During a defined critical period of development, if the target of a neuronal population is removed, there is a massive decrease in the number of neurons that survive into adulthood. Previous studies have found that source neuron number is a strictly linear function of target size. The current work extends these observations to the inferior olive-->Purkinje cell projection. Three distinct model systems have been used: (i) lurcher<-->wild-type aggregation chimeras, (ii) staggerer<-->wild-type chimeras and (iii) naturally occurring polymorphisms in Purkinje cell number found in different inbred mouse strains. Total neuron numbers were counted in the inferior olive and plotted as a function of the number of Purkinje cells in the contralateral cerebellar cortex. In lurcher mutants and chimeras, the relationship between these values is well described by a straight line. This suggests that, like the granule-->Purkinje cell circuit, the olive-->Purkinje cell circuit uses a linear algorithm to achieve a numerical balance. The results from the two other model systems were not as clear cut. In the staggerer chimeras, we found only a rough correlation between neuron and target numbers and in the inbred strains there was no discernible relationship at all. These findings indicate that in the final analysis, there are multiple factors involved in the determination of the number of olive cells surviving into adulthood. The potential contribution of sustaining collaterals and afferent inputs is discussed as well as the possible existence of different subcircuits of olivocerebellar connections, each with its own numerical matching function.