The crystalch emistry and crystal structure of naturalky anite crystals from the Eclogitic Micaschists Complex of the Sesia-Lanzo Zone, Western Italian Alps, have been investigated by means of optical microscopy, wavelength dispersive X-ray microanalysis, and single-crystal X-ray diffraction. The association of kyanite + garnet + phengitic-mica + chloritoid suggests that the eclogite-facies stages occurred at P ≤ 2.1 GPa and T ≤ 650ºC. Kyanite grains are large (cm-sized) porphyroblasts grown dynamically during one of the deformational events related to the subduction of the Austroalpine continentalcr ust. Under the polarizing microscope, kyanite grains show almost homogeneous cores, whereas rims are sometimes symplectitic aggregates of quartz and kyanite, confirming at least two stages of growth most likely related to the multistage deformational history of these rocks. Chemical analysis shows that Fe3+ is the major substituting cation for Al3+, ranging between 0.038 and 0.067 a.p.f.u.

The single-crystal X-ray diffraction investigation of the kyanites shows severely textured patterns on the (h0l)*-plane. Such evidence is not observed in the unwarped diffraction patterns on (0kl)* and (hk0)*. The most significant difference between the structuralp arameters refined in this study, with respect to those of previously published unstrained gem-quality crystals, concerns the displacement parameters. The anisotropic displacement ellipsoids of all the atomic sites are significantly larger than those previously described, and systematically oriented with the largest elliptical section almost perpendicular to [010]. The larger ellipsoids in the kyanite crystal investigated here reflect the displacement of the centre of gravity of the electron distribution, rather than an anomalous atomic thermal motion. The magnitude and orientation of the displacement parameters and the textured/strained diffraction pattern may be the result of two combined effects: (1) that the kyanite crystals are actually composed of several blocks; (2) the crystals are affected by a pervasive residual strain, as a result of tectonometamorphic plastic deformations and re-crystallization.

The review focuses on the evolution of five contiguous peralkaline salic complexes in the south-central Kenya Rift Valley, stressing new developments of general significance to peralkaline magmatism. The complexes have evolved dominantly by combinations of fractional crystallization and magma mixing; volatile-melt interactions, remobilization of plutonic rocks and crystal mushes, and carbonate-silicate liquid immiscibility have been additional petrogenetic processes. Geochemical and experimental studies have shown that pantelleritic magmas can be generated by fractional crystallization of trachyte and high-silica rhyolite. Melts of comenditic composition were also formed by fractionation of trachyte but also locally by partial meltingof syenites. Studies of apparent partition coefficients have provided some of the first data on element distribution between phenocrysts and peralkaline silicic melts. Compositional zonation has been ubiquitous in the complexes, probably a result of the very low viscosity of the magmas.

Understanding the deformation mechanisms that may operate in pyrite (FeS2) across a range of P-Tconditions is important in deciphering the history of deformed ore deposits. Pyrite has frequently been considered a hard mineral, which deforms by cataclastic flow or diffusive processes, if at all, at temperatures <425°C. However, utilizing SEM-based orientation-contrast (OC) imaging and electron-backscatter diffraction (EBSD) techniques, plastic deformation can now be readily identified within pyrite grains. In this study, a series of pyrite-rich polymetallic ore deposits, deformed at low temperature metamorphic conditions (∼200—420°C), have been investigated. Results indicate that pyrite grains in all the ore deposits preserve internal lattice ‘distortion’ or ‘bending’ and therefore plastic deformation mechanisms have operated. Many pyrite grains in the ore deposits also contain low-angle (∼2°) sub-grain boundaries or ‘dislocation walls', indicating that both dislocation glide and creep have been the dominant deformation mechanisms at peak metamorphic conditions within the pyrite grains. These results suggest that the brittle-ductile transition in pyrite occurs at temperatures potentially as low as ∼200°C, far lower than implied from previous studies or the current pyrite deformation-mechanism map.

Magmas undergoing shear are prime examples of flows that involve the transport of solids and gases by a separate (silicate melt) carrier phase. Such flows are called multiphase, and have attracted much attention due to their important range of engineering applications. Where the volume fraction of the dispersed phase (crystals) is large, the influence of particles on the fluid motion becomes significant and must be taken into account in any explanation of the bulk behaviour of the mixture. For congested magma deforming well in excess of the dilute limit (particle concentrations >40% by volume), sudden changes in the effective or relative viscosity can be expected. The picture is complicated further by the fact that the melt phase is temperature- and shear-rate-dependent. In the absence of a constitutive law for the flow of congested magma under an applied force, it is far from clear which of the many hundreds of empirical formulae devised to predict the rheology of suspensions as the particle fraction increases with time are best suited. Some of the more commonly used expressions in geology and engineering are reviewed with an aim to home in on those variables key to an improved understanding of magma rheology. These include a temperature, compositional and shear-rate dependency of viscosity of the melt phase with the shear-rate dependency of the crystal (particle) packing arrangement. Building on previous formulations, a new expression for the effective (relative) viscosity of magma is proposed that gives users the option to define a packing fraction range as a function of shear stress. Comparison is drawn between processes (segregation, clustering, jamming), common in industrial slurries, and structures seen preserved in igneous rocks. An equivalence is made such that congested magma, viewed in purely mechanical terms as a high-temperature slurry, is an inherently nonequilibrium material where flow at large Pe´clet numbers may result in shear thinning and spontaneous development of layering.

Two crystals from a sample of titanite from the Heftetjern granitic pegmatite, Tørdal, southern Norway, were extracted for structure analysis and shown to have triclinic symmetry. Unit-cell parameters are as follows: a = 7.0696(4) Å, b = 8.7167(5) Å, c = 6.5695(3) Å, α = 89.7372(11)°, β = 113.7607(10)°, γ = 90.2929(13)°, V = 370.52(6) Å3 for one crystal and a = 7.0612(5) Å, b = 8.7102(6) Å, c = 6.5628(4) Å, α = 89.7804(16)°, β = 113.7713(13)°, γ = 90.2502(16)°, V = 369.39(7) Å3 for the other. The interaxial angles α and γ deviate from the value of 90° required for monoclinic symmetry by ~200–250 standard deviations. The single-crystal X-ray intensities were averaged in both monoclinic and triclinic Laue symmetries, giving R(merge) values of ~14% and ~1.3% respectively. For both crystals, more than 50 reflections with I > 3σI violated the criterion for the presence of the a-glide required for monoclinic A2/a symmetry. Both crystals were refined in the space group A with Z = 4, and final R1 indices are 4.4% and 4.7% (wR2 = 8.4 and 8.9%) respectively. The composition of one crystal was determined by electron microprobe analysis: Ca[Ti0.623Ta0.105Nb0.018Al0.137Fe0.0463+Sn0.0834+]Σ=1.012(SiO4)O. The characteristic corner-sharing [MO5] chains of identical octahedra observed in monoclinic titanite become chains of alternating M(1) and M(2) octahedra of different size, with the stronger X-ray scattering constituents concentrated at the M(2) site. Short-range bond-valence considerations suggest that the M cations will order as Al—O—Ta in adjacent octahedra, and when present in sufficient amounts, will couple along the chain to break long-range monoclinic symmetry.

A natural quartz, annealed first at 800°C and then irradiated with a 3.5 MeV electron beam, has been investigated by single-crystal electron paramagnetic resonance (EPR) spectroscopy at X- and W-band frequencies from 110 K to 298 K. The W-band EPR spectra allow better separation of two previously reported radiation-induced defects (D and E) and improved determinations of their spin Hamiltonian parameters. These defects have similar g tensors with the gmax axes approximately along an O—O pair and gmin axes perpendicular to the short Si—O bonds, but different 27Al hyperfine structures. Centre E is also characterized by a 29Si hyperfine structure (A/geβe = ~0.4 mT). These spin Hamiltonian parameters, together with results from density functional theory (DFT) calculations, suggest centre E to be a new variant of peroxy radicals in quartz, whereas a peroxy radical model for centre D remains tentative. Thermal stabilities and decay kinetics of centres D and E have been investigated by use of isochronal and isothermal annealing experiments on a neutron-irradiated quartz and six smoky quartz crystals in druses from a U deposit.

Perovskite compositions are used to investigate the relationship between the minor components (i.e. LREE, Fe3+ and Nb) and the oxygen fugacity (fo2) of perovskite in four different kimberlite lithofacies from the Dutoitspan pipe, Kimberley, South Africa, which range from diamondiferous to barren. The perovskite textures and chemical variations provide insight into magmatic and eruptive processes. Some crystals display cores with rims separated by a sharp boundary. The cores contain larger Na and LREE contents relative to the rims, which show a large increase in Fe3+ and Al. The mid-grade and barren kimberlites have bi-modal cores, reflected in the mineral chemistry, signifying multiple batches of magma and magma mixing. The fo2 of the magma is determined by an Fe-Nb oxygen barometer. The most diamondiferous kimberlite has the greatest Fe3+ content and highest fo2 (NNO –3.6 to –1.1). The kimberlite containing large diamonds has the smallest Fe3+ content and lowest fo2 (NNO –5.2 to –3.0). The barren and mid-grade kimberlites display a wide range of fo2,(NNO –5.3 to –1.5) as a result of perovskites forming in different melts and subsequently mixing together. Chemical and petrological evidence suggests that the volatile content, degassing, decompression and rate of crystallization can influence the rate at which the magma is erupted. One possibility is that the most oxidized magma, containing the highest volatile content, is therefore erupted much more rapidly, preserving diamond as a consequence.

Nyholmite, Cd3Zn2(AsO3OH)2(AsO4)2·4H2O, from the Block 14 Opencut, Broken Hill, New South Wales, Australia, is a new Cd-Zn arsenate species, isostructural with the minerals of the hureaulite group. The mineral occurs in a quartz-garnet-arsenopyrite matrix as white globules, tufted aggregates of fibrous crystals and radiating hemispheres of thin, colourless, bladed crystals. Associated minerals are goldquarryite, lavendulan-sampleite, scorodite-strengite and gypsum. Individual crystals are up to 0.2 mm in length and 0.05 mm across. The mineral is transparent to translucent with a vitreous lustre. It is brittle with an uneven fracture and a white streak. The Mohs hardness is 3–3.5 and the calculated density is 4.23 g cm–3 for the empirical formula. Electron microprobe analyses yielded CdO 34.58, ZnO 9.72, MnO 3.59, CuO 3.39, Al2O3 0.20, CaO 0.16, PbO 0.37, As2O5 34.55, P2O5 6.29 totalling 92.85 wt.%. The empirical formula, based on 20 oxygen atoms, is Ca0.03Pb0.02 Cd2.80Al0.04Zn1.24-Cu0.44Mn0.53[(AsO4)3.13(PO4)0.92]Σ4.05H1.91·3.79H2O. Nyholmite is monoclinic, C2/c, a = 18.062(4) Å, b = 9.341(2) Å, c = 9.844(2) Å, β = 96.17(3)°, V = 1651.2(6) Å3 (single-crystal data, at 123 K). The six strongest lines in the X-ray powder diffraction pattern are [d(Å),I,(hkl)]: 8.985,30,(200); 8.283,85,(110); 6.169,25,(111); 4.878,25,(002); 3.234,100,(2, 420); 3.079,65,(222, 511); 2.976’45’(113). The crystal structure was solved by Patterson methods and refined using 2045 observed reflections to R1(F) = 3.73%. The structure is characterized by a kinked, five-membered chain of edge-sharing Mφ6 (φ = unspecified anion) octahedra, or pentamer, that extends in the a direction. The pentamers link by sharing corners to form a sheet in the (001) plane. Pentamers are also linked, via corner-sharing, by (As,P)O4 groups forming thick slabs in the (001) plane. The slabs link in the c direction by cornersharing between octahedra and tetrahedra to form a dense heteropolyhedral framework. Moderate to weak hydrogen-bonding provides additional linkage between the slabs.

The crystal structure of stephanite fromthe type locality, Freiberg District, Saxony, Germany, was refined in the space group Cmc21, up to a final R index of 0.0427. Unit-cell parameters are: a 7.8329(6) Å, b 12.458(1) Å, c 8.5272(7) Å, V 832.1(1) Å3; Z = 4. The previously reported structural model is confirmed, but a higher-precision refinement was achieved herein by the introduction of thirdorder non-harmonic Gram-Charlier tensors for one Ag atom. In the structure of stephanite, Sb forms isolated SbS3 pyramids, which typically occur in sulphosalts, and Ag occupies sites with coordination ranging fromtriangular to almost tetrahedral. Both the Sb–S and Ag–S bond distances closely match the values commonly observed in the structures of other Ag sulphosalts and sulphides.

The use of non-harmonic parameters for Ag allowed a better description of the electron density related to Ag, which is usually difficult to refine in good ionic conductors. A careful analysis of the energy barriers between the Ag sites defines preferred ion-diffusion pathways within the crystal structure of stephanite. The diffusion of Ag ions occurs preferentially along the sites Ag1 and Ag2, giving rise to two-dimensional nets of Ag atoms in which the ion conduction probably takes place.

Recent advances in ion microprobe instrumentation and techniques have enabled the mapping of C isotope ratios across the whole of a polished plate of a natural diamond from Guaniamo, Venezuela. The resultant map of C isotope variation closely matches the cathodoluminescence image of the growth structure of the diamond and, therefore, indicates an extremely limited scale of diffusion of C atoms sincethetimeof diamond formation. This result is compatible with thelimite d mobility of N atoms shown by theIaAB aggregation stateof thediamond. Inclusions in thediamond aree clogitic, in common with many Guaniamo diamonds with temperatures of formation of around 1200ºC. At such temperature the IaAB aggregation state indicates a mantle residence time on the order of 1 Ga. Such temperatures of formation and mantle residence times are common to many natural diamonds; thus the extremely limited diffusion of C isotopes shown by the mapping indicates that many diamonds will retain the C isotope compositions of their initial formation.

Alumoåkermanite, (Ca,Na)2(Al,Mg,Fe2+)(Si2O7), is a new mineral member of the melilite group from the active carbonatite-nephelinite-phonolite volcano Oldoinyo Lengai, Tanzania. The mineral occurs as tabular phenocrysts and microphenocrysts in melilite-nephelinitic ashes and lapilli-tuffs. Alumoåkermanite is light brown in colour; it is transparent, with a vitreous lustre and the streak is white. Cleavages or partings are not observed. The mineral is brittle with an uneven fracture. The measured density is 2.96(2) g/cm3. The Mohs hardness is ~4.5–6. Alumoåkermanite is uniaxial (–) with ω = 1.635(1) and ε = 1.624–1.626(1). In a 30 mm thin section (+N), the mineral has a yellow to orange interference colour, straight extinction and positive elongation, and is nonpleochroic. The average chemical formula of the mineral derived from electron microprobe analyses is: (Ca1.48Na0.50Sr0.02 K0.01)(Si1.99Al0.01O7). Alumoåkermanite is tetragonal, space group P421m with a = 7.7661(4) Å, c = 5.0297(4) Å, V = 303.4(1) Å3 and Z = 2. The five strongest powder-diffraction lines [d in Å, (I/Io), hkl] are: 3.712, (13), (111); 3.075, (25), (201); 2.859, (100), (211); 2.456, (32), (311); 1.757, (19), (312). Single-crystal structure refinement (R1 = 0.018) revealed structure topology typical of the melilite-group minerals, i.e. tetrahedral [(Al,Mg)(Si2O7)] sheets interleaved with layers of (CaNa) cations. The name reflects the chemical composition of the mineral.

High-pressure-temperature (P-T) experiments were conducted in an attempt to determine the diffusion rates of C atoms in diamond, and the possibility of changes in the isotope compositions of diamond at high P-T in the Earth’s mantle. The starting material consisted of a polished plate of natural diamond (very largely 12C), which had been coated with 13C diamond by chemical-vapourdeposition to form a sharp interface between 12C and 13C diamond. Three experiments were performed at 1800, 2000 and 2300ºC, all at 7.7 GPa, for0.5 –20 h. Isotopic profiles obtained by ion microprobe before and after each experiment showed no evidence of relaxation of the sharp interface between 12C and 13C, and so diffusion must have been on a scale less than the ~32 nm depth resolution for this technique. Using 32 nm as the maximum length scale of diffusion across the interface, the maximum ln D (diffusion coefficient) values for the experiments were calculated to be in the range –38 to –42. Unlike previous experimental data, these results show that changes in the isotopic compositions of diamond on long time scales in the Earth’s upper mantle are unlikely. Furthermore, the results support empirical evidence from mapping of C isotope distributions in natural diamonds that C isotope compositions reflect diamond growth compositions.

The composite Eskdalemuir dyke was formed by the intrusion of rhyolitic magma into partly crystallized basaltic magma, the proportion of the silicic component increasing towards the dyke centre. The less silicic parts were effectively supercooled and crystallization was halted when the residual melt quenched to rhyolitic glass which constitutes up to 50% of the dyke's centre. Pyroxene compositions in the mixed magma rocks include augite, pigeonite and orthopyroxene of very variable composition. Individual crystals in the matrix as small as 200 μm can contain compositions occupying a considerable part of the Di-Hd-En-Fs quadrilateral. Pyroxene crystallization took place under extreme disequilibrium conditions, as shown by highly variable zonation patterns; apparently opposite zoning trends can occur in adjacent grains. Small-scale variations in melt composition were probably the dominant control over the compositional and textural complexity in the pyroxenes.

In the southern Aosta Valley (Italian Western Alps), several massive magnetite bodies occur within serpentinized ultramafic rocks belonging to the Mesozoic meta-ophiolite nappe. The ultramafic rocks consist of lherzolite with minor dunite bodies and show a high pressure metamorphic overprint. The results of a multi-methodological study, based on optical microscopy, electron microprobe analysis and single-crystal X-ray diffraction, are reported here in order to give new insights into (1) the mineralogy and crystal chemistry of spinels and silicates and (2) the genesis of the massive magnetite bodies. Chromium-rich relict cores inside the magnetite grains suggest a derivation from primary chromite concentrations. The major-element behaviour shows the presence of two chromite types: a Cr2O3-rich (Al2O3-poor) type and a Cr2O3-poor (Al2O3-rich) type. Magnetite ore deposits probably represent the product of transformation from a chromite proto-ore which formed in ultramafic rocks pertaining to an ophiolite suite. The transformation of chromite to magnetite occurred during multiple stages: the pre- metamorphic setting of the ultramafics and the petrographic evidence suggest that metasomatism started before the onset of the alpine metamorphism and was active during the early alpine, eclogite- facies metamorphic overprint related to a subduction process under high fluid activity.

The [AlO4/M+]0 (where M = H, Li, Na and K) defects in α-quartz have been investigated by ab initio calculations at the density functional theory (DFT) level, using the CRYSTAL06 code, 72-atom supercells, and all-electron basis sets. Our DFT calculations yielded substantially improved results than previous cluster calculations with minimal basis sets. For example, the [AlO4/M+(a<)]0 defects with M = H, Li and Na have been shown to be more stable than their [AlO4/M+(a>)]0 structural analogues (where a> and a< denote the location of the charge-compensating ion on the long-bond and short-bond side respectively), correctly predicting the common occurrence of paramagnetic [AlO4/M+(a>)]+ centres. In addition, the [AlO4/K+]0 defects have been investigated for the first time and are shown to be stable in quartz. Moreover, our calculations confirm previous suggestions that incorporation of the [AlO4/M+]0 defects results in significant structural relaxations that extend at least to the nearest Si atoms and give Li—O and Na—O bond distances in better agreement with the experimentally obtained values. The present theoretical results on the [AlO4/M+]0 defects provide a more complete picture for the coupled Al3+—M+ substitutions and hence new insights into crystal-chemical controls on the uptake of Al in quartz.

Arsenic is a very common by-product of the processing of Cu, Au and polymetallic ores worldwide, where the ore is roasted (calcined) to remove volatile elements. In southwest England, a diverse range of As-mineral species occur as efflorescent secondary mineral growths on historic calciner buildings. Gypsum occurs as abundant dendritic growths comprising either interlocking blades or tabular crystals. Ca-arsenate minerals are locally very abundant as white colloform masses. Positively identified Ca arsenates include pharmacolite, weilite and haidingerite. Other secondary minerals include arsenolite, scorodite, bukovskyite and an As-bearing potassium alum, together with a wide variety of unidentified minerals, including an Al-As-S phase and As-rich F-bearing phases. Gypsum contains As concentrations up to ~7 wt.%. Efflorescent growth at sites exposed to the prevailing weather systems is less abundant than at sheltered sites. This is interpreted as being due to ‘pressure washing’ of exposed sites by driving rain. Successive concentric growths of gypsum and Ca arsenate on masonry are interpreted as being the result of seasonal crystallization.

Understanding both current and historicalmining and mineralprocessing methods is criticalin the evaluation of the potential impact on the modern environment. In particular, due to the abundance of As-bearing minerals in a wide range of ore types, many buildings worldwide are potentially significantly contaminated with As even though few are directly related to As production or handling. Characterizing the secondary As mineralspecies present at mine and mineralprocessing sites is critical in understanding the potentialheal th risk these sites might pose.

Sector zoning in solid solutions of Tutton’s salts (NH4)2M(SO4)2.6H2O (M = Co, Ni, Zn, Mg) and nitrates of divalent metals Me(NO3)2 (Me = Sr, Pb, Ba) grown from low-T aqueous solutions under controlled conditions is studied. The effect of crystal composition and growth rate (supersaturation) on the intersector difference in concentrations of isomorphic components is examined experimentally and modelled theoretically based on the phenomenological approach to crystallization in a ‘solid solution—aqueous solution’ system. Calculations show a predominating contribution of the selective adsorption phenomena in the formation of a sector-zoned crystal with usually minor, but sometimes a major contribution of growth kinetics.

Mavlyanovite, ideally Mn5Si3, is a new mineral from a lamproite diatreme close to the upper reaches of the Koshmansay river, Chatkal ridge, Uzbekistan. It occurs together with unnamed manganese siliciphosphide and manganese silicicarbide minerals in round to ovoid segregations, up to 10 cm in diameter, in volcanic glass. Segregations of hexagonal prismatic mavlyanovite up to 1–2 mm occur in interstices in the matrix and tiny inclusions (1–2 μm) of alabandite and khamrabaevite occur within mavlyanovite. It is opaque with a metallic lustre, has a dark-grey streak, is brittle with a conchoidal fracture and a near-perfect basal cleavage. VHN100 is 1029–1098 kg/mm2 (Mohs hardness ~7). In plane-polarized reflected light, mavlyanovite is a pale-brownish-grey against the accompanying unnamed manganese silicicarbide (white). Reflectance values and colour data are tabulated. Average results of 19 electronmicroprobe analyses give Mn70.84, Fe 6.12, Si 22.57, Ti 0.15, P 0.18, total 99.86 wt.% leading to an empirical formula of (Mn4.66Fe0.40)5.06(Si2.91Ti0.01P0.02)2.94 based on8 a.p.f.u. The calculated density is 6.06 g/cm3, (on the basis of the empirical formula and unit-cell parameters from the structure determination). Mavlyanovite is hexagonal (P63/mcm) with a 6.8971(7), c 4.8075(4) Å, V 198.05(3) Å3 and Z = 2. The structure has been determined and refined to R1 = 0.017, wR2 = 0.044, GoF = 1.16. Mavlyanovite is the naturally-occurring analogue of synthetic Mn5Si3 which is the parent aristotype structure of the Nowotny intermetallic phases studied extensively by the material-science community. It is also the Mn-dominant analogue of xifengite Fe5Si3. The mineral name honours Academician Gani Arifkhanovich Mavlyanov (1910–1988), for his contributions to the understanding of the geology of Uzbekistan.

The crystal structure of nechelyustovite, ideally Na4Ba2Mn1.5☐2.5Ti5Nb(Si2O7)4O4(OH)3F(H2O)6, a 5.447(1) Å, b 7.157(1) Å, c 47.259(9) Å, α 95.759(4)°, β 92.136(4)°, γ 89.978(4)°, V 1831.7(4) Å3, space group P, Z = 2, Dcalc. 3.041 g cm–3, from Lovozero alkaline massif, Kola Peninsula, Russia, has been solved and refined to R1 = 13.9% on the basis of 1745 unique reflections (Fo > 15σF). Electron microprobe analysis yielded the empirical formula (H20)6.01, Z = 2, calculated on the basis of 42 (O + F) a.p.f.u., H2O and OH are calculated from structure refinement (H2O = 6 p.f.u.; F + OH = 4 p.f.u.). The crystal structure of nechelyustovite is a combination of a TS (titanium silicate) block and an I (intermediate) block. The TS block consists of HOH sheets (H-heteropolyhedral, O-octahedral). The TS block exhibits linkage and stereochemistry typical for Group III (Ti = 3 a.p.f.u.) of Ti-disilicate minerals: two H sheets connect to the O sheet such that two (Si2O7) groups link to the trans edges of a Ti octahedron of the O sheet. There are two distinct TS blocks of the same topology, TS1 and TS2, that differ in the cations of the O sheet, [(Na1.5Mn1☐0.5)Ti] and [(Na2Mn0.5☐0.5)Ti] (4 a.p.f.u.) respectively. The TS1 and TS2 blocks have two different H sheets, H1,2 and H3,4, where (Si2O7) groups link to [5]- and [6]-coordinated (Ti,Nb) polyhedra respectively. There are three peripheral sites, AP(1—3), occupied mainly by Ba (less Sr and K) at 96, 86 and 26% and one peripheral site AP(4) occupied by Na at 50%. There are two I blocks: the I1 block is a layer of Ba atoms; the I2 block consists of H2O groups and AP(3) atoms. TS blocks alternate with I blocks or link through hydrogen bonds (as in epistolite). There is a sequence of four TS blocks and three I blocks per the c cell parameter: TS2 — I1 — TS1 — I2 — TS1 — I1 — TS2.