The glass vessel is first sterilized in the autoclave, and then it must be coated on the inside with a thin layer of paraffin wax. The whole success of this method depends upon this wax coating being absolutely complete right up to the tip of the cannula at the bottom. If the tiniest area of glass be left exposed in the cannula, the process will fail. The production of this perfect wax coating used to be exceedingly difficult of attainment without very frequent practice. The apparatus was first raised to exactly the right temperature; sterile, melted paraffin was then put into it, and distributed evenly over the surface, excess being allowed to run out. The apparatus was then cooled down, and could be put away in a sterile towel ready for use, great care being taken that the lumen of the cannula was patent and not blocked with excess of wax. A simplification of the process was introduced by the use of a saturated solution of wax in ether. This solution is put into the vessel, which must not be heated, and is made to run all over the surface, excess as before being allowed to escape through the lower opening. The ether quickly evaporates, leaving a very thin and perfect film of wax over the surface of the glass. As before, it must be ascertained that the lumen of the cannula is patent. The apparatus is then ready for use. The donor and recipient need not be lying close together, but they must be in the same room. A vein is exposed in the arm of each by dissection under a local anæsthetic. The operator then picks up the vein with a pair of dissecting forceps, and makes an oblique cut into the lumen as in the diagram on p. 131. A flap is thus made which is held in the dissecting forceps in the left hand or is picked up with a fine-pointed pair of artery forceps. The Kimpton’s tube is taken in the right hand, and the point of the cannula is introduced into the vein; that part of the lumen lying opposite the flap serves as a gutter which guides the cannula directly into the lumen, so that it is introduced without any fumbling or delay. The cannula is pushed on so that its widest part engages the whole circumference of the vein, forming a joint through which blood does not leak. The cannula having been pushed well up into the vein, the forceps holding the venous flap may be let go. At the same time an assistant grips the donor’s upper arm, or some form of tourniquet of the necessary degree of tightness is applied, so that the veins become congested without obliteration of the arterial pulse. Blood now flows rapidly into the tube, and the venous pressure is always sufficient to overcome the counter-pressure of the increasing head of fluid in the tube. It is unnecessary, therefore, to produce any negative pressure within the tube with a reversed Higginson’s syringe or an exhaustion pump, which has been used by some workers. Blood is allowed to flow into the tube until the requisite amount has been obtained. The venous