IN 1947 GOODALE 1 et alia discovered thrombotic lesions in the hearts of every one of the twentyeight dogs used in catheterization studies. Further work by this same school confirmed these distressing complications. Now comes a report (Ellis et alia) 2 of similar studies carried out under the most careful conditions, to make the experiment comparable to human cardiac catheterization conditions, which confirm the earlier findings. Sixteen dogs were used and nine were found with thrombotic lesions, often including mural thrombosis of the right atrium and ventricle : no lung emboli were found. 'No apparent causal relationship was established between (a) stiffness of catheter and estimated degree of trauma or cardiac size on the one hand, and (b) the presence or absence of lesions on the other.' Johnson 3 reports on mural thrombi in a cyanotic infant who died one month after cathcterization.