In this article, subjects diagnosed with alcoholic liver disease are shown to have lower concentrations of several polyunsaturated fatty acids (PUFAs), including 18:2n6, 18:3n6, 20:3n6, 18:3n3, 22:5n3, and 22:6n3, but not 20:4n6 and 22:4n6, nor 22:5n6, in the total lipid extracts of their livers compared with findings for specimens obtained from patients diagnosed with primary biliary cirrhosis and from control subjects. Findings of studies in animals have demonstrated that prolonged alcohol consumption reduces liver polyunsaturate content. However, the effect of ethanol on the elongation/desaturation of essential fatty acids is complex, as in vitro study results indicate that the direction of the effect of alcohol may be related to the dose of alcohol. Findings of studies in hepatocyte cell culture indicate that ethanol increased delta-5 and delta-6 desaturase activities throughout a broad concentration range. In contrast, lower liver desaturase activity has been reported in animals consuming high concentrations of alcohol (36%-40% energy) over a period of several months. Findings from in vivo isotope tracers studies in nonhuman primates and felines indicate that prolonged periods of moderate (mean consumption 2.6 g kg(-1) d(-1) and 1.2 g kg(-1) d(-1), respectively) alcohol consumption had no effect on the uptake of either linoleic (18:2n6) or alpha-linolenic (18:3n3) acids into the plasma and lead to an increased incorporation of these deuterated precursors into 20:4n6 and 22:6n3. Thus, this likely reflects a stimulated, rather than an inhibited, production of long-chain PUFAs. In numerous studies in various species, investigators have documented that alcohol consumption can increase the level of lipid peroxidation in tissues, and sustained periods of ethanol-induced peroxidation can deplete tissues of PUFAs. A hypothesis to rationalize the long-term effects of alcohol consumption on liver PUFA concentration that takes into consideration the effect of ethanol on essential fatty acid metabolism is presented.