The purpose of this study was to compare the in vivo and in vitro phototoxicity potentials of 13 fragrances. We used the 3T3 neutral red uptake phototoxicity (3T3 NRU PT) test and the photohaemolysis test as in vitro phototoxicity assays. In the 3T3 NRU PT test, all of the fragrances were non-phototoxic. Six fragrances were phototoxic in the photohaemolysis test. Three of the six photohaemolytic fragrances were phototoxic in the guinea-pig photoirritation test. These phototoxic fragrances did not cause cellular phototoxicity, but showed a photohaemolytic reaction. The photohaemolysis test was more sensitive than the 3T3 NRU PT test for screening for the phototoxicity of fragrances. The accuracy of this in vitro phototoxicity test battery was 82%. It is thought that the major phototoxic mechanism of fragrances is cell membrane damage. We suggest that a battery composed of the 3T3 NRU PT test and the photohaemolysis test is a simple and effective model for the in vitro phototoxicity assay of fragrances.