Three amplification protocols were analyzed for error rate and generation of polymorphisms during RAPD analysis. Using a set of 240 primers, the protocols detected similar frequencies of polymorphisms in two inbred sugar beet lines. The error rate was investigated by including a 1:1 mixture of DNA from the two lines in all analyses. Similar error rates, approximately 18%, were detected by the three protocols. Thus, altered amplification conditions did not substantially affect the error rate during RAPD analysis. For each of the three possible pairs of protocols, a positive correlation was obtained for primer and number of polymorphisms. Thus, a set of highly polymorphic RAPD primers can be used effectively, without prior screening, to detect polymorphisms for each protocol.