Abstract

The ongoing controversy about the possibility of directed mutations in bacteria is examined for its methodological, and thereby philosophical, implications. The method of fluctuation analysis, widely used to investigate whether mutations are random or directed, is described and subjected to a conceptual critique which shows that it cannot decide whether some mutations are directed while most are random. In this context, recent experiments that exploit this possibility to suggest that directed mutations occur in bacteria are described. Interpretive and experimental responses to such claims are briefly analyzed. Finally it is argued that mere statistical analysis, such as fluctuation analysis, cannot resolve this dispute. What will probably be required is an investigation of mechanisms of mutagenesis, an investigation that is intrinsically reductive.