The study reported here employed a mixed factorial design to experimentally investigate the effects of message format on memory for the source of information. Political messages were presented in 3 types of formats: conventional political ads, news-like political ads, and news stories. Memory for the source of information was measured directly after exposure and a week later. The results of the experiment suggest that format and time had a significant effect on memory for the source. Subjects identified the source of information with about the same level of accuracy across formats right after exposure. A week later, subjects were significantly more inept at attributing information contained in news-like ads to its source than doing so for conventional ads and news stories. At that point, information presented in news-like ads was incorrectly attributed to news about 70% of the time.