The effect of blood contamination on the gel-acetylcholinesterase (AChE) test used in the diagnosis of fetal open neural-tube defects was studied with amniotic fluid samples artificially contaminated with fetal or maternal blood in concentrations covering a range exceeding that usually found in clinical practice. Amniotic fluid samples contaminated with maternal blood gave negative gel-AChE results at all concentrations. Contamination with fetal blood yielded positive results if the erythrocyte concentration was greater than about 60 x 10(6) cells/ml. Thus contamination of amniotic fluid with blood is only likely to cause false positive gel-AChE results if this critical concentration is exceeded. Such samples will occur only rarely in clinical practice but when they do the diagnosis should be made with caution.