Abstract It is pointed out that the application of capillary pressure curves obtainedby drainage or desaturation processes to the calculation of the fluiddistribution in interphase transition zones involves a number of difficulties;namely:the development of very low nonwetting phase saturations appears tobe in contradiction with the lack of mobility of such distributions indicatedby permeability-saturation curves,dispersed non wetting phases arethermodynamically unstable, anddiscontinuous phases should not be subjectto hydrostatic equilibrium requirements. While these difficulties could beobviated by assuming that the capillary pressure drainage curve has an initialhorizontal segment, they are automatically circumvented by application of theimbibition capillary pressure curve to the lower part of the water-oiltransition zone. These generally show zero displacement pressure at onlypartial wetting-phase saturations. The countercurrent upward flow of oil intothe main oil-saturated pay and downward drainage of water also suggests thatwetting-phase imbibition processes will control the saturation distributionimmediately above the water-saturated section. Similar considerations, with ageneralized interpretation of the apparent wetting-phase behavior of the oiland gas phases, provides a basis for constructing the curve for fluiddistribution in the oil-gas transition zone. In the transition zones soderived, the oil begins with a nonvanishing saturation at the water-oil contactand terminates with a similar saturation at the top of the gas-oil contact. Thegas-oil transition zone begins with an equivalent nonvanishing gassaturation. Introduction It has been a common assumption for some years that the nature of the fluiddistribution in virgin reservoirs, and in particular that in the transitionzones between the oil and water and between the oil and gas sections, can becomputed by a simple application of capillary pressure data. The results ofsuch calculations appear to have been first reported by Leverett. The latter, however, indicated only the numerical values of the parameters used in thecomputations, without explicitly describing the procedure. Leverett'sillustrative calculated transition zones are reproduced in Fig 1. Although nocritical study of the apparently obvious method of calculation has beenpublished, one cannot proceed very far in the computation without encounteringsome rather fundamental questions not yet answered in the literature. This paper is not written under the pretense that the whole transition-zoneproblem has been fully and satisfactorily solved. Its purpose, rather, is todiscuss suggestions for its treatment, much of it admittedly hypothetical, andemphasize the nature of the uncertainties arising therein; for the literature, as now available, gives little indication that there are any problems stilloutstanding. T.P. 2405