Vision incapacitation and blindness associated with retinal degeneration affect millions of people worldwide. Cell based therapy and specifically transplantation of human adult bone marrow-derived stem cells (hBM-MSCs) present possible treatment strategy. Subretinal transplantation of human or rat BM-MSCs was shown previously to improve retinal function in Royal College Surgeons (RCS) rats. In those studies cells were transplanted via a transscleral-transchoroidal approach, creating a localized subretinal bleb. Limited number cells could be injected and photoreceptor rescue was restricted to areas in proximity to the injection site. Here we describe a new surgical method for subretinal transplantation that facilitates uniform distribution of transplanted cells as a thin layer along most of the subretinal space. We assessed the therapeutic effect of hBM-MSCs on RCS rats when transplanted either subretinally or intravitreally. We also examined whether a second transplantation can prolong the therapeutic effect. A cell suspension of 2.5 × 10(6) cells in 5 μl was injected subretinally or intravitreally in RCS rats at 28 days postnatal. In the subretinal group, hBM-MSCs were transplanted posterior to the limbus in the superotemporal part of the eye through a longitudinal triangular scleral tunnel reaching the choroid. In the intravitreal group, the cells were injected into the superotemporal part of the vitreous cavity. In cross sections of subretinally transplanted eyes, removed 2 h following transplantation, hBM-MSCs were distributed as a near-homogenous thin layer along most of the subretinal space. In some animals the cells were also detected in the choroid. In the intravitreal injection group, hBM-MSCs were clustered in the vitreous cavity. Transplanted cells could be detected up to 2 weeks after transplantation but not at later time points. Retinal function and structure were assessed by electroretinogram (ERG) and histology analysis, respectively. Six weeks post transplantation, the mean maximal scotopic ERG b-wave amplitude response recorded in RCS control eyes was 1.2 μV. By contrast, in transplanted eyes mean responses of 56.4 μV and 66.2 μV were recorded in the intravitreally and subretinally transplanted eyes, respectively. In the subretinal group, retinal function was significantly higher in transplanted compared with control eyes up to 20 weeks following transplantation. By contrast, in the intravitreal group, rescue of retinal function persisted only up to 12 weeks following transplantation. Histological analysis revealed that 8 weeks following subretinal transplantation, the retinas of control eyes were dystrophic, with outer nuclear layer (ONL) containing a single cell layer. An extensive photoreceptor rescue was demonstrated in transplanted eyes at this time point, with 3-4 cell layers in the ONL along the entire retina. A second subretinal transplantation at 70 days postnatal did not enhance or prolong the therapeutic effect of hBM-MSCs. No immunosuppressants were used and long-term safety analysis demonstrated no gross or microscopic adverse effects. Taken together our findings suggest that transplantation of hBM-MSCs as a thin subretinal layer enhances the therapeutic effect and the safety of cell transplantation.