This study explores behaviour of subjects in an experiment on dynamic entry games with regime switch as they are used to explain speculative attacks or banking crises. The dynamic nature of the games is implemented by payoffs changing over time in a predictable way such that monetary incentives to enter early or late are controlled. In addition, subjects are always informed about the number of previous entrants. One of the aims of this study is to measure the relative strength of a pre-emption motive in comparison with signalling and information gathering motives and financial incentives for the time of entry. A second goal is to relate the results of dynamic games to their subgame perfect equilibria and to the predictions of the associated one-shot games. We want to know under which conditions one or the other delivers a good prediction of behaviour in the dynamic game. Thirdly, we test some hypotheses regarding the effects of treatment parameters on behaviour.