Relationships between very low-grade metamorphism …hera.ugr.es/doi/15087864.pdf · Relationships between very low-grade metamorphism and tectonic deformation: examples from the

Journal of the Geological Society, London, Vol. 158, 2001, pp. 953–968. Printed in Great Britain.

Relationships between very low-grade metamorphism and tectonic deformation:examples from the southern Central Iberian Zone (Iberian Massif, Variscan Belt)

DAVID MARTÍNEZ POYATOS1, FERNANDO NIETO2, ANTONIO AZOR1 &JOSEu FERNANDO SIMANCAS1

1Departamento de Geodinámica, Universidad de Granada, Campus de Fuentenueva, 18071 Granada, Spain(e-mail: [email protected])

2Departamento de Mineralogıa y Petrologıa, Instituto Andaluz de Ciencias de la Tierra, Universidad de Granada,Consejo Superior de Investigaciones Cientıficas, Campus de Fuentenueva, 18071 Granada, Spain

Abstract: We have studied the syn-kinematic very low-grade metamorphism in a polyphase Variscandeformed region using X-ray diffraction techniques. Two phases of regional metamorphism are related totheir respective episodes of penetrative deformation in the southern Central Iberian Zone. The dataobtained suggest that the rocks did not reach metamorphic equilibrium, but strain favoured the progressof mineral reactions in the more deformed parts. The first deformation is Devonian in age and consists ina heterogeneous ductile shearing coeval with large-scale recumbent folding that were produced underhigh-anchizone to epizone metamorphic conditions. The heterogeneity of the shearing originated straingradients that can be said to enhance the growth of new minerals and the illite polytype transformationin the highly strained overturned limb of the preserved pile of recumbent folds, but illite crystallinityremained constant throughout the structure. The second deformation is Mid-Carboniferous in age andconsists in an upright folding that took place under late diagenesis to low-anchizone metamorphicconditions. The distribution of mineral parageneses and illite crystallinity across one of the upright foldssuggests that strain gradients favoured the metamorphic reaction progress from the hinge (low strain)towards the limbs (high strain). Other characteristics of the region such as a metamorphic gap associatedwith an unconformity at the base of the Lower Carboniferous rocks, or cryptic contacts aureolessurrounding volcanic intercalations and a large granitic batholith, are also studied.

Keywords: very low-grade metamorphism, structural geology, strain, deformation-enhanced reactionprogress, Iberian Massif.

Integrated studies of metamorphic petrology and structuralgeology constitute a powerful tool for understanding thetectonic evolution of the inner zones of orogens. In thiscontext, the relationships between the growth of metamorphicminerals and different deformation phases are important. Inmedium- and high-grade metamorphic terrains, these studiescan be adequately accomplished with the aid of optical mi-croscopy. In contrast, in low- and very low-grade metamorphicterrains, it is difficult to establish the relationships between thegrowth of metamorphic minerals and the deformation history.Despite the difficulties, attempts have nevertheless been madeto study very low-grade metamorphic terrains, e.g. Kasig &Späth (1975) in the Venn anticline, Frey et al. (1980) in theCentral Alps, Roberts & Merriman (1985) in the Pennantanticline, Bevins & Robinson (1988) and Roberts et al. (1989)in the Welsh Basin, Johnson & Oliver (1990) in the LesserHimalaya, Merriman et al. (1990, 1995) in the SouthernUplands, Gutiérrez Alonso & Nieto (1996) in the northernIberian Massif, Warr et al. (1996) in the ScandinavianCaledonides and Offler et al. (1998) in the Lachlan Fold Belt.

The increase in the grade of regional metamorphism with theintensity of deformation is a generally established fact. Ingeological areas deformed under low- and very low-grademetamorphism, the correlation between the intensity of defor-mation and the metamorphic grade has been examined bysome authors. Roberts & Merriman (1985), in a tight anticline,interpreted a causative relationship between strain and meta-morphic grade (as apparent from illite crystallinity) from the

association between the isocryst pattern and the fold geometry.Robinson et al. (1990) questioned this detailed isocryst pattern,although they still recognized a general trend showing higherillite crystalinity towards the core of the anticline. The work byGutiérrez Alonso & Nieto (1996) has established a semiquan-titative relationship between illite crystallinity and finite strainrelated to distance from an important thrust. Orozco et al.(1998), in a tight kilometre-scale recumbent anticline, showedthe existence of an increasing crystallinity trend from the limbstowards the inner core of the fold.

Nonetheless, the mechanisms by which deformation canenhance the metamorphic grade are not fully understood. Thestudy of the interaction between deformation and metamor-phic crystallization processes has been approached in relationto cleavage development and mylonitization. Investigations ofmineral growth during cleavage development have shown thatdeformation significantly influences metamorphic reactions(e.g. Etheridge & Hobbs 1974; White & Knipe 1978; Knipe1981) and the crystal growth of fabric-forming phyllosilicates(Merriman et al. 1995). In a dynamic metamorphic environ-ment, the stored strain energy and the mechanical energydissipated during deformation can provide the energy sourcedriving these chemical or mineral reactions (Wintsch 1985). Insome cases, compositional changes in mylonitic zones resultfrom significant fluid flow (Beach 1976) that may be chan-nelized by shearing and mylonitization (Selverstone et al.1991; Tobisch et al. 1991; Leloup & Kiénast 1993; Azor &Ballèvre 1997; Upton 1998), thus favouring metamorphic

953

reactions in highly strained zones. In this context, the principalforces driving a dynamic metamorphic system towardsequilibrium can be linked to the kinetic effects of deformation,thereby giving rise to the hypothesis that a particular mineralassemblage records the state of reaction progress (Robinson &Merriman 1999). Such a kinetic interpretation seems to offer abetter explanation for certain aspects of mineralogical devel-opment at low temperature than the effects of temperaturealone.

This work presents two examples of low- to very low-grademetamorphism associated with well-known geologicalstructures resulting from different types and intensities ofdeformation. The study has been performed by combiningclassic structural geology techniques (detailed structural map-ping, study of fabric characteristics, strain analysis) withmethods for identifying low- and very low-grade metamorphicminerals. X-ray diffraction (XRD) analysis of minerals hasenabled us to accurately characterize the metamorphism thataffected the rocks in the southernmost Central Iberian Zone,part of the southern branch of the Variscan Belt in the IberianMassif (Fig. 1). The Central Iberian Zone contains largeareas affected by only very low-grade metamorphism that arerelatively well-known structurally (e.g. Dıez Balda et al. 1990;Martınez Poyatos 1997), though poorly understood in terms oftheir metamorphic evolution. The rocks examined here areaffected by low- to very low-grade metamorphism and themost frequent parageneses are made up of white mica+chlorite+quartz�albite. These parageneses are stable underconditions ranging from diagenesis up to the incoming ofbiotite. XRD in pelitic rocks enables the identification ofminerals not easily recognizable with optical microscopy(fine-grained muscovite, paragonite, pyrophyllite, kaolinite,mixed-layer phyllosilicates, etc.) and the measurement ofcrystallographic parameters (i.e. mineral crystallinity, b celldimension), which provide good qualitative (and even semi-quantitative) tools for very low-grade metamorphic research.In combination with structural studies, such investigations are

considered to be an important approach for a more successfulunderstanding of orogenic areas deformed under low- to verylow-grade metamorphic conditions.

The aim of this work is twofold: (1) to characterize themetamorphism in a polyphase deformed area; and (2) toexplain some specific metamorphic characteristics related todeformation features such as the existence of heterogeneousductile shear coeval with folding, strain intensity, type andpenetrativity of the tectonic fabric and the distribution ofmetamorphism across folds.

Geological setting

The Badajoz–Córdoba Shear Zone (Burg et al. 1981), asredefined by Azor et al. (1994a), separates the Ossa–MorenaZone from the Central Iberian Zone (Julivert et al. 1972) (Fig.1) and it is one of the suture contacts which can be recognizedin the Variscan Belt in the Iberian Massif. The Variscantectonothermal evolution of this part of the Iberian Massifincludes (Azor et al. 1994a; Simancas et al. 2001): (1) An initialcompressional stage, during the Mid–Late Devonian, due to aconvergence and crustal-scale thrusting of the Central IberianZone over the Ossa–Morena Zone, producing large recumbentfolds in the walls (Ossa–Morena Zone and Central IberianZone) and high-pressure metamorphism in the interveningBadajoz–Córdoba Shear Zone. (2) During the Early Carbon-iferous, the thickened lithosphere became unstable and anorth-directed oblique-extensional collapse took place in therear of the thickened zone. Both the thrusting at the front andthe extension at the rear contributed to the exhumation of thehigh-pressure metamorphic rocks included in the Badajoz–Córdoba Shear Zone. During the extensional stage, subsidencein the hanging wall (the southern part of the Central IberianZone) allowed the accumulation of a thick syn-orogenicsedimentary sequence (Martınez Poyatos 1997).

In the southern part of the Central Iberian Zone, thesequence of pre-orogenic rocks comprises, from bottom to top:

Fig. 1. (a) Geological sketch of thesouthwestern Iberian Massif showing themain zones. (b) Schematic geologicalmap of the boundary between theOssa–Morena and Central IberianZones. Areas corresponding to Figures 2(Hornachos sector) and 3 (Espiel sector)are indicated.

954 D. MARTIuNEZ POYATOS ET AL.

schists and metagreywackes with interlayered amphibolites,gneisses and black quartzites (Late Precambrian); a volcano-sedimentary unit made up of slates, greywackes, conglomer-ates, and plutonic-volcanic rocks (Vendian–Early Cambrian);detrital and carbonate formations (Early–Mid-Cambrian); andfinally, quartzite and phyllosilicate-rich formations ranging inage from Early Ordovician to Mid-Devonian. In terms of thestructure of the southern part of the Central Iberian Zone, twounits can be distinguished (Fig. 1b): an allochthonous one(located close to the Badajoz–Córdoba Shear Zone) and aparautochthonous one (located to the NE of the first one),separated by a NW–SE-striking thrust with top-to-the-NEsense of movement (Figs 2 & 3), which was active during theMid-Carboniferous (Martınez Poyatos et al. 1998a).

The pre-orogenic rocks of the allochthonous unit areaffected by two kilometre-scale recumbent folds verging to theNE that formed synchronously with the main schistose fabric(Fig. 2). A detailed structural cartography and geologicalcross-sections of these recumbent folds have been described inrecent works (Azor et al. 1994b; Martınez Poyatos et al. 1995a,1998b). The folds strike NW–SE and have sub-horizontal axes,showing considerable hinge thickening. The major overturnedlimb indicates a SW–NE shortening of c.15 km (Fig. 2). Theserecumbent folds are related to a planar–linear fabric associatedwith ductile shearing; the foliation is axial-planar to therecumbent folds, and the stretching lineation is parallel to thefold axes. XZ sections commonly contain asymmetric struc-tures, indicating an important strain component of simple

Fig. 2. Simplified map and cross-sections of the Hornachos sector (see location in Fig. 1b). Circles contain mineral parageneses and IC of eachsample. Mineral symbols after Kretz (1983); MP: muscovite/paragonite mixed layers.

VERY LOW-GRADE METAMORPHISM, IBERIAN MASSIF 955

shearing with top-to-the-SE sense of movement. This ductileshearing is very prominent in the overturned limb of therecumbent folds, as indicated by a ductile thinning of therock-sequence, strong development of the foliation and finitestrain (see below). This main deformation phase was coevalwith a regional metamorphism (M1 hereafter) that produced amineral growth parallel to the foliation planes. M1 is ofvariable grade, although in the study area it does not exceedlow-grade. López Munguira et al. (1991) consider it to be oflow to medium pressure (c.4–5 kbar) using the b cell dimensiondata of white mica. Radiometric isotopic ages for M1 are notavailable, but field relationships enable the age of the recum-bent folds and M1 to be constrained. Rocks as young asMid-Devonian are affected by these folds, but Lower Carbon-iferous sediments unconformably overlie them. The recumbentfolding and shearing resulted from the transpressional evol-ution of the Central Iberian Zone/Ossa–Morena Zone suture,being interpreted as conjugate structures developed in thehanging wall of the main SW-vergent thrust that superposedthe Central Iberian Zone over the Ossa–Morena Zone duringMid–Late Devonian times. Drawing on metamorphic, struc-tural and sedimentary records, Merriman & Frey (1999) haveproposed a classification of geotectonic settings, to which anyregional pattern of very low-grade metamorphism can beadscribed. In our case, all the data can be suitably explainedand integrated within an Alpine collisional type geotectonicsetting.

Unconformably overlying the pre-Carboniferous rocks ofthe allochthonous unit, Early Carboniferous syn-orogenicsediments appear. These sediments are made up to c.2 km ofalternating slates and greywackes with intercalated conglom-erates, limestones and volcanic rocks towards the base of thesequence. These Carboniferous sediments are only affected byupright folds striking NW–SE (Fig. 2), with associated axial-planar slaty cleavage produced at very low-grade metamorphic

conditions (M2 hereafter). In the underlying rocks (previouslydeformed by recumbent folding) there is a local crenulationcleavage associated with the upright folds.

In the parautochthonous unit of the Central Iberian Zone,Late Precambrian–Early Carboniferous rocks crop out, withthe Lower Carboniferous rocks lying conformably over theDevonian ones (Fig. 3). In this unit, the Lower Carboniferousrocks are similar to those of the allochthonous unit, but theyhave a thickness of up to c.7 km. As a whole, the Carbonifer-ous subsidence and associated volcanism in the southern partof the Central Iberian Zone have been interpreted as beingrelated to the simultaneous north-directed extensional collapsein the Badajoz–Córdoba Shear Zone (Martınez Poyatos 1997).The structure of the parautochthonous unit (Figs 2 & 3)consists of NW–SE-striking open to tight upright folds (andconcomitant very low-grade metamorphism) that can be cor-related with the upright folds affecting the allochthonous unit(Martınez Poyatos et al. 1995b). Hence, very low-grade meta-morphism associated with the upright folding in the para-utochthonous unit will also be referred to as M2. Axial-planarto these upright folds, there is a rough slaty cleavage. Radio-metric datings for M2 are not available, but its age can be saidto be constrained (Late Westphalian) from field evidence:Mid-Westphalian sediments are affected by the uprightfolds, but the metamorphic aureole produced by the intrusionof Los Pedroches Batholith (c.300 Ma; Fig. 1b) postdates thefoliation related to the upright folds in the country rocks. Thisupright folding was subsequent to the brittle thrusting of theallochthonous unit over the parautochthonous unit, and bothstructures can be related to a Mid-Carboniferous SW–NEcompression in the SW Iberian Massif. The patterns of struc-tures and associated very low-grade metamorphism producedduring this Mid-Carboniferous compression are well inte-grated, as in the case of M1, in an Alpine collisional typegeotectonic setting (Merriman & Frey 1999).

Fig. 3. Map and cross-section of theEspiel sector (see location in Fig. 1b).IC of each sample is shown.


Microstructural description and strain analysis

During the recumbent folding and its ductile shearing in theallochthonous unit, a planar–linear fabric developed, whichconstitutes the main fabric in the pre-Carboniferous rocks. Thefoliation in the metapelites is defined by alternating quartz-richand predominant phyllosilicate-rich domains. Phyllosilicates(mostly white mica and chlorite) are parallel to the foliation, aswell as elongate quartz grains, which display larger grain sizeand more dynamic recrystallization in the quartz-rich domainsthan in the phyllosilicate-rich ones. The foliation in the quartz-ites is defined by elongate quartz grains and minor phyllo-silicates. The lineation is both mineral and stretching (seestrain analysis below).

When the fabric associated with the recumbent folds isstudied in more detail, differences between the normal andoverturned limbs can be detected. In the metapelites, thefoliation is a slaty cleavage in the normal limbs, and a slatycleavage to schistosity in the overturned limb. Thus, themetapelites of the overturned limb can be defined as truephyllites showing their characteristic lustre, while these fea-tures are observed to be lacking in the normal limbs. Further-more, competent rocks furnish a more adequate lithology inorder to explore any differences in tectonic fabric, since theyare not as easily deformed as pelitic rocks. Quartzites from theoverturned limb (Fig. 4c, d) show a well-developed myloniticfoliation defined by highly elongate and partially recrystallizedquartz ribbons embedded in a fine-grained and well-foliatedquartz–micaceous matrix. Foliation in the quartzites from thelower normal limb (Fig. 4e, f) is much less penetrative and isdefined by weakly deformed quartz grains and by foliation-subparallel phyllosilicates. Finally, a tectonic fabric in thequartzites from the upper normal limb (Fig. 4a, b) ispractically lacking, as shown by rounded quartz grains anddisoriented phyllosilicates. The differences on the tectonicfabric development outlined above are related to the hetero-geneity of the ductile shearing associated with the recumbentfolding. Thus, the greater development of the fabric in theoverturned limb strongly suggests that shearing was moreintense in that limb, while the repercussion of this process inthe normal limbs was lesser.

A finite strain analysis was performed in order to establishthe existence of strain gradients across the different limbs ofthe recumbent folds (Azor et al. 1994b; Martınez Poyatos 1997and unpublished data). To avoid interference with the Carbon-iferous deformation (upright folding), measurements weremade only in those outcrops where the crenulation phase isrepresented by open folds without cleavage development. Thefollowing markers were used: K-feldspar phenocrysts inorthogneisses, pyroclasts in volcanic rocks, pebbles in con-glomerates, quartz grains in sandstones, Skolithos and miner-alized nodules in slates. To measure the finite strain, we usedthe harmonic mean method (Lisle 1977), the Rf/� method(Dunnet 1969), and the Fry method (Fry 1979). The calculatedstrain ellipsoids show the short axis (Z direction) perpendicu-lar to the cleavage, the long axis (X direction) parallel to thestretching lineation, and the intermediate axis (Y direction)perpendicular to X and included in the cleavage planes. Thestrain ellipsoids (as defined by the K parameter of Flinn (1965);K=log(X/Y)/log(Y/Z)) plot in the plane deformation field(K=1) or in the constriction field (K>1), with values of Kranging from 0.8 to 6. Assuming no volume change duringdeformation, the calculated ellipsoids indicate finite stretchingalong the X direction of up to 120%. These K values and

stretching along X are compatible with the existence of simpleshearing parallel to the fold axes during the recumbent folding.The existence of strain gradients across the recumbent foldshas been demonstrated. The X/Y ratios range from 1.5 to 2 inthe upper normal limb and from 2 to 3.2 in the overturnedlimb. The X/Z ratios typically range from 1.1 to 1.5 in theupper normal, from 2.3 to 5 in the overturned limb, and from1.1 to 2.2 in the lower normal limb. These differences in finitestrain indicate an intensification of the deformation from boththe upper and lower normal limbs towards the interveningoverturned limb, which is in agreement with the greatercleavage development in the latter.

To sum up, variations in tectonic fabric and finite strainindicate that ductile shearing was more intense in the over-turned limb than in the normal limbs during the recumbentfolding.

With regard to the Carboniferous upright folding, there is arough slaty cleavage without any stretching lineation. In thepre-Carboniferous rocks of the allochthonous unit (previouslyaffected by the recumbent folds), the upright folding producedlocal, millimetric-spaced crenulation cleavage in the meta-pelites. In the rocks affected only by the upright folds(parautochthonous unit and Carboniferous rocks of theallochthonous unit), the foliation in the metapelites is a slatycleavage defined by oriented phyllosilicates and thin veins filledin with micaceous minerals and Fe oxides. Development of thiscleavage is highly variable, ranging from persistence of thebedding-parallel fabric (compaction), through pencils, up tocomplete obliteration of the bedding. In the quartzites, foli-ation of any type is usually lacking; in just a few cases cana centimetric- to decimetric spaced partition cleavage beobserved. A finite strain analysis was performed on con-glomerates from the parautochthonous unit (MartınezPoyatos 1997) using the afore-mentioned methods. The strainellipsoid is oblate (K=0.1), showing the Z axis to be perpen-dicular to the cleavage. X/Y � 1, and the X/Z � 1.5. Thesedata reveal a slight SW–NE flattening associated with theupright folding.

Materials and methods

Samples aimed of characterizing the M1 and M2 metamor-phisms were taken from the allochthonous and parautoch-thonous units in the southern part of the Central Iberian Zone(Fig. 1b). Two sectors (i.e. Hornachos and Espiel) were chosendue to well-exposed outcrops for obtaining pelitic samples. Inthe Hornachos sector (Fig. 2), most of the samples belong tothe allochthonous unit, while in the Espiel sector (Fig. 3) theybelong to the parautochthonous unit.

Samples were collected from pelitic lithologies as compo-sitionally homogeneous as possible. These were taken far fromfaults and joints in clean outcrops without any evident signs ofmeteoric alteration. After washing and cleaning of patinas,oxides and mineralizations in cracks and veins, they werecrushed to a <2 mm fraction. Oriented aggregates were pre-pared with whole-rock and <2 µm fractions (the latter sep-arated following the recommendations of the ‘IGCP 294 ICworking group’; Kisch 1991). The aim of separating a fine-grained fraction is to minimize the content of detrital micasnon-re-equilibrated during very low-grade metamorphism,which are generally larger than 2 µm. The samples wereanalysed with a Phillips PW 1710 X-ray diffractometerequipped with a graphite monochromator and automatic


divergence slit, using CuK� radiation. The resulting diffractiondiagrams provided the following data:

Identification of the different minerals based on theircharacteristic reflections. All the minerals identified have beenroughly classified by order of abundance, according to therelative intensity of their reflections. Poor-quality reflections

(multiple peaks, poorly defined peaks, etc.) are considered tobe due to minerals altered or present in small percentages.

Illite crystallinity index. Illite crystallinity indices weremeasured (IC, width at half peak height; Kübler 1968) inhigh-resolution scanned diffraction diagrams specifically ob-tained for the 10 Å reflection of white mica. In those samples

Fig. 4. Microstructure of quartzites sampled from different limbs of the recumbent folds. (a), (b): upper normal limb (width of view 1.1 mm).(c), (d): overturned limb (width of view 3.7 mm). (e), (f): lower normal limb (width of view 0.9 mm).


containing sodium micas (paragonite or muscovite/paragonitemixed layers), and after a detailed examination of the shape ofthe 10 Å potassium mica peak, IC was only measured when thepeak interference observed did not affect the white mica IC, i.e.sodium micas being scarce enough so that the widening of the10 Å white-mica peak is located far below the half peak height.

The intensity of the reflections from which the crystallinity ismeasured must also be taken into account. Thus, an ICmeasured from weak peaks (<400 counts) has only been takento be valid when the value obtained is coherent with the resultsof the group of samples to which it belongs.

The IC values obtained have been recalculated to theinternational scale of crystallinity indices (CIS, ‘crystallinityindex standard’; Warr & Rice 1994) via the equation: ourdata=0.674·CIS+0.052 (regression coefficient 0.999), obtainedin our laboratory with standard samples provided by Warr &Rice (Nieto & Sánchez Navas 1994). Using this standard scale,the anchizone boundaries are the traditional ones of 0.25 and0.42 ��2� proposed by Kübler (1968) (Warr & Rice 1994). Theanchizone is subdivided into low (IC>0.30 ��2�) and high(IC<0.30 ��2�) (Merriman & Peacor 1999). The epizone ischaracterized by values lower than 0.25 ��2� (which can becorrelated to the low-grade metamorphism or the greenschistfacies; Kisch 1987), whereas late diagenesis is revealed byvalues higher than 0.42 ��2�. The thermal range for theanchizone is c.200–300 �C, although the IC cannot beconsidered as a true geothermometer (Frey 1987; Kisch 1987).More properly, IC values represent an indicator of reactionprogress and inferred temperatures should only be taken asapproximate (Merriman & Peacor 1999).

IC values were measured in both whole-rock and <2 µmsample fractions in order to evaluate the detrital mica contentin the samples studied. By comparing the IC values obtainedfrom both fractions, the existence of high-crystalline non-re-equilibrated detrital micas during very low-grade metamor-phism is evidenced when the IC value for the whole-rockfraction is less than for the <2 µm fraction. In such cases, thelatter IC value is considered to be a more realistic indication ofmetamorphic conditions.

Qualitative estimation of illite polytypes. An estimate of theproportion of 1Md (characteristic of lower temperature) and2M1 (characteristic of higher temperature) illite polytypes canbe obtained from X-ray diffraction diagrams. Several methodshave been proposed to determine illite polytype ratios, buttheir accuracy is strongly influenced by the difficult prep-aration of disoriented powder samples and by the calibrationof the method (see Dalla Torre et al. 1994 for a review). In ourcase, we have limited ourselves to making a simple qualitativeestimate of the existence of illite polytypes by comparing twoillite peak heights: the 2.58 Å reflection (present in bothpolytypes) and the 2.80 Å reflection (which appears only in the2M1 polytype; Maxwell & Hower 1967). When the former ismuch more intense than the latter then, apart from the 2M1,there may be a significant proportion of the 1Md polytype.Experiments have shown that the prograde reaction 1Md �2M1 seems to occur at 200–350 �C (approximately anchizone)and PH2O� 2 kbar (Frey 1987). From a compilation of fieldstudies, Frey (1987) concluded that this reaction is completedapproximately at the anchizone/epizone boundary. By con-trast, Merriman & Peacor (1999) conclude that there is not apredictable, accurate correlation with temperature and thatpolytypic sequences should only be used as indicators of (mostlikely strain-induced) reaction progress.

Basal spacing of white mica. Precise measurements of the basalspacing of white mica have been made, using an aggregate ofthe <2 µm fraction with <2 µm quartz added as an internalstandard. The basal spacing of phyllosilicates is related to theircompositional characteristics (Guidotti et al. 1992). The basalspacing of white mica is, in theory, related to the paragoniticNa/K substitution, thereby approximately reflecting the tem-perature of formation. Nevertheless, the basal spacing mayalso be affected by other factors such as the phengite content(Guidotti 1984; Guidotti et al. 1992), the presence of minorquantities of NH4

+ (Juster et al. 1987), F− (Robert et al. 1993),or OH�O− substitution (Ackermann et al. 1993).

White mica b cell dimension. The b cell dimension of the whitemica has been obtained using a polished slice of rock cutperpendicular to the foliation, in order to enhance the (060)reflection (Sassi & Scolari 1974; Guidotti & Sassi 1986; Frey1987). As an internal standard we have used the quartzreflection at 59.96 �2� (after repeated calibration with quartzreflections at 20.56 �2� and 26.66 �2�). The b cell dimension hasbeen calculated by measuring the difference in spacing betweenthese peaks ((060) for white mica and the corresponding one of59.96 �2� for quartz). Although this measure corresponds to aconvolution of (060) and other white-mica peaks, no signifi-cant error is introduced and the method remains valid (Riederet al. 1992; Wang et al. 1996). The b cell dimension of whitemica depends quite exclusively on the phengite substitution ofmica, which is related to the pressure conditions at the time offormation/recrystallization of the mica in the rock. Thus, asemiquantitative relationship has been established between theb cell dimension and the metamorphic pressure gradient: bvalues lower than 9.000 Å are typical of low-pressure andvalues higher than 9.040 Å of high-pressure facies meta-morphism (Guidotti & Sassi 1986). The lack of a limitingassemblage is the main problem of this approach (Massone &Schereyer 1987) but, as in many geological terrains, no othercriteria exist for the estimation of pressure. In low-grademetamorphic rocks, approximate burial pressures are derivedfrom the P–T–b diagram of Guidotti & Sassi (1986). Padanet al. (1982) considered the possibilities of application of the bcell dimension to subgreenschist facies terrains, which islimited by the persistence of detrital micas. Despite the limi-tations, in the case of very low-grade metamorphic or latediagenetic rocks, many authors extend the b curves into thesubgreenschist facies P–T space (e.g. Underwood et al. 1993;Offler et al. 1998).

XRD results

According to the structure of the study area, the samples canbe grouped as follows (Tables 1 & 2).

(a) In the allochthonous unit we have separated the samplesfrom the Carboniferous rocks (only affected by upright foldsand M2) from those collected in the pre-Carboniferoussequence (affected by the recumbent folding and M1).Obviously, the pre-Carboniferous rocks were afterwards affectedby the upright folds and M2 but, as will be discussed below, weconsider that the metamorphic data of the pre-Carboniferousrocks can be attributed to at least minimum P/T conditionsduring M1. Samples affected by M1 were grouped into thosebelonging to the normal limbs and those belonging to the over-turned limb, in order to check whether the above-describeddifferences in tectonic fabric development and finite strain couldhave influenced their metamorphic evolution.


Table 1. Characteristic phyllosilicates and crystallographic parameters of samples studied

Sample

Characteristic minerals*Standardized

IC (��2�) Basalspacing(Å) Ms b (Å) MsMs Chl Pg Cld Prl Kln MP W-rock <2 µm

M1 (recumbent folding)Normal limbs (pre-Carboniferous rocks, allochthonous unit, Hornachos sector)106 � � 0.19 10.025 8.992107 � � 0.28 0.29 9.979 9.018120 � � 0.25 0.23 9.992121a � � 0.20 0.22 9.997 9.004121b � � 0.20 0.23 9.999122 � � 0.25 0.35 9.976 9.013123 � 0.26 0.31 9.996 9.003Mean 0.23 0.27 9.995 9.006Standard deviation 0.03 0.05 0.015 0.009Overturned limb (pre-Carboniferous rocks, allochthonous unit, Hornachos sector)14 � � � � 9.99415 � � 0.19 0.25 10.00016 � � � � 10.018 8.98823 � � 0.20 0.23 9.99597a � � � � 0.20 0.28 9.97697b � � � � 0.19 0.28 9.98598a � � 0.25 0.25 10.00298b � � 0.22 0.22 10.002 8.99499 � � 0.19 0.26 9.989 9.005100 � � 0.19 0.29 9.996 8.995101 � � 0.20 0.26 9.987 8.994102 � � � 10.000105 � 0.22 0.25 9.970 9.025108 � � � 9.997 9.038109 � � 0.25 0.26 10.001 9.017116 � � � � � 10.030117 � � � 0.26 0.26 10.016 8.986Mean 0.21 0.26 9.998 9.005Standard deviation 0.02 0.02 0.014 0.017M2 (upright folding)Carboniferous rocks (allochthonous unit, Hornachos sector)91 � 0.34 0.32 9.97992 � � 0.26 0.29 10.002 9.00596 � � 0.23 0.26 10.000 9.007111 � � 0.25 0.28 10.002 9.005118 � � � 0.44 0.53 10.008 8.996119 � � 0.29 0.23 9.997 9.010124 � � � 0.26 0.28 10.009 9.004Mean 0.30 0.31 10.000 9.004Standard deviation 0.07 0.09 0.009 0.004Parautochthonous unit (Hornachos sector)103 � � � 0.29 0.34 10.013104a � � � � 0.41 0.61 10.012104b � � � � 0.35 0.50 10.015112 � � 0.34 0.28 10.010113 � � � 0.47 0.49 10.012 8.967114 � � 0.37 0.41 10.014 8.992Mean 0.37 0.44 10.013 8.979Standard deviation 0.06 0.11 0.002 0.012Pre-Carboniferous rocks (parautochthonous unit, Espiel sector)66a � � 0.31 0.40 10.01066b � � 0.35 0.55 10.01372a � � � 0.44 0.59 10.00772b � � � 0.31 0.52 10.00073 � � 1.05 9.999 8.98875 � � 0.55 0.44 10.003 8.98981 � � � 0.28 0.47 10.004 9.00382 � � 0.41 0.59 9.987 9.00183a � � 0.47 0.64 9.984 9.01183b � � 0.46 0.55 9.984 9.01684 � � 0.49 0.65 10.013 8.995


(b) In the parautochthonous unit we have distinguished theCarboniferous rocks from the underlying ones, in order todetermine whether there is any metamorphic evidence for anunconformity at the base of the Carboniferous rocks.

M1 metamorphismIn the pre-Carboniferous rocks of the allochthonous unit,the IC difference between the whole-rock fraction and the<2 µm fraction is slight (c.0.05 ��2�; Table 1). In the graph of

Table 1. continued

Sample

Characteristic minerals*StandardizedIC (��2 �) Basal

spacing(Å) Ms b (Å) MsMs Chl Pg Cld Prl Kln MP W-rock 2 µm

M2 (upright folding)Pre-Carboniferous rocks (parautochthonous unit, Espiel sector)85 � � � � � 10.00686 � � � 9.990 8.991Mean 0.41 0.59 10.000 8.999Standard deviation 0.09 0.16 0.010 0.020Carboniferous rocks (parautochthonous unit, Espiel sector)67 � � 0.38 0.47 10.011 9.00068 � � 0.28 0.32 10.002 8.98871 � � 0.37 0.41 9.999 8.99274 � � � 0.37 0.59 10.000 8.99876 � � � 0.34 0.34 9.99177a � � 0.40 0.46 10.008 8.98477b � � � 0.37 0.49 10.008 8.99478 � � 0.32 0.32 9.98179 � � � � 0.35 0.41 10.013 8.98680 � � � 0.29 0.37 10.01087 � � � 9.98388 � � 0.37 0.43 10.00589a � � 0.38 0.43 9.99389b � � 0.37 0.47 9.98690 � � 0.34 0.52 9.996 8.998Mean 0.35 0.43 9.999 8.992Standard deviation 0.03 0.07 0.010 0.006

* Mineral symbols after Kretz (1983). MP: Ms/Pg mixed layers.� Abundant.� Scarce.� Possibly from meteoric alteration or affected by it.

Table 2. Metamorphic and structural features related to recumbent and upright folding phases in the area studied

Para-genesis* IC (��2�)†

Illitepolytypes

Metamorphicgrade b (Ms)†

Foliation types Strain

Quartzites Pelites X/Y X/Z

Recumbent folding (M1)Pre-Carboniferous rocks (allochthonous unit)Normal limbs Ms�Chl 0.27 (0.05) 1Md, 2M1 High anchizone

to epizone9.006

(0.009)Rough

cleavageSlaty cleavage 1.5–2 1.1–2.2

Overturnedlimb

Ms�Chl�MP�Pg�Cld�Prl

0.26 (0.02) 2M1 High anchizoneto epizone

9.005(0.017)

Myloniticfoliation

Slaty cleavageto schistosity

2–3.2 2.3–5

Upright folding (M2)Carboniferous rocks (allochthonous unit)

Partitioncleavage

Slaty cleavage 1 1.5

Ms+Chl�Kln

0.31 (0.09) 1Md, 2M1 Low to highanchizone

9.004(0.004) }Parautochthonous unit

Hornachossector

Ms+Kln�MP�Prl

0.44 (0.11) 1Md, 2M1 Late diagenesisto low an-

chizone

8.979(0.012)

Espiel sector Ms�Chl�Kln�MP

0.50 (0.14) 1Md, 2M1 Late diagenesis 8.996(0.009)

* In order of abundance, in addition to quartz and minor feldspar, hematite, goethite and calcite; MP: Ms/Pg mixed layers.† Mean value (standard deviation).


Figure 5, samples in this group are projected on or close to thediagonal line representing IC (<2 µm fraction) = IC (whole-rock fraction). This indicates that the detrital micas wereconsiderably re-equilibrated towards the prevailing meta-morphic conditions.

Normal limbs. Samples in this group (Table 1, Fig. 2) showparageneses containing muscovite � chlorite. The chlorite issometimes scarce and partially altered (it has wide and poorlydefined reflection peaks). The average IC value is 0.27 ��2� forthe <2 µm fraction. The lower-temperature white-mica poly-type (1Md) appears in addition to the 2M1 one. These featuresare representative of high-anchizone to epizone metamorphicconditions.

Overturned limb. The parageneses identified in this group(Table 1, Fig. 2) are quite variable and show several meta-morphic minerals: apart from muscovite � chlorite, there isparagonite, chloritoid, pyrophyllite and muscovite/paragonitemixed layers. Only the high-temperature illite polytype(2M1) has been recognized. Mean IC is 0.26 ��2� (approx.anchizone–epizone boundary) for the <2 µm fraction. Thepresence of pyrophyllite and chloritoid (although in differentsamples) may indicate entrance within the chloritoid stabilityfield, defined by the continuous reaction pyrophyllite+chlorite� chloritoid+quartz+H2O (Frey 1987), even though there arealso other reactions producing chloritoid. The kaolinite found(a typical mineral of sedimentary processes) cannot be para-genetic with the rest of the metamorphic minerals in thesamples from this limb. Therefore, we consider this mineral to

have formed during a process of retrogression (i.e. during M2)or as an alteration subsequent to the metamorphic peak.

Pressure conditions. The parageneses identified are not helpfulfor estimating the pressure conditions during the recumbentfolding in the Hornachos sector. We have determined the b celldimension of the white mica (Table 1), obtaining values thatshow a relatively good correlation with its basal spacing (Fig.6). This relationship indicates that the basal spacing of whitemica is mainly controlled by the phengitic substitution andthat the influence of other factors (e.g. the paragonitic substi-tution) is secondary. As stated above, the detrital micas werere-equilibrated towards the prevailing metamorphic conditions(M1), thus indicating that the determination of the b celldimension using whole-rock specimens is suitable for meta-morphic purposes. Due to the small vertical distance betweenthe three limbs of the recumbent folds (Fig. 2), differences inpressure between them are not expected, and our data confirmthis (Table 1). The mean value of the b cell dimension for allthe samples affected by M1 is 9.005 Å (standard deviation0.015), indicating an intermediate-pressure gradient, near theboundary with the low pressure gradient. Overall, consideringthat the metamorphic grade was about high-anchizone toepizone (for which a temperature of c.300–400 �C can beassumed), pressures of at least c.2–3 kbar can be suggested(M1 arrow in Figure 7).

M2 metamorphism

The metamorphism associated with the Carboniferous uprightfolding has been studied in several sectors; namely, in theCarboniferous rocks of the allochthonous unit and in theHornachos and Espiel sectors of the parautochthonous unit.

The IC difference between the whole-rock fraction and the<2 µm fraction is highly variable, going from negligible in theCarboniferous rocks of the allochthonous unit to high(c.0.20 ��2�) in the pre-Carboniferous rocks of the parautoch-thonous unit in the Espiel sector (Table 1). Furthermore, in thesamples affected only by M2, the greater the IC value, the

Fig. 5. Plot of IC (whole-rock fraction) versus IC (<2 µm fraction)of all the samples studied. Note that the difference between thesetwo values (i.e. detrital content of white mica) decreases asmetamorphism increases.

Fig. 6. Basal spacing versus b cell dimension plot of white mica, withregression lines in the different groups of samples considered. n:number of samples; r: regression coefficients.


greater the difference in IC between the two fractions will be(Fig. 5). This relationship suggests that a progressive decreasein the proportion of non-re-equilibrated detrital micas wasproduced as the metamorphism increased in intensity from thediagenesis to the anchizone.

Carboniferous rocks of the allochthonous unit. Samples fromthis group (Table 1, Fig. 2) have a paragenesis consisting ofmuscovite�chlorite (sometimes altered), kaolinite being alsooccasionally present. The average IC for the <2 µm fraction is0.31 ��2� and both illite polytypes appear (1Md and 2M1).These features indicate a low- to high-anchizone metamorphicgrade. Taking into account the data concerning only M2 in theparautochthonous unit (described in the following para-graphs), we consider that, as will be discussed below, thismiddle-anchizone metamorphic grade is overestimated due tothe thermal effect of the volcanic rocks intercalated in theCarboniferous sedimentary sequence. Therefore, metamor-phism in the Carboniferous rocks of the allochthonous unitcannot be concluded to be representative of M2.

Hornachos sector (parautochthonous unit). The samples fromthis sector (Table 1, Fig. 2) have parageneses containingkaolinite+muscovite�muscovite/paragonite mixed layers�pyrophyllite. The average IC is 0.44 ��2� for the <2 µmfraction, and both illite polytypes are present. These dataindicate a metamorphic grade of late diagenesis, near theanchizone boundary. More precisely, the co-existence of kao-linite and pyrophyllite indicates the lower boundary of thepyrophyllite stability field, defined by the reaction kaolinite +quartz � pyrophyllite + H2O, which occurs at a calculatedtemperature of 280–315 �C (Frey 1987).

Espiel sector (parautochthonous unit). The samples from thissector (Table 1, Fig. 3) contain muscovite�chlorite (often

altered)�kaolinite. There can also be found scarce muscovite/paragonite mixed layers and pyrophyllite. Crystallinity is low(IC=0.50 ��2� for the <2 µm fraction) and both illite polytypesoccur. These data indicate that this sector did not exceed latediagenetic conditions.

Upon closer examination of the data, some differencesbetween the pre-Carboniferous and the Carboniferous sampleshave been detected. In the pre-Carboniferous samples, kaoli-nite predominates over chlorite, and illite crystallinity is verylow (mean IC=0.59 ��2�). In some samples, there are reflec-tions corresponding to large basal spacings (>10 Å), likely tohave been caused either by illite/smectite mixed layers orsimply by alterations. These samples were treated withethylene-glycol, and samples 73 and 75 showed expandablephases of illite/smectite (with an R3 or ISII layering pattern,i.e., three layers of illite and one of smectite). In the Carbon-iferous samples, chlorite is much more abundant than kaoli-nite, illite crystallinity is higher (IC=0.43 ��2�), and there areno illite/smectite mixed layers. The above differences suggestthat the pre-Carboniferous materials underwent typical diage-netic conditions, while the Carboniferous materials reached thelate diagenesis–low anchizone boundary. The fact that thepre-Carboniferous samples do not show a higher metamorphicgrade than the Carboniferous ones is in accordance with thelack of an unconformity between these two groups of rocks.Paradoxically, the metamorphism in the Carboniferous rocksseems to be slightly higher than in the deeper stratigraphiclevels (see discussion below).

Pressure conditions. The white mica b cell dimension has beendetermined in the samples affected only by M2. The use of thismethod in such low temperature pelites is limited mainly by thepersistence of variable amounts of detrital micas (Padan et al.1982), which can make the interpretations only orientative.The b values are relatively low (Table 1): the average value is8.997 Å and the standard deviation is 0.011, suggesting thatthe metamorphic pressure gradient during the upright foldingwas low, near the intermediate pressure gradient boundary.

As in the case of M1, there is a good correlation betweenthe white mica basal spacing and its b cell dimension(Fig. 6), lower b values (indicating low phengitic substitution)corresponding to greater basal spacings. This correlation indi-cates that the basal spacing is mainly controlled by the degreeof phengitization and that the influence of the paragoniticsubstitution or other factors is secondary. As a semi-quantitative approximation, considering late diagenesis to lowanchizone metamorphic grade during M2 (<300 �C), minimumpressures of c.1–1.5 kbar can be suggested (M2 arrow inFig. 7).

Discussion

Retrogression of M1 during M2

The pre-Carboniferous sequence of the allochthonous unit wasmetamorphosed during two phases of deformation: first (M1)in the Devonian during recumbent folding and associatedductile shearing, and second (M2) in the Mid-Carboniferousduring upright folding. The first deformation was undoubtedlymuch more intense than the second one. The second defor-mation produced crenulation of the previous fabric and rarecrystallization associated with the local development of amillimetric-spaced crenulation cleavage. M2 conditions were

Fig. 7. Qualitative comparison between metamorphic conditionscorresponding to M1 and M2 (metamorphism associated with therecumbent and upright folds, respectively). White mica b cellparameters have been projected in a P–T diagram (Guidotti & Sassi1986). Aluminium silicate triple point after Holdaway (1971);reaction Kln+Qtz=Prl+H2O after Frey (1987).


of late diagenesis and did not exceed the diagenesis–anchizoneboundary, as shown by our results in the parautochthonousunit (only affected by the upright folds and M2). M1 paragen-eses and the IC might have been modified by M2 to someextent. However, we consider that the metamorphic gradeduring M2 was low enough to preserve the higher-gradeparageneses previously formed during M1. At most, M1parageneses could have been partially retrogressed (forexample, the kaolinite found in the overturned limb of therecumbent folds can be explained as having formed duringM2). Therefore, we consider that the results concerning meta-morphism from the pre-Carboniferous samples of the alloch-thonous unit are representative of the M1 conditions or, ifsome retrogression took place during M2, they would thenrepresent the minimum conditions of M1.

M1 differences between normal and overturned limbs(allochthonous unit)

Clear differences in cleavage development, mylonitization andfinite strain exist between the normal and overturned limbs ofthe recumbent folds, which provides evidence for an intensifi-cation of the ductile shearing in the overturned limb. In termsof metamorphism, our XRD results also show some differ-ences: mineral parageneses in the two normal limbs onlyinclude muscovite and chlorite, while in the overturned limbparagonite, muscovite–paragonite mixed layers, pyrophylliteand chloritoid are also found. These mineralogical differencesare not considered to be controlled by the chemical compo-sition of the original sediments, since the samples come fromsimilar lithologies. In addition to these differences in para-geneses, a qualitative estimate of illite polytypes shows thatboth are present in the normal limbs, but in the overturnedlimb only the high-temperature one can be found, thussuggesting that the latter reached a higher temperature thanthe normal limbs.

Concerning the IC values, our results do not show statisti-cally distinguishable differences between the normal and over-turned limbs (Fig. 5). The mean IC value of the overturnedlimb (0.26 ��2�) is very similar to that of the lower normallimb (0.27 ��2�), with only one value being available from theupper normal limb (0.29 ��2�). Since temperature is consid-ered to be the main factor controlling the IC (Frey 1987), ourresults strongly suggest that the whole recumbent structuremust have formed under the same temperatures (approxi-mately anchizone–epizone boundary). Chloritoid is a typicalepizonal index mineral, but its appearance in two samples fromthe overturned limb cannot be unambiguously regarded as anindicator of the epizone, since chloritoid can occur, althoughrarely, in the anchizone (Kisch 1983). In this context, theparageneses and illite polytypes of the overturned limb cannotbe properly explained in terms of higher temperature. Analternative interpretation is that strain and fluid pathwaysalong highly sheared zones could enhance the growth ofapparently higher-grade metamorphic minerals and illite poly-types. Thus, the metamorphic conditions in the whole struc-ture could have remained constant while, in the overturnedlimb, deformation must have favoured the approach to equi-librium conditions. Metamorphic fluids and energy suppliedthrough deformation in the highly sheared overturned limballowed, through reaction progress, the growth of new meta-morphic minerals other than muscovite and chlorite, and thecomplete 1Md � 2M1 illite polytype transformation.

Thermal influence of igneous intercalations in theCarboniferous sedimentary sequence (allochthonousunit)

The IC values obtained from the Lower Carboniferoussamples of the allochthonous unit indicate a higher metamor-phic grade (low to high anchizone) than the grade in the othersectors where M2 has been studied (late diagenesis; Table 2).Moreover, the difference in IC value between the <2 µmfractions and the whole-rock fractions suggests that, in thecase of the Carboniferous rocks of the allochthonous unit, thedetrital micas were re-equilibrated during metamorphism,whereas in the other sectors, non-re-equilibrated detrital micasare always present (Fig. 5). These particular features of theCarboniferous rocks of the allochthonous unit can be ex-plained by the existence of diverse subvolcanic and volcanicrocks intercalated in the sedimentary sequence. In the Carbon-iferous outcrop south of Hornachos (Fig. 2), there are c.300 mof igneous rocks (two rhyolitic episodes separated by a basicone containing basalts and gabbros), and in the outcrop southof Campillo (Fig. 2) there are several decimetric- to metricintercalations of basalts. Despite the fact that the distancebetween igneous rocks and the sites for sample collection is atleast several tens of metres, the heat supplied by these igneousintercalations could have produced low-temperature contactmetamorphism (cryptic aureole; Merriman & Frey 1999) in thesurrounding sediments, thus enhancing the crystallinity oftheir phyllosilicates and re-equilibrating the detrital micastowards the thermal conditions during this Lower Carbonifer-ous igneous activity. Mid-Carboniferous M2 (late diagenesis)was not intense enough to re-equilibrate the phyllosilicatestowards such very low metamorphic conditions. One exceptionis the Carboniferous outcrop east of Campillo (Fig. 2), whereigneous rocks have not been found. Although only one samplefrom this outcrop has been studied (sample 118), its low ICvalue (0.53 ��2�) and the presence of non-re-equilibrated de-trital micas (Fig. 5) are in accordance with the late diageneticconditions during M2.

Distribution of M2 in the Espiel sector(parautochthonous unit); its relationship with foldgeometry

The results in the Espiel sector suggest an apparently highermetamorphic grade in the Carboniferous rocks (diagenesis–anchizone boundary with predominance of chlorite over kao-linite) than in the underlying pre-Carboniferous ones (latediagenesis with predominance of kaolinite over chlorite, andpresence of illite/smectite mixed layers) (Table 1). Severalarguments concerning these paradoxical differences need to bediscussed.

(1) In the Carboniferous slates there are several inter-calations of basalts (some being as thick as a few hundreds ofmetres), which might also have been responsible for the highercrystallinity observed. Samples were collected at least atc.100 m from these igneous intercalations. Apart from thisdistance, a thermal influence can be rejected in view of Figure3: IC values of the samples collected from the southernCarboniferous outcrops (which include igneous rocks) arestatistically higher (mean 0.47 ��2�; lower crystallinity) thanthose from the northeastern outcrops (without igneous rocks;mean 0.40 ��2�; higher crystallinity).

(2) Another hypothesis to consider is the existence ofinverted metamorphism. However, during the upright folding


and related M2 it is very difficult to imagine an invertedgeothermal gradient in the uppermost crust. The unique ther-mal source directed downwards during the Carboniferousmight be the NE-directed thrusting of the allochthonous unitonto the parautochthonous unit. This thrusting occurred in theMid-Carboniferous after the Visean sedimentation and priorto the upright folding phase (Late Westphalian; MartınezPoyatos et al. 1998a). This hypotesis can in turn be rejectedsince no crystallization is observed along the brittle fault zone.Futhermore, IC values from the southern Carboniferous out-crops in Figure 3 (which are located closer to the thrust sheet)are higher than those from the northern ones, thus precludingthis interpretation.

(3) A more detailed examination of the distribution acrossthe antiformal structure (i.e. in the SW–NE direction) high-lights new interesting features (Fig. 8). Although the data aresomewhat scattered, there is a clear tendency of decreasing ICvalues from the hinge zone towards both the southern and thenorthern limbs. Samples close to the fold hinge have IC valuesof c. 0.50–0.65 ��2�, while those of the limbs are lower(c. 0.40–0.50 ��2� for the southern limb and c. 0.30–0.50 ��2�for the northern limb). The most noteworthy exceptions aresamples 73 (which, as expected from its position near thehinge, has a high IC value of 1.05 ��2�) and 75 (also locatednear the hinge but with an IC value of 0.44 ��2�) (Fig. 8).Furthermore, the mineral parageneses also show a distributionacross the fold (Fig. 8). Most of the samples from the limbscontain chlorite but not kaolinite. By contrast, most of thesamples located close to the hinge contain kaolinite (and theillite/smectite mixed layers found in samples 73 and 75) but nochlorite. These differences suggest that the metamorphicgrade during the folding might have been slightly higher in thelimbs than in the hinge. However, although temperature is

considered to be the main factor controlling the illite crystal-linity and mineral reactions, other factors such as strain canalso influence it (Frey 1987). The relationship between defor-mation and illite crystallinity has been approached by severalauthors in outcrop-scale folds (e.g. Flehmig & Langheinrich1974; Nyk 1985; Fernández Caliani & Galán 1992), buterroneously assuming that strain is greater in the hinge than inthe limbs. Flexural flow is the most important folding mech-anism in phyllosilicate-rich beds and it produces a concen-tration of the finite strain in the limbs rather than in the hinge(Twiss & Moores 1992; Fischer & Jackson 1999), which is alsoin accordance with the field evidence that cleavage is morepenetrative in the limbs than in the hinges. In the case at issuehere, strain determinations across the fold studied are notavailable, but the distribution of the cleavage intensity (lesspenetrative in the hinge of the antiform) is compatible with afolding mechanism producing greater finite strain in the limbsthan in the hinge. Therefore, the distribution of the IC valuesand mineral parageneses shown in Figure 8 can be interpretedas the effect of the strain on the metamorphic reactions acrossthe fold. Strain-related processes intensified in the limbs couldhave favoured, through kinematically driven reaction progress,the rocks to attain or come closer to equilibrium conditions(Merriman et al. 1995; Merriman & Frey 1999).

Finally, we should note the asymmetry of the IC valuesacross the fold (Fig. 8): while those from the southern limb arealways higher than 0.40 ��2�, several samples from the north-ern limb have IC values as low as 0.32 ��2�. These low ICvalues of samples from the northern limb may be due to itsproximity to the Pedroches Batholith. The Pedroches Batho-lith is a huge Upper Carboniferous intrusion NE of the Espielsector (Fig. 1) having a cartographic contact aureole 500–2000 m wide, with parageneses containing biotite, andalusiteand cordierite. In addition, a cryptic contact aureole can beestablished, wider than the other one, in which only recrystal-lization of the pre-existing micas occurs, with no appreciableneoformation of minerals (Merriman & Frey 1999). Thenorthern limb of the fold studied is located c.10 km from thesouthern boundary of the batholith (see horizontal scale in Fig.8), a distance that allows the low IC values to be explained asdue to the long-distance thermal influence of the intrusion.

Conclusions

In accordance with the sequence of events that took place inthe study region during the Variscan Orogeny, the relation-ships between deformation and very low-grade metamorphismare summarized in the following conclusions (Table 2, Fig. 9).

(1) The recumbent folds (pre-Carboniferous in age) in theHornachos sector of the allochthonous unit formed in highanchizone to epizone (c.300–400 �C) metamorphic conditions(M1), with complete re-equilibration of the detrital micas.Pressure has been estimated at a minimum of 2–3 kbar, whichcould correspond to the minimum original pile of recumbentfolds (partially eroded in the lowermost Carboniferous).

(2) During the recumbent folding, coeval ductile shearingwas especially prominent in the overturned limb, generating awell-developed mylonitic planar–linear fabric and more intensestrain. In terms of metamorphism, IC data from the over-turned and normal limbs do not show significant differences,indicating a uniform metamorphic temperature for all the(preserved) recumbent folds. However, relevant differences interms of mineral parageneses and illite polytypes have been

Fig. 8. Distribution of IC and some metamorphic index mineralsaccros the upright antiform in the Espiel sector. The horizontal scalestarts at the southern boundary of The Pedroches Batholith(see Fig. 1b).


found between the normal and overturned limbs: while thenormal limbs only contain muscovite � chlorite and both illitepolytypes 1Md and 2M1, the overturned limb also containsparagonite, muscovite/paragonite mixed layers, pyrophyllite,chloritoid and only the 2M1 illite polytype. We interpret thesedifferences in relation to the intense shearing and mylonitiz-ation concentrated in the overturned limb. Metamorphic fluidsand/or energy supplied through strain could have enhanced,via reaction progress, the growth of new metamorphic miner-als and the complete transformation of the illite polytypes.

(3) After a thick synorogenic sedimentation and volcanism(Early Carboniferous) and brittle thrusting (Mid-Carboniferous) of the allochthonous unit onto the parautoch-thonous unit, upright folds were produced in the whole regionin diagenesis to anchizone (<300 �C) metamorphic conditions(M2). Pressure has been estimated at a minimum of1–1.5 kbar, which could correspond to the minimum lithostaticload due to the thick Carboniferous sedimentary sequence.

(4) In the allochthonous unit, there is an unconformitybetween the Carboniferous rocks (affected only by uprightfolds and M2) and the underlying sequence (also affected byprevious recumbent folds with ductile shearing and M1). Thisunconformity is also reflected by the metamorphic gap be-tween the pre-Carboniferous rocks and the Carboniferousones, the latter having only low to high anchizone-grademetamorphism. This metamorphic gap is minimized by thethermal effect due to the existence of volcanic rocks interlay-ered in the Carboniferous sequence. In the parautochthonousunit, there is no structural evidence for an unconformity at thebase of the Carboniferous rocks. Additional support forthe lack of this unconformity is provided by the fact that thepre-Carboniferous sequence does not display a higher-grademetamorphism than the Carboniferous one.

(5) Metamorphic grade and estimated pressure were higherfor M1 than for M2, which indicates that deformation duringM1 was produced at a deeper structural level than during M2.The exhumation of the rocks affected by M1 towards ashallower structural level must have been produced by erosionof part of a pile of recumbent folds and by thrusting of theallochthonous unit onto the parautochthonous one.

(6) In the Espiel sector of the parautochthonous unit, ametamorphic gradient across a kilometre-scale antiform hasbeen detected. In the hinge zone, IC values are typical of thelate diagenesis and kaolinite is present, while in the limbs ICvalues correspond to the diagenesis–anchizone boundary andthe chlorite zone is reached. These differences can be inter-preted in terms of metamorphic disequilibrium rather thandifferent temperatures. Therefore, a greater cleavage develop-ment and strain in the limbs could have kinematically favouredthe metamorphic reaction progress, thus allowing the rocks ofthe limbs to attain, or come closer to, equilibrium conditions.

Martin Frey, José Barrenechea and two anonymous reviewers arethanked for their helpful comments on an earlier version of themanuscript, and Ray Burgess for his editorial assistance. We gratefullythank Christine Laurin and F. Gonzálvez Garcıa for their help withthe English version. Financial support was provided by the CICYT(Spain) Projects PB93/1149/Co3/01, PB96/1452/Co3/01 and BTE2000/1490/Co2/01.

ReferencesA, L., L, K. & R, M. 1993. Germanium muscovites with

excess hydroxil water, KAl2[Ge3-xAl+xO10-x(OH)*(OH)2] and thequestion of excess OH in the natural muscovites. European Journal ofMineralogy, 5, 19–20.

A, A. & BÈ, M. 1997. Low-Pressure Metamorphism in the SierraAlbarrana Area (Variscan Belt, Iberian Massif). Journal of Petrology, 38,35–64.

A, A., GÁ L, F. & S, J.F. 1994a. Tectonic evolutionof the boundary between the Central Iberian and Ossa-Morena Zones(Variscan Belt, southwest Spain). Tectonics, 13, 45–61.

A, A., G L, F., MÍ P, D. & S, J.F.1994b. Regional significance of kilometric-scale NE-vergent recumbentfolds associated with E-to SE-directed shear on the southern border ofthe Central Iberian Zone (Hornachos-Oliva region, Variscan belt, IberianPeninsula). Geologische Rundschau, 83, 377–387.

B, A. 1976. The interrelations of fluid transport, deformation, geo-chemistry and heat flow in early Proterozoic shear zones in the Lewisiancomplex. Philosophical Transactions of the Royal Society of London, A280,569–604.

B, R.E. & R, D. 1988. Low grade metamorphism of theWelsh Basin Lower Palaeozoic succession: an example of diastathermalmetamorphism? Journal of the Geological Society of London, 145,363–366.

Fig. 9. Sequential schematic drawing summarizing the tectonometamorphic evolution of the southern Central Iberian Zone.


B, J.P., I, M., L, P., M, P. & R, A. 1981. Variscanintracontinental deformation: The Coimbra-Córdoba Shear Zone (SWIberian Peninsula). Tectonophysics, 78, 161–177.

D T, M., S, W.B. & F, M. 1994. Determination of whiteK-mica polytype ratios: comparison of different XRD methods. ClayMinerals, 29, 717–726.

DÍ B, M.A., V, R. & G L, F. 1990. Structure in theCentral-Iberian Zone. In: D, R.D. & MÍ GÍ, E. (eds)Pre-Mesozoic Geology of Iberia. Springer Verlag, Berlin-Heidelberg, 172–188.

D, D. 1969. A technique of finite strain analysis using elliptical particles.Tectonophysics, 7, 117–136.

E, M.A. & H, B.E. 1974. Chemical and deformational controls onrecrystallization of mica. Contributions to Mineralogy and Petrology, 43,111–124.

FÁ C, J.C. & GÁ, E. 1992. Influence of tectonic factors onillite crystallinity: a case study in the Iberian pyrite belt. Clay Minerals, 27,385–388.

F, M.P. & J, P.B. 1999. Stratigraphic controls on deformationpatterns in fault-related folds: a detachment fold example from the SierraMadre Oriental, northeast Mexico. Journal of Structural Geology, 21,613–633.

F, W. & L, G. 1974. Beziehung zwischen tektonischerDeformation und Illit-kristallinität. Neues Jahrbuch fur Geologie undPaläontologie Abhandlungen, 146, 325–326.

F, D. 1965. On the symmetry principle and the deformation ellipsoid.Geological Magazine, 102, 36–45.

F, M. 1987. Very low-grade metamorphism of clastic sedimentaryrocks. In: F, M. (ed.) Low temperature metamorphism. Blackie,Glasgow, 9–58.

F, M., TÜ, R., M, J., K, B., B, A., G,U. & S, B. 1980. Very low-grade metamorphism in external partsof the Central Alps, illite “cristallinity”, coal rank and fluid inclusion data.Eclogae Geologicae Helvetiae, 73, 173–203.

F, N. 1979. Random point distributions and strain measurement in rocks.Tectonophysics, 60, 89–105.

G, C.V. 1984. Micas in metamorphic rocks. In: B, S.W. (ed.) Micas.Mineral Society of America, Reviews in Mineralogy, 13, 357–467.

G, C.V. & S, F.P. 1986. Classification and correlation ofmetamorphic facies series by means of muscovite b0 data from low-grade metapelites. Neues Jahrbuch fur Mineralogie, Abhandlungen, 153,363–380.

G, C.V., M, C., S, F.P. & B, J.G. 1992. Composi-tional controls on the cell dimensions of 2M1 muscovite and paragonite.European Journal of Mineralogy, 4, 283–297.

GÉ A, G. & N, F. 1996. White-mica ‘crystallinity’, finite strainand cleavage development across a large Variscan structure, NW Spain.Journal of the Geological Society, London, 153, 287–299.

H, M.J. 1971. Stability of andalusite and the aluminium silicate phasediagrams. American Journal of Science, 271, 97–131.

J, M.R.W. & O, G.J.H. 1990. Precollision and postcollision thermalevents in the Himalaya. Geology, 18, 753–756.

J, M., FÉ, J.M., R, A. & C, L. 1972. Mapa tectónicode la Penınsula Ibérica y Baleares. E 1:1.000.000. Instituto Geológico yMinero de España.

J, T.C., B, P.E. & B, S.W. 1987. NH4-bearing illite in very lowgrade metamorphic rocks associated with coal, northeastern Pennsylvania.American Mineralogist, 72, 555–565.

K, W. & S, G. 1975. Neue Ergebnisse über die Geologie der Kern-undMantelschichten des Hohen Venns auf Grund von Profilaufnahmen bei derVerlegung der Erdgasleitung Aachen-Reinfelden. Zeitschrift der deutschengeologischen Gesellschaft, 126, 1–14.

K, H.J. 1983. Mineralogy and petrology of burial diagenesis (burial meta-morphism) and incipient metamorphism in clastic rocks. In: L, G. &C, G.V. (eds) Diagenesis in Sediments and Sedimentary Rocks. 2.Elsevier, Amsterdam, 289–493.

K, H.J. 1987. Correlation between indicators of low-grade metamorphism.In: F, M. (ed.) Low temperature metamorphism. Blackie, Glasgow,227–300.

K, H.J. 1991. Illite ‘crystallinity’: recommendations on sample preparation,X-ray diffraction settings and interlaboratory samples. Journal of Metamor-phic Geology, 9, 665–670.

K, R.J. 1981. The interaction of deformation and metamorphism in slates.In: L, G.S., B, H.J., W, K. & Z, H.J. (eds) The Effect ofDeformation on Rocks. Tectonophysics, 78, 249–272.

K, R. 1983. Symbols for rock-forming minerals. American Mineralogist, 65,277–279.

KÜ, B. 1968. Evaluation quantitative du métamorphisme par la cristallinitéde l’illite. Bulletin du Centre de Recherches Pau-SNPA, 2, 385–397.

L, P.H. & KÉ, J.R. 1993. High-temperature metamorphism in amajor strike-slip shear zone: the Ailao-Red River, People’s Republic ofChina. Earth and Planetary Science Letters, 118, 213–234.

L, R.J. 1977. Estimations of tectonic strain ratio from the mean shape ofdeformed elliptical markers. Geologie en Mijnbouw, 56, 140–144.

LÓ M, A., N, F., SÁ P, E. & V, N. 1991.The composition of phyllosilicates in Precambrian low-grade-metamorphic,clastic rocks from the Southern Hesperian Massif (Spain) used as anindicator to metamorphic conditions. Precambrian Research, 53, 267–279.

MÍ P, D. 1997. Estructura del borde meridional de la ZonaCentroibérica y su relación con el contacto entre las Zonas Centroibérica y deOssa-Morena. Tesis Doctoral, Univ. Granada.

MÍ P, D., S, J.F., A, A. & GÁ L, F.1995a. La estructura del borde meridional de la Zona Centroibérica en elsector suroriental de la Provincia de Badajoz. Revista de la SociedadGeológica de España, 8, 41–50.

MÍ P, D., A, A., GÁ L, F. & S, J.F.1995b. Timing of the Variscan structures on both sides of the Ossa-Morena/Central Iberian contact (southwest Iberian Massif). Comptes Rendus del’Academie des Sciences de Paris, Serie IIa, 321, 609–615.

MÍ P, D., S, J.F., A, A. & GÁ L, F.1998a. Evolution of a Carboniferous piggy-back basin in the southernCentral Iberian Zone (Variscan Belt, SW Spain). Bulletin de la SocietéGeologique de France, 169, 573–578.

MÍ P, D., S, J.F., A, A. & GÁ L, F.1998b. La estructura del borde meridional de la Zona Centroibérica(Macizo Ibérico) en el Norte de la Provincia de Córdoba. Revista de laSociedad Geológica de España, 11, 87–94.

M, H.J. & S, W. 1987. Phengite geobarometry based on thelimiting assemblage with K-feldspar, phlogopite and quartz. Contributionsto Mineralogy and Petrology, 96, 212–224.

M, D.T. & H, J. 1967. High-grade diagenesis and low-grademetamorphism of illite in the Precambrian Belt Series. American Mineral-ogist, 52, 843–857.

M, R.J. & F, M. 1999. Patterns of very low-grade metamorphism inmetapelitic rocks. In: F, M. & R, D. (eds) Low-Grade Meta-morphism. Blackwell Science, Oxford, 61–107.

M, R.J. & P, D.R. 1999. Very low-grade metapelites: mineralogy,microfabrics and measuring reaction progress. In: F, M. & R,D. (eds) Low-Grade Metamorphism. Blackwell Science, Oxford, 10–60.

M, R.J., R, B. & P, D.R. 1990. A transmission electronmicroscope study of white mica crystallite size distribution in a mudstone toslate transitional sequence, North Wales, U.K. Contributions to Miner-alogy & Petrology, 106, 27–40.

M, R.J., R, B., P, D.R. & H, S.R. 1995. Strain-relateddifferences in the crystal growth of white mica and chlorite: a TEM andXRD study of the development of metapelitic microfabrics in the SouthernUplands thrust terrane, Scotland. Journal of Metamorphic Geology, 13,559–576.

N, F. & SÁ N, A. 1994. A comparative XRD and TEM study ofthe physical meaning of the white mica “crystallinity” index. EuropeanJournal of Mineralogy, 6, 611–621.

N, R. 1985. Illite crystallinity in Devonian slates of the Meggen Mine(Rhenish Massif). Neues Jahrbuch fur Mineralogie. Monatshefte, 1985,268–276.

O, R., M, ML., G, D.R., F, D.A. & B, R. 1998.Crystallinity and b0 spacing of K-white micas in a Paleozoic accretionarycomplex, Eastern Australia: metamorphism, paleogeotherms, and struc-tural style of an underplated sequence. Journal of Geology, 106, 495–509.

O, M., A C, F.M. & N, F. 1998. Development of largenorth-facing folds and their correlation to crustal extension in the Alborándomain (Alpujarras region, Betic Cordilleras, Spain). Tectonophysics, 298,271–295.

P, A., K, H.J. & S, R. 1982. Use of the lattice parameter b0 ofdioctahedral illite/muscovite for the characterization of P/T gradients ofincipient metamorphism. Contributions to Mineralogy and Petrology, 79,85–95.

R, J.L., B, J.M., D V, G. & H, M. 1993. Fluorine inmicas: crystal-chemical control of the OH-F distribution between triocta-hedral and dioctahedral sites. European Journal of Mineralogy, 5, 7–18.

R, B. & M, R.J. 1985. The distinction between Caledonian burialand regional metamorphism in metapelites from North Wales: an analysisof isocryst patterns. Journal of the Geological Society of London, 142,615–624.

R, B., E, J.A., M, R.J. & S, M. 1989. Discussionon Low grade metamorphism of the Welsh Basin Lower Palaeozoicsuccession: an example of diastathermal metamorphism? Journal of theGeological Society of London, 146, 885–890.

R, D. & M, R.J. 1999. Low-temperature metamorphism: anoverview. In: F, M. & R, D. (eds) Low-Grade Metamorphism.Blackwell Science, Oxford, 1–9.

R, D., W, L.N. & B, R.E. 1990. The illite ‘crystallinity’technique: a critical appraisal of its precision. Journal of MetamorphicGeology, 8, 333–344.

S, F.P. & S, A. 1974. The b0 value of the potassium white micas as abarometer indicator in low-grade metamorphism of pelitic schist. Contri-butions to Mineralogy and Petrology, 45, 143–152.

S, J., M, G. & S, J.M. 1991. Fluid channelling duringductile shearing: transformation of granodiorite into aluminous schist in


the Tauern Window, Eastern Alps. Journal of Metamophic Geology, 9,419–431.

S, J.F., MÍ P, D., EÓ, I., A, A. & GÁL, F. 2001. Structure of the SW Iberian Massif on Both Sides ofa Variscan Suture: the Ossa-Morena / Central Iberian Contact. Tectono-physics, 332, 295–308.

T, O.T., B, M.D., V, R.H. & P, S.R. 1991. Fluid-enhanced deformation: transformation of granitoids to banded mylonites,western Sierra Nevada, California, and southeastern Australia. Journal ofStructural Geology, 13, 1137–1156.

T, R.J. & M, E.M. 1992. Structural Geology. W.H. Freeman andCompany, New York.

U, B.M., L, M.M. & K, S.M. 1993. A comparisonamong organic and inorganic indicators of diagenesis and low-temperaturemetamorphism, Tertiary Shimanto Belt, Shikoku, Japan. In: U,M.B. (ed.) Thermal evolution of the Tertiary Shimanto Belt, southwestern

Japan: An example of ridge-trench interaction. Geological Society ofAmerica Special Papers, 273, 45–61.

U, P. 1998. Modelling localization of deformation and fluid flow in acompressional orogen: implications for the Southern Alps of New Zealand.American Journal of Science, 298, 296–323.

W, H., F, M. & S, W.B. 1996. Diagenesis and metamorphism of clayminerals in the Helvetic Alps of Eastern Switzerland. Clays and ClayMinerals, 44, 96–112.

W, L.N. & R, H.N. 1994. Interlaboratory standarization and calibrationof clay mineral “crystallinity” and crystallite size data. Journal of Meta-morphic Geology, 12, 141–152.

W, L.N., G, R.O. & Z, E. 1996. Thrust-related verylow grade metamorphism in the marginal part of an orogenic wedge,Scandinavian Caledonides. Tectonics, 15, 1213–1229.

W, S.H. & K, R.J. 1978. Microstructure and cleavage development inselected slates. Contributions to Mineralogy & Petrology, 66, 165–174.

Received 17 March 2000; revised typescript accepted 30 April 2001.Scientific editing by Ray Burgess.


Documents

Relationships between very low-grade metamorphism …hera.ugr.es/doi/15087864.pdf · Relationships between very low-grade metamorphism and tectonic deformation: examples from the