Preservation in Vitro of Living Erythrocytes. 123 



no preservative effect; agar proved toxic, as shown by early 

 laking; and dextrin had only slight preservative qualities, except 

 in concentrations which caused a browning of the blood pigment, 

 presumably to methemoglobin. Soluble starch, the watery extract 

 of coagulated blood serum, beef albumen, an aqueous solution of 

 the alcohol-soluble constituents of blood serum, and even serum 

 water made up to isotonicity with sodium chloride were all non- 

 preservative. The sugars alone have proved well suited to our 

 purpose. In a dextrose-Ringer's mixture containing 2Y2 per cent, 

 of dextrose the erythrocytes of the sheep have been kept intact 

 for a period of two months; and those of man for a month in a 

 somewhat similar solution. The limits of the method have not 

 yet been reached. The sugars which have proved best are 

 saccharose and glucose. Isotonic mixtures are better than hyper- 

 tonic. For the blood of each species a different preservative 

 solution is required. Thus, the blood of the dog keeps best in a 

 sugar-Ringer's-dextrin mixture, whereas for other bloods dextrin 

 is useless if not harmful. It is interesting that though sugars 

 and dextrin are preservative they are not protective; red cells 

 handled in solutions of them undergo as much injury as in ordinary 

 Ringer's. 



Are the cells kept intact in the preservative mixtures to be 

 considered as surviving? They can be washed repeatedly; will 

 take up and give off oxygen; and those of the sheep when used 

 for the Wassermann reaction behave exactly as do the freshly 

 drawn cells of the same animal. But this is not sufficient evidence 

 of viability. To determine the matter bleedings followed by 

 transfusions have been performed, the blood of a number of 

 animals being replaced so far as possible with red cells preserved 

 for many days in vitro and suspended in salt solution. The 

 results of such experiments with rabbits show that the washed 

 and preserved blood cells remain alive. Following a disturbance 

 in the blood count in the first few hours, associated with the 

 replacement of more than half the total blood and due to dis- 

 crepancy between the number of red cells taken out and that put 

 in, to readjustment of the blood volume, etc., there is practically 

 no change in the count, or in the readings of hemoglobin. Bile 

 has never been found in the urine after the transfusions, nor hemo- 



