A fluid mechanical hypothesis for the formation of salt domes is presented, for which the basic assumptions are: (1) that the prime motive force for the formation of domes is the density difference between the salt and the surrounding sediments; and (2) that both the salt and the surrounding sediments behave as highly viscous liquids and slowly flow through long geological time.
A simple analysis of the behavior to be expected under the above assumptions shows that a "peripheral sink" will be formed. This will cut off the supply of salt flowing into the dome but this cut-off does not depend on the salt being actually pinched off by the meeting of rocks originally above and below the salt. It may occur at any stage in the development of the peripheral sink, depending on the strength or viscosity of the overburden.
Expressions for the volume relations and relative dimensions of the dome are given in terms of the thickness of the salt and radius of the peripheral sink. Several numerical examples are tabulated.
A series of qualitative experiments show the flow, under a wide range of relative viscosities, of two liquids of different densities with the lighter liquid originally below the heavier liquid. The experiments illustrate the modifications of the flow produced by the peripheral sink and the manner in which the cut-off by the peripheral sink is controlled by the relative viscosities of the two liquids involved.
A series of diagrams is presented to show the hypothetical history of the formation of domes by fluid flow with the further assumption that the flow takes place in a time comparable with the time of deposition of the overlying sediments. From these diagrams it is evident that some of the most striking geological features of salt domes, such as overhang, rim synclines and down-faulted blocks next to the salt are natural consequences of the fluid hypothesis. Also the wide range in form and volume of salt domes in the Gulf Coast and in the interior of Texas and Louisiana, as indicated by the wide range in gravity effects which they produce, can be reasonably accounted for by the fluid hypothesis outlined.