The present experiment investigated clustering and free recall when alternative clustering cues were available in the same word list. Specifically, the list composition allowed 5 to cluster items in associatively related pairs, rhyming pairs, or both. Associative clustering was chosen overwhelmingly. Further, though rhyming was used as a cue for clustering, the effect was clear only when there were no other nonsubjective cues. The incidence of rhymed clustering was considerably reduced when another, stronger, organizational basis was available.