To investigate the role of the liver in the detoxication of anaesthetics, two main methods are available. First, the liver may be damaged by some drug such as carbon tetrachloride; if the anaesthetic has a greater effect than before, it is probable that it is normally detoxicated by the liver. This method is open to at least two objections. In the first place, it is impossible accurately to assess the amount of liver damaged by this means; and further it is possible that the carbon tetrachloride has some action of its own on the central nervous system. The second method, which we have used, is to remove by operation a large proportion of the liver and observe whether the effect of the anaesthetic is increased. This method has the advantage that the amount of liver tissue removed can be estimated fairly accurately, and it is unlikely that the operation has g.6 any complicating sideeffects.Higgins and - Anderson (1931) quent intervals the rat was laid on its side, and when it remained in this position it was considered to be asleep. As soon as it got to its feet it-was considered awake, and the sleeping-time was thus determined. At least five days later partial hepatectomy was carried out by the method of Higgins and Anderson (1931). The animals withstood the operation well, the mortality being about 10 per-cent, and by the third day after operation appeared perfectly normal. For 24 hours after operation they were fed on bread and milk and 20 per cent glucose. Thereafter they were given the normal laboratory diet. On a chosen day after hepatectomy, the rat was given the same dose of drug as before, and the sleeping-time again determined in the same way. The experiment was planned to give at least 10 results on each of the following days after hepatectomy: 6th-7th, 8th-9th, 10th, 12th, 15th, 17th-18th, 20th, 22nd, and 26th-28th. The mean sleeping-time of the rats on each of these days after