The gold balance in an ore deposit where the ore is treated by cyanide is the sum of the ‘visible gold’ that is amenable to cyanidation and ‘visible gold’ and the ‘invisible gold’, which are not amenable to cyanidation. Petrographic analyses, electron and ion microprobe as well as scanning electron microscope studies of ore from the Golden Sunlight deposit, Montana, suggest that periods of relatively poor gold recoveries are primarily due to the presence of inclusions, <25 µm in size, of native gold, petzite, calaverite, buckhornite and krennerite. These are encapsulated in cyanide insoluble grains of pyrite, chalcopyrite and tennantite and are present in the tailings. This contribution probably accounts for 3–25% of the unrecoverable gold processed during the life of the mine. Minor amounts (6–7%) of ‘invisible gold’, as indicated by ion microprobe studies and the presence of up to 5% ‘visible gold’ in buckhornite, which is rare in nature, appears to account for the remainder of the gold budget.

Benleonardite, ideally Agg(Sb,As)Te253, occurs in ore specimens from the Mayflower and Gies epithermal gold-silver telluride deposits in Montana commonly spatially associated with hessite and tetrahedrite. In these deposits, it is Cu-bearing (up to 2.7 wt.%) and exhibits a slight deficiency in Ag+Cu coupled with a slight excess in S. A cervelleite-like mineral coexists with benleonardite at Mayflower and is unusual in composition in that it is Se-bearing suggesting the possibility of solid solution with aguilarite (Ag4SeS).

Interlayered mafic–telsic intrusions from the Mineral Lake intrusive complex in northwest Wisconsin reflect the typical bimodal basalt–rhyolite compositional pattern of the Midcontinent Rift flood basalt province in the Lake Superior region. The later felsic intrusions were emplaced between the mafic intrusions and overlying basalt flows, and postemplacement fractional crystallization produced gradational mineralogical and geochemical variations. Isotopic and trace-element data for the Mineral Lake intrusions are consistent with mantle sources for both mafic and felsic intrusions, with compositional differences explained by the extent of fractional crystallization and crustal contamination or mantle source characteristics.εNd–εSr plots of analyzed Midcontinent Rift igneous rocks define three largely separate isotopic fields that suggest separate sources. However, the spread in isotopic data and a spider diagram plot of mafic samples from the εNd = εSr = 0 field suggest a crustal component and derivation from depleted rather than chondritic mantle. Evolved felsic rocks plotting in two negative εNd – positive εSr fields can be explained by derivation from separate enriched mantle sources or crustal contamination or both.