ABSTRACT  We report here the results of fetal cell enrichment from maternal blood in 58 pregnant women by the use of magnetic activated cell sorting (MACS) with erythroblast‐specific and/or maternal cell specific antibodies. Two approaches were compared; one‐step MACS to enrich CD71+ (a membrane‐bound marker) or GPA+ (another marker, glycophorin A) fetal cells versus two‐step MACS to deplete CD14+ maternal cells and subsequently to enrich fetal (CD71+ or GPA+) cells. The existence of fetal cells was ensured by both FISH with Y‐specific probes and karyotyping of respective anuniotic and/or chorionic vullus cells, the results being applied for comparison of detection rate for XY fetuses between the two MACS procedures.In 24 (38.8%) of the 58 blood samples examined, Y‐positive cells were observed by FISH, whereas there were 38 true XY fetuses later confirmed by karyotyping, including two cases of 47, XY,+21. On the other hand, in Y‐negative cells by FISH, there were two cases of 47, XX,+18. The average number of cells sorted did not differ among one‐step MACS procedures with anti‐CD14, anti‐CD71 and anti‐GPA antibodies. With the latter, 12 (75%) of 16 Y‐positive fetuses were detected, while only one (20%) of 5 Y‐positive fetuses was detected by two‐step MACS with anti‐CD14/anti‐GPA antibodies. The detection rate significantly varied (p = 0.0024) between the two procedures, although the numbers of cases examined were small. There was no statistical difference (p > 0.05) between one‐step and two‐step MACS with other combinations of antibodies. These findings indicate that one‐step MACS using the anti‐GPA antibody is more effective than two step MACS for enrichment of fetal cells from maternal blood.