Statisticians R.A. Fisher and Joseph Berkson have become infamous for ending up on the "wrong" side of the debate over the evidence linking smoking and lung cancer during the 1950s, and scholars have speculated about their personal motives in the controversy. But there were many senior biostatisticians and epidemiologists voicing similar concerns about the quality of the evidence at the time, albeit with less inflammatory rhetoric. This debate occurred during a time when epidemiological research methods commonly used today were understood by few and were only just beginning to work their way into public health and medicine. All of the participants in the debate over smoking and lung cancer saw the need for explicit and rigorous standards for evaluating etiological hypotheses, but they held conflicting views about what those standards should be. The differing opinions on the evidence reflected two different models of etiological research--controlled experiment as the crucial, objective test of a causal hypothesis versus inferential judgment based on a diverse body of evidence. This debate has relevance for current epidemiological practice, as tension between these two views still remains.