In order to study the natural history and pathogenesis of acute and chronic HCV infections in more detail, infectious clones of HCV RNA have been constructed and used for experimental infection of chimpanzees. A number of lessons have been learned from these experiences. Initially, a full-length viral RNA was constructed from fragments of liver-derived material, but when a full-length cDNA was used for intrahepatic injection of chimpanzees, it was not infectious (Major & Feinstone, 2000). Since the intrahepatic viral RNA originally cloned for this work may have been shorter than a genome length replication intermediate, cloning was done again using a patient's serum as starting material. Compared with the liver-derived material, the serum-derived clone had an additional 98 bp at the 3′ end, within an UTR that was highly conserved among HCV isolates at the primary and secondary structural levels (Tanaka et al., 1995; Kolykhalov et al., 1996), suggesting that these sequences may be important for replication. This was confirmed by showing that some mutations within the 3′ 98 bp untranslated region rendered an otherwise infectious clone uninfectious in chimpanzees (Yanagi et al., 1999). Another striking observation was that RT/PCR amplification of full-length clones from serum were also noninfectious, in all probability because of point mutations introduced during the amplification procedures. Based upon this assumption, a “consensus” clone was made, and only this proved to be infectious. Experimental infection of two chimpanzees with the “consensus” RNA by intrahepatic infection resulted in the establishment of chronic infections with mild (periportal) hepatitis.