The high failure rate of interspecific pregnancy is a major obstacle to the successful interspecific cloning of mammals. Embryo transfer between rats and mice provides a unique model for studying the causes of such failures. Previous research has shown that the upper time limit for the survival of rat embryos in mouse uteri was the seventh day of pregnancy (Day 7). To study the reasons for the failure of interspecific pregnancy between rats and mice, we transferred rat blastocysts into mouse uteri on the third day of pseudopregnancy. Unexpectedly, intact rat embryos could still be observed in mouse uteri on Day 9 and the implantation rate was as high as 30.6%. However, compared with mouse embryos, the further development of transferred rat embryos in mouse uteri was retarded. On Day 10, transferred rat embryos shrank with much blood. From Day 11 on, they lost their intact structure and the recipient uteri developed dropsy. On Day 12, the embryos shrank further and completely separated from the mouse uteri. By Day 13, they had been absorbed without any remains. In an in vitro co-culture (CT) system, the attachment rate of rat embryos on a monolayer of mouse uterine epithelial cells was similar to that of mouse embryos, but the outgrowth rate of rat embryos was significantly lower. Further investigation by gelatin zymography showed that matrix metalloproteinase-2 (MMP-2) and metalloproteinase-9 (MMP-9) activities in transferred rat embryos was significantly less than in mouse embryos. The same result was obtained in the in vitro CT assay. These results suggest that rat embryos can complete adhesion but not the invasion when transferred into mouse uteri. The reduced invasive ability, and especially, the associated reduction of MMP-2 and -9 activity, is one of the reasons for the failure of interspecific pregnancy.