The authors critically review the problem of Hodgkin's disease (HD) in the light of new morphological, immunohistochemical, kinetic, genotypic, and virological findings. These support the lymphoid origin of neoplastic Hodgkin's and Reed-Sternberg cells, because of regular expression of the CD30 lymphoid activation antigen and frequent detection of B- or T-cell phenotypic and/or genotypic markers. It is possible to hypothesize the release of cytokines by tumoral elements as well as the presence of specific cytokine receptors on their surface. This might explain some clinical and pathological features, such as fever, loss of weight, eosinophilia and attraction of reactive elements that make up the composite cellular milieu of typical HD. Integration of monoclonal EBV in the genoma of neoplastic elements has repeatedly been shown, and this might play an essential role in the pathogenesis of the disease. On the basis of present concepts, the borderlines between HD and some categories of non-Hodgkin's lymphomas--especially the anaplastic large cell forms--have become somewhat blurred. Additional research work in the field of HD is desirable and might pave the way for new and more effective therapeutic approaches, designed on the basis of the natural history of the neoplasm.