The estimation of two parameters in the transmission of Babesia microti by the tick Ixodes trianguliceps amongst small mammals, (1) the duration of infectivity in natural hosts and (2) the probability of transmission from an infected to a susceptible vole, is described. When B. microti was maintained by direct tick transmission, the probability of a complete cycle of transmission via the larval-nymphal and nymphal-adult transstadial routes was 1·0 and 0·71 respectively, but only if the larvae or nymphs had engorged, as distinct from feeding slowly, while the source parasitaemia exceeded 2 or 0·2% respectively, but had not yet passed the peak level. The duration of this condition for infectivity in voles infected by nymphal bites was only 1–4 days, whilst infections delivered by adult ticks barely reached the threshold level necessary for successful transmission. When syringe passage was introduced into the parasite maintenance schedule (a) the probability of transmission declined markedly and (b) the time-course of the parasitaemia was altered. If these parameter values are put into a simple model, together with field data on tick and host survival rates, it becomes apparent that additional factors, such as the highly aggregated distribution of ticks on their hosts, must account for the maintenance of B. microti at the levels seen in wild small mammal populations.