A review of data on haemolytic disease of the newborn (HDN) collected in Newcastle upon Tyne over 25 years revealed 194 pregnancies in which anti-Kell was the only antibody detected. Sixteen affected babies were born. None was hydropic, three had very severe disease but all survived. There were also three stillbirths, none of which had post-mortem appearances of HDN. The highest recorded anti-Kell titres in individual patients ranged from 1/1 to 1/2048 and bore no relation to the severity of the disease. Of the eight pregnancies in which amniotic fluid examination predicted a high risk of stillbirth, half resulted in unaffected babies. We suggest that haemolytic disease caused by anti-Kell is less severe than suggested by some workers. The use of guidelines developed from the study of Rhesus disease to determine the need for intervention in women with anti-Kell may be inappropriate.