The selection of patients with mitral and aortic valve disease for surgery can usually be made from the clinical, radiological, and electrocardiographic data. In a significantly large proportion of cases, however, more direct measurements of the haemodynamic effects of the valve lesion add valuable information. This communication concerns the use of percutaneous needle puncture of the chambers of the left heart, the aorta, and the pulmonary artery in the assessment of some cases of valvular disease. It is based on an analysis of more than 300 diagnostic procedures.The suprasternal route described by Radner (1955) has been used to reach the aorta, pulmonary artery, and left atrium, and that described by Brock, Milstein, and Ross (1956) (Radner, 1955;Brock et al., 1956). It has been our practice to employ general anaesthesia on the grounds that this makes the procedure more acceptable to the patient as well as providing a steadier haemodynamic state. The measurements can be made in about 10 minutes, and the quality of the pressure records is strikingly superior to that obtained through a catheter.The nmethod has not been widely adopted until recently, presumably because of the potential risks associated with blind needle puncture of the great vessels and heart. In the past two years, however, several authors have reported series with an acceptably small incidence of complications (Willman, Hanlon, Symbas, Kelly, and Mudd, 1964; Hansen, Fabricius, Pedersen, and Sand0e, 1962;Blackburn and Fleming, 1964). Certainly the occurrence of serious incidents is less than that reported after alternative methods of left heart catheterization (Susmano and Carleton, 1964;