Gallium 67 scanning in the malignant lymphomas has been done, with variable success, for over 20 years. After initial enthusiasm, the technique fell into disrepute and it was not until the early 1980s that it enjoyed a revival. There have been many major contributions to the literature, both favourable and unfavourable. The reasons for the latter include: poor instrumentation (only single-pulse height analysis), low gallium 67 doses, impatient and careless scanning techniques, timing of the study after treatment (chemotherapy, radiation) and insensitive methods of confirmation of the presence or absence of disease ("truth"). Anatomical diagnostic techniques (computed tomography, plain X-radiography, magnetic resonance imaging and others) are incapable of distinguishing viable tumour in normal-size lymph nodes or necrotic/fibrotic residual masses. With improvements in instrumentation (triple-pulse height analysis, gamma camera resolution and tomographic techniques) gallium 67 can detect active tumour in residual masses and in normal-size nodes. This is due to gallium 67's unique ability to localize in viable tumour cells. It has greater than 90% sensitivity, specificity, accuracy and positive predictive value in patients with lymphoma. Its major contributions are in: staging (changing management of mediastinal disease, obviating the need for a laparotomy and clearly identifying stage IV disease); detecting relapse or residual, progressive disease (it establishes true complete remission and is often the first and only evidence of relapse before clinical evidence); predicting response to therapy (failure to convert to a negative scan post-treatment signals a poor prognosis and alternative therapy is required); and predicting outcome--prognosis (it is the only diagnostic modality to predict outcome accurately).