To improve the efficiency of targeted gene replacement (TGR), a dual screen (DS) system with gusA gene as negative selective marker (GUS-DS) was developed in Magnaporthe oryzae. First, we tested the endogenous β-glucuronidase (GUS) activities of 78 fungal strains. All tested strains were GUS-, only with 3 exceptions. Whereas, after the gusA being introduced in, M. oryzae, Fusarium oxysporum and Colletotrichum lagenarium acquired high GUS activities. The gusA is thus usable as a selective maker in most fungal species. With gusA as the negative marker, HPH gene as the positive marker, and the peroxisomal targeting signal receptor genes MGPEX5 and MGPEX7 as 2 instances of target genes, we established the GUS-DS system. After transformation, we collected the transformants from hygromycin B screen media and then tested the GUS activities of them. The GUS– ones were selected as potential mutants and checked in succession by PCR and Southern blotting to identify the true mutants and calculate the efficiency of GUS-DS. As a result, GUS-DS improved the screen efficiency for Δmgpex5 from 65.8% to 90.6%, and for Δmgpex7 from 31.2% to 82.8%. In addition, we established a multiple PCR (M-PCR) method for mutant confirmation. By amplifying the different regions at the targeted locus, M-PCR differentiated the wild type, the ectopic transformants and the mutants effectively and rapidly, and had the same reliability as Southern blotting. In conclusion, GUS-DS and M-PCR are useful tools to improve the efficiency of TGR and would be helpful for fungal genomics.