Journal of Petrology Advance Access originally published online on March 18, 2005
Journal of Petrology 2005 46(8):1585-1601; doi:10.1093/petrology/egi027
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Spatial Constraints on Magma Chamber Replenishment Events from Textural Observations of Cumulates: the Rum Layered Intrusion, Scotland
DEPARTMENT OF EARTH SCIENCES, UNIVERSITY OF CAMBRIDGE, DOWNING STREET, CAMBRIDGE CB2 3EQ, UK
RECEIVED FEBRUARY 27, 2004; ACCEPTED FEBRUARY 14, 2005
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONThe Eastern Layered Series of the Rum Layered Suite is formed of 16 macro-units each comprising a lower peridotite and an overlying feldspar-rich layer (the local term is allivalite). The origin of the peridotite layers is unresolved, with two contrasting models. The earlier of the two is based on repeated replenishment of an open-system magma chamber with deposition of fractionated material on the chamber floor. The second is based on the early formation of a troctolitic complex, which is then repeatedly intruded by sills of replenishing picritic magma to form the peridotite horizons. The lack of resolution of this fundamental problem is a consequence of the reliance of previously published studies on field observations. I present evidence to show that the clinopyroxene in the allivalites preserves information about the distribution of the last melt to solidify, permitting determination of not only the extent of super-solidus textural equilibration but also the sub-solidus history of the allivalite horizons. Comparison of profiles of clinopyroxene–plagioclase–plagioclase median dihedral angle across allivalite units demonstrates that it is possible to distinguish between those that were intruded by later picrite sills and those adjacent to peridotite horizons formed by replenishment and subsequent deposition of fractionated crystals above the pre-existing pile. In the region studied, only the main peridotite body of Unit 9 was intruded into a pre-existing allivalitic mush.
KEY WORDS: Rum Layered Intrusion; chamber replenishment; dihedral angles; cumulates
|
INTRODUCTION |
|---|
Magma chambers play an important role in determining the composition and timing of volcanic eruptions. Although it is possible to infer much about the workings of magma chambers from an examination of volcanic rocks (e.g. Bacon & Metz, 1984
; Murphy et al., 2000), direct observations of active chambers are necessarily limited in scope. Studies of dissected magma chambers, or igneous intrusions, offer a perhaps more direct route to understanding magma chamber evolution, although the problem becomes one of unravelling an often complex and sustained history of sub-solidus textural modification (e.g. Hunter, 1996).
Holness et al. (2005) have shown that the information recorded by the population of dihedral angles at the junctions of interstitial material with cumulus grains can be used to make inferences about the late-stage pore structure in solidifying systems and their subsequent sub-solidus thermal history. In this contribution I give an example of how such information can be used, and show how the location of replenishing magma in an open-system chamber can be inferred from detailed studies of the spatial variation of dihedral angles. The specific example I discuss is the Paleocene layered intrusion of Rum, Scotland, which poses a hitherto unresolved problem concerning the geometry of magma chamber replenishment (Bédard et al., 1988; Volker & Upton, 1990, 1991; Bédard & Sparks, 1991).
|
GEOLOGICAL SETTING |
|---|
The Rum Layered Intrusion (Fig. 1) comprises a series of well-defined layers of alternating peridotite (strictly these olivine cumulates are feldspathic peridotites) and feldspar-rich rocks. The latter comprise troctolites and gabbros, and are collectively known by the local name of allivalite. The layered series is subdivided into the Eastern Layered Series (ELS), the Western Layered Series (WLS) and the Central Series (CS), together with numerous subsidiary peridotite, gabbro and dolerite plugs (Emeleus et al., 1996
; Emeleus, 1997). This study concentrates on the ELS, specifically on Units 8–11 (Fig. 1). The ELS is composed of a 750 m thick succession of peridotites and allivalites, divided into 16 (Volker & Upton, 1990) major macro-rhythmic units, each having a peridotite base, overlain by troctolitic (or more rarely gabbroic) allivalite cumulates. In detail, the allivalite horizons of each of the 16 units are texturally and compositionally varied, and contain many olivine-rich layers, varying in thickness from a few centimetres to several metres (Brown, 1956; Emeleus et al., 1996). Each of these olivine-rich layers is compositionally and texturally indistinguishable from the underlying major peridotite unit, and some of these minor layers have well-developed load casts at the base and/or evidence of localized metasomatic replacement of the overlying allivalite (Volker, 1983; Butcher et al., 1985; Morse et al., 1987; Renner & Palacz, 1987).
|
Brown (1956)
, followed by Dunham & Wadsworth (1978), developed a model involving fractionation and crystal settling from a mafic replenishing magma, with a crystallization sequence olivine (with spinel) 
olivine + plagioclase olivine + plagioclase + clinopyroxene. Accumulation of successive crops of crystals at the base of the chamber from a single batch of magma would create a cumulate sequence passing upwards through peridotite to troctolite and gabbro. Each macro-scale unit can thus form from a single magma injection, with the overlying unit formed by another injection, which enters the chamber and rises to a level above the developing crystal mush. This early work has been supplemented by that of McClurg (1982), Tait (1984, 1985) and Upton et al. (2002), who presented evidence that the magma that periodically replenished the Rum chamber was most probably slightly hydrous, with about 18 wt % MgO.
This essentially sedimentary process has been corroborated recently by work on the peridotite horizons in the WLS (Worrell, 2002; Worrell et al., 2003), and on allivalites from the ELS (Tepley & Davidson, 2003), which is consistent with the deposition in gravity currents of cumulus crystals grown elsewhere. However, the allivalite parts of many units in the ELS contain thin (<10 m), transgressive peridotite layers, some of which show outcrop evidence for intrusive relationships, with infiltration and assimilation of the overlying allivalite (Volker, 1983; Butcher et al., 1985; Morse et al., 1987; Renner & Palacz, 1987; Bédard et al., 1988; Volker & Upton, 1990). Additionally, some of the major peridotite horizons contain allivalite blocks apparently residual after assimilation of solid allivalite by hot intruding picrite (Butcher et al., 1985; Morse et al., 1987; Volker & Upton, 1990). This poses the interesting (and geologically important) question as to how widespread was the intrusion of replenishing picritic magmas into a pre-existing crystal pile at the base of the chamber.
Evidence for intrusion of at least some of the major peridotite bodies on Rum is incontrovertible. Numerous, late-stage, plugs of peridotite intrude both the central part of the Complex and the Torridonian sediments to the north (Volker & Upton, 1990; Wadsworth, 1994; Holness, 1999). Similarly, the peridotite body identified by Brown (1956) as the Unit 9 peridotite is discontinuous, forming a series of lens-like bodies, with common allivalitic xenoliths. The SE end of the lens-like body centred on grid reference NM (397 968) (labelled A in Fig. 1) is particularly clearly intrusive, with margins that truncate at high angle the layering in the adjacent allivalite [with detailed field evidence presented by Bédard et al. (1988)]. Butcher et al. (1985) mapped this region as a peridotite plug distinct from the essentially conformable, adjacent, peridotite body. Bédard et al. (1988) suggested that the plug is actually an integral part of the essentially conformable body, with the corollary that the entire Unit 9 peridotite is intrusive into a pre-existing allivalite [which is usually known as the two separate allivalites of Units 8 and 9 after Brown (1956)].
After a careful study of field relations, Bédard et al. (1988) took this idea a step further by postulating that many, if not most, of the major peridotite horizons in the ELS were intruded as picrite sills into a pre-existing allivalite crystal mush at the base of the chamber. Although the evidence is compelling that many minor peridotite layers were late sills, and that many of the contacts of the major bodies have features indicating at least localized intrusive relationships, the extension of this model beyond the Unit 9 peridotite to all major peridotite horizons, with its associated implications for the development of the Rum magma chamber, is controversial.
Volker & Upton (1990) put forward a list of criteria to distinguish intrusive peridotites from conformable, sedimentary peridotites [again dominated by field-scale characteristics, and disputed by Bédard & Sparks (1991)]. They concluded that the only major peridotite horizon that is intrusive is that of Unit 9. The exchange between Volker & Upton (1991) and Bédard & Sparks (1991) shows that the same field observations can be used to support either an intrusive or a sedimentary origin for major peridotite bodies. Although the intrusive nature of the Unit 9 peridotite has never been under question, there is a clear need to find a general, non-field based, discriminant between intrusive and sedimentary peridotite bodies to resolve the question of whether other major layer-parallel bodies are also intrusions.
In this contribution I present the results of a detailed textural study of allivalite horizons in the Rum ELS cumulates. First, I build on the textural observations of Holness et al. (2005) to show that much of the clinopyroxene is pseudomorphing the last melt within the solidifying crystal mush, with at least some late-stage melt pseudomorphing by clinopyroxene even in rocks in which the clinopyroxene is predominantly cumulus. This means that we can determine not only the distribution of late-stage melts, but also spatial variations in the rate of cooling of the cumulate pile. The latter is because there is a significant difference between the angles subtended at melt–solid–solid junctions (which typically have median values of 60° or less; Holness et al., 2005) and the equilibrium clinopyroxene–solid–solid dihedral angle (
cpp), which lies in the range 110–122° expected for solid-state values (Vernon, 1968, 1970; Holness et al., 2005). Because the replacement of melt by a solid silicate phase will result in an inherited population of low dihedral angles, sub-solidus textural adjustment will result in an increase in dihedral angle towards the equilibrium solid-state values (Holness et al., 2005). The extent to which this occurs is controlled by the sub-solidus thermal history (Holness et al., 2005). We can thus use observed cpp variation to distinguish regions of the crystal pile that experienced anomalous cooling histories. By collating textural information across entire allivalite horizons, the thermal effects of nearby peridotite bodies can be discerned, and large-scale intrusive and cumulate peridotites can be distinguished.
|
POST-ACCUMULATION THERMAL HISTORIES OF ALLIVALITE |
|---|
Regardless of the ultimate origin of the cumulus crystals in the mush at the base of the chamber, whether grown in situ (e.g. Campbell, 1978
; Tait & Jaupart, 1996), or deposited in density currents (e.g. Worrell, 2002; Tepley & Davidson, 2003; Worrell et al., 2003), the interstitial material must have grown in situ as a result of conductive loss of heat downwards. In an essentially undisturbed crystallizing pile, in which the temperature of the solidification front is that of the eutectic in the system, the sub-solidus thermal history of any horizon is a function of the rate of upwards movement of the solidification front. For a constant upwards movement of the solidification front, the dihedral angle population for the interstitial phases will not change with stratigraphic height. Importantly, in the Rum magma chamber in which the formation of allivalite is consequent to the arrival of plagioclase on the liquidus during closed-system fractionation of initially picritic magma, there is no significant change in temperature of the crystal mush as it changes from primary growth of olivine (either in situ or as cumulus) to cotectic growth of olivine and plagioclase. This means that in a sequence primarily controlled by cooling-driven fractionation, the thermal history of the uppermost peridotite layer of any macro-unit will be indistinguishable from that of the base of the overlying, related, allivalite. The interstitial phase dihedral angle population will thus not change across the boundary of the peridotite with the overlying allivalite.
If a replenishment event places hot, primitive, picritic, liquid above the allivalite layer, the thermal history at the top of the allivalite will be discernibly different from that in the bulk. In the absence of remelting by the replenishing magma, completely solidified horizons close to the top of the allivalite will experience an input of heat that will increase the dihedral angles of the interstitial phases. For incompletely solidified allivalite, the resident, evolved, cool, gabbroic interstitial melt may be replaced by downwards infiltrating, dense, picritic melts (e.g. Kerr & Tait, 1985), which will cool and crystallize until the three-phase cotectic is reached. The subsequent thermal history for each horizon of the infiltrated allivalite would then be controlled by the distance to the top of the crystal mush. If picrite injection resulted in a burst of crystallization and growth of an olivine-rich crystal mush at the base of the chamber, then the increase in dihedral angle in the underlying allivalite may be less than in an environment in which the picrite was convecting turbulently, maintaining high temperatures at the chamber floor [as suggested by Huppert & Sparks (1980)].
In the case of injection of a picrite sill into a pre-existing allivalitic crystal pile, the thermal effects at the lower contact of the sill would be similar to those expected for the top of an allivalite layer in a normal cumulate sequence. However, a difference would be discernible at the upper surface of the sill. As the mechanically most plausible scenario is the intrusion of a picrite sill just above the level of the solidification front (C. H. Emeleus, personal communication, 2003), upwards infiltration of picrite into the overlying allivalitic mush is likely. In such a case, one might expect a significant change in dihedral angles in the immediate vicinity of the sill compared with that for a normally accumulating layer. Such changes are also likely to be coupled with metasomatism of the overlying allivalite.
|
ANALYTICAL TECHNIQUE |
|---|
Following the metallurgists, who pioneered work in this field using opaque materials, geologists have generally measured dihedral angles using a conventional, flat-stage, microscope or by analysis of images generated by scanning electron microscopy. This method relies on measuring a population of angles on a randomly oriented two-dimensional section through the material. For samples containing a single true value of the dihedral angle, it can be shown that the median of a population of these angles is within 1° of the true three-dimensional angle (Harker & Parker, 1945
; Riegger & Van Vlack, 1960). For samples with anisotropic interfacial energies, which includes all of geological interest (e.g. Kretz, 1966; Vernon, 1968, 1970; Laporte et al., 1997; Cmíral et al., 1998), sophisticated statistical techniques are required to constrain the range of true equilibrium angles (e.g. Jurewicz & Jurewicz, 1986). However, the problem can be neatly circumvented by the use of the Universal Stage on an optical microscope (Vernon, 1997). I used a four-axis Leitz Universal Stage, which permits up to 90° rotation of the thin-section from the horizontal, and allows accurate measurement of the majority of the dihedral angles present in the sample, with an error on each measurement of a few degrees. I measured between 30 and 110 clinopyroxene–plagioclase–plagioclase angles in each sample, with a magnification of x300.
|
LOCATION OF SAMPLES |
|---|
For this study I collected samples along closely spaced vertical traverses across Units 8, 9 and 10 (Fig. 1), to augment the pre-existing sample suites of Brown (1956)
(Unit 10), and Hallworth (1998) (Unit 9). Two additional traverses were made across a small, lensoid, allivalite body that forms a pair of short scarps between grid references NM (39345, 96460) and NM (39470, 96469). This lens lies stratigraphically above the chromite-capped pavement marking the top of Unit 11 on the col between Hallival and Barkeval, and is separated from the underlying Unit 11 allivalite horizon by a significant thickness (
50 m) of peridotite. Likewise, it is separated from the overlying Unit 12 allivalite by a significant thickness of peridotite (some tens of metres). This lensoid body is not part of Brown's (1956) schema and, because of pre-GPS mapping errors, there are discrepancies between the published map (Emeleus, 1994) and my observations. The exact position (determined using GPS) of the allivalite lens (marked by a pair of scarps on the topographic map) occurs within the Unit 13 peridotite on the map of Emeleus (1994), and I show its correct stratigraphic position schematically in Fig. 1b. I refer to this allivalite body as Unit 11a.
In general, the peridotitic parts of the macro-scale units on Rum are poorly exposed, whereas the allivalite crops out well and forms a shelf and escarpment. Although it is relatively common to find peridotite exposed at the base of the scarps, and thus to collect an accurately measured traverse upwards from the base of the allivalite, the shelf-like outcrops characteristic of the upper parts of the allivalite make it difficult to be sure that the vertical position of samples is accurately measured. In some cases it is also not possible to know whether the very top of the unit has been sampled. The base of each of Traverses a, e, f, g and h is well located, whereas the peridotite underlying Traverses b, d and i is not exposed. The top of each of Traverses b, g, h, and i is well located (and sampled), whereas the tops of the others are not well defined because of lack of outcrop. Most of the samples of Traverses g and h were collected and first described by Hallworth (1998). Those of Traverse a were collected and described by Brown (1956).
The Units 8 and 9 of Brown (1956) occur roughly in the middle of the currently exposed ELS stratigraphy. Although the peridotite of Unit 8 is continuous over a wide area, that of Brown's Unit 9 is laterally impersistent, with variable thickness (0–25 m). Where the Unit 9 peridotite is absent, there is a continuous succession from the Unit 8 allivalite through the Unit 9 allivalite (Fig. 1b). Because the Unit 9 peridotite is an intrusion, these two allivalite horizons were formed during a continuous episode of allivalite formation and I will refer to them collectively as the Unit 8/9 allivalite. The thickness of the allivalite overlying the Unit 9 peridotite varies between 15 and 25 m. Mapping by Bédard et al. (1988) demonstrated that this thickness variation is primarily due to downcutting by the overlying Unit 10 peridotite. Five traverses were made within Unit 8/9 (Fig. 1b).
Brown (1956) chose Unit 10 as the type unit, because of its great thickness and exceptional exposure. His decision to make this the type unit has resulted in the concentration of previous research efforts on Unit 10 (e.g. Dunham & Wadsworth, 1978; Tait 1984, 1985). The allivalite of Unit 8/9 in the region of Traverses e, f, g and h has been studied extensively as the contact between the uppermost gabbro and the underlying troctolite is undulatory here, cutting layering with little or no deflection. This enigmatic feature, generally known as the Wavy Horizon, has been ascribed by Bédard et al. (1988) to a process of metasomatic ‘gabbroitization’ of original troctolitic mush by downwards percolating late-stage melts, although Hallworth (1998) suggested it was more probably due to a metasomatic event within an original gabbro caused by the influence of a hot picrite intruded below.
|
PETROGRAPHIC DESCRIPTIONS |
|---|
All the rocks are fresh, with only a minor proportion (<1%) of alteration products.
Peridotite
All peridotite samples studied are formed of a touching framework of olivine grains, typically 1–2 mm in diameter, surrounded by interstitial plagioclase and clinopyroxene. Both interstitial minerals show considerable variation in modal proportion and spatial distribution. The olivine grains are generally equidimensional, with subhedral or rounded habit, although some elongated or rod-like morphologies occur. Layering, either modal or formed by preferred alignment of slab-like olivine grains, is evident in some peridotites. Cumulus, euhedral, Cr-spinel is common, either enclosed by intercumulus plagioclase and clinopyroxene, or as inclusions within olivine.
The peridotites commonly contain two generations of plagioclase. One forms oikocrysts enclosing olivine grains. The other forms much smaller, interstitial, grains, which are commonly zoned, consistent with growth during the last stages of solidification when the remaining interstitial liquid forms isolated pockets. The differentiation of plagioclase into two distinct generations demonstrates that the earlier oikocrystic growth resulted in the formation of extended pockets of residual melt (Fig. 2a), which were filled by later growth of zoned plagioclase. This feature is also seen at the margins of clinopyroxene oikocrysts in the same rock (Fig. 2b), which develop apophyses extending from the main body of the oikocryst along boundaries between olivine and plagioclase grains. By comparison with the analogous plagioclase texture, I interpret these apophyses as pseudomorphs of late melt by clinopyroxene.
|
Further evidence of the distribution of late-stage melts in the peridotites is found in sample H1-5 (collected from the top of the Unit 10 peridotite above Traverses g and h). This sample is very fresh, and the few primary hydrous minerals present (e.g. amphibole and mica) are concentrated in cuspate pockets between individual oikocrysts of clinopyroxene and plagioclase, and closely associated with zoned plagioclase. I interpret these pockets to have formed from the very last melt to have solidified.
Allivalite
The rocks mapped as ‘allivalites’ can be divided into troctolites and gabbros, both of which are represented in Units 8–10, although there is no cumulus clinopyroxene in the allivalite of Unit 11a.
Troctolite
The troctolite comprises 60–80% cumulus plagioclase, 15–40% olivine and <7% clinopyroxene, with accessory Cr-spinel (1%). Tabular plagioclase grains form 0·1–0·5 mm interlocking laths (Fig. 3a), commonly with strong, layer-parallel foliation that wraps around olivine grains and/or clinopyroxene oikocrysts. Olivines are generally subhedral, tabular to equant grains, with a size range of 1–3 mm. In detail, many olivine grains display small apophyses extending down plagioclase–plagioclase grain boundaries. Diopsidic pyroxene occurs as patches of optically continuous, interstitial wedges between plagioclase laths (Fig. 3a and b). Notably it also forms rinds, 15–70 µm thick, separating olivine grains from plagioclase (Fig. 3a and b). These rinds generally comprise only a single grain, although rarely rinds are made of two or three grains, separated by low-angle grain boundaries. The rinds either completely or partially surround olivine grains, with no systematic distribution of the rinds in relation to the foliation or orientation of the olivine grain—rinds may wrap entirely around an individual grain (in a 2-D cross-section), or occur at one or more sides. Clinopyroxene appears to concentrate in olivine-rich regions (Fig. 3b).
|
Cr-spinel occurs as sparsely distributed euhedral grains (
0·2 mm). Layering is commonly defined by a preferred orientation of plagioclase grains, but modal layering is also present in places, with occasional olivine-rich laminae 2–4 mm thick.
Gabbro
Gabbroic parts of the allivalites are generally composed of an adcumulate assemblage of diopsidic augite (30%), plagioclase (60%), olivine (up to 10%), and minor Cr-spinel. Two distinct lithofacies occur: a fine- to medium-grained massive granular gabbro in which pyroxene occurs as equant, anhedral (cumulus) grains; and a medium- to coarse-grained poikilitic facies, with large (<2 cm) pyroxene oikocrysts that envelop both olivine and plagioclase (Fig. 3c). Enclosed crystals are generally finer grained than those outside, and commonly more rounded (Fig. 3c). They have no preferred orientation. The poikilocrysts of clinopyroxene tend to concentrate in narrow, layer-parallel horizons. Cumulus grains of clinopyroxene commonly have angular projections down intersecting grain boundaries between plagioclase grains (Fig. 3d). These angular projections are also present on the margins of cumulus olivine grains, but generally are not so well developed, and partially obscured by minor serpentinization of the grain boundaries.
In both lithofacies, plagioclase exhibits strong, layer-parallel foliation and wraps around olivine and pyroxene grains. The gabbro displays modal layering in the form of thin (2–4 mm) olivine-rich laminae and centimetre-scale micro-rhythmic alternations of gabbro, troctolite, anorthosite and pyroxenite, texturally identical to that in the troctolite.
|
CLINOPYROXENE AS AN INDICATOR OF MELT TOPOLOGY |
|---|
It is possible to infer the distribution and extent of textural equilibration of late-stage melts from textural observations of entirely solid rocks if the last solidifying phases pseudomorph the final melt (e.g. Platten, 1981
; Hunter, 1987; Holness & Clemens, 1999; Sawyer, 2001; Marchildon & Brown, 2002; Holness et al., 2005). For interstitial material growing in pore space during late-stage crystallization, the angle subtended at pore corners is inherited from that of the melt (e.g. Platten, 1981; Harte et al., 1991; Holness & Clemens, 1999; Sawyer, 1999, 2001; Rosenberg & Riller, 2000; Holness & Watt, 2001; Holness et al., 2005). Holness et al. (2005) have shown that much information about the geometry of the pore structure immediately prior to complete solidification (and the subsequent sub-solidus thermal history) is recorded in the populations of dihedral angles subtended at the corners of pseudomorphing interstitial phases.
Clinopyroxene is the last phase to arrive on the liquidus in the Rum magma (apart from minor hydrous phases), and textures in the Rum allivalites are consistent with late, interstitial clinopyroxene growth. It should be noted that clinopyroxene is not the only phase forming during the last stages of solidification. The other two phases at the eutectic, olivine and plagioclase, must also be forming, with the last porosity filled by a mixture of olivine, plagioclase and clinopyroxene in eutectic proportions. Cumulus olivine grains do have apophyses extending down plagioclase–plagioclase grain boundaries, demonstrating that olivine does pseudomorph the adjacent porosity to some extent. However, the ubiquitous low levels of serpentinization of olivine prevent accurate determination of olivine–plagioclase–plagioclase dihedral angles in most of the samples. The presence of plagioclase pseudomorphs is impossible to detect, given the ease of grain boundary migration, and consequent textural maturation, in monomineralic aggregates (e.g. Hunter, 1987). Therefore, the population of clinopyroxene–plagioclase–plagioclase dihedral angles, or cpp, in the Rum allivalites is the best for recording information about late-stage cumulate history.
For comparison with cpp in the Rum allivalites, the equilibrium value of cpp was obtained from a metabasic granulite, sample 11787 from the Harker Collection at Cambridge University. This came from Perugar, Madura district, South India, and has the unimodal grain size, straight or gently curved grain boundaries and well-defined dihedral angles diagnostic of textural equilibrium. A population of 100 true 3-D angles varies from 68° to 150°, demonstrating a significant crystallographic control (i.e. interfacial energy anisotropy) on the position of the grain boundaries in clinopyroxene–plagioclase aggregates. The population is unimodal, with a broad main peak at 120°, and a subsidiary peak at 65–70° (Fig. 4a). The mean of the population is 114°, with a mode of 116° and a median value of 114°. The standard deviation is 17°. These results are consistent with previously published measurements of cpp in granulites (Vernon, 1968, 1970).
|
In contrast, the population of true dihedral angles in each of the samples of Rum allivalites has a wide range, offset to lower angles compared with the equilibrium population (Fig. 4a). Median values of the true 3-D
cpp in the Rum allivalites vary from a minimum of 62° to a maximum of 110°, with an inverse relationship between the median and the standard deviation. Those samples with a low median value of cpp have standard deviations of 25–30°, whereas those with high median values have standard deviations generally <20° (Fig. 5). The data thus form a trend between the solid-state value of cpp and populations associated with textures formed by impingement of growing grains (Fig. 5). This led Holness et al. (2005) to conclude that the structure of the original melt-filled porosity in the Rum allivalites was dominated by growth of impinging plagioclase, with no apparent control by melt–solid textural equilibration. Subsequent sub-solidus modification of this inherited texture resulted in the increase of cpp towards the equilibrium population observed in granulites [reported here and by Vernon (1968, 1970)] (Holness et al., 2005).
|
Samples containing poikilitic clinopyroxene have two distinct populations of
cpp. In an analogous manner to the poikilitic clinopyroxene enclosing olivine grains in the Rum peridotites (Holness et al., 2005), the population of cpp in the central parts of each poikilocryst has a significantly higher median and lower standard deviation than that at the margins. As an example, angles measured on totally enclosed plagioclase grains within poikilocrysts in the poikilitic gabbros overlying the southeastern end of the peridotite intrusion (from Traverse e, Fig. 1b) have a median close to the highest measured on Rum at 106° (with a standard deviation of 15° for 35 measurements). The median value of cpp measured on the margins of the same poikilocrysts is 71° (with a standard deviation of 22° for 50 measurements). Following Holness et al. (2005) I interpret this as a result of sub-solidus textural equilibration occurring at a rate commensurate with growth of the poikilocryst.
Apart from the important difference between the central and marginal regions of clinopyroxene poikilocrysts, there is no relationship between the median and standard deviation of cpp and the type of clinopyroxene, i.e. whether it is interstitial, cumulus or on the margins of poikilocrysts. This means that not only is the interstitial clinopyroxene pseudomorphing melt-filled pores, but also that the marginal regions of cumulus and poikilitic clinopyroxene are inheriting a shape from melt-filled pores. The elongate apophyses and cuspate margins of cumulus clinopyroxene (Fig. 3d) are thus melt pseudomorphs.
The clinopyroxene rims surrounding olivine grains in the troctolitic allivalites (Fig. 3a and b) suggest that melt wetted olivine–plagioclase boundaries in the last stages of solidification. Given that the melt-filled porosity was dominated by impingement (Holness et al., 2005), the rims are most probably a disequilibrium feature. Although there is no experimental information on the distribution of melt in texturally equilibrated olivine–plagioclase aggregates, I am aware of no natural examples of olivine–plagioclase boundaries being replaced by stable, thick, films of melt. Melt-free olivine–plagioclase boundaries are common in crystal clots in basaltic lavas, for example.
During shear, thick melt films may form on grain boundaries, even in systems with non-zero melt–solid–solid dihedral angles (e.g. Urai et al., 1986; Jin et al., 1994). It is thus possible that melt films formed during late shearing of the crystal pile (C. H. Emeleus, personal communication, 2004), but this would perhaps mean that clinopyroxene rims would predominate on olivine–plagioclase grain boundaries at high angles to the foliation. Although clinopyroxene certainly concentrates in pressure shadows adjacent to olivine grains, demonstrating a strong control on residual porosity by compaction, rims show no spatial correlation with the foliation.
Similar rims are present in the Rum peridotites, where they are formed by both plagioclase and clinopyroxene (Fig. 2). I suggest that these features are a result of diffusion-limited growth kinetics, i.e. mass transport to grow the plagioclase is fastest in the centres of the melt channels separating the olivine grains, leaving behind a skin of plagioclase-depleted melt in which growth is inhibited. It is not clear how this mechanism would operate in the allivalites however, and the rims remain enigmatic.
The range of angles observed in the Rum allivalites demonstrates a variable amount of sub-solidus textural equilibration in these rocks (Holness et al., 2005). By putting this variation in a spatial context, comparing the amount of textural adjustment across individual allivalite bodies, we can constrain variations in sub-solidus thermal history and locate the heat source responsible (such as a replenishing picritic melt).
|
SPATIAL VARIATION OF cpp IN THE ALLIVALITES
|
|---|
The
cpp measurements are presented in terms of the median of the measured population in each sample. Whereas the median of a population of angles measured in random 2-D cross-sections of a sample containing a single-valued dihedral angle is close to that single-valued angle (Riegger & Van Vlack, 1960), there is no corresponding meaning to the median of a population of true 3-D angles. However, the median of such a population gives a qualitative idea of the approach to sub-solidus textural equilibrium (Fig. 5), and is a convenient short-hand descriptor.
In samples containing poikilitic clinopyroxene, I report only the median cpp for interstitial clinopyroxene grains, and not that for fully enclosed plagioclase clusters within the poikilocrysts.
Unit 10
The median value of cpp is almost constant through most of Traverse a, with a range of 86–91° with standard deviations of 20° (Fig. 6). Similar values are found within the top 8 m of the underlying Unit 10 peridotite. The two hand specimens from the top of Brown's traverse are missing from the rock collections of both Oxford and Cambridge Universities, although the original thin-sections remain. Because of historically valuable annotations on the labels these sections could not be cut down to fit the Universal Stage. However, the topmost sample contains no clinopyroxene, and measurements of sharply defined dihedral angles on the other sample (i.e. those at a high angle to the plane of the thin-section) using a conventional microscope stage yielded a population of 32 angles with a median value of 91° (standard deviation 22°). Two other samples from the top of Unit 10 were collected by Brown (1956), although no exact locations and only sketchy descriptions of their stratigraphic position were given. One of these samples, 5049 (collected from the topmost layer of the Unit) has a median angle of 101° (standard deviation 16°), and is the only sample from Brown's collection with a median outside the range 86–91°. Although no exact stratigraphic position is given, it appears that this increase in angle occurs within a few feet of the top of Unit 10.
|
Unit 11a
Median values of populations of
cpp in the Unit 11a allivalites from Traverses b and d generally fall in the range 79–85° (Fig. 7). Exceptions to this are the single sample collected within a few centimetres of the contact with the overlying peridotite, and a cluster of samples 8 m from the base of the allivalite in Traverse d.
|
Unit 8/9
Traverse f crosses the complete allivalite section from Unit 8/9 (Fig. 1). The median value of
cpp is 88–93° across the entire traverse, apart from the lower three samples (Fig. 8). Traverse i is also within this range in the lower part, but shows a steep increase to 106° within a few metres of the overlying Unit 9 peridotite body (Fig. 9).
|
|
The traverses of the Unit 8/9 allivalite both above and below the Unit 9 peridotite are very different from those described previously (Fig. 9). The traverse below the peridotite (Traverse i) shows an increase of median angle some metres below the contact, whereas those described previously show an increase in angle only so close to the contact (a few tens of centimetres at the most) that it is likely to be sampled and observed only if the contact itself is exposed. The traverses above the peridotite show a wide variation in angle, with a broad downwards trend in their basal portions, followed by an upwards trend. These sinusoidal curves dip downwards from, and return to, the range of median values in the bulk of Traverse f. In the case of Traverse e, the upwards trend flattens out to a relatively constant value of 87–94°, whereas Traverses h and g show an increase up to values of 110° before a steep decrease to
90° within a few centimetres of the top of the allivalite, where exposed. Although the initial downwards trend of Traverse e is exactly coincident with that of Traverses g and h, the subsequent upwards trend of the former is more shallow. This difference may not be significant. Traverse e was collected over a wide area, as the outcrop forms many stepped pavements, and so the vertical distance to the base of the allivalite is not clear. It is entirely possible that the Unit 9 peridotite domes up under the exposed allivalites of Traverse e, thus making the upwards trend on all three traverses identical.
|
DISCUSSION |
|---|
The essentially flat dihedral angle profiles of the traverses across the allivalites of Units 10 and 11a (Figs 6 and 7) are consistent with these horizons having formed by steady accumulation of crystals at the base of a magma chamber, with a constant rate of upwards movement of the solidification front in each. The 10° lower average angle in the bulk of Unit 11a compared with Unit 10 points to a somewhat faster rate of solidification of the former. The excursion to median angles as low as 74° in Traverse d is enigmatic. As it occurs over only 20 cm it is possible that this excursion is also present in Traverse b but was not sampled. It occurs in the centre of an otherwise unremarkable and uniform layer of foliated troctolite, and suggests either an infiltration of late melts or a period of unusually rapid crystallization. The lack of change in angles towards the base of the traverses across Units 10 and 11a, and the similarity of dihedral angles in the underlying peridotite in the case of Unit 10, are strong evidence that the underlying peridotite bodies are cumulate in character, and not intrusive. The localized upwards deflections at the top of each of these two units, where the upper contact is exposed, is most plausibly the result of the introduction of the picrite that formed the overlying macro-scale peridotites.
Importantly, there is no correlation between dihedral angle and the detailed stratigraphy of the allivalite units [with that of Unit 10 recorded by Brown (1956) and that of Unit 11a recorded as part of this study]. The presence of minor peridotite layers (up to 2 m thick in the case of Traverse d), or changes of the proportion of clinopyroxene, the grain size or the strength of preferred alignment of the plagioclase, does not affect the dihedral angle populations. Notably, the detailed stratigraphy of Unit 11a changes between the two exposures, with no obvious correlation between the two. This does not appear to have affected the dihedral angles, suggesting that solidification rates were uniform, despite changes in modal composition, over lateral distances of 100 m.
Comparison of Traverse f (Fig. 8) across the Unit 8/9 allivalite with those of Units 10 and 11a demonstrates many points of similarity. The median values of cpp in much of Traverse f are within the same range as in Unit 10 for most of the traverse. Traverse f differs from those across Units 10 and 11a in that the lowermost samples have abnormally low dihedral angles and the topmost samples do not show an upwards deflection. The lack of an upwards deflection may simply be due to lack of sampling of the (unexposed) uppermost few tens of centimetres. The departure from the expected pattern at the base of the allivalite may be due to an abnormally high solidification rate early in the accumulation of this allivalite, or to the infiltration of late-stage melts during compaction of the underlying peridotite (e.g. Tait, 1985). In any case, the dihedral angle profile for Traverse f is consistent with the overlying peridotite of Unit 10, and the underlying peridotite of Unit 8, resulting from a normal replenishment event.
In contrast, the dihedral angle profiles in the samples collected in the vicinity of the intrusive Unit 9 peridotite differ in important respects from those attributed to normal accumulation in a fractionating system. The profile in Traverse i is indistinguishable from that of Traverse f in its lower regions, whereas the well-defined increase up to the peridotite occurs over a distance of more than 3 m (Fig. 9). This contrasts with the more localized upwards deflection in the other allivalites, which is observed only in traverses in which the actual contact can be sampled. Additionally, the sigmoidal variation in dihedral angles above the intrusive body is strikingly different from the profile expected for normally accumulating allivalite horizons, and points to a very different evolution and thermal history. I suggest that these differences are due to the effects accompanying intrusion of picrite into allivalite.
The intrusion of picritic magma into a troctolitic mush is likely to have major thermal and chemical consequences, given that the magma is likely to have only olivine on the liquidus. The much greater scale of the upwards deflection of median cpp in the allivalite underlying the Unit 9 peridotite, compared with that at the top of the other allivalite units, points to a much greater thermal effect of an intrusive picrite body on the underlying allivalite compared with that associated with the emplacement of picrite above the allivalite pile. If the picrite sill intruded just above the level of the solidification front, this suggests that the difference in scale of the upwards deflection may be linked to the completely solid nature of the allivalite underlying the sill compared with the partially solidified layers of allivalite during the other replenishment events.
If the picrite sill did intrude at the level of the solidification front, the effects on the material overlying the Unit 9 peridotite are likely to be significant. Intrusion of minor (1 m thick) picrite layers into completely solid troctolite is believed to have resulted in localized assimilation of the troctolite to form finger structures (Butcher et al., 1985; Morse et al., 1987; Renner & Palacz, 1987) and the 10 m thick intrusive peridotite body in Unit 14 on Trallval (SW of Hallival, and not shown in Fig. 1) is associated with assimilation and clinopyroxene mobilization of the adjacent allivalite (Volker & Upton, 1990). The intrusion of a 25 m thick body is therefore likely to have effects on a commensurately larger scale.
The departure of the cpp profile from that expected for a normally accumulating allivalite extends 10 m above the peridotite body, with the midpoint of the sinusoidal curve of Traverses g and h, and e occurring just below the stratigraphic level of the Wavy Horizon (Fig. 10). I suggest that this link between the dihedral angles (caused by thermal and chemical effects of intrusion of an underlying picrite) and the position of the Wavy Horizon supports the idea that the Wavy Horizon is the field-scale manifestation of upwards mobilization of clinopyroxene caused by picrite intrusion into a partially solidified allivalitic mush (Hallworth, 1998).
|
What factors control whether a major body of replenishing picrite intrudes the pre-existing crystal pile at the base of the magma chamber? It may be related to a greater than usual thickness of the mushy zone. At 45 m (Traverse f) the complete Unit 8/9 allivalitic layer is certainly much thicker than the 15–20 m thickness of the other allivalites from Units 7, 10 and 11a. Interestingly, the other allivalite bodies that contain intrusive peridotite bodies [Unit 14, Renner & Palacz (1987)
, Volker & Upton (1990), and Unit 3, Brown (1956)] are also thicker than 15–20 m. However, field evidence of slumped layers suggests that the thickness of the upper mushy zone in the Rum allivalites was generally of the order of at least a few metres (Renner & Palacz, 1987; C. H. Emeleus, personal communication, 2003). The presence of vertical, 5 m high, feldspathic streamers in peridotite horizons on Trallval (Volker & Upton, 1990) is suggestive of the pipe-like structures expected in a crystallizing mush with compositional convection (e.g. Tait & Jaupart, 1996), suggesting a mushy zone up to 5 m thick in the peridotites. However, similar clinopyroxene streamers in the Trallval allivalites occur only on the centimetre scale (Volker & Upton, 1990), suggesting a much thinner mushy layer. Further evidence for the thickness of the mushy zone in Unit 8/9 is provided by the dihedral angle profiles.
The median cpp angle in Traverses e and f are nearly coincident within the upper 7–8 m of Unit 8/9 (Fig. 9). Given that the intrusion of the picrite most probably occurred in a partially consolidated mush, and that a median value of cpp of about 90° is related to the sub-solidus thermal history of normally accumulating allivalite, the return of the median cpp to the normal range in the upper reaches of the Unit 8/9 allivalite implies that the underlying material had solidified, with resumption of normal conditions within the mush. This gives an upper bound to the possible range of mushy layer thickness of 12 m at Traverse e, which, given the likelihood that the actual distance to the peridotite is less than this, reduces to 8 m if we consider Traverses g and h. [Note that the sample collected from the top of Traverse h (clearly identifiable as the outcrop is overlain by a thin skin of peridotite) has a high median value of cpp: I interpret this as analogous to the increases in angle seen elsewhere in the topmost regions of Units 10 and 11a]. For comparison, the intrusive body of peridotite 5 m from the top of the Unit 14 allivalite was interpreted by Renner & Palacz (1987) to be an apophysis of the overlying Unit 15 peridotite, pointing to a 5 m thickness of mushy material. It may be that intrusion of replenishing picrite is favoured by a thicker than usual (i.e. >5 m) mushy layer. Under conditions where the solidification front is close to the top of the crystal pile, replenishing magma is emplaced on top.
|
CONCLUSIONS |
|---|
In this contribution I have shown that it is possible to obtain novel and useful information about the super- and sub-solidus thermal history of cumulates from a detailed examination of textures; in particular, measurement of dihedral angles of the interstitial phases. When placed in a spatially well-defined, field-scale, context, this information can be used to distinguish between normal accumulation of a chemically stratified crystal pile at the base of a magma chamber and the juxtaposition of horizons of different composition by intrusion.
Application of this technique to the Rum Layered Intrusion has shown that much of the material in the large-scale peridotite bodies exposed on Hallival was formed by crystal accumulation during fractionation, in agreement with previous work on Rum (e.g. Brown, 1956; Huppert & Sparks, 1980; Tait, 1985; Volker & Upton, 1990). Within the context of the limited area examined, the only macro-scale peridotite body formed by intrusion of picritic magma into a pre-existing allivalite mush was that of Unit 9. This intrusive replenishment event may be related to the unusually thick (10 m instead of 1 m) mushy layer present in the chamber at that time, and had profound effects on the overlying allivalite, forming the so-called Wavy Horizon.
The Rum Layered Intrusion has, however, had a complex history, and it is likely that intrusion of picrite into the pre-existing pile was ubiquitous, although perhaps not on the scale seen in Unit 8/9. This was recognized by Renner & Palacz (1987) and Volker & Upton (1990) in their studies of the Unit 14 allivalite. Similarly, there are clearly several minor intrusions of peridotite in the Unit 10 allivalite on Cnapan Breaca (Fig. 1b; Bédard et al., 1988), some of which appear to merge with the underlying peridotite (J. Bédard, personal communication, 2004). I think, therefore, that an understanding of the mode of replenishment of the Rum Layered Series is thus fundamentally one of scale. Intrusions of replenishing picritic magma were indeed very common, but in the middle section of the ELS major replenishment events were dominated by emplacement of the new magma on top of the crystal pile.
|
ACKNOWLEDGEMENTS |
|---|
This research was funded in part by the European Community's Human Potential Programme under contract HPRN-CT-2002-000211 [EUROMELT]. Scottish Natural Heritage granted permission to conduct fieldwork on the Isle of Rum (project no. RG 458). I thank Mike Bickle, Mike Cheadle, Henry Emeleus, Stephen Siklos and Mark Hallworth for stimulating discussions. Mark Hallworth, Joseph Barraud and the staff of the Oxford University Museum of Natural History loaned samples, and Steve Laurie was invaluable in providing access to material in the Harker Collection. Eleanor Holness and Stephen Siklos provided assistance in the field.
* Corresponding author. E-mail: marian{at}esc.cam.ac.uk
|
REFERENCES |
|---|
Bacon, C. R. & Metz, J. (1984). Magmatic inclusions in rhyolites, contaminated basalts and compositional zonation beneath the Coso volcanic field, California. Contributions to Mineralogy and Petrology 85, 346–365.[CrossRef][Web of Science]
Bédard, J. H. & Sparks, R. S. J. (1991). Comment on: ‘The structure and petrogenesis of the Trallvall and Ruinsival areas of the Rhum ultrabasic complex’. By J. A. Volker & B. J. G. Upton. Transactions of the Royal Society of Edinburgh: Earth Sciences 82, 389–390.[Web of Science]
Bédard, J. H., Sparks, R. S. J., Renner, R., Cheadle, M. J. & Hallworth, M. A. (1988). Peridotite sills and metasomatic gabbros in the Eastern Layered Series of the Rhum Complex. Journal of the Geological Society, London 145, 207–224.
Brown, G. M. (1956). The layered ultrabasic rocks of Rhum, Inner Hebrides. Philosophical Transactions of the Royal Society of London, Series B 240, 1–53.[CrossRef]
Butcher, A. R., Young, I. M. & Faithfull, J. W. (1985). Finger structures in the Rhum complex. Geological Magazine 122, 491–502.[Abstract]
Campbell, I. M. (1978). Some problems with the cumulus theory. Lithos 11, 311–323.[CrossRef][Web of Science]
Cmíral, M., Fitz Gerald, J. D., Faul, U. H. & Green, D. H. (1998). A close look at dihedral angles and melt geometry in olivine-basalt aggregates; a TEM study. Contributions to Mineralogy and Petrology 130, 336–345.[CrossRef][Web of Science]
Dunham, A. C. & Wadsworth, W. J. (1978). Cryptic variation in the Rhum Layered Intrusion. Mineralogical Magazine 42, 347–356.[Web of Science]
Emeleus, C. H. (1994). 1:20 000 solid geology map of Rum, 2nd edn. Edinburgh: Scottish Natural Heritage.
Emeleus, C. H. (1997). Geology of Rum and the adjacent islands. Memoir of the British Geological Survey, Sheet 60 (Scotland).
Emeleus, C. H., Cheadle, M. J., Hunter, R. H., Upton, B. G. J. & Wadsworth, W. J. (1996). The Rum Layered Suite. In: Cawthorn, R. G. (ed.) Layered Intrusions. Amsterdam: Elsevier, pp. 403–439.
Hallworth, M. A. (1998). Laboratory investigations of geological fluid flows. Ph.D. thesis, Anglia Polytechnic University, Cambridge.
Harker, D. & Parker, E. R. (1945). Grain shape and grain growth. Transactions of the American Society of Metals 34, 156–195.
Harte, B., Pattison, D. R. M. & Linklater, C. M. (1991). Field relations and petrography of partially melted pelitic and semi-pelitic rocks. In: Voll, G., Töpel, J., Pattison, D. R. M. & Seifert, F. (eds) Equilibrium and Kinetics in Contact Metamorphism: the Ballachulish Igneous Complex and its Aureole. Heidelberg: Springer, pp. 181–210.
Holness, M. B. (1999). Contact metamorphism and anatexis of Torridonian arkose by minor intrusions of the Rum Igneous Complex, Inner Hebrides, Scotland. Geological Magazine 136, 527–542.[Abstract]
Holness, M. B. & Clemens, J. D. (1999). Partial melting of the Appin Quartzite driven by fracture-controlled H2O infiltration in the aureole of the Ballachulish Igneous Complex, Scottish Highlands. Contributions to Mineralogy and Petrology 136, 154–168.[CrossRef][Web of Science]
Holness, M. B. & Watt, G. R. (2001). Quartz recrystallisation and fluid flow during contact metamorphism: a cathodoluminescence study. Geofluids 1, 215–228.[CrossRef]
Holness, M. B., Cheadle, M. J. & McKenzie, D. (2005). On the use of changes in dihedral angle to decode late-stage textural evolution in cumulates. Journal of Petrology 46, 1565–1583.
Hunter, R. H. (1987). Textural equilibrium in layered igneous rocks. In: Parsons, I. (ed.) Origins of Igneous Layering. Dordrecht: Reidel, pp. 473–503.
Hunter, R. H. (1996). Texture development in cumulate rocks. In: Cawthorn, R. G. (ed.) Layered Intrusions. Developments in Petrology 15, 77–101.
Huppert, H. E. & Sparks, R. S. J. (1980). The fluid dynamics of a basaltic magma chamber replenished by an influx of hot, dense, ultrabasic liquid. Contributions to Mineralogy and Petrology 75, 279–289.[CrossRef]
Jin, Z.-M., Green, H. W. & Zhou, Y. (1994). Melt topology in partially molten mantle peridotite during ductile deformation. Nature 372, 164–167.[CrossRef]
Jurewicz, S. R. & Jurewicz, A. J. G. (1986). Distribution of apparent angles on random sections, with emphasis on dihedral angle measurements. Journal of Geophysical Research 91, 9277–9282.
Kerr, R. C. & Tait, S. R. (1985). Convective exchange between pore fluid and an overlying reservoir of denser fluid: a post-cumulus process in layered intrusions. Earth and Planetary Science Letters 75, 147–156.[CrossRef][Web of Science]
Kretz, R. (1966). Interpretation of the shape of mineral grains in metamorphic rocks. Journal of Petrology 7, 68–94.
Laporte, D., Rapaille, C. & Provost, A. (1997). Wetting angles, equilibrium melt geometry, and the permeability threshold of partially molten crustal protoliths. In: Bouchez, J.-L., Hutton, D. H. & Stephens, W. E. (eds) Granite: from Segregation of Melt to Emplacement Fabrics. Norwell, MA: Kluwer Academic, pp. 31–54.
Marchildon, N. & Brown, M. (2002). Grain-scale melt distribution in two contact aureole rocks: implications for controls on melt localization and deformation. Journal of Metamorphic Geology 20, 381–396.[CrossRef][Web of Science]
McClurg, J. E. (1982). Petrology and evolution of the northern part of the Rhum Ultrabasic Complex. Ph.D. thesis, University of Edinburgh.
Morse, S. A., Owens, B. E. & Butcher, A. R. (1987). Origin of finger structures in the Rhum Complex: phase equilibrium and heat effects. Geological Magazine 124, 205–210.[Abstract]
Murphy, M. D., Sparks, R. S. J., Barclay, J., Carroll, M. R. & Brewer, T. S. (2000). Remobilisation of andesite magma by intrusion of mafic magma at the Soufrière Hills Volcano, Montserrat, West Indies. Journal of Petrology 41, 21–42.
Platten, I. M. (1981). Partial melting of feldspathic quartzite around late Caledonian minor intrusions in Appin, Scotland. Geological Magazine 119, 413–419.[Web of Science]
Renner, R. & Palacz, Z. (1987). Basaltic replenishment of the Rhum magma chamber: evidence from Unit 14. Journal of the Geological Society, London 144, 961–970.
Riegger, O. K. & Van Vlack, L. H. W. (1960). Dihedral angle measurement. Transactions of the Metallurgical Society of the AIME 218, 933–935.
Rosenberg, C. L. & Riller, U. (2000). Partial melt topology in statically and dynamically recrystallised granite. Geology 28, 7–10.
Sawyer, E. W. (1999). Criteria for the recognition of partial melting. Physics and Chemistry of the Earth 24, 269–279.[CrossRef][Web of Science]
Sawyer, E. W. (2001). Melt segregation in the continental crust: distribution and movement of melt in anatectic rocks. Journal of Metamorphic Geology 19, 291–309.[CrossRef][Web of Science]
Tait, S. R. (1984). Fluid dynamical processes in the formation of latered igneous rocks. Ph.D. thesis, University of Cambridge.
Tait, S. R. (1985). Fluid dynamic and geochemical evolution of cyclic Unit 10, Rhum, Eastern Layered Series. Geological Magazine 122, 469–484.[Abstract]
Tait, S. R. & Jaupart, C. (1996). The production of chemically stratified and adcumulate plutonic igneous rocks. Mineralogical Magazine 60, 99–114.[Abstract]
Tepley, F. J., III & Davidson, J. P. (2003). Mineral-scale Sr-isotope constraints on magma evolution and chamber dynamics in the Rum layered intrusion, Scotland. Contributions to Mineralogy and Petrology 145, 628–641.[CrossRef][Web of Science]
Upton, B. G. J., Skovgaard, A. C., McClurg, J., Kirstein, L., Cheadle, M., Emeleus, C. H., Wadsworth, W. J. & Fallick, A. E. (2002). Picritic magmas and the Rum ultramafic complex, Scotland. Geological Magazine 139, 437–452.
Urai, J. L., Speirs, C. J., Zwart, H. J. & Lister, G. S. (1986). Weakening of rock salt by water during long-term creep. Nature 324, 554–557.[CrossRef]
Vernon, R. H. (1968). Microstructures of high-grade metamorphic rocks at Broken Hill, Australia. Journal of Petrology 9, 1–22.
Vernon, R. H. (1970). Comparative grain-boundary studies of some basic and ultrabasic granulites, nodules and cumulates. Scottish Journal of Geology 6, 337–351.
Vernon, R. H. (1997). Comment: On the identification of textural disequilibrium in rocks using dihedral angle measurements. By Elliott, M. T. & Cheadle, M. J. Geology 25, 1055.
Volker, J. A. (1983). The geology of the Trallvall area, Rhum, Inner Hebrides. Ph.D. thesis, University of Edinburgh.
Volker, J. A. & Upton, B. G. J. (1990). The structure and petrogenesis of the Trallvall and Ruinsival areas of the Rhum ultrabasic complex. Transactions of the Royal Society of Edinburgh: Earth Sciences 81, 69–88.[Web of Science]
Volker, J. A. & Upton, B. G. J. (1991). Reply to comments by J. H. Bédard & R. S. J. Sparks. Transactions of the Royal Society of Edinburgh: Earth Sciences 82, 391.[Web of Science]
Wadsworth, W. J. (1994). The peridotite plugs of northern Rum. Scottish Journal of Geology 30, 167–174.[Web of Science]
Worrell, L. M. (2002). The origin of igneous cumulates: integrated studies of peridotites from the Western Layered Series of the Rum Layered Intrusion. Ph.D. thesis, University of Liverpool.
Worrell, L. M., Cheadle, M. J., Coogan, L. A., Prior, D. J., Wheeler, J. & Toplis, M. J. (2003). A multidisciplinary approach to understanding the origin of peridotite cumulates. EOS Transactions, American Geophysical Union 84(46), Fall Meeting Supplement, V11F-02.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. R. NICOLL, M. B. HOLNESS, V. R. TROLL, C. H. DONALDSON, E. P. HOLOHAN, C. H. EMELEUS, and D. CHEW Early mafic magmatism and crustal anatexis on the Isle of Rum: evidence from the Am Mam intrusion breccia Geological Magazine, May 1, 2009; 146(3): 368 - 381. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. HOLNESS and B. WINPENNY The Unit 12 allivalite, Eastern Layered Intrusion, Isle of Rum: a textural and geochemical study of an open-system magma chamber Geological Magazine, May 1, 2009; 146(3): 437 - 450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Holness, S. A. Morse, and C. Tegner Response to Comment by McBirney, Boudreau and Marsh J. Petrology, January 1, 2009; 50(1): 97 - 102. [Full Text] [PDF]
|
|
|
|
|
|
B. O'Driscoll, C. T. E. Stevenson, and V. R. Troll Mineral Lamination Development in Layered Gabbros of the British Palaeogene Igneous Province: A Combined Anisotropy of Magnetic Susceptibility, Quantitative Textural and Mineral Chemistry Study J. Petrology, June 1, 2008; 49(6): 1187 - 1221. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. R. TROLL, G. R. NICOLL, C. H. DONALDSON, and H. C. EMELEUS Dating the onset of volcanism at the Rum Igneous Centre, NW Scotland Journal of the Geological Society, May 1, 2008; 165(3): 651 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Holness, C. Tegner, T. F. D. Nielsen, G. Stripp, and S. A. Morse A Textural Record of Solidification and Cooling in the Skaergaard Intrusion, East Greenland J. Petrology, December 1, 2007; 48(12): 2359 - 2377. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Holness, A. T. Anderson, V. M. Martin, J. Maclennan, E. Passmore, and K. Schwindinger Textures in Partially Solidified Crystalline Nodules: a Window into the Pore Structure of Slowly Cooled Mafic Intrusions J. Petrology, July 1, 2007; 48(7): 1243 - 1264. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.B. Holness Textural immaturity of cumulates as an indicator of magma chamber processes: infiltration and crystal accumulation in the Rum Eastern Layered Intrusion Journal of the Geological Society, May 1, 2007; 164(3): 529 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Holness, M. A. Hallworth, A. Woods, and R. E. Sides Infiltration Metasomatism of Cumulates by Intrusive Magma Replenishment: the Wavy Horizon, Isle of Rum, Scotland J. Petrology, March 1, 2007; 48(3): 563 - 587. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. O'Driscoll, C. H. Donaldson, V. R. Troll, D. A. Jerram, and C. H. Emeleus An Origin for Harrisitic and Granular Olivine in the Rum Layered Suite, NW Scotland: a Crystal Size Distribution Study J. Petrology, February 1, 2007; 48(2): 253 - 270. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Holness, T. F. D. Nielsen, and C. Tegner Textural Maturity of Cumulates: a Record of Chamber Filling, Liquidus Assemblage, Cooling Rate and Large-scale Convection in Mafic Layered Intrusions J. Petrology, January 1, 2007; 48(1): 141 - 157. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. HOLNESS, M. J. CHEADLE, and D. McKENZIE On the Use of Changes in Dihedral Angle to Decode Late-stage Textural Evolution in Cumulates J. Petrology, August 1, 2005; 46(8): 1565 - 1583. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||