Response to John Cairns: The Contribution of Transiently Hypermutable Cells to Mutation in Stationary Phase

IN his letter, John Cairns reiterates the model described in his Appendix (Cairns 1999) to Rosche and Foster (1999) and seems to be concerned that we have not adopted it as an explanation for our observations (Bull et al. 2000). His model is that most Lac+ adaptive mutants arise in one transiently mutating cell population while concurrently, secondary (unselected) mutations, and a smaller number of Lac+ adaptive mutants associated with them, arise in a different transiently hypermutating subpopulation. The two populations were proposed to produce mutations via different mechanisms, using different gene products (Rosche and Foster 1999). In fact, we did not reject a multiple-population model (Bull et al. 2000). We favor the simpler model of a single transiently mutable subpopulation producing both Lac+ and secondary mutations (Hall 1990; Torkelson et al. 1997) partly because the data that can be used to distinguish between the two models are very sparse such that other interpretations of those data remain possible. In the absence of more extensive data, a simpler model is more attractive.