S.KEY636] RHEUMATIC DISEASE IN CHILDREN 245 more scratchy than that of a valvular murmur and, moreover, it does Diagnosis not follow the well known murmur paths. It is important also to notice if the friction extends round the edges of the heart (external pericarditis) and between the heart and the lungs. In this event the friction will be influenced by the respirator; as well as Psncars^ltls by the cardiac rhythm. Such external pericarditis adds much to the danger of the future development of adhesive mediastinitls. In association with acute pericarditis there are usual!}' abnormal signs Associated at the base of the left lung, in the lower axilla, and posteriorly. There may be a direct extension of the inflammation from the pericardium to the pleura, giving rise to friction or effusion, i.e. rheumatic pleurisy. Commoner are signs of congestion, broncho-pneumonic consolidation and collapse (partial or complete) in the lower lobe of the left lung. These pulmonary conditions appear to depend upon compression of the lung by the greatly enlarged heart rather than directly upon the presence of acute pericarditis; but in a peculiar form of consolidation running an afebrile course, which he termed the 'rheumatic lung', A, E. Naish demonstrated histological changes of the rheumatic type. This suggests that perhaps rheumatic infection pla\ s a part in the production of these various pulmonary conditions. Similar changes at the base of the right lung are rare. Acute pericarditis may subside in the dry stage: then the friction will Pericardia! disappear as the temperature falls and the chiid improves. Often, how- e&USJOn ever, it passes on to a stage of pericardial effusion: then the friction will disappear while the raised temperature persists and the child remains unimproved. Effusion in rheumatic pericarditis in a child is nowadays of liitle Significance practical importance. The effusion is small in amount, it is not in itself a danger or a cause of death, and it is neither necessary nor safe to attempt its aspiration. The dangers of pericarditis, even in its stage of effusion, still come from the immense dilatation of the heurt from the severe associated myocarditis. Consequently the recognition of the presence of effusion is of no great moment, and this is fortunate* because the signs of its presence are equivocal. The condition is one in which there is a small effusion overlying a greatly dilated heart, and the signs of its presence are not distinctive as in tuberculous pericardial effusion, in which there is a large effusion surrounding a heart of normal size. Various physical signs of pericardial effusion are described, hut Signs of in rheumatic cases, for the reasons given, none is of great value. The €ffifshn most suggestive are the great enlargement of the cardiac dullness, especially upwards, the distant character of the heart sounds, and the resistant dullness over the praecordia. Radiological examination should not be undertaken in such gravely ill children. The diagnosis of rheumatic pericardia! effusion is best made not by Diagnosis of these signs but by consideration of the progress of the case. When ' friction disappears while the temperature is sustained and the child unimproved, especially if there is a further increase in the area of cardiac