Eighteen years after dome-forming eruptions ended in 1986, and with little warning, Mount St. Helens began to erupt again in October 2004. During the ensuing two years, the volcano extruded more than 80×10⁶ m³ of gas-poor, crystal-rich dacite lava. The 2004-6 dacite is remarkably uniform in bulk-rock composition and, at 65 percent SiO₂ , among the richest in silica and most depleted in incompatible elements of the magmas erupted at Mount St. Helens during the past 500 years. Since shortly after the first spine of lava appeared, samples have been collected using a steel box dredge (“Jaws”) suspended 20-35 m below a helicopter and, occasionally, by hand sampling. As of the spring of 2006, 25 age-controlled samples have been collected from the seven spines of the new lava dome. Samples were obtained from both the interiors of spines and from their carapaces, which are composed of fault gouge and cataclasite 1-2 m thick. The dacite lava is crystal rich, with 40-50 percent phenocrysts. The groundmass is extensively crystallized to a cotectic assemblage of quartz, tridymite, and Na- and K-rich feldspar microlites, raising the total crystal content to more than 80 percent on a vesicle-free basis in all but the earliest erupted samples. Early samples and those collected from near the spine margin are more glassy and vesicular that those collected later and from the interior of the spines. Oxide thermobarometer determinations for the earliest erupted samples we collected cluster at temperatures of approximately 850°C and at an oxygen fugacity one log unit above the nickel-nickel oxide (NNO) buffer curve. In contrast, samples from relatively glass-poor samples erupted in late 2004 and early 2005 have zoned oxides with apparent temperatures that range to greater than 950°C. The higher temperatures in these microlite-rich rocks are attributed to latent heat evolved during extensive and rapid groundmass crystallization. Low volatile contents of matrix glasses and presence of tridymite and quartz in the high-silica rhyolite matrix glass indicate extensive shallow (<1 km) crystallization of the matrix, driven by degassing of water and solidifying the magma below the level of the vent. The mode of eruption of the dacite as a series of fault-gouge-mantled spines is explained by this process of extensive subvent degassing and solidification. Although the dacite from this eruption is more silica rich than 1980-86 dome rocks, most major and trace element concentrations of the 1980-86 and 2004-6 magma batches are similar, and magmatic gas emissions have been low and have had similar ratios to those of the 1980s, raising the possibility that the magma might be residual from the 1980–86 reservoir. However, titanium and chromium are enriched slightly relative to the most recent 1980-86 and Goat Rocks (A.D. 1800-1857) eruptive cycles, and heavy rare-earth-element abundances are slightly depleted relative to those erupted during the past 500 years at Mount St. Helens. These data suggest either addition of new gas-poor dacite magma or tapping of a region of the preexisting reservoir that was not erupted previously. A relatively low pressure of last phenocryst growth suggests that the magma was derived from near the apex of the Mount St. Helens magma reservoir at a depth of about 5 km. Viewed in the context of seismic, deformation, and gas-emission data, the petrologic and geochemical data can be explained by ascent of a geochemically distinct batch of magma into the apex of the reservoir during the period 1987-97, followed by upward movement of magma into a new conduit beginning in late September 2004. The question of new versus residual magma has implications for the long-term eruptive behavior of Mount St. Helens, because arrival of a new batch of dacitic magma from the deep crust could herald the beginning of a new long-term cycle of eruptive activity. It is also important to our understanding of what triggered the eruption and its future course. Two hypotheses for triggering are considered: (1) top-down fracturing related to the shallow groundwater system and (2) an increase in reservoir pressure brought about by recent magmatic replenishment. With respect to the future course of the eruption, similarities between textures and character of eruption of the 2004-6 dome and the long-duration (greater than 100 years) pre-1980 summit dome, along with the low eruptive rate of the current eruption, suggest that the eruption could continue sluggishly or intermittently for years to come.