The results described demonstrate the conditions which the gas and oil relative permeabilities are independent of the connote (initial) water saturation. Summary Gas and oil relative permeabilities were measured on water-wet and mixed-wet Berea sandstone at various connate (initial) water saturation values, including zero, using the gas flood method. The centrifuge method was used to extend the measured oil relative permeability curves to lower total liquid saturations than could be permeability curves to lower total liquid saturations than could be achieved by gas flood. Composite cores were used in the gasfloods, whereas individual plugs were tested in the centrifuge. The results show that gas and oil relative permeabilities are independent of the connate water saturation in water-wet and mixed-wet Berea sandstone provided certain conditions are satisfied. Introduction Core preparations for flow tests can be difficult and time consuming. In fact, the preparation often requires more time than the actual now measurements. Among the different preparation considerations, achieving a specific value of initial water saturation in a core is especially difficult. The objective of this study is to quantify the effect that connate water saturation has on gas-oil relative permeabilities for both water-wet and mixed-wet Berea rock. A brief qualitative discussion of the effect that water saturation has on gas-oil relative Permeability measurements was presented by Owens et al. in 1956 . Permeability measurements was presented by Owens et al. in 1956 . Although no data were shown, the authors stated that the presence of water saturation has no effect if the water is immobile. In 1965, Knopp published his results of gasflood tests performed on extracted cores without connate water and with restored water saturations." He observed no effect on the gas-oil relative permeability ratio due to the presence of the water phase. In permeability ratio due to the presence of the water phase. In 1987, Delclaud et al. presented the results of a comparison on a single core sample. No difference in production was observed between the gasflood with low water saturation and the gasflood performed with no water saturation. performed with no water saturation. In this paper, we use the term "water wet" to describe the wettability condition of Berea sandstone as received from the quarry We visualize the water-wet state as one in which a continuous film of water is always present on the rock surface; this film prevents direct contact of the oil with the surface. We use the term "mixed wet" to describe the wettability state visualized by Salathiel, in which the smaller pores are water wet and the larger pores are oil wet. Water cannot come into direct contact with the rock surface in these larger oil-wet pores. Salathiel postulated that the oil-wet pores are connected over an appreciable distance, so that oil can continue to flow at low saturations. Procedures The porous rock used in this study was Berea sandstone. A number of 2- and 2.5-inch-long (5.08- and 6.35m) plugs were cut from a three-foot (0.9-m) length of Berea stock that was 1.5 inches (3.81 cm) in diameter. Test plugs were randomly selected from those obtained from a given three-foot length of stock. Several three-foot lengths were used during this work. Table 1 gives the routine porosity and permeability of the individual plugs. Relative permeabilities were measured by two methods: gasflood and centrifuge. In the gasflood (unsteady-state) method, injected gas displaced oil from the core. The production and differential pressure data were analyzed using the Johnson-Bossler-Naumann (JBN) pressure data were analyzed using the Johnson-Bossler-Naumann (JBN) method to obtain gas and oil relative permeabilities. In the centrifuge method, the core was enclosed in a sleeve and centrifuged in air at constant speed. The volume of oil produced was recorded versus time. The production data were analyzed using a method similar to the technique described by Hagoort to obtain oil relative permeabilities. Gas relative permeability cannot be calculated from centrifuge data because the analysis assumes that no pressure gradient exists in the gas phase.