Objective:To examine the multiplication efficiency Japanese encephalitis virus(JEV)genotype Ⅰ(G Ⅰ) and genotype Ⅲ(GⅢ) of different cell lines which originated from human,porcine,mosquitoes in order to prove mechanism of JEV G Ⅰ replacement JEV GⅢ since it emerging in nature recent decades.Methods:The mixture of Gi and GⅢ JEV isolates was inoculated on human rhabdomyosarcoma(RD).pig kidney epithelial(PS) and Aedes albopictus C6/36 clone(C6/36) which originated from human,porcine and mosquitoes,respectively.Plaque assays were performed to calculate virus titer and real-time RT-PCR with GⅠand GⅢspecific primer sets to quantify the number of GⅠ and GUI RNA copies.Results:The highest virus titer reached at the 3rd day of post infection when G Ⅰand GⅢ mixture was inoculated on RD and PS and that of C636 was at the 4~(th)day.JEVs were amplified and maintained by C6/36 cells after 10 passages whereas that by RD and PS only limited within 8 and 6 passages,respectively.GⅠ strain amplified and maintained more efficiently on C6/36 and PS but not RD.whereas GⅢ strain amplified and maintained more efficiently on RD.Conclusions:There is a correlation between the multiplication efficiency of GⅠ and GⅢ JEV strains when these two genotype strains co-infected on different cell lines with the predominance of GⅠstrains in C6/36 and PS and the limited detection of G 1 strains in RD cells proving a possible mechanism of shift JEV genotypes in nature recent decades since GⅠ emerging. Objective: To examine the multiplication efficiency of Japanese encephalitis virus (JEV) genotype Ⅰ (G Ⅰ) and genotype Ⅲ (G Ⅲ) of different cell lines which originated from human, porcine, mosquitoes in order to prove mechanism of JEV G Ⅰ replacement JEV GⅢ it emerging in nature recent decades. Methods: The mixture of Gi and GIII JEV isolates was inoculated on human rhabdomyosarcoma (RD). Pig kidney epithelial (PS) and Aedes albopictus C6 / 36 clone (C6 / 36) which originated from human, porcine and mosquitoes, respectively. Plaque assay were performed to calculate virus titer and real-time RT-PCR with GⅠ and GⅢ specific primer sets to quantify the number of GⅠ and GUI RNA copies. Results: The highest virus titer reached at the 3rd day of post infection when G I and G III mixture was inoculated on RD and PS and that of C636 was at the 4 th (th) day. JEVs were amplified and maintained by C6 / 36 cells after 10 passages where that by RD and PS only limited within 8 and 6 passages, respectively.G strain amplified and maintained more efficient on C6 / 36 and PS but not RD. whereas G III strain amplified and maintained more efficient on RD. Conclusions: There is a correlation between the multiplication efficiency of GⅠ and G Ⅲ JEV strains when these two genotype strains co- infected on different cell lines with the predominance of G1strains in C6 / 36 and PS and the limited detection of G 1 strains in RD cells proving a possible mechanism of shift JEV genotypes in nature recently decades since GI emergencies.