This review examines the strengths and weaknesses of animal models of human placentation and pays particular attention to the mouse and non-human primates. Analogies can be drawn between mouse and human in placental cell types and genes controlling placental development. There are, however, substantive differences, including a different mode of implantation, a prominent yolk sac placenta, and fewer placental hormones in the mouse. Crucially, trophoblast invasion is very limited in the mouse and transformation of uterine arteries depends on maternal factors. The mouse also has a short gestation and delivers poorly developed young. Guinea pig is a good alternative rodent model and among the few species known to develop pregnancy toxaemia. The sheep is well established as a model in fetal physiology but is of limited value for placental research. The ovine placenta is epitheliochorial, there is no trophoblast invasion of uterine vessels, and the immunology of pregnancy may be quite different. We conclude that continued research on non-human primates is needed to clarify embryonic-endometrial interactions. The interstitial implantation of human is unusual, but the initial interaction between trophoblast and endometrium is similar in macaques and baboons, as is the subsequent lacunar stage. The absence of interstitial trophoblast cells in the monkey is an important difference from human placentation. However, there is a strong resemblance in the way spiral arteries are invaded and transformed in the macaque, baboon and human. Non-human primates are therefore important models for understanding the dysfunction that has been linked to pre-eclampsia and fetal growth restriction. Models that are likely to be established in the wake of comparative genomics include the marmoset, tree shrew, hedgehog tenrec and nine-banded armadillo.