Abstract: The study confirmed the effect of donor and recipient and embryo stage etc on sheep embryo transfer. (1)Poll Dorset (n=163) sheep were treated by superovulation (SOV), the number of embryos was evaluated in virgin ewes (n=76) and in ewes-once delivery (n=87), respectively. The mean number of embryo and useful embryo recovery was significantly higher in virgin ewes than in ewes-once delivery (5.33,3.91 versus 7.55,6.34, P<0.05). SOV efficiency was not significantly different in the spring and in the fall, and repeated-SOV had no significant effect on embryo recovery (P>0.05). (2) Estrus synchronization (ES) rate of recipient (n=876) was 56.37% and 89.02% in the spring and in the fall, utility ratio of recipient was 84.67 % and 88.77%, respectively. (3) 318 recipient pregnancy and 396 lambs were born from total 545 recipients transferred, recipient pregnancy and lambing rates were 58.35% and 57.47% respectively. Among all recipients, small-tail sheep had the highest pregnancy rate (61.2%), the second of pregnancy rate were tong sheep (54.62%) and Mongolia fine-wool sheep (56.58%), No significant differences were observed in different sort of recipient (P>0.05), the corpora lutea number of recipient had some side-effect on pregnancy and lambing rate (although not significantly). the embryo stage also contributed to pregnancy rate, recipient of blastocyst transferred got the highest pregnancy rate (64.76%), significantly higher than that of early morula (P<0.05). There was no significant difference in lambing rate per embryo from the morula stage to blastocyst (P>0.05).