| dc.description.abstract | The Neoarchæan greenstone belt of western Kenya forms part of the Tanzania Craton of East Africa. Two groups of supracrustal rock sequences are recognised in this greenstone belt, namely a dominantly volcanic supergroup (the Nyanzian) and a dominantly sedimentary one (the Kavirondian); this is typical of many other Neoarchæan greenstone belts. The Nyanzian volcanics include mafic lavas (rarely ultramafic) which could be part of an original oceanic plateau sequence, but comprising an ‘island-arc’ sequence of basaltic andesites, dacites and rhyolites. These supracrustal rocks are intruded by later granites, and it is the petrogenesis of the granitoids that is the focus of this paper.
Geochemical studies indicate that the granites are all I-type, having calc-alkaline affinity, compositionally varying between monzodiorite, through granodiorite and quartz monzonite, and having chemical trends comparable to those of granites found in present day Andean-type continental arc environments. The granitoids are very primitive in terms of isotropic ratios, and most have high Ba and Sr contents and low Y and HREE typical of many trondhjemite-tonalite-granodiorite (TTG) plutonic suites that dominate the Archæan, but are more potash-rich than most such suites. They carry a high proportion of mafic enclaves. There is little doubt that the rocks are related in their petrogenesis, but are not crustal melts. While major element modelling is superficially compatible with the calc-alkaline trends being derived by crystal fractionation, it is more likely they are derived through melting of a mafic source with residual garnet (to account for the HREE and Y depletion). This paper also speculates on how the granitoids could be related to the calc-alkaline volcanics and to the mafic volcanic sequence | en |
