Compared with the neuroleptic nonresponsive (NNR) mouse line, the neuroleptic responsive (NR) line has a significantly higher number of striatal cholinergic neurons (Hitzemann et al., 1993). We now report additional information on this genetic association. At the fifth selected generation, a new selection of the NR and NNR lines differed 5-fold in their ED50 values (1 vs. 5 mg/kg) for haloperidol-induced catalepsy and 20% in the number of striatal cholinergic neurons (higher in the NR line). This association was further examined in 10 standard inbred mouse strains; eight of the strains had been crossed to form the heterogeneous stock from which the new NR and NNR lines were selected. In this panel, we detected no significant association between number of cholinergic neurons and haloperidol response. To examine the similarities and differences in the modes of inheritance for the two phenotypes, we formed a full Mendelian cross from the C57BL/6 (B6) and DBA/2 (D2) mouse strains. The B6 and D2 strains differ 9-fold in their haloperidol ED50 values (3.9 vs. 0.4 mg/kg) and more than 30% in the number of cholinergic neurons (higher in the D2 strain). Haloperidol-induced catalepsy was described by a simple additive genetic model; the narrow sense heritability was 0.60. In contrast, for the number of cholinergic neurons, the B6 genotype was dominant and heterosis was detected in the F1 cross. Despite the differences in heritability, among B6D2 F2 individuals, increasing haloperidol sensitivity was associated with increasing numbers of striatal cholinergic neurons. The BXD recombinant inbred series (25 strains) showed a 16-fold range of variation in the haloperidol ED50 and a greater than 50% variation in the number of striatal cholinergic neurons. However, we detected no significant association between haloperidol response and number of cholinergic neurons. Overall, the data suggest that the genetic association between the phenotypes is modest, complex and detectable only with some genetic strategies.