Information concerning tape-recorded interviews by police detectives was obtained from nine Metropolitan Police stations in England. Seven key variables were assessed: strength of evidence against the suspect, interviewer perception of offense seriousness, offense type, suspect age, suspect sex, criminal history, and use of legal advice. Three main interrogation outcomes were examined: confession, damaging admission, and exercise of the right to silence. Data from 1,067 cases showed that 16 percent used either selective or complete silence. The use of silence was reliably associated with offense severity, legal advice, criminal history, and police station where suspects were interrogated. The incentive to use silence was greater with more serious offenses. Police officers complained that legal advice consistently encouraged the use of silence by suspects. Silence tended to be used strategically by those who felt compelled or advised to gain what advantage they could from not talking. The use of silence did not directly influence the decision to charge a suspect. Of 174 cases in which silence was employed, 113 suspects (65 percent) were subsequently charged or detained pending further inquiry. The link between the use of silence and question type is discussed, as well as how police officers cope with silent suspects. 8 figures