The traditional psychiatric belief that homosexual women are emotionally unstable (3, 10, 21) has been challenged by Armon (1), Freedman (11), and Hopkins (13). The contention that such women are neurotic has typically been voiced by clinicians reporting on their own therapy patients (3, 10, 19). One exception is the recent psychometric investigation by Kenyon (15) who studied a non-clinical group of English homosexual women, and concluded that they were higher in neuroticism than a comparison group of heterosexuals. In contrast to the ‘illness' notion of homosexuality, the authors of three psychometric studies (1, 11, 13) dealing with non-clinical homosexuals and heterosexuals reported that heterosexual women were not better adjusted than homosexuals. The paucity of research in this area is exemplified by the fact that a total of only four studies, noted above, have been found to date by this author. Even the clinical literature, which is replete with case studies and therapeutic discussions concerning male homosexuality, is strikingly sparse in the area of lesbianism (19). The present study was therefore conducted to add to the small body of data we now have on the adjustment of homosexual versus heterosexual women.

The subject completing a repertory grid test records a large number of comparative judgements about himself and people known to him (the elements), using descriptions (constructs) elicited from him. The analysis of this grid provides extensive numerical data requiring interpretation by the tester. In previous work we have reported the use of grids in the investigation of psychiatrically disturbed students (Ryle, 1967, 1969) and have used the grids to predict and assess the effects of psychotherapy (Ryle and Lunghi, 1969). These interpretations of grid data were initially largely subjective and were usually related to clinical formulations based upon a psycho-analytic model. With experience, the assumptions underlying the interpretation of grid data have become more explicit and the time is clearly ripe for these assumptions to be subjected to more rigorous testing.

Psychological factors have long been posited as having a role in determining a person's response to open-heart surgery. As early as 1956 Bolton and Bailey reported a high correlation between a history of psychiatric problems and psychiatric complications following cardiac surgery.

Illicit drug taking, particularly by young people, has increased substantially in recent years. Cannabis offences increased from 51 in 1957 to 4,683 in 1969, and offences involving amphetamines and LSD increased from 1,216 in 1966 to 3,762 in 1969. Most drug taking is undetected by the police, and is hence underestimated by conviction rates. Weiner (1969) found by questionnaire an incidence of drug-taking by London schoolchildren of 5·39 per cent, and Webb (1969) found an incidence of 24 per cent on a university campus.

The aim of this paper is to present an enumerative survey of abuse of heroin and methadone by persons of 30 years of age and under, known to have resided at some time within the Oxford City boundaries during the year 1969.

In the clinical evaluation of drug treatment, medical practitioners have long appreciated the fact that the response does not merely reflect the pharmacological effect of the drug. The manner in which the treatment is presented to the patient by his therapist is important and there is ample proof in the literature of the therapeutic effects of inert substances—the so-called placebo effects (1, 2). In psychiatric practice particularly, the halo effects of such factors are of considerable importance.

Learning is said to be state-dependent when a response which is learned in a given internal state shows disproportionate decrement in retention when the internal state of the learner is changed. Research on this phenomenon has a long history. It was first noted by investigators who were concerned with the problem of whether overt responses were necessary for the formation of classically conditioned responses. Curare was used to eliminate the overt response, and it was found that:

A sedative or tranquillizing effect may be important in the treatment of mild depressive illness not requiring admission to hospital or ECT (Wheatley, 1969), and Hare et al. (1962, 1964) have suggested that antidepressant drugs may exert their effect by virtue of their sedative action. Other workers also have found similar effects from antidepressant drugs and tranquillizers (Overall et al., 1964; Paykel, 1969).

In order to try to overcome the common clinical experience of patients who do not take the drugs prescribed for them (Wilcox et al., 1965) a sustained release form of amitriptyline (Lentizol) has recently been made available. This trial was undertaken to test the efficacy of this formulation given once daily at night compared to standard amitriptyline given thrice daily.

Early studies (21, 22, 25, 12) have pointed to a genetic component in the aetiology of manic-depressive disorders. More recently, careful clinical observation has suggested the division of the affective disorders into two distinct groups: the so-called bipolar or manic-depressive group and the unipolar type with recurrent depressions only (14, 27, 1). Although genetic differences in terms of family risk have been demonstrated between the so-called bipolar and unipolar psychoses, it is not yet evident whether each group constitutes a homogeneous entity. Moreover, the diagnosis of unipolar depressive disease for the authors cited above includes such syndromes as involutional psychotic reactions, psychotic depressive reactions, and probably also psychoneurotic depressions.

Although depression is well known as a sequel to influenza, mania is not. We describe a patient with a severe psychotic illness who first showed features of a confusional state but then developed a manic psychosis. The occurrence of functional affective psychoses during or following systemic physical illness has been well described (Bonhoeffer, 1909; Kraepelin, 1921; Kiloh, 1961; Post, 1965), but the nature of the relationship has never been clear. In some instances, the physical illness may simply act as a form of non-specific stress precipitating the mental disorder; evidence for a more direct patho-physiological relationship has been lacking. When, in addition to a background of physical illness, a patient shows disorientation and clouding of consciousness together with signs of a manic psychosis, the diagnostic problem is especially difficult, particularly so since a confusional state can occur as part of a severe manic illness (Maudsley, 1895; Kraepelin, 1921). Conversely, hypomanic symptoms can occur in an organic psychosis (Bonhoeffer, 1909).

Manic-depressive psychosis is characterized by severe swings of mood, which may be depressive, hypomanic or manic in type: occasionally a circular form is seen, with successive mood fluctuations in both directions in the same individual. Kraepelin indicated the frequency of these various types (depression alone 48·9 per cent, mania alone 16 6 per cent, circular 34·5 per cent), but also emphasized how difficult it is to predict the precise course of the disease in any individual patient. He stated: ‘we cannot speak of even an approximate regularity in the course of the disease’, and ‘the kind and direction of the attack and the intervals do not by any means remain the same in the individual case’ (Kraepelin, 1921).

The classification of depressive illness has long been a source of controversy in British psychiatry. In particular, phenomenological and statistical studies have failed to agree about the existence of a valid distinction between ‘endogenous' and ‘reactive’ depression (Carney, Roth and Garside (2); Eysenck (3); Kendell (7); Maxwell (9)).

We have recently had the opportunity to investigate clinical and behavioural characteristics of a pair of 17-year-old twin boys who were discovered to have the XYY karyotype (18). Since these twins, known to this department since infancy, had been studied at the age of three, early direct findings as well as retrospective data are available regarding their symptoms and development.

From a sex-chromatin survey of patients in hospitals for the subnormal, Hunter (1966) concluded that the great height and large build of XYY (and XXYY) males might present such a frightening picture that when such a person was convicted of an offence, judges and psychiatrists would be biased towards directing them to Special Hospitals in the interests of community safety. In a study of 464 defective delinquents we found 97 inmates who were at least 72 inches tall (Welch, Borgaonkar and Herr, 1967). There were two 47, XYY and four 47, XXY males in this group. Nielsen and Tsuboi (1970) have indicated that patients with tall stature have a comparatively pronounced disposition to character disorder and criminality. However, Baker et al. (1970) believe that the role of extraordinary height in generating both antisocial behaviour and selective commitment to maximum security facilities remains undefined. Although all these reports have been concerned with adult offenders, we considered about a possible relationship between 47, XYY karyotype, height and admission to institutions of juvenile delinquents.

The following study has been made on the background of the finding by Kramp and Nielsen (1968) of enlarged short arms of a chromosome in the 21-22 group in four out of five psychoinfantile members of a family with a case of Down's syndrome.

We are continually nagging patients. In a hospital which specializes in the rehabilitation of the chronic psychiatric patient (Morgan, Cushing and Manton, 1965) this is understandable. It goes on informally between nurse or doctor and patient, and at formal case conferences. Good advice is given on these occasions by a variety of people, in a variety of ways, and the patient responds with a variety of results.

Among the theories of making rational choices under uncertainty, the one which originated with an eighteenth-century English clergyman, the Reverend Thomas Bayes, and which uses observed evidence to modify prior judgements concerning different possible ‘causes' of the evidence, is logically nearest to the paradigm of clinical diagnosis. A simple illustration of this theorem is the classical Bertrand Box problem (Fry, 1928), and its wider implications for decision making have been discussed by Birnbaum and Maxwell (1960). An example of the use of the theorem in clinical classification has been given by Maxwell (1961).

In spite of the importance of sleep disorder in clinical practice, and the widespread use of hypnotics (Johnson and Clift, 1968), no simple technique for objectively recording sleep patterns in clinical situations is available. EEG recording during sleep (Oswald, 1969) or the use of motility beds (Hinton and Marley, 1959) are sophisticated techniques for use largely in laboratory conditions. Many clinical investigations of sleep disorders (Haider, 1968) or trials of hypnotics (Matthew et al., 1969), rely upon subjective assessments of the quality and duration of sleep by the patient, together with a night nurse's assessment largely based upon observed restlessness. Errors in the estimation of duration of sleep are particularly likely to occur in depressed patients where the illness itself produces an altered awareness of the passage of time (Orme, 1969), while nursing observations on sleep are unreliable (Kupfer et al., 1970). Similarly patients suffering from insomnia due to chronic pain or secondary to diffuse organic cerebral disease may misjudge the passage of time and their period of wakefulness during the hours of isolation during the night. Another group who particularly misjudge the passage of time in relation to sleep are those obsessional personalities, with overvalued ideas about the necessity for adequate sleep, who account for so many of the requests for hypnotics in general practice. Hypnotic drug-dependent patients overvalue their insomnia during their withdrawal from drugs, and again make inaccurate assessment of their sleep patterns.