Apparently, then, we are confronted with this result,—that citrate plasma which causes little or no constriction of the stretched artery ring, little or no slowing of the flow through the frog perfusion preparation, while the corresponding serum produces a marked effect on both preparations, will affect the intestine or uterus preparation practically in the same way as the corresponding serum. Hirudin plasma and serum exert on the intestine and uterus preparations practically the same effect, causing a marked increase of tone. On the artery ring preparation, there is a difference although it is not so strongly marked as in the case of the citrate material. The frog perfusion preparation, as regards the effect of the hirudin material, seems to occupy an intermediate position between the intestine and uterus on the one hand, and the artery ring on the other. Of the three plasmas, the peptone plasma most closely resembles serum in its action on the artery rings. Like the other plasmas its effect on the intestine and uterus does not differ appreciably from that of the serum. From these observations the following conclusions seem justified. A change occurs in shed blood which confers on it the property of constricting artery rings and of slowing the flow through the perfused frog preparation. If this property is in any degree possessed by circulating blood it is at least markedly increased after the blood is shed. The change, whatever it may be, does not entail any essential alteration in the action of the blood on intestine and uterus segments. The tone-increasing property developed in the shed blood may, therefore, so far as the four test objects included in the present survey are concerned, be looked upon as especially affecting the blood vessels and probably their smooth muscle directly. This need not imply that the pressor substance, if it is a single definite substance developed in the shed blood, exerts no action on the smooth muscle of the intestine and uterus preparations, but merely that its action on these objects is masked by the general action of the serum and plasma, so that in the presence of the other constituents common to serum and plasma, its effect is inconspicuous or not to be detected at all, while on the blood vessel preparations, especially the artery rings, the effect of the pressor substance is the dominant one, and the general action of the serum and plasma is feeble or undetectable. It is not the clotting process as such, i.e., the actual change of fibrinogen into fibrin, that is responsible for the differences between serum and plasma revealed by the biological tests employed but some process which precedes or accompanies the clotting and which may or may not be causally related to it. Changes in formed elements of the blood under the influence of the changed conditions (contact with foreign bodies, restriction of gaseous exchange, etc.). which blood encounters as soon as it leaves the living vessels, are known to occur. Among these it is to be assumed are the changes which condition the differences between plasma and serum under discussion.16 Even if these changes represent preliminary stages in coagulation (liberation of the factors necessary to the formation of thrombin, for instance), they may still occur to a greater or less extent in blood which is prevented in certain ways from clotting since it is known that the procedures by which the various non-coagulable plasmas are obtained break at different points the chain of events which normally ends in coagulation. A procedure which simply supplies sufficient antithrombin to neutralize the thrombin which has been allowed to form in normal amount may not interfere at all with the changes in the formed elements, and the pressor property of the resulting plasma may then be as marked as that of the serum. On the other hand, a procedure which binders clotting by preventing or diminishing the alterations in the cells, for example, the addition of a substance which acts as a preservative for blood platelets, will very likely yield a plasma with little or no pressor effect in comparison with the serum. The mere prevention of clotting, then, except in so far as it is an index of the prevention of changes in the formed elements may have little significance in preventing the development of the pressor property. Indeed it is conceivable that the alterations in the cells which are connected with the development of this property may even be wholly or partially independent of the cell changes concerned in coagulation. In this case it might be possible to obtain blood which would clot without developing the pressor property. The main results which seem to follow from our observations may be thus summarized: 1. The substance, or property, developed in shed blood by which it causes constriction of artery rings, is not developed, or at least not mainly developed, in connection with the actual change of fibrinogen to fibrin, since the constrictor action of different plasmas differs greatly, while the absence of coagulation is common to all. 2. The development of the constrictor substance, or property, is associated with changes undergone probably by formed elements of the blood when it is shed. These changes may be identical with the alterations, or with some of the alterations, preliminary to clotting. It is possible, however, that there may be changes connected with the development of the pressor property which, although concomitant with the liberation of the substances concerned in the production of thrombin, are yet quite independent of the clotting process. Our observations do not enable us to decide definitely between these possibilities. 3. The constrictor substance, or property, developed in shed blood acts especially on blood vessels and does not equally affect the other organs examined. This follows from the fact that a plasma and serum which differ markedly in their action on the blood vessels may have...