The nature, incidence and pathogenesis of some clinical features in fifty patients with extreme obesity have been analyzed, utilizing several kinds of relevant information. Exertional dyspnea (84 per cent), joint pains (72 per cent) and somnolence (52 per cent) were common symptoms. The incidence of blood pressure elevation (58 per cent) and peripheral edema (64 per cent) was relatively high, and roentgenographic evidence of cardiomegaly was quite regularly present, as indicated by comparison of the observed transverse cardiac diameter with that predicted for the ideal body weight by the Ungerleider-Clark table. Since one-third of the patients were normotensive, it was concluded that blood pressure elevation is by no means an invariable accompaniment of obesity, even when extreme. No correlation was found between the level of systemic blood pressure and the amount of excess weight. Comparison of blood pressure measurements by direct intra-arterial recording and the standard cuff method yielded falsely elevated values using the cuff 25 per cent of the time, but in only 8 per cent of the cases was this discrepancy large. Actually, the cuff method gave a falsely low estimate as frequently as it did a high one, so that a high reading obtained with the cuff was usually a valid indication of systemic arterial hypertension. The frequency and severity of peripheral edema could not be correlated with the amount of excess weight or with the level of the central venous pressure. Although elevated intra-abdominal pressure and varicose veins may play a role, mechanisms of edema formation in obesity seem poorly defined at present. Incidence and degree of somnolence were not correlated with amount of excess weight, arterial oxygen saturation, central venous pressure, or level of cerebral blood flow and oxygen consumption. Analysis of the relation between arterial carbon dioxide tension and degree of somnolence showed that all six patients with carbon dioxide retention had a tendency to somnolence. However, neither the presence of somnolence nor its severity could be related to the level of arterial carbon dioxide tension. The occurrence of pronounced somnolence without carbon dioxide retention has been taken as evidence denying a causal relationship involving carbon dioxide retention alone. It has been concluded that while carbon dioxide retention may aggravate a tendency to somnolence, the chief mechanisms responsible for development of somnolence in the obese state are not well defined at this time. Polycythemia was not observed in this series, although its occurrence in obese patients is well recognized. On the basis of our experience and published reports of larger series it has been concluded that the incidence of polycythemia is 5 per cent or less in patients with uncomplicated obesity. Mechanisms of dyspnea in the obese state have been reviewed in the light of demands upon the respiratory apparatus and its response. There was poor correlation between the severity of the dyspnea and the observed body weight or the amount of excess weight. A concept of the sequence of events leading to cardiomegaly and congestive heart failure in patients with extreme obesity has been advanced, and the so-called Pickwickian syndrome of obesity has been discussed with particular reference to the development of hypoventilation and cor pulmonale. Presently available data suggest that the incidence of hypoventilation is of the order of 10 per cent in patients with uncomplicated extreme obesity, as found in this study, and therefore this feature is representative at best of only a small segment of this obese population. Congestive heart failure, although uncommon, may occur in extremely obese patients with or without elevation of systemic arterial pressure. It is usually characterized by high cardiac output and insufficiency of both ventricles, predominantly the left. It has been concluded further that there is as yet no convincing evidence to indicate that uncomplicated obesity of extreme degree gives rise to the development of isolated cor pulmonale.