A number of experimental studies have demonstrated significant responses of s. c. solid tumours to flavone acetic acid (FAA). Clinical studies to date have been disappointing, with no objective responses being seen. The present study demonstrated that the tumour site is important for the anit-tumour action of FAA against two transplantable adenocarcinoma lines (MAC) in NMRI mice. Responses were achievable only when the tumours were implanted s. c. Ascitic or systemic tumours did not respond to FAA. Experimentally achievable plasma levels of FAA were not sufficient to induce significant cell kills in either MAC 15A or MAC 26 cell lines in vitro. A poor correlation exists between in vitro and in vivo responses, as the clonogenic assay could not predict the respose of the solid MAC tumours grown s. c. The in vitro data indicated that the length of exposure to FAA was important, with long exposure times being necessary for cytotoxicity to develop, in these tumour cell lines. These studies imply that more than one mechanism is involved, and it is likely that the activity of FAA against s. c. tumours relies at least in part on a specific biological feature of tumours in this site. However, it may still be possible to achieve systemic tumour cell kill in vivo by increasing drug-exposure times.