The various diseases that follow experimental infection with the acute and non-acute avian oncoviruses are discussed with special reference to the pathogenesis of avian erythroblastosis. One view, based on in vitro studies, sees erythroblastosis as the product of a failure in the differentiation of virus-infected stem cells to mature erythrocytes, as a result of cell 'transformation'. The results of some in vivo studies, however, point to a resemblance of the disease to a haemolytic anaemia, where cellular death is an important component. It seems probable that the disease is the result of transformation of cells of the erythroblastic series followed by the death of many of these cells due to influences that have not yet been determined. Determination of the causes of this cellular death may prove to be as important for our understanding of the problem of leukaemia as the work that has already been accomplished in explaining the causes of cell transformation. It is also suggested that the tendency of gs amino acid sequences of the avian leukosis viruses and mouse leukaemia viruses to form fusion proteins with a variety of proto-oncogenes may be part of a wider phenomenon, and that these sequences may fuse with other proteins, altering their properties. More work is required on the possibility that there is an undiscovered immunological component in the progression of the L/S diseases.