Wheat (Triticum aestivum L.) is one of the most important staple crops world-wide. Gene-derived 'functional' markers (FMs, also called perfect or diagnostic markers elsewhere) would be the ideal tools for marker-assisted breeding of wheat but, at present, their utility is restricted by the limited availability of genes that control agronomic characters. This bottleneck will be overcome in the next decade by ongoing genomics projects. Another obstacle for assigning sequence to trait variation is the high level of linkage disequilibrium (LD) found in elite materials. Thus, although laborious, TILLING seems to be the most promising approach for targeting sites in genes of interest for FM development. Once larger numbers of FMs become available for wheat breeding, they might be useful in exploiting the fixed genetic variation that is present in regions of high LD.