Transmission Electron Microscopy Study of Very Low-Grade

ClayMinerals (2003)38, 459–481

Transmissionelectronmicroscopystudyof verylow-grademetamorphicevolutionin NeoproterozoicpelitesofthePuncoviscana formation(CordilleraOriental, NWArgentina)

1, 2 M . D O C A M PO * AN D F. N I E TO

1 Institutode Geocronolog ´ga y Geolog´gaIsoto´picaand Facultadde Ciencias Exactas y Nautrales,U.B.A., Pabello ´n INGEIS,Ciudad Universitaria, 1428-Buenos Aires, Argentina, and 2 Departamentode Mineralog ´gayPetrolog ´ga and I.A.C.T.,Universidadde Granada-CSIC, Avda. Fuentenueva s/ n,18002-Granada, Spain

(Received 17 April 2003; revised 19 May 2003 )

ABSTRACT:The Puncoviscana Formation, largely cropping out in NWArgentina, is mainly composed of apelite-greywacke turbidite sequence affected by incipient regional metamorphism and polyphase deformation. Metapelites, composed mainly of quartz, albite, dioctahedral mica and chlorite, were sampled in the Lules-Puncoviscana and Choromoro belts. Lattice-fringe images, selected area electron diffraction and analytical electron microscopy analyses, coupled with previous data from white mica crystallinity index, indicate astate of reaction progress for Puncoviscana slates consistent with medium anchizone- to epizone-grade metamorphism. The 2 M polytype prevails in dioctahedral micas, coexisting in afew cases with the 1 Md polytype as aconsequence of lack of equilibrium. The 2 M polytype coexists with 3 T in two slates and long-range four-layer and ten-layer stacking sequences were identified in another sample. Samples with 3 T and long-range stacking sequences present b values characteristic of intermediate–high pressure metamorphism and ordered chlorites (1 L, 2L, 3L and 7L) prevail. Based on the Sicontents of dioctahedral micas and considering peak temperatures of ~350 –400ºC, pressures from 5kbar and 5 –7kbar were derived for metapelites from the Lules- Puncoviscana and Choromoro belts, respectively. These values agree with facies series derived from the b values. Micas with awide range of phengitic substitution, as evidenced by Fe +Mg and Si contents, coexist. These variations could not arise from the disturbing effect of detrital white K-mica because TEMevidence indicates that they are absent or represent <10% of the mica population. Thus, compositional variations suggest that dioctahedral micas of individual slates crystallized at different pressure conditions in response to the P-T path of the metamorphism. Moreover, in several biotite-free slates the illite crystallinity (IC) values lead to an underestimation of the metamorphic grade attained in these rocks. The coexistence of ICcorresponding to anchizone and the occurrence of biotite in some slates and View metadata, citation and similar papers at core.ac.uk brought to you by CORE felsic metavolcanicprovided by Digital.CSIC rocks intercalated in the Puncoviscana metasediments are interpreted to be the result of ametamorphic path including arelatively high-pressure/ low-temperature (H P/LT) event, followed by alower-pressure overprint possibly at higher temperatures than the H P/LT event. Small micas formed during the high-pressure stage would prevail in the <2 mmfraction, producing anchizone IC.

KEYWORDS:CentralAndes, Puncoviscana Formation, intermediate Na-K dioctahedral micas, wonesite, mixedlayers, illite crystallinity.

*E-mail: [email protected] DOI: 10.1180/0009855033840109

# 2003The Mineralogical Society 460 M. Do Campo and F.Nieto

Thereconstruction of P-T pathsfor medium- to datawould pro videa constraintfor furthe r high-graderock sisa commonobject ivein geodynamicmodels of the area. petrologicalstudies. However, in clastic sequences TheHRTEM andAEM techniqueshave been affectedby very low- to low-grade metamorphism employedin recent years to study numerous thesevariables and particula rlytheir temporal sequencesaffected by incipient metamorphi sm evolutionare difficult to establish, often because (Lee et al.,1984;Lee & Peacor,1985; Lee et al., ofa lackof reliable indicator minerals. Moreover, 1986;Dalla Torre et al.,1996;Dong & Peacor, considerablecontrov ersyexists about whether 1996;Lo ´pezMunguira & Nieto,2000; Warr & equilibriumconditions are reached in incipient Nieto,1998; Schmidt and Livi, 1999). A compre- metamorphism.Merriman&Peacor(1999) hensivereview can be found in Merriman & Peacor compiledconsiderable evidence of chemical and (1999). texturaldisequilibrium in pelites and metapelites formedat sub-greenschist conditions that invalidate GEOLOGYANDSAMPLE theuse of clay-mineral-rel ated‘ geothermometers’ DESCRIPTIONS foraccurated eterminationoftemperature. However,the same a uthorsrec ognizedthat ThePuncoviscana Formation (Turner, 1960), which mineraltransitio nsin pelites generall yfollow constitutesthe basement of thestudy area, is mainly well-constrainedsequ encesas a functionof composedof a pelite-greywacketurbiditesequence, metamorphicgrade. They pointed out that there is withsubordinate sandstone and locally interbedded ageneraltrendto de creasingdisord erwith conglomerates,shelf limestone and volcanic rocks increasinggrade, which is consistent with a series (Omarini,1983; Omarini & Baldis,1984; Jezek, ofprogradetransitions in whichpelites pass through 1990).This unit was affect edby polyph ase asequenceof metastable states as they approach deformationand has eastern and western tectonic stablech emicalequili brium.In this contex t, units,tending NS whichshow contrasting tectonic Merriman& Peacor(1999) postulated that the evolution(Mon & Hongn,1991, 1996) (Fig. 1). The specificstate of a givensystem can be characterized easternLules-Puncoviscan abeltis characterized by ascorresponding to a ``stateof reaction progress ’’ asymmetricfolds overturned to the west, with definedby data obtained by X-ray diffraction concomitantd evelopmento fana xialplane (XRD) ortransmission electron microscopy (TEM). cleavage.A secondsup erposedde formation Fine-grainedrocks, in which intergrown phyllo- produceda crenulationcleavage and associated silicatesare the dominant constituent s,require microfolding,which affects the bedding and the sophisticatedtechniques for study, such as high- axialplane cleavage. The western Choromoro belt resolutiontransmissionelectronmicroscopy ismarked by tight chevron folds overturned to the (HRTEM)andanalyti calelectro nmicroscopy east,accompanied by an east-dipping axial plane (AEM). foliation.Superimposed on these folds are other Thepurpose of this paper is to establish the P-T cleavagesurfaces that are best developed in the conditionsformetapeliticrocksofthe westernpart of the Choromoro belt. The main NeoproterozoicPuncoviscana Formation, affected deformationof the Pu ncoviscanaFormation byincipient regional metamorphism, based on the occurredduring the latest Proteroz oic–earliest detailedstudy of representative samples at TEM CambrianBraziliano orogeny (Ramos & Basei, scalecoupled with previously published data on the 1997;Do Campo, 1999b). crystallinityindex and b parameterof white mica TheSanta Rosa de Tastil Granitic Complex is an (Do Campo et al.,1998;Do Campo, 1999a,b). epizonalpost-tectonic body that developed a narrow ThePuncoviscana Formation crops out exten- contactzone that comprises an inner hornfels sivelyin the Cordillera Oriental and to a lesser aureoleand an outer zone of semi-metamorphose d extenton the eastern border of Puna (Salta and pelites(Kilmurray et al.,1974).However, only part JujuyProvinces), in NW Argentina.The compre- ofthis complex batholith belongs to the basement, hensionofits metamo rphice volutionis an becauserecent investigations have demonstrated importantke yforth eunderstandingoft he thatthe red granite facies intrude Eopalaeozoic Neoproterozoic-CambricevolutionofSouth sediments(Hongn et al.,2001b).The Puncoviscana America.In addition, a betterknowledge of its Formationis uncomformably overlain by Cambrian metamorphicevolution outlined from mineralogical sandstoneof the Meson Group. TEMstudy of pelites of the Puncoviscana Formation 461

67º 66º 65º BOLIVIA PARAGUAY BRASIL B O L I V I A Tarija N

URUGUAY 22º

C T

LOCATION E MAP B

REPUBLICA

ARGENTINA O

60º R

74º 25 º

O M 23º 0 500 Km

9 0º

L R

C H I L E E O

La Colorada B H El Niño 42 C

Muerto Hill 39 A

36 N A

C 24º

S Jujuy I

26 V

A R G E N T I N A O

Mojotoro C

Range N

U 20 51 Puncoviscana 48 P Formation Salta - Precambrian- S 25º Cambrian granites E Ordovician L magmatism U L Fault Guachipas Range 3 Sample

0 50 100 Km 3 4 26º

FIG.1. Geological sketch and location map.

Illitecrystallinity values for Lules Puncoviscana 0.25ºD2y),(Table1). The medium anchizonal to beltand the Puna sector of Choromoro belt are in epizonalgrade indicated by IC datais in agreement therange of high anchizone and the anchizone– withprevious results (Toselli & Rosside Toselli, epizoneboundary (Table 1, data from Do Campo et 1982;Toselli, 1990). al.,1998;Do Campo, 1999a) whereas values for the Thepresence of prograde biotite in several CordilleraOriental section of Choromoro belt metapelitesfrom the Sierra del Cobre area and correspondto anchizone showing a slightlyhigher felsicmetavol canicrocks interc alatedin the averagevalue and more scattering (range 0.36 to sedimentarysequence at El Nin˜oMuertoHill 462 M. Do Campo and F.Nieto indicatesthat greenschist-fac iesconditions were northernmostMojotoro Range. Metalimolite PU51 reachedin the eastern border of Puna. However, showscurved crystals of trioctahedral mica up to accordingto Hongn et al.(2001a),Ordovician 350 mmlongand interleaved phyllosilicate grains magmaticbodies produced a strongmetamorphic composedof chlorite + trioctahedralmica as well as overprinton volcano-s edimentaryrocksin La chlorite+ phengite(Do Campo, 1999b). Samples Coloradaarea 15 –25km no rthof Cerril los PU20,PU26 and PU42 correspond to the Choromoro School;therefore the presence of biotite could beltin the Cordillera Oriental area; the last one was alsobe interpreted as the result of Ordovician chosenbecause it presents a markedlyhigher IC overprint. value(see Table 1), it wasalso the only metapelite in Then ines tudiedmetapelitesaremainly whichminor amounts of detrital K-feldspar were composedof detrital quartz and albite plus newly identifiedat SEM scale.In turn, samples PU36 and formedphyllo silicates(Table 1). Clasts were PU39are representative of greenschist-facies slates reducedin size by pressure solution, alteration fromthe eastern border of Puna. The occurrence of andbreakdown, which produced elongated quartz progradebiotite in several slates from this area, andplagioclase grains with preferred orientation. includingslate PU36 (Cerrill osSchool), was Newlyformed white mica and chlorite flakes, corroboratedby quantitative analysis at SEM scale 10 –50 mminlength, show preferred alignment (DoCampo, 1999b). thatgive place to a continuousslate cleavage in phyllosilicate-richmetapelites (PU4, PU26, PU36 TRANSMISSIONELECTRON andPU39) and a discontinuousschistosit yin M I C R O S C O P Y quartz-plagioclaserich samples (PU3, PU20, PU42 andPU51). Metamorphic biotite lying sub-parallel Representativesamples for TEM observationswere tothe main foliation was optically identified in chosenbased on prior observations by optical severalslates from the eastern border of Puna and microscopyand scannin gelectronmicrosco py thencorroborated by quantitative analysis at SEM (SEM).Sampleswerep reparedasCanada andTEM scale(Do Campo, 1999b). Interleaved Balsam-mountedthin sections oriented approxi- phyllosilicategrains composed of chlorite and matelynormal to bedding and slaty cleavage. muscoviteor chlorite and biotite can be optically Theywere further thinned using a Gatan600 ion recognizedin many rocks. As accessory minerals, milland carbon-coated for TEM observationwith a zircon,apatite, ilmenite, hematite, iron-titanium PhilipsCM-20 scanning transmis sionelectro n oxidesand monazite are present. Figure 2 presents microscope(STEM) equippedwith an ultra-thin BSE imageswhich illustrate the typical texture of windowEDX detector(Centro de Instrumentacio ´n ananchizonal and a biotite-bearingPuncoviscana Cientõ´fica,C.I.C., Granada University). metapelite. Quantitativeanalyses (AEM) wereobtained only SamplesPU3 andPU4 arerepresentative of slates fromthin edges, using a 70A Ê beamdiameter and a fromthe Puncoviscana Formation in the Guachipas 10006200 AÊ scanningarea, with the long axis Range,while PU48 and PU51 correspond to the orientedparallel to the phyllosilicate packets. The

TABLE 1. Mineral data for Puncoviscana Formation metapelites. Illite crystallinity index from Do Campo (1999a), expressed in CISscale (Warr &Rice, 1994). Limiting values for the low-grade and high-grade boundary of the anchizone are 0.42ºand 0.25ºD2y respectively. ( n)=number of samples. Mineral associations do not necessarily represent equilibrium assemblages. Mineral abbreviations after Kretz (1983).

Tectonic Geological Mineralogy IC (ºD2y) unit zone Range Average (n)

Lules- Cordillera Phen +Chl ±Sm 0.31 –0.24º0.28º ± 0.02 (9) Puncoviscana belt Oriental Choromoro belt Cordillera Phen + Chl 0.36 –0.25º0.30º ± 0.03 (17) Oriental Puna Phen +Chl ±Bt± Sm 0.30 –0.24º0.27º ± 0.02 (8) TEMstudy of pelites of the Puncoviscana Formation 463

FIG.2. BSEimages of Puncoviscana metapelites: (a) biotite flake parallel to slate cleavage in an epizonal slate, (b) interleaved phyllosilicate grains of chlorite +muscovite in an anchizonal slate. Grains of intermediate contrast are quartz and albite. samplewas tilted 20º towards the detector, giving Lattice-fringe images and electron diffraction anX-ray take-off angle of 34º . Standardsused to patterns obtainthe K-factors for the transformati onof intensityratios to concentrat ion,following the Thephyllosilicates identified in all samples were methodsofCliff&Lorimer(1975)and dioctahedralmica and chlorite + smectitein several Champness et al. (1981),were: albite, biotite, cases.Mo rec ommondefectsare low-an gle spessartine,muscovite, olivine, titanite, CaSO 4 and boundariesbetween packets of the same mineral MnSO4.Theultra-thin window detector allows (Fig.3) or between mica and chlorite packets; less oxygenqualitativ eidentification,but since its frequently,coherent adjacent packets can also be numberof X-ray counts is strongly affected by observed.The dominant polytype in dioctahedral smalldifferences in sample thickness, it was not micas is 2M, although 1Md isalso present in quantitativelymeasure d.Atomic concent ration appreciableamounts (Table 2); in a fewcases, ratioswere converted into formulae according to partial 10 AÊ periodicityoccurs in k = 3n rows stoichiometry(number of oxygens in the theoretical withina lineof streaking (Fig. 3, inset). Muscovite formulaeof minerals). Alkali loss is a significant occursas defect-free packets with straight and in problemin the TEM analysisof clay minerals, generalcontinuous 10 or 20 A Ê latticefringes, and particularlyfor defect-r ichminerals (Van der thetypica lmottledtexture (Fig. 4). Besides , Pluijm et al.,1988).The comparison of analyses phengitewith three-layer sequences, interpreted as obtainedfor 15 to 100 s showedthat shorter 3T polytypewere identif iedin two samples countingtimes gave improved reproducibility for (Fig.5a). Some isolated layers at 14 A Ê and at theseelements; therefore, counting times of 15 s 24 AÊ ,interleavedinside dominant mica packets, wereused as a compromisefor major alkali wereobserve dina fewcases. Mica packet analysis(Nieto et al., 1996). thicknessvaries conside rablyin each sample (Table2). In slates PU3, PU4, PU20, PU42 and PU48,thin mica packets in the range 200 –500 AÊ T E M R E S U L T S predominate.Fu rthermore,the distrib utionis Asynthesisof the results obtained in the TEM skewedtowards greater thickness. No sharp break studyis presented in Table 2, while the AEM inthickness distribution that could be attributed to resultsare detailed in Tables 3, 4, 5 and6, for theoccurrence of detrital mica was registered. Mica mica,chlorite ,smectiteand mixed-la yerclay packetsof >2000 A Ê ,whichcould correspond to minerals,respectively. detritalmica, are absent or represent <10% of the 464

TABLE 2. Main data obtained with TEM. The numbers in the polytype rows refer to the frequency of occurrence.

Lules –Puncoviscana Belt Choromoro Belt PU3 PU4 PU48 PU51 PU20 PU26 PU42 PU36 PU39 Ku¨bler index 0.28 0.30 0.26 0.25 0.27 0.27 0.36 0.29 0.26

Diocthaedral mica Polytype 1Md 2 – 1 – 3 4 3 – – 1M 1 – 1 – 2 – 1 – – 2M 5 4 4 3 7 9 7 5 5

Others – – – 1 (3T) – 2 (4L, 10L) – 2 (3T) – M.

Thickness 270 to 400 to – 250 to 160 400 to 100 to 1000 to 200 to Do (in AÊ ) 4900 3500 1900 1650 >3600 >5000 1300* 1600 ap Campo

Trioctahedral mica x x Chlorite n and Polytype Random 2 – – – – – – – –

Semi- 5 2 1 – 2 1 4 1 1 F.

Random Nieto Ordered 1 (1L) 2 (1L, 2L) 2 (1L) 1 (1L) 1 (1L) 3 (1L, 3L) 1 (1L) 5 (1L, 2L, 7L) (3 1L) Thickness 130 to 500 to 2600** 380 to 375 to 200 to 670 to 250 to 100 to (in AÊ ) 3600 1500 950 >1300 2700 5100 3400 750 Mixed-layers Kind of layers 14-14{ 14-14{ – 10-14{ – 10-14{ – 10-14{ – Order Ordered x x Disordered x x x x x Smectite x x – x – x – – x

* Only two packets were measured ** Only one packet was measured { Mixed-layers identified in SAED patterns and lattice-fringe images { Mixed-layers identified only in lattice-fringe images TEMstudy of pelites of the Puncoviscana Formation 465

wasNa, it could be termed wonesite. In the lattice- fringeimage, three packets of mica 25, 50 and 75 layersthick with straight and continuous 10 A Ê latticefringes could be recognized. AEM analysis revealedthe presence of smectite associated with thiswonesite crystal. Two SAED patternsrepresenting long-range mica stackingsequences were obtained in sample PU26. Oneof themhad points with a 40A Ê periodicityin k = 3n rowsof reflections and a 10A Ê periodicityin k = 3n rows,i.e. a four-layerstacking sequence (Fig.6a); in some cases three successive points displaythe same intensity, while the fourth is fainter.The correspon dinglattice-f ringeimage illustratesa packetmore than 1000 A Ê thick with straightand continuous 10 A Ê latticefringes. Theother long-range mica stacking sequence showsreflec tionscorres pondingto a 100A Ê periodicityin k = 3n rows,whereas the k =3 n reflectionshave a 10A Ê periodicity,i.e. a ten-layer stackingsequence (Fig. 6b). This muscovite occurs asdefect-free packets with straight 10 A Ê lattice fringes,except for abundan tdislocationsthat interruptthe basal planes in two packets. Chloritewith both semi-random and ordered FIG.3. SAEDpattern view along [100] (inset) and stackingwerei dentifiedi nSAEDpatterns; lattice-fringe image of 1 M partially disordered phen- orderedpolytypes prevail in samples PU39, PU36, gite. Arrows indicate low-angle boundaries between PU26,PU48 and PU51, while in PU3, PU4, PU42 packets. Sample PU3. andPU20 semi-random stacking predominates. PU3 isthe only metapelite in which disordered chlorite micapopulation. In samples PU26, PU36 and wasalso identified. In general, chlorite exhibits PU39,mica packets thousands of A Ê thick are straightand continuous, defect-free 14 A Ê lattice moreabundant. fringes(Fig. 7). Figure 8 revealsone packet of Aninterlayer-defi cienttrioctahedral mica 1 Md semi-randomchlorite>2600 A Ê thick,no open polytype,distinguished from muscovite by AEM layerswere observed, and the only evidence of (PU51-6,in Table 3), was identified in sample disorderis the existence of scarce interleaved 10 A Ê PU51.Because its predominant interlayer cation and 24 AÊ layers.In chlorite from samples PU3 and

FIG.4. SAEDpattern view along 110 (inset) and lattice-fringe image of 2 M muscovite showing 10 A Ê layers. Arrows indicate low-angle boundaries between packets. Sample PU26. 466 TABLE 3. AEM analyses of dioctahedral and trioctahedral micas.

Wonesite Element PU 3-6 PU 3-8 PU 3-9 PU3-15 PU3-24 PU4-8 PU4-12 PU4-15 PU 48-1 PU51-1 PU51-4 PU51-6 PU51-10 PU51-14

Si 3.27 3.32 3.67 3.40 3.14 3.23 3.24 3.49 3.32 3.12 3.06 2.62 3.42 3.16 AlIV 0.73 0.68 0.33 0.60 0.86 0.77 0.76 0.51 0.68 0.88 0.94 1.38 0.58 0.84

AlVI 1.36 1.45 1.51 1.44 1.70 1.73 1.56 1.46 1.43 1.58 1.85 0.55 1.47 1.54 Mg 0.35 0.37 0.36 0.32 0.22 0.15 0.34 0.27 0.39 0.22 0.12 1.02 0.33 0.28 Fe 0.39 0.35 0.21 0.30 0.14 0.15 0.25 0.33 0.23 0.31 0.09 1.40 0.22 0.22 Ti 0.02 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.02 0.02 0.02 0.00 0.02 0.03 Mn 0.00 0.00 o.oo o.oo 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.03 0.00 0.00 SOct 2.12 2.16 2.08 2.06 2.09 2.03 2.14 2.06 2.06 2.13 2.08 2.97 2.03 2.08 .M. K 0.82 0.67 0.50 0.83 0.81 0.89 0.75 0.68 0.93 0.79 0.84 0.22 0.79 0.93 oDo Na 0.00 0.00 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.54 0.07 0.00 Ca 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Campo SInt 0.82 0.67 0.58 0.53 0.81 0.89 0.75 0.66 0.93 0.79 0.54 0.76 0.86 0.93 n and

Element PU51-15 PU20-6 PU20-9 PU42-6 PU42-8 PU 42-9 PU26-1 PU26-2 PU 26-5 PU26-7 PU26-14 PU26-15 PU26-21 PU26-22 F. Nieto Si 3.11 3.25 3.25 3.37 3.25 3.42 3.42 3.13 3.25 3.35 3.59 3.23 3.30 327 AlIV 0.89 0.75 0.75 0.63 0.75 0.58 0.58 0.87 0.75 0.65 0.41 0.77 0.70 0.73

AlVI 1.57 1.50 1.45 1.49 1.51 1.59 1.54 1.24 1.46 1.48 1.57 1.48 1.53 1.47 Mg 0.21 0.31 0.31 0.34 0.40 0.29 0.13 0.23 0.27 0.27 0.27 0.28 0.20 0.39 Fe 0.32 0.24 0.24 0.18 0.20 0.13 0.28 0.60 0.34 0.30 0.23 0.31 0.31 0.22 Ti 0.03 0.05 0.02 0.02 0.04 0.00 0.05 0.02 0.03 0.02 0.00 0.02 0.03 0.01 Mn 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

SOct 2.13 2.09 2.02 2.03 2.14 2.01 2.00 2.09 2.10 2.07 2.08 2.08 2.06 2.09 K 0.78 0.81 0.63 0.93 0.77 0.36 0.75 0.99 0.65 0.77 0.51 0.87 0.79 0.83 Na 0.00 0.00 0.42 0.04 0.00 0.52 0.00 0.00 0.16 0.00 0.20 0.00 0.00 0.09 Ca 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 SInt 0.78 0.81 1.05 0.97 0.77 0.88 0.75 0.99 0.81 0.77 0.71 0.87 0.79 0.92 TEMstudy of pelites of the Puncoviscana Formation 467

PU4,dislocations changing lattice-fringe direction 7

1 Ê - andoccasional interleaved 10 A layers,sometimes 9 1 6 9 9 2 0 6 2 0 0 2 9 0 9 7 1 0 0 0 0 9 0 0 9 3 ...... open,were observed. Two packetsof chlorite with U 3 0 1 0 0 0 0 2 0 0 0 0 P low-angleboundaries between them are shown in 8

- Fig.9, while in the contact, dislocati onsthat 6 4 9 7 1 0 0 5 1 0 0 1 9 1 8 7 1 1 0 0 0 9 0 0 9 3 ...... resemblemicro kink-bands appear. 3 0 1 0 0 0 0 2 0 0 0 0 U

P Amongordered chlorite sone-(Fig. 5b and Fig.7, inset) and two-layer sequences (Fig. 5c) 5 - 9 1 9 3 2 2 0 6 9 0 0 9 9

1 8 7 1 1 0 0 0 7 0 0 7 wereidentified in several samples, while a three- 3 ...... 3 0 1 0 0 0 0 2 0 0 0 0 U layerstacking sequence was only registered in slate P PU26. The 3L polytypechlorite presents k = 3n 4

- Ê 0 0 3 2 7 2 0 4 9 0 0 9 reflectionswith a 42A periodicity,whereas the k = 9 0 0 8 1 0 0 0 0 9 0 0 9 3 ...... Ê 3 1 1 0 0 0 0 2 0 0 0 0

U 3n reflectionshave a 14A periodicity(Fig. 6c). P Moreover,in slate PU36 a chloritewith long-range

7 Ê

e stackingwa sidentified.It exh ibitsa 98A 1 t 9 1 5 0 7 4 0 6 5 0 0 5 - i t 6 7 2 2 2 2 1 0 5 9 0 0 9 ...... o n 3 periodicityin k = 3 rows,indicating a seven- i 2 1 0 1 1 0 0 2 0 0 0 0 U B

P layerstacking sequence (Fig. 5d). Only five points intercalatedbetween 00 l reflectionswere observed, 3 1 - 1 9 9 1 2 5 0 7 3 0 0 3

6 probablydue to low intensity in some super-order 1 8 3 3 3 0 0 0 0 0 0 0 ...... 3 3 0 1 0 0 0 0 2 1 0 0 1

U reflections.Ordered as wellas semi-randomchlorite P showsstraight and continuous 14 A Ê andoccasion- 1

1 Ê

- 4 6 2 2 6 3 0 2 8 0 0 8 ally 28 A latticefringes, Fig. 7 showsa defect-free 6 2 7 5 2 2 0 0 0 9 0 0 9 ......

3 Ê

3 0 1 0 0 0 0 2 0 0 0 0 packetof more than 2300 A .Inseveral packets, U

P scarceinterlea ved10 A Ê or 24 AÊ layers were

8 identified,in other cases small dislocations were - 0 0 6 9 2 1 0 7 7 4 0 2 6 3 7 5 1 2 0 0 9 9 0 0 0 3 ...... observed.In slate PU36 lateral changes were 3 0 1 0 0 0 0 1 0 0 0 1 U Ê Ê P observedfrom a 14A layerto a 10A layer or from a 14 AÊ layerto a 24A Ê layer,the latter 7 e - t 8 2 0 3 8 7 2 8 5 6 0 1 i 6

t frequentlyassociated with dislocations. 6 3 4 3 1 0 0 9 5 1 0 7 3 ...... o i 2 1 0 1 1 0 0 2 0 0 0 0 U Samplesfrom Sierra de Guachipas and slate B P PU20gave SAED patternsshowing points with

4 Ê Ê - superimposed10 A and 14 A periodicities(Fig. 10, 5 5 2 5 1 3 0 1 6 8 0 3 6 4 5 4 2 3 0 0 0 7 0 0 8 3 ...... inset).The corresponding HRTEM image(Fig. 10) 3 0 1 0 0 0 0 2 0 0 0 0 U P showsthin packets of muscoviteintercalated among chloritepackets, and low-angle boundaries between 3 - 7 3 8 2 1 3 0 4 7 2 0 9 6 0 9 4 2 3 0 0 0 8 2 0 0

3 them.A lattice-fringeimage reveals some inter- ...... 3 0 1 0 0 0 0 2 0 0 0 1 U Ê

P leaved 14 A layerswithin one of the muscovite packets.In western samples (PU26, PU36 and 1 - 9 8 4 7 4 0 3 9 0 0 9

6 PU39)thick defect-free mica and chlorite packets 7 4 2 2 0 0 0 9 0 0 9 3 ...... 0 1 0 0 0 0 2 0 0 0 0 1 U aremore frequent. 2 P 3 Biotite,distinguished from muscovite by AEM 4

2 (PU36-17in Table 3), occurs as a 1 M ordered - 0 0 1 6 1 2 0 0 1 0 0 1 6 3 7 5 2 2 0 0 0 0 0 0 0 ......

2 polytypein sample PU36 (Fig. 11); its b parameter 3 0 1 0 0 0 0 2 1 0 0 1 U Ê P of 9.2 A isalso consistent with a trioctahedral

3 phyllosilicate.Figure 11 shows several packets 150 2 - 4 6 2 2 5 1 0 0 4 0 0 4 ) 6 2 7 5 3 2 0 0 1 8 0 0 8 to290 layers thick that are coherently related or ...... 2 d 3 0 1 0 0 0 0 2 0 0 0 0 t U exhibitlow-angle boundaries with straight and n P o c continuous10 A Ê latticefringes. Lateral transitions ( 3 t from a 10 AÊ layerto a 14A Ê layerlinked with n E e t L I t

c dislocationscould be observed. m V B I V n g n e l l a O a I A i l i e Severalpackets of 10 –14 AÊ mixedlayers were T E S A A M F T M S K N C S identifiedin slate PU36. Figure 5e shows a 468 M. Do Campo and F.Nieto

TABLE 4. Representative AEManalyses of chlorites.

Element PU3-28PU4-6 PU4-7 PU48-2 PU48-3 PU48-4 PU48-5 PU48-6 PU20-3 PU20-14

Si 2.77 2.70 2.57 2.69 2.82 2.88 2.79 2.69 2.59 2.64 AlIV 1.23 1.30 1.43 1.31 1.18 1.12 1.21 1.31 1.41 1.36 AlVI 1.14 1.19 1.17 1.29 1.10 1.11 1.07 1.21 1.51 1.33 Mg 3.03 2.42 2.29 2.21 2.53 2.57 2.63 2.00 1.61 2.01 Fe 1.83 2.37 2.53 2.52 2.37 2.33 2.35 2.85 2.84 2.67 Ti 0.00 0.00 0.00 0.00 0.04 0.00 0.02 0.00 0.00 0.00 Mn 0.02 0.04 0.07 0.00 0.00 0.00 0.00 0.00 0.00 0.00 SOct 6.02 6.02 6.06 6.01 6.04 6.01 6.07 6.05 5.95 6.01 K 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Na 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ca 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Element PU26-12PU26-26 PU36-9 PU36-6 PU36-10 PU36-18 PU36-19 PU39-7 PU39-10 PU39-14

Si 2.78 2.65 2.74 2.70 2.68 2.74 2.56 2.68 2.66 2.64 AlIV 1.22 1.35 1.26 1.30 1.32 1.26 1.44 1.32 1.34 1.36 AlVI 1.30 1.18 0.97 1.12 1.19 1.26 1.22 1.41 1.47 1.44 Mg 2.10 2.47 3.02 2.70 2.86 2.40 2.43 1.79 1.94 1.72 Fe 2.56 2.43 2.13 2.23 2.22 2.30 2.40 2.74 2.50 2.75 Ti 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Mn 0.00 0.00 0.03 0.04 0.00 0.02 0.03 0.00 0.02 0.01 SOct 5.96 6.08 6.15 6.09 6.07 5.98 6.08 5.94 5.92 5.93 K 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.03 0.00 0.02 Na 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ca 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 representativeSAED patternin which the following of0.60 % (Bailey,1982) was obtained for this reflectionswere measured, expressed in A Ê : 24, 12, pattern,indicating an ordered 10 –14 AÊ mixed 7.9,7.3, 6.0, 4.8, 4.0, 3.4, 3.0, 2.8, 2.7, 2.4, 2.2, 2.0, layer.Reflection sat7.3 and 2.8 A Ê were not 1.9,1.7, 1.6, 1.5 and 1.4. A coefficientof variation consideredin the calculation as they probably

TABLE 5. Representative AEManalyses for smectites.

Element PU3-12PU3-14 PU3-22 PU4-9 PU4-16 PU51-7 PU51-8 PU26-13 PU39-9 PU39-16

Si 3.79 4.03 3.60 3.68 3.63 3.61 3.67 3.56 3.41 3.69 AlIV 0.21 –0.03 0.40 0.32 0.37 0.39 0.33 0.44 0.59 0.31 AlVI 1.54 1.40 1.55 1.04 0.79 1.21 1.72 1.59 1.55 1.62 Mg 0.34 0.43 0.38 0.39 0.22 0.41 0.18 0.29 0.24 0.13 Fe 0.20 0.18 0.25 0.72 0.92 0.50 0.21 0.25 0.37 0.19 Ti 0.00 0.00 0.00 0.00 0.04 0.01 0.02 0.00 0.00 0.02 Mn 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 SOct 2.08 2.01 2.18 2.15 1.96 2.14 2.12 2.13 2.16 1.96 K 0.30 0.30 0.49 0.05 0.31 0.10 0.12 0.09 0.23 0.12 Na 0.00 0.00 0.00 0.12 0.18 0.24 0.00 0.26 0.13 0.20 Ca 0.00 0.04 0.00 0.04 0.09 0.01 0.00 0.00 0.00 0.10 SInt 0.30 0.34 0.49 0.21 0.58 0.36 0.12 0.35 0.36 0.42 TEMstudy of pelites of the Puncoviscana Formation 469

TABLE 6. AEManalyses of mixed-layer minerals, mineralformed of chlorite-like layers and trioctahe- structural formulae calculated for 25 (14 +11) dral 10 AÊ layers.Minor amounts of random 10 A Ê / oxygens per formula unit. 14 AÊ mixedlayers were also identified in slate PU26(PU26-11 in Table 6). In this case AEM Element PU3-26PU26-11 PU51-2 PU36-20 analysisdisplays an Fe +Mg/Al ratioslightly higherthan chlorite and a highSi/ Al ratio.These Si 6.12 6.11 5.84 5.89 facts,in addition to the low alkali content, are IV Al 1.88 1.89 2.16 2.11 compatiblewith a chlorite-smectitemixedlayer. AlVI 1.26 1.68 1.59 1.63 Thesemixed-layer minerals will be described in Mg 4.22 3.64 3.23 4.43 detailin a separatepaper (Do Campo & Nieto,in Fe 3.68 3.87 3.81 2.87 prep.). Ti 0.00 0.00 0.02 0.05 Smectitethat occurs in several slates is always in Mn 0.06 0.00 0.00 0.00 packetsclearly discordant with other phyllosilicates SOct 9.21 8.99 8.85 8.97 orin pores of the rock. Discrete smectite packets, K 0.09 0.14 0.59 0.45 subparallelto other phyllos ilicates,were not Na 0.00 0.00 0.65 0.00 observed.The dominant interlayer cation could be Ca 0.00 0.04 0.00 0.00 SInt 0.09 0.18 1.24 0.45 KorNa; Ca present in minor amounts or is absent. Inspite of its higher IC, at TEM scalesample Fe+Mg/Al 2.52 2.05 1.88 1.95 PU42is similar to sample PU20, although they Si/Al 1.95 1.71 1.56 1.58 differin their packet-thickne ssdistribution. Mica packetsof 250 –500 AÊ arepredominant in sample PU20,while packets up to 250 A Ê prevailover those correspondto small chlorite packets interleaved inthe range 250 –500 AÊ inPU42. In addition, in withthe mixed-l ayerand recogn izedin the slatePU42 there are some subordinate mica packets correspondinglattice-f ringeimage. The SAED thickerthan 1100 A Ê . pattern,whose orient ationis c*-b*, presents reflectionswith a 24A Ê periodicityin all levels, AEMcomposition of phyllosilicates suggestingordered stacking and a coherentcrystal- lographicrelation between the two types of layers. Dioctahedralmicas. Table3 showsrepresentative Thevalue of the b parameter(9.2 A Ê )indicatesa analysesof dioctahedral micas. In the interlayer trioctahedralphyllosilicate.The HRTEM image charge-Si/Al diagram(Fig. 12a), illitic substitution displaystwo packets, 600 and 900 A Ê thick, with canbe recognized by the relationship between the straightand continuous apparent 24 A Ê layers. An lowinterlayer charge and the replacement of Si by intermediatelattice fringe corresponding to the 10 Al.Illitic substitution is clearly present in some and 14 AÊ layerscan sometimes be recognized. The micasof samples PU3 andPU4 andin minor 10 –14 AÊ unitis repeated up to seven successive amountsin slate PU26. timesin one packet and ten times in the other. Severalmica analyses belonging to different Interleavedbetweenth esethi n,mi xed-layer samplesshow different degrees of paragonit ic ‘packets’,oneortwo 1 4A Ê layersappear. substitution.Two analyses(PU42-9, PU20-9, see Microanalysisof these packet s(PU36-20in Table3) stand out due to their Na/ (K+Na)ratios Table6) displays Fe +Mg/Al andSi/ Al ratios of0.60 and 0.40. These intermediate Na-K micas closeto the values for a theoreticalbiotite 0.5/ willbe discussed in detail in alatersection, because chlorite0.5mixture. Therefor e,the chemica l theexistence of a miscibilitygap in the muscovite- compositionas well as the b parametersuggest a paragonitesystem has been well corroborated based trioctahedralcharacter not only for the 14 A Ê layers, onnatural and experimental data (Guidotti, 1984; butalso for the 10 A Ê layers.The low interlayer Guidotti et al., 1994a). chargein thestructural formulae could be explained Analysesin general show a largevariation in the bythe excess chlorite layers observed in the lattice- Fe +Mgcontents of dioctahedralmicas within each fringeimage in addition to some loss of K. sample.In the (Fe + Mg)-Sidiagram (Fig. 12b), TheSAED patternsindicative of 10 –14 AÊ mixed mostof the analyses plot above the line for ideal layerswere also identified in slate PU51 (PU51-2 in Tschermaksubstitution ((Mg, Fe 2+)VI, SiIV = AlVI, Table6), the AEM analysesindicate a mixed-layer AlIV),indicatingthat, besides phengitic substitution, 470 M. Do Campo and F.Nieto

FIG.5. SAEDpatterns. (a) [100] image of muscovite three-layer polytype; (b) [100] image of ordered chlorite one-layer stacking sequence; (c) [100] image of ordered chlorite two-layer stacking sequence; (d) ordered chlorite, probably seven-layer stacking sequence; and (e) [100] or equivalent image of 10 A Ê –14 AÊ mixed layer. Sample PU36. aferrimuscovite(Fe 3+ substitutingfor Al) compon- component,with phengit icand ferrimus covite entis a lsopresen t.Guidott i et al. (1994b) substitutionsbeing more relevant. demonstratedthat even in medium redox para- Thefrequency distributions of the Si contentsof geneses,containing ilmenite + magnetite,the Fe 3+/ phengitesare shown in Fig. 13a,b. Only AEM

Fetot ratiois >0.60. Although magnetit ewas analyseswith an interlayer charge of >0.75 atoms positivelyidentified only in one case, in most of performula unit (a.p.f.u.) were included in the thesamples ilmenite was identified by SEM, and histogramsto avoid drastic effects of illitic hematiteby XRD, whichalso indicate oxidizing substitution,which also produce excess Si. In conditions.Therefore, the structural formulae in addition,analyses with an octahedral sum >2.15 +3 Table3 werecorrected according to Fe /Fetot = werenot included. b parametervalues from Do 0.70.This correction produced a moreplausible Campo(1999a,b) were included in thelower part of sumof octahedral cations. thefigure for comparison (Fig. 13). To evaluate the Thus,with the exception of samples PU3 and effectof ferrimuscovitic substitution on white mica PU4,dioctahedr almicas indicate a lowillitic b-geobarometrythe correlation between Si contents TEMstudy of pelites of the Puncoviscana Formation 471

FIG.7. SAEDpattern view along [100] (inset) of chlorite 1-layer polytype and corresponding lattice- fringe image of adefect-free packet showing 14 A Ê layers. Sample PU26.

and b parameterdata from Do Campo (1999b) was carriedout in two ways. Firstly, using the formulae of Guidotti et al. (1989),to calculate the b parameterfrom Si contentsshown as open bars. Secondly,we employed the formulae of Guidotti et al. (1992),based on Fe +Mgcontents, to calculate b parametersshown as grey bars (Fig. 13). The higher b valuesobtained in the second case are a consequenceof the high Fe 3+ contentof these phengites,i.e. the b parameteris affected not only byphengitic but also by ferrimuscovite substitution FIG.6. SAEDpatterns. (a) [100] image of muscovite (Guidotti et al., 1989). four-layer polytype, depicting superstructure reflec- TheSi contentsof phengite from the Lules- tions with 40 A Ê periodicities in rows with k = 3n; (b) [100] or equivalent image of amuscovite ten-layer Puncoviscanabelt varies between 3.06 and 3.50 polytype; 100 A Ê periodicities are visible in k = 3n exhibitinga weakmaximum in the range 3.10 4 Si rows; (c) [100] or equivalent image of ordered chlorite <3.15(Fig. 13). On the other hand, the Si in three-layer polytype, 42 A Ê periodicities are apparent in phengitefrom the Choromoro belt varies from 3.00 k = 3n rows. Sample PU26. to3.45, exhibiting a normaldistribution, they define

FIG.8. SAEDpattern view along [100] (inset) and lattice-fringe image of semi-random chlorite showing 14 A Ê layers and afew interleaved 10 A Ê (short white lines) and 24 A Ê layers. Sample PU48. 472 M. Do Campo and F.Nieto

greaterthan the change in Al IV.Inseveral samples, Fe/(Fe+ Mg)varies by <0.1, suggesting that chemicalequilibrium was reached. In metapelites PU3 andPU48, Fe/ (Fe+ Mg)vary by >0.1, suggestinga lackof chemical equilibrium. For othersamples, this evaluation was not possible becausetoo few AEM analysesof pure chlorites wereobtained.

D I S C U S S I O N State of reaction progress

TheTEM resultsindicate a ‘stateof reaction progress’for the Puncoviscana slates consistent withmedium-anchiz oneto epizone-grad emeta- morphism.This conclu sionis based on the absenceof I-S mixedlayers, the prevalence of the 2M polytypein dioctahedral mica, the thickness of themica and chlorite packets, and the scarcity of crystallinedefects (Merriman & Peacor,1999). Furthermore,the characteristics of the phyllosili- catesin slates PU26, PU36 and PU39 correspond to thefeatures enumerated by Peacor (1992) for epizonalslates, which, according to this author FIG.9. Lattice-fringe image of two chlorite packets with low-angle boundaries between them showing andKisch (1987), do not differ from greenschist- 14 AÊ layers. In the contact, dislocations in the form graderocks. For these samples, IC valueslead to an of micro kink-bands arise (indicated by arrow). Sample underestimationof the metamorphic grade that, in PU20. thecase of slate PU36, could be explained by the broadeningeffect of biotite on the 10 A Ê peak. In amodein the range of 3.25 4 Si <3.35(Fig. 13). theother samples no disturbing effect on the 10 A Ê Thetwo highest Si contentsof 3.42 and 3.45 are peakwas detected, possibly reflecting the influence phengitefrom samples PU26 and PU36. Most of ofstrain features (Giorgetti et al., 2000) or a theSi valueslower than the mode are in mica from particularmica thickness distribution. samplePU39, which is in accordance with its low Directsize measurements made on lattice fringe phengiticcontent proved by AEM analyses(Table 3 imagesshowed that mica packet thickness varies andFig. 12b). considerablyin each sample (Table 2). In slates Chlorite. Table4 showsrepresentative analyses PU3,PU4, PU20, PU42 and PU48, thin mica ofchlorite. Analyses derived from packets that packetsin the range 200 –500 AÊ predominates provedinterlayering with other phyllosilicates in whilein samples PU26, PU36 and PU39 mica lattice-fringeimages were not plotted .Many packetsthousands of A Ê thickare more abundant. analysessh owa trendto id ealTschermak Averages,calculated for samples with at least ten substitutionby plotting close to the (Al VI+ 2Ti + values,vary between 785 and 1500 A Ê . Similar Cr) – 1 vs. AlIV – 1line(Fig. 14a). However, valueswere related by Nieto & Sa´nchezNavas aboutone third of the analyses plot below this line (1994)with high-anchiz onalto epizonal meta- suggestingthat part of the Fe ispresent as Fe 3+ morphismin a systematicstudy on mica crystallite substitutingfor Al inbalancing the negative charge sizesand illite crystallinity. However, other studies onthe tetrahedral sheet. correlatethis entire range of crystallite sizes with Thescatter of analyses on an Al IV – 1 vs. Fe/(Fe low-epizoneconditions (see Fig. 2.1 in Merriman& +Mg)(Fig. 14b) plot indicates no correlation Peacor,1999). betweentetrahedral substitution and (Fe + Mg) Schmidt et al.(1997),based on fluid inclusion contents.The variation in Fe/ (Fe+ Mg)is slightly dataand stable isotope thermometry, correlated IC TEMstudy of pelites of the Puncoviscana Formation 473

FIG.10. SAEDpattern (inset) showing reflections with superimposed 10 A Ê and 14 AÊ periodicities and corresponding lattice fringe showing thin packets of muscovite intercalated among chlorite packets in sample PU4. Arrows indicate low-angle boundaries. frommedium-anchizone conditions with tempera- morphictemperatures of >270º C forpeak condi- tures>270º C. Thus,as the IC valuesfor the tionscan be inferred. An independent estimate PuncoviscanaFormation largely correspon dto basedon the concordance between K-Ar dates over highanchizone and epizone condition s,meta- finefractions and whole rocks (Do Campo, 1999b;

FIG.11. SAEDpattern of 1 M biotite view along [100] (inset) and corresponding lattice-fringe image showing 10 AÊ layers, dislocations (black arrow) and individual layers with different orientations (indicated by white arrow). This image is bidimensional, the cross fringes correspond to b* or 110 showing continuity across twenty layers. Sample PU36. 474 M. Do Campo and F.Nieto

FIG.12. Compositional diagrams for dioctahedral micas.

Do Campo et al.,1999)indicat esthat peak Althoughthe orientation of biotite flakes, parallel metamorphictemperaturesof~350ºCwere toslate cleavage (Fig. 2) supports this interpreta- reachedat least in some localities of the basin. tion,the effect of an Ordovician thermal overprint Moreover,the occurrence of prograde biotite in oneastern Puna Puncoviscana rocks (Hongn et al., severalslates from the eastern border of Puna 2001a,b)cannot be completely ruled out. impliestemperatures of ~400 –450ºC (Bucher& Smectiteoccurring in several slates in packets Frey,1994) for this area. Biotite formation probably clearlydiscordant with other phyllosili catesis occurredaccording to the reaction: interpretedas a retrogradeproduct (Nieto et al., 1994; Zhao et al.,1999).Discrete smectite packets, 8phengite+ chlorite ? subparallelto mic aorc hlorite,were never muscovite+ 3biotite+ 7quartz+ H 2O

FIG.13. Frequency distribution of Sicontents of phengites from slates of Choromoro and Lules-Puncoviscana Belts, determined by AEManalyses. b parameter values from Do Campo (1999b) are shown in the lower part of the figure for comparison. Correlation between the b parameter and the Sicontents were calculated from Fe+Mg contents, grey bars (Guidotti et al.,1992) and Sicontents, open bars (Guidotti et al., 1989). See text for further explanation. TEMstudy of pelites of the Puncoviscana Formation 475

FIG.14. Compositional diagram for chlorites (Laird, 1988). Same symbols as in Fig. 12. observed.In addition, rare interleaved layers at folds.Deforma tionfeatur escould have been 10 AÊ wereidentified in some chlorite packets in producedby the second folding phase, related to samplesPU3 andPU4; in a fewcases layers at thedevelopment of a crenulationcleavage, which 10 AÊ aremore abundant, giving place to random wouldnot produce mineral growth. 10 AÊ /14 AÊ mixedlayers. The occurrence of open Thecomposition of the Na-rich trioctahedral layersin these packets, together with the Si/ Al and micapresent in sample PU51 can be described as (Fe+ Mg)/Al ratiosobtained (PU3-26 in Table 6), anintermediate member of thepreiwekite-aspido lite indicatedthe existence of vermiculite-like layers. seriesexhibiting a solidsolution towards the K Accordingto TEM andXRD observations,these trioctahedralphlogopite-east oniteseries. However, slatesare at least anchizonal and therefore the consideringthat this specimen displays a moderate occurrenceof individual vermiculite-like layers or interlayerdeficiency we favour the name wonesite chlorite/vermiculitemixed-layers is incompatible followingthe IMA nomenclatureof micas (Rieder withprograde metamorphism. Their formation is et al.,1998).However, the original wonesite moreprobably related to the interaction between describedby Spear et al.(1981)is much more Si- rocksand groundwater, under oxidizing conditions, richand interlayer-defic ientthan recorded here. afterthe uplift of the Puncoviscana Formation, Moreover,a furtherTEMstudybyVeb len similarto the retrograde alteration of chlorite to (1983a,b)revealed a complexlamellar intergrowth smectiteat a regionalscale described by Nieto et al. withtalc in the trioctahedral micas studied by (1994)for slates of sub-greenschist facies in Sierra Spear.This situation suggests that further studies Espunã(BeticCordilleras, Spain). A 1-1regular willbe necessary to understand the relations among chlorite-vermiculitemixed layer is the predominant Na-richand K trioctahedralmicas to establish the productof chlorite weathering in natural samples possibleexistence of miscibility gaps between studiedby Banfield & Murakami(1998). This them. inferenceis supported by a LowerOrdovician Rb- Thechlorite-biotite mixed layers identified in Sr whole-rockisochrondate (Do Campo and slatePU36 are probably metamorphic in origin, as Cagnoni2001), discordant with previous K-Ar progradebiotite with no evidence of deformation datesthat these authors interpreted as the time of wasidentified by SEM andTEM inthis sample. upliftof the Puncoviscana Formation. Chlorite-biotitemixed layers probably represent a Crystalsof mica and chlorite with features metastableproduct, an Ostwald intermediate step, indicativeof deformation, such as bending and thatwould probably give place to discrete biotite microkink-bands (Fig. 9), were observed at TEM andchlorit einrespons etoincreas ingmeta- scalecoinciding with evidence of bending and morphism. microfoldingunder the optical microscope and SEM. Bearingin mind that the tectonic studies of Stacking sequencesof phyllosilicates Mon& Hongn(1996) identified two episodes of foldingin the study area, we postulate that the Eventhough the 2 M polytypeclearly prevails in phyllosilicatesrecrystallized syntectonically during dioctahedralmicas, in several cases it coexists with thefirst folding event, which produced the regional the 1Md polytype.Optical and SEM observations 476 M. Do Campo and F.Nieto suggesta metamorphicorigin for the white mica 2L and 7L chloriteas wellas two 3 T phengiteswere 2M polytypebased on the degree of recrystalliza- identified.Both samples also present high b values tionshown by the slates. The 1 Md polytypeusually (9.037and 9.042, respec tively)and high Si predominatesin diagenetic samples (Merriman & contents,as revealed by AEM, whichis character- Peacor,1999). Alternatively, the very low-grade isticof intermediate– high pressure metamorphism. metamorphic2 M polytypeappears in white micas Assuggested by Jullien et al. (1996)and Sassi et inrocks that have been subjected to stress. As the al.(1994)the total pressure of metamorphism, or 2M polytypeis more abundant at higher tempera- the P/T ratio,could in this case be the factor tures,some authors (Velde, 1965; Sassi et al., 1994) controllingthe stacking sequences of phyllosili- haveconsidered that this variable controls its cates.However, the role of temperature cannot be stability.Nevertheless, the coexistence of these underestimatedsince other investiga tionshave twopolytypes over a widerange of temperature is provedits importance in the degree of stacking probablya consequenceof lack of equilibrium orderin trioctahedral chlorites (Schmidt and Livi, (Lo´pezMunguira & Nieto,2000). Consequently, a 1999).The possible influence of pressure in the temperaturefor the 1 Md to 2M transformations stackingorder of chlorites is still an open question. couldnot be defined and therefore could not be employedas a reliablegeothermometer. Intermediate Na/Kdioctahedral micas Moreover,in the Puncoviscana Formation, the 2M polytypecoexists with 3 T phengitein two Apparentintermedi atecomposit ionsbetween samples(PU51 and PU36) in addition, long-range paragoniteand muscovite, similar to those reported four-layerand ten-layer stacking sequences were inthis work (Table 3), have been ascribed by identifiedinanothermetapelite(PU26). differentauthors to averaging between paragonite Coexistencebetween 2 M andlong-range polytypes andmuscovite intergrowths at a verysmall scale hasrecently been reported by Lo ´pezMunguira & (Shau et al.,1991),averagin gofmuscovit e/ Nieto(2000) for potassic white micas from epizonal paragonitemixed layers (Frey, 1969), or to samples.According to Sassi et al. (1994),the factor homogeneousmetastable Na/ Kmuscovite(Jiang thatstabilizes the 3 T polytypein metamorphic &Peacor,1993). More recently, Livi et al. (1997) phengitesis a high P/T ratio. demonstratedthat XRD cannotdistinguish between Inrelation to the stacking sequences of chlorite, paragonite-muscovitemixed layers and an homo- insamples PU39, PU36, PU26, PU48 and PU51 geneous,compositionally intermediate solid solu- orderedpolytypes prevail, while in PU3, PU4, tion.Moreover, Livi et al.(1997)studied by TEM PU42and PU20 semi-random stacking predomi- andEMP sampleMF-925, in which Frey (1978) nates.PU3 isthe only metapel itein which determinedfrom XRD a6:4mixture of paragonite- disorderedchlorite was also identified. In contrast muscovitealong with macroscopic paragonite and tomicas,semi-random stacking is themost frequent muscovite.The EMP andAEM analysesof this typein chlorites. Therefore, the occurrence of many sampleindicated that mixtures of Na-K micas exist orderedchlorites (1 L, 2L, 3L and 7L) in the atscales of 5 mm.In crystals containing sub-equal Puncoviscanasamples is remarkable. butvariable amounts of Na and K, X-raymaps Factorsthat control the stacking sequences of indicatedthat Na is often concentrated on the outer phyllosilicatesinclude pressure, temperature, the edgesof the grains. The Na-rich/ K-richmica crystalgrowth mechani sms,mineral chemica l boundariesin these cases were mostly normal to compositionand deformation history (Schmidt and theapparent basal planes. With this evidence, Livi Livi,1999). Jullien et al. (1996)proposed that et al.(1997)postulated the existence of irregularly pressuremight play an important role in the shapeddomains of Na-and K-rich mica at scalesof stackingsequences of cookeite from pegmatitic <100 AÊ . andmetamorph icenvironme nts.These authors Inthis work, intermediate Na/ Kmicaswere pointedout that ordered polytypes of cookeite revealedby AEM analyses.These micas present behaveas polymorphs remarkably sensitive to total straightand continuous 10 A Ê latticefringes and the pressure. mottledtexture typical of potassic white micas. In Weshould emphasize that in sample PU26, one ourstudy we have found no evidence of different 3L chloriteand two long-range (4 L and 10L) domainsininterm ediateNa-K micas, which phengitescoexist. Furthermore, in sample PU36, supportsthe hypothesis of Jiang & Peacor(1993) TEMstudy of pelites of the Puncoviscana Formation 477 thatthey constitute metastable phases that are evidencedby Fe +Mg,as well as Si contents. replacedby discrete muscovite and paragonite Thesevariations could not arise from the disturbing whenthe metamorphic grade increases. However, effectof detrital white K-mica because TEM weconsider that our data are not conclusive. A evidenceindicates that they are absent or represent simulationof HRTEM imagesof intermediateNa-K <10%of the mica population. Thus, compositional micasat differen tfocalconditio nswould be variationssuggest that dioctah edralmicas of necessaryto determine whether changes in contrast individualslates crystallized at different pressure betweenNa-rich and K-rich layers allow them to be conditionsin response to the P-T path of the distinguished. metamorphism.Moreover, in several biotite-free slates(PU26, PU39, PU42), the IC valueslead to an underestimationof the metamorphic grade attained Physicochemical evolution of phengites inthese rocks. through aclockwiseP-T-t metamorphicpath Ina TEM-AEM andelectron microprobe (EMP) Theassociation of phengite in a non-limiting studyofanchizone-gradeshalesfromthe assemblage,with chlorite here means that the FranciscanComplex (Diablo Range, California), phengitebarometerapplyin gtheSi isopleths Dalla Torre et al. (1996)found that some white diagram(Massone & Schreyer,1987) can be used K-micapackets were zoned, having phengitic cores toderive minimum formation pressures. For peak andmuscovitic rims. These authors postulated that temperaturesof 4 350 –400ºC, pressuresof 5 to themetamorp hicevoluti onof the Francis can 7kbarare derived for the metamorphic rocks from Complexrocks includ eda high-pressure/low- Choromorobelt. These values agree with facies temperature(H P/LT)eventrelated to a subduc- seriesderived from the b values(Do Campo, tion-zonetectonic regime, followed by a lower- 1999a),notwithstanding the fact that the contribu- pressureoverprint possibly at higher temperatures tionof ferrimuscovite substitution to this parameter than the HP/LT event. The HP/LT eventproduced hasbeen corroborated (Fig. 13). This fact reinforces smallp hengiticmica cryst als,which, be ing theusefulness of both methods for semi-quantitativ e enrichedin the <2 mmfraction,would produce barometricpurposes. Nevertheless, numerous Si anchizoneIC values.Therefore, the real maximum valueslie above as well as below the cited range temperatureresponsible for the larger muscovite (Fig.13). Phengite with the highest Si 3.42and crystalswould not be registered by the IC values. 3.45,from slates PU26 and PU36, which have the WhenKisch (1987) established a generalcorrela- largest b parameters(Do Campo, 1999a), implies tionbetween differe ntindicat orsin incipie nt pressuresof <10 kbar. However, for temperatures metamorphism,he emphasized that it was deduced closeto 400ºC, thesevalues point to a high-pressure forintermediate-pre ssureterrane and that therefore metamorphism,which is unsupported by textural or thebehaviour of illite crystallinity in higher- or mineralogicalevidence. When considering these lower-pressureconditions was unknown and subject highresults, it must be remembered that the tofurther verification. accuracyof AEM analysesis less than that of the Thecoexistenc eofIC predominantlycorre- electronmicroprobe. Most of the Si valuesplaced spondingto anchizone and the occurrenc eof underthe mode correspond to muscovites of slate biotitein some slates and metavolcanic rocks PU39,in agreement with the low b parameterfor intercalatedin the Puncoviscana metasediments at thissample. El NinõMuertoHill (Do Campo, 1999b) indicates a Inthe samples from the Choromoro belt, 56% of similarscenario to that described by Dalla Torre et thephengite has Si 5 3.25.In contrast, in the al. (1996)for the Franciscan Complex. Phengitic Lules-Puncoviscanabelt, Si valuesof <3.25 micain the Puncoviscana Formation could have predominate,indica tingpressur esof ~5 kbar. formedduring the maximum pressure event, which Overall,these data suggest that pressures were wasprobably followed by decompression and a rise probablyhigher in the Choromoro belt than in the intemperature. At thatstage, muscovitic white easternLules-Puncoviscan abelt,which is consistent micasand biotite would be expected to grow. Mica withthe contrasting folding styles of these belts, formedat the high-pressure stage would prevail in mentionedabove. the <2 mmfraction,producing an IC indicativeof Inseveralsamples, dioctahedral mica with a wide anchizonein these rocks, as in those of the rangeofphengiticsu bstitutioncoexists,a s FranciscanComplex. Considering that the studied 478 M. Do Campo and F.Nieto slatesprobably represent metastable equilibrium, ACKNOWLEDGMENTS thepersistence of the former phengitic micas after therelatively H T eventwould be controlled by ThehelpofM.M.Abad-Ortega(Centrode kineticfactors. Instrumentacio ńCient õ´fica, University of Granada) Theassociation of anchizonal illite crystallinities with the TEMand AEM,of J.D. Montes Rueda with the ion mill and of A.Molina Illescas with the anda clockwise P-T-t path,which attained high- photographic laboratory was essential for the present pressureconditions followed by greenschist-fac ies, work. C.Laurin is responsible for the language revision hasbeen recorded for the Alpujarride Complex in of the text. Financial support was supplied by Research theBetic Cordilleras (Spain) and for the Verrucano Project BTE2000-0582 from the Ministerio Espanõl de Groupin the Apennines (Italy). Although tempera- Ciencia yTecnolog õáand the Research Group RNM- turesof ~400º C andpressures of 7 –9 kbar were 0179 of the Junta de Andaluc õá. The stay of M.Do establishedfor the Alpujarride Complex (Azan õń & Campo at the University of Granada was supported by Crespo-Blanc,2000) these rocks had anchizone to aUnesco/ICSU/TWASshort-term fellowship awarded epizoneIC values(Azan õń,1994; Orozco et al., in 1997. The authors also thank F.Hongn and M.Pilar 1998). Mata Campo for their constructive reviews of the paper. Thermobarometricdata presented by Giorgetti et al.(1998)for Verrucano metasediments in the Apenninesestablished that they experienced rela- REFERENCES tively HP/LT metamorphismwith peak conditions AzanõńJ.M. (1994) Metamorfismo de alta presioń/baja rangingfrom 350º C at8 kbarto 420º C at10 kbar. temperatura, baja presioń/alta temperatura ytecto´- Thisnotwithstanding, anchizone to epizone IC data nica del Complejo Alpuja´rride (Cordilleras Be´tico – weredetermined for these rocks (Franceschelli et Rifenãs).PhDthesis, Universidad de Granada, al.,1991,1994). These discrepanc iesbetween Spain, 327 pp. thermobarometricconditions and IC datamay be AzanõńJ.M. &Crespo-Blanc A.(2000) Exhumation aconsequenceof the initial H P/LT event of the during acontinental collision inferred from the metamorphicprocess, as postulated in the case of tectonometamorphic evolution of the Alpujarride thePuncoviscana Formation. The IC underestimates Complex in the central Betics (Alboran Domain, SE themetamorphic grade compared to estimates from Spain). Tectonics, 19, 549 –565. thermobarometricdata because smaller phengites Bailey S.W. (1982) Nomenclature for regular inter- formedduring the H P/LT eventare concentrated in stratifications. American Mineralogist , 67, 394 –398. Banfield J.F. &Murakami T.(1998) Atomic-resolution the <2 mm fraction. transmission electron microscope evidence for the Basedon ou rdatafor th ePuncoviscana mechanism by which chlorite weathers to 1:1 semi- Formation(Argentina), those from the Franciscan regular chlorite-vermiculite. American Mineralogist , Complex(California) (Dalla Torre et al., 1996) and 83, 348 –357. theinconsiste nciesin the geologica lliterature Bucher K.&Frey M.(1994) Petrogenesis of regardingthe Alpujarride Complex (Spain) and Metamorphic Rocks .Springer-Verlag, Berlin. Verrucano(Italy), we conclude that high-anchizone / Cliff G.&Lorimer G.W. (1975) The quantitative low-epizoneIC dataof rocks in which mineral analysis of thin specimens. Journal of Microscopy , assemblagesindicate higher-grade conditions may 103, 203 –207. belinked with a metamorphicevolution that Champness P.E.,Cliff G.&Lorimer G.W. (1981) includesan initial relatively H P/LT event. Quantitative analytical electron microscopy. Bulletin Themetamorphic field geothermal gradient for de Mine´ralogie, 104, 236 –240. peakconditions of the P-T-t pathcan be calculated Dalla-Torre M., Livi K., Veblen D.R.& Frey M.(1996) from the P-T conditionsestimated in this study and White K-mica evolution from phengite to muscovite consideringthe average rock density of 2960 kg/ m 3 in shales and shale matrix melange, Diablo Range, California. Contributions to Mineralogy and derivedfrom a gravityfield study of the Cordillera Petrology, 123, 390 –405. Oriental(Omarini et al.,1999).The gradient varies Do Campo M.(1999a) Metamorfismo del basamento en from15º C/ kmduring the maximum pressure event la Cordillera Oriental yborde oriental de la Puna. Pp. to26º C/ kmduring the decompression with a risein 41-51 in: Relatorio XIVCongreso Geolo´gico temperature.The outlined metamorphic evolution of Argentino Geolog ´gadel Noroeste Argentino (G. thePuncoviscana Formation would allow geody- Gonza´lez Bonorino, R.Omarini &J. Viramonte, namicmodels for the area to be improved. editors). Universidad de Salta, Argentina. TEMstudy of pelites of the Puncoviscana Formation 479

Do Campo M.(1999b) Mineralog´ga, geoqu´gmica y (1992) Compositional controls on the cell dimen- geocronolog ´gade la FormaciońPuncoviscana sions of 2M1 muscovite and paragonite. European (Neoproterozoico) entre los 23º30’ y 25º50’ de Journal of Mineralogy , 4, 283 –297. Latitud Sur, Noroeste de Argentina . PhD thesis Guidotti C.V.,Sassi F.P.&Blencoe J.G. (1989) (unpublished), Buenos Aires University, Argentina. Compositional controls on the a and b cell dimen- Do Campo M.&Cagnoni M.(2001) The Rb-Sr isochron sions of 2M1 muscovite. European Journal of method applied to low-grade metamorphic rocks, the Mineralogy, 1, 71 –84. case of the Puncoviscana Formation, NWArgentina. Guidotti C.V., Sassi F.P.,Blencoe J.G. &Selverstone J. III South American Symposium on Isotope Geology, (1994a) The paragonite –muscovite solvus: I.P-T-X Pucoń, Actas, 26 –29. limits derived from the Na –Kcompositions of Do Campo M.,Nieto F.&Omarini R.(1998) natural, quasibinary paragonite –muscovite pairs. Mineralogõáde arcillas ymetamorfismo de la Geochimica et Cosmochimica Acta , 58, 2269 –2275. FormaciońPuncoviscana en localidades de la Guidotti C.V., Yates M.G., Dyar M.D. &Taylor M.E. Cordillera Oriental yPuna, Argentina. X Congreso (1994b) Petrogenetic implications of the Fe +3 con- Latinoamericano de Geolog ´ga, Buenos Aires , Actas, tent of muscovite in pelitic schists. American 2, 217 –223. Mineralogist , 79, 793 –795. Do Campo M., Nieto F., Omarini R.&Ostera H.(1999) Hongn F.D., AcunãP.A., Mon R.&Kirschbaum A.M. Neoproterozoic K-Ar ages for the metamorphism of (2001a) Deformaciońpaleozoica en el a´rea de La thePuncoviscanaFormation,Northwestern Colorada. Sierra de Cobres (NOde la Argentina). X Argentina. IISouth American Symposium on Congreso Argentino de Geolog ´ga Econo´mica, Salta, Isotope Geology, Carlos Paz , Actas, 48 –53. Actas, 1, 57 –63. Dong H.&Peacor D.R.(1996) TEMobservations of Hongn F.D., Tub õáJ.M., Aranguren A., Mon R.& coherent stacking relations in smectite, I/Sand illite Battaglia R.(2001b) Intrusiońdel granito rojo del of shales: Evidence for MacEwan crystallites and batolito de Tastil en areniscas eopaleozoicas en el dominance of 2M 1 polytypism. Clays and Clay angosto de la Quesera, Cordillera Oriental, Salta. Minerals, 44, 257 –275. Revista de la Asociacio ń Geolo´gica Argentina , 56, Franceschelli M.,Leoni L.&Sartori F.(1991) 249 –252. Crystallinity distribution and crystallinity-b 0 rela- Jezek P.(1990) Ana´lisis sedimentolo´gico de la tionships in white K-micas of Verrucano metasedi- FormaciońPuncoviscana entre TucumańySalta. ments (Northern Apennines, Italy). Schweizerische Pp. 9 –36 in: ElCiclo Pampeano en el Noroeste Mineralogische und Petrographische Mitteilungen , Argentino (F.G. Acenõlaza, H.Miller &A.J. Toselli, 71, 161 –167. editors).Universidad Nacional de Tucumań, Franceschelli M., Pandeli E., Puxeddu M., Porcu R.& Tucumań, Argentina. Fadda S.(1994) Illite crystallinity in pelitic and Jiang W.T.&Peacor D.R.(1993) Fomation and marly rocks from the Northern Apennines (southern modification of metaestable intermediate sodium Tuscany and Umbria, Italy). Neues Jahrbuch fu¨r potassium mica, paragonite and muscovite in hydro- Mineralogie Monatshefte , 8, 367 –384. thermally altered metabasites from North Wales. Frey M.(1969) Amixed-layer paragonite/phengite of American Mineralogist , 78, 782 –793. low grade metamorphic origin. Contributions to Jullien M., Baronnet A.&Goffe´B.(1996) Ordering of Mineralogy and Petrology , 24, 63-65. the stacking sequence in cookeite with increasing Frey M.(1978) Progressive low-grade metamorphism of pressure. An HRTEMstudy. American Mineralogist , ablack shale formation, Central Swiss Alps, with 81, 67 –78. special reference to pyrophyllite and margarite Kilmurray J.O., Merodio J.C. &Rapela C.W. (1974) Las bearing assemblages. Journal of Petrology , 19, metamorfitas cordier õ´ticas del a´rea Estacioń 95 –135. Incahuasi-Santa Rosa de Tastil, Provincia de Salta: Giorgetti G.,Goffe´B., Memmi I. &Nieto F.(1998) Rasgos petrolo´gicos ygeoqu õ´micos. Revista de la Metamorphic evolution of Verrucano metasediments AsociaciońGeolo´gica Argentina , 29, 425 –442. in northern Apennines: new petrological constraints. Kisch H.J. (1987) Correlation between indicators of very European Journal of Mineralogy , 10, 1295 –1308. low-grade metamorphism. Pp. 227 –300 in: Low- Giorgetti G., Memmi I.&Peacor D.R.(2000) Retarded temperature Metamorphism (M.Frey, editor). illite crystallinity caused by stress-induced sub-grain Blackie, London. boundaries in illite. Clay Minerals , 35, 693 –708. Kretz R.(1983) Symbols for rock-forming minerals. Guidotti C.V.(1984) Micas in metamorphic rocks. Pp. American Mineralogist , 68, 277 –279. 357 –467 in: Micas (S.W. Bailey, editor). Reviews in Laird J.(1988) Chlorites: metamorphic petrology. Pp. Mineralogy, 13.Mineralogical Society of America, 405 –496 in: Hydrous Phyllosilicates (S.W. Bailey, Washington, D.C. editor). Reviews in Mineralogy, 19.Mineralogical Guidotti C.V., Mazzoli C., Sassi F.P.& Blencoe J.G. Society of America, Washington, D.C. 480 M. Do Campo and F.Nieto

Lee J.H. &Peacor D.R.(1985) Ordered 1:1 interestra- diagenesis through incipient metamorphism in sedi- tification of illite and chlorite: atransmission and ments of the Basque-Cantabrian Basin. Clays and analytical electron microscopy study. Clays and Clay Clay Minerals , 44, 304 –323. Minerals, 33, 463 –467. Omarini R.H.(1983) Caracterizacio ńlitolo´gica, difer- Lee J.H., Peacor D.R., Lewis D.D.& Wintsch R.P. enciaciońygeńesis de la FormaciońPuncoviscana (1984) Chlorite-illite/ muscovite interlayered and entre el Valle de Lerma yla Faja Eruptiva de la interstratified crystals: ATEM/STEMstudy. Puna.PhDthesis (unpublished). Salta University, Contributions to Mineralogy and Petrology , 88, Argentina, 202 pp. 372 –385. Omarini R.&Baldis B.A.J. (1984) Sedimentolog õá y Lee J.H., Peacor D.R., Lewis D.D.& Wintsch R.P. mecanismosdeposicionales de la Formacioń (1986) Evidence for syntectonic crystallization for Puncoviscana (Grupo Lerma, Preca´mbrico- the mudstone to slate transition at Lehigh Gap, Ca´mbrico) del noroeste argentino. IX Congreso Pennsylvania, U.S.A. Journal of Structural Geology , Geolo´gico Argentino, Actas I: 384 –398. 8, 767 –780. Omarini R., Sureda R.J., Go¨tze H.J., Seilacher A.& Livi K.J.T., Veblen D.R., Ferry J.M. &Frey M.(1997) Pflu¨ger F.P.(1999) Puncoviscana folded belt in the Evolution of 2:1 layered silicates in low-grade Northwestern Argentina: testimony of Late metamorphosedLiassic shalesofCentral Proterozoic Rodinia fragmentation and pre- Switzerland. Journal of Metamorphic Geology , 15, Gondwana collisional episodes. International 323 –344. Journal of Earth Sciences , 88, 76 –97. Lo´pez Munguira A.&Nieto F.(2000) Transmission Orozco M., Alonso-Chaves F.M.& Nieto F.(1998) electron microscopy study of very low-grade meta- Development of large north-facing folds and their morphic rocks in Cambrian sandstones and shales, relation to crustal extension in the Alborańdomain Ossa-Morena Zone, southwest Spain. Clays and Clay (Alpujarrasregion, Betic Cordilleras, Spain). Minerals, 48, 213 –223. Tectonophysics , 298, 271 –295. Massone H.J. &Schreyer W.(1987) Phengite geobaro- Peacor D.R.(1992) Diagenesis and low-grade meta- metry based on the limiting assemblage with morphism of shales and slates. Pp. 335 –380 in: K-feldspar, phlogopite and quartz. Contributions to Minerals and Reactions at the Atomic Scale: Mineralogy and Petrology , 96, 212 –224. Transmission Electron Microscopy (P.R. Buseck, Merriman R.J. &Peacor D.R.(1999) Very low-grade editor). Reviews in Mineralogy, 27. Mineralogical metapelites: mineralogy microfabrics and measuring Society of America, Washington, D.C. reaction progress. Pp. 10 –60 in: Low-grade Ramos V.A.& Basei M.(1997) Gondwanan, perigond- Metamorphism (M. Frey &D.Robinson, editors). wanan, and exotic terranes of southern South Blackwell Science, Oxford, UK. America. South-American Symposium on Isotope Merriman R.J. Roberts B.&Peacor, D.R.(1990) A Geology, Actas, 250 –252. transmission electron microscope study of white Rieder M.,Cavazzini G., D’yakonov Y.S., Frank- mica crystallite size distribution in amudstone to Kamenetskii V.A., Gottardi G., Guggenheim S., slate transitional sequence, North Wales, UK. Komal P.V., Mu¨ller G., Neiva A.M.R., Radoslovich Contributions to Mineralogy and Petrology , 106, E.W., Robert J.L., Sassi F.P.,Takeda H., Weiss Z.& 27 –40. Wones D.R.(1998) Nomenclature of the micas. Mon R.&Hongn F.D.(1991) The structure of the Clays and Clay Minerals , 46, 586 –595. Precambrian and Lower Paleozoic basement of the Sassi F.P.,Guidotti C.V., Rieder M.&De Pieri R. Central Andes between22ºand 32ºS Lat. (1994) On the occurrence of metamorphic 2M 1 Geologische Rundschau , 80, 745 –758. phengites: some thoughts on polytypism and crystal- Mon R.&Hongn F.D.(1996) Estructura del basamento lization conditions of 3T phengites. European proterozoico ypaleozoico inferior del norte argen- Journal of Mineralogy , 6, 151 –160. tino. Revista de la AsociaciońGeolo´gica Argentina , Schmidt D.&Livi K.J.T. (1999) HRTEMand SAED 51, 3 –14. investigations of polytypism, stacking disorder, Nieto F.&Sańchez-Navas A.(1994) Acomparative crystal growth, and vacancies in chlorites from XRDand TEMstudy of the physical meaning of the subgreenschistfaciesoutcrops. American white mica ‘crystallinity’index. European Journal Mineralogist, 84, 160 –170. of Mineralogy , 6, 611 –621. Schmidt D., Schmidt S.Th., Mullis J., Mahlmann R.F.& Nieto F., Velilla N., Peacor D.R.& Ortega Huertas M. Frey M.(1997) Very low grade metamorphism of the (1994) Regional retrograde alteration of sub-greens- Taveyanne formation of western Switzerland. chist facies chlorite to smectite. Contributions to Contributions to Mineralogy and Petrology , 129, Mineralogy and Petrology , 115, 243 –252. 385 –403. Nieto F.,Ortega-Huertas M., Peacor D.R.& Arostegui J. Shau Y.H., Feather M.E., Essene E.J. &Peacor D.R. (1996) Evolution of illite/smectite from early (1991) Genesis and solvus relations of submicrosco- TEMstudy of pelites of the Puncoviscana Formation 481

pically intergrown paragonite and phengite in a Analytical electron microscopy and the problem of blueschist from northern California. Contributions to potassium diffusion. Clays and Clay Minerals , 36, Mineralogy and Petrology , 106, 367 –378. 498 –504. Spear F.S.,Hazan R.M. &Rumble III D.(1981) Veblen D.R.(1983a) Exsolution and crystal chemistry Wonesite: anew rock-forming silicate from the ofthesodiummicawonesite. American Post Pond Volcanics, Vermont. American Mineralogist , 68, 554 –565. Mineralogist , 66, 100 –105. Veblen D.R.(1983b) Microstructures and mixed layer- Toselli A.J. (1990) Metamorfismo del Ciclo Pampeano. ing in intergrown wonesite, chlorite, talc, biotite, and Pp. 181 –197 in: ElCiclo Pampeano en el Noroeste kaolinite. American Mineralogist , 68, 566 –580. Argentino (F.G. Acenõlaza, H.Miller &A.J. Toselli, Velde B.(1965) Experimental determination of musco- editors).Universidad Nacional de Tucumań, vite polymorph stabilities. American Mineralogist , Tucumań, Argentina. 50, 436 –449. Toselli A.J.&Rossi de ToselliJ.N.(1982) Warr L.L.& Nieto F.(1998) Crystalline size distribu- Metamorfismo de la FormaciońPuncoviscana en tions in very low grade metamorphic pelites: a las provincias de Salta yTucumań, Argentina. V HRTEMand XRDstudy of clay mineral crystallinity Congreso Latinoamericano de Geolog ´ga, Buenos index standards. Canadian Mineralogist , 36, Aires, Actas, 2, 37 –52. 1453 –1474. Turner, J.C.M. (1960) Estratigraf õáde la Sierra de Santa Zhao G., Peacor D.R.and McDowell D.S.(1999) Victoria yadyacencias. Bolet´gnAcademia Nacional Retrograde diagenesis of clay minerals in the de Ciencias de Cordoba , 41, 163 –196. Precambrian Freda sandstone, Wisconsin. Clays Van der Pluijim B.A., Lee J.H. &Peacor D.R.(1988) and Clay Minerals , 47, 119 –130.