© 2015. Orogenic gold deposits of all ages, from Paleoarchean to Tertiary, show consistency in chemical composition. They are the products of aqueous-carbonic fluids, with typically 5-20mol% CO2, although unmixing during extreme pressure fluctuation can lead to entrapment of much more CO2-rich fluid inclusions in some cases. Ore fluids are typically characterized by significant concentrations of CH4 and/or N2, common estimates of 0.01-0.36mol% H2S, a near-neutral pH of 5.5, and salinities of 3-7wt.% NaCl equiv., with Na>K>>Ca,Mg. This fluid composition consistency favors an ore fluid produced from a single source area and rules out mixing of fluids from multiple sources as significant in orogenic gold formation. Nevertheless, there are broad ranges in more robust fluid-inclusion trapping temperatures and pressures between deposits that support a model where this specific fluid may deposit ore over a broad window of upper to middle crustal depths.Much of the reported isotopic and noble gas data is inconsistent between deposits, leading to the common equivocal interpretations from studies that have attempted to define fluid and metal source areas for various orogenic gold provinces. Fluid stable isotope values are commonly characterized by the following ranges: (1) δ18O for Precambrian ores of +6 to +11 % and for Phanerozoic ores of +7 to +13% (2) δD and δ34S values that are extremely variable; (3) δ13C values that range from -11 to -2% and (4) δ15N of +10 to +24% for the Neoarchean, +6.5 to +12 for the Paleoproterozoic, and +1.5 to + 10 % for the Phanerozoic. Secular variations in large-scale Earth processes appear to best explain some of the broad ranges in the O, S, and N data. Fluid:rock interaction, particularly in ore trap areas, may cause important local shifts in the O, S, and C ratios. The extreme variations in δD mainly reflect measurements of hydrogen isotopes by bulk extraction of waters from numerous fluid inclusion generations, many which are not related to ore formation. Radiogenic isotopes, such as those of Pb, Sr, Nd, Sm, and Os, measured on hydrothermal minerals are even more difficult to interpret for defining metal source, particularly as the low-salinity ore fluids transport limited amounts of these elements and significant amounts of these may be locally added to the minerals during alteration reactions at the sites of gold deposition. Noble gas and halogen data are equally equivocal.Fluid exsolution from granitoids emplaced into the upper and middle crust, metamorphism of the crust, or fluids entering trans-crustal fault zones from below the crust all remain as permissive scenarios associated with orogenic gold formation, as the abundant geochemical data are equivocal. However, geological and geochronological data weigh heavily against a magmatic-hydrothermal model in the upper to middle crust. There is no universal temporal association between orogenic gold and magmatism, and where there is an overlap in age, there is no specific type of magmatism consistently associated with gold formation, nor element zonation around any specific pluton. A crustal metamorphic model for fluid and metal sources is very consistent with geological, geochronological, and geochemical data, although metamorphism on a regional scale that releases these components into major fault zones can be associated with many processes along active continental margins. These can include crustal thickening and radiogenic heating, slab rollback and heating during crustal extension, or subduction of a spreading ridge heating the base of an accretionary prism. In rare examples where Phanerozoic orogenic gold deposits are hosted in Precambrian high-grade metamorphic terranes, fluids and metals must, however, enter a transcrustal fault system from a sub-crustal source. This could either be a devolatilized, subducted, relatively flat, perhaps stalled slab and its overlying sediment, or the corner of the fertilized mantle wedge that releases a fluid during a thermal event.