The reliability and utility of diffuse reflectance spectra are briefly but critically reviewed. The results of measurements of diffuse reflectance over the wavelength range 200 < λ < 2500 nm are reported for wüstite, magnetite, hematite, maghemite, ilmenite, ulvöspinel, and α-FeO · OH (goethite), β-FeO 7sd OH, γ-FeO · OH (lepidocrocite), and δ-FeO · OH. The spectra have been assigned by reference to simplified molecular-orbital energy-level diagrams derived from recent SCF-Xα calculations.

The specular reflectances reported in the Quantitative Data File (Henry, 1977) are related to the diffuse reflectance spectra in a rational way. Minerals that absorb strongly throughout the visible display little dispersion of specular reflectance, and their powders are dark (wüstite, magnetite, ilmenite, ulvöspinel); those that absorb much more strongly in the near ultraviolet than in the visible have specular reflectances that decrease monotonically from blue to red according to a simple dispersion relation derived by combining the Sellmeier dispersion and Fresnel reflexion equations; their powders are strongly coloured (hematite, maghemite, lepidocrocite, goethite) and their optical anisotropy is closely related to crystal structure.

From the microscopic examination of lightly etched, cleaved (111) surfaces of banded blue fluorite (Blue John, from Castleton, Derbyshire) it is concluded that the coloured lamellae are not necessarily or exclusively associated with lattice-line imperfections. Elemental analyses, by electron dispersive methods, of fresh cleavage (111) surfaces in the immediate vicinity of small inclusions (c. 10 µm diameter) possessing associated coloured haloes detected no appreciable concentrations of impurities. From the evidence available, it is suggested that the zones of blue colour consist of colloidal calcium resulting from radiation damage caused by the intermittent deposition of radioactive material on the surfaces of fluorite during crystal development. The dispersion of colloidal calcium produced is particularly stable as a consequence of the close correspondence between lattice spacings in calcium fluoride and in calcium metal.

Ore specimens from the abandoned copper mine at Seathwaite Tarn, Cumbria, were studied in polished section by reflected light microscopy. The mineral paragenesis is described and related to current views on Lake District ore genesis. A common origin with the neighbouring deposit at Coniston is proposed. Wittichenite was identified in the ore, and is the first authenticated occurrence in Great Britain.

Petrofabric analysis of five specimens of laminated gabbro from the Centre 3 igneous complex, Ardnamurchan, was carried out by measuring orientations of grains of plagioclase, olivine, and augite with a Universal-stage. Linear alignments of primary precipitate mineral grains within the laminar fabrics were revealed by plots of data for coexisting plagioclase and olivine or augite. The plagioclase grains involved have tabular forms elongated parallel to their a-axes (rather than their c-axes). The olivine grains have flattened ovoid forms and the augite grains have sub-ophitic forms developed about prismatic primocrysts. In all specimens plagioclase displayed alignment of long grain-axes directed approximately down the dip of the lamination, save for one case in which a weak oblique alignment was observed. These lineations were reproduced by accompanying olivine and augite. The alignments are tentatively ascribed to flowage of a viscous boundary zone to the magma body in which crystal growth was taking place.

Electron microprobe analyses and X-ray data are given for iron-rich carrollites, cobaltian violarites, and other members of the linnaeite group. The iron-rich carrollites (8.5 wt% Fe) occur as an unmixed phase in a copper linnaeite host from the Zambian Copperbelt. Primary and secondary cobaltian violarites are reported from Rhodesia and Australia respectively with the latter recording up to 19.2 wt% Co. A diagrammatic review is given of the extremes of composition reported within the group together with the more reliable cell sizes.

An olivine-titanomagnetite-apatite-clinopyroxene-mica-nepheline-feldspar assemblage occurs in late-stage vesicles in a small outcrop of olivine leucitite at Cosgrove, Victoria. The vesicles were formed by exsolution of volatiles at an early stage in the cooling history of the lava. Subsequently, a volatile-rich residual liquid filled cavities and fractures, giving rise to a coarse-grained pegmatoid rock type similar in over-all mineralogy to the vesicles. The volatiles facilitating crystallization in both the vesicles and the pegmatoid were probably enriched in F, CO2, and P. A number of geothermometers applied to the vesicle assemblage failed to agree on likely crystallization temperatures.

Intrusion of the Bee Mountain soda microsyenite plug in southwestern Brewster County, Texas established conditions of low grade contact metamorphism in the upper Cretaceous Boquillas and Pen Formation sedimentary rocks. It is unlikely that load pressure exceeded 330 bars or that temperatures exceeded 470 °C at the intrusive contact. Thermally unaltered parent materials predominantly contain ortho- and allochemical low-magnesium calcite, silt-sized quartz, calcium montmorillonite, kaolinite, and volcanic glass. Incipient metamorphism is marked by the appearance of xanthophyllite followed by prehnite formed at the expense of xanthophyllite and montmorillonite. Near the soda microsyenite contact prehnite has decomposed to yield grossular and epidote group minerals, and apophyllite has replaced calcite where fluorine-bearing solutions have reacted with the parent material. Textural modifications include the recrystallization of calcite, redistribution of some of the mineral phases, and decreased friability in the marly layers.

Illite crystallinity determinations on Palaeozoic pelitic rocks, whose stratigraphic range runs from Lower Cambrian to Westphalian, indicate that anchimetamorphism has affected both the Lower and Upper Palaeozoic sequences. Two metamorphic episodes are in evidence, with the earlier, Caledonian, being of slightly higher grade. The higher anchizone crystallinity values are recorded from the Fishguard area in which mineral assemblages of the prehnite-pumpellyite facies have recently been recognized in basic igneous rocks. The later metamorphic episode has affected rocks to the south of the Variscan front. Here crystallinity values are low anchizone but some straddle the boundary with the diagenetic state. The Pembroke coalfield lies in this southern area and has coals largely of anthracite rank with volatile matter contents of between 10.1 and 5% Grade of metamorphism as indicated by crystallinity and by coal rank data from the Pembroke coalfield shows anomalous results to that described from the main South Wales coalfield. A neo-formed 2M illite from the Variscan spaced cleavage is described with analytical and X-ray diffraction data.

Ejecta associated with the Xalapasco de la Joya maar 33 km northeast of San Luis Potosi, SLP, Mexico, contain a diverse assemblage of nodules of postulated lower-crust/upper-mantle origin. Spinel lherzolites constitute the most abundant group of nodules and display chemical evidence that crystallization or equilibration took place at temperatures and pressures consistent with upper mantle genesis. Estimated equilibration temperatures and pressures range from 825 to 1025 °C and 7 to 19 kb.

The third terrestrial occurrence of hibonite is reported from granulite-facies rocks in the Furua Granulite Complex in southern Tanzania. The mineral forms yellowish-brown lath-shaped crystals in a grossular-anorthite rock containing subordinate sphene (clino)zoisite, hercynite, apatite, ilmenite, and corundumilmenite intergrowths.

Electron-microprobe analyses indicate a generalized formula (Ca1−xREx)[(Al,Fe3+)12−2a+x(Ti,Si)(Ti,Si)a−x(Fe2+, Mg)a]O19, with RE = Ce + La + Nd, x = 0.2, and a = 0.8. Individual mineral analyses show a cation substitution of Ca + Ti + Fe3+ = RE + 2Al. Relatively high RE and Fe contents represent the main chemical differences with meteoritic hibonite. The hexagonal unit cell has a = 5.61 Å, c = 22.18, in good agreement with the other terrestrial hibonites.

Three compositional types of (clino)zoisite are distinguished: 1.8–3.1 wt% Fe2O3 (orthorhombic and monoclinic), 3.9–6.0 wt% Fe2O3 (monoclinic), and 5.8–7.9 wt% Fe2O3 with an average of 6.3 wt% RE2O3 (monoclinic).

Thermometric and barometric data for coexisting pyroxenes and garnet from adjacent rocks indicate granulite-facies equilibration conditions of 750 to 850 °C and 6 to 11 kb. During retrogression with increasing partial H2O pressures, hibonite reacted with plagioclase and garnet to form spinel, sphene, and RE-bearing clinozoisite. Corundum-ilmenite inter-growths probably resulted from the breakdown of an Fe-högbomite.

Goethites of well-crystallized appearance from a sedimentary ore deposit have been studied by Mössbauer spectroscopy and other methods following HCl corrosion, particle size fractionation, and annealing treatments.

The goethites contained an appreciable amount of Si, but were not significantly Al-substituted. The observed broad and asymmetric resonant lines of the Mössbauer spectra are considered to result from this Si which, interspersed in the goethite structure, would interrupt magnetic coupling and give rise to a reduced and inhomogeneous magnetic hyperfine field.

Zussmanite has been found at only one locality: the Laytonville Quarry, Mendocino County, California. There are, however, present at this locality, two separate, but apparently inter-related minerals that from the evidence of chemistry, and limited diffraction information, appear to be zussmanite-type species. Their relative structural similarities are demonstrated within two rocks from the quarry in which a manganese concentration gradient has allowed ferrous-iron-rich zussmanites to develop partly contiguous overgrowths of one or other of these two minerals, one of which is a new form that has a provisionally determined ideal formula of KAlMn3−5-8Si17O42(OH)14 and that is separated from ideal zussmanite compositions (of the form Si17O42(OH)14) by an immiscibility gap. The other ‘zussmanite-type’ mineral has a composition that closely resembles a manganese-rich form of minne-sotaite.

The first zussmanite-type species (ZU2) has been separated and not unambiguous diffraction information obtained of its cell dimensions and powder lines, which are similar to those of zussmanite but appear to have an 8 % smaller cell-base and only a two-layer repeat (as in some of the zussmanite polytypes, and as in the talc structure). It is therefore considered possible that ZU2 has an altered compatibility between the tetrahedral and octa-hedral sheet overlap, perhaps from 13 (as in zussmanite) to 12. Whilst zussmanite appears to be a blueschist-eclogite mineral, ZU2 occurs under conditions at the low-pressure side of the blueschist facies.

An intermediate between zussmanite and the manganoan ‘minnesotaite’ is found in one rock in which the rims of zussmanite have been leached of potassium. As minnesotaite is more of a range of compositions than a structure (there is mounting evidence that it is not a simple talc-analogue) a consideration of the Laytonville manganoan minnesotaite as a zussmanite mineral is not unreasonable.

The mineral deerite is restricted in occurrence to blueschist facies meta-ironstones. From the ideal formula of Si12O40(OH)10 there are a limited range of substitutions: Ti and V for Fe3+ and Mn on the Fe2+ site. In the present survey of deerite compositions from the majority of deerite localities the maximum substitutions have been 2Ti and 3Mn in the above formula. There is an at present unexplained anomaly in the totals of a number of the analyses indicating that there may be variations away from the ideal formula towards increased hydroxyl and decreased cation contents.

From differences between the mineral compatibilities of deerites from the higher-grade Franciscan exotic block blueschist localities and those from more conformable Alpine blueschists, in conjunction with previous work on the experimental stabilities of the low-pressure/low-temperature hydrous iron silicate minerals, it has been possible to map the form of the low-pressure deerite stability field involving reactions of greenalite or minnesotaite or grunerite with magnetite and quartz. Within the deerite stability field, further deerite-forming reactions involve the breakdown of riebeckite and the hydrous incompatibility of magnetite with quartz; both are very dependent on the activity of sodium. A low-pressure stability to 4 kb at 200 °C and 6 kb at 300 °C are estimated from these low-temperature breakdown reactions. This fits in well with the high-pressure deerite stability determinations of Langer et al. (1977).

The sodic amphiboles possess two independent chemical substitution series (Fe3+-Al and Fe2+-Mg) that combine to provide a ‘plane’ of compositions. Yet at no single T and P are compositions covering the whole plane stable: (i) pure riebeckite exists under low-P conditions but breaks down in normal blueschists to give deerite; (ii) ferro-glaucophane is in competition at all except the lowest blueschist temperatures with almandine garnet; (iii) magnesio-riebeckite is stable at high-T and low-P but within the blueschist facies is replaced by the alternative higher density aegirine-talc assemblage; and (iv) glaucophane is stable only at high-P.

At higher T and P than those of the blueschists, competition from NaCa pyroxenes, garnets, and deerite first erodes, and then removes, nearly all sodic amphibole compositions. At low-P the normal sodic amphibole-forming reactions from stilpnomelane and chlorite (in the presence of iron oxides, albite, etc.) produce an initial ‘rie-beckitic’ amphibole that subsequently becomes more glaucophanitic with increasing P. Under certain conditions, perhaps connected with hydrous fluid overpressures, these reactions become transposed such that crossitic compositions become replaced whilst ferroglaucophane to glancophane compositions remain stable.

The chemical, mineralogical, and textural changes involved in the weathering of basalt have been traced through various stages from fresh rock (which has a cation exchange capacity of 10 meq/100 g due to the presence of a swelling chlorite mineral) to reddened basaltic rubble consisting of interstratified montmorillonite-illite, hematite, and anatase. The cation exchange capacities of the rocks increase progressively with the formation of secondary clay from labradorite as Al, Fe, and Ti accumulate and Si, Mg, Ca, and Na are depleted—much of the K is retained in the secondary clay mineral. The weathering is considered to be contemporaneous with the formation of the Antrim bauxites but not so intense.

The wittichenite (BM 1977, 172) discovered in the course of a reflected light study of polished sections from the Seathwaite deposit, Cumbria (described earlier in this journal by Stanley and Criddle), is the first authenticated occurrence of the species in Great Britain; for this reason it was examined in detail and compared with a specimen (BM 1437) from Wittichen, the type locality. Chemical analyses, cell dimensions, visible-spectrum reflectances, quantitative colour values, and VHN's are provided. They are in agreement with previously published data for the species.

The mineral howieite has a stability range that extends down to pressures at, or even below, those normally considered to represent blueschist facies metamorphism. Its structure, lying somewhere between a chain and a sheet silicate, can accommodate almost complete iron for manganese substitution as well as the more extraordinary ferrous iron for ferric iron substitution, that is tolerated by changes in the hydroxyl and water content of the unit cell. The ideal formula is NaM12 Si12(O,OH)44, where M is principally manganese and iron but may include limited amounts of aluminium and magnesium. Four new localities for howieite have been discovered by the author and the howieites from the eight localities that have been analysed here help complete the compositional spectrum. The most curious crystallographic feature of howieite is that although the structure seems capable of considerable chemical variation, it is only those howieites of approximate NaMn3Fe8(AlMg)1Si12(O,OH)44 composition that are well crystallized, often growing to over 10 mm in length.

The composition of scarbroite, written in the form Al5(OH)13CO3·5H2O, suggests close crystal-chemical relations to gibbsite and to hydrotalcite and kindred minerals. Parametral relations a(S) ≃ 2b(G) and 3b(S) ≃ 5a(G), S = scarbroite, G = gibbsite, are obtained and are consistent with a hydroxy Al layer structure and interlayer carbonate anions and water molecules. Thermal transformations show two discrete hydrated phases with layer spacings near 8.5 and 7.65 Å and a third phase with initial spacing 6.65 Å which decreases progressively to 5.5 Å over the temperature range 100°–250 °C as decomposition proceeds. No clear break is observed between dehydroxylation and decarbonation.

The field relationships, petrography and mineralogy, including probe analyses of olivines, pyroxenes, and plagioclases, of the Karroo dolerite dike swarm of southern Malawi, are described. Rock analyses are given and define a typical tholeiitic trend of iron enrichment. The most magnesian rocks are thought to represent fairly picritic liquids. A number of authors have shown that there are significant geochemical differences between the basaltic rocks of the northern and southern parts of the Karroo outcrop in southern Africa, notably for elements such as Al, P, K, Ti, and Ba. The Malawi data presented here, however, show trends that are the reverse of those found hitherto, the Malawi dolerites being closely comparable with the rocks of the central Karroo basin of South Africa and Basutoland. The striking regional variation in the geochemistry of the Karroo basaltic rocks over southern Africa is independent of the tectonics or position vis-à-vis the central Karroo basin, and so probably reflects mantle heterogeneity.