The existence of an important genetic contribution to the aetiology of schizophrenia is well established from genetic epidemiological studies. However, the mode of transmission is complex and non-Mendelian. The main approaches used to identify susceptibility genes are linkage and association studies and the study of cytogenetic abnormalities associated with or linked to schizophrenia. Many linkage studies have been reported but have failed as yet to produce unequivocal, replicated demonstrations of linkage. However, modest evidence for several regions has been reported in more than one data set. Areas implicated include chromosome 22q11-12, 6p24-22, 6q, 8p22-21, 13q14.1-q32 and 1q21-q22, but in every case there are positive as well as negative findings. Most candidate gene studies have been based upon neuropharmacological studies suggesting that abnormalities in monoamine neurotransmission play a role in the aetiology of schizophrenia. Overall, the results have been disappointing, but it should be noted that the sample sizes in many of the older studies would now generally be regarded as inadequate. Finally, recent work has suggested that velo-cardio-facial syndrome (VCFS) is associated with rates of psychosis possibly as high as 30%. VCFS is caused by small interstitial deletions of chromosome 22q11 in 80-85% of individuals. Work is now under way to try and identify whether a gene or genes within the deleted region are of more general relevance to schizophrenia. Future directions in schizophrenia research include collecting larger samples to increase power of findings and applying novel methods for large-scale genotyping of single-nucleotide polymorphisms.