ALBERT

Notes: Extensive examination of large numbers of spatially orientated thin sections of orientated samples from orogens of all ages around the world has demonstrated that porphyroblasts do not rotate relative to geographical coordinates during highly non-coaxial ductile deformation of the matrix subsequent to their growth. This has been demonstrated for all tectonic environments so far investigated. The work also has provided new insights and data on metamorphic, structural and tectonic processes including: (1) the intimate control of deformation partitioning on metamorphic reactions; (2) solutions to the lack of correlation between lineations that indicate the direction of movement within thrusts and shear zones, and relative plate motion; and (3) a possible technique for determining the direction of relative plate motion that caused orogenesis in ancient orogens.

Notes: Abstract Textural ‘unconformities’or truncations are common in porphyroblasts with complex inclusion trails. They reflect cycles of successive foliations that develop against competent porphyroblasts during orogenesis and are preserved by successive growth increments. Their truncational character results from shear and dissolution along a particular foliation generating a differentiated crenulation cleavage. The increment of porphyroblast growth that follows a textural ‘unconformity’may or may not mark a significant compositional change, depending on the amount of movement of the rock through P–T space between cleavage-forming events.Although historically interpreted to result from a significant metamorphic hiatus, most textural unconformities indicate that the reactions involved in the formation of these minerals are episodic during continuous prograde metamorphism, starting and stopping as a function of the stage of crenulation of the matrix foliation and the pattern of deformation partitioning. Such episodic reaction behaviour can only occur for multivariant reactions, or successive but different univariant reactions. The reason why garnet is the most common porphyroblast to exhibit evidence for episodic reactions is probably the fact that it grows by multivariant reactions over a much wider P–T range than most other common porphyroblast phases.Porphyroblast growth is micrometasomatic. It is episodic because a significant reduction of strain occurs within domains of progressive shortening each time continuous progressive shearing domains form on their margins. This stops microfracture development across the progressive shortening domains, thereby preventing rapid access and interaction of fluid, ions and complexes with porphyroblast boundaries. Shifting patterns of deformation partitioning and resulting small-scale juxtaposition of different compositional layers spreads the duration and location of multivariant reactions and causes differential timing of porphyroblast growth along a particular stratigraphic horizon. It may also locally preserve metastable metamorphic assemblages.In regionally metamorphosing/deforming pelites, near-simultaneous cessation of porphyroblast growth on all rims, once continuous differentiated progressive shearing domains have formed nearby, precludes fluid recirculation as a significant process for removal of material during cleavage development. Alternatively, diffusion of simple molecules and dissociated ions along actively shearing and micro-gaped phyllosilicates, with recomplexing in fluid-filled microfractures, readily explains the control of deformation partitioning on reaction site and reaction duration.

Notes: Dissolution and solution transfer during deformation/metamorphism are controlled by the partitioning of deformation into progressive shearing and shortening components. Progressive shearing is readily accommodated by slip on the planar crystal structure of phyllosilicates and graphite without accumulating dislocation density gradients across grain boundaries.Progressive shortening is accommodated by the cores of most other minerals (including sulphides). These minerals develop strain, and hence dislocation density gradients, on their rims due to progressive shearing along grain boundaries. These gradients are particularly large when the mineral abuts phyllosilicate or graphite. The resulting chemical potential gradients between the core and rim drive dissolution, causing removal of the highly strained grain margins.Removal of dissolved material by solution transfer is aided by the geometry of shearing of phyllosilicates and graphite around other grains in an active anastomosing foliation. Interlayers and interfaces on boundaries lying at a low angle to the direction of shearing, and oriented relative to the sense of shear such that they can open, gape by small amounts. Water present in these interlayer spaces becomes destructured, considerably enhancing diffusion rates along the foliation.Penetrative volume loss, especially in deforming/metamorphosing pelitic rocks, is large at all metamorphic grades, increasing and becoming more penetrative with depth to at least the transition into granulite and eclogite facies. Transference of material by fluid flow from deep to high levels in the earth's crust is precluded because thousands to tens of thousands of rock volumes of fluid are required, necessitating continual recirculation of fluid from shallow to deep crustal levels in one large or several small sets of cells, unless some extremely large-scale form of fluid channelling is possible. Reassessment of diffusion mechanisms, and hence rates, during deformation and pervasive foliation generation in large volumes of rock where fluid channeling cannot provide enough fluid, indicates that diffusion can proceed with sufficient rapidity that massive recirculation of fluid is no longer required. The amount of fluid can be reduced sufficiently to allow large volume losses by a one-way flow of fluid to the earth's surface, in deforming/metamorphosing environments where the fluid pressure equals or exceeds the hydrostatic pressure.Deformation partitioning-controlled dissolution progressively changes the bulk chemistry of a rock containing phyllosilicates or graphite during deformation/metamorphism because matrix minerals, other than phyllosilicates and graphite, are preferentially removed. The large size of porphyroblasts, if present, tends to preserve them from dissolution. Hence, the bulk chemistry operative during subsequent porphyroblast growth can have changed considerably from that operative when the first porphyroblasts grew, in rocks in which bedding is still well preserved.

Notes: Detailed microstructural analysis of inclusion trails in hundreds of garnet porphyroblasts from rocks where spiral-shaped inclusion trails are common indicates that spiral-shaped trails did not form by rotation of the growing porphyroblasts relative to geographic coordinates. They formed instead by progressive growth by porphyroblasts over several sets of near-orthogonal foliations that successively overprint one another. The orientations of these near-orthogonal foliations are alternately near-vertical and near-horizontal in all porphyroblasts examined. This provides very strong evidence for lack of porphyroblast rotation.The deformation path recorded by these porphyroblasts indicates that the process of orogenesis involves a multiply repeated two-stage cycle of: (1) crustal shortening and thickening, with the development of a near-vertical foliation with a steep stretching lineation; followed by (2) gravitational instability and collapse of this uplifted pile with the development of a near-horizontal foliation, gravitational spreading, near-coaxial vertical shortening and consequent thrusting on the orogen margins. Correlation of inclusion trail overprinting relationships and asymmetry in porphyroblasts with foliation overprinting relationships observed in the field allows determination of where the rocks studied lie and have moved within an orogen. This information, combined with information about chemical zoning in porphyroblasts, provides details about the structural/metamorphic (P-T-t) paths the rocks have followed.The ductile deformation environment in which a porphyroblast can rotate relative to geographic coordinates during orogenesis is spatially restricted in continental crust to vertical, ductile tear/transcurrent faults across which there is no component of bulk shortening or transpression.

Notes: In the Woodroffe Thrust mylonite zone, central Australia, recrystallization in plagioclase and K-feldspar involved subgrain rotation, assisted by grain-boundary or kink band boundary bulging, without contribution from a change in the chemical composition from host grains to new grains. The size of subgrains and new grains changes across the mylonite zone, apparently as a function of the strain rate and the H2O content of the rock.The partitioning of deformation into zones of progressive shearing and progressive shortening controls the sites of recovery and recrystallization in feldspar during mylonitization. The size of feldspar porphyroclasts in well developed mylonites is governed by the scale of deformation partitioning reached in the earlier stages of mylonitization, before the formation of a large proportion of fine-grained matrix that can accommodate the progressive shearing component of the deformation.Recrystallization occurs in microcline, apparently without involving a translation to a monoclinic structure, as microcline-twinned new grains are common adjacent to microcline-twinned host grains. K-feldspar triclinicity values calculated from XRD traces increase from the margins to the interior of the mylonite zone, in conjunction with deformation intensity. K-feldspar host grains locally have cores of orthoclase or untwinned microcline, surrounded by mantles of twinned microcline, suggesting a relationship between the presence of microcline twinning and the degree of K-feldspar triclinicity.

Notes: Abstract Reactivation of early foliations accounts for much of the progressive strain at more advanced stages of deformation. Its role has generally been insufficiently emphasized because evidence is best preserved where porphyroblasts which contain inclusion trails are present. Reactivation occurs when progressive shearing, operating in a synthetic anastomosing fashion parallel to the axial planes of folds, changes to a combination of coarse- and finescale zones of progressive shearing, some of which operate antithetically relative to the bulk shear on a fold limb. Reactivation of earlier foliations occurs in these latter zones.Reactivation decrenulates pre-existing or just-formed crenulations, generating shearing along the decrenulated or rotated pre-existing foliation planes. Partitioning of deformation within these foliation planes, such that phyllosilicates and/or graphite take up progressive shearing strain and other minerals accommodate progressive shortening strain, causes dissolution of these other minerals. This results in concentration of the phyllosilicates in a similar, but more penetrative manner to the formation of a differentiated crenulation cleavage, except that the foliation can form or intensify on a fold limb at a considerable angle to the axial plane of synchronous macroscopic folds.Reactivation can generate bedding-parallel schistosity in multideformed and metamorphosed terrains without associated folds. Heterogeneous reactivation of bedding generates rootless intrafolial folds with sigmoidal axial planes from formerly through-going structures. Reactivation causes rotation or ‘refraction’of axial-plane foliations (forming in the same deformation event causing reactivation) in those beds or zones in which an earlier foliation has been reactivated, and results in destruction of the originally axial-plane foliation at high strains. Reactivation also provides a simple explanation for the apparently ‘wrong sense’, but normally observed ‘rotation’of garnet porphyroblasts, whereby the external foliation has undergone rotation due to antithetic shear on the reactivated foliation. Alternatively, the rotation of the external foliation can be due to its reactivation in a subsequent deformation event.Porphyroblasts with inclusion trails commonly preserve evidence of reactivation of earlier foliations and therefore can be used to identify the presence of a deformation that has not been recognized by normal geometric methods, because of penetrative reactivation. Reactivation often reverses the asymmetry between pre-existing foliations and bedding on one limb of a later fold, leading to problems in the geometric analysis of an area when the location of early fold hinges is essential. The stretching lineation in a reactivated foliation can be radically reoriented, potentially causing major errors in determining movement directions in mylonitic schistosities in folded thrusts.Geometric relationships which result from reactivation of foliations around porphyroblasts can be used to aid determination of the timing of the growth of porphyroblasts relative to deformation events. Other aspects of reactivation, however, can lead to complications in timing of porphyroblast growth if the presence of this phenomenon is not recognized; for example, D2-grown porphyroblasts may be dissolved against reactivated S1 and hence appear to have grown syn-D1.

Notes: Investigation of microstructural relationships in major movement zones in metamorphic rocks, where the sense of displacement is known from regional geological relationships, indicates numerous problems with current concepts of shear-sense criteria and their application. The direction of apparent shearing commonly conflicts from one criterion to another (e.g. from the symmetry of quartz c-axis orientation diagrams to the asymmetry of extensional crenulation cleavages). This implies that interpretations of shear sense along foliations from some mesoscale and microscale criteria have been erroneous.A new approach to interpreting shear sense, involving the use of strain fields, resolves conflicts in mesoscopic and microscopic criteria and provides a method for determining coherent shear-sense histories extending back before the last shearing event for ‘any foliated metamorphic rock’. It also provides a powerful tool for determining the structural/metamorphic path that a rock has followed within an orogen. For determination of the shear sense on the last foliation developed in a rock, this approach uses geometries developed around competent heterogeneities such as quartz pebbles, pegmatite pods, veins, porphyroclasts, porphyroblasts and breccia clasts. A shear-sense history is derived by applying this approach to earlier foliations preserved within the heterogeneities and their strain shadows.

Notes: Abstract In regional metamorphic rocks, the partitioning of deformation into progressive shearing and progressive shortening components results in strain and strain-rate gradients across the boundaries between the partitioned zones. These generate dislocation density gradients and hence chemical potential gradients that drive dissolution and solution transfer. Phyllosilicates and graphite are well adapted to accommodating progressive shearing without necessarily building up large dislocation density gradients within a grain, because of their uniquely layered crystal structure. However, most silicates and oxides cannot accommodate strain transitions within grains without associated dislocation density gradients, and hence are susceptible to dissolution and solution transfer.As a consequence, zones of progressive shearing become zones of dissolution of most minerals, and of concentration of phyllosilicates and graphite. Exceptions are mylonites, where strain-rates are commonly high enough for plastic deformation to dominate over diffusion rates and therefore over dissolution and solution transfer. Porphyroblastic minerals cannot nucleate and grow in zones of active progressive shearing, as they would be dissolved by the effects of shearing strain on their boundaries. However, they can nucleate and grow in zones of progressive shortening and this is aided by the propensity for microfracturing in these zones, which allows rapid access of fluids carrying the material presumed to be necessary for nucleation and growth. Zones of progessive shortening also have a number of characteristics that help to lower the activation energy barrier for nucleation, this includes a build up of stored strain-energy relative to zones of progressive shearing, in which dissolution is occuring.Porphyroblast growth is generally syndeformational, and previously accepted criteria for static growth are not valid when the role of deformation partitioning is taken into account.Porphyroblasts in a contact aureole do not grow statically either, as microfracturing, associated with emplacement, allows access of fluids in a fashion that is similar to microfracturing in zones of progressive shortening.The criteria used for porphyroblast timing can be readily accommodated in terms of deformation partitioning, reactivation of deforming foliations, and a general lack of rotation of porphyroblasts, with the spectacular exception of genuinely spiralling garnet porphyroblasts.

Notes: Abstract Most porphyroblasts never rotate during ductile deformation, provided they do not internally deform during subsequent events, with the exception of relatively uncommon but spectacular examples of spiralling garnets. Instead, the surrounding foliation rotates and reactivates due to partitioning of the deformation around the porphyroblast. Consequently, porphyroblasts commonly preserve the orientation of early foliations and stretching lineations within strain shadows or inclusion trails, even where these structures have been rotated or obliterated in the matrix due to subsequent deformation. These relationships can be readily used to help develop an understanding of the processes of foliation development and they demonstrate the prominent role of reactivation of old foliations during subsequent deformation. They can also be used to determine the deformation history, as porphyroblasts only rotate when the deformation cannot partition and involves progressive shearing with no combined bulk shortening component.

Notes: Abstract The mid-Tertiary blueschists, eclogites and eclogitic gneisses of northern New Caledonia are the products of four phases of regional metamorphism and deformation (D1–D4). Omphacite, lawsonite and Mn-rich garnet isogradic surfaces were developed during the second deformation (D2) under prograde pressure and temperature conditions. Subsequent deformations (D3–D4) folded these D2 isogradic surfaces. However, within the P-retrograde, T-prograde metamorphic environment of the D4 phase, omphacite altered to albite and chlorite; as a result, a late-stage sub-horizontal isogradic surface developed for omphacite-out where this mineral preserved as relics within syn-D4 albite porphyroblasts. Other minerals that crystallized for the first time (epidote) or had rim additions (almandine phengite) during D4, also form nearly horizontal isogradic surfaces. Porphyroblastic garnet and albite contain inclusion trails, which allow their microstructural development and crystallization of the matrix to be traced from D2 to D4.Late syn-D4 the temperature increased markedly in association with an extensive exothermic decarbonation, even though the rocks were in a state of pressure retrogression. This caused considerable neocrystallization, recrystallization and growth of mattix and porphyroblasts such that, although S2 foliation crenulated by D3 and D4 is readily observable, almost all signs of stored strain due to D3 and D4 have been removed, and the deeper schists and eclogitic gneisses superficially appear to have undergone a drastic annealing recrystallization, post-dating deformation.