Previous studies showed that nonsense mutations in either of two genes (capR or capS) or an undefined mutation in a third gene (capT) led to pleiotropic effects: (i) increased capsular polysaccharide synthesis (mucoid phenotype); (ii) increased synthesis of enzymes specified by at least four spatially separated operons involved in synthesis of capsular polysaccharide including the product of the galE gene, UDP-galactose-4-epimerase (EC 5.1.3.2) in capR mutants. The present study demonstrated that the entire galactose (gal) operon (galE, galT, and galK) is derepressed by mutations in either the capR or the capT genes, but not by mutation in capS. Double mutants (capR9 capT) were no more derepressed than the capR9 mutant, indicating that capR9 and capT regulate the gal operon via a common pathway. Isogenic double mutants containing either galR(+), galR(-), galR(s), or galO(c) in combination with either capR(+) or capR9 were prepared and analyzed for enzymes of the gal operon. The results demonstrated that capR9 caused derepression as compared to capR(+) in all of the combinations. Strains with a galR(s) mutation are not induced, for the gal operon, by any galactose compound including d-fucose, and this was confirmed in the present study using d-fucose. Nevertheless, the derepression of galR(s) capR9 compared to galR(s) capR(+) was four- to sixfold. The same derepression was observed when galR(+)capR9 was compared to galR(+)capR(+). The data eliminate the explanation that internal induction of the gal operon by a galactose derivative was causing increased gal operon enzyme synthesis in capR or capT mutants. Furthermore, the same data suggest that the galR and capR genes are acting independently to derepress the gal operon. A modified model for the structure of the gal operon is proposed to explain these results. The new feature of the model is that two operator sites are suggested, one to combine with the galR repressor and one to combine with the capR repressor.