As long as the manufacture of plasma preparations depends upon blood donors, we are forced to use artificial colloidal solutions. On the basis of available results it seems unlikely that plasma substitutes equal to or better than dextran can be manufactured on a gelatin or starch base. Should it become evident that the risk of hepatitis transfer is not present with deep-frozen erythrocytes, one should investigate whether solutions of erythrocytes with artificial colloids can be used. Clarification of this possibility in the near future seems appropriate, since in many countries large amounts of erythrocytes are available in the manufacture of plasma; only a small portion of these erythrocytes are used clinically. It is also conceivable that in the future artificial colloids can be developed or so modified that a certain oxygen transport and release capacity can be obtained. The discovery of such a substance would make it possible to manufacture an almost perfect blood replacement substance for initial treatment of shock. Unfortunately, there is as yet no basis for believing that this might soon be the case. Until such a time we will continue searching for new ways to reduce the risks of blood transfusions, and we will have to restrict the administration of blood to such cases where it is indicated as urgent.