Geode

Abstract: Geochemical studies of the six lowermost lava flows of the Cretaceous Serra Geral Formation (Parana volcanic province) in Quarai (Brazil) and Artigas (Uruguay) were combined with flow-by-flow field studies of structures and scintillometric profiles to establish a consistent regional stratigraphic framework over at least 100 km. This greatly improves exploration capability for amethyst and agate geodes. A basalt, colada Mata Olho (Alegrete facies, Serra Geral Formation), was the first lava to flow over the ancient Botucatu desert in the region, but an andesite, colada Catalan, overstepped this basalt in many places, perhaps palaeohighs. Four basaltic andesites complete the lava stratigraphy in this formation, adding up to 300 m of lavas. The stratigraphic sequence of contrasting lava compositions is 51.0 wt% SiO2 in the first lava, followed by 57.5, 52.5, 56.0, 53.0 and finally 54.5 wt% SiO2. Overall MgO variation is between 2 and 7 wt%. All lavas in the two districts are low-Ti (<2.0 wt% TiO2) of the Gramado type. The characteristic contents of most major and trace elements (124 rock samples analysed) allow the ready identification of each lava. Contrasting rock chemistry also results in strong variation in scintillometric values (270 points measured in the field and nineteen continuous borehole profiles); from bottom to top of the stratigraphy, the cps values are 49±3.2, 123±10.3, 62±4.7, 94±4.6, ~45 and ~85. Colada Catalan has the structure of aa lava, particularly the contorted igneous banding and autobreccias in the upper and lower crusts. In some places, a 2 m thick, silicified sandstone layer lies on top of some coladas, and silicified sandstone forms breccias with volcanic rocks. Geochemistry of the six lavas indicates complex evolution, involving melting of lithospheric mantle, injection into the crust and assimilation of crust followed by fractional crystallization. This study indicates the possibility of world-class deposits of amethyst geodes on the Brazilian side of the border with Uruguay.

Abstract: IN a study of authigenic silica in the Cretaceous Edwards Limestone of Travis County, Texas, J. S. P.1 noted in passing that some of the chalcedony had a peculiar 60° grid structure and was also length-slow as tested by insertion of the gypsum plate; this is in contrast to normal chalcedony which is length-fast. (In this context, “length-slow” means that the slower polarized light component produced in double refraction is the one vibrating parallel to the crystal length; “length-fast” can be defined similarly for the faster component.) These chalcedony masses, which grade outward into megaquartz, came from an area of collapse structures apparently caused by solution of gypsum/anhydrite- nodules with some replacement by authigenic silica. West2 has also found length-slow chalcedony in Jurassic evaporitic deposits in England. It therefore seemed possible that there was a connexion between length-slow chalcedony and sulphate evaporite minerals and comprehensive literature search revealed more examples of length-slow chalcedony—all associated with evaporites.