Studies of hyperspecific transfer of processing in indirect memory tests are reviewed. A procedure for deriving a comprehensive assessment of priming of indirect memory is then proposed. The procedure is illustrated in Experiment 1, in which prior study of randomly selected words presented with no item-specific context and what could be construed as neutral instructions primed their identification more in a perceptual (fragment completion) task than in a conceptual (semantic cuing) task. Experiments 2 and 3 failed to provide evidence for hyperspecific (i.e., sublexical) transfer of processing in an indirect memory task that called for rapid identification of gradually presented words. Experiment 4 also failed to provide evidence of hyperspecific transfer of processing, despite following more closely the procedure of an experiment (Hayman & Tulving, 1989, Experiment 4) that has provided such evidence. It appears that hyperspecific transfer of processing may be more elusive than sometimes assumed.