The content of 472U to 472C revertant virus in serotype 3 oral poliomyelitis monovalent bulk vaccines can be quantified by MAPREC (Mutant Analysis by PCR and Restriction Enzyme Cleavage). Besides other wildtype reversions identified in propagated type 3 Sabin strain populations, the 472U to 472C reversion correlates most prominently with neurovirulence in the monkey neurovirulence test. Therefore, the results can be used for the discrimination of 'good' and 'bad' vaccines on the molecular level. In international collaborative studies it has been well established that vaccine lots containing revertant genomes below a critical threshold pass the in vivo monkey neurovirulence test (MNVT), while vaccine lots containing more revertants fail the MNVT. In this communication we show that the MAPREC test is a sensitive tool for quality control and the demonstration of consistency in large scale production. Furthermore, MAPREC offers a possibility to assess the effect of changed production conditions on the rate of reversion and to find conditions for consistent production with low reversion rates.