Benfontein

Three sills of kimberlite outcrop over an area of 4.5x2.5 km and are emplaced in shales of the upper Dwyka Formation that have been intruded by a large Karoo dolerite sill. There are Upper, Middle and Lower sills with numerous small veinlets. The sills are respectively 2-6, 1.5-2 and 0.25-0.8 m thick. The sills are unusual not only in comprising kimberlite but because near the upper contacts and at various horizons within the black kimberlite white calcite-rich layers occur that vary from 5 mm to 20 cm in thickness and are thickest near the upper contacts of the Middle and Upper Sills. These layers do not persist horizontally but vary from lenticular masses up to 5 m long to trough-shaped channels 4 to 5 cm wide. There are also numerous veins of fibrous calcite. In the upper part of the Lower Sill there is a zone of small-scale diapiric structures consisting of tubular bodies rooted in a calcite-rich layer and having diameters of 0.5-1 cm and up to 3 cm high. These unique features are fully described and illustrated by Dawson and Hawthorne (1973). The kimberlites are conspicuously layered on macro and micro scales and, apart from the calcite layers, the layering is produced by variations in the proportions of olivine, spinel and perovskite with olivine, olivine-spinel and olivine-spinel-perovskite cumulates as well as mineral- and size-graded layering. Mineralogically the kimberlites consist of cumulus grains of olivine, rarer phlogopite, opaque spinel, perovskite and apatite in a matrix mainly of calcite, with some patches of serpentine and rare second generation phlogopite. The calcite is texturally very variable some crystals being long and platy and some tabular. In the calcite layers platy phenocrysts up to 5 mm long may have length to width ratios of 40:1; others are dendritic or skeletal while columnar aggregates of dendritic crystals also occur. The importance of the sills lies in the association of kimberlite with calcite rocks which are clearly carbonatitic i.e. carbonate rocks of igneous origin, and thus this is an important locality for discussion of the relationship between kimberlites and carbonatites. There have been several papers investigating the mineralogy in detail to try to elucidate the problem of whether the carbonate rocks are carbonatites per se. Boctor and Boyd (1981), for instance, looked at the Fe-Ti oxide minerals in both silicate and carbonate layers and from their data of chemical trends in spinel, ilmenite and perovskite, which appeared to be similar to those in kimberlites but not in carbonatites, concluded that the Benfontein sills do not support a genetic relationship between kimberlites and carbonatites. Gaspar and Wyllie (1984) found that in the calcite layers the sequence of spinels and ilmenites correspond to those of carbonatites from alkaline complexes or have characteristics intermediate between those of carbonatites and kimberlites. Primary baddeleyite has been described from the kimberlite by Scatena-Wachel and Jones (1984) which is taken to favour a genetic link between the kimberlite and carbonatite. The arguments concerning the kimberlite-carbonatite association are developed in these papers and others cited by them. They also contain much mineralogical data while an analysis of calcite rock, with some trace element data (Sr = 2795, Ba = 1480, La = 200 p.p.m.), is given by Dawson and Hawthorne (1973).