Three tumour-specific monoclonal antibodies (MoAbs) showed localisation in human tumour xenografts in nude mice, although the tumour discrimination was limited by the survival of a greater proportion of the MoAb in the blood and body as a whole. An attempt was made to increase tumour discrimination by the subsequent administration of syngeneic idiotypic-specific MoAbs (anti-id) directed against the first antibodies, in the expectation of clearing excess of the first MoAb from the circulation. With one MoAb (NCRC-2), its anti-id (NCRC-60) did effectively clear it from the blood, and, at least within a few hours, the tumour-to-blood ratios were increased. After longer periods, however, the tumour levels of NCRC-2 were also reduced, and the tumour discrimination was no longer increased. With another MoAb (NCRC-23) the tumour levels were reduced to a greater extent than were the blood levels in mice treated with its anti-id (NCRC-59), so that rather than being increased the tumour discrimination was actually reduced to about a third of that in control mice. With a third MoAb (NCRC-48), there was no effect on the tumour or blood levels within a few hours of injection of its anti-id (NCRC-62), and so there was no short-term effect on tumour discrimination. Subsequently, however, the tumour levels were slightly reduced, while the blood levels increased in mice treated with anti-id compared with control mice, so that the tumour-to-blood ratios decreased. These studies have shown a variety of effects of syngeneic anti-id MoAbs on tumour discrimination by their target MoAbs, and overall the indication is that such anti-ids selected indiscriminately are likely to be of little value for enhancing target recognition by murine or human MoAbs given for diagnostic or therapeutic purposes.