The recordings of 47 confession-verified deceptive subjects, 33 verified truthful subjects, 24 nonverified deceptive subjects, and 21 nonverified truthful subjects were analyzed. The recordings of the actual polygraph examinations were analyzed with a PM Analyzer, an audiomicroprocessor that produces an oscillographic envelope display and fundamental frequency of sound. This produced measurements of the voices' fundamental frequencies, amplitudes, and syllabic durations. Digital readings of the speech parameters were statistically analyzed and tested for significance. Findings indicate that the fundamental frequency of the human voice is useful in detecting deception in field situations using the control question theory. Changes in amplitude did not reliably distinguish between people who lied or told the truth, and data did not support the use of the syllabic duration as a means of detecting deception using the control question theory. Findings did show, however, that the syllabic duration for deceptive subjects was significantly greater than the syllabic duration for truthful subjects, irrespective of the type of questions (control-relevant). 46 data tables and 48 references.