Abstract: Geographic patterns of species diversity and their underlying mechanisms have long been the focus of macro-ecology and biogeography. Recently, the mid-domain effect (MDE) hypothesis has been proposed to explain geographical diversity patterns. The hypothesis states that if the ranges of the species are randomly distributed within a bounded domain then more ranges will overlap near the middle of the domain than at the edges, and thus decreasing species richness will be observed from the mid-domain to the edges. Many studies have shown that the MDE is an important mechanism affecting geographic richness pattern. However, its relative role in such patterns differs markedly depending on many factors. In this paper, we introduced the assumptions and basic models of the hypothesis and illustrate that the models differ in their predictions as a result of different assumptions. We also review the debate on the MDE hypothesis, and discuss the limitations of present mid-domain models. Although the hypothesis has improved our understanding of the effects of geometric constraints and random process on geographic richness gradients, current MDE models are too simplistic to describe biodiversity patterns in the real world. Improvements to mid-domain models should be based on a better understanding of the mechanisms underlying species distribution.