Many mechanisms have been developed to solve the free-rider problem in private public good provision. All those mechanisms have strengths and weaknesses and are, therefore, suited or unsuited for particular public good environments. I present in this work a new class of mechanisms which are, because of their dynamic incentive structure, particularly suited for repeated public good environments. These mechanisms, which I call Conditional Contribution Mechanisms (CCM), give all agents the possibility to condition their contribution on the total level of contribution provided by all agents. I prove for a very general class of environments that all equilibria of these mechanisms are Pareto efficient under a new variant of Better Response Dynamics, called Unexploitable Better Response Dynamics (UBRD). I further present a first experimental study of one mechanism's performance compared to the performance of the Voluntary Contribution Mechanism. In an environment with binary contribution and linear valuations subjects play the mechanisms in a repeated setting. The mechanisms are compared in one case of complete information and homogeneous valuations and in a second case with heterogeneous valuations and incomplete information. In both cases a significantly higher contribution rate can be observed when the CCM is used. Furthermore, all stable outcomes of the CCM, which are observed in the experiment, are in line with the prediction of UBRD.