The efficiency of the fluorescent antibody (FA) test in detecting rabies virus antigen in decomposed specimens was evaluated in simulated conditions of the safety test recommended for the assessment of residual virus in inactivated rabies vaccines. The CVS-infected mice were submitted to different treatments combining time and temperature in order to cause different stages of carcass decomposition and, the FA test was carried out sequentially at pre-determined time intervals. For the materials stored at 25 ºC, greater difficulties for prompt recognition of the inclusion bodies were found after 12 - 18h, whilst the specimens maintained at 4ºC, the inclusions were easily visualized for up to 48h. Brain smears of carcasses kept at -20ºC were suitable for adequate identification after 720 h of storage. In carcasses that had been maintained at 25ºC for 10 h with additional storage at 4 or -20ºC, rabies antigenicity could not be detected, respectively after 10 and 24 h, due to tissue decomposition. The authors recommend that the FA test, when applied as an additional tool for the control of the safety test of inactivated rabies vaccine using mice, care must be taken in order to avoid the use of decomposed materials.