The identification of activated oncogenes as the basic biochemical difference between tumour cells and normal cells has opened up the possibility for development of truly tumour specific chemotherapy. It may be hypothesized that malignant cells would revert to a more normal phenotype and might even be triggered into terminal differentiation if expression of the appropriate oncogenes were inhibited. Although, at present, it is not possible to anticipate what form future therapy based upon this approach would take, it is clear that the immediate priority must be to establish the general validity of the hypothesis with a variety of tumour cell types in vitro. For this purpose antisense oligonucleotide analogues appear to offer considerable promise as sequence specific inhibitors of oncogene expression. However, no analogue structure yet devised fulfils all the requirements of an ideal antisense effector in terms of biological stability, cell uptake, non-toxicity, hybridization efficiency and mechanism of action on target nucleic acids. Inhibition of oncogene expression in certain cell types has been reported using antisense oligonucleotides but the technique is not universally applicable and more detailed biochemical investigations of the interactions of oligonucleotides with intact cells are required before improved structures may be developed.