Yttrotungstite occurs at Kramat Pulai mine, and at Tapah, Kinta, Perak, Malaysia, as yellow earthy material and as monoclinic laths, elongated along [001], flattened, and always twinned, on {100} to pseudo-orthorhombic symmetry, and frequently bevelled by {110}; crystals are very rarely terminated by {101}. γ = [010], α:[001] = 26°, probably in β acute, and 2Vα ≃ 68°. A thermal weight loss curve and an infra-red absorption spectrum are given and discussed. Accessory minerals include raspite and stolzite.

The yttrotungstite unit cell has a 6·95, b 8·64, c 5·77, β 104° 56′, space group P21/m; cell contents are (Yt, Ln, Ca, Mg)2(W, Al, Si, Ti, Fe)4(O, OH)14(OH)2·2H2O. Indexed X-ray powder data are given. Microprobe studies show that Al and Si replace W and are concentrated in zones showing larger values of a sin β (6·75 Å instead of 6·72 Å) but with no other appreciable difference in cell dimensions. Crystal structure studies show that the structure consists of WO6 octahedra sharing non-opposite edges in zig-zag chains running parallel to [010]. Yttrium is in approximately trigonal prismatic coordination between the chains, with the water molecule as a seventh neighbour at one prism face. The water molecule is accommodated in the angle between zig-zags in the WO6 chains; it is probably hydrogen bonded to chain oxygen atoms, and becomes coordinated to the yttrium by a shear between chains away from strict close-packing of the oxygen atoms. The shear is related to the change in a sin β in (Al, Si)-rich zones; arguments based on this and on details of the chemical analysis suggest that SiO4 replaces WO6 with a local oxygen deficit.

Oxygen atom peaks on partial Fourier difference syntheses for our data for yttrotungstite are only slightly larger than the troughs in the syntheses due to experimental error. Methods of testing the significance of the positive peaks are described in an appendix.

This paper is primarily concerned with the interpretation of phenomena associated with phase transformations in single crystals, as observed by current experiments using electron-optical techniques. An analysis of permissible metastable behaviour in such single crystals, in terms of lattice theory, is provided. The dominant role of symmetry in defining the constraints on permissible be-haviour of this kind is explained. The resulting theory indicates that, at low temperatures, permissible thermal fluctuations in local chemical composition, or order, become strongly degenerate and lead to the development of non-Bragg diffraction effects. The theory and practical study of the phenomena by electron microscopy are also described.

The theory of metastable behaviour in a single crystal subject to symmetry constraints is then tested by re-examining the experimental data available for two systems that have been studied recently where such behaviour might be suspected. The first of these concerns the low-temperature behaviour of adularia, and the second deals with incipient exsolution phenomena in the alkali feldspar from a pantellerite. The data in each case appear to be compatible with the general theory proposed. Finally the theory is used to analyse the low-temperature behaviour of the intermediate plagioclase feldspars. In this case the theory indicates that both antiphase phenomena and schiller effects are associated with metastable behaviour in a system subject to strong symmetry constraints at low temperature. The implications of the analysis in relation to peristerite exsolution phenomena are discussed.

A simple method has been developed for calculating the d-orbital energy levels of transition-metal ions in coordination polyhedra with both orthogonal and non-orthogonal distortions, using equations based on those derived by Ballhausen (1954). The input data are atomic coordinates, a standard value of the crystal field splitting parameter Δ at known metal-ligand distance, and the ratio of radial integrals B2/B4, which is approximately constant for a given ion. The method can be applied to polyhedra containing different ligands.

Application of the equations to the Mn3+ (M3) site in piemontite and the Fe2+ (M2) site in orthopyroxene gives calculated transition energies in good agreement with the observed band energies.

The calculations permit definite assignment of the great majority of d-d absorption bands even in multi-site phases, and enable discrimination of crystal-field and charge-transfer bands in mineral spectra. They also throw light on the fine structures of both oxygen → metal and metal → metal charge-transfer bands, and allow the calculation of crystal-field stabilization enthalpy and electronic entropy. The latter is a previously neglected energy term that contributes significantly to the energetics of reactions within and between phases containing transition-metal ions.

Zapatalite, Cu3Al4(PO4)3(OH)9·4H2O, was first found at a small prospect north-west of Cerro Morita, 27 kilometers south-west of Agua Prieta, Sonora. It occurs in silicified and brecciated limestone with libethenite, chenevixite, beaverite, alunite, and pseudomalachite.

Zapatalite is pale (faience) blue with pale blue streak; H = 1½, G = 3·016±0·026. It is soluble in cold dilute acids and decomposed by 20% KOH. Duplicate analyses on samples of about 9 and 15 mg gave CuO 27·92, 25·77%; Al2O3 20·16, 20·26%; P2O5 20·69, 22·51%; H2O 15·46, 15·79%; insol. (baryte and quartz), 15·46, 15·79%. This leads to the formula Cu3Al4(PO4)3(OH)9·4H2O.

The cell can be indexed as tetragonal with a 15·22Å, c 11·52Å, and a volume of 2669Å3. If Z = 6, empirical cell contents are Cu19·4Al22·8(PO4)17·6(OH)55·422·2H2O, and the calculated G = 3·017. Strongest lines of the powder pattern are 7·62 (10), 11·60 (10), 5·75 (7), 6·82 (7), 3·04 (5), 2·53 (5), 2·95 (4), and 2·88 (4).

Pale green in thin section with feeble dichroism in green, with ε > ω. The good basal cleavage is a length slow direction. Indices for Na-D are ε = 1·635, ω = 1·646. Usually uniaxial (—), but may be biaxial (—) with variable 2 V.

Named for Emiliano Zapata (1879–1919), a popular hero of the Mexican revolution. Type specimens to be left with British Museum (Natural History) and the University of Arizona.

Infra-red examination of a weathered biotite and of biotites that have been converted to vermiculites and subsequently oxidized, indicates that oxidation of octahedral ferrous ions to ferric ions is associated with a reversible conversion of hydroxyl ions to oxide ions. Subsequently, in high-iron biotites, there is an irreversible loss of ferric ions from the octahedral layer, resulting in an increased number of dioctahedral sites. Electron microscopy and X-ray diffraction indicate that ejected ferric ions form either amorphous interlayer oxides or, when bromine is used as an oxidant, a crystalline external phase of β-FeOOH. The high refractive index of some oxidized vermiculites is shown to be due largely to submicroscopic iron oxides.

The chemical factors controlling the rare appearance of garnet in Connemara amphibolites are elucidated, including the importance of low oxidation ratio (w), low Mg/(Mg+Fe+Mn) (mg), and high MnO. Pressure is believed to be critical in determining whether garnet is common in amphibolites in any one terrain and it is suggested that the ratio mg × w/MnO (called g) in serial analyses of garnetiferous amphibolites can be used as a measure of the pressure of metamorphism.

89 Connemara amphibolite analyses with plagioclase and hornblende are considered including 31 new analyses, which are reported. Of the 89 analyses, 10 contain garnet, 6 cummingtonite, 16 clinopyroxene including 3 with garnet and clinopyroxene and 2 with garnet and cummingtonite. The coexisting garnet and hornblende in 3 amphibolites have been analysed, including one sample in which brown-green and green hornblende, cummingtonite, plagioclase, garnet, and ilmenite have been analysed. Cummingtonite does not coexist with clinopyroxene or sphene and its presence is controlled by pancity of Ca as well as medium to low pressure in metamorphism.

Based on nearly 1500 published amphibole analyses the maximum possible Alvi in hornblendes is shown to increase with increase of Aliv. New analyses of hornblendes from amphibole-corundum rocks, with and without anorthite, are given and after critical examination of the available data it is concluded that the maximum verified Alvi-rich calciferous amphibole that approaches the closest to hypothetical tschermakite comes from a kyanite-bearing aluminous high-pressure-crystallized schist from Lukmanier, Switzerland. Pure natural edenite or ferroedenite is unknown, but a new analysis of the nearest known natural edenite, from Mysore, India, agrees with the postulated view that extraordinarily low temperatures are needed for edenite-ferroedenite crystallization, much below that possible in magmas and only rarely achieved in metamorphic rocks containing amphiboles. The limit of the approach of igneous hornblendes to edenite-ferroedenite and tremolite-ferroactinolite is outlined.

At least 1100 °C is required for complete expulsion of water from some amphiboles.

The person after whom forsterite was named has been confirmed as Jacob Forster (1739–1806), a mineral collector and dealer of London. The discovery of palladium and rhodium by William Hyde Wollaston and of iridium and osmium by Smithson Tennant appears to have been made on crude platinum ore from South America supplied by Forster.

An account of the eastern part of the Galway Granite is presented together with the first detailed geological map. An aphyric medium-grained alkali granite occurs at the margin of the pluton followed by a coarse porphyritic granite, which becomes more basic towards the granite centre, consistent with findings in the western and northern parts of the batholith. Westwards the coarse porphyritic granite develops a foliation in which xenoliths, and occasionally potash feldspar phenocrysts, are aligned. Chemical fractionation trends illustrate the consanguinity of the granites and aplites. Field, petrographic, and chemical evidence suggest that most of the xenoliths are cognate.

The zonation of the granite is thought to have resulted from crystallization of a magma in which compositional gradients were set up during the early crystallization period. A temperature gradient, decreasing from the centre of the magma chamber outwards into the country rocks, resulted in the migration of water, accompanied by alkalis and other volatiles, towards the granite margin; also the early-crystallizing minerals displaced the residual magma outwards. Crystallization was followed by shearing in the deeper parts of the pluton to produce the granite foliation. A total of 166 rocks have been chemically analysed for 31 constituents.

Holtite, a new mineral allied to dumortierite, occurred as pebbles with stibiotantalite and tantalite on an alluvial tin lease near Greenbushes, Western Australia, and is named after the late H. E. Holt, Prime Minister of Australia.

The mineral is orthorhombic with a 11·905 Å, b 20·355 Å, c 4·690 Å, space group Pmcn, weak supercell 2a, 2b, c developed. Crystals are elongated along c, D 3·90 ± 0·02, hardness 8½, fluorescent. Optical properties α 1·743−1·746, mainly yellow, ‖ [001], β 1·756−1·759, colourless, γ 1·758−1·761, colourless, 2Vα 49−55°, r < v. X-ray powder data are given.

Chemical analysis gave SiO2 20·30, Sb2O5 4·61, Al2O3 46·43, Ta2O5 11·24, Nb2O5 0·76, Fe2O3 0·27, MnO 0·05, TiO2 0·09, BeO 0·05, B2O3 1·82, Sb2O3 13·89, H2O+ 0·38, H2O− 0·08, sum 99·97%. On a water-free basis the unit cell contains Al24·5Sb2·56‴Ta1·36Sb0·76vNb0·16Fe0·10‴Be0·05Ti0·03Mn0·02B1·40‴Si9·09O66·85. Compared with dumortierite, 4[(A1,Fe)7BSi3O18] or 4 (X11O18), the holtite unit cell contains approximately 4(X10O17).

Type material is preserved at the Government Chemical Laboratories, Perth, Western Australia.

1666 moldavites from 19 localities in Bohemia and 617 moldavites from 10 localities in Moravia were measured to yield parameters L (length), B (breadth), and T (thickness). A digital computer was programmed to calculate values T/L, (L-B)/(L-T), and according to Sneed and Folk (1958) and values q = B/L, p = T/B, and F = p/q according to Zingg (1935). Averages of large numbers of values characterize Bohemian and Moravian localities. Moldavites from most localities in Bohemia are often drop-like, flat or elongate with average values of between 0·53 and 0·66 and F between 0·68 and 0·91. Moldavites from Moravia are frequently massive to spheroidal with average in the range 0·60 to 0·75 and F between 0·87 and 1·08. Among the localities in Bohemia, four have moldavites with morphological features of Moravian moldavites. The data obtained permit an interpretation that localities of moldavites with low average sphericity contain less heated material. Less heated and thus more viscous glass could have flown along shorter trajectories (Bohemia). More heated glass was ejected to geographically more distant places (Moravia) or travelled along relatively long steeper trajectories (several localities in Bohemia).

In the Four Peaks region of Labrador Precambrian hornblende granulite facies metamorphosed basic dykes discordantly cross the foliation of garnet-hornblende-pyroxene granulites with sillimanite-garnet-biotite-quartz-feldspar gneiss. Most of the dykes are schistose or banded, but some are massive and undeformed with blastophitic textures. Mineral analyses are given for the equilibrium assemblage in one analysed clinopyroxene-orthopyroxene-amphibole-garnet-plagioclase- quartz blastophitic dyke. Dyke intrusion occurred after formation of the host rock foliation, while the metamorphism was about 1400 Myr ago.

The solidus temperatures (IPTS-68) and melilite compositions (mole %) determined for some four-phase assemblages in the system CaO-MgO-Al2O3-SiO2 are: anorthite, diopside, spinel melilite (Åk81Geh19), 1236·5 °C; diopside, spinel, forsterite, melilite (Åk83Geh17), 1238 °C; foresterite, spinel, monticellite, melilite (Åk82Geh18), 1351 °C; monticellite, merwinite, spinel, melilite (Åk52Geh48), 1378 °C; dicalcium-silicate, Ca5MgSi3O12, spinel, melilite (Åk5Geh95), ca. 1412 °C; dicalcium-silicate, Ca6MgAl8SiO21, spinel, melilite (Åk4Geh96), ca. 1347 °C.

Embreyite is a new mineral that has been found only in old specimens collected at Berezov, Siberia. The composition is Pb4·97(CrO4)2·00(PO4)1·91.0·75H2O, or Pb5(CrO4)2(PO4)2. H2O based on both wet and electron-probe analyses. Z = I, Dmeas 6·45, Deale 6·41. Crystals are monoclinic with a 9·755 Å, b 5·636, c 7·135, β 103°5′ the space group may be P21/m. No single crystals were found. α = 2·20, β = γ = 2·36. Colour in various shades of orange with a yellow streak. H = 3½; no cleavages observed. Occurs with vauquelinite, crocoite, and phoenicochroite in the oxide zone assemblage from Berezov.

The crystal structure of coalingite (Mg10Fe2(OH)24(CO3)·2H2O) has been determined using single-crystal X-ray methods. The mineral is trigonal, with space group Rm, aH = 3·12, cH = 37·4 Å, Z = ½, and (0001) cleavage. The structure is of a layer type, and is based on a structural element about 12·5 Å thick in the c-direction and consisting of two brucite-like layers and one disordered layer containing carbonate ions and water molecules and resembling those in sjögrenite and pyroaurite. The unit cell comprises three of these structural elements stacked together in the c-direction. The Mg2+ and Fe3+ ions are randomly distributed among all the octahedral sites of the brucite-like layers. The structure closely resembles those of sjögrenite and pyroaurite, but has two brucite-like layers between each CO32−−H2O layer where these have one. There is a tendency to random interstratification, and the crystals appear to contain intergrown regions of brucite and of sjögrenite or pyroaurite. Coalingite-K probably has a similar structure, but with three brucite-like layers between each -H2O layer; its idealized formula is probably Mg16Fe2(OH)36(CO3).2H2O.

K-feldspar variation is described for four salic plutonic igneous complexes that show normative mineral variation in keeping with the course of fractionation predicted from experimental studies. These are a sodic syenite (Loch Ailsh), potassic syenites (Loch Loyal), a peralkaline syenite and soda granite complex (Puklen), and a calc-alkaline pluton (Foyers). Additional data are given for examples from the literature. Microcline becomes more abundant relative to orthoclase as members of these intrusive sequences approach thermal minima on the liquidus in the system Ab-Or-An-Qz. This relation exists irrespective of the compositional range of the rock suites, the bulk composition of the alkali feldspars, grain size, or field relations. Early orthoclase-bearing rocks retain orthoclase when enclosed as xenoliths in microcline-bearing rocks, or when cut by minor bodies of microcline-bearing rocks. The K-feldspar variation may reflect increasing water content or the peralkaline or peraluminous character of the fractionating magmas. These factors affect the feldspar structure at the time of initial crystal growth and dictate whether microcline will develop on cooling.

The local development of strongly oxidizing conditions during crystallization of a recent basalt lava lake on Piton de la Fournaise, Réunion, led to the production of approximately 15 % residual rhyodacite liquid. This was segregated into the vesicle cavities as the lava solidified, and congealed as a glass containing microlites of feldspar, clinopyroxene, and hematite. The rhyodacite is compared chemically with the residual glasses recorded from other basaltic and doleritic occurrences, and with the Mauna Kuwale rhyodacite of Oahu.

Without direct evidence that cations favour particular sites in the octahedral layers of phlogopite-biotite micas, it is usually assumed that the cations are distributed randomly. Such a distribution, however, implies appreciable populations of cation combinations that seem electrostatically improbable. A computer program was written that rearranges a random distribution so that the improbable combinations of cations are eliminated by predetermined restrictions. Its use shows that conditions deduced from Pauling's electrostatic valence rule lead to grouping of trivalent cations around vacancies; for some chemical compositions, chains of alternate trivalent cations and vacancies extend through the structure.

The crystal structure of kilchoanite, Ca6(SiO4)(Si3O10), has been determined from three- dimensional X-ray methods (r = 0·089 on 348 independent reflections). It is closely related to that of γ-Ca2SiO4, which is olivine-like. In kilchoanite, slabs of γ-Ca2SiO4 structure alternate with ones composed of Ca ions and finite chains of three tetrahedra; these have the composition Si3O108− and have rarely been observed previously. A related, synthetic phase, 8CaO. 5SIO2, probably has a similar structure but with a higher proportion of γ-Ca2SiO4 regions, and should thus be formulated as Ca8(SiO4)2(Si3O10). Disordered synthetic phases also occur that appear to have similar structures with still higher proportions of γ-Ca2SiO4 regions. These phases collectively form an interesting contrast to the chondrodite minerals.