Whilst the potential of gene therapy is considerable, current applications have been restricted by the limitations of available vectors. As yet, no vector is able to produce the desired safe, targeted and efficient transfer of genetic material with regulation of the new gene in the targeted cell. Notwithstanding these limitations, more than 65 clinical gene transfer protocols have been approved in the US. The majority of these are open to patients with malignant disease, in whom the risk:benefit ratio is most appropriate. Current progress and problems in gene transfer are illustrated by reference to gene transfer into haemopoietic stem cells (HSC), an area that has attracted particular attention, both because of the logistic advantages of these cells and because of the wide range of pathologies that may be corrected in the HSC itself or in its progeny. Because of the low efficiency of transfer into HSC, initial studies have involved transfer of marker genes to determine the origin of relapse after autologous bone marrow transplantation and to learn more about the conditions that enhance gene transfer and expression in haemopoietic tissue. Information gained from these studies is already guiding the practice of autologous and allogeneic marrow transplantation and has contributed to the development of gene therapy protocols for the treatment of malignant disease, immune deficiency syndromes and lysosomal storage disorders. Over the next decade, as the technology of gene transfer advances, many further clinical applications of the approach will become evident.