EBV infects B lymphocytes in vivo and establishes a life-long persistent infection in the host. The latent infection is controlled by EBV-specific MHC class 1-restricted CTL. Immunosuppression reduces CTL activity, and this facilitates outgrowth of EBV+ve B cell lymphoproliferative disease (BLPD). BLPD are aggressive lesions with high mortality. This review presents some key facets in the development of EBV-associated BLPD and in vivo studies on its pathogenesis. The animal models used to date include the common marmoset (Callithrix jacchus), the cottontop tamarin (Saguinus oedipus oedipus), rhesus monkey, murine herpesvirus 68 (MHV68), and the severe combined immunodeficient (scid) mouse, each of which has been used to address particular aspects of EBV biology and BLPD development. Scid mice inoculated i.p. with PBMC from EBV-seropositive individuals develop EBV+ve BLPD-like tumours. Thus this small animal model (hu-PBMC-scid) is currently used by many laboratories to investigate EBV-associated diseases. We and others have studied BLPD pathogenesis in the hu-PBMC-scid model and shown that EBV+ve B cells on their own do not give rise to tumours in this model without inclusion of autologous T cell subsets in the inoculum. Based on the findings that (1) established tumours do not contain T cells and (2) tumour cells express a variety of B cell growth factors, a stepwise model of lymphomagenesis in the scid mouse model can be defined. Additionally, the hu-PBMC-scid model can be used to assess novel therapeutic regimes against BLPD before introduction into a clinical setting.