Abstract
A comparison between a genetic and a complementation map of the pyrimidine-3 locus of Neurospora is presented. The exceptions to co-linearity are numerous. Several pairs of mutants shown to be allelic by one criterion are nonallelic by the other criterion; strikingly, homoallelic multisite mutants complement. One possible explanation of this finding is discussed. The complementation pattern of one mutant can be changed by ultraviolet induction of a mutation that resembles an intragenic-suppressor. The new mutation has not been separated from the pyr-3 locus in more than 750 progeny tested thus far. It is specific for the KS-23-KS-43 combination, and it does not change either the reverse mutation rate of KS-23 or the map distances between KS-23 and other pyr-3 mutants.