It is probably true that by far the most research in the field of psychiatry today still starts with an end-product, a sick patient, and seeks to unravel the sequence of events, psychological and physiological, that appear to have led to his becoming sick. This results in many suggestive hypotheses but, like any single method of enquiry, has its limitations. One of the hallmarks of an advancing science is exploitation of as many methods as can be devised. When in physiological medicine research was expanded to include the systematic investigation of one or another probable pathogen and its effects, a great harvest of knowledge was garnered. Adolf Meyer, we know, looked forward to the day when the same would be possible in psychiatry. "When we know better what to look out for," he wrote in 1903, "we may undertake studies of developing abnormalities which are not insanity yet, and follow them out so as to accumulate material of actual observationon which to build a solid theory...."Because of its practical and scientific implications, the study of responses to loss of mother figure in the early years might have appealed to Adolf Meyer. On the practical side he might have been attracted by the vision of our becoming able to develop measures to prevent at least some forms of mental ill-health. On the scientific side he would, no doubt, have valued the opportunities that stem from the identification of an experience of childhood that is probably pathogenic, can be clearly defined, and the effects of which on the developing personality can be systematically studied by direct observation.There are, of course, many other experiences of childhood besides loss that there is good reason to believe also contribute to the development of disturbed personality and psychiatric illness. Examples are the child's experience of one or another of the various sorts of parental attitude that have long been the subject of concern and therapeutic endeavour in child psychiatric clinics. For each the research task is, first, to define the experience, secondly, to locate a sample of cases in which it is occurring so that its effects on psychological development may be studied, and, finally, to relate the processes that are found to be set in train by it to processes present in patients with declared illness. The consequences of such an expansion of research are far reaching. It is my hope that the illustration of its adoption that has been given will encourage others to try the same route.Abstract Teaser