The purpose of this study was to examine the qualitative variation of human fetal-cord sera (HCS) and to accept the sera in human lVF-ET program. One hundred and sixteenth RCS were tested with 1772 2-cell embryos of F1 (C57BL x CBA) virgin mice, Ten to sixteenth embryos were cultured in m-KRB medium with a aliquot of each serum (10%, v/v) or with bovine serum albumin(O.4%, w/v) as a control medium. Embryonic development were recorded at every 24hr for 4 days by such events as cellular compaction, cavitation, and hatching. In the control groups of eight assays, 98.1%(106/ 108) of 2-ce1l embryos developed above expanded blastocyst and the embryonic development was unified through the tests. But the developmental pattern in medium with each serum was various. Namely, the sera that supported development of 100% 2-cell embryos to above morula, early blastocyst, expanded blastocyst and hatching blastocyst was 45,7%(53/116) , 35.3%(41/116), 15.5%08/116.) and 6.9-%(8/116), respectively. And the sera that supported development of above 80% 2-cell embryos to the each embryonic stage was 92.2% (107/116), 83.6%(97/116), 63.8%(74/116) and 36.2%(42/116), respectively. Meanwhile two kinds of toxic pattern to the embryonic development were observed in some sera. The first pattern is that some sera arrested development of most embryos in pre- or post-stage of morula or blastocyst. The second pattern is that some sera promoted or arrested a part of embryos in the same dish. The ability of serum was depended on the batch of serum. Finally we could accept 69%(80/116) of the tested sera for human IVF-ET program. The base line for acceptance was the ability that supported above 80% 2-ce1l embryos to blastocyst. But some deterious sera were contained in this range. We cut off about 10% of the sera (83.6% , 97/116) that passed the baseline. This final percent of sera was similar to that of grade N of this study.