Feto-maternal transfusion following chorionic villus sampling (CVS) in the first trimester of pregnancy was evaluated by alpha-fetoprotein (AFP) level determination in maternal serum before and after sampling. Some fetal haemorrhage was suggested in 72% of 283 continuing pregnancies by a significant increase of maternal AFP level. Fetal bleeding appeared to stop a short time after CVS, and did not complicate detection of neural tube defects (NTDs) in the second trimester. The change in the maternal serum AFP level was correlated with the size of the chorionic tissue specimen, but no association was observed between fetal and neonatal outcome. The risk of maternal rhesus (Rh) iso-immunization must be taken into account, and anti-D immunoglobulin administrated after CVS. Maternal Rh immunization should be considered as a contraindication to CVS.