Notes: Diatom ooze and diatomaceous mudstone overlie terrigenous mudstone beds at Leg 19 Deep Sea Drilling Project sites. The diatomaceous units are 300-725 m thick but most commonly are about 600 m. Diagenesis of diatom frustules follows a predictable series of physical and chemical changes that are related primarily to temperature (depth of burial and local geothermal gradient). During the first 300-400 m of burial frustules are fragmented and undergo mild dissolution. By 600 m dissolution of opal-A (biogenic silica) is widespread. Silica reprecipitates abundantly as inorganic opal-A between 600 and 700 m sub-bottom depth. Inorganic opal-A is rapidly transformed by crystal growth to opal-CT. The result is formation of silica cemented mudstone and porcelanite beds.A regional acoustic reflector (called the bottom-simulating reflector, or BSR) occurs near 600 m depth in the sections. This acoustic event marks the upper surface where silicification (cementation) is active. In Bering Sea deposits, opal-A is transformed to opal-CT at temperatures between 35° and 50°C. This temperature range corresponds to a sub-bottom depth of about 600 m and is the area where silicification is most active. Thus, the BSR represents an isothermal surface; the temperature it records is that required to transform opal-A to opal-CT. Deposition of at least 500 m of diatomaceous sediment was required before the temperature at the base of the diatomaceous section was appropriate (35°-50°C) for silica diagenesis to occur. Accordingly, silica diagenesis did not begin until Pleistocene time. Once silicification began, in response to sediment accumulation during the Quaternary, the diagenetic front (the BSR) moved upsection in pace with the upward migrating thermal boundary.X-ray diffractograms and SEM photographs show three silica phases, biogenic opal-A, inorganic opal-A’, and opal-CT. These have crystallite sizes of 11-16 A, 20-27 A, and 40-81 A, respectively, normal to 101. The d(101) reflection of opal-CT decreases with depth of burial at DSDP Site 192. This occurs by solid-state ordering and requires at least 700 m of burial.Most clinoptilolite in Leg 19 cores forms from the diagenesis of siliceous debris rather than from the alteration of volcanic debris as is commonly reported.

Notes: Three methods, (a)dessication, (b)gas extraction and (c)immiscible-liquid extraction, are described and evaluated for removing interstitial water from coarse-grained sedi- ments for semi-quantitative and quantitative chemical analyses. The dessication method is useful for chlorinity or chlorosity determinations and (with a correction of 5%) gives results probably accurate to within about ± 2O% Extraction of interstitial water with gas of high relative humidity gives results which are correct to within an error of ± l.O% The chlorosity of interstitial water extracted by forcing a high- viscosity epoxy plasticizer (immiscible with water) through the sediment falls within about ± 0.1% of the correct value. The immiscible-extraction method is therefore recommended for work requiring quantitative knowledge of interstitial water chemistry.

Notes: Continuum models are used to investigate the large-scale deformation associated with the subduction of aseismic ridges. Formulated in the horizontal plane using thin viscous sheet theory, these models measure the horizontal transmission of stress through the arc lithosphere accompanying ridge subduction. Modelling was used to compare the Tonga arc and Louisville ridge collision with the New Hebrides arc and d’Entrecasteaux ridge collision, which have disparate arc-ridge intersection speeds but otherwise similar characteristics. Models of both systems indicate that diffuse deformation (low values of the effective stress-strain exponent n) are required to explain the observed deformation. Deformation is somewhat insensitive to the vertically integrated strength of the arc (inversely proportional to the Argand number Ar), but indicates that the arc lithosphere is not extremely weak (Ar 〈 100). Low values of both Ar and n suggest that the thermal structure is typical of ’cold’ or ’normal’ arcs and that deformation is dominated by flow in the lower crust and mantle. In addition, low values of n (approaching Newtonian flow) may indicate that specific deformation mechanisms dictate deformation of the arc lithosphere. Possible mechanisms include low-stress, grain-size dependent creep, pyroxenite-controlled rheology and mechanisms associated with water weakening. Changes in the boundary conditions greatly affect deformation within island arcs. High rates of arc-ridge intersection speed (Tonga-Louisville system) yield arc-parallel tension and crustal thickening in the wake of ridge subduction. In contrast, low rates of arc-ridge intersection speed (New Hebrides-d’Entrecasteaux system) yield compressional deformation directly arcward of the collision zone and transverse strike-slip faulting adjacent to the region of compressional deformation. Localized regions of extensional deformation along the frontal part of the arc adjacent to the collision zone may contribute to the formation of re-entrants.