C. Henry Emeleus

Age and origin of the major syenite centers in the Gardar province of south Greenland: Rb-Sr studies

Rb-Sr whole-rock isochron data for nine intrusive centers (including the major oversaturated and undersaturated syenite bodies) in the Precambrian Gardar alkalic igneous province of south Greenland complete a systematic isotopic study of the major complexes of the region (15 distinct igneous centers). The new data show that magmatic activity occurred throughout the region over a period in excess of 150 m.y. and that there is a clear distinction between two major magmatic events — an early Gardar (∼1,300 m.y. old) event and a more widespread late Gardar (∼1,160 m.y. old) event, which also involved extensive east-northeast-trending dike emplacement. Two relatively small intrusions in the western part of the province have middle Gardar (∼1,250 m.y. old) ages. There does not appear to be any relationship between emplacement age and geographic location, mode of emplacement, or the recognized petrologic division into saturated or undersaturated types; indeed, some adjacent complexes with strikingly similar petrologic characteristics have originated during different events, and, conversely, adjacent complexes with very different compositions are chronologically indistinguishable with present techniques. Initial 87 Sr/ 86 Sr ratios for the Gardar rocks are broadly divisible into three groups. Low (0.702 to 0.704) ratios are characteristic of most of the intrusions in the province, including most of the syenites (both undersaturated and oversaturated) and gabbros —which indicates a primitive (mantle) origin for the parental Gardar magmas. Intermediate ratios (0.704 to 0.707) are observed in fewer centers, particularly in the Nunarssuit complex (gabbro, oversaturated syenites, and granites). For these centers, the possibility of closed-system enrichment in 87 Sr (by decay of 87 Rb during a complex molten history extending throughout 150 m.y. of Gardar time) is rejected as unlikely, as is a direct origin by partial melting of older crust; field, chemical, and isotopic evidence presented here strongly support a bulk-assimilation hypothesis. Only two complexes have high (>0.707)ratios; both show evidence of introduction of radiogenic 87 Sr into their magmas at a late stage of development through selective leaching of radiogenic 87 Sr from older crust, and both of these complexes are associated with mineralization.

Trace element and isotope constraints on crustal anatexis by upwelling mantle melts in the North Atlantic Igneous Province: an example from the Isle of Rum, NW Scotland

Sr and Nd isotope ratios, together with lithophile trace elements, have been measured in a representative set of igneous rocks and Lewisian gneisses from the Isle of Rum in order to unravel the petrogenesis of the felsic rocks that erupted in the early stages of Palaeogene magmatism in the North Atlantic Igneous Province (NAIP). The Rum rhyodacites appear to be the products of large amounts of melting of Lewisian amphibolite gneiss. The Sr and Nd isotopic composition of the magmas can be explained without invoking an additional granulitic crustal component. Concentrations of the trace element Cs in the rhyodacites strongly suggests that the gneiss parent rock had experienced Cs and Rb loss prior to Palaeogene times, possibly during a Caledonian event. This depletion caused heterogeneity with respect to 87Sr/86Sr in the crustal source of silicic melts. Other igneous rock types on Rum (dacites, early gabbros) are mixtures of crustal melts and and primary mantle melts. Forward Rare Earth Element modelling shows that late stage picritic melts on Rum are close analogues for the parent melts of the Rum Layered Suite, and for the mantle melts that caused crustal anatexis of the Lewisian gneiss. These primary mantle melts have close affinities to Mid-Oceanic Ridge Basalts (MORB), whose trace element content varies from slightly depleted to slightly enriched. Crustal anatexis is a common process in the rift-to-drift evolution during continental break-up and the formation of Volcanic Rifted Margins systems. The ‘early felsic–later mafic’ volcanic rock associations from Rum are compared to similar associations recovered from the now-drowned seaward-dipping wedges on the shelf of SE Greenland and on the Voring Plateau (Norwegian Sea). These three regions show geochemical differences that result from variations in the regional crustal composition and the depth at which crustal anatexis took place.

Sr and Nd isotope evidence for successive crustal contamination of Slieve Gullion ring-dyke magmas, Co. Armagh, Ireland

The Palaeogene Slieve Gullion Igneous Centre in southern Armagh, Northern Ireland, consists of a layered central intrusive complex surrounded by a prominent and slightly older ring-dyke that intrudes both Lower Palaeozoic sedimentary rocks and the Caledonian Newry Granodiorite pluton (452 Ma). The ring-dyke comprises two major rock types: porphyritic felsite and porphyritic granophyre. We analysed both ring-dyke lithologies, both types of country rock, and a local Palaeogene basalt dyke sample for Sr and Nd isotopes. Trace element and whole rock data for this suite suggest that there are two distinct groups of both felsite and granophyre: one Si-rich and one Si-poor, most likely representing two magmas from a zoned chamber and their mushy chamber wall equivalents (McDonnell et al. 2004). Isotope data show the low-Si rocks to be higher in radiogenic Sr than the high-Si rocks, which is inconsistent with a simple AFC-scenario of increasing sediment assimilation with higher degrees of differentiation. However, using MORB-type basalt as a starting composition, the low-Si ring-dyke rocks can be modelled through AFC with Lower Palaeozoic sedimentary rock as the contaminant. The decreasing 87 Sr/ 86 Sr trend from low-Si to high-Si dyke rocks, in turn, represents a second stage of contamination. Selective assimilation of the most fusible portions of Newry Granodiorite, which is lower in radiogenic Sr than the local sedimentary rocks, appears to be the most plausible solution. The Sr and Nd data are consistent with (a) at least a two-stage contamination history during upper crustal residence and storage, whereby fractionating magmas of basaltic and intermediate composition are contaminated by local sedimentary rocks, giving rise to rhyolite magmas that experience additional shallow contamination by Newry Granodiorite, and (b) a zoned rhyolite magma chamber where high-Si magma is stored in the upper part of the chamber where crystallization and crustal contamination are most extensive.

Early mafic magmatism and crustal anatexis on the Isle of Rum: evidence from the Am Màm intrusion breccia

The Rum Igneous Centre comprises two early marginal felsic complexes (the Northern Marginal Zone and the Southern Mountains Zone), along with the later central ultrabasic–basic layered intrusions. These marginal complexes represent the remnants of near-surface to eruptive felsic magmatism associated with caldera collapse, examples of which are rare in the North Atlantic Igneous Province. Rock units include intra-caldera collapse breccias, rhyolitic ignimbrite deposits and shallow-level felsic intrusions, as well the enigmatic ‘Am Mam intrusion breccia’. The latter comprises a dacitic matrix enclosing lobate basaltic inclusions (~1–15 cm) and a variety of clasts, ranging from millimetres to tens of metres in diameter. These clasts comprise Lewisian gneiss, Torridonian sandstone and coarse gabbro. Detailed re-mapping of the Am Mam intrusion breccia has shown its timing of emplacement as syn-caldera, rather than pre-caldera as previously thought. Textural analysis of entrained clasts and adjacent, uplifted country rocks has revealed their thermal metamorphism by early mafic intrusions at greater depth than their present structural position. These findings provide a window into the evolution of the early mafic magmas responsible for driving felsic magmatism on Rum. Our data help constrain some of the physical parameters of this early magma–crust interaction and place it within the geochemical evolution of the Rum Centre.

Petrographic, mineralogic, and x-ray fluorescence analysis of lunar igneous-type rocks and spherules.

Three lunar rocks show almost identical mineralogy but grain sizes that vary from basaltic to gabbroic. Clinopyroxene is zoned from augite to subcalcic ferroaugite compositions and is accompanied by decrease in Cr, Al, and Ti. Plagioclase is zoned from 93 to 78 percent anorthite. Olivine (68 percent forsterite) is present in one rock and apatite is rare. Cristobalite, ilmenite with Ti-rich lamellae, ulv�spinel (often Cr-rich), troilite, and kamacite are low in trace elements. Glassy spherules are of basaltic or feldspathic (92 percent anorthite) composition but contain abundant iron spheres of taenite composition (13 percent Ni). Four rock analyses by x-ray fluorescence show affinity with terrestrial basalts but with anomalous amounts of Ti, Na, Cr, Zr, Y, Rb, Nb, Ni, Cu, and Zn.

Lateral versus vertical emplacement in shallow-level intrusions? The Slieve Gullion Ring-complex revisited

Recent studies on shallow-level arcuate intrusions have identified numerous examples of horizontal mineral fabrics. These are commonly interpreted as reflecting considerable lateral flow during magma emplacement, thus querying established ‘semi-vertical’ ring-dyke models. We question the recent lateral emplacement model proposed for the Palaeocene Slieve Gullion Ring-complex, NE Ireland, where the absence of steep fabrics in parts of the ring-complex has been used to support a shallow, semi-horizontal sheet intrusion mechanism. We argue that such simple flow models cannot be applied to explosive ring-fissure eruptions and that fabric data alone do not warrant rejection of the ring-dyke model. Moreover, the apparent ‘absence of steep intrusive contacts’ along the intrusion’s perimeter is readdressed and we present numerous examples of outcrops (27) with steep-sided geometries. The Camlough Breccias are reinterpreted as the product of gas-driven tuffisites injected along the active ring-fault (rather than of purely tectonic origin). Crucially, the porphyritic microgranite and porphyritic rhyolite ring-dyke rocks exhibit geochemical and petrographic signatures of contamination by the geographically restricted Palaeozoic Newry granodiorite and are best explained through crustal interaction vertically beneath the ring-complex. Subsequently, these silicic magmas rose into ignimbrite feeders along a caldera ring-fault system that was emplaced into near-surface vent-filling breccias.

Incremental Construction of the Unit 10 Peridotite, Rum Eastern Layered Intrusion, NW Scotland

The Rum Eastern Layered Intrusion (ELI) is the product of part of a ~60 Ma open-system magma chamber. The 16 coupled peridotite/troctolite ± gabbro macro-rhythmic units it contains represent crystallisation of multiple batches of basaltic and picritic magma. Within the ELI, Unit 10 has been considered the type example of batch fractionation of magma on Rum for more than 50 years, successively producing peridotite, troctolite and olivine gabbro. Detailed field observations and logs of the Unit 10 peridotite cumulate are presented here, together with mineralogical and textural analyses of Cr-spinel seams and their peridotite host rocks. Numerous harrisite layers are commonly associated with diffuse, laterally discontinuous platinum-group element (PGE) enriched Cr-spinel seams. Multiple millimetre–centimetre thick Cr-spinel seams occur at the bases, tops and within harrisite layers. These relationships are inconsistent with simple batch fractionation of magma. Critically, the harrisite layers also exhibit centimetre to metre-scale, upward oriented apophyses that point to injection of magma into the overlying cumulate, indicating an intrusive origin for the harrisite. Quantitative textural and chemical analysis suggests that the Cr-spinel seams formed via in situ crystallisation within the crystal mush together with the intrusive peridotites from an assimilation reaction between the replenishing magma and peridotitic crystal mush. Intrusive magma replenishment in Unit 10 caused significant compositional disequilibrium between the crystallising phases in response to the postcumulus migration of reactive liquid, resulting in chemical zoning of intercumulus plagioclase crystals. We propose that the Unit 10 peridotite is intrusive and that repeated small volume magma replenishments are responsible for incremental construction of a large proportion of the peridotite body, similar to recent interpretations of parts of Unit 12 and Unit 14. Moreover, it is suggested that some or all of the injections of magma occurred into the crystal mush, rather than at the magma chamber floor. This new model of intra-mush Cr-spinel formation may have significant economic implications for PGE enrichment in other layered intrusions, such as the peridotite-hosted chromitites of the Stillwater Complex Ultramafic Series (Montana, USA). It is also worth noting that thin platiniferous chromitite seams considered to have formed in situ occur below the Merensky Reef of the Bushveld Complex (South Africa).

Linking In Situ Crystallization and Magma Replenishment via Sill Intrusion in the Rum Western Layered Intrusion, NW Scotland

The construction of layered mafic-ultramafic intrusions has traditionally been attributed to gravitydriven accumulation, involving the mechanical settling of crystals onto the magma chamber floor, at the interface between the crystal mush at the base and overlying replenishing magma, such that the layered sequence of cumulates (i.e., the crystal mush) at the floor aggrades upwards. The Rum Western Layered Intrusion (WLI) is a ~250 m sequence of layered peridotite cumulates comprising the structurally lowest portion of the Rum Layered Suite (RLS). As such, it is taken to represent the oldest sequence in the RLS and has been assumed to young upwards. The WLI hosts the largest proportion of harrisite, a cumulate composed of skeletal olivine that formed by in situ crystallisation, in the Rum layered intrusion. Harrisite layers in the WLI ubiquitously exhibit extremely irregular upward-oriented apophyses, up to several metres high and metres across, alongside laterally extensive dome-like structures; features consistent with intrusive, sill-like emplacement of harrisite. The distribution and abundance of harrisite therefore points to chaotic sill-like emplacement of the magmas that produced at least half of the WLI cumulate. This probably occurred various ambient crystal mush temperatures and punctuated intervals during cumulate formation. The harrisite layers are associated with numerous Cr-spinel seams occurring along the tops, bases, and interiors of these layers, suggesting they formed in situ alongside harrisite sills within the crystal mush. Detailed quantitative textural and mineral chemical analysis of Cr-spinel seams support a simple in situ crystallisation process for their formation. It is suggested the Cr-spinel seams form within melt channels that develop along the same hot tears that allowed the harrisite parental melts to enter the crystal mush. The chemistry and texture of Cr-spinel is controlled by the volume of through-flow of melt through the melt channel. Where melt flux through channels was high, sulphide and platinumgroup minerals are more abundant, highlighting the key economic implications of this model for the platinum-group element enrichment of chromitite horizons in layered intrusions. We also highlight the role of infiltration metasomatism at multiple levels of the WLI, where porous percolation of interstitial melt and reactive liquid flow played a key role in cumulate formation, supporting the notion of layered intrusion growth by incremental sill emplacement.