Two-dimensional gel electrophoresis can be an expensive technology, and many studies for this reason are based on a modest number of replicates. However, it is important that the power of the study is sufficient to detect protein expression differences of interest. This paper discusses this matter, reviewing the application of standard power calculation methods and then considering how issues such as the use of regularised tests and spot selection based on false discovery rates and related approaches affect the choice of sample size. The important distinction between biological and technical replication is made, and the superiority of the former emphasised