most dominant silica-saturated domains having Grt1, Grt3 and Grt4 was taken from XRF analysis (expressed in mol% in Table 1). H2O content corresponds to the loss on ignition (LOI). To estimate the Fe2O3 bulk content, a T-X(Fe2O3) pseudosection was calculated at a pressure of 9 kbar (Fig. 9a) which is the most approximate pressure condition during peak metamorphism in the HC according to published data (e.g. Dharmapriya et al., 2015a, b, 2016).

In the T-X(Fe2O3) diagram (Fig. 9a), the interpreted peak matrix assemblage melt -plagioclase - alkali-feldspar - garnet - ilmenite - sillimanite - quartz is predicted to be stable at X(Fe2O3) < 0.33. At higher X(Fe2O3) values, first rutile and then hematite would be stable instead of ilmenite. At > 860 °C, the observed assemblage is stable for X(Fe2O3) down to 0, while at T between 860 and 775 °C, it would require the progressive increase of ferric iron component to be stable. Microtextural observations revealed that after peak T (during the retrogression) rutile was produced by reaction (10) at the expense of ilmenite and no biotite formed. This suggests that X(Fe2O3) was < 0.08.

The P-T pseudosection of Fig 9b displays the phase relations for the investigated khondalite calculated at low X(Fe2O3) = 0.01. The observed peak assemblage is stable for a wide P-T range between 830-1000 °C and 6-10.5 kbar. At P< 6 kbar, cordierite is predicted to appear. At P> 9-10.5 kbar, depending on T, rutile would replace ilmenite (the presence of minor rutile as inclusion phases in Grt2 (Fig. 5c, d, j) and Grt3 (Fig. 6b), and their absence in the peak metamorphic assemblage conform above argument). The appearance of biotite at T < 830 °C defines the lower T limit. Although pseudosection modelling provides the most powerful thermobarometric tools compare to conventional thermobarometric calculations (e.g. Powell and Holland, 2008; Palin et al., 2016), interpreted P-T conditions using pseudosection modelling also contains uncertainties relating to end-member data within the thermodynamic datasets (e.g. Powell and Holland, 2008) and determination of effective bulk composition, including via XRF (Palin et al., 2016).

The operation of the reaction Grt + Crn = Spl + Sil (reaction 9), inferred from inclusions formed in specific micro-domains under silica under-saturated conditions, once these were trapped by garnet and isolated from the silica-rich rock matrix, allows to better constrain the P-T at the peak of the metamorphism. In the simple FASH chemical system, reaction (9) occurs at 7.0, 8.5, 10.0 kbar at 800, 900, 1000 °C, respectively (e.g., Hiroi et al., 1994). However, the location of the same reaction (9) in ZnFAS system (Shulters and Bohlen, 1988) is shifted to lower T and higher P and obtains a higher dP/dT. Since the spinals in Grt2 contains considerable amount of Zn (up to 5.3 wt.%), the ZnFAS system is most suitable to calculate the T conditions. Reaction (9), calculated using Grt2 mantle composition (next to the corundum and spinel inclusion bearing sillimanite in Grt2 in Fig. 5e, f and g, and the average composition of spinel inclusions in sillimanite in Grt2, yields P of 8.3, 9.2, and 10.1 kbar at T of 840, 880 and 920 °C , respectively for Zn Spl-Grt-Crn-Sil thermometer after Shulters and Bohlen (1988). This reaction line intersects the rutile-out boundary at T of

ca. 890 °C and P of ca. 9.5 kbar (Fig. 9b), defining a minimum peak T around 900 °C and a maximum P around 9.5 kbar.

Textural observations indicate that after the peak of the metamorphism rutile and Grt4 formed at the expense of ilmenite, sillimanite and quartz through reaction (10). From the pseudosection topology, it is very likely that reaction (10) occurred during isobaric cooling after peak T. This implies that the minimum P at peak T was ~9 kbar. However, the Grt3 contain considerable amount of MgO (5.00 wt.%) which could act as a component to evaluate the P-T stability of the reaction (9) and (10). Also number of authors argued that the uncertainty in conventional thermobarometry are analytical imprecision and the systematic error associated with experimental calibration techniques (Hodges and McKenna, 1987: Powell and Holland, 2008). Palin et al. (2016) argued that when a single pair of equilibria is used for comparative purposes, the systematic experimental-related errors cancel, leaving only random uncertainty associated with analytical imprecision as a contributing factor. Hence, these remaining errors are cited to be at least ±50 °C and ± 1 kbar (2o; Palin et al., 2016). Therefore, the uncertainty of the interpreted peak metamorphic conditions of the khondalite can be affected by up to ±50 °C and ± 1 kbar (2a) in P-T space.

7.2. Prograde metamorphic conditions

The reconstruction of the prograde metamorphic conditions of granulites is particularly challenging because of common overprinting of early phases by peak/post-peak mineral assemblages, resetting of the original mineral compositions during cooling and/or changes in the bulk composition due to the addition or loss of melts and fluids (e.g., Harley 2008; Kelsey, 2008). Nevertheless, inclusions present in garnet porphyroblasts may provide information on at least a portion of the prograde history. The specific inclusion features observed in four types of garnet porphyroblasts in this study allow the identification of different mineral assemblages stable at different metamorphic conditions, the establishment of a series of reactions which took place during the early stages of the prograde history, as well as the reconstruction, at least qualitatively, of the petrographical evolution of the rock.

However, to quantitatively constrain the prograde path in P-T space is still challenging. A conventional geothermobarometric strategy would likely fail because of the inability to assess the equilibrium compositions of the minerals involved in the observed reactions with reasonable certainty. Alternatively, a pseudosection approach depends on the ability to estimate the original bulk rock composition at sub-solidus conditions. In the case of

the investigated khondalite, estimated peak conditions of T ~ 900°C and P around 9-9.5 kbar suggest that the rock may have changed its original bulk composition due to the loss of melt during its prograde evolution. This is consistent with the residual bulk composition measured by XRF. Compared to Post-Archaean Average Shale (e.g., Taylor and McLennan, 1985), the highly aluminous khondalite composition suggests melt loss of 30% or more. To have a plausible approximation of the protolith composition at sub-solidus conditions, the approach of White et al. (2004) was followed for stepwise reintegration of melt until reaching a composition that allows the wet solidus to be calculated (Table 4). However, one has to be aware that this approach offers a good approximation for most of the major oxides but cannot predict variations in Ti and ferric iron because these components are not included in the melt model for pelitic composition available so far (White et al., 2007).

The reintegration of melt back to the measured XRF bulk composition (CXRF) was performed in four different steps. Melt composition and volume at each stage have been calculated at given P-T conditions using the software "werami" of the package Perplex_X_07. Amounts (in vol.%) and compositions of melt added at each step (M1 - M4), P-T conditions at which the calculation was performed and the new calculated bulk rock compositions after melt reintegration (CM1 - CM4) are given in Table 4. After this procedure, which implied the reintegration of about 30 vol.% melt, we obtained an estimate of the original, pre-melting, bulk rock composition (CM4 in Table 4). At this point, we assumed H2O to be in excess and we calculated a sub-solidus P-T pseudosection using the calculated proportions of rock forming oxides in the system NCKFMASHTO (Fig. 10). To explore the metamorphic evolution above the solidus, we calculated a new P-T pseudosection using the amount of H2O predicted to be present immediately after crossing the wet solidus (Fig. 12). The large temperature jump of the rock solidus at P > 7.5 kbar between Figs. 10 and 12 reflects the inferred melt escape, which effectively lowers the amount of volatile components and thus raises the T needed for further melting. Both pseudosections have been calculated at X(Fe2O3) = 0.01. In order to visualize the phase relations within both pseudosections, we also calculated mineral modes for each phase involved in the rock evolution (Figs. 11 and 13).

Figs. 10 and 11 display the phase relations in P-T space at conditions below the wet solidus. The occurrence of staurolite, quartz and sillimanite inclusions in the Grt1 mantle suggests that the earliest recognizable stage of the metamorphic evolution occurred at T around 575 °C and P of 4.5 kbar which implies that the rock has evolved through relatively high-temperature and low-pressure amphibolites facies conditions. At these conditions, the modeling predicts the formation of staurolite and plagioclase and consumption of biotite,

muscovite and quartz. The subsequent formation of sillimanite and biotite, as inferred by reaction (3), is best explained by a P-T path with slope of ~0.03 kbar/ °C (Fig. 11). The occurrence of kyanite inclusions in Grt1 rims indicates that the P-T path entered the kyanite field, probably at ~ 650 °C and 6.5 kbar, before crossing the wet solidus at 675 °C and 7.5 kbar (the star in Figs. 10 and 12).

As illustrated in Figs. 12 and 13, the metamorphic evolution above the wet solidus started in the kyanite field. According to the model, muscovite dehydration melting, leading to the complete consumption of white mica and decrease in quartz and plagioclase and the contemporaneous formation of alkali feldspar, kyanite and melt, occurred at > 725 °C and 9 kbar. The presence of kyanite and rutile inclusions within garnet and matrix sillimanite suggests that the P-T path reached peak pressure at > 10 kbar and around 800 °C, before starting decompressing during further heating.

Although there is no applicable geobarometer to determine the peak prograde pressure in the studied khondalite, based on conventional thermobarometric calculations on UHT granulites collected 6 km southwest and 1.5 km northeast of the present sampling locality, Dharmapriya et al. (2015a,b) reported similar peak pressures of 11-12 kbar. Hence, the studied khondalite could also have reached at least 11 kbar during the prograde evolution. After peak P, the rock underwent a prograde decompression stage and re-entered the sillimanite field, probably around 825 °C and 10.5 kbar. The prograde evolution of the khondalite culminated at ca. 900 °C at ~9-9.5 kbar.

7.3. Post-peak metamorphic conditions

As explained above, the textural observations suggest that Grt4 and rutileformed after the peak of the metamorphism via reaction (10). In the P-T pseudosection of Fig. 9b, the appearance of rutile replacing ilmenite requires that after peak T the rock followed an isobaric cooling path down to < 875 °C at ~9 kbar. The experimental calibration of reaction (10) in the FASHT chemical system by Bohlen et al. (1983), calculated using Grt4 core composition, yielded P = 8.8, 9.1, 9.5 at nominal T = 840, 880 and 920 °C (temperature uncertainties of ± 50 °C result in maximum errors in inferred pressure of about 0.5 kbar), respectively (Fig. 14), in good agreement with the pseudosection results.

The last recognizable event in the investigated khondalite is the formation of pyrophyllite moats around sillimanite by reaction (11). In the system ASH, reaction (11) takes place at T< 400 °C, implying that the final low-grade history experienced by the sample

was characterized by infiltration of H20-rich fluids at low temperatures. Fig. 14 summarizes the entire P-Ttrajectory followed by the khondalite.

8. Zircon U-Pb geochronology

After crushing of -1.5 kg of sample, zircon grains were separated by gravimetric and magnetic methods. Further purification was done by hand picking under a binocular microscope at the Centre for Earth Sciences, Indian Institute of Science, Bangalore, India. Grains were mounted in epoxy resin discs and polished to reveal mid-sections. After gold sputter coating, images were taken under transmitted and reflected light. Thereafter, the samples were imaged using cathodoluminescence (CL) to determine zircon morphology and internal structures and to choose potential target sites for U-Pb analyses at the Key Laboratory of Continental Dynamics at the Department of Geology, Northwest University, Xi'an, China. U-Pb isotopes in zircons were measured using a Geolas-193 UV laser ablation system coupled with an Agilent 7500a ICP-MS at the same lab. Helium was utilized as the carrier gas with a laser beam of 32 [j,m in diameter and a frequency of 6 Hz. Data acquisition time was 40s. Zircon 91500 was used as the standard for U-Pb isotopic ratio determination and age calibration. The standard NIST610 was employed as an external standard and Si as the internal standard during trace element analysis. Common Pb was corrected using measured 204Pb. The degree of discordence of data was <10%. The concentrations are calculated using the GLITTER software (Macquarie University).

The zircon U-Pb analytical results are given in Table 5 (spot ages were calculated for 1 o error). Representative CL images of the zircons from the studied khondalite are shown in Fig. 15a, while age data are plotted in Fig. 15b and c (weighted means were calculated for 2a error).

Zircon crystals are colourless and transparent to translucent. Most of the grains are anhedral to subhedral and show near-spherical to stubby or irregular morphology, interpreted as metamorphic grains. A number of zircons show clear core-rim texture with dominantly dark cores and bright rims (Fig. 15a). Some grains are structureless and display a homogeneous gray colour. Zircon lengths are 25-130 [j,m while length to width ratios varying from 2:1 to 1:1. A total of 35 spots from 35 zircons were analyzed from this rock.

All of the age data were found to plot along the Concordia (Fig. 15b). Twenty seven

206 238

spots with " Pb/" U age range from 538 ± 5 Ma to 548 ± 5 Ma form a coherent tight group that yields a weighted mean 206Pb/238U age of 542 ± 2 Ma (MSWD = 0.24; Fig. 15b and c) with a narrow range of Th/U (0.01-0.03). Seven spots analyzed on mainly rim domains

formed another tight group with 206Pb/238U ages between 507 ± 4 Ma and 517 ± 4 Ma, yielding a weighted mean 206Pb/238U age of 514 ± 3 Ma (MSWD = 0.50, Th/U = 0.01-0.05;

206 238

Fig. 15b and c). One spot analyzed within a core domain shows a

206Pb/238U

age of 569 ± 4

Ma with a Th/U ratio of 0.01.

9. DISCUSSION

9.1. Different garnet generations in the khondalite

According to the enclosed phase assemblages and the metamorphic evolution deduced from them, the garnet can be subdivided into those that formed in silica-saturated domains (Grt1, Grt3 and Grt4) vs. those that formed in locally silica-deficient domains (Grt2). In addition, Grt1 and Grt2 are probably pre-tectonic with respect to the main foliation (S2), whereas Grt3 is syn-tectonic and Grt4 is post-tectonic. The latter is clear from the breakdown of matrix silimanite parallel to S2 and L2 forming Grt4.

Needle shaped sillimanite and ribbon quartz in the matrix coexist with Grt1 (Fig. 4a) and define the major rock lineation that lies within the main gneissic layering. By contrast, oriented quartz and minor ilmenite inclusions define an internal foliation (Si) in Grt1, which is discontinuous and oblique to the matrix foliation (S2). Since staurolite inclusions in the mantle area of Grt1 (see right hand side of Grt1 in Fig. 4a) are oriented parallel to quartz and ilmenite, Si probably represents an early deformation phase that operated in the staurolite stability field during prograde metamorphism. The elongation of Grt1 parallel to the external foliation could indicate Grt1 growth during early D2. The orientation and grain size of sillimanite inclusions in Grt1 differ from matrix sillimanite, suggesting an earlier origin. In view of the inclusion phases in Grt1 (Fig. 4) and the pseudosection calculated after melt reintegration up to the wet solidus (Figs. 10 and 11), we conclude that that the rock evolved from the sillimanite to the kyanite field during Grt1 growth.

The XMg of staurolite in the Grt2 core (close to corundum and kyanite) is higher than XMg in the mantle to rim of Grt1 (0.29 vs. 0.23). This may indicate that the local silica-deficient domains had a higher bulk XMg than the more common silica-saturated domains. A number of studies found that an increasing XMg of staurolite enhances its stability especially at elevated pressures (e.g., Schreyer, 1988; Peacock and Goodge, 1995; Fockenberg, 1998; Shimpo et al., 2006) in silica-deficient rock domains. This supports the idea that the core area of Grt2 formed at elevated P-T conditions relative to the formation of mantle and rim of Grt1.

The absence of growth zoning during the evolution (from sillimanite to kyanite stability field with increasing P) of Grt1 can probably be explained by the chemical homogenization of prograde garnet during HT/UHT metamorphism due to the fact that growth zoning is frequently obliterated in high-grade garnets due to volume diffusion (e.g., Blackburn, 1969; Grant and Weiblen, 1971; Yardley, 1977, Fernando et al., 2003; Caddick et al., 2010).

Further, the variability of inclusion assemblages in Grt2 (kyanite, corundum and quartz bearing) indicates that Grt2 may have grown across silica-saturated and silica-deficient microdomains having different effective bulk compositions during prograde metamorphism. Dharmapriya et al. (2016c) argued that such contrasting compositional microdomains of the rock could represent different layers in the precursor sediments or differentiated crenulation cleavages. Due to strong ductile deformation associated with HT/UHT metamorphism, metamorphic differentiation and crystallization of locally produced melt may have obliterated the evidence for such microdomains in the matrix.

9.2. P-T path of the khondalite in relation to the HC evolution

The metamorphic history of the studied khondalite inferred from detailed microtextural observations, pseudosections and relevant experimentally calibrated reactions, allows us to determine a possible PT path for the HC and put the studied rock in the context of the Srilankan tectono-metamorphic evolution.

9.2.1. Prograde evolution

Prograde compression: the earliest recognizable stage of the prograde path started in the sillimanite stability field (in Sil-Ky-And phase diagram) and continued in the kyanite field due to a pressure increase. Reactions 1, 2, 3 and 4 coupled with mineral stability fields of the P-T pseudosection calculated with the melt-reintegration technique until to have a composition that allows the wet solidus (Fig.10) and the mineral mode diagram of Fig.11 provide evidence for prograde burial of the khondalite. This interpretation is consistent with early sillimanite needles pre-dating kyanite inclusions in garnet porphyroblasts reported from pelitic granulites in the eastern HC (Raase and Schenk, 1994; Fig. 1).

The transition from the sillimanite to the kyanite stability field and the occurrence of kyanite + staurolite in garnet indicates that during the early metamorphic evolution the pressure considerably increased up to peak P with only minor heating (average dP/dT ~30

bar/°C). This could suggest a setting of crustal subduction-related continent-continent collision tectonics, similar to present-day Pamir (e.g., Searle et al, 2001) Peak pressure conditions: the occurrence of rutile in garnet limits the minimal peak P at around 10 kbar (Fig. 13). A more precise determination of the peak pressure experienced by the khondalite is hampered by the lack of P-sensitive reactions or assemblages. The P-T pseudosection calculated of melt re-integration modeling above wet-solidus (Fig. 12) and calculated mineral modes (Fig. 13) indicated that the maximum pressure occurred at P > 10 kbar and T around 780 °C. Comparisons with other studies carried out in the proximity of the sampling area (Dharmapriya et al., 2015a, b) suggest that peak P was probably around 10.5-12 kbar. A similar peak P of ~11-12 kbar at T of ~820-850 °C has been inferred by Dharmapryia et al. (2015b) using conventional thermobarometric methods applied to sapphirine granulites collected about 1.5 km eastward from the present sampling locality.

Prograde decompression: The re-appearance of sillimanite replacing kyanite (reaction 6) shows that after peak P the rock entered again into the sillimanite field. Both dehydration melting reactions (7) and (8) have taken place in the sillimanite field during prograde metamorphism. Textural observations coupled with metamorphic evolution above the wet-solidus (Figs. 12 and 13) also indicate that the rock has undergone a period of prograde decompression with increasing temperature up to peak metamorphism. Mineral textures related to the net transfer reaction (9) in Grt2 (representing a silica-deficient domain of the khondalite) indicate that the rock evolved from the garnet-corundum stability field to the spinel-sillimanite field at this stage.

Petrological evidence for prograde decompression in the HC has been also reported in earlier studies (e.g., Hiroi et al., 1994; Karunarathna et al., 2002; Osanai et al., 2006; Wickramasinghe and Perera, 2014; Dharmapriya et al., 2015a,b). The inferred prograde decompression could be due to lateral extrusion of the crust, which followed the crustal overthickening due to collisional tectonics (e.g., Tapponnier et al., 1981; Ellis, 1987). Tapponnier et al. (1981) proposed that duringcontinent-continent collision, an overthickened crust can spread laterally while the collisional event is still going on. Ellis (1987) argued that due to the temperature dependent rheology of the lower crust, extension could be expected after convergence if the temperature exceeds some critical value. In addition, extension could result in increasing temperature during slight decompression and could be followed by isobaric cooling (Harley, 1991). This process could be accelerated by the presence of melt. The incorporation of curved quartz inclusions of core and inner mantle areas in Grt3 via reaction (7) and (8) could provide evidence for increasing temperature and melt production

towards peak metamorphism under non-coaxial strain. Such melt-assisted non-coaxial strain at lower crustal levels is consistent with lower crustal extension (e.g., Hopper, 1996). Also Dharmapriya et al. (2016c) reported containing of rotated crenulation lineation demarcated by biotite and tiny sillimanite needles from prograde garnets (has formed at near peak conditions via melting reactions) of UHT granulites in the HC. These crenulation lineation is oblique to the major matrix lineation of the UHT granulites indicating the rotation of garnet during their growth (Dharmapriya et al., 2016c). Many earlier studies on the HC granulites suggested that the formation of the prominent foliation and lineation occurred under non-coaxial flow simultaneously with peak T during D2 (e.g., Kriegsman, 1991, 1993; Kehelpannala, 1993, 1997). This D2 event is demarcated by local extreme stretching factors up to 20 (Kriegsman, 1993; Kleinschrodt, 1994), by quartz ribbons, and by elongate aggregates of recrystallized feldspar which commonly form the major mineral lineation throughout the HC granulites.

9.2.2. Peak metamorphism

Conventional thermobarometric calculations and pseudosections reveal that the khondalite reached a peak T around 900 °C at P 9 - 9.5 kbar (with possible ±50 °C and ± 1 kba uncertainties associated with P-T calculation) which approximately represent the HT/UHT boundary (e.g., Harley, 1998).

9.2.3. Post-peak evolution

Although evidence to reconstruct the retrograde segment of the P-T path for the HC is limited (e.g., Hiroi et al, 1994; Raase and Schenk, 1994; Kriegsman and Schumacher, 1999), many studies on the retrograde evolution of metaigneous (e.g., Schumacher et al., 1990; Schenk et al., 1991; Prame, 1991) and metasedimentary (e.g., Perera, 1984, 1987; Ogo et al., 1992; Hiroi et al., 1994; Dharmapriya et al., 2014, 2015a,b) granulites show that, after peak T, the HC experienced a period of near-isobaric cooling followed by isothermal decompression.

Isobaric cooling (IBC): in the studied khondalites, the formation of fine Grt4 and rutile via reaction (10) after peak T indicates that the post-peak evolution started with a period of near-IBC. A similar feature was first reported from pelitic granulites in the HC by Perera (1987), who described the occurrence of tiny, post-peak garnets that formed during IBC and are compositionally different from early, coarse-grained garnet porphyroblasts.

Even though a phase of isobaric cooling after peak T in the HC is poorly constrained in metapelitic rocks (e.g.; Perera, 1987; Raase and Schenk, 1994; Dharmapriya et al., 2014,

2015a) and calc-silicates (Mathavan and Fernando, 2001), it has been largely documented from reaction textures in intermediate and mafic granulites (Schumacher et al., 1990; Faulhaber and Raith, 1991; Prame, 1991; Schenk et al., 1991).

However, close to the current sampling locality (6 km south-west), IBC after peak T was also inferred from corundum-bearing khondalites (Dharmapriya et al., 2015a). In these pelitic granulites, the peak assemblage formed in the sillimanite - spinel stability field, similar to the investigated khondalites, and garnet + corundum crystallised after peak T during IBC at UHT conditions.

We interpret the formation of oriented ilmenite needles within Grt2 porphyroblasts in the studied khondalite (Fig. 5k) as features resulting from isobaric cooling. Many experimental and natural studies have shown that TiO2 candissolve into silicates like garnet at high temperatures. During isobaric cooling and/or decompression, TiO2 may then start to precipitate forming needle-like ilmenite and/or rutile within garnet (e.g., Song et al., 2004; Wark and Watson, 2006; Kawasaki and Osanai, 2007; Ague and Eckert, 2012; Dharmapryia et al., 2015). Since the oriented ilmenite needles display two well defined directions of crystallization in corundum-bearing porphyroblastic garnet, they do not define an early foliation but represent mimetic growth during exsolution from host garnets during IBC. Similar exsolutions have been reported by Dharmapryia et al. (2015a) from garnet porphyroblasts contained in high-grade metapelites from the HC.

Isothermal decompression (ITD): In the studied khondalites there is no direct textural evidence for isothermal decompression. However, several studies on the HC granulites inferred a stage of ITD after IBC (e.g., Perera, 1987, 1994, Prame, 1991; Raase and Schenk, 1994; Mathavan and Fernando, 2001; Dharmapriya et al., 2014, 2015b) and evidence for decompression is very strong in interlayered metabasites (e.g., Schumacher and Faulhaber, 1994; Sajeev et al., 2007) and sapphirine granulites (Kriegsman and Schumacher, 1999; Sajeev and Osanai, 2004b; Dharmapriya et al., 2015b). White et al. (2007) suggested that mineral assemblages in dry rocks would hardly react in response to P-T changes. Therefore, in the absence of fluids/melts infiltrating through the rock after the IBC event, no ITD reaction textures have formed in most of the dry metapelitic granulites.

9.2.4. Late stage of the retrograde evolution

Late pyrophyllite moats around both matrix sillimanite and sillimanite inclusions in Grt2 and Grt3 represent the last detectable stage in the khondalite and suggest that sillimanite interacted with H2O-rich fluids at T < 400 °C. Since H2O-rich fluids are commonly available during retrogression of regional metamorphic rocks at middle to upper crustal levels (e.g., Bucher and Grapes, 2011; Huizenga et al., 2014), it is likely that H2O-rich fluids infiltrated the khondalite and reacted with matrix sillimanite and sillimanite + quartz inclusions in garnet along cracks during a very late stage of cooling after decompression (e.g. Hiroi et al., 2014). Reaction (11) operates in the andalusite or kyanite stability field, but involved breakdown of metastable sillimanite.

Several authors (e.g., Ogo et al., 1992; Hiroi et al., 1994; Raase and Schenk, 1994) suggested that the latest part of the retrograde path in the HC occurred in the andalusite stability field. For instance, Hiroi et al. (1994) reported the formation of andalusite, siderite, and quartz at the expense of cordierite or garnet in the presence of CO2. Raase and Schenk, (1994) described the formation of andalusite, chlorite and quartz at the expense of cordierite in the presence of H2O in HC metapelites. Generally, most of the andalusite forming reactions were observed in the southwestern part of the HC, where low P conditions prevailed. However, Hiroi et al. (1994) reported andalusite in the central HC towards the northeastern part, close to the present sampling locality. Hence, reaction (11) could also have taken place in the and alusite field at a late stage of the P-T trajectory.

9.3. Geochronological evolution: implications for P-T-t evolution of the HC metasediments

The Th/U values of all zircons analyzed are < 0.06 (Table 5), typical of metamorphic

206 238

zircons (e.g., Rubatto, 2002). The Pb/ U dates form a tight group with a weighted mean 206Pb/238U age of 542 ± 2 Ma (MSWD = 0.24, Th/U=0.01-0.03; Fig. 13). Recently Dharmapriya et al. (2015b, 2016) and Takamura et al. (2016) suggested that the peak UHT metamorphism in the HC in the age range from ca. 580 - 530 Ma. Hence, the obtained weighted mean average 542 probably represent the time of peak metamorphism of the studied khondalite. This peak zircon population could be a result of melt loss and Zr oversaturation of the rock.

The second206Pb/238U age group ranges from 507 ± 4 Ma to 517 ± 4 Ma (Fig. 15b and c) and forms a tight cluster with a weighted mean 206Pb/238U age of 514 ± 3 Ma (MSWD = 0.50). These zircons are also of typical metamorphic origin considering their low Th/U ratio,

however, it is likely them to represent a cooling age of the khondalite, possibly leading to final melt crystallization near the wet solidus during retrograde metamorphism.

The U-Pb ages reported here do not constrain the timing of stages along the prograde path. Recently, Dharmapriya et al. (2015b, 2016) analysed detrital zircons from Neoarchean to Neoproterozoic (ca. 2800 Ma to 720 Ma) from UHT metasedimentary rocks of the HC. Hence, those authors argued that the metasediments of the HC derived from areas with multiple provenance ages and were deposited during the Neoproterozoic. This is consistent with Sajeev et al. (2010), who reported U-Pb age clusters of c. 1700 Ma and 1040-830 Ma from a sapphirine granulite outcropping southwest of Gampola (2.5 km north east of the present sampling locality) Takamura et al. (2016) who report detrital zircons from 3500 to < 700 Ma.

Dharmapriya et al. (2016) observed oldest metamorphic populations with weighted

206 238

206Pb/238U

age of 665 ± 5 Ma close to the center of the HC (close to Nildandahinna),

206 238

and populations with a weighted mean Pb/ U age of 633 ± 6 Ma from the western margin of the HC (close to Gampola, 2 km North east of the present sampling locality). Recently, Osanai et al. (2016b) also reported metamorphic zircon ages from ~640 - 535 Ma from UHT granulites in the HC. Those authors argued that UHT granulites in the HC were formed by multiple thermal episodes during Ediacaran to Cambrian period approximately at 620-580 and 565-525 Ma. But there is no any clear textural evidence so far being reported relevant to two separate HT metamorphic events from the HC. Instead, textural evidence indicates that the HC rocks have evolved through clockwise P-T trajectory from early medium grade amphibolite facies (during prograde evolution) to HT granulite facies (at peak metamorphic conditions).

We would like to interpret these older metamorphic zircon populations could be coeval with the early, medium P and medium to high T amphibolite facies metamorphism of the HC reported in this study. These less abundance older zircons may have preserved (during HT/UHT peak metamorphism) due to incorporation of them into minerals such as garnet prophyroblases (coeval with their growth) during prograde metamorphism. The peak HT/UHT metamorphism at around 580 - 530 Ma while IBC could have taken place ca. 500 Ma.

In combination, the evidence suggests that the entire P-T trajectory, from prograde evolution to near IBC, discussed in this study, could occur in the time span from the Late Cryogenian to the Early Cambrian (ca. 650 - 500 Ma), is coeval with the assembly of Gondwana Supercontinent.

10 Conclusion

The earliest recognizable stage of the studied khondalite occurred in the sillimanite field at around 575 °C at 4.5 kbar. Then the rock has entered in to kyanite stability field around ~660 °C at 6.5 kbar with an average dP/dTof ~30 bar/ °C, before crossing the wet-solidus at around 675 °C at 7.5 kbar. After that, the khondalite has reached to the highest pressure occurred at P> 10 kbar and T around 780 °C. The peak pressure condition was at high pressure granulite facies condition (> 10 kbar) with a garnet-corundum-kyanite ± staurolite assemblage. At 825 °C and 10.5 kbar, the rock re-entered into the sillimanite field during the prograde decompression associated with increasing temperature. The temperature peaked at 900 °C at ca. 9-9.5 kbar. Subsequent near-isobaric cooling led to the growth of Grt4 and rutile at T ~880 °C. Subsequent near isothermal decompression, well-known from the Highland Complex, is not recorded in this sample probably due to absence of fluids/melts infiltrating through the rock after the isobaric cooling event. Local pyrophyllite rims around sillimanite suggest a late stage of rehydration at T< 400 °C, which probably occurred after uplift to upper crustal levels.

We conclude that the Highland Complex metasediments have evolved through relatively high-temperature, low-pressure amphibolite facies conditions in the sillimanite stability field (in Sil-Ky-And phase diagram) during its prograde path and subsequently entered into the kyanite stability field during overthickening of the crust during subduction related continent - continent collision associated with assembly of Gondwana. This prograde compression continuous until occurrences of the peak pressure condition of the Highland Complex rocks at high-pressure granulite facies (at least 11 kbar). Thereafter, the rock re-entered into the sillimanite stability field where peak metamorphism was accompanied by ductile deformation at HT/UHT granulite facies conditions. A subsequent stage of near isobaric cooling occurred still at lower crustal levels. Finally the rock suite was rapidly uplifted to upper crustal levels along an isothermal decompression path accompanied by folding and thrusting.

The P-T trajectory from prograde to near isobaric cooling after peak metamorphism seems to occur within a time span from Late Cryogenian to Early Cambrian (ca. 665 - 500 Ma), coeval with the assembly of the Gondwana Supercontinent.

Acknowledgements

We are grateful to the National Research Council (NRC) of Sri Lanka (grant No 15089) and the Ministry of Technology and Research (MTR/TRD/AGR/3/1/04) and Department of Science and Technology (DST), India, for funding this project, and the Department of Geology, University of Peradeniya for providing necessary facilities. The first author acknowledges a one month Martin Fellowship to work at Naturalis Biodiversity Center, Leiden, the Netherlands and specially thanks Hans de Groot for EPMA and SEM analysis and EDS mapping and Hanco Zwaan for Laser-Raman analysis. C.B. Dissanayake and N.D. Subhasinghe are kindly acknowledged for providing additional laboratory facilities at the National Institute of Fundamental Studies, Kandy, and to O. K. S. Opatha for various assistance and Thilini Harischandra for support in preparing rock thin sections. L.M.K. acknowledges support by the Stichting Dr Schurmannfonds (Grants Nos. 88/2012, 94/2013 and 101/2014).

References

Ague, J.J., Eckert J.O., 2012. Precipitation of rutile and ilmenite needles in garnet: Implications for extreme metamorphic conditions in the Acadian Orogen, U.S.A. American Mineralogist 97, 840-855.

Almond, D.C., 1991.Arena gneiss and Kandy gneiss - a proposed subdivision of the Highland Series around Kandy and its significance. Journal of Geological Society of Sri Lanka 3, 41-50.

Amarasinghe, U.B., Collins, S.A., 2011. Highland Complex and Wanni Complex of Sri Lanka formed in one Neoproterozoic orogenic mobile belt during assembly og Gondwana: La-ICPMS U-Pb zircon geochronology, (Abstract) 8th International symposium on Gondwana to Asia, IAGR Conference Series 12, p 33. Armstrong, J.T., 1988. Quantitative analysis of silicates and oxide minerals: Comparison of Monte-Carlo, ZAF and Phi-Rho-Z procedures, Microbeam Analysis 239-246.

Baur, N., Kroner, A., Todt,W., Liew, T.C., Hofman, A.W., 1991. U-Pb isotopic systematics of zircons from prograde and retrograde transition zones in high grade orthogneisses, Sri Lanka. Journal of Geology 99, 527-545.

Berger, A. R., Jayasinghe, N.R., 1976. Precambrian structure and chronology in the Highland Series of Sri Lanka. Precambrian Research 3, 559-576.

Blackburn.W.H., 1969.Zoned and unzoned garnets from the Grenville Gneisses around Cananoque, Ontario. Canadian Mineralogists 9, 691-698.

Bohlen, S.R., Wall, V.J., Boettcher, A.L., 1983. Experimental investigations and geological applications of equilibria in the system FeO-TiO2-A12O3-SiO2-H2O. American Mineralogist 68, 1049-1058.

Bolder-Schrijver, L.J.A., Kriegsman, L.M., Touret, J.L.R., 2000. Primary carbonate/CO2 inclusions in sapphirine-bearing granulites from central Sri Lanka. Journal of Metamorphic Geology 18, 259-269.

Braun, I., Kriegsman, L.M., 2003. Proterozoic crustal evolution of southernmost India and Sri Lanka.In: Yoshida, M., Windley, B.F. Dasgupta, S. (Eds.) Proterozoic East Gondwana: Supercontinents Assembly and Breakup., Geological Society, London, Special Publications 206, 169-202.

Carmichael, D.M., 1969. On the mechanism of prograde metamorphic reactions in quartz-bearing pelitic rocks. Contributions to Mineralogy and Petrology 20,244-267.

Cooray, P.G., 1962. Charnockites and their associated gneisses in the Precambrian of Ceylon.Journal of Geological Society of London 118, 239- 273.

Cooray, P.G., 1984. An Introduction to the Geology of Sri Lanka (Ceylon), 2nd revised edition. pp 340.

Cooray, P.G., 1994. The Precambrian of Sri Lanka: a historic review. In: Raith, M., Hoernes, S. (Eds.). Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka, Precambrian Research, 66, 3-18.

Dharmapriya, P.L., Malaviarachchi, S.P.K., Galli, A., Ben-Xun Su., Subasinghe, N.D., Dissanayake, C.B., Nimalsiri, T.B., Zhu, B., 2014. P-T evolution of a spinel + quartz bearing khondalite from the Highland Complex, Sri Lanka: Implications for non-UHT metamorphism. Journal of Asian Earth Sciences 95, 99-113.

Dharmapriya, P.L, Malaviarachchi, S.P.K., Galli, A., Ben-Xun Su, Subasinghe, N.D., Dissanayake, C.B., 2015a. Rare evidence for formation of garnet + corundum during isobaric cooling of UHT meta-pelites: New insights for retrograde P-T trajectory of the Highland Complex, Sri Lanka. Lithos 220, 300-317.

Dharmapriya, P.L, Malaviarachchi, S.P.K., Santosh, M., Tang, L., Sajeev, K., 2015b.Late-Neoproterozoic ultra high temperature metamorphism in the Highland Complex, Sri Lanka. Precambrian Research 271, 311-333.

Dharmapriya, P.L., Malaviarachchi, S.P.K., Sajeev, K., Chengli Zhang., 2016a. New LA-ICPMS U-Pb geochronology of detrital zircons from the Highland Complex: insights into Sri Lanka-India Gondwana linkage. International Geology Review 58, 1856-1883.

Dharmapriya, P.L., Malaviarachchi, S.P.K., Kriegsman, L.M., Galli, A., Sajeev, K., Osanai, Y., Subasinghe, N. D., Dissanayake, C. B., under review. Different mineral reactions in silica-saturated and silica-deficient microdomains within single garnet in UHT granulites: an example from Sri Lanka. Geoscience Frontiers.

Droop, G.T.R., 1989. Reaction history of garnet-sapphirine granulites and conditions of Archaean high-pressure granulite facies metamorphism in the Central Limpopo Mobile Belt, Zimbabwe. Journal of Metamorphic Geology 7, 383-403.

Ellis, D.J., 1987. Origin and evolution of granulites in normal and thickened crust. Geology 15, 167-170.

Faulhaber, S., Raith, M., 1991. Geothermometry and geobarometry of high-grade rocks: a case study on garnet- pyroxene granulites in southern Sri Lanka. Mineralogical Magazine 55, 17- 40.

Fernando, G.W.A.R., Hauzenberger, C.A. Baumgartner, L.P., Hofmeister, W., 2003. Modeling of retrograde diffusion zoning in garnet: evidence for slow cooling of granulites from the Highland Complex of Sri Lanka. Mineralogy and Petrology 78, 5371.

Fockenberg, T., 1998.An experimental investigation on the P-T stability of Mg-staurolite in the system MgO-Al2O3-SiO2-H2O.Contributions to Mineralogy and Petrology 130, 187- 198.

Geological Survey Department of Sri Lanka, 1996. Geological Map of Sri Lanka Colombo: Geological Survey Department of Sri Lanka. Scale 1:100000, sheet No14.

Grant. J.A., Weiblen, P W. ., 1971. Retrogradezoning in garnet near the second sillimanite Isograd American Journal of Science 270, 281-296.

Harley, S.L., 1991. The crustal evolution of some East Antarctic granulites. In: Thomson, M.R.A., Crame, J.A.,Thomson, W.J.(Eds.), Geological evolution of Antarctica Cambridge University Press, 7-12.

Harley, S.L., Kelly, N.M. Möller, A., 2007. Zircon and thethermal histories of hot mountain belts. Elements 3, 25-30.

Harley, S.L., 2008. Refining the P-T records of UHT crustal metamorphism. Journal of Metamorphic Geology 26, 125-154.

He, X.F., Santosh, M., Tsunogae, T., Malaviarachchi, S.P.K., 2015. Early to late Neoproterozoic magmatism and magma mixing - mingling in Sri Lanka: Implications for convergent margin processes. Gondwana Research 32, 151-180.

He, X.F., Santosh, M., Tsunogae, T., Malaviarachchi, S.P.K., Dharmapriya, P.L., 2016. Neoproterozoic arc accretion along the 'eastern suture' in Sri Lanka during Gondwana assembly, Precambrian Research279, 57-80.

Hensen, B. J., 1987. P-T grids for silica-undersaturated granulites in the systems MAS (n+4) and FMAS (n+3): tools for the derivation of the P-T paths of metamorphism. Journal of Metamorphic Geology 5, 255-271.

Hiroi, Y., Ogo, Y., Namba, K., 1994. Evidence for prograde metamorphic evolution of Sri Lankan pelitic granulites, and implications for the development of continental crust. In: Raith, M., Hoernes, S. (Eds.),Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka. Precambrian Research 66, 245263.

Hiroi, Y., Yanagi, A., Kato, M., Kobayashi, T., Prame, B., Hokada, T., Satish-Kumar, M., Ishikawa, M., Adachi, T., Osanai, Y., Motoyoshi, Y., Shiraishi, K., 2014. Supercooled melt inclusions in lower-crustal granulites as a consequence of rapid exhumation by channel flow. Gondwana Research 25, 226-234.

Hodges, K.V., McKenna, L.W., 1987. Realistic propagation of uncertainties in geologic

thermobarometry. American Mineralogist 72, 671-680.

Hoffbauer, R., Hoernes, S., Fiorentini, E.,1994. Oxygen isotope thermometry based on a refined increment method and its application to granulite-grade rocks from Sri Lanka. In: Raith, M., Hoernes, S.(Eds.), Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka, Precambrian Research 66, 199-220.

Holland, T.J.B., Powell, R., 1998. An internally consistent thermodynamic dataset for phases of petrological interest.Journal of Metamorphic Geology 16, 309-343.

Hö1zl, S., Kohler, H., Kröner, A., Jaeckel, P., Liew, T. C., 1991. Geochronology of Sri Lanka basement. In: Kröner, A.(Eds.),The Crystalline Crust of Sri Lanka, Part I. Summary of Research of the German-Sri Lankan Consortium, Geological Survey Department of Sri Lanka, Professional. Paper 5, 237-257.

Hö1zl, S., Hofmann, A.W., Todt, W., Köhler, H., 1994. U-Pb geochronology of the Sri Lanka basement. In: Raith, M. and Hoernes, S.(Eds.), Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka, Precambrian Research 66, 123-149.

Hopper, J.R., 1996. The effect of lower crustal flow on continental extension and passive margin formation. Journal of Geophysical Research 10, 175-194.

Hounslow, A.W., Moore, J.M., 1967. Chemical petrology of Grenville schists near Fernleigh,'Ontario.Journal of Petrology 8, l-28.

Karunarathne, D. N., Perera, L.R.K., Fernando, G.W.A.R., 2002. The missing link between Prograde decompression(?) and Retrograde Isobaric Cooling in the Highland Complex Granulites of Sri Lanka. (Abstract) In: Proceedings. 18th Annuel Technical Sessions of the Geologica Socity of Sri Lanka, p. 3.

Kawasaki, T., Osanai, Y., 2008.Empirical thermometer of TiO2 in quartz for ultrahigh-temperature granulites of East Antarctica. In: Satish- Kumar, M., Motoyoshi, Y., Osanai, Y. Hiroi, Y., Shiraishi, K.(Eds.), Geodynamics evolution of East Antarctica: A key to the East-West Gondwana connection , Geological Society of London Special Publications 308, 419-430.

Kehelpannala, K.V.W., 1991. Structural evolution of high-grade terrains in Sri Lanka with special reference to the areas around Dudangaslanda and Kandy. In: (Kröner, A. (Eds.),The Crystalline Crust of Sri Lanka, Part I. Summary of Research of the German-Sri Lankan Consortium, Geological Survey Department, Sri Lanka Professional Paper 5, 69-88.

Kehelpannala, K.V.W., 1997. Deformation of a high-grade Gondwana fragment, Sri Lanka. Gondwana Research 1, 47-68.

Kehelpannala, K.V.W., 2003. Structural evolution of the middle to lower crust in Sri Lanka — a review. Journal of the Geological Society of Sri Lanka 11, 45-86.

Kehelpannala, K.V.W., 2004. Arc accretion around Sri Lanka during the assembly of Gondwana. Gondwana Research 7, 1323-1328.

Kelsey, D.E., 2008. On ultrahigh-temperature crustal metamorphism, Gondwana Research 13, 1-29.

Kleinschrodt, R., 1994. Large scale thrusting in the lower crustal basement of Sri Lanka. In: Raith, M., Hoernes, S.(Eds.),Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka, Precambrian research 66, 39-57.

Kleinschrodt, R., Voll, G., 1994. Deformation and metamorphic evolution of a large-scale fold in the lower crust: The Dumbara synform, Sri Lanka, Journal of Structural Geology 16, 1495-1507.

Kriegsman, L., 1991. Structural geology of the Sri Lankan basement- a preliminary review. In: Kroner, A. (Eds.) The crystalline crustof Sri Lanka, Part I, Summary of Research of the German-SriLankan Consortium. Geological Survey Department, Sri Lanka, Prof. Paper 5, 52-68.

Kriegsman, L.M., 1993. Structural and petrological investigations into the Sri Lankan highgrade terrain - geodynamic evolution and setting of a Gondwana fragment. Geology Ultraiectina, 114, 208 pp.

Kriegsman, L.M., 1994. Evidence for a fold nappe in the high-grade basement of central Sri Lanka: terrane assembly in the Pan-African lower crus In: Raith, M., Hoernes, S.(Eds.),Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka. Precambrian Research 66, 59- 76.

Kriegsman, L.M., 1995. The Pan-African events in East Antarctica: a review from Sri Lanka and the Mozambique Belt. Precambrian Research 75, 263-277.

Kriegsman L.M., 1996. Divariant and trivariant reaction line slopes in FMAS and CFMAS: Theory and applications. Contributions to Mineralogy and Petrology 126, 38-50.

Kriegsman, L.M., Schumacher, J.C., 1999. Petrology of sapphirine bearing and associated granulites from central Sri Lanka. Journal of Petrology 40, 1211-1239.

Kröner, A., Williams, I.S., Compston, W., Baur, N., Vithanage, P.W., Perera, L.R.K., 1987. Zircon ion microprobe dating of high grade rocks in Sri Lanka. Journal of Geology 95, 775-791.

Kröner, A., Cooray, P. G., Vitanage, P.W., 1991.Lithotectonic subdivision of the Precambrian basement in Sri Lanka. In: Kröner,A. (Eds.) The Crystalline Crust of Sri Lanka, Part-1. Summary of Research of the German-Sri Lankan Consortium. Geological Survey Department, Sri Lanka, Professional Paper 5, 5-21.

Kröner, A., Williams, I.S., 1993.Age of metamorphism in the high-grade rocks of SriLanka. Journal of Geology 101, 513-521.

Kröner, A., Jaeckel, P., Williams, I. S., 1994. Pb-loss patterns in zircons from a high-grade metamorphic terrain as revealed by different dating methods: U-Pb and Pb-Pb ages for igneous and metamorphic zircons from northern Sri Lanka. In: Raith, M., Hoernes, S.(Eds.),Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka. Precambrian Research, 66,151-181.

Kröner, A., Kehelpannala, K.V.W., Hegner, E., 2003. Ca. 700-1000 Ma magmatic events and Grenvillian-age deformation in Sri Lanka: relevance for Rodinia supercontinent

formation and dispersal, and Gondwana amalgamation. Journal of Asian Earth Sciences 22, 279-300.

Kröner, A., Rojas-Agramonte, Y., Kehelpannala, K.V.W., Zack, T., Hegner, E., Geng, H. Y., Wong, J., Barth, M., 2013. Age, Nd-Hf isotopes, and geochemistry of the Vijayan Complex of eastern and southern Sri Lanka: a Grenville-age magmatic arc of unknown derivation. Precambrian Research 234, 288-321

Liew, T.C., Milisenda, C.C., Hölzl, S., Köhler, H., Hofmann, A. W., 1991. Isotopic characterisation of the high-grade basement rocks of Sri Lanka. In: Kröner, A.(Eds.)The Crystalline Crust of Sri Lanka, Part I. Summary of Research of the German-Sri Lankan Consortium, Geological Survey Department, Sri Lanka, Professional Paper 5, 258-267.

Madugalla, T.B.N.S., Pitawala, H.M.T.G.A. and Karunaratne, D.G.G.P., 2014. Use of Carbonatites in the Production of Precipitated Calcium Carbonate: A Case Study from Eppawala, Sri Lanka. Natural Resources Research 23, 217-229.

Malaviarachchi, S. P. K., Takasu, A., 2011. Electron Microprobe dating of Monazites from Sri Lanka.Journal of Geological Society of Sri Lanka 14, 81-90.

Malaviarachchi, S.P.K., Dharmapriya, P.L., 2015. Revisiting ultrahigh temperature granulites of Sri Lanka: new prograde and retrograde mineral textures from the Highland Complex.Journal of Indian Institute of Science 95, 159-171.

Mathavan, V., Prame, W.K.B.N., Cooray, P. G., 1999. Geology of the high grade Proterozoic terrains of Sri Lanka and the assembly of Gondwana: an update on recent developments. Gondwana Research 2, 237-250.

Mathavan, V. Fernando, G.W.A.R., 2001. Reactions and textures in grossular-wollastonite-scapolite calc-silicate granulites from Maligawila, Sri Lanka.Lithos 59, 217-232.

Milisenda, C.C., Liew, T. C., Hoffman, A. W., Kröner, A., 1988. Isotopic mapping of age provinces in Precambrian high-grade terrains: Sri Lanka. Journal of Geology 96, 608615.

Milisenda, C. C., Liew, T. C., Hofmann, A. W., Köhler, H., 1994. Nd isotopic mapping of the Sri Lankan basement: update, and additional constraints from Sr isotopes. In: Kröner, A. (Eds.), Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka. Precambrian Research, 66, 95-110.

Newton, R.C,Charlu, T.V., Kleppa,O. J., 1980.Thermochemistryof the high structural state plagioclases.GeochemicaCosmochimicaActa44, 933-941.

Ogo, Y., Hiroi, Y., Prame, K.B.N., Motoyoshi, Y., 1992. A new insight of possible correlation between the Lützow-Holm bay granulites (East Antarctica) and Sri Lankan granulites. In: Yoshida, Y., Kaminuma, K., Shiraishi, K. (Eds.)Proceedings of the Sixth International Symposium on Antarctic Earth Sciences.Recent progress in Antarctic Earth science, 75-86.

Osanai, Y., 1989. A preliminary report on sapphirine / kornerupine granulite from Highland series, Sri Lanka (Extended abstract). Seminar on recent advantages in Precambrian Geology of Sri Lanka, IFS Kandy, Sri Lanka.

Osanai, Y., Ando, K. T., Miyashita, Y., Kusachi, I., Yamasaki, T., Doyama, D., Prame, W.K.B.N., Jayatilake, S., Mathavan, V., 2000. Geological field work in the southwestern and central parts of the Highland Complex, Sri Lanka during 1998-1999, special reference to the highest grade metamorphic rocks. Journal of Geoscience 43, 227-247.

Osanai, Y., Sajeev, K., Owada, M., Kehelpannala, K.V.W., Prame, W.K.B., Nakano, N., Jayatileke, S., 2006. Metamorphic evolution of ultrahigh-temperature and high-pressure granulites from Highland Complex, Sri Lanka. Journal of Asian Earth sciences 28, 2037.

Osanai, Y., Sajeev, K., Nakano, N., Kitano, I., Kehelpannala, K.V.W., Kato, R., Adachi, T. and Malaviarachchi, S.P.K. (2016a) UHT granulites of the Highland Complex, Sri Lanka II: Geochronological constraints and implications for Gondwana correlation. Journal of Mineralogical and Petrological Sciences 111,157-169.

Osanai, Y., Sajeev, K., Nakano, N., Kitano, I., Kehelpannala, K.V.W., Kato, R., Adachi, T.

and Malaviarachchi, S.P.K. (2016b) UHT granulites of the Highland Complex, Sri Lanka I:

Geological and petrological background. Journal of Mineralogical and Petrological Sciences,

111,145-156.

Ouzegane, K., Guiraud, M., Kienast, J.R., 2003. Prograde and retrograde evolution in high-temperature corundum granulites (FMAS and KFMASH systems) from In Ouzzal terrane (NW Hoggar, Algeria), Journal of Petrology 44, 517-545.

Peacock, S. M., Goodge, J.W., 1995. Eclogite-facies metamorphism preserved in tectonic blocks from a lower crustal shear zone, central Transantarctic Mountains, Antarctica, Lithos 36, 1- 13.

Perera, L.R.K., 1984. Co-existing cordierite-almandine-- a key to the metamorphic history of Sri Lanka. Precambrian Resesearch 25, 349-364.

Perera, L.R.K., 1987. Petrogenisis of granulite-facies metamorphic rocks in Sri Lanka, Unpublished M.Phil.Thesis, University of Peradeniya, 124 pp.

1206 Perera, L.R.K., 1994. P-T-t path vs isotopic ages in correlations, anti-correlations and

1207 pseudo-correlations of regional metamorphic terrains, a case study from Sri Lanka.

1208 Journal of the Geological Society of Sri Lanka 5, 27-41.

1210 Perera, L.R.K., Kagami, H., 2011. Centimetre- and metre-scale Nd and Sr

1211 isotopichomogenization in Kadugannawa Complex, Sri Lanka.Journal of Geological

1212 Socity of Sri Lanka 14,129-141.

1214 Pitawala, A., Lottermoser, B.G., 2012. Petro genesis of the Eppawalacarbonatites, Sri Lanka:

1215 A cathodoluminescence and electron microprobe study. Mineralogy and Petrology 92,

1216 57-70.

1218 Pohl, J.G. Emmermann, R., 1991. Chemical composition of theSri Lanka Precambrian

1219 Basement. In: A. Kröner (Eds.) The Crystalline Crust of Sri Lanka, Part 1. Geological

1220 Survey Department of Sri Lanka. Professional. Paper 5.94-123.

1222 Powell, R., Holland, T.J.B., 2008. On thermobarometry. Journal of MetamorphicGeology 26,

1223 155-179.

1225 Prame, W.K.B.N., 1991. Petrology of the Kataragama Complex, Sri Lanka: evidence for high

1226 P-T granulite facies metamorphism and subsequent isobaric cooling. In: Kröner, A.

1227 (Eds.) The Crystalline Crust of Sri Lanka, Part 1. Summary of Research of the German-

1228 Sri Lankan Consortium. Geological Survey Department Sri Lanka, Professional Paper 5,

1229 200-224.

1231 Raase, R., Schenk, V., 1994. Petrology of granulite facies metapelites of the Highland

1232 Complex, Sri Lanka : implications for metamorphic zonation and P-T path. In: Raith, M.,

1233 Hoernes, S.(Eds.)Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks,

1234 With Special Emphasis on Sri Lanka. Precambrian Research, 66, 265-294.

1236 Sajeev, K., Osanai, Y., 2004a. Ultrahigh-temperature metamorphism (1150 0C, 12 kbar) and

1237 multi-stage evolution of Mg, Al rich granulites from the central Highland Complex, Sri

1238 Lanka, Journal of petrology 45, 1821-1844.

1240 Sajeev, K., Osanai, Y., 2004b. Osumilite and spinel+quartz from Sri Lanka: implications for

1241 UHT conditions and retrograde P-T path, Journal of Mineralogical and Petrological

1242 Sciences 99, 320-327.

1244 Sajeev, K., Osanai, Y., Connolly, J. A. D., Suzuki, S., Ishioka, J., Kagami, H., Rino, S., 2007.

1245 Extreme crustal metamorphism during a Neoproterozoic event in Sri Lanka, A study of

1246 dry mafic granulites, Journal of Geology 115, 563-582.

Sajeev, K., Osanai, Y., Kon, Y., Itaya, T., 2009. Stability of pargasite during ultrahightemperature metamorphism: a consequence of titanium and REE partitioning? American Mineralogist 94, 535-545.

Sajeev, K., Williams, I.S., Osanai, Y., 2010. Sensitive high-resolution ion microprobeU-Pb dating of prograde and retrograde ultrahigh-temperature metamorphismas exemplified by Sri Lankan granulites. Geology 38, 971-974.

Santosh, M., Tsunogae, T., Malaviarachchi, S. P. K., Zhang, Z., Ding, H., Tang, L., Dharmapriya, P. L., 2014. Neoproterozoic crustal evolution in Sri Lanka: in- sights from petrologic, geochemical and zircon U-Pb and Lu-Hf isotopic data and implications for Gondwana assembly. Precambrian Research 255, 1-29.

Schenk, V., Raase, P., Schumacher, R., 1991.Metamorphic zonation and P-T history of the Highland Complex in Sri Lanka. In: Kröner, A. (Eds.)The Crystalline Crust of Sri Lanka, Part I. Summary of Research of the German-Sri Lankan Consortium, Geological Survey Department, Sri Lanka, Professional Paper 5, 150-163.

Schreyer, W., 1988. Experimental studies on metamorphism of crustal rocks under mantle pressures, Mineralogical Magazine 52, 1- 26.

Schreyer, W., Chinner, G.A., l966. Staurolite-quartzite bands in kyanite quartzite at Big Rock, Rio Arriba County, New Mexico. Contributions to Mineralogy and Petrology, 12, 223-244.

Schumacher, R., Schenk, V., Raase, P., Vitanage, P.W., 1990. Granulite facies metamorphism of metabasic and intermediate rocks in the Highland Series of Sri Lanka. In: Brown, M., Ashworth, J.R. (Eds.)High-Grade Metamorphism and Crustal Anatexis, Allen and Unwin, London 235-271.

Schumacher, R., Faulhaber, S., 1994.Summary and discussion of P-T estimates from garnet-pyroxene-plagioclase-quartz-bearing granulite-facies rocks from Sri Lanka. In: Raith, M. & Hoernes, S.(Eds.), Tectonic, Metamorphic and Isotopic Evolution of Deep Crustal Rocks, With Special Emphasis on Sri Lanka, Precambrian Research 66, 295-30.

Searle, M., Hacker, B. R., Bilham, R., 2001. The Hindu Kush Seismic Zone as a Paradigm for the Creation of Ultrahigh-Pressure Diamond- and Coesite-Bearing Continental Rocks. Journal of Geology 109, 143-153.

Shimpo, M., Tsunogae, T., Santosh, M., 2006. First report of garnet-corundum rocks from southern India: implications for prograde high-pressure (eclogite facies?) metamorphism. Earth and Planetary Science Letters 242, 111-129.

1291 Shulters, J., Bohlen, S., 1988. The stability of hercynite and hercynite-gahnite spinels in

1292 corundum- bearing or quartz-bearing assemblages.Journal of petrology 30, 1017 -

1293 1031.

1295 Song, S., Zhang, L., Niu, Y., 2004.Ultra-deep origin of garnet peridotite from the North

1296 Qaidam ultrahigh-pressure belt, northern Tibetan Plateau, NW China. American

1297 Mineralogist 89, 1330-1336.

1299 Stüwe, K., 1997. Effective bulk composition changes due to cooling: a model predicting

1300 complexities in retrograde reaction textures. Contributions to Mineralogy and

1301 Petrology 129:43-52.

1303 Tajcmanovâ, L., Connolly, J.A.D., Cesare, B., 2009. A thermodynamic model for titanium

1304 and ferric iron solution in biotite. Journal of Metamorphic Geology 27, 153-165.

1306 Takamura, Y., Tsunogae, T., Santosh, M., Malaviarachchi, S.P.K., Tsutsumi, Y., 2015,

1307 Petrology and zircon U-Pb geochronology of metagabbro from the Highland

1308 Complex, Sri Lanka: Implications for the correlation of Gondwana suture zones:

1309 Journal of Asian Earth Sciences 113, 826-841.

1311 Takamura, Y., Tsunogaea, T., Santosh, M., Malaviarachchi, S.P.K., Tsutsumi, Y., 2016. U-Pb

1312 geochronology of detrital zircon in metasediments from Sri Lanka: Implications for the

1313 regional correlation of Gondwana fragments. Precambrian Research 281, 434-452.

1315 Tapponnier, P., Mercier, J. =L., Armijo, R., Toglin, H., Ji. Z., 1981. Field evidence for active

1316 normal faulting in Tibet. Nature 294, 410-414.

1318 Taylor S. R., McLennan S. M., 1985. The Continental Crust: Its Composition and Evolution.

1319 Blackwell Scientific Publication, Carlton, 312 p

1321 Vitanage, P.W., 1972. Post-Precambrian uplift and regional neotectonic movements in

1322 Ceylon. 24th Int. Geological Congress, Montreal, Section 3, 642-654.

1324 Voll, G., Kleinschrodt, R., 1991. Sri Lanka: Structural, magmatic and and metamorphic

1325 development of a Gondwana fragment. . In: Kröner, A. (Eds.), The Crystalline Crust of

1326 Sri Lanka, Part I. Summary of Research of the German-Sri Lankan Consortium,

1327 Geological Survey Department, Sri Lanka, Professional Paper 5, Geological Survey

1328 Department, Sri Lanka, Professional Paper 5, 22-52.

1330 Voll, G., Evangelakakis, C., Kroll, H., 1994. Revised two-feldspar geothermometry applied

1331 to Sri Lankan feldspars. Precambrian Research 66, 351-378.

Waldbaum, D.R., Thompson, J.B.J., 1969. Mixing properties of sanidine crystalline solutions: IV. Phase diagrams from equation of state. American Mineralogist 54, 1274-1298.

Wark, D.A., Watson, E.B., 2006. TitaniQ: a titanium-in-quartz geothermometer. Contributions to Mineralogy and Petrology 152, 743-754.

Waters, D., 1986. Metamorphic history of sapphirine-bearing and related magnesian gneisses from Namaqualand, South Africa. Journal of Petrology, 27, 541-565.

Weerakoon, M.W.K., Miyazaki, T. Shuto, K., Kagamil, H., 2001. Rb-Sr and Sm-Nd geochronology of the Eppawala metamorphic rocks and carbonatite, Wanni Complex, Sri Lanka. Gondwana Research 4, 409-420.

White, R.W, Powell, R, Holland, T.J.B., Worley, B.A., 2000. The effect of TiO2 andFe2O3 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.

White, R.W., Powell, R., Clarke, G.L., 2002. The interpretation of reaction textures in Fe-rich metapelitic granulites of the Musgrave Block, central Australia: constraints from mineral equilibria calculations in the system K2O-FeO-MgO-Al2O3-SiO2-H2O-TiO2-Fe2O3. Journal of Metamorphic Geology 20, 41-55.

White, R.W., Powell, R., Halpin, J.A., 2004.Spatially-focussed melt formation in aluminous metapelites from Broken Hill, Australia. Journal of metamorphic Geology 22, 825845.

White, R. W., Powell, R., Holland, T.J.B., 2007. Progress relating to calculation of partial melting equilibria for metapelites. Journal of Metamorphic Geology 25, 511-527.

Wickramasinghe G.G.M.S., Perera, L.R.K., 2014. P-T-t Path of the southwestern Highland Complex around Balangoda, Sri Lanka. (Abstract) In Proceedings of the Peradeniya Univ. International Research Sessions, Sri Lanka 18, 132.

Willbold, M., Hegner, E., Kleinschrodt, R., Stosch, H.G., Kehelpannala, K.V.W., Dulski, P., 2004.Geochemical evidence for a Neoproterozoic magmatic continental margin in Sri Lanka - relevance for the Rodinia-Gondwana supercontinental cycle. Precambrian Research 130, 185-198.

Witney, D.L., Dilek, Y., 1997. Core complex development in central Anatolia. Geology 25, 1023-1026.

Yang, Y., Liu, M., 2009. Crustal thickening and lateral extrusion during the Indo-Asian collision: A 3D viscous flow model. Tectonophysics, 465, 128-135.

Yardley, B.W.D., 1977. An empirical study of diffusion in garnet. American Mineralogist 62, 793-800.

Figure and table captions

Figure captions

Figure 1. (a) Geological subdivision of the Sri Lankan basement (after Cooray, 1994). The sampling locality is shown by a star along with other reported UHT localities: 1. Osanai (1989) (>900 °C); 2. Kriegsman and Schumacher (1999) (830 °C, here corrected to 950 °C); 3. Sajeev and Osanai (2004a) (950 °C): 4. Sajeev and Osanai (2004b) (1150 °C); 5. Osanai et al. (2006) (>1000 °C); 6. Sajeev et al. (2007) (925 °C); 7. Dharmapriya et al. (2014) (950-975 °C); 8. Sajeev et al. (2009) (950 °C); 9. Dharmapriya et al. (2015a) (950 -975 °C); 10. Takamura et al. (2015) (940 °C); 11-13. Dharmapriya et al. (2015b) (900-975 °C); (b) detailed map of the area south of Gampola (Geological Survey and Mines Bureau 1996, map sheet No. 14, used with permission), showing the khondalite localities within a zone of metasediments.

Figure 2. Comparison of the previously proposed P-T paths for the Highland Complex granulites with published paths (a) after Perera (1987, 1994); (b) Hiroi et al. (1994), where A and B are estimated peak metamorphic conditions for rocks from around Kandy (A) and Colombo (B), and C indicates retrograde metamorphic conditions for rocks carrying the andalusite+siderite+cordierite assemblage. Isobaric cooling path suggested for metaigneous

rocks by by Schenk et al. (1991), Prame (1991) and Schumacher et al. (1991) is also shown; (c) Kriegsman (1993, 1996); (d) Raase and Schenk (1994), showing Eastern Highland Complex (EHC) and Western Highland Complex (WHC, now called Wanni Complex); (e) Osanai et al. (2006) for UHT granulites and Matavan and Fernando (2001) for cack-silicate rocks; (f) Dharmapriya et al. (2015a) and Takamura et al. (2015).

Figure 3. (a) Field occurrence of the khondalite, (b) a hand specimen of the khondalite cut perpendicular to the foliation, (c) a sketch block diagram of the khondalite representing Garnet3 and Garnet4.

Figure 4.Inclusion textures in Grt1. (a) a sketch diagram showing inclusion phases; (b) oriented quartz grains in the rim area; (c) coexisting prismatic sillimanite and quartz in the mantle area; (d) isolated sillimanite inclusions in the mantle area; (e) isolated biotite and staurolite inclusion towards the rim area; (f) CPL view of kyanite with pseudomorphed margins; (g) distribution of inclusion phases from rim to core under PPL. Mineral abbreviations: Ky-pseu - kyanite pseudomorphs; Sil - sillimanite; Grt - garnet: Ilm -ilmenite: Qtz - quartz; St - staurolite; Bt - biotite.

Figure 5. Inclusion textures in Grt2. (a) Anhedral corundum and kyanite pseudomorph within core area of garnet; (b) aggregate of fibrolitic sillimanite pseudomorphing kyanite in core area of garnet; (c) isolated kyanite inclusion; (d) medium- to coarse-grained prismatic sillimanite and medium-grained spinel between corundum and garnet (PPL view); (e) tiny spinel inclusions and corundum inclusions in coarse prismatic sillimanite, with staurolite present on the right bottom; (f) EDS-Xray image of Al in (e); (g) Mg in (e); (h) cluster of fine grained K-feldspars and rare quartz inclusions in the rim area; (i) formation of pyrophyllite

1429 towards the rim; (j) oriented ilmenite needles and apatite rods, with rutile also present; (k)

1430 preferredly oriented ilmenite needles in two directions in garnet.

1432 Figure 6. Petrography of Grt3 and Grt4 (a, b, c and d are Grt3 and e, f and g are Grt4). (a) PPL

1433 view of the rim to mantle area; (b) PPL view of Grt3 perpendicular to the rotation axis; (c) co-

1434 existence of sillimanite+biotite in the presence of spinel in the core; (d) pyrophyllite rim

1435 around sillimanite in the presence of quartz (PPL view); (e) garnet containing sillimanite

1436 inclusions oriented parallel to the matrix sillimanite; (f) garnet with ilmenite, rutile,

1437 sillimanite and quartz inclusions; (g) Rutile needle in the matrix close to garnet; , (h)

1438 inclusions of rutile, ilmenite, quartz and sillimanite in fine grained garnet, (i) partially

1439 broken-down sillimanite, ilmenite and quartz rains are present in the matrix adjacent to

1440 garnet-rutile intergrowth.

1443 Figure 7.Petrography of the matrix minerals in the khondalite. (a) Elongate quartz and

1444 preferredly oriented sillimanite grains within the main foliation; (b) fine sillimanite inclusions

1445 in anhedral quartz grain; (c) cluster of prismatic sillimanite and stretched quartz in the matrix,

1446 with ilmenite and recrystallized K-feldspar also present; (d) euhedral coarse K-feldspars; (e)

1447 co-existence of prismatic sillimanite and garnet; (f) pyrophyllite rims around matrix

1448 sillimanite under the present of quartz.

1450 Figure 8. Raman spectra with characteristics Raman peaks. (a) kyanite enclosed by Grti; (b)

1451 kyanite enclosed by Grt2; (c) staurolite enclosed by Grt1; (d) staurolite enclosed by Grt2.

Figure 9. (a) T-X(Fe2O3) pseudosection calculated in the NCKFMASHTO system at P = 9 kbar for the khondalite; (b) P-T pseudosection calculated at X(Fe2O3) = 0.01 for the quartz khondalite. Solution-phase models used in the calculations are: garnet - Gt(WPH): White et al. (2007); biotite - Bio(TCC): Tajcmanovâet al. (2009); ilmenite -Ilm(WHP): Whiteet al. (2000); alkali-feldspar - San: Waldbaum and Thompson (1969); plagioclase - Pl(h): Newton et al. (1980); spinel - Sp(WPC): White et al. (2002); cordierite - hCrd: Holland and Powell (1998); melt - melt (HP): White et al. (2007). Mineral abbreviations after Holland and Powell (1998)

Figure 10.P-Tpseudosection calculated with the melt-reintegration technique of White et al. (2004) in the system NCKFMASHTO at X(Fe2O3) = 0.01. Solution-phase models used in the calculations are: garnet - Gt(WPH): White et al. (2007); biotite - Bio(TCC): Tajcmanovâ et al. (2009); ilmenite -Ilm(WHP); White et al. (2000); alkali-feldspar - San: Waldbaum and Thompson (1969); plagioclase - Pl(h): Newton et al. (1980); melt - melt (HP): White et al. (2007). Mineral abbreviations after Holland and Powell (1998).

Figure 11. P-T-diagrams showing the modal proportion of (a) garnet, (b) biotite, (c) muscovite, (d) staurolite, (e) sillimanite, (f) kyanite, (g) ilmenite, (h) rutile, (i) quartz, (j) plagioclase, (k) K-feldspar, (l) melt expressed in vol.% calculated of melt re-integration until to have a composition that allows the wet solidus in the system NCKFMASHTO. The arrow is showing the P-T trajectory followed by the rock

Figure 12. P-T pseudosection calculated of melt re-integration modeling above wet-solidus in the system NCKFMASHTO at X(Fe2O3) = 0.01. Solution-phase models used in the calculations are: garnet - Gt(WPH): White et al. (2007); biotite - Bio(TCC): Tajcmanovâet

al. (2009); ilmenite -Ilm(WHP); Whiteet al. (2000); alkali-feldspar - San: Waldbaum and Thompson (1969); plagioclase - Pl(h): Newton et al. (1980); white mica - wm (HP): Holland and Powell (1998); melt - melt (HP): White et al. (2007). Mineral abbreviations after Holland and Powell (1998)

Figure 13.P-T-diagrams showing the modal proportion of (a) garnet, (b) biotite, (c) muscovite, (d) staurolite, (e) sillimanite, (f) kyanite, (g) ilmenite, (h) rutile, (i) quartz, (j) plagioclase, (k) K-feldspar, (l) melt expressed in vol.% calculated of melt re-integration modeling above wet-solidus in the system NCKFMASHTO. The arrow is showing the P-T trajectory followed by the rock.

Figure 14. A P-T diagram outlining the overall shape of the inferred P-T path of the khondalite

Figure 15.Zircon U-Pb geochronology: (a) Cathodoluminescence (CL) images of representative zircons in khondalite (analytical point are also given); (b) concordia diagram

206 238

showing Weighted mean Pb/ U age of zircons from the khondalite; (c) age data

206 238

histogram and Probability Density plot (calculated for Pb/ U ages) of khondalite.

Table Captions

Table 1.Representative bulk chemical composition of the khondalite from XRF. Table 2. Representative electron microprobe analyses of garnets and staurolite in the khondalite

Table 3. Representative electron microprobe analyses of biotite, spinel, corundum, K-feldspar, pyrophylite, rutile and ilmenite in the khondalite

1503 Table 4.Melt compositions used in the pseudosection calculation after melt reintegration.

1504 CXRF: bulk rock composition measured by XRF; CM1- CM4: rock composition calculated after

1505 melt reintegration; M1 - M4: composition of melt reintegrated. Melt amount (vol. %), T (°C)

1506 and P (kbar): PT conditions and amount of melt reintegrated at each step.

1508 Table 5.Zircon U-Pb analytical data by LA-ICPMS

Table 1. Representative bulk chemical compositions of the khondalite from XRF (major _ elements are in mol%; trace elements are in ppm)

Element Quantity

SiO2 72.88

TiO2 0.60

Al2O3 12.44

FeO 6.71

MnO 0.09

MgO 2.37

CaO 1.02

Na2O 1.31

K2O 2.71

&2O3 1.90

P2O5 0.03

H2O* 0.73

Total 100

XMg 0.26

Rb 90.5

Ba 919.3

Sr 159.7

Nb 10.8

Zr 294.9

Y 44.1

Zn 64.6

Cu 7.4

Ni 0.0

Cr 670.4

V 100.1

Sc 8.7

Th 19.6

H2O* based on loss on ignition (LOI)

Table 2. Representative electron microprobe analyses of garnets and staurolite in the khondalite (major elements are in wt.%).

Grt1_ _Grt2_ _Grt3_ _Grt4_ _St

Core Rim Core Rim Core Rim Core Rim In Grt1 In Grt2

SiO2 36.44 36.20 37.00 36.65 37.62 36.41 36.31 36.48 23.59 23.29

TiO2 0.03 0.02 0.03 0.04 0.25 0.02 0.04 0.05 1.59 1.55

Al2O3 21.57 21.46 21.00 21.34 21.42 21.62 21.53 21.61 57.81 57.87

Cr2O3 0.05 0.01 0.02 0.02 0.07 0.00 0.02 0.01 0.10 0.13

FeO 34.52 34.67 33.60 34.47 33.12 34.68 34.33 34.19 11.54 11.23

MnO 0.73 0.70 3.55 3.17 1.13 0.94 0.72 0.70 0.05 0.05

MgO 5.26 5.25 3.38 3.73 4.82 4.59 5.66 5.51 1.91 2.55

CaO 1.39 1.29 2.13 1.55 1.49 1.39 1.35 1.35 0.01 0.05

Na2O 0.01 0.01 0.03 0.02 0.00 0.02 0.00 0.00 0.00 0.00

K2O 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

ZnO 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.07 1.60 1.56

Total 100.00 99.62 100.73 100.98 99.92 99.68 99.96 99.96 98.20 98.28

O a.f.u Si 12 12 12 12 12 12 12 12 23 23

2.97 2.91 2.97 2.94 2.99 2.92 2.91 2.92 3.28 3.23

Ti bdl bdl bdl bdl 0.02 bdl bdl bdl 0.17 0.16

Al 2.04 2.04 1.99 2.01 2.01 2.04 2.03 2.04 9.46 9.46

Cr bdl bdl bdl bdl bdl bdl bdl bdl 0.01 0.01

Fe 2.31 2.33 2.25 2.31 2.20 2.35 2.30 2.29 1.34 1.30

Mn 0.05 0.05 0.24 0.22 0.08 0.06 0.05 0.05 0.01 0.01

Mg 0.63 0.63 0.40 0.45 0.57 0.55 0.68 0.66 0.40 0.53

Ca 0.12 0.11 0.18 0.13 0.13 0.12 0.12 0.12 bdl 0.01

Na bdl bdl 0.01 bdl bdl bdl bdl bdl bdl bdl

K bdl bdl bdl bdl bdl bdl bdl bdl bdl bdl

Zn bdl bdl bdl bdl bdl bdl bdl bdl 0.16 0.16

Total cation 8.06 8.07 8.04 8.06 7.99 8.06 8.08 8.06 14.82 14.87

Alm 0.74 0.74 0.79 0.73 0.74 0.76 0.73 0.73

Spe 0.02 0.02 0.08 0.07 0.03 0.02 0.02 0.02

Pyr 0.21 0.21 0.13 0.15 0.19 0.18 0.22 0.22

Grs 0.04 0.04 0.06 0.05 0.04 0.04 0.04 0.04

XMg 0.21 0.21 0.15 0.16 0.21 0.19 0.23 0.22 0.23 0.29

bdl - below the detection limit; wt.% - weight% oxide; a.f.u - atoms per formula unit

Table 3. Representative electron microprobe analyses of biotite, spinel, corundum, K-feldspar, pyrophylite, rutile and ilmenite in the khondalite (major elements are in wt.%).

Bt in Bt in Kfs- Kfs- Prl- Ilm-

Grt1 Grt3 Spl in Grt2 Spl in Grt3 Crn in Grt2 Grt3 X. Matrix Grt3 Ru-Grt4 Grt4

SiO2 33.55 33.80 0.03 0.03 0.01 65.04 65.55 58.96 0.34 0.00

TiO2 8.64 9.91 0.04 0.03 0.00 0.04 0.03 0.01 98.82 52.93

Al2O3 16.76 15.93 56.21 59.07 99.59 18.69 18.94 30.22 0.01 0.06

Cr2O3 0.08 0.05 0.40 0.37 0.03 0.01 0.00 0.05 0.09 0.05

FeO 18.55 19.70 33.09 25.87 0.65 0.01 0.03 3.80 0.72 45.86

MnO 0.00 0.01 0.11 0.05 0.00 0.00 0.00 0.04 0.00 0.20

MgO 7.92 7.19 4.07 4.31 0.00 0.00 0.00 1.09 0.00 1.32

CaO 0.00 0.05 0.00 0.02 0.00 0.24 0.31 0.35 0.10 0.01

Na2O 0.07 0.09 0.00 0.00 0.00 2.85 3.72 0.17 0.00 0.00

K2O 9.62 9.32 0.11 0.00 0.00 12.13 10.84 0.95 0.00 0.00

ZnO 0.00 0.00 5.27 10.31 0.00 0.00 0.00 0.01 0.11

Total 95.19 96.04 99.33 100.05 100.29 99.03 99.43 95.64 100.18 100.53

O 22 22 4 4 3 8 8 22 2 3

Si 5.15 5.16 bdl bdl bdl 2.99 2.99 7.31 0.01 0.00

Ti 1.00 1.14 bdl bdl bdl bdl bdl bdl 0.99 0.99

Al 3.03 2.87 1.92 1.98 1.99 1.01 1.02 4.42 bdl bdl

Cr 0.01 0.01 0.01 0.01 bdl bdl bdl bdl bdl bdl

Fe 2.38 2.52 0.80 0.62 0.01 bdl bdl 0.39 0.01 0.96

Mn bdl bdl bdl bdl bdl bdl bdl bdl bdl bdl

Mg 1.81 1.64 0.18 0.18 bdl bdl bdl 0.20 bdl 0.05

Ca 0.001 0.01 Bdl bdl bdl 0.01 0.02 0.04 bdl bdl

Na 0.02 0.02 bdl bdl bdl 0.25 0.33 0.04 bdl bdl

K 1.88 1.82 bdl bdl bdl 0.71 0.63 0.15 bdl bdl

Zn bdl bdl 0.11 0.22 bdl bdl bdl bdl bdl bdl

Total cation 15.29 15.18 2.92 2.79 2.00 4.98 4.98 12.57 1.10 2.01

Fe+2 0.80 0.62

Fe+3 0.00 0.00

An 0.01 0.02

Ab 0.26 0.34

Or 0.73 0.65

*Mg 0.43 0.39 0.24 0.18 0.34 0.05

bdl - below the detection limit; wt.% - weight% oxide; a.f.u - atoms per formula unit

Table 4: Bulk rock and melt compositions used in the pseudosection calcualtion. CXRF: bulk rock composition measured by XRF; CMi - CM4: rock composition calculated after melt reintegration; Mi - M4: composition of melt reintegrated. Melt amount (vol.%), T (°C) and P (kbar): PTconditions and amount of melt reintegrated at each step.

mol% CXRF M1 Cm1 M2 CM2 M3 CM3 M4 CM4

SiO2 72.82 62.94 72.46 61.86 71.71 58.48 70.61 57.38 69.26

TiO2 0.60 0.00 0.60 0.00 0.60 0.00 0.60 0.00 0.60

AI2O3 12.34 7.89 12.18 7.81 11.87 7.27 11.49 7.04 11.17

FeO 6.70 0.46 6.47 0.40 6.05 0.28 5.56 0.27 5.23

Fe2O3 0.07 0.00 0.07 0.00 0.07 0.00 0.07 0.00 0.05

MgO 2.38 0.12 2.29 0.09 2.14 0.07 1.97 0.06 1.87

CaO 1.03 0.64 1.01 0.65 0.99 0.48 0.95 0.29 0.91

Na2O 1.31 3.19 1.38 3.26 1.51 3.49 1.68 3.94 1.75

K2O 1.89 3.17 1.94 3.10 2.02 2.83 2.09 2.38 2.08

H2O* 0.73* 21.60 1.50 22.84 2.99 27.11 5.01 28.64 4.34

^Mg 0.26 0.21 0.26 0.18 0.26 0.20 0.26 0.19 0.26

Melt amount 3.70 7.00 8.35 10.20

(vol.%)

T (°C) 820 812 777 757

P (kbar) 9.5 10 10.5 10

* H2O content form loss on ignition (LOI)

Table 5 Zircon U-Pb analytical data by LA-ICPMS.

Sample Spots

Element concentration (PPm)

Isotope ratios!' ± ig)

Age(Ma±1g)

07Pb/206Pb

Pb/ Th

07Pb'206Pb

06Pb/238U

08Pb/232Th

GAMP-01 GAMP-03 GAMP-05 GAMP-06 GAMP-07 GAMP-08 GAMP-09 GAMP-11 GAMP-12 GAMP-13 GAMP-14 GAMP-15 GAMP-16 GAMP-19 GAMP-20 GAMP-21 GAMP-23 GAMP-24 GAMP-25 GAMP-26 GAMP-27 GAMP-30 GAMP-31 GAMP-32 GAMP-33 GAMP-35 GAMP-37 GAMP-38 GAMP-40 GAMP-41 GAMP-44 GAMP-45 GAMP-48 GAMP-49 GAMP-50 GAMP-52 GAMP-55

8.73 16.25 10.32 13.56 5.77 7.05 16.31 16.18 8.27 9.83 6.95 10.48 10.75 14.29 6.48 9.97

16.82 7.21 11.82 10.18 18.78

7.74 6.83 10.02 7.95 9.87 9.74 8.33 9.7

11.63 12.74 11.61 15.44 17.86 8.99 7.92

781.3 1145.37 702.42 1029.82 527.24 500.34 1163.76

725.73 681.75 645.61 724.83 572.07 797.3 580.92 654.9 792.34

623.74 571.44 680.73 1135.89

719.11 865.24

248.12 491.11

672.3 734.91 649.36 621.24 886.49 336.42 256.78 430.42 767.46

569.04 671.78 619.42 750.55

0.01 0.01 0.01 0.01 0.01

0.01 0.02 0.01

0.02 0.01

0.02 0.01

0.02 0.01 0.01

0.01 0.03

0.01 0.01 0.01 0.02 0.03 0.01 0.01 0.01 0.02 0.02 0.01

0.03 0.05 0.03 0.02 0.03 0.03 0.01 0.01

0.0601 0.05835 0.05836 0.05708 0.0593 0.06184 0.0611 0.06002 0.05963 0.05737 0.05778 0.05706 0.06218 0.05887 0.06121 0.06058 0.05753 0.06006 0.07027 0.05868 0.05991 0.05993 0.06011 0.05806 0.05924 0.05996 0.05931 0.05899 0.06031 0.05868 0.05781 0.05867 0.05996 0.05965 0.05837 0.05989 0.06056

0.00126 0.00114 0.00148 0.00141 0.00168 0.00162 0.00112 0.00141 0.00156 0.00138 0.00135 0.00147 0.00131 0.00141 0.00167 0.00144 0.00145 0.00144 0.00146 0.00118 0.00144 0.00121 0.00219 0.00267 0.00129 0.0014 0.00156 0.00163 0.00144 0.00228 0.0027 0.00209 0.00125 0.00155 0.00187 0.00135 0.00146

0.76429 0.66948 0.7038 0.68872 0.72051 0.74697 0.73756 0.72262 0.67342 0.69232 0.6977 0.68808 0.74772 0.67732 0.73922 0.73789 0.69674 0.6883 0.85189 0.7164 0.72913 0.72582 0.72465 0.70257 0.71552 0.72557 0.67834 0.6778 0.72746 0.70633 0.6645 0.70468 0.7262 0.71751 0.70517 0.725 0.74135

0.01465 0.01188 0.01671 0.01586 0.01924 0.01838 0.0121 0.01575 0.0166 0.01553 0.01521 0.01663 0.01441 0.0152 0.01905 0.01633 0.01647 0.01535 0.01619 0.01317 0.01638 0.01342 0.02544 0.03125 0.01445 0.01574 0.01676 0.01776 0.01622 0.02646 0.03014 0.02406 0.01393 0.01755 0.02158 0.01523 0.01672

0.09221 0.0832 0.08745 0.08749 0.0881 0.0876 0.08753 0.0873 0.0819 0.08751 0.08756 0.08745 0.0872 0.08343 0.08758 0.08833 0.08782 0.0831 0.0879 0.08852 0.08826 0.08782 0.08742 0.08774 0.08758 0.08775 0.08294 0.08332 0.08747 0.08728 0.08334 0.08709 0.08782 0.08722 0.08759 0.08778 0.08876

0.00075 0.00066 0.00078 0.00077 0.00083 0.0008 0.00068 0.00075 0.00074 0.00075 0.00075 0.00078 0.00072 0.00072 0.00082 0.00077 0.00078 0.00072 0.00074 0.00071 0.00077 0.00071 0.00095 0.00115 0.00073 0.00075 0.00075 0.00078 0.00077 0.00101 0.00105 0.00094 0.00072 0.00079 0.0009 0.00074 0.00078

0.05054 0.02712 0.02982 0.03446 0.03134 0.06153 0.05611 0.03336 0.03579 0.02799 0.0289 0.03552 0.04173 0.03313 0.03502 0.03641 0.02917 0.04001 0.25314 0.03671 0.03144 0.03116 0.02743 0.03373 0.02817 0.02056 0.03381 0.02709 0.06749 0.03248 0.03822 0.03104 0.0285 0.02733 0.02939 0.02874 0.03311

0.00401 0.00208 0.00308 0.00327 0.00489 0.00479 0.00267 0.00244 0.00414 0.00295 0.00397 0.00307 0.00321 0.00242 0.0049 0.0036 0.00406 0.00213 0.00898 0.00284 0.00309 0.00186 0.00351 0.00734 0.00274 0.00334 0.00319 0.00302 0.00507 0.00394 0.00367 0.00311 0.0024 0.00206 0.00253 0.00293 0.0037

668.4 642.7

604.3 590.1 505.3 521.3 493.1 680.3

646.5 624.3

511.5 605.7 936.3

555.3 600.1

607.4 531.7

576 602.1

554.7 602

599.5 623.5

44.59 42.14

54.44 53.95

60.23 55.09 38.82 49.89 55.9

50.77 56.29

44.24 51.53 57.66

54.78 50.87 42.03 43.26 51.24 43.14 77.06 98.14

46.74 49.62

56 59.22 50.73 82.69 99.58

44.42 55.29 68.56 48.1 51.06

541.1 532 551

566.5 561

522.8 534.2

566.9 525.2 561.9

561.2 536.8 531.8

548.5 556

554.1 553.4

540.3 548 554

542.6 517.4 541.6 554.3

541.9 553.6 563.2

9.53 11.36 10.68 7.07 9.28 10.08 9.32

9.1 10

9.2 11.12

9.54 9.85

9.23 8.88 7.79 9.62 7.89 14.98 18.64

8.55 9.26 10.14 10.75 9.53 15.74 18.38 14.33 8.19 10.38 12.85

8.97 9.75

515.2 540.4

541.3 540.9 539.6

540.4 539

541.2 545.6 542.6 514.6

545.2 542.6

540.3 542.2 541.2

539.4 516.1

4.45 3.9 4.61

4.54 4.94 4.74 4.03

4.45 4.43

4.46 4.42

4.26 4.29 4.87

4.6 4.29 4.37

4.56 4.2

5.62 6.83 4.32

4.47 4.49 4.65 4.54 5.99

6.23 5.58

4.27 4.69 5.34 4.41

996.6 77.14

540.8 40.85

593.9 60.48

684.7 63.93

623.8 95.77 1207 91.18

1103.4 51.12

663.2 47.73

710.8 80.84

557.9 58.07 575.9 77.97

705.5 59.87

826.4 62.36 658.9 47.28

695.6 95.62

722.8 70.27

581.2 79.72

792.9 41.39 4561 144.86

728.7 55.44 625.6 60.49

620.1 36.41 547 69.08

670.5 143.45

561.5 53.85

411.3 66.06

672.2 62.4 540.2 59.39 1320.2 96.01

646.1 77.19

758.2 71.4 617.9 61.02 568.1 47.23 544.9 40.56

585.6 49.59 572.6 57.62

658.4 72.33

1100 500

1100 500

Temperature (°c)

S 100-

E 75-(0 a.

^^L /vi

1000 600 Raman Shift (cm')

c 60" re E ro a.

1400 1000 600

Raman Shift (cm1)

^ f \J lAA ;

1000 600 Raman Shift (cm ')

1000 600 200 Raman Shift (cm"1)

NCKFMASHTO

I: bi liq pi ksp gt ilm ky q ru I: bi pi ksp gt ilm ky q ru 5: bi liq pi ksp gt ky q ru

bi liq pi ksp mu gt ky q ru >: bi pi ksp mu gt ky q ru

>NCKFMASHTO

1: bi pi ksp g ilm sil q ru 2: bi liq pi ksp g ilm sil q ru ■ 3: crd pi ksp g sp ilm sil q , 4: crd pi ksp g sp ilm sil q 5: liq crd pi ksp g sp ilm sil q

X(Fez03) = 0.01

liq pi ksp gt ilm sill q

0.2 0.3

XlFeiOa)

T(°C)

NCKFMASTO (H2O saturated)

1: bi pi mu gt ky q 2: bi liq pi mu gt ky q 3: bi liq pi gt ilm ky q 4: bi liq gtilmkyq 5: bi liq crd pi ksp gt and q 6: bi liq crd pi gt and q 7: bi liq crd pi gt ilm and q 8: bi ad pJ ksp gt ilm and q 9: bi liq crd pi ksp gt ilm and q 10: bi liq ad pi ksp gt ilm q

bi pi mu gtqru

bi pi mu gt ilm q

bi pi mu gt ilm kyq-

X(Fe2Q3) =0.0 1-

liq mu gt ky q ru

bi liq mu gt ky q ru —

bi liq pi mu gt iim q

bi liq mu t ilm ky q

bi liq pi mu gt ilm ky q

bi liq

ilm sillq

bi st pi mu gt ilm and q"

bi liq pi gt sill q

bi pi ksp mu gt sill qx

bi pi mu gt ilm and q

bi liq pi ksp gt sill q

bi liq crd tjl gt ilm q

"525 545 565 585 605 625 645 665 685 705 725

T(°C)

675 725

NCKFMAST0(H20 not saturated)

"i-1-1-1-r

bi pi ksp mu gt ky q ru

X(Fe2C>3)=0.01 -1-

bi pi mu gtkyqru

bi pi mu gtilmkyq

¿¿¡¡¡r bi liq pi ksp gt ky <

bi liq pi ksp gt ky q ru

bi liq pi ksp gt ilm ky q

liq pi ksp gt ky q ru

liq pi ksp_ gt sill q ru~

1 U /- A J bi liq pi mu gt sill q

bi liq pi ksp ""mugtsillq

bi liq pi ksp gt sill q

bi liq pi ksp gt sill q ru

bi liq pi ksp gt ilm sill q

liq pi ksp gtilm sill q-

liq crd pi ksp gt ilm sill q -

■"" bi liq crd pi ksp gtilmq

J_I ^ i

liq crd pi ksp gt ilm q

Irbistplmu gt ilmkyq 2: bi pi mu gt ilm ky q H20 3: bi pi mu gt ilm sill q H20 4: bi pi mu gt sill q H20 5: bi pi ksp gt sill q H20 6: bi pi ksp gt and q H20 7: bi crd pi ksp gt and q H20 8: bi crd pi ksp gt ilm and q M20

650 670 690 710 730 750 770

790 810 830

Highlights

1. Preservation of textural evidence for early prograde to late retrograde P-T path has been presented.

2. First attempt of P-T calculation using thermodyanamic modeling of early prograde P-T trajectory followed by the Highland Complex metasediments.

3. Peak metamorphic P-T-t conditions are around ~9 kbar, 900 0C and 540 Ma respectively

4. Implications of P-T-t evolution of metasediments in the Highland Complex of Sri Lanka.