During recent years, DNA microarrays have become the method of choice to monitor the expression level of a large number of genes. Depending on the focus of the study and the method of microarray fabrication, a number of different strategies for probe selection may be most appropriate. One consideration concerns the length of the probe, ranging from some 25 residues used for oligonucleotide arrays to complete cDNAs. Unless resources are truly unlimited, an important decision to be made is the amount of effort to be put into the selection of genes and gene fragments. While high-throughput cDNA arraying projects usually will select from a collection of existing cDNA clones, smaller projects focusing on a number of selected genes can afford to selectively amplify fragments optimised for that purpose. This paper discusses the full scope of probe selection strategies, highlighting the problems that may be encountered in the various systems.