In this reply to the criticisms of Fletcher of the three papers published in this issue, we have outlined three studies by Fleischmann and co-workers, which should have been considered by Deutscher and Fletcher in the development of their own work. This neglect of the literature probably accounts for the strange and incorrect conclusions, which these authors have drawn from their investigations. The major cause of these incorrect conclusions appears to lie in the attribution of the macroscopic nucleation rate constant to the initial region of the probability distributions of the elementary stochastic systems. It is this attribution that has led in turn to the postulates of 'nucleation rate dispersion' and 'nucleation persistence', concepts, which appear to us to be physically unsound. In appendix A we have also outlined the relevant parts of two earlier studies: (a) the nucleation of beta-PbO2 in microcrystalline layers of PbSO4 by Fleischmann and Thirsk (b) the double-pulse method for studying the kinetics of electrocrystallisation of alpha-PbO2 on Pt substrates by Fleischmann and Liler. These two investigations have been criticised previously by Deutscher and Fletcher and we ask interested readers to assess whether there could have been any validity in these criticisms. Our own conclusions were that it would not have been useful to refer to the work by Deutscher and Fletcher (nor to revise the texts published in this issue in the light of the conclusions of Deutscher and Fletcher) at the present time as any such reference would have required an extensive discussion of their work. We believed (and we still believe) that such a discussion should have been delayed until proper tests of the ergodicity of the systems had been carried out and the probability distributions of the elementary systems had been related to those of the compound systems, which, in turn determine the macroscopically measured nucleation rate constant.