Demystifing Migmatites

Applied Earth Science
Transactions of the Institutions of Mining and Metallurgy: Section B
ISSN: 0371-7453 (Print) 1743-2758 (Online) Journal homepage: http://www.tandfonline.com/loi/yaes20
Demystifying migmatites: introduction for field-

based geologist
M. Pawley, A. Reid, R. Dutch & W. Preiss
To cite this article: M. Pawley, A. Reid, R. Dutch & W. Preiss (2015) Demystifying migmatites:
introduction for field-based geologist, Applied Earth Science, 124:3, 147-174
To link to this article: http://dx.doi.org/10.1179/1743275815Y.0000000014
Published online: 18 Nov 2015.
Submit your article to this journal
Article views: 2
View related articles
Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=yaes20
Download by: [SA Aquatic Sciences Centre], [Mark Pawley] Date: 01 February 2016, At: 14:38
REVIEW
Demystifying migmatites: an introduction for
the field-based geologist
M. Pawley*1, A. Reid1,2, R. Dutch1,2 and W. Preiss1
Migmatites are heterogeneous, medium- to high-grade metamorphic rocks with at least one
component formed by partial melting. As migmatites form under P–T conditions common in
the middle and lower crust, they are widespread and likely to be encountered in the field. They are
complex rocks that can be confusing, and the present paper provides a background for the
field-based geologist. Migmatites comprise several main components. Neosome is newly formed
Downloaded by [SA Aquatic Sciences Centre], [Mark Pawley] at 14:38 01 February 2016
by partial melting and includes leucosome and residuum (or melanosome). These essentially
represent the liquid and solid components of the melt reaction. Palaeosome is the part of the
migmatite that did not undergo partial melting and is typically of a more refractory composition.
Migmatites can be divided into two main types. Metatexites have a lower melt fraction and
preserve coherent pre-partial melting textures in the palaeosome. Diatexites are dominated by
the neosome, and coherent pre-partial melting structures are absent. These can be further
subdivided, with the different metatexites based on where the melt has ponded to form the
leucosomes, while diatexites are subdivided based on the geometry of the palaeosome and/or
residuum or its absence. The different types of migmatites provide information about the
proportion and distribution of melt and the type and degree of strain. Facing direction can be
obtained from migmatites, using criteria such as cauliflower structures and vein orientation and
distribution. The leucosome margins can provide information about the rheological contrasts
between the leucosomes and their host and their timing relative to anatexis and deformation.
Consequently, these complex rocks can be important tools in understanding deformation within
the middle to lower crust.
Keywords: Migmatites, Metamorphism, Partial melting, Deformation
Introduction in the field, where their complexity can confuse and inti-
midate the geologist. However, as well as forming spec-
Migmatites are complex, medium- to high-grade meta- tacular exposures, migmatites can also provide a large
morphic rocks formed by partial melting. They have amount of geological information.
been recognised across most of geological time, develop The aim of the present paper is to provide an intro-
in most tectonic settings, and can affect a wide range of duction to the processes that form migmatites, the ter-
protoliths. The pressure (P) and temperature (T ) con- minology applied to them, and their usefulness. The first
ditions under which rocks undergo partial melting require section provides the background necessary to understand
the perturbation of a typical geothermal gradient. This and describe these rocks, including how they form, the
can be achieved through tectonic processes such as crustal parts of a migmatite, and the different types of migmatitic
thickening, enhanced mantle heat flow, mafic under- rocks. The second section provides some practical infor-
plating, or the advection of heat through emplacement of mation, including what information can be extracted
magmas into the middle crust (Brown, 1994; Clemens, from a migmatite, dating of these rocks, and some ideas
2013). Burial of highly radiogenic crust may also increase on how to deal with migmatitic rocks in the field. The
the geothermal gradient and lead to widespread migma- final section includes a series of case studies from
tite formation (Gerdes et al., 2000; McLaren et al., 2015). South Australia, which will give the reader some idea of
Consequently, migmatites can commonly be encountered what information can be extracted from migmatitic rocks.
The present paper is not meant to be an exhaustive
analysis of the wide range of empirical, theoretical and
1
experimental work that has been done on migmatites.
Geological Survey of South Australia, Department of State Development
Level 4, 101 Grenfell Street, Adelaide, SA, 5000, Australia
Instead, it is hoped that the present paper will give
2
School of Physical Sciences, The University of Adelaide, Adelaide, SA, geologists enough background to collect meaningful
5000, Australia data from migmatitic rocks, whether in outcrop or drill
*Corresponding author, email Mark.Pawley@sa.gov.au core, and confidently interpret these observations. For
Ñ 2015 Institute of Materials, Minerals and Mining and The AusIMM

Published by Maney on behalf of the Institute and The AusIMM
Received 31 March 2015; accepted 21 June 2015
DOI 10.1179/1743275815Y.0000000014 Applied Earth Science (Trans. Inst. Min. Metall. B) 2015 VOL 124 NO 3 147
Pawley et al. Demystifying migmatites
further information, the reader is referred to Sawyer contain hydrous mineral phases, such as mica-rich
(2008a) and the references therein. schists, will generally tend to be more fertile and melt at
lower temperatures. This contrasts with comparatively
dry rocks (i.e. those lacking chemically-bound fluid),
Formation of migmatite such as quartzite and marble, which will generally
A migmatite has been defined by Sawyer (2008a) as: require higher temperatures to begin melting (in the
absence of free H2O).
a rock found in medium- and high-grade
The availability of free H2O (or an H2O-rich fluid)
metamorphic areas that can be heterogeneous at the
also has a major effect on controlling the temperature at
microscopic to macroscopic scale and that consists of
which partial melting may begin and the volume and
two, or more, petrographically different parts...one
mobility of the melt produced (e.g. Clemens, 1984).
of which must have formed by partial melting.
Many natural and experimental observations have
The partial melting, or anatexis, of crustal rocks is a demonstrated that free H2O can considerably lower the
complex process controlled by a number of variables temperature of the solidus and affect the stability of
including the P–T conditions, the composition or many hydrous mineral phases (e.g. Holland and Powell,
fertility of the rock protolith and the presence or absence 2001; Thompson, 2001; White et al., 2001). In rocks of
of free H2O or other volatile phases. typical pelitic composition from the Eyre Peninsular,
Partial melting, like all metamorphic reactions, is South Australia, H2O saturation can lower the solidus to
controlled by the chemical stability of a given mineral temperatures around 650uC at 5 kbar (Fig. 1) producing
assemblage at particular P–T conditions. As a rock a H2O-saturated melt. The control of H2O on the solidus
moves through the crust, the P–T conditions change, can be demonstrated using phase equilibria modelling
bringing about disequilibrium conditions and inducing and varying the amount of H2O in the system. A tem-
the mineral assemblage to alter as it attempts to mini- perature–molar% H2O pseudosection generated using
mise the Gibbs free energy of the system (Spear, 1993). the program THERMOCALC (Powell and Holland,
The temperature of a system has a major control on 1988) for a pelitic rock sampled from Shoal Point on
inducing partial melting and the amount of melt pro- southern Eyre Peninsula, South Australia (Fig. 2;
duced because most melt-producing reactions have a Dutch, 2009) demonstrates the behaviour that may be
steep slope (i.e. a large DP/DT ) in P–T space.
The propensity for a given rock to start melting at
given P–T conditions, otherwise known as melt fertility,
is controlled by the bulk composition of the rock
(Spear, 1993). Partial melting reactions in crustal rocks
are strongly controlled by the modal proportions of
quartz and feldspar and the amount of hydrous minerals
present such as muscovite, biotite and hornblende
(Thompson, 1996). Melting reactions, like all chemical
reactions, are controlled by the stoichiometry of the
reaction and will therefore be limited by the availability
of the required reactants. Melting initially occurs
through the process of dehydration–melting reactions
that generally take the form
Hydrous mineral+ feldsparþquartz ¼ H2O under-

saturated melt þ crystal residue (Thompson, 2001).
Typically for felsic crustal rocks, this involves muscovite

and biotite to produce either a ferromagnesian
(e.g. garnet or cordierite) or aluminosilicate (e.g. sillima-
nite) mineral and melt, for example via the reactions
Muscovite þ quartz ¼ garnet þ biotite þ K-

feldspar þ melt (Spear et al., 1999) and
Biotite þ sillimanite þ quartz ¼ garnet þ -
cordierite þ K-feldspar þ melt (Spear et al., 1999). 1 P–T pseudosection for metapelitic rock from southern
Eyre Peninsula calculated with H2O saturated bulk
At progressively higher T conditions, or as hydrous
composition using phase equilibria modelling program
phases are consumed, melt production will continue via
THERMOCALC (Powell and Holland, 1988) in chemical
the breakdown of more refractory minerals such as
system MnNCKFMASH (bulk composition in figure is in
Hornblende þ quartz ¼ orthopyroxene þ clinopyro- molar%). V refers to variance or degrees of freedom within
xene þ plagioclase þ melt (Spear et al., 1999) in given stability field. Dashed line is solidus, univariant line
mafic lithologies, or where first melt occurs. Narrow field to up-temperature
Garnet þ sillimanite ¼ spinel þ quartz þ melt side of solidus is small stability field with H2O saturated
(Patiño Douce and Johnston, 1991) in pelitic rocks. silicate melt. At 5 kbar, first melt occurs through dehy-
dration reactions involving muscovite, whereas at lower
Because dehydration reactions are so important for pressures (below 4 kbar) it occurs via biotite breakdown
producing melt at lower temperatures, rocks that reactions, which occur at slightly higher temperatures
148 Applied Earth Science (Trans. Inst. Min. Metall. B) 2015 VOL 124 NO 3
strongly partitioned into the melt phase, if melt is not

removed via an open system it will cause the H2O con-
tent of the system to remain constant. As a consequence,
the rock package will undergo significant retrogression
during recrystallisation of the melt phase and the sub-
sequent rehydration of the rock. The amount of melt
loss required to preserve granulite mineral assemblages
is strongly dependent on the composition of the
rock package. Thermodynamic modelling of an alumi-
nous metapelitic assemblage by White and Powell (2002)
suggests that the removal of nearly all the melt is
required to preserve the peak mineral assemblage,
whereas in a subaluminous metapelite, the melt loss
required is roughly half that for an aluminous assem-
blage. The fact that granulite facies mineral assemblages
are preserved indicates that a significant amount of melt
has been lost from the system by the process of melt
segregation into leucosomes and subsequent migration
from the system. Thus, melt loss and open system

behaviour is fundamental to the preservation of weakly
retrogressed granulite facies rocks.
Evidence of removal of material from a rock mass
2 Temperature–molar% H2O pseudosection calculated for
may not be easily recognised as melt loss from the
pelitic metasedimentary rock from Shoal Point on Eyre
system can be difficult to identify. It is possible to do this
Peninsula. Calculated using phase equilibria modelling
using geochemistry (e.g. Sawyer, 1991), but discussing
program THERMOCALC (Powell and Holland, 1988) in
this approach is outside the scope of the present paper.
chemical system NCKFMASHTO (bulk composition in
However, a good example of this process can be found in
figure is in molar%). V refers to variance or degrees of
the metapelites at Round Hill, east of Broken Hill
freedom within given stability field. Molar% H2O varies
(White et al., 2004). The rocks in this area contain two
along x-axis from 7.0 molar-% at 0 (left) to 2.5 mol-% at
types of leucosomes; some are dominated by very large
1 (Cartwright and Buick, 2000). Solidus (purple line)
garnet grains that are up to 15 cm in diameter (Fig. 3a),
represents first appearance of melt. Field labelled 1 is
whereas others are coarse-grained and garnet poor
H2O-saturated, and solidus is at ,7008C. First melt pro-
(Fig. 3b). The garnet-bearing leucosomes have been
duced here (field 2) are H2O-saturated, liberating free
interpreted as in situ melts, with the garnet representing
H2O. Free H2O leaves system along green univariant
the residuum, and the garnet-poor leucosomes are
lines. Under H2O-undersaturated conditions, temperature
interpreted as segregated and migrating melt. Both types
of solidus jumps to ca. 7758C. As rock is made pro-
of leucosome are generally parallel to a composite, layer-
gressively more anhydrous towards right of pseudosec-
parallel solid-state foliation, although locally discordant
tion, temperature of solidus progressively increases
leucosomes form a network. In places, the large garnet
expected for any rock of a generally pelitic composition. grains also form augen structures with the leucosome
The bulk composition has been kept the same, but the preserved in the pressure shadows (Fig. 3c). It is
molar proportion of H2O varies from 7 molar-% on likely that the melt originally formed more extensive
the left (labelled as 0 in Fig. 2) to 2.5 molar-% on the layer-parallel leucosomes, before the extraction of the
right (labelled as 1 in Fig. 2). At 7 molar-% H2O, the melt that was locally squeezed out from between the
rock begins to melt below 700uC. Free H2O leaves the foliation planes, leading to volume loss in the rock. The
system at *6 mol-% H2O, with the temperature of garnet-poor leucosomes are likely part of a more
the solidus rising to *775uC. As H2O content decreases, extensive melt migration network. The removal of the
the solidus continually rises to nearly 850uC at melt from around the garnet was important for the
2.5 molar-% H2O. textural and mineralogical evolution of the rock, as it
Upon the initiation of melting, small isolated patches prevented the rehydration reactions that would
and tubes of melt will form at the junctions and along have caused the garnet to breakdown in retrogression
grain boundaries of the reactant phases (Mehnert et al., (White et al., 2004).
1973). Under compressive stresses, the melt will tend to Since rock packages are chemically heterogeneous at a
form patches and tubes that are aligned perpendicular variety of scales, the process of migmatisation is corre-
to the least compressive stress (Arzi, 1978; van der spondingly heterogeneous throughout a rock package
Molen and Paterson, 1979). Less than 2% melt volume is and across a metamorphic terrane. This heterogeneity is
required in felsic systems to allow the melt pockets particularly evident within metasedimentary packages,
to join and for the rock to become permeable where even subtle compositional variation between
(Vigneresse et al., 1996; Lupulescu and Watson, 1999). different lithologies permits some layers to melt at lower
The melt can then begin to segregate and move through temperatures than adjacent layers. The melting of more
the solid framework, beginning the formation of fertile lithologies within a rock package may liberate
migmatites. H2O and other volatile phases, which may then migrate
The formation and migration of hydrous melt out of a into less fertile lithologies nearby and induce partial
rock body is vital for the preservation of anhydrous melting there, further propagating melt production
granulite facies mineral assemblages. Because H2O is within the larger system.
Applied Earth Science (Trans. Inst. Min. Metall. B) 2015 VOL 124 NO 3 149
a stromatic metatexite migmatite with large garnet porphyroblasts (red lensoidal crystals) in in situ leucosomes (lighter
coloured layers); b migmatite following melt extraction, with in situ leucosome (lighter coloured material) only preserved beside
large garnet porphyroblasts (large red grains). There is oblique, garnet-poor leucosome in lower part of photograph, which has
segregated from its source and is interpreted as in-source leucosome; c detail of migmatite from which most of melt has been
extracted. Remnants of stromatic leucosomes are preserved in pressure shadows next to large garnet porphyroblasts, which
are starting to fragment, resulting in augen structures
3 Migmatites from Round Hill, near Broken Hill, New South Wales
Components of migmatite typically of a more refractory composition, such as calc-

silicate or quartzite. The term mesosome may also be
As migmatitic rocks are heterogeneous by definition, applied to describe intermediate-coloured relatively
they consist of a number of components that represent unreacted protolith.
the various stages of the partial melting process. Resistors are refractory phases in the palaeosome that
Protolith is the name given to the rock that has are particularly resistant to partial melting or micro-
undergone partial melting (Johannes, 1985). Strictly structural change (Read, 1957; Sawyer, 2008a). These
speaking, this term is applied to the pre-anatectic phases are typically competent and can include quart-
lithology, because once melting starts, the rock will be zite, calc-silicate, and metamafic rocks (as opposed to
converted to neosome (Sawyer, 2008a; 2008b). However, the more melt-fertile lithologies, such as pelite, which
for the sake of convenience, a migmatite that was contain hydrous minerals; Sawyer, 2008b).
derived by partial melting of a pelite can be described as Palaeosome and resistors can be important for
having a pelitic protolith. unravelling migmatites, as they may represent the original
The palaeosome is the part of a migmatitic rock that layering (e.g. bedding), thereby forming marker units.
was not affected by partial melting (Fig. 4) and which Furthermore, because they are more competent than the
preserves structures (e.g. foliations, folds and layering) host, they can provide information about strain.
that are older than the anatectic event (Olsen, 1985; For example, layers can be boudinaged or folded, or
Sawyer, 2008a). The microstructure of the rock may be blocks can be rotated to form asymmetric features that
unchanged, or there could be minor changes in the size, are often analogous to porphyroclasts in mylonites,
form, and orientation of the grains that would reflect e.g. s- and d-porphyroclasts, and mica fish (see Passchier
subsolidus metamorphic processes. The palaeosome is and Trouw, 2005 for microscopic examples).
The leucosome will not necessarily have the composition

of an anatectic melt, as fractional crystallisation and
separation of the fractionated melt may have occurred.
Leucosomes can be subdivided, depending on whether
they are in situ, or they have segregated and migrated
from their source (see the section on ‘Leucosome
segregation and migration features’).
Margins of leucosomes
The margins of leucosomes can provide important
information about the rheological contrasts between the
leucosomes and their host, and the timing of leucosome
development relative to anatexis and deformation. These
observations can help the geologist to better understand
the development of the migmatitic rock.
There are several types of sharp margins, with each
4 Migmatite from Kirton Point, Port Lincoln, South Australia, type providing information about the rheology of the
showing two components of neosome. Paler material rep-
rock. The context of the injected melt is important in

resents leucosome (A), the dark selvedge composed of mafic understanding the process:
minerals; (B) beside the leucosome would be melanosome (i) a sharp, stepped margin following the edges of
(type of residuum), and mesocratic host would be palaeo- individual crystals suggests that the host was not
some (C). Rock is part of Donington Suite, which underwent completely crystallised, and there was a film of
partial melting during ca. 1730–1690 Ma Kimban Orogeny silicate melt at the grain boundaries. This film
permitted the rock to part along the boundaries
The neosome comprises the parts of a migmatite that of individual grains (Fig. 5a) rather than frac-
are newly formed by, or reconstituted by, partial melting turing across the grains
of the protolith (Sawyer, 2008a; 2008b). The neosome (ii) in stromatic (or layered) metatexitic gneisses,
essentially includes the solid (i.e. residuum and mela- leucosomes with a sharp, straight margin parallel
nosome) and liquid (i.e. leucosome) components of the to the gneissosity may indicate that the leuco-
melt-producing metamorphic reaction (see the section some formed relatively early in the anatectic
on ‘Formation of migmatite’). The parts of the neosome event. The earlier leucosomes would have been
may, or may not, have undergone segregation into sep- subjected to more strain, which resulted in
arate melt and solid fractions. For example, at lower modification and transposition of the margins,
degrees of partial melting, the rock may not be relative to the later leucosomes (Fig. 5a)
permeable and the melt may occur along the grain (iii) a sharp, straight margin can also indicate that the
margins. There are several parts to the neosome (Fig. 4), host was relatively solid, with the grains suffi-
including the residuum, the melanosome and leucosome. ciently bound that the injection of the vein or
The residuum is the part of the neosome that rep- dyke resulted in brittle fracturing and cracking of
resents the solid fraction left after partial melting and the crystals (Fig. 5b). There will not necessarily
extraction of some, or all, of the melt fraction. There be any solid-state strain associated with the
may be microstructural evidence that partial melting has intrusive sheets
occurred; see Holness (2008) and Sawyer (2008a) for a (iv) alternatively, parallel veins or dykes with sharp,
review of the petrographic evidence for partial melting. straight margins may be originally oblique or
The melanosome is the darker coloured part of the subparallel leucosomes, veins or dykes that were
neosome, which is enriched in dark minerals such as biotite, subjected to a later solid-state structural over-
garnet, cordierite, orthopyroxene, clinopyroxene, amphi- print that led to their transposition (Passchier
bole, and olivine (Fig. 4). The melanosome represents the et al., 1990). It should be possible to distinguish
solid fraction that is left after some, or all, of the melt this, as the leucosome, vein or dyke will tend to
fraction has been extracted (Sawyer, 2008a). There may be have a solid-state foliation that is parallel to a
microstructural evidence for partial melting. similar foliation in the host. In some cases, the
The definitions for residuum and melanosome appear partitioning of simple shear into a leucosome,
to be quite similar, which can be confusing. The easiest vein or dyke may cause an angular relationship
way to think of them is that they are both the solid with the foliation in the host.
product of the melt reaction, with residuum a more Some margins of leucosomes can be described as dif-
general term, whereas melanosome specifically refers to fuse or gradational where it is difficult to pinpoint
residuum that is composed of dark minerals. precisely the edge of the leucosome (Fig. 5c). This may
The leucosome is the lighter coloured part of the occur at the boundary of an in situ leucosome where
neosome in a migmatite that is predominantly composed there is a transition from a leucosome to the residuum,
of feldspar and quartz (Fig. 4), although it can also via a zone where the melt fills the grain boundaries in
contain other minerals such as garnet, cordierite or the residuum.
tourmaline (Sawyer, 2008a). The leucosome is derived Leucosomes can be considered to have feathery
from segregated partial melt and represents the crystal- margins when there are a series of fine apophyses that can
lised liquid (i.e. melt) of the melt-producing reaction. be traced into the foliation in the host leading to a
Consequently, the leucosomes may contain microstruc- feathered or comb-like geometry (Fig. 5d). This feature
tures that indicate crystallisation from a melt or magma. suggests local derivation, with the melt moving along the
5 a Close-up of shear band from eastern Musgrave Province (South Australia) illustrating difference between early, straight-edged
leucosomes and parallel, later, coarser grained leucosomes that have sharp stepped margins that follow grain boundaries. It is
possible to distinguish two generations of parallel leucosomes by cross-cutting relations. Straight-edged leucosomes are cut and
deflected by oblique shear band, which is filled with leucosome that is texturally continuous with coarser grained leucosome
(suggesting they are the same generation). This is close-up of part of Fig. 25; b example from east Albany-Fraser Orogen, Western
Australia of sharp, straight-edged dyke intruding solid rock. Rapakivi granite is cut by pale microgranite dyke, which has sharp,
margins that cut across large mantled feldspar grains in host. This suggests that host rock was completely crystallised and able to
crack when microgranite magma intruded; c garnet-bearing leucosome from Cape Carnot, Port Lincoln, South Australia, showing
diffuse/gradational margin; d small melt patches (light-coloured) from Wanna, Port Lincoln, South Australia, showing feathery
margins where fine apophyses can be traced along foliation into darker host rock
Leucosome segregation and migration features

Leucosomes can be subdivided based on the degree of
separation from the site of melt production
(Sawyer, 2008b). The distance travelled from the source
can affect the compositional and textural evolution of
the leucosome, causing it to progress from
something that may be a minimum melt composition to
something that has evolved and resembles a granite
sensu stricto. These groups also represent a progression
in scale from melt that is locally segregating at the
grain-scale up to large-scale magma, or melt, migration
6 Schematic representation of migmatitic layer showing (Fig. 6):
difference between in situ, in-source and leucocratic vein (i) in situ leucosomes are crystallised anatectic melt,
or dykes (modified from Sawyer, 2008b) or part of a melt, that have segregated from the
residuum but remained at the site where the
melting occurred
foliation planes and into the low-pressure site, (ii) in-source leucosomes are crystallised anatectic
represented by a shear band or other dilational site. melt, or part of a melt, which has migrated away
Feathery margins can be a relatively small-scale feature from the place where it formed but is still within
with the apophyses occurring at the millimetre scale. the confines of its source layer
(iii) leucocratic veins or dykes are crystallised ana- Rheology

tectic melt, or part of a melt, which has migrated Rheology is the study of the deformation or flow of a
out of its source layer and been injected into material and describes the relationship between stress (the
another rock that is still in the region affected by forces acting upon a body) and strain (the change in
the anatectic event volume, shape or position of that body) for a given rock
(iv) granite dykes, sills, and plutons are crystallised felsic (Stüwe, 2007). The rheology of a migmatitic rock will vary
melt that has migrated out of its source region depending on several factors, including the proportion of
completely and been injected into host rocks of lower melt, the distribution of melt, and any external stresses.
metamorphic grade or non-metamorphosed rocks. Magmas are non-Newtonian fluids that can have
As the last three types of migration features have moved, or complex rheological behaviour. Newtonian fluids are
started to move from their source, there is the likelihood substances that have a linear, proportional relationship
that they will be transgressive and cut across in situ or between differential stress and shear strain rate. This
earlier migmatitic structures as they migrate. Furthermore, means that the viscosity of the fluid will not change,
as the melt moves from the source, there will be a change in regardless of any external stresses. In contrast, non-
the melt fraction of the host rock—from an abundance of Newtonian fluids are characterised by a non-linear
melt in the anatectic zone to no melt in lower metamorphic relationship between differential stress and the shear
grade or non-metamorphosed rocks. This change will strain rate. As a result, the viscosity of the fluid will
affect the interaction between the migrating melt and the change when external stresses are applied. For example,
host, which can be reflected in the margins of the trans- shear thickening can occur when the viscosity increases
ferring material. For example: as strain rate increases due to the mix of fluid and rigid
(i) the in situ and in-source leucosomes will likely particles. At low velocities, the fluid dominates the
have a feathery or diffuse/gradation margin with behaviour as it is able to flow and fill the spaces between
the neosome as the host is undergoing partial the particles because they are not moving fast. At higher
melting and will contain a component of melt. If velocities, the fluid cannot keep up with the particle
the residuum has a relatively low melt fraction, movement and it is unable to fill the spaces between
the leucosomes may have sharp and stepped them. This causes the particles to rub against each other,
margins that follow the boundaries of the crys- possibly causing the magma to crack if it is stressed or
tals in the host. Alternatively, the leucosomes flows too quickly.
may have sharp, relatively straight margins with Shear thickening can also affect migmatites. When the
resistor units melt is pervasively distributed at the grain scale, an increase
(ii) the margin of a leucocratic vein or dyke will vary in the strain rate can result in the development of dilational
from sharp and stepped to sharp and straight, features, such as small-scale shear zones. The melt will flow
depending on the proportion of melt in the host into these dilational features as they will have lower
(iii) the felsic melt that has been injected into a lower pressure relative to the main rock mass. The rheology of a
grade or non-metamorphosed rock to form a melting rock will be further examined in the section on
dyke or sill will tend to have a sharp, straight ‘Transition from metatexite to diatexite’.
intrusive contact, since the host will be solid and
able to crack (e.g. Figure 5b). If the temperature
Magmatic versus submagmatic flow
contrast is great enough, there may also be a
chilled margin to the intruding body. The terms magmatic flow and submagmatic flow can
be applied to the different types of migmatites
described below.
Magmatic flow is inferred from the preferred orien-
Some additional concepts tation (magmatic foliation) of minerals in a rock derived
Anatexis is a general term used to describe melting or from a magma, which was acquired during flow when the
partial melting of rocks (Sawyer, 2008a). It does not crystals were suspended in the magma and free to rotate
specifically refer to the degree of partial melting and can with very little or no interaction with nearby crystals.
therefore be applied to all stages, from incipient Experiments and models have indicated that this will
partial melting to complete fusion. Crustal anatexis is occur when the magma contained v55% crystals (Vig-
generally accompanied by deformation, which will neresse et al., 1996). Magmatic flow is defined by the
facilitate a series of physical processes, including melt alignment of euhedral to subhedral elongate to tabular
segregation, melt migration, fractional crystallisation, crystals or mafic schlieren within a massive groundmass.
and magma flow. Submagmatic flow is inferred from the preferred
Segregation is the process whereby the anatectic melt orientation of minerals in a rock that was acquired when
phase is separated from the residuum in the neosome. the magma had sufficient crystals that interactions
Note, however, that not all neosomes have undergone commonly occurred between them as they rotated in the
segregation of the melt. flow (submagmatic foliation). Experiments and models
The liquid component of the neosome can also be have indicated that this will occur when the magma
subdivided into melt and magma. The melt is a silicate contained w55% crystals (Vigneresse et al., 1996).
liquid without solids. Magma is a silicate liquid that Indicators of submagmatic flow include the tiling
contains suspended solids. The solid component may be (or imbrication) of crystals and the healing of cracked
crystals that either crystallised from the melt (i.e. liqui- crystals by a silicate liquid. Unfortunately, as the inter-
dus phases), be the solid product of the melt reaction action between crystals increases, the textural evidence
(i.e. known as peritectic products), or represent refrac- for magmatic and submagmatic flow can be lost.
tory crystals (e.g. zircon cores). Alternatively, the solids The concept of magmatic and submagmatic state
may be rock fragments from the source area. foliations does not just apply to partially melted rocks.
It is also an important concept to keep in mind when Paterson (1979) experimentally determined that the
looking at syn-tectonic/syn-kinematic granites. The sys- strength of a partially melted granite will decrease
tematic alignment and tiling of euhedral to subhedral gradually from 5 volume-% melt, with a rapid decrease
phenocrysts in an igneous body indicates that stresses at 24% melt, before a transition from framework-
were being applied to the body, while it was still in the controlled flow to suspension-like behaviour at
magmatic state, with the minerals able to freely rotate 30–35 volume-% melt. The lower limit to this range is
and align. During pure shear, the phenocrysts can be similar to the 0.07 melt fraction where it is predicted that
aligned perpendicular to the shortening direction. the films at the grain boundaries will interconnect (called
During simple shear, the phenocrysts can be aligned the melt connectivity transition), allowing grain bound-
either parallel to the C or the S plane. Alternatively, the ary sliding, and rotation and movement of the minerals
phenocrysts can be aligned parallel to the boundaries of (Rosenberg and Handy, 2005). These authors also noted
the conduit the magma is flowing through. As well as a solid-to-liquid transition at a melt fraction of 0.4, which
providing important structural information (e.g. foli- corresponds to the breakdown of the solid crystal fra-
ation, lineation, kinematic indicators, strain markers, mework. Grain interactions will continue after the
stress directions), these are excellent targets for breakdown of the framework, and it is not until a melt
geochronology, as the age of the igneous rock will also fraction of 0.6 that the solid components will be sus-
constrain the age of the deformation. pended in the melt and no longer interacting with each
other. Ultimately, it appears that the interconnection of
melt, rather than the break-up of the solid crystal fra-

Types of migmatites mework, results in the most significant drop in strength
Migmatites have generally been divided into metatexites and may lead to the onset of disaggregation of the
and diatexites. A metatexite is a migmatite that is het- partially melted rock (Rosenberg and Handy, 2005).
erogeneous at outcrop scale, with coherent, pre-partial Syn-anatectic strain is an important factor in
melting structures preserved in the palaeosome the transition from a metatexite to a diatexite
(Sawyer, 2008a). In metatexites, the neosome is generally (Fig. 7; Sawyer, 2008a), as higher strain or increased
segregated into leucosome and melanosome. Essentially, strain rates can lead to the disaggregation of the crystal
a metatexite is a migmatitic rock that preserves structural framework at relatively lower melt fractions (Sawyer,
integrity, whether this is primary layering in the protolith, 1996; Vigneresse and Tikoff, 1999). The difference in
earlier structural elements, or syn-anatectic structures, syn-anatectic strain also tends to be an important factor
such as layering. It is important to note that the stresses in the second order subdivisions (Fig. 7). For example,
associated with deformation can precede partial melt. high shear strain and the resulting flow will result in
Consequently, it is not unusual to develop an early foli- more efficient disaggregation and better alignment of
ation that can then be exploited once the rock starts to minerals, schollen (rafts), and schlieren in diatexites.
melt and would be considered to be pre-anatexis. This can often lead to the development of compositional
A diatexite is a migmatite dominated by pervasive layering or foliations (see Fig. 8), which can be truncated
neosome (Sawyer, 2008a). Pre-partial melting structures at a low angle during shear flow (Fig. 9).
are absent from the neosome and commonly replaced by
syn-anatectic flow structures or isotropic neosome. The Metatexites
neosome is variable in appearance, as it typically Metatexites are dominated by the palaeosome and/or
includes leucosome and melanosome in varying pro- residuum, retain evidence of pre-partial melt structures,
portions. The palaeosome can occur as rafts or schollen, and have coherency of the residuum. Due to the struc-
or it may be absent. tural integrity of these rocks, they are often layered or
banded and can appear gneissic. Metatexites can be
Transition from metatexite to diatexite further subdivided (Fig. 7b), depending on the sites
The difference between a metatexite and a diatexite is where the melt has ponded to form the leucosomes.
generally a function of the melt fraction and the amount of
syn-anatectic strain (Fig. 7; Sawyer, 2008a). The pro- Patch metatexite migmatites
portion of melt that corresponds to the breakdown of the Patch metatexite migmatites form when melting occurs
structural framework of the rock and the transition from a at discrete sites to produce small, scattered patches of
metatexite to a diatexite are difficult to quantify. Some non-foliated, in situ neosome (Sawyer, 1991). The neo-
models (Fig. 7a) treat the solid crystals as uniform, rigid somes are generally round or oval in shape (Fig. 10a)
spheres (URS) and suggest that the breakdown of the solid and are characteristic of the incipient stages of partial
crystal network will occur at a melt fraction of 0.26 melting, although they can be difficult to distinguish in
(the URS model, e.g. Sawyer, 2008a). This is similar to the some lithologies (e.g. a coarse-grained, non-foliated
rheological critical melt percentage of 20+ 10 volume-%, granite). However, as the melt fraction increases, the
where the greatest change in the viscosity was predicted to neosome grows, and the patches can coalesce to form
occur (Arzi, 1978), and the melt escape threshold of 20–25% irregular, lobate shapes (Fig. 10b; Sawyer, 2008a).
melt predicted for the onset of the migration of melt and
residuum over larger distances (Vigneresse et al., 1996). Dilation metatexite migmatites
More realistic models that treat the crystals as solid In dilation metatexite migmatites, the leucosome occurs
particles that are non-uniform in size and shape (the within dilatant (i.e. relatively low pressure) sites, such as
non-uniform particle (NUP) model) have determined boudin necks, pressure shadows, shear bands, or frac-
that the transition from a metatexite to diatexite can tures in more competent layers (Figs. 11 and 12; Collins
occur over a wider melt fraction range of 0.16–0.6 and Sawyer, 1996; Oliver and Barr, 1997; Sawyer,
(Fig. 7a; Renner et al., 2000). Van der Molen and 2008a). The dilational sites are generally of limited
a first-order division of migmatites into metatexite and diatexite migmatites is a function of fraction of melt and properties of
solid grains in partially melted rock. This will depend on the nature of the crystal model, which include URS model and NUP
model. See text for description; b second-order morphologies of metatexite and diatexite migmatites on plot of syn-anatectic
strain versus melt fraction. Diagram is shaded for URS model, but vertical dashed lines indicate where boundaries are for
transitional zone in NUP model
7 Classification scheme for migmatitic rocks, modified from Sawyer (2008a)
8 Schlieric diatexite migmatite from Shaw Granitoid

Complex, Pilbara Craton, Western Australia, showing
irregular, mesocratic to melanocratic layers and lenses
that are defined by trains of mafic minerals. Layering is
locally defined by leucocratic lenses of coarse-grained 9 Layering, defined by subparallel leucosomes in migmatite
feldspar and quartz (in centre of photo), which likely from the Birksgate Complex in eastern Musgrave
represents attenuated leucosomes. Layering would Province, South Australia, which are locally cut at low
be result of magmatic flow, which has drawn out and angle by melt-filled shear bands that developed during
disaggregated neosome shear flow
10 a Patch metatexite migmatite from Wanna, Port Lincoln, South Australia, showing irregular, discontinuous leucosomes in
migmatised orthogneisses of the Donington Suite. Leucosomes have feathered margins, with fine apophyses that can be
traced into darker host, suggesting local derivation; b patch metatexite migmatite from Kirton Point, Port Lincoln,
South Australia showing irregular, generally discontinuous leucosomes in migmatised orthogneiss of the Donington Suite.
Well-developed melanosome selvedges indicate in situ partial melting, and relatively large size of leucosomes and lobate
boundaries suggest relatively high melt fraction and coalescence of melt patches. Locally, they have connected to form
stroma that would have accommodated melt migration. There are also some discrete stromata with well-developed
melanocratic selvedges (right of centre). Development of parallel leucosomes suggests that there were external stresses
extent and may be restricted to particular, relatively Net-textured or net-structured metatexite migmatites
competent layers. The dilational structures imply that Net-structured metatexite migmatites contain leuco-
there is a competency contrast in the host rock, with the somes that occur in two or more systematic sets,
more competent layers often composed of less fertile intersecting to form a net-like pattern that outlines
(e.g. greywacke in a pelite) or resistant or refractory polygonal blocks of the palaeosome and/or residuum
lithologies (e.g. calc-silicate, quartzite, metamafic rocks, (Fig. 13; Sawyer, 1991; 2008a; Brown, 1994). The two
pegmatite). The competent layers could also be newly sets of leucosomes may be texturally continuous with
formed residual layers, which are rich in competent no obvious cross-cutting relations, or cross-cutting
minerals such as garnet or pyroxene (Sawyer, 2008a). relations may be present. However, the latter situation
does not necessarily mean that there were two phases of
partial melting, as they may be progressive and represent
different stages of the same anatectic event.
Geometrically, the sets may provide evidence for
syn-anatectic deformation. For example, there may be
leucosomes that are parallel to the compositional
layering or foliation and extensional shear bands that
also host the melt (Sawyer, 1991; Brown, 1994). This
relationship can be distinguished by the deflection of the
compositional layering or foliation into the shear bands
and may indicate that the bulk rock underwent short-
ening perpendicular to the layering and layer-parallel
extension.
A range of progressive textural development can be
a Melt can pond in the necks of boudins that develop in recognised in the net-structured migmatites (Sawyer,
more competent layers, such as palaeosome resistors or 2008a):
melanosome; b melt can pond in extensional shear (i) at the early stages of partial melting, the leuco-
bands, with both synthetic and antithetic shears shown in some is narrow with a high aspect ratio and
figure; c melt ponding in asymmetrical boudins; d melt bordered by melanosome. The leucosomes
can pond in several sites that are associated with parallel outline polygonal blocks that have centres
folds. Melt can migrate into fold hinges to form layer- composed of palaeosome (i.e. they have not
parallel leucosomes. Melt can also form oblique leuco- undergone melting)
somes that are either parallel to axial plane of fold (if layer
(ii) with further partial melting, the rocks become
is less competent) or in extensional cracks that are
perpendicular to folded layer (if layer is more competent);
progressively dominated by neosome. If the melt
e melt can flow into shear zones that can often develop in is extracted from the blocks, they will be com-
hinge zones during folding of migmatitic rocks. Figure is posed of melanosome. If melt is injected into the
modified from Sawyer (2008a) blocks from the leucosomes, the palaeosome may
11 Diagram showing some of structural sites that can host be mesocratic
leucosome in dilation metatexite migmatites. Solid areas (iii) with increased melt fraction, the net-structured
represent leucosome, and dashed lines represent traces migmatites will disaggregate and progress
of bedding or solid-state foliation towards a schollen diatexite migmatite.
12 a Dilation metatexite migmatite from the Shaw Granitoid Complex, Pilbara Craton, Western Australia, where relatively
competent mafic layer within tonalitic orthogneisses has been boudinaged, with melt ponding in boudin neck. There also
appears to be more melt accumulated on right side of mafic layer (east side), relative to left (west side). Melt is represented
by coarser grained, lighter-coloured material and to the right side contains thin, discontinuous screens of mafic rock that
can be seen to be deflected upwards into boudin neck. These relations suggest that this is also asymmetric vein cluster
that indicates way-up was to the west during partial melting; b dilation metatexite migmatite from the Shaw Granitoid
Complex, Pilbara Craton, Western Australia, where more mafic (darker, finer grained) layer in tonalitic orthogneiss has
sheared parallel to axial plane of small-scale ‘z’ fold. At smaller scale, fine, discontinuous leucosomes are cutting some of
the thin, more competent, mafic layers. At first glance, mafic layers appear to be boudinaged, but on closer inspection,
leucosomes are not divergent, but instead they are parallel to axial plane of fold; c dilation metatexite migmatite from
Fishery Bay, Gawler Craton, South Australia, with melt ponding to form series of vertical sheets in sheared axial planes of
upright folds. Folded layering also contains layer-parallel leucosomes that are i) locally deflected and truncated by shears
and ii) texturally continuous with melt in shears. This indicates progression from earlier, layer-parallel leucosome devel-
opment to folding with flow most likely occurring from later leucosomes into axial plane parallel shears; d dilation
metatexite migmatite from East Albany Fraser Orogen, Western Australia, with melt ponding in extensional shear band that
is at moderate angle to foliation and parallel leucosomes. Leucosome in shear band has very irregular, ‘feathered’ edge
and is texturally continuous with layer-parallel leucosomes. This suggests that melt is moving along foliation planes and
into low-pressure tear represented by shear band
13 a Net-structured metatexite migmatite from Shaw Granitoid Complex, Pilbara Craton, Western Australia, showing two
systematic sets of leucosome. Relatively earlier, foliation parallel, sharp-edged leucosomes (parallel to hammer head) are
deflected by shear bands with diffuse margins (subparallel to hammer handle); b net-structured metatexite migmatite from
Red Banks, southern Eyre Peninsula, South Australia, showing two systematic sets of leucosome. Relatively earlier,
foliation parallel, sharp-edged leucosomes (parallel to pen) are deflected at moderate angle by shear bands with diffuse
margins. Two sets are texturally continuous, indicating their development was progressive
Stromatic metatexite migmatites Diatexites

Stromatic metatexite migmatites are characterised by In contrast to the metatexites, diatexites are dominated
numerous thin and laterally continuous leucosome layers by the melt fraction, with few preserved pre-partial
that are oriented parallel to the major plane of anisotropy melting structures. This reflects the progressive dis-
in the palaeosome, typically compositional layering or a aggregation and dispersal of the palaeosome and/or
solid-state foliation (Figs. 10b and 14). The individual residuum within the melt phase, which can occur during
leucosome layers can have melanosome on both sides, buoyant flow or regional deformation. Diatexites can
only on one side, or they can be absent (Sawyer, 2008a). also be subdivided, depending on the geometry of the
Several models have been proposed for the origin of palaeosome and/or residuum or its absence.
stromatic migmatites. Leucosomes with melanosome sel-
vedges have been interpreted to represent in situ partial
Nebulite diatexite migmatites
melts (e.g. Sawyer, 1991; Oliver and Barr, 1997), although
this may not necessarily always be the case, as the mela- Nebulite diatexite migmatites have neosome that is
nosome selvedges do not always have suitable residual diffuse and difficult to differentiate from the palaeo-
compositions. Instead, these selvedges may represent some. This can happen when patch metatexite migma-
reaction zones between the melt and the host rocks (Jung tites reach higher melt fractions in the relative absence of
et al., 1999). Alternatively, stromatic metatexites have any external stresses (Fig. 7; Sawyer, 2008a). External
also been interpreted as injection migmatites, where there stress will lead to either structural control on the dis-
is lit par lit, literally bed by bed intrusion of a migrating tribution of the melt phase resulting in systematically
melt (e.g. Lucas and St-Onge, 1995). arranged leucosomes and a metatexite or disruption of
the migmatite to form a diatexite.
Schollen diatexite migmatite

Schollen or raft-structured diatexite migmatites are
characterised by the presence of raft-like blocks that are
isolated within leucocratic to mesocratic neosome. The
blocks can be either remnants of palaeosome, resistant
lithologies, or melanosome (Fig. 15). Schollen diatexite
migmatites commonly occur at the transition from
metatexite to diatexite and represent the disaggregation
of the palaeosome, melanosome and resister lithologies
into discrete blocks, thereby disrupting the structural
continuity (Fig. 15; Solar and Brown, 2001). Initially,
the blocks are large, have high aspect ratios (reflecting
the compositional layering and foliation), and rounded
ends. At this stage, the blocks have typically not
undergone rotation. With continued flow and
14 Stromatic metatexite migmatite from Birksgate Complex, disaggregation, there is a progressive decrease in the size,
eastern Musgrave Province, South Australia. Rock is aspect ratio, and number of the rafts, which typically
characterised by parallel gneissosity and leucosomes, become more rounded, rotated and dispersed through-
with the later forming continuous, coarse- to medium- out the leucocratic neosome. The leucocratic neosome
grained layers may have a flow foliation defined by the alignment of
platy or tabular minerals.
15 a Schollen diatexite migmatite from Shaw Granitoid Complex, Pilbara Craton, Western Australia, containing large blocks of
melanosome. Tips of some of the blocks are being attenuated to form schlieren; b schollen diatexite migmatite from the
same outcrop, which has undergone further disaggregation, resulting in scattered, smaller blocks of melanosome in
groundmass of heterogeneous, leucocratic to mesocratic neosome. Rock also contains relatively common schlieren,
defined by trains of mafic minerals. There is broad layering, extending from top right to bottom left, defined by aligned
schollen and schlieren, as well as compositional layering in neosome, which would likely be flow foliation; c schollen
diatexite migmatite from Birksgate Complex, eastern Musgrave Province, South Australia. Residuum and palaeosome form
common isolated blocks that have been rotated, so that there is no structural coherency between them. Furthermore,
blocks do not appear to be aligned, suggesting there was little to no large-scale magmatic flow at this outcrop,
with movement occurring at local scale
16 a Diatexite migmatite from Shaw Granitoid Complex, Pilbara Craton, Western Australia, showing dominance of neosome.
Rare pre-melting textures are preserved as layering on right side of photo, but rock is dominated by broad-scale, sinuous,
magmatic or submagmatic state flow foliation and layering. Flow foliation is frequently truncated by magmatic flow or
magma injection. The latter is best developed on left-hand side of photo, where a sheet of more evolved, blue-grey, flow-
banded granitic material (likely monzogranite) has intruded diatexite and truncated layering; b diatexite migmatite from
Birksgate Gneiss, Musgrave Province, South Australia that is well on the way to becoming granite. Outcrop is generally
homogeneous with granitic textures, although origin as migmatitic rock is suggested by rare mafic schlieren and thin,
leucocratic, coarse-grained layers that likely represent late-stage leucosomes. Equally important is local context, as there
is spectrum of migmatitic textures, ranging from metatexite, to diatexite, to dirty granite in this area
Schlieric diatexite migmatite microstructure. Diatexite migmatites are gradational from

Schlieric diatexite migmatites are characterised by well- schollen and schlieric migmatites through an increase in
developed, flow-induced structures, called schlieren, melt fraction (i.e. neosome to palaeosome ratio) and from
which are defined by thin layers or trains of aligned platy, nebulitic migmatites through an increase in syn-anatectic
tabular, or prismatic minerals (Fig. 8). The schlieren are strain (Fig. 7; Sawyer, 2008a).
most commonly defined by biotite but can also include Most diatexites contain leucocratic veins and
plagioclase, sillimanite, orthopyroxene, or amphibole and patches, composed of quartz, K-feldspar and plagio-
likely represent entrained and attenuated neosome, most clase, which post-date the magmatic foliations and flow
commonly the residuum (Sawyer, 2008a). structures (Sawyer, 2008a). These late leucosomes are
Schlieren tend to be orientated parallel to each other interpreted to represent residual melt preserved within
(Fig. 8), although in some outcrops the foliation, com- the pore space of the crystallising diatexite. As the
positional layering, or schlieren can be truncated at a crystal fraction increases and the crystals start to interact
low angle by similar looking but often more leucocratic and form a relatively solid framework, heterogeneous
(i.e. melt-rich) layers (Fig. 9). These features are deformation can lead to the differential stresses
evidence of syn-anatectic shear zones or flow dis- within the framework. This will result in dilatant sites
continuities in the partial melted rock, with greatest that will have lower pressure, relative to the main body
strain partitioned into the zones with the highest melt of the rock, thereby drawing in the residual, interstitial
fraction (Sawyer, 2008a). Folding can also occur during melt.
syn-magmatic flow, resulting in rootless, isoclinal folds, At the extreme end of this spectrum, diatexites are
and sheath-like geometries. transitional to granites. Such diatexites can be relatively
Schollen of the palaeosome, resistor lithologies, or homogeneous with granitic textures (such as interlock-
melanosome may also be present, but they are less ing euhedral to subhedral crystals). However, the pre-
abundant than in schollen diatexite migmatites. The sence of schollen and mafic schlieren, thin, leucocratic,
transition from schollen migmatites to schlieric migma- coarse-grained layers that likely represent late-stage
tites represents an increase in the melt fraction or neo- melts, and xenocrysts rather indicate its origin as a
some to palaeosome ratio. Strain generally has little migmatitic rock (Fig. 16b). Whereas many granites
effect on this transition, although the increased melt are derived from partial melting of the crust
fraction will reduce the viscosity of the system, leading (e.g. Brown, 2013), diatexite migmatites represent an
to a greater tendency to flow and attenuation of the intermediate stage of the process. Subsequent migration
residuum. and flow of the magma is necessary for the continued
disaggregation of the residual features, leading to hom-
Diatexite migmatite ogenisation of the magma and the development of a
Diatexite migmatites are dominated by neosome, and granite sensu stricto (e.g. Brown 2013; Clemens, 2013).
relicts of palaeosome and resistors form scattered schollen, Diatexites can also provide information about the
which may contain pre-melting structures, and schlieren deformation at the time of partial melting. If the flow is a
(Fig. 16; Sawyer, 2008a). The neosome can contain a response to regional deformation, which is likely as strain
magmatic or submagmatic state flow foliation, defined by is often partitioned into these ductile rocks, the diatexite
the alignment of platy, tabular, or prismatic minerals, such can form melt-rich shear zones (e.g. Blumenfeld and
as mica and feldspar. There may also be a flow banding, Bouchez, 1988; Brown and Solar, 1998). In these struc-
defined by variable mineral proportions, grain size, or tures, the blocks and schlieren may have asymmetric
geometries, analogous to microstructures in mylonites,

which can be used to determine the sense of shear.
Other information that can be extracted

from migmatite
Dating of migmatites
Dating of migmatitic rocks usually focuses on con-
straining either the formation age of the metamorphosed
protolith or the timing of partial melt formation.
Any migmatitic zircon-bearing rock can be dated using
U–Pb isotopic methods. Zircon grains extracted from
migmatitic rocks typically preserve evidence for either
solid state recrystallisation or new zircon growth, and
with in situ methods like sensitive high resolution ion
microprobe (SHRIMP), it is possible to target both the
inner portion of zircon grains that are part of the pro-
tolith and metamorphic rims formed during partial

melting. In this way, zircon-bearing migmatites can be
very useful rocks to date, as they provide information on
the primary rock type, be it sedimentary or igneous, and
on the timing of metamorphism.
An example is from the Christie Domain of the Gawler
Craton, where the Mount Christie drill hole CD-1 inter-
sected magnetite gneiss and migmatitic metapelitic rocks.
A sample of garnet–cordierite migmatitic gneiss
(1562986) was taken for SHRIMP geochronology, and
the data are presented in Reid et al. (2014). The zircon
grains in this sample preserve small, oscillatory-zoned
cores surrounded by ubiquitous sector zoning interpreted 17 a Photograph of Christie Gneiss, Mulgathing Complex,
to be formed by metamorphic recrystallisation (Fig. 17). in drill core CD-1, sample 1562986; b representative
SHRIMP dating of core and rim populations from this cathode luminescence image of zircons from sample
sample reveals that the cores have a range of ages, from as 1562986, showing difference in age between metasedi-
old as ca. 2980 Ma to as young as ca. 2480 Ma. This range mentary cores and metamorphic rims
of ages suggests that the protolith is sedimentary in origin,
with a maximum depositional age of 2480 Ma. The information about the palaeo-horizontal during partial
metamorphic rims have ages ranging from 2470 to melting (Burg, 1991; Burg and Vanderhaege, 1993). The
2415 Ma, which suggest that metamorphism was not a way-up indicators are primarily driven by the buoyancy
single event but resulted in progressive stages of zircon contrast between the melt and the surrounding rock,
growth, presumably linked to different P–T conditions with the buoyant ascent of the melt resulting in asym-
(Jagodzinski et al., 2009). metric features, including asymmetric vein clusters,
As the leucosomes can form discrete markers in a cauliflower structures, and branching fractures (Fig. 18).
sequence of structural events, samples of the leucosome
material has been taken in order to place constraints on
Asymmetric vein clusters
the timing of the different structural elements.
An example is from the Albany–Fraser Orogen, Asymmetric vein clusters are collections of parallel
Western Australia, where Kirkland et al. (2011) dated leucosome veins that have accumulated and crystallised
i) leucosomes aligned parallel to the gneissosity that were beneath the lower contact of a more competent
folded into northwesterly trending isoclinal folds and layer, such as palaeosome or resistors (Fig. 18a).
ii) leucosomes that were aligned parallel to the axial The leucosomes are interpreted to have been melts
planes of the folds. Overprinting relations would indicate that ascended buoyantly—either at the grain scale via
that two events were recorded in the rocks. However, the percolation or as in-source leucosomes or leucocratic
SHRIMP zircon age data from the folded leucosomes veins—and have been trapped below the impermeable,
shows that they crystallised at 1676+ 6 Ma, whereas the more competent layer (Burg, 1991; Burg and
cross-cutting leucosomes formed at 1679+ 6 Ma Vanderhaege, 1993).
(Kirkland et al., 2011). Hence, the two different leuco- The more competent layer may be disrupted during
some generations are within geochronological uncer- deformation, boudinaged for example. In which case,
tainty, and the structures bracketed by these leucosomes the buoyant melt would start to flow through the brea-
were interpreted to have formed during sequential ched layer (Burg, 1991; Burg and Vanderhaege, 1993).
deformation during a single orogenic event. This would also result in the asymmetric distribution of
leucosomes with crystallised melt occurring below the
layer, in the boudin neck, and also possibly forming
Way-up indicators in migmatitic rocks veins that continue upwards from the neck.
Several studies have shown that migmatitic structures An example of asymmetric vein clusters and boudi-
can be used as ‘way-up’ indicators as well as provide naging (Fig. 12a) is from an outcrop on the western side
a asymmetric vein clusters, where percolating or migrating melt ponds underneath impermeable layer; b Cauliflower
structures, with white layer representing leucosome with lower density r and viscosity n that ascends into the overlying layer;
c branching fractures, where melt ponding under competent layer raises fluid pressure and results in hydraulic fracturing of
overlying layer. Figure modified from Burg and Vanderhaege (1993)
18 Diagram showing different types of syn-anatectic way-up indicator
of the Shaw Granitoid Complex, a granite dome in the granite dome, which led to tilting of the layering to its
eastern Pilbara Craton, Western Australia. At this current steep attitude (Pawley et al., 2004).
location, tonalitic orthogneiss contains a mafic layer that
is aligned parallel with the granite–greenstone contact. Cauliflower structures
The mafic layer is acting as a resistor, with melt ponding
Cauliflower structures are lobate structures that form at
on the eastern side, i.e. the side closest to the core of the
the upper margin of a layer-parallel leucosome
granite dome (on the right side of Fig. 12a). This has
(or intrusive sheet), whereas the lower margin of the
resulted in a zone of melt that contains discontinuous
leucosome is typically more planar (Fig. 18b). Cauli-
layers of the mafic rock and the host orthogneiss.
flower structures are often transgressive and can cut
The mafic layer has been boudinaged and breached,
across the foliation in the overlying layer, or the cauli-
with melt migrating into the boudin necks (one can be
flowers can deflect the foliation to form a pinch and
seen just above the notebook), and locally flowing along
swell geometry. Cauliflower structures are interpreted to
the top side of the layer. Towards the top of the
be the result of a gravitational instability, with the less
photograph, the melt is starting locally to overwhelm
dense, less viscous melt in the layer-parallel leucosome
and disaggregate the mafic sheet, resulting in a series of
ascending buoyantly and intruding into the denser
rafts within the leucosome. These observations suggest
overlying rock (Burg, 1991; Burg and Vanderhaege,
that the melt was migrating to the left (Fig. 12a).
1993). The process can be considered as small-scale
This interpretation is consistent with regional obser-
incipient diapirism.
vations from the granites and surrounding greenstones
that gneissosity and parallel leucosomes were originally
horizontal during the early stages of partial melting and Branching fractures
softening of the crust. Softening of the crust facilitated Branching fractures are melt-filled veins that project
greenstone sinking and associated development of the from one side of layer-parallel leucosome into a more
competent layer, with the veins interpreted to be the suggesting they were contemporaneous, or do
upward termination of the fracture network (Burg and they cross-cut each other, suggesting progressive
Vanderhaege, 1993). The fractures are considered to injection, or even several events?
form when the accumulation of a buoyant melt below (iv) what do the margins of the leucosomes look like?
a competent layer causes the local fluid pressure Are they sharp, diffuse, feathery?
to increase, eventually overcoming the tensile strength of These observations are important as they allow classifi-
the layer (Clemens and Mawer, 1992). This is shown in cation of the metatexite, but they also help to under-
Fig. 18c, where the increasing fluid pressure causes the stand the processes that formed them, such as the degree
Mohr circle to move to the left, eventually intersecting of strain or fraction of partial melting.
the failure envelope for the competent layer. Failure and
cracking of the overlying layer creates a pressure gra- If migmatite is diatexite
dient that draws the melt into the crack (Burg and The following observations can be made at an outcrop
Vanderhaege, 1993). of diatexite:
(i) the proportion of rafts, layers, schleiren and melt
phase can estimated. This can be as a percentage,
Working with migmatitic rocks in field or it can be qualitative, i.e. are the blocks
common, rare. However, the use of qualitative
An outcrop of migmatitic rocks should be viewed as a
terms will require consistency
whole before detailed examination is attempted, as some

(ii) what is the composition of the rafts/layers/schlie-
features are easier to see at the broad scale. Initially, the
ren? Are they composed of residuum or palaeo-
authors ascertain whether the rock is dominated by
some? What types of palaeosome are present? This
leucosome, with disrupted and dispersed palaeosome,
is important, as it may provide evidence about the
residuum, or melanosome and is therefore a diatexite, or
original rock, such as the composition of the pro-
if melt is less abundant with coherent structures between
tolith and whether it was homogeneous
the melt patches/layers and is therefore a metatexite.
(iii) what are the size, shape, and aspect ratios of the
There could also be features at the outcrop scale that
rafts, layers and schlieren?
may not be so obvious under closer inspection. These
(iv) are the rafts, layers and schlieren aligned or orien-
include broad compositional layering, represented by
ted randomly? If they are aligned, are there trun-
subtle differences in the colour of the palaeosome, or
cations, folds. Are the rafts tiled or imbricated?
variation in the proportion of leucosome, which could
These observations can provide information about
reflect differences in the composition of the protolith, i.e.
the flow regime and stresses during anatexis
bedding. An overview could also reveal folding that may
(v) are the rafts, layers and schlieren evenly dis-
not be so obvious once the eye has started to examine
tributed, or do they occur within zones? This
the detail in the exposure.
may provide information about any original
Colour can be important for distinguishing the various
compositional layering, as trains of more resist-
components. There is a good chance that the melt phase
ant lithologies may indicate bedding
will be leucocratic, and the palaeosome and residuum will
(vi) What does the melt phase look like?
be darker (i.e. mesocratic to melanocratic). The resistant
N what mineral phases are present?
lithologies will range in colour, from leucocratic for
quartzite, to melanocratic for mafic rocks. N is it relatively homogenous, or is hetero-
geneous?
If the origin of a part of the migmatite is uncertain, the
N how is it heterogeneous?
colour, texture and geometry can still be described, e.g.
‘the thick, fine-grained, grey layer is boudinaged with N is there evidence for flow, such as schlieren or
mineral alignment?
medium-grained, massive, leucocratic material ponding in
the boudin neck’. This type of first order observation N is there evidence for injected melt/magma?
This could be syn-anatectic injection of external
adequately describes the features of the rock, without
melt/magmas, which could have irregular
relying on technical jargon that can be resolved later.
diffuse margins and distinct compositions or
If migmatite is metatexite textures. Or it could be late-stage pegmatite
veins or injected magmas that have relatively
The neosomes, palaeosomes, and relationships between
sharp margins?
them should be described at an outcrop of metatexite.
These observations are important as they allow classifi-
This could include the following observations:
cation of the diatexite and an understanding of the
(i) what are shapes of the leucosomes? Are they
processes that formed them (e.g. the degree of strain,
equant, elongate, sheet-like, planar, curviplanar?
fraction of partial melting, external strain).
(ii) what are the orientations of the leucosomes, and
how do they relate to the foliation or compo-
sitional layering in the host? For example, are the What to record at outcrop of migmatite
leucosomes parallel to the axial planes of folded As for any other exposure, the orientations of structural
palaeosome, or do they fill shear zones (exten- elements should be measured. The type of structure
sional, compressional) or have they intruded measured, and what they are defined by (e.g. leucosome,
boudin necks? This is important, as it will pro- schlieren, palaeosome, solid-state foliation, bedding),
vide insights into the stresses acting on the mig- should be recorded. This second point can be important,
matite during partial melting as a melt-filled shear zone is likely to be con-
(iii) if there are several sets of leucosomes in different temporaneous with anatexis, whereas a mylonitic foli-
orientations, how do the different sets relate to ation (which results from recrystallisation of the
each other? Are they texturally continuous, minerals) would have occurred later, either late in the
event when the rocks have cooled or during a later

separate event. Some of the structural elements that can
be measured include:
(i) foliations and/or layering, making sure to note
what the feature is defined by, e.g. solid-state fo-
liations in the palaeosome, leucosomes, schlieren
(ii) fold axes and axial planes, recording whether
these structures are melt-filled or not
(iii) shears (measured as a plane), recording the apparent
sense of shear. If possible, measure the lineation that
would represent the transport direction.
This may not be possible if the leucosome is massive
and isotropic
(iv) if boudins are present, the orientation of the
layer that is being boudinaged plus, if possible,
the plunge of the boudin neck are measured, as
these will give an indication of the principal stress
directions during the anatectic event.
It is useful looking closely at the leucosomes for solid-state

textures, such as recrystallised quartz or feldspar grains, as
these would possibly indicate that the stress continued
after cooling and crystallisation of the migmatite. Impor-
tantly, migmatitic and solid-state structural elements may
not be in the same orientation, as the different rheology
may cause the rock to deform differently, but they would 19 Location of field localities that are described in the pre-
be reconcilable with the same stress field (see Pawley and sent paper, shown against South Australia state digital
Collins, 2002 for an example of the textural evolution of a elevation model image. White dashed line shows bound-
crystallising syn-kinematic magma). aries of major tectonic elements of South Australia
Field examples of migmatitic rocks from McGee et al., 2010). This has been interpreted to be a
South Australia function of pre-metamorphic hydrous alteration within the
ore zone, possibly as part of an epi- or mesothermal gold
Christie Gneiss, Mulgathing Complex around mineral system (McFarlane et al., 2007). The alteration of
Mt Christie and Challenger Mine, Gawler Craton the host rocks and the generation of a hydrous mineral
The Mulgathing Complex is a Neoarchaean to earliest assemblage likely affected the protolith heterogeneity and
Palaeoproterozoic (ca. 2555–2420 Ma) belt of rocks influenced the melt fertility of the system. In this example,
located in the central Gawler Craton (Fig. 19). It con- the recognition of migmatite and the variable intensity of
sists of a wide variety of lithologies from ultramafic partial melt within a sequence of migmatites have direct
intrusions, komatiites, basaltic and felsic volcanics, exploration significance.
clastic and chemical metasedimentary rocks. There are
felsic to mafic intrusions including some that predate
and some that are synchronous with the early Palaeo- Migmatites of Kimban Orogeny
proterozoic Sleafordian Orogeny (Reid et al., 2014). (,1700 Ma) within Sleaford Complex on
The Christie Gneiss is a unit that encompasses all of the southern Eyre Peninsula, Gawler Craton
metasedimentary lithologies of the Mulgathing Complex
deposited between ca. 2555 and ca. 2485 Ma The Sleaford Complex is the southern belt of Neoarchaean
(Jagodzinski et al., 2009). The Christie Gneiss preserves to early Palaeoproterozoic rocks with similar rock types
an abundance of migmatite lithology types. At the and ages to the Mulgathing Complex, of which it is con-
Mt Christie type locality (Fig. 20), the exposed meta- sidered a lateral equivalent (Daly and Fanning, 1993; Daly
banded iron formation is interlayered at the scale of tens et al., 1998; Swain et al., 2005). The Sleaford Complex
to hundreds of metres with stromatic migmatite that underwent high temperature metamorphism during the
contains abundant coarse-grained, poikiloblastic garnet, Sleafordian Orogeny, similar to the Mulgathing Complex
cordierite, K-feldspar, biotite and quartz (Fig. 20a). (Daly and Fanning, 1993; Dutch et al., 2010; Jagodzinski
Such mineral assemblages are common also at et al., 2012). However, the Sleaford Complex underwent a
Challenger Gold Mine, where stromatic migmatite second phase of high temperature metamorphism and
includes gold-bearing leucosomes (Fig. 20b). Peak associated deformation during the ca. 1730–1690 Ma
metamorphic poikioblastic garnet within the migmatite Kimban Orogeny (Dutch et al., 2008; 2010). Therefore,
possess gold inclusions, supporting the notion that gold large portions of the Sleaford Complex are poly-
mineralisation predates migmatite formation at metamorphic and deciphering which minerals or structural
Challenger and that gold mobilisation during peak elements formed during each event requires careful exam-
metamorphism played a role in localising the gold within ination. As a result of the Kimban Orogeny, many spec-
the migmatite (Tomkins and Mavrogenes, 2002). tacular examples of migmatite developed along the
At Challenger, there is a greater intensity of partial melt southern coastal outcrops of Eyre Peninsula, particularly
present in and around the ore body in comparison to those around Cape Carnot, Shoal Point and Point Sir Isaac
adjacent barren gneiss (Fig. 20c; McFarlane et al., 2007; (Fig. 21).
a stromatic metatexite migmatite with garnet-bearing, in-source leucosomes, at outcrop near Mt Christie; b typical stromatic
metatexite migmatite that is host to gold mineralisation in diamond core from Challenger Mine. Visible gold is often present
within coarse blue-quartz rich leucosomes; c underground photograph from Challenger Mine near M1 ore shoot. In this
photograph, upper right portion of gneiss clearly has more leuocosomes than the lower left portion. This differentiation into
greater or lesser degrees of partial melt has been interpreted to be in part related to localisation of melt into zone of enhanced
water content by pre-metamorphic hydrous alteration (see McFarlane et al., 2007; McGee et al., 2010)
20 Examples of migmatite from Mulgathing Complex, central Gawler Craton, South Australia
At the western headland of the Cape Carnot discordant structures, such as boudin necks (Fig. 23c)
locality, there are two broad rock types. These are and shear bands (Fig. 23d) in dilation metatexite mig-
garnet–cordierite-bearing megacrystic granitic gneiss matites. At greater degrees of partial melt production, the
and metapelitic gneiss. SHRIMP dating of the granitic rocks disaggregated to form schollen diatexite migma-
gneiss indicates that the granite protolith was emplaced tites, which have been folded (Fig. 23e). There is little
at 2431+ 7 Ma, with metamorphic zircon growth at palaeosome preserved in this unit, with a large proportion
1715+ 6 Ma (Jagodzinski et al., 2012). The granitic of light coloured leucogranitic melt and only minor dark
gneiss preserves net-textured migmatites related to foli- coloured, garnet-bearing melanosome and residuum.
ation boudinage (Fig. 22a and b), and examples that are The migmatites on the Coffin Bay Peninsula at Point
garnet+ cordierite-bearing are in-source leucosomes that Sir Isaac (Fig. 21) developed in both mafic and felsic
become diatexites in places (Fig. 22c). igneous protoliths. The Point Sir Isaac coast consists of
Shoal Point, on the central south coast of Eyre generally undeformed late Sleafordian-aged (2414+ 6
Peninsula, contains spectacular examples of metatexite Ma) garnet+ cordierite bearing S-type granite, locally
and diatexite in a variety of metasedimentary and intruded by a number of mafic dykes ranging from 5 to
igneous lithologies. These units underwent two phases of 50 m in width (Dutch et al., 2008). This area was sub-
high-T metamorphism during the Sleafordian and sequently metamorphosed during the Kimban Orogeny
Kimbanorogenies (Dutch et al., 2010). Peak conditions at conditions reaching 750uC and 10 kbar. Importantly,
during the Sleafordian reached 750uC at 5–6 kbar, the majority of the granite, due to its relatively anhy-
whereas during the Kimban conditions reached over drous assemblage, was unaffected by the later meta-
800uC at up to 10 kbar. The peak Sleafordian assem- morphism. Deformation and migmatisation were limited
blages are only preserved as boudins in the high-strain to discrete zones, interpreted to be hydrated fracture
Kimban fabrics where they preserve net structure mig- zones or alteration zones, or the relatively hydrous
matites with large garnet porphyroblasts forming within hornblende-bearing mafic dykes (Dutch et al., 2008).
the leucosomes (Fig. 23a). Within the Kimban-aged High grade melting within the felsic shear zones has
fabrics, leucosomes formed parallel layers in stromatic produced dilational metatexites with leucosomes occur-
metatexite migmatites, with some of the thicker layers ring both concordant with and cross-cutting the shear
forming transfer zones (Fig. 23b). There are also fabric and within shear bands (Fig. 23f ). Melting within
21 Migmatite localities described in the present paper from southern Eyre Peninsula, shown against regional total magnetic
intensity (TMI) image
the mafic lithologies was limited to small-scale patch This can be seen in Fig. 23g where small amounts of leuco-
migmatisation probably produced by the reaction cratic melt contain garnet and clinopyroxene porphyroblasts.
That the melting within the granite was limited to hydrous
Hornblende þ plagioclase þ quartz ¼ melt þ garnet+ zones and that the hornblende-rich mafic units melted at all
clinopyroxene reflect the strong role of H2O in increasing melt fertility.
a net-structured and stromatic metatexite migmatite with K-feldspar megacrystic granite protolith; b in situ and in-source
leucosomes forming in dilational sites in K-feldspar megacrystic granite protolith; c transgressive garnet–cordierite-bearing
in-source leucosome that grades into diatexite migmatite to upper right of field of view
22 Examples of migmatite from Cape Carnot, southern Eyre Peninsula, South Australia
a net-structured metatexite migmatite from Shoal Point, with large garnet porphyroblasts forming within leucosomes. These
leucosomes formed during Sleafordian metamorphism and are now only preserved in low strain boudins within high-strain
Kimban fabrics; b stromatic metatexite migmatite from Shoal Point, with thick leucogranitic veins channelling melt migration
parallel to compositional layering within migmatite; c dilation metatexite migmatite from Shoal Point, with leucocratic melt
localised into dilational sites, in this case within neck zones of domino-type boudins of mafic lithology, which can be con-
sidered palaeosome. Asymmetry of these domino boudins indicates sinistral shear; d Kimban-aged, ,1700 Ma leucosome-
filled, dextral shear bands in net-structured metatexite migmatite from Shoal Point; e folded schollen diatexite migmatite from
Shoal Point. Rock has high degree of neosome and only minor residuum is preserved; f net-structured metatexites mig-
matites within a sheared ,2415 Ma granitic protolith at Point Sir Isaac; g in situ leucosome containing garnet and clin-
opyroxene porphyroblasts, within patch metatexite migmatite at Point Sir Isaac. Protolith was mafic dyke
23 Examples of migmatite from Shoal Point and Point Sir Isaac, southern Eyre Peninsula, South Australia
Olarian (,1600 Ma) migmatites of Curnamona compositional layering (Fig. 24a). Such leucosomes are
Group, Olary Domain, Curnamona Province commonly folded by second generation and younger
The southern Curnamona Province is divided into the folds (Fig. 24b). Locally, partial melting has developed
Broken Hill Domain in the east (mostly in NSW) and to a greater extent and stromatic migmatite grades into
Olary Domain in the west (Fig. 19). The oldest exposed schlieric and diatexite migmatite (Fig. 24c), or stromatic
rocks are a late Palaeoproterozoic (ca. 1720–1710 Ma) migmatite is intruded by flow-foliated, syn-tectonic leu-
metasedimentary succession, the Willyama Supergroup, cogranite sheets (Fig. 24d). Where the parent metasedi-
with lesser volcanic components that were moderately to mentary rocks are albite-rich, which is commonly the
highly deformed and metamorphosed during the ca. case in the Wiperaminga Subgroup, the leucosomes are
1620–1590 Ma Olarian Orogeny (Page et al., 2005). The also albitic.
metamorphic style of the Olarian Orogeny is broadly In contrast, the coarse-grained two-mica granites of
high temperature–low pressure, ranging from granulite the Bimbowrie Suite are S-type, mainly potassic granite,
facies metamorphic conditions (e.g. Willis et al., 1983; and are mostly undeformed, except where they are cut
White et al., 2005) in the Broken Hill Domain to by late, retrograde shear zones. Many granitic dykes
greenschist facies in the northern-central portion of the intruded along bedding planes, whereas others, par-
Curnamona Province. The region of granulite facies ticularly the larger granite bodies, appear to cross-cut
extends westward into South Australia, where meta- the regional structural trends and have sharp intrusive
contacts. As such, it appears that these larger granite
morphic conditions of 700–800uC and 4.6–5.6 kbar have

been estimated (Webb and Crooks, 2005). In the Olary bodies have formed late in the orogenic process (Fricke,
Domain, metamorphic grade is somewhat lower, 2008), whereas most migmatisation occurred early under
decreasing from upper to lower amphibolite facies high heat flow conditions. At Bimbowrie therefore, the
towards the northwest. This is interpreted to reflect authors observe:
differential exhumation, along late Olarian shear zones (i) compositional controls on the degree of partial
involving tectonically stacked nappes of Willyama melting, with greater partial melt formed within
Supergroup metasedimentary rocks. Migmatitic rocks more psammopelitic lithologies compared to
are located in the regions with upper amphibolite facies those metasedimentary rocks of a more psam-
assemblages or higher metamorphic grade rocks, mainly mitic, quartz-rich composition
in the southeast (Clarke et al., 1987; Crooks, 2001; (ii) that partial melting is confined to the lower
Webb and Crooks, 2005), although there is also local stratigraphic units and decreases up-section, and
migmatitisation in the northwest that may be related to (iii) the presence of large plutons of mostly unde-
intrusion of the Crocker Well Suite granitoids. Shear- formed late Olarian, two-mica granite that post-
bounded areas of lower amphibolite facies metasedi- dates peak metamorphism.
mentary rocks, which are significantly less recrystallised The first observation can be explained simply as a
and have andalusite as the dominant aluminosilicate, function of the control of bulk rock chemistry on the
generally do not show evidence of migmatisation. timing and degree of partial melting. The second
In the upper amphibolite facies rocks, evidence of partial observation indicates that early heating during burial
melting is found in lithologies of appropriate bulk was a significant process in the Olarian Orogeny.
composition, mainly in the lower part of the Willyama Migmatite formation during burial metamorphism is
Supergroup stratigraphy (Curnamona Group). likely to have been a major factor in the formation of
The Wiperaminga Subgroup (lower Curnamona Group) is migmatite in the lower stratigraphic units of the
the most intensely affected, especially in its oldest units, Curnamona Province (Clarke et al., 1987).
whereas migmatisation in the overlying Ethiudna Sub- Large bodies of truly syn-tectonic granite have not
group is more incipient (Conor and Preiss, 2008). There is been found in the Olary Domain, but migmatisation of
little migmatite development in the upper Willyama supracrustal rocks occurred early in the deformation
Supergroup in the Olary Domain, although it is common in history, and the granitic magmas thus generated show
the Broken Hill Domain. evidence of only incipient mobilisation and pooling on a
Many gold and copper deposits in the Curnamona limited scale. Therefore, whereas some of the late granite
Province are associated with pegmatitic and other plutons in the Curnamona Province, e.g. the Bimbowrie
leucocratic veins (Cooke, 2003; Burtt et al., 2004). Some Suite, have geochemical signatures suggesting derivation
occur in high-grade regions close to the source of ana- from partial melting of the Willyama Supergroup, others
texis (e.g. White Dam), whereas others were formed at have formed via fractionation from mantle-derived
much higher crustal levels in low-grade metasedimentary melts (Barovich and Foden, 2002; Fricke, 2008). This
rocks (e.g. Portia, Kalkaroo). Such veins are likely to be suggests that there may have been a second phase of
related, at least in part, to partial melting during the melting of supracrustal rocks in tandem with mantle
Olarian Orogeny (Skirrow, 2003). High heat flow during processes to produce late- to post-orogenic granites that
this orogeny partially melted Willyama Supergroup were generated at a lower crustal level than is currently
rocks and also resulted in the remobilisation of metals exposed and intrude a variety of structural levels within
during regional fluid flow and related structural events. deformed Willyama Supergroup rocks.
As an example of migmatite in the Curnamona Pro-
vince, the following observations come from the Mulga Musgravian Orogeny (, 1200– 1160 Ma)
Creek–Oonartra Creek area in Bimbowrie Conservation migmatites of eastern Musgrave Province
Park. Within Wiperaminga Subgroup rocks, partial melt The eastern Musgrave Province (Fig. 19) is interpreted to
generally constitutes less than 15% of the rock mass be a ca. 1600–1550 Ma magmatic arc (Wade et al., 2006).
and typically forms discrete, in situ and in-source leu- These rocks were deformed during the ca. 1200–1160 Ma
cosomes that are parallel or subparallel to the main Musgrave Orogeny, which was accompanied by the ca.
a coarse-grained in-source leucosomes within psammopelitic protolith. Leucosomes are generally parallel to layering, but note
that some leucosomes are also developed in en echelon dilational arrays that suggest subhorizontal, top to the left shear (present-
day reference frame); b in-source leucosomes preserved within individual psammopelite layers, folded here by open, probable
third generation fold; c transitional metatexite to diatexite migmatite. Some portions of this migmatite consist entirely of medium-
grained leucosome that has engulfed residuum to form schollen. Region to the right of photo has greater degree of melanocratic
layers and schlieren; d melt-rich metatexite migmatite (bottom) that is intruded by weakly foliated, syn-tectonic granite sheet (top)
24 Olarian-aged (,1600 Ma) migmatites of Wiperaminga Subgroup, Curnamona Province, Bimbowrie Conservation Park,
South Australia
1200–1140 Ma, anhydrous, charnockite series Pitjantjat- steep dip (Fig. 26), although there is local variation that
jara Supersuite (Camacho and Fanning, 1995; Smithies can be attributed to outcrop-scale folding and the
et al., 2010). The province was then intruded by ca. 1080– apparent realignment of the layering beside large faults
1040 Ma rocks of the Giles Event (Glikson et al., 1996; (Fig. 26). Mafic selvedges to the leucosomes are rare,
Edgoose et al., 2004), before being reworked during the and their relatively sharp boundaries (see below) suggest
low-grade ca. 600–530 Ma Petermann Orogeny that they are mostly in-source leucosomes that have
(Scrimgeour and Close, 1999). The Musgrave Orogeny was segregated and migrated from the site of partial melting.
a regional amphibolite to granulite facies metamorphic Rare leucosomes have mafic selvedges however, indi-
event that resulted in partial melting of arc-related cating that at least some partial melting may have
magmatic rocks and a variety of sedimentary lithologies in occurred at the current exposure level.
the eastern Musgrave Province, which together are inclu- Based on their textures and contact relations, the
ded within the gneisses of the Birksgate Complex (Major layer-parallel leucosomes can be subdivided into two
and Conor, 1993). The tectonic setting of the Musgrave types (Fig. 25b): medium-grained leucosomes with
Orogeny is unclear. However, based on the long-lived, relatively sharp, straight margins and coarser grained
widespread nature of the event, and the high-temperature leucosomes with less sharp, but still straight, margins.
granites of the Pitjantjatjara Supersuite, it is considered to The margins of the latter type are often stepped, fol-
have been associated with extension and/or upwelling in an lowing the grain boundaries of the individual crystals in
intracontinental setting (Smithies et al., 2010; 2011). the residuum. The relatively sharp, straight margins of
On the Agnes Creek 1:100 000 map sheet (Fig. 19), the the medium-grained, gneiss-parallel leucosome suggests
Birksgate Complex rocks have a layering that is gener- that they may have been earlier and have experienced
ally defined by parallel leucosomes classifying it as a more strain (i.e. the margins have been transposed),
stromatic metatexite migmatite, although there are also relative to the parallel, coarse-grained leucosomes that
locally developed dilation metatexite migmatites, have stepped, or relatively unmodified margins
characterised by melt-filled shear bands and boudins that follow grain boundaries. While the possibility exists
(Fig. 25a). On the regional scale, the gneissosity and that these two leucosome generations formed in
parallel leucosomes are north–northeast-striking with a different melt-producing events, it is likely they
a reverse, melt-filled shear band; b close-up illustrating

difference between early, straight-edged leucosomes
(labelled 1) and parallel, later, coarser grained leuco-
somes that have sharp stepped margins that follow grain
boundaries (labelled 2). Straight-edged leucosome is
deflected and cut by oblique shear band (labelled 3),
which is filled with leucosome that is texturally continuous
with coarser grained leucosome, indicating that they are
the same generation
25 Outcrop of Birksgate Complex, eastern Musgrave
Province, South Australia
developed progressively during the same event. Precise

geochronology is required to distinguish between these
two possibilities.
In addition to the layer-parallel leucosomes, there are a
series of oblique, coarse-grained, leucosomes, which have
irregular margins and which cut the gneissosity at low to
moderate angles (Fig. 25). The medium-grained leuco-
somes are truncated and deflected by the oblique leuco-
somes. The coarse-grained, gneiss-parallel leucosomes also
appear to be deflected by the oblique leucosomes, but closer
examination reveals that they are also texturally continu-
ous with them. The oblique leucosomes are interpreted to
be syn-anatectic, melt-filled, shear bands that have been
observed to occur in three orientations: steep northeast- 26 Equal area stereonets for a gneissosity (contoured
striking dextral, steep south–southeast-striking sinistral, poles to planes) and b melt-filled shear bands on central
and north–northeast-striking reverse shears that dip part of Agnes Creek 1:100 000 map sheet (as great
moderately to the northwest (Fig. 26c). The deflection of circles); c summary stereonet that shows relationship
the leucosomes suggests they are overprinted by the melt- between average orientations of various shear zones
filled shear bands, although the textural continuity between and average gneissosity (thick black line), with
the coarse-grained layer-parallel and oblique leucosomes proposed shortening direction indicated by large
and the absence of cross-cutting relations suggest that they black arrows
a stromatic metatexite migmatite with layer-parallel partial melt focused within certain sedimentary layers; b coarse-grained
stromatic metatexite migmatite with layer-parallel in-source leucosomes in biotite-rich residuum, which was folded by upright
probable F2 folds; c some zones of this migmatite have undergone significant partial melting and can be considered tran-
sitional diatexite migmatite in layers in centre of field of view. Narrow leucosome vein projects subvertically from these melt-
rich zones (labelled 1) and entire migmatite is cut by shallow dipping granite dyke (labelled 2)
27 Examples of migmatitic Kanmantoo Group metasediments near Reedy Creek, South Australia
formed at the same time with the melt from the layer-par- northwest-directed tectonic transport (Preiss, 1995; Foden
allel leucosomes likely flowing into a low pressure site et al., 2006). Metamorphism above greenschist facies is
represented by the shear band. At one locality, the gneissic found only in the eastern and southern Mount Lofty
fabric is cut by a conjugate pair of melt-filled sinistral and Ranges and Kangaroo Island. The highest metamorphic
dextral shear bands. grade reaches the sillimanite stability field and is locally
These observations suggest the 1600–1550 Ma protolith accompanied by migmatisation. Such rocks occur within a
underwent approximately east–southeast–west–north- narrow zone of high T–low P metamorphism trending
west-directed shortening (based on present orientations) north–northwestward from the Barossa Ranges in
during the Musgravian Orogeny producing the pre- South Australia to the Glenelg River in western Victoria
dominantly north-northeast-striking gneissosity and layer- (Kemp and Gray, 1999). Within this zone, Neoproterozoic
parallel leucosomes. The orientations of contemporaneous, and early Cambrian metasedimentary rocks include
melt-filled shear bands described above (Fig. 26c) are psammopelitic and lesser pelitic schist, metasandstone,
consistent with this shortening direction and are interpreted quartzite, marble and calc-silicate. Evidence of partial
as a set of conjugate structures. The relationship between melting is restricted to the stratigraphically lowest unit of
the various structures also suggests that locally at least, the the Kanmantoo Group (Carrickalinga Head Formation).
area was subjected to pure shear rather than simple shear In its type area on the west coast of Fleurieu Peninsula,
during the Musgravian Orogeny. this formation is an essentially unmetamorphosed dirty
sandstone–siltstone alternation, but in the eastern Mount
Delamerian migmatites in Kanmantoo Group, Lofty Ranges it is strongly migmatised, with excellent
eastern Mount Lofty Ranges exposure preserved at Reedy Creek (Fig. 19).
The Kanmantoo Group is an extremely thick succession of At Reedy Creek, anatexis occurred early in the defor-
shallow and deep-water mainly clastic sediments deposited mation history, with the leucosomes following typically a
rapidly in an early Cambrian extensional basin near the strong layer-parallel foliation (S1), related to northwest-
eastern margin of Gondwana after the Neoproterozoic directed thrusting (Fig. 27a), to produce a stromatic
break-up of Rodinia (Flottmann et al., 1998). Deep burial metatexite migmatites. This foliation and accompanying
by sedimentation was immediately followed by onset of the leucosomes were folded by upright north–south-trending
Delamerian Orogeny (ca. 510–490 Ma), involving overall F2 folds of probable transpressional origin (Fig. 27b) and
later northwest-trending F3 folds, comparable to structures Brown, M. 2013. Granite: from genesis to emplacement, Geol. Soc. Am.
seen in the low-grade areas of the Delamerian Orogeny. Bull., 125, 1079–1113.
Brown, M. and Solar, G. S. 1998. Shear-zone systems and melts:
The migmatites were then cut by shallow-dipping sheets of feedback relations and self-organization in orogenic belts,
undeformed, post-tectonic granite (Fig. 27c). J. Struct. Geol., 20, (2-3), 211–227.
The observation that partial melting is recorded only Burg, J.-P. 1991. Syn-migmatization way-up criteria, J. Struct. Geol.,
in the lowest formation of the Kanmantoo Group may 13, 617–623.
Burg, J.-P. and Vanderhaege, O. 1993. Structures and way-up criteria in
reflect a combination of factors. First, the depth of migmatites, with application to the Velay dome (French Massif
burial may have been important, as the Carrickalinga Central), J. Struct. Geol., 15, 1293–1301.
Head Formation was likely buried to at least 10 km, Burtt, A., Conor, C. H. H. and Robertson, R. S. 2004. Curnamona—an
possibly up to 15 km deep. Furthermore, the geothermal emerging Cu-Au province, MESA J., 33, 9–17.
gradient may still have been high as a result of syn- Camacho, A. and Fanning, C. M. 1995. Some isotopic constraints on
the evolution of the granulite and upper amphibolite facies
depositional rifting of the Kanmantoo Trough. Second, terranes in the eastern Musgrave Block, central Australia,
the free water content in the rocks may have been high, Precambrian Res., 71, (1-4), 155–181.
as these sediments were extremely rapidly deposited Cartwright, I. and Buick, I. S. 2000. Fluid generation, vein formation
shortly before the onset of metamorphism and may and the degree of fluid-rock interaction during decompression
of high-pressure terranes: the Schistes Lustrés, Alpine Corsica,
therefore still have contained significant pore fluids. France, J. Metamorph. Geol., 18, 607–624.
As described above, the presence of free water is significant Clarke, G. L., Guiraud, M., Powell, R. and Burg, J.-P. 1987.
for lowering the solidus for metasedimentary rocks. Metamorphism in the Olary Block, South Australia: compression
In contrast, the non-migmatised Neoproterozoic sedi- with cooling in a Proterozoic fold belt, J. Metamorph. Geol., 5,
ments that underlie the Kanmantoo Group were already 291–306.
Clemens, J. D. 1984. Water contents of silicic to intermediate magmas,
up to 300 m.y. old at the time of the Delamerian Orogeny, Lithos, 17, (4), 273–287.
had been thoroughly dewatered and hence would be unli- Clemens, J. D. 2013. Granitic magmatism, from source to
kely to have undergone fluid-induced partial melting. emplacement: a personal view, Appl. Earth Sci. (Trans. Inst.
Third, the melt fertility of the protolith may also have been Min. Metall. B), 121, (3), 107–136.
Clemens, J. D. and Mawer, C. K. 1992. Granitic magma transport by
a factor, as these relatively immature, clay-rich sandy fracture propagation, Tectonophysics, 204, 339–360.
sediments, with much detrital feldspar and mica, were of Collins, W. J. and Sawyer, E. W. 1996. Pervasive granitoid magma
the right bulk composition to undergo dehydration-melt- transfer through the lower-middle crust during non-coaxial
ing partial melting at relatively low temperatures. compressional deformation, J. Metamorph. Geol., 14, (5),
565–579.
Conor, C. H. H. and Preiss, W. V. 2008. Understanding the 1720-1640
Ma Palaeoproterozoic Willyama Supergroup, Curnamona
Summary Province, Southeastern Australia: implications for tectonics,
Migmatites are a fascinating rock type, widespread basin evolution and ore genesis, Precambrian Res., 166, 297–317.
Cooke, A. 2003. White Dam—an exciting new gold project in the
through geological time and present within many geo- Curnamona Province, MESA J., 31, 4–5.
logical provinces. The examples presented in this review Crooks, A. F. 2001. Olary-Broken Hill Domain boundary—MINGARY
show that understanding and identifying migmatites in 1:100 000 map area, Curnamona Province, MESA J., 20, 44–45.
the field and in drill core are relevant for many geologists Daly, S. J. and Fanning, C. M. 1993. Archaean, in The geology of
working in high-grade terranes. While only brief, it is South Australia, Vol. 1, The Precambrian, Bulletin,
(eds. J. F. Drexel, W. V. Preiss and A. J. Parker), Vol. 54, 32–49;
hoped that this introduction to the formation and Adelaide, Geological Survey of South Australia.
development of migmatitic rocks will inspire the geol- Daly, S. J., Fanning, C. M. and Fairclough, M. C. 1998. Tectonic
ogist to look at these rocks with fresh eyes. The examples evolution and exploration potential of the Gawler
presented in this review show that migmatites can pro- Craton, South Australia, AGSO J. Aust. Geol. Geophys., 17, (3),
145–168.
vide a considerable range of geological data. They play a Dutch, R. A. 2009. Reworking the Gawler Craton: metamorphic and
key role in the generation of granites and the differen- geochronologic constraints on Palaeoproterozoic reactivation of
tiation of the crust and can help the geologist unravel the southern Gawler Craton, Australia. PhD Thesis, University
structural histories and stress patterns. of Adelaide, Adelaide, Australia.
Dutch, R. A., Hand, M. and Kelsey, D. E. 2010. Unravelling the
tectonothermal evolution of reworked Archean granulite facies
Acknowledgements metapelites using in situ geochronology: an example from
The authors would like to thank N. Phillips for inviting us to the Gawler Craton, Australia, J. Metamorph. Geol., 28, (3),
293–316.
submit the present paper. This is an expanded version of an Dutch, R. A., Hand, M. and Kinny, P. D. 2008. High-grade
article first published in Volume 69 of the MESA Journal Paleoproterozoic reworking in the southeastern Gawler Craton,
(2013), and it also incorporates information published in the South Australia, Aust. J. Earth Sci., 55, (8), 1063–1081.
DMITRE Report Book (2013/00016). The authors would Edgoose, C. J., Scrimgeour, I. R. and Close, D. F. 2004. Geology of the
Musgrave Block, Northern Territory. Report 15, Darwin,
like to thank the two anonymous reviewers. The authors are Northern Territory Geological Survey.
also grateful to all of their colleagues who have participated Flottmann, T., et al., 1998. Formation and reactivation of the
over the years in many excellent discussions on migmatites, Cambrian Kanmantoo Trough, SE Australia: implications for
both in South Australia and further afield. early Palaeozoic tectonics at eastern Gondwana’s plate margin,
J. Geol. Soc., 155, 525–539.
Foden, J. D., Elburg, M. A., Dougherty-Page, J. and Burtt, A. 2006.
References The timing and duration of the Delamerian Orogeny: correlation
Arzi, A. A. 1978. Critical phenomena in the rheology of partially with the Ross Orogen and implications for Gondwana assembly,
melted rocks, Tectonophysics, 44, (1-4), 173–184. J. Geol., 114, 189–210.
Blumenfeld, P. and Bouchez, J.-L. 1988. Shear criteria in granite and Fricke, C. E. 2008. Definitions of Mesoproterozoic igneous rocks of the
migmatite deformed in the magmatic and solid states, J. Struct. Curnamona Province: the Ninnerie Supersuite. Report Book /04,
Geol., 10, (4), 361–372. Adelaide, Department of Primary Industries and Resources,
Brown, M. 1994. The generation, segregation, ascent and emplacement of South Australia.
granite magma: the migmatite-to-crustally-derived granite Gerdes, A., Wörner, G. and Henk, A. 2000. Post-collisional granite
connection in thickened orogens, Earth Sci. Rev., 36, (1-2), 83–130. generation and HT-LP metamorphism by radiogenic heating:
the Variscan South Bohemian Batholith, J. Geol. Soc., 157, Pawley, M. J., Van Kranendonk, M. J. and Collins, W. J.
577–587. 2004. Interplay between deformation and magmatism
Glikson, A. Y., et al., 1996. Geology of the western Musgrave Block, during doming of the Archaean Shaw Granitoid Complex,
central Australia. with particular reference to the mafic- Pilbara Craton, Western Australia, Precambrian Res., 131,
ultramafic Giles Complex, Bulletin 239; Canberra, Australian 213–230.
Geological Survey Organisation. Powell, R. and Holland, T. J. B. 1988. An internally consistent dataset
Holland, T. and Powell, R. 2001. Calculation of phase relations with uncertainties and correlations: 3. Applications to
involving haplogranitic melts using an internally consistent geobarometry, worked examples and a computer program,
thermodynamic dataset, J. Petrol., 42, (4), 673–683. J. Metamorph. Geol., 6, (2), 173–204.
Holness, M. B. 2008. Decoding migmatite microstructures, in Working Preiss, W. V. 1995. Delamarian Orogeny, in The geology of
with migmatites, (eds. E. W. Sawyer and M. Brown), 57–76; South Australia, Vol. 2, The Phanerozoic, Bulletin 54,
Quebec, Mineralogical Association of Canada. (eds. J. F. Drexel, W. V. Preiss and A. J. Parker), 45–59;
Jagodzinski, E. A., Reid, A. J. and Fraser, G. L. 2009. Compilation of Adelaide, Geological Survey of South Australia.
SHRIMP U–Pb geochronological data for the Mulgathing Read, H. H. 1957. The granite controversy; London, Thomas
Complex, Gawler Craton, South Australia, 2007–2009. Report Murby and Co.
Book 2009/16, Adelaide, Department of Primary Industries and Reid, A. J., et al., 2014. SHRIMP U-Pb zircon age constraints on the
Resources, South Australia. tectonics of the Neoarchean to early Paleoproterozoic transition
Jagodzinski, E. A., Reid, A. J. and Hand, M. 2012. SHRIMP U-Pb within the Mulgathing Complex, Gawler Craton,
Geochronology of Archaean to Palaeoproterozoic rocks from South Australia, Precambrian Res., 250, 27–49.
the southern Eyre Peninsula. Report Book 2012/015, Adelaide, Renner, J., Evans, B. and Hirth, G. 2000. On the rheologically critical
Department for Manufacturing, Innovation, Trade, Resources melt fraction, Earth Planet. Sci. Lett., 191, (4), 585–594.
and Energy, South Australia. Rosenberg, C. L. and Handy, M. R. 2005. Experimental deformation of
Johannes, W. 1985. The significance of experimental studies for the partially melted granite revisited: implications for the continental
formation of migmatites, in Migmatites, (ed. J. R. Ashworth), crust, J. Metamorph. Geol., 23, 19–28.
36–85; Glasgow, Blackie. Sawyer, E. W. 1991. Disequilibrium melting and the rate of melt-
Jung, S., Hoernes, S., Masberg, P. and Hoffer, E. 1999. The residuum separation during migmatization of mafic rocks from
petrogenesis of some migmatites and granites (Central Damara the Grenville Front, Quebec. J. Petrol., 32, 701–738.
Orogen, Namibia): evidence for disequilibrium melting, Sawyer, E. W. 1996. Melt segregation and magma flow in migmatites:
wall-rock contamination and crystal fractionation, J. Petrol., 40, implications for the generation of granite magmas, Trans.
(8), 1241–1269. R. Soc. Edinburgh: Earth Sci., 87, (1-2), 85–94.
Kemp, A. I. S. and Gray, C. M. 1999. Geological context of crustal anatexis Sawyer, E. W. 2008a. Atlas of Migmatites; Ottawa, Ontario, NRC
and granitic magmatism in the northeastern Glenelg River Complex, Research Press.
western Victoria, Aust. J. Earth Sci., 46, (3), 407–420. Sawyer, E. W. 2008b. Working with migmatites: nomenclature
Kirkland, C. L., et al., 2011. On the edge: U-Pb, Lu-Hf, and Sm-Nd of the constituent parts, in Working with migmatites,
data suggests reworking of the Yilgarn craton margin during (eds. E. W. Sawyer and M. Brown), 1–28; Quebec, Canada,
formation of the Albany-Fraser Orogen, Precambrian Res., 187, Mineralogical Association of Canada.
(3-4), 223–247. Scrimgeour, I. R. and Close, D. F. 1999. Regional high-
Lucas, S. B. and St-Onge, M. R. 1995. Syn-tectonic magmatism and the pressure metamorphism during intracratonic deformation:
development of compositional layering, Ungava Orogen the Petermann Orogeny, central Australia, J. Metamorph. Geol.,
(northern Quebec, Canada), J. Struct. Geol., 17, 475–491. 17, 557–572.
Lupulescu, A. and Watson, E. B. 1999. Low melt fraction connectivity of Skirrow, R. G. 2003. Fe-Oxide Cu-Au deposits: potential of the
granitic and tonalitic melts in a mafic crustal rock at 800 degrees C Curnamona Province in an Australian and global context,
and 1 GPa, Contrib. Mineral. Petrol., 134, (2-3), 202–216. in Broken Hill Exploration Initiative: Abstracts from the
Major, R. B. and Conor, C. H. H. 1993. The Musgrave Block, in The July 2003 conference, (ed. M. Peljo), 158–161; Canberra,
geology of South Australia, Vol. 1, The Precambrian, Bulletin 54, Geoscience Australia.
(eds. J. F. Drexel, W. V. Preiss and A. J. Parker), 156–167; Smithies, R. H., et al., 2010. Geochemistry, geochronology and
Adelaide, Geological Survey of South Australia. petrogenesis of Mesoproterozoic felsic rocks in the western
McFarlane, C. R. M., Mavrogenes, J. A. and Tomkins, A. G. 2007. Musgrave Province of central Australia, and implication for the
Recognizing hydrothermal alteration through a granulite facies Mesoproterozoic tectonic evolution of the region. Report 106;
metamorphic overprint at the Challenger Au deposit, Perth, Geological Survey of Western Australia.
South Australia, Chem. Geol., 243, 64–89. Smithies, R. H., et al., 2011. High-temperature granite magmatism,
McGee, B., Giles, D., Kelsey, D. E. and Collins, A. S. 2010. Protolith crust-mantle interaction and the Mesoproterozoic
heterogeneity as a factor controlling the feedback between intracontinental evolution of the Musgrave Province,
deformation, metamorphism and melting in a granulite-hosted Central Australia, J. Petrol., 52, (5), 931–958.
gold deposit, J. Geol. Soc., 167, 1089–1104. Solar, G. S. and Brown, M. 2001. Petrogenesis of migmatites in Maine,
McLaren, S., Sandiford, M. and Powell, R. 2015. Contrasting styles of USA: possible source of peraluminousleucogranite in plutons?,
Proterozoic crustal evolution: a hot-plate tectonic model for J. Petrol., 42, (4), 789–823.
Australian terranes, Geology, 33, (8), 673–676. Spear, F. S. 1993. Metamorphic phase equilibria and pressure-
Mehnert, K. R., Busch, W. and Schneider, G. 1973. Initial melting at temperature-time paths; Washington, D.C, Mineralogical
grain boundaries of quartz and feldspar in gneisses and Society of America.
granulites, Neues Jahrb. Mineral. Monatsh., 4, 165–183. Spear, F. S., Kohn, M. J. and Cheney, J. T. 1999. P-T paths from
Oliver, N. H. S. and Barr, T. D. 1997. The geometry and evolution of anatecticpelites, Contrib. Mineral. Petrol., 134, (1), 17–32.
magma pathways through migmatites of the Halls Creek Orogen, Stüwe, K. 2007. Geodynamics of the lithosphere: an introduction;
Western Australia, Mineral. Mag., 61, 3–14. Berlin/Heidelberg, Springer-Verlag.
Olsen, S. N. 1985. Mass balance in migmatites, in Migmatites, Swain, G. M., et al., 2005. Provenance and tectonic development of the
(ed. J. R. Ashworth), 145–179; Glasgow, Blackie. late Archaean Gawler Craton, Australia; U-Pb zircon,
Page, R. W., Stevens, B. P. J. and Gibson, G. M. 2005. Geochronology geochemical and Sm-Nd isotopic implications, Precambrian Res.,
of the sequence hosting the Broken Hill Pb-Zn-Ag Orebody, 141, 106–136.
Australia, Econ. Geol., 100, (4), 633–661. Thompson, A. B. 1996. Fertility of crustal rocks during anatexis, Trans.
Passchier, C. W., Myers, J. S. and Kröner, A. 1990. Field geology of R. Soc. Edinburgh: Earth Sci., 87, (1-2), 1–10.
high-grade Gneiss Terrains; Berlin/Heidelberg, Springer-Verlag. Thompson, A. B. 2001. Clockwise P-T paths for crustal melting and
Passchier, C. W. and Trouw, R. A. J. 2005. Microtectonics; Berlin/ H2O recycling in granite source regions and migmatite terrains,
Heidelberg, Springer-Verlag. Lithos, 56, (1), 33–45.
Patiño Douce, A. E. and Johnston, A. D. 1991. Phase equilibria and Tomkins, A. G. and Mavrogenes, J. A. 2002. Mobilization of gold as a
melt productivity in the pelitic system—implications for the polymetallic melt during peliteanatexis at the Challenger deposit,
origin of peralunimous granitoids and aluminous granulites, South Australia; a metamorphosed Archean gold deposit, Econ.
Contrib. Mineral. Petrol., 107, (2), 202–218. Geol., 97, 1249–1271.
Pawley, M. J. and Collins, W. J. 2002. The development of contrasting Van der Molen, I. and Paterson, M. S. 1979. Experimental deformation
structures during the cooling and crystallisation of a syn- of partially-melted granite, Contrib. Mineral. Petrol., 70, (3),
kinematic pluton, J. Struct. Geol., 24, 469–483. 299–318.
Vigneresse, J. L., Barbey, P. and Cuney, M. 1996. Rheological transitions rocks from Broken Hill, Australia, J. Metamorph. Geol., 23,
during partial melting and crystallization with application to Felsic 579–602.
Magma segregation and transfer, J. Petrol., 37, (6), 1579–1600. White, R. W. and Powell, R. 2002. Melt loss and the preservation of
Vigneresse, J. L. and Tikoff, B. 1999. Strain partitioning during partial granulite facies mineral assemblages, J. Meta. Geol., 20, 621–632.
melting and crystallizing felsic magma, Tectonophysics, 312, (2-4), White, R. W., Powell, R. and Halpin, J. A. 2004. Spatially-focussed
117–132. melt formation in aluminous metapelites from Broken Hill,
Wade, B. P., et al., 2006. Evidence for early Mesoproterozoic arc Australia, J. Metamorph. Geol., 22, (9), 825–845.
magmatism in the Musgrave Block, Central Australia: White, R. W., Powell, R. and Holland, T. J. B. 2001. Calculation of
implications for Proterozoic crustal growth and tectonic partial melting equilibria in the system Na2O-CaO-K2O-FeO-
reconstructions of Australia, J. Geol., 114, (1), 43–63. MgO-Al2O3-SiO2-H2O (NCKFMASH), J. Metamorph. Geol.,
Webb, G. and Crooks, A. F. 2005. Metamorphic investigation of the 19, (2), 139–153.
Palaeoproterozoic metasediments of the Willyama Inliers, Willis, I. L., Brown, R. E., Stroud, W. J. and Stevens, B. P. J. 1983.
southern Curnamona Province—a new isograd map, MESA J., The Early Proterozoic Willyama Supergroup: stratigraphic
37, 53–57. subdivision and interpretation of high to low-metamorphic
White, R. W., Pomroy, N. E. and Powell, R. 2005. An in situ rocks in the Broken Hill Block, New South Wales, J. Geol. Soc.
metatexite-diatexite transition in upper amphibolite facies Aust., 30, 195–224.

Demystifing Migmatites

Uploaded by

Copyright:

Available Formats

Demystifing Migmatites

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Demystifing Migmatites

Uploaded by

Copyright:

Available Formats

Applied Earth Science

Transactions of the Institutions of Mining and Metallurgy: Section B

ISSN: 0371-7453 (Print) 1743-2758 (Online) Journal homepage: http://www.tandfonline.com/loi/yaes20

Demystifying migmatites: introduction for field-

M. Pawley, A. Reid, R. Dutch & W. Preiss

To link to this article: http://dx.doi.org/10.1179/1743275815Y.0000000014

Published online: 18 Nov 2015.

Submit your article to this journal

View related articles

Full Terms & Conditions of access and use can be found at

Ñ 2015 Institute of Materials, Minerals and Mining and The AusIMM

Hydrous mineral+ feldsparþquartz ¼ H2O under-

Typically for felsic crustal rocks, this involves muscovite

Muscovite þ quartz ¼ garnet þ biotite þ K-

strongly partitioned into the melt phase, if melt is not

from the system. Thus, melt loss and open system

Components of migmatite typically of a more refractory composition, such as calc-

The leucosome will not necessarily have the composition

rock. The context of the injected melt is important in

Leucosome segregation and migration features

(iii) leucocratic veins or dykes are crystallised ana- Rheology

melt, rather than the break-up of the solid crystal fra-

8 Schlieric diatexite migmatite from Shaw Granitoid

Stromatic metatexite migmatites Diatexites

Schollen diatexite migmatite

Schlieric diatexite migmatite microstructure. Diatexite migmatites are gradational from

geometries, analogous to microstructures in mylonites,

Other information that can be extracted

tolith and metamorphic rims formed during partial

whole before detailed examination is attempted, as some

event when the rocks have cooled or during a later

It is useful looking closely at the leucosomes for solid-state

morphic conditions of 700–800uC and 4.6–5.6 kbar have

a reverse, melt-filled shear band; b close-up illustrating

developed progressively during the same event. Precise

You might also like