METHODS.

Full-grow-n rabbits weighing 2-2 kg. from dealers' mixed stocks were used throughout. Save where some variant of the standard procedure is specifically mentioned. each individual test has made use of an independent pair of rabbits: a donor (D) and a recipient (R).(In one of the two groups of eye-implantation experiments. most of the recipients were linked together in pairs by sharing a donor between them. In these particular experiments the variable under investigation was the degree of vascularization of the grafts. As this is a function of operation technique and post-operative history. and not of the immunological relationship between donor and recipient. the procedure is unobjectionable.) Each recipient was immunized by the standard operation (Medawar. 1944, 1945) in which 8 large pinch grafts. each weighing modally O O45-O-o55 g.. were cut from the thigh of D and accurately fitted to separate raw areas of the appropriate size on the skin overling left side of the chest of R. Fifteento 2I1 days after this heavy preliminary immunization, single further skin grafts from D were trans-planted (a) to the centre of a 2 cm. x 2 cm. raw area stripped above the level of the panniculus carnosus from the skin overlying the right side of the chest: and (b) to one of the following positions: the left cerebral hemisphere: the sub-integumentary tissue between the panniculus carnosrs and the body wall in the chest regionand the anterior chamber of one eve. In one group of experiments six days. and in the remainder ten (lays. was allowedl to pass bettween this second operation and autopsy. The grafts were then examined in situ and fixed in formol-HgCl, for sectioning. The behaviour of the orthotopic control graft on the chest (ie the skin graft in skin positioni) pro-ided a standard to which the behaviour of the others could be individually referred. In the majority of cases its long-st-anding necrosis stood witness to tihe completeness of immunization. But in some members of the six-day autopsy group (thetime interval being chosen for this reason) the control graft showed some degree of partial survival. These experiments made possible a particularly accurate point-for-point comparison between the degree of survival of the control grafts and that of their contemporaries in anatomically unnatural positioins elsewhere.

The control grafts. and those transplanted to the brain and below the integu-ment, were rectangles ranging from 2 mm. x 3 mm. to21 mm. x 4 mm. in lengths of side. and so cut with a straight-edged scalpel blade as to include with the epidermis only the upper part of the compact connective tissue of the corium.