©2015 Society of Economic Geologists, Inc. Economic Geology, v. 110, pp. 1295–1312 The Nickeliferous Archean Madziwa Igneous Complex, Northern Zimbabwe: Petrological Evolution, Magmatic Architecture, and Ore Genesis M. D. Prendergast 1,† and A. H. Wilson 2 1 Guesachan, Shielhill Road, Kirriemuir, Angus, Scotland DD8 4PA, United Kingdom 2 School of Geosciences, University of the Witwatersrand, 1 Jan Smuts Avenue, Braamfontein, 2000, Johannesburg, South Africa Abstract The nickeliferous Madziwa Igneous Complex consists of several lenticular, mafc-ultramafc enclaves within late Archean gneissic terrain. The lenses of igneous rock are locally deformed and metamorphosed to lower greenschist grade. They are here interpreted to be remnants of a large, composite, magmatic structure with two main components: (1) a set of narrow (10s–100s of meters wide) dikes intruding the gneissic foliation, and (2) a large lopolith made up of leuconorite. Some dikes contain a differentiated and vertically oriented layered suite comprising a central pyroxenite layer, plus norite as both a continuous marginal layer and intermittent layers within the pyroxenite. The pyroxenite-norite suite has a preliminary U-Pb zircon age of 2684 Ma. Other dikes are made up of diorite and/or ferrodiorite; in places, the dioritic rocks also intrude the pyroxenite-norite suite. The leuconorite lopolith transects both the dikes and the gneissic country rocks, its basal contacts with the pyroxenite-norite suite varying locally from intrusive to gradational to (magmatic) erosional. Although modi- fed in places by secondary mobilization, disseminated Ni-Cu sulfdes are primarily hosted within the central (ortho- to mesocumulate) pyroxenite adjacent to internal norite layers. Whole-rock geochemical data establish the comagmatic origin of the principal rock types and indicate an Archean, D-type, basaltic source magma with ca. 8% MgO, the two dioritic rock types representing late, immiscible, silica- and Fe-Ti-P-rich derivatives of pyroxenite-norite(-leuconorite) crystallization. In addition to the major rock types, peridotites occur in several outlying dikes and may represent fractionates of a komatiitic precursor to the basalt. Geologic and geochemical evidence points to bulk assimilation of country rock gneiss by the Madziwa magma and sulfde segregation trig- gered by felsic contamination. These processes did not occur locally within the dikes but rather in the conduit system prior to the emplacement of the magma charged with sulfde droplets. The unusual vertical layering of the pyroxenite-norite sequence and the localization of the sulfde ores are attributed to the strong outward, cooling gradient across such narrow dikes and to large-scale, lateral movement of sulfde droplets through the solidifying, pyroxene crystal framework ahead of an advancing postcumulus plagioclase crystallization front. Introduction The Madziwa Igneous Complex in northern Zimbabwe comprises a regional group of nickeliferous, mafc-ultramafc lenses set within an Archean gneiss terrain (Figs. 1, 2; Stidolph, 1977). Between 1969 and late 2000, about 12 million tons of low-grade, cupriferous, nickel sulfde ore were produced from several deposits within the Madziwa Igneous Complex. Previ- ous work (Birch, 1985; Birch and Buchanan, 1989) suggested that local, felsic, wall-rock contamination played the principal causal role in sulfde segregation at Madziwa mine and the camp was later cited as a type-example of sulfde ore genesis by interaction of mafc magma with a felsic crustal melt (e.g., Naldrett, 1997). Nonetheless, uncertainties about the primary host-rock structure (e.g., Birch and Buchanan, 1989) and the age and geotectonic setting of the Madziwa Igneous Complex (e.g., Stidolph, 1977; Birch and Buchanan, 1989; Malenga, 1996; Dirks and Jelsma, 2006) remain a critical impediment to a full understanding of ore genesis at Madziwa. This paper reexamines these and other important aspects of the Madziwa Igneous Complex in light of investigations car- ried out before fnal mine closure (including detailed map- ping, walk-over surveys, logging of 1990s exploration drill core, and reassessment of original, late 1960s drilling data plus mine assay and geologic plans: Prendergast, unpub. data, 1992–1995, 1997) as well as whole-rock geochemistry and preliminary U-Pb zircon geochronology. 0361-0128/15/4328/1295-18 1295 Submitted: July 16, 2014 Accepted: January 14, 2015 † Corresponding author: e-mail, marprend@hotmail.com M Mr P E B Eo Bulawayo Mutare Harare B o t s w a n a M o z a m b i q u e South Africa Z a m b i a 100 km Fig.2 Great Dyke Greenstone belt Zimbabwe craton Granitoid, orogenic belt and cover Madziwa batholith Zambezi belt 32°E 28°E 26°E 17°S 19°S 21°S 30°E Fig. 1. Geologic map of Zimbabwe. Inset shows position of Zimbabwe in Africa. Note (1) location of Madziwa batholith almost completely surrounded by Bindura-Shamva, Mount Darwin, and Dindi greenstone belts (see Fig. 2), and (2) location and trend of Zambezi metamorphic belt. B = Belingwe belt. Lettered Ni deposits belong to a probable 2.70 to 2.68 Ga nickeliferous magmatic event: E = Empress mine; Eo = Elmo claims; M = Madziwa mine; Mr = Muir claims; P = Perseverance mine.