Currently, there are two contrasting models concerning the state in which rhyolite lava reaches the surface: (1) the permeable foam model, where the lava reaches the surface as an expanded foam and collapses into dense obsidian during lava flow; and (2) the traditional model, where decreased load pressure leads to vesiculation of the upper surface of the dome.In order to test these models, textural parameters were documented on outcrops of the Ben Lomond rhyolite dome, which has two flow lobes comprising an eroded dome (c. 100 ka old) cut by the Whangamata Fault. A composite stratigraphy of the upper 60 m of the flow, assembled from three sections (roadcut, airstrip and fault scarp), comprises, irom the flow top: finely vesicular pumice, black aphyric obsidian, and a spherulitic transition zone above a central crystalline rhyolite core. An explosion breccia occurs as a pod at a depth of c. 35 m and cross-cuts the upper two units us an inverted cone-shaped deposit-an infilled explosion pit.Eight physical parameters (density and porosity, void aspect ratios, spherulite size and proportion, microlite size and abundance, and volatile contents) were measured throughout the exposed thickness of the flow. Results show that primary vesicles-those that deform pre-existing flow banding-are most numerous within the finely vesicular pumice layer but are suppressed c. 10-15 m below the surface. Vesicularity and volatile contents are anomalously high in explosion breccia fabrics. Furthermore, microlites are ubiquitous in the flow but absent in the explosion breccia. Incomplete degassing of the ascending magma took place, resulting in volatiles being distributed within the flow thickness during lava emplacement. The lava flow carapace revesiculated during extrusion (primary vesiculation), whereas secondary vesiculation (spherulites and lithophysae) was associated with groundmass crystallisation of the flow centre, causing localised volatile enrichments responsible ior the formation of explosion breccias. An emplacement model, formulated for the Ben Lomond dome, shows that

This study examines the depletion of ferromagnesian silicate minerals from a sequence of thin, distal, mainly rhyolitic tephra layers of Holocene age preserved in an acid peat bog (Kopouatai), North Island, New Zealand. The rate of such depletion has been fast, as indicated by the complete loss of biotite from one tephra layer (Kaharoa Tephra), in which it is normally dominant, in only ca. 770 yr. Chemical dissolution is advocated as the likely cause for the depletion, with amphiboles and other mineral grains commonly showing etch pits, microcaves, and other characteristic surface solution features. Theoretical thermodynamic and kinetic models Icwrfi1Ishow a marked increase in the rate of dissolution of all ferromagnesian minerals under conditions of low pH (< 4), but that where silica concentrations in solution are high the relative proportions of minerals remaining are unaffected. However, where concentrations of dissolved silica are low, as in most bog environments, the relative proportions of ferromagnesian minerals are affected as well as absolute amounts being decreased. Amphiboles are depleted relative to pyroxenes, consistent with kinetic studies.The results show that the identification and correlation of tephras on the basis of relative abundances of ferromagnesian minerals alone may be unreliable, and emphasise the need to use multiple criteria in such studies.Quaternary kleme

No abstract

Use of the elemental geochemistry of carbonates for characterizing and distinguishing between temperate and tropical carbonate facies is in its infancy. In particular, apart from several Tasmanianexamples, few elementaldata exist for temperate carbonates.

A B S T R A C T:The kinetics of clay formation in buried paleosols developed from late Quaternary rhyolitic tephra layers near Rotorua, New Zealand, can be described in terms of a combination of parabolic and linear kinetics, reflecting the hydration of glass, and the formation of clay minerals, respectively. Such a model is consistent with the formation of clay minerals showing an Arrhenian temperature dependence and suggests, on the basis of calculated activation energies, that the process of formation of Al-rich allophane (imogolite) is diffusion controlled, whereas the rate of formation of Si-rich allophane is controlled by the chemical processes at the site of reaction.The weathering behaviour of glass in the natural environment has been of interest to archaeologists because of their concern with dating obsidian artefacts and, in an attempt to define the kinetics of the process, short-term dissolution experiments undertaken in laboratories by geochemists have been useful. Our study has been concerned with a more advanced form of weathering: the formation of clay minerals from volcanic glass. The starting materials for our investigation were glass shards and derived clay minerals in buried paleosols developed on rhyolitic tephra deposits of known age and time for weathering (Fig. 1). From the proportions of clay minerals and glass, kinetic models have been developed to help explain the occurrence and neoformation of allophane in tephra deposits.Laboratory investigations of the weathering of glass have consistently shown that, at least in the early stages, cations are released into solution, and the glass hydrates according to socalled parabolic kinetics (e.g. White & Claassen, 1980). This can be represented:( 1) where C is the concentration of some species in solution, or the thickness of a hydrated layer, or some other appropriate indicator of the progress of the process, t is the time, and kp the rate constant. For rather longer periods of reaction (weeks rather than days), while the release of Na and K continue to display parabolic kinetics, the release of Si and AI are observed to display "linear kinetics" (e.g. White, 1983). "Linear" is probably an unfortunate term, because, as Lerman (1979) points out, such a mechanism would lead after a long time to either total dissolution of the solid, or to impossibly high concentrations in solution. In fact, linear kinetics are probably approximations to first-order reactions:or its more usual integrated form 9 1990 The Mineralogical Society