Journal of Petrology Advance Access published online on April 3, 2007
Journal of Petrology, doi:10.1093/petrology/egm016
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Textures in Partially Solidified Crystalline Nodules: a Window into the Pore Structure of Slowly Cooled Mafic Intrusions
1Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EQ, UK
2Department of the Geophysical Sciences, University Of Chicago, 5734 S. Ellis Ave., Chicago, IL 60637, USA
3Department of Earth Sciences, University Of Durham, Durham DH1 3LE, UK
4School of Geosciences, University of Edinburgh, West Mains Road, Edinburgh EH9 3JW, UK
Received August 2, 2006; Revised typescript accepted March 8, 2007
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONAbundant glass is present along grain boundaries in coarse-grained, glass-bearing, crystalline gabbroic and peridotitic nodules entrained and erupted in lavas from Iceland, Santorini and Mauna Loa (Hawaii), even when the total porosity is less than a few volume per cent. The glass films vary from a few microns to a few tens of microns thick, and are associated with strings of small lensoid grain boundary pockets formed by impingement during crystal growth. Additional porosity occurs as extensive liquid-filled pockets adjacent to included grains within oikocrysts and as large triangular pockets formed by impingement of planar-sided grains. Interstitial material within glass films, and the irregularity of film thickness along a single grain boundary, suggest that the present pore structure is representative of the pore structure before entrainment and eruption. Pore geometry is consistent with a dominant control by crystal growth during solidification, with little or no evidence for control by minimization of internal energies driven by textural equilibration. Similarities between liquid distribution in the crystalline nodules and that of late-stage, interstitial phases in fully solidified mafic cumulates from the Rum and Skaergaard intrusions demonstrate that the crystalline nodules provide information about the latest stages of solidification in slowly cooled mafic plutons. The highly permeable network of intersecting liquid films, lenses and pockets may promote in situ crystallization in the solidifying mush, explaining the common presence of adcumulates in such intrusions.
KEY WORDS: textures; liquid distribution; mafic cumulates; crystalline nodules
|
INTRODUCTION |
|---|
Given the importance of mafic intrusions in the formation and evolution of the Earth's crust, it is essential to understand the physical and chemical evolution of slowly cooled, mafic, crystal mushes. A particular lacuna in our understanding of these bodies concerns the last stages of solidification, when the volume fraction of liquid is small. This poses problems for attempts to understand the mechanism of adcumulate formation, for which significant mass movement is required even at vanishingly low porosities (e.g. Morse, 1986
; Tait & Jaupart, 1996; Meurer & Meurer, 2006). It is also thrown into relief when considering recent work demonstrating the importance of late-stage liquid migration in solidifying mafic crystal mushes caused by liquid expulsion from underlying horizons by compaction (e.g. McKenzie, 1984; Shirley, 1987). Passage of chemically evolved liquids through a mush comprising early-formed primocrysts may result in metasomatism and recrystallization (e.g. McBirney & Sonnenthal, 1990; Boudreau & McCallum, 1992; Sonnenthal, 1992; Tait & Jaupart, 1992, 1996; Bergantz, 1995; Mathez, 1995; Boudreau & Meurer, 1999). These processes may be further complicated by the development of immiscibility in the interstitial liquid (e.g. Jakobsen et al., 2005), which may play a role in the development of commercially important ore deposits (e.g. Andersen et al., 1998). Critical to our understanding of these processes is liquid distribution and connectivity in the later stages of solidification.
On a large scale (100 m), compositional convection in the crystal mush on a magma chamber floor may result in the creation of widely spaced, high-permeability, chimney structures, which channel the bulk of the mobile liquid phase (Tait & Jaupart, 1992). Meurer & Claeson (2002) and Meurer & Meurer (2006) documented perhaps analogous centimetre-scale features caused by channelling of late-stage flow in almost completely solidified cumulates. However, although such large-scale channels undoubtedly play an important role in cumulate evolution, it is the pore structure and permeability of the bulk mush between such channels that is of interest here.
The early stages of pore occlusion in solidifying granites have been demonstrated using 3-D mapping of serial sections (Bryon et al., 1995, 1996), and in mafic and ultramafic cumulates from consideration of dihedral angles at pore corners (Holness et al., 2005a). As a result of significant sub-solidus textural modification in mafic cumulates (e.g. Hunter, 1987; Boorman et al., 2004; Holness et al., 2005a), efforts to read back through the final texture to gain insight into the pore geometry and connectivity in rocks in which the liquid fraction is below 
5 vol. % are difficult. Pore geometries in such liquid-poor rocks are generally investigated experimentally (e.g. Faul et al., 1994; Laporte & Watson, 1995; Minarik & Watson, 1995; Lupulescu & Watson, 1999; Wark et al., 2003) or by computation (e.g. von Bargen & Waff, 1986; Cheadle et al., 2004). Such studies are mainly directed at understanding connectivity during melting (although see Cheadle et al., 2004), which does not necessarily result in the same textures as the reverse process of solidification. Furthermore, these studies are all based on various assumptions concerning the extent of textural equilibrium and may not reflect what actually occurs in naturally solidifying systems. In this contribution, we attempt to circumvent these difficulties by direct observation of natural examples of partially solidified, slowly cooled, mafic rocks provided by glass-bearing gabbroic and peridotitic crystalline nodules entrained and erupted by mafic magmas.
Glass-bearing nodules occur in association with basaltic scoria in all major tectonic environments: ocean ridge, hotspot and subduction zone. Although some of these nodules occur in alkaline basalts and are probably of mantle or high-pressure origin, others occur in tholeiitic basalt, are of comparatively shallow origin and formed in the subvolcanic environment. Many studies of such shallow-level nodules have been published (e.g. Becker, 1977; Hermes & Cornell, 1981; Tait, 1988; de Silva, 1989; Tait et al., 1989; Turbeville, 1992, 1993; Mattioli et al., 2003; Holness et al., 2005b; Holness & Bunbury, 2006; Martin et al., 2007a) but generally these studies are concerned with fractionation trends and unravelling the history of chamber replenishment and eruption triggers, rather than looking in detail at the progressive occlusion of porosity in the crystallizing mush. Here we present an observational study of pore distribution and late-stage crystallization in suites of coarse-grained, glass-bearing but almost completely crystallized, crystalline nodules collected from several volcanoes in a variety of tectonic settings (Iceland—ocean ridge; Hawaii—hotspot; Santorini—subduction zone). These nodules are thought to represent almost completely solidified crystalline material from the margins of magmatic conduits and chambers, and provide a snapshot of textural evolution during the last stages of solidification in an otherwise inaccessible environment.
Differences in the relative rates of crystal nucleation, crystal growth, and textural modification by minimization of internal energies may result in differences in the late-stage pore structure in slowly cooled compared with more rapidly cooled rocks. We concentrate on textures related to residual liquid in coarse-grained nodules that are of a similar grain size to, and thus likely to be the direct analogues of, completely solidified large mafic intrusions. From direct observations of glass distribution in such incompletely crystallized material we can infer where the late-stage porosity may have been in completely solidified cumulates in large mafic intrusions. In this contribution we concentrate on the Rum Layered Intrusion in Scotland and the Skaergaard Intrusion of East Greenland, although we have found the same features in many other such intrusions (e.g. Stillwater, Bushveld, and the Freetown Complex of Sierra Leone).
|
GEOLOGICAL SETTINGS—NODULE LOCALITIES |
|---|
Kameni, Santorini
Santorini Volcano, Greece, lies above the subduction zone in the Eastern Mediterranean (Fig. 1). After the Minoan eruption 3600 years ago (Hammer et al., 1987
), volcanic activity resumed within the caldera, forming a 2·5 km3 intra-caldera volcano, the summit of which rises 500 m above the submarine caldera floor (Druitt et al., 1999) and 126 m above sea level. After breaking the surface in 197 BC, to form the Kameni Islands, there have been at least nine subaerial eruptions of consistently dacitic magma, the last of which was in 1950. The volume of magma erupted during each of these episodes was 0·07 x 106 to 140 x 106 m3 (Higgins, 1996). The lava flows contain a variety of crystal-rich glassy mafic enclaves, the majority of which are andesitic. These enclaves are interpreted to be the remnants of a layer of replenishing magma ponded at the base of the chamber, which was disrupted and overturned immediately prior to each eruption (Martin et al., 2007a). The 1950 lava dome also contains these glass-rich andesitic enclaves but is unique in containing a significant number of small, almost wholly crystalline nodules (which are present, albeit rare, in the earlier flows). The nodules are troctolitic or gabbroic, coarse-grained (with millimetre- to centimetre-sized crystals), with a low glass and vesicle content, and may be up to 5 cm in diameter, with a usual size range of 1–2 cm. They are sub-rounded to angular and many are often enveloped in less dense, highly vesicular crystal-rich material similar to that of the more common type of enclaves.
|
Damon's Cone, Mauna Loa, Hawaii
The active volcano of Mauna Loa is near the southeastern end of the age-progressive chain of volcanoes forming the Hawaiian Islands and the Emperor Seamount Chain. Mauna Loa is the largest of the five coalesced shield volcanoes forming the island of Hawaii (Fig. 2). It has a summit caldera and, in common with most of the other volcanoes of the Hawaiian Islands, elongate, arcuate rift zones. Most of the lava forming the surface of Mauna Loa erupted from the caldera and the SW and NE rift zones (Langenheim & Clague, 1987
, and earlier works referenced therein). The lavas are divided into three age groups: the oldest Ninole Basalt [which is up to about a few hundred thousand years old (Clague & Dalrymple (1987), and is not shown in Fig. 2]; the unconformably overlying Kahuku Basalt; and the youngest Kau Basalt, which covers most of the surface of Mauna Loa and is separated from the Kahuku Basalt by the Pleistocene Pahala Ash (Langenheim & Clague, 1987).
|
Damon's Cone is the remains of a cinder cone, unnamed on the (1:24 000) Papa Quadrangle topographic map, at an elevation of 265 m and located about 1 km SE of Puu Ohohia and about 1 km west of Mauna Loa's SW rift zone (Fig. 2). It lies on undifferentiated lava grouped together by Lockwood & Lipman (1987
) as their Group III (1·5–0·75 ka in age). Much of the scoria has been dispersed by the wind. This wind erosion formed a deflation armour containing abundant, 1–3 cm in diameter, nodules. The nodule population comprises both porous gabbros and dense peridotites. These nodules were very numerous in the 1970s when the bulk of the sample collection was made. However, encroachment of settlements into the area, and the use of off-road vehicles on Damon's Cone, has resulted in the almost complete destruction of the natural deflation pavement. Sampling was concentrated on the dense peridotitic nodules, because these are unusual, especially for a tholeiitic volcano.
The basalts of Mauna Loa are characteristically silica-rich (49–52 wt % SiO2), olivine porphyritic and many contain orthopyroxene, augite and plagioclase phenocrysts (MacDonald, 1949; Wright, 1971). Two of the historic basalts (1843 and 1950) are relatively differentiated and these have plausibly evolved ‘by removal of about 5–15% of silicates other than olivine’ (Wright, 1971, p. 28).
Saxi, Fontur and Brandur tuff cones, Iceland
Saxi, Fontur and Brandur are large tuff cones located to the east of Þórisvatn, in the northern part of the Veiðivötn–Vatnaöldur fissure system (Fig. 3), and have been proposed as candidates for the source of the giant, 8600-year-old (Hjartarson, 1988), Þjórsá basaltic flow (Hansen & Grönvold, 2000; Thordarson et al., 2003). Saxi and Fontur are separated by a linear, 1·5 km long, fissure, which is oriented parallel to others in the Veiðivötn–Vatnaöldur system. Brandur is located some 3 km to the west of the fissure line on the edge of Austurbotnar. All three cones are of similar size, rising only a few tens of metres from the surrounding plains, and with well-developed central craters. They are composed predominantly of a glassy, poorly consolidated ash (Hansen & Grönvold, 2000), containing a mixture of volcanic bombs between 10 and 100 mm across (glassy, scoriaceous matrix with olivine, plagioclase and clinopyroxene phenocrysts and glomerocrysts), together with fresh basaltic scoria, fragments of bright orange hyaloclastite (possibly country rock), and abundant sub-rounded gabbroic nodules.
|
The gabbroic nodules liberally cover the sides and tops of the cones, but are less abundant on the plains between the cones, indicating minimal erosional transport. Representative suites of samples were collected from different areas on each cone (see Table 1 for grid references and descriptions of sample localities). The nodules vary in size (10–100 mm in length), grain size (<0·5 mm overall to >10 mm for mono-crystalline feldspars), and modal mineral proportions. There is little or no variation in grain size towards the margins.
|
|
GEOLOGICAL SETTINGS—MAFIC INTRUSIONS |
|---|
Skaergaard, East Greenland
The Skaergaard Intrusion is part of a group of gabbroic and syenitic bodies formed in East Greenland during the opening of the Atlantic at about 55 Ma (Deer, 1976
; Brooks & Nielsen, 1982; Tegner et al., 1998). It is a shallow-level body, 8 km x 11 km x 4 km in size (Nielsen, 2004), of relatively evolved tholeiitic basaltic magma intruded at the contact between underlying Precambrian gneisses and an overlying 2 km thick sequence of Tertiary plateau lavas. It fractionated as a closed system to form the prime example of shallow magmatic differentiation.
Tilting of the eastern coast of Greenland by about 20° (Wager & Brown, 1968; McBirney, 1996), associated with regional stretching, has resulted in the almost continuous exposure of >3·5 km of stratigraphy, which dips gently to the SE. The intrusion has been divided into three major units: the Layered Series, formed on the floor of the intrusion; the Marginal Border Series, crystallized inwards from the walls; and the Upper Border Series, which grew downwards from the roof. The Layered Series is subdivided into Lower, Middle and Upper Zones by the disappearance of abundant primary olivine at the base of Middle Zone and its reappearance at the base of the Upper Zone. Further details of the intrusion have been given by McBirney (1996).
Rum, Inner Hebrides, Scotland
The Tertiary (60·53 ± 0·08 Ma, Hamilton et al., 1998) layered intrusion on Rum was emplaced at <0·5 kbar (Holness, 1999) into Precambrian arkose of the Torridonian Group on the east, and into the earlier granite on the west. Early activity on Rum was dominated by acid magmatism, with a granitic body intruded during doming and uplift within a ring fault. Subsequent caldera-forming collapse was followed by further central uplift and emplacement of a layered series of ultramafic and mafic rocks.
The Rum Layered Suite, which extends over an ellipsoidal area 10 km x 5 km, formed in an episodically replenished sill-like magma chamber of perhaps 200 m depth overlying a considerable crystal pile (Emeleus et al., 1996). The Layered Suite comprises olivine-rich cumulates (feldspathic peridotites), troctolites and gabbros (known collectively by the local term allivalite), and is divided into a Western, Central, and Eastern Layered Intrusion. The Eastern Layered Intrusion comprises 16 macro-units [defined originally as a set of 15 by Brown (1956) with an additional unit added by Volker & Upton (1990)], each comprising a lower peridotite and an overlying allivalitic horizon. The alternating rock types resulted from accumulation of crystals on the chamber floor from a sequence of magma injections into the Rum chamber (e.g. Tepley & Davidson, 2003; Worrell et al., 2003). The majority of these periodic replenishments were by picritic liquids (Upton et al., 2002), although some were basaltic (Renner & Palacz, 1987). Further details of the geological history, and the primary published sources of this information, have been given by Emeleus et al. (1996) and Emeleus (1997).
|
PETROGRAPHY OF THE NODULES |
|---|
Kameni, Santorini
The suite of samples examined here are predominantly from the 1950 dome, but two nodules were collected from the 1925–1928 lava flow (in which such nodules are very rare). The nodules fall into several compositionally and texturally distinct groups. The gabbroic nodules are medium-grained (Fig. 4a) and contain 60–70 vol. % plagioclase and 16–24% clinopyroxene, with some olivine crystals. The plagioclase is tabular and commonly zoned, with cores of An87–89 and rims of An86–90 (Martin, 2005
). Plagioclase encloses clumps of sub-rounded to subhedral (primocrystic) clinopyroxene grains, and rare, rounded, primocrystic olivine grains (unzoned, Fo76–79). Some clinopyroxene oikocrysts enclose abundant rounded olivine (Fig. 4b). Brown glass comprises <10 vol. % of the nodules and contains abundant microlites of plagioclase, oxides and pyroxene. The liquid (which is now glass) occurs in interstitial pores within the clumps of pyroxene grains (Fig. 4a), in triangular pores formed by impingement of plagioclase grains (Figs 4c and 5a), or in small scattered pores on the grain boundaries between plagioclase grains. (These pores are most clearly visible in nodules in which the liquid does not contain microlites. For illustrated examples of these pore geometries, the reader is referred to the sections describing the Icelandic and Hawaiian nodules.)
|
|
In the lower porosity (less than a few volume per cent) gabbroic nodules, clino- and orthopyroxene form compact grains with elongate cuspate extensions down adjacent plagioclase–plagioclase grain boundaries (Fig. 4d), indicating interstitial growth subsequent to the incorporation of pyroxene primocrysts in the mush. Clinopyroxene is locally replaced by brown amphibole. In one of these nodules, rare, primocrystic, partially resorbed olivine grains are rimmed by orthopyroxene (Fig. 4e), and are locally replaced by a symplectite of oxide grains and orthopyroxene via a non-isochemical reaction of the type olivine + TiO2 + SiO2 = orthopyroxene + magnetite + ilmenite (Drüppel et al., 2001
). In another nodule of this type, the partially resorbed, symplectite-bearing, olivine grains are surrounded by large clinopyroxene grains (Fig. 4f). The troctolitic nodules from the 1950 dome commonly comprise 75–85 vol. % coarse (up to 5 mm), zoned (cores An75–93, rims An60), plagioclase grains with multiple and complex twinning, growing from a central nucleation site, with compositionally zoned growth bands marked by concentrations of melt inclusions (Fig. 6a), and terminal growth facets. These large crystals enclose numerous small, sub-rounded, zoned grains of olivine (cores of Fo78–82, rims of Fo73–80) and, more rarely, marginal grains of clinopyroxene. Liquid in these nodules is mostly contained in small pores and liquid films on plagioclase grain boundaries (Fig. 6b, and see images of Icelandic and Hawaiian nodules). Rarely, liquid-filled pores occur on plagioclase–olivine grain boundaries, and interstitially within clumps of included olivine grains (Fig. 6b).
|
A further type of troctolitic nodule comprises coarse-grained, randomly oriented plagioclase, with abundant rounded olivine and sub-rounded clinopyroxene primocrysts. Liquid forms films and pockets on irregular plagioclase–plagioclase grain boundaries. Liquid films also commonly separate olivine from plagioclase, and these films contain small grains of orthopyroxene nucleating on the olivine wall (Fig. 5d–f). Pyroxene also grows on olivine surfaces adjacent to the large liquid-filled pores formed by impingement of tabular plagioclase (Fig. 6c and d).
Damon's Cone, Mauna Loa, Hawaii
The nodules fall into five clearly defined types. The predominant type is gabbroic, rich in vesicular glass, with a framework of randomly oriented, tabular, plagioclase grains together with euhedral grains of orthopyroxene and clinopyroxene (Fig. 7a). Chromite, olivine and sulphide blebs are minor components. All grains are rich in melt inclusions (Fig. 7b). There are no quench-related overgrowths, and the original impingement texture is unmodified (see Holness et al., 2005a).
|
The second type is defined by accumulations of olivine grains with minor faceted to sub-rounded chromite grains, and abundant interstitial liquid (Fig. 7c and d). Olivine grains in contact with liquid are rounded or subhedral, with some approach to liquid–solid equilibrium at olivine–olivine–liquid junctions (Fig. 7c and d). Olivine grains are commonly cemented by orthopyroxene oikocrysts, and many such nodules comprise 1–3 large (5–10 mm) oikocrysts. The olivine grains in orthopyroxene-cemented nodules generally have rounded and embayed outlines indicative of resorption (Fig. 7e and f). In general, the enclosure of the olivine [and the minor chromite (Fig. 8e)] grains is not total, with elongate lenses and rounded liquid pockets on one or more sides of the included grains (Figs. 7e, f, 8, and 9). Where in contact with liquid, olivine grains are faceted (Fig. 8d). Orthopyroxene–liquid interfaces are invariably non-faceted, and may be cuspate (Figs 8 and 9). Clinopyroxene nucleates on the orthopyroxene walls of liquid pockets, with a topotactic relationship (Fig. 8c and f). There is generally a continuous, or semi-continuous, liquid film of variable thickness separating adjacent orthopyroxene oikocrysts.
|
|
The third type contains very little liquid (< 2 vol. %), and is dominated by aggregates of rounded, or semi-faceted olivine (±clinopyroxene, orthopyroxene) primocrysts enclosed by plagioclase oikocrysts (Fig. 10a), with interstitial clinopyroxene (Fig. 10b). Planar-sided liquid pockets occur between adjacent plagioclase grains (Figs 9a and 10b), although there are no elongate liquid pockets along olivine–plagioclase grain boundaries.
|
All remaining nodules are either accumulates of rounded to subhedral fine-grained orthopyroxene primocrysts (together with minor plagioclase, clinopyroxene, olivine and chromite) with abundant vesicular brown glass (Fig. 10c), or coarse-grained (several millimetres), partially sintered, vesicular glass-bearing aggregates of subhedral olivine and orthopyroxene (of which the orthopyroxene grains commonly contain inclusions of partially resorbed olivine, Fig. 10d). In the latter, the grain boundaries (either olivine–olivine, or olivine–orthopyroxene) commonly contain elongate liquid lenses or triangular pockets formed by the impingement of non-planar growth faces (Figs 8 and 9).
The composition of the olivines in the nodules falls in a narrow range, with average compositions of Fo80 to Fo75. Within any single nodule the compositional range is <1 mol % Fo. The average composition of the orthopyroxenes is En76·5Fs19Wo4·5 and that of the clinopyroxenes is En50Fs11·5Wo38·7. The plagioclases have a larger range of composition than the olivines and pyroxenes, with average compositions ranging from An70·1 to An77·6, with a maximum compositional range in a single nodule of 2·4 mol % An. The plagioclase compositions correlate with the percentage of interstitial glass, with the more glass-rich nodules containing more anorthitic plagioclase.
Saxi, Fontur and Brandur tuff cones, Iceland
All nodules contain 50–90 vol. % primocrystic, tabular to sub-rounded, melt inclusion-rich, complexly zoned plagioclase (An93–82) (Fig. 11a–c). The most plagioclase-rich nodules occur at Brandur. Plagioclase primocrysts show signs of partial resorption in some nodules, with sieve-like rims (Fig. 11d) or a scattering of rounded grains around a more compact core, suggestive of some recrystallization after remelting (Fig. 11e). Grain size may be uniform, or highly variable on a thin-section scale. Other primocrysts comprise rare grains of sub-rounded to subhedral cumulus olivine (Fo86–79), and abundant (up to 30 vol. %) clinopyroxene (Mg-number 82), which may also be interstitial (Fig. 11b and c). The primocryst framework contains variable amounts of brown vesicular glass, which may contain plumose intergrowths of plagioclase and clinopyroxene nucleated on plagioclase primocrysts (Fig. 11f). Fine-grained interstitial clinopyroxene clumps also occur (Fig. 11d). The plagioclase primocrysts commonly have chemically and optically distinct faceted rims.
|
Plagioclase–plagioclase grain boundaries in the coarse-grained nodules contain abundant lensoid liquid-filled pores, formed by the impingement of non-planar growth faces (Fig. 12a and b). Clinopyroxene grows in these pores (Fig. 12b). Plagioclase–plagioclase boundaries in both coarse- and fine-grained nodules contain continuous liquid films of thickness varying from a few microns (in which case the film has been quenched in a partially broken-down state, Fig. 12c) to some tens of microns (Fig. 12d). These thicker films may contain clinopyroxene (Fig. 12d).
|
The boundaries between plagioclase and olivine primocrysts are mainly liquid-free with isolated pockets near the end of the boundary (Fig. 12e), bounded by a rounded olivine face and a planar plagioclase face. The liquid pockets may be pseudomorphed by clinopyroxene. Clinopyroxene only rarely forms an extensive rim to the olivine (Fig. 12f).
|
SUMMARY AND INTERPRETATION OF LIQUID DISTRIBUTION IN THE NODULES |
|---|
In general, the observations of pore geometry in the nodules confirm the conclusions of Holness et al. (2005a)
, who suggested that plagioclase-dominated cumulates have a pore structure that is essentially controlled by the impingement of planar-sided grains (the ‘impingement texture’ of Holness et al. (2005a), shown in this contribution in Figs 4c, 5a, 7a, b, 9a, and 11a–c), whereas olivine-dominated cumulates have a porosity controlled more by internal energy minimization (e.g. Figs 7c, d, and 10a, b). However, the new observations throw further light on this generalization.
The impingement texture of Holness et al. (2005a), dominated by planar-sided polygonal pores, is confined to highly porous plagioclase-rich nodules with fine- to medium grain size. Those nodules with low porosity and a coarse plagioclase grain size (and where the plagioclase has lost its earlier tabular form as a result of extensive, in situ, growth) tend to have a porosity concentrated along grain boundaries, as either isolated lenses or films. Olivine grains in olivine-dominated mushes are certainly rounded but at least some of this rounding may be a result of growth habit controlled by diffusion-limited growth (e.g. Figs 9a and 12e) rather than minimization of liquid–olivine interfacial energy.
The new observations of porosity in almost completely solidified mushes can be subdivided according to an (inferred) mode of formation, as follows.
Impingement of non-planar growth faces to form a grain boundary perpendicular to the mutual growth directions
The progressive approach of two growing grains to form an impingement boundary perpendicular to the growth direction generally results in the formation of isolated (at least in the 2-D sections involved here) grain boundary liquid pockets with a form dependent on the shape of the impinging growth faces and the relative orientation of the impinging grains. In all three sample suites, large (>0·1 mm) plagioclase grains grow with rounded steps on their boundaries. The impingement of two such stepped grains results in liquid-filled lenses (Fig. 12a and b). Similar lenses occur on olivine–olivine grain boundaries (Fig. 9b) and on boundaries between dissimilar phases (olivine–pyroxene: Figs 8a, b, 9b, d, and 10d; olivine–plagioclase: Figs 5c, d, and 12f). Although these features appear to be formed by impingement, and are thus part of an impingement texture, we consider them sufficiently different from the initial concept of an impingement texture (Elliott et al., 1997; Holness et al., 2005a) to merit a different name: ‘impingement lenses’.
Development of poly-phase grain boundaries parallel to the growth direction
Where grains of different phases grow together to form a grain boundary, the progressive occlusion of porosity by further growth parallel to the new grain boundary depends on grain morphology. In gabbroic nodules, plagioclase typically grows with planar surfaces whereas olivine and, to a lesser extent, pyroxene form rounded grains. This leads to the development of liquid-filled grooves at the junction (Figs 6d, 9a, b, and 12e). It is possible that the groove depth may be increased by diffusion-limited growth in confined spaces (e.g. Holness et al. 2005b).
Impingement of planar-sided grains
Tetrahedral pores bounded by planar crystal surfaces form a triangular pore in the plane of the thin-section (Fig. 5a–c). Subsequent occlusion of this pore by further growth of one of the bounding grains results in features such as that illustrated in Fig. 5b. Deep grooves at two-grain boundaries (Fig. 5a) are indicative of diffusion-limited growth in pore corners (Holness et al., 2005b), and the depletion of plagioclase components in such regions of the glass undoubtedly encourages nucleation of other phases.
Oikocrystic growth
Grain growth resulting in the enclosure of grains of another phase generally results in the formation of large pockets of liquid (Figs 7e, f, and 8). It is tempting to speculate that these pockets form on the downstream side (i.e. that opposite to where the included grain and the oikocryst first impinge) of the inclusions because of the incomplete closure of the oikocryst.
It is important to note that the liquid pockets around the olivine grains in the Hawaiian oikocrystic nodules are not related to any reaction consuming olivine. Although some olivine grains certainly show evidence of resorption by some reaction of the type olivine + liquid = orthopyroxene, the faceting of the olivine where it is in contact with liquid reveals that this resorption reaction preceded inclusion by the pyroxene oikocryst. Additional evidence that the liquid pockets surrounding the olivine inclusions are a primary and fundamental feature of such oikocrystic enclosure is given by the presence of identical features surrounding inclusions of chromite grains within the same oikocryst (Fig. 8e).
Grain boundary liquid films
Liquid films are very common in the nodules suite, ranging from the thin, unstable, films on plagioclase–plagioclase grain boundaries (Fig. 12c) to 50 µm thick films in which other phases may have nucleated and grown (Figs 5c–f, 8f, 9b–d, 10d, and 12d). These films may be of uniform thickness along the length of the grain boundary (e.g. Figs 9c and 12d) or of varying thickness (Fig. 5d–f). Films occur between two grains of the same phase, or on two-phase boundaries.
|
ARE THESE FEATURE PRIMARY? |
|---|
The relevance of the above observations to fully solidified mafic cumulates depends on the primary nature of the porosity distribution; that is, whether what we are seeing actually reflects the in situ liquid distribution in the pre-eruptive crystal mush. The process of nodule entrainment is poorly understood. It must involve fracturing, but the extent to which this fracturing affects regions between the fractures (which form the margins of the nodules) is not known. Following entrainment, decompression-related expansion of gas and volatile-bearing liquid can distort and fracture crystal mushes (e.g. Martin, 2005
; Holness & Bunbury, 2006; Martin et al., 2007b). Also, decompression melting can result in fracture (Holness & Siklos, 2000). However, we do not think any such factors played a significant role in determining the liquid distributions described here. The nodules contain open, glass-free, parallel-sided fractures (not shown in the figures): these clearly distinguishable features are plausible candidates for fracture caused by decompression. There are no fractures containing the host lava, suggesting that the entrainment process did not form fractures through the nodules. The grain boundary liquid films within the nodules that we describe cannot have formed during either entrainment or decompression, because they commonly have non-parallel margins (i.e. they cannot be fitted together without leaving voids, Figs 5, 9 and 10d). Both the common presence of cusps on these margins (Fig. 9) and the absence of large variations in solid–liquid interfacial orientation relative to the plane of the thin section mean that the non-parallel nature of the film in thin-section cannot be a cut effect. The thicker liquid films may also be partially filled by other phases (Figs 5e, f, and 12d, f), suggestive of a primary origin for the films. Those Kameni liquid films containing small grains of orthopyroxene nucleating on the olivine wall at the wider points of the film are possibly the result of a reaction olivine + liquid = orthopyroxene [note that plagioclase is unlikely to be participating in this reaction as orthopyroxene grains are observed in melt rims bounded by smooth, planar (and thus unreacting) plagioclase growth faces], again suggestive of a primary origin for the liquid film.
Crystallization related to entrainment and/or decompression is distinguishable from early, pre-entrainment crystallization. Fine plagioclase microlites thinner than a few microns in the glass (particularly prominent in the Kameni nodules) have the same composition as that of optically well-defined rims on the plagioclase primocrysts (Martin et al., 2007a). Similarly, post-entrainment (and perhaps ascent-related) rims on plagioclase in other nodule suites are distinguishable optically. Two clinopyroxene populations in the Icelandic nodules can be differentiated in terms of grain size, with both finer-grained aggregates and the plumose intergrowths with plagioclase likely to be late-stage features associated with entrainment and ascent. Late-stage resorption of plagioclase is also distinguishable.
Liquid in the nodules is unlikely to be derived from the host lava. The force exerted by expanding gas bubbles within the nodules inhibits melt from entering. We conclude that the pore structure and distribution described in the previous sections are representative of that present during the latest stages of solidification of slowly cooled mafic crystal mushes, and justify a direct comparison with fully solidified cumulates.
|
TEXTURES IN FULLY SOLIDIFIED CUMULATES |
|---|
The Rum peridotites are formed of accumulations of rounded to subhedral olivine grains with interstitial plagioclase and clinopyroxene. For those with a relatively low volumetric proportion of olivine, the interstitial plagioclase and clinopyroxene form oikocrysts (Fig. 13a). The Rum troctolites comprise a framework of plagioclase and cumulus olivine with interstitial clinopyroxene (Fig. 13b). The plagioclase-rich regions of the Rum troctolites have a variably developed foliation, wrapping around the olivine primocrysts.
|
The gabbros of the lower Layered Series in the Skaergaard Intrusion are dominated by a framework of plagioclase primocrysts (Fig. 14a), supplemented by primocrysts of olivine, clinopyroxene (and rarely orthopyroxene) and Fe–Ti oxides. In the lower parts of the Layered Series clino- and orthopyroxene are interstitial, with orthopyroxene rims around early, partially resorbed olivine grains (Fig. 13f). Foliation is weak or absent. In the Middle Zone of the Layered Series, olivine is not a cumulus phase although it forms polycrystalline rims separating oxide and clinopyroxene grains, and monocrystalline rims on oxide–plagioclase and clinopyroxene–plagioclase grain boundaries (Fig. 13e). Biotite is a common primary hydrous phase in the lowermost 175 m of the Layered Series [i.e. Lower Zone a as defined by McBirney (1996)
], forming large interstitial grains (Fig. 14c) and rims around oxide grains (Fig. 14d). In Lower Zone a, olivine is locally replaced by symplectites of Fe–Ti oxides and orthopyroxene (Fig. 13f). Granophyric pockets occur between plagioclase grains in the upper parts of the Layered Series (Fig. 14b).
|
|
CAN NODULES TELL US ABOUT LATE POROSITY IN CUMULATES? |
|---|
A direct comparison of nodules and fully solidified cumulates must take into account possible differences in their evolution. The nodules are likely to have been entrained from the inner margins of magma chambers or conduits, whereas the holocrystalline cumulates solidified primarily on the floors of magma chambers: such environmental differences could lead to differences in textural evolution. Well-developed foliation in holocrystalline cumulates can be attributed to either compaction of thick, undisturbed, crystal piles (Meurer & Boudreau, 1998
), shearing as a result of magma flow (Wager & Brown, 1968) or a combination of both (Higgins, 1991). Such processes, which are most likely to occur on the floor of large magma chambers, will play an important role in determining porosity (e.g. Boorman et al., 2004) and geochemical evolution (e.g. Meurer & Meurer, 2006). None of the nodules examined as part of this study display evidence for compaction or shearing (such as deformed or preferentially oriented grains). Similarly, the interstitial liquid in thick cumulate piles may convect (e.g. Tait & Jaupart, 1992), but such a process may not have taken place in the mush growing on a conduit margin.
Because we cannot directly examine natural, partially solidified, cumulates we cannot assess the extent to which the residual porosity is controlled by environmental differences. However, the commonly observed pseudomorphing of porosity by late-crystallizing phases as shown, for example, by Fig. 12b and d (and assumed by previous studies such as those of Platten, 1981; Harte et al., 1991; Pattison & Harte, 1991; Holness & Clemens, 1999; Sawyer, 1999, 2001; Rosenberg & Riller, 2000; Holness & Watt, 2001; Seyler et al., 2001; Marchildon & Brown, 2002; Holness et al., 2005a), provides an opportunity to make inferences about the late-stage liquid distribution in holocrystalline cumulates. Although the picture thus obtained can only be partial because compaction and grain coarsening (particularly in mono-mineralic regions: Hunter, 1987; Boorman et al., 2004) in large intrusions can obliterate much of the finer detail of the early sub-solidus history, we present possible analogues in cumulates for all the types of pore structure we have identified in the nodules.
|
INFERRED PORE GEOMETRY IN MAFIC CUMULATES |
|---|
An example of infilling of residual porosity surrounding primocrysts of olivine and clinopyroxene in mafic cumulates is provided by the common triangular extensions, or elongate apophyses, of these primocrysts at their junctions with plagioclase–plagioclase grain boundaries (Fig. 13c). Evidence that these extensions formed by the continued growth of the cumulus grain into late-stage pore space created by the impingement by the adjacent plagioclase grains is provided by Fig. 5a–c.
The mono-mineralic (and commonly also mono-crystalline) rims of clinopyroxene that commonly surround cumulus olivine crystals in Rum troctolites (Fig. 13b and d, with similar examples shown in Figs 13e, f, and 14d–f) have been interpreted as pseudomorphs of late-stage liquid (e.g. Holness, 2005). Continuous, grain boundary liquid films are common in coarse-grained gabbroic crystal mushes (e.g. Figs 5c–f, 9, 10d, 11b, and 12c, d). That late-crystallizing phases may grow in these liquid films is shown by Figs 6d, 10b, and 12d, f, although we have not found nodules containing the rims of uniform thickness so prominent in Rum and Skaergaard. More commonly, the rims of pyroxene surrounding olivine primocrysts in the nodules have a more irregular width than those observed in the Rum cumulates (e.g. Figs 6c, d, and 12f). However, this may simply relate to the early stage of textural development captured by the nodules: irregular rims become more regular as the texture matures to one with a lower interfacial area.
The early developmental stages of Ca-poor pyroxene rims, which grow at the partial expense of olivine primocrysts in the Skaergaard intrusion (Fig. 13f), are shown by the Kameni nodules. Orthopyroxene is attached to some olivine walls and evidently nucleated and grew there from liquid films that still lie along olivine–plagioclase grain boundaries in some Kameni nodules (Fig. 5e and f). Similarly, the coarse-grained, replacive intergrowths of oxides and orthopyroxene observed in an early stage of development at Kameni (Fig. 4e and f) are fully developed in the Skaergaard examples (Fig. 13f). Importantly, the Kameni example suggests that this replacive reaction occurs in the absence of grain boundary liquid, consistent with other interpretations of this type of symplectite as a relatively low temperature (700–800°C) feature (e.g. Barton & van Gaans, 1988; Barton et al., 1991), although the nearby presence of primary liquid in the Kameni examples of orthopyroxene–oxide symplectites suggests that temperatures as low as 500°C (suggested by Drüppel et al., 2001) are unlikely.
Closely related to the continuous mono-mineralic rims in holocrystalline cumulates are the elongate extensions of interstitial clinopyroxene along olivine–plagioclase grain boundaries (Rum troctolites, Fig. 13b and d) or similar extensions of olivine along oxide–plagioclase/clinopyroxene grain boundaries (Skaergaard, Fig. 13e). Such extensions may be a consequence of different growth habits, possibly coupled with diffusion-limited growth (e.g. Fig. 12e). Plagioclase primocrysts tend to have planar faces, whereas olivine and pyroxene commonly do not. The mutual boundary between plagioclase and either olivine or pyroxene will thus contain a deep-filled trough, which is likely to be accentuated by diffusion-limited growth of the primocrysts. Pseudomorphing of the liquid will then result in the apophyses illustrated in Fig. 13b and d. The elongate extensions are thus likely to be a growth feature, rather than indicating the stable presence of deep liquid-filled grooves indicative of a low solid–liquid dihedral angle.
Extensive mono-mineralic rims are ubiquitous in both Rum and Skaergaard cumulates. Their distribution is not affected by the extent of preferred alignment of the framework-forming plagioclase (and hence compaction). Additionally, in those rocks displaying a high degree of plagioclase alignment, there is no spatial pattern of rim distribution relative to the foliation, suggesting that compaction does not determine the distribution of thick liquid films. There does, however, appear to be a mineralogical control, with oxide grains, in particular, being generally surrounded by mono-mineralic rims. In the lower parts of the Skaergaard Layered Series, biotite (Fig. 14d and f) or plagioclase rims separate oxides from other phases, whereas in Rum cumulate oxides are surrounded by plagioclase or clinopyroxene (Fig. 14e). The rims surrounding the Rum oxides have been attributed to reaction between oxides, melt and olivine (e.g. Brown, 1956; Henderson & Suddaby, 1971; Henderson, 1975), but the variety of minerals that form rims around oxide grains in both Rum and Skaergaard suggest that these rims formed from the pseudomorphing of late-stage liquid films.
An impingement texture (as originally envisaged by Elliott et al., 1997; Holness et al., 2005a), formed of a framework of planar-sided plagioclase grains with the interstitial liquid replaced by clinopyroxene, is very common in dolerites and gabbros from rapidly cooled intrusions (e.g. Fig. 14a). A similar texture also occurs in the upper reaches of the Skaergaard Layered Series, where the interstitial material is coarse-grained granophyre (Fig. 14b). There is no sign of any textural maturation of these pockets, which retain the shape inherited from the impingement of the framework-forming plagioclase. Impingement textures in plagioclase mushes are also indicated by the distribution of late-crystallizing magmatic, hydrous phases such as biotite in the lower parts of the Skaergaard Layered Series (Fig. 14c), which infill planar-sided pores defined by the random juxtaposition of plagioclase primocrysts. Abundant interstitial biotite (and associated high apatite content) is common in the lower parts of the Skaergaard Intrusion (McBirney, 1989). Because these impingement textures occur exclusively in relatively high-porosity, medium-grained, plagioclase-rich nodules, this suggests that the nucleation of the late-stage pseudomorphing phases in the cumulates occurred before significant grain coarsening and compaction of the plagioclase framework.
Liquid pockets formed by incomplete enclosure of olivine and oxide grains by orthopyroxene oikocrysts (Fig. 8) may be a common feature of oikocrystic growth. Such a pore structure is consistent with variations in trace element concentration in amphibole oikocrysts, which have been used as evidence for the development of progressively more channelized interstitial liquid during oikocryst growth (Meurer & Claeson, 2002). Additionally, clinopyroxene oikocrysts in Rum allivalites rarely show small, intra-grain, variations in optical birefringence, resulting from minor compositional differences, permitting the distinction of different growth episodes of the pyroxene (Fig. 15a). These appear to demonstrate the progressive wrapping-round of solid inclusions as the oikocryst grows, in a similar manner to that shown by the Hawaiian examples [a similar example was also illustrated by Higgins & Roberge (2003)]. Further evidence supporting the generality of this growth history is provided by the common presence of plagioclase or biotite surrounding inclusions of Fe–Ti oxide in pyroxenes of the Skaergaard intrusion (Fig. 15b).
|
These examples demonstrate the close textural relationship between the liquid pockets shown in Fig. 8 and the formation of liquid inclusions containing what may appear to be daughter minerals. Progressive inclusion of grains of a second mineral may promote trapping of liquid (Fig. 8c). Progressive growth of an oikocryst to surround inclusions of a second mineral (Fig. 15a), rather than the preservation (as pseudomorphs) of liquid pockets (e.g. Fig. 15b), plausibly is a consequence of the connectedness of the liquid pockets with a larger pool of liquid in three dimensions.
Plagioclase-dominated horizons of mafic layered intrusions commonly are texturally equilibrated, with smooth (i.e. constant mean curvature, Thomson, 1887; Bulau et al., 1979) plagioclase–plagioclase grain boundaries (Hunter, 1987), in contrast to the highly porous and liquid-rich plagioclase–plagioclase grain boundaries observed in the nodules. The difference in extent of textural equilibration may be accounted for by the different evolution of cumulates and nodules. There is no evidence for compaction in the coarse-grained nodules, whereas the texturally equilibrated and inclusion-free grain boundaries in many mafic cumulates are probably the result of extensive recrystallization associated with compaction. However, preliminary investigations of coarse-grained anorthosites, in which the plagioclase grains are randomly oriented (i.e. plausible analogues for the non-compacted coarse-grained nodules), reveal clearly identifiable impingement lenses, now filled with quartz, which occur at steps and jogs in the plagioclase–plagioclase grain boundaries, together with elongate quartz grains on the straighter parts of plagioclase grain boundaries (Fig. 15c and d). It should be noted that the position of the pseudomorphed impingement lenses at places where the host grain boundary changes curvature could be interpreted as a consequence of grain boundary migration with localized pinning by inclusions. However, we suggest that these features are primary and do not relate to subsolidus textural adjustment.
|
DISCUSSION AND CONCLUSIONS |
|---|
A major conclusion of this study is that pore geometry and permeability of solidifying mafic plutonic rocks are controlled by crystal growth. Very few of the liquid-filled pores in the nodules described here can be ascribed to the minimization of internal energies during textural equilibration: the great majority are almost certainly out of equilibrium, with shapes controlled by the kinetics of crystal growth. An example is provided by Fig. 8b. The glass-filled lenses are on the same olivine–pyroxene grain boundary, with the same lattice orientations of the two solid phases. If they were an equilibrium feature they would have the same angle subtended at their margins. They do not, demonstrating that their shape was controlled by growth processes alone.
The rare texturally equilibrated pore geometries include some grain boundary lenses and triple junction pores. In particular, some impingement lenses on plagioclase–plagioclase grain boundaries may have liquid–plagioclase–plagioclase dihedral angles approaching the equilibrium value (e.g. Fig. 12b). This may reflect a significant reduction in growth rate of the bounding plagioclase grains (and consequent occlusion of the pores) as a result of the chemical isolation of the liquid pocket and depletion in plagioclase components in the melt, thus permitting shape adjustment under the driving force of textural equilibration.
Our observations confirm the common presence of liquid on grain boundaries, either as pockets or as extensive films of varying thickness. Such films may provide an explanation for the anomalous compositions of olivine in the immediate vicinity of grain boundaries in dunite nodules erupted from Hualalai volcano, Hawaii (Waff & Holdren, 1981). Late-stage grain boundary liquid films have previously been invoked to explain the reverse zoning on the extreme margins of plagioclase grains in Kiglapait (Morse & Nolan, 1984). The pervasive nature of hydrous partial melting in the lower oceanic crust (Koepke et al., 2005a, 2005b) may also be promoted by the presence of grain boundary liquid films, which provide rapid diffusion pathways. Compositional variations on plagioclase–plagioclase grain boundaries as a function of proximity to cumulus oxide grains (Butcher & Merkle, 1991) are consistent with our inference that inter-phase boundaries involving an oxide phase commonly contain liquid.
We infer, based on textural similarities with glass-bearing nodules, the presence of abundant grain boundary liquid during the final stages of crystallization of large mafic intrusions. The textural similarities support late-stage crystallization, metasomatism and reaction on grain boundaries between primocrystic phases (e.g. Morse & Nolan, 1984; Batiza & Vanko, 1985; Koepke et al., 2005a, 2005b). Importantly, the late-stage grain boundary liquid films result in a higher permeability of the almost completely solidified crystal mush than would be predicted assuming an equilibrated pore structure (see Cheadle et al., 2004). A highly interconnected liquid phase at vanishingly low liquid fractions in gabbroic cumulates has been suggested by Meurer & Meurer (2006) on the basis of apatite compositions, and the grain boundary liquid films described here show how such high permeabilities might have been achieved. The liquid films and related grain boundary textures of glass-bearing nodules and holocrystalline counterparts provide a basis for quantitative tests of relations between crystallization, liquid fraction, mineral composition, permeability and porosity of magma near its solidus.
|
ACKNOWLEDGEMENTS |
|---|
We are indebted to Ilya Veksler and Troels Nielsen for stimulating discussions that led to this study. Troels Nielsen and Christian Tegner lent samples from the Skaergaard intrusion. E.P. is supported by an NERC studentship, and V.M. was supported by an NERC studentship while collecting the Kameni nodules. J.M. thanks NERC for a postdoctoral fellowship. Helpful and constructive reviews from Michael Higgins, Dougal Jerram and Bill Meurer improved an earlier version of the manuscript.
|
FOOTNOTES |
|---|
Present address: 38 Spencer Place, Lewisburg, PA 17837, USA. 
*Corresponding author. E-mail: marian{at}esc.cam.ac.uk
|
REFERENCES |
|---|
Andersen JCØ, Rasmussen H, Nielsen TFD, Ronsbø JG. (1998) The Triple Group and the Platinova gold and palladium reefs in the Skaergaard Intrusion: stratigraphical and petrographic relations. Economic Geology 93:488–509.
Barton M and van Gaans C. (1988) Formation of orthpyroxene–Fe–Ti oxide symplectites in Precambrian intrusives, Rogaland, southwestern Norway. American Mineralogist 73:1046–1059.[Abstract]
Barton M, Sheets JM, Lee WE, van Gaans C. (1991) Occurrence of low-Ca clinopyroxene and the role of deformation in the formation of pyroxene–Fe–Ti oxide symplectites. Contributions to Mineralogy and Petrology 108:181–195.[CrossRef][Web of Science]
Batiza R and Vanko DA. (1985) Petrologic evolution of large failed rifts in the Eastern Pacific: petrology of volcanic and plutonic rocks from the Mathematician Ridge area and the Guadalupe Trough. Journal of Petrology 26:564–602.
Becker HJ. (1977) Pyroxenites and hornblendites from the Maar-type volcanoes of the Westeifel, Federal Republic of Germany. Contributions to Mineralogy and Petrology 65:45–52.[CrossRef][Web of Science]
Bergantz GW. (1995) Changing techniques and paradigms for the evaluation of magmatic processes. Journal of Geophysical Research 100:17603–17613.[CrossRef]
Boorman S, Boudreau A, Kruger FJ. (2004) The Lower Zone–Critical Zone transition of the Bushveld Complex: a quantitative textural study. Journal of Petrology 45:1209–1235.
Boudreau AE and McCallum IS. (1992) Infiltration metasomatism in layered intrusions—an examples from the Stillwater Complex, Montana. Journal of Volcanology and Geothermal Research 52:171–183.[CrossRef][Web of Science]
Boudreau AE and Meurer WP. (1999) Chromatographic separation of the platinum-group elements, gold, base metals and sulfur during degassing of a compacting and solidifying igneous crystal pile. Contributions to Mineralogy and Petrology 134:174–185.[CrossRef][Web of Science]
Brooks CK and Nielsen TFD. (1982) The E. Greenland continental margin: a transition between oceanic and continental magmatism. Journal of the Geological Society, London 139:265–275.
Brown GM. (1956) The layered ultrabasic rocks of Rhum, Inner Hebrides. Philosophical Transactions of the Royal Society of London, Series B 240:1–53.
Bryon DN, Atherton MP, Hunter RH. (1995) The interpretation of granitic textures from serial thin-sectioning, image analysis and three-dimensional reconstruction. Mineralogical Magazine 59:203–211.[Abstract]
Bryon DN, Atherton MP, Cheadle MJ, Hunter RH. (1996) Melt movement and the occlusion of porosity in crystallising granitic systems. Mineralogical Magazine 60:163–171.[Abstract]
Bulau JR, Waff HS, Tyburczy JA. (1979) Mechanical and thermodynamic constraints on fluid distribution in partial melts. Journal of Geophysical Research 84:6102–6108.
Butcher AR and Merkle RKW. (1991) Unusual textures and structures associated with a magnetitite layer in the Bushveld Complex: a contribution to the adcumulus debate. Mineralogical Magazine 55:465–477.[CrossRef][Web of Science]
Cheadle MJ, Elliott MT, McKenzie D. (2004) Percolation threshold and permeability of crystallising igneous rocks: the importance of textural equilibrium. Geology 32:757–760.
Clague DA and Dalrymple GB. (1987) In Decker RW, Wright TL, Stauffer PH (Eds.). The Hawaiian–Emperor volcanic chain. Part I. Geological evolution. Volcanism in Hawaii. US Geological Survey, Professional Papers 1350:5–54.
Deer WA. (1976) Tertiary igneous rocks between Scoresby Sund and Kap Gustav Holm, East Greenland. In Esher A and Watt WS (Eds.). Geology of Greenland(Geological Survey of Greenland, Copenhagen) pp. 405–429.
De Silva SL. (1989) The origin and significance of crystal rich inclusions in pumices from two Chilean ignimbrites. Geological Magazine 126:159–175.[Abstract]
Druitt TH, Edwards L, Mellors RM, Pyle DM, Sparks RSJ, Lanphere M, Davies M, Barreiro B. (1999) Santorini Volcano. Geological Society of London, Memoirs. 19:.
Drüppel K, von Seckendorff V, Okrusch M. (2001) Subsolidus reaction textures in the anorthositic rocks of the southern part of the Kunene Intrusive Complex, NW Namibia. European Journal of Mineralogy 13:289–309.
Easton RM. (1987) In Decker RW, Wright TL, Stauffer PH (Eds.). Stratigraphy of Kilauea Volcano. Volcanism in Hawaii. US Geological Survey, Professional Papers 1350:243–260.
Elliott MT, Cheadle MJ, Jerram DA. (1997) On the identification of textural equilibrium in rocks using dihedral angle measurements. Geology 25:355–358.
Emeleus CH. (1997) Geology of Rum and the adjacent islands. Memoir of the British Geological Survey 170 Sheet 60 (Scotland).
Emeleus CH, Cheadle MJ, Hunter RH, Upton BGJ, Wadsworth WJ. (1996) The Rum Layered Suite. In Cawthorn RG (Ed.). Layered Intrusions(Elsevier, Amsterdam) pp. 403–439.
Faul UH, Toomey DR, Waff HS. (1994) Intergranular basaltic melt is distributed in thin, elongated inclusions. Geophysical Research Letters 21:29–32.[CrossRef][Web of Science]
Fiske RS and Jackson ED. (1972) Orientation and growth of Hawaiian volcanic rifts: the effect of regional structure and gravitational stresses. Proceedings of the Royal Society of London, Series A 329:299–326.
Hamilton MA, Pearson DG, Thompson RN, Kelley SP, Emeleus CH. (1998) Rapid eruption of Skye lavas inferred from precise U–Pb and Ar–Ar dating of the Rum and Cuillin plutonic complexes. Nature 394:260–263.[CrossRef]
Hammer CU, Clausen HB, Frierich WL, Tauber H. (1987) The Minoan eruption of Santorini in Greece dated to 1645 BC? Nature 328:517–517.[CrossRef]
Hansen H and Grönvold K. (2000) Plagioclase ultraphyric basalts in Iceland: the mush of the rift. Journal of Volcanology and Geothermal Research 9:81–32.
Harte B, Pattison DRM, Linklater CM. (1991) Field relations and petrography of partially melted pelitic and semi-pelitic rocks. In Voll G, Töpel J, Pattison DRM, Seifert F (Eds.). Equilibrium and Kinetics in Contact Metamorphism: the Ballachulish Igneous Complex and its Aureole(Springer, Heidelberg) pp. 181–210.
Henderson P. (1975) Reaction trends shown by chrome-spinels of the Rhum Layered Intrusion. Geochimica et Cosmochimica Acta 39:1035–1044.[CrossRef][Web of Science]
Henderson P and Suddaby P. (1971) The nature and origin of the chrome-spinel of the Rhum Layered Intrusion. Contributions to Mineralogy and Petrology 33:21–31.[CrossRef][Web of Science]
Hermes OD and Cornell WC. (1981) Quenched crystal mush and associated magma compositions as indicated by intercumulus glasses from Mt. Vesuvius, Italy. Journal of Volcanology and Geothermal Research 9:133–149.[CrossRef][Web of Science]
Higgins MD. (1991) The origin of laminated and massive anorthosite, Sept Iles intrusion, Quebec, Canada. Contributions to Mineralogy and Petrology 106:340–354.[CrossRef][Web of Science]
Higgins MD. (1996) Magma dynamics beneath Kameni Volcano, Thera, Greece, as revealed by crystal size and shape measurements. Journal of Volcanology and Geothermal Research 70:37–48.[CrossRef][Web of Science]
Higgins MD and Roberge J. (2003) Crystal size distributions (CSD) of plagioclase and amphibole from Soufrière Hills volcano, Montserrat: evidence for dynamic crystallization/textural coarsening cycles. Journal of Petrology 44:1401–1411.
Hjartarson Á. (1988) Þjórsárhraunið mikla-stærsta nútímahraun jarðar. Náttúrufrædingurinn 58:1–16 (in Icelandic).
Holness MB. (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 MB. (2005) Spatial constraints on magma chamber replenishment events from textural observations of cumulates: the Rum Layered Intrusion, Scotland. Journal of Petrology 46:1585–1601.
Holness MB and Bunbury JM. (2006) Insights into continental rift-related magma chambers: igneous nodules from the Kula Volcanic Province, Western Turkey. Journal of Volcanology and Geothermal Research 153:241–261.[CrossRef][Web of Science]
Holness MB and Clemens JD. (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 MB and Siklos STC. (2000) Rates of textural equilibration in fluid-bearing systems: kinetic limitations to surface-energy controlled permeability. Chemical Geology 162:137–153.[CrossRef][Web of Science]
Holness MB and Watt GR. (2001) Quartz recrystallisation and fluid flow during contact metamorphism: a cathodoluminescence study. Geofluids 1:215–228.[CrossRef]
Holness MB, Cheadle MJ, McKenzie D. (2005a) On the use of changes in dihedral angle to decode late-stage textural evolution in cumulates. Journal of Petrology 46:1565–1583.
Holness MB, Martin VM, Pyle DM. (2005b) Information about open-system magma chambers derived from textures in magmatic enclaves: the Kameni Islands, Santorini, Greece. Geological Magazine 142:637–649.
Hunter RH. (1987) Textural equilibrium in layered igneous rocks. In Parsons I (Ed.). Origins of Igneous Layering(D. Reidel., Dordrecht) pp. 473–503.
Jakobsen JK, Veksler IV, Tegner C, Brooks CK. (2005) Immiscible iron- and silica-rich melts in basalt petrogenesis documented in the Skaergaard intrusion. Geology 33:885–888.
Koepke J, Fieg ST, Snow J. (2005a) Hydrous partial melting within the lower oceanic crust. Terra Nova 17:286–291.[CrossRef][Web of Science]
Koepke J, Feig ST, Snow J. (2005b) Late-stage magmatic evolution of oceanic gabbros as a result of hydrous partial melting: evidence from the ODP Leg 153 drilling at the Mid-Atlantic Ridge. Geochemistry, Geophysics, Geosystems 6: 2004GC000805.
Langenheim VAM and Clague DA. (1987) In Decker RW, Wright TL, Stauffer PH (Eds.). The Hawaiian–Emperor volcanic chain. Part II. Stratigraphic framework of volcanic rocks of the Hawaiian Islands. Volcanism in Hawaii. US Geological Survey, Professional Papers 1350:55–84.
Laporte D and Watson EB. (1995) Experimental and theoretical constraints on melt distribution in crustal sources: the effect of crystalline anisotropy on melt interconnectivity. Chemical Geology 124:161–184.[CrossRef][Web of Science]
Lockwood JP and Lipman PW. (1987) In Decker RW, Wright TL, Stauffer PH (Eds.). Holocene eruptive history of Mauna Loa Volcano. Volcanism in Hawaii. US Geological Survey, Professional Papers 1350:509–536.
Lupulescu A and Watson EB. (1999) Low melt fraction connectivity of granitic and tonalitic melts in a mafic crustal rock at 800°C and 1 GPa. Contributions to Mineralogy and Petrology 134:202–216.[CrossRef][Web of Science]
MacDonald GA. (1949) Hawaiian petrographic province. Geological Society of America Bulletin 60:1541–1596.
Macdonald GA, Abbott AT, Peterson FL. (1983) Volcanoes in the Sea: the Geology of Hawaii 2nd edn (University of Hawaii Press, Honolulu) pp. 517.
Marchildon N and Brown M. (2002) Grain-scale melt distribution in two contact aureole rocks: implications for controls on melt localisation and deformation. Journal of Metamorphic Geology 20:381–396.[CrossRef][Web of Science]
Martin VM. (2005) Geochemical and textural analysis of mafic nodules from Nea Kameni, Santorini, Greece. Ph.D. thesis University of Cambridge.
Martin VM, Holness MB, Pyle DM. (2007a) Textural analysis of magmatic enclaves from the Kameni Islands, Santorini, Greece. Journal of Volcanology and Geothermal Research 154:89–102.[CrossRef][Web of Science]
Martin VM, Pyle DM, Holness MB. (2007b) The role of crystal frameworks in the preservation of enclaves during magma mixing. Earth and Planetary Science Letters 248:787–799.[CrossRef][Web of Science]
Mathez EA. (1995) Magmatic metasomatism and formation of the Merensky reef, Bushveld Complex. Contributions to Mineralogy and Petrology 119:277–286.[Web of Science]
Mattioli M, Serri G, Salvioli-Mariani E, Renzulli A, Holm PM, Santi P, Venturelli G. (2003) Sub-volcanic infiltration and syn-eruptive quenching of liquids in cumulate wall-rocks: the example of the gabbroic nodules of Stromboli (Aeolian Islands, Italy). Mineralogy and Petrology 78:201–230.[CrossRef][Web of Science]
McBirney AR. (1989) The Skaergaard Layered Series: I. Structure and average compositions. Journal of Petrology 30:363–399.
McBirney AR. (1996) The Skaergaard Intrusion. In Cawthorn RG (Ed.). Layered Intrusions(Elsevier, Amsterdam) pp. 147–179.
McBirney AR and Sonnenthal EL. (1990) Metasomatic replacement in the Skaergaard Intrusion, East Greenland: preliminary observations. Chemical Geology 88:245–260.[CrossRef][Web of Science]
McKenzie D. (1984) The generation and compaction of partially molten rock. Journal of Petrology 25:713–765.
Meurer WP and Boudreau AE. (1998) Compaction of igneous cumulates Part II: compaction and the development of igneous foliations. Journal of Geology 106:293–304.[Web of Science]
Meurer WP and Claeson DT. (2002) Evolution of crystallizing interstitial liquid in an arc-related cumulate determined by LA ICP-MS mapping of a large amphibole oikocryst. Journal of Petrology 43:607–629.
Meurer WP and Meurer MES. (2006) Using apatite to dispel the ‘trapped liquid’ concept and to understand the loss of interstitial liquid by compaction in mafic cumulates: an example from the Stillwater Complex, Montana. Contributions to Mineralogy and Petrology 151:187–201.[CrossRef][Web of Science]
Minarik WG and Watson EB. (1995) Interconnectivity of carbonate melt at low melt fraction. Earth and Planetary Science Letters 133:423–437.[CrossRef][Web of Science]
Morse SA. (1986) Convection in aid of adcumulus growth. Journal of Petrology 27:1183–1214.
Morse SA and Nolan KM. (1984) Origin of strongly reversed rims on plagioclase in cumulates. Earth and Planetary Science Letters 68:485–498.[CrossRef][Web of Science]
Nielsen TFD. (2004) The shape and volume of the Skaergaard Intrusion, East Greenland: implications for mass balance and bulk composition. Journal of Petrology 45:507–530.
Pattison D. RM and Harte B. (1991) Petrography and mineral chemistry of pelites. In Voll G, Topel J, Pattison D. RM, Seifert F (Eds.). Equilibrium and Kinetics in Contact Metamorphism: the Ballachulish Igneous Complex and its Aureole(Springer, Heidelberg) pp. 135–180.
Platten IM. (1981) Partial melting of feldspathic quartzite around late Caledonian minor intrusions in Appin, Scotland. Geological Magazine 119:413–419.[Web of Science]
Porter SC. (1979a) Quaternary stratigraphy and chronology of Mauna Kea, Hawaii: a 380,000 year record of mid-Pacific volcanism and ice-cap glaciation. Geological Society of America Bulletin 90:980–1093.
Porter SC. (1979b) Geological map of Mauna Kea volcano, Hawaii. Geological Society of America Map and Chart Series MC-30 scale 1:50 000.
Renner R and Palacz Z. (1987) Basaltic replenishment of the Rhum magma chamber: evidence from unit 14. Journal of the Geological Society, London 144:961–970.
Rosenberg CL and Riller U. (2000) Partial melt topology in statically and dynamically recrystallised granite. Geology 28:7–10.
Sawyer EW. (1999) Criteria for the recognition of partial melting. Physics and Chemistry of the Earth 24:269–279.[CrossRef][Web of Science]
Sawyer EW. (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]
Seyler M, Toplis MJ, Lorand J-P, Luguet A, Cannat M. (2001) Clinopyroxene microtextures reveal incompletely extracted melts in abyssal peridotites. Geology 29:155–158.
Shirley D. (1987) Differentiation and compaction in the Palisades Sill, New Jersey. Journal of Petrology 28:835–865.
Sonnenthal EL. (1992) Geochemistry of dendritic anorthosites and associated pegmatites in the Skaergaard Intrusion, East Greenland: evidence for metasomatism by a chlorine-rich fluid. Journal of Volcanology and Geothermal Research 52:209–230.[CrossRef][Web of Science]
Tait SR. (1988) Samples from the crystallising boundary layer of a zoned magma chamber. Contributions to Mineralogy and Petrology 100:470–483.[CrossRef][Web of Science]
Tait S and Jaupart C. (1992) Compositional convection in a reactive crystalline mush and melt differentiation. Journal of Geophysical Research 97:6735–6756.
Tait S and Jaupart C. (1996) The production of chemically stratified and adcumulate plutonic igneous rocks. Mineralogical Magazine 60:99–114.[Abstract]
Tait SR, Wörner G, Van Den Bogaard P, Schminke H-U. (1989) Cumulate nodules as evidence for convective fractionation in a phonolite magma chamber. Journal of Volcanology and Geothermal Research 37:21–37.[CrossRef][Web of Science]
Tegner C, Duncan RA, Bernstein S, Brooks CK, Bird DK, Storey M. (1998) Ar40–Ar39 geochronology of Tertiary mafic intrusions along the East Greenland rifted margin: relation to flood basalts and the Iceland hotspot track. Earth and Planetary Science Letters 156:75–880.[CrossRef][Web of Science]
Tepley FJ III and Davidson JP. (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]
Thomson W. (1887) On the division of space with minimum partitional area. Philosophical Magazine 24:503–514.
Thordarson T, Self S, Miller DJ, Larsen G, Vilmundardóttir EG. (2003) In Oppenheimer C, Pyle DM, Barclay J (Eds.). Sulphur release from flood lava eruptions in the Veiðivötn, Grimsvötn and Katla volcanic systems, Iceland. Volcanic Degassing. Geological Society, London, Special Publications 213:103–121.[CrossRef]
Turbeville BN. (1992) Relationships between chamber margin accumulates and pore liquids: evidence from arrested in situ processes in ejecta, Latera caldera, Italy. Contributions to Mineralogy and Petrology 110:429–441.[CrossRef][Web of Science]
Turbeville BN. (1993) Sidewall differentiation in an alkalic magma chamber: evidence from syenite xenoliths in tuffs of the Latera caldera, Italy. Geological Magazine 130:453–470.[Abstract]
Upton BGJ, Skovgaard AC, McClurg J, Kirstein L, Cheadle M, Emeleus CH, Wadsworth WJ, Fallick AE. (2002) Picritic magmas and the Rum ultramafic complex, Scotland. Geological Magazine 139:437–452.
Vilmundardóttir EG, Snorrason SP, Larsen G, Gudmundsson A. (1988) Geological Map, Sigalda–Veiðivötn 3340 B, 1:50 000. (National Energy Authority, Hydro Power Division and National Power Company, Reykjavik).
Vilmundardóttir EG, Gudmundsson A, Snorrason SP, Larsen G. (1990) Geological map, Botnafjöll, 1913 IV, 1:50 000. (Iceland Geodetic Survey, National Energy Authority and National Power Company, Reykjavik).
Volker JA and Upton BGJ. (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]
Von Bargen N and Waff HS. (1986) Permeabilities, interfacial areas and curvatures of partially molten systems: results of numerical computations of equilibrium microstructures. Journal of Geophysical Research 91:9261–9276.
Waff HS and Holdren GR Jr. (1981) The nature of grain boundaries in dunite and lherzolite xenoliths: implications for magma transport in refractory upper mantle material. Journal of Geophysical Research 86:3677–3683.
Wager LR and Brown GM. (1968) Layered Igneous Rocks(Oliver & Boyd, Edinburgh).
Wark DA, Williams CA, Watson EB, Price JD. (2003) Reassessment of pore shapes in microstructurally equilibrated rocks, with implications for permeability of the upper mantle. Journal of Geophysical Research 108: B1, 2050, doi:10.1029/2001JB001575.
Williams E, Boudreau AE, Boorman S, Kruger FJ. (2006) Textures of orthopyroxenites from the Burgersfort Bulge of the eastern Bushveld Complex, Republic of South Africa. Contributions to Mineralogy and Petrology 151:480–492.[CrossRef][Web of Science]
Worrell LM, Cheadle MJ, Coogan LA, Prior DJ, Wheeler J, Toplis MJ. (2003) A multidisciplinary approach to understanding the origin of peridotite cumulates. EOS Transactions, American Geophysical Union 84:46 Fall Meeting Supplement V11F-02.
Wright TL. (1971) Chemistry of Kilauea and Mauna Loa in space and time. US Geological Survey, Professional Papers 735:39.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. Schiavi, N. Walte, and H. Keppler First in situ observation of crystallization processes in a basaltic-andesitic melt with the moissanite cell Geology, November 1, 2009; 37(11): 963 - 966. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Tegner, P. Thy, M. B. Holness, J. K. Jakobsen, and C. E. Lesher Differentiation and Compaction in the Skaergaard Intrusion J. Petrology, May 1, 2009; 50(5): 813 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. S. Humphreys Chemical Evolution of Intercumulus Liquid, as Recorded in Plagioclase Overgrowth Rims from the Skaergaard Intrusion J. Petrology, January 7, 2009; (2009) egn076v1. [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]
|
|
|
|
|
|
J. Maclennan Concurrent Mixing and Cooling of Melts under Iceland J. Petrology, November 1, 2008; 49(11): 1931 - 1953. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Holness and E. W. Sawyer On the Pseudomorphing of Melt-filled Pores During the Crystallization of Migmatites J. Petrology, July 1, 2008; 49(7): 1343 - 1363. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||