The perception of American English epenthetic and underlying stops (as in prin[t]ce~prints) was examined in a forced-choice identification experiment that controlled for word frequency and familiarity, closure duration and presence of burst. Previous production data have shown durational differences between epenthetic and underlying [t]. The present results support this generalization, but only for familiar words: listeners appear more sensitive to distinctions between short and long closure durations, tending to categorize those with short duration as “nce” words.