AimsTo use a sensitive forensic test to measure blood contamination of used Siqveland matrix bands following routine cleaning and sterilisation procedures in general dental practice. Materials and methods Sixteen general dental practices in the West of Scotland participated. Details of instrument cleaning procedures were recorded for each practice. A total of 133 Siqveland matrix bands were recovered following cleaning and sterilisation and were examined for residual blood contamination by the Kastle-Meyer test, a wellrecognised forensic technique. Results Ultrasonic baths were used for the cleaning of 62 (47%) bands and retainers and the remainder (53%) were hand scrubbed prior to autoclaving. Overall, 21% of the matrix bands and 19% of the retainers gave a positive Kastle-Meyer test, indicative of residual blood contamination, following cleaning and sterilisation. In relation to cleaning method, 34% of hand-scrubbed bands and 32% of handscrubbed retainers were positive for residual blood by the Kastle-Meyer test compared with 6% and 3% respectively of ultrasonically cleaned bands and retainers (P < 0.001). Conclusions If Siqveland matrix bands are re-processed in the assembled state, then adequate pre-sterilisation cleaning cannot be achieved reliably. Ultrasonic baths are significantly more effective than hand cleaning for these items of equipment.