The early Eocene to the Cretaceous (48-148 Ma) was a period in the Earth's history where the climate was much warmer than the present day, with no permanent ice sheets and atmospheric CO2 levels higher than the present day. Using the climate model HadCM3L coupled to a dynamic vegetation model, this thesis aims to analyse vegetation-climate interactions during these past greenhouse climates, and how the climate, vegetation and climate sensitivity of these time periods are influenced by changes in palaeogeography and CO2 . The results of these model simulations are also evaluated against climatologically-sensitive geological proxies. Past modelling studies for the early Eocene have struggled to model the shallow equator to pole temperature gradient that data suggests was present during this time. However, most models have neglected vegetation feedbacks and incorporating these may help to reduce the model-data discrepancy. In this thesis, vegetation climate interactions during the early Eocene are modelled and analysed, and the results compared to available proxy data. The model-data discrepancies for temperatures are also reduced when vegetation feedbacks were included (compared to simulations with static vegetation), although there are still differences, particularly at high latitudes. This suggests that the models are still missing important processes or the data is not being interpreted correctly. In addition, twelve consistent simulations are carried out , each representing a different stage of the Cretaceous. Each simulation has the same atmospheric CO2 level, allowing the effect of palaeogeography on climate, climate sensitivity and vegetation to be analysed. It was found that, in general, the temperature trends that occurred in the mid-Cretaceous simulations were consistent with data. However, the data record does not extend to the earliest Cretaceous, and in the late Cretaceous the results deviate from the data. The model results suggest that, in order for the model to be consistent with the data there must have been a decline in CO2 from the early to late Cretaceous, which is supported by the CO2 proxy record. More data from the early Cretaceous needs to be collected in order to carry out a more robust model-data comparison for this time period.