Low-salinity fluids (T > 200°C reduced S, modest CO2) and high geothermal gradients are common to many gold deposits and provinces. In contrast, host rocks, hosting structures, depth of formation (in the crust during deposition), subsequent metamorphic overprint, alteration mineralogy and isotopic signatures can vary dramatically within single deposits or provinces. Gold deposits with co-product base metals are an exception to the above comments, and probably relate to saline fluids.

The low salinity fluids inferred for major gold-only deposits are not easily explained by seawater, basinal brines, meteoric fluid or common magmatic processes. In contrast, metamorphic devolatilisation of mafic/greywacke rocks is one effective way to produce low-salinity metamorphic fluids with characteristics matching the gold fluids. Such an origin also explains the link to geothermal gradients.

The transition from chlorite—albite—carbonate assemblages to amphibole-plagioclase assemblages (commonly greenschist—amphibolite facies boundary) involves considerable loss of metamorphic fluid whose composition is buffered by the mineral assemblage, and is a function of P and T. This low salinity, H2O-CO2 fluid is evolved at T > 400°C commonly carries reduced sulphur, and may contain Au complexed with this sulphur. This auriferous fluid is likely to mix with other fluid types during times of elevated temperature, especially magmatic fluids at depth, and upper crustal fluids at higher levels.

Gold deposits in Archaean greenstone belts exhibit good evidence of low salinity, H2O-CO2 fluids of T > 300°C these include examples from Canada, Australia, Brazil, Zimbabwe, India, and South Africa. Turbidite-hosted (slate-belt) deposits exhibit similar evidence for such fluids but commonly with appreciable CH4; the Victoria and Juneau (Alaska) goldfields are examples. The Witwatersrand goldfields also show evidence of low salinity, H2O-CO2 fluids carrying reduced sulphur and gold, but their distribution and timing are not well established. Epithermal (sensu lato) gold deposits have evidence for low salinity fluids carrying Au and S, but are much more diverse in character than those from the previously mentioned gold provinces: this probably arises from mixing of several fluid types at high crustal levels. Together these four types of gold provinces account for over 80% of the primary gold mined to date.

Mesothermal, gold-bearing quartz veins are widespread within allochthonous terranes of Alaska that are composed dominantly of greenschist-facies metasedimentary rocks. The most productive lode deposits are concentrated in south-central and southeastern Alaska; small and generally nonproductive gold-bearing veins occur upstream from major placer deposits in interior and northern Alaska. Oreforming fluids in all areas are consistent with derivation from metamorphic devolatilisation reactions, and a close temporal relationship exists between high-T tectonic deformation, igneous activity, and gold mineralization. Ore fluids were of consistently low salinity, CO2-rich, and had δ18O values of 7‰- 12‰ and δD values between −15‰ and −35‰. Upper-crustal temperatures within the metamorphosed terranes reached at least 450-500°C before onset of significant gold-forming hydrothermal activity. Within interior and northern Alaska, latest Paleozoic through Early Cretaceous contractional deformation was characterised by obduction of oceanic crust, low-T/high-P metamorphism, and a lack of gold vein formation. Mid-Cretaceous veining occurred some 50-100 m.y. later, during a subsequent high-T metamorphic/magmatic event, possibly related to extension and uplift. In southern Alaska, gold deposits formed during latter stages of Tertiary, subduction-related, collisional orogenesis and were often temporally coeval with calc-alkaline magmatism.

There is extensive textural evidence for sulphidation in the Middelvlei Reef of the Carletonville Goldfield, South Africa. Secondary iron sulphides have replaced sand- and pebble-sized elasts in conglomerate, as well as matrix material in both conglomerate and quartzite. Within the Middelvlei Reef there is a close correspondence between the intensity of sulphidation and areas of greater modal muscovite, and there is less sulphidation where chloritoid dominates. Both pyrite and minor pyrrhotite are widespread. The timing of the sulphidation appears to be post-depositional but the relative importance of diagenetic and syn-metamorphic processes cannot be determined using textural criteria. The absence of retrograde effects suggests that sulphidation occurred prior to or during peak metamorphism. The distribution and intensity of sulphidation and phyllosilicate alteration indicate that these processes may be more significant in the Witwatersrand than has generally been thought. These processes should be considered in all future genetic models.

Multi-element crush-leach analysis of H2O-CO2 inclusion fluids from a suite of six vein samples from gold-quartz veins in the Brusson district demonstrates that their solute chemistry (c. 5 wt.% NaCl equivalent) is dominated by sodium chloride with lesser amounts of calcium bicarbonate, potassium chloride and sodium bicarbonate. The samples have been analysed both for gas species (CO2, H2O, N2 and H2S) and for Na, K, Li, Rb, Ca, Mg, Sr, Ba, Fe, Mn, Zn, Pb, Cu, Al, As, B, SO42−, F, Cl, Br and I. The fluids contain appreciable H2S (>10−3 molal), which correlates with the contents of As, CO2 and B. Concentrations of many cations remain similar irrespective of wall rock, but there is evidence of leaching of Li, and possibly I, from some wall rocks. Large variations in the K-content of the fluid may result from precipitation of sericite. The bicarbonate concentrations in the fluids, estimated from charge imbalance, are substantially less than their total CO2 content when trapped as single phase fluids, indicating a low pH. Sulphate : sulphide ratios suggest relatively reducing conditions, which is consistent with Fe concentrations significantly greater than Mn.

The gold-quartz veins formed as H2O-CO2 fluids of modest salinity and very uniform composition ascended from depth. Halogen ratios of the fluids are consistent with an ultimate origin for these fluids from deep-penetrating surface or connate waters although such a model requires extremely low fluid : rock ratios, to account for the hydrogen isotope composition of many similar deposits. There is as yet insufficient reference data to use halogen ratios as a rigorous test for the alternative model of an origin for the fluid by metamorphic devolatilisation.

The Hill goldfield, NSW, Australia, is an example of a syntectonic, slate-belt gold deposit formed in a multiply deformed, Late Silurian slate-metagreywacke turbidite sequence. Gold is confined to bedding-parallel veins and discordant leader veins composed of as many as four generations of quartz, accompanied by phyllosilicates, carbonates and minor sulphides. Vein formation and gold deposition was apparently synchronous with Early Carboniferous metamorphism and deformation. Homogenisation temperatures (Th) for fluid inclusions in vein quartz demonstrate five groupings in the temperature intervals 350-280°C 280-250°C 250-190°C 190-150°C and 150-110°C corresponding to a variety of primary and secondary inclusions developed during four periods of vein quartz deposition under a generally declining temperature regime. Inclusion fluids are characterised by a low salinity of around 0.1 to 3.6 wt. % NaCl equivalent. The dominant gas phase present in the inclusion fluids varies from N2 in the early stages of the paragenesis, through CH 4 during the main episode of gold deposition, to CO2- rich fluids associated with late-stage mineralisation. δ18O values for vein quartz (range 15.1-17.1‰) and vein carbonate (range 11.3-13.4‰) are typical of metamorphic mineralisation. δD composition of hydrous minerals and inclusion fluids (range −53 to −138‰) suggest an influx of meteoric water in the later mineralising fluids. This conclusion is supported by δ13C data for vein calcite (range −2.5 to −9.7%0). δ34S composition of vein pyrrhotite and pyrite ranges from 6.9 to 7.8‰ early in the paragenesis, to lighter values (around 4.2 to 5.8%0) accompanying late gold deposition from more oxidising fluids. Sulphur isotope data imply a sulphur source from underlying turbidites and an increase in fluid oxidation state during mineralisation . Lead isotope measurements on vein pyrite, arseno py rite, galena and gold are characterised by two isotope populations with 207Pb/206Pb ratios of 0.862 and 0.860, which define two discrete mineralising events during vein formation. Consistency between data from vein minerals and lead isotope signatures for potential source rocks indicate that lead was derived from the sedimentary pile.

Vein formation occurred throughout a deformation sequence which involved early transpressive ductile deformation through to late-kinematic transpressive brittle structures which host a series of gold prospects. Fluid inclusion data from (S1) fabric parallel veins associated with early deformation suggest that a low-salinity aqueous fluid, with a mean salinity of 6.4 wt.%, was present during peak metamorphism, Pelite mineralogy and isochores constrain peak metamorphism to the lowermost part of the upper greenschist facies at 325 to 425°C and 1.4 to 3.4 kbar.

Fluid inclusion data from auriferous and barren late-kinematic quartz veins, both containing unmixing assemblages of aqueo-carbonic inclusions with low salinities of ≈2.7 wt.% NaCl equiv., indicate unmixing occurred at 300°C and 1.5 kbar.

Volatiles (CO2, N2, CH4) are observed in all the late-kinematic veins. The N2 contents of veins with elevated gold grades are typically higher than those with low gold grades. N2 reaches 8.7 mole% in a vein with 0.49−4.6 p.p.m. Au compared to <1 mole% in a vein with <0.05 p.p.m. Au. The CH4 content of late kinematic veins is generally less than 1 mole% and shows no relative enrichment in mineralised veins. The generation of N2 in the mineralising fluid most likely results from interaction of fluid with the ammonium ion, NH4+, in micas and feldspars. This interaction could take place either at source, due to metamorphic devolatisation reactions, or along those structures which acted as fluid conduits due to fluid-rock interaction.

The Rhenohercynian thrust- and fold-belt of Central Europe hosts a syn- to post-kinematic Variscan vein mineralisation, which is restricted to the Rhenish Massif, Germany. It is formed during four major stages and siderite is the principal ore mineral of the 'main stage'. The latter can be traced throughout the massif, but is mainly developed in the Siegerland district as synkinematic lodes which are hosted by low-grade metamorphic pelites of Lower Devonian age.

Fluid inclusion studies prove large-scale homogeneous ore forming fluids of low salinity (≤5 wt.% NaCI equiv.) and Na(-K > -Fe ≫ Mg)-C1 composition, which are CO2 undersaturated (XCO2 = 0.003-0.1). Siderite precipitation at 220−≤320°C and 0.7-1.4 kbar can be deduced by microthermometry, chlorite themometry, REE fractionation and experimental data. Thereby the maximum formation temperature comes close to or even exceeds the peak metamorphic temperature. From Mn contents and δ13C-δ18O variation of siderite a trend is apparent in which formation temperatures gradually decrease from S.E. to N.W. across the belt. This trend correlates with decreasing degree of host-rock deformation and decreasing metamorphic grade. Fluid composition as well as stable isotopes and REE fractionation of siderite point to ore-forming solutions, which were generated and equilibrated at depth during prograde metamorphism.

Structural characteristics of the lodes, age determinations, and P-T estimations prove fluid ascent and siderite precipitation during and/or immediately after peak metamorphism predating the postkinematic magmatism of the Rhenohercynian belt. Main(-siderite)-stage mineralisation of the Rhenish Massif is classified as metamorphogeneous.

Single crystals of magnetite and of hematite have been dissolved at atmospheric pressure in superheated melts in the systems CaO-MgO-Al2O3-SiO2 and CaO-Al2O3-SiO2, and in a basalt. The crystals were suspended in alumina crucibles containing ca 3.5 cm 3 of melt. Quenched run products were examined optically and by electron probe analysis to establish the distribution of Fe in the glassy charges. There is usually a concentration of Fe at the base of a run product, consistent with flow of dissolved matter from the crystal to the floor. One or more columns of brown, Fe-rich glass may extend from the underside of a relic crystal towards the floor. In CMAS run products, such columns typically extend this entire distance, whereas in the CAS and basalt run products, the columns are either detached from the crystal or do not reach the floor. In the CMAS melt (viscosity ~1 poise) there is apparently continuous release of Fe-bearing melt from around a dissolving crystal, whereas in the CAS and basalt melts (viscosities 7000 and 300 poise, respectively) release is intermittent. In CMAS run products the Fe content is usually greatest, in glass, at the base, and declines gradually upwards; in CAS and basalt runs the bottom of a crucible is occupied by discrete, sharply bounded pillows of Fetich glass, with only slight, or no, gradation in composition. Rising gas bubbles can elevate small blobs of the denser, Fe-bearing melt from around a dissolving crystal, and trains of bubbles in the CAS melt may guide this Fe-bearing melt, against gravity, to the surface of the charge. When the bubbles burst at the surface, this dense melt is left in an unstable location and releases diapirs which descend to the bottom of the crucible. In spite of the evidence that convective fractionation occurs in these haplomagmas and in the basalt, it remains to be demonstrated that it will occur during sidewall crystallization or during the growth of minerals in a cumulus mush to cause magmatic differentiation.

Herzenbergite, SnS, was found in dykes in three granite pegmatite fields in north-central Sweden. It occurs as erratic replacements of cassiterite. The formation of herzenbergite was probably triggered by local interactions with fluids from sulphide-bearing graphitic schists located in the immediate vicinity of the pegmatite dykes.

A carbonatite dyke from the Sarnu-Dandali alkaline complex, Rajasthan, India, contains a remarkable suite of rare earth, strontium-rich minerals with spectacular primary textures.

Sr, Mn-rich calcite in the outer 5 mm of the dyke contains exsolved lamellae of carbocernaite, (Ca,Na)(Sr,Ce,Ba)(CO3)2, orientated parallel to its twin and cleavage planes. The amount of exsolved carbocernaite increases away from the dyke margin as the Sr content of the calcite increases to a maximum 13 wt.%. Sr levels as high as this in calcite have previously been recorded only in experimental work. The carbocernaite exsolution suggests that Sr-rich calcium carbonate can be a host for major amounts of REE in carbonatite magma.

Separated by a sharp internal boundary, is a complex possibly cotectic intergrowth of carbocernaite and Sr-rich calcite with late Ca-rich strontianite (19 wt.% CaO). Other minerals in the dyke include baryte, pyrrhotite, alabandite, sphalerite and occasional bastnäsite-(La) and thorite. Bands of late britholite-(Ce) traverse the dyke.

The host rock for the dyke is fenitized melanephelinite which is itself traversed by narrow, <1 mm, carbonatite veins beleived to predate the carbonatite dyke. Allanite, britholite-(Ce) and rare monazite-(Ce), developed at the boundary between the carbonatite dyke and the fenite, may have been produced by a reaction between the dyke and the wall rock, or may be related to the later britholite mineralisation.

The textures and mineral compositions indicate primary crystallisation. They are unique amongst rare earth-rich carbonatites which are usually late-stage phenomena with signs of secondary alteration.

Comparison with experimental data available for the calcite-strontianite system suggests conditions of 500°C and 2 kbar for coexisting Sr-rich calcite and Ca-rich strontianite. A smaller scale intergrowth of calcite containing only 2.9 wt.% SrO and coexisting Ca-strontianite may correspond to a further unmixing at 350°C and 2 kbar. Since no experimental data are available for a calcite-carbocernaitestrontianite system, mineral chemistries and the interpreted sequence of crystallisation have been used to construct a hypothetical phase diagram.

A partly destabilised Na-richterite has been found in an olivine-nephelinite from Morocco. The riehterite crystal (600 × 420 μm) is surrounded by a reaction zone (400-700 μm) of K- and Si-rich glass containing small (<50 μm) olivine (Fo80-83%) and endiopside crystals. Outwards, another zone is formed of normal magmatic minerals and circumscribes the original crystal, indicating that the destabilisation event took place at the end of the crystallisation sequence. Estimated ascent time of about 100 hours would have completely decomposed an isolated richterite crystal, which suggests that the amphibole was originally included in a xenolith. A mass-balance calculation shows that the fichterite isovolumic decomposition was accompanied by exchanges with the magma. The loss of Na from the reaction zone and the gain of AI from the magma allowed the precipitation of an analcime-rich zone observed around the destabilised amphibole and the concentration of K in the reaction zone glass. Compositional variations, Fe and Ti increase and Mg, Ca and F decrease at the richterite edge are interpreted as the result of a diffusion process. No alkali gradients are observed. The diffusion phenomenon lasted less than 100 hours and ceased to be operative at a temperature of 900-950°C i.e. just below the solidus temperature. Diffusion coefficients for the amphibole are proposed: e.g 10−9 cm2 s−1 for K2O and 10−10 for FeO at 900°C