The extent to which marine or brackish water conditions of sedimentation are requisite to the later formation of oil, as is suggested in the above quotation, has long been a debatable question. It may be noted that certain oil shales formed in fresh water basins contain abundant organic matter which is undoubtedly suitable for the generation of oil and gas, and that these shales on distillation yield oil essentially like that obtained from oil shales of marine origin; that certain important oil-bearing sands of the younger Appalachian formations were laid down in waters which are believed to have been only slightly saline; that natural gas is present in fresh water basins; and that it has not been demonstrated that salt in appreciable amounts is necessary for the geologic, any more than for the artificial, distillation of oil. Most of the great oil fields have been in regions of marine or other saline water deposits, but it has not been proved that this is a necessary condition. White11 says: "At the present stage of our knowledge, fresh-water basins appearing otherwise to meet the requirements should be wildcatted without prejudice."
The principal oil-bearing horizons in any locality are comparatively few, and it is ordinarily easy to determine by stratigraphic methods the presence or absence of a favorable geologic horizon. By knowing the succession and thicknesses of the beds in a given region it is possible to infer from surface outcrops the approximate depth below the surface at which the desired horizon can be found. To do this, however, the conditions of sedimentation, the initial irregularities of the beds, the structural conditions, including unconformities, and other factors must be studied.
In exploration for oil the determination of the existence and location of the proper horizon is but an initial step. For instance, the oil of the Midcontinent field of the United States is in the beds of the Pennsylvanian, which are known to occupy an enormous area extending from Illinois and Wyoming south to the Gulf of Mexico. This information is clearly not sufficiently specific to limit the location of drill holes. Sometimes