Abstract
Attention is directed to the age-specific and sex-specific incidence rates of manic depressive psychosis in relation to age. Using Malzberg's (1955) statistics for New York State, it is shown that certain statistical features of this mental disorder closely resemble comparable features of diseases that are widely believed to have an autoimmune aetiology. The age pattern of incidence rates, and the familial evidence, indicate that predisposition to manic depressive psychosis is confined to individuals of a sub-population that is characterized by a particular genotype. There are more females at risk than males. The specific genetic requirements appear to be: one dominant allele at an X-linked locus, and one dominant allele at an autosomal locus. Phenotypic initiation of the disease depends upon the accumulation in a carrier of three specific random events. The average rate of each type of event is constant from 10 years of age (or earlier), to the end of the life span, although the rate of one event is twice as high in females as in males. It is concluded that these random events are most probably some form of somatic gene mutation, and that one mutation affects an X-linked gene while the other two affect an autosomal gene or genes. Burnet's forbidden-clone hypothesis offers a possible explanation as to how a few somatic gene mutations can be "amplified" into a general systemic or mental disorder. The statistics indicate that the average interval, or latent period, between the occurrence of the final initiating somatic mutation and first admission to hospital is about 2.5 years in males and about 5 years in females.