The minor intrusions of the Skye Central Complex

From Earthwise
Jump to navigation Jump to search
From: Bell, B.R. and Harris, J.W. An excursion guide to the geology of the Isle of Skye : Geological Society of Glasgow, 1986. © 1986 B.R. Bell & J.W. Harris. All rights reserved.
Figure 11 Geological sketch-map of the Creagan Dubh and Kilchrist pyroclastic deposits
Figure 4 Geological sketch-map of the Cuillin Complex
Figure 19b Excursion 5 Kilchrist. Ignimbrite outcrops, location map

Chapter 9 The minor intrusions[edit]

(A) Introduction[edit]

Numerous different types of minor intrusion of Lower Tertiary age are found associated with the Skye Centre. They are extremely variable, both in form, for example, dykes, sills, cone-sheets and plugs, and in composition, from peridotites and allivalites through to pitchstones and rhyolite breccias. The time-relationships of these minor intrusions are often obscure, principally because of a lack of clear field evidence. Nevertheless, a wealth of information on the intrusions has been presented by several researchers, notably Harker (1904) and Anderson and Dunham (1966). In the following sections of this chapter attention will be paid only to the salient details of each of the catagories defined. The groups chosen are solely for descriptive purposes and their order should not be taken to imply relative ages. Where such evidence is available it is presented.

(B) The regional dyke swarm and associated subswarms[edit]

The main regional linear dyke swarm associated with the Skye Centre has a general NW-SE trend and invades an area of at least 4000km2 (Speight et al. 1982). The majority of these dykes are less than 2m wide and are either alkali olivine dolerites or tholeiites, although other less-common types have been identified (see below). The thickest dykes occur closest to the intrusive centres. In association are subswarms which have a completely different trend (NE-SW). Two have been identified: the Scalpay Subswarm and the Glenbrittle Subswarm. As these are colinear, it is possible that they are part of one larger unit. Field evidence suggests that the subswarm(s) and the main regional swarm are approximately the same age, although Speight et al. (1982) note that dykes of the main swarm are typically younger. Dykes of the regional swarm intrude the Cuillin Complex but, in contrast, the majority of these dykes pre-date the Srath na Creitheach, Western Red Hills and Eastern Red Hills Centres.

Both Speight et al. (1982) and Wilson et al. (1982) consider that the main swarm and subswarms were intruded over a relatively short period of time and acted as fissure feeders to the plateau lavas of the district (see Section (3D) of Chapter 3). Most of the dykes are vertical, or near vertical, and show little sign of tectonic disturbance. In general, the dykes are discordant to the main units which they cut, although, locally, adopt trends which are controlled by their host-rocks (for example, where bedding or foliation planes allowed easier passage of the magma). Typically, the dykes stand proud of the rocks which they intrude and give rise to a very irregular topography where they occur in large numbers (see Section (4A) of Chapter 4). Close to the intrusive centres the average crustal dilation (or extension) caused by the main NW-SE trending swarm is of the order of 15%, decreasing to the NW and SE, along strike. The amount of crustal dilation decreases markedly in a direction perpendicular to the strike of the swarm, typically over a distance of 10–15km.

Multiple dykes also occur and are commonest close to the intrusive centres. Speight et al. (1982) record that up to 60% of the NW-SE trending dykes exposed along the Strathaird Peninsula, SE of the Cuillin Complex, are of this type. The commonness of this feature may be attributed to the ease with which a second or subsequent magma pulse used the pre-existing weaknesses developed by the initial intrusion.

Radial and tangential sets of dykes, in close proximity to the Cuillin Complex, have been reported by Harker (1904). However, their validity as a distinct group (and not as components of the Scalpay and Glenbrittle Subswarms) is in doubt (Speight et al. 1982), as Harker's detailed mapping did not extend any great distance from the intrusive centres.

The origin of the main NW-SE -trending swarm of the Skye Centre is related to a distinct axis of crustal dilation which developed during Lower Tertiary times. This axis has much in common with those of the Rhum, Ardnamurchan, Mull, Arran and St. Kilda volcanic centres (see Chapter 13). Injection of magma was most likely in a vertical direction from an elongate, NW-SE -trending chamber at depth (20–50km, Speight et al. 1982), as there is little evidence of lateral migration of magma within the swarm. Tensile stresses would have been oriented perpendicular to the planar structure of the swarm, whilst compressive stresses developed parallel to the dominant 'grain', both vertically and horizontally.

(C) The compositions of the basic dykes of the regional swarm[edit]

The basic dykes of the regional swarm have been analysed for major-and trace-elements by Mattey et al. (1977) and B.R. Bell (1984b). Those studied by Mattey et al. (1977) were from four traverses across the general trend of the main swarm, namely: along the SE coast of Harris, Outer Hebrides; on the Waternish Peninsula of Skye; on the Sleat Peninsula of Skye; and, along the north coast of Loch nan Uamh, near Arisaig on the Scottish mainland. The dykes studied by B.R. Bell (1984b) were specifically from the Eastern Red Hills district on Skye. Generally, all of the dykes are porphyritic, with randomly distributed phenocrysts within the central portions, flanked by relatively aphyric, fine-grained, marginal facies.

Mattey et al. (1977) distinguish three groups of basic dykes, each attributable to distinct parental magmas. They are: (1) The Preshal Mhor type, with characteristic low alkali-element concentrations, high CaO (11–12wt.%), Ti/Zr ≃ 110–120, Y/Zr > 0.40 and light rare-earth- element (REE) -depleted chondrite-normalised patterns; (2) The Fairy Bridge type, with similar Ti/Zr and Y/Zr ratios to the Preshal Mhor type, but not showing the characteristic low alkali-element and high CaO concentrations, and with flat REE chondrite-normalised patterns; and, (3) Dykes with compositions similar to the Skye Main Lava Series (SMLS) (see Section (3D) of Chapter 3), with CaO < 9.5wt.%, Ti/Zr < 110 and varying degrees of enrichment in the light REE.

Approximately 70% of the dykes investigated by Mattey et al. (1977) are members of group (1) and appear to have their greatest concentration in the area nearest to the intrusive centres, where crustal dilation is at its greatest (see Section (9B), above). Lavas with similar compositions are found within the upper part of the plateau sequence (Osdale Group, see Section (3D) of Chapter 3) and the group (1) dykes most likely acted as fissure feeders to these lavas. The remainder of the dykes are either of the SMLS or the Fairy Bridge types. Both types are represented by lavas within the plateau sequence, although the former is more common.

The systematic distribution of the groups led Mattey et al. (1977) to suggest that the swarm itself might be regarded as being of a compound nature.

Although not common, dykes of all three groups intrude the granites of the Western and Eastern Red Hills Centres (Mattey et al. 1977; B.R. Bell 1984b). More significant within the granites is a fourth group of dykes, the so-called Beinn Dearg Type (Harker 1904). The Beinn Dearg dykes are characterised by high alkali-element contents and light REEenriched, chondrite-normalised patterns. They are particularly well represented on both of the red hills which are called Beinn Dearg Mhor, one in the Western Red Hills (the 'type-locality') and one in the Eastern Red Hills. They are readily identified by their brown coloration on weathered surfaces. Dykes of the Beinn Dearg Type appear to be restricted to a relatively late stage in the evolution of the Skye Centre.

(D) The cone-sheets of the Cuillin Complex[edit]

Cone-sheets intrude the coarse-grained basic and ultrabasic rocks of the Cuillin Complex, as well as the Coire Uaigneich Granite (see Chapter 4) and parts of the surrounding country-rocks. They have a focal point below Meall Dearg (at an estimated depth of 2–3km), at the southern end of Glen Sligachan. The Skye cone-sheets were first investigated by Harker (1904), who referred to them as inclined, basic sheets. Subsequent work by Bailey et al. (1924) on Mull led to the use of the term cone-sheet. Detailed mineralogical and geochemical studies, using modern analytical techniques, have not, as yet, been carried out on these intrusions.

The cone-sheets are most common in the units just above and just below the Druim nan Ramh Eucrite (see Section (4H) of Chapter 4), where they dip inwards at an angle of 35–40°. Exterior to the eucrite, in lower 'stratigraphic units', they dip at shallower angles (10–20°), whilst in higher 'stratigraphic units' they dip at higher angles (50–65°). In general, the cone-sheets are concordant with any layering present within the basic and ultrabasic units of the Cuillin Complex.

The majority of the cone-sheets are olivine-free, non-porphyritic dolerites with tholeiitic affinities and weather to a dark grey (Harker 1904). They are generally less than 1m thick. The dominant minerals are plagioclase and augite, in an ophitic to subophitic arrangement. Harker (1904) also reports xenocrysts of plagioclase and xenoliths of gabbro within some of these cone-sheets (for example, on Sgurr a' Bhasteir, on Druim nan Ramh, and, below the "Inaccessible Pinnacle" of Sgurr Dearg).

A second variety of cone-sheet is the porphyritic dolerite type, also free of olivine. Harker (1904) records their presence in Coire' a' Chruidh and east of Gars-bheinn, in the southern part of the Cuillin Hills. The phenocrysts within this type are zoned plagioclases, with compositions typical of the labradorite range, whilst the groundmass consists of plagioclase, augite and Fe-Ti oxides. Locally, cone-sheets of this type contain phenocrysts of plagioclase up to 6mm in length, with rounded margins, set in a granulitic groundmass of plagioclase, augite and Fe-Ti oxides.

Harker (1904) also identifies a porphyritic olivine dolerite variety of cone-sheet, which is restricted to the Sgurr Thuilm area in the Western Cuillin Hills. This type is relatively fine-grained and contains, in addition to olivine microphenocrysts, abundant phenocrysts of plagioclase.

A fourth type of cone-sheet, identified by Harker (1904), is similar to the dominant non-porphyritic dolerite noted above, but is, in contrast, more basic and contains abundant groundmass olivine and bytownite. This mineralogical difference is readily noted in the field as these intrusions weather to a distinctive rusty-red. The rock which forms the two summits of Blaven is a particularly good example of this type of cone-sheet.

From field relationships it is clear that the cone-sheets pre-date the majority of the granites associated with the Skye Centre. However, cone-sheets cut the Coire Uaigneich Granite (see Section (4L) of Chapter 4) and this led Richey (1932) to conclude that granites of distinctly different ages are present within the Skye Centre. This conclusion has been substantiated by the studies reported in Chapters 5, 6 and 7, above.

(E) The tholeiite sheets of the Western Cuillin Complex[edit]

Within the gabbros (s.l.) of the Western Cuillin Hills several sheets of amygdaloidal tholeiite and associated volcanic breccias are recognised (Hutchison 1966a). These rocks were originally described by Harker (1904), who concluded that they were remnants of the plateau lava sequence which had been incorporated into the gabbro by foundering of large slabs into the magma chamber. Bailey (1952) re-investigated these rocks and suggested that they were normally-crystallised, tholeiite intrusions. This conclusion was also reached by Hutchison (1966a) and it is from this work that the following descriptions are taken.

Two suites of intrusions have been recognised (Figure 4): an Outer Complex, consisting of three near-horizontal, N-S -trending sheets west of Coire na Banachdich associated with many more inwardly-inclined (50–60°) sheets forming the lower parts of Coire na Banachdich; and, an Inner (Main Ridge) Complex, consisting of numerous sheets also dipping inwards at 60°, which crop out along the main Cuillin ridge.

Within the Outer Complex, marginal facies of the sheets are either distinctly fine-grained, or show no decrease in grain-size. These sheets intrude marginal eucrites and gabbros (see Section (4B) of Chapter 4). Xenoliths of 'country-rock' Ring Eucrite are common and columnar jointing is locally developed.

Sheets belonging to the Inner Complex are intimately associated with pyroclastic rocks (for example on Sgurr Alasdair and Sgurr Sgumain), with the tholeiite acting as a matrix material. Hutchison (1966a) reports diverse rock-types, from welded breccias to xenolithic intrusive rocks. These sheets intrude rocks both of the Border Group and the Outer Layered Allivalite Series (see Sections (4C) and (4E) of Chapter 4).

The non-xenolithic portions of the tholeiite sheets range from coarse-grained dolerites through to fine-grained basalts, the latter sometimes containing phenocrysts of plagioclase. The groundmass is typically composed of plagioclase and augite in an ophitic to subophitic arrangement. Amygdales and interstitial patches of quartz and secondary, hydrous silicates, such as chlorite and epidote, are also present.

Alteration of the Ring Eucrite by the Outer Complex tholeiite sheets is evidenced by the sericitisation of plagioclase feldspars and the schillerization of pyroxenes, together with the replacement of olivine and clinopyroxene by serpentine and a fibrous amphibole, respectively (see Section (4B) of Chapter 4).

According to Hutchison (1966a), the tholeiite sheets released volatiles during emplacement and this caused brecciation and fragmentation of the host-rocks. Vesiculation occurred as the confining pressure reached critical values. Subsequently, vesicules were infilled by secondary (amygdale) mineral assemblages. There is also the possibility that, in terms of composition and mode of emplacement, a close connection exists between the tholeiite intrusions, the cone-sheets (see Section (D), ), and the volcaniclastic pipes of the Cuillin Complex (see Section (4N) of Chapter 4). To date, however, no comparative studies have been undertaken.

(F) The Gars-bheinn Ultrabasic Sill[edit]

The Gars-bheinn Ultrabasic Sill crops out on the southern side of the peak of that name at 300m O.D. and has been described in detail by Weedon (1960) and Bevan and Hutchison (1984). It is composed of feldspathic peridotite and the upper 15m of this 80m-thick intrusion shows the development of alternating bands of feldspar-rich and olivine-rich material. The sill is intruded into Torridonian strata and Lower Tertiary plateau lavas. Marginal chill facies are present, but are not obvious in the field. The base of the sill is not exposed, although the lower portion appears to be connected with a dyke of similar composition.

The feldspar-rich bands in the upper part of the sill commonly show a pegmatoid texture, with individual crystals extending across the complete thickness of individual bands (1–2cm). Clinopyroxene is interstitial to the feldspars. Feldspars from the bands have compositions similar to those found in the non-layered main (lower part) of the intrusion. Individual bands thicken upwards. The olivine-rich layers contain interstitial plagioclase and an Fe-rich clinopyroxene.

Compositional data from the chilled marginal facies of the dyke located below the sill indicates that a picritic magma was involved (Weedon 1960), with affinities to some of the ultrabasic dykes associated with the Cuillin Complex. Within the upper part of the sill some olivine-rich layers have mineralogical and petrological features in common with these ultrabasic dykes (see Section (9I), below).

According to Weedon (1960), the formation of the sill, including the well-developed layering in the upper part and the presence of pegmatoid structures in the plagioclase-rich portions, is dominated by crystal-liquid fractionation processes involving crystal settling. The main portion o he sill, composed of unlayered feldspathic peridotite, was formed by avitational settling of olivine crystals and intercumulus growth of pla oclase and clinopyroxene. Any liquid not involved in this intercumulus rowth process was removed, either by diffusion or convective mec anisms, thus enriching the magma above the crystal-liquid interface Ca, Al, K, Na and volatiles. This peculiar magma composition might, then precipitate the pegmatoidal feldspar-rich layers referred to above. Further cumulus growth of olivine within the normal liquid would give rise to the subsequent olivine-rich layer. This process would be repeated several times. Increases in the thicknesses of individual feldspar-rich bands, upwards, might be attributed to an overall build-up of volatiles within the residual magma. The model of Weedon (1960), therefore, is temperature-independent and strongly controlled by local variations in magma composition.

On the basis of a detailed re-investigation of the field relationships and mineralogy of the sill, Bevan and Hutchison (1984) conclude that the feldspar-rich layers were introduced in the form of fluid 'sills' which were injected into already-consolidated peridotite. These fluids appear to be related to the magma which crystallised to form the Ring Eucrite (see Section (4B) of Chapter 4) and are not co-genetic with the peridotite portions of the sill.

(G) The layered ultrabasic dyke of Stac Suisnish[edit]

Intruded into Lower Jurassic sedimentary rocks on the east side of Loch Slapin, at Stac Suisnish, is a 10m-wide, vertical dyke which is composed of alternating bands (10–20cm thick) of plagioclase-rich and olivine- and pyroxene-rich material which are flat-lying. Little is known about this intrusion, although the layering most likely developed through crystal-liquid fractionation involving gravitational settling of crystals.

(H) The basic sill complex of North Skye[edit]

The Jurassic sedimentary rocks of north Skye are intruded by sills from a large sill complex which is believed to have a focal point locally, at depth (Anderson and Dunham 1966). Only rarely do parts of the sill complex invade the overlying plateau lavas, for example, at Oisgill Bay, on the west side of the Duirinish Peninsula. Faulting and tilting affect the outcrop pattern of the complex, although it is probable that it was originally flat-lying. Sill transgressions are common, for example at Loch Mealt, 4km SE of Staffin Bay.

At different localities total sill thicknesses are relatively constant at approximately 250m, with individual sills achieving thicknesses in excess of 90m (for example, at Meall Tuath, north of Duntulm on the Trotternish Peninsula).

Columnar jointing is a common feature and often gives rise to spectacular scarp slopes on the east side of north Skye. For example, the Kilt Rock, north of Loch Mealt, has columns in excess of 30m long. Generally, these columns are oriented at 90° to the relatively flat-lying Jurassic bedding surfaces and, therefore, are inclined at a high angle. Fan-jointing has also been recorded and may be attributed to either the presence of locally irregular bedding surfaces into which the sill is intruded or unusual heat loss conditions during the cooling of the sill. The most spectacular example of this structure is seen at Lub a' Sgiathain (Tay of the Wing'), east of Rubha Hunish, at the northern end of the Trotternish Peninsula.

Walker (1932) and Anderson and Dunham (1966) identify five distinct rock compositions within this sill complex: (1) olivine dolerite (for example, the sill at the Loch Mealt waterfall, the so-called Kilt Rock); (2) marginal tachylitic and basaltic variants (the upper contact of the lower sill at Rubha Hunish); (3) picrite and picrodolerite (Ru Bornesketaig, Kilmuir); (4) crinanitic and teschenitic variants (the lower sill at Rubha Hunish and the sill at Tulm Bay, north of Duntulm); and, (5) pegmatitic dolerite (which occurs within the Kilt Rock). All five rock-type groups contain varying amounts of: forsteritic olivine, diopsidic augite, calcic plagioclase, Fe-Ti oxides and zeolites (analcime and thomsonite). Where augite and plagioclase dominate the rock-types are (1) and (2), grading into (3) with increasing olivine content, and (4), where zeolites become significant. The development of (5) appears to have been more specific, being dependent upon the build-up of volatiles during late-stage crystallisation processes.

Walker (1932) suggested that variations in composition throughout the sill complex could be attributed to in situ crystal-liquid fractionation by gravitational settling of early precipitated phases. Anderson and Dunham (1966), however, conclude that there is little good evidence for this model, although they do propose that the variations in mineralogy between individual sills throughout the district may be the result of such processes having operated in some magma chamber which existed at a depth below the present level of erosion.

From an investigation of the mineralogy of individual sills it is clear that they all have affinities with alkali olivine basalt magmas and that these rocks can be derived by fractional crystallisation processes. Essentially, the crystallisation of such an alkali olivine basalt magma, which has suffered relatively little crystal-liquid fractionation, could be equated with the normal olivine dolerites ((1), above). Rapid cooling of such a magma would give the tachylitic and basaltic variants. Cumulates rich in olivine might be equated with the picrites and picrodolerites, whilst build-ups of volatiles and volatiles plus alkali-elements would produce pegmatitic dolerites and crinanite/teschenite variants, respectively.

In a subsequent re-investigation of these associations, Simkin (1968) proposed that some form of flow differentiation process had taken place. As magma was injected along flat-lying fissures, phenocrysts would move towards the centre of the conduit and there would be rapid chilling of magma at the fissure margins resulting in the tachylitic and basaltic facies. The porphyritic magma within the central portion of the conduit, when spreading out laterally, would be influenced by gravity and hence facilitate the formation of olivine cumulates (the picrites and picrodolerites).

A banded dyke (the so-called Mystery Dyke), with many features in common with the sills, has been described from Camas Mor, near Kilmuir (on the west side of the Trotternish Peninsula) by Dreyer (1969).

(I) The ultrabasic (picritic) dykes and sills of Central Skye[edit]

The ultrabasic dykes which intrude the Cuillin Complex (see Chapter 4), the plateau lavas of Strathaird and the pre-Tertiary country-rocks of the surrounding area, including the island of Soay, were originally studied by Harker (1904). Subsequently, Bowen (1928) concluded that many of the marginal off-shoots of these intrusions were doleritic (but see below), with olivine typically concentrated in the central portions. Further studies by Dreyer and Johnston (1958), Wyllie and Dreyer (1963), Gibb (1966, 1967, 1968, 1969, 1971) and Gibb and Henderson (1971) have provided much additional data. The descriptions below are taken from the more recent works.

Gibb (1968) defines two distinct groups of intrusions: (1) The Coire Lagan type, typically the smaller intrusions and devoid of any xenoliths; and, (2) The Ben Cleat type, typically the larger intrusions and containing xenoliths. The first group is described by Dreyer and Johnston (1958), whilst the publications of Gibb (see above) consider in detail the Ben Cleat type. Unlike the sills of north Skye (see Section (9H), above), none of these intrusions show alkaline affinities.

The Coire Lagan dykes and sills are typically less than 3m thick and generally have fine-grained, picritic, marginal facies. Examples include: near Leac nan Fasileann, on Soay; Camas nan Gall, on Soay; 2km east of Kinloch, in central Sleat; in the Allt an t-Sithein, north of Loch Sligachan; and, in the Cuillin Hills, in Coire Lagan itself (NW of the main track).

Crystals of olivine dominate these intrusions. They are unzoned and have a relatively constant composition of Fo89 (Dreyer and Johnston 1958). In general, crystal sizes increase from the margins towards the centres. The groundmass is typically variolitic to subvariolitic and is frequently associated with a second generation of much smaller olivine crystals. The proportion of groundmass plagioclase (An40–87, Dreyer and Johnston 1958) to pyroxene (augite) is relatively constant, although the absolute amounts of these two minerals is dependent upon the olivine content of the particular intrusion under consideration.

Dreyer and Johnston (1958) suggest that the Coire Lagan intrusions were initially derived by the selective fusion of pre-existing ultrabasic rocks, to form an eucritic magma which then fractionated forsteritic olivine. The composition of the magma is most likely represented by the fine-grained, marginal facies of some of these intrusons (see above) and, if so, appears to be transitional (with respect to the Critical Plane of Silica Undersaturation of Yoder and Tilley 1962). Flow differentiation processes (see Section (9H), above) may have contributed to the mineral distribution patterns present within these intrusions.

Considerable heat accompanied the intrusion of this group of dykes and sills. Torridonian sedimentary rocks at the margin of the sill at Loch Doir' a' Chreamha, on the west side of Soay, were, locally, up to 92% melted, precipitating tridymite, cordierite, hypersthene and magnetite upon cooling (Wyllie 1959, 1961). The remainder of the liquid was quenched to a glass.

Unlayered, non-xenolithic, ultrabasic dykes and sills occur south of Beinn an Dubhaich, in Strath. Field and mineralogical evidence suggests that these rocks may have affinities with the Coire Lagan type of intrusions.

The Ben Cleat type of picrites occur only as dykes and typically contain cognate xenoliths of ultrabasic material. Gibb (1968) reports details on over 30 dykes on the Strathaird Peninsula and throughout the Cuillin Hills. From his studies, Gibb shows that the mineralogy and whole-rock compositions of the non-xenolithic component of these dykes are very similar to that reported from the Coire Lagan type (see above). Olivine (Fon., Gibb 1968) is the dominant mineral, together with lesser amounts of clinopyroxene, calcic plagioclase (An52–85) and chrome spinel. In general, olivine is a phenocryst phase and most of the dykes are distinctly porphyritic, although Gibb (1968) reports that non-porphyritic variants also exist. The olivine tends to be enriched in the central portions of the dykes (see below).

The most accessible dykes in this group crop out on the Strathaird Peninsula, due east and due west of the summit of Ben Cleat, and on the SW side of the valley between Ben Cleat and Ben Meabost. Others crop out south of Sgurr na Gobhar and along the main Cuillin ridge (Gibb 1968). The Cuillin dykes show a radiate distribution pattern about the Layered Peridotite Series of Sgurr Dubh (see Section (4D) of Chapter 4). Also, there is a positive correlation between dyke width and absolute olivine content; local changes in width do not greatly affect this factor.

The xenoliths which are dispersed throughout these dykes range from 1–60cm across, vary in shape from sub-equant to elongate, and may be angular or rounded (Gibb 1968). Contacts between xenoliths and dyke-rock are sharp. The same minerals which dominate the dyke-rocks are present within the xenoliths, namely: olivine, clinopyroxene, plagioclase, chrome spinel, and magnetite. The relative proportions of each of these minerals within an individual xenolith are quite variable. Most of the xenoliths from one particular dyke studied by Gibb (1969), east of the summit of Ben Cleat on the Strathaird Peninsula, were of feldspathic peridotite, with lesser amounts of dunite, peridotite, allivalite and picrite also present. There is little evidence of axial concentration of the xenoliths. Gibb (1969) concludes that the xenoliths were most likely derived by the disaggregation of some already-consolidated ultrabasic mass at depth and, therefore, are cognate.

In a later publication, Gibb (1976) suggests that the dyke materials were introduced as a magmatic suspension consisting of olivine phenocrysts and ultrabasic xenoliths in an eucritic magma; olivine distribution patterns developing as a result of fractionation by flow differentiation.

(J) The ultrabasic (allivalitic) dykes of North Skye[edit]

The allivalite and anorthite-bearing gabbro dykes of north Skye have been described by Anderson and Dunham (1966) and Donaldson (1977). The majority of these intrusions crop out in the area north of Bracadale, between the village and Beinn a' Chleirich. Donaldson (1977) identifies numerous exposures within this area, together with similar small masses to the south of Bracadale and north of Loch Ravag. Dykes of similar composition also crop out in south Skye at Am Mam, east of Camasunary Bay. Most of the intrusions are elongate and have near-vertical margins, indicating their dyke-like form. They have thicknesses of up to 80m and form elongate, rounded hummocks which trend NW-SE and can be traced for distances of up to 0.5km. They cut the plateau lavas of the Bracadale Group (see Section (3D) of Chapter 3) and are often cut by dykes of the regional swarm (see Section (9B), above).

These dykes are composed of a central, coarse-grained facies consisting of anorthite-bytownite and olivine megacrysts, set in a groundmass of allivalitic composition. Marginal facies may be regarded as feldsparphyric basalts containing sparse plagioclase (An87–93, 20 vol.%) and olivine (Fo78.5–87, 1–10 vol.%) megacrysts set in a fine-grained groundmass (Donaldson 1977). This marginal rock grades over a few tens of centimetres into a central, coarse-grained facies which contains the same megacrysts with the same compositions. Within each dyke, the amount of megacrysts increases towards the centre, although maxima are not found at the centre, but on either side of it. According to the studies of Gibb (1968), this type of crystal distribution pattern may be attributed to flow differentiation processes.

Donaldson (1977) reports xenoliths of the following rock-types from these dykes: olivine-anorthite rock; augite-anorthite rock; and, fluxion-banded troctolite. Also present are xenocrysts (2–2.5cm) of plagioclase (An92), olivine (Fo82–84) and augite.

The two intrusions on either side of the Broisgillmore Burn in Bracadale show clearly many of the features present in these dykes. Likewise, the intrusions at Creag Dhubh and NW of Dun Garsin are instructive, and have easier access.

By subtracting modal percentages of plagioclase and olivine present within marginal facies of these dykes, Donaldson (1977) has calculated that the groundmass composition is hypersthene-normative.

The initial development of these dykes involved the intrusion of the marginal facies, essentially a megacryst-poor, tholeiitic basalt magma, which then crystallised inwards to a coarser-grained rock. Subsequently, a crystal-rich magma was intruded and underwent flow differentiation. The dykes, therefore, consist of megacryst-depleted, fine-grained, marginal facies and a megacryst-rich central zone. Significantly, detailed mineralogical studies indicate that the megacrysts and the groundmass mineral assemblages of individual dykes are not cognate (Donaldson 1977).

Donaldson (1977) also concludes that the 'high-lime liquids' which were involved in the formation of the mineral assemblages of these dykes may have been similar to those involved in the formation of the White Allivalite of the Border Group of the Cuillin Complex (see Section (4C) of Chapter 4 and Section (12C) of Chapter 12) and the crystallisation of the Ca-rich, Ti- and P-poor Preshal Mhor lavas of the plateau sequence (see Section (3D) of Chapter 3).

(K) The trachyte dykes[edit]

Two distinct groups of trachyte dykes are associated with the Skye Centre: (1) The Drynoch Group, at the head of Loch Harport; and, (2) The Broadford and Sleat Group, on the Sleat Peninsula and in the eastern part of the district of Strath (Harker 1904). Dykes from both groups are readily identified by their dull grey colour and platy fracture, together with their generally non-porphyritic texture.

Members of the Drynoch Group are exposed in the Allt Coir' a' Ghobhainn and the Allt na Guile, both at the head of Loch Harport. Typically, these dykes are less than 5m thick and are oriented parallel to the NW-SE -trending regional dyke swarm (see Section (9B), above). They are dominated by oligoclase and alkali feldspar (0.1–0.2mm), in a well-developed trachytic texture, together with accessory augite, Fe-Ti oxides and biotite.

The trachytes of the Broadford and Sleat Group are also oriented parallel to the regional dyke swarm and show similar petrographic features, but additionally exhibit a marginal, spherulitic facies which has a "rodded" structure. This lineation dips at an average angle of 45° to the horizontal and has been interpreted as the direction of magma migration through the fissure during dyke formation. The individual spherules are 5–10mm in diameter and are composed of radiating aggregates of oligoclase and alkali feldspar. Within these marginal facies, as well as in the central portions of certain dykes, there are developed amygdales of various zeolites. Examples of trachyte dykes which show these features include: the dyke near the footpath which crosses the Broadford River, east of Coire-chat-achan; and, the dykes in the Allt Reidhe Ghlais, at the head of Loch Eishort, approximately 1km west of the road. Another member of this group, but which does not show the spherulitic structure, is located 800m NNE of the summit of Ben Suardal in Strath. It is a sparsely porphyritic rock, buff-coloured, and shows a crude trachytic texture.

In the case of the Broadford and Sleat Group, there is no obvious connection between these dykes and any extrusive trachytes. However, the trachytes of the Drynoch Group are in close spatial association with trachytic plateau lavas of the Bracadale Group (see Section (3D) of Chapter 3). It is possible that such dykes represent feeders to these lavas, although, as yet, no comparative geochemical studies have been undertaken.

(L) The augite-andesite dykes[edit]

Dykes of augite-andesite composition crop out in the district of Strath and on the Sleat Peninsula (Harker 1904). These rocks may be distinguished from more basic and acid types by their distinctive mineralogy, consisting of phenocrysts of labradorite and augite set in a groundmass of variable composition ranging between a relatively acid glass and its fine-grained devitrification product. As a typical example, Harker (1904) describes a dyke on Glas Eilean, opposite Harrapool in Broadford Bay, which is readily identified in the field because of its dull grey and somewhat 'rotted' appearance. It is dominated by phenocrysts of sodic labradorite, in a groundmass consisting of feldspar, augite and glass; the latter two now completely altered. An example of a more glassy type crops out on the north side of Loch Eishort, west of the abandoned settlement of Boreraig. This dyke consists of phenocrysts of sodic labradorite (5–6mm) and augite (1–3mm), set in a black, glassy groundmass containing small crystals of plagioclase and altered augite.

With decreasing amounts of plagioclase and augite, and an increase in silica content in the glassy groundmass, there appears to be a transition between the augite-andesite dykes and the pitchstone dykes described in Section (9M), below. This link was alluded to by Harker (1904), but has not been followed up using modern geochemical methods of analysis.

(M) The pitchstone dykes[edit]

In comparison to other igneous centres within the province, Skye contains very few pitchstones, either intrusive or extrusive. This rock-type forms thin dykes which crop out along a narrow corridor of ground running from Glamaig in the NW, to the Allt Duisdale, near Isleornsay, in the SE. Harker (1904) identified two spherulitic pitchstone dykes on the western slopes of Glamaig. The most prominent one is located at 400m O.D. and trends NW-SE, parallel to the regional dyke swarm (see Section (9B), above). This multiple dyke is 60cm wide and is readily identified as a dark green, glassy rock. In thin-section the glass is a pale yellow and there are occasional patches which show the alignment of feldspar microlites, giving a flow-banded appearance. Minute crystals of augite and hornblende are also preserved.

Other dykes in this group crop out on: the eastern slopes of Glas Bheinn Mhor; the northern and eastern slopes of Beinn na Caillich; and, in the Allt Duisdale, west of Isleornsay.

Harker (1904) also reports a group of dykes which he regards as altered pitchstones. They are most abundant east of Beinn na Caillich and are referred to as the Coire-chat-achan type. They are readily identified by their dull grey-brown colour and contain spherules along marginal facies, giving rise to a "rodded" structure. In thin-section the bulk of these dykes consists of pale yellow, altered glass, with scattered microphenocrysts of sanidine and augite.

The main difference between the Coire-chat-achan pitchstone dykes and the others noted above may be one of age, as the former most likely pre-date the granites, and therefore have been subjected to the intense hydrothermal alteration associated with these large intrusions (see Section (12F) of Chapter 12).

(N) The leidleite dyke of the Tungadal River[edit]

In north Skye, a single leidleite dyke is recorded by Anderson and Dunham (1966) from the Tungadal River, 7km east of the village of Bracadale. This bluish rock contains numerous vesicules and is similar to the stony type of leidleite described from Mull (Bailey et al. 1924). Petrographically, it consists of feathery growths of alkali feldspar, together with yellow-brown amphibole microphenocryts (0.1–0.2mm). Glass is absent.

(O) The brecciated rhyolite and acid breccias of the Kilchrist area[edit]

Five small brecciated acid intrusions (Figure 11) occur within the pyroclastic rocks of the Kilchrist area (see Section (8D) of Chapter 8). They were studied by Ray (1962, 1972) and re-investigated by B.R. Bell (1982). They are referred to here as: (1) The Cnoc nam Fitheach Breccia; (2) The Coire Forsaidh Brecciated Rhyolite; and, (3) The Allt Slapin Breccia(s).

Intrusion (1) crops out in the stream bed west of Cnoc nam Fitheach (Figure 11) and is approximately oval-shaped (85m x 35m), with its long-axis running N-S. It is a brecciated quartz-alkali feldspar porphyry which has been severely disrupted. Along the vertical eastern and southern margins there is a flow-banded facies, suggesting an intrusive relationship between the porphyry and the surrounding pyroclastic rocks. The centre of the intrusion consists of rare angular to rounded fragments of porphyritic felsite, set in a fine-grained matrix composed of comminuted quartz and alkali feldspar crystals.

Intrusion (2) is exposed in the Allt Coire Forsaidh, SE of the waterfall (Figure 11), and forms an elongate mass (115m x 40m) with its long axis trending NW-SE. The contacts of this intrusion with the Coire Forsaidh Hybrid (see Section (7B) of Chapter 7) to the NW, and with pyroclastic rocks on all other sides are not exposed. Relative age relationships, therefore, are conjectural. Exposure of the Coire Forsaidh Brecciated Rhyolite is limited to the stream bed and its steeply-dipping east bank. The dominant rock-type is a very pale, greyish-green breccia containing large (up to 5cm) fragments of flow-banded rhyolite. Both the rhyolite fragments and the matrix material are extremely altered, making the rock very friable; nowhere can totally fresh material be obtained. The rhyolite fragments contain microphenocrysts of quartz and alkali feldspar, set in a flow-banded groundmass dominated by the same two minerals. Hydrothermal alteration has produced low-temperature mineral assemblages, including: montmorillonite, chlorite, sericite, epidote, albite and cryptocrystalline silica, together with haematite and goethite, after pyrite.

The acid breccias of (3), above, are exposed only where cut by the Allt Slapin and, therefore, the complete extent of their outcrops cannot be determined. Three intrusions have been mapped (Figure 19c). First, a vertical strip of rhyolite breccia, intrusive towards agglomerates to the NE and also to another acid breccia to the SW. It varies in width between 1 and 5m and contains large (up to 1m across) xenoliths of rhyolitic and pyroclastic material (some of which are tuffaceous). The rhyolite fragments show a well-developed flow-banding, disrupted by spherulitic aggregates of quartz and alkali feldspar, and are set in a pale-coloured (grey-green) tuffaceous matrix consisting of comminuted crystals of quartz and alkali feldspar. The rock has a flinty fracture and in places is cut by thin (up to 1mm), anastomosing veins of quartz, suggesting that a certain amount of silicification has taken place.

The second intrusion is a rhyolite tuff breccia, SW of the intrusion described above, and is, on average, 5–10m wide with vertical contacts. It is composed of angular fragments of flow-banded, tuffaceous rhyolite, randomly dispersed throughout a dark, grey-green matrix.

The third of the Allt Slapin Breccias is located further to the SW and has a common strike with the two intrusions described above. It is a vertical sheet, at least 1m thick, of tuffaceous breccia and exhibits a distinctive flow-banding, containing abundant fragments of rhyolitic tuff. These fragments are similar to the material of the rhyolite tuff breccia (intrusion (2), above) and, therefore, suggests that the tuffaceous breccia is younger. The whole mass is strongly fractured and jointed, allowing its extent to be determined readily in the field.

(P) Deformed Lower Tertiary dykes and sills[edit]

Deformed basic and acid igneous sheets within Cambro-Ordovician dolostones (see Section (2D) of Chapter 2) around the margins of the Beinn an Dubhaich granite intrusion (see Section (7F) of Chapter 7) have been described by Nicholson (1970, 1985) and Longman and Coward (1979). These minor intrusions are boudinaged and folded and this deformation has been attributed to radial compressive forces emanating from the western part of the granite intrusion, presumably during its emplacement. Boudin necks within these dykes (for example, at Camas Malag) show evidence of chilling, suggesting that the deformation took place before the intrusions had completely crystallised. Other examples may be examined at Dun Beag, on Loch Slapinside, and around the old manse at Kilchrist.


Appendix 1: Glossary of petrological names and terms[edit]

Appendix 2: Glossary of fossil names[edit]

Appendix 3: Glossary of place names and grid references[edit]

At all times follow: The Scottish Access Codeand Code of conduct for geological field work