Transformation of Andalusite to Kyanite in the Alpujarride ...grupo179/pdf/Sanchez Navas 2012.pdfA,...

Transformation of Andalusite to Kyanite in the Alpujarride Complex (Betic Cordillera,Southern Spain): Geologic ImplicationsAuthor(s): Antonio Sánchez-Navas, Rita de Cassia Oliveira-Barbosa, Antonio García-Casco, andAgustín Martín-AlgarraReviewed work(s):Source: The Journal of Geology, Vol. 120, No. 5 (September 2012), pp. 557-574Published by: The University of Chicago PressStable URL: http://www.jstor.org/stable/10.1086/666944 .Accessed: 26/09/2012 12:41

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to TheJournal of Geology.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=ucpress

http://www.jstor.org/stable/10.1086/666944?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

[The Journal of Geology, 2012, volume 120, p. 557–574] � 2012 by The University of Chicago.All rights reserved. 0022-1376/2012/12005-0005$15.00. DOI: 10.1086/666944

557

Transformation of Andalusite to Kyanite in the Alpujarride Complex(Betic Cordillera, Southern Spain): Geologic Implications

Antonio Sanchez-Navas,1,* Rita de Cassia Oliveira-Barbosa,1

Antonio Garcıa-Casco,1 and Agustın Martın-Algarra2

1. Departamento de Mineralogıa y Petrologıa, Instituto Andaluz de Ciencias de la Tierra, Universidad deGranada, Consejo Superior de Investigaciones Cientıficas, E-18071 Granada, Spain; 2. Departamento de

Estratigrafıa y Paleontologıa, Instituto Andaluz de Ciencias de la Tierra, Universidad de Granada,Consejo Superior de Investigaciones Cientıficas, E-18071 Granada, Spain

A B S T R A C T

The crystal growth features of andalusite and the transformation of andalusite to kyanite allow recognition of pre-Alpine and Alpine tectonometamorphic histories in the metapelites of the Alpujarride Complex. Two types of mineralsegregations occur in relation to the andalusite r kyanite transformation. One type of segregation consists of a mantleof muscovite around a pre-Alpine andalusite chiastolite core that is partially transformed to fine-grained Alpinekyanite within a matrix particularly rich in biotite � quartz. The second type of segregation consists of Alpinemuscovite � kyanite domains that form after dissolution of pre-Alpine andalusite � biotite domains within thematrix. The formation of these two types of mineral segregations involves similar reactions between the fluid andthe local mineral assemblage. These reactions progress simultaneously. Each of them acts as source (or sink) for theions, and intermediate mineral phases are consumed (or produced) by the other reactions, so that a combination ofindividual reactions produces the andalusite r kyanite net reaction. This reaction is catalyzed by the muscovite andbiotite of the matrix, whose dehydration provides the chemical driving force needed to break Si-O bonds for theandalusite r kyanite transformation. Surfaces perpendicular to F-type {110} faces of the andalusite chiastolites as-sociated with layeritic crystal growth mechanisms constitute fast reaction pathways for the andalusite r kyanitereaction. The transformation of pre-Alpine andalusite to Alpine kyanite constitutes the first solid textural evidenceof the existence of a polymetamorphic history in the rocks of the Alpujarride Complex (Betic Cordillera, southernSpain).

Introduction

The textural relations among the Al2SiO5 poly-morphs and other commonly related key meta-morphic minerals—such as staurolite, biotite, mus-covite, cordierite, and garnet—are an importantsource of information to establish the PT paths andtectonometamorphic evolution of metapelites (e.g.,Rumble 1973; Holdaway 1978; Kerrick and Spear1988; Mottana et al. 1990; Pattison 1992; Martınezet al. 2001; Whitney 2002). The coexistence of theAl2SiO5 polymorphs in the same rock sample is acommon feature of many metamorphic terrains.Such coexistence normally represents a nonequi-librium feature favored because of small entropy

Manuscript received October 14, 2011; accepted May 29,2012.

* Author for correspondence; e-mail: [email protected].

change of the polymorphic transformations. There-fore, the activation energy barrier is difficult toovercome, leading up to strong metastability of theinvolved polymorphs (Walther and Wood 1984; Ker-rick 1990).

The occurrence of And � Sil and of Ky � Sil(abbreviations after Whitney and Evans 2010) arequite common. In both cases, sillimanite typicallyforms later in the metamorphic evolution (i.e.,And/Ky r Sil) as a consequence of prograde meta-morphism under low-PT and medium- to high-PTgradients, respectively. In the same way, the Ky r

And transformation is easy to account for during aprograde metamorphic evolution at relatively lowP. In addition, Ky r Sil transformation has beenproposed for the Alpujarride Complex in relation

mailto:[email protected]

558 A . S A N C H E Z - N A V A S E T A L .

to isothermal decompression (Garcıa-Casco et al.1993; Garcıa-Casco and Torres-Roldan 1996, 1999).The kinetic energy barriers of “retrograde” trans-formations of Sil r And, Sil r Ky, and And r Kyare more difficult to overcome during a single meta-morphic cycle. Consequently, these transforma-tions have been less commonly described and rarelydiscussed (however, for case studies of the Ky r Silr And transformations occurring along one meta-morphic path, see Garcıa-Casco and Torres-Roldan1996, 1999; Kim and Ree 2010).

Diverse natural examples of the And r Ky trans-formation are described in the literature (e.g., Whit-ney 2002; Sepahi et al. 2004; Kim and Ree 2010).However, this reaction has received different pet-rogenetic explanations. In the Chiwaukum schists,North Cascades, Washington, Evans and Berti(1986) interpreted this transformation in relationto a polymetamorphic history. These rocks wereaffected by (1) a dynamic contact metamorphicevent associated with the emplacement of theMount Stuart batholith, which produced andalusitechiastoliths in the surrounding thermal aureole;this was followed by (2) a Barrovian regional meta-morphic event that produced staurolite and kyanitepostdating andalusite. In NW Iberia, Martınez et al.(2001) interpreted this transformation according tothe existence of a prograde event within the an-dalusite stability field, followed by retrogression inthe kyanite field at somewhat higher P. Recently,Kim and Ree (2010) have described this transfor-mation in metapelites that followed a clockwiselow-T and medium-P path related to a rapid con-traction event.

The And r Ky transformation studied herecomes from medium- to high-grade metapelites be-longing to the upper tectonic units of the Alpujar-ride Complex, Betic Cordillera, southern Spain (fig.1), where the three Al2SiO5 polymorphs commonlycoexist in the same rock (Garcıa-Casco and Torres-Roldan 1996, 1999). This coexistence was ex-plained according to the late-orogenic isothermaldecompression stage of Alpine age that evolvedfrom the Ky through the Sil to the And stabilityfields (Garcıa-Casco et al. 1993; Garcıa-Casco andTorres-Roldan 1996, 1999). Such interpretation wasbased on textural/structural relations (prekine-matic Ky, synkinematic Sil, postkinematic And,textural replacement of Ky by And, and inclusionsof Ky within And) and in agreement with similartextural-mineralogical development in metapelitesof Mesozoic deposition age of some Alpujarrideunits (e.g., Torres-Roldan 1974, 1978; Azanon andCrespo-Blanc 2000; De Oliveira Barbosa 2010). Wedescribe here, for the first time, the And r Ky trans-

formation in Alpujarride rocks. Special attention isdevoted to the textural features related to the re-action mechanisms involved in this transformationand, in particular, to the crystal growth features inandalusite chiastolites. Our findings bear importantimplications for the geologic evolution of the Al-pujarride Complex, since they may represent directevidence of pre-Alpine metamorphic evolution.

Geological Setting

During the Alpine orogenic evolution of the West-ern Mediterranean Alpine chains, a metamorphicbelt, referred by Bouillin et al. (1986) as theAlKaPeCa (Alboran-Kabylia-Peloritani-Calabria)belt, was formed at the beginning of the Miocene.This belt resulted from the Alpine collision be-tween the Iberian, African, and Mesomediterraneanplates formed during the Mesozoic breakup of thePangea supercontinent (Martın-Algarra 1987; Guer-rera et al. 1993; Martın-Algarra et al. 2009b). Theorogenic belt was later fragmented during and afterthe Late Miocene, when the Western Mediterra-nean basins opened, thus producing the outward(foreland) migration of its fragments (Martın-Algarra et al. 2000). One of these fragments con-stitutes the Betic Internal Zones, which, togetherwith the Internal Zones of the Rif (Morocco) andthe basement of the Alboran Sea, form the so-calledAlboran Domain (Balanya and Garcıa Duenas1987). The westward migration of this domain andits collision against the surrounding basins andmargins produced the Gibraltar Arc (fig. 1A) andthe orogenesis of the External and Flysch Domainsof the Betic Cordillera (fig. 1B).

The Betic Internal Zones are formed by a com-plex stack of Alpine thrust nappes (fig. 1B) of (es-sentially) Miocene age (quite probably preceded bylate Cretaceous to early Tertiary tectonometa-morphic events; see Puga et al. 2000, 2002; refer-ences therein). This stack was strongly thinned bymid- to late-Miocene orogenic extensional collapse(Balanya et al. 1993; Platt et al. 1996, 1998, 2003;Argles et al. 1999a) coeval with the disintegrationof the AlKaPeCa belt (Martın-Algarra et al. 2000)and is unconformably covered by Burdigalian de-posits (Vinuela Group; fig. 1C; Martın-Algarra1987; Vera 2004; references therein). In addition toa group of frontal units (Martın-Algarra et al. 2009a;Mazzoli and Martın-Algarra 2011; referencestherein), the nappe stack is classified, from bottomto top, in the Nevadofilabride, Alpujarride, andMalaguide Complexes (fig. 1B). The rocks of theAlpujarride Complex are typically grouped inLower, Intermediate, and Upper Alpujarride units,

Journal of Geology T R A N S F O R M A T I O N O F A N D A L U S I T E T O K Y A N I T E 559

Figure 1. A, Betic Cordillera in its context within the Western Mediterranean Alpine belts. B, Tectonic sketch mapof the Betic Cordillera. C, Alpujarride Complex in the study area, with location of the sampled site within the UpperAlpujarride Torrox Unit (modified from Martın-Algarra et al. 2009b; Vera 2004).

with slightly different lithostratigraphy and meta-morphic grade of Alpine age (Vera 2004; referencestherein). A typical Alpujarride succession isformed, from bottom to top, by dark-colored gra-phitic micaschists, quartzites, and gneisses of pre-Mesozoic protolith age, followed by light-coloredschists and quartzites of assumed Permian(?)-Triassic protolith age, in upward stratigraphic tran-

sition to Triassic and locally younger carbonates(Vera 2004; references therein).

The Upper Alpujarride tectonic units were af-fected by Alpine plurifacial metamorphism char-acterized by an early intermediate PT gradient fol-lowed by a low to very low PT gradient interpretedas resulting from near isothermal decompressionrelated to the Miocene extensional collapse of the


Betic Belt (e.g., Torres Roldan 1974, 1981; Balanyaet al. 1993, 1997; Garcıa-Casco et al. 1993; Monieet al. 1994; Garcıa-Casco and Torres-Roldan 1996,1999; Platt et al. 1996, 1998, 2003; Argles et al.1999a). The decompression would be related to thedevelopment of a foliation predating the mega-scopic brittle extensional tectonic structures thataffected the Alpujarride Complex in the Miocene(Azanon and Crespo-Blanc 2000).

Geochronological data indicate that the peak ofthe low- to medium-PT event in the Alpujarrideunits was reached around 25–19 Ma (Priem et al.1979; Zeck et al. 1989; Monie et al. 1991) and wasfollowed by cooling and uplift at extremely fastrates that have been interpreted in relation to theMiocene extensional collapse of the whole meta-morphic belt (e.g., Zeck et al. 1992; Monie et al.1994; Platt and Whitehouse 1999; Platt et al. 2003).According to the current literature (e.g., Garcıa-Casco and Torres-Roldan 1996, 1999; Azanon et al.1998; Azanon and Crespo-Blanc 2000), the mineralparagenesis formed in medium-grade metapelitesduring the early medium-PT event of Alpine ageconsists of garnet � kyanite � staurolite-bearingassemblages associated with the development of anold foliation predating the main foliation, both in-terpreted as developed during Tertiary times.

The studied samples come from rocks of pre-Mesozoic depositional age of the Torrox Unit,which is an Upper Alpujarride tectonic unit locatedin the central sector of the Betic Cordillera (fig. 1C).The Torrox Unit bears, at its base, a kilometer-sizedgneissic body (Torrox Gneiss; Garcıa-Casco et al.1993) surrounded by a monotonous series ofgraphite-bearing micaschists with fibrolite � gar-net � staurolite � kyanite � andalusite � cor-dierite that grade upward into graphite-bearing mi-caschists with garnet � staurolite � kyanite �andalusite � cordierite. According to Garcıa-Cascoet al. (1993) and Garcıa-Casco and Torres-Roldan(1996, 1999), the thermobarometric conditions inthe metapelites of the Torrox Unit were 10–12 kbarand 550�–650�C during the first metamorphicevent, the paragenesis formed during the main fo-liation contained Bt � Sil (in fact, fibrolite), and alltypes of assemblages were overprinted by post-kinematic And-Bt � Crd assemblages formed at !4kbar and similar temperature.

Evidence for pre-Alpine tectonometamorphicevents has been found in high-grade (migmatitic/gneissic) rocks of some Alpujarride units, includingthe Torrox Unit (Acosta-Vigil 1997; Zeck andWhitehouse 1999, 2002; Zeck and Williams 2001).In particular, Variscan ages obtained in bothgneissic and metapelitic rocks of the Torrox Unit

have been interpreted as resulting from metamor-phic and magmatic events (Zeck and Whitehouse1999, 2002). However, the Variscan tectonometa-morphic processes are not yet well constrained be-cause of the strong Alpine metamorphic overprintthat affected this unit. Consequently, most authorshave not dealt with the pre-Alpine events in theAlpujarride complex (see, however, Boulin 1970;Sanchez-Navas 1999).

Material and Methods

The graphite-bearing micaschists of the TorroxUnit typically contain abundant Al2SiO5 poly-morphs, mainly andalusite porphyroblasts, whichwere carefully studied both in the field and underthe optical microscope. Macroscopically, the(sub-)centimetric chiastolitic andalusite crystalsappear strongly deformed, and as presented below,some of them are replaced by fine-grained kyanite.Detailed analytical and textural electron micros-copy studies were performed in selected andalusitechiastolites in which peculiar crystal growth/re-placement features are well preserved.

Backscattered electron images were obtainedwith a Leo 1430VP scanning electron microscope,coupled with an energy-dispersive x-ray spectros-copy (EDX) microanalysis (Inca 350, ver. 17; OxfordInstruments; Centro de Instrumentacion Cientı-fica, University of Granada). Mineral compositionand x-ray maps were acquired with the same ma-chine operated at 500 pA filament current, 20 keVbeam energy, and 10 eV/ch resolution for pointanalyses and 1 nA filament current and 20 eV/chresolution for x-ray map acquisition. High-resolu-tion maps (1024 pixels # 768 pixels) were obtainedfor andalusite porphyroblast with 500 frames anda dwell time of 10 ms (16-h acquisition). Phasemaps were compiled from elemental distributionmaps by applying the Phasemap tool implementedin the Inca 350 version used in this work. Phasemap software allowed identification of phases usingternary element plots of specific pixel informationfrom x-ray maps.

A Leo GEMINI-1530 electron microscopeequipped with an Inca Crystal detector was usedfor the electron backscattered diffraction (EBSD)study. Software MATTER was employed, and theexperimental conditions were as follows: 15–20crystallographic planes of And and Ky were selectedfrom the crystallographic files, six bands were de-tected in the diffraction patterns, and a minimumof four bands were indexed by the program to iden-tify the crystallographic orientation at the pinpointanalyses. Crystallographic information files (CIFs)


Figure 2. Field views of the Torrox dark schists. A, Aggregates of randomly oriented andalusite prisms on the foliation(probably S2) within a graphite-rich metapelitic band. B, Stretching lineation defined by elongated andalusite por-phyroblasts surrounded by whitish aureoles of sericite (arrow). C, Stretched (boudinated and cracked) pink andalusiteprism (arrow), with cracks filled up with Qz and Ms aggregates, within mylonitized graphite-rich schists. D, Microfoldsrelated to the development of the S3 foliation, which affects both the S2 foliation and post-S2 quartz and albite veins.E, Strongly sheared quartz and albite veins, including crushed and reoriented pink andalusite crystals within myloniticschists. A color version of this figure is available in the online edition or from the Journal of Geology office.

come from the American Mineralogist CrystalStructure Database. The diffraction patterns of Andand Ky were indexed using their corresponding CIFfiles and then integrated to perform phase maps,orientation maps, and pole figures. Orientationmaps may be viewed along three directions in thesample: the normal to the polished thin section,the transverse direction (strike), and the rolling di-rection (dip), with the two later parallel to the planeof the section. In the studied samples, the trans-verse direction corresponds to the trace of the mainor more pervasive foliation in the sample. The ori-entation maps represent the crystallographic direc-tions parallel to one of the three former directionsin the sample. The colors assigned to point analyses

in the orientation maps of kyanite are white for[001]Ky, green for [100]Ky, and blue for [010]Ky.

Field Relations

The studied samples are representative of the meta-sedimentary Alpujarride rocks whose pre-Mesozoicprotoliths come from a pelitic-psammitic rhythmicsuccession rich in organic matter. During meta-morphism and deformation, the succession wastransformed into decimeter- to meter-thick alter-nating bands of strongly folded and foliated me-dium- to fine-grained quartzites and fine-graineddark (graphite-rich) metapelites that contain abun-dant millimeter- to centimeter-sized andalusite

Figure 3. A, Augen-like And chiastolite elongated following S3 planes within a fine-grained mylonitized shearedmatrix. B, And overgrowing a previous St crystal, the whole being surrounded by the S3 foliation, which is definedby a fine-grained and sheared matrix mainly composed of Fi, commonly intergrown with Bt. C, Deformed And crystalpartially transformed into Fi along shear planes following the S3 foliation. D, Backscattered electron image of thesquared area in C, showing the transformation of the broken parts of the And crystal into Fi. Note that this trans-formation starts by penetrating the neoformed Fi along the exfoliation planes of the And crystal to finally becomethe resulting Fi, reoriented along the foliation (shear) planes. E, Deformed And prisms, reoriented along the S3 foliation.Note the crushed aureole that surrounds And prisms. F, Close-up of the right-hand part of the larger And prism shownin E but under crossed nicols, making evident the transformation of the former And single crystal into a mosaic ofsubgrains due to the strong strain that affected the rock.


Figure 4. A, Skeletal aggregate of And, Bt, and Qz within the schistose matrix that changes into fine-grained Kyplus Ms domains in the sheared zone visible in the upper right part of the image. B, Same as A, crossed nicols. C,Close-up of the Ms � Ky domains. D, Same as C, crossed nicols.

porphyroblasts. In the studied samples, the anda-lusite crystals appear as prisms, commonly chias-tolites, up to 1 cm thick and several centimeterslong, especially in the lower part of the succession.

The andalusite prisms commonly occur ran-domly oriented on a foliation, hereafter named S2

(fig. 2A). However, a closer examination of thelarger crystals reveals that the chiastolites are de-formed and surrounded by a whitish aureole of ser-icite (arrow in fig. 2B). In some cases, the andalusitechiastolites are strongly deformed and reoriented(fig. 2B–2E), showing boudinage (fig. 2C), foldingand microcrenulation (fig. 2D), and even crushingand elongation on a new foliation, hereafter namedS3. The foliation S3 may transpose the earlier foli-ation S2, with the And prisms defining a NNW-SSE-to N-S-oriented stretching lineation (fig. 2B).

In addition, the schists typically include centi-meter-thick veins filled with quartz and albite,with subordinate muscovite, which also contain

millimeter- to centimeter-thick, clean and pink,prismatic andalusite crystals (fig. 2D, 2E). Theseveins crosscut the foliation S2 (fig. 2D), but a de-tailed observation shows that, as well as their schis-tose matrix, the veins and the included pink an-dalusite prisms are deformed (folded and stretched;fig. 2E) along a new mylonitic foliation (S3).

Textural Features of the Andalusite-Bearing Schists

Andalusite chiastolites preserve their crystalgrowth features in spite of later deformation D3 (fig.3A). And includes corroded St grains (fig. 3B) andis surrounded by fibrolite (Fi). Fi forms fibers andneedle sheafs mainly related to the youngest foli-ation (S3), which it defines together with Bt and Ms(fig. 3B). In some cases, Fi appears as bunches offibers that crosscut And porphyroblasts along sur-faces of crystal cleavage (fig. 3C, 3D). Fi is exclu-


Figure 5. A, Two deformed And chiastolites within a fine-grained matrix under plane parallel light. The lower crystalshows an Ms-rich aureole, whereas the upper chiastolite does not. Note the existence of dirty areas defining thechiastolite arms. B, Image with crossed nicols of the lower left-hand chiastolite shown in A, which appears nowconstituted by a mosaic of subgrains due to deformation. C, Close-up of the large chiastolite in the lower part of A,with location of the selected areas for the electron backscattered diffraction study shown in figure 7. These areasshow indented boundaries with the surrounding Ms-rich aureole and are partially transformed to Ky, which penetrateswithin the chiastolite along a direction perpendicular to the {110} faces of the And prism.

sively present at the base of the metapelite suc-cession, and it disappears upward.

Upward in the succession, where Fi disappears,the foliation S3 is defined preferentially by fine-grained Ms, Bt, and Qz. As mentioned above, thelargest And crystals appear stretched and stronglydeformed by D3. This is evident from fragmentation

of the And porphyroblasts into a mosaic of sub-grains (fig. 3E, 3F), with an important subgrain sizereduction mostly at porphyroblasts boundaries, al-though the previous chiastolite growth features arecommonly preserved (fig. 3A). Skeletal And withinthe schistose matrix is associated with Bt fromwhich it grew, forming And � Bt domains (fig. 4A,


Figure 6. A, Backscattered electron image of the chiastolite shown in figure 5C and of the mineral segregates aroundit. B, Triangular diagram indicating the composition of the phases represented in C, which correspond to Al-silicate(And � Ky; red), Qz (pink), Ms (light green), Bt (dark green), or mixtures of Al silicate � Qz (blue). C, Phase map ofthe chiastolite and of its surrounding aureole and enclosing matrix.

4B). Fine-grained Ms � Ky domains (fig. 4C, 4D)also occur associated with the S3 foliation alongshear zones crosscutting the And � Bt domains.

Two deformed And chiastolites are visible in fig-ure 5A. The largest chiastolite is strongly flattenedand crushed, resulting in a mosaic of subgrains (fig.5B). It has its lower arm sheared and shows a light-colored aureole (fig. 5A, 5C). This aureole is absentin the smallest chiastolite (fig. 5A), which shows adirty and fine-grained rim. The light-colored au-reole of the largest chiastolite is rich in Ms, and itsurrounds only the two right-hand arms of the chi-astolite, which show indented boundaries with thesurrounding matrix. Within the aureole, Bt is ab-sent, whereas it concentrates in the surroundingmatrix. Dirty dark rims with high relief are visiblein the left-hand arm of the same crystal, and they

are similar to the rim of the smaller chiastolite infigure 5A and, especially, to that of the augen-like,clearly sheared And chiastolite shown in figure 3A.The Ms forming the aureole penetrates within thechiastolite arms along the indented boundaries, andthe arms of the chiastolite between the Ms inden-tations are replaced by Ky (fig. 5C).

Electron Microscopy Study of theReaction Texture

The main results obtained from scanning electronmicroscope, EDX, and EBSD studies of chiastolitecrystals are summarized in figures 6 and 7. Thetextural relations among the And, Ms, Bt, Qz, andKy within the metamorphic segregations (in thesense of Fisher [1970]) associated with chiastolites


Figure 7. Electron backscattered diffraction study of the And r Ky transformation. A, Transmitted plane parallellight optical image of the zone 1 squared in figure 5C, with indication of the parts of the chiastolite that have beenpartially or totally replaced by Ky and those where the And is yet preserved. B, Electronic image of the same zoneas A, with indication of the trace of the {110} prismatic faces of the And. Arrows point to the Ms-rich bands interlayeredwith Ky and perpendicular to the prismatic faces of the former And crystal. C, Orientation map of the Ky in the areashown in A following the rolling direction in the sample. D, Phase map of the area shown in A showing the Ky. E,Phase map of Ky (pink) that replaces And (yellow) in zone 2 of the chiastolite shown in figure 5C. F, (001) or equivalentpole figure of the deformed And chiastolite, showing at least three (1, 2, 3) different crystallographic orientationswithin the formerly unique single crystal. G, (010) pole figure of the And. H, (100) pole figure of the And.

are clarified by the EDX and EBSD analyses. Thebackscattered electron image (fig. 6A) clearly showsthe chemical zonation related to the formation ofmetamorphic segregations around chiastolites. TheEDX analysis and x-ray maps allow establishing thedistribution of five different compositional phaseswithin and around the chiastolite (fig. 6B, 6C). Theselected phases are (1) the Al2SiO5 polymorphs de-fining the chiastolite, (2) the Qz-rich zones formingmainly its dirty dark rims, (3) the Ms-rich phase ofthe aureole, and (4) the Bt and (5) the Qz of thesurrounding matrix.

The EBSD analysis confirms the existence of Ky(fig. 7A) within the upper and right-hand arms ofthe chiastolite. The Ky clearly replaces the And ofthe chiastolite, and this replacement progresses fol-lowing intracrystalline surfaces perpendicular tothe {110} crystal faces of the And from the boundaryof the chiastolite toward its interior (fig. 7B). Theorientation map of the Ky along the rolling direc-tion indicates that the fine-grained aggregate of Kyreplacing And does not show any preferential ori-entation (fig. 7C–7E). In addition, the former Andchiastolite has been transformed into a polycrys-talline aggregate of And, as clearly visible in thepole figures (fig. 7F–7H).

Discussion

Crystal Growth. The above described chiastoliticAnd porphyroblasts commonly show some com-mon peculiar crystal growth features that can besummarized as follows: (1) the development ofpolyhedral forms where edges and corners seemsto be absent, (2) the preferential crystal growth nor-mal to the flat faces (F-faces; Hartman and Perdock1955), and (3) the formation of cross-like inclusionpatterns. Such crystal growth features are producedby a common spiral-growth mechanism that canbe accounted for by low supersaturation or low un-dercooling conditions (e.g., Kuroda et al. 1977; Kirk-patrick 1981).

The model proposed in figure 8A explains theobserved inclusion pattern perpendicular to F-faces

in And chiastolites. The starting point of the modelis the presence of numerous screw dislocationsemerging at the center of an F-face, so that thegrowing steps migrate from the center of the faceto the edges (fig. 8A, 1). The layers spread outwardfrom the center of the emergent screw dislocation,pushing impurities outside of the crystal (fig. 8A,2). This crystal growth mechanism will be hereafterreferred to as layeritic growth (Kerrick 1990, p. 304),the opposite of dendritic growth, where more layersare generated at the corner or edges and spread to-ward the center of the crystal face (Kirkpatrick1981). The existence of numerous screw disloca-tions is necessary to explain why the layeriticgrowth mechanism proposed here produces the Qz-rich inclusion pattern observed in And. Accordingto this mechanism, the layers related to indepen-dent adjacent dislocations converge, thus trappingimpurities inside the crystal (fig. 8A, 3). The quartzinclusions, defining intracrystalline surfaces per-pendicular to prismatic F-faces within And crystals,result from dissolution and reprecipitation of theQz of the matrix during crystal growth.

Because of the strong crystallization pressure inthe direction normal to the F-faces, the completedissolution of the matrix minerals at the advancingAnd crystal faces tends to drive the impurities awayat their front, thus producing the accumulation ofgraphite that is observed at the ends of the skeletalarms of And chiastolites (fig. 8B). The formation ofnonplanar crystal surfaces (spherulites, dendrites,and hopper crystals; fig. 8B) is typical of crystalgrowth under high supersaturation conditions(Sunagawa 1987). Nevertheless, the crystal surfacesare planar when crystals grow at lower supersatu-ration because of the presence of numerous emer-gent screw dislocations. As an extreme case for fastgrowth rates under high thermal activation energy,the resulting polyhedron lacks their edges and cor-ners (fig. 8B). These crystal growth features havebeen preferentially described in relation to the or-igin of chiastolite (Rast 1965; Spry 1969; Petreus1974; Rubenach and Bell 1988; Kerrick 1990; Riceand Mitchell 1991), but they have also been ob-


Figure 8. A, Sequential model to explain (1) the location of screw dislocations (sites of spiral growth) at the centerof F-faces, (2) the normal and lateral growth of F-face in relation to layeritic mechanisms, and (3) the developmentof inclusion patterns perpendicular to the crystal faces. B, Dominant mechanism during growth of a crystal face,depending on supersaturation. {110} primatic faces of And resulting from a layeritic growth mechanism. V, growthrate; S, supersaturation degree. Vertical line separates surface nucleation-controlled growth at relatively large super-saturation from growth controlled by screw dislocations emergent at the center of the F-faces at low supersaturation(modified from Kuroda et al. 1977).

served in the trapiche emerald (Nassau and Jackson1970), ruby (Sunagawa et al. 1999), and garnets (Wil-bur and Ague 2006).

Reaction Pathways of the And r Ky Transformation.The textural data presented above demonstrate thatthe studied And r Ky transformation progressesfrom crystal (chiastolite) boundaries toward theirinterior following intracrystalline surfaces perpen-dicular to the {110} And crystal faces (zones 1 and2 in fig. 5C). This constitutes a natural example ofreaction controlled by the crystal growth featuresof the reactant mineral (And). Intracrystalline sur-faces perpendicular to crystal faces are much morereactive than stable crystal faces because they arenot structurally controlled and, consequently, theyfavor the beginning of the studied reaction (fig. 7).In this sense, two less-deformed arms of the chi-astolite, preserving original crystal growth features,develop the And r Ky transformation (figs. 5C, 6A,6C). The textural and compositional analysis byEBSD and EDX of the And chiastolites indicatesthat Ms precipitated and Bt dissolved around re-acting And to produce fine-grained Ky and that thistransformation was controlled by the inherited

crystal growth features of the previous And crystal(figs. 6, 7). A model for the And r Ky reaction atthe chiastolite boundaries (in particular, zones 1and 2 in fig. 5C) is presented in figure 9A–9C. TheAnd r Ky transformation is catalyzed by Msthrough the reactions (KASH system)

3 3� �And � Qz � 1K � 2H r 1Ms, (1)

2 2

3 3� �1Ms r Ky � Qz � 1K � 2H . (2)

2 2

The sum of the two reactions is

3 3And r Ky. (3)

2 2

In these elementary and overall reactions (in thesense of Carmichael [1969], Fisher [1970], andFisher and Lasaga [1981]), the oxygen is not bal-anced, and potassium and hydrogen are introducedas mobile components (K�, H�) derived from Msdissolution. To explain the formation of an Ms-richaureole, the Bt of the matrix surrounding the Andchiastolite must be dissolved from the aureole


Figure 9. Model proposed to explain the And r Ky transformation. A, Initial stage of development of the And r Kyreaction along discontinuities perpendicular to {110} faces within the And chiastolite. B, Dissolution of And to formthe Ms catalyzer at the reaction front through reaction (1). C, Formation of Ky through reaction (2). D, Model thatexplains the formation of the Ms-rich metamorphic segregate of the aureole, with indications of the sites wherereactions (4) and (5) should account. E, Sketch model explaining the location of reaction (4) within the schistosematrix according to the texture defined by the And � Bt domains and Ms � Ky domains, as observed in figure 4Cand 4D.

through the following reaction in the KFASH sys-tem (cf. figs. 6C, 9D):

5 3 2�And � 1Bt r Ky � 1Ms � 1Qz � 3Fe . (4)2 2

This reaction also occurs within the matrix to formMs � Ky domains from dissolution of And � Btdomains (cf. figs. 4C, 4D, 9E).

Neoformed Bt is segregated toward the matrixthat surrounds the Ms aureole (see fig. 6B, 6C)through the reaction

1 2 4� � 2�Ms � 2Qz � K � H � 3Fe r 1Bt. (5)

3 3 3

Considering Bt as a catalyzer together with Ms, theoverall reaction And r Ky is obtained by the fol-lowing sum of elementary reactions:

reaction (3) p

2reaction (2) � reaction (4) � reaction (5).

3

From these reactions, it becomes evident that Msand Bt are catalyzers that allow the And r Ky trans-formation, since they are simultaneously con-sumed and regenerated.

The inclusion of ionic species in the reactionsimplies the presence of water-rich fluids producedby dehydration and/or dehydroxylation of hydratedminerals, such as Ms and Bt. The soluble Fe2� pro-duced during the dissolution of Bt (together withother ionic species, such as Mg, Al, and Ti) is trans-ported from the Ms-rich aureole to the surroundingmatrix to produce a new Bt (�ilmenite). Similarmetamorphic processes are well known in the lit-erature (Carmichael 1969; Fisher 1970, 1973). Ac-cording to Lasaga (1986), local equilibria, such as


Figure 10. Catalysis of the Ms-mediated And r Kytransformation. A, Diagram of the activation energyneeded for Si-O bonds breaking through the progress ofthe reaction. Ea, activation energy without catalyzer; ,′Ea

activation energy in the presence of a catalyzer. B, For-mation of the K2O basic oxide—or of the K�.e� pair—from Ms dissolution. C, Formation of hydroxile ionswhen the basic oxide is dissolved in water produced byMs dehydration. Hydroxile ions formed from the acid-base reaction in C or directly from dehydroxylation ofthe Ms represent a high-energy form of the water, andtherefore their presence increases the energy of the sys-tem. D, Lengthening and breaking of the Si-O bonds ofsilicates are due to the increase in chemical energy afterreaction of Al2SiO5 with alkaline solutions. The processis represented by a simple acid-base reaction similar tothose proposed for the depolymerization in silicates. Acolor version of this figure is available in the online edi-tion or from the Journal of Geology office.

those represented by the subreactions above, leadto domains with a selective mineralogy. Local equi-librium conditions between a fluid and the mineralspecies of each domain within the rock determinea diffusion-controlled process of the reaction rate.

The surface reaction processes described here forthe And r Ky transformation should be fast be-cause their catalyzation by Ms and Bt provides wa-ter that allows an Si-O bond breaking through asolid-solid solution-mediated transformation (fig.10A). The role of water in Si-O bond breaking iswell known for silicate depolymerization (e.g., Gill1997, p. 180). The water may acts as a base (in thesense of Lewis), since it constitutes an electron do-nor to the system. The same accounts for the elec-tropositive metallic elements, which also favor de-

polymerization of silicate structures through theSi-O bond breaking. In the studied case, K2O andH2O are produced by dissolution of Ms and Bt (fig.10B, 10C). Breaking of Si-O bonds accounts throughthe population of high-energy empty antibondingmolecular orbitals (Burdett 1980, p. 265; Tosselland Vaughan 1992). The population of the Si-O an-tibonding orbitals can be measured by the numberof valence electrons per atom (electron count inBurdett 1980, p. 265). The electron count is definedby the ratio between the number of the electronsof the anions divided by the number of O and Siatoms of the silicate anion group (the most elec-tronegative atoms in the silicate structure).

In the And r Ky transformation, K2O acts as abase during its reaction with H2O and provides hy-droxile ions to the solution. Hydroxile ions bearthe chemical energy (measured by the electroncount; fig. 10C; modified from Albright et al. 1984,p. 124) that is needed for Si-O bond breakingthrough a new acid-base reaction, and finally waterrecovers its low-energy state (fig. 10D). This ex-plains why the neoformed Ky after And dissolutionalways coexists with Ms, which is the catalyzer ofthe And r Ky transformation.

Evidences of a Polymetamorphic History:Geological Implications

The And r Ky transformation is recognized for thefirst time in the Alpujarride Complex in this studyboth in And chiastolites and in the And � Bt do-mains of the S3-sheared schistose matrix. As dis-cussed below, these features reveal a deformation-metamorphic history with two main stages: (1)static first stage related to a prograde blastesis oflarge And porphyroblasts by spiral growth mecha-nisms and (2) dynamic second stage related to adeformational history that produced mylonitiza-tion and grain size reduction of the previous largeminerals and partial transformation of And chias-tolites to fine-grained Ky. The first stage here isinterpreted as pre-Alpine; the second one is inter-preted to be Alpine.

When well preserved, large porphyroblasts ofAnd postdating the foliation S2 show evidence oflayeritic crystal growth mechanisms under non-stress-induced (static) conditions. This type of fastcrystal growth can be accounted for by low super-saturation conditions related to slow diffusion ratesand therefore low matter supply. This reveals thatthe blastesis of the And porphyroblasts was con-trolled by thermally activated processes, probablyunder a high PT gradient.


The second part of the tectonometamorphic his-tory was clearly related to the mylonitic defor-mation of the And porphyroblasts (And chiasto-lites, large pink andalusite prisms withinquartz-albite veins, and skeletal And within theschistose matrix). This And is clearly affected byat least one younger deformation event that is cer-tainly Alpine in age. It produced strong folding andmylonitization of the successions as well as thedevelopment of the Alpine foliation, here namedS3.

The formation of S3 was associated with theblastesis of fine-grained Ms, Bt, Ky, and Fi, the latterfound only in the lowest part of the studied Al-pujarride succession. A similar replacement of Andchiastolites by fine-grained Ky is commonly foundin the Nevadofilabride Complex. There, it is inter-preted as resulting from a pre-Alpine tectonometa-morphic evolution that produced large porphyrob-lasts of andalusite (and also of St, Grt, and Ctd) thatwere transformed in fine-grained Alpine Ky, St,Ctd, and Grt (Estevez and Perez-Lorente 1974; Go-mez-Pugnaire and Sassi 1983; Puga et al. 2002).

The above-mentioned evolution is here inter-preted as (early) Alpine in age, in agreement withthe commonly accepted models for the early tec-tonometamorphic evolution of the Upper Alpujar-ride Units (Torres-Roldan 1979, 1981; Monie et al.1991; Tubıa and Gil-Ibarguchi 1991; Tubıa et al.1997; Sanchez-Rodrıguez and Gebauer 2000; Maz-zoli and Martın-Algarra 2011). This early Alpineevolution is usually interpreted in relation to a sub-duction-related P increase. This higher-P event wasfollowed by exhumation, with development of low-P and high-T conditions and partial melting in

some Alpujarride Units—but not in the studiedrocks—during the Early Miocene (e.g., Zeck et al.1989; Monie et al. 1994; Platt and Whitehouse1999; Platt et al. 2003, 2006; Mazzoli and Martın-Algarra 2011). The existence of a polymetamorphichistory in the studied Alpujarride rocks is also inagreement with the chemical zoning of Grt por-phyroblasts with overgrowths belonging to the Up-per Alpujarride Los Reales Unit, which Argles etal. (1999b) dated by Sm/Nd at Ma. This235.1 � 1.7age can be interpreted as an intermediate age be-tween pre-Alpine (possibly Variscan) and Alpine(probably Lower Miocene) tectonometamorphicevents.

To conclude, this is the first study where theAlpine And r Ky transformation from pre-Alpineandalusite has been observed in the basement (pre-Mesozoic) rocks of the Alpujarride Complex. Theproposed model explains why the (pre-Alpine) Andchiastolites and prisms are so strongly strained andpartially transformed to Ky and Fi during the Al-pine tectonometamorphic evolution.

A C K N O W L E D G M E N T S

This work is supported by grants CGL-2009-09249(Direccion General de Investigacion Cientıfica yTecnica, Spain) and P11-RNM-7067 of the Junta deAndalucıa (Spain) and by the research groups RNM-208, 179, and 3715 (Junta de Andalucıa). We ac-knowledge the critical reading of D. Aerden and R.Compagnoni of a previous version of the manu-script. Critical reading and comments of D. B. Row-ley and an anonymous reviewer are gratefullyacknowledged.

R E F E R E N C E S C I T E D

Acosta-Vigil, A. 1997. Estudio de los fenomenos de fusioncortical y generacion de granitoides asociados a lasPeridotitas de Ronda. PhD dissertation, University ofGranada.

Albright, T. A.; Burdett, J. K.; and Whangbo, M.-H. 1984.Orbital interactions in chemistry. New York, Wiley.

Argles, T. W.; Platt, J. P.; and Waters, D. J. 1999a. Atten-uation and excision of a crustal section during exten-sional exhumation: the Carratraca Massif, Betic Cor-dillera, southern Spain. J. Geol. Soc. Lond. 156:149–162.

Argles, T. W.; Prince, C. I.; Foster, G. L.; and Vance, D.1999b. New garnets for old? cautionary tales fromyoung mountain belts. Earth Planet. Sci. Lett. 172:301–309.

Azanon, J. M., and Crespo-Blanc, A. 2000. Exhumationduring a continental collision inferred from the tec-

tonometamorphic evolution of the Alpujarride com-plex in the central Betics (Alboran Domain, SE Spain).Tectonics 19:549–565.

Azanon, J. M.; Garcıa-Duenas, V.; and Goffe, B. 1998.Exhumation of high-pressure pelites and coevalcrustal extension in the Alpujarride complex (BeticCordillera). Tectonophysics 285:231–252.

Balanya, J. C.; Azanon, J. M.; Sanchez-Gomez, M.; andGarcıa-Duenas, V. 1993. Pervasive ductile extension,isothermal decompression and thinning of the Jubri-que unit in the Paleogene (Alpujarride Complex, west-ern Betics Spain) C. R. Acad. Sci. Paris 316:1595–1601.

Balanya, J. C., and Garcıa-Duenas, V. 1987. Les directionsstructurales dans le Domaine d’Alboran de part etd’autre du Detroit de Gibraltrar. C. R. Acad. Sci. Paris304:929–933.

Balanya, J. C.; Garcıa-Duenas, V.; Azanon, J. M.; and San-


chez-Gomez, M. 1997. Alternating contractional andextensional events in the Alpujarride nappes of theAlboran Domain (Betics, Gibraltar Arc). Tectonics 16:226–238.

Bouillin, J. P.; Durand-Delga, M.; and Oliver, P. 1986.Betic-Rifian and Tyrrhenian arcs: distinctive features,gneisses and development stages. In Wezzel, F. C., ed.The origin of arcs. Amsterdam, Elsevier, p. 281–304.

Boulin, J. 1970. Les zones internes des Cordilleres Be-tiques de Malaga a Motril (Espagne Meridionale). PhDdissertation, University of Paris.

Burdett, J. K. 1980. Molecular shapes. New York, Wiley.Carmichael, D. M. 1969. On the mechanism of prograde

metamorphic reactions in quartz-bearing peliticrocks. Contrib. Mineral. Petrol. 20:244–267.

De Oliveira Barbosa, R. de C. 2010. Estudios texturalesen rocas de alto grado de los Alpujarrides Superior eIntermedio: una aproximacion cristaloquımica. PhDdissertation, University of Granada.

Estevez, A., and Perez-Lorente, F. 1974. Estudio geologicodel sector de Canar-Soportujar, vertiente meridionalde Sierra Nevada. Estud. Geol. 30:515–541.

Evans, B. W., and Berti, J. W. 1986. A revised metamor-phic history for the Chiwaukum schist, north Cas-cades, Washington. Geology 14:695–698.

Fisher, G. W. 1970. The application of ionic equilibria tometamorphic differentiation: an example. Contrib.Mineral. Petrol. 29:91–103.

———. 1973. Nonequilibrium thermodynamics as amodel for diffusion-controlled metamorphic pro-cesses. Am. J. Sci. 273:897–924.

Fisher, G. W., and Lasaga, A. C. 1981. Irreversible ther-modynamics in petrology. Rev. Mineral. Geochem. 8:171–207.

Garcıa-Casco, A.; Sanchez-Navas, A.; and Torres-Roldan,R. L. 1993. Disequilibrium decomposition and break-down of muscovite in high P-T gneisses, Betic Alpinebelt (southern Spain). Am. Mineral. 78:158–177.

Garcıa-Casco, A., and Torres-Roldan, R. L. 1996. Dis-equilibrium induced by fast decompression in St-Bt-Grt-Ky-Sil-And metapelites from the Betic belt (SSpain). J. Petrol. 37:1207–1240.

———. 1999. Natural metastable reactions involving gar-net, staurolite and cordierite: implications for petro-genetic grids and the extensional collapse of the Betic-Rif Belt. Contrib. Mineral. Petrol. 136:131–153.

Gill, R. 1997. Chemical fundamentals of geology. Lon-don, Unwin Hyman.

Gomez-Pugnaire, M. T., and Sassi, F. P. 1983. Pre-Alpinemetamorphic features and Alpine overprints in someparts of the Nevado-Filabride basement (Betic Cordil-leras, Spain). Mem. Sci. Geol. 36:49–72.

Guerrera, F.; Martın-Algarra, A.; and Perrone, V. 1993.Late Oligocene-Miocene syn-late-orogenic succes-sions in western and central Mediterranean chainsfrom the Betic Cordillera to the southern Apennines.Terra Nova 5:525–544.

Hartman, P., and Perdock, W. G. 1955. On the relationsbetween structure and morphology of crystals. ActaCrystallogr. 8:49–52.

Holdaway, M. J. 1978. Significance of chloritoid-bearingand staurolite-bearing rocks in the Picuris Range, NewMexico. Geol. Soc. Am. Bull. 89:1404–1414.

Kerrick, D. M. 1990. The Al2SiO5 polymorphs. Reviewsin Mineralogy 22. Chantilly, VA, Mineralogical Soci-ety of America.

Kerrick, D. M., and Spear, A. J. 1988. The role of minorelement solid solution on the andalusite-sillimaniteequilibrium in metapelites and peraluminous granit-oids. Am. J. Sci. 288:153–192.

Kim, H. S., and Ree, J. H. 2010. P-T modeling of kyaniteand sillimanite paramorphs growth after andalusite inlate Paleozoic Pyeongan supergroup, South Korea: im-plication for metamorphism during the Mesozoic tec-tonic evolution. Lithos 118:269–286.

Kirkpatrick, R. J. 1981. Kinetics of crystallization of ig-neous rocks. In Lasaga, A. C., and Kirkpatrick, R. J.,eds. Kinetics of geochemical processes. Rev. Mineral.8:321–396.

Kuroda, T.; Irisawa, T.; and Ookawa, A. 1977. Growth ofa polyhedral crystal from solution and its morpholog-ical stability. J. Crystal Growth 42:41–46.

Lasaga, A. C. 1986. Metamorphic reaction rate laws anddevelopment of isograds. Mineral. Mag. 50:359–373.

Martın-Algarra, A. 1987. Evolucion geologica alpina delcontacto entre las Zonas Internas y las Zonas Externasde la Cordillera Betica. PhD dissertation, Universityof Granada.

Martın-Algarra, A.; Mazzoli, S.; Perrone, V.; and Rodrı-guez-Canero, R. 2009a. Variscan tectonics in the Ma-laguide complex (Betic Cordillera, southern Spain):stratigraphic and structural Alpine versus pre-Alpineconstraints from the Ardales area (province of Malaga).II. Structure. J. Geol. 117:263–284.

Martın-Algarra, A.; Mazzoli, S.; Perrone, V.; Rodrıguez-Canero, R.; and Navas-Parejo, P. 2009b. Variscan tec-tonics in the Malaguide complex (Betic Cordillera,southern Spain): stratigraphic and structural Alpineversus pre-Alpine constraints from the Ardales area(province of Malaga). I. Stratigraphy. J. Geol. 117:241–262.

Martın-Algarra, A.; Messina, A.; Perrone, V.; Russo, S.;Maate, A.; and Martın-Martın, M. 2000. A lost realmin the internal domains of the Betic-Rifian orogen(Spain and Morocco): evidence from Oligo-Aquitanianconglomerates and consequences for Alpine geody-namic evolution. J. Geol. 108:447–467.

Martınez, F. J.; Reche, J.; and Arboleya, M. L. 2001. P-Tmodeling of the andalusite-kyanite-andalusite se-quence and related assemblages in high-Al graphiticpelites: prograde and retrograde paths in a late kyanitebelt in the Variscan Iberia. J. Metamorph. Geol. 19:661–677.

Mazzoli, S., and Martın-Algarra, A. 2011. Deformationpartitioning during transpressional emplacement of a“mantle extrusion wedge”: the Ronda peridotites,western Betic Cordillera, Spain. J. Geol. Soc. Lond.168:373–382.

Monie, P.; Galindo-Zaldıvar, J.; Gonzalez-Lodeiro, F.;Goffe, B.; and Jabaloy, A. 1991. 40Ar/39Ar geochronol-


ogy of Alpine tectonism in the Betic Cordilleras(southern Spain). J. Geol. Soc. Lond. 148:288–297.

Monie, P.; Torres-Roldan, R. L.; and Garcıa-Casco, A.1994. Cooling and exhumation of the Western BeticCordilleras, 40Ar/39Ar thermochronological con-straints on a collapsed terrane. Tectonophysics 238:353–379.

Mottana, A.; Crespi, R.; and Liborio, G. 1990. The oc-currence of three aluminosilicates at Corenno Plinio(upper Lake Como region, Italy). Mem. Sci. Geol. 42:361–379.

Nassau, K., and Jackson, K. A. 1970. Trapiche emeraldfrom Chivor and Muzo, Colombia. Am. Mineral. 55:416–427.

Pattison, D. R. M. 1992. Stability of andalusite and sil-limanite and the Al2SiO5 triple point: constraints fromthe Ballachulish aureole, Scotland. J. Geol. 100:423–446.

Petreus, I. 1974. The divided structure of crystals. II. Sec-ondary structures and habits. Neues Jahrb. Mineral.Abh. 122:314–338.

Platt, J. P.; Anczkiewicz, R.; Soto, J. I.; and Kelley, S. P.2006. Early Miocene continental subduction and rapidexhumation in the western Mediterranean. Geology34:981–984.

Platt, J. P.; Argles, T. W.; Carter, A.; Kelley, S. P.; White-house, M. J.; and Lonergan, L. 2003. Exhumation ofthe Ronda peridotite and its crustal envelope: con-straints from thermal modelling of a P-T-time array.J. Geol. Soc. Lond. 160:655–676.

Platt, J. P.; Soto, J. I.; and Comas, M. C. 1996. Decom-pression and high-temperature-low-pressure meta-morphism in the exhumed floor of an extensional ba-sin, Alboran Sea, western Mediterranean. Geology 24:447–540.

Platt, J. P.; Soto, J. I.; Whitehouse, M. J.; Hurford, A. J.;and Kelley, S. P. 1998. Thermal evolution, rate of ex-humation, and tectonic significance of metamorphicrocks from the floor of the Alboran Extensional Basin,western Mediterranean. Tectonics 17:671–689.

Platt, J. P., and Whitehouse, M. J. 1999. Early Miocenehigh-temperature metamorphism and rapid exhuma-tion in the Betic Cordillera (Spain): evidence from U-Pb zircon ages. Earth Planet. Sci. Lett. 171:591–605.

Priem, H. N. A.; Boelrijk, N. A. I. M.; Hebeda, E. H.; Oen,E. H.; and Verdurmen, E. A. T. 1979. Isotopic datingof the emplacement of the ultramafic masses in theSerranıa de Ronda, southern Spain. Contrib. Mineral.Petrol. 70:103–109.

Puga, E.; Diaz De Federico, A.; and Nieto, J. M. 2002.Tectonostratigraphic subdivision and petrologicalcharacterisation of the deepest complexes of the BeticZone: a review. Geodin. Acta 15:23–43.

Puga, E.; Nieto, J. M.; and Dıaz de Federico, A. 2000.Contrasting P-T paths in eclogites of the Betic Ophi-olitic Association, Mulhacen complex, southeasternSpain. Can. Miner. 38:1137–1161.

Rast, N. 1965. Nucleation and growth of metamorphicminerals. In Pitcher, W. S., and Flinn, G. W., eds. Con-trols of metamorphism. New York, Wiley, p. 73–102.

Rice, A. H. N., and Mitchell, J. I. 1991. Porphyroblasttextural sector-zoning and matrix displacement. Min-eral. Mag. 55:379–396.

Rubenach, M. J., and Bell, T. H. 1988. Microstructuralcontrols and the role of graphite in matrix/porphy-roblast exchange during synkinematic andalusitegrowth in granitoid aureole. J. Metamorph. Geol. 6:651–666.

Rumble, D. 1973. Andalusite, kyanite, and sillimaniteform the Mount Moosilauke region, New Hampshire.Geol. Soc. Am. Bull. 84:2423–2430.

Sanchez-Navas, A. 1999. Sequential kinetics of a mus-covite-out reaction: a natural example. Am. Mineral.84:1270–1286.

Sanchez-Rodrıguez, L., and Gebauer, D. 2000. Mesozoicformation of pyroxenites and gabbros in the Rondaarea (southern Spain), followed by early Miocene sub-duction metamorphism and emplacement into themiddle crust: U-Pb sensitive high-resolution ion mi-croprobe dating of zircon. Tectonophysics 316:19–44.

Sepahi, A. A.; Whitney, D. L.; and Baharifar, A. A. 2004.Petrogenesis of andalusite-kyanite-sillimanite veinsand host rocks, Sanandaj-Sirjan metamorphic belt,Hamadan, Iran. J. Metamorph. Geol. 22:119–134.

Spry, A. 1969. Metamorphic textures. London, Pergamon.Sunagawa, I. 1987. Morphology of minerals. In Suna-

gawa, I. ed. Morphology of crystals, part B. Tokyo,Terra Science, p. 509–587.

Sunagawa, I.; Bernhardt, H. J.; and Schmetzer, K. 1999.Texture formation and element partitioning in trap-iche ruby. J. Crystal Growth 206:322–330.

Torres-Roldan, R. L. 1974. El metamorfismo progresivoy la evolucion de la serie de facies en las metapelitasalpujarrides al S.E. de Sierra Almijara (sector Centralde las Cordilleras Beticas, S. de Espana). Cuad. Geol.5:21–77.

———. 1978. Scapolite-bearing and related calcsilicatelayers from the Alpujarride series (Betic Cordilleras ofsouthern Spain): a discussion on their origin and somecomments. Geol. Rundsch. 67:342–255.

———. 1979. The tectonic subdivision of the Betic zone(Betic Cordilleras, southern Spain): its significance andone possible geotectonic scenario for the westernmostAlpine belt. Am. J. Sci. 279:19–51.

———. 1981. Plurifacial metamorphic evolution of theSierra Bermeja peridotite aureole (southern Spain). Es-tud. Geol. 37:115–133.

Tossell, J. A., and Vaughan, D. J. 1992. Theoretical geo-chemistry: application of quantum mechanics in theearth and mineral sciences. Oxford, Oxford UniversityPress.

Tubıa, J. M.; Cuevas, J.; and Gil-Ibarguchi, J. I. 1997. Se-quential development of the metamorphic aureole be-neath the Ronda peridotites and its bearing on thetectonic evolution of the Betic Cordillera. Tectono-physics 279:227–252.

Tubıa, J. M., and Gil-Ibarguchi, J. L. 1991. Eclogites ofOjen nappe: a record of subduction in the Alpujarridecomplex (Betic Cordilleras, southern Spain). J. Geol.Soc. Lond. 148:801–804.


Vera, J. A. 2004. Geologıa de Espana. Madrid, SGE-IGME.Walther, J. V., and Wood, B. J. 1984. Rate and mechanism

in prograde metamorphism. Contrib. Mineral. Petrol.88:246–259.

Whitney, D. L. 2002. Coexisting andalusite, kyanite, andsillimanite: sequential formation of three Al2SiO5 pol-ymorphs during progressive metamorphism near thetriple point, Sivrihisar, Turkey. Am. Mineral. 87:405–426.

Whitney, D. L., and Evans, B. W. 2010. Abbreviations fornames of rock-forming minerals. Am. Mineral. 95:185–187.

Wilbur, D. E., and Ague, J. J. 2006. Chemical disequilib-rium during garnet growth: Monte Carlo simulationsof natural crystal morphologies. Geology 34:689–692.

Zeck, H. D.; Albat, F.; Hansen, B. T.; Torres-Roldan, R.L.; Garcıa-Casco, A.; and Martın-Algarra, A. 1989. A

Ma age for the termination of the ductile Al-21 � 2

pine deformation in the internal zone of the Betic Cor-dilleras, south Spain. Tectonophysics 169:215–220.

Zeck, H. P.; Monie, P.; Villa, I. M.; and Hansen, B. T.1992. Very high rates of cooling and uplift in the Al-pine belt of the Betic Cordilleras, southern Spain. Ge-ology 20:79–82.

Zeck, H. P., and Whitehouse, M. J. 1999. Hercynian, Pan-African, Proterozoic and Archean ion-microprobe zir-con ages for a Betic-Rif core complex, Alpine belt, WMediterranean: consequences for its P-T-t path. Con-trib. Mineral. Petrol. 134:134–149.

———. 2002. Repeated age resetting in zircons from Her-cynian-Alpine polymetamorphic schists (Betic-Riftectonic belt, S. Spain): a U-Th-Pb ion microprobestudy. Chem. Geol. 182:275–292.

Zeck, H. P., and Williams, I. S. 2001. Hercynian meta-morphism in nappe core complexes of the AlpineBetic-Rif belt, western Mediterranean: a SHRIMP zir-con study. J. Petrol. 42:1373–1385.

Transformation of Andalusite to Kyanite in the Alpujarride ...grupo179/pdf/Sanchez Navas 2012.pdfA,...

Documents

Transcript of Transformation of Andalusite to Kyanite in the Alpujarride ...grupo179/pdf/Sanchez Navas 2012.pdfA,...