Assuming that many hydrothermal ore deposits are formed from emanations given off from a magma at depth while it cools through the interval in which latent heat of crystallization is generated, it is shown that this cooling interval for magmatic bodies of moderate size must be measured in tens or hundreds of thousands of years. Emanations from such a magma should change at the source with time: relatively insoluble volatiles should depart early and the more soluble ones late; the general order is probably sulfur gases and oxides of carbon, water, chlorides, and fluorides. Experimental and field evidence indicates that this order approximates the increasing solubility of these gases in natural magmas. Theoretical considerations show that within a hydrothermal conduit a relatively small gradient would soon be established between the magma and the surface. A small gradient suggests that temperature drop is a minor factor in precipitating substances in solution, whereas a drop in pressure and reaction with wall rocks or with material precipitated from earlier emanations would be of major importance. The sulfur-rich early emanations tend to react with indigenous iron of the country rock, or with iron carried by carbon dioxide-rich solutions to where a marked pressure drop occurs; either of these reactions will form abundant early iron sulfide. Later sulfur-deficient emanations, which then carry soluble halides of ore metals, react with this early iron sulfide to precipitate the ore mineral sulfides by replacement and deposition with loss of iron to the solution. Precipitation of much sulfide ore is thus commonly accomplished by sulfur that was fixed near the site of the ore body by earlier emanations from the magmatic source; a large amount of ore, however, may be precipitated from late-stage magmatic solutions where they mingle with early-stage sulfur-bearing solutions from a different magmatic source.