Reliability and availability are very important trends in the development process of distributed systems. In order to improve these features, object replication mechanisms have been introduced. Programming replication policies for a given application is not an easy task, and this is the reason why transparency for the programmer has been one of the most important properties offered by all replication models. However, this transparency for the programmer is not always desirable. In this paper we present a replication model, JReplica, based on Aspect Oriented Programming (AOP). JReplica allows the separated specification of the replication code from the functional behaviour of objects, providing not only a high degree of transparency, as done by previous models, but also the possibility for programmers to introduce new behaviour to specify different fault tolerance requirements. Moreover, the replication aspect has been introduced at design time, and in this way, UML has been extended in order to consider replication issues separately when designing fault tolerance systems.