Information on how promptly food recalls of U.S. Food and Drug Administration (FDA)-regulated products are disseminated to retailers is not well documented. Store managers were surveyed after recalls were declared to estimate the proportion aware of a recall, to describe the methods by which they learned of the recall, and to ascertain how they would prefer to be notified of recalls in the future. From 1 January 2008 to 31 December 2009, we identified FDA Class I products recalled because of potential contamination with an infectious agent such as Salmonella, which were sold in New York City. After each recall, a sample of retailers who carried the products was contacted, a standardized questionnaire was administered to store managers, and a sample of stores was inspected to determine if the product had been removed. Among nine recalls evaluated, 85 % (range, 12 to 100 % ) of managers were aware of the recall affecting a product at their store. Chain store managers were more aware of recalls than were independent store managers (93 versus 78%, P < 0.0001). More chain store managers first heard about the recall via e-mail as compared with independent store managers (35 versus 4%, P < 0.0001). E-mail notification was preferred by large chain store managers (38 versus 8%, P < 0.0001); on inspection, chain stores were more likely to have removed the item than were independent stores (85 versus 56%, P = 0.0071). Although recall information reaches many stores, faster electronic notifications are not effective at reaching small, independent stores, which may lack computers or fax machines. Alternate means to disseminate recall notifications rapidly are needed for stores without electronic communication capabilities.