The present investigation seeks to assess to what extent some Dutch phonemes can serve as cues in auditory voice identification of telephone speech. Four Dutch consonants, /x/, /r/, /s/ and /p/, occurring in three different contexts in 24 monosyllabic words, were selected on the basis of their different auditory interspeaker variability. The words were pronounced by 16 speakers in two guises, ordinary chest voice and falsetto. Although none of the consonants appeared to perform significantly better, an interaction was found between type of consonant and position in the word. It also appears that the confidence ratings assigned to the responses vary significantly as a function of the type of consonant and its position in the word.