Schizophrenia was unknown before the 19th C, then became common in the 19th and 20th C, but now seems on the decrease. The simplest explanation for this is that those with genes predisposing to schizophrenia used to die in infancy, then survived but with increased morbidity, but now survive with better psychological health. The main factor in high rates of IM in the general population was the care that infants received, especially for feeding, and the better the psychological health of the mother and the lower the genetic load, the better the survival. One would therefore predict that the social class of the survivors from long ago would have been of high social class, unlike the present situation where schizophrenia in concentrated in the lower social classes. The situation with infantile autism may well be similar, and indeed there was a striking tendency for cases before the second WW to come from the upper classes. So was the same upper class bias the case for schizophrenia?