MDV latency is defined as the persistence of the viral genome in the absence of production of infectious virus except during reactivation. A number of systems for studying MDV latency exist, and most involve the use of lymphoblastoid cells or tumors. It has been difficult to divorce latency and transformation. Understanding the relationship between these two states remains a major challenge for the MDV system. Based on their patterns of expression, the MDV LATs are apt to be important in the balance between latent and lytic infections. The LATs are a complex group of transcripts. The profile of gene expression that characterizes latency differs among all herpesviruses, and MDV is no exception. MDV LATs bear little resemblance to LATs of other alphaherpesviruses or to the LATs of other lymphotropic herpesviruses. LAT splicing patterns are complex and the relationships among various spliced species or between these species and the large 10-kb transcript are unknown. In addition, the existence of any protein gene products of significance is unknown at this time. More work is needed to further investigate the significance and function of these RNAs. Better technology to construct mutants in the MDV system is badly needed, since the analysis of mutants in the chicken is a powerful and unique advantage of the MDV system.