JOURNAL OF PETROLOGY VOLUME 38 NUMBER 1
Low-Pressure Metamorphism in the
Sierra Albarrana Area (Variscan Belt,
Iberian Massif )
A. AZOR1∗ AND M. BALLÈVRE2
1
DEPARTAMENTO DE GEODINÁMICA, UNIVERSIDAD DE GRANADA, FACULTAD DE CIENCIAS,
CAMPUS FUENTENUEVA S/N, E-18002 GRANADA, SPAIN
2
LABORATOIRE DE PÉTROLOGIE–GÉOCHIMIE, GÉOSCIENCES RENNES (UPR–CNRS 4661), UNIVERSITÉ RENNES I,
F-35042 RENNES CEDEX, FRANCE
RECEIVED FEBRUARY 2, 1996 REVISED TYPESCRIPT ACCEPTED AUGUST 2, 1996
A low-pressure metamorphic zonation ranging from biotite to mig-
matite zones occurs in the Sierra Albarrana area (Variscan Belt of
southwestern Iberian Peninsula) in uppermost Precambrian to Lower
Palaeozoic metasedimentary rocks. The principal deformation in
this area is related to a major ductile shear zone whose central part
is localized immediately to the southwest of the Sierra Albarrana
Quartzites. The metamorphism is synchronous with respect to this
deformation. The metamorphic zones are symmetrically distributed
with respect to the Sierra Albarrana Quartzites. Pressure–temperature
(P–T) conditions are ~3·5–4 kbar and range from ~400°C
(biotite zone) to 500°C (staurolite–garnet zone) up to 650–700°C
(migmatite zone). We have not detected pressure variations along
the different metamorphic zones. Relic kyanite is observed in the
form of inclusions in andalusite within veins in the lower-grade part
of the staurolite–andalusite zone. The low-pressure metamorphism of
the Sierra Albarrana area arises from a two-stage history including
moderate crustal thickening followed by subsequent localization of
deformation in a transcurrent shear zone during peak P–T conditions.
Channelized fluid flow within the major ductile shear zone may
have contributed to the heat budget of the low-pressure metamorphism.
KEY WORDS: fluid flow; Iberian Massif; low-pressure metamorphism;
shear zone; Sierra Albarrana area
INTRODUCTION
Regions of low-pressure and medium- to high-
temperature metamorphism are widespread in many
∗Corresponding author. Telephone: (34) 58 24 29 00. Fax: (34) 58 24
33 52. e-mail: azor@platon.ugr.es
PAGES 35–64 1997
orogenic belts, including the Variscan Belt of Southwest
Europe. De Yoreo et al. (1991) presented a review of some
possible tectonic settings of low-pressure metamorphism.
Those workers considered low-pressure metamorphism
to be that in which peak temperature is above 500°C
and pressure does not exceed that of the aluminium
silicate (Als) triple point. The following four tectonic
situations were envisaged for the low-pressure meta-
morphism (De Yoreo et al., 1991): (1) magmatic arcs;
(2) regions of crustal extension; (3) continent–continent
collision zones where thermal relaxation occurs after
crustal thickening; (4) regions with significant fluxes of
aqueous fluids, especially those located above subduction
zones.
This paper is concerned with the description of a high-
temperature–low-pressure facies series terrane (the Sierra
Albarrana area, Variscan Belt, southwestern Iberian Mas-
sif ) (Figs 1 and 2). In this area, a metamorphic zonation
from the biotite zone up to the migmatite zone developed
at low pressure (Garrote, 1976; González del Tanago &
Peinado, 1990). As will be shown here, the metamorphic
zonation is not related to post-thickening thermal re-
laxation but rather exhibits a roughly symmetric pattern
around a major ductile shear zone. In addition, syn-
metamorphic magmatism is not recognized. For these
reasons, the low-pressure metamorphism of the Sierra Al-
barrana area may provide a good example of moderate
crustal thickening followed by significant heat advection
by aqueous- or silicate-rich fluids along a major ductile
shear zone.
 Oxford University Press 1997
 JOURNAL OF PETROLOGY
Fig. 1. (a) Geologic map of the southwestern Iberian Massif showing the major zones. (b) Geologic map of the Ossa–Morena–Central Iberian
Zones boundary in which the Sierra Albarrana area is located.
GEOLOGICAL SETTING OF THE
SIERRA ALBARRANA AREA
The Sierra Albarrana area is located in the southwestern
part of the Iberian Massif, immediately to the southwest of
a crustal-scale shear zone known as the Badajoz–Córdoba
Shear Zone (Burg et al., 1981) (Fig. 1a and 1b). This
ductile shear zone is the boundary between two major
zones of the Iberian Massif: the Central Iberian Zone to
the northeast and the Ossa–Morena Zone to the south-
west. The Badajoz–Córdoba Shear Zone represents one
of the sutures of the Variscan Belt (e.g. Burg et al., 1981;
Matte, 1991; Azor, 1994; Azor et al., 1994). The Sierra
Albarrana area is separated from the Badajoz–Córdoba
Shear Zone by a subvertical semibrittle left-lateral fault
known as the Azuaga Fault (Fig. 1b).
Pre-Carboniferous rocks of the Sierra
Albarrana area
Lithostratigraphic sequence
The area studied (Figs 1b and 2) is made up mainly of
VOLUME 38 NUMBER 1 JANUARY 1997
metasedimentary rocks. The lithostratigraphic sequence
can be divided into three groups of rocks separated by
tectonic contacts (Azor, 1994; Azor et al., 1994).
The first group consists of Upper Proterozoic to Lower
Cambrian rocks, in which three formations can be dis-
tinguished. These are, from bottom to top (i.e. from
southwest to northeast), as follows:
(1) A volcanosedimentary succession of Upper
Proterozoic age (Vendian), known as the Malcocinado
Formation (Fricke, 1941; Delgado Quesada, 1971).
(2) A Lower Cambrian succession of limestones and
dolostones with slate intercalations, known as the
Pedroche Formation (Liñán, 1978; Azor, 1994).
(3) A Lower Palaeozoic (probably Lower to Middle
Cambrian) monotonous succession of slates and
metagreywackes with some quartzite intercalations,
known as the Villares Formation (Liñán, 1978; Azor,
1994).
The second group of rocks crops out in a band
located to the northeast of the first group and separated
from it by a late brittle fault known as the Onza Fault
36
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 2. Geologic map and cross-section of the Sierra Albarrana area. The various lithologic units are described in the text. 1, Malcocinado
Formation; 2, Pedroche Formation; 3, Villares Formation; 4, Albariza Micaschists; 5, migmatitic gneisses with minor amphibolites; 6, Sierra
Albarrana Quartzites; 7, migmatitic gneisses, micaschists and metagreywackes; 8, Lower Carboniferous undeformed sediments of the Valdeinfierno
basin; 9, Badajoz–Córdoba Shear Zone; 10, Los Ojuelos Gabbro; 11, La Cardenchosa granite; AB, geologic cross-section; SP, principal foliation;
SC, crenulation foliation. The shear zone mainly corresponds to the migmatitic gneisses located to the southwest of the Sierra Albarrana
Quartzites. Its location has been indicated on the cross-section (see text for further explanation).

37
 JOURNAL OF PETROLOGY VOLUME 38
(Figs 1b and 2) (Azor, 1994). This second group,
known as the Albariza Micaschists, is made up of a
succession of schists and metasandstones with am-
phibolite intercalations, the age of which is probably
Lower Palaeozoic (Azor, 1994).
The third group of rocks crops out to the northeast of
the second one (Fig. 2). The contact between these two
groups of rocks coincides with a narrow band of intensely
deformed rocks, interpreted as a ductile shear zone
(Azor, 1994; Azor et al., 1994). Three formations can be
distinguished in this group of rocks. They are, from
bottom to top, as follows:
(1) Migmatitic gneisses with minor amphibolite,
metagreywacke and quartzite intercalations.
(2) Feldspathic quartzites with intercalated pa-
ragneisses, schists, metagreywackes and amphibolites
(Sierra Albarrana Quartzites, Delgado Quesada, 1971).
These quartzites are of Lower Palaeozoic age according
to palaeontological evidence (Marcos et al., 1991). The
outcrop area of these quartzites is a small ridge known
as the Sierra Albarrana. Way-up criteria at the southwest
and northeast contacts of this succession indicate a stra-
tigraphic top to the northeast.
(3) Migmatitic gneisses and schists with minor quartzites
and metagreywackes.
Structure
The pre-Carboniferous rocks of the Sierra Albarrana
area are affected by a principal phase of penetrative
deformation responsible for the development of tight
kilometre-scale upright folds and a ductile shear zone
located immediately to the southwest of the Sierra Al-
barrana Quartzites (Fig. 2). Southwest of the shear zone,
the fabric developed is generally planar and constitutes
the axial planar foliation of the folds recognized (Fig.
2). The ductile shear zone mainly coincides with the
migmatitic gneisses located to the southwest of the Sierra
Albarrana Quartzites. Within this zone, the fabric is
strongly planar and linear. Northeast of the shear zone,
the intensity of the planar–linear fabric progressively
decreases and kilometre-scale folds related to this de-
formation are recognized in the Sierra Albarrana Quartz-
ites (Fig. 2).
In the whole area at issue here, the principal foliation
is subvertical or steeply dipping in its present orientation
and has a NW–SE strike. The stretching lineation is
subhorizontal or gently plunging to the southeast or
northwest, except in the southeastern end of the area
studied, where plunges of 40–70° to the southeast are
found (Fig. 3).
Within the ductile shear zone and immediately to the
southwest and northeast, different types of shear criteria
can be observed in sections parallel to the stretching
38
NUMBER 1 JANUARY 1997
lineation and perpendicular to the foliation, which
indicate a non-coaxial component for the principal
deformation. The shear criteria recognized [see Simpson
& Schmid (1983) and Hanmer & Passchier (1991) for a
review] are S–C structures, asymmetric tails in feldspathic
veins and rotation of the foliation in synkinematic garnet,
andalusite and staurolite porphyroblasts. These criteria
consistently indicate a dextral sense of movement (Azor,
1994; Azor et al., 1994).
The age of the principal phase of deformation and
metamorphism must be post-Lower Palaeozoic, as it
affects rocks that are palaeontologically dated as Lower
Palaeozoic (Marcos et al., 1991). 40Ar/39Ar radiometric
dating (Dallmeyer & Quesada, 1992) indicates ages of
390 Ma for amphiboles and 360–350 Ma for muscovites.
Dallmeyer & Quesada (1992) have suggested that these
ages record a thermal rejuvenation during the Variscan
orogeny of minerals that grew during a Late Precambrian
(i.e. Cadomian) event. However, because the Palaeozoic
age of some rocks of this area has been documented,
and because only one tectonothermal event has been
identified (see below), we interpret these ages as cooling
ages after an Upper Devonian–Lower Carboniferous
metamorphism (i.e. Variscan in age).
A second phase of deformation can be recognized in
the Sierra Albarrana area. It consists of folds that affect
the above structures and locally generate a subvertical
crenulation cleavage (Fig. 2). This phase is almost coaxial
with the principal deformation phase and is responsible
for the present steep dips of the principal foliation and
for the folds affecting the southern end of the Sierra
Albarrana Quartzites (Fig. 2).
The latest Variscan deformation was an episode of
brittle faulting that took place during the Carboniferous
and generated the low-angle normal fault located to the
northwest of Sierra Albarrana and several steeply dipping
left-lateral faults (Fig. 2). The Onza Fault separating the
Albariza Micaschists from the Villares Formation belongs
to the latter set and, in addition to the left-lateral move-
ment, displays a component of downthrowing of the
southwestern block.
The Valdeinfierno basin
Undeformed and unmetamorphosed sediments of Lower
Carboniferous age [Upper Tournaisian to Lower Visean
according to Wagner (1978), Garrote & Broutin (1979)
and Roldán (1983)] rest unconformably on low-grade
schists and slates immediately to the northwest of the
Sierra Albarrana, in the Valdeinfierno basin (Figs 1b and
2). This basin is bounded by a synsedimentary normal
fault, and the pebbles from the conglomeratic layers
within the basin record the progressive exhumation of
the metamorphic rocks of the Sierra Albarrana area
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 3. Structural map of the Sierra Albarrana area showing strike and plunge of the stretching lineation (arrows). The Sierra Albarrana
Quartzites and the Valdeinfierno basin are depicted by the same pattern as in Fig. 2. The dotted line represents the external limit of the
metamorphic contact aureole associated with La Cardenchosa granite.

(Roldán, 1983; Gabaldón et al., 1983; Gabaldón & Que- IDENTIFICATION OF

sada, 1986; Azor, 1994).
La Cardenchosa granite
The area studied is bounded to the east by La Car-
denchosa granite (Figs 1b and 2). The intrusion postdates
the regional ductile deformation and metamorphism of
the Sierra Albarrana area and is associated with a contact
aureole (Garrote, 1976; Garrote & Sánchez Carretero,
1979). The age of La Cardenchosa granite with respect
to the Valdeinfierno basin is not known yet, but some
workers have suggested that the granite is younger (i.e.
Namurian–Westphalian) than the Lower Carboniferous
sediments (Delgado Quesada et al., 1985).
In summary, the tectonothermal evolution of the area
studied is two-fold. An intense ductile deformation as-
sociated with low-pressure metamorphism occurred dur-
ing a major dextral shearing. The exhumation of the
metamorphic rocks took place during the Lower Car-
boniferous, thus being synchronous with sedimentation
within the Valdeinfierno basin.
METAMORPHIC ZONES
In the area at issue here, metapelitic rocks constitute the
major part of the stratigraphic sequence (see Geological
Setting). We have been concerned with assemblages from
metapelites alone and have examined about 300 thin
sections. The distribution of low-variance assemblages as
well as some higher-variance ones is shown in Fig. 4.
The distribution of aluminium silicate polymorphs is
shown in Fig. 5. The relationships between the different
metamorphic phases and the deformation history are
shown in Fig. 6. To constrain the P–T conditions in the
different metamorphic zones, low-variance assemblages
(i.e. those with three or more AFM phases) were studied
with the electron microprobe. Table 1 shows the location
of the 15 samples studied and their mineralogy.
Biotite zone
Pelitic rocks from the biotite zone are slates or fine-grained
schists. These rocks generally show a slaty cleavage that

39
 JOURNAL OF PETROLOGY VOLUME 38 NUMBER 1 JANUARY 1997

Table 1: Mineral parageneses and location of the samples analysed; the location of the samples on map is
shown in Figs 4 and 5; the metamorphic zone of each sample is also indicated

Sample UTM coordinates Kfs Ms Grt St Chl And Sil P Sil F Ilm Met. zone

AA-166 30S TH 885113 St–Grt

AA-175 30S TH 795184 St–Grt
AA-53 30S TH 901202 St–Grt
AA-85 30S TH 818184 St–Grt
AA-137 30S TH 905169 St–Grt
AA-171 30S TH 859140 St–Grt
AA-223 30S TH 925136 St–Grt
AA-233 30S TH 889127 St–Grt
AA-80 30S TH 820195 Sil
AA-138 30S TH 896154 Sil
AA-139 30S TH 899156 Sil
AA-141 30S TH 894170 Sil
AA-18 30S TH 891152 Mig
AA-20 30S TH 879158 Mig
AA-104 30S TH 864190 Mig

Sil P, prismatic sillimanite; Sil F, fibrolitic sillimanite; Mig, migmatite zone. Other mineral abbreviations after Kretz (1983).

constitutes the principal foliation. Locally, a crenulation
cleavage overprints the principal foliation. The mineral
parageneses of these rocks are, in addition to quartz and
muscovite, Bt + Chl or Bt. Accessory minerals are
ore minerals and tourmaline. Chloritoid was found by
Garrote (1976) and Contreras et al. (1984), but its exact
location is not known.
Staurolite–garnet zone
Pelitic rocks from the staurolite–garnet zone are medium-
to fine-grained schists in which garnet, biotite and at
times staurolite can be recognized in hand specimen. In
the field, the principal foliation is a slaty cleavage or
schistosity. The principal foliation is occasionally affected
by a later crenulation cleavage and is generally the first
penetrative deformation as the sedimentary layering can
be recognized. The lowest-variance assemblages from
this zone are Ms + Grt + St + Bt and Ms + Grt +
Bt + Chl, but higher-variance assemblages such as Ms
+ Grt + Bt are also observed. Common accessory
minerals are ilmenite, zircon and tourmaline. Chloritoid
has also been reported in this zone (Garrote, 1976;
Contreras et al., 1984).
Garnet (0·1–2 mm) normally forms euhedral or sub-
hedral inclusion-free crystals, although sometimes it has
inclusions of quartz and minor ilmenite. Most garnet
grains have pressure shadows. In some samples (e.g.
sample AA-222), quartz inclusions define an internal
foliation, which is generally linear and discontinuous but
at high angles with respect to the external foliation (Fig.
7). Because the sense (clockwise) and the amount (about
20–30°) of rotation of the internal foliation is constant
at the scale of the thin section, garnet growth appears to
be synkinematic, although affected by late rotation. In
the most altered samples, garnet is transformed along
margins and fractures to chlorite, biotite and Fe-oxides.
Staurolite crystals up to 1 cm in size contain quartz
and ilmenite inclusions, defining an internal foliation. In
most cases (e.g. sample AA-175), the internal foliation is
rotated with respect to the external foliation, although
both foliations are in continuity at the border of the
porphyroblasts (Fig. 8a). These relationships indicate that
staurolite porphyroblasts are rotated with respect to the
matrix.
Two types of biotite can be recognized: (1) elongated
crystals which, together with muscovite, define the fo-
liation, and (2) up to 1 mm porphyroblasts with the
cleavage parallel or oblique to the principal foliation (e.g.
samples AA-166 and AA-222). In the latter case, the
shape of the porphyroblasts and the angle between the
foliation and the {001} cleavage indicate a dextral shear
sense (Fig. 7). The biotite porphyroblasts are altered to
chlorite. If a late deformation is present, biotite por-
phyroblasts are kinked when their cleavage is oblique to
the principal foliation.
Some chlorite grains are parallel to the foliation, which
can be taken to suggest that chlorite was in equilibrium
with garnet, muscovite and biotite. Chlorite is also found

40
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 4. Distribution of AFM-mineral assemblages in the Sierra Albarrana area. Andalusite-in, sillimanite-in and migmatite-in isograds have
been marked (dashed lines). Other symbols are as in Fig. 3. All the samples analysed are numbered. These numbers correspond to the numbering
in the text, tables and other figures. Other samples cited are also located.

within pressure shadows around garnet (Fig. 7), replacing
garnet and biotite, or as porphyroblasts overgrowing the
foliation (Fig. 8b).
Staurolite–andalusite zone
This zone is defined by the coexistence of staurolite and
andalusite (Figs 4–6). Metapelitic rocks from this zone
are fine- to medium-grained schists containing por-
phyroblasts of staurolite (up to 2 cm) and andalusite
(up to 5 cm). Crystals of biotite, muscovite and garnet
generally do not exceed 2 mm. The principal foliation
in these schists is mainly a schistosity, although in some
cases a crenulation cleavage is observed. Locally, a mil-
limetre-spaced crenulation cleavage overprints the prin-
cipal foliation. The most abundant assemblage is Ms +
Grt + St + Bt + And. Some pelitic rocks, however,
contain higher-variance assemblages such as Ms + Grt
+ Bt, Ms + Grt + St + Bt or Ms + Grt + Bt +
And. Accessory minerals are ilmenite, zircon, tourmaline
and apatite.
Staurolite forms euhedral to subhedral porphyroblasts
with an internal foliation marked by quartz inclusions.
Staurolite also includes euhedral ilmenite and garnet. In
some cases (e.g. sample AA-171) the internal foliation is
continuous with the external foliation (i.e. the principal
foliation), and staurolite porphyroblasts are rotated with
respect to the matrix. In other samples (e.g. samples AA-
85 and AA-223) the internal foliation is a crenulated
cleavage, recording an earlier stage of development of
the foliation now observed in the matrix. In these cases,
transposed microfolds can be seen in the matrix, showing
that the principal foliation corresponds to a crenulation
cleavage. Andalusite appears as subhedral porphyroblasts
with numerous quartz, muscovite, biotite and ilmenite
inclusions sometimes defining an internal foliation. As in
the staurolite porphyroblasts, the internal foliation is
sometimes a crenulated cleavage (e.g. sample AA-233),
which we interpret as having recorded the earliest stages
of foliation development within the matrix. Euhedral
garnet is occasionally included in andalusite, but staurolite
inclusions are not observed.

41
 JOURNAL OF PETROLOGY VOLUME 38 NUMBER 1 JANUARY 1997

Fig. 5. Distribution of aluminium silicate polymorphs in the Sierra Albarrana area. Andalusite-in, andalusite-out and sillimanite-in regional
isograds have been depicted (dashed lines). Distribution of aluminium silicate polymorphs in the contact aureole of La Cardenchosa Granite is
also shown. Sample location is indicated with numbers as in Fig. 4. Other symbols are as in Fig. 3.

The fact that the principal foliation is either a schistosity
or a crenulation cleavage can be interpreted as evidence
for polymetamorphism with two different generations of
staurolite and andalusite porphyroblasts. Nevertheless,
this is not the case, as this fact can be related to the
existence of a non-coaxial component in the principal
deformation as indicated by the development of a planar–
linear fabric and shear criteria. In this regard, the su-
perposition of different fabrics in a shear zone is fairly
common owing to the rotation affecting all surfaces and
lines included in it (Ramsay & Huber, 1987). If the angle
with the boundaries of the shear zone is adequate,
earliest developed fabrics can be folded during progressive
deformation and the principal foliation finally observed
would be a crenulation cleavage. In other cases, the
fabric would not rotate, owing to its initial orientation
with respect to the boundaries of the shear zone, and
the final foliation observed would be a schistosity.
Garnet and biotite present the same textural char-
acteristics as in the staurolite–garnet zone. Muscovite
crystals are parallel to the foliation. In some samples (e.g.
sample AA-223), large muscovite grains develop around
the andalusite porphyroblasts.
Chlorite develops in the more altered samples at the
expense of biotite and garnet. It is also present as por-
phyroblasts up to 1 mm in size that overgrow the principal
foliation (e.g. samples AA-85 and AA-223). The growth
of the chlorite porphyroblasts postdates the crystallization
of the other phases as well as the principal deformation.
Textural equilibrium between staurolite, andalusite,
biotite, garnet, plagioclase and muscovite is observed. All
these phases are synkinematic with respect to the principal
foliation as deduced from pressure shadows and internal
foliation–external foliation relationships in staurolite, an-
dalusite and garnet porphyroblasts.
Sillimanite zone
The sillimanite zone is marked by the disappearance of
staurolite (staurolite-out isograd), which seems to coincide
on a regional scale with the incoming of sillimanite
(sillimanite-in isograd) in most metapelites (Figs 4–6).

42
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 6. Sequence of synkinematic phases within metapelites, late to postkinematic phases in metapelites, and mineral assemblages observed
within veins in the different metamorphic zones.

Andalusite porphyroblasts are still present in a narrow
band adjacent to the staurolite–andalusite zone (Fig. 5).
The modal amount of garnet decreases upgrade from
the sillimanite-in isograd.
Metapelites from the sillimanite zone are fine- to
medium-grained schists in which fibrolitic sillimanite and
biotite are recognized in hand specimen. The principal
foliation is a schistosity occasionally affected by a cren-
ulation cleavage. The most abundant assemblages in this
zone are Ms + Grt + Bt + And + Sil and Ms + Bt
+ Sil. The three-phase AFM assemblage Ms + Grt +
Bt + Sil is occasionally present. Accessory minerals are
tourmaline, apatite, zircon and ilmenite. Relic kyanite is
observed as inclusions within biotite crystals in a rare
cordierite-bearing layer (González del Tanago & Peinado,
1990).
Andalusite porphyroblasts have characteristics similar
to those of the staurolite–andalusite zone, and sometimes
contain garnet pseudomorphs (Fig. 9). Prismatic sil-
limanite progressively develops either within fractures in
the andalusite porphyroblasts (sample AA-141, Fig. 9) or
as aggregates between andalusite grains (sample AA-139).
Fibrolitic sillimanite is sometimes present in small clusters
(sample AA-80).
Garnet is less common than in the staurolite–andalusite
zone and appears as euhedral to subhedral inclusion-free
crystals of several microns to 1 mm in size. Reddish
brown biotite and muscovite lamellae define the principal
foliation. Some muscovite grains develop around an-
dalusite (sample AA-141). Synkinematic phases with re-
spect to the principal foliation are sillimanite, biotite,
garnet, plagioclase and muscovite.

43
 JOURNAL OF PETROLOGY VOLUME 38
Fig. 7. Subhedral garnet with internal foliation marked by quartz
inclusions (sample AA-222, staurolite–garnet zone). The internal fo-
liation is rotated with respect to the principal foliation indicating a
dextral sense of shearing (arrows); the angle between {001} cleavage
in biotite porphyroblasts and the principal foliation also indicates a
dextral sense of shearing. Garnet diameter is 1·28 mm.
Migmatite zone
Metapelites from the migmatite zone show a gneissic
layering defined by millimetre-scale alternating mi-
caceous and quartzofeldspathic domains. The quartzo-
feldspathic domains have a coarse-grained granitic min-
eralogy and form essentially continuous layers or small
boudinaged pods. The micaceous domains are made
up by biotite, fibrolitic sillimanite and at times minor
muscovite. Fibrolitic sillimanite sometimes forms elong-
ated whitish nodules up to 1·5 cm in size parallel to the
stretching lineation. We interpret these textural char-
acteristics as being due to partial melting. In this zone,
most rocks are of quartzitic or quartzofeldspathic com-
position, thus making it difficult to map the boundaries
of the zone, as partial melting affects only the metapelitic
lithologies. To the southwest of the Sierra Albarrana
Quartzites, the migmatite zone extends to the contact
with the Albariza Micaschists. Migmatites are sometimes
present within the Sierra Albarrana Quartzites, where
they represent interbedded metapelites. Finally, mig-
matites are largely developed to the northeast of the
Sierra Albarrana Quartzites, the zone boundary being
located within the formation of migmatitic gneisses and
schists.
In most migmatitic metapelites (e.g. samples AA-18
and AA-20), complex relationships between coexisting
phases can be observed. Large grains of K-feldspar
contain euhedral inclusions of biotite and, more rarely,
fibrolite. Matrix biotite is strongly corroded by fibrolitic
sillimanite (Fig. 10a). Myrmekitic intergrowths are fre-
quently present at the contact between plagioclase and
K-feldspar. Muscovite grains constitute either small cor-
roded grains or large (up to 1 cm in size) porphyroblasts
44
NUMBER 1 JANUARY 1997
which grow over the foliation, although sometimes they
appear deformed. Fibrolitic sillimanite is present within
all phases (Fig. 10a). Accessory minerals are tourmaline,
apatite and zircon. The lack of ilmenite is also char-
acteristic of this zone (Fig. 6).
Sample AA-104 is slightly different from the other
samples of this zone, because it shows a textural equi-
librium between biotite, K-feldspar, prismatic sillimanite,
muscovite, quartz and tourmaline (Fig. 10b): the grains
of the different phases have approximately the same size
and are in mutual contact. Some sillimanite grains have
been replaced by fine-grained muscovite aggregates.
DISTRIBUTION OF METAMORPHIC
ZONES
A metamorphic zonation (Figs 4–6) ranging from the
biotite zone to the migmatite zone developed during the
principal deformation in the Sierra Albarrana area. This
zonation has been noted in previous studies (Laurent,
1974; Garrote, 1976; González del Tanago & Peinado,
1990; González del Tanago, 1993). According to these
workers, the metamorphic zones are roughly symmetrical
with respect to the Sierra Albarrana Quartzites and the
metamorphic grade decreases towards the southwest as
well as towards the northeast up to the chlorite zone.
This simple pattern requires modification in the light of
our data. Because late granite intrusion and brittle faulting
have modified the distribution of metamorphic zones, it
is only in the central part of the area studied where the
initial distribution is preserved. We shall now turn to
discussion of the relevance of the late modifications before
considering the relationships between the distribution of
metamorphic zones and the ductile deformations.
Three kinds of late modifications are recognized:
(1) Structural investigations to the southwest of the
Sierra Albarrana reveal the existence of a left-lateral
brittle fault with downthrowing of the southwestern block
(the Onza Fault, Azor, 1994). This kinematics is consistent
with the distribution of the metamorphic zones: the
biotite zone is located to the south of the Onza Fault, in
the downthrown block, whereas higher-grade zones (from
staurolite–garnet up to the migmatite zones) can be
observed to the north of the fault, i.e. in the upthrown
block (Fig. 4). The southeasternmost part of the meta-
morphic domain is cut across by the Onza Fault, as
shown by the occurrence of garnet-bearing micaschists
south of the fault. These latter rocks can be interpreted
as high-variance assemblages belonging to the staurolite–
garnet zone, or as assemblages from a garnet zone,
transitional between the biotite and staurolite–garnet
zones. The garnet-bearing micaschists have not been
identified in the southwestern part of the area studied
probably because they are cut across by the Onza Fault.
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 8. (a) Photomicrograph of euhedral staurolite and garnets from sample AA-175 (staurolite–garnet zone). Internal foliation in staurolite is
marked by quartz inclusions and is continuous with the external foliation, which is defined by biotite and muscovite. Rotation of the internal
foliation in staurolite indicates a sinistral sense of shearing in the photograph, which, in map view, corresponds to a dextral sense of shearing.
Plane-polarized light. Diameter of the garnet located in the lower left corner of the photograph is 1·24 mm. (b) Photomicrograph of chlorite
porphyroblasts from sample AA-170 (staurolite–garnet zone). Chlorite porphyroblasts postdate the principal foliation, which is defined by biotite
and muscovite. Staurolite and garnet are also present in this sample. Plane-polarized light. The chlorite porphyroblast located to the left in the
photograph is 0·36 mm wide.

45
 JOURNAL OF PETROLOGY
Fig. 9. Garnet pseudomorph included in andalusite porphyroblast from sample AA-141 (sillimanite zone). Prismatic sillimanite appears along
fractures in the andalusite porphyroblast. Garnet pseudomorph is made up of quartz, muscovite and biotite. Cross-polarized light. Garnet
pseudomorph diameter is 1·46 mm.
(2) The metamorphic zonation of the Sierra Albarrana
area is cut across towards the northwest by a low-angle
brittle normal fault (Figs 1b and 4), which is responsible
for the location of the Lower Carboniferous Valdeinfierno
basin (Azor, 1994). The rocks in the hangingwall of the
fault belong to the Villares Formation, corresponding to
metamorphic conditions of the biotite zone.
(3) The contact aureole associated with La Car-
denchosa granite cuts across the regional metamorphic
zonation (Garrote, 1976) (Fig. 4). The development of
hornfelses largely overprints the regional metamorphic
zonation. Assemblages belonging to the staurolite–garnet
zone and lower-grade ones, if ever present, have been
replaced by contact metamorphic assemblages. Con-
sequently, no isograd associated with the regional meta-
morphism can be mapped near La Cardenchosa granite,
and the northern boundary of the staurolite–andalusite
zone probably corresponds to the external limit of the
contact aureole (Fig. 4).
The distribution of metamorphic zones is well pre-
served in the central part of the area studied, i.e. on both
sides of the Sierra Albarrana. At the sample scale, mineral
growth is synchronous with the principal foliation. This
means that the Sierra Albarrana area does not constitute
a thermal dome postdating the development of the prin-
cipal foliation, as proposed by Quesada & Munhá (1990).
At a regional scale, a close correlation can be observed
VOLUME 38 NUMBER 1 JANUARY 1997
between the direction and plunge of the stretching lin-
eation (Fig. 3) and the shape of the isograds (Figs 4 and
5). When the stretching lineation is subhorizontal, the
isograds run parallel to the lithologic boundaries and to
the shear zone (e.g. the andalusite-in and sillimanite-in
isograds to the southwest of the Sierra Albarrana). In
addition, the sillimanite-in and migmatite isograds close
where the plunge of the stretching lineation sharply
increases.
DISTRIBUTION AND MINERAL
ASSEMBLAGES OF VEINS AND
PEGMATITES
Veins are especially abundant in the Sierra Albarrana
area. The mineral content and the spatial distribution of
these veins are summarized in Figs 6 and 11, respectively.
These data are consistent with previous observations
(Garrote et al., 1980; Ortega Huertas et al., 1982).
Only quartz veins are present in the biotite and stau-
rolite–garnet zones. The amount and size of these veins
increase in the staurolite–andalusite zone, where they may
also contain large crystals of muscovite and andalusite (up
to 10 cm). In one locality (Era de la Charneca, Fig. 11),
a large number of mineral assemblages can be observed.
46
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 10. (a) Photomicrograph of fibrolitic sillimanite related to biotite from sample AA-18 (migmatite zone). (Note sillimanite needles surrounding
biotite and included in K-feldspar.) Plane-polarized light. Width of the field is 1·4 mm. (b) Photomicrograph of Ms–Kfs–Bt–Sil assemblage from
sample AA-104 (migmatite zone). Apparent textural equilibrium between the different phases is shown. Plane-polarized light. Width of the field
is 1·4 mm.

Most veins contain various combinations of quartz, mus- andalusite–muscovite, as well as a minor amount of either
covite and andalusite, especially andalusite–quartz and garnet or staurolite. Interestingly, a few kyanite crystals

47
 JOURNAL OF PETROLOGY
Fig. 11. Distribution of coarse-grained veins and pegmatites. It should be noted that they are preferentially distributed in the higher-grade part
of the area, and that their mineralogy is consistent with the metamorphic grade of the enclosing metapelites. Other symbols are as in Fig. 2.
are observed as inclusions within andalusite (sample AA-
415, Fig. 12). These observations are consistent with
previous reports (Abad Ortega, 1993; González del Tan-
ago, 1993) and show that kyanite is a relic phase.
Veins containing quartz, muscovite and andalusite
partly replaced by sillimanite are present in the lower-
grade part of the sillimanite zone (e.g. sample AA-
276). In the higher-grade part of the sillimanite zone,
aluminium silicate is not present in the veins, which
essentially consist of coarse-grained quartz, plagioclase,
muscovite and tourmaline crystals. Biotite is a minor
phase in some veins.
In the migmatite zone, pegmatite veins are extremely
abundant and are essentially made up of quartz, albite,
orthoclase and tourmaline. Graphic intergrowths be-
tween K-feldspar or plagioclase and quartz are frequent.
In addition, some veins contain giant biotite and beryl
crystals (up to 1 m). The mineralogy of the pegmatites
has been dealt with in a number of recent studies (e.g.
González del Tanago, 1991; Abad Ortega, 1993; Abad
Ortega et al., 1993; Abad Ortega & Nieto, 1995a, 1995b).
VOLUME 38 NUMBER 1 JANUARY 1997
The volume and the orientation of the veins are
dependent on both the metamorphic grade and the
nature of the host rocks. Veins are larger in the higher-
grade zones (up to 500 m long and 70–100 m thick)
compared with the lower-grade ones (up to 10–20 m
long and 1 m thick). On the overall outcrop scale, they
parallel the strike of the foliation when the enclosing
rocks are slates or schists, even if their margins locally
cut across the foliation. Veins elongated perpendicular
or subperpendicular to the strike of the foliation are only
observed within the Sierra Albarrana Quartzites or within
the quartzites interlayered within the gneissic rocks loc-
ated to the southwest of the Sierra Albarrana (e.g. Cerro
de la Sal, Fig. 11). Nevertheless, many of the quartz and
pegmatitic veins are deformed (i.e. foliated, folded and/
or boudinaged) whereas others are undeformed. This
indicates that the veins are probably synchronous with
the principal deformation and metamorphism.
In summary, the volume of veins increases with the
metamorphic grade and their mineralogical content is
closely related to the paragenesis in the metapelitic host
48
 AZOR AND BALLÈVRE
Fig. 12. Kyanite relic included within andalusite from veins in Era de la Charneca (staurolite–andalusite zone).
rocks. These data suggest that they derive from local
subsolidus dehydration reactions or melting reactions,
depending on the grade. Alternatively, at least the peg-
matitic veins could be related to a granitic intrusion
located below the higher-grade part of the metamorphic
domain. These two possibilities will be considered in
more detail in the final discussion on the origin of the
metamorphism.
MINERAL CHEMISTRY OF
COEXISTING PHASES
The chemical composition of the phases in selected
samples was analysed with an electron microprobe
(Microsonde Ouest, Brest, France) using the PAP cor-
rection. Analytical conditions were 15 kV accelerating
voltage, 15 nA sample current and 6 s counting time.
Standards were albite (Na), orthoclase (K), corundum
(Al), wollastonite (Ca, Si), forsterite (Mg), ilmenite (Mn,
Ti), Fe2O3 (Fe) and ZnS (Zn). Zn was detected with the
La ray, and all others using the Ka ray.
Garnet
The garnets analysed (see Table 2 and Fig. 13) are
essentially almandine (64–82 mol %)–spessartine (5–27
mol %)–pyrope (4–10 mol %) solid solutions with low to
very low grossularite (0–6 mol %) and andradite (0–2
LP METAMORPHISM IN SIERRA ALBARRANA
mol %) contents. Microprobe analyses reveal two types
of variations: (1) zoning patterns within individual grains
and (2) systematic variations of the chemistry of garnet
rims between different metamorphic zones.
Zoning profiles are shown in Fig. 13. Most garnets from
the staurolite–garnet and staurolite–andalusite zones are
characterized by a decrease in MnO and an increase in
FeO and MgO contents from core to rim. Garnets from
the sillimanite zone are unzoned (Fig. 13, sample AA-
138) or show flat profiles with an increase in MnO and
a decrease in MgO contents at the rims (Fig. 13, sample
AA-80). This evolution of zoning profiles with increasing
grade is similar to the one documented by Dempster
(1985) in a medium-pressure metamorphic series, and is
interpreted in a similar way [see also Tracy (1982),
Loomis (1983) and Chakraborty & Ganguly (1990)].
Garnets from the staurolite–garnet and staurolite–
andalusite zones preserve growth zoning. In the sil-
limanite zone, more efficient diffusion appears to have
erased the growth zoning, and there is evidence for a
slight down-temperature reequilibration of garnet rims.
The spessartine content of cores decreases from the
staurolite–garnet zone (~15 mol %) to the sillimanite
zone (~8–10 mol %), but remains relatively high, which
suggests that garnet is stabilized by MnO. The pyrope
content increases from the staurolite–garnet zone (5 mol
%) to the sillimanite zone (~10 mol %). Accordingly,
the Mg/(Mg + Fe) ratio slightly increases from the
staurolite–garnet (0·06–0·07) to the staurolite–andalusite
49
 JOURNAL OF PETROLOGY VOLUME 38 NUMBER 1 JANUARY 1997

Table 2: Representative garnet analyses and structural formulae on a

basis of 12 oxygens and 8 cations; almandine, spessartine, pyrope,
grossular, andradite and uvarovite percentages are indicated; for each
sample, core and rim analyses are distinguished and the metamorphic
zone is indicated

Sample: AA-166 AA-175 AA-137 AA-223 AA-80

Met. zone: St–Grt St–Grt St–And St–And Sil
rim rim rim rim core

SiO2 36·84 36·95 36·73 37·32 37·00

TiO2 0·02 0·00 0·00 0·00 0·00
Al2O3 20·96 20·87 20·91 20·80 20·86
Cr2O3 0·01 0·11 0·05 0·12 0·00
FeO 33·01 35·40 37·14 32·73 34·45
MnO 6·84 3·43 2·44 4·53 4·81
MgO 1·32 2·34 1·85 2·23 2·65
CaO 1·61 1·52 1·78 1·86 0·93
Na2O 0·03 0·05 0·01 0·00 0·00
K2O 0·00 0·00 0·00 0·00 0·02
Total 100·64 100·64 100·90 99·59 100·71
Structural formulae on a basis of 12 oxygens and 8 cations
Si 2·985 2·975 2·961 3·024 2·976
AlIV 0·015 0·025 0·039 0·000 0·024
AlVI 1·987 1·956 1·948 1·987 1·955
Ti 0·001 0·000 0·000 0·000 0·000
Cr 0·001 0·007 0·003 0·007 0·000
Fe3+ 0·011 0·037 0·049 0·000 0·045
Fe2+ 2·227 2·348 2·456 2·218 2·273
Mn 0·470 0·234 0·167 0·311 0·328
Mg 0·159 0·280 0·222 0·269 0·317
Ca 0·140 0·131 0·154 0·161 0·080
Na 0·005 0·007 0·002 0·000 0·000
K 0·000 0·000 0·000 0·000 0·002
Almandine 74·33 78·45 81·91 74·96 75·83
Spessartine 15·68 7·81 5·55 10·50 10·93
Pyrope 5·32 9·37 7·41 9·09 10·58
Grossular 4·25 2·92 3·41 5·20 1·17
Andradite 0·40 1·23 1·62 0·00 1·49
Uvarovite 0·02 0·23 0·10 0·25 0·00
Mg/(Mg + Fe) 0·07 0·11 0·08 0·11 0·12

(~0·12) zones, and shows the same value in the sillimanite
zone as in the staurolite–andalusite zone.
Staurolite
The structural formulae of staurolite have been calculated
on a basis of 46 oxygens. Staurolite composition is very
similar in all the samples studied (Table 3). MnO and
ZnO contents are very low (0–0·4 and 0·1–1·1 wt %,
respectively). No zoning has been observed in staurolite.
The staurolite is slightly more magnesian [Mg/(Mg +
Fe) ratio of ~0·15–0·17] than the coexisting garnet.
Biotite
Typical compositions of biotites in the different meta-
morphic zones are shown in Table 4. Most of the biotites

50
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 13. Compositional zoning in garnets from the Sierra Albarrana metapelites. Each profile extends from rim to rim, through the core of the garnet.

51
 JOURNAL OF PETROLOGY VOLUME 38 NUMBER 1 JANUARY 1997

Table 3: Representative staurolite analyses and structural

formulae on a basis of 46 oxygens; the metamorphic zone
of each sample is indicated

Sample: AA-175 AA-137 AA-223

Met. zone: St–Grt St–And St–And

SiO2 27·97 27·91 27·26

TiO2 0·51 0·42 0·40
Al2O3 53·44 53·82 55·10
Cr2O3 0·00 0·05 0·00
FeO 13·81 14·97 13·02
MnO 0·17 0·10 0·34
MgO 1·61 1·51 1·46
ZnO 0·00 0·12 0·74
CaO 0·00 0·00 0·08
Na2O 0·08 0·00 0·01
K 2O 0·00 0·00 0·02
Total 97·59 98·90 98·44
Structural formulae on a basis of 46 oxygens
Si 7·778 7·704 7·527
Al 17·525 17·517 17·941
Ti 0·106 0·087 0·083
Cr 0·000 0·011 0·000
Fe2+ 3·213 3·456 3·008
Mn 0·040 0·023 0·080
Mg 0·669 0·621 0·599
Zn 0·000 0·024 0·152
Ca 0·000 0·000 0·022
Na 0·044 0·000 0·006
K 0·000 0·000 0·008
29·375 29·445 29·427
Mg/(Mg + Fe) 0·17 0·15 0·17

analysed, especially those from the staurolite–garnet zone,
show low K2O contents, probably owing to a partial
alteration to chlorite. Within the staurolite–andalusite
zone, biotite inclusions in andalusite are altered to a
lesser degree than matrix biotites, as shown by the higher
K2O contents. Biotites from the staurolite–andalusite
zone are slightly more magnesian than biotites from the
sillimanite and migmatite zones (Table 4 and Fig. 14).
Ti contents increase with metamorphic grade (Fig. 14)
regardless of whether ilmenite is present or not.
Chlorite
Chlorite is present in samples AA-85 (Sta–Grt zone) and
AA-175 (Sta–And zone) as porphyroblasts cutting across
the foliation (Fig. 8b) and in sample AA-166 (Sta–Grt
zone) as an alteration product of biotite and garnet.
Chlorites in these three samples have the same com-
position.
Muscovite
Most of the samples studied contain primary muscovite,
except some of the migmatite zone. No compositional
difference can be established between the crystals parallel
to the foliation, those included within andalusite, and
those appearing as porphyroblasts. Their celadonite con-
tents are consistently low or very low, as Si contents vary
between 3·03 and 3·10 per formula unit (p.f.u.) (Table
4). In most samples, MgO and FeO contents range from
0·3 to 0·6 and from 0·7 to 1·5 wt %, respectively.
Muscovites from sample AA-104 are richer in total FeO
(2·22–2·74 mol %) and poorer in Al2O3, which indicates
that most of the iron is ferric rather than ferrous. The

52
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Table 4: Representative biotite and muscovite analyses and structural formulae on a basis of 11 oxygens; the
metamorphic zone of each sample is indicated

Sample: AA-223 AA-80 AA-138 AA-18 AA-104 AA-166 AA-137 AA-138 AA-18 AA-104
Met. zone: St–And Sil Sil Mig Mig Grt St–And Sil Mig Mig
Mineral: Bt Bt Bt Bt Bt Ms Ms Ms Ms Ms

SiO2 35·94 34·99 35·99 35·15 35·28 46·50 47·35 46·68 45·32 52·74
TiO2 1·43 2·44 2·08 3·73 2·96 0·26 0·00 0·00 0·94 0·00
Al2O3 20·18 19·58 19·14 18·96 18·33 36·37 37·21 35·93 35·41 28·66
Cr2O3 0·21 0·04 0·05 0·00 0·00 0·02 0·02 0·03 0·10 0·01
FeO 20·46 21·51 21·39 20·00 21·09 1·52 0·68 0·73 0·90 2·22
MnO 0·12 0·04 0·15 0·04 0·04 0·02 0·06 0·00 0·12 0·00
MgO 8·74 7·66 8·07 7·58 8·11 0·49 0·39 0·37 0·55 2·58
CaO 0·03 0·00 0·00 0·00 0·00 0·03 0·00 0·00 0·00 0·13
Na2O 0·15 0·24 0·38 0·13 0·14 1·10 1·94 1·32 0·47 0·16
K 2O 8·65 8·74 8·15 9·87 10·28 8·90 7·96 9·99 11·16 9·41
Total 95·93 95·23 95·40 95·46 96·22 95·20 95·61 95·05 94·97 95·90
Structural formulae on a basis of 11 oxygens
Si 2·709 2·681 2·737 2·687 2·697 3·072 3·087 3·098 3·039 3·450
AlIV 1·291 1·319 1·263 1·313 1·303 0·928 0·913 0·902 0·961 0·550
AlVI 0·503 0·449 0·454 0·395 0·349 1·904 1·948 1·909 1·839 1·660
Ti 0·081 0·141 0·119 0·214 0·170 0·013 0·000 0·000 0·047 0·000
Cr 0·013 0·003 0·003 0·000 0·000 0·001 0·001 0·002 0·005 0·001
Fe2+ 1·290 1·379 1·361 1·278 1·349 0·084 0·037 0·041 0·051 0·121
Mn 0·008 0·002 0·010 0·003 0·002 0·001 0·003 0·000 0·007 0·000
Mg 0·982 0·875 0·915 0·864 0·924 0·048 0·038 0·037 0·055 0·251
Ca 0·002 0·000 0·000 0·000 0·000 0·002 0·000 0·000 0·000 0·009
Na 0·022 0·035 0·056 0·020 0·021 0·141 0·245 0·170 0·061 0·020
K 0·832 0·854 0·791 0·963 1·003 0·750 0·662 0·846 0·955 0·786
7·734 7·738 7·708 7·736 7·819 6·944 6·935 7·004 7·019 6·848
Mg/(Mg + Fe) 0·43 0·39 0·40 0·40 0·41

paragonite content of the muscovites ranges from 5 to
30 mol %. Fine-grained, secondary muscovites growing
at the expense of prismatic sillimanite (sample AA-104)
or K-feldspar (sample AA-138 ) have higher Si contents
(up to 3·45 p.f.u.) and, accordingly, higher MgO contents.
Feldspars
Plagioclases from the staurolite–garnet and staurolite–
andalusite zones are generally oligoclase (samples AA-
137 and AA-175), but nearly pure albite is also observed
(samples AA-53 and AA-223) (Table 5). It is not known
whether these albitic compositions result from the late
alteration of more calcic primary plagioclases, or rep-
resent compositions across the peristerite gap.
Plagioclases from the sillimanite zone have anorthite
contents equal to or higher than 18–20 mol %, except
in sample AA-104. Primary K-feldspar was analysed in
three samples (AA-18, AA-104 and AA-138) and its
composition approximates Or90Ab10 (Table 5).
Andalusite and sillimanite
Minor components were not detected in andalusite and
sillimanite. Only very low contents in Fe2O3 (0–0·8 wt
%) are present.
Oxides
Primary oxides in the samples studied are ilmenites with
MnO contents between 0·28 and 2·83 mol %. In the
matrix, most of the ilmenites are altered to a TiO2-rich
phase, most probably anatase as reported by Hébert &
Ballèvre (1993) from staurolite-bearing micaschists in the
Cadomian Belt of northern Brittany. Ilmenites included
in andalusite, staurolite and garnet are normally not

53
 JOURNAL OF PETROLOGY VOLUME 38
Fig. 14. Ti content (p.f.u.) vs Mg/(Fe + Mg) diagram for biotites of
the three higher-grade zones. The progressive increase in Ti content
from the staurolite–andalusite zone to the migmatite zone should be
noted.
altered. In the samples belonging to the migmatite zone
(AA-18, AA-20, AA-104), no ore minerals are present,
probably because all the TiO2 in the rock was in-
corporated into biotite (Fig. 14).
PHASE RELATIONS AND REACTION
HISTORY
The observed mineral assemblages, their relationships
with respect to the deformation history and their meas-
ured chemical compositions are used to decipher the
reaction history. Because the studied rocks invariably
contain quartz and muscovite or K-feldspar, we will use
the AFM projection (Thompson, 1957) to depict phase
compatibilities (Fig. 15). A vapour (Vap) phase (or a melt
phase in the case of the migmatite zone) is assumed to
be in excess. Because MnO is an essential constituent
for garnet (e.g. Giaramita & Day, 1991; Symmes & Ferry,
1992), it is taken into account in the AFM projection
(Fig. 15).
Phase relations in low-pressure metapelites have been
investigated using essentially two methods. Some workers
have calculated stable KFMASH reactions using ex-
perimentally based internally consistent thermodynamic
datasets for the mineral end-members, combined with
estimated solution models (e.g. Spear & Cheney, 1989;
Powell & Holland, 1990; Dymoke & Sandiford, 1992;
Xu et al., 1994). Other workers have constructed reaction
grids based on repeated occurrences of natural mineral
54
NUMBER 1 JANUARY 1997
assemblage sequences, combined with Schreinemakers’
rules (e.g. Hensen, 1971; Harte, 1975; Thompson, 1976;
Harte & Hudson, 1979; Pattison & Harte, 1985; Pattison
& Tracy, 1991).
A crucial point in the interpretation of the mineral
assemblages from the Sierra Albarrana area is the location
of some reactions with respect to the And–Sil equilibrium.
No general agreement has been reached in depicting
phase relations in low-pressure metapelites based strictly
on thermodynamic databases. In addition, comparison
with natural assemblages is plagued by departures from
the model KFMASH system owing to the incorporation
of non-KFMASH components, especially CaO and MnO
within garnet and possibly also Fe2O3 and TiO2 into
biotite. In this regard, a good example is shown by the
common occurrence of the assemblage Grt–
St–Bt–And–Ms–Qtz, where Grt is stabilized by the in-
corporation of MnO (see Giaramita & Day, 1991;
Pattison & Tracy, 1991; Symmes & Ferry, 1992; Droop
& Harte, 1995; Pitra & Guiraud, 1996).
Despite these difficulties, the petrogenetic grid of Pat-
tison & Tracy (1991) agrees well with the sequence of
mineral reactions deduced from our own observations,
and is used below. Figure 16 shows part of the KFMASH
grid of Pattison & Tracy (1991), but some modifications
have been made to take into account the additional
constituent MnO. Assuming that MnO enters only into
garnet, some invariant points and equilibrium curves are
displaced with respect to their position in the KFMASH
grid. In particular, the invariant point I1 will be displaced
within the andalusite stability field (Fig. 16). A sequence
of continuous and discontinuous model reactions can be
reconstructed with increasing grade.
The entrance in the staurolite–garnet zone
Because the transition between the biotite zone and the
staurolite–garnet zone is obscured by the Onza Fault,
the metamorphic reactions responsible for that transition
cannot be established accurately. The appearance of
garnet is most probably related to the continuous
FeMnMg reaction
Chl + Ms + Qtz = Grt + Bt + Vap. (1)
The entrance of staurolite is related to the continuous
FeMnMg reaction
Grt + Chl + Ms + Qtz = St + Bt + Vap (2)
or to the continuous FeMg reaction
Chl + Ms + Qtz = St + Bt + Vap. (3)
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Table 5: Representative plagioclase and K-feldspar analyses and structural formulae on

a basis of eight oxygens; the metamorphic zone of each sample is indicated

Sample: AA-137 AA-223 AA-80 AA-138 AA-18 AA-18

Met. zone: St–And St–And Sil Sil Mig Mig
Mineral: Pl Pl Pl Kfs Pl Kfs

SiO2 63·35 68·94 62·01 64·09 61·86 64·37

TiO2 0·02 0·05 0·00 0·03 0·00 0·00
Al2O3 22·90 19·46 23·88 19·33 23·96 18·68
Cr2O3 0·00 0·04 0·00 0·09 0·02 0·14
FeO 0·00 0·03 0·00 0·14 0·00 0·00
MnO 0·04 0·01 0·00 0·01 0·04 0·00
MgO 0·02 0·06 0·00 0·00 0·00 0·01
CaO 4·03 0·13 5·32 0·00 5·43 0·08
Na2O 8·76 11·90 8·69 0·29 8·75 1·50
K2O 0·00 0·06 0·15 15·19 0·17 15·48
Total 99·12 100·68 100·04 99·17 100·22 100·26
Structural formulae on a basis of 8 oxygens
Si 2·814 2·994 2·749 2·970 2·741 2·970
AlIV 1·186 0·996 1·248 1·030 1·252 1·016
AlVI 0·013 0·000 0·000 0·027 0·000 0·000
Ti 0·001 0·002 0·000 0·001 0·000 0·000
Cr 0·000 0·001 0·000 0·003 0·001 0·005
Fe2+ 0·000 0·001 0·000 0·005 0·000 0·000
Mn 0·002 0·000 0·000 0·000 0·001 0·000
Mg 0·001 0·002 0·000 0·000 0·000 0·001
Ca 0·192 0·006 0·253 0·000 0·258 0·004
Na 0·755 1·002 0·747 0·026 0·752 0·134
K 0·000 0·003 0·009 0·898 0·010 0·911
4·963 5·008 5·005 4·961 5·014 5·042
Ca/(Ca + Na) 0·20 0·00 0·25 0·26

These reactions account for the disappearance of primary,
synkinematic chlorite in the staurolite–garnet zone and
the decrease in spessartine content within garnet at
increasing grade.
The andalusite-in isograd
The beginning of the staurolite–andalusite zone is defined
by the appearance of andalusite. AFM-phase com-
patibilities (Fig. 15) suggest that the model KFMASH
discontinuous reaction
St + Chl + Ms + Qtz = Bt + And + Vap (4)
was responsible for andalusite formation. Garnet may
not have been involved in the andalusite-producing re-
action, as suggested by its euhedral shape when included
within andalusite.
The sillimanite-in isograd
Staurolite breakdown (i.e. the entrance within the sil-
limanite zone) took place either by means of the model
continuous reaction
St + Ms + Qtz = Als + Bt + Vap (5)
or the model discontinuous reaction
St + Ms + Qtz = Grt + Bt + Als + Vap (6)
as suggested by the AFM topologies seen in the staurolite–
andalusite and sillimanite zones (Fig. 15). At map scale,
the staurolite-out isograd coincides with the sillimanite-
in one (Figs 4 and 5), whereas andalusite is still re-
cognizable in the lower-grade part of the sillimanite zone.
Two models are compatible with these observations:
(1) The staurolite-breakdown reactions (5) and (6) could
have occurred within the andalusite stability field, with
the transformation of andalusite into sillimanite taking

55
 JOURNAL OF PETROLOGY VOLUME 38 NUMBER 1 JANUARY 1997

Fig. 15. Al2O3–(FeO+MnO)–MgO projection from quartz, muscovite and vapour of coexisting minerals in the Sierra Albarrana metapelites.
For the sample AA-223, in which the four-phase AFM assemblage Grt–St–Bt–And is present, an FeO–MnO–MgO projection from aluminium
silicate, quartz, muscovite and vapour is also shown. For the migmatite zone (sample AA-18) the projection has been done from K-feldspar in
instead of muscovite.

56
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 16. Petrogenetic grid for a portion of the KFMASH system [slightly modified after Pattison & Tracy (1991)] showing the sequence of
continuous (light lines bounding Fe-richer end-member reactions) and discontinuous reactions (heavy lines) with increasing grade. Equilibria
between aluminium silicate polymorphs are also indicated with light dashed lines. Dashed arrow indicates the proposed P–T conditions for the
staurolite–andalusite zone (left), sillimanite zone (centre), and migmatite zone (right). Melting reactions in the NKASH system after Thompson
& Algor (1977). (See text for further explanations.)

place during a later temperature increase. In this model,
a Grt–Bt–And zone must be observed before the first
appearance of sillimanite. This is not the case in our
study. Indeed, Grt–Bt–And assemblages are observed in
some metapelites (Fig. 4) but Grt stabilization is due to
its high MnO content.
(2) An alternative model assumes that reactions (5) and
(6) took place in the sillimanite stability field. In that
case, the replacement of andalusite by sillimanite must
have taken place before the staurolite breakdown. Be-
cause the Gibbs energy difference between andalusite
and sillimanite is extremely small, the reaction
And = Sil
is kinetically sluggish. On the contrary, the dehydration
reactions (5) and (6) are associated with large Gibbs
energy differences, which means that they are kinetically
easier (see Pattison, 1992). Thus, sillimanite growth is
most probably due to reaction (5) or (6) rather than to
reaction (7), which would explain the coincidence of
staurolite disappearance with sillimanite appearance.
The sillimanite zone is also marked by a strong decrease
in the modal abundance of garnet, which may be related
to the model continuous equilibrium
Grt + Ms = Als + Bt + Vap. (8)
The same reaction could also explain the garnet pseudo-
morphs observed in some andalusite porphyroblasts (Fig.
9).
(7)
Melting reactions
The migmatite zone is characterized by the widespread
occurrence of sillimanite and K-feldspar instead of mus-
covite and the large amount of leucosome segregations

57
 JOURNAL OF PETROLOGY VOLUME 38
(up to 20–30 % in volume), which are thought to derive
from in situ partial melting.
The model reaction
Ms + Ab + Qtz = Als + Kfs + Vap (9)
may account for the development of Kfs + Sil but not
of liquid (L). The onset of partial melting is therefore
probably due to one of the following three model re-
actions:
Ms + Kfs + Ab + Qtz + Vap = L (10)
Ms + Ab + Qtz + Vap = L + Als (11)
Ms + Qtz + Ab = Als + Kfs + L. (12)
The coincidence of Kfs–Sil and migmatites provides
good evidence that dehydration-melting of muscovite
[reaction (12)] largely controlled partial melting. Further
indications in favour of this model are the following. The
lack of Kfs in most metapelites before the migmatite
zone indicates that production of partial melts through
reaction (10) is unlikely. Moreover, the large volume of
leucosome suggests that fluid-present partial melting [i.e.
reactions (10) and (11)] is unlikely. Finally, the lack
of garnet within migmatitic metapelites indicates that
temperatures were not high enough to lead to de-
hydration-melting of Ms + Bt + Qtz or Bt + Sil +
Qtz.
Some of the rocks studied (e.g. sample AA-104) show
the four-phase assemblage Ms + Qtz + Kfs + Sil in
apparent textural equilibrium (Fig. 10b). This texture
suggests that either the H2O activity was buffered by the
four-phase assemblage or, alternatively, infiltrating H2O
maintained a high H2O activity.
In most rocks, especially those presenting evidence of
partial melting (e.g. sample AA-18), primary muscovite
is generally lacking, which suggests that fluid-absent
partial melting took place by reaction (12). In these rocks,
fibrolitic sillimanite develops at the expense of biotite
(Fig. 10a), and large muscovite porphyroblasts with in-
clusions of sillimanite needles overgrow the foliation but
are also slightly deformed (undulose extinction, kink-
folds). These textures are consistent with an increase in
H2O activity at the end of the deformation history, which
can be taken to result from the crystallization of nearby
melts or, alternatively, from fluid infiltration during cool-
ing from peak conditions.
P–T CONDITIONS
Geobarometry
The use of the plagioclase–garnet–Al2SiO5–quartz geo-
barometer (Newton & Haselton, 1981; Koziol & Newton,
58
NUMBER 1 JANUARY 1997
1988; Powell & Holland, 1988) is suspect in the rocks
studied, because the CaO and MnO contents in garnet
are very low and high, respectively. This fact by itself
accords well with low-pressure conditions (e.g. Hébert &
Ballèvre, 1993). Owing to the low anorthite content of
plagioclase, calculated pressure values are erratic, except
in the sillimanite zone where values range from 2·4 to
2·7 kbar. Previous work in metapelites from the sillimanite
zone gives P estimates of 4·9±0·5 kbar using the Hodges
& Spear (1982) calibration of the garnet–
plagioclase–Al2SiO5–quartz geobarometer (González del
Tanago & Peinado, 1990). Pressure estimates within
garnet-bearing amphibolites from the staurolite–
andalusite zone give values of the order of 4·3±0·5 kbar
(González del Tanago & Arenas, 1991). These values
are not significantly different from our own estimate of
4·0±0·5 kbar (see below).
Geothermometry
The geothermometer based on FeMg exchange between
garnet and biotite (Thompson, 1976; Ferry & Spear,
1978; Williams & Grambling, 1990) has been applied to
the rocks studied. The main problem to be faced with
this geothermometer relates to the fact that biotite is
partially altered to chlorite in most samples.
Using the less altered biotite compositions and garnet
rims for the staurolite–garnet and staurolite–andalusite
zones, and garnet cores for the sillimanite zone, garnet–
biotite geothermometry yields increasing temperatures
from the staurolite–garnet zone (500±50°C) to the sil-
limanite zone (600±50°C) (Fig. 17). In individual
samples, no systematic variation between the different
calibrations of the geothermometer is observed, except
in sample AA-166, where the spessartine content of
garnet is higher than in the other samples and the
calibration of Williams & Grambling (1990) gives higher
temperatures than the calibrations of Thompson (1976)
and Ferry & Spear (1978). The trend in estimated tem-
peratures is generally in accordance with the sequence
of mineral assemblages tied to the petrogenetic grid (Figs
16 and 17).
P–T HISTORY
The sequence of mineral assemblages in the studied
area is characteristic of the andalusite–sillimanite type of
Miyashiro (1961), found at pressures below the aluminium
silicate triple point. Several attempts have been made to
clarify relative pressure in such domains (e.g. Hietanen,
1967; Carmichael, 1978; Pattison & Tracy, 1991). On
the basis that (1) andalusite is found at lower grades than
sillimanite, (2) staurolite breakdown probably occurred in
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA

Fig. 17. Diagram showing temperature estimates obtained with the garnet–biotite geothermometer. Three calibrations of this geothermometer
have been used: T: Thompson (1976); FS: Ferry & Spear (1978); WG: Williams & Grambling (1990). The progressive temperature increase
from the staurolite–garnet zone to the sillimanite zone should be noted.

the sillimanite stability field and (3) migmatite formation
is due to muscovite breakdown, the Sierra Albarrana
metamorphism corresponds to facies series 2b of Pattison
& Tracy (1991). As stated above, it is implicitly assumed
that the succession of isograds in the Sierra Albarrana
area records an isobaric section through the crust. The
question that then arises is whether this pattern [‘the
metamorphic field gradient’ of England & Richardson
(1977)] results from a nearly isobaric heating–cooling
cycle (as in most contact aureoles), or from the pre-
servation of peak-temperature assemblages within rocks
having undergone a clockwise or anticlockwise P–T loop.
This question is an important one, because it has sig-
nificant implications for the origin of the low-pressure
metamorphism in the Sierra Albarrana area. One clue
is provided by the kyanite occurrences, which are dis-
cussed in detail below.
Significance of kyanite
Kyanite has been reported in some rocks from the
staurolite–andalusite and sillimanite zones (Garrote,
1976; González del Tanago & Peinado, 1990; Abad
Ortega, 1993; González del Tanago, 1993). Despite
the fact that kyanite is unusually common in similar
metamorphic series [see review by Pattison & Tracy
(1991)], its significance is not well understood.
Two types of kyanite occurrences are known in the
area studied. First, kyanite is found in veins in the
staurolite–andalusite zone (Abad Ortega, 1993; González
del Tanago, 1993), where it is usually overgrown by
andalusite (Fig. 12). Second, kyanite is also reported in
some cordierite-bearing schists from the higher-grade
rocks (González del Tanago & Peinado, 1990). According
to González del Tanago & Peinado (1990), kyanite occurs
as relic crystals enclosed within biotite. The rock contains
an assemblage consisting of quartz, plagioclase, K-feld-
spar, cordierite and prismatic sillimanite. These ob-
servations clearly indicate that the aluminium silicate
succession in the Sierra Albarrana area is first kyanite,
then andalusite and finally sillimanite.
The relic character of kyanite in the Sierra Albarrana
area precludes the possibility of an anticlockwise P–T
path, similar to the one described in Mount Isa
(Reinhardt, 1992). Another explanation for kyanite oc-
currence is that the rocks studied have been submitted
to a clockwise P–T path. If this was the case, severe
constraints on the shape of the P–T path can be derived
from the observed assemblages. Specifically, the P–T
path has to enter the andalusite stability field before the
first aluminium silicate producing reaction in metapelites,
i.e. reactions (4) and (5) (Fig. 16). This explanation is
consistent with the occurrence of kyanite overprinted by
andalusite in veins, and with the absence of kyanite in
metapelites, which indicates that in these rocks the first
aluminium silicate producing reaction took place in the
andalusite stability field. To sum up, we consider that the
few kyanite occurrences reported in the Sierra Albarrana
area record the earliest part of the P–T history. They are
compatible with either a nearly isobaric heating–cooling

59
 JOURNAL OF PETROLOGY
Fig. 18. Simplified P–T diagram for the Sierra Albarrana area. The
horizontal, dashed line shows the ‘metamorphic field gradient’. Two
clockwise P–T paths are proposed for the staurolite–andalusite (St–And)
and migmatite (Mig) zones. It should be noted that the P–T path of
the staurolite–andalusite zone first enters the kyanite stability field, then
the andalusite stability field, in accordance with the occurrence of
kyanite relics within andalusite. The location of the aluminium silicate
triple point and the melting reactions are taken from Holdaway &
Mukhopadhyay (1993) and Thompson & Algor (1977), respectively.
cycle or a clockwise path characterized by a very slight
increase or decrease in pressure at increasing tem-
peratures (Fig. 18).
P–T path
Two isobaric paths can reproduce the sequence of mineral
assemblages observed. They are parallel and located on
each side of the invariant point I1′. The higher-pressure
path (Fig. 16) is in closer agreement with the data
available, provided that sillimanite growth would have
occurred owing to reaction (5) or (6). Determining
whether P varies across metamorphic boundaries, how-
ever, presents greater difficulties. No definitive argument
in favour of a P variation across strike has been found.
In particular, observed assemblages within the migmatite
zone imply that pressure must have been lower than that
of the invariant point I3. In principle, the relative position
of invariant points I1 and I1′ on the one hand, and I2
and I3 on the other hand (Fig. 16), might indicate
whether a P increase or decrease has occurred from
the staurolite–andalusite zone to the migmatite zone.
However, because the exact P–T location of the invariant
points is dependent on the MnO content within garnet
(I1 and I1′) and H2O activity (I2 and I3), P variation cannot
VOLUME 38 NUMBER 1 JANUARY 1997
be determined with any certainty. If we assume that a P
variation did actually occur, such variation must have
been lower than the uncertainties and/or real dis-
placement of the invariant points owing to minor com-
ponents (i.e. only of the order of 0·5–1 kbar).
Given that the rocks studied were sediments deposited
on the Earth’s surface and because the synmetamorphic
deformation is compressional, we suggest that the most
probable P–T path involves a clockwise cycle with a very
slight increase in pressure during heating, followed by
cooling at slightly decreasing pressures (Fig. 18). Peak
P–T conditions therefore may define a metamorphic field
gradient which is nearly isobaric or presents a very gentle
positive slope. The location of the aluminium silicate triple
point (Kerrick, 1990; Bohlen et al., 1991; Hemingway et
al., 1991; Pattison, 1992; Holdaway & Mukhopadhyay,
1993) and of the invariant point I3 roughly constrain
pressures around 3·5±0·5 kbar in the staurolite–
andalusite zone and 4±0·5 kbar in the migmatite zone,
corresponding to a burial depth of the order of 10–12
km.
DISCUSSION AND CONCLUSION
Main constraints
Any model which seeks to account for the low-pressure
metamorphism in the Sierra Albarrana area should take
into account the main conclusions of this study, which
are summarized below.
The metamorphism in the Sierra Albarrana area is
synchronous with the principal deformation, which is
characterized by upright folds with subhorizontal axes
associated with a steeply dipping foliation and with a
subhorizontal stretching lineation. The intensity of this
deformation increases towards the migmatitic gneisses
located to the southwest of the Sierra Albarrana Quartz-
ites. This strongly deformed band is interpreted here as a
major ductile shear zone. Stratigraphic, palaeontological
and geochronological data constrain the tectono-meta-
morphic evolution of this area to be Variscan in age.
The regional metamorphism is characterized by in-
creasing grade towards the central part of the area, giving
way to a roughly concentric pattern of isograds around
the Sierra Albarrana Quartzites. The shear zone mainly
corresponds to the southwestern part of the migmatite
zone. To the southwest of the shear zone, metamorphic
grade progressively decreases through the sillimanite,
staurolite–andalusite, staurolite–garnet and biotite zones.
The highest metamorphic grade (migmatite zone) de-
velops to the northeast of the shear zone in a broad band
centred around the Sierra Albarrana Quartzites. To the
northeast of the migmatite zone, metamorphic grade
decreases through the sillimanite and staurolite–
andalusite zones.
60
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA
The metamorphism developed at relatively low pres-
sures (~3·5–4 kbar) and temperatures ranging from
~500°C in the staurolite–garnet zone to ~650°C in the
migmatite zone. No pressure difference can be found
across the studied area, suggesting an almost isobaric
section through the metamorphic terrane. Rare kyanite
relics are interpreted as belonging to the earliest part of
the prograde history. Therefore, no evidence for a
medium- or high-pressure metamorphism before the low-
pressure event has been found.
Voluminous coarse-grained veins and pegmatites occur
throughout the area, and their mineralogy is consistent
with the metamorphic grade in the enclosing rocks,
ranging from quartz–andalusite–muscovite in the
staurolite–andalusite zone to quartz–orthoclase–
albite–tourmaline in the migmatite zone.
Possible models
Taking into consideration the above constraints, several
models for explaining the low-pressure metamorphism
in the studied area are possible. These models are now
briefly discussed.
As a first model, the low-pressure metamorphism could
result from a thermal perturbation related to a large
number of granitoid intrusions at mid- to upper-crustal
depths (e.g. Barton & Hanson, 1989). Unfortunately,
plutonic rocks are relatively scarce or absent in the studied
area, i.e. south of the Badajoz–Córdoba shear zone (Fig.
1). Moreover, when present, granitoid intrusions (the
Cardenchosa granite) postdate the low-pressure meta-
morphism of the Sierra Albarrana area. Consequently,
this simple model is not appropriate for the studied area.
The second model for the development of low-pressure
metamorphism involves late-orogenic extension after
crustal thickening (e.g. Reinhardt & Kleemann, 1994;
Escuder Viruete et al., 1994). This model does not fit
well with the characteristics of the Sierra Albarrana area.
In particular, because no medium- or high-pressure relics
have been found, the studied area should be located in
the upper plate, i.e. above a detachment zone. In this
case, the principal foliation should be gently dipping,
which is not observed in the Sierra Albarrana area.
Therefore, this tectonic setting can be ruled out on the
basis of both structural and petrological data.
In a third model, the low-pressure metamorphism of the
Sierra Albarrana area could be related to a transcurrent
crustal-scale shear zone. This is consistent with the strain
pattern observed (steeply dipping foliation and sub-
horizontal stretching lineation). A more elaborated ver-
sion of this model results from a combination of the
following processes: an initial stage of moderate crustal
thickening, accounting for the burial of the Lower
Palaeozoic sediments; a second stage marked by the
61
development of a transcurrent shear zone, which is
synchronous with the peak P–T conditions; and a third
stage accounting for the exhumation of the Sierra Al-
barrana metamorphic rocks, during the sedimentation in
the Valdeinfierno basin (Lower Carboniferous).
In trancurrent shear zones, the heat source for low-
pressure metamorphism has been claimed to be shear
heating and/or heat advection by channelized fluid flow
or synkinematic magmatic intrusions (e.g. Hanmer et al.,
1982; Leloup & Kiénast, 1993). As regards the Sierra
Albarrana area, the model proposed here is liable to two
variants:
(1) The initial phase of thickening is associated with
prograde, dehydration, reactions at depth. The large
amount of H2O-rich fluid resulting from these de-
hydration reactions could have been channelled during
the second phase along the shear zone, and this results
in heat advection along the shear zone.
(2) One cannot preclude that the studied area is located
just above a synkinematic intrusion of granitoid com-
position, and that the post-metamorphic exhumation has
not been sufficient to unroof this magmatic body. Under
this hypothesis, the numerous pegmatite veins would
provide evidence for the existence of a granitoid body at
depth. The large amounts of fluid released during their
crystallization were probably channelized along the shear
zone.
Further research needs to be carried out to determine
the possible contribution of both the fluid flow and the
magmatic intrusion to the heat budget, but the Sierra
Albarrana area provides valuable insights into the de-
velopment of low-pressure metamorphism in terranes not
previously submitted to a higher-pressure event.
ACKNOWLEDGEMENTS
We gratefully thank Marcel Bohn for technical assistance
with microprobe work. Jean L. Sanders and Francisco
Gonzálvez Garcı́a considerably improved the English
version of the manuscript. Michel Lautram drew some
of the figures. Mario Sánchez Gómez is kindly thanked
for computer assistance in the preparation of some figures.
Many thanks are due to D. R. M. Pattison and S. Harley
for their detailed and critical review of the manuscript.
Discussions with J. F. Simancas and F. González Lodeiro
were very helpful in clarifying some aspects of the geo-
logical history of the area studied. A.A. received financial
support from the CICYT (Spain), Projects PB-90/
C0860/C03/01 and PB 93/1149/C03/01. M.B. grate-
fully acknowledges the hospitality of the Residence ‘Car-
men de la Victoria’ at the University of Granada.
 JOURNAL OF PETROLOGY
REFERENCES
Abad Ortega, M., 1993. Mineralogı́a y geoquı́mica de las pegmatitas
de Sierra Albarrana (Provincia de Córdoba). Tesis Doctoral, Un-
iversidad de Granada.
Abad Ortega, M. & Nieto, F., 1995a. Extension and closure of the
compositional gap between Mn- and Mg-rich chlorites toward Fe-
rich compositions. European Journal of Mineralogy 7, 363–367.
Abad Ortega, M. & Nieto, F., 1995b.Genetic and chemical relationships
between berthierine, chlorite and cordierite in nodules associated to
granitic pegmatites of Sierra Albarrana (Iberian Massif, Spain).
Contributions to Mineralogy and Petrology 120, 327–336.
Abad Ortega, M., Fenoll Hach-Alı́, P., Martı́n Ramos, J. D. & Ortega
Huertas, M., 1993. The feldspars of the Sierra Albarrana granitic
pegmatites, Córdoba, Spain. Canadian Mineralogist 31, 185–202.
Azor, A., 1994. Evolución tectonometamórfica del lı́mite entre las
Zonas Centroibérica y de Ossa–Morena (Cordillera Varisca, SO de
España). Tesis Doctoral, Universidad de Granada.
Azor, A., González Lodeiro, F. & Simancas, J. F., 1994. Tectonic
evolution of the boundary between the Central Iberian and Ossa–
Morena Zones (Variscan Belt, SW Spain). Tectonics 13, 45–61.
Barton, M. D. & Hanson, R. B., 1989. Magmatism and the development
of low-pressure metamorphic belts: implications from the western
United States and thermal modeling. Geological Society of America
Bulletin 101, 1051–1065.
Bohlen, S. R., Montana, A. & Kerrick, D. M., 1991. Precise de-
terminations of the equilibria kyanite = sillimanite and kyanite =
andalusite and a revised triple point for Al2SiO5 polymorphs. American
Mineralogist 76, 677–680.
Burg, J. P., Iglesias, M., Laurent, Ph., Matte, Ph. & Ribeiro, A., 1981.
Variscan intracontinental deformation: the Coimbra–Córdoba Shear
Zone (SW Iberian Peninsula). Tectonophysics 78, 161–177.
Carmichael, D. M., 1978. Metamorphic bathozones and bathograds:
a measure of the depth of post-metamorphic uplift and erosion on
a regional scale. American Journal of Science 278, 769–797.
Chakraborty, S. & Ganguly, J., 1990. Compositional zoning and cation
diffusion in garnets. In: Ganguly, J. (ed.) Diffusion, Atomic Ordering, and
Mass Transport: Selected Problems in Geochemistry. Advances in Physical
Geochemistry 10, 120–175.
Contreras, M. C., Garrote, A. & Sánchez-Carretero, R., 1984. Aso-
ciaciones minerales metamórficas con cloritoide en Sierra Albarrana
(Córdoba). Cuaderno Laboratorio Xeoloxigo Laxe 8, 221–234.
Dallmeyer, R. D. & Quesada, C., 1992. Cadomian vs. Variscan
evolution of the Ossa–Morena Zone (SW Iberia): field and 40Ar/
39
Ar mineral age constraints. Tectonophysics 216, 339–364.
Delgado Quesada, M., 1971. Esquema geológico de la hoja N°878 de
Azuaga (Badajoz). Boletin Geológico y Minero 82, 277–286.
Delgado Quesada, M., Garrote, A. & Sánchez Carretero, R., 1985. El
magmatismo de la alineación La Coronada–Villaciosa de Córdoba
en su mitad oriental, Zona de Ossa–Morena. 5a Reunión del Grupo
de Ossa–Morena. Temas Geológico Mineros. Madrid: IGME, pp.
41–64.
De Yoreo, J. J., Lux, D. R. & Guidotti, C. V., 1991. Thermal modelling
in low-pressure/high-temperature metamorphic belts. Tectonophysics
188, 209–238.
Dempster, T. J., 1985. Garnet zoning and metamorphism of the
Barrovian type area, Scotland. Contributions to Mineralogy and Petrology
89, 30–38.
Droop, G. T. R. & Harte, B., 1995. The effect of Mn on the
phase relations of medium-grade pelites: constraints from natural
assemblages on petrogenetic grid topology. Journal of Petrology 36,
1549–1578.
VOLUME 38 NUMBER 1 JANUARY 1997
Dymoke, P. & Sandiford, M., 1992. Phase relationships in Buchan
facies series pelitic assemblages: calculations with application to
andalusite–staurolite parageneses in the Mount Lofty Ranges, South
Australia. Contributions to Mineralogy and Petrology 110, 121–132.
England, P. C. & Richardson, S. W., 1977. The influence of erosion
upon the mineral facies of rocks from different metamorphic en-
vironments. Journal of the Geological Society of London 134, 201–213.
Escuder Viruete, J., Arenas, R. & Martı́nez Catalán, J. R., 1994.
Tectonothermal evolution associated with Variscan crustal extension
in the Tormes Gneiss Dome (NW Salamanca, Iberian Massif, Spain).
Tectonophysics 238, 117–138.
Ferry, J. M. & Spear, F. S., 1978. Experimental calibration of the
partitioning of Fe and Mg between biotite and garnet. Contributions
to Mineralogy and Petrology 66, 113–117.
Fricke W., 1941. Die Geologie des Grenzgebietes zwischen nor-
dostlichen Sierra Morena und Extremadura. Ph.D Thesis, University
of Berlin.
Gabaldón, V. & Quesada, C., 1986. Exemples de bassins houillers
limniques du sud-ouest de la Péninsule Ibérique: évolution séd-
imentaire et contrôle structural. Mémoires de la Société Géologique de
France 149, 27–36.
Gabaldón, V., Garrote, A. & Quesada, C., 1983. Las Cuencas de
Valdeinfierno y Benajarafe (Tournaisiense–Viseense). Ca-
racterización sedimentológica y implicaciones regionales. Dominio
de Sierra Albarrana (Zona de Ossa–Morena). Comunicações dos Serviços
Geológicos de Portugal 69, 209–218.
Garrote, A., 1976. Asociaciones minerales del núcleo metamórfico de
Sierra Albarrana (Prov. de Córdoba). Sierra Morena Central. Me-
mórias e Notı́cias, Publicações do Museu e Laboratório Mineralógico e Geológico
da Universidad de Coimbra 82, 17–39.
Garrote, A. & Broutin, J., 1979. Le bassin tournaisien de Benajarafe
(Province de Cordoue, Espagne). Géologie et premières données
paléobotaniques et palynologiques. Comptes Rendus du 104ème Congrès
National des Sociétés Savantes 1, 175–184.
Garrote, A. & Sánchez Carretero, R., 1979. Granitos postcinemáticos
de tendencia alcalina en Ossa–Morena: el stock de La Cardenchosa
(Provincia de Córdoba). Acta Geológica Hispánica 14, 90–96.
Garrote, A., Ortega Huertas, M. & Romero, J., 1980. Los yacimentos
de pegmatitas de Sierra Albarrana (Provincia de Córdoba, Sierra
Morena). 1a Reunión sobre la geologı́a de Ossa–Morena. Temas Geológico
Mineros. Madrid: IGME, pp. 145-168.
Giaramita, M. J. & Day, H. W., 1991. The four-AFM assemblage
staurolite–aluminum silicate–biotite–garnet: extra components and
implications for staurolite-out isograds. Journal of Petrology 32, 1203–
1229.
González del Tanago, J., 1991. Las pegmatitas graniticas de Sierra
Albarrana (Córdoba, España): mineralizaciones de berilio. Boletin
Geológico y Minero 102, 578–603.
González del Tanago, J., 1993. El Núcleo Metamórfico de Sierra
Albarrana y su campo pegmatı́tico granı́tico asociado, Macizo Ib-
érico, Córdoba, España. Tesis Doctoral, Universidad Complutense
de Madrid.
González del Tanago, J. & Arenas, R., 1991. Anfibolitas granatı́feras
de Sierra Albarrana (Córdoba). Termobarometrı́a e implicaciones
para el desarrollo del metamorfismo regional. Revista de la Sociedad
Geológica de España 4, 251–269.
González del Tanago, J. & Peinado, M., 1990. Contribución al estudio
del metamorfismo de Sierra Albarrana (Z.O.M., Córdoba, España).
Boletin Geológico y Minero 101, 678–700.
Hanmer, S. & Passchier, C., 1991. Shear-sense indicators: a review.
Geological Survey of Canada, Paper 90-17.
Hanmer, S., Le Corre, C. & Berthé, D., 1982. The role of Hercynian
granites in the deformation and metamorphism of Brioverian and
62
 AZOR AND BALLÈVRE LP METAMORPHISM IN SIERRA ALBARRANA
Palaeozoic rocks of Central Brittany. Journal of the Geological Society of
London 139, 85–93.
Harte, B., 1975. Determination of a pelite petrogenetic grid for the
eastern Scottish Dalradian. Carnegie Institution of Washington, Yearbook
74, 438–446.
Harte, B. & Hudson, N. F. C., 1979. Pelite facies series and the
temperatures and pressures of Dalradian metamorphism in E Scot-
land. In: Harris, A. L., Holland, C. H. & Leake, B. E. (eds) The
Caledonides of the British Isles Reviewed. Geological Society of London Special
Publication 8, 323–337.
Hébert, R. & Ballèvre, M., 1993. Petrology of staurolite-bearing me-
tapelites from the Cadomian Belt, Northern Brittany (France): con-
straints on low-pressure metamorphism. Bulletin de la Société Géologique
de France 164, 215–228.
Hemingway, B. S., Robie, R. A., Evans, H. T. & Kerrick, D. M.,
1991. Heat capacities and entropies of sillimanite, fibrolite, an-
dalusite, kyanite, and quartz and the Al2SiO5 phase diagram. American
Mineralogist 76, 1597–1613.
Hensen, B., 1971. Theoretical phase relations involving cordierite
and garnet in the system MgO–FeO–Al2O3–SiO2. Contributions to
Mineralogy and Petrology 33, 191–214.
Hietanen, A., 1967. On the facies series in various types of meta-
morphism. Journal of Geology 75, 187–214.
Hodges, K. V. & Spear, F. S., 1982. Geothermometry, geobarometry
and the Al2O3 triple point at Mt. Moosilauke, New Hampshire.
American Mineralogist 67, 1118–1134.
Holdaway, M. J. & Mukhopadhyay, B., 1993. A reevaluation of the
stability relations of andalusite: thermochemical data and phase
diagram for the aluminum silicates. American Mineralogist 78,
298–315.
Kerrick, D. M., 1990. The Al2O3 polymorphs. Mineralogical Society of
America, Reviews in Mineralogy 22, 1–406.
Koziol, A. M. & Newton, R. C., 1988. Redetermination of the anorthite
breakdown reaction and improvement of the plagioclase–
garnet–Al2O3–quartz geobarometer. American Mineralogist 73, 216–
223.
Kretz, R., 1983. Symbols for rock-forming minerals. American Mineralogist
68, 277–279.
Laurent, P., 1974. Structure et pétrologie de la Bande Blastomylonitique
de Badajoz–Córdoba (Chaı̂ne hercynienne sud-ibérique à l’est
d’Azuaga, Espagne). Description et interprétation de la déformation
dans les blastomylonites. Thèse Troisième Cycle, Université des
Sciences et Techniques du Languedoc, Montpellier.
Leloup, P. H. & Kiénast, J. R., 1993. High-temperature metamorphism
in a major strike-slip shear zone: the Ailao–Red River, People’s
Republic of China. Earth and Planetary Science Letters 118, 213–234.
Liñán, E., 1978. Bioestratigrafı́a de la Sierra de Córdoba. Tesis Doc-
toral, Universidad de Granada.
Loomis, T. P., 1983. Compositional zoning of crystals: a record of
growth and reaction history. In: Saxena, S. K. (ed.) Kinetics
and Equilibrium in Mineral Reactions. Advances in Physical Geochemistry 3,
1–60.
Marcos, A., Azor, A., González Lodeiro, F. & Simancas, J. F., 1991.
Early Phanerozoic trace fossils from the Sierra Albarrana Quartzites
(Ossa–Morena Zone, Southwest Spain). Scripta Geologica 97,
47–53.
Matte, Ph., 1991. Accretionary history and crustal evolution of the
Variscan belt in Western Europe. Tectonophysics 196, 309–337.
Miyashiro, A., 1961. Evolution of metamorphic belts. Journal of Petrology
2, 277–311.
Newton, R. C. & Haselton, H. T., 1981. Thermodynamics of the
garnet–plagioclase–Al2O3–quartz geobarometer. In: Newton, R. C.,
63
Navrotsky, A. & Wood, B. J. (eds) Thermodynamics of Minerals and
Melts. New York: Springer-Verlag, pp. 131–147.
Ortega Huertas, M., Garrote, A., Rodrı́guez, J. & Fenoll Hach-Alı́, P.,
1982. Pegmatitic assemblages in the metamorphic core of Sierra
Albarrana (Córdoba, Spain). Proceedings 13th General Meeting Inter-
national Mineralogical Association (Varna), pp. 641–649.
Pattison, D. R. M., 1992. Stability of andalusite and sillimanite and
the Al2O3 triple point: constraints from the Ballachulish Aureole,
Scotland. Journal of Geology 100, 423–446.
Pattison, D. R. M. & Harte, B., 1985. A petrogenetic grid for pelites
in the Ballachulish Aureole and other Scottish thermal aureoles.
Journal of the Geological Society of London 142, 7–28.
Pattison, D. R. M. & Tracy, R. J., 1991. Phase equilibria and thermo-
barometry of metapelites. In: Kerrick, D. M. (ed.) Contact Meta-
morphism. Mineralogical Society of America, Reviews in Mineralogy 25,
105–206.
Pitra, P. & Guiraud, M., 1996. Probable anticlockwise P–T evolution
in extending crust: Hlinsko region, Bohemian Massif. Journal of
Metamorphic Geology 14, 49–60.
Powell, R. & Holland, T. J. B., 1988. An internally consistent dataset
with uncertainties and correlations: 3. Applications to geobarometry,
worked examples and a computer program. Journal of Metamorphic
Geology 6, 173–204.
Powell, R. & Holland, T. J. B., 1990. Calculated mineral equilibria in
the pelitic system, KFMASH (K2O–FeO–MgO–Al2O3–SiO2–H2O).
American Mineralogist 75, 367–380.
Quesada, C. & Munhá, J., 1990. Part V: Ossa–Morena Zone, 5:
Metamorphism. In: Dallmeyer, R. D. & Martı́nez Garcı́a, E. (eds)
Pre-Mesozoic Geology of Iberia. Berlin: Springer-Verlag, pp. 314–
320.
Ramsay, J. G. & Huber, M. I., 1987. The Techniques of Modern Structural
Geology. Volume 2: Folds and Fractures. London: Academic Press, pp.
595–640.
Reinhardt, J., 1992. Low-pressure, high-temperature metamorphism
in a compressional tectonic setting: Mary Kathleen Fold Belt, north-
eastern Australia. Geological Magazine 129, 41–57.
Reinhardt, J. & Kleemann, U., 1994. Extensional unroofing of granulitic
lower crust and related low-pressure, high-temperature meta-
morphism in the Saxonian Granulite Massif, Germany. Tectonophysics
238, 71–94.
Roldán, F., 1983. La cuenca tournaisiense de Valdeinfierno, in-
terpretación tectónica y paleoambiental (Córdoba, España). In:
Lemos de Sousa, M. J. (ed.) Contributions to the Carboniferous Geology
and Palaeontology of the Iberian Peninsula. Porto: Universidade de Porto,
pp. 41–50.
Simpson, C. & Schmid, S. M., 1983. An evaluation of criteria to
deduce the sense of movement in sheared rocks. Geological Society of
America Bulletin 94, 1281–1288.
Spear, F. S. & Cheney, J. T., 1989. A petrogenetic grid for pelitic schists
in the system SiO2–Al2O3–FeO–MgO–K2O–H2O. Contributions to
Mineralogy and Petrology 101, 149–164.
Symmes, G. H. & Ferry, J. M., 1992. The effect of whole-rock
MnO content on the stability of garnet in pelitic schists during
metamorphism. Journal of Metamorphic Geology 10, 221–237.
Thompson, A. B., 1976. Mineral reactions in pelitic rocks: II. Cal-
culations of some P–T–X(Fe–Mg) phase relations. American Journal of
Science 276, 425–454.
Thompson, A. B. & Algor, J. R., 1977. Model systems for anatexis of
pelitic rocks. I—Theory of melting reactions in the system KAlO2–
NaAlO2–Al2O3–SiO2–H2O. Contributions to Mineralogy and Petrology 63,
247–269.
Thompson, J. B., Jr, 1957. The graphical analysis of mineral assemblages
in pelitic schists. American Mineralogist 42, 842–858.
 JOURNAL OF PETROLOGY VOLUME 38
Tracy, R. J., 1982. Compositional zoning and inclusions in metamorphic
minerals. In: Ferry, J. M. (ed.) Characterization of Metamorphism through
Mineral Equilibria. Mineralogical Society of America, Reviews in Mineralogy
10, 355–397.
Wagner, R. H., 1978. The Valdeinfierno sequence (prov. Córdoba):
its tectonic, sedimentary and floral significance. Annales de la Société
Géologique du Nord 98, 59–66.
64
NUMBER 1 JANUARY 1997
Williams, M. L. & Grambling, J. A., 1990. Manganese, ferric iron, and
the equilibrium between garnet and biotite. American Mineralogist 75,
886–908.
Xu, G., Will, T. M. & Powell, R., 1994. A calculated petrogenetic grid
for the system K2O–FeO–MgO–Al2O3–SiO2–H2O, with particular
reference to contact-metamorphosed pelites. Journal of Metamorphic
Geology 12, 99–119.