The Yellowstone plateau volcanic field is less than 2 million years old, lies in a region of intense tectonic and hydrothermal activity, and probably has the potential for further volcanic activity. The youngest of three volcanic cycles in the field climaxed 600,000 years ago with a voluminous ashflow eruption and the collapse of two contiguous cauldron blocks. Doming 150,000 years ago, followed by voluminous rhyolitic extrusions as recently as 70,000 years ago, and high convective heat flow at present indicate that the latest phase of volcanism may represent a new magmatic insurgence. These observations, coupled with (i) localized postglacial arcuate faulting beyond the northeast margin of the Yellowstone caldera, (ii) a major gravity low with steep bounding gradients and an amplitude regionally atypical for the elevation of the plateau, (iii) an aeromagnetic low reflecting extensive hydrothermal alteration and possibly indicating the presence of shallow material above its Curie temperature, (iv) only minor shallow seismicity within the caldera (in contrast to a high level of activity in some areas immediately outside), (v) attenuation and change of character of seismic waves crossing the caldera area, and (vi) a strong azimuthal pattern of teleseismic P-wave delays, strongly suggest that a body composed at least partly of magma underlies the region of the rhyolite plateau, including the Tertiary volcanics immediately to its northeast. The Yellowstone field represents the active end of a system of similar volcanic foci that has migrated progressively northeastward for 15 million years along the trace of the eastern Snake River Plain (8). Regional aeromagnetic patterns suggest that this course was guided by the structure of the Precambrian basement. If, as suggested by several investigators (24), the Yellowstone magma body marks a contemporary deep mantle plume, this plume, in its motion relative to the North American plate, would appear to be "navigating" along a fundamental structure in the relatively shallow and brittle lithosphere overhead. The concept that a northeastward-propagating major crustal fracture controls the migration path of the major foci of volcanisim is at least equally favored by existing data, as Smith et al. (19) noted.