Oil and gas are not at rest in the sedimentary mantle of the earth. They are not in equilibrium, whether they are finely dispersed in a potential source rock or whether they are concentrated in a trap in a reservoir rock. A wide variety of possible escape mechanisms exists; these include diffusion, continuous single phase flow, solution of oil in gas or gas in oil, and solution in water derived from compaction, clay diagenesis, or meteoric sources. The problem is to quantify the possible mechanisms and to rank their relative importance under a given set of physical, chemical, and geologic conditions. The quantitative importance of the various proposed mechanisms can vary by orders of magnitude, depending on the physical, chemical, and geologic conditions.

During the past decade, oil-to-source correlations have become reliable and the timing of peak generation and concomitant migration has been sufficiently quantified to allow the geologist/geochemist to make estimates of when and how much petroleum moved from one location to another. Combined with a knowledge of the physical, chemical, and geologic conditions at the time of migration, such quantitative descriptions of subsurface petroleum transfer permit an empirical test of the applicability of the various proposed migration mechanisms. The application of this technique to selected areas suggests that most of the major commercial oil accumulations of the world left their source rock in a continuous oil phase. When bitumen concentrations in the rock are too low for continuous phase flo to exist, other migration mechanisms, which always are operative, will increase in both absolute and relative intensity. Solution of oil in gas may become significant in thick Tertiary delta systems, and meteoric water may be a surprising asset in some very specific geologic settings. However, it is unlikely that solution of oil in water derived from compaction or from dehydration of clay has much to do with the origin of many of the major oil accumulations of the world.