1.1. The Connemara Metamorphic Complex

The bedrock geology of Connemara consists of a fold and thrust sequence, known as the Connemara Metamorphic Complex (CMC) (Leake & Tanner Reference Leake and Tanner1994), which is intruded by a series of Siluro-Devonian granitoids (Fig. 1). The hanging wall of the CMC consists of Dalradian strata that are cross-cut and metamorphosed by Ordovician orthogneiss and metagabbros associated with the Grampian Orogeny (Friedrich et al. Reference Friedrich, Hodges, Bowing and Martin1999a, Reference Friedrich, Bowring, Martin and Hodgesb). This sequence of rocks was thrust southwards over the Delaney Dome Formation along the Mannin Thrust during the closure of the Iapetus Ocean (Leake et al. Reference Leake, Tanner, Singh and Halliday1983; Leake Reference Leake1986; Tanner et al. Reference Tanner, Dempster and Dickin1989). The CMC underwent five phases of deformation (D1–D5) before initiation of late Caledonian magmatism (Leake & Tanner Reference Leake and Tanner1994); D4 and D5 being of most significance to the current work. Strain imparted during D4 formed the regional scale Connemara Antiform, a gently ESE-plunging fold across central Connemara. D5 formed N–S-trending open folds and is associated with substantial shearing along large NNW–SSE and NNE–SSW conjugate faults. Here, reactivation of D5 faults during the Silurian is associated with concurrent regional sinistral strike slip (Leake & Tanner Reference Leake and Tanner1994).

1.2. The Galway Granite complex defined

The Siluro-Devonian Galway granites, often referred to as the late Caledonian Galway Granites (Graham et al. Reference Graham, Feely and Callaghan2000; Suzuki et al. Reference Suzuki, Feely and O'Reilly2001; Selby et al. Reference Selby, Creaser and Feely2004), refer to a suite of granitoids that were emplaced into the CMC between 425 Ma and 380 Ma (Feely et al. Reference Feely, Selby, Hunt and Conliffe2010). These include the ‘Main Batholith’ and associated ‘Satellite Plutons’; i.e., the Omey, Inish, Letterfrack and Roundstone plutons (Fig. 1). The ‘Main Batholith’ is composed of two distinctive parts. An oval shaped NNW–SSE oriented intrusive body, recently described as the “Western Ring Complex of the Galway Granite Batholith” (Leake Reference Leake2011), forms the western portion of the Main Batholith (Fig. 1(5)). Molybdenum Re–Os geochronology shows this part of the pluton to be ∼410 Ma (Selby et al. Reference Selby, Creaser and Feely2004). To the east, a larger (∼1200 km²), younger, ESE–WNW orientated oval intrusion forms the larger portion of the Main Batholith (Fig. 1(6)). Zircon U–Pb geochronology shows this part of the pluton to be ∼400 Ma (Feely et al. Reference Feely, Coleman, Baxter and Miller2003). The observations of Max et al. (Reference Max, Long and Geoghegan1978), Wright (Reference Wright1964) and Leake (Reference Leake2011) also report compositional and structural differences between the east and west parts of the Main Batholith. Therefore, the ‘Main Batholith’ is here defined as a composite batholith of two discrete parts; the earlier ‘Carna Pluton’ in the west (Fig. 1(5)) and the adjoining ‘Kilkieran Pluton’ to the east (Fig. 1(6)).

A misleading phrase used in reference to the Galway Granites is ‘Satellite Plutons’. The prefix ‘satellite’ implies that these plutons intruded around, and were inherently satellites to, the Main Batholith. This is not the case, as the Main Batholith is ∼410–380 Ma (Feely et al. Reference Feely, Coleman, Baxter and Miller2003, Reference Feely, Selby, Hunt and Conliffe2010) and the ‘Satellite’ plutons are understood to be ∼425–420 Ma (Elias et al. Reference Elias, Macintyre and Leake1988; Buchwaldt et al. Reference Buchwaldt, Kroner, Toulkerides, Todt and Feely2001; Feely et al. Reference Feely, Selby, Conliffe and Judge2007). Therefore, the use of the term ‘satellite’ is incorrect and unnecessarily confusing. We propose that the Inish, Letterfrack, Omey and Roundstone plutons should be referred to collectively as the ‘Earlier Plutons’ as this term accurately depicts these as standalone plutonic bodies that predate the Main Batholith. Magmatism in the Galway Granite Complex (GGC) initiated during Caledonian regional transpression (Dewey & Strachan Reference Dewey and Strachan2003; Soper & Woodcock Reference Soper and Woodcock2003) and terminated 10 Ma after the Acadian Orogeny ( Meere & Mulchrone Reference Meere and Mulchrone2006; Woodcock et al. Reference Woodcock, Soper and Strachan2007). Therefore, magma was generated and emplaced concurrently with two major orogenic events over a period of at least 40 Ma. Thus, we suggest that the suite of Silurian–Devonian granites in Connemara, be referred to as the ‘Galway Granite Complex’, as the current data set suggest it is not all of Caledonian origin.

2.1. Facies definition

Following Townend (Reference Townend1966), detailed remapping of the intrusion (Fig. 2) has resulted in considerable modifications in terms of facies definition, distribution and contact relationships. The Omey Pluton is a composite intrusion consisting of three main facies, G1, G2 and G3, previously termed the Omey Adamellite, the Aughrus More Adamellite and the Island Adamellite respectively (Townend Reference Townend1966). The Glassillaun Granodiorite, previously described as a sub-facies of G1 (Townend Reference Townend1966), is now recognised as a series of E–W-striking porphyritic microgranite dykes hosted within G1 and not a distinct facies of the pluton (Fig. 2). Townend (Reference Townend1966) also described the petrology and nature of lamprophyre and hornblende plagioclase porphyritic dykes which occur throughout the pluton, trending NNW–SSE or WSW–ENE and cross-cutting the granite and which, therefore, are <422 Ma.

Omey Island and a large proportion of the Aughrus Peninsula are composed of the G1 facies which makes up ∼75 % of the surface area of the pluton (Fig. 2). An array of small rocky sea stumps and islands lie 100–1000 m offshore to the W and N; the majority of these are also composed of G1. In the east, near Barnahallia Lough, G1A is defined by a notable increase in the abundance of 3-mm prismatic dark green hornblende (2 %). South of Aughrusbeg Lough ([055770, 258230]) and north of Tonashindilla Pier ([055790, 256650]), a circular body of G3 (diameter ∼1.2 km) forms the core of the Aughrus Peninsula and the highest point within the pluton, Aughrus More (146 m, [056475, 257626]). In the west, Cruagh Island and Friar Island are composed predominantly of G3. G2 is only found between G1 and G3. The largest exposures of G2 skirt around the NW and NE perimeter of the large exposure of G3 in the centre of the Aughrus Peninsula. The G2 facies is also present on the eastern tip of both Friar Island and Cruagh Island, and immediately north of Tonashindilla Pier (Fig. 2).

G1 is a medium-grained (8-mm) equigranular biotite hornblende monzogranite with rare 2 cm-long tabular phenocrysts of zoned plagioclase with seriticised calcic spikes (Fig. 3a, b). Rock mode is 30 % alkali feldspar, 34 % oligoclase-andesine, 30 % quartz, 3 % biotite, >2 % hornblende and 1 % accessories. Overall, this facies contains 72 SiO₂ wt.% (Leggo et al. Reference Leggo, Compston and Leake1966). Typically, subhedral K-feldspar occurs as 10-mm crystals amongst slightly finer-grained quartz and biotite. Prismatic titanite and zircon, along with apatite and rutile, are common accessory minerals.

Figure 3 Characteristic petrographic features of facies G1 and G2. Seriticised calcic spikes within plagioclase (Plag) (a, b) and poikilitic K-feldspar (Ortho) crystals which exhibit inclusions of 0.5–2-mm quartz (Qtz), euhedral feldspar and oxides which accreted to, or nucleated on, the former outer margins of the phenocryst (c, d). G2 is distinguished from G1 by the presence of larger K-feldspars (c) and rounded quartz and plagioclase phenocrysts (e, f).

G2 is a medium-grained biotite hornblende monzogranite and is very similar to G1, having a mode of 29 % alkali feldspar, 33 % andesine, 34 % quartz, 3 % biotite and 1 % accessories. G2 is finer grained (∼6 mm), contains a lower abundance of hornblende (0–1 %) and exhibits very rare large phenocrysts of tabular K-feldspar up to 4 cm long. K-feldspar phenocrysts exhibit inclusions of quartz, biotite and plagioclase which are arranged and aligned in concentric zones within the crystal and against its faces via synneusis (Fig. 3c, d) (Vance Reference Vance1969). Quartz and feldspar commonly occur as rounded 6–8-mm grains (Fig. 3e, f), or as slightly elongate lobes which, with biotite, define subtle foliations when fabrics are present.

G3 is a medium- to fine-grained (∼2–6 mm) pink quartz-phyric monzogranite (Fig. 4). Accessory minerals include zircon, apatite, rutile and fluorite. Rock mode is 30 % K-feldspar, 32 % oligoclase, 35 % quartz, 2 % biotite, >1 % muscovite and >1 % accessories. Biotite is far less abundant, muscovite is present but rare and hornblende is absent; oligoclase is the dominant feldspar. Magnetite is present in lower abundances than those observed in earlier facies and a higher proportion of hematite is present (see ). This facies has a slightly higher SiO₂ content (75 SiO₂ wt.% (Leggo et al. Reference Leggo, Compston and Leake1966)) and fewer mafic minerals than G1 and G2, and is interpreted to represent a more evolved magma. The G3 facies characteristically exhibits a porphyritic texture defined by ≤15-mm tabular plagioclase and orthoclase (Fig. 4a, b) and ≤12-mm studs of anhedral or rounded quartz (Fig. 4c) set in a 2–5-mm monzogranite groundmass. Quartz phenocrysts exhibit minor undulose extinction and are approximately spherical. Fluorite associated with biotite layering on Cruagh Island is associated with a concentration of late-stage residual fluids rich in incompatible elements near the upper portions of the intrusion (Townend Reference Townend1966).

Figure 4 Characteristic petrographic features of G3. Porphyritic texture defined by ≤15-mm tabular plagioclase (Plag) and perthitic orthoclase (Ortho) (a, b) and ≤12-mm studs of anhedral or rounded quartz (Qtz) (c). Seriticised plagioclase is concentrated along calcic rich cores and calcic zones (d).

All facies plot towards the lower right of the monzogranite field, with a general trend from G1 to G3 becoming more quartz rich and less mafic. G1 and G2 are texturally and petrographically quite similar and are best distinguished by the presence of larger K-feldspars (Fig. 3c, d) and rounded quartz in G2 (Fig. 3e), and by a higher abundance of hornblende in G1. In both cases, prismatic green hornblende (∼2–4 mm) is typically fresh, while biotite (2–6 mm) is almost always euhedral (outside of shear zones) and extensively chloritised. Plagioclase is partially altered and commonly extensively replaced by sericite (Fig. 3a). A broad variety of zoning textures are noted in both sodic and potassium-rich feldspars, including oscillatory, normal and rarely boxy zoning (Figs 3a, b, 4d). K-feldspar regularly exhibits a dramatic micro-perthite texture with thin exsolution lamellae of albite flames, which may be parallel or irregularly orientated (Fig. 3c). Although rare, the conversion of orthoclase to microcline is recorded and spatially associated with minor wedged twins in plagioclase, kinking and smearing of biotite. These features are associated with syn-post emplacement strain, a topic discussed further below.

2.2. External contact relationships

The external contacts of the pluton are best described relative to the axial trace of the D4 Connemara Antiform (Fig. 2). South of the axial trace, from Barnahallia southwest to Fountainhill Beach ([058293, 255265]) and on Cruagh Island, the granite is in contact with the steeply inclined (∼80°) southward-dipping limb. North of the axial trace, from Barnahallia northwest to Rossadillask Beach ([058064, 259035]) the granite is in contact with the moderately inclined (∼55°) northward-dipping limb of the antiform. Within the proximity of the fold hinge, at Barnahallia in the east and on Friar Island in the west, the granite is in contact with the gently eastward-plunging (∼10°) axial trace.

Exposures of the western granite contact are limited to Cruagh Island and Friar Island (Fig. 2). On Cruagh Island, bedding dips steeply to the S (∼75°) and strikes WNW–ESE; while on Friar Island, bedding dips 10–25° to the E and strikes NNE–SSW. Thus, Cruagh Island and Friar Island are the offshore representations of the southern limb and fold hinge of the Connemara Antiform respectively, although Friar Island probably sits slightly south of the actual fold hinge. Host rock strata is currently stratigraphically unassigned (Leake & Tanner Reference Leake and Tanner1994; Long & McConnell Reference Long and McConnell1995), and is dominated by well-bedded pelites and psammites. Townend (Reference Townend1966) described the presence of a narrow quartzite unit and several quartz pebble bed horizons on Cruagh Island; an observation supported here. Similarities between the quartz conglomerate units observed here and observations made by Kilburn et al. (Reference Kilburn, Shackleton and Pitcher1965) at Cleggan Head ([059600, 259900]), suggest that the stratigraphic units on these Islands can be correlated with the Cleggan Boulder Bed Formation.

2.2.1. Southern limb

The granite contact extends along an E–W trajectory from Fountain Hill Beach ([058260, 255260]) to Grallaghan ([059060, 255270]), before swinging north toward Barnahallia (Fig. 2). Here, the Streamstown Schist Formation (Badley Reference Badley1976; Treloar Reference Treloar1977, Reference Treloar1982; Leake Reference Leake1986; Leake & Tanner Reference Leake and Tanner1994) forms the bounding country rock.

Immediately south of Fountain Hill (Fig. 2), a well-defined steeply-dipping contact parallels the southward-dipping southern limb of the Connemara Antiform. Several intrusions of G1 and G1A extend from the pluton up to 50 m into the host rock, exploiting joints and compositional layering that ultimately define sheets which are largely concordant to the folded symmetry of the country rock (Fig. 5a, b). Abrupt undulations in the granite–country rock contacts in this area occur parallel to regional joint sets (NNW–SSE, WNW–ESE), and illustrate that localised discordant injection of magma was facilitated by pre-existing joint planes. These features are regarded as broken bridges (Hutton Reference Hutton2009; Schofield et al. Reference Schofield, Heaton, Holford, Archer, Jackson and Jolley2012), where magma wedging preferentially exploits a vulnerable plane of weakness (bedding) and periodically sheets discordantly along a second pre-existing anisotropy (joints) (Fig. 5c–e).

Figure 5 The southeastern pluton contact is, overall, concordant to the Dalradian country rock which dips steeply to the south (a, b). The folded stratigraphy is locally transgressed by sheets of magma which exploit joints and stope off small fragments of schist and psammite (c–e); these features are regarded as broken bridges (as defined by Hutton Reference Hutton2009).

Owing to better topographic relief and exposure, the sheeted nature of the granite contact is best observed in Cruagh Island. Situated on the southern limb of the Connemara Antiform, steeply-dipping (70–60° S) units of the Cleggan Boulder Bed Formation are intruded by concordant sheets of G3 which penetrate up to 50m into the country rock. Some large (1–5 m) detached blocks are present and local subtle chilled margins are observed.

2.2.3. Along the fold hinge

At the southeast granite contact, the axis of the Connemara Antiform is located just north of Barnahallia Lough. At this locality, the inclination of country rock bedding is slightly steeper than that observed further east along the fold hinge (10–20° vs. 5–10°). This implies that bedding has been vertically displaced at this locality, probably as a result of bifurcation of the host rock during inflation of granite sheets as magma intruded. At Barnahallia Lough, the Streamstown Schist Formation lies in direct contact with the Omey Pluton, while to the NW at [059240, 256450] the same can be said for the Lakes Marble Formation (Fig. 2). It is, therefore, clear that intruding granite cross-cuts the Streamstown Schist Formation at map scale (Fig. 2).

Due to abrupt topographic relief and good exposure, the attitude of the pluton's contact near the fold hinge of the Connemara Antiform is best observed on Friar Island (Fig. 6). The majority of the granite bedrock is G3; however, a small proportion of the eastern bedrock is part of the G2 facies (Fig. 2). The contact between G2 and G3 on Friar Island is gradational. An overall E–W reduction in grain size occurs toward the country rock contact, along with increasing numbers of small (usually cm-m scale) stoped blocks. A weak contact parallel foliation, steeply inclined to the west, is recorded. In the western extremity of the island, the Cleggan Boulder Bed Formation dips gently (∼10°) to the east and strikes NNE–SSW. As observed at Barnahallia Lough in the east, the inclination of the country rock bedding increases slightly (∼15–20°) eastwards toward the granite contact. At the contact, sheets of granite, some over 4 m thick, are plainly observed intruding into and above units of the Cleggan Boulder Bed Formation and along inclined joints (Fig. 6a, b). A schematic vertical profile looking south shows bedded schists, overlain by a thick granite sill, then a 3-m sequence of schists which is again overlain by a granite sheet (Fig. 6c). This suggests that vertical displacement of the gently inclined country rock was achieved in order to accommodate the lateral intrusion and thickening of granite sheets.

Figure 6 (a, b) Gently (10–20° E) eastward-inclined units of the Cleggan Boulder Bed Formation on Friar Island (see Fig. 2) show G3 intruded into and over the Cleggan Boulder Bed Formation, concordant to the primary schistosity and discordantly along inclined joint sets, along the fold axis of the Connemara Antiform. (c) A schematic section (looking south), depicting the perceived nature of the granite contact.

2.3. Internal contacts

Facies contacts are gradational, with the exception of some minor G3 sheets (∼1 m) that cross-cut G2 to the NE of Aughrus More; these show that G3 post-dates G2. The G1–G3 contact can be observed toward the western end of the Aughrus Peninsula at [055193, 257203], where a gradational contact over 1 m dips gently toward the west (<5°). Critically, G3 lies on top of G1, suggesting that G3 was emplaced sub-horizontally over G1. A gradational contact over ∼4 m is observed between G2 and G3 on Cruagh Island, Friar Island and near the summit of Aughrus More.

To the east and west of Aughrus More, mapped G1–G2 and G2–G3 contacts are approximately contour-parallel (Fig. 2), and so appear to be flat-lying. To the north and south of Aughrus More, mapped gradational contacts cross-cut topography and dip gently to the north and south, respectively. An ENE–WSW depression between Aughrus More and Aughrusbeg Lough reveal that G1–G2/G2–G3 contacts dip moderately to the west in this area.

2.4. Fabric development and sub-magmatic deformation

Characteristically weak, laterally discontinuous foliations defined by K-feldspar and euhedral platy biotite are observed sporadically within the pluton and define an overall concentric foliation (Fig. 2). This foliation is observed in G1 and G2, but not in G3. No systematic increase in fabric intensity is recorded in proximity to pluton margins. In thin section, quartz is typically anhedral and may exhibit weak undulose extinction. Chloritised biotite is euhedral, feldspars are idiomorphic and essentially unstrained, twinning in plagioclase is straight and only very rare examples of wedged twins are found. These features are characteristic of a dominantly magmatic or high-temperature weak sub-magmatic deformation, indicating that the observed fabrics formed prior to and during crystallisation (Vernon Reference Vernon2004; Passchier & Trouw Reference Passchier and Trouw2005). These foliations are interpreted as pluton inflation fabrics which resulted from the continued injection of magma into the site of emplacement.

Townend (Reference Townend1966) identified two distinctive fault sets (NNW–SSE and E–W) which cross-cut all pluton facies (Fig. 2). Remapping has identified several prominent NNW–SSE and E–W topographic lows that cut through Aughrus More, Omey Island and define lakes, scarps and the shape of the coastline about the Omey Pluton (Fig. 2). E–W-trending faults are denoted by undulations in topography and sometimes weak fabric development. Porphyritic microgranite sheets which show variably unchilled and chilled margins intrude these structures (as do Carboniferous dolerite dykes and the lamprophyre dykes described by Townend (Reference Townend1966)).

NNW–SSE faults are more prominent and are denoted by minor intrusions and ∼5 m-wide zones of moderate to strong sub-vertical NNW–SSE foliation (Fig. 7a). No consistent sense of shear has been identified along these faults. Marginal NNW–SSE sub-vertical foliations are defined by elongate quartz, aligned or smeared biotite and partially aligned feldspars, and intensify toward the centre of a shear zone. Brittle deformation occurs at the core of these structures and is expressed as intense, closely-spaced micro-fractures. In thin section, quartz ribbons are composed of multiple sub-grains which exhibit moderate undulose extinction, and biotite is variably kinked and smeared around euhedral feldspar phenocrysts (Fig. 7c). Feldspars incorporated into the foliation exhibit weak undulose extinction, orthoclase is sometimes partially converted to microcline and fractures are infilled with primary magmatic quartz (Fig. 7e, f). Such features are characteristic of sub-magmatic to solid-state deformation (Vernon Reference Vernon2004).

Figure 7 (a) A NNW–SSE shear zone from the northwestern portion of the Omey Pluton (see Fig. 2). (b) A moderate petrofabric within the shear zone parallels fine-scale micro fracturing, giving rise to a strong textural anisotropy. Petrographic observations show quartz ribbons (Qtz) and smeared biotite (Bt). (c) Marginal dynamic recrystallisation of plagioclase (Plag) along the foliation plane (d). Sub-magmatic deformation indicated by brittle fractures in biotite (e) and plagioclase (f), with primary quartz and biotite infill.

The field relationships and petrographic descriptions described show that the shear zones were established before the granite had cooled. These observations imply a concurrent relationship between shearing, which occurred mainly along NNW–SSE faults but also along E–W faults, and pluton construction.

3.1. AMS results

Results from a total of 113 out of 121 sample sites were accepted (Table 1); data from eight sites were rejected owing to localised hydrothermal overprinting. A summary of AMS data compiled from each facies is presented in Table 2. A review of the parameters used to evaluate the AMS tensor can be found in Tarling & Hrouda (Reference Tarling and Hrouda1993); here, the principal parameters used to characterise magnetic anisotropy are:

(1) $$\rm K_m = [K1 + K2 + K3] / 3$$

(2) $$\rm P_j = \{\exp \sqrt 2 [( \box 1 - \box )2 + ( \box 2 - \box ) 2 + ( \box 3 - \box ) 2]\}$$

(3) $$\rm T_j = [2 \box 2 - \box 1 - \box 3 ] / [\box 1 - \box 3 ]$$

(4) $$\rm H = [K1 - K3] / K_{mean}$$

where K_m is the mean susceptibility, P_j and T_j are the Corrected Anisotropy Degree and the Shape Factor respectively (Jelinek Reference Jelinek1981) and H is the Total Anisotropy (Owens Reference Owens1974). The T_j parameter is used to distinguish oblate (0<T_j<1) from prolate fabrics (0>T_j>−1) (Jelinek Reference Jelinek1981). The shape of highly oblate AMS tensors (i.e., a magnetic foliation) can be visualised as a flattened ellipsoid in the K1–K2 plane (K3≪K2 ≈ K3). In this scenario, sub-samples return comparable K3 axes orientations (normal to the plane), while K1–K2 axes are distributed along the plane of flattening. Where the AMS tensor is highly prolate (K1≫K2≈K3), sub-specimen K1 susceptibility axes cluster and a girdle may be defined by K2 and K3 axes. Both a lineation and foliation can be determined where all principal susceptibility axes are well defined. If a composite fabric is present, or where an intermediate fabric occurs due to structural overprinting, the stereograph projection of the AMS tensor commonly shows broad 95 % confidence ellipses and a low degree of consistency between normalised and un-normalised stereograph tensor projections (Owens Reference Owens2000).

Table 1 Anisotropy of Magnetic Susceptibility data from the Omey Pluton, Western Ireland

Table 2 Summary of AMS results

AMS results yield mean susceptibility (K_mean) values that vary from 0.166×10^–3 to 17.82×10^–3 (SI units) across the pluton (mean=9.69×10^–3±4.94×10^–3). A sharp contrast in K_mean values is noted between earlier (G1=10.93×10^–3; G2=9.29×10^–3) and later (G3=1.94×10^–3) facies (Table 2). A contour map of K_mean (Fig. 8a) reveals a low magnetic anomaly over the Aughrus Peninsula, Friar Island and Curagh Island which coincides with the mapped extent of the G3 facies. There is a similar, but less apparent, anomaly associated with the extent of the G2 facies. The corrected degree of anisotropy (P_j) varies between 1.01 and 1.07 (mean=1.04±0.013), and the overall Degree of Anisotropy (H) varies between 1.2 % and 10.1 % (mean=5.33 %±1.8 %). The maximum variance of mean P_j and H values between each facies is 0.01 % and 0.28 % respectively. As illustrated by a contour map of P_j (Fig. 8b), these data show that magnetic anisotropy does not systematically vary across facies contacts. The same observation is made for T_j values; G1 and G2 samples exhibit a spectrum of AMS ellipsoid shapes that range from well-defined oblate (−0.71), to tri-axial to strongly prolate (0.95) (Table 2). Samples taken from G3 return only weak to moderate oblate ellipsoids (T_j min=0.0024; max=0.016) and adjacent G2 and G1 sites also return similar shape and strength of anisotropy parameters. These data show that K_mean systematically fluctuates across facies contacts and is controlled by the relative concentration of magnetite in each facies, and that magnetic anisotropy is not inherent to granite facies.

Figure 8 (a) Contour map showing magnetic susceptibility (K_mean) remains ≥10^–3 SI across the pluton. (b) A contour map of Corrected Degree of Anisotropy (P_j) shows no relationship between facies contacts and the degree of anisotropy.

A map of the compiled AMS data, with representative stereographic projections of averaged AMS ellipsoids (95 % confidence ellipse), is presented in Figure 9. Either lineations or foliations are plotted from sites where strongly oblate or prolate tensors were detected; both symbols are used for sites where both a magnetic lineation and foliation were detected.

Figure 9 Results of AMS analysis with representative southern hemisphere projections of averaged AMS tensors. Each data point is plotted as a foliation, lineation or both, depending on calculated anisotropy values.

A summary of AMS data is presented in Figure 10. Two distinct AMS foliation patterns are identified (Fig. 10a). At the core of the intrusion, moderately-inclined, outward-dipping oblate fabrics define a strong concentric pattern. In the east of the pluton, where the granite is in contact with the Lakes Marble Formation and the Streamstown Schist Formation, foliations strike contact-parallel and are slightly oblique to the orientation of the host rock structure. Similarly, on Friar Island and Cruagh Island, magnetic foliations dip moderately outwards and are overall contact-parallel, but are often oblique to the local sheeted intrusions which intrude parallel to compositional layers in the folded host rock. Therefore, AMS data contrast field observations that clearly indicate that magma sheeted into the country rock along the folded limbs of the D4 Connemara Antiform and pre-granite regional joint sets. We interpret these magnetic foliations as inflation fabrics which represent the forceful input of magma into the pluton; an observation consistent with interpretations made on data derived from similar intrusions (Molyneux & Hutton Reference Molyneux and Hutton2000; Hutton & Siegesmund Reference Hutton and Siegesmund2001).

Figure 10 A summary of AMS data shows that the concentric foliation is cross-cut by NNW–SSE tectonic shear zones. (A) Two sets of fabrics are identified: (1) concentric foliations, associated with laccolith inflation; and (2) cross-cutting NNW–SSE foliations, attributed to submagmatic shearing along NNW–SSE shear zones (labelled ‘tectonic foliation’). (B) A polar plot of all AMS data points shows that K1 axes are aligned parallel to NNW–SSE axes. ‘Tectonic foliation’ K3 axes show a very consistent NNW–SSE subvertical foliation orientation. A much broader distribution of ‘inflation foliation’ K3 axes reflects the concentric fabric which was partially overprinted by tectonic shearing. (C) Polar plot of Tj vs Pj data shows the spectrum of AMS tensors detected across the pluton.

A different set of magnetic foliations are identified in close proximity to faults which have been identified by field mapping and petrographic observations (Fig. 10a). Magnetic foliations in these areas strike NNW–SSE, are discordant to internal and external pluton contacts, are steeply inclined and become more intense within the NNW–SSE fault zones. K1 vectors, i.e., magnetic lineations, maintain an overall NNW–SSE trend across the pluton and exhibit the highest anisotropy within the NNW–SSE fault zones. A stereographic projection of all AMS data shows that K3 axes from samples spatially associated with shear zones are tightly clustered in the ESE and WSW of the stereonet, while K3 axes sampled >250 m from the fault zones define a concentric pattern and are more evenly distributed (Fig. 10b). K1 axes throughout the pluton consistently plot along a NNW–SSE axis, regardless of proximity to faults (Fig. 10b).

Several examples of magnetic fabrics which bear no immediate relationship to either concentric or NNW–SSE fabrics are noted on Omey Island and immediately east in the Fountain Hill area (Fig. 9). The discordant nature of the mean tensor relative to geological contacts, the contrasting normalised and un-normalised AMS tensor projections, and the broad 95 % confidence ellipse (Owens Reference Owens2000), indicate that these are structurally intermediate fabrics and may be the result of interaction between shearing along less prominent E–W faults and inflation fabrics or NNW–SSE shear structures.