A number of phonologists have observed the frequent occurrence of ‘devoiced’ obstruents in word-final position in children's utterances. Although such transcription-based accounts have provided evidence of the phenomenon, physical phonetic properties of devoiced consonants have not been described. Therefore, characteristics of voicing control for ‘voiced’ and ‘voiceless’ stops produced by five English-speaking children aged 2; 6–3; 0, five aged 4; 0–4; 6, and five adults were investigated in terms of (1) the percentage of all stop productions evidencing ‘devoicing’ during consonant closure, and (2) in the case of devoiced stops, the proportion of consonant closure evidencing voicing. Both groups of children revealed substantial amounts of consonantal devoicing when compared with the adults; in addition, the proportion of stop closure evidencing voicing was considerably less for the children than for the adults. It was also observed, however, that the children's ‘devoiced’ stops revealed significantly more voicing than their phonemically ‘voiceless’ stop productions.