In 2 self-paced reading experiments, the authors investigate the processing characteristics of unfamiliar metaphorical subject-predicate structures. The literal first hypothesis predicts that processing metaphorical expressions of the type an x is a y… will proceed more slowly than in the case of literal statements of the same type. This prediction is confirmed: At the position of the metaphorical term, RTs were indeed higher for the metaphorical conditions than for the literal ones. This result was obtained both without (Experiment 1) and with a supportive context sentence (Experiment 2). In Experiment 2, a distinction also emerges between apt and non-apt instances, such that RTs for apt metaphors are no longer significantly higher towards the end of the clause containing them. This suggests that, when embedded in a rich context, the interpretation of unfamiliar apt metaphors can be completed by the end of a fragment that can serve as a clause.