Cannabinoid ligands have wide ranging neural and behavioral effects; therefore, they are of substantial therapeutic interest. The levels of cannabinoids are tightly controlled in brain infusion and in vitro methodologies, although the studied dose-ranges are extremely wide (e.g. 0.4-470 nmol in brain infusion studies). The brain levels reached after systemic administration are virtually unknown. To investigate this issue, we injected intraperitoneally (3)H-labeled WIN-55,212 and SR141716A (0.3, 1 and 3 mg/kg) and estimated their accumulation in the blood, adipose tissue and brain. Accumulation was dose-dependent. The largest amounts were found in the adipose tissue, while the levels seen in the blood and brain were approximately similar. The accumulation of SR141716A was markedly more pronounced than that of WIN-55,212 in all three tissues. The brain distribution of WIN-55,212 showed large regional differences. Such differences were significant but much smaller with SR141716A. The largest brain levels noticed after intraperitoneal injections did not exceed 2.5 nmol/g. This is larger than the brain level of the endocannabinoid anandamide but smaller than that of 2-arachidonoyl glycerol. Yet, the CB1 receptor affinity of WIN-55,212 and SR-141716A is two orders of magnitude larger than that of 2-arachidonoyl glycerol, suggesting that the exogenously administered compounds were functionally more active. Our findings also suggest that brain infusion and in vitro techniques employing considerably larger doses than 2.5 nmol should be dealt with caution. It appears that measuring brain levels after systemic injections increases our understanding of cannabinoid effects, and provides important clues for the comparison of results obtained with different methodologies.