Search

Serpentinites (frequently cross cut by gabbroic dikelet), collected from Northern Central Indian Ridge (NCIR), Indian Ocean, contain both Cr-rich (Group I) and Cr-poor (Group II) variety of spinel. Based on mineralogy they can be classified as non-residual and residual spinel, respectively. While, non residual spinel (TiO2 up to 0.5 wt %; Group I) display the evidences of peridotite-gabbro interaction, residual spinel (TiO2

The Ronda peridotite massif of southern Spain exposes subcontinental lithospheric mantle that records pressure–temperature data and microstructures formed during exhumation beneath the rapidly extending Alboran domain. The peridotite is zoned from garnet- and spinel-bearing mylonites at the structural top, to spinel-bearing tectonites, to melt-percolated spinel-bearing granular peridotites, to plagioclase-bearing tectonites at the structural base. We find microstructural evidence of melt present in the spinel zones prior to the deformation event which exhumed the peridotites, and we therefore reinterpret the spinel tectonites as being a result of deformational overprinting of part of the granular domain. We also reinterpret garnet intergrown with spinel in the mylonite zone as part of the pre-mylonitic porphyroclast assemblage, rather than as a syn-mylonite assemblage. This places mylonite formation within the spinel field, rather than right on the garnet-spinel transition (18 kb). Two-dimensional thermal modeling indicates that these conditions require removal of lithospheric mantle below 100 km followed by exhumation along a low angle shear zone. Excision of material during exhumation is required to explain the steep thermal gradients observed. These results shed light on the mechanisms of back-arc extension, as well as the emplacement of orogenic lherzolites.

A Late Cretaceous lamprophyre dyke in the Villany Mts (S Hungary), situated in the Tisza unit, contains abundant spinel lherzolite xenoliths with porphyroclastic textures. Mineral chemistry suggests a relatively fertile mantle, which experienced only 5–7% melt extraction. Differences in porphyroclast and neoblast chemistry and thermobarometric calculations suggest that the mantle section represented by the xenoliths experienced recrystallization at lower PT as it was transported to shallow mantle depths close to the plagioclase stability field, followed by later relaxation. Based on volcanological and sedimentological constraints from the Villany Mts and the neighboring Mecsek Mts, we suggest that the uprise of the subcontinental mantle material was related to a Cretaceous rifting event and lithospheric deformation of the southwestern part of the Tisza unit. Mantle upwelling and formation of lamprophyre melts can be related to generation or reactivation of deep fractures of the lithosphere, during a period of lithospheric extension between the major nappe emplacements (Albian–Cenomanian and Paleocene) of the region.

The mylonitization of the Pankenushi gabbro in the Hidaka metamorphic belt of central Hokkaido, Japan, occurred along its western margin at ≈600 MPa and 660–700 °C through dynamic recrystallization of plagioclase and a retrograde reaction from granulite facies to amphibolite facies (orthopyroxene + clinopyroxene + plagioclase + H2O = hornblende + quartz). The reaction produced a fine-grained (≤100 μm) polymineralic aggregate composed of orthopyroxene, clinopyroxene, quartz, hornblende, biotite and ilmenite, into which strain is localized. The dynamic recrystallization of plagioclase occurred by grain boundary migration, and produced a monomineralic aggregate of grains whose crystallographic orientations are mostly unrelated to those of porphyroclasts. The monomineralic plagioclase aggregates and the fine-grained polymineralic aggregates are interlayered and define the mylonitic foliation, while the latter is also mixed into the former by grain boundary sliding to form a rather homogeneous polymineralic matrix in ultramylonites. However in both mylonite and ultramylonite, plagioclase aggregates form a stress-supporting framework, and therefore controlled the rock rheology. Crystal plastic deformation of pyroxenes and plagioclase with dominant (100)[001] and (001)1/2 1 1 ¯ 0 > slip systems, respectively, produced distinct shape- and crystallographic-preferred orientations of pyroxene porphyroclasts and dynamically recrystallized plagioclase grains in both mylonite and ultramylonite. Euhedral to subhedral growth of hornblende in pyroxene porphyroclast tails during the reaction and its subsequent rigid rotation in the fine-grained polymineralic aggregate or matrix produced clear shape- and crystallographic-preferred orientations of hornblende grains in both mylonite and ultramylonite. In contrast, the dominant grain boundary sliding of pyroxene and quartz grains in the fine-grained polymineralic aggregate of the mylonite resulted in their very weak shape- and crystallographic-preferred orientations. In the fine-grained polymineralic matrix of the ultramylonite, however, pyroxene and quartz grains became scattered and isolated in the plagioclase aggregate so that they were crystal-plastically deformed leading to stronger shape- and crystallographic-preferred orientations than those seen in the mylonite.

The role of fluids in the deformation of continental serpentinites is investigated from structural, microstructural and petrographic analyses applied to a km-scale porphyroclast mantled in a viscous matrix of amphibolites. The clast is sited within a shear zone of the Palaeozoic Maures massif, France. Syntectonic fluid–rock interactions occurred from km to mm scales, at first on the clast borders (along the main rheological boundaries) then within the clast. They are accommodated macroscopically by slickenfibers faults and microscopically by shear microcracks within crack-seal veins, typifying an intermediate, brittle–ductile behaviour of serpentinites. Three main stages of deformation–serpentinisation processes occurred in relation with the left-lateral movement of the hosted shear zone. They developed under metamorphic conditions evolving from amphibolites to green-schists facies conditions (∼ 400 MPa/550 °C to ∼ 200 MPa/

The island of Corsica (France) plays a central role in any reconstruction of Western Mediterranean geodynamics and paleogeography but several key aspects of its geological evolution are still uncertain. The most debated topics include the interpretation of the Corsican orogen as the result of an east- or west-directed subduction, and the actual involvement of the Variscan basement of Corsica in the Alpine orogenic cycle. This study integrates 40 Ar– 39 Ar laserprobe, mesostructural, microtextural, and microchemical analyses and places relevant constraints on the style, P – T conditions, and timing of Alpine-age, pervasive ductile shear zones which affected the Variscan basement complex of central Corsica, a few kilometers to the west of the present-day front of the Alpine nappes. Shear zones strike ~ NNE–SSW, dip at a high angle, and are characterized by a dominant sinistral strike-slip component. Two of the three investigated shear zones contain two texturally and chemically resolvable generations of white mica, recording a prograde (burial) evolution: (1) deformed celadonite-poor relicts are finely overgrown by (2) a celadonite-rich white mica aligned along the main foliation. White mica from a third sample of another shear zone, characterized by a significantly lower porphyroclast/matrix ratio, exhibits a nearly uniform high-celadonite content, compositionally matching the texturally younger phengite from the nearby shear zones. Mineral-textural analysis, electron microprobe data, and pseudosection modeling constrain P – T conditions attained during shearing at ~ 300 °C and minimum pressures of ~ 0.6 GPa. In-situ 40 Ar– 39 Ar analyses of coexisting low- and high-celadonite white micas from both shear zones yielded a relatively wide range of ages, ~ 45–36 Ma. Laser step-heating experiments gave sigmoidal-shaped age profiles, with step ages in line with in-situ spot dates. By contrast, the apparently chemically homogenous high-celadonite white mica yielded concordant in-situ ages at ~ 34 Ma, but a hump-shaped age spectrum, with maximum ages of ~ 35 Ma and intermediate- to high-temperature steps as young as ~ 33–32 Ma. Results indicate that the studied samples consist of an earlier celadonite-poor white mica with a minimum age of ~ 46 Ma, overgrown by a synshear high-celadonite white mica, developed at greater depth between ~ 37 and 35 Ma; faint late increments in shearing occurred at ≤ 33–32 Ma, when white mica incipiently re-equilibrated during exhumation. Results suggest that ductile shearing with a dominant strike-slip component pervasively deformed the Corsican basement complex during the emplacement and progressive thickening of the Alpine orogenic wedge and broaden the extent of the domain affected by the Alpine tectonometamorphic events. Integration of petrological modeling and geochronological data shows that the Variscan basement of central Corsica, close to the Alpine nappes, was buried during the late Eocene by ≥ 18 km of Alpine orogenic wedge and foreland deposits. Our results, combined with previously published apatite fission-track data, imply an overburden removal ≥ 15 km from the late Eocene (Priabonian) to the early Miocene (Aquitanian), pointing to a minimum average exhumation rate of 1.3–1.5 mm/a.

Two spinel lherzolite xenoliths from Hungary that contain pyroxene–spinel symplectites have been studied using EPMA, Laser ablation ICP-MS and universal stage. Based on their geochemical and structural characteristics, the xenoliths represent two different domains of the shallow subcontinental lithospheric mantle beneath the Pannonian Basin. The occurrence of symplectites is attributed to the former presence and subsequent breakdown of garnets due to significant pressure decrease related to lithospheric thinning. This implies that both mantle domains were once part of the garnet lherzolitic upper mantle and had a similar history during the major extension that formed the Pannonian Basin. Garnet breakdown resulted in distinct geochemical characteristics in the adjacent clinopyroxene crystals in both xenoliths. This is manifested by enrichment in HREE, Y, Zr and Hf towards the clinopyroxene porphyroclast rims and also in the neoblasts with respect to porphyroclast core compositions. This geochemical feature, together with the development and preservation of the texturally very sensitive symplectites, enables us to determine the relative timing of mantle processes. Our results indicate that garnets had been metastable in the spinel lherzolite environment and their breakdown to pyroxene and spinel is one of the latest processes that took place within the upper mantle before the xenoliths were brought to the surface. © 2006 Published by Elsevier B.V.

The Tres Arboles ductile fault zone in the Eastern Sierras Pampeanas, central Argentina, experienced multiple ductile deformation and faulting events that involved a variety of textural and reaction hardening and softening processes. Much of the fault zone is characterized by a (D2) ultramylonite, composed of fine-grained biotite + plagioclase, that lacks a well-defined preferred orientation. The D2 fabric consists of a strong network of intergrown and interlocking grains that show little textural evidence for dislocation or dissolution creep. These ultramylonites contain gneissic rock fragments and porphyroclasts of plagioclase, sillimanite and garnet inherited from the gneissic and migmatitic protolith (D1) of the hangingwall. The assemblage of garnet + sillimanite + biotite suggests that D1-related fabrics developed under upper amphibolite facies conditions, and the persistence of biotite + garnet + sillimanite + plagioclase suggests that the ultramylonite of D2 developed under middle amphibolite facies conditions. Greenschist facies, mylonitic shear bands (D3) locally overprint D2 ultramylonites. Fine-grained folia of muscovite + chlorite ± biotite truncate earlier biotite + plagioclase textures, and coarser-grained muscovite partially replaces relic sillimanite grains. Anorthite content of shear band (D3) plagioclase is c. An30, distinct from D1 and D2 plagioclase (c. An35). The anorthite content of D3 plagioclase is consistent with a pervasive grain boundary fluid that facilitated partial replacement of plagioclase by muscovite. Biotite is partially replaced by muscovite and/or chlorite, particularly in areas of inferred high strain. Quartz precipitated in porphyroclast pressure shadows and ribbons that help define the mylonitic fabric. All D3 reactions require the introduction of H+ and/or H2O, indicating an open system, and typically result in a volume decrease. Syntectonic D3 muscovite + quartz + chlorite preferentially grew in an orientation favourable for strain localization, which produced a strong textural softening. Strain localization occurred only where reactions progressed with the infiltration of aqueous fluids, on a scale of hundreds of micrometre. Local fracturing and microseismicity may have induced reactivation of the fault zone and the initial introduction of fluids. However, the predominant greenschist facies deformation (D3) along discrete shear bands was primarily a consequence of the localization of replacement reactions in a partially open system.

With increasing mylonitization of aplite veins and their adjacent granite in the Hatagawa Shear Zone of NE Japan, the modal content and grain size of feldspar porphyroclasts decrease, while the modal content of fine-grained (≤50 μm) polymineralic aggregate of plagioclase, K-feldspar, quartz and biotite increases up to 60–80% in the granite and aplite ultramylonites. Deformation-induced myrmekite lobes are ubiquitously developed around K-feldspar porphyroclasts. The occurrence and chemical composition of plagioclase and K-feldspar in fine-grained polymineralic aggregates indicate that the plagioclase is mostly derived from myrmekite, while the interstitial K-feldspar originates from isolated porphyroclast rims between growing myrmekite lobes and/or precipitations from solution. Modal and bulk chemical compositions of granite and aplite indicate that the mylonites are preferentially developed in aplite veins richer in K-feldspar than the surrounding granite. Deformation-induced K-feldspar replacement by myrmekite is then more favored during deformation to produce more abundant fine-grained polymineralic aggregates. Subsequent reaction softening leads to more enhanced deformation, which promotes further myrmekite-forming reaction. Such a positive feedback of deformation and myrmekite-forming reaction in aplite veins and their adjacent granite may be responsible for strain localization in the granite and aplite mylonites along the Hatagawa Shear Zone.

Survival of plagioclase in the residual melting column during melting can have a significant impact on the melting process beneath a mid-oceanic ridge [Asimow et al., Phil. Trans. R. Soc. London Ser. A 355 (1997) 255^281]. Here we investigate the origin of plagioclase that occurs in some rare mantle xenoliths from Oahu, Hawaii. The xenoliths are harzburgitic with less than 2 modal% clinopyroxene and are characterized by strong foliation and porphyroclastic texture. Olivine and orthopyroxene are common porphyroclasts; and only one xenolith (77PAII-9) contains a single large clinopyroxene porphyroclast with thick exsolved orthopyroxene lamellae. The strongly foliated groundmass shows well-developed triple-point junctions and is dominantly composed of olivine (ol85� 90 opx7� 14 cpx61� 2 plag3� 5 spineltrace). Spinel grains are small and dispersed through the groundmass and show extreme variation in Cr/Al ratio within individual thin sections, indicating that they are out of equilibrium with the other phases in these xenoliths. A highly anorthitic plagioclase (An92� 96) occurs only in the groundmass and its modal abundance (V3^5%) is too high relative to the abundance of clinopyroxene (commonly 6 1%) for it to be a residual phase, implying that plagioclase may have an exotic origin. The porphyroclasts show strong compositional zoning near the rims and appear to be relict phases (as are all the spinel grains) that had once equilibrated with melts within the stability field of spinel peridotite (pressure V1^3 GPa). Clinopyroxene neoblasts and the single porphyroclast in PAII-9 are all characterized by lithosphere-like strongly depleted light rare earth element (chondrite-normalized) patterns. The clinopyroxene porphyroclast in PAII-9 is zoned in Al, Eu, Cr, and Na. The porphyroclastic ortho- and clinopyroxenes give a homogenized (host+exsolution) temperature of 1300‡C, which is inferred to be the temperature at which the porphyroclast cores equilibrated with a MORB melt in the spinel peridotite stability field. Last subsolidus equilibration of these xenoliths occurred at a temperature of V1000‡C and a pressure of 0.7 GPa based on groundmass (and rim of porphyroclast) equilibration. Diffusion calculations at groundmass equilibration pressure and temperature conditions show that the Al zoning and Eu anomaly in the rim of the clinopyroxene porphyroclast probably developed within about 80 kyr of subsolidus equilibration with plagioclase. These xenoliths show decoupled behavior of Sr and Os isotopic systems with some of the highest Koolau-like 87 Sr/ 86 Sr ratios (whole rock) but lowest

The metamorphic evolution of the garnet peridotite body of Alpe evolution of these rocks as deduced from their mineral assemblages and textures can potentially provide imArami, Central Alps, is a matter of current controversy. In this paper, the interand intragrain distribution of major and trace portant constraints on orogenic processes (e.g. Van der Wal & Vissers, 1993; Zhang et al., 1994; Altherr & Kalt, elements obtained by electron and ion probe microanalyses is used to better constrain the P–T evolution of this peridotite. Using the 1996; Van Roermund & Drury, 1999). One of the most intriguing examples of a subduction-related ultrahighcompositions of homogeneous porphyroclast cores, peak metamorphic conditions of 1180 ± 40°C and 5·9 ± 0·3 GPa are estimated, pressure peridotite–eclogite association is exposed in the Central Alps at Alpe Arami (AA), Ticino, Switzerland based on consistent results from the application of several independent thermometers (Fe–Mg exchange between garnet, pyroxenes and (Grubenmann, 1908; O’Hara & Mercy, 1966; Mockel, 1969; Rost et al., 1974; Ernst, 1977, 1978, 1981; Evans olivine, Ni exchange between garnet and olivine, Co and Ni exchange between orthopyroxene and clinopyroxene), the Al-in-orthopyroxene & Trommsdorff, 1978; Trommsdorff, 1990; Pfiffner & Trommsdorff, 1998). barometer and the Ca–Cr systematics of garnet. Orthopyroxene and clinopyroxene porphyroclasts are, however, not in equilibrium with Whereas the age of the ultrahigh-pressure metamorphism that affected the AA rocks is now reasonably respect to some elements with low diffusivities, such as Ca, Ti, Cr, V and Sc. This disequilibrium appears to be the main cause for well constrained at 43–35 Ma (Sm–Nd on garnet– the lower P–T values suggested by some of the previous workers. clinopyroxene–whole rock: Becker, 1993; SHRIMP On the other hand, there is no evidence for an ultradeep (>200 km) U–Pb on zircon: Gebauer, 1996, 1999), the P–T evolution origin of the Alpe Arami body as postulated recently. Chemical of the AA garnet peridotite has been the subject of a zonation profiles across mineral grains suggest that during retrograde continuing controversy (Ernst, 1978, 1981; Evans & evolution a near-isothermal decompression was followed by acTrommsdorff, 1978; Becker, 1993; Dobrzhinetskaya et celerated cooling. al., 1996, 1999; Brenker & Brey, 1997; Green et al., 1997a, 1997b; Risold et al., 1997; Ulmer & Trommsdorff, 1997; Pfiffner & Trommsdorff, 1998; Bozhilov et al.

Cataclastic rocks and pseudotachylytes are exposed along the Uchinoura shear zone, a normal fault zone cutting the middle Miocene (14 Ma) Osumi granodiorite in southern Kyushu, Japan. Cataclastic rocks include non-foliated clast-supported to matrix-supported cataclasite and foliated clast-supported cataclastic granodiorite. In these rocks, fracturing and comminution played a major role, but dissolution and recrystallization of quartz, and plastic deformation of quartz and biotite were also active processes, especially in foliated granodiorite. Two types of pseudotachylyte are distinguished: a foliated-type characterized by a planar arrangement of clasts and microlites, and a spherulitic-type characterized by clasts surrounded by microlite overgrowths. Both types are of melt origin, as attested by the presence of microlites and rounded or embayed clasts, and by the scarcity of biotite clasts. Unlike spherulitic-type pseudotachylyte, which solidified without being deformed, the foliated-type pseudotachylyte underwent flow before complete solidification. This deformation is thought to reflect post-seismic strain accommodation immediately following the main slip episode. Kinematic indicators, which consist of Riedel-type secondary fractures branching on primary fractures, shear bands offsetting the foliation of foliated granodiorite, or asymmetrical porphyroclast systems within pseudotachylyte veins, show that all fault rocks were generated during N–S- to NW–SE-directed extensional deformation. Pseudotachylyte is closely associated both in time and space with cataclastic rocks, thus indicating that the behaviour of the Uchinoura fault zone alternated between comminution and frictional melting. Given the slow strain rates which characterize dissolution and recrystallization processes detected in cataclasites, the juxtaposition of pseudotachylytes and foliated cataclasites provides an example of aseismic and seismic displacements within the same shear zone.

The Late Archean Striding–Athabasca mylonite zone, western Canadian Shield, is a 400-km-long, linked system of granulite facies mylonite belts developed in the continental crust. The granulite facies mylonites (800–1000°C at 0.8–1.1 GPa) developed in subsets of the anhydrous mineral assemblage clinopyroxene–orthopyroxene–garnet–plagioclase–quartz±hornblende. Comparison of the microstructures of the Striding–Athabasca mylonites with published descriptions of natural and experimental deformation of geological and analogue materials at elevated homologous temperatures provides some insight into the role played by different processes in the development of the Striding–Athabasca mylonites. In addition to dislocation creep and dynamic recrystallisation, extensive mass transfer occurred contemporaneously with brittle fracture and cataclasis during granulite facies metamorphism. The microstructure and extreme phase dispersal in anhydrous polymineralic matrix mosaics is indicative of efficient grain-scale and aggregate-scale diffusion, grain boundary mobility, and mass transfer, at relatively slow strain rates. Despite their annealed appearance, the granoblastic matrix mosaics developed synkinematically; deformation-induced dilation may enhance metamorphic reactions where products are more voluminous than reactants. In the porphyroclast population, highly elongate (>20:1) monocrystalline orthopyroxenes appear to be fragments of dismembered, kinked parent grains, rather than stretched porphyroclasts. Granulite facies mylonites should not be treated as direct analogues of greenschist facies mylonites. In particular, it is essential to evaluate the potential positive feedback between structural and metamorphic processes in highly strained, high-temperature shear zone rocks.

A unique suite of basalt-borne sheared and granular spinel lherzolites from the Yitong graben provides an insight into the nature of the lithospheric mantle adjacent to the Tancheng-Lujiang fault, northeastern China. Data obtained by ion microprobe (multi-elemental) and laser-fluorination (δ18O ratios) techniques have been integrated with thermobarometric data. Shallow (25–35 km), cool (742–967°C), sheared periototites contain clinopyroxenes with LREE-depleted profiles and minor enrichments in La. A marked difference in the δ18O ratios of porphyroclast and neoblast olivine in these sheared peridotites may be due to thermal re-equilibration or recent fluid activity. Deeper (50–60 km), hotter (950–1073°C), granular peridotites contain LREE-depleted and enriched clinopyroxenes, the latter depleted in Ti and Zr. Melts in equilibrium with these LREE-enriched and TiZr-depleted pyroxenes are very similar to potassic glasses in wehrlite xenoliths from the Yitong graben. On the basis of these data it is proposed that toward the base of the lithosphere hot, granular ‘refractory’ peridotites predominate and that they have been variably enriched by intergranular migration of small volume potassic melts. It is believed that the cold, sheared, peridotites from near the Moho were derived from a hot, granular, refractory precursor but that movement on the Tancheng-Lujiang fault, a major intracontinental shear zone, generated a ‘sheared’ texture and vertical displacement relocated these mantle rocks in a cooler part of the lithosphere.

Samples collected across the NW Grenville Front in Ontario, Canada, have been used to document the change in feldspar microstructures produced during shearing under metamorphic conditions that range from upper amphibolite to lower greenschist grade. The differential displacement across this 5 km thick section of the Grenville Front tectonic zone is at least 10 km. At upper to mid-amphibolite conditions both plagioclase and potassium feldspars behave plastically. Grainsize reduction occurs through grain boundary migration and/or subgrain rotation and produces a core-and-mantle structure. At middle to lower amphibolite grade, grain-size reduction is accomplished through grain boundary migration. At lower amphibolite to upper greenschist grade, grain-size reduction occurs by the nucleation and growth of new grains with a different composition. The abundance of intracrystalline deformation at this grade is inversely proportional to the amount of fine-grained matrix. In ultramylonite with a high matrix to porphyroclast ratio few intracrystalline strain features occur other than ubiquitous undulatory extinction. Myrmekite is well developed on the high stress boundaries in K-feldspar. At upper to mid-greenshist conditions, feldspars deform mainly by slip on synthetic shear fractures. Deformation twins, undulatory extinction, deformation bands and kink bands occur in plagioclase, and perthite flames and myrmekite occur principally on high stress boundaries in K-feldspar. Grain-size reduction in feldspar is accomplished through cataclasis. At middle to lower greenschist facies, deformation in feldspars is dominated by antithetic fractures. The transition from the dominance of synthetic to antithetic fractures is gradational. Below lower greenschist grade, rocks deform by pervasive cataclasis of all minerals.

Microstructures from a traverse through Early Proterozoic gneiss and Adelaidean and Cambrian metasediments on southern Fleurieu Peninsula, South Australia, show strong mylonitic textures in their common shallowly east‐dipping foliation. The intensity and style of mylonite development is strongly linked to rock type: quartz‐biotite schist has undergone crystal plastic deformation including twinning, deformation banding, intragranular creep, subgrain formation and dynamic recrystallization. Grain boundary sliding and slip on basal planes of cleavage forming micas are dominant in metapelite. Gneiss in the Early Proterozoic inlier shows plastic deformation of quartz augen, brittle deformation of feldspar porphyroclasts and basal slip on phyllosilicates. The Adelaidean calc‐silicate rocks structurally beneath the Early Proterozoic inlier are the weakest unit on the traverse and accommodate the biggest displacement. The sense of shear has been derived statistically from porphyroclast asymmetries, S‐C planes and...

A suite of granulites including a meta-ironstone, pyroxenites, and spinel-lherzolites from East Tonagh Island, Enderby Land, Antarctica, preserve exsolution-recry-stallization features consistent with a shared metamorphic evolution that involves marked cooling from initial metamorphic temperatures of nearly 1000°C. Reintegrated pre-exsolution and pre-reaction grain compositions in the meta-ironstone indicate the former coexistence of metamorphic pigeonite (Wo12En38Fs50) and ferroaugite (Wo35En31Fs34) at temperatures in excess of 980°C for pressures of 7 kbar (0.7 GPa) using pyroxene quadrilateral thermometry (Lindsley, 1983). Intra-grain lamellae relationships indicate the exsolution of a second pigeonite (Wo12En35Fs53) from the ferroaugite at temperatures in the range 930–970°C, prior to the c. 720–600°C exsolution of orthopyroxene and clinopyroxene (100) lamellae and later partial recrystallization at similar temperatures. Although pyroxenitic and iherzolitic granulites preserve a much less complete history, reintegrated porphyroclast compositions in these yield temperature estimates which approach those inferred from the metaironstone. Pyroxene thermometry based on neoblast compositions suggests that recrystallization post-dating a late, low intensity, deformation phase (D3) occurred at temperatures greater than 600°C. These results are consistent with the independent evidence obtained from studies of metapelitic and felsic rock types for very high temperature metamorphism throughout the Napier Complex followed by near-isobaric cooling and later deformation under lower-grade granulite facies conditions. Comparison with similar pyroxene data from Fyfe Hills (Sandiford & Powell, 1986) demonstrates further the regional significance of these high temperatures, and implies broadly isothermal metamorphic conditions over a large area (∼ 5000 km2) and thickness (6–9 km) of lower crust at c. 3070 Ma.

The Josephine Peridotite has experienced deformational episodes over a wide range of temperatures. Deformation within the stability field of a serpentine group mineral appears invariably accompanied by retrogressive metamorphism (serpentinization). Two general types of deformed serpentinites have been identified: (1) incohesive serpentinites (most common), which appear to have experienced low-temperature, predominantly brittle deformation and (2) serpentinite mylonites, which have experienced plastic flow near the upper limit of antigorite stability. The distinction between these two is based on textural and mineralogical criteria. Serpentinite mylonites are cohesive, homogeneously foliated, and typically strongly lineated rocks composed primarily of antigorite and magnetite. They occur in planar-bounded zones enclosed within the Josephine Peridotite that were the loci of both noncoaxial deformation and serpentinization. We discuss the development of two important shear-sense indicators, porphyroclast systems and oblique shear-band foliations as observed in these rocks. Porphyroclast systems in serpentinite mylonites are often comprised of cores and mantles of different mineralogy. Cores are usually composed of primary pyrogenetic minerals, whereas mantles, including tails, are composed of products of the serpentinization reaction. These systems are therefore termed "porphyroctasts with neocrystallization tails.". Oblique shear bands are common micro-structures in serpentinite mylonites. Shear bands which offset the mylonitic foliation along narrow, discrete fractures are termed "f-type," whereas those which offset the mylonitic foliation along narrow, ductile "micro-shear zones" are termed "d-type". The latter are considered more reliable for shear-sense determinations. Strains accomplished by movement on shear bands are probably small. Strain resulting from a volume increase dur-ing serpentinization while holding the shear-zone width constant is sufficient to produce the shear bands. Therefore, we suggest that oblique shear bands may develop to accom-modate volume increases associated with deformation.

Electron probe analyses show that the pyroxenes of the peridotite-mylonites are significantly enriched in alumina when compared with the pyroxenes in the dunites described by Ross et al (1954), but are comparable to the Lizard pyroxenes and to those in olivine nodules of basalts. The closest comparison of the latter is with nodules of a mugearite from Ichinomegata, Akita, Japan. In the groundmass of the amphibole peridotite-mylonites, clinopyroxene is extremely rare, and the variation of the amphibole content is estimated to range from

Unusually alumina-poor orthopyroxene is found in a spinel peridotite from the Horoman Peridotite Complex, Japan. Al2O3 ,C r 2 O 3and CaO contents in the low-Al orthopyroxene (named Low-Al OPX hereafter) are 5025 wt%, 5004 wt% and 503 wt%, respectively, and are distinctively lower than those in orthopyroxene porphyroclasts. The Low-Al OPX occurs in two modes, both at the margin of olivine. The first mode of occurrence is as the rim of a large orthopyroxene porphyroclast in contact with olivine. This type of Low-Al OPX occurs only locally (15mm 45mm), and the orthopyroxene rim in contact with olivine more commonly has normal Al2O3 contents (42 wt%). In the second mode of occurrence, the Low-Al OPX occurs as a thin film, 5mm 50mm in dimension, at a grain boundary between olivine and clinopyroxene. Trace element compositions of porphyroclast clinopyroxene in the sample indicate that the sample having the Low-Al OPX underwent metasomatism although there are no hydrous minerals around the Low-Al OPX. Petrographic observations and trace element compositions of clinopyroxene combined with an inferred P‐T history of the Horoman peridotite suggest that the Low-Al OPX was formed through a very local reaction between peridotite and invasive fluids, probably formed by dehydration of a subducted slab, in a late stage of the history of the Horoman peridotite. Crystallization of orthopyroxene, representing addition of silica to mantle lherzolite via a CO2a H2O-bearing fluid phase, is a mechanism for metasomatic alteration of mantle wedge peridotite.

Subhorizontal shear sense along subvertical mylonite zones marking the southeast and northwest flanks of the Pine Mountain belt in Alabama, i.e., the Towaliga, Bartletts Ferry, and Goat Rock fault zones, has been deduced from S-C composite planar fabrics, extensional shear bands, displaced broken grains, asymmetric folds, and porphyroclast systems. Quartz and feldspar elongation lineations are generally subhorizontal and closely correspond to estimated sliplines. Each of the fault zones records dominantly dextral shear; the Towaliga has an apparent minor oblique, down-to-the-north normal component, and the Goat Rock has a minor down-to-the-south normal component. The mylonite zones postdate the early to middle Paleozoic schistosity in rocks outside the shear zones and thus are considered to be late Paleozoic in age. Results imply persistence of the late Paleozoic (Alleghanian?) dextral shear system into the southernmost exposures of the Appalachian orogen.