>Magister Scientiae - MSc / This study is focused on the Akanani prospect area, approximately 25 km north-west of the town of Mokopane, Limpopo Province where exploration geologists at the study area have classified the ‘pyroxenitic’ units into P1, P2, P3 and P4 units upward in order of succession with height based on their textures, mineralogy and colour. The primary aim of this study is to distinguish the distinctive geochemical and mineralogical characteristics that can be used to identify each unit (P1 to P4) and in so doing create major geochemical, petrographic and mineralogical variables that will help or facilitate the exploration for and recovery of PGE and BMS mineralisation. Geochemical and mineralogical variation studies were carried out on the cores from ZF044, ZF045, ZF048, ZF057, ZF078, ZF082 and M0023, located in the Platreef at the Akanani Prospect area on the farms Moordkopje 813LR and Zwartfontein 814LR. Using a combination of various multivariate statistical techniques (factor, cluster and discriminant analysis) and mineralogical studies (CIPW norm, microprobe analysis, petrography), the outcomes of the study have demonstrated that the Platreef at Akanani comprise at least four lithological units i.e. the basal pyroxenite portion referred to as the P1 unit comprises chromitite, pyroxenites and feldspathic pyroxenites with associated Cr, TiO2, chromite, pyroxenes, hematite and Fe2O3, the mineralized section of the P2 unit is characterized by harzburgite, serpentinized harzburgite and in places orthopyroxenites are present consistent with high MgO and LOI contents, the feldspathic portion referred here as the P3 unit is characterized by a feldspathic pyroxenite containing higher Al2O3, Na2O, K2O, albite, hypersthene and SiO2 and the top most portion of the P4 unit comprising CaO, Diopside, ilmenite, anorthite, apatite and P2O5 that can be interpreted to have formed by three separate magma pulses. Considering the possibility that the P4 unit is a hybrid melt of assimilated Platreef that interacted with intruding Main Zone magma, this reduces the number of magma pulses to two. The classification of P1, P2, P3 and P4 units of the Platreef at Akanani shows that the criteria used by mining personnel to classify the four lithological units is not definitive and therefore are not highly reliable. Although various multivariate statistical techniques were employed relatively similar elemental associations were obtained highlighiting the importance of this approach. The strongly positive correlation between sulphides, PGEs and chromite at Akanani is consistent with an orthomagmatic deposit that had been disturbed by significant hydrothermal activity, while in places a good BMS-PGE relationship is commonly associated with the main chromitite stringers in P1. Mineral and whole rock compositions of silicate rocks highlight the strongly magnesian nature of the ultramafic P2 unit. Mineral chemistry studies of chromite, orthopyroxene, olivine, clinopyroxene and plagioclase are consistent with the multi- emplacement model. Convective exchange resulted in the enrichment of iron at the bottom of the stagnant chamber, while incompatible elements migrated upwards consistent with iron depletion with stratigraphic height. Injection of P1 magma and subsequent mixing with country rocks gave rise to the formation of chromitites and addition of plagioclase component to the intruding magma. A normal fractionation trend is suggested between P2 and P3 consistent with enrichment of MgO in P2 and enrichment of Al2O3, Na2O, SiO2 and K2O in P3. The An% of 84.4 of plagioclase coupled with CaO enrichment in P4 is suggestive of some Main Zone influence and can be interpreted as resulting from partial melting and recrystallization of P3 in response to the intrusion of the Main Zone magma is suggested for the formation of the P4 unit. There exists a good correlation between the modal mineralogy and mineral chemistry as determined optically, the norm as determined by the CIPW norm and the whole-rock geochemical results as determined by multivariate statistics and conventional methods.