An investigation of lower crustal deformation: Evidence for channel ﬂow
and its implications for tectonics and structural studies
Paul F. Williams
, Dazhi Jiang
Department of Geology, University of New Brunswick, Fredericton, NB, Canada E3B 5A3
Department of Earth Sciences, University of Western Ontario, London, ON, Canada N6A 5B7
Received 28 April 2004; received in revised form 8 March 2005; accepted 1 April 2005
Available online 29 June 2005
High-grade metamorphic terranes generally have similar deformation fabrics characterised by a shallowly dipping transposition foliation
and recumbent folds. Deformation paths are typically non-coaxial and strain magnitudes extreme. We refer to this association of structures
and metamorphism as the high-grade nappe association (HGNA), and argue that it is common, and represents crustal-scale (kilometres thick)
shear zones. The structure of such rocks is generally interpreted as comprising stacked thrust sheets separated by thrusts. We concluded,
however, that the ‘thrusts’, which are ubiquitously parallel to the transposition foliation, are not thrusts in the normal sense, but are various
discontinuities, rotated by large pervasive shear strains. The thrust-like appearance may be further enhanced by late localised shear strain.
The sense of shear may be constant across an HGNA shear zone, giving it the geometry and kinematics of a crustal-scale (kilometres thick)
detachment zone. Alternatively it may reverse across the body, consistent with channel ﬂow. Structural evidence therefore supports current
ideas on the behaviour of the middle to lower crust during orogeny, and illustrates the deformation mechanisms involved. We describe the
association with special reference to the Monashee complex, and discuss the implications of our interpretation for the kinematics of high-
strain zones, palinspastic reconstruction, and interpretation of deformation fabrics at various scales.
q2005 Elsevier Ltd. All rights reserved.
Keywords: Shear zone; Channel ﬂow; Strain; Vorticity; Kinematics; Nappe
High-grade regional metamorphic rocks, in our experi-
ence, generally have a horizontal or shallowly dipping
composite foliation (see also Fyson, 1971 and references
therein; Mattauer, 1973) unless the foliation is reoriented by
later structures. Where reoriented, the composite foliation
generally has an enveloping surface that is shallowly
dipping, suggesting that the dip of the foliation was initially
shallow. Steeply dipping high-grade shear zones are an
exception, but they are mostly ﬂanked by shallowly dipping
high-grade rocks, and are rarely voluminous compared with
the latter. These shallowly dipping rocks have similar
fabrics. They are characterised structurally by their
prominent composite foliation, abundant intrafolial folds,
and various kinematic indicators, which indicate that the
foliation developed by transposition during a non-coaxial
deformation. They are the infrastructure rocks of Wegmann
On a large scale, sequences of such rocks may be
divisible into individual nappes, recognisable by differences
in lithostratigraphy or metamorphic grade, or by the
presence of faults or localised high-strain zones (mylonites
and phyllonites), between different groups of rocks. Such
discontinuities are generally parallel to the transposition
foliation. The similarities of fabric lead us to group these
rocks into a high-grade structural association, which we
refer to here as the high-grade nappe association (HGNA).
We believe that many high-grade regional metamorphic
rocks belong to this association.
The discontinuities in rocks of the HGNA are commonly
interpreted as thrusts (e.g. Gee, 1975a,b; Coward, 1980;
Boullier and Quenardel, 1981; McClay and Coward, 1981;
Boyer and Elliott, 1982; Butler, 1982, 1983 with discussion
by Duncan, 1984; Platt, 1984; McDonough and Parrish,
1991; Brown et al., 1992; Gibson et al., 1999), the analogy
Journal of Structural Geology 27 (2005) 1486–1504
0191-8141/$ - see front matter q2005 Elsevier Ltd. All rights reserved.
Corresponding author. Tel.: C1 506 452 6035; fax: C1 506 453 5055.
E-mail address: firstname.lastname@example.org (P.F. Williams).
being drawn with brittle thrusting as exempliﬁed by the
Rocky Mountain thrust and fold belt. Rocky Mountain
thrust and fold belt terms such as ‘sole thrust’, ‘roof thrust’
and ‘duplex’ have been adopted to describe the structure of
some high-grade areas (e.g. in the southern Canadian
Cordillera, Brown et al., 1986; Brown and Journeay, 1987;
McNicoll and Brown, 1995; Crowley, 1999; Gibson et al.,
1999). In HGNA rocks of the Canadian Cordillera, a thrust,
the Monashee de
´collement, has been proposed (Brown,
1980; Read and Brown, 1981; Brown and Read, 1983;
Brown et al., 1986; Journeay, 1986), and interpreted as a
sole thrust toward which all Rocky Mountain thrusts
converge (Brown et al., 1992; Cook et al., 1992). This
implies that the deformation mechanisms and the associated
styles and geometries of structures of the Rocky Mountain
thrust and fold belt are applicable to rocks of the HGNA.
Based on geometrical, mechanical and ﬁeld constraints, we
argue here that this interpretation is fundamentally in error.
Deformation at the mid- to deep-crustal level is character-
ised primarily by penetrative ﬂow, not by stacking of
discrete thrust sheets.
A number of writers have related nappe formation to
ductile thrusting, separating the nappes by the shear zones,
or even allowing internal deformation of the nappes by
shear (e.g. Henderson, 1981; Krabbendam et al., 1997;
Gibson et al., 1999; Treagus, 1999; Rose and Harris, 2000).
We explore this approach further and argue that the shear
strain is important and the whole HGNA fabric evolved in
crustal-scale shear zones. Shear strains are large, but the
thrust-like discontinuities are not mechanisms of defor-
mation, and are not even necessarily the locus of above-
average shear. They are inherited and, in general, involve
far less thrust-like displacement than what is suggested by
the mismatch across the discontinuity. We correlate the
HGNA with the broad zones of detachment and channel
ﬂow being explored by modellers (Davidson et al., 1997;
Beaumont et al., 2001a,b; Jamieson et al., 2002).
After characterising the association, we interpret the
structural development and tectonic setting, and then
discuss some of the implications of our interpretation. We
make special reference to the Monashee complex of the
Canadian Cordilleran Omineca Belt (Gabrielse et al., 1991),
as a typical example of the association. This complex
comprises two structural highs, the northern Frenchman Cap
culmination and the southern Thor–Odin culmination. Our
work is concentrated in Thor–Odin.
2. The high-grade nappe association
2.1. General description
As outlined above, the deﬁning characteristics of the
HGNA are: high-grade metamorphism, the horizontal to
shallow dipping transposition foliation (or its enveloping
surface), a non-coaxial deformation history, and where
divisable into sub units, sheet-like bodies (individual nappes)
with boundaries that are parallel to the transposition foliation.
This broad description fails to cover many details of the
structure as well as the variations that occur from one area to
another. We discuss this more fully here and note that despite
variation, there is an amazing similarity between areas.
The transposition foliation is deﬁned by compositional
layering, which is typically of mixed origin. Some layers
may represent primary compositional variation such as
sedimentary layers, plutons and lava ﬂows. Others represent
dykes and veins that have been intruded at various times
prior to the end of transposition. All are in various stages of
being folded into asymmetrical intrafolial folds (e.g. Fig. 1),
as a result of the progressive deformation. The rocks are
commonly migmatitic, and leucosome is also generally
folded and rotated into the transposition foliation. Orthog-
neisses are generally tabular, parallel to the transposition
foliation and intensely deformed, indicating that their shape
is secondary and a product of the deformation.
Folds in the HGNA are usually abundant and vary greatly
in scale (e.g. Fig. 2). Microscopic to outcrop-scale folds are
generally present, and larger folds including regional-scale
structures occur in some areas (e.g. Fig. 2), but are absent or
difﬁcult to recognise in others. For example, in the
Monashee complex the large folds have amplitudes up to
20 km (Figs. 3 and 4), and various authors (e.g. Thomas,
1979; Treagus, 1999) show folds in the Tay Nappe of
Scotland with amplitudes of the order of 40–50 km. At
Little Broken Hill, Australia, folds with amplitudes
measurable in hundreds of metres have sheath-fold
geometry (Williams, 1967) as do regional folds in the
Foxe fold belt of northern Canada (Henderson, 1981). The
fold hingelines are commonly approximately parallel to a
stretching lineation, or in the case of sheath folds, the
lineation parallels the axis of the sheath.
HGNA fabrics are those typical of shear zones—
environments of large ﬁnite strain accumulated non-
coaxially. There is commonly a grain-shape fabric inclined
to the transposition layering at angles of the order of 308.
This is interpreted as an S-foliation (commonly a steady-
state foliation, cf. Means, 1981), in normal shear-zone usage
´et al., 1979), whereas the transposed layering (S
approximately parallel to the shear plane (C-foliation;
´et al., 1979). Locally, where folds are incompletely
transposed, there may be an axial plane foliation, which is
inclined to both the S-foliation and the transposition
). Kinematic indicators such as C/Srelation-
ships, rotated porphyroblasts/porphyroclasts, fold asymme-
try, shear bands (C0of Berthe
´et al., 1979), and rotated
boudins all indicate a non-coaxial deformation history and
the same sense of vorticity through large parts of, if not
throughout, an entire zone. For example in the Tay Nappe
the sense of rotation of garnets neither changes across the
axial planes of folds (Williams et al., 2000), nor across the
full thickness of the nappe (cf. Krabbendam et al., 1997;
Rose and Harris, 2000). In the Thor–Odin culmination of the
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1487
Monashee complex (Fig. 3) the asymmetry of mesoscopic
folds is constant throughout most of the sequence, as is the
asymmetry of coeval C–S–C0fabrics (Jones, 1959; Johnston
et al., 2000; Spark, 2001).
The discontinuities that separate individual nappes may
be faults or high-strain zones. The latter may be mylonitic or
phyllonitic, and may be of similar metamorphic grade to the
nappes (coeval with the nappe fabric), or may be retrograde
Fig. 1. Example of transposed migmatised gneisses and schists from the Thor–Odin culmination of the Monashee complex, Canadian Cordillera. Larger folds
outlined in the tracing below. The folds shown here are F
and the transposition foliation is present at all scales. It is deﬁned by the alignment of the fold
limbs and discontinuities at all scales; the microscopic foliation, locally visible, is deﬁned by transposed chevrons and crenulations. Person for scale.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041488
Fig. 2. Transposition in the Svartisen area, central Norway (location indicated by star south of Bodø on the inset map), showing variation in the scale of folds related to transposition. (a) Regional synformal
structure of Holandsfjord area after Rutland and Nicholson (1965). Locations of maps ((b) and (c)) are indicated. (b) Detailed map of peninsula between Skarsfjord and Tjongsfjord showing two generations (F
) of tight regional scale folds. Based on mapping of Kuipers (1982). (c) Detailed map (after Williams, 1983) of large mesoscopic folds at the head of Holandsfjord below the Engabreen. Dips of fold limbs
are steep to vertical and folds plunge moderately towards west. (d) Small mesoscopic folds (after Williams, 1983) from the Holandsfjord area.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1489
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041490
(post-nappe fabric). They may contain all the same
structures and show the same deformation history, differing
only in being more strongly transposed. For example in the
Swedish Caledonides (Williams and Zwart, 1977) folds in
the nappes have hingelines that deﬁne a girdle that lies in the
transposition foliation. The more open folds tend to be
perpendicular to a mineral lineation (interpreted as a
stretching direction) and tighter (more evolved) folds are
progressively closer in orientation to the mineral lineation,
the tightest being approximately parallel. In the high-strain
zones between the nappes, fold morphologies and distri-
bution are qualitatively the same. Quantitatively, however,
tighter, and locally sheath folds, parallel or nearly parallel to
the lineation are the most common, indicating a greater
magnitude of strain than in the nappes themselves.
The nappes generally have an extremely thin aspect ratio,
being truly sheet-like. For example in the Caledonides of
central Scandinavia the nappe pile, comprising at least four
major nappes and several smaller ones, is generally only 3 km
thick (maximum thickness 7 km), but covers an area
approximately 250 by 250 km. These nappes are recognisable
because of variation in lithology, stratigraphy and meta-
morphic grade, and have high-strain zones between them
(Gee, 1975a,b). Elsewhere, through-going discontinuities are
less abundant. For example, in the Thor–Odin culmination of
the Monasheecomplex, rocks belonging tothe associationare
more than 10 km thick, but there are no recognisable regional
discontinuities (Reesor and Moore, 1971; Johnston et al.,
2000; Spark, 2001; Kruse et al., 2004) other than late brittle
faults. Regional-scale recumbent isoclinal folds (Figs. 3 and 4;
see also Reesor and Moore, 1971) were interpreted by Duncan
(1984) as nappes, separated by thrusts. Truncation of
structures (Fig. 5) does occur and a lack of continuity in the
folded sequence indicates that discontinuities are present, but
none are regionally persistent and they are themselves folded
locally. They are sharp withthe appearance of brittle faults but
are completely healed (Fig. 5), i.e. there are no fractures
preserved. Exceptional continuity of outcrop makes it possible
to recognise the discontinuities as asymmetrical boudinage
and shear surfaces following S
and C0, at various scales,
rather than discrete nappe-scale faults or shear zones.
We restrict the association to high-grade metamorphic
rocks, including amphibolite facies, high-pressure equiva-
lents and anything of higher temperature. A similar
structural association may occur in some lower-grade
rocks (for example the Morcles Nappe of the Alps (e.g.
Casey and Dietrich, 1997) and the Ko
¨li Nappe of Sweden
(Gee, 1975a,b)), but at lower grades continuity is commonly
better preserved and upright folds and foliations are
generally the norm. In HGNA rocks, metamorphism is
typically late relative to the transposition. For example, near
Blanket Mountain in Thor–Odin (Fig. 3), kyanite crystals
overprint early folds and are folded by late folds, both of
which are associated with transposition, and are then
pseudomorphed by sillimanite, which was still stable during
a later crustal-scale extensional deformation (cf. Johnston
et al., 2000). Elsewhere in the Monashee complex,
assemblages that we associate with the transposition have
been shown to represent pressures at least as high as 8–10 kb
and temperatures around 750–800 8C(Norlander et al.,
2002). They are overprinted by lower pressure (!5 kb)
assemblages at approximately the same temperature (750–
800 8C) (Norlander et al., 2002), which we associate with
the onset of crustal extension. Microstructural observations
indicate that high-temperature conditions outlasted ductile
deformation and there was abundant leucosome develop-
ment throughout the transposition and during the ductile
phase of the crustal extension.
As in the Swedish example above, sequences of rocks
belonging to the HGNA are commonly several kilometres
thick and kilometres to tens or even hundreds of kilometres
in aerial extent. In our experience most regional high-grade
rocks belong to this association (cf. Fyson, 1971 and
references therein). They therefore represent signiﬁcant
volumes of the Earth’s crust. Examples with which we are
familiar include the Caledonian Nappes of central Scandi-
navia (e.g. Trouw, 1973; Williams and Zwart, 1977;
Williams, 1983), the Trans-Hudson–Orogen Kisseynew
domain of Manitoba and Saskatchewan (Lucas et al.,
1994; Norman et al., 1995; Kraus and Williams, 1999),
the Teslin Zone of the Yukon (de Keijzer et al., 1999), the
Thor–Odin culmination of the Monashee complex (Reesor
and Moore, 1971; Johnston et al., 2000), the Little Broken
Hill area of NSW, Australia (Williams, 1967) and the Sesia–
Lanzo of the Italian Alps (Williams and Compagnoni,
1983). In this paper we are concerned primarily with the
high-temperature/medium-pressure as opposed to high-
temperature/ high-pressure members of the association (to
which the Sesia-Lanzo example belongs).
2.2. Typical deformational history of the HGNA: an
example from the Monashee Complex
The deformation history of HGNA rocks in general, as
indicted by overprinting of structures, is monotonously
similar. It involves progressive deformation with several
generations of folds associated with the transposition (e.g.
Central Scandinavia, Little Broken Hill, the Trans-Hudson
Orogen and the Monashee complex; see references above).
We brieﬂy describe the transposition folds of the Thor–Odin
culmination of the Monashee complex as a typical example
of the HGNA.
Fig. 3. Map of the Thor–Odin culmination of the Monashee complex of the Omineca Belt, Canadian Cordillera (based on Kruse et al., 2004). The inset map
shows the location of the whole Monashee complex (MC), comprising Frenchman Cap to the north and Thor–Odin to the south. Only Thor–Odin is represented
in the detailed map. Zones delineated and numbered 1–5 on the inset map represent the Insular Belt, Coast Belt, Intermontane Belt, Omineca Belt and Foreland
Belt of the Canadian Cordillera, respectively. TCH: Trans-Canada Highway. Section line A–B (Fig. 4) and source area of data presented in Fig. 7 are indicated.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1491
Rocks of the Monashee complex can be divided into core
and cover sequences, the base of the cover commonly being
marked by the base of a thick quartzite. The boundary
between the two is interpreted as an unconformity (e.g.
Fyles, 1970; Høy and McMillan, 1979; Journeay, 1986;
Ross and Parrish, 1991) and it is, therefore, signiﬁcant that
the transposition foliation is everywhere parallel to the
unconformity, both above and below. Previously the area
has been divided into an eastern para-authochthon and a
western allochthon (the Selkirk Allochthon) separated by
the Monashee de
´collement (Brown, 1980; Read and Brown,
1981; Brown and Read, 1983; Brown et al., 1986; Journeay,
1986; Gibson et al., 1999; Crowley et al., 2001). However,
extensive mapping across the proposed de
Thor–Odin (cf. Kruse et al., 2004) has failed to reveal any
structural discontinuity other than late brittle faults
(Johnston et al., 2000; Spark, 2001). The transition from
Monashee complex to overlying Selkirk Allochthon along
the west side of Thor–Odin is complicated by the faults
(Fig. 3;Kruse et al., 2004), but in the south where the
sequence is continuous, rocks of the Selkirk Allochthon and
the Monashee complex cover are inter-folded, with the same
transposition fabric present in both, and a gradual change
from Monashee complex below to Selkirk Allochthon
above. One argument for the de
´collement is the presence
on Joss Mountain (Fig. 3) of Devonian intrusives (Johnston
et al., 2000), which are found elsewhere in the Selkirk
Allochthon, but not in the Monashee complex (e.g.
Woodsworth et al., 1991, p. 495) suggesting that the Selkirk
Allochthon is allochthonous. However, in view of the large
amplitude folds (Fig. 4) and the normal faults separating
Joss Mountain rocks from the Monashee complex, it is not
necessary to propose a de
´collement to explain how they
could have been widely separated in the Devonian. Thus, we
interpret the whole of the Monashee complex, and that part
of the Selkirk Allochthon covered by the map (Fig. 3), as a
single crustal-scale zone of transposition. It is still useful to
distinguish rocks west of the normal fault passing through
Greenbush Lake from those to the east of the fault. We
therefore refer to them as the Joss Mountain complex, east
of the fault being the Monashee complex.
In Thor–Odin, three generations of folds are recognised
(Fig. 6). F
folds are tight to isoclinal, generally
dismembered and oriented with their axial surfaces
subparallel to the transposition foliation and their hinges
typically parallel or acutelyinclinedtoastretching
lineation, which forms a partial girdle with its maximum
in the south-west quadrant (Fig. 7c). They commonly occur
as asymmetrical fold pairs and occur in all sizes up to
regional scale structures (Figs. 1, 3, 4, 6 and 8). They may
have a foliation (the transposition foliation) parallel to their
axial plane or may be overprinted by a shape foliation. The
only sure way of distinguishing F
folds is by
overprinting. Where they can be distinguished, the F
(Fig. 6) tend to be more open than F
and their axial surfaces
and hinges more inclined to the local transposition foliation
and stretching lineation, respectively. Map-scale F
appear to have few metre-scale parasitic folds. Small folds
do occur on their limbs but the asymmetry is mostly the
same on both limbs, consistent with interpretation of the
large folds as early (see below) and the small folds as
folds (Figs. 6, 8 and 9) occur at all scales with
amplitudes up to 2 km. They deform the transposition
foliation and generally lack an axial plane foliation. Where
an axial plane foliation is present, it is inclined to the
transposition foliation and generally restricted to symmetri-
cal folds (Fig. 8c), or to the hinges and common limb of
asymmetrical fold pairs. It is not normally developed on
long regional fold limbs, and even where best developed, is
patchy (Fig. 8c). When present it is generally a crenulation
cleavage or is deﬁned by transposed micro-chevrons, and is
progressively rotated into the transposition foliation. F
folds are typically more open than F
folds (Figs. 6, 8 and
9), but there is a gradation between the two, both in
orientation and in style. The tight F
folds are distinguish-
able from F
folds only by virtue of their continuity with
more open F
folds or by overprinting relationships. F
lie on the same partial girdle as deﬁned by F
but spread between the normal to the stretching lineation
and a westerly orientation, which overlaps the spread of the
stretching lineation (Fig. 7b).
Fig. 4. Interpretive section through the south end of Thor–Odin (for position
of section see A–B in Fig. 3). Dark shading represents the cover and light
shading the basement. The dashed line represents the axial surface trace of a
fold, refolded by a regional F
in the centre of the section.
The overall antiformal appearance is due to a north-trending upright
fold. The section passes rapidly from the F
hinge, which it cuts
obliquely, into the steep F
limb (see Fig. 3). This makes the regional F
fold appear almost upright, whereas it actually has an axial surface dipping
shallowly to the SW. To the west the rocks pass into an F
synform and to
the east they pass into the Columbia River Fault. No vertical exaggeration.
Fig. 5. Discontinuities in quartzite (dark grey), laminated feldspathic
quartzite (grey) and calc-silicate (form surface only) rocks from the Thor–
Odin culmination of the Monashee complex. Black areas represent
pegmatite. Outcrop viewed looking west. Note the truncated F
in the calc-silicate unit. The only evidence of the discontinuities is the
observed truncations of fold and foliations.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041492
folds in Thor–Odin (Fig. 9) have long limbs
mostly dipping shallowly in the south-west quadrant, with
rare parasitic F
structures and shorter limbs with abundant
folds and an enveloping surface that dips
steeply in the north-east quadrant. The axial surfaces of
these parasitic F
folds vary considerably in dip, mostly
between vertical, through southwesterly dipping, to
shallowly northeasterly dipping. The steep dips are most
common in the rare parasitic folds on the shallowly dipping
regional fold limbs and recumbent attitudes are most
common on the steep regional limbs (Figs. 8 and 9).
In most of Thor–Odin, asymmetrical F
predominantly north-northeasterly (Jones, 1959; Johnston
et al., 2000). The vergence of F
folds is variable, due to
their rotation towards the stretching lineation. However,
taking into account sense of asymmetry and orientation,
they are consistent with the same vergence, and rare F
sheath folds also verge toward the north-northeast. In the
Joss Mountain complex the same styles and generations of
folds are recognised as in the Monashee complex and the
vergence of F
is the same. F
changes to southwesterly.
The boundaries of this crustal-scale shear zone do not
occur in Thor–Odin or in the Joss Mountain complex.
However, the upper boundary is recognisable in the work of
Campbell (1970) and Murphy (1987), at a higher structural
level, in the Selkirk Allochthon of the neighbouring Cariboo
Mountains. They report upright folds that become progress-
ively recumbent downwards, the structural transition
coinciding with a metamorphic transition from greenschist
to amphibolite facies. A similar situation is reported by
Zwart (1979), from the Pyrenees, at the boundary between
the low grade superstructure and high-grade infrastructure
rocks. Upright folds in the superstructure deﬂect into
recumbent folds of the infrastructure and the latter has the
characteristics of the HGNA.
From the preceding description it is obvious that strain
magnitudes are large in the Monashee complex and the Joss
Mountain complex, but no accurate quantiﬁcation is
possible. However, an impression of the minimum
magnitude can be obtained. For example, in a simple-
progressive deformation, periclinal folds develop
approximately parallel to the vorticity axis with axial planes
approximately perpendicular to the ﬂow direction (Fig. 10).
Fig. 6. Overprinting between F
folds in quartz–feldspar–biotite
migmatitic gneiss, from the Thor–Odin culmination of the Monashee
complex. Black areas represent leucosome of various ages. Outcrop viewed
looking NW. F
fold is seen in proﬁle, other folds seen in oblique sections. Fig. 7. Orientation data (equal area projection) for area immediately NW of
Blanket Mt (see Fig. 3) of the Thor–Odin culmination of the Monashee
complex. (a) Poles to the transposition foliation forming a girdle about a
fold. (b) Plunge of mesoscopic F
folds; these folds
developed as dragfolds about an approximately west-northwest-trending
vorticity vector and have started to rotate towards a south-southwesterly
plunging stretching direction. (c) Stretching lineation spreading between
the original top-to-the-north north-northeasterly ﬂow direction and a late
) west-southwesterly crustal extensional direction. (d) F
various stages of rotation towards the SSW plunging stretching direction.
These folds have z-asymmetry if viewed looking west-northwest. The
stretching lineation is interpreted as developing with a south-southwesterly
plunge parallel to the early, top-to-the-north-northeasterly ﬂow. Metre-
folds mostly developed as dragfolds parallel to the vorticity
vector, and therefore, initially plunged west-northwesterly. The folds
rotated initially towards the stretching lineation, but then locally the
lineation and folds were rotated towards the late (D
) crustal extension
direction. Thus the stretching lineation spreads on a girdle between the
early and late extension directions, but since positive identiﬁcation of folds
is difﬁcult where D
is well developed, there are no fold measurements in
such areas. The plots of F
folds therefore do not reﬂect the D
It is generally sufﬁcient to consider simple shear only, both here and
below. The existence of a pure-shear component only changes the amount
of strain required to rotate pre-existing planes or lines to certain angles with
respect to the shear plane and direction, not the resulting geometry.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1493
For a fold plunging 108(i.e. the fold axis is inclined to the
assumed horizontal vorticity axis by 108), a shear strain of
20 is required to rotate the fold axis to within 168
(a reasonable estimate of the angle commonly observed)
of the projection of the shear direction on the axial plane (cf.
Skjernaa, 1989). The closer the orientation of the initial fold
axis to the vorticity axis, the larger the shear strain required
to rotate it a given amount (Fig. 10b).
Such a shear strain is compatible with the observation
that foliations on both sides of the unconformity in the
Monashee complex are parallel. Even if the initial situation
was one of an angular difference of up to 908across the
unconformity (Fig. 11a), a shear strain O10 will render the
difference too small to recognise in the ﬁeld (Fig. 11d), for
most orientations of the original markers. However, even a
shear strain of 20, in simple shear (maximum principal
stretch of 20.2), is far too small to explain commonly
observed isolated metre scale boudins in large outcrops
). So there is evidence of even larger strains and
we are conﬁdent, therefore, that a shear strain of 20 is a
minimum value, within the limitations of the simplifying
assumptions. Similar estimates can be made for the other
examples of the HGNA listed above. For example in the
outcrops represented in Fig. 2c, isoclinal F
amplitudes an order greater than their wavelength) are
parallel to the lineation and can be traced through a series of
boudins (i.e. both limbs occur in each boudin) where the
lengths of the boudins and distances between them are
2.3. Timing of deformation in the Monashee complex
There has been extensive dating in the Monashee
complex (e.g. Carr, 1992; Parrish, 1995 and references
therein; Crowley and Parrish, 1999; Gibson et al., 1999;
Crowley et al., 2001), and U–Pb dates primarily from zircon
and monazite, are generally interpreted as representing peak
metamorphism and deformation. The dates young down-
wards from w170 Ma at a high structural level in the
Selkirk Allochthon, to w78 Ma at the highest level in the
Monashee complex, to w49 Ma in the core of the Monashee
complex. At the deepest levels of the Monashee complex, in
the Frenchman Cap culmination, the rocks are regarded as
having experienced essentially no deformation since the
Palaeoproterozoic (Crowley, 1999), despite the lack of any
obvious change in fabric, and apparently no metamorphic
peak temperature high enough to reset zircon, monazite and
titanite. The downward younging of peak metamorphism is
explained by progressive burial by thrust sheets advancing
from the west (Parrish, 1995). The Monashee complex,
being the lowest sheet, was the last to be buried (Parrish,
1995; Crowley and Parrish, 1999; Gibson et al., 1999;
Crowley et al., 2001). However, the lack of regional scale
thrusts does not support this interpretation.
We suggest a different interpretation that is consistent
with the structural observations, involving detachment and
channel ﬂow in warm middle to lower crustal rocks,
sandwiched between strong upper crust and strong
Palaeoproterozoic basement. Since deformation is outlasted
by peak metamorphism it may have started considerably
earlier than the peak so that part of the history may not be
represented by the dates (cf. Kuiper, 2003). Initially a broad
detachment zone developed and evolved into a zone of
channel ﬂow. At this stage the upper boundary may have
been migrating upwards (cf. Kuiper, 2003). The channel
tunnelled through cooler rocks (cf. Jamieson et al., 2002, for
example), which became involved in deformation as they
warmed. This, combined with cooling from above due to
progressive unrooﬁng, resulted in both channel boundaries
migrating downwards. Finally, conditions reverted to
detachment ﬂow which by then involved basement rocks,
but ended within the basement, where warm and cool
basement rocks were juxtaposed (Crowley, 1999). We
Fig. 8. F
mesoscopic folds on the steep limb of a regional F
fold in the
Blanket Mt area (see Fig. 3). Both viewed towards the west. (a) and (b)
Overprinting of F
folds by F
. (c) F
folds showing variable tightness
and irregular development of an axial plane foliation.
Fig. 9. F
fold on the east side of Mt Thor (see Fig. 3) overprinting an earlier
fold, viewed towards the west. Visible amplitude of F
estimated at 70 m. Inset drawings show interpretative history of folds. An
early upright dragfold, overprinting an earlier isocline, is modiﬁed by NE
directed simple shear into an overturned structure, with a younger buckle
fold on its steep shortened limb.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041494
suggest that this is the sort of complexity that might be
expected in channel ﬂow zones in general.
3. Development of the HGNA
The salient deformational features that occur in the
HGNA as a group (not necessarily recognised in all
examples), which have to be explained, are as follows:
features found in all examples include (1) a crustal-scale
shear zone characterised by a transposition fabric resulting
from folding before and/or during progressive non-coaxial
deformation, (2) either a constant vergence across the full
width of exposed rocks or a simple reversal, and (3) high-
grade metamorphism that generally continued late relative
to the deformation. Features only found in some examples,
(1) regional folds that are so large that they cannot be
explained by any simple process of buckling or bending in
response to shortening (there is insufﬁcient room in the crust
(cf. Fig. 4), and (2) regional thrust-like discontinuities.
Because the association is commonly interpreted in terms
of the Rocky Mountain thrust and fold belt type thrusting
model, we ﬁrst discuss the inadequacies of this model as
applied to the fabric of HGNA rocks, and then present an
alternative model that is capable of explaining the
3.2. Inadequacies of the foreland thrust and fold belt model
when applied to the HGNA
Although nappes separated by discontinuities can
resemble the structure of a foreland thrust and fold belt,
with the nappes resembling the thrust sheets and the
discontinuities the thrust faults, there are fundamental
differences between the two structural associations aside
from metamorphic grade. The main difference is that Rocky
Mountain thrust and fold belt type thrust sheets are only
weakly deformed, so that continuity of internal primary
structure is preserved, whereas primary structure in the
HGNA is completely transposed, indicating intense internal
deformation. In the Rocky Mountain thrust and fold belt
situation, deformation is generally brittle, and is essentially
concentrated on the thrust plane. In the HGNA, deformation
is penetrative and generally (but not exclusively) ductile,
Fig. 10. (a) Relationship in simple shear, between the initial plunge (a) of an upright fold with axis AF and axial plane ABCF and the ﬁnal orientation of the
passively rotated axis and axial plane, as a function of shear strain (g). The shear plane is AA0BB0and the shear direction AA0. (b) Graph of the angle (b)
between the rotated fold axis (A0F) and the projection of the shear direction on the axial plane of the fold, as a function of initial plunge (a) and shear strain (g).
Fig. 11. Modiﬁcation of upright folds and an angular unconformity by simple shear. At a shear strain as low as 12, all markers are inclined to one another at
sufﬁciently small angles as to make it difﬁcult to recognise non-parallelism in the ﬁeld. (d) Details of the rectangular area indicated in (c); it contains several
fold hinges as well as the unconformity.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1495
and there is not necessarily any recognisable variation in the
intensity of deformation across nappe boundaries. Folds are
not a penetrative feature of the Rocky Mountain thrust and
fold belt structure at outcrop- or smaller-scale. Intrafolial
folds, however, are characteristic of the HGNA, at outcrop-
to micro-scale and commonly at regional scale also. It
follows that, although the aspect ratio of Rocky Mountain
thrust sheets may have changed, they have always been
sheet-like with their long boundaries parallel to original
primary fabric. On the other hand, the original shape of the
HGNA nappes is unknown because of intense folding and
The Rocky Mountain thrust and fold belt model has been
applied to the Monashee complex in Thor–Odin for example
(McNicoll and Brown, 1995), but despite almost continuous
outcrop there are no mappable thrusts separating their
individual horses, nor any fault–bend folds. There are,
however, large isoclinal folds that do not ﬁt the interpret-
ation. Boyer and Elliott (1982) applied the Rocky Mountain
thrust and fold belt model to the Alpine Nappes, but their
interpretation is refuted by Casey and Dietrich (1997).
Similarly, Ramsay (1997) has criticised model-driven
application of Rocky Mountain thrust and fold belt type
thrusting in general, and with particular reference to
development of the ductile fabric of the area of the Moine
Thrust in NW Scotland.
The major problem with applying the Rocky Mountain
thrusting model to high-grade rocks is, therefore, that the
model offers no explanation of the intense deformation
observed within the nappes. In the Rocky
Mountain situation, the thrusts and related structures,
including folds, can all be explained as products of thrusting
and there is a sound mechanical basis for the process (e.g.
Elliott, 1976). However, in the HGNA, the fabric indicates
that all rocks were deforming penetratively during the
deformation process, and therefore changing their aspect
ratio and shape. We are unable to envisage a mechanically
sound process, which involves ductile thrusting, of
internally ﬂowing thin sheets, over one another.
3.3. Alternative interpretation of the HGNA
All features indicate that rocks of the HGNA have
formed in regional-scale (typically kilometres thick and tens
to hundreds of kilometres in extent) horizontal to shallowly
dipping shear zones and we interpret the various structures
in the context of this general deformation environment. We
start by considering what might be expected to happen to
various pre-existing structures if subjected to such defor-
mation (cf. Krabbendam et al., 1997; Williams, 1999).
Fig. 12 shows three simple combinations of folds and
faults before and after applying varying amounts of simple
shear. In each example, pre-existing discontinuities that are
initially inclined to layering are rotated until almost parallel
to layering. Irrespective of their origin (e.g. normal or
transcurrent fault) they develop thrust-like geometry as all
markers are rotated towards parallelism with the shear
plane. With a suitable initial geometry, old layers may be
placed on young (Fig. 12 a and b), so that after a large shear
strain, the whole sequence has the appearance of a stack of
thrust sheets (Fig. 12c). At a shear strain of 20 all original
features are so nearly parallel to one another that any
discrepancy would be difﬁcult to recognise in the ﬁeld.
Fault-drag folds, in a horizontal shear zone, become tight
Fig. 12. Modiﬁcation of early structures by large magnitude simple shear. (a)–(d) Early normal faults (a) are converted progressively (b) to nappe-like
structures (c). Initial ‘fault-drag’ (a) is converted to tight recumbent folds with attenuated limbs between alternate fold-pairs ((c) and (d)). Pre-existing upright
folds (e) are converted to recumbent folds and intervening discontinuities again develop thrust geometry (g). Initial sub-equant markers, such as the shaded
rectangle at the LH end of (e), are converted to thin tabular bodies (g); this is the expected fate of granitoids believed to be the protolith of tabular granitic
gneisses. Alternate limbs of the upright folds (e) undergo continuous stretching or initial shortening followed by stretching. This may lead to secondary
structures such as buckle folds (cf. Fig. 9), boudinage and shear bands (f). Late shearing parallel to the early thrust-like discontinuities and/or parallel to S
produce faults, mylonites or shear zones with thrust geometry (h) and apparent thrust-related fabrics. These features are expected to form as S
parallelism with the shear plane, but are shown earlier here because of the problem of representing larger shear strains.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041496
recumbent folds with alternate limbs apparently attenuated
(Fig. 12d), and pre-existing upright folds (Fig. 12e) are
rotated towards the shear plane to become recumbent
isoclines (Fig. 12g). One limb is in the extensional ﬁeld
throughout shearing, but the other shortens initially.
Consequently, parasitic buckling and an axial plane
foliation can be expected on the latter (Figs. 9 and 12f).
Once the transposition foliation is approximately parallel to
the shear plane, local perturbations can amplify into
dragfolds (cf. Bons, 1993), which then rotate and contribute
to the transposition. Fold hinges rotate progressively
towards the shear direction.
The interpretation of folds as presented here differs
signiﬁcantly from interpretations based on the slip fold
model (Weiss, 1959; Ramsay, 1960) in which the folds are a
product of shear-zone-parallel shear (e.g. Casey and
Dietrich, 1997; Gibson et al., 1999). At the strain
magnitudes believed to pertain in many examples of the
association considered here, such folds would reverse in
vergence and then unfold (e.g. Ramsay et al., 1983). The
inherited folds and dragfolds of our interpretation rotate
towards the shear plane but never truly reach it unless
overprinted and rotated by other structures such as later
dragfolds. Thus, they become tighter and more attenuated
but maintain the same vergence. Another difference
between the two models is that the one presented here
does not require changes in shear sense across every axial
plane as is required by the slip fold model. Our
interpretation is thus consistent with the observation that
the shear sense is constant across many axial planes, through
thick packages of transposed rock.
The gradation in style and orientation as exempliﬁed for
example by the Monashee complex of Thor–Odin indicates
folds are consistent with a progressive
deformation in a ﬂat-lying shear zone. A simplistic
interpretation of the Monashee complex is that F
were inherited from earlier deformation and modiﬁed by
shearing (Figs. 12e–g and 13) to produce a transposition
was then perturbed by F
dragfolds that, on
tightening, further contributed to the development of S
folds developed and deformation stopped before
their limbs had been completely transposed back into S
hingelines began to rotate as they developed, as described
above. The most rotated are the F
folds and the least
rotated are F
represent a progressive
deformation in which early folds have completed the
cycle of perturbation–transposition and late folds have
been frozen in the perturbation stage. In reality, many F
folds may be dragfolds, but in view of the fact that
individual folds can be traced across the upper boundary of
the shear zone, in the Omineca Belt (Murphy, 1987) and in
the Pyrenees (Zwart, 1979) at least some F
folds must be
inherited. They may be much older or may represent an
early stage of the deformation that resulted in development
of the shear zone (see Fig. 13). It is also possible that more
than one generation of folds is inherited.
Two types of HGNA shear zone have been described,
one with constant sense of shear and another across which
the sense of shear reverses. We cannot rule out the
possibility that the latter is the norm and the former simply
an artefact of inadequate exposure. For example, one side of
the shear zone may be eroded or the other may be too deep
to be exposed. However, it seems likely that there is a
progression from one type to the other (see Discussion
below) so that we might expect both to exist. Consequently,
the non-coaxial, high-strain structural characteristics and
the high-grade of metamorphism lead us to interpret the
association in terms of (1) detachment ﬂow within the crust
and/or at the base of the crust, where the sense of shear is
constant, and (2) channel ﬂow (e.g. Beaumont et al., 2001a,b),
where the sense of shear reverses across the channel.
We now discuss the nature and signiﬁcance of
discontinuities that are commonly interpreted as thrusts, in
Despite detailed mapping in the Monashee complex, we
have been unable to ﬁnd through-going thrust-like dis-
continuities. Discontinuities occur at various scales but
there are no regionally persistent surfaces that dissect the
complex into thrust sheets or horses. It is possible, however,
to arbitrarily divide the complex into various nappes deﬁned
by individual regional recumbent folds (cf. Fig. 3). If it were
Fig. 13. End-member behaviour during shortening of the crust (white) over
upper mantle or strong crust (grey). This represents for example, what is
happening to the crust and mantle in Fig. 15b. (a) Initial conﬁguration. (b)
The situation assuming a frictionless boundary between crust and strong
substrate. The crust deforms in bulk pure shear by folding and potentially
reverse faulting. (c) and (d) The situation assuming that the crust is attached
to mantle. In (c) both weak crust and substrate deform and in the end
member they have the same strength and deform equally. In (d) the
substrate is rigid and the weak crust deforms in bulk pure shear near the
surface, but has to undergo a non-coaxial deformation at depth in order to
maintain continuity with the substrate. Early folds develop as upright
structures, but locally they are modiﬁed by shear to form overturned folds.
(e) Simpliﬁed representation of initial and ﬁnal form of the weak crust,
according to the end member represented in (d). See text for further
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1497
not for the continuity of glacier-polished outcrop, one might
imagine that these folds were separated by through-going
discontinuities since whole segments of various units are
truncated. However, because of the exceptional outcrop we
are able to see that discontinuities, though numerous, are
local. It may be that this is a more common situation than
realised, because few regions are as well exposed. Of those
listed above that we are familiar with, only the Svartisen
area is comparable and it also lacks through-going
discontinuities in the area mapped.
We have shown that where thrust-like discontinuities do
exist, they may be a product of the interaction of the strain
and pre-existing discontinuities. This implies that such
structures have existed throughout the shear-zone defor-
mation, and we would therefore expect that they would
themselves be folded in areas such as the Monashee
complex, where large dragfolds are present. There is,
however, the possibility that faults develop during the shear-
zone deformation as large shear-band type structures or if
the shear zone is widening (in the sense of Means (1995)),
they might develop as reverse faults. Either would
ultimately be transposed into a thrust-like structure. Some
faults or high-strain zones may develop parallel to the shear
Some discontinuities have mylonite fabrics consistent
with their development parallel to the shear plane, but many
are ambiguous. However, if a mylonite is retrograde for
example, it cannot have been inherited and preserved
through high-grade metamorphism. Nevertheless, we
suggest that these are not evidence of thrusting in the
normal sense, but rather indicate reactivation of an earlier
surface or zone that has been rotated into approximate
parallelism with the shear plane (Fig. 12h). The early
features may be, for example, inherited faults, shear zones,
lithological contacts, or weak lithological units that have
acted as planes or zones of weakness. In shear-zone
terminology these would be C-orY-shears. The shear strain
associated with the mylonite might be small, but because of
the pre-existing discontinuity the structure would have the
appearance of a major thrust. The mylonite below the Glarus
nappe of the Swiss Alps may be an example of such a
Schmid (1975) concluded that an unreasonable strain rate
would have been necessary for the mylonite below the
Glarus nappe to accommodate nappe emplacement, given
the known constraints. He discussed various possible
solutions to the problem, including the possibility that
shear deformation was not restricted to the narrow mylonite,
but occurred across a broader zone. An alternative
possibility is that the Glarus thrust is a rotated earlier
discontinuity and the mylonite is due to later reactivation.
The geometry described, with cleavage inclined to the
discontinuity and with folds parallel to the stretching
lineation close to the discontinuity and perpendicular to
the lineation further away, is consistent with the HGNA,
despite the lower (greenschist) metamorphic grade.
In view of all the possibilities we believe that there is no
basis for interpreting discontinuities between nappes as
thrusts. This has implications for palinspastic reconstruc-
tions as discussed below.
4.2. Implications for timing of deformation
The concept of diachroneity of fold generations is not
new (e.g. Hobbs et al., 1976, p. 355; Williams and Zwart,
1977) and diachroneity is likely to be acute in shear zones
where local rates of deformation may vary considerably and
where the boundaries of the zone may be migrating,
progressively or intermittently, into or out of the adjacent
country rock. We ﬁnd it useful to use the designations F
, etc. where the generations are identiﬁed by overprinting,
but stress that F
in one outcrop can be younger than F
, in another (Williams, 1985). The folds have
developed progressively and F
maturity of development as opposed to a less mature F
Where shear-zone boundaries migrate through time and
activity varies in rate from place to place within the shear
zone, there will be no simple relationship between maturity
and absolute age.
In environments such as Benioff zones it is conceivable
that the same non-coaxial ﬂow may be generated repeatedly
in different packages of rocks. Thus, adjacent packages may
be deformed at very different times, but the kinematic
picture may be the same for each, and thus, the fabric is the
same in appearance and orientation. Similarly, if horizontal
zones of ductile detachment or channel ﬂow occur deep in
orogens in general, as suggested by the abundance of
HGNA rocks, it is possible that a detachment or channel
ﬂow fabric generated by one mountain building episode
could conceivably be reactivated by a later one. If the whole
zone was reactivated, evidence of the earlier ﬂow might be
annihilated. However, if only part of the zone was
reactivated it is possible that fabrics of very different age
would be geometrically indistinguishable. A further
complication might occur where, as a channel approaches
the surface, plug-like ﬂow might develop due to extreme
localization of deformation into channel margins, with a
central portion of the channel being carried by the
surrounding ﬂow without internal deformation.
4.3. Implications for kinematic vorticity determination in
Our observation that the compositional layering in
HGNA is a transposition foliation has implications for the
kinematic vorticity determination in shear zones. All
vorticity number determination methods, besides other
assumptions (e.g. Jiang and White 1995), rely on the
existence and recognition of a ﬂow extensional eigenvector
(fabric attractor of Passchier (1997)). Compositional
layering in shear zones is commonly assumed to represent
the ﬂow extensional eigenvector, but since the layering is
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041498
generally a transposition foliation (S
) this is unjustiﬁed. In
the case of a rotating S
toward the shear zone boundary
orientation, the total vorticity with respect to the shear zone
boundary is the sum of the vorticity relative to S
vorticity accommodated by the spin of the layer with respect
to the zone boundary. Therefore, the kinematic vorticity
number determined from the rock fabric assuming that S
the fabric attractor, might be expected to be signiﬁcantly
less than the bulk kinematic vorticity number. This may
explain the low kinematic vorticity numbers consistently
obtained by many authors (Simpson and De Paor, 1997;
Xypolias and Koukouvelas, 2001; Bailey et al., 2004;
Grasemann et al., 1999; Law et al., 2004). These authors
have interpreted the low kinematic vorticity numbers as
indicating large values of pure shear component in natural
shear zones. We regard these kinematic vorticity numbers as
not being representative of the bulk kinematics of the zone.
4.4. Implications for palinspastic reconstructions
If ‘thrusts’ in HGNA rocks do not necessarily have
signiﬁcant movement on them, there are important
consequences for palinspastic reconstruction. A simple
comparison between our interpretation and the thrust model
is made in Fig. 14, where homogeneous simple shear is
assumed across the zone. The difference between the
original width and the ﬁnal width of the zone is greater
for the normal klippe-to-fenster (e.g. Hobbs et al., 1976,
p. 309) method of reconstruction than it is for the ﬂow
model presented here (compare Fig. 14c (ii) with c (iii)).
Pre-nappe-formation shortening is ignored in both models,
even though early folding is probably a pre-requisite of the
ﬂow model. Unfortunately, we have no way of addressing
this problem, other than to note that in the ﬂow model a
shortening in excess of 0.5 is probably required for
development of the folds.
The klippe-to-fenster method always gives the larger
value for the original width of the zone (Fig. 14), but all
other values being equal, the difference between the two
models decreases as the aspect ratio increases. For a given
aspect ratio, increasing gor N(Fig. 14) also increases the
difference between the predictions of the two models. The
predictions of the klippe-to-fenster model can exceed those
of the ﬂow model by an order of magnitude, but for wide
zones combined with small values of gand N, the difference
can be negligible. However, gis difﬁcult to determine and
unless it can be shown to be small, there will be
considerable uncertainty in palinspastic reconstruction, in
addition to that which generally exists, due to imperfect
knowledge of the geometry of the orogenic zone.
For example, a 200 km (W
) wide zone with a thickness
(H) of 10 km, combined with 10 nappes and gZ10
(Fig. 14b (ii)), according to the thrust model, gives an
initial width of 1105 km, compared with an initial width of
100 km predicted by the ﬂow model. Even if a pre-ﬂow
shortening of 75% is assumed for the ﬂow model, the initial
width would only be 400 km. Increasing the shear to 20
almost doubles the initial width predicted by the thrust
model, so that W
Z2102 km as opposed to the more
realistic estimate of 400 km.
Another problem for palinspastic reconstruction is one of
provenance. If the thrusts associated with nappes are in fact
inherited discontinuities, it becomes important from the
point of view of reconstructing provenance to know their
early history. For example, if a ‘thrust’ is really a rotated
transcurrent fault, any interpretation based on the assump-
tion that it is a thrust is misleading. There may have been
considerable mass movement perpendicular to the direction
of detachment or channel ﬂow. In addition, a history
involving varying amounts of detachment ﬂow and channel
ﬂow, which are likely to be difﬁcult to quantify, makes
interpretation of provenance difﬁcult, even if there is no
movement perpendicular to the ﬂow direction.
4.5. Tectonic implications
The HGNA is common and it is argued that regional
high-grade metamorphic rocks in general belong to the
association. High-pressure members of the association
probably develop in Benioff zones (cf. Fyson, 1971;
Mattauer et al., 1981) where the prime requirements of
non-coaxial ﬂow, large strains and high-grade metamorph-
ism exist. Most members, however, are too hot (relative to
pressure) to develop in Benioff zones and we suggest that
they are evidence of the subhorizontal ﬂow of the middle to
Fig. 14. Palinspastic reconstruction: comparison of penetrative ﬂow versus
thrust-stacking origin of nappes. Each part shows a section through a block
of crust divided by discontinuities into 10 units with nappe-like geometry
(ii). According to the ﬂow model proposed here, the present geometry
resulted from shear modiﬁcation of the original geometry. (i) Is obtained by
reversing the shear. (i) Is thus the restoration of (ii) according to the ﬂow
model. Restoration of the section according to the thrust-stacking model is
achieved by reversing the thrusts individually (iii), according to the
klippen-to-fenster method (i.e. each nappe is placed end to end with its
neighbour. Only one thrust is restored in (b (iii)) because of space
limitations; all are restored in (c (iii)). Aspect ratios and shear strains are as
follows: (a) ARZ5, gZ10. (b) ARZ10, gZ10. (c) ARZ50, gZ10. The
original width of the deformed zone (W
), for the ﬂow model, is given by
is the ﬁnal width of the deformed zone), and for
the thrust model, by WOZðN2H2CðgNHKgHCWPÞ2Þ1=2. Comparison
of (a)–(c) indicates the effect of aspect ratio on the reconstruction according
to the thrust model.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1499
lower crust that has been postulated by modellers (e.g.
Beaumont et al., 2001a,b). We believe that the structural
evidence in the HGNA overwhelmingly indicates that the
middle to lower crust is generally weak on the orogen scale,
and that it is a locus of ﬂow within the lithosphere. This does
not preclude the possibility of the detachment zone
extending down to, or even into the mantle, or of the
existence of other zones, but it does indicate that part of the
crust is generally involved.
Mountain chains are a product of crustal shortening, and
geological evidence indicates that the shortening is achieved
at shallow levels by upright folding and reverse faulting,
generally under greenschist facies or lower conditions. As
we go deeper into middle to lower crust amphibolite facies
conditions, we generally move into a regime of shallowly
dipping fabrics (cf. Fyson, 1971; Mattauer, 1973, p. 185).
This is consistent with shear resulting from shortening of the
crust over strong basement crust or mantle (Fig. 15b). At
shallow levels shortening is by a bulk-pure-shear process,
involving upright folding and reverse faulting, and at deep
levels there is an increasing inﬂuence of the comparatively
strong mantle. Fig. 13 shows three end member situations.
In (b) the boundary between the crust and mantle is
frictionless and the whole crust deforms in bulk pure shear.
In (c) the mantle is no stronger than the crust and both
shorten in pure shear. In (d) there is perfect coupling
between the crust and mantle, and the mantle is rigid. If any
end member applies it is likely to be (d), but more likely is
some combination of the three. In this detachment situation,
the shear will be localised in any rocks that are signiﬁcantly
weaker than average. Thus, it could be concentrated
anywhere in the crust rather than immediately above the
mantle (cf. Jamieson et al., 2002). Typically, the weak rocks
will be high-temperature and/or wet rocks, but we submit
that the shear must occur whatever the conditions, so long as
there is shortening of the crust. This may explain why
strains and strain paths typical of the HGNA are found in
some low-grade rocks (e.g. the Morcles and Ko
If the contrast in strength is sufﬁciently high and the crust
Fig. 15. Tectonic model for the development of detachment ﬂow and channel ﬂow. (a) Factors inﬂuencing the geometry of the orogen. (b) Tectonic erosion or
downward ﬂow of the mantel, and localised thickening of the crust result in detachment shear at the base of the crust. (c) An increase in convergence rate and
input of crustal material result in further thickening of the crust and increased thermal weakening of its base. This results in the evolution of detachment ﬂow
into channel ﬂow. (d) A decrease in convergence rate and reduction in input of crustal material results in collapse of the orogen and channel ﬂow revertsto
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041500
is thick enough (Fig. 15c), detachment ﬂow may progress
into a form of extrusive ﬂow, or channel ﬂow (Beaumont
et al., 2001a,b), as evidenced by the reversal of shear in the
Himalaya (Burchﬁel et al., 1992; Grujic et al., 1996) and the
Monashee complex. The fact that the vergence of folds in
the Monashee only changes in the late stages of shear
deformation (i.e. during F
, but not during F
consistent with this interpretation. It is signiﬁcant in this
regard that high temperatures, high pressure, and develop-
ment and emplacement of melt are synchronous with
transposition (see above). Subsequently, a slowing down
of the convergence rate or a reduction in the amount of
material being accreted will result in collapse and thinning
of the crust and channel ﬂow might be expected ultimately
to revert to detachment ﬂow (Fig. 15d) driven by the
This interpretation is consistent with the modelling of
Beaumont et al. (2001a). The detachment stage is
represented, for example, by tZ7.5 My in ﬁg. 4a of
Beaumont et al. (2001a) where the shear sense is the same
across the full width of the high-strain zone (F
Monashee complex and Selkirk Allochthon). As the slab
thickens, the shear zone develops into a zone of channel
ﬂow (e.g. tZ52.5 My in ﬁg. 5a of Beaumont et al., 2001a)
so that locally the sense of shear reverses (F
in the upper
Monashee complex and the Selkirk Allochthon).
A potential test of the detachment model is to check
observed minimum-strain magnitudes against model pre-
dictions. Fig. 13e diagrammatically shows the detachment
model in the form that maximises shear strain. For (1) a
shortening of 60%, which combined with erosion would
keep the crust to a reasonable thickness, (2) a ﬁnal
detachment zone thickness of 10 km, as estimated for the
Monashee complex, and (3) a shear strain of 42 at E
(Fig. 13e) decreasing linearly to 0 at D (Fig. 13e) to give
comparable values to those recorded in the Monashee
complex; the length of the zone (ED) is 700 km. If we
assume that E represents the approximate centre of the
mountain chain, the full width of the chain is w1400 km.
Further, the minimum shear strain suggested for the
Monashee complex (gZ20) would only exist out as far as
350 km from the centre. Compared with observations, the
width of the orogen is large, the pure shear shortening is
reasonable, but the shear is minimal, suggesting that for the
model to be viable, there has to be some other factor such as
tectonic erosion of the mantle at the Benioff zone. Once
channel ﬂow develops there is not the same dependency of
shear-strain magnitude on the width of the mountain chain
and the pure-shear shortening.
As modelled by Beaumont et al. (2001a,b), the channel
ﬂow is much better developed on the ‘pro-side’ (i.e. on the
subducting plate side) than on the ‘retro-side’ of the orogen.
However, there is probably sufﬁcient ﬂow on either side to
explain the developmentof the HGNA. Further, by varying the
boundary conditions, different variations on this theme are
possible. Beaumont et al. (2001a,b) have shown, for example,
the importance of erosion (see also Zeitler et al., 2001)in
enhancing the ﬂow. Also, in their model, shortening of the
crust on the retro-side of the suture is superﬁcial and does not
involve shortening of the lower crust and mantle. This is a
model constraint, and if it were to be relaxed, for example by
allowing tectonic erosion of the mantle below the hanging-
wall plate at the suture, channel ﬂow could be greatly
increased on the retro-side of the orogen.
deformation of the Monashee complex and the
Joss Mountain complex has been interpreted in terms of
extrusive ﬂow previously, which in terms of local
kinematics can be identical to channel ﬂow. However, as
envisaged, the extrusion is a static process, in which a
wedge of rock is simply squeezed out, without loss of
continuity, deforming and thinning as it goes. The problem
with this is that the driving force required to keep the
thinning wedge moving has to increase rapidly (e.g. Jaeger,
1964, pp. 140–143). The channel ﬂow model (Beaumont
et al., 2001a,b) includes the bigger picture, and because of
the thickening of the crust and continuous feeding of
material into the channel; there is no need for thinning. In
fact, the channel may thicken. Thus, the channel ﬂow model
is a dynamic model involving ongoing crustal shortening
and it is believed to be a better representation of the general
From a metamorphic point of view both detachment and
channel ﬂow can explain the inverted metamorphic
sequences commonly observed in orogenic regions (e.g.
Spear, 1993; Jamieson et al., 1996). Detachment ﬂow may
transport warm rocks from the interior of the orogen over
cooler rocks, towards its margins. In channel ﬂow, warm
rocks may ‘tunnel’ through cooler rocks (e.g. Jamieson
et al., 2002). Signiﬁcantly, inverted gradients are reported in
the Monashee complex (Journeay, 1986) and in the Joss
Mountain complex and Selkirk Allochthon gradients are
normal (Campbell, 1970; Murphy, 1987; Johnston, 1998).
In qualitative terms, therefore, temperature decreases sym-
metrically away from the centre of the proposed channel.
The HGNA is a product of crustal-scale shear zones,
characterised by horizontal to shallowly dipping trans-
position foliation, non-coaxial deformation histories with a
constant shear sense across much if not all of the zone, high-
grade metamorphism, and in some examples regional-scale
recumbent folds and/or nappes. It is common and rocks
belonging to the association represent a signiﬁcant volume
of orogenic mountain belts. We suggest that many high-
grade metamorphic rocks belong to the HGNA.
Nappes are generally explained in terms of thrust
tectonics with thrusting on discontinuities or narrow shear
zones. However, such an interpretation ignores much of the
structural evidence (speciﬁcally the internal transposition
fabric), which it is incapable of explaining.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1501
The HGNA is best explained in terms of large-strain
crustal-scale shear zones in which the axial planes of earlier
(inherited and/or transposition related) upright folds, and
various discontinuities, are rotated towards the shear plane.
The folds are progressively ampliﬁed so that their ultimate
amplitude is more closely related to shear-strain magnitude
than to the initial folding process. Earlier discontinuities
develop thrust geometry without the necessity of any dis-
placement appropriate to thrusting. Thus the nappes and
their related ‘thrusts’ are structures modiﬁed in the HGNA
shear zones rather than the cause of the HGNA fabric.
Structural evidence indicates that crustal-scale shear
zones not only occur as Benioff zones but are common-
place in lower pressure environments. This supports current
ideas of channel ﬂow. These shear zones start as detachment
zones, and if there is sufﬁcient thickening of the crust and
rocks become weak enough, channel ﬂow develops where
the rocks are weakest. This typically coincides with
migmatitic conditions. Channel ﬂow may later revert to
detachment ﬂow, due to late collapse of the orogen.
In some realistic situations detachment ﬂow is a neces-
sity and, therefore, is not necessarily restricted to high-grade
rocks. This explains the existence of fabrics similar to those
described from the HGNA in Greenschist facies rocks.
Earlier discontinuities or lithological contacts, once
rotated close to the orientation of the shear plane, may
undergo C-(Y-) shear and thus give rise to late mylonites. In
the situation where the discontinuity is a pre-shear fault, the
mylonite is not likely to accurately reﬂect the total dis-
placement that has occurred on the ‘thrust’. It is unlikely,
however, that the discrepancy will be recognisable in general.
It is conceivable that HGNA shear zones will be active
over long periods of time or that the same boundary
conditions will re-occur at the same or different levels, at
different times. This could result in extreme diachronism of
structures that would be grouped, on the basis of style and
overprinting, into common generations.
Compositional layering, being in general a transposition
foliation in high-strain zones, cannot be assumed to
represent the ﬂow extensional eigenvector of the zone.
Making such an assumption in determining the kinematic
vorticity number of natural shear zones may underestimate
the bulk kinematic vorticity number of the zone and lead to
incorrect conclusions regarding the signiﬁcance of the pure-
shear component in the zone.
Palinspastic reconstructions based on the assumption that
the thrust-like geometry is a product of thrusting, are likely,
at best, to be highly misleading if applied to mid- to lower-
We are grateful to Stefan Kruse, Paul McNeill and
Yvette Kuiper for discussion and critical reading of the
manuscript, to Andy Parmenter and Sharon Carr for
discussion, to Dick Brown for introducing us to the area
and problems, and to Mike Williams for a constructive
review. We thank Angel Gomez for assistance with drafting.
The research was ﬁnanced by an NSERC PDF to DJ and an
NSERC research grant to PFW.
Bailey, C.M., Francis, B.E., Fahrney, E.E., 2004. Strain and vorticity
analysis of transpressional high-strain zones from Virginia Piedmont,
USA. In: Alsop, G.I., Holdsworth, R.E., McCaffrey, K.J.W., Hand, M.
(Eds.), Flow Processes in Faults and Shear Zones Special Publications,
vol. 224. Geological Society, London, pp. 249–264.
Beaumont, C., Jamieson, R.A., Nguyen, M.H., Lee, B., 2001a. Mid-Crustal
channel ﬂow in large hot orogens: results from coupled thermal-
mechanical models. In: Cook, F., Erdmer, P. (Eds.), Slave—Northern
Cordillera Lithospheric Evolution (SNORCLE) and Cordilleran
Tectonics Workshop; Report of 2001 Combined Meeting Lithoprobe
Report, 79, pp. 112–170.
Beaumont, C., Jamieson, R.A., Nguyen, M.H., Lee, B., 2001b. Himalayan
tectonics explained by extrusion of a low-viscosity channel coupled to
focused surface denudation. Nature 414, 738–742.
´, D., Choukroune, P., Jegouzo, P., 1979. Orthogneiss, mylonite and
non-coaxial deformation of granites: the example of the South
Armorican shear zone. Journal of Structural Geology 1, 31–42.
Bons, P.D., 1993. Experimental deformation of polyphase rock analogues.
Geologica Ultrajectina: Mededelingen van de Faculteit Aardwe-
tenschappen der Universiteit Utrecht 110, pp. 1–207.
Boullier, A.-M., Quenardel, J.-M., 1981. The Caledonides of northern
Norway: relation between preferred orientation of quartz lattice, strain
and translation of the nappes. In: McClay, K.R., Price, N.J. (Eds.),
Thrust and Nappe Tectonics Geological Society Special Publication,
vol. 9, pp. 185–195.
Boyer, S.E., Elliott, D., 1982. Thrust Systems. Bulletin of the American
Association of Petroleum Geologists 66, 1196–1230.
Brown, R.L., 1980. Frenchman Cap dome, Shuswap complex, British
Columbia: a progress report. Current Research, Part A, Geological
Survey of Canada Paper 80-1A, pp. 47–51.
Brown, R.L., Journeay, J.M., 1987. Tectonic denudation of the Shuswap
metamorphic terrane of southeastern British Columbia. Geology 15,
Brown, R.L., Read, P.B., 1983. Shuswap Terrane of British Columbia: a
Mesozoic core complex. Geology 11, 164–168.
Brown, R.L., Journeay, J.M., Lane, L.S., Murphy, D.C., Rees, C.J., 1986.
Obduction, backfolding and piggyback thrusting in the metamorphic
hinterland of the southeastern Canadian Cordillera. Journal of
Structural Geology 8, 255–268.
Brown, R.L., Carr, S.D., Johnson, B.J., Coleman, V.J., Cook, F.A.,
Varsek, J.L., 1992. The Monashee de
´collement of the southern
Canadian Cordillera: a crustal-scale shear zone linking the Rocky
Mountain foreland belt to lower crust beneath accreted terranes. In:
McClay, K.R. (Ed.), Thrust Tectonics. Chapman and Hall, London, UK,
Burchﬁel, B.C., Chen, Z., Hodges, K.V., Liu, Y., Royden, L.H., Deng, C.,
Xu, J., 1992. The South Tibetan Detachment system: extension
contemporaneous with and parallel to shortening in a collisional
mountain belt. Geological Society of America, Special Paper 269.
Butler, R.W.H., 1982. A structural analysis of the Moine Thrust Zone
between Loch Eriboll and Foinaven, NW Scotland. Journal of
Structural Geology 4, 19–29.
Butler, R.W.H., 1983. Balanced cross-sections and their implications for
deep structure of the northwest Alps. Journal of Structural Geology 5,
Campbell, R.B., 1970. Structural and metamorphic transitions from
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041502
infrastructure to suprastructure, Cariboo Mountains, British
Columbia. In: Wheeler, J.O. (Ed.), Structure of the southern
Canadian Cordillera Special Paper, vol. 6. The Geological
Association of Canada, pp. 67–72.
Carr, S.D., 1992. Tectonic setting and U–Pb geochronology of the early
Tertiary Ladybird leucogranite suite, Thor–Odin–Pinnacles area,
southern Omineca Belt, British Columbia. Tectonics 11, 258–278.
Casey, M., Dietrich, D., 1997. Overthrust shear in mountain building. In:
Sengupta, S. (Ed.), Evolution of Geological Structures in Micro- to
Macro-scales. Chapman and Hall, London, pp. 119–142.
Cook, F.A., Varsek, J.L., Clowes, R.M., Kanasewich, E.R., Spencer, C.S.,
Parrish, R.R., Brown, R.L., Carr, S.D., Johnson, B.J., Price, R.A., 1992.
Lithoprobe crustal reﬂection cross-section of the southern Canadian
Cordillera 1. Foreland thrust and fold belt to Fraser River fault.
Tectonics 11, 12–35.
Coward, M.P., 1980. The Caledonian thrust and shear zones of NW,
Scotland. Journal of Structural Geology 2, 11–17.
Crowley, J.L., 1999. U–Pb geochronologic constraints on Paleoproterozoic
tectonism in the Monashee complex, Canadian Cordillera: elucidating
an overprinted geologic history. Bulletin of the Geological Society of
America 111, 560–577.
Crowley, J.L., Parrish, R.R., 1999. U–Pb isotopic constraints on
diachronous metamorphism in the northern Monashee complex,
southern Canadian Cordillera. Journal of Metamorphic Geology 17,
Crowley, J.L., Brown, R.L., Parrish, R.R., 2001. Diachronous deformation
and a strain gradient beneath the Selkirk allochthon, northern Monashee
complex, southeastern Canadian Cordillera. Journal of Structural
Geology 23, 1103–1121.
Davidson, C., Grujic, D.E., Hollister, L.S., Schmid, S.M., 1997.
Metamorphic reactions related to decompression and synkinematic
intrustion of leucogranite, High Himalayan Crystallines, Bhutan.
Journal of Metamorphic Geology 15, 593–612.
de Keijzer, M., Williams, P.F., Brown, R.L., 1999. Kilometre-scale folding
in the Teslin zone, northern Canadian Cordillera, and its tectonic
implications for the accretion of the Yukon–Tanana terrane to North
America. Canadian Journal of Earth Sciences 36, 479–494.
Duncan, I.J., 1984. Structural evolution of the Thor–Odin gneiss dome.
Tectonophysics 101, 87–130.
Elliott, D., 1976. The motion of thrust sheets. Journal of Geophysical
Research 81, 949–963.
Fyles, J.T., 1970. Structure of the Shuswap metamorphic complex in the
Jordan River area, northwest of Revelstoke, British Columbia. In:
Wheeler, J.O. (Ed.), Structure of the Southern Canadian Cordillera
Special Paper, vol. 6. The Geological Association of Canada, pp. 87–98.
Fyson, W.K., 1971. Fold attitudes in metamorphic rocks. American Journal
of Science 270, 373–382.
Gabrielse, H., Monger, J.W.H., Wheeler, J.O., Yorath, C.J., 1991.
Morphogeological belts, tectonic assemblages, and terranes. In:
Gabrielse, H., Yorath, C.J. (Eds.), Geology of the Cordilleran Orogen
in Canada Geological Survey of Canada, Geology of Canada, vol. 4,
pp. 15–59 (also Geological Society of America, The Geology of North
America, v. G-2).
Gee, D.G., 1975a. A tectonic model for the central part of the Scandinavian
Caledonides. American Journal of Science 275, 468–515.
Gee, D.G., 1975b. A geotraverse through the Scandinavian Caledonides—
¨stersund to Trondheim. Sveriges Geologiska Underso
¨kning C, A
Gibson, H.D., Brown, R.L., Parrish, R.R., 1999. Deformation-induced
inverted metamorphic ﬁeld gradients: an example from the southeastern
Canadian Cordillera. Journal of Structural Geology 21, 751–767.
Grasemann, B., Fritz, H., Vannay, J.C., 1999. Quantitative kinematic ﬂow
analysis from the Main Central Thrust Zone (NW Himalaya, India):
implications for a decelerating strain path and the extrusion of orogenic
wedges. Journal of Structural Geology 21, 837–853.
Grujic, D., Casey, M., Davidson, C., Hollister, L.S., Ku
¨ndig, R., Pavlis, T.,
Schmid, S., 1996. Ductile extrusion of the Higher Himalaya Crystalline
in Bhutan: evidence from quartz microfabrics. Tectonophysics 260,
Henderson, J.R., 1981. Structural analysis of sheath folds with horizontal
X-axes, northeast Canada. Journal of Structural Geology 3, 203–210.
Hobbs, B.E., Means, W.D., Williams, P.F., 1976. An Outline of Structural
Geology. Wiley, New York.
Høy, T., McMillan, W.J., 1979. Geology in the vicinity of Frenchman Cap
gneiss dome. In: Geological ﬁeldwork 1979. British Columbia
Department of Mines and Petroleum Resources, Paper, 1980-1,
Jaeger, J.C., 1964. Elasticity Fracture and Flow. Methuen, London.
Jamieson, R.A., Beaumont, C., Hamilton, J., Fullsack, P., 1996. Tectonic
assembly of inverted metamorphic sequences. Geology 24, 839–842.
Jamieson, R.A., Beaumont, C., Nguyen, M.H., Lee, B., 2002. Interaction of
metamorphism, deformation and exhumation in large convergent
orogens. Journal of Metamorphic Geology 20, 9–24.
Jiang, D., White, J.C., 1995. Kinematics of rock ﬂow and the interpretation
of geological structures, with particular reference to shear zones.
Journal of Structural Geology 17, 1249–1265.
Johnston, D.H., 1998. Structural and thermal evolution of the northwest
Thor–Odin dome, Monashee complex, southeast British Columbia.
PhD thesis, University of New Brunswick.
Johnston, D.H., Williams, P.F., Brown, R.L., Crowley, J.L., Carr, S.D.,
2000. Northeastward extrusion and extensional exhumation of crystal-
line rocks of the Monashee complex, southeastern Canadian Cordillera.
Journal of Structural Geology 22, 603–625.
Jones, A.G., 1959. Vernon Map Area, British Columbia, Geological Survey
of Canada, Memoir 296 1959.
Journeay, J.M., 1986. Stratigraphy, internal strain, and thermo-tectonic
evolution of northern Frenchman Cap dome, and exhumed basement
duplex structure, Omineca hinterland, southeastern Canadian Cordil-
lera. PhD thesis, Carleton University.
Krabbendam, M., Leslie, A.G., Crane, A., Goodman, S., 1997. Generation
of the Tay Nappe, Scotland, by large-scale SE-directed shearing.
Journal of the Geological Society, London 154, 15–24.
Kraus, J., Williams, P.F., 1999. Structural development of the Snow Lake
Allochthon and its role in the evolution of the southeastern Trans-
Hudson Orogen in Manitoba, central Canada. Canadian Journal of Earth
Sciences 36, 1881–1899.
Kruse, S., McNeill, P., Williams, P.F., 2004. Geological map of the Thor–
Odin dome, southern Monashee complex, BC, http://www.unb.ca/
Kuiper, Y.D., 2003. Isotopic constraints on timing of deformation and
metamorphism in the Thor–Odin dome, Monashee complex, south-
eastern British Columbia. PhD thesis, University of New Brunswick.
Kuipers, T.T., 1982. De endogene geologie van het schiereiland tussen
Skarsfjord en Tjongsfjord, Noorwegen. Doctoral thesis, University of
Law, R.D., Searle, M.P., Simpson, R.L., 2004. Strain, deformation
temperatures and vorticity of ﬂow at the top of the Greater Himalayan
Slab, Everest Massif, Tibet. Journal of the Geological Society, London
Lucas, S.B., White, D., Hajnal, Z., Lewry, J., Green, A., Clowes, R.,
Zwanzig, H., Ashton, K., Schledewitz, D., Stauffer, M., Norman, A.,
Williams, P.F., Spence, G., 1994. Three-dimensional collisional
structure of the Trans–Hudson Orogen, Canada. Tectonophysics 232,
Mattauer, M., 1973. Les de
´formations des mate
´riaux de l’e
Mattauer, M., Faure, M., Malavieille, J., 1981. Transverse lineation and
large-scale structures related to Alpine obduction in Corsica. Journal of
Structural Geology 3, 401–409.
McClay, K.R., Coward, M.P., 1981. The Moine Thrust Zone: an overview.
In: McClay, K.R., Price, N.J. (Eds.), Thrust and Nappe Tectonics
Geological Society Special Publication, vol. 9, pp. 241–260.
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–1504 1503
McDonough, M.R., Parrish, R.R., 1991. Proterozoic gneisses of the Malton
Complex, near Valemount, British Columbia: U–Pb ages and Nd
isotopic signatures. Canadian Journal of Earth Sciences 28, 1202–1216.
McNicoll, V.J., Brown, R.L., 1995. The Monashee de
´collement at Cariboo
Alp, southern ﬂank of the Monashee complex, southern British
Columbia, Canada. Journal of Structural Geology 17, 17–30.
Means, W.D., 1981. The concept of steady state foliation. Tectonophysics
Means, W.D., 1995. Shear zones and rock history. Tectonophysics 247,
Murphy, D.C., 1987. Suprastructure–infrastructure transition, east-central
Cariboo Mountains, British Columbia: geometry, kinematics and
tectonic implications. Journal of Structural Geology 9, 13–29.
Norlander, B.H., Whitney, D.L., Teyssier, C., Vanderhaeghe, O., 2002.
Partial melting and decompression of the Thor–Odin dome, Shuswap
metamorphic core complex, Canadian Cordillera. Lithos 61, 103–125.
Norman, A.R., Williams, P.F., Ansdell, K.M., 1995. Early Proterozoic
deformation along the southern margin of the Kisseynew gneiss belt,
Trans–Hudson Orogen: a 30 million year progressive deformation
cycle. Canadian Journal of Earth Sciences 32, 875–894.
Parrish, R.R., 1995. Thermal evolution of the southeastern Canadian
Cordillera. Canadian Journal of Earth Sciences 32, 1618–1642.
Passchier, C.W., 1997. The fabric attractor. Journal of Structural Geology
Platt, J.P., 1984. Balanced cross-sections and their implications for deep
structure of the northwest Alps: discussion. Journal of Structural
Geology 6, 603–606.
Ramsay, J.G., 1960. The deformation of early linear structures in areas of
repeated folding. Journal of Geology 68, 75–93.
Ramsay, J.G., 1997. The geometry of a deformed unconformity in the
Caledonides of NW Scotland. In: Sengupta, S. (Ed.), Evolution of
Geological Structures in Micro- to Macro-Scales. Chapman and Hall,
London, pp. 445–472.
Ramsay, J.G., Casey, M., Kligﬁeld, R., 1983. Role of shear in development
of the Helvetic fold-thrust belt of Switzerland. Geology 11, 439–442.
Read, P.B., Brown, R.L., 1981. Columbia River fault zone: southeastern
margin of the Shuswap and Monashee complexes, southern British
Columbia. Canadian Journal of Earth Sciences 18, 1127–1145.
Reesor, J.E., Moore Jr., J.M., 1971. Petrology and structure of Thor–Odin
gneiss dome, Shuswap metamorphic complex, British Colombia.
Geological Survey of Canada, Bulletin 195.
Rose, P.T.S., Harris, A.L., 2000. Evidence for the Lower Palaeozoic age of
the Tay Nappe: the timing and nature of Grampian events in the Scottish
Highland sector of the Laurentian Margin. Journal of the Geological
Society, London 157, 381–391.
Ross, G.M., Parrish, R.R., 1991. Detrital zircon geochronology of
metasedimentary rocks in the southern Omineca Belt, Canadian
Cordillera. Canadian Journal of Earth Sciences 28, 1254–1270.
Rutland, R.W.R., Nicholson, R., 1965. Tectonics of the Caledonides in part
of Nordland, Canadian Cordillera, Norway. Quarterly Journal of the
Geological Society, London 121, 73–109.
Schmid, S.M., 1975. The Glarus Overthrust: ﬁeld evidence and mechanical
model. Eclogae Geologicae Helvetiae 68, 247–280.
Simpson, C., De Paor, D.G., 1997. Practical analysis of general shear zones
using the porhyroclast hyperbolic distribution method: an example from
the Scandinavian Caledonides. In: Sengupta, S. (Ed.), Evolution of
Geological Structures in Micro- to Macro-scales. Chapman and Hall,
London, pp. 69–184.
Skjernaa, L., 1989. Tubular folds and sheath folds: deﬁnitions and
conceptual models for their development, with examples from the
Grapesvare area, northern Sweden. Journal of Structural Geology 11,
Spark, R.N., 2001. Crustal thickening and tectonic denudation within the
Thor–Odin culmination, Monashee complex, southern Canadian
Cordillera. PhD thesis, University of New Brunswick.
Spear, F.S., 1993. Metamorphic phase equilibria and pressure-temperature-
time paths, Monograph Series. Mineralogical Society of America,
Thomas, P.R., 1979. New evidence for a Central Highland root zone. In:
Harris, A.L., Holland, C.H., Leake, B.E. (Eds.), The Caledonides of the
British Isles Special Publication, vol. 8. Geological Society of London,
Treagus, J.E., 1999. A structural reinterpretation of the Tummel Belt and a
transpressional model for evolution of the Tay Nappe in the Central
Highlands of Scotland. Geological Magazine 136, 643–660.
Trouw, R., 1973. Structural geology of the Marsfja
¨llen area, Caledonides of
¨sterbotten, Sweden. Sveriges Geologiska Underso
Wegmann, C.E., 1935. Zur Deutung der Migmatite. Geologischen
Rundschau 26, 305–350.
Weiss, L.E., 1959. Geometry of superposed folding. Bulletin of the
Geological Society of America 70, 91–106.
Williams, P.F., 1967. Structural analysis of the Little Broken Hill area of
New South Wales. Journal of the Geological Society of Australia 14,
Williams, P.F., 1983. Large scale transposition by folding in northern
Norway. Geologische Rundschau 78, 589–604.
Williams, P.F., 1985. Multiply deformed terrains—problem of correlation.
Journal of Structural Geology 7, 269–280.
Williams, P.F., 1999. Fold Nappes and the problem of model driven
interpretation. Specialist Group in Tectonics and Structural Geology
Field Conference, Halls Gap, Victoria, Australia (February 14–19,
1999). Geological Society of Australia Abstracts Series 53, 267–268.
Williams, P.F., Compagnoni, R., 1983. Deformational and metamorphic
history of the Sesia Lanzo Zone around Bard, Val D’Aosta. Journal of
Metamorphic Geology 1, 117–140.
Williams, P.F., Zwart, H.J., 1977. A model for the development of the
Seve-Koli Caledonian nappe complex. In: Saxena, S.K.,
Bhattacharji, S. (Eds.), Energetics of Geological Processes. Springer,
Berlin, pp. 169–187.
Williams, P.F., Stringer, P., Jiang, D., 2000. Garnet rotation and the
interpretation of the Tay Nappe. In: Skilbeck, C.G., Hubble, T.C.T.
(Eds.), Understanding Planet Earth: Searching for a Sustainable Future
Abstracts of the 15th Australian Geological Convention, University of
Technology, Sydney, July 3–7, vol. 59, p. 542.
Woodsworth, G.J., Anderson, R.G., Armstrong, R.L., with contributions
from Struik, L.C. and van der Heyden, P., 1991. Plutonic regimes. In:
Gabrielse, H., Yorath, C.J. (Eds.), Geology of the Cordilleran Orogen in
Canada. Geological Survey of Canada, Geology of Canada 4, pp. 15–59
(see also Geological Society of America, The Geology of North
America, v. G-2).
Xypolias, P., Koukouvelas, I.K., 2001. Kinematic vorticity numbers and
strain rate patterns associated with ductile extrusion in the Chelmos
shear zone (External Hellenides, Greece). Tectonophysics 338, 59–77.
Zeitler, P.K., Meltzer, A.S., Koons, P.O., Craw, D., Hallet, B.,
Chamberlain, C.P., Kidd, W.S.F., Park, S.K., Seeber, L., Bishop, M.,
Schroder, J., 2001. Erosion, Himalayan Geodynamics, and the
Geomorphology of Metamorphism. GSA Today 11, 4–9.
Zwart, H.J., 1979. The geology of the central Pyrenees. Leidse Geologische
P.F. Williams, D. Jiang / Journal of Structural Geology 27 (2005) 1486–15041504