Abstract In this study, we have used tectonic-diffusion model calculations to estimate resources of bedrock-hosted granitic tin deposits of Phanerozoic age. These deposits range from proximal skarns, greisens and pegmatites in or near plutonic granites, to distal veins and disseminations, many of which are associated with subvolcanic rhyolites. Tin is also found in residual and placer deposits, some of which are associated with bedrock deposits. Most of the bedrock deposits formed at temperatures of 200–500 °C from fluids with a wide range of salinities and CO 2 contents. Limited information suggests that they formed over a relatively continuous range of depths from a few hundred metres to as much as 6 km, with an average of about 2 km. Information was obtained on 547 deposits, including 435 with measured or estimated ages and 301 with estimated tin contents, almost all of which are bedrock rather than residual-placer deposits. The bedrock deposits have an average size of 58 000 t, and the largest five deposits, including one in Bolivia and four in China, account for 46% of the total tin resources in all deposits. Tectonic-diffusion model calculations indicate that approximately 14 520 granitic tin deposits remain in the crust and that they contain 8.4×10 8 t of tin. If about 50% of the deposits above depths of 1 km can be discovered and mined, current tin production of about 260 000 t annually can continue for about another 200 years. Supplementary material: Data on granitic tin deposits are available at www.geolsoc.org.uk/SUP18688

Abstract We have estimated the magnitude of resources available to support world production of gold into the next millennium. The estimate was made using the tectonic-diffusion computational model in which ore deposits move through depth-time space in response to global tectonism, and it is based on a global compilation of ages and gold contents for each type of deposit. The method was applied to the most important hydrothermal deposits that yield gold, including Carlin-type, epithermal, iron oxide-copper-gold, orogenic, porphyry copper, skarn, and volcanogenic massive sulfide. As production from the Witwatersrand deposits has declined, these types of hydrothermal deposits have supplied a growing fraction of global gold production, and it is likely that this pattern will continue. Estimates were made for gold resources to crustal depths of 1 and 3 km, which are likely depth limits for most mineral exploration and production. Our results indicate that porphyry copper and epithermal deposits will be the most important hosts of gold produced in the future. The contribution to future production from orogenic gold deposits is likely to decrease relative to other types of deposits because orogenic gold deposits do not increase in abundance as rapidly downward through the uppermost crust as do epithermal and porphyry copper deposits, which form at much shallower crustal depths. Although the gold resource estimated here, about 1 million metric tons (Mt) to a depth of 1 km and almost 5 Mt to a depth of 3 km, is large relative to current estimates of gold reserves, recoverable gold will probably be much smaller, possibly by as much as 50 percent, because of cultural, geologic, and mining-processing factors. Recoverable gold resources to a depth of about 3 km in the crust could supply current world mine production of gold for about 1,000 yr. Although this is a long period of time, it is short relative to the ~7,000–yr history of gold mining. These estimates highlight the fact that a growing fraction of world gold supply will have to come from buried deposits, many below postore cover, and from deposits in which gold is a co- or by-product.