This research seeks to determine the parameters of the stress cycle of earthquakes along the Central American subduction zone. The work will integrate observed temporal and spatial seismicity patterns with theoretical models of plate subduction. The models will be constrained with bathymetric, seismicity, and fault mechanism data from local and global seismographs. The relative influences of geometry and coupling on observed seismicity will be evaluated via comparisons of the relatively simple regime of the Middle America Trench offshore Mexico with the more complex regions offshore Central America and Chile. This research has as its main goal the definition of the earthquake cycle in anticipation of utilizing the resultant model for prediction of earthquakes at subduction zones. It is therefore directly applicable to the National Earthquake Hazard Reduction Program.