Microinjection of cholinergic agonists into the pontine tegmentum was used to evoke a state which was polygraphically and behaviorally similar to desynchronized (D) sleep. This study was designed to test the hypothesis that the production of this pharmacologically induced D sleep-like state (D-ACh) was dependent upon the anatomical locus of drug administration within the pontine tegmentum. Four dependent variables of D sleep were measured: D latency, D percentage, D duration and D frequency. Multiple regression analysis and analysis of variance were performed to evaluate the relationship between the three-dimensional coordinates of the injection site (posterior, vertical and lateral) and these 4 dependent measures. The intrapontine site of drug administration accounted for a statistically significant amount of the variance in D latency, D percentage and D duration. There was no significant relationship between the anatomical site of saline injection and the dependent measures of D sleep. A significant increase in D frequency following microinjection of cholinergic agonists was found to be independent of injection site. Pontine injection sites which yielded the shortest D latencies were found to be the same sites from which the highest D percentages were evoked. Rostrodorsal pontine tegmental injection sites were most effective in producing the highest percentages of D-ACh with the shortest latencies to onset. Injections made more caudally and ventrally within the pontine tegmentum produced lower percentages of D-ACh with longer latencies to onset. These data suggest the existence of an anatomical gradient within the pontine tegmentum for the cholinoceptive evocation of a D sleep-like state, and further support the concept that D sleep is generated, in part, by pontine cholinergic mechanisms.