In exploration, seismology surface sound sources and detectors are utilized to indicate variations in the elastic properties of the subsurface of the earth and from these infer the geological structure. It is helpful to pose the basic problem in a form that is common to most scientific investigations. One possesses data (time series from known seismic sources and detectors) plus a conceptual model (elastic wave propagation) with unknown parameters (the elastic variables as a function of space). The problem is to estimate the unknown parameters so as to get some best fit between the model and the data. At the present time, it seems impractical to attempt to solve this total problem directly. Instead, conceptual approximations are made that lead to a number of consecutive problems, each of which normally involves input data, some particular simple model and unknown parameters that need to be estimated before providing output data. A digital computer is a flexible device with which it is feasible to fit a variety of simple models to data, estimate parameters and transform input data to output data. This has caused a major revolution in the processing of seismic data during the last few years and has undoubtedly improved the over-all effectiveness of seismic exploration. Examples will be given of some of the processing.