HeLa/MV cells which have previously been found to contain measles antigen in more than 90% of cells were very similar to normal HeLa cells in their morphology and growth. Although almost all of HeLa/MV cells may be infected, only 10% of them released infectious virus particles. The hemadsorption test, however, showed that most of the infected cells produced by hemagglutinin. Two kinds of clones were obtained by cloning HeLa/MV cells in the presence of anti-measles serum. One was the virusreleasing clone and the other the uninfected clone which did not contain any measles antigen. The proportions of virus-releasing clones to all clones varied between 20 to 54%, and did not show any increase even after recloning of the virus-releasing clones. The susceptibility of the uninfected clones to the standard measles virus was not different from that of normal HeLa cells.Persistent infection with measles virus was reported by Rustigian [11,12] and Norrby [9]. Rustigian [12] reported that all clones derived from the carrier culture as well as varying proportions of subclones therefrom showed evidence of infection, when tested without addition of antibody. Carrier cultures of a non-cytocidal virus such as parainfluenza virus HA2 were cloned in the presence of antibody and all clones clearly contained viral antigen In the author's laboratory, HeLa cell culture persistently infected with measles virus (HeLa/MV) was obtained as follows (Minagawa, unpublished data). HeLa cells were infected by measles virus, Toyoshima strain. After infection almost all the infected cells were destroyed but a few number of the infected cells survived and multiplied after changing the medium. The cells were transferred and passaged for a few months during which they were unstable and showed cytopathic effects characteristic for measles virus. Thereafter the cells became stable and have been passaged for over 4 years showing measles antigen in more than 90% of cells as detected by fluorescent antibody. In this carrier system , infected cells were able to divide and give rise to virus-releasing clones suggesting that the system might belong to a carrier state 325