It is impossible to make any sort of comparative evaluation of the various treatment methods which have been recommended and tried for severe, chronic emotional and behavioural disorders in children and young people. Although many programs exist, and many more have existed and been reported, the case material has seldom been clearly defined, outcome studies have been limited and longer-term follow-up almost nil. To take simply the few programs which have been discussed in this paper, it is not known whether the young people treated in the California Youth Project, Aycliffe School, the Cotswold Community and the Alberta Parent Counsellors program are at all similar. All programs claim to be treating seriously disturbed children, but more detailed descriptions are needed. Achievement Place claims it deals with “pre-delinquent” youths, while clearly St. Charles Youth Treatment Centre, Aycliffe School and the California Youth Project treat serious established delinquents. So it may indeed be true, as Hoghughi (21) has suggested, that methods that work in certain situations are not readily transferred to others. Balbernie (8) seems to be on the right lines when he calls for precise diagnosis with an accurate definition of what the problem is and of who is supposed to be doing what about it, and with what aims. Similarly precise requirements seemed to be the policy of Hoghughi at Aycliffe School, when this was visited. Despite the problems of evaluating the different therapeutic approaches, certain points do seem clear from this review and from visits to centres. 1. In many cases treatment of the seriously disturbed, previously intractable, child is a very long-term proposition. A commitment to work with the boy, girl or family for several years, is often necessary. 2. Improvement achieved in residential settings, and while active treatment is in progress, is not always maintained subsequently. There is need for much more investigation of what determines whether improvement is maintained, but many programs provided little data about the aftercare given and the longer-term follow-up of the children treated. 3. Intensive treatment, whether residential or not, only makes sense in the context of a long-continuing program of management. Yet many programs, even the best ones, seem to work in relative isolation. 4. Sequential treatment seems to have much to recommend it, and is used, though in a somewhat different way, by all the four British programs that were visited. 5. Some severely disturbed children can be treated in alternative family settings, but which ones, and with what long-term results, is quite vague. These programs do however have several advantages: they keep children in the community, if not in their own homes; they avoid the dangers of institutionalization and the contaminant effects of living with a delinquent peer group; and they approximate more closely the sort of situation (that is, normal family life) which treatment should be helping children to adapt. They are also much less expensive than residential treatment. 6. There is a role for secure units. All who are familiar with the clinical group we are discussing are aware of the existence of a sub-group of very aggressive and violent children who must first be contained. Some of these children can only be constructively treated in a highly secure and very well-staffed unit, but in such a setting it seems that there is a prospect of providing them with some real help. The British “Youth Treatment Centre” concept does seem a useful one. Many points are unclear. These include the following: 1. Does family therapy have a significant part to play in these cases? There is suggestive evidence that it may in some, but many of these children have no families, or at least none with whom they are in contact, and often have been in institutions for much of their lives. 2. What future is there in “intermediate treatment” and community work? Is it in any way realistic to expect to help severely disturbed children by work in the community of which they are part? 3. Can a community approach like that of the California Youth Project make a real contribution to the problem? It seems that in many cases it is better than traditional institutional treatment, but that itself has great limitations. 4. Which of the many residential programs that have been tried is best for which type of problem? 5. How can residential programs be integrated with services in the child's own community to best advantage? 6. What should be the longer term aims of treatment? The various reports of different programs rarely consider this. In conclusion, two points stand out. The first is the need for properly planned and executed research into the treatment of these disorders. It is amazing that so much has been spent on treatment and so little on its evaluation. Perhaps residential treatment is often seen more as a way of getting difficult children out of their communities. The second conclusion is that surely more effort should be made to prevent these disorders. Relatively few of the children under consideration have been brought up in stable, loving homes by their two natural parents. The apparently progressive deterioration of family life, the abandonment of children to day care, the abrogation by many parents of real responsibility for their children and the loss of moral values and religious beliefs are alarming features of contemporary life. Bronfenbrenner (12) has recently commented on how “the American family is falling apart”, and expressed alarm about the current tendency of people to do their “own thing”, to the exclusion of the interests of others. While most children seem to be able to grow up healthily even in contemporary society, the number who become severely disturbed seems likely to increase as these changes in society occur. At the very least we should give a high priority to giving the very best alternative care to children deprived of normal family life.