Using corpora of spoken American English conversations, the present study exam- ines the use of discourse markers in different spoken registers. Three conversational cor- pora were selected for analysis: 12 family conversations, 11 professor-student conversa- tions during office hours, and 10 sen'er-customer conversations. Twelve discourse markers were identified based on previous literature, and their occurrences in context were ana- lyzed using the Monoconc concordancing program. Quantitative and qualitative analyses show that there are considerable differences in the frequency distributions of discourse markers. These distribution patterns are interpreted in light of the functions of each dis- course marker interacting with the typical characteristics ofdifferent conversational regis- ters.