The proper selection of cases, the settling of the suitable position for the tapping, and the choosing of the suitable time for it, are more difficult, and not within the scope of the present work. On these subjects much information may be obtained from the papers of Dr. Bowditch of Boston, of Dr. Hughes and Mr. Cock,[139] and an exceedingly interesting and valuable paper by Dr. Warburton Begbie.[140] _Where_ is it to be performed? Not _above_ the sixth rib, else the opening is not sufficiently dependent; very rarely _below_ the eighth on the right side, and the ninth on the left. The intercostal space generally bulges outwards if fluid is present, and this bulging acts as an aid to diagnosis. As the intercostal artery lies under the lower edge of the upper rib in each space, the trocar should be entered not higher than the middle of the space; and because the artery is largest near the spine, and also the space is there deeply covered with muscle, the tapping should never be _behind_ the angle of the rib. In most of the manuals we are told to select a spot midway between the sternum and spine for the puncture; but Bowditch, Cock, and Begbie, who have had large experience, prefer, and I believe rightly, a position considerably behind this, _an inch_ or two below the angle of the scapula, between the seventh and eighth, or between the eighth and ninth ribs. The operation may be performed with a simple trocar and canula, round, about an eighth of an inch in diameter, and at least two inches in length. The point must be sharp, and it must be pushed in with considerable quickness, so as to penetrate, not merely push forwards, the pleura, which may be tough, and thicker than usual. Once the skin is pierced, the instrument must be directed obliquely upwards, so as to make the opening and position of the trocar dependent. When the trocar is withdrawn the fluid may be allowed to flow so long as it keeps in a full equable stream; whenever it becomes jerky and spasmodic, the canula should be removed _before_ the sucking noise of air entering the chest is heard. In more chronic cases, where the quantity of fluid is large, and especially if it is thick and curdy, the exhausting syringe of Mr. Bowditch is an improvement on the simple trocar and canula. It consists of a powerful syringe, which fits accurately to the trocar with which the puncture is made. There is a stop-cock between the trocar and syringe, and another at right angles to the syringe. The trocar being introduced, it is held firmly in position by an assistant, by means of a strong cross handle; the first stop-cock is then opened, and the syringe worked slowly till it is filled with fluid through the trocar, the other delivery stop-cock being closed. The first is then closed, and the second opened; the syringe is then emptied through the second into a basin. By a repetition of this process, the fluid can be removed at pleasure, without any risk of the entrance of air.