The change in esophageal carcinoma from predominantly squamous carcinoma associated with tobacco and alcohol use to predominantly adenocarcinoma associated with reflux has to be one of the most dramatic epidemiologic shifts in human neoplasia. Despite a heightened awareness, we have only begun our understanding of the pathophysiology of this disease. The study by Sabel et al addresses the long-standing controversy of whether all adenocarcinomas of the esophagus arise within the Barrett epithelium and thus progress through a metaplasia, dysplasia, carcinoma sequence. Unfortunately, the tools presently available to answer this question are crude, namely, demographic and symptomatic differences and comparison of survival statistics. The authors have compared these parameters in patients with esophageal adenocarcinoma with and without surrounding Barrett epithelium. The primary findings of this study were a surprisingly low prevalence of reflux-related symptoms, both with and without identifiable Barrett epithelium; smaller and earlier-stage tumors in patients with identifiable Barrett epithelium; and no significant difference in survival stage for stage. These findings can be readily explained without the need to conclude that there are pathophysiologic differences. The presence of symptoms is proportional to the diligence with which they are looked for, the type of symptoms attributed to reflux (such as asthma and cough), and the fact that patients often minimize reflux symptoms in the presence of new ones related to the cancer. The likelihood that larger tumors overgrow the area of Barrett epithelium confounds this and similar studies, and is unavoidable. Dividing the study groups based on criteria present in the tissue surrounding the tumor can be expected, by chance alone, to result in the absence of the criteria given large tumors. The authors do not give us insight into whether any of the patients who had BE were enrolled in a surveillance program, although it seems unlikely that none of the 66 would have been. If so, this would also result in earlier-stage tumors in the BE group. They conclude that all esophageal adenocarcinomas arise in Barrett epithelium. This is almost certainly true, although proof is lacking. A better understanding of the genetic events leading to BE and its neoplastic progression will one day answer the question.