It has been hypothesized that the amnesic effects of alcohol are through selective disruption of hippocampal function. Delay and trace fear conditioning are useful paradigms to investigate hippocampal-dependent and independent forms of memory. With delay fear conditioning, learning of explicit cues does not depend on normal hippocampal function, whereas learning explicit cues in trace fear conditioning does. In both delay and trace fear conditioning, the hippocampus is involved in learning to contextual cues, but it may not be entirely necessary. The present study investigates the effects of alcohol on the acquisition of delay and trace fear conditioning in mice, using freezing as a measure of learning. Male C57BL/6J mice were injected with 0.8 or 1.6 g/kg of 20% v/v alcohol and were immediately exposed to eight tone-footshock pairings in which the conditional stimulus (CS) either coterminated with a footshock unconditional stimulus (US) (delay conditioning) or was separated from the footshock by a 30-s trace interval (trace conditioning). During trace, but not delay fear conditioning, 0.8 g/kg alcohol impaired learning to a tone CS. This dose also impaired context-dependent learning in both procedures (although only slightly for trace fear conditioning). The 1.6 g/kg alcohol exerted a nonselective impairment on learning. The impairment by alcohol of learning to a tone CS when it is hippocampus-dependent, but not when it is hippocampus-independent provides further support for the hypothesis that alcohol exerts a selective effect on hippocampus-dependent learning.