We present results of powder X-ray-diffraction and microstructural studies of analcime from the volcanic rocks of the Crowsnest Formation, Alberta, Canada. The powder X-ray diffractogram of only one of the analcime samples contains a peak inconsistent with the cubic symmetry that has been attributed to conversion from leucite. We re-evaluate this criterion, and suggest that primary igneous analcime may be cubic as well; caution thus is required in deducing paragenesis from crystal symmetry. A more compelling argument for the primary nature of the analcime is that fresh fracture-surfaces are clean and planar, the characteristics suggested by previous workers to be indicative of crystallization from a melt or fluid. The fracture surfaces lack the porous microstructure previously attributed to the conversion of leucite to analcime. SOMMAIRE Nous présentons les résultats d'études en diffraction X (méthode des poudres) et microstructurales de l'analcime des roches volcaniques de la formation de Crowsnest, en Alberta, Canada. Un des trois échantillons contient une raie en violation de la symétrie cubique, antérieurement attribuée à une transformation de l'analcime à partir de la leucite. Nous ré-évaluons la validité de ce critère, et nous préférons l'hypothèse que la symétrie cubique pourrait aussi caractériser l'analcime ignée primaire. Il est donc nécessaire d'utiliser ce critère paragénétique avec circonspection. Il semble plus sûr de se fier sur la microstructure des phénocristaux. Les surfaces de fractures sont propres et planaires dans le cas de l'analcime fraiche et primaire, indication d'une croissance à partir d'un bain fondu ou bien d'une phase fluide. Les surfaces de fractures sont dépourvues des microstructures poreuses attribuées dans la littérature à la transformation de la leucite en analcime.(Traduit par la Rédaction)

Alkaline igneous rocks of the Crowsnest Formation in southwestern Alberta and in the Howell Creek area in southeastern British Columbia have been suggested previously to be cogenetic. To test this hypothesis, samples of both suites were characterized petrographically and their major and trace element geochemistry was determined. A subset of the samples was analyzed for whole-rock Sr and Nd isotope geochemistry. The samples of the two suites are latites, trachytes, and phonolites based on the International Union of Geological Sciences (IUGS) total alkalis versus silica (TAS) diagram. Samples from both suites show similar patterns on mantle-normalized trace element diagrams, being enriched relative to mantle values but depleted in the high field-strength elements Nb, Ta, and Ti relative to the large-ion lithophile elements. The chondrite-normalized rare-earth element (REE) patterns for both suites are light REE enriched, with no Eu anomaly and flat heavy REE. The isotope geochemistry of both suites is characterized by low initial 87Sr/86Sr (SrT = 0.704 to 0.706) and low εNdT (–7 to –16). The Howell Creek samples have lower εNdT and higher SrT than do the Crowsnest samples. Based on the intra- and intersuite differences in the isotope geochemistry, we conclude that these samples are not cogenetic, but rather represent samples that have experienced similar evolutionary histories from a heterogeneous source region in the subcontinental lithospheric mantle.

No abstract