During reuse of formaldehyde sterilized Kiil-dialysers, red cell survival, measured by means of 51Cr t/2, was significantly reduced (p less than 0.001) in 16 patients with anti-N-like positive sera, when compared with 19 antibody negative control patients (mean +/- SD: 16.5 +/- 2.7 versus 22.4 +/- 3.1 days.) In antibody negative patients (n = 10) replacement of formaldehyde sterilized dialysers by ethylene-oxide sterilized disposable dialysers resulted in a significant increase (p less than 0.002) of 51Cr t/2 (Mean +/- SD, days: Kiildialyser 16.3 +/- 1.9; disposable dialyser 20.3 +/- 3.5). This improvement took place, although antibody titres persisted during the 51Cr-measurements and declined thereafter only slowly. In antibody negative patients (n = 6) red cell survival did not increase, when formaldehyde as a sterilant was avoided. In antibody positive patients mean haematocrit rose significantly (p less than 0.05), whereas in none of the antibody negative patients a definite change of haematocrit occurred. The data demonstrate, that formaldehyde sterilisation of dialysers may cause antibody-mediated haemolysis contributing to the extent of renal anaemia. This immunohaemolysis may be corrected, in spite of continuing antibody persistance, when formaldehyde exposure is totally avoided, or possibly when minimized.