In the first experiment, 72 (35 females and 37 males) introductory psychology students read descriptions of a vehicular homicide trial. Of the five witnesses, three offered circumstantial evidence, and one provided eyewitness testimony. An equal number of jurors read trial descriptions where the eyewitness' age was 6, 10, or 30. Each juror rated each witness' credibility and the defendant's degree of guilt. To determine if the results of the first experiment could be generalized to a different type of trial, the second experiment portrayed a murder trial using a format similar to the first experiment. To determine if the results of the first two experiments would be reproduced in a more realistic trial depiction, the third experiment used a videotaped mock trial based on the scenario of the first experiment. The mock jurors for the third experiment were solicited through newspaper ads and club memberships to produce a sample comparable to actual jury members in the county. Across all three experiments, jurors judged children to be less credible eyewitnesses than adults; however, eyewitness age did not determine the degree of guilt attributed to the defendant. Results were similar regardless of the sample tested (college students versus a more heterogeneous group), the type of trial, or the means used to depict the trial. 2 tables and 32 references.