||Tar sand deposits are widely distributed over the United States of America. These deposits vary from small to very large, and some of the latter may prove to be of economic significance. Curiously, virtually all of the largest known deposits occur in Utah, and this state is believed to possess reserves of bitumen totaling 4.2 Gm3 (27 billion barrels) (Ritzma, 1979). However, only a small portion of the Utah reserves are of sufficiently shallow depth to permit recovery by the techniques currently used in Alberta, Canada. The majority will therefore have to be recovered by some in-situ technique. However, such an in-situ process must by some means reduce the very high viscosity, greater than 104 mPa.s (10" cp), of the bitumen. Two techniques to accomplish this have been utilized bythe Laramie Energy Technology Center(United States Department of Energy). These are fireflooding and steamflooding. So far, three recovery projects have been completed by the Laramie Energy Technology Center. All were conducted in the Rimrock Member of the Mesa Verde Formation of the Northwest Asphalt Ridge (Utah) tar sand deposit. Two of the recovery projects, the results of which were particularly significant, will be discussed in this paper. The first was the reverse/ forward-combustion project (TS-2C) conducted from August 28, 1977, to February 27. 1978 (183 days). This project produced 92 m3 (580 barrels) of oil or 25 percent of that originally in-place. The second was the steamflood project (TS-1S) conducted from April 23 to September 29, 1980 (160 days). This project produced 183 m3(l 150 barrels) of oil, or 5.1 percent of that originally in-place. The design and conduct of these two projects have been described in detail in previous publications (Johnson et al., 1980 and 1981. respectively) and will only be discussed briefly here. To evaluate these two processes fully, it is necessary that both the engineering data (pressures, temperatures, flow rates, etc.) and the chemical and physical properties of the products (primarily the oil) be known. Only by being able to relate the properties of the produced oil to the engineering data can the utility of the process be understood. For example, if an operational change is initiated on the surface (e.g., increasing the air or steam injection pressure) the effect of this change on the properties of the produced oil must be noted and understood. Even if no change is evident in the product, that fact is still valuable information.