Thin apple-green crusts on several specimens from Hingston Down Consols mine, Calstock, Cornwall, proved to consist of a new mineral, alone or intimately mixed with pharmacosiderite (which it closely resembles), or with an unidentified mineral of the alunite-beudantite group having a 7·04, c 16·6 Å, or with both. The new mineral, for which the name arthurite is proposed, for Sir Arthur Russell and Mr. Arthur W. G. Kingsbury, who independently collected material and suggested that it might be new, gives X-ray powder photographs with their three strongest lines at 4·28, 4·81, and 6·97 Å ; the photographs could be satisfactorily indexed on a monoclinic unit-cell with a 10·09, b 9·62, c 5·55o Å (all ± 0·01 Å), β 92·2° ± 0·2° and containing [Cu2Fe4(AsO4)3(OH)7.6H2O]. Chemical analysis on 1·1 mg gave: CuO 16·8, Fe2O3 32·4, As2O5 34·3, H2O 16·5 % (reduced to 100 % after deduction of 27·4 % quartz) ; sp. gr. 3·2 (calc. 3·07). Under the microscope arthurite is pale olive-green and very finely granular; n 1·78, birefringence low to moderate.

A new cobalt-iron mineral of composition close to CoFe has been found associated with awaruite (Ni3Fe) in the Red Hills serpentinites of the Wairau Valley, South Island, New Zealand. The name wairauite, after this locality, is proposed. Electron-probe analysis of the small grains so far discovered has shown that the nickel content is very low (0·5 ± 0·1 wt. %). Wairauite is optically very similar to awaruite but a tendency to form euhedral grains (octahedron and cube) and a slightly lower refiectivity are an aid to identification. No X-ray data are available at present, but it is suggested that the structure may correspond to that of the ordered CsCl type of the synthetic alloy of the same composition.

The lattice parameters of twenty calcium-bearing anorthoclases and potassium-bearing acid plagioclases have been determined by powder methods. The 201̄ spacing of the homogenized anorthoclases gives a good measure of the Or content irrespective of the amount of calcium present ; this spacing combined with α* (010): (001) may be used to determine the ternary composition of homogeneous feldspars more potassic than approximately Or15. The lattice parameters of feldspars less potassic than Or15 cannot be used for the determination of their ternary composition.

Tremolite and actinolite form rhythmic overgrowths over hornblende in appinites associated with pipes of explosion-breccia and represent growth under alternating physical conditions arising from increase of gas pressure and its explosive release. The number of overgrowths present in each appinite mass is related to the number of explosions required to breach the particular structural trap that hindered the upward movement of the volatiles.

Experiments have been conducted to determine the optimum conditions for the hydrothermal growth of ruby. The solubility of corundmn in various solutions has been determined and compared with the respective growth obtained in each medium. The discrepancy between solubility and effective growth is discussed.

The need for a high degree of chemical purity has resulted in the development of platinum liners for the autoclaves. The effects of impurities and of other factors on the growth of ruby crystals have been studied. The crystal habit of hydrothermal ruby is also described.

Redetermination of the unit-cell dimensions as a 10·406 Å, b 10·813 Å, c 8·926 Å, α 104° 56′, β 96° 52′, γ 125° 19′, has shown that the hitherto accepted unit-cell of Gossner and Spielberger (1929) is in error. The morphological unit-cell of Palache (1933) is related to the new unit-cell, to a high degree of precision, by the transformation Palache → Kelsey-MeKie [001∕01½½∕½00]. Fifteen published and two new analyses indicate that the unit-cell contents approach Na+4 [Fe2-10- Ti4+2] Si4+12O2-40. The common twin axis constitutes the diad of a pseudo-monoclinic cell, the (010) plane of which is dimensionally closely related to that of pyroxenes and amphiboles. It is suggested that the aenigmatite structure is based on silicate chains of pyroxene type cross-linked by Ti4+ and Fe2+ on distinct 6-fold sites, with Na+ on sites of higher coordination number. Aenigmatite from volcanic in contrast to that from plutonic parageneses is often submicroscopically twinned and may be inverted from a high temperature monoclinic polymorph.

The ordinary quartz wedge is adapted for use by the colour blind by the superposition of a two-colour isotropic auxiliary plate. A birefringent plate of suitable retardation may also be used in conjunction with the same auxiliary plate to replace the standard sensitive tint plate. In both cases, diagnostic colour changes are replaced by relative intensity changes.

The linkage of Archimedean polyhedra has been studied to provide trial models for tetrahedrally-coordinated structures such as zeolites. The cuboctahedron, rhombicuboctahedron, snub cube, icosidodecahedron, rhombicosidodecahedron, and snub dodecahedron cannot be used because four edges meet at a corner. The truncated tetrahedron, truncated cube, and truncated dodecahedron have triangular faces, and are unlikely to occur in silicate frameworks because of the instability of 3-rings. The truncated icosahedron and truncated icosidodecahedron have fivefold symmetry and cannot form four-connected frameworks with lattice symmetry. The truncated octahedron and truncated cuboctahedron may be linked either directly, in combination, or with square, hexagonal, or octagonal prisms. There are nine structures of which four are represented by sodalite, Linde A, faujasite, and ZK-5. The other five have the following properties: truncated octahedra linked H(H-S), Fd3m, a 17·5 Å; truncated octahedra linked H′(H-S) and (H-H), P63/mmc, a 12·4, c 20·5 Å; truncated octahedra linked H′(H-S) and (H-H), P63/mmc, a 17·5, c 28·5 Å; truncated cuboctahedra linked O′(S-S), Im3m, a 15·1 Å; truncated octahedra linked to truncated cuboctahedra H′(S-S), Fm3m, a 31·1 Å. The first symbol specifies the linking unit (H hexagon, H′ hexagonal prism, O′ octagonal prism) while the symbols in brackets specify the type of faces opposing across the contact (S square).

The theoretical basis for the stability of binary and quasi-binary solutions is discussed with special emphasis on miscibility relations. Solutions of the distribution equations are presented for the case of two and three coexisting regular solutions and this model is used to illustrate the energetics of miscibility relations. The same principles are then extended to give a qualitative interpretation to sequences of mineral assemblages consisting of pyroxenes, amphiboles, micas, and feldspars. A formulation is presented for the intrinsic stability of a solution, which depends on the presence or absence of an excess free energy of mixing, and the extrinsic stability of a solution, which depends on the standard free energy of the component end members.

The transformation of groutite (α-MnOOH) by heating has been studied at 300° C in air, by single-crystal and powder X-ray methods. At this temperature groutite transforms topotactically into pyrolusite (MnO2), the a, b, and c axes of groutite becoming respectively the a, b, and c axes of pyrolusite (in pyrolusite b = a). At various stages of the transformation other weak and diffuse spots were observed on X-ray oscillation photographs, which could not be ascribed to pyrolusite. Some of these extra spots fit well to an α-Mn2O3 structure (isostructural with hematite), with c 14·3 and a 4·9 Å; the other few spots could not be identified.

The transformation of α-MnOOH into MnO2 is explained by a homogeneous mechanism, with migration of protons and electrons to the crystal surface. A detailed interpretation of this mechanism is presented on the basis of the close-packing characteristics of these two structures.

An occurrence of stilpnomelane in calcite veins cutting an Ordovician sedimentary iron-ore body has yielded good material for optical and chemical study; data for both ferro- and ferri-stilpnomelane are presented. Associated pyrrhotine, ilmenite, and apatite are briefly described, and a chemical analysis of the apatite shows it to be an extremely pure fluorapatite.

A new universal stage extinction technique is presented. It is based on the special case of coplanar and conical extinction curves where an optic axis lies on the wave-normal locus. It may be used in favourable circumstances to determine 2V when neither optic axis is directly observable.