Two hypotheses for dike emplacement are: (1) magma flows into and dilates pre-existing fractures; or (2) magma flows into and dilates self-generated fractures. In the first case dikes should be parallel to an element of the rock fabric; in the second, they should be perpendicular to the least compressive stress. The two hypotheses suggest different dike intrusion and fissure eruption mechanisms and therefore different strategies for monitoring igneous events at Long Valley. We derive a method to distinguish the two mechanisms, a priori, from in-situ stress measurements and estimates of magma pressure. Estimates of relative dilation and slip across a dike plane from models constrained by surface displacement data provide a method to distinguish the two mechanisms, a posteriori. Joints cluster near dike contacts just as microcracks cluster near laboratory fractures. Such clusters define a process zone of secondary cracking that forms at the tip of a primary crack. For basaltic dikes in sedimentary host rocks of the Colorado Plateaus, the process zone size is about 10 m and the number of joints is in the range 10 to 100. The mechanical energy release rate for propagation of these dikes is estimated to be 10 to 100 times that for a single laboratory fracture. Data from .proposed drill holes through rhyolite dikes under the Inyo Domes will elucidate propagation mechanisms and process zone characteristics. As a dike nears the Earth's surface, two sets of ground cracks open parallel to the dike trend. A "rule of thumb" is that the depth to the dike top is one-third to one-half the spacing between the innermost surface cracks of each set. Surface structures on and near the Inyo Domes suggest a NNE trend for shallow (< several hundred meters) dike segments, but dome alignment suggests a NNW trend for the feeder dikes at depth.