There is increasing evidence that steroid hormones derived from vitamin D act through classical nuclear receptors (nVDR), as well as specific binding sites on the plasma membrane of target cells that are coupled to signal transduction systems. These sites are referred to as Membrane Associated, Rapid Response Steroid (MARRS) binding proteins or complexes. In the case of the seco-steroid 1,25-dihydroxyvitamin D3 [1,25(OH)2D3], the two 'receptors' appear to be different proteins with distinct affinities for vitamin D analogs. These differences may be useful in a number of clinical situations. In treating leukemias, it would be useful to promote the actions of the nVDR for differentiation to macrophages while blocking the 1,25D3-MARRS in intestine, which might contribute to the undesirable side effect of hypercalcemia. In contrast, stimulation of the intestinal 1,25D3-MARRS would be desirable in the elderly, since this signalling system appears to decline with age in model systems, potentially contributing to diminished intestinal absorption of calcium and associated bone loss. Bone itself is known to have osteoblasts that respond to 1,25(OH)2D3 through both nVDR and 1,25D3-MARRS mechanisms. Both systems are required for bone-building activities. Osteoclasts lack the nVDR, but may become activated through the 1,25D3-MARRS, offering another site of drug intervention in the treatment of osteoporosis. Finally, during tooth mineralization, immunohistochemical studies reveal an absence of the nVDR and a marked appearance of the 1,25D3-MARRS. In addition to our growing knowledge of 1,25(OH)2D3, the physiological actions of a lesser studied metabolite of vitamin D, 24,25(OH)2D3, are coming to light and may offer additional targets for pharmaceutical modulation.