Abstract
The replication of scientific experiments is discussed stressing the problem of communication between the originator of an experiment and a scientist intending to replicate it. Models of communication are set up, with reference to established fields. A more marginal field is then investigated in the light of these models and it is concluded that scientists in the latter field should not be seen as engaged in replicating original experiment, but in negotiating the rules of replication, and hence the nature of the phenomenon under investigation.