To observe form and function changes of vascular endothelial cells (VEC) which were transplanted to the posterior surface of rhesus monkey cornea without Descemet's Membrane by anterior chamber injection, explore the feasibility of transplanting VEC to treat corneal endothelial injury, and find new method of corneal endothelial cell transplantation. Cultured VEC to proliferate in vitro.Rhesus monkeys were randomly divided into two groups: Experimental group (6) and control group (6) according to a random number table. The experimental group:transplant the culured VEC to the posterior surface of rhesus monkey cornea without Descemet's membrane by anterior chamber injection. The control group:Tear out the Descemet's membrane by capsulotomy needle without VEC transplantation. A ultrasound apparatus was adopted to measure the postoperative thicknes of the cornea and Goldman intraocular pressure meter measuring intraocular pressure in the postoperative 7 days, 14 days, 30 days, 60 days, 90 days, organize the data and entered into the computer , applicated the software of SPSS 11.5 for data independent samples t-test and analysis of variance. The eyes were removed respectively in postoperative 30, 60, 90 days to do pathological HE dyeing and scanning electron microscopy (SEM) observation of VEC in the posterior surface of cornea graft. Corneal transpareney:In the experimental observation period (3 months), the experimental group had better transparency than the control group with normal anterior chamber depth and without bullous keratopathy. And the corneal neovascularization was exist in the cornea graft in experimental group in the third month. After 3 months, the corneal thickness of experimental group (500 ± 14) µm was significantly lower than the control group (618 ± 11) µm, Corneal thickness values between experimental and control groups were statistically significant differences in changes (all P < 0.05, t-values were -3.256, -4.419, -12.896 postoperative 1, 2, 3 months). Changes of intraocular pressure difference between the experimental group and the control was not statistically significant (all P > 0.05, t-values were -1.179, -2.166, -2.536 postoperative 1, 2, 3 months). The pathological:the cell layer was visible in the posterior surface of cornea graft. The control group:can't find the cell sample structure.SEM:Experimental group showed that VEC with irregular shape uniformly distributed on the inner surface of cornea and growing well, a small amount of white blood cells can be seen between VEC, and part of cellular debris exist in the trabecular meshwork. Control group showed a fiber material without VEC. Ultrasonic emulsification can established a repeatability and simple model of corneal endothelial injury in Rhesus monkeys.VEC can be transplanted to the corneal surface by Anterior chamber injection and the cells can grow on the surface and play a barrier role in maintaining the state of dehydration and transparency of the cornea to a certain extent. After transplantation, organizational structure and morphology of the anterior chamber angle does not produce pathological effects in the short term. Prompt that transplant the culured VEC to the posterior surface of rhesus monkey cornea without Descemet's Membrane by anterior chamber injection to substitute the function of the corneal endothelial cells may be a new idea for treatment of corneal endothelial damage.