The self diffusion of oxygen has been studied in prepared natural crystals of albite, grossular garnet, quartz, and ruffle by isotope exchange with hydrothermal water enriched in 18O, and subsequent analysis by ion microprobe. Measured oxygen diffusion coefficients (D) in quartz (‖c) may be described by D = 1.08 × 10−11 exp(−31.5 kcal/RT) cm2s−1 at 600−750°C and 1 kbar water pressure. For grossular, D = 2.5 × 10−16 cm2s−1 at 1050°C and 8 kbar, and D = 4.8 × 10−17 cm2s−1 at 850 °C and 2 kbar. All ruffle crystals exhibited variable amounts of corrosion, and an approximate diffusion coefficient of D(‖c) = 3.16 × 10−15 cm2s−1 cm2s−1 was obtained at 1050 °C and 1 kbar. Oxygen diffusion coefficients in albite, perpendicular to (001) faces, have been determined as a function of pressure at 600 °C Between 0.5 and 8.0 kbar pressure no systematic variation in the results was observed and most of the data may be described by D = 4.1 (±0.5) × 10−15 cm2s−1. Slow oxygen diffusion rates in quartz and garnet suggest that these minerals should have high ‘closure temperatures’ for oxygen exchange, and may provide reliable oxygen isotope exchange geothermometers.

The significance of metastable and equilibrium pyroxene fractionation trends in tholeiitic magmas is discussed, and the development of sector zoning and skeletal growth are considered in relation to Nakamura's (1973) hypothesis of protosites on growing clinopyroxene crystal surfaces. At the Galapagos Spreading Centre (GSC) basalts and ferrobasalts investigated behave paradoxically in that the slower-cooled basalts follow the quench trend while the faster-cooled ferrobasalts define a much closer approach to the equilibrium trend.

It is concluded that under metastable conditions fractionation trends in Ca-rich pyroxenes may be strongly influenced by textural features such as cotectic crystallization of plagioclase and the onset of liquid immiscibility, the latter leading to the development of strongly Fe-enriched ferroaugites. Even under plutonic conditions metastable crystallization can develop and severely reduce the pyroxene miscibility gap. A model for metastable crystallization is presented. These considerations are then addressed to the remarkable correspondence of the Skaergaard and Thingmuli pyroxene fractionation trends.

New field observations of certain harrisite occurrences are presented, including: interruption of layers; splitting of very thick layers into several smaller ones; existence of small isolated lenses and pods of harrisite and of upwardly extending tongues of harrisite; harrisite forming parts of the matrix of breccias; and isolated pods of harrisite along the western margin of the intrusion. These layers, lenses, tongues, and pods seem to have crystallized within the cumulus mush rather than at the mush magma boundary. It is proposed that the rock in these instances represents segregated, intercumulus melt which ‘ponded’ beneath relatively impermeable layers in the cumulus mush. Several ways in which supercooling may have arisen to cause skeletal olivine growth are considered and previous estimates are thought to need reduction by 10–20°C. It is suggested that segregation of upward-filtering melt in other layered intrusions might produce layers indistinguishable from ‘uniform’ cumulate.

Whole-rock, minor element, rare earth, and electron microprobe data are presented for basaltic lavas from the western Kangerdlugssuaq area of East Greenland. Samples were obtained from Professor W. A. Deer's 1936 collection at Triangular Nunataks and Gardiner Plateau, and additional material obtained by sampling moraines on the surface of Kangerdlugssuaq Glacier. Both undersaturated and tholeiitic lavas are present at the Triangular Nunatak locality but the glacier suite is dominantly tholeiitic. The tholeiitic suite is less evolved than tholeiites from the Scoresby Sund area. Undersaturated lavas show enrichment in light rare earth elements and tholeiitic lavas show fiat chondrite-normalized patterns. Tholeiites from the Gardiner Plateau show no Eu anomaly but others show a slight negative Eu anomaly. Chemical data and considerations of regional geology are consistent with Cox's (1980) model of flood basalt vulcanism.

A local area of the Kap Edvard Holm Upper Layered Series gabbro has been converted into a quartz-albite-epidote ilmenite rock by leaching of iron and with a net mass loss. Among the textural changes is the dissolution of the magnetite matrix of magnetite-ilmenite lamellar intergrowths. This is an extreme example of a series of low-temperature changes occurring throughout the intrusion due to the reaction of the gabbro with a volatile-rich phase which separated during the magmatic stage and remained trapped within the walls of the intrusion. The Fe-rich fluid formed by this autometasomatic process is similar to those responsible for the genesis of Cornwall-type magnetite deposits. The probable temperature range for the process is 400–550°C.

The Kapalagulu intrusion displays the following sequence of cumulus phase layering in a stratigraphic sequence of 1400 m: Basal Zone (BZ) olivine ± chromite → Intermediate Zone (IZ) olivine + plag + opx →, olivine + plag + opx + cpx → Main Zone (MZ) plag + opx + cpx → plag + cpx + Fe/Ti oxide + apatite. The corresponding cryptic variation is olivine Fo83 − 77 (limited to BZ and IZ), orthopyroxene En82 − 56, clinopyroxene Ca46Mg45Fe9 to Ca43Mg37Fe21 and plagioclase An88 − 80. Reversals of the cryptic variation occur at the base of MZ (minor reversal) and in the middle of MZ (major reversal), and are attributed to the influx of relatively primitive magma. The major reversal indicates that progressive mixing of fresh and residual magmas occurred. Because of the major reversal, inverted pigeonite appears twice in the layered sequence, but at different compositions (En65 and En56). Unlike the cumulus olivine and pyroxene, cumulus plagioclase exhibits a wide range of composition (5−10% An) in individual rocks and even in single crystals.

Density measurements of eight silicate liquids containing substantial amounts of Fe2O3 have been made over a range of 250 °C. These have been combined with published density measurements on multicomponent silicate liquids to yield (by multiple regression) partial molar volumes of SiO2, TiO2, Al2O3, Fe2O3, FeO, MgO, CaO, Na2O, and K2O. The data on Fe2O3-liquids are neither precise nor abundant enough to show a compositional dependence of . In a liquid of constant composition and temperature, the pressure dependence of the oxygen fugacity is given by

which, if ΔV is independent of pressure, necessitates an increase in fO2 with increasing pressure of about 1 log10 unit for 10 kbars.

Combining an equation relating oxygen fugacity to composition, T, and Fe2O3 at 1 bar (Sack et al., 1980) with the results for partial molar volumes, the oxygen fugacity of any magma can be calculated as a function of P and T. If basic magmas have their Fe2O3/FeO set in the source regions, and ascend isochemically, then the calculated oxygen fugacities in the mantle increase as pressure increases and silica activity decreases. A P-T grid has been constructed to show the calculated oxygen fugacities in a source region which has equilibrated with some common lava types, based on their FeO and Fe2O3 contents.

Major and trace element data are presented for the basic, intermediate, and acid rocks of the Beinn Chaisgidle Centre of the Tertiary igneous complex on the Isle of Mull. The variation in the bulk chemical composition, the iron enrichment trend in the clinopyroxenes, the systematic decrease in concentrations of the transition metal cations, and the increase in the incompatible elements suggest that the magma from which these rocks formed underwent crystal fractionation. The rocks show a coherence of rare earth element patterns for all the rocks in the sequence, and there is a progressive increase in negative Eu anomaly with increasing silica content.

The Pendennis minette has an exceptionally high K2O/Na2O ratio and is strongly peralkaline. Its constituent minerals, which are described in detail, are sanidine, biotite, two carbonates (dolomite and ferroan magnesite), two amphiboles (riebeckite and a new variety, potassium magnesio-arfvedsonite), and quartz. The major and trace dement composition of the minette indicates close affinities with the lamproitic lavas.

The water-undersaturated melting relationships of an orendite (with 1.23% H2O as shown by chemical analysis) from the Leucite Hills, Wyoming, have been determined at pressures up to 30 kbar. The dominant liquidus and near-liquidus phases are leucite, olivine, orthopyroxene, clinopyroxene, and garnet. Leucite is stable only at pressures below 5 kbar, but at 27 kbar, minor olivine, orthopyroxene, clinopyroxene, and garnet crystallize simultaneously at or near the liquidus. The following reaction relationships occur with falling temperature in the orendite magma: (a) a reaction between olivine and melt to yield orthopyroxene at pressures above 12 kbar; (b) a reaction between olivine and melt to yield phlogopite at pressures below 12 kbar; (c) a reaction between olivine, orthopyroxene and melt to yield phlogopite and probably clinopyroxene at pressures above 12 kbar; (d) a reaction between leucite and melt to yield sanidine at pressures below 5 kbar. Electron microprobe analyses demonstrate that the ortho- and clinopyroxenes crystallized from orendite are aluminium-poor; the clinopyroxenes contain insufficient aluminium to balance sodium and titanium (Al < Na+2Ti) and these elements must either be partly balanced by (undetermined) chromium or ferric iron or be involved in substitutions which do not require trivalent ions for charge balance. The experimental results indicate that relatively silica-rich potassic magmas such as orendite form under water-undersaturated (essentially carbon dioxide free) conditions at pressures of about 27 kbar by small degrees of melting of phlogopite-garnet-lherzolite or by larger degrees of melting of peridotite which has been enriched in potassium and incompatible elements. The peralkalinity of some potassic magmas (such as orendite and wyomingite) could reflect a primary geochemical characteristic of the source rock, but could also result from the melting of phlogopite in the presence of residual pyroxenes. The association of silica-poor, mafic madupites and relatively silica-rich orendites and wyomingites in the Leucite Hills can be explained in terms of the relative effects of water and carbon dioxide on melting processes within the upper mantle.