In 1956, Wise suggested that retinal hypoxia stimulated retinal neovascularization in the ischemic proliferative retinopathies. Although not directly proven, this theory is strongly supported by a wealth of circumstantial information. Two treatment modalities, vitrectomy and panretinal photocoagulation, have been shown to be effective against retinal neovascularization in diabetics. Both of these treatment modalities improve retinal oxygenation, and we propose that this is the mechanism through which they halt retinal neovascularization. The mechanism for improving retinal oxygenation is different for the two treatment modalities. In the case of panretinal photocoagulation, the new oxygen supply comes from the choroid through the laser scar in the outer retina. In the case of vitrectomy, it comes from the vitreous cavity itself, but the end result is the same. We have expanded Wise's hypothesis to include these two treatment modalities, which were not known at the time of Wise's original paper. © 1990 Springer-Verlag.