Because half its genes are inherited from the father, the embryo during pregnancy has been frequently compared to an allograft transplanted into the maternal uterus. The question why the embryo is, therefore, not rejected by the mother's immune response in the same way as other allografts has still not been satisfactorily answered. Recent investigations indicate that this analogy may not be as close as it initially appears. One major difference is the expression of an unusual combination of HLA class I molecules by the placental trophoblast cells which invade into uterine decidua during implantation. Furthermore, the lymphoid population in the uterus potentially capable of interacting with these trophoblast HLA class I antigens also appears to be unique. These observations have led to the suggestion that implantation might utilise a novel immune system which is different to that seen in clinical transplantation immunology. Much research effort is now directed towards the elucidation of the mechanisms involved and the outcome of this interaction because this is likely to be how implantation is controlled.