Species distribution models are used increasingly in biogeography and conservation biology to predict how species are distributed and to understand attributes of species' environmental requirements. Whilst previous work has established that standard statistical approaches, such as General Linear Models anc General Additive Models, can be useful in modelling species' distribution classification trees are both more flexible and perform better in a range of scenarios and are thus being adopted as 'state-of-the-art'. In this thesis I investigate the relative performance of a range of four different classification tree modelling techniques namely CART, Bagging, Random Forests and Boosted Regression Trees.