Identifying the target cells of a hormone is a key step in understanding its function. Once the molecular nature of the receptors for a hormone has been established, researchers can use several techniques to detect these receptors. Here I will review the different tools used over the years to localize melatonin receptors and the problems associated with each of these techniques. The radioligand 2-[125I] iodomelatonin was the first tool to allow localization of melatonin receptors on tissue sections. Once the MT1 and MT2 receptors were cloned, in situ hybridization could be used to detect the messenger RNA for these receptors. The deduced amino acid sequences for MT1 and MT2 receptors allowed the production of peptide immunogens to generate antibodies against the MT1 and MT2 receptors. Finally, transgenic reporters driven by the promoter elements of the MT1 and MT2 genes have been used to map the expression of MT1 and MT2 in the brain and the retina. Several issues have complicated the localization of melatonin receptors and the characterization of melatonin target cells over the last three decades. Melatonin receptors are expressed at low levels, leading to sensitivity issues for their detection. The second problem are specificity issues with antibodies directed against the MT1 and MT2 melatonin receptors. These receptors are G protein-coupled receptors and many antibodies directed against such receptors have been shown to present similar problems concerning their specificity. Despite these specificity problems which start to be seriously addressed by recent studies, antibodies will be important tools in the future to identify and phenotype melatonin target cells. However, we will have to be more stringent than previously when establishing their specificity. The results obtained by these antibodies will have to be confronted and be coherent with results obtained by other techniques.