Magmatic sulfides are generally accepted as forming by segregation of an immiscible sulfide liquid from a host silicate melt. Immiscible sulfides have been observed in many types of igneous rocks; however, some types of plutonic and volcanic rocks lack sulfides. We have examined a suite of samples from Mount Pinatubo (Philippines), Volcán Popocatépetl (Mexico), Satsuma-Iwojima (Japan) and Mount St. Helens, Bingham Canyon, Tintic District, and Clear Lake (U.S.A.). The samples reflect a range of crystallization histories and compositions; they range from rhyolite to basalt to trachyandesite, with f(O 2) at the time of eruption ranging from below the fayalite-magnetite + quartz (FMQ) buffer to well above the nickel-nickel oxide (NNO) buffer. Textural and chemical evidence from our suite of samples indicate that sulfides initially were present, but were modified prior to complete cooling of the parent melt, giving rise to Fe-oxide globules. The globules formed through: (1) segregation of an immiscible Fe-SO melt, and possibly, further separation of immiscible Fe-S and Fe-O liquids, and (2) undersaturation with respect to sulfide, causing removal of S from the immiscible sulfide melt. Sulfide undersaturation may have been caused by magma degassing (passively or during eruption), or magma mixing. The recognition of modified magmatic sulfides is important because, with extensive degassing, base and precious metals (e.g., Cu, Au) could be stripped from a melt by a S-rich magmatic volatile phase and entrained into a magmatic-hydrothermal fluid, ultimately giving rise to porphyry-type or related mineralization. For a melt containing 0.01 modal % magmatic sulfides, efficient degassing of only 10 km 3 of magma could yield enough Cu to form a giant deposit.