For some years, the pace of progress in clinical neuroscience has progressively quickened but none more so than with molecular biological techniques. Clinical psychiatrists have been promised (some say forewarned) that the systematic application of these techniques will swiftly cut through the multifactorial aetiologies of many mental illnesses and revolutionise diagnosis, treatment and, possibly prevention. Not surprisingly, given the fact that Down's syndrome and Alzheimer's neuropathological changes (senile plaques and neurofibrillary tangles) are so tightly linked, understanding of Alzheimer's disease (AD) was the first mental illness to benefit from these new methods. Once the amyloid β protein component of the senile plaque had been isolated and its 39–43 constituent amino acids sequenced, then it became almost a routine matter to locate the gene and describe comprehensively the much larger (approximately 710 amino acids) amyloid β protein precursor (APP). Almost simultaneously, the gene responsible for familial pre-senile Alzheimer's disease (FAD) was located, like the APP gene, on chromosome 21 (Tanzi et al, 1989). Soon, a claim was made that these (FAD and APP) were the same gene, and, in a manner akin to the presumed causal gene dosage effects in Down's syndrome, Alzheimer's disease was attributed to excess production of amyloid (by way of APP). However, this was quickly refuted and data to support a gene dosage effect in AD were not confirmed. The trail then seemed to go cold. Several studies indicated that FAD was genetically heterogeneous and distinct from senile AD (St George-Hyslop et al, 1990), and the problems of prion disease in animals and man secured more attention (Westaway et al, 1989).