ebsd-measured crystal preferred orientation of eclogites ... · their constituent minerals (nicolas...

Eur. J. Mineral.2016, 28, 1–14Published online

Paper presented at theXI Eclogite Conference, Dominican Republic,

2–5 February 2015

EBSD-measured crystal preferred orientation of eclogitesfrom the Sanbagawa metamorphic belt, central Shikoku, SW Japan

HAFIZ UR REHMAN1,*, DAVID MAINPRICE2, FABRICE BAROU2, HIROSHI YAMAMOTO1

AND KAZUAKI OKAMOTO3

1 Graduate School of Science and Engineering, Kagoshima University, Japan*Corresponding author, e-mail: [email protected]

2 Géosciences Montpellier, UMR CNRS 5243, Université de Montpellier, France3 Department of Earth Sciences, Faculty of Education, Saitama University, Japan

Abstract: Electron back-scattered diffraction (EBSD) maps and crystal-preferred orientation (CPO) of eclogite-facies (omphaciteand garnet) and amphibolite-facies (hornblende and actinolite) phases are reported for understanding the rheological behaviour ofcrust during subduction. Two types of eclogites from the subduction-related high-pressure/low-temperature type Sanbagawametamorphic belt, Japan, have been investigated. Type-I eclogite (sample Sb-1) is composed of garnet, omphacite, secondaryactinolite and hornblende. Type-II eclogite (samples Sb-2 and Sb-2a) are mainly composed of omphacite, garnet, and retrogradehornblende with no actinolite. Omphacite, the peak eclogite-facies phase, exhibits L-type CPO (maximum density of [001] axesparallel to and high density of {110} poles normal to the lineation) in Type-I eclogite, suggesting intra-crystalline plasticity with [001]{110} and ⟨110i{110} active slip systems, indicating a constrictive strain regime at mantle depths. Omphacite in Type-II eclogiteexhibits a similar fabric but with much weaker CPO. Using the LS-index symmetry analysis (one for the end-member L-type, zerofor the end-memberS-type, and intermediate values forLS-types), a progressive change inLS indexof 0.80 forType-I and 0.61 to0.44 forType-II eclogites is observed. These values suggest a transition from axial extension parallel to the lineation for Sb-1 and weaker CPOassociatedwith pure or simple shear for Sb-2 andSb-2a.Garnet, the second dominant phase in the eclogite-facies stage, exhibitsweak andcomplex fabric patterns in all eclogite types, behaved like rigid bodies and does not show plastic deformation. Amphibolite-facies phases(e.g., hornblende and actinolite) exhibit more than two types of CPO. Hornblende and actinolite in Type-I eclogite have a strong CPOalong [001] axes aligned parallel to the lineation, indicating homotactic crystal growth probably by the replacement of omphaciteduring the early stages of retrogression. Type-II eclogites have weak CPO in hornblende but with characteristic alignment of [001]parallel to the lineation and other poles to planes (100), (010), and {110} normal to the lineation. This fabric might have resulted froma cataclastic deformation and could be related to the late-“D1” deformation stage in the Sanbagawa metamorphic belt.

Key-words: Sanbagawa metamorphic belt; high-pressure/low-temperature eclogites; EBSD; crystal preferred orientation; CPO;omphacite; garnet; deformation; rheology.

1. Introduction

Eclogites, rocks now exposed at the Earth's surface, aregenerally interpreted as remnant mafic parts of the oceanicand/or continental crust which have once been subductedto upper mantle depths, underwent high- and ultrahigh-pressure (HP/UHP) metamorphism, and were exhumedto the Earth's surface without significant retrogression(e.g., Helmstaedt et al., 1972; MacGregor & Manton,1986). Garnet and clinopyroxene (mainly omphacite)are the two major minerals, which are volumetricallyimportant for constraining the rheology of the bulk Earth.Garnet, the main constituent of eclogite at the base ofcontinental crust, within subducted slabs and in themantle (in peridotites and as majoritic garnet), is a keymetamorphic mineral for constraining the rheology ofcrust and mantle (e.g., Karato et al., 1995;Mainprice et al.,

eschweizerbart_xxx

DOI: 10.1127/ejm/2016/0028-2574

2004; Storey & Prior, 2005). In addition, garnet maybehave as rigid bodies in the presence of omphacite, thesecond most abundant mineral in eclogites. In such cases,plastic deformation in eclogites would be accommodatedin omphacite, which is a framework-supporting mineralin medium- and high-temperature eclogites at ca. 500–750 °C, >1.5GPa (van Roermund, 1983; Godard & vanRoermund, 1995; Abalos, 1997). Therefore, the fabricof omphacite is widely used to determine rheologicalbehaviour of the eclogites during HP metamorphism.Plastic deformation of rocks is often recorded by thedevelopment of crystal-preferred orientation (CPO) intheir constituent minerals (Nicolas & Poirier, 1976;Bascou et al., 2002). The CPO symmetry patterns inlow-symmetry minerals tend to coincide with the activeslip direction and planes (e.g., Mainprice & Nicolas,1989; Mainprice et al., 2011).

0935-1221/16/0028-2574 $ 6.30© 2016 E. Schweizerbart'sche Verlagsbuchhandlung, D-70176 Stuttgart

mailto:[email protected]

http://dx.doi.org/10.1127/ejm/2016/0028-2574

[001][010]

S-type fabric

[001][010]

L-type fabric

X

Z

Y

Fig. 1. Typical omphacite CPO in naturally deformed eclogites (redrawn from Abalos, 1997). Structural coordinates of X, Yand Z representlineation, foliation and normal to the foliation, respectively. Horizontal line across the circle is along structural X.

2 H.U. Rehman et al.

The characteristic CPOs of omphacite are: (1) S-type(flattening fabric); (2) L-type (constriction fabric); (3)annealed fabric; (4) composite fabrics; and (5) misorientedfabrics (e.g., Helmstaedt et al., 1972; Godard & vanRoermund, 1995; Abalos, 1997). The most common andearlier interpretation of the omphacite CPO (Fig. 1: typicalomphacite CPO) in eclogites was correlated with theflattening (planar: S-type) and constriction (linear: L-type)fabrics (Helmstaedt et al., 1972). The S-type fabrics aremarked by a strong concentration of [010] axes normalto the foliation (Z) and [001] axes by a great-circle girdleperpendicular to [010] (e.g., Godard & van Roermund,1995; Abalos, 1997; Bascou et al., 2001, 2002).The L-type fabrics have a relatively strong [001]-axesmaximum parallel to lineation (X) and [010] axes forma girdle perpendicular to X (Helmstaedt et al., 1972).Literature data indicate that multiple dislocation-slipsystems, grain-boundary migration, and mass-transferare the main microstructural processes that control thedevelopment of CPO in omphacite (e.g., van Roermund,1983; Abalos, 1997). Later studies indicated that CPOs inomphacite are due to the dominant dislocation creep, whichgenerally activates 1/2⟨110i{110}, [001]{110} and [001](100) slip systems (e.g., van Roermund & Boland, 1981;Godard & van Roermund, 1995). Most studies report polefigureswith [001] parallel to the lineation and the (010) polenormal to the foliation (e.g., Wang et al., 2010; Cao et al.,2013), interpreted as [001] (010) slip. However it hasbeen established by Bascou et al. (2001, 2002) and Ulrich& Mainprice (2005), using viscoplastic self-consistent(VPSC) modelling, that pole figures with [001] axesparallel to the lineation and the (010) pole normal to thefoliation can be perfectly reproduced using the experimen-tally determined slip systems 1/2⟨110i{110}, [001]{110},and [001](100) (Raterron & Jaoul, 1991; Raterron et al.,1991) or observed by the transmission electron microscope(vanRoermund&Boland, 1981; vanRoermund, 1983).Weinvestigated eclogite-facies and amphibolite-facies miner-als in naturally deformed eclogites from the Sanbagawametamorphic belt, central Shikoku, SW Japan, to identifythe types of fabrics and tried to correlate the rheologicalbehaviour of these rockswith the subduction-relatedHP/LT(low-temperature) and post-eclogite-facies regional meta-morphism.

eschweizerbart_xxx

2. Geological setting

The Sanbagawa belt extends for>800 km along southwestJapan (Fig. 2). It is bounded to the north by the Ryoke belt(a Cretaceous HT/LP regional metamorphic belt) along amajor strike-slip fault known as the Median Tectonic Line(Isozaki & Maruyama, 1991; Maruyama et al., 1996) andforms the well-known paired metamorphic belts ofMiyashiro (1961). In the south, it is bounded by the Chichibu(a Jurassic accretionary complex) and the Shimanto(aCretaceous accretionary complex) belts. Itsmetamorphismis characterized by the HP/LT intermediate grade.

The belt is mainly composed of basic, quartzose, andpelitic-psammitic schists with several eclogite andultramafic bodies (Hide, 1961; Banno, 1964; Bannoet al., 1978; Enami, 1982; Takasu, 1989; Aoya, 2001;Terabayashi et al., 2005). The mafic unit, known as theIratsu mass (subdivided into Lower and Upper Iratsumass), generally shows epidote-amphibolite-facies grade,with locally preserved eclogite lenses or layers (Fig. 2).To the south, ultramafic rocks of the Higashi-Akaishimass are tectonically juxtaposed with the Iratsu mass(Terabayashi et al., 2005). The protolith of eclogites,ultramafic rocks, and surrounding schists were consideredas accreted materials from an oceanic domain includingpillow basalts (e.g., Terabayashi et al., 2005). Utsunomiyaet al. (2011) interpreted eclogites with lower SiO2 contents(45–49wt.%) to have formed from an oceanic island arcsource, and those with relatively higher SiO2 contents(53–65wt.%) to have been derived from a silica-richresidual liquid when the Highashi-Akaishi mass wasformed. However, the deformation regimes in which therocks formed and later evolved are not well constrained.

3. Deformation events in the Sanbagawa belt

Three major regional deformation events have beenrecognized in the Sanbagawa metamorphic belt (Haraet al., 1977; Faure, 1983): D1, represented by beddingschistosity and E–W trending horizontal stretching minerallineation, associated with late retrogression stage (Haraet al., 1992; Wallis, 1995); D2, the Ozu-Nagahama phase(Hara et al., 1977), represented by S-verging overturned

Mt. Higashi-Akaishi-yama

Mt. Gongen-yama

Uryuno

Seba

Upper Iratsu mass

Sample locality

Lower Iratsu mass

Higashi-Akaishi mass

Dozan-gawa

river

0 200kmNPhilippine Sea plate

Pacificplate

Eurasian plate

Ryoke belt

a North American plate

Study area

Shimanto belt

MTL

MTL Sanbagawa belt

133°23'E

33°54'N

133°26'E

133°26'Eb

1 km

N

eclogiteperidotite

ep-amphibolite

pelitic gneissfelsic gneiss

pelitic-psammiticschists

metacarbonate

basic schist

quartz schistSb-1Sb-2 & 2a

33°51'N

fault & shear zones

talus

talus

legend

Fig. 2. (a) Index map of the Sanbagawa and surrounding belts in western Japan (after Miyashiro, 1961). MTL: Median Tectonic Line.(b) Geological map of the Sanbagawa metamorphic belt (after Yamamoto et al., 2004 and references therein) (Online version in colour).

EBSD data of Sanbagawa eclogites, SW Japan 3

folds and thrust faults; and D3, the Hajikawa phase (Haraet al., 1977), represented by outcrop-scale upright openfolds with E–W trending horizontal axes. Terabayashi et al.(2005), based on detailed mapping, suggested a duplexstructure for the Sanbagawa belt where eclogites andmetagabbro–peridotites occur in the form of lenses or layerswithin the pelitic schists. The deformation history explainedabove mainly describes late-stage events in the Sanbagawaarea and no study has addressed deformation within and/orduring peak eclogite-facies metamorphism. Our studyprovides constraints on the rheology at the eclogite-faciesstage in the subduction-related HP/LT environment.

4. Sample description and petrographic features

Eclogites in the study area have been subdivided into twotypes based on field observations, petrographic features,and chemical composition. Type-I eclogite is dark green,contains abundant mafic minerals (omphacite and horn-

eschweizerbart_xxx

blende) and was named by Banno (1964) as “hornblende-eclogite”. Type-II eclogite is light green with silica-richportions and was named “quartz-bearing eclogite”.Threesamples, one from Type-I and two from Type-II eclogite,were selected for this study. Sample Sb-1 (Type I) wascollected from the Lower Iratsu mass, west of GongenYama (33°5204700 N, 133°2302600 E, see Fig. 2 for location)and Samples Sb-2 and Sb-2a (Type II) were collectedfurther southwest (33°5204000 N, 133°2302100 E, Fig. 2).

4.1. Sample Sb-1 (Type-I eclogite)

Sample Sb-1 is dark green (Fig. 3a–c) and composed ofomphacite, garnet, hornblende, actinolite, rare epidote,minor quartz, muscovite, and accessory rutile and titanite.Garnet (43 vol%) and omphacite (35 vol%) make themajor framework of this rock (Table 1). This type ofeclogite exhibits a granoblastic texture in which medium-grained garnet is embedded in clustered omphacite grains


(Fig. 3b and c). An apparent foliation, defined by omphaciteshape-preferred orientation of long axes, can be observed(Fig. 3c). Some of the grains have no retrogression whereasothers are transformed in places to hornblende (5 vol%)and actinolite (7 vol%). Garnet porphyroblasts, ranging in

eschweizerbart_xxx

2mm

2mm2mm

2mm

GrtOmp

Hbl

Omp

Zo

Act

Act

Grt

Omp

Omp

Grt

Hbl

(b)

(c)

(e)

(f )

(a) (d)

Fig. 3. Hand specimen and photomicrographs of Type I (a–c) andType II (c–e) eclogites. Most of the garnet porphyroblasts in bothtypes are heavily cracked where secondary hornblende, epidote, andchlorite have been formed at places. Apparent foliation is marked bythe elongated direction of omphacite (direction of black arrowhead)(Online version in colour).

Table 1. Modal abundance (%) of mineral phases in the measured sam

Sample Sb-1

Mineral Mode % Points (n) Mode %

Grt 43.00 549 809 37.00Omp 35.04 441 044 41.11Qz 0.01 106 0.01Hbl 4.50 5784 11.00Ms 0.09 1112 0.07Ep 0.10 1233 0.02Zo 0.63 8085 1.50Rt 0.48 6131 0.25Ttn 0.03 352 0.03Al 0.11 1402 0.07Ilm 0.00 20 –Act 7.40 94 534 –NI 9.00 114 655 9.70Total 100.38 100.76Indexed 91.38 91.06

Modal abundance for each phase was calculated from the measured ENI means not indexed.

size from 0.2 to 1mm, have numerous cracks wheresecondary hornblende, epidote and chlorite have developed.Hornblende and actinolite form composite grains thatsurround most of the garnet grains.

4.2. Sample Sb-2 (Type-II eclogite)

Sample Sb-2 is light green (Fig. 3d–f) and composed ofomphacite (41 vol%), garnet (37 vol%), secondary horn-blende (11 vol%), epidote (2 vol%), quartz, muscovite,and accessory rutile and titanite. Garnet porphyroblastshave identical features to those observed in Sb-1, butmost of the porphyroblasts are heavily fractured anddisplay pull-apart structures (Fig. 3e and f). Omphaciteoccurs in two forms, oblate to sub-rounded porphyro-blasts forming the major framework and as fine-grainedmatrix surrounding the porphyroblastic garnet andgranoblastic omphacite. Hornblende occurs as sur-rounded grains around omphacite and garnet, and somegrains fill the cracks.

4.3. Sample Sb-2a (Type-II eclogite)

Sample Sb-2a is composed of omphacite (42 vol%), garnet(38 vol%), secondary hornblende (10 vol%) and epidote(<2 vol%), less common quartz and muscovite withaccessory rutile and titanite. This sample is identicalpetrographically to Sb-2.

4.4. Pressure–temperature conditions

The P–T estimates obtained from the garnet–clinopyroxene thermometer for Type-I eclogite were1.4–2.5GPa and 500–770 °C (Ota et al., 2004).For Type-II eclogite conditions of up to 3.2GPa and920 °C were proposed by Enami & Miyamoto (2001).Ota et al. (2004) postulated that the Type-II eclogites may

ples.

Sb-2 Sb-2a

Points (n) Mode % Points (n)

125 050 38.00 139 674136 901 42.10 154 628

23 0.01 2435 913 10.00 38 598

234 0.06 22681 0.04 138

4890 1.20 4507837 0.13 48792 0.03 94238 0.07 250– – –– – –

32 565 8.60 31 679100.2391.63

BSD data.

Table 2. Details of analytical procedure and data processing.

Sample Sb-1 Sb-2 Sb-2aRock type Eclogite Quartz-bearing eclogite Quartz-bearing eclogiteThin section Circular (25mm) Circular (25mm) Circular (25mm)

Polish 0.3m (diamond paste) 0.4m (diamond paste) 0.5m (diamond paste)Finishing SYTON SYTON SYTONCoating No No NoDirection XZ XZ XZSEM JEOL JSM 5600 JEOL JSM 5600 JEOL JSM 5600Tilt (deg) 70 70 70Acc. volt. (kV) 17 17 17Working distance (mm) 25 25 25EBSD Aztec Oxford Aztec Oxford Aztec OxfordAnalysis Multiphase map Multiphase map Multiphase mapStep (mm) 14 30 25Cleaning Yes Yes YesNoise reduction (wild spike) Yes Yes YesNeighbours zero solution 5 5 5


have reached the UHP stability field, as suggested by thepresence of quartz exsolution in omphacite. However, noUHP minerals have been confirmed.

5. Analytical procedure

Three circular polished thin sections (2.5 cm in diameter)from the eclogite samples were analysed for electron back-scattered diffraction (EBSD) patterns using the scanningelectron microscope (SEM) JEOL JSM 5600 at Geo-sciences Montpellier, Université de Montpellier. Analyti-cal details for the studied samples are shown in Table 2.The samples were marked with X-direction (lineation) andZ-direction (normal to the foliation), on the basis ofapparent elongated directions of omphacite grains. In theanalysed samples diffraction patterns for garnet, ompha-cite, hornblende, actinolite, quartz, muscovite, epidote,rutile, and titanite were determined. All the data wereacquired and presented in the form of EBSD maps, andthe CPO data were plotted using the MTEX 4.1, anopen-source MATLAB toolbox for quantitative textureanalysis (Hielscher & Schaeben, 2008). Detailed analyti-cal procedure, including EBSD measurement parameters,data acquisition and processing, construction of polefigures, are given in Supplementary material (AppendixS1), freely available online at GSWwebsite of the journal,http://eurjmin.geoscienceworld.org/.The measured CPO in minerals of the peak eclogite-

facies stage and the retrograde stage are presented onequal-area, lower-hemisphere projections as pole figures(PFs), in the traditional structural frame (X, lineation; Z,foliation normal; Y, perpendicular to X and Z). Thestrength of the CPO (scale bar on PFs) is represented bycontours in multiples of uniform distribution (mud) witha value of 1 for a uniform distribution. The symmetry ofomphacite CPO is shown by the LS-index (one for the end-member L-type, zero for the end-member S-type, andintermediate values for LS-types) introduced by Ulrich &

eschweizerbart_xxx

Mainprice (2005). The PFs of all minerals werecharacterized by the texture index referred to as PF-J.In addition, the Orientation Distribution Function (ODF)was also calculated, shown as ODF-J. Misorientationalong high- and low-angle boundaries, referred to as theboundary misorientation function (BMF), for omphaciteand garnet was also constructed using the MTEX (seeAppendix S1 for details).

6. Results

6.1. Fabric of the peak eclogite-facies minerals

6.1.1. Omphacite CPO

In the three samples from two types of eclogites, theomphacite CPO (Fig. 4) are characterized by strongconcentrations of (010) poles and [001] axes, respectively,normal to the foliation (Z) and parallel to the lineation (X).The {110} poles are concentrated in a girdle perpendicularto the lineation with a maximum density close to theZ-direction. As can be seen in Fig. 4, omphacite CPO inSb-1 exhibit stronger concentrations and higher densitiescompared with those measured in Sb-2 and Sb-2a. Wecan globally describe the CPO strength by using theODF-J, where Sb-1, Sb-2, and Sb-2a have values of 2.73,1.59, and 1.79, respectively (Table 3). However, botheclogite types generally exhibit similar fabrics for theomphacite. It is interesting to note that the intensity of PFof omphacite in Sb-1 (having higher modal percentageof garnet than omphacite, see Table 1) is stronger with amaximum mud (>5) for [001] directions compared withthose in Sb-2 and Sb-2a, which have relatively lowermud (up to 3) (Fig. 4).

6.1.2. Garnet CPO

Garnet CPO from the three samples (Fig. 5) show weakfabric with complex pole figure patterns and relativelylow densities never exceeding much above 3mud. Overall,

http://eurjmin.geoscienceworld.org/

(100 ) (010 ) {110 } [001 ]

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

0.5

1

1.5

2

2.5

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Sb-1

Omphacite

(100 ) (010 ) {110 } [001 ]

0.4

0.6

0.8

1

1.2

1.4

1.6

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

0.5

1

1.5

2

2.5

3

Sb-2

(100 ) (010 ) {110 } [001 ]Max:3.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

0.6

0.8

1

1.2

1.4

1.6

0.5

1

1.5

2

2.5

3

Sb-2a

Max:2.32

Min:0.31

Min:0.11

Min:0.22

Min:0.13

Min:0.53

Min:0.36

Min:0.51

Min:0.27

Min:0.32

Min:0.36

Min:0.47

Min:0.22

Max:2.83

Max:2.16

Max:5.09

Max:1.71

Max:2.40

Max:1.60

Max:3.06

Max:1.88

Max:2.36

Max:1.73

Max:3.15

PF texture index = 1.23 PF texture index = 1.47 PF texture index = 1.28 PF texture index = 2.17



LS-index = 0.80

LS-index = 0.61

LS-index = 0.44

X

Z

Fig. 4. Omphacite CPO from the analyzed samples shown in equal area projections, lower hemisphere. Foliation (XY plane; full line) isvertical, lineation (X) is horizontal in this plane, and Z is normal to foliation. Contours in multiples of uniform distribution (mud).


Sb-1 has lower maximum pole figures densities for {100},{110}, and {111} poles than Sb-2 and Sb-2a. Generally,the maximum densities are not aligned with structuraldirections, such as the lineation X, pole to the foliation Z,or the foliation plane XY. The major exception is Sb-2a,which has the maximum density of 2.4mud of {110} polesexactly parallel to Z and a maximum density of 2.3mudin ⟨111i directions in the XY plane, about 10° from X.The garnet CPO strength is given by ODF-J with valuesof 1.51, 1.97, and 2.05 for Sb-1, Sb-2, and Sb-2a,respectively (Table 3). The results indicate that Sb-2ahas strongest CPO but the difference between samples isvery small.Garnet in the Sanbagawa Type-I eclogite displays

relatively weak fabric having maximum densities of2.6 to 1.8mud for ⟨111i directions and {110} poles,respectively. Garnet in Type-II eclogite have slightly

eschweizerbart_xxx

higher maximum densities of 3.3 to 2.3mud for ⟨111idirections and {110} poles, respectively. Only in SampleSb-2a, the ⟨111i directions are in the foliation (XY) planewith the maximum density (2.3mud) and the {110} polesmaximum are exactly parallel to Z.

6.2. Fabric of the post-eclogite facies minerals

Hornblende CPO (Fig. 6) shows that point-maximum for[001] coincides with the point-maximum for [001] inomphacite (Fig. 4). The CPO have a strong alignment of[001] parallel to structural X-direction (4mud in Sb-1),with other poles (100), (010), and {110} normal to X withmaxima near the XYplane. Samples Sb-2 and Sb-2a havesignificantly weaker CPO, with a maximum [001] of 1.90and 1.10mud, respectively. In the Type-II eclogite, thehornblende CPO are very similar, exhibiting partial girdles

eschweizerbart_xxx

Table 3. Parameters of the CPO for all minerals discussed in thetext.

Omphacite

Pole figures min max PF-J ODF-J LS-index

Sb-1(100) 0.31 2.32 1.23 2.73 0.80(010) 0.11 2.83 1.47(110) 0.22 2.16 1.28[001] 0.13 5.09 2.17Sb-2(100) 0.53 1.71 1.10 1.59 0.61(010) 0.47 2.40 1.15(110) 0.51 1.60 1.09[001] 0.27 3.06 1.37Sb-2a(100) 0.32 1.88 1.16 1.79 0.44(010) 0.36 2.36 1.23(110) 0.47 1.73 1.11[001] 0.22 3.15 1.42

Garnet

Pole figures min max PF-J ODF-J

Sb-1(100) 0.34 2.61 1.28 1.51(110) 0.48 1.77 1.06[111] 0.19 2.12 1.12Sb-2(100) 0.19 3.04 1.25 1.97(110) 0.39 2.77 1.15[111] 0.21 3.35 1.24Sb-2a(100) 0.12 3.03 1.39 2.05(110) 0.38 2.41 1.09[111] 0.16 2.31 1.20

Actinolite


Sb-1(100) 0.27 3.22 1.35 2.78(010) 0.21 2.98 1.45(110) 0.30 2.14 1.21[001] 0.15 4.15 1.84

Hornblende


Sb-1(100) 0.13 3.12 1.34 3.19(010) 0.17 3.29 1.36(110) 0.26 2.60 1.26[001] 0.09 4.00 1.78Sb-2(100) 0.61 1.55 1.06 1.25(010) 0.63 1.26 1.02(110) 0.69 1.57 1.04[001] 0.42 1.90 1.10Sb-2a(100) 0.61 1.88 1.05 1.30(010) 0.57 1.60 1.04(110) 0.67 1.47 1.03[001] 0.36 1.74 1.10


for [001] near the XYplane and have the maximum densityparallel to X and other poles (100), (010) and {110}maximum near the foliation normal Z. Note that Sb-2 andSb-2a have significantly weaker and slightly differentCPO for omphacite than Sb-1.

Actinolite is only present in Type-I eclogite (Sb-1),exhibits similar CPO (Fig. 7) to those of omphacite witha maximum [001] of 4.15mud, and point maximum for(010) normal to foliation. The actinolite {110} poles havea maximum normal to the lineation X and the pole to thefoliation Z whereas omphacite (Fig. 4) {110} poles have amaximum parallel to Z.

7. Discussion

7.1. Analysis of boundary misorientation functions

Crystal preferred orientations in omphacite generallyindicate the strain regime of a deformation event(Helmstaedt et al., 1972; van Roermund & Boland,1981; van Roermund, 1983; Godard & van Roermund,1995; Abalos, 1997; Bascou et al., 2001, 2002; Mauleret al., 2001; Zhang & Green, 2007). To identify the strainregime and rheological behaviour of omphacite in theSanbagawa eclogites, which have very coherent and strongCPO patterns in all three samples, we discuss the boundarymisorientation. The BMF are shown for low angles(v= 5°, 10°, or 15°), as high-angle grain boundaries aregenerally accepted to between 10° and 15°, and hencedislocation sub-grain boundaries have lower angles(Fig. 8). The low-angle boundary misorientation axesfor pure tilt and twist boundaries are generally associatedwith slip systems. For a pure tilt boundary themisorientation axis is given by the cross product for thetilt-axis = [uvw]� (hkl), where [uvw] is Burgers vectordirection of the edge dislocations in tilt-boundary and (hkl)is the slip plane. For a pure twist boundary themisorientation axis was given by the cross product withthe twist-axis = [u1v1w1]� [u2v2w2], where [u1v1w1] and[u2v2w2] are the two screw dislocation Burgers vectordirections in the twist boundary and twist-axis is normal totwist boundary. To interpret the BMF sections at selectedangles we used the following slip systems: [001](100),[001](110), ½001�ð110Þ; ½110�ð110Þ; ½110�ð110Þ, and [100](010), known to be the dominant slip systems in TEMstudies (van Roermund & Boland, 1981; van Roermund,1983; Godard & van Roermund, 1995) or having shownhigh activity in the VPSC simulations (Bascou et al., 2001,2002; Ulrich & Mainprice, 2005). We found four tiltboundary misorientation axes M1 to M4, which aredirections: M1 [010] for [001](100), M2 [221] for [001](110) and [001](1–10), M3 [001] for ½110�ð110Þ and½110�ð110Þ, and M4 [106] for [100](010). For pure twistboundaries, we found six possibilities with misorientationaxes R1 to R6, which are poles to twist boundary planes;R1 (110) for [001] and ½110�, R2 ð110Þ for [001] and[110], R3 (010) for [001] and [100], R4 (001) for ½110�and [110], R5 (001) for ½110� and [100], and R6 (001) for[100] and [110].

Sb-1

Almandine

Sb-2

Sb-2a

PF texture index = 1.28 PF texture index = 1.06 PF texture index = 1.12



0.40.60.811.21.41.61.822.22.4

0.6

0.8

1

1.2

1.4

1.6

0.20.40.60.811.21.41.61.82

{100 } {110}

0.5

1

1.5

2

2.5

0.40.60.811.21.41.61.822.2

0.20.40.60.811.21.41.61.822.2

Max:2.12

0.5

1

1.5

2

2.5

3

0.5

1

1.5

2

2.5

0.5

1

1.5

2

2.5

3

Min:0.19

Max:1.77

Min:0.48

Max:2.61

Min:0.34

Max:3.04

Min:0.19

Max:2.77

Min:0.39

Max:3.35

Min:0.21

Max:3.03

Min:0.12

Max:2.41

Min:0.38

Max:2.31

Min:0.16

{100 } {110}

{100 } {110} <111>X

Z

<111>

<111>

Fig. 5. Garnet CPO from the analyzed samples shown in equal area projections, lower hemisphere. Foliation (XYplane; full line) is vertical,lineation (X) is horizontal in this plane, and Z is normal to foliation. Contours in mud.


InFig. 8, theBMFsectionshave thesemisorientationaxesmarked for omphacite. First, we remark that the BMFsections for Sb-1 have a much higher maximum density(>14mud) than in Sb-2 and Sb-2a (3mud). For sample Sb-1, at a misorientation of 5°, there are maxima up to 3mudnear the poles (221), (110), and ð110Þ, all characteristic of[001]{110} slip. There is also a maximum near the [001]axis, poles (001), and (106), characteristic of [100](010) and⟨110i{110} slip; the latter is a high-temperature slip system.At v =10°, there is a strong peak (10mud) near pole (221),characteristic of [001]{110}slip.Atv =15°, there is a strongpeak (14mud) near [001], characteristic of ⟨110i{110} slip.The highest misorientation densities (10–14mud) in Type-Ieclogite (Sb-1) can be associated with twist boundaries(221)and tiltmisorientationaxis [001],which indicate [001]{110} and ⟨110i{110} slip systems, respectively.Samples Sb-2 and Sb-2a have similar patterns of

misorientation, but slightly different from Sb-1. At v = 5°,the only significant peak is near [010] of 3.5mud,

eschweizerbart_xxx

characteristic of [001](100). At v = 10°, the peak near[010] is still present, with a weak maximum >2mud near{110} poles, characteristic of the activity of [001]and ⟨110i slip directions, forming twist boundaries. Atv = 15°, the peaks are less pronounced than at 5° and 10°,and overall the pattern is weak, except to a new peak at[001] of >2mud, which could be produced by the tiltboundary formed by ⟨110i{110} slip. The Type-IIeclogites have a strong maximum of >3mud for v = 5°and v = 10°, and a tilt misorientation axis of [010],characteristic of pure tilt boundary formed by [001](100)slip. The densities of >2mud near [001] in Sb-2a suggestsmall activity of ⟨110i{110}, but this is absent in Sb-2.

Garnet, the first or second most abundant mineral ineclogite, also provides information on the strain regime,especially on the plastic deformation which usually occursat high temperature (Ando et al., 1993; Mainprice et al.,2004). Plasticity in garnet in the studied samples wouldonly be expected during the peak eclogite-facies

eschweizerbart_xxx

Sb-1

Hornblende

Sb-2

Sb-2a

Max:3.12

Min:0.13

Min:0.17

Min:0.69

Min:0.09

Min:0.61

Min:0.63

Min:0.69

Min:0.42

Min:0.61

Min:0.57

Min:0.67

Min:0.36

Max:3.29

Max:2.60

Max:4.00

Max:1.55

Max:1.26

Max:1.57

Max:1.90

Max:1.88

Max:1.60

Max:1.47

Max:1.10




(100 ) (010 ) {110 } [001 ]

0.5

1

1.5

2

2.5

3

0.5

1

1.5

2

2.5

3

0.5

1

1.5

2

2.5

0.5

1

1.5

2

2.5

3

3.5

(100 ) (010 ) {110 } [001 ]

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

0.7

0.8

0.9

1

1.1

1.2

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

0.6

0.8

1

1.2

1.4

1.6

1.8

(100 ) (010 ) {110 } [001 ]

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

1.6

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

0.4

0.6

0.8

1

1.2

1.4

1.6

X

Z

Fig. 6. Hornblende CPO from the analyzed samples shown in equal area projections, lower hemisphere. Foliation (XY plane; full line) isvertical, lineation (X) is horizontal in this plane, and Z is normal to foliation. Contours in mud.

Sb-1

ActinoliteMax:3.22

Min:0.27

Min:0.21

Min:0.30

Min:0.15

Max:2.98

Max:2.14

Max:4.15


(100 ) (010 ) {110 } [001 ]

0.5

1

1.5

2

2.5

3

0.5

1

1.5

2

2.5

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0.5

1

1.5

2

2.5

3

3.5

4X

Z

Fig. 7. Actinolite CPO from sample Sb-1 shown in equal area projections, lower hemisphere. Foliation (XY plane; full line) is vertical,lineation (X) is horizontal in this plane, and Z is normal to foliation. Contours in mud.


ω = 5°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106)

ω = 10°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106)

ω = 15°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106)0.5

1

1.5

2

2.5

3

Sb-1

Omphacite - Boundary Misorientation Function Sections

Sb-2

Sb-2a

ω = 5°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106)

ω = 10°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106)

ω = 15°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106) 1

1.5

2

2.5

3

3.5

ω = 5°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106)

ω = 10°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106)

ω = 15°

(100)

[010]

[001] (001)

(110)( 1̄10)

(221)

(106) 2

4

6

8

10

12

14

Fig. 8. Omphacite boundary misorientation functions for angles 5°, 10° and 15°. Reference directions (221), (106), [001], (001), [010], (110)are referred to in the text (Online version in colour).


metamorphism. We examined the misorientation axes,which are produced by the ⟨111i{110} slip system ingarnet. For pure tilt dislocation boundary the tilt-axis is½112� ¼ ½111� � ð110Þ, and for pure twist dislocationboundary the twist-axis is ½112� ¼ ½111� � ð110Þ, where[112] and [101] are directions within the fundamental unittriangle for cubic ½m3m� symmetry. Figure 9 shows theBMF sections for garnet. The frequency of tilt-boundariesin cubic polycrystals, having random boundaries with anallowed deviation from pure tilt geometry of 1°, exhibit39.0% tilt boundaries. For a tolerance of 5°, the tilt can beup to 98.6% of all boundaries (Morawiec, 2010). Hence,we can expect near tilt boundaries to be well representedeven in the case of random boundary distribution. In Fig. 9,we find no indication of peak at [112] or [101]. All sectionsfor Sb-1, Sb-2, and Sb-2a at 5°, 10°, and 15° show a broadmaximum of about 22mud at the centre of the diagram andlow values (<5mud) for [112] or [101]. In the studiedsamples, it is hard to envisage a dominant slip system of1/2⟨111i(110) in garnet as is usually reported fromexperimentally deformed garnets (e.g., Ando et al., 1993;Karato et al., 1995). The volume fractions of garnet in Sb-1, Sb-2, and Sb-2a eclogites are 43, 37, and 38 vol%,respectively. In Sb-1, garnet is the most abundant mineraland omphacite is the second. In contrast, in Sb-2,omphacite is the most abundant and garnet second. Thismeans garnet and omphacite were the main load-bearingphases in both eclogite types. However, the preferredorientation of omphacite (Fig. 4) is much stronger than

eschweizerbart_xxx

garnet (Fig. 5), the fabric pattern of garnet being differentfor the three samples, whereas omphacite in all threesamples has an almost identical fabric pattern. Theprincipal glide direction and plane in garnet are ⟨111i and{110}; they tend to align with the shear direction andplane, respectively (Mainprice et al., 2004). Only theSb-2a (Fig. 5) has the {110} poles with maximum density(mud>2.4) at the expected position of Z normal to the XYfoliation, and the ⟨111i directions show moderate con-centration (mud >2.14) at the expected position of X withmaximum concentration in foliation 10° from X of2.3mud. However the misorientation analysis in Fig. 9shows no clear evidence for ⟨111i{110} slip in any of thethree samples.

Two observations suggest that there are some differ-ences between Type-II eclogites Sb-2 and Sb-2a, despitethe fact that they come from the same outcrop. First, thegarnet CPO are slightly different, Sb-2a has highdensities of ⟨111i and {110} parallel to X and Z,respectively, whereas Sb-2 does not. Second, theomphacite misorientation plots (Fig. 8) show thepresence of the [001] axis in Sb-2a, which is totallyabsent in Sb-2. Both these observations suggest somedegree of heterogeneity in the samples. The observationthat Type-I eclogite has a weaker CPO in garnet thanthose of Type II is probably a more robust result.Omphacite CPO in Type-I eclogite (Sb-1) has an ODF-Jof 2.73mud (Table 3), which is higher than the ODF-J forType II (Sb-2: 1.59, and Sb-2a: 1.79), suggesting that

Sb-1

Sb-2

Sb-2a

ω = 5°

[101]

[111]

[001]

[112]

ω = 10°

[101]

[111]

[001]

[112]

ω = 15°

[101]

[111]

[001]

[112]

5

10

15

20

25

ω = 5°

[101]

[111]

[001]

[112]

ω = 10°

[101]

[111]

[001]

[112]

ω = 15°

[101]

[111]

[001]

[112]

5

10

15

20

25

ω = 5°

[101]

[111]

[001]

[112]

ω = 10°

[101]

[111]

[001]

[112]

ω = 15°

[101]

[111]

[001]

[112]

5

10

15

20

25

Almandine - Boundary Misorientation Function Sections

Fig. 9. Garnet (almandine) boundary misorientation functions for angles 5°, 10° and 15°. Reference directions [112], and [101] are referred toin the text (Online version in colour).


high values of ODF-J of omphacite are correlated withlow values of ODF-J of garnet. It seems reasonable toconclude that the stress level is significantly above theplastic yield stress for the plastic deformation ofomphacite, resulting in the development of strongCPO. In garnet, the stress level is at or below the plasticyield stress, resulting in weak to generally incoherentCPO patterns with respect to the lineation and foliation.The misorientation analysis gives a clear indication thatomphacite in Type-I eclogite deformed by [001]{110}and ⟨110i{110} slip which are generally considered to behigh-temperature systems (Raterron & Jaoul, 1991;Raterron et al., 1991). The misorientation analysis forType-II eclogite gives clear [001](100) slip in bothsamples, but only small activity ⟨110i{110} in Sb-2a, and

eschweizerbart_xxx

none in Sb-2. It would appear that the deformation inomphacite in Type-II eclogite was characteristic of lowertemperature, as the high-temperature ⟨110i{110} slip isless active than in Type I.

7.2. Deformation regime of the eclogite-facies stage

In the Sanbagawa metamorphic belt several studiesprovide fabric data, e.g., garnet and clinopyroxene CPOfrom the Higashi-Akaishi peridotite body (Muramotoet al., 2011), inclusion-trail geometry of albite in basicschists (Okamoto, 1998), and misorientation of garnetaggregates in vein in the mafic schists (Okamoto &Michibayashi, 2006). However, there is no information onthe internal fabric of the Sanbagawa eclogites. This


research provides the first report on the fabric of eclogite-facies and retrograde-stage minerals as well as usefulinsights into the rheological mineral behaviour duringeclogite formation and exhumation.Our results show that no coherent garnet CPO patterns

occur in the three samples from the Sanbagawa eclogitesand all the garnet fabrics are weak. Further, there is nosupporting evidence for activation of the ⟨111i{110} slipsystem in garnet. Thus garnet crystals likely behaved asrigid bodies and are not significantly plastically deformed.Omphacite shows clear evidence of plastic deformationwith elongated grain shapes, strong fabrics, and highdensities of misorientation in BMF sections. Hence mostof the plastic strain in the rock was accommodated byomphacite. The fabric symmetry of omphacite shows anevolution of LS-Index from 0.80, 0.61, and 0.44 insamples Sb-1, Sb-2, and Sb-2a, respectively. Thisobservation suggests a change from axial extensionparallel to X for Type-I to either pure or simple shearfor Type-II eclogites. There is also a strong contrastbetween Type-I (Sb-1 ODF-J of 2.73 for Sb-1), and Type-II eclogites (ODF-J of 1.59 for Sb-2 and ODF-J of 1.79 forSb-2a). The strength of CPO and BMF sections is muchhigher for Type-I than for Type-II eclogites. Themisorientation analysis shows that evidence is strongerfor the high-temperature slip systems ⟨110i{110} and[001]{110} slip for Type I. In Sb-2 and Sb-2a [001](100)slip appears more dominant, indicating that Type-IIeclogites might have been deformed at relatively lowertemperature than Type I.

7.3. Deformation of the late-stage phases

The CPO patterns in hornblende and actinolite were usedto define the rheology in minerals during the retrogradestage. As shown in Figs. 6 and 7 , both of these late-stagephases exhibit identical CPO (i.e., displaying maximumdensity for [001] parallel to X and girdle for (001) and(110) subnormal to foliation) patterns to those found inomphacite. These features suggest homotactic growth ofamphibolite-facies phases (actinolite and hornblende) atthe expense of omphacite. Similar results were reportedearlier by Engels (1972), and more recently described byXu et al. (2015). However, hornblende in Type-II eclogitesdisplay distinct strong girdles aligned along [001]. Similarpatterns were recently observed by Ko & Jung (2015) as“type III” from their experimental work on hornblende.They interpreted that the three dominant types of CPO inhornblende form under different temperature and stressconditions (for details see Ko & Jung, 2015). Grain-sizereduction or cataclastic flow (due to faulting) wasconsidered a major factor for this type of deformationin hornblende. We interpret the hornblende CPO in Type-Ieclogite with maximum for [001] axes to representhomotactic growth during the early stage of retrogression.In contrast, the CPO in Type-II eclogites, exhibiting astrong girdle with the maximum of [001] axes along X,could have developed by cataclastic deformation related tothe E-W-trending horizontal stretching mineral lineation

eschweizerbart_xxx

associated with the development of schistosity during theD1 stage of deformation (Hara et al., 1977) in theSanbagawa metamorphic belt. Our interpretations aresupported by the textural features; including developmentof abundant cracks in major phases and the cataclastictexture, which is severe in Type-II eclogites and lesspronounced in Type I.

7.4. Types of CPO associated with the deformationregime

There is general agreement that L-type fabric in omphacitedevelops in the constriction strain regime (Helmstaedtet al., 1972; Abalos, 1997). Therefore, we conclude thatthe L-type fabric in omphacite in the Sanbagawa Type-Ieclogites (Sb-1) could be related to constriction, parallel tothe lineation with LS-index = 0.80. Type-II eclogites (Sb-2and Sb-2a) have LS-indices of 0.61 and 0.44 that are morecompatible with pure or simple shear (see Ulrich &Mainprice, 2005). The fabric indicating a dominant slipdirection along [001] was probably activated within thesubduction-related high-stress regime. These results areconsistent with the fabric developed in the Higashi-Akaishi peridotite massif (Muramoto et al., 2011). Thisevidence clearly indicates that the Sanbagawa eclogitesexperienced the same deformational event as that observedin the Higashi-Akaishi peridotite massif when theselithologies were in the mantle-wedge. The origin forType-II eclogites was regarded as volcaniclastic materialassociated with the trench-filled sediments (Okamoto et al.,2004).However, thenearly identicalCPOwith aprogressionfrom L- to LS-type indicates that both eclogite types weredeformed at more or less similar depths under a constrictionstress regime regardless of what their protolith was.

8. Conclusions

The CPOs in the Sanbagawa eclogites most likelydeveloped during the peak eclogite-facies metamorphism.The CPO pattern in omphacite in Type-I eclogite is L-typeand progressively transforms to LS-type in Type-IIeclogites, indicating constriction strain regime for Sb-1,and pure or simple shear for Sb-2 and Sb-2a. Garnetbehaved as rigid bodies and shows no significant plasticdeformation. Most of the strain was apparently accom-modated by omphacite. The CPOs of retrograde-stageminerals (e.g., hornblende and actinolite) indicate at leasttwo stages of their formation. Those similar to thatobserved in omphacite with maximum density of [001]axes aligned sub-parallel to lineation suggest that theirdevelopment was due to homotactic growth by thereplacement of omphacite. In contrast, those showing astrong girdle along [001], suggest late-stage deformation,probably related to the D1 stage reported for theSanbagawa metamorphic belt. These results are consistentwith the petrographic observations where sub-roundedgrains of hornblende are found around garnet andomphacite porphyroblasts.


Acknowledgements:We highly appreciate critical reviewand constructive comments by Bill McClelland, GastonGodard, and an anonymous reviewer, which greatlyimproved the manuscript. The authors are indebted toChristophe Nevado and Doriane Delmas for preparingcarefully polished thin-sections for the EBSD analysis.The Geosciences EBSD-SEM national facility in Mont-pellier is supported by the Institut National de Sciencesde l'Univers (INSU) du Centre National de la RechercheScientifique (CNRS, France). The research was alsosupported by the Kagoshima University's YoungResearchers Visiting Program and partly by the Kakenhito HUR (Kiban C: #15K05316).

References

Abalos, B. (1997): Omphacite fabric variations in the Cabo Ortegaleclogite (NW Spain): relationship with strain symmetry duringhigh-pressure deformation. J. Struct. Geol., 19, 621–637.

Ando, J.J., Fujino, K., Takeshita, T. (1993): Dislocation micro-structures in naturally deformed silicate garnets. Phys. EarthPlanet. Int., 80, 105–116.

Aoya, M. (2001): P–T–D path of eclogite from of the Sambagawabelt deduced from combination of petrological and microstruc-tural analyses. J. Petrol., 42, 1225–1248.

Banno, S. (1964): Petrologic studies of the Sanbagawa crystallineschists, in the Bessi-Ino district, central Shikoku, Japan. TokyoUniv. J. Fac. Sci., 15, 203–319.

Banno, S., Higashino, T., Otsuki, M., Itaya, T., Nakajima, T. (1978):Thermal structure of the Sanbagawa belt in central Shikoku. J.Phys. Earth, 26, 345–356.

Bascou, J., Barruol, G., Vauchez, A., Mainprice, D., Egydio-Silva,M. (2001): EBSD-measured lattice-preferred orientations andseismic properties of eclogites. Tectonophysics, 342, 61–80.

Bascou, J., Tommasi, A., Mainprice, D. (2002): Plastic deformationand development of clinopyroxene lattice-preferred orientationin eclogites. J. Struct. Geol., 24, 1357–1368.

Cao, Y., Jung, H., Song, S. (2013): Petro-fabrics and seismicproperties of blueschist and eclogite in the North Qilian suturezone, NW China: Implications for the low-velocity upper layerin subducting slab, trench-parallel seismic anisotropy andeclogite detectability in the subduction zone. J. Geophys. Res.,118, 1–22.

Engels, J.S. (1972): The catazonal polymetamorphic rocks of CaboOrtegal (NW Spain), structural and petrofabric study. LeidseGeologische mededelingen (Ph.D. thesis).

Enami, M. (1982): Oligoclase–biotite zone of the Sanbagawametamorphic terrain in the Bessi district, central Shikoku,Japan. J. Geol. Soc. Jpn., 88, 887–900 (in Japanese, withEnglish abstract).

Enami, M. & Miyamoto, A. (2001): Revised thermobarometry ofkyanite–eclogite in the Sanbagawa metamorphic belt of theBessi area, Japan. In Abstracts for Annual Meeting of theJapanese Association of Mineralogy Petrology and EconomicGeology, 207 (in Japanese).

Faure, M. (1983): Eastward ductile shear during the earlytectonic phase in the Sanbagawa belt. J. Geol. Soc. Jpn., 89,319–329.

Godard, G. & van Roermund, H.L.M. (1995): Deformation-inducedclinopyroxene from eclogites. J. Struct. Geol., 17, 1425–1443.

eschweizerbart_xxx

Hara, I., Hide, K., Takeda, K., Tsukuda, E., Tokuda, M., Shiota, T.(1977): Tectonic movement in the Sambagawa belt. in “TheSambagawa belt”, K. Hide, ed., Hiroshima University Press,Hiroshima, p. 309–390.

Hara, I., Shiota, T., Hide, K., Kanai, K., Goto, M., Seki, S., Kaikiri,K., Takeda, K., Hayasaka, Y., Miyamoto, T., Sakurai, Y.,Ohtomo, Y. (1992): Tectonic evolution of the Sambagawaschists and its implications in convergent margin processes. J.Sci. Hiroshima Univ. Ser. C, 9, 495–595.

Helmstaedt, H., Anderson, O.L., Gavasci, A.T. (1972): Petrofabricstudies of eclogite, spinel-websterite, and spinel-lherzolitexenoliths from kimberlite-bearing breccia pipes in southeasternUtah and northeasternArizona. J. Geophys. Res., 77, 4350–4365.

Hide, K. (1961): Geologic structure and metamorphism of theSanbagawa crystalline schists of the Bessi-Shirataki miningdistrict in Shikoku, southwest Japan. Geol. Rep. HiroshimaUniv., 9, 1–87 (in Japanese with English abstract).

Hielscher, R. & Schaeben, H. (2008): A novel pole figure inversionmethod: specification of the MTEX algorithm. J. Appl.Crystallogr., 41, 1024–1037.

Isozaki, Y. & Maruyama, S. (1991): Studies on orogeny based onplate tectonics in Japan and new geotectonic subdivision of theJapanese Islands. J. Geogr., 100, 697–761 (in Japanese, withEnglish abstract).

Karato, S.,Wang, Z.C., Liu, B., Fujino,K. (1995): Plastic deformationof garnets: systematic and implications for the rheology of themantle transition zone. Earth Planet. Sci. Lett., 130, 13–30.

Ko, B. & Jung, H. (2015): Crystal preferred orientation of anamphibole experimentally deformed by simple shear. Nat.Commun., 6, 7586.

MacGregor, I. & Manton, W.I. (1986): Roberts Victor eclogites:ancient oceanic crust. J. Geophys. Res., 91, 14063–14079.

Mainprice, D. & Nicolas, A. (1989): Development of shape andlattice preferred orientations: application to the seismicanisotropy of the lower crust. J. Struct. Geol., 11, 175–189.

Mainprice, D., Bascou, J., Cordier, P., Tommasi, A. (2004): Crystalpreferred orientations of garnet: comparison between numericalsimulations and electron back-scattered diffraction (EBSD)measurements in naturally deformed eclogites. J. Struct. Geol.,26, 2089–2102.

Mainprice, D., Hielscher, R., Schaeben, H. (2011): Calculatinganisotropic physical properties from texture data using theMTEX open source package. in “Deformation Mechanisms,Rheology and Tectonics: Microstructures: Mechanics andAnisotropy”, D.J. Prior, E.H. Rutter, D.J. Tatham, eds., Geol.Soc. Lond. Spec. Publ., 360, 175–192.

Maruyama, S., Liou, J.G., Terabayashi, M. (1996): Blueschists andeclogites of the world and their exhumation. Int. Geol. Rev., 38,485–594.

Mauler, A., Godard, G., Kunze, K. (2001): Crystallographicfabrics of omphacite, rutile and quartz in Vendée eclogites(Armorican Massif, France): consequences for deformationmechanisms and regimes. Tectonophysics, 342, 81–112.

Miyashiro, A. (1961): Evolution of metamorphic belts. J. Petrol., 2,277–311.

Morawiec, A. (2010): Frequencies of tilt and twist boundaries amongrandom grain boundaries. Solid Sate Phenom., 160, 95–99.

Muramoto, M., Michibayashi, K., Ando, J., Kagi, H. (2011):Rheological contrast between garnet and clinopyroxene in themantle wedge: An example from Higashi-akaishi peridotitemass, SW Japan. Phys. Earth Planet. Int., 184, 14–33.


Nicolas, A. & Poirier, J.P. (1976): Crystalline plasticity and solidstate flow in metamorphic rocks. Wiley, New York.

Okamoto, A. & Michibayashi, K. (2006): Misorientations of garnetaggregate within a vein: an example from the Sanbagawametamorphic belt, Japan. J. Metamorph. Geol., 24, 353–366.

Okamoto, K. (1998): Inclusion-trail geometry of albite porphyro-blasts in a fold structure in the Sambagawa belt, central Shikoku,Japan. Island Arc, 7, 283–294.

Okamoto, K., Shinjoe, H., Katayama, I., Terada, K., Sano, Y. (2004):SHRIMP U–Pb dating of quartz-bearing eclogite from theSanbagawa belt, south-west Japan: implications formetamorphicevolution of subducted protolith. Terra Nova, 16, 81–89.

Ota, T., Terabayashi, M., Katayama, I. (2004): Thermobaricstructure and metamorphic evolution of the Iratsu eclogitebody in the Sanbagawa belt, central Shikoku, Japan. Lithos, 73,95–126.

Raterron, P. & Jaoul, O. (1991): High temperature deformation ofdiopside single crystal: 1, mechanical data. J. Geophys. Res., 96,14277–14286.

Raterron, P., Doukhan, N., Jaoul, O., Doukhan, J.C. (1991): Hightemperature deformation of diopside. IV: predominance of{110} glide above 1000 °C. Phys. Earth Planet. Int., 82,209–222.

Storey, C.D. & Prior, D.J. (2005): Plastic deformation andrecrystallization of garnet: a mechanism to facilitate diffusioncreep. J. Petrol., 46, 2593–2613.

Takasu, A. (1989): P–T histories of peridotite and amphibolitetectonic blocks in the Sambagawa metamorphic belt, Japan. in“The evolution of metamorphic belts”, J.S. Daly, R.A. Cliff, B.W.D. Yardley, eds., Geol. Soc. Lond. Spec. Publ., 43, 533–538.

Terabayashi, M., Okamoto, K., Yamamoto, H., Kaneko, Y., Ota, T.,Maruyama, M., Katayama, I., Komiya, T., Ishikawa, A., Anma,R., Ozawa, H., Windley, B.F., Liou, J.G. (2005): Accretionarycomplex origin of the mafic–ultramafic bodies of the Sanbagawabelt, central Shikoku, Japan. Int. Geol. Rev., 47, 1058–1073.

eschweizerbart_xxx

Ulrich, S. & Mainprice, D. (2005): Does cation ordering inomphacite influence development of lattice-preferred orienta-tion? J. Struct. Geol., 27, 419–431.

Utsunomiya, A., Jahn, B.M., Okamoto, K., Ota, T., Shinjoe, H.(2011): Intra-oceanic island arc origin for Iratsu eclogites of theSanbagawa belt, central Shikoku, southwest Japan. Chem.Geol., 280, 97–114.

van Roermund, H.L.M. (1983): Petrofabrics and microstructures ofomphacites in a high temperature eclogite from the SwedishCaledonides. Bull. Minéral., 106, 709–713.

van Roermund, H.L.M. & Boland, J.N. (1981): The dislocationsubstructures of natural deformed Omphacite. Tectonophysics,78, 403–418.

Wallis, S.R. (1995): Vorticity analysis and recognition of ductileextension in the Sanbagawa metamorphic belt, SW Japan.J. Struct. Geol., 17, 1077–1093.

Wang, L., Jin, Z.M., Kusky, T., Xu, H.J., Liu, X.W. (2010):Microfabric characteristics and rheological significance of ultra-high pressure metamorphosed jadeite-quartzite and eclogitefrom Shuanghe, Dabie Mountains, China. J. Metamorph. Geol.,28, 163–182.

Xu, H., Zhang, J., Zong, K., Liu, L. (2015): Quartz exsolutiontopotaxy in clinopyroxene from the UHP eclogite of Weihai,China. Lithos, 226, 17–30.

Yamamoto, H., Okamoto, K., Kaneko, Y., Terabayashi, M. (2004):Southward extrusion of eclogite-bearing mafic–ultramaficbodies in the Sanbagawa belt, central Shikoku, Japan.Tectonophysics, 387, 151–168.

Zhang, J. & Green, H.W. (2007): Experimental investigation ofeclogite rheology and its fabrics at high temperature andpressure. J. Metamorph. Geol., 25, 97–115.

Received 29 October 2015Modified version received 29 April 2016Accepted 7 June 2016

ebsd-measured crystal preferred orientation of eclogites ... · their constituent minerals (nicolas...

Documents