It is well known that the duration of a stop consonant in the intervocalic, poststressed position may serve as a cue to that stop's voicing characteristic [L. Lisker, Lang. 33, 42–49 (1957)]. More recent research has suggested that when stop-closure durations are examined in a variety of positions, the voiceless stops have greater duration than the voiced stops only in the intervocalic, poststressed position [L. Lisker in A. Valdman [Ed.], Papers in Linguistics and Phonetics to the Memory of Pierre Delattre, 339–343 (1972)]. The present research was designed to provide additional data on stop-closure durations as a function of (1) voicing, (2) stress, (3) vowel context, and (4) place of articulation. Six subjects produced a series of nonsense disyllables of the form CVCVC in a carrier phrase; for each disyllable, one consonant was a “test consonant” and the other two were controls. The test consonant was systematically varied with respect to voicing, stress and position-in-disyllable. Vowels in the disyllable were also varied systematically within the stressed syllable, and included /i/, /l/, u/, and /æ/. Preliminary analyses show that for some subjects, the dorsal stops are affected by stress and position differently than are the apicals and bilabials.