New chemical analyses, cell-sizes, and specific gravities are given for 12 garnets, and TiO2 values, cell-sizes, and refractive indices are presented for a further 33 garnets all belonging to the andradite-melanite-schorlomite series. The increase of refractive index, cell-size, and specific gravity with increasing titania is shown graphically. A plot of Ti against Si atoms in the unit cell suggests that Ti replaces Si rather than Fe3+ in the structure, while electron-probe results indicate that Si or (Si+Fe) may be replaced by Ti. The chemical and infra-red absorption data are considered in a discussion of the structural role of titanium in these garnets : both Ti4+ and Ti3+ may be present.

The literature contains considerable disagreements on the relative stabilities of the members of the copper hydroxyl sulphate family. Titration of copper sulphate with sodium hydroxide is claimed by some to produce only brochantite, while other reports indicate that antlerite and a dihydrate of antlerite are produced in the titration. Most stability field diagrams show that antlerite is the more stable stoichiomer at pH 4 and sulphate activity of 0.05–1. We have reexamined this stoichiometric family by titration of aqueous copper sulphate with sodiumhydroxide and sodium carbonate, reverse titration of sodiumhydroxide with copper sulphate and simultaneous addition of copper sulphate and sodium hydroxide at a variety of mole ratios, concentrations, temperatures and reaction times. We have also explored the reaction of copper hydroxide with copper sulphate and the reaction of weak bases, such as sodiumacetate, sodiumcarbonate and urea, with copper sulphate. Our work indicates that: (1) antlerite is not formed in reactions of 0.05 to 1.2 M CuSO4 with 0.05–1.0 M NaOH or Na2CO3 at room temperature; (2) antlerite is formed in the addition of small concentrations of base (≤0.01 M) to 1 M CuSO4 at 80°C, but not at roomtem perature or with 0.01 M CuSO4 at 80°C; (3) the formation of Cu5(SO4)2(OH)6·4H2O occurs at large Cu2+ to base mole ratios; (4) the compound described in the literature as antlerite dihydrate is actually Cu5(SO4)2(OH)6.4H2O; (5) at mole ratios of Cu2+ to OH– ranging from 2:1 to 1:2 the predominant product is brochantite; and (6) brochantite and Cu5(SO4)2(OH)6.4H2O are converted to antlerite in the presence of 1 M CuSO4 (the latter requires temperatures of 80°C or greater).

The Ksp (ion activity product) values of antlerite and brochantite were determined to be 2.53 (0.01)⨯10−48 and 1.01 (0.01)⨯10−69, respectively, using atomic absorption spectroscopy and Visual MINTEQ after equilibration in solutions of varying ionic strength and pH for six days. These values are in good agreement with those from the literature. However, after 6 months, antlerite in contact with solution is partially converted to brochantite and hence is metastable with a relatively low conversion rate. The Ksp value for antlerite must therefore be considered approximate. The relative stabilities of the copper hydroxyl sulphates are rationalized using appropriate equations and Gibbs energy calculations. A Gibbs free energy of formation for Cu5(SO4)2(OH)6.4H2O of –3442 kJ/mol was obtained from the simple salt approximation.

Many natural minerals and synthetic materials display twin microstructures resulting from displacive phase transitions. These microstructures may be removed temporarily from the sample by heating above the relevant transition temperature, though the twinning generally returns on subsequent cooling.

In anorthoclase, the spatial distributions of twins before and after brief annealing above TC are often identical. This property appears to be a common feature in many materials which undergo ferroelastic phase transitions, and is known as ‘twin memory’. The atomic mechanisms responsible for this twin memory may be investigated by studying the annealing regimes required to remove the memory effect; how long must a sample be annealed, and at what temperature, to induce ‘twin amnesia’.

High-resolution X-ray diffraction (XRD) has been used to investigate twin memory and twin amnesia in anorthoclase. In anorthoclase, the primary constraint on twin amnesia is thermodynamic, rather than kinetic. The critical temperature to induce amnesia correlates well with the top of the (Na, K) solvus in disordered alkali feldspar. For this reason, the proposed mechanism for twin memory involves the segregation of alkali cations in thin lamellae at the twin boundaries.

Serpentinised ultramafic bodies containing zoned chromite grains occur in the Cordillera Frontal Range, western Argentina. Chromites show Zn-rich Al-chromite cores (4.04 wt.% ZnO) surrounded by ferritchromite rims (1.3 wt.% ZnO) and outer Cr-magnetite rims. In intensely altered chromites which are spatially related to sulphide mineralisation, the primary chromite cores have been replaced by Zn-rich ferritchromit (7 wt.% ZnO) and they are rimmed by Cr-magnetite. In the Al-chromites the mean [Cr/(Cr + Al] ratio is 0.53 and the [Mg/(Mg + Fe)] ratio is 0.53; they plot in the field of Alpinetype intrusions. Ferritchromit has a mean [Cr/(Cr + Al)] ratio of 0.93 and [Mg/(Mg + Fe)] ratio of 0.4. In Cr-magnetites the mean [Cr/(Cr + Al) is 0.98 and the [Mg/(Mg + Fe)] ratio is 0. Ferritchromit is always surrounded by a Cr-magnetite rim and it was formed as a reaction product owing to the irreversible dissolution of primary chromite cores. The dissolution of these cores provided the essential components for ferritchromit growth. Zn was introduced into the chromite cores and ferritchromit rims during the formation of the latter. Step-scan profiles have shown that the Zn content in the cores increases from the centre to their outer border, where they show the highest Zn values. It is suggested that Zn was introduced by the fluid phase involved in the alteration process that affected the cores and lead to the formation of the zoned chromite grains. This alteration process was also responsible for the changes in [Cr/(Cr + Al)] and [Mg/(Mg + Fe)] ratios.

In many areas of south and south-eastern Asia, concentrations of As in ground water have been found to exceed the WHO maximum concentration limit of 10 μg/l. This is adversely affecting the health of millions of people and has grave current and future health implications. It has recently been suggested that extensive abstraction of ground water in these areas may accelerate the release of As to ground water. This study uses geochemical and isotopic data to assess this hypothesis. The area investigated in this study is in the Chakdaha block of the Nadia District, West Bengal. The ground water is predominantly of the Ca-Mg-HCO3 type, although some samples were found to contain elevated concentrations of Na, Cl and SO4. This is thought to reflect a greater degree of water-rock interaction at the locations of these particular samples. Arsenic concentrations exceeded the national limit of 50 μg/l in 13 of the 22 samples collected. Four of the 13 samples with high As were recovered from tubewells with depths of 60 m or more. Shallow ground water samples were found to have a stable isotopic composition which falls subparallel to the Global Meteoric Water Line. This probably represents a contribution of evaporated surface water to the ground water, possibly from surface ponds or re-infiltrating irrigation water. Deep ground water, conversely, was shown to have a composition that closely reflects that of meteoric water. The data presented in this study suggest that, whilst the drawdown of surface waters may drive As release in shallow ground waters, it is not responsible for driving As release in deep ground water. However, local abstraction may have resulted in changes in the ground water flow regime of the area, with contaminated shallow ground waters being drawn into previously uncontaminated deep aquifers.

Gallium abundances, determined by laser ablation-inductively coupled plasma-mass spectrometry, are presented for phenocrysts and glassy matrices from a metaluminous trachyte and five peralkaline rhyolites from the Greater Olkaria Volcanic Complex, Kenya Rift Valley. Abundances in the glasses range from 28.9 to 33.3 ppm, comparable with peralkaline rhyolites elsewhere. Phenocryst Ga abundances (in ppm) are: sanidine 31.5–45.3; fayalite 0.02–0.22; hedenbergite 3.3–6.3; amphibole 12; biotite 72; ilmenite 0.56–0.72; titanomagnetite 32; chevkinite-(Ce) 364. The mafic phases and chevkinite-(Ce) are enriched in Ga relative to Al, whereas Ga/Al ratios in sanidine are smaller than in coexisting glass. Apparent partition coefficients range from <0.01 in fayalite to 12 in chevkinite-(Ce). Coefficients for hedenbergite, ilmenite and titanomagnetite decrease as melts become peralkaline. The sharp increase in Ga/Al in the more fractionated members of alkaline magmatic suites probably results from alkali feldspar-dominated fractionation. Case studies are presented to show that the Ga/Al ratio may be a sensitive indicator of such petrogenetic processes as magma mixing, interaction of melts with F-rich volatile phases, mineral accumulation and volatile-induced crustal anatexis.

The Prins Christians Sund rapakivi granite pluton in South Greenland is a member of the early Proterozoic ‘Rapakivi Suite’ and is emplaced into early Proterozoic Ketilidian migmatites. The pluton is composed predominantly of black or dark brown monzonites and quartz monzonites (collectively, rapakivi granites), although a localised white facies is developed adjacent to metasedimentary xenoliths. The back rapakivi granites are extremely fresh and have an anhydrous primary mafic mineralogy of olivine and orthopyroxene, with rare inverted pigeonite and clinopyroxene; minor amounts of biotite and amphibole occur in most fayalite-bearing rapakivi granites. Feldspars in these rocks are black and non-turbid. The white rapakivi granites have a wholly hydrous mafic silicate assemblage and turbid, white or cream-coloured feldspars. Electron microprobe analyses of the mafic silicates in the black rapakivi granites show that they are Fe-rich, comprising fayalite (Fa93−96.5), orthopyroxene (Fs77−81), ferro-pargasitic and ferro-edenitic hornblende (Fe/(Fe + Mg) = 0.72−0.93), and biotite (Fe/(Fe + Mg) = 0.77−0.88). Both biotite and amphibole crystallised subsolidus, and often adopt symplectic morphologies. Biotite has formed in response to a fayalite-consuming reaction at temperatures below 650–700°C and fO2, of 10−16.5 to 10−17.5 bars, and continued to grow under reducing conditions below the QFM buffer to temperatures below 450–500°C. Orthopyroxene formed in response to a low-pressure fayalite-consuming reaction in the melt. The correlation of black, pristine feldspar with anhydrous mafic silicates, and of turbid feldspar with hydrous phases suggests either that the feldspars reflect the anhydrous nature of the parent magma, or more likely that the mafic mineralogy of the white rapakivi granites is secondary.

The powder diffractograms of twenty wet chemically analysed samples in the isomorphous triphylite-lithiophilite series and five synthesized members with Fe/(Fe + Mn) = 1, 0.75, 0.50, 0.25, and 0.0, were recorded. Their unit cell dimensions were accurately refined in order to find a reliable method for semi-quantitative determination of the divalent cation content of these minerals. A multivariate best fit analysis based on Kummell’s procedure shows the marked influence of Fe2+ and Mg2+ on the cell dimensions, as well as that of small amounts of Fe3+ substituting for Mn2+, following LiMn2+ → □ Fe3+. The best representation of the correlation between chemical composition and cell parameters is given by the equations:

a = 6.1041 − 0.0245 Fetot − 0.049 Mg2+

b = 10.4511 − Fetot − 0.082 Mg2+

c = 4.7400 − 0.0130(Fe2+ + Mg2+) − 0.025 Fe3+.

No evidence of non-linearity has been found for the variation of the three cell dimensions with the chemical composition. Assuming the absence of appreciable amounts of Mg2+, the following set of equations is proposed: Fetot = 41(6.104–a); Fetot = 35(10.451–b); Fetot = 77(4.740–c) in which the c dimension gives a relatively poor estimate.

Two sets of determinative graphs were constructed, one based on the cell parameter variation, and the other on the 311, 222, and 142 reflection angular positions, v. the total iron content of these minerals. These two methods, whose reliability is examined, can be used for determination of the divalent cation content, provided the samples contain less than about 0.5 wt.% Na2O, 0.5 wt.% CaO, and 3 wt. % Fe2O3, and are homogeneous.

Two new minerals – manganoblödite (IMA2012–029), ideally Na2Mn(SO4)2·4H2O, and cobaltoblödite (IMA2012–059), ideally Na2Co(SO4)2·4H2O, the Mn-dominant and Co-dominant analogues of blödite, respectively, were found at the Blue Lizard mine, San Juan County, Utah, USA. They are closely associated with blödite (Mn-Co-Ni-bearing), chalcanthite, gypsum, sideronatrite, johannite, quartz and feldspar. Both new minerals occur as aggregates of anhedral grains up to 60 μm (manganoblödite) and 200 μm (cobaltoblödite) forming thin crusts covering areas up to 2 × 2 cm on the surface of other sulfates. Both new species often occur as intimate intergrowths with each other and also with Mn-Co-Ni-bearing blödite. Manganoblödite and cobaltoblödite are transparent, colourless in single grains and reddish-pink in aggregates and crusts, with a white streak and vitreous lustre. Their Mohs' hardness is ∼2½. They are brittle, have uneven fracture and no obvious parting or cleavage. The measured and calculated densities are Dmeas = 2.25(2) g cm−3 and Dcalc = 2.338 g cm−3 for manganoblödite and Dmeas = 2.29(2) g cm−3 and Dcalc = 2.347 g cm−3 for cobaltoblödite. Optically both species are biaxial negative. The mean refractive indices are α = 1.493(2), β = 1.498(2) and γ = 1.501(2) for manganoblödite and α = 1.498(2), β = 1.503(2) and γ = 1.505(2) for cobaltoblödite. The chemical composition of manganoblödite (wt.%, electron-microprobe data) is: Na2O 16.94, MgO 3.29, MnO 8.80, CoO 2.96, NiO 1.34, SO3 45.39, H2O (calc.) 20.14, total 98.86. The empirical formula, calculated on the basis of 12 O a.p.f.u., is: Na1.96(Mn0.44Mg0.29Co0.14Ni0.06)Σ0.93S2.03O8·4H2O. The chemical composition of cobaltoblödite (wt.%, electron-microprobe data) is: Na2O 17.00, MgO 3.42, MnO 3.38, CoO 7.52, NiO 2.53, SO3 45.41, H2O (calc.) 20.20, total 99.46. The empirical formula, calculated on the basis of 12 O a.p.f.u., is: Na1.96(Co0.36Mg0.30Mn0.17Ni0.12)Σ 0.95S2.02O8·4H2O. Both minerals are monoclinic, space group P21/a, with a = 11.137(2), b = 8.279(1), c = 5.5381(9) Å, β = 100.42(1)°, V = 502.20(14) Å3 and Z = 2 (manganoblödite); and a = 11.147(1), b = 8.268(1), C = 5.5396(7) Å, β = 100.517(11)°, V = 501.97(10) Å3 and Z = 2 (cobaltoblödite). The strongest diffractions from X-ray powder pattern [listed as (d,Å(I)(hkl)] are for manganoblödite: 4.556(70)(210, 011); 4.266(45)(01); 3.791(26)(11); 3.338(21)(310); 3.291(100)(220, 021), 3.256(67)(211, 21), 2.968(22)(21), 2.647(24)(01); for cobaltoblödite: 4.551(80)(210, 011); 4.269(50)(01); 3.795(18)(11); 3.339(43)(310); 3.29(100)(220, 021), 3.258(58)(11, 21), 2.644(21)(01), 2.296(22)(122). The crystal structures of both minerals were refined by single-crystal X-ray diffraction to R1 = 0.0459 (manganoblödite) and R1 = 0.0339 (cobaltoblödite).

Structures of calcium-rare-earth (Ca-RE) fluorocarbonate minerals from southwest China have been investigated using selected area electron diffraction (SAED) and high-resolution transmission electron microscopy (HRTEM). They are described as combinations of layers of bastnäsite-(Ce), CeFCO3 (B layers), and synchysite-(Ce), CeFCO3·CaCO3 (S layers) (Donnay and Donnay, 1953). We report here the discovery of six new, regular, mixed-layer structures in parisite-(Ce) observed using SAED and HRTEM. The symmetry, cell parameters, chemical formulae and stacking of the structural unit layers, etc., were determined for each. The regular, mixed-layer structures are formed by stacking unit layers of bastnäsite-(Ce) and synchysite-(Ce) in varying proportions along the c axis. HRTEM shows that there are different distribution modes for the Ce-F ion layers, the CO3 ion groups between the Ce-F ion layers and the CO3 ion groups between the Ce-F and Ca ion layers. The minerals may be regarded as polymorphs with the same chemical composition and the same spacing of the unit layers, but different arrangements of ion layers in the structural unit layers and different stacking sequences of the structural unit layers.

Kintoreite is a new lead iron phosphate mineral in the alunite-jarosite family, from Broken Hill, New South Wales, Australia. It is the phosphate analogue of segnitite and the iron analogue of plumbogummite. Kintoreite occurs as clusters and coatings of cream to yellowish green rhombohedral crystals up to 2 mm high and with the principal form {112}. The mineral also forms waxy, yellowish green globular crusts and hemispheres on other phosphate minerals. These associated species include pyromorphite, libethenite, rockbridgeite/dufrenite, apatite and goethite. Kintoreite formed during oxidation of primary ore rich in galena, in the presence of solutions with high P/(As + S) ratios. The mineral is named for the locality, the Kintore opencut, in which it is most common. A mineral closely resembling kintoreite in composition has also been found at several mines in Germany. Type material is preserved in the Museum of Victoria and the South Australian Museum.

Electron microprobe analysis showed a nearly complete spread of compositions across the P-dominant portion of the segnitite-kintoreite series. The selected type specimen has an empirical formula of Pb0.97(Fe2.95Zn0.13Cu0.02)Σ3.10[(PO4)1.30(AsO4)0.39(SO4)0.18(CO3)0.11]Σ1.98(OH)5.45·0.74H2O, calculated on the basis of 14 oxygens and with all Fe trivalent. The simplified formula is PbFe3(PO4)2(OH,H2O)6. Kintoreite crystals are translucent with a vitreous to adamantine lustre, with globules appearing waxy. The streak is pale yellowish green and the Mohs hardness is ∼ 4. Crystals show good cleavage on {001} and are brittle with a rough fracture. The calculated density is 4.34 g cm−3. Kintoreite crystals are uniaxial negative with RIs between 1.935 and 1.955 and show light yellowish green to medium yellow pleochroism.

The strongest lines in the X-ray powder pattern are (dobs, Iobs, hkl) 3.07(100) 113; 5.96(90)101; 3.67(60)110; 2.538(50)024; 2.257(50)107; 1.979(50)303; 1.831(40)220. The X-ray data were indexed on a hexagonal unit cell by analogy with beudantite, giving a = 7.325(1) Å, c = 16.900(3) Å, V = 785.3(5) Å3 and Z = 3. The probable space group is Rm, by analogy with beudantite and other members of the alunite-jarosite family. Powder X-ray diffraction data for several intermediate members suggest that the segnitite-kintoreite series may not represent ideal solid solution.

During the study of kintoreite, part of the type specimen of lusungite from Zaïre was obtained and shown to be goyazite. The IMA's Commission on New Minerals and Mineral Names has voted to discredit lusungite as a species, and has approved the renaming of the ‘lusungite’ group as the segnitite group. However, as relationships between crystal structure, order-disorder and solid solution in the Pb-rich minerals of the alunite-jarosite family are not well documented, the nomenclatural changes resulting from this study should be seen as interim only.

Non-metamict betafite, a pyrochlore-group mineral with general formula A2−mB2XO6Y(O,OH,F)1−n·pH2O and 2 Ti > Nb+Ta and U > 20% at the A site, has been found at Le Carcarelle, (Latium, Italy). It occurs within miarolitic cavities of a foid-bearing syenitic ejectum enclosed within the pyroclastic formation known as “ignimbrite C”, which belongs to the main effusive phase of the Vico volcanic complex. The host rock is composed of K-feldspar, biotite, augitic clinopyroxene, magnetite and minor sodalite. Electron microprobe analyses gave the following crystal-chemical formula: (Ca1.24Na0.17U0.49REE0.03)Σ=1.93 (Ti1.05Nb0.76Zr0.14Fe0.04Ta0.01)Σ=2.00O6(O,OH). Compared with other occurrences reported in the mineralogical literature, betafite from Le Carcarelle is extremely enriched in U and depleted in Th. Due to its young age of formation (∼150 k.y.), this betafite sample is highly crystalline, thus allowing structure refinement of unheated material. Betafite from Le Carcarelle is cubic Fdm, with a = 10.2637(13) Å, and V = 1081.21(35) Å3, and has a smaller A site (consistent with the higher U content), and a larger and more distorted B site (consistent with the higher Ti content) than calciobetafite from Campi Flegrei, Italy (Mazzi and Munno, 1983). Analysis of the atomic displacement parameters provides evidence for static disorder at the X site.

The (110) surfaces of arsenopyrite (FeAsS) and enargite (Cu3AsS4) have been modelled using a Density Functional Theory (DFT) plane-wave pseudopotential method (CASTEP) in order to better understand aspects of the geometric and electronic structures of these minerals, which have important implications for the release of arsenic in acid mine drainage environments. In this study, bulk calculations of these minerals have been conducted to give consistent geometries for surface models and to establish reference states for changes at the surfaces in these models. Surface structure experimental data for enargite were collected using low-energy electron diffraction which confirmed an unreconstructed 161 (110) lattice, with a and b values of 6.38±0.3 Å and 9.92±0.5 Å, respectively. Surface calculations demonstrate geometric and electronic relaxation of both arsenopyrite and enargite (110) surfaces. Changes in atomic positions, interatomic distances, Mulliken charges and electronic configurations are reported. Enargite has a surface energy of ∼0.02 eV/Å2 compared with arsenopyrite which has a surface energy of ∼0.11 eV/Å2, indicating that the enargite (110) surface is more energetically stable than that of arsenopyrite. The most stable surfaces are those which relax to restore the surface coordination and partial charge balance. For both minerals this is achieved by the formation of covalent bonds. Arsenic has the most positive Mulliken charge of all the surface atoms and is, therefore, predicted to be the most reactive atom at the arsenopyrite and enargite (110) surfaces. This implies that, according to these calculations, arsenic is most likely to react with oxidative species such as O2 and H2O in environments such as those associated with acid mine drainage, potentially releasing oxides and acids of arsenic into the environment.

In this work we report on a complete crystal-chemical characterization of a near end-member riebeckite from Malawi, and use the available data to critically compare information obtained from different analytical methods. The sample occurs as well-formed and very large single crystals in pegmatitic rocks. Accurate site-populations were determined by combining single-crystal structure refinement and electron microprobe analysis (EMPA). The Fe3+/Fe2+ ratio was obtained from Mössbauer spectroscopy. Lithium was quantified by Laser Ablation Inductively Coupled Plasma Mass Spectroscopy (LA-ICP-MS).

Fourier-Transform Infrared (FTIR) spectra, collected both on powders and single crystals, are presented and discussed. FTIR spectra in the NIR region are also presented for the first time for this amphibole. The FTIR data are compatible with complete local ordering of A cations close to F, and complete Fe2+/Mg disorder at M(1,3). Polarized Raman-scattering data collected from single crystals confirm this conclusion. In addition, it was found that FTIR data collected on powders provide the best agreement with the site occupancies derived from chemical (EMPA and LA-ICP-MS) and crystal-chemical data, possibly because they do not depend on experimental issues such as orientation and polarization.

In July 1930 a fragment of this meteorite was received at the Government Laboratory, Perth, with a request from the sender that he should be informed of its nature and value. Realizing its scientific interest, one of the authors (E. S. S.) took advantage of a visit to Beneubbin (lat. 30° 48′ S., long. 117° 51′ E. ; about 150 miles NE. of Perth in the South-West Division) to interview the finder and suggest that the whole mass should be handed over to the Government Laboratory for examination, a specimen detached for the British Museum, and the main mass finally presented to the Western Australian Museum in Perth. This arrangement was agreed to.

Shimazakiite occurs as greyish white aggregates up to 3 mm in diameter. Two polytypes, shimazakiite-4M and shimazakiite-4O, have been identified, the former in nanometre-sized twin lamellae and the latter in micrometre-sized lamellae. Shimazakiite was discovered in an irregular vein in crystalline limestone near gehlenite-spurrite skarns at Fuka mine, Okayama Prefecture, Japan. Associated minerals include takedaite, sibirskite, olshanskyite, parasibirskite, nifontovite, calcite and an uncharacterized hydrous calcium borate. The mineral is biaxial (–), with the following refractive indices (at 589 nm): α = 1.586(2), β = 1.650(2), γ = 1.667(2) and 2Vcalc = 53º [shimazakiite-4M]; and α = 1.584(2), β = 1.648(2), γ = 1.670(2) and 2Vcalc = 54.88º [shimazakiite-4O]. Quantitative electronmicroprobe analyses (means of 28 and 25 determinations) gave the empirical formulae Ca2B1.92O4.76(OH)0.24 and Ca2B1.92O4.76(OH)0.24 for shimazakiite-4M and shimazakiite-4O, respectively. The crystal structure refinements: P21/c, a = 3.5485(12), b = 6.352(2), c = 19.254(6) Å , β = 92.393(13)°, V = 433.6(3) Å3 [for shimazakiite-4M]; and P212121, a = 3.55645(8), b = 6.35194(15), c = 19.2534(5) Å , V = 434.941(18) Å3[for shimazakiite-4O], converged into R1 indices of 0.1273 and 0.0142, respectively. The crystal structure of shimazakiite consists of a layer containing B2O5 units (two near-coplanar triangular corner-sharing BO3 groups) and 6- and 7-coordinate Ca atoms. Different sequences in the c direction of four layers are observed in the polytypes. The five strongest lines in the powder-diffraction pattern [listed as d in Å (I)(hkl)] are: 3.02(84)(022); 2.92(100)(10) 2.81(56)(104); 2.76(32)(113); 1.880(32)(11,12,126,118) [for shimazakiite-4M]; and 3.84(33)(014); 3.02(42)(022); 2.86(100)(104); 2.79(29)(113); 1.903(44)(126,118) [for shimazakiite-4O].

The chain of Late Subhercynian granitoids can be traced along the Srednogorie-Istranca-Pontid belt. The Demirköy pluton outcrops in the Istranca segment. The rocks of the pluton range from diorite through grandiorite to perthite granite with granodiorite predominating. On the basis of 25 chemical analyses, the intrusive setting of the granitoid has been investigated. Calcic to calc-alkaline and peraluminous to metaluminous character indicate a subduction-related origin. Low to moderate concentrations of the large-ion lithophiles (LIL), high field strength elements (HFS), light rare earth elements (La, Ce) and low HFS/LIL ratios indicate a mantle-derived magma with subduction-related enrichment. Trace element discrimination diagrams such as Rb/Zr-Nb, Rb/Zr-Y, Rb-SiO2 and Rb-(Nb + Y), are particularly indicative of normal arc-setting for the samples.

The Demirköy granitoid is chemically indistinguishable from the Upper Cretaceous granitoids of Strednogorie to the west and the Pontids to the east. This extensive maximum of plutonism can be considered as a time marker in the northern Tethys subduction system.

A number of geochemical plots are suggested by means of which metamorphosed quartzofeldspathic sedimentary rocks (meta-sandstones, meta-arkoses, quartzites, etc.) can be distinguished from their metamorphosed igneous mineralogical analogues (metarhyolite, meta-granite, etc.). These plots are applied to an enigmatic, partly mylonitized, quartzofeldspathic series, the Delaney Dome Formation, which is involved in a major thrust zone. The formation is shown to be igneous and is probably a meta-rhyolite. The plots proposed should have general application in identifying the precursors of both high-grade and low-grade quartzofeldspathic rocks of unknown origin.