To test several hypotheses related to source credibility, comprehension, and homophily of the source, a study was designed. The 124 college students from a basic communication course who served as subjects were presented a message on the effects of slavery in America by either a homophilous source, who wore his hair standing out from his head or who wore very casual clothes, and a heterophilous source, who wore his hair flattened down or who wore formal clothes. After the message, subjects were given a questionnaire containing 12 semantic differential scales measuring source credibility and a 10-question multiple choice test. Results of a 2-by-2 (hair style by dress style) analysis of variance showed that the audience did not perceive the long-haired homophilous speaker as more credible, although they did perceive the casually-dressed homophilous speaker as significantly more credible. No significant effect on comprehension was produced by the independent variables. (SH)