An optical test for measuring null correctors has been developed that uses a rotationally symmetric computer-generated hologram (CGH) to synthesize the wavefront that would be reflected by a perfect primary mirror. The test of a null lens is performed by measuring the CGH through the null corrector. Agreement between the null lens and the CGH indicates a high probability that both the null lens and the CGH are correct. This paper presents results that confirm the ease and accuracy of the test. Three null correctors for 3.5-m primary mirrors were measured at the Steward Observatory Mirror Lab using this test. Two of the null correctors were shown to have the correct conic constants within the measurement uncertainty of +/- 78 ppm. The test of the third null corrector was limited by gross flaws in the holograms caused by a fabrication error. A proposed solution to the fabrication problem is presented.