With the price of gold having climbed to above $1400 oz^-1 while social revolutions develop in North Africa and the Middle East, articles on how gold deposits form will get a wider readership than they would have during its doldrum years in the late 20^th century – the price has increased 7-fold since 2000. The bulk of gold nowadays can be mined profitably from ores in which it cannot be seen, except using a microscope, at grades well below 1 g t^-1 (1 part per million) thanks to cheap heap-leaching with sodium cyanide of lightly milled ore. The epitome of low-grade gold is that produced at huge open-pit mines in Nevada from sedimentary host rocks. The gold is far too fine grained to have been deposited as placers, and also occurs dissolved in pyrite (Fe₂S), so most experts regard it as having been introduced by hydrothermal fluids. Yet that covers two possibilities: by deep penetration of groundwater or from magmatic waters, and it is hard to decide which, again because the mineralisation is too fine grained to allow conclusive studies of fluid inclusions and stable isotopes. Also, such evidence as there is suggests low temperature fluids (~200° C) with low salinity; ambiguous data.

By using a synergy of ore-mineral chemistry, experimental data and ages of magmatism and mineralisation, Nevadan geologists have developed a convincing model for these ‘Carlin-type’ deposits (Muntean, J.L. et al. 2011. Magmatic-hydrothermal origin of Nevada’s Carlin-type gold deposits. Nature Geoscience, v. 4, p. 122-127). First, the mineralisation is of Eocene age and was introduced in Lower Palaeozoic sediments. The Eocene in the western USA saw the end of a period of compressional tectonics related to subduction since the Jurassic, fluids from which gave rise to partial melting of the overlying mantle wedge. This was succeeded by extensional tectonics and further intrusive magmatism dated between 40 to 36 Ma. This provided thermal energy and passageways for fluid migration. The second line of evidence is that hydrogen- and oxygen isotopes from fluid inclusions in hydrothermal gangue minerals show evidence that both mantle-derived and meteoric water mixed in the ore-forming fluids, and sulfur isotopes are similarly evidence of dual origin. Thirdly, the authors reasonably postulate from experimental data that basaltic back-arc magmas of Jurassic to Eocene age may have repeatedly added metals, including gold, to the mantle wedge that underpinned Nevada during subduction over a 175 Ma period. Thus later extension-related magmatism sourced in the wedge would itself have become metal enriched from this ‘fertile’ source. Moreover, conditions would have been ripe for highly oxidised conditions in the magmas and high concentrations of water in their fractionated descendants. Under such conditions gold and other metals favour entry into hydrothermal fluids. Given the extensional tectonic conditions such fluids could rise efficiently. Initially highly saline and very hot, rapid rise of the fluids would eventually result in them cooling adiabatically and separating into a dense salty liquid or brine and remaining vapour. That would force down the chlorine content in the vapour, favour some metals (Fe, Ag, Pb, Zn and Mn) ending up in the brine, while others (Au, Cu, As and Sb together with S) would remain in the vapour phase together with dissolved CO₂ in large amounts, making the vapour acidic. Able to pass into the fractured Palaeozoic cover, the fluids widened fractures in the carbonate sediments and facilitated their own precipitation of minerals, the foremost being gold-bearing pyrite. Nevada is probably unique, but my goodness it is a big gold province; >6000 t of gold in tham thar hills.