Vast amounts of geologic and geochemical data bearing on depths of oil origin and primary migration have been published during the past 30 years. Although concepts and understanding have evolved substantially, contrasting views also have been reinforced.

Serious obstacles exist to theories of shallow "in-place" origin of reef oil and tar-sand oil. In reefs, environmental conditions would have dissipated and widely disseminated most of the organic matter. In sands, humic material could have accumulated abundantly, but its transformation to enormous quantities of heavy oil under the thin overburden characteristic of most tar-sand diagenesis is difficult to explain chemically.

Investigations of a variety of modern sediments demonstrate: (1) sparsity of liquid hydrocarbons and free hydrocarbon precursors; (2) absence or traces only of many hydrocarbon and other bitumen components which are common in ancient rocks and crude oil; (3) dilute occurrence of dissolved organic matter and only traces of liquid hydrocarbons in interstitial waters; and (4) major upward movement of water to the surface, representing a serious loss to proposed shallow primary migration mechanisms.

Experimental evidence indicates that effective barriers to mobility of liquid hydrocarbons and hydrocarbon precursors exist in source-type sediments under shallow burial. If hydrocarbons are in oil form, they apparently are immobilized by capillary attraction. In either oil or more finely divided occurrences, hydrocarbons are attracted to the bulk organic matter, which in turn is attracted to clays and other minerals. Hydrocarbon precursors in bitumen are subject to both organic and mineral barriers. At shallow depths, catalytic processes appear inadequate to explain any really significant hydrocarbon formation and primary migration.

It is concluded that most crude oil and its constituents resulted from thermocatalytic transformations and primary migration at depths ranging from a few thousand to about 10,000 ft. Chemical alteration of organic matter should be greatly accelerated in the relatively high temperature-pressure regime at these depths. Energy input from heat and pressure should greatly increase the mobility of bitumen constituents formed, including hydrocarbons. Sufficient water should be available as a migration medium where expandable clays are abundant. Clay mineral transformation should release water and organic matter, and improve source-bed drainage. Large amounts of colloid-forming substances would be produced, and gases would be abundant products of organic alterations. Primary migration of liqu d hydrocarbons in colloidal, true-water, and/or gas solution would obviate the high-permeability requirements of migration in the form of oil.

Thermal investigations of oil shale and disseminated kerogen show that the heavy nonhydrocarbon part of the bitumen is produced initially, and at higher temperatures the bitumen is altered in part to hydrocarbons. The fact that relatively high temperatures are needed for these transformations, even in the presence of catalysts, argues for rather deep corresponding transformations in the sedimentary section. Bitumen contains colloid-forming fractions which would greatly increase the solubility of hydrocarbons in water. Carboxylated (colloid-forming) organic material is relatively resistant to temperature increase in a water-wet environment, even where catalysts (clays) are present. Hence colloids should remain active to considerable depth. Laboratory experiments indicate that thermocat lytic processes can best explain the origin of liquid normal paraffin hydrocarbons of oil from carboxylated organic matter.

Several studies of depth patterns in nonreservoir (source type) rocks of the geologic section reveal that both bitumen and the hydrocarbon component of bitumen increase markedly in the depth range of a few thousand to several thousand feet. The increase in hydrocarbon abundance is particularly impressive. In some examples, a zone of sharply decreasing bitumen and hydrocarbon occurrence has been observed directly below the zone of abundance, indicating that primary migration has evidently taken place. Much primary migration apparently is related to stages of "dehydration" of expanded clay minerals, the most important stage being temperature-dependent.

Coordinated studies of geology, geochemistry, and oil occurrence, particularly in the Soviet Union and northwestern Germany, support major origin-primary migration episodes at depths of a few thousand to several thousand feet or more. They also suggest continuous or recurrent primary migration through rather long time and broad depth intervals.

Ultra-deep origin of mobile oil, and upward migration through thousands of feet of shale are shown to be very unlikely. In examples such as the South Sumatra, Los Angeles, and Ventura basins, where the ultra-deep concept has been applied, the data can be explained better by primary migration at intermediate depths.