The behavior therapists have accepted Mowrer's two-factor theory of avoidance as an experimental paradigm of phobia development. The inadequacy of this theory, reviewed in detail by Herrnstein, has been further emphasized. The adaptive nature of conditioned avoidance responses has been contrasted with the maladaptive nature of phobias. Alternative theories to the conditioning one have been briefly noted.1 Requests for reprints should be sent to C.

In the paper criticized by Wolpe and by Powell and Lumia, it was argued that conditioned avoidance responses did not provide a plausible experimental paradigm of phobias. These critics have not concerned themselves directly with this argument. Instead, they argue in favor of classically conditioned emotional responses, self-punitive behavior, and learned helplessness as paradigms of phobia acquisition and maintenance. This is of no help in any further valuation of the original proposal.

hospital psychiatric unit within a specified ten-month period were examined. A multiple regression equation (7) was used to determine the relationship of the following variables to the administration of ECT: sex, month of hospital admission, age, marital status, religion, education, method of admission, occupation and diagnosis.The composite variable made up of these predictors with administration of ECT as the criterion produced an R" of 0.073, i.e. accounted for only 7 per cent of the variance. With diagnosis eliminated as a predictor, an R 2 of only 0.014 was obtained (F = 1.278; df = 13; p = 0.226). A second analysis using the patients' physicians as an added predictor produced an R" of 0.139. Using doctors as the sole predictors an R" of 0.090 was obtained (F = 1.075; df = 13; P = 0.382). Thus the data indicate that neither the diagnosis, the doctor nor the non-illness variables investigated are significantly related to the administration of ECT. As these analyses yielded no significant data the raw data were re-examined.A Chi-square (6) was done to examine the relationship between depressed patients who received ECT (twenty-seven) and those who did not (twenty-five) versus those with other diagnoses who received ECT (twentyfive) and those with other diagnoses who did not receive ECT (one hundred and thirty-seven). The relationship was found to be not significant (x" = 2.28 p > .05).Within the depressive category the relationship between reactive and endogenous depression and ECT administration was examined. ECT was administered to eighteen patients with reactive depression and to nine with endogenous depression -it was not administered to fifty-four patients with reactive depression or to thirty-nine with endogenous depression. A Chi-square showed the relationship to be not significant (x" = 0.32, P > .05). Four depressive 325

Recent reports in the lay literature and professional house publications both have brought into question the validity of electroconvulsive shock as a therapy, and highlighted the iatrogenic effects of the treatment. The failure to complete a study of the therapeutic effectiveness of ECT is reported. The study incorporated two conditions thought to be essential for an adequate evaluation of the treatment: a placebo-ECT group; and double-blind procedures. The failure of the study revealed the difficulty of conducting to completion an adequately controlled investigation of a treatment already accepted by clinicians. Other placebo-controlled double-blind studies are reviewed, and alternative study designs are discussed.

The Pilowsky Questionnaire and a number of self-report measures of general depression were administered to 95 patients, mostly consecutive admissions to the psychiatric unit of a general hospital. There were high correlations (mean = .72) between the scores on the measures of general depression and the scores on the reactive and endogenous scales of the Pilowsky Questionnaire. The correlations between the scores on the measures of general depression and the classification score from the Pilowsky Questionnaire were generally low (mean = .33). The implications of these findings for the issue concerning the unitary or binary nature of depression are discussed .