The adequacy of research presented in abstract form for presentation at scientific meetings remains a concern, as indicated by the proportion of such research appearing in scientific journals after peer review. This report examines the publication of research presented at the pediatrics meetings and the effect of a change in selection procedure for abstracts presented at the Ambulatory Pediatric Association meetings for the years 1979 and 1980. Throughout the period, about half of the presentations were published, usually in general pediatric journals, with an average time to publication of about 20 months. The change in selection procedure resulted in little change in the proportion published among either presentations or abstracts submitted but not selected for presentation, but it did lead to a reduction in time to publication. This experience suggests a mechanism for enhancing the timeliness of publication of research presented at scientific meetings without adversely affecting the process of selection.