The residual liquids of many tholeiitic basalts and andesites, on cooling, split into iron-rich and silica-rich fractions, which may quench to brown glassy globules and clear glass respectively. More commonly, however, cooling is sufficiently slow for the iron-rich liquid to crystallize to globular single crystals of pyroxene. Depending on the cooling rate, these crystallized globules range in shape from spheres to elongated globules bounded by crystal faces. The fine grain size of the mesostasis of most tholeiites is partly due to these small crystallized globules. The silica-rich fraction, on the other hand, is more commonly quenched to a glass, and when preserved as globules in the crystallized iron-rich fraction, it may be bounded by negative crystal faces of the surrounding pyroxene. Globules of the iron-rich liquid commonly nucleate on the surface of the plagioclase crystals where they can become trapped, later crystallizing to spherical pyroxene grains that mostly contain a minor opaque phase. In contrast, iron-rich globules that form next to pyroxene grains commonly become attached to these crystals, giving them lobate boundaries. The immiscible silica-rich liquid becomes trapped between these lobes and, with sufficiently slow cooling, results in finger-like quartzo-feldspathic inclusions extending in from the margins of the pyroxene grains. These textures can provide evidence of immiscibility in a wide range of volcanic and hypabyssal rocks, even when no glass is present.

Several discordant, garnet-bearing bodies are described from the St. Just part of the Land's End granite aureole, Cornwall. It is suggested that these formed by a combination of metasomatic replacement and dilation with concomitant precipitation. Other Ca-Fe aluminosilicates (containing large amounts of Zn, Sn, B, and Be) are also present in the bodies. The chemical, petrographic, and field evidence suggests that the introduced material was transported by hydrothermal fluids. These fluids could well be related to the intrusion of the granite and the earliest phases of ore mineralization. Chemical analyses of garnets, calcic amphiboles and one apatite, epidote, diopside, and axinite are given.

Partial replacement of original igneous augite crystals by omphacite during blueschist metamorphism of a dolerite from the Mihalliççik area of north-west Turkey has been studied by transmission electron microscopy. The replacement occurred topotactically, apparently by ion exchange with a fluid phase, which left the basic pyroxene structure unchanged. Cation ordering in the omphacite caused a symmetry change from C-face centred to primitive with the formation of fine-scale antiphase domains. Selected-area diffraction provides evidence for P2 and P2/c space groups for the ordered omphacite though the best ordered areas show a tendency towards P2/n (reflections violating the n-glide are very weak) and also contain fine, wavy, disordered precipitates approximately parallel to (too).

It is suggested that the replacement temperature was below the cation-ordering temperature and that the omphacite grew in a metastable, disordered state. Subsequent ordering occurred under irreversible conditions via a series of intermediate structures. The ordering sequence may illustrate the relationship between different ordering schemes in other blueschist pyroxenes.

An occurrence of aegirine in Middle Devonian lacustrine beds in Caithness is described. The mineral occurs as euhedral to subhedral prismatic crystals within a finely laminated siltstone. Microprobe analyses reveal a remarkably ‘pure’ composition consisting almost entirely of Na, Fe3+, Si, and O. X-ray diffractometry yielded the unit cell data: a 9·657 Å, b 8·800 Å, c 5·296 Å, and β 107·37°. Comparison with other natural aegirine compositions shows marked differences from those of igneous or metamorphic origin but a close similarity to a unique occurrence of authigenic aegirine in the lacustrine Green River Formation of Wyoming, U.S.A. Consideration of the nature of the Caithness occurrence, and comparison of the Devonian Orcadian basin with the Eocene Green River basin indicate similar geological environments. It is suggested that the mineral formed in hypersaline conditions in which sodium may have been derived from contemporaneous alkaline volcanism within the Orcadian basin.

Sector-zoning is brought out by mineral inclusions (plagioclase, biotite, hornblende, sphene, and apatite) in 2 to 5 cm microcline megacrysts of the Godani Valley pluton, Nigeria. The inclusions are concentrated in the megacryst rims beneath the faces {110} and {20}; the megacryst cores are not sector-zoned. Comparison with other occurrences (Portugal and Czechoslovakia) shows that this type of sector-zoning is not controlled by growth rate. Chemical sector-zoning does not occur in the Godani megacrysts. Growth from magma is proposed, with preference for inclusions on faces at a high angle to the feldspar a-chain. A superficially similar type of sector-zoning occurs in microcline megacrysts of both pluton and wall rock at Papoose Flat, California, suggesting that this feature can be generated either by growth from magma or by replacement.

Surface microtopography of habit faces, and internal textures of corresponding growth sectors of adularia from the Rhonegletscher, Switzerland, and the Seikoshi mine, Japan, were studied by means of optical polarizing microscopy, reflection interference contrast, electron microscopy, and infra-red absorption spectroscopy. Thin sections parallel to the growth plane were prepared from the specimen whose as-grown face was pasted on a glass slide. Most internal textures seen optically in the thin sections correlated well with the growth patterns for both specimens (fig. 1). Thin foils prepared by ion-thinning from the Rhonegletscher adularia show, under the electron microscope, coexisting heavily cross-hatched areas giving diffuse diffraction spots and homogeneous areas giving sharp diffraction spots. As a whole, adularia from the Seikoshi mine gives more homogeneous and much weaker cross-hatching than the Rhonegletscher specimens. Also, infra-red absorption spectra, which correlate very well with the Al/Si ordering and the optics, show that the {110} growth sector consists of triclinic disordered K-feldspar and the {001} and {01} sectors monoclinic high or low sanidine or orthoclase.

These observations indicate that the characteristic internal textures seen on all optical micrographs are formed during the process of crystal growth, and are not due to monoclinic-triclinic transition. Although the transition process has taken place to a slight extent, the original growth textures remain well preserved throughout geological time. It is suggested from these observations that the degree of Al/Si ordering is mainly controlled by the different rates of growth and the different two-dimensional atomic arrangements exposed on the various growth surfaces. The higher the growth rate the more disordered are the phases that form. The symmetry of the two-dimensional atomic arrangement is kept during growth regardless of the degree of ordering; {001} and {01} growth sectors whose crystal surfaces are parallel to b-axis are monoclinic, and the {110} sector whose surface is inclined to the b-axis is triclinic.

Magnetite is common in carbonatites, and usually has octahedral habit. Abnormal crystals from Galapo, Tanzania, were long thought to be unique, but such crystals are now known to occur at seven localities in four countries. The magnetite is titaniferous, but not unusual in composition for its provenance. The abnormal crystals typically have large growth pyramids extending in one direction (occasionally in two or more directions). They appear to be formed by superimposition of macroscopic growth sheets, parallel to octahedral and other planes, giving a planar structure to the growth pyramid. The edges of the sheets may form low-index or more complex surfaces. Some sheets project beyond the bounding planes of the growth pyramid, which is not repeated in crystallographically equivalent orientations. This unusual mode of growth will be called stratiform. Recognition of these distinctive crystals elsewhere might reveal the presence of hitherto unsuspected carbonatites.

Electron-probe analyses and cell dimensions are reported for twenty members of the tapiolite group. The minerals can be placed in ordered and disordered structural groups, the latter with a significantly larger c dimension. The a and c dimensions for both groups are dependent mainly on their titanium contents. The disordered set form part of a discontinuous series through strüverite/ilmenorutile to rutile. The chemical and structural data are incompatible with the existence of ferrotantalite or mossitc as valid species.

An uncommon sample of a pale-blue Madagascar cordierite and its mineral inclusions of different kinds have been studied. The physical properties of the embedding cordierite are: α = 1·528, β = 1·522, γ = 1·537, δ = 0·009, 2Vα = 96°, density = 2·534 g/cm3, a = 17·082 Å, b = 9·728, c = 9·352, Δ = 0·26. The electron probe analysis gave SiO2=48·64%, Al2O3 = 35·08, FeO = 0·70, MgO = 12·35, H2O+ = 2·50. This is therefore a hydrous magnesian cordierite. Its structural state is related to the shift in α, β, γ, 2Vα, and Δ with heating. Probe analyses were performed on the inclusions: hydro-xyapatite, dolomite, and rutile. The occurrence of dolomite suggests the conditions of formation of the host mineral.

Garnet containing up to 26% SnO2, and pyroxene and spinel containing 12% and 10% SnO2 respectively, occur in a tin-refining slag in association with cassiterite. These tin contents probably represent minimum values for the limit of ionic replacement of Sn4+ in the structure of these minerals. The absence of tin from coexisting melilite suggests that Sn4+ is only able to substitute for ions in octahedral co-ordination.

The Dhajala meteorite shower, the latest recorded fall in India, comprises pieces fragmented near the lower limit of the break-up zone. The meteorite is a chondrite consisting of silica-rich chondrules making up to 34 vol.%, and fine-grained, Fe-Ni metal-rich groundmass admixed with a few irregular lithic fragments. Dhajala contains olivine of variable compositions, ortho- and clinopyroxenes, troilite, kamacite, taenite, chromite, and clear to opaque glass; magnetite occurs only in the fusion crust. The chondrules, which are of variable shape, mineral composition, and texture, represent different stages of quenching, leading to incomplete crystallization of minerals and some degree of disparity between norm and mode. Lithophile elements are less in Dhajala than in average chondrites. Two chemical analyses of Dhajala are presented and it has 27·10% total iron and 2·38 % sulphur (maximum). Chemical and petrological data indicate that it is an H3 olivine-bronzite chondrite. Evidence for the crystallization of chondrules from melt is overwhelming. Some chondrules have been permeated by later troilite and NiFe from the groundmass, which might have crystallized directly from a gaseous environment.

Several types of microscope-mounted drills have been described in literature to obtain pure mineral material from thin sections or polished sections (e.g. Granigg, 1915; Moritz, 1929; Haycock, 1931; Russanow, 1937; Machairas, 1967), but all of them appear to be of limited mechanical stability and reliability. This paper describes an improved drill to extract under the microscope minute quantities of pure material from thin or polished sections. The drilling is performed in a liquid (silicon-oil) to minimize loss of the drilled-out mineral powder. With the help of a micromanipulator the liquid, together with the borings, is drawn up into a microcapillary tube (less than 30 µm in diameter) made of Lindemann glass. For identification of the mineral powder the microcapillary is mounted in a Debye-Scherrer X-ray diffraction camera.

Ellipsometric measurements in the spectral range 250–850 nm have been made on polished single crystals of natural cuprite from Namibia and of artificial cuprite grown by a total-oxidation method from pure copper. Derived values of the refractive indices, absorption coefficients, normal reflectances, and the complex dielectric constants for the two materials are all closely similar, suggesting that the stoichiometry and purity are essentially the same. Values of the absorption coefficient at the red end of the spectrum are inconsistent with the observed transparency of the crystals, and are compatible with a 10 nm surface oxidation film of cupric oxide. This study is part of a wider survey of the effects of doping and stoichiometry on the optical parameters of cuprite.

A study of the post-metamorphic hydrothermal mineralization in the gabbros of the Lizard has already been published (Seager, 1971). The present paper extends the investigation of mineralization in the peninsula to an adjacent area, on the NE. coast around Porthkerris Point. Table 1 gives the more important events of the paragenesis. Calcite and adularia are the only common minerals; all others are scarce or rare.

The calcite that precedes and succeeds adularia is called early calcite and late calcite, respectively. Both early and late calcite occur in several generations differing in habit, fluorescence, and phosphorescence.

Ophiolitic rocks occur as a tectonic mélange in the Mutki area of the Eastern Taurus Mountains of south-eastern Turkey. They form the upper part of a Tethyan ophiolite-flysch complex, which is thrust southward over sedimentary rocks of the Arabian foreland (Hall, 1976). The tectonic mélange has a matrix of serpentinite and includes blocks of basic volcanics, gabbros, picrites, and rodingites, most of which have suffered metamorphism and metasomatism. The volcanic rocks have been metamorphosed under conditions transitional between the glaucophane-lawsonite schist facies and the greenschist or greenschist-amphibolite transitional facies of Turner (1968). The picrites have escaped any significant metamorphism, while the gabbros have been partially or completely recrystallized under greenschist facies conditions. Both picrites and gabbros have also suffered calcium metasomatism resulting in the alteration of some of the gabbros to rodingites. Pyroxenes from eight separate blocks from the mélange have been analysed by microprobe (fig. 1) to determine if the pyroxene chemistry is consistent with an igneous origin (as suggested by textural evidence) or if there have been changes due to metamorphism and metasomatism.