932BJ Letters in the absence of knowledge of the kinetics of the individual steps in coupling and uncoupling, we can therefore state that if the number of competing acceptors (uncoupler molecules) exceeds the number of ATP synthase molecules then we always have the possibility of apparent pool behaviour. This will be exacerbated as the turnover number of the ATP synthases is decreased, such that if we need a ratio of 10 uncouplers to 1 ATP synthase to effect uncoupling ofthat synthase we will always obtain behaviour indistinguishable from pool behaviour, whatever the reality (Rich, 1984).In our own photophosphorylation work (cited above), we confined ourselves to situations in which the titre for full uncoupling did not exceed approx. 3 the number of redox chains or ATP synthases present. In the work of Van der Bend et al. (1985), the (extrapolated) titres for full uncoupling (which was not in fact obtained) corresponded to more than 30 times the number of F0F, molecules present.Thus the claim of Van der Bend et al. (1985), using superstoichimetric uncouplers and poorly coupled coreconstituted systems, that the observation of apparent pool behaviour in their titration experiments not only might provide a means of distinguishing localized from delocalized coupling, but indicated delocalized coupling, is erroneous.