Summary.-Ss in 2 experiments recalled 12 sentences in either the active or passive voice and containing or not containing an adjective modifying the categorizable noun. Both free recall and clustering increased over trials. Sentences not containing an adjective were recalled more frequently than sentences containing an adjective but clustering was not affected by the presence or absence of adjectives. Sentence voice did not affect recall or clustering. Following the fourth recall uial, Ss sorted the sentences into categories. Ss who learned sentences in passive voice more frequently sorted by acted-upon noun. Andre and Kulhavy (1971) found thac the recall and category clustering of sentences containing categorizable nouns appeated to increase over trials in a free-recall experiment. Sentence voice had no effecc on either recall or clustering. However, since Andre and Kulhavy had presented the sentences in constant order and blocked by categories over trials, it is possible that Ss recalled the sentences in the presented order inscead of clustering. Andre and Kulhavy had also expected passive sentences to be clustered more than active sentences since the passive voice emphasized the cluscerable element. However, the blocked presentation may have made the categories so obvious that the effecc of sentence voice was overridden. The present experiments eliminated the ambiguities of the Andre and Kulhavy study and tested the hypothesis that passive sentences produce greater recall and clustering. In addition, since Cofer, et al. (1969) had found that adjective modification of cluscerable nouns reduces recall and clustering, this variable was also manipulated. METHOD The two experiments conducced were alike in all respects save one: the acted-upon nouns thac could be categorized differed. In Exp. 1 these nouns were low-frequency associates of the categories earth formations, four-legged animals, and occupations taken from the Battig and Montague (1969) norms. In Exp. 2 the conceptual subsets medical doctors, farm animals, and depressions in the earth, from the same categories were used. Four groups of Ss recalled 12 declarative sentences whose acted-upon element could be classified into one of the categories given above. Each group learned a different version of the sentences; the modified active and passive groups learned active or passive sentences containing an adjective modifying the acted-upon noun. The noc-modified active and passive groups learned the same 'A version of this paper was read at the American Educational Research Association Convention, Chicago, 1972.