Nevertheless, it must be admitted that citrate has not yet been absolutely proved to be the cause of this slight reaction in all the cases in which it occurs. Evidence has, indeed, been brought forward by Lewisohn and by Meleney to show that citrate is definitely not responsible for the reaction. The statement is made that some reaction occurs after 10 per cent. of all transfusions, and that this percentage is unaffected whether whole blood or citrated blood is used. Lewisohn has himself investigated the effects in a long series of parallel cases in which different methods were employed, and he reports that the results following the use of citrated blood were as good as with any other method. Drinker states that reactions follow the use of citrated blood slightly more often than they do that of whole blood, but this has not been confirmed. He was unable to find any impurity in the citrate that might be held responsible. It is quite possible that all the reactions observed are in reality caused by the “minor agglutinins” mentioned on p. 73. Meleney has noticed that the blood of some donors is more likely to produce a reaction than that of others; this suggests that the responsibility rests with the blood and not with the citrate. The occurrence of a toxic reaction constitutes the only real objection to the use of citrated blood that has yet been brought forward, but even this has not yet been fully substantiated; in any case, the reaction is of so little importance that it is greatly outweighed by the numerous advantages that are conferred by the use of citrate. The possibility that a citrate reaction does sometimes occur may be taken as an indication in favour of using the smaller amount recommended by Lewisohn rather than the larger dose used by Robertson. The experience of a great many observers has established the fact that citrated blood is quite as effective as whole blood in its therapeutic effects. It is convenient to have the sodium citrate in a form ready for immediate use. I have therefore been in the habit of keeping it in the solid form in small stoppered bottles, each containing 1 gramme of the salt. These are sterilized at 130° C., and can be kept indefinitely until wanted. If 450 cc. of blood or less are to be drawn, the contents of one bottle is shaken into the transfusion flask; 50 cc. (approximately 2 oz.) of sterile warm water are added, in which the citrate will rapidly dissolve. If more than 450 cc. of blood is to be used, the contents of two bottles must be dissolved in 100 cc. or 4 ozs. of water. Alternatively a concentrated solution of citrate may be kept in sealed ampoules, but the salt is less stable in solution, and I prefer to keep it in the solid form.